oEPA
United States Office of Water EPA-842-D-22-001
Environmental Protection 4304T April 2022
Agency
DRAFT
AQUATIC LIFE AMBIENT WATER Ql ALITY CRITERIA
for
PERFLUOROCKTANOIC ACID
(PFOA)
April 2<>22
U.S. I jn ironniental Protection Agency Office of Water, Office of Science and
Technology. Health and Ecological Criteria Division
Washington, D.C.
-------
Acknowledgements
Technical Analysis Leads:
James R. Justice, Office of Water, Office of Science and Technology, Health and Ecological
Criteria Division, Washington, DC
Amanda Jarvis, Office of Water, Office of Science and Technology, Health and Ecological
Criteria Division, Washington, DC
Brian Schnitker, Office of Water, Office of Science and Technology. I lealth and Ecological
Criteria Division, Washington, DC
Mike Elias, Office of Water, Office of Science and Technology, Healih and Ideological Criteria
Division, Washington, DC
Reviewers:
Kathryn Gallagher and Elizabeth Behl, Office of Water. Office of Science and Technology,
Health and Ecological Criteria Division, Washington, DC
EPA Scoping Workgroup Reviewers:
Gerald Ankley, Laurence Burkhard, Russ I-rick son. Manlier l.llcrson. Russ Hockett, Dale Hoff,
Sarah Kadlec, Dave Mount, Carlie LaLone. and Dan Yilleneuve, Office of Research and
Development, Center for Computational Toxicology and Exposure, Great Lakes Toxicology and
Ecology Division, Duluth, MX
Anthony Williams, Office of Research and Development, Center for Computational Toxicology
and Exposure. Chemical Characterization and Exposure Division, Durham, NC (Research
Triangle Park)
Colleen Elonen. Office of Research and Development, Center for Computational Toxicology and
Exposure. Scientific Computing and Data Curation Division, Duluth, MN
Robert Burgess. Office of Research and Development, Center for Environmental Measurement
and Modeling. Atlantic Coastal Environmental Sciences Division, Narragansett, RI
Sandy Raimondo, Office of Research and Development, Center for Environmental Measurement
and Modeling, Guld Ecosystem Measurement and Modeling Division, Gulf Breeze, FL
Susan Cormier, Office of Research and Development, Center for Environmental Measurement
and Modeling, Watershed and Ecosystem Characterization Division, Cincinnati, OH
Mace Barron, Office of Research and Development, Center for Environmental Solutions and
Emergency Response, Homeland Security and Materials Management Division, Gulf Breeze, FL
Cindy Roberts, Office of Research and Development, Office of Science Advisor, Policy, and
Engagement, Science Policy Division, Washington, DC
11
-------
EPA Peer Reviewers:
Jed Costanza, Office of Chemical Safety and Pollution Prevention, Office of Pollution
Prevention and Toxics, Existing Chemical Risk Assessment Division, Washington, DC
Alexis Wade, Office of General Counsel, Water Law Office, Washington, DC
Richard Henry, Office of Land and Emergency Management, Office of Superfund Remediation
and Technology Innovation, Edison, NJ
Kelly O'Neal, Office of Land and Emergency Management, Office of Superfund Remediation
and Technology Innovation, Washington, DC
Russ Hockett, Office of Research and Development, Center for Computational Toxicology and
Exposure, Great Lakes Toxicology and Ecology Division, Duluth, MN
Karen Kesler and Lars Wilcut, Office of Water, Office of Science and Technology. Standards
and Health Protection Division, Washington, DC
Rebecca Christopher and Jan Pickrel, Office of Water, Office of Wastewater Management,
Water Permits Division, Washington, DC
Rosaura Conde and Danielle Grunzke, Office of W tiler. Office of Wetlands, Oceans, and
Watersheds, Watershed Restoration, Assessment, and Protection I)i\ ision, Washington, DC
Dan Arsenault, Region I. W ater Division, Boston, MA
Brent Gaylord, Region 2. Wtiter l)i\ ision, New York, NY
Hunter Pates. Region .v W ater l)i\ision. Philadelphia, PA
Reneti 11 til I. Joel 1 Itinsel. Lauren Petter, andKathryn Snyder, Region 4, Water Division, Atlanta,
GA
Aaron Johnson and Sydney Weiss, Region 5, Water Division, Chicago, IL
Russell Nelson. Region (\ Water Division, Dallas, TX
Ann Lavaty, Region 7. Water Division, Lenexa, KS
Tonya Fish and Maggie Pierce, Region 8, Water Division, Denver, CO
Terrence Fleming, Region 9, Water Division, San Francisco, CA
Mark Jankowski, Region 10, Lab Services and Applied Sciences Divisions, Seattle, WA
iii
-------
Action Development Process (ADP) Workgroup Members:
Tyler Lloyd, Office of Chemical Safety and Pollution Prevention, Office of Pollution Prevention
and Toxics, New Chemicals Division, Washington, DC
Thomas Glazer, Office of General Counsel, Water Law Office, Washington, DC
Stiven Foster and Kathleen Raffaele, Office of Land and Emergency Management, Office of
Program Management, Washington, DC
Kelly O'Neal, Office of Land and Emergency Management, Office of Superfund Remediation
and Technology Innovation, Washington, DC
Sharon Cooperstein, Office of Policy, Office of Regulatory Policy and Management, Policy and
Regulatory Analysis Division, Washington, DC
Cindy Roberts and Emma Lavoie, Office of Research and Development, Office of Science
Advisor, Policy, and Engagement, Science Policy Division. Washington, DC
Kay Edly and Sydney Weiss, Region 5, Water Division, Chicago, 1L
We would like to thank Russ Erickson, Da\ e Mount, and Russ I lockett, Office of Research and
Development, Center for Computational Toxicology and Exposure. Great Lakes Toxicology and
Ecology Division, Duluth, M1V lor their technical support and contribution to this document.
We would like to thank Sandy Raimondo and Crystal Lilavois, Office of Research and
Development, Center lor l-n\ ironnienial Measurement and Modeling, Gulf Ecosystem
Measuring and Modeling l)i\ ision. Gulf Breeze, FL, for their work assisting the Office of Water
in developing the esluarine murine benchmarks using Interspecies Correlation Estimates (ICE).
iv
-------
Table of Contents
Acknowledgements ii
Table of Contents v
List of Tables vii
List of Figures ix
List of Appendices xii
Acronyms xiii
Notices xvi
Foreword xvii
Executive Summary xix
1 INTRODUCTION AM) BACKGROUND 1
1.1 Previously Derived PFOA Toxicity Values and Thresholds 2
1.2 Overview of Per- and Polyfluorinated Substances (PFAS) 8
1.2.1 Physical and Chemical Properties of PFOA 11
2 PROBLEM FORMULATION 14
2.1 Overview of PFOA Sources 14
2.1.1 Manufacturing of PFOA 14
2.1.2 Sources of PFOA to Aquatic I-n\ironments 16
2.2 Environmental Fate and Transport of PI-OA in the Aquatic IEn\ ironment 17
2.2.1 Environmental Fate of PFOA in the Aquatic I Environment 17
2.2.2 Environmental Transport of PFOA in the Aquatic I Environment 18
2.3 Transformation and Degradation of PFOA Precursors in the Aquatic Environment 19
2.3.1 Biodegradation of fluorotelomer-based precursors 20
2.3.2 Biodegradation of side-chain polymers 21
2.3.3 Biodegradation of other polyfluoroalkyl substances 22
2.3.4 Non-microbial biodegradation of other polyfluoroalkyl substances 24
2.4 I Environmental Monitoring of PFOA in Abiotic Media 24
2.4.1 PFOA Occurrence and Detection in Ambient Surface Waters 25
2.5 Eiioiiccumulation and liiomagnification of PFOA in Aquatic Ecosystems 28
2.5.1 PI-OA Bioaccunuilation in Aquatic Life 28
2.5.2 Factors Influencing Potential for PFOA Bioaccumulation and Biomagnification in
Aquatic Ecosystems 30
2.5.3 Environmental Monitoring of PFOA in Biotic Media 32
2.6 Exposure Pathways of PFOA in Aquatic Environments 36
2.7 Effects of PFOA on Biota 36
2.7.1 Mechanisms of PFOA Toxicity 36
2.7.2 Potential Interactions with Other PFAS 38
2.8 Conceptual Model of PFOA in the Aquatic Environment and Effects 39
2.9 Assessment Endpoints 42
2.10 Measurement Endpoints 43
v
-------
2.10.1 Overview of Toxicity Data Requirements 43
2.10.2 Measure of PFOA Exposure Concentrations 44
2.10.3 Measures of Effect 49
2.11 Analysis Plan 52
2.11.1 Derivation of Water Column Criteria 52
2.11.2 Derivation of Tissue-Based Criteria 53
2.11.3 Translation of Chronic Water Column Criterion to Tissue Criteria 54
3 EFFECTS ANALYSIS FOR AQUATIC LIFE 56
3.1 Toxicity to Aquatic Life 56
3.1.1 Summary of PFOA Toxicity Studies Used to Derive the Aquatic Life Criteria 57
3.2 Derivation of the PFOA Aquatic Life Criteria 86
3.2.1 Derivation of Water Column-based Criteria 86
3.2.2 Derivation of Tissue-Based Criteria 93
3.3 Summary of PFOA Aquatic Life Criteria 98
4 EFFECTS CHARACTERIZATION FOR AQl ATIC I.IFE 100
4.1 Influence of Using Non-North American Resident Species on PFOA Criteria 100
4.1.1 Freshwater Acute Water Criterion with Resident Organisms 101
4.1.2 Freshwater Chronic Water Criterion with North American Resident Organisms.. 103
4.2 Consideration of Relatively Sensi ti \ e Qualitatively Acceptable Water Column-Based
Toxicity Data 105
4.2.1 Consideration of Qualitatively Acceptable Acute Data 106
4.2.2 Consideration of Qualitatively Acceptable Chronic Data 112
4.3 Evaluation of the Acute Insect Minimum Data Requirement through Interspecies
Correlation Estimates (ICE) 119
4.4 Acute to Chronic Ratios 125
4.5 Tissue-based Toxicity Studies Compared to the Chronic Tissue-based Criteria 127
4.6 F.ffects on Aquatic Plants. . 129
4.7 Summary of the I'l OA Aquatic I .ife Criteria and the Supporting Information 130
5 ri;n;riinces 132
vi
-------
List of Tables
Table Ex-1. Draft Recommended Freshwater Perfluorooctanoic acid (PFOA) Aquatic Life
Ambient Water Quality Criteria xxi
Table Ex-2. Draft Recommended Acute Perfluorooctanoic Acid (PFOA) Benchmark for the
Protection of Aquatic Life in Estuarine/Marine Waters xxi
Table 1-1. Previously Derived PFOA Toxicity Values and Thresholds 4
Table 1-2. Two Primary Categories ofPFAS1 9
Table 1-3. Classification and Chemical Structure of Perfluoroalkvl Acids (PFAAs).1 10
Table 1-4. Chemical and Physical Properties of PFOA 12
Table 2-1. Summary of Assessment Endpoints and Measures of Effect Used in the Criteria
Derivation for PFOA 51
Table 2-2. Evaluation Criteria for Screening Bioaccumulation Factors (BAI-'s) in the Public
Literature 56
Table 3-1. Summary Table of Minimum Data Requirements per the 1985 Guidelines
Reflecting the Number of Acute and Chronic (ienus and Species Level Mean
Values in the Freshwater and Saltwater Toxicity Datasets for PFOA 58
Table 3-2. The Four Most Sensitive Genua Used in Calculating the Acute Freshwater
Criterion (Sensitivity Rank 1-4) 60
Table 3-3. Ranked Freshwater Genus Mean Acute Values 67
Table 3-4. Estuarine/Marine Acute PFOA Genera 70
Table 3-5. Ranked Estuarine/Marine Genus Mean Acute Values 74
Table 3-6. The Most Sensitive Genera Used in Calculating the Chronic Freshwater Criterion
(Sensitivity Rank 1-4) 75
Table 3-7. Ranked Freshwater Genus Mean Chronic Values 85
Table 3-8. Freshwater Final Acute Value and Criterion Maximum Concentration 87
Table 3-9. Freshwater Final Chronic Value and Criterion Continuous Concentration 89
Table 3-10. Summary Statistics for PFOA BAFs in Invertebrate Tissues and Various Fish
Tissues1 94
Table 3-1 I Recommended I 'reshwater Perfluorooctanoic acid (PFOA) Aquatic Life
Ambient Water Quality Criteria 100
Table 4-1. Ranked I 'reshwater Genus Mean Acute Values with North American Resident
Organisms 102
Table 4-2. Freshwater Exploratory Final Acute Value and Acute Water Column
Concentration with North American Resident Organisms (zebrafish included) 103
Table 4-3. Ranked Freshwater Genus Mean Chronic Values with Resident Organisms 104
Table 4-4. Freshwater Exploratory Final Chronic Value and Chronic Water Column
Concentration with North American Resident Organisms 105
Table 4-5. All ICE models available in Web-ICE v3.3 for predicted insect species based on
surrogates with measured PFOA 121
Table 4-6. ICE-estimated Insect Species Sensitivity to PFOA 124
vii
-------
Table C-l. EC so to ECio ratios from all quantitatively acceptable chronic concentration-
response curves with species similar to H. azteca (i.e., small members of the
subphylum Crustacea) and with endpoints that were based on reproduction
per female C-5
Table L-l. Surrogate Species Measured Values for PFOA and Corresponding Number of
ICE Models for Each Surrogate L-7
Table L-2. Comparison of ICE-predicted and measured values of PFOA for species using
both scaled values (entered as mg/L) and values potentially beyond the model
domain (entered as (J,g/L) (Raimondo et al. in prep)... L-9
Table L-3. All ICE Models Available in Web-ICE v3.3 for Saltwater Predicted Species
Based on Surrogates with Measured PFOA L-15
Table L-4. ICE-estimated Species Sensitivity to PFOA L-17
Table L-5. Ranked Estuarine/Marine Genus Mean Acute Values L-20
Table L-6. Estuarine/Marine Final Acute Value and Protective Aquatic Acute lienchmark. ...L-21
Table M-l. Correlations of paired nominal and measured PFO A concentrations across
various experimental conditions in freshwater toxicity tests M-6
Table M-2. Percentage of Measured PFOA Concentrations Falling Outside of 20% of
Corresponding Nominal Concentrations as well as the Minimum and Maximin
of Measured as Percent of Nominal Concentrations
(i.e., Measured/Nominal*100) across Range of I-\perimental Conditions M-14
Table M-3. Paired Nominal and Measured PFOA Concentrations from Quantitatively and
Qualitatively Acceptable Freshwater Toxicity Tests that Reported Measured
PFOA Concentrations M-20
Table M-4. Paired Nominal and Measured PFOA Concentrations from Quantitatively and
Qualitatively Acceptable Saltwater Toxicity Tests that Reported Measured
PI OA Concentrations M-25
Table N-l. Measured Perfluorooctanoic acid (PFOA) Concentrations in Surface Waters
Across the United States N-l
Table O-l Characteristics of adult fish sampled for the calculation of PFOA reproductive
tissue BAFs 0-2
Table 0-2. Summary Statistics for PFOA Freshwater BAFs in Additional Fish Tissues1 0-3
Table 0-3. PFOA Concentrations for Additional Fish Tissue Values.1'2 0-4
viii
-------
List of Figures
Figure 1-1. Chemical Structure of the Linear Isomer of Perfluorooctanoic acid (PFOA) 11
Figure 2-1. Synthesis of Perfluorooctanoic acid (PFOA) by Electrochemical Fluorination
(ECF) 15
Figure 2-2. Map Indicating Sampling Locations for Perfluorooctanoic acid (PFOA)
Measured in Surface Waters Across the United States (U.S.) Based on Data
Reported in the Publicly Available Literature 26
Figure 2-3. Distribution of the minimum and maximum concentrations (ng/L) of
Perfluorooctanoic acid (PFOA) measured in surface waters for each state or
waterbody (excluding the Great Lakes) with reported data in the publicly available
literature 27
Figure 2-4. Conceptual Model Diagram of Sources, Compartmental Partitioning, and Trophic
Transfer Pathways of Perfluorooctanoic acid (PFOA) in the Aquatic I-nvironment
and its Bioaccumulation and Effects in Aquatic Life and Aquatic-dependent
Wildlife 41
Figure 3-1. Ranked Freshwater Acute PFOA GMAVs Fulfilling the Acute Family MDR 69
Figure 3-2. Acceptable Estuarine/Marine GMAVs 74
Figure 3-3. Freshwater Genus Mean Chronic Values for PFOA 86
Figure 3-4. Ranked Freshwater Acute PFOA GMAVs used for the Criterion Calculation
and the Qualitative Value for the Insect MDR Group 88
Figure 3-5. Freshwater Quantitative GMCVs used for the Criterion Calculation 90
Figure L-l. Example ICF. Model for Rainbow Trout (surrogate) and Atlantic Salmon
(predicted) L-4
Figure L-2. Ranked l-stuaiine Marine Acute PFOA GMAVs Used for the Aquatic Life
Acute Benchmark Calculation L-21
Figure L-3. Americamysts bahia (\-a\is) and Daphnia magna (Y-axis) regression model
used for ICI- predicted \alues L-26
Figure I .-4 . \mericamysis bahia (X-axis) and Lepomis macrochirus (Y-axis) regression
model used for ICI- predicted values L-26
Figure L-5 . \mcncamysis bahia (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used lor ICE predicted values L-27
Figure L-6. American lysis bahia (X-axis) and Pimephales promelas (Y-axis) regression
model used for ICE predicted values L-27
Figure L-7. Danio rerio - embryo (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values L-28
Figure L-8. Danio rerio - embryo (X-axis) and Lepomis macrochirus (Y-axis) regression
model used for ICE predicted values L-28
Figure L-9. Danio rerio - embryo (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for ICE predicted values L-29
IX
-------
Figure L-10. Danio rerio - embryo (X-axis) and Pimephalespromelas (Y-axis) regression
model used for ICE predicted values L-29
Figure L-l 1. Daphnia magna embryo (X-axis) and Americamysis bahia (Y-axis) regression
model used for ICE predicted values L-30
Figure L-l2. Daphnia magna (X-axis) and Lampsilis siliquoidea (Y-axis) regression model
used for ICE predicted values L-30
Figure L-13. Daphnia magna (X-axis) and Lepomis macrochirus (Y-axis) regression model
used for ICE predicted values L-31
Figure L-14. Daphnia magna (X-axis) and Lithobates catesbeianus (Y-axis) regression
model used for ICE predicted values L-31
Figure L-l5. Daphnia magna (X-axis) and Oncorhynchus mykiss (Y-a\is) regression
model used for ICE predicted values L-32
Figure L-16. Daphnia magna embryo (X-axis) and Pimephales promelas (Y-a\is)
regression model used for ICE predicted values L-32
Figure L-l7. Lampsilis siliquoidea (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values L-33
Figure L-18. Lampsilis siliquoidea (X-axis) and Lepomis macrochirus (Y-axis) regression
model used for ICE predicted \ allies L-33
Figure L-19. Lampsilis siliquoidea (X-axis) and / igmiiia recla (Y-a\is) regression model
used for ICE predicted values L-34
Figure L-20. Lampsilis siliquoidea (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for 1C11 predicted values L-34
Figure L-21. Lampsilis siliquoidea (X-axis) and Pimephales promelas (Y-axis) regression
model used for ICE predicted values L-35
Figure L-22. Lepomis macrochirus (X-axis) and . Imericamysis bahia (Y-axis) regression
model used lor ICE predicted \ allies L-35
Figure I .-23 Lepomis macrochirus (\-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values L-36
Figure I .-24 Lepomis macrochirus (X-axis) and Lampsilis siliquoidea (Y-axis) regression
model used for ICI- predicted values L-36
Figure L-25. Lepomis macrochirus (X-axis) and Lithobates catesbeianus (Y-axis)
regression model used for ICE predicted values L-37
Figure L-26. Lepomis macrochirus (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for ICE predicted values L-37
Figure L-27. Lepomis macrochirus embryo (X-axis) and Pimephales promelas (Y-axis)
regression model used for ICE predicted values L-3 8
Figure L-28. Ligumia recta (X-axis) and Lampsilis siliquoidea (Y-axis) regression model
used for ICE predicted values L-3 8
Figure L-29. Lithobates catesbeianus (X-axis) and Daphnia magna (Y-axis) regression
model used for ICE predicted values L-39
x
-------
Figure L-30. Lithobates catesbeianus (X-axis) and Lepomis macrochirus (Y-axis)
regression model used for ICE predicted values L-39
Figure L-31. Lithobates catesbeianus (X-axis) and Oncorhynchus mykiss (Y-axis)
regression model used for ICE predicted values L-40
Figure L-32. Lithobates catesbeianus (X-axis) and Pimephalespromelas (Y-axis)
regression model used for ICE predicted values L-40
Figure L-33. Oncorhynchus mykiss (X-axis) and Americamysis bahia (Y-axis) regression
model used for ICE predicted values L-41
Figure L-34. Oncorhynchus mykiss (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values L-41
Figure L-35. Oncorhynchus mykiss (X-axis) and Lampsilis s/ln/iionlea (Y-a\is) regression
model used for ICE predicted values L-42
Figure L-36. Oncorhynchus mykiss (X-axis) and Lepomis macrochirus (Y-a\is) regression
model used for ICE predicted values L-42
Figure L-37. Oncorhynchus mykiss (X-axis) and Lithobates catesbeianus (Y-axis)
regression model used for ICE predicted values L-43
Figure L-38. Oncorhynchus mykiss (X-axis) and Pimephales promelas (Y-axis) regression
model used for ICE predicted \ allies L-43
Figure L-39. Pimephales promelas (X-axis) and . \iiiericamysis baliia (Y-axis) regression
model used for ICE predicted values ... L-44
Figure L-40. Pimephales promelas (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values L-44
Figure L-41. Pimephales promelas (X-axis) and Lampsilis siliquoidea (Y-axis) regression
model used for ICE predicted values L-45
Figure L-42. Pimephales promelas (X-axis) and I
-------
Figure M-3. PFOA measured vs. nominal concentrations for freshwater tests conducted in
plastic/steel test vessels (Panel A, top) and freshwater tests conducted in glass
test vessels (Panel B, bottom) data M-10
Figure M-4. Measured concentrations as a percent of corresponding nominal concentrations
with horizontal lines to denote where the relative ratio (i.e., Y-axis) differs by
more than 20% and 30% M-12
Figure N-l. Distribution of the minimum and maximum concentrations (ng/L) of
Perfluorooctanoic acid (PFOA) measured in surface water samples collected
from the Great Lakes as reported in the publicly available literature N-8
Figure N-2. Comparison of relatively high (A; greater than 30 ng/1.) and low (B; less than
30 ng/L) maximum Perfluorooctanoic acid (PFOA) concentrations (ng/L)
measured in surface water samples collected across the United Suites (U.S.) as
reported in the publicly available literature N-l3
List of Appendices
Appendix A Acceptable Freshwater Acute PFOA Toxicity Studies A-l
Appendix B Acceptable Estuarine/Marinc Acute PFOA Toxicity Studies B-l
Appendix C Acceptable Freshwater Chronic PI'OA Toxicity Studies C-l
Appendix D Acceptable Estuarine/Mari no Chronic PFOA Toxicity Studies D-l
Appendix E Acceptable Freshwater Plant PFOA Toxicity Studies E-l
Appendix F Acceptable Estuarine/Marine Plant PFOA Toxicity Studies F-l
Appendix G Other Freshwater PFOA Toxicity Studies G-l
Appendix H Other Estuarine/Marine PFOA Toxicity Studies H-l
Appendix I Acute-to-Chronic Ratios 1-1
Appendix .T I Inused PFOA Toxicity Studies J-l
Appendix k EPA Methodology lor l-itting Concentration-Response Data and Calculating
Effect Concentrations K-l
Appendix I. Derivation of Acute Protective PFOA Benchmarks for Estuarine/Marine
Waters through a New Approach Method (NAM) L-l
Appendix M Mela-Analysis of Nominal Test Concentrations Compared to Corresponding
Measured Test Concentrations M-l
Appendix N Occurrence of PFOA in Abiotic Media N-l
Appendix O Translation of The Chronic Water Column Criterion into Other Fish Tissue
Types 0-1
Appendix P Bioaccumulation Factors (BAFs) Used to Calculate PFOA Tissue Values P-l
Appendix Q Example Data Evaluation Records (DERs) Q-l
xii
-------
Acronyms
8:2 FTAC
fluorotelomer acrylate
8:2 FTMAC
fluorotelomer methacrylate
8:2 FTOH
fluorotelomer alcohol
8:2 FTS
fluorotelomer stearate
8:2 HMU
aliphatic diurethane ester
ACR
Acute-to-Chronic Ratio
ADP
Action Development Process
AFFF
aqueous film-forming foam
AIC
Akaike information criteria
ASTM
American Society for Testing and Materials
AWQC
National Recommended Ambient Water Oualily Criteria
BAF
bioaccumulation factor
BMF
biomagnification factors
C8/C8-PFPIA
Bis(perfluorooctyl) phosphinic acid
C8-PFPA
Perfluorooctyl phosphonic acid
CAS
Chemical Abstracts Service
CASRN
Chemical Abstracts Service Registry Number
C-F
carbon fluorine
C-R
concentration-response
CC
Chronic Criterion
CCC
Criterion Continuous Concentration
CMC
Criterion Maximum Concenlialion
CWA
Clean Water Act
DER
Data Evaluation Record
dpf
days post fertilization
diPAPs
polyfluoroalkyl phosphoric acid dieslers
drc
dose-response curve
dw
dry weight
ECF
1 ¦ 1 ec 1 roc 11 em i ca 1 ll u ori n ati on
ECOTOX
ECOTOXicology database
ECx
Effect concentration at x percent level
EPA
U.S. Em ironmental Protection Agency
EU
F.uropean I nion
FACR
I'inal Acule-to-Chronic Ratio
FAV
I'inal Acute Value
FCV
I'inal Chronic Value
FTEOs
fluorotelomer ethoxylates
FTCAs
fluorotelomer carboxylates
FTOH
fluorotelomer alcohol
GLI
Great Lakes Initiative
GMAV
genus mean acute value
GMCV
genus mean chronic value
GSD
genus sensitivity distribution
GSI
gonadal somatic index
hpf
hours post fertilization
xiii
-------
HIS
ICE
ICx
Kd
Koc
Kow
LCx
LOECs
LOQ
MATC
MC
MDRs
NAMS
NOECs
NPDES
OECD
OCSPP
ORD
OW
ppt
PFAAs
PFAS
PFCAs
PFdiCAs
PFdiSAs
PFECAs
PFESAs
PFO
PFOA
PFO A An is
PFOS
PFOSI
PFPAs
PFPIAs
PFSAs
PFSIAs
pKa
QACs
SMACR
SMAV
SMCV
SOP
hepatic somatic index
Interspecies Correlation Estimation
Inhibitory concentration at x percent level
partitioning coefficients
Organic carbon water partitioning coefficient
n-octanol-water partition coefficient
Lethal concentration at x percent level
Lowest Observed Effect Concentrations
limit of quantification
Maximum Acceptable Toxicant Concentration
Maximum Criterion
minimum data requirements
New Approach Methods
No Observed Effect Concentrations
National Pollutant Discharge Elimination System
Organization for Economic Cooperation and Development
Office of Chemical Safety and Pollution Prevention
Office of Research and Development
Office of Water
parts per thousand
perfluoroalkyl acids
per- and polyfluorinatcd substances
Perfluorocarboxylic acids, perlluomulkvl cait>o\\ lates or perfluoroalkyl
carboxylic acids
Perfluoroalkyl dicarboxylic acids
Perfluoroalkane disulfonic acids
Perfluoroalkyl ether carboxylic acids
Perfluoroalkylether sulfonic acids
Perfluorooctanoate
Pulluoiooctanoic acid, penladecafluoro-l-octanoic acid,
pentadecafluoro-n-octanoic acid, octanoic acid, pentadecafluoro-,
perfluomcaprylic acid, pentadecafluorooctanoic acid, perfluoroalkyl
carboxylic acid or pertluoroheptanecarboxylic acid
perfluorooctaneamido quaternary ammonium salt
Perfluorooctane sulfonic acid or perfluorooctane sulfonate
Perfluorooctane sulfinic acid
Pcilluoioalkyl phosphonic acids
PerlTuoroalkyl phosphinic acids
Perfluoroalkane sulfonic acids or perfluorokane sulfonates
Perfluoroalkyl sulfinic acids
acid dissociation constant
quaternary ammonium polyfluoroalkyl surfactants
species mean acute-to-chronic ratio
species mean acute value
species mean chronic value
standard operating procedure
xiv
-------
SSD
species sensitivity distribution
TMDLs
Total Maximum Daily Loads
TSCA
Toxic Substances Control Act
U.S.
United States
web-ICE
Web-based Interspecies Correlation Estimation
WQS
water quality standards
WW
wet weight
WWTPs
wastewater treatment plants
XV
-------
Notices
This draft document provides information to states and tribes authorized to establish
water quality standards under the Clean Water Act (CWA), to protect aquatic life from toxic
effects of Perfluorooctanoic acid (PFOA). Under the CWA, states and tribes are to establish
water quality criteria to protect designated uses. State and tribal decision makers retain the
discretion to adopt approaches that are scientifically defensible thai differ from these criteria to
reflect site-specific conditions. While this document contains the I -n\ ironmenlal Protection
Agency's (EPA) draft scientific recommendations regarding ambient concentrations of PFOA
that protect aquatic life, the draft PFOA Criteria Document does not substitute lor the Clean
Water Act or the EPA's regulations; nor is it a regulation itself Thus, the document when final
would not impose legally binding requirements 011 the EPA, states, tribes, or the regulated
community, and might not apply to a particular situation hased upon the circumstances. The EPA
intends to finalize this document in the future This draft document has been approved for
publication by the Office of Science and Technology, Office of Water, U.S. Environmental
Protection Agency.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use. This document can be downloaded from:
https://www cpa uo\ wqc/aqualic-life-criteria-perfluorooctanoic-acid-pfoa.
xvi
-------
Foreword
The Clean Water Act Section 304(a)(1) (P.L. 95-217) directs the Administrator of the
EPA to publish water quality criteria that accurately reflect the latest scientific knowledge on the
kind and extent of all identifiable effects on health and welfare that might be expected from the
presence of pollutants in any body of water, including groundwater This document is a draft
ambient water quality criteria (AWQC) document for the protection of aquatic life based upon
consideration of all available information relating to effects of perfluorooctanoic acid on aquatic
organisms.
The term Water Quality Criteria is used in two sections of the CWA, Section 304(a)(1)
and Section 303(c)(2). The term has different meanings in each section. Under CWA section
304, the term represents a non-regulatory, scientific assessment of ecological and human health
effects. Criteria presented in this draft document are such a scientific assessment of ecological
effects. Under CW A section 3<)3. when water quality criteria associated with specific surface
water uses are adopted In a state or authorized tribe and approved by EPA as water quality
standards, they become the CWA water quality standards applicable in ambient waters within
that state or authorized tribe Water quality criteria adopted in state/authorized tribal water
quality standards could ha\ e the same numerical values as recommended criteria developed
under CWA section 3<)4 I lowever, in some situations, states/authorized tribes might want to
adjust water quality criteria developed under CWA section 304 to reflect local water chemistry
or ecological conditions. Alternatively, states and authorized tribes may develop numeric criteria
based on other scientifically defensible methods that are protective of designated uses.
Guidelines to assist the states and authorized tribes in modifying the criteria presented in this
draft document are contained in the Water Quality Standards Handbook (U.S. EPA 2014).
xvii
-------
This document presents draft recommendations only. It does not establish or affect legal
rights or obligations. It does not establish a binding requirement and cannot be finally
determinative of the issues addressed. The EPA will make decisions in any particular situation
by applying the CWA and the EPA regulations on the basis of specific facts presented and
scientific information then available.
Deborah (i \auk-
Director
Office of Science and Technology
xviii
-------
Executive Summary
U.S. Environmental Protection Agency (EPA) developed the draft recommended
perfluorooctanoic acid (PFOA) aquatic life ambient water quality criteria in accordance with the
provisions of section304(a) of the Clean Water Act. This document provides EPA's basis for and
derivation of the draft national PFOA ambient water quality criteria recommendations for fresh
and saltwater environments to protect aquatic life. EPA has dm fled the PFOA aquatic life criteria
to be consistent with methods described in EPA's "Guidelines for Deriving Numerical National
Water Quality Criteria for the Protection of Aquatic Organisms and Their I ses" (U.S. EPA
1985).
PFOA is an organic, human-made perfluorinated compound, consisting of a seven-carbon
backbone and a carboxylate functional group PI-OA (and oilier related chemicals in the
perfluorocarboxylic acids, PFCAs) is used primarily in specialized applications associated with
surface coatings in a \ ariety of industrial and commercial products. This document provides a
critical review of toxicity data identified in EPA's literature search for PFOA, including the
anionic form (CAS \o 452S5-5 I -(•>). the acid form (CAS No. 335-67-1), and the ammonium salt
(CAS No 3S25-20-1). It also quantifies the toxicity of PFOA to aquatic life, and provides draft
criteria to protect aquatic life in freshwater from the acute and chronic toxic effects of PFOA.
The draft Aquatic 1 .ile Ambient Water Quality Criteria for the draft PFOA document
includes water column-based acute and water column-based chronic criteria, as well as chronic
tissue-based criteria for freshwaters. Quantitatively-acceptable estuarine/marine toxicity data
only fulfilled three of the eight minimum data requirements (MDRs) for deriving an acute
estuarine/marine criterion, and none of the eight MDRs for deriving a chronic estuarine/marine
criterion per the 1985 Guidelines. EPA did, however, include an acute aquatic life benchmark for
estuarine/marine environments in Appendix L, using available estuarine/marine species toxicity
xix
-------
data and the New Approach Methods (NAMS) application of EPA Office of Research and
Development's (ORD) peer-reviewed web-based Interspecies Correlation Estimate tool (Web-
ICE; Version 3.3; https://www.epa.gov/webice/). Both the freshwater criteria and the acute
estuarine/marine benchmark are draft recommendations for states/authorized tribes to consider as
protective values in their state water quality protection programs. ITo\\c\ er, the acute
estuarine/marine benchmark magnitude is less certain than the freshwater criteria since it was
based on both empirical and estimated acute toxicity data (Appendix I.)
The draft freshwater acute water column-based criterion magnitude is -N nig/L, and the
draft chronic water column-based chronic criterion magnitude is 0.094 mg/L. The draft chronic
freshwater criterion also contains tissue-based criteria with magnitudes of 6.10 mg/kg wet weight
(ww) for fish whole-body, 0.125 mg/kg \\\\ lor lisli muscle tissue, and I 11 mg/kg ww for
invertebrate whole-body tissue. All criteria are intended to he equally protective against adverse
PFOA effects and are intended to he independently applicable. The three tissue criteria
magnitudes (for lisli and in\ ertehrale tissues) are translations of the chronic water column
criterion for freshwater using hioaccunuilation factors (BAFs) derived from a robust national
dataset of li.\l"s (liurkhard 2<>21) The assessment of the available data for fish, invertebrates,
amphibians, and plants indicates these criteria are expected to protect the freshwater aquatic
community
xx
-------
Table Ex-1. Draft Recommended Freshwater Perfluorooctanoic acid (PFOA) Aquatic Life
Ambient Water Quality Criteria
Type/Media
Acute \Yaid-
Co! ii in n
(CMC)14
Chronic
\\ aid-
Co! ii in n
(CCCV5
Chronic
Inverlebrale
\\ hole-
liod v 12
Chronic
lisli
\\ hole-
IJod v1-2
Chronic Fish
Muscle12
Magnitude
49 mg/L
0.094 mg/L
1.11 mg/kg
WW
6.10 mg/kg
WW
0.125 mg/kg
WW
Duration
One hour average
Four day
average
Instantaneous3
Frequency
Not to be
exceeded more
than once in three
years on average
Not to be
exceeded more
than once in
three years on
average
Not to be exceeded more than once in ten
years on average
1 All five of these water column and tissue criteria are intended to be independently applicable and no one criterion takes
primacy. All of the above recommended criteria (acute and chronic water column and tissue criteria) are intended to be
protective of aquatic life. These criteria are applicable throughout the year.
2 Tissue criteria derived from the chronic water column concentration (CCC) with the use of bioaccumulation factors and are
expressed as wet weight (ww) concentrations.
3 Tissue data provide instantaneous point measurements that reflect integrative accumulation of PFOS over time and space in
aquatic life population(s) at a given site.
4 Criterion Maximum Concentration; applicable throughout the water column.
5 Criterion Continuous Concentration; applicable throughout the water column.
Table Ex-2. Draft Recommended Acute Perlliiorooclanoic Acid (PFOA) Benchmark for
the Protection of Aquatic l.ife in Estuarine/Marine Waters.
Type/Media
Acute W ater Column Benchmark
Magnitude
7.0 mu 1.
Duration
One hour a\ eraue
Frequency
Not lo he exceeded more than once in three years on average
xxi
-------
1 INTRODUCTION AND BACKGROUND
National Recommended Ambient Water Quality Criteria (AWQC) are established by the
EPA under the CWA. Section 304(a)(1) states that aquatic life criteria serve as recommendations
to states and tribes by defining ambient water concentrations that will protect against
unacceptable adverse ecological effects to aquatic life resulting from exposure to pollutants
found in water. Once EPA publishes final CWA section 304(a) recommended water quality
criteria, states and authorized tribes may adopt these criteria into their water quality standards
(WQS) to protect the designated uses of water bodies. States and authorized irihes may also
modify these criteria to reflect site-specific conditions or use other scientifically defensible
methods to develop criteria before adopting these into standards. After adoption, states/
authorized tribes are to submit new and re\ ised WQS to EPA lor review and approval or
disapproval. When approved by EPA, the state's tribe's WQS become the applicable WQS for
CWA purposes. Such purposes include identification of impaired waters and establishment of
Total Maximum Daily Loads (TMI)l.s) under CWA section 303(d) and derivation of water
quality-based effluent limitations in permits issued under the CWA section 402 National
Pollutant Discharge l-limination System (NPDES) program. EPA would recommend the
adoption of both the acute and chronic water column criteria as well as the chronic tissue-based
criteria to ensure the protection of aquatic life through all exposure pathways, including direct
aqueous exposure and hioaccumulation. The draft estuarine/marine benchmarks are provided in
Appendix L as additional protective values that states and tribes may consider in their water
quality programs.
This document provides a critical review of toxicity data identified in EPA's literature
search for PFOA, including the anionic form (CAS No. 45285-51-6), the acid form (CAS No.
335-67-1), and the ammonium salt (CAS No. 3825-26-1). It also quantifies the toxicity of PFOA
1
-------
to aquatic life and provides draft criteria to protect aquatic life in freshwater from the acute and
chronic toxic effects of PFOA.
EPA derived the draft recommended criteria using the best available data to reflect the
latest scientific knowledge on the toxicological effects of PFOA on aquatic life. EPA developed
the criteria following the general approach outlined in the EPA's "Guidelines for Deriving
Numerical Water Quality Criteria for the Protection of Aquatic (hgamsms and Their Uses"
(U.S. EPA 1985). The draft PFOA criteria are expected to be protecth e of most aquatic
organisms in the community (i.e., approximately 95 percent of tested aquatic organisms
representing the aquatic community) and are deri\ ed lo he protective of aquatic life designated
uses established by states and authorized tribes for freshuaters The draft estuarine/marine
benchmarks are also intended to be proteeli\ e of aquatic life designated uses, but they are based
on fewer empirical PFOA toxicity data and. therefore. ha\ e greater inherent uncertainty. The
draft criteria presented herein are l-IW's best estimate (if the maximum concentrations of PFOA,
with associated frequency and duration specifications, that would protect sensitive aquatic life
from unacceptable acute and chronic effects of PI'OA.
1.1 Previously Derived PFOA Toxicity Values and Thresholds
Other jurisdictions (eg. stales, countries, etc.) have previously published PFOA acute
and chronic criteria, benchmarks, or thresholds, including values for both freshwater and marine
systems. These values locus exclusively on water column-based values only; no other
jurisdiction has previously derived tissue-based values. Within the United States, no states or
tribes have CWA Section 303(c) approved PFOA water quality standards for the protection of
aquatic life. However, two states have acute and chronic protective values that were developed to
be numerical translations of CWA Section 303(c) narrative water quality criteria (e.g., Michigan
2
-------
and Minnesota) and other states have published draft/interim acute and chronic ecological
screening level values/benchmarks for the protection of aquatic life (e.g., Texas, Florida,
California).
These publicly available freshwater acute values range from 4.47 mg/L in Texas (TCEQ
2021) to 20 mg/L in Florida (Stuchal and Roberts 2019) (Table 1-1). No acute estuarine/marine
criteria, benchmarks, or protective values have been established lor PI-OA Publicly available
freshwater chronic values for other jurisdictions range from 0.22 mu l.in Anslralia/New Zealand
(95% species protection level; CRC CARE 2017; EPAV 2016; HEPA 202<). Table 1-1) to 2.27
mg/L in Texas (TCEQ 2021)
Previously published estuarine/marine chronic values are available for Australia/New
Zealand with a chronic protective value of <>.22 niu'I. (95% species protection level) and
California with a chronic "interim final screening le\ el \ nine" of <> 54 mg/L (Table 1-1).
3
-------
Table 1-1. Previously Derived PFOA Toxicity Values and Thresholds.
Stale/Country
of
Applicability
Aquatic l.ilo Protective
Value (mg/L)
Criteria or licnchinark and Calculation Approach
Source
Freshwater Acute
Texas
4.47
Based on NOAELs, LOAELs, or similar values from specific toxicological
studies. Contact the TCF.Q for more information This is an acute surface water
benchmark and does not represent a CWA Section .><>.>( c) approved water
quality standard for PF() A
TCEQ 2021
Michigan
7.7
Calculated from a species sensi li \ ilv distribution (SSD) consisting of two
species-specific values. The final Acute Value (FAV) was based on the lowest
ECso divided In a safety factor of 1.1 (loll owing US EPA Great Lakes Initiative
[GLI; US EPA 1995a]). This protecli\ e \ alue is a translation of narrative water
quality criteria and does not represent a CWA Section 303(c) approved water
quality standard for PFOA.
EGLE 2010
Minnesota
15
Calculated from a species sensitivity distribution (SSD) consisting of three
species-specific values. The Maximum Criterion (MC) was based on the lowest
l-Cs" di\ ided hy a safety factor of 13 (following US EPA Great Lakes Initiative
| (il. 1. I S IPX 1995a]). This protective value is a translation of narrative water
quality criteria and does not represent a CWA Section 303(c) approved water
quality standard for PFOA.
STS/MPCA
2007
Florida
ZD
Secondary Acute Value (SAV) calculated using US EPA Great Lakes Initiative
(GLI; USLIW 1995) Tier 11 Methodology. FAV calculated as the lowest
GMAV (unspecified) divided by a safety factor of 5.2. This value was released
in a White Paper sponsored by Florida Department of Environmental Protection
and is considered a draft eco-based surface water screening level, it is not a
CWA Section 303(c) approved water quality standard.
Stuchal and
Roberts
2019
4
-------
State/Country
ol'
Applicability
Aquatic l.il'e Protective
\ ;ilno (mg/l.)
Criteria or Benchmark and Calculation Approach
Source
Freshwater Chronic
Australia, New
Zealand
0.019
(99% species protection - high
conservation value systems)
Guidelines calculated from a species sensitivity distribution (SSD) consisting of
12 species-specific values for fish, insects, crustaceans, rotifers, algae, and
plants following the guidance of \Yarne et al. (2017) and Batley et al. (2014)
CRCCARE
2017,
EPAV
2016,
HEP A 2020
0.22
(95% species protection -
slightly to moderately disturbed
systems)
0.632
(90% species protection - highly
disturbed systems)
1.824
(80% species protection - highly
disturbed systems)
California
0.54
(99% species protection)
HC1 calculated from an acute and chronic NOEC-based SSD as reported in
SERDP Project l-R 1S-1 (->14 (SI -RI)P 2d 1 lJ). Acute NOEC values were
converted to chronic values using mean acute-to-chronic ratios derived from
Giesy et al. (2010). This value represents an "Interim Final Environmental
Screening Level" and does not represent a CWA Section 303(c) approved water
quality Standard for PFOA.
San
Francisco
Bay
RWQCB
2020;
SERDP
2019
Michigan
0 SS
Final Chronic Value (FCV) was calculated as the FAV Final Acute: Chronic
ratio (ACR) (following the GLI; US EPA 1995a). This protective value is a
translation of narrative water quality criteria and does not represent a CWA
Section 303(c) approved water quality standard for PFOA.
EGLE 2010
5
-------
Slate/Country
ol'
Applicability
Aquatic Life Protective
\ ;ilno (mg/L)
Criteria or Benchmark and Calculation Approach
Source
Florida
1.3
Secondary Chronic Value (SCV) calculated using US EPA Great Lakes
Initiative (GLI; USEPA 1995) Tier 11 Methodology with acute-to-chronic ratio
(ACR)of 15.3. SCV = SAV (20,000 |ig/L) ACR (15.3) = 1,300 |ig/L or 1.3
mg/L. This value was released in a White Paper sponsored by Florida
Department of Environmental Protection and is considered a draft eco-based
surface water screening level, it is not a CWA Section 3<)3(c) approved water
quality standard.
Stuchal and
Roberts
2019
Minnesota
1.7
Chronic Criterion (CC) calculated as the FAV ^ a generic ACR following
Minnesota Rules Chapter 7<)5<) No species-specific ACRs were available at the
time to calculate the FACR This protective value is a translation of narrative
water quality criteria and does not represent a CWA Section 303(c) approved
water quality standard for PFOA.
STS/MPCA
2007
Texas
2.77
Based on NOAELs, LOAELs, or similar \ alues from specific toxicological
studies. Contact the TCEQ for more information. This is a chronic surface
water benchmark and does not represent a CWA Section 303(c) approved water
quality standard for PFOA.
TCEQ 2021
Marine Chronic
California
0.54 (W.i species
protection)
1IC1 calculated from an acute and chronic NOEC-based SSD as reported in
SI-RDP Project I-RIK-IM4 (SERDP 2019). Acute NOEC values were
coin cried lo chronic \ allies using mean acute-to-chronic ratios derived from
Giesy el al. (2010). This value represents an "Interim Final Environmental
Screening 1 ,e\ el" and does not represent a CWA Section 303(c) approved water
quality Standard for PFOA.
San
Francisco
Bay
RWQCB
2020;
SERDP
2019
6
-------
Slsito/Coiinlrv
ol'
Applicability
Aqunlic l.il'c Protective
\ ;ilno (mg/l.)
Critcri:i or licnchmnrk mid ( nlculntioii Approach
Source
Australia, New
Zealand
0.019
(99% species protection - high
conservation value systems)
Freshwater values are to lx- used on an interim hasis until final marine guideline
values can be set using the nationally agreed process under the Australian and
New Zealand Guidelines for 1'ivsh and Marine Water Ouality.
HEP A 2020
0.22
(95% species protection -
slightly to moderately disturbed
systems)
0.632
(90% species protection - highly
disturbed systems)
1.824
(80% species protection - highly
disturbed systems)
7
-------
1.2 Overview of Per- and Polyfluorinated Substances (PFAS)
Perfluorooctanoic acid (PFOA), and its salts, belong to the per- and polyfluorinated
substances (PFAS) group of chemicals. EPA's Office of Pollution Prevention and Toxics
(OPPT) defines a PFAS as: any chemical substance or mixture that structurally contains the unit
R-(CF2)-C(F)(R')R". Both the CF2 and CF moieties are saturated carbons. None of the R groups
(R, R' or R") can be hydrogen. The carbon-fluorine (C-F) bond is strong and stable due to the
strong electronegativity and intermediate atomic size of fluorine. The chemical structure of the
perfluoroalkyl moiety make PFAS water and oil repellent, chemically and thermally stable, and
exhibit surfactant properties. Due to these properties. PI-AS have been used in a wide range of
industrial and consumer products since the 1940s with common uses including wetting agents,
lubricants, corrosion inhibitors, firefightinu loams, and stain-resistant treatments to leather,
paper, and clothing.
There are many families of PFAS, and each contain many individual homologues and
isomers (Buck et al 2<)| I) These PI'AS families can be divided into two primary categories:
nonpolymers and polymers The nonpolymer PFAS include perfluoroalkyl and polyfluoroalkyl
substances Polymer PI AS include lluoropolymers, perfluoropolyethers, and side-chain
fluorinated polymers (Table 1-2).
8
-------
Table 1-2. Two Primary Categories of PFAS1.
PI'AS Non-polymers:
Slriiclurnl Klenienls:
r.xnmplc PI' AS l-;imilies:
Perfluoroalkyl
Substances
Compounds in which all carbon-hydrogen
bonds, except those on the functional
group, are replaced with carbon-fluorine
bonds
Perfluoroalkyl acids,
perfluoroalkane sulfonamides,
perfluoroalkane sulfonyl
fluorides
Polyfluoroalkyl
Substances
Compounds in which all carbon-hydrogen
bonds on at least one carbon (but not all)
are replaced with carbon-fluorine bonds
Polyfluoroalkane sulfonamido
derivatives, semifluorinated n-
alkanes and alkenes
PI'AS Polymers:
Slriiclurnl Kleinenls:
l.xnmple PI'AS I'nmilics:
Fluoropolymers
Carbon only polymer backbone with
fluorines directly attached
Polyetrafluoroethylene,
polva inylidene fluoride
Perfluoropolyethers
Carbon and oxygen polymer backbone
with fluorines directly attached
F-(CmF2niO-)nCF3, where the
CmF2mO represents -CF2O, -
CF2CF2O, and or -
CF(CF.i)CF:() distributed
randomly along polymer
backbone
Side-chain fluorinated
polymers
Non-fluorinated polymer backbone with
fluorinated side chains with \ariable
composition
Fluorinated acrylate and
methacrylate polymers,
ll norinated urethane polymers,
and fluorinated oxetane
polymers
1 Modified from P.ucK el ;i1 ('2011 >
PFOA belongs to the peril uoroalkyl acids (PFAAs) of the non-polymer perfluoroalkyl
substances category of PFAS (Table 1-2) PIAAs are among the most researched PFAS (Wang
etal. Z<)|7) TIk- family PI-'AAs includes periluoroalkyl carboxylic, sulfonic, sulfinic,
phosphonic. and phosphinic acids (Table 1-3). PFAAs are highly persistent and are frequently
found in the en\ ii onment ( Alli ens 2011; Wang et al. 2017). PFAAs may dissociate to their
anions in aqueous en\ iionmenlal media, soils, or sediments depending on their acid strength
(pKa value). Although the protonated and anionic forms may have different physiochemical
properties the anionic form is the dominant form in the aquatic environment, including in the
toxicity tests used to derive the PFOA criteria.
9
-------
Table 1-3. Classification and Chemical Structure of Perfluoroalkyl Acids (PFAAs).1
Classification
I'linclional (.roup
1'. XII 111 |)lc
Perfluoroalkyl carboxylic acids
(PFCAs)
Or
Perfluoroalkyl carboxylates (PFCAs)
-COOH
Perfluorooctanoic acid (PFOA)2
-COO"
Perfluorooctanoate (PFO)
Perfluoroalkane sulfonic acids (PFSAs)
Or
Perfluorokane sulfonates (PFSAs)
-SOsH
Pci lluorooclane sulfonic acid (PFOS)
-SO3-
Perfluorooclanc sulfonate (PFOS)
Perfluoroalkyl sulfinic acids (PFSIAs)
-SO2H
Perfluorooctanc sulfinic acid (PFOSI)
Perfluoroalkyl phosphonic acids
(PFPAs)
-P(=0)(OH)2
Perfluorooctyl phosphonic acid (C8-
PFPA)
Perfluoroalkyl phosphinic acids
(PFPIAs)
-P(=OyOH)(Cn,F2n,ll)
Bis(perfluorooctyl) phosphinic acid
(CX C8-PFPIA)
Perfluoroalkylether carboxylic acids
(PFECAs)
CF3(OCF2>,COO
periluoro (3,5,7-trioxaoctanoic) acid
Perfluoroalkylether sulfonic acids
(PFESAs)
CF3 (OCF2)n S O3H
(v2 Cl-PFESA
Perfluoroalkyl dicarbow lie aeids
(PFdiCAs)
1 IOOC-CnF2n-COOII
9:3 Fluorotelomer betaine
Perfluoroalkane disulfonic acids
(PFdiSAs)
1103S-CI1F211-S03H
Perfluoro-l,4-disulfonic acid
Modified from P.uck el ;il (2(>| 11 ;md OliCI) (2021)
2: At most em iroiinieiilalls rele\ ;inl pi I conditions. PI'OA occurs in the anionic form.
Pu lluoroalkx I ciirhivxylie acicls (PFCAs), including PFOA, consist of a general chemical
structure of iCOOH This chemical structure makes PFOA (see Figure 1-1) extremely
strong and slaMc. and rcsislanl to hydrolysis, photolysis, microbial degradation, and metabolism
(Ahrens 2011; Beach el al. 2006; Buck et al. 2011).
10
-------
F
HO
Figure 1-1. Chemical Structure of the Linear Isomer of Perfluorooctanoic acid (PFOA).
(Source: United States EPA Chemistry Dashboard; https://comptox.epa.gov/dashboard).
1.2.1 Physical and Chemical Properties of PFOA
Physical and chemical properties along with other reference information for PFOA are
provided in Table 1-4. These physical and chemical properties helped to define the
environmental fate and transport of PFOA in the aquatic environment. In the environment,
PFOA rapidly ionizes in water to its anionic form (perfluorooctanoate, PFO). PFOA is highly
stable and is resistant to hydrolysis, photolysis, volatilization, and biodegradation (UNEP 2015).
11
-------
Table 1-4. Chemical and Physical Properties of PFOA.
Properly
PI-OA. acidic form1
Source
Chemical
Abstracts Service
Registry Number
(CASRN)
335-67-1
Chemical
Abstracts Index
Name
2,2,3,3,4,4,5,5,6,6,7,7,8,8,8-
pentadecafluorooctanoic acid
Synonyms
PFOA; Pentadecafluoro-1-
octanoic acid; Pentadecafluoro-n-
octanoic acid; Octanoic acid,
pentadecafluoro-;
Perfluorocaprylic acid;
Pentadecafluorooctanoic acid;
Perfluoroheptanecarboxylic acid;
Chemical
Formula
C8HF15O2
Molecular Weight
(grams per mole
[g/moll)
414.07
PubChem Identifier (CID 9554) (URL:
https://pubeliem.ncbi.nlm.nih.gov/compound/9554);
I.ide (2007)
Color/Physical
State
White powder (ammonia salt)
PuK'hem Identifier (CID 9554) (URL:
hllns://Dubchem.ncbi.nlm.nih.gov/comDOund/9554)
Boiling Point
1Q2 4 °C
HSDB (2012); Lide (2007); SRC (2016)
Melting Point
54 3 T
HSDB (2012); Lide (2007); SRC (2016)
Vapor Pressure
i) 525 111111 1 lu al 25 ('
(measured)
1") 0r>2 111111 1 lg al 5l> 25 ('
(111 ciisn reel)
Hekster et al. (2003); HSDB (2012); SRC (2016)
AT SDR (2015); Kaiser et al. (2005)
Kaw
0.0() 102 (e\periineiilii 11 \
determined. et|Lii\iilenl lo
Henry's Law Constant of
0.000028 Pa-111 Vmol at 25 °C)
Li et al. (2007)
Kow
Nol measurable
UNEP (2015)
Organic carbon
water partitioning
coefficient (Koc)
2.00
Higgins and Luthy (2006)
pKa
3.15 (mean measured)
Burns et al. (2008) and 3M (2003) as reported in
EPA Chemistry Dashboard (URL:
https://comptox.epa.gov/dashboard/dsstoxdb/results
?search=DTXSID8031865#DroDerties )
Solubility in
Water
9,500 mg/L (estimated);
3,300 mg/L at 25 °C (measured)
Hekster et al. (2003);
Inoue et al. (2012)
12
-------
Properly
PI-OA. noidie form1
Source
Half-Life in
Water
Stable
UNEP (2015)
Half-Life in Air
Stable
UNEP (2015)
1 :PFOA is most commonly produced as an ammonia salt (CASRN 2795-39-3). Properties specific to the salt are not
included.
PFOA is water soluble, nonvolatile, and stable, with a low vapor pressure and is a solid at
room temperature (UNEP 2015). EPA's chemistry dashboard reported a mean experimental acid
dissociation constant (pKa) for PFOA of 3.15 calculated from pka values determined from
Burns et al. (2008) and 3M (2003). Burns et al. (2008) measured an acid dissociation constant
(pKa) for PFOA of 3.8 using a standard water-methanol mixed solvent approach, u hich indicates
PFOA is a moderate acid, while 3M Company (2003) reported a measured PFOA pKa of 2.5.
Due to the surfactant properties of PI-OA. it forms three layers when added to octanol and
water in a standard test system used to measure a n-octanol-uater partition co-efficient (Kow),
thus preventing direct measurement (EFSA 2<)()K. (iiesy et al 2010). Although a Kow cannot be
directly measured, a k< >\\ lor PI-OA has been estimated from its individual water and octanol
solubilities (estimated PI OA kow range 2 oO 6.3; UNEP 2015); however, the veracity of
such estimates is uncertain (I Al-P 2d I 5) T.acking a reliable Kow for PFOA precludes
application of T\< >w-based models commonly used to estimate various physiochemical properties
for organic compounds, includi tig bioconcentration factors and soil adsorption coefficients.
Further, the unusual characteristics of PFOA would bring into question the use of Kow as a
predictor of environmental behavior; for example, bioaccumulation of PFOA is thought to be
mediated via binding to proteins rather than partitioning into lipids (EFSA 2008; Giesy et al.
2010), the latter being the theoretical basis for Kow-based prediction of bioaccumulation.
13
-------
2 PROBLEM FORMULATION
A problem formulation provides a strategic framework for water quality criteria
development under the CWA by focusing on the most relevant chemical properties and
endpoints. In the problem formulation, the purpose of the assessment is stated, the problem is
defined, and a plan for analyzing and characterizing risk is developed. The structure of this
problem formulation was consistent with EPA's Guidelines for Ixolouical Risk Assessment
(U.S. EPA 1998).
2.1 Overview of PFOA Sources
2.1.1 Manufacturing of PFOA
PFOA is primarily produced through Electrochemical Huorination (E(T) in which an
organic raw material, in the case of PFOA as oclanoyl fluoride (OlIisCOF), undergoes
electrolysis in anhydrous hydrogen fluoride solution This electrolysis leads to the replacement
of all the hydrogen atoms by fluorine atoms and results in peril uorooctanoyl fluoride
(C7F15COF), which is the major raw material used to manufacture PFOA and PFOA salts (Figure
2-1; Buck et al. 20] I) l-leclrochemical I'liioiination typically results in a mixture of branched
and linear isomers
14
-------
CyHisCOCl
C7H15COF
(Octanoyl Fluoride)
HF, e
O
o
o
-b
W
O
"03
S-H
O
2
HF. e
1
C7F15COF
(Perfluorooctanoyl
Fluoride)
Treatment with Strong
Acid and Base
C?1 l<()2"M-
(PFOA Salts)
Figure 2-1. Synthesis of Perfluorooetaiioir sir id (I'L-'OA) hy Kleetrorhemieal Fluorination
(ECF).
Modified from Buck el a I (2' 11 I)
Initial production of PI OA started in the 1940s and commercial production and use as
protecti\ e coatings starting in the mid-1950s. From 1951 - 2004 the total global historic industry
wide emission of PFCAs (including PFOA) from all sources (i.e., direct and indirect sources
such as manufacture, use, consumer products, and PFCA precursors) ranged from 3200 tons to
7300 tons (Prevedouros et al. 2006). In 2006, EPA initiated the 2010/2015 PFOA Stewardship
Program, resulting in major PFOA producers committing to a 95% reduction in PFOA facility
emissions and product contents across the globe by 2010. The 2010/2015 PFOA Stewardship
Program further aimed to eliminate PFOA emissions and product content by 2015 (U.S. EPA
2006).
15
-------
2.1.2 Sources of PFOA to Aquatic Environments
PFCAs, including PFOA are primarily released into water (95% of PFCAs are emitted to
water; 3M Company 2000) and can enter the aquatic environment from both industrial and
consumer products during manufacturing, along the supply chains, during product use and/or
disposal (Ahrens and Bundschuh 2014; Kannan 2011). Occurrence of PFOA in the aquatic
environment arises from both direct and indirect sources (Ahrens et al. 2011). However, the
quantitative assessments of their production, direct and indirect emissions, and environmental
measurements are lacking (Ahrens and Bundschuh 2014; Prevedouros et al. 2006).
The direct sources of PFOA to the aquatic en\ ironment include both municipal and
industrial wastewater treatment plants (WWTPs), landfill Icachate, and runoff from contaminated
biosolids (Renner 2009). WWTPs in particular arc an important source of PFOA to aquatic
systems (Ahrens el al 2<)i)1))
Indirect sources of PI OA to aqualic cn\ ironments include dry and wet atmospheric
deposition, runoff from contaminated soils, and consumer product use and disposal (Kannan
2011). Identification of indirect sources of PI OA and understanding their relative contribution to
aqualic ecosystems is difficult ()\ erall. the presence of indirect sources of PFOA and their
contributions are dependent on the system and the nearby land uses. Overall PFAS
concentrations, including PI OA, in the environment are positively correlated with population
density. Overall. PI OA occurrence in aquatic environments is driven by legacy PFOA sources
because use of PFOA in the United States was largely phased out by 2010, and completely
phased out by 2015 in accordance with EPA's 2010/2015 PFOA Stewardship Program.
In addition to direct discharge, environmental breakdown of precursor compounds
containing a seven-member perfluoro moiety can provide an additional source of PFOA.
Metabolic transformation of PFAS precursors, such as PAPs, FTCAs, FTUCAs, FTSAs, and
16
-------
FASAAs, and the degradation of volatile PFAS, such as FTOHs, FASAs, and FASEs, can be
potential sources of PFOA as these compounds can transform into more persistent PFAS,
including PFCAs and PFOA (Ahrens and Bundschuh 2014). For example, fluoroacrylate
polymers can breakdown in soil and release fluorotelomer alcohol (FTOH) which can further
degrade into PFOA (Russel et al. 2008). Similarly, polyfluoroalkyl phosphoric acid diesters
(diPAPs) are used in commercial applications, such as food packaging, and can be found in
WWTP sludge and contaminated biosolids. In environmental media, dilWI's can release FTOH
that further degrades into PFOA (Lee et al. 2010; Sinclair and Kannan 2000; Washington et al.
2009). Current understanding of these transformation processes remains limited, and additional
work is needed to fully understand these processes and their role in generation of sources of
PFOA to aquatic environments (Lau et al. 2007).
PFOA can also be re-emitted to the aquatic cn\ iionment from ice melt and sediment
transport. Release of PI-OA will continue into the future from the transformation of other PFAS
and the historical pi'odncls still in use (eg, consumer goods manufactured and/or obtained
before the PFOA discontinuation)
2.2 Environmental Fate and Transport of PFOA in the Aquatic Environment
2.2.1 I ¦ n\ ironmental Fate of PFOA in the Aquatic Environment
In natural waters near neutral pH, PFOA rapidly dissociates into ionic components. In
aquatic environments. PI-OA has an affinity to remain in the water column rather than sediments,
but can also adsorb to sediments in the presence of organic carbon, with the partitioning
coefficients (Kd) increasing with salinity (Environment Canada 2012; Hekster et al. 2003).
Because of its water solubility and preferential binding to proteins, once PFOA enters a
waterbody it tends to remain dissolved in the water column, where it is mobile, unless it adsorbs
to organic particulate matter or is assimilated by organisms.
17
-------
PFOA has low volatility in the ionized form but can adsorb to particles in air where it can
be transported globally, including remote locations (Del Vento et al. 2012; Shoeib et al. 2006).
PFOA is water soluble and has been found in surface water, ground water, and drinking water.
Because of the relatively low log Koc of PFOA, it does not easily adsorb to sediments and tends
to stay in the water column.
In the water column, and other environmental compartments. PFOA is stable and
resistant to hydrolysis, photolysis, volatilization, and biodegradation (I liuuins and Luthy 2006;
UNEP 2015). The persistence of PFOA has been attributed to the strong carbon-lluorine (C-F)
bond. Additionally, there are limited indications that naturally occurring defluorinating enzymes
exist that can break a C-F bond. Consequently, no biodegradation or abiotic degradation
processes for PFOA are known. In aquatic en\ ironments, the only dissipation mechanisms for
PFOA are physical mechanisms, such as em ironmenlal dilution and sorption.
2.2.2 Environmental Transport of PI-OA in the Aquatic Ijnironment
The physiochemical properties discussed in Section 1.2.1 above enable PFOA to be
highly persistent in the aquatic en\ ironment PFOA tends to be distributed in waters and in the
atmosphere ( Alliens 2<)| I. Yamashita et al 2008). PFOA concentrations in seawater are
typically greater than PI OS. u hich has been attributed to the relatively lower bioaccumulation
potential, lower sorption to sediments, and greater water solubility (Ahrens 2011; Ahrens et al.
2009).
Numerous uncertainties exist in the understanding of environmental transport of PFOA in
aquatic systems. Both point and non-point sources contribute PFOA to the aquatic environment.
PFOA can be transported from these sources into rivers, streams, lakes, and marine
environments. There is a general decrease in PFOA concentrations along a transport pathway
resulting from dilution in the water column. For example, measured PFOA concentrations in
18
-------
WWTP effluents were generally an order of magnitude greater than riverine concentrations, with
the upper end of riverine concentrations being similar to WWTP effluents. Although minimum
and maximum PFOA concentrations in coastal waters (ranging from hundreds of pg/L to several
ng/L) were below corresponding measurements in riverine systems, coastal PFOA
concentrations in general were largely similar to riverine concentrations. Open oceans contained
the lowest PFOA concentrations resulting from immense dilution ()\ erall, open ocean
concentrations of PFOA were roughly 2.5 orders of magnitude lower lhaii those reported in
WWTP effluents (Ahrens 2011).
PFOA has been found in a diversity of en\ ironmenls. including in the arctic and
Antarctic, despite the limited number of manufacturing facilities and/or small population sizes
typically found in these areas (Del Vento el al 2<>12: Schoeih el al 2006). Although PFOA has
low volatility, particularly in the ionized form, it can ahstx h to air particles before being
deposited via atmospheric deposition to these remote regions. For example, Kim and Kannan
(2007) reported PFOA in snow in the I nited States ranging from below the limit of
quantification to 2d ng I. Similarly. Young el al (2007) reported mean PFOA concentrations in
snow in Canada ranging from
-------
above). Polyfluoroalkyl substances are one type of precursor substance that have the potential to
be transformed abiotically or biotically into PFCAs such as PFOA (Buck et al. 2011). On a
global scale, production volumes of polyfluoroalkyl substances, many of which are likely
polyfluoroalkyl precursor substances that ultimately degrade or transform to PFOA, greatly
exceed direct emissions of PFOA through its manufacture, use and disposal (Butt et al. 2014; Liu
and Mejia Avendano 2013). According to the OECD (2006), there were approximately one
thousand polyfluorylalkyl chemicals commercially produced at the time thai could conceivably
degrade to PFCAs such as PFOA. For example, Buck et al. (201 1) identified 42 families of
compounds and numerous individual PFAS detected in environmental and human matrices,
many of which have not been evaluated for their biodegradability (Liu and Mejia Avendano
2013). Any or all members of these PFAS ha\ e the ability to be Hans formed or degraded to
PFAAs such as PFCAs or perfluoroalkane sulfonic acids (PI'SAs).
Critical ie\ iews In Ikill el al. (2014) and I.in and Mejia Avendano (2013) provided a
comprehensive summary of I lie <.|ualilalive and quantitative relationships between biodegradation
and transformation of polylluoroalkyl precursors and generation of PFOA and other PFCAs. The
most ucll-sludicd polylluoroalkyl precursor substances are fluorotelomer-based compounds,
which are produced through lelomerization technology and are associated with PFOA as the final
product (Buck el al 2<)| I)
2.3.1 Biodegradation of lluorotelomer-based precursors
The aerobic biodegradation pathway of fluorotelomer alcohols (8:2 and 6:2 FTOH) have
been thoroughly studied. Dinglasan et al. (2004) was among the first to investigate the
biotransformation of 8:2 FTOH in a mixed microbial culture. Additional studies of the aerobic
microbial degradation of 8:2 FTOH by Liu et al. (2010) and Wang et al. (2005, 2009, and 2012)
have since confirmed the formation of PFOA via this pathway. The observed half-lives of 8:2
20
-------
FTOH ranged from <2 days to 30 days in these laboratory studies. Molar yield of PFOA ranged
anywhere from 0.5 to 40% depending on type of microbes or microcosm used in the study, with
Wang et al. (2009) observing relatively higher PFOA yield in aerobic soils relative to PFOA
yield in pure bacterial culture (Liu et al. 2010). Thus, aerobic microbial degradation of 8:2 FTOH
can be a significant source of PFOA in some environmental compartments. Anaerobic microbial
degradation of 8:2 FTOH, on the other hand, is inefficient and likely an insignificant source of
PFOA to the environment (Zhang et al. 2013b). Additional studies are needed, however, to
understand anaerobic biodegradability of FTOHs and related compounds in general (Liu and
Mejia Avendano 2013).
Aerobic biodegradation of several other types of fluoroielomer-based, polyfluoroalkyl
precursor substances definitively linked 1o PI-OA formation include fluorotelomer stearate (8:2
FTS) with observed half-life in aerobic soils of 5-2S days and molar yield of about 01.7-4%
(Dasu et al. 2012, 2<)| 3). lluorotelomer acrylale (S 2 FT AC) and fluorotelomer methacrylate (8:2
FTMAC) monomers with oltsei\ ed hall-life in aerobic soils of 3-5 days and 15 days and molar
yields of 7.8 and I <>"<>. respecli\ ely (Rover 2<)| I). fluorotelomer ethoxylates (FTEOs) with
obser\ ed liall-lile in unlllleied WW IP eflluent of approximately one day and molar yield of
about ii 3".. (I'romel and Knepper 2oi180 days and molar yield of
0.9% (Dasu 2011)
2.3.2 Biodegradation of side-chain polymers
At present, a crucial need exists to understand the potential degradation of side-chain
fluorinated polymers in natural environments because they currently represent a high percentage
of all commercial and industrial PFAS sales products (Liu and Mejia Avendano 2013). Side-
chain polymers are those with polyfluoroalkyl or perfluoroalkyl chains attached to non-
21
-------
fluorinated backbones (Buck et al. 2011). Russell et al. (2008) investigated the biodegradation of
a high molecular weight (-40,000 amu, 100-300 nm in diameter) polyacrylate polymer aqueous
dispersion product in four aerobic soils over two years. Two approaches (molar mass balance
and kinetic modeling) gave conflicting results. The molar mass balance approach indicated no
evidence of biodegradation because the PFOA generated was mostly accounted for by impurity
(residual non-polymerized PFAS) degradation. Conversely, the kinetic modeling approach,
estimated half-lives of PFOA to be around 1,200-1,700 years among the lour soils tested. Upon
further investigation using a low molecular weight (-3,500 amu) polyurethane polymer product
and a similar approach, Russel et al. (2010) clearly demonstrated biodegradabilily of the low
molecular weight polyurethane polymer product compared to the polyacrylate polymer, as the
levels of PFOA produced were several orders of magnitude greater than what the impurities
could account for. Applying a similar kinetic modeling approach, the half-lives of the
polyurethane polymer were estimated to range from 28 to 241 years among the four test soils.
Given the large disparity in half-life prediction between the two studies, however, additional
research is needed to clarify the contributions of polyfluoroalkyl polymers to PFOA formation
due to the high percentage of side-chain fluorinated polymers that exist in commercial and
industrial sales products.
2.3.3 Biodegradation of other polyfluoroalkyl substances
Recently, Mcjia-.\ \ endano et al. (2016) examined the formation of PFOA from aerobic
biotransformation of quaternary ammonium polyfluoroalkyl surfactants (QACs). Capitalizing on
several recent studies focused on identifying specific PFAS in major PFAS-based aqueous film-
forming foam (AFFF) formulations, all the newly identified PFAS were polyfluoroalkyl
compounds. These compounds have perfluoroalkyl carbon chain lengths varying from four to 12
and possess functionalities such as sulfonyl, thioether, tertiary amine, quaternary ammonium,
22
-------
carboxylate, sulfonate, amine oxide, and betaine, etc. (Mejia-Avendano et al. 2016). Importantly,
the identified cationic PFAS in these studies contain either tertiary amine or quaternary
ammonium groups. In this first study of the fate of polyfluoroalkyl cationic surfactants used in
aqueous AFFF formulations, the biotransformation of perfluorooctaneamido quaternary
ammonium salt (PFOAAmS) was characterized by a DT50 value (time necessary to consume
half of the initial mass) of 142 days and significant generation of peril uoroalkyl carboxylic acid
(PFOA) at a yield of 30% (mol) by day 180. Three novel biotransformation intermediates were
identified for PFOAAmS, and it was demonstrated lluil despite o\ erall high sliihilily of QACs
and their biocide nature, the ones with perfluoroalkyl chains can he substantially biotransformed
into perfluoroalkyl acids in aerobic soil.
The above microbial biotransformation and degradation pathways are all dependent on
environmental conditions, degradation kinetics, and the chemical structures and properties of the
individual polyfluoroalkyl precursors (Buck et al 2<>11. liutt et al. 2014; Liu and Mejia
Avendano 2013). Of particular importance is the environmental stability of key chemical
linkages (such as esters and ethers) as the stability of these chemical linkages determines the
stability of the o\ em 11 PI'AS (l.iu and Mejia Avendano 2013). It is evident through these studies
that the biotransformation and biodeuradability of polyfluoroalkyl precursor substances is due to
the breakdow n of the non-lluorinated functionality of the precursor substances, which precedes
the breakdown of the periluorinated carbons. In contrast, perfluoroalkyl chemicals in general
resist biotransformation and defluorination under natural conditions. Using 14C-labeled PFOA to
examine five different microbial communities, a range of electron donors for reductive
defluorination processes, and the possibility of co-metabolism during reductive dechlorination of
23
-------
trichloroethene, Liou et al. (2010) was able to confirm that PFOA is highly resistant to microbial
degradation in natural environments.
2.3.4 Non-microbial biodegradation of other polyfluoroalkyl substances
Butt et al. (2014) reviewed the current state of knowledge regarding the
biotransformation of fluorotelomer-based, polyfluoroalkyl precursor substances that degrade to
form PFCAs (PFOA) in microbial systems, rats, mice, and fish Consistent with information
presented above, the majority of biotransformation studies thus far used N 2 FTOH (a
fluorotelomer alcohol) as the substrate; only a few studies of non-FTOH biotransformation exist.
The biotransformation studies of 8:2 FTOH metabolism universally show the formation of
PFOA. As above, the overall yield of PFOA is low, presumably because of the multiple branches
in the biotransformation pathways, including conjugation reactions in animal systems which are
capable of phase II metabolism and results in the formation of conjugated metabolites such as
glucuronide, sulfate, and glutathione metabolites IJutt et al (2
-------
2.4.1 PFOA Occurrence and Detection in Ambient Surface Waters
PFOA is one of the dominant PFAS detected in ambient surface waters, along with PFOS
(Ahrens 2011; Benskin et al. 2012; Dinglasan-Panlilio et al. 2014; Nakayama et al. 2007;
Remucal 2019; Zareitalabad et al. 2013). Most of the current, published PFOA occurrence
studies have focused on a handful of broad geographic regions, many times targeting sites with
known manufacturing or industrial uses of PFAS, such as the (iron l I .a kes, the Cape Fear River
and waterbodies near Decatur, Alabama (Figure 2-2. Boukinuer el ill 2<)i)4. Cochran 2015;
Hansen et al. 2002; Konwick et al. 2008; Nakayama el al 2<)()7. ,i\[ Company 2<)nl).
25
-------
Figure 2-2. Map Indicating Sampling Locations for Perfluorooctanoic acid (PFOA)
Measured in Surface Waters Across the United States (U.S.) Based on Data Reported in the
Publicly Available Literature.
Colorado sampling coordinates were not available, these data are represented by the dash marks to
indicate measured PFOA surface water concentrations are available in Colorado.
Concentrations of PFOA in surface waters vary wi dely (Figure 2-3), with observed
concentrations ranging over seven orders of magnitude and detected generally between pg and
ng per liter with some sites with reported concentrations in ug/L (Zareitalabad et al. 2013). For
the purposes of this overview and comparison, all concentrations reported here are in ng per liter
(ng/L). Unlike other contaminants commonly found in aquatic ecosystems, PFAS are synthetic
compounds and therefore have no natural source. Thus, the occurrence of any PFAS in the
environment is an indication of anthropogenic sources, including consumer and industrial use,
26
-------
long-range transport, atmospheric deposition, surface water runoff, and general persistence in the
environment.
10000
1000
^ 100
<
2 io
Oi
53
O
0.1
0.01
0.001
Figure 2-3. Distribution of the minimum and maximum concentrations (ng/L) of
Perfluorooctanoic acid (PFOA) measured in surface waters for each state or waterbody
(excluding the Great Lakes) with reported data in the publicly available literature.
The distribution is arranged alphabetically by state and waterbody.
PFOA concentrations in surface water tend to increase with levels of urbanization.
Across the Great Lakes region, PFOA was higher in the downstream lakes of Erie and Ontario
and lower in the upstream lakes of Superior, Michigan, and Huron (Remucal 2019). Similarly,
Zhang et al. (2016) observed measured PFOA concentrations in urban areas (urban average
PFOA concentration = 10.17 ng/L; n = 20) to be more than three time greater than concentrations
in rural areas (rural average PFOA concentration = 2.95 ng/L; n = 17) within New Jersey, New
York, and Rhode Island. Temporal variation of PFOA in surface waters remains largely
unknown due to data limitations. See Appendix N for further discussion of PFOA occurrence in
surface waters and other abiotic media such as aquatic sediments, groundwater, air, and ice.
27
¦
n |—i
U «
~ 1
¦
AL CA CO DE FL GA LA MI MN NJ NM NY NC EI SC IN H WA
River
-------
2.5 Bioaccumulation and Biomagnification of PFOA in Aquatic Ecosystems
PFOA is found in aquatic ecosystems around the globe (e.g., Ankley et al. 2020; Giesy
and Kannan 2001; Houde et al. 2008). Although they were used predominantly in more
populated areas, these compounds are resistant to hydrolysis, photolysis, and biodegradation,
which facilitates their long-range transport to aquatic ecosystems in the remote arctic and mid-
oceanic islands (Haukas et al. 2007; Houde et al. 2006). Several physical-chemical properties of
PFAS contribute to their bioaccumulation within aquatic and aquatic-dependent species once
they have entered an aquatic ecosystem.
2.5.1 PFOA Bioaccumulation in Aquatic Life
In contrast to many persistent organic pollutants u hich tend to partition to fats, PFOA
preferentially binds to proteins (Martin et al. 2003a, 2003h) Within the body, PFOA tends to
bioaccumulate within protein-rich tissues, such as the Mood serum proteins, liver, kidney, and
gall bladder (De Silva et al. 2009; Jones et al. 2003; Martin et al. 2003a, 2003b). PFOA may also
bind to ovalbumin, and the transfer of PI ( )A to such albumin in eggs can be an important
mechanism for depuration in female o\ iparous species, as well as a mechanism for maternal
transfer (Jones et al 2<)i)3. Kannan et al 2005).
The stability of PI OA contributes to its bioaccumulation potential, as PFOA has not been
found to undergo biotransformation within the organism and is primarily depurated through
excretion in urine or across gill surfaces (De Silva et al. 2009; Martin et al. 2003a). Within an
organism, PFOA may undergo enterohepatic recirculation, in which PFOA is excreted from the
liver in bile to the small intestine, then reabsorbed and transported back to the liver (Goecke-
Flora and Reo 1996). Among PFAS, this process becomes increasingly more efficient the longer
the perfluorinated chain length, resulting in longer half-lives for chemicals like PFOA with
relatively long chain length, as they are less readily excreted. PFAS with carboxylate head
28
-------
groups, such as PFOA, are less efficiently resorbed by the small intestine and transported back to
the liver than sulfonate PFAS, resulting in lower bioaccumulation levels (Hassell et al. 2020;
Martin et al. 2003b).
Sex differences in the elimination rate of PFOA chemicals have been observed in some
species. Lee and Schultz (2010) observed that the elimination rate of PFOA from blood plasma
was ten times faster in female fathead minnows compared to males The faster elimination rate
may be related to sex hormones (i.e., androgen and estrogen) levels, as the elimination rate in
females decreased four-fold following exposure to the androgen trenbolone (I .ee and Schultz
2010). This pattern has also been observed in rals. where the elimination ofPFOA \\as70 times
faster in females than males, and was attributed to sex-related differences in the expression of
organic anion transporters in kidneys resulting in higher excretion rates (Kudo et al. 2002). The
mechanism for the higher elimination rate in female fathead minnows has not been determined,
and the degree to w liich gender-related differences in elimination rate apply to other fish species,
or other taxonomic groups, is unknou n I louever, it does suggest that the sex of the organism
should be considered u hen assessing ecosystem level bioaccumulation, and that there may be
another mechanism in addition to egg production that can result in lower concentration of PFAS
in females
The structure ofPFOA also contributes to its bioaccumulation potential, with linear
forms being more Moaccumulative than branched forms (De Silva et al. 2009; Hassell et al.
2020). The preferential accumulation of linear PFOA occurs because the elimination rate, of
branched isomers ofPFOA is higher, particularly across gill surfaces (De Silva et al. 2009). This
pattern has also been observed in the field, as the proportion of branched isomers was higher in
29
-------
water and sediment compared to fish tissue in Taihu Lake, China (Fang et al. 2014) and Lake
Ontario (Houde et al. 2008).
2.5.2 Factors Influencing Potential for PFOA Bioaccumulation and Biomagnification in
Aquatic Ecosystems
PFOA binding to the surface of sediment organic matter and biofilms is influenced by
both hydrophobic and electrostatic effects, resulting from the hydrophobicity of the
perfluorinated chain and the hydrophilicity of the carboxylatc head groups (Higgins and Luthy
2006). In a series of laboratory studies, Higgins and Luthy (2006) demonstrated that PFOA
sorption to sediments increased with increasing organic content, increasing calcium ions, and
decreasing pH. The strongest effect was observed in response to increasing organic content,
demonstrating the importance of hydrophobic effects, while the increased sorption in response to
calcium ions and decreasing pH demonstrated the role of electrostatic effects (Higgins and Luthy
2006). Across all PFAS, sorption to sediments increased with increasing perfluorinated chain
length, and for a gi\ en chain length. PI AS such as PFOS, had approximately 1.7 times the
sorption capacily as peilluoroalkyl cait>o\ylie acids (PFCA) such as PFOA (Higgins and Luthy
2006). The capacity of PI OA to bind to particulate matter increases with increasing salinity.
Jeon et al (2010) obser\ ed that water column PI OA partitioned more readily to particulate
organic matter as salinity increased from 10 to 34 ppt, resulting in increased uptake of PFOA in
Pacific oysters (('rassostrca ^i^as). In a recent review, Li et al. (2018) found no single parameter
strongly predicted PFOA sorption to sediments and Li et al. (2019) reported that the protein
content of soil was a better predictor of sorption than organic carbon. Overall, these results
suggest that sorption to sediments should be an important mechanism for PFOA entry into an
aquatic ecosystem.
30
-------
Evidence of the PFOA sediment pathway in aquatic ecosystems, although mixed, overall
demonstrates the importance of bioaccumulation from sediments and biofilms via diet into
aquatic invertebrates. In laboratory studies, PFOA concentrations in sediment were positively
correlated to PFOA tissue concentrations in Lumbriculus variegatus (Lasier et al. 2011), but not
for Chironomusplumosus (Wen et al. 2016) or the amphipods Gammarus fossarum and G. pulex
(Bertin et al. 2016). In field studies PFOA concentrations were posiii\ ely correlated between
sediments and biofilms and benthic feeding organisms (Lescord et al 2<> 15: T.oi et al. 2011;
Martin et al. 2004; Penland et al. 2020). In addition, the distribution of PI AS in sediments was
more similar to their distribution in the tissues of benthic invertebrates (Lescord el al 2015) and
benthic-feeding fish (Thompson et al. 2011) than they were to their distribution in pelagic
organisms.
PFOA can also enter aquatic organisms directly from the water column through
respiration. Because of its Mndinu affinity to proteins, PI OA can enter the body of gill-breathing
organisms by binding lo proteins in the Mood at gill surfaces (De Silva et al. 2009; Jones et al.
2003; Martin et al 2<)()3a. 2<)()3|->)
The relath e distribution of PI OA in tissues is related to the primary route of exposure
(dietary or respiratory). In rainbow trout, the rank order of PFOA concentrations following
aqueous exposure was blood kidney>liver (Martin et al. 2003b). In contrast, their rank order
following dietary exposure was liver>blood>kidney (Goeritz et al. 2013). Hong et al. (2015)
observed the highest concentrations of PFAS in the intestines of green eel goby; soft tissues,
shell, and legs of shore crabs; and gills and intestines of oysters, suggesting bioaccumulation
through both dietary and aqueous uptake in invertebrates, and primarily dietary uptake in fish.
31
-------
Data from multiple field studies suggest trophic biomagnification potential of PFOA is
low, and is often not observed, particularly with respect to aquatic organisms. In a review of
PFOA and PFOS concentration data across major taxonomic groups, Ahrens and Bundschuh
(2014) found that maximum PFOA and PFOS concentrations were similar for invertebrates, but
that maximum PFOS concentrations in fish were nearly an order of magnitude greater than
PFOA, and several orders of magnitude greater for aquatic-dependent birds and mammals.
When individual aquatic species pairs were considered, biomagnification factors (BMF)
greater than one, indicating biomagnification, have been observed for PFOA (eg. Tang et al.
2014; Penland et al. 2020; Tomy et al. 2009), suggesting trophic biomagnification I lowever,
when ecosystem-level biomagnification is assessed using trophic biomagnification factors
(TMF), which measures the change in the concentration of a chemical per trophic level within a
food web, PFOA is nearly always shown not to hiomaunily (I ,oi et al. 2011; Martin et al. 2004;
Tomy et al. 2004; \ii et al 2" 14. Zhou et al. 2d 12). unless aquatic-dependent species, such as
aquatic-dependent hi ids. are included in the food web model (Houde et al. 2006b; Kelly et al.
2009; Tomy et al 2<)i)1)) The o\ erall lack of biomagnification in PFOA relative to PFOS is
attributed to its ph\ sical-chemical properties, including a shorter perfluorinated chain length and
the carhow late head group, both of u hich are associated with less efficient assimilation into
tissues and faster excretion rates (e.g., Martin et al. 2003a, 2003b).
2.5.3 Environmental Monitoring of PFOA in Biotic Media
Generally, PFOA is one of the dominant PFAS detected in aquatic ecosystems, along
with PFOS (Ahrens 2011; Benskin et al. 2012; Dinglasan-Panlilio et al. 2014; Nakayama et al.
2007; Remucal 2019; Zareitalabad et al. 2013). PFAS were first detected in human serum
samples in the late 1960s, and subsequent studies across several continents demonstrated the
global distribution of PFAS in humans (Giesy and Kannan 2001; Houde et al. 2006a). Since
32
-------
then, the global distribution of PFAS in tissues of aquatic and aquatic-dependent species has
been demonstrated in studies conducted in freshwater and marine environments across every
continent, including remote regions far from direct sources, such as the high arctic, Antarctica,
and oceanic islands (Giesy and Kannan 2001; Houde et al. 2006a).
In lentic surface waters of the United States, one of the most comprehensive studies of
PFOA concentrations included fish muscle tissue data from 157 near shore sites across the Great
Lakes selected following a probabilistic design as part of the 2010 National Coastal Condition
Assessment (Stahl et al. 2014). In this study, PFOA was measured in fish collected at 12% of the
sites, with a 90th centile concentration of 0.16 ng g wet weight (ww), and a maximum
concentration of 0.97 ng/g ww (Stahl et al. 2014). Lake trout (3 1% of samples), smallmouth bass
(14%), and walleye (13%) were the most commonly sampled species from the Great Lakes
samples.
Martinet al (2< >( »4) measured PFOA in whole body samples of invertebrates and fish in
Lake Ontario, near the tow n of Niauara-on-the-Lake. PFOA concentrations were much higher in
the benthic amphipod / hporciu hoya (l)<) ng g w w) than in the more pelagic Mysis relicta (2.5
ng/g w w ). suggesting sediments are an important source of PFOA in this area (Martin et al.
2004) Among the four fish species sampled, PFOA concentrations were highest in the slimy
sculpin (44 ng g w w ), which feeds onM relicta and D. hoya. Although lake trout occupy the
highest trophic le\ el at this site, their PFOA concentrations were the lowest of all sampled fish
species (1.0 ng/g ww) (Martin et al 2004). PFOA concentrations were lower in lake trout than in
alewife (1.6 ng/g ww), which comprise 90% of the lake trout diet, suggesting a lack of PFOA
biomagnification in this system (Martin et al. 2004).
33
-------
Guo et al. (2012) measured PFOA in lake trout muscle tissues in Canadian waters of
Lakes Ontario, Erie, Huron, and Superior, as well as Lake Nipigon, Ontario. The average PFOA
concentration across all sites was 0.045 ng/g ww and was not significantly different (P<0.1)
across the different lakes (Guo et al. 2012). Finally, Delinsky et al. (2010) sampled bluegill,
black crappie, and pumpkinseed muscle tissues in 59 lakes in Minnesota, including four lakes in
the Minneapolis-St. Paul metropolitan area, and did not detect PI OA in any of the samples (limit
of quantification = 0.77 ng/g ww; see Table 2 of Delinsky et al. 2001))
In flowing surface waters of the United States, one of the most comprehensive studies of
PFOA concentrations included fish muscle tissue data from 164 urban river sites (5lh order or
higher) across the coterminous U.S. selected following a probabilistic design as part of the 2008-
2009 National Rivers and Streams Assessment and the National Coastal Condition Assessment
(Stahl et al. 2014). Largemouth bass (34% of samples), small month bass (25%), and channel
catfish (11%>) were the most commonly sampled species from the urban stream sites (Stahl et al.
2014). PFOA was not detected in any of the urban river sites (Stahl et al. 2014). The lack of
detection may ha\ e been related to the method detection limit of 2.37 ng/g ww, which was
higher than the highest PI 'OA concentration measured in the Great Lakes coastal survey
described aho\ e. which also followed a probabilistic sampling design (Stahl et al. 2014).
In 2()i)5, Ye et al (2<)0S) detected average PFOA concentrations of 0.17 ng/g ww and 0.2
ng/g ww from whole body composite samples of multiple fish species from the Ohio River and
Mississippi River, respectively. PFOA was not detected (<1.0 ng/g ww) in whole body
composite fish samples collected from the Missouri River (Ye et al. 2008). Delinsky et al. (2010)
sampled PFOA in bluegill, black crappie, and pumpkinseed muscle tissues at eleven locations
34
-------
along the upper Mississippi River in 2007, and did not detect it at any location, including the
heavily impacted Pool Two site in the Minneapolis-St. Paul metropolitan area.
In a more recent study, Penland et al. (2020) measured PFAS concentrations in
invertebrates and vertebrates along the Yadkin - Pee Dee River, in North and South Carolina in
2015. PFOA was detected in whole body tissues of unionid mussels (741 ng/g ww) and aquatic
insects (10.68 ng/g ww), but was not detected in Asian clam, snails, or crayfish. PFOA was
measured in muscle tissue of two of the 11 sampled fish species, the channel catfish (21.19 ng/g
ww) and notchlip redhorse (45.66 ng/g ww). PFOA was not delected in the euus of a robust
redhorse sample, which had the highest measured PI- OS concentration (482.9 nu u w w) of any
sample from the Penland et al. (2020) study.
Houde et al. (2006b) measured whole body PI-OA in six fish species in Charleston
Harbor, South Carolina, and whole body PI OA of zooplankton and five fish species in Sarasota
Bay, Florida. Charleston I larbor was the more de\ eloped of the two sites and had higher overall
PFOA concentrations PI'OA was detected in four of the six fish species in Charleston Harbor
and ranged lVoni n 5 nu u w w in spot to I S nu'u ww in spotted seatrout. In Sarasota Bay, PFOA
concentrations a\ eraued <> 3 nu u w w in zoo plankton, and was not detected in any of the fish
species (I londe et al. 2006b)
Overall, these results illustrate the distribution of PFOA in biotic media collected from
invertebrate and fish samples In contrast to PFOS, PFOA concentrations in biotic media are
often low, or below detection levels, highlighting the lower overall bioaccumulation potential for
this chemical, based on its physical-chemical properties, including a shorter perfluorinated chain
length, and a carboxylate head group. In addition, trophic biomagnification is rarely observed
with PFOA, as concentrations in invertebrates are often similar to concentrations in fish.
35
-------
2.6 Exposure Pathways of PFOA in Aquatic Environments
There are multiple potential exposure pathways of PFOA in the aquatic environment,
including: (1) direct aqueous (dermal and respiratory) exposure, (2) direct exposure from
contaminated sediment (for benthic organisms), (3) diet (e.g., bioaccumulation and
biomagnification), and (4) maternal transfer (Ankley et al. 2020). Exposure of PFOA through
water and sediment occurs through direct contact with the respecti\ e media, such as water
passing across the gills, or consumption of suspended and deposited sediments (Prosser et al.
2016). Elevated PFOA concentrations in eggs of fish and piscivorous birds suggests that PFOA
may maternally transfer to offspring. Given these exposure pathways, aquatic organisms, such as
fish and aquatic invertebrates, are exposed to PFOA when it is present in the environment. This
exposure occurs through multiple exposure routes including water, sediment, diet, and maternal
transfer.
2.7 Effects of PI OA on Biota
Currently, PI OA aquatic ecotoxicity data are primarily available for freshwater fish,
aquatic invertebrates, plants, and algae. Section 3 and Section 4 provide study summaries of
individual publicly a\ ailahle ecotoxicity studies, and Appendix A through Appendix H
summarize the current PFOA aquatic lile ecotoxicity data.
2.7.1 Mechanisms of PFOA Toxicity
The mechanisms underpinning the toxicity of PFOA to aquatic organisms, like other
PFAS, is an active and on-going area of research. Much work is still needed from a mechanistic
perspective to better understand how the different modes of action elicit specific biological
responses. Molecular disturbance at the cellular and organ-level resulting in effects on
reproduction, growth and development at the individual-level are associated with the sex-related
endocrine system, thyroid-related endocrine system, and neuronal-, lipid-, and carbohydrate-
36
-------
metabolic systems (see Ankley et al. 2020 and Lee et al. 2020 for the latest reviews on the
subject). The underlying mechanisms of PFOA toxicity to aquatic animals, and fish in particular,
appear to be related to oxidative stress, apoptosis, thyroid disruption, and development-related
gene expression (Lee et al. 2020). The published research suggests that many of these molecular
pathways interact with each other and could be linked. For example, for several PFAS including
PFOA, oxidative stress appears correlated with effects on egg hatching and larval formation,
linking reproductive toxicity, oxidative stress, and developmental toxicity (I ,ee et al. 2020). The
actual mechanism(s) through which PFAS induce oxidati\ e stress require additional study, but
increased B-oxidation of fatty acids and mitochondrial toxicity are proposed triggers (Ankley et
al. 2020).
Of particular importance to this document is that PFOA exposure-related disruption of
the sex-related endocrine system (e.g., androgen and estrogen) at the molecular, tissue and organ
levels appears to ha\ e ad\ erse re productive outcomes in fish and invertebrates, and likely in both
freshwater and saltwater and \ ia multiple exposure routes, i.e., waterborne and dietary (Lee at al.
2020). The reproducti\ e effects were ohsei \ ed in the Fo,Fi and F2 generations of zebrafish,
Danio rcrio. in the multi-generational PFOA exposure reported by Lee et al. (2017).
It is clear that PFOA. and many other PFAS, cause a wide range of adverse effects in
aquatic organisms, including reproductive failure, developmental toxicity; androgen, estrogen
and thyroid hormone disruption; immune system disruption; and, neuronal and developmental
damage. Study of the systematic interactions among the relevant biological pathways in fish is a
research need, as well as a better understanding of several knowledge gaps in non-fish aquatic
organisms where mechanistic-based investigations need to be prioritized.
37
-------
2.7.2 Potential Interactions with Other PFAS
PFAS may occur as mixtures in the environment. Occurrence studies document the
presence of complex mixtures of PFAS in surface waters in the U.S. and across the globe
(Ahrens 2011; Ahrens and Bundschuh 2014; Giesy and Kannan 2002; Houde et al. 2006; Keiter
et al. 2012; Wang et al. 2017; see Section 2.4.1). Although EPA's PFOA criteria are based solely
on single chemical exposure aquatic toxicity tests, it is recognized thai PFAS are often
introduced into the environment as end-use formulations comprised of mixtures of PFAS and or
PFAS-precursors, the ecological effects of which are poorly understood (Ankley el al. 2020). It
is useful, therefore, to briefly summarize the types of interactions that might be expected based
on the few PFAS mixtures studies involving PFOA and one or more PFAS to dale. Note that for
purposes of this document, the reader is referred to Ankley el al (2020) and elsewhere for more
comprehensive reviews of PFAS mixtures in general, and the challenges they are expected to
present in ecological risk assessment. Findings of the studies described below are as reported by
the study authors without any additional interpretation or analysis of uncertainty.
At both the organismal and cellular levels, studies on zebrafish (Danio rerio\ Ding et al.
2013). a u ater Ilea (/ ki/>/niia Yang et al. 2019), a bioluminescent cyanobacterium
(,Anabcicini sy>.; Rodea-Palomares et al 2012), or with cultured hepatocytes of the cyprinid,
Gobiocypns rams (Wei et al 2009), demonstrate that the effects observed from in vivo and in
vitro tests on PFAS mixtures vary and can have unpredictable, exposure and species-specific
effects. For example, in a single in vivo exposure of zebrafish (I). rerio) embryos, synergism,
additivity and antagonism were all reported for different combinations/ratios of PFOA and PFOS
and endpoints (Ding et al. 2013), thereby illustrating the complexity and uncertainty associated
with mixture studies. Importantly, neither the concentration addition model nor the independent-
action model could predict the combined effects when strong interactive effects existed. More
38
-------
recently, Yang et al. (2019) exposed the water flea, Daphnia magna, to single and binary
mixtures of PFOA and PFOS. The authors reported synergism in acute and chronic toxic effects.
Conversely, Rodea-Palomares et al. (2012) showed binary PFOA and PFOS mixture as having
an antagonistic interaction at the whole range of effect levels tested using the bioluminescent
cyanobacterium, Anabaena.
In tests with cultured hepatocytes of the cyprinid G. rams, co-exposure of PFOA with a
mixture of five other PFAS [PFNA, PFDA, PFDoA, PFOS, 8:2 F'l'OH | tillered genes involved in
multiple biological functions and processes, including fatty acid metabolism and I ran sport,
xenobiotic metabolism, immune response, and oxidative stress Additionally, greater than 80%
of the altered genes in both the PFOA- and PFOS-dominaiU mixture groups were of the same
gene set. Finally, U.S. EPA (2021, unpublished) observed PI OA and PFOS interacting in an
additive manner to reduce pup body weight, pup li\ er weight, and maternal liver weight in the
Sprague-Dawley ml
2.8 Conceptual Model of PI'OA in the Aquatic Environment and Effects
A conceptual model depicts the relationship between a chemical stressor and ecological
compartments, linking exposure characteristics to ecological endpoints. The conceptual model
provided in I'igiire 2-4 summarizes sources, potential pathways of PFOA exposure for aquatic
life and aquatic-dependent wildlife and possible toxicological effects.
PFOA initially enters the aquatic environment through direct discharge from wastewater
treatment facilities, atmospheric deposition, and runoff from contaminant surfaces such as PFAS
disposal sites or contaminated biosolids. PFOA enters the aquatic environment primarily in the
dissolved form and to a lesser extent, particle-bound forms. Exposure pathways for the biological
receptors of concern (i.e., aquatic organisms) and potential effects (e.g., impaired survival,
39
-------
growth, and reproduction) in those receptors are represented in the conceptual model (Figure
2-4). Both direct (i.e., exposure from the water column which is represented by *) and indirect
(i.e., bioconcentrated by producers and bioaccumulated by consumers in higher trophic levels
represented by **) pathways are represented in the conceptual model.
40
-------
6C
a
U
PFOA Source
Point Sources
(from municipal and industrial
dischargers from applications
such as surfactants, textile stain
and soil repellents )
PFOA in Water
Dissolved & Particle-Bound
Degradation &
Metabolism of
Other PFASs
¦4
PFOA in Sediment
Aquatic Life
PFOA Source
Nonpoint Sources
(from landfill leaclrate, land
application of biosolids)
Producers
1" Trophic Transfer
(from phytoplankton, periphyton, macrophytes; e.g., algae, cyanobacteria, waterweed/common eelgrass)
-
o
P.
u
tu
Pi
Consumers
2"d Tr0phjc Transfer
(to zooplankton, macroinvertebrates;
e.g., cladocerans/copepods &
mayflies/ribbed mussels)
Consumers
3rd Trophic Transfer
(to predatory fish:
e.g., longnose dace/
American shad)
>
Consumers
4th Trophic Transfer
(to predatory fish;
e.g., largemouth bass/
striped bass)
Figure 2-4. Conceptual Model Diagram of Sources, Compartmental Partitioning, and
Trophic Transfer Pathways of Perfluorooctanoic acid (PFOA) in the Aquatic Environment
and its Bioaccumulation and Effects in Aquatic Life and Aquatic-dependent Wildlife.
PFOA sources represented in ovals, compartments w ithin the aquatic ecosystem represented by rectangles, and
effects (on trophic levels of aquatic-dependent wildlife, represented by shaded box) in pentagons. Examples of
organisms in each trophic transfer provided as freshwater/marine. Movement of PFOA from water to receptors
indicated by two separate pathways: bioconcentration by producers (*) and direct exposure to all trophic levels
within box (**). Relative proportion of PFOA transferred between each trophic level is dependent on life history
characteristics of each organism.
41
-------
2.9 Assessment Endpoints
Assessment endpoints are defined as the explicit expressions of the environmental values
to be protected and are comprised of both the ecological entity (e.g., a species, community, or
other entity) and the attributes or characteristics of the entity to be protected (U.S. EPA 1998).
Assessment endpoints may be identified at any level of organization (e.g., individual, population,
community). In context of the CWA, aquatic life criteria for toxic suhslaiices are typically
determined based on the results of toxicity tests with aquatic organisms, lor which adverse
effects on growth, reproduction, or survival are measured. This information is typically compiled
into a sensitivity distribution based on genera and representing the impact on taxa across the
aquatic community. Criteria are based on the 5th percentile of genera and are, thus intended to be
protective of approximately 95 percent of aquatic genera to ensure aquatic communities are
protected. Assessment endpoints consistent with the criteria developed in this document are
summarized in Table 2-1
The use of laboratory toxicity tests to protect bodies of water and resident aquatic species
was based on the theory that effects occurring to a species in appropriate laboratory tests will
generally occur to the same species in comparable field situations. Since aquatic ecosystems are
complex and diverse, the 11>X5 Guidelines recommend that acceptable data be available for at
least eight genera with a specified taxonomic diversity (the standard eight minimum data
requirements, or MDRs) The intent of the eight MDRs is to serve as a typical surrogate sample
community representative of the larger and generally much more diverse natural aquatic
community, not necessarily the most sensitive species in a given environment. The 1985
Guidelines note that since aquatic ecosystems can tolerate some stress and occasional adverse
effects, protection of all species at all times and places are not deemed necessary (the intent is to
42
-------
protect 95 percent of a group of diverse taxa, and any commercially and recreationally important
species).
2.10 Measurement Endpoints
2.10.1 Overview of Toxicity Data Requirements
To ensure the protection of various components of an aquatic ecosystem, EPA collects
acute toxicity test data from a minimum of eight diverse taxonomic groups.
• Acute freshwater criteria require data from the following eight taxonomic groups:
a) the family Salmonidae in the class Osteichthyes
b) a second family in the class Osteichthyes, preferably a commercially or
recreationally important warmwater species (e.g., bluegill, channel catfish)
c) a third family in the phylum Chordata (may be in the class Osteichthyes or may
be an amphibian)
d) a planktonic crustacean (e g . cladoceran, copepod)
e) a benthic crustacean (e.g., ostracod. isopod, amphipod, crayfish)
f) an insect (e.g., mayfly, dragonfly, damselflv, stolidly. caddisfly, mosquito,
midge)
g) a family in a phylum other than Ailhropoda or Chordata (e.g., Rotifera, Annelida,
Mollusca)
h) a family in any order of insect or any phylum not already represented
• Acute estuarine/marine criteria require data from the following taxonomic groups:
a) two families in the phylum Chordata
h>) a family in a phylum other than Arthropoda or Chordata
c) a family from either Mysidae or Penaeidae
d) three other families not in the phylum Chordata (may include Mysidae or
Penaeidae, u hichever was not used above)
e) any other family
Additionally, to ensure the protection of various components of the aquatic ecosystem
from long term exposures chronic toxicity test data are recommended for the same minimum of
eight diverse taxonomic groups that are recommended for freshwater acute criterion derivation.
If the eight diverse taxonomic groups are not available to support chronic criterion derivation
using a genus distribution approach, the chronic criterion may be derived using an acute-to-
43
-------
chronic ratio (ACR) approach. To apply an ACR approach to derive a chronic freshwater
criterion a minimum of three taxa are recommended, with at least one chronic test being from an
acutely sensitive species. Acute-to-chronic ratios (ACRs) can be calculated with data for aquatic
organisms.
• Chronic aquatic life criteria require data from the following taxonomic groups:
a) At least one is a fish
b) At least one is an invertebrate
c) At least one is an acutely sensitive freshwater species. Ibr freshwater chronic
criterion (the other two may be saltwater species)
d) At least one is acutely sensitive saltwater species for estuarine murine chronic
criterion (the other two may be freshwater species)
The 1985 Guidelines also specified at least one quantitative test with a freshwater alga or
vascular plant. If plants are among the most sensitive aquatic organisms, toxicity test data from a
plant in another phylum should also be a\ ailaMe Aquatic plant toxicity data are examined to
determine whether aquatic plants are likely to be adversely affected by the concentration
expected to be protective for other aquatic organisms.
2.10.2 Measure of PI OA l-xposure Concentrations
These PI OA ambient water quality criteria are for the protection of aquatic life. This
criteria document pro\ ides a critical re\ iew of all data identified in EPA's literature search for
PFOA. including
• the anionic form (CAS No. 45285-51-6),
• the acid form (CAS No. 335-67-1), and;
• the ammonium salt (CAS No. 3825-26-1).
Based on EPA's data review, PFOA toxicity studies typically used the linear PFOA
isomer for dosing with fewer studies using the branched isomer. Data for possible inclusion in
the PFOA criteria were obtained from published literature reporting acute and chronic exposures
of PFOA that were associated with mortality, growth, and reproduction. This set of published
44
-------
literature was identified using the ECOTOXicology database (ECOTOX;
https://cfpub.epa.gov/ecotox/) as meeting data quality standards. ECOTOX is a source of high-
quality toxicity data for aquatic life, terrestrial plants, and wildlife. The database was created and
is maintained by the EPA, Office of Research and Development, Center for Computational
Toxicology and Exposure. The ECOTOX search generally begins with a comprehensive
chemical-specific literature search of the open literature conducted according to ECOTOX
Standard Operating Procedures (SOPs; Elonen 2020). The search terms are often comprised of
chemical terms, synonyms, degradates and verified Chemical Abstracts Sen ice (CAS) numbers.
After developing the literature search strategy, ECOTOX curators conduct a series of searches,
identify potentially applicable studies based on title and ahsii iici, acquire potentially applicable
studies, and then apply the applicability criteria for inclusion in ECOTOX. Applicability criteria
for inclusion into ECOTOX generally include:
1. The toxic effects are related to single chemical exposure (unless the study is being
considered as part of a mixture effects assessment);
2. There is a biological effect on live, whole organisms or in vilro preparation including
gene chips or omics data on adverse outcome pathways potentially of interest;
3. Chemical test concentrations are reported;
4. There is an explicit duration of exposure;
5 Toxicology information that is relevant to OW is reported for the chemical of concern;
6. The paper is published in the English language;
7. The paper is available as a full article (not an abstract);
8. The paper is publicly a\ ailable;
9. The paper is the primary source of the data;
10. A calculated endpoint is reported or can be calculated using reported or available
information.
11. Treatment(s) are compared to an acceptable control;
12. The location of the study (e.g., laboratory vs. field) is reported; and
13. The tested species is reported (with recognized nomenclature).
Following inclusion in the ECOTOX database, toxicity studies were subsequently
evaluated by Office of Water. All studies were evaluated for data quality as described by U.S.
EPA (1985), EPA's Office of Chemical Safety and Pollution Prevention (OPP)'s Ecological
45
-------
Effects Test Guidelines (U.S. EPA 2016c), and EPA OW's internal data quality standard
operating procedure (SOP), which is consistent with OPP's data quality review approach (U.S.
EPA 2016c). Office of Water completed a Data Evaluation Record (DER) for each species by
chemical combination from the PFOA studies identified by ECOTOX. This in-depth review
ensured the studies used to derive the criteria resulted in robust scientifically defensible criteria.
Example DERs are shown in Appendix Q with the intent to coin e\ I lie meticulous level of
evaluation, review, and documentation each PFOA study identified In IX'OIOX was subject to.
Studies that did not fully meet the data quality objectives outlined I -PA SOP were not
considered for inclusion in the criteria derivation, including some studies with oilier PFAS
exposures, but were considered qualitatively as supporting information and are characterized in
the Effects Characterization. These studies are listed in Appendix (i and Appendix H.
Furthermore, only single chemical toxicity tests with PI OA were considered for possible
inclusion in criteria deri\ ation. studies that tested chemical mixtures, including mixtures with
PFAS were excluded from criteria deri\ ation. Both controlled laboratory experiments and field
observations/studies were included
The ll>K5 (inidelines recommend only toxicity tests focused on North American resident
species he considered. Due to l-PA's interest in using all available quality data, particularly for a
data-sparse chemical like PI OA (relative to chemicals such as cadmium or ammonia), toxicity
studies were considered lor possible inclusion regardless of the test species residential status in
North America. Use of non-North American residential species is also consistent with other
published aquatic life criteria (U.S. EPA 2018b). Non-North American resident species also
serve as taxonomically-related surrogate test organisms for the thousands of untested resident
species. Supporting analyses to evaluate the influence of including non-resident species on the
46
-------
freshwater criteria magnitudes were conducted by limiting toxicity datasets to North American
resident species with established populations in North America (see Section 4.1). These analyses
provided an additional line-of-evidence that supports inclusion of non-resident species in PFOA
criteria derivation.
Toxicity tests used in many previous EPA aquatic life criteria documents are typically
based on measured chemical concentrations only. For PFOA, F.PA has examined the issue of
whether nominal (unmeasured) and measured concentrations are in close agreement with each
other (see Appendix M). Briefly, approximately 24.3% of the 152 freshwater acute and chronic
toxicity tests determined to be quantitatively or qualitatively acceptable reported measured
PFOA concentrations in at least one treatment. Approximately 57.1% of the 14 saltwater acute
and chronic toxicity tests determined to be quantitatively or qualitatively acceptable reported
measured PFOA concentrations in at least one treatment Pairs of nominal and corresponding
measured PFOA concentrations were compared to one another through: (1) linear correlation
analysis and: (2) an assessment of measured concentrations as a percent of its paired nominal
concentration Linear correlation between measured and corresponding nominal concentrations
suggests a high degree of precision between paired observations across all test conditions and
83% of measured freshwater concentrations and all of measured saltwater concentrations fell
within 20% of paired nominal concentrations, which represent the test acceptability threshold
identified by EP.Vs OCSPP's Ecological Effects Test Guidelines. Instances where measured
concentrations were not within 20% of nominal were isolated to a few studies. In these isolated
cases, suspected dosing errors, unexplained phenomena, and/or presence of substrate (e.g.,
sediment) may have contributed to observed differences. Overall, PFOA concentrations in test
waters are expected to remain relatively constant over the course of acute and chronic exposures
47
-------
given its ability to resist breakdown and transformation (Ahrens et al. 2011). Because suspected
dosing errors were a relatively rare occurrence and toxicity tests with substrate and nominal
concentrations only were not used quantitatively in PFOA criteria derivation, EPA determined
nominal test concentrations adequately represent actual PFOA exposures in standard acute and
chronic laboratory-based toxicity tests. Consequently, PFOA toxicity tests were not excluded
from quantitative use in criteria derivation on the basis of unmeasured lest concentrations alone.
Typically, per the 1985 Guidelines acute toxicity data from all measured flow-through
tests would be used to calculate species mean acute values (SMAVs), unless data from a
measured flow-through test were unavailable, in which case the acute criterion would be
calculated as the geometric mean of all the available acute values (i.e., results of unmeasured
flow-through tests and results of measured and unmeasured sialic and renewal tests). Chronic
unmeasured flow-through tests, as well as measured and unmeasured static and renewal tests are
not typically considered to calculate chronic \ allies In the ease of the PFOA, static, renewal, and
flow-through experiments were considered for possible inclusion for both species mean acute
and chronic values regardless of whether PI OA concentrations were measured because PFOA is
a highly stable compound, resistant to hydrolysis, photolysis, volatilization, and biodegradation
(Section 12 1) and, therefore, expected to vary only minimally in the course of a toxicity test.
Additionally, chronic \ alues were based on endpoints and exposure durations that were
appropriate to the species Thus, both life- and partial life-cycle tests were utilized for the
derivation of the chronic criterion. However, it should be noted that the 1985 Guidelines specify
life-cycle chronic tests are typically used for invertebrates. The chronic studies used in the
derivation of the chronic water column-based PFOA criterion followed taxon-specific exposure
duration requirements from various test guidelines (i.e., EPA's 1985 Guidelines and EPA's
48
-------
OCSPP's Ecological Effects Test Guidelines) when available. For example, only chronic
daphnid studies of 21 days were considered in the chronic criterion derivation because the EPA
1985 Guidelines states daphnid tests should begin with young < 24-hours old and last at least 21
days. When taxon-specific exposure duration requirements were not available for a particular test
organism in the PFOA toxicity literature, both life- and partial life-cycle tests were considered in
the derivation of the chronic criterion.
PFOA toxicity in aquatic life is manifested as effects on survi\ al. growth, and
reproduction. Measurements of fish tissue may be linked to the chronic ad\ eise effects of PFOA,
since PFOA is highly persistent and potentially bioaccumulative.
2,10,3 Measures of Effect
Each assessment endpoint requires one or more "measures of ecological effect," which
are defined as changes in the attributes of an assessment end point itself or changes in a surrogate
entity or attribute in response to chemical exposure Ideological effects data were used as
measures of direct and indirect effects to growth, reproduction, and survival of aquatic
organisms.
2.10..V I Acute Measures of I-fleet
The acute measures of effect on aquatic organisms are the lethal concentration (LCso),
effect concentration (I vCso). or inhibitory concentration (ICso) estimated to produce a specific
effect in 50 percent of the test organisms. LCso is the concentration of a chemical that is
estimated to kill 50 percent of the test organisms. EC so is the concentration of a chemical that is
estimated to produce a specific effect in 50 percent of the test organisms. And the ICso is the
concentration of a chemical that is estimated to inhibit some biological process (e.g., enzyme
inhibition associated with an apical endpoint such as mortality) in 50 percent of the test
organisms.
49
-------
2.10.3.2 Chronic Measures of Effect
The endpoint for chronic exposures is the effect concentration estimated to produce a
chronic effect on survival, growth, or reproduction in 10 percent of the test organisms (ECio).
EPA selected an ECio to estimate a low level of effect that would be both different from controls
and not expected to be severe enough to cause effects at the population level for a potentially
bioaccumulative contaminant, such as PFOA. The use of the ECi... instead of anEC2o, is also
consistent with the use of this metric for the bioaccumulative pollutant selenium in the recent
2016 Selenium Freshwater Aquatic Life Criteria (II S EPA 2016a). Use of a 1 <»" <» effect
concentration for deriving chronic criteria magnitudes is also consistent with the harmonized
guidelines from OECD and the generally preferred effect le\ el lor countries such as Canada,
Australia, and New Zealand (CCMC 20( >7. Warne et al. 2UIS)
Regression analysis was used preferentially to characterize a concentration-effect
relationship and lo estimate concentrations at which chronic effects are expected to occur (i.e.,
point estimate). Reported (No ()hser\ ed Effect Concentrations) (NOECs) and (Lowest Observed
Effect Concentrations) (I OIX's) were only used for the derivation of a chronic criterion when a
robust IX'i" could nol he calculated lor the genus. ANOECisthe highest test concentration at
which none of the ohscr\ ed effects are statistically different from the control. A LOEC is the
lowest test concentration at \\ liich the observed effects are statistically different from the control.
When LOECs and NOECs were used, a Maximum Acceptable Toxicant Concentration (MATC)
was calculated, which is the geometric mean of the NOEC and LOEC. For the calculation of a
chronic criterion, point estimates were selected for use as the measure of effect in favor of
MATCs, as MATCs are highly dependent on the concentrations tested. Point estimates also
provide additional information that is difficult to determine with an MATC, such as a measure of
effect level across a range of tested concentrations. A decision rule was also applied to the PFOA
50
-------
toxicity data when an author-reported NOEC or LOEC was used in conformity with the 2013
Ammonia Freshwater Aquatic Life Criteria (U.S. EPA 2013) such that "greater than" values for
concentrations of a relatively low magnitude compared to the other available toxicity data, and
"less than" values for concentrations of relatively high magnitude were considered to add little
significant information to the analyses and were not used quantitali\ ely Conversely, if data from
studies with relatively low "less than" values indicated a significant effect or studies with
relatively high "greater than" values only found an incomplete response lor a chronic endpoint
(indicating low toxicity of the test material), those data significantly enhanced the understanding
of PFOA toxicity. Thus, the decision rule was applied as follows: "greater than" ( ) high toxicity
values and "less than" (<) low toxicity values were used quantitatively to derive the chronic
water column-based PFOA criterion (U.S N\\ 2<> 13) Data that met the quality objectives and
test requirements were utilized quantitati\cl\ in dei i\ inu freshwater criteria for aquatic life and
are presented in TaMe 3-3 and TaMe 3-7.
Table 2-1. Summary of Assessment K ml points and Measures of Effect Used in the Criteria
Derivation for PI-OA.
Assessment Kndpoinls for (lie Aquatic
(oniniunitY
Measures of KITect
Aquatic Life. Sui\i\al, growth, and
reproduction of freshwater and
estuarine/marine aquatic life (i e . fish,
amphibians, aquatic in\ei tehiates)
For effects from acute exposure:
1. LCso, EC50, or IC50 concentrations in water
2. NOEC and LOEC concentrations in water
For effects from chronic exposure:
1. EC10 concentrations in water
2. NOEC and LOEC concentrations in water;
Only used when an EC 10 could not be calculated
for a genus.
NOEC = No observed effect concentration
LOEC = Lowest observed effect concentration
ECio = 10% Effect Concentration
51
-------
2.10.3.3 Summary of Independent Calculation of Toxicity Values
Toxicity values, including LCso and ECio values, were independently calculated from the
data presented in the toxicity studies meeting the inclusion criteria described above when
adequate concentrations-response data were published in the study or could be obtained from
authors. When concentration-response data were not presented in toxicity studies, concentration-
response data were requested from study authors to independently calculate toxicity values. In
cases where study authors did not respond to EPA's request lor data or were unable to locate
concentration-response data, the toxicity values were not independently calculated by EPA, and
the reported toxicity values were retained for crileria dc\ iation. Where concentration-response
data were available, they were analyzed using the statistical software program R (version 3.6.2)
and the associated dose-response curve (die) package. The R die package has various models
available for modeling a concentration-response relationship lor each toxicity study. The specific
model used to calculate toxicity \ alues was selected following the details provided in Appendix
K, and the models performed well on most or all statistical metrics. The independently calculated
toxicity values used to deri\ e the PI ()A aquatic life criteria were included in each study
summary below and were used to derive criteria for aquatic life, where available. Details relating
to the independent verification of toxicity values for each toxicity study used to derive the
criteria were included in Appendix A.2 and Appendix C.2.
2.11 Analysis Plan
2.11,1 Derivation of Water Column Criteria
During CWA section 304(a) criteria development, EPA reviews and considers all
relevant toxicity test data. Information available for all relevant species and genera were
reviewed to identify: 1) data from acceptable tests that meet data quality standards; and 2)
whether the acceptable data meet the MDRs as outlined in EPA's 1985 Guidelines (U.S. EPA
52
-------
1985). The taxa represented by the different MDR groups represent taxa with different
ecological, trophic, taxonomic and functional characteristics in aquatic ecosystems, and are
intended to be a representative subset of the diversity within a typical aquatic community. MDRs
for derivation of acute and chronic freshwater criterion were met for seven of the eight MDRs.
Because nearly all MDRs were met for deriving both acute and chronic criteria, EPA derived the
acute and chronic freshwater column criteria based on the se\en MDRs PFOA insect toxicity
testing is an active and ongoing area of research within the ecotoxicolouical scientific
community that will likely provide information to evaluate the sensitivity of insects to acute and
chronic PFOA exposures before the PFOA criteria document is finalized.
Acute and chronic MDRs for PFOA estuarine/marine criteria derivation were not met
and, consequently, acute and chronic estuarine marine criteria were nol derived. EPA is,
however, including an acute aquatic life benchmark lor estuarine marine environments (see
Appendix L), using a\ ailahlc estuarilie/marine species toxicity data and application of ORD's
peer-reviewed ch-ICI-I tool A minimal number of tests from acceptable studies of aquatic algae
and vascular plants were also a\ ailable lor possible derivation of a Final Plant Value. However,
the rdali\ e sensili\ ily of freshwater plants to PFOA exposures indicated plants are less sensitive
than aquatic \ crtchrates and in\ ertehrates so plant criteria were not developed.
2,11,2 Derivation of Tissue-liased Criteria
Chronic toxicity studies (both laboratory and field studies) were further screened to
ensure they contained the relevant chronic PFOA exposure conditions to aquatic organisms (i.e.,
dietary, or dietary and waterborne PFOA exposure), measurement of chronic effects, and
measurement of PFOA in tissue(s). EPA considered deriving tissue-based criteria using
empirical toxicity tests with studies that exposed organisms to PFOA in water and/or diet and
reported exposure concentrations based on measured tissue concentrations. This approach would
53
-------
also correspond with the 2016 Selenium Aquatic Life Freshwater Criterion, which is the only
304(a) aquatic life criterion with tissue-based criterion elements. However, the freshwater
chronic PFOA toxicity data with measured tissue concentrations were limited, with no
quantitatively acceptable tissue-based tests. Qualitatively acceptable tissue-based tests were
reported for four species (three fish species and one amphibian) across five publications.
Therefore, there were insufficient data to derive tissue-based criteria using a GSD approach from
empirical tissue data from toxicity studies. EPA thus developed protecti\ e tissue-based criteria
through a bioaccumulation factor approach (Burkhard 2021).
2,11,3 Translation of Chronic Water Column Criterion lo Tissue Criteria
Because there were insufficient chronic toxicity d;il;i with measured tissue concentrations
to derive chronic PFOA tissue criteria using a (iSI) approach. I-IW derived PFOA chronic
tissue-based criteria by translating the chronic freshwater column criterion (see Section 3.2.1.3)
into tissue-based criteria magnitudes using Moaccumulation factors and the following equation:
Tissue Criteria = Chronic Water Column Criterion x BAF (EquationX-l)
The resulting tissue-bused criteria magnitudes correspond to the tissue type from the BAF used
in the equation (see Section 2 I I 3 I ).
2.11 ..v I Aquatic I .ife Bioaccumulation Factors
A Moaccumulation I actor (BAF) is determined from field measurements and is calculated
using the equation
BAF = (X-2)
Cwater
Where:
Cbiota = PFOA concentration in the organismal tissue(s)
Cwater = PFOA concentration in water where the organism was collected
54
-------
EPA considered BAF data from field measurements to capture all PFOA exposure routes,
i.e., dietary, water, contact with sediments via dermal exposure and ingestion, and maternal
transfer. Depending on the tissue residue measurement, BAFs can be based upon residues in the
whole organisms, muscle, liver, or any other tissue.
Searching for literature reporting on PFOA was implemented by developing a series of
chemical-based search terms. These terms included chemical names and Chemical Abstracts
Service registry numbers (CASRN or CAS), synonyms, tradenames, and oilier relevant chemical
forms (i.e., related compounds). Databases searched were Current Contents, h oQuest CSA,
Dissertation Abstracts, Science Direct, Agricola. TO.WIF.T. and UNIFY (database internal to
U.S. EPA's ECOTOX database). The literature search yielded numerous citations and the
citation list was further refined by excluding citations on analytical methods, human health,
terrestrial organisms, bacteria, and where PI OA was not a chemical of study. The citations
meeting the search criteria were reviewed for reported B \l s and/or reported concentrations in
which BAFs could be calculated lor freshwater and estuarine/marine species. BAFs from both
freshwater and estnarine marine species were considered because; (1) inclusion of
estuarine marine Ii.\I¦"s expanded the relati\ely limited PFOA BAF dataset and (2) Burkhard
(2021) did not specifically observe notable differences in PFAS BAFs between freshwater and
estuarine/marine systems, instead stating additional research is needed to formulate conclusions.
Data from papers with appropriate BAF information were further screened for data
quality. Four factors were evaluated in the screening of the BAF literature: (1) number of water
samples, (2) number of organism samples, (3) water and organism temporal coordination in
sample collection, and (4) water and organism spatial coordination in sample collection.
Additionally, the general experimental design was evaluated. Table 2-2 below outlines the
55
-------
screening criteria for study evaluation and ranking. Only BAFs of high and medium quality were
used to derive the tissue criteria (Appendix P). For further details on BAFs compilation and
ranking, see Burkhard (2021).
Table 2-2. Evaluation Criteria for Screening Bioaccumulation Factors (BAFs) in the Public
Literature.
Screening l-'actor
High Qualify
Medium Quality
Low Quality
Number of Water Samples
>3
2-3
1
Number of Organism Samples
>3
2-3
1
Temporal Coordination
Concurrent
collection
Within one year
Collection period >1 year
Spatial Coordination
Collocated
collection
Within 1 - 2 km
Significantly different
locations
(>2 km)
General Experimental Design
Mixed species tissues
samples
Modified from Burkhard (2021).
3 EFFECTS ANALYSIS FOR AQl ATK LIFE
3.1 Toxicity to Aquatic Life
All available studies relating lo the acute and chronic toxicological effects of PFOA on
aquatic life were considered in the deri\ alion of these national recommended PFOA criteria.
Data lor possible inclusion in these PI OA criteria were obtained from published literature
reporting acute and chronic exposures of PFOA that were associated with mortality, survival,
growth, and reproduction. Acute and chronic data meeting the quality objectives and test
requirements were utilized quantitatively in deriving these criteria for aquatic life and are
presented in Appendix A. Acceptable Freshwater Acute PFOA Toxicity Studies; Appendix B:
Acceptable Estuarine/Marine Acute PFOA Toxicity Studies; Appendix C: Acceptable
Freshwater Chronic PFOA Toxicity Studies, and; Appendix D: Acceptable Estuarine/Marine
Chronic PFOA Toxicity Studies.
56
-------
3.1.1 Summary of PFOA Toxicity Studies Used to Derive the Aquatic Life Criteria
Quantitatively acceptable acute PFOA toxicity data were available for 25 freshwater
species, representing 18 genera and 16 families in five phyla, and four estuarine/marine species,
representing four genera and three families in three phyla (Table 3-1). Quantitatively acceptable
chronic PFOA toxicity data were available for 11 freshwater species, representing 10 genera and
nine families in three phyla. There were no quantitatively accepliihle chronic studies with
estuarine/marine organisms. The following study summaries present the key acute and chronic
freshwater toxicity data with effect values that were used quantitatively to deri\ e ilie acute and
chronic freshwater and estuarine/marine criteria to protect aquatic life. Study summaries for the
most sensitive taxa are presented below and are grouped by acute or chronic exposure and sorted
by sensitivity to PFOA.
Acute and chronic values were presented as reported by the study authors for each
individual study, unless stated otherwise. EIW independently calculated these toxicity values if
sufficient raw data were available for EPA to conduct statistical analyses. EPA's independently-
calculated toxicity values were used preferentially, where available. Author-reported toxicity
values and EIWs independently calculated values (where available) were included in each study
summary and in appendices, as applicable. The results of all toxicity values, such as LC values,
EC values. NOI 'Cs. LOEC's. and species- and genus-mean values, are given to four significant
figures to prevent round-olT error in subsequent calculations, not to reflect the precision of the
value. The specific toxicity value utilized in the derivation of the corresponding PFOA criteria is
stated for each study at the end of the summaries below and in the respective appendices.
57
-------
Table 3-1. Summary Table of Minimum Data Requirements per the 1985 Guidelines
Reflecting the Number of Acute and Chronic Genus and Species Level Mean Values in the
Freshwater and Saltwater Toxicity Dataset
ts for PFOA.
MDU1' h
l-'reshwaler
GMAV
S.M.W
CMC V
SMCV
I annis Salnionidae 111 die class Osleichth) es
1
1
1
1
Second family in the class Osteichthyes, preferably
a commercially or recreationally important
warmwater species
3
3
2
2
Third family in the phylum Chordata (may be in
the class Osteichthyes or may be an amphibian,
etc.)
5
10
2
2
Planktonic Crustacean
3
5
2
3
Benthic Crustacean
1
1
1
1
Insect
()¦'
IT
1
1
Family in a phylum other than Arthropoda or
Chordata (e.g., Rotifera, Annelida, orMollusca)
4
4
1
1
Family in any order of insect or any phylum not
already represented
1
1
0b
0b
Total
IS
25
10
11
MDU
Salt water'
GMAV
S.MAY
CMCV
SMCV
Family in the phylum Chordata
i)
i)
0
0
Family in the phylum Chordata
i)
0
0
0
Either the Mysidac or Pcnacidac family
2
0
0
Family in a phylum other than Arthropoda or
Chordata
1
1
0
0
Family in a phylum other than Chordata
1
1
0
0
Family in a phylum other than Chordata
0
0
Family inaphs I11111 nther than Clmi'dala
0
0
Any other famiK
0
0
Total
4
4
0
0
a One acute MDR. lor aquatic insects. w;is noi fulfilled. Of the available qualitatively-acceptable insect data, only
Yang ol al (2014) conduced a lesi I'm- die standard 96 acute exposure. Other qualitatively acceptable insect
toxicity dala u ere based on either chronic or sub-chronic exposure durations Yang et al. (2014) specifically
conducted a ''(>-lk>ur renewal lesi u ith measured PFOA concentrations on the midge, Chironomusplumosus. This
study was not acceptable for quantitative use due to the potential problematic source of the organisms but was
retained for qualilali\ e use The reported LC50 was 402.24 mg/L PFOA. PFOA insect toxicity testing is an active
and ongoing area of research u itliin the ecotoxicological scientific community that will likely provide additional
information to fully evaluate 011 the sensitivity of insects to acute PFOA exposures before the PFOA criteria
document is finalized.
b One chronic MDR, for any order of aquatic insects or any phylum not already represented, was not fulfilled with
quantitatively acceptable chronic data. PFOA insect toxicity testing is an active and ongoing area of research
within the ecotoxicological scientific community that will likely provide information to evaluate the sensitivity of
insects to chronic PFOA exposures before the PFOA criteria document is finalized.
c The 1985 Guidelines require that data from a minimum of eight families are needed to calculate an
estuarine/marine criterion. Insufficient data exist to fulfill all eight of the taxonomic MDR groups. Consequently,
EPA cannot derive an estuarine/marine acute criterion, based on the 1985 Guidelines. However, EPA has
developed draft estuarine/marine benchmarks through use of surrogate data to fill in missing MDRs using EPA's
WeblCE tool and other New Approach Methods. These benchmarks are provided in Appendix L.
58
-------
3.1.1.1 Summary of Acute PFOA Toxicity Studies Used to Derive the Freshwater Aquatic Life
Criterion
The acute data set for PFOA contains 18 genera representing seven of the eight
taxonomic MDR groups. Quantitatively-acceptable data for acute PFOA toxicity were available
for four freshwater fish species, representing four genera and three families and fulfilled two of
the eight MDRs. Quantitatively acceptable data for acute PFOA toxicity were also available for
11 freshwater invertebrate species, representing nine genera and eight families, and fulfilled four
of the eight MDRs. Data for acute PFOA toxicity were available for 10 freshwater amphibian
species, representing five genera and five families fulfilling one of the MDRs. The missing MDR
is a representative from an insect family. Therefore, qualitalis ely acceptable data in Appendix G
were examined to determine if any "Qualitati\ c Data" provided information on the relative
sensitivity of aquatic insects. Yang et al. (2<)|4) conducted a test for the standard 96 hour acute
exposure. Other qualitali\ ely acceptable insect toxicity data were based on either chronic or sub-
chronic exposure durations Yang el al (2014) specifically conducted a 96-hour renewal test
with measured PFOA concentrations on the midge, Chironomusplumosus. This study was not
acceptable lor quantilali\ e use due to the potential problematic source of the organisms but was
retained lor qualitative use The reported LCso was 402.24 mg/L PFOA. PFOA insect toxicity
testing is an acti\ e and ongoing area of research within the ecotoxicological scientific
community that will likely provide additional information to fully evaluate the sensitivity of
insects to acute PFOA exposures before the PFOA criteria document is finalized. Summaries of
studies for the most sensitive acute genera are describe below, with the four most sensitive
genera provided in Table 3-2.
59
-------
Table 3-2. The Four Most Sensitive Genera Used in Calculating the Acute Freshwater
Criterion (Sensitivity Rank 1-4).
Ranked Below from A
dost to Least Sensitive
Knnk
Genus
(/MAN
(inss/l-)
Species
Comment
1
Chydorus
93.17
Cladoceran
(Chydorus sphaericus)
North American resident
species
2
Daphnia
144.1
Cladoceran
(Daphnia carinata)
Non-North American
resident species
Cladoceran
(Daphnia magna)
North American resident
species
Cladoceran
(Daphnia pultcarta)
North American resident
species
3
Brachionus
150.0
Rotifer
(Brachionus calyciflortis)
North American resident
species
4
Ligumia
161.0
Black sandshell mussel
(Ligumia recta)
North American resident
species
3.1.1.1.1 Most acutely sensitive genus: Chydorus (cladoceran)
Le and Peijnenburg (2013) performed a 4K-hour static unmeasured acute PFOA toxicity
test with the cladoceran, Chydorus sphaericus The authors reported the 48-hour EC so was 0.22
mM PFOA (91.11) mu I.) I-IW performed concentration-response (C-R) analysis for the test and
calculated a LC;.. of l)3 I 7 mu I. PI-OA that is acceptable for quantitative use. No other
quantitatively acceptable acute toxicity data were a\ ailable for Chydorus sphaericus or other
members of the genus (liyilonis Therefore, the I.Cfu (i.e., 93.17 mg/L) from this test served
directly as the ('hydorus sphaericus SM AV and the Chydorus Genus Mean Acute Value
(GMAV).
3.1.1.1.2 Second most acutely sensitive genus: Daphnia (cladoceran)
Logeshwaran et al. (2021) conducted an acute PFOA test with the cladoceran, Daphnia
cannula, and PFOA (95% purity, purchased from Sigma-Aldrich Australia) following OECD
guidelines (2000a) with slight modifications. Authors used nominal test concentrations (0, 0.5, 1,
2.5, 5, 10, 20, 30, 40, 50, 100, 150, 200 and 250 mg/L PFOA) with three replicates per treatment.
No mortality occurred in the controls. The author-reported 48-hour ECso was 78.2 mg/L PFOA.
60
-------
The EPA-cal culated 48-hour EC so value was 66.80 mg/L, which was acceptable for quantitative
use. No other quantitatively acceptable acute tests were available for this species and the EC so of
66.80 mg/L from Logeshwaran et al. (2021) served directly as the Daphnia carinata SMAV.
Boudreau (2002) performed a 48-hour static PFOA (CAS # 335-67-1, >97% purity)
acute test with Daphniapulicaria, following ASTM E729-96 (1999). Five unmeasured test
concentrations plus a negative control were used with 3-4 replicates per treatment and 10
daphnids per replicates. Nominal concentrations were 0 (negative control). 2(-> 3, 52.6, 105, 210
and 420 mg/L. Mortality of daphnids in the negative control was not reported, but the protocol
followed by authors (i.e., ASTM E729-96) required > 90% survival in negative controls. The 48-
hour D. pulicaria EC so reported in the publication was 203.7 mg/L, which was acceptable for
quantitative use. No other quantitatively acceptable acute tests were available for this species and
the ECso of 203.7 mg/L from Boudreau (2002) served directh as the D. pulicaria SMAV.
Boudreau (2002) also performed a 4S-hour static PFOA (CAS # 335-67-1, >97% purity)
acute test with 1 kip/inni magna follow inu the same methods used in the I), pulicaria acute test.
The 48-hour D. magna l-C*.. reported in the publication was 223.6 mg/L, which was acceptable
for qnantitati\ e use
Colombo et al. (2008) conducted a 48-hour static PFOA (ammonium salt, CAS # 3825-
26-1, 99.7" 0 purity) acute lest on Daphnia magna. Authors stated the test followed OECD test
guideline 202 (ll^2) The nominal test concentrations included control, 100, 178, 316, 562 and
1,000 mg/L, with four replicates/treatment and five animals/replicate. No mortality was observed
in the controls, and the 48-hour EC50 reported in the study was 480 mg/L, which was acceptable
for quantitative use.
61
-------
Ji et al. (2008) performed a 48-hour static acute test of PFOA (CAS # 335-67-1, purity
unreported; obtained from Sigma Aldrich, St. Louis, MO) on I), magna. Authors stated the test
followed U.S. EPA/600/4-90/027F (2002). The test involved four replicates of five daphnids
each in five unmeasured test concentrations plus a negative control. Nominal concentrations
were 0 (negative control), 62.5, 125, 250, 500 and 1,000 mg/L. Mortality of daphnids in the
negative control was not reported, although EPA/600/4-90/027F requires at least 90% survival
for test acceptability. The author-reported 48-hour ECso for the study was 47(-> 52 mg/L (95% C.I.
= 375.3 - 577.7 mg/L). EPA performed C-R analysis for the test. The EPA-calculated ECso was
542.5 mg/L PFOA and was acceptable for quanliUili\ e use
Li (2009) conducted a 48-hour static PFOA (ammonium salt, >98%) purity) acute test
with Daphnia magna. Authors stated the lest generally followed OI-CD 202 (1984). The test
employed five replicates of six daphnids each in ll\ e lest concentrations (nominal range = 31 -
250 mg/L) plus a neuati\ e control No control daphnids were immobile at the end of the test. The
author-reported 4S-hour I X';.. Ibr the study was 181 mg/L (95%> C.I.: 166-198 mg/L) which was
averaged across three tests I-PA performed C-R analysis for each individual test. All three tests
had acceptable cur\ es with NW-calculated ECso values of 220.8 mg/L, 157.9 mg/L, and 207.3
mg/L.. which were acceptable for quantitative use.
Yiing ol ;il. (2014) conducted a 48-hour acute test of PFOA (CAS # 335-67-1, 99%>
purity) with Daphnia magna, following ASTM E729 (1993). The test employed three replicates
of 10 daphnids each in six test concentrations plus a negative and solvent control. Nominal
concentrations were 0 (negative and solvent controls), 50, 80, 128, 204.8, 327.68 and 524.29
mg/L. Test concentrations were measured in low and high treatments only. Negative control and
solvent control mortality were 0%> each. The author-reported 48-hour LCso was 201.85 mg/L
62
-------
(95% C.I. = 134.68 - 302.5 mg/L). EPA performed C-R analysis for the test and fit an acceptable
curve with an EPA-calculated LCso of 222.0 mg/L PFOA, which was acceptable for quantitative
use.
Barmentlo et al. (2015) performed a 48-hour static, measured acute test of PFOA (CAS
# 335-67-1, >96%) with Daphnia magna following OECD 202 (2004) test guidelines. The test
involved four to six replicates of five daphnids each in five tesl concentrations plus a negative
control. Nominal concentrations were not provided, but PFOA was measured in the control,
lowest, and highest test concentrations. Based on these measurements, the authors interpolated
all test concentrations to be: 0.053 (negative control), 81. 128. 202, 318 and 503 mu I,. The
author-reported 48-hour EC so was 239 mg/L (95% C.I. - 190-287 mg/L). EPA performed C-R
analysis for the test and fit an acceptable cur\ e with an EPA-calculaled ECso of 215.6 mg/L
PFOA, which was acceptable for quantitati\ e use
Ding et al. (2012a) conducted a 48-hour static, purl in lly measured acute test on PFOA
(CAS # 335-67-1; W,> purity from Sigma Aldrich) with/), magna. The test generally followed
OECD test guideline 2< >2 (2<)i)4) Authors employed four replicates of five daphnids each in six
test concentrations plus a neuati\ e control Nominal concentrations were 0 (negative control),
144.9. l(o 186.3, 207.0. 227 7, and 248.4 mg/L. Concentrations of PFOA were confirmed in
the highest and lowest concentrations, though only nominal concentrations were reported. It was
stated that the verified concentrations were "well in line with nominal concentrations". The 48-
hour ECso was reported as 211.6 mg/L (95% C.I. = 184.7 - 255.5 mg/L). EPA performed C-R
analysis for the test. The EPA-calculated ECso was 216.1 mg/L PFOA, which was acceptable for
quantitative use.
63
-------
Lu et al. (2016) evaluated the acute toxicity of PFOA (CAS# 335-67-1, 98% purity,
purchased from Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) on Daphnia magna
immobilization. The test was conducted following modified OECD standard test procedure 202,
whereby five concentration treatments (3, 10, 30, 100 and 300 mg/L) plus a blank control were
employed with three replicates per treatment. Authors reported immobility/survival to be a more
sensitive endpoint than survival alone. The author-reported 48-hour IX "so for immobility/survival
was 110.7 mg/L and the EPA-calculated 48-hour ECso was 1 14.6 niu I.. u liich was acceptable
for quantitative use.
Yang et al. (2019) evaluated the acute effects of PI OA (CAS# 335-67-1. purchased from
Sigma-Aldrich in St. Louis, MO) on Daphnia magna in a 48-hour unmeasured static exposure.
Authors stated the protocol for all testing followed OF.CD (iuideline 202. Nominal acute test
concentrations included 0 (control), 66.67, 7l) l)2. <>(¦>. 115 1. 138.3, and 166.0 mg/L PFOA,
with four replicates per treatment Authors reported a LC;.. of 120.9 mg/L PFOA. The EPA-
calculated 48-hour I.Cs" was 117 2 niu L, which was acceptable for quantitative use.
Quantitatively acceptable IK magna acute values from Boudreau (2002; ECso = 223.6
mg/L). Colombo et al (2<)()S. I ¦;("*¦¦ 480 mg/L), Ji et al. (2008; ECso = 542.5 mg/L), Li et al.
(200l>. I¦:(¦«„ 220.8, 157.9. and 207 3 mg/L), Yang et al. (2014; LCso = 222.0 mg/L), Barmentlo
et al. (2015. I215 6 niu I.), Ding et al. (2012a; ECso =216.1 mg/L), Lu et al. (2016; ECso
=114.6 mg/L), and Yang et al. (2019; LCso = 117.2 mg/L) were taken together as a geometric
mean value to calculate the D. magna SMAV of 220.0 mg/L. The D. carinata SMAV (i.e., 66.80
mg/L), D. pulicaria SMAV (i.e., 203.7 mg/L), and/), magna SMAV (i.e., 220.0 mg/L) were
used to determine the Daphnia GMAV of 144.1 mg/L.
64
-------
3.1.1.1.3 Third most acutely sensitive genus: Brachionus (rotifer)
Zhang et al. (2013a) performed a 24-hour static test of PFOA (CAS # 335-67-1, 96%
purity) with Brachionus calyciflorus. Organisms were neonates less than two hours old at test
initiation. All animals were parthenogenetically-produced offspring of one individual from a
single resting egg. PFOA concentrations were not measured in the rotifer exposures, but rather,
in a side experiment that showed that the concentration of PFO A measured every eight hours
over a 24-hour period in rotifer medium with green algae inclined minium I change. C-R data
were not included in the publication and could not be obtained for independent calculation of the
test-specific LCso. Therefore, the study reported 24-hour LC so of 150.0 mg/L was considered
acceptable for quantitative use. No other quantitatively acceptable acute toxicity data were
available for Brachionus calyciflorus or other members of the uenus Brachionus. Therefore, the
LCso (i.e., 150.0 mg/L) from this test served directly as the Brachionus calyciflorus SMAV and
the Brachionus GM AV.
3.1.1.1.4 Fourth most acutely sensitive genera: Ligumia (mussel)
Hazelton el al. (2012. 2013) e\ aluated the acute effects of PFOA (96% purity) on the
freshwater mussel. J.i^umia recta Acute toxicity was observed under static conditions over a 24-
hour period ( 24-hour old ulochidia) or renewal conditions over a 96-hour period (four to six
week-old ju\ eniles). Authors stated the tests followed ASTM E2455-06 (2006). Measured test
concentrations of PI - O A were within 10% of target in water from acute tests. Mortality of
mussels in the negali\ e control was <10% in all exposures. C-R data were not reported in the
publication and could not be obtained for independent calculation of EC50 values from these
tests. The 24-hour EC50 reported forZ. recta glochidia. was 161.0 mg/L (95% C.I. 135.0-192.7
mg/L). The 96-hour LC50 value for the juvenile L. recta was greater than the highest test
concentration (500 mg/L). The juvenile life stage was determined to be relatively tolerant to
65
-------
acute PFOA exposures and, therefore, the LCso from the acute juvenile mussel test was not used
to derive the acute PFOA criterion. No other quantitatively acceptable acute toxicity data were
available for Ligumia recta or other members of the genus Ligumia. Therefore, the EC so (i.e.,
161.0 mg/L) from this test served directly as the Ligumia recta SMAV and the Ligumia GMAV
(the fourth most sensitive GMAV).
3.1.1.1.5 Missing Insect MDR
The acute data set for PFOA contained 18 genera (Tahlc 3-.i) representing seven of the
eight taxonomic MDR groups. The missing MDR was a representative from an insect family.
The EPA examined data in Appendix G to determine if any qualitatively accepliihle freshwater
PFOA toxicity studies could be used to evaluate the missing MDR group. Yang el al. (2014)
conducted a 96-hour renewal, measured I'l OA acute test with the midge, Chironomusplumosus,
which is described in greater detail in Appendix G2 I 5 The source of the test organisms was
the Beijing City Big Forest Flower Market, which potentially was a problematic source given
uncertainties associated with prc\ ions exposures to PFOA. Consequently, this study was not
considered acceptable lor quanlilali\ e use but was considered qualitatively by providing relative
species sensitivity information The reported LCso was 402.24 mg/L PFOA indicating this insect
species may not be one of the more sensitive taxonomic groups (Figure 3-1). EPA will continue
to seek additional acute PFOA insect data to further evaluate the sensitivity of insects. PFOA
insect toxicity testing is an active and ongoing area of research within the ecotoxicological
scientific community that will likely provide additional information to fully evaluate on the
sensitivity of insects to acute PFOA exposures before the PFOA criteria document is finalized.
EPA calculated the PFOA Criterion Maximum Concentration (CMC) using all acceptable
quantitative studies from Appendix A, but did not include the insect data in the criterion
66
-------
calculation (i.e., the relatively tolerant insect LCso value was not included in the total count ("n")
of Genus Mean Acute Values in the criterion calculation).
Table 3-3. Ranked Freshwater Genus Mean Acute Values.
Rank"
(i.MAV
(ing/l. PI-OA)
MI)K
Group'-''
Genus
Species
S.M.W h
(ing/l. PI-OA)
1
i>3 17
1)
('Imloriis
CkukKvian.
Chydorus spin icriciis
93.17
2
144.1
D
Daphnia
Cladoceran,
Daphnia carinata
66.80
Cladoceran,
Daphnia magna
220.0
Cladoceran,
Daphnia pulicaria
203.7
3
150.0
H
Brachionus
Roliler.
Brachioi///.v cat vciflorus
150.0
4
161.0
G
Ligumia
Black sandshdl,
/ igumia recta
161.0
5
164.4
G
Lampsilis
l-'almucket,
/ ampsilis si !i quo idea
164.4
6
166.3
1)
.\ loina
Cladoceran,
Moina macrocopa
166.3
7
377.0
(
Xcnopiis
Frog,
Xenopus sp.
377.0
8
3S3 6
II
/ Uigcs/a
Planaria,
Dugesia japonica
383.6
9
413 2
li
I'tmephales
Fathead minnow,
Pimephales promelas
413.2
10
43 1 5
i:
Neocaridina
Green neon shrimp,
Neocaridina denticulata
431.5
11
572.4
li
Danio
Zebrafish,
Danio rerio
572.4
12
646.2
C
Hyla
Gray treefrog,
Hyla versicolor
646.2
13
664.0
B
Lepomis
Bluegill,
Lepomis macrochirus
664.0
14
681.1
G
Physella
Bladder snail,
Physella acuta
681.1
15
689.4
C
Ambystoma
Jefferson salamander,
Ambystoma jeffersonianum
1,070
67
-------
Uiink"
(i.MAV
(ing/l. PI-OA)
MI)K
Group'-''
(iOllllS
Species
S.M.W h
(ing/l. PIOA)
Small-mouthed salamander,
Ambystoma texanum
407.3
Eastern tiger salamander,
Ambystoma tigrinum
752.0
16
793.9
C
Anaxyrus
American toad,
Anaxyrus americanus
793.9
17
951.5
C
Lithobates
American bullfrog,
Lithobates catesbeiana
1,020
Green frog,
A ithobates da mite ii is
1,070
Northern leopard frog.
A ithobates pi pi ens
751.7
Wood frog,
A ithobates sy/vatica
999
18
1,682
A
Oncorhynchus
Raiilium trout,
()ih ¦( >rh\ i K hus my kiss
1,682
a Ranked from the most sensitive to the most resistant based on Genus Mean \cute Value,
b From Appendix A: Acceptable Freshwater Acule PI 'O-X Toxicity Studies
c MDR Groups - Freshwater:
A. the family Salmonidae in the class Osleiclillis cs.
B. a second family in the class Osteichthycs. preferable a commercially or recreationally important
warmwaler species (e.g.. blucgill. channel catfish, etc.)
C. a third family in the phylum Chordata (may be in the class Osteichthycs or may be an amphibian, etc.)
D. a planktonic crustacean (e.g.. cladoccran. copcpod. etc.)
E. a bcnthic crustacean (e.g.. ostracod. isopod. amphipod. crayfish, etc.)
F. an insect (e.g.. mayfly, dragonfly, damsclfly. stonefly. caddisfly, mosquito, midge, etc.)
G. a family in a phylum other than Arthropoda or Chordata (e.g., Rotifera, Annelida, Mollusca, etc.)
H. a family in any order of insect or any phylum not already represented.
d Of the available qualitatively-acceptable insect data, only Yang et al. (2014) conducted a test for the standard 96
hour acute exposure. Other qualitatively acceptable insect toxicity data were based on either chronic or sub-
chromc exposure durations Yaim el al. (2014) specifically conducted a 96-hour renewal test with measured PFOA
concentrations on the midge, < 'hinmomus plumosus. This study was not acceptable for quantitative use due to the
potential problematic source of I lie organisms but was retained for qualitative use. The reported LC5o was 402.24
mg/L PFOA. PFOA insect to\icil> testing is an active and ongoing area of research within the ecotoxicological
scientific commiimix i hat will likely provide additional information to fully evaluate on the sensitivity of insects
to acute PFOA exposures before llie PFOA criteria document is finalized.
68
-------
s
3
•w
s
S3
si
a
Z
-------
Interspecies Correlation Estimation (WeblCE) tool (Raimondo et al. 2010). These benchmarks
are provided in Appendix L.
The following section provides information and summaries of studies for the sensitive
estuarine/marine taxa, based on the limited available data (Table 3-4). Study summaries for the 4
most sensitive genera are provided below.
Table 3-4. Estuarine/Marine Acute PFOA Genera.
Ranked Below from Most to Least Sensitive
(i.MAV
Uank
(ienus
(mg/L)
Species
Comment
NotaNorlh American resident
1
Siriella
15.5
Mysid
(,Siriella anna la)
species, but a member of the
Mysidae Family and serves as a
surrogate for untested mysid
species residing in North America.
Mediterranean mussel
2
Mytilus
17.58
O Iv/ihis
galloprovincialis)
North American resident species
3
StrongvloLvniroiiis
2<~) 63
Purple sea urchin
(Slrongylocenii < >iiis
purpnralns)
North American resident species
4
American lis is
24
Mysid
(. hiicricaiiivsis bahia)
North American resident species
3.1.1.2.1 Most sensitive estuurine marine genus: Siriella (mysid)
Mhadhbi et al. (2012) performed a 96-hour static, unmeasured acute test with PFOA
(96% purity) oil l he mysid. Siriella armata. Mysids were exposed to one of ten nominal PFOA
treatments (0.1. n 5. 1. 2. 5. I <). 20, 30, 40 and 80 mg/L). Neonates were fed 10-15 Artemia
salina nauplii daily and mortality was recorded after 96 hours. The 96-hour LCso reported in the
study was 15.5 mg/L PFOA and was acceptable for quantitative use. No other quantitatively
acceptable acute toxicity data were available for Siriella armata or other members of the genus
Siriella. Therefore, the LCso (i.e., 15.5 mg/L) from this test served directly as the Siriella
70
-------
GMAV. Although S. armata is not a North American resident species, it is a member of the
Mysidae Family and serves as a surrogate for untested mysid species residing in North America.
3.1.1.2.2 Second most sensitive estuarine/marine genus: Mytilus (mussel)
The acute toxicity of PFOA (purity not provided) on the Mediterranean mussel, Mytilus
galloprovincialis, which occurs in California and other parts of the Pacific Northwest (Green
2014), was evaluated by Fabbri et al. (2014). The endpoint was the percent reduction of normal
D-larvae in each well. Authors noted that controls had >80% normal l)-lar\ ae across all tests,
meeting the >75% acceptability threshold outlined by ASTM (2004). PFOA was only measured
once in one treatment which was similar to the nominal concentration. The percentage of normal
D-larva decreased with increasing test concentrations The \IOF,C and LOEC reported for the
study were 0.00001 and 0.0001 mg/L, rcspecti\ ely Although authors report -27% effect at the
LOEC (i.e., 0.0001 mg/L), the test concentrations tailed to elicit 5<»"<» malformations in the
highest test concentration, and an F.Cso was not determined Therefore, the EC so for the study
was greater than the highest test concentration (1 mg/L). The 48-hour ECso based on
malformation of >1 mg I. was <.|iianlilali\ely acceptable.
Hayman et al. (2021) reported the results of a 48-hour static, measured acute PFOA
(CAS # 335-67-1, 95% pui it\. purchased from Sigma-Aldrich, St. Louis, MO) test on the
Mediterranean mussel, Mytilus galloprovincialis. Authors note that tests followed U.S. EPA
(1995b) and ASTM (2004) protocols. Six test solutions were made in 0.45 |im filtered seawater
(North San Diego Bay, CA) with PFOA dissolved in methanol. The highest concentration of
methanol was 0.02% (v/v) and each treatment solution contained five replicates. At test
termination (48 hours), larvae were enumerated for total number of larvae that were alive at the
end of the test (normally or abnormally developed) as well as number of normally-developed (in
the prodissoconch "D-shaped" stage) larvae. There were no significant differences between
71
-------
solvent control and filtered seawater, suggesting no adverse effects of methanol. The author
reported 48-hour ECso, based on normal survival larvae was 9.98 mg/L PFOA. The EPA-
calculated 48-hour EC so value was 17.58 mg/L, which was acceptable for quantitative use.
Although the 48-hour EC so based on malformation of >1 mg/L from Fabbri et el. (2014)
met EPA's data quantitatively objectives, it was not used directly in the calculation of theM
galloprovincialis SMAV because it was a "greater than LCso" \ aluc and a definitive LCso was
available for the same species as reported by Hayman et al. (2021). The definitive LCso value
from Hayman et al. (2021) of 17.58 mg/L served directly as the Mytilus galloprovn/cialis SMAV
and as the Mytilus GMAV.
3.1.1.2.3 Third most sensitive estuarine/marine genus: Sirongylocentrotus (urchin)
Hayman et al. (2021) reported the results of a 96-hour static, measured PFOA (CAS #
335-67-1, 95% purity, purchased from Sigma-Aldrich, St. Louis, MO) test with the purple sea
urchin, Strongylocentrotuspurpuratus. Authors note that tests followed U.S. EPA (1995b) and
ASTM (2004) protocols. Six test solutions were made in 0.45 |im filtered seawater (North San
Diego Bay, CA) with PFOA dissolved in methanol and each treatment was replicated five times.
The highest concentration of methanol was 0.02% (v/v). At test termination (96 hours), the first
100 larvae were enumerated and observed for normal development (organisms distinguished as
being in the four arm pluteus stage). There were no significant differences between solvent
control and filtered seawater, suggesting no adverse effects of methanol. The author reported 96-
hour ECso, based on normal development, was 19 mg/L PFOA. The EPA-calculated 96-hour
ECso value was 20.63 mg/L, which was acceptable for quantitative use. The EC50 value of 20.63
mg/L was the only acceptable acute value for Strongylocentrotus purpuratus or any members of
the genus Strongylocentrotus. Therefore, it served directly as the Strongylocentrotus purpuratus
SMAV and the Strongylocentrotus GMAV.
72
-------
3.1.1.2.4 Fourth most sensitive estuarine/marine genus: Americamysis (mysid)
Hayman et al. (2021) conducted a 96-hour static, measured test to assess effects of
PFOA (CAS # 335-67-1, 95% purity, purchased from Sigma-Aldrich, St. Louis, MO) on the
mysid, Americamysis bahia. Authors note that tests followed U.S. EPA (2002) protocols. Six test
solutions were made in 0.45 |im filtered seawater (North San Diego Bay. CA) with PFOA
dissolved in methanol. The highest concentration of methanol was 0.02% (v/v) and each test
solution was replicated six times with five mysids per replicate. There were no significant
differences between solvent control and filtered seawater, suggesting no adverse effects of
methanol. No organisms were found dead in the controls at test termination. EPA was unable to
fit a concentration-response model with significant parameters and relied on the author-reported
96-hour LCso of 24 mg/L PFOA as the quantitatively acceptable acute value. The LC50 value of
24 mg/L was the only acceptable acute value for Americamysis bahia or any members of the
genus Americamysis. Therefore, it served directly as the Ann 1 icamysis bahia SMAV and the
Americamysis GMAY
The esliiaiine murine acute data set lor PFOA contained four genera (Figure 3-2)
representing only three of the eight taxonomic MDR groups. The missing MDR groups included
two families in the phylum Chordata. two families in a phylum other than Chordata, and any
other family not already represented (Table 3-5). As noted above, EPA used the available acute
toxicity data and ORD's peer-reviewed web ICE tool to develop aquatic life benchmarks for
consideration by states and tribes (see Appendix L).
73
-------
Table 3-5. Ranked Estuarine/Marine Genus Mean Acute Values.
GMAV
MDR
SMAVb
Rank3
(mg/L PFOA)
Group0
Genus
Species
(mg/L PFOA)
1
15.5
C
Siriella
Mysid,
Siriella armata
15.5
2
17.58
D
Mytilus
Mediterranean mussel,
Mytilus galloprovincialis
17.58
3
20.63
F
Strongylocentrotus
Purple sea urchin,
Strongylocentrotus
purpuratus
20.63
4
24
C
Americamysis
Mysid,
Americamysis bahia
24
a Ranked from the most sensitive to the most resistant based on Genus Mean Acute Value,
b From Appendix B: Acceptable Estuarine/Marine Acute PFOA Toxicity Studies
c MDR Groups identified in Footnote C of Table 3-1.
1.0
0.9
a
.2 0.:
0.7 --
C3
£
a<
>
« 0.6
3
s
3 0.5
g 0.4
X
S 0.3 +
a
ai
S 0.2
0.1
0.0
I Invertebrate (Other)
~ Invertebrate (Mollusk)
Americamysis
Strongylocentrotus
Mytilus
Siiiella
10
Genus Mean Acute Value (mg/L PFOA)
Figure 3-2. Acceptable Estuarine/Marine GMAVs.
100
74
-------
3.1.1.3 Summary of Chronic PFOA Toxicity Studies Used to Derive the Freshwater Aquatic
Life Criterion
Acceptable chronic PFOA toxicity data in freshwater were available for a total of 11
species representing 10 genera and nine families in three phyla. A total of seven of the eight
required MDRs were fulfilled. Data for one MDR group (any other phylum not already
represented or a second insect order not already represented) remains unfulfilled. PFOA insect
toxicity testing is an active and ongoing area of research within the ecotoxicological scientific
community that will likely provide information to evaluate the sensitivity of insects to chronic
PFOA exposures before the PFOA criteria document is finalized. The following section provides
information and summaries of studies for the sensitive taxa with effect values used in the
quantitative calculation of the chronic freshwater PFOA criterion (Table 3-6). Study summaries
for the four most sensitive genera are provided below to describe all sensitive genera up to the
point where four established North American species are represented.
Table 3-6. The Most Sensitive Genera Used in Calculating the Chronic Freshwater
Criterion (Sensitivity Rank 1-4).
Ranked Below from Most to Least Sensitive
Rank
Genus
GMCV
(mg/L)
Species
Comment
1
Hyalella
0.147
Amphipod
(Hyalella azteca)
North American resident
species
2
Lithobates
0.288
American bullfrog
(Lithobates catesbeiana)
North American resident
species
3
Daphnia
0.3700
Cladoceran
(Daphnia carinata)
Not a resident species
Cladoceran
(Daphnia magna)
North American resident
species
4
Brachionus
0.7647
Rotifer
(Brachionus calyciflorus)
North American resident
species
3.1.1.3.1 Most chronically sensitive genus: Hyalella (amphipod)
Bartlett et al. (2021) evaluated the chronic effects of PFOA (CAS# 335-67-1, 96%
purity, solubility in water at 20,000 mg/L, purchased from Sigma-Aldrich) on Hyalella azteca
75
-------
via a 42-day static-renewal, measured study. Methods for this study were adapted from
Borgmann et al. (2007), and organisms were two to nine days old at the test initiation. A 100
mg/L stock solution was prepared to yield measured test concentrations of 0 (control), 0.84, 3.3,
8.9, 29 and 97 mg/L PFOA. Two separate tests were performed with five replicates per
concentration and 20 amphipods per replicate. At test termination (day 42), adults were sexed
and weighed, as well as their young counted. The 42-day author-reported I .Cio value for survival
was 23.2 mg/L PFOA. The author-reported ECio values for growth and reproduction were 0.160
mg/L and 0.0265 mg/L, respectively. EPA only performed C-R analysis for the growth and
reproduction-based endpoints for this test, given the apparent tolerance of the sur\ i\ a I-based
endpoint. EPA calculated ECio values for the 42-day growth endpoint (i.e., control normalized
wet weight/amphipod) and the 42-day reproduction endpoint (i e . number of juveniles per
female). The 42-day growth-based ECio of " 4SS mu I. was not selected as the primary endpoint
from this test because it was more tolerant than the reproduction-based ECio of 0.147 mg/L,
which was acceptable lor <.|ualitati\ e use. The ECio value of 0.147 mg/L was the only acceptable
chronic value for llyalclla ml cat or any members of the genus Hyalella. Therefore, it served
directly as the llyalclla az/eca SMCV and the Hyalella GMCV.
3.1.1.3.2 Second most chronically sensitive genus: Lithobates (frog)
Flynn et al. (2019) e\ aluated the chronic effects of PFOA (CAS# 335-67-1, purchased
from Sigma-Aldrich) on the American bullfrog (Lithobates catesbeiana, formerly, Rana
catesbeiana) during a 72-day static-renewal unmeasured exposure. The authors tested a negative
control and two treatment concentrations (i.e., 0.144 and 0.288 mg/L), which were the only three
PFOA-only treatments within the larger factorially-designed experiment. Each treatment
contained 10 tadpoles (Gosner stage 25) and treatments were replicated four times. On day 72 of
the experiment, all tadpoles were euthanized and measured (snout vent length and mass). The
76
-------
most sensitive chronic endpoint was growth (snout-vent length), with a 72-day NOEC and LOEC
of 0.144 mg/L and 0.288 mg/L, respectively. EPA could not independently calculate an ECio
value because there were minimal effects observed across the limited number of treatment
concentrations tested. Consequently, EPA used the LOEC of 0.288 mg/L as the chronic value
from this chronic test. The LOEC was used preferentially to the MATC from this test because a
-7% reduction in snout-vent length relative to control responses was observed at the LOEC (i.e.,
0.288 mg/L), which is a similar effect level to the chronic 10% effect level (i.e., ECio) used
preferentially to derive the chronic criterion. The LOEC value of 0.288 mg/L was the only
acceptable chronic value for Lithobates catesbeiana or any members of the genus Li I ho hates.
Therefore, it served directly as the Lithobates catesbeiana SMC V and the Lithobates GMCV.
3.1.1.3.3 Third most chronically sensitive iyims: Daphnia (dadoceran)
Logeshwaran et al. (2021) conducted a PFOA (95% purity, purchased from Sigma-
Aldrich Australia) chronic toxicity test with the cladoceran, Daphnia carinata. Authors stated the
chronic test protocol followed OECD guidelines (2012). Authors tested a negative control and
five PFOA concentrations (i.e., 0.001, 0.01, 0.1, 1.0 and 10 mg/L PFOA). Each test treatment
was replicated 10 times with one daphnid (six tol2 hours old) per treatment. At test termination
(21 days) test endpoints included survival, days to first brood, average offspring in each brood
and total live offspring. No mortality occurred in the controls or lowest test concentration. Of the
three endpoints measured, average offspring in each brood and total live offspring were the more
sensitive endpoints with 21-day NOEC and LOEC values of 0.01 and 0.1 mg/L PFOA,
respectively. EPA was unable to calculate statistically robust ECio estimates from C-R models
for these endpoints, largely because of the 10X dilution series across five orders of magnitude.
The LOECs for these endpoints were not selected as the chronic value because the LOECs
produced a 29.23% reduction in the average number of offspring per brood relative to controls
77
-------
and a 39.89% reduction in the total living offspring relative to controls. Therefore, the MATC
(i.e., 0.03162 mg/L) was selected as the quantitatively acceptable chronic value form this test.
The MATC value of 0.03162 mg/L was the only acceptable chronic value for Daphnia carinata
and it served directly as the Daphnia carinata SMCV.
Colombo et al. (2008) conducted a 21-day renewal measured chronic test on PFOA with
the daphnid, Daphnia magna. Authors stated the toxicity test was conducted followed OECD test
guideline 211. Average number of live young was the most sensitive end point reported by
Colombo et al. (2008), with a NOEC of 20 mg/L. Based on the author-reported l-X'so for the
average number of live young, the LOEC was 44 2 niu'I. and the MATC was 2^ 73 mg/L. EPA
performed C-R analysis for each reported endpoint. The most sensitive endpoint with an
acceptable C-R curve was average number of li\ e young, with an l-PA-calculated ECio of 20.61
mg/L PFOA and was acceptable for quantitative use
Ji et al. (200S) conducted a chronic life-cycle test on l he effects of PFOA with Daphnia
magna. Authors slated thai the IK magna test followed OECD 211 (1998). The most sensitive
endpoint for D. magna reported in the publication was days to first brood with a 21-day NOEC
of 6.25 mu l.(LOLC 12 5 mu I.: MATC = 8.839 mg/L); however, number of young per
starting female (an endpoint not reported in the publication, which only assessed number of
young per sui \ i \ inu female) was calculated by EPA and considered to be a more sensitive
endpoint with an I ¦ PA-calculated ECio of 7.853 mg/L. Therefore, the EPA-calculated ECio of
7.853 mg/L PFOA for D. magna (number of young per starting female) was considered
quantitatively acceptable.
Li (2010) conducted an unmeasured chronic life cycle 21-day test on the effects of PFOA
on Daphnia magna. Authors stated the test followed OECD 211 (1998). The D. magna 21-day
78
-------
NOEC (reproduction as number of young per female, broods per female, and mean brood size)
was 10 mg/L (LOEC = 32 mg/L; calculated MATC = 17.89 mg/L). EPA performed C-R analysis
for each reported endpoint. EPA also revaluated all endpoints that were based on number of
surviving females to be based on the number of starting females. This recalculation was done
with the intent to account for starting females that were unable to contribute to the population as
reproduction/female due to mortality. The most sensitive endpoini with an acceptable C-R curve
was the number of young per starting female with an EPA-calculated IX'i- of 12.89 mg/L PFOA
and was acceptable for quantitative use.
Yang et al. (2014) evaluated the chronic 21 -day renew nl. measured test of PI OA with
Daphnia magna, following ASTM E729 (1993). The author-reported/). magna 21-day ECio for
reproduction (total number of spawning) was 7 <)2 mgl. EPA performed C-R analysis for each
reported endpoint. Both chronic survival and reproduction endpoints resulted in acceptable C-R
curves. The EPA-ca I ciliated l:X'\- lor reproduction as total number of spawning events was 6.922
mg/L, similar to the I X'i- reported In the authors (i e , 7.02 mg/L). Chronic survival was more
sensitive than reproduction, with an NW-calculaled ECio of 5.458 mg/L PFOA. Therefore, the
survi\ al based I-X"i¦ ¦ calculated In N\\ (i e . 5 45S mg/L) was acceptable for quantitative use.
I.ii ol al. (2016) evaluated the chronic toxicity of PFOA (CAS# 335-67-1, 98% purity,
purchased from Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) on Daphnia magna
immobilization, growth and reproduction in a 21-day semi-static test with unmeasured treatment
solutions. Authors stated the test protocol followed OECD Test Method 211. Neonates (<24 h
old) were exposed to PFOA in one of six PFOA treatments (i.e., 0 [control] 0.032, 0.16, 0.8, 4
and 20 mg/L), with 20 replicates for each treatment. The 21-day growth and reproductive NOEC
and LOEC values were 0.032 and 0.16 mg/L PFOA, respectively. EPA was unable to fit a C-R
79
-------
model with significant parameters to the chronic data associated with reproduction from this test.
The EPA-calculated EC 10 values for mean intrinsic rate of increase (r) and growth (as length)
were 0.0173 mg/L and 0.0124 mg/L, respectively. Both ECio values were nearly two times lower
than the NOEC of 0.032 mg/L and four times lower than the LOEC value (i.e., 0.16 mg/L) where
only 15.2% and 11.9% reductions in intrinsic rate of natural increase (r) and length were
observed, respectively. As a result, the MATC of 0.0716 mg/L for growth and reproduction was
selected as the most appropriate chronic value for quantitative use lo in deri\ ing the chronic
water column-based criterion.
Yang et al. (2019) evaluated the chronic effects of PI OA (CAS# 335-07-1. purchased
from Sigma-Aldrich in St. Louis, MO) on Daphma nni^iui \ ia a 21 -day unmeasured, static-
renewal test that assessed reproductive el'lccls Protocol for testing followed OECD Guideline
211. Authors tested a negative control and lour PI OA treatment concentrations (6.708, 10.10,
15.11, and 22.61 mg I.), with each treatment replicated l<> times and each replicate containing
one neonate (12-24 hours old) in a I mi ml. glass beaker. The reproductive NOEC and LOEC
values were 6 70S and I <> I <> mg I. PI OA. respectively. EPA performed C-R analysis for the test.
The Ll\\-calculaled l-Ci.. bused on mean offspring at 21-days as a proportion of the control
response was S 0S4 mg/L and was used quantitatively to derive the draft chronic water column
criterion.
The chronic \ allies from Colombo et al. (2008; ECio = 20.61 mg/L; endpoint = average
number of live young), Ji et al. (2008; ECio = 7.853 mg/L; endpoint = number of live young per
starting female), Li (2010; ECio =12.89 mg/L; endpoint = number of young per starting female),
Yang et al. (2014; ECio = 5.458 mg/L; endpoint = survival), Lu et al. (2016; MATC = 0.0716
mg/L; endpoint = length and rate of natural increase), and Yang et al. (2019; ECio = 8.084 mg/L;
80
-------
endpoint = mean offspring as a proportion of control response) were taken together as a
geometric mean to serve as the Daphnia magna SMCV (i.e., 4.330 mg/L). The D. carinata
SMCV (i.e., 0.03162 mg/L) and the D. magna SMCV (i.e., 4.330 mg/L) were used to calculate
the Daphnia GMCV of 0.3700 mg/L.
3.1.1.3.4 Fourth most chronically sensitive genus: Brachionus (rotifer)
Zhang et al. (2013a) conducted a chronic life-cycle renewal lest of PFOA (CAS # 335-
67-1, 96% purity) with Brachionus calyciflorus. The test consisted of a negative control and four
PFOA concentrations (0.25, 0.5, 1.0, 2.0 mg/L PFOA). For each treatment le\ el. fifteen amictic
rotifers were placed individually into culture plate wells containing two mL of test solution that
was renewed daily. Numbers of eggs produced and starting rotifer lifetimes were recorded for
every individual, and the test was conducted until every starting rotifer from every treatment
level died, which occurred around 200 hours after test initiation. Data from this test were used to
construct survivorship and fertility tables using conventional life-history techniques, which were
used to calculate net reproducti\ e rate, generation time, and intrinsic rate of natural increase.
EPA calculated ECios from C-R data reported in the publication, and the most sensitive endpoint
with an acceptable C-R air\ e was the intrinsic rale of natural increase, with an ECio of 0.5015
mg/L. I'l OA
The intrinsic rate of natural increase (d"1) is a population level endpoint that accounts for
births and deaths o\ er time In Zhang et al. (2013a), the intrinsic rate of natural increase was
calculated as the natural log of the lifetime net reproductive rate for all individuals within a
population (defined here as a PFOA treatment level) divided by the average generation time of
those individuals.
The EC 10 calculated for the intrinsic rate of natural increase was similar to the ECio value
for average net reproductive rate (0.514 mg/L). Zhang et al. (2013a) also reported significant
81
-------
reductions in egg size, with an EPA-calculated ECio = 0.193 mg/L. However, this endpoint
displayed a relatively poor concentration response relationship and may not be relevant for
assessing population level effects. For these reasons, it was not selected as the primary effect
concentration from this study. Zhang et al. (2013a) also reported effects to average juvenile
period, which was a relatively tolerant endpoint. Juvenile period decreased with increasing
exposure concentration, with the average juvenile period being ahout 16% faster than the control
responses in the highest treatment concentration (2.0 mg/L). Effects lo chronic apical endpoints
in this publication and Zhang et al. (2014) generally appear as a threshold effect from 0.25 mg/L
to 1.0 mg/L, providing further support for selection of the F.C i" value (i.e., 0.5015 mu/L) based
on rate of natural increase as the primary chronic value for quantitative use from Zhang et al.
(2013a).
In addition to the life cycle exposure /lianu el al (2<> 13a) also conducted a second multi-
generational 28-day study lo measure effects of PI-OA on growth patterns, population density,
and population dynamics The 2S-day test consisted of a negative control and two PFOA
concentrations 25. and 2 n mu I. PI OA) Population densities were lower than controls at both
PFOA treatment le\els. ho\\c\er. because this study was limited to two treatment levels, it was
considered to he of secondary importance compared to the life-cycle test.
Zhang et al. (2014h) describes the results of three experiments involving Brachionus
calyciflorus exposures to PTOA (CAS # 335-67-1, 96% purity). The effects of PFOA
concentration on mictic ratios of B. calyciflorus was examined by placing individual neonates in
culture plate wells with two mL of medium containing of two PFOA concentrations (0.25 mg/L
and 2.0 mg/L) plus a control. Each treatment level, as well as the control, was replicated three
times. All eggs produced by these exposed individuals were individually incubated in culture
82
-------
wells with 1 mL control medium. The mature Fi offspring were subsequently identified as
producing mictic or amictic eggs, and these data were used to calculate mictic ratios. The
proportions of mictic eggs increased with increasing PFOA concentration (0.56 - control, 0.72 -
0.25 mg/L, 0.75 - 2.0 mg/L), and the results were statistically significant (p<0.05). In contrast,
mictic ratios were not affected by PFOA concentrations in Zhang et al. (2013a). Because of the
inconsistent result in the mictic ratio endpoint between Zhang el ill (2<>13a) and Zhang et al.
(2014), it was not selected as the representative endpoint from either publication.
The effects of PFOA concentration on resting egg production of B. calvci/loriis was
examined by exposing rotifers to one of five PFOA concentrations (plus control) in the dark for
six days. Resting eggs were collected on the sixth day and then hatched in control medium 30
days later. Resting egg production decreased with increasing PI'OA concentration. The EPA-
calculated ECio calculated from C-R data reported in I'iuinv I of Zhang et al. (2014b) was 0.076
mg/L. Because there was only one replicate (as implied In lack of error bars in Figure 1 of the
publication, no clear description of replicates in the methods section, and no author-reported
statistical analysis of lliis cndpoinl). resting egg production from this study was not considered
quantitati\ ely acceptable hut was retained lor qualitative use. In a second resting egg exposure
stud). resting eggs were produced under control conditions, then allowed to hatch while exposed
to one of five PI OA concentrations plus a control. In this study, the effects of PFOA exposure
on resting egg hatching rate were not statistically significant.
Finally, the effects of PFOA concentration on B. calyciflorus population growth was
examined during a four-day study in which 10 neonates were placed into chambers with 10 mL
of medium containing one of eight PFOA concentrations, plus a control. Each treatment level, as
well as the control, was replicated at least six times. After four days, the total numbers of rotifers
83
-------
in each chamber were counted, and these data were used to calculate the intrinsic rate of natural
increase (d"1), the most sensitive acceptable endpoint from this study, with an EPA-calculated
ECio of 1.166 mg/L. Beyond Zhang et al. (2013a) and Zhang et al. (2014b), no other
quantitatively acceptable chronic tests were available for Brachionus calyciflorus. The EPA-
calculated ECio values from Zhang et al. (2013a) (i.e., 0.5015 mg/L; endpoint = rate of natural
increase) and Zhang et al. (2014b) (i.e., 1.166 mg/L; endpoint ~ rale of natural increase) were
taken together as a geometric mean to serve as the Brachionus calyciflorus SMCV (i.e., 0.7647
mg/L). No other quantitatively acceptable chronic toxicity data were available lor other members
of the genus Brachionus and the Brachionus calyciflorus SMCV (i.e., 0.7647 mg I.) served
directly as the Brachionus GMCV.
3.1.1.3.5 Missing Minimum Data Ret/uiiriih.-ius
The chronic data set for PFOA based on quantilali\ ely acceptable data contains 10 genera
representing seven of the eight taxonomic MDR groups (Table 3-7). The MDR group missing is
the any other phylum not already represented, or a second insect order not already represented.
The EPA examined data in Appendix (i and did not identify any qualitatively acceptable studies
with species that could be used to inform the relative sensitivity of the missing MDR group. Two
qualitati\ ely acceptable chronic sub-chronic tests withDugesiajaponica, a species that falls into
the missing MDR category. were available to inform the potential sensitivity of the missing
MDR. Overall, these two tests (Yuan et al. 2016b, Yuan et al. 2017) did not measure apical
effects associated with growth, survival and reproduction. Therefore, the qualitatively acceptable
data provided by these studies did not provide information about the relative sensitivity of this
species. Because nearly all MDRs were met, EPA derived the chronic freshwater column
criterion based on the genus sensitivity distribution (GSD) of the 10 genera representing seven
MDRs. EPA will continue to seek additional chronic PFOA insect data to further evaluate the
84
-------
sensitivity of insects. PFOA insect toxicity testing is an active and ongoing area of research
within the ecotoxicological scientific community that will likely provide information to evaluate
the sensitivity of insects to chronic PFOA exposures before the PFOA criteria document is
finalized.
Table 3-7. Ranked Freshwater Genus Mean Chronic Values.
Uank"
c\i( vh
(m«/l. Pro A)
MI)U
Croup'
(¦onus
Species
S.MCV1'
(ing/l. PI-OA)
1
i) 147
i:
llyalclla
A ill phi pod.
//yah.-/la an ecu
i) 147
2
0.288
c
Lithobates
American bullfrog.
A ilhohalc's catesbeiam i
0.288
3
0.3700
D
Daphnia
Cladoceran,
/ htpliiiia carinata
0.03162
(ladoccran,
/ kip/ima magna
4.330
4
0.7647
G
Braclnoniis
Rotifer.
Hrachionus catvciflorus
0.7647
5
2.194
D
Moil ni
Cladoceran,
Moina macrocopa
2.194
6
9.487
C
Oryzias
Medaka,
Oryzias latipes
9.487
7
>30
B
(jobiocypris
Rare minnow,
Gobiocypris rarus
>30
8
40
A
()ncorhynchus
Rainbow trout,
Oncorhynchus mykiss
>40
9
>76
13
l'imephales
Fathead minnow,
Pimephales promelas
>76
10
SS 32
F
Chironomus
Midge,
Chironomus dilutus
88.32
a Ranked from the niosi scnsili\ c in i lie most resistant based on Genus Mean Chronic Value,
b From Appendix C. Acceptable lTos>hwater Chronic PFOA Toxicity Studies,
c MDR Groups identified in I \u>i note C of Table 3-3.
85
-------
1.0 T
0.9 --
¦» 0.8 --
o
«
^ 0.7
a>
>
« 0.6 --
3
=
£0.5 +
S 0.4 +
C3
—
s 03 "
+->
fl
a 0.21
41
8-1
0.1 --
0.0
¦ Invertebrate
~ Fish
A Amphibian
• Insect
Chironomus <
Piniephales (non-definitive, greater than value) ~
Oncorhynchus (non-definitive, greater than value) ~
Gobiocypris (non-definitive, gr eater than value) ~
~ Oiyzias
¦ Moina
Brachionus
¦ Daphnia
~ Lithobates
Hyalella
0.1
1 10
Genus Mean Chronic Value (mg/L PFOA)
100
Figure 3-3. Freshwater Genus Mean Chronic Values for PFOA.
3.1.1.4 Summary of Chronic PFOA Toxicity Studies Used to Derive the Estuarine/Marine
Aquatic Life Criterion
There are no quantitatively acceptable chronic estuarine/marine PFOA toxicity studies at
this time.
3.2 Derivation of the PFOA Aquatic Life Criteria
3.2.1 Derivation of Water Column-based Criteria
3.2.1.1 Derivation of Acute Water Criterion for Freshwater
The acute data set for PFOA contains 18 genera representing seven of the eight
taxonomic MDR groups. The missing MDR is a representative from an insect family. GMAVs
for the four most sensitive genera were within a factor of 1.7 of each other. The freshwater Final
Acute Value (FAV) (i.e., the 5th percentile of the genus sensitivity distribution, intended to
address 95 percent of the genera) for PFOA is 97.14 mg/L, calculated using the procedures
86
-------
described in the 1985 Guidelines (Table 3-8). The FAV is lower than all the GMAVs for the
tested species, except the Chydorus GMAV (i.e., 93.17 mg/L). The FAV was then divided by
two to obtain a concentration yielding a minimal effects acute criterion value. Based on the
above, the FAV/2, which is the freshwater acute criterion water column magnitude (criterion
maximum concentration, CMC), is 49 mg/L PFOA (rounded to two significant figures) and is
expected to be protective of 95% of freshwater genera potentially exposed to PFOA under short-
term conditions of one-hour of duration, if the one-hour a\ eraue magnitude is not exceeded more
than once in three years (Figure 3-4).
Table 3-8. Freshwater Final Acute Value and Criterion Maximum Concentration.
Calculated I'leshwaler I'.
\V based on 4 lowest \allies
olal Number ofGMAYs in Data Set IS
G.MAY
Knnk
(ienus
(mg/L)
ln(CMAY)
Ih((;ma\ )2
l»=U/(N+l)
siirt(P)
1
Chydorus
93.17
4.53
20.56
0.053
0.229
2
Daphnia
144.1
4.97
24 71
0.105
0.324
3
Brachiomis
| 5
-------
1.0
0.9
a A
o 0.
0.7 -
o
£
QJ
>
« 0.6
*3
s
3 0.5
a
a
&
0.4
0.3
0.2
0.1
0.0
-
~ Oncorhynchus
--
~ Lithobates
~ Anaxyrus
A Ambystoma
• Physella
-
~ Lepomis
-
~ Hyla
-
~ Danio
-
¦ Neocaridina
-¦
~ Pimephales
~ Chironomus
-
¦ Dugesia
-
~ Xenopus
¦ Invertebrate (Other)
--
¦ Moina
• Invertebrate (Mollusk)
-
• Lampsilis
~ Fish
--
• Ligimiia
¦ Brachionus
~ Amphibian
~ Insect - Qualitative Data
-¦
¦ Daphnia
CMC
¦ Chydoras
i i i i i 1 i i i i i i i i |
1 1 1
i i i
10
100 1,000
Genus Mean Acute Value (mg/L PFOA)
10,000
Figure 3-4. Ranked Freshwater Acute PFOA GMAVs used for the Criterion Calculation
and the Qualitative Value for the Insect MDR Group.
Note: Chirortomus is only displayed as a visual representation to show relative rank of the species based on the
qualitative data. The Chirortomus GMAV is not used in the Final Acute Value calculation.
3.2.1.2 Derivation of Acute Water Criterion for Estuarine/Marine Water
The 1985 Guidelines state that data from a minimum of eight families are needed to
calculate an estuarine/marine FAV. Insufficient data exist to fulfill all eight of the taxonomic
MDR groups. Notably, no acceptable test data on fish species were available. Since data were
available for only three families, an estuarine/marine FAV could not be derived (and
consequently, the EPA cannot derive an estuarine/marine acute criterion). EPA has, however,
developed an acute benchmark value using available empirical data and EPA/ORD's web-ICE
tool to estimate missing data. The acute estuarine/marine benchmark is provided in Appendix L.
88
-------
3.2,1.3 Derivation of Chronic Water Criterion for Freshwater
The freshwater Final Chronic Value (FCV) (i.e., the 5th percentile of the genus sensitivity
distribution, intended to address 95 percent of the genera) for PFOA is 0.09404 mg/L, calculated
using the procedures described in the 1985 Guidelines (Table 3-9). The freshwater chronic
criterion water column magnitude (CCC) is the FCV rounded to two significant figures, or 0.094
mg/L PFOA, and is expected to be protective of 95% of freshw ater genera potentially exposed to
PFOA if the four-day average magnitude is not exceeded more than once in three years.
Table 3-9. Freshwater Final Chronic Value and Criterion Continuous Concentration.
Calculated I'leshwaler 1 CY based on 4 lowest \allies
otal Number o
"(iMCYs in Data Set 1"
(i.MCY
Rank
(iCIlllS
(mg/l.)
ln((i.M('Y)
ln((;M( A )2
P=R/(N+1)
sqrl(P)
1
llyulcllu
U. 147
-1.917
3.07b
0.091
0.302
2
Lithobates
0.288
-1 245
1 550
0.182
0.426
3
Daphnia
0.3700
-i) 004
o.(ws
0.273
0.522
4
Brachionus
0.7647
-() 2bS
0.072
0.364
0.603
£ (Sum):
-4.42
6.29
0.91
1.85
S2 =
27.54
S = slope
L =
-3.538
I. X-a\is intercept
A =
-2.364
A InlCY
FCV =
0.09404
P cu nui lative probability
CCC =
0.094 mg/l. PI-OA (rounded to two significant figures)
89
-------
4J
>
"w
«
s
S
3
u
-£
S
es
s
a
K
1.0 -T
0.9 -
0.8 -
0.7 -
0.6 -
0.5 -
0.4 -
0.3 -
0.2 -
0.1 -
0.0 --
¦
Invertebrate
~
Fish
A
Amphibian
•
Insect
-ccc
Chironomus 4
Pimephales (non-definitive, greater than value) ~
Oncorhynchus (non-definitive, greater than value) ~
Gobiocypris (non-definitive, greater than value) ~
~ Oiyzias
¦ Mo ina
¦ Brachionus
¦ Daphnia
A Lithobates
Hyalella
0.01
0.1 1 10
Genus Mean Chronic Value (mg/L PFOA)
100
Figure 3-5. Freshwater Quantitative GMCVs used for the Criterion Calculation.
90
-------
3.2,1.4 Deriving A Protective Duration Component of the Water Column-Based Chronic
Criterion
The EPA 1985 Guidelines set the standard chronic duration at four days for two primary
reasons. The 1985 Guidelines state, "An averaging period of four days seems appropriate for use
with the CCC for two reasons. First, it is substantially shorter than the 20 to 30 days that is
obviously unacceptable. Second, for some species it appears that the results of chronic tests are
due to the existence of a sensitive life stage at some time during the test "
Among tests with chronically sensitive genera, Bartlett et al. (2<>21) measured effects of
PFOA on H. azteca (the most chronically sensitive GMCV) survival at 7, 14. 21. 2S. 35, and 42
days and concluded, "Toxicity increased approximately l\\o-lbld over the duration of exposure,
with LCsos of 110 mg/L after seven days and 51 mg/L after 42 days." Based on Table S6 of
Bartlett et al. (2021), LC50 values deceased from se\ en to 21-ckiys and remained generally stable
from 28 to 42 days, suggesting the chronic PI OA I .("*¦¦ \ alue hecame time-independent between
21 and 28 days. Although II. aiiccu sui \ ival was tolerant al seven days, clear time-dependent
toxicity may not occur lor more sensiti\ e endpoinls such as reproduction.
Bartlett et al (2<>21) determined effects to reproduction to be more sensitive than long-
term sur\ i\ al. but only measured effects to reproduction after 42 days of exposure, which is
substantially longer than the lour day chronic duration. Bartlett et al. (2021) noted, effects to
amphipod reproduction are typically the result of effects to growth under the premise, "larger
amphipods have a greater reproductive output" and "reduced growth delays sexual maturity."
However, results observed in Bartlett et al. (2021), suggested "the effects on reproduction may
also have occurred independently of growth." Therefore, the reproductive-specific effects
observed by Bartlett et al. (2021) may not have been caused by the long-term effects of reduced
growth but were possibly the result of a sexually-developing and uniquely-sensitive life stage
91
-------
that existed during a relatively brief duration within the longer 42-day test. Such instances are
among the two primary reasons why the 1985 Guidelines prescribed the standard four-day
chronic duration.
PFOA effects observed for other chronically sensitive and short-lived species further
suggests a four-day chronic duration was appropriate. For example, the SMCV for Brachionus
calyciflorus and the Brachionus GMCV (fourth most sensitive genus) are both the geometric
mean of a full life-cycle test (i.e., up to 200 hours) test by Zhang et al (2') I ,ni) and a four day
test by Zhang et al. (2014b). Chronic values for both studies correspond to the effect on
population intrinsic growth rate. The full life-cycle test lasted up to 200 hours and yielded an
EPA-calculated ECio of 0.5015 mg/L. The four day test yielded an EPA-calculated ECio of 1.166
mg/L, suggesting chronic PFOA effects 1o shorl-li\ cd species may occur after four days and
increase with exposure duration.
Similarly, the SMCV lor Moinamacvocopa and the Mount GMCV (fifth most sensitive
genus) are based on a se\en day lile-cycle lest by Ji et al. (2008), which also suggested
reproducti\ e effects (cndpoinl of number of young per starting female) occur in as little as seven
days
()\ erall. no chronic PI-OA toxicity tests systematically evaluated time-to-effect, reported
effect data al lime intervals at a high enough resolution to model the speed of toxic action,
assessed time variable PI OA exposures, or assessed the potential for latent toxicity. However,
chronic tests, including life cycle tests with relatively short-lived species suggest chronic effects
may occur at durations shorter than those of standard toxicity tests (e.g., 28 days) and a chronic
four day duration component of the water column criterion was considered protective for these
species/genera. Therefore, EPA has set the duration component of the PFOA chronic water
92
-------
column criterion at four days to reflect the chronic criterion duration recommended in the 1985
Guidelines. This four day duration component of the chronic water column criterion is also
consistent with U.S. EPA (1991), which considered the default four day chronic averaging period
as "the shortest duration in which chronic effects are sometimes observed for certain species and
toxicants," and concludes that four day averaging "should be fully protective even for the fastest
acting toxicants."
3.2.1.5 Derivation of Chronic Water Criterion for Estuarine/Marine Water
There are no quantitatively acceptable GMCVs for estuarine/marine genera.
Consequently, the EPA could not derive an estuarine/marine chronic criterion.
3.2.2 Derivation of Tissue-Based Criteria
Chronic PFOA toxicity data with measured tissue concentrations were limited. There
were no aquatic life tissue-based toxicity studies considered acceptable for quantitative use.
Therefore, there were not sufficient data to derive chronic tissue criteria using a sensitivity
distribution approach Instead, the water column chronic criterion was transformed into
corresponding tissue-bused criteria through a Ii.\I" approach, as outlined in Section 2.11.3. The
chronic PI OA tissue-based criteria were cleii\ ed by translating the chronic freshwater column
criterion (i e . <> <>lM mg/L; see equation X-I in Section 3.2.1.3) into corresponding tissue-based
criteria. The resulting tissue criterion corresponded to the tissue type from the 20th percentile
BAF used in the equation (see Section 2.11.3). The 20th centile BAF was used to derive tissue-
based criteria as a relatively conservative BAF estimate in order to protect species across taxa
and across water bodies with variable bioaccumulation conditions. That is, use of the 20th centile
BAF protects species and conditions where the bioaccumulation of PFOA and resultant tissue-
93
-------
based exposures is relatively low as well as those conditions with the bioaccumulation potential
of PFOA is relatively high.
3.2.2.1 PFOA Bioaccumulation Factors (BAFs)
Section 2.11.3.1 above summarizes the literature search, calculation, and evaluation of
the PFOA BAFs for aquatic life. These BAFs were compiled by and can be found in Burkhard
(2021). BAFs used in the derivation of the PFOA tissue-based criteria consisted of two or more
water and organism samples each and were collected within one year and 2 km distance of one
another. In order to derive more protective tissue criteria across and within water bodies, the
distributions of BAFs used to derive tissue criteria were based on the lowest species-level BAF
reported at a site. When more than one BAF was available for the same species within the same
waterbody, the species-level BAF was calculated as the geometric mean of all BAFs for that
species at that site. Summary statistics for the PFOA BAFs used in derivation of the tissue-based
criteria are presented in Table 3-10 and individual BAFs are provided in Appendix P.
Table 3-10. Summary Statistics for PFOA BAFs in Invertebrate Tissues and Various Fish
Tissues1.
20th
Geometric
Median
Centile
Mean BAF
BAF
BAF
Minimum
Maximum
(L/kg-wet
(L/kg-wet
(L/kg-wet
(L/kg-wet
(L/kg-wet
Category
n
weight)
weight)
weight)
weight)
weight)
Invertebrates
21
105.3
84.8
11.76
0.985
9,680
Fish Muscle
17
7.152
7.94
1.331
0.292
656
Fish Whole Body
25
198.6
219
64.93
1
16,273
1 Based on the lowest species-level BAF measured at a site (i.e., when two or more BAFs were available for the
same species at the same site, the species-level geometric mean BAF was calculated, and the lowest species-level
BAF was used).
3.2.2.2 Deriving Tissue-Based Criteria from the Chronic Water Column Criterion
Invertebrate whole-body, fish whole-body, and fish muscle tissue criteria were derived
separately by multiplying the freshwater chronic water column criterion by the respective 20th
94
-------
centile of the distribution of BAFs described in Section 3.2.2.1, using Equation X-I from Section
2.11.3. The use of a 20th centile BAF results in more protective tissue criteria than those derived
from a BAF based on a central tendency measure (e.g., geometric mean or median), which would
only be protective on average or approximately 50% of the time.
The invertebrate whole-body tissue chronic criterion was calculated by multiplying the
20th centile BAF of 11.76 L/kg wet weight and the PFOA freshwater chronic water criterion of
0.094 mg/L, which resulted in a chronic invertebrate whole-body tissue criterion of 1.11 mg/kg
wet weight. The fish muscle tissue chronic criterion was calculated by multiplyinu the 20th
centile BAF of 1.331 L/kg wet weight and the PFOA freshwater chronic water criterion of0.094
mg/L, which resulted in a chronic fish muscle-based chronic criterion of 0.125 mg/kg wet
weight. The fish whole-body tissue chronic criterion was calculated by multiplying the 20th
centile BAF of 64.93 L/kg wet weight and the PI-OA freshwater chronic water criterion of 0.094
mg/L, which resulted in a chronic fish whole-body tissue criterion of 6.10 mg/kg wet weight.
The chronic tissue-based criteria are expected to be protective of 95% of freshwater genera
potentially exposed to PI-'OA under long-term exposures if the tissue-based criteria magnitudes
are not exceeded more than once in ten years
I-PA acknowledges that there is uncertainty in deriving protective tissue criteria
magnitudes by transforming the chronic water column criterion (which was based on tests that
only added PFOA to the water column) into tissue concentrations through field-measured
bioaccumulation data of paired water and tissue concentrations in waterbodies. Nevertheless, the
chronic water column criterion is based on chronic toxicity tests that fed test organisms. In these
tests, PFOA can directly affect species based on direct water column exposure and/or sorb to
added food that is consumed by test organisms before eliciting chronic effects from dietary
95
-------
exposure. Therefore, the chronic water column criterion magnitude accounts for water column-
based and, to a possible lesser extent, dietary-based effects, while the field-based BAFs account
for water column-based and dietary-based PFOA exposure in tissues.
The tissue criteria will provide information to states, tribes, and stakeholders on potential
effects to aquatic organisms based on aquatic tissue monitoring data. No available quantitatively
acceptable data on the effects of dietary exposures to aquatic species were available, thus EPA
elected to develop protective values for aquatic organism tissues based 011 the observed
relationship between water column concentrations and tissue concentrations and observed PFOA
toxicity in chronic tests where PFOA was only added directly to the water column
3.2,2.3 Deriving Protective Duration and Exceedance Frequencies for the Tissue-based Chronic
Criteria
3.2.2.3.1 Duration: Chronic Tissue-Basal ('rucria
PFOA concentrations in tissues are generally expected to change only gradually over
time in response to en\ ironmental lluctuations. The chronic tissue-based criteria averaging
periods, or duration components, were therefore specified as instantaneous, because tissue data
provide point, or instantaneous, measurements that reflect integrative accumulation of PFOA
over time and space in populalion(s) at a ui\en site
3.2.2.3.2 l-'nyiiciicy: Chrome Tissue-Based Criteria
Ecological recovery times following chemical disturbances are situational-specific, being
largely dependent 011 (I) biological variables such as the presence of nearby source populations
or generational time of ta\a affected; (2) physical variables such as lentic and lotic habitat
considerations where recovery rates in lentic systems may be slower than lotic systems where the
pollutant may be quickly flushed downstream, and; (3) chemical variables such as the persistence
of a chemical and potential for residual effects. Given the large variation in possible biological
and physical variable influencing ecological recovery, EPA focused on the known chemical
96
-------
attributes of PFOA to inform the recommended ten-year exceedance frequencies for the chronic
tissue-based criteria.
Metals and other chemical pollutants may be retained in the sediment and biota, where
they can result in residual effects over time that further delay recovery. Few studies are available
concerning PFOA elimination or depuration half-life in aquatic animals, however the data that
exist indicate a short half-life. For example, the elimination half-lilc lor PFOA in adult rainbow
trout exposed to PFOS for 28 days via the diet followed by 28 days depuration was estimated to
be seven days in muscle tissue (Falk et al. 2015), while the terminal half-life in rainbow trout
receiving a one-time intra-arterial injection of PFOA was 12 6 days (Consoer el al 2
-------
reproductive failure and deformities in fish, were still measurable (as fish deformities) in 1992
(seven years later) and in 1996 (ten years later). Lemly (1997, pg. 280) estimated based on these
data that "the timeframe necessary for complete recovery from selenium contamination from
freshwater reservoirs can be on the order of decades."
Beyond bioaccumulation, chemical-specific considerations such as degradation vs.
persistence may also provide a mechanism influencing ecological recovery rates. The persistence
of PFOA has been attributed to the strong C-F bond, with no know n Modegradation or abiotic
degradation processes for PFOA (see Section 2.3). Similarly, metals do nol degrade and may
persist in aquatic systems following elevated discharge The persistence of metals may explain
why metals had the second longest median recovery time of any disturbance described in a
systematic review of aquatic ecosystem reco\ cry ((icrgs et al 2<> I (•>) Gergs et al. (2016) showed
recovery times following metal disturbances ranged from roughly six months to eight years
(median recovery lime one year. 75"'centile lhreeyears.il 20).
The bioaccunuilali\ e nature and persistence of PFOA in aquatic systems, in combination
with the documented icco\ery times of pollutants with similar chemical attributes (Lemly 1997;
Gergs et al 2<) I (•>). suggested I < > years was a protective exceedance frequency for the tissue-
based PI OA criteria. The tissue-based criteria are protective if they are not exceeded more than
once in ten years to allow sufficient time for PFOA concentrations built up in tissues and source
reservoirs in the freshwater system to diminish while simultaneously providing freshwater
organisms adequate time to recover following elevated PFOA exposures in tissues.
3.3 Summary of PFOA Aquatic Life Criteria
This Aquatic Life Ambient Water Quality Criteria for PFOA document includes water
column based acute and chronic criteria and tissue-based criteria for freshwaters. Acute and
98
-------
chronic water column criteria magnitudes for estuarine/marine waters could not be derived at this
time due to data limitations; however, an acute estuarine/marine benchmark is provided for
states/authorized tribal consideration (see Appendix L). The freshwater acute water column-
based criterion magnitude is 49 mg/L, and the chronic water column-based chronic criterion
magnitude is 0.094 mg/L. The fish whole-body tissue criterion magnitude is 6.10 mg/kg wet
weight, the fish muscle tissue criterion magnitude is 0.125 mg'ku wet weight and the invertebrate
whole-body tissue criterion magnitude is 1.11 mg/kg wet weight (Table 3-1 I) The assessment of
the available data for fish, invertebrates, amphibians, and plants indicates these criteria will
protect the freshwater aquatic community.
The freshwater chronic water column criterion is more strongly supported than the
chronic tissue-based criteria because the u ater column-based chronic criterion was derived
directly from the results of empirical toxicity tests The chronic tissue-based criteria are
relatively less certain because they were derived by trail storming the chronic water column
criterion into tissue concentrations through BAFs, with any uncertainty and variability in the
underlying BAFs then inopauatinu into the resultant tissue-based criteria magnitudes.
99
-------
Table 3-11. Recommended Freshwater Perfluorooctanoic acid (PFOA) Aquatic Life
Ambient Water Quality Criteria.
Tvpc/.Mcdia
Acule \Ysilcr
Column (CMC)14
Chronic W'siler
Column (CCC)1^
Inverlebrale
Whole-
Body12
lisli
Whole-
Body12
I'isli Muscle12
Magnitude
49 mg/L
0.094 mg/L
1.11 mg/kg
WW
6.10 mg/kg
WW
0.125 mg/kg
WW
Duration
One hour average
Four day average
Instantaneous3
Frequency
Not to be exceeded
more than once in
three years on
average
Not to be exceeded
more than once in
three years on
average
Not to be exceeded more than once in ten
years on a\ eraue
1 All five of these water column and tissue criteria are intended to be independently applicable and no one criterion takes
primacy. All of the above recommended criteria (acute and chronic water column and tissue criteria) are intended to be
protective of aquatic life. These criteria are applicable throughout the year.
2 Tissue criteria derived from the chronic water column concentration (CCC) with the use of bioaccumnlalion factors and are
expressed as wet weight (ww) concentrations.
3 Tissue data provide instantaneous point measurements that reflect integrative accumulation of PFOS over time and space in
aquatic life population(s) at a given site.
4 Criterion Maximum Concentration; applicable throughout the water column.
5 Criterion Continuous Concentration; applicable throuahout the water column.
4 EFFECTS CHARACTERIZATION FOR AQUATIC LIFE
This section describes supporting information for I lie derivation of these PFOA aquatic
life criteria. Specifically, this chapter (I) assesses the iniluenee of including non-North
American resident species in criteria deri\ ation (Section 4.1); (2) considers relatively sensitive
toxicity data from qualitati\ ely acceptable studies that were used as supporting information
(Section 4 2). (3) e\ alnation of the acute insect MDR (Section 4.3); (4) describes the available
PFOA ACRs (Section 4.4). (5) compares the tissue-based criteria magnitudes to the empirical
tissue-based effect concentrations available (Section 4.5); and (5) evaluates aquatic plant
tolerance to PFOA exposures (Section 4.6).
4.1 Influence of Using Non-North American Resident Species on PFOA
Criteria
EPA conducted an additional analysis of the water column-based criteria by limiting the
toxicity datasets to organisms that are residents to the conterminous United States that have
100
-------
established populations to evaluate the influence of including non-North American resident
species in criteria derivation.
4.1.1 Freshwater Acute Water Criterion with Resident Organisms
Three species, the green neon shrimp (Neocaridina denticulata), the cladoceran {Daphnia
carinata) and the planarian (Dugesia japonica), are not resident or reproducing in the
conterminous United States, while it remains uncertain if there are established resident zebrafish
(Danio rerio) populations in the conterminous United States (USFWS 2<) IS) Nevertheless,
zebrafish are common ecotoxicity test organisms that serve as taxonomic surrogates for untested
fish species and are also considered in effects assessments conducted under the Toxic Substances
Control Act (TSCA) and the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA).
Removal of the green neon shrimp, Daphma carinata. and J higcsia japonica, while retaining
zebrafish truncated the freshwater acute dataset lo 22 species representing 16 genera (Table 4-1).
The freshwater acute dataset truncated lo North American resident species only was missing two
MDR groups (a hcnlhic crustacean and an insect) While the SMAV for Daphnia carinata was
the most sensiti\ e genus. its remo\ al and remo\ al of the other non-North American resident
species had limited impact on the exploratory I AY and subsequent acute water column
concentration (Table 4-2). The acute water column concentration based on North American
resident species only, including zebrafish, was 47 mg/L PFOA, which was slightly lower than
the CMC (49 mg/l.) bused on both North American and non-North American species. Had
zebrafish also been removed from the exploratory FAV and acute water column concentration
based on North American resident species only, the four most sensitive genera would have
remained the same, the number of genera in the dataset would have decreased by one, and the
resultant exploratory FAV and acute water column concentration would have been 92.97 mg/L
PFOA and 46 mg/L PFOA, respectively. The exploratory FAV and CMC based on North
101
-------
American resident species only with zebrafish excluded were both similar to the FAV and CMC
described in Section 3.2.1.1.
Table 4-1. Ranked Freshwater Genus Mean Acute Values with North American Resident
Organisms.
Rank"
(i.MAV
(nig/I. PI-OA)
MI)U
Group'
Genus
Species
S.M.W h
(in«/l. PI-OA)
1
93.17
D
Chydorus
Cladoceran.
Chydorus sphaericiis
93.17
2
150.0
H
Brachionus
Rotifer,
Brachionus calycifh >i lis
150.0
3
161.0
G
Ligumia
Black sandshell,
Ligumia recta
161.0
4
164.4
G
Lampsilis
Fatmucket,
Lampsilis siliquoidea
164.4
5
166.3
D
Moina
Cladoceran.
Moma macrocopa
166.3
6
211.7
D
Daphniu
Cladoceran.
/ kiphnia magna
220.0
Cladoceran,
/ kiphnia pnlicaria
203.7
7
377.0
(
Xenopus
Frog,
Xenopus sp.
377.0
8
413.2
B
I'micpltales
Fathead minnow,
Pimephales promelas
413.2
9
572 4
B
/ kinio
Zebrafish,
1 kinio rerio
572.4
10
M6.2
(
l/yla
Gray treefrog,
Hyla versicolor
646.2
11
oM i)
B
Lepomis
Bluegill,
Lepomis macrochirus
664.0
12
681 1
G
Physella
Bladder snail,
Physella acuta
681.1
13
689.4
C
Ambystoma
Jefferson salamander,
Ambystoma jeffersonianum
1,070
Small-mouthed salamander,
Ambystoma texanum
407.3
Eastern tiger salamander,
Ambystoma tigrinum
752.0
14
793.9
C
Anaxyrus
American toad,
Anaxyrus americanus
793.9
15
951.5
C
Lithobates
American bullfrog,
Lithobates catesbeiana
1,020
102
-------
GMAV
MDR
SMAVb
Rank3
(mg/L PFOA)
Group0
Genus
Species
(mg/L PFOA)
Green frog,
Lithobates clamitcms
1,070
Northern leopard frog,
Lithobates pipiens
751.7
Wood frog,
Lithobates sylvatica
999
16
1,682
A
Oncorhynchus
Rainbow trout,
Oncorhynchus mykiss
1,682
a Ranked from the most sensitive to the most tolerant based on Genus Mean Acute Value,
b From Appendix A: Acceptable Freshwater Acute PFOA Toxicity Studies,
c MDR Groups identified inFootnote C of Table 3-3.
Table 4-2. Freshwater Exploratory Final Acute Value and Acute Water Column
Concentration with North American Resident Organisms (zebrafish included).
Calculated Freshwater FAY based on 4 lowest values:r
"otal Number of GMAVs in Data Set = 16
GMAV
Rank
Genus
(mg/L)
ln(GMAV)
ln(GMAV)2
P=R/(N+1)
sqrt(P)
1
Chydorus
93.17
4.53
20.56
0.059
0.243
2
Brachionus
150.0
5.01
25.11
0.118
0.343
3
Ligumia
161.0
5.08
25.82
0.176
0.420
4
Lampsilis
164.4
5.10
26.03
0.235
0.485
£ (Sum):
19.73
97.52
0.59
1.49
S2 =
6.59
S = slope
L =
3.975
L = X-axis intercept
A =
4.549
A = InFAV
FAV =
94.58
P = cumulative probability
CMC =
47 mg/L PFOA (rounded to two significant figures)
4.1.2 Freshwater Chronic Water Criterion with North American Resident Organisms
Three species, the cladoceran {Daphnia carinata), the rare minnow (Gobiocypris varus)
and the medaka (Oryzias latipes), are not resident or reproducing in the conterminous United
States. Removal of these species truncated the freshwater chronic dataset to eight species
representing eight genera (Table 4-3). The freshwater chronic dataset truncated to North
American resident species was missing one MDR group (another phylum or a second insect
order not already represented). Calculating the FCV based on the chronic GSD comprised of
103
-------
North American resident species only resulted in an exploratory FCV of 0.04825 mg/L and a
chronic water column criterion of 0.048 mg/L (rounded to two significant figures; Table 4-4).
The exploratory chronic water column concentration CCC (0.048 mg/L PFOA) based on North
American species only was about half of the CCC based on North American and non-North
American species (0.094 mg/L PFOA) and is lower than the all of the quantitatively-acceptable
GMCVs (Table 3-7). The reduction in the exploratory FCV based on North American resident
species only was primarily an artifact of the FCV calculation procedure rather than inclusion of
more sensitive toxicity data. That is, the reduced used in the exploratory criterion calculation
and the increase in the Daphnid GMCV (which was the result of excluding/), carina la SMCV),
increased the slope of the GSD which decreased the extrapolated FCV. The EPA retained the
chronic water column criterion which includes North American and non-North American
species, with a magnitude of 0.094 mg/L, lo ensure llie fullest, high quality dataset available is
used to represent the thousands of untested aquatic taxa present in U.S. ecosystems when
deriving the chronic criterion lor PI-OA.
Tsihle 4-3. Usinkeri I roshwnlor (icniis Menu Chronic \ sillies with Resident Organisms.
Usink"
(;\K vh
(ing/l. PI OA)
MI)U
(iroiii)'-'1
(ienns
Species
SMCV1'
(ing/l. I'l OA)
1
i) 147
H
llyalella
Amphipod,
Hyalella azteca
0.147
2
() 2SS
C
Lithobates
American bullfrog,
Lithobates catesbeiana
0.288
3
0.7647
G
Brachionus
Rotifer,
Brachionus calyciflorus
0.7647
4
2.194
D
Moina
Cladoceran,
Moina macrocopa
2.194
5
4.330
D
Daphnia
Cladoceran,
Daphnia magna
4.330
6
>40
A
Oncorhynchus
Rainbow trout,
Oncorhynchus mykiss
>40
7
>76
B
Pimephales
Fathead minnow,
Pimephales promelas
>76
104
-------
Rank"
CMC V"
(ing/l. PI OA)
MI)U
Croup'-'1
CcilllS
Species
S.MCV1'
(m«/l. PI OA)
8
88.32
F
Chironomus
Midge,
Chironomus dilutus
88.32
a Ranked from the most sensitive to the most resistant based on Genus Mean Chronic Value,
b From Appendix C: Acceptable Freshwater Chronic PFOA Toxicity Studies,
c MDR Groups identified inFootnote C of Table 3-3.
Table 4-4. Freshwater Exploratory Final Chronic Value and Chronic Water Column
Concentration with North American Resident Organisms.
Calculated Freshwater 1 CY based on 4
owes! \allies.
Total Number of (iMCYs in
Data Set 8
cmcy
Knnk
(iCIlllS
(mg/I.)
in((;M( \)
in((;M( \ )2
P=U/(N+I)
sqrl(P)
1
Hyalella
0.147
-1.917
3.676
0.1 1 1
0.333
2
Lithobates
0.288
-1.245
1.550
0.222
0.471
3
Brachionus
0.7647
-0.268
n 072
0.333
0.577
4
Moina
2.194
0.786
i)ol7
0.444
0.667
£ (Sum):
-2.64
5.91
1.11
2.05
S2 =
67.46
S slope
L =
-4.868
1. \-a\is intercept
A =
-3.031
A = InFCY
FCV =
0.04825
P = cumulative probability
CCC =
0.048 mg/I. PI'"OA (rounded to two significant figures)
4.2 Consideration of Relatively Sensitive Qualitatively Acceptable Water
Column-Based Toxicity Data
A multitude of studies were identified as not meeting EPA's data quality guidelines for
inclusion in the criteria deri\ ation. howc\ er, these studies were used qualitatively as supporting
information to the I'l OA criteria derived to protect aquatic life and provide additional evidence
of the observed toxicity and effects of PFOA, including the relative sensitivities. Most of these
studies produced relatively tolerant effect concentrations relative to the criteria and are reported
in Appendix G. The key studies used qualitatively in the derivation of the PFOA criteria were
identified as being from either relatively sensitive genera or relatively sensitive tests and are
described below. That is, qualitatively acceptable data for tests with species among the four most
105
-------
sensitive genera that were used to qualitatively derive the criteria magnitudes are discussed.
Qualitatively acceptable tests with relevant exposure durations and apical effects that also
observed effect concentrations less than or similar to (e.g., factor of two) the corresponding
criteria magnitude are discussed. Additionally, qualitatively-acceptable data for species within
the one unfulfilled chronic MDR (i.e., another phylum or a second insect order not already
represented) are evaluated relative to the chronic water column criterion. Qualitatively
acceptable studies described below were separated by acute (Section 4 2 1) and chronic (Section
4.2.2) data and only included those studies that reported apical endpoints. The toxicity values
summarized as part of this Effects Characterization were not used in any quanli liili \ e analyses or
in the numerical derivation of the PFOA aquatic life criteria.
4.2.1 Consideration of Qualitatively Acceptable Acute Data
4.2.1.1 Qualitatively Acceptable Acute Data for Species Among the Tour Most Sensitive
Genera Used to Derive the Acute Water Column Criterion
4.2.1.1.1 Most acutely sensitt i v genus, (Itydorus
There were no qualitatively acceptable acute tests for species within the genus, Chydorus.
4.2.1.1.2 Second most acutely sensitive genus, Daphnia
3AI Co. (2000) exposed P. magna to PFOA (CAS # 335-67-1) in a 48-hour static,
unmeasured acute toxicity test that followed USEPA-TSCA Guideline 797.1300. The toxicant
was part of the 3 VI production lot number 269 and was characterized as mixture of PFOA (96.5-
100% of the com pound) and C7 and C9 perfluoro homologue compounds (0-3.5% of the
compound). The 48-hour reported EC50, based on death/immobility, was 360 mg!L PFOA. This
test was not acceptable for quantitative use because of possible mixture effects from other
perfluoro homologue compounds in the test substance but was retained for qualitative use.
3M Co. (2000) summarized four 48-hour static, unmeasured APFO (CAS # 3825-26-1)
acute toxicity tests with the cladoceran, Daphnia magna and APFO. The toxicant was part of the
106
-------
3M production lot number 37 and was characterized as mixture of APFO (96.5-100% of the
compound) and C6, C7 and C9 perfluoro analogue compounds (0-3.5% of the compound). The
48-hour EC50 determined from tests conducted in May 1982, based on death/immobility, was
>1,000 mg/L APFO, while the EC50 for a subsequent test in June 1982 was reported to be 126
mg/L APFO. Possible mixture effects of other perfluoro analogue compounds did not make these
tests acceptable for quantitative use and they were retained for qualitative use.
3M Co. (2000) summarized a 48-hour static, unmeasured acute toxicity test with the
cladoceran, Daphnia magna and APFO (CAS # 3825-26-1). The toxicant was pari of the 3M
production lot number 390 and was characterized as mixture of APFO (78-93% of the
compound) and C5, C6 and C7 perfluoro analogue compounds (7-22% of the compound). The
author-reported 48-hour EC50, based on mortality, was 221 mu I. APFO. The possible mixture
effects of APFO with other perfluoro analogue compounds in the test material did not make this
test acceptable for quanlilali\ e use This test w as retained for qualitative use only.
3M Co. (2000) summarized a 4X-hour static, unmeasured acute toxicity test with the
cladoceran. Daphnia magna and API O (CAS 3825-26-1). The acute test followed USEPA-
TSCA Guideline 7l)7 I3<)() protocol The toxicant was part of the 3M production lot number
HOG I- 2<)5 and was not sufficiently characterized but was considered a mixture of APFO (30%
of the compound) and water (80% of the compound). The author-reported 48-hour EC50, based
on mortality, was 1.2<)i) mu L test substance. The authors reported that the test substance was
considered a mixture of APFO and other impurities, so the EC50 does not accurately reflect the
toxicity of APFO and therefore the value was not acceptable for quantitative use but was retained
for qualitative use.
107
-------
3M Co. (2000) summarized a 48-hour static, unmeasured acute toxicity test with the
cladoceran, Daphnia magna and APFO (CAS # 3825-26-1). The acute test followed test
guidance from OECD 202. The toxicant was part of the 3M production lot number 2327 and was
characterized as mixture of APFO (<45% of the compound), water (50% of the compound), inert
perfluorinated compound (<3% of test substance), and Cs and C7 perfluoro analogue compounds
(1-2%) of the compound). The author-reported 48-hour EC50, based 011 death/immobility, was 584
mg/L test substance. Because of the possible mixture effects of the inert peril uorinated
compounds and other perfluoro analogue compounds the test was not acceptable lor quantitative
use but was retained for qualitative use.
3M Co. (2000) summarized a 21-day static-renewal, unmeasured chronic toxicity test
with the cladoceran, Daphnia magna, and API O (CAS M- 3825-2(->-1) and also briefly described a
corresponding acute test with a reported 4S-hour ! .('-¦ of 2o6 mu L APFO. Very few details were
provided about the acute test methodology. The test compound was assumed to be that of the
chronic test, where the toxicant was part of the 3M production lot number 37 and was
characterized as mixture of API O (lH->.5-1 <)<)% of the compound) and C6, C7 and C9 perfluoro
analogue compounds (0-3 5".. of the compound). The 48-hour EC50 from this test was not used
quanti tiiti \ ely because of missing study details and the possible presence of additional PFAS, but
the study was retained for qualitative use.
Overall, these / K magna acute effect concentrations were all greater than the FAV (i.e.,
97.12 mg/L) and the acute criterion magnitude (i.e., 49 mg/L). These additional data suggest the
Daphnia GMAV (i.e., 144.1 mg/L) used to derive the acute criterion was sufficiently protective.
4.2.1.1.3 Third most acutely sensitive genus, Brachionus
There were no qualitatively acceptable acute tests for species within the genus,
Brachionus.
108
-------
4.2.1.1.4 Fourth most acutely sensitive genus, Ligumia
There were no qualitatively acceptable acute tests for species within the genus, Ligumia.
4,2,1,2 Consideration of Relatively Sensitive Freshwater Tests based on Qualitatively
Acceptable Data
This section focuses on qualitatively acceptable tests that were most relevant to informing
the appropriateness of the acute freshwater criterion. Specifically, those tests used to
qualitatively inform the acute freshwater criterion magnitude were identified as most relevant if
they met all parameters listed below:
1. reported effect concentrations that were less than or similar to ( e u . factor of two) the
acute criterion magnitude;
2. evaluated an animal species;
3. conducted the test for a relevant acute exposure duration (e.g., -48 hours to -96 hours);
4. evaluated apical effects (i.e., acute mortality/inhibition), and;
5. not already discussed in the previous section (i.e., not a species discussed among the four
most sensitive genera).
The toxicity values summarized below were not used quantitatively to derive the acute
PFOA criterion. Results of each indi\ idual study (as well as the rationale why a study was not
quantitatively acceptable) were considered relative to the acute criterion magnitude to ensure the
acute PI OA criterion was not underpioductive and to provide additional supporting evidence of
the potential toxicity of PI OA to aquatic organisms.
4.2.1.2.1 ( jams: Dugesia iplanarian)
Yuan ol al. (2015) conducted a 96-hour, unmeasured renewal acute test on PFOA (96%
purity) with Dugesta /aponica The study reported 96-hour LCso was 39.35 mg/L (95% C.I. =
32.38 - 46.32 mg/L). The test was not acceptable for quantitative use because the test organisms
were collected from a fountain in Quanhetou, Boshan, China, where there may have been
potential exposures to PFAS from the source of the test organisms. Overall, three additional tests
were available for this species (Li 2008, 2009), which resulted in a SMAV of 383.6, suggesting
109
-------
this species is tolerant to acute PFOA exposures. The apparent sensitivity of the LCso reported by
Yuan et al. (2015) may have been the result of compounding chemical stressors originating from
the source of the test organisms (i.e., fountain in Quanhetou, Boshan, China)
4.2.1.2.2 Genus: Danio (zebrafish)
Truong et al. (2014) evaluated the sub-chronic effects of 1,060 compounds (U.S. EPA
ToxCast phase 1 and 2) on zebrafish, Danio rerio, through the use of high-throughput
characterization of multidimensional in vivo effects. The effects ofAPIO and PFOA on
mortality, growth, behavior, morphology, histology and physiology were obsei\ ed until 120 hpf
(114-hour test duration) with the water quality conditions not reported. The most sensitive
endpoint was mortality with a reported LOEC of 0.02751) mu I. APFO. There were no effects of
PFOA on mortality for zebrafish embryos with a reported NOI-C of 26.50 mg/L PFOA. This test
was not used quantitatively and retained for qimlitati\ e use only because the exposure durations
were too long for an acute test and too short for a chronic test
Dasgupta el al. (2020) e\ aluated the acute effects of PFOA (CAS # 335-67-1, 96%
purity, purchased from Acros Oi uanics) on zebrafish (Danio rerio) via a 66-hour unmeasured,
static study At test terniination there were no significant effects on survival or development of
zebrallsh embryos. The M-hour NOI'C of 20.70 mg/L PFOA, based on survival, was acceptable
for qualitati \ e use only due to the short exposure period (i.e., 66-hour exposure instead of the
established 96-hour acute exposure for this species).
Wasel et al. (2020) reported the results of an unmeasured, renewal acute toxicity test
with larval D. rerio and PFOA (>99% purity). The 90-91 hour LCso reported by the authors was
57.6 mg/L for the unbuffered test solution. Based on the starting age of the organisms (embryo,
five to six hpf), the acute test was too short to be used quantitatively, so values were acceptable
110
-------
for qualitative use only, especially since other acute quantitatively acceptable tests for this
species were available.
Quantitatively acceptable acute tests used to calculate the SMAV (Hagenaars et al. 2011;
Godfrey et al. 2017a; Stengel et al. 2017) suggested/), rerio is relatively tolerant to acute PFOA
exposures with a D. rerio SMAV of 572.4 mg/L. Another quantitatively acceptable acute test
with the zebrafish was conducted by Corrales et al. (2017). FTowe\ cr. the I Cso value from this
test was excluded from the SMAV calculation because a comparative assessment between this
LCso value and the other five quantitatively-acceptable values available indicated the LCso was
an outlier, falling out more than an order of magnitude lower than the other fi\e I values. It is
expected that D. rerio will be tolerant to acute PFOA exposures.
4.2.1.2.3 Genus: Pimephales (fathead minnow >
3M Co. (2000) reported the results of a lH->-hour static, unmeasured acute toxicity test
with the fathead minnow. Pimephalespromelas and API O (C AS .1825-26-1). The toxicant was
part of the 3M production lot number S3 and was characterized as mixture of APFO (96.5-100%
of the compound) and ('- and ("¦¦ peril uoro analogue compounds (0-3.5% of the compound).
No specific test protocol was identified The authors extrapolated the concentration-response
data graphically to estimate a lH->-hour I.Cso of 70 mg/L APFO. This test was not acceptable for
quantita1i\ e use because of the lack of replicates and lack of observed effects in the test, as well
as the possible mixture effects of other perfluoro analogue compounds. Although the LCso
reported in this test is sensitive relative to the FAV (i.e., 97.14 mg/L), quantitatively acceptable
acute data with this species (Corrales et al. 2017) suggest P. promelas is tolerant with a SMAV
of 413.2 mg/L
111
-------
4.2.2 Consideration of Qualitatively Acceptable Chronic Data
4.2.2.1 Qualitatively Acceptable Chronic Data for Species Among the Four Most Sensitive
Genera Used to Derive the Chronic Water Column Criterion
4.2.2.1.1 Most chronically sensitive genus, Hyalella
There were no qualitatively acceptable chronic tests for species within the genus,
Hyalella.
4.2.2.1.2 Second most sensitive genus, Lithobates
Hoover et al. (2017) tested PFOA (96% purity) toxicity oil the northern leopard frog,
Lithobatespipiens (formerly, Ranapipiens) in a chronic renewal test toxicity using measured
PFOA treatment concentrations. The 40-day NOEC was > 1.0 mg/L PFOA based on Gosner
stage reached at test termination and snout-vent length The test used water renewals rather than
the required flow-through design for chronic ALC development: however, leopard frogs
commonly do not tolerate flow-through test systems and the use of renewal system was
appropriate for this study organism. Also, PI OA was detected in the control organisms at
concentrations 1 luce orders of magnitude lower than any PI OA treatment groups, indicating the
trace contamination in controls may not be considered a significant issue. The 40-day NOEC of
>1.0 mg/L was classified as acceptable lor quantitative use based on meeting data quality
objecti\ es. however, it was not used to deri\ e the chronic criterion because the study showed no
adverse effects at the highest treatment concentrations (i.e., 1.0 mg/L). Because the highest
treatment group that showed no effects was a relatively low treatment concentration, including
this NOEC value in the criterion calculation would have resulted in the criterion magnitude being
influenced by the relatively low-test concentration selected by study investigators (that did not
produce an adverse response), rather than a concentration-response relationship. Therefore, this
test was not used quantitatively and was considered as qualitatively acceptable for use in
criterion derivation.
112
-------
Flynn et al. (2021) evaluated the chronic effects of PFOA (CAS # 335-67-1, >96%
purity, purchased from Sigma-Aldrich) on northern leopard frogs, Lithobatespipiens (formerly
Ranapipiens), via a 30-day sediment-spiked, static outdoor mesocosm study. At test termination
(30 days) there was no effect on survival or growth (snout-vent length and weight). The 30-day
NOEC, based on survival and growth, was 0.066 mg/L. However, on test-day five and at test
termination all frogs in the spiked sediment mesocosm were less de\ eloped, based on Gosner
stage, than the control mesocosms. The study was not acceptable for quantitative use because the
test design was an outdoor spiked sediment mesocosm exposure with algal and xooplankton
communities present and because of the relatively low NOF.C value that did not quantitatively
inform criteria derivation based on an exposure-response effect
Overall, these two studies showed minimal effects to the northern leopard frog at the
concentrations tested, while the indoor laboratory lesl In I'lynn el al. (2019; used to derive the
Lithobates catesbeiana SMCV) showed a 7'\> reduction in SYL after 72-day exposures at 0.288
mg/L. Although I loo\ or el al (2' > I 7) reported a NOEC of 1.0 mg/L, the tests only consisted of a
40-day exposure, w hich may not ha\ e been long enough to elicit the chronic effects to SVL
obser\ ed by I'lynn el al (2dI1)) after 72 days. For example, Flynn et al. (2019) reported effects
of PFOS on / nhobates catcsbeiana tadpole mass after 21, 42, 56, 63, 70, and 72 days, with
PFOS dose-dependenl effects only becoming apparent at 56 days. Results of Flynn et al. (2021)
could not meani nuful ly i nlbrm the appropriateness of the chronic criterion or the Lithobates
catesbeiana SMCV (0.288 ing/L) because Flynn et al. (2021) did not observe effects of PFOA
on Lithobates pipiens, with a relatively low NOEC of 0.066 mg/L.
4.2.2.1.3 Third most sensitive genus, Daphnia
3M Co. (2000) summarized a 21-day static-renewal, unmeasured chronic toxicity test
with the cladoceran, Daphnia magna, and APFO (CAS # 3825-26-1). The toxicant was part of
113
-------
the 3M production lot number 37 and was characterized as mixture of APFO (96.5-100% of the
compound) and C6, C7 and C9 perfluoro analogue compounds (0-3.5% of the compound). The
test followed U.S. EPA (1982) and OECD (1997) test protocols. The 21-day NOEC and LOEC,
based on reproduction and survival were 22 and 36 mg/L APFO, respectively with a
corresponding MATC of 28.14 mg/L. This test was acceptable for qualitative use only because
of the possible mixture effects of other perfluoro analogue compounds but does not suggest/).
magna will be chronically sensitive relative to the chronic freshwater criterion
Seyoum et al. (2020) evaluated the chronic effects of PFOA (CAS 335-07-1, >99%,
purchased from Sigma) on Daphnia magna neonates \ in a 21 -day unmeasured, static-renewal
study. The study authors did not report following any specific protocol. The 21-day reproductive
(fecundity) LOEC of 0.4141 mg/L PFOA was reported by the study authors, where a —38.25%
reduction in mean number of daphnids relati\ e to the control was observed. EPA was unable to
fit a model with significant parameters to the reproduction-based concentration-response data
due to a lack of clear concentration-dependent effects beyond the LOEC. The reproduction-based
LOEC (i.e., D4I4I mu |.) was selected as the chronic value from this test; however, it was not
considered acceptable lor quantilali\ e use because chronic responses in this test did not display
concentration-dependent effects beyond the LOEC despite a 25X increase in treatment
concentrations Moreover, additional EC10 values from other, quantitatively acceptable tests,
were available to inform the chronic sensitivity of Daphnia magna in criteria derivation.
4.2.2.1.4 Fourth most chronically sensitive genus, Brachionus
The qualitatively acceptable chronic value tor Brachionus from Zhang et al. (2014b)
was discussed in greater detail in Section 3.1.1.3.4. Briefly, Zhang et al. (2014b) conducted a full
life-cycle test using renewal conditions for approximately four days on the rotifer, Brachionus
calyciflorus. Zhang et al. (2014b) reported several endpoints, including intrinsic rate of natural
114
-------
increase, which was selected as the primary endpoint for criterion derivation. Zhang et al.
(2014b) also observed effects to resting egg production. Resting egg production is an
ecologically important endpoint for this species because it represents the final result of sexual
reproduction. NOEC and LOEC values were not reported for resting egg production, but 0.25
mg/L PFOA produced more than a 50% reduction in resting egg production. Based on the
authors description of results in the text, it was assumed the B. culycijlorus four-day NOEC for
resting egg production was 0.125 mg/L and the LOEC was 0.25 mg/L and the calculated MATC
was 0.1768 mg/L, suggesting resting egg production may be a relatively sensiti\ e endpoint.
Because there was only one replicate (as implied by lack of error bars in Figure 1 of the
publication, no clear description of replicates in the methods section, and no author-reported
statistical analysis of this endpoint), resting egg production from this sludy was not considered
quantitatively acceptable and was instead considered in a qualitative manner. Overall, effects to
chronic apical endpoints lor this genus. reported in this publication and Zhang et al. (2013a),
generally appear as a threshold effect from n 25 mg/L to 1.0 mg/L, providing support for the
endpoint and effect le\ el selected lor quantilati\ e use in criterion derivation (i.e., intrinsic rate of
natural increase), and further suggests the chronic criterion magnitude is adequately protective of
the genus. Brachionus.
4,2,2.2 Consideration of Relatively Sensitive Freshwater Tests based on Qualitatively
Acceptable Data
This section focuses on qualitatively acceptable chronic tests that were most relevant to
informing the appropriateness of the chronic freshwater criterion. Specifically, those tests used to
qualitatively inform the chronic freshwater criterion magnitude were identified as most relevant
if they met all parameters listed below:
115
-------
1. reported effect concentrations that were less than or similar to (e.g., factor of two) the
chronic freshwater criterion magnitude;
2. evaluated an animal species;
3. conducted the test for a non-acute exposure duration (e.g., greater than seven days);
4. evaluated apical effects (i.e., long-term survival, growth, and/or reproduction), and;
5. not already discussed in the previous section (i.e., not a species discussed among the four
most sensitive genera).
The toxicity values summarized below were not used quanliUili\ ely lo derive the chronic
PFOA freshwater criterion. Results of each individual study (as well as the rationale why a study
was not quantitatively acceptable) were considered relative to the chronic criterion magnitude to
ensure the chronic PFOA criterion was not underproductive and to provide additional supporting
evidence of the potential toxicity of PFOA to aquatic organisms.
4.2.2.2.1 Genus: Chironomus (midge)
Stefani et al. (2014) conducted a chronic (I n generation) test of PFOA with a midge,
Chironomus riparius. The 10 generations (each approximately 2<~) to 28 days) were tested under
static conditions The NOI-C for the study, based on effect on emergence, reproduction or sex
ratio at the only concentration tested, was flflS1) mu |. PFOA. Marziali et al. (2019) provides
further analysis of the same chronic test conducted by Stefani et al. (2014; further described in
Appendix (i 2 2 3) by reporting measurements of alterative endpoints/responses. The LOEC
based on de\ elopmental time, adult weight was 0.0098 mg/L (time-weighted average; NOEC
<0.0098 mg I.) W hile Stefani et al. (2014) reported no effects across the chronic test, Marziali et
al. (2019) reported effects lo select generations. Overall, however, effects were sporadic with
reductions in growth observed in several generations. There were no effects on "survival,
development, or reproduction" and Marziali et al. (2019) concluded "no effects at population
level (population growth rate) were proved, thus a toxicity risk in real ecosystems at the tested
concentrations seems unlikely." The results from these studies were deemed not acceptable for
116
-------
quantitative use because of limited test concentrations assessed, and uncertainty pertaining to
sediment characteristics, as well as poor control survival in four of the 10 generations.
Quantitatively acceptable midge data (McCarthy et al. 2021) were available to derive the chronic
PFOA freshwater criterion and these chronic data further suggest Chironomus is relatively
tolerant to chronic PFOA exposures.
4.2.2.2.2 Genus: Danio (zebrafish)
Jantzen et al. (2017b) evaluated the effects of PFOA on the morphometric, behavioral
and gene expression in D. rerio exposed via five-day static unmeasured exposures . Zebrafish
embryos were exposed at three-hours post fertilization (hpf) to PFOA for 120-houi s in what is
equivalent to a rapid early-life stage test. The observation period in clean water was extended
beyond the exposure time points from 12<> hpf u> 14 days post fertilization (dpf) to assess
possible latent effects. The five-day (plus nine-days lor observation) chronic value for growth-
based endpoints, including body length, was an M.VI'C of') (->325 mg'L (NOEC = 0.2 mg/L;
LOEC = 2.0 mg/L). Inn the M.VI'C for swimming activity, a non-apical endpoint, was reported
as 0.06325 mg/L (M)l-(' <).i)2 mg I.. I.()!¦(' = 0.2 mg/L). The reported chronic values based on
growth and swimming acti \ ity were not considered quantitatively acceptable because of the
relati \ el\ brief chronic (i e . Ii\ e-day) exposure duration compared to other acceptable acute
exposures that indicated D. ivrio was tolerant to brief (i.e., 96-hours) PFOA exposures.
4.2.2.2.3 Genus: ()r\~ias l medaka)
Ji et al. (2008) e\ aluated the chronic toxicity of PFOA (CAS # 335-67-1, purity not
provided) to the Japanese medaka, Oryzias latipes, via renewal unmeasured exposures. For the
F0 fish exposure study, breeding medaka pairs were exposed to PFOA for 14 days. Eggs were
counted every day, and the eggs spawned on the seventh day were saved for the F1 generation
exposure study. For the F1 fish exposure study, fertilized eggs collected from F0 fish were
117
-------
exposed until all living embryos had hatched. Newly hatched larvae were then randomly
transferred to 100 mL beakers and observed daily for swim-up success and survival for an
additional two weeks. Larvae were fed Artemia nauplii ad libitum twice daily. After 14 days,
replicates from each treatment group were transferred to beakers without PFOA for observation
through 100 days post hatch. The F0 (parental generation) adult survival, condition factor and
adult male and female GSI and HSI 14-day LOECs were all -in mu |. PFOA. For the F1
(progeny generation), the LOEC for larval survival was <> I mu I., while the corresponding
NOEC was considered <0.1 because effects were observed in the lowest concentration tested.
This test was not used quantitatively because uncertainties associated with the responses across
the range of concentrations tested. In many instances, authors did not report increasing chronic
effects with increasing concentrations thai differed hy an order of magnitude. Additionally,
endpoints associated with longer term effects to ju\ eniles were also be rejected because of
pseudoreplication resulting from a lack of replicates in the hatching stage. Since this test is a
static-renewal unmeasured test. I -PA chose to rely exclusively on the test by Lee et al. (2017) to
derive the SMC A' lor this species since l.ee et al (2017) was a flow-through measured test with
fewer concerns pertaining to test design (i e . no pseudoreplication) and results (lack of
increasing effects despite a ]<)-lbld increase in exposure concentrations).
4.2,2.3 Consideration of Qualitatively Acceptable Data from Missing Chronic MDRs
4.2.2.3.1 Another I'hv/imi or a Second Insect Order not Already Represented
Yuan et al. (2016b) conducted a 10-day renewal, unmeasured test on PFOA with the
planarian, Dugesia japonica (a non-North American species). No apical endpoints were
measured as the study focused on neural genes expression and neuronal morphology in the
planarian. The lowest test concentration, 0.5 mg/L, decreased the mRNA expression levels of
neural genes DjFoxD, DjotxA and DjotxB. Due to a lack of apical endpoints and insufficient test
118
-------
duration, the LOEC was not used quantitatively; however, it was considered acceptable for
qualitative use by providing relevant toxicity information to inform relative species sensitivity as
well as potential sublethal affects to inform mode of action and adverse outcome pathway (AOP)
considerations.
Yuan et al. (2017) conducted another 10-day static, unmeasured test on with the
planarian, Dugesiajaponica. No apical endpoints were measured as the study focused on stress
responses. The lowest test concentration, 0.5 mg/L, exhibited elevated lipid peroxidation,
increased mRNA expression levels of HSP40 and HSP70, two stress response genes. Due to a
lack of apical endpoints and test duration, the LOEC was not used quantitatively hul considered
acceptable for qualitative use by providing relevant toxicity information as well as potential
sublethal affects to inform mode of action and AOP considerations
Chronic toxicity values for Dugesia /a/'onica reported by Yuan et al. (2016b) and Yuan
et al. (2017) were not hasccl on apical effects associated with growth, survival or reproduction.
Therefore, the qualitali\ e data pi o\ ided by these studies did not provide useful information about
the relative sensiti\ ily of this species As a result, the MDR for another phylum or a second
insect order not already represented remains unfulfilled. However, authors did not report
noticeable effects to apical endpoints at the test concentration of 0.5 mg/L, suggesting the
chronic water column criterion (i.e., 0.094 mg/L) is protective of Dugesia japonica and
potentially other members of the unfulfilled MDR.
4.3 Evaluation of the Acute Insect Minimum Data Requirement through
Interspecies Correlation Estimates (ICE)
The acute data set for PFOA contained 18 genera (Table 3-3) representing seven of the
eight taxonomic MDR groups. The missing MDR was a representative from an insect family.
Evaluation of qualitatively acceptable insect data (i.e., Yang et al. 2014) relative to the acute
119
-------
criterion magnitude was the primary line of evidence used to inform insect sensitivity to acute
PFOA exposures (see section 3.1.1.1.5). Acute insect LCso data were estimated using Web-ICE
and compared to the acute criterion as a secondary line of evidence to evaluate insect sensitivity
to acute PFOA exposures.
EPA's web-ICE tool is described in detail in Appendix L.l.l. Briefly, ICE models are
log-linear regressions of the acute toxicity (EC50/LC50) of two species across a range of
chemicals, thus representing the relationship of inherent sensitivity between those species
(Raimondo et al. 2010). ICE models can be used predict the sensitivity of ail untested taxon
(predicted taxa are represented by the y-axis) from the known, measured sensi ti \ ity of a
surrogate species (represented by the x-axis). This analysis focused on all possible ICE models
that used insects as a predictor species (i.e . \ -axis) and a corresponding surrogate input species
(i.e., x-axis) for which a SMAV (see Table 3-3) was a\ ailahle These models are shown in Table
4-5 along with use classifications lor each individual model based on a host of statistical metrics
described by Willmi nu et al (2' 11 (\ see box one of Appendix L. 1.1 for additional discussion on
model use criteria).
120
-------
Table 4-5. All ICE models available in Web-ICE v3.3 for predicted insect species based on surrogates with measured PFOA.
Model parameters are used to evaluate prediction robustness. Cross-validation success is the percentage of all model data that were predicted within 5-fold of the
measured value through leave-one-out cross-validation (Willming et al. 2016). Taxonomic distance describes the relationship between surrogate and predicted
species (e.g., 1 = shared genus, 2 = shared family, 3 = shared order, 4 = shared class, 5 = shared phylum. 6 = shared kingdom).
Pn-dii'li-d S|K-iii-s
SuiTii»;ik-
Spi-iii-\
Sllipi"
Ink-rii-pl
l)i-»ivi-s
ill'
l"ivi-(liim
-
\ ;iliii-
Mi-sin
S(|ii;i ri-
ll mil-
(MSI!)
SuiTii»;ik-
miidi-l
minimum
\ ;ilui-
imk'D
Siiitii»:i li-
nn idi-l
iii;i\iniiiin
\ :d in-
Cniss-
\ ;ilid;iliull
Suiii-ss
< " I
T;i\iiniimii'
Diskinii-
I si-
( l:issilli":i 1 iiin
Atherix variegata
Lepomis
macrochirus
0.85
0.9
2
0.91
0.0428
0.08
0.36
59.53
100
6
Accepted
qualitatively
Atherix variegata
Oncorhynchus
mykiss
0.94
0.73
2
0.91
0.0439
0.08
0.61
59.27
100
6
Accepted
qualitatively
Chironomus plumosus
Americamysis
bahia
0.64
1.1
9
0.65
0.0026
0.97
0.01
5083
45
5
Rejected
Chironomus plumosus
Daphnia
magna
0.63
1.05
19
0.5
0.0002
1.14
0.13
i9000
29
5
Rejected
Chironomus plumosus
Lepomis
macrochirus
0.74
0.37
21
0.43
0.0006
I.I
0.77
45166
26
6
Rejected
Chironomus plumosus
Oncorhynchus
mykiss
0.78
0.3
21
0.5
0.0001
1.04
0.82
140000
35
6
Rejected
Chironomus plumosus
Pimephales
promelas
1.03
-0.46
15
0.64
0.0001
0.99
2.27
97000
35
6
Rejected
Chironomus tentans
Daphnia
magna
0.83
0.94
"
0.79
0.0011
1.03
0.32
472000
33
5
Rejected
Chironomus tentans
Lepomis
macrochirus
0.95
0.05
6
0.8
0.0027
0.88
2.85
517825
25
6
Accepted
Chironomus tentans
Oncorhynchus
mykiss
1.11
-0.64
>
0.81
0.005
0.95
11.24
905704
29
6
Accepted
qualitatively
Chironomus tentans
Pimephales
promelas
1.21
-1.04
>
0.8
0.006
1.34
19.63
766452
57
6
Rejected
Claassenia sabulosa
Americamysis
bahia
0.34
0.4
0.77
0.049
0.04
0.04
8.85
100
5
Rejected
Claassenia sabulosa
Lepomis
macrochirus
0.4
-0.34
-
0.63
0.0102
0.19
0.36
7326
78
6
Rejected
Claassenia sabulosa
Oncorhynchus
mykiss
0.42
-0.43
/
0.55
0.0213
0.23
0.61
1638
67
6
Rejected
Claassenia sabulosa
Pimephales
promelas
0.33
-0.62
6
0.63
0.0182
0.22
1.24
110000
75
6
Rejected
Paratanytarsus
dissimilis
Daphnia
magna
0.57
2.17
8
0.41
0.0441
1.96
0.66
1190000
50
5
Rejected
121
-------
Pivdiclcd Species
SlllTii»;ile
Species
Slope
IlllCI'Ccpl
l)c»rccs
ill'
I'lVl-lllim
-------
Table 4-6 shows model outputs from all the rejected, qualitatively acceptable, and
acceptable ICE models listed in Table 4-5. PFOA acute values are typically reported as mg/L and
are, therefore, often greater than the toxicity values used to develop an ICE model, meaning the
input PFOA LCso value of the surrogate was typically outside the model domain. In these cases,
the input toxicity value could be entered as [j,g/L and model would be allowed to extrapolate
beyond its range or the input toxicity value could be a "scaled" mu I. \ alue (i.e., estimate the
value as mg/L). Table 4-6 includes a column to denote whether the input toxicity data were |ig/L
or a "scaled" mg/L value for individual models. Please see Appendix L for further discussion on
the selection process for identifying whether a |iu I. or a "scaled" niu/L value ^as used as the
input toxicity value for individual models.
Within Table 4-6, bolded and underlined \ allies in the "llslimaled Toxicity" column
represent the estimated LC50 values from the acceptable models Only estimated toxicity values
from acceptable models were used to develop the estimated insect SMAVs reported in Table 4-6.
When more than one acceptable 1(11 model was a\ ailable for an individual predicted insect
species, all the acceptable estimated toxicity \ allies (i.e., LCso values) were taken together as a
geometric mean to represent the estimated SMAV
123
-------
Table 4-6. ICE-estimated Insect Species Sensitivity to PFOA.
Values in bold and underlined are used for estimated insect SMAYs
( 111111111 111
VlllK-
Pmliik-d Spiiiis
SuiTii»;ik- Spi'di's
ii»/1.
or
m»/l.
inpul
llsiiniiiii-d
lo\kii\
(in^/1.)
Ciiiilldi-iui-
Inli-nids (m»/l.)
SMAY
Snipefly
Atherix variegata
Lepomis macrochirus
mg/L
2027.47'1'
35.46- 115910.82
NA
Oncorhynchus mykiss
mg/L
6271.94ab
38.26 - 1028076.54
Midge
Chironomus plumosus
Americamysis bahia
Hg/L
8.33bc
0.44 - 159.47
NA
Daphnia magna
Hg/L
28.83bc
2.61-318.13
Lepomis macrochirus
Hg/L
51.17k'
2.95 - 888.56
Oncorhynchus mykiss
Hg/L
155.221"'
8.12 -2966.13
Pimephales promelas
Hg/L
210.74bc
12.78 - 3492.24
Midge
Chironomus tentans
Daphnia magna
Hg/L
251.45°
15.1 -4186.53
575.87
Lepomis macrochirus
mg/L
575.87
86.87- 3817.34
Oncorhynchus mykiss
mg/L
868.79a
97.49 - 7742.26
Pimephales promelas
Mg/L
621.10ac
12.55 - 30731.5
Stonefly
Claassenia sabulosa
Americamysis bahia
Hg/L
0.083bc
0.002 - 3.32
NA
Lepomis macrochirus
Hg/L
0.10bc
0.006- 1.8
Oncorhynchus mykiss
Hg/L
0.16bc
0.004 - 6.07
Pimephales promelas
Hg/L
0.01.7'*
0.002-0.13
Midge
Paralanylarsus dissimilis
Daphnia magna
Hg/L
168.49111'
5.34 -5318.67
1557.92
Lepomis macrochirus
mg/L
575.87
388.32- 12682.13
l.ilhobales catesbeianus
Hg/L
1627.49ac
30.27-87514.09
Oncorhynchus mykiss
mg/L
11063.24
3183.00- 38452.77
Pimephales promelas
Hg/L
593.51
110.45 - 3189.21
Midge
Parakinvlarsus
parthenogenelicus
Daphnia magna
Hg/L
568.96
313.37- 1033.01
890.25
Lepomis macrochirus
Hg/L
3147.45a
152.44-64987.96
Oncorhynchus mykiss
Hg/L
6405.37ac
55.24 - 742788.68
Pimephales promelas
Hg/L
1392.97
351.36- 5522.46
Stonefly
1'leronarcella baclia
Americamysis bahia
mg/L
67.06
10.47-429.35
67.06
Lepomis macrochirus
Hg/L
0.44bc
0.01 - 16.72
Oncorhynchus mykiss
Hg/L
3.27bc
0.15-69.67
Pimephales promelas
Hg/L
0.033bc
0.01-0.1
Stonefly
Pteronarcys californica
Daphnia magna
Hg/L
13.96bc
1.19-163.96
NA
Lepomis macrochirus
Hg/L
5.55bc
0.45 - 68.82
Oncorhynchus mykiss
Hg/L
10.6bc
0.52-215.73
a Both confidence intervals >1.5 order magnitude.
b Input data outside model range.
0 Guidance for model mean square error, R2, and/or slope not met.
124
-------
Overall, acceptable ICE models and empirical acute PFOA LCso values as model input
data were available to support the estimation of SMAVs for four individual insect species (Table
4-6). Estimated insect SMAVs ranged from 67.06 mg/L for the stonefly, Pteronarcella badia, to
1,557.92 mg/L for the midge, Paratanytarsus dissimilis. Beyond the P. badia estimated SMAV,
the remaining three estimated SMAVs were greater than the FAV (i.e., 97.14 mg/L; see section
3.2.1.1) by more than a factor of five. The FAV was greater than the /'. badia estimated SMAV
by nearly a factor of 1.5, suggesting P. badia may be relatively sen si li \ e lo acute PFOA
exposures. However, the P. badia estimated SMAV was based on a single estimated LCso value
of 67.06 mg/L with corresponding 95% confidence inlci\als thai ranged from 1 <» 47-429.35
mg/L.
Three of the four estimated insect SMAVs were greater than the FAV. Further,
qualitatively acceptable empirical insect toxicity data (i e . Yang el al. 2014; see section
3.1.1.1.5) suggested insects may not he among the most sensith e genera to acute PFOA
exposures. PFOA insect toxicity testing is an active and ongoing area of research within the
ecotoxicological scientific community that will likely provide information to evaluate the
sensiti\ ity of insects to acute and chronic PFOA exposures before the PFOA criteria document is
finalized I V.\ will continue to seek additional acute PFOA insect data to further evaluate the
sensitivity of insects
4.4 Acute to Chronic Ratios
When sufficient empirical data are not available, the 1985 Guidelines allow the use of a
Final Acute-to-Chronic Ratio (FACR) to convert the FAV to a FCV as an alternative approach to
derive the chronic criterion (U.S.EPA 1985). An ACR approach was not used for the derivation
of the chronic freshwater PFOA criterion. For descriptive purposes, 11 individual ACRs for four
125
-------
invertebrate species, one fish, and one amphibian could be calculated from the quantitatively
acceptable acute and chronic toxicity data (Appendix A and Appendix C). Appendix I includes
the ACRs for freshwater aquatic species with quantitatively acceptable acute values for which
comparable quantitatively acceptable chronic values were reported from the same study or same
investigator and laboratory combination. For each species where more than a single ACR was
calculated, Species Mean Acute-Chronic Ratios (SMACRs) w ere also calculated as the
geometric mean value of individual ACRs. In the case of a single ACR uiihin a species, that
ACR was the SMACR.
Individual ACRs ranged from 14.5 to 3,49} across all species and SMACRs ranged from
<17.68 to 3,493. Except for D. magna, all SMACRs consisted of a single ACR. Lithobates
catesbeiana had the largest ACR (i.e., 3,4l)3) The denominator of the /,. catesbeiana ACR was a
LOEC where authors reported a significant effect to snout \ enl length (SVL) despite a reduction
of only -7% relati\c to control responses (Flynn et al. 2d I1)) The -7% decrease to SVL
observed at the T.OI-C was a relati\ ely mild effect level compared to the denominator (i.e.,
chronic value) of most other I'l OA ACRs. with chronic effect levels typically being ECio values
orMATCs that had corresponding I .OI X s that produced a >10% effect. Consequently, the
relati\ el\ mild effect to SVI. observed by Flynn et al. (2019) may have contributed to an
artificially high ACR relati\ e in the other PFOA ACRs available.
Daphnia cannula had the second highest ACR (2,113) and the denominator was based on
a MATC, with the corresponding NOEC and LOEC (which reduced reproduction by -40%) that
differed by a factor of 10 (Logeshwaren et al. 2021). The 10X difference between the I),
carinata NOEC and LOEC published by Logeshwaran et al. (2021) likely contributed to an
artificially low MATC that, in turn, produced an artificially high ACR.
126
-------
Four out of the five D. magna ACRs ranged from 14.50 to 69.08, with the remaining
ACR from Lu et al. (2016) being 1,602. The D. magna ACR from Lu et al. (2016) was removed
from the SMACR calculation because it was an order of magnitude greater than other ACRs for
the species. Overall, the range of SMACRs was greater than a factor of 100. There was an
apparent relationship between SMACRs and SMAVs, but only after excluding the L. catesbeiana
ACR. The 1985 Guidelines do not provide for calculation of a I ACR under these circumstances.
However, if one were calculated as the geometric mean of the six SMACRs reported in
Appendix I, it would be 207.5.
4.5 Tissue-based Toxicity Studies Compared to the Chronic Tissue-based
Criteria
Tissue-based PFOA toxicity dala were reported for lour species (three fish and one frog
species) across five publications, all of which were classified as t|Lial i lati vely acceptable. Feng et
al. (2015) conducted a 96-hour study with ju\ enile goldfish (('arassms auratus) and observed no
effects of PFOA on mortality or antioxidant enzyme activity in the highest aqueous PFOA
treatment concentration (4 mu I.. measured), which corresponded to liver, gill, and muscle
PFOA concentrations of 17 11. 35 l.v and 6.07 mg/kg wet weight, respectively.
(iiari et al. (2< >!(•>) measured PI OA in several tissues of two-year-old common carp
(Cyprinus car/>n>) exposed to nominal PFOA water concentrations of 2 mg/L for 56 days.
Corresponding tissue PI OA concentrations in blood, liver, and muscle were 0.0649, 0.0281, and
0.0075 mg/kg wet weight, respectively. No effects of mortality, condition factor, hepatic somatic
index (HSI) or gonadal somatic index (GSI) were observed. Manera et al. (2017) performed a
separate study that replicated the study design of Giari et al. (2016), in which two-year-old
common carp (Cyprinus carpio) were exposed to nominal PFOA water concentrations of 2 mg/L
for 56 days. PFOA liver concentrations in fish exposed to 2 mg/L for 56 days were 0.0284 mg/kg
127
-------
wet weight, similar to Giari et al. (2016). No apical endpoints were reported; however, evidence
of degenerative liver morphology in PFOA exposed fish was observed.
Hagenaars et al. (2013) exposed adult zebrafish (D. rerio) to aqueous PFOA for 28 days.
Several reproductive and biochemical endpoints were measured. Whole-body PFOA
concentrations in the highest concentration (1 mg/L PFOA, nominal) after 28 days averaged
0.550 mg/kg wet weight in males and 0.301 mg/kg wet weighl in females No statistically
significant differences were observed in reproductive endpoints (total cuu production, fertilized
egg production, and hatching rate) for any treatment levels compared to controls Statistically
significant effects were observed among non-apical endpoints Decreased whole body glycogen
content was lower in male and female fish across all exposure treatments, and liver
mitochondrial electron transport activity was lower in males exposed to the highest PFOA
concentration. Differences in several liver proteins of PI-OA exposed males and females were
also observed.
Hoover et al (2<> I 7) exposed juvenile (Gosner stage 26) northern leopard frogs
(Lithobatespipicns. formerly. Rana pipicns) to three PFOA concentrations (10.5, 10.92, and
1,110 mu I. PI OA. iespecti\ely) Ibr4<) days Survival, growth (snout-vent length), and
developmental time were measured Whole body PFOA concentrations in frogs exposed to the
highest aqueous treatment le\ el averaged 3.87 mg/kg dry weight after 40 days. Tadpole moisture
content was not reported I n order to convert the reported dry weight concentrations to wet
weight concentrations, so that they would be more directly comparable to the whole-body fish
tissue criterion, a whole-body moisture content of 72.1% was applied, calculated as the average
for all fish collected as part of the USGS National Contaminant Biomonitoring Program (NCBP
Fish Database). The resulting whole-body concentration at the highest treatment level after 40
128
-------
days was 1.08 mg/kg wet weight. No effects of PFOA on mortality, growth, or development time
were reported.
Tissue PFOA concentrations reported in these qualitative studies were lower than the
tissue-based criteria calculated from BAFs, with the exceptions of the 96-hour liver- and muscle-
based NOECs of 17.11 mg/kg and 6.07 mg/L, respectively, reported by Feng et al. (2015), which
were greater than the corresponding liver tissue value of 0.221 mu ku and muscle tissue criterion
magnitude of 0.125 mg/kg. However, the liver- and muscle-based NOI X's reported by Feng et al.
(2015) were from an acute duration (96-hour exposure), whereas the tissue-based values were
derived to protect species from longer-term chronic exposures, where effects lo sensitive species
at concentrations lower than the whole body-based NOLC reported by Feng et al. (2015) may
occur.
Although all other tissue-based concentrations were lower than the corresponding tissue-
based criteria, no statistically significant effects of apical endpoints were observed in any of
these studies. Results of these studies do not provide any evidence that the aquatic community
will experience unacceptable chronic effects at tissue-based criteria magnitudes.
4.6 Hffects on Aquatic Plants
A\ ailahlc data for aquatic plants and algae were reviewed to determine if aquatic plants
were likely to be ad\ ersely affected by PFOA and if they were likely to be more sensitive to
PFOA than aquatic animals ( see Section 4 and Appendix E: Acceptable Freshwater Plant PFOA
Toxicity Studies). Toxicity values for freshwater plants reported in Appendix E were all greater
than the chronic freshwater criterion (i.e., 0.094 mg/L PFOA), with the exception of a green alga
(Chlorellapyrenoidosa) with a 96-hour growth-based NOEC of >0.1 mg/L (Li et al. 2021a).
Excluding the low NOEC reported by Li et al. (2021; NOEC >0.1 mg/L), effect concentrations
129
-------
for freshwater plants and algae ranged from 5.7 to 745.7 mg/L relative to animal chronic values
of 0.03162 to 88.32 mg/L (Appendix C). Therefore, it was not necessary to develop a criterion
based on the toxicity of PFOA to aquatic plants and the PFOA freshwater criteria are expected to
be protective of freshwater plants.
4.7 Summary of the PFOA Aquatic Life Criteria and the Supporting
Information
The PFOA aquatic life criteria were developed to protect aquatic life against adverse
effects, such as mortality, altered growth, and reproductive impairments associated with acute
and chronic exposure to PFOA. This Aquatic Life Ambient Water Quality Criteria lor
Perfluorooctanoic acid (PFOA) document includes water column based acute and chronic criteria
and tissue-based criteria for freshwaters. Acute and chronic water column criteria magnitudes for
estuarine/marine waters could not be deri\ ed at this time due to data limitations; however acute
estuarine/marine benchmarks are provided in Appendix I. The freshwater acute water column-
based criterion magnitude is -N mu I., and the chronic water column-based chronic criterion
magnitude is 0.094 mu I. The fish u hole-body tissue criterion magnitude is 6.10 mg/kg wet
weight, the iisli muscle tissue criterion magnitude is 0.125 mg/kg wet weight and the invertebrate
whole-body tissue criterion magnitude is I I I mg/kg wet weight (Table 3-11). Although
empirical PI OA toxicity data lor estuarine/marine species were not available to fulfill the eight
MDRs directly. I-PA included an acute aquatic life benchmark for estuarine/marine
environments in Appendix L, using available estuarine/marine species toxicity data and a NAM
application of ORD's peer-reviewed weblCE tool. The estuarine/marine acute water column-
based benchmark magnitude is 7.0 mg/L; this value provides information on a concentration that
should be protective of aquatic estuarine/marine life from acute PFOA exposures. As noted
130
-------
earlier, the benchmark value has greater uncertainty than the freshwater PFOA criteria, due to the
paucity of empirical data of PFOA effects on estuarine/marine organisms.
EPA evaluated the influence of including non-North American resident species on the
acute and chronic criteria magnitudes and concluded their inclusion did not substantiality affect
the criteria magnitudes. These PFOA aquatic life criteria are expected to be protective of aquatic
life, such as fish and aquatic invertebrates, on a national basis.
131
-------
5 REFERENCES
3M Company. 2000. Information on Perfluorooctanoic Acid and Supplemental Information on
Perfluorooctane Sulfonates and Related Compounds [FC-26 DATA], EPA/OTS Doc. #FYI-
OTS-0500-1378:4297 p.
3M Company. 2000. Information on Perfluorooctanoic Acid and Supplemental Information on
Perfluorooctane Sulfonates and Related Compounds [FC-143 DATA], EPA/OTS Doc. #FYI-
OTS-0500-1378:4297 p.
3M Company. 2000. Information on Perfluorooctanoic Acid and Supplemental Information on
Perfluorooctane Sulfonates and Related Compounds [FC-126 DATA], EPA/OTS Doc. #FYI-
OTS-0500-1378:4297 p.
3M Company. 2000. Information on Perfluorooctanoic Acid and Supplemental Information on
Perfluorooctane Sulfonates and Related Compounds [FC-1015 DATA], EPA/OTS Doc. #FYI-
OTS-0500-1378:4297 p.
3M Company. 2000. Information on Perfluorooctanoic Acid and Supplemental Information on
Perfluorooctane Sulfonates and Related Compounds [FX-1003 DATA], EPA/OTS Doc. #FYI-
OTS-0500-1378:4297 p.
3M Company. 2000. Information on Perfluorooctanoic Acid and Supplemental Information on
Perfluorooctane Sulfonates and Related Compounds [FX-1001 DATA], EPA/OTS Doc. #FYI-
OTS-0500-1378:4297p
3M Company. 2000 Voluntary I se and Exposure Information Profile Perfluorooctanic Acid and
Salts; U.S. EIW Administrate c Record ,\R226-<)51>5; U.S. Environmental Protection Agency:
Washington. DC. Zooo
3M Company 2<)o| Ln\iionmcnlal Monitoring-Multi-City Study Water, Sludge, Sediment,
POTW Lfflucnl and Landfill l.cachalc Samples. 3M Environmental Laboratory.
3M Company 2<>03. Perfluorooctanoic Acid Physiochemical Properties and Environmental Fate
Data. 3M Center. St Paul, \l\ EPA. Office of Pollution Prevention and Toxics (OPPT); OPPT-
2003-00 12.
Ahrens, L. 2011. Polyll uoroalkyl compounds in the aquatic environment: a review of their
occurrence and fate. J. Environ. Monit. 13(1): 20-31.
Ahrens, L. and M. Bundschuh. 2014. Fate and effects of poly- and perfluoroalkyl substances in
the aquatic environment: a review. Environ. Toxicol. Chem. 33(9): 1921-1929.
Ahrens L, S. Felizeter, R. Sturm, Z. Xie, and R. Ebinghaus. 2009. Polyfluorinated compounds in
waste water treatment plant effluents and surface waters along the River Elbe, Germany. Mar.
Pollut. Bull. 58: 1326-1333.
132
-------
Ahrens, L., M. Shoeib, S. Del Venlo, G. Codling, and C. Halsall. 2011a. Polyfluoroalkyl
compounds in the Canadian Arctic atmosphere. Environ. Chem. 8: 399-406.
Ahrens L., M. Shoeib, T. Harner, S.C. Lee, R. Guo, and E. J. Reiner. 201 lb. Wastewater
Treatment Plant and Landfills as Sources of Polyfluoroalkyl Compounds to the Atmosphere.
Environ. Sci. Technol. 45: 8098-8105
Allen, M.B. and D.I. Arnon. 1955. Studies on nitrogen-fixing blue green algae. I Growth and
nitrogen fixation by Anabaena cylindrica Lemm. Plant Physiol. 30: 366-372.
Anderson, R. H., G. C. Long, R. C. Porter, and J. K. Anderson. 2016. Occurrence of select
perfluoroalkyl substances at U.S. Air Force aqueous film-forming foam release sites other than
fire-training areas: Field-validation of critical fate and transport properties. Chemosphere. 150:
678-685.
Ankley, G.T., P. Cureton, R. A. Hoke. M. Houde, A. Kumar, J. Kurias, R. Lanno, C. McCarthy,
J. Newsted, C. J. Salice, B. E. Sample, M. S. Sepulveda, J. Steevens, and S. Valsecchi. 2020.
Assessing the ecological risks of per- and polyfluoroalkyl substances: Current state-of-the
science and a proposed path forward. Environ. Toxicol. Chem. 00(00): 1-42.
Annunziato, K.M. 2018. Low molecular weight PI-AS alternate es (C-6) result in fewer cellular
and behavioral alterations than long chain (C-K (-9) PFASS in larval zebrafish. Ph.D. Thesis,
Rutgers, The State University of New Jersey, New Brunswick, NJ: 188 p.
Armitage, J. M.. I Schenker. M Scheringer, J. W. Martin, M. MacLeod and I. T. Cousins.
2009. Modeling the global fine and transport of perfluorooctane sulfonate (PFOS) and precursor
compounds in relation to temporal trends in wildlife exposure. Environ. Sci. Technol.. 43(24):
9274-9280.
Arukwe, A and A S Moi tensen. 2d I I Lipid peroxidation and oxidative stress responses of
salmon fed a diet containing peril uoroocUine sulfonic- or perfluorooctane carboxylic acids.
Comp liiochem. Physiol Part (' 154 2SX-2l)5.
ASTM (American Society for Testing and Materials). 1993. Standard guide for conduction acute
toxicity tests with fishes, macroinvertebrates, and amphibians. Annual Book of ASTM
Standards. 88-721)
ASTM (American Society for Testing and Materials). 1999. Standard guide for conducting acute
toxicity tests on test materials with fishes, macroinvertebrates, and amphibians: designation E
729-96. In: Annual book of ASTM standards: water and environmental technology. 11th edition.
224-244.
ASTM (American Society for Testing and Materials). 2006. Standard guide for conducting
laboratory toxicity tests with freshwater mussels. E2455-06. In Annual Book of ASTM
Standards. Philadelphia, PA.
133
-------
ATSDR (Agency for Toxic Substances and Disease Registry). 2015. Toxicological Profile for
Perfluoroalkyls. Draft for Public Comment. Agency for Toxic Substances and Disease Registry,
Public Health Service, United States Department of Health and Human Services, Atlanta, GA.
Accessed May 2016. http://www.atsdr.cdc.gov/ToxProfiles/tp200.pdf.
Awkerman, J.A., S. Raimondo, C.R. Jackson, andM.G. Barron. 2014. Augmenting species
sensitivity distributions with interspecies toxicity estimation models. Environ. Toxicol. Chem.
33: 688-695.
Barmentlo, S.H., J.M. Stel, M. van Doom, C. Eschauzier, P. de Voout and M.H.S. Kraak. 2015.
Acute and chronic toxicity of short chained perfluoroalkyl substances to Daphnia magna.
Environ. Pollut. 198: 47-53.
Bartlett, A.J., A.O. De Silva, D.M. Schissler, A.M. Hedges, L.R. Brown, k Shires, J. Miller, C.
Sullivan and C. Spencer. 2021. Lethal and sublethal toxicity of perfluorooctanoic acid (PFOA) in
chronic tests with Hyalella azteca (amphipod) and early-life stage tests with J'tnu-phales
promelas (fathead minnow). Ecotoxicol. Environ. Sal" 2<">7' l<~)
Beach, S. A., J. L. Newsted, K. Coady and J. P. Giesy 2<)<)(•> Ixotoxicological evaluation of
perfluorooctanesulfonate (PFOS). Reviews of environmental contamination and toxicology. 133-
174.
Bejarano, A.C. and J.R Wheeler. 2020. Scientific basis lor expanding the use of interspecies
correlation estimation models. Integr. Environ. Assess Manage. 16(4): 525-530.
Bejarano, A.C., S. Raimondo, and M.G. Barron. 2017. Framework for optimizing selection of
interspecies correlation estimation models to address species diversity gaps in an aquatic
database. Environ. Sci. Teclinol 51 SI 5S-S1(->5.
BenningholV. A I), \V 11 liisson. I) (' Koch. D.J. Ehresman, S.K. Kolluri and D.E. William.
2011 F.stiogen-Like Acli\ ily of Perlluoioalkyl Acids In Vivo and Interaction with Human and
Rainbow Trout Estrogen Receptors In Vitro Toxicol. Sci. 120(1): 42-58.
BenningholV. A I)., G.A. Orner, C.H. Buchner, J.D. Hendricks, A.M. Duffy and D.E. Williams.
2012. Promotion of Hepatocarcinogenesis by Perfluoroalkyl Acids in Rainbow Trout. Toxicol.
Sci. 125(1): 69-78.
Benskin, J.P., D C. Muir, B.F. Scott, C. Spencer, A.O. De Silva, H. Kylin, J W. Martin, A.
Morris, R. Lohmann, G. Tomy, B. Rosenberg, S. Taniyasu, andN. Yamashita. 2012.
Perfluoroalkyl acids in the Atlantic and Canadian Arctic Oceans. Environ. Sci. Technol. 46(11):
5815-5823.
Bernardini, I., V. Matozzo, S. Valsecchi, L. Peruzza, G.D. Rovere, S. Polesello, S. Iori, M.G.
Marin, J. Fabrello, and M. Ci. 2021. The New PFAS C604 and Its Effects on Marine
Invertebrates: First Evidence of Transcriptional and Microbiota Changes in the Manila Clam
Ruditapesphilippinarum. Environ. Int. 152: 17 p.
134
-------
Berlin, D., P. Labadie, B. J. D. Ferrari, A. Sapin, J. Garric, O. Geffard, H. Budzinski and M.
Babut. 2016. Potential exposure routes and accumulation kinetics for poly- and perfluorinated
alkyl compounds for a freshwater amphipod: Gammarus spp. (Crustacea). Chemosphere. 155:
380-387.
Borgmann, U., D.T. Bennie, A.L. Ball, and V. Palabrica. 2007. Effect of a mixture of seven
pharmaceuticals on Hyalella azteca over multiple generations. Chemosphere. 66(7): 1278-1283.
Boudreau, T.M. 2002. Toxicity of perfluorinated organic acids to selected freshwater organisms
under laboratory and field conditions. Chapter 3: Toxicology of peril uoroalkyl carboxylic acids
(PFCAs) in relation to carbon-chain length. Masters of Science Thesis. I diversity of Guelph,
Ontario, Canada. December 17, 2002.
Boudreau, T. M., P. K. Sibley, S. A. Mabury, D. G. C. Muir, and K. R. Solomon 2003.
Laboratory Evaluation of the Toxicity of Perfluorooclane Sulfonate (PFOS) on Sclanas/rum
capricornatum, Chlorella vulgaris, Lemna gibba, Daphnia magna, and Daphnia puticaria. Arch
Environ Contam Toxicol. 44: 307-313.
Boulanger, B., J. Vargo, J.L. Schnoor and T\ C TTornbuckle 2<>'»4 Detection of perfluorooctane
surfactants in Great Lakes water. Environ Sei Technol. 38( 15) 4< )64-4070.
Bouwmeester, M.C., S. Ruiter, T. Lommelaars, J. Sippel. 11 M. Hodemaekers, E.J. van den
Brandhof, J.L.A. Pennings, J.H. Kamstra, J. Jelinek, J.I'.J. Issa, J. Legler and L.T.M. van der
Ven. 2016. Zebrafish embryos as a screen for DNA methylation modifications after compound
exposure. Toxicol. Applied Pharmacol 201: 84-96
Brill, J.L., S.E. Behinuer. .1 Chancy. S I) Dyer, S. Raimondo, M.G. Barron, and C.A. Pittinger.
2016. Development of algal inter-species correlation estimation (ICE) models for chemical
hazard assessment I jniion Toxicol ( hem. 35. 2368-2378.
Buck. R (' . .1 I'rank I in. I lieruer. .1 M Conder, I.T. Cousins, P. de Voogt, A.A. Jensen, K.
Kannan, S A Mabury and S P .1. van Leeuwen. 2011. Perfluoroalkyl and polyfluoroalkyl
substances in the environment Terminology, classification, and origins. Integrat. Environ.
Assess. Manag 7(4) 513-541
Burkhard, L. P. 2021. V.\ al nation of published bioconcentration factor (BCF) and
bioaccumulation factor (BAF) data for per-and polyfluoroalkyl substances across aquatic
species. Environ. Toxicol. Chem..
Burns, D. C., D.A. Ellis, H. Li, C.J. McMurdo, and E. Webster. 2008. Experimental pKa
determination for perfluorooctanoic acid (PFOA) and the potential impact of pKa concentration
dependence on laboratory-measured partitioning phenomena and environmental modeling.
Environ. Sci. Technol. 42(24): 9283-9288.
135
-------
Butt, C.M., U. Berger, R. Bossi, and G.T. Tomy. 2010. Levels and trends of poly- and
perfluorinated compounds in the arctic environment. Sci. Total. Environ. 48(2010): 2936-2965.
Butt, C.M., D.C.G. Muir, and S.A. Mabury. 2014. Biotransformation pathways of fluorotelomer-
based polyfluoroalkuyl substances: A review. Environ. Toxicol. Chem. 33(2): 243-267.
CCME (Canadian Council of Ministers of the Environment). 2007. Protocol for the derivation of
water quality guidelines for the protection of aquatic life. Winnipeg, Manitoba: Canadian
Council of Ministers of the Environment. Available online at: http://ceqg-
rcqe.ccme.ca/ downl oad/en/220.
Coats, K., G. Long, and J. Adelson. 2017. Quality Systems Manual (QSM) for Environmental
Laboratories. Department of Defense, Washington, DC.
Cochran, R. S. 2015. Evaluation of perfluorinated compounds in sediment, water, and passive
samplers collected from the Barksdale Air Force Base. MS Thesis. Texas Tech I diversity,
Lubbock, TX United States. https://ttu-ir.tdl.org/bitstream/handle/2346/63633/COCHR.AN-
THESIS-2015. pdf? sequence= 1 &i s Allowed=v
Colombo, I., W. de Wolf, R.S. Thompson. D G Farrar, R A I loke. and J. L'Haridon. 2008.
Acute and chronic aquatic toxicity of ammonium perfluorooctanoale (APFO) to freshwater
organisms. Ecotoxicol. Environ. Saf. 71: 74l>-75(\
Colorado Department of Public Health and the Environment 2<)20. PFAS sampling project
results: surface water.
Consoer, D.M. 2017 A mechanistic in\ estigation of perfluoroalkyl acid kinetics in rainbow trout
{Oncorhynchus myktss) A dissertation submitted to the faculty of the University of Minnesota.
Consoer. I) M .AD I lolVman. P.\ I'itzsimmons, P.A. Kosian, and J.W. Nichols. 2014.
Toxicokinetics of peilluoiooclanoalc (PI OA) in rainbow trout (Oncorhynchus mykiss). Aquat.
Toxicol 15o (->5-73
Corrales, J., I..A kristofco, VV.B Steele, G.N. Saari, J. Kostal, E.S. Williams, M. Mills, E.P
Gallagher, T.J ka\ anagh, N. Simcox, L.Q. Shen, F. Melnikov, J.B. Zimmerman, A.M.
Voutchkova-Kostal. P T Anastas, and B.W. Brooks. 2017. Toward the design of less hazardous
chemicals: Exploring comparative oxidative stress in two common animal models. Chem. Res.
Toxicol. 30: 893-904
CRC CARE (Cooperative Research Centre for Contamination Assessment and Remediation of
the Environment). 2017. Assessment, management and remediation for PFOS and PFOA Part 4:
application of HSLs and ESLs. Technical Report No. 38.
https://www.crccare.com/files/dmfile/CRCCARETechReport38Part4 Assessmentmanagementan
dremediationforPFOSandPFOA ApplicationofESLsandHSLs2.pdf
136
-------
Cui, Y., W. Liu, W. Xie, W. Yu, C. Wang, and H. Chen. 2015. Investigation of the effects of
perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) on apoptosis and cell
cycle in a zebrafish (Danio rerio) liver cell line. Int. J. Environ. Res. Public Health 12(12):
15673-15682.
Dasgupta, S., A. Reddam, Z. Liu, J. Liu, and D.C. Volz. 2020. High-Content Screening in
Zebrafish Identifies Perfluorooctanesulfonamide as a Potent Developmental Toxicant. Environ.
Pollut. 256: 113550.
Dasu, K. 2011. Evaluating the biotransformation potential of commercial model fluorotelomer
monomers in soils. PhD Thesis, Purdue University.
Dasu, K., J. Liu, andL.S. Lee. 2012. Aerobic soil biodegradation of S 2 fluorotelomer stearate
monoester. Environ. Sci. Technol. 46: 3831-3836.
Dasu, K., L.S. Lee, R.F. Turco, andL.F. Nies. 2013 Aerobic biodegradation of S 2
fluorotelomer stearate monoester and 8:2 fluorotelomer citrate triester in forest soil
Chemosphere. 91: 399-405.
Delinsky, A.D., M.J. Strynar, S.F. Nakayama. .T T.. Varus. \ 1} Ye. P.J. McCann, and A.B.
Lindstrom. 2009. Determination of ten perfluorinated compounds in bluegill sunfish (Lepomis
macrochirus) fillets. Environ. Res. 109(8) l)75-l>X4
Delinsky, A.D., M.J. Strynar, P.J. McCann, J.L. Varus, L. McMillan, S.F. Nakayama, and A.B.
Lindstrom. 2010. Geographical distribution of perfluorinated compounds in fish from Minnesota
lakes and rivers. Environ. Sci. Teclinol. 44: 2549-2554.
Del Vento, S., C. Halsall, R. Gioia, K. Jones, and J. Dachs. 2012. Volatile per- and
polyfluoroalkvl compounds in the remote atmosphere of the western Antarctic Peninsula: an
indirecl source of peril uoroulkv1 acids lo Antarctic waters? Atmos. Poll. Res.3(4): 450-455.
Department of Defense. Strategic I ji\ iron mental Research and Development Program (SERDP).
2019. Guidance for Assessing the Ecological Risks of PFASs to Threatened and Endangered
Species at Aqueous Tilrn Forming Foam-Impacted Sites SERDP Project ER18-1614, July 2019.
De Silva, A.O., P .1 Tseng, and S.A. Mabury. 2009. Toxicokinetics of Perfluorocarboxylate
Isomers in Rainbow Trout Environ. Toxicol. Chem. 28: 330-337.
De Silva, A. O., C. Spencer, B.F. Scott, S. Backus, and D.C. Muir. 2011. Detection of a cyclic
perfluorinated acid, perfluoroethylcyclohexane sulfonate, in the Great Lakes of North America.
Environ. Sci. Technol. 45(19): 8060-8066.
Ding, G.H., T. Fromel, E.J. van den Brandhof, R. Baerselman, and W.J.G.M. Peijnenburg.
2012a. Acute toxicity of poly- and perfluorinated compounds to two cladocerans, Daphnia
magna and Chydorus sphaericus. Environ. Toxicol. Chem. 31(3): 605-610.
137
-------
Ding, G., M. Wouterse, R. Baerselman and W.J.G.M. Peijnenburg. 2012b. Toxicity of
polyfluorinated and perfluorinated compounds to lettuce (.Lactuca sativa) and green algae
(Pseudokirchneriella subcapitata). Arch. Environ. Contam. Toxicol. 62: 49-55.
Ding, G., J. Zhang, M. Wang, Y. Chen, G. Luo, and D. Xiong. 2012c. Evaluation and Prediction
of Mixture Toxicity of PFOS and PFOA to Zebrafish (Danio rerio) Embryo. Adv. Mater. Res.
485: 297-300.
Ding, G., J. Zhang, Y. Chen, L. Wang, M. Wang, D. Xiong, and Y. Sun. 2013. Combined effects
of PFOS and PFOA on zebrafish {Danio rerio) embryos. Arch. En\ iron. Contam. Toxicol. 64(4):
668-675.
Dinglasan, M.J.A., Y. Ye, E.A. Edwards, and S.A. Mabury. 2004. Fluorolelomer alcohol
biodegradation yields poly- and perfluorinated acids. Environ. Sci. Technol 38 2857-2864.
Dinglasan-Panlilio, J. M., S.S. Prakash, and J.E. Baker. 2014. Perfluorinated compounds in the
surface waters of Puget Sound, Washington and Clayoquot and Barkley Sounds, British
Columbia. Mar. Pollut. Bull. 78(1-2): 173-180.
Du, G., H. Huang, J. Hu, Y. Qin, D. Wu, T. Song, Y. Xia, and X. Wang. 2013. Endocrine-related
effects of perfluorooctanoic acid (PFOA) in /chrafish, H295R steroidenesis and receptor reporter
gene assays. Chemosphere 91: 1099-1 lOo
DuPont Haskell Laboratory. 2000. Summaries of studies conducted at DuPont Haskell
Laboratory with ammonium peril uoiooctanoate and perfluorononanoate (with cover letter dated
05252000).
Dyer, S.D., D.J Yerslccu. S I- lielanuer. .1 G Chancy, andF.L. Mayer. 2006. Interspecies
correlation estimates predict protect i\ e en\ ironmenlal concentrations. Environ. Sci. Technol. 40:
3102-31 I I.
Dyer. S l).l) J. Vcrstceu. S I- Belanuer. J.G. Chaney, S. Raimondo, andM.G. Barron. 2008.
Comparison of species sensili\ ity distributions derived from interspecies correlation models to
distributions used to derive water quality criteria. Environ. Sci. Technol. 42 (8): 3076-3083.
EFSA (European I'ood Safely Authority). 2008. Opinion of the scientific panel on contaminants
in the food chain on peril uorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) and
their salts. EFSA Journal 653:1-131.
EGLE (Michigan Department of Environment, Great Lakes, and Energy). 2010. Rule 57 water
quality values. Surface Water assessment Section.
El on en, C. ECOTOXicology Knowledgebase System User Guide Version 5.3. US
Environmental Protection Agency, Cincinnati, OH, EPA/600/R-20/087, 2020.
138
-------
EN ISO. 1993. Water Quality ----- Fresh Water Algal Growth Inhibitor Test with Scenedesmus
subspicatus and Selenastrum capricorutum (ISO 8692:1993). European Committee for
Standardization, Brussels.
EN ISO. 1995. Water Quality ----- Marine Algal Growth Inhibition Test with Skeletonema costatum
and Phaeckiactylum tricornutum (ISO 10253:1995). European Committee for Standardization,
Brussels.
Environment Canada and Health Canada. 2012. Screening Assessment Report, Perfluorooctanoic
Acid, its Salts, and its Precursors. Accessed May 2016.
EPAV (Environment Protection Authority Victoria). 2016. Commonwealth Environmental
Management Guidance on Perfluorooctane Sulfonic Acid (PFOS) and Periluorooctanoic Acid
(PFOA). Australian Government, Department of the Environment and Energy.
Fabbri, R., M. Montagna, T. Balbi, E. Raffo, F. Palumbo, and L. Canesi. 2014. Adaptation of the
bivalve embryotoxicity assay for the high throughput screening of emerging contaminants in
Mytilus galloprovincialis. Mar. Environ. Res. 99: 1-8.
Fairbrother, A. Risk Management Safety Factor. Tn Encyclopedia of Ecology vol 4,
Ecotoxicology; J0rgensen, S.R., Fath, B.I). Lds. F.lsevier: Oxford. 2008.
Falk, S., K. Failing, S. Georgii, H. Brunn, and T. Stahl. 2d I 5 Tissue specific uptake and
elimination of perfluoroalkyl acids (PFAAs) in adult rainbow irout (Oncorhynchus mykiss) after
dietary exposure Cheniosphere 129 150-156.
Fang, S., X. Chen. S Zhao. Y /hang, W. Jiang, L. Yang, and L. Zhu. 2014. Trophic
magnification and isomer lYaclionalion of perfluoroalkyl substances in the food web of Taihu
Lake, China Liniron Sci Technol 4S 2I73-2IS2.
Fang. X . Y Wei. Y l.iu. .1 Wang. and.I Dai. 2010. The identification of apolipoprotein genes
in rare minnow ((iohiocypns ranis) and their expression following perfluorooctanoic acid
exposure Comp. Biochem Physiol. Part C 151: 152-159.
Feng, C.l. . L W u. S I). Dyer. U. Change, and X.L. Zhao. 2013. Derivation of freshwater quality
criteria for zinc using interspecies correlation estimation models to protect aquatic life in China.
Chemosphere. 90: I 177-1 I S3.
Feng, M., Q. He, L. Meng, X. Zhang, P. Sun, and Z. Wang. 2015. Evaluation of single and joint
toxicity of perfluorooctane sulfonate, perfluorooctanoic acid, and copper to Carassius auratus
using oxidative stress biomarkers. Aquat. Toxicol. 161: 108-116.
Fernandez-Sanjuan, M., M. Faria, S. Lacorte, and C. Barata. 2013. Bioaccumulation and effects
of perfluorinated compounds (PFCs) in zebra mussels (Dreissenapolymorpha). Environ. Sci.
Pollut. Res. 20: 2661-2669.
139
-------
Flynn, R.W., M.F. Chislock, M.E. Gannon, S.J. Bauer, B.J. Tornabene, J.T. Hoverman, and M.
Sepulveda. 2019. Acute and Chronic Effects of Perfluoroalkyl Substance Mixtures on Larval
American Bullfrogs (Rana catesbeiana). Chemosphere. 236: 7 p.
Flynn, R.W., M. Iacchetta, C. De Perre, L. Lee, M.S. Sepulveda, and J.T. Hoverman. 2021.
Chronic Per-/Polyfluoroalkyl Substance Exposure Under Environmentally Relevant Conditions
Delays Development in Northern Leopard Frog {Ranapipiens) Larvae. Environ. Toxicol. Chem.
40(3): 711-716.
Fojut, T.L., A.J. Palumbo, and R.S. Tjeerdema. 2012a. Aquatic life water criteria derived via the
UC Davis Method: II. Pyrethroid Insecticides. In. Tjeerdema, R S . ed . Springer, NY, NY. Rev.
Environ. Contam. Toxicol. 216: 51-104.
Fojut, T.L., A.J. Palumbo, and R.S. Tjeerdema. 2012b. Aquatic life water criteria derived via the
UC Davis Method: III. Diuron. In. Tjeerdema, R.S., ed., Springer, NY, NY Re\ Environ.
Contam. Toxicol. 216:105-142.
Fromel, T. and T.P. Knepper. 2010. Fluorotelomer ethoxvlates Sources of highly lluorinated
environmental contaminants Part I: Biotransformation. Chemosphere. 80: 1387-92.
Furdui, V.I., P.W. Crozier, E.J. Reiner, and S A Vlabury. 2008 Trace level determination of
perfluorinated compounds in water by direct injection. Chemosphere. 73(1 Suppl): S24-30.
Gergs, A., S. Classen, T. Strauss, R. Ottermanns, T. Brock, H. Ratte, U. Hommen, and T. Preuss.
2016. Ecological Recovery Potential of Freshwater Organisms: Consequences for Environmental
Risk Assessment of Chemicals. Rev. Environ. Con. Toxicol. . 236: 259-294.
Gewurtz, S B., S.M. Backus. A () l)e Silva, L. Ahrens, A. Armellin, M. Evans, S. Fraser, M.
Gledhill, P. Guerra, T. I lamer. P A I lelm, H. Hung, N. Khera, M.G. Kim, M. King, S.C. Lee,
R.J. Letcher, P. Martin. (' \lar\ in. I) .1. McGoldrick, A.L. Myers, M. Pelletier, J. Pomeroy, E. J.
Reiner. \1. Rondeau, M (' Same. M Sekela, M. Shoeib, D.W. Smith, S.A. Smyth, J. Struger, D.
Spry. .1 Svrgiannis, and .1 Wallho Z<)|3 Perfluoroalkyl acids in the Canadian environment:
multi-media assessment of current status and trends. Environ. Int. 59: 183-200.
Giari, L., I' Yincenxi. S. Badini, C. Guerranti, B.S. Dezfuli, E.A. Fana, and G. Castaldelli. 2016.
Common carp ('\priims car/>i<> responses to sub-chronic exposure to perfluorooctanoic acid.
Environ. Sci. Pollut Res 23: 15321-15330.
Giesy, J. P. and K. Kannan. 2001. Global distribution of perfluorooctane sulfonate in wildlife.
Environmental science & technology. 35(7): 1339-1342.
Giesy, J. P. and K. Kannan. 2002. Perfluorochemical Surfactants in the Evironment. Envrion Sci
Technol. 147-152.
Giesy, J. P., J.E. Naile, J.S. Khim, P.D. Jones, and J.L. Newsted. 2010. Aquatic toxicology of
perfluorinated chemicals. Rev. Environ. Contam. Toxicol. 202: 1-52.
140
-------
Godfrey, A.E. 2017. Endocrine disrupting effects of halogenated chemicals on fish. Chapter 4:
Sex specific endocrine disrupting effects of halogenated chemicals in Japanese medaka. Ph.D.
Thesis Dissertation, Purdue University, Department of Forestry and Natural Resources, West
Lafayette, IN.
Godfrey, A., A. Abdel-moneim, and M.S. Sepulveda. 2017a. Acute mixture toxicity of
halogenated chemicals and their next generation counterparts on zebrafish embryos.
Chemosphere 181: 710-712.
Godfrey, A., B. Hooser, A. Abdelmoneim, K.A. Horzmann, J.L. l-'reeman, and M.S. Sepulveda.
2017b. Thyroid disrupting effects of halogenated and next generation chemicals on the swim
bladder development of zebrafish. Aquat. Toxicol. 193: 228-235.
Goecke-Flora, C M. and N. V. Reo. 1996. Influence of carbon chain length on the hepatic effects
of perfluorinated fatty acids. A 19F- and 31P-NMR investigation. Chem. Res. Toxicol.9(4):689-
95.
Goeritz, I., S. Falk, T. Stahl, C. Schafers, and C. Schlechliicm 2013. Biomagnification and
Tissue Distribution of Perfluoroalkyl Substances (PFASs) in Market-Size Rainbow Trout
(Oncorhynchus mykiss). Environ. Toxicol Chun 32(9): 2078-2<)SS
Gonzalez-Naranjo, V. andK. Boltes. 2014. Toxicity of il->ii|">iolcn and perfluorooctanoic acid for
risk assessment of mixtures in aquatic and terrestrial en\ iionmenls Int. J. Environ. Sci. Technol.
11: 1743-1750.
Gorrochategui, E., S l.acoile. R Tucker. andF.L. Martin. 2016. Perfluoroalkylated substance
effects inXenopus laevis \(-> kidney epithelial cells determined by ATR-FTIR spectroscopy and
chemometric analysis Chem. Res Toxicol 2^.^24-932.
Green, A. 2014. Invasive species report Mediterranean mussel. University of Washington,
Seattle, WA. Accessed January 2020: https://depts.washington.edu/oldenlab/wordpress/wp-
content/uploads/2015/09/Mytilus galloprovincialis Green 2014.pdf
Greenberg, B.M.. \-D Huang, and D.G. Dixon. 1992. Applications of the higher aquatic plants
Lemna gibba for ecotoxicological risk assessment. J. Aquat. Ecosyst. Health. 1:147-155.
Guo, R., E.J. Reiner, S.IJ. Bhavsar, P.A. Helm, S.A. Mabury, E. Braekevelt, and S.A. Tittlemier.
2012. Determination of polyfluoroalkyl phosphoric acid diesters, perfluoroalkyl phosphonic
acids, perfluoroalkyl phosphinic acids, perfluoroalkyl carboxylic acids, and perfluoroalkane
sulfonic acids in lake trout from the Great Lakes Region. Anal. Bioanal. Chem. 404: 2699-2709.
Hagenaars, A., L. Vergauwen, W. De Coen, and D. Knapen. 2011. Structure-activity relationship
assessment of four perfluorinated chemicals using a prolonged zebrafish early life stage test.
Chemosphere 82: 764-772.
141
-------
Hagenaars, A., L. Vergauwen, D. Benoot, K. Laukens, and D. Knapen. 2013. Mechanistic
toxicity study of perfluorooctanoic acid in zebrafish suggests mitochondrial dysfunction to play a
key role in PFOA toxicity. Chemosphere 91: 844-856.
Hansen, K. J., H. Johnson, J. Eldridge, J. Butenhoff, and L. Dick. 2002. Quantitative
characterization of trace levels of PFOS and PFOA in the Tennessee River. Environ. Sci.
Technol. 36(8): 1681-1685.
Hanson, M.L., J. Small, P.K. Sibley, T.M. Boudreau, R.A. Brain, S.A. Mabury, and K.R.
Solomon. 2005. Microcosm evaluation of the fate, toxicity, and risk lo aquatic macrophytes from
perfluorooctanoic acid (PFOA). Arch. Environ. Contam. Toxicol 4l) 3<~>7-316.
Hassell, K.L., T.L. Coggan, T. Cresswell, A. Kolobaric, K. Berry, N.l) Crosbie, J. Blackbeard,
V.J. Pettigrove, and B.O. Clarke. 2020. Dietary Uptake and Depuration Kinetics of
Perfluorooctane Sulfonate, Perfluorooctanoic Acid, and Hexafluoropropy lene Oxide Dimer Acid
(GenX) in a Benthic Fish. Environ. Toxicol. Cheni 30(3): 595-603.
Haukas, M., U. Berger, H. Hop, B. Gulliksen, and G. W Gahiielsen. 2007. Bioaccumulation of
per- and polyfluorinated alkyl substances (PFAS) in selected species from the Barents Sea food
web. Environ. Pollut. 148(1): 360-371.
Hayman, N.T., G. Rosen, M.A. Colvin, J Condor, and .1 A Arblaster 2021. Aquatic toxicity
evaluations of PFOS and PFOA for five standard murine endpoinls. Chemosphere 273: 7 p.
Hazelton, P.D. 2013. Emerging methods for emerging contaminants: novel approaches to
freshwater mussel toxicity testing. University of Georgia.
Hazelton, P.D., W.G. Cope, T.J. Pandolfo, S. Mosher, M.J. Strynar, M.C. Barnhart, and R.B.
Bringolf. 2012a Partial Life-Cycle and Acute Toxicity of Perfluoroalkyl Acids to Freshwater
Mussels l-miron Toxicol. Chem., 3 1: 1611-1620.
HekskT. I'M. R.W. Laane. and P. de Voogt. 2003. Environmental and toxicity effects of
perfluoroalkx lated substances llev. Environ. Contam. Toxicol. 179: 99-121.
HEPA (Heads of I- PAs Australia and New Zealand). 2020. PFAS National Environmental
Management Plan Version 2 <) National Chemicals Working Group of the Heads of EPAs
Australia and New Zealand (https://www.awe.gov.au/sites/default/files/documents/pfas-nemp-
2.pdf)
Higgins C. and R. Luthy. 2006. Sorption of Perfluorinated Surfactants on Sediments. Environ.
Sci. Technol. 40(23): 7251-7256.
Holth, T.F., M. Yazdani, A. Lenderink, and K. Hyllan. 2012. Effects of fluoranthene and
perfluorooctanoic acid (PFOA) on immune functions in Atlantic cod (Gadus morhua). Abstracts
Comp. Biochem. Physiol. Part A. 163: S39-S42.
142
-------
Hong, S., J.S. Khim, T. Wang, J.E. Naile, J. Park, B.O. Kwon, S.J. Song, J. Ryu, G. Codling,
P.D. Jones, Y. Lu, and J.P. Giesy. 2015. Bioaccumulation characteristics of perfluoroalkyl acids
(PFAAs) in coastal organisms from the west coast of South Korea. Chemosphere. 129:157-63.
Hoover, G.M., M.F. Chislock, B.J. Tornabene, S.C. Guffey, Y.J. Choi, C.D. Perre, J.T.
Hoverman, L.S. Lee, and M.S. Sepulveda. 2017. Uptake and depuration of four
per/polyfluoroalkyl substances (PFASS) in Northern leopard frog Ranapipiens tadpoles.
Environ. Sci. Technol. Letters 4(10): 399-403.
Houde, .M, A. Trevor, D. Bujas, J. Small, R.S. Wells, P.A. Fair, G.D. Bossart, K.R. Solomon, D.
Muir. 2006. Biomagnification of perfluoroalkyl compounds in the bolllenose dolphin (Tursiops
truncatus) Food Web. Environ. Sci. Technol. 40(13): 4138-4144.
Houde, M., G. Czub, J.M. Small, S. Backus, X. Wang, M. Alaee, and D. (' Muir. 2008.
Fractionation and bioaccumulation of perfluorooctane sulfonate (PFOS) isomers in a Lake
Ontario food web. Environ. Sci. Technol. 42(24): 9397-9403.
Houde, M., J.W. Martin, R.J. Letcher, K.R. Solomon and D.C. Muir. 2006. Biological
monitoring of polyfluoroalkyl substances: A review. Environ. Sci. Technol. 40(11): 3463-3473.
HSDB (Hazardous Substances Data Bank) 2<) 12 Perfluorooclanoic acid. Accessed May 2016.
Hu, C., Q. Luo, and Q. Huang. 2014. Ecotoxicological effects of pertluorooctanoic acid on
freshwater microalgae Chlamydomonas reinhanhn and SccuciL-smus ohliquus. Environ. Toxicol.
Chem. 33(5): 1I21M 134
Huset, C.A., A.C Chiaia. I) I' liarolsky, N. Jonkers, H. P.E. Kohler, C. Ort, W. Giger, and J.A.
Field. 2008. Occurrence and mass flows of fluorochemicals in the Glatt Valley watershed,
Switzerland l-miron Sci Technol 42( 17). (->369-6377.
Inouc. Y.N I lashi/ume. \ Yakala. 11 Murakami, Y. Suzuki, E. Kikushima, and M. Otsuka.
2012. I nique |ih\ sieoehemieal properties of perfluorinated compounds and their
bioconcenli alion in common carp Cypnnus carpio L. Arch. Environ. Contam. Toxicol. 62(4):
672-680.
Jantzen, C.E., K M Annun/ialo, and K.R. Cooper. 2016. Behavioral, morphometric, and gene
expression effects in adult zebrafish (Danio rerio) embryonically exposed to PFOA, PFOS, and
PFNA. Aquat. Toxicol. 180: 123-130.
Jantzen, C.E., F. Toor, K.A. Annunziato, and K.R. Cooper. 2017a. Effects of chronic
perfluorooctanoic acid (PFOA) at low concentration on morphometries, gene expression, and
fecundity in zebrafish (Danio rerio). Reproduct. Toxicol. 69: 34-42.
Jantzen, C.E., K.A. Annunziato, S.M. Bugel, and K.R. Cooper. 2017b. PFOS, PFNA and PFOA
sub-lethal exposure to embryonic zebrafish have different toxicity profiles in terms of
morphometries behavioral and gene expression. Aquat. Toxicol. 175: 168-170.
143
-------
Jeon, J., K. Kannan, H.K. Lim, H.B. Moon, J.S. Ra, and S.D. Kim. 2010a. Bioaccumulation of
perfluorochemicals in Pacific oyster under different salinity gradients. Environ. Sci. Technol.
44(7): 2695-2701.
Jeon J., K. Kannan, H.K. Lim, H.B. Moon, and S.D. Kim. 2010b. Bioconcentration of
perflourinated compounds in blackrock fish, Sebastes schlegeli, at different salinity levels.
Environ. Toxicol. Chem. 29: 2529-2535.
Ji, K., Y. Kim, S. Oh, B. Ahn, H. Jo, and K. Choi. 2008. Toxicity of periluorooctane sulfonic
acid and perfluorooctanoic acid on freshwater macroinvertebrales {I kiphnia magna and Moina
macrocopa) and fish (Oryzias latipes). Environ. Toxicol. Chem. 27( 1<)): 2159-2168.
Jones, P.D., W. Hu, W. De Coen, J.L. Newsted, and J.P. Giesy. 2003. Binding of perfluorinated
fatty acids to serum proteins. Environmental Toxicology and Chemistry: An International
Journal. 22(11): 2639-2649.
Kaiser, M.A., B.S. Larsen, C-P.C. Kao, and R.C. Buck. 2<)()5 Vapor pressures of
perfluorooctanoic, -nonanoic, -decanoic, undecanoic, and dodecanoic acids. Jo. Chem. Eng.
Data. 50(6): 1841-1843.
Kalasekar, S.M., E. Zacharia, N. Kessler. \ A Ducharme. J Guslalsson, I.A. Kakadiaris, andM.
Bondesson. 2015. Identification of environmental chemicals that induce yolk malabsorption in
zebrafish using automated image segmentation. Reproducl Toxicol. 55: 20-29.
Kang, J.S., T.G. Ahn. and .1 \Y Park 2<~>19 Perfluorooctanoic Acid (PFOA) and Perfluooctane
Sulfonate (PFOS) Induce Different Modes of Action in Reproduction to Japanese Medaka
{Oryzias latipes). J 11 a/a id Mater 3oS 1)7-I03.
Kannan, K. 2') I I Perlluoioalky1 and polylluoroalkyl substances: current and future perspectives.
Environ Chem. 8(4)
Kannan, K.I. Tao, E. Sinclair. S I) Pastva, D.J. Jude, and J.P. Giesy. 2005. Perfluorinated
compounds in aquatic organisms at various trophic levels in a Great Lakes food chain. Arch.
Environ. Contain Toxicol 48(4): 559-566.
Keiter, S., L. Baumann. 11 Farber, H. Holbech, D. Skutlarek, M. Engwall, and T. Braunbeck.
2012. Long-Term Effects of a Binary Mixture of Perfluorooctane Sulfonate (PFOS) and
Bisphenol A (BPA) in Zebrafish (Danio rerio). Aquat. Toxicol. 118/119: 116-129.
Kelly, B.C., M.G. Ikonomou, J.D. Blair, B. Surridge, D. Hoover, R. Grace, and F.A. Gobas.
2009. Perfluoroalkyl Contaminants in an Arctic Marine Food Web: Trophic Magnification and
Wildlife Exposure. Environ. Sci. Technol.(43): 4037-4043.
144
-------
Khan, E.A., X. Zhang, E.M. Hanna, F. Yadetie, I. Jonassen, A. Goksoyr, and A. Arukwe. 2021.
Application of Quantitative Transcriptomics in Evaluating the Ex Vivo Effects of Per- and
Polyfluoroalkyl Substances on Atlantic Cod (Gadus morhua) Ovarian Physiology. Sci. Total
Environ. 755(1): lip.
Kim, M., J. Son, M.S. Park, Y. Ji, S. Chae, C. Jun, J.S. Bae, T.K. Kwon, Y.S. Choo, H. Yoon, D.
Yoon, J. Ryoo, S.H. Kim, M.J. Park, and H.S. Lee. 2013. In vivo evaluation and comparison of
developmental toxicity and teratogenicity of perfluoroalkyl compounds using Xenopus embryos.
Chemosphere 93: 1153-1160.
Kim, S.-K. and K. Kannan. 2007. Perfluorinated acids in air, rain, snow, surface runoff, and
lakes: relative importance of pathways to contamination of urban lakes F.nviron. Sci. Technol.
41(24): 8328-8334.
Kim, W.K., S.K. Lee, and J. Jung. 2010. Integrated assessment of biomarker responses in
common carp (Cyprinus carpio) exposed to perfluorinaled organic compounds .1 I laz. Mat. 180:
395-400.
Konwick, B.J., G.T. Tomy, N. Ismail, J.T. Peterson, R.J l'au\ or, D. Higginbotham, and A.T.
Fisk. 2008. Concentrations and patterns of perfluoroalkyl acids in (reorgia, USA surface waters
near and distant to a major use source. Em iron Toxicol. Chem 27( 10); 2011-2018.
Kudo, N., M. Katakura, Y. Sato, and Y. Kawashima. 2002. Sex hormone-regulated renal
transport of perfluorooctanoic acid. Chem. Biol. Interact. 139(3): 301-16.
Kwadijk, C., P. korylar. and A koelmans. 2010. Distribution of perfluorinated compounds in
aquatic systems in the Netherlands I jiviron. Sci. Technol. 44(10): 3746-3751.
Lasier, P.J., J W W ashington. S M I lassan. and T.M. Jenkins. 2011. Perfluorinated chemicals in
surface waters and sediments from northwest Georgia, USA, and their bioaccumulation in
Lumbriculus varieuatiis Ijniron Toxicol Chem. 30(10): 2194-2201.
Latala, A.M. Nedzi, and P Stepnowski. 2009. Acute toxicity assessment of perfluorinated
carboxylic acids towards the lialtic microalgae. Environ. Toxicol. Pharmacol. 28: 167-171.
Lath, S., E.R. Knight. I) A Navarro, R.S. Kookana, and M. J. McLaughlin. 2019. Sorption of
PFOA onto different laboratory materials: Filter membranes and centrifuge tubes. Chemosphere
222: 671-678.
Lau, C., K. Anitole, C. Hodes, D. Lai, A. Pfahles-Hutchens, and J. Seed. 2007. Perfluoroalkyl
acids: a review of monitoring and toxicological findings. Toxicol. Sci. 99(2): 366-394.
Le, T. T.Y., and W. J. G. M. Peijnenburg. 2013. Modeling toxicity of mixtures of
perfluorooctanoic acid and triazoles (triadimefon and paclobutrazol) to the benthic cladoceran
chydorus sphaericus. Environ. Sci. Technol. 47(12): 6621-6629.
145
-------
Lee, B.C., J.D. Hendricks, and G.S. Bailey. 1991. Toxicity of microtoxins in the feed of fish. In:
Mycotoxins and Animal Feedstuff: Natural Occurrence, Toxicity, and Control (Smith JE, ed.).
Boca Raton, FL: CRC Press, 607-626.
Lee, FL, J. D'eon, and S.A. Mabury. 2010. Biodegradation of polyfluoroalkyl Phosphates as a
Source of Perfluorinated Acids to the Environment. Environ. Sci. Technol. 44(9): 3305-3310.
Lee, J.J. and I.R. Schultz. 2010. Sex Differences in the uptake and disposition of
perfluorooctanoic acid in fathead minnows after oral dosing. Environ. Sci. Technol. 44(44): 491-
496.
Lee, J.W., J.W. Lee, K. Kim, Y.J. Shin, J. Kim, S. Kim, H. Kim, P. Kim. and K. Park. 2017.
PFOA-induced metabolism disturbance and multi-generational reproducti\ e toxicity in Oryzias
latipes. J. Haz. Mat. 340: 231-240.
Lee, J.W., P.K. Seong, S.D. Yu, and P. Kim. 2020. Adverse effects of perfluoroalkvl acids on
fish and other aquatic organisms: A review. Sci. Total Environ. 707: 135334.
Lee, W. and Y. Kagami. 2010. Effects of perfluorooctanoic acid and perfluorooctane sulfonate
on gene expression profiles in medaka (Oryzias latipes). Abstracts Toxicol. Letters 196S: S37-
S351.
Lemly, A. D. 1997. Ecosystem recovery following selenium contamination in a freshwater
reservoir. Ecotox. and Environ. Safety. 36(3): 275-281.
Lescord, G.L., K.A. Kidd, A.O. De Silva, M. Williamson, C. Spencer, X.W. Wang, andD.C.G.
Muir. 2015. Perfluorinated and polyfluorinated compounds in lake food webs from the Canadian
High Arctic. Environ. Sci. Technol. 49: 2694-2702.
Li, F., X.L. I'ang, / M Zhou, X li l.iao, J. Zou, B.L. Yuan, and W.J. Sun. 2019. Adsorption of
perfluorinated acids onto soils' Kinetics, isotherms, and influences of soil properties. Sci. Tot.
Environ 649-504-514
Li, F., V. Yu. M Guo, Y. Lin. Y. Jiang, M. Qu, X. Sun, Z. Li, Y. Zhai, andZ. Tan. 2021a.
Integrated analysis of physiological, transcriptomics and metabolomics provides insights into
detoxication disruption of PI-OA exposure in Mylilus edulis. Ecotoxicol. Environ. Saf. 214: 11
pp.
Li, H., D. Ellis, and D. Mackay. 2007. Measurement of low air-water partition coefficients of
organic acids by evaporation from a water surface. Jour. Chem. Engin. Data. 52(5): 1580-1584.
Li, M.H. 2008. Effects of nonionic and ionic surfactants on survival, oxidative stress, and
cholinesterase activity of planarian. Chemosphere. 70(10): 1796-1803.
Li, M.H. 2009. Toxicity of perfluorooctane sulfonate and perfluorooctanoic acid to plants and
aquatic invertebrates. Environ. Toxicol. 24(1): 95-101.
146
-------
Li, M.H. 2010. Chronic effects of perfluorooctane sulfonate and ammonia perfluorooctanoate on
biochemical parameters, survival and reproduction of Daphnia magna. J. Health Sci. 56(1): 104-
111.
Li, M.H. 2011. Changes of cholinesterase and carboxylesterase activities in male guppies,
Poecilia reticulate, after exposure to ammonium perfluorooctanoate, but not to perfluorooctane
sulfonate. Fresenius Environ. Bull. 20(8a): 2065-2070.
Li, Y., T. Fletcher, D. Mucs, K. Scott. C.H. Lindh, P. Tallving, and K. Jakobsson. 2017. Half-
lives of PFOS, PFHxS and PFOA after end of exposure to contaminated drinking water. Occup.
Environ. Med. 75: 46-51.
Li, Y.S., D.P. Oliver, and R.S. Kookana. 2018. A critical analysis of published data to discern
the role of soil and sediment properties in determining sorption of per and polylluoroalkyl
substances (PFASs). Sci. Tot. Environ. 628-629: 1 in. 120.
Li, Y., X. Liu, X. Zheng, M. Yang, X. Gao, J. Huang, L. /hanu. and Z. Fan. 2021 b. Toxic effects
and mechanisms of PFOA and its substitute GenX on the photosynthesis of Chlorella
pyrenoidosa. Sci. Tot. Environ.765: 14443 I
Liang, X. and J. Zha. 2016. Toxicogenomic applications of Chinese rare minnow (Gobiocypris
rarus) in aquatic toxicology. Comp. Biochcm Physiol Pai l D 19' 174-180.
Lide, D.R. 2007 CRC I landhook of Chemistry and Physics S8lh ed. CRC Press, Taylor &
Francis, Boca Ralon. II. .1 412
Liou, J.S.C., B. Szoslek. C M DeRilo. and F. L. Madsen. 2010. Investigating the
biodegradability of peilliioiooclanoic acid Chemosphere. 80: 176-83.
Liu, C and K.Y II Gin 2<)|S Imnuinoloxicity in green mussels under perfluoroalkyl substance
(PFAS) exposure Re\ eisiMe response and response model development. Environ. Toxicol.
Chem 37(4) 1138-1145.
Liu, C., Y l)u. and li. Zhou 2<)07a. Evaluation of estrogenic activities and mechanism of action
of perfluorinated chemicals determined by vitellogenin induction in primary cultured tilapia
hepatocytes. Aquat Toxicol. 85: 267-277.
Liu, C., K. Yu, X. Shi, J. Wang, P.K.S. Lam, R.S.S. Wu, and B. Zhou. 2007b. Induction of
oxidative stress and apoptosis by PFOS and PFOA in primary cultured hepatocytes of freshwater
tilapia (Oreochromis niloticus). Aquat. Toxicol. 82: 135-143.
Liu, C., V.W. Chang, and K.Y. Gin. 2013. Environmental Toxicity of PFCs: An Enhanced
Integrated Biomarker Assessment and Structure-Activity Analysis. Environ. Toxicol. Chem.
32(10): 2226-2233.
147
-------
Liu, C., V.W.C. Chang, K.Y.H. Gin, and V.T. Nguyen. 2014a. Genotoxicity of perfluorinated
chemicals (PFCs) to the green mussel (Perna viridis). Sci. Tot. Environ. 487: 117-122.
Liu, C., V.W.C Chang, and K.Y.H. Gin. 2014b. Oxidative toxicity of perfluorinated chemicals in
green mussel and bioaccumulation factor dependent quantitative structure-activity relationship.
Environ. Toxicol. Chem. 33(10): 2323-2332.
Liu, C., K.Y.H Gin and V.W.C. Chang. 2014c. Multi-biomarker responses in green mussels
exposed to PFCs: Effects at molecular, cellular, and physiological levels. Environ. Sci. Pollut.
Res. 21: 2785-2794.
Liu, J. and S. Mejia Avendano. 2013. Microbial degradation of polyfluoroalkyl chemicals in the
environment: A review. Environ. Intern. 61: 98-1 14.
Liu, J., N. Wang, B. Szostek, R.C. Buck, P.K. Panciroli, P.W. Folsom, L.M. Sulecki, and C.A.
Bellin. 2010. 6-2 Fluorotelomer alcohol aerobic biodegradation in soil and mixed bacterial
culture. Chemosphere. 78: 437-444.
Liu, W., S. Chen, X. Quan, and Y.H. Jin. 2008a. Toxic effect of serial perfluorosulfonic and
perfluorocarboxylic acids on the membrane system of a freshwater alga measured by flow
cytometry. Environ. Toxicol. Chem. 27(7) I 5^7-1604.
Liu, Y., J. Wang, Y. Wei, H. Zhang, Y. Liu, and J. Dai 2<)i)Sb Molecular characterization of
cytochrome P450 1A and 3 A and the effects of perfluorooctanoic acid on their mRNA levels in
rare minnow {Gobiocypris rams). Aquat. Toxicol. 88: 183-l^n
Liu, Y., J. Wang, Y. Liu, H. Zhang, M. Xu, and J. Dai. 2009. Expression of a novel cytochrome
9450 4T gene in rare minnow ((Jobiocypris rams). Comp. Biochem. Physiol. Part C 150: 57-64.
Logeshwaran, P , A K. Si\aram. A Surapaneni, K. Kannan, R. Naidu, andM. Megharaj. 2021.
Exposure to perfluorooclanesulfonale (PFOS) but not perfluorooctanoic acid (PFOA) at ppb
concent rati on induces chronic toxicity in Daphnia carinata. Sci. Tot. Environ.769: 8 p.
Loi, I- I.I. \V Yeung, S. Taniyasu, P.K. Lam, K. Kannan, andN. Yamashita. 2011. Trophic
Magnification of Poly- and Perfluorinated Compounds in a Subtropical Food Web. Environ. Sci.
Technol.(45): 55«»o-5513
Loos, R., J. Wollgast, T. Huber, and G. Hanke. 2007. Polar herbicides, pharmaceutical products,
perfluorooctanesulfonate (PFOS), perfluorooctanoate (PFOA), and nonylphenol and its
carboxylates and ethoxylates in surface and tap waters around Lake Maggiore in Northern Italy.
Anal. Bioanal. Chem. 387(4): 1469-1478.
Loos, R., G. Loroco, T. Huber, J. Wollgast, E.H. Cristoph, A. de Jager, B.M. Gawlik, G. Hankee,
G. Umlauf, and J-M. Zaldivar. 2008. Analysis of perfluorooctanoate (PFOA) and other
perfluorinated compounds (PFCs) in the river PO watershed in N-Italy. Chemosphere. 71: 306-
313.
148
-------
Loos, R., B.M. Gawlik, G. Locoro, E. Rimaviciute, S. Contini, and G. Bidoglio. 2009. EU-wide
survey of polar organic persistent pollutants in European river waters. Environ. Pollut. 157(2):
561-568.
Lu, G.H., B.H. Ma, S. Li, and L.S. Sun. 2016. Toxicological Effects of Perfluorooctanoic Acid
(PFOA) on Daphnia magna. Mat. Sci. Environ. Eng.: 559-564.
MacDonald, M.M., A.L. Warne, N.L. Stock, S.A. Mabury, K.R. Solomon, andP.K. Sibley.
2004. Toxicity of perfluorooctane sulfonic acid and perfluorooctanoic acid to Chironomus
tentans. Environ. Toxicol. Chem. 23(9): 2116-2123.
Mahapatra, C.T., N.P. Damayanti, S.C. Gufifey, J.S. Serafin, J. Irudayaraj. and M.S. Sepulveda.
2017. Comparative in vitro toxicity assessment of perfluorinated carboxylic acids J. Applied
Toxicol. 37: 699-708.
Manera, M., L. Giari, F. Vincenzi, C. Guerranti, J.A. DePasquale, and G. Castaldclli 2017.
Texture analysis in liver of common carp (Cyprinns carpio) sub-chronically exposed to
perfluorooctanoic acid. Ecol. Indicators 81: 54-64.
Martin, J.W., S.A. Mabury, K.R. Solomon, and DC. Muir. 2003a. Dietary accumulation of
perfluorinated acids in juvenile rainbow trout (Oncorhynchus mykiss). Environ. Toxicol. Chem.
22(1): 189-195.
Martin, J.W., S.A. Mabury, K.R. Solomon, and D.C.G. Muir. 2003b. Bioconcentration and tissue
distribution of perfluorinated acids in rainbow trout (Oncorhynchus mykiss). Environ. Toxicol.
Chem. 22: 196-204
Martin, J.W., M.M. Smilhwick, B.M. Braiine. P I-'. Hoekstra, D.C. Muir, and S.A. Mabury. 2004.
Identification oflonu-chain perfluorinated acids in biota from the Canadian Arctic. Environ. Sci.
Tech 38(2)' 373-3Xn
Martin, .1 \V . .1 A Brian, S IJeeson, J.D. Benskin, and M.S. Ross. 2010. PFOS of PreFOS? Are
perfluorooclanc sulfonate precursors (PreFOS) important determinants of human and
environmental peril uorooctoune sulfonate (PFOS) exposure? J. Envrion. Monit. 12:1979-2004.
Martin, J.W., S.A. Mahurx. K.R. Solomon and D.C.G. Muir. 2013. Progress toward
understanding the bioaccumulation of perfluorinated alkyl acids. Environ. Toxicol. Chem.
32(11): 2421-2423.
Marwood, C.A., K.R. Solomon, and B.M. Greenberg. 2001. Chlorophyll fluorescence as a
bioindicator of effects on growth in aquatic macrophytes from mixtures of polycyclic aromatic
hydrocarbons. Envrion. Toxicol. Chem. 20: 890-898.
149
-------
Marziali, L., F. Rosignoli, S. Valsecchi, S. Polesello, andF. Stefani. 2019. Effects of
perfluoralkyl substances (PFASs) on a multigenerational scale: A case study with Chironomus
riparius (Diptera, Chironomidae). Environ. Toxicol. Chem. 38(5): 988-999.
McCarthy, C.J., S.A. Roark, D. Wright, K. O'Neal, B. Muckey, M. Stanaway, J. Rewerts, J.
Field, T. Anderson, and C.J. Sa. 2021. Toxicological response of Chironomus dilutus in single
chemical and binary mixture exposure experiments with 6 perfluoralkyl substances. Environ.
Toxicol. Chem. 40(8): 2319-2333.
Mejia-Avendano, S., S. VoDuy, S. Sauve and J. Liu. 2016. Generation of perfluoroalkyl acids
from aerobic biotransformation of quaternary ammonium polvfluoroalkx 1 surfactants. Environ.
Sci. Technol. 50(18): 9923-9932.
Mhadhbi, L., D. Rial, S. Perez, and R. Beiras. 2012. Ecological risk assessment of
perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) in marine environment
using Isochrysis galbana, Paracentrotus lividus, Siric/la aiwata and Psetui maxima J. Environ.
Monit. 14(5): 1375-1382.
Miranda, A.F., C. Trestrail, S. Lekamge, and D. Nugegoda 2<)20. Effects of Perfluorooctanoic
Acid (PFOA) on the Thyroid Status, Vitellogenin, and Oxidant-Antioxidant Balance in the
Murray River Rainbowfish. Ecotox.29(2): 163-174.
MPCA (Minnesota Pollution Control Agency). 200S PCIs in Minnesota's Ambient
Environment: 2008 Progress Report.
Moody, C.A., J.W. Martin, W.C. Kwan, D.C. Muir, and S.A. Mabury. 2002. Monitoring
perfluorinated surfactants in biota and surface water samples following an accidental release of
fire-fighting foam into Etobicoke Creek. Environ. Sci. Technol. 36(4): 545-551.
Moody, ( A . (i \ I lehert. S.H Strauss, and J.A. Field. 2003. Occurrence and persistence of
perfluorooclanesulfonate and other perfluorinated surfactants in groundwater at a fire-training
area at Wurtsmith Air Force Base, Michigan, USA. J. Environ. Monit. 5(2): 341-345.
Mortensen, A S . K.J. Letcher. M.V. Cangialosi, S. Chu, and A. Arukwe. 2011. Tissue
bioaccumulation patterns, \enobioticbiotransformation and steroid hormone levels in Atlantic
salmon (Salmo salar) fed a diet containing perfluoroactane sulfonic or perfluorooctane
carboxylic acids. Chemosphere. 83: 1035-1044.
Nakata, H., K. Kannan, T. Nasu, H.-S. Cho, E. Sinclair, and A. Takemura. 2006. Perfluorinated
contaminants in sediments and aquatic organisms collected from shallow water and tidal flat
areas of the Ariake Sea, Japan: environmental fate of perfluorooctane sulfonate in aquatic
ecosystems. Environ. Sci. Technol. 40(16): 4916-4921.
Nakayama, S., M.J. Strynar, L. Helfant, P. Egeghy, X. Ye, and A. B. Lindstrom. 2007.
Perfluorinated compounds in the Cape Fear drainage basin in North Carolina. Environ. Sci.
Technol. 41(15): 5271-5276.
150
-------
Nakayama, S. F., M. J. Strynar, J. L. Reiner, A. D. Delinsky, and A. B. Lindstrom. 2010.
Determination of Perfluorinated Compounds in the Upper Mississippi River Basin. Environ. Sci.
Technol. 44:4103-4109.
Newsted, J.L., R. Holem, G. Hohenstein, C. Lange, M. Ellefson, W. Reagen, and S. Wolf. 2017.
Spatial and temporal trends of poly- and perfluoroalkyl substances in fish fillets and water
collected from pool 2 of the Upper Mississippi River. Environ. Toxicol. Chem. 36(11): 3138-
3147.
Nguyen, V. T., M. Reinhard, and G. Y. Karina. 2011. Occurrence and source characterization of
perfluorochemicals in an urban watershed. Chemosphere. 82(9): 1277-1285.
NJDEP. (New Jersey Department of Environmental Protection). 2019. ln\ csliuation of Levels of
Perfluorinated Compounds in New Jersey Fish, Surface Water, and Sediment New Jersey
Department of Environmental Protection Division of Science, R., and Environmental
Health,SRI 5-010. pp. 1-46.
https://www.ni. gov/dep/dsr/publications/Investigation%2<)pp'„20Levels%20of%20l->erfluorinated
%20Compounds%20in%20New%20Jersev%20Fish.%2<)Sui-|'ace%20Water.%20and%20Sedime
nt.pdf
NMED (New Mexico Environment Depart menu 2< >21 PI -AS. Data Accessed January 2021.
Available online at: https://www.env.nm uo\ pl'as data
Norwegian Institute for Air Research. 2007a. ]S|}\ l)78-S2-425-1962-7. Available at:
www.nilu.no.
Norwegian Institute for Air Research 2<)()7b TSBN 978-82-425-1984-9. Available at:
www.nilu.no.
Nowak. (' . T. Hankeln. I- R Schmidt, and Tv. Schwenk. 2006. Development and localization of
microsatellite markers lor the si hi i nu species Chironomus riparius and Chironomus piger
(Diplera. Chironomidae). Mol Ecol. Notes 6: 915-917.
Nowak, ('. I) .lost. ('. Vogt. M. Oetken, K. Schwenk, and J. Oehlmann. 2007a. Consequences of
inbreeding and reduced genetic variation on tolerance to cadmium stress in the midge
Chironomus riparius \<.|iiat. Toxicol. 85: 278-284.
Nowak, C., A. Czeikowitz, C. Vogt, M. Oetken, B. Streit, and K. Schwenk. 2008. Variation in
sensitivity to cadmium among genetically characterized laboratory strains of the midge
Chironomus riparius. Chemosphere 71: 1950-1956.
Nowak, C., C. Vogt, M. Pfenninger, K. Schwenk, J. Oehlmann, B. Streit, and M. Oetken. 2009.
Rapid genetic erosion in pollutant-exposed experimental Chironomid populations. Environ.
Pollut. 157: 881-886.
151
-------
NRC (National Research Council). 2013. Assessing risks to endangered and threatened species
from pesticides; National Academies Press, National Research Council. Washington, DC. p 142.
Nunez, O., J.D. Hendricks, D.N. Arbogast, A.T. Fong, B.C. Lee, and G.S. Bailey. 1989.
Promotion of aflatoxin B1 hepatocarcinogenesis in rainbow trout by 17P-estradiol. Aquat.
Toxicol. 15: 289-302.
Oakes, K.D, P.K. Sibley, K.R. Solomon, S.A. Mabury, and G.J. Van Der Kraak. 2004. Impact of
perfluorooctanoic acid on fathead minnow (Pimephalespromelas) fatty acyl-CoA oxidase
activity, circulating steroids, and reproduction in outdoor microcosms. Environ. Toxicol. Chem.
23(8): 1912-1919.
OECD. 1984. Guideline for testing of chemicals 202 -Daphnia sp., Aculc I in mobilization Test
and Reproduction Test.
OECD. 1992. Test No. 203: Fish, Acute Toxicity Tesl OECD Guidelines for the Testing of
Chemicals, Section 2, OECD Publishing, Paris, Imps doi.org/10.1787/9789264069961-en.
OECD. 1997. Guideline 211: Daphnia magna reproduction test.
OECD. 1998. Daphnia magna reproduction test. OECD Guideline 211. Paris, France.
OECD. 2000a. Guideline 202: Daphnia magna, Acute Immobilisation Test, Updated Guideline,
October 2000. Guidelines for the testing of chemicals.
OECD. 2000b. Guidelines for the testing of chemicals, No. 23: Guidance Document on Aquatic
Toxicity Testing of Difficult Substances and Mixtures. PDF Edition, p 53, ISBN:
9789264078406.
OECD. 2004a. Guideline for Testing of Chemicals - Daphnia sp., Acute Immobilization Test
202.
OECD. 2004b. Chironomid toxicity test using spiked water; Guideline for Testing of Chemicals
No. 219. In: Paris: Organisation for Economic Cooperation and Development.
OECD. 2004c. Chironomid toxicity test using spiked sediment; Guideline for Testing of
chemicals No. 218. In: Paris: Organisation for Economic Cooperation and Development.
OECD. 2006. Lists of PFOS, PFAS, PFOA, PFCA, related compounds and chemicals that may
degrade to PFCA. Report ENV/JM/MONO (2006). Organisation for Economic Co-operation and
Development; 2007. p. 15.
OECD. 2012. Test no. 211: Daphnia magna reproduction test. OECD Guidelines for the Testing
Chemicals, Section 2: Effect on Biotic Systems. OECD Publishing, Paris.
152
-------
OECD. 2019. Guidance Document on Aqueous-phase Aquatic Toxicity Testing of Difficult Test
Chemicals. Report ENV/JM/MONO(2000)6/REV1. Organisation for Economic Co-operation
and Development. Series on Testing and Assessment. No 23 (Second Edition).
OECD. 2021. Reconciling Terminology of the Universe of Per- and Polyfluoroalkyl Substances:
Recommendations and Practical Guidance, OECD Series on Risk Management, No. 61, OECD
Publishing, Paris
Oliaei, F., D. Kriens, and R. Weber. 2013. PFOS and PFC releases and associated pollution from
a PFC production plant in Minnesota (USA). Environ. Sci. Pollut. Res. 20: 1977-1992.
Orner, G.A., C. Mathews, J.D. Hendricks, H.M. Carpenter, G.S. Bailey, andD.E. Williams.
1995. Dehydroepiandrosterone is a complete hepatocarcinogen and potent tumor promoter in the
absence of peroxisome proliferation in rainbow trout. Carcinogenesis. 16: 2893-2898.
Padilla, S., D. Corum, B. Padros, D.L. Hunter, A. Beam, K.A. Houck, N. Sipcs. Y Kleinstreuer,
T. Knudsen, D.J. Nix, and D.M. Reif. 2012. Zebrafish developmental screening of the
ToxCast™ Phase I chemical library. Reprod. Toxicol. 33: 174-187.
Palumbo, A.J., P.L. TenBrook, T.L. Fojul. T R Faria, and R.S. Tjeerdema. 2012. Aquatic life
water criteria derived via the UC Davis Method I. Organophosphate Insecticides. In. Tjeerdema,
R.S., ed., Springer, NY, NY. Rev. Environ Conlam. Toxicol. 216. Pp. 1-50.
Pan, Y., H. Zhang, Q. Cui, N. Sheng, L. W. Y. Yeung, Y. Sun, Y. Guo, and J. Dai. 2018.
Worldwide Distribution of Novel Perfluoroether Carboxylic and Sulfonic Acids in Surface
Water. Environ. Sci Technol 52(14)' 7621-7629.
Pecquet,A.M., A. Maier. S kasper. S. Sumanas, and J. Yadav. 2020. Exposure to
Perfluorooctanoic Acid (PI OA) Decreases Nenli ophil Migration Response to Injury in Zebrafish
Embryos Ii\1C Res \olesl3( I) (•> p
Penland. T \ . W.G Cope. T .1 kuak. M J. Strynar, C.A. Grieshaber, R.J. Heise, and F.W.
Sessions 2<)20. Trophodynamics of Per- and Polyfluoroalkyl Substances in the Food Web of a
Large Atlantic Slope River 1'iiviron. Sci. Technol. 54(11): 6800-6811.
Plumlee, M.H., .1 I .aralx-e. and M. Reinhard. 2008. Perfluorochemicals in water reuse.
Chemosphere. 72( 1 <)). I 541 -1547.
Popovic, M., R. Zaja, K. Fent, and T. Smital. 2014. Interaction of environmental contaminants
with zebrafish organic anion transporting polypeptide, Oatpldl (Slcoldl). Toxicol. Appl.
Pharmacol. 280(1): 149-158.
Prevedouros, K., I.T. Cousins, R.C. Buck, and S.H. Korzeniowski. 2006. Sources, fate and
transport of perfluorocarboxylates. Environ. Sci. Technol. 40(1): 32-44.
153
-------
Prosser, R.S., K. Mahon, P.K. Sibley, D. Poirier. and T. Watson-Leung. 2016. Bioaccumulation
of perfluorinated carboxylates and sulfonates and polychlorinated biphenyls in laboratory-
cultured Hexagenia spp., Lumbriculus variegatus and Pimephalespromelas from field-collected
sediments. Sci. Tot. Environ. 543: 715-726.
Raimondo, S. and M.G. Barron. 2020. Application of Interspecies Correlation Estimation (ICE)
models in estimating species sensitivity to pesticides. SAR and QSAR in Environmental
Research. 31: 1-18.
Raimondo, S., C.R. Jackson, and M.G. Barron. 2010. Influence of laxonomic relatedness and
chemical mode of action in acute interspecies estimation models for aquatic species. Environ.
Sci. Technol. 44: 7711-7716.
Raimondo, S., C.R. Jackson, and M.G. Barron. 2015. Web-based Interspecies Correlation
Estimation (Web-ICE) for Acute Toxicity: User Manual. Version 3.3, EPA/61 )<) R-l 5/192, U. S.
Environmental Protection Agency, Office of Research and Development, Gulf Lcology Division.
Gulf Breeze, FL.
Raimondo, S., C.R. Lilavois, andM. Elias. In prep. Application of Interspecies Correlation
Estimation models for derivation of water quality standards of PI AS chemicals.
Rainieri, S., N. Conlledo, T. Langerholc, I- Madon an. M Sala. and A Barranco. 2017. Toxic
Effects of Perfluorinated Compounds at! Iiinian Cellular l.evel and on a Model Vertebrate. Food
Chem. Toxicol. 104: 14-25.
Remucal, C. K. 2<> 11-> Spatial and temporal variability of pertluoroalkyl substances in the
Laurentian Great I .tikes Liniron Sci Process. Impacts. 21(11): 1816-1834.
Renner, R. 2<~>n<) LPA finds record PI-OS. PI OA levels in Alabama grazing fields. Environ. Sci.
Technol. 43(3) 1245 l24(->
Rewerts, J.N., EC. Christie, A.E. Robel, T.A. Anderson, C. McCarthy, C.J. Sal ice, and J. A.
Field. 2021. Key Considerations for Accurate Exposures in Ecotoxicological Assessments of
Perfluorinated Carboxylates and Sulfonates. Environ. Toxicol. Chem. 40: 677-688.
Road, P., Y. Wei, J. Dai, M. Liu, J. Wang, M. Xu, J. Zha, and Z. Wang. 2007. Estrogen-like
properties of perfluorooctanoic acid as revealed by expressing hepatic estrogen-responsive genes
in rare minnows (Gobiocypris rarus). Environ. Toxicol. Chem. 26: 2440-2447.
Rodea-Palomares, F.F., C. Gonzales-Garcia, and F. Legares. 2009a. Use of Lux-marked
cyanobacterial bioreporters for assessment of individual and combined toxicities of metals in
aqueous samples in: Handbook on Cyanobacteria: Biochemistry, Biotechnology, and
Applications. Nova Science Publishers Inc. New York, USA.
154
-------
Rodea-Palomares, I, C. Gonzales Garcia, F. Legares, and F. Fernandez Pinas. 2009b. Effect of
pH, EDTA, and anions on heavy metal toxicity toward a bioluminescent cyanobacterial reporter.
Arch. Environ. Contain. Toxicol. 57: 477-487.
Rodea-Palomares, I., F. Leganesa, R. Rosal, andF. Fernandez-Pinas. 2012. Toxicological
interactions of perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) with
selected pollutants. J. Hazard. Mater. 201-202: 209-218.
Rodea-Palomares, I., M. Makowski, S. Gonzalo, M. Gonzalez-Pleiter, F. Leganes, and F.
Fernandez-Pinas. 2015. Effect of PFOA/PFOS pre-exposure on the toxicity of the herbicides 2,4-
D, atrazine, diuron and paraquat to a model aquatic photosynlhelic microorganism.
Chemosphere. 139: 65-72.
Rosal, R., I. Rodea-Palomares, K. Boltes, F. Fernandez-Pinas, 1'. Leganes. and A. Petre. 2010.
Ecotoxicological assessment of surfactants in the aquatic environment: combined toxicity of
docusate sodium with chlorinated pollutants. Chemosphere. 81: 288-293.
Rotondo, J.C., L. Giari, C. Guerranti, M. Tognon, G. Caslaldelli, E.A. Fano, and I' Martini.
2018. Environmental doses of perfluorooctanoic acid change the expression of genes in target
tissues of common carp. Environ. Toxicol Cheni 37(3): 942-lMS
Royer, L.A. 2011. An investigation of the biodegradation potential of S 2 fluorotelomer esters in
environmentally relevant systems. PhD Thesis, Purdue I ni\ ersity, Purdue, IN.
Russell, M.H., \V R lierli. 1} S/.oslek. and R.C. Buck. Zoos Investigation of the biodegradation
potential of a fhioroacrx late polymer product in aerobic soils. Environ. Sci. Technol. 42: 800-
807.
Russell, M.H . W R lierli. li S/.oslek. and R C. Buck. 2010. Evaluation of PFO formation from
the biodegradation of a lluorolelomer-based urethane polymer product in aerobic soils. Polym.
Degrad Stab. 95: 7l)-S5.
Saez M.. I) Y. Moreno, J. Begona., and S. Van Leeuwen. 2008. Uncommon PFC-profile in arctic
ice samples lYom Russia. Organohalogen Compd. 70:1870-1873.
Saito, N., K. Sasaki. K. Nakalome, K. Harada, T. Yoshinaga, and A. Koizumi. 2003.
Perfluorooctane sulfonate concentrations in surface water in Japan. Arch. Environ. Contam.
Toxicol. 45(2): 149-158.
San Francisco Bay Regional Water Quality Control Board (RWQCB). TRANSMITTAL
MEMORANDUM: Transmittal of Interim Final Environmental Screening Levels (ESLs) for
Two Per- and Polyfluoroalkyl Substances (PFAS): Perfluorooctane Sulfonate (PFOS) and
Perfluorooctanoate (PFOA); May 27, 2020; Alec Naugle, Chief, Toxics Cleanup Division.
155
-------
Sanderson, H., T.M. Boudreau, S.A. Mabury, and K.R. Solomon. 2003. Impact of
perfluorooctanoic acid on the structure of the zooplankton community in indoor microcosms.
Aquat. Toxicol. 62: 227-234.
Scott B.F., C.A. Moody, C. Spencer, J.M. Small, D.G. Muir, and S.A. Mabury. 2006. Analysis
for perfl uorocarboxylic acids/anions in surface waters and analysis of PFOA from large volume
samples. Environ. Sci. Tech. 40:6405-6410.
Scott, B.F., C. Spencer, E. Lopez, and D. C. Muir. 2009. Perfluorinated alkyl acid concentrations
in Canadian rivers and creeks. Water Qual. Res. Jour.. 44(3): 263-277.
Scott, B.F., A.O. De Silva, C. Spencer, E. Lopez, S.M. Backus, and I) (' (i Muir. 2010.
Perfluoroalkyl acids in Lake Superior water: Trends and sources. Journal of Great Lakes
Research. 36(2): 277-284.
Sedlak, M.D., J.P. Benskin, A. Wong, R. Grace, and I) .1 Greig. 2017. Per- and polyfluoroalkyl
substances (PFASs) in San Francisco Bay wildlife: Temporal trends, exposure pathways, and
notable presence of precursor compounds. Chemospheie 185 1217-1226.
Seyoum, A., A. Pradhan, J. Jass, andP.E. Olsson. 2020. IVilluoiinated Alkyl Substances Impede
Growth, Reproduction, Lipid Metabolism and I .ill-span in Daphnia magna. Sci. Tot.
Environ.737: 12 p.
Shoeib, M., T. Harner, and P. Vlahos. 20<>5 Ratio of perfluorochemical concentrations as a tracer of
atmospheric deposition to surface waters I-miron. Sci. Technol. 39(22): 8678-8683.
Sinclair, E. and K kannan 2<)i)(-. Mass loading and fate of perfluoroalkyl surfactants in
wastewater treatment plants I jniion Sci Technol. 40(5): 1408-1414.
Spachmo. li and Arukwe, A 2012. Endocrine and developmental effects in Atlantic salmon
(Salmo salur) exposed to periluorooctane sulfonic or perfluorooctane carboxylic acids. Aquat.
Toxicol. 108 I 12-124
SRC (Syracuse Research Corporation). 2016. PHYSPROP Database. Accessed May 2016.
http://www.srcinc.coin/what-we-do/environmental/scientific-databases.html.
Stahl, L.L., B.D. Snyder, A.R. Olsen, T.M. Kincaid, J.B. Wathen, andH.B. McCarty. 2014.
Perfluorinated compounds in fish from U.S. urban rivers and the Great Lakes. Sci. Tot. Environ.
499:185-195.
Stefani, F., M. Rusconi, S. Valsecchi, and L. Marziali. 2014. Evolutionary ecotoxicology of
perfluoralkyl substances (PFASs)inferred from multigenerational exposure: A case study with
Chironomus riparius (Diptera, Chironomidae). Aquat. Toxicol. 156: 41-51.
156
-------
Stengel, D., S. Wahby, and T. Braunbeck. 2017. In search of a comprehensible set of endpoints
for the routine monitoring of neurotoxicity in vertebrates: Sensory perception and nerve
transmission in zebrafish (Danio rerio) embryos. Environ. Sci. Pollut. Res. Int. 12: 19 p.
Stevenson, C.N., L.A. MacManus-Spencer, T. Luckenbach, R.G. Luthy, and D. Epel. 2006. New
perspectives on pefluorochemical ecotoxicology: inhibition and induction of an efflux transporter
in marine mussel, Mytilus californianus. Environ. Sci. Technol. 40: 5580-5585.
Stinckens, E., L. Vergauwen, G.T. Ankley, R. Blust, V.M. Darras. 1)1. Yilleneuve, H. Witters,
D.C. Volz, and D. Knapen. 2018. An AOP-Based Alternative Testing Strategy to Predict the
Impact of Thyroid Hormone Disruption on Swim Bladder Inflation in Zebrafish. Aquat.
Toxicol.200:1-12.
StockNL, V.I. Furdui, D.C.G. Muir, S. A. Mabury. 2007. Perfluoroalkyl contaminants in the
Canadian Arctic: Evidence of atmospheric transport and local contamination Ln\ iron. Sci.
Technol. 41:3529-3536.
STS Consultants, Ltd. 2007. Surface water quality criterion for perfluorooctanoic acid. Prepared
for Minnesota Pollution Control Agency Si Paul, Minnesota
https://www.pca.state.mn.us/sites/defaull llles pl'oa-report.pdl'
Stuchal, L., and S. Roberts. 2019. PFAS- IYo\ isional Cleanup Target Levels and Screening
Levels. University of Florida, Center for Environ menial and I liinian Toxicology, Contaminated
Media Forum. September. 2<>ll)
Tang, J., X. Jia, \ (.ao. Y \\ u. / l.iu. X Lu. Q Du, J. He, N. Li, B. Chen, J. Jiang, W. Liu, Y.
Ding, W. Zhu, and II /hang. 2<)|S Role of the VI2-ARE pathway in perfluorooctanoic acid
(PFOA)-induced hepatotoxicity in Rana iii^romaciikua. Environ. Pollut. 238: 1035-1043.
TCEQ (Texas Commission on Ln\ironmcntal Quality). 2021. 2021 Ecological Screening
Benchmarks. a\ailaMe at Imps www tceq.texas.gov/remediation/eco (2021 Benchmarks),
accessed HI 13.22.
Thienpont, J J., A Tingaud-Sequeira, E. Prats, C. Barata, P.J. Babin, andD. Raldua. 2011.
Zebrafish eleutheroemhryos provide a suitable vertebrate model for screening chemicals that
impair thyroid hormone synthesis. Environ. Sci. Technol. 45(17): 7525-7532.
Thompson, J., A. Roach, G. Eaglesham, M. E. Bartkow, K. Edge, and J. F. Mueller. 2011.
Perfluorinated alkyl acids in water, sediment and wildlife from Sydney Harbour and
surroundings. Mar. Poll. Bull.(62): 2869-2875.
Tilton, S.C, G.A. Orner, A.D. Benninghoff, H.M. Carpenter, J.D. Hendricks, C.B. Pereira, and
D.E. Williams. 2008. Genomic profiling reveals an alternate mechanism for hepatic tumor
promotion by perfluorooctanoic acid in rainbow trout. Environ. Health Perspect. 116(8): 1047-
1055.
157
-------
Tomy, G.T., S.A. Tittlemier, V.P. Palace, W.R. Budakowski, E. Braekevelt, L. Brinkworth, and
K. Friesen. 2004. Biotransformation of TV-Ethyl Perfluorooctanesulfonamide by Rainbow Trout
(Onchorhynchus mykiss) Liver Microsomes. Environ. Sci. Technol. 38: 758-762.
Tomy, G. T., K. Pleskach, S. H. Ferguson, J. Hare, G. Stern, G. Macinnis, C. H. Marvin, and L.
Losefo. 2009. Trophodynamics of Some PFCs and BFRs in a Western Canadian Arctic Marine
Food Web. Environ. Sci. Technol. 43: 4076-4081.
Tornabene, B.J., M.F. Chislock, M.E. Gannon, M.S. Sepulveda, and .1 T. llo\erman. 2021.
Relative acute toxicity of three per- and polyfluoroalkyl sub si ances on nine species of larval
amphibians. Integr. Environ. Assess. Manag. 17(4): 684-689.
Truong, L., D.M. Reif, L. St Mary, M.C. Geier, H.D. Truong, and R.L. Tanguay. 2014.
Multidimensional In Vivo Hazard Assessment Using Zebrafish. Toxicol. Sci 137( 1): 212-233.
Ulhaq, M., S. Orn, G. Carlsson, J. Tallkvist, and L. Norrgren 2012. Perfluorooctanoic acid
toxicity in zebrafish (Danio rerio). Abstracts Toxicol, Letters 21 1S: S43-S216.
Ulhaq, M., G. Carlsson, S. Orn, and L. Xorruren 2013. Comparison of developmental toxicity of
seven perfluroalkyl acids to zebrafish embryos I-nviron. Toxicol Pharmacol. 36: 423-426.
UNEP (United Nations Environmental Program). 201 5 Proposal to list pentadecafluorooctanoic
acid (CAS No: 335-67-1, PFOA, perfluorooctanoic acid), its salts and PFOA-related compounds
in Annexes A, B and/or C to the Stockholm Convention on Persistent Organic Pollutants.
U.S. EPA (United States Environmental Protection Agency). 1982. Standard Evaluation
Procedure, Dciphnici magna Life-cycle (21 -day renewal) Chronic Toxicity Test EPA 540/9-86-
141. Environmental Protection Agency, office of Pesticidy Programs. Washington D.C.
U.S. EPA (United States I jivironmenial Protection Agency). 1985. Guidelines for deriving
numerical national water quality criteria for the protection of aquatic organisms and their uses.
National Technical Information Service No. PB85-227049.
U.S. EPA (United States En\ ironmental Protection Agency). 1991. Technical support document
for water quality-based toxics control. U.S. EPA, Office of Water, Washington, DC. EPA/505/2-
90-001, PB91-12741 5.
U.S. EPA (United States Environmental Protection Agency). 1995a. Final water quality
guidance for the Great Lakes system. 60 Federal Register 15366-15425 (March 23, 1995)
(40 CFR Parts 9, 122, 123, 131, and 132).
U.S. EPA (United States Environmental Protection Agency). 1995b. Short-term Methods
for Estimating the Chronic Toxicity of Effluents and Receiving Waters to West Coast
Marine and Estuarine Organisms. Environmental Monitoring and Support Laboratory,
Cincinnati, OH. EPA/600/R-95/136.
158
-------
U.S. EPA (United States Environmental Protection Agency). 1996a. OPPTS 850.1075 Fish
Acute Toxicity Test, Freshwater and Marine (EPA 712-C-96-118). Office of Prevention,
Pesticides, and Toxic Substances. Washington, D.C.
U.S. EPA (United States Environmental Protection Agency). 1996b. Ecological effects test
guidelines. OPPTS 850.4200: seed germination/root elongation toxicity test. EPA 712-C-96-154.
Washington, DC
U.S. EPA (United States Environmental Protection Agency). 19lM (midelines for ecological
risk assessment. EPA/630/R-95/002F. Risk Assessment Forum Office of Research and
Development, Washington, D.C.
U.S. EPA (United States Environmental Protection Agency). 2006. PFOA Stewardship
Program (EPA-HQ-OPPT-2006-0621). Office of Prevention, Pesticides, and Toxic Substances.
Washington, D.C.
U.S. EPA. (U.S. Environmental Protection Agency). 2012 Ixological Effects Test Guidelines.
OCSPP 850.4500: Algal Toxicity. Office of Chemical Safety and Pollution Prevention.
Washington, DC. EPA 712-C-006-18-002.
U.S. EPA. (U.S. Environmental Protection Agency) 2<~)13 Aquatic l.ile Ambient Water Quality
Criteria for Ammonia - Freshwater. Office of Water. Office of Science and Technology.
Washington, DC. EPA 822-R-l8-002.
U.S. EPA (United States Environmental Protection Agency). 2014. Water quality standards
handbook. EPA-820-B-14-008. Office of Water, Washington, DC. Available online at:
https://www.epa.gov/wqs-tech/water-quality-standards-handbook.
U.S. EPA (I nited Slates r.nvironmental Protection Agency). 2016a. Recommended
Aquatic I.ife Ambient Water Quality Criterion for Selenium in Freshwater. 81 Federal
Register 45285-45287 (July 13, 2<)|(->)
U.S. EPA (I nited States En\ i ronmental Protection Agency. 2016b. OCSPP 850.1000
Background and Special Considerations-Tests with Aquatic and Sediment-Dwelling Fauna and
Aquatic Microcosms (T.PA-I IQ-OPPT-2009-0154-0042). Office of Chemical Safety and
Pollution Prevention Washington, D.C.
U.S. EPA (U.S. Environmental Protection Agency). 2016c. Series 850 - Ecological Effects Test
Guidelines. Office of Chemical Safety and Pollution Prevention, Washington, DC. Accessed
March 2021. https://www.epa.gov/test-guidelines-pesticides-and-toxic-substances/series-850-
ecological-effects-test-guidelines.
U.S. EPA (United States Environmental Protection Agency). 2018b. Final Aquatic Life
Ambient Water Quality Criteria for Aluminum - 2018 (EPA-822-R-18-001). Office of Water.
Washington, D.C.
159
-------
U.S. EPA (United States Environmental Protection Agency). 2021. EPA Internal Report,
Preliminary Results from a USEPA ORD Study: Cumulative Developmental Toxicity of
Combined Exposure to Ammonium Perfluorooctanoic Acid (PFOA, CASRN: 3825-26-1) and
Potassium Perfluorooctane Sulfonate (PFOS, CASRN: 2795-39-3) in the Sprague-Dawley Rat.
Internal Report. Office of Research and Development, Center for Public Health and
Environmental Assessment, Public Health and Integrated Toxicology Division, Reproductive
and Developmental Toxicology Branch. August 26, 2021.
U.S. FWS. (U.S. Fish and Wildlife Service). 2018. Zebra Danio (Ikimo rcrio) Ecological Risk
Screening Summary.
U.S. FWS (U.S. Fish and Wildlife Service). 2020. Japanese Medaka ((hyzias latipes) Ecoloical
Risk Screening Summary.
Vedagiri, U.K., R.H. Anderson, H.M. Loso, and CM Schwach. 2018. Ambient le\ els of PFOS
and PFOA in multiple environmental media. Rem. J 2S(2)' 9-5 |.
Vogs, C., G. Johanson, M. Naslund, S. Wulff, M. Sjodin, \l I lellstrandh, J. Lindberg, and E.
Wincent. 2019. Toxicokinetics of Perfluorinaled Alkyl Acids Influences Their Toxic Potency in
the Zebrafish Embryo (Danio rerio). En\ iron Sci Technol. 53(7) 3898-3907.
Vogt, C., C. Nowak, J.B. Diogo, M. Oetken, K. Schuenk. and .1 Oehlmann. 2007b. Multi-
generation studies with Chironomus riparius - Effects of low trihutyltin concentrations on life
history parameters and genetic di\ ersity. Chemosphere 67( I I2192-2200.
Vogt, C., A. Pupp, C. Nowak. I,S Jagodzinski, J. Baumann, D. Jost, M. Oetken, and J.
Oehlmann. 2007c. Interaction |->et\\een genetic diversity and temperature stress on life-cycle
parameters and genetic \ ai iaMlitv in midue ('huanomus riparius populations. Clim. Res. 33:
207-214
Vogt, (' VI . HeP, C. ISowak. .1 1} Diouo. J. Oehlmann, andM. Oetken. 2010. Effects of
cadmium of life-cycle parameters in a multi-generation study with Chironomus riparius
following a pi e-exposure of populations to two idfferent tributyltin concentrations for several
generations. l-coto\icol 1^(7) 1174-1182.
Wang, N., B. Szostek. R (' Buck, P.W. Folsom, L.M. Sulecki, V. Capka, W.R. Berti, and J.T.
Gannon. 2005. Fluorotelomer alcohol biodegradation-direct evidence that perfluorinated carbon
chains breakdown. Environ. Sci. Technol. 39: 7516-7518.
Wang, N., B. Szostek, R.C. Buck, P.W. Folsom, L.M. Sulecki, and J.T. Gannon. 2009. 8-2
fluorotelomer alcohol aerobic soil biodegradati on: pathways, metabolites, and metabolite yields.
Chemosphere. 75: 1089-1096.
160
-------
Wang, N., R.C. Buck, B. Szostek, L.M. Sulecki and B.W. Wolstenholme. 2012. 5:3
Polyfluorinated acid aerobic biotransformation in activated sludge via novel "one-carbon
removal pathways". Chemosphere. 87: 527-534.
Wang, W.J., C. DeWitt, C.P. Higgins, and I.T. Cousins. 2017. A never-ending story of per- and
polyfluoroalkyl substances (PFASs)?. Environ. Sci. Technol. 5(51): 2508-2518.
Wang, X., B. Fan, M. Fan, S. Belanger, J. Li, J. Chen, X. Gao, and Z. Liu. 2020. Development
and use of interspecies correlation estimation models in China for potential application in water
quality criteria. Chemosphere. 240: 124848
Warne, M.St.J., G.E. Batley, R.A. van Dam, J.C. Chapman, 1) R I ox. (' \V Hickey, and J.L.
Stauber. 2018. Revised method for deriving Australian and New Zealand water quality guideline
values for toxicants - update of 2015 version. Prepared for the revision of I lie Australian and
New Zealand Guidelines for Fresh and Marine Water Quality. Australian and New Zealand
Governments and Australian state and territory governments, Canberra, 48 pp.
Wasel, O., K.M. Thompson, Y. Gao, A.E. Godfrey, J. Gao, C.T. Mahapatra, L.S. Lee, M.S.
Sepulveda, and J.L. Freeman. 2020. Comparison of Zebrafish in Vitro and in Vivo
Developmental Toxicity Assessments of Perfluoroalkyl Acids (PI'AAs). J. Toxicol. Environ.
Health Part A84:125-136.
Washington, J. W., J.J. Ellington, T.M. Jenkins, J.J. Evans, H. Yoo, S.C. Hafner. 2009.
Degradability of an Acrylate-Linked, Fluorotelomer Polymer in Soil. Environ. Sci. Technol.
43(17): 6617-6623
Wei, Y., J. Dai, M I.in. J Wanu. M \u. .1 /ha, and Z. Wang. 2007. Estrogen-like properties of
perfluorooctanoic acid as re\ ea I eel In expressing hepatic estrogen-responsive genes in rare
minnows (Gobiocypris ranis) l-n\imn Toxicol Chem. 26(11): 2440-2447.
Wei, Y . Y. Liu, J. \Yanu. Y Tao. and .1 Dai 2008a. Toxicogenomic analysis of the hepatic
effects of perfluorooclanoic acid on rare minnows (Gobiocypris rarus). Toxicol. Appl.
Pharmacol 226: 285-297.
Wei, Y., I. I. Chan. D. Wanu. IL Zhang, J. Wang, and J. Dai. 2008b. Proteomic analysis of
hepatic protein profiles in rare minnow (Gobiocypris rarus) exposed to perfluorooctanoic acid. J.
Proteome Res. 7. 172lM739.
Wei, Y., X. Shi, H. Zhang, J. Wang, B. Zhou, and J. Dai. 2009. Combined Effects of
Polyfluorinated and Perfluorinated Compounds on Primary Cultured Hepatocytes from Rare
Minnow (Gobiocypris rarus) Using Toxicogenomic Analysis. Aquat. Toxicol. 95: 27-36
(Supplemental Journal Materials).
Wen, W., X. Xia, X. Chen, H. Wang, B. Zhu, H. Li and Y. Li. 2016. Bioconcentration of
perfluoroalkyl substances by Chironomusplumosus larvae in water with different types of
dissolved organic matters. Environ. Pollut. 213: 299-307.
161
-------
Williams, T.D., A. Diab, F. Ortega, V.S. Sabine, R.E. Godfrey, F. Falciani, J.K. Chipman, and
S.G. George. 2008. Transcriptomic Responses of European Flounder (Platichthys flesus) to
Model Toxicants. Aquat. Toxicol.90(2): 83-91.
Willming, M. M., C. R. Lilavois, M.G. Barron, and S. Raimondo. 2016. Acute toxicity
prediction to threatened and endangered species using Interspecies Correlation Estimation (ICE)
models. Environ. Sci. Technol. 50: 10700-10707.
Wu, J., Z. Liu, Z. Yan, X. Yi. 2015. Derivation of water quality criteria of phenanthrene using
interspecies correlation estimation models for aquatic life in China 1 ji\ iron. Sci. Pollut. Res. 22:
9457-9463.
Wu, J., Z. Yan, X. Yi, Y. Lin, J. Ni, X. Goa., Z. Liu, and X. Shi. 2016. Comparison of species
sensitivity distributions constructed with predicted acute toxicity data from inlcrspecies
correlation estimation models and measured acute data for Benzo[a]pyrene. Chemosphere. 144:
2183-2188.
Xia, X., X. Chen, X. Zhao, H. Chen, and M. Shen. 2012 I-fleets of carbon nanotubes, chars, and
ash on bioaccumulation of perfluorochemicals In ('hironomns plumosus larvae in sediment.
Environ. Sci. Technol. 46: 12467-12475.
Xia, X., A.H. Rabearisoa, X. Jiang, and Z Dai 2<> 13 liioaccunuilation of perfluoroalkyl
substances by Daphnia magna in water with different types and concentrations of protein.
Environ. Sci. Technol. 47: 10955-10963.
Xia, X., Z. Dai, A.H. Rabearisoa. P Zhao, and X. Jiang. 2015a. Comparing humic substance and
protein compound effects on the Moaccunuilation of perfluoroalkyl substances by Daphnia
magna in water Chemosphcre I I ^ l)7X-l)X(-<
Xia, X . ATI Ralxicrisoa. / Dai. \ Jiang, P. Zhao, and H. Wang. 2015b. Inhibition effect of
Na+ and Ca2+ on the bioaeeiinuilation of perfluoroalkyl substances by Daphnia magna in the
presence of protein. Environ Toxicol. Chem. 34(2): 429-436.
Xiao, F. 2017. 1 -merging poly- and perfluoroalkyl substances in the aquatic environment: A
review of current literature Water Res. 124:482-495.
Xiao, F., M.F. Simcik, T.R. Halbach, and J.S. Gulliver. 2015. Perfluorooctane sulfonate (PFOS)
and perfluorooctanoate (PFOA) in soils and groundwater of a U.S. metropolitan area: Migration
and implications for human exposure. Water Res. 72: 64-74.
Xu, D., C. Li, H. Chen, and B. Shao. 2013. Cellular response of freshwater green algae to
perfluorooctanoic acid toxicity. Ecotoxicol. Environ. Saf. 88: 103-107.
Xu, J., C.S. Guo, Y. Zhang, and W. Meng. 2014. Bioaccumulation and trophic transfer of
perfluorinated compounds in a eutrophic freshwater food web. Environ. Pollut. 184: 254-261.
162
-------
Yamashita, N., S. Taniyasu, G. Petrick, S. Wei, T. Gamo, P.K. Lam, and K. Kannan. 2008.
Perfluorinated acids as novel chemical tracers of global circulation of ocean waters.
Chemosphere. 70(7): 1247-1255.
Yang, H. B., Z. Ya-Zhou, Y. Tang, G. Hui-Qin, F. Guo, S. Wei-Hua, L. Shu-Shen, H. Tan, and
F. Chen. 2019. Antioxidant Defence System is Responsible for the Toxicological Interactions of
Mixtures: A Case Study on PFOS and PFOA in Daphnia magna. Sci. Total Environ. 667: 435-
443 (Supplemental Journal Materials).
Yang, J. 2010. Perfluorooctanoic acid induces peroxisomal fatty acid oxidation and cytokine
expression in the liver of male Japanese medaka (Oryzias latipes). Chemosphere. 81: 548-552.
Yang, S., F. Xu, F. Wu, S. Wang, andB. Zheng. 2014. Development of lJl OS and PFOA criteria
for the protection of freshwater aquatic life in China. Sci. Total Environ. 47<)-471 677-683.
Ye, X., M. J. Strynar, S. F. Nakayama, J. Varns, L. Helfant. J I.azorchak, and A I) Lindstrom.
2008. Perfluorinated compounds in whole fish homogenales from the Ohio, Missouri, and Upper
Mississippi Rivers, USA. Environ. Pollut. 156: 1227-1232.
Young, C.J., V.I. Furdui, J. Franklin, R.M koeiner. D.C.G. Muir. and S.A. Mabury. 2007.
Perfluorinated acids in arctic snow: new e\ ideuce for atmospheric formation. Environ. Sci.
Technol. 41(10):3455-3461.
Yuan, Z., J. Zhang, Y. Zhang, 11 /hen, and Y. Sun. 201 5 The effect of perfluorooctanoic acid
on the planarian Dugesia japonica Pol J Environ. Stud. 24(2): 801-807.
Yuan, Z., J. Zhang, C Tn. / \Yanu. and W. Xin. 2016a. The protective effect of blueberry
anthocyanins against peilluoiooclanoic acid-induced disturbance in planarian (Dugesia
japonica) Ixoloxicol Ijniion Sal" 127 170-174.
Yuan./.. .1 /hang, 1} Zhao. / Miao. and X. Wu. 2016b. Effects of perfluorooctanoic acid on
neural genes expression and neuronal morphology in the planarian Dugesia japonica. Chem.
Ecol. 32(0) 575-582.
Yuan, Z., Z. Miao. \ Gonu. 1} Zhao, Y. Zhang, H. Ma, J. Zhang, and B. Zhao. 2017. Changes
on lipid peroxidation, enzymatic activities and gene expression in planarian (Dugesia japonica)
following exposure to periluorooctanoic acid. Ecotoxicol. Environ. Saf. 145: 564-568.
Zareitalabad, P., J. Siemens, M. Hamer and W. Amelung. 2013. Perfluorooctanoic acid (PFOA)
and perfluorooctanesulfonic acid (PFOS) in surface waters, sediments, soils and wastewater - A
review on concentrations and distribution coefficients. Chemosphere. 91(6): 725-732.
Zhang, H., W. Fang, D. Wang, N. Gao, Y. Ding, and C. Chen. 2014a. The role of interleukin
family in perfluorooctanoic acid (PFOA)-induced immunotoxicity. J. Hazard. Materials 280:
552-560.
163
-------
Zhang, H., J. He, N. Li, N. Gao, Q. Du, B. Chen, F. Chen, X. Shan, Y. Ding, W. Zhu, Y. Wu, J.
Tang, and X. Jia. 2019b. Lipid accumulation responses in the liver of Rana nigromaculata
induced by perflurooctanoic acid (PFOA). Ecotoxicol. Environ. Saf. 167: 29-35.
Zhang, H., L. Shen, W. Fang, X. Zhang, and Y. Zhong. 2021. Perfluorooctanoic Acid-Induced
Immunotoxicity Via NF-Kappa B Pathway in Zebrafish (Danio rerio) Kidney. Fish Shellfish
Immunol. 113:9-19.
Zhang, J., B. Wang, B. Zhao, Y. Li, X. Zhao, and Z. Yuan. 2019a. Blueberry anthocyanin
alleviate perfluorooctanoic acid-induced toxicity in planarian (Dh^csici japonica) by regulating
oxidative stress biomarkers, ATP contents, DNA methylation and niRNA expression. Environ.
Pollut. 245: 957-964.
Zhang, L., J. Niu, Y. Li, Y. Wang andD. Sun. 2013a. Evaluating the sub-lclhal toxicity ofPFOS
and PFOA using rotifer Brachionus calyciflorus. F.nviion. Pollut. 180: 34-40
Zhang, L., J. Niu, Y. Wang, J. Shi and Q. Huang. 2014b Chronic effects of PFOA and PFOS on
sexual reproduction of freshwater rotifer Brachionus calyci/lorns. Chemosphere. 114: 114-120.
Zhang, S., B. Szostek, P.K. McCausland. li \V Wolstenholme. \ l.n. \ Wang, andR.C. Buck.
2013b. 6:2 and 8:2 Fluorotelomer alcohol anaerobic biotransformation in digester sludge from a
WWTP under methanogenic conditions. Environ. Sci. Technol 47: 4227-4235.
Zhang, S. L. Wang, Z. Want, 1) I'an, L. Shi, and J. Lui. 2<> I 7 Derivation of freshwater water
quality criteria for dibutyltin dilain ate from measured data and data predicted using interspecies
correlation estimation models. Chemosphere. 171: 142-148.
Zhang, X . R T.ohmann, C. Dassuncao, X. C. Hu, A. K. Weber, C. D. Vecitis and E. M.
Sunderland 2<) I (¦> Source attribution of poly- and perfluoroalkyl substances (PFASs) in surface
waters from Rhode Island and the New York Metropolitan Area. Environ Sci Technol Lett. 3(9):
316-321
Zheng, X.M . II L Liu, W Shi. S. Wei, J.P. Giesy, and H.X. Yu. 2012. Effects of perfluorinated
compounds on dc\clopmenl of zebrafish embryos. Environ. Sci. Pollut. Res. 19(7): 2498-2505.
Zhou, Y. and Z.S /lianu 1989. The Method of Aquatic Toxicity Test. Agricultural Press,
Beijing.
Zhou, Z., Y. Shi, and W. Li. 2012. Perfluorinated Compounds in Surface Water and Organisms
from Baiyangdian Lake in North China: Source Profiles, Bioaccumulation and Potential Risk.
Bull. Environ. Contam. Toxicol. 89; 519-524 (2012).
164
-------
Appendix A Acceptable Freshwater Acute PFOA Toxicity Studies
A. 1 Summary Table of Acceptable Quantitative Freshwater Acute PFOA Toxicity Studies
Species (lifestage)
Method3
Test
Duration
Chemical /
Purity
i)H
Temp.
(°C)
Effect
Author
Reported
Effect
Cone.
(mg/L)
EPA
Calculated
Effect
Cone.
(mg/L)
Final Effect
Cone.
(mg/L)e
Species
Mean
Acute
Value
(mg/L)
Reference
Planaria (0.9 cm),
Dugesia japonica
S,U
96 hours
PFOA
>98%
-
25
LCso
458
-
458
-
Li 2008
Planaria (0.9 cm),
Dugesia japonica
s,u
96 hours
PFOA
>98%
-
25
LCso
337h
321.8
321.8
-
Li 2009
Planaria (0.9 cm),
Dugesia japonica
s,u
96 hours
PFOA
>98%
-
25
LCso
337h
383.0
383.0
383.6
Li 2009
Fatmucket
(glochidia, <24 hours),
Lampsilis siliquoidea
S, M
24 hours
PFOA
96%
8.46
20
ECso
(viability)
164.4
-
164.4
-
Hazelton et al.
2012, 2013
Fatmucket (juvenile, 4-6 weeks),
Lampsilis siliquoidea
R, M
96 hours
PFOA
96%
8.46
20
LC50
>500
-
>500f
164.4
Hazelton et al.
2012, 2013
Black sandshell
(glochidia, <24 hours),
Ligumia recta
S, M
24 hours
PFOA
96%
8.46
20
ECso
(viability)
161.0
-
161.0
-
Hazelton et al.
2012, Hazelton
2013
Black sandshell
(juvenile, 4-6 weeks),
Ligumia recta
RM
96 hours
PFOA
96%
8.46
20
LCso
>500
-
>500f
161.0
Hazelton et al.
2012, Hazelton
2013
Pewter Physa (mixed age),
Phvsella acuta
(formerly, Phvsa acuta)
S,U
96 hours
PFOA
>98%
-J
25
LCso
672h
762.0
762.0
-
Li 2009
Pewter Physa (mixed age),
Phvsella acuta
S,U
96 hours
PFOA
>98%
-
25
LCso
672h
659.9
659.9
-
Li 2009
Pewter Physa (mixed age),
Phvsella acuta
S,U
96 hours
PFOA
>98%
-
25
LCso
672h
628.3
628.3
681.1
Li 2009
Rotifer (<2-hour old neonates),
Brachionus calvciflorus
s,ub
24 hours
PFOA
96%
-
20
LCso
150.0
-
150.0
150.0
Zhang et al. 2013a
A-l
-------
Species (lilesliiue)
Method-'
Tesl
l)iir;ilion
( hemiciil /
PuriU
pll
Temp.
<°C)
I-'. 11 eel
Author
Kcporlcd
l-!ITeel
(one.
(iiiii/l.)
r.p\
( iileuliiled
EITecl
Cone.
(111^/1.)
l-iiiiil 1-1 ITeel
(one.
Speeies
Mesin
Acute
Value
(inii/l.)
Reference
( ladoa.T;ni ( 24 hum's).
Chydorus sphaericus
S, U
48 hours
mm
Unreported
2U
IX
(death/immobility)
1 lu
93.17°
93.17°
93.17
Le and
Peijnenburg 2013
Cladoceran (6-12 hours),
Daphnia carinata
s,u
48 hours
PFOA
95%
-
21
ECso
(death/immobility)
78.:
66.80
66.80
66.80
Logeshwaran et
al. 2021
Cladoceran (<24 hours),
Daphnia magna
s,u
48 hours
PFOA
>97%
-
21
ECso
(iinrnobility)
223.6
-
223.6
-
Boudreau 2002
Cladoceran
(STRAUS-clone 5; 6-24 hours),
Daphnia magna
s,u
48 hours
APFO
99.7%
-
18-22
IX
480d
-
480d
-
Colombo et al.
2008
Cladoceran (<24 hours),
Daphnia magna
s,u
48 hours
PFOA
Unreported
-
21
EC..
(iinmobilily)
476 52
542.5
542.5
-
Ji et al. 2008
Cladoceran (<24 hours),
Daphnia magna
s,u
48 hours
PFOA
>98%
7.82-
7.91
25
IX
181
220.8
220.8
-
Li 2009
Cladoceran (<24 hours),
Daphnia magna
s,u
48 hours
PFOA
>98%
7.82-
7.91
25
IX"
181h
157.9
157.9
-
Li 2009
Cladoceran (<24 hours),
Daphnia magna
s,u
48 hours
PFOA
>98%
7 82-
7,'H
25
LC50
18 lh
207.3
207.3
-
Li 2009
Cladoceran (<24 hours),
Daphnia magna
S,M
48 hours
PFOA
96%
-
20
ECso
(death/immobility)
211.6°
216.1°
216.1c
-
Ding et al. 2012a
Cladoceran (<24 hours),
Daphnia magna
S,M
48 hours
PFOA
99%
7
::
1 'C-50
201.85
222.0
222.0
-
Yang et al. 2014
Cladoceran (<24 hours),
Daphnia magna
S.\T
4S hours
PFOA
>%"»
7.0-
7.82
20
ECso
(immobility)
239
215.6
215.6
-
Barmentlo et al.
2015
Cladoceran (<24 hours),
Daphnia magna
S. L
48 hours
PF() \
98",,
-
20
EC50
(death/immobility)
110.7
114.6
114.6
-
Lu et al. 2016
Cladoceran (12-24 hours),
Daphnia magna
S, U
48 hours
PR) \
Unreported
6-8.5
20
LC50
120.9°
117.2°
117.2C
220.0
Yang et al. 2019
Cladoceran (<24 hours),
Daphnia pulicaria
S,U
48 hmi in
PFOA
>97%
-
21
ECso
(immobility)
203.7
-
203.7
203.7
Boudreau 2002
Cladoceran (<24 hours),
Moina macrocopa
S,U
48 hours
PFOA
Unreported
-
25
ECso
(immobility)
199.51
166.3
166.3
166.3
Ji et al. 2008
A-2
-------
Species (lilesliiue)
Method'
Tesl
Diimlion
( hemiciil /
PuriU
pll
Temp.
<°C)
r.lTeel
Author
Reported
HITccl
(one.
(niii/l.)
r.p\
( iileuliiled
KITccl
Cone.
(111^/1.)
l-iiiid KITeel
(one.
98%
25
LC50
454
499.7
499.7
-
Li 2009
Green neon shrimp,
Neocaridina denticulata
s,u
96 hours
PFOA
>98%
-
25
LC50
454
428.1
428.1
-
Li 2009
Green neon shrimp,
Neocaridina denticulata
s,u
96 hours
PFOA
>98%
-
25
T.r
454
375.5
375.5
431.5
Li 2009
Rainbow trout (2.8 cm, 0.21 g),
Oncorhynchus mykiss
S,M
96 hours
APFO
99.4%
7.1-
7.2
11.8
i.(
4.001
-
4,001
-
DuPont Haskell
Laboratory 2000
Rainbow trout
(juvenile, 40-50 mm),
Oncorhynchus mykiss
S,U
96 hours
APFO
99.7%
6.0-
8.5
P-17
LC*.
707
-
707d
1,682
Colombo et al.
2008
Zebrafish (embryo),
Danio rerio
S,U
96 hours
PFOA
>97%
7.2-
7.5
26
I.(
5(»1
-
>500
-
Hagenaars et al.
2011
Zebrafish (embryo),
Danio rerio
s,u
96 hours
PFOA
TTnrcportcd
7.5
26-28
l.(
24.6
22.77
22.778
-
Corrales et al.
2017
Zebrafish (embryo, 4 hpf),
Danio rerio
R, U
96 hours
H<)\
Unreported
5
28
LC50
473h
548.00
548.0
-
Godfrey et al.
2017a
Zebrafish (embryo, 4 hpf),
Danio rerio
R, U
96 hours
PFOA
Unreported
7-7 5
28
LC50
473h
508.5
508.5
-
Godfrey et al.
2017a
Zebrafish (embryo, 4 hpf),
Danio rerio
R, U
96 hours
PFOA
Unreported
7-7.5
:x
1 /C50
473h
547.0
547.0
-
Godfrey et al.
2017a
Zebrafish (embryo),
Danio rerio
r. r
% hours
PFOA
Unreported
-
26
LC50
759
806.6
806.6
572.4
Stengel et al. 2017
Fathead minnow (larva),
Pimephales promelas
s. u
')<> hours
PR) \
Unreported
7.5
25
LC50
413.2
-
413.2
413.2
Corrales et al.
2017
Bluegill (2.1 cm, 0.228 g),
Lepomis macrochirus
s,u
96 hours
API'O
99%
6.9-
7.4
21.4-
22.1
LC50
634
664.0
664.0
664.0
DuPont Haskell
Laboratory 2000
American toad
(larva, Gosner stage 26),
Anaxyrus americanus
s,u
96 hours
PFOA
Unreported
-
21
LC50
711h
781.4
781.4
-
Tornabene et al.
2021
A-3
-------
Species (lilesliiue)
Method'
Tesl
Diimlion
( hemiciil /
PuriU
pll
Temp.
<°C)
r.lTeel
Author
Reported
HITccl
(one.
(inii/l.)
r.p\
( iileuliiled
KITccl
Cone.
(111^/1.)
I iii;il KITeel
(one.
4<>:
646.2
646.2
Tornabene et al.
2021
American bullfrog
(tadpole, Gosner stage 25),
Lithobates catesbeiana
(formerly, Rana catesbeiana)
s,u
96 hours
PFOA
Unreported
-
21
LCm,
1,004
1,006
1,006
-
Flynn etal. 2019
American bullfrog
(larva, Gosner stage 26),
Lithobates catesbeiana
s,u
96 hours
PFOA
Unreported
-
:i
1 ,C"
1 .<)(•(>
1,035
1,035
1,020
Tornabene et al.
2021
Green frog
(larva, Gosner stage 26),
Lithobates clamitans
(formerly, Rana clamitans)
s,u
96 hours
H<)\
I iiivpniial
-
21
l.(
1,070
-
1,070
1,070
Tornabene et al.
2021
Northern leopard frog
(larva, Gosner stage 26),
Lithobates pipiens
(formerly, Rana pipiens)
s,u
1KMI is
H<)\
I nrcpuricd
-
:i
1 'C50
752
751.7
751.7
751.7
Tornabene et al.
2021
Wood frog
(larva, Gosner stage 26),
Lithobates sylvatica
(formerly, Rana sylvatica)
s,u
lK> 1 KM lis
H<)\
Unrcpmia.1
-
:i
LC50
999
-
999
999
Tornabene et al.
2021
Frog (embryo stage 8.5),
Xenopus sp.
R,U
96 Ikui i s
PFOA
Unreported
-
23
LC50
377°
-
37T
377
Kim et al. 2013
Jefferson salamander
(larva, Harrison stage 40),
Ambystoma jeffersonianum
s,u
96 hours
PFOA
Unreported
-
21
LC50
1,070
-
1,070
1,070
Tornabene et al.
2021
A-4
-------
Species (lil'cstniic)
Method'
Tesl
Diinition
( Ik-iii ic;il /
Piiriu
pll
l oin p.
(°C)
r.nwi
Author
Kcporiod
i:riw-t
(one.
(iii'^/l.)
r.p.\
( iilculiilod
KITcct
Cone.
(m Si/1 - >
l iiiiil KITcct
(one.
till"/!.)'
Species
Mciin
Acule
\ ;iluc
tiiiu/l.)
Reference
Small-mouthed salamander
(larva, Harrison stage 40),
Ambystoma texanum
s,u
96 hours
PFOA
Unreported
-
21
LC50
4 "4
4U7.3
407.3
-
Tornabene et al.
2021
Small-mouthed salamander
(larva, Harrison stage 45),
Ambystoma texanum
s,u
96 hours
PFOA
Unreported
-
21
!•( 50
-
l,070f
407.3
Tornabene et al.
2021
Eastern tiger salamander
(larva, Harrison stage 40),
Ambystoma tigrinum
s,u
96 hours
PFOA
Unreported
-
21
L(
752
-
752
752
Tornabene et al.
2021
a S=static, R=renewal, F=flow-through, U=unmeasured, M=measured, T=total, D=dissolvcd. Diet dicl;n\. M l =maternal transfer
b Chemical concentrations made in a side-test representative of exposure and verified stability of concentrations ol PFOA in the range of concentrations tested under similar
conditions. Daily renewal of test solutions,
c Reported in moles converted to milligram based on a molecular weight of 414.07 mu niniol
d Concentration of APFO determined as the anion (PFO-).
e Values in bold used in the SMAV calculation,
f Only the most sensitive life-stage used in the SM \V calculation
g Value is considered an outlier and not used in S \ 1 \V calculation
h Author pooled test of lifestages.
A-5
-------
A.2 Detailed PFOA Acute Toxicity Study Summaries and
Corresponding Concentration-Response Curves (when calculated)
The purpose of this section is to present detailed study summaries for tests that were
considered quantitatively acceptable for criterion derivation, with summaries grouped and
ordered by genus sensitivity. C-R models developed by EPA that were used to determine acute
toxicity values used for criterion derivation are also presented. C-R models included here with
study summaries were those for the four most sensitive genera. In many cases, authors did not
report concentration-response data in the publication/supplemental materials and or did not
provide concentration-response data upon EPA request. In such cases, EPA did nol
independently calculate toxicity values and the author-reported effect concentrations were used
to derive the criterion.
A.2.1 Most acutely sensitive genus - Chvilorus
Le and Peijnenburg (2013) performed a 48-hour static unmeasured test on PFOA
(unreported purity) with the cladoceran. Chydorus sphaencus. Authors stated the test followed
the protocol of the '"Chydotox toxicity test" developed by the National Institute for Public Health
and the En\ ironment. The Netherlands In-house cultures of neonates (<24 hours) were exposed
to 25<) ul. of test solutions in 2 ml. \ ials of unreported material. Each vial contained five
neonates and each test concentration was replicated four times. No solvent was used in the test
solutions with I K-2<) test concentrations. C. sphaericus was cultured at 20 ± 1°C and a cycle of
16-hour: 8-hour light.dark without the addition of food. At test termination vials were shaken
slightly and the mobility of the neonates was determined. The author-reported 48-hour ECso was
0.22 mM PFOA (91.10 mg/L). EPA performed concentration-response (C-R) analysis for the test
and calculated a LCso of 93.17 mg/L PFOA (95% C.I. = 82.52 - 103.8 mg/L) that was acceptable
for quantitative use.
A-6
-------
Publication: Le and Peijnenburg (2013)
Species: Cladoceran (Chydorus sphaericus)
Genus: Chydorus
EPA-Calculated LCso: 93.17 mg/L (95% C.I. = 82.52 - 103.8 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
3.007
0.4165
7.2181
5.27e"13
d
0.9560
0.0248
38.5566
<2.2e"16
e
5.9057
5.9057
15.7767
<2.2e"16
Concentration-Response Model Fit:
Le and Peijnenberg 2013
Chydorus sphaericus
Log Logistic type L, 3 para
PFOA ( mg'L )
A.2.2 Second most acutely sensitive genus - Dayhnia
Logeshwaran et al. (2021) conducted acute and chronic toxicity tests with the
cladoceran, Daphnia carinata, and PFOA (95% purity, purchased from Sigma-Aldrich
Australia). In-house cultures of daphnids were maintained in 2 L glass bottles with 30% natural
spring water in deionized water, 21°C and a 16-hour:8-hour light:dark photoperiod. The acute
test protocol followed OECD guidelines (2000a) with slight modifications. A PFOA stock
solution (100 mg/L) was prepared in deionized water. Cladoceran culture medium was used to
prepare the PFOA stock and test solutions. Ten daphnids (six to 12 hours old) were transferred to
A-7
-------
polypropylene containers containing one of 14 nominal test concentrations (0, 0.5, 1, 2.5, 5, 10,
20, 30, 40, 50, 100, 150, 200 and 250 mg/L PFOA). Each test treatment was replicated three
times and held under the same conditions as culturing. At test termination (48 hours) immobility
was determined after 15 seconds of gentle stirring. No mortality occurred in the controls. The
author-reported 48-hour ECso was 78.2 mg/L PFOA. The EPA-calculated 48-hour ECso value
was 66.80 mg/L (95% C.I. = 57.10 - 76.50 mg/L), which was acceptable for quantitative use.
Publication: Logeshwaran et al. (2021)
Species: Cladoceran {Daphnia carinata)
Genus: Daphnia
EPA-Calculated LC50: 66.80 mg/L (95% C.I. = 57.10 - 76.50 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
1.6249
0.1565
10.3860
<2.2e"16
e
83.6974
5.9263
14.1230
<2.2e"16
Concentration-Response Model Fit:
Logeshwaran et al. 2021
Daphnia carinata
Weibull type 1, 2 para
PFOA (mg/L )
A-8
-------
Boudreau (2002) performed a 48-hour static unmeasured test on PFOA (CAS # 335-67-
1, >97% purity) with Daphnia magna and Daphniapulicaria as part of a Master's thesis at the
University of Guelph, Ontario, Canada. The results were subsequently published in the open
literature (Boudreau et al. 2003). Authors stated the test followed ASTM E729-96 (1999).
Daphnids used for testing were less than 24-hours old at test initiation. D. magna were obtained
from a brood stock (Dm99- 23) at ESG International (Guelph. ON. Canada). D. pulicaria were
acquired from a brood stock maintained in the Department of Zoology al I lie I niversity of
Guelph. Dilution water was clean well water obtained from ESG International I lardness was
softened by addition of distilled deionized water to achieve a range of 200-225 mu I. of CaC03.
Photoperiod was 16-hours of illumination under cool-while, fluorescent light between 380 and
480 lux. Laboratory-grade distilled water was used for all solutions with maximum
concentrations derived from stock solutions no greater than 45<) nig/L. Test vessels consisted of
225 mL polypropylene disposable containers containing I 5<) mL of test solution. All toxicity
testing involved three to lour replicates of 10 daphnids each in five unmeasured test
concentrations plus a neuati\ e control Nominal concentrations were 0 (negative control), 26.3,
52.6, 1115. 211) and 42<) mu I. I-\periments uere conducted in environmental chambers at a test
temperature of 21 ± 1JC. Authors note that temperature and pH were measured at the beginning
and end of the study, but this information is not reported. Mortality of daphnids in the negative
control was also not reported, although ASTM E729-96 requires at least 90% survival for test
acceptability. The 48-hour D. magna ECso reported in the publication was 223.6 mg/L. The 48-
hour D. pulicaria EC so reported in the publication was 203.7 mg/L.
A-9
-------
Publication: Boudreau (2002)
Species: Cladoceran {Daphnia pulicaria)
Genus: Daphnia
EPA-Calculated LCso: Not calculable, concentration-response data not available
Concentration-Response Model Fit: Not Applicable
Publication: Boudreau (2002)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: Not calculable, concentration-response data not available
Concentration-Response Model Fit: Not Applicable
Colombo et al. (2008) conducted a 48-hour static unmeasured aculc lest on PFOA
(ammonium salt, CAS # 3825-26-1, 99.7% puril\) with (he daphnid, Daphnia magna The
authors stated that the toxicity test was conducted follow i nu ()l -CD test guideline 202 (1992).
Neonates, six to 24-hours old, were acclimuled lo test conditions lor six-hours before test
initiation with test solutions made in reconstituted \I4 media. There were four replicates for each
test treatment containing five animals each. Exposure vessel material and size were not reported.
Based on loading, exposure \ essels contained at least 100 mL test solution. Nominal test
concentrations were used bused on the know n stability of the test substance in water. The
nominal test concentrations included control. 100, 178, 316, 562 and 1,000 mg/L. Dissolved
oxygen was 60% saturation and temperature was maintained between 18-22°C. Illumination
invoh ed lo-hours of light with an unreported intensity. No mortality was observed in the
controls. C-R data were available for this acute test; however, EPA was unable to fit a model
with significant parameters and relied on the 48-hour ECso reported in the study of 480 mg/L,
which was acceptable for quantitative use.
Publication: Colombo et al. (2008)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: Not calculable, unable to fit a model with significant parameters
Concentration-Response Model Fit: Not Applicable
A-10
-------
Ji et al. (2008) also performed a 48-hour static, unmeasured acute test of PFOA (CAS #
335-67-1, purity unreported; obtained from Sigma Aldrich, St. Louis, MO) with/), magna.
Authors stated that the test followed U.S. EPA/600/4-90/027F (2002). I). magna used for testing
were obtained from brood stock cultured at the Environmental Toxicology Laboratory at Seoul
National University, Korea. Test organisms were less than 24-houi s old al test initiation. Dilution
water was moderately hard reconstituted water (hardness typical l\ Si)-1 <») nig/L as CaCCte).
Experiments were conducted in glass jars of unspecified size and fill volume Pholoperiod was
assumed as 16-hour:8-hour, light:dark, the same conditions as the daphnid cultures Preparation
of test solutions was not described. The test involved four replicates of five daphnids each in five
unmeasured test concentrations plus a negative control. Nominal concentrations were 0 (negative
control), 62.5, 125, 250, 500 and 1,000 mu I. Test temperature was maintained at 21 ± 1°C.
Authors noted water quality parameters (pH, temperature, conductivity, and dissolved oxygen)
were measured 48-hours after exposure, but the information was not reported. Mortality of
daphnids in the negati\ e control was not reported, although EPA/600/4-90/027F requires at least
90% sur\ i\ al for test acceptability The author-reported 48-hour ECso for the study was 476.52
mg/L. (^5"oC.I. = 375 3 - 577 7 mu I.) N\\ performed C-R analysis for the test. TheEPA-
calculated was 542.5 mu L PTOA (95% C.I. = 461.1 - 623.8 mg/L), which was acceptable
for quantita1i \ e use
A-ll
-------
Publication: Ji et al. (2008)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 542.5 mg/L (95% C.I. = 461.1 - 623.8 mg/L),
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
3.7248
1.8230
2.0432
0.0410
d
0.8985
0.0393
22.8879
< 2.0 e"16
e
598.5588
66.9972
8.9341
< 2.0 e"16
Concentration-Response Model Fit:
Ji et al. 2008
Daphnia magna
WeibuH type L 3 para
PFOA ( mgtL )
Li (2009) conducted a 48-hour static unmeasured acute test on PFOA (ammonium salt,
>98% purity) with Daphnia magna. The authors stated that the test followed OECD 202 (1984)
with slight modifications. D. magna used for the test were less than 24-hours old at test
initiation. Dilution water was dechlorinated tap water. The photoperiod consisted of 12-hours of
illumination at an unreported light intensity. A primary stock solution was prepared in dilution
water and did not exceed 400 mg/L. Exposure vessels were polypropylene of unreported
dimensions and 50 mL fill volume. The test employed five replicates of six daphnids each in five
A-12
-------
test concentrations plus a negative control. Based on water solubility of test chemicals and
preliminary toxicity results, nominal test concentrations were in the range of 31-250 mg/L for
PFOA. The test was conducted in a temperature incubator at 25 ± 2°C. Water quality parameters
including water pH, conductivity, and dissolved oxygen were measured at the beginning and at
the end of each test. Initial values of pH were 7.82 ± 0.12 and 7.91 <") <)3 after 48-hours. At the
start of the bioassays, dissolved oxygen and specific conductivity were (->7 7 ± 6.8% saturation
and 101.8 ± 6.8 |iS/cm. After the 48-hour testing period, dissolved oxygen and specific
conductivity were 55.6 ± 1.26% saturation (implying 4.56 mg/L) and 109.1 3.5 uS/cm,
respectively. None of the control animals became immobile at the end of the test. The author-
reported 48-hour ECso for the study was 181 mg/L (95"0 C. 1.. 166-198 mg/L) which was
averaged across three tests. EPA performed C-R analysis for each individual test. All three tests
had acceptable curves with EPA-calculated !¦('=..s of 22<) S mg/L (95% C.I. = 191.8 - 250.0
mg/L), 157.9 mg/L (^5"o (' I. 135 9- 18<> n mu I.), and 207.3 mg/L (95% C.I. = 176.1 -238.5
mg/L), which were acceptable lor quantitative use
A-13
-------
Publication: Li (2009)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 220.8 mg/L (95% C.I. = 191.8 - 250.0 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
3.2035
0.6834
4.6875
2.766 e6
e
247.6075
17.6535
14.0260
< 2.2 e16
Concentration-Response Model Fit:
Li 2009
Daphnia magna
Weibull type 1,2 para
PFOA ( mg'L )
A-14
-------
Publication: Li (2009)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 157.9 mg/L (95% C.I. = 135.9 - 180.0 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
2.8137
0.4368
6.4423
1.177 e10
e
179.9061
12.3471
14.5707
< 2.2 e"16
Concentration-Response Model Fit:
Li 2009
Daphnia magna
Weibull type 1,2 para
PFOA ( mg'L )
A-15
-------
Publication: Li (2009)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 207.3 mg/L (95% C.I. = 176.1 -238.5 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
2.5732
0.4477
5.7479
9.036 e9
e
239.0336
19.0886
12.5223
< 2.2 e"16
Concentration-Response Model Fit:
Li 2009
Daphnia magna
Weibutl type 1, 2 para
PFOA ( mg.'L )
Yang et al. (2014) conducted a 48-hour measured acute test of PFOA (CAS # 335-67-1,
99% purity) with Daphnia magna, following ASTM E729 (1993). Although the authors termed
the test conditions "static", they also mentioned PFOA measurements before and after renewal;
based on this distinction the test was assumed to be renewed at least once. Daphnids used for the
test were donated by the Chinese Research Academy of Environmental Sciences. The daphnids
were less than 24-hours old at test initiation. Dilution water was dechlorinated tap water (pH, 7.0
± 0.5; dissolved oxygen, 7.0 ± 0.5 mg/L; total organic carbon, 0.02 mg/L; and total hardness,
190.0 ±0.1 mg/L as CaC03). The photoperiod consisted of 12-hours of illumination at an
A-16
-------
unreported intensity. A primary stock solution was prepared by dissolving PFOA in deionized
water and solvent, DMSO, and proportionally diluted with dilution water to prepare the test
concentrations. Exposure vessels were 200 mL beakers of unreported material type containing
100 mL of test solution. The test employed three replicates of 10 daphnids each in six test
concentrations (measured in low and high treatments only) plus a negative and solvent control.
Nominal concentrations were 0 (negative and solvent controls). 5<). Si). 128. 204.8, 327.68 and
524.29 mg/L. The authors provided mean measured concentrations before and after renewal:
49.62 and 43.93 mg/L (lowest concentration) and 526.9 and 476.41 mg/L (highest
concentration). Analyses of test solutions were performed using HPLC/MS and negati\ e
electrospray ionization. The concentration of PFOA was calculated from standard curves (linear
in the concentration range of 1-800 ng/ml.). and the average extraction efficiency was in the
range of 70-83%. The concentrations and chromatographic peak areas exhibited a significant
positive correlation (r n l^S7. p : 0.01), and the water sample-spiked recovery was 99%. The
temperature. DO. and pi I were reported as having been measured every day during the acute
test, but results were not reported Negati\ e control and solvent control mortality were 0% each.
The author-reported 4S-houi' I Ibr the study was 201.85 mg/L (95% C.I. = 134.7 - 302.5
mg/I.) I-PA performed C-R analysis for the test and had an acceptable curve with an EPA-
calculated I .("*¦¦ of 222.0 mg I. PFOA (95% C.I. = 190.5 -253.5 mg/L). The acute value was
acceptable for quantitati\ e use.
A-17
-------
Publication: Yang et al. (2014)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 222.0 mg/L (95% C.I. = 190.5 -253.5 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
1.1031
0.1773
6.2226
4.89 e10
e
309.4319
36.3820
8.5051
< 2.2 e"16
Concentration-Response Model Fit:
Yang et. al. 2014
Daphnia magna
Weibull type 1, 2 para
PFOA ( mg/L )
Barmentlo et al. (2015) performed a 48-hour static, measured acute test of PFOA (CAS
# 335-67-1, >96%) with Daphnia magna. Authors stated the test followed OECD 202 (2004)
guidelines for testing. D. magna used for testing were obtained from Grontmij, Amsterdam, and
cultured in M4 media according to OECD 211 (2008). Test organisms were less than 24-hours
old at test initiation. Dilution water was ISO medium. Experiments were conducted in 50 mL
polypropylene tubes with 20 mL of test solution. The photoperiod consisted of 16-hours of
illumination at an unreported intensity. PFOA stock was made with demineralized water. The
test involved four to six replicates of five daphnids each in five test concentrations plus a
A-18
-------
negative control. Nominal concentrations were not provided, but PFOA was measured in the
control, lowest and highest test concentrations. Based on these measurements, the authors
interpolated all test concentrations, 0.053 (negative control), 81, 128, 202, 318 and 503 mg/L.
Test temperature was maintained at 20 ± 1°C, pH ranged from 7.00-7.82, and D.O. ranged from
8.54-9.42 mg/L. Mortality of daphnids in the negative control was not reported. The author-
reported 48-hour ECso for the study was 239 mg/L (95% C.I.: 11><) - 2S7 mg/L). EPA performed
C-R analysis for the test and had an acceptable cin \ c with an l-IW-calailalcd EC50 of 215.6
mg/L PFOA (95% C.I. = 181.7 - 249.5 mg/L). The acute \ nine was acceptable lor quantitative
use.
A-19
-------
Publication: Barmentlo et al. (2015)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 215.6 mg/L PFOA (95% C.I. = 181.7 - 249.5 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
2.3893
0.4288
5.5728
2.507 e8
e
251.3332
19.6475
12.7921
< 2.2 e"16
Concentration-Response Model Fit:
Barmentlo et al. 2015
Daphnia magna
Weibull type 1,2 para
PFOA ( mg'L )
Ding et al. (2012a) conducted a 48-hour static, partially measured acute test on PFOA
(CAS # 335-67-1; 96% purity from Sigma Aldrich) with/), magna. The test was performed
following OECD test guideline 202 (2004) with slight modifications. D. magna used for testing
were purchased from local suppliers and cultured for two months prior to use. Test organisms
were less than 24-hours old at test initiation. Dilution water was M4 solution prepared following
the OECD test guideline. The photoperiod consisted of a 16 hour:8 hour light:dark cycle at an
unreported light intensity. A primary stock solution was prepared in dilution (reconstituted M4)
water. Exposure vessels were 50 mL polypropylene disposable tubes containing 20 mL of test
A-20
-------
solution. The test involved four replicates of five daphnids each in six test concentrations plus a
negative control. Nominal concentrations were 0 (negative control), 0.35, 0.4, 0.45, 0.5, 0.55 and
0.6 mM PFOA, or 0, 144.9, 165.6, 186.3, 207.0, 227.7, and 248.4 mg/L after conversion by
multiplying the mM concentration by a molecular weight of 414.07 g/mol for PFOA. The
subsequent concentrations are reported in the converted units of mg/L. Concentrations of PFOA
were confirmed in the highest and lowest concentrations, though only nominal concentrations
were reported. It was stated that the verified concentration was "well in line with nominal
concentrations". Test temperature was maintained at 20 ± 1 °C. Observations u ere made at 24-
hours and 48-hours after test initiation. EC so values were reported for both observational time
periods. The 48-hour ECso was reported as 211.6 mg/L with the 95% confidence levels of 184.7 -
255.5 mg/L and aNOEC of 207.0 mg/L. I-PA performcd C-R analysis for the test. The EPA-
calculated ECso was 216.1 mg/L PFOA (^5"o C I 2
-------
Publication: Ding et al. (2012a)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 216.1 mg/L PFOA (95% C.I. = 206.1 - 225.9 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std. Error
t-stat
p-value
b
7.5008
1.3478
5.5650
2.621 e8
e
0.5478
0.0128
42.6800
< 2.2 e16
Concentration-Response Model Fit:
Ding et al. 2012
Daphnia magna
Weibull type 1,2 para
l.oo-
0.75-
't
m
g 0.50-
•-P
o
£X
O
&
025-
0.00-
le-04 le-03 le-OQ le-01
PFOA ( mM )
Lu et al. (2016) evaluated the acute toxicity of PFOA (CAS# 335-67-1, 98% purity,
purchased from Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) on Daphnia magna
immobilization. Reconstituted daphnia culture media was used for both culturing and test
solution preparation as described in OECD Test Guideline 202. D. magna cultures (originally
obtained from the Chinese Center for Disease Control and Prevention (Beijing, China) were fed
with the green algae Scenedesmus obliquits daily, maintained at 20°C and a light/dark
photoperiod of 16 h/8 h and the medium renewed three times weekly. The 48-hour static
unmeasured acute test was conducted via a modified OECD standard test procedure 202,
whereby five concentration treatments (3, 10, 30, 100 and 300 mg/L) plus a blank control were
A-22
-------
employed. Ten neonates (<24-hours old) from a designated brood were placed in a 100 mL glass
beaker containing 45 mL test solution for each test concentration and control. Test daphnids
were not fed during the testing period and each treatment was replicated three times. The status
of immobilization and mortality was checked at 48 hours (daphnids unable to swim within 15
seconds after gentle agitation of the test container are considered to be immobile and those
animals whose heartbeats have stopped are considered dead). Authors reported immobility/
survival to be a more sensitive endpoint than survival alone. 1'he authoi-iqxmed 48-hour ECso
for immobility/survival was 110.7 mg/L and the NW-calailalcd 4S-hourL(';.. was 114.595
mg/L (95% C.I. = 93.71 - 135.5 mg/L).
A-23
-------
Publication: Lu et al. (2016)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 114.595 mg/L (95% C.I. = 93.71 - 135.5 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
0.5649
0.1050
5.3783
7.517 e8
e
219.2463
68.2427
3.2127
0.0013
Concentration-Response Model Fit:
Lu et al. 2016
Daphnia magna
Weibull type 1, 2 para
PFOA (mg/L )
Yang et al. (2019) evaluated the acute effects of PFOA (CAS# 335-67-1, purchased from
Sigma-Aldrich in St. Louis, MO) on Daphnia magna via a 48-hour unmeasured static exposure.
D. magna cultures were originally obtained from the Institute of Hydrobiology of Chinese
Academy of Science in Wuhan, China. Organisms were cultured in Daphnia Culture Medium
according to the parameters specified in OECD Guideline 202. Protocol for all testing followed
OECD Guideline 202. Cladocerans were cultured in artificial freshwater maintained at 20 ± 1°C
A-24
-------
under a 16-hour:8-hour light:dark photoperiod and a light intensity of 1,000-1,500 lux at the
surface of the water. Cultures were fed Scenedesmus obliquus daily and the water was changed
twice weekly. Reported water quality parameters include total hardness of 140-250 mg/L as
CaCCte and pH of 6-8.5. Acute test concentrations included 0 (control), 0.000161, 0.000193,
0.000232, 0.000278, 0.000334 and 0.000401 mol/L (or 0 (control), 66.67, 79.92, 96.06, 115.1,
138.3, and 166.0 mg/L given the molecular weight of the form of PI OA used in the study, CAS
# 335-67-1, of 414.07 g/mol). Five neonates (12-24 hours old) were placed randomly in 100 mL
glass beakers filled with 60 mL test solution, with four replicates per concentration. Organisms
were observed for mortality at 48 hours, and the authors reported a LCso of 0.0002^2 mol/L, or
120.9 mg/L PFOA. The EPA-calculated 48-hour LCso was 117.192 mg/L (95% C.I. = 112.2 -
122.2 mg/L).
A-25
-------
Publication: Yang et al. (2019)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated LCso: 117.192 mg/L (95% C.I. = 112.2 - 122.2 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
6.0229
1.0808
5.5724
2.512 e8
d
0.9998
0.0103
96.6600
< 2.2 e16
e
117.1921
4.8691
24.0685
< 2.2 e16
Concentration-Response Model Fit:
Yang et al. 2019
Daphnia magna
Log Logistic type 1, 3 para
PFOA (mg/L)
A.2.3 Third most acutely sensitive genus - Brachionus
Zhang et al. (2013a) performed a 24-hour static test of PFOA (CAS # 335-67-1, 96%
purity) with Brachionus calyciflorus. Organisms were neonates less than two-hours old at test
initiation. All animals were parthenogenetically-produced offspring of one individual from a
single resting egg collected from a natural lake in Houhai Park (Beijing, China). The rotifers
were cultured in an artificial inorganic medium at 20°C (16-hour:8-hour, light:dark; 3,000 lux)
for more than six months before toxicity testing to acclimate to the experimental conditions. All
A-26
-------
toxicity tests were carried out in the same medium and under the same conditions as during
culture (i.e., pH, temperature, illumination). Solvent-free stock solutions of PFOA (1,000 mg/L)
were prepared by dissolving the solid in deionized water via sonication. After mixing, the
primary stock was proportionally diluted with dilution water to prepare the test concentrations.
Exposures were in 15 mL, six-well cell culture plates (assumed plastic) each containing at total
of 10 mL of test solution. The test employed seven measured lest concentrations plus a negative
control. Each treatment consisted of one replicate plate of 10 rotifers each in individual cells and
repeated six times. Nominal concentrations were 0 (negative control), 60, 80. I no. 120, 140, 160,
and 180 mg/L. PFOA concentrations were not measured in the rotifer exposures, bill rather, in a
side experiment using HPLC/MS. The side experiment showed that the concentration of PFOA
measured every eight-hours over a 24-hour period in rotifer medium with green algae incurs
minimal change in the concentration range from o 25 lo 2 o mu I. 1'he acute test was conducted
without green algae added lo the exposure medium Although this rotifer species has a short life
span, a 24-hour unled lest is not expected to cause starvation and 0% mortality was observed at
test termination in the neuali\ e control The study reported 24-hour LCso was 150.0 mg/L. The
acute value was acceptable lor <.|iianlilative use.
Publication: Zhang et al. (2013a)
Species Rotifer (Brachionus calyciflorus)
Genus Hrachionus
EPA-Calculntcd T.On Not calculable, concentration-response data not available
Concentraiion-Ucsponse Model Fit: Not Applicable
A.2.4 Fourth and fifth most acutely sensitive genera - Ligumia and Lampsilis (mussels)
Hazelton et al. (2012) and Hazelton (2013) evaluated the acute effects of PFOA (96%
purity) on two freshwater mussels: Lampsilis siliquoidea and Ligumia recta. Acute toxicity was
observed under static conditions over a 24-hour period (<24-hour old glochidia) or renewal
conditions over a 96-hour period (four to six-week-old juveniles). Authors stated the tests
A-27
-------
followed ASTM E2455-06 (2006). Dilution water was hard reconstituted water. Photoperiod and
light intensity were not reported. No details were provided regarding primary stock solution and
test solution preparation. Experiments were conducted in 3.8 L glass jars of unspecified fill
volume. The test employed three replicates of 150 glochidia or seven juvenile mussels each in
six measured test concentrations plus a negative control (10 juveniles for the control treatment).
Nominal concentrations were 0 (negative control), 0.005, 0.05. n 5. 5. 50. and 500 mg/L, while
corresponding mean measured concentrations were less than the limit of quantification (LOQ,
specifics not provided), 0.0051, 0.0484, 0.490, 4.8, 51, and 476 mg/L PFOA. respectively.
Analyses of test solutions were performed at the U.S. EPA National Exposure Research
Laboratory in Research Triangle Park, NC using HPLOMS. Measured test concentrations of
PFOA were within 10% of target in water from acute tests. Reco\ cry of PFOA standards ranged
from 91.2-108%. For all acute tests, alkalinity ranged from ^7 to 110 mg/L as CaC03 with a
mean of 104.4 mu I.. lolal hardness ranged from 132 to I (->2 mg/L as CaC03 with a mean of
149.6 mg/L; conducli\ ily ranged from 514 to 643 |iS/cm with a mean of 556.5 |iS/cm; pH
ranged from S <)5 lo S 5o with a mean of X 4(\ and dissolved oxygen ranged from 8.16 to 9.46
mg/L. with a mean of S (->2 mu I. (n 12 for alkalinity and total hardness, n = 55 for all other
parameters) Exposures were conducted at 20°C. Mortality of mussels in the negative control
was <10" o in all exposures The 24-hour EC so reported for glochidia of L. siliquoidea was 164.4
mg/L (95% C.L.: I !(•><)- 232.8 mg/L) and fori, recta, 161.0 mg/L (95% C.L.: 135.0 - 192.7
mg/L). The 96-hour LCso values for the juvenile L. siliquoidea and L. recta were greater than the
highest test concentration (500 mg/L). The study reported 24-hour ECsos for L. siliquoidea and
for L. recta represent acute values acceptable for quantitative use for the two mussel species. The
juvenile life stage is less sensitive, such that its LCsos were not used quantitatively in SMAVs.
A-28
-------
Text pertaining to C-R modeling below is only described for L. recta because this species was
among the four most sensitive acute genera; L. siliquoidea was among the fifth most sensitive
genus and C-R curves were only displayed for those species within the four most sensitive acute
genera.
Publication: Hazelton et al. (2012, 2013)
Species: Black sandshell, {Ligumia recta)
Genus: Ligumia
EPA-Calculated LC50: Not calculable, concentration-response data not available
Concentration-Response Model Fit: Not Applicable
A.2.5 Sixth most acutely sensitive genus -Moina
Ji et al. (2008) performed a 48-hour static. 1111 measured acute test of PFOA (CAS # 335-
67-1, purity unreported; obtained from Sigma Aldrich. Si I .ouis. MO) with Moina macrocopa.
Authors stated the test followed U.S. EPA (¦><)<) 4-l)<)/U27F (2t)t)2) \ 1. macrocopa used for testing
were obtained from brood stock cultured al the I -n\ ironmenlal Toxicology Laboratory at Seoul
National Universily. Korea Test organisms were less than 24-hours old at test initiation. Dilution
water was moderately hard reconstituted water (total hardness typically 80-100 mg/L as CaCCte).
Experiments were conducted in glass jars of unspecified size and fill volume. Photoperiod was
assumed as lO-hour S-hour. light dark, the same conditions as the daphnid cultures. Preparation
of test solutions was not described The test involved four replicates of five daphnids each in five
unmeasured test concentrations plus a negative control. Nominal concentrations were 0 (negative
control), 62.5, 125. 25<). 5<)<) and 1,000 mg/L. Test temperature was maintained at 25 ± 1°C.
Authors noted water quality parameters (pH, temperature, conductivity, and dissolved oxygen)
were measured 48-hours after exposure, but the information was not reported. Survival of
daphnids in the negative control was not reported, although EPA/600/4-90/027F requires at least
90% survival for test acceptability. The author-reported 48-hour EC50 for the study was 199.51
mg/L (95% C.I. = 163.9 - 245.1). EPA performed C-R analysis for the test. The EPA-calculated
A-29
-------
ECso was 166.3 mg/L PFOA (95% C.I. = 138.6 - 194.1 mg/L) and was acceptable for
quantitative use.
A.2.6 Seventh most acutely sensitive genus -Xenoyus
Kim et al. (2013) conducted a 96-hour renewal unmeasured assay with perfluorooctanoic
acid (PFOA) using the frog embryo teratogenesis assay -Xenopus (FETAX). PFOA stock
solutions were prepared by dissolving PFOA in dimethyl sulfoxide (DMSO), and then diluting in
FETAX medium for exposure solutions (DMSO did not exceed 0.15%) Adult Xenopus were
purchased from Nasco (Fort Atkinson, WI) and housed in clear plastic aquariums with
dechlorinated tap water at 18 ± 2°C with a 12-hour light cycle and fed three times a week.
Ovulation was induced by injecting 1,000 IU of human chorionic gonadotropin just under the
skin of a female in the evening. The next day. females laid eggs in 60 mm plastic dishes. The
eggs were immediately fertilized in 0.1X modified liai lh solution (MBS) (Xenopus testes were
obtained from sacrificed males) Following successful fertilization, the jelly coat was removed
by swirling the enilnyos in a 2"(> I .-cysteine solution. The embryos were then transferred to IX
MBS containing 3"o I'icoll 4o<) I 'nfertilized eggs and dead embryos were removed and
maintained at 22 n 5 ('. finely clea\ ed embryos in the blastula stage (stage 8.5) were selected,
with 2<)-25 embryos used per concentration (nominal concentrations of 100, 500, 750, 1000 and
1,250 |iM PI OA. or 41.4, 2<)7 0, 310.6, 414.1, and 517.6 mg/L PFOA). DMSO alone (0.1%) and
FETAX medium alone were used as controls. Embryos were incubated at 23°C until the end of
the assay. The media were changed every day, and dead embryos were removed. At the end of
the experiments, embryo mortality was recorded and surviving embryos were fixed in 4%
formaldehyde to check for malformation. Head-tail lengths and malformations analyzed to
measure growth inhibition. The authors reported a 96-hour LCso of 377 mg/L PFOA and the
value was acceptable for quantitative use.
A-30
-------
A.2.7 Eighth most acutely sensitive genus -Dugesia
Li (2008) conducted a 96-hour static, unmeasured acute toxicity test on PFOA (CAS #
3825-26-1, >98% purity) with the planarian, Dugesia japonica (a non-North American species).
The test organisms were originally collected from Nan-shi stream located in Wu-lai, Taipei
County, Taiwan in 2004 and maintained in the laboratory in dechlorinated tap water. The
planarians had a body length of 0.9 ± 0.1 cm attest initiation. The dilution water was
dechlorinated tap water and a primary stock solution of PFOA was prepared in the same dilution
water. The photoperiod consisted of 12-hours of illumination at an unreported intensity.
Exposure vessels were polypropylene beakers of unreported dimensions with a 5<) ml. Ill I
volume. The test employed five replicates of five planarians each in at least five lest
concentrations plus a negative control. Nominal test concentrations were in the range of 100-750
mg/L PFOA. The test temperature was maintained at 25 I °C. No other water quality
parameters were reported for test solutions. Survival of negative control animals was not
reported. The study reported a lH->-hour I.('so was 458 mg/L (95% C.I. = 427 - 491 mg/L). The
acute value was acceptable lor <.|iianlilati\ e use
l.i (2009) conducted a second lWi-hour static, unmeasured acute test of PFOA
(ammonium salt, >lW() purity) with / >ni*csiajaponica. Again, the tested individuals were
originally collected from Nan-shi stream located in Wu-lai, Taipei County, Taiwan in 2004 and
maintained in the In born lory in dechlorinated tap water. The planarians had a body length of 0.9
±0.1 cm at test initiation. The dilution water was dechlorinated tap water and a primary stock
solution of PFOA was prepared in the same dilution water. The photoperiod consisted of 12-
hours of illumination at an unreported intensity. Exposure vessels were made of polyethylene
with unreported dimensions and 50 mL fill volume. The test employed three replicates of 10
planarians each in at least five test concentrations plus a negative control. The test was repeated
A-31
-------
three times with different test concentrations. Nominal test concentrations were in the range of
150-750 mg/L PFOA. The test temperature was maintained at 25 ± 2°C. Water quality
parameters including pH, conductivity, and D.O. were reported as having been measured at the
beginning and end of each test, but the information is not provided. Organisms were not fed, and
no mortality was observed in the control groups in any of the three tests The author-reported 96-
hour LCso was 337 mg/L (95% C.I. = 318-357 mg/L) which was a\ eraued across the three tests.
EPA performed C-R analysis for each individual test. Two of the tests had acceptable curves
with EPA-calculated LCso values of 321.8 mg/L PFOA (95% C.l. = 290.0 353 I mg/L) and
383.0 mg/L PFOA (95% C.l. = 347.8 - 418.2 mg/L) which were acceptable for quantitative use.
The third curve had a poor concentration response and the LCso (427.7 mg/L; 95% C.I. = 251.4 -
604.1 mg/L) was, therefore, not used quantitatively but considered lor qualitative use only.
A.2.8 Ninth most acutely sensitive izenus - Pimephales
Corrales et al. (2017) evaluated the acute toxicity of PFOA to the fathead minnow
(Pimephalespromelas) I jnlnyos were exposed lo ITOA for 96-hours employing static
unmeasured procedures (IS. N\\ Ol-('l) 2d 13). Fish were housed in a flow-through
system supplied with aged, dechlorinaled tap water at a constant temperature of 25 ±1°C under a
16 h:S h light'dark pholoperiod They were fed twice daily with brine shrimp (Artemia sp.
nauplii) and TelraMin Tropical Flakes. Individuals were aged to at least 120 days before
breeding at which lime lliey were placed in tanks in a 1:4-5 male to female ratio. Embryos were
collected, and within 24-hours post hatched larvae were used for toxicity studies. Glass beakers
were used as experimental units; 10 fathead minnow larvae were placed in each 500 mL beaker
containing 200 ml test solution. Before the start of each experiment, all solutions were titrated to
pH 7.5 following standard methods. General water chemistry measures (e.g., alkalinity, total
hardness, dissolved oxygen, and temperature) were also routinely monitored (assume same
A-32
-------
culture and test physico-chemical test conditions). The reported 96-hour LCso was 413.2 mg/L
PFOA and was determined to be quantitatively acceptable for criterion derivation.
A.2.9 Tenth most acutely sensitive genus -Neocaridina
Li (2009) conducted a 96-hour acute test on PFOA (ammonium salt, >98% purity) with
the freshwater shrimp species, Neocaridina denticulata (a non-North American species). Test
conditions were static (no solution renewal), and test concentrations were unmeasured. Test
organisms were obtained from an unspecified local supplier and acclimated in the laboratory for
at least seven days prior to the experiments. N. denticulata of unspecified auc were used at test
initiation and were reported to be 1.3 ± 0.2 cm long. Dilution water was dechlorinalcd tap water.
The photoperiod consisted of 12-hours of illumination at an unreported light intensity. A primary
stock solution was prepared in dilution ^ater Exposure vessels were polypropylene beakers of
unreported dimensions and 1 L fill volume The lesl employed ll\ e replicates of six organisms
each in at least fi\ e lesl concentrations plus a neuati\ e control liach treatment was tested three
different times. Nominal test concentrations were in the range of 50-1,000 mg/L PFOA. The test
temperature was maintained at 25 2 (' Water quality parameters including pH, conductivity,
and D () were reported as ha\ inu Ixvn measured at the beginning and end of each test, but the
information is not reported Mortality of negative control animals was 10% for one treatment, but
0% in others The author-reported 96-hour LCso reported in the study was 454 mg/L (95% C.I.:
418-494 mg/L) which was a\ eraged across three tests. EPA performed C-R analysis for each
individual test. All three tests had acceptable curves with EPA-calculated LCsos of 499.7 mg/L
(95% C.I. = 457.4 - 542.1 mg/L), 428.1 mg/L (95% C.I. = 396.3 - 459.9 mg/L), and 375.5 mg/L
(95% C.I. = 296.5 - 454.4 mg/L), which were acceptable for quantitative use.
A-33
-------
A.2.10 Eleventh most acutely sensitive genus - Danio
The acute effects of PFOA on the zebrafish, Danio rerio, have been reported by
numerous researchers. Hagenaars et al. (2011) exposed D. rerio embryos to PFOA (CAS #335-
67-1, purity >97%) under static unmeasured conditions for 120-hours. The PFOA was dissolved
in medium-hard reconstituted laboratory water, which was aerated and kept at 26°C until use (no
solvent). Adult wildtype zebrafish (breeding stock) were obtained from a commercial supplier
(Aqua hobby, Heist-op-den-berg, Belgium) and kept in aerated and biologically filtered medium-
hard reconstituted freshwater. Four males and four females were used for egg production, with
fertilized eggs collected in egg traps within 30 minutes of spawning. Eggs were li ansleired to the
test solutions within 60 minutes after spawning. Eggs with anomalies or damaged membranes
were discarded, and fertilized eggs were separated from the non-fertilized eggs using a
stereomicroscope. Twenty normally shaped fertilized eggs per exposure concentration were
divided over a 24-well plastic plate and each embryo was placed individually in 2 mL of the test
solution. The remaining lour wells were Hlied with clean water and used for the control eggs.
Two replicate plates were used for each exposure concentration resulting in 40 embryos per
exposure condition al the beginning of the experiment. The 24-well plates were covered with a
self-adhesi\ e foil, placed in an incubation chamber at 26 ± 0.3°C, pH 7.2-7.5 and subjected to a
14-hour:10-hour, light dark cycle. A test was considered valid if more than 90% of the controls
successfully hatched and showed neither sublethal nor lethal effects. The authors reported a 96-
hour LCso of >500 mg/L PFOA and was classified as quantitative.
D. rerio embryos (4 hours post-fertilization; hpf) were also subjected to PFOA by
Godfrey et al. (2017a) in a 96-hour acute toxicity test using static renewal exposures that were
not analytically confirmed. Stock solutions were prepared by dissolving PFOA in 1 L of reverse
osmosis (RO) water containing 12.5 |iL Replenish (Seachem Laboratories Inc.) and then
A-34
-------
adjusted to neutral pH (7-7.5). Adult zebrafish, AB wild-type, were maintained at a water
temperature of 28 ± 1°C and a photoperiod of 14-h L: 10-h D. Fish were fed twice daily, Artemia
nauplii in the morning and Tetramin in the afternoon, and genders were kept separate overnight
at a ratio of 2 males: 1 female. Randomly collected embryos (20 per concentration, gastrula stage,
4.5- hpf) were placed in plastic petri dishes containing 25 mL of exposure solution for 96-hours
at 28°C. Each test consisted of a minimum of two replicates per dose and test solutions were
renewed daily (nominal exposure solutions ranged from 250-1,000 mu I. I'TOA). The author-
reported 96-hour LCso was 473.0 mg/L PFOA which was averaged across four tests. EPA
performed C-R analysis for each individual test. Three tests had acceptable cm \ es w ith EPA-
calculated LCsos of 548.0 mg/L (95% C.I. = 530.6 - 565.5 mg/L), 508.5 mg/L (95% C.I. = 471.4
- 545.6 mg/L), and 547.0 mg/L (95% C.l 5 lo n 578 0 mu I.), which were acceptable for
quantitative use. The fourth test had an unacceptable cui\ e and therefore the EPA-calculated
LCso of 560.1 mu I. PI OA (^5"o (' T = 556.4 5o3 8 mu I.) was not used.
Stengel el ;il. (2017) exposed / K rerio embryos to PFOA for 96-hours using renewal
unmeasured procedures as specified in OI X'I) (2' >13) guidelines. PFOA stock and exposure
solutions were prepared in reconstituted laboratory water. All adult zebrafish used for breeding
were wild-type descendants of the "\Yestaquarium" strain and obtained from the Aquatic
Ecology and Toxicology breeding facilities at the University of Heidelberg. Details of zebrafish
maintenance, egu production and embryo rearing are provided as described previously (Kimmel
et al. 1995, 1988; Nagel 2002; Spence et al. 2006; Wixon 2000) and have been updated for the
purpose of the zebrafish embryo toxicity test by Lammer et al. (2009). Embryos were exposed at
the latest from 1 hpf in glass vessels, which had been preincubated (saturated) for at least 24-
hours, to a series of nominal PFOA dilutions (0, 400, 512, 640, 800 and 1,000 mg/L). After
A-35
-------
verifying fertilization success, embryos were individually transferred to the wells of 24-well
plates, which had been pre-incubated with 2 mL of the test solution per well for 24-hours prior to
the test start, and kept in an incubator at 26.0 ± 1.0°C under a 14-hour: 10-hour light:dark regime.
In order to prevent evaporation or cross-contamination between the wells, the plates were sealed
with self-adhesive foil. Embryo tests were classified as valid if the mortality in the negative
control was <10%, and if the positive control (3,4-dichloroanilinc) showed mortalities between
20% and 80%. All fish embryo tests were run in three independent replicates The author-
reported 96-hour LCso was 759 mg/L PFOA. EPA performed C-R analysis for the lest and had
an acceptable curve with an EPA-calculated LC50 of 806.6 mg/L (95% C.I. = 773 (•> 839.6
mg/L) and was determined to be quantitatively acceptable for criterion derivation.
Corrales et al. (2017) exposed 1). rcrio embryos to PI OA lor 96-hours employing static
unmeasured procedures (U.S. EPA 2002, OECD 2013) Tropical 5D wild type adult zebrafish
were kept at a density of less than lour fish per liter in a /-mod recirculating system with water
(pH 7.0, 260 ppm Instant Ocean) maintained at 26-28°C and a 16-hour:8-hour light/dark cycle.
Zebrafish were led twice daily with brine shrimp {Anemia sp. nauplii) and once per day with
TetraMin Tropical Hakes Sexually mature lish were bred to produce embryos for toxicity
studies (ilass beakers were used as experimental units; 15 zebrafish embryos in 100 mL beakers
containing 3d ml test solution. Before the start of each experiment, all solutions were titrated to
pH 7.5 following standard methods. General water chemistry measures (e.g., alkalinity, total
hardness, dissolved oxygen, and temperature) were also routinely monitored (assume same
culture and test physico-chemical test conditions). The author-reported 96-hour LC50 was 24.6
mg/L PFOA. EPA performed C-R analysis for this test with an EPA-calculated LC50 of 22.77
mg/L PFOA (95% C.I. = 13.30 - 32.20 mg/L) which was acceptable for quantitative use. This
A-36
-------
LCso value, however, was excluded from derivation of the acute criterion because a comparative
assessment between this LCso value and the other five quantitatively-acceptable zebrafish LCso
values available (Godfrey et al. 2017a; Hagenaars et al. 2011; Stengel et al. 2017), indicated the
LCso reported by Corrales et al. (2017) was an outlier, falling out more than an order of
magnitude lower than the other four LCso values.
A.2.11 Twelfth most acutely sensitive genus - Hyla
Tornabene et al. (2021) conducted an acute PFOA (purchased lVom Sigma Aldrich,
Catalog # 171468-25G; purity not provided) toxicity test with the gray I reel You. Hyla versicolor.
The acute test followed standard 96-hour acute toxicity lest guidance (U.S. EPA 2<)(P; ASTM
2008, 2017). Frog egg masses were collected from 1he field in the wetlands of Indiana near the
campus of Purdue University. Collected egg masses were raised outdoors in 200 L polyethylene
tanks filled with well water. Experiments began when iVous reached (iosner stage 26, defined as
when larvae are free swimming and feeding IJelbre lest initiation larvae were acclimated to test
conditions (21°C and 12-hour 12-hour light dark photoperiod) for 24 hours. A stock solution of
PFOA (2,000 mg/L) was made in I V-liltered well water and diluted with well water to reach
test concentrations (ranged from <)-2.<)()() mg LPFOA). Test concentrations were not measured in
test solutions based on pre\ ions research that demonstrated limited degradation under similar
conditions I ,ar\ a w ere transferred individually to 250 mL plastic cups with 200 mL of test
solution and were not led during the exposure period. There were nine tolO replicates for each
treatment and no mortality occurred in the controls. The author reported 96-hour LCso was 557
mg/L and the EPA-calculated 96-hour LCso values was 646.2 mg/L (95% C.I. = 588.0 - 704.4
mg/L), which was acceptable for quantitative use. Note, the authors also reported a qualitatively
acceptable test for the same species (Gosner stage 40) that is described in Appendix G.
A-37
-------
A.2.12 Thirteenth most acutely sensitive genus - Lepomis
The DuPont Haskell Laboratory (2000) evaluated the acute toxicity of ammonium
perfluorooctanoate (APFO, 99% purity) to the bluegill sunfish, Lepomis macrochirus. The static
unmeasured GLP study exposed 2.1 cm fish for 96-hours (dilution water not identified). Fish
used in this study were not fed approximately 24-hours prior to and during the test. Bluegill
sunfish were assigned to the test chambers using random numbers Nominal APFO
concentrations were 262, 328, 410, 512, 640, 800 and 1,000 mg I. (ilass aquaria (20L)
containing 10 L of test solution were employed. Positions of test chambers in the water bath used
to maintain constant temperature were assigned using random numbers. Ten iish were added to
each replicate using random numbers (2 replicates per concentration; total 20 fish per
concentration). A photoperiod of 16-hours light (3 12-344 lu\) \ ersus eight-h darkness was
employed with 25 minutes of transitional light ( 2 I 5 lu\) preceding and following the 16-hour
light interval. Observations for mortality and behavioral effects were made daily. All chemical
and physical parameters were within expected ranges. Total alkalinity and EDTA total hardness
of the dilution water control were 7^ mg I. CaCOi and 76 mg/L CaC03, respectively. During the
test, dissoK ed oxygen concentrations ranged from 6.7-8.5 mg/L, pH ranged from 6.9-7.4, and
temperature ranged from 21 4-22 I (' No fish died in the controls. The authors reported a 96-
hour LCso of 034 mg/L API O EPA performed C-R analysis for the test and had an acceptable
curve with an EPA-calculated LC50 of 664.0 mg/L (95% C.I. = 631.4 - 696.7 mg/L), which was
determined to be quantitatively acceptable for criterion derivation.
A.2.13 Fourteenth most acutely sensitive genus -Physella
Li (2009) conducted a 96-hour static unmeasured acute test on PFOA (ammonium salt,
>98% purity) with the snail species, Physella acuta (Note: formerly called Physa acuta). The test
organisms were collected from a ditch located in Shilin of Taipei City in June 2005. Snails were
A-38
-------
fed with lettuce and half of the culture medium was changed with dechlorinated water every two
weeks, implying a holding time of greater than two weeks. Snails of mixed ages (shell length 0.6
± 0.2 cm) were used at test initiation. The dilution water was dechlorinated tap water, and a
primary stock solution of PFOA was prepared in the same dilution water. The photoperiod
consisted of 12-hours of illumination at an unreported intensity. Exposure vessels were made of
polyethylene with unreported dimensions and 1 L fill volume The test employed five replicates
of six snails each in at least five test concentrations plus a negative control Nominal test
concentrations were in the range of 100-1,000 mg/L PFOA. The test temperature was maintained
at 25 ± 2°C. Water quality parameters including pH, conductivity, and D.O. were reported as
having been measured at the beginning and end of each test, but the information is not reported.
Organisms were not fed, and no animals died in the control groups The author-reported 96-hour
LCso was 672 mg/L (95% C.I.: 635-711 mu I.(which was a\ eraued across three tests. EPA
performed C-R analysis lor each individual test. All three tests had acceptable curves with EPA-
calculatedLCsos of 7(C 1) mu I. (^5".. (' I 706.1 - 817.9 mg/L), 659.9 mg/L (95% C.I. = 607.9
- 711.8 mg/L), and (->28 .1 111 u I. (^5"o CI 582 9 - 673.7 mg/L), which were acceptable for
quantitati\ e use.
A.2.14 Fifteenth most acutely sensili\ e uenus - Ambystoma
Tornuhcne et al. (2021) conducted acute toxicity tests with three species of salamanders
in the genus Ambysioma and I'l-'OA (purchased from Sigma Aldrich, Catalog # 171468-25G;
purity not provided). Acute tests followed standard 96-hour acute toxicity test guidance (U.S.
EPA 2002; ASTM 2008, 2017). The three test species (Jefferson salamander, Ambystoma
jeffersonianum\ small-mouthed salamander, A. texanum\ eastern tiger salamander, A. tigrinum)
were collected from the field in the wetlands of Indiana near the campus of Purdue University.
Collected egg masses were raised outdoors in 200 L polyethylene tanks filled with well water.
A-39
-------
Experiments began when salamanders reached Harrison stage 40, defined as when larvae are free
swimming and feeding. Before test initiation larvae were acclimated to test conditions (21°C and
12-hour: 12-hour light:dark photoperiod) for 24 hours. An additional acute test with Harrison
stage 45 small-mouthed salamanders was run to determine if toxicity varied between life stages.
A stock solution of PFOA (2,000 mg/L) was made in UV-filtered well water and diluted with
well water to reach test concentrations (ranged from 0-2,000 mu I. PI-OA). Test concentrations
were not measured in test solutions based on previous research that demonstrated limited
degradation under similar conditions. Larva were transferred individually to 25<) niL plastic cups
with 200 mL of test solution and were not fed during the exposure period. The number of
replicates varied by species, lifestage and treatment; five replicates per treatment for Jefferson
salamander and Harrison stage 45 small-mouthed salamander. Ii\ e lo seven replicates per
treatment for Harrison stage 40 small-mouthed salamander, and 20 replicates in the control and
10 replicates in each treatment lor eastern tiger salamander Only one salamander larva died in
the controls across all tests (eastern tiger salamander test). Acute values from the four tests
include:
• Jefferson salamander The author-reported 96-hour LCso was 1,070 mg/L. EPA
was unable to lit a C-R model with significant parameters and relied on the
author reported value as quantitatively acceptable.
• I Inn •ison stage 40 small-mouthed salamander: The author-reported 96-hour
LCso was 474 mg/L. The EPA-calculated LC50 was 407.3 mg/L (95% C.I. =
303.7 - 0.510.9 mg/L), which was acceptable for quantitative use.
• Harrison stage 45 small-mouthed salamander: The author-reported 96-hour
LCso was 1,000 mg/L. EPA was unable to fit a C-R model with significant
A-40
-------
parameters and relied on the author-reported value as quantitatively acceptable;
however, the LCso from this test was more than two times greater than the
Harrison stage 40 small-mouthed salamander indicating the Harrison stage 45
was a relatively tolerant life stage. As a result, the LCso from this test was not
used in the SMAV calculation for A. texanum.
• Eastern tiger salamander: The author-reported lH->-hour LCso was 752 mg/L.
Concentration-response data from this test lacked partial effects and EPA was
unable to fit a C-R model with significant parameters and relied on the author
reported value as quantitatively acceptable
A.2.15 Sixteenth most acutely sensitive genus - A naxyrus
Tornabene et al. (2021) conducted acute PFOA (purchased from Sigma Aldrich, Catalog
# 171468-25G; purity not provided) toxicity tests with the American toad, Anaxyrus americanus.
The acute tests followed standard %-hour acute toxicity lest guidance (U.S. EPA 2002; ASTM
2008, 2017). The load euu masses were collected from the field in the wetlands of Indiana near
the campus of Purdue I ni\ eisily Collected euu masses were raised outdoors in 200 L
polyethylene tanks 111 led with well water. Experiments began when frogs reached Gosner stage
26, dellned as when larvae are free swimming and feeding. An additional acute test with Gosner
stage 41 was conducted to determine if toxicity varied between life stages. Before test initiation
larvae were acclimated to lest conditions (21°C and 12-hour: 12-hour light:dark photoperiod) for
24 hours. A stock solution of PFOA (2,000 mg/L) was made in UV-filtered well water and
diluted with well water to reach test concentrations (ranged from 0-2,000 mg/L PFOA). Test
concentrations were not measured in test solutions based on previous research that demonstrated
limited degradation under similar conditions. Larva were transferred individually to 250 mL
plastic cups with 200 mL of test solution and were not fed during the exposure period. The
A-41
-------
number of replicates varied by treatment for both tests; 10 replicates for each all treatments
except the 1,750 mg/L PFOA exposure which had nine replicates. No mortality occurred in any
of the control groups. The authors did not find a significant difference between the life stages of
the American toad, so results of the two tests were pooled to determine the 96-hour author-
reported LC5oof711 mg/L. The EPA-calculated 96-hour LCso value was781.4 mg/L (95% C.I. =
748.3 - 814.4 mg/L) for the 26 Gosner stage test and was 806 6 (l)5"n C T = 760.6 - 852.6 mg/L)
mg/L for the 41 Gosner stage test, both of which were quantitatively acceptable for use.
A.2.16 Seventeenth most acutely sensitive genus - Lithobates
Flynn et al. (2019) evaluated the acute effects of PFOA (CAS# 335-07-1. purchased
from Sigma-Aldrich) on the American bullfrog (Lilhobaics caicsbeiana, formerly, Rana
catesbeiana) during a 96-hour static unmeasured study. Testing full owed Purdue University's
Institutional Animal Care and Use Committee Guidelines Protocol I (->010013551. American
bullfrog eggs were taken from a permanent pond in the Martell Forest outside of West Lafayette,
Indiana. The eggs from a single egg mass were acclimated in 100 L outdoor tanks filled with 70
L of aged well water and co\ ered with a 7<)'\> shade cloth. Once hatched, tadpoles were fed
rabbit chow and Teti aMin ml libimm and were acclimated to laboratory conditions for 24 hours
before testing (21°C and a 12-hour 12-hour light:dark photoperiod). A 2,000 mg/L PFOA stock
solution was prepared with re\ erse osmosis water to produce 12 nominal test concentrations of
PFOA [0 (control). I<). I dm. 250, 500,750, 1,000, 1,250, 1,500, 1,750, 2,000 and 2,500 mg/L],
Each test treatment contained 10 replicates with one Gosner Stage 25 tadpole in each 250 mL
plastic tub. Mortality was monitored twice daily. The author reported a 96-hour LCso value of
1,004 mg/L PFOA. EPA performed C-R analysis for the test and the EPA-calculated 96-hour
LCso was 1,006 mg/L (95% C.I. = 992.8 - 1,018 mg/L).
A-42
-------
Tornabene et al. (2021) conducted acute PFOA (purchased from Sigma Aldrich, Catalog
# 171468-25G; purity not provided) toxicity tests with four species of frogs in the genus
Lithobates (formerly, Rana). Acute tests followed standard 96-hour guidance (U.S. EPA 2002;
ASTM 2008, 2017). The four test species (American bullfrog, Lithobates catesbeiana; green
frog, L. clamitans; northern leopard frog, L. pipiens; wood frog, L. sylvatica) were collected from
a field in the wetlands of Indiana near the campus of Purdue Vn'w ersily Collected egg masses
were raised outdoors in 200 L polyethylene tanks filled with well water I l\pci iinents began
when frogs reached Gosner stage 26, defined as when larvae are free swimming and feeding.
Before test initiation larvae were acclimated to test conditions (21 °C and 12-hour 12-hour
light:dark photoperiod) for 24 hours. A stock solution of PFOA (2,000 mg/L) was made in UV-
filtered well water and diluted with well water to reach test concentrations (ranged from 0-2,000
mg/L PFOA). Test concentrations were not measured in test solutions based on previous research
that demonstrated limited degradation under similar conditions. Larva were transferred
individually to 250 ml. plastic cups with 2<)() mL of test solution and were not fed during the
exposure period. The number of replicates \ aried by species and treatment; 30 replicates in the
control and H\ e to 2d replicates in each treatment for American bullfrog, 10 replicates for each
treatment lor green frog, northern leopard frog and wood frog. No mortality occurred in any of
the control groups Acute allies from the four tests include:
• American bullfrog The author-reported 96-hour LCso was 1,060 mg/L. The
EPA-calculated LCso was 1,035 mg/L (95% C.F = 1,020 - 1,049 mg/L), which
was acceptable for quantitative use.
A-43
-------
• Green frog: The author-reported 96-hour LCso was 1,070 mg/L. EPA was unable
to fit a C-R model with significant parameters and relied on the author reported
value as quantitatively acceptable.
• Northern leopard frog: The author-reported 96-hour LCso was 752 mg/L. The
EPA-calculated LCso was 751.7 mg/L (95% C.I. = 713 ri 790.5 mg/L), which
was acceptable for quantitative use.
• Wood frog: The author-reported 96-hour LCso was 99l) mu I. I-PA was unable
to fit a C-R model with significant parameters and relied on the author reported
value as quantitatively acceptable
A.2.17 Eighteenth most acutely sensitive genus - Oncorhyiiclms
The acute effects of ammonium periluorooctanoate (API (). 99 4% purity) to
Oncorhynchus mykiss was investigated by researchers al the DiiPont Haskell Laboratory
(2000). The static measured Cil.l' study exposed 2 S cm fish for 96-hours (dilution water not
identified). Rainbow trout used in this study were not fed approximately 29-hours prior to and
during the test Rainbow trout were assigned to the test chambers using random numbers.
Addition offish to the test solutions was initiated approximately 41 minutes after test solution
mixing was completed. Mean measured concentrations of ammonium perfluorooctanoate were
554, 1,090, 2.2X0. 4.560 and lU60 for the 625, 1,250, 2,500, 5,000, and 10,000 mg/L nominal
dose levels, respecti\ ely (measured directly by high performance liquid chromatography/tandem
mass spectrometry). Control solutions showed no detectable concentrations of ammonium
perfluorooctanoate. All test substance solutions were clear and colorless with no insoluble test
substance present during the test. Test chambers were stainless steel aquaria that held
approximately 9 L of test solution. Two replicate test chambers were used per test concentration
A-44
-------
with 10 fish in each chamber (total of 20 fish per concentration). Each chamber was covered
with a glass plate to prevent fish from escaping. Mortality and behavioral observations were
made at test start, every 24-hours thereafter, and at test end. All chemical and physical
parameters for the definitive test were within expected ranges. Total alkalinity and EDTA total
hardness of the dilution water control were 49 mg/L CaCCb and 122 mg/L CaCCte, respectively.
During the test, dissolved oxygen concentrations ranged from 7 5-11 2 mg/L, pH ranged from
7.1-7.2, and mean temperature was 11.8°C (range 11.6-12 I (') A photopcriod of 16-hours light
(approximately 199-450 lux) and eight-hours darkness was employed, which included 30
minutes of transitional light (11-157 lux) preceding and following the 16-hour light interval. The
authors reported a 96-hour LCso of 4,001 mg/L APFO and no mortality or sublethal effects were
observed at concentrations <2,500 mg/L API O This study was classified as quantitatively
acceptable for use in criterion derivation.
Colombo ol ;il. (2008) also evaluated the acute toxicity of ammonium perfluorooctanoate
(APFO, CAS # 3N25-20-1. 7" pin i tv) to O. mykiss. Authors stated that the unmeasured static
96-hour GI.P test followed OIX'I) test guideline 203 and EU Commission Directive 92/69/EEC.
APFO stock solutions were prepared In dissolving the test substance directly in the dilution
water and then diluting the stock solution to provide a geometric series of test concentrations
(nominal concentrations of 3 I 3, 62.5, 125, 250, 500 and 1,000 mg/L APFO). Analyses to
confirm the APFO test concentrations were not performed during the acute tests based on the
known stability of the test substance in water, but tests concentrations were measured in the
chronic test. One replicate test chamber containing seven fish was used for the control and each
test solution concentration. Test organisms were randomly assigned to the test solutions after a
pre-test acclimation period of 12 days. Test organisms loading was 0.76 g/L during the study
A-45
-------
with juvenile fish length ranging from 40 to 50 mm. Dilution water was filtered, dechlorinated
tap water that was treated by a softening system to obtain the desired total hardness of 150 ± 20
mg/1 as CaCCte, and pH of 6.0-8.5. A light:dark cycle of 16-hours:8-hours, a temperature of 13-
17 ± PC, and dissolved oxygen greater than 60% saturation were used for acclimation and
testing. Fish were fed trout chow twice daily during the acclimation period, but were not fed
during the 24-hour pre-test period or during acute testing. Fish were observed for mortality and
visible abnormalities at 0, 2, 4, 24, 48, 72 and 96-hours. The study reported a 06-hour LCso of
707 mg/L as PFO". The authors note the contribution of ammonia from API () exposure indicates
that un-ionized ammonia could be a potential contributor to the observed acute toxicity of APFO.
EPA does not believe ammonia was a significant contributor of toxicity (see section C.2.8).
Furthermore, Table 7 of Colombo et al. (2<)i)X) reports unionized ammonia at the calculated
APFO [chronic] NOEC of 40 mg/L to be O.o| 3 mu I. mi-ionized ammonia, nearly half the
unionized ammonia concentration described in the suppoi ti\e text (i.e., 0.021 mg/L). Therefore,
this acute rainbow trout toxicity test is classified as quantitatively acceptable.
A-46
-------
Appendix B Acceptable Estuarine/Marine Acute PFOA Toxicity Studies
B.l Summary Table of Acceptable Quantitative Estuarine/Marine Acute PFOA Toxicity Studies
Species (lifcshiiio)
Method-'
Tesl
Dui'iilioii
( hemiciil /
PuriU
pll
l oin p.
<°C)
S;ilini(\
(ppl)
IITccl
Author
Keporied
IITccl
(one.
(ni}»/l.)
IP A
( iilculiiled
IITccl
(one.
(Ill!i/I.)
l-'iiiiil I ITccl
(one.
(ill"/!.)'
Species
Mc;in
Aciilc
\ iilue
(iiiii/l.)
Reference
Purple sea urchin (embryo),
Strongylocentrotus purpuratus
S,M
96 hours
PFOA
95%
15
30
i:( 50
(normal
development)
I<>
:i) In
20. (i3
:o(..
1 l;i> man el
al. 2U21
Mediterranean mussel (larva),
Mytilus galloprovincialis
S,U
48 hours
PFOA
Unreported
7.9-
8.1
16
36
i:( 50
(mallbrmalion)
>1
-
>lc
-
Fabbri et al.
2014
Mediterranean mussel
(embryo),
Mytilus galloprovincialis
S,M
48 hours
PFOA
95%
-
15
.0
i:( 50
(normal and
surviving)
9.98
17.58
17.58
17.58
Hayman et
al. 2021
Mysid (3-days old),
Americamysis bahia
S,M
96 hours
PFOA
95%
-
:u
}()
i.(
24
-
24
24
Hayman et
al. 2021
Myside (neonate, <24 hours),
Siriella armata
S,U
96 hours
H<)\
-
:u
-
LC50
15.5
-
15.5
15.5
Mhadhbi et
al. 2012
a S=static, R=renewal, F=flow-through, U=unmeasurcd. \1 measured. I inial. I) dissnhcd. Dicl=dietary, MT=maternal transfer
b Values in bold used in the SMAV calculation
B-l
-------
B.2 Detailed Study Summaries of Acute Saltwater PFOA Toxicity
Studies Considered for Use in Saltwater Criterion Derivation
The purpose of this section is to present detailed study summaries for acute
estuarine/marine tests that were considered quantitatively acceptable for criterion derivation,
with summaries grouped and ordered by genus sensitivity. Unlike Appendix A.2 and Appendix
C.2, EPA-calculated C-R models were not presented below for the lour most sensitive
estuarine/marine genera because an estuarine/marine criterion was not de\ eloped exclusively
based on these empirical data. Rather, an estuarine/marine benchmark was deri\ ed using a NAM,
which is further described in Appendix L.
B.2.1 Most acutely sensitive estuarine/marine genera - Siric/hi (mysid)
Mhadhbi et al. (2012) performed a 96-hour static, unmeasured acute test with PFOA
(96% purity) on the mysid, Siriella armani A slock solution of PFOA was made either with
filtered sea water from the Ria of Vigo (Iberian Peninsula) for low exposure concentrations, or
with DMSO for high PFOA concentrations (a final maximum DMSO concentration of 0.01%
(v/v) in the test medium) I lowe\ er. the authors do not indicate what is considered a high test
concentration, so it's unclear which test concentrations actually used DMSO as a solvent. If
DMSO was used, a sol\ ent control was also included. Mysids were exposed to one of ten
nominal PI-'OA treatments (<> 1. 0.5, 1, 2, 5, 10, 20, 30, 40 and 80 mg/L). Mysids were also
collected from the same source as the dilution water and quarantined before use in 100 L plastic
tanks with circulating sand-filtered seawater. The adult stock was fed daily and maintained at
laboratory conditions (17-18°C, salinity between 34.4-35.9 ppt, and oxygen 6 mg/L). Twenty
neonates (<24-hours old) were used per each treatment. To prevent cannibalism, a single
individual was added to each glass vial with 2-4 mL of test solution. Vials were incubated at
20°C with a 16-hour light period. Neonates were fed 10-15 Artemia salina nauplii daily and
B-2
-------
mortality was recorded after 96 hours. The 96-hour LCso reported in the study was 15.5 mg/L
PFOA and was acceptable for quantitative use.
B.2.2 Second most acutely sensitive estuarine/marine genus -Mytilus (mussel)
The acute toxicity of perfluorooctanoic acid (PFOA, purity not provided) on the
Mediterranean mussel, Mytilus galloprovincialis, which occurs in California and other parts of
the Pacific Northwest (Green 2014), was evaluated by Fabbri ef al. (2014). Sexually mature
mussels were purchased from an aquaculture farm in the Ligiiiian Sea (I .a Spezia, Italy) and held
for two days for gamete collection. Gametes were held in artificial sea water ( ASW) made of
analytical grade salts and at a constant temperature of 16 ± 1 °C. It is assumed thai the gametes
were held at the same environmental conditions as the adults, so test salinity was assumed to be
36 ppt with a pH of 7.9-8.1. Embryos were transferred to 90-well microplates with a minimum of
40 embryos/well. Each treatment had six replicates I jnlnyos were incubated with a 16-hour:8-
hour light:dark photoperiod and exposed to one of six: nominal PFOA concentrations (0.00001,
0.0001, 0.001, () i) |. 0 |. | niu |.) or controls. The PFOA stock was made with ethanol, and ASW
control samples run in parallel This included ethanol at the maximal final concentration of
0,01°.. l-acli experiment was repeated four times. At test termination (48 hours), the endpoint
was ihe percentage of normal l)-lar\ ae in each well, including malformed larvae and pre-D
stages. The acceptability of test results was based on controls for a percentage of normal D-shell
stage larvae of 75".. ( \ST\1 2004). Authors noted that controls had >80% normal D-larvae
across all tests. PFOA was only measured once in one treatment which was similar to the
nominal concentration, 0.000081 mg/L versus the nominal concentration of 0.0001 mg/L. PFOA
was below the limit of detection in the control ASW (0.04 ng/L). The percentage of normal D-
larva decreased with increasing test concentrations. The NOEC and LOEC reported for the study
were 0.00001 and 0.0001 mg/L, respectively. However, the test concentrations failed to elicit
B-3
-------
50% malformations in the highest test concentration, and an EC so was not determined. Therefore,
the ECso for the study was greater than the highest test concentration (1 mg/L). The 48-hour EC50
based on malformation of >1 mg/L was acceptable for quantitative use.
Hayman et al. (2021) report the results of a 48-hour static, measured test on the effects
of PFOA (CAS # 335-67-1, 95% purity, purchased from Sigma-Aldrich, St. Louis, MO) on the
Mediterranean mussel, Mytilus galloprovincialis. Authors note thai the tests followed U.S. EPA
(1995b) and ASTM (2004) protocols. Mussels were collected in the field (Sand Diego Bay, CA)
and conditioned in a flow-through system at 15°C. Mussels were induced to spaw 11 by heat-shock
and approximately 250 embryos (2-cell stage) were added to 2<~> mL borosilicate glass
scintillation vials with 10 mL of test solution. There were five replicates per test concentration.
Test conditions were 30 ppt, 15°C and a I (-"-hour S-hour light dark pholoperiod. Six test solutions
were made in 0.45 |im filtered seawater (North San Diego liny. C.\) with PFOA dissolved in
methanol. The highest concentration of methanol was <>.<>2% (\ \). Measured test concentrations
ranged from 1.5-52 nig I. Controls were made in the same seawater and the acute test also
included a solvent control At test termination (4S hours), larvae were enumerated for total
number of lar\ ae that w ere ali\ e at the end of the test (normally or abnormally developed) as
well as number of normally-developed (in the prodissoconch "D-shaped" stage) larvae. There
were no signillcant differences between solvent control and filtered seawater, suggesting no
adverse effects of methanol The author reported 48-hour EC50, based on normal survival larvae
was 9.98 mg/L PFOA. The EPA-calculated 48-hour EC50 value was 17.58 mg/L (95% C.I. =
13.73 - 21.43 mg/L), which was acceptable for quantitative use.
B.2.3 Third most acutely sensitive estuarine/marine genus - Stronsvlocentrotus (sea urchin)
Hayman et al. (2021) report the results of a 96-hour static, measured test on the effects
of PFOA (CAS # 335-67-1, 95% purity, purchased from Sigma-Aldrich, St. Louis, MO) on the
B-4
-------
purple sea urchin, Strongylocentrotuspurpuratus. Authors noted tests followed U.S. EPA
(1995b) and ASTM (2004) protocols. Sea urchins were collected in the field (Sand Diego Bay,
CA) and conditioned in a flow-through system at 15°C. They were induced to spawn by KC1
injection and approximately 250 embryos (2-cell stage) were added to 20 mL borosilicate glass
scintillation vials with 10 mL of test solution. There were five replicates per test concentration.
Test conditions were 30 ppt, 15°C and a 16-hour:8-hour light dark photoperiod. Six test solutions
were made in 0.45 |im filtered seawater (North San Diego Bay, CA) with PI OA dissolved in
methanol. The highest concentration of methanol as n.02% (v/v). Measured test concentrations
ranged from 1.5-52 mg/L. Controls were made in the same seawater and the acute lest also
included a solvent control. At test termination (96 hours), the first 100 larvae were enumerated
and observed for normal development (four-anil pluteus stage) There were no significant
differences between solvent control and filtered seawater. suggesting no adverse effects of
methanol. The author reported lH->-liour ECso, based on normal development, was 19 mg/L
PFOA. The EPA-calculated W-hour i:Oo value was 20.63 mg/L (95% C.I. = 19.74-21.52
mg/L), which was acceptable for quantitati\e use
B.2.4 fourth most acutely sensili\c cstuarine/marine genus - Americamysis (mysid)
I layman et al. (2021) conducted a 96-hour static, measured test to assess effects of
PFOA (CAS 335-07-1, 95".. purity, purchased from Sigma-Aldrich, St. Louis, MO) on the
mysid, Americamysis bahm Authors noted tests followed U.S. EPA (2002) protocols. Mysids
were purchased from a commercial supplier (Aquatic Research Organisms, Hampton, NH) and
acclimated to test conditions (30 ppt, 20°C and a 16-hour: 8-hour light:dark photoperiod). Six test
solutions were made in 0.45 jam filtered seawater (North San Diego Bay, CA) with PFOA
dissolved in methanol. The highest concentration of methanol was 0.02% (v/v). Measured test
concentrations ranged from 1.1-29 mg/L. The highest test concentration (61.7 mg/L) was
B-5
-------
reported as nominal only because the sample was mistakenly not sent to the lab for verification.
Controls were made in the same seawater and the acute test also included a solvent control. Five
mysids (three-days old) were added to 120 mL polypropylene cups and 100 mL of test solution
with six replicates per treatment. Living mysids were counted and dead organisms were removed
daily. There were no significant differences between solvent control and filtered seawater,
suggesting no adverse effects of methanol. No organisms were found dead in the controls at test
termination. EPA was unable to fit a concentration-response model with significant parameters
and relied on the author-reported 96-hour LCso ol"24 mu I. PI OA as the <.|iianlilali\civ
acceptable acute value.
B-6
-------
Appendix C Acceptable Freshwater Chronic PFOA Toxicity Studies
C.l Summary Table of Acceptable Quantitative Freshwater Chronic PFOA Toxicity Studies
Species (lifestage)
Method3
Test
Duration
Chemical /
Purity
pH
Temp.
(°C)
Chronic Value
Endpoint
Author
Reported
Chronic
Value
(mg/L)
EPA
Calculated
Chronic
Value
(mg/L)
Final
Chronic
Value
(mg/L)c
Species
Mean
Chronic
Value
(mg/L)
Reference
Rotifer
(<2-hours old neonates),
Brachionus calvciflorus
R,Ub
Up to
200
hours
PFOA
96%
20
ECio
(intrinsic rate of natural
increase)
0.3536
0.5015
0.5015
-
Zhang et al.
2013a
Rotifer
(<2-hours old neonates),
Brachionus calvciflorus
R,Ub
4 days
PFOA
96%
-
20
ECio
(intrinsic rate of natural
increase)
2.828
1.166
1.166
0.7647
Zhang et al.
2014b
Cladoceran (6-12 hours old),
Daphnia carinata
RU
21 days
PFOA
95%
-
21
MATC
(average # of offspring per
brood and total # of living
offspring)
0.03162
-
0.03162
0.03162
Logeshwaran
et al. 2021
Cladoceran (STRAUS-clone
5; 6-24 hours old),
Daphnia magna
RM
21 days
APFO
99.7%
-
18-22
ECio
(average # of live young)
29.73
20.61
20.61d
-
Colombo et al.
2008
Cladoceran,
Daphnia magna
RU
21 days
PFOA
Unreported
21
ECio
(# young/starting female)
17.68
7.853
7.853
-
Ji et al. 2008
Cladoceran (<24 hours old),
Daphnia magna
RU
21 days
APFO
>98%
-
20
ECio
(# young/starting female)
17.89
12.89
12.89
-
Li 2010
Cladoceran (<24 hours old),
Daphnia magna
RM
21 days
PFOA
99%
7
22
ECio
(survival)
7.02g
5.458
5.458
-
Yang et al.
2014
Cladoceran (<24 hours old),
Daphnia magna
S,U
21 days
PFOA
98%
-
20
MATC
(growth and reproduction)
0.07155
-
0.07155
-
Lu et al. 2016
Cladoceran (12-24 hours old),
Daphnia magna
RU
21 days
PFOA
Unreported
6-
8.5
20
ECio
(# of offspring)
8.23 lf
8.084f
8.084f
4.330
Yang et al.
2019
Cladoceran (<24 hours old),
Moina macrocopa
RU
7 days
PFOA
Unreported
-
25
ECio
(# young/starting female)
4.419
2.194
2.194
2.194
Ji et al. 2008
Amphipod (2-9 days old),
Hvalella azteca
RM
42 days
PFOA
96%
8.1
25
ECio
(# of juveniles/female)
0.0265
0.147
0.147
0.147
Bartlett et al.
2021
C-l
-------
Spocios (lili'shiiic)
Mclliod'1
Tcs(
Diimlion
( hcmiciil /
Pu ri( \
pll
Temp.
(°C)
Chi'onic N'iiluc
llndpoiiii
Aullior
Kcporicd
Chronic
V;iliic
(in Si/I.)
I.PA
Ciilculiilcd
( lironic
\ iduc
(in Si/I.)
limii
(lironic
V;iliic
(ill"/!.)*
Species
Mc;in
Chronic
Viilnc
(insi/l.)
Reference
\lidue (2-da> old lar\aei.
(Ivronomus di/iilus
k. \1
1 da> s
HO\
'J~"u
(. s-
s -
:oo-
240
i:c
(survival)
s
ss
SS.32
ss
\1c( ai'llis el
al. 2u21 '
Rainbow trout
(embryo-larval-juvenile),
Oncorhynchus mykiss
F, M
85 days
(ELS)
APFO
99.7%
6.0-
8.5
11.1-
14.4
LOEC
(grow ill and mortality)
40
-
>40d
>40
Colombo et al.
2008
Rare minnow (adult),
Gobiocypris rarus
F, U
28 days
PFOA
98%
-
25
I.OEC
(survival)
>30
-
>30
>30
Wei et al. 2007
Fathead minnow (<18 hpf),
Pimephales promelas
R, M
21 days
PFOA
96%
7.4-
7.8
25
LOEC
(mortality and growth)
>76
-
>76
>76
Bartlett et al.
2021
Medaka
(adult-FO, embryo-Fl, F2),
Oryzias latipes
F, M
259 days6
PFOA
Unreported
7.5
25
MATC
(F2: sac-Try survival; l'O. l-'l.
1'2: lecundily)
9.487
-
9.487
9.487
Lee et al. 2017
American bullfrog
(tadpole, Gosner stage 25),
Lithobates catesbeiana
(formerly, Rana catesbeiana)
R, U
72 da> s
HO\
I ureporied
-
:i
LOEC
(snout vent length)
0.288
-
0.288
0.288
Flynn et al.
2019
a S=static, R=renewal, F=flow-through, U=unmeasuied. \1 measured. I loial. I) dissol\ed. Diel=dietary, MT=maternal transfer
b Chemical concentrations made in a side-test rcpieseuiali\e exposure ;md \ on lied siahilily of concentrations of PFOA in the range of concentrations tested under similar
conditions. Daily renewal of test solutions.
c Values in bold used in SMCV calculation.
d Concentration of APFO determined as llie anion < 1*1 ¦'()-)
e Total exposure period across FO, Fl. and I'2 ueueralioii!,.
-P
Reported in moles, converted to grams based mi a molecular wemlil of 414.07 g/mol.
g Value represents an ECio based on reproduciion
C-2
-------
C.2 Detailed PFOA Chronic Toxicity Study Summaries and
Corresponding Concentration-Response Curves (when calculated)
The purpose of this section is to present detailed study summaries for tests that were
considered quantitatively acceptable for criteria derivation, with summaries grouped and ordered
by genus sensitivity. C-R models developed by EPA that were used to determine chronic toxicity
values used for criterion derivation are also presented. C-R models included here with study
summaries were those for the four most sensitive genera. Tn many cases, authors did not report
concentration-response data in the publication/supplemental materials and or did not provide
concentration-response data upon EPA request. In such cases, EPA did not independently
calculate toxicity values and the author-reported effect concentrations were used to derive the
criterion.
C.2.1 Most chronically sensitive izenus - Hvalella
Bartlett et al. (2021) evaluated the chronic effects of PI OA (CAS# 335-67-1, 96%
purity, solubility in water at 20.000 mg/I.. purchased from Sigma-Aldrich) onHyalella azteca
via a 42-day static-renewal, measured study Methods for this study were adapted from
Borgmann et al (2<)i)7). and organisms were two to nine days old at the test initiation.
Experiments were conducted in standard artificial media with water quality characteristics of 52
to 60 mg I. alkalinity as CaCO;, average specific conductivity of 0.41 mS/cm, dissolved oxygen
of 5.2 to 8.8 mg I., total hardness of 120 to 140 mg/L CaCCte, average pH of 8.1 and average
temperature of 25°C A I no mg/L stock solution was prepared to yield measured test
concentrations of 0 (control), 0.84, 3.3, 8.9, 29 and 97 mg/L PFOA. Two separate tests were
performed with five replicates per concentration and 20 amphipods per replicate in 2-L HDPE
containers filled with 1 L of testing solution, 2.5 mg of ground TetraMin and one piece of 5x5
cm cotton gauze. Test organisms were fed 2.5 mg TetraMin three times a week during weeks one
C-3
-------
and two, 5 mg TetraMin three times a week during weeks three and four, and 5 mg TetraMin five
times a week during weeks five and six. At test termination (day 42), adults were sexed and
weighed, as well as their young counted. The 42-day author-reported LCio value for survival was
23.2 mg/L PFOA. The author-reported ECio values for growth and reproduction were 0.160
mg/L and 0.0265 mg/L, respectively. EPA only performed C-R analysis for the growth and
reproduction-based endpoints for this test, given the apparent tolerance of the survival-based
endpoint. EPA calculated ECio values for the 42-day growth endpoinl (i e . control normalized
wet weight/amphipod) and the 42-day reproduction endpoint (i.e., number of jn\ eniles per
female). The 42-day growth-based ECio of0.48N mu I. (95% C 1. = 0.319 - 0.(->57 mu/L) was not
selected as the primary endpoint from this test because it was more tolerant than the
reproduction-based ECio of 0.147 mg/L (95% CI - 147 n 147 nig'L). The EPA-calculated
ECio was 0.147 mg/L with a corresponding !¦('=.. of <>.91 | muL While the ECio was relatively
uncertain due to a lack of partial effects around the 10°/.. e fleet level, the EC so estimate remained
relatively certain gi\ en the 47" effect observed in the lowest treatment concentration. EC so to
ECio ratios from all <.|iiantilali\ely acceptable chronic concentration-response curves with similar
species (i e . small members (if the suhphvlum Crustacea) and endpoints (i.e., offspring/female)
were e\ aluated to understand the ariability in the ECso:ECio ratios and provide further context
to the reasonableness of the II. azteca ECio estimate. Overall, three quantitatively acceptable
chronic concentration-response curves with similar species/endpoints were available. See Table
C-l below for a description of the individual C-R curves and resultant ECso:ECio ratios from
each C-R curve.
C-4
-------
Table C-l. EC50 to ECio ratios from all quantitatively acceptable chronic concentration-
response curves with species similar to H. azteca (i.e., small members of the subphylum
Crustacea) and with endpoints that were based on reproduction per female.
C'iliilion
Species
Knilpoinl
i:i»a-
C'iik-iihiled
IX ?n (m«i/l.)
i:i»a-
('iilciihiled
IX'mi (m«>/l.)
IX 5„: IX in
Uiitio
Ji et al. 2008
Daphnia
magna
(# young/starting
female)
61.67
7.853
7.853
Li et al. 2008
Daphnia
magna
(# young/starting
female)
40.75
12.89
3.161
Ji et al. 2008
Moina
macrocopa
(# young/starting
female)
12.77
2 194
5.819
ECsoiECio ratios from the three tests with similar species/endpoinls ranged from 3.161 to
7.852 with a geometric mean ration of 5.247. Dividing the H. azteca reproduction-based ECso
(i.e., 0.911 mg/L) by the geometric mean ECsoiECio ratio (i.e.. 5.247) produced an estimated H.
azteca ECio of 0.174 mg/L, which was similar to l-Cio value calculated directly from the//.
azteca C-R curve (i.e., 0.147 mg/L). The I X"i¦ ¦ \ alue calculated directly from the H. azteca C-R
dataset was, therefore, hypothesized to pro\ ide a robust estimate of a 10% reproductive-based
effect concentration despite the lack of partial low-level effects observed along the C-R curve.
The 42-day a\ eiage number of young per female ECio value of 0.147 mg/L (95% C.I. = 0.147 -
0.147 mg I.) calculated from C-R data reported by Bartlett et al. 2021 was retained for
quantitati\ e use
C-5
-------
Publication: Bartlett et al. (2021)
Species: Hyalella azteca
Genus: Hyalella
EPA-Calculated ECio: 0.147 mg/L (95% C.I. = 0.147 - 0.147 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std. Error
t-stat
p-value
b
1.0325
2.1536 e6
479432
1.328 e6
d
1.7400
1.3029 e7
1335511
4.767 e7
e
1.2996
2.5393 e6
511775
1.244 e6
Concentration-Response Model Fit:
Bartlett et al. 2021
Hyalella azteca
Weibull type 1, 3 para
PFOA ( mg/L )
C.2.2 Second most chronically sensitive genus - Lithobates
Flynn et al. (2019) evaluated the chronic effects of PFOA (CAS# 335-67-1, purchased
from Sigma-Aldrich) on the American bullfrog {Lithobates catesbeiana, formerly, Rana
catesbeiana) during a 72-day static-renewal unmeasured exposure. Testing followed Purdue
University's Institutional Animal Care and Use Committee Guidelines Protocol #16010013551.
American bullfrog eggs were taken from a permanent pond in the Martell Forest outside of West
C-6
-------
Lafayette, Indiana. The eggs from a single egg mass were acclimated in 100 L outdoor tanks
filled with 70 L of aged well water and covered with a 70% shade cloth. Once hatched, tadpoles
were fed rabbit chow and TetraMin ad libitum and were acclimated to laboratory conditions for
24 hours before testing (21°C and a 12-hour: 12-hour light:dark photoperiod). A 2,000 mg/L
PFOA stock solution was prepared with RO water to produce three concentrations for the
chronic test (0, 0.144 and 0.288 mg/L). Each chronic test treatment contained 10 tadpoles
(Gosner stage 25), replicated four times, in 15-L plastic tubs filled with I n I. ol'aged UV-
irradiated, filtered well water. Complete water changes were performed e\ cry ihrec to four days,
at which time chemical treatments were reapplied. Each experimental unit was led daily at a
constant rate (10% per capita) based on tadpole wet biomass in the control treatment to assure
that food was not limiting. On day 72 of the experiment, all tadpoles were euthanized, measured
(snout vent length and mass) and staged. The most sensiti\ e chronic endpoint was growth (snout-
vent length), with a 72-day NOI-C and LOEC of <> 144 mu I. and 0.288 mg/L, respectively. EPA
could not independently calculate an I value because there were minimal effects observed
across the limited number of treatment concent rations tested. Consequently, EPA used the LOEC
of 0.2SS mu I. as the chronic \ allie from this chronic test. The LOEC was used preferentially to
the MATC from this test because a -7% reduction in snout-vent length relative to control
responses was observed at the LOEC (i.e., 0.288 mg/L).
Publication I 'lynn et al. (2019)
Species: American bullfrog (Lithobates catesbeiana)
Genus: Lithobates
EPA-Calculated ECio: Not calculable, unable to fit a model with significant parameters
Concentration-Response Model Fit: Not Applicable
C-7
-------
C.2.3 Third most chronically sensitive genus - Daphnia
Logeshwaran et al. (2021) conducted a PFOA (95% purity, purchased from Sigma-
Aldrich Australia) chronic toxicity tests with the cladoceran, Daphnia carinata. In-house
cultures of daphnids were maintained in 2 L glass bottles with 30% natural spring water in
deionized water, 21°C and a 16-hour: 8-hour light:dark photoperiod. The chronic test protocol
followed OECD guidelines (2012). A PFOA stock solution (100 mu I.) was prepared in
deionized water. Cladoceran culture medium was used to prepare llie PI-OA stock and test
solutions. One daphnid (6-12 hours old) was transferred to each 100 ml. polypropylene container
containing 50 mL of the nominal test solution (0, 0.001, 0.01,0.1, 1.0 and 10 mu I. PI ( )A). Each
test treatment was replicated 10 times with test solutions renewed and daphnids fed daily. At test
termination (21 days) test endpoints included sur\ ival, days lo first brood, average offspring in
each brood and total live offspring. No mortality occurred in the controls or lowest test
concentration. Of the three endpoints measured. a\ erage offspring in each brood and total live
offspring were the more sensiti\ e endpoints with 21-day NOEC and LOEC values of 0.01 and
0.1 mg/LPFOA. respecti\el\ N\\ was unable to calculate statistically robust ECioestimates
from C-R models for these endpoints. largely because of the 10X dilution series across five
orders of magnitude The I.OIX's for these endpoints were not selected as the chronic value
because the I .Ol-Cs produced a 29.23% reduction in the average number of offspring per brood
relative to controls and a SlJ% reduction in the total living offspring relative to controls.
Therefore, the MATC (i.e., 0.03162 mg/L) was selected as the quantitatively acceptable chronic
value form this test.
Publication: Logeshwaran et al. (2021)
Species: Cladoceran (Daphnia carinata)
Genus: Daphnia
EPA-Calculated ECio: Not used, unable to fit a statistically robust model
Concentration-Response Model Fit: Not Applicable
C-8
-------
Ji et al. (2008) conducted a chronic life-cycle test on the effects of PFOA (CAS # 335-
67-1, purity was not reported; obtained from Sigma Aldrich, St. Louis, MO, USA) with Daphnia
magna. The test was done under renewal conditions over a 21-day period and test solutions were
not analytically confirmed. Authors stated that the D. magna test followed OECD 211 (1998). I),
magna used for testing were obtained from brood stock cultured ill the Environmental
Toxicology Laboratory at Seoul National University, Korea, l est organisms were less than 24-
hours old at test initiation. Dilution water was moderately hard reconstituted water (total
hardness typically 80-100 mg/L as CaCCte). Experiments were conducted in glass jars of
unspecified size and fill volume. Photoperiod was assumed lo be 16-hours of illumination, the
same conditions as the daphnid cultures used as the source of 1 lie experimental organisms.
Preparation of test solutions was not described The lesl in\ ol\ ed 10 replicates of one daphnid
each in five nominal lesl concentrations plus a negative control. Nominal concentrations were 0
(negative control), 3 125. (•> 25. 12 5. 25. and 50 mg/L and test solutions were renewed three
times per week Test lenipci alnre was 21 I C for the D. magna test. Authors note that the
water quality parameters (pi I. temperature, conductivity, and dissolved oxygen) were measured
after changing the medium, but the information is not reported. Survival of daphnids in the
negative control was 100% The most sensitive endpoint for D. magna reported in the
publication was days to lust brood with a 21-day NOEC of 6.25 mg/L (LOEC = 12.5 mg/L;
MATC = 8.839 mg/L); however, number of young per starting female (an endpoint not reported
in the publication, which only assessed number of young per surviving female) was calculated by
EPA and considered to be a more sensitive endpoint with an EPA-calculated ECio of 7.853 mg/L
C-9
-------
(95% C.I. = 4.253 - 11.45 mg/L). Therefore, the EPA-calculated ECio of 7.853 mg/L PFOA for
D. magna (number of young per starting female) was considered quantitatively acceptable.
Publication: Ji et al. (2008)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated ECi0: 7.853 mg/L (95% C.I. = 4.253 - 11.45 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std. Error
t-stat
p-value
b
-0.4609
0.0633
-7.2762
0.0054
d
83.0500
3.3807
24.5662
0.0001
e
37.5761
6.5642
5.7244
0.0106
Concentration-Response Model Fit:
Ji et al. 2008
Daphnia magna
Weibull type 2, 3 para
PFOA { mg/L )
Li (2010) conducted an unmeasured chronic life cycle 21-day test on the effects of PFOA
(ammonium salt, >98% purity) on Daphnia magna. Authors stated that the test followed OECD
211 (1998). D. magna used for the test were maintained in the laboratory for more than one year
and were less than 24-hours old at test initiation. Dilution water was distilled water with ASTM
C-10
-------
medium salts added (0.12 g/L CaS04.2H20, 0.12 g/L MgS04, 0.192 g/L NaHCOs, and 0.008 g/L
KC1). The calculated total hardness was 169 mg/L as CaCCte. The photoperiod had 16-hours of
illumination with an unreported light intensity. A primary stock solution (1,000 mg/L) was
prepared in ASTM medium. Exposure vessels were 50 mL polypropylene culture tubes with 50
mL fill volume. The test involved 10 replicates of one daphnid each in five nominal test
concentrations plus a negative control and each test was repeated three times. Nominal
concentrations were 0 (negative control), 1, 3.2, 10, 32, and I'm nig/1. Test temperature was
maintained at 20 ± 1°C. Water quality parameters measured in test solutions were not reported.
Survival of daphnids in the negative control u as 7" () across all three tests. The / K magna 21-
day NOEC (reproduction as number of young per female, broods per female, and mean brood
size) was 10 mg/L (LOEC = 32 mg/L; calculated MATC = I 7 Sl) nigT.). EPA performed C-R
analysis for each reported endpoint. EPA also re\ alualed all endpoints that were based on
number of survi\ inu females to be based on the number of starting females. This recalculation
was done with the intent to account lor starting females that were unable to contribute to the
population as reproduclion female due to mortality. The most sensitive endpoint with an
acceptable C-R cur\ e was the number of young per starting female with an EPA-calculated ECio
of 12 Kl) mu I. PI 'OA (95% (' I = 8.292 - 17.49 mg/L) and was acceptable for quantitative use.
C-ll
-------
Publication: Li (2010)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated ECio: 12.89 mg/L (95% C.I. = 8.292 - 17.49 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
1.2765
0.2577
4.9540
0.0158
d
145.8089
3.0147
48.3655
1.946 e5
e
57.4122
8.7260
6.5795
0.0071
Concentration-Response Model Fit:
Li 2010
Daphnia magna
Weibull type 1, 3 para
PFOA (mg/L )
Yang et al. (2014) evaluated the chronic 21-day renewal, measured test of PFOA (CAS #
335-67-1, 99% purity) with Daphnia magna, following ASTM E729 (1993). Daphnids used for
the test were donated by the Chinese Research Academy of Environmental Sciences. The
daphnids were less than 24-hours old at test initiation. Dilution water was dechlorinated tap
water (pH, 7.0 ± 0.5; dissolved oxygen, 7.0 ± 0.5 mg/L; total organic carbon, 0.02 mg/L; and
total hardness, 190.0 ±0.1 mg/L as CaCCte). The photoperiod consisted of 12-hours of
illumination at an unreported intensity. A primary stock solution was prepared by dissolving
C-12
-------
PFOA in deionized water and DMSO solvent, and proportionally diluted with dilution water to
prepare the test concentrations. Exposure vessels were 200 mL beakers of unreported material
type containing 100 mL of test solution. The test employed ten replicates of one daphnid each in
six test concentrations plus a negative and solvent control. Nominal concentrations were 0
(negative and solvent controls), 5, 7.5, 11.25, 16.88, 25.31 and 37.97 mg/L and were renewed at
48-hour intervals. Test concentrations were measured in low and high treatments only. The
authors provided mean measured concentrations before and after renewal 4 and 4.49 mg/L
(lowest concentration) and 37.66 and 32.88 mg/L (highest concentration). Analyses of test
solutions were performed using HPLC/MS and ncuali\ e eleetrospray ionization The
concentration of PFOA was calculated from standard air\ es (linear in the concentration range of
1-800 ng/mL), and the average extraction efficiency was in the range of 70-83%. The
concentrations and chromatographic peak areas exhibited a signilicant positive correlation (r =
0.9987, p < 0.01). and the water sample-spiked recovery was w%. The temperature, DO, and pH
were reported as ha\ ing been measured every day during the test, but results are not provided.
Negative control and sol\ enl control sur\ i\ al were 90% and 100%, respectively. The author-
reported / K ma^iia 21 -day I X" i- lor reproduction (total number of spawning) was 7.02 mg/L.
EPA performed ( -R analysis for each reported endpoint. Both chronic survival and reproduction
endpoints resulted in acceptable C-R curves. The EPA-calculated ECio for reproduction as total
number of spawning e\ enls was 6.922 mg/L (95% C.I. = 4.865 - 8.979 mg/L), similar to the
ECio reported by the authors (i.e., 7.02 mg/L). Chronic survival was more sensitive than
reproduction, with an EPA-calculated ECio of 5.458 mg/L PFOA (95% C.I. = 3.172 - 7.743
mg/L). Therefore, the survival based ECio calculated by EPA (i.e., 5.458 mg/L) was acceptable
for quantitative use.
C-13
-------
Publication: Yang et al. (2014)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated ECio: 5.458 mg/L (95% C.I. = 3.172 - 7.743 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
1.2765
0.2577
4.9540
0.0158
d
145.8089
3.0147
48.3655
1.946 e5
0.0158
57.4122
8.7260
6.5795
0.0071
Concentration-Response Model Fit:
Yang et al. 2014
Daphnia magna
Weibull type 2,3 para
PFOA ( mg'l )
Colombo et al. (2008) conducted a 21-day renewal measured chronic test on PFOA
(ammonium salt, CAS # 3825-26-1, 99.7% purity) with the daphnid, Daphnia magna. Authors
stated that the toxicity test was conducted followed OECD test guideline 211. There were 10
replicates for each test treatment containing one neonate, six to 24-hours old, each. Exposure
vessel material and size were not reported but filled with 50 mL of test solution. Stock solutions
of APFO were prepared by dissolving the test substance directly in M4 media. The stock was
diluted with M4 media to make the nominal test concentrations: control, 6.25, 12.5, 25, 50 and
C-14
-------
100 mg/L. Test solutions were analyzed by ion chromatography with electrochemical detection.
Measured concentrations were 60% saturation and temperature was maintained between 18-22°C. Illumination included 16-
hours of light with an unreported intensity. Test solutions were typically renewed every three
days and daphnids were fed daily. Control survival met the minimum survival guidance (80%).
Average number of live young was the most sensitive endpoinl reported by Colombo et al.
(2008), with a NOEC of 20 mg/L. Based on the author-reported ECso for llie average number of
live young, the LOEC was 44.2 mg/L and the MATC was 29.73 mg/L. EPA performed C-R
analysis for each reported endpoint. The most sensitive endpoint with an acceptable C-R. curve
was average number of live young, with an EPA-calculated ECio of 20.61 mg/L PFOA (95% C.I.
= 11.29 - 29.93 mg/L), which was acceptable for quantitative use. Although the D. magna C-R
curve from Colombo et al. (2008) displayed a relatively wide 95% confidence bands, the C-R
curve was retained lor use because the ECio is just beyond the NOEC, where effects quickly
increase from 0% to nearly Although there is a lack of partial effects, there appears to be a
threshold effect that occurs abo\ e the NOI-C
C-15
-------
Publication: Colombo et al. (2008)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated ECio: 20.61 mg/L (95% C.I. = 11.29-29.93 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std. Error
t-stat
p-value
b
3.3397
0.7219
4.6265
0.0190
d
66.9304
2.0761
32.2383
6.559 e5
e
43.8667
0.8789
49.9126
1.771 e5
Concentration-Response Model Fit:
Colombo et al. 2008
Daphnia magna
WabuH type 2, 3 para
PFOA (mg'L )
Lu et al. (2016) evaluated the chronic toxicity of PFOA (CAS# 335-67-1, 98% purity,
purchased from Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) on Daphnia magna
immobilization, growth and reproduction. Reconstituted daphnia culture media was used for both
culturing and test solution preparation as described in OECD Test Guideline 202. D. magna
cultures (originally obtained from the Chinese Center for Disease Control and Prevention
(Beijing, China) were fed with the green algae Scenedesmus obliquits daily, maintained at 20°C
and a light/dark photoperiod of 16-hours/8-hours and the medium renewed three times weekly.
The 21-day chronic test endpoints were assessed by a semi-static unmeasured test according to
C-16
-------
OECD Test Method 211. Neonates (<24-hours old) were exposed to six concentrations of PFOA
[0 (control), 0.032, 0.16, 0.8, 4 and 20 mg/L] maintained at 20 ± 1°C. One test organism was
exposed in a 100 mL glass beaker filled with 45 mL of test solution, and there were 20 replicates
for each exposure concentration. The daphnids were fed lxlO6 cells of Scenedesmus obliquus per
animal per day, and the test solution was renewed every other day. Survival, growth and
reproduction (fecundity) was determined during the 21-day exposure The 21-day growth and
reproductive NOEC and LOEC were 0.032 and 0.16 mg/L PFOA, respectix el\. EPA was unable
to fit a C-R model with significant parameters to the chronic data associated with reproduction
from this test. The EPA-calculated ECio values for mean intrinsic rate of increase (i ) and growth
(as length) were 0.0173 mg/L (95% C.I. = 0.0170 - 0.0177 mg, L) and 0.0124 mg/L (95% C.I. =
0.0048 - 0.0200 mg/L), respectively. Bolli of these ECio values were nearly two times lower
than the NOEC of 0.032 mg/L and four times lower than the LOEC value (i.e., 0.16 mg/L) where
only 15.2%) and 11,^".. reductions in intrinsic rale of natural increase (r) and length were
observed, respectively As a result, the MATC of 0.07155 mg/L for growth and reproduction was
selected as the most appropriate chronic \ alue for quantitative use to in deriving the chronic
water eolumn-hased criterion
Publication: Lu et al (2010)
Species Cladoceran (Paplima magna)
Genus Ikiphnia
EPA-Calculntcd F.Cni Not used, unable to fit a statistically robust model
Concentraiion-Ucsponse Model Fit: Not Applicable
Yang et al. (2019) evaluated the chronic effects of PFOA (CAS# 335-67-1, purchased
from Sigma-Aldrich in St. Louis, MO) on Daphnia magna via a 21-day unmeasured, static-
renewal test that assessed reproductive effects. D. magna cultures were originally obtained from
the Institute of Hydrobiology of Chinese Academy of Science in Wuhan, China. Organisms were
C-17
-------
cultured in Daphnia Culture Medium according to the parameters specified in OECD Guideline
202. Protocol for all testing followed OECD Guideline 211. Cladocerans were cultured in
artificial freshwater maintained at 20 ± 1°C under a 16-hour:8-hour light:dark photoperiod and a
light intensity of 1,000-1,500 lux at the surface of the water. Cultures were fed Scenedesmus
obliquus daily and the water was changed twice weekly. Reported water quality parameters
include total hardness of 140-250 mg/L as CaCCte and pH of 6-8 5 The 21 -day chronic study had
nominal concentrations of 0 (control), 0.0000162, 0.0000244, 0.0000365 and n.0000546 mol/L
(or 0 (control), 6.708, 10.10, 15.11, and 22.61 mg/L given the molecular weight of the form of
PFOAused in the study, CAS # 335-67-1, of 414 <>7 g niol) One neonate (12-24 hours old) was
placed in a 100 mL glass beaker, replicated 10 times, and each container filled with 80 mL of test
solution maintained at 20 ± 1°C and a 16-hour 8-hour 1ight:dark pholoperiod with a light
intensity of 1,000-1,500 lux. D. magna were led \. ob/n/mis and lest solutions were renewed
every 72 hours. Test organisms were counted daily, with any young removed. The reproductive
NOEC andLOEC were <> <)<)<)<) I (->2 and 0 0000244 mol/L, or 6.708 and 10.10 mg/LPFOA,
respectively EPA performed C-R analysis lor the test. The EPA-calculated ECio based on mean
offspring at 21-days as a proportion of the control response was 8.084 mg/L (95% C.I. = 7.830 -
8.334 mg I.) and was used <.|iiantita1i\ ely to derive the draft chronic water column criterion.
C-18
-------
Publication: Yang et al. (2019)
Species: Cladoceran {Daphnia magna)
Genus: Daphnia
EPA-Calculated ECio: 8.084 mg/L (95% C.I. = 7.830 - 8.334 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std. Error
t-stat
p-value
b
-0.9632
0.1065
-9.0420
0.0029
e
19.2161
0.9269
20.7320
0.0002
Concentration-Response Model Fit:
Yang et al. 2019
Daphnia magna
Weibull type 2, 2 para
PFOA (mg/L)
C.2.4 Fourth most chronically sensitive genus - Brachionus
Zhang et al. (2013a) conducted a chronic life-cycle renewal test of PFOA (CAS # 335-
67-1, 96% purity) with Brachionus calyciflorus. The test duration was up to 200-hours in a full-
life cycle test (primary emphasis), and 28 days in a population growth test (secondary emphasis:
only two concentrations plus a control). Test organisms were less than two-hours old at test
initiation. All animals were parthenogenetically-produced offspring of one individual from a
single resting egg collected from a natural lake in Houhai Park (Beijing, China). The rotifers
were cultured in an artificial inorganic medium at 20°C (16-hours:8-hours, light:dark; 3,000 lux)
C-19
-------
for more than six months before toxicity testing to acclimate to the experimental conditions.
Culture medium was an artificial inorganic medium and all toxicity tests were carried out in the
same culture medium and under the same conditions as during culture (i.e., pH, temperature,
illumination). Solvent-free stock solutions of PFOA (1,000 mg/L) were prepared by dissolving
the solid in deionized water via sonication. After mixing, the primary stock was proportionally
diluted with dilution water to prepare the test concentrations. Exposures were carried out in 24-
well cell culture plates (assumed plastic) containing 2 mL of test solution per cell. The test
employed four measured test concentrations plus a negative control. Each treatment consisted of
one replicate plate of 15 rotifers, with one rotiler per cell. Treatments were repeated six times.
Nominal concentrations were 0 (negative control), 0.25, 0.5, 1.0, and 2.0 mg/L. PFOA
concentrations were not measured in the rotiler exposures, but rather, in a side experiment using
HPLC/MS. The side experiment showed that the concentration of lJFOA measured every 8-hours
over a 24-hour period in rotiler medium with green algae incurs minimal change in the
concentration range from n 25 to 2 n mg I. I < i0% survival was observed at 24 hours in the
negative control in the corresponding acute test but survival information is not provided for the
life-cycle test /hang et al (2d I .ni) demonstrated rotifer body size and mictic ratio after 28-days
were relati \ el\ tolerant end points with reported NOECs of > 1.0 mg/L and 2.0 mg/L,
respecti\el\ I .IW performed C-R analysis for the remaining reported endpoints from C-R data
reported in the publication The most sensitive endpoint with an acceptable C-R curve was the
intrinsic rate of natural increase with an EPA-calculated ECio of 0.5015 mg/L PFOA (95% C.I. =
0.1458 - 0.8572 mg/L), which was acceptable for quantitative use. The intrinsic rate of natural
increase (d"1) is a population level endpoint that accounts for births and deaths over time. In this
study, the intrinsic rate of natural increase was defined as the natural log of the lifetime net
C-20
-------
reproductive rate for all individuals within a population (defined here as a PFOA treatment level)
divided by the average generation time of those individuals. The effect associated with intrinsic
rate of natural increase is similar to other chronic apical effects reported by Zhang et al. (2013a).
For example, Zhang et al. (2013a) also reported net reproductive rate and juvenile period which
produced an EPA-calculated ECio value of 0.514 mg/L (95% C.I. = 0.1958 - 0.8329 mg/L).
Zhang et al. (2013a) also reported effects to average juvenile period, u hich was a relatively
tolerant endpoint. Juvenile period decreased with increasing exposure concentration, with the
average juvenile period being about 16% faster than the control responses in the highest
treatment concentration (2.0 mg/L; EPA was unable to fit a statistically-robust C-R model for
this endpoint). Zhang et al. (2013a) reported significant reductions in egg size with an EPA-
calculated ECio = 0.193 (95% C.I. = -0.1 (¦><)(¦> n 5466 mg/L). ho\\e\ er. this endpoint displayed a
relatively poor concentration response relationship and may not be relevant for assessing
population level effects and was. therefore, not selected as the primary effect concentration from
this test. Effects to chronic apical endpoints in this publication and Zhang et al. (2014) generally
appear as a threshold effect from n 25 mu I. to 1 0 mg/L, providing further support for the
endpoint and effect le\ el selected for <.|iianlitative use from Zhang et al. (2013a).
C-21
-------
Publication: Zhang et al. (2013a)
Species: Rotifer (Brachionus calyciflorus)
Genus: Brachionus
EPA-Calculated ECio: 0.5015 mg/L (95% C.I. = 0.1458 - 0.8572 mg/L)
Concentration-Response Model Estimates:
Parameter
Kslimale
Sul. Error
l-slal
p-value
b
-0.6515
0.1231
-5.2930
0.0339
d
0.5058
0.0144
35.0478
0.0008
e
1.8042
0.2240
8.0546
OOI5I
Concentration-Response Model Fit:
Zhang et al, 2013
BracMonus calyciflorus
Weitral type 2; 3 para
i 0.4
w
cd
O
'S3
•S „
e O.j
<
\
0.0!
0.10
PFOA ( mg/L )
/hang et al. (2014h) reports the results of a similar chronic life-cycle test of PFOA
(CAS # 3."?5-07-1. W o pui'ilv) with Brachionus calyciflorus. The full life-cycle test used renewal
conditions for approxi niiilcK four days. B. calyciflorus used for the test were less than two-hours
old at test initiation. All animals were parthenogenetically-produced offspring of one individual
from a single resting egg collected from a natural lake in Houhai Park (Beijing, China). The
rotifers were cultured in an artificial inorganic medium at 20°C (16-hours:8-hours, light:dark;
3000 lux) for more than six months before toxicity testing to acclimate to the experimental
conditions. Culture medium was an artificial inorganic medium and all toxicity tests were carried
C-22
-------
out in the same culture medium and under the same conditions as during culture (i.e., pH,
temperature, illumination). Solvent-free stock solutions of PFOA (1,000 mg/L) were prepared by
dissolving the solid in deionized water via sonication. After mixing, the primary stock was
proportionally diluted with dilution water to prepare the test concentrations. Exposure vessels
and size were not reported for the four-day reproductive assay, but were likely 6-well cell culture
plates (assumed plastic) each containing at total of 10 mL of lesl solution. The test employed
eight test concentrations plus a negative control. Each treatment consisted of six replicates of 10
rotifers each in individual cells. The numbers of living rotifers were counted after four days for
each treatment level. Nominal concentrations were 0 (negative control), 0.125, n 25. 0.50, 1.0,
2.0, 4.0, 8.0, and 16.0 mg/L. PFOA concentrations were not measured in the rotifer exposures,
but rather in a side experiment using HPI.(' MS The side experiment showed that the
concentration of PFOA measured every eight-hours o\ cr a 24-hour period in rotifer medium with
green algae incurs minimal change in the concentration, ranging from 0.25 to 2.0 mg/L. Negative
control survival was not pro\ ided lor the life-cycle test.
Resting egg production is an ecologically important endpoint for this species because it
represents the linal result of sexual reproduction. Based on authors description of results in the
text, "PI-OA exposure signilicantly reduced resting egg production of B. calyciflorus females
during the three-day period "" NOEC and LOEC values were not reported, but 0.25 mg/L PFOA
produced more than a 5<>".. reduction in resting egg production. Therefore, it is assumed The B.
calyciflorus four-day NOEC for resting egg production was 0.125 mg/L and the LOEC was 0.25
mg/L, with a calculated MATC is 0.1768 mg/L. Concentration response data from Figure 1 of
C-23
-------
Zhang et al. (2014b) were estimated (WebPlotDigitizer1) and used to derive an EPA-calculated
ECio of 0.076 mg/L (95% C.I. = 0.054 - 0.098 mg/L), further suggesting resting egg production
may be a relatively sensitive endpoint. Because there was only one replicate (as implied by lack
of error bars in Figure 1 of the publication, no clear description of replicates in the methods
section, and no author-reported statistical analysis of this endpoint). resting egg production from
this study was not considered quantitatively acceptable but was retained lor qualitative use.
Beyond resting egg production, PFOA did not clearly affect hatching rate of resting eggs when
exposed to PFOA during the formation or hatching period, enhanced hatching rale relative to
controls in most treatments (nominal test concentration range <> 2.0 mg/L; see figures 3 and 4
of Zhang et al. 2014b) and displayed no clear concentration-response relationship, suggesting
rotifer hatching rate was a relatively tolerant endpoinl from this publication. In contrast to Zhang
et al. (2013a), which oltser\ ed no effect of PFOA on mictic ratio after 28 days at a nominal
concentration as high as 2 n mg I. PI-OA. Zhang el al (2014b) stated PFOA significantly
increased the F1 mictic ratio from n 5o in llie control treatment to 0.75 and 0.72 in nominal
PFOA test concentrations of <) 25 mg I. and 2.0 mg/L, respectively. Given conflicting results of
PFOA on rotifer mictic ratio, it was not selected as the primary endpoint from Zhang et al.
(2014b). The most sensiti\e quantitatively acceptable endpoint was the intrinsic rate of natural
increase. The intrinsic rate of natural increase (d"1) is a population level endpoint that accounts
for births and deaths over time. In this study, the intrinsic rate of natural increase was defined as
the natural log of the net increase in the number of rotifers (surviving parents and offspring) for
each PFOA treatment level over a four-day exposure period. This endpoint was conceptually
1 WebPlotDigitizer is an online application used to convert values shown in figures to numerical values. This
application was used to obtain numerical concentration-response data when they were only reported in figures. The
application is free and available online (WebPlotDigitizer - Extract data from plots, images, and maps
(automeris.io')').
C-24
-------
equivalent to the intrinsic rate of natural increase endpoint calculated by Zhang et al. (2013 a) but
was a simplification of the calculations performed in Zhang et al. (2013b), in that it only applied
to the four-day observational period, whereas the intrinsic rate of natural increase calculated in
Zhang et al. (2013a) represented the full lifetimes of all individuals within each population (i.e.,
exposure concentration). The EPA-calculated ECio for this endpoint was 1.166 mg/L (95% C.I. =
0.7720- 1.559 mg/L).
Publication: Zhang et al. (2014b)
Species: Rotifer (Brachionus calyciflorus)
Genus: Brachionus
EPA-Calculated ECi0: 1.166 mg/L (95% C.I. = 0.7720 - 1.559 mg/L)
Concentration-Response Model Estimates:
Parameter
Estimate
Std.Error
t-stat
p-value
b
1.1913
0.2118
5.6236
0.0014
d
0.2253
0.0081
27.9366
1.392 e7
e
7.7080
0.8191
9.4102
8.183 e5
Concentration-Response Model Fit:
Zhang et al. 2014b
Brachionus calyciflorus
Weibull t>pe 1, 3 para
PFOA ( mg'L )
C-25
-------
C.2.5 Fifth most chronically sensitive genus -Moina
Ji et al. (2008) conducted a chronic life-cycle test on the effects of PFOA (CAS # 335-
67-1, purity unreported; obtained from Sigma Aldrich, St. Louis, MO, USA) with Moina
macrocopa. Tests were done under renewal conditions over a seven-day period and test solutions
were not analytically confirmed. Authors stated that theM macrocopa test followed a protocol
developed and reported by S.R. Oh (2007) (Master's thesis, Seoul National University, Seoul,
Korea), which is similar to OECD 211 (1998), but with slight modi Ilea lion (i.e., shorter test
duration, exposure temperature and different feeding regime: 100 |iL yea si ccrophyll:tetramin
mixture and 200 |iL algae suspension per day). M macrocopa used for testing were obtained
from brood stock cultured at the Environmental Toxicology I .ahoratory at Seoul National
University, Korea. Test organisms were less llian 24 hours old al lest initiation. Dilution water
was moderately hard reconstituted water (lolal hardness typically K<)-l<>0 mg/L as CaC03).
Experiments were conducted in glass jars of unspecified size and fill volume. Photoperiod for the
test was not reported hut was assumed to he 16-hours of illumination, the same conditions as the
daphnid cultures reported in this same publication. Preparation of test solutions was not
described The lest in\ ol\ ed ten replicates of one individual each in five nominal test
concentrations plus a negati\ e control Nominal concentrations were 0 (negative control), 3.125,
6.25, 12 5. 25. and 50 mg/l. and test solutions were renewed three times per week. Test
temperature w as 25 I (' lor \ I macrocopa. Authors note that the water quality parameters
(pH, temperature, conductivity, and dissolved oxygen) were measured after changing the
medium, but the information was not reported. Survival of daphnids in the negative control was
100%. TheM macrocopa seven-day NOEC (reproduction: number of young per adult) was
3.125 mg/L, the LOEC was 6.25 mg/L and the MATC is 4.419 mg/L. EPA performed C-R
analysis for this study. The most sensitive endpoint with an acceptable C-R curve was the
C-26
-------
number of young per starting female. The EPA-calculated ECio was 2.194 mg/L PFOA (95%
C.I. = -0.7120 - 5.010 mg/L) for M. macrocopa. The lowest treatment concentration produced a
greater than 10% effect which forced the ECio calculation to extrapolate beyond the lowest
treatment concentration (i.e., not the control, but the nominal treatment of 3.25 mg/L). However,
the resultant ECio value (i.e., 2.194 mg/L) was considered acceptable for quantitative use
because it was largely in agreement with the 14.3% effect obser\ cd al the test concentration of
3.125 mg/L.
C.2.6 Sixth most chronically sensitive genus - Oryzias
Lee et al. (2017) conducted a multiple generation exposure to determine the effects of
PFOA (CAS # 335-67-1, purity was not reported) on the reproductive toxicity and metabolic
disturbances to Oryzias latipes. Fish were originally recci\ cd from the Department of Risk
Assessment of the National Institute of En\ ironmcntal Research (MI.R; South Korea) and
maintained according lo the following conditions: dissoK ed oxygen 7-8 mg/L, pH 7.5 ± 0.2,
water temperature 25 I ('. I o-hour light, 8-hour dark photoperiod and total hardness 55-57
mg/L (as CaCOo (I km pes were led .\ncmia salina once daily, based on the OECD test
guideline 24<) feeding schedule Adults (about 13 weeks old, when genders could be visually
differentiated) through the 12 generation were exposed to three nominal concentrations of PFOA
(0.3, 3 and 3d nig I. PFOA) for a total exposure period of 259 days. At test initiation, four pairs
of both genders were introduced into the test chambers (8 L glass tank) of a flow-through
exposure system. PFOA solutions were replenished five times daily to keep the same water
quality as fish maintaining condition. PFOA exposure continued for three weeks, during which
eggs produced by mating of F0 fish were removed from test chamber and counted daily for
fecundity. During test week four, (spawning period), the F1 generation eggs (n = 192) were
obtained per each concentration. Right after the spawning period was finished, F0 fish were used
C-27
-------
to evaluate metabolism disturbance. The F1 eggs were pooled and redistributed into an
incubation chamber containing PFOA solution. After the hatching was completed, the test
organisms were returned into test chambers and raised under flow-through PFOA exposure
conditions until they reached adult stage (at about 13 weeks old), during which sac-fry survival
rate, hatching rate, and abnormality of F1 were analyzed. When F1 fish reached the adult stage,
sex ratio of total F1 fish was determined and 32 individuals ofF I fish were used to analyze
gonadosomatic (GSI), hepatosomatic (HSI), condition factor (K), VT(i expression, and
histological alterations. In addition, other F1 fish (32 male and female fish) were used for
obtaining the F2 generation eggs same as the FO generation The exposure conditions to F2 fish
were carried out in the same manner as F1 fish. Consequently, 10 fish were exposed to PFOA for
four weeks and F1 and F2 fish were exposed to PI-OA across all life cycle stages without
exposure pause. The exposure regime was applied equally in all lest groups. The 259-day MATC
of 9.487 mg/LPI OA was reported lor 1-2 sac-lYy survhal and fecundity for the FO, F1 and F2
generations and represented the most sensitive endpoints from the study. Reproductive responses
reported by T.ee el al (2d I 7) appear lo he control normalized; however, use of control
normalized data in this study does not alter conclusions from hypothesis-based testing (i.e., use
of a M)l-(\ I.OI-C, or MATC) Beyond F2 survival (which had control mortality), EPA
attempted C-R analysis for all endpoints reported by Lee et al. (2017). Given the large dilution
factor between PI OA treatments, C-R models could either not be fit, or when models could be
fit, they performed poorly on statistical metrics and were not used. Therefore, the 259-day
MATC of 9.487 mg/L PFOA was considered to be quantitatively acceptable for criterion
derivation. The large dilution factor from this test does not support concentration response
C-28
-------
modeling, consequently EPA relied on an MATC (i.e., 9.487 mg/L) as the chronic effect level
from this test.
C.2.7 Seventh most chronically sensitive genus - Gobiocyyris
The chronic toxicity of PFOA (98% purity) on the rare minnow (Gobiocypris rarus; not
North American resident species) was investigated by Wei et al. (2007) using flow-through
unmeasured exposure conditions. Two hundred and forty mature mule and female rare minnows
(about nine months old, 1.4 ± 0.4 g, 47.7 ± 3.6 mm) were oblained from a laboratory hatchery
and randomly assigned to eight 20 L glass tanks (30 individuals per tank) I 'ish were supplied
with dechlorinated tap water under continuous flow-illrough conditions at 25 ^ 2 (' and a
photoperiod of 16-hours:8-hours light:dark. During the 2S-da\ exposure period, fish were fed a
commercial granular food (Tetra) at a daily rale of 0.1% hock weight. Waste and uneaten food
were removed daily. After a one-week acclimation period. 30 randomly selected male and 30
female rare minnows (gender determined by observing the shape of the abdomen and the
distance between the abdomen fin and the stern fin) were assigned to one of the four nominal
PFOA exposures (<>. 3. I n or 3d mg I. PI OA) l:.ach treatment was performed in duplicate tanks.
The How rate of the test solution was S I. hour, and actual PFOA concentrations in the tanks
were not \ eri I led by chemical analysis During the exposure period, there were separate inputs
for water and PI OA and the mixer helped mix PFOA and water before flowing into the tanks.
The concentration of mixed solution flowing out from the mixer were kept at 3, 10, or 30 mg/L
PFOA by adjusting the input flow rate of concentrated PFOA and water, respectively. After 14-
day and 28-day exposure periods, fish were anesthetized on ice, and liver samples were taken
and immediately frozen in liquid nitrogen and stored at -80°C until analyzed. No mortality was
observed in any treatments. The 28-day LOEC (survival) was >30 mg/L PFOA and was
C-29
-------
acceptable for use as a high-unbounded value from a high-quality study which provides relevant
sensitivity information for this fish species.
C.2.8 Eighth most chronically sensitive genus - Oncorhynchus
Colombo et al. (2008) evaluated the chronic effects of ammonium perfluorooctanoate
(APFO, CAS #3825-26-1, 99.7% purity) to embryos of the rainbow trout, Oncorhynchus mykiss.
Stock solutions of APFO were prepared by dissolving the test substance directly in the test media
or dilution water and then diluting the stock solution to provide a geometric series of test
concentrations (nominal concentrations of 3.13, 6.25, 12.5, 25 and 50 nig I. API O). The early-
life-stage (ELS) test was performed under flow-illrough conditions and in compliance with
OECD test guideline 210. Unfertilized trout eggs and sperm were received from a commercial
supplier and the eggs were fertilized in the laboratory. One hundred and eighty newly fertilized
eggs were randomly selected and allocated. (•><> eggs per replicate, to the three replicate test
vessels for each control and test concentration Authors stated that the number of surviving fish
was reduced randomly to 3d per replicate just alter the end of the hatching period (day 26) in the
control. The number of sur\ i\ ing fish was again reduced randomly to 15 per replicate when
swim-up and feeding began on day 5<) Actively feeding juveniles were fed trout chow two to
four times per day, corresponding to approximately 4% of their body weight per day, from day
50 to the end of the S5-day test Test solutions were continuously renewed during the study by
pumping the stock solutions into flowing dilution water with a peristaltic pump system at a
replacement rate of 5 7(-> times the test vessel volume per day. Dilution water pH was 6.0-8.5,
total hardness was 150 mg/L as CaC03, and water temperature was kept between 11.1 and
12.5°C for embryos and between 11.6 and 14.4°C for larvae and juvenile fish. The dissolved
oxygen concentration was greater than 60% air saturation, the light/dark cycle was maintained at
constant darkness until seven days after hatching, then 16-hours light and eight-hours dark
C-30
-------
through test end. Observations were made daily as follows: eggs-marked loss of translucency and
change in coloration, white opaque appearance; embryos-absence of body movement or
heartbeat; larvae and juvenile fish-immobility, absence of respiratory movement or heartbeat,
white opaque coloration of the central nervous system, lack of reaction to mechanical stimulus,
and abnormalities. The reported 85-day growth and mortality NOEC was 40 mg/L PFO";
however, the authors' note that the contribution of ammonia from API'O exposure indicates that
un-ionized ammonia could be a potential contributor to the observed toxicity of APFO. Although
the authors cite EPA (1999) for un-ionized ammonia toxicity values, that document (and the
subsequent U.S. EPA [2013] criteria document) expressed toxicity in terms of a relationship
between total ammonia nitrogen and pH and temperature. For rainbow trout, U.S. EPA (1999)
declined to specify a chronic value, due to inconsistencies between tests. However, U.S. EPA
(2013) set the rainbow trout chronic value at (•> (•>(•> mu TAN I. (Total Ammonia Nitrogen/L) at pH
= 7. Using the normalization equations in U.S. N\\ (2013), the rainbow trout chronic value
translated to 3.6i) mu \ I. at the authors' assumed chronic test pH of 7.8 (see table 7 of Colombo
et al. [2008 |). u liich in turn translated to a rainbow trout chronic value as un-ionized ammonia of
0.064 mu un-ionized ammonia I. at pi I 7 S and a reported test of temperature (13°C). Table 7 of
Colombo et al (2008) listed the un-ionized ammonia concentration at their APFO NOEC as
0.013 mg un-ionized ammonia L, which is 4.9-fold lower than EPA's chronic value for rainbow
trout re-expressed as an un-ionized ammonia concentration for the test condition. Therefore,
EPA does not believe ammonia was a confounding factor in this test and the study was
determined to be quantitatively acceptable for criterion derivation.
C.2.9 Ninth most chronically sensitive genus - Pimephales
Bartlett et al. (2021) also evaluated the chronic effects of PFOA (CAS# 335-67-1, 96%
purity, solubility in water at 20,000 mg/L, purchased from Sigma-Aldrich) on fathead minnows
C-31
-------
(Pimephalespromelas) via a 21-day early-life stage static-renewal, measured study. The authors
followed OECD Test Guideline 210, except that the test ended at 16 days post-hatch (dph)
compared to 28 dph for the standard OECD test. Test water (i.e., stock solutions, exposure
solutions and controls) was charcoal-filtered UV-sterilized Burlington City water from Lake
Ontario (total hardness 120-130 mg/L, alkalinity 89-93 mg/L, pH 7.4-7.8), and was maintained
in a header tank prior to use in testing. Fathead minnow eggs ( I S-hour post fertilization) were
purchased from Aquatox Labs (Guelph, ON) and exposed to nine nominal PI ( )A concentrations:
0.01, 0.032, 0.1, 0.32, 1, 3.2, 10, 32, and 100 mg/L. The tests were divided into low
concentration (0.01-10 mg/L) and high concentration (32-1 <~>f) mg/L) tests, with fi\ e days in the
egg stage and 16 days in the larval fish stage. Tests were initiated with eggs from five to ten egg
batches (from different fathead minnow breeding groups) to maximize genetic diversity and
variability. There were 20 eggs per beaker, with eight replicates of controls and four replicates of
each PFOA concentration in each of the two tests I jnbi yos and larvae were held in glass, Nitex
mesh bottomed (mesh size 5<)o mil) egg cups within 800-mL HDPE beakers filled to 700 mL
with test solution lieakers containing fathead minnow eggs/larvae were aerated, loosely covered,
and held in a 25 (' incubator with a photoperiod of 16-hours light:8-hours dark. Larvae were fed
10 |iJ. fish (<) 9 dph) and 2d uL/fish (lM6 dph) of newly hatched brine shrimp slurry per day.
The first feedi ng (hall' of the dally aliquot) was two hours prior to the daily solution changeover
(to remove excess food and waste), and the second feeding (the other half of the daily aliquot)
was after solution changeover, so that food was available at all times during the tests. Endpoints
evaluated were survival to hatch, time to hatch, hatching success, deformities at hatch, uninflated
swim bladder, survival from the egg until nine and 16 dph, and weight, length, tail length, and
condition factor of larvae at nine and 16 dph. The reported 21-day NOEC for mortality, weight,
C-32
-------
length, and condition factor was 76 mg/L PFOA. EPA could not independently calculate an ECio
value because no effects were observed across the range of concentrations tests. Because the
NOEC of 76 mg/L was a relatively tolerant NOEC value it was considered quantitatively
acceptable for criteria derivation.
C.2.10 Tenth most chronically sensitive genus - Chironomus
McCarthy et al. (2021) conducted a 19-day chronic PI'O A (^7% purity, purchased from
Sigma-Aldrich) toxicity test on the midge, Chironomus dilutns The PI-OA stock solution was
dissolved in reconstituted moderately hard water without the use of a sol\ enl and stored in
polyethylene at room temperature until use. Authors reported that they followed standard
protocols (ASTM 2005; U.S. EPA 2000) with slight modifications. Exposure vessels for both
experiments were 1 L high-density polyethylene beakers containing natural-field collected
sediment with 60 mL of sediment and 105 nil. of test solution. PI-OA lest solutions were added
via pipette to the beakers with the tip just above the sediment substrate. Nominal test
concentrations were 0, 26, 87, 1-N. 210 and 272 mg/L PFOA, respectively. Test concentrations
were based on the results of a I <)-day range finding test conducted by McCarthy et al. (2021),
which is further described in Appendix (i l-gg cases were obtained from Aquatic Biosystems or
USGS Columbia I-n\ ironmeiital Research Center and held as free-swimming hatched embryos
(<24 hour after hatch) before testing. Each beaker held 12 organisms with five replicates per
exposure treatment Solutions were renewed every 48 hours. PFOA treatment concentrations
were measured on days I". 15 and 20 in the 20-day exposure. Mean measured PFOA
concentrations in the 20-day exposure were 0 (control), 19.9, 59.4, 145, 172 and 227 mg/L
PFOA. Percent survival in the control treatment was 82%. The most sensitive endpoint was
survival with an author reported 19-day ECio of 89.8 mg/L PFOA. The EPA-calculated survival-
C-33
-------
based ECio was 88.32 mg/L (95% C.I. = 15.40 - 161.3 mg/L), which was acceptable for
quantitative use.
C-34
-------
Appendix D Acceptable Estuarine/Marine Chronic PFOA Toxicity Studies
No data at this time.
D-l
-------
Appendix E Acceptable Freshwater Plant PFOA Toxicity Studies
E. 1 Summary Table of Acceptable Quantitative Freshwater Plant PFOA Toxicity Studies
Species
Method3
Test
Duration
Chemical
/ Purity
PH
Temp.
(°C)
Effect
Reported
Effect
Concentration
(mg/L)
Reference
Green alga,
Chlamvdomonas reinhardtii
S,U
96 hours
PFOA
>96%
6.8
25
ECso
(growth)
51.9
Huetal. 2014
Green alga,
Chlamvdomonas reinhardtii
s,u
8 days
PFOA
>96%
6.8
25
MATC
(cell number)
28.28
Huetal. 2014
Green alga (7.0 x 105 cells/inL),
Chlorella pvrenoidosa
S, M
96 hours
PFOA
>98%
-
25
ECso
(growth)
190.99
Xuetal. 2013
Green alga (9 x 105 cells/inL),
Chlorella pvrenoidosa
S,U
96 hours
PFOA
>95%
-
25
NOEC
(growth)
0.1
Li et al. 2021b
Green alga (1.5 x 104 cells/inL),
Chlorella vulgaris
S,U
96 hours
PFOA
95%
23
ICs,
(cell density)
115.5
Boudreau 2002
Green alga (1.5 x 104 cells/inL),
Raphidocelis subcapitata
(formerly Pseudokirchneriella subcapitata
and Selenastrum capricornutum)
S,U
96 hours
PFOA
95%
23
ICs,
(cell density)
123.4
Boudreau 2002
Green alga (log phase growth),
Raphidocelis subcapitata
S, M
96 hours
APFO
99.7%
-
21-25
MATC
(biomass and growth rate)
16.07
Colombo et al.
2008
Green alga (7.0 x 105 cells/inL),
Raphidocelis subcapitata
S, M
96 hours
PFOA
>98%
25
ECso
(growth)
207.46
Xuetal. 2013
Green alga,
Scenedesmus obliquus
S,U
96 hours
PFOA
>96%
6.8
25
ECso
(growth)
44.0
Huetal. 2014
Green alga,
Scenedesmus oblic/uus
S,U
8 days
PFOA
>96%
6.8
25
NOEC
(cell number)
40
Huetal. 2014
Green alga,
Scenedesmus quadricauda
S, M
96 hours
PFOA
99%
7
22
ECso
(growth inhibition rate)
269.63
Yang et al. 2014
Water milfoil (4 cm apical shoots),
Mvriophvllum sibiricum
S, M
14 days
PFOA
Unreported
8.3-8.7
17.8-
22.0
ECio
(dry weight)
8.7
Hanson et al.
2005
E-l
-------
Species
Method-'
Tesl
l)ii r;i lion
C "hem ic;il
/ I'uriM
I'"
Temp.
<°C)
HITccl
Reported
i:ileel
( oneeiili'iilioii
(niii/l.)
Kel'eivnee
Water milfoil (4 cm apical shoots),
Myriophyllum sibiricum
S,M
21 days
PFOA
Unreported
8.3-8.7
17.8-
22.0
ECio
(dry weight)
7.9
Hanson et al.
2005
Water milfoil (4 cm apical shoots),
Myriophyllum sibiricum
S,M
35 days
PFOA
Unreported
8.3-8.7
17.8-
22.0
EC,0
(wet weight)
21.6
Hanson et al.
2005
Water milfoil (4 cm apical shoots),
Myriophyllum spicatum
S,M
14 days
PFOA
Unreported
8.3-8.7
17.8-
22.0
EC,,,
(dry weight)
18.1
Hanson et al.
2005
Water milfoil (4 cm apical shoots),
Myriophyllum spicatum
S,M
21 days
PFOA
Unreported
8.3 S "
1 7.8-
17 0
ECio
(plant length)
5.7
Hanson et al.
2005
Water milfoil (4 cm apical shoots),
Myriophyllum spicatum
S,M
35 days
PFOA
Unreported
8.3-8."
ra-
il ii
ECio
(dry weight)
19.7
Hanson et al.
2005
Lettuce (seed),
Lactuca sativa
S,U
5 days
PFOA
96°/,
-
-
EC50
(rool elongation)
745.7b
Ding et al.
2012b
a S=static, R=renewal, F=flow-through, U=unmeasured, M=measured, T=total, I) dissol\ ed. \k nol ivpoiied
b Reported in moles converted to milligram based on a molecular weight of 414 t>~ mu niniol
E-2
-------
E.2 Summary of Quantitatively Acceptable Plant PFOA Toxicity
Studies
E.2.1 Green alga. Chlamydomonas reinhardtii
Hu et al. (2014) evaluated the growth inhibition of PFOA (>96% purity) with
Chlamydomonas reinhardtii in 96-hour and eight-day static exposures. Authors stated that the
tests followed OECD test guidance 201 (OECD 2006). Chlamydomonas reinhardtii were
supplied by UTEX Culture Collection of Algae, University of Texas al Austin. Dilution medium
was described as modified high-salt media at a pH of 6.S Algae in exponential growth phase
were exposed to nominal concentrations of 0 (negative control), 1, 3.16, 10, 3 I (v I mi. 316, and
1,000 mg/L in the 96-hour exposure, and 0, 5, 10, 20, and 4<) nig/L in the eight-day exposure.
Experiments were initiated by inoculating equal cell numbers of I x 104 cells/mL in the 96-hour
exposure and 5xl06 cells/mL in the eight-day exposure into 250 ml. Ilasks containing a total
volume of 100 mL of algal cell suspension per llask There were live replicates per each
treatment in the lH->-hour exposure and three replicates in the eight-day exposure. Algae were
incubated at 25' (' under cool-w hite fluorescence lights at 85-90 |imol photons/[m2 x s]
irradiance with a lo-hour S-hour light dark cycle The 96-hour growth ECso (inhibition based on
optical density) was 5 I ^ mg I. The S-day MATC hased on cell number was 28.28 mg/L (NOEC
and 1 .Ol-C are 2d and 40 mg I.. respectively). The plant values from the study were acceptable
for quantitati\e use
E.2.2 Green alga. Chlorellapyrenoidosa
Xu et al. (2013) performed a 96-hour static, measured algal growth inhibition test on
PFOA (>98% purity) with Chlorella pyrenoidosa. Algae were obtained from the Aquatic
Organism Research Institute of the Chinese Academy of Science and precultured for three
generations prior to initiating the test. Dilution medium consisted of number one culture medium
E-3
-------
supplemented with aquatic number four nutrient solution (Zhou and Zhang 1989). Algae in
logarithmic growth phase (7.0xl05 cells/mL) were inoculated in medium containing PFOA at 0
(negative control), 30, 60, 90, 120, 150, 180, 240, 300 and 360 mg/L. Tests were conducted in
100 mL conical flasks with 50 mL of solution with each concentration replicated three times.
Exposure concentrations were verified via UHPLC-MS/MS using the Agilent 1290 Infinity
UHPLC system interfaced with an Agilent 646-0 Triple Quadrupole mass. Algae were exposed
under a 12-hour: 12-hour light:dark cycle at 3,000-4,000 lux and 25UC Chlorophyll concentration
and permeability of cell membranes was determined after lH->-hours of exposure The reported 96-
hour growth ECso (inhibition based on optical density) was 11)<) mg/L and was considered to
be acceptable for quantitative use.
Li et al. (2021b) conducted a 12-day sialic, unmeasured U>\icitv test with PFOA (>95%
purity, purchased from Sigma-Aldrich) on the green alua. ('liloivlfapyrenoidosa. The FACHB-9
strain of the green alua was purchased from the Institute of I lydrobiology, Chinese Academy of
Sciences. The alua was ail lured in IKi-11 medium at 25°C under a 12-hour: 12-hour light:dark
photoperiod (2(")oo hi\) and shaken manually e\ cry 12 hours. Two PFOA solutions (0.100 and
100 (.iu I.) were prepared in sterile water and control solutions were sterile water only. Test
solutions were added to flasks containing an initial density of 9xl05 cells/mL growing in BG-11
medium, iih three llasks for each treatment. The variation in algal density was observed every
day over the 12-day exposure period and chlorophyll pigment content and photosynthetic activity
was observed on days 3, 6, 9 and 12. During the first half of the test there was no significant
difference in growth of PFOA treatments and the control. On day 12, the growth was reduced by
6.76 and 14.4% relative to the control, in the 0.1 and 100 |ig/L PFOA treatments, respectively.
Later time points (i.e., >4-12 days of exposure) were not used quantitatively use because the
E-4
-------
exposure duration was too long (U.S. EPA 2012). The four-day cell density-based endpoint with
a NOEC of 0.1 mg/L was acceptable for quantitative use.
E.2.3 Green alga. Chlorella vulgaris
Boudreau (2002) performed a 96-hour static algal growth inhibition test on PFOA (acid
form, CAS # 335-67-1, >97% purity) with Chlorella vulgaris as part of a Master's thesis at the
University of Guelph, Ontario, Canada. Authors stated that the algal growth inhibition tests
followed protocols found in ASTM E 1218-97a (ASTM 1991)) and (ids el al (2000). Chlorella
vulgaris (UTCC 266 strain) used for testing were obtained as slants from the I ni\ ersity of
Toronto Culture Collection (UTCC; Toronto, Canada) Stock concentrations were prepared in
laboratory-grade distilled water with a maximum concentration that did not exceed the critical
micelle concentration for PFOA of 450 nig I. Dilution medium was Bristol's algal growing
media. Toxicity testing consisted of a range-finder lesl and at leasl two definitive tests. Nominal
test concentrations were ( neuali\ e control). 6.7, 12.5, 25. 5<). 1U0, 200, and 400 mg/L. Tests
were conducted in on \ I 5 111111 polyethylene disposable Petri dishes containing 20 mL of test
solution. Each Petri dish was inoculated with 1 5x 104 cells/mL at initiation and replicated four
times per test concentration Tests were continuously illuminated with cool-white, fluorescent
light between 3,800 and 4,2<)i) lu\ and incubated at 23 ± 1°C. Replicate Petri dishes were
manually shaken twice a day during testing. Toxicity test endpoints included cell density and
chlorophyll-a content The reported IC10, IC25 and IC50 based on growth inhibition (measured as
either chlorophyll-a or cell density) were 0.014 M (95% Confidence Interval, C.I.: 0.013-0.016),
0.034 M (95% C.I.: 0.032-0.040) and 0.279 M (95% C.I.: 0.249-0.320). Note that the ICx's for
PFOA were reported in molar (M) units, but EPA judged the units were misreported and were
actually millimolar (mM) units. This judgement was based on the reported test concentrations in
Table 3.1 of the publication and the reported effect concentrations (ICx) would not fall within
E-5
-------
this range unless the values were in mM units. Accordingly, the ICx reported as mM were
converted to mg/L by multiplying the mM concentration by a molecular weight of 414.07 g/mol
for PFOA. The calculated 96-hour ICio, IC25 and IC50 expressed as mg/L from the study were
5.797, 14.07 and 115.5, respectively and acceptable for quantitative use.
E.2.4 Green alga. Rayhidocelis subcayitata
(formerly known as Selenastrum capricornutum and I'seudokiix hiicncl/a subcapitata)
Boudreau (2002) also performed a 96-hour static algal growth inhibition test on PFOA
(acid form, CAS # 335-67-1, >97% purity) with Raphidncclis subcapitata as pail of the Master's
thesis at the University of Guelph, Ontario, Canada. Authors stated that the algal growth
inhibition test with R subcapitata similarly followed protocols found in ASTME 1218-97a
(ASTM 1999) and Geis et al. (2000). R. subcapitata (UTCC 37 strain) used for testing were
obtained as slants from the University of Toronto Culture Collection (I JTCC; Toronto, Canada).
Stock concentrations were prepared in laboratory-grade distilled water with a maximum
concentration that did not exceed the critical micelle concentration for PFOA of 450 mg/L.
Dilution medium was Bristol's algal growing media. Toxicity testing consisted of a range-finder
test and at least two dellniti\ e tests. Nominal test concentrations were 0 (negative control), 6.7,
12.5, 25. 5i). 100, 200, and 4<)i) nig I. Tests were conducted in 60 x 15 mm polyethylene
disposable Petri dishes with 2d mL of test solution. Each Petri dish was inoculated with 1.5xl04
cells/mL at initiation and replicated four times per test concentration. Tests were continuously
illuminated with cool-white, fluorescent light between 3,800 and 4,200 lux and incubated at 23 ±
1°C. Replicate Petri dishes were manually shaken twice a day during testing. Toxicity test
endpoints included cell density and chlorophyll-a content. The reported IC10, IC25 and IC50 based
on growth inhibition (measured as either chlorophyll-a or cell density) were 0.130 M (95% C.I.:
0.020-0.162), 0.197 M (95% C.I.: 0.166-0.231) and 0.298 M (95% C.I.: 0.274-0.317). As noted
E-6
-------
above, although the ICx for PFOA were reported in molar (M) units in the thesis, EPA judged
the units were misreported and were actually millimolar (mM). This judgement was based on the
reported test concentrations in Table 3.1 of the publication and the reported effect concentrations
(ICx) would not fall within this range unless the values were in mM units. Accordingly, the ICx
reported as mM were converted to mg/L by multiplying the mM concentration by a molecular
weight of 414.07 g/mol. The calculated 96-hour ICio, IC25 and TCs>. expressed as mg/L from the
study were 53.83, 81.57 and 123.4, respectively and acceptable for quanliUili\ e use.
More recently, Xu et al. (2013) conducted a 96-hour static, measured algal growth
inhibition test on PFOA (acid form, >98% purity) with Raphidocelis subcapiiaia Algae were
obtained from the Aquatic Organism Research Institute of the Chinese Academy of Science and
precultured for three generations prior to initialing the test. Dilution medium consisted of number
one culture medium supplemented with aquatic number lour nutrient solution (Zhou and Zhang
1989). Algae in logarithmic growth phase (7.0\ ID' cells ml.) were inoculated in medium
containing nominal concentrations of PI 'OA at 0 (negative control), 30, 60, 90, 120, 150, 180,
240, 300 and 36D mu |. Tests were conducted in 100 mL conical flasks with 50 mL of solution.
Each test concentration and the control were replicated three times. Exposure concentrations
were \ eiilicd \ ia IIHPLC-MS MS using the Agilent 1290 Infinity UHPLC system interfaced
with an Agilent Mm) Triple Ouadrupole mass. Algae were exposed under a 12-hour: 12-hour
light:dark cycle at 3.doo-4.d00 lux and 25°C. Chlorophyll concentration and permeability of cell
membranes was determined after 96 hours of exposure. The reported 96-hour growth EC50
(inhibition based on optical density) was 207.46 mg/L and was considered to be acceptable for
quantitative use.
E-7
-------
Colombo et al. (2008) evaluated growth inhibition with Raphidocelis subcapitata on
ammonium perfluorooctanoate (APFO, the ammonium salt of PFOA, CAS # 3825-26-1, 99.7%
purity). Authors stated that the 96-hour algal growth inhibition test followed OECD test guidance
201 and European Commission directive 92/69/EEC. The source of R. subcapitata used for
testing was not reported, but presumably from an in-house culture as the medium reported to be
used for both culturing and testing was reconstituted water recommended via the French algae
test guideline (AFNOR T 90-304). The media differs slightly from the Ol-('l) recommended
media with regard to concentrations of P, N, and chelators. Stock solutions of API O were
prepared by dissolving the test substance directly in the test media and diluting to provide a
geometric series of test concentrations. A range-finding and two definitive tests were conducted.
Definitive tests included six negative control replicates and three replicates at each PFOA
concentration. Tests were initiated via inoculation with I \ in1 cclls/mL from an algal culture in
log phase growth and carried out under continuous illumination with approximately 2,000 lux
and at 21-25°C. Test solutions were agitated to keep algae in suspension during the 96-hour
exposure and growth was determined al 24-hour intervals by counting an aliquot of test solution
from each replicate test chamber Test concentrations measured in the second definitive algal test
and A\ere <) (negative control). 5.76, I 1.37, 22.70, 46.33, 95.87, 180.67, and 369.67 mg/L. APFO
was determined as PI OA from a calibration curve of peak area against APFO concentrations in
standard solutions The limit of quantification (LOQ) of the analytical method was 1 mg/L.
Linearity was checked with a resulting coefficient of determination for the calibration curve of
greater than 0.999 in the range of 1-100 mg/L. Accuracy and precision were demonstrated by
analyzing six solutions containing nominal concentrations of 2.03 and 50.7 mg/L APFO in Milli-
Q water. The mean measured concentrations were 2.02 and 53.7 mg/L, respectively, with
E-8
-------
calculated precision of 6% and 2% and accuracy of 99% and 106%, respectively. The reported
96-hour NOEC, based on biomass and growth rate, was 11.37 mg/L. The reported 96-hour
LOEC was 22.70 mg/L. The calculated MATC was 16.07 mg/L and was considered to be
acceptable for quantitative use.
E.2.5 Green alga. Scenedesmus obliguus
Hu et al. (2014) evaluated algal growth inhibition of PI OA ( l)6% purity) with
Scenedesmus obliquus in both a 96-hour and eight-day static unmeasured exposures. Authors
stated that the tests followed OECD test guidance 201 (OECD 2006). \. ob/ujiins were supplied
by UTEX Culture Collection of Algae, University of Texas at Austin. Dilution medium was HB-
4 media adjusted to a pH of 6.8. Algae in exponential growth phase were exposed to nominal
concentrations of 0 (negative control), 1.3 K\ I <>. 3 1.6, I <)<). 31 (\ and 1,000 mg/L in the 96-hour
exposure, and 0, 5, 10, 20, and 40 mg/L in the eiuhl-day exposure l-.xperiments were initiated by
inoculating equal cell numbers of 5x103 cells ml. in the lH-«-hour exposure and 5xl06 cells/mL in
the eight-day exposure into 25<) nil. Ilasks containing a total volume of 100 mL of algal cell
suspension per flask There were ll\ e replicates per each treatment in the 96-hour exposure and
three replicates in the eiuhl-day exposure. Algae were incubated at 25°C under cool-white
fluorescence lights al S5-l)<) umol pholons/[m2 x s] irradiance with a 16-hour:8-hour light:dark
cycle. The lH->-hour growth I¦('*.) (inhibition based on optical density) was 44.0 mg/L. The eight-
day NOEC based on cell number was 40 mg/L (the highest test concentration). The plant values
from the study were acceptable for quantitative use.
E.2.6 Green alga. Scenedesmus guadricauda
Yang et al. (2014) conducted a 96-hour renewal, measured test on the growth effects of
PFOA (acid form, CAS #335-67-1, 99%) with the green alga, Scenedesmus quadricauda. Algae
were obtained from in-house cultures originally supplied by the Chinese Research Academy of
E-9
-------
Environmental Sciences. The algae used for testing were inoculated at a cell density equal to
2.0xl04 cells/mL in 50 mL beakers. PFOA was dissolved in deionized water and DMSO (amount
not provided) and then diluted with M4 medium. Algae in logarithmic growth phase were
exposed to 0 (solvent control), 80.00, 128.00, 204.80, 327.68, 524.29, and 838.86 mg/L. Each
treatment was replicated three times. Measured concentrations ranged from 75.68 mg/L (before
renewal) to 78.8 mg/L (after renewal) in the lowest treatment, and from 764.13 (before renewal)
to 831.45 mg/L (after renewal) in the highest treatment. The experiments u civ conducted at 22 ±
2°C with a 12-hour: 12-hour light:dark cycle. The initial pH of the test solution u as 7.0 ± 0.5,
total hardness was 190 ±0.1 mg/L as CaCCb, and total organic carbon was 0.02 mu I,. Algae
concentrations in the beakers were measured daily with a microscope. The 96-hour growth
inhibition ECso was reported as 269.63 mu I. and was acceptable lor quantitative use.
E.2.7 Watennilfoil. Myrioyhyllum sy.
Hanson et al. (2005) conducted a 35-day microcosm study on PFOA (sodium salt
donated by 3M Co., purity not pro\ idcd) with the submerged watermilfoils, Myriophyllum
spicatum andM sibiricnm The study was conducted in 12,000 L outdoor microcosms at the
University of (ruelph Microcosm I'acility located in Ontario, Canada using in-house cultures of
Myriophyllum spp. Each microcosm w as below ground and was flush with the surface. Plastic
trays filled with sediment (I I.I mixture of sand, loam and organic matter, mostly manure) were
placed in the bottom of each microcosm. The total carbon content of the sediment was 16.3%.
Ten apical shoots, 4 cm in length, from in-house cultures using the same sediment were
transferred to each microcosm, with three separate microcosms used for each treatment (nominal
concentrations 0, 0.3, 10, 30, and 100 mg/L). Endpoints of toxicity that were monitored on days
14, 21 and 35 of the study included growth in plant length, root number, root length, longest root,
node number, wet mass, dry mass and chlorophyll-a and -b content. PFOA treatments were
E-10
-------
dissolved in the same water (well water) used to supply the microcosms. Results showed that
measured concentrations remained similar to nominal concentrations throughout the entire
exposure period and did not change appreciably over the course of the study. The time-weighted
average measured concentrations were 0 (negative control), 0.27, 0.65, 23.9, and 74.1 mg/L.
Water quality over the length of the 35-day microcosm experiment was: dissolved oxygen: 7.3-
8.5 mg/L; temperature: 17.8-22.0°C; pH: 8.3-8.7; total hardness 217 5 mu'L as CaCCte. The
light:dark cycle was outdoor ambient cycles beginning June 13, 2000 ((mclph. Ontario). The
watermilfoil species were equally sensitive to P] O A The 35-day I X'io (based oil weight) was
21.6 mg/L for M. sibiricum and 19.7 mg/L for /. spicamm The plant values were acceptable for
quantitative use.
E.2.8 Lettuce. Lactuca sativa
Ding et al. 2012b conducted a microcosm study where water lettuce, Lactuca sativa, was
exposed to PFOA (CAS# 335-67-1) for 5 days Authors stated the test protocols followed U.S.
EPA (1996b). Test \ essels were plastic and test solutions were static and unmeasured. The test
employed six exposure concentrations and a negative control, with each treatment being
replicated three times Petri dishes containing lettuce seeds were placed in a plant test chamber
with a constant room temperature of 18°C ± 2°C and a photoperiod or 16-hours light and 8-hours
dark. After 5 chiys. the number of germinated seeds was counted, and the length of the roots was
measured with a ruler to the closest millimeter. The author reported ECso (endpoint = root
elongation) was 1.801 mM, which was converted to 745.7 mg/L and was retained for
quantitative use.
E-ll
-------
Appendix F Acceptable Estuarine/Marine Plant PFOA Toxicity Studies
No data at this time.
F-l
-------
Appendix G Other Freshwater PFOA Toxicity Studies
G. 1 Summary Table of Qualitative Freshwater PFOA Toxicity Studies
Species (lifestage)
Method3
Test
Duration
Chemical /
Purity
PH
Temp.
(°C)
Effect
Chronic
Limits
(NOEC-
LOEC)
(mg/L)
Reported
Effect Cone.
(mg/L)
Deficiencies
Reference
Cyanobacteria,
Anabaena sp.
S, M
24 hours
PFOA
96%
-
ECso
(bioluminescence
inhibition)
-
19.81
Duration too
short for a plant
test, missing
some exposure
details, non-
apical endpoint
Rodea-Palomares
et al. 2012
Cyanobacteria,
Anabaena sp.
S,U
24 hours
PFOA
96%
7.8
28
ECso
(bioluminescence
inhibition)
-
78.88
Duration too
short for a plant
test, missing
some exposure
details, non-
apical endpoint
Rodea-Palomares
et al. 2015
Green alga,
Raphidocelis subcapitata
(formerly, Selenastrum
capri cornutum)
S,U
96 hours
PFOA
96.5-100%
2.3-
10.3
-
ECso
(cell density and growth
rate)
-
90
Possible mixture
effects of other
perfluoro
homologue
compounds and
the amount of
isopropanol, wide
pH range.
3M Company
2000
Green alga,
Raphidocelis subcapitata
S,U
96 hours
APFO
96.5-100%
. -
23
ECio
(cell count)
-
5.3
Possible mixture
effects of other
perfluoro
homologue
compounds
3M Company
2000
Green alga,
Raphidocelis subcapitata
S,U
7 days
APFO
96.5-100%
-
23
ECio
(cell count)
-
3.3
Possible mixture
effects of other
perfluoro
homologue
compounds
3M Company
2000
Green alga,
Raphidocelis subcapitata
S,U
10 days
APFO
96.5-100%
-
23
ECio
(cell count)
-
2.9
Possible mixture
effects of other
perfluoro
homologue
compounds
3M Company
2000
Green alga,
Raphidocelis subcapitata
s,u
14 days
APFO
96.5-100%
-
23
ECio
(cell count)
-
5
Possible mixture
effects of other
perfluoro
homologue
compounds
3M Company.
2000
G-l
-------
Species (lifestage)
Method3
Test
Duration
Chemical /
Purity
PH
Temp.
(°C)
Effect
Chronic
Limits
(NOEC-
LOEC)
(mg/L)
Reported
Effect Cone.
(mg/L)
Deficiencies
Reference
Green alga,
Raphidocelis subcapitata
S,U
96 hours
APFO
Unknown
-
ECso
(cell count)
-
1,980
Test substance is
considered a
mixture of APFO
and other
impurities
3M Company.
2000
Green alga,
Raphidocelis subcapitata
S, M
72 hours
PFOA
96%
-
21-24
ECso
(growth)
96.2
Duration too
short for a plant
test, missing
some exposure
details
Rosal et al. 2010
Green alga,
Raphidocelis subcapitata
S,U
4.5 hours
PFOA
96%
-
-
ECso
(photosynthetic efficiency)
-
748.2°
Duration too
short for a plant
test, missing
some exposure
details, non-
apical endpoint
Ding et al. 2012b
Green alga
(104 cells/inL),
Scenedesmus obliquus
S,U
72 hours
PFOA
Unreported
7.5
22
NOEC
(growth rate)
-
828.1°
Duration too
short for a plant
test
Liu et al. 2008a
Duckweed,
Lemna gibba
S,U
7 days
PFOA
95%
A-
-
IC50
(wet weight)
-
79.92
Culture water not
characterized,
missing some
exposure details
Boudreau 2002
Tubificid worm
(0.03g, 0.8cm),
Limnodrilus hoffineisteri
S, M
96 hours
PFOA
99%
7
22
LCso
-
568.20
Atypical source
of organisms
Yang et al. 2014
Planaria (0.9 cm),
Dugesia japonica
S,U
96 hours
PFOA
>98%
25
LC50
-
427.7
Poor
concentration-
response curve
Li 2009
Planaria (10-12 mm),
Dugesia japonica
R, U
96 hours
PFOA
96%
-
20
LC50
-
39.35
Atypical source
of the test
organisms
Yuanet al. 2015
Planarian,
Dugesia japonica
R, U
10 days
PFOA
96%
-
20
LOEC
(decrease mRNA
expression levels of neural
genes DiFoxD, DiotxA and
DjotxB)
<0.5-0.5
0.5
Duration too long
for an acute test
and too short for
a chronic test,
non-apical
endpoint
Yuanet al. 2016b
G-2
-------
Spocios (lil'cshilic)
Molliod'1
losl
Dui'iilioii
( hcmic;il /
l*iiril>
pll
Temp.
<°C)
l.llecl
Chronic
l.imils
(NOI'.C-
i.or.ci
(lllli/l.)
Reported
KITccl ( one.
Deficiencies
Ucl'crcncc
Planarian,
Dugesia japonica
S,U
10 days
PFOA
96%
20
LOEC
(elevated lipid
peroxidation; increased
mRNA expression levels of
HSP 40 and HSP 70)
<0.5-0.5
0.5
Duration loo long
for an acute test
and too short for
a chronic test,
non-apical
endpoint
Yuanetal. 2017
Mud snail (4.0 g, 2.0 cm)
Cipangopaludina
cathayensis
S,M
96 hours
PFOA
99%
7
22
LC50
-
740.07
Atypical source
of organisms
Yang et al. 2014
Rotifer
(<2-hour old neonates),
Brachionus calyciflorus
R,Ud
4 days
PFOA
96%
-
:u
EC,..
(resting egg production)
0.125-
0 25
0.1768
(EPA-
Calculated ECi0:
0.07758)
Only one
replicate
Zhang et al. 2014b
Cladoceran,
Daphnia magna
S,U
48 hours
PFO\
9(i 5-1
" 5-
S 4
1<> 4-
:u:
EC*.
(death immobility)
-
360
Possible mixture
effects of other
perfluoro
homologue
compounds and
the amount of
isopropanol
3M Company
2000
Cladoceran,
Daphnia magna
S,U
4X hum's
\H<)
1,000
Possible mixture
effects of other
perfluoro
analogue
compounds
3M Company
2000
Cladoceran,
Daphnia magna
S,U
4X Ikmiis
\H<)
96.5-luo",,
-
-
EC50
(death/immobility)
-
126
Possible mixture
effects of other
perfluoro
analogue
compounds
3M Company
2000
Cladoceran (<24 hours
old),
Daphnia magna
S,U
48 hours
A,P| ( )
"S-93%
8.0-
8.1
21
EC50
(death/immobility)
-
221
Possible mixture
effects of other
perfluoro
analogue
compounds
3M Company
2000
Cladoceran (<24 hours
old),
Daphnia magna
S,U
48 hours
AIJFO
Unreported
8.1-
8.3
20.3-
20.8
LC50
-
1,200
Test substance is
considered a
mixture of APFO
and other
impurities
3M Company
2000
G-3
-------
Spocios (lil'cshilic)
Method'
losl
Dui'iilioii
( hcmic;il /
l*iiril>
nil
Tom p.
(°C)
r.iTcci
('limine
Limits
(NOI'.C-
i.or.ci
(lllli/l.)
Reported
HITccl ( one.
(iii'^/l.)
Deficiencies
Ucl'crcncc
Cladoceran (<24 hours
old),
Daphnia magna
s,u
48 hours
APFO
Unreported
00 °0
19.5-
20.1
EC*,
(death immobility)
-
584
Possible mixture
effects of the
inert
perfluorinated
compounds and
other perfluoro
analogue
compounds
3M Company
2000
Cladoceran,
Daphnia magna
R,U
48 hours
APFO
96.5-100%
-
-
EC*,
(death immobility)
-
266
Missing test
details, possible
mixture effects of
other perfluoro
analogue
compounds
3M Company
2000
Cladoceran,
Daphnia magna
R,U
21 days
APFO
96.5-100%
-
-
\IATC
(survival and reproduction)
::-36
28.14
Missing test
details, possible
mixture effects of
other perfluoro
analogue
compounds
3M Company
2000
Cladoceran (<24 hours
old),
Daphnia magna
R, U
21 days
H<)\
" 5
23
LOEC
(fecundity)
<0.4141-
0.4141
0.4141°
Chronic
responses in this
test did not
display
concentration-
dependent effects
beyond the
LOEC despite a
25X increase in
treatment
concentrations
Seyoum et al.
2020
Oriental river prawn
(0.30 g, 4.0 cm),
Macrobrachium
nipponense
S, M
hours
PFO \
99".,
"
22
LC50
-
366.66
Atypical source
of organisms
Yang et al. 2014
Midge (larva, 10 days
old),
Chironomus dilutus
R, U
10 days
NOV
- 97%
-
23
NOEC
(survival and growth)
-
100
Range-finding
experiment;
duration too long
for an acute test
and too short for
a chronic test
MacDonald et al.
2004
G-4
-------
Spocios (lil'cshitic)
Molliod'1
Test
Dui'iilioii
( licmiciil /
l*iiril>
nil
Temp.
<°C)
l.llecl
('limine
Limits
<\OI.( -
i.or.ci
(lllli/l.)
Koporied
IHTocl ( one.
(inii/1.)
Deficiencies
Ucl'crcncc
Midge (larva, 10 days
old),
Chironomus dilutus
S,M
10 days
PFOA
97%
-
\1 \T(
(mortality)
84.10
Rango-llnding
experiment;
duration too long
for an acute test
and too short for
a chronic test
McCarthy et al.
2021
Midge
(multi-generational),
Chironomus riparius
S,M
-20-38 days
/ generation
PFOA
Pure (unspecified)
oo
l< 00
20
NOEC
(emergence. reproduction,
so.\ ralio)
-
0.0089
Only one
exposure
concentration,
static chronic
exposure
Stefani et al. 2014
Midge
(multi-generational),
Chironomus riparius
S,M
-20-38 days
/ generation
PFOA
Pure
(unspecified)
oo
l< 00
:u
\oi:c
(increased nuilalion rale)
-
0.0089
Only one
exposure
concentration,
static chronic
exposure
Stefani et al. 2014
Midge (larva, 1st instar),
Chironomus riparius
S,M
-1 yearb
PFO \
I iiicportcd
7.5-
S 2
:u i
i.oir
(1; 10 developmental time,
adull weight, exuvia length)
;0.0098-
0.0098
0.0098
Only one
exposure
concentration,
static chronic
exposure, low
control survival
in 4 of the 10
generations
Marziali et al.
2019
Midge (0.05 g, 1.2 cm),
Chironomus plumosus
S,M
9<> hours
H<)\
"
::
LC50
-
402.24
Atypical source
of organisms
Yang et al. 2014
Rainbow trout (fry),
Oncorhynchus mykiss
Diet, U
~ii da> s
H<)\
I iiicportcd
-
14
MATC
(liver somatic index)
200-1,800
(mg/kg)
600
(mg/kg diet)
Non-apical
endpoint
Tilton et al. 2008
Rainbow trout (fry),
Oncorhynchus mykiss
Diet, U
6 months
PFO \
I nrepmied
-
14
MATC
(palmitoyl CoA p-oxidation
- liver enzyme)
200-1,800
(mg/kg)
600
(mg/kg diet)
Non-apical
endpoint
Tilton et al. 2008
Rainbow trout (juvenile),
Oncorhynchus mykiss
Diet, U
15 da> s
PFO \
I iircportod
-
12
MATC
(increase plasma
vitellogenin)
5-50
(mg/kg)
15.81
(mg/kg diet)
Test design and
lack of exposure
details
Benninghoff et al.
2011
Rainbow trout (fry, 10-
15 weeks old),
Oncorhynchus mykiss
Diet, U
6 months
PFOA
Unreported
-
12
LOEC
(increase tumor multiplicity
and size)
<2,000-
2,000
2,000
(mg/kg diet)
Test design and
lack of exposure
details
Benninghoff et al.
2012
G-5
-------
Spocios (lil'cshilic)
Molliod'1
losl
Dui'iilioii
( hcmic;il /
l*iiril>
pll
Temp.
(°C)
l.llecl
( limine
l.imils
(NOI'.C-
i.or.ci
(inii/1.)
Rc'pork'd
IHTocl ( one.
Deficiencies
Ucl'crcncc
Atlantic salmon
(embryo-larval),
Salmo salar
F, U
52 days
PFOA
95%
5-7
]S( )l ( -I.()l (
(growill - weight and
length)
0 I- 0 I
>0.1
No elleets at the
Spachmo and
Arukwe 2012
highest
concentration
tested resulting in
a relatively low
greater than value
that does not
inform species
sensitivity
Goldfish (6.0 g, 7.0 cm),
Carassius auratus
S,M
96 hours
PFOA
99%
7
22
LCm,
-
606.61
Atypical source
of organisms
Yang et al. 2014
Goldfish
(juvenile, 27.85 g),
Carassius auratus
S,M
96 hours
PFOA
>98%
7.25
Antioxidanl cn/\ me
acii\ n\
-
>4.93 lc
Only two
exposure
concentrations,
non-apical
endpoint
Feng et al. 2015
Common carp (juvenile,
~12 cm, ~20 g),
Cyprinus carpio
F, M
96 hours
PFO \
')') S"„
6 9
I.()l (
(vitellogenin (VTG)
activity)
-
6.582
Broad range of
test treatments,
non-apical
endpoint
Kim et al. 2010
Common carp
(adult - 2 years old),
Cyprinus carpio
F, U
(tissue)
56 days
HO\
(.
SO
10-15
LOEC
(PCNA-positive hepatocyte
abundance)
0.0002-2
2
Poor test design,
only two
exposure
concentrations,
non-apical
endpoint
Giari et al. 2016
Common carp
(adult - 2 years old),
Cyprinus carpio
F, U
(tissue)
5<> da> s
(.
SO
10-15
LOEC
(liver biomarkers)
0.0002-2
2
Only two
exposure
concentrations,
non-apical
endpoint
Manera et al. 2017
Zebrafish (embryo),
Danio rerio
R,U
96 hours
API ¦'()
98".,
-
26
LC50
-
386.3°
Inability to verify
LC50
Ding et al. 2012c,
2013
Zebrafish (embryo),
Danio rerio
S,U
72 hours
I'FOA
<>5",,
8.3
28.5
LC50
-
262
Duration too
short for acute
test
Zheng et al. 2012
Zebrafish (embryo),
Danio rerio
S,U
96 hours
I'RIV
95%
8.3
28.5
EC50
(malformation)
-
198
Non-apical
endpoint
Zheng et al. 2012
G-6
-------
Spocios (lil'cshilic)
Molliod'1
losl
Dui'iilioii
( licmiciil /
l*iiril>
nil
Temp.
<°C)
l.llecl
('limine
Limits
(NOI'.C-
i.or.ci
(lllli/l.)
Reported
HITccl ( one.
(iii'^/l.)
Deficiencies
Ucl'crcncc
Zebrafish
(embryo, 4 hpf),
Danio rerio
R,U
120 hours
PFOA
Unreported
28
NOEC-LOEC
(increase relative mRNA
expression of hhex and pax
8)
0 1-0:
-
Duration loo
short for a
chronic test and
too long for an
acute test, non-
apical endpoint,
only three
exposure
concentrations
Du et al. 2013
Zebrafish (adult),
Danio rerio
R,U
(tissue)
28 days
PFOA
96%
-
26
NOEC
(reproduction: fecundity,
fertility and hatching)
1-: 1
I
Not a true ELS
test
Hagenaars et al.
2013
Zebrafish (adult),
Danio rerio
R, U
(tissue)
28 days
PFOA
96%
-
26
LOEC
(alterations of gene
transcripts)
<0.1-0.1
0.1
Not a true ELS
test, non-apical
endpoint
Hagenaars et al.
2013
Zebrafish
(embryo, 4 cell stage),
Danio rerio
S,U
Fertilization
to 144 hours
post-
fertilization
(6 days)
PFOA
Unreported
7.2-
7.6
26
EG,.
(lethal and sublethal
endpoinl)
-
350
Static chronic
exposure
Ulhaq et al. 2013
Zebrafish
(embryo, 6 hpf),
Danio rerio
S,U
114 hours
\H ()
I lll'cpoi'lcd
-
-
LOEC
(mortality)
<0.02759-
0.02759
0.02759°
Duration too long
for acute test
Truong et al. 2014
Zebrafish
(embryo, 6 hpf),
Danio rerio
S,U
114 hours
H<)\
I mvpmicd
-
-
NOEC
(mortality)
26.50-
>26.50
26.50°
Duration too long
for acute test
Truong et al. 2014
Zebrafish (adult),
Danio rerio
R, U
: 1 da> s
HO\
')(<
-
MATC
(decrease in inflammatory
cytokines (IL-lfl and IL-21)
in spleen)
0.05-0.1
0.0707
Duration too
short for a
chronic test, non-
apical endpoint
Zhang et al. 2014a
Zebrafish
(embryo, 2 days pf),
Danio rerio
S,U
"2 hours
PFO \
96".,
-
28.5
LC50
-
157.3°
Duration too
short for an acute
test
Kalasekar et al.
2015
Zebrafish (embryo),
Danio rerio
S,U
72 hours
H<)\
I lll'cpoi'lcd
-
26
NOEC
(embryo toxicity)
-
132.5°
Only one
exposure
concentration and
duration too short
for an acute test
Bouwmeester et
al. 2016
G-7
-------
Spocios (lil'cshilic)
Molliod'1
losl
Dui'iilioii
( licmiciil /
l*iiril>
pll
Temp.
<°C)
l.llecl
('limine
Limits
(NOI'.C-
i.or.ci
(lllli/l.)
Kcpnrk-ri
lllTccl ( one.
(inii/1.)
Deficiencies
Ucl'crcncc
Zebrafish
(gastrula stage, 4.5 hpf),
Danio rerio
R,U
Embryo
development
to 28 days
post-hatch
PFOA
Unreported
28
LOEC
(swim bladder
development)
-
4.7
Unconventional
test design, diet
and water
concentrations
were not
measured
Godfrey et al.
2017b
Zebrafish (3 hpf),
Danio rerio
S,U
5 day + 9
day
observation
PFOA
Unreported
7.2-
7.7
26-28
MATC
(grow ill - total body length,
intorocular distance, yolk
sac area)
(i
0.6325
Duration too long
for an acute test,
Jantzen et al.
2017b
Zebrafish (3 hpf),
Danio rerio
S,U
5 day + 9
day
observation
PFOA
Unreported
7.2-
7.7
26-28
\1 \TC
(sw ininiing activity -
distance traveled)
0.02-0.2
0.06325
Duration too long
for an acute test,
non-apical
endpoint
Jantzen et al.
2017b
Zebrafish
(embryo, 72 hpf),
Danio rerio
S,M
48 hours
PFOA
Unreported
-
:_
i.(
-
>500
Duration too
short for an acute
test
Rainieri et al.
2017
Zebrafish
(embryo, 3 hpf),
Danio rerio
S,U
117 hours +
9 days
observation
PFOA
Unreported
7.2-
7.7
\1 \T(
(morphology)
X281-
0 US281
0.02619
Duration too long
for an acute test
Annunziato 2018
Zebrafish (embryo),
Danio rerio
S,U
168 hours
H<)\
I mvpmicd
-
-
LC50
-
362.5
Duration too long
for an acute test
Stinckens et al.
2018
Zebrafish
(embryo, 2 hpf),
Danio rerio
R,M
118 hours
H<)\
-
28
ECso
-
210.8°
Duration too long
for an acute test,
no true
replication
Vogs etal. 2019
Zebrafish
(embryo, 2 hpf),
Danio rerio
R, M
1 1 S Ikilll's
NOV
')(<••„
-
:s
ec20
(deformities)
-
147.2°
Duration too long
for an acute test,
no true
replication
Vogs etal. 2019
Zebrafish
(embryo, 6 hpf),
Danio rerio
S,U
o<> hours
PFO \
96".,
-
28
NOEC
(survival and development)
20.70-
>20.70
20.70°
Duration too
short for an acute
test, only one
exposure
concentration
Dasgupta et al.
2020
Zebrafish
(embryo, 1 hpf),
Danio rerio
S,U
48 hours,
PI (JA
I uicpoiled
-
28
LC50
-
300
Duration too
short for an acute
test
Pecquet et al. 2020
Zebrafish
(embryo, 1 hpf),
Danio rerio
S,M
24 hours
PFOA
Unreported
-
28
LOEC
(increase neutrophil
migration)
<0.685-
0.685
0.685
Duration too
short for an acute
test, atypical
endpoint
Pecquet et al. 2020
G-8
-------
Spocios (lil'cshilic)
Molliod'1
losl
Dui'iilioii
( hcmic;il /
l*iiril>
pll
Temp.
<°C)
l.llecl
('limine
Limits
(NOI'.C-
i.or.ci
(lllli/l.)
Kcpnrk-ri
HITccl ( one.
(inii/1.)
Deficiencies
Ucl'crcncc
Zebrafish
(embryo, 5-6 hpf),
Danio rerio
R, U
90-91 hours
PFOA
>99%
28
l.(
-
57.6
Duration too
short for an acute
test
Wasel et al. 2020
Zebrafish
(embryo, 5-6 hpf),
Danio rerio
R, U
90-91 hours
PFOA
>99%
7
28
l.(
-
487.4
Duration too
short for an acute
test
Wasel et al. 2020
Zebrafish,
Danio rerio
R, U
21 days
PFOA
Unreported
-
-
LOEC
(mRNA gene expression in
kidneys)
0.05-O 1
0.1
Atypical
endpoint
Zhang et al. 2021
Rare minnow (male, 9
months old, 1.4 g, 47.7
cm),
Gobiocypris rarus
F, U
28 days
PFOA
98%
-
25
\1 \T(
(increase relative mRNA
expression of AhR in gills)
3.0-10
5.477
Non-apical
endpoint
Liu et al. 2008b
Rare minnow (female,
9 months old, 1.4 g, 47.7
cm),
Gobiocypris rarus
F, U
28 days
PFOA
98".,
-
25
MATC
(decrease relative mRNA
expression of CYPla and
increase relative mRNA
expression of PXR in gills)
10.0-30
17.32
Non-apical
endpoint
Liu et al. 2008b
Rare minnow
(adult, 9 months old),
Gobiocypris rarus
F, U
28 days
H<)\
<>S"„
-
25
LOEC
(polymerase chain reaction
(PCR)
alterations of genes in liver)
<3-3
3
Non-apical
endpoint
Wei et al. 2008a
Rare minnow
(adult, 9 months old),
Gobiocypris rarus
F, U
2X da> s
H<)\
<>X"„
-
-
MATC
(change in m-RNA M-H-
FABP)
3.0-10
5.477
Non-apical
endpoint
Wei et al. 2008b
Rare minnow
(adult, 9 months old),
Gobiocypris rarus
F, U
:s da> s
HO\
98%
-
-
LOEC
(protein spots identified by
MALD1-TOF/TOF)
<3-3
3
Non-apical
endpoint
Wei et al. 2008b
Rare minnow (9 months
old female, 1.4 g, 47.7
mm),
Gobiocypris rarus
F, U
2X da> s
PFO \
98".,
-
25
MATC
(increase relative mRNA
expression of PPARy in
gills)
3.0-10
5.477
Non-apical
endpoint
Liu et al. 2009
Rare minnow (9 months
old male, 1.4 g, 47.7
mm),
Gobiocypris rarus
F, U
28 days
I'lOV
98%
-
25
MATC
(increase relative mRNA
expression of PPARy and
PPARa in gills and
CYP4T11 in liver)
10.0-30
17.32
Non-apical
endpoint
Liu et al. 2009
G-9
-------
Spocios (lil'cshilic)
Molliod'1
losl
Dui'iilioii
( hcmic;il /
l*iiril>
pll
Temp.
(°C)
l.llecl
Chronic
l.imils
(NOI'.C-
i.or.ci
(lllli/l.)
Reported
KITccl ( one.
Deficiencies
Ucl'crcncc
Rare 111111110W (9 months
old male, 1.3 g),
Gobiocypris rarus
F, U
14 days
PFOA
98%
25
LOHC
(apolipoprotein gene
expression)
-
3
Duration loo
short for a
chronic test, non-
apical endpoint
Fang et al. 2010
Fathead minnow
(juvenile),
Pimephales promelas
s,u
96 hours
APFO
Unknown
7.2-
7.9
21.8-
22.5
LC50
-
:,470
Test substance
was considered a
mixture of APFO
and other
impurities
3M Company
2000
Fathead minnow,
Pimephales promelas
s,u
96 hours
PFOA
96.5-100%
-
-
l.(
-
440
Lack of exposure
details, possible
mixture effects of
other perfluoro
homologue
compounds
3M Company
2000
Fathead minnow,
Pimephales promelas
s,u
96 hours
PFO \
9(> 5-1 ()()"„
-
-
l.(
-
140
Possible mixture
effects of other
perfluoro
homologue
compounds and
the amount of
isopropanol, low
initial pH (3.0-
4.3) in highest
test concentration
3M Company
2000
Fathead minnow,
Pimephales promelas
s,u
9<> hours
\P1()
'X, 5-|u0°/„
-
-
LC50
-
70
Possible mixture
effects of other
perfluoro
analogue
compounds, lack
of replication
3M Company
2000
Fathead minnow,
Pimephales promelas
s,u
9<> hours
APR)
%.5-]ou"„
"l>-
8.0
19
LC50
-
776
Possible mixture
effects of other
perfluoro
homologue
compounds, lack
of replication
3M Company
2000
Fathead minnow,
Pimephales promelas
s,u
96 hours
APFO
96.5-100%
7.9-
8.0
19
LC50
-
754
Possible mixture
effects of other
perfluoro
homologue
compounds, lack
of replication
3M Company
2000
G-10
-------
Spocios (lil'cshilic)
Molliod'1
losl
Dui'iilioii
( hcmic;il /
l*iiril>
I'"
Tom p.
(°C)
l.llecl
('limine
Limits
(NOI'.C-
i.or.ci
(lllli/l.)
Reported
lllTccl ( one.
(iii'^/l.)
Deficiencies
Uelerence
Fathead minnow,
Pimephales promelas
S,U
96 hours
APFO
78-93%
7.7-
8.0
20
l.(
-
301
Possible mixture
effects of other
perfluoro
analogue
compounds
3M Company
2000
Fathead minnow,
Pimephales promelas
s,u
96 hours
PFOA
95-98%
7.5-
7.7
19-20
-LC50
-
843
Possible mixture
effects of other
perfluoro
impurities
3M Company
2000
Fathead minnow
(juvenile),
Pimephales promelas
s,u
96 hours
APFO
Unreported
7.4-
8.4
21-22
I.C
-
>1,000
Possible mixture
effects of the
inert
perfluorinated
compounds and
other perfluoro
analogue
compounds
3M Company
2000
Fathead minnow
(embryo, 48 hpf),
Pimephales promelas
F, U
30 days post
hatch
APFO
96.5-luo",,
7.0-
7.3
25
\oi:c
(hatch. survival and
grow ih)
I00->100
100
Possible mixture
effects of other
perfluoro
analogue
compounds, lack
of replication
3M Company
2000
Fathead minnow (64
days old),
Pimephales promelas
S,M
13 days
\H<)
9(> 5-11111".,
-
-
BCF
-
1.8
(L/Kg)
Steady state not
documented,
static uptake
study
3M Company
2000
Fathead minnow
(adults),
Pimephales promelas
S,M
da> s
H<)\
l<> 4"o'
S 5
1 0
26.6
LOEC
(mean total egg production)
74.1-
>74.1
>74.1
Atypical
exposure, started
with adults, not a
true ELS test
Oakes et al. 2004
Topmouth gudgeon
(4.0 g, 4.0 cm),
Pseudorasbora parva
S,M
hours
PFO \
99".,
"
22
LC50
-
365.02
Atypical source
of test organisms
Yang et al. 2014
Topmouth gudgeon
(4.0 g, 4.0 cm),
Pseudorasbora parva
R,M
30 days
PFO \
7
22
EC10
(survival)
-
11.78
Not a true ELS
test (started with
older life stage),
atypical source of
organisms
Yang et al. 2014
G-ll
-------
Spocios (lil'cshilic)
Molliod'1
losl
Dui'iilioii
( licmiciil /
l*iiril>
pll
Temp.
<°C)
l.llecl
('limine
Limits
(NOI'.C-
i.or.ci
(lllli/l.)
Reported
lllTccl ( one.
Deficiencies
Ucl'crcncc
Bluegill,
Lepomis macrochirus
S,U
96 hours
APFO
96.5-100%
7.8-
8.0
18-19
-
569
Only one
replicate per
treatment,
possible mixture
effects of other
perfluoro
analogue
compounds
3M Co. 2000
Murray River
rainbowfish
(male, adult, 1 year old),
Melanotaenia fluviatilis
R,M
14 days
PFOA
>96%
7.1-
7.4
23
\oi:c
(growill and mortality)
9.0->9.0
9.0
Duration too long
for acute test and
too short for a
chronic test, not
NA species
Miranda et al.
2020
Medaka (<6 hpf),
Oryzias latipes
R, U
Embryo
development
to 48 hours
post-hatch
PFOA
Unreported
-
25
T.OHC
(swim bladder
development)
-
4.7
Only one
exposure
concentration
Godfrey 2017
Medaka (adult, male),
Oryzias latipes
R, U
14 days
H<)\
I iiicpni'icd
-
25
\< )i :c
(adull survival, GSI%,
HSI%, K%)
10->10
10
Duration too long
for acute test and
too short for a
chronic test
Ji et al. 2008
Medaka (adult, female),
Oryzias latipes
R, U
14 days
HO\
Unreported
-
25
NOEC
(adult survival, GSI%,
HSI%, K%)
10->10
10
Duration too long
for acute test and
too short for a
chronic test
Ji et al. 2008
Medaka (F1 generation,
<12 hours old, embryo),
Oryzias latipes
R, U
7-14 days
(assumed)
PI <) \
U ii re purled
-
25
NOEC
(% hatchability)
10->10
10
Duration too long
for acute test and
too short for a
chronic test
Ji et al. 2008
Medaka (F1 generation,
<12 hours old, embryo),
Oryzias latipes
R, U
7-14 days
(assumed)
PFO \
Unreported
-
25
MATC
(time to hatch)
1.0-10
3.162
Duration too long
for acute test and
too short for a
chronic test
Ji et al. 2008
Medaka (F1 generation,
<12 hours old, embryo),
Oryzias latipes
R, U
28 da> s
post-halch
(assumed)
PFO \
I iircported
-
25
NOEC
(swim up success)
10->10
10
Pseudoreplication
that occurred at
hatching stage
Ji et al. 2008
Medaka (F1 generation,
<12 hours old, embryo),
Oryzias latipes
R, U
100 days
post-hatch
HOY
Unreported
-
25
NOEC
(growth - length and
weight)
10->10
10
Pseudoreplication
that occurred at
hatching stage
Ji et al. 2008
G-12
-------
Spocios (lil'cshilic)
Molliod1
losl
Diimlion
( hcmic;il /
I'uriU
nil
Temp.
<°C)
l.llecl
( limine
l.imils
(NOI'.C-
i.or.ci
(mii/l.)
Reported
lllTccl ( one.
(mji/l.)
Deficiencies
Ucl'crcncc
\1alaka (I I ueiieialinn.
<12 hours old, embryo),
Oryzias latipes
R, U
loo days
post-hatch
PFOA
Unreported
25
NOLL
(male/female GSI% and
HSI%)
0.1->0.1
0.1
Pseudoreplicalion
that occurred at
hatching stage
Ji et al. 2008
Medaka (F1 generation,
<12 hours old, embryo),
Oryzias latipes
R, U
28 days
post-hatch
PFOA
Unreported
-
25
LOEC
(larval survival)
- (1 1-0 1
0.1
Pseudoreplication
that occurred at
hatching stage
Ji et al. 2008
Medaka
(adult, 16 weeks old,
0.38 g),
Oryzias latipes
R, U
21 days
PFOA
96%
-
25
T.OFC
(Ibcundily)
<10-|u
10
Only one
exposure
concentration,
tolerant LOEC
value
Kang et al. 2019
Northern leopard frog
(larva, Gosner 26),
Lithobates pipiens
(formerly, Rana pipiens)
R, M
40 days
PFOA
96%
-
20
\or(
(snoul-vcnl Icnglli
and Ciosnor stage at 40 d)
1->1
1
No effects at the
highest
concentration
tested resulting in
a relatively low
greater than value
that does not
inform species
sensitivity
Hoover etal. 2017
Northern leopard frog
(larva, Gosner 25),
Lithobates pipiens
S,M
30 days
H<)\
" s
26.2
NOEC
(survival and growth)
0.066-
>0.066
0.066
Mesocosm
exposure
Flynnetal. 2021
Gray treefrog (larva,
Gosner 40),
Hyla versicolor
S,U
9 hours
Pi t) \
99".,
7
22
LC50
-
114.74
Atypical source
of organisms
Yang et al. 2014
Asiatic toad (tadpole,
1.8 cm, 0.048 g),
Bufo gargarizans
R, M
30 days
H<)\
W,,
7
22
EC10
(survival)
-
5.89
Not a true ELS
test, atypical
source of
organisms
Yang et al. 2014
a S=static, R=renewal, F=flow-through, U=unmeasured, M=mcasured, T=total, D=dissolved, Diet=dietary, MT=maternal transfer
b CI year corresponds to the total duration of the 10-generations study. Most generations did not show statistically significant effects.
c Reported in moles converted to milligram based on a molecular weight of 414.07 g/mol PFOA or 431.1 g/mol APFO.
G-13
-------
d Chemical concentrations made in a side-test representative of exposure and verified stability of concentrations of PFOA in the range of concentrations tested under similar
conditions. Daily renewal of test solutions.
G-14
-------
G.2 Qualitatively Acceptable PFOA Toxicity Study Summaries for
Tests Not Described in the Effects Characterization
G.2.1 Summary of Acute PFOA Toxicity Studies Used Qualitatively in the Freshwater Aquatic
Life Criterion Derivation
G.2.1.1 Worms
Yang et al. (2014) conducted a 96-hour measured acute test of PFOA (CAS # 335-67-1,
99% purity) with the annelid worm, Limnodrilus hoffmeisteri. Although the authors termed the
test conditions "static", they also mentioned PFOA measurements before and after renewal;
based on this distinction the test was assumed to be renewed at least once Authors stated that the
test followed ASTM E729 (1993). L. hoffmeisteri (0.03 g, 0.8 cm) used for the test were obtained
from Beijing City Big Forest Flower Market and allowed lo acclimate for at least seven days
before testing. Dilution water was dechloiinaled lap water (pi I. 7 t) ± 0.5; dissolved oxygen, 7.0
± 0.5 mg/L; total organic carbon, 0.02 mg I., and lolal hardness, h'n ii t 0.1 mg/L as CaC03).
The photoperiod consisted of 12-hours of illumination at an unreported intensity. A primary
stock solution was prepared In dissoK ing PFOA in deionized water and solvent, DMSO, and
proportionally diluted with dilution water to prepare the test concentrations. Exposure vessels
were l><) cm Petri dishes containing I n ml. of test solution. The test employed three replicates of
10 worms each in six test concentrations plus a negative and solvent control. Nominal
concentrations were 0 (negati\e and solvent controls), 300, 390, 507, 659.1, 856.83 and 1,113.88
mg/L. Test concentrations were measured in low and high treatments only. The authors provided
mean measured concentrations before and after renewal: 295.3 and 259.31 mg/L (lowest
concentration) and 1,098.05 and 987.37 mg/L (highest concentration). Analyses of test solutions
were performed using high performance liquid chromatography with mass spectrometric
detection (HPLC/MS) and negative electrospray ionization. The concentration of PFOA was
calculated from standard curves (linear in the concentration range of 1-800 ng/mL), and the
G-15
-------
average extraction efficiency was in the range of 70-83%. The concentrations and
chromatographic peak areas exhibited a significant positive correlation (r = 0.9987, p < 0.01),
and the water sample-spiked recovery was 99%. The temperature, D.O., and pH were reported as
having been measured every day during the acute test, but results are not reported. Negative
control and solvent control mortality was 0% and 3%, respectively. The 96-hour LCso was 568.2
mg/L (95%) C.I.: 476.3-677.8 mg/L). The acute value was no1 acceptable for quantitative use
because of the uncertainties and potential exposures to PFAS from the source of the test
organisms but was retained for qualitative use by providing relevant toxicity information,
Li (2009) conducted a second 96-hour static, unmeasured acute test of PI OA
(ammonium salt, >98%> purity) with Dugesia japonica, which is described in greater detail in
Appendix A.2.7: Eighth most acutely sen si li \ e uciuis Dugcsm The author-reported 96-hour
LCso was 337 mg/L (95% C.I.: 318-357 mu I.) which was u\ crimed across the three tests. EPA
performed C-R uniilysis lor each individual Icsl Two of llic lesis had acceptable curves while the
third curve had 11 poor concentration response and the LC50 (427.7 mg/L) was, therefore, not
acceptable for qiuintitati\ e use hut was retained for qualitative use only.
Yusin el :il. (2015) also conducted 11 %-hour, unmeasured acute test on PFOA (96%
purity) with / Uigcsiajaponica. with daily solution changes. Planarians used for testing were
originally collected from a fountain in Quanhetou, Boshan, China, and acclimated in the
laboratory for an unspecified time period before use. The planarians had a body length of 10-12
mm at test initiation. Dilution water was aerated tap water. No details were provided regarding
photoperiod or light intensity. A primary stock solution was prepared by dissolving the salt in
dimethyl sulfoxide (DMSO). The control and exposed planarians received 0.005%) DMSO (v/v).
Exposure vessels' material, dimensions and fill volume were not reported. The test employed
G-16
-------
three replicates of 10 planarians each in seven test concentrations plus a solvent control. Nominal
test concentrations were 0 (solvent control), 10, 30, 35, 40, 45, 50 and 55 mg/L PFOA. The test
temperature was reported as 20°C. No other water quality parameters were reported. Mortality of
solvent control animals was also not reported. The study reported 96-hour LCso was 39.35 mg/L
(95% C.I. = 32.38 - 46.32 mg/L). The test was not acceptable for quantitative use because of the
uncertainties and potential exposures to PFAS from the source of 1 lie lest organisms.
G.2.1.2 Mollusks
Yang et al. (2014) conducted a 96-hour measured acute test of PFOA (CAS # 335-67-1,
99% purity) with a non-North American snail species, Cipan^opaludina cathawnsis Although
the authors termed the test conditions "static", they also mentioned PFOA measurements before
and after renewal; based on this distinction the lest was assumed to be renewed at least once.
Authors stated that the test followed ASTM I-721) (11^3) The lest organisms (4.0 g, 2.0 cm)
were purchased from the lieijinu Dahongmen Jingshen Seafood Market and allowed to acclimate
for at least seven days before testing Dilution water was dechlorinated tap water (pH, 7.0 ± 0.5;
dissolved oxygen. 7 t) t) 5 mg I., lolal organic carbon, 0.02 mg/L; and total hardness, 190.0 ±
0.1 mg I. as CaCO;) The pholopcriod consisted of 12-hours of illumination at an unreported
intensity A primary stock solution was prepared by dissolving PFOA in deionized water and
solvent, DMSO. and proportionally diluted with dilution water to prepare the test concentrations.
Exposure vessels were 2<)() mL beakers of unreported material type containing 100 mL of test
solution. The test employed three replicates of 10 snails each in six test concentrations plus a
negative and solvent control. Nominal concentrations were 0 (negative and solvent controls),
300, 420, 588, 823.2, 1,152.48 and 1,613.47 mg/L. Test concentrations were measured in low
and high treatments only. The authors provided mean measured concentrations before and after
G-17
-------
renewal: 293.55 and 261.77 mg/L (lowest concentration) and 1,596.62 and 1,468.08 mg/L
(highest concentration). Analyses of test solutions were performed using HPLC/MS and negative
electrospray ionization. The concentration of PFOA was calculated from standard curves (linear
in the concentration range of 1-800 ng/mL), and the average extraction efficiency was in the
range of 70-83%. The concentrations and chromatographic peak areas exhibited a significant
positive correlation (r = 0.9987, p < 0.01), and the water sample-spiked recovery was 99%. The
temperature, D.O., and pH were reported as having been measured every day during the acute
test, but results were not reported. Negative control and solvent control mortality were 0% each.
The 96-hour LCso was 740.07 mg/L (95% C.I.: 5l)7 716 4 mg/L). The acute \ nine was not
acceptable for quantitative use because of the uncertainties and potential exposures to PFAS
from the source of the test organisms but was retained to for used in a qualitative manner.
G. 2.1.3 Planktonic crustaceans
3M Company (2000) exposed/), ruaifiia to PFOA (CAS # 335-67-1) in a 48-hour static,
unmeasured acute toxicity test The toxicant was part of the 3M production lot number 269 and
was characterized as mixture of PI OA (lH-> 5-|oo% of the compound) and C6, C7 and C9
perfluoro homoloune compounds (<)-3 5% of the compound). The substance was dissolved in a
50:5o water isopropanol solution to make a primary solution of 1,000 mg/L test substance and
isopropanol Another toxicity test conducted by the authors showed no mortality or sublethal
effects at 390 mu I. isopropanol on the same species. It was then diluted with reconstituted water
to make five nominal test concentrations (130, 200, 360, 600 and 1,000 mg/L test substance or
65, 110, 180, 300 and 500 mg/L PFOA) plus a control (reconstituted water only). The test
followed USEPA-TSCA Guideline 797.1300. Exposures were conducted in 300 mL glass
beakers with 250 mL of test solution with 10 daphnids per beaker. There were two replicates for
G-18
-------
each treatment. Test conditions throughout the experiment varied little (D.O.: 8.7-8.8 mg/L; pH:
7.5-8.4; 19.4-20.2°C). No mortality occurred in the control treatment and PFOA treatments <180
mg/L. The 48-hour reported ECso, based on death/immobility, was 720 mg/L test substance and
isopropanol or 360 mg/L PFOA. This test was not acceptable for quantitative use because of
possible mixture effects from other perfluoro homologue compounds in the test substance, but
was retained for qualitative use.
3M Company (2000) summarized four 48-hour static, unmeasured AITO (CAS # 3825-
26-1) acute toxicity tests with the cladoceran, Daphnia magna and APFO. "J'lie toxicant was part
of the 3M production lot number 37 and was characterized as mixture of APFO (lK-< 5-100% of
the compound) and C6, Ci and C9 perfluoro analogue compounds (0-3.5% of the compound). No
specific test protocol was identified. Solutions of APFO were made in carbon filtered well water.
Exposures were conducted in 250 mL glass beakers with 2<>0 mL of test solution and 10-20
daphnids per beaker There were two replicates for each treatment and test. The authors noted
that the results of the lour \ arious tests were inconsistent and that the inconsistencies in the effect
concentrations may he due to diet (specifics not provided). The 48-hour ECsos determined from
tests conducted in May llM2. bused 011 death/immobility, were >1,000 mg/L APFO, while the
EC50 lor a subsequent test in June 1982 was reported to be 126 mg/L APFO. The results of the
acute tests were reported by the study authors to have been superseded by a more recent study
reported in 3M Co (2<)(P) These inconsistencies and the possible mixture effects of other
perfluoro analogue compounds rendered these tests unacceptable for quantitative use and they
were retained for qualitative use.
3M Company (2000) summarized a 48-hour static, unmeasured acute toxicity test with
the cladoceran, Daphnia magna and APFO (CAS # 3825-26-1). The toxicant was part of the 3M
G-19
-------
production lot number 390 and was characterized as mixture of APFO (78-93% of the
compound) and Cs, C(, and C7 perfluoro analogue compounds (7-22% of the compound). No
specific test protocol was identified. Solutions of APFO were made in carbon filtered well water
and included five nominal test concentrations (100, 180, 320, 560, and 1000 mg/L test substance)
plus a control (well water only). Exposures were conducted in 250 mL glass beakers with 200
mL of test solution and 10 daphnids (<24 hours old) per beaker There were two replicates for
each treatment. Test conditions throughout the experiment varied minimally (D O.: 8.8-9.0
mg/L; pH: 8.0-8.1; 21°C). No mortality occurred in the control and 100 mg. L. API O treatments.
The author-reported 48-hour EC50, based on mortality. was 221 mg/L APFO. The possible
mixture effects of APFO with other perfluoro analogue com pounds in the test material precluded
this test for quantitative use, and it was therefore retained for t|Lialilali\ e use only.
3M Company (2000) summarized a 4S-honr sialic, unmeasured acute toxicity test with
the cladoceran, / kip/inni niaifiia and APFO (CAS # 3825-20-1) The toxicant was part of the 3M
production lot mini her 11( K >1 ¦ 2< >5 and was not sufficiently characterized but was considered a
mixture of APFO (3d"., of the compound) and water (80% of the compound). The acute test
followed I Si:i\\-TSC A Guideline 7l)7 13<)i) protocol. Solutions of the test substance were made
in reconsliuiled water and included li\ e nominal test concentrations plus a control. Exposures
were conducted in 300 mL glass beakers with 250 mL of test solution and 10 daphnids (<24
hours old) per beaker. There were two replicates for each treatment. Test conditions throughout
the experiment varied minimally (D.O.: 8.1-9.1 mg/L; pH: 8.1-8.3; 20.3-20.8°C). No mortality
occurred in the control treatment and in 730 mg/L test substance treatment. The author-reported
48-hour EC50, based on mortality, was 1,200 mg/L test substance. The authors reported that the
test substance was considered a mixture of APFO and other impurities, so the EC50 does not
G-20
-------
accurately reflect the toxicity of APFO and therefore the value was not acceptable for
quantitative use, but was retained for qualitative use.
3M Company (2000) summarized a 48-hour static, unmeasured acute toxicity test with
the cladoceran, Daphnia magna and APFO (CAS # 3825-26-1). The toxicant was part of the 3M
production lot number 2327 and was characterized as mixture of APFO (<45% of the
compound), water (50% of the compound), inert perfluorinated compound (<3% of test
substance), and Cs and C7 perfluoro analogue compounds (1-2% of 1 lie compound). The acute
test followed test guidance from OECD 202. Solutions of APFO were made in well water and
included five nominal test concentrations (150, 25<). 4<)i). 6f>n and 1,000 mg/L test substance)
plus a control (well water only). Exposures were conducted in 250 mL glass beakers with 200
mL of test solution and 5 daphnids (<24 hours old) per beaker There were four replicates for
each treatment. Test conditions throughout the experiment \ aiied little (D.O.: 9.0-9.5 mg/L; pH:
8.1-8.4; 19.5-20.1°(. ) \o mortality occurred in the control and treatments <400 mg/L. The
author-reported 48-hour IX';.. Ixised 011 death/immobility, was 584 mg/L test substance. Because
of the possible mixture effects of the inert perfluorinated compounds and other perfluoro
analogue compounds the test was not acceptable for quantitative use but was retained for
qualitati\ e use
3M C oiiipiiny (2000) summarized a 21-day static-renewal, unmeasured chronic toxicity
test with the cladoceran. / kiplinia magna, and APFO (CAS # 3825-26-1), and also briefly
described a corresponding acute test with a reported 48-hour EC50 of 266 mg/L APFO. Very
little details were provided about the acute test methodology. The test compound was assumed to
be that of the chronic test, where the toxicant was part of the 3M production lot number 37 and
was characterized as mixture of APFO (96.5-100%) of the compound) and C6, C7 and C9
G-21
-------
perfluoro analogue compounds (0-3.5% of the compound). The 48-hour EC so from this test was
not used quantitatively because of missing study details and the possible presence of additional
PFAS, but the study was retained for qualitative use.
G. 2.1.4 Benthic crustaceans
Yang et al. (2014) conducted a 96-hour measured acute test of PFOA (CAS # 335-67-1,
99% purity) with the freshwater prawn species, Macrobrachinm ui/>/x>nense (a non-North
American species). Although the authors termed the test conditions "static", they also mentioned
PFOA measurements before and after renewal; based on this distinction the test was assumed to
be renewed at least once. Authors stated that the test followed ASTM E729 (ll^3) /.
nipponense (0.30 g, 4.0 cm) used for the test were purchased from the Beijing Dahongmen
Jingshen Seafood Market and allowed to acclimate for at least se\ en days before testing.
Dilution water was dechlorinated tap water (pi I. 7 t) t) 5. dissolved oxygen, 7.0± 0.5 mg/L;
total organic carbon. <> <>2 mu I., and total hardness. 190 n n 1 mg/L as CaCCte). The
photoperiod consisted of 12-hours of illumination at an unreported intensity. Aprimary stock
solution was prepared In dissoK inu PI OA in deionized water and solvent (DMSO), and
proportionally diluted with dilution water to prepare the test concentrations. Exposure vessels
were 2 I. beakers of unreported material type containing 1.5 L of test solution. The test employed
three replicates of I <> prawns each in six test concentrations (measured in low and high
treatments) plus a neuati\ e and solvent control. Nominal concentrations were 0 (negative and
solvent controls), 200.00, 300.00, 450.00, 675.00, 1,012.50 and 1,518.75 mg/L. Test
concentrations were measured in low and high treatments only. The authors provided mean
measured concentrations before and after renewal: 179.46 and 196.25 mg/L (lowest
concentration) and 1,344.28 and 1,492.75 mg/L (highest concentration). Analyses of test
G-22
-------
solutions were performed using high performance liquid chromatography with mass
spectrometric detection (HPLC/MS) and negative electrospray ionization. The concentration of
PFOA was calculated from standard curves (linear in the concentration range of 1-800 ng/mL),
and the average extraction efficiency was in the range of 70-83%. The concentrations and
chromatographic peak areas exhibited a significant positive correlation (r = 0.9987, p < 0.01),
and the water sample-spiked recovery was 99%. The temperature. DO. and pH were reported as
having been measured every day during the acute test, but results were not reported. Negative
control and solvent control mortality were 0% and 3%, respectively. The 90-hour LCso was
366.66 mg/L (95% C.I.: 253.09-531.18 mg/L). The acute value was not considered lor
quantitative use because of the uncertainties and potential exposures to PFAS from the source of
the test organisms but was retained for qutililtili\ e use by pro\ idinu relevant toxicity information.
G.2.1.5 Insects
Yang et al. (2014) performed a 96-houi" measured acule lesl of PFOA (CAS # 335-67-1,
99% purity) willi the mi due. ('/iimiioiinis phimosns. Authors stated that the test followed ASTM
E729 (1993) Although the authors termed the test conditions "static", they also mentioned
PFOA measurements before and tiller renewal; based on this distinction the test was assumed to
be renewed til least once. (pliimosns (0.05 g, 1.2 cm) used for the test were purchased from the
Beijing City liiu I'orest Flower Market and allowed to acclimate for at least seven days before
testing. Dilution wtiter was dechlorinated tap water (pH, 7.0 ± 0.5; dissolved oxygen, 7.0 ± 0.5
mg/L; total organic carbon, 0.02 mg/L; and total hardness, 190.0 ±0.1 mg/L as CaC03). The
photoperiod consisted of 12-hours of illumination at an unreported intensity. A primary stock
solution was prepared by dissolving PFOA in deionized water and solvent (DMSO), and
proportionally diluted with dilution water to prepare the test concentrations. Exposure vessels
G-23
-------
were 90 cm Petri dishes containing 10 mL of test solution. The test employed three replicates of
10 midges each in six test concentrations (measured in low and high treatments) plus a negative
and solvent control. Nominal concentrations were 0 (negative and solvent controls), 200, 300,
450, 675, 1,012.5 and 1,518.75 mg/L. Test concentrations were measured in low and high
treatments only. The authors provided mean measured concentrations before and after renewal:
196.25 and 178.48 mg/L (lowest concentration) and 1,488.4 and L.iM 97 mg/L (highest
concentration). Analyses of test solutions were performed using high performance liquid
chromatography with mass spectrometric detection (IIPLC/MS) and negali\ e electrospray
ionization. The concentration ofPFOA was calculated from standard curves (linear in the
concentration range of 1-800 ng/mL), and the average extraction efficiency was in the range of
70-83%. The concentrations and chromatographic peak areas exhibited a significant positive
correlation (r = 0.9987, p < 0.01), and the water sum pie-spiked recovery was 99%. The
temperature, D.O.. and pi I were reported as ha\ inu been measured every day during the acute
test, but results were not reported Neuati\ e control and solvent control mortality were 0% and
3%, respectively The lH->-hoiir l.('s" was 4<>2 24 mg/L (95% C.I.: 323.8-499.6 mg/L). The acute
value was not acceptable lor quantilali\ e use because of the uncertain source and unreported
previous exposure to PFAS of the test organisms but was retained for qualitative use.
G.2.1.6 ( \]>riniil fishes
Yang et :il. (2014) evaluated the toxicity of the acidic form of perfluorooctanoic acid
(PFOA, CAS #335-67-1, 99% purity) with Carassius auratus for 96-hours using measured
conditions (the authors note that the experiments followed ASTM standards and U.S. EPA
procedures for deriving water quality criteria). Although the authors termed the test conditions
"static", they also mentioned PFOA measurements before and after renewal; based on this
G-24
-------
distinction the test was assumed to be renewed at least once. The goldfish (6.0 g, 7.0 cm) were
purchased from the Beijing Chaoyang Spring Flower Market, which is considered an atypical
source. The organisms were allowed to acclimate for at least seven days before testing, and the
test was conducted at 22 ± 2°C with 12-hours of illumination, 10 fish per replicate, and three
replicates per concentration. Beakers used for exposure are assumed to be made of glass but was
not specified by study authors. PFOA was dissolved in deionizcd water and the carrier solvent
dimethylsulfoxide (DMSO) to obtain a 7 mg/mL stock solution, and then diluted with
dechlorinated tap water to yield nominal exposure concentrations of 200, 320. 512 0, 819.2,
1,311 and 2,097 mg/L PFOA. Water quality parameters reported were pH = 7.0 ') 5, dissolved
oxygen = 7.0 ± 0.5 mg/L, total organic carbon = 0.02 mg/L and total hardness = 190.0 ±0.1
mg/L as CaCCte. The supplemental data pro\ ided lor the study included a comparison of
measured PFOA concentrations before and after solution renewal in the low and high test
concentrations. PFOA concentrations in the test water did not fluctuate by more than 15% during
experiments The 9(i-liour I reported for the study, 606.6 mg/L PFOA, was not acceptable for
quantitative use hecausc the test organisms were obtained from an atypical source and previous
exposure to contaminants was unknown hut was retained for qualitative use.
The effect of PFOA (CAS #335-67-1, >98% purity) on oxidative stress enzyme responses
of C. auraius ju\ eniles (27 S5 g) was evaluated by Feng et al. (2015). The 96-hour static
measured exposure was conducted at a temperature of 23°C, pH of 7.25, dissolved oxygen of 6.5
mg/L and total hardness of 174.3 mg/L as CaC03. The fish were purchased from a local aquatic
breeding base and acclimated in dechlorinated tap water for at least for 10 days, with the total
mortality near zero. After acclimatization, five fish were randomly selected and placed in each
glass tank (two replicates for treatments and five control replicates) containing 20 L of test
G-25
-------
solution (nominal concentrations of 1.21 or 12.10 |iM PFOA, or approximately 0.5 and 5.0 mg/L
PFOA) or 20 L of dechlorinated tap water. PFOS was dissolved in DMSO to prepare a 120.77
mmol/L stock solution. The tanks were continuously aerated, and water was refreshed to
minimize the contamination from metabolic wastes. Antioxidant enzyme activity (CAT, SOD
and GPx) and lipid peroxidation were not adversely impacted at 4.931 mg/L PFOA at test
termination, but these data were not considered quantitatively because of the atypical endpoints
reported and only two exposure concentrations were evaluated. This study was retained for
qualitative use by providing non- apical endpoint which may inform mode of action and AOP
considerations.
Kim et al. (2010) evaluated the effects of PFOA (99.8% purity) to biomarker responses
exhibited by Cyprinus carpio exposed for lH->-hoiii's under flow-through, measured conditions.
PFOA stock solutions were prepared in KVIimethvllbrmamide ( <100 mg/L) and diluted with
carbon-filtered and dechloiinaled tap water to gi\ e nominal concentrations of 0.050, 0.500, 5.000
and 50.00 mg/L. Dechloiinaled tap water was used as a control. The measured exposure
concentrations of PI-OA ranged from SI to 138% of the nominal concentrations (0, 0.041, 0.483,
6.582. 55 57 nig I.), and w here appropriate, the average of the PFOA measured concentrations
was used to calculate endpoints when not within ± 20% of the nominal concentrations. The carp
were obtained from the Chungcheongnam-do Experimental Station for Inland Waters
Development (Republic of Korea) and held in 2,000 L tanks with flowing dechlorinated tap
water at 23 ± 2°C, which was also used in the study (pH, 6.9; alkalinity, 28.0 mg/L as CaC03;
total hardness, 47.8 mg/L as CaC03). Ten juvenile fish (-12 cm; -20 g) were held in each 100 L
glass exposure tank (assume one replicate per concentration) under 16-hours of illumination, and
water temperature maintained at 23 ± 1°C. At test termination, the fish were removed from the
G-26
-------
tanks and evaluated for biochemical and genetic responses. Vitellogenin activity was determined
to be the most sensitive endpoint, with a 96-hour LOEC of 6.582 mg/L. This study was not used
quantitatively due to the non-apical endpoint reported. This study was retained for qualitative use
by providing non- apical endpoint which may inform mode of action and AOP considerations.
Ding et al. (2012c, 2013) evaluated the acute effects of APFO (CAS # 3825-26-1, 98%
purity, purchased from Sigma-Aldrich) to Danio rerio embryos \ i a a %-hour static-renewal,
unmeasured exposure. Adult AB strain zebrafish were cultured in aerated and biologically-
filtered reconstituted freshwater at 26 ± 1°C. The day before a test, male and female zebrafish, at
a ratio of 1:1, were placed in spawning tanks before the onset of darkness. Mating, spawning,
and fertilization take place within 30 minutes after light onset in the morning. Eggs were
collected from spawn traps and washed with clean OF.CD water I nfertilized or abnormal eggs
were removed under a stereomicroscope. API () was dissoK cd in reconstituted water to achieve
the desired test concentrations. no sol\ cuts were used. For the toxicity test, the authors stated that
six exposure concentrations were performed with three replicates each. Graphically, seven PFOA
concentrations are show n as loui..( mol I.) concentrations. These were converted to mg/L (105.1,
204.0. 3SI (\ (->81 7. 812 i). 1238. and 1.314 mg/L, respectively) given the molecular weight of
the form of PFOA used in the study (CAS # 3825-26-1; molecular weight of 431.1 g/mol).
Response data lor the control treatment were not presented graphically or reported in the text.
Twenty fertilized euus per exposure concentration were divided into a 24-well plate with one
embryo per well, containing 2 mL test solution. The remaining four wells were filled with
control water and a single embryo. An embryo was considered dead when one of four end points
(i.e., coagulation of the embryo, non-detachment of the tail, non-formation of somites and non-
detection of the heartbeat) was observed. The survival rates were monitored and documented at
G-27
-------
72- and 96-hour post fertilization (hpf). The test solutions were half renewed every 24 hours. The
author-reported 96-hour LCso was 0.896 mM, or 386.3 mg/L APFO. The graphical presentation
of concentration-response data in the publication without including the control response was
considered problematic. EPA attempted to contact the study authors on July 13, 2021, to request
the data from the paper, but did not hear back as of October 26, 2021. Therefore, the limited
level of data presented in the paper and the apparent lack of data from the control treatment
precluded this test from being used quantitatively. The reported LCso from the study was
considered acceptable for qualitative use only.
A 72-hour exposure of PFOA (CAS # 335-07-I. l)5"n puriiy) to D. reno embryos was
conducted by Zheng et al. (2012) following OECI) (1methodology. No solvent was used
for PFOA because of its high water solubility (l->.5on mg'L). therefore, exposure solutions were
diluted from the stock solutions with embryonic water Adult wild-type zebrafish were obtained
from the Model Animal Research ("enter of Nanjing Uni\ ersity and kept in a semiautomatic
rearing system I tap water), with ll\ e females and ten males in each 10 L tank at 28 ± 1°C. Water
was exchanged at a rate of I 3 daily and the lighting was 14-hours of illumination at 1,000 lux.
Spaw ning and fertilization took place within 30 minutes after the lights were turned on in the
morning Ijnhryos were transferred to exposure solutions (reconstituted embryo water)
immediately after fertilization and examined under a stereomicroscope. Damaged or unfertilized
embryos were discarded /ebrafish embryos were exposed in 24-well cell culture plates (material
not identified) with 2 mL solution per well (pH of 8.3 ± 0.2, dissolved oxygen concentration of
6.07 ± 0.24 mg/L at the beginning and end of experiments). Twenty normally shaped fertilized
embryos were assigned to each treatment (nominal concentrations of 0, 150, 200, 212, 225, 240,
255, 270 mg/L). All concentrations were repeated in triplicate at different days with different
G-28
-------
batches of eggs. Embryos were cultured in an incubator at 28.5°C after exposure. Control
performance was reported as >80% proportion of normal embryos. The reported 72-hour LCso
and 96-hour ECso (malformations) were 262 and 198 mg/L PFOA, respectively. However, the
data were not considered in a quantitative manner because the duration was too short for an acute
exposure and the data were instead considered qualitatively acceptable
Du et al. (2013) investigated the effect of PFOA (CAS " .Vo-(->7-l, unreported purity) on
the survival, malformation and suppression of steroidogenic enzyme synthesis of D. rerio
embryos exposed via renewal unmeasured conditions for 120-hours. PFOA slock solutions were
prepared in DMSO at a concentration of 1 M and stored at -2<~>°C. They were diluted to desired
concentrations in culture medium immediately before use, and the final concentration of DMSO
in the culture medium did not exceed 0.1".. (\ \ ). Wild-type adult male and female zebrafish,
obtained from the Model Animal Center of Nanjing I ni\ ersity. w ere maintained on a 14-
hour: 10-hour light dark cycle at 2S (' under semi-static conditions with charcoal-filtered water.
Spawning was induced in the morning when the lights were turned on. Fertilized eggs were
collected 3<~> minutes later and examined under the microscope. Only those that had developed
normally uere selected IjiiIhyos uere incubated with reconstituted embryo medium in Petri
dishes lor subsequent experiments. Zebrafish embryos at four hours post-fertilization were
exposed (three replicates, 4<) fish per replicate) to 0.100, 0.200 and 0.500 mg/L PFOA and
0.001% DMSO (control) at 28°C, with daily renewal of the embryo medium. Embryo survival
and stage of embryonic development were recorded daily until test termination (120 hours post-
fertilization). The five-day NOEC and LOEC for the increase in relative mRNA expression of
hhex and pax eight were reported as 0.100 and 0.200 mg/L, respectively. Since only non-apical
G-29
-------
endpoints were reported, and the five-day exposure period, these data were not classified as
acceptable for quantitative use but retained for qualitative use.
A six-day static unmeasured test was utilized by Ulhaq et al. (2013) to determine the
toxicity of PFOA (CAS # 335-67-1, purity not reported) to D. rerio. PFOA stock solutions were
freshly prepared in reconstituted water in concentrations below the limit for water solubility.
Adult zebrafish (AB strain) were held in charcoal-filtered tap water. Breeding groups including
three males and two females were placed in 10 L glass aquaria equipped with spawning nets
separating the parental fish from the eggs. A half an-hour after the onset of lights the eggs were
collected, rinsed for removal of debris, and then only normally developed fertilized eggs at least
in the four-cell stage were selected. The zebrafish embryos were then (within 15 minutes after
collection) exposed to a series of concern rat ions of the test substance dissolved in reconstituted
water (exposure medium). Fertilized eggs (lour-cell stage) were randomly distributed
individually into ilat bottom. 4N-well polystyrene plates along with 750 |iL of the exposure
medium. PFOA was tested at six consecutive nominal concentrations differing by a factor of 3.3
based on logarithmic scale lilting (3-1 .noo mg I.) For each test, four 48-well plates were used,
with a total of 24 embryos per concentration as well as 24 in the water control group. Each
treatment group was equally distributed to each of the four well plates (i.e., six
embryos/coiiccniralion/plale). giving a total of 168 embryos. The plates were covered with
parafilm, and the embryos were exposed to the chemical until 144-hour post fertilization (hpf).
Fish laboratory conditions throughout the study were kept at pH 7.2-7.6, a water temperature of
26 ± 1°C and a light cycle of 14-hours. Observations of mortality and sublethal endpoints were
made after 24, 48, 120 and 144 hours post-fertilization using a stereomicroscope. Sublethal
endpoints such as presence of edema, malformations, not-hatched eggs, lack of circulation and
G-30
-------
reduced pigmentation were also observed. Heart rate was recorded at 48 hours post-fertilization
and hatching time was determined using time-lapse photography. The 144-hour LCso was 430
mg/L PFOA, and the EC so (lethal and sublethal effects) was 350 mg/L PFOA. Neither value was
used in a quantitative manner because of test duration, but both were considered in a qualitative
manner.
Bouwmeester et al. (2016) tested various chemicals, including PFOA (CAS 335-67-1)
in single-chemical exposures to evaluate the potential of I), rerio embryos as a screening
tool to examine DNA methylation modifications after xenobiotic exposure. Emlnyos were
exposed from 0 to 72 hours post fertilization. In the single PFOA chemical tested (i e , 320 |iM,
which converts to 132 mg/L based on the molecular weight of PFOA (414.07 |ig/|iM) authors
reported no embryonic effects. The reported \OI-C of 132 nig I. was not acceptable for
quantitative use because the test did not measure acute apical effects and the test duration was
too short. This study was retained lor qualitath e use.
PFOA (CAS 335-07-I. purity) acute exposure toD. rerio embryos was evaluated
by Kalaseknr el sil. (2015) \ ia a 72-liour static unmeasured exposure. PFOA was dissolved and
diluted in DMSO to make a I.ooik stock solution. Adult wild type (TAB 14) zebrafish were
maintained in 3 5 L tanks in a Tecniplast system supplied continuously with circulating filtered
water at 2S 5 C with 14-houis of illumination. Embryos were harvested after spawning and
allowed to develop in a Petri dish at 28.5°C containing reconstituted E3 media. At two days post
fertilization (dpf), a clutch of 20 embryos were transferred into each well of a six-well plate
containing 4 mL of E3. The embryos were exposed to PFOA dissolved in dimethylsulfoxide
(DMSO) or to DMSO alone (vehicle-control). At five dpf, the larvae were transferred to 96-well
plates (one embryo/well in approximately 100 |iL E3), manually imaged on an Olympus 1X51
G-31
-------
inverted fluorescence microscope using a 4x objective, and images captured using an Olympus
XM10 camera with CellSens Dimension vl.9 software. Exposures were not performed directly
in 96 well plates because the small volume could result in motility restriction, thereby causing
decreased yolk absorption. The larvae were not anaesthetized, and thus swam in an upright
position, facilitating image capture of the ventral side of the larvae. Chemical exposure
experiments for all concentrations were repeated two to four limes, with one to two sets of 20
embryos per experiment. To determine LCso values and morphological malformations, a clutch
of 20 embryos were exposed at two dpf, and lethality and morphology were scored at five dpf.
The 72-hour LCso was reported as 3.8 x 10"4 M PFOA (I 57 3 mg/L PFOA) and was not
classified as quantitatively acceptable due to the short test duration but was retained for
qualitative use.
Rainieri et al. (2017) evaluated the acute effects of PFOA (purchased from Acros, Geel,
Belgium) on zebrafish (I kinio tvrio) in a 4K-hour static measured study. A 2 mg/ml stock
solution was prepared l\v dissoK inu PFOA in methanol and storing the stock solution in darkness
at <4°C until use W ild type fish were obtained from AZTI Zebrafish Facility and maintained in
60 L tanks at 27 (' and a 12-hour light '12-hour dark cycle, and were fed twice daily with
commercial feed Embryos were held in embryo water for 72 hours before testing. Twenty-five
hatched embryos (72 hpf) were exposed to 10 mL of test solution in glass petri dishes 6 cm in
diameter at 27°C under a 12-hour: 12-hour light-dark photoperiod for 48 hours. Triplicate
exposures ranged from 10 to 500 mg/L with a maximum of 0.45% DMSO in any exposure.
Samples of each exposure solution were taken at the beginning and at the end of the test to
determine PFOA concentrations. The reported 48-hour LCso value was >500 mg/L PFOA, but
G-32
-------
the value was acceptable for qualitative use only because of the short test duration (i.e., only 48
hours rather than the established 96-hour acute exposure period for this species).
Vogs et al. (2019) evaluated the acute effects of PFOA (>96% purity, CAS # 335-67-1,
0.002 mg/L solubility at 25°C) on zebrafish (Danio rerio) embryos in a 118-hour measured,
static-renewal study. AB strain fish used in this study were provided by the Zebrafish Core
Facility at Comparative Medicine, Karolinska Institute. Three mule and three female adults were
grouped, and embryos were collected in E3 medium directly after spaw ninu Study authors
reported following OECD TG 236 for fish embryo toxicity testing. A stock solution was
prepared by dissolving PFOA into dimethyl sulfoxide to achieve initial measured concentrations
of 21, 41 and 340 |iM PFOA. Thirty embryos (two hpf) were placed in 30 mL of E3 exposure
medium in 50 mL glass petri dishes maintained at 28 1 I °C under dark conditions throughout the
exposure. A 118-hour ECso value of 210.8 mu I. (5<>*¦•> iiM ) PFOA was reported for mortality, and
the EC20 for non-inllated swim Madders, pericardial and yolk sac edemas, and scoliosis was
147.2 mg/L (355 6 u\l) The I 18-hour cumulative EC?o was qualitatively acceptable because
118-hour exposure duration was longer than standard 96-hour exposure duration prescribed by
OECI) Test No 23o l-'ish I-mlnyo Acute Toxicity (FET) Test. Additionally, true replication was
not used because authors conducted replicates at different times (confirmed by through personal
communication with the corresponding author [C. Vogs] on 9/3/2021).
Dasgupta ol al. (2020) evaluated the acute effects of PFOA (CAS # 335-67-1, 96%
purity, purchased from Acros Organics) on zebrafish (Danio rerio) via a 66-hour unmeasured,
static study. A stock solution was prepared with either DMSO or NaOH and stored in 5 mL glass
vials and kept at room temperature. The working solution (50 mM PFOA) was freshly prepared
by spiking stock solutions into water derived from the recirculating water system. Adult,
G-33
-------
wildtype (5D) zebrafish were maintained and bred in the same water system. Eight embryos (six
hpf) were incubated and exposed to 10 mL of either a solvent control or 50 |iM PFOA until 72
hpf at 28°C under a 14-hour: 10-hour light-dark photoperiod. At test termination there was no
significant effect of survival or development on zebrafish embryos. The 66-hour NOEC of 20.70
mg/L (50 |iM) PFOA, based on survival, was acceptable for qualitative use only due to the short
exposure period (i.e., 66-hour exposure instead of the established lM-hour acute exposure for this
species).
Pecquet et al. (2020) conducted acute exposures with zebrafish embryos lo determine the
LCso of PFOA, and to examine sublethal exposures of PFO A on neutrophil migration in response
to wounding. A stock solution of PFOA (1 mg/L) was prepared in DMSO and diluted with fish
water containing 60 mg/L Instant Ocean sails I11 the acute I .("*¦¦ tests. 2d embryos (one hpf) were
exposed to one of seven PFOA concentrations (<)-1 .odd mu |. PFOA) with three replicates for
each test concent rati 011 llach test was replicated li\ e times and results were pooled. Tests
followed the OI 1(1) l-'ish Fmlnyo Toxicity Test Guideline 236, and the U.S. EPA Fish Early-life
Stage Toxicity Test Guideline 2 in Fxposurcs were static and lasted for 48 hours at 28°C. In a
separate experiment, embryos (one hpf) were exposed for 24 hours to 0, 0.5 or 5.0 mg/L PFOA
for 24 hours and checked for mortality and deformity. Measured PFOA concentrations were
0.685 and (¦> I (¦> mu I. PFOA for the test treatments and 0.089 mg/L PFOA was detected in the
DMSO control. Afterwards PTU (l-phenyl-2-thiourea) (0.003%) was added to the remaining
viable embryos to inhibit melanin formation (pigmentation), and the embryos were incubated for
an additional 24 hours in their respective PFOA or control treatments. Fish were subsequently
wounded and allowed to recover for three hours to facilitate neutrophil recruitment. The reported
48-hour LCso was 300 mg/L PFOA, but the test duration was too short (i.e., only 48-hours rather
G-34
-------
than the established 96-hour acute exposure period for this species) to be used quantitatively. In
the separate experiment, neutrophil migration was significantly increased in the 0.685 mg/L
PFOA exposure compared to the unwounded control fish. Results of this publication were
acceptable for qualitative use.
Wasel et al. (2020) reported the results of toxicity tests with the zebrafish, Danio rerio,
and PFOA in either whole larvae (in vivo) or as zebrafish embryonic cell line (in vitro). The
results of the in vivo tests are summarized from this study since they represent apical endpoints
with whole animals. PFOA (CAS # 335-67-1, >99% purity, purchased from Siuma-Aldrich, St.
Louis, MO) stock solution was prepared in reverse osmosis water and diluted with embryo
medium to make test concentrations ranging from 10-9,000 mg/ L PFOA. Two acute toxicity
tests were run, one where pH in test solutions was not buffered and another where test solutions
were buffered with NaHC03 to pH 7. Wild-type zebrafish from in-house cultures were used to
supply embryos (fi\ e to six hours post fertilization, hpf) lor testing. Exposures were conducted in
culture plates with one embryo per well and 20 embryos per plate. For each experiment there
were at least three plates for each treatment I-mhryos were observed for development until 96
hpf under test conditions (solutions renewed daily, 28±1°C, and a 14-hour: 10-hour light:dark
photoperiod) The 90-91 hour LCsos reported by the authors were 57.6 and 487.4 mg/L PFOA,
for the unbuffered and buffered test solutions, respectively. Based on the starting age of the
organisms, the acute test was too short to be used quantitatively, so values were acceptable for
qualitative use only, especially since other acute quantitatively acceptable tests for this species
were available.
3M Company (2000) summarized a 96-hour static, unmeasured acute toxicity test with
the fathead minnow, Pimephalespromelas, and APFO (CAS # 3825-26-1). The toxicant was part
G-35
-------
of the 3M production lot number HOGE 205 and was not sufficiently characterized but was
considered a mixture of APFO (30% of the compound) and water (80% of the compound). The
acute test followed USEPA-TSCA Guideline 797.1400 protocol. Solutions of the test substance
were made in reconstituted water and included five nominal test concentrations (530, 830, 1330,
2100, and 3330 mg/L test substance) plus a control. Exposures were conducted in 20 L glass
tanks with 15 L of test solution and 20 juvenile fish per tank. There were two replicates for each
treatment. Test conditions throughout the experiment ranged as follow s DO 5.7-9.2 mg/L; pH:
7.2-7.9; and temperature: 21.8-22.5°C. No mortality occurred in the control and treatments with
<830 mg/L test substance. The author-reported 96-hour I .Cwas 2,470 mg/L test substance. The
authors reported that the test substance was considered a mixture of APFO and other impurities,
so the LCso does not accurately reflect the toxicity of APFO and therefore the value was not
acceptable for quantitative use but was retained lor qualitati\ e use.
3M Conipsmv (20(H)) reported the results of two lH->-hour static, unmeasured acute
toxicity test with the fathead minnow. I'imcphalespromelas and PFOA (CAS # 335-67-1). The
toxicant was part of the 3\l production lot number 269 and was characterized as mixture of
PFOA (lM 5-1 do" n of the compound) and C7 and C9 perfluoro homologue compounds (0-
3.5% of the compound). The lust test lacks important test details including the dilution water,
stock preparation, source of organisms, number of organisms per replicate and exposure vessels.
Six nominal test concentrations (control, 50, 125, 250, 375 and 500 mg/L PFOA) were used,
with each test treatment only replicated one time. The measured dissolved oxygen in the control
and 375 mg/L test concentration was also low; ranging from 3.8-5.7 over the test period.
Therefore, the authors questioned the health of the test fish, even though there was 100%
survival at the end of the test period except the highest concentration with zero survivors. The
G-36
-------
author-reported 96-hour LCso for the test was 440 mg/L, but the lack of test details and the
possible mixture effects of other perfluoro homologue compounds did not make the value
acceptable for quantitative use. In the second acute test, with the same PFOA compound, authors
dissolved the test substance in a 50:50 water:isopropanol solution to make a primary solution of
1,000 mg/L test substance and isopropanol. Another toxicity test conducted by the authors
showed no mortality or sublethal effects at 500 mg/L isopropanol on the same species. The stock
solution was then diluted with reconstituted water to make five nominal test concentrations (130,
220, 360, 600 and 1,000 mg/L test substance or 65, 1 10, 180, 300 and 500 mu I. PI 'OA) plus a
control (reconstituted water only). The test followed USF.PA-TSCA Guideline 1^1 1400.
Exposures were conducted in 20 L glass aquaria with 15 L of test solution and 10 fish per tank
(0.23 g/L loading). There were two replicates for each t realm en l The pH of the three highest test
solutions was very low at test initiation (3.0-4 3) u liieh would adversely affect fish survival. No
mortality occurred in the control and treatments with <1 I n mu.L PFOA. The 96-hour reported
LCso was 280 mg' I. test substance and isopropanol or 140 mg/L PFOA. The test was not
acceptable for quantilali\ e use because of the possible mixture effects of other perfluoro
homologue compounds and the low pi I of test: solutions. Results of both tests described here
were retained for qualitath e use.
3M Com puny (20(H)) reported the results of a 96-hour static, unmeasured acute toxicity
test with the fathead minnow. I'imephalespromelas and APFO (CAS # 3825-26-1). The toxicant
was part of the 3M production lot number 83 and was characterized as mixture of APFO (96.5-
100% of the compound) and C6, C7 and C9 perfluoro analogue compounds (0-3.5% of the
compound). No specific test protocol was identified. The test lacks important test details
including the stock preparation, source of organisms, number of organisms per replicate and
G-37
-------
exposure vessels. Six nominal test concentrations (control, 10, 20, 30, 40 and 50 mg/L APFO)
were made in carbon-filtered well water. Each test treatment only replicated one time. The
measured dissolved oxygen in the control and 50 mg/L test concentration was low; ranged from
4.0-5.9 over the test period. There was 90% survival in the control at test termination, 100%
survival at <40 mg/L and 80% survival at the highest test concentration (50 mg/L). The authors
extrapolated graphically to estimate a 96-hour LCso of 70 mg,'[. API () This test was not
acceptable for quantitative use because of the lack of replicates and lack of observed effects in
the test, as well as the possible mixture effects of other perfluoro analogue compounds. This test
was retained for qualitative use only.
3M Company (2000) reported the results of a 96-hour static, unmeasured acute toxicity
test with the fathead minnow, Pimephales piomdas and API O (CAS " 3825-26-1). The toxicant
was part of the 3M production lot number 37 and was characterized as mixture of APFO (96.5-
100%) of the compound) and ('- and C9 perfluoro analogue compounds (0-3.5%) of the
compound). The lest followed I SI-PA 660/3-75-009 (1975) test protocol. Preparation of the
stock was not pio\ ided. but the dilution water was noted as carbon filtered well water. The acute
test included six nominal test concentrations (0, 560, 650, 750, 870 and 1,000 mg/L APFO).
Exposures were conducted in glass aquaria with 16 L of test solution and 12 fish per tank (0.5
g/L loading) There were two replicates for each treatment. Test conditions throughout the
experiment ranged by the following: D.O.: 4.0-5.0 mg/L; pH: 7.9-8.0; temperature: 19°C. The
authors report a 96-hour LC50 for each replicate (776 and 754 mg/L), with an average LC50 of
766 mg/L APFO. Results of this test were not acceptable for quantitative use because of possible
mixture effects of other perfluoro homologue compounds and the lack of true replicates and
exposure details. Results of this test were retained for qualitative use only.
G-38
-------
3M Company (2000) summarized a 96-hour static, unmeasured acute toxicity test with
the fathead minnow, Pimephalespromelas, and APFO (CAS # 3825-26-1). The toxicant was part
of the 3M production lot number 390 and was characterized as mixture of APFO (78-93% of the
compound) and Cs, C6 and C7 perfluoro analogue compounds (7-22% of the compound). No
specific test protocol was identified. Solutions of APFO were made in carbon filtered well water
and included five nominal test concentrations (100, 180, 320, 560. aiicl 1000 mg/L test substance)
plus a control (well water only). Exposures were conducted in 4 L glass honkers with 3 L of test
solution and six fish per beaker (0.6 g/L fish loading). There were two replicates lor each
treatment. Test conditions throughout the experiment varied little (D.O.: 5.6-7.4 mu I,; pH: 7.7-
8.0; 20°C). No mortality occurred in the control and APFO treatments <180 mg/L. The author-
reported 96-hour LC50 was 301 mg/L APFO. liccause of possible mixture effects of other
perfluoro analogue compounds, this test was not acceptable for quantitative use but was retained
for qualitative use.
3M Company (2000) summarized a 96-hour static, unmeasured acute toxicity test with
the fathead minnow. I'imcphalcspromdas. and PFOA (CAS # 335-67-1). The toxicant was part
of the 3\l production and was characterized as mixture of PFOA (95-98%) of the compound) and
perfluorochemical inert compounds (I -5% of the compound). No specific test protocol was
identified. A stock of PFO A w as made by dissolving the test substance with NaOH and diluting
the stock with carhon-liltered well water to make five test concentrations (690, 750, 810, 870 and
930 mg/L) plus a control (well water only). Exposures were conducted in glass beakers with 5 L
of test solution and five fish per beaker (0.5 g/L fish loading). There were two replicates for each
treatment. Test conditions throughout the experiment varied little (D.O.: 6.1-7.7 mg/L; pH: 7.5-
7.7; 19-20°C). No mortality occurred in the control and PFOA treatments <750 mg/L. The
G-39
-------
author-reported 96-hour LCso was 843 mg/L PFOA. Because of possible mixture effects of other
perfluoro analogue compounds, this test was not acceptable for quantitative use but was retained
for qualitative use.
3M Company (2000) summarized a 96-hour static, unmeasured acute toxicity test with
the fathead minnow, Pimephalespromelas, and APFO (CAS # 3825-20-1). The toxicant was part
of the 3M production lot number 2327 and was characterized as mixture of APFO (<45% of the
compound), water (50% of the compound), inert perfluorinated compound ( 3% of test
substance), and Cs and C7 perfluoro analogue compounds (1-2% of the compound I. The acute
test followed test guidance from OECD 203. Solutions of APFO were made in well water and
included five nominal test concentrations (150, 250, 400, 600, and 1,000 mg/L test substance)
plus a control (well water only). Exposures were conducted in I I. glass aquaria with 15 L of
test solution and 10 fish (juveniles) per aquaria (<> 3<) g I. fish loading). There were two replicates
for each treatment Test conditions throughout the experiment \ aried little (D.O.: 6.1-9.2 mg/L;
pH: 7.4-8.4; 21.H-22 o (') \o mortality occurred in the control and any treatment involving the
test substance. The author-reported lH->-hour I .("*¦¦ was >1,000 mg/L test substance. Because of
possible mixture effects of other periluoro analogue compounds, this test was not acceptable for
quantitiitix e use but was retained for qualitative use.
Topmouth gudgeons. 1'seudorasboraparva, were exposed to PFOA (99% purity) for 96-
hours by Yang ol al. (2014) \ ia a static measured exposure (the authors note that the
experiments followed ASTM standards and USEPA procedures for deriving water quality
criteria). The topmouth gudgeon (4.0 g, 4.0 cm) were purchased from the Beijing Chaoyang
Spring Flower Market, which was considered an atypical source. The organisms were allowed to
acclimate for seven days before testing, and the test was conducted at 22 ± 2°C with a light:dark
G-40
-------
cycle of 12-hour: 12h-hour, with 10 fish per replicate and three replicates per concentration.
Beakers used for exposure were assumed glass but was not specified by the study authors. PFOA
was dissolved in deionized water and carrier solvent DMSO to obtain a 7 mg/mL stock solution,
and then diluted with dechlorinated tap water to yield nominal exposure concentrations of 100,
140, 196, 274.4, 384.2 and 537.8 mg/L PFOA. Water quality parameters reported were pH = 7.0
± 0.5, dissolved oxygen = 7.0 ± 0.5 mg/L, total organic carbon n <)2 ing/L and total hardness =
190.0 ± 0.1 mg/L as CaCCte. The supplemental data provided for the study includes a comparison
of measured PFOA concentrations before and after solution renewal in the low and high acute
and chronic test concentrations. PFOA concentrations in the test water did not lliictuate by more
than 15% during experiments. The 96-hour LCso reported for the study was 365.0 mg/L PFOA,
which was not used quantitatively because of the uncertainties and potential exposures to PFAS
from the source of the test organisms. The test was instead acceptable for qualitative use.
G.2.1.7 Adriank hi In n k ic fis/h-s
The effects of PI OA (CAS 335-07-1, purity not provided) on swim bladder
development of Oryzia* Iaii/h-s was in\ estimated by Godfrey (2017). The stock solution was
prepared In dissoKinu PI OA in I I. of ie\ eise osmosis water containing 12.5 mL Replenish,
with the pi I adjusted to neutral (7-7 5) The stock solution was then diluted to obtain 4.7 mg/L
PFOA (only one concentration evaluated) based on data from a previous study. Adult, see-
through Japanese niedaka SK2MC strain were maintained in a controlled recirculating system,
with a 14-hour: 10-hour light:dark photoperiod and a temperature of 25 ± 1°C. Adult fish were
fed ad libitum twice daily with a combination of hatched Artemia nauplii and commercial food
(Otohime). Adult fish were bred by placing them in breeding tanks (1:1 male to female ratio) in
an environmental chamber with the same conditions described earlier. Fertilized eggs (<6-hours
G-41
-------
post-fertilization) were collected from the bottom of the tank or gently brushed off from females.
Eggs were then immediately disinfected in a 0.005% bleach solution and moved to six-well
plastic plates containing 10 mL of the designated exposure or control media and no more than
five embryos per well. Test solutions were fully changed every other day. The total length of the
exposure lasted until 48-hours post-hatch. Fish were sexed with aid of leucophores along the
body axis of males only prior to hatching. The LOEC for swim Madder development was 4.7
mg/L PFOA, which was not used quantitatively because only one concentration was tested;
however, the tests results were retained for qualilali\ e use
G.2.1.8 Centrarchidae fishes
3M Company (2000) summarized two %-hour sialic, unmeasured acute toxicity tests
with the bluegill sunfish, Lepomis macrochinis. and APFO (CAS 3825-26-1). The toxicant was
part of the 3M production lot number 83 and was characterized as mixture of APFO (96.5-100%
of the compound) and ('- and ('<; perfluoro analogue compounds (0-3.5%) of the compound).
No specific test protocol was identified in either acute test. In the first test, solutions of APFO
were made in carbon filtered well water and included five nominal test concentrations (135, 180,
240, 32<). and 42<) mu I. test substance) plus a control (well water only). Exposures were
conducted in lo L tanks with 20 fish per tank. There were no replicates for each treatment. Test
conditions throughout the experiment varied little (D.O.: 5.1-6.9 mg/L; pH: 7.8-8.0; 18-19°C).
No mortality occurred in the control and treatments with <240 mg/L of test substance. The
author-reported 96-hour LCso was >420 mg/L. Since only one fish died in the highest test
treatment, the experiment was re-run with higher test treatments (0, 420, 560, 750, 1000, and
1350 mg/L test substance). The 96-hour LCso for the second test was 569 mg/L test substance.
G-42
-------
Both effect concentrations were not acceptable for quantitative use because of possible mixture
effects of other perfluoro analogue compounds, but they were retained for qualitative use.
G.2.1.9 Amphibians
Tornabene et al. (2021) conducted an acute PFOA (purchased from Sigma Aldrich,
Catalog # 171468-25G; purity not provided) toxicity tests with the gray treefrog, Hyla
versicolor. The acute test followed standard 96-hour acute toxicity test guidance (U.S. EPA
2002; ASTM 2008, 2017). Frog egg masses were collected from a field in the wetlands of
Indiana near the campus of Purdue University. Collected egg masses were raised outdoors in 200
L polyethylene tanks filled with well water. The experiments began when frogs reached Gosner
stage 40. Before test initiation larvae were acclimated to test conditions (21°C and 12-hour: 12-
hour light:dark photoperiod) for 24 hours A slock solution of PI-OA (2,000 mg/L) was made in
UV-filtered well water and diluted with well water to leach test concentrations (ranged from 0-
2,000 mg/L PF( )A) Test concentrations were not measured in test solutions based on previous
research that demonstrated limited degradation under similar conditions. Larva were transferred
individually to 25<) ml. plastic cups with 2<)() ml. of test solution and were not fed during the
exposure period There were six to se\ en replicates for each treatment and two of the seven frogs
died in the controls (i.e., 2S'\. control mortality). The author-reported 96-hour LCso was 191
mg/L PFO A. w liich was not acceptable for quantitative use because of high control mortality but
was retained for qualitati\ e use. Note, the authors also reported a quantitatively acceptable test
for the same species (Gosner stage 26) that is described in A.2.11.
Yang et al. (2014) evaluated the acute toxicity of acidic form of perfluorooctanoic acid
(PFOA, CAS #335-67-1, 99% purity) to the Asiatic toad, Bufo gargarizans via 96-hour renewal
measured exposures (the authors note that the experiments followed ASTM standards and U.S.
G-43
-------
EPA procedures for deriving water quality criteria). The tadpoles (0.048 g, 1.8 cm) were
purchased from the Beijing Olympic Park, which is considered an atypical source. The
organisms were allowed to acclimate for seven days before testing, and the test was conducted at
22 ± 2°C with a light:dark cycle of 12-hours: 12-hours, with 10 tadpoles per replicate and three
replicates per concentration. Beaker material used for exposure was not specified by study
authors. PFOA was dissolved in deionized water and carrier sol\ enl I)VISO to obtain a 7 mg/mL
stock solution, and then diluted with dechlorinated tap water to yield nominal exposure
concentrations of 35, 56, 89.60, 143.36, 229.38 and 367 mg/L PFOA. Water quality parameters
reported were pH = 7.0 ± 0.5, dissolved oxygen 7 i) <> 5 mg/L, total organic carbon = 0.02
mg/L and total hardness = 190.0 ±0.1 mg/L as CaCOi. The supplemental data provided for the
study includes a comparison of measured PI OA concentrations before and after solution renewal
in the low and high acute and chronic test concentrations PIOA concentrations in the test water
did not fluctuate by more lluin I 5".. during experiments The ^6-hour LCso of 114.74 mg/L
PFOA was not used quanlilali\ ely due lo the atypical test organism source but was retained for
qualitative use
G.2.2 Summary of Chronic PI OA Toxicity Studies Used Qualitatively in the Freshwater
Aquatic I .il'e Crilerion Derivation
G.2.2.1 Worms
Yuan ol al. (2016b) conducted a 10-day renewal, unmeasured test on PFOA (96%
purity) with the planarian. / higesia japonica (a non-North American species). The test organisms
were originally collected from a fountain in Quanhetou Boshan, China, and cultivated in the
laboratory for an unspecified time period before use. Dilution water was aerated tap water. No
details were provided regarding photoperiod or light intensity. A primary stock solution was
prepared by dissolving the salt in dimethyl sulfoxide (DMSO). The control and exposed
G-44
-------
planarians received 0.005% DMSO (v/v). Exposure vessels were beakers of unreported material
type and dimensions and a 50 mL fill volume. The test employed three replicates of 10
planarians each in five test concentrations: 0 (solvent control), 0.5, 5, 10, and 20 mg/L PFOA.
The test temperature was reported as 20°C. No other water quality parameters were reported as
having been measured in test solutions. Survival of solvent control animals was not reported. No
apical endpoints were measured as the study focused on neural genes expression and neuronal
morphology in the planarian. The lowest test concentration, 0.5 mg/l.. decreased the mRNA
expression levels of neural genes DjFoxD, DjotxA and DjotxB. Due to a lack of apical
endpoints, insufficient test duration, and uncertainties associated with the source of the test
organism, the LOEC was not acceptable for quantitative use but was retained for qualitative use.
Yuan et al. (2017) conducted another I <>-day static, unmeasured test on PFOA (purity
not provided) with the planarian, Dugesia lapomca Manx of the exposure details were similar to
the previous experiment The lest organisms were collected from a Quanhetou stream (Zibo,
China), and culli\ ated in the laboratory lor two weeks before use. IntactD. japonica (>1 cm)
were starved for se\ en days before exposures Dilution water was aerated tap water. No details
were pro\ ided regarding pholoperiod or light intensity. A primary stock solution was prepared
by dissoK ing the salt in dimethyl sulfoxide (DMSO). The control and exposed planarians
received 0 < )< )5" n DMSO (\ \) Exposure vessels were glass tanks of unreported material,
dimension, and 111 I \ oliime The test employed three replicates of 100 planarians each in five
nominal test concentrations: 0 (solvent control), 0.5, 5, 10, and 20 mg/L PFOA. The test
temperature was reported as 20°C. No other water quality parameters were reported as having
been measured in test solutions. Survival of solvent control animals was not reported. No apical
endpoints were measured as the study focused on stress responses. The lowest test concentration,
G-45
-------
0.5 mg/L, exhibited elevated lipid peroxidation and increased mRNA expression levels of HSP40
and HSP70, two stress response genes. Due to a lack of apical endpoints and insufficient test
duration, the LOEC was not acceptable for quantitative use but was retained for qualitative use.
G.2.2.2 Planktonic crustaceans
3M Company (2000) summarized a 21-day static-renewal, unmeasured chronic toxicity
tests with the cladoceran, Daphnia magna, and APFO (CAS ?" 3X25-20-1) The toxicant was part
of the 3M production lot number 37 and was characterized as mixture olWPI () (96.5-100% of
the compound) and C6, Ci and C9 perfluoro analogue compounds (0-3.5% of the compound).
The test followed USPEA (1982) and OECD (1997) test protocols. Solutions of API O were
made in carbon filtered well water and include six test treatments (5.0, 8.0, 13, 22, 36, and 60
mg/L APFO) plus a control. Exposures were conducted in 25<) ml. glass beakers with 200 mL of
test solution and five daphnids per beaker. There were lour replicates for each treatment and each
test solution was renewed e\ cry two days. Over the 21 day exposure period none of the daphnids
died in the controls. Inn K<)% of daphnids died in the highest test treatment. Reproduction was
also affected by API O with a signilicanl decrease in the mean live young per adult at >36 mg/L
APFO at lest termination The 21 -day YOEC and LOEC, based on reproduction and survival
were 22 and 3o mg/L APFO. respecti\ ely, with a corresponding MATC of 28.14 mg/L. This test
was acceptable lor qualita1i\ e use only because of the possible mixture effects of other perfluoro
analogue compounds. Note as part of this publication the authors reported a 48-hour EC50 of 266
mg/L APFO but provide very little details about test methodology. Because of possible mixture
effects of other perfluoro analogue compounds, this test was not acceptable for quantitative use
but was retained for qualitative use.
G-46
-------
Seyoum et al. (2020) evaluated the chronic effects of PFOA (CAS# 335-67-1, >99%,
purchased from Sigma) on Daphnia magna neonates via a 21-day unmeasured, static-renewal
study. The study authors did not report following any specific protocol. D. magna ephippia were
purchased from MicroBioTests Inc. (Belgium) and were activated by rinsing in tap water.
Ephippia were hatched by incubating at 20-22°C for 72 to 90 hours in standard freshwater under
a continuous light intensity (6,000 lux). Newly hatched neonates ( 24-hours old) were fed a
suspension of Spirulina micro-algae two hours before testing. Nominal concent rations of 0
(control), 1, 10 and 25 |iM (or 0 (control), 0.4141, 4.141, and 10.35 mg/L gi\ en the molecular
weight of the form of PFOA used in the study, CAS # 335-67-1. of 414.07 g/niol) were prepared
by mixing the respective amounts of PFOA in dimethyl sulfoxide (DMSO). Ten <24-hour old
neonates, exposed in triplicate, were placed into 25<~> ml. crystallization dishes with 100 mL of
test solution. A mean temperature of 23 = I °(\ dissoK ed oxygen of 8 to 9 mg/L, total hardness of
250 mg/L as CaCO;. pi I of 7 5 < > 25 and salinity of 0.< )2".. u ere reported in the exposure water.
D. magna were fed a mixture ol'Spirnlma microalgae and yeast (Saccharomyces cerevisiae)
daily during the test, and 5<»" <» ofthelest solution was changed every other day. Neonates were
counted daily and remo\ ed The 21 -day reproductive (fecundity) LOEC of 1 |iM, or 0.4141
mg/I. PI OA was reported In the study authors, where a -38.25% reduction in mean number of
daphnids relatix e to the control was observed. EPA was unable to fit a model with significant
parameters to the reproduction-based concentration-response data due to a lack of clear
concentration-dependent effects beyond the LOEC. The reproduction-based LOEC (i.e., 0.4141
mg/L) was selected as the chronic value from this test; however, it was not considered acceptable
for quantitative use because chronic responses in this test did not display concentration-
dependent effects beyond the LOEC despite a 25X increase in treatment concentrations.
G-47
-------
Moreover, additional ECio values from other, quantitatively acceptable tests, were available to
inform the chronic sensitivity of Daphnia magna.
G.2.2.3 A quatic insects
MacDonald et al. (2004) conducted a 10-day renewal, unmeasured range-finding test of
PFOA (>97% purity) with the midge, Chironomus dilutus. PFOA data were limited since no
effects were observed at the highest treatment concentration (1 nn.ooo ug/L). Limited details
were provided about the range-finding test. Authors stated that the test followed the general
guidance given by EPA-600-R99-064 (USEPA 2002) and ASTM E 1706-00 (ASIVT 2000).
These were methods for measuring the toxicity and bioacciimulation of sediment-associated
contaminants with freshwater invertebrates and have different exposure durations than those
typically considered for aqueous exposures, as well as different control survival requirements
and recommendations. C. dilutus used for the test were I "-day old larvae from in-house cultures.
Dilution water was reconstituted hard water according to AS I'M (2002), but specifics were not
provided. Temperature and DO concentrations were remained within acceptable ranges (21.0-
23.0°C; D.O >5 <> mu I.) The pholopcriod consisted of 16-hours of illumination, at an
unreported intensity A primary stock solution was proportionally diluted with dilution water to
prepare the lest concentrations Exposure vessels were 250 mL polypropylene beakers containing
240 mL of test solution and 5<> mL of clean cultured sand. Assuming the range-finding test
followed that later dellniti\ e test with PFOS; the PFOA test employed at least two replicates of
10 midges each in unmeasured PFOA concentrations that ranged from 1-100 mg/L PFOA. No
effect on survival or growth were observed at the highest test concentration (100 mg/L PFOA) so
the authors did not follow up with a definitive test. The 10-day NOEC of 100 mg/L, based on
G-48
-------
survival and growth, was not acceptable for quantitative use but was retained for qualitative use
by providing toxicity information on the relative tolerance of insects to PFOA exposure.
McCarthy et al. (2021) conducted a 10-day sub-chronic PFOA (97% purity, purchased
from Sigma-Aldrich) test on the midge, Chironomus dilutus. The PFOA stock solution was
dissolved in reconstituted moderately hard water without the use of a solvent stored in
polyethylene at room temperature until use. The 10-day exposure was considered a range finding
test for a follow up 19-day duration test (which is further described in (' 2 MM and included
concentrations spaced by ~100x with only mortality measured. Exposure \ essels were 1 L high-
density polyethylene beakers containing natural-field collected sediment with (¦»< > nil. of sediment
and 105 mL of test solution. PFOA in test solutions was added via pipette to the beakers with the
tip just above the sediment substrate. Nominal test concentrations were 0, 0.05, 26, 2,600, 26,000
|ig/L. Egg cases were obtained from Aquatic liios\ stems or I S(iS Columbia Environmental
Research Center and held lor l<) days. Each test beaker held 12 organisms with five replicates
per exposure treatment Solutions were renewed e\ cry 48-72 hours in the 10-day exposure.
Water samples of test concentrations were measured on day one and day 10 and measured test
concentrations ranged from of nominal Based on nominal concentrations, the author-
reported NOI -C and LOEC (endpoint = mortality) values were 26 and 272 mg/L PFOA,
respecti\el\ Results of this test were not used quantitatively because the exposure duration was
too short for a chronic test and too long for an acute test; however, results were retained as
qualitatively acceptable for use.
Stefani et al. (2014) conducted a chronic (10 generation) test of PFOA (form and purity
not reported) with a midge, Chironomus riparius. The 10 generations (each approximately 20 to
28 days) were tested under static conditions. Authors stated that the test followed OECD 218
G-49
-------
(2004c), OECD 219 (2004b), OECD 233 (2010), and a specific protocol for multigenerational
assays using C. riparius developed and published by Nowak et al. (2006, 2007a, 2008, 2009) and
Vogt et al. (2007b, 2007c, 2010). The specific protocol for multigenerational assays was
designed to highlight neutral evolutionary responses caused by exposure to contaminants. A
native population collected in the Lambro River (Milan, Lombardy, Ttaly) was used as a starting
population for the test. C. riparius used to initiate the test were I.I (I list instar) larvae. Dilution
water was reconstituted water according to U.S. EPA (2000); test hardness was not specified, pH
7.8-8.2. Photoperiod was 16-hours:8-hours, light:dark with an intensity 5
-------
because they fulfilled the validity criteria for survival according to OECD guideline 218 (OECD,
2004a). In the control group, most vessels in all generations reached the emergence of at least
70% individuals. There were no significant effects on mutation rate, emergence, reproduction, or
sex ratio. The NOEC for the study was 0.0089 mg/L PFOA. The results from this study were not
acceptable for quantitative use because of the lack of details pertaining to the characteristics of
the sediment used in the exposure.
In a companion study to Stefani et al. (2014), Marziali et al. (2019) similarly conducted
a chronic (10 generation) test of PFOA (form and purity not reported) with the midge,
Chironomus riparius. The test was done under static conditions for 10 generations, each
approximately 36 days (or 1/10 of this year-long, 10 generation test). The test followed OECD
218 and 233 (OECD 2004, 2010), with slight variations. C. riparius used for testing were from
in-house cultures originating from a native population collected in the Lambro River (Milan,
Lombardy, Italy). C. riparius used to initiate the test were first instar larvae. Dilution water was
reconstituted water according to U.S. EPA/600/R-711 99/064 (U.S. EPA 2000); the hardness was
not specified, pH 7.8-8.2. The photoperiod included 16-hours of unspecified illumination.
Authors tested a single treatment (0.010 mg/L nominal) and solvent control. Authors stated there
were two replicates per treatment. PFOA was dissolved in pure methanol (>99%) in order to
achieve stock solutions at 1 g/L of PFOA. Each stock solution was then diluted in reconstituted
water in order to achieve a nominal concentration of 0.01 mg/L. Exposure vessels were glass
tanks (19 cm x 19 cm x 18 cm) containing 1 L of test solution and 1 cm of formulated sediment
(75%) of the volume aquarium quartz sand and 25%> of sterilized natural sediment). The reported
time-weighted measured concentration was 0.0098 mg/L and PFOA was found primarily in the
water column. Water temperature, dissolved oxygen and pH were measured every three to five
G-51
-------
days in two to three replicates per treatment. Test temperature was controlled at 20.1 ± 0.7°C,
dissolved oxygen remained >66% saturation, and pH stayed within the range of 7.8-8.2. Each
generation test was considered valid if emergence in the control was >70% in at least six
replicates (i.e., vessels) of the 10 included. Emergence in the control groups by generation was as
follows: 88 (primary emphasis for criteria development), 71, 53, 61.6, 78.6, 91.9, 62, 53.5, 79.1,
75.5. Generations one, two, five, six, nine, and 10 met control survival acceptability. The LOEC
based on developmental time, adult weight, was 0.0098 mg/L (time-weighted average; NOEC
and MATC <0.0098 mg/L). Marziali et al. (2019) reported effects to select generations. Overall,
however, effects were sporadic with reductions in growth observed in several generations. There
were no effects on "survival, development, or reproduction" and Marziali et al. (2019) concluded
"no effects at population level (population growth rate) were proved, thus a toxicity risk in real
ecosystems at the tested concentrations seems unlikely." The results from this study were not
acceptable for quantitative use because of limited test concentrations assessed, and uncertainty
pertaining to sediment characteristics, and poor control survival in four of the 10 generations.
G. 2.2.4 Salmonul Jishcs
The chronic toxicity of periluorooctanoic acid (PFOA, purity not reported) to
Oncorhynchiis mvkiss via a dietary exposure was evaluated by Tilton et al. (2008). Two separate
experiments were performed I) A tumor experiment involved feeding 10-week post-hatch fry
200 or 1,800 ppm PI OA lor six months (30 weeks); and 2) A microarray experiment involved
exposing 12-18-month-old juvenile fish trout diets containing 500 or 1,800 ppm PFOA for 14
days. PFOA in diets was not measured in either experiment, and tissues were not measured to
ensure body burdens reflected doses used. Mt. Shasta strain rainbow trout were hatched and
reared at the Oregon State University Sinnhuber Aquatic Research Laboratory in 14°C flowing
G-52
-------
well water on a 12-hour: 12-hour light:dark cycle. In the tumor experiment, approximately 1,000
fry were initiated at 10 weeks post-hatch with an aqueous exposure to 0.01 ppm aflatoxin B1
(AFB1) for 30 min. Sham-exposed trout were exposed to vehicle alone (0.01% ethanol) and
served as noninitiated controls for each treatment. After initiation, fry were fed Oregon Test Diet
(OTD), a semi-purified casein-based diet, for three months (Lee et al 19Q1) Trout were then
randomly (within initiator group) divided into experimental treal menl groups (140
animals/treatment) and fed experimental diets containing 200 or 1,800 ppm PFOA, 1,800 ppm
clofibrate (CLOF), or 1,800 ppm dehydroepiandrosterone (DHEA) ad libitum (2 S-5.6% body
weight) five days/week for six months, a protocol similar to that previously described for DHEA
(Orner et al. 1995). The PFOA concentrations in the diet for 200 and 1,800 ppm were equivalent
to 5 and 50 mg/kg/day, respectively. Diets were prepared monthly and stored frozen at -20°C
until 2-4 days prior to feeding, when diets were allowed to thaw at 4VC. At nine months post-
initiation, juvenile fish were euthanized by deep anesthesia with 250 ppm tricaine methane
sulfonate and sampled lor li\ er tumors over a two-day period. Livers were fixed in Bouin's
solution for two to se\ en days for histologic identification and examination of tumors with
hematoxylin and eosin In the microarray experiment, juvenile trout, 12-18 months of age, were
maintained in separate 375 I. tanks (three tanks) for each treatment, with five fish per tank.
Animals \a ere led a maintenance ration (2.8% wet weight) of OTD. Administration of
experimental diets containing 500 or 1,800 ppm PFOA, 1,800 ppm CLOF, 750 ppm DHEA, 5
ppm E2, or 0.1% dimethyl sulfoxide (DMSO) vehicle control was carried out for 14 days. The
concentrations of 17P-estradiol (E2) and DHEA were chosen based on their ability to maximally
induce vitellogenin (VTG) and/or act as hepatic tumor promoters in trout (Nunez et al. 1989). On
day 15, fish were euthanized by deep anesthesia with 250 ppm tricaine methane sulfonate.
G-53
-------
Approximately 100 mg liver tissue from individual fish was minced, stored in TRIzol Reagent
and quick-frozen in liquid nitrogen for gene expression analysis. The rest of the liver was quick-
frozen in liquid nitrogen for enzyme assays. The 10-week MATC (liver somatic index) and 6-
month MATC (palmitoyl CoA P-oxidation-liver enzyme) were both 600 mg/kg PFOA diet
(NOEC = 200; LOEC = 1,800 mg/kg). The test was not acceptable for quantitative use based on
the non-apical test endpoints. This study was retained for qua1ilali\ e use by providing non-apical
endpoints which may inform mode of action and AOP considerations
Benninghoff et al. (2011) evaluated the chronic effects of PFOA (C AS ,vi5-67-l,
purchased from Sigma-Aldrich in St. Louis, MO) on rainbow trout (Oncorhynchns mykiss)
juveniles in a 15-day static, unmeasured study. Mount Shasta strain rainbow trout were hatched
and reared in the Sinnhuber Aquatic Research I .ahoratory al Oregon State University in
Corvallis, OR. Fish were maintained at 12VC and a 12-hour 12-hour light:dark cycle in a 375 L
tank filled with carbon filtered lap water. Two weeks before testing, fish were fed a semipurified
casein-based dietwilh menhaden oil al a rate of 2% body weight. A stock solution was prepared
by dissolving PFOA in dimethyl sulfoxide (DMSO) that was then added to oil in the fish diet to
prepare a nominal concentration of 250 ppm PFOA wet weight. Eleven-month-old fish weighing
approximately 70 g were placed in treatment groups, six fish per group, and were fed either 0.1,
1 or 5 mg/kg body weight/day PFOA laced food five times per week for 15 days: correlating to a
diet concentration of 5. 5<) or 250 ppm PFOA per day. Four replicates were included for each
dietary treatment, along with a negative control group, and a vehicle control group (treated
dietarily with 0.5 ppm DMSO). A positive control group (treated dietarily with 5 ppm estradiol)
was also included with an additional twelve fish. Fish were sacrificed and weighed on day 15.
The 50 and 250 ppm PFOA test diets caused a significant increase in plasma vitellogenin (Vtg)
G-54
-------
levels. The lack of description of the dilution water and the test methodology (dietary exposure)
rendered this study unacceptable for quantitative use and was retained for qualitative use only.
Benninghoff et al. (2012) also evaluated the chronic effects of PFOA (CAS No. 335-67-
1, analytical grade purchased from Sigma Aldrich in St. Louis, MO) on rainbow trout
(Oncorhynchus mykiss) in a ~9-month unmeasured study. Mount Shasta strain rainbow trout
were hatched and reared in the Sinnhuber Aquatic Research Laboratory at Oregon State
University in Corvallis, OR. Fish were raised in a 375 L tank filled with carbon filtered tap water
and maintained at 12°C and a 12-hour: 12-hour light:darl< cycle. Fry (10-15 weeks post spawn)
were exposed to a cancer-causing agent for 30 minutes, then fed a semi purified casein-based
diet for one month. Fish were fed experimental diets containing 2,000 ppm PFOA
(approximately 50 mg/kg body weight/day) li\ e-days per week lor a period of six months. Fish
were sacrificed at test termination (12.5 months post spaw n) and examined for tumor presence.
The dietary PFOA irealmenl significant increased tumor multiplicity and size. The lack of
description of the dilution water and the test methodology (dietary exposure) deemed this study
unacceptable for quanlilati\ e use and was retained for qualitative use only.
Atlantic salmon (Sa/nio sulur) embryos were evaluated by Spachmo and Arukwe (2012)
via a 52-day llow-through unmeasured exposure to PFOA (95% purity). Eggs were obtained
from Lundamo I latcheries, Norway (Aquagen) and transported to the Norwegian University of
Science and Technology Centre of Fisheries and Aquaculture in Trondheim, Norway. The eggs
were kept in plastic tanks (25 L) at 5-7°C with filtered, re-circulating and aerated water.
Approximately one-third of the water volume was changed once per week. At age 404-679 day
degrees (dd: number of days multiplied by degree Celsius), the eggs and larvae were exposed to
PFOA (0.10 mg/L). PFOA was dissolved in methanol (carrier solvent: 0.01%) and the control
G-55
-------
group was exposed to the carrier solvent only. Hatching occurred at 20 calendar days after start
of exposure, at an effective developmental age of 504 dd, after which a riverbed environment
was simulated by tank bed gravel and continuous water flow. Fish sampling was performed at
21, 35, 49 and 56 calendar days after exposure, or at respective developmental age of 549, 597,
679 and 721 dd. The exposure was terminated at 679 dd, and 712 dd represents the end of a one-
week exposure-free recovery period. Thus, day 49 sampling was performed 24-hours after
terminating the exposure and no exposure related differences in hatching rale were observed. The
52-day growth NOEC and LOEC were 0.10 and >0.10 mg/L PFOA, respecti\ ely These data
were deemed acceptable for quantitative use based on meeting data quality objecti\ es; however,
were not used in deriving the chronic criterion because the study only included one treatment
group that showed no adverse effects. Because the one treatment group that showed no effects
was a relatively low treatment concentration, including this NOI-C value in the criterion
calculation would ha\ e resulted in the criterion magnitude being strongly influenced by the low
test concentration selected by study investigators (that did not produce an adverse response),
rather than a concentration-response relationship
G.2.2.5 ('yprmiil fishes
C>i:i ri el sil. (2016) e\ aluated the effects of PFOA (96% purity) on tissue accumulation
and histological alterations of specific organs of adult Cyprinus carpio. Thirty-one two-year-old
common carp were purchased from a local fish farm and acclimated according to OECD 305 test
guidelines (1996). Four weeks prior to the start of the experiment, fish were transferred into the
test tanks to adapt to the exposure environmental conditions. The fish were fed pelleted feed
three times per week with feeding stopped two days before fish were killed. Feeding was carried
out manually to ensure rapid and complete consumption. Uneaten feed and feces were removed
G-56
-------
from the tank to minimize PFOA sequestration by these organic substances. Fish were placed
into each of three 120 L glass aquaria filled with tap water; with a continuous supply of fresh
water provided at a flow-through rate of 500 mL/min. PFOA stock solutions were continuously
dispensed and diluted to deliver the tested concentrations to the tanks over a period of 56 days.
Exposures were conducted at 0.0002 mg/L (n = 10) and 2 mg/L (n = 11) levels based on
environmental reports (Loos et al. 2008, 2009) and on reported experimental data, respectively
(Oakes et al. 2004; Wei et al. 2007). A control group of 10 carp were held in kip water only. The
stock solutions were prepared by dissolving PFOA in distilled water and delix ered into the
treatments tanks by a peristaltic pump at a flow rate of <">.42 ml ./min. At time zero of exposure,
an initial volume of the stock solution was added to the treatments tanks to immediately achieve
the desired exposure concentration. Water parameters were monitored and recorded three times
weekly for temperature (10-15°C), pH (6.7<>-S and oxygen saturation (>80 %). PFOA
concentration in the lest water from each test tank was measured by liquid chromatography-mass
spectrometry three times during the exposure period. The results of these analyses indicate that
PFOA concentration was maintained at Si) of each nominal concentration. At the end of the
56-day exposure, the fish were anesthetized with MS-222, killed by a deep cut through the neck,
and dissected Sex was recorded, and animals were measured for total length, body wet weight,
and liver and gonad weight. Condition factor, hepato-somatic index, and gonado-somatic index
were also calculated No mortality was observed in the controls or either treatment level. The 56-
day LOEC (PCNA-positive hepatocyte abundance) was reported as 2.0 mg/L PFOA, which was
not used quantitatively because only two exposure concentrations were tested, lack of
replication, and the test initiated with insensitive two-year-old adults. Results were retained for
qualitative use.
G-57
-------
The chronic toxicity of PFOA to Cyprinus carpio was also evaluated by Manera et al.
(2017) under the same conditions as Giari et al. (2016). Stock solutions were prepared by
dissolving PFOA (96% purity) in distilled water. Thirty two-year-old carp (19.3 ± 2.5 cm; 104.8
± 27.8 g) were obtained from a local fish farm, randomly divided into three groups/tank (two
PFOA treated tanks and one unexposed tank) of 10 fish each. Fish were acclimated for four
weeks before starting the experiment and treated according to Ol-('l) (2<> 12) test guidelines. The
carp were fed with a commercial pellet food (Tetra Pond Pellets Mini) three li nies per week at
2% of their total body weight. Waste and uneaten food were removed regularly A How-through
exposure test was conducted for 56 days by a system that continuously delivered PFOA to the
test tanks to maintain concentrations of 0.0002 mg/L or 2 mg/ L. Test tanks were 120 L glass
aquaria filled with a continuous supply of lap water at a flo^-illrough rale of 500 mL/min. The
tested concentrations were selected, respecti\ el\. on the hasis (if environmental reports (Loos et
al. 2008, 2009) and experimental data from the literature (Oakes et al. 2004; Road et al. 2007;
Wei et al. 2007) The slock solution was deli\ ered into the treatments tanks by a peristaltic pump
at a flow rate of n 42 ml. mi n At time zero of exposure, an initial volume of the stock solution
was added to the treatments tanks to immediately achieve the desired exposure concentration.
Water parameters were monitored and recorded three times weekly for temperature (10-15°C),
pH (6.70-8.OD), niul oxygen saturation (> 80%). At the end of the 56-day exposure, the fish were
anesthetized with MS-222. pithed, dissected and sexed. In each group of carp, the sex ratio was
approximately 1:1. The 56-day LOEC (liver biomarkers) was 2 mg/L PFOA, which was not used
quantitatively because only two exposure concentrations were tested, lack of replication, and the
test initiated with insensitive two-year-old adults. Results were retained for qualitative use.
G-58
-------
The chronic effects of PFOA on the zebrafish, Danio rerio, have been reported by
numerous researchers. However, all data available for this species were classified as qualitatively
acceptable. Hagenaars et al. (2013) exposed D. rerio adults to PFOA (purity 96%) under
renewal unmeasured conditions for 28 days. Adult zebrafish were obtained from a commercial
supplier (Aqua hobby, Heist-op-den-berg, Belgium) and acclimated for four weeks prior to
treatment in aerated medium-hard reconstituted freshwater (OF.CI) 2d;,) at 26 ± 0.5°C. Fish were
subjected to a 14-hour light cycle and fed daily with Sera flakes at a rate of 2".. of their mean
body weight. After acclimation, fish were exposed to nominal concentrations of <) I, 0.5 and 1
mg/L PFOA for four and for 28 days while the control fish were kept in clean ^ titer Separate
experiments with an identical setup were performed for four and 28 days. Every 48-hours, the
water was totally renewed by water with the correct nominal PI-OA concentrations. For both
experiments, three different 25 L aquaria per exposure concenirtilion were used with each
aquarium containing eight nuile tiiul eight female zebrafish Alter respectively four and 28 days,
fish were decapitated and dissected The li\ ers of six male fish and six female fish were pooled
separately per aquarium and snap frozen in liquid nitrogen. The livers of fish exposed for 28
days to <>. o | and I mg I. PI OA were used for transcriptomic and proteomic analyses. Whole
bodies of all exposure concentrations were used for biochemical analyses (28 day) as well as for
the determination of PFOA concentrations (four and 28 days). Based on the results of the first
experiment, a second experiment was conducted to study the mitochondrial dysfunction caused
by PFOA in more detail. The activity of the mitochondrial electron transport chain was
measured. Male zebrafish were exposed to 1 mg/L PFOA for 14 days using the same exposure
conditions as in the previous experiments. Three biological replicates were used for both the
control and the PFOA exposed fish, with each aquarium containing 18 males. Males were chosen
G-59
-------
in this experiment as they accumulated more PFOA than females. The liver was dissected to
assess the electron transport chain (ETC) activity. ETC activity was also measured at the
beginning of the experiment (time point zero) in three biological replicates consisting of 18
males each. The 28-day LOEC (reproduction-based endpoints, including fecundity, fertility, and
hatching) was >1 mg/L PFOA, which was not used quantitatively because the test was initiated
with a relatively tolerant adult life stage. A 28-day LOEC (non-apical alterations of gene
transcripts) of 0.1 mg/L PFOA was also reported but was not used quaniiuui\ ely due to the non-
apical endpoint. Results of this study were, therefore, only considered quuliiaii\ ely acceptable.
Truong et al. (2014) evaluated the sub-chronic effects of 1,060 compounds (U.S. EPA
ToxCast phase 1 and 2) on zebrafish, Danio rerio, through the use of high-throughput
characterization of multidimensional in vivo effects The test design and results of the toxicity
test with APFO and PFOA are highlighted here A slock solution of ALJFO and PFOA were
made with 100% DMSO al a concentration of 2" niM (final DMSO concentrations was 0.64%
vol/vol) and diluted with standard em Inyo medium to make one of six test concentrations (0,
0.0064, 0.064. i) M. (•> 4 and M tiM) lor each chemical of interest. One zebrafish embryo (six
hpf) from in-house cultures was placed i ncli \ idually in a well along with 90 |iL of test solution in
a 96-uell culture plate. Thiri\ -two replicates were used for each test treatment. The effects of
APFO and PI OA on mortality, growth, behavior morphology, histology and physiology were
observed until 12<) hpf (I 14-hour test duration) with the water quality conditions not reported.
The most sensitive endpoint was mortality with a reported LOEC of 0.064 |iM APFO, or
0.02759 mg/L APFO, based on a molecular weight of 431.1 g/mol for APFO. There were no
effects of PFOA on mortality for zebrafish embryos with a reported NOEC of 64 |iM PFOA or
26.50 mg/L PFOA based on a molecular weight of 414.07 g/mol for PFOA. This test was not
G-60
-------
used quantitatively and retained for qualitative use only because the exposure durations were too
long for an acute test and too short for a chronic test.
Zhang et al. (2014a) exposed adult D. rerio to PFOA (CAS # 335-67-1, 96% purity) for
21 days via renewal unmeasured exposure conditions. PFOA was dissolved directly in dilution
water to obtain nominal concentrations (likely prepared a stock solution first). Adult wild-type
zebrafish were obtained from a local fish dealer and acclimatized lor three weeks prior to
treatment in well-aerated tap water at 27 ± 0.5°C. Fish were subjected to a 12-hour light/dark
cycle and fed live bloodworms and fish flakes (Tetramin) twice a day. After acclimation, fish
were exposed to nominal concentrations of 0.05, n 1.05 and I mg/L PFOA. Separate
experiments with an identical setup were performed. Every 24-hours, aquarium water was
completely replenished with a water solution containing the correct nominal PFOA
concentrations. For three experiments, five different (•> I. aquaria per exposure concentration were
used with each aquarium containing 15 male and I 5 female zebrafish. After 21 days, fish were
decapitated and dissected The spleens from 1 5 male and 15 female fish were pooled and frozen
in liquid nitrogen The 21 -day M.VI'C (decrease in inflammatory cytokines, IL-1B and IL-21, in
spleen) was o i)7t)7 mg I. PI OA (\OI-(" <) < i5, LOEC = 0.1 mg/L). The chronic value was not
acceptable for quantitative use because exposure was initiated with of adult fish and only non-
apical endpoints were reported Consequently, results of this study were only retained for
qualitative use.
The effects of PFOA (CAS # 335-67-1, unreported purity) on Danio rerio thyroid
disruption and subsequent swim bladder development was investigated by Godfrey et al.
(2017b) Stock solutions were prepared by dissolving PFOA in 1 L of reverse osmosis water
containing 12.5 |iL Replenish (Seachem Laboratories Inc.) and then adjusted to neutral pH (7-
G-61
-------
7.5). The stock solution was then diluted to 1% of the respective LCso based on data from a
previous acute test (96-hour LCso = 473 mg/L PFOA) (Godfrey et al. 2017a). Thus, the tested
exposure concentration was 4.7 ppm PFOA. Adult zebrafish, AB wild-type, were maintained at a
water temperature of 28 ±1°C and a photoperiod of 14-hours L: 10-hours D. Fish were fed twice
daily, Artemia nauplii in the morning and Tetramin in the afternoon, and genders were kept
separate overnight at a ratio of two males to one female. Embryos were collected (gastrula stage,
4.5-hpf) and randomly placed into Petri dishes containing 25 mL of test solution which was
renewed daily throughout the duration of the exposure Each Petri dish contained 20 embryos
and each test consisted of a minimum of three replicates per dose with experiments repeated
three times. In order to cover the complete period of sw i m Madder development, zebrafish
embryos were exposed starting immediately after fertilization either suhchronically for six days
(zero to six days post fertilization, dpf) or chronically for 2K days (zero to 28 dpf). For the
subchronic exposures, embryos were maintained in Petri dishes for six days, after which they
were imaged, and llash frozen for qPCR analysis I-'or the chronic exposures, larvae were moved
after six days to a 5<)i) ml. glass mason jar containing 200 mL solution. Fish were not fed during
the sub-chronic exposures since they rely on their yolk sac until swim-up. From six to 14 dpf
larvae were led ml libitum paramecin once a day, and from 15 to 28 dpf larvae were fed Artemia
nauplii in the morning and Tetramin in the afternoon. Embryos and larvae were maintained in an
environmental chamber at a temperature of 28 ± 1°C and a photoperiod of 14-hour: 10-hour
light:dark. The 28-day LOEC (swim bladder development) was 4.7 mg/L PFOA, which was not
used quantitative since there was only one concentration tested; however, this value was retained
for qualitative use.
G-62
-------
Jantzen et al. (2017b) evaluated the effects of PFOA (purity not provided) on the
morphometries behavioral and gene expression in I). rerio exposed via five-day static
unmeasured exposures (OECD Method 212, 2011). The AB strain of zebrafish (Zebrafish
International Resource Center, Eugene, OR) were used for all experiments. Breeding stocks were
bred and housed in recirculating systems under a 14-hour light cycle. System water was obtained
by carbon/sand filtration of municipal tap water and water quality was maintained at a pH
between 7.2 and 7.7, and water temperature between 26 and 28UC. Zehralish embryos were
exposed at 3-hpf to PFOA at concentrations of 0, 0.02, 0.2 or 2.0 [jJVl (or U. n 02. t).20, and 2.0
mg/L as reported by the authors) for 120-hours (four replicates. 30-38 fish per replicate). All
compounds were dissolved in water. After this time, fish were transferred to non-treated system
water and fed two times daily with Zeigler I .ar\ al AP50 Therefore, the only exposure was
through the water from 3-hpf to 120-hpf ( li\ e days), w liieh corresponds to an embryonic to
yolk sac larval exposure At 12<)-hpf, morphometric measurements were recorded, and gene
expression analyzed Morphometry measurements were also taken at 7-dpf and 14-dpf. At 14-
dpf, gene expression data and swim acti\ ity endpoints were collected. Each treatment compound
and corresponding control group was set up as indi\ idual experiments, and the sample size was
dependent on number of emlnyos produced from the stock breeding sets. No experiment had
mortality greater than 20% of the starting sample size. The five-day (plus nine-days for
observation) chronic \ alue lor growth based apical endpoints, including body length, was a
MATC of 0.6325 mg/L (NOEC = 0.2 mg/L; LOEC = 2.0 mg/L). A MATC for swimming
activity, a non-apical endpoint, was also reported as 0.06325 mg/L (NOEC = 0.02 mg/L; LOEC
= 0.2 mg/L). The reported chronic values based on growth and swimming activity were not
considered quantitatively acceptable because of the relatively brief chronic (i.e., five-day)
G-63
-------
exposure duration compared to other acceptable acute exposures that indicated D. rerio was
tolerant to brief (i.e., 96-hours) PFOA exposures.
Annunziato (2018) investigated the cellular and behavioral alterations of zebrafish
exposed to PFOA (purity not provided, purchased from Sigma Aldrich) for 333 hours. Zebrafish
from in-house cultures were maintained at a pH of 7.2-7.7, temperature of 27±1°C and a 14-
hour: 10-hour light:dark cycle, and were fed twice daily a diet of. 1 ricniia in the mornings and
aquatox/tetramin flake mix in the evenings. A stock solution of PFOA (2.<)()() uM) was prepared
in egg water. Twenty-five embryos (3-hpf) were exposed under static conditions lo one of four
nominal treatments (0, 0.02, 0.2 and 2.0 |iM PFOA) for five days and then transferred to control
water for morphometric analysis of stained cartilage and yolk membranes of hatched larvae at 14
dpf. Control survival was >85%. At 14 dpi" the stained neural area in larvae was less than the
control fish and the normalized neural strain inlcnsilv u as greater at the two highest test
concentrations. The 333-hour (117 hour plus nine-day ohser\ alion) NOEC and LOEC based on
morphology were 0 <>2 and <> 2 u\l PI-OA. respectively; or 0.008281 and 0.08281 mg/L PFOA
based on a molecular weight of 414 <>7 g mol The study was acceptable for qualitative use only
because of the i elati\ ely short test duration lor a chronic test.
Slinckcns et al. (2018) report the results of 12 chemicals using an adverse outcome
pathway testing strategy on xehrafish, Danio rerio. Twenty-four viable embryos (wild type),
from in-house cultures, were transferred to polystyrene 24-well plates (one embryo/well; two
mL/well). The first column of each plate consisted of a negative control (reconstituted water),
resulting in 20 exposed embryos per plate and four control embryos per plate. Four plates were
used for the median effect estimation (e.g., LCso). Seven nominal test concentrations (0, 10, 50,
100, 250, 500 and 1000 mg/L PFOA) were used to determine the LCso of zebrafish held under
G-64
-------
static conditions for 168 hours. No solvent was used in stock preparation. The 168-hour LCso
was 362.5 mg/L PFOA, which was not acceptable for quantitative use because of the atypical
exposure duration (too long for an acute test and too short for a chronic test). The test was
retained for qualitative use only.
Zhang et al. (2021) summarized the immunotoxicity of PFOA via the NF-kP pathway in
zebrafish, Danio rerio, kidney. PFOA (CAS # 335-67-1, 95% purity) purchased from Sigma-
Aldrich (St. Louis, MO) was mixed with dechlorinated tap water to create li\ e nominal test
concentrations (0, 0.05, 0.1, 0.5 and 1 mg/L PFOA) /ebrafish from a commercial supplier
(specifics not provided) were acclimated for seven days in glass aquaria. Over the course of the
21-day exposure, fish were fed daily in a 12-hour: 12-hour light dark photoperiod and test
solutions were renewed daily. At test termination 4<"> fish were sampled from each treatment and
kidneys were excised. The specific number of replicates and number of organisms per replicate
was not provided. PI OA kidney concentrations increased with increasing exposure durations.
Additionally, as PI'OA concentrations increased the kidneys became enlarged and the color
faded. The mRY\ expression le\ els of 11A and I L-ip peaked when zebrafish were exposed to
0.1 mg I. PI OA. increasing In I5<)"„ and 170%, respectively, compared to the control. The
mRN A expression level was reduced when the PFOA concentration was higher than 0.1 mg/L.
The mRNA expression le\el of IL-ip was lower than the control group at 1 mg/L PFOA. The
21-day LOEC of <> I nig I. (endpoint = mRNA gene expression levels in the kidney) was not
acceptable for quantitative use because it was a non-apical endpoint, but it was retained for
qualitative use.
3M Company (2000) summarized a flow-through, unmeasured chronic toxicity tests
with the fathead minnow, Pimephalespromelas, and APFO (CAS # 3825-26-1). The toxicant
G-65
-------
was part of the 3M production lot number 83 and was characterized as mixture of APFO (96.5-
100% of the compound) and C6, Ci and C9 perfluoro analogue compounds (0-3.5% of the
compound). The test followed USEPA (1972) proposed chronic test protocols. Solutions of
APFO were made in aerated well water and included six test treatments (0, 6.2, 12.5, 25, 50, and
100 mg/L APFO). Exposures started with 60 eggs (48 hours post fertilization) sourced from U.S.
EPA (Duluth, MN) in oscillating eggs cups. Each test treatment hud two test replicates (aquaria).
After hatch, 40 fry from each cup were placed in their respective aquarium and exposures
continued for an additional 30 days under flow-through conditions. Fry were led two to three
times per day. Test conditions throughout the test were D O >95% saturation, pi I 7 d-7.3 and
temperatures of 25 ± 1°C. At test termination there were 110 significant effects on egg
hatchability, and fry survival and growth The 3<)-day post hatch \OF.C of 100 mg/L APFO,
based on all test endpoints, was not acceptable lor quantilali\ e use because of the possible
mixture effects of other peril uoro analogue compounds and lack of observed chronic effects.
Results of this test were retained lor qualitative use only.
3M Conipnnv (2000) summarized the uptake/depuration of APFO (CAS # 3825-26-1)
by the fathead minnow. I'liiicplnik-spromelas, in a 13-day sub-chronic exposure. The toxicant
was characterized as a mixture of API O (96.5-100%) of the compound) and C6, C7 and C9
perfluoro analogue compounds (0-3.5%) of the compound). The exposure included one test
concentration of API O (25 mg L) dissolved in carbon-filtered well water plus a control.
Exposures were static and included a 13-day uptake phase followed by a 15-day depuration
phase containing no PFOA. Thirty fathead minnows (64 days old) were obtained from Aquatic
Biosystems (Fort Collins, CO) and held in five gallon high density polyethylene tanks containing
15 L of test solution (0.1246 g/L fish loading). No treatments were replicated. Measured
G-66
-------
concentrations of APFO ranged from 25.0-25.9 mg/L in test solutions and <1 mg/L in control
solutions. Fish were fed daily and five fish were sampled on day eight and day 13 in the uptake
phase and days one, four, seven, and 10 in the depuration phase for whole body APFO
concentrations. Average APFO measured in whole bodies of fathead minnows was 46.7 |ig/g wet
weight at test day 13, with a calculated BCF of 1.8 L/Kg. Authors reported no effects of survival
and growth in the single treatment concentration evaluated, resulting in a N'OEC of >25 mg/L.
Results of this study were not acceptable for quantitative use because of possible mixture effects
of other perfluoro analogue compounds, the lack of replicates, and because of the relatively short
exposure duration. Results of this test were retained for qualitative use only.
Oakes et al. (2004) exposed fathead minnows to lJFOA in an outdoor microcosm
experiment. The University of Guelph Microcosm Facility is located at the Guelph Turfgrass
Institute (ON, Canada) and consists of 30 artificial ponds of approximately 12,000 L. The
microcosms were constructed below grade to a depth of 1.2 meters using galvanized steel panels
lined with food-grade polya inylchloride. Each microcosm had a diameter of 3.9 meters, was
filled with water to a depth of approximately I meter and was flush with ground level. The water
supply for the microcosms was an irrigation pond supplied by a well located on site. Sediment
trays containing a 1:1:1 (v/\ \ ) mixture of sand, loam, and organic matter, as well as potted
macrophytes ( \ lyriophylhtm spicalum) were added to each microcosm. Prior to being treated
with PFOA, water was circulated among all microcosms for two weeks at a flow rate of
approximately 12 m3/d, ensuring homogeneous water chemistry, zooplankton, and algae
assemblages. PFOA (19.4% wet-weight aqueous solution from the 3M Company; however,
purity of PFOA that constituted the 19.4 wet weight was not reported) was added to the
microcosms once by subsurface injection. Water samples from each microcosm were obtained
G-67
-------
using a metal depth-integrating water-column sampler at one-hour and one, two, four, seven, 14,
21, 28 and 35 days after PFOA addition to calculate time-weighted mean PFOA concentrations.
Microcosms were treated in triplicate at nominal concentrations of 0.3, 1.0, 30 and 100 mg/L
PFOA (mean time weighted average concentrations of 0.27, 0.65, 23.9 and 74.1 mg/L PFOA).
Three additional microcosms served as controls and did not receive any PFOA. Fathead
minnows were purchased from Silhanek Baitfish Farms (Bobcaueon. ON'. Canada) and
acclimated in the adjacent irrigation pond for 10 days prior to PFOA exposure (under a natural
photoperiod). Breeding pairs were held in two wooden frames with 5 mm aperture
polyvinylchloride mesh cages. Each microcosm held two cages, with each PFOA concentration
replicated in three microcosms. Cages were divided into four quadrants, and each quadrant
contained a single breeding pair for a total of I (¦> fish per microcosm Fish were initially sexed
prior to exposure based on size and presence of secondary sex characteristics. Sexes were
subsequently confirmed al the conclusion of the exposure alter the fish were killed. A 15 cm
piece of 10 cm round poly \ inyl chloride pipe cut in half lengthwise served as a breeding
substrate within each quadrant and was examined for egg deposition daily. Both egg production
and o\ iposilion (spawning) frequency were recorded and used for the subsequent calculation of
egg and o\ iposilion frequency per female, per microcosm, and cumulatively per dose. At the
conclusion of the 3l)-day exposure, measurements of total length, total weight, gonad weight, and
liver weight were taken, and uonadosomatic indices (GSI), liver-somatic indices (LSI), and
condition factor (K) were calculated. Mean water-quality parameters (collected mid-depth)
sampled over the course of the exposure include dissolved oxygen (7.6 mg/L), temperature
(21.7°C), pH (8.5) and alkalinity (112.9 mg/L as CaC03). The 39-day NOEC was 74.1 mg/L
PFOA based on mean total egg production over the course of the experiment. The large outdoor
G-68
-------
microcosm experiment contained sediment, algae, macrophyte and zooplankton and therefore the
chronic value was not used quantitatively but was considered in a qualitative manner.
Wei et al. (2008a) later evaluated the hepatic protein profiles in Gobiocypris rarus
exposed to PFOA (98% purity) for 28 days. Adult male and female rare minnows (~nine months
old) with an average body weight of 1.4 ± 0.4 g and an average total length of 47.7 ±3.6 mm
were obtained from a laboratory hatchery and held in 20 L glass tanks ( -2 u body weight/L).
Fish were acclimated and treated as previously described (Wei et al. 2<)i)7) Briefly, fish were
supplied with dechlorinated tap water under continuous flow-through conditions at 25 ± 2°C
with a photoperiod of 16-hours:8-hours (light/dark). Commercial granular food (Teli a) was
supplied at a daily rate of 0.1% body weight. Waste and uneaten food were removed daily.
Gender determination was based on the shape of the abdomen and the distance between the
abdominal fin and the tail fin. After a one-week acclimation period, equal numbers of randomly
selected male and female rare minnows were assigned 1o one of four treatment groups: PFOA
exposure at 0, 3. I n or 3d mu I. These concentrations were selected based on previous studies
(Oakes et al 2<~)i)4). ho\\e\ er. the actual PI OA concentrations in the tanks were not verified by
chemical analysis luich treatment group contained ten male and ten female minnows in duplicate
tanks The flow rate of the test solution was 8 L/hour. At the end of the 28-day exposure period,
all fish were anesthetized on ice. Gonadal tissues from all fish and hepatic tissues from four male
and four female fish per treatment group were quickly dissected and fixed in 10% formalin for
histological analysis. The livers from the remaining six males and six females per treatment
group were removed and immediately frozen in liquid nitrogen and stored at -80 °C until
analysis. The 28-day LOEC (PCR alterations of genes in liver) was 3 mg/L PFOA, which was
not used quantitatively because it was a non-apical endpoint, but was qualitatively acceptable to
G-69
-------
potentially inform mode of action and AOP considerations. Following the same methodology
described above, Wei et al. (2008b) reported a 28-day MATC (change in m-RNA muscle-heart-
fatty acid binding protein, a hepatic protein) of 5.477 mg/L PFPA, and a 28-day LOEC (protein
spots identified by matrix-assisted laser desorption/ionization (MALDI) tandem time-of-flight
mass spectrometry (TOF/TOF) analysis) of 3 mg/L PFOA. These data were also not used
quantitatively but were retained for qualitative use.
Liu et al. (2008b) evaluated the effects of PFOA (98%) on the niRN A levels in the rare
minnow, Gobiocypris rarus. The flow-through unmeasured test was conducted lor 28 days. All
rare minnows were obtained from a laboratory hatchery. Two hundred and forty mature males
and females (about nine months old, 1.4 ± 0.4 g, 47.7 ± 3.0 111111) were randomly assigned to
eight 20 L glass tanks (30 individuals per lank) and acclimated lor one week. Fish were supplied
with dechlorinated tap water under continuous llow-through conditions at 25 ± 2°C and
subjected to a pholopcriod of I 0-hours:8-hours light:dai k I 'isli were fed a commercial granular
food (Tetra) at a daily rale of 0 I"., body weight. Waste and uneaten food were removed daily.
After a one week acclimation period. I 5 randomly selected male and 15 randomly selected
female rare minnows (gender determined by observing the shape of the abdomen and the
distance between the abdomen tin and the stern fin) were assigned to each of the four groups (0,
3, 10 or 30 mu I. PI OA). Lacli treatment was carried out in duplicate tanks (15 male and 15
female fish in each tank) The ilow rate of the test solution was 8 L/hour. After a 28-day
exposure period, fish were anesthetized on ice. Gills from four male and four female fish from
each treatment group were quickly dissected and fixed in 10% formalin for histological analysis.
The gills from six male and six female fish in each group were removed and immediately frozen
in liquid nitrogen and stored at -80°C until further real-time PCR analysis. Various tissues,
G-70
-------
including muscle, liver, brain, gonad, gill, and intestine from two male and two female fish in the
control group were used for semi-quantitative PCR analysis. The remaining samples were used
for proteomic analysis. The 28-day MATC (increase relative mRNA expression of AhR in gills)
was 5.477 mg/L PFOA for males, and the MATC (decrease relative mRNA expression of CYPla
and increase relative mRNA expression of PXR in gills) for females was 17.32 mg/L PFOA.
Both toxicity values were not used quantitatively because they were based on non-apical
endpoints but were considered qualitatively acceptable.
Liu et al. (2009) again investigated the effects of PFOA (98%) on the mR\A levels in
the gill and liver of the rare minnow, Gobiocypris rams, but for males and females separately.
Minnows (about nine months old), with a body mass of 1.4 -l. U.4 g and total length of 47.7 ±3.6
mm, were obtained from a laboratory hatchery. Briefly, male and female rare minnows were
randomly allocated into four treatment groups (<). 3. I <) or 3d mg L PFOA). Six males and six
females were included in each treatment group After a 2X-da\ exposure in dechlorinated tap
water under flow -illrough conditions at 25 2°C with a 16-hour light, 8-hour dark photoperiod,
fish were anesthetized on ice and then sampled The gills and livers from fish in the PFOA
treatment groups and the muscle. Ii\ ers. brains, gonads, gills, and intestines from fish in the
control group were removed, immediately frozen in liquid nitrogen, and stored at -80°C until
analysis. The 2X-da\ MATC (increase relative mRNA expression of PPARy in gills) was 5.477
mg/L PFOA for females, and the MATC (increase relative mRNA expression of PPARy and
PPARa in gills and CYP4T11 in liver) for males was 17.32 mg/L PFOA. Both toxicity values
were not used quantitatively because they were based on non-apical endpoints but were
considered qualitatively acceptable.
G-71
-------
The effects of PFOA (98% purity) on the gene expression of Gobiocypris rarus was also
evaluated by Fang et al. (2010). Nine-month-old male rare minnows with an average body mass
of 1.3 ± 0.3 g were obtained from a laboratory hatchery. Fish were kept in an indoor aquaria
system with flowing dechlorinated water at 25 ± 2°C and a photoperiod of 16-hours:8-hours
(light:dark). After acclimation for one week to ensure the absence of disease, fish were randomly
assigned to 20 L glass tanks (10 individuals per tank) and exposed under flow-through conditions
to various concentrations (0, 3, 10 or 30 mg/L) of PFOA for 14 days l-ach treatment was in
duplicate tanks. The flow rate of the test solution (8 L/hour), dissolved oxygen ( S0%), water
temperature (25 ± 2°C) and the functioning of the delivery system were monitored throughout
the study. Commercial granule food (Tetra) was supplied at a rate of 0.1% body weight per day
during the experiment. Waste and uneaten food were remo\ed daily \'o decrease in food
consumption or other adverse effects was ohser\ ed during the experiment. At the end of the
exposure, ten fish per group were anesthetized on ice. The li\ ers were removed, and six livers
from each group were immediately frozen in liquid nitrogen and stored at -80°C until analysis.
Hepatic tissues of ilie oilier lour indi\ iduals per group were quickly dissected and fixed in 10%
formalin lor histological examination. The 14-day LOEC (apolipoprotein gene expression) was 3
mg/I. PI OA. w liich was nol used quantitatively because of the test duration and non-apical
endpoint; howe\ er. the results were retained for qualitative use.
Yang et al. (2014) evaluated the toxicity of PFOA (CAS # 335-67-1, 99% purity) to
Pseudorasboraparva via a 30-day renewal measured exposure (the authors note that the
experiments followed ASTM standards and USEPA procedures for deriving water quality
criteria). The topmouth gudgeon (4.0 g, 4.0 cm) were purchased from the Beijing Chaoyang
Spring Flower Market, which was considered an atypical source. The organisms were allowed to
G-72
-------
acclimate for at least seven days before testing, and the test was conducted at 22 ± 2°C with a
light cycle of 12-hours, 10 fish per replicate and three replicates per concentration. Beakers used
for exposure were assumed glass but was not specified by study authors. PFOA was dissolved in
deionized water and carrier solvent DMSO to obtain a 7 mg/mL stock solution, and then diluted
with dechlorinated tap water to yield nominal exposure concentrations of 10, 15, 22.5, 33.75,
50.63 and 75.94 mg/L PFOA. Blank and solvent controls were also included. Water quality
parameters reported were pH = 7.0 ± 0.5, dissolved oxygen = 7.0 ± 0 5 mu I.. total organic
carbon = 0.02 mg/L and total hardness = 190.0 ± 0.1 mg/L CaCCb. The supplemental data
provided for the study included a comparison of measured PFOA concentrations before and after
solution renewal in the low and high acute and chronic test concentrations. PFOA concentrations
in the test water did not fluctuate by more than I 5".. duri ng experiments The 30-day survival
ECio of 11.78 mg/L PFOA was not used qiiantitiitix ely due to the atypical fish source and the
older/unspecified lile stage of the test organisms at test initiation. Results of this test were,
instead, considered t|Lialitalix elx acceptable
G.2.2.6 fchtnolucnmluc jishcs
Mirsimlsi ol ;il. (2020) cxaluatcd the sub-chronic effects of PFOA (CAS# 335-67-1,
>96% purity, purchased from Sigma-.Vldrich) on the Murray River rainbowfish (Melanotaenia
fluviatilis) One-year-old male adult fish were purchased from a commercial supplier and held in
a flow-through system in 5<) I. tanks with carbon filtered aerated water to acclimate to test
conditions. During the exposure period, the test temperature was 23°C, pH was 7.1-7.4 s.u. and
dissolved oxygen was >80% saturation. Stock solutions were made with Milli-Q water and
diluted to one of four test treatments (0.01, 0.1,1 and 10 mg/L). Four fish were assigned to 12 L
tanks, with three tanks per treatment, for a total of 12 fish per test concentration or control. Water
G-73
-------
was changed and fish were fed frozen brine shrimp daily during the 14-day exposure period.
Measured test concentrations at test initiation and after the first 24-hour renewal averaged <0.01,
0.01, 0.1, 0.91 and 9.0 mg/L PFOA, respectively. There was no effect on mortality, condition
factor or gonadosomatic index (GSI) between any treatment or the control at test termination.
The hepatosomatic index was significantly reduced at the three highest test concentrations as
compared to the control fish, suggesting increasing energy demands of the exposed fish. The
NOEC of 9.0 mg/L PFOA, based on growth and mortality, was acceptable lor qualitative use
only because of the short test duration for a chronic exposure
G.2.2.7 Adrianichthyidae fishes
Ji et al. (2008) evaluated the chronic toxicity of PI OA (CAS # 335-67-1, purity not
provided) to the Japanese medaka, Oryzias km pes. \ in renew a I unmeasured exposures. Solvent-
free stock solutions of PFOA (2,000 mg/l.) were prepared by dissolving the solid in MilliQ®
water via sonication Chemical measurements were not made, and nominal concentrations were
used throughout the study Medaka were maintained in the laboratory for several years at 25 ±
1°C, a 16-hour S-hour light dark photoperiod and led w ith Artemia nauplii (<24-hours after
hatching) twice daily I or the l'<> fish exposure study, breeding medaka pairs (-2.5 cm) were
maintained at 25 1UC for at least se\ en days in 1 L beakers filled with dechlorinated tap water,
which was prepared In serial liltration through a sediment and two granular activated carbon
filters. Thirty-six mating pairs that spawned more than eight eggs per breeding and bred more
than five times per week were selected and randomly separated into four groups. Nine mating
pairs were assigned to each treatment group and the control. Definitive test PFOA concentrations
were 0.1, 1 and 10 mg/L, based on the preliminary range-finding results using adult medaka. The
exposure duration for F0 fish was limited to 14 days, during which the fish were fed Artemia
G-74
-------
nauplii (<24-hours after hatching) ad libitum twice daily. The exposure medium was renewed at
least three times per week. Dead fish were removed as soon as possible. Eggs were counted
every day, and the eggs spawned on the seventh day were saved for the F1 generation exposure
study. On day 14, all surviving fish were euthanized, and body length and weight were
measured, from which the condition factor (K) was calculated. The gonads and livers were also
measured, and the gonadosomatic index (GSI) and the hepatosomatic index (HSI) were
calculated. For the F1 fish exposure study, fertilized eggs collected from I'd llsli exposed to each
concentration of PFOA and the control were randomly separated into groups of 25 eggs each and
then assigned to varying concentrations of PFOA (<).<). I. I or 10 mg/L), with only one replicate
per treatment. Because eggs were compiled into a single replicate for the hatching stage, results
reported beyond hatching (even when lar\ ae ju\ cniles were separated into replicates) are based
on pseudoreplication. During the egg stage lor the I ' I generation, investigators maintained all
possible combinations of I 'd \ I ' I exposure concentrations for a given compound. Exposure was
initiated in 50 ml. beakers less than 12 hours after fertilization. The developing embryos were
observed daily under a stereoscopic microscope, and dead embryos were removed. This
procedure was repeated until all li\ inu embryos had hatched. Hatching was defined as the
disruption of the chorion. New ly hatched larvae were then randomly transferred to 100 mL
beakers and ohser\ eel daily for swim-up success and survival for an additional two weeks.
Larvae were fed . \ricmia nauplii ad libitum twice daily. After 14 days, replicates with five fry
each were randomly selected from each treatment group and transferred to 1 L beakers for the
100-day post hatch observation. All survivors were sacrificed 100 days after hatching, and body
length and weight were measured. The gonads and livers were weighed to determine GSI and
HSI. The F0 (parental generation) adult survival, condition factor and adult male and female GSI
G-75
-------
and HSI 14-day LOECs were all >10 mg/L PFOA. For the F1 (progeny generation), the percent
hatchability, swim-up success and larval growth NOECs were all 10 mg/L PFOA, while the
NOEC for the time to hatch endpoint was 1.0 mg/L, suggesting these were tolerant endpoints
relative to the criterion magnitude of 0.094 mg/L. The LOEC for larval survival was 0.1 mg/L,
while the corresponding NOEC was considered <0.1 because effects were observed in the lowest
concentration tested. This test was not used quantitatively because uncertainties associated with
the responses across the range of concentrations tested. In many instances, authors did not report
increasing chronic effects with increasing concentrations that differed by an order of magnitude.
Additionally, endpoints associated with longer term effects to juveniles were also lx- rejected
because of pseudoreplication resulting from a lack of replicates in the hatching stage. Since this
test is a static unmeasured test, EPA chose to rely exclusively on the test by Lee et al. (2017) to
derive the SMCV for this species since Lee et al. (2< >17) u as a flow through measured test with
fewer concerns pcrlaininu to test design (i.e., no pseudoreplication) and results (lack of
increasing effects despite a l<)-lbkl increase in exposure concentrations).
Kang ef al. (2019) e\ aluated the chronic effects of PFOA (96% purity, CAS # 335-67-1
purchased from Sigma Aklrich. St. Louis. MO) on Japanese medaka (Oryzias latipes) in a 21-
day unmeasured, static-renewal stuck A stock solution was prepared by dissolving PFOA into
dimethyl sulfoxide and stored at 4°C. The 10 mg/L working solution was prepared by diluting
the stock solution in fish culture water (carbon-filtered dechlorinated tap water). Adult fish (16 ±
2 weeks, 0.38 ± 0.06 g) were obtained from the fish culture facility at the Korea Institute of
Technology in Jinju, Gyeongnam, South Korea. Fish were acclimated for 7 days in carbon-
filtered dechlorinated tap water at 25°C with a 14-hour: 10-hour light-dark photoperiod. Eight
male and eight female fish were introduced to a 20 L glass tank filled with 15 L of working
G-76
-------
solution at concentrations of 0 (control), 0 (solvent control) and 10 mg/L PFOA. Fish were fed
brine shrimp and Tetramin daily, and the working solution was renewed twice weekly. Authors
reported following OECD 229 exposure guidance, with conditions maintained the same as
during the acclimation period. Eggs were harvested and counted twice daily at seven, 14 and 21
days. A significant reduction in fecundity was shown for all time periods, but there was no
abnormal behavior or mortality observed. The 21-day fecundity I .()!¦(' was 10 mg/L PFOA and
only considered acceptable for qualitative use because it only reported a relali\ ely tolerant
LOEC value.
G.2.2.8 Amphibians
Hoover et al. (2017) tested chronic PFOA (purity) toxicity on the northern leopard
frog, Lithobatespipiens (formerly, Rana/>//>/i.7/\) in a chronic renewal test using measured
PFOA treatment concentrations. Stock solutions consisted of 1 g of chemical dissolved in 1 L of
Milli-Q water, followed In pi I adjustment to (¦> ^5-7 05, and lastly vacuum-filtration before
storage in polycarbonate bottles I juIh northern leopard frog egg masses were collected during
early spring from a temporary pond at the Purdue Wildlife Area in West Lafayette, IN, and
randomly assigned to outdoor ~ I no |. wading pools. After hatching, larvae were checked daily
for mortality and led Purina Rabbit Chow ad libitum. Treatments consisting of control and
exposure to PI OA at three concentrations (nominally 0.01, 0.1, and 1.0 mg/L) were placed in
two replicates on adjacent shelves within an environmental chamber. Experimental units
consisted of 15 L plastic aquaria filled with 7.5 L of filtered, UV-irradiated well water. Tadpoles
(n=35 per aquarium) were randomly assigned to the experimental units. Prior to addition to
aquaria, a subset of animals was examined to confirm development at Gosner stage 26, when
hind limb buds start to develop. Tadpoles with visible irregularities in morphology, coloration, or
G-77
-------
behavior were excluded. Animals were maintained at 20 ± 2°C with a 12-hour: 12-hour light:dark
photoperiod for 10 days to acclimate to indoor conditions and were fed a Tetramin slurry ad
libitum. Water changes (100%) were conducted every four days. Tadpoles were exposed for 40
days and were monitored daily for survival and abnormalities. A water sample (~5 mL) was
taken immediately prior to and after each water change to monitor concentration of test
chemicals. Every 10 days, six animals were randomly collected from each aquarium. The
animals were euthanized, measured (total length at 10 days, snout-vent length otherwise), and
staged (Gosner) prior to storage at -20°C for chemical analyses. After 40 days, the depuration
phase was initiated by removing animals, cleaning each aquarium with a methanol-soaked
sponge, and rinsing to remove adsorbed compound. Aquaria were refilled with clean water;
animals were returned to the same aquarium and monitored as described above. Water changes
were carried out every four days with fresh water, and a water sample was taken prior to each
water change. Two tadpoles were sampled every in days for an additional 30 days. Survival was
>90% for all treatments and no significant sublethal effects were observed. The 40-day NOEC
was >1.0 mg I. PI OA based oil (iosner stage reached at test termination and snout-vent length.
The test used water renewals rather than the required flow-through design for chronic ALC
development, however, leopard frogs commonly do not tolerate flow-through test systems and
the use of renewal system was appropriate for this study organism. Also, PFOA was detected in
the control organisms at concentrations three orders of magnitude lower than any PFOA
treatment groups, indicating the trace contamination in controls may not be considered a
significant issue. The 40-day NOEC of >1.0 mg/L was classified as acceptable for quantitative
use based on meeting data quality objectives; however, it was not used to derive the chronic
criterion because the study showed no adverse effects at the highest treatment concentrations
G-78
-------
(i.e., 1.0 mg/L). Because the highest treatment group that showed no effects was a relatively low
treatment concentration, including this NOEC value in the criterion calculation would have
resulted in the criterion magnitude being influenced by the relatively low test concentration
selected by study investigators (that did not produce an adverse response), rather than a
concentration-response relationship. Therefore, this test was not used quantitatively and was
considered as qualitatively acceptable for use in criterion deri\ a lion
Flynn et al. (2021) evaluated the chronic effects of PFOA (CAS /o5-67-1, >96%
purity, purchased from Sigma-Aldrich) on Northern leopard frogs, Lithubau-s pipicns (formerly
Rana pipiens), via a 30-day sediment-spiked, static outdoor mesocosm study. I 'rou egg masses
were collected from an ephemeral pond at the Purdue Wildlife area in West Lafayette, Indiana.
Egg masses were held in covered 190-L outdoor Uihs containing So I. of well water. Once
hatched, the larvae were fed ad libitum with Purina rabbit chow A control treatment (replicated
four times) and three nominal sediment exposure concentrations of 10, 100 and 1,000 ppb PFOA
(each replicated li\ e times) were set up in I S()-L plastic wading pools filled with 75 L well
water. The stock solution was made In dissoK ing 2.0 g PFOA into 1 L of reverse osmosis Milli-
Q water in polycarbonate bottles Sediment was collected from the upper 5 to 8 cm of a
permanent pond i n the same w i Idlife area. The sediment was air dried for eight days, with 10.1
kg of the dried homogenized sediment placed in each experimental unit. Sediment was spiked
with the assigned PI OA dose by adding the appropriate volume of stock solution to 6 L of water,
stirred for five minutes, and then allowed to equilibrate for seven days. Once equilibrated, 75 L
of water was added to the experimental chamber and allowed to sit for an additional three days.
The water was then inoculated with algae and zooplankton from local pond water and allowed to
establish for five days, after which Gosner stage 25 frog larvae were added to each tank.
G-79
-------
Reported average water quality conditions include a pH of 7.8 and temperature of 26.2°C.
Overlying water PFOA measured water concentrations were 2.9, 7.3 and 66 |ig/L for the nominal
sediment concentrations of 10, 100 and 1,000 ppb, respectively. At test termination (30 days)
there was no effect on survival and growth (snout-vent length and weight). The 30-day NOEC,
based on survival and growth, was 66 |ig/L (or 0.066 mg/L). Howe\ ei\ on test-day five and at
test termination all frogs in the spiked sediment mesocosm were less de\ eloped, based on Gosner
stage, than the control mesocosms. The study was acceptable for qualilali\ e use only because the
test design was an outdoor spiked sediment mesocosm exposure with algal and xooplankton
communities present.
Yang et al. (2014) evaluated the chronic toxicity of PFOA (CAS #335-67-1, 99% purity)
to the Asiatic toad, Bufo gargarizans via a 3<)-da\ renewal measured exposure (the authors note
that the experiments followed ASTM standards and I SI -PA procedures for deriving water
quality criteria). The tadpoles (<) 114s u, 1.8 cm) were purchased from the Beijing Olympic Park,
which was considered an aly pical source. The organisms were allowed to acclimate for seven
days before testing. and the lest was conducted at 22 ± 2°C with a light:dark cycle of 12-
hours 12-hours There were 10 tadpoles per replicate and three replicates per concentration.
Beakers used lor exposure were assumed glass but was not specified by study authors. PFOA
was dissol\ ed in deionized water and carrier solvent DMSO to obtain a 7 mg/mL stock solution,
and then diluted ith dechlorinated tap water to yield nominal exposure concentrations of 5, 7.5,
11.25, 16.88, 25.31 and 37.97 mg/L PFOA. Water quality parameters reported were pH = 7.0 ±
0.5, dissolved oxygen = 7.0 ± 0.5 mg/L, total organic carbon = 0.02 mg/L and total hardness =
190.0 ±0.1 mg/L as CaC03. The supplemental data provided for the study included a
comparison of measured PFOA concentrations before and after solution renewal in the low and
G-80
-------
high acute and chronic test concentrations. PFOA concentrations in the test water did not
fluctuate by more than 15% during experiments. The 30-day ECio (survival) reported for the
study, 5.89 mg/L PFOA, was not used quantitatively due to the atypical test organism source
(obtained from Beijing Olympic Park) and limited details pertaining to the source of the test
organisms and any potential previous exposure to PFOA or any other contaminant.
Consequently, this study was considered qualitatively acceptable
G.2.3 Summary of Qualitatively Acceptable Plant PFOA Toxicity Studies
G. 2.3.1 Cyanobacteria, Anabaena sp.
Rodea-Palomares et al. (2012) examined the toxicity of PFOA (acid form. CAS # 335-
67-1, 96% purity) with the bioluminescent cyanobacleiium. . Inabaena sp. (CPB4337 strain)
following OECD Guidelines No. 23 (OECD 2000b) and Rodea-Palomares et al. (2009). The
inhibition of constitutive luminescence was examined o\ er a 24-hoiir test period. Very little
information was provided about the exposure details (i.e . test media, test vessel, cell density per
replicate, water quality parameters) PI-OA was dissolved in the exposure media with no solvent
and was measured in the highest test concentration and one concentration near the reported ECso.
The cyanolxictei ia were exposed to li\ e to seven test concentrations with replicate samples. Each
test was repented three times The reported ECso was 19.81 mg/L based on bioluminescence
inhibition and was not acceptable for quantitative e use, based on the short test duration and lack
of exposure details. This test was, therefore, considered qualitatively acceptable for use in
criteria derivation.
Rodea-Palomares et al. (2015) conducted a similar 24-hour static, unmeasured test on
PFOA (acid form, 96% purity) with the bioluminescent cyanobacterium, Anabaena sp.
(CPB4337 strain). The test was performed with 1.5 mL of cyanobacterial growth media
(AA/8+N, Allen and Arnon 1955) in transparent 24-well microtiter plates. The pH of the growth
G-81
-------
media was 7.8. The plates were incubated at 28°C under continuous illumination on a rotary
shaker. The cyanobacteria in the log-growth phase (initial optical density at 750 nm = 0.1) were
exposed to seven nominal concentrations ranging from 0-200 mg/L. A description of test
solution preparation was lacking, but it does not appear that a solvent was used. Each test was
repeated three times. The reported ECso was 78.88 mg/L based on bioluminescence inhibition
was not acceptable for quantitative use given the short test duration This test was considered
qualitatively acceptable.
G. 2.3.2 Green alga, Raphidocelis subcapitati i
(Formerly known as Selenastrum capricornuiiini and I'scinlokirclineriella subcapuuhi)
3M Company (2000) exposed the green alua. Raplmlocclis subcapitata (formerly,
Selenastrum capricornutum) to PFOA (C AS 335-67-1) in a l)6-houi' static, unmeasured acute
toxicity test. The toxicant was part of the 3\l production lot number 269 and was characterized
as a mixture of PI () A (5-1 < n)" <> of the compound) and C'u, C: and C9 perfluoro homologue
compounds (0-3 5".. of the compound). The substance was dissolved in a 50:50
water:isopropanol solution to make a primary solution of 1,000 mg/L test substance and
isopropanol A separate lest conducted by the authors showed no growth effects at 1,000 mg/L
isopropanol 011 the same species. The primary solution was diluted with algal medium to make
five nominal test concentrations (63, 125, 250, 500 and 1,000 mg/L test substance or 32, 63, 130,
250 and 500 mg/l. PI OA) plus a control (algal medium). The report stated the test followed
USEPA-TSCA Guideline 797.1050. Exposures were conducted in 250 mL glass beakers with 50
mL of test solution and an initial cell loading of 10,000 cells/mL. There were three replicates for
each treatment. The pH of the highest test concentration was low (2.3-3.0) over the course of the
experiment as compared to the control (7.4-10.3). The mean number of cells increased in the
G-82
-------
control and <125 mg/L test substance treatments but decreased at all other treatments. The 96-
hour reported ECso, based on cell density and growth rate, was 180 mg/L test substance and
isopropanol or 90 mg/L PFOA. Because of the possible mixture effects of other perfluoro
homologue compounds, the toxicity value from the study was not acceptable for quantitative use
but was retained for qualitative use.
3M Company (2000) provides the results of four separate toxicity tests completed with
the green alga, Raphidocelis subcapitata (formerly Selenastrum caprkorimmni), and APFO
(CAS # 3825-26-1). The toxicant was part of the 3M production lot number 37 and was
characterized as mixture of APFO (96.5-100% of the compound) and C6, C7 and ("¦¦ perfluoro
analogue compounds (0-3.5% of the compound). The toxicity tests followed a protocol modified
from USEPA-600/9-78-018 (1978) and ASTM-I >35 23 (1981) There were four separate
exposure regimes: 1) a four-day exposure I "-day reco\ cry period; 2) a seven-day exposure +
seven-day recovery period. 3) a I "-day exposure + four-day recovery period; and 4) a 14-day
continuous exposure A bacteria-lYee culture of the alga was obtained from the USEPA
(Corvallis, OR) and stored in the dark until testing. Seven-day old stock cultures with an initial
density of I \ I n1 cells nil. were placed in 250 mL flasks with 50 mL of test solution. There were
three replicates for each of the six nominal test concentrations (100, 180, 320, 560, 1000 and
1800 mg/l.) and control. Nutrient medium was used as the dilution media for all test treatments
and were not renewed during the exposure. Algae were grown at 23°C and continuously shaken
at 100 rpm. The author-reported EC10, based on cell counts, was 5.3, 3.3, 2.9, and 5 mg/L, for the
four, seven, 10 and 14 day exposures, respectively. Note: the authors did not specify if the ECios
were determined after the exposure period or the post observation period. Because of the
G-83
-------
possible mixture effects of other perfluoro analogue compounds in the tests these toxicity values
were not acceptable for quantitative use but were retained for qualitative use.
3M Company (2000) exposed Raphidocelis subcapitata (formerly, Selenastrum
capricornutum) to APFO (CAS # 3825-26-1) in a 96-hour static, unmeasured acute toxicity test.
The toxicant was part of the 3M production lot number HOGE205 and was not sufficiently
characterized but was considered a mixture of APFO (30% of the compound), other impurities,
and water. A stock solution (3330 mg/L) was made without the use of a sol\ enl, which was then,
then diluted with algal medium (USEPA 1978) to make five nominal test concentrations (210,
430, 830, 1670 and 3330 mg/L APFO) plus a control (algal medium). The test followed USEPA-
TSCA Guideline 797.1050 and OECD 201. Exposures were conducted in 250 mL glass beakers
with 100 mL of test solution and an initial cell loading of 10.<)<)<) cells mL. There were three
replicates for each treatment. The 96-hour reported !¦('=... haseel on cell count, was 1,980 mg/L
APFO. The authors reported that the test substance is considered a mixture of APFO and other
impurities and stated the l-C*.. may not accurately reflect the toxicity of APFO. Therefore, the
value was not acceptable for quantilali\ e use and was retained for qualitative use.
Kossil ol al. (2010) performed a 72-hour static, measured growth inhibition test with
perfluorooctanoic acid (96° n purity) on the green alga, Raphidocelis subcapitata following
OECD TG 2<)| Protocol. While limited details were provided about the exposure, the authors
state they were following the OECD protocol. The algae were cultured in 96-well microplates
with a total volume of 200 |iL. No solvents were used to make test solutions. Specific test
concentrations were not provided, but the authors noted that nominal and measured
concentrations did not have significant deviations. The 72-hour growth inhibition (based on
G-84
-------
biomass) ECso of 96.2 mg/L PFOA was not acceptable for quantitative use because of the short
test duration but was considered qualitatively acceptable.
Ding et al. (2012b) performed a rapid, 4.5-hour static algal growth inhibition test on
PFOA (acid form, CAS # 335-67-1, 96% purity) with Raphidocelis subcapitata. The test follows
the protocol of PAM Test: Acute effects on photosynthesis in algae developed in the Dutch
National Institute of Public Health, RIVM (Verweij et al. 2009) Dilution medium consisted of
Dutch standard water. R subcapitata used for testing were from an in-housc culture. Algae at a
cell density of 3xl06 cells/mL were inoculated in medium containing PFOA al nominal
concentrations of 0 (negative control), 1, 1.5, l.S. 2. 2 2. 2 5. 3 mM PFOA, or 414 1. 621.1,
745.3, 828.1, 911.0, 1,035, and 1,242 mg/L PFOA when converted by multiplying the reported
mM concentration by a molecular weighl ol"4l4 <>7 g'mol. There were two replicates for each
exposure concentration. No details were pro\ ided lor lighting, temperature, or other dilution or
test solution parameters The reported 4.5-hour l:X'=- based on photosynthetic efficiency was
1.807 mM (748 mg I.) The I¦("*¦¦ was not quantitatively acceptable due to the short duration but
was retained lor qimlilali\ e use
G.2.3 (irccn a/ifa. Scciicik'siiiiis ob/n/mis
I.in et al. (2008a) conducted a 72-hour unmeasured exposure with Scenedesmus obliquus
to evaluate the effects of PI OA (acid form, CAS #335-67-1, purity not reported) at the cellular
level, measured by How cytometry. Authors stated that the test followed OECD (2002)
methodology with S. obliquus that were obtained from the Freshwater Algae Culture Collection,
Institute of Hydrobiology, Chinese Academy of Sciences (Beijing). The algal test medium was
prepared according to OECD (2002) using deionized water and analytically pure chemicals,
adjusted to pH 7.5 ± 0.2. The authors did not provide details regarding how the PFOA treatments
G-85
-------
were prepared. Algae in exponential growth phase and a cell density of lxlO4 cells/mL were
inoculated in medium containing PFOA at nominal concentrations of 0 (negative control), 500,
1,000, 1,500, and 2,000 |iMPFOA, or 0.0, 207.0, 414.1, 621.1, and 828.1 mg/L when converted
based on molecular weight of PFOA, 414.07 g/mol. Experiments were initiated in replicated 50
mL flasks (n = 3 per treatment) containing a total volume of 20 mL of algal cell suspension.
Algal cells were incubated at 22 ± 1°C under cool-white lights ("(¦>.ddd lux) with a 14-hour:10-
hour light:dark cycle. The 72-hour NOEC (growth rate reduction, based on optical density) of
PFOA was 828.1 mg/L; the LOEC was >828.1 mg/L. The plant value from the study was not
used quantitatively because of the short exposure duration (less than 96-hours) and the missing
exposure details. The test was instead considered qualitatively acceptable.
G. 2.3.4 Duckweed, Lemna gibba
Boudreau et al. (2003) performed a 7-day static acute algal growth inhibition test on
PFOA (acid form. CAS .>."?5-o7-1. l)7% purity) with duckweed, Lemna gib ba. The study was
part of a Master's thesis at the I ni\ ersity of Guelph, Ontario, Canada. Authors stated that the
test followed protocols found in \ST\I l-1415-^l (ASTM 1991), Greenberg etal. (1992) and
Mam ood et al (2'><> I) Duckweed was obtained from laboratory culture maintained according to
Marwood et al (2001), and originally acquired from University of Waterloo. All treatment
concentrations were prepared in laboratory-grade distilled water. Toxicity testing consisted of six
test treatments plus a neuati\ e control (0, 10, 30, 50, 100, 300, and 500 mg/L) in 10 mL of
Hunter's growing media in 60 x 15 mm polyethylene disposable petri dishes. There were three to
four replicates per treatment, but the number of plants and fronds per plant were not reported.
Tests were continuously illuminated with cool-white, fluorescent light between 5,800 and 6,200
lux and incubated at 25 ± 1°C. Endpoints used to determine inhibition of growth were mean
G-86
-------
frond number and biomass, measured as wet weight. The reported ICio, IC25 and IC50, based on
wet weight, were 0.052 M (95% C.I.: 0.042-0.065), 0.127 M (95% C.I.: 0.117-0.146) and 0.193
M (95% C.I.: 0.142-0.210), respectively. Note that although ICx for PFOA were reported in
molar (M) units, EPA judged the units were misreported and were actually millimolar (mM).
This judgement was based on the reported test concentrations in Table 3.1 of the publication and
the reported effect concentrations (ICx) would not fall within this range unless the values were in
mM units. Accordingly, ICx now considered as mM, were converted lo nig I. hy molecular
weight of 414.07 g/mol PFOA. The calculated 7-day ICio, IC25 and IC50 expressed as mg/L from
the study were 21.53, 52.59, and 79.92, respectively and were not acceptable for quantitative use
given the lack of exposure details and uncertainties with the reported units but were retained for
qualitative use.
G-87
-------
Appendix H Other Estuarine/Marine PFOA Toxicity Studies
H.l Summary Table of Qualitative Estuarine/Marine PFOA Toxicity Studies
Species (lifestage)
Method3
Test
Duration
Chemical
/ Purity
pH
Temp.
(°C)
Salinity
(ppt)
Effect
Chronic
Limits
(NOEC-
LOEC)
(mg/L)
Reported
Effect
Cone.
(mg/L)
Deficiencies
Reference
Bacterium,
Vibrio fischeri
S, M
15 minutes
PFOA
96%
18
-
ECso
(bioluminescence
inhibition)
524
Duration too short for a
plant test, missing some
exposure details, non-
apical endpoint
Rosal et al.
2010
Cyanobacterium,
Anabaena sp.
S, M
24 hours
PFOA
96%
-
28
-
ECso
(bioluminescence
inhibition)
-
72.3
Duration too short for a
plant test, missing some
exposure details, non-
apical endpoint
Rosal et al.
2010
Cyanobacterium,
Geitlerinema
amphibium
S,U
72 hours
PFOA
Unreported
7.6-
7.8
20
8
ECso
(growth)
-
248.4b
Duration too short for a
plant test, missing some
exposure details
Latala et al.
2009
Dinoflagellate,
Pyrocvstis lunula
S, M
24 hours
PFOA
95%
19
EC50
(bioluminescence
inhibition)
-
18
Duration too short for a
plant test, atypical
endpoint
Hayman et al.
2021
Golden brown alga,
Isochrysis galbana
S,U
72 hours
PFOA
96%
-
20
¦¦
ECso
(growth inhibition)
-
163.6
Duration too short for a
plant test, missing some
exposure details
Mhadhbi et al.
2012
Green alga,
Chlorella vulgaris
S,U
72 hours
PFOA
Unreported
7.6-
7.8
20
8
ECso
(growth)
-
977.2b
Duration too short for a
plant test, missing some
exposure details
Latala et al.
2009
Diatom,
Skeletonema marinoi
S,U
72 hours
PFOA
Unreported
7.6-
7.8
20
8
ECso
(growth)
-
368.5b
Duration too short for a
plant test, missing some
exposure details
Latala et al.
2009
Purple sea urchin
(fertilized eggs),
Paracentrotus lividus
S,U
48 hours
PFOA
96%
-
20
-
ECso
(growth inhibition)
-
110.0
Duration too short for an
acute test
Mhadhbi et al.
2012
Blue mussel,
Mytilus edulis
R, U
21 d
PFOA
Unknown
-
16-19
-
LOEC
(catalase activity)
<0.2-0.2
0.2
Atypical endpoint
Li et al. 2021a
H-l
-------
Species (lifestage)
Method3
Test
Duration
Chemical
/ Purity
i)H
Temp.
(°C)
Salinity
(ppt)
Effect
Chronic
Limits
(NOEC-
LOEC)
(mg/L)
Reported
Effect
Cone.
(mg/L)
Deficiencies
Reference
Green mussel
(60-65 mm),
Perna viridis
R, M
7 days
PFOA
96%
-
25
25
MATC
(relative condition
factor)
0.0114-
0.099
0.03359
Exposure duration too
short for chronic test and
too long for acute test,
non-apical endpoint
Liu et al. 2013;
2014c
Green mussel (adult),
Perna viridis
RM
7 days + 7
days
observation
PFOA
96%
-
25
30
ECso
(integrative
genotoxicity)
0.093-
0.950
0.5940
Exposure duration too
short for chronic test and
too long for acute test,
non-apical endpoint
Liu et al. 2014a
Green mussel (adult),
Perna viridis
RM
7 days
PFOA
96%
-
25
25
MATC
(CAT and SOD
activity)
0.099-
1.12
0.3330
Exposure duration too
short for chronic test and
too long for acute test,
non-apical endpoint
Liu et al. 2014b
Green mussel,
Perna viridis
RM
7 days + 7
days
observation
PFOA
96%
8
25
30
MATC
(hemocyte cell
viability)
0.0114-
0.099
0.03359
Exposure duration too
short for chronic test and
too long for acute test,
non-apical endpoint
Liu and Gin
2018
Manila clam
(3.64 cm),
Ruditapes
philippinarum
RM
21 days
PFOA
Unreported
-
12
35
NOEC
(mortality)
0.00093-
>0.00093
0.00093
Only one exposure
concentration, apical
endpoints are not the
focus of study
Bernardini et al.
2021
Japanese medaka
(adult),
Oryzias latipes
RU
7 days
PFOA
ammonium
salt
98%
-
25
NOEC
(survival, condition
factor)
100->100
100
Exposure duration too
short for chronic test and
too long for acute test
Yang 2010
Turbot (embryo),
Scophthalmus
maximus
(formerly, Psetta
maxima)
RU
6 days
PFOA
96%
-
18
-
LCso
-
11.9
Exposure duration too
short for chronic test and
too long for acute test
Mhadhbi et al.
2012
a S=static, R=renewal, F=flow-through, U=unmeasured, M=measured, T=total, D=dissolved, Diet=dietary, MT=maternal transfer
b Reported in moles converted to milligram based on a molecular weight of 414.07 mg/mmol.
H-2
-------
H.2 Estuarine/Marine Qualitatively Acceptable PFOA Toxicity Study
Summaries
H.2.1 Summary of Acute PFOA Toxicity Studies Used Qualitatively
H.2.1.1 Invertebrates
Mhadhbi et al. (2012) conducted a 48-hour static, unmeasured acute test with PFOA
(96% purity) on the sea urchin, Paracentrotus lividus (a non-North American species). A stock
solution of PFOA was made either with filtered sea water from the Ria (if Vigo (Iberian
Peninsula) for low exposure concentrations, or with DMSO for high PI OA concentrations (at a
final maximum DMSO concentration of 0.01% (v/v) in the test medium). Ho\\c\ cr. authors did
not indicate what was considered a high test concentration 11" a DMSO was used, a solvent
control was also included with the test. Sea urchin embryos were exposed to one of ten nominal
PFOA treatments (1, 2, 5, 10, 20, 50, 100. Zoo. 5<)<) and 750 mg I.) I-'our hundred fertilized eggs
(within 30 minutes of fertilization) were transferred lo glass \ ials containing 10 mL of test
solutions with four rcplicales per PI 'OA treatment and five replicates per control. Vials were
incubated at 20' ( in the dark lor 4N-hours Al lest termination samples were fixed in formalin
and 35 lan ae per \ ial was measured lor growth (length). The 48-hour ECso (growth inhibition)
was I I ii 'i mg/L and was not acceptable for t|Lianlitalive use due to the atypical acute endpoint
and short test duration but was retained for qualitative use
H.2.1.2 lish
Mhadhbi cl al. (2012) conducted a six-day renewal, unmeasured acute test with PFOA
(96% purity) on the turbot, Scopthalmus maximus (formerly, Psetta maxima; a non-North
American species). A stock solution of PFOA was made either with filtered sea water from the
Ria of Vigo (Iberian Peninsula) for low exposure concentrations, or with DMSO for high PFOA
concentrations (a final maximum DMSO concentration of 0.01% (v/v) in the test medium).
H-3
-------
However, authors did not indicate what was considered a high test concentration. If a DMSO
was used a solvent control was also included. Fish were exposed to one of eight nominal PFOA
treatments (1.5, 3, 5, 10, 12, 24, 100 and 200 mg/L). Turbot eyed eggs from a single stock of
adults were supplied by a nearby fish hatchery (PESCANOVA Insuina) and acclimated to
laboratory conditions before use. At 72-hpf, the floating fertilized euus were collected and the
non-fertilized eggs at the bottom discarded. Embryos that had reached I lie blastula stage were
used for testing. Fifty normal embryos were added to glass beakers containing 500 mL of test
solution. Each treatment had four replicates and were incubated in the dark for six days at 18°C
with no food or aeration provided. Dead embryos and larvae were removed daily I jidpoints
included dead embryos, malformation, hatch success at 48-hours and larvae survival (missing
heartbeat and a non-detached tail) at six days The 6-day LC;.. of I I ^ mg/L PFOA was not used
quantitatively because of the atypical acute lest duration but was considered qualitatively
acceptable.
H.2.2 Summary ol'Chionic PI OA Toxicity Studies Used Qualitatively
H.2.2.1 Moltnsks
l.icl al. (202In) e xamined the physiological, transcriptomic, and metabolomic responses
to PI-OA in the blue mussel. \ lyiilns cihilis Mussels were collected from the Jinhuang Gulf
(Yellow Sea. China) and acclimated in flow-through tanks at 16-19°C and a 12-hour light:dark
photoperiod for se\ en days PI 'OA was dissolved in DMSO and diluted with natural seawater to
create three test treatments (20, 200 and 2000 |ig/L PFOA) plus a control. Each treatment was
replicated three times with sixty mussels per each replicate. Water in exposure tanks was
replaced daily and ten mussels were subsampled on test day 0, 1, 3, 7. 21. At test termination
enzyme activity was measured and CAT and SOD activity decreased in the 200 and 2000 |ig/L
PFOA treatments groups as compared to the controls. The 21-day LOEC, based on catalase
H-4
-------
activity, was 0.2 mg/L PFOA. The lack of apical effect assessed precluded this study from being
acceptable for quantitative use, but it was retained for qualitative use.
Liu et al. (2013, 2014c) evaluated the chronic effects of PFOA (96% purity, purchased
from Sigma-Aldrich) on green mussels, Perna viridis, via a seven-day measured, static-renewal
study. The mussels were obtained from a local farm in Singapore, and subsequently acclimated
to laboratory conditions for seven days before testing. Mussels were kept at a salinity of 25 ppt
(artificial seawater) and a temperature of 25°C. Forty mussels (60-65 111111 length) per 50-L
polypropylene tank, each duplicated, were exposed to measured PFOA concentrations of 0
(control), 0.08, 1.2, 11.4, 99 and 1,120 |ig/L. Mussels were fed a commercial murine micro-algae
purchased from Reed Mariculture on renewal days, which occurred every two days, two hours
before the solution renewal. PFOA concentrations were verified through water and muscle tissue
samples via liquid chromatography-tandem mass spectrometry Weights and lengths were
determined on days zero and se\ en A NOEC of II 4 |iu I. and a LOEC of 99 |ig/L was
determined for a decrease in the relati\ e condition factor (RCF). The study was acceptable for
qualitative use only because of the atypical test duration, which is too long for an acute test and
too short for a chronic test Additionally, the PFOA test displayed a questionable concentration-
response patten where there was no difference between the RCF at the LOEC (i.e., 99 ug/L) and
the highest test concentration, which contained a PFOA concentration that was more than 10X
greater (i.e., 1120 uu I.). The large magnitude between these two concentrations in combination
with the lack of effects to the RCF observed between the LOEC and the highest treatment
concentration suggests a true concentration-response relationship was not observed for PFOA in
this test.
H-5
-------
Lui et al. (2014a,b) and Liu and Gin (2018) conducted a series of seven-day renewal,
measured experiments with perfluorooctanoic acid (PFOA, 96% purity) on the green mussel,
Perna viridis (a non-North American species). All of these studies utilized a similar test design,
but with each publication providing a different level of test details. In Liu et al. (2014a), green
mussels were obtained from a local fish farm and acclimated to laboratory conditions prior to
PFOA exposure. Adult organisms were exposed in 70 L polypropylene tanks in artificial
seawater at a temperature of 25°C and at salinity of 30 ppt. Mussels w ere exposed to one of five
nominal PFOA concentrations (0.0001, 0.001, 0.010, 0.1 and 1.0 mg/L) or a control. Each tank
contained 60-65 mussels, with two tanks per exposure concentration or control. During
exposures, mussels were fed with microalgae and each tank was cleaned and refilled every two
days. After seven days of exposure and se\ en days of depuration. \ arious biomarkers were
measured. The ECso (integrative genotoxicily) was reported as n 594 mg/L PFOA and was based
on three genotoxic endpoints (l)Y\ fragmentation and single strand breaks (comet assay),
chromosomal breaks (microiuicleus lest) and apoptosis (DNA diffusion assay). Results of this
study were not used <.|iiantilali\ ely due to the short exposure duration and endpoint but was
considered <.|iialitati\ely acceptable
In I jii ol al. (2014b). the oxidative damage of PFOA (96% purity) to green mussels was
assessed after se\ en days under similar conditions as Lui et al. (2014a). Green mussels were
obtained from a local fish farm in Singapore and acclimated to laboratory conditions for one
week prior to exposures. Organisms (60-65 mm) were exposed in polypropylene tanks
containing artificial seawater at a temperature of 25°C and at salinity of 25 ppt. Mussels were
exposed to one of six nominal PFOA concentrations (0.0001, 0.001, 0.010, 0.1, 1.0 and 10.0
mg/L) or a control. Nominal concentrations were similar to measured concentration (0.00008,
H-6
-------
0.0012, 0.0114, 0.099, 1.120 and 9.630 mg/L, respectively) and no PFOA was detected in the
controls. Each treatment was replicated with an unreported number of mussels per replicate.
Again, during the exposure mussels were fed with microalgae and each tank was cleaned and
refilled every two days. The most sensitive parameters to PFOA were activation of antioxidant
enzymes (catalase [CAT] and superoxide dismutase [SOD]), which is an adaptive response to the
excessive reactive oxygen species. Significant effects were obser\ cd al 0.099 mg/L PFOA, but
not at 1.12 mg/L. The seven-day MATC (CAT and SOD activity) was <> 333<) ing/L, which was
not used quantitatively because of the atypical endpoint and duration but was considered
qualitatively acceptable for use.
Liu and Gin (2018) employed the same test design and nominal PFOA concentrations as
Lui et al. (2014a). The most sensitive biomarker endpoint reported was hemocyte cell viability
with a NOEC and LOEC of 0.0114 and 0 ii')1) mu I.. respectively Again, the MATC of 0.03359
mg/L PFOA was not used quantilali\ ely due to the atypical test duration but was considered
qualitatively acceptable
Bernardiiii ol ;d. (2021) reported the results of a 21-day chronic study with the Manila
clam. A'iii/iia/K-\/>hi/i/>/>marimi. and ITOA (CAS # 335-67-1). Clams were collected from the
field (Venice I .auoon, Italy) and acclimated to laboratory conditions (aerated natural seawater,
salinity 35 I pptandl2: n 5DC) for one week. Ninety healthy individuals (3.64 cm shell
length) were evenly di\ ided amongst two aquaria exposed to either PFOA at a nominal 1 |ig/L or
control seawater for 21 days. Clams were fed and test solutions were renewed every other day.
Subsamples of clams (n=20) were collected on test days seven and 21 for soft tissue PFOA
concentrations and haemolymph analysis. No significant effects of mortality were observed in
the single treatment group throughout the exposure. The measured concentration of PFOA was
H-7
-------
0.93 jug/L. At the end of the experiment over 113 genes were upregulated and 362 genes were
downregulated in exposed compared to control clams. The 21-day NOEC, based on mortality,
was 0.00093 mg/L PFOA, which was not acceptable for quantitative use because apical
endpoints were the focus of the study and only one exposure concentration was employed which
did not result in apical effects at a relatively low concentration. Results of this study were
retained for qualitative use only.
H.2.2.2 Fish
Yang (2010) performed a seven-day, renewal unmeasured toxicity test with the
ammonium salt of perfluorooctanoic acid (98% purity) using the medaka, Oryzias talipes (a non-
North American species). Male medaka were gradually acclimated to a high saline condition
(specific salinity not reported) in the laboratory o\ or I 5 days l-'reshw uter was mixed with equal
parts seawater, with half of the volume replaced with senwiiler e\ ery day. No mortality was
observed during 1 his acclimation period. The I 3 generation from these parental fish were used in
the experiments. Twel\ e adult male medaka were added to unreplicated 2 L glass tanks and
exposed to one of three nominal PI OA concentrations (10, 50, 100 mg/L) or controls for seven
days Solutions were renewed daily and fish were maintained at 25 ±1°C under a constant
photoperiod of I (i-hours^-hours (light:dark). After seven days, fish were euthanized and various
endpoints were measured: condition factor (K), gonadosomatic index (GSI), liver somatic index
(LSI), and other enzymatic responses. Across all treatments none of the fish died, and there were
no significant differences between PFOA treatments and controls for K, GSI and LSI. The
survival-based seven-day NOEC of 100 mg/L PFOA was not acceptable for quantitative use
because of the atypical exposure duration and lack of replication in the experiment but was used
qualitatively.
H-8
-------
H.2.3 Summary of Qualitatively Acceptable Data for Saltwater Plants
H.2.3.1 Bacterium, Vibrio fischeri
Rosal et al. (2010) conducted a 15-minute static, measured bioluminescence inhibition
test with perfluorooctanoic acid (PFOA, 96% purity) on the bacterium, Vibrio fischeri following
ISO 11348-3 standard protocol. While limited details were provided about the exposure, the
authors stated they followed the standard protocol. The experiment used a commercially
available Biofix Lumi test (Macherey-Nagel, Germany), where the bacterium is supplied freeze-
dried. It was reconstituted and incubated at 3°C for five min before use. The experiment
employed a 0.34 M NaCl (2% w/v) test medium conducted at 18°C. No solvents were used to
make test solutions. Specific test concentrations were not pro\ ided, but the authors noted that
nominal and measured concentrations did not ha\ e significant de\ iations. The 15-minute ECso,
based on bioluminescence inhibition, was 524 mu I. The test \\ as not used quantitively because
of the short test duration and lack of exposure details but was considered qualitatively
acceptable.
H.2.3.2 Cyanobacicriiini. Anabaena
Rosal ol al. (2010) also conducted a 24-hour static, measured bioluminescence inhibition
test with PI'OA (96° n purity) on the c\ anobacterium, Anabaena sp. Limited details were
provided about the exposure, but the authors stated they were following the test design in Rodea-
Palomares et al (2<")l>) The c\ anobacterium Anabaena, CPB4337 strain, was grown at 28°C on
a rotary shaker in 50 mL AA/8 media supplemented with nitrate (5 mM) in 125 ml Erlenmeyer
flasks and 10 mg/mL of neomycin sulphate. No solvents were used to make test solutions.
Specific test concentrations were not provided, but the authors noted that nominal and measured
concentrations did not have significant deviations. The 24-hour EC so, based on bioluminescence
H-9
-------
inhibition was 72.3 mg/L and was not used quantitatively because of the short test duration and
lack of exposure details but was retained for qualitative use.
H.2.3.3 Cyanobacterium, Geitlerinema amphibium
Latala et al. (2009) performed a 72-hour static algal growth inhibition test on PFOA
(acid form, CAS # 335-67-1) with Geitlerinema amphibium. The purity of PFOA was not
reported, but the authors stated they used the highest grade commercially available. Authors
stated the algal growth inhibition tests followed protocols found in ISO 10253 (ISO 1995) and
ISO 8692 (ISO 1993), but used f/2 medium, different test species and a different photoperiod
than the test protocols. The blue-green alga G. amphibium 15A-13 strain was isolated from Baltic
Sea coastal waters and maintained as monoalgal cultures in the Culture Collection of Baltic
Algae (CCBA) at the Institute of Oceanography. I ni\ ersity of (idansk Algae were batch-
cultured in f/2 medium prepared in distilled water and lnought to a salinity of 8 PSU was using
Tropic Marin® sea salt Cultures were acclimated lor ten days at 20°C and a 16-hour
photoperiod (25 umol photons m": s"1) The pli was stabilized at 7.6-7.8 with NaOH. These
conditions were maintained throughout the test Aliquots (9.5 cm3) of cell cultures in the log
growth phase were added to conical glass llasks to which different concentrations of PFOA or
distilled water was added. Nominal test concentrations ranged from 0.000005-50 mM, or 0.0 to
20,703.5 mg I. PI OA. After 72-hours the number of cells was determined by measuring optical
density spectrophotometrically. The 72-hour ECso for growth inhibition was reported as 0.60
mM, or 248.4 mg/L PFOA. The EC so of 248.4 mg/L PFOA was not acceptable for quantitative
use, due to the short test duration but was retained for qualitative use.
H-10
-------
H.2.3.4 Dinoflagellate, Pyrocystis lunula
Hayman et al. (2021) conducted a short-term, sublethal 24-hour exposure to examine the
effects of PFOA on the bioluminescent dinoflagellate (Pyrocystis lunula) following ASTM
E1924-97 (ASTM 2004). Test solution water was 0.45 |im filtered seawater collected from North
San Diego Bay, CA spiked with PFOA. Spiking consisted of the addition of stock solutions of
PFOA (CAS # 335-67-1; 95% purity) dissolved in methanol; highest methanol concentration of
0.02% (v/v). Concentrations of PFOA for the toxicity tests were determined iVoin a range finding
study. Measured concentrations for PFOA were 0 (control and solvent control). I 5, 4.7, 11, 16,
29, and 52 mg/L. Approximately 3,000 cells of/'. lunula were added to 2.5 ml. of lest solution in
acrylic test cuvettes, with six replicates per treatment concentration. P. lunula were exposed for
24 hours in a 19°C incubator with a reversed (eg. dark during the typical "day" period) 12-
hour: 12-hour light:dark cycle. Test cuvettes were remo\ ed Irom the incubator after 24 hours and
after being in the dark period lor approximately three hours, inserted and analyzed in a
specialized spectrometer (Ouikl.ile 2"0 Biosensor System, Assure Controls, Carlsbad, CA) and
the light output was recorded. I.ess light output relative to concurrently evaluated controls is
indicative of an ad\ else effect The 24-hour bioluminescence EC so for P. lunula was determined
to be IS mg I. PFOA. The chronic value was used qualitatively and was not acceptable for
quantita1i\ e use because of the short exposure duration and lack of apical endpoints.
H.2.3.5 Golden brown alga, Isochrysis galbana
A 72-hour static, unmeasured algal growth inhibition test on PFOA (96% purity) with
Isochrysis galbana was performed by Mhadhbi et al. (2012) following OECD (2006) test
methodology. Golden brown algae were provided by Estacion de Ciencias Marinas de Toralla
(ECIMAT). The cultures were maintained in 250 mL glass Erlenmeyer flasks with autoclaved
filtered sea water and EDTA-free f/2 culture medium. PFOA stock solutions were prepared in
H-ll
-------
DMSO and added to the dilution water with a maximum DMSO concentration of 0.01% (v/v).
Nominal test concentrations were solvent control, 25, 50, 100, 200 and 400 mg/L PFOA. Each
flask was inoculated at a density of 10,000 cells/mL, with the algae in the exponential growth
phase. Each treatment was replicated three times. Flasks were kept at 20°C with a 24-hour light
period, and manually shaken daily. Cell counts were carried out e\ cry 24-hours, with a reported
72-hour ECso based on growth inhibition of 163.6 mg/L PFOA Results of this test were not
acceptable for quantitative use, due to the short test duration but was retained for qualitative use.
H. 2.3.6 Green alga, Chlorella vulgaris
Latala et al. (2009) conducted a 72-hour static algal growth inhibition test 011 I'FOA
(acid form, CAS # 335-67-1) with Chlorella vulgaris. The purity of PFOA was not reported, but
the authors stated they used the highest grade commercially a\ ailable Authors stated that the
algal growth inhibition test followed protocols found in ISO l<)253 (ISO 1995) and ISO 8692
(ISO 1993), but used I" 2 medium, different test species and a different photoperiod than the test
protocols. The green alga (vulgaris IJA-02 strain was isolated from Baltic Sea coastal waters
and maintained as monoalgal cultures in the Culture Collection of Baltic Algae (CCBA) at the
Institute of Oceanography: I ni\ ersity of Gdansk. Algae were batch-cultured in f/2 medium
prepared in distilled water and brought to a salinity of 8 PSU was using Tropic Marin® sea salt.
Cultures ^ere acclimated for ten days at 20°C and a 16-hour photoperiod (25 |imol photons m"2
s"1). The pH was stabilized at 7.6-7.8 with NaOH. These conditions were maintained throughout
the test. Aliquots (9.5 cm1) of cell cultures in the log growth phase were added to conical glass
flasks to which different concentrations of PFOA or distilled water was added. Nominal test
concentrations ranged from 0.000005-50 mM, or 0.0 to 20,703.5 mg/L PFOA. After 72-hours the
number of cells was determined by measuring optical density spectrophotometrically. The 72-
H-12
-------
hour ECso for growth inhibition was reported as 2.36 mM, or 977.2 mg/L PFOA. The EC50 of
977.2 mg/L PFOA was not acceptable for quantitative use, due to the short test duration but was
instead classified at qualitatively acceptable.
H.2.3.7 Diatom, Skeletonema marinoi
Latala et al. (2009) conducted a 72-hour static growth inhibition test on PFOA (acid
form, CAS # 335-67-1) with the diatom, Skeletonema marinoi The purity of PFOA was not
reported, but the authors stated they used the highest grade commercially a\ ail able. Authors
stated that the algal growth inhibition tests followed protocols found in ISO I "253 (TSO 1995)
and ISO 8692 (ISO 1993), but used f/2 medium, different test species and a different photoperiod
than the test protocols. The diatom S. marinoi BA-98 strain was isolated from Baltic Sea coastal
waters and maintained as monoalgal cultures in the C ulliire Collodion of Baltic Algae (CCBA)
at the Institute of Oceanography, University of (idansk Diatoms were batch-cultured in f/2
medium prepared in distilled water and brought to a salinity of 8 PSU was using Tropic Marin®
sea salt. Cultures were acclimated for ten days at 20°C and a 16-hour photoperiod (25 |imol
photons m"2 s"1) The pi I was stabilized at 7 (•> 7 8 with NaOH. These conditions were maintained
throughout the test Aliquols (l) 5 cm') of cell cultures in the log growth phase were added to
conical glass llasks to which different concentrations of PFOA or distilled water was added.
Nominal test concentrations ranged from 0.000005-50 mM, or 0.0 to 20,703.5 mg/L PFOA.
After 72-hours the number of cells was determined by measuring optical density
spectrophotometrically. The 72-hour EC50 for growth inhibition was reported as 0.89 mM, or
368.5 mg/L. The EC50 of 368.5 mg/L PFOA was not used quantitatively due to the short test
duration but was retained for qualitative use.
H-13
-------
Appendix I Acute-to-Chronic Ratios
1.1 Acute-to-Chronic Ratios from Quantitatively Acceptable Tests.
Species
Chemical
/ Purity
Acute
Test
Duration
Chronic
Test
Duration
Acute
Effect
Chronic Effect
Acute
Effect
Cone.
(mg/L)
Chronic
Effect
Cone.
(mg/L)
ACR
SMACR
Reference
Rotifer,
Brachionus
calyciflorus
PFOA
96%
24 hours
Up to
200
hours
LCso
EC10
(intrinsic rate of natural
increase)
150
0.5015
299.1
299.1
Zhang et al.
2013a
Water flea,
Daphnia carinata
PFOA
95%
48 hours
21 days
ECso
MATC
(# of average number of
offspring per brood and
total # of living offspring)
66.8
0.03162
2,113
2,113
Logeshwaran
et al. 2021
Cladoceran,
Daphnia magna
APFO
99.7%
48 hours
21 days
ECso
EC10
(average # of live young)
480
20.61
23.29
-
Colombo et
al. 2008
Cladoceran,
Daphnia magna
PFOA
Unreported
48 hours
21 days
ECso
(immobility)
EC10
(# young/starting female)
542.5
7.853
69.08
-
Ji et al. 2008
Cladoceran,
Daphnia magna
PFOA
>98%
48 hours
21 days
LCso
EC10
(# young/starting female)
193.3a
12.89
15.00
-
Li 2009,
2010
Cladoceran,
Daphnia magna
PFOA
99%
48 hours
21 days
LCso
EC10
(survival)
222.0
5.458
40.67
-
Yang et al.
2014
Cladoceran,
Daphnia magna
PFOA
98%
48 hours
21 days
ECso
MATC
(growth and reproduction)
114.6
0.07155
l,602b
-
Lu et al.
2016
Cladoceran,
Daphnia magna
PFOA
Unreported
48 hours
21 days
LCso
EC10
(# of offspring)
117.2
8.084
14.50
26.96
Yang et al.
2019
Cladoceran,
Moina macrocopa
PFOA
Unreported
48 hours
7 days
ECso
(immobility)
EC10
(mean young/adult)
166.3
2.194
75.80
75.80
Ji et al. 2008
1-1
-------
Species
Chcmic:il
/ Purity
Acute
Test
Duration
Chronic
Test
Durnlion
Acute
K fleet
Chronic KITect
Acute
KITect
Cone.
(mg/1.)
Chronic
KITect
Cone,
(mg/l.)
ACU
SMACK
Reference
Rainbow trout,
Oncorhynchus
mykiss
APFO
99.7%
96 hours
85 days
(ELS)
LCso
NOEC
(growth and mortality)
707
40
<17.68
<17.68
Colombo et
al. 2008
American bullfrog,
Lithobates
catesbeiana
PFOA
Unreported
96 hours
72 days
LC50
i.oi:c
(snout vcnl length)
1,006
0.288
3,493
3,493
Flynn et al.
2019
a Geometric mean of three LC50 values.
b Value not used in the SMACR calculation, because the value is an order of magnitude m eaier I lian other ACRs for the species.
1-2
-------
Appendix J Unused PFOA Toxicity Studies
J. 1 Summary of Unused PFOA Toxicity Studies
Author
Citation
Reason Unused
Arukwe, A. and A.S. Mortensen
2011. Lipid peroxidation and oxidative stress responses of salmon fed a diet
containing perfluorooctane sulfonic- or perfluorooctane carboxylic acids.
Comp. Biochem. Physiol. Part C 154: 288-295.
Force-fed (oral gavage); only one exposure
concentration
Consoer, D.M.
2017. A mechanistic investigation of perfluoroalkyl acid kinetics in rainbow
trout (Oncorhvnchus mvkiss). A dissertation submitted to the faculty of the
University of Minnesota.
Injected toxicant; only one exposure
concentration
Consoer, D.M., A.D. Hoffman, P.N.
Fitzsimmons, P.A. Kosian and J.W. Nichols
2014. Toxicokinetics of perfluorooctanoate (PFOA) in rainbow trout
(Oncorhvnchus mvkiss). Aquat. Toxicol. 156: 65-73.
All fish were surgically altered (dorsal
aortic cannula, plus a urinary catheter); no
controls; non-apical endpoints only
Cui, Y.. W. Liu, W. Xie, W. Yu, C. Wang and
H. Chen
2015. Investigation of the effects of perfluorooctanoic acid (PFOA) and
perfluorooctane sulfonate (PFOS) on apoptosis and cell cycle in a zebrafish
(Danio rerio) liver cell line. Int. J. Environ. Res. Public Health 12(12): 15673-
15682.
Excised cells (liver cell line)
De Silva, A.O., P.J. Tseng and S.A. Mabury
2009. Toxicokinetics of perfluorocarboxylate isomers in rainbow trout.
Environ. Toxicol. Chem. 28(2): 330-337.
Study involved a mixture of ECF PFOA,
linear PFNA, and isopropyl PFNA added to
diet
Fernandez-Sanjuan, M., M. Faria, S. Lacorte
and C. Barata
2013. Bioaccumulation and effects of perfluorinated compounds (PFCs) in
zebra mussels (Dreissenapolvmorphd). Environ. Sci. Pollut. Res. 20:2661-
2669.
Mixture
Gonzalez-Naranjo, V. and K. Boltes
2014. Toxicity of ibuprofen and perfluorooctanoic acid for risk assessment of
mixtures in aquatic and terrestrial enviromnents. Int. J. Environ. Sci. Technol.
11: 1743-1750.
Severe lack of exposure details (cannot
judge against data quality objectives)
Gorrochategui, E., S. Lacorte, R. Tucker and
F.L. Martin
2016. Perfluoroalkylated substance effects in A'enopus laevis A6 kidney
epithelial cells determined by ATR-FTIR spectroscopy and chemometric
analysis. Chem. Res. Toxicol. 29: 924-932.
The tests were performed on cell cultures
obtained from an outside source; whole
organisms were not investigated
Holth, T.F., M. Yazdani, A. Lenderink and K.
Hyllan
2012. Effects of fluoranthene and perfluorooctanoic acid (PFOA) on immune
functions in Atlantic cod (Gadus morhua). Abstracts Comp. Biochem. Physiol.
Part A. 163: S39-S42.
Abstract only; cannot judge against data
quality objectives
Jantzen, C.E., K.M. Annunziato and K.R.
Cooper
2016. Behavioral, morphometric, and gene expression effects in adult zebrafish
(Danio rerio) embryonically exposed to PFOA, PFOS, and PFNA. Aquatic
Toxicology. 180:123-130.
Single concentration test where exposure to
PFOA was of an acute (117-hours) duration
but endpoints were measured at 6 months of
age
Jantzen, C.E., F. Toor, K.M. Annunziato and
K.R. Cooper
2017a. Effects of chronic perfluorooctanoic acid (PFOA) at low concentration
on morphometries, gene expression, and fecundity in zebrafish (Danio rerio).
Reproduct. Toxicol. 69: 34-42.
Unable to determine dietary exposure
concentration
J-l
-------
Author
( iliiliun
Ko;isoii I niisod
Khan, E.A., X. Zhang, E.M. Hanna, F.
Yadetie, I. Jonassen, A. Goksoyr, and A.
Arukwe
2021. Application of Quantitative Transcriptomics in Evaluating the Ex Vivo
Effects of Per- and Polyfluoroalkyl Substances on Atlantic Cod (Gadus
morhua) Ovarian Physiology. Sci. Total Environ.755(l): 11 pp.
In vitro exposure
Lee, W. and Y. Kagami
2010. Effects of perfluorooctanoic acid and perfluorooctanc sulfonate on gene
expression profiles inmedaka (Oryzias latipes). Abstracts. Toxicol. Letters
196S: S37-S351.
Abstract only, cannot judge against data
quality objectives
Li, M.H.
2011. Changes of cholinesterase and carboxylcslcrasc activities in male
guppies, Poecilia reticulata, after exposure to ammonium perflikH'iuvlauoate,
but not to perfluorooctane sulfonate. Fresenius Environ. Bull. 20(8;i) 2(N>5-
2070.
Each treatment group for PFOA was ran
two times at separate times (not
simultaneously) and the sample size for
each treatment group was unclear.
Liang, X. and J. Zha
2016. Toxicogenomic applications of Chinese rare minnow (Gobiocypris
rarus) in aquatic toxicology. Comp. Bioclicm Physiol. Pari D 19: 174-1 so
Review paper
Liu, C., Y. Du and B. Zhou
2007a. Evaluation of estrogenic activities and mechanism of action of
perfluorinated chemicals determined by vitellogenin induction in primary
cultured tilapia hepatocytes. Aquat. Toxicol. 85: 267-277.
In vitro, cultured hepatocytes
Liu, C., K. Yu, X. Shi, J. Wang, P.K.S. Lam,
R.S.S. Wu and B. Zhou
2007b. Induction of oxidative stress and apoptosis by PI '< )S and PFOA in
primary cultured hepatocytes nf freshwater tilapia (Orcm hmmi.*. niloticus).
Aquat. Toxicol. 82: 135-143
In vitro, cultured hepatocytes
Mahapatra, C.T., N.P. Damayanti, S.C.
Guffey, J.S. Serafin, J. Irudayaraj, and M.S.
Sepulveda
2017. Comparative in vitro to\ialv assessment of perfliionnated carboxylic
acids. Journal of Applied Toxicology 37: 699-708
In vitro exposure, zebrafish liver cell
cultures
Martin, J.W., S.A. Mabury, K.R. Solomon and
D.C.G. Muir
2003b. Bioconcentralion and tissue distribution of pcrlTuorinaled acids in
rainbow trout (Oncorhvnchus mvkiss). Environ. Toxicol. Chem. 22: 196-204.
Bioaccumulation (steady state not
documented); only 12 days
Martin, J.W., S.A. Mabury, K.R. Solomon and
D.C.G. Muir
2013a. Progress toward understanding the bioaccumulation of perfluorinated
alkvl acids. Environ. Toxicol. Chcin. 32(11): 2421-2423
Review paper
Mortensen, A.S., R.J. Letcher, M.V.
Cangialosi, S. Chu, and A. Arukwe
2011. Tissue bioaccumulatiou pallcius. xenobiotic biotransformation and
steroid hormone levels in Allauiic salmon (Salmo salar) fed a diet containing
perfluoroactanc sulfonic or perfluorooctane carboxylic acids. Chemosphere 83:
1035-1044.
One dietary dosage level provided over a 6-
day period; not intended as a toxicity test
Padilla, S., D. Coram, B. Padros, D 1. 1111nici.
A. Beam, K.A. Houck, N. Sipes, N.
Kleinstreuer, T. Knudsen, D.J. Nix and DM
Reif
2012. Zcbnilish developmental screening of the ToxCastTM Phase I chemical
library. Repiod Toxicol. 33: 174-187.
Severe lack of exposure details, only one
exposure concentration
Popovic, M, R. Zaja, K. Fent and T. Smital
2(i 14. Inteniclinn of environmental contaminants with zebrafish organic anion
lianspiHiing polypeptide, Oatpldl (Slcoldl). Toxicol. Appl. Pharmacol.
280(h 149-158.
Excised cells
Prosser, R.S., K. Mahon, P.K. Sibley, D.
Poirier and T. Watson-Leung
2016. Bioaccumulation of perfluorinated carboxylates and sulfonates and
polychlorinated biphenyls in laboratory-cultured Hexagenia spp., Lumbriculus
variegatus and Pimephales promelas from field-collected sediments. Sci. Total
Environ. 543: 715-726.
Mixture (filed collected sediment, contained
PFAS mixtures and PCBs)
J-2
-------
Author
( iliiliun
Reason I nusod
Rotondo, J.C, L. Giari, C. Guerranti, M.
Tognon, G. Castaldelli, E. A. Fano and F.
Martini
2018. Environmental doses of perfluorooctanoic acid change the expression of
genes in target tissues of common carp. Environ. Toxicol. Chem. 37(3): 942-
948.
Two exposure concentrations 10,000-fold
apart; atypical endpoint
Sanderson, H., T.M. Boudreau, S.A. Mabury
and K.R. Solomon
2003. Impact of perfluorooctanoic acid on the structure of the zooplanklon
community in indoor microcosms. Aquat. Toxicol. 62: 227-234.
Poor experimental design/performance
Stevenson, C.N., L.A. MacManus-Spencer, T.
Luckenbach, R.G. Luthy and D. Epel
2006. New perspectives on pefluorochemical ccoloxicology: inhibition and
induction of an efflux transporter in marine mussel, Mvtilus califbmianus.
Environ. Sci. Technol. 40: 5580-5585.
Excised cells (gills)
Tang, J., X. Jia, N. Gao, Y. Wu, Z. Liu, X. Lu,
Q. Du, J. He, N. Li, B. Chen, J. Jiang, W. Liu,
Y. Ding, W. Zhu and H. Zhang
2018. Role of the Nrf2-ARE pathway in perfluorooctanoic acid (PF() \ )-
induced hepatotoxicity inRana nigromaculaia. Environ. Pollut. 238: 1 <>"5-
1043.
No apical endpoints were measured; control
survival was not reported; test duration of
14 days relatively short for a chronic
amphibian study; not NA species
Thienpont, B., A. Tingaud-Sequeira, E. Prats,
C. Barata, P.J. Babin and D. Raldua
2011. Zebrafish eleutheroembryos provide a smiahlc \ cricbrale model for
screening chemicals that impair thyroid hormone s\ nihesis. Environ. Sci.
Technol. 45(17): 7525-7532.
Only one exposure concentration; no apical
endpoints
Ulhaq, M., S. Orn, G. Carlsson, J. Tallkvist
and L. Norrgren
2012. Perfluorooctanoic acid toxicity in zebrafish (Danio rerio). Abstracts.
Toxicol. Letters 21 IS: S43-S216.
Abstract only, cannot judge against data
quality objectives
Williams, T.D., A. Diab, F. Ortega, V.S.
Sabine, R.E. Godfrey, F. Falciani, J.K.
Chipman, and S.G. George
2008. Transcriptomic Responses of European llounder
-------
Author
( iliiliun
Ko;isoii I misod
Zhang, H., J. He, N. Li, N. Gao, Q. Du, B.
Chen, F. Chen, X. Shan, Y. Ding, W. Zhu, Y.
Wu, J. Tang and X. Jia
2019b. Lipid accumulation responses in the liver of Rana nigromaculata
induced by perflurooctanoic acid (PFOA). Ecotoxicol. Environ. Saf. 167: 29-
35.
No apical endpoints were measured; control
survival was not reported; test duration of
14 days relatively short for a chronic
amphibian study; not NA species
J-4
-------
Appendix K EPA Methodology for Fitting Concentration-
Response Data and Calculating Effect Concentrations
Toxicity values, including LCso and ECio values, were independently calculated from the
data presented in the toxicity studies meeting the inclusion criteria described above (see Section
2.10.3.3) and when adequate concentrations-response data were published in the study or could
be obtained from authors. When concentration-response data were 1101 presented in toxicity
studies, concentration-response data were requested from study authors lo independently
calculate toxicity values. In cases where study authors did not respond to \:.I'.Vs request for data
or were unable to locate concentration-response data, the toxicity values were not independently
calculated by EPA, and the reported toxicity values ^ere retained for criteria deviation. EPA also
retained author-reported effect concentrations u hen data a\ ailahility did not support effect
concentration calculation by EPA. This retention was done to be consistent with use of author-
reported toxicity values in previous criteria documents and retain informative toxicity values
(that would have olherw ise not been used only on the basis of lacking the underlying C-R data).
Where concentration-response data were a\ ailablc. they were analyzed using the statistical
software program R (\ersion 3 (•> 2) and the associated dose-response curve (drc) package.
In some cases, the author-reported toxicity values were different than the corresponding
effect concentrations calculated by EPA. Overall, the magnitude of such discrepancies was
limited and largely occurred lor several potential reasons such as: (1) instances where authors
were presumed to calculate effect concentrations using replicate level data but EPA only had
access to treatment mean data; (2) the model selected to fit a particular set of C-R data, and; (3)
the software used to fit a model to C-R data and calculate an effect concentration.
K-l
-------
K.l Fitting Concentration Response Data in R
Concentration-response data were obtained from quantitatively-acceptable toxicity
studies when reported data were available. In many scenarios, toxicity studies report treatment-
level mean concentrations and mean organismal responses; however, individual-replicate data
may also be reported. When fitting C-R curves, replicate-level data were preferred over
treatment-level data, if both types of data were available. Within R. the drc package can fit a
variety of mathematical models to each set of C-R data.
K.l.l Fitting Acute Mortality Data
K. 1.1.1 Dichotomous Data
Dichotomous data are binary in nature (e u . Ii\ e dead or <> I) and are typical of survival
experiments. They are usually represented as a proportion snr\ i\ ed
K.1.2 Fitting Chronic Growth. Reproduction, and Snr\ival Data
K. 1.2.1 Continuous Data
Continuous data lake on any \ alue along the real number line (e.g., biomass).
K. 1.2.2 Count I hint
Count data lake on only integer \ allies (e g., number of eggs hatched).
K. 1.2.3 / hchotomons / kna
Dichotomous data are binary in nature (e.g., live/dead or 0/1) and are typical of survival
experiments. They are usually represented as a proportion survived.
K.2 Determining Most Robust Model Fit for Each C-R curve
The R drc package was used to fit a variety of models to each individual C-R dataset. A
single model was then selected from these candidate models to serve as the representative C-R
model. The selected model represented the most statistically-robust model available. To
K-2
-------
determine the most-statistically-robust model for a C-R dataset, all individual model fits were
assessed on a suite of statistical metrics.
K.2.1 Selecting Candidate Models
Initially, models were ranked according to the Akaike information criteria (AIC). The AIC
provides a measure of the amount of information lost for a given model by balancing goodness
of fit with model parsimony. The models with the lowest AIC, relali \ e to other models based on
the same data, tend to be optimal. In some instances, however, the model with the lowest AIC
possessed a questionable characteristic that suggested said model was not the most appropriate.
Rather than selecting a model based solely on the lowest AIC, the initial ranking slop was only
used to identify a subset of candidate models that were more closely examined before selecting a
model fit for each C-R dataset.
K.2.2 Assessment of Candidate Models lo Determine the Most Appropriate Model
Candidate models (i.e., models with low AIC scores relative to other models produced for
a particular C-R dataset) were further evaluated based 011 additional statistical metrics to
determine a single, statistically robust curve for each quantitatively-acceptable toxicity test.
These additional statistical metrics were evaluated relative to the other candidate curve fits
produced lor each C-R dataset Of these statistical metrics, residual standard errors, confidence
inten als relati\ e to effects concentration estimates, and confidence bands carried the most
weight in determining the most appropriate model to be representative of an individual C-R
dataset. These additional statistical metrics included:
K. 2.2.1 Comparison of residual standard errors
As with AIC, smaller values were desirable. Residual standard errors were judged
relative to other models.
K-3
-------
K. 2.2.2 Width of confidence intervals for EC estimates
Confidence intervals were assessed on standard error relative to estimate and confirming
that the intervals were non-negative. Judged in absolute and relative to other models.
K. 2.2.3 Width of confidence bands around the fitted model
A general visual inspection of the confidence bands for the fitted model. Wide bands in
the area of interest were undesirable. Judged in absolute and rc1aii\ e lo other models.
K. 2.2.4 P-values ofparameters estimates and goodness of fit tests
Hypothesis tests of parameter values to determine whether an 68111111110 is significantly
different from zero. Goodness of fit tests were used to judge the overall performance of the
model fit. Typically, the level of significance was set at 0.05. There may have been occasional
instances where the 0.05 criterion may not lx- met. but there was lilllc recourse for choosing
another model. Judged in absolute terms.
K.2.2.5 Residue il pit >is
Residuals ^ere examined lor homoscedasticity and biasedness. Judged in absolute and
relative to other models
K.2.2. (> ()vcrly influential observations
()hser\ alions were judged based on Cook's distance and leverage. When an observation
was deemed o\ erly influential. it was not reasonable to refit the model and exclude any overly
influential observations gi\ en the limited data available with typical C-R curves. Judged in
absolute terms.
K.3 Determining Curve Acceptability for use in Criteria Derivation
The final curve fits selected for each of the quantitatively-acceptable toxicity tests were
further evaluated and classified to determine whether the curves were: 1) quantitatively-
K-4
-------
acceptable for use, 2) qualitatively acceptable for use, or 3) unacceptable. To determine curve
acceptability for use in deriving an effect concentration, each individual curve was considered
based on the statistical metrics described above and assessed visually to compare how the
calculated effect concentration aligned with the underlying raw C-R data. Instead of evaluating
curves fits relative to other curve fits for the same data (as was previously described to select the
most-robust curve for each test), curve fit metrics were used to assign each curve a score:
• Quantitatively Acceptable Model. Model performed well on mosi/all statistical
metrics and resultant effect concentrations were typically used in a quantitative
manner.
• Qualitatively Acceptable Model. Model generally performed well on statistical
metrics; however, the model presented some chai aclerislic( s) that called estimates
into question. Such models were considered with caution. These problems may have
consisted of any number of issues such as a para meter with a high p-value, poor
goodness of 111 p-\ nine, wide confidence hands for fit or estimate interval, or
residuals thai indicate model assumptions are not met. Broadly, effect concentrations
from models that were deemed qualilati\ ely acceptable were not used numerically in
criteria derivation if quantitatively acceptable models for different endpoints or tests
from the same publication were available. If quantitatively acceptable models for
different endpoints or tests from the same publication were not available, effect
concentrations from the qualitatively acceptable model were used numerically in
criteria derivation on a case-by-case basis.
• Unacceptable Model. Model poorly fit the data. These models were not used for
criteria derivation.
K-5
-------
No single statistical metric can determine a given model's validity or appropriateness. Metrics
should be considered as a whole. As such, there is a slightly subjective component to these
evaluations. That said, this assessment methodology was developed to aid in evaluating models
as to their quantitative or qualitative attributes in a transparent and relatively repeatable manner.
K-6
-------
Appendix L Derivation of Acute Protective PFOA Benchmarks
for Estuarine/Marine Waters through a New Approach
Method (NAM)
L. 1 Background
The 1985 Guidelines for Deriving Numerical National Water Quality Criteria for the
Protection of Aquatic Organisms and Their Uses (U.S. EPA 1985) recommend that data for a
minimum of eight families be available to fulfill taxonomic minimum data requirements (MDRs)
to calculate criteria values, including to calculate estuarine/marine aquatic life criteria. Acute
estuarine/marine test data are currently available for only three families, a mysid (Siriella and
Americamysis), a sea urchin (Strongylocentrotus), and a mussel (Mytihis), addressing only three
of the eight MDRs; thus, EPA was not able to derive an acute estuarine/marine criterion element
for PFOA based on the 1985 Guidelines \ II)R specifications (Section 2.10.1). However, EPA
was able to develop a draft acute PFOA protecti\ c benchmark using a New Approach Methods
(NAMS) process. \ ia the application of Interspecies Correlation Estimation (ICE) models
(Raimondo et al 2<)|<)) Although not a criterion hascd on 1985 Guidelines specifications,
because of gaps in a\ ailaMc data for sc\ cral of the taxonomic MDRs listed in the 1985
Guidelines for the dcri\ ation of aquatic lile criteria, this benchmark represents an aquatic life
value deri\ cd to he protec1i\ e of aquatic communities. The ICE model predictions supplement
the available test dataset to help fill missing MDRs and allow the derivation of acute
estuarine/marine hcnchmark recommendations for aquatic life using procedures consistent with
those in the 1985 Guidelines. This is important as it provides an approach by which values that
are protective of aquatic life communities can be developed, even when MDRs are not fulfilled
by direct PFOA test data. This approach is consistent with both the 1985 Guidelines "good
science" clause, EPA's interest in providing useful information to states and tribes regarding
L-l
-------
protective values for aquatic life, and EPA's intention to reduce the use of animal testing via
application of NAMS (https://www.epa.gov/chemical-research/epa-new-approach-methods-
work-plan-reducing-use-animals-chemical-testing).
L.l.l Introduction to Web-ICE
ICE models, developed by EPA's Office of Research and Development, are log-linear
regressions of the acute toxicity (EC50/LC50) of two species across a range of chemicals, thus
representing the relationship of inherent sensitivity between Ihosc species (Raimondo et al.
2010). Each model is derived from an extensive, standardized database of acute toxicity values
by pairing each species with every other species for which acceptable toxicity data are available.
Once developed, ICE models can be used predict the sensiii\ iiy of an untested taxon (predicted
taxa are represented by the y-axis) from the know 11. measured sensitivity of a surrogate species
(represented by the x-axis; Figure L-l).
ICE models have been developed lor a hroad range of different chemicals (e.g., metals
and other inorganics, pesticides. sol\ cuts, and reactive chemicals) and across a wide range of
toxicity values. There are approximately 3.4<~><~> significant ICE models for aquatic animal and
plant species in the most recent \ersion of Web-ICE (v3.3, https://www3.epa.gov/webiceA last
updated June 2016; Raimondo et al 201 5)
Models were validated using leave-one-out cross validation, which formed the basis for
the analyses of uncertainly and prediction robustness. For this process, each datapoint within the
model (representing the relative sensitivity of two species for a particular chemical) is
systematically removed, one at a time. The model is then redeveloped with the remaining data
(following each removal) and the removed value of the surrogate species is entered into the
model. The estimated value for the predicted species is then compared to the measured value for
that species (Raimondo et al 2010; Willming et al. 2016).
L-2
-------
ICE models have high prediction accuracy when values are derived from models with
robust parameters (e.g., mean square error, R2), that fall within a defined range of acceptability,
and with close prediction confidence intervals that facilitate evaluating the fit of the underlying
data (Brill et al. 2016; Raimondo et al. 2010; Willming et al. 2016). Results of these analyses
provide the basis of the user guidance for selecting ICE predicted toxicity with high confidence
(Box 1).
ICE models have undergone extensive peer review and their use luis been supported for
multiple applications, including direct toxicity estimation for endangered species (NRC 2013;
Willming et al. 2016) and development of Species Sensitivity Distributions (SSDs) (Awkerman
et al. 2014; Bejarano et al. 2017; Dyer et al. 2006, 2008, Raimondo et al. 2010, 2020). The
application of ICE-predicted values to de\ clop protective at|imlic life values by multiple
independent, international groups confirms thai \ allies de\ eloped from ICE-generated SSDs
provides a level of protection thai is consistent with using measured laboratory data (Dyer et al.
2008; Feng et al. 201 .V I 'ojut et al 2d 12a. 2d 12b; Palumbo et al. 2012; Wu et al. 2015, 2016;
Wang et al 2D2D; /hang et al 2d I 7) A recent external review of ICE models additionally
supports their use in regulatory applications based on the reliability of underlying data, model
transparency, stati stical robustness, predictive reliability, proof of principle, applicability to
probabilistic approaches, and reproducibility of model accuracy by numerous independent
research teams (licjarano and Wheeler 2020).
L-3
-------
c
_g
4-
c
2
-1 -
0 2 4
6
Z
Rainbow trout (log LC50)
Figure L-l. Example ICE Model for Rainbow Trout (surrogate) and Atlantic Salmon
(predicted).
Each model datapoint is a common chemical that was tested in both species to develop a log-linear regression.
Box 1. ICE Model User Guidance Recommended for Listed Species (Willming et al 2016):
Close taxonomic distance (within class)
• Low Mean Squared Error (<~ 0.95)
• High R2 (>~ 0.6)
• High slope (>~ 0.6)
• Prediction confidence intervals should be used to evaluate the prediction using
professional judgement for the application (Raimondo et al. in prep).
• For models between vertebrates and invertebrates, using those with lower MSE or
MOA-speciftc models (not available for PFAS) has been recommended for listed
species predictions (Willming et al. 2016).
L.1.2 Application of Web-ICE with PFOA
As previously discussed, ICE models have been developed using a diversity of
compounds (e.g., metals and other inorganics, pesticides, solvents, and reactive chemicals)
L-4
-------
across a wide range of toxicity values; however, PFAS are not included in Web-ICE v3.3 due to
the lack of available toxicity data at that time. PFAS acute values (typically reported as mg/L)
can be greater than those used to develop an ICE model (ICE database toxicity range IE"4 to IE8
(j,g/L) such that the input PFAS value of the surrogate would be outside the model domain. In
these cases, a user can either enter the value as [j,g/L and allow the model to extrapolate beyond
its range or enter the toxicity as a "scaled" value (i.e., estimate the \ alLie as mg/L). The principal
assumptions of ICE models are: 1) they represent the relationship of in herein sensitivity between
two species, which is conserved across chemicals, mechanisms of action, and ranges of toxicity
and; 2) the nature of a contaminant that was tested 011 the surrogate reflects the nature of the
contaminant in the predicted species (e.g., effect concentration | ECso] or lethal concentration
[LC50]), percentage of active ingredient, technical grade: Raimondo et al. 2010). While neither of
these assumptions are violated by either extrapolating heyond the range of the model or using
scaled toxicity data, the uncertainty of using ICI- models in either manner had not been
thoroughly evaluated Additionally, since PI AS were not included in the database used to
develop Web-ICI- \ 3 3. the \ alidation of ICI' models to accurately and specifically predict to
these com pounds has not heen pre\ iously explored. We address both these topics in the sections
below
L.2 Prediction Accuracy of Web-ICE for Scaled Toxicity and Values
Beyond the Model Domain
The accuracy of using scaled toxicity data as input into ICE models was evaluated using
an analysis with the existing ICE models (v3.3) and is described in detail in Raimondo et al. (in
prep). Briefly, ICE models containing a minimum of 10 datapoints and spanning at least five
orders of magnitude were separated into two subsets: 1) a lower subset that contained all paired
chemical data corresponding to values below the 75th percentile of surrogate species values and;
L-5
-------
2) an upper subset containing paired chemical data above the 75th percentile of surrogate values.
The lower subset was used to develop "truncated" ICE models. The surrogate species values in
the upper subset were converted to mg/L and entered into the truncated ICE models. The
predicted mg/L value was compared to the respective value of the measured predicted species.
Prediction accuracy was determined as the fold difference (maximum of the predicted/measured
and measured/predicted) between the predicted and the measured \ alue, consistent with
previously published evaluations of ICE models (Raimondo el ill 201'). Willmiug et al. 2016).
Accuracy of using scaled toxicity as input into ICE models was compared lo o\ em 11 TCE
prediction accuracy as previously reported and prediction accuracy of the respeeli\ e upper subset
data points that were entered into the models as |ig/L (i.e., values beyond the model domain). A
total of 3,104 datapoints from 398 models were ewiluated. A match-paired comparison showed
that the average fold differences of toxicity \ allies predicted using scaled toxicity was not
significantly different than the respective average Ibid differences of all cross-validated data
points reported in ^N\ i 11 nii nu et al (2d 16) (Wilcoxon paired rank sum test, V = 42741, p-value
0.11). Additionally. Raimondo et al (2<)|0) aiicl Willming et al (2016) showed a consistent and
reproducible relationship Ix-tween the taxonomic distance of the predicted and surrogate species,
which was also reproduced using scaled values; the percentage of datapoints predicted using
scaled toxicity was within ll\ e-Pold of the measured value for over 94% of all validated
datapoints for species pairs within the same order, with a reduction in accuracy coinciding with
decreasing taxonomic relatedness (Raimondo et al. in prep). Comparison of scaled values with
those predicted from [j,g/L values beyond the model domain showed that predicted values varied
by a factor of 10 for models with slopes ranging from 0.66 - 1.33. Toxicity values predicted
from models with slopes within this range had a median fold difference of 2.4 using mg/L values
L-6
-------
and 2.8 using |ig/L values (Wilcoxon paired rank sum test, V = 1334749, p-value 0.77). These
results and a detailed review of ICE model assumptions are provided in Raimondo et al. (in
prep).
L.3 Direct Comparison of Web-ICE and Measured Toxicity Values
Since limited PFOA toxicity test data are available for estuarine/marine species, the
ability of ICE models to predict PFOA toxicity was evaluated using direct comparisons of
freshwater species sensitivity as reported in the draft criteria document and predicted by Web-
ICE. In this comparison, the measured SMAVs for PFOA reported in Appendix A.l and
Appendix B. 1 were used as values for surrogate species to predict to all possible species that also
had a measured PFOA SMAV reported. The available SMAVs for PFOA that could be used as
ICE surrogate values, along with the number of ICE models corresponding to each surrogate, are
shown in Table L-l.
Table L-l. Surrogate Species Measured Values for PFOA and Corresponding Number of
ICE Models for Each Surrogate.
For example there are 53 species for which Daphnia magna can predict toxicity.
Broad Taxon
Species
PFOA SMAV
(mg/L)
Number of ICE
Models
Common Name
Scientific
Amphibian
Bullfrog
Lithobates catesbeiana a
1020
9
Amphibian
Clawed frog
Xenopus sp. b
377
2
Crustacean
Mysid
Americamysis bahia
24
28
Crustacean
Water flea
Daphnia magna
220
53
Fish
Zebrafish
Danio rerio
572.4
2 (juvenile models)
6 (embryo models)
Fish
Bluegill
Lepomis macrochirus
664
68
Fish
Rainbow trout
Oncorhynchus mykiss
1682
77
Fish
Fathead minnow
Pimephales promelas
413.2
74
Mollusc
Fatmucket
Lampsilis siliquoidea
164.4
29
Mollusc
Black sandshell
Ligamia recta
161
1
a Lithobates catesbeianus was used in Web-ICE
b Xenopus laevis was used in Web-ICE
L-7
-------
Table L-2 shows direct comparisons for PFOA measured and ICE-predicted values. The
regressions for these comparisons are provided in Section L.6. Comparisons are limited by the
number of measured toxicity values and models available. To be included in this comparison, a
measured value was needed for both species in an ICE model pair. For direct comparison of
predicted and measured PFOA values, the measured SMAV of the surrogate species is entered
into a model for which the measured SMAV for the intended predicted species is also known.
The PFOA toxicity predicted to this model is then compared to the measured SMAV for the
predicted species listed in Appendix A.l, Appendix B. 1 and Table L-l. This allows both species
of an ICE model to serve as either the predicted or surrogate species. The exception to this was
in cases involving zebrafish embryos, as Web-ICE v3.3 only included models for which
zebrafish embryos were used as surrogates Accuracy of 1CI- predictions are presented as the
"fold-difference" between the measured and the predicted species, such that fold difference is the
maximum of the ratio of the predicted LCso/measured I.C*.. or measured LCso/predicted LCso.
Analyses of ICE prediction accuracy ha\ e show 11 that ICE models over- and under-estimate
toxicity values randomly, i e . there is 110 systematic bias associated with the models (Table L-2,
Raimondo et al. 2<)|o. Raimondo et al in prep) I-'or accuracy assessments, the fold difference
provides a si 111 piilied metric to easily see how close predictions are to measured values at a
glance. A fix e-lbld difference has been demonstrated to be the average interlaboratory variability
of acute aquatic toxicity tests and represents a conservative amount of variance under
standardized test conditions for a given life stage (Fairbrother 2008; Raimondo et al. 2010). This
inter-test variation can increase significantly where experimental variables differ between tests;
however, all ICE models are based on standardized life stages to minimize extraneous variability
(Raimondo et al. 2010).
L-8
-------
Table L-2. Comparison of ICE-predicted and measured values of PFOA for species using both scaled values (entered as mg/L) and values
potentially beyond the model domain (entered as jig/L) (Raimondo et al. in prep).
Measured SMAVs are for the predicted species as listed in Appendix A. I, Appendix B. 1 and Table L-l. Footnotes indicate where predictions or models do not meet one or more
Prcdiclcd Species
SiirmusiU* Species
lo\icil\ \'sillies Pulcnlhilh lk'\
-------
l o\icil\ \ .lines Pokn(i;ill\ lk\ond Model l)om;iin
Sc;ilcd
l'o\icil\ Yiilncs
Measured
\\eh-KT.
( (inCidcncc
Measured
Weli-ICI-:
SMAY
Prcdiclcd
lnlcr\;d
l-nld
SMAN
Predicled
Confidence
l-old
Prcdiclcd Species
Simto»;iIc Species
<.u»/l.)
(H!i/I.)
(uii/l.)
Difference
(mu/l.)
(mu/l.)
lnlcr\iil (inii/l.)
Difference
Bluegill
(Lepomis macrochirus)
Bullfrog
(Lithobates catesbeianus)
664,000
1501135.53
114395.51 -
19698393.38
2.3V1'
664
601.13
166.16-2174.76
1.1
Daphnid
(Daphnia magna)
81671.31
52906.69 -
126074.85
8.1
801.26
627.74 - 1022.74
1.2
Fathead minnow
(Pimephales promelas)
155867.21
82472.16-
294579.23
4.3
255.38
171.26- 380.83
2.6
Fatmucket
(Lampsilis siliquoidea)
408413.42
32355.12-
5155335.38
1.6ac
295.92
21.91 -3996.50
2.2ac
Mysid
(Americamysis bahia)
23663.56
13808.02 -
40553.53
28.1
266.08
195.07- 362.94
2.5
Rainbow trout
('Oncorhynchus mykiss)
1388883.84
1036631.76 -
1860832.75
2.1
2138.11
1903.46-2401.68
3.2
Zebrafish embryo
(Danio rerio - embryo)
347447.26
41366.56-
2918288.94
l.9a
274.69
48.73 - 1548.44
2.4
Rainbow trout
(Oncorhynchus mykiss)
Bluegill
(Lepomis macrochirus)
1,682,000
353682.57
274243.78 -
456131.98
4.8
1,682
524.72
466.57- 590.13
3.2
Bullfrog
(Lithobates catesbeianus)
781571.06
313951.07 -
1945695.88
2.2
376.93
177.31 - 801.30
4.5
Daphnid
(Daphnia magna)
56931.33
35836.37 -
90443.76
29.5C
630.92
485.50- 819.91
2.T
Fathead minnow
(Pimephales promelas)
145155.57
92010.78 -
228996.44
11.6
179.81
133.78-241.68
9.4
Fatmucket
(Lampsilis siliquoidea)
736514.01
14043.59 -
38626348.23
2 ^abc
201.6
39.91 -1018.32
8.3C
Mysid
(.Americamysis baliia )
17330.67
9740.08 -
30836.72
97.1
192.8
137.35 -270.65
8.7
Zebrafish embryo
(Danio rerio - embryo 1
410453.8
77544.98 -
2172575.38
4.1
121.34
42.07- 349.98
13.9
Fathead minnow
(Pimephales promelas)
African clawed frog
(Xenopus laevis)
413.200
524824.76
4876.97 -
56477848.18
1.3ab
413.2
257.81
16.14-4116.88
1.6a
Bluegill
(Lepomis macrochirus)
387148.61
201381.26-
744279.98
1.1
1427.45
1024.71 - 1988.48
3.5
Bullfrog
(Lithobates catesbeianus)
1012520.93
299871.50-
3418793.08
2.5
758.77
320.24 - 1797.81
1.8
Daphnid
(Daphnia magna)
111877.81
66019.39-
189590.43
3.7C
1709.52
1209.11 -2417.01
4.1c
Fatmucket
(Lampsilis siliquoidea)
675590
54952.40 -
8305767.53
1.6ac
3450.51
576.51 -20651.62
8.4C
Mysid
(Americamysis bahia)
34114.93
11148.28 -
104395.31
12.1°
635.95
321.17- 1259.27
1.5C
Rainbow trout
(Oncorhynchus mykiss)
1557418.83
896689.52 -
2705009.20
3.8
3950.04
3110.36- 5016.41
9.6
L-10
-------
Predicted Species
Simto»;iIc Species
To\icil\ \'sillies Polcnlhilh licwind Model Doniiiin
Sciiled l o\icil\ \ ;ilnes
Measured
SMW
(.u»/l.)
\\eh-KT.
Predicted
(H!i/I.)
( (incidence
lnlcr\;il
(uii/l.)
l-nld
Difference
Mciisured
SMW
(nili/l.)
Weli-ICI-:
Predicted
(mu/l.)
( (inCidcncc
lnlcr\iil (inii/l.)
l-old
Difference
Zebrafish embryo
(Danio rerio - embryo)
498531.23
249260.29 -
997083.75
1.2
901.61
385.51 -2108.61
2.2
Fatmucket
(Lampsilis siliquoidea)
Black sandshell
(Ligumia recta)
164,400
153582.08
28669.47 -
822737.61
1.1
164.4
109.67
9.40 - 1278.86
1.5a
Bluegill
(Lepomis macrochirus)
97837.12
12334.24 -
776059.00
1 7ac
733.53
111.00-4847.23
4.5C
Daphnid
(Daphnia magna)
69944.45
23967.21 -
204121.63
2.4
406.14
160.97- 1024.70
2.5
Fathead minnow
(Pimephales promelas)
17586.46
2323.63 -
133103.22
9 3ac
78.68
7.86 -787.28
2.1ac
Rainbow trout
(Oncorhynchus mykiss)
26218.81
2516.16-
273203.88
6.3abo
729.52
264.03 -2015.66
4.4C
Black sandshell
(Ligumia recta)
Fatmucket
(Lampsilis siliquoidea)
161,000
171536.98
34311.02 -
857594.23
1.1
161
237.72
25.46 -2219.07
1.5a
a Confidence interval >1.5 order magnitude.
b Input data outside model range.
c Guidance for model mean square error, R2, and/or slope not met.
L-ll
-------
These comparisons are consistent with Web-ICE user guidance, previously published
reports on ICE model accuracy (Raimondo et al. 2010; Willming et al. 2016), and the above
described uncertainty analysis of using scaled toxicity as model input. ICE models predict with
acceptable accuracy for PFOA when invertebrates were used to predict to invertebrate species
and vertebrates were used to predict to vertebrate species in these comparisons. Models validated
across a wide range of species, chemicals, and toxicity values show an acceptable level of
prediction accuracy (>90% values predicted within five-fold of measured \ alue) when adhering
to the model guidance listed in Box 1 (Raimondo et al. 2010; Willming et al. 2'Ho). The results
summarized in Sections L.l and L.2 and described more thoroughly in Raimondo el al. (in prep)
demonstrate that the relationship of inherent sensitivity is preserved across taxa, chemicals, and
range of toxicity values when using robust l(T! models While the current analysis uses
freshwater species to predict to estuarine/maiine species. pre\ ions model validation and
uncertainty analyses did not indicate the habitat of the species to be an influential source of ICE
model uncertainty (Raimondo et al 2<) I o. Willming et al. 2016).
L.4 Prediction of Kstuarine/Marine Species Sensitivity to PFOA
A \ nine of PI OA scnsili\ily was predicted with Web-ICE v3.3 for all possible species
using all a\ ailahlc surrogate species (Table L-l). Predicted values were obtained by entering all
available surrogate species into the Web-ICE SSD generator, which predicts to all possible
species from all a\ ailahlc surrogates simultaneously and exports results into an excel
spreadsheet. Web-ICE results were generated using both mg/L and |ig/L values to evaluate the
full set of possible predictions using both units of measure against the model domain, confidence
intervals, and model parameters. First, all available models were evaluated based on the
parameter (MSE, R2, slope) guidance in Box 1, which are the same for an ICE species pair
L-12
-------
regardless of input value (Table L-3). Models that did not meet the parameter criteria in Box 1
were rejected in this first pass. In the next step, values that were predicted using |ig/L were
evaluated against the model domain and selected for the next tier of evaluation when the
surrogate value was within the range of data used to develop the model. If the surrogate value
reported as [j,g/L was beyond the model domain, the mg/L value was evaluated if it was within
the model domain and if the model slope was between 0.66-1 33 (Raimondo et al. in prep). Cases
in which both units were outside the model domain were not included quantitatively, but the
value with the narrowest confidence intervals was included for qualitative considerations. Values
(using either [^g/L or mg/L input value) were excluded quantitatively from the SM AYs but
retained for qualitative consideration if an evaluation of confidence intervals, model parameters,
and the model domain indicated the relationship between surrogate and predicted species was not
informed by robust underlying data. At this stage, specific predictions should be based on
holistic evaluation of nil a\ailaMe information provided In the model, confidence interval, and
data used to develop the model Decisions to exclude a prediction from the SMAV are clarified
in footnotes Because the scnsili\ ily of a single species can be predicted by multiple surrogates,
we c ; llculated the SMAY where multiple robust models were available for a predicted species.
Each predicted species was then assigned to the appropriate saltwater MDRs as defined in the
1985 Guidelines
a) Family in the phylum Chordata
b) Family in the phylum Chordata
c) Either the Mysidae or Penaeidae family
d) Family in a phylum other than Arthropoda or Chordata
e) Family in a phylum other than Chordata
f) Family in a phylum other than Chordata
g) Family in a phylum other than Chordata
h) Any other family
L-13
-------
The acute sensitivity of estuarine/marine species to PFOA is presented in Table L-4. A
total of 48 models representing 21 estuarine/marine species were available in Web-ICE to predict
the toxicity of PFOA to saltwater species (Table L-4). Of these, 14 models were initially rejected
based on model parameters not meeting the guidance in Box 1, reducing the number of predicted
species to 19 represented by 34 models. Further evaluation of ICE predictions resulted in 13
SMAVs representing crustaceans, molluscs, and fish and demonstrated the potential to meet the
saltwater MDRs. The range of sensitivity for the predicted taxa is consistent with the range of
sensitivity of freshwater species for this compound
L-14
-------
Table L-3. All ICE Models Available in Web-ICE v3.3 for Saltwater Predicted Species Based on Surrogates with Measured PFOA.
Model parameters are used to evaluate prediction robustness. Cross-validation success is the percentage of all model data that were predicted within 5-fold of the measured value
through leave-one-out cross-validation (Willming et al. 2016). Taxonomic distance describes the relationship between surrogate and predicted species (e.g., 1 = shared genus, 2 =
1*1edicled spede*
*>iNTii!!:ilr specie*
slope
Mlld'Ccpl
Ml
1- rcedoin
|\ 2i
K;
|> MlllIC
Menu
S(|lliirc
llTMl
(Msl.i
Sinto!>hIc
Model
Minimum
\ nine
III!! 1 l
*»lllTii!Mle
Model
M:i\inmni
\ :ilne ill!! 1 i
( ross
Milidiilion
S||(('f<«.
("..i
1 :i\ononiii
l)i-1;inie
1 -i- in ( rilerin
Acartia tonsa
Daphnia magna
0.59
1.31
2
0.91
0.0443
0.17
2.24
38514.70
50
5
Rejected
Allorchestes compressa
Daphnia magna
0.83
1.59
3
0.80
0.0390
0.12
5.00
184.54
100
5
Accepted qualitatively
Allorchestes compressa
Pimephales promelas
0.84
0.15
3
0.96
0.0028
0.02
163.05
26895.72
100
6
Accepted
Americamysis bahia
Daphnia magna
0.83
0.02
160
0.68
<0.001
0.93
0.07
840000.00
64
5
Accepted
Americamysis bahia
Lepomis macrochirus
1.01
-0.92
138
0.66
<0.001
0.95
0.13
290000.00
59
6
Accepted
Americamysis bahia
Oncorhynchus mykiss
0.92
-0.50
150
0.60
<0.0001
1.08
0.06
1100000.00
57
6
Rejected
Americamysis bahia
Pimephales promelas
0.95
-1.12
46
0.55
<0.001
1.75
2.27
70200000.00
35
6
Rejected
Chelon labrosus
Lampsilis siliquoidea
1.27
1.50
1
0.99
0.0403
0.00
19.01
281.00
na
6
Accepted
Chelon macrolepis
Pimephales promelas
1.51
-1.04
2
0.97
0.0114
0.05
26.00
2533.38
100
4
Accepted qualitatively
Crassostrea virginica
Americamysis bahia
0.44
1.76
114
0.34
<0.001
0.88
0.003
117648.20
55
6
Rejected
Crassostrea virginica
Daphnia magna
0.44
1.54
1 16
0.28
<0.001
1.08
0.08
137171.43
58
6
Rejected
Crassostrea virginica
Lampsilis siliquoidea
0.82
-0.28
0.95
0.0041
0.06
30.00
22000.00
100
4
Accepted
Crassostrea virginica
Lepomis macrochirus
0.66
0.71
1 12
0.51
<0.001
0.64
0.36
290000.00
69
6
Rejected
Crassostrea virginica
Oncorhynchus mykiss
0.59
0.97
120
0.50
<0.001
0.68
0.02
570000.00
68
6
Rejected
Crassostrea virginica
Pimephales promelas
0.75
0.44
24
0.61
<0.001
0.68
1.24
206300.75
69
6
Accepted
Cyprinodon bovinus
Lepomis macrochirus
0.66
0.70
1
0.99
0.0326
0.00
7.43
7326.20
na
4
Accepted
Cyprinodon bovinus
Oncorhynchus mykiss
0.72
0.80
2
0.91
0.0427
0.08
4.93
1637.92
100
4
Accepted qualitatively
Cyprinodon bovinus
Pimephales promelas
0.67
0.65
2
0.99
0.0043
0.00
10.49
7847.42
100
4
Accepted
Cyprinodon variegatus
Americamysis bahia
0.57
1.88
88
0.56
<0.001
0.67
0.003
182000.00
64
6
Rejected
Cyprinodon variegatus
Daphnia magna
0.53
1.79
84
0.49
<0.001
0.72
0.08
304000.00
64
6
Rejected
Cyprinodon variegatus
Lampsilis siliquoidea
0.72
0.76
1
0.99
0.0392
0.00
30.00
22000.00
na
6
Accepted
Cyprinodon variegatus
Lepomis macrochirus
0.74
0.87
82
0.65
<0.001
0.47
0.77
157000.00
82
4
Accepted
Cyprinodon variegatus
Oncorhynchus mykiss
0.75
0.90
87
0.65
<0.001
0.56
0.82
12700000.00
75
4
Accepted
Cyprinodon variegatus
Pimephales promelas
0.69
0.98
24
0.74
<0.0001
0.43
2.27
16500000.00
77
4
Accepted
Farfantepenaeus
duorarum
Americamysis bahia
1.03
0.06
6
0.81
0.0022
0.55
0.01
720.00
50
4
Accepted
L-15
-------
I'lVlliclrll
IllU'lTt'pl
Ml
1' ivciImiii
|\ 2 i
K;
|> MlllIC
Menu
llTMl
iMslii
*«im rii«iilr
Model
Minimum
\ nine
ill!! 1 i
*«im riiiiiiic
MimI.I
\ Li s i i ii ii i ii
\ lllllc III!! 1 l
( row
\;ili
-------
Table L-4. ICE-estimated Species Sensitivity to PFOA.
Values in bold and underlined are used for SMAV.
( (iiniiKiii \;imc
Prcdiclcd Species
Siimiiiiile Species
Inpul
I nil
l-'sliiii;iletl
~r«i\icil>
(mii/l.)
95" ii Confidence
lnler\;ils (niu/l.)
SMAV
Calanoid copepod
Acartia tons a
Daphnia magna
M-g/L
ill y~
0.84 - 1138.99
NA
Amphipod
Allorchestes compressa
Daphnia magna
mg/L
3(.os
604.53 -21536.48
225.75
Pimephales promelas
mg/L
225.75
| 15.43 -441.47
Mysid
Americamysis bahia
Daphnia magna
Hg/L
30.88
15^- 60.63
52.37
Lepomis macrochirus
mg/L
88.82
(.1 oy - 129.12
Oncorhynchus mykiss
hr/l
171,63bc
59.3 - 4l><>.74
Pimephales promelas
MU 1.
18.02°
3.25 - yy.94
Thicklip mullet
Chelon labrosus
Lampsilis siliquoidea
mg'J ¦
21448.80
5726.99 - 80330.23
21448.80
Bigscale mullet
Chelon macrolepis
Pimephales promelas
mg/L
X5I i9d
248.17-2920.75
NA
Eastern oyster
Crassostrea virginica
Americamysis bahia
."g/L
5 us '
2.55 - 10.1
96.96
Daphnia magna
iiii'I.
S l(.
3.27 -20.37
Lampsilis siliquoidea
mu 1.
34.9(,
13.16-92.90
Lepomis macrochirus
Hg/I.
-Sl>c
15.7 - 90.87
Oncorhynchus mykiss
M-g/L
5i.55bc
20.64 - 128.77
Pimephales promelas
mg/L
268.94
124.11 -582.78
Leon springs pupfish
Cyprinodon bovinus
Lepomis macrochirus
mg/L
385.84
100.03 - 1488.19
321.5
On corh vn ch us mykiss
mg/L
1405.12ab
117.55 - 16795.20
Pimephales promelas
mg/L
267.81
163.81 -437.84
Sheepshead minnow
Cyprinodon variegatus
Americamysis bahia
Mg/L
24.42°
12.11 -49.24
300.4
Daphnia magna
Mg/L
43.97°
18.29 - 105.67
Lampsilis siliquoidea
mg/L
236.12
42.61 - 1308.27
Lepomis macrochirus
mg/L
975.20
695.11 - 1368.14
Oncorhynchus mykiss
Mg/L
432.19
164.38 - 1136.32
Pimephales promelas
Mg/L
81.82
24.12-277.55
Pink shrimp
Farfantepenaeus duorarum
Americamysis bahia
mg/L
31.16
5.31 - 182.77
31.16
Daphnia magna
Mg/L
825.20b°
47.76 - 14258.5
Lepomis macrochirus
Mg/L
350.34b°
11.8 - 10398.14
Oncorhynchus mykiss
Mg/L
1468.52b°
47.52 -45386.61
L-17
-------
( (iiniiKiii Niiino
Prcdiclcd Species
Siiito»;i(c Species
Inpul
I nil
I'.sliiiiiiled
To\ici(\
(inii/l.)
95V'» ( onlitleiice
lnlei'\iils (inii/l.)
SMAY
Banana prawn
Fenneropenaeus merguiensis
Daphnia magna
mg/L
2395.84
^3.14 - 15382.93
2395.84
Threespine stickleback
Gasterosteus aculeatus
Lepomis macrochirus
mg/L
1867.23ab
2 ^9.79 - 14539.92
NA
Oncorhynchus mykiss
mg/L
4934.49"'
357.05 -68194.76
Polychaete
Hydroides elegans
Daphnia magna
M-g/L
17.26bc
1.79 - 166.56
NA
Oncorhynchus mykiss
M-g/L
3.90bc
1.96-7.76
Pinfish
Lagodon rhomboides
Lepomis macrochirus
muT.
342.29d
12 | 48 - 964.44
NA
Blue slirimp
Litopenaeus stylirostris
Americamysis bahia
mu 1.
28.40a
^ 14.22
NA
Inland silverside
Menidia beryllina
Lepomis macrochirus
mu 1.
1373.42
509.9') - i8.68
1373.42
Atlantic silverside
Menidia menidia
Lepomis macrochirus
mu 1.
419.03
151.91 - 1155.88
491.5
Oncorhynchus mykiss
mg/L
576.41
110.75 -2999.99
Tidewater silverside
Menidia peninsulae
Americamysis bahia
mg/L
(.1 ^a
8.51 -444.50
279.81
Lepomis macrochirus
mg/L
2"79.SI
97.91 -799.61
Oncorhynchus mykiss
IlliZ/l.
818.72'ib
11.88 - 56382.63
Mysid
Metamysidopsis insularis
Daphnia magna
mg/L
S94.87
209.96 -3814.04
853.2
Lampsilis siliquoidea
mg/L
813.42
361.18 - 1831.90
Striped mullet
Mugil cephalus
Lepomis macrochirus
mg/L
730.34ab
36.69 - 14537.49
NA
Oncorhynchus mykiss
mg/L
18685.50ab
81.57 -4280001.49
Harpacticoid copepod
Tigriopus japonicus
I.epomis macrochirus
mg/L
2812.24d
976.85 - 8096.09
1810.69
Pimepha/es prome/as
mg/L
1810.69
533.03 -6150.86
Harpacticoid copepod
Tisbe battagliai
Daphnia magna
mg/L
1923.22ab
204.62 - 18075.62
NA
NA = Not Available
a Both confidence intervals >1.5 order mamiilude
b Input data outside model range
c Guidance for model mean square error, R:. and or slope not niel
d Does not meet slope criteria for using scaled lo\ial\ (0 66-1 ^ ^ i
L-18
-------
L.5 Derivation of Acute Water Benchmark for Estuarine/Marine
Water
The Web-ICE predicted acute data set for PFOA contains 10 genera representing the
eight MDR groups that would be necessary for developing an estuarine/marine criterion.
However, the EPA supplemented this dataset with acceptable quantitative study data (discussed
in Section 3.1.1.2). In scenarios where both empirical LCso values and estimated LCso values
were available for the same species, only the empirical data were used lo derive the species mean
acute value. The ranked GMAVs for these combined data along with the MDR met by each
GMAV is summarized in Table L-5. From this dataset, an acute benchmark was calculated using
procedures consistent with the 1985 Guidelines and willi those used for the derivation of
freshwater criterion values for PFOA. GM AVs lor the four mosl sensitive genera were within a
factor of 1.5 of each other (Table L-5). The csluurinc murine F.W (llie 5th percentile of the genus
sensitivity distribution) for PFOA is 14.07 mg/L (Table I .-(•>). The I AV was lower than all of the
GMAVs for both the tested species and lor values derived using Web-ICE. The FAV is then
divided by two to obtain a concentration yielding a minimal effects acute effect value. Based on
the abo\ e. the FAX' 2. w liich is the estuarine/marine acute water column benchmark magnitude,
is 7.D mu I. PFOA (rounded to two significant figures) and is expected to be protective of 95% of
saltwater genera potentially exposed to PFOA under short-term conditions of one-hour of
duration, if the one-hour a\ crime magnitude is not exceeded more than once in three years
(Figure L-2). This draft acute benchmark for estuarine/marine aquatic life is lower than the
recommended acute freshwater criterion (49 mg/L), suggesting that estuarine/marine species
may be more acutely sensitive to PFOA and emphasizing the importance of having a separate
benchmark value for the protection of estuarine/marine aquatic life.
L-19
-------
Table L-5. Ranked Estuarine/Marine Genus Mean Acute Values.
Values in bold are derived from empirical PFOA toxicity tests with the species.
MDR
(.roup
( oiniiKin Name
Species
SMAY
(inii/l.)
CMAY
(m»/l.)
Rank
Percentile
C
Mysid
Siriella armata
15.5
15.5
1
0.07
D
Mediterranean mussel
Mytilus galloprovincialis
17.6
17.6
2
0.14
F
Purple sea urchin
Strongylocentrotus purpuratus
20.63
20.63
3
0.21
C
Mysid
Americamysis bahia
24
24
4
0.29
F
Pink shrimp
Farfantepenaeus duorarum
31.16
31.16
5
0.36
D
Eastern oyster
Crassostrea virginica
96.96
96.96
6
0.43
E
Amphipod
Allorchestes compressa
225.8
225.8
7
0.50
A
Leon springs pupfish
Cyprinodon bovinus
321 5
^ It) S
8
0.57
Sheepshead minnow
Cyprinodon variegatus
300.4
B
Inland silverside
Menidia beryllina
1,373
573.8
0.64
Atlantic silverside
Menidia menidia
491.5
Tidewater silverside
Menidia peninsulae
279.8
C
Mysid
Metamysidopsis insularis
853.2
853.2
|u
0.71
G
Harpacticoid copepod
Tigriopus japonicus
1,811
1,811
11
0.79
F
Banana prawn
Fenneropenaeus merguiensis
2.^96
2,396
12
0.86
H
Thicklip mullet
Chelon lahrosus
21.449
21,449
13
0.93
1: Estuarine/Marine MDR Groups
a) Family in the phylum Chordata
b) Family in the phylum Chordata
c) Either the Mysidae or Penaeidae family
d) Family in a phylum other than \rlliropoda nr ( hmdala
e) Family in a phylum other than Chordata
f) Family in a phylum other than Chordata
g) Family in a phylum mlier than ( hurdala
h) Any other fam i ly
L-20
-------
Table L-6. Estuarine/Marine Final Acute Value and Protective Aquatic Acute Benchmark.
Calculated Estuarine/Marine FAY based on 4 lowest values; n=13
Rank
Genus
GMAV
(mg/L)
ln(GMAV)
ln(GMAV)2
P=R/(N+1)
sqrt(P)
1
Siriella
15.5
2.74
7.51
0.071
0.267
2
Mytilas
17.6
2.87
8.22
0.143
0.378
3
Strongylocentrotus
20.63
3.03
9.16
0.214
0.463
4
Americamysis
24
3.18
10.10
0.286
0.535
E (Sum):
11.81
35.00
0.71
1.64
S2 =
L =
A =
FAV =
PVAL=
2.73 S = slope
2.275 L = X-axis intercept
2.644 A = InFAV
14.07 P = cumulative probability
7.0 mg/L PFOA (rounded to two significant figures)
>
"3
s
s
u
u
e
cs
fi
0»
3
Ch
0.6
"I
:
O Chelon
~ Fenneropenaeus
~ Tigriopus
--
~ Metamysidopsis
:
O Menidia
O Cyprinodon
~ Allorchestes
-
A Crassostrea
¦
~ Farfantepenaeus
O Fish (WeblCE)
¦
¦ Americamysis
¦ Invertebrate (Empirical)
-
¦
~ Invertebrate (WeblCE)
--
¦ Strongylocentrotus
~ Mollnsk (Empirical)
-
~ Mytilus
A Mollnsk (WeblCE)
¦ Siriella
Acute Benchmark
ill i i i i i i i i 1 i i i i i i i i 1 i
. i i i i i i I . . i i i i i i
10 100 1,000
Genus Mean Acute Value (mg/L PFOA)
10,000
100,000
Figure L-2. Ranked Estuarine/Marine Acute PFOA GMAVs Used for the Aquatic Life
Acute Benchmark Calculation.
L-21
-------
L.5.1 Estuarine Marine/Benchmark Uncertainty
Epistemic uncertainty of individual ICE estimates used for SMAV calculation was
quantified through the calculation of corresponding 95% confidence intervals for each ICE
estimate. Of the individual models and resultant ICE-estimated LCso values from the available
and quantitatively acceptable models (see bolded and underlined values in Table L-4; n =21), the
range of individual 95% CIs (i.e., 95% CI range = upper 95% CI low er 95% CI) as a percent of
the corresponding LCso estimate (i.e., = [95% CI range/LC50 eslimale|:;= I'm) ranged from
69.01%) to 626.49%). The ICE model with the lowest 95% CI range relati\ e lo the LCso estimate
(i.e., 69.01%) employed Lepomis macrochirus as the predictor species and Cyprmoilon
variegatus as the predicted species. The ICE model with the largest 95% CI range relative to the
LCso estimate (i.e., 626.49%) employed / ki/>lmia magna as the predictor species and
Fenneropenaeus merguiensis as the predicted species Nineteen of the 21 ICE-predicted values
in Table L-4 that were used for SMAV calculation had ^5"..('l ranges that were greater than the
corresponding LC5('estimate (i e. ^5".. CI range was > 100% of the LC50 estimate). The
relatively wide ranging ^5".. CIs demonstrate the underlying uncertainty in the PFOA
estuarine marine benchmark
I'oiir of the 1.1 (iM.W's used to derive the acute PFOA estuarine/marine benchmark were
based on empirical toxicity tests. Interestingly, the four GMAVs based on empirical data were
also the four most sensiti\ e (iMAVs in the GSD (Figure L-2), meaning final
estuarine/benchmark magnitude was primarily based on relatively certain empirical toxicity tests
and the inherent uncertainty in the ICE models had little influence on the final acute
estuarine/marine benchmark magnitude. It is unclear if ICE-estimated data were greater than
empirical data because of a simple coincidence or a systematic mechanistic reason. A systematic
mechanistic reason why ICE-estimated acute values were greater than empirical acute values
L-22
-------
could be attributed to the use of freshwater species to predict to estuarine/marine species in the
ICE regressions. For example, estuarine/marine LCso values from quantitatively acceptable
studies (Appendix B. 1) were typically smaller than acute LCso values for freshwater species
(Appendix A. 1). The apparent increase in PFOA toxicity in estuarine/marine environments
relative to freshwaters may represent a unique toxicological consideration of PFOA (and
possibly other PFAS) that was not a toxicological attribute of I lie oilier chemicals used to build
the supporting ICE models, which would result in artificially high PFOA I estimates for
estuarine/marine species.
The estuarine/marine benchmark still appears adequately protective based oil the
available high quality empirical data (Appendix B I) The acute PI OA estuarine/marine
benchmark (i.e., 7.0) is more than two times lower than the lowest G\1AV (i.e., 15.5 mg/L),
which was based on empirical data for Siric/la I-PA further evaluated the appropriateness of the
estuarine/marine benchmark In comparing it to empirical, hut qualitatively acceptable, data for
estuarine/marine species N\\ specifically focused on qualitatively-acceptable estuarine/marine
tests reported in Table 11 I that (I) tested an animal species, (2) exposed test organisms to a
PFOA lor a duration that was reasonably similar to standard acute exposures (e.g., 48 hours to
seven days). (3) reported acute apical effects, and (4) reported effect concentrations that were
lower than the acute estuarine marine benchmark final acute value (i.e., 14 mg/L). EPA
identified three indi\ idual tests in Table H.l as meeting the previous criteria:
1. Liu et al. (2013, 2014c) evaluated the chronic effects of PFOA (96% purity,
purchased from Sigma-Aldrich) on green mussels, Perna viridis, via a seven-day
measured, static-renewal study. A NOEC of 0.0114 mg/L and a LOEC of 0.099
mg/L was determined for a decrease in the relative condition factor (RCF). The
L-23
-------
study was acceptable for qualitative use only because of the atypical test duration,
which is too long for an acute test and too short for a chronic test. Additionally,
the PFOA test displayed a questionable concentration-response patten where there
was no difference between the RCF at the LOEC (i.e., 0.099 mg/L) and the
highest test concentration, which contained a PFOA concentration that was more
than 10X greater (i.e., 1.120 mg/L). The large magnitude between these two
concentrations in combination with the lack of effects to the RCF observed
between the LOEC and the highest treatment concentration suggests a true
concentration-response relationship was not observed for PFOA in this test.
2. Bernardini et al. (2021) reported the results of a 21 -day chronic study with the
Manila clam, Ruditapespliilippmarimi Subsamples of clams (n=20) were also
collected at test day seven. No significant effects of mortality were observed in
the single treatment group throughout the exposure, including attest day seven.
The se\ en-day \()!¦('. bused on mortality, was 0.00093 mg/L PFOA. Although
the se\ en-day NOI X' is less than the acute estuarine/marine benchmark, the
authors did not report any significant effects to mortality and this study was not
useful in understanding the relative protectiveness of the acute PFOA
estuarilie/marine benchmark.
3. Mhadhhi et al. (2012) conducted a six-day acute test on the turbot, Scophthalmus
maximus (a non-North American species). Endpoints included dead embryos,
malformation, hatch success at 48-hours and larvae survival (missing heartbeat
and a non-detached tail) at six days. The reported six-day LCso of 11.9 mg/L
PFOA was not used quantitatively because of the test duration was longer than the
L-24
-------
standard 96 hour acute exposure. Nevertheless, this six-day tests suggests early
life stages of S. maximus may be sensitive to acute PFOA exposures. EPA
concluded the acute PFOA estuarine/marine benchmark to be protective on the
six-day LCso reported by Mhadhbi et al. (2012) because (1) it was reasonably
similar to the most sensitive GMAV used to derive the acute estuarine/marine
benchmark (i.e., Siriella GMAV = 15.5 mg/L) and (2) lhe 96 hour LCso that
corresponds to the six-day LCso reported by Mhadhbi el al (2d12) was
hypothesized to be greater than or equal to the six-day LOj under the premise
that acute effect concentrations typically decrease with exposure lime (until an
incipient lethal concentration is reached).
Overall, results of quantitatively- and qualilali\ el\ - acceptable empirical toxicity studies with
estuarine/marine organisms do not provide any e\ idence thai the aquatic estuarine/marine
community will experience unacceptable chronic effects al ihe acute estuarine/marine PFOA
benchmark.
L-25
-------
L.6 ICE Regressions Supporting the Acute Estuarine/Marine
Benchmark
CD
£Z
° S"
E °
*= o
CD _l
d Oil fNJ
JZ o
Q. _l
CD ^
Cl
Americamysis bahia
(Log LC50)
Figure L-3. Americamysis bahia (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values.
0 O fT)
1 3
E d,
o
Q.
(D 1_
Americamysis bahia
(Log LC50)
Figure L-4. Americamysis bahia (X-axis) and Lepomis macrochirus (Y-axis) regression
model used for ICE predicted values.
L-26
-------
-2 0 2 4
Americamysis bahia
(Log LC50)
Figure h-5. Americamysis bahia (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for ICE predicted values.
-2 0 2 4 6
Americamysis bahia
(Log LC50)
Figure L-6. Americamysis bahia (X-axis) and Pimephales promelas (Y-axis) regression
model used for ICE predicted values.
L-27
-------
2 3 4 5 6 7
Danio rerio- embryo
[Log LC50)
Figure L-7. Danio rerio - embryo (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values.
O .
o o
|S,
13
5 - m
Q.
(D
CM
Danio rerio- embryo
(Log LC50)
Figure L-8. Danio rerio - embryo (X-axis) and Lepomis macrochirus (Y-axis) regression
model used for ICE predicted values.
L-28
-------
Danio rerio- embryo
[Log LC50)
Figure L-9. Danio rerio - embryo (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for ICE predicted values.
2 3 4 5 6 7
Danio rerio- embryo
(Log LC50)
Figure L-10. Danio rerio - embryo (X-axis) and Pimephales promelas (Y-axis) regression
model used for ICE predicted values.
L-29
-------
Daphnia magna
(Log LC50)
Figure L-ll. Daphnia magna embryo (X-axis) and ,4mericamysis bahia (Y-axis) regression
model used for ICE predicted values.
Daphnia magna
(Log LC50)
Figure L-12. Daphnia magna (X-axis) and Lampsilis siliquoidea (Y-axis) regression model
used for ICE predicted values.
L-30
-------
o o ^r
" Y\
| "
¦g O
o
Cl
-------
-2 0 2 4 6
Daphnia magna
(Log LC50)
Figure L-15. Daphnia magna (X-axis) and Oncorhynchus mykiss (Y-axis) regression model
used for ICE predicted values.
ra
CD
E
O o
-s
ill
Daphnia magna
(Log LC50)
Figure L-16. Daphnia magna embryo (X-axis) and Pimephales promelas (Y-axis) regression
model used for ICE predicted values.
L-32
-------
2 3 4 5 6
Lampsilis siliquoidea
(Log LC50)
Figure L-l 7. Lampsilis siliquoidea (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values.
2 3 4 5 6
Lampsilis siliquoidea
(Log LC50)
Figure L-18. Lampsilis siliquoidea (X-axis) and Lepomis macrochirus (Y-axis) regression
model used for ICE predicted values.
L-33
-------
23 4
5
6
Lampsilis siliquoidea
(Log LC50)
Figure L-19. Lampsilis siliquoidea (X-axis) and Ligumia recta (Y-axis) regression model
used for ICE predicted values.
¦- o
tn id
Lampsilis siliquoidea
(Log LC50)
Figure L-20. Lampsilis siliquoidea (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for ICE predicted values.
L-34
-------
2 3 4 5 6
Lampsilis siliquoidea
(Log LC50)
Figure L-21. Lampsilis siliquoidea (X-axis) and Pimephalespromelas (Y-axis) regression
model used for ICE predicted values.
-1 0 1 2 3 4 5
Lepomis macrochirus
(Log LC50)
Figure L-22. Lepomis macrochirus (X-axis) and Americamysis bahia (Y-axis) regression
model used for ICE predicted values.
L-35
-------
0 2 4 6
Lepomis macrochirus
(Log LC50)
Figure L-23. Lepomis macrochirus (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values.
CO -
03
CD
TD kL~ -
O _
=> O
ct m
= °
CO I ¦^1" —
cd
= o
£ _l
!_L '
£ ^ -
03
CN -
2 4 6
Lepomis macrochirus
(Log LC50)
Figure L-24. Lepomis macrochirus (X-axis) and Lampsilis siliquoidea (Y-axis) regression
model used for ICE predicted values.
L-36
-------
CD
« S
CD ^
CD ^ CO
O —1
to CD
Q) O
I
CD ^
J~1
O
CM
Lepomis macrochirus
(Log LC50)
Figure L-25. Lepomis macrochirus (X-axis) and Lithobates catesbeianus (Y-axis) regression
model used for ICE predicted values.
O
CO Li~
2 Q
£=
>-
Lepomis macrochirus
(Log LC50)
Figure L-26. Lepomis macrochirus (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for ICE predicted values.
L-37
-------
0 2 4 6
Lepomis macrochirus
(Log LC50)
Figure L-27. Lepomis macrochirus embryo (X-axis) and Pimephales prontelas (Y-axis)
regression model used for ICE predicted values.
-O
o _
=> O
ur ltj
= °
co | -*r
.£2 o
— o
s.d.
E
03 en
Ligumia recta
(Log LC50)
Figure L-28. Ligumia recta (X-axis) and Lampsilis siliquoidea (Y-axis) regression model
used for ICE predicted values.
L-38
-------
o
2 4 6
Lithobates catesbeianus
(Log LC50)
Figure L-29. Lithobates catesbeianus (X-axis) and Daphnia magna (Y-axis) regression
model used for ICE predicted values.
1 2 3 4 5
Lithobates catesbeianus
(Log LC50)
Figure L-3G. Lithobates catesbeianus (X-axis) and Lepomis macrochirus (Y-axis) regression
model used for ICE predicted values.
L-39
-------
2 4 6
Lithobates catesbeianus
(Log LC50)
Figure L-31. Lithobates catesbeianus (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for ICE predicted values.
2 4 6
Lithobates catesbeianus
(Log LC50)
Figure I .-32. Lithobates catesbeianus (X-axis) and Pimephales promelas (Y-axis) regression
model used for ICE predicted values.
L-40
-------
0 2 4 6
Oncorhynchus mykiss
(Log LC50)
Figure L-33. Oncorhynchus mykiss (X-axis) and A mericamysis bahia (Y-axis) regression
model used for ICE predicted values.
Oncorhynchus mykiss
(Log LC50)
Figure I.-34. Oncorhynchus mykiss (X-axis) and Daphn'ui magna (Y-axis) regression model
used for ICE predicted values.
L-41
-------
1 2 3 4 5
Oncorhynchus mykiss
(Log LC50)
Figure L-35. Oncorhynchus mykiss (X-axis) and Lampsilis siliquoidea (Y-axis) regression
model used for ICE predicted values.
O O
o «
CT3 O
E
CO ^
E ° N
o
Q.
OJ
Oncorhynchus mykiss
(Log LC50)
Figure L-36. Oncorhynchus mykiss (X-axis) and Lepomis macrockirus (Y-axis) regression
model used for ICE predicted values.
L-42
-------
0
2
4
6
Oncorhynchus mykiss
(Log LC50)
Figure L-37. Oncorhynchus mykiss (X-axis) and Lithobates catesbeianus (Y-axis) regression
model used for ICE predicted values.
0 2 4 6
Oncorhynchus mykiss
(Log LC50)
Figure L-38. Oncorhynchus mykiss (X-axis) and Pimephales promelas (Y-axis) regression
model used for ICE predicted values.
L-43
-------
2 4 6 8
Pimephales promelas
(Log LC50)
Figure L-39. Pimephales prornelas (X-axis) and Americamysis bahia (Y-axis) regression
model used for ICE predicted values.
CD
O
CD
O
Li"
c O
CD _l
a cd
jz o
Cl _l
CD " i
Q
Pimephales promelas
(Log LC50)
Figure L-40. Pimephales promelas (X-axis) and Daphnia magna (Y-axis) regression model
used for ICE predicted values.
L-44
-------
3 4 5 6 7
Pimephales prornelas
(Log LC50)
Figure L-41. Pimephales prornelas (X-axis) and Lampsilis siliquoidea (Y-axis) regression
model used for ICE predicted values.
0 2 4 6
Pimephales prornelas
(Log LC50)
Figure L-42. Pimephales prornelas (X-axis) and Lepomis macrochirus (Y-axis) regression
model used for ICE predicted values.
L-45
-------
0 2 4 6
Pimephales prornelas
(Log LC50)
Figure L-43. Pimephales prornelas (X-axis) and Lithobates catesbeianus (Y-axis) regression
model used for ICE predicted values.
0 2 4 6
Pimephales prornelas
(Log LC50)
Figure L-44. Pimephales prornelas (X-axis) and Oncorhynchus mykiss (Y-axis) regression
model used for ICE predicted values.
L-46
-------
2 3 4 5
Pimephales prornelas
(Log LC50)
Figure L-45. Pimephales prornelas (X-axis) and Xenopus laevis (Y-axis) regression model
used for ICE predicted values.
2 3 4 5
Xenopus laevis
(Log LC50)
Figure L-46. Xenopus laevis (X-axis) and Pimephales prornelas (Y-axis) regression model
used for ICE predicted values.
L-47
-------
Appendix M Meta-Analysis of Nominal Test Concentrations
Compared to Corresponding Measured Test
Concentrations
M. 1 Analysis
M. 1.1 Introduction
Of the freshwater acute toxicity tests considered quantitatively acceptable for criteria
derivation, 15.4% reported measured test concentrations while the remaining 84.6% only
reported nominal test concentrations (see Appendix A). Of the freshwater chronic toxicity tests
considered quantitatively acceptable for criteria derivation, 41.2% reported measured test
concentrations while the remaining 58.8% only reported nominal test concentrations (see
Appendix C). Therefore, EPA determined if nominal and measured PFOA concentrations were
typically in close agreement (i.e., measured within 2<~>% of nominal) to inform whether nominal
concentrations from unmeasured tests provide reasonable approximations of actual PFOA
exposures. If nominal and measured PFOA concentrations were systematically similar across
tests, then EPA retained unmeasured PI'OA toxicity tests for quantitative use in criteria
derivation.
M. 1 2 Mela-.\nalvsis Methods
All freshwater and saltwater acute and chronic tests with animals that were determined to
be quantitati\ ely or qualitali\ ely acceptable for criteria derivation (i.e., studies documented in
Appendices A. 15. ('. I), (i and 11) were reviewed to identify those that reported both nominal and
measured PFOA concentrations in at least one treatment concentration. Approximately 24% of
the 152 freshwater toxicity tests and 57% of the 14 saltwater toxicity tests reported measured
PFOA concentrations in at least one treatment. Pairs of nominal and measured concentrations
were not considered when the nominal concentration was 0.0 (i.e., controls). Additionally, all
nominal and measured pairs available from publications were used regardless of when the
M-l
-------
measured concentration was collected or if the measured concentration was an average value
because of broad inconsistencies in the timepoints concentrations were measured and reported in
publications. Authors typically either reported measured concentrations at various time points or
reported measured concentrations as an average value. The database of paired nominal and
measured concentrations as well as the timepoint each measured concentration was determined
are reported in Appendix M.2. Pairs of nominal and corresponding measured PFOA
concentrations were compared to one another through (1) linear correlation analysis and (2) an
assessment of measured concentrations as a percent of its paired nominal concentration
M1.2.1 Linear Correlation Analysis
The linear correlation analysis plotted nominal concentrations on the X-axis and
corresponding measured concentrations on the Y-a\is and assessed correlation between nominal
and measured concentrations across all freshwater studies and again with all saltwater studies.
Keeping freshwater and saltwater studies separate, the linear correlation analysis was then
systematically repeated on subsets of studies based on specific experimental conditions to
determine if specific experimental conditions may attribute to discrepancies between nominal
and measured PI OA concentrations Specific experimental conditions that were subsequently
reassessed through linear correlation analysis included: (1) acute studies; (2) chronic studies; (3)
unfed studies, (4) led studies. (5) studies that used solvent vehicles; (6) studies that did not
employ a solvent \ chicle. (7) studies with substrate; (8) studies without substrate; (9) studies that
used glass test vessels, (lUj studies that used plastic test vessels, and; (11) studies with test
vessels of an unspecified material.
M-2
-------
M.l.2.2 Assessment ofMeasured Concentrations as a Percent of Nominal
The assessment of percent differences was used to identify the proportion of paired
observations in which the measured concentration was not within 20% or 30% of the
corresponding nominal concentration. Measured concentrations within 20% of corresponding
nominal concentrations were considered in close agreement within one another based on EPA's
Office of Chemical Safety and Pollution Prevention (OCSPPYs Ixological Effects Test
Guidelines. For example, when describing test acceptability rules, I S l-IW (2016b) states,
"measured concentration of test substance at each treatment level remains i \ iiluii plus or minus
(±) 20% of the time-weighted average concentration for the duration of the test'' Similarly, U.S.
EPA (1996a) states, "In any case there must be evidence thai lest concentrations remained at
least 80 percent of the nominal concentrations throughout the test or that mean measured
concentrations are an accurate representation of exposure levels maintained throughout the test
period' when describing data quality. Finally, the Organization for Economic Cooperation and
Development (Ol-('l) 2<)|1)) defines a stable exposure concentration as, "A condition in which
the exposure conceniraiion remains u iilun W-/ 20% of nominal or mean measured values over
the enure exposure perioil" Recently, in a study of key considerations for accurate exposures in
ecotoxicolouical assessments of peril uorinated carboxylates and sulfonates, Rewerts et al. (2021)
used a threshold of 30% to agree with nominal concentration for both stock and exposure
solutions, as specified In the guidelines in the consolidated Quality Systems Manual for
Environmental Laboratories set by the US Department of Defense and the US Department of
Energy (Coats et al. 2017). The proportion of instances where measured concentrations differed
from corresponding nominal concentrations by > either 20% or 30% was used to inform whether
nominal and measured concentrations are expected to be systematically similar to one another.
M-3
-------
Similar to the linear correlation analysis experimental conditions, the proportion of paired
observations with measured concentration within 20% or 30% of nominal was also assessed
across specific experimental conditions to determine if any test attributes consistently explained
such discrepancies.
M.1.3 Results
M1.3.1 Linear Correlation Analysis
Overall, nominal and measured concentrations from Ihc freshwater I'FOA toxicity
literature indicated a high degree of correlation; however, measured concentrations were
generally slightly lower than nominal concentrations. For example, the geometric mean
measured/nominal ratios across freshwater and saltwater pairs were 0.8741 and 0.9065,
respectively. Figure M-l (Panel A) displays the strong correlation (correlation = 0.9995) of the
124 pairs of nominal and measured concentrations from freshwater studies, with the
measured/nominal ratios mostly falling in a tight range (close to 1.0) with a geometric mean of
0.8741. Similarly, Figure M-l (Panel li) displays the strong correlation (correlation = 0.9999) of
the 12 pairs of nominal and measured concentrations from saltwater studies, with the
measured nominal ratios mostly falling in a tight range (close to 1.0) with a geometric mean of
0.96<->5
M-4
-------
Figure M-l. PFOA measured vs. nominal concentrations for freshwater (Panel A, top) and
saltwater (Panel B, bottom) data.
M-5
-------
Nominal and measured concentration pairs in freshwater were further compared in terms
of experimental conditions. In general, the nominal and measured concentrations displayed a
high degree of correlation across experimental conditions. In particular, strong correlations were
observed across all experimental conditions in freshwater tests (correlations > 0.98; Table M-l).
Nominal and measured concentration pairs in saltwater were not luri her compared across various
experimental conditions because of the three saltwater tests identified with measured PFOA
treatments, two were performed by the same authors and followed the same methodologies. The
only difference between the three exposures (saltwater tests) was the test vessel material; thus
there were inadequate numbers of nominal and measured concentration pairs in saltwater to
make any supportable conclusions.
Table M-l. Correlations of paired nominal and measured 1'i'OA concentrations across
various experimental conditions in freshwater toxicity tests.
l-'rcsh water
Experimental
# of Paired
Geometric Mean of
Median Percent
Condition
Observations
Correlation
Measured/Nominal
Difference
Acute
51
o.yyy6
0.y572
8.80
Chronic
73
o.ysyy
0.8204
11.00
Unfed
51
o.yyy6
0.y572
8.80
Fed
73
o.ysyy
0.8204
11.00
Solvent
52
o.yy82
0.y561
9.01
No solvent
72
0.9991
0.8iy3
11.18
Substrate
25
0.y853
0.6560
17.65
No substrate
yy
o.yyys
o.y3yy
y.y2
Glass
23
o.yysy
0.y653
4.80
Plastic/Steel
45
0.9991
0.77y8
12.50
Unspecified material
56
o.yy84
o.yiyy
10.02
Correlations displayed in Table M-l were high between paired observations in
freshwaters across experimental conditions. Therefore, experimental conditions did not influence
the correlation between nominal and measured concentrations. For example, Figure M-2 presents
M-6
-------
log-scale plots of freshwater nominal and measured concentrations with data for acute tests (with
51 pairs; Panel A) and chronic tests (with 73 pairs; Panel B). Additionally, all acute tests
displayed in Figure M-2 (Panel A) constituted all the unfed tests while all the chronic tests
displayed in Figure M-2 (Panel B) constituted all the fed tests. Measured and nominal
concentrations under acute (and unfed) and chronic (and fed) test conditions were highly
correlated (0.9996 and 0.9899, respectively).
M-7
-------
Figure M-2. PFOA measured vs. nominal concentrations for freshwater acute (Panel A,
top) and freshwater chronic (Panel B, bottom) data.
M-8
-------
As a secondary example, the nominal and measured concentrations from different test
vessel types (e.g., plastic/steel or glass test vessels) are displayed in Figure M-3 below. Nominal
and measured PFOA concentrations pairs were highly correlated in both the plastic/steel test
vessels (45 pairs; correlation = 0.9997), glass test vessels (23 pairs; correlation = 0.9989) and
unspecified test vessels (56 pairs; correlation = 0.9984).
M-9
-------
Figure M-3. PFOA measured vs. nominal concentrations for freshwater tests conducted in
plastic/steel test vessels (Panel A, top) and freshwater tests conducted in glass test vessels
(Panel B, bottom) data.
M-10
-------
M.l.3.2 Assessment ofMeasured Concentrations as a Percent of Nominal
In freshwater, 21 (or 16.9%) of the 124 measured samples differed by corresponding
nominal concentrations by more than 20%. Of these 21 pairs, 17 measured samples were <80%
of the corresponding nominal concentration, while only 4 sample was >120% of the
corresponding nominal concentration (Figure M-4). The trend of measured concentrations being
less than nominal concentrations was also observed across the 124 paired observations in
freshwaters, where measured concentrations were less than nominal concentrations in 80.6% of
paired observations. Additionally, measured concentration as a percent of nominal
concentrations was not influenced by the magnitude of the nominal concentrations, suggesting
relatively low- or high-test concentrations did not systematically produce discrepancies between
nominal and measured concentrations (Figure M-4) I-'or com pari son, only 11 of the 124
measured samples differed from corresponding nominal concentrations by more than 30% (data
not shown).
M-ll
-------
All PFOA Relative Ratio (meas/nom) x 100
Fresh Water
130
120
100
& 80'
03
> 10
§ °9 1 n O U fl gO go Oq iP ° c
0.1 10.0 1,000.0
Nominal Concentration (mg'L)
Figure M-4. Measured concentrations as a percent of corresponding nominal
concentrations with horizontal lines to denote where the relative ratio (i.e., Y-axis) differs
by more than 20% and 30%.
The horizontal line coding: 100% (dash line); +/- 20% (solid lines); +/- 30% (dash/dot lines).
In saltwater, none of the 12 measured samples differed from nominal concentrations by
more than 20%. Similar to freshwater, 9 (or 75.0%) of the 12 measured concentrations were less
than corresponding nominal concentrations, suggesting measured concentrations in both
freshwater and saltwater may be systematically lower than nominal concentrations; however, the
relative difference between measured and corresponding nominal concentrations was relatively
minimal across all paired observations.
Measured as a percent of paired nominal concentrations from freshwater tests were
further compared across experimental conditions, but similar comparisons were not performed
for saltwater tests because of a lack of differences in experimental conditions across the limited
number of saltwater tests available for such a comparison (Table M-4). Acute tests (which were
the same as unfed tests), tests with solvent, tests without substrate, and tests conducted in glass
test vessels or test vessels of an unspecified material had relatively low proportions of measured
M-12
-------
concentrations that were not within 20% of nominal concentrations (range = 4.3% - 10.7%;
Table M-2) and correspondingly lower proportion not within 30%, suggesting these test
conditions may represent test scenarios where nominal concentrations accurately depict actual
test concentrations in unmeasured tests.
Of the 21 measured PFOA concentrations that were not within 20% of nominal
concentrations, 18 paired measured and nominal concentrations occurred across six different
toxicity tests reported in four publications (Oakes et al. 2004, Colombo el ill 2008, McCarthy et
al. 2021, Pecquet et al. 2020). These 18 pairs also all largely reoccurred across the remaining test
condition categories with a high proportion of measured concentrations that were not within 20%
of nominal concentrations (i.e., chronic/fed tests, tests without solvent, tests with substrate, and
tests conducted in plastic test vessels; Table M-2) As a result, the six tests reported in Oakes et
al. (2004), Colombo et al. (2008), McCarthy et ill (2<>21). and Pecquet et al. (2020) were further
evaluated to determine if there were unique aspects of these individual tests that produced
discrepancies between paired concentrations or if the discrepancies were explained by a
systematic error that is expected to occur across PFOA toxicity tests in similar experimental
conditions
M-13
-------
Table M-2. Percentage of Measured PFOA Concentrations Falling Outside of 20% of
Corresponding Nominal Concentrations as well as the Minimum and Maximin of
Measured as Percent of Nominal Concentrations (i.e., Measured/Nominal* 100) across
Freshwater
Total #
Percent of Measured Concentrations
Minimum of Measured
Maximum of Measured
r.\|K-rinu-nl;il
Nominal and
outside of 20% of Paired Nominal
Concentration as a
Concentration as a
Condition
Measured Pairs
Concentrations (%)¦'
Percent of Nominal (%)
Percent of Nominal (%)
All
124
16.9
0.38
144
Acute
51
5.9
82 0
137
Chronic
73
24.7
() 3S
144
I nled
51
5.9
82 ()
137
1 ed
73
24.7
i.i. 3 S
144
Sol\ ent
52
5.8
82.1)
137
\o Sol\ ent
72
25.0
0.38
144
Substrate
25
48.0
0.38
110
No Substrate
99
9.1
76.0
144
(ilass
23
4.3
82.0
132
Plastic Steel
45
31.1
0.38
144
I 'nspedlied
56
10.7
69.0
137
a Bold values represent lest conditions with a high pioporiion of measured concern rations that were not within 20% of
nominal concentrations.
McCarthy et al. (2021) conducted a 10-day test and a separate 20-day test on C. dilutes
(note, organisms in the 20-day test were not exposed until 4 dph, so actual exposure was only 16-
days in the 20-day test I-or consistency with study authors, "20-day" test was used here). For
both tests. PFOA stock solutions were dissoked in reconstituted moderately hard water without
the use of a solvent and stored in polyethylene at room temperature until use. In both tests, the
exposure \ essels were 1 L hiuh-density polyethylene beakers containing natural-field collected
sediment with (¦»<» ml. of sedi nient and 105 mL of test solution. PFOA in test solutions of both the
10-day and 20-day tests was added via pipette to the beakers with the tip just above the sediment
substrate.
• McCarthy at al (2021) 10-Day test. Solutions were renewed every 48-72 hours
and test concentrations were measured on day 1 and day 10. Measured test
concentrations ranged from 0-97% of nominal. At day one of the exposure, five of
M-14
-------
the seven measured concentrations were lower than corresponding paired nominal
concentrations by more than 20%. Of these five pairs, four pairs had measured
concentrations that were lower than nominal by more than 30% (see data in
Appendix M.2).
• McCarthy at al. (2021) 20-Day Test. Solutions were renewed every 48 hours and
PFOA treatment concentrations were measured on days I fi. 15 and 20. At day 20,
only one of the nine samples had a measured PFOA concentration that was larger
than the corresponding nominal concentration (percent difference in this instance
was 101.3%)). Of the eight remaining measurements, four measured
concentrations were less than corresponding nominal concentration by more than
20%o, including one instance u here the difference exceeded 30% (see data in
Appendix M.2).
McCarthy el al (2' >21) acknowledged the challenges associated with accounting for the
differences between measured and nominal concentrations but could not credibly offer a reason
why measured concentrations were consistently lower than nominal concentrations. It could
represent a clear systematic dosing error, added substrate could have bound to PFOA thereby
reducing the water column concentrations, and/or there could be other, unexplained, reasons why
measured PI O A concentrations in McCarthy et al. (2021) were lower than paired nominal
concentrations.
Oakes et al. (2004) exposed fathead minnows to PFOA in an outdoor microcosm
experiment that contained sediment trays and potted macrophytes. With the exception of the
lowest treatment concentration, all measured PFOA treatments were not within 20%> of the
nominal concentrations; however, Oakes et al. (2004) noted minimal variation in PFOA
M-15
-------
concentrations throughout the course of the exposure period compared to those concentrations
determined after one hour. The relatively minimal discrepancy between measured and nominal
concentrations following one hour suggests there was possibly an error in the nominal test
solutions themselves and/or the added PFOA rapidly sorbed to the sediment, organic matter, and
macrophytes in the microcosms. Additionally, PFOA may have rapidly sorbed to the test vessel
which was described as galvanized steel panels lined with food-grade I'VC. rather than glass test
vessels. For example, Lath et al. (2019) determined PFOA was more likely lo sorb to plastic test
vessels than glass, stating "Contrary to suggestions in the hicrainrc. our results indicated that
the greatest sorption losses for PFOA occurred on I'l' \ polypropylene], whereas losses on glass
tubes were much lower
Colombo et al. (2008) conducted acute and chronic freshwater aquatic toxicity studies
with algae (Pseudokirchneriella subcapitaia). / hi/>/inia magna, and rainbow trout. Treatment
waters were no1 measured in the acute tests because oV'ilie known stability of the test substance
in water" but treatment waters were measured in the chronic tests on the three species. Paired
measured and nominal concentrations from the algae test were not specifically considered in this
comparati\ e analysis of measured and nominal concentrations because a Final Plant Value was
not deri\ ed. and the intent was to focus on toxicity tests with animals.
In reference to the 21 -day Daphnia magna test (test vessel material was unspecified),
authors stated, "/// the / ktphnia chronic test, some of the measured test concentrations were not
within 20% of the nominal values but the maximum coefficient of variation of the measured
concentrations was 8.5%.'" Specifically, the measured concentration in the two lowest treatments
(excluding control) were 69.0% and 73.3% of nominal concentrations, respectively. All other
treatments were within 20% of corresponding nominal concentrations. In reference to the 85-day
M-16
-------
rainbow trout test, Colombo et al. (2008) stated, "the measured test concentrations again were
not within 20% of the nominal concentrations in some instancesSpecifically, the measured
concentration in the two lowest treatments (excluding control) were 69.6% and 71.7% of
nominal concentrations, respectively. All other treatments were within 20% of corresponding
nominal concentrations.
Interestingly, the two lowest treatment concentrations in both the rainbow trout and
Daphnia magna tests had the greatest discrepancies between measured and nominal
concentrations. Colombo et al. (2008) does not provide a reason for the discrepancies, but did
note that the algal chronic test had all measured concentrations within 20% of the nominal
concentrations, which suggests potential dosing errors or unknown phenomena did not occur
across all tests conducted by Colombo et al (2' >< >8). Unlike (hikes el al (2004) which reported
the use of PVC lined test vessels and McCarthy which reported I IDI'B test vessels, Colombo et
al. (2008) did not report the lesl \ essel material.
Finally, Pecquet el al (2<)2
-------
nominal) between concentrations across the remaining 89% of paired observations suggests
dosing errors are relatively rare and no specific test condition(s) systematically produce
discrepancies between measured and nominal concentrations.
M.1.4 Meta-Analvsis Conclusions
Linear correlation between measured and corresponding nominal concentrations suggests
a high degree of precision between paired observations across all test conditions. Nearly 83% of
freshwater measured concentrations fell within 20% of paired nominal concentrations and 91%
fell within 30% of paired nominal concentrations, which represent the test acceptability
thresholds identified by EPA's OCSPP's Ecological I-fleets Test Guidelines and DoD's Quality
Systems Manual for Environmental Laboratories, respect i\ el y For example, Rewerts et al.
(2021) states, "To agree with nominal concentration, measured concentrations for both stock
and exposure solutions were required to fall within the margin of 100 ± 30%, as specified by the
guidelines in the consolidated Quality Systems Manual for Environmental Laboratories set by
the US Department of Defense and the US Department of Energy (Coats et al.2017)."
Instances where measured concentrations were not within 20% of nominal were isolated
to a few studies In these cases, suspected dosing errors, unexplained phenomena, and/or
presence of substrate (i e , McCarthy et al. 2021; Oakes et al. 2004) may have contributed to
observed differences Dosing errors were not considered to be a systemic issue with PFOA
toxicity tests since large discrepancies from suspected dosing errors were only observed in a
small subset of obscr\ ed pairs in isolated toxicity tests. Use of substrate in tests may contribute
to discrepancies between measured and nominal concentrations; however, no unmeasured tests
with added substrate were used to derive the PFOA criteria. Therefore, PFOA toxicity tests were
not excluded from quantitative use in criteria derivation on the basis of unmeasured test
concentrations alone.
M-18
-------
Although measured concentrations were relatively similar to nominal concentrations,
measured concentrations were typically less than nominal concentrations, meaning PFOA
toxicity tests with unmeasured treatments may overestimate exposure, thereby underestimating
PFOA toxicity. As a result, it was particularly important to ensure tests with sensitive species
were not excluded solely because treatment concentrations were not measured. For example, a
relatively sensitive GMCV (i.e., Brachionus) was based on two unmeasured tests (Zhang et al.
2013a; Zhang et al. 2014b). Exclusion of these tests would have been inappropriate given the
relative similarities between measured and nominal concentrations and would ha\ c also resulted
in water column- and tissue-based chronic criterion magnitudes that were under productive.
M-19
-------
M.2 Paired Nominal and Measured PFOA Concentrations Data Table Used to Inform the Meta-
Analysis of Nominal and Measured Concentrations
Table M-3. Paired Nominal and Measured PFOA Concentrations from Quantitatively and Qualitatively Acceptable
Freshwater Toxicity Tests that Reported Measured PFOA Concentrations.
Treatment concentrations where the nominal concentration was 0.0 mg/L (i.e.. controls) were not included.
Chemical /
Purity
Test
Durati
on
Test Vessel
Material
Solvent
Used
Measurement
Method
Time Measured
Nominal
Cone.
(mg/L)
Measured
Cone.
(mg/L)
Reference
PFOA / 96%
96
hours
Glass (assumed
based on reference
to glass beakers in
the chronic PFOS
test within this
article)
None
High Performance
Liquid
Chromatography /
Mass Spectrometry
0 hours
0.005
0.0051
Hazelton et al.
2012, 2013
0 hours
0.05
0.0484
0 hours
0.5
0.49
0 hours
5
4.8
0 hours
50
51
0 hours
500
476
PFOA / 99%
48
hours
Glass (assumed
based on authors
reference to beakers
in acute test
methods)
DMSO
High Performance
Liquid
Chromatography /
Mass Spectrometry
After Renewal
50
49.62
Yang et al.
2014
Before Renewal
50
43.93
After Renewal
524.29
526.9
Before Renewal
524.29
476.41
APFO / 99%
96
hours
Stainless steel
None
High Performance
Liquid
Chromatography /
Tandem Mass
Spectrometry.
Details about when
samples are taken are not
provided
625
554
DuPont
Haskell
Laboratory
2000
1250
1090
2500
2280
5000
4560
10000
9360
APFO / 99.7%
21 days
Not reported.
None
Ion
Chromatography
Mean measured
concentrations taken from
14 samples, details of
when samples were taken
are not provided.
6.25
4.31
Colombo et al.
2008
12.5
9.16
25
20
50
44.2
100
88.6
PFOA / 99%
21 days
Glass (assumed)13
DMSO
High Performance
Liquid
After Renewal
5
4.96
Yang et al.
2014
Before Renewal
5
4.49
After Renewal
37.97
37.66
M-20
-------
Tesl
Nomiiiiil
Measured
(hemie;il /
Diinili
Tesl Vessel
Sol\enl
Meiisiiremenl
(one.
(tine.
PuriU
on
Miilcriiil
I sod
Method
Time Measured
(m»/l.)
(m»/l.)
Referenee
Chioniatomaphs
Mass Spectrometry
l.cl'oic Renewal
i~
-------
(hemie;il /
PuriU
Tesl
Diinili
on
Tesl Vessel
Miilcriiil
Sol\enl
I sod
Meiisiiremenl
Method
Time Measured
Nomiiiiil
(one.
(m»/l.)
Measured
(tine.
(in»/l.)
Referenee
( hiiHiialnmaphs
Mass S|Xviroiueli'\
1'icfniv renewal
:u
-------
Tosl
Nomiiiiil
Mo;isurod
(homio;il /
Diinili
Tosl Vessel
Sol\onl
Moiisiiromonl
(one.
Cone.
PuriU
on
Msilorisil
I sod
Method
1 imo Mo;isurod
(iiiii/l.)
(inii/l.)
Roforonco
0.01
0.01
0.032
0.046
0.1
0.091
PFOA
96%
0.32
0.28
Bartlett et al.
2021
21 days
HDPE
None
LC-MS/MS
ii a
1
0.88
3.2
2.7
lu
12
26
100
76
2.6
1.66
2.6
1.4
PFOA
97%
None
(assumed)
0.026
0.0001
McCarthy et
al. 2021
10 days
HDPE
l.( MS
1 das
26
19.9
26
25.1
272
184
272
265
0 day
26
19.7
20 day
26
20
0 day
87.5
68.6
20 day
87.5
50.1
PFOA
16 days
iidpi:
None
LC/MS
0 day
149
140
McCarthy et
97%
(assumed)
20 day
149
151
al. 2021
0 day
210
180
20 day
210
165
0 day
272
230
20 day
272
224
PFOA
96
Glass
DMSO
High Performance
See Note
0.501
0.4513363
Feng et al.
>98%
hours
Liquid
See Note
5.010
4.9315737
2015
M-23
-------
(hemie;il /
PuriU
Tesl
Diinili
on
Tesl Vessel
Msilerisil
Sol\enl
I sod
Meiisiiremenl
Method
Time Measured
Nomiiiiil
(one.
(m»/l.)
Measured
(tine.
(inii/l.)
Referenee
C luonialograpli> \l
ass Spectrometry
PFOA
Not reported
24
hours
Not provided
DMSO
USEPA Method 537
Noi icpoi'ial
0.5
0.685
Pecquet et al.
2020
5
6.16
APFO
96.5-100%
13 days
Polypropylene
None
Ion-selec1i\ c
electrode < l -i
Ohr
25
25
3M Company
2000
192 hr
25
25.5
^12 hr
25
25.9
PFOA
>96%
14 days
Not provided
None
l.( -\1S-\1S
Ohr
10
8.9
Miranda et al.
2020
24 hr before renewal
10
9.1
0 hr
1
0.9
24 hr before renewal
1
0.92
0 hr
0.1
0.09
24 hr before renewal
0.1
0.11
0 hr
0.01
0.01
24 hr before renewal
0.01
0.01
n/a = not applicable
a Assumed glass based on test outline for ihe dnninc c\posiiiv
b Specific details not provided.
M-24
-------
Table M-4. Paired Nominal and Measured PFOA Concentrations from Quantitatively and Qualitatively Acceptable Saltwater
Toxicity Tests that Reported Measured PFOA Concentrations.
Treatment concentrations where the nominal concentration was 0.0 mg/L (i.e., Controls) were not included.
( homiciil /
PuriU
losl
Dui'iilioii
IVsl Vessel
Miiloriiil
Sol\i*nl I sod
Moiisiiivinoiil
Method
1 inn' Moiisuml
Nomiiiid
(ttne.
(inu/l.)
Moiisuivd
(one.
(mii/l.)
RcIVivikv
PFOA / 96%
7 days
Polypropylene
None
Liquid
Chromatography -
tandem mass
spectrometry
Evcrv 24 hours
1)0001
0.00008
Liu et al. 2014a
0 001
0.00095
0.01
0.0091
0.1
0.093
1
0.95
PFOA / 96%
7 days
Polypropylene
None
Liquid
Chromaloaraphy -
tandem mass
specimmem
Samples were
lakcn every 2
da>s Iwelve
samples per
cuiiceiiiiaiKHi
0.0001
0.00008
Liu et al. 2014b,c;
Liu and Gin 2018
0.001
0.0012
0.01
0.0114
0.1
0.099
1
1.120
10
9.630
PFOA / Not
reported
21 days
Glass
\oiic
I.C-MS
See iinie
0.001
0.00093
Bernardini et al.
2021
M-25
-------
Appendix N Occurrence of PFOA in Abiotic Media
N. 1 PFOA Occurrence in Surface Waters
Table N-l. Measured Perfluorooctanoic acid (PFOA) Concentrations in Surface Waters
Across the United States.
Sliilc
\\;Kcrho(l\1
Arilhmolic
M on 11 PI-OA
( OIKTIIII'illioil
(liu/l -
Mi'riiiin PI-OA
( OIKTIlll'illioil
(US'/I.)
K;in»col PI-OA
( oiK'oiili'iilions
(iiii/l.)
Reference'
Lake Erie
18.33
15
13-27
Sinclair and Kamiaii
2006
35.75
33.5
^0-46
Boulanger et al. 2004
5.460
5.852
3.16
De Silva et al. 2011
1.9
1.9
1.6-2 2
Furdui et al. 2008
Lake Huron
3.222
3.475
0.656-4 "2
De Silva et al. 2011
0.592
0.433
0.1-1.1
Furdui et al. 2008
Lake Michigan
1.840
1 S4<)
0.28-3.4
Simcik and
Dorweiler 2005
4.100
3.788
3.579-5.243
De Silva et al. 2011
Lake Ontario
not pro\ ided
21
18-34
Sinclair et al. 2006
44.75
48 5
27-55
Boulanger et al. 2004
4.4o5
4 I5()
3.226-6.710
De Silva et al. 2011
3.773
2 win
1.8-6.7
Furdui et al. 2008
Lake Superior
0.255
0.236
0.095-0.395
De Silva et al. 2011
0.233
0.3
0.1-0.3
Furdui et al. 2008
() 246
0.124
0.074-0.996
Scott et al. 2010
Alabama
WalerboiK in Decalur
" 5 \ 25
7.5 in \1nhile
55 5
57.0
26.5-83
3M Company 2001
Pond in \1nhile
:i (.3
21.63
21.63
Tennessee River
(upstream of Baker's
Creek)
<25
<25
<25
Hansen et al. 2002
Tennessee River
(downsiream nf
Baker's ( reek)
335.2
355.0
<25-598
California
Upper Silver Creek
not provided
not provided
10-36
Plumlee et al. 2008
Coyote Creek
not provided
not provided
<4-13
Colorado
Animas River
<0.76
<0.76
<0.76
Colorado Department
of Public Health and
the Environment
2020
Arkansas River
1.58
0.69
0.36-3.90
Arvada Blunn
Reservoir
0.80
0.80
0.80
Barker Reservoir
<0.78
<0.78
<0.78
Bessemer Ditch
2.60
2.60
2.60
Big Thompson River
2.90
2.90
2.90
N-l
-------
Arilhmolic
Mo;in PI-OA
Mi'riiiin PI-OA
Kiin»col PI OA
( oiiiTiilnilion
( oiicoiilriilion
Coiicoiili'iilions
Sliilc
W.iU'i'hmlt1
(n«/l.
(iiii/l.)
(nii/l.)
Reference
Blue River
1.40
1.40
1.40
Boulder Feeder Canal
<0.71
<0.71
<0.71
Boyd Lake
1.50
1.50
1.50
Cache la Poudre River
6.68
6.68
<0.72-13.0
Clear Creek
3.05
3.05
3.00-3.10
Colorado River
0.77
0.93
(i 77-1.00
Coon Creek
<0.76
<0.76
<0.76
Eagle River
1.40
1.40
1 40
East Plum Creek
<0.67
<0.67
(1 (.7
Erie Lake
0.81
0.81
0.81
Fairmount Reservoir
<0.81
<0.81
<0.81
Fountain Creek
11.3
13.0
4.30-15.0
Fraser River
1.10
1.10
1.10
Gore Creek
2.00
2.00
2.00
Gunnison River
<0."X
<0.78
<0.78
Horsetooth Reservoir
<0."l
(i "1
<0.71
Jackson Creek
<0.~"
(i "(i
<0.70
Jerry Creek
-------
Slsile
\\;ilerho(l\1
Arillimclic
M o;i 11 PI-OA
( oncenlnilion
(n«/l.
Mcriiiin PI-OA
( oiicoiilriilion
(iiii/l.)
Kiin»col PI OA
Coiicoiili'iilions
(nii/l.)
Reference
Severy Creek
<0.74
<0.74
<0.74
Somerville Flowline
<0.75
<0.75
<0.75
South Boulder Creek
0.74
0.74
0.74
South Platte River
9.68
11.0
4.60-14.0
St. Vrain River
5.40
5.40
5.40
Strontia Springs
<0.80
<0.80
0.80
Taylor River
<0.70
<0.70
0 "0
Uncompahgre River
(delta)
0.73
0.73
0
Welton Reservoir
1.20
1.20
1.20
White River
<0.73
<0.73
<0.73
Yampa River
<0.74
0.74
<0.74
DE, NJ, PA
Delaware River
5.95
5.24
2.12-14.9
Pan el al. 2018
Florida
Waterbody in
Pensacola
<7.5
< ~ s
<7.5
3M Company 2001
Pond in Pensacola
<7.5
<7.5
<7.5
Waterbody in Port St.
Lucie
7.5<\ 25
"5 \ 25
" 5
^ (i
not provided
not provided
Houde et al. 2006
Georgia
Wateibodv in
Columbus
22 l>2
26.00
<25-26.5
3M Company 2001
Pond in Columbus
~ s
<7.5
<7.5
Conasauua Kin or
4"S
366.5
32.4-1,150
Konwick et al. 2008
\liamaiia ki\er
-- IK,"
3.1
3-3.1
Sueams and ponds mi
Dalton
ri 7
169.5
51.80-296.0
Oostanaula River
115.7
113
100-134
Laiseretal. 2011
Coosa River
104
104
104
Louisiana
Wale Indies
(local ions of concern i
near Barksdale \ F.B.
62.67
30.00
<10-370
Cochran 2015
Reference walorbodies
nearBarksdale A.F.B.
<10
<10
<10
Michigan
Raisin River
14.7
14.7
14.7
Kannan et al. 2005
St Clair River
4.467
4.4
4-5
Siskiwit Lake
0.576
0.576
0.558-0.594
Scott et al. 2010
Minnesota
Upper Mississippi
River
119.4
2.80
<2-3,600
Newsted et al. 2017;
Oliaei et al. 2013
Lake of the Isles
0.46
0.46
0.46
Simcik and
Dorweiler 2005
Lake Calhoun
19.44
19.44
19.44
N-3
-------
Slsile
\\;ilerho(l\1
Arillimclic
Mo;in PI-OA
( oncenlnilion
(n«/l.
Mcriiiin PI-OA
( oiicoiilriilion
(iiii/l.)
Kiin»col PI OA
Coiicoiili'iilions
(nii/l.)
Reference
Lake Harriet
3.38
3.38
3.38
Minnesota River
1.2
1.2
1.2
Lake Tettegouche
0.47
0.47
0.47
Lake Nipisiquit
0.14
0.14
0.14
Lake Loiten
0.7
0.7
0.7
Little Trout Lake
0.31
0.31
0.31
New Jersey
Echo Lake Reservoir
4.9
4.9
4
\J DEP 2019
Passaic River
13.55
13.55
1 -14 1
Raritan River
8.7
8.7
8.7
Metedeconk River
31.1
31.1
28.3-33.9
Pine Lake
13.6
13.6
13.6
Horicon Lake
1.9
1
1.9
Little Pine Lake
25.9
25
25.9
Mirror Lake
13 2
13.2
13.2
Woodbury Creek
7.2
7.2
" 2
Fenwick Creek
10 5
In 5
10.5
Cohansey River
4.6
4.0
4.3-4.9
Harborlown Road
3.738
3.738
3.738
Zhang et al. 2016
Passaic River
18.65
13.24
U871-47.25
New Mexico
Alamogordo DumcMic
Water Svslem
1
<1
<1
New Mexico
Environment
Department 2020,
2021
Animas River
0 >)7
<0.95
<0.89-
Lagoon(i
2,450
2,450
2,450
Holloman AFB
Outfall
74.6
74.6
74.6
Holloman AFB
Sewage Lagoon
941
941
941
Karr Canyon Estates
<0.93
<0.93
<0.93
La Luz MDWCA
<1.3
<1.3
<1.3
Lake Holloman
1,297
1,300
990-1,600
Mountain Orchard
MDWCA
<0.89
<0.89
<0.89
N-4
-------
Slsile
\\;ilerho(l\1
Arillimclic
Mo;in PI-OA
C cincoii 1 r;ilion
(n«/l.
Mcriiiin PI-OA
( oiicoiilriilion
(iiii/l.)
Kiin»col PI OA
Coiicoiili'iilions
(nii/l.)
Reference
Pecos River
0.628
<0.96
<0.94-0.936
Rio Chama
<0.98
<0.98
<0.96-<1.0
Rio Grande
0.791
0.474
<0.86-1.95
Rio Puerco
<1.3
<1.3
<1.3
San Juan River
<1.06
<0.96
<0.89-<1.9
Tularosa Water
System
<0.89
<0.89
<0.89
New York
Washington Park
Lake
10.1
10.4
4S3-I5.8
KimandKannan
:oo7
Rensselaer Lake
6.79
7.2
3 6
Iroquois Lake
7.365
not provided
not pro\ ided
Unnamed lake 1
outside Albany, NY
2.246
not provided
not pro\ ided
Unnamed lake 2
outside Albany, NY
4.341
nui pro\ ided
not provided
Niagara River
19.67
l<>
18-22
Sinclair and Kannan
2006
Finger Lakes
not pro\ idcd
14
11-20
Lake Onondaga
50.(>~
49.00
39-64
Lake Oneida
1<>
19
19
Erie Canal
38.0
'mi
25-59
Hudson River
not provided
35
22-173
Lake Champlam
no) provided
24
10-46
Lower NY Harbor
: u:
2.02
2.02
Zhang et al. 2016
Statcn Island
4U49
4.049
4.049
Hudson River
7.333
7.333
2.805-11.86
North Carolina
Capo I 'ear ki\cr
4V4
12.6
<0.2-287
Nakayama et al. 2007
Rhode Island
Narragaiibell IJa>
12
1.2
1.2
Benskin et al. 2012
Vllen Cove 1 nllou
3.784
3.784
3.784
Zhang et al. 2016
P.rislol Harbor
1.168
1.170
1.014-1.320
1 '.rook at Mill Co\ e
36.81
36.81
36.81
L!ucke\e Rrook
8.455
8.455
8.455
Chickasheen 1 irook
1.006
1.006
1.006
EG Town Dock
1.972
1.972
1.972
Fall River
0.64
0.64
0.64
Green Falls River
0.6470
0.6470
0.5860-0.7080
Hunt River
6.978
6.978
6.978
Mill Brook
9.237
9.237
9.237
Narrow River
0.9850
0.8985
0.6630-1.480
Pawcatuck River
16.98
16.98
14.99-18.97
Pawtuxet River
7.546
7.546
7.546
N-5
-------
Arithmetic
Mo;in PI-OA
Mcriiiin PI-OA
Kiiniicol- PI OA
( onccnlnilion
( onccnlr;ilion
Conccnlr;ilions
Slsile
\\;ilerho(l\1
(n«/l.
(iiii/l.)
(nii/l.)
Reference
Queens River
0.898
0.898
0.898
Sand Hill Brook
6.905
6.905
6.905
Secret Lake - Oak
Hill Brook
0.849
0.849
0.849
Slack's Tributary
2.363
2.363
2.363
South Ferry Road Pier
0.267
0.267
it.267
Southern Creek
10.08
10.08
10.08
Woonasquatucket
River
7.034
7.034
5 2 16-8 832
South Carolina
Charleston Harbor
9.5
not provided
noi pirn ided
Houde et al. 2006
Tennessee
Waterbody near
Cleveland
7.5
-------
and upstream lake within the great lakes system, Lake Superior, consistently had lower PFOA
concentrations than the other Great Lakes, with mean concentrations reported in the literature
ranging from 0.23 to 0.25 ng/L (Furdui et al. 2008; Scott et al. 2010; De Silva et al. 2011).
The higher PFOA concentrations in Lakes Erie and Ontario are likely due to higher levels
of urbanization around these lakes (Boulanger et al. 2004; Remucal 2019). A mass balance
constructed for Lake Ontario by Boulanger et al. (2005) indicated thai surface water and
wastewater effluent are the major sources of PFOA to the lake. In contrast. inputs from Canadian
tributaries and atmospheric deposition of PFAS were the major contributing sources of PFOA to
Lake Superior. Inputs from Canadian tributaries and atmospheric deposition were estimated to
contribute 59% and 35% of PFOA inputs into Lake Superior, respectively (Scott et al. 2010).
Within the Great Lakes, Remucal (2<)ll>) noted there were limited PFOA data to evaluate
temporal trends in surface waters. If the dataset was I i in i led to I.ake Ontario, which is one of the
most well-studied walei hodies lor PI-OA occurrence in the I S. (with data from 2002 to 2010)
there appears to lx- a mild decrease in PFOA concentrations overtime. This decrease was likely
due to the reduction in PI OA manufacturing, however, the downward PFOA trend in Lake
Ontario was statistically insignilicant. with authors noting additional data over longer time scales
were needed to fully inform conclusions (reviewed in Remucal 2019).
N-7
-------
100
10
U 0.1
0.01
Lake Lake Lake Lake Lake
Erie Huron Michigan Ontario Superior
Figure N-l. Distribution of the minimum and maximum concentrations (ng/L) of
Perfluorooctanoic acid (PFOA) measured in surface water samples collected from the
Great Lakes as reported in the publicly available literature.
This distribution is arranged alphabetically by waterbody.
N.1.2 PFOA Occurrence and Concentrations in the Midwestern U.S.
Similar PFOA concentrations are reported in the publicly available literature for
waterbodies in urban areas across the Midwest and northeastern U.S. along with lower PFOA
concentrations associated with remote areas in the same states (Newsted et al. 2017; NJ DEP
2019; Simcik and Dorweiler 2005; Sinclair et al. 2006). In Minnesota, Simcik and Dorweiler
(2005) observed PFOA concentrations ranging between 0.46 and 19 ng/L in urban areas near
Minneapolis and concentrations ranging between 0.14 to 0.7 ng/L in remote areas in northern
Minnesota.
N.1.3 PFOA Occurrence and Concentrations in the Northeastern U.S.
In New York, Sinclair et al. (2006) quantified (limit of quantification = 2 ng/L) PFOA in
all waters sampled across the state. Unlike PFOS, PFOA was detected at relatively elevated
N-8
-------
concentrations across all sites with comparatively little variability (median PFOA concentrations
across nine sites ranged from 14 to 49 ng/L) (Sinclair et al. 2006). Additionally, New Jersey
Department of Environmental Protection (NJ DEP) measured PFOA in surface water samples
collected from 14 different sites across the state. PFOA concentrations ranged from 1.9 ng/L to a
high of 33.9 ng/L, which was quantified in the Metedeconk River downstream of a wastewater
treatment plant. NJ DEP (2019) also indicated the Metedeconk Ri\ or is impacted by PFOA-
contaminated groundwater originating from an industrial source. Zhang el al (2016) reported the
surface water median PFOA concentration was 4.05 ng/L (n = 9; Zhang el al 2d I (•>) in 2014 in
the New York City Metropolitan area, including sites in New Jersey.
N.1.4 PFOA Occurrence and Concentrations in the Southeastern U.S.
Measured PFOA concentrations in surface waters among southeastern states of the U.S.
are highly variable with some of the highest ohsei'\ ed concentrations occurring in specific
waterbodies near areas uilh PFOA manufacturing In 2<)i)| the 3\l Company conducted a multi-
city study measuring PI OA concentrations across waterbodies with known manufacturing and/or
industrial uses of PI OA In the 3\1 Company's 2001 report, PFOA concentrations from sites
with know n PI-OA manufacturing uses were compared to PFOA concentrations in waterbodies
with no know n sources of any PI-AS (3M Company 2001). In this comparison study, cities with
PFOA manufacturing uses included Mobile and Decatur, Alabama, Columbus, Georgia, and
Pensacola, Florida Measured PFOA concentrations in surface waters, including lentic systems,
ranged from not detected (with a detection limit of 7.5 ng/L stated in the report; 3M Company
2001) to 83 ng/L in the cities with known PFOA use in manufacturing processes. These PFOA
concentrations were compared to those measured in control cities with no known PFOA
applications in manufacturing. These control cities were Cleveland, Tennessee and Port St.
Lucie, Florida. PFOA concentrations ranged from not detected to not quantified (limit of
N-9
-------
quantification = 25 ng/L; 3M Company 2001) in flowing surface waters. PFOA concentrations,
however, ranged from 97 ng/L to 748.5 ng/L in lentic systems (i.e., ponds, lakes, and reservoirs;
3M Company 2001) in St. Lucie, Florida. Lentic water samples were not collected at the other
city described as a "control," Cleveland, Tennessee. At the time of the report from the 3M
Company, the source of PFOA in lentic waters near Port St. Lucie, Florida was unknown;
however, the report noted the presence of visible litter, a greenish film on the water, and
contributions from a culvert creating a grayish plume as it entered the ualeituKly (3M Company
2001). Aside from the samples collected in Port St. Lucie, Florida, this report demonstrated that
measured PFOA concentrations in surface waters tend to be higher in areas with PFOA
manufacturing and/or industrial use (3M Company 2001).
In separate studies, PFOS and PFOA concentrations were measured in surface waters by
Hansen et al. (2002) near Decatur, Alabama and konuick el al (2008) in Georgia. Hansen et al.
(2002) studied a stretch of llic Tennessee river near Decatur. Alabama and Konwick et al. (2008)
focused on the Conasauua Ri\ er in Cieorgia as areas with known PFOA exposure and use.
Hansen et al (2<)iP) reported discharge from a lluorochemical manufacturing facility entered the
Tennessee Ri\ er towards the middle of the sampling area of the study, allowing for a comparison
of PR) A concentrations in relation to the fluorochemical manufacturing facility. In contrast,
Konwick el a I (2<)i)K) compared the PFOA concentrations measured in the Conasauga River
with those from reference sites (i.e., not impacted) along the Altamaha River. In both studies,
mean PFOA concentrations were higher in the study areas near manufacturing sources of organic
fluorochemicals. Specifically, Hansen et al. (2002) did not detect PFOA above the limit of
quantification (i.e., 25 ng/L) upstream of the fluorochemical manufacturing facility. Downstream
of the facility, PFOA concentrations ranged from below the limit of quantification at two sites
N-10
-------
immediately downstream of the facility to 598 ng/L with a mean concentration of 335.2 ng/L.
Similarly, Konwick et al. (2008) observed higher measured PFOA concentrations in the
Conasauga River, which ranged from 32.4 to 1,150 ng/L, compared to those in the Altamaha
River, ranging between 3.0 and 3.1 ng/L. Consistent with the report from the 3M Company
summarized above, the effluent from manufacturing facilities were determined to be the source
of increased PFOA concentrations in both the Tennessee and Conasauga rivers (Hansen et al.
2002; Konwick et al. 2008). PFOA concentrations in contaminated areas of ilie Conasauga River
and Altamaha River were relatively consistent with those measured in Alabama and Georgia
(3M Company 2001).
Similarly, Nakayama et al. (2007) and Cochran (2" I 5) measured PFAS, including PFOA,
in the Cape Fear Drainage Basin in North Carolina and waterhodies on Barksdale Air Force Base
in Bossier City, Louisiana; respectively. PI OA and PI-OS were found to be the dominant PFAS
detected in both studies Nakayama et al. (2007) reported PI-OA exceeded the level of
quantification (i.e . I nu I.) i n S2 3".. of samples. PFOA concentrations in the Cape Fear
Drainage Basin ranged between I (the lower limit of quantification) and 287 ng/L with a mean
concentration of 43 4 nu I. Cochran (2015) detected PFOA in 64% of all water samples
collected with an average concentration of 62.67 ng/L. As in other studies summarized above,
lower PF AS concentrations were found in the smallest upland tributaries and highest in the
middle reaches of the Cape Fear River. WWTP effluents were identified as the source of PFAS
to the study area (Nakayama et al. 2007).
N.1.5 PFOA Occurrence and Concentrations in the Western U.S.
PFOA concentrations in urbanized areas of western U.S. states were consistent with
concentrations measured in northeastern states (Sinclair et al. 2006; Zhang et al. 2016) but
remained lower than contaminated areas of southeastern states (3M Company 2001). Plumlee et
N-ll
-------
al. (2008) measured PFOA in surface water samples collected from Coyote Creek and a tributary
of Upper Silver Creek in San Jose, California. PFOA concentrations in Coyote Creek ranged
from below the detection limit (4 ng/L) to 13 ng/L and 10 ng/L to 36 ng/L in Upper Silver Creek.
Plumlee et al. (2008) postulated a combination of atmospheric deposition of volatile precursors
and surface runoff are likely sources of PFOA to both Coyote and Upper Silver Creeks.
Dinglasan-Panlilio et al. (2014) measured PFOA concentrations along the Puget Sound in
Washington, as well as Clayoquot and Barkley Sounds in lirilish Columbia. Canada. Broadly,
sampling locations spanned these inland marine systems and included freshwaters and
estuarine/marine waters. Overall, PFOA was detected at all sampling locations (PI OA
concentration range = 0.16 ng/L - 8.2 ng/L), but concentrations were lower than those observed
from sites with known manufacturing and or industrial PFOA uses. These concentrations were
consistent with those reported in the publicly a\ ail able literature for remote areas, such as the
northern Great 1 .akes and rural Minnesota (Simcik and I)or\\oiler 2005). Dinglasan-Panlilio et al.
(2014) speculate these specific regional sources as well as atmospheric deposition are likely
contributors of PI OA to these remote areas (Dinglasan-Panlilio et al. 2014).
N.1.6 Summary »!' PI OA Occurrence and Concentrations across the U.S.
Despite the wide use and persistence of PFOA in aquatic ecosystems, current information
on the en\ironmenlal distribution of PFOA in surface waters across the U.S. remains limited.
Present data are largely focused from freshwater systems collected along the east coast,
southeast, and upper Midwest, focusing primarily on study areas with known manufacturing or
industrial uses of PFAS. Current data indicate that PFOA concentrations measured in U.S.
surface waters vary widely across five orders of magnitude (Figure N-2). Additional data,
particularly in saltwater systems, are needed to better understand PFOA occurrence in aquatic
ecosystems.
N-12
-------
\ 10000
1000
100
10
-------
N.1.7 Comparison to Global Occurrence of PFOA in Surface Waters
Similar to PFAS occurrence in surface waters in the U.S., generally PFOS and PFOA
were the dominant PFAS detected in surface waters around the world (Ahrens 2011). On a global
scale, concentrations of PFOA measured in surface waters generally range between pg/L and
ng/L with some concentrations in the |ig/L range. PFOA concentrations measured in the U.S.
appear to be similar to those detected in studies with sampling sites in other countries. Global
surface water PFOA concentrations reported in the public literature range between not detected
and 11,300 ng/L near a PFAS spill site (as summarized below), and ZarcilalalxKl el: al. (2013)
reported a median PFOA concentration in surface water of 24 ng/L across Canada, f'.ui'ope, and
Asia.
In Canada elevated concentrations of PI-OA in surface waters were generally distributed
among urbanized areas, suggesting that uihan and industrial areas with high population densities
contributed to the elevated concentrations of PFOA in surface waters (Gewurts et al. 2013; Scott
et al. 2009). In a systematic. cross-Canada study of PFAS in surface waters, Scott et al. (2009)
observed that PFOS and PI OA were the predominant PFAS detected and that generally PFOS
concentrations were higher o\ erall. ranging between < 0.02 and 34.6 ng/L, than PFOA
concentrations, which ranged between n <44 and 9.9 ng/L. From the systems sampled in Canada,
Scotl et al (2<»)^) indicated that PFOA concentrations measured in Canadian surface waters
were lower than those measured in the U.S., Europe, and Asia. However, elevated PFOA
concentrations were observed in Etobicoke Creek, a tributary to Lake Ontario, after an accidental
spill of a fire-retardant foam containing perfluorinated surfactants at L.B. Pearson International
Airport in Toronto, Ontario in June 2000. The measured concentrations of PFOA ranged
between not quantified (with a quantification limit of 9 ng/L) to 11,300 ng/L (Moody et al.
2002).
N-14
-------
PFOA concentrations measured in surface waters across Europe are similar to those
observed in the U.S. Specifically, in a European Union (EU)-wide study of polar organic
persistent pollutants, Loos et al. (2009) detected PFOA in 97% of samples with a median
concentration of 3 ng/L in surface waters sampled across a wide range of sampling sites
(including contaminated and pristine rivers and streams of various sizes). However, relatively
high PFOA concentrations of nearly 200 ng/L were detected in the Po River, Italy. Mean PFOA
concentrations observed by Pan et al. (2018) were similar to those reported in Loos et al. (2009)
and across the U.S. with mean surface water concentrations from waterbodics across western
Europe, specifically the Thames River, Malarcn l.ake. and Rhine River, ranging between 2.31
ng/L to 8.51 ng/L, with a maximum concentration of 1 1.7 ng/ L detected in the Thames River.
Kwadijk et al. (2010) detected PFOA in all surface water samples collected from 20 locations
across the Netherlands, with concentrations ranging from (•> 5 to 43 ng/L. Huset et al. (2008)
measured similar PI OA concentrations in three rivers in the (ilatt Valley Watershed,
Switzerland, and reported a\ erages from three rivers ranging from 7.0 to 7.6 ng/L. Like in the
U.S. and Canada. ele\ ated concentrations of PFOA in surface waters across Europe are higher in
urbani/ed areas and sources ha\ e been attributed to waste water treatment plant effluent, AFFF
spills, and lluorochemical manufacturing facilities (Ahrens 2011; Huset et al. 2008; Kwadijk et
al. 2010; Loos et al 2007 and 2009; Pan et al. 2016).
PFOA concentrations observed in surface waters across eastern Asia were broadly similar
to the US, Canada, and Europe. In Japan, Saito et al. (2003) observed PFOA concentrations
ranging between 0.1 and 456 ng/L in surface waters samples collected from various locations.
Similarly, Nguyen et al. (2011) reported PFOA concentrations ranging between 5 and 31 ng/L
collected from an urbanized section of the Marina catchment in Singapore. Pan et al. (2018)
N-15
-------
reported PFOA concentrations from 112 samples across eastern Asia ranging from 0.15 ng/L to
52.8 ng/L. These 112 samples were collected from eight different water bodies, including; the
Yangtze River (median PFOA = 12.2 ng/L; n = 35), Yellow River (median PFOA = 2.45 ng/L; n
= 15), Pearl River (median PFOA = 1.82 ng/L; n = 13), Liao River (median PFOA = 9.39 ng/L;
n = 6), Huai River, (median PFOA = 6.01 ng/L; n = 9), Han River (median PFOA = 3.69 ng/L; n
= 6), Chao Lake (median PFOA = 8.17 ng/L; n = 13) and Tai Lake (median = 17.95 ng/L; n =
15).
Overall, these studies show the widespread distribution and variability of PI-OA
concentration in surface waters around the world and that surrounding land use and urbanization
with high population densities have a large influence on the overall occurrence of PFOA in
surface waters (Ahrens 2011; Gewurtsetal. 2<)|3. I.oosetal 2<)i)7: I.oosetal. 2009; Scott etal.
2009).
N.2 PFOA Occurrence and Detection in Aquatic Sediments
Although aquatic sediments are not anticipated to be a major PFOA sink (Ahrens 2011;
Ahrens et al 2fi
-------
The median concentration of PFOA across all sites was 2.45 |ig/kg, with a maximum
concentration of 950 |ig/kg (Anderson et al. 2016). Lasier et al. (2011) measured PFOA in
sediment from the Coosa River, Georgia watershed, upstream and downstream of a land-
application site of municipal/industrial wastewater, at concentrations ranging from 0.06-1.97
|ig/kg dry weight. Values reported in various locations across San Francisco Bay ranged from
below detection to 0.292 |ig/kg dry weight (San Francisco Bay. C.\. Sedlak et al. 2017).
Internationally, values ranged from below detection in areas with relali\ ely low population
density to |ig/kg wet weight in areas of higher human population density ( L( )(). (iufunes Bay,
Iceland; Butt et al. 2010); (I0); (0.96 |ig/kg wet weight, Tidal
Flats of Ariake Sea, Japan; Nakata et al. 2006); (0.29-0.36 |ig/l
-------
0.1 |ig/L 1.4 km from the PFAS disposal site (Xaio et al. 2015). Despite not having been an
active-fire training area, PFOA was still present on various U.S. Air Force Installations where
there is a known history of use of AFFF to extinguish hydrocarbon-based fires. Anderson et al.
(2016) measured groundwater samples between March and September of 2014 at the ten
locations with PFOA concentrations detected in 90% of samples. The median concentration of
PFOA across all sites was 0.41 |ig/L, with a maximum concentration of 250 |ig/L (Anderson et
al. 2016). These concentrations are consistent with groundwater samples from I lolloman Air
Force Base in New Mexico measured in 2017 with PFOA groundwater concentrations in
evaporation ponds and fire training areas ranging from 746 -254 |ig/L (NMI -1) 2<>2 I).
PFOA was detected in groundwater samples across Minnesota in 2006/2007,
approximately five years after the 3M Corporation phased out PI-OS production in Minnesota in
2002 (MPCA 2008). Analyses of samples collected from \ ulnerable, shallow aquifers in both
urban and agricn I Ui ra I areas across Minnesota, with a variety of potential contamination sources
(i.e., industrial and municipal stormwater. pesticides, land application of contaminated biosolids
and atmospheric deposition). indicated that periluorinated chemicals were present in
concentrations of potential concern in areas beyond the disposal sites and aquifers associated
with them (MPCA 2008). (iroundwater samples ranged from <0.001 to 0.0324 |ig/L with a
reporting limit of <) <>25 ug I.
N.4 PFOA Occurrence and Detection in Air and Rain
Air concentrations of PFOA in the atmosphere is widely distributed globally. In an urban
area in Albany, NY perfluorinated acids were measured in air samples in both the gas and
particulate phase in May and July 2006 (Kim and Kannan 2007). PFOA in the gas phase had a
mean concentration of 3.16 pg/m3 (range: 1.89-6.53) and in the particulate phase had with a
N-18
-------
mean concentration of 2.03 pg/m3 (range: 0.76-4.19) (Kim and Kannan 2007). Kim and Kanaan
(2007) also reported mean PFOA concentrations of 2.53 ng/L and 4.89 ng/L in rain and snow,
respectively. In an urban area in Minneapolis, MN, PFOA was measured in both the particulate
and gas phase. PFOA in the particulate phase ranged from 1.6-5.1 pg/m3 and from 1.7-16.1
pg/m3 in the gas phase across the five samples (MPCA 2008). The mean concentration value
reported from a location in Resolute Bay, Nunavut, Canada w as I 4 pu m3 (Stock et al. 2007).
These concentrations are greater than PFOA concentrations measured in the particle phase of air
samples measured in Zeppelinstasjonen, Svalbard (Butt et al. 2010). PFOA was measured in
September and December of 2006 and August and December of 2007, with mean concentrations
of 0.44 pg/m3 (Norwegian Institute for Air Research, 2007a,b).
N.5 PFOA Occurrence and Detection in Ice
Very little information is provided about PI-OA concentrations in ice. The PFOA
concentration from a Russian Arctic ice core sampled in 2<>(»7 was 131 pg/L (Saez et al. 2008;
Martin et al. 201 <>) During the spring of 2<)i)5 and 2<>06 surface snow was collected and the
following values were reported lor the Canadian Arctic and Greenland, respectively: 13.1-53.7
pg/L and 5<)i)-52<)pg I. (Young el ill 2<)()7)
N-19
-------
Appendix O Translation of The Chronic Water Column
Criterion into Other Fish Tissue Types
The PFOA freshwater aquatic life criteria (summarized in Section 3.3) includes chronic
tissue criteria for fish whole body, fish muscle, and invertebrate whole-body tissues. Additional
values for fish liver, fish blood, and fish reproductive tissues were also calculated by
transforming the freshwater chronic water column criterion (i.e . n 1)^4 mg,L) into representative
tissue concentrations using tissue-specific bioaccumulation factors (li.M 's) Fish liver, fish blood,
and fish reproductive BAFs were identified following the same approaches used to identify fish
wholebody, mush muscle, and inverterbrate whole body BAFs, which are described in delial in
section 2.11.3.1. Briefly, BAFs were determined from Held measurements and calculated using
the equation:
BAF = (Eq. 0-1)
Cwater
Where:
Cbiota = J'l ( 11 concern 1 uiion 111 orgaiiismal tissne(s)
Cwater = l'l( 11 coiicciiiiuiioii in wmer where the organism was collected
To identify the icpicscnlali\ e Ii \I s. a literature search for reporting on PFOA
bioacainuilalion was implemented by developing a series of chemical-based search terms to
identify studies llial were re\ iewed for reported BAFs and/or reported concentrations in which
BAFs could be calculated lor both freshwater and estuarine/marine species. BAFs from both
freshwater and estuarine/marine species were considered because; (1) inclusion of
estuarine/marine BAFs expanded the relatively limited PFOA BAF dataset and (2) Burkhard
(2021) did not specifically observe notable differences in PFAS BAFs between freshwater and
estuarine/marine systems, instead stating additional research is needed to formulate conclusions.
0-1
-------
Sources with relevant BAF information were further screened to determine if the reported BAF
information from each source was of low, medium, or high quality. Only BAFs of high and
medium quality were used to derive the tissue-specific BAFs and corresponding tissue-based
values described below.
BAFs based on reproductive tissues identified by Burkhard (2< >21) were further screened
to ensure only BAFs based on adult females were considered, because female reproductive
tissues are most relevant to potential maternal transfer to offspring. This subset of reproductive-
based BAFs and corresponding species and sampling locations are described in Table 0-1.
Table O-l. Characteristics of adult fish sampled for the calculation of PFOA reproductive
tissue BAFs.
All sampled fish were adults, and all reproducli\ e I issues idem i lied as unuad Weights, lengths, and BAFs are
Author
Species
Collection
Dale
11
Sex
Age
<\r.)
Weight
(g-\v\v)
Length
(cm)
ISA 1'"
(1 -/Kg)
Ahrens et al.
2015
European perch
(Perca fluviatilis)
io i: :ui:
3
l;
N.R.
N.R.
3.1
Shi et al.
2015,2018
Crucian carp
(Carassius carassius)
July 2014
30
24 r
o M
\ k
~*9A (F)
o0.5 (M)
15.0 (F)
13.7 (M)
6.59
Shi et al.
2015,2018
Crucian carp
(Carassius carassius)
July 2014
13
9 F
4 M
N.R.
352.3 (F)
320.7 (M)
24.6 (F)
24.8 (M)
4.64
Wang et al.
2016
Crucian carp
(Carassius carassius)
\pril 2iH4
8
N.R.
N.R.
(16.8-
65.1)1
(10.0-
14.7)1
85.4
N.R. = Not Reported
'Range.
Additional details on li.VF compilation and ranking can be found in Section 2.11.3.1 and
Burkhard (2021) The distributions of fish liver, fish blood, and fish reproductive BAFs
identified in the literature used to calculate tissue-specific BAFs were determined in the same
manner as invertebrate, fish muscle, and fish whole body BAFs (Section 3.2.2). Briefly,
distributions of BAFs used to derive additional tissue values were based on the lowest species-
level BAF reported within a waterbody. When more than one BAF was available for the same
species at the same site, the species-level BAF was calculated as the geometric mean of all BAFs
0-2
-------
for that species at that site. Summary statistics for the PFOA BAFs used in the derivation of the
additional tissue-based values are presented below (Table 0-2) and individual BAFs are
provided in Appendix P.
Table Q-2. Summary Statistics for PFOA Freshwater BAFs in Additional Fish Tissues1.
20"'
(Geometric
Median
(entile
Mean liA 1
HA I-"
HA I-"
Minimum
.Maximum
(1 ./kg-wet
(1 ./kg-wet
(L/kg-wet
(l./kg-wet
(1 ./kg-wet
Category
n
weigh 1)
weight)
weight)
weight)
weight)
Liver
13
15.59
10
2 34l)
0.732
1,109
Blood
5
80.71
34.1
14.90
14
636
Reproductive
Tissue
4
9.488
5.62
3.1
3.1
S5 41
1- Based on the lowest species-level BAF measured at a site (i.e.. w lien l\\ o or more BAFs were a\ ailable for the
same species at the same site, the species-level geometric mean B AI" was calculated, and the lowest species-level
BAF was used).
Equation 0-2 was used to transform the chronic freshwater column criterion (see Section
3.2.1.3) into tissue values using the 20lh centile BAFs from the distributions of BAFs described
above and summarized in Table 0-2
Tissue Value = Chronic Water Column Criterion x 20th Centile BAF (Eq. 0-2)
The resulting tissue \ allies that correspond to the 20th centile tissue-specific BAFs used in
Equation ()-2 are reported in Table ()-3 The values reported in Table 0-3 represent tissue-based
concentrations that offer a le\ el of protection that is equal to the magnitude components of the
chronic water column criterion as well as the fish whole body, fish muscle, and invertebrate
whole body tissue-based criteria; however, the tissue-based values reported in Table 0-3 are only
presented for comparative purposes and are not recommended criteria.
0-3
-------
Table 0-3. PFOA Concentrations for Additional Fish Tissue Values.1'2
Category
PI-'OA Concentration (m«/k« ww)
Liver
0.2208
Blood
1.401
Reproductive Tissue
0.2914
1 These PFOA concentrations are provided as supplemental information and are not recommended criteria
2 Tissue concentrations are expressed as wet weight (ww) concentrations.
0-4
-------
Appendix P Bioaccumulation Factors (BAFs) Used to Calculate PFOA Tissue Values
P.l Summary Table of PFOA BAFs used to calculate tissue criteria and supplemental fish tissue
values
Common Name
Scientific Name
Tissue
Log
BAF
(L/kg-
ww)
BAF
(L/kg-
ww)
Ranking
Location
Reference
goldfish
Carassius auratus
Blood
2.786
611.0
high
17 sites in six major rivers, Korea
Lam et al. 2014
mandarin
Siniperca scherzeri
Blood
2.869
739.0
high
17 sites in six major rivers, Korea
Lam et al. 2014
common carp
Cvprinus carpio
Blood
1.930
85.11
high
Xiaoqing River, China
Pan et al. 2017
Crucian carp
Carassius carassius
Blood
1.267
18.50
high
Tangxun Lake
Shi et al. 2018
Crucian carp
Carassius carassius
Blood
1.532
34.06
high
Xiaoqing River
Shi et al. 2018
Crucian carp
Carassius carassius
Blood
2.803
635.6
high
Beijing Airport, China
Wang et al. 2016
European perch
Perca fluviatilis
Blood
1.146
14.00
medium
Lake Halmsjon near Stockholm,
Sweden
Alirens et al. 2015
Crucian carp
Carassius carassius
Gonad
0.667
4.641
high
Tangxun Lake
Shi et al. 2018
Crucian carp
Carassius carassius
Gonad
0.819
6.594
high
Xiaoqing River
Shi et al. 2018
Crucian carp
Carassius carassius
Gonad
1.932
85.41
high
Beijing Airport, China
Wang et al. 2016
European perch
Perca fluviatilis
Gonad
0.491
3.100
medium
Lake Halmsjon near Stockholm,
Sweden
Alirens et al. 2015
common shiner
Notropis cornutus
Liver
1.390
24.55
high
Spring/Etobicoke Creek, Toronto,
Canada
Awad et al. 2011
European chub
Leuciscus cephalus
Liver
1.000
10.00
high
Orge River, near Paris, France
Labadie and
Chevreuil 2011
common carp
Carassius auratus
Liver
2.127
134.0
high
17 sites in six major rivers, Korea
Lam et al. 2014
mandarin
Siniperca scherzeri
Liver
2.779
601.0
high
17 sites in six major rivers, Korea
Lam et al. 2014
common carp
Cvprinus carpio
Liver
1.240
17.38
high
Xiaoqing River, China
Pan et al. 2017
Bream
Parabramis pekinensis
Liver
2.900
794.3
high
Pearl River Delta, China
Pan et al. 2014
goldfish
Carassius auratus
Liver
3.000
1000
high
Pearl River Delta, China
Pan et al. 2014
Common carp
Cvprinus carpio
Liver
3.400
2512
high
Pearl River Delta, China
Pan et al. 2014
Chub
Hypophthahnichthys
molitrix
Liver
2.600
398.1
high
Pearl River Delta, China
Pan et al. 2014
Tilapia
Tilapia aurea
Liver
2.300
199.5
high
Pearl River Delta, China
Pan et al. 2014
Snakehead
Ophicephalus argus
Liver
2.300
199.5
high
Pearl River Delta, China
Pan et al. 2014
P-l
-------
( 0111111 on \:ime
Scientific \:ime
Tissue
l.oji
IJAI
\\\\)
IJAI
(l./kji-
\\\\)
Kiinkin^
Locution
Reference
Lcnthcr cntfish
Clarias fuscus
Liver
2.200
158.5
high
Pearl River Delia. China
Pan el al. 2014
grass carp
Ctenopharyngodon idellus
Liver
3.600
3981
high
Pearl River Delia. China
Pan el al. 2014
Crucian carp
Carassius carassius
Liver
0.800
6.316
high
Tangxun Lake
Shi et al. 2018
Crucian carp
Carassius carassius
Liver
0.937
8.654
high
Xiaoqing Ri\ cr
Shi et al. 2018
Silver perch
Bidyanus bidyanus
Liver
1.146
14.00
high
Shoalhavcn icuinu. Australia
Terechovs et al.
2019
Crucian carp
Carassius carassius
Liver
3.045
1109
high
Beijing Airport. China
Wang et al. 2016
European perch
Perca fluviatilis
Liver
0.851
7.100
medium
Lake Halmsjon. near Stockholm,
Sweden
Ahrens et al. 2015
Mozambique
tilapia
Oreochromis mossambicus
Liver
-0.051
0.888
medium
iiM\ oti, Estuary Mouth
Fauconier et al.
2020
Mozambique
tilapia
Oreochromis mossambicus
Liver
-0.13(.
u.732
medium
iiM \ oli. Gledhow
Fauconier et al.
2020
Mozambique
tilapia
Oreochromis mossambicus
Liver
0.846
7.017
medium
aMalikulu, N2 Bridge
Fauconier et al.
2020
Cape stumpnose
Rhabdosargus holubi
Li\ or
0.434
2.714
medium
aMatikulu, N2 Bridge
Fauconier et al.
2020
tilapia
tilapia
Liver
1 S26
67.00
medium
Key River, Taiwan
Linetal. 2014
tilapia
tilapia
1 .iver
1 "24
53.00
medium
Key River, Taiwan
Linetal. 2014
tilapia
tilapia
l.ivcr
1 "40
55.00
medium
Key River, Taiwan
Linetal. 2014
Mud carp
Cirrhinus molitorella
l.i\cr
^ 'JIM I
7943
medium
Pearl River Delta, China
Pan et al. 2014
European perch
Perca fluviatilis
Muscle
1.568
37.00
high
Lake Halmsjon, near Stockholm,
Sweden
Ahrens et al. 2015
minnow
Hemiculter leucisculus
Muscle
2.130
135.0
high
Taihu Lake, China
Fang et al. 2014
silver carp
Hypophthalmichlhvs
molitrix
Muscle
1.153
14.21
high
Taihu Lake, China
Fang et al. 2014
white bait
Reganisalanx
brachyrostra/is
Muscle
2.245
175.6
high
Taihu Lake, China
Fang et al. 2014
Japanese crucian
carp
Carassius cuvieri
Muscle
1.988
97.24
high
Taihu Lake, China
Fang et al. 2014
Lake Saury
Coilia mystus
\luscle
2.496
313.0
high
Taihu Lake, China
Fang et al. 2014
common carp
Cyprinus carpio
Muscle
2.328
213.0
high
Taihu Lake, China
Fang et al. 2014
Mongolian culter
Culter mongolicus
Muscle
2.252
178.7
high
Taihu Lake, China
Fang et al. 2014
P-2
-------
Com 111011 N:i 1110
Scientific \:ime
Tissue
l.oji
IJAI
\\\\)
IJAI
(l./kji-
\\\\)
k;inkin<>
Locution
Reference
mudfish
lisgurnus
angu i //i caudal us
Muscle
: :<>:
I'K. 1
I"-'1
Tailiu Lake. China
1 aim el al :<>I4
Chinese bitterling
Rhodeus sinensis Gunther
Muscle
2.023
105.5
high
Tailm Lake. China
Fang et al. 2014
gobies
Ctenogobius giurinus
Muscle
1.656
45.28
high
Tailiu Lake. China
Fang et al. 2014
eel
Anguilla anguilla
Muscle
1.120
13.18
high
Netherlands
Kwadijk et al.
2010
Crucian carp
Carassius cuvieri
Muscle
0.159
1.442
high
Asan Lake. South Korea
Lee et al. 2020
Adult char
Salvelinus alpinus
Muscle
0.770
5.882
liidi
Merclla Lake. Canadian High Arctic
Lescord et al.
2015
Adult char
Salvelinus alpinus
Muscle
1.571
37.23
liigh
Resolute Lake, Canadian High Arctic
Lescord et al.
2015
Anchovy
Engraulis encrasicolus
Muscle
1.929
85 (10
high
(iii'iuidc estuary, SW France
Munoz et al. 2017
Common seabass
Dicentrarchus labrax
Muscle
1.995
<)X X1)
high
(iii'iuidc estuary, SW France
Munoz et al. 2017
Spotted seabass
Dicentrarchus punctatus
Muscle
2.280
190.6
hi »h
(iii'iuidc estuary, SW France
Munoz et al. 2017
common carp
Cyprinus carpio
Muscle
0.460
2.884
lndi
Xiaoqing River, China
Pan et al. 2017
Crucian carp
Carassius carassius
Muscle
0.192
1.557
lndi
Tangxun Lake
Shi et al. 2018
Crucian carp
Carassius carassius
Muscle
0 T44
2.207
lndi
Xiaoqing River
Shi et al. 2018
Silver perch
Bidyanus bidyanus
Muscle
u <>54
9.000
high
Shoalhaven region, Australia
Terechovs et al.
2019
goby
Gobio gobio
Muscle
: xr
655.6
medium
Roter Main, Upper Franconia,
Germany
Becker etal. 2010
Black Crappie
Pomoxis nigrornaculalus
Muscle
1.400
25.12
medium
Lake Niapenco, Ontario, Canada.
Bhavsar et al.
2016
Brown Bullhead
Ameiurus nebulosus
Muscle
I oOO
10.00
medium
Lake Niapenco, Ontario, Canada.
Bhavsar et al.
2016
Channel Catfish
Ictalurus punctatus
Muscle
l.UUO
10.00
medium
Lake Niapenco, Ontario, Canada.
Bhavsar et al.
2016
Common Carp
Cyprinus carpio
Muscle
1.100
12.59
medium
Lake Niapenco, Ontario, Canada.
Bhavsar et al.
2016
Largemouth Bass
Micropterus salmoides
Muscle
0.900
7.943
medium
Lake Niapenco, Ontario, Canada.
Bhavsar et al.
2016
Northern Pike
Esox lucius
Muscle
0.900
7.943
medium
Lake Niapenco, Ontario, Canada.
Bhavsar et al.
2016
Pumpkinseed
Lepomis gibbosus
Muscle
1.000
10.00
medium
Lake Niapenco, Ontario, Canada.
Bhavsar et al.
2016
P-3
-------
Com 111011 N:i 1110
Scientific \:ime
Tissue
l.oji
IJAI
\\\\)
IJAI
(l./kji-
\\\\)
k;inkin<>
Locution
Reference
Siiki 11 iiioiiIli r.;iss
licropterus dolomieu
Muscle
0 <>()()
medium
Lake \iapeuai. Ouiaiin. Canada
l!ha\saiel al
:<)|<>
White Crappie
Pomoxis annularis
Muscle
1.100
12.59
medium
Lake Niapcnco. Ontario, Canada.
Bhavsar et al.
2016
Yellow Perch
Percaflavescens
Muscle
0.900
7.943
medium
Lake Niapcnco. Ontario. Canada.
Bhavsar et al.
2016
Mozambique
tilapia
Oreochromis mossambicus
Muscle
-0.535
0.292
medium
uMvoti. Estuary Mouth
Fauconier et al.
2020
Mozambique
tilapia
Oreochromis mossambicus
Muscle
-0.420
0.380
medium
uMvoti, Glcdhow
Fauconier et al.
2020
Slender glassy
Ambassis natalensis
Muscle
0.257
1 809
medium
aMalikulu, Estuary Mouth
Fauconier et al.
2020
Mozambique
tilapia
Oreochromis mossambicus
Muscle
0.157
1.4-4
medium
aMalikulu, N2 Bridge
Fauconier et al.
2020
Cape stumpnose
Rhabdosargus holubi
Muscle
0.066
1.163
medium
aMatikulu, N2 Bridge
Fauconier et al.
2020
tilapia
tilapia
Muscle
1.708
51.00
medium
Key River, Taiwan
Linetal. 2014
tilapia
tilapia
Muscle
1 (>5'
45.00
medium
Key River, Taiwan
Linetal. 2014
tilapia
tilapia
Muscle
1 (.SI
48.00
medium
Key River, Taiwan
Linetal. 2014
common shiner
Notropis cornutus
WIS
I) SSI)
" 5S(i
high
Spring/Etobicoke Creek, Toronto,
Canada
Awad et al. 2011
medaka
Oryzias latipes
WIS
: 5 1
V,(M)
high
Seven locations across Japan
Iwabuchi et al.
2015
Juvenile char
Salvelinus a/pinus
WIS
1 SS"
77.06
high
Meretta Lake, Canadian High Arctic
Lescord et al.
2015
Juvenile char
Salvelinus a/pinus
WIS
"5.731
5387
high
Resolute Lake, Canadian High Arctic
Lescord et al.
2015
Juvenile char
Salvelinus alpinus
WIS
2.638
434.8
high
9-Mile Lake, Canadian High Arctic
Lescord et al.
2015
Spotted seabass
Dicentrarchus punctalus
WIS
2.535
343.0
high
Gironde estuary, SW France
Munoz et al. 2017
Crucian carp
Carassius carassius
WB
0.570
3.713
high
Xiaoqing River
Shi et al. 2018
Crucian carp
Carassius carassius
WB
0.443
2.775
high
Tangxun Lake
Shi et al. 2018
lake trout
Salvelinus namaycush
WB
1.740
55.00
high
Lake Superior
De Silva et al.
2011
P-4
-------
Com 111011 N:i 1110
Scientific \:ime
Tissue
l.oji
IJAI
\\\\)
IJAI
(l./kji-
\\\\)
k;inkin<>
Locution
Reference
hike li'oni
Salvelinus namaycush
WIS
1 4<>l
OO
I"-'1
Lake i:ne
Do Sil\ ;i el al
:<)| 1
walleye
Sander vitreus
WB
2.425
266.0
high
Lake 1 Tie
De Silva et al.
2011
lake trout
Salvelinus namaycush
WB
2.604
402.0
high
1 ;ike ()nl;ii'k>
De Silva et al.
2011
Crucian carp
Carassius carassius
WB
2.807
641.9
high
Beijing Airport. China
Wang et al. 2016
Common carp
Cyprinus carpio
WB
2.260
182.0
hi eh
Baivangdian Lake. China
Zhou et al. 2012
European perch
Perca fluviatilis
WB
0.000
1.000
medium
Lake Halmsjon. near Stockholm,
S\\ eden
Ahrens et al. 2015
Grass carp
Ctenopharyngoclon idellus
WB
3.912
8.160
medium
1 '.auiou Reservoir, Xiamen Sea, China
Dai and Zeng
2019
Chameleon goby
Tridentiger
trigonocephaly
WB
4.211
medium
(mil Park. Xiamen Sea, China
Dai and Zeng
2019
Lake Trout
Salvelinus namaycush
WB
3.300
1995
medium
Lake Superior
Furdui et al. 2007
Lake Trout
Salvelinus namaycush
WB
3.600
3981
medium
Lake Huron
Furdui et al. 2007
Lake Trout
Salvelinus namaycush
WB
2.900
794.3
medium
Lake Erie
Furdui et al. 2007
Lake Trout
Salvelinus namaycush
WB
:
398.1
medium
Lake Ontario
Furdui et al. 2007
Lake Trout
Salvelinus namaycush
WB
^ 40(1
2512
medium
Lake Michigan
Furdui et al. 2007
herring
Clupea harengus membras
WIS
: un
2188
medium
Baltic Sea
Gebbink et al.
2016
spat
Sprattus
\vi:
: 5:u
-I 1
medium
Baltic Sea
Gebbink et al.
2016
Sea Bass
Lateolahrax
\vi:
2 4<.<>
294.1
medium
Omuta River mouth and estuary,
Japan
Kobayashi et al
2018
Grey mullet
Mugil cephalus
wis
2.582
382.4
medium
Omuta River mouth and estuary,
Japan
Kobayashi et al
2018
Yellowfin goby
Acanthogobius flavimanus
WB
3.388
2441
medium
Omuta River mouth and estuary,
Japan
Kobayashi et al
2018
Chinese icefish
Neosalanx tangkahkeii
taihuensis
WIS
1.792
61.90
medium
Lake Chaohu, China
Pan et al. 2019
Common carp
Cyprinus carpio
WB
2.200
158.5
medium
Xerta, Ebro Delta, Spain
Pignotti et al.
2017
Mullet
Liza
WB
2.000
100.0
medium
Xerta, Ebro Delta, Spain
Pignotti et al.
2017
P-5
-------
l.oji
IJAI
IJAI
(l./kji-
Com 111011 N:i 1110
Scientific \:ime
Tissue
\\\\)
\\\\)
k;inkin<>
Locution
Reference
knach
Ruli/us ruli/us
WIS
: loo
125
medium
Xoila. 1 !hi\> Delia. Spam
huikHii el al
20 r
Rudd
Scardinius erythrophtalmus
WB
1.900
79.43
medium
Xcrla. Ebro Delia. Spain
Pignotti et al.
2017
European catfish
Silurus glanis
WB
2.100
125.9
medium
Xcrla. Ebro Delta. Spain
Pignotti et al.
2017
Ebro chub
Squalius laietanus
WB
2.000
100.0
medium
Xcrla. Ebro Delia. Spain
Pignotti et al.
2017
Bleak
Alburnus alburnus
WB
2.300
199.5
medium
Xcrla, Ebro Delia, Spain
Pignotti et al.
2017
grass goby
Zosterisessor
ophiocephalus
WB
2.125
1-.4
medium
\( Site. Orbetell lagoon, Italy
Renzi et al. 2013
grass goby
Zosterisessor
ophiocephalus
WB
1.987
T o"
medium
\( Sue. Orbetell lagoon, Italy
Renzi et al. 2013
grass goby
Zosterisessor
ophiocephalus
WB
2.123
i "
medium
IT Sue. Orbetell lagoon, Italy
Renzi et al. 2013
zooplankton
zooplankton
111VC I'l
1 "4"
55.91
high
Taihu Lake, China
Fang et al. 2014
zooplankton
zooplankton
hive it
l.'Uo
87.10
high
Taihu Lake, China
Xu et al. 2014
zooplankton
zooplankton
ln\ orl
: ;oo
199.5
medium
Baltic Sea
Gebbink et al.
2016
zooplankton
zooplankton
lii\ orl
l.OOS
11.70
medium
Mai Po Marshes, Hong Kong
Loi et al. 2011
amphipod
Gammarus, Hyalella
ln\ orl
3.413
2591
high
Welland River, Hamilton, Ontario,
Canada
De Solla et al.
2012
freshwater mussel
Unionidac
lll\ oi l
1.177
15.04
high
Taihu Lake, China
Fang et al. 2014
pearl mussel
Unionidae
lll\ oi l
1.678
47.64
high
Taihu Lake, China
Fang et al. 2014
Copepods
Copcpoda
lll\ oi l
0.398
2.500
high
Gironde estuary, SW France
Munoz et al. 2019
mysids
Mysidacea
Invert
0.398
2.500
high
Gironde estuary, SW France
Munoz et al. 2019
white shrimp
Palaemon longirosiris
Invert
0.362
2.300
high
Gironde estuary, SW France
Munoz et al. 2019
brown shrimp
Crangon crangon
Invert
0.398
2.500
high
Gironde estuary, SW France
Munoz et al. 2019
Oyster
Crassostrea gigas
Invert
1.301
20.00
high
Gironde estuary, SW France
Munoz et al. 2017
snails
Bithynia tentaculata
Invert
1.672
47.01
high
Hogsmill, Chertsey Bourne,
Blackwater Rivers
Wilkinson et al.
2018
amphipod
Gammarus pulex
Invert
1.048
11.16
high
Hogsmill, Chertsey Bourne,
Blackwater Rivers
Wilkinson et al.
2018
P-6
-------
Com 111011 N:i 1110
Scientific \:ime
Tissue
l.oji
KM
(l./k«-
\\\\)
KM
(l./k«-
\\\\)
k;inkin<>
Locution
Reference
I'aalic (K sicr
('rassostrea gigas
lii\ eil
. •«(.
Pacific Oyster
Crassostrea gigas
Invert
3.807
6410.00
medium
Jimci Bridge. Xiamen Sea, China
Dai and Zeng
2019
Ghost Crab
Ocypode stimpsoni
Invert
3.872
7440.00
medium
Fcnglin. Xiamen Sea. China
Dai and Zeng
2019
Ghost Crab
Ocypode stimpsoni
Invert
3.812
6490.00
medium
Jimci Bridge. Xiamen Sea, China
Dai and Zeng
2019
Orange-striped
hermit crab
Clibanarius infraspinatus
Invert
3.797
6263.26
medium
Jimci Bridge, Xiamen Sea, China
Dai and Zeng
2019
Shrimp
Caridea
Invert
-0.0(T
0 98
medium
u\1\ oti. Gledhow
Fauconier et al.
2020
Snail
Gastropoda
Invert
0.584
^ S4
medium
u\1\ mi. Gledhow
Fauconier et al.
2020
Snail
Gastropoda
Invert
1.025
|u 5S
medium
aMatikulu, N2 Bridge
Fauconier et al.
2020
Snail
Cerithidea rhizophorarum
ln\ eil
: 4:x
267.6
medium
Oniuta River mouth and estuary,
Japan
Kobayashi et al
2018
waterlouse, water
boatmen,
amphipods,
roundworm
Isopoda, Hemiptera,
amphipoda, nematoda
ln\ eil
:: -
171.0
medium
site A Stockholm Arlanda Airport
Koch et al. 2019
Fresh water
amphipods
Amphipoda
ln\ eil
: "16
520.0
medium
site R Ronneby Airport
Koch et al. 2019
Mayflies,
Caddisflies,
Dragonflies,
Water boatmen,
Waterlouse, Fresh
water amphipods
Ephcmeroptcrn.
Trichoptem. (hlnuata.
Hemiptera, Isopuda.
Amphipoda
ln\ eil
i (i'J'j
50.0
medium
site K the Kvarntorp area
Koch et al. 2019
worms
Sabellidae
lll\ Oi l
1.646
44.21
medium
Mai Po Marshes, Hong Kong
Loi et al. 2011
Copepods
Copepoda
Invert
2.554
358.0
medium
Gironde estuary, SW France
Munoz et al. 2017
Mysids
Mysidacea
Invert
2.033
108.0
medium
Gironde estuary, SW France
Munoz et al. 2017
Gammarids
Gammarus
Invert
3.097
1250
medium
Gironde estuary, SW France
Munoz et al. 2017
White shrimp
Palaemon longirostris
Invert
2.576
377.0
medium
Gironde estuary, SW France
Munoz et al. 2017
P-7
-------
l.oji
ISAI
IJAI
(l./kji-
( 0111111 on \:ime
Scientific \:ime
Tissue
\\\\)
\\\\)
Kiinkin^
Locution
Reference
Brown shrimp
Crangon crangon
Invert
2.661
458.0
medium
Gironde csluarv. SW France
Munoz cl al. 2017
bivalve
Mytilus galloprovincialis
Invert
1.074
11.86
medium
PC Site. Orbclcll lagoon. Ilalv
Rcn/.i cl al. 2013
bivalve
Mytilus galloprovincialis
Invert
2.108
128.3
medium
AC Site. Orbclell lagoon, Italy
Renzi et al. 2013
bivalve
Mytilus galloprovincialis
Invert
2.516
327.9
medium
NC Site. Orbclcll lagoon, Italy
Renzi et al. 2013
bivalve
Mytilus galloprovincialis
Invert
2.383
241.8
medium
FC Site. Orbclcll lagoon, Italy
Renzi et al. 2013
bivalve
Ruditapes decussatus
Invert
2.335
216.4
medium
PC Site. Orbclcll lagoon. Italy
Renzi et al. 2013
bivalve
Ruditapes decussatus
Invert
2.145
139.8
medium
LC Site. Orbclcll lagoon. Italy
Renzi et al. 2013
bivalve
Ruditapes decussatus
Invert
2.008
102.0
medium
M Site. Orbctcll lagoon. Italy
Renzi et al. 2013
bivalve
Ruditapes decussatus
Invert
1.928
84.78
medium
AC Site. Orbetcll lagoon, Italy
Renzi et al. 2013
bivalve
Ruditapes decussatus
Invert
1.794
62.24
medium
NC Site, Orbetell lagoon, Italy
Renzi et al. 2013
bivalve
Ruditapes decussatus
Invert
1.586
^8 56
medium
FC Site, Orbetell lagoon, Italy
Renzi et al. 2013
crab
Carcinus aestuarii
Invert
2.323
210.2
medium
PC Site. Orbetell lagoon, Italy
Renzi et al. 2013
crab
Carcinus aestuarii
Invert
2.370
234.2
medium
LC Site. Orbetell lagoon, Italy
Renzi et al. 2013
crab
Carcinus aestuarii
Invert
2.024
105.6
medium
AC Site. Orbetell lagoon, Italy
Renzi et al. 2013
crab
Carcinus aestuarii
Invert
1.760
57.55
medium
NC Site, Orbetell lagoon, Italy
Renzi et al. 2013
crab
Carcinus aestuarii
Invert
1.675
47.34
medium
FC Site. Orbetell lagoon, Italy
Renzi et al. 2013
prawn
Palaemon serratus
Invert
2.270
186.4
medium
PC Site, Orbetell lagoon, Italy
Renzi et al. 2013
prawn
Palaemon serratus
Invert
: ^ n
205.6
medium
LC Site, Orbetell lagoon, Italy
Renzi et al. 2013
prawn
Palaemon serratus
Invert
:
P-8
-------
P.2 Summary of PFOA BAFs used to calculate tissue criteria and
supplemental fish tissue values
Field measured BAFs used to calculate fish and invertebrate PFOA tissue criteria (fish
muscle, fish whole body, invertebrate whole body) and supplemental fish tissue values (blood,
reproductive tissue, liver) are shown in Appendix P. 1. Summary statistics for the BAFs from this
table used to derive tissue criteria and additional tissue values (i.e.. lowest species-level BAF
from each site) are reported in Table 3-10 and Table 0-2, respectively Rankings for individual
BAFs were determined by Lawrence (2021), who devised a ranking system based on fi\ e
characteristics: 1) number of water samples; 2) number of (issue samples; 3) spatial coordination
of water and tissue samples; 4) temporal coordination of water and tissue samples; and 5) general
experimental design. For the first four characteristics, a score of one to three was assigned, based
on number of samples or how closely the water and tissue measurements were paired. For the
experimental design characteristic, a delimit value of zero was assigned; unless the measured
tissues were composites of mixed species, in w liich case it was assigned a three (Lawrence
2021). These sub-scores were then summed and assigned a rank based on the final score. Studies
with high quality rankings had scores of four or li\ e. studies with medium quality rankings had
scores of li\ e or six, and studies with low quality rankings had scores of seven or higher
(Lawrence 2021) Parameters for the scores assigned to the five characteristics are listed in Table
2-2, and additional details can be found in Burkhard (2021). Only BAFs from studies with high
or medium quality rankings were included for the final BAF geometric mean calculations used to
derive tissue criteria (Table 3-11) and supplemental tissue values (Table 0-3).
P-9
-------
P.3 PFOA BAFs References
Ahrens, L., K. Norstrom, T. Viktor, A.P. Cousins, S. Josefsson. 2015. Stockholm Arlanda
Airport as a source of per- and polyfluoroalkyl substances to water, sediment and fish.
Chemosphere 129: 33-38.
Awad, E., X. Zhang, S.P. Bhavsar, S. Petro, P.W. Crozier, E.J. Reiner, R. Fletcher, S.A.
Tittlemier, E. Braekevelt. 2011. Long-Term Environmental Fate of Perfluorinated Compounds
after Accidental Release at Toronto Airport. Environmental Science and Technology 45: 8081 -
8089.
Becker, A.M., S. Gerstmann, H. Frank. 2010. Perfluorooctanoic Acid and Perfluorooctane
Sulfonate in Two Fish Species Collected from the Roter Main River, Bavieuth, Germany.
Bulletin of Environmental Contamination and Toxicology 84: 132-135
Bhavsar, S.P., C. Fowler, S. Day, S. Petro, N. Gandhi, S.B. Gewurtz, C. Hao. \ /hao, K.G.
Drouillard, D. Morse. 2016. High levels, partitioning and fish consumption based water
guidelines of perfluoroalkyl acids downstream of a former firefighting training facility in
Canada. Environment international 94: 415-423.
Dai, Z. and F. Zheng. 2019. Distribution and Moacaimulalion of peril uoroalkyl acids in Xiamen
coastal waters. Journal of Chemistry 36: I -S
De Silva, A. O., C. Spencer, B. F. Scott, S Backus and I) (' Muir 2011. Detection of a cyclic
perfluorinated acid, perfluoroethylcyclohexane sulfonate, in the Great Lakes of North America.
Environ Sci Technol. 45(19): 8060-8066
De Solla, S.R., A.O. De Silva, R.J. Letcher. 2012. Highly elevated levels of perfluorooctane
sulfonate and other perfluorinated acids found in biota and surface water downstream of an
international airport. Hamilton. Ontario, Canada. Environment International 39: 19-26.
Fang. S . X. Chen, S. Zhao. Y Zhang, W. Jiang, L. Yang, L. Zhu. 2014. Trophic magnification
and isomer fractionation ol" perlluoroalkyl substances in the food web of Taihu Lake, China.
Environmental Science & Technology 48: 2173-2182.
Fauconier, G., T Groffen, Y Wepener, and L. Bervoets. 2020. Perfluorinated compounds in the
aquatic food chains of two subtropical estuaries. Sci. Total Environ. 719: 135047
Furdui, V.I., N.L. Stock, D A. Ellis, C.M. Butt, D M. Whittle, P.W. Crozier, E.J. Reiner, D.C.G.
Muir, S.A. Mabury. 2007. Spatial Distribution of Perfluoroalkyl Contaminants in Lake Trout
from the Great Lakes. Environ Sci Technol. 41(5): 1554-1559.
Gebbink, W.A., A. Bignert, U. Berger. 2016. Perfluoroalkyl Acids (PFAAs) and Selected
Precursors in the Baltic Sea Environment: Do Precursors Play a Role in Food Web Accumulation
of PFAAs? Environ Sci Technol. 50(12): 6354-6362.
P-10
-------
Iwabuchi, K., N. Senzaki, S. Tsuda, H. Watanabe, I. Tamura, H. Takanobu, N. Tatarazako. 2015.
Bioconcentration of perfluorinated compounds in wild medaka is related to octanol/water
partition coefficient. Fundam. Toxicol. Sci. 2(5): 201-208.
Kobayashi, J., Y. Maeda, Y. Imuta, F. Ishihara, N. Nakashima, T. Komorita, T. Sakurai. 2018.
Bioaccumulation Patterns of Perfluoroalkyl Acids in an Estuary of the Ariake Sea, Japan.
Bulletin of Environmental Contamination and Toxicology 100: 536-540.
Koch, A., A. Karrman, L.W.Y. Yeung, M. Jonsson, L. Ahrens, and T. Wang. 2019. Point source
characterization of per- and polyfluoroalkyl substances (PFASs) and intractable organofluorine
(EOF) in freshwater and aquatic invertebrates. Environmental Science Process and Impacts 21:
1887-1898.
Kwadijk, C., P. Korytar and A. Koelmans. 2010. Distribution of perfluorinated compounds in
aquatic systems in the Netherlands. Environmental science & technology. 44( I < >) 3746-3751.
Labadie, P. and M. Chevreuil. 2011. Partitioning behaviour of perfluorinated alkyl contaminants
between water, sediment and fish in the Orge River (nearby Paris, France). Environmental
Pollution 159: 391-397.
Lam, N.-H., C.-R. Cho, J.-S. Lee, H.-Y. Soli. 1} -(' l.ee, J.-A. Lee. Tatarozako, K. Sasaki, N.
Saito, K. Iwabuchi, K. Kannan, H.-S. Cho Z<)|4 Perfluorinated alkyl substances in water,
sediment, plankton and fish from Korean rivers and lakes: A nationwide survey. Science of the
Total Environment 491-492: 154-162.
Lee, Y-.M., J.-Y l.ee. M -K Kim. 11 Yang, J.-E. Lee, Y. Son, Y. Kho, K. Choi, K.-D. Zoh.
2020. Concentration and distribution of per- and polyfluoroalkyl substances (PFAS) in the Asan
Lake area of South Korea Journal of I la/ardous Materials 381: 120909.
Lescord. (i. I. . K A Kidd. A () l)e Sil\a, M. Williamson, C. Spencer, X. W. Wang andD. C.
G. Muir 2<) I 5 Peril uoiinated and polyfluorinated compounds in lake food webs from the
Canadian I liuli Arctic. l-n\iron Sci Technol. 49: 2694-2702.
Lin, A Y -C. SC. Panchanuani, Y.-T. Tsai, T.-H. Yu. 2014. Occurrence of perfluorinated
compounds in the aquatic en\ ironment as found in science park effluent, river water, rainwater,
sediments, and biolissucs I ji\ ironmental monitoring and assessment 186: 3265-3275.
Loi, E. I., L. W. Yeung, S. Taniyasu, P. K. Lam, K. Kannan andN. Yamashita. 2011. Trophic
Magnification of Poly- and Perfluorinated Compounds in a Subtropical Food Web. Environ. Sci.
Technol.(45): 5506-5513.
Munoz, G., H. Budzinski, M. Babut, H. Drouineau, M. Lauzent, K.L. Menach, J. Lobry, J.
Selleslagh, C. Simonnet-Laprade, P. Labadie. 2017. Evidence for the trophic transfer of
perfluoroalkylated substances in a temperate macrotidal estuary. Environmental Science &
Technology 51: 8450-8459.
P-ll
-------
Munoz, G., H. Budzinski, M. Babut, J. Lobry, J. Selleslagh, N. Tapie, P. Labadie. 2019.
Temporal variations of perfluoroalkyl substances partitioning between surface water, suspended
sediment, and biota in a macrotidal estuary. Chemosphere 233: 319-326.
Pan, C.-G., J.-L. Zhao, Y.-S. Liu, Q.-Q. Zhang. 2014. Bioaccumulation and risk assessment of
per- and polyfluoroalkyl substances in wild freshwater fish from rivers in the Pearl River Delta
region, South China. Ecotoxicology and Environmental Safety 107: 192-199.
Pan, Y., H. Zhang, Q. Cui, N. Sheng, L.W.Y. Yeung, Y. Guo, Y. Sun, J. Dai. 2017. First Report
on the Occurrence and Bioaccumulation of Hexafluoropropylene Oxide Trimer Acid: An
Emerging Concern. Environmental Science and Technology 51 9553-0560
Pan, X., J. Ye, H. Zhang, J. Tang, andD. Pan. 2019. Occurrence, remowil and bioaccumulation
of perfluoroalkyl substances in Lake Chaohu, China. Int. J. Environ. Res Public TTealthl6(10):
1692.
Pignotti, E., G. Casas, M. Llorca, A. Tellbuscher, D. Almeida. E. Dinello, M. Fane. I). Barcelo.
2017. Seasonal variations in the occurrence of perfluoroalkyl substances in water, sediment and
fish samples from Ebro Delta (Catalonia, Spain). Science of the Total Environment 607-608:
933-943.
Renzi, M., C. Guerranti, A. Giovani, G. IVrra. S I- I'ocardi 2d 13 IV-i llnorinated compounds:
Levels, trophic web enrichments and human dietary intakes in transitional water ecosystems.
Marine Pollution Bulletin 76: 146-157.
Shi Y., R. Vestergren, T.H. Nost, Z. Zhou, Y. Cai. 2018. Probing the differential tissue
distribution and bioaccumulation behavior of per-and polyfluoroalkyl substances of varying
chain-lengths, isomeric structures and functional groups in crucian carp. Environmental Science
& Technology 52: 4592-4600.
Shi, Y . R Vestergren, Z Zhou, \ Song. L. Xu, Y. Liang, Y. Cai. 2015. Tissue distribution and
whole body burden of the chlorinated polyfluoroalkyl ether sulfonic acid F-53B in crucian carp
(Carassius carassius): Evidence for a highly bioaccumulative contaminant of emerging concern.
Environmental Science and Technology 49:14156-14165.
Terechovs, A. k I-.. A .1 Ansari, J.A. McDonald, S.J. Khan, F.I. Hai, N.A. Knott, J. Zhou, L.D.
Nghiem. 2019. Occurrence and bioconcentration of micropollutants in Silver Perch (Bidyanus
bidyanus) in a reclaimed water reservoir. Science of the Total Environment 650 (Part 1): 585-
593.
Wang, Y., R. Vestergren, Y. Shi, D. Cao, L. Xu, X. Zhao, F. Wu. 2016. Identification, Tissue
Distribution, and Bioaccumulation Potential of Cyclic Perfluorinated Sulfonic Acids Isomers in
an Airport Impacted Ecosystem. Environmental Science and Technology 50: 10923-10932.
P-12
-------
Wilkinson, J.L., P.S. Hooda, J. Swinden, J. Barker, S. Barton. 2018. Spatial (bio) accumulation
of pharmaceuticals, illicit drugs, plasticisers, perfluorinated compounds and metabolites in river
sediment, aquatic plants and benthic organisms. Environmental pollution 234: 864-875.
Xu, J., C. S. Guo, Y. Zhang and W. Meng. 2014. Bioaccumulation and trophic transfer of
perfluorinated compounds in a eutrophic freshwater food web. Environ Pollut. 184: 254-261.
Zhou, Z., Y. Shi, L. Xu, Y. Cai. 2012. Perfluorinated Compounds in Surface Water and
Organisms from Baiyangdian Lake in North China: Source Profiles, Bioaccumulation and
Potential Risk. Bulletin of Environmental Contamination and Toxicology 89: 519-524.
P-13
-------
Appendix Q Example Data Evaluation Records (DERs)
Background: This set of published literature was identified using the ECOTOXicology
database (ECOTOX; https://cfpub.epa. gov/ecotox/) as meeting data quality standards. ECOTOX
is a source of high-quality toxicity data for aquatic life, terrestrial plants, and wildlife. The
database was created and is maintained by the EPA, Office of Research and Development,
Center for Computational Toxicology and Exposure. The ECOTOX search generally begins with
a comprehensive chemical-specific literature search of the open literature conducted according to
ECOTOX Standard Operating Procedures (SOPs). The search terms are often comprised of
chemical terms, synonyms, degradates and verified Chemical Abstracts Service (CAS) numbers.
After developing the literature search strategy. IX'OTOX curators conduct a series of searches,
identify potentially applicable studies based 011 title and abstract, acquire potentially applicable
studies, and then apply the applicability criteria lor inclusion in ECOTOX. Applicability criteria
for inclusion into IX'OIOX generally include:
1 The toxic effects are related to single chemical exposure (unless the study is being
considered as part of a mixture effects assessment);
2 There is a biological effect 011 live, whole organisms or in vitro preparation including
gene chips or oniics data on adverse outcome pathways potentially of interest;
.1 Chemical test concentrations are reported;
4. There is an explicit duration of exposure;
5. Toxicology information that is relevant to OW is reported for the chemical of
concern.
6. The paper is published in the English language;
7. The paper is available as a full article (not an abstract);
8. The paper is publicly available;
9. The paper is the primary source of the data;
10. A calculated endpoint is reported or can be calculated using reported or available
information;
11. Treatment(s) are compared to an acceptable control;
12. The location of the study (e.g., laboratory vs. field) is reported; and
13. The tested species is reported (with recognized nomenclature).
Q-i
-------
Following inclusion in the ECOTOX database, toxicity studies are subsequently
evaluated by the Office of Water. All studies were evaluated for data quality generally as
described by U.S. EPA (1985) in the 1985 Guidelines and in EPA's Office of Chemical Safety
and Pollution Prevention (OCSPP)'s Ecological Effects Test Guidelines (U.S. EPA 2016c), and
EPA OW's internal data quality SOP, which is consistent with OCSPP's data quality review
approach (U.S. EPA 2018). These toxicity data were further screened lo ensure that the observed
effects could be primarily attributed to PFOS exposure. Office of Water completed a DER for
each species by chemical combination from the PFOS studies identified by ECOTOX. Example
DERs are presented here to convey the meticulous level of evaluation, review, and
documentation each PFOS study identified by ECOTOX was subject to. Appendix Q.l shows an
example fish DER and Appendix Q.2 shows an example aquatic invertebrate DER.
Q-2
-------
Q. 1 Example Fish DER
Part A: Overview
I. Test Information
Chemical name:
CAS name: CAS Number:
Purity: Storage conditions:
Solubility in Water (units):
Controlled Experiment Field Study/Observation (I'hu e X by One)
(imanipulated) {not manipulated)
Primary Reviewer: Date: I' PA Contractor {Place X by One)
Secondary Reviewer: Date: I'PA Contractor {PlaceXby One)
{At least one reviewer should be from EPA for sensitive taxa)
Citation: Indicate: author (s), year, study title, journal, volume, and pages.
(e.g., Slonim, A.R. 1973. Acute toxicity of beryllium sulfate to the common guppy. J. Wal. Pollul. Conlr. Fed. 45( 10): 2110-2122)
Companion Papers: Identify any companion papers associated i villi this paper using I he citation format above.
{Ifyes, list file names of
Were other DERs completed lor Companion I'sipcrs? Yes No DERs below)
Study Classification for Aquatic I .il'e ('rilcria Development: Place X by One Based on Highest Use
Acceptable for Ouanlilali\c I sc
Acceptable for Onalilali\e I sc
Not Acceptable lor I se'Unused
General Notes: Provide any necessary details regarding the study's use classification for all pertinent endpoints,
including non-apical endpoints within the study (e.g., note all study classifications for each endpoint if the use varies)
Major Deficiencies (note any stated exclusions): Check all that apply. Checking any of these items make the study "Not
Acceptable for Use"
A I, , . , , . No Controls (for controlled experiments
Mixture (tor controlled experiments only) only)
Excessive Control Mortality (> 10% for acute and > 20% for chronic)
. , . . . . Bioaccumulation: steady state not
Dilution water not adequately characterized , ,
reached
Dermal or Injection Exposure Pathway
Review paper or previously published without modification
Q-3
-------
Other: (if any, list here)
POTENTIAL CHEMICAL MIXTURES : Describe any potential chemicals mixtures as characterized by study
authors (including any confirmation of chemical mixtures).
DESCRIPTION OF DILUTION WATER: Describe concerns with characterization of and/or major deficiencies
with dilution water.
General Notes:
Minor Deficiencies: List and describe any minor deficiencies or other concerns with lest, 't hese items may make the study
"Acceptablefor Qualitative Use" (exceptions may apply as noted)
For Field Studies/Observations: A field study/observaiion mav he considered "Acceptable for Quantitative Use" if it
consisted of a range of exposure concentrations and the observed effects are justifiably contributed lo a single chemical
exposure
Mixture (observed effects not justifiably contribulcd lo single chemical exposure)
Uncharacterized Reference Sites/Conditions
POTENTIAL CHEMICAL MIXTURES PRESENT AT SITE: Describe any potential chemicals mixtures present at
the site as characterized by study authors (including any confirmation of chemicals present at study site).
EXPOSURE VARIABILITY ACROSS STUDY SI I E(S): Describe any exposure variability across study site(s)
as characterized by study authors (i.e.. description of study design with reference and contaminated sites).
General Notes:
Reviewer's Comments: Provide additional comments that do not appear under other sections of the DER.
Q-4
-------
ABSTRACT: Copy and paste abstract from publication.
SUMMARY: Fill out and modify as needed.
Acute:
Species (lilesliiiie)
Method'
Tesl
Dunilion
( hemiciil
/ Piiriu
pll
1 em p.
<°C)
Ihirdncss
(111li/l. ilS
CsiCO.0
or
S;ilini(\
(DDII
DOC
(lliu/l -)
r.ll'ecl
Reported
r.llecl
( oiicenlriilion
(m»/l.)
Verified
HITccl
( (iiicenlriilion
(mi»/l.)
Cliissificiilion
Quantitative /
Qualitative / Unused
a S=static, R=renewal, F=flow-through, U=unmeasured, M=measured, T=total, D=dissolved, Dict=diel;ii\. M l maternal transfer
Chronic:
Species < liTesliii^e)
Method-'
lesl
l)ui';ilioii
( hemiciil
/ Piiriu
pll
Temp.
<°C)
lliii'diiess
(111li/l. ilS
CaCO.o
or
S;ilini(\
(ppl)
DOC
(111^/1.)
Chronic
l.imils
Reported
C hronic
Value
(in^/l. or
Verified
( lironic
Value
(illli/l. or
C h ron ic
Value
l.ndpoinl
( liissil'iciilion
Quantitative /
Qualitative /
Unused
a S=static, R=renewal, F=flow-throudi. 1 unmeasured. M measured. I luial. I) dissolved, Diet=dietary, MT=maternaltransfer
Q-5
-------
II. Results Provide results as reported in the publication (including supplemental materials). Include screen shots of tables and/or
figures reporting results from the article following tabulated data table in each associated results section for all studies. Complete
tabulated data tables for all studies for studies marked "Acceptable for Quantitative Use" and "Acceptable for Qualitative Use".
Water Quality Parameters: If only general summary data of water quality parameters is provided by study authors (i.e., no
specific details of water quality parameters on a treatment level is provided), summarize any information regarding water quality
parameters under General Notes below and indicate data not provided in Table A.II.l.
General Notes: For aquatic life criteria development, measured water quality parameters in the treatments nearest the toxicity
test endpoint(s), e.g., LC50, EC20, etc., are most relevant.
•
Table A.II.1. Measured Water Quality Parameters in Test Solutions.
Dissolved oxygen, temperature, pH and [other parameters (hardness, snhiiiiv. !)()( )| 111 lesi solutions during the /Ay-day
exposure of [test organism] to [concentration of treatments)] of [test substance / under I static renewal/flow-through]
conditions.
Parameler
Treatment
Mean
Range
Dissolved
Oxygen
(% saturation
or mg/L)
[1]
[2]
j
j
Temperature
(O
[1]
[2]
j
j
pi"
HI
12/
J
J
Oilier (e.g..
hardness,
salinity, DOC)
[1]
12]
J
j
Q-6
-------
Chemical Concentrations: Summarize the concentration verification data from test solutions/media. Expand table to include
measured concentration data for each media type (i.e., water, diet, muscle, liver, blood, etc.).
General Notes: Provide any necessary detail regarding the measured concentrations, including any identified cause for
substantial differences between nominal and measured concentrations, if samples were collected on separate days (and if so provide
details), and any potential cross contamination.
Table A.II.2. Measured (and Nominal) Chemical Concentrations in Test Solutions/Media.
[Analytical Method] verification of test and control concentrations during an |X|-d;i\ exposure of [test organism] to [test
substance] under [static renewal/flow-through] conditions.
|Mi*;in|
Nil ill hoi' of
| Si ;i nthi r«l
Nomiiiiil
Mo;isiiiv(l
Siimplcs
l)c\ iiilion or
( (IIKTIlll'illioil
( niHTiili'iilion
Number of
Non-
Ik'low Son-
Siiindiinl
1 iviilmonl
( ii nils)
( ii iii Is)
S;i in pies
DolocC
Dclcd
Krroil
KiiiiUO
Control
[1]
[2]
[3]
[4]
[5]
[6]
./'
aNon-Detect: 0 = measured and detected; 1= measured and noi detected; if noi measured or reported enter as such
Q-7
-------
Mortality: Briefly summarize mortality results (if any).
General Notes: Comment on concentrations response relationship and slope of response ifprovided. Compare mortality in
treatments with control group and/or the reference chemical.
Table A.II.3. Mean Percent [Mortality or Survival].
Mean percent mortality [or number of immobilized, survival] of [test organism] exposed to [test substance] for [test duration]
under [static/renewal/flow-through] conditions.
| Menu "i.
|Si;iii«I;ir«l l)c\ iiiiion
1 iviilmcnl
Mor(;ili(\ |
or Siiindiii'd l-'rr«ir|
Control
[11
[21
[31
[41
[51
[61
[LCxl
NOEC
LOEC
a Use superscript to identify the values reported U> he simnlicaiilly different from control.
Q-8
-------
Growth: Briefly summarize growth results (if any).
General Notes: Comment on concentrations response relationship and slope of response ifprovided. Compare growth endpoints
in treatments with control group and/or the reference chemical.
Table A.II.4. Mean [Growth].
Mean growth [length and/or weight] of [test organism] exposed to [test substance] for [test duration] under
[static/renewal/flow-through] conditions.
Menu (.row111
Moiin IVivcnl
|l. en»( h/\\oiiih(|
|Si;inthird IK'\i;i(ion
( hiinjic in ll.on^lh/
|Si;iii«l;ird l)c\ iiiiion
1 iviilmonl
( ii nils)
or Si;iii«hir«l l'!rror|
liioiiiiissj
or Si;iii(hird I'lrrorj
Control
[1]
[2]
[3]
[4]
[5]
[6]
./'
[ECxl
NOEC
LOEC
aUse superscript to identify the values reported In he simnlicaiilU differeni linm cnniinl
Q-9
-------
Reproductive: Briefly summarize reproduction endpoint results (if any). For multi-senerational studies, copy and paste Table
A.II.5 below for each generation with reproductive effects data.
General Notes: Comment on concentrations response relationship and slope of response ifprovided. Compare reproductive
endpoints in treatments with control group and/or the reference chemical.
Table A.II.5. Mean [Reproductive] Effect.
Mean [reproductive] effects for [generation] of [test organism] exposed to [test substance] for [test duration] under
[static/renewal/flow-through] conditions.
|Sliiildill'd
|Siiindiird
| M Oil n
| Si iin (hi r«l
|Sliiiul;ir(l
l)o\ iiilion
l)o\ iiilion
IlillCll
l)o\ iiilion
|\loiin
l)o\ iiilion or
| Moiin
OI*
|Moiin
or
Poroonl
or
Troiilmonl
Number of
Sliindiinl
Nil in hoi' of
Sliindiinl
Porconl
Sliindiird
Su r\ i\ill
Sliindiinl
(iinils)
S|):i\\ ns |
r.rrorl
r.rror|
ll;ik-h|
r.rrorl
Posll
r.rrorl
Control
[11
[21
[31
[41
[51
[61
i
[ECx]
NOEC
LOEC
a Use superscript to identify the values reported to he simnlicaiilK dilTerenl from control.
Q-10
-------
Sublethal Toxicity Endpoints: Include other sublethal effect(s), including behavioral abnormalities or other signs of toxicity,
if any. Copy Table A.II.6 as needed to provide details for each sublethal effect observed.
General Notes: Briefly summarize observed sublethal effects otherwise not captured in the results table(s) below.
Table A.II.6. Mean [Sublethal] Effect.
Mean /"Sublethal effect (e.g., behavioral abnormalities, etc.)] in [test organism] during [test duration (acute/chronic)]
exposure to [test substance] under [static/renewal/flow-through] conditions.
|Mc;iii Suhlolhiil
Rc'sponsi'l
|Si;iii(l;iI'd l)o\i;ili(iil or
llViKllKIII
( ii nils)
Siiimliinl l'.mir|
Control
[1]
[2]
[3]
[4]
[5]
[6]
./'
lECxl
NOEC
LOEC
a Use superscript to identify the \ allies ivpoi icd In be siginl ic;iiill> different from control
Reported Statistics: Copy and paste statistical section from publication
Q-ll
-------
Part B: Detailed Review
I. Materials and Methods
Protocol/Guidance Followed: Indicate if provided by authors.
Deviations from Protocol: If authors report any deviations from the protocol noted above indicate here.
Study Design and Methods: Copy and paste methods section from publication.
TEST ORGANISM: Provide information under Details and any relevant or related information or clarifications in Remarks.
Parameler
Delails
Remarks
Species:
Common Name:
Scientific Name:
North American species?
Surrogate I'm- \orlh American
Taxon?
(Place X if applicable)
Strain/Source:
• Wild caught from unpolluted areas [1]
o Quarantine for at least 14 days or until they are
disease free, before acclimation [1]
• Must originate from same source and population [1]
• Should not be used:
o If appeared stressed, such as discoloration or
unusual behavior [1]
o If more than 5% die during the 48 hours before
test initiation [1]
o If they were used in previous test treatments or
controls [2]
• No treatments of diseases may be administered:
o Within 16 hour of field collection [1]
o Within 10 days or testing or during testing 111
Age at Study Initiation:
Acute:
• Juvenile stages preferred [1]
Chronic:
• Life-cycle lesi:
o Embryos or newly hatched voting < 48 hours old
[2]
• Partial life-cycle test:
o Immature juveniles at least 2 months prior to
active gonad development [21
• Early life-stage test:
o Shortly after fertilization [2]
Was body weight or length recorded al
test initiation?
Yes No
Was body weight or length recorded at
regular intervals?
Yes No
If yes, describe regular intervals:
Q-12
-------
STUDY PARAMETERS: Provide information under Details and any relevant information of deficiencies in Remarks.
Complete for both Controlled Experiments and Field Studies/Observations.
Pummel or
Delails
Remarks
Number of Replicates per Treatment
Group:
• At least 2 replicates/treatment recommended for
acute tests [1]
• At least 2 replicates/treatment recommended for
chronic tests [3]
Control(s):
Treatment(s):
Number of Organisms per Replicate/
Treatment Group:
• At least 10 organisms/treatment recommended [3]
• At least 7 organisms/treatment acceptable [4]
Control(s):
Treatment(s):
Exposure Pathway:
(i.e., water, sediment, gavage, or diet).
Note: all other pathways (e.g., dermal, single dose via
gavage, and injection) are unacceptable.
Exposure Duration:
Acute
• Should be 96 hours [2]
Chronic
• Life-cycle tests:
o Ensure that all life stages and life processes are
exposed [2]
o Begin with embryos (or newly hatched young),
continue through maturation and reproduction, and
should end not less than 24 days (90 days for
salmonids) after the hatching of the next
generation [2]
• Partial life-cycle tests:
o Allowed with species that require >1 year to reach
sexual maturity, so that all major life stages can be
exposed to the test material in <15 months [2]
o Begin with immature juveniles al least 2 months
prior to active gonad development, continue
through maturation and reproduction, and end not
less than 24 days (90 days for salmonids) alter the
hatching of the next generation 121
• Early life-cycle tests:
o 28 to 32 day (60 day post hatch for salmonids)
exposures from shortly after fertilization through
embryonic, larval, and earlv juvenile development
m
Acute
I'ariial Life ( \ele
LarK Life Siaue
l ull Life ( \ele
Other (please remark):
Test Concentrations (remember units):
Recommended test concentrations include ai leasi three
concentrations other than the control; four or more will
provide a better statistical analysis [3]
Nummal:
Measured:
Media measured in:
Observation Intervals:
• Should be an appropriate number of observations
over the study to ensure water quality is being
properly maintained [4]
Q-13
-------
CONTROLLED EXPERIMENT STUDY PARAMETERS: Provide information under Details and any relevant
information of deficiencies in Remarks. Complete for Controlled Experiments only.
-*
Paramcler
Details
Remarks
Acclimation/Holding:
• Should be placed in a tank along with the water in
which they were transported
o Water should be changed gradually to 100%
dilution water (usually 2 or more days) [1]
o For wild-caught animals, test water temperature
should be within 5°C of collection water
temperature [1]
o Temperature change rate should not exceed 3°C
within 72 hours [1]
• To avoid unnecessary stress and promote good
health:
o Organisms should not be crowded [1]
o Water temperature variation should be limited [1]
o Dissolved oxygen:
¦ Maintain between 60 - 100% saturation [1]
¦ Continuous gentle aeration if needed [1]
o Unionized ammonia concentration in holding and
acclimation waters should be < 35 jag/L [1]
Duration:
Feeding:
Water type:
Temperature (°C):
Dissolved Oxygen (mg/L):
Health (any mortality observed?):
Identify number of individuals excluded from testing and/or
analysis (if any):
Acclimation followed published guidance?
Describe, if any
Yes \n
If yes, indicate which guidance:
Test Vessel:
• Test chambers should be loosely covered [1]
• Test chamber material:
o Should minimize sorption of test chemical from
water [1]
o Should not contain substances that can be leached
or dissolved in solution and are free of substances
that could react with exposure chemical [1]
o Glass, No. 316 stainless steel, nylon screen and
perfluorocarbon (e.g. Teflon) are acceptable 111
o Rubber, copper, brass, galvanized mclal. epoxy
glues, lead and flexible tubing should not come
into contact with test solution, dil. water, or slock
[i]
• Size/volume should maintain acceptable biomass
loading rates (see Biomass Loading Rate below) 111
Material:
Size:
l ill Volume
Briefly describe the test vessel:
Test Solution Delivery Systein/Melhod:
• Flow-through preferred for some highly volatile,
hydrolysable or degradable materials [2|
o Concentrations should be measured often enough
using acceptable analytical methods [2]
• Chronic exposures:
o Flow-through, measured tests required |2|
Tom ( niiceiilialmiis Measured
Yci No
Tom Solution Delivery System:
Static
Renewal
Indicate Interval:
Flow-through
Indicate Type of Diluter:
Source of Dilution Water:
• Freshwater hardness range should be < 5 mg/L or <
10% of the average (whichever is greater) [1]
• Saltwater salinity range should be < 2 g/kg or < 20%
of the average (whichever is greater) [1]
• Dilution water must be characterized (natural surface
water, well water, etc.) [3]
o Distilled/deionized water without the addition of
appropriate salts should not be used [2]
• Dilution water in which total organic carbon or
particulate matter >5 mg/L should not be used [2]
o Unless data show that organic carbon or particulate
matter do not affect toxicity [2]
Dilution Series (e.g., 0.5x, 0.6x, etc.):
Q-14
-------
Paramcler
Details
Remarks
Dissolved Oxygen (mg/L):
Dilution Water Parameters:
Measured at the beginning of the experiment or
averaged over the duration of the experiment (details of
pH:
Temperature (°C):
Hardness (mg/L as CaCCh):
water quality parameters measured in test solutions
should be included under the results section)
Salinity (ppt):
Total Organic Carbon (mg/L):
Dissolved Organic Carbon (mg/L):
Aeration:
• Acceptable to maintain dissolved oxygen at 60 -
100% saturation at all times [1]
• Avoid aeration when testing highly oxidizable,
reducible and volatile materials [1]
• Turbulence should be minimized to prevent stress on
test organisms and/or re-suspend fecal matter [1]
• Aeration should be the same in all test chambers at all
times [1]
Yes No
Describe Preparation of Test
Concentrations (e.g., water exposure,
diet):
Test Chemical Solubility in Water:
List units and conditions (e.g., 0.01% at 20°C)
s>
Were concentrations in water or diet
verified by chemical analysis?
Measured test concentrations should be reported in
Table A.II.2 above.
Yes No
Indicate media:
-
Were test concentrations verified by
chemical analysis in tissue?
Measured test concentrations can be verified in test
organism tissue (e.g., blood, liver, muscle) alone if a
dose-response relationship is observed.
Measured test concentrations should be reported in
Table A.II.2 above.
Yes \n
Indicate tissue type:
If test concentrations were verified in test organism
tissue, was a dose-response relationship observed?
Were stability and homogeneity of test
material in water/diet determined?
Yes No
Was test material regurgitated/avoided?
Yes \n
Solvent/Vehicle Type (Water or Dictan )
• When used, a carrier solvent should be kept to a
minimum concentration [1]
• Should not affect either survival or growth oTlcsl
organisms 111
• Should be reagent grade or better [1J
• Should not exceed 0.5 ml/L (static) or 0.1 nil I. (Mow
through) unless it was shown that higher
concentrations do not aJlecl toxicity [3]
Negative Control:
Yes No
Reference Toxicant Testing:
Yes No
If Yes, identify substance:
Other Control: If any (e.g. solvent control)
Q-15
-------
Biomass Loading Rate:
• Loading should be limited so as not to affect test
results. Loading will vary depending on temperature,
type of test (static vs. flow-through), species,
food/feeding regime, chamber size, test solution
volume, etc. [1]
• This maximum number would have to be determined
for the species, test duration, temperature, flow rate,
test solution volume, chamber size, food, feeding
regime, etc.
• Loading should be sufficiently low to ensure:
o Dissolved oxygen is at least 60% of saturation
(40% for warm-water species) [1,5]
o Unionized ammonia does not exceed 35 |ig/L [1]
o Uptake by test organisms does not lower test
material concentration by > 20% [1]
o Growth of organisms is not reduced by crowding
• Generally, at the end of the test, the loading (grams of
organisms; wet weight; blotted dry) in each test
chamber should not exceed the following:
o Static tests: > 0.8 g/L (lower temperatures); > 0.5
g/L (higher temperatures) [1]
o Flow through tests: > 1 g/L/day or > 10 g/L at any
time (lower temperatures); >0.5 g/L/day or > 5
g/L at any time (higher temperatures) [1]
• Lower temperatures are defined as the lower of 17°C
or the optimal test temperature for that species [ 11
Q-16
-------
PiiniiiKMi'i-
IH'l.iils
Ki'iiiiirks
-*
l<"eediug:
• Unacceptable for acute tests [2]
o Exceptions:
¦ Data indicate that the food did not affect the
toxicity of the test material [2]
¦ Test organisms will be severely stressed if they
are unfed for 96 hours [2]
¦ Test material is very soluble and does not sorb
or complex readily (e.g., ammonia) [2]
Yes No
Lighting:
• Depends on the type of test (acute or chronic) and
endpoint (e.g., reproduction) of interest.
o Embryos should be incubated under dim
incandescent lighting (< 20 fc) or total darkness
during early life-stage toxicity testing
o Embryos must not be subjected to prolonged
exposure to direct sunlight, fluorescent lighting, or
high intensity incandescent lighting
• Generally, ambient laboratory levels (50-100 fc) or
natural lighting should be acceptable, as well as a
diurnal cycle consisting of 50% daylight or other
natural seasonal diurnal cycle.
• Artificial light cycles should have a 15 - 30-minute
transition period to avoid stress due to rapid increases
in light intensity [1]
Study Design/Methods Classification: (Place Xby One Based on Overall Study Design/Methods Classification)
Provide details of Major or Minor Deficiencies/Concerns uith Study Design in Associated Sections of Part A: Overview
This classification should be taken into consideration for the overall study classification for aquatic life criteria development in Part A.
Study Design Acceptable for Quantitative Use
Study Design Acceptable for Qualitative Use
Study Design Nol AccqMable for Use
Additional Notes: Provide additional considerations for the classification of study use based on the study design.
Q-17
-------
OBSERVATIONS: Provide information under Details and any relevant information in Remarks. This information should be
consistent with the Results Section in Part A.
Parameler
Delails
Remarks
Parameters measured including sublethal
effects/toxicity symptoms:
Common Apical Parameters Include:
Acute
• EC50 based on percentage of organisms exhibiting
loss of equilibrium plus the percentage of organisms
immobilized plus percentage of organisms killed [2]
0 If not available, the 96-hr LC50 should be used [2]
Chronic
• Life-cycle/Partial Life-cycle test:
0 Survival and growth of adults and young,
maturation of males and females, eggs spawned
per female, embryo viability (salmonids only), and
hatchability [2]
• Early life-cycle test:
0 Survival and growth T21
List parameters:
Was control survival acceptable?
Acute
• > 90% control survival at test termination [2]
Chronic
• > 80% control survival at test termination [2]
Yes \n
Control survival
Were individuals excluded from the
analysis?
Yes \n
If yes, describe justification provided:
Was water quality in test chambers
acceptable?
• If appropriate, describe any water quality issues
(e.g., dissolved oxygen level below 60% of
saturation)
Yes V<
Availability of concentration-response
data:
• Were treatment level concentration-response
data included in study publication (can be from
tables, graphs, or supplemental materials)?
specify endpoints in remarks
• Were replicate level concentration-response
data included in study publication (can be from
tables, graphs, or supplemental materials)'?
specify endpoints in remarks
Yes \n
Yes No
• If treatment and/or replicate level
concentration-response data were included, how
was data presented? (check all that applyj
Tables
Graphs
Supplemental Files
• Were concentration-response data estimated
from graphs study publication or supplemental
materials?
Yes No
If yes, indicate software used:
Yes No
• Should additional concentration-response data
be requested from study authors?
If concentration-response data are available, complete
Verification of Statistical Results (Part C) for sensitive
species.
Requested by:
Request date:
Date additional data received:
Q-18
-------
Part C: Statistical Verification of Results
I. Statistical Verification Information: Report the statistical methods (e.g., EPA TRAP, BMDS, R, other) used to verify the
reported study or test results for the five (5) most sensitive genera and sensitive apical endpoints (including for tests where such
estimates were not provided). If values for the LC50, LT50 and NOEC are greater than the highest test concentration, use the "> "
symbol.
Primary Reviewer: Date: EPA Contractor {Place X by One)
Secondary Reviewer: Date: EPA Contractor {Place X by One)
{At least one reviewer should be from EPA for sensitive taxa)
Endpoint(s) Verified:
Additional Calculated Endpoint(s):
Statistical Method (e.g., TRAP, BMDS, R, other):
II. Toxicity Values: Include confidence intervals if applicable
NOEC:
LOEC:
MATC:
ECs:
EC10:
EC20:
ECso or LCso
Dose-Response Curve Classification: (Place X by One)
This classification should be taken into consideration for llie overall study classification for aquatic life criteria development in Part A
Dose-Response Cim\ e AccqHaMe for Quantitative Use
Dose-Response Cur\ e AcccplaMc lor Qualitative Use
Dosc-Rcsponsc Cur\e \ol AccqMablc for Use
Summary of Statistical Veri Ileal ion: I'nividc .summary of methods used in statistical verification.
Additional Notes:
Attachments:
1. Provide attachments to ensure all data used in Part C are captured, whether from study results reported in the publication
and/or from additional data requested from study authors
• Data from study results of the publication should be reported in Results section of Part A
• Additional data provided upon request from study authors should be reported in Table C.II.l below and original
correspondence with study authors should be included as attachments
2. Model assessment output (including all model figures, tables, and fit metrics)
3. Statistical code used for curve fitting
Q-19
-------
III. Attachments: Include all attachments listed above after the table below.
Additional Data Used in Response-Curve: Provide all data used to fit dose-response curve not captured in Results section of PER above in Part A. Add rows as needed.
First row in italicized text is an example.
Table C.II.1 Additional Data Used in Dose-Response Curve.
( 11 I'M" II)
Spi-iii-s
r.iHipiiini
Tiv;iliiu-nl
| Si ;i n il:i I'll
IX'\ i;iiiun
HI"
Sl;i ihI;i nl
Krnir|
# of
Sun i\ urs
N'
k1
11 ¦
kl'spilllsi-
kl'spilllsi-
I nil
ClIIH'
('mil' iinils
Alchronicl
Ceriodaphnia dubia
#of
young/female
0
6
10
10
I
IS
count
0.03
mg/L
aN = number of individuals per treatment; k= number of replicates per treatment le\ el. n number ol individuals per replicate
Q-20
-------
Part D: References to Test Guidance
1. ASTM Standard E 739, 1980. 2002. Standard guide for conducting acute toxicity tests on
test materials with fishes, macroinvertebrates, and amphibians. ASTM International,
West Conshohocken, PA.
2. Stephan, C.E., D.I. Mount, D J. Hansen, I.H. Gentile, G.A. Chapman and W.A. Brungs. 1985.
Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic
Organisms and their Uses. PB85-227049. National Technical Information Service, Springfield,
VA.
3. Stephan, C.E. 1995. Review of results of toxicity tests with aquatic organisms. Draft.
U.S. EPA, MED. Duluth, MN. 13 pp.
4. OECD 203. 1992. Test No. 203: Fish, Acute Toxicity Test. OECD Guidelines for the
Testing of Chemicals, Section 2, OECD Publishing, Paris,
https://doi.org/10.1787/9789264069961-en.
5. American Public Health Association (APHA). 2012. Standard methods lor the
examination of water and wastewater. Part 8000 - Toxicity. APHA. Washington, DC.
Q-21
-------
Q.2 Example Aquatic Invertebrate DER
Part A: Overview
I. Test Information
Chemical name:
CAS name:
Purity:
Solubility in Water (units):
Controlled Experiment
(,manipulated)
Primary Reviewer:
Secondary Reviewer:
(At least one reviewer should be from EPA for sensitive taxa)
CAS Number:
Storage conditions:
Field Study/Observation
(inot manipulated)
Date:
Date:
(I'/acc X by One)
l-'.PA
LI'A
Con tractor (Place X by One)
Con tractor (Place X by One)
Citation: Indicate: author(s), year, study title, journal, volume, and/>ages.
(e.g., Keller, A.E and S.G. Zam. 1991. The acute toxicity of selected metals to the freshwater mussel. - Inoc/onta imbecilis. Environ. Toxicol. Chem. 10(4): 539-546.)
Companion Papers: Identify any companion papers associated i villi this paper using the citation format above.
(Ifyes, list file names of
Were other DERs completed lor Companion Papers? Yes No DERs below)
Study Classification for Aquatic Life Criteria Development:
AcccplaMc lor Ouanlilalive Use
Acceptable for ()iuilitati\c I se
Not Acceptable for I se/Unused
General Notes: Provide any necessary details regarding the study's use classification for all pertinent endpoints, including
non-apical endpoints within the study (e.g., note all study classifications for each endpoint if the use varies)
Major Deficiencies (note any stated exclusions): Check all that apply. Checking any of these items make the study "Not
Acceptable for Use"
, , ,,, ,, , . , . . No Controls (for controlled experiments
Mixture (tor controlled experiments only) only)
Excessive Control Mortality (> 10% for acute and > 20% for chronic)
. , . . . . Bioaccumulation: steady state not
Dilution water not adequately characterized , ,
n J reached
Dermal or Injection Exposure Pathway
Review paper or previously published without modification
Q-22
-------
Other: (if any, list here)
POTENTIAL CHEMICAL MIXTURES : Describe any potential chemicals mixtures as characterized by study
authors (including any confirmation of chemical mixtures).
DESCRIPTION OF DILUTION WATER: Describe concerns with characterization of and/or major deficiencies
with dilution water.
General Notes:
Minor Deficiencies: List and describe any minor deficiencies or other concerns with lest. These items may make the study
"Acceptablefor Qualitative Use" (exceptions may apply as noted)
For Field Studies/Observations: A field study/observation mav he considered "Acceptable for Quantitative Use" if it
consisted of a range of exposure concentrations and the observed effects are justifiably contributed to a single chemical
exposure
Mixture (observed effects not justifiably contribulcd In single chemical exposure)
Uncharacterized Reference Sites/Conditions
POTENTIAL CHEMICAL MIXTURES PRESENT AT SITE: Describe any potential chemicals mixtures present at
the site as characterized by study authors (including any confirmation of chemicals present at study site).
EXPOSURE VARI ABILITY ACROSS STUDY SITE(SV Describe any exposure variability across study site(s)
as characterized by study authors (i.e.. description of study design with reference and contaminated sites).
General Notes:
Reviewer's Comments: Provide additional comments that do not appear under other sections of the template.
Q-23
-------
ABSTRACT: Copy and paste abstract from publication.
SUMMARY: Fill out and modify as needed.
Aaile
Species (lilesliiiie)
Method'1
Test
dunilion
( hemiciil
/ Piiriu
pll
1 cmp.
<°C)
Hardness
(111li/l. ilS
CaiCO.o
or
S;ilini(\
(DDII
DOC
(lliu/l -)
i:iTcci
Kcporicd
i:ilcc(
( onceiil r;ition
(nii»/l.)
Verified
r.ii'cci
( onccnlr;ilion
(mii/l.)
CI;issific;ilion
Quantitative / Qualitative /
Unused
3 S=static, R=renewal, F=flow-through, U=unmeasured, M=measured, T=total, D=dissolved, Diel diclars. M l maternal transfer
Chronic:
Species (lil'cslii^c)
Melliod-'
lesl
(lui'iilioii
( hemiciil
/ Piiriu
|)ll
Temp.
<°C)
lliirdness
(lllfi/l. ilS
CaCO.o
or
S;ilinil\
(ppl)
DOC
(mii/l.)
Chronic
Limits
Kcporicd
Chronic
\ iilne
(mii/l. or
Verified
C hroilic
VillllC
(mii/l. or
Chronic
\ illllC
I'lndpoinl
CI:issiric;ilion
Quantitative /
Qualitative / Unused
a S=static, R=renewal, F=flow-through, U=unmeasuied. \1 measured. I U>i;il. I) dissnhcd. Dicl=dietary, MT=maternal transfer
Q-24
-------
II. Results Provide results as reported in the publication (including supplemental materials). Include screen shots of tables and/or
figures reporting results from the article following tabulated data table in each associated results section for all studies. Complete
tabulated data tables for all studies for studies marked "Acceptable for Quantitative Use" and "Acceptable for Qualitative Use".
Water Quality Parameters: If only general summary data of water quality parameters is provided by study authors (i.e., no
specific details of water quality parameters on a treatment level is provided), summarize any information regarding water quality
parameters under General Notes below and include data not provided in Table A. II. 1.
General Notes: For aquatic life criteria development, measured water quality parameters in the treatments nearest the toxicity
test endpoint(s), e.g., LC50, EC20, etc., are most relevant.
Table A.II.1. Measured Water Quality Parameters in Test Solutions.
Dissolved oxygen, temperature, pH and [other parameters (hardness, s; il i 11 i i\. I)()(') | 111 lesi solutions during the [X]-day
exposure of [test organism] to [concentration of treatments) / of [test substance/ under I sialic renewal/flow-through]
conditions.
Parameler
Treatment
Mean
Range
Dissolved
oxygen
(% saturation
or mg/L)
[1]
[2]
j
j
Temperature
(O
[I]
[2]
j
j
pi"
HI
12/
i
J
Oilier (e.g..
hardness,
salinity, DOC)
11J
12]
J
J
Q-25
-------
Chemical Concentrations: Summarize the concentration verification data from test solutions/media. Expand table to include
each measured concentration data for each media type (i.e., muscle, liver, blood, etc.).
General Notes: Provide any necessary detail regarding the measured concentrations, including any identified cause for
substantial differences between nominal and measured concentrations, if samples were collected on separate days (and if so provide
details), and any potential cross contamination.
Table A.II.2. Measured (and Nominal) Chemical Concentrations in Test Solutions/Media.
[Analytical Method] verification of test and control concentrations during an |X|-d;i\ exposure of [test organism] to [test
substance] under [static renewal/flow-through] conditions.
|Mi*;in|
Nil ill hoi' of
| Si ;i nthi r«l
Nomiiiiil
Mo;isiiiv(l
Siimplcs
l)c\ iiilion or
( (IIKTIlll'illioil
( niHTiili'iilion
Number of
Non-
Ik'low Son-
Siiindiinl
1 iviilmonl
( ii nils)
( ii iii Is)
S;i in pies
DolocC
Dclcd
Krroil
KiiiiUO
Control
[1]
[2]
[3]
[4]
[5]
[6]
./'
aNon-Detect: 0 = measured and detected; l=]iicasuivd and noi detected; if noi measured or reported enter as such
Q-26
-------
Mortality: Briefly summarize mortality results (if any).
General Notes: Comment on concentrations response relations and slope of response ifprovided. Compare mortality with control
treatment and/or the reference chemical.
Table A.II.3. Mean Percent [Mortality or Survival].
Mean percent mortality [or number of immobilized] or survival of [test organism] exposed to [test substance] for [test
duration] under [static/renewal/flow-through] conditions.
| Menu "i.
|Si;iii«I;ir«l l)c\ iiiiion
1 iviilmcnl
Mor(;ili(\ |
or Siiindiii'd l-'rr«ir|
Control
[11
[21
[31
[41
[51
[61
[LCxl
NOEC
LOEC
a Use superscript to identify the values reported U> he simnlicaiilly different from control.
Q-27
-------
Growth: Briefly summarize growth results (if any).
General Notes: Comment on concentrations response relations and slope of response if provided. Compare growth endpoints with
control treatment and/or the reference chemical.
Table A.II.4. Mean [Growth].
Mean growth [length and/or weight] of [test organism] exposed to [test substance] for [test duration] under
[static/renewal/flow-through] conditions.
Menu (.row111
Moiin IVivcnl
|l. en»( h/\\oiiih(|
|Si;inthird IK'\i;i(ion
( hiinjic in ll.on^lh/
|Si;iii«l;ird l)c\ iiiiion
1 iviilmonl
( ii nils)
or Si;iii«hir«l I jtoi'I
liioiiiiissj
or Si;iii(hird I'lrrorj
Control
[1]
[2]
[3]
[4]
[5]
[6]
./'
lECxl
NOEC
LOEC
aUse superscript to identify the values reported In he simnlicaiilU differeni linm cnniinl
Q-28
-------
Reproductive: Briefly summarize reproduction endpoint results (if any). For multi-senerational studies, copy and paste Table
A.II.5 below for each generation with reproductive effects data.
General Notes: Comment on concentrations response relations and slope of response if provided. Compare reproduction
endpoints with control treatment and/or the reference chemical.
Table A.II.5. Mean [Reproductive] Effect.
Mean [reproductive] effects for [generation] of [test organism] exposed to [test substance] for [test duration] under
[static/renewal/flow-through] conditions.
|Si:iii(l;ir(l
ISliindiird
| Mciin
ISliindiird
l)c\ hilion
l)o\ iiilion
Nil m her
l)c\ iiilion or
|Moiin
or
|Mciin
or
1 IVillllKMII
ol'
Sliindiird
Number ol°
Sliindiird
Number of
Sliindiird
(ii nils)
S|)ii\\ns|
r.rrorl
I''litis |
r.rror|
OITspriuiil
i:rror|
Control
[1]
[2]
[3]
[4]
[5]
[6]
./'
[ECx]
NOEC
LOEC
a Use superscript to identify the values reported to he simnlicaiilK dilTeiviil I'mm control.
Q-29
-------
Sublethal Toxicity Endpoints: Include other sublethal effect(s), including behavioral abnormalities or other signs of toxicity,
if any. Copy Table A.II.6 as needed to provide details for each sublethal effect observed.
General Notes: Briefly summarize observed sublethal effects otherwise not captured in the results table(s) below.
Table A.II.6. Mean [Sublethal] Effect.
Mean /"Sublethal effect (e.g., behavioral abnormalities, etc.)] in [test organism] during [test duration (acute/chronic)]
exposure to [test substance] under [static/renewal/flow-through] conditions.
|Mcan Sublethal
Rc'sponsi'l
|S(audai (l lk'\ia(ion or
Treadmill
( ii nils)
Siiindiinl I'.itoi-|
Control
[1]
[2]
[3]
[4]
[5]
[6]
./'
|ECxl
NOEC
LOEC
a Use superscript to identify the \ allies ivpoi icd In be siginl icanll> different from control
Q-30
-------
Reported Statistics: Copy and paste statistical section from publication.
Q-31
-------
Part B: Detailed Review
I. Materials and Methods
PROTOCOL/GUIDANCE FOLLOWED: Indicate ifprovided by authors.
DEVIATIONS FROM PROTOCOL: If authors report any deviations from the protocol noted above indicate here.
Study Design and Methods: Copy and paste methods section from publication.
TEST ORGANISM: Provide information under Details and any relevant or related information or clarifications in Remarks.
Parameter
Details
Remarks
Species:
Common Name:
Scientific Name:
North American species?
Surrogate for North American
Taxon?
(Place A' if applicable)
Strain/Source:
• Wild caught from unpolluted areas [1]
o Quarantine for at least 7 days or until they are
disease free, before acclimation [1]
• Must originate from same source and population [1]
• Should not be used:
o If appeared stressed, such as discoloration or
unusual behavior [1]
o If more than 5% die during the 48 hours before
test initiation [ 1 ]
o If they were used in previous test treatments or
controls [2]
• No treatments of diseases may be administered:
o Within 16 hours of field collection [1]
o Within 10 days of testing or during testing [11
Age at Study Initiation:
Acute:
• Larval stages preferred [1]
• Mayflies and Stoneflies
o Early instar [1]
• Daphnids/cladocerans:
o < 24-hr old [1]
• Midges:
o 2nd or 3rd instar larva [1]
• Hyalella azteca (chronic exposure)
o Generally, 7-8 days old [3]
• Freshwater mussels (chronic exposure)
o Generally, 2 month old juveniles [4]
• Mysids (chronic exposure)
o < 24-hr old Til
Was body weight or length recorded at
test initiation and/or at regular intervals?
Yes No
Was body weight or length recorded at
regular intervals?
Yes No
If yes, describe regular inten'als:
Q-32
-------
STUDY PARAMETERS: Provide information under Details and any relevant information of deficiencies in Remarks.
Complete for both Controlled Experiments and Field Studies Observations.
Paramcler
Delails
Remarks
Number of Replicates per Treatment
Group:
• At least 2 replicates/treatment recommended for
acute tests [1]
• At least 2 replicates/treatment recommended for
chronic tests [5]
Control(s):
Treatment(s):
Number of Organisms per Replicate/
Treatment Group:
• At least 10 organisms/treatment recommended.
Control(s):
Treatment(s):
1
Exposure Pathway:
(i.e., water, sediment, or diet). Note: all other pathways
(e.g., dermal, injection) are unacceptable.
Exposure Duration:
Acute
• Cladocerans and midges should be 48 hours [2]
o Longer durations acceptable if test species not fed
and had acceptable controls [2]
• Freshwater mussel glochidia should be a maximum
of 24 hours [4]
o Shorter durations (6, 12, 18 hours) acceptable so
long as 90% survival of control animals achieved
(see below) [4]
• Embryo/larva (bivalve mollusks, sea urchins,
lobsters, crabs, shrimp and abalones) should be 96
hours, but at least 48 hours [2]
Acute
( limine
<
• Other invertebrate species should be 96 hours
Other (please remark):
¦i
Chronic
• Daphnids/cladocerans should be 21 days (3-brood
test) [2]
o Exception 7 days acceptable for Ceriodaphnia
dubia [2]
• Freshwater juvenile mussels should be al least 28
days [4]
• Hyalella azteca should be at least 42 days
o Beginning with 7-8 day old animals |31
• Mysids should continue until 7 days pasl the median
time of first brood release in the controls 14|
Test Concentrations (remember iniils):
Nmmiial
Recommended test concentrations include ai least three
concentrations other than the control: four or more will
provide a better statistical analysis.
Measured:
Media measured in:
Observation Intervals:
• Should be an appropriate number of observations
over the study to ensure waler quality is being
properly maintained f 11
Q-33
-------
CONTROLLED EXPERIMENT STUDY PARAMETERS: Provide information under Details and any relevant
information of deficiencies in Remarks. Complete for Controlled Experiments only.
ParnimMcr
Details
Remarks
Acclimation/Holding:
• Should be placed in a tank along with the water in
Duration:
Identify number of individuals excluded from testing and/or
analysis (if any):
which they were transported [1]
o Water should be changed gradually to 100%
dilution water (usually 2 or more days) [1]
o For wild-caught animals, test water temperature
should be within 5°C of collection water
Feeding:
Water:
temperature [1]
o Temperature change rate should not exceed 3°C
Temperature (°C):
within 72 hours [1]
• To avoid unnecessary stress and promote good
health:
o Organisms should not be crowded [1]
o Water temperature variation should be limited
o Dissolved oxygen:
¦ Maintain between 60 - 100% saturation [1]
¦ Continuous gentle aeration if needed [1]
o Unionized ammonia concentration in holding and
acclimation waters should be < 35 jag/L [1]
Dissolved Oxygen (mg/L):
Health (any mortality observed?):
-
Acclimation followed published guidance?
Describe, if any
Yes \n
-*
If yes, indicate which guidance:
s
Test Vessel:
• Test chambers should be loosely covered [1]
Material:
Briefly describe the test vessel here
o Should minimize sorption of test chemical from
water [1]
o Should not contain substances that can be leached
Size:
£
or dissolved in solution and free of substances that
could react with exposure chemical [1]
o Glass, No. 316 stainless steel, nylon screen and
Fill Volume'
perfluorocarbon (e.g. Teflon) arc acceptable 111
o Rubber, copper, brass, galvanized nielal. epoxv
glues, lead and flexible tubing should noi come
into contact with test solution, dilution water or
stock [1]
• Size/volume should maintain acceptable bioniass
loading rates (see below) 111
• Substrate:
o Required for some species (e.g.. Hvalclla azicca)
[3]
o Common types: stainless steel screen, nylon
screen, quail/, sand, cotton gauze and maple leaves
[3]
o More inert substances preferred over plant
material, since plants may break down during
testing and promote bacterial growth [3]
o Consideration should be given between substrate
and toxicant [3]
¦ Hydrophobic organic compounds in particular
can bind strongly to Nitex® screen, reducing
exposure concentrations, especially for studies
using static or intermittent renewal exposure
methods [31
Sllllsll';||e I sod (ifapplicable)'.
Q-34
-------
Parameter
Details
Remarks
Test Solution Delivery System/Method:
• Flow-through preferred for some highly volatile,
hydrolyzable or degradable materials [2]
o Concentrations should be measured often enough
using acceptable analytical methods [2]
• Chronic exposures:
o Flow-through, measured tests required [2]
o Exception: renewal is acceptable for daphnids [2]
Test Concentrations Measured
Yes No
Test Solution Delivery System:
Static
Renewal
Indicate Interval:
Flow-through
Indicate Type of Diluter
Source of Dilution Water:
• Freshwater hardness range should be < 5 mg/L or <
10% of the average (whichever is greater) [1]
• Saltwater salinity range should be < 2 g/kg or < 20%
of the average (whichever is greater) [1]
• Dilution water must be characterized (natural surface
water, well water, etc.) [2]
o Distilled/deionized water without the addition of
appropriate salts should not be used [2]
• Dilution water in which total organic carbon or
particulate matter exceed 5 mg/L should not be used
o Unless data show that organic carbon or particulate
matter do not affect toxicity [2]
• Dilution water for tests with Hyalella azteca
o Reconstituted waters should have at least 0.02 mg
bromide/L; natural ground or surface water
presumed to have sufficient bromide [3]
o Recommended that control/dilution waters have
chloride concentrations at or above 15 mg/L [3]
Dilution Series (e.g., 0.5x, 0.6x, etc.):
Dilution Water Parameters:
Measured at the beginning of the experiment or
averaged over the duration of the experiment (details of
water quality parameters measured in test solutions
should be included under the results section)
Dissolved Oxygen ung/L):
pH:
Temperature (°C):
Hardness (mg/L as CaC03):
Salinity (ppt):
Total Organic Carbon (mg/L):
1 )issolved Organic Carbon (mg/L):
Aeration:
• Acceptable to maintain dissolved oxygen al 60 -
100% saturation al all limes [1]
• Avoid aeration when testing highly oxidizable.
reducible and volatile materials
• Turbulence should be minimized to prevent stress on
test organisms and/or re-suspend fecal mailer 111
• Aeration should be the same in all lesl chambers al all
times [1]
Yes No
Describe Preparation of Test
Concentrations (e.g., water exposure,
diet):
Q-35
-------
Parameter
Details
Remarks
Test Chemical Solubility in Water:
• List units and conditions (e.g., 0.01% at 20°C)
Were concentrations in water or diet
verified by chemical analysis?
Measured test concentrations should be reported in
Table A.II.2 above.
Yes No
Indicate media:
Were test concentrations verified by
chemical analysis in tissue?
Measured test concentrations can be verified in test
organism tissue (e.g., blood, liver, muscle) alone if a
dose-response relationship is observed.
Measured test concentrations should be reported in
Table A.II.2 above.
Yes No
If test concentrations were verified in test organism
tissue, was a dose-response relationship observed?
Indicate tissue type:
Were stability and homogeneity of test
material in water/diet determined?
Yes \o
Was test material regurgitated/avoided?
Yes V-
Solvent/Vehicle Type:
• When used, a carrier solvent should be kept to a
minimum concentration [1]
• Should not affect either survival or growth of test
organisms [1]
• Should be reagent grade or better [1]
• Should not exceed 0.5 ml/L (static), or 0.1 ml/L (flow
through) unless it was shown that higher
concentrations do not affect toxicity [5]
Negative Control:
Yes No
Reference Toxicant Testing:
Yes No
If yes, identify substance:
Other Control: If any (e.g. solvent control I
Biomass Loading Rate:
• Loading should be limited so as nol lo alfccl test
results. Loading will vary depending on icmpcrnlurc.
type of test (static vs. flow-through), species,
food/feeding regime, chamber size, test solution
volume, etc. [1]
• This maximum number would have lo be determined
for the species, test duration, temperature. How rale,
test solution volume, chamber size. Ibod. leediim
regime, etc.
• Loading should be sufficiently low to ensure:
o Dissolved oxygen is at least 60% of saturation
(40% for warm-water species) [1,6]
o Unionized ammonia does not exceed 35 |ig 1. 111
o Uptake by test organisms does not lower tesl
material concentration by 20% [1]
o Growth of organisms is nol reduced by crowding
• Generally, at the end of the test, the loading (grams of
organisms; wet weight; blotted dry) in each test
chamber should not exceed the following:
o Static tests: > 0.8 g/L (lower temperatures); > 0.5
g/L (higher temperatures) [1]
o Flow through tests: > 1 g/L/day or > 10 g/L at any
time (lower temperatures); >0.5 g/L/day or > 5
g/L at any time (higher temperatures) [1]
o Lower temperatures are defined as the lower of
17°C or the optimal test temperature for that
species. [1]
Q-36
-------
Feeding:
• Unacceptable for acute tests [2]
o Exceptions:
¦ Data indicate that the food did not affect the
-*
toxicity of the test material [2]
Yes No
c.
¦ Test organisms will be severely stressed if they
are unfed for 96 hours [2]
¦ Test material is very soluble and does not sorb
>
or complex readily (e.g., ammonia) [2]
Lighting:
• No specific requirements for lighting
• Generally, ambient laboratory levels (50 - 100 fc) or
£
natural lighting should be acceptable, as well as a
diurnal cycle consisting of 50% daylight or other
natural seasonal diurnal cycle
• Artificial light cycles should have a 15 - 30 minute
-C.
transition period to avoid stress due to rapid increases
in light intensity [1]
• Depends on the type of test (acute or chronic) and
endpoint (e.g., reproduction) of interest.
Study Design/Methods Classification: (Place Xby One Based on Overall Study Design/Methods Classification)
Provide details of Major or Minor Deficiencies/Concerns with Study Design in Associated Sections of Part A: Overview
This classification should be taken into consideration for the overall study classification for aquatic life criteria development in Part A.
Study Design Acceptable for Quantitative Use
Study Design Acceptable for Qualitative Use
Study Design Not Acceptable for I [se
Additional Notes: Provide additional considerations far the classification of study use based on the study design.
Q-37
-------
OBSERVATIONS: Provide information under Details and any relevant information in Remarks. This information should be
consistent with the Results Section in Part A.
Paramclcr
Delails
Remarks
Parameters measured including sublethal
effects/toxicity symptoms:
Common Apical Parameters Include:
Acute
• Daphnids/cladocerans:
o EC50 based on percentage of organisms
immobilized plus percentage of organisms killed
[2]
• Embryo/larva (bivalve molluscs, sea urchins, lobsters,
crabs, shrimp, and abalones):
0 EC50 based on the percentage of organisms with
incompletely developed shells plus the percentage
of organisms killed [2]
¦ If not available, the lower of the 96 hour EC50
based on the percentage of organisms with
incompletely developed shells and the 96-hr
LC50 should be used [2]
• Freshwater mussel (glochidia and juveniles):
0 Glochidia: EC50 based on 100 x number closed
glochidia after adding NaCl solution - number
closed glochidia before adding NaCl solution) /
Total number open and closed glochidia after
adding NaCl solution [4]
0 Juvenile: EC50 based on percentage exhibiting foot
movement within a 5-min observation period [4]
• All other species and older life stages:
0 EC50 based on the percentage of organisms
exhibiting loss of equilibrium plus the percentage
of organisms immobilized plus the percentage of
organisms killed [2]
¦ If not available, the 96 hour LC50 should be
used [2]
Chronic
• Daphnid:
0 Survival and young per female [2]
• Mysids:
0 Survival, growth and young per female 121
List parameters:
Was control sun i\ al acceptable?
Acute
• > 90% control survival al lost termination |2|
0 Glochidia 90% alter 24 hours, or. the next longest
duration less than 24 hours that had al least 90° 0
survival [4]
Chronic
• > 80% control survival at lesi termination |2|
0 80% in 42 day tesi with Hyalclla azteca. slighlly
lower in tests substantially longer than 42 davs |3|
Yes No
Cmili'ol survival (%):
Q-38
-------
Paramcler
Delails
Remarks
Were individuals excluded from the
analysis?
Yes No
If yes, describe justification provided:
Was water quality in test chambers
acceptable?
• If appropriate, describe any water quality issues
(e.g., dissolved oxygen level below 60% of
saturation)
Yes No
Availability of concentration-response
data:
• Were treatment level concentration-response
data included in study publication (can be from
tables, graphs, or supplemental materials)?
specify endpoints in remarks
Yes No
• Were replicate level concentration-response
data included in study publication (can be from
tables, graphs, or supplemental materials)?
specify endpoints in remarks
•
Yes No
• If treatment and/or replicate level
concentration-response data were included, how
was data presented? (check all that apply)
Tables
Graphs
Supplemental Fi les
• Were concentration-response data estimated
from graphs study publication or supplemental
materials?
Yes No
If yes, indicate software used:
'les \n
Should additional concentration-response data be
requested from study authors?
If concentration-response data are available, complete
Verification of Statistical Results (Part C) lor sensitive
species.
Requested by:
Request date:
Dale additional data received:
Q-39
-------
Part C: Statistical Verification of Results
I. Statistical Verification Information: Report the statistical methods (e.g., EPA TRAP, BMDS, R, other) used to verify the
reported study or test results for the five (5) most sensitive genera and sensitive apical endpoints (including for tests where such
estimates were not provided). If values for the LC50, LT50 and NOEC are greater than the highest test concentration, use the "> "
symbol.
Primary Reviewer: Date: EPA Contractor {Place X by One)
Secondary Reviewer: Date: EPA Contractor {Place X by One)
{At least one reviewer should be from EPA for sensitive taxa)
Endpoint(s) Verified:
Additional Calculated Endpoint(s):
Statistical Method (e.g., TRAP, BMDS, R, other):
II. Toxicity Values: Include confidence intervals if applicable
NOEC:
LOEC:
MATC:
ECs:
EC10:
EC20:
ECso or LCso
Dose-Response Curve Classification: (Place X by One)
This classification should be taken into consideration for llie overall study classification for aquatic life criteria development in Part A
Dose-Response Cim \ e Acceptable for Quantitative Use
Dose-Response Cin\ e Acceptable lor Qualitative Use
Dose-Response Cui \e Not Acceptable for Use
Summary of Statistical Ver ideal ion: I'nividc .summary of methods used in statistical verification.
Additional Notes:
Attachments:
1. Provide attachments to ensure all data used in Part C is captured, whether from study results reported in the publication
and/or from additional data requested from study authors
• Data from study results of the publication should be reported in Results section of Part A
• Additional data provided upon request from study authors should be reported in Table C.II.l below and original
correspondence with study authors should be included as attachments
2. Model assessment output (including all model figures, tables, and fit metrics)
3. Statistical code used for curve fitting
Q-40
-------
III. Attachments: Include all attachments listed above after the table below.
Additional Data Used in Response-Curve: Provide all data used to fit dose-response curve not captured in Results section of PER above in Part A, rows as needed. First
row in italicized text is an example.
Table C.II.1 Additional Data Used in Dose-Response Curve.
( 11 I'M" II)
Spi-iii-s
r.iHipiiini
Tiv;iliiu-nl
| Si ;i n il:i I'll
IX'\ i;iiiun
HI"
Sl;i ihI;i nl
Krnir|
# of
Sun i\ urs
N'
k1
11 ¦
kl'spilllsi-
kl'spilllsi-
I nil
ClIIH'
('mil' iinils
Alchronicl
Ceriodaphnia dubia
#of
young/female
0
6
10
10
I
IS
count
0.03
mg/L
aN = number of individuals per treatment; k= number of replicates per treatment le\ el. n number ol individuals per replicate
Q-41
-------
Part D: References to Test Guidance
6. ASTM Standard E 739, 1980. 2002. Standard guide for conducting acute toxicity tests on
test materials with fishes, macroinvertebrates, and amphibians. ASTM International,
West Conshohocken, PA.
7. Stephan, C.E., D.I. Mount, D J. Hansen, J.H. Gentile, G.A. Chapman and W.A. Brungs. 1985.
Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic
Organisms and their Uses. PB85-227049. National Technical Information Service, Springfield,
VA.
8. Mount, D.R. and J.R. Hockett. 2015. Issue summary regarding test conditions and
methods for water only toxicity testing with Hyalella azleca. Memorandum to Kathryn
Gallagher, U.S. EPA Office of Water. U.S. EPA Office of Research and Development.
MED. Duluth, MN. 9 pp.
9. Bringolf, R.B., M.C. Barnhart, and W.G. Cope. 2013. Determining the appropriate duration of
toxicity tests with glochidia of native freshwater mussels. Submitted to I ^Iwaixl I lammer. U.S.
EPA. Chicago, IL, May 8, 2013. 39 pp.
10. Stephan, C.E. 1995. Review of results of toxicity tests with aquatic organisms. Draft.
U.S. EPA, MED. Duluth, MN. 13 pp.
11. American Public Health Association (APHA). 2<) 12. Standard methods for the
examination of water and wastewater. Part 8000 - Toxicity APHA. Washington, DC.
Q-42
------- |