xvEPA
EPA/635/R-ll/001Fa
United States www.epa.gov/iris
Environmental Protection
Agency
TOXICOLOGICAL REVIEW
OF
METHANOL (NONCANCER)
(CAS No. 67-56-1)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
September 2013
U.S. Environmental Protection Agency
Washington, DC
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
-------
CONTENTS TOXICOLOGICAL REVIEW OF
METHANOL (Noncancer)(CAS NO. 67-56-1)
CONTENTS TOXICOLOGICAL REVIEW OF METHANOL (Noncancer)(CAS NO. 67-56-1) iii
LIST OF TABLES v
LIST OF FIGURES vii
LIST OF ABBREVIATIONS AND ACRONYMS viii
AUTHORS, CONTRIBUTORS, AND REVIEWERS xvii
EXECUTIVE SUMMARY xxi
INTRODUCTION xxi
CHEMICAL AND PHYSICAL INFORMATION xxii
TOXICOKINETICS xxii
HAZARD IDENTIFICATION xxiv
DOSE-RESPONSE ASSESSMENT AND CHARACTERIZATION xxvi
RELATIONSHIP OF THE RFC AND RFD TO BACKGROUND METHANOL BLOOD LEVELS AND
MONKEYBLOOD LEVELS ASSOCIATED WITH EFFECTS OF UNCERTAIN ADVERSITY xxviii
1... INTRODUCTION 1-1
2...CHEMICAL AND PHYSICAL INFORMATION 2-1
3... TOXICOKINETICS 3-1
3.1. OVERVIEW 3-1
3.2. KEY STUDIES 3-11
3.3. HUMAN VARIABILITY IN METHANOL METABOLISM 3-18
3.4. PHYSIOLOGICALLY BASED PHARMACOKINETIC MODELS 3-20
3.4.1. Model Requirements for EPA Purposes 3-20
3.4.2. MethanolPBPK Models 3-26
3.4.3. Selected Modeling Approach 3-27
3.4.4. Monkey PK Data and Analysis 3-32
3.4.5. Summary and Conclusions 3-33
4...HAZARD IDENTIFICATION 4-1
4.1. STUDIES IN HUMANS - CASE REPORTS, OCCUPATIONAL AND CONTROLLED STUDIES 4-1
4.1.1. Case Reports 4-1
4.1.2. Occupational Studies 4-3
4.1.3. Controlled Human Studies 4-5
4.2. ACUTE, SUBCHRONIC AND CHRONIC STUDIES IN ANIMALS - ORAL AND
INHALATION 4-7
4.2.1. Oral Studies 4-7
4.2.2. Inhalation Studies 4-11
4.3. REPRODUCTIVE AND DEVELOPMENTAL STUDIES - ORAL AND INHALATION 4-20
4.3.1. Oral Reproductive and Developmental Studies 4-20
4.3.2. Inhalation Reproductive and Developmental Studies 4-23
4.3.3. Other Reproductive and Developmental Studies 4-40
4.4. NEUROTOXICITY 4-47
4.4.1. Oral Neurotoxicity Studies 4-47
4.4.2. Inhalation Neurotoxicity Studies 4-50
-------
4.4.3. Neurotoxicity Studies Employing i.p. and in vitro Methanol Exposures 4-58
4.5. IMMUNOTOXICITY 4-63
4.6. SYNTHESIS OF MAJOR NONCANCER EFFECTS 4-68
4.6.1. Summary of Key Studies in Methanol Toxicity 4-68
4.7. NONCANCER MOA INFORMATION 4-77
4.7.1. Role of Methanol and Metabolites in the Developmental Toxicity of Methanol 4-78
4.7.2. Role of Folate Deficiency in the Developmental Toxicity of Methanol 4-83
4.7.3. Methanol-Induced Formation of Free Radicals, Lipid Peroxidation, and Protein Modifications 4-84
4.7.4. Exogenous Formate Dehydrogenase as a Means of Detoxifying the Formic Acid that Results
from Methanol Exposure 4-88
4.7.5. Summary and Conclusions Regarding MOA for Developmental Toxicity 4-89
4.8. EVALUATION OF CARCINOGENICITY 4-90
4.9. SUSCEPTIBLE POPULATIONS AND LIFE STAGES 4-90
4.9.1. Possible Childhood Susceptibility 4-90
4.9.2. Possible Gender Differences 4-91
4.9.3. Genetic Susceptibility 4-92
5...DOSE-RESPONSE ASSESSMENTS 5-1
5.1. INHALATION REFERENCE CONCENTRATION (RFC) 5-1
5.1.1. Choice of Principal Study and Critical Effect(s) 5-1
5.1.2. Methods of Analysis for Identifying the POD—Application of PBPK and BMD Models 5-7
5.1.3. RfC Derivation - Including Application of Uncertainty Factors 5-16
5.1.4. Previous RfC Assessment 5-24
5.2. ORAL REFERENCE DOSE (RFD) 5-25
5.2.1. Choice of Principal Study and Critical Effect-with Rationale and Justification 5-25
5.2.2. RfD Derivation-Including Application of Uncertainty Factors 5-28
5.2.3. Previous RfD Assessment 5-29
5.3. UNCERTAINTIES IN THE INHALATION RFC AND ORAL RFD 5-30
5.3.1. Choice of Study/Endpoint 5-31
5.3.2. Choice of Model for BMDL Estimation 5-35
5.3.3. Route-to-Route Extrapolation 5-35
5.3.4. Statistical Uncertainty at the POD 5-36
5.3.5. Choice of Species/Gender 5-36
5.3.6. Relationship of the RfC and RfD to Background Levels of Methanol in Blood 5-38
5.3.7. Relationship of the RfC and RfD to Methanol Blood Levels In Monkeys Associated with
Unquantifiable Effects of Uncertain Adversity 5-40
5.4. CANCER ASSESSMENT 5-42
6...REFERENCES 6-1
IV
-------
LIST OF TABLES
Table 2-1 Relevant physical and chemical properties of methanol 2-1
Table 3-1 Background blood methanol and formate levels in human studies 3-2
Table 3-2 Human blood methanol and formate levels following methanol exposure 3-3
Table 3-3 Monkey blood methanol and formate levels following methanol exposure 3-4
Table 3-4 Mouse blood methanol and formate levels following methanol exposure 3-5
Table 3-5 Rat blood methanol and formate levels following methanol exposure 3-6
Table 3-6 Plasma methanol concentrations in monkeys 3-16
Table 3-7 Plasma formate concentrations in monkeys 3-17
Table 3-8 Serum folate concentrations in monkeys 3-17
Table 3 -9 Routes of exposure optimized in models - optimized against blood concentration data 3-27
Table 3-10 Key methanol kinetic studies for model validation 3-30
Table 4-1 Mortality rate for subjects exposed to methanol-tainted whisky in relation to their level of
acidosis 4-2
Table 4-2 Reproductive and developmental toxicity in pregnant Sprague-Dawley rats exposed to
methanol via inhalation during gestation 4-25
Table 4-3 Reproductive parameters in Sprague-Dawley dams exposed to methanol during pregnancy, and
then allowed to deliver their pups 4-27
Table 4-4 Embryonic and Developmental effects in CD-I mice after methanol inhalation 4-30
Table 4-5 Benchmark doses at two added risk levels 4-31
Table 4-6 Developmental Phase-Specific Embryotoxicity and Teratogenicity in CD-I mice after
methanol inhalation 4-32
Table 4-7 Developmental phase-specific embryotoxicity in CD-I mice induced by methanol inhalation
(15,000 ppm) during neurulation 4-33
Table 4-8 Reproductive parameters in monkeys exposed via inhalation to methanol during prebreeding,
breeding, and pregnancy 4-36
Table 4-9 Mean serum levels of testosterone, luteinizing hormone, and corticosterone (± SD) in male
Sprague-Dawley rats after inhalation of methanol, ethanol, n-propanol or n-butanol at
threshold limit values 4-38
Table 4-10 Maternal and litter parameters when pregnant female C57BL/6J mice were injected i.p. with
methanol 4-41
Table 4-11 Developmental studies of rodent embryos exposed to methanol 4-44
Table 4-12 Reported thresholds concentrations (and author-estimated ranges) for the onset of embryotoxic
effects when rat and mouse conceptuses were incubated in vitro with methanol, formaldehyde,
and formate 4-46
Table 4-13 Brain weights of rats exposed to methanol vapors during gestation and lactation 4-56
Table 4-14 Intraperitoneal injection neurotoxicity studies 4-61
Table 4-15 Effect of methanol on Wistar rat acetylcholinesterase activities 4-63
Table 4-16 Effect of methanol on neutrophil functions in in vitro and in vivo studies in male Wistar rats 4-64
-------
Table 4-17 Effect of intraperitoneally injected methanol on total and differential leukocyte counts and
neutrophil function tests in male Wistarrats 4-65
Table 4-18 Effect of methanol exposure on animal weight/organ weight ratios and on cell counts in
primary and secondary lymphoid organs of male Wistarrats 4-67
Table 4-19 The effect of methanol on serum cytokine levels in male Wistarrats 4-68
Table 4-20 Summary of noncancer effects reported in repeat exposure and developmental studies of
methanol toxicity in experimental animals (oral) 4-69
Table 4-21 Summary of repeat exposure and developmental studies of methanol toxicity in experimental
animals (inhalation exposure) 4-70
Table 4-22 Developmental outcome on GD10 following a 6-hour 10,000 ppm (13,104 mg/m3) methanol
inhalation by CD-mice orformate gavage (750 mg/kg) onGDS 4-79
Table 4-23 Summary of ontogeny of relevant enzymes in CD-I mice and humans 4-80
Table 4-24 Dysmorphogenic effect of methanol and formate in neurulating CD-I mouse embryos in
culture (GD8) 4-81
Table 4-25 Time-dependent effects of methanol administration on serum liver and kidney function, serum
ALT, AST, BUN, and creatinine in control and experimental groups of male Wistar rats 4-87
Table 4-26 Effect of methanol administration on male Wistar rats on malondialdehyde concentration in the
lymphoid organs of experimental and control groups and the effect of methanol on
antioxidants in spleen 4-87
Table 5-1 Summary of studies considered most appropriate for use in derivation of an RfC 5-7
Table 5-2 The EPA PBPK model estimates of methanol blood levels (AUC) adjusted for background
(control) levels in rat dams following methanol inhalation exposures and reported mean brain
weights of 6-week-old male pups 5-12
Table 5-3 Maximum methanol blood levels (Cmax) adjusted for background, in mice following
inhalation exposures to methanol along with the corresponding incidence of extra cervical ribs
observed 5-14
Table 5-4 Summary of PODs for critical endpoints, application of UFs and conversion to candidate RfCs
using PBPK modeling 5-16
Table 5-5 Comparison of the lowest rodent and monkey methanol blood LOAELs (excluding
background) observed in developmental neurotoxicity studies 5-23
Table 5-6 Summary of PODs for critical endpoints, application of UFs and conversion to candidate RfDs
using PBPK modeling 5-29
Table 5-7 Summary of uncertainties in methanol noncancer assessment 5-31
VI
-------
LIST OF FIGURES
Figure 3-1 Methanol metabolism and key metabolic enzymes in primates and rodents 3-9
Figure 3-2 Folate-dependent formate metabolism. Tetrahydrofolate (THF)-mediated one carbon
metabolism is required for the synthesis of purines, thymidylate, and methionine 3-10
Figure 3-3 Plot of fetal (amniotic) versus maternal methanol concentrations in GD20 rats 3-13
Figure 3-4 Conceptus versus maternal blood AUC values for rats and mice 3-24
Figure 3-5 Schematic of the PBPK model used to describe the inhalation, oral, and i.v. route
pharmacokinetics of methanol 3-31
Figure 4-1 Exposure response array for noncancer effects reported in animals from repeat exposure and
developmental studies of methanol (Oral) 4-72
Figure 4-2 Exposure response array for noncancer effects reported in animals from repeat exposure and
developmental studies of methanol (Inhalation) 4-73
Figure 5-1 Fit of the Hill model to decreased mean brain weight in male rats at 6 weeks age using
estimated AUC of methanol in blood (adjusted for background) as the dose metric. The BMD
is estimated based on a BMR of one SD change from the control mean 5-13
Figure 5-2 Fit of the nested logistic (NLogistic) model- to the incidence of extra cervical rib in fetal mice
versus Cmax adjusted for background of methanol in blood from a GD6-GD15 inhalation study
in mice. The BMD is estimated based on a BMR of 0.05 extra risk 5-15
Figure 5-3 Projected impact of daily peak RfD and RfC exposures on sample background methanol blood
levels (mg MeOH/Liter [mg/L] blood) in humans 5-40
Figure 5-4 Relationship of monkey blood levels associated with effects of uncertain adversity with
projected impact of daily peak RfC and RfD exposures on sample background methanol blood
levels (mg MeOH/Liter [mg/L] blood) in humans 5-42
VII
-------
LIST OF ABBREVIATIONS AND ACRONYMS
ACGIH
ADH
ADH1
ADH3
AIC
ALD
ALDH2
ALT
ANOVA
AP
AST
ATP
ATSDR
AUC
P-NAG
Bav
BMD
BMD1SD
BMDL
BMDL1S
1SD
BMDS
BMR
BSD
BUN
BW,bw
Ci pool
C-section
CA
CAR
CASRN
CAT
CERHR
American Conference of Governmental and
Industrial Hygienists
alcohol dehydrogenase
alcohol dehydrogenase-1
formaldehyde dehydrogenase-3
Akaike Information Criterion
aldehyde dehydrogenase
mitochondrial aldehyde dehydrogenase-2
alanine aminotransferase
analysis of variance
alkaline phosphatase
aspartate aminotransferase
adenosine triphosphate
Agency for Toxic Substances and Disease
Registry
area under the curve, representing the
cumulative product of time and
concentration for a substance in the blood
N-acetyl-beta-D-glucosaminidase
oral bioavailability
benchmark dose(s)
BMD for response one standard deviation
from control mean
lower limit of a one-sided 95% confidence
interval on BMD (benchmark dose)
BMDL for response one standard deviation
from control mean
benchmark dose software
benchmark response
butathione sulfoximine
blood urea nitrogen
body weight
one carbon pool
peak concentration of a substance in the
blood during the exposure period
Cesarean section
chromosomal aberrations
conditioned avoidance response
Chemical Abstracts Service Registry
Number
catalase
Center for the Evaluation of Risks to
Human Reproduction at the NTP
CH3OH
CHL
CI
Cls
*~inax
CNS
C02
con-A
CR
CSF
Css
CT
CVB
CvBbg
CvBmb
CYP450
d,5,A
D2
DA
DIPE
DMDC
DNA
DNT
DOPAC
DPC
DTH
EFSA
EKG
EO
EPA
ERF
EtOH
F
Fo
Fi
F2
F344
FAD
FAS
FD
methanol
Chinese hamster lung (cells)
confidence interval
clearance rate
peak concentration
central nervous system
carbon dioxide
concanavalin-A
crown-rump length
Cancer slope factor
steady-state concentration
computed tomography
concentration in venous blood
background concentration in venous blood
concentration in venous blood minus
constant background
cytochrome P450
delta, difference, change
dopamine receptor
dopamine
diisopropyl ether
dimethyl dicarbonate
deoxyribonucleic acid
developmental neurotoxicity test(ing)
dihydroxyphenyl acetic acid
days past conception
delayed-type hypersensitivity
European Food Safety Authority
electrocardiogram
Executive Order
U.S. Environmental Protection Agency
European Ramazzini Foundation
ethanol
fractional bioavailability
parental generation
first generation
second generation
Fisher 344 rat strain
folic acid deficient
folic acid sufficient
formate dehydrogenase
VIM
-------
FP folate paired
FR folate reduced
FRACIN fraction inhaled
FS folate sufficient
FSH follicular stimulating hormone
y-GT gamma glutamyl transferase
g gravity
g, kg, mg, ug gram, kilogram, milligram, microgram
G6PD glucose-6-phosphate dehydrogenase
GAP43 growth-associated protein (neuronal growth
cone)
GD gestation day
GFR glomerular filtration rate
GI gastrointestinal track
GLM generalized linear model
GLP good laboratory practice
GSH glutathione
HAP hazardous air pollutant
HCHO formaldehyde
HCOO formate
Hct hematocrit
HEC human equivalent concentration
HED human equivalent dose
HEI Health Effects Institute
HERO Health and Environmental Research Online
(database system)
HH hereditary hemochromatosis
5-HIAA 5-hydroxyindolacetic acid
HMGSH S-hydroxymethylglutathione
Hp haptoglobin
HPA hypothalamus-pituitary-adrenal (axis)
HPLC high-performance liquid chromatography
HSDB Hazardous Substances Databank
HSP70 biomarker of cellular stress
5-HT serotonin
IL interleukins
i.p. intraperitoneal (injection)
IPCS International Programme on Chemical
Safety
IQ intelligence quotient
IRIS Integrated Risk Information System
IUR inhalation unit risk
i.v. intravenous (injection)
kj first-order urinary clearance
km
klv
KLH
KLL
Km
ksl
L, dL, mL
LD50
LDH
LH
LLF
LMI
LOAEL
M, mM, uM
MeOH
MLE
M-M
MN
MOA
4-MP
MRI
mRNA
MTBE
MTX
N2O/O2
NAD+
NADH
NET
NCEA
ND
NEDO
NIEHS
first-order urinary clearance scaling
constant; first order clearance of methanol
from the blood to the bladder for urinary
elimination
first order uptake from the intestine
first order methanol oral absorption rate
from stomach
rate constant for urinary excretion from
bladder
respiratory /cardiac depression constant
keyhole limpet hemocyanin
alternate first order rate constant
apparent Michaelis-Menten constant;
substrate concentration at half the
maximum velocity (Vmax)
first order transfer between stomach and
intestine
liter, deciliter, milliliter
median lethal dose
lactate dehydrogenase
luteinizing hormone
(maximum) log likelihood function
leukocyte migration inhibition (assay)
lowest-observed-adverse-effect level
molar, millimolar, micromolar
methanol
maximum likelihood estimate
Michaelis-Menten
micronuclei
mode of action
4-methylpyrazole (fomepizole)
magnetic resonance imaging
messenger RNA
methyl tertiary butyl ether
methotrexate
nitrous oxide
nicotinamide adenine dinucleotide
reduced form of nicotinamide adenine
dinucleotide
nitroblue tetrazolium (test)
National Center for Environmental
Assessment
not determined
New Energy Development Organization (of
Japan)
National Institute for Environmental Health
Sciences
IX
-------
NIOSH National Institute for Occupational Safety
and Health
nmol nanomole
NOAEL no-observed-adverse-effect level
NOEL no-observed-effect level
NP nonpregnant
NR not reported
NRC National Research Council
NS not specified
NTP National Toxicology Program at NIEHS
NZW New Zealand White (rabbit strain)
OR osmotic resistance
ORD Office of Research and Development
OSF oral slope factor
OU oculus uterque (each eye)
OXA oxazolone
P, p probability
PB blood:air partition coefficient
PBPK physiologically based pharmacokinetic
model
PC partition coefficient
PEG polyethylene glycol
PFC plaque-forming cell
PK pharmacokinetic
PMN polymorphonuclear leukocytes
PND postnatal day
POD point of departure
ppb, ppm parts per billion, parts per million
PR body :blood partition coefficent
PWG Pathology Working Group of the NTP of
NIEHS
Q wave the initial deflection of the QRS complex
QCC cardiac output scaling constant
QP pulmonary (alveolar) ventilation
QRS portion of electrocardiogram corresponding
to the depolarization of ventricular cardiac
cells.
R2 square of the correlation coefficient, a
measure of the reliability of a linear
relationship.
RBC red blood cell
RfC reference concentration
RfD reference dose
RNA ribonucleic acid
R0bg zero-order endogenous production rate
ROS reactive oxygen species
S9 microsomal fraction from liver
SAP serum alkaline phosphatase
s.c. subcutaneous
SCE sister chromatid exchange
S-D Sprague-Dawley rat strain
SD standard deviation
S.E. standard error
SEM standard error of mean
SGPT serum glutamate pyruvate transaminase
SHE Syrian hamster embryo
SOD superoxide dismutase
SOP standard operating procedure(s)
T; T,/2, t/2 time; half-life
T wave the next deflection in the electrocardiogram
after the QRS complex; represents
ventricular repolarization
TAME tertiary amyl methyl ether
TAS total antioxidant status
Tau taurine
THF tetrahydrofolate
TLV threshold limit value
TNFa tumor necrosis factor-alpha
TNP-LPS trinitrophenyl-lipopolysaccharide
TRI Toxic Release Inventory
U83836E vitamin E derivative
UF(s) uncertainty factor(s)
UFA UF associated with interspecies (animal to
human) extrapolation
UFD UF associated with deficiencies in the
toxicity database
UFH UF associated with variation in sensitivity
within the human population
UFS UF associated with subchronic to chronic
exposure
Vd volume of distribution
Vmax pseudo-maximal velocity of metabolism
VmaxC multiplier for allometric scaling of Vmax
VDR visually directed reaching test
VitC vitamin C
VPR ventilation perfusion ratio
v/v volume of solute/volume of solution
VYS visceral yolk sac
WBC white blood cell
WOE weight of evidence
w/v weight (mass of solute)/volume of solution
X2 chi square
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Assessment Team
Jeffrey Gift, Ph.D. (Chemical Manager)
J. Allen Davis, MSPH
U.S. EPA/ORD/NCEA
Research Triangle Park, NC
Paul Schlosser, Ph.D.
U.S. EPA/ORD/NCEA
Washington, DC
Scientific Support Team
Jane Caldwell, Ph.D.
*J. Michael Davis, Ph.D.
Robert Dewoskin, Ph.D., DABT
* Angela Howard, Ph.D.
Jennifer Jinot, Ph.D.
Eva McLanahan, Ph.D.
Connie Meacham, M.S.
Reeder Sams, II, Ph.D.
U.S. EPA/ORD/NCEA
Research Triangle Park, NC
*Stanley S. Barone, Jr., Ph.D.
TedBerner, M.S.
*Chad Thompson, Ph.D., MBA
Paul D. White, Ph.D.
U.S. EPA/ORD/NCEA
Washington, DC
Marina Evans, Ph.D.
John Rogers, Ph.D.
U.S. EPA/ORD/NHEERL
Research Triangle Park, NC
*Hugh Barton, Ph.D.
U.S. EPA/ORD/NCCT
Research Triangle Park, NC
*Formerly at EPA-NCEA.
XVII
-------
Production Team
* Ellen Lorang, M.S.
*Deborah Wales
Richard N. Wilson
U.S. EPA/ORD/NCEA
Research Triangle Park, NC
*J. Sawyer Lucy
U.S. EPA/ORD/NCEA
Student Services Contractor
Research Triangle Park, NC
Kenneth J. Breito
*Mark Greenberg, Ph.D.
Barbara Wright
U.S. EPA/ORD/NCEA
Senior Environmental Employment
Program
*Formerly at EPA-NCEA.
Contractor Support
Susan Goldhaber, M.S.
Frank Stack
Errol Zeiger, Ph.D.
Alpha-Gamma Technologies, Inc.
Torka Poet, Ph.D.
Justin TeeGuarden, Ph.D.
Battelle, Pacific Northwest National
Laboratories
Bruce C. Allen, M.S.
Bruce Allen Consulting
Robinan Gentry, M.S.
ENVIRON International
George Holdsworth, Ph.D.
Lisa Lowe, Ph.D.
Kan Shao, Ph.D.
Lutz Weber, Ph.D., DABT
Oak Ridge Institute for Science and
Education
Annette lannucci, M.S.
Sciences International, Inc.
XVIII
-------
Executive Direction
Kenneth Olden, Ph.D., Sc.D., L.H.D. U.S. EPA/ORD/NCEA
T-V i TT? i i A ,ro Washington, DC
Debra Walsh, M.S. 6
Lynn Flowers, Ph.D., DABT
Vincent Cogliano, Ph.D.
Samantha Jones, Ph.D.
T v, w A u our, U.S. EPA/ORD/NCEA
John Vandenberg, Ph.D. „ , _ . , „ . -.-,„
0 Research mangle Park, NC
Ila Cote, Ph.D., DABT
Reeder Sams, Ph.D.
Lyle Burgoon, Ph.D.
Reviewers
The methanol (noncancer) assessment was provided for review to scientists in EPA's Program and
Region Offices. Comments were submitted by:
Office of Air Quality and Planning Standards, Research Triangle Park, NC
Office of Children's Health Protection, Washington, DC
Office of Policy, Economics, and Innovation, Washington, DC
Office of Solid Waste and Emergency Response, Washington, DC
Office of Water, Washington, DC
The methanol (noncancer) assessment was provided for review to other federal agencies and the
Executive Office of the President. Comments were submitted by:
Agency for Toxic Substances Disease Registry, Centers for Disease Control and Prevention,
Department of Health & Human Services
Council on Environmental Quality, Executive Office of the President
National Institute for Occupational Safety and Health, Centers for Disease Control and
Prevention, Department of Health & Human Services
Office of Management and Budget, Executive Office of the President
United States Department of Defense
XIX
-------
The methanol (noncancer) assessment was released for public comment in April 2011 and a revised
assessment was released for public comment in May 2013. A summary and EPA's disposition of the
comments from the public is included in Appendix A. Comments were received from the following
entities:
Kimberly Wise, Ph.D. American Chemistry Council
Paul Noe American Forest & Paper Association
Robert Glowinski American Wood Council
Patrick Beany, Ph.D., DABT American Petroleum Institute
Greg Dolan Methanol Institute
Steve Howell National Biodiesel Board
Andrew G. Salmon M.A., D.Phil.* Private Citizen
Lisa M. Sweeney, Ph.D., DABT* Private Citizen
George Cruzan, Ph.D., DABT ToxWorks
*Members of the 2011 peer review panel, who also provided public comments on the 2013 revised draft of the
methanol (noncancer) toxicological review.
The methanol (noncancer) assessment was peer reviewed by independent expert scientists external to
EPA and a peer-review meeting was held on July 22, 2011. A follow-up peer review was completed in
July 2013 to obtain feedback from members of the original 2011 peer review panel (identified with an
asterisk below) on the 2013 revised draft methanol (noncancer) toxicological review and EPA's response
to the 2011 peer review comments. The original and follow-up external peer-review comments are
available on the IMS Web site. A summary and EPA's disposition of the comments received from the
independent external peer reviewers is included in Appendix A.
Stephen Roberts, Ph.D. (Chair)** University of Florida
Gainesville, FL
Janusz Z. Byczkowski, Ph.D.** Independent Consultant
Fairborn, OH
Thomas M. Burbacher, Ph.D. University of Washington
Seattle, WA
David C. Dorman, Ph.D. North Carolina State University-College of
Veterinary Medicine
Raleigh, NC
Kenneth McMartin, PhD.** Louisiana State University Health Sciences Center
Shreveport, LA
Andrew Salmon, Ph.D. California EPA- Office of Environmental Health
Hazard Assessment
Lafayette, CA
Lisa M. Sweeney, Ph.D. Henry M. Jackson Foundation for the Advancement of
Military Medicine, Naval Medical Research Unit-Dayton
Kettering, OH
**Members of the original 2011 peer review panel, who also reviewed the 2013 revised draft of the
methanol (noncancer) toxicological review.
xx
-------
EXECUTIVE SUMMARY
Introduction
Methanol is a high production volume chemical with many commercial uses. It is a basic
building block for numerous chemicals. Many of its derivatives are used in the construction,
housing or automotive industries. Consumer products that contain methanol include varnishes,
shellacs, paints, windshield washer fluid, antifreeze, adhesives, and deicers.
Methanol can be formed in the mammalian organism as a metabolic byproduct.
Endogenous background levels [naturally generated from within the body] are not the same as
exogenous exposure (exposure from a source outside the body), but the combination of
endogenous background levels of methanol plus exogenous methanol exposure can lead to
toxicity. Diet can contribute to background levels of methanol, principally from the ordinary
ingestion of fruits and vegetables. This Toxicological Review provides scientific support and
rationale for a hazard identification and dose-response assessment of the noncancer effects
associated with chronic exposures to exogenous sources of methanol that add to background
levels of methanol. For the purpose of this methanol (noncancer) assessment, EPA estimates that
a diet that includes fruits and vegetables would not increase methanol blood levels above 2.5
mg/L (see discussion in Section 5.3.6). Thus, for a population with background blood levels of
methanol at or below 2.5 mg/L, the inhalation reference concentration (RfC) and oral reference
dose (RfD) that are derived in this assessment represent estimates (with uncertainty spanning
perhaps an order of magnitude) of daily exposures to the human population (including sensitive
subgroups) that are likely to be without an appreciable risk of deleterious effects during a
lifetime. In Section 5 (Dose Response Assessments), the basis for a RfC of 2x 101 mg/m3 and a
RfD of 2 mg/kg-day are described.
This health assessment does not assess the potential carcinogenicity of methanol, or the
health effects associated with background levels of methanol that arise from metabolic and
dietary sources such as vegetables, fruits and juices that naturally contain methanol or have
components (e.g., plant pectin) that convert to methanol. Hence, as discussed in Section 3.4.3.2
(Model Structure), responses observed in oral and inhalation studies of laboratory animals
exposed to methanol are evaluated against blood concentrations of methanol after subtracting an
estimate of the background blood levels in control animals.
XXI
-------
Chemical and Physical Information
Methanol is the smallest member of the family of aliphatic alcohols. Also known as
methyl alcohol or wood alcohol, among other synonyms, it is a clear, colorless, very volatile, and
flammable liquid. Methanol is widely used as a solvent in many commercial and consumer
products. It is freely miscible with water and other short-chain aliphatic alcohols but has little
tendency to distribute into lipophilic media.
Toxicokinetics
Due to its very low oil:water partition coefficient, methanol is taken up efficiently by the
lung or the intestinal tract and distributes freely in body water (blood volume, extracellular and
intracellular fluid, etc.) without any tendency to accumulate in fatty tissues. Methanol can be
metabolized completely to CO2, but may also, as a regular byproduct of metabolism, enter the
formic acid Ci-pool (1-carbon unit pool), and become incorporated into biomolecules. Animal
studies indicate that blood methanol levels increase with the breathing rate and that metabolism
becomes saturated at high exposure levels. Because of its volatility methanol can be exhaled with
air, and also excreted unchanged via urine. As discussed in Section 3.1 (Toxicokinetics
Overview), the enzymes responsible for metabolizing methanol are different in rodents and
primates (Figure 3-1). Several published rat, mouse, and human PBPK models which attempt to
account for these species differences are described in Section 3.4.2 (Methanol PBPK Models).
The development of methanol PBPK models was organized around a set of criteria,
described in Section 3.4.1.2 (Criteria for the Development of Methanol PBPK Models), that take
into account the dose routes used in key toxicity studies, the availability of pharmacokinetic
information necessary for PBPK model development and the most likely toxicological mode of
action (MOA). Specifically, EPA developed new PBPK models or modified the existing ones,
which allowed for the estimation of monkey and rat internal dose metrics. A human model was
also developed to extrapolate those internal metrics to inhalation and oral exposure
concentrations that would result in the same internal dose in humans (human equivalent
concentrations [HECs] and human equivalent doses [HEDs]). The procedures used for the
development, calibration and use of these EPA models are summarized in Section 3.4
(Physiologically Based Pharmacokinetic Models), with further details provided in Appendix B,
"Development, Calibration and Application of a Methanol PBPK Model."
Developmental malformations and anomalies in gestationally exposed fetal mice (and
developmental neurotoxicity, as indicated by reduced absolute brain weight, in gestationally and
lactationally exposed fetal and neonate rats) observed in inhalation studies are sensitive
endpoints considered in the derivation of an RfC. However, questions remain regarding the
XXII
-------
relative involvement of parent methanol, formaldehyde, and reactive oxygen species (ROS) in
the MOA for these developmental effects. Given the reactivity of formaldehyde and the lack of
relevant pharmacokinetic information, PBPK models that predict levels of formaldehyde (or
subsequent metabolites of formaldehyde) in the blood would be difficult to validate.1 However,
the high reactivity of formaldehyde (see Section 3.1 [Toxicokinetics Overview]) would limit its
unbound and unaltered transport as free formaldehyde from maternal to fetal blood (see
discussion in Section 3.4.1.1 [MOA and Selection of a Dose Metric] and 4.7.1 [Role of Methanol
and Metabolites in the Developmental Toxicity of Methanol]), and the ROS MOA requires the
presence of methanol to alter embryonic catalase activity. Hence, it is likely that all of these
MO As require methanol to be present at the target site. For this reason, and because adequate
pharmacokinetic information was available, PBPK models that estimate levels of parent
methanol in blood were developed and validated for rats and humans. Because actual measured
internal blood methanol levels suitable for use as estimates of peak concentrations (Cmax) in mice
were provided in the Rogers et al. (1993b) study, and these data were considered better than a
predictive model, the mouse PBPK model was not used or discussed in detail in this
toxicological review. A simple PK model for monkey methanol kinetics was also developed and
used to evaluate the results of monkey developmental studies (Burbacher et al., 2004a: 2004b:
1999a:
A pregnancy-specific PBPK model does not exist for methanol and limited data exist for
the development and validation of a fetal/gestational/conceptus compartment. For this reason,
and because levels of methanol in non-pregnant and pregnant adult females, and fetal blood (all
measures of maternal exposure) are expected to be similar following the same oral or inhalation
methanol exposure (see discussion in Section 3.4.1.2 [Criteria for the development of Methanol
PBPK Models]), EPA developed and used non-pregnancy models for the appropriate species and
routes of exposure for the derivation of candidate RfCs and RfDs. It is recognized that these
models may not accurately represent neonate blood levels following the gestation, lactation and
inhalation exposure regimen used in one of the key rat studies (NEDO, 1987), but they are
considered appropriate for use in deriving HEC values from this study assuming the ratio of
maternal to offspring blood methanol would be similar in rats and humans (see discussion in
Sections 5.1.3.2.2 [Animal-to-Human Extrapolation UFA]).
The rat and human methanol PBPK models fit multiple data sets for inhalation, oral, and
i.v. exposures, from multiple research groups using consistent parameters that are representative
of each species but are not varied within species or by dose or source of data. Also, a simple PK
1 The PBPK models developed by EPA estimate total amount of methanol cleared by metabolic processes, but this
has limited value as a metric of formaldehyde or formate dose since it ignores metabolic processes that may differ
between species and between the mother and the fetus/neonate.
XXIII
-------
model calibrated to non-pregnant (NP) monkey data, which were shown to be essentially
indistinguishable from pregnant monkey PK data, was used to estimate blood methanol area
under the curve (AUC) values (internal doses) in that species. In the case of the mouse, a PK
model developed from in vivo blood methanol levels in (Rogers et al., (1993b) resulted in more
reliable estimates compared to the PBPK model and was used for derivation of effect levels in
this species. Section 5 (Dose Response Assessments and Characterization) describes how the
human PBPK model was used in the derivation of candidate RfCs and RfDs.
Hazard Identification
In humans, acute central nervous system (CNS) toxicity can result from relatively low
ingested doses (as low as 3-20 mL of methanol), which can metabolize to formic acid and lead to
metabolic acidosis. The resulting acidosis can potentially cause lasting nervous system effects
such as blindness, Parkinson-like symptoms, and cognitive impairment. These effects have been
observed in humans with blood methanol levels as low as 200 mg/L (Adanir et al., 2005).
CNS effects have not been observed in rodents following acute exposures to methanol,
and NEDO (1987) reported that methanol blood levels around 5,000 mg/L were necessary to
cause clinical signs and CNS changes in cynomolgus monkeys. The species differences in
toxicity from acute exposures appear to be the result of a limited ability of humans to metabolize
formic acid.
Occupational studies and case reports offer valuable information on the effects of
methanol following acute human exposures, but the relatively small amount of data for
subchronic, chronic, or in utero human exposures are inconclusive. However, a number of
reproductive, developmental, subchronic, and chronic toxicity studies have been conducted in
mice, rats, and monkeys.
Data regarding effects from oral exposure in experimental animals exist, but they are
more limited than data from the inhalation route of exposure (see Sections 4.2 [Acute,
Subchronic, and Chronic Studies in Animals - Oral and Inhalation], 4.3 [Reproductive and
Developmental Studies - Oral and Inhalation], and 4.4 [Neurotoxicity]). Two oral studies in rats
(Soffritti et al., 2002; TRL, 1986), one oral study in mice (Apaja, 1980) and several inhalation
studies in monkeys, rats and mice (NEDO, 1987, 1985a, b) of 90-days duration or longer have
been reported. Some noncancer effects of methanol exposure were noted in these studies,
principally in the liver and brain tissues, but they occurred at relatively high doses.
A number of studies have used the inhalation route of exposure to assess the potential of
reproductive or developmental toxicity of methanol in mice, rats, and monkeys (see Section 4.3.2
[Inhalation Reproductive and Developmental Studies]). These studies indicate that fetal and
XXIV
-------
neonate toxicity occurs at lower doses than maternal toxicity. At exposure concentrations of
5,000 ppm or above, methanol has been shown to cause an increase in litters with resorptions
(Bolon et al., 1993), and severe malformations (exencephaly and cleft palate) in mice, the most
sensitive gestational days being GD6 and GD7 (i.e., early organogenesis) (Rogers and Mole,
1997; Rogers et al., 1993a: Rogers et al., 1993b). Increased occurrences of ossification
disturbances and skeletal anomalies were observed at exposure concentrations of 2,000 ppm in
mice (Rogers et al., 1993b) and at 10,000 ppm in rats (Nelson et al., 1985). NEDO (1987)
conducted a series of developmental and reproductive studies, including a two generation and a
follow up one generation reproductive toxicity study in rats, which used exposure times of 20
hours/day or more at concentrations between 100 and 5,000 ppm. Details were not reported (e.g.,
means, variances, sample sizes, pup-to-litter correlations) that would allow for an analysis of the
findings from this study. However, a follow-up one-generation study conducted by NEDO (1987)
contained enough information to confirm and quantify the primary endpoint identified, pup brain
weight changes. This developmental neurotoxicity study is discussed in Section 4.4.2 (Inhalation
Neurotoxicity Studies). Section 4.4.2 also describes another key developmental neurotoxicity
study conducted in pregnant cynomolgus monkeys exposed to 200-1,800 ppm methanol for 2.5
hours/day throughout pre-mating, mating, and gestation (Burbacher et al., 2004a: 2004b: 1999a:
1999b). Potential compound-related effects noted were a shortening of the gestation period by
less than 5%, and developmental neurotoxicity (particularly delayed sensorimotor development)
in the monkeys.
As discussed in Section 4.6.1.2 (Key Studies, Inhalation), due largely to the lack of clear
dose-response information, the data from the monkey developmental study are not conclusive,
and there was insufficient evidence to determine if the primate fetus is more sensitive, or less
sensitive, than rodents to the developmental or reproductive effects of methanol. Taken together,
however, the NEDO (1987) rat study and the Burbacher et al. (2004a: 2004b: 1999a: 1999b)
monkey study suggest that prenatal exposure to methanol can result in adverse effects on
developmental neurology pathology and function, which can be exacerbated by continued
postnatal exposure. Among an array of findings indicating developmental neurotoxicity and
developmental malformations and anomalies that have been observed in rodents, a decrease in
the brain weights of gestationally and lactationally exposed neonatal rats (NEDO, 1987) and an
increase in the incidence of cervical ribs of gestationally exposed fetal mice (Rogers et al.,
1993b) are considered the most robust endpoints for the purposes of RfD and RfC derivation.
See Section 4.6 (Synthesis of Major Noncancer Effects) for a more extensive summary of the
dose-related effects that have been observed following subchronic or chronic exposure.
Sections 4.7 (Noncancer MOA Information) and 5.3.5 (Choice of Species/Sex), provide a
discussion of the uncertainty regarding human relevance of the mouse and rat developmental
xxv
-------
studies due to differences in the way humans and rodents metabolize methanol. Adult humans
metabolize methanol principally via alcohol dehydrogenase (ADH1) and rodents via catalase and
ADH1. Recent studies in mice have demonstrated that high catalase activity can reduce, and low
catalase activity can enhance, methanol's embryotoxic effects. However, the MO A for these
effects, and the role of catalase, have not been determined. Further, while catalase does not
appear to be involved in adult human methanol metabolism, less is known about the metabolism
of methanol in human infants (see Section 3.3 [Human Variability in Methanol Metabolism]).
Thus, the effects observed in rodents are considered relevant for the assessment of human health.
Dose-Response Assessment and Characterization
As discussed above and in Section 5.1.1 (Choice of Principal Study and Critical
Effect[s]), reproductive and developmental effects are considered the most sensitive and
quantifiable effects reported in studies of methanol. Because the oral reproductive and
developmental studies employed single and comparatively high doses (i.e., oral versus
inhalation), the developmental effects observed in the inhalation studies were used to derive the
RfC and, using a route-to-route extrapolation, the RfD.
Clearly defined toxic endpoints at moderate exposure levels have been observed in
inhalation studies of reproductive and developmental toxicity (see Section 5.1.1.2 [Selection of
Critical Effect[s]). Three endpoints from inhalation developmental toxicity studies were critically
evaluated for derivation of the RfC: (1) increased occurrences of ossification disturbances and
skeletal abnormalities (i.e., formation of cervical ribs) in CD-I mice exposed to methanol during
organogenesis (Rogers et al., 1993b): (2) reduced brain weights in rats exposed to methanol from
early gestation through 8 weeks of postnatal life (NEDO, 1987); and (3) deficits in sensorimotor
development in the offspring of monkeys exposed to methanol throughout gestation (Burbacher
et al.. 2004a: 2004b: 1999a:
Rogers et al. (1993b) exposed CD-I mice to air concentrations of 0, 1,000, 2,000, and
5,000 ppm methanol for 7 hours/day on GD7 to GDI7. A lower limit of a one-sided 95%
confidence interval on the BMD (BMDL) of 43 mg/L was estimated for the internal peak blood
methanol (Cmax) associated with 5% extra risk for the formation of cervical ribs (see Section
5.1.2.3 [BMD Approach Applied to Cervical Rib Data in Mice] and Appendix D [RfC Derivation
Options]). This BMDL05 was then divided by 100 to account for uncertainties associated with
human variability (UFH), the animal-to-human extrapolation (UFA) and the database (UFD), and
to reduce it to a level that is within the range of blood levels for which the human PBPK model
was calibrated (see discussion in Section 5.1.3.2 [Application of UFs]). The PBPK model was
then used to convert this adjusted internal BMDL0s of 0.43 mg/L to a human equivalent
XXVI
-------
candidate RfC of 20.0 mg/m3 (see Section 5.1.3 [RfC Derivation - Including Application of
Uncertainty Factors]) and a candidate RfD of 1.9 mg/kg-day (see Section 5.2.2 [RfD Derivation
- Including Application of Uncertainty Factors]).
NEDO (1987) exposed fetal Sprague-Dawley rats and their dams to air concentrations of
0, 500, 1,000 and 2,000 ppm methanol from the first day of gestation (GDI) until 8 weeks of
age, and brain weights were determined at 3, 6, and 8 weeks of age. ABMDL of 858 mg-hr/L
was estimated for the area under the curve (AUC) internal blood methanol dose, associated with
a brain weight reduction at 6 weeks equal to one standard deviation (SD) from the control mean
(see Section 5.1.2.2 [BMD Approach Applied to Brain Weight Data in Rats], and Appendix D
[RfC Derivation Options]). This BMDLiso was then divided by 100 to account for uncertainties
associated with human variability (UFH), the animal-to-human extrapolation (UFA) and the
database (UFD), and to reduce it to a level that is within the range of blood levels for which the
human PBPK model was calibrated (see discussion in Section 5.1.3.2 [Application of UFs]). The
PBPK model was then used to convert this adjusted internal BMDLiso of 8.58 mg-hr/L to a
human equivalent candidate RfC of 17.8 mg/m3 (see Section 5.1.3 [RfC Derivation - Including
Application of Uncertainty Factors]) and a candidate RfD of 5.2 mg/kg-day (see Section 5.2.2
[RfD Derivation - Including Application of Uncertainty Factors]).
Burbacher et al. (2004a; 2004b: 1999a: 1999b) exposed M. fascicularis monkeys to
0, 200, 600, or 1,800 ppm methanol 2.5 hours/day, 7 days/week during pre-mating/mating and
throughout gestation (approximately 168 days). ABMDLso of 19.6 mg/L was estimated for the
blood methanol Cmax associated with a one SD delay in sensorimotor development in the
offspring as measured by a visually directed reaching (VDR) test (see Appendix D [RfC
Derivation Options]). However, only the unadjusted VDR response for females exhibited a
response that could be modeled and the dose-response was marginally significant, with only the
high dose exhibiting a response significantly different from controls. Although, the metabolism
of methanol in monkeys is comparable to humans (Section 3.1 [Toxicokinetics Overview]) and a
delay in VDR is a potentially relevant CNS effect (Section 4.4.2 [Inhalation Neurotoxicity
Studies]), EPA concluded that the use of this data for RfC/D derivation was not preferable, given
the availability of more reliable dose-response data from the Rogers et al. (1993b) and NEDO
(1987) rodent studies.
In summary, after the evaluation of different species, different endpoints, different
protocols and different data sources, the Rogers et al. (1993b) mouse, NEDO (1987) rat, and
Burbacher et al. (2004a; 2004b: 1999a: 1999b) monkey studies exhibited developmental effects
at similar doses, providing consistent results. As described in Sections 5.1.1.2 (Selection of
Critical Effects) and 5.2.1.1 (Expansion of the Oral Database by Route-to-Route Extrapolation),
because the Rogers et al. (1993b) and NEDO (1987) studies identified relevant effects in relevant
XXVII
-------
species that could be adequately quantified in a dose-response analysis, they are considered the
most appropriate studies for use in the RfC and RfD derivation. The candidate RfC of 2x 101
mg/m3 based on decreased brain weight observed in the NEDO (1987) rat developmental study
(see Table 5-4 [Summary of POD values for critical endpoints, application of UFs and
conversion to candidate RfCs using PBPK modeling]) was selected as the RfC for methanol. The
candidate RfD of 2 mg/kg-day based on the formation of extra cervical ribs observed in the
Rogers et al. (1993b) mouse developmental study (see Table 5-6 [Summary of POD values for
critical endpoints, application of UFs and conversion to candidate RfDs using PBPK modeling])
was selected as the RfD for methanol. As described in Sections 5.1.3 (RfC Derivation -
Including Application of Uncertainty Factors) and 5.2.2 (RfD Derivation - Including Application
of Uncertainty Factors), the UFs employed for both the RfC and RfD derivations include a UFH
of 10 for intraspecies variability, a UFA of 3 to address pharmacodynamic uncertainty and a UFD
of 3 for database uncertainty.
Relationship of the RfC and RfD to Background Methanol
Blood Levels and Monkey Blood Levels Associated with
Effects of Uncertain Adversity
In Section 5.3.6, PBPK model predictions for the increase in methanol levels in blood
resulting from exposure to methanol at the level of the RfC or RfD are compared to background
blood levels of methanol estimated from (1) daily endogenous production and dietary exposure
estimates from the U.K. report (COT, 2011) and (2) a sample background distribution derived
from relevant study groups in Table 3-1 of this toxicological review. Both the EPA and the U.K.
data are consistent with approximately 2.5 mg/L representing a high end of the range of
background (as defined in Section 5.3.6) methanol blood levels. EPA estimates that the shift in
EPAs sample background methanol blood level distribution that would be associated with daily
exposures of the entire population to methanol at the RfC or the RfD would increase the number
of individuals with peak methanol blood levels at or above 2.5 mg/L from -7% to -14%. EPAs
PBPK model predicts that a continuous daily methanol exposure at the RfD or RfC would raise
the peak methanol blood level of an individual with a high end background methanol blood level
of 2.5 mg/L to just under 3 mg/L. As discussed in Section 5.3.7, this 3 mg/L methanol blood
level is at the low end of the range of methanol blood levels that have been reported in monkey
chronic and gestational exposure studies to be associated with CNS and
reproductive/developmental effects of uncertain, but potential adversity.
XXVIII
-------
1.INTRODUCTION
This document presents background information and justification for the Integrated Risk
Information System (IRIS) Summary of the hazard and dose-response assessment of methanol.
IRIS Summaries may include oral reference dose (RfD) and inhalation reference concentration
(RfC) values for chronic and other exposure durations, and a carcinogenicity assessment.
The RfD and RfC, if derived, provide quantitative information for use in risk assessments
for noncancer health effects known or assumed to be produced through a nonlinear (presumed
threshold) mode of action (MOA). The RfD (expressed in units of milligrams per kilogram per
day [mg/kg-day]) is defined as an estimate (with uncertainty spanning perhaps an order of
magnitude) of a daily exposure to the human population (including sensitive subgroups) that is
likely to be without an appreciable risk of deleterious effects during a lifetime. The inhalation
RfC (expressed in units of milligrams per cubic meter [mg/m3]) is analogous to the oral RfD but
provides a continuous inhalation exposure estimate. The inhalation RfC considers toxic effects
for both the respiratory system (portal-of-entry) and for effects peripheral to the respiratory
system (extrarespiratory or systemic effects). Reference values are generally derived for chronic
exposures (up to a lifetime), but may also be derived for acute (< 24 hours), short-term
(>24 hours up to 30 days), and subchronic (>30 days up to 10% of lifetime) exposure durations,
all of which are derived based on an assumption of continuous exposure throughout the duration
specified. Unless specified otherwise, the RfD and RfC are derived for chronic exposure
duration.
Development of these hazard identification and dose-response assessments for the
noncancer effects of methanol has followed the general guidelines for risk assessment as set forth
by the National Research Council (NRC) (1983). EPA Guidelines and Risk Assessment Forum
Technical Panel Reports that may have been used in the development of this assessment include
the following: Guidelines for the Health Risk Assessment of Chemical Mixtures (U.S. EPA,
1986b), Guidelines for Mutagenicity Risk Assessment (U.S. EPA, 1986a), Recommendations for
and Documentation of Biological Values for Use in Risk Assessment (U.S. EPA, 1988),
Guidelines for Developmental Toxicity Risk Assessment (U.S. EPA, 1991), Interim Policy for
Particle Size and Limit Concentration Issues in Inhalation Toxicity Studies (U.S. EPA, 1994a),
Methods for Derivation of Inhalation Reference Concentrations and Application of Inhalation
Dosimetry (U.S. EPA, 1994b), Use of the Benchmark Dose Approach in Health Risk Assessment
(U.S. EPA, 1995), Guidelines for Reproductive Toxicity Risk Assessment (U.S. EPA, 1996),
Guidelines for Neurotoxicity Risk Assessment (U.S. EPA, 1998a), Science Policy Council
Handbook: Risk Characterization (U.S. EPA, 2000a), Supplementary Guidance for Conducting
1-1
-------
Health Risk Assessment of Chemical Mixtures (U.S. EPA, 2000b), A Review of the Reference
Dose and Reference Concentration Processes (U.S. EPA, 2002), Guidelines for Carcinogen Risk
Assessment (U.S. EPA, 2005a), Supplemental Guidance for Assessing Susceptibility from
Early-Life Exposure to Carcinogens (U.S. EPA, 2005b), Science Policy Council Handbook: Peer
Review (U.S. EPA, 2006b), and A Framework for Assessing Health Risks of Environmental
Exposures to Children (U.S. EPA, 2006a), Recommended Use of Body Weight % as the Default
Method in Derivation of the Oral Reference Dose (U.S. EPA, 20 lib), and benchmark Dose
Technical Guidance Document (U.S. EPA, 2012a).
Primary, peer-reviewed literature identified through January 2013 was included where
that literature was determined to be relevant to the assessment. The relevant literature included
publications on methanol that were identified through Toxicology Literature Online (TOXLINE),
PubMed, the Toxic Substance Control Act Test Submission Database (TSCATS), the Registry of
Toxic Effects of Chemical Substances (RTECS), the Chemical Carcinogenesis Research
Information System (CCRIS), the Developmental and Reproductive Toxicology/Environmental
Teratology Information Center (DART/ETIC), the Hazardous Substances Data Bank (HSDB),
the Genetic Toxicology Data Bank (GENE-TOX), Chemical abstracts, and Current Contents.
Other peer-reviewed information, including health assessments developed by other
organizations, review articles, and independent analyses of the health effects data were retrieved
and included in the assessment where appropriate. Studies that had not been peer-reviewed and
were potentially critical to the conclusions of the assessment were separately and independently
peer-reviewed. Any pertinent scientific information submitted by the public to the IRIS
Submission Desk or by reviewers during internal and external peer reviews was also considered
in the development of this document. It should be noted that references added to the
Toxicological Review after the external peer review in response to peer reviewer's comments
have not changed the overall qualitative and quantitative conclusions.
An initial keyword search was based on the Chemical Abstracts Service Registry Number
(CASRN) and several common names for methanol. The subsequent search strategy focused on
the toxicology and toxicokinetics of methanol, particularly as they pertain to target tissues,
effects at low doses, different developmental stages, sensitive subpopulations, and background
levels from endogenous and exogenous sources. A more targeted search was completed for the
construction and parameterization of a methanol physiologically-based pharmacokinetic (PBPK)
model. The focus of this targeted search included existing PBPK models for primary alcohols
and pharmacokinetic information for major metabolites and related enzymes. Both the general
and targeted searches identified a multitude of studies that used methanol for laboratory
procedures. Exclusion terms such as 'extract of methanol' were used in order to cull such
irrelevant studies. The literature keyword searches are narrowed down further by manual review.
1-2
-------
Selection of studies for inclusion in the Toxicological Review was based on consideration of the
extent to which the study was informative and relevant to the assessment and general study
quality considerations. In general, the relevance of health effect studies was evaluated as outlined
in EPA guidance [A Review of the Reference Dose and Reference Concentration Processes (U.S.
EPA, 2002) and Methods for Derivation of Inhalation Reference Concentrations and Application
of Inhaled Dosimetry (U.S. EPA, 1994b)]. All animal studies of methanol involving repeated
oral, inhalation, or dermal exposure that were considered to be of acceptable quality, whether
yielding positive, negative, or null results, were considered in assessing the evidence for health
effects associated with chronic exposure to methanol. In addition, animal toxicity studies
involving short-term duration and other routes of exposure were evaluated to inform conclusions
about health hazards. The references considered and cited in this document, including
bibliographic information and abstracts, can be found on the Health and Environmental Research
Online (HERO) website.2
On December 23, 2011, The Consolidated Appropriations Act, 2012, was signed into law
(U.S. Congress, 2011). The report language included direction to EPA for the IRIS Program
related to recommendations provided by the National Research Council (NRC) in their review of
EPAs draft IRIS assessment of formaldehyde (NRC, 2011). The NRC's recommendations,
provided in Chapter 7 of their review report, offered suggestions to EPA for improving the
development of IRIS assessments. The report language included the following:
The Agency shall incorporate, as appropriate, based on chemical-specific datasets
and biological effects, the recommendations of Chapter 7 of the National
Research Council's Review of the Environmental Protection Agency's Draft IRIS
Assessment of Formaldehyde into the IRIS process .... For draft assessments
released in fiscal year 2012, the Agency shall include documentation describing
how the Chapter 7 recommendations of the National Academy of Sciences (NAS)
have been implemented or addressed, including an explanation for why certain
recommendations were not incorporated.
The NRC's recommendations, provided in Chapter 7 of the review report (NRC, 2011),
offered suggestions to EPA for improving the development of IRIS assessments. Consistent with
the direction provided by Congress, documentation of how the recommendations from Chapter 7
of the NRC report have been implemented in this assessment is provided in the tables in
Appendix E. Where necessary, the documentation includes an explanation for why certain
recommendations were not incorporated. The IRIS Program's implementation of the NRC
2HERO is a database of scientific studies and other references used to develop EPA's risk assessments aimed at
understanding the health and environmental effects of pollutants and chemicals. It is developed and managed in
EPA's Office of Research and Development (ORD) by the National Center for Environmental Assessment (NCEA).
The database includes more than 700,000 scientific articles from the peer-reviewed literature. New studies are added
continuously to HERO.
1-3
-------
recommendations is following a phased approach that is consistent with the NRC's "Roadmap
for Revision" as described in Chapter 7 of the formaldehyde review report (NRC, 2011). The
NRC stated that, "the committee recognizes that the changes suggested would involve a multi-
year process and extensive effort by the staff at the National Center for Environmental
Assessment and input and review by the EPA Science Advisory Board and others." The IRIS
methanol (noncancer) assessment is in Phase 1 of implementation, which focuses on a subset of
the short-term recommendations, such as editing and streamlining documents, increasing
transparency and clarity, and using more tables, figures, and appendices to present information
and data in assessments. Phase 1 also focuses on assessments near the end of the development
process and close to final posting. Chemical assessments in Phase 2 of the implementation will
address all of the short-term recommendations from Appendix E, Table E-l. The IRIS Program is
implementing all of these recommendations but recognizes that achieving full and robust
implementation of certain recommendations will be an evolving process with input and feedback
from the public, stakeholders, and external peer review committees. Chemical assessments in
Phase 3 of implementation will incorporate the longer-term recommendations made by the NRC
as outlined in Appendix E, Table E-2, including the development of a standardized approach to
describe the strength of the evidence for noncancer effects. On May 16, 2012, EPA announced
(U.S. EPA, 2012b)3 that as a part of a review of the IRIS Program's assessment development
process, the NRC will also review current methods for weight-of-evidence analyses and
recommend approaches for weighing scientific evidence for chemical hazard identification. This
effort is included in Phase 3 of EPAs implementation plan.
For other general information about this draft assessment or other questions relating to
IRIS, the reader is referred to EPAs IRIS Hotline at (202) 566-1676 (phone), (202) 566-1749
(fax), or hotline.iris@epa.gov.
! EPA Announces NAS' Review of IRIS Assessment Development Process (http://www.epa.gov/iris).
1-4
-------
2.CHEMICAL AND PHYSICAL INFORMATION
Methanol is also known as methyl alcohol, wood alcohol; Carbinol; Methylol; colonial
spirit; Columbian spirit; methyl hydroxide; monohydroxymethane; pyroxylic spirit; wood
naphtha; and wood spirit. Some relevant physical and chemical properties are listed in Table 2-1
below (HSDB. 2009: IPCS. 1997).
Table 2-1 Relevant physical and chemical properties of methanol.
Property
CASRN:
Empirical formula:
Molecular weight:
Vapor pressure:
Vapor Density:
Specific gravity:
Boiling point:
Melting point:
Water solubility:
Log octanol-water partition coefficient:
Conversion factor (in air):
Value
67-56-1
CH3OH
32.04
160mmHgat30°C
1.11
0.7866 g/mL (25 °C)
64.7 °C
-98 °C
Miscible
-0.82 to -0.68
1 ppm = 1.31 mg/m3; 1 mg/m3= 0.763 ppm
Methanol is a clear, colorless liquid that has an alcoholic odor (TPCS, 1997). Endogenous
levels of methanol are present in the human body as a result of both metabolism4 and dietary
sources such as fruit, fruit juices, vegetables and alcoholic beverages,5 and can be measured in
exhaled breath and body fluids (Turner et al.. 2006: NTP-CERHR. 2004: IPCS. 1997). Dietary
exposure to methanol also occurs through the intake of some food additives. The artificial
sweetener aspartame and the beverage yeast inhibitor dimethyl dicarbonate (DMDC) release
methanol as they are metabolized (Stegink et al., 1989). In general, aspartame exposure does not
contribute significantly to the background body burden of methanol (Butchko et al., 2002). The
United Kingdom (U.K.) Food Standards Agency has stated that endogenous methanol production
4 Methanol is generated metabolically through enzymatic pathways such as the methyltransferase system (Fisher et
al.. 20001.
5 Fruits and vegetables contain methanol (Cal/EPA. 2012). Further, ripe fruits and vegetables contain natural pectin,
which is degraded to methanol in the body by bacteria present in the colon (SiragusaetaL 1988). Increased levels
of methanol in blood and exhaled breath have also been observed after the consumption of ethanol (Fisher etal.
2000).
2-1
-------
ranges from 300 to 600 mg/day (Lindinger et al., 1997) and that diet can contribute up to an
additional 1,000 mg/day, principally from fruits and vegetables (COT, 2011). Oral, dermal, or
inhalation exposure to methanol in the environment, consumer products, or workplace also
occur.
Methanol is a high production volume chemical with many commercial uses and it is a
basic building block for hundreds of chemical products. Many of its derivatives are used in the
construction, housing or automotive industries. Consumer products that contain methanol include
varnishes, shellacs, paints, windshield washer fluid, antifreeze, adhesives, de-icers, and Sterno
heaters. In 2009, the Methanol Institute (2009b) estimated a global production capacity for
methanol of about 35 million metric tons per year (close to 12 billion gallons), a production
capacity in the United States (U.S.) of nearly 3.7 million metric tons (1.3 billion gallons), and a
total U.S. demand for methanol of over 8 million metric tons. Methanol is among the highest
production volume chemicals reported in the U.S. EPA's Toxic Release Inventory (TRI).6 It is
among the top chemicals on the 2008 TRI lists of chemicals with the largest total on-site and off-
site recycling (6th), energy recovery (2nd) and treatment (1st) (U.S. EPA, 2009a). TRI also
reports that approximately 135,000,000 pounds of methanol was released or disposed of in the
United States in 2008, making methanol among the top five chemicals on the list entitled "TRI
On-site and Off-site Reported Disposed of or Otherwise Released in pounds for facilities in All
Industries for Hazardous Air Pollutant Chemicals U.S. 2008" (U.S. EPA. 2009c).
While production has switched to other regions of the world, demand for methanol is
growing steadily in almost all end uses. A large reason for the increase in demand is its use in the
production of biodiesel, a low-sulfur, high-lubricity fuel source. Global demand for biodiesel is
forecast to increase by 32% per year, rising from 30 million gallons in 2004, to 150 million
gallons by 2008, and to 350 million gallons by 2013 (Methanol Institute. 2009a). Power
generation and fuel cells could also be large end users of methanol in the near future (Methanol
Institute. 2009b).
6 The information in TRI does not indicate whether (or to what degree) the public has been exposed to toxic
chemicals. Therefore, no conclusions on the potential risks can be made based solely on this information (including
any ranking information). For more detailed information on this subject refer to The Toxics Release Inventory (TRI)
and Factors to Consider When Using TRI Data (U.S. EPA. 2009b).
2-2
-------
3.TOXICOKINETICS
3.1. Overview
As has been noted, methanol occurs naturally in the human body as a product of
metabolism and through intake of fruits, vegetables, and alcoholic beverages (Cal/EPA, 2012;
Turner et al., 2006; NTP-CERHR, 2004; IPCS, 1997). Table 3-1 summarizes background blood
methanol levels in healthy humans which were found to range from 0.25-5.2 mg/L. Formate, a
metabolite of methanol, also occurs naturally in the human body (IPCS, 1997). Table 3-1 outlines
background levels of formate in human blood. In most cases, methanol and formate blood levels
were measured in healthy adults following restriction of methanol-producing foods from the
diet.7
The absorption, excretion, and metabolism of methanol are well known and have been
consistently summarized in reviews such as NTP-CERHR (2004). IPCS (1997). U.S. EPA
(1996). Kavet and Nauss (1990). HEI (1987). and Tephly and McMartin (1984). Therefore, the
major portion of this toxicokinetics overview is based upon those reviews.
Studies conducted in humans and animals demonstrate rapid absorption of methanol by
inhalation, oral, and dermal routes of exposure. Table 3-2 outlines increases in human blood
methanol levels following various exposure scenarios. Blood levels of methanol following
various exposure conditions have also been measured in monkeys, mice, and rats, and are
summarized in Tables 3-3, 3-4, and 3-5, respectively. Once absorbed, methanol pharmacokinetic
(PK) data and physiologically based pharmacokinetic (PBPK) model predictions indicate rapid
distribution to all organs and tissues according to water content, as an aqueous-soluble alcohol.
Tissue:blood concentration ratios for methanol are predicted to be similar through different
exposure routes, though the kinetics will vary depending on exposure route and timing (e.g.,
bolus oral exposure versus longer-term inhalation). Because smaller species generally have faster
respiration rates relative to body weight than larger species, they are predicted to have a higher
rate of increase of methanol concentrations in the body when exposed to the same concentration
in air.
7 Background levels among people who are on normal/non-restricted diets may be higher than those on restricted
diets.
3-1
-------
Table 3-1 Background blood methanol and formate levels in human studies.
Description of human subjects
12 adults who drank no alcohol for 24 hr
12 adults who drank no alcohol for 24 hr
12 males on restricted diet (no
methanol-containing or methanol-
producing foods) for 12 hr
4 adult males who fasted for 8 hr, drank
no alcohol for 24 hr, and took in no fruits,
vegetables, or juices for 18 hr
8 adults who had no fruit, alcohol or
drugs for 48 hr
3 males who ate a breakfast with no
aspartame-containing cereals and no
juice
5 males who ate a breakfast with no
aspartame-containing cereals and no
juice (second experiment)
22 adults on restricted diet (no
methanol-containing or methanol-
producing foods) for 24 hr
35 adults who drank no alcohol for
1 week, fasted 4 hours
12 adults fasted 5 hours
30 fasted adults
24 fasted infants
30 adults. No dietary restrictions. Blood
levels were estimated from
concentrations in breath.
18 males, fasted 3 hr, no other dietary
restrictions
Methanol (mg/L)
mean ± SDa
(Range)
1.7 ±0.9
(0.4-4.7)
1.8 ±0.7
(No range data)
0.570 ± 0.305
(0.25-1.4)
1.75 ±0.65
(1.2-2.6)
No mean data
(0.3-2.4)
1.82± 1.21
(0.57-3.57)
1.93 ±0.93
(0.54-3.15)
1.8 ±2.6
(No range data)
0.64 ± 0.45
(No range data)
1.1
(0.4-2.2)
<4
(No range data)
<3.5
(No range data)
1.25±0.29a
(0.45-1.7)
2.62 ± 1.33
(0.7-5.2)
Formate (mg/L)
mean ± SD
(Range)
No data
No data
3.8± 1.1
(2.2-6.6)
No data
No data
9.08 ± 1.26
(7.31-10.57)
8.78 ± 1.82
(5.36-10.83)
11.2±9.1
(No range data)
No data
No data
19.1
(No range data)
No data
No data
No data
Reference
Batterman and
Franzblau (1997)
Batterman et al. (1998)
Cook et al. (1991)
Davoli et al. (1986)
Ernstgard et al. (2005)
Lee et al. (1992)
Lee et al. (1992)
Osterloh et al. (1996):
Chuwers et al. (1995)
Sarkola and Eriksson
(2001)
Schmutte et al. (1988)
Stegink et al. (1981)
Stegink et al. (1983)
Turner et al. (2006)
Woo et al. (2005)
Arithmetic mean and standard deviation calculated from mean values listed in Table 1 of Turner et al. (2006).
3-2
-------
Table 3-2 Human blood methanol and formate levels following methanol exposure.
Human subjects;
type of sample Exposure
collected a'b route
5 adults
7 adults Oral
Fasted 5 hours
Males; post , , , ...
exposure samples lnhalatlon
Males and females
with exercise; post Inhalation
exposure samples
Males and females;
post exposure Inhalation
samples
Males without
exercise; post Inhalation
exposure samples
Males with exercise;
post exposure Inhalation
samples
Females; post , , , ...
exposure samples lnhalatlon
Exposure
duration or
method
1 dose in
water
75 min
2hr
4hr
6hr
6hr
8hr
"Unless otherwise specified, it is assumed that whole
blnformation about dietarv restrictions is included in T;
Methanol
exposure
concentration
7 mg/kg bw
12.5 mg/kg bw
0 ppm
191 ppm
0 ppm
100 ppm
200 ppm
0 ppm
200 ppm
0 ppm
200 ppm
0 ppm
200 ppm
0 ppm
800 ppm
Blood methanol
mean
(mg/L)
9.04
0.570
1.881
0.64
3.72
7.82
1.8
6.5
1.82
6.97
1.93
8.13
1.8
30.7
Blood
formate
mean
(mg/L)
No data
3.8
3.6
No data
11.2
14.3
9.08
8.70
8.78
9.52
No data
Reference
Schmutte et al.
(1988)
Cook et al.
(1991)
Ernstgard et al.
(2QQ5)
Osterloh et al.
(1996)
Lee et al.
(1992)
Batterman et
al. (1998)
blood was used for measurements.
able 3-1.
3-3
-------
Table 3-3 Monkey blood methanol and formate levels following methanol exposure.
Strain -sex
Monkey; cynomolgus;
female; mean blood
methanol and range of
plasma formate at 30 min
post daily exposure during
premating, mating, and
pregnancy
Monkey; cynomolgus;
female; Lung only
inhalation of anesthetized
monkeys post exposure a
Monkey; Rhesus male;
post exposure blood level
Exposure Exposure
route duration
2.5 hr/day,
7days/wk during
, , , .. premating,
lnhalatlon mating, and
gestation
(348 days)
Inhalation 2 hr
Inhalation 6 hr
Methanol
exposure
concentration
0 ppm
200 ppm
600 ppm
1,800 ppm
10 ppm
45 ppm
200 ppm
900 ppm
900 ppm - FD
200 ppm
1,200 ppm
2,000 ppm
Blood
methanol
mean
(mg/L)
2.4
5
11
35
0.021
0.096
0.67
3.4
6.8
3.9
37.6
64.4
Blood
formate
mean or
range
(mg/L)
8.7
8.7
8.7
10
0.0032
0.012
0.11
0.13
0.44
5.4-13.2
at all doses
Reference
Burbacher et
al. (2004a;
1999a)
Dorman et
al. (1994)
Morton et al.
(1992)
FD=folate deficient
aMethanol and formate blood levels obtained from radiolabeled methanol and do not include background levels of methanol or
formate.
3-4
-------
Table 3-4 Mouse blood methanol and formate levels following methanol exposure.
Exposure
Species/strain/sex route
Mouse;CD-1;female; peak Injection
concentration (Cmax) (i.v.)
Mouse;CD-1;female; peak Q .
concentration (Cmax)
Mouse;CD-1;female; post
exposure plasma methanol Inhalation
and peak formate level
Mouse;CD-1;female; post
exposure blood methanol Inhalation
level
Mouse;CD-1;female; mean
post exposure plasma Inhalation
methanol level
Mouse;CD-1;female; plasma Oral-
level 1 hr post dosing Gavage
Mouse;CD-1;female; peak Oral-
plasma level Gavage
Methanol
Exposure exposure
duration concentration
100 mg/kg bw
GD18 500 mg/kg bw
2,500 mg/kg bw
GD18 2,500 mg/kg bw
10,000 ppm
6 hr 10,000 ppm
on GD8 (+ 4-MP)
15,000 ppm
2,500 ppm
5,000 ppm
n hr
10,000 ppm
15,000 ppm
0
1,000 ppm
2,000 ppm
/ hr/day on .- __,,
nnfi_nniR 5,000 ppm
7,500 ppm
10,000 ppm
15,000 ppm
GD6-GD15 4,000 mg/kg bw
1,500 mg/kg bw
GD8 1,500 mg/kg bw
(+ 4-MP)
Blood
methanol
mean
(mg/L)
252
869
3,521
3,205
2,080
2,400
7,140
1,883
3,580
6,028
11,165
1.6
97
537
1,650
3,178
4,204
7,330
3,856
1,610
1,450
Blood
formate
mean
(mg/L )
No data
No data
28.5
23
34.5
No data
No data
No data
35
43
Reference
Ward et al.
(1997)
Ward et al.
(1997)
Dorman et al.
(1995)
Pollack and
Brouwer (1996):
Perkins et al.
(1995a)
Rogers et al.
(1993b)
(1995)
4-MP=4-methylpyrazole (fomepizole)
3-5
-------
Table 3-5 Rat blood methanol and formate levels following methanol exposure.
Species; strain/sex: type Exposure
of sample collected route
Rat; Sprague-Dawley;
female; post exposure . , . ...
blood methanol level on mnaiation
3 days
Rat; Sprague-Dawley;
female; post exposure Inhalation
blood methanol level
Rat; LongEvans; female;
post exposure plasma Inhalation
level on GD7-GD12
Rat; LongEvans; female; 1
hr post exposure blood Inhalation
level
Rat; Long-Evans; male
and female; 1 hr post . , . ..
exposure blood level in '"halation
pups
Rat/Fischer-344 male; post , , , ..
, , , , ' r Inhalation
exposure blood level
Rat; Long-Evans; male; . , . ...
post- exposure serum level lnnalatlon
Rat/Fischer-344 male;
25 min post exposure
blood level for 4-wk |nha|atjon
animals; -250 mm post
exposure for 104-wk
animals
Rat/Fischer-344 female;
25 min post exposure
blood level for 4-wk . , . ,.
animals; -250 min post Inhalation
exposure for 104-wk
animals
Rat; Long-Evans; male; . , . ..
peak blood formate level lnnalatlon
Exposure
duration
7 hr/day for
19 days
8hr
7 hr/day on
GD7-GD19
6 hr/day on
GD6-
PND21
6 hr/day on
PND1-
PND21
6hr
6hr
19. 5 hr/day
for 4/1 04 wk
19 hr/day for
4/1 04 wk
6hr
Methanol
exposure
concentration
5,000 ppm
10,000 ppm
20,000 ppm
1,000 ppm
5,000 ppm
10,000 ppm
15,000 ppm
20,000 ppm
0 ppm
15,000 ppm
4,500 ppm
4,500 ppm
200 ppm
1,200 ppm
2,000 ppm
200 ppm
5,000 ppm
10,000 ppm
0 ppm
10 ppm
100 ppm
1,000 ppm
0 ppm
10 ppm
100 ppm
1,000 ppm
0 ppm FS
0 ppm FR
1,200 ppm-FS
1,200 ppm-FR
2,000 ppm-FS
2,000 ppm-FR
Blood
methanol
mean or
range (mg/L)
1,000-2,170
1,840-2,240
5,250-8,650
83
1,047
1,656
2,667
3,916
1.8-2.7
3,169-3,826
555
1,260
3.1
26.6
79.7
7.4
680-873
1,468
4.01 /3.78
1.56/3.32
3.84/3.32
53.59/12.08
13.39/3.60
6.73/3.70
4.34/4.32
88.33/8.50
No data
Blood
formate
mean or
range
(mg/L)
No data
No data
No data
No data
No data
5.4-1 3.2 at
all doses
No data
No data
No data
8.3
10.1
8.3
46.0
8.3
83.0
Reference
Nelson et al.
(1985)
Pollack and
Brouwer
(1996);
Perkins et al.
(1995a)
Stanton et al.
(1995)
•Weiss et al.
(1996)
Morton et al.
(1992)
Cooper et al.
(1992)
NEDO (1985b)
NEDO (1985b)
Lee et al.
(1994)
3-6
-------
Table 3-5 (Continued): Rat blood methanol and formate levels following methanol
exposure.
Species; strain/sex: type Exposure
of sample collected route
Rat; Spraque-Dawley; . . ...
, , , / t. Injection
female; peak concentration .. ,
ir- \ l'-v-)
\^max}
Rat;Sprague-Dawley; |niection
female; peak concentration ,.' ,
/r \ l'-v-)
^maxj
Rat; Long-Evans; male; Q .
peak blood methanol and
i . gavage
formate 3 3
Methanol
Exposure exposure
duration concentration
1 00 mg/kg bw
GD14
500 mg/kg bw
100 mg/kg bw
GD20
500 mg/kg bw
2,000 mg/kg bw
FS
2,000 mg/kg bw
FR
3,000 mg/kg bw-
FS
Single dose 3-°00 * bw
3,500 mg/kg bw-
FS
3,500 mg/kg bw-
FP
3,500 mg/kg bw-
FR
Blood
methanol
mean or
range (mg/L)
123.7
612.9
149.0
663.6
No data
4,800
4,800
4,800
Blood
formate
mean or
range
(mg/L)
No data
No data
9.2
538
9.2
718
9.2
38.2
860
Reference
Ward et al.
(1997)
Ward et al.
(1997)
Lee et al.
(1994)
FS = Folate sufficient; FR = Folate reduced; FP = Folate paired
At doses that do not saturate metabolic pathways, a small percentage of methanol is
excreted directly in urine. Because of the high blood:air partition coefficient for methanol and
rapid metabolism in all species studied, the bulk of clearance occurs by metabolism, though
exhalation and urinary clearance become more significant when doses or exposures are
sufficiently high to saturate metabolism (subsequently in this document, "clearance" refers to
elimination by all routes, including metabolism, as indicated by the decline in methanol blood
concentrations). Metabolic saturation and the corresponding clearance shift have not been
observed in humans and nonhuman primates because doses used were limited to the linear range,
but the enzymes involved in primate metabolism are also saturable.
The primary route of methanol elimination in mammals is through a series of oxidation
reactions that form formaldehyde, formate, and carbon dioxide (Figure 3-1). As noted in
Figure 3-1, methanol is converted to formaldehyde by alcohol dehydrogenase-1 (ADH1) in
primates and by catalase (CAT) and ADH1 in rodents. Although the first step of metabolism
occurs through different pathways in rodents and nonhuman primates, Kavet and Nauss (1990)
report that the reaction proceeds at similar rates (Vmax= 30 and 48 mg/hr/kg in rats and
nonhuman primates, respectively). In addition to enzymatic metabolism, methanol can react with
3-7
-------
hydroxyl radicals to spontaneously yield formaldehyde (Harris et al., 2003). Mannering et al.
(1969) also reported a similar rate of methanol metabolism in rats and monkeys, with 10 and
14% of a 1 g/kg dose oxidized in 4 hours, respectively; the rate of oxidation by mice was about
twice as fast, 25% in 4 hours. In an HEI study by Pollack and Brouwer (1996), the metabolism of
methanol was 2 times as fast in mice versus rats, with a Vmax for elimination of 117 and
60.7 mg/hr/kg, respectively. Despite the faster elimination rate of methanol in mice versus rats,
mice consistently exhibited higher blood methanol levels than rats when inhaling equivalent
methanol concentrations (See Tables 3-4 and 3-5). Possible explanations for the higher methanol
accumulation in mice include faster respiration (inhalation rate/body weight) and increased
fraction of absorption by the mouse (Perkins et al., 1995a). Sweeting et al. (2010) examined
methanol dosimetry in CD-I mice, New Zealand white (NZW) rabbits, and cynomolgus
monkeys, and found that peak plasma concentrations are not significantly different, but clearance
in rabbits is approximately half that of mice following a single 0.5 or 2 g/kg i.p. (intraperitoneal)
injection. This suggests that rabbit clearance is similar to that in rats and monkeys, since
Mannering et al. (1969) found that rat and monkey clearance rates are also about half that in
mice. Sweeting et al. (2010) did not report the clearance rates from monkeys, but the 6-hour
AUC in monkeys was similar to that in rabbits. Because smaller species generally have faster
breathing rates than larger species, humans would be expected to absorb methanol via inhalation
more slowly than rats or mice inhaling equivalent concentrations. If humans eliminate methanol
at a comparable rate to rats and mice, then humans would also be expected to accumulate less
methanol than those smaller species. However, if humans eliminate methanol more slowly than
rats and mice, such that the ratio of absorption to elimination stays the same, then humans would
be expected to accumulate methanol to the same internal concentration but to take longer to
reach that concentration.
In all species, formaldehyde is rapidly converted to formate, with the half-life for
formaldehyde being ~1 minute. Formaldehyde is oxidized to formate by two metabolic pathways
(Teng et al., 2001). The first pathway (not shown in Figure 3-1) involves conversion of free
formaldehyde to formate by the so-called low-affinity pathway (affinity = l/Km = 0.002/|jM)
mitochondrial aldehyde dehydrogenase-2 (ALDH2). The second pathway (Figure 3-1) involves a
two-enzyme system that converts glutathione-conjugated formaldehyde
(^-hydroxymethylglutathione [HMGSH]) to the intermediate ^-formylglutathione, which is
subsequently metabolized to formate and glutathione (GSH) by S-formylglutathione hydrolase.8
The first enzyme in this pathway, formaldehyde dehydrogenase-3 (ADH3), is rate limiting, and
the affinity of HMGSH for ADH3 (affinity = l/Km = 0.15/nM) is about a 100-fold higher than
8 Other enzymatic pathways for the oxidation of formaldehyde have been identified in other organisms, but this is
the pathway that is recognized as being present in humans (Caspi et al.. 2006: http://metacvc.org).
3-8
-------
that of free formaldehyde for ALDH2. In addition to the requirement of GSH for ADH3 activity,
oxidation by ADH3 is (NAD+ [nicotinamide adenine dinucleotide])-dependent. Under normal
physiological conditions NAD+ levels are about two orders of magnitude higher than NADH,
and intracellular GSH levels (mM range) are often high enough to rapidly scavenge
formaldehyde (Svensson et al., 1999; Meister and Anderson, 1983): thus, the oxidation of
HMGSH is favorable. In addition, genetic ablation of ADH3 results in increased formaldehyde
toxicity (Deltour et al., 1999). These data indicate that ADH3 is likely to be the predominant
enzyme responsible for formaldehyde oxidation at physiologically relevant concentrations,
whereas ALDHs likely contribute to formaldehyde elimination at higher concentrations (Dicker
and Cedebaum, 1986).
Primates
K
Alcohol dehydrogenase \
(ADH1) /
V
K
Formaldehyde dchvdrogenasc \^
(ADH3) /
V
K
; '\
S-lor m\ i^liiiiithiime In ill uhisc
v'
K
' \
Folatc-dcpcndcnt pathway N.
(see Figure 3-2) /
i /
V
CHX)H
(Methanol)
i
HCHO
(Formaldehyde)
|( + GSH)
HMGSH
(hydroxymethyl-GSH)
X
•*•
(S-formyl glutathione)
> |<-CSH)
HCOO (Formate)
1 <
CO2 (Carbon dioxide)
Rodents
A
/ Catalase (CAT)
\ andADHl
N
A
/ Formaldehyde dchvdrogenase
\ (ADH3)
2.
( S-form> lj;l u hith icmc hyclrolusc
N/1
/ CAT-peroxide and
Kolatc-clcpcndcnt pathway
N
Source: IPCS (1997).
Figure 3-1 Methanol metabolism and key metabolic enzymes in primates and rodents.
Rodents convert formate to carbon dioxide (CO2) through a folate-dependent enzyme
system and a CAT-peroxide system (Dikalova et al., 2001). Formate can undergo adenosine
triphosphate- (ATP-) dependent addition to tetrahydrofolate (THF), which can carry either one or
two one-carbon groups. Formate can conjugate with THF to form jV10-formyl-THF and its isomer
A^-formyl-THF, both of which can be converted to TV5, jV70-methenyl-THF and subsequently to
other derivatives that are ultimately incorporated into DNA and proteins via biosynthetic
3-9
-------
pathways (Figure 3-2). There is also evidence that formate generates CCV radicals, and can be
10
metabolized to CO2 via CAT and via the oxidation ofW -formyl-THF (Dikalovaetal.. 2001).
Cytoplasm
Mitochondria
methenylTHF THF
10-formylTHF
methenylTHF
methyleneTHF
I MTHFR
5-methylTHF
MS "
homocysteme
AdoHyc
PURINES
dTMP
Methionine
AdoMet
Source: Reprinted with permission of the American Society for Biochemistry and Molecular Biology Montserrat et al. (2006).
Figure 3-2 Folate-dependent formate metabolism. Tetrahydrofolate (THF)-mediated one
carbon metabolism is required for the synthesis of purines, thymidylate, and
methionine.
Unlike rodents, formate metabolism in primates occurs solely through a folate-dependent
pathway. Black et al. (1985) reported that hepatic THF levels in monkeys are 60% of that in rats,
and that primates are far less efficient in clearing formate than are rats and dogs. Studies of
human subjects involving [14C]formate suggest that -80% is exhaled as 14CC>2, 2-7% is excreted
in the urine, and -10% undergoes metabolic incorporation (Hanzlik et al., 2005, and references
therein ). Sweeting et al. (2010) have reported that formic acid accumulation in primates within
6 hours of a 2 g/kg i.p. exposure to methanol was 5-fold and 43-fold higher than in rabbits and
mice, respectively. Mice deficient in formyl-TFIF dehydrogenase exhibit no change in LD50 (via
intraperitoneal [i.p.]) for methanol or in oxidation of high doses of formate. Thus it has been
suggested that rodents efficiently clear formate via high capacity folate-dependent pathways,
peroxidation by CAT, and by an unknown third pathway; conversely, primates do not appear to
exhibit such capacity and are more sensitive to metabolic acidosis following methanol poisoning
(Cook etal.. 2001).
3-10
-------
Blood methanol and formate levels measured in humans under various exposure
scenarios are reported in Table 3-2. As noted in Table 3-2, 75-minute to 6-hour exposures of
healthy humans to 200 ppm methanol vapors, the American Council of Governmental Industrial
Hygienists (ACGIH) threshold limit value (TLV) for occupational exposure (ACGIH, 2000),
results in increased levels of blood methanol but not formate. A limited number of monitoring
studies indicate that levels of methanol in outdoor air are orders of magnitude lower than the
TLV (IPCS, 1997). Table 3-3 indicates that exposure of monkeys to 600 ppm methanol vapors
for 2.5 hours increased blood methanol but not blood formate levels. Normal dietary exposure to
aspartame, which releases 10% methanol during metabolism, is unlikely to significantly increase
blood methanol or formate levels (Butchko et al., 2002). Exposure to high concentrations of
aspartame is unlikely to increase blood formate levels; no increase in blood formate levels were
observed in adults ingesting "abusive doses" (100-200 mg/kg) of aspartame (Stegink et al.,
1981). Kerns et al. (2002) studied the kinetics of formate in 11 methanol-poisoned patients
(mean initial methanol level of 57.2 mmol/L or 1.83 g/L) and determined an elimination half-life
of 3.4 hours for formate. Kavet and Nauss (1990) estimated that a methanol dose of 11 mM or
210 mg/kg is needed to saturate folate-dependent metabolic pathways in humans. There are no
data on blood methanol and formate levels following methanol exposure of humans with reduced
ADH activity or marginal folate tissue levels, a possible concern regarding sensitive populations.
As discussed in greater detail in Section 3.2, a limited study in folate-deficient monkeys
demonstrated no increase in blood formate levels following exposure to 900 ppm methanol
vapors for 2 hours. In conclusion, limited available data suggest that typical occupational,
environmental, and dietary exposures are likely to increase baseline blood methanol but not
formate levels in most humans.
3.2. Key Studies
Toxicokinetic and metabolism studies (Burbacher et al., 2004b: Burbacher et al., 1999b:
Medinsky et al., 1997; Pollack and Brouwer, 1996; Dorman et al., 1994) provide key information
on interspecies differences, methanol metabolism during gestation, metabolism in the nonhuman
primate, and the impact of folate deficiency on the accumulation of formate.
As part of an effort to develop a physiologically based toxicokinetic model for methanol
distribution in pregnancy, Pollack and Brouwer (1996) conducted a large study that compared
toxicokinetic differences in pregnant and nonpregnant (NP) rats and mice. Methanol disposition9
9 Methanol concentrations in whole blood and urine were determined by gas chromatography with flame ionization
detection (Pollack and Kawagoe. 1991).
3-11
-------
was studied in Sprague-Dawley rats and CD-I mice that were exposed to 100-2,500 mg/kg of
body weight pesticide-grade methanol in saline by i.v. or oral routes. Exposures were conducted
in NP rats and mice, pregnant rats on gestation days (GD) GD7, GDI4, and GD20, and pregnant
mice on GD9 and GDIS. Disposition was also studied in pregnant rats and mice exposed to
1,000-20,000 ppm methanol vapors for 8 hours. Three to five animals were examined at each
dose and exposure condition.
• Based on the fit of various kinetic models to methanol measurements taken from all
routes of exposure, the authors concluded that high exposure conditions resulted in
nonlinear disposition of methanol in mice and rats.10 Both linear and nonlinear
pathways were observed with the relative contribution of each pathway dependent on
concentration. At oral doses of 100-500 mg/kg of body weight, methanol was
metabolized to formaldehyde and then formic acid through the saturable nonlinear
pathway. A parallel, linear route characteristic of passive-diffusion accounted for an
increased fraction of total elimination at higher concentrations. Nearly 90% of
methanol elimination occurred through the linear route at the highest oral dose of
2,500 mg/kg of body weight.
• Oral exposure resulted in rapid and essentially complete absorption of methanol. No
significant change in blood area under the curve (AUC) methanol was seen between
NP and GD7, GDI4 and GD20 rats exposed to single oral gavage doses of 100 and
2,500 mg/kg, nor between NP and GD9 and GDIS mice at 2,500 mg/kg. The data as a
whole suggested that the distribution of orally and i.v. administered methanol was
similar in rats versus mice and in pregnant rodents versus NP rodents with the
following exceptions:
• There was a statistically significant increase in the ratio of apparent volume of
distribution (Vd) to fractional bioavailability (F) by -20% (while F decreased but not
significantly), between NP and GD20 rats exposed to 100 mg/kg orally. However,
this trend was not seen in rats or mice exposed to 2,500 mg/kg, and the result in rats
at 100 mg/kg could well be a statistical artifact since both Vd and F were being
estimated from the same data, making the model effectively over-parameterized.
• There were statistically significant decreases in the fraction of methanol absorbed by
the fast process (resulting in a slower rise to peak blood concentrations, though the
peak is unchanged) and in the Vmax for metabolic elimination between NP and GDIS
mice. No such differences were observed between NP and GD9 mice.
10 A model incorporating parallel linear and nonlinear routes of methanol clearance was required to fit the data from
the highest exposure groups.
3-12
-------
The authors estimated a twofold higher Vmax for methanol elimination in mice versus
rats following oral administration of 2,500 mg/kg methanol, suggesting that similar
oral doses would result in lower methanol concentrations in the mouse versus rat.
Methanol penetration from maternal blood to the fetal compartment was examined in
GD20 rats by microdialysis.11 A plot of the amniotic concentration versus maternal
blood concentration (calculated from digitization of Figure 17 of Pollack and
Brouwer (1996) report) is shown in Figure 3-3. The ratio is slightly less than 1:1
(dashed line in plot) and appears to be reduced with increasing methanol
concentrations, possibly due to decreased blood flow to the fetal compartment.
Nevertheless, this is a very minor departure from linearity, consistent with a substrate
such as methanol that penetrates cellular membranes readily and distributes
throughout total body water.
_ 500°
31
Ł 4000
c
o
"§ 3000
•4-"
(V
c 2000 H
o
o
'I 1000 H
0
~^
y =-4E-05)? + 1.0782X
R2 = 0.9919
0 1000 2000 3000 4000 5000
Maternal blood concentration (mg/L)
Source: Reprinted with permission of the Health Effects Institute, Boston, MA; Pollack and Brouwer (1996).
Note: Data extracted from Figure 17 by digitization, and amniotic concentration estimated as ("Fetal Amniotic Fluid/Maternal
Blood Methanol") x ("Maternal Methanol").
Figure 3-3 Plot of fetal (amniotic) versus maternal methanol concentrations in GD20 rats.
11 Microdialysis was conducted by exposing the uterus (midline incision), selecting a single fetus in the middle of
the uterine horn and inserting a microdialysis probe through a small puncture in the uterine wall proximal to the
head of the fetus.
3-13
-------
Inhalation exposure resulted in less absorption in both rats and mice as concentrations of
methanol vapors increased, which was hypothesized to be due to decreased breathing rate and
decreased absorption efficiency from the upper respiratory tract.12 Based on blood methanol
concentrations measured following 8-hour inhalation exposures to concentrations ranging from
1,000-20,000 ppm, the study authors (Pollack and Brouwer, 1996) concluded that, across this
range, methanol accumulation in the mouse occurred at a two- to threefold greater rate compared
to the rat. They speculated that faster respiration rate and more complete absorption in the nasal
cavity of mice may explain the higher methanol accumulation and greater sensitivity to certain
developmental toxicity endpoints (see Section 4.3.2).
The Pollack and Brouwer (1996) study was useful for comparing effects in pregnant and
NP rodents exposed to high doses, but the implication of these results for humans exposed to
ambient levels of methanol is not clear (NTP-CERHR. 2004).
Sweeting et al. (2011; 2010) studied methanol and formic acid pharmacokinetics in male
C57BL/6 mice, male C3H mice, male CD-I mice, male NZW rabbits and male cynomolgus
monkeys (Macacafascicularis) following a 0.5 or 2 g/kg i.p. exposure to methanol. Blood
samples were taken over the entire methanol elimination period for rabbits (48 hours) and CD-I
mice (12 hours for 0.5 g/kg exposure; 24 hours for 2 g/kg exposure), over a 12-hour exposure
window for the C57BL/6 and C3H mice and a 6-hour post exposure window for monkeys.
Following the 2 g/kg dose, methanol blood levels exhibited saturated elimination kinetics in all
three species, and peak methanol concentrations were similar (68, 87 and 79 ± 10 mmol/L in
C57BL/6, C3H and CD-I mice, respectively; 114 ± 7 mmol/L in rabbits and 94 ± 14 mmol/L in
monkeys), though the peak concentrations in C57BL/6 (p < 0.01) and CD-I (p < 0.05) mice were
significantly lower than rabbits. Methanol clearance rates were 2.5-fold higher in CD-I mice
than in rabbits after the 2 g/kg exposure, and 2-fold higher after the 0.5 g/kg exposure. When
measured over the entire elimination period, plasma methanol AUCs in the rabbits were 175 ± 27
after a 0.5 g/kg dose and 1,893 ± 345 mmol-hr/L after a 2 g/kg dose. Comparable plasma
methanol AUCs in CD-I mice were more than 2-fold lower (71 ± 9 after a 0.5 g/kg dose, and
697 ± 50 mmol-hr/L after a 2 g/kg dose). At 12-hours (after a 2 g/kg dose), the plasma methanol
AUC values for C57BL/6, C3H and CD-I mice were 465 ± 14, 550 ± 30 and 640 ± 33
mmol-hr/L, respectively, and rabbits had an AUC value of 969 ± 77 mmol-hr/L. The elimination
period for plasma formic acid AUCs in the rabbits were 3.02 ±1.3 mmol-hr/L after a 0.5 g/kg
12 Exposed mice spent some exposure time in an active state, characterized by a higher ventilation rate, and the
remaining time in an inactive state, with lower (~i/2 of active) ventilation. The inactive ventilation rate was
unchanged by methanol exposure, but the active ventilation showed a statistically significant methanol-
concentration-related decline. There was also some decline in the fraction of time spent in the active state, but this
was not statistically significant.
3-14
-------
dose, and 10.6 ±1.4 mmol-hr/L after a 2 g/kg dose. In comparison, plasma formic acid AUCs in
CD-I mice were nearly 6-fold lower at 0.5 g/kg (71 ± 9 mmol-hr/L) and more than 3-fold lower
at 2 g/kg (697 ± 50 mmol-hr/L). Twelve hours after a 2 g/kg (i.p.) dose, the plasma formic acid
AUC values for C57BL/6, C3H, and CD-I mice were 2.1 ± 0.3, 1.6 ± 0.2, and 1.9 ± 0.2
mmol-hr/L, respectively, and rabbits had a formic acid AUC value of 3.0 ± 0.3 mmol-hr/L. All of
the 12-hour formic acid AUCs for the mice were significantly lower (p < 0.05) than the rabbit,
but none of the mouse strains differed from each other (p < 0.05). Formic acid accumulation at
6-hours post-exposure in monkeys (7.75 ±3.5 mmol-hr/L) was 5-fold and 43-fold higher than in
rabbits (1.5 ± 0.2 mmol-hr/L) and CD-I mice (0.15 ± 0.04 mmol-hr/L), respectively.
Burbacher et al. (2004a; 1999a) examined toxicokinetics in Macacafascicularis monkeys
prior to and during pregnancy. As part of the report (Reproductive and Offspring Developmental
Effects Following Maternal Inhalation Exposure to Methanol in Nonhuman Primates [which
includes the commentary of the Institute's Health Review Committee]), the HEI review
committee (Burbacher et al., 1999a) noted that this was a quality study using a relevant species.
The study objectives were to assess the effects of repeated methanol exposure on disposition
kinetics, determine whether repeated methanol exposures result in formate accumulation, and
examine the effects of pregnancy on methanol disposition and metabolism. Reproductive,
developmental, and neurological toxicity associated with this study were also examined and are
discussed in Sections 4.3.2 and 4.4.2. In a 2-cohort design, 48 adult females
(6 animals/dose/group/cohort) were exposed to 0, 200, 600, or 1,800 ppm methanol vapors
(99.9% purity) for 2.5 hours/day, 7 days/week for 4 months prior to breeding and during the
entire breeding and gestation periods. Six-hour methanol clearance studies were conducted prior
to and during pregnancy. Burbacher et al. (2004a; 1999a) reported that:
• At no point during pregnancy was there a significant change in baseline methanol
blood levels, which ranged from 2.2-2.4 mg/L throughout (Table 3-6).
• PK studies were performed initially (Study 1), after 90 days of pre-exposure and prior
to mating (Study 2), between GD66 and GD72 (Study 3), and again between GD126
and GDI32 (Study 4). These studies were analyzed using classical PK (one-
compartment) models.
• Disproportionate mean, dose-normalized, and net blood methanol dose-time profiles
in the 600 and 1,800 ppm groups suggested saturation of the metabolism-dependent
pathway. Data from the 600 ppm group fit a linear model, while data from the
1,800 ppm group fit a Michaelis-Menten model.
3-15
-------
• Methanol elimination rates modestly increased between Study 1 and Study 2 (90 days
prior to mating). This change was attributed to enzyme induction from the subchronic
exposure.
• Blood methanol levels were measured every 2 weeks throughout pregnancy, and
while there was measurement-to-measurement variation, there was no significant
change or trend over the course of pregnancy (Table 3-6). An upward trend in
elimination half-life appeared to correspond with a downward trend in blood
methanol clearance between Studies 2, 3, and 4. However, the changes were not
statistically significant and the time-courses for blood methanol concentration
(elimination phase) appeared fairly similar.
• Significant differences between baseline and plasma formate levels (p = 0.005), and
between prebreeding and pregnancy (p = 0.0001) were observed but were not dose
dependent (Table 3-7).
• Significant differences in serum folate levels between baseline and prepregnancy
(p = 0.02), and between prepregnancy and pregnancy (p = 0.007) were not dose
dependent (Table 3-8).
Table 3-6 Plasma methanol concentrations in monkeys.
Exposure Group
Control (n=11)
200 ppm (n=12)
600 ppm (n=11)
1,800 ppm (n=12)
Mean3
Baseline
2.3 ±0.1
2.2 ±0.1
2.4 ±0.1
2.4 ±0.1
plasma methanol level
Pre-breeding
2.3±0.1
4.7 ±0.1
10.5 ±0.3
35.6 ± 1.0
(mg/L) during each
Breeding
2.3 ±0.1
4.8 ±0.1
10.9 ±0.2
35.7 ±0.9
exposure period
Pregnancy13
2.7 ±0.1
5.5 ±0.2
11.0±0.2
35.5 ±0.9
Values are presented as means ± SE in mg/L.
n = 9 for control, 200 ppm, and 600 ppm pregn;
Source: Reprinted with permission of the Health Effects Institute, Boston, MA; Burbacheret al. (1999a).
br> = 9 for control, 200 ppm, and 600 ppm pregnancy groups; n = 10 for 1,800 ppm pregnancy group.
3-16
-------
Table 3-7 Plasma formate concentrations in monkeys.
Mean3 plasma formate level (mg/L) during each exposure period
Exposure Group
Control (n=11)
200 ppm (n=12)
600 ppm (n=11)
1,800 ppm (n=12)
Baseline
8.3 ±9.2
7.4 ±0.9
6.9 ±0.5
6.4 ±0.9
Pre-breeding
7.8 ±0.5
8.3 ±0.5
7.8 ±0.5
8.7 ±0.5
Breeding
10.1 ±0.9
9.7 ±0.5
9.2 ±0.5
11 ±0.5
Pregnancy13
8.3 ± 1.4
7.8 ±0.5
8.7 ± 1.4
10± 1.4
aValues are presented as means ± SE in mg/L; transformed from mM, for consistency.
bn = 9 for control, 200 ppm, and 600 ppm pregnancy groups; n = 10 for 1,800 ppm pregnancy group.
Source: Reprinted with permission of the Health Effects Institute, Boston, MA; Burbacheret al. (1999a).
Table 3-8 Serum folate concentrations in monkeys.
Mean3 serum folate level (ug/L) during each exposure period
Exposure Group
Control (n=11)
Baseline
14.4 ± 1.0
Day 70
Pre-pregnancya
14.0 ± 1.2
Day 98
Pre-pregnancya
13.4 ± 1.2
Day 55
Pregnancy3
16.0 ± 1.1
Day 113
Pregnancyb|C
15.6 ± 1.1
200ppm(n=12) 11.9±1.3 13.2±1.6 12.9±1.3 15.5±1.5 13.4±1.3
600ppm(n=11) 12.5± 1.4 15.4± 1.2 13.4± 1.0 14.8±1.1 16.4±1.0
1,800 ppm (n=12) 12.6±0.7 14.8±1.2 15.3±1.1 15.9±1.2 15.7±1.0
aValues are presented as means ± SE in ug/L.
bNumber of days exposed to methanol
°n = 9 for control, 200 ppm, and 600 ppm pregnancy groups; n = 10 for 1,800 ppm pregnancy group.
Source: Reprinted with permission of the Health Effects Institute, Boston, MA; Burbacheret al. Q999a).
A series of studies by Medinsky et al. (1997) and Dorman et al. (1994) examined
metabolism and pharmacokinetics of [14C]methanol and [14C]formate in normal and folate-
deficient cynomolgus, M. fascicularis monkeys that were exposed to [14C]methanol through an
endotracheal tube while anesthetized. In the first stage of the study, 4 normal 12-year-old
cynomolgus monkeys were each exposed to 10, 45, 200, and 900 ppm [14C]methanol vapors
(>98% purity) for 2 hours. Each exposure was separated by at least 2 months. After the first stage
of the study was completed, monkeys were given a folate-deficient diet supplemented with 1%
succinylsulfathiazole (an antibacterial sulfonamide used to inhibit folic acid biosynthesis from
intestinal bacteria) for 6-8 weeks in order to obtain folate concentrations of <3 ng/mL serum and
<120 ng/mL erythrocytes. Folate deficiency did not alter hematocrit level, red blood cell count,
mean corpuscular volume, or mean corpuscular hemoglobin level. The folate-deficient monkeys
3-17
-------
were exposed to 900 ppm [14C]methanol for 2 hours. The results of the Medinsky et al. (1997)
and Dorman et al. (1994) studies showed:
• Dose-dependent changes in toxicokinetics and metabolism did not occur as indicated
by a linear relationship between inhaled [14C]methanol concentration and
end-of-exposure blood [14C]methanol level, [14C]methanol AUC and total amounts of
exhaled [14C]methanol and [14C]carbon dioxide.
• Methanol concentration had no effect on elimination half-life (<1 hour) and percent
urinary [14C]methanol excretion (<0.01%) at all doses.
• Following exposure to 900 ppm methanol, urinary excretion or exhalation of
[14C]methanol did not differ significantly between monkeys in the folate sufficient
and deficient state. There was no significant [14C]formate accumulation at any dose.
• Peak blood [14C]formate levels were significantly higher in folate-deficient monkeys,
but did not exceed endogenous blood levels reported by the authors to be between 0.1
and 0.2 mmol/L (4.6-9.2 mg/L).
An HEI review committee (Medinsky et al., 1997) noted that absolute values in this study
cannot be extrapolated to humans because the use of an endotracheal tube in anesthetized
animals results in an exposure scenario that is not relevant to humans. However, the data in this
study suggest that a single exposure to methanol (10- 900 ppm for 2 hours) is unlikely to result in
a hazardous elevation in formate levels, even in individuals with moderate folate deficiency.
3.3. Human Variability in Methanol Metabolism
The ability to metabolize methanol may vary among individuals as a result of genetic,
age, and environmental factors. Reviews by Agarwal (2001), Burnell et al.(1989), Bosron and Li
(1986), and Pietruszko (1980), discuss genetic polymorphisms for ADH. Class IADH, the
primary ADH in human liver, is a hetero- or homodimer composed of randomly associated
polypeptide units encoded by three separate gene loci (ADH1 A, ADH1B, and ADH1C).
Polymorphisms have been found to occur at the ADH1B (ADH1B*2, ADH1B*3) and ADH1C
(ADH1C*2) gene loci; however, no human allelic polymorphism has been found in ADH1 A.
The ADH1B*2 phenotype is estimated to occur in -15% of Caucasians of European descent,
85% of Asians, and <5% of African Americans. ADH1C*1 is also highly prevalent in Asians, but
has only been examined in a few studies of Chinese and Korean samples (Eng et al., 2007).
Fifteen percent of African Americans have the ADH1B*3 phenotype, while it is found in <5% of
Caucasian Europeans and Asians. To date, there are two reports of polymorphisms in ADH3
3-18
-------
(Cichoz-Lach et al., 2007; Hedberg et al., 2001), yet the functional consequence(s) for these
polymorphisms remains unclear.
Although racial and ethnical differences in the frequency of the occurrence of ADH
alleles in different populations have been reported, ADH enzyme kinetics (Vmax and Km) have not
been reported for methanol. There is an abundance of information pertaining to the kinetic
characteristics of the ADH dimers to metabolize ethanol in vitro. Methanol blood concentrations
of 2.62 ± 1.33 mg/L (Table 3-1) in 18 Korean males (Woo et al., 2005) were considerably higher
than the sample U.S. background distribution of 1.36 mg/L and 0.77 mg/L estimated in Section
5.3.6. However, the functional and biological significance is not well understood due to the lack
of data documenting metabolism and disposition of methanol or ethanol in individuals of known
genotype. Thus, while potentially significant, the contribution of ethnic and genetic
polymorphisms of ADH to the interindividual variability in methanol disposition and metabolism
cannot be reliably quantified at this time.
Because children generally have higher baseline breathing rates and are more active, they
may receive higher methanol doses than adults exposed to equivalent concentrations of any air
pollutant (NTP-CERHR, 2004). There is evidence that children under 5 years of age have
reduced ADH activity. A study by Pikkarainen and Raiha (1967) measured liver ADH activity
using ethanol as a substrate and found that 2-month-old fetal livers have -3-4% of adult ADH
liver activity. ADH activity in 4 to 5 month old fetuses is -10% of adult activity, and an infant's
activity is -20% of adult activity. ADH continues to increase in children with age and reaches a
level that is within adult ranges at 5 years of age. Adults were found to have great variation in
ADH activity (1,625 to 6,530/g liver wet weight or 2,030 to 5,430 mU/100 mg soluble protein).
Smith et al. (1971) also compared liver ADH activity in 56 fetuses (9 to 22 weeks gestation),
37 infants (premature to <1 year old), and 129 adults (>20 years old) using ethanol as a substrate.
ADH activity was 30% of adult activity in fetuses and 50% of adult activity in infants. There is
evidence that some human infants are able to efficiently eliminate methanol at high exposure
levels, however, possibly via CAT (Tran et al., 2007).
ADH3 exhibits little or no activity toward small alcohols, thus the previous discussion is
not relevant to the ontogeny of formaldehyde elimination (clearance). While such data on ADH3
activity does not exist, ADH3 mRNA is abundantly expressed in the mouse fetus (Ang et al.,
1996) and is detectible in human fetal tissues (third trimester), neonates and children (Hines and
McCarver, 2002: Estonius et al., 1996).
As noted earlier in this section, folate-dependent reactions are important in the
metabolism of formate. Individuals who are commonly folate deficient include those who are
pregnant or lactating, have gastrointestinal (GI) disorders, have nutritionally inadequate diets, are
alcoholics, smoke, have psychiatric disorders, have pernicious anemia, or are taking folic acid
3-19
-------
antagonist medications such as some antiepileptic drugs (NTP-CERHR, 2004; IPCS, 1997).
Groups which are known to have increased incidence of folate deficiencies include Hispanic and
African American women, low-income elderly, and mentally ill elderly (NTP-CERHR, 2004).
A polymorphism in methylene tetrahydrofolate reductase reduces folate activity and is found in
21% of Hispanics in California and 12% of Caucasians in the United States. Genetic variations in
folic acid metabolic enzymes and folate receptor activity are theoretical causes of folate
deficiencies.
3.4. Physiologically Based Pharmacokinetic Models
In accordance with the needs of this human health assessment, particularly the derivation
of human health effect benchmarks from studies of the developmental effects of methanol
inhalation exposure in mice (Rogers et al., 1993b), monkeys (Burbacher et al., 2004b: Burbacher
et al., 1999b) and rats (NEDO, 1987) models were evaluated for their ability to estimate mouse,
monkey and rat internal dose metrics. A human model was developed to extrapolate those
internal metrics to inhalation and oral exposure concentrations that would result in the same
internal dose in humans (HECs and HEDs). The procedures used for the development,
calibration and use of these models are summarized in this section, with further details provided
in Appendix B, "Development, Calibration and Application of a Methanol PBPK Model."
3.4.1. Model Requirements for EPA Purposes
3.4.1.1. MOA and Selection of a Dose Metric
Dose metrics closely associated with one or more key events that lead to the selected
critical effect are preferred for dose-response analyses compared to metrics not clearly
correlated. For instance, internal (e.g., blood, target tissue) measures of dose are preferred over
external measures of dose (e.g., atmospheric or drinking water concentrations), especially when,
as with methanol, blood methanol concentrations increase disproportionally with dose (Rogers et
al., 1993b). This is likely due to the saturable metabolism of methanol. In addition, respiratory
and GI absorption may vary between and within species. Mode of action (MOA) considerations
can also influence whether to model peak concentrations (Cmax) or a time-dependent metric such
as area under the curve (AUC), and whether to model the parent compound with or without its
metabolites for selection of the most adequate dose metric.
3-20
-------
As discussed in Section 4.3, developmental effects following methanol exposures have
been noted in both rats and mice (Rogers et al.. 1993a: Rogers etal.. 1993b: NEDO, 1987:
Nelson et al., 1985), but are not as evident or clear in primate exposure studies (Burbacher et al.,
2004b: Clary, 2003; Andrews et al., 1987). The report of the New Energy Development
Organization (NEDO, 1987) of Japan, which investigated developmental effects of methanol in
rats, indicated that there is a potential that developing rat brain weight is reduced following
maternal and neonatal exposures. These exposures included both in utero and postnatal
exposures. The methanol PBPK models developed for this assessment do not explicitly describe
these exposure routes. Mathematical modeling efforts have focused on the estimation of human
equivalent external exposures that would lead to an increase in maternal blood levels of methanol
or its metabolites presumed to be associated with developmental effects as reported in rats
(NEDO, 1987), mice (Rogers et al., 1993b) and monkeys (Burbacher et al., 2004b: Burbacher et
al., 1999b). PBPK models were developed for all species, but because measured internal blood
methanol levels suitable for use as estimates of peak concentrations (Cmax) are provided by
Rogers et al. (1993b), a mouse PBPK model is not used or discussed in this toxicological review.
However, limited discussion of the mouse models is included, as they are useful in evaluating
model structure.
In a recent review of the reproductive and developmental toxicity of methanol, a panel of
experts concluded that methanol, not formate, is likely to be the proximate teratogen and
determined that blood methanol level is a useful biomarker of exposure (NTP-CERHR, 2004;
Dormanetal., 1995). The NTP-CERHR Expert Panel based their assessment of potential
methanol toxicity on an assessment of circulating blood levels (NTP-CERHR, 2004). While
recent in vitro evidence indicates that formaldehyde is more embryotoxic than methanol and
formate (Harris et al., 2004; 2003), the high reactivity of formaldehyde would limit its unbound
and unaltered transport as free formaldehyde from maternal to fetal blood (Thrasher and Kilburn,
2001), and the capacity for the metabolism of methanol to formaldehyde is likely lower in the
fetus and neonate versus adults (see discussion in Section 3.3). Thus, even if formaldehyde is
ultimately identified as the proximate teratogen, methanol would likely play a prominent role, at
least in terms of transport to the target tissue.
Given the reactivity of formaldehyde, models that predict levels of formaldehyde in the
blood are difficult to validate. However, production of formaldehyde or formate following
exposure to methanol can be estimated by summing the total amount of methanol cleared by
metabolic processes.13 This metric of formaldehyde or formate dose has limited value since it
ignores important processes that may differ between species, such as elimination (all routes) of
13 This assumption is more likely to be appropriate for formaldehyde than formate as formaldehyde is a direct
metabolite of methanol.
3-21
-------
these two metabolites, but it can be roughly equated to the total amount of metabolites produced
and may be the more relevant dose metric if formaldehyde is found to be the proximate toxic
moiety. Thus, both blood methanol and total metabolism metrics are considered to be important
components of the PBPK models. Dose metric selection and MOA issues are discussed further in
Section 4.7.
3.4.1.2. Criteria for the Development of Methanol PBPK Models
The development of methanol PBPK models that would meet the needs of this
assessment was organized around a set of criteria that reflect: (1) the MOA(s) being considered
for methanol; (2) absorption, distribution, metabolism, and elimination characteristics; (3) dose
routes necessary for interpreting toxicity studies or estimating HECs; and (4) general parameters
needed for the development of predictive PK models.
The criteria with a brief justification are provided below:
• (1) Must simulate blood methanol concentrations and total methanol metabolism.
Blood methanol is the recommended dose metric for developmental effects, but total
metabolism may be a useful metric.
• (2) Must be capable of simulating experimental blood methanol and total metabolism
for the inhalation route of exposure in rats (a) and humans (b), and the oral route in
humans (c). These routes are important for determining dose metrics in the most
sensitive test species under the conditions of the toxicity study and in the relevant
exposure routes in humans.
• (3) The model code should easily allow designation of respiration rates during
inhalation exposures. A standard variable in inhalation route assessments is
ventilation rate. Blood methanol concentrations will depend strongly on ventilation
rate, which varies significantly between species.
• (4) Must address the potential for saturable metabolism of methanol. Saturable
metabolism has the potential to bring nonlinearities into the exposure: tissue dose
relationship.
• (5) Model complexity should be consistent with modeling needs and limitations of the
available data. Model should adequately describe the biological mechanisms that
determine the internal dose metrics (blood methanol and total metabolism) to assure
that it can be reliably used to predict those metrics in exposure conditions and
3-22
-------
scenarios where data are lacking. Compartments or processes should not be added
that cannot be adequately characterized by the available data.
Although existing rat models are useful for the evaluation of the dose metrics associated
with methanol's developmental effects and the relevant toxicity studies, including gestational
exposures, no pregnancy-specific PBPK model exists for methanol, and limited data exists for
the development and validation of a fetal/gestational/conceptus compartment. However, EPA
determined that nonpregnancy models for the appropriate species and routes of exposure could
prove to be valuable because, as discussed in Section 3.2, levels of methanol in NP, pregnant and
fetal blood are expected to be similar following the same oral or inhalation exposure. Pollack and
Brouwer (1996) determined that methanol distribution in rats and mice following repeated oral
and i.v. exposures up to day 20 of gestation is "virtually unaffected by pregnancy, with the
possible exception of the immediate perinatal period." Ward et al. (1997) report a "nonlinear"
relationship between the maternal blood and conceptus, but the nonlinear perception given by
Figure 8 in their paper is the result of the data being plotted on a log-y/linear-x scale. Replotting
the data from their Table 5 (AUC) shows the results to be linear, especially in the low-dose
region which is of the greatest concern (Figure 3-4).
3-23
-------
10000
">
re
3
f 1000
3
o
D
<
« 100
1 <
o
i
0 10 J
t
A
**-- f
v'
*
>
^
>
D 1000 2000 3000
•
• rat
• mouse
- y = x
4000 5000
Maternal ADC (ug/mL-day)
ocnn
Conceptus ADC (ug/mL-day)
->•->• IV) IV) CO G
en o en o en o o
o o o o o o c
3OOOOOOC
k*. I II II
B
•/' °
•
^'
+ '
*
D
+ rat
• mouse
- y=x
U ' i i i
0 1000 2000 3000 4000 5000
Maternal ADC (ug/mL-day)
Source: Adapted with permission of Elsevier; Ward et al. (1997).
Note: Plotted (A) on a log-linear scale, as in Figure 8 of Ward et al. (1997), and (B) on a linear-linear scale. In both panels the
line y = x is plotted (dashed line) for comparison.
Figure 3-4 Conceptus versus maternal blood AUC values for rats and mice.
3-24
-------
The critical window for methanol induction of cervical rib malformations in CD-I mice
has been identified as occurring between GD6 and GD7 (Rogers and Mole, 1997; Rogers et al.,
), a developmental period roughly equivalent to week 3 of human development (Chernoff
and Rogers, 2004). Methanol blood kinetics measured during and after inhalation exposure in NP
and pregnant mice on GD6-GD10 and GD6-GD15 (Perkins et al.. 1996: Dormanetal.. 1995:
Perkins et al., 1995a: Rogers et al., 1993b) are also similar. Further, the available data indicate
that the maternal blood:fetal partition coefficient is approximately 1 at dose levels most relevant
to this assessment (Ward et al., 1997: Horton et al., 1992). Further supporting data exist for
ethanol, which is quite similar to methanol in its partitioning and transport properties. In rats
(Zorzano and Herrera, 1989: Guerri and Sanchis, 1985), sheep (Brien etal., 1985: Cumming et
al., 1984), and guinea pigs (Clarke et al., 1986), fetal and maternal blood concentrations of
ethanol are virtually superimposable; maternal to fetal blood ratios are very close to 1, including
during late gestation. Also, fetal brain concentrations in guinea pigs (Clarke etal., 1986) were
very similar to the maternal concentrations. Consequently, fetal methanol concentrations are
expected to be roughly equivalent to that in the mother's blood. Thus, pharmacokinetics and
blood dose metrics for NP rats and humans are expected to provide reasonable approximations of
pregnancy levels and fetal exposure, particularly during early gestation, that improve upon
default estimations from external exposure concentrations.
In addition to the absolute maternal-fetal concentration similarity noted above, it is
common practice to use blood concentrations as an appropriate metric for risk extrapolation via
PBPK modeling for effects in various tissues, based on the reasonable expectation that any
tissue:blood differences will be similar in both the test species and humans. For example, even if
the brain:blood ratio was around 1.2:1 in the mouse or rat, because tissue:blood ratios depend on
tissue composition which is expected to be quite similar in rats and humans, the brain:blood
levels in humans is also expected to be close to 1.2:1. Therefore, the potential error that might
occur by using blood instead of brain concentration in evaluating the dose-response in rats will
be cancelled out by using blood instead of brain concentration in the human. Measured fetal
blood levels are virtually identical to maternal levels for methanol (and ethanol) thus indicating
that the rate of metabolism in the fetus is not sufficient to significantly reduce the fetal
concentration of methanol versus maternal. Use of a PBPK model to predict maternal levels will
give a better estimate of fetal exposure than use of the applied dose or exposure, because there
are animal-human differences in adult PK of methanol for which the model accounts, based on
PK data from humans as well as rodents.
3-25
-------
3.4.2. Methanol PBPK Models
As has been discussed, methanol is well absorbed by both inhalation and oral routes and
is readily metabolized to formaldehyde, which is rapidly converted to formate in both rodents
and humans. As was discussed in Section 3.1, the enzymes responsible for metabolizing
methanol are different in adult rodents and humans. Several rat, mouse and human PBPK models
that attempt to account for these species differences have been published (Fisher et al., 2000;
WardetaL 1997: Perkins et al.. 1995a: HortonetaL 1992). Two methanol PK models
(Bouchard et al., 2001; Ward etal., 1997) were identified as potentially appropriate for use in
animal-to-human extrapolation of methanol metabolic rates and blood concentrations. An
additional methanol PBPK model by Fisher et al. (2000) was considered principally because it
had an important feature - pulmonary compartmentalization (see below for details) - worth
adopting in the final model.
3.4.2.1. Ward etal. (1997)
The PBPK model of Ward et al. (1997) describes inhalation, oral and i.v. routes of
exposure and is parameterized for both NP and pregnant mice and rats (Table 3-9). The model
has not been parameterized for humans.
Respiratory uptake of methanol is described as a constant infusion into arterial blood at a
rate equal to the minute ventilation times the inhaled concentration and includes a parameter for
respiratory bioavailability, which for methanol is <100%. This simple approach is nonstandard
for volatile compounds but is expected to be appropriate for a compound like methanol, for
which there is little clearance from the blood via exhalation. Oral absorption is described as a
biphasic process, dependent on a rapid and a slow first-order rate constant.
Methanol elimination in the Ward et al. (1997) model is primarily via saturable hepatic
metabolism. The parameters describing this metabolism come from the literature, primarily
previous work by Ward and Pollack (1996) and Pollack et al. (1993). A first-order elimination of
methanol from the kidney compartment includes a lumped metabolic term that accounts for both
renal and pulmonary excretion.
The model adequately fits the experimental blood kinetics of methanol in rats and mice
and is therefore suitable for simulating blood dosimetry in the relevant test species and routes of
exposure (oral and i.v.). The Ward et al. (1997) model meets criteria 1, 2a, 2c, 3, 4, and 5
outlined in Section 3.4.1.2. The most significant limitation is the absence of parameters for the
oral and inhalation routes in the human. A modified version of this model that includes human
parameters and a standard PBPK lung compartment might be suitable for the purposes of this
assessment.
3-26
-------
3.4.2.2. Bouchard et al. (2001)
The Bouchard et al. (2001) model is not actually a PBPK model but is an elaborate
classical PK model, since the transfer rates are not determined from blood flows, ventilation,
partition coefficients, and the like. The Bouchard et al. (2001) model uses a single compartment
for methanol: a central compartment represented by a volume of distribution where the
concentration is assumed to equal that in blood. The model was developed for inhalation and i.v.
kinetics only. Methanol is primarily eliminated via saturable metabolism. The model adequately
simulates blood kinetics in NP rats and humans following inhalation exposure and in NP rats
following i.v. exposure; there is no description for oral absorption. Because methanol distributes
with total body water (Ward et al., 1997; Horton et al., 1992), this simple model structure is
sufficient for predicting blood concentrations of methanol following inhalation and i.v. dosing.
The Bouchard et al. (2001) model has the advantage of simplicity, reflecting the
minimum number of compartments necessary for representing blood methanol pharmacokinetics.
Because volume of distribution can be easily and directly estimated for water-soluble compounds
like methanol or fit directly to experimental kinetics data, concern over the scalability of this
parameter is absent. The model has been parameterized for a required human exposure route,
inhalation (Table 3-9). The model meets criteria 1, 2b, 3, 4, and 5 described in Section 3.4.1.2.
However, the Bouchard model has a specific and significant limitation. The model has not been
parameterized for the oral route in humans. As such, the model cannot be used to conduct the
necessary interspecies extrapolation.
Table 3-9 Routes of exposure optimized in models - optimized against blood
concentration data.
Route
Injection (i.v.)
Inhalation
Oral
Ward et al. (1997)
Mouse Rat Human
P/NP P/NP
P/NP
P/NP NP
Bouchard et al. (2001)
Mouse Rat Human
NP
NP NP
-
P = Pregnant NP = Nonpregnant
Source: Ward et al. (1997): Bouchard et al. (2001).
3.4.3. Selected Modeling Approach
As discussed earlier regarding model criteria, fetal methanol concentrations can
reasonably be assumed to equal maternal blood concentration. Thus, methanol pharmacokinetics
3-27
-------
and blood dose metrics for NP laboratory animals and humans are expected to improve upon
default extrapolations from external exposures as estimates of fetal exposure during early
gestation. The same level of confidence cannot be placed on the whole-body rate of metabolism,
in particular as a surrogate for formaldehyde dose. Because of formaldehyde's reactivity and the
limited fetal metabolic (ADH) activity (see Sections 3.3 and 4.9.1), fetal formaldehyde
concentration increases (from methanol) will probably not equal maternal increases in
formaldehyde concentration. But since there is no model that explicitly describes formaldehyde
concentration in the adult, let alone the fetus, the metabolism metric is the closest one can come
to predicting fetal formaldehyde dose. This metric is expected to be a better predictor of
formaldehyde dose than applied methanol dose or even methanol blood levels, which do not
account for species differences in conversion of methanol to formaldehyde.
Most of the published rodent kinetic models for methanol describe the metabolism of
methanol to formaldehyde as a saturable process but differ in the description of metabolism to
and excretion of formate (Bouchard et al.. 2001: Fisher et al.. 2000: Wardetal.. 1997). The
model of Ward et al. (1997) used one saturable and one first-order pathway to describe methanol
elimination in mice. The saturable pathway described in Ward et al. (1997) can specifically be
ascribed to metabolic formation of formaldehyde in the liver, while the renal first-order
elimination described in that paper represents nonspecific clearance of methanol (e.g.,
metabolism, excretion, or exhalation), since it was not fit to route-specific elimination data.
However, Pollack and Brouwer (1996) obtained a rate constant for the urinary elimination rate
from rat urine excretion data, so it can be made specific to that route by use of that parameter.
The model of Ward et al. (1997) does not describe kinetics of formaldehyde subsequent to its
formation and does not include any description of formate.
Bouchard et al. (2001) employed a metabolic pathway for conversion of methanol to
formaldehyde and a second pathway described as urinary elimination of methanol in rats and
humans. They then explicitly describe two pathways of formaldehyde transformation, one to
formate and the other to "other, unobserved formaldehyde byproducts." Finally, formate removal
is described by two pathways, one to urinary elimination, and one via metabolism to CC>2 (which
is exhaled). All of these metabolic and elimination steps are described as first-order processes,
but the explicit descriptions of formaldehyde and formate kinetics significantly distinguish the
model of Bouchard et al. (2001) from that of Ward et al. (1997), which only describes methanol.
There are two other important distinctions between the Ward et al. (1997) and Bouchard
et al. (2001) models. The former is currently capable of simulating blood data for all exposure
routes in mice but not humans, while the latter is capable of simulating human inhalation route
blood pharmacokinetics but not those in mice. The Ward et al. (1997) model has more
compartments than is necessary to adequately represent methanol disposition but has been fit to
3-28
-------
PK data in pregnant and NP mice for all routes of exposure (i.v., oral, and inhalation). The Ward
et al. (1997) model has also been fit to i.v. and oral route PK data in rats. Based primarily on the
extensive amount of fitting that has already been demonstrated for this model, it was determined
that a modified Ward et al. (1997) model, with the addition of a lung compartment as described
by Fisher et al. (2000), should be used for the purposes of this assessment. The ability of the
Ward et al. (1997) mouse PBPK model to describe dosimetry in that species supports the
biological basis for this model structure; and hence, the expectation that it can be used to predict
dosimetry in humans. However, as mentioned previously, the mouse parameterized PBPK model
is not used in this assessment. See Appendix B for a more complete discussion of the selected
modeling approach and modeling considerations.
3.4.3.1. Available PK Data
Although limited human data are available, several studies exist that contain PK and
metabolic data in mice, rats, and nonhuman primates for model parameterization (Table 3-10).
3.4.3.2. Model Structure
As described in detail in Appendix B, a model was developed which includes
compartments for alveolar air/blood methanol exchange, liver, fat, bladder (human simulations)
and the rest of the body (Figure 3-5). This model is a revision of the model reported by Ward et
al. (1997), reflecting significant simplifications (removal of compartments for placenta,
embryo/fetus, and extra-embryonic fluid) and two elaborations (addition of a second GI lumen
compartment to the existing stomach lumen compartment and addition of a bladder
compartment), while maintaining the ability to describe methanol blood kinetics in rats and
humans. A fat compartment was included because it is the only tissue with a tissue:blood
partitioning coefficient appreciably different than 1, and the liver is included because it is the
primary site of metabolism. A bladder compartment was also added for use in simulating human
urinary excretion to capture the difference in kinetics between changes in blood methanol
concentration and urinary methanol concentration. The model code describes inhalation, oral,
and i.v. dose routes, and data exist from studies (Table 3-10) that were used to fit parameters and
evaluate model predictions for all three of those routes. In humans, inhalation exposure data an
i.v. study and a single short-duration oral PK study were available for model calibration and
validation.
3-29
-------
Table 3-10 Key methanol kinetic
i.v. dose
Reference (mg/kg)
Batterman & Franzblau
(1997)
Batterman et al. (1998)
Ernstgard et al. (2005)
Haffner et al. (1992) 10
Osterloh et al. (1996):
Chuwers et al. (1995):
D'Alessandro et al. (1994)
Schmutte et al. (1988)
Sedivec et al. (1981)
Burbacher et al. (2004a):
Burbacher et al. (2004b)
Medinsky et al. (1997):
Dorman et al. (1994)
Morton etal. (1992) ™°Jrats
Perkins etal. (1996,
1995a, b)
Pollack and Brouwer
(1996) 100-2,500
Pollack et al. (1993)
Ward etal. (1997) 100,500
Ward and Pollack (1996) (Rat)
Rogers and Mole (1997)
Rogers et al. (1993b)
studies for model validation.
Inhalation
(ppm)
800 (8 hr)
100(2hr)
200 (2hr)
200 (4 hr)
78-231 (8 hr)
0-1,800(2.5
hr, 4 mo)
10-900 (2 hr)
50-2,000
(6hr)
1,000-20,000
(8hr)
1,000-20,000
(8hr)
1,000-15,000
(7hr, 10 days)
Oral/
dermal/
i.p. Species
Dermal MaSemale
Human
Male/female
Human
Male/female
Human males
Human
Male/female
°ral .. Human
1.1 mg/L
Human
Male
Monkeys
Cynomolgus
Pregnant, NP
Monkeys
Cynomolgus
Folate deficient
Monkey Rhesus,
and Rat Fischer-
344
Mouse and Rat
Oral- Rat: sPrague-
100-2,500 Dawley,& Mouse;
mg/kg CD-1 Pregnant,
Mouse CD-1,
Oral: 2,500 GD18; Rat
mg/kg Sprague-Dawley,
GD14&GD20
Mouse CD-1
Pregnant
Samples
Blood
Blood,
Urine,
Exhaled
Blood,
Exhaled
Blood
Blood, Urine
Blood
Urine, Blood
Blood
Blood,
Urine,
Exhaled
Blood,
Urine,
Exhaled
Blood, Urine
Blood
Blood,
Conceptus
Blood
Digitized
figures3
Figure 1
Figure 1
Figure 1in
Osterloh et
al. (1996)
Figure 1
Figures 2,
3, 6, 7, 8
Figure 7
"Data obtained from the reported figure, from the corresponding reference.
The approach to model calibration and specific data sets used for Sprague-Dawley (S-D)
rats and humans are described in detail in Appendix B. The metabolism of methanol was
described using Michaelis-Menten kinetics. Simulated metabolic elimination of methanol is not
linked in the PBPK model to production of formaldehyde or formate; it is simply another route
of methanol elimination. Metabolism of formaldehyde (to formate) is not explicitly simulated by
3-30
-------
the model, and this model tracks neither formate nor formaldehyde. Since the metabolic
conversion of formaldehyde to formate is rapid (<1 minute) in all species (Kavet and Nauss,
1990), the rate of methanol metabolism may approximate a formate production rate, though this
has not been verified.
Inhalation Fracin
exposure
Exhaled
air
Bladder
(human only)
Urine
Metabolism
Oral
exposure
Bav
Endogenous
production
Note: Parameters: Fracin (FRACIN), fraction of exposure concentration reaching gas exchange region in lungs; Bav, oral
bioavailability; kas, first-order oral absorption rate from stomach; k^, first-order uptake from 2nd GI compartment; ksi, first-order
transfer between stomach and 2nd GI; Vmax and Km , apparent Michaelis-Menten rate constants for metabolism in liver; kb first-
order rate constant for urinary elimination; kH, rate constant for urinary excretion from bladder. For the rat only, high levels of
methanol in the body compartment lead to respiratory and cardiac depression, indicated by the dashed line. Rat data were
consistent with Bav = 100%, but humans with Bav = 83%.
Figure 3-5 Schematic of the PBPK model used to describe the inhalation, oral, and i.v.
route pharmacokinetics of methanol.
The primary purpose of this assessment is for the determination of noncancer risk
associated with exogenous oral or inhalation exposure to methanol that add to background levels
of methanol derived from a diet that includes fruits and vegetables. However, because
background methanol levels can impact model parameter estimation and internal dose
predictions, the PBPK models developed for this assessment incorporate a zero-order liver
infusion term for methanol designed to approximate reported background levels. The PBPK
model estimate of background levels was then subtracted for benchmark dose (BMD) modeling.
3-31
-------
For example, when the metric is blood AUC, BMD analysis used the PBPK-predicted difference,
AUC (exposed rats) -AUC (control rats), as the dose metric. In short the level of effect was
correlated with the internal dose above background in the test animal (PODAuc)- The human
PBPK model was then used to estimate the human equivalent oral dose (PODHEo) or inhalation
concentration (PODnEc) associated with this internal dose. To do this, the human PBPK model
used an upper bound background level (discussed in Section 5.3.6) of 2.5 mg/L (except when
study specific data were available during model calibration) and the PODHED or PODHEc was
selected such that the predicted increase in human blood levels over this background matched the
PODAUC.
3.4.3.3. Model Parameters
The EPAmethanol model uses a consistent set of physiological parameters obtained
predominantly from the open literature (Appendix B, Table B-l); the Ward et al. (1997) model
employed a number of data-set specific parameters.14 Parameters for blood flow, ventilation, and
metabolic capacity were scaled as a function of body weight raised to the 0.75 power, according
to the methods of Ramsey and Andersen (1984).The process by which the rat and human
inhalation and oral models were calibrated and analyzed for parameter sensitivity is discussed in
Appendix B, "Development, Calibration and Application of a Methanol PBPK Model." An
evaluation of the importance of selected parameters on the model estimates of blood methanol
was performed using the subroutines within acslX v2.3 (Aegis Technologies, Huntsville,
Alabama).
3.4.4. Monkey PK Data and Analysis
In order to estimate internal doses (blood Cmax and AUC values) for the monkey health-
effects study of Burbacher et al. (1999b) and further elucidate the potential differences in
methanol pharmacokinetics between NP and pregnant individuals (2nd and 3rd trimester), a
focused reanalysis of the data of Burbacher et al. (1999a) was performed. The monkeys in this
study were exposed for 2.5 hours/day, with the methanol concentration raised to approximately
the target concentration for the first 2 hours of each exposure and the last 30 minutes providing a
chamber "wash-out" period, when the exposure chamber concentration was allowed to drop to 0.
Blood samples were taken and analyzed for methanol concentration at 30 minutes, 1, 2, 3, 4, and
6 hours after removal from the chamber (or 1, 1.5, 2.5, 3.5, 4.5, and 6.5 hours after the end of
14 Some data sets provided in the Ward et al. (1997) model code were corrected to be consistent with figures in the
published literature describing the experimental data.
3-32
-------
active exposure). These data were analyzed to compare the PK in NP versus pregnant animals,
and fitted with a simple PK model to estimate 24-hour blood AUC values for each exposure
level. Details of this analysis are provided in Appendix B. The chamber concentrations for
"pregnancy" exposures recorded by Burbacher et al. (1999a: Table 2) and average body weights
for each exposure group at the 2nd trimester time point were used along with the model
described in Appendix B to calculate Cmax above background and 24-hour blood methanol AUC
above background (Table B-6) for the dose-response analysis of data from the Burbacher et al.
(1999a: 1999b) developmental study in monkeys described in Appendix D.
3.4.5. Summary and Conclusions
Rat and human versions of a methanol PBPK model have been developed and calibrated
to data available in the open literature. The model simplifies the structure used by Ward et
al.(1997), while adding specific refinements such as a standard lung compartment employed by
Fisher et al. (2000) and a two-compartment GI tract.
Although the developmental endpoints of concern are effects which occur following in
utero and (to a lesser extent) lactational exposure, no pregnancy-specific PBPK model exists for
methanol and limited data exists for the development and validation of a
fetal/gestational/conceptus compartment. The fact that the unique physiology of pregnancy and
the fetus/conceptus are not represented in a methanol model would be important if methanol
pharmacokinetics differed significantly during pregnancy or if the observed partitioning of
methanol into the fetus/conceptus versus the mother showed a concentration ratio significantly
greater than or less than 1. Methanol pharmacokinetics during GD6-GD10 in the mouse are not
different from NP mice (Pollack and Brouwer, 1996), and the maternal blood:fetus/conceptus
partition coefficient is reported to be near 1 (Ward et al., 1997; Horton et al., 1992). Maternal
blood kinetics in monkeys differs little from those in NP animals (see Section 3.2 for details).
Further, in both mice and monkeys, to the extent that late-pregnancy blood levels differ from NP
for a given exposure, they are higher; i.e., the difference between model predictions and actual
concentrations is in the same direction. These data support the assumption that the ratio of actual
target-tissue methanol concentration to (predicted) NP maternal blood concentrations will be
about the same across species, and hence, that using NP maternal blood levels in place of fetal
concentrations will not lead to a systematic error when extrapolating risks.
The critical gestational window for the reduced brain weight effect observed in the
NEDO (1987) rat study is broader than for the mouse cervical rib effect. In addition, NEDO
(1987) rats were exposed not only to methanol gestationally but also lactationally and via
inhalation after parturition. The findings in the mice and rats up to GD20 (similar blood
3-33
-------
methanol kinetics between NP and pregnant animals and a maternal blood:fetal partition
coefficient close to 1) are assumed to be applicable to the rat later in pregnancy. However, the
additional routes of exposure to the pups in this study present uncertainties (see additional
discussion in Sections 5.1.2.2 and 5.1.3.2.2) and suggest that average blood levels in pups might
be greater than those of the dam.
Methanol is transported directly from the maternal circulation to fetal circulation via the
placenta, but transfer via lactation involves distribution to the breast tissue, then milk, then
uptake from the pup's GI tract. Therefore blood or target-tissue levels in the breast-feeding infant
or rat pup are likely to differ more from maternal levels than do fetal levels. In addition, the
health-effects data indicate that most of the effects of concern are due to fetal exposure, with a
relatively small influence due to postnatal exposures. Therefore, it would be extremely difficult
to distinguish the contribution of postnatal exposure from prenatal exposure to a given effect in a
way that would allow the risk to be estimated from estimates of both exposure levels, even if one
had a lactation/child PBPK model that allowed for prediction of blood (or target-tissue) levels in
the offspring. Finally, one would still expect the target-tissue concentrations in the offspring to be
closely related to maternal blood levels (which depend on ambient exposure and determine the
amount delivered through breast milk), with the relationship between maternal levels and those
in the offspring being similar across species. Further, as discussed in Section 5.1.3.2.2, it is likely
that the difference in blood levels between rat pups and dams would be similar to the difference
between mothers and human offspring. Therefore, it is assumed that the potential differences
between pup and dam blood methanol levels do not have a significant impact on this assessment
and the estimation of HECs.
Therefore, the development of a lactation/child PBPK model is not necessary, given the
minimal change that is likely to result in risk extrapolations, and use of NP maternal blood levels
as a measure of risk in the offspring is considered preferable over use of default extrapolation
methods. In particular, the existing human data allow for predictions of maternal blood levels,
which depend strongly on the rate of maternal methanol clearance. Since bottle-fed infants do
not receive methanol from their mothers, they are expected to have lower or, at most, similar
overall exposures for a given ambient concentration than the breast-fed infant, so that use of
maternal blood levels for risk estimation should also be adequately protective for that group.
The final rat and human methanol PBPK models fit multiple data sets for inhalation, oral,
and i.v. (rat only) exposures, using consistent parameters that are representative of each species
but are not varied within species or by dose or source of data. Also, a simple PK model calibrated
to early gestation monkey data, which were shown to be essentially indistinguishable from NP
and late-gestation pregnant monkey PK data, was used to estimate blood methanol peak
concentrations (internal doses) in that species. The models are used to estimate chronic human
3-34
-------
exposure concentrations from internal dose metrics for use in the RfC and RfD derivations
discussed in Section 5.
3-35
-------
4.HAZARD IDENTIFICATION
4.1. Studies in Humans - Case Reports, Occupational and Controlled
Studies
4.1.1. Case Reports
An extensive library of case reports has documented the consequences of acute
accidental/intentional methanol poisoning. Nearly all have involved ingestion, but a few have
involved percutaneous and/or inhalation exposure.
As many of the case reports demonstrate, the association of Parkinson-like symptoms
with methanol poisoning is related to the observation that lesions in the putamen are a common
feature both in Parkinson's disease and methanol overexposure. These lesions are commonly
identified using Computed Tomography (CT) or by Magnetic Resonance Imaging (MRI). Other
areas of the brain (e.g., the cerebrum, cerebellum, and corpus callosum) also have been shown to
be adversely affected by methanol overexposure. The associated effects are further discussed in
Appendix C, Human Case Studies.
Various therapeutic procedures [e.g., infusion of ADH1 inhibitors ethanol or fomepizole
(4-methylpyrazole)], sodium bicarbonate or folic acid administration, and hemodialysis) have
been used in many of these methanol overexposures, and the reader is referred to the specific
case reports for details in this regard (see Appendix C). The reader also is referred to Kraut and
Kurtz (2008) and Barceloux et al. (2002) for a more in-depth discussion of the treatments in
relation to clinical features of methanol toxicity.
Most cases of accidental/intentional methanol poisoning reveal a common set of
symptoms, many of which are likely to be presented upon hospital admission. These include:
• blurred vision and bilateral or unilateral blindness
• convulsions, tremors, and coma
• nausea, headache, and dizziness
• abdominal pain
• diminished motor skills
• acidosis
4-1
-------
• dyspnea
• behavioral and/or emotional deficits
• speech impediments
Acute symptoms generally are nausea, dizziness, and headache. In the case reports cited
in Appendix C, the onset of symptom sets as well as their severity varies depending upon how
much methanol was ingested, whether or not and when appropriate treatment was administered,
and individual variability. A longer time between exposure and treatment, with few exceptions,
results in more severe outcomes (e.g., convulsions, coma, blindness, and death). The diminution
of some acute and/or delayed symptoms may reflect concomitant ingestion of ethanol or how
quickly therapeutic measures (one of which includes ethanol infusion) were administered in the
hospital setting.
Those individuals who are in a metabolic acidotic state (e.g., pH <7.0) are typically the
individuals who manifest the more severe symptoms. Many case reports stress that, unlike blood
pH levels <7.0, blood levels of methanol are not particularly good predictors of health outcome.
According to a publication of the American Academy of Clinical Toxicology (Barceloux et al.,
2002), "the degree of acidosis at presentation most consistently correlates with severity and
outcome."
Table 4-1 Mortality rate for subjects exposed to methanol-tainted whisky in relation to
their level of acidosis.
Subjects"
All patients
Acidotic (CO2 <20 mEq)
Severely acidotic (CO2 <10 mEq)
Number
323
115
30
Percent deaths
6.2
19
50
"These data do not include those who died outside the hospital or who were moribund on arrival.
Source: Reprinted with permission of Lippincott, Williams & Wilkins; Bennett et al. (1953).
As the case reports (Appendix C) demonstrate, those individuals who present with more
severe symptoms (e.g., coma, seizures, and severe acidosis) generally exhibit higher mortality
(even after treatment) than those without such symptoms. In survivors of poisoning, persistence
or permanence of vision decrements and particularly blindness often have been observed.
Because of the strong correlation between outcomes of methanol poisoning with severity of
acidosis (e.g., Table 4-1), formate is usually assumed to be the proximal cause of the acute
effects of methanol. Most of the symptoms of methanol poisoning (listed in the individual
4-2
-------
studies in Appendix C) are common to the several other types of metabolic acidosis (Berkow and
Fletcher, 1992). It has been postulated that formaldehyde may be the toxic moiety for the
symptoms of methanol poisoning that are seemingly distinct from acidotic symptoms (Hayasaka
et al., 2001). However, the predominant role of formic acid as the major metabolic agent for
methanol ocular toxicity has been demonstrated in monkeys, who experienced ocular toxicity
following methanol exposure that was essentially identical to that produced in monkeys exposed
to formate (McMartin et al., 1979). Since formaldehyde has a very short half life, it is unlikely to
be distributed from the liver to the brain or eye fast enough to cause CNS or ocular damage.
Nevertheless, methanol is distributed to multiple organ systems and there is evidence that it can
be metabolized to formaldehye in situ by other organ systems, including studies that have found
ADH activity in non-liver cells (Jelski et al., 2006; Motavkin et al., 1988; BuhleretaL 1983)
and a rat study that reports dose-dependent increases of formaldehyde DNA adducts derived
from exogenous methanol exposure in multiple tissues such as liver, lung, spleen, thymus, bone
marrow, kidney, and WBC (exogenous adduct levels were less than 10% of endogenous adduct
levels for most organ systems) (Luetal., 2012).
Correlation of symptomatology with blood levels of methanol has been shown to vary
appreciably between individuals. Blood methanol levels in the case reports involving ingestion
ranged from values of 300 to over 10,000 mg/L. The lowest value (200 mg/L) reported (Adanir
et al., 2005) involved a case of percutaneous absorption (with perhaps associated inhalation
exposure) that led to vision and CNS deficits after hospital discharge. In one case report
(Rubinstein et al., 1995) involving ingestion, coma and subsequent death were associated with an
initial blood methanol level of 360 mg/L.
Upon MRI and CT scans, the more seriously affected individuals typically have focal
necrosis in both brain white matter and more commonly, in the putamen. Bilateral hemorrhagic
and nonhemorrhagic necrosis of the putamen is considered by many radiologists as the most
well-known sequelae of methanol overexposure.
4.1.2. Occupational Studies
Occupational health studies have been carried out to investigate the potential effects of
chronic exposure to lower levels of methanol than those seen in acute poisoning cases such as
those described in Appendix C. For example, Frederick et al. (1984) conducted a health hazard
evaluation on behalf of the National Institute for Occupational Safety and Health (NIOSH) to
determine if vapor from duplicating fluid (which contains 99% methanol) used in mimeograph
duplicating machines caused adverse health effects in exposed persons. A group of 84 teacher's
aides were selected for study, 66 of whom responded with a completed medical questionnaire. A
4-3
-------
group of 297 teachers (who were not exposed to methanol vapors to the same extent as the
teacher's aides) completed questionnaires as a control group. A 15-minute breathing zone sample
was taken from 21 duplicators, 15 of which were greater than the NIOSH-recommended short
term ceiling concentration of 800 ppm (1,048 mg/m3). The highest breathing zone concentrations
were in the vicinity of duplicators for which no exhaust ventilation had been provided
(3,080 ppm [4,036 mg/m3] was the highest value recorded). Upon comparison of the self-
described symptoms of the 66 teacher's aides with those of 66 age-matched teachers chosen from
the 297 who responded, the number of symptoms (potentially related to methanol) was
significantly higher in the teacher's aides. These included blurred vision (22.7 versus 1.5%),
headache (34.8 versus 18.1%), dizziness (30.3 versus 1.5%), and nausea (18 versus 6%). By
contrast, symptoms that are not usually associated with methanol exposure (painful urination,
diarrhea, poor appetite, and jaundice) were similar in incidence among the groups.
To further investigate these disparities, NIOSH physicians (not involved in the study)
defined a hypothetical case of methanol toxicity by any of the following four symptom
aggregations: (1) visual changes; (2) one acute symptom (headache, dizziness, numbness,
giddiness, nausea or vomiting) combined with one chronic symptom (unusual fatigue, muscle
weakness, trouble sleeping, irritability, or poor memory); (3) two acute symptoms; or (4) three
chronic symptoms. By these criteria, 45% of the teacher's aides were classified as being
adversely affected by methanol exposure compared to 24% of teachers (p < 0.025). Those
teacher's aides and teachers who spent a greater amount of time using the duplicators were
affected at a higher rate than those who used the machines for a lower percentage of their work
day.
Tanner (1992) reviewed the occupational and environmental causes of Parkinson!sm,
spotlighting the potential etiological significance of manganese, carbon monoxide, repeated head
trauma (such as suffered by boxers), and exposure to solvents. Among the latter, Tanner (1992)
discussed the effects of methanol and n-hexane on the nervous system. Acute methanol
intoxication resulted in inebriation, followed within hours by GI pain, delirium, and coma.
Tanner (1992) pinpointed the formation of formic acid, with consequent inhibition of
cytochrome oxidase, impaired mitochondrial function, and decreased ATP formation as relevant
biochemical and physiological changes for methanol exposure. Nervous system injury usually
includes blindness, Parkinson-like symptoms, dystonia, and cognitive impairment, with injury to
putaminal neurons most likely underlying the neurological responses.
Kawai et al. (1991) carried out a biomarker study in which 33 occupationally exposed
workers in a factory making methanol fuel were exposed to concentrations of methanol of up to
3,577 ppm (4,687 mg/m3), as measured by personal samplers of breathing zone air. Breathing
zone exposure samples were correlated with the concentrations of methanol in urine at the end of
4-4
-------
the shift in 38 exposed individuals and 30 controls (r = 0.82). Eleven of 22 individuals who
experienced high exposure to methanol (geometric mean of 459 ppm [601 mg/m3]) complained
of dimmed vision during work while 32% of this group of workers experienced nasal irritation.
These incidences were statistically significant compared to those of persons who worked in low-
exposure conditions (geometric mean of 31 ppm [41 mg/m3]). One 38-year-old female worker
who had worked at the factory for only 4 months reported that her visual acuity had undergone a
gradual impairment. She also displayed a delayed light reflex.
Lorente et al. (2000) carried out a case control study of 100 mothers whose babies had
been born with cleft palates. Since all of the mothers had worked during the first trimester,
Lorente et al. (2000) examined the occupational information for each subject in comparison to
751 mothers whose babies were healthy. Industrial hygienists analyzed the work histories of all
subjects to determine what, if any, chemicals the affected mothers may have been exposed to
during pregnancy. Multivariate analysis was used to calculate odds ratios, with adjustments made
for center of recruitment, maternal age, urbanization, socioeconomic status, and country of
origin. Occupations with positive outcomes for cleft palate in the progeny were hairdressing
(OR = 5.1, with a 95% confidence interval [CI] of 1.0-26) and housekeeping (OR = 2.8, with a
95% CI of 1.1-7.2). Odds ratios for cleft palate only and cleft lip with or without cleft palate
were calculated for 96 chemicals. There seemed to be no consistent pattern of association for any
chemical or group of chemicals with these impairments, and possible exposure to methanol was
negative for both outcomes.
4.1.3. Controlled Human Studies
Two controlled studies have evaluated humans for neurobehavioral function following
exposure to -200 ppm (262 mg/m3) methanol vapors in a controlled setting. The occupational
TLV established by the American Conference of Governmental Industrial Hygienists (2000) is
200 ppm (262 mg/m3). In a pilot study by Cook et al. (1991), 12 healthy young men (22-32 years
of age) served as their own controls and were tested for neurobehavioral function following a
random acute exposure to air or 191 ppm (250 mg/m3) methanol vapors for 75 minutes. The
majority of results in a battery of neurobehavioral endpoints were negative. However, statistical
significance was obtained for results in the P-200 and N1-P2 component of event-related
potentials (brain wave patterns following light flashes and sounds), the Sternberg memory task,
and subjective evaluations of concentration and fatigue. As noted by the Cook et al.(1991),
effects were mild and within normal ranges. Cook et al. (1991) acknowledged limitations in their
study design, such as small sample size, exposure to only one concentration for a single duration
4-5
-------
time, and difficulties in masking the methanol odor from experimental personnel and study
subjects.
In a randomized double-blind study, neurobehavioral testing was conducted on 15 men
and 11 women (healthy, aged 26-51 years) following exposure to 200 ppm (262 mg/m3)
methanol or water vapors for 4 hours (Chuwers et al., 1995): subjects served as their own
controls in this study. Exposure resulted in elevated blood and urine methanol levels (up to peak
levels of 6.5 mg/L and 0.9 mg/L, respectively) but not formate concentrations. The majority of
study results were negative. No significant findings were noted for visual, neurophysiological, or
neurobehavioral tests except for slight effects (p < 0.05) on P-300 amplitude (brain waves
following exposure to sensory stimuli) and Symbol Digit testing (ability to process information
and psychomotor skills). Neurobehavioral performance was minimally affected by methanol
exposure at this level. Limitations noted by Chuwers et al. (1995) are that studies of alcohol's
affect on P-300 amplitude suggest that this endpoint may be biased by unknown factors and
some experimenters and subjects correctly guessed if methanol was used.
Although the slight changes in P-200 and P-300 amplitude noted in both the Chuwers et
al. (1995) and Cook et al. (1991) studies may be an indication of moderate alterations in
cognitive function, the results of these studies are generally consistent and suggest that the
exposure concentrations employed were below the threshold for substantial neurological effects.
This is consistent with the data from acute poisoning events which have pointed to a serum
methanol threshold of 200 mg/L for the instigation of acidosis, visual impairment, and CNS
deficits.
Mann et al. (2002) studied the effects of methanol exposure on human respiratory
epithelium as manifested by local irritation, ciliary function, and immunological factors. Twelve
healthy men (average age 26.8 years) were exposed to 20 and 200 ppm (26.2 and 262 mg/m3,
respectively) methanol for 4 hours at each concentration; exposures were separated by 1-week
intervals. The 20 ppm (26.2 mg/m3) concentration was considered to be the control exposure
since previous studies had demonstrated that subjects can detect methanol concentrations of
20 ppm (26.2 mg/m3) and greater. Following each single exposure, subclinical inflammation was
assessed by measuring concentrations of interleukins (IL-8, IL-lp, and IL-6) and prostaglandin
E2 in nasal secretions. Mucociliary clearance was evaluated by conducting a saccharin transport
time test and measuring ciliary beat frequency. Interleukin and prostaglandin data were evaluated
by a 1-tailed Wilcoxon test, and ciliary function data were assessed by a 2-tailed Wilcoxon test.
Exposure to 200 (262 mg/m3) versus 20 ppm (26.2 mg/m3) methanol resulted in a statistically-
significant increase in IL-lp (median of 21.4 versus 8.3 pg/mL) and IL-8 (median of 424 versus
356 pg/mL). There were no significant effects on IL-6 and prostaglandin E2 concentration,
ciliary function, or on the self-reported incidence of subjective symptoms of irritation. The
4-6
-------
authors concluded that exposure to 200 ppm (262 mg/m3) methanol resulted in a subclinical
inflammatory response.
In summary, adult human subjects acutely exposed to 200 ppm (262 mg/m3) methanol
have experienced slight neurological (Chuwers et al., 1995) and immunological effects
(increased subclinical biomarkers for inflammation) with no self-reported symptoms of irritation
(Mann et al., 2002). These exposure levels were associated with peak methanol blood levels of
6.5 mg/L (Chuwers et al., 1995). Nasal irritation effects have been reported by adult workers
exposed to 459 ppm (601 mg/m3) methanol (Kawai et al., 1991). Frank effects such as blurred
vision, bilateral or unilateral blindness, coma, convulsions/tremors, nausea, headache, abdominal
pain, diminished motor skills, acidosis, and dyspnea begin to occur as blood levels approach
200 mg methanol/L, while 800 mg/L appears to be the threshold for lethality. Data for
subchronic, chronic or in utero human exposures are very limited and inconclusive.
4.2. Acute, Subchronic and Chronic Studies in Animals - Oral and
Inhalation
A number of studies in animals have investigated the acute, subchronic, and chronic
toxicity of methanol. Most are via the inhalation route. Presented below are summaries of the
noncancer effects reported in these bioassays. Carcinogenic effects are not described, evaluated,
or discussed in this assessment.
4.2.1. Oral Studies
4.2.1.1. Acute Toxicity
Although there are few studies that have examined the short-term toxic effects of
methanol via the oral route, a number of median lethal dose (LD50) values have been published
for the compound. As listed in Lewis (1992), these include 5,628 mg/kg in rats, 7,300 mg/kg in
mice, and 7,000 mg/kg in monkeys.
4.2.1.2. Subchronic Toxicity
An oral repeat dose study was conducted by the U.S. EPA (TRL, 1986) in rats. Sprague-
Dawley rats (30/sex/dose) at no less than 30 days of age were gavaged with 0, 100, 500, or
2,500 mg/kg-day of methanol. Six weeks after dosing, 10 rats/sex/dose group were subjected to
interim sacrifice, while the remaining rats continued on the dosing regimen until the final
4-7
-------
sacrifice (90 days). This study generated data on weekly body weights and food consumption,
clinical signs of toxicity, ophthalmologic evaluations, mortality, blood and urine chemistry (from
a comprehensive set of hematology, serum chemistry, and urinalysis tests), and gross and
microscopic evaluations for all test animals. Complete histopathologic examinations of over
30 organ tissues were done on the control and high-dose rats. Histopathologic examinations of
livers, hearts, and kidneys and all gross lesions seen at necropsy were done on low-dose and mid-
dose rats. There were no differences between dosed animals and controls in body weight gain,
food consumption, or upon gross or microscopic evaluations. Elevated levels (p < 0.05 in males)
of serum alanine transaminase (ALT)15 and serum alkaline phosphatase (SAP), and increased
(but not statistically significant) liver weights in both male and female rats suggest possible
treatment-related effects in rats bolus dosed with 2,500 mg methanol/kg-day despite the absence
of supportive histopathologic lesions in the liver. Brain weights of high-dose group
(2,500 mg/kg-day) males and females were significantly less than those of the control group at
terminal sacrifice. The only histopathology noted was a higher incidence of colloid in the
hypophyseal cleft of the pituitary gland in the high-dose versus control group males
(13/20 versus 0/20) and females (9/20 versus 3/20). Based on these findings, 500 mg/kg-day of
methanol is considered an NOAEL from this rat study.
4.2.1.3. Chronic Noncancer Toxicity
A report by Soffritti et al. (2002) summarized a European Ramazzini Foundation (ERF)
chronic duration experimental study of methanol16 in which the compound was provided to
100 Sprague-Dawley rats/sex/group ad libitum in drinking water at concentrations of 0, 500,
5,000, and 20,000 ppm (v/v). The animals were 8 weeks old at the beginning of the study. In
general, ERF does not randomly assign animals to treatment groups, but assigns all animals from
a given litter to the same treatment group (Bucher, 2002). All rats were exposed for up to
104 weeks, and then maintained until they died naturally. Rats were housed in groups of 5 in
Makrolon cages (41 x 25 x 15 cm) in a room that was maintained at 23 ± 2°C and 50-60%
relative humidity. The in-life portion of the experiment ended at 153 weeks with the death of the
last animal. Mean daily drinking water, food consumption, and body weights were monitored
weekly for the first 13 weeks, every 2 weeks thereafter for 104 weeks, then every 8 weeks until
the end of the experiment. Clinical signs were monitored 3 times/day, and the occurrence of
15 Also known as serum glutamate pyruvate transaminase (SGPT)
16 Soffritti et al. (2002) report that methanol was obtained from J.T. Baker, Deventer, Holland, purity grade 99.8%.
4-8
-------
gross changes was evaluated every 2 weeks. All rats were necropsied at death then underwent
histopathologic examination of organs and tissues.17
Soffritti et al. (2002) reported no substantial dose-related differences in survival, but no
data were provided. Using individual animal data available from the ERF website,18 Cruzan
(2009) reports that male rats treated with methanol generally survived better than controls, with
50% survival occurring at day 629, 686, 639 and 701 in the 0, 500, 5,000, and 20, 000 mg/L
groups, respectively. There were no significant differences in survival between female control
and treatment groups, with 50% survival occurring at day 717, 691, 678 and 708 in the 0, 500,
5,000, and 20,000 mg/L groups, respectively. Body weight and water and food consumption
were monitored in the study, but the data were not documented in the published report.
Soffritti et al. (2002) reported that water consumption in high-dose females was reduced
compared to controls between 8 and 56 weeks and that the mean body weight in high-dose males
tended to be higher than that of control males. Overall, there was no pattern of compound-related
clinical signs of toxicity, and the available data did not provide any indication that the control
group was not concurrent with the treated group (Cruzan, 2009). Soffritti et al. (2002) further
reported that there were no compound-related signs of gross pathology or histopathologic lesions
indicative of noncancer toxicological effects in response to methanol.
Apaja (1980) performed dermal and drinking water chronic bioassays in which male and
female Eppley Swiss Webster mice (25/sex/dose group; 8 weeks old at study initiation) were
exposed 6 days per week until natural death to various concentrations of malonaldehyde and
methanol. The stated purpose of the study was to determine the carcinogen!city of
malonaldehyde, a product of oxidative lipid deterioration in rancid beef and other food products
in advanced stages of degradation. However, due to its instability, malonaldehyde was obtained
from the more stable malonaldehyde bis (dimethylacetal), which was hydrolyzed to
malonaldehyde and methanol in dilute aqueous solutions in the presence of a strong mineral acid.
In the drinking water portion of this study, mice were exposed to 3 different concentrations of the
malonaldehyde/methanol solution and three different control solutions of methanol alone,
0.222%, 0.444% and 0.889% methanol in drinking water (222, 444 and 889 ppm, assuming a
density of 1 g/mL), corresponding to the stoichiometric amount of methanol liberated by
hydrolysis of the acetal in the three test solutions. The methanol was described as Mallinckrodt
17 Histopathology was performed on the following organs and tissues: skin and subcutaneous tissue, brain, pituitary
gland, Zymbal glands, parotid glands, submaxillary glands, Harderian glands, cranium (with oral and nasal cavities
and external and internal ear ducts) (5 sections of head), tongue, thyroid and parathyroid, pharynx, larynx, thymus
and mediastinal lymph nodes, trachea, lung and mainstem bronchi, heart, diaphragm, liver, spleen, pancreas,
kidneys, adrenal glands, esophagus, stomach (fore and glandular), intestine (four levels), urinary bladder, prostate,
gonads, interscapular fat pad, subcutaneous and mesenteric lymph nodes, and any other organs or tissues with
pathologic lesions.
18 http://www.ramazzini.it/fondazione/foundation.asp.
4-9
-------
analytical grade. No unexposed control groups were included in these studies. However, the
author provided pathology data from historical records of untreated Swiss mice of the Eppley
colony used in two separate chronic studies, one involving 100 untreated males and 100
untreated females (Toth etal., 1977) and the other involving 100 untreated females
histopathological analyzed by Apaja (Apaja, 1980).
Mice in the Apaja (1980) study were housed five/plastic cage and fed Wayne Lab-Blox
pelleted diet. Water was available ad libitum throughout life. Liquid consumption per animal was
measured at 3 times/week. The methanol dose in the dermal study (females only) was 21.3 mg
(532 mg/kg-day using an average weight of 0.04 kg as approximated from Figure 4 of the study),
three times/week. The methanol doses in the drinking water study were reported as 22.6, 40.8
and 84.5 mg/day (560, 1,000 and 2,100 mg/kg-day using an average weight of 0.04 kg as
approximated from Figures 14-16 of the study) for females, and 24.6, 43.5 and 82.7 mg/day
(550, 970, and 1,800 mg/kg-day using an average weight of 0.045 kg as approximated from
Figures 14-16 of the study) for males, 6 days/week. The animals were checked daily and body
weights were monitored weekly. The in-life portion of the experiment ended at 120 weeks with
the death of the last animal. Like the Soffritti et al. (2002) study, test animals were sacrificed and
necropsied when moribund.19
The authors reported that survival of the methanol exposed females of the drinking water
study was lower than untreated historical controls (p < 0.05), but no significant differences in
survival was noted for males. An increase in liver parenchymal cell necrosis was reported in the
male and female high-dose groups, with the incidence in females (8%) being significant
(p < 0.01) relative to untreated historical controls. Incidence of acute pancreatitis was higher in
high-dose males (p <0.001), but did not appear to be dose-related in females, increasing at the
mid- (p <0.0001) and low-doses (p <0.01) when compared to historical controls but not
appearing at all in the high-dose females. Significant increases relative to untreated historical
controls were noted in amyloidosis of the spleen, nephropathy and pneumonia, but the increases
did not appear to be dose related.
19 The following tissues were fixed in 10% formalin (pH 7.5), embedded in paraffin, sectioned, stained routinely
with hematoxylin and eosin (special stains used as needed) and histologically-evaluated: skin, lungs, liver spleen,
pancreas, kidneys, adrenal glands, esophagus, stomach, small and large intestines, rectum, urinary bladder, uterus
and ovaries or testes, prostate glands and tumors or other gross pathological lesions.
4-10
-------
4.2.2. Inhalation Studies
4.2.2.1. Acute Toxicity
Lewis (1992) reported a 4-hour median lethal concentration (LCso) for methanol in rats of
64,000 ppm (83,867 mg/m3).
Japan's NEDO sponsored a series of toxicological tests on monkeys (M. fascicularis),
rats, and mice, using inhalation exposure.20 These are unpublished studies; accordingly, they
were externally peer reviewed (ERG, 2009).21A short-term exposure study evaluated monkeys
(sex unspecified) exposed to 3,000 ppm (3,931 mg/m3), 21 hours/day for 20 days (1 animal),
5,000 ppm (6,552 mg/m3) for 5 days (1 animal), 5,000 ppm (6,552 mg/m3) for 14 days (2
animals), and 7,000 and 10,000 ppm (9,173 and 13,104 mg/m3, respectively) for up to 6 days
(1 animal at each exposure level) (NEDO, 1987). Most of the experimental findings were
discussed descriptively in the report, without specifying the extent of change for any of the
effects in comparison to seven concurrent controls. However, the available data indicate that
clinical signs of toxicity were apparent in animals exposed to 5,000 ppm (all exposure durations)
or higher concentrations of methanol. These included reduced movement, crouching, weak
knees, involuntary movements of hands, dyspnea, and vomiting. In the discussion section of the
summary report, the authors stated that there was a sharp increase in the blood levels of methanol
and formic acid in monkey exposed to >3,000 ppm (3,931 mg/m3) methanol. They reported that
methanol and formic acid concentrations in the blood of monkeys exposed to 3,000 ppm or less
were 80 mg/L and 30 mg/L, respectively.22 In contrast, monkeys exposed to 5,000 ppm or higher
concentrations of methanol had blood methanol and formic acid concentrations of 5,250 mg/L
and 1,210 mg/L, respectively. Monkeys exposed to 7,000 ppm and 10,000 ppm became critically
ill and had to be sacrificed prematurely. Food intake was said to be little affected at 3,000 ppm,
but those exposed to 5,000 ppm or more showed a marked reduction. Clinically, the monkeys
exposed to 5,000 ppm or more exhibited reduced movement, weak knees, and involuntary
movement of upper extremities, eventually losing consciousness and dying.
20 In their bioassays, NEDO (NEDO. 1987) used inbred rats of the F344 or Sprague-Dawley strain, inbred mice of
the B6C3F1 strain and wild-caught M. fascicularis monkeys imported from Indonesia. The possibility of disease
among wild-caught animals is a concern, but NEDO (NEDO. 1987) state that the monkeys were initially
quarantined for 9 weeks and measures were taken throughout the studies against the transmission of pathogens for
infectious diseases. The authors indicated that "no infectious disease was observed in monkeys" and that "subjects
were healthy throughout the experiment."
21 An external peer review (ERG. 2009) was conducted for EPA in 2009 to evaluate the accuracy of experimental
procedures, results, and interpretation and discussion of the findings presented in these study reports.
22 Note that Burbacher et al. (1999a) and Burbacher et al. (20Q4a) measured blood levels of methanol and formic
acid in control monkeys of 2.4 mg/L and 8.7 mg/L, respectively (see Table 3-3).
4-11
-------
There were no significant changes in growth, with the exception of animals exposed to
the highest concentration, where body weight was reduced by 13%. There were few compound-
related changes in hematological or clinical chemistry effects, although animals exposed to 7,000
and 10,000 ppm showed an increase in white blood cells. A marked change in blood pH values at
the 7,000 ppm and 10,000 ppm levels (values not reported) was attributed to acidosis due to
accumulation of formic acid. The authors reported that no clinical or histopathological effects of
the visual system were apparent, but that exposure to 3,000 ppm (3,931 mg/m3) or more caused
dose-dependent fatty degeneration of the liver, and exposure to 5,000 ppm (6,552 mg/m3) or
more caused vacuolar degeneration of the kidneys, centered on the proximal uriniferous tubules.
A range of histopathologic changes to the CNS was apparently related to treatment. Severity of
the effects was increased with exposure concentration. Lesions included characteristic
degeneration of the bilateral putamen, caudate nucleus, and claustrum, with associated edema in
the cerebral white matter. CNS effects reported in this and the NEDO chronic monkey inhalation
study are discussed in greater detail in Section 4.4.2 "Inhalation Neurotoxicity Studies."
The NEDO (1987) studies in nonhuman primates, including the chronic study discussed
below, have multiple deficiencies that make them difficult to interpret. The reports lack a full
description of the materials and methods and raw data from the experiments. The data gaps
(e.g., materials and methods, statistical methods, data) are profound and the group sizes are too
small to support rigorous statistical analysis. At best, they provide a descriptive, rather than
quantitative, evaluation of the inhalation toxicity of methanol (ERG, 2009).
4.2.2.2. Subchronic Toxicity
A number of experimental studies have examined the effects of subchronic exposure to
methanol via inhalation. For example, Sayers et al. (1944) employed a protocol in which 2 male
dogs were repeatedly exposed (8 times daily for 3 minutes/exposure) to 10,000 ppm
(13,104 mg/m3) methanol for 100 days. One of the dogs was observed for a further 5 days before
sacrifice; the other dog was observed for 41 days postexposure. There were no clinical signs of
toxicity, and both gained weight during the study period. Blood samples were drawn on a regular
basis to monitor hematological parameters, but few if any compound-related changes were
observed. Ophthalmoscopic examination showed no incipient anomalies at any point during the
study period. Median blood concentrations of methanol were 65 mg/L (range 0-280 mg/L) for
one dog, and 140 mg/L (70-320 mg/L) for the other.
White et al. (1983) exposed 4 male Sprague-Dawley rats/group, 6 hours/day, 5 days/week
to 0, 200, 2,000, or 10,000 ppm (0, 262, 2,621, and 13,104 mg/m3) methanol for periods of 1, 2,
4, and 6 weeks. Additional groups of 6-week-exposure animals were granted a 6-week
4-12
-------
postexposure recovery period prior to sacrifice. The lungs were excised intact and lavaged
6 times with known volumes of physiological saline. The lavage supernatant was then assayed
for lactate dehydrogenase (LDH) and TV-acetyl-^-D-glucosaminidase (/?-NAG) activities. Other
parameters monitored in relation to methanol exposure included absolute and relative lung
weights, lung DNA content, protein, acid RNase and acid protease, pulmonary surfactant,
number of free cells in lavage/unit lung weight, surface protein, LDH, and/?-NAG. As discussed
by the authors, none of the monitored parameters showed significant changes in response to
methanol exposure.
Andrews et al. (1987) carried out a study of methanol inhalation in five Sprague-Dawley
rats/sex/group and three M. fascicularis monkeys/sex/group, 6 hours/day, 5 days/week, to 0, 500,
2,000, or, 5,000 ppm (0, 660, 2,620, and 6,552 mg/m3) methanol for 4 weeks. Clinical signs were
monitored twice daily, and all animals were given a physical examination once a week. Body
weights were monitored weekly, and animals received an ophthalmoscopic examination before
the start of the experiment and at term. Animals were sacrificed at term by exsanguination
following i.v. barbiturate administration. A gross necropsy was performed, weights of the major
organs were recorded, and tissues and organs taken for histopathologic examination. As
described by the authors, all animals survived to term with no clinical signs of toxicity among
the monkeys and only a few signs of irritation to the eyes and nose among the rats. In the latter
case, instances of mucoid nasal discharges appeared to be dose related. There were no
differences in body weight gain among the groups of either rats or monkeys, and overall,
absolute and relative organ weights were similar to controls. The only exception to this was a
decrease in the absolute adrenal weight of female high-concentration monkeys and an increase in
the relative spleen weight of mid-concentration female rats. These changes were not considered
by the authors to have biological significance. For both rats and monkeys, there were no
compound-related changes in gross pathology, histopathology, or ophthalmoscopy. These data
suggest a NOAEL of 5,000 ppm (6,600 mg/m3) for Sprague-Dawley rats and monkeys under the
conditions of the experiment.
Two studies by Poon et al. (1995; 1994) examined the effects of methanol on Sprague-
Dawley rats, when inhaled for 4 weeks. The effects of methanol were evaluated in comparison to
those of toluene and toluene/methanol mixtures (Poon et al., 1994), and to gasoline and
gasoline/methanol mixtures (Poon etal., 1995). In the first case (Poon etal., 1994), 10 Sprague-
Dawley rats/sex/group were exposed via inhalation, 6 hours/day, 5 days/week to 0, 300, or
3,000 ppm (0, 393, 3,930 mg/m3) methanol for 4 weeks. Clinical signs were monitored daily, and
food consumption and body weight gain were monitored weekly. Blood was taken at term for
hematological and clinical chemistry determinations. Weights of the major organs were recorded
at necropsy, and histopathologic examinations were carried out. A 10,000 xg Hver supernatant
4-13
-------
was prepared from each animal to measure aniline hydroxylase, aminoantipyrine N-demethylase,
and ethoxyresorufin-O-deethylase activities. For the most part, the responses to methanol alone
in this experiment were unremarkable. All animals survived to term, and there were no clinical
signs of toxicity among the groups. Body weight gain and food consumption did not differ from
controls, and there were no compound-related effects in hematological or clinical chemistry
parameters or in hepatic mixed function oxidase activities. However, the authors described a
reduction in the size of thyroid follicles that was more obvious in female than male rats. The
authors considered this effect to possibly have been compound related, although the incidence of
this feature for the 0, 300, and 3,000 ppm-receiving females was 0/6, 2/6, and 2/6, respectively.
The second experimental report by Poon et al. (1995) involved the exposure of
15 Sprague-Dawley rats/sex/group, 6 hours/day, 5 days/week for 4 weeks to 0 or 2,500 ppm
(0 and 3,276 mg/m3) to methanol as part of a study on the toxicological interactions of methanol
and gasoline. Many of the toxicological parameters examined were the same as those described
in Poon et al. (1994) study. However, in this study urinalysis featured the determination of
ascorbic and hippuric acids. Additionally, at term, the lungs and tracheae were excised and
aspirated with buffer to yield bronchoalveolar lavage fluid that was analyzed for ascorbic acid,
protein, and the activities of gamma-glutamyl transferase (y-GT), AP and LDH. Few if any of the
monitored parameters showed any differences between controls and those animals exposed to
methanol alone. However, two male rats had collapsed right eyes, and there was a reduction in
relative spleen weight in females exposed to methanol. Histopathologic changes in methanol-
receiving animals included mild panlobular vacuolation of the liver in females and some mild
changes to the upper respiratory tract, including mucous cell metaplasia. The incidence of the
latter effect, though higher, was not significantly different than controls in rats exposed to
2,500 ppm (3,267 mg/m3) methanol. However, there were also signs of an increased severity of
the effect in the presence of the solvent. No histopathologic changes were seen in the lungs or
lower respiratory tract of rats exposed to methanol alone.
4.2.2.3. Chronic Noncancer Toxicity
Information on the chronic noncancer toxicity of inhalation exposure to methanol has
come from NEDO (1987) which includes the results of experiments on (1) monkeys exposed for
up to 3 years, (2) rats and mice exposed for 12 months, (3) mice exposed for 18 months, and
(4) rats exposed for 2 years. These are unpublished studies; accordingly, they were externally
4-14
-------
peer reviewed (ERG, 2009)23 Neurotoxic effects reported in the monkey studies are discussed in
more detail in Section 4.4.2.
In the monkeys, 8 animals (sex unspecified) were exposed to 10, 100, or 1,000 ppm
(13, 131, and 1,310 mg/m3) methanol, 21 hours/day, for 7 months (2 animals), 19 months,
(3 animals), or 29 months (3 animals). There was no indication in the NEDO (1987) report that
this study employed a concurrent control group. One of the 3 animals receiving 100 ppm
methanol and scheduled for sacrifice at 29 months was terminated at 26 months. Clinical signs
were monitored twice daily, body weight changes and food consumption were monitored weekly,
and all animals were given a general examination under anesthetic once a month. Blood was
collected for hematological and clinical chemistry tests at term, and all animals were subject to a
histopathologic examination of the major organs and tissues.
While there were no clinical signs of toxicity in the low-concentration animals, there was
some evidence of nasal exudate in monkeys in the mid-concentration group. High-concentration
(1,000 ppm) animals also displayed this response and were observed to scratch themselves over
their whole body and crouch for long periods. Food and water intake, body temperature, and
body weight changes were the same among the groups. NEDO (1987) reported that there was no
abnormality in the retina of any monkey. When animals were examined with an
electrocardiogram, there were no abnormalities in the control or 10 ppm groups. However, in the
100 ppm group, one monkey showed a negative change in the T wave. All 3 monkeys exposed to
1,000 ppm (1,310 mg/m3) displayed this feature, as well as a positive change in the Q wave. This
effect was described as a slight myocardial disorder and suggests that 10 ppm (13.1 mg/m3) is a
NOAEL for chronic myocardial effects of methanol and mild respiratory irritation. There were
no compound-related effects on hematological parameters. However, 1 monkey in the 100 ppm
(131 mg/m3) group had greater than normal amounts of total protein, neutral lipids, total and free
cholesterol, and glucose, and displayed greater activities of ALT and aspartate transaminase
(AST). The authors expressed doubts that these effects were related to methanol exposure and
speculated that the animal suffered from liver disease.24
Histopathologically, no degeneration of the optical nerve, cerebral cortex, muscles, lungs,
trachea, tongue, alimentary canal, stomach, small intestine, large intestine, thyroid gland,
pancreas, spleen, heart, aorta, urinary bladder, ovary or uterus were reported (neuropathological
findings are discussed in Section 4.4.2). Most of the internal organs showed no compound-
related histopathologic lesions. However, there were signs of incipient fibrosis and round cell
infiltration of the liver in monkeys exposed to 1,000 ppm (1,310 mg/m3) for 29 months. NEDO
23 An external peer review (ERG. 2009) was conducted by EPA in 2009 to evaluate the accuracy of experimental
procedures, results, and interpretation and discussion of the findings presented in these study reports.
24 Ordinarily, the potential for liver disease in test animals would be remote, but may be a possibility in this case
given that these monkeys were captured in the wild.
4-15
-------
(1987) indicated that this fibrosis occurred in 2/3 monkeys of the 1,000 ppm group to a "strictly
limited extent." They also qualitatively reported a dose-dependent increase in "fat granules" in
liver cells "centered mainly around the central veins" at all doses, but did not provide any
response data. The authors state that 1,000 ppm (1,310 mg/m3) represents a chronic lowest-
observed-adverse-effect level (LOAEL) for hepatic effects of inhaled methanol, suggesting that
the no effect level would be 100 ppm (131 mg/m3). However, this is a tenuous determination
given the lack of information on the pathological progression and significance of the appearance
of liver cell fat granules at exposures below 1,000 ppm and the lack detail (e.g., time of sacrifice)
for the control group.
Dose-dependent changes were observed in the kidney; NEDO (1987) described the
appearance of Sudan-positive granules in the renal tubular epithelium at 100 ppm (131 mg/m3)
and 1,000 (1,310 mg/m3) and hyalinization of the glomerulus and penetration of round cells into
the renal tubule stroma of monkeys exposed to methanol at 1,000 (1,310 mg/m3). The former
effect was more marked at the higher concentration and was thought by the authors to be
compound-related. This would indicate a no effect level at 10 ppm (13.1 mg/m3) for the chronic
renal effects of methanol. The authors observed atrophy of the tracheal epithelium in four
monkeys. However, the incidence of these effects was unrelated to dose and therefore, could not
be unequivocally ascribed to an effect of the solvent. No other histopathologic abnormalities
were related to the effects of methanol. Confidence in these determinations is considerably
weakened by limited study details (e.g., materials and methods, statistical methods, data), small
group sizes and uncertainty over whether a concurrent control group was used in the chronic
study.25 In general, external peer reviewers of the NEDO (1987) monkey studies stated that the
deficiencies in these reports were broad and significant, precluding the use of these studies for
quantitative dose-response assessment (ERG, 2009). Although the limited information available
from the NEDO (1987) summary report suggests that 100 ppm (131 mg/m3) may be an effect
level for myocardial effects, renal effects and neurotoxicity (see Section 4.4.2) following
continuous, chronic exposure to methanol, NOAEL and LOAEL values are not derived for any
of the NEDO (1987) monkey studies.
NEDO also performed 12-months inhalation studies in rats and mice (NEDO, 1987), an
18-month inhalation study in mice (NEDO, 1985a) and a 24-month inhalation study in rats
(NEDO, 1985b). External peer reviewers generally indicated that these rodent studies used good
experimental designs, group sizes, endpoints and quality assurance procedures that were
consistent with the OECD guidelines in place at the time. However, the reports available for the
chronic studies (NEDO, 1985a, b) were far more detailed than the summary reports available for
25 All control group responses were reported in a single table in the section of the NEDO (1987) report that describes
the acute monkey study, with no indication as to when the control group was sacrificed.
4-16
-------
the 12-month studies (NEDO, 1987), which suffered from many of the same reporting issues
identified for the NEDO monkey studies, including a lack a full description of the materials and
methods and raw data from the experiments. For all of the NEDO (1987) mouse, rat and monkey
studies, parameters should have been assessed by one way analysis of variance (ANOVA), rather
than the t-test comparisons with controls that were apparently performed (ERG, 2009).
NEDO (1987) describes a 12-month inhalation study in which 20 F344 rats/sex/group
were exposed to 0, 10, 100, or 1,000 ppm (0, 13.1, 131, and 1,310 mg/m3) methanol,
approximately 20 hours/day, for a year. Clinical signs of toxicity were monitored daily; body
weights and food consumption were recorded weekly for the first 13 weeks, then monthly. Blood
samples were drawn at term to measure hematological and clinical chemistry parameters.
Weights of the major organs were monitored at term, and a histopathologic examination was
carried out on all major organs and tissues. Survival was high among the groups; one high-
concentration female died on day 337 and one low-concentration male died on day 340. As
described by the authors, a number of procedural anomalies arose during this study. For example,
male controls in two cages lost weight because of an interruption to the water supply. Another
problem was that the brand of feed was changed during the study. Fluctuations in some clinical
chemistry and hematological parameters were recorded. The authors considered the fluctuations
to be minor and within the normal range. Likewise, a number of histopathologic changes were
observed, which, in every case, were considered to be unrelated to exposure level or due to
aging.
A companion experiment featured the exposure of 30 B6C3F1 mice/sex/group for 1 year
to the same concentrations as the F344 rats (NEDO, 1987). Broadly speaking, the same suite of
toxicological parameters was monitored as described above, with the addition of urinalysis.
10 mice/sex/group were sacrificed at 6 months to provide interim data on the parameters under
investigation. A slight atrophy in the external lacrimal gland was observed in both sexes and was
significant in the 1,000 ppm male group compared with controls. An apparently dose-related
increase in moderate fatty degeneration of hepatocytes was observed in males (1/20, 4/20, 6/20
and 8/20 in the 0, 10, 100, and 1,000 ppm dose groups, respectively) which was significantly
increased over controls at the 1,000 ppm dose. However a high (10/20) incidence of moderate to
severe fatty degeneration was observed in untreated animals maintained outside of the chamber.
In addition, there was a clear correlation between fatty degeneration and body weight (a change
which was not associated with treatment at 12 months); heavier animals tended to have more
severe cases of fatty degeneration. Thus, methanol's role in fatty liver degeneration in mice is
questionable, especially given the failure to confirm the finding in the 18-month study described
below (ERG, 2009). The possibility of renal deficits due to methanol exposure was suggested by
the appearance of protein in the urine. However, this effect was also seen in controls and did not
4-17
-------
display a dose-response effect. Therefore, it is unlikely to be a consequence of exposure to
methanol. NEDO (1987) reported other histopathologic and biochemical (e.g., urinalysis and
hematology) findings that do not appear to be related to treatment, including a number of what
were considered to be spontaneous tumors in both control and exposure groups.
NEDO (1987. 1985a)26 exposed 52 male and 53 female B6C3F1 mice/group for
18 months at the same concentrations of methanol (0, 10, 100 and 1,000 ppm) and with a similar
experimental protocol to that described in the 12-month studies.27 Animals were sacrificed at the
end of the 18-month exposure period. NEDO (1985a) reported that "there was no
microbiological contamination that may have influenced the result of the study" and that the
study included an assessment of general conditions, body weight change, food consumption rate,
laboratory tests (urinalysis, hematological, and plasma biochemistry) and pathological tests
(pathological autopsy,28 organ weight check and histopathology29). As stated in the summary
report (NEDO, 1987), a few animals showed clinical signs of toxicity, but the incidence of these
responses was not related to dose. Likewise, there were no compound-related changes in body
weight increase, food consumption,30 urinalysis, hematology, or clinical chemistry parameters.
High-concentration males had lower testis weights compared to control males. Significant
differences were detected for both absolute and relative testis weights. One animal in the high-
dose group had severely atrophied testis weights, approximately 25% of that of the others in the
dose group. Exclusion of this animal in the analysis still resulted in a significant difference in
absolute testis weight compared to controls but resulted in no difference in relative testis weight.
High-concentration females had higher absolute kidney and spleen weights compared to controls,
but there was no significant difference in these organ weights relative to body weight. At
necropsy, there were signs of swelling in spleen, preputial glands, and uterus in some animals.
Some animals developed nodes in the liver and lung although, according to the authors, none of
these changes were treatment-related. NEDO (1985a) reported that all non-neoplastic changes
were "nonspecific and naturally occurring changes that are often experienced by 18-month old
26 This study is described in a summary report (NEDO. 1987) and a more detailed, eight volume translation of the
original chronic mouse study report (NEDO. 1985a). The translation was submitted to EPA by the Methanol
Institute and has been certified by NEDO as accurate and complete (Hashimoto. 2008). An external peer review
(ERG. 2009) was conducted by EPA in 2009 to evaluate the accuracy of experimental procedures, results, and
interpretation and discussion of the findings presented in these study reports.
27 The authors reported that "[t]he levels of methanol turned out to be ~4 ppm in low level exposure group (10 ppm)
for ~11 weeks from week 43 of exposure due to the analyzer malfunction" and that "the average duration of
methanol exposure was 19.1 hours/day for both male and female mice."
28 Autopsy was performed on all cases to look for gross lesions in each organ.
29 Complete histopathological examinations were performed for the control group and high-dose (1,000 ppm)
groups. Only histopathological examinations of the liver were performed on the low- and medium-level exposure
groups because no chemical-related changes were found in the high-level exposure group and because liver changes
were noted in the 12-month mouse study (NEDO. 1987).
30 NEDO (NEDO. 1985a) reports sporadic reductions in food consumption of the 1,000 ppm group, but no
associated weight loss or abnormal test results.
4-18
-------
B6C3F1 mice" and that fatty degeneration of liver that was suspected to occur dose-dependently
in the 12-month NEDO (1987) study was not observed in this study NEDO (1985a).
Another study reported in NEDO (1987, 1985b)31 was a 24-month bioassay in which
52 F344 rats/sex/group were kept in whole body inhalation chambers containing 0, 10, 100, or
1,000 ppm (0, 13.1, 131, and 1,310 mg/m3) methanol vapor. Animals were maintained in the
exposure chambers for approximately 19.5 hours/day for a total of 733-736 days (males) and
740-743 days (females). Animals were monitored once a day for clinical signs of toxicity, body
weights were recorded once a week, and food consumption was measured weekly from a
24-animal subset from each group. Urinalysis was carried out on the day prior to sacrifice for
each animal, the samples being monitored for pH, protein, glucose, ketones, bilirubin, occult
blood, and urobilinogen. Routine clinical chemistry and hematological measurements were
carried out and all animals were subject to necropsy at term, with a comprehensive
histopathological examination of tissues and organs.32
There was some fluctuation in survival rates among the groups in the rat study, though
apparently unrelated to exposure concentration.33 In all groups, at least 60% of the animals
survived to term. A number of toxicological responses were described by the authors, including
atrophy of the testis, cataract formation, exophthalmia, small eye ball, alopecia, and paralysis of
the hind leg. However, according to the authors, the incidence of these effects were unrelated to
dose and more likely represented effects of aging. NEDO (1985b) reported a mild, nonsignificant
(4%) body weight suppression among 1,000 ppm females between 51 and 72 weeks, but that
body weight gain was largely similar among the groups for the duration of the experiment. Food
consumption was significantly lower than controls in high-concentration male rats during the day
210-365 time interval, but no corresponding weight loss was observed. Among hematological
parameters, mid- and high-concentration females had a significantly (p < 0.05) higher differential
leukocyte count than controls, but dose dependency was not observed. Serum total cholesterol,
triglyceride, free fatty acid, and phospholipid concentrations were significantly (p < 0.05) lower
in high-concentration females compared to controls. Likewise, serum sodium concentrations
were significantly (p < 0.05) lower in mid- and high-concentration males compared to controls.
31 This study is described in a summary report (NEDO. 1987) and a more detailed, 10-volume translation of the
original chronic rat study report (NEDO. 1985b). The translation was submitted to EPA by the Methanol Institute
and has been certified by NEDO as accurate and complete (Hashimoto. 2008). An external peer review (ERG. 2009)
was conducted by EPA in 2009 to evaluate the accuracy of experimental procedures, results, and interpretation and
discussion of the findings presented in these study reports.
32 Complete histopathological examinations were performed on the cases killed on schedule (week 104) among the
control and high-exposure groups, and the cases that were found dead/ killed in extremis of all the groups. Because
effects were observed in male and female kidneys, male lungs, as well as female adrenal glands of the high-level
exposure group, these organs were histopathologically examined in the low- and mid-exposure groups.
33Survival at the time of exposure termination (24 months) was 69%, 65%, 81%, and 65% for males and 60%, 63%,
60% and 67% for females of the control, low-, mid- and high-exposure groups, respectively.
4-19
-------
High-concentration females had significantly lower (p < 0.05) serum concentrations of inorganic
phosphorus but significantly (p < 0.05) higher concentrations of potassium compared to controls.
Glucose levels were elevated in the urine of high-concentration male rats relative to controls, and
female rats had lower pH values and higher bilirubin levels in mid- and high-concentration
groups relative to controls. In general, NEDO (1987, 1985b) reported that these variations in
urinary, hematology, and clinical chemistry parameters were not related to chemical exposure.
NEDO (1987) reported that there was little change in absolute or relative weights of the
major organs or tissues. When the animals were examined grossly at necropsy, there were some
signs of swelling in the pituitary and thyroid, but these effects were judged to be unrelated to
treatment. The most predominant effect was the dose-dependent formation of nodes in the lung
of males (2/52, 4/52, 5/52, and 10/52 \p < 0.01] for control, low-, mid-, and high-concentration
groups, respectively). Histopathologic examination pointed to a possible association of these
nodes with the appearance of pulmonary adenoma (1/52, 5/52, 2/52, and 6/52 for control, low-,
mid- and high-concentration groups, respectively) and a single pulmonary adenocarcinoma in the
high-dose group (1/52).
The authors identified a tentative NOAEL of 100 ppm for the 12-month and 18-month
mouse and the 12-month and 24-month rat studies on page 293 of their summary report (NEDO,
1987). However, peer reviewers of these studies expressed reservations about the dose-response
data quality (e.g., histopathology was only performed on the 10 and 100 ppm groups if the 1,000
ppm group demonstrated statistically significant difference from controls) and interpretation
(e.g., statistical methods were incompletely described and, in some cases, improperly applied)
(ERG, 2009). In addition, the evidence for dose-related effects at 1,000 ppm was weak for both
the mouse and rat studies. Thus, EPA assigns a low weight-of evidence determination to the
1,000 ppm LOAEL identified for these chronic mouse and rat studies.
4.3. Reproductive and Developmental Studies - Oral and Inhalation
Many studies have been conducted to investigate the reproductive and developmental
toxicity of methanol. The purpose of these studies was principally to determine if methanol has a
similar toxicology profile to another widely studied teratogen, ethanol.
4.3.1. Oral Reproductive and Developmental Studies
Three studies were identified that investigated the reproductive and developmental effects
of methanol in rodents via the oral route (Fu et al., 1996; Sakanashi etal., 1996; Rogers et al.,
4-20
-------
1993b). Two of these studies also investigated the influence of folic acid-deficient (FAD) diets
on the effects of methanol exposures (Fu et al., 1996; Sakanashi etal., 1996).
Rogers et al. (1993b) conducted a developmental toxicity study in which methanol in
water was administered to pregnant female CD-I mice via gavage on GD6-GD15. Eight test
animals received 4 g/kg-day methanol given in 2 daily doses of 2 g/kg; 4 controls received
distilled water. By analogy to the protocol of an inhalation study of methanol that was described
in the same report, it is assumed that dams were sacrificed on GDI7, at which point implantation
sites, live and dead fetuses, resorptions/litter, and the incidences of external and skeletal
anomalies and malformations were determined. In the brief summary of the findings provided by
the authors, it appears that cleft palate (43.5% per litter versus 0% in controls) and exencephaly
(29% per litter versus 0% in controls) were the prominent external defects following maternal
methanol exposure by gavage. Likewise, an increase in totally resorbed litters and a decrease in
the number of live fetuses per litter were evident. However, it is possible that these effects may
have been caused or exacerbated by the high bolus dosing regimen employed. It is also possible
that effects were not observed due to the limited study size. The small number of animals in the
control group relative to the test group limits the power of this study to detect treatment-related
responses.
Sakanashi et al. (1996) tested the influence of dietary folic acid intake on various
reproductive and developmental effects observed in CD-I mice exposed to methanol. Starting
5 weeks prior to breeding and continuing for the remainder of the study, female CD-I mice were
fed folic acid free diets supplemented with 400 (low), 600 (marginal), or 1,200 (sufficient) nmol
folic acid/kg. After 5 weeks on their respective diets, females were bred with CD-I male mice.
On GD6-GD15, pregnant mice in each of the diet groups were given twice-daily gavage doses of
2.0 or 2.5 g/kg-day methanol (total dosage of 4.0 or 5.0 g/kg-day). On GDIS, mice were weighed
and killed, and the liver, kidneys, and gravid uteri removed and weighed. Maternal liver and
plasma folate levels were measured; and implantation sites, live and dead fetuses, and
resorptions were counted. Fetuses were weighed individually and examined for cleft palate and
exencephaly. One third of the fetuses in each litter were examined for skeletal morphology. They
observed an approximate 50% reduction in liver and plasma folate levels in the mice fed low
versus sufficient folic acid diets in both the methanol exposed and unexposed groups. Similar to
Rogers et al. (1993b), Sakanashi et al. (1996) observed that an oral dose of 4-5 g/kg-day
methanol during GD6-GD15 resulted in an increase in cleft palate in mice fed sufficient folic
acid diets, as well as an increase in resorptions and a decrease in live fetuses per litter. They did
not observe an increase in exencephaly in the folic acid sufficient (FAS) group at these doses,
and the authors suggest that this may be due to diet and the source of CD-I mice differing
between the two studies.
4-21
-------
In the case of the animals fed the folate deficient diet, there was a 50% reduction in
maternal liver folate concentration and a threefold increase in the percentage of litters affected by
cleft palate (86.2% versus 34.5% in mice fed sufficient folic acid) and a 10-fold increase in the
percentage of litters affected by exencephaly (34.5% versus 3.4% in mice fed sufficient folic
acid) at the 5 g/kg methanol dose. Sakanashi et al. (1996) speculate that the increased methanol
effect from the FAD diet could have been due to an increase in tissue formate levels (not
measured) or to a critical reduction in conceptus folate concentration following the methanol
exposure. Plasma and liver folate levels at GDIS within each dietary group were not
significantly different between exposed versus unexposed mice. However, these measurements
were taken 3 days after methanol exposure. Dorman et al. (1995) observed a transient decrease in
maternal red blood cells (RBCs) and conceptus folate levels within 2 hours following inhalation
exposure to 15,000 ppm methanol on GD8. Thus, it is possible that short-term reductions in
available folate during GD6-GD15 may have affected fetal development.
Fu et al. (1996) also tested the influence of dietary folic acid intake on reproductive and
developmental effects observed in CD-I mice exposed to methanol. This study was performed
by the same laboratory and used a similar study design and dosing regimen as Sakanashi et al.
(1996), but exposed the pregnant mice to only the higher 2.5 g/kg-day methanol (total dosage of
5.0 g/kg-day) on GD6-GD10. Like Sakanashi et al. (1996), Fu et al. (1996) measured maternal
liver and plasma folate levels on GDIS and observed similar, significant reductions in these
levels for the FAD versus FAS mice. However, Fu et al. (1996) also measured fetal liver folate
levels at GDIS. This measurement does not address the question of whether methanol exposure
caused short-term reductions in fetal liver folate because it was taken 8 days after the
GD6-GD10 exposure period. However, it did provide evidence regarding the extent to which a
maternal FAD diet can impact fetal liver folate levels in this species and strain. Significantly, the
maternal FAD diet had a greater impact on fetal liver folate than maternal liver folate levels.
Relative to the FAS groups, fetal liver folate levels in the FAD groups were reduced 2.7-fold for
mice not exposed to methanol (1.86 ± 0.15 nmol/g in the FAD group versus 5.04 ± 0.22 nmol/g
in the FAS group) and 3.5-fold for mice exposed to methanol (1.69 ± 0.12 nmol/g in the FAD
group versus 5.89 ± 0.39 nmol/g in the FAS group). Maternal folate levels in the FAD groups
were only reduced twofold both for mice not exposed (4.65 ± 0.37 versus 9.54 ± 0.50 nmol/g)
and exposed (4.55 ± 0.19 versus 9.26 ± 0.42 nmol/g). Another key finding of the Fu et al. (1996)
study is that methanol exposure during GD6-GD10 appeared to have similar fetotoxic effects,
including cleft palate, exencephaly, resorptions, and decrease in live fetuses, as the same level of
methanol exposure administered during GD6-GD15 (Sakanashi et al., 1996; Rogers et al.,
1993b). This is consistent with the hypothesis made by Rogers et al. (1993b) that the critical
period for methanol-induced cleft palate and exencephaly in CD-I mice is within GD6-GD10.
4-22
-------
As in the studies of Sakanashi et al. (1996) and Rogers et al. (1993b), Fu et al. (1996) reported a
higher incidence of cleft palate than exencephaly.
4.3.2. Inhalation Reproductive and Developmental Studies
Nelson et al. (1985) exposed 15 pregnant Sprague-Dawley rats/group to 0, 5,000, 10,000,
or 20,000 ppm (0, 6,552, 13,104, and 26,209 mg/m3) methanol (99.1% purity) for 7 hours/day.
Exposures were conducted on GDI-GDI9 in the two lower concentration groups and
GD7-GD15 in the highest concentration group, apparently on separate days. Two groups of
15 control rats were exposed to air only. Day 1 blood methanol levels measured 5 minutes after
the termination of exposure in NP rats that had received the same concentrations of methanol as
those animals in the main part of the experiment were 1.00 ± 0.21, 2.24 ± 0.20, and
8.65 ± 0.40 mg/mL for those exposed to 5,000, 10,000, and 20,000 ppm methanol, respectively.
Evidence of maternal toxicity included a slightly unsteady gait in the 20,000 ppm group during
the first few days of exposure. Maternal bodyweight gain and food intake were unaffected by
methanol. Dams were sacrificed on GD20, and 13-30 litters/group were evaluated. No effect was
observed on the number of corpora lutea or implantations or the percentage of dead or resorbed
fetuses. Statistical evaluations included analysis of variance (ANOVA) for body weight effect,
Kruskal-Wallis test for endpoints such as litter size and viability and Fisher's exact test for
malformations. Fetal body weight was significantly reduced at concentrations of 10,000 and
20,000 ppm by 7% and 12-16%, respectively, compared to controls. An increased number of
litters with skeletal and visceral malformations were observed at > 10,000 ppm, with statistical
significance obtained at 20,000 ppm. Numbers of litters with visceral malformations were 0/15,
5/15, and 10/15 and with skeletal malformations were 0/15, 2/15, and 14/15 at 0, 10,000, and
20,000 ppm, respectively. Visceral malformations included exencephaly and encephaloceles. The
most frequently observed skeletal malformations were rudimentary and extra cervical ribs. The
developmental and maternal NOAELs for this study were identified as 5,000 ppm (6,552 mg/m3)
and 10,000 ppm (13,104 mg/m3), respectively.
NEDO (1987) sponsored a teratology study in Sprague-Dawley rats that included an
evaluation of postnatal effects in addition to standard prenatal endpoints. Thirty-six pregnant
females/group were exposed to 0, 200, 1,000, or 5,000 ppm (0, 262, 1,310, and 6,552 mg/m3)
methanol vapors (reagent grade) on GD7-GD17 for 22.7 hours/day. Statistical significance of
results was evaluated by t-test, Mann-Whitney U test, Fisher's exact test, and/or Armitage's
X2 test.
Contrary to the Nelson et al. (1985) report of a 10,000 ppm NOAEL for this rat strain, in
the prenatal portion of the NEDO (1987) study, reduced body weight gain and food and water
4-23
-------
intake during the first 7 days of exposure were reported for dams in the 5,000 ppm group.
However, it was not specified if these results were statistically significant. One dam in the
5,000 ppm group died on GDI9, and one dam was sacrificed on GDIS in moribund condition.
On GD20, 19-24 dams/group were sacrificed to evaluate the incidence of reproductive deficits
and such developmental parameters as fetal viability, weight, sex, and the occurrence of
malformations. The reported reproductive and fetal effects are summarized in Table 4-2. The
authors suggest that adverse effects (an increase in late-term resorptions, decreased live fetuses,
reduced fetal weight, and increased frequency of litters with fetal malformations, variations, and
delayed ossifications) were limited to the 5,000 ppm group. However, dose-response analyses
indicate statistically significant linear trends for more than one reproductive/fetal effect in the
FI rats, including number of pre-implantation resorptions (p < 0.01), pre-implantation resorption
rate (p < 0.01) and bifurcated vertebral center (p < 0.01) (ERG. 2009).
Postnatal effects of methanol inhalation were evaluated in the remaining 12 dams/group
that were permitted to deliver and nurse their litters. Again, the authors suggest that effects were
limited to the 5,000 ppm group, including a 1-day prolongation of the gestation period and
reduced post-implantation survival, number of live pups/litter, and survival on PND4 (Table 4-3).
However, dose-response analyses indicate statistically significant linear trends for post-
implantation embryo survival rate (p < 0.01) and number of surviving pups on postnatal day 4
(p < 0.03) (ERG, 2009). When the delay in parturition was considered, methanol treatment had
no effect on attainment of developmental milestones such as eyelid opening, auricle
development, incisor eruption, testes descent, or vaginal opening. There were no adverse body
weight effects in offspring from methanol treated groups. The weights of some organs (brain,
thyroid, thymus, and testes) were reduced in 8-week-old offspring exposed to 5,000 ppm
methanol during prenatal development.
4-24
-------
Table 4-2 Reproductive and developmental toxicity in pregnant Sprague-Dawley rats
exposed to methanol via inhalation during gestation.
Exposure concentration (ppm)
Effect
0
200
1,000
5,000
Reproductive effects
Number of pregnant females
examined
Number of corpora lutea
Number of implantations
No. of pre-implantation resorptions
Early resorption
Late resorption
Number of live fetuses
Sex ratio (M/F)
Fetal weight (male)
Fetal weight (female)
Total resorption rate (%)
Pre-implantation resorption rate
(%)d
Pre-implantation resorption rate
(%)e
Early resorption rate (%)
Late resorption rate (%)
19
17.0 ±2.6
15.7± 1.6
0.79 ±0.85
0.68 ±0.75
0.11 ±0.32
14.95 ± 1.61
144/140
3.70 ± 0.24
3.51 ±0.19
11.2±9.0
6.6 ±8.2
4.9 ±5.2
4.3 ±4.7
0.6 ± 1.9
24
17.2 ±2.7
15.0 ±3.0
0.71 ± 1.23
0.71 ± 1.23
0.0 ±0.0
14.25 ±3.54
177/165
3.88 ±0.23
3.60 ±0.25
15.6±21.3
11. 8± 18.7
5.4 ± 12.1
5.4 ± 12.1
0.0 ±0.0
22
16.4 ± 1.9
15.5 ± 1.2
0.95 ±0.65
0.91 ±0.61
0.05 ±0.21
14.55± 1.1
164/156
3.82 ±0.29
3.60 ±0.30
10.6 ±8.4
4.9 ±7.9
6.1 ±4.0
5.8 ±3.9
0.3± 1.3
21
16.5 ±2.4
14.5 ±3.3
1.67 ±2.03
0.67 ± 0.97
1.00± 1.79
12.86 ±4.04a
134/136
3.02 ± 0.27C
2.83 ± 0.26C
23.3±22.7a
12.7 ± 16.5
14.5 ±23.3
4.2 ±6.1
10.4±23.4a
Soft tissue malformations
Number of fetuses examined
Abnormality at base of right
subclavian
Excessive left subclavian
Ventricular septal defect
Residual thymus
136
0.7 ±2.87(1)
0
0
2.9 ±5.91 (4)
165
0
0
0.6 ±2.92(1)
2.4 ± 5.44 (4)
154
0
0
0
2.6 ±5.73 (4)
131
0
3.5 ±9.08 (3)
47.6 ±36.51 (16)b
53.3 ± 28.6 (20)b
Serpengious urinary tract
43.0 ±24.64 (18) 35.2 ± 31.62 (19) 41.8 ± 38.45 (15) 22.1 ± 22.91 (13)
4-25
-------
Table 4-2 (Continued) Reproductive and developmental toxicity in pregnant Sprague-
Dawley rats exposed to methanol via inhalation during gestation.
Exposure concentration (ppm)
Effect
0
200
1,000
5,000
Skeletal abnormalities
Number of fetuses examined
Atresia of foramen
costotransversarium
Patency of foramen
costotransversium
Cleft sternum
Split sternum
Bifurcated vertebral center
Cervical rib
Excessive sublingual neuropore
Curved scapula
Waved rib
Abnormal formation of lumbar
vertebrae
148
23.5 ±5.47 (3)
0
0
0
0.8 ±3.28(1)
0
0
0
0
0
177
7.7 ± 1.3(8)
0
0
0
1.6 ±5.61 (2)
0
0
0
0
0
165
3.5 ±8.88 (4)
0.6 ±2.67(1)
0
0
3.0 ±8.16 (3)
0
0
0
0
0
138
45.2 ± 25.18 (20)b
13.7 ±20.58 (7)
5.6 ± 14.14(3)
7.0 ± 14.01 (5)
14.5± 16.69(11)b
65.2 ± 25.95 (19)b
49.9 ±27.31 (19)
0.7 ±3.19(1)
6.1 ± 11.84(5)
0.7 ±3.19(1)
ap<0.05
bp<0.01
°p < 0.001, as calculated by the authors.
dPre-implantation resorption/corpora lutea x 100 (%)
e(Early + late resorption) / implantation x 100 (%)
Values are means ± SD Values in parentheses are the numbers of litters.
Source: NEDO (1987).
4-26
-------
Table 4-3 Reproductive parameters in Sprague-Dawley dams exposed to methanol during
pregnancy, and then allowed to deliver their pups.
Exposure concentration (ppm)
Parameter
Number of dams
Duration of gestation (days)
Number of implantations
Number of pups
Number of live pups
Number of live pups on PND4
Sex ratio (M/F)
Postimplantation embryo
survival rate
0
12
21. 9 ±0.3
15.8 ± 1.6
15.2 ± 1.6
15.2 ± 1.6
15.0 ± 1.7(2)
88/94
96.3 ± 4.2
200
12
21.9±0.3
14.8± 1.2
14.4± 1.3
14.1 ± 1.4
13.8 ± 1.5(3)
87/85
94.9 ±5.1
1,000
12
21.9±0.3
15.3± 1.3
14.5± 1.4
14.3± 1.4
14.2± 1.6(1)
103/703
93.6 ±6.1
5,000
12
22.6±0.5C
14.6± 1.1a
13.1 ±2.2a
12.6±2.5b
10.3 ± 2.8 (9)c
75/81
86.2±16.2a
ap<0.05
bp<0.01
°p < 0.001
p values presented as calculated by the authors.
Values are means ± SD Values in parentheses are the numbers of litters.
Source: NEDO (1987).
NEDO (1987) contains an account of a two-generation reproductive study that evaluated
the effects of pre- and postnatal methanol (reagent grade) exposure (20 hours/day) on
reproductive and other organ systems of Sprague-Dawley rats. The F0 generation (30 males and
30 females per exposure group)34 was exposed to 0, 10, 100, and 1,000 ppm (0, 13.1, 131, and
1,310 mg/m3) from 8 weeks old to the end of mating (males) or to the end of lactation period
(females). The FI generation was exposed to the same concentrations from birth to the end of
mating (males) or to weaning of F2 pups 21 days after delivery (females). Males and females of
the F2 generation were exposed from birth to 21 days old (one animal/sex/litter was exposed to
8 weeks of age). NEDO (1987) noted reduced brain, pituitary, and thymus weights, and early
testicular descent in the offspring of FO and FI rats exposed to 1,000 ppm methanol. The early
testicular descent is believed to be an indication of earlier fetal development as indicated by the
observation that it was correlated with increased pup body weight. However, no histopathologic
effects of methanol were observed. As discussed in the report, NEDO (1987) sought to confirm
the possible compound-related effect of methanol on the brain by carrying out an additional
study in which Sprague-Dawley rats were exposed to 0, 500, 1,000, and 2,000 ppm (0, 655,
34 A second control group of 30 animals/sex was maintained in a separate room to "confirm that environmental
conditions inside the chambers were not unacceptable to the animals."
4-27
-------
1,310, and 2,620 mg/m3) methanol from the first day of gestation through the FI generation (see
Section 4.4.2).
Rogers et al. (1993b) evaluated development toxicity in pregnant female CD-I mice
exposed to air or 1,000, 2,000, 5,000, 7,500, 10,000, or 15,000 ppm (0, 1,310, 2,620, 6,552,
9,894, 13,104, and 19,656 mg/m3) methanol vapors (> 99.9% purity) in a chamber for
7 hours/day on GD6-GD15 in a 3-block design experiment. The numbers of mice exposed at
each dose were 114, 40, 80, 79, 30, 30, and 44, respectively. During chamber exposures to air or
methanol, the mice had access to water but not food. In order to determine the effects of the
chamber exposure conditions, an additional 88 control mice were not handled and remained in
their cages; 30 control mice were not handled but were food deprived for 7 hours/day on
GD6-GD15. Effects in dams and litters were statistically analyzed using the General Linear
Models procedure and multiple ^-test of least squares means for continuous variables and the
Fisher's exact test for dichotomous variables. An analysis of plasma methanol levels in
3 pregnant mice/block/treatment group on GD6, GD10, and GDIS revealed a dose-related
increase in plasma methanol concentration that did not seem to reach saturation levels, and
methanol plasma levels were not affected by gestation stage or number of previous exposure
days. Across all 3 days, the mean plasma methanol concentrations in pregnant mice were
approximately 97, 537, 1,650, 3,178, 4,204, and 7,330 |ig/mLinthe 1,000, 2,000, 5,000, 7,500,
10,000, and 15,000 ppm exposure groups, respectively.
The dams exposed to air or methanol in chambers gained significantly less weight than
control dams that remained in cages and were not handled. There were no methanol-related
reductions in maternal body weight gain or overt signs of toxicity. Dams were sacrificed on
GDI 7 for a comparison of developmental toxicity in methanol-treated groups versus the chamber
air-exposed control group. Fetuses in all exposure groups were weighed, assessed for viability,
and examined for external malformations. Fetuses in the control, 1,000, 2,000, 5,000, and
15,000 ppm groups were also examined for skeletal and visceral defects. Incidence of
developmental effects is listed in Table 4-4. A statistically significant increase in cervical
ribs/litter was observed at concentrations of 2,000, 5,000, and 15,000 ppm. At doses of
>5,000 ppm the incidences of cleft palates/litter and exencephaly/litter were increased with
statistical significance achieved at all concentrations with the exception of exencephaly which
increased but not significantly at 7,500 ppm.35 A significant reduction in live pups/litter was
noted at > 7,500 ppm, with a significant increase in fully resorbed litters occurring at
> 10,000 ppm. Fetal weight was significantly reduced at > 10,000 ppm. Rogers et al. (1993b)
identified a developmental NOAEL and LOAEL of 1,000 ppm and 2,000 ppm, respectively.
35 Due to the serious nature of this response and the relative lack of a response in controls, all incidence of
exencephaly reported in this study at 5,000 ppm or higher are considered biologically significant.
4-28
-------
They also provide BMD maximum likelihood estimates (referred to by the authors as MLE) and
estimates of the lower 95% confidence limit on the BMD (BMDL; referred to by the authors as
benchmark dose [BMD]) for 5% and 1% added risk, by applying a log-logistic dose-response
model to the mean percent/litter data for cleft palate, exencephaly and resorption. The BMD0s
and BMDLos values for added risk estimated by Rogers et al. (1993b) are listed in Table 4-5.
From this analysis, the most sensitive indicator of developmental toxicity was an increase in the
proportion of fetuses per litter with cervical rib anomalies. The most sensitive BMDL and BMD
from this effect for 5% added risk were 305 ppm (400 mg/m3) and 824 ppm (1,080 mg/m3),
respectively.36
36 The BMD analysis of the data described in Section 5 was performed similarly using, among others, a similar
nested logistic model. However, the Rogers et al. (1993b) analysis was performed using added risk and external
exposure concentrations, whereas the analyses in Section 5 used extra risk and internal dose metrics that were then
converted to human equivalent exposure concentrations.
4-29
-------
Table 4-4 Embryonic and Developmental
effects in
CD-I mice after methanol inhalation.
Exposure concentration (ppm)
Effects
0
1,000
2,000
5,000
7,500
10,000
15,000
Endpoint
No. live pups/litter
No. fully resorbed litters
Fetus weight (g)
Cleft palate/ litter (%)
Exencephaly/litter (%)
9.9
0
1.20
0.21
0
9.5
0
1.19
0.65
0
12.0
0
1.15
0.17
0.88
9.2
0
1.15
8.8b
6.9a
8.6b
3
1.17
46.6C
6.8
7.3C
5a
1.04C
52.7C
27.4C
2.2C
14C
0.70C
48.3C
43.3C
Anomalies
Cervical ribs/litter (%)
Sternebral defects/litter (%)
Xiphoid defects/litter (%)
Vertebral arch defects/litter (%)
Extra lumbar ribs/litter (%)
Ossifications (values are means of litter
Sternal
Caudal
Metacarpal
Proximal phalanges
Metatarsals
Proximal phalanges
Distal phalanges
Supraoccipital score+
28
6.4
6.4
0.3
8.7
means)
5.96
5.93
7.96
7.02
9.87
7.18
9.64
1.40
33.6
7.9
3.8
ND
2.5
5.99
6.26
7.92
7.04
9.90
7.69
9.59
1.65
ap<0.05
bp<0.01
cp< 0.001
p values as calculated by the authors.
ND = Not determined. + = on a scale of 1-4, where 1 is fully ossified
Source: Reprinted with permission of John Wiley & Sons; Rogers et
49.6b
3.5
4.1
ND
9.6
5.94
5.71a
7.96
7.04
9.87
6.91
9.57
1.57
74.4C
20.2C
10.9
1.5
15.6
5.81
5.42
7.93
6.12
9.82
5.47
8.46b
1.48
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
60.0a
100C
73.3C
33.3C
40.0C
5.07C
3.20a
7.60b
3.33C
8.13C
Oc
4.27C
3.20C
and 4 is unossified.
al. (1993b).
4-30
-------
Table 4-5 Benchmark doses at two added risk levels.
Endpoint
Cleft Palate (CP)
Exencephaly (EX)
CP and EX
Resorptions (RES)
CP, EX, and RES
Cervical ribs
BMDos (ppm)
4,314
5,169
3,713
5,650
3,667
824
BMDLos(ppm)
3,398
3,760
3,142
4,865
3,078
305
BMDoi (ppm)
2,717
2,122
2,381
3,749
2,484
302
BMDUi (ppm)
1,798
784
1,816
2,949
1,915
58
Source: Reprinted with permission of John Wiley & Sons; Rogers et al. (1993b).
Bolon et al. (1993) performed an inhalation exposure developmental study in CD-I mice
under conditions similar to Rogers et al. (1993b). To determine the determine the developmental
phase specificity of methanol induced fetal effects, they evaluated developmental toxicity in
CD-I mice (n = 20-27/group) following inhalation exposure (6 hr/day) to 5,000, 10,000, or
15,000 ppm methanol either throughout organogenesis (GD 6-15), during the period of neural
tube development and closure (GD 7-9), or during a time of potential neural tube reopening
(GD9-GD11). To better define the critical gestational window of susceptibility, mice
(n = 8-15/group) were exposed to 15,000 ppm on GD 7, GD8 or GD9 or for 2 days on GD7-GD8
or GD8-GD9. The results of the dose-response portion of the study are shown in Table 4-6 and
the results of the "window of susceptibility" portion of the study are shown in Table 4-7.
4-31
-------
Table 4-6 Developmental Phase-Specific Embryotoxicity and Teratogenicity in CD-I mice
after methanol inhalation.
Gestational Days of Exposure
Methanol Concentration
No. of pregnant dams
No. implants/litter3
GD7 to GD9
0
22
12.5 ±0.4
5,000 ppm
27
11.6±0.5
10,000 ppm
20
12.8 ±0.4
15,000 ppm
20
13.4 ±0.4
GD9toGD11
15,000 ppm
17
12.8 ±0.4
Embryotoxicity
% Resorptions/litterb
% Litters with > 1 Resorptionb
No. (%) of live fetuses/I ittera'd
Fetal body wt (GD 17)a, in
grams(g)
Maternal body wt (GD 17)a, in
grams (g)
2.7
27.3
12.0 ±0.4 (98)
0.92 ±0.05
51.2 ±0.9
10.5
55.6C
10.8 ±0.5 (99)
0.96 ±0.01
49.7 ±0.8
16.6
75.0C
11.2 ±0.6 (100)
0.91 ±0.01
51.1 ± 1.1
46.2C
90.0C
7.9 ± 1.1 (91)
0.82 ±0.02
45.9 ± 1.8
6.9
41.2
10.5 ±0.9 (87)
0.83 ±0.01
51.1 ± 1.1
Developmental Toxicity
No. of litters examined
Neural tube defects8
Cleft palate8
22
0
9(0.7)
27
0
4 (0.3)
20
30 (3.6)
50C(14.6)
17
65C(14.7)
88C (50.4)
17
0
53(20.1)
Renal pelvic dilatation
Cavitation8
Hydronephrosis8
Ocular defects8
Limb anomalies8
Tail anomalies8
41 (4.3)
0
0
0
0
100C(49.4)
7 (0.9)
0
0
0
90C(31.2)
45C(13.9)
10C(1.3)
5(0.5)
40C(8.8)
75C (44.9)
53C(11.3)
53C(17.2)
0
65C(15.1)
100(36.9)
18(5.9)
0
41 (24.7)
71 (12.4)
aValues represent mean ± standard error.
bEmbryos from 3/20 litters completely resorbed at 15,000 ppm.
°Denotes lowest dose that was significantly different from control by Shirley's test, p < 0.05
d(Percentage of live fetuses per number of fetuses born).
Percentage of affected litters (Percentage affected fetuses)
Source: Reprinted with permission of Oxford University Press; Bolon et al. (1993).
4-32
-------
Table 4-7 Developmental phase-specific embryotoxicity in CD-I mice induced by
methanol inhalation (15,000 ppm) during neurulation.
Gestational Days of Exposure
No. of pregnant dams
No. of implants/litter3
% Resorptions/litter
% Litters with
> 1 Resorptionb
No. (%) of live
fetuses/litter3
Fetal body wt
(GD17)a, (grams[g])
Control
(GD7-GD9)b
22
12.5 ±0.4
2.7
27.3
12.0 ±0.04
(98.3)
0.92 ± 0.05
GD7
15
11.3±0.9
38.6 c
86.7 c
7.7 ± 1.2C
(92.3)
0.99 ±0.03
GD8
13
12.9 ±0.6
4.2
30.8
12.2 ±0.6
(98.9)
0.93 ±0.02
GD9
8
13.2 ±0.8
2.3
25.0
12.9 ±0.8
(99.1)
0.99 ±0.02
GD7-GD8
14
12.9 ±0.5
41. 9C
100C
8.4± 1.0
(95.5)
0.81 ±0.02
GD8-GD9
11
12.7± 1.1
10.7
45.5
11. 7 ± 1.3
(98.7)
0.90 ±0.03
GD7-GD9b
20
13.4 ±0.4
46.2C
90.0 c
7.9± 1.1
(91.0)
0.82± 0.02C
Maternal body wt (GD
17)a, (grams [g])
Dam with uterus
Dam minus uterus
Neural tube defectsd
51.2±0.9
36.9 ±2.1
0
45.3 ±2.0
34.8 ±0.9
8(1.4)
54.0 ± 1.3
35.8 ±0.4
15(2.2)
54.3 ±2.5
34.3 ± 1.4
0
46.1 ± 1.8
33.5 ±0.8
67C(15.6)
52.9 ±2.5
35.1 ± 1.0
27(1.9)
45.9 ± 1.8
Not Done
65C(14.7)
"Values represent mean ± standard error
"Values from Table 4-6
°Significantly different from controls by Dunn's test, ac = 0.05
Percentage affected litters (Percentage affected fetuses)
Source: Reprinted with permission of Oxford University Press; Bolon et al. (1993).
Bolon et al. (1993) reported that transient neurologic signs and reduced body weights
were observed in up to 20% of dams exposed to 15,000 ppm. Embryotoxicity (increased
resorptions, reduced fetal weights, and/or fetal malformations) was apparent at 10,000 and
15,000 ppm, while 3-day exposures at 5,000 ppm yielded only an increase in the percentage of
litters with one or more resorptions. Developmental toxicity included neural and ocular defects,
cleft palate, hydronephrosis, deformed tails, and limb (paw and digit) anomalies at 10,000 ppm
(GD 7-9). The only endpoint increased at 5,000 ppm was renal pelvic dilatation (cavitation).
Neural tube defects and ocular lesions occurred after methanol inhalation between GD7-GD9,
while limb anomalies were induced only during GD9-GD11; cleft palate and hydronephrosis
were observed after exposure during either period. Table 5 (of the Bolon et al. study) shows that
neural tube effects are most likely to develop from exposure on GD8 and resorptions are most
likely to occur from exposure on GD7. These findings indicate that the spectrum of teratogenic
effects depended upon both the timing (i.e., stage of embryonic development) and the number of
methanol exposures.
Bolon et al. (1994) observed a spectrum of cephalic neural tube defects in near-term
(gestation day 17 [GDI7]) mouse fetuses following maternal inhalation of methanol at a high
4-33
-------
concentration (15,000 ppm) for 6 hr/day during neurulation (GD7-GD9). Their results suggest
that (1) exposure to a high concentration of methanol injures multiple stem cell populations in
the neurulating mouse embryo and (2) significant neural pathology may remain in older
conceptuses even in the absence of gross lesions.
Rogers and Mole (1997) investigated the critical period of sensitivity to the
developmental toxicity of inhaled methanol in the CD-I mouse by exposing 12-17 pregnant
females to 0 or 10,000 ppm (0 and 13,104 mg/m3), 7 hours/day on 2 consecutive days during
GD6-GD13, or to a single exposure to the same methanol concentration during GD5-GD9.
Another group of mice received a single 7-hour exposure to methanol at 10,000 ppm. The latter
animals were sacrificed at various time intervals up to 28 hours after exposure. Blood samples
were taken from these animals to measure the concentration of methanol in the serum. Serum
methanol concentrations peaked at ~4 mg/mL 8 hours after the onset of exposure. Methanol
concentrations in serum had declined to pre-exposure levels after 24 hours. All mice in the main
body of the experiment were sacrificed on GDI 7, and their uteri removed. The live, dead, and
resorbed fetuses were counted, and all live fetuses were weighed, examined externally for cleft
palate, and then preserved. Skeletal abnormalities were determined after the carcasses had been
cleaned and eviscerated. Cleft palate, exencephaly, and skeletal defects were observed in the
fetuses of exposed dams. For example, cleft palate was observed following 2-day exposures to
methanol on GD6-GD7 through GD11-GD12. These effects also were apparent in mice receiving
a single exposure to methanol on GD5-GD9. This effect peaked when the dams were exposed on
GD7. Exencephaly showed a similar pattern of development in response to methanol exposure.
However, the data indicated that cleft palate and exencephaly might be competing
malformations, since only one fetus displayed both features. Skeletal malformations included
exoccipital anomalies, atlas and axis defects, the appearance of an extra rudimentary rib on
cervical vertebra No.7, and supernumerary lumbar ribs. In each case, the maximum time point
for the induction of these defects appeared to be when the dams were exposed to methanol on or
near GD7. When dams were exposed to methanol on GD5, there was also an increased incidence
of fetuses with 25 presacral vertebrae (26 is normal). However, an increased incidence of fetuses
with 27 presacral vertebrae was evident when dams were exposed on GD7. These results indicate
that gastrulation and early organogenesis is a period of increased embryonic sensitivity to
methanol.
Burbacher et al. (1999a: 1999b) carried out toxicokinetic and reproductive/developmental
studies of methanol in M. fascicularis monkeys that were published by the Health Effects
Institute (HEI) in a two-part monograph. Some of the data were subsequently published in the
open scientific literature (Burbacher et al., 2004a: Burbacher et al., 2004b). The experimental
protocol featured exposure to 2 cohorts of 12 monkeys/group to low exposure levels (relative to
4-34
-------
the previously discussed rodent studies) of 0, 200, 600, or 1,800 ppm (0, 262, 786, and
2,359 mg/m3) methanol vapors (99.9% purity), 2.5 hours/day, 7 days/week, during a premating
period and mating period (~180 days combined) and throughout the entire gestation period
(-168 days). The monkeys were 5.5-13 years old. The study included an evaluation of maternal
reproductive performance and tests to assess infant postnatal growth and newborn health,
reflexes, behavior, and development of visual, sensorimotor, cognitive, and social behavioral
function (see Section 4.4.2 for a review of the developmental neurotoxicity findings from this
study). Blood methanol levels, clearance, and the appearance of formate were also examined and
are discussed in Section 3.2.
With regard to reproductive parameters, there was a statistically significant decrease
(p = 0.03) in length of pregnancy in all treatment groups, as shown in Table 4-8. Cesarean
section (C-section) deliveries performed in the methanol exposure groups did not impact this
finding (decreased length of pregnancy was observed in vaginally delivered animals). C-section
deliveries were performed in response to "signs of difficulty" in the pregnancy, but it is not clear
whether this is an indication of either reproductive dysfunction or fetal risk due to methanol
exposure. Maternal menstrual cycles, conception rate, and live birth index were all unaffected by
exposure. There were also no signs of an effect on maternal weight gain or clinical toxicity
among the dams.
While pregnancy duration was virtually the same in all exposure groups, there were some
indications of increased pregnancy duress only in methanol-exposed monkeys. C-sections were
done in 2 monkeys from the 200 ppm group and 2 from the 600 ppm group due to vaginal
bleeding, presumed, but not verified, to be from placental detachment.37 A monkey in the
1,800 ppm group also received a C-section after experiencing nonproductive labor for 3 nights.
In addition, signs of prematurity were observed in 1 infant from the 1,800 ppm group that was
born after a 150-day gestation period. The consistent reduction in length of pregnancy observed
in may reflect a treatment effect on the fetal neuroendocrine system. The authors suggested that
the shortened gestation length could be due to a direct effect of methanol on the fetal
hypothalamus-pituitary-adrenal (HPA) axis or an indirect effect of methanol on the maternal
uterine environment. Other fetal parameters such as crown-rump length and head circumference
were unchanged among the groups. Infant growth and tooth eruption were unaffected by prenatal
methanol exposure.
37 Burbacher, et al. (20041^ and Burbacher et al. (2004a) note, however, that in studies of pregnancy complication in
alcohol- exposed human subjects, an increased incidence of uterine bleeding and abruptio placenta has been
reported.
4-35
-------
Table 4-8 Reproductive parameters in monkeys exposed via inhalation to methanol
during prebreeding, breeding, and pregnancy.
Exposure (ppm) Conception rate
0
200
600
1,800
9/11
9/12
9/11
10/12
Weight gain (kg) Pregnancy duration (days)3
1.67 ±0.07
1.27±0.14
1.78 ±0.25
1.54 ±0.20
168 ±2
160 ±2b
162 ±2b
162 ±2b
Live born delivery rate
8/9
9/9
8/9
9/10
aLive-born offspring only;
bp < 0.05, as calculated by the authors.
Values are means ± SE.
Source: Reprinted with permission of Elsevier; Burbacher et al. (2004b).
In later life, 2 females out of the total of 9 offspring in the 1,800 ppm group experienced
a wasting syndrome at 12 and 17 months of age. Food intake was normal and no cause of the
syndrome could be determined in tests for viruses, hematology, blood chemistry, and liver,
kidney, thyroid, and pancreas function. Necropsies revealed gastroenteritis and severe
malnourishment. No infectious agent or other pathogenic factor could be identified. Thus, it
appears that a highly significant toxicological effect on postnatal growth can be attributed to
prenatal methanol exposure at 1,800 ppm (2,300 mg/m3).
In summary, the Burbacher et al. (1999a: 1999b) studies suggest that methanol exposure
can cause reproductive effects, manifested as a shortened mean gestational period, pregnancy
complications that precipitated delivery via a C-section, and developmental neurobehavioral
effects which may or may not be related to the shortened gestational period (see Section 4.4.2).
The low exposure of 200 ppm may signify a LOAEL for reproductive effects. However, the
decrease in gestational length was marginally significant. Also, this effect did not appear to be
dose related, the greatest gestational period decrease having occurred at the lowest (200 ppm)
exposure level. Thus, a clear NOAEL or LOAEL cannot be determined from this study.
In a study of the testicular effects of methanol, Cameron et al. (1984) exposed 5 male
Sprague-Dawley rats/group to methanol vapor, 8 hours/day, 5 days/week for 1, 2, 4, and 6 weeks
at 0, 200, 2,000, or 10,000 ppm (0, 262, 2,620, and 13,104 mg/m3). The authors examined the
possible effects of methanol on testicular function by measuring blood levels of testosterone,
luteinizing hormone (LH), and follicular stimulating hormone (FSH) using radioimmunoassay.
When the authors tabulated their results as a percentage of the control value for each duration
series, the most significant changes were in blood testosterone levels of animals exposed to
200 ppm methanol, the lowest concentration evaluated. At this exposure level, animals exposed
for 6 weeks had testosterone levels that were 32% of those seen in controls; however, higher
concentrations of methanol were associated with testosterone levels that were closer to those of
4-36
-------
controls. The lack of a clear dose-response is not necessarily an indication that the effect is not
related to methanol. The higher concentrations of methanol could be causing other effects
(e.g., liver toxicity) which can influence the results. Male rats exposed to 10,000 ppm methanol
for 6 weeks displayed blood levels of LH that were about 3 times higher (mean ± SD) than those
exposed to air (311 ± 107% versus 100 ± 23%). In discussing their results, the authors placed
greater emphasis on the observation that an exposure level equal to the ACGIH TLV (200 ppm)
had caused a significant depression in testosterone formation in male rats.
A follow-up study report by the same research group (Cameron et al., 1985) described the
exposure of 5 male Sprague-Dawley rats/group, 6 hours/day for either 1 day or 1 week, to
methanol, ethanol, n-propanol, or n-butanol at their respective TLVs. Groups of animals were
sacrificed immediately after exposure or after an 18-hour recovery period, and the levels of
testosterone, LH, and corticosterone measured in serum. As shown in Table 4-9, the data were
consistent with the ability of these aliphatic alcohols to cause a transient reduction in the
formation of testosterone. Except in the case of n-butanol, rapid recovery from these deficits can
be inferred from the 18-hour postexposure data.
4-37
-------
Table 4-9 Mean serum levels of testosterone, luteinizing hormone, and corticosterone
(± SD) in male Sprague-Dawley rats after inhalation of methanol, ethanol,
n-propanol or n-butanol at threshold limit values.
TLV
Condition (ppm)
Single-day exposure
18 hr
End of exposure postexposure
One-week exposure
18 hr
End of exposure postexposure
Testosterone (as a percentage of control)
Control
Methanol
Ethanol
n-Propanol
n-Butanol
200
1,000
200
50
100 ± 17
41 ± 16a
64 ± 12a
58 ± 15a
37±8a
100 ±20
98 ± 18
86 ± 16
81 ± 13
52 ± 22a
100 ±26
81 ±22
88 ± 14
106 ±28
73 ±34
100± 17
82 ±27
101 ± 13
89 ± 17
83 ± 18
Luteinizing hormone
Control
Methanol
Ethanol
n-Propanol
n-Butanol
200
1,000
200
50
100 ±30
86 ±32
110±22
117±59
124 ±37
100 ±35
110±40
119±54
119±83
115±28
100 ±28
78 ± 13
62 ±26
68 ±22
78 ±26
100 ±36
70 ± 14
81 ± 17
96 ±28
98 ±23
Corticosterone
Control
Methanol
Ethanol
n-Propanol
n-Butanol
200
1,000
200
50
100 ±20
115± 18
1 1 1 ± 32
112±21
143+ na
ND
ND
ND
ND
ND
100 ±21
74 ±26
60 ±25
79 ± 14
85 ±26
ND
ND
ND
ND
ND
ap < 0.05, as calculated by the authors.
ND = No data.
Source: Reprinted with permission of Springer-Verlag; Cameron et al. (1985).
In a series of studies that are relevant to the reproductive toxicity of methanol in males,
Lee et al. (1991) exposed 8-week-old male Sprague-Dawley rats (9-10/group) to 0 or 200 ppm
(0 and 262 mg/m3) methanol, 8 hours/day, 5 days/week, for 1, 2, 4, or 6 weeks to measure the
possible treatment effects on testosterone production. Study results were evaluated by one factor
ANOVA followed by Student's ^-test. In the treated rats, there was no effect on serum
testosterone levels, gross structure of reproductive organs, or weight of testes and seminal
vesicles. Lee et al. (1991) also studied the in vitro effect of methanol on testosterone production
from isolated testes, but saw no effect on testosterone formation either with or without the
addition of human chorionic gonadotropin hormone.
In a third experiment from the same report, Lee et al. (1991) examined testicular
histopathology to determine if methanol exposure produced lesions indicative of changing
testosterone levels; the effects of age and folate status were also assessed. This is relevant to the
4-38
-------
potential toxicity of methanol because folate is the coenzyme of tetrahydrofolate synthetase, an
enzyme that is rate limiting in the removal of formate. Folate deficiency would be expected to
cause potentially toxic levels of formate to be retained. The same authors examined the relevance
of folate levels, and by implication, the overall status of formate formation and elimination in
mediating the testicular functions of Long-Evans rats. Groups of 4-week-old male Long-Evans
rats were given diets containing either adequate or reduced folate levels plus 1%
succinylsulfathiazole, an antibiotic that, among other activities,38 would tend to reduce the folate
body burden. At least 9 rats/dietary group/dose were exposed to 0, 50, 200, or 800 ppm (0, 66,
262, and 1,048 mg/m3) methanol vapors starting at 7 months of age while 8-12 rats/dietary
group/dose were exposed to 0 or 800 ppm methanol vapors at 15 months of age. The methanol
exposures were conducted continuously for 20 hours/day for 13 weeks. Without providing
details, the study authors reported that visual toxicity and acidosis developed in rats fed the low
folate diet and exposed to methanol. No methanol-related testicular lesions or changes in testes
or body weight occurred in rats that were fed either the folate sufficient or deficient diets and
were 10 months old at the end of treatment. Likewise, no methanol-lesions were observed in
18-month-old rats that were fed diets with adequate folate. However, the incidence but not
severity of age-related testicular lesions was increased in the 18-month-old rats fed folate-
deficient diets. Subcapsular vacuoles in germinal epithelium were noted in 3/12 control rats and
8/13 rats in the 800 ppm group. One rat in the 800 ppm group had atrophied seminiferous tubules
and another had Leydig cell hyperplasia. These effects, as well as the transient decrease in
testosterone levels observed by Cameron et al. (1985; 1984), could be the result of chemically-
related strain on the rat system as it attempts to maintain hormone homeostasis.
Dorman et al. (1995) conducted a series of in vitro and in vivo studies of developmental
toxicity in ICRBR (CD-I) mice associated with methanol and formate exposure. The studies
used HPLC grade methanol and appropriate controls. PK and developmental toxicity parameters
were measured in mice exposed to a 6-hour methanol inhalation (10,000 or 15,000 ppm),
methanol gavage (1.5 g/kg bw) or sodium formate (750 mg/kg by gavage) on GD8. In the in vivo
inhalation study, 12-14 dams/group were exposed to 10,000 ppm methanol for 6 hours on GD8,39
with and without the administration of fomepizole to inhibit the metabolism of methanol by
ADH1. Dams were sacrificed on GD10, and folate levels in maternal RBC and conceptus
(decidual swelling) were measured, as well as fetal neural tube patency (an early indicator of
methanol-induced dysmorphogenic response). The effects observed included a transient decrease
in maternal RBC and conceptus folate levels within 2 hours following exposure and a significant
38 Succinylsulfathiazole antibiotic may have a direct impact on the effects being measured, the extent of which was
not addressed by the authors of this study.
39 Dorman et al. (1995) state that GD8 was chosen because it encompasses the period of murine neurulation and the
time of greatest vulnerability to methanol-induced neural tube defects.
4-39
-------
(p < 0.05) increase in the incidence of fetuses with open neural tubes (9.65% in treated versus
0 in control). These responses were not observed following sodium formate administration,
despite peak formate levels in plasma and decidual swellings being similar to those observed
following the 6-hour methanol inhalation of 15,000 ppm. This suggests that these methanol -
induced effects are not related to the accumulation of formate. As this study provides information
relevant to the identification of the proximate teratogen associated with developmental toxicity in
rodents, it is discussed more extensively in Section 4.7.1.
4.3.3. Other Reproductive and Developmental Studies
Additional information relevant to the possible effects of methanol on reproductive and
developmental parameters has been provided by experimental studies that have exposed
experimental animals to methanol during pregnancy via i.p. injections (Sweeting et al., 2011;
Degitz et al., 2004b: Rogers et al., 2004). Relevant to the developmental impacts of the chemical,
a number of studies also have examined the effects of methanol when included in whole-embryo
culture (Miller and Wells. 2011: Hansen et al.. 2005: Harris etal.. 2003: Andrews et al.. 1998:
Andrews et al., 1995: Andrews et al., 1993).
Pregnant female C57BL/6J mice received two i.p. injections of methanol on GD7
(Rogers et al., 2004). The injections were given 4 hours apart to provide a total dosage of 0, 3.4,
and 4.9 g/kg. Animals were sacrificed on GDI7 and the litters were examined for live, dead, and
resorbed fetuses. Rogers et al. (2004) monitored fetal weight and examined the fetuses for
external abnormalities and skeletal malformations. Methanol-related deficits in maternal and
litter parameters observed by Rogers et al. (2004) are summarized in Table 4-10.
4-40
-------
Table 4-10 Maternal and litter parameters when pregnant female C57BL/6J mice were
injected i.p. with methanol.
Methanol dose (g/kg)
Parameter
No. pregnant at term
Wt gain GD7-GD8 (g)
WtgainGD7-GD10(g)
Live fetuses/litter
Resorbed fetuses/litter
Dead fetuses/litter
Fetal weight (g)
0
43
0.33 ±0.10
1.63 ±0.1 8
7.5 ±0.30
0.4 ±0.1
0.1 ±0.1
0.83 ±0.02
3.4
13
0.37 ±0.15
2.20 ± 0.20
6.3±0.5a
1.3±0.4a
0
0.82 ±0.03
4.9
24
-0.24±0.14a
1.50 ±0.20
3.7±0.4a
4.4 ± 0.4a
0.1 ±0.1
0.70 ± 0.02a
ap < 0.05, as calculated by the authors.
Values are means ± SEM.
Source: Reprinted with permission of Elsevier; Rogers et al. (2004).
Rogers et al. (2004) used a number of sophisticated imaging techniques, such as confocal
laser scanning and fluorescence microscopy, to examine the morphology of fetuses excised at
GD7, GD8, and GD9. They identified a number of external craniofacial abnormalities, the
incidence of which was, in all cases, significantly increased in the high-dose group compared to
controls. For some responses, such as microanophthalmia and malformed maxilla, the incidence
was also significantly increased in animals receiving the lower dose. Fifteen compound-related
skeletal malformations were tabulated in the report. In most cases, a dose-response effect was
evident, resulting in statistically significant incidences in affected fetuses and litters, when
compared to controls. Apparent effects of methanol on the embryonic forebrain included a
narrowing of the anterior neural plate, missing optical vesicles, and holoprosencephaly (failure of
the embryonic forebrain to divide). The authors noted that there was no sign of incipient cleft
palate or exencephaly, as had been observed in CD-I mice exposed to methanol via the oral and
inhalation routes (Rogers et al., 1993b).
In order to collect additional information on cell proliferation and histological changes in
methanol-treated fetuses, Degitz et al. (2004b) used an identical experimental protocol to that of
Rogers et al. (2004) by administering 0, 3.4, or 4.9 g methanol/kg in distilled water i.p. (split
doses, 4 hours apart) to C57BL/6J mice on GD7. Embryos were collected at various times on
GD8 and GD10. Embryos from dams exposed to 4.9 g/kg and examined on GD8 exhibited
reductions in the anterior mesenchyme, the mesenchyme subjacent to the mesencephalon and the
base of the prosencephalon (embryonic forebrain), and in the forebrain epithelium. The optic pits
were often lacking; where present their epithelium was thin and there were fewer neural crest
cells in the mid- and hindbrain regions. At GD9, there was extensive cell death in areas
4-41
-------
populated by the neural crest, including the forming cranial ganglia. Dose-related abnormalities
in the development of the cranial nerves and ganglia were seen on GD7. In accordance with an
arbitrary dichotomous scale devised by the authors, scores for ganglia V, VIII, and IX were
significantly (not otherwise specified) reduced at all dose levels, and ganglia VII and X were
reduced only at the highest dose. At the highest dose (4.9 g/kg), the brain and face were poorly
developed and the brachial arches were reduced in size or virtually absent. Flow cytometry of the
head regions of the embryos from the highest dose at GD8 did not show an effect on the
proportion of cells in S-phase.
Cell growth and development were compared in C57BL/6J and CD-I mouse embryos
cultured in methanol (Degitz et al., 2004a). GD8 embryos, with 5-7 somites, were cultured in
0, 1, 2, 3, 4, or 6 mg methanol/mL for 24 hours and evaluated for morphological development.
Cell death was increased in both strains in a developmental stage- and region-specific manner at
4 and 6 mg/mL after 8 hours of exposure. The proportions of cranial region cells in S-phase were
significantly (p < 0.05) decreased at 6 mg/mL following 8- and 18-hour exposures to methanol.
After 24 hours of exposure, C57BL/6J embryos had significantly (p < 0.05) decreased total
protein at 4 and 6 mg/kg. Significant (p < 0.05) developmental effects were seen at 3, 4, and
6 mg/kg, with eye dysmorphology being the most sensitive endpoint. CD-I embryos had
significantly decreased total protein at 3, 4, and 6 mg/kg, but developmental effects were seen
only at 6 mg/kg. It was concluded that the C57BL/6J embryos were more severely affected by
methanol in culture than the CD-I embryos.
Sweeting et al. (2011) performed a series of experiments in NZW rabbits, C57BL/6J mice
and C3H mice to compare plasma pharmacokinetics of methanol and formic acid and
embryotoxicity. For the teratology portion of the study, pregnant female mice and rabbits were
given two i.p. doses of 2 g methanol/kg body weight on GD7 or GD8, for a total daily dose of
4 g methanol/kg body weight, or two i.p. doses of a saline vehicle control. Methanol exposure
did not significantly impact fetal body weights for any of the species and strains tested. No
statistically significant effects were reported on rabbit growth parameters and mortality.
A 4.4-fold increase in tail abnormalities per litter, including shortening and absence, was
reported in rabbit fetuses. However, due to the variability of this endpoint among litters, this
difference was not statistically significant. Non-significant increases were reported in exposed
rabbit litters for several other effects that were not observed in controls, including two fetuses
with open posterior neuropores, one with an abdominal wall defect (prune belly), and three with
frontal nasal hyperplasia. In C3H mice, methanol in utero exposure caused a 2-fold increase in
fetal resorptions, but this increase was not statistically significant over saline treated controls
(p < 0.01). In C57BL/6, methanol caused a 66% incidence of fetal ophthalmic abnormalities
(p < 0.001) compared to a non-significant 3% incidence in C3H mice. Ophthalmic anomalies
4-42
-------
were not observed in saline-exposed controls of either strain. Methanol also caused a 17%
increase in fetal cleft palates in C57BL/6 mice (p < 0.05) compared to 0% in saline controls, and
0% in C3H mice treated with either methanol or saline. No increase in cephalic NTDs, an
endpoint commonly observed in CD-I mice, was observed in C57BL/6 or C3H mice. The
different teratological results across these mouse strains could not be explained by differences in
methanol or formic acid disposition (the pharmacokinetic results of this study are described in
Section 3.2). The authors hypothesize that these differences in embryotoxicity could be due to
strain differences in ADH activity and the amount of catalase available for ROS detoxification,
or differences in other pathways that involve ROS formation. Sweeting et al. (2011) suggest that
their findings indicate that rabbits are resistant to the teratogenic effects of methanol. However,
because the critical gestational window for developmental effects could be different for rabbits
versus mice, this claim needs to be verified over several gestational days, as has been done for
mice. Postpartum lethality was nearly 2-fold higher in the methanol exposed (11%) versus
control (5%) rabbit fetuses, and stillbirths were also increased (4% versus 0%). Though these
increased incidences were not statistically significant, they may prove to be biologically
significant given that postpartum lethality ("wasting syndrome") and a shortened gestational
period were possible adverse outcomes observed in methanol exposed monkeys (see discussion
of Burbacher et al., (2004b; 1999b) in Section 4.3.2).
Table 4-11 displays the results of three studies of whole rodent embryos exposed to
methanol (Miller and Wells, 2011; Hansen et al., 2005; Andrews et al., 1993). These data suggest
that mouse embryos are more sensitive than rat embryos to the developmental effects of
methanol. The Miller and Wells (2011) results also demonstrate that developmental effects from
methanol exposure are increased in acatalasemic (aCat)-expressing mouse embryos over their
wild type controls (C3HWT) and decreased in mouse embryos expressing human catalase (hCat)
over their wild type controls (C57WT). These results suggest that embryonic catalase activity
may be a determinant for teratological risk in mice following methanol-exposure.
4-43
-------
Table 4-11 Developmental studies of rodent embryos exposed to methanol.
Species/Strain/GD
Mouse/CD-1/GD8
Rat/Sprague-
Dawley/GD9
Mouse/CD-1/GD8
Rat/Sprague-
Dawley/GD10
Mouse/wild-type
control (C57WT)/GD9
Mouse/C57BL/6 with
human catalase
(hCat)/GD9
Mouse/wild-type
control C3HeB/FeJ
(C3HWT)/GD9
Mouse/CSGa.Cg-
Catb/J acatalasemic
(aCat)/GD9
Embryo Culture Dose
& Duration
0, 2, 4, 6, or 8 mg/mL
for 24 hrs
0,2,4, 8, 12 or 16
mg/mL for 24 hrs
4-12 mg/mL for 24 hrs
8 - 20 mg/mL for 24 hrs
0 or 4 mg/L for 24 hrs
0 or 4 mg/L for 24 hrs
0 or 4 mg/L for 24 hrs
0 or 4 mg/L for 24 hrs
Effect
Decrease in developmental score and crown-rump
length at 4 mg/mL and above. Embryo lethality at
8 mg/mL.
Decrease in somite number, head length, and
developmental score at 8 mg/mL and above.
Embryo lethality at 12 mg/mL.
Reduced VYS DMA and rotation at 4 mg/mL;
reduced embryo DMA and protein, neural tube
closure and viability at 8 mg/L; reduced VYS protein
at 1 0 mg/L
Reduced embryo protein and rotation at 8 mg/mL;
reduced VYS DMA and protein, embryo DMA, and
neural tube closure at 8 mg/L; reduced viability at 16
mg/L
Decreased somites developed and turning, and
increased heart rate at 4 mg/L relative to 0 mg/L.
Decreased neuropore closure at 4 mg/L relative to
0 mg/L and hCat
Increased crown rump length and heart rate relative
to 0 mg/L. Increased somites at 4 mg/L relative to
C57WT
Decreased somites developed at 4 mg/L relative to
0 mg/L.
Decreased somites developed at 4 mg/L relative to
0 mg/L. Reduced anterior neuropore closure and
head length at 4 mg/L relative to 0 mg/L and
C3HWT. Lower yolk sac diameters at 4 mg/L
relative to C3HWT.
Reference
Andrews
et al. (1993)
Andrews
et al. (1993)
Hansen et al.
(2QQ5)
Hansen et al.
(2QQ5)
Miller and
Wells (2011)
Miller and
Wells (2011)
Miller and
Wells (2011)
Miller and
Wells (2011)
GD = Gestation Day; WT = Wild Type; VYS = visceral yolk sac
In contrast to the in vitro and in vivo findings of Dorman et al. (1995), Andrews et al.
(1995) demonstrated that formate can induce similar developmental lesions in whole rat and
mouse conceptuses. Using a similar experimental system as Andrews et al. (1993) to examine the
developmental toxicity of formate and formic acid in comparison to methanol, Andrews et al.
(1995) report that the formates are embryotoxic at doses that are four times lower than equimolar
doses of methanol. Among the anomalies observed were open anterior and posterior neuropores,
plus rotational defects, tail anomalies, enlarged pericardium, and delayed heart development.
Andrews et al. (1998) showed that exposure to combinations of methanol and formate was less
embryotoxic than would be expected based on simple toxicity additivity, suggesting that the
embryotoxicity observed following low-level exposure to methanol is mechanistically different
from that observed following exposure to formate.
4-44
-------
The whole embryo study by Hansen et al. (2005) also determined the comparative
toxicity of methanol and its metabolites, formaldehyde and sodium formate, in GD8 mouse
(CD-I) and GD10 rat (Sprague-Dawley) conceptuses. Whole embryos were incubated for
24 hours in media containing methanol (mouse: 4-12 mg/mL; rat: 8-20 mg/mL), formaldehyde
(mouse: 1-6 jig/mL; rat: 1-8 |ig/mL) and sodium formate (mouse: 0.5-4 mg/mL; rat:
0.5-8 mg/mL). In other experiments, the chemicals were injected directly into the amniotic
space. The embryos were examined morphologically to determine growth and developmental
parameters such as viability, flexure and rotation, crown-rump length, and neuropore closure.
For each response, Table 4-12 provides a comparison of the concentrations or amounts of
methanol, formaldehyde, and formate that resulted in statistically significant changes in
developmental abnormalities compared to controls. For a first approximation, these
concentrations or amounts may be taken as threshold-dose ranges for the specific responses
under the operative experimental conditions. The data show consistently lower threshold values
for the effects of formaldehyde compared to those of formate and methanol. The mouse embryos
were more sensitive towards methanol toxicity than rat embryos, consistent with in vivo
findings, whereas the difference in sensitivity disappeared when formaldehyde was administered.
Hansen et al. (2005) hypothesized that, while the MO A for the initiation of the organogenic
defects is unknown, the relatively low threshold levels of formaldehyde for most measured
effects suggest formaldehyde involvement in the embryotoxic effects of methanol. Consistent
with this hypothesis, formate, a subsequent metabolite of methanol and putative toxicant for the
acute effects of methanol poisoning (acidosis, neurological deficits), did not appear to reproduce
the methanol-induced teratogenicity in these whole embryo culture experiments.
4-45
-------
Table 4-12 Reported thresholds concentrations (and author-estimated ranges) for the onset
of embryotoxic effects when rat and mouse conceptuses were incubated in vitro
with methanol, formaldehyde, and formate.
Parameter
Methanol
Mouse
Formaldehyde
Formate
Methanol
Rat
Formaldehyde
Formate
In vitro incubation (mg/mL)
Viability (%)
Normal rotation (%)
CR a length
Neural tube closure (%)
Reduced embryo
protein
Reduced VYS b protein
Reduced embryo DNA
Reduced VYS DNA
8.0
4.0
No change
8.0
8.0
10.0
8.0
4.0
Microinjection (author-estimated dose
Viability (%)
Normal rotation (%)
CR a length
Neural tube closure (%)
Reduced embryo
protein
Reduced VYS b protein
Reduced embryo DNA
Reduced VYS b DNA
46-89
1-45
No change
1-45
1-45
135-178
46-89
1-45
0.004
0.003
No change
0.001
0.003
0.004
0.003
0.001
ranges in ug)
0.003-0.5
0.003-0.5
No change
0.003-0.5
0.501-1.0
1.01-1.5
0.501-1.0
0.003-0.5
NS
0.5
No change
2.0
4.0
4.0
No change
0.5
1.01-1.5
0.03-0.5
No change
1.01-1.5
No change
No change
No change
0.03-0.5
16.0
8.0
No change
12.0
8.0
12.0
12.0
12.0
46-89
46-89
No change
No change
No change
No change
No change
No change
0.006
0.003
No change
No change
0.004
0.004
0.003
0.003
1.01-1.5
1.01-1.5
No change
No change
1.51-2.0
No change
No change
No change
2.0
4.0
No change
No change
2.0
NR
NR
NR
1.51-4.0
0.51-1.0
No change
1.01-1.5
0.51-1.0
1.01-1.5
0.51-1.0
0.51-1.0
aCR = crown-rump length,
bVYS = visceral yolk sac.
NR = not reported
Source: Adapted with permission of John Wiley and Sons; Hansen et al. (2005): and of Springer; Harris et al. (2004).
Harris et al. (2003) provided biochemical evidence consistent with the concept that
formaldehyde might be the ultimate embryotoxicant of methanol by measuring the activities of
enzymes that are involved in methanol metabolism in mouse (CD-I) and rat (Sprague-Dawley)
whole embryos at different stages of development. Specific activities of the enzymes ADH1,
ADH3, and CAT, were determined in rat and mouse conceptuses during the organogenesis period
of 8-25 somites. Activities were measured in heads, hearts, trunks, and VYS from early- and
late-stage mouse and rat embryos. While CAT activities were similar between rat and mouse
embryos, mouse ADH1 activities in the VYS were significantly lower throughout organogenesis
when compared to the rat VYS or embryos of either species. ADH1 activities of heads, hearts,
and trunks from mouse embryos were significantly lower than those from rats at the 7-12 somite
stage. However, these interspecies differences were not evident in embryos of 20-22 somites.
4-46
-------
ADH3 activities were lower in mouse versus rat VYS, irrespective of the stage of development.
However, while ADH3 activities in mouse embryos were markedly lower than those of rats in
the early stages of development, the levels of activity were similar to at the 14-16 somite stage
and beyond. A lower capacity to transform formaldehyde to formate might explain the increased
susceptibility of mouse versus rat embryos to the toxic effects of methanol. The hypothesis that
formaldehyde is the ultimate embryotoxicant of methanol is supported by the demonstration of
diminished ADH3 activity in mouse versus rat embryos and by the demonstration by Hansen
et al. (2005) that formaldehyde has a far greater embryotoxicity than either formate or methanol
itself.
4.4. Neurotoxicity
A substantial body of information exists on the toxicological consequences to humans
who consume or are exposed to large amounts of methanol. As discussed in Section 4.1,
neurological consequences of acute methanol intoxication in humans include Parkinson-like
responses, visual impairment, confusion, headache, and numerous subjective symptoms. The
occurrence of these symptoms has been shown to be associated with necrosis of the putamen
when neuroimaging techniques have been applied (Salzman, 2006). Such profound changes have
been linked to tissue acidosis that arises when methanol is metabolized to formaldehyde and
formic acid through the actions of ADH1 and ADH3. However, the well-documented impact of
the substantial amounts of formate that are formed when humans and animals are exposed to
large amounts of methanol may obscure the potentially harmful effects that may arise when
humans and animals are exposed to smaller amounts. Human acute exposure studies (Chuwers et
al.. 1995: CooketaL 1991) (See Section 4.1.3) at TLV levels of 200 ppm would indicate that
some measures of neurological function (e.g., sensory evoked potentials, memory testing and
psychomotor testing) were impaired in the absence of measurable formate production.
4.4.1. Oral Neurotoxicity Studies
As discussed in Section 4.2.1.2, an oral subchronic (90 days, beginning at roughly
30 days of age) gavage study noted reduced brain weight in high-dose group (2,500 mg/kg-day)
male and female S-D rats (30/sex/dose) (TRL, 1986). They also reported a higher incidence of
colloid in the hypophyseal cleft of the pituitary gland in the high-dose versus control group
males (13/20 versus 0/20) and females (9/20 versus 3/20). Based on these findings, a 500 mg/kg-
day NOAEL was identified for this study
4-47
-------
Two rodent studies investigated the neurological effects of developmental methanol
exposure via the oral route (Aziz et al., 2002; Infurna and Weiss, 1986). One of these studies also
investigated the influence of FAD diets on the effects of methanol exposures (Aziz et al., 2002).
In the first, Infurna and Weiss (1986) exposed 10 pregnant female Long-Evans rats/dose to 2%
methanol (purity not specified) in drinking water on either GDIS-GDI? or GD17-GD19. Daily
methanol intake was calculated at 2,500 mg/kg-day by the study authors. Dams were allowed to
litter and nurse their pups. Data were analyzed by ANOVA with the litter as the statistical unit.
Results of the study were equivalent for both exposure periods. Treatment had no effect on
gestational length or maternal bodyweight. Methanol had no effect on maternal behavior as
assessed by the time it took dams to retrieve pups after they were returned to the cage following
weighing. Litter size, pup birth weight, pup postnatal weight gain, postnatal mortality, and day of
eye opening did not differ from controls in the methanol treated groups. Two neurobehavioral
tests were conducted in offspring. Suckling ability was tested in 3-5 pups/treatment group on
PND1. An increase in the mean latency for nipple attachment was observed in pups from the
methanol treatment group, but the percentage of pups that successfully attached to nipples did
not differ significantly between treatment groups. Homing behavior, the ability to detect home
nesting material within a cage containing one square of shavings from the pup's home cage and
four squares of clean shavings, was evaluated in 8 pups/group on PND10. Pups from both of the
methanol exposure groups took about twice as long to locate the home material and took less
direct paths than the control pups. Group-specific values differed significantly from controls.
This study suggests that developmental toxicity can occur at this drinking water dose without
readily apparent signs of maternal toxicity.
Aziz et al. (2002) investigated the role of developmental deficiency in folic acid and
methanol-induced developmental neurotoxicity in PND45 rat pups. Wistar albino female rats
(80/group) were fed FAD40 and FAS diets separately. Following 14-16 weeks on the diets, liver
folate levels were estimated and females exhibiting a significantly low folic acid level were
mated. Throughout their lactation period, dams of both the FAD and the FAS group were given
0, 1,2, or 4% v/v methanol via drinking water, equivalent to approximately 480, 960 and
1,920 mg/kg-day.41 Pups were exposed to methanol via lactation from PND1-PND21. Litter size
was culled to 8 with equal male/female ratios maintained as much as possible. Liver folate levels
were determined at PND21 and neurobehavioral parameters (motor performance using the
spontaneous locomotor activity test and cognitive performance using the conditioned avoidance
response [CAR] test), and neurochemical parameters (dopaminergic and cholinergic receptor
40 Along with the FAD diet, 1% succinylsulfathiazole was also given to inhibit folic acid biosynthesis from intestinal
bacteria.
41 Assuming that Wistar rat drinking water consumption is 60 mL/kg-day (Rogers et al.. 2002X 1% methanol in
drinking water would be equivalent to 1% x 0.8 g/mL x 60 mL/kg-day = 0.48 g/kg-day = 480 mg/kg-day.
4-48
-------
binding and dopamine levels) were measured at PND45. The expression of growth-associated
protein (GAP 43), a neuro-specific protein in the hippocampus that is primarily localized in
growth cone membranes and is expressed during developmental regenerative neurite outgrowth,
was examined using immunohistochemistry and Western blot analysis.
A loss in body weight gain was observed at PND7, PND14, and PND21 in animals
exposed to 2% (11, 15 and 19% weight gain reduction) and 4% (17, 24 and 29% weight gain
reduction) methanol in the FAD group and only at 4% (9, 14 and 17% weight gain reduction)
methanol in the FAS group. No significant differences in food and water intake were observed
among the different treatment groups. Liver folate levels in the FAD group were decreased by
63% in rats prior to mating and 67% in pups on PND21.
Based on reports of Parkinson-like symptoms in survivors of severe methanol poisoning
(see Section 4.1), Aziz et al. (2002) hypothesized that methanol may cause a depletion in
dopamine levels and degeneration of the dopaminergic nigrostriatal pathway.42 Consistent with
this hypothesis, they found dopamine levels were significantly decreased (32% and 51%) in the
striatum of rats in the FAD group treated with 2% and 4% methanol, respectively. In the FAS
group, a significant decrease (32%) was observed in the 4% methanol-exposed group.
Methanol treatment at 2% and 4% was associated with significant increases in activity, in
the form of distance traveled in a spontaneous locomotor activity test, in the FAS group (13%
and 39%, respectively) and more notably, in the FAD group (33% and 66%, respectively) when
compared to their respective controls. Aziz et al. (2002) suggest that these alterations in
locomotor activity may be caused by a significant alteration in dopamine receptors and
disruption in neurotransmitter availability. Dopamine receptor (D2) binding in the hippocampus
of the FAD group was significantly increased (34%) at 1% methanol, but was significantly
decreased at 2% and 4% methanol exposure by 20% and 42%, respectively. In the FAS group,
D2 binding was significantly increased by 22% and 54% in the 2% and 4% methanol-exposed
groups.
At PND45, the CAR in FAD rats exposed to 2% and 4% methanol was significantly
decreased by 48% and 52%, respectively, relative to nonexposed controls. In the FAS group, the
CAR was only significantly decreased in the 4% methanol-exposed animals and only by 22% as
compared to their respective controls. Aziz et al. (2002) suggest that the impairment in CAR of
the methanol-exposed FAD pups may be due to alterations in the number of cholinergic
(muscarinic) receptor proteins in the hippocampal region of the brain. Muscarinic receptor
binding was significantly increased in the 2% (20%) and 4% (42%) methanol-exposed group in
42 The nigrostriatal pathway is one of four major dopamine pathways in the brain that are particularly involved in the
production of movement. Loss of dopamine neurons in the substantia nigra is one of the pathological features of
Parkinson's disease (KimetaL 2003).
4-49
-------
FAD animals, while FAS group animals had a significant increase in cholinergic binding only in
the 4% methanol exposed group (21%). High concentrations of methanol may saturate the body's
ability to remove toxic metabolites, including formaldehyde and formate, and this may be
exacerbated in FAD pups having a low store of folate.
Immunohistochemistry showed an increase in the expression of GAP-43 protein in the
dentate granular and pyramidal cells of the hippocampus in 2% and 4% methanol-exposed
animals in the FAD group. The FAS group showed increased expression only in the 4%
methanol-exposed group. The Western blot analysis also confirmed a higher expression of
GAP-43 in the 2% and 4% methanol-exposed FAD group rats. Aziz et al. (2002) suggested that
up-regulation of GAP-43 in the hippocampal region may be associated with axonal growth or
protection of the nervous system from methanol toxicity.
The Aziz et al. (2002) study provides evidence that hepatic tetrahydrofolate is an
important contributing factor in methanol-induced developmental neurotoxicity in rodents.
The immature blood-brain barrier and inefficient drug-metabolizing enzyme system make the
developing brain a particularly sensitive target organ to the effects of methanol exposure.
4.4.2. Inhalation Neurotoxicity Studies
A review by Carson et al. (1981) has summarized a number of older reports of studies on
the toxicological consequences of methanol exposure. In one example relevant to the potential
for neurotoxicity from repeat, low level exposure to methanol, the review cites a research report
of Chen-Tsi (1959) who exposed 10 albino rats/group (sex and strain unstated) to 1.77 and
50 mg/m3 (1.44 and 40.7 ppm) methanol vapor, 12 hours/day, for 3 months. Deformation of
dendrites, especially the dendrites of pyramidal cells, in the cerebral cortex was included in the
description of histopathological changes observed in adult animals following exposure to
50 mg/m3 (40.7 ppm) methanol vapor. One out often animals exposed to the lower methanol
concentration also displayed this feature.
Information on the neurotoxicity of methanol inhalation exposure in adult cynomolgus
monkeys (M fascicularis) has come from NEDO (1987) which describes the results of a number
of inhalation experiments that have already been discussed in Section 4.2.2. The monkey studies
that will be discussed here with respect to their neurotoxicity implications include an acute study,
a chronic study, and a repeated exposure experiment (of variable duration depending upon
exposure level), followed by recovery period (1-6 months), and an experiment looking at chronic
formaldehyde exposure (1 or 5 ppm), a metabolite of methanol. This last experiment was only a
pilot study and included only one monkey per exposure condition.
4-50
-------
As noted in Section 4.2.2.1, histopathologic changes to the CNS reported in monkeys
following acute exposure to methanol included characteristic degeneration of the bilateral
putamen, caudate nucleus, and claustrum, with associated edema in the cerebral white matter
(NEDO, 1987). These lesions increased in severity with increasing exposure. Necrosis of the
basal ganglia was noted following exposure to 5,000 ppm for 5 days (1 animal) and 14 days
(1 animal). The authors reported that at 3,000 ppm, the monkeys experienced little more than
minimal fibrosis of "responsive stellate cells" of the thalamus, hypothalamus, and basal
ganglion. This effect was also observed following chronic exposure and is discussed more
extensively below.
In the chronic experiment, 8 monkeys were included per exposure level (control, 10, 100,
1,000 ppm or 13, 131, and 1,310 mg/m3, respectively, for 21 hours/day); however, animals were
serially sacrificed at 3 time points: 7 months, 19 months, or >26 months. This design reduced
the number of monkeys at each exposure level to 2 subjects at 7 months, and 3 subjects at the
subsequent time points (see Section 4.2.2). One of the 3 animals receiving 100 ppm methanol
and scheduled for sacrifice at 29 months was terminated at 26 months.
Histopathologically, no overt degeneration of the retina, optical nerve, cerebral cortex, or
other potential target organs (liver and kidney) was reported in the chronic experiment.
Regarding the peripheral nervous system, 1/3 monkeys exposed to 100 ppm (131 mg/m3) and
2/3 exposed to 1,000 ppm (1,310 mg/m3) for 29 months showed slight but clear changes in the
peroneal nerves. The most pervasive effect noted across the exposure concentrations and
durations was "fibrosis of responsive stellate cells," characterized as "neurological disease" in
the NEDO (1987) summary report. These "stellate cells" are likely to be astrocytes, star-shaped
glial cells in the brain that are among the most numerous cells in all regions of the CNS. As was
noted in an independent peer review of this study (ERG, 2009), the degree of fibrosis of
responsive stellate cells is an appropriate CNS endpoint of consideration given that stellate
astroglia are believed to play a key role in the pathogenesis of CNS disorders and an essential
role in response to tissue injury and inflammation by hypertrophy, proliferation, production of
growth factors and cytokines, and involvement in extracellular matrix deposition characteristic of
fibrosis (De Keyser et al., 2008) and the presence of hypertrophic astrocytes is considered
evidence of CNS injury (Sofroniew and Vmters, 2010; O'Callaghan and Sriram, 2005). A peer
reviewer also recommended that, because there did not appear to be an effect of duration on the
incidence of this neurological endpoint, the results can be pooled across durations to obtain a
clearer view of dose-response results (ERG, 2009). As reported in "appended Table 3" of the
NEDO (1987) report, the incidence of stellate cell fibrosis at 10 ppm (13.1 mg/m3), 100 ppm
(131 mg/m3) or 1,000 ppm (1,310 mg/m3) for exposure durations of 7 months or longer were:
[3/8, 7/8 and 7/8 within the cerebral white matter]; [0/8, 3/8 and 3/8 inside the nucleus of the
4-51
-------
thalamus]; [3/8, 6/8 and 4/8 in the hypothalamus]; [4/8, 7/8 and 7/8 in the mesencephalon central
gray matter]; [2/8, 7/8 and 7/8 in the pons tegmentum]; and [0/8, 5/8 and 4/9 in the medulla
oblongata tegmentum]. All monkeys that had degeneration of the inside nucleus of the thalamus
also had degeneration of the cerebral white matter.
According to NEDO (1987), the stellate cell response "disappeared as soon as exposure
was stopped" in groups exposed to 1,000 ppm or less for 7 months and was "not characteristic of
degeneration." However, neurodegeneration can persist after a glial cell reaction to nervous
system damage is no longer evident (O'Callaghan and Sriram, 2005). Further, because the RfD
and RfC represent "lifetime" exposures, the transient nature of this response upon cessation of
exposure is not as relevant. The authors also noted that the stellate cell response represented an
"abnormal increase" in 3 of 8 monkeys exposed to 10 ppm methanol, was "widely observed" in
monkeys exposed to 100 ppm and more, was "nearly absent in normal monkeys in the control
group" and that "in the groups exposed to a large quantity of methanol or for a long time their
presence tended to become permanent, so a relation to the long term over which the methanol
was inhaled is suspected." There is a question concerning whether an appropriate, concurrent
control was used as all control group responses are reported in a single table in the section of the
NEDO (1987) report that describes the acute monkey study, with no indication as to when the
control group was sacrificed. However, responses in the mid- and high- dose groups appeared to
be increased over responses in the low-dose groups.
In the recovery experiment, monkeys were exposed for 7 months to 1,000 ppm
(3 animals), for 20 days to 2,000 ppm (3 animals), for 20 days to 3,000 ppm (4 animals), for
5-14 days to 5,000 ppm (5 animals) or for 6 days to 7,000 ppm (2 animals) methanol, followed
by recovery periods of various durations. Monkeys exposed to 3,000 ppm for 20 days followed
by a 6-month recovery period experienced relatively severe fibrosis of responsive stellate cells
and elucidation of the medullary sheath. However, resolution of some of the glial responses was
noted in the longer duration at lower exposure levels, with no effects observed on the cerebral
white matter in monkeys exposed for 7 months to 1,000 ppm methanol followed by a 6-month
recovery period. In general, the results from the recovery experiment corroborated results
observed in the chronic experiment. NEDO (1987) interpreted the lack of glial effects after a
6-month recovery as an indication of a transient effect. However, glial responses to neural
damage do not necessarily persist following resolution of neurodegeneration (Aschner and
Kimelberg, 1996). In addition, the reported data do not fully support that changes in cerebral
white matter were transient (ERG, 2009). Two of three monkeys exposed to 2,000 ppm exhibited
stellate cell changes in at least one lobe after 1 and 11 months recovery. Also, the only monkey
exposed 7 months with a 1 month recovery period exhibited such changes at autopsy. While the
monkeys exposed to 1,000 ppm for 7 months with a 5 month 20 day recovery period were devoid
4-52
-------
of stellate cell changes, the small sample size (n=2) does not allow for the stellate cell effect to
be characterized as transient.
The limited information available from the NEDO (1987) summary report suggests that
100 ppm (131 mg/m3) may be an effect level following continuous, chronic exposure to
methanol. However, as noted in Section 4.2.2.1, the NEDO (1987) studies in nonhuman
primates, have multiple reporting deficiencies and data gaps that make them difficult to interpret.
In addition, confidence in the dose-response data from this study is weakened by the apparent
lack of a concurrent control group and the small number of animals at each exposure level for
each serial sacrifice (2-3 monkeys/time point/exposure level). In general, peer reviewers of this
study stated that it provides descriptive, rather than quantitative, support for the evaluation of the
inhalation toxicity of methanol (ERG, 2009). Thus, a clear NOAEL or LOAEL cannot be
determined from these monkey studies.
Weiss et al. (1996) exposed 4 cohorts of pregnant Long-Evans rats (10-12 dams/
treatment group/cohort) to 0 or 4,500 ppm (0 and 5,897 mg/m3) methanol vapor (high-
performance liquid chromatography [HPLC] grade), 6 hours/day, from GD6 to PND21. Pups
were exposed together with the dams during the postnatal period. Average blood methanol levels
in pups on PND7 and PND14 were about twice the level observed in dams. However, methanol
exposure had no effect on maternal gestational weight gain, litter size, or postnatal pup weight
gain up to PND1843. Neurobehavioral tests were conducted in neonatal and adult offspring; the
data generated from those tests were evaluated by repeated measures ANOVA. Three
neurobehavioral tests conducted in 13-26 neonates/group included a suckling test, conditioned
olfactory aversion test, and motor activity test. In contrast to earlier test results reported by
Infurna and Weiss (1986), methanol exposure had no effect on suckling and olfactory aversion
tests conducted on PND5 and PND10, respectively. Results of motor activity tests in the
methanol group were inconsistent, with decreased activity on PND18 and increased activity on
PND25. Tests that measured motor function, operant behavior, and cognitive function were
conducted in 8-13 adult offspring/group. Some small performance differences were observed
between control and treated adult rats in the fixed wheel running test only when findings were
evaluated separately by sex and cohort. The test requires the adult rats to run in a wheel and
rotate it a certain amount of times in order to receive a food reward. A stochastic spatial
discrimination test examined the rats' ability to learn patterns of sequential responses. Methanol
exposure had no effect on their ability to learn the first pattern of sequential responses, but
methanol-treated rats did not perform as well on the reversal test. The result indicated possible
43 The fact that this level of exposure caused effects in the Sprague-Dawley rats of the NEDO (1987) study but did
not cause a readily apparent maternal effect in Long-Evans rats of this study could be due to differences in strain
susceptibility.
4-53
-------
subtle cognitive deficits as a result of methanol exposure. A morphological examination of
offspring brains conducted on PND1 and PND21 indicated that methanol exposure had no effect
on neuronal migration, numbers of apoptotic cells in the cortex or germinal zones, or
myelination. However, neural cell adhesion molecule (NCAM) 140 andNCAM 180 gene
expression in treated rats was reduced on PND4 but not 15 months after the last exposure.
NCAMs are glycoproteins required for neuron migration, axonal outgrowth, and establishing
mature neuronal function patterns.
Stanton et al. (1995) exposed 6-7 pregnant female Long-Evans rats/group to 0 or
15,000 ppm (0 and 19,656 mg/m3) methanol vapors (> 99.9% purity) for 7 hours/day on
GD7-GD19. Mean serum methanol levels at the end of the 1st, 4th, 8th, and 12th days of
exposure were 3,836, 3,764, 3,563, and 3,169 |ig/mL, respectively. As calculated by authors,
dams received an estimated methanol dose of 6,100 mg/kg-day. A lower body weight on the first
2 days of exposure was the only maternal effect; there was no increase in postimplantation loss.
Dams were allowed to deliver and nurse litters. Parameters evaluated in pups included mortality,
growth, pubertal development, and neurobehavioral function. Examinations of pups revealed that
two pups from the same methanol-exposed litter were missing one eye; aberrant visually evoked
potentials were observed in those pups. A modest but significant reduction in body weight gain
on PND1, PND21, and PND35 was noted in pups from the methanol group. For example, by
PND35, male pups of dams exposed to methanol had a mean body weight of 129 grams versus
139 grams in controls (p < 0.01). However, postnatal mortality was unaffected by exposure to
methanol. The study authors did not consider a 1.7-day delay in vaginal opening in the methanol
group to be an adverse effect. Preputial separation was not affected by prenatal methanol
exposure. Neurobehavioral status was evaluated using 8 different tests on specific days up to
PND160. Tests included motor activity on PND13-PND21, PND30, and PND60, olfactory
learning and retention on PND18 and PND25, behavioral thermoregulation on PND20-21,
T-maze delayed alternation learning on PND23-PND24, acoustic startle reflex on PND24, reflex
modification audiometry on PND61-PND63, passive avoidance on PND73, and visual evoked
potentials on PND160. A single pup/sex/litter was examined in most tests, and some animals
were subjected to multiple tests. The statistical significance of neurobehavioral testing was
assessed by one-way ANOVA, using the litter as the statistical unit. Results of the
neurobehavioral testing indicated that methanol exposure had no effect on the sensory, motor, or
cognitive function of offspring under the conditions of the experiment. However, given the
comparatively small number of animals tested for each response, it is uncertain whether the
statistical design had sufficient power to detect small compound-related changes.
NEDO (1987) sponsored a teratology study that included an evaluation of postnatal
effects in addition to standard prenatal endpoints in Sprague-Dawley rats. Thirty-six pregnant
4-54
-------
females/group were exposed to 0, 200, 1,000, or 5,000 ppm (0, 262, 1,310, and 6,552 mg/m3)
methanol vapors (reagent grade) on GD7-GD17 for 22.7 hours/day. Statistical significance of
results was evaluated by t-test, Mann-Whitney U test, Fisher's exact test, and/or Armitage's
X2 test.
Postnatal effects of methanol inhalation were evaluated in the remaining 12 dams/group
that were permitted to deliver and nurse their litters. Effects were only observed in the
5,000 ppm. There were no adverse effects on offspring body weight from methanol exposure.
However, the weights of some organs (brain, thyroid, thymus, and testes) were reduced in
8-week-old offspring following prenatal-only exposure to 5,000 ppm methanol. An unspecified
number of offspring were subjected to neurobehavioral testing or necropsy, but results were
incompletely reported.
As described in Section 4.3.2, NEDO (1987) performed a two-generation reproductive
study that evaluated the effects of pre- and postnatal methanol exposure (20 hours/day) on
reproductive and other organ systems of Sprague-Dawley rats and in particular the brain. They
reported reduced brain, pituitary, and thymus weights, in the offspring of F0 and FI rats exposed
to 1,000 ppm methanol. In particular, they noted a reduction in absolute brain weights in Fl pups
at 8 weeks (male and female), 16 weeks (males) and 24 weeks (females) and in F2 pups at 8
weeks (male and female). Details were not reported (e.g., means, variances, sample sizes, pup-to-
litter correlations) that would allow for further analysis of these findings.
Seeking to confirm the possible compound-related effect of methanol on the brain NEDO
(1987) conducted an additional developmental study in which Sprague-Dawley rats were
exposed to 0, 500, 1,000, and 2,000 ppm (0, 655, 1,310, and 2,620 mg/m3) methanol from the
first day of gestation through the FI generation. According to NEDO (1987 page 201 ), another
purpose of the supplementary study was "to know from what period after birth such changes
would appear." Information important for a determination of possible litter correlations (e.g., pup
litter assignments) was not reported for the supplemental experiment. However, the number of
pups per dose group per "period afterbirth" was reported (11-14/sex/dose/postnatal period) and
it is reasonable to assume that, consistent with the standard culling protocol used for both the Fl
and F2 generations of the two-generation study (NEDO, 1987 pages 185 and 189 ), the pups for
each gender, dose and exposure time combination came from a different litter (to avoid problems
associated with litter correlation). Brain weights were measured in the 11-14 offspring/sex/group
at 3, 6, and 8 weeks of age. As illustrated in Table 4-13, brain weights were significantly reduced
in 3-week-old males and females exposed to > 1,000 ppm. At 6 and 8 weeks of age, brain
weights were significantly reduced in males exposed to > 1,000 ppm and females exposed to
2,000 ppm. Due to the toxicological significance of this postnatal effect, the brain weight
changes observed by NEDO (1987) following gestational and postnatal exposures and following
4-55
-------
gestation-only exposure (in the teratology study discussed above) are evaluated quantitatively
and discussed in more detail in Section 5 of this review.
Table 4-13 Brain weights of rats exposed to methanol vapors during gestation and
lactation.
Brain weight (g)
Offspring age
3 weeks3
3 weeks3
6 weeks3
6 weeks3
8 weeks3
8 weeks3
8 weeks'3
8 weeks'3
Sex
Male
Female
Male
Female
Male
Female
Male
Female
Mean ± SD
0 ppm
1.45 ±0.06
1.41 ±0.06
1.78 ±0.07
1.68 ±0.08
1.99 ±0.06
1.85 ±0.05
2.00 ±0.05
1.86 ±0.08
Mean ± SD, (% of control) at each exposure level
200 ppm 500 ppm
1.46 ±0.08
(101%)
1.41 ±0.07
(100%)
1.74 ±0.09
(98%)
1.71 ±0.08
(102%)
1.98 ±0.09
(99%)
1.83 ±0.07
(99%)
(100%)
(103%)
1,000 ppm
1.39±0.05C
(96%)
1.33±0.07d
(94%)
1.69±0.06d
(95%)
1.62 ±0.07
(96%)
1.88±0.08d
(94%)
1.80 ±0.08
(97%)
1.99 ±0.07
(100%)
1.90 ±0.08
(102%)
2,000 ppm
1.27±0.06e
(88%)
1.26±0.09e
(89%)
1.52±0.07e
(85%)
1.55±0.05e
(92%)
1.74±0.05e
(87%)
1.67±0.06e
(90%)
-
-
5,000 ppm
-
-
-
-
-
-
1.81 ±0.16d
(91%)
1.76± 1.09
(95%)
aExposed throughout gestation and F! generation.
""Exposed on gestational days 7-17 only.
°p < 0.05;
dp<0.01;
ep< 0.001;
p values as calculated by the authors. Values are means ± SD
Source: NEDO (1987).
Burbacher et al. (1999a; 1999b) carried out toxicokinetic, reproductive, developmental
and postnatal neurological and neurobehavioral studies of methanol in M fascicularis monkeys
that were published by HEI in a two-part monograph. Some of the data were subsequently
published in the open scientific literature (Burbacher et al., 2004a: Burbacher et al., 2004b). The
experimental protocol featured exposure to 2 cohorts of 12 monkeys/group to low-exposure
levels (relative to the previously discussed rodent studies) of 0, 200, 600, or 1,800 ppm (0, 262,
786, and 2,359 mg/m3) methanol vapors (99.9% purity), 2.5 hours/day, 7 days/week, during a
premating period and mating period (-180 days combined) and throughout the entire gestation
period (-168 days). The monkeys were 5.5-13 years old. The outcome study included an
evaluation of maternal reproductive performance (discussed in Section 4.3.2) and tests to assess
infant postnatal growth and newborn health, neurological outcomes included reflexes, behavior,
4-56
-------
and development of visual, sensorimotor, cognitive, and social behavioral function. Blood
methanol levels, elimination, and the appearance of formate were also examined and are
discussed in Section 3.2. The effects observed were in the absence of appreciable increases in
maternal blood formate levels.
Neurobehavioral function was assessed in 8-9 infants/group during the first 9 months of
life (Burbacher et al., 2004b: Burbacher et al., 1999b). Although results in 7/9 tests were
negative, 2 effects were possibly related to methanol exposure. The Visually Directed Reaching
(VDR) test is a measure of sensorimotor development and assessed the infants' ability to grasp
for a brightly colored object containing an applesauce-covered nipple. Beginning at 2 weeks after
birth, infants were tested 5 times/day, 4 days/week. Performance on this test, measured as age
from birth at achievement of test criterion (successful object retrieval on 8/10 consecutive trials
over 2 testing sessions), was reduced in all treated male infants. The times (days after birth) to
achieve the criteria for the VDR test were 23.7 ± 4.8 (n = 3), 32.4 ± 4.1 (n = 5), 42.7 ± 8.0
(n = 3), and 40.5 ± 12.5 (n = 2) days for males and 34.2 ± 1.8 (n = 5), 33.0 ± 2.9 (n = 4),
27.6 ± 2.7 (n = 5), and 40.0 ± 4.0 (n = 7) days for females in the control to 1,800 ppm groups,
respectively. Statistical significance was obtained in the 1,800 ppm group when males and
females were evaluated together (p = 0.04) and in the 600 ppm (p = 0.007) for males only.
However, there was no significant difference between responses and/or variances (indicating lack
of a dose-response trend) among the dose levels for males and females combined (p = 0.244), for
males only (p = 0.321) and for males only, excluding the high-dose group (p = 0.182). However,
there was a significant dose-response trend for females only (p = 0.0265). The extent to which
VDR delays were due to a direct effect of methanol on neurological development or a secondary
effect due to the methanol-induced decrease in length of pregnancy and subsequent prematurity
is not clear. Studies of reaching behavior have shown that early motor development in pre-term
human infants without major developmental disorders differs from that of full-term infants
(Fallang et al., 2003). Clinical studies have indicated that the quality of reaching and grasping
behavior in pre-term infants is generally less than that in full-term infants (Fallang et al., 2003;
Plantinga et al., 1997). For this reason, measures of human infant development generally involve
adjustment of a child's "test age" if he or she had a gestational age of fewer than 38 weeks, often
by subtracting weeks premature from the age measured from birth (Wilson and Cradock, 2004).
When this type of adjustment is made to the Burbacher et al. (2004b; 1999b) VDR data, the
dose-response trend for males only remains unacceptable (p = 0.448) and, while the dose-
response trend for the females only remains adequate (p = 0.009), the variance in the data could
not be modeled adequately. Thus, only the unadjusted VDR response for females only exhibited
a dose-response that could be adequately modeled (see Appendix D).
4-57
-------
At 190-210 days of age, the Pagan Test of infant intelligence was conducted. The
paradigm makes use of the infant's proclivity to direct more visual attention to novel stimuli
rather than familiar stimuli. The test measures the time infants spend looking at familiar versus
novel items. Deficits in the Pagan task can qualitatively predict deficits in intelligence quotient
(IQ) measurements assessed in children at later ages (Pagan and Singer, 1983). Control monkey
infants in the Burbacher et al. (2004b; 1999b) study spent more than 62% ± 4% (mean for both
cohorts) of their time looking at novel versus familiar monkey faces, while the treated monkeys
did not display a statistically significant preference for the novel faces (59% ± 2%, 54% ± 2%
and 59% ± 2% in 200, 600 and 1,800 ppm groups, respectively). Unlike the VDR results
discussed previously, results of this test did not appear to be gender specific and were neither
statistically significant (ANOVA/? = 0.38) nor related to exposure concentration. The findings
indicated a cohort effect which appeared to reduce the statistical power of this analysis. The
authors' exploratory analysis of differences in outcomes between the 2 cohorts indicated an
effect of exposure in the second cohort and not the first cohort due to higher mean performance
in controls of cohort 2 (70% ± 5% versus 55% ± 4% for cohort 1). In addition, this finding could
reflect the inherent constraints of this endpoint. If the control group performs at the 60% level
and the most impaired subjects perform at approximately the 50% chance level (worse than
chance performance would not be expected), the range over which a concentration-response
relationship can be expressed is limited. Because of the longer latency between assessment and
birth, these results would not be confounded with the postulated methanol-induced decrease in
gestation length of the exposed groups of this study. Negative results were obtained for the
remaining seven tests that evaluated early reflexes, gross motor development, spatial and concept
learning and memory, and social behavior. Infant growth and tooth eruption were unaffected by
methanol exposure.
4.4.3. Neurotoxicity Studies Employing i.p. and in vitro Methanol Exposures
Table 4-14 describes three i.p. injection studies that attempt to determine the biochemical
changes associated with the effects of repeat methanol exposures on the brain, retina, optic nerve
(Rajamani et al., 2006; Gonzalez-Quevado et al., 2002) and the hypothalamus-pituitary-adrenal
(HPA) axis of the neuroendocrine system (Parthasarathy et al., 2006b). The goal of the Gonzalez-
Quevado et al. (2002) study was to determine whether a sustained increase in formate levels,44 at
concentrations below those known to produce toxic effects from acute exposures, can induce
44 Formate levels were increased by treating test rats with methotrexate (MTX), which depletes folate stores by
interfering with tetrahydrofolate (THF) regeneration (Dorman et al.. 1994).
4-58
-------
biochemical changes in the retina, optical nerve, or certain regions of the brain. 45 The amino
acids aspartate, glutamate, asparagine, serine, histidine, glutamine, threonine, glycine, arginine,
alanine, hypotaurine, gamma-aminobutyric acid (which is also a neurotransmitter), and tyrosine
were measured in blood, brain, and retinal regions.
The increased level of aspartate in the optic nerve of animals treated with MTX-methanol
and Tau-MTX-methanol may indicate a relation to formate accumulation. The authors note that
L-aspartate is a major excitatory amino acid in the brain and that increased levels of excitatory
amino acids can trigger neuronal cell damage and death (Albin and Greenamyre, 1992).
Increased levels of aspartate and glutamine in the hippocampus could provide an explanation for
some of the CNS symptoms observed in methanol poisonings on the basis of their observed
impact on cerebral arteries (Huang et al., 1994). The observation that these increases resulted
primarily from methanol without MTX could be significant in that it indicates methanol can
cause excitotoxic effects without formate mediation. The neurotransmitters serotonin (5-HT) and
dopamine (DA) and their respective metabolites, 5-hydroxyindolacetic acid (5-HIAA) and
dihydroxyphenylacetic acid (DOPAC), were also measured in various brain regions. The levels
of these monoamines were not affected by formate accumulation, as the only increases were
observed for 5-HT and 5-HIAA following methanol-only exposure. DA and DOPAC levels were
not altered by the treatments in any of the areas measured. The posterior cortex did not show any
changes in monoamine levels for any treatment group.
Rajamani et al. (2006) examined several oxidative stress parameters in male Wistar rats
following methotrexate-induced folate deficiency. The optic nerve, retina, and brain were
collected and the brain was dissected into the following regions: cerebral cortex, cerebellum,
mid-brain, pons medulla, hippocampus, and hypothalamus. Each region was examined for
indicators of oxidative stress including increases in the free radical scavengers: superoxide
dismutase (SOD), CAT, glutathione peroxidase (GPx); and reduced GSH levels. The levels of
protein thiols, protein carbonyls, and amount of lipid peroxidation were also measured. More
recently, investigators from the same laboratory measured increased methanol blood levels and
corresponding increases in these indicators of oxidative stress in discrete regions of the brain in
Wistar strain albino rats exposed to 75 mg/kg/day aspartame (lyyaswamy and Rathinasamy,
2012). Overall, the results reported in these studies suggest that folate-deficient rats exposed to
methanol exhibit signs of oxidative stress (e.g., increased SOD, GPx and CAT activity and
decreased levels of GSH and protein thiol) in discrete regions of the brain, retina, and optic
nerve.
45 A subset of exposed rats were also exposed to taurine, which plays an important role in the retina and optical
nerve, to explore its possible protective effect (Gonzalez-Quevado et al.. 2002).
4-59
-------
To determine the effects of methanol on the HPAaxis, Parthasarathy et al. (2006b)
evaluated a combination of oxidative stress, immune and neurobehavioral parameters following
methanol exposure. Oxidative stress parameters examined included SOD, CAT, GSH peroxidase,
GSH, and ascorbic acid (Vitamin C). Plasma corticosterone levels were measured, and lipid
peroxidation was measured in the hypothalamus and the adrenal gland. An assay for DNA
fragmentation was conducted in tissue from the hypothalamus, the adrenal gland and the spleen.
Immune function tests conducted included the footpad thickness test for delayed type
hypersensitivity (DTH), a leukocyte migration inhibition assay, the hemagglutination assay
(measuring antibody titer), the neutrophil adherence test, phagocytosis index, and a nitroblue
tetrazolium (NET) reduction and adherence assay used to measure the killing ability of
polymorphonuclear leukocytes (PMNs). The open field behavior test was used to measure
general locomotor and explorative activity during methanol treatment in the 30-day treatment
group, with tests conducted on days 1, 4, 8, 12, 16, 20, 24, and 28.
The results for this study shown in Table 4-14 suggest that exposure to methanol-induced
oxidative stress, disturbs HPA-axis function, altering corticosterone levels and producing effects
in several nonspecific and specific immune responses.
4-60
-------
Table 4-14 Intraperitoneal injection neurotoxicity studies.
Species/Strain/N Dose & Duration
Effect Relative to Control
Reference
Rat/Sprague-
Dawley/
(5-7 per group;
100-150 g)
Control: tap water (wk 1);
saline s.c. (wks 2-4)
MeOH: tap water (wk 1);
s.c. saline (wk2);
2 g/kg-day MeOH i.p.
(wks 3-4)
Increased blood formate (<2-fold); Increased
aspartate, glutamine and Tau in hippocampus;
Increased 5-HT and 5-HIAA in hippocampus;
Increased 5-HT in retina
MIX: tap water (wk 1);
0.2 mg/kg-day MIX s.c.
(wk2); 0.1 mg/kg-day MIX
s.c. & saline i.p. (wks 2-4)
No change in blood formate or any other measured
parameter
MTX-MeOH: tap water
(wk1);
0.2 mg/kg-day MIX s.c.
(wk2); 0.1 mg/kg-day MIX
s.c. & 2 g/kg -day MeOH
i.p. (wks 3-4)
Increased blood formate (>3-fold); Increased aspartate
in optic nerve; Increased aspartate and Tau in
hippocampus
Gonzalez-
Quevado
et al. (2002)
Tau: 2% Tau in DW
(wks 1-4);
saline s.c. (wks 2-4)
No change in blood formate; Increased blood histidine
and Tau
Tau-MTX-MeOH: 2% Tau
in DW (wks 1-4);
0.2 mg/kg-day MTX s.c.
(wk2); 0.1 mg/kg-day MTX
s.c. & 2 g/kg-day MeOH
i.p. (wks 3-4)
Increased blood formate (>3-fold) and Tau; Increased
aspartate in optic nerve; Increased aspartate,
glutamine and Tau in hippocampus
Rat/Wistar/
6 per group
Control: saline i.p. (day 8)
MTX: 0.2 mg/kg-day MTX
(wk 1);
saline i.p. (day 8)
Increased SOD, CAT, GSH peroxidase, oxidized
GSH, protein carbonyls and lipid peroxidation in all
brain regions; Decreased GSH and protein thiols in all
brain regions; Increased HSP70 in hippocampus
MTX-MeOH: 0.2 mg/kg-
day MTX (wk 1);
3 g/kg-day MeOH i.p.
(day 8)
Increased SOD, CAT, GSH peroxidase, oxidized
GSH, protein carbonyls and lipid peroxidation in all
brain regions over control and MTX group; Decreased
GSH and protein thiols in all brain regions over control
and MTX group; Increased HSP70 in hippocampus
Rajamani
et al. (2006)
4-61
-------
Table 4-14 (Continued): Intraperitoneal injection neurotoxicity studies.
Species/Strain/N Dose & Duration Effect Relative to Control Reference
Rat/Wistar/ 0 or 2.37 g/kg-day MeOH All antioxidants increased at 1-day, but decreased at
6 per group i.p. for 1, 15 or 30 days 15 and 30 days; Increased lipid peroxidation in
hypothalamus and adrenal gland at 1, 15, and
30 days; Increased leukocyte migration and antibody
titer at all time points; Decreased footpad thickness at
15 and 30 days; Decreased neutrophil adherence at 1
and 30 days. Decreased NET reduction and
adherence in PMNs at 30-days versus PMNs at 15- Parthasarathy
days; Decrease in ambulation from 4th day on; et al. (2006b)
Decrease in rearing and grooming from 20th day on.
Increase in immobilization from 8th day on; Increase
fecal bolus from 24th day on; Increase in
corticosterone levels at 1 and 15 days; Decrease in
corticosterone levels at 30 days; Fragmentation of
DMA from hypothalamus, adrenal gland, and spleen at
30 days.
wk = week; MeOH = methanol; s.c. = subcutaneous injection; i.p.= intraperitoneal injection; MIX = methotrexate; Tau = taurine; DW
= drinking water ad libitum exposure
There is some experimental evidence that the presence of methanol can affect the activity
of acetylcholinesterase (Tsakiris et al., 2006). Although these experiments were carried out on
erythrocyte membranes in vitro, the apparent compound-related changes may have implications
for possible impacts of methanol and/or its metabolites on acetylcholinesterase at other centers,
such as the brain. Tsakiris et al. (2006) prepared erythrocyte ghosts from blood samples of
healthy human volunteers by repeated freezing-thawing. The ghosts were incubated for 1 hour at
37°C in 0, 0.07, 0.14, 0.6, or 0.8 mmol/L methanol, and the specific activities of
acetylcholinesterase monitored. Respective values (in change of optical density units/minute-mg
protein) were 3.11 ± 0.15, 2.90 ± 0.10, 2.41 ± 0.10 (p < 0.05), 2.05 ± 0.11 (p < 0.01), and
1.81 ± 0.09 (p < 0.001). More recently, Simintzi et al. (2007) carried out an in vitro experiment
to investigate the effects of aspartame metabolites, including methanol, on 1) a pure preparation
of acetylcholinesterase, and 2) the same activity in homogenates of frontal cortex prepared from
the brains of (both sexes of) Wistar rats. The activities were measured after incubations with 0,
0.14, 0.60, or 0.8 mmoles/L (0, 4.5, 19.2, and 25.6 mg/L) methanol, and with methanol mixed
with the other components of aspartame metabolism, phenylalanine and aspartic acid. After
incubation at 37 °C for 1 hour, the activity of acetylcholinesterase was measured
spectrophotometrically. As shown in Table 4-15, the activities of the acetylcholinesterase
preparations were reduced dose dependently after incubation in methanol. Similar results were
also obtained with the other aspartame metabolites, aspartic acid, and phenylalanine, both
individually or as a mixture with methanol. While the implications of this result to the acute
neurotoxicity of methanol are uncertain, the authors speculated that methanol may bring about
4-62
-------
these changes through either interactions with the lipids of rat frontal cortex or perturbation of
proteinaceous components.
Table 4-15 Effect of methanol on Wistar rat acetylcholinesterase activities.
..... . .. Acetylcholinesterase activity (AOD/min-mq)
Methanol concentration — —
(mmol/L) Frontal cortex Pure enzyme
Control 0.269 ±0.010 1.23 ±0.04
0.14 0.234 ±0.007a 1.18 ±0.06
0.60 0.223 ± 0.009b 1.05±0.04b
0.80 0.204 ±0.008b 0.98±0.05b
ap < 0.01.
bp< 0.001.
Values are means ± SD for four experiments. The average value of each experiment was derived from three determinations of each
enzyme activity.
Source: Reprinted with permission of Elsevier; Simintzi et al. (2007).
4.5. Immunotoxicity
Parthasarathy et al. (2005b) provided data on the impact of methanol on neutrophil
function in an experiment in which 6 male Wistar rats/group were given a single i.p. exposure of
2,370 mg/kg methanol mixed 1:1 in saline. Another group of 6 animals provided blood samples
that were incubated with methanol in vitro at a methanol concentration equal to that observed in
the in vivo-treated animals 30 and 60 minutes postexposure. Total and differential leukocyte
counts were measured from these groups in comparison to in vivo and in vitro controls.
Neutrophil adhesion was determined by comparing the neutrophil index in the untreated blood
samples to those that had been passed down a nylon fiber column. The cells' phagocytic ability
was evaluated by their ability to take up heat-killed Candida albicans. In another experiment,
neutrophils were assessed for their killing potential by measuring their ability to take up then
convert NET to formazan crystals.46 One hundred neutrophils/slides were counted for their total
and relative percent formazan-positive cells.
The blood methanol concentrations 30 and 60 minutes after dosing were 2,356 ±162 and
2,233 ±146 mg/L, respectively. The mean of these values was taken as the target concentration
for the in vitro methanol incubation. In the in vitro studies, there were no differences in total and
46 Absence of NET reduction indicates a defect in some of the metabolic pathways involved in intracellular
microbial killing.
4-63
-------
differential leukocyte counts, suggesting that no lysis of the cells had occurred at this methanol
concentration. This finding contrasts with the marked difference in total leukocytes observed as a
result of methanol incubation in vivo, in which, at 60 minutes after exposure, 16,000 ± 1,516
cells/mm3 were observed versus 23,350 ± 941 in controls (p < 0.001). Some differences in
neutrophil function were observed in blood samples treated with methanol in vitro and in vivo.
These differences are illustrated for the 60-minute postexposure samples in Table 4-16.
Table 4-16 Effect of methanol on neutrophil functions in in vitro and in vivo studies in
male Wistar rats.
Parameter
Phagocytic index (%)
Avidity index
NET reduction (%)
Adherence (%)
In vitro studies (60
Control
89.8 ±3.07
4.53 ±0.6
31. 6 ±4.6
50.2 ±5.1
minutes)
Methanol
81.6±2.2a
4.47 ±0.7
48.6±4.3b
39.8 ± 2.4a
In vivo studies
Control
66.0 ±4.8
2.4 ±0.1
4.6 ± 1.2
49.0 ±4.8
(60 minutes)
Methanol
84.0±7.0b
3.4±0.3a
27.0±4.6b
34.6±4.0b
ap < 0.01 .
bp< 0.001.
Values are means ± SD for six animals.
Source: Reprinted with permission of Taylor & Francis; Parthasarathy et al. (2005b).
Parthasarathy et al. (2005b) observed differences in the neutrophil functions of cells
exposed to methanol in vitro versus in vivo, most notably in the phagocytic index that was
reduced in vitro but significantly increased in vivo. However, functions such as adherence and
NET reduction showed consistency in the in vitro and in vivo responses. The authors noted that,
by and large, the in vivo effects of methanol on neutrophil function were more marked than those
in cells exposed in vitro.
Another study by Parthasarathy et al. (2005a) also exposed 6 male Wistar rats/group i.p.
to methanol at approximately 1/4 the LD50 (2.4 g/kg). The goal was to further monitor possible
methanol-induced alterations in the activity of isolated neutrophils and other immunological
parameters. The exposure protocol featured daily injections of methanol for up to 30 days in the
presence or absence of sheep RBCs. Blood samples were assessed for total and differential
leukocytes, and isolated neutrophils were monitored for changes in phagocytic and avidity
indices, NET reduction, and adherence. In the latter test, blood samples were incubated on a
nylon fiber column, then eluted from the column and rechecked for total and differential
leukocytes. Phagocytosis was monitored by incubating isolated buffy coats from the blood
samples with heat-killed C. albicans. NET reduction capacity examined the conversion of the
dye to formazan crystals within the cytoplasm. The relative percentage of formazan-positive cells
4-64
-------
in each blood specimen gave a measure of methanol's capacity to bring about cell death. As
tabulated by the authors, there was a dose-dependent reduction in lymphoid organ weights
(spleen, thymus, and lymph node) in rats exposed to methanol for 15 and 30 days via i.p.
injection, irrespective of the presence of sheep RBCs. Methanol also appeared to result in a
reduction in the total or differential neutrophil count. These and potentially related changes to
neutrophil function are shown in Table 4-17.
Table 4-17 Effect of intraperitoneally injected methanol on total and differential leukocyte
counts and neutrophil function tests in male Wistar rats.
Without sheep red blood cell treatment
Parameter
Control
15-day
methanol
30-day
methanol
With sheep red blood cell treatment
Control
15-day
methanol
30-day
methanol
Organ weights (mg)
Spleen
Thymus
Lymph node
1,223 ±54
232 ± 12
32 ±2
910±63a
171 ±7a
24±3a
696 ± 83a'b
121 ± 10a'b
16 ± 2a'b
1,381 ±27
260 ±9
39 ±2
1,032±39a
172 ± 10a
28 ± 1a
839 ± 35a'b
130±24a'b
23 ± 1a'b
Leukocyte counts
Total leukocytes
% neutrophils
% Lymphocytes
Neutrophil function
Phagocytic
index (%)
Avidity index
NET reduction
(%)
Adherence (%)
23,367
±946
24 ±8
71 ±7
tests
91.0±2.0
2.6 ±0.3
6.3 ±2.0
49.0 ±5.0
16,592
± 1,219a
21 ±3
76 ±3
80.0±4.0a
3.2± 0.5a
18.2±2.0a
44.0 ±5.0
13,283
± 2,553a'b
16±3a
79 ±5
79.0±2.0a
3.2±0.1a
15.0 ± 1.0a'b
29.5 ± 5.0a'b
18,633
±2,057
8±3
89 ±4
87.0 ±4.0
4.1 ±0.1
32.0 ±3.3
78.0 ±9.2
16,675
± 1,908
23±4a
78.5 ±4a
68.0 ± 3.0a
2.6±0.3a
22.0 ± 3.0a
52.0 ± 9.0a
14,067
± 930a'b
15±5a'b
82 ±6
63.0±4.0a
2.1 ±0.3a
19.0±2.4a
30.0 ± 4.3a'b
ap < 0.05 from respective control.
bp < 0.05 between 15-and 30-day treatment groups.
Values are means ± SD (n = 6).
Source: Reprinted with permission of Taylor & Francis; Parthasarathy et al. (2005a).
The study provided data that showed altered neutrophil functions following repeated
daily exposures of rats to methanol for periods up to 30 days. This finding is indicative of a
possible effect of methanol on the immunocompetence of an exposed host.
Parthasarathy et al. (2006a) reported on additional immune system indicators as part of a
study to determine the effects of methanol intoxication on the HPAaxis. As described in
Section 4.4.3, immune function tests conducted included the footpad thickness test for DTH, a
leukocyte migration inhibition assay, the hemagglutination assay (measuring antibody titer), the
4-65
-------
neutrophil adherence test, phagocytosis index, and a NET reduction and adherence assay used to
measure the killing ability of PMNs.
Leukocyte migration and antibody titer were both significantly increased over controls
for all time points, while footpad thickness was significantly deceased in 15- and 30-day treated
animals. Neutrophil adherence was significantly decreased after 1 and 30 days of exposure. A
significant decrease in the NET reduction and adherence was found when comparing PMNs from
the 30-day treated animals with cells from the 15-day methanol-treated group.
Parthasarathy et al. (2007) reported the effects of methanol on a number of specific
immune functions. As before, 6 male Wistar rats/group were treated with 2,370 mg/kg methanol
in a 1:1 mixture in saline administered intraperitoneally for 15 or 30 days. Animals
scheduled/designated for termination on day 15 were immunized intraperitoneally with 5 x 109
sheep RBCs on the 10th day. Animals scheduled for day 30 termination were immunized on the
25th day. Controls were animals that were not exposed to methanol but immunized with sheep
RBCs as described above. Blood samples were obtained from all animals at sacrifice and
lymphoid organs including the adrenals, spleen, thymus, lymph nodes, and bone marrow were
removed. Cell suspensions were counted and adjusted to 1 x 108 cells/mL. Cell-mediated
immune responses were assessed using a footpad thickness assay and a leukocyte migration
inhibition (LMI) test, while humoral immune responses were determined by a hemagglutination
assay, and by monitoring cell counts in spleen, thymus, lymph nodes, femoral bone marrow, and
in splenic lymphocyte subsets. Plasma levels of corticosterone were measured along with levels
of such cytokines as TNF-a, IFN-y, IL-2, and IL-4. DNA damage in splenocytes and thymocytes
was also monitored using the Comet assay.
Table 4-18 shows decreases in the animal weight/organ weight ratios for spleen, thymus,
lymph nodes and adrenal gland as a result of methanol exposure. However, the splenocyte,
thymocyte, lymph node, and bone marrow cell counts were time-dependently lower in methanol-
treated animals.
4-66
-------
Table 4-18 Effect of methanol exposure on animal weight/organ weight ratios and on cell
counts in primary and secondary lymphoid organs of male Wistar rats.
Immunized
Organ
Control
15 days
30 days
Animal weight/organ weight ratio
Spleen
Thymus
Lymph node
Adrenal
3.88 ±0.55
1.35 ±0.29
0.10±0.01
0.14 ±0.01
2.85 ± 0.36a
0.61 ± 0.06a
0.08±0.01a
0.15 ±0.01
2.58±0.45a
0.63±0.04a
0.06±0.02a
0.12±0.01a'b
Cell counts
Splenocytes (x 108)
Thymocytes (x 108)
Lymph node (x 107)
Bone marrow (x 107)
5.08 ±0.06
2.66 ±0.09
3.03 ±0.04
4.67 ±0.03
3.65±0.07a
1.95±0.03a
2.77 ± 0.07a
3.04 ± 0.09a
3.71 ± 0.06a
1.86±0.09a
2.20±0.06a'b
2.11 ±0.05a'b
Values are means ± six animals. ap < 0.05 versus control groups, p < 0.05 versus 15-day treated group.
Source: Reprinted with permission of Springer; Parthasarathy et al. (2007).
Parthasarathy et al. (2007) also documented their results on the cell-mediated and
humoral immunity induced by methanol. Leukocyte migration was significantly increased
compared to control animals, an LMI of 0.82 ± 0.06 being reported in rats exposed to methanol
for 30 days. This compares to an LMI of 0.73 ± 0.02 in rats exposed for 15 days and 0.41 ± 0.10
in controls. By contrast, footpad thickness and antibody titer were decreased significantly in
methanol-exposed animals compared to controls (18.32 ± 1.08, 19.73 ± 1.24, and 26.24 ± 1.68%
for footpad thickness; and 6.66 ± 1.21, 6.83 ± 0.40, and 10.83 ± 0.40 for antibody titer in 30-day,
15-day exposed rats, and controls, respectively).
Parthasarathy et al. (2007) also provided data in a histogram that showed a significant
decrease in the absolute numbers of Pan T cells, CD4, macrophage, major histocompatibility
complex (MHC) class II molecule expressing cells, and B cells of the methanol-treated group
compared to controls. The numbers of CDS cells were unaffected. Additionally, as illustrated in
the report, DNA single strand breakage was increased in immunized splenocytes and thymocytes
exposed to methanol versus controls. Although some fluctuations were seen in corticosterone
levels, the apparently statistically significant change versus controls in 15-day exposed rats was
offset by a decrease in 30-day exposed animals. Parthasarathy et al. (2007) also tabulated the
impacts of methanol exposure on cytokine levels; these values are shown in Table 4-19.
4-67
-------
Table 4-19 The effect of methanol on serum cytokine levels in male Wistar rats.
Cytokines (pg/mL)
IL-2
IL-4
TNF-a
IFN-Y
Control
1,810 ±63.2
44.8 ±2.0
975 ±32.7
1,380 ±55.1
Immunized
15 days
1, 303.3 ±57.1a
74.0±5.1a
578.3 ± 42.6a
961.6±72.7a
30 days
1, 088.3 ±68.8a'b
78.8±4.4a
585 ± 45a
950 ± 59.6a
Bp < 0.05 versus control groups.
bp < 0.05 versus 15-day treated group.
Values are means ± six animals.
Source: Reprinted with permission of Springer; Parthasarathy et al.(2007).
Drawing on the results of DNA single strand breakage in this experiment, the authors
speculated that methanol-induced apoptosis could suppress specific immune functions such as
those examined in this research report. Methanol appeared to suppress both humoral and cell-
mediated immune responses in exposed Wistar rats.
4.6. Synthesis of Major Noncancer Effects
4.6.1. Summary of Key Studies in Methanol Toxicity
A substantial body of information exists on the toxicological consequences to humans
who consume or are acutely exposed to large amounts of methanol. Neurological and
immunological effects have been noted in adult human subjects acutely exposed to as low as
200 ppm (262 mg/m3) methanol (Mann et al.. 2002: Chuwers et al.. 1995). Nasal irritation effects
have been reported by adult workers exposed to 459 ppm (601 mg/m3) methanol. Frank effects
such as blurred vision and bilateral or unilateral blindness, coma, convulsions/tremors, nausea,
headache, abdominal pain, diminished motor skills, acidosis, and dyspnea begin to occur as
blood levels approach 200 mg methanol/L, and 800 mg/L appears to be the threshold for
lethality. Data for subchronic, chronic, or in utero human exposures are very limited.
Determinations regarding longer term effects of methanol are based primarily on animal studies.
An end-point-by-end-point survey of the primary noncancer effects of methanol in
experimental animals is given in the following paragraphs. Tabular summaries of the principal
toxicological studies that have examined the noncancer effects of methanol when experimental
animals were exposed to methanol via the oral or inhalation routes are provided in Tables 4-20
and 4-21. Figures 4-1 and 4-2 graphically depict the oral and inhalation exposure-response
information from these studies, illustrating the relationship between NOAELs and LOAELs that
4-68
-------
have been identified. Most studies focused on developmental and reproductive effects. A large
number of the available studies were performed by routes of exposure (e.g., i.p.) that are less
relevant to the assessment. The data are summarized in separate sections that address oral
exposure (Section 4.6.1.1) and inhalation exposure (Section 4.6.1.2).
Table 4-20 Summary of noncancer effects reported in repeat exposure and developmental
studies of methanol toxicity in experimental animals (oral).
Species, strain,
number/sex
Rat
Sprague-Dawley
3 0/s ex/group
Rat
Sprague-Dawley
100/sex/group
Mouse
Swiss
NOAEL
Dose/duration (mg/kg-day)
0, 100, 500, and
2,500 mg/kg-day for 500
13 wk
0, 500, 5,000, or
20,000 ppm (v/v) in
drinking water, for
104 wk. Doses were
approx. 0, 46.6, 466, ND
and 1,872 mg/kg-day
(male) and 0, 52.9,
529, and 2, 101
mg/kg-day (female)
560, 1,000 and 2,100
mg/kg/day (female)
and 550, 970, and
1,800 mg/kg/day a/u ''UUU
(male), 6 days/wkfor
life
LOAEL
(mg/kg-day) Effect
Reduction of brain
weights, increase in the
2,500 serum activity of ALT
and AP. Increased liver
weights
...-. No noncancer effects
were reported
Increased incidence of
1,800-2,100 liver parenchymal cell
necrosis
Reference
TRL (1986)
Soffritti et al.
(2002)
Apaja (1980)
Reproductive/developmental toxicity studies
Rat
Long-Evans
10 pregnant
females/group
Mouse
CD-1
8 pregnant
females and 4
Controls
0 and 2,500 mg/kg-
day on either ...
GD15-GD17or
GD17-GD19.
4 g/kg-day in 2 daily ...
dosesonGD6-GD15
Neurobehavioral
2 500 deficits (such as
homing behavior,
suckling ability
Increased incidence of
totally resorbed litters,
4 000 cleft palate and
exencephaly. A
decrease in the number
of live fetuses/litter
Infurna and
Weiss (1986)
Rogers et al.
(1993b)
NA = Not applicable; ND = Not determined; M= male, F=female.
4-69
-------
Table 4-21 Summary of repeat exposure and developmental studies of methanol toxicity in
experimental animals (inhalation exposure).
Species, strain,
number/sex
Monkey
M. fascicularis,
1 or 2 animals/group
Dog (2)
Rat
Sprague-Dawley
5 males/ group
Rat
Sprague-Dawley
5 males/ group
Rat
Sprague-Dawley
5/sex/group
Monkey
M. fascicularis
3/s ex/group
Rat
Sprague-Dawley
10/sex/group
Rat
Sprague-Dawley
1 5/sex/group
Monkey
M. fascicularis
2 or 3 animals/
group/time point
Rat
F344
20/sex/group
Mouse
B6C3F-,
3 0/s ex/group
Mouse
B6C3F-,
52-53/sex/group
Rat
F344
52/sex/group
Dose/duration
0, 3,000, 5,000, 7,000,
or 10,000 ppm, 21
hr/day, for up to 14
days
10,000 ppm for 3 min,
8 times/day for 100
days
0, 200, 2000, or
10,000 ppm, 8 hr/day,
5 days/wkfor up to 6
wk
0, or 200 ppm,
6 hr/day, for either 1 or
7 days
0, 500, 2,000, or
5,000 ppm, 5 days/wk
for 4 wk
0, 500, 2,000, or
5,000 ppm, 5 days/wk
for 4 wk
0, 300, or 3, 000 ppm, 6
hr/day, 5 days/wk for 4
wk
0 or 2, 500 ppm, 6
hr/day, 5 days/wk for 4
wk
0, 10, 100, or
1,000 ppm, 21 hr/day
for either 7, 19, or 29
mo
0, 10, 100, or
1,000 ppm, 20 hr/day,
for 12 mo
0, 10, 100, or
1,000 ppm, 20 hr/day,
for 12 mo
0, 10, 100, or
1,000 ppm, 20 hr/day,
for 18 mo
0, 10, 100, or
1,000 ppm, -20 hr/day
for 2 yr
NOAEL LOAEL
(ppm) (ppm) Effect
Clinical signs of toxicity, CMS
changes, including degeneration of
ND ND a the bilateral putamen, caudate
nucleus, and claustrum. Edema of
cerebral white matter.
NA NA None
... „-. Transient reduction in plasma
testosterone levels
MA 9nn Transient reduction in plasma
testosterone levels
5,000 NA No compound-related effects
5,000 NA No compound-related effects
NA 300 Reduction in size of thyroid follicles
Reduction of relative spleen weight
NA 9 500 in females> histopathologic
changes to the liver, irritation of the
upper respiratory tract
Limited fibrosis of the liver;
ND ND a Possible myocardial and renal
effects; ; Fibrosis of responsive
stellate cells in the brain
NA NA No compound-related effects
... ... No clear-cut compound-related
NA NA effects
Increase in absolute kidney and
100 1,000 b spleen weight, decrease in
absolute and relative testis weight
fluctuations in urinalysis,
100 1,000 b hematology, and clinical chemistry
parameters
Reference
NEDO (1987)
Sayers et al.
(1944)
Cameron et
al. (1984)
Cameron et
al. (1985)
Andrews et al.
(1987)
Poon et al.
(1994)
Poon et al.
(1995)
NEDO (1987)
4-70
-------
Table 4-21 (Continued): Summary of repeat exposure and developmental studies of
methanol toxicity in experimental animals (inhalation exposure).
Species, strain,
number/sex
Rat
Sprague-Dawley
15/pregnant
females/group
Rat
Sprague-Dawley
36/pregnant
females/group
Rat
Sprague-Dawley
F-i and F2 generations
of a two-generation
study
Rat
Sprague-Dawley
Follow-up study of
brain weights in F-i
generation of
10-14/sex/group in F-i
generation
Mouse
CD-1
30-114 pregnant
females/group
Mouse
CD-1
12-17 pregnant
females/group
Rat
Long-Evans
6-7 pregnant
females/group
Rat
Long-Evans
10-12 pregnant
females/group
Monkey
M. fascicularis
12 monkeys/group
NOAEL
Dose/duration (ppm)
0, 5,000, 10,000, or
20,000 ppm, 7 hr/day
on either GD1-GD19 or '
GD7-GD15.
0,200, 1,000, or
5,000 ppm,
22.7 hr/day, on ''uuu
GD7-GD17
0, 10, 100, or
1,000 ppm, 20 hr/day;
F-i- birth to end of 100
mating (M) or weaning
(F); F2- birth to 8 wks
0, 500, 1,000, and
2,000 ppm, 20 hr/day;
GDO through F-i 8 wks
500
0, 1,000,2,000, 5,000,
7,500, 10,000, or
15,000 ppm, 7 hr/day '
onGD6-GD15.
0 and 10,000 ppm
7 hr/day, 2 consecutive
days during GD6-GD1 3 NA
or on one day during
GD5-GD9
0 or 15,000 ppm,
7hr/dayonGD7-GD19 NA
0 or 4, 500 ppm from
GD10toPND21. NA
0, 200, 600, or
1,800 ppm, 2.5 hr/day,
7 days/wk, during Nn
premating, mating and
gestation
LOAEL
(ppm)
10,000
5,000
1,000
1,000
2,000
10,000
15,000
4,500
NDC
Effect
Reduced fetal body weight,
increased incidence of visceral and
skeletal abnormalities, including
rudimentary and extra cervical ribs
Late-term resorptions, reduced
fetal viability, increased frequency
of fetal malformations, variations
and delayed ossifications.
Reduced weight of brain, pituitary,
and thymus at 8, 16 and 24 wk
postnatal in F-i and at 8 wk in F2
Reduced brain weight at 3 wk and
6 wk (males only). Reduced brain
and cerebrum weight at 8 wk
(males only)
Increased incidence of extra cervi-
cal ribs, cleft palate, exencephaly;
reduced fetal weight and pup
survival, Delayed ossification
Cleft palate, exencephaly, skeletal
malformations
Reduced pup weight
Subtle cognitive deficits
Shortened period of gestation; may
be related to exposure (no dose-
response), neurotoxicological
deficits including reduced
performance in the VDR test; may
be related to premature births.
Reference
Nelson et al.
(1985)
NEDO (1987)
Rogers et al.
(1993b)
Rogers and
Mole (1997)
Stanton et al.
(1995)
Weiss et al.
(1996)
Burbacher et
al. (2004a;
2004b; 1999a;
1999b)
"Effects in the brain and other organs were noted at exposures as low as 100 ppm (131 mg/m ), but due to substantial uncertainties
associated with these results, EPA was not able to identify a NOAEL or LOAEL from this study.
bA LOAEL of 1,000 ppm was identified by the authors of these studies, but the weight-of evidence is low (see Section 4.2.2.3).
The shortened gestation period was noted dams exposed to as low as 200 ppm (263 mg/m3) and signs of possible developmental
neurotoxicity were noted in the offspring of dams exposed to as low as 600 ppm (789 mg/m3). However, because of uncertainties
associated with these results, including the lack of a clear dose-response, EPA was not able to identify a NOAEL or LOAEL from
this study.
ND = Not determined due to study limitations such as small number of animals /time point/ exposure level
NA = Not applicable.
4-71
-------
ILow Dose T High Dose A NOAEL TLOAEL
2-9 Gestation Days
91-Day Subcronic
~ 5 10,000
II
= co
§0
1,000
100
10
1
Rat
Lifespan
Rat
1= developmental neurobehavioral deficits (Infurna and Weiss. 1986)
2 = resorbed litters, fetal death, cleft palate, exencephaly (Rogers et al.. 1993b)
3 = reduced brain weight (TRL. 1986)
4 = increased liver weight and serum activity of ALT/AP (TRL. 1986)
5 = parenchymal cell necrosis (Apaia. 1980)
6 = (cancer study): no noncancer effects were reported (Soffritti et al., 2002)
Figure 4-1 Exposure response array for noncancer effects reported in animals from repeat
exposure and developmental studies of methanol (Oral).
4-72
-------
II
§&
§§
"«
O
I
-*
I
100
1 Low Dose T High Dose A NOAEL T LOAEL
1-40 Gestation Days
Monkey
T
1
Mouse
f\
I
A
I
I - -J
2
1-80 GDH
Rat
Y
/
~^~
f
345
Rat
PND 10-50 D»y»
/~^*-
A
\
r^
/
6
7
Monkey Rat
_,_ __
j
-A-
• I
89 10 11 12
500-1 000 Days
Monkey Mouse Rat
j
-
i. j
^ L
t-
\
13 14 15
Note: Oval shapes in the array indicate principal studies used in reference value determinations.
Effects (bolded effects from principal studies were used in reference value determinations):
1= reproductive (shortened gestation) and developmental neurotoxicity (delayed VDR) (Burbacher et al.. 2004b: Burbacher et al..
1999b): N(L)OAEL not determined
2 = extra cervical ribs, cleft palate, excencephaly, reduced fetal weight & pup survival, delayed ossification (Rogers and Mole. 1997:
Rogers etal.. 1993b):
3 = reduced fetal weight, visceral and skeletal abnormalities, including rudimentary and extra cervical ribs (Nelson etal.. 1985):
4 = late-term resorptions, reduced fetal viability, fetal malformations, variations and delayed ossifications (NEDO. 1987):
5 = reduced pup weight (Stanton et al.. 1995)
6 = reduced weight of brain, pituitary, and thymus at 3, 6,8,16 and 24 wk postnatal in F1 and at 8 wk in F2 generation (NEDO.
1987)
7 = subtle cognitive deficits (Weiss et al.. 1996)
8 = clinical signs of toxicity, CNS changes in bilateral putamen, caudate nucleus, and claustrum. edema of cerebral white matter
(NEDO. 1987)
9 = no methanol-related effects (Andrews et a I.. 1987)
10 = transient reduction in plasma testosterone levels (Cameron et al.. 1985: Cameron et al.. 1984)
11 = no methanol-related effects (Andrews et al.. 1987)
12 = reduction in size of thyroid follicles (Poon etal.. 1995: Poon et al.. 1994)
13 = limited fibrosis of the liver, possible myocardial and renal effects; fibrosis of responsive stellate cells in the brain (NEDO. 1987)
14 = increased absolute kidney, spleen weight; decreased relative testes weight (NEDO. 1987): tentative LOAEL due to low weight-
of-evidence.
15 = fluctuations in urinalysis, hematology, and clinical chemistry parameters (NEDO. 1987): tentative LOAEL due to low weight-of-
evidence..
Figure 4-2 Exposure response array for noncancer effects reported in animals from repeat
exposure and developmental studies of methanol (Inhalation).
4.6.1.1. Oral
There have been very few subchronic, chronic, or in utero experimental studies of oral
methanol toxicity. In one such experiment, an EPA-sponsored 90-day gavage study in Sprague-
Dawley rats suggested a possible effect of the compound on the liver CTRL, 1986). In the
absence of gross or histopathologic evidence of toxicity, fluctuations on some clinical chemistry
markers of liver biochemistry and increases in liver weights at the highest administered dose
(2,500 mg/kg-day) justify the selection of the mid-dose level (500 mg/kg-day) as a NOAEL for
this effect under the operative experimental conditions. That the bolus effect may have been
important in the induction of those few effects that were apparent in the subchronic study is
4-73
-------
suggested by the outcome of lifetime drinking water study of methanol that was carried out in
Sprague-Dawley rats by Soffritti et al. (2002). According to the authors, no noncancer
toxicological effects of methanol were observed at drinking water concentrations of up to
20,000 ppm (v/v). Based on default assumptions on drinking water consumption and body
weight gain assumptions, the high concentration was equivalent to a dose of 1,780 mg/kg-day in
males and 2,177 mg/kg-day in females. In the stated absence of any changes to parameters
reflective of liver toxicity in the Soffritti et al. (2002) study, the slight impacts to the liver
observed in the subchronic study (TRL, 1986) at 2,500 mg/kg-day suggest the latter dose to be a
minimal LOAEL. Logically, the true but unknown threshold would at the high end of the range
from 500 (the default NOAEL) to 2,500 mg/kg-day for liver toxicity via oral gavage.
Two studies have pointed to the likelihood that oral exposure to methanol is associated
with developmental neurotoxicity or developmental deficits. When Infurna and Weiss (1986)
exposed pregnant Long-Evans rats to 2% methanol in drinking water (providing a dose of
approximately 2,500 mg/kg-day), they observed no reproductive or developmental sequelae
other than from 2 tests within a battery of fetal behavioral tests (deficits in suckling ability and
homing behavior). In the oral section of the Rogers et al. (1993b) study, such teratological effects
as cleft palate and exencephaly and skeletal malformations were observed in fetuses of pregnant
female mice exposed to daily gavage doses of 4,000 mg/kg methanol during GD6-GD15.
Likewise, an increase in totally resorbed litters and a decrease in the number of live fetuses/litter
appear likely to have been an effect of the compound. Similar skeletal malformations were
observed by Rogers and Mole (1997), Rogers et al.(1993b), and Nelson et al. (1985) following
inhalation exposure.
4.6.1.2. Inhalation
Some clinical signs, gross pathology, and histopathological effects of methanol have been
seen in experimental animals including adult nonhuman primates exposed to methanol vapor.
Results from an unpublished study (NEDO, 1987) of M. fascicularis monkeys, chronically
exposed to concentrations as low as 10 ppm for up to 29 months, resulted in histopathological
effects in the liver, kidney, brain and peripheral nervous system. These results were generally
reported as subtle or transient. However, brain effects, such as responsive stellate cells in
cerebral white matter, were observed as many as 11 months after the cessation of exposure.
Confidence in the methanol-induced findings of effects in adult nonhuman primates is limited
because this study utilized a small number (2-3) of animals/dose level/time of sacrifice and
inadequately reporting of results (e.g., limited details on materials and methods, lack of clear
documentation of a concurrent control group). Due to these concerns NOAEL and LOAEL
4-74
-------
values could not be identified and the NEDO (1987) monkey studies have limited utility in
derivation of an RfC.
A number of studies have examined the potential toxicity of methanol to the male
reproductive system (Lee etal., 1991; Cameron et al., 1985; Cameron et al., 1984). The data
from Cameron et al. (1985; 1984) showed a transient but not necessarily dose-related decrease in
serum testosterone levels of male Sprague-Dawley rats. Lee et al. (1994) reported the appearance
of testicular lesions in 18-month-old male Long-Evans rats that were exposed to methanol for
13 weeks and maintained on folate-deficient diets. Taken together, the Lee et al. (1994) and
Cameron et al. (1985; 1984) study results could indicate chemically-related strain on the rat
system as it attempts to maintain hormone homeostasis. However, the available data are
insufficient to definitively characterize methanol as a toxicant to the male reproductive system.
When Sprague-Dawley rats were exposed to methanol, 6 hours/day for 4 weeks, there
were some signs of irritation to the eyes and nose. Mild changes to the upper respiratory tract
were also described in Sprague-Dawley rats that were exposed for 4 weeks to up to 300 ppm
methanol (Poon etal., 1995). Other possible effects of methanol in rats included a reduction in
size of thyroid follicles (Poon et al., 1994), panlobular vacuolation of the liver, and a decrease in
spleen weight (Poon etal., 1995). NEDO (1987) reported dose-related increases in moderate
fatty degeneration in hepatocytes of male mice exposed via inhalation for 12 months, but this
finding was not observed in the NEDO (1987) 18-month mouse inhalation study. Nodes reported
in the liver of mice from the 18-month study may have been precancerous, but the 18-month
study duration was not of sufficient duration to make a determination.
One of the most definitive and quantifiable toxicological impacts of methanol when
administered to experimental animals via inhalation is related to the induction of developmental
abnormalities in fetuses exposed to the compound in utero. Developmental effects have been
demonstrated in a number of species, including monkeys, but particularly rats and mice. Most
developmental teratological effects appear to be more severe in the latter species. For example,
in the study of Rogers et al. (1993b) in which pregnant female CD-I mice were exposed to
methanol vapors on GD6-GD15 at a range of concentrations, reproductive and fetal effects
included an increase in the number of resorbed litters, a reduction in the number of live pups, and
increased incidence of exencephaly, cleft palate, and the number of cervical ribs. While the
biological significance of the cervical rib effect has been the subject of much debate (See
discussion of Chernoff and Rogers (2004) in Section 5), it appears to be the most sensitive
indicator of developmental toxicity from this study, with a NOAEL of 1,000 ppm (1,310 mg/m3).
In rats, however, the most sensitive developmental effect, as reported in the NEDO (1987)
two-generation inhalation studies, was a postnatal reduction in brain weight at 3, 6 and 8 weeks
postnatally, which was significantly lower than controls when pups and their dams were exposed
4-75
-------
to 1,000 ppm (1,310 mg/m3) during gestation and throughout lactation. The NOAEL reported in
this study was 500 ppm (655 mg/m3).
Rogers and Mole (1997) addressed the question of which period of gestation was most
critical for the adverse developmental effects of methanol in CD-I rats. Such malformations and
anomalies as cleft palate, exencephaly, and a range of skeletal defects, appeared to be induced
with a greater incidence when the dams were exposed on or around GD6. These findings were
taken to indicate that methanol is most toxic to embryos during gastrulation and in the early
stages of organogenesis. However, NEDO (1987) gestation-only and two-generation studies
showed that significant reductions in brain weight were observed at a lower exposure levels
when pups and their dams were exposed during lactation as well as gestation, indicating that
exposure during the later stages of organogenesis, including postnatal development, can
significantly contribute to the severity of the effects in this late-developing organ system.
In comparing the toxicity (NOAELs and LOAELs) for the onset of developmental effects
in mice and rats exposed in utero, there is suggestive evidence from the above studies that mice
may be more susceptible to methanol than rats. Supporting evidence for this proposition has
come from in vitro studies in which rat and mouse embryos were exposed to methanol in culture
(Andrews et al., 1993). Further evidence for species-by-species variations in the susceptibility of
experimental animals to methanol during organogenesis has come from experiments on monkeys
(Burbacher et al., 2004a: 2004b: 1999a: 1999b). In these studies, exposure of monkeys to
methanol during premating, mating, and throughout gestation resulted in a shorter period of
gestation in dams exposed to as low as 200 ppm (263 mg/m3). Though statistically significant,
the finding of a shortened gestation length may be of limited biological significance. Gestational
age, birth weight and infant size observations in all exposure groups were within normal ranges
for M. fascicularis monkeys, and other "signs of possible difficulty in the maintenance of
pregnancy" reported, such as vaginal bleeding, are considered normal within 1-4 days of delivery
and do not necessarily imply a risk to the fetus [as cited in NTP-CERHR (2004)1. As discussed in
Section 4.4.2, there is also evidence from this study that methanol caused neurobehavioral effects
in exposed monkey infants that may be related to the gestational exposure. However, the data are
not conclusive, and a dose-response trend is not robust. There is insufficient evidence to
determine if the primate fetus is more or less sensitive than rodents to methanol teratogenesis.
The use of a cohort design necessitated by the complexity of this study may have limited its
power to detect effects. Because of the uncertainties associated with these results, including the
lack of a clear dose-response for decreased in gestational length and neurological effects, EPA
was not able to identify a definitive NOAEL or LOAEL from this study. This study does support
the weight of evidence for developmental neurotoxicity in the hazard characterization of low-
level methanol exposure.
4-76
-------
Weiss et al. (1996) and Stanton et al. (1995) evaluated the developmental and
developmental neurotoxicological effects of methanol exposure on pregnant female Long-Evans
rats and their progeny. In the latter study, exposure of dams to 15,000 ppm (19,656 mg/m3),
7 hours/day on GD7-GD19 resulted in reduced weight gain in pups, but produced little other
evidence of adverse developmental effects. The authors subjected the pups to a number of
neurobehavioral tests that gave little if any indication of compound-related changes. This study,
while using high exposure levels, was limited in its power to detect effects due to the small
number of animals used. In the Weiss et al. (1996) study, exposure of pregnant female Long-
Evans rats to 0 or 4,500 (0 and 5,897 mg/m3) methanol from GD6 to PND21 likewise provided
fluctuating and inconsistent results in a number of neurobehavioral tests that did not necessarily
indicate any compound-related impacts. The finding of this study indicated subtle cognitive
defects not on the learning of an operant task but in the reversal learning. This study also
reported exposure-related changes in neurodevelopmental markers of NCAMs on PND4.
NCAMs are a family of glycoproteins that is needed for migration, axonal outgrowth, and
establishment of the pattern for mature neuronal function.
Taking all of these findings into consideration reinforces the conclusion that the most
appropriate endpoints for use in the derivation of an inhalation RfC for methanol are associated
with developmental neurotoxicity and developmental toxicity. Among an array of findings
indicating developmental neurotoxicity and developmental malformations and anomalies that
have been observed in the fetuses and pups of exposed dams, an increase in the incidence of
cervical ribs of gestationally exposed mice (Rogers et al., 1993b) and a decrease in the brain
weights of gestationally and lactationally exposed rats (NEDO, 1987) appear to be the most
robust and most sensitive effects.
4.7. Noncancer MOA Information
There is controversy over the possible roles of the parent compound, metabolites, reactive
oxygen species (from methanol metabolism competitively inhibiting other catalase activity) and
folate deficiency (potentially associated with methanol metabolism) in the developmental
toxicity of methanol. Experiments that have attempted to address these issues are reviewed in the
following paragraphs.
4-77
-------
4.7.1. Role of Methanol and Metabolites in the Developmental Toxicity of
Methanol
Dorman et al. (1995) conducted a series of in vitro and in vivo studies that provide
information for identifying the proximate teratogen associated with developmental toxicity in
CD-I mice. The studies used CD-I ICR BR (CD-I) mice, HPLC grade methanol, and
appropriate controls. PK and developmental toxicity parameters were measured in mice exposed
to sodium formate (750 mg/kg by gavage), a 6-hour methanol inhalation (10,000 or 15,000 ppm),
or methanol gavage (1.5 g/kg). In the in vivo inhalation study, 12-14 dams/ group were exposed
to 10,000 ppm methanol for 6 hours on GD8,47 with and without the administration of
fomepizole to inhibit the metabolism of methanol by ADH1. Dams were sacrificed on GD10, and
fetuses were examined for neural tube patency. As shown in Table 4-22, the incidence of fetuses
with open neural tubes was significantly increased in the methanol group (9.65% in treated
versus 0 in control) and numerically but not significantly increased in the group treated with
methanol and fomepizole (7.21% in treated versus 0 in controls). Rodents metabolize methanol
via both ADH1 and CAT (as discussed in Section 3.1) which, when coupled with the Dorman et
al. (1995) observation that maternal formate levels in blood and decidual swellings (swelling of
the uterine lining) did not differ in dams exposed to methanol alone or methanol and fomepizole,
suggest that the role of ADH1 relative to CAT and nonenzymatic methanol clearance is not of
great significance in adult rodents.
47 Dorman et al. (1995) state that GD8 was chosen because it encompasses the period of murine neurulation and the
time of greatest vulnerability to methanol-induced neural tube defects.
4-78
-------
Table 4-22 Developmental outcome on GD10 following a 6-hour 10,000 ppm
(13,104 mg/m3) methanol inhalation by CD-mice or formate gavage
(750 mg/kg) on GD8.
Treatment
Air
Air/fomepizole
Methanol
Methanol/fomepizole
Water
Formate
No. of litters
14
14
12
12
10
14
Open neural tubes (%)
2.29 ± 1.01
2.69± 1.19
9.65±3.13a
7.21 ±2.65
0
2.02 ± 1.08
Head length (mm)
3.15 ±0.03
3.20 ±0.05
3.05 ±0.07
3.01 ±0.05
3.01 ±0.07
2.91 ±0.08
Body length (mm)
5.89 ±0.07
5.95 ±0.09
5.69 ±0.13
5.61 ±0.11
5.64 ±0.11
5.49 ±0.12
ap < 0.05, as calculated by the authors.
Values are means ± SD
Source: Adapted with permission of John Wiley & Sons; Dorman et al. (1995).
The data in Table 4-22 suggest that the formate metabolite is not responsible for the
observed increase in open neural tubes in CD-I mice following methanol exposure. Formate
administered by gavage (750 mg/kg) did not increase this effect despite the observation that this
formate dose produced the same toxicokinetic profile as a 6-hour exposure to 10,000 ppm
methanol vapors (48.33 mg/L formate in maternal blood and 2.0 mM formate/kg in decidual
swellings). However, the data are consistent with the hypotheses that the formaldehyde
metabolite of methanol may play a role. Both CAT and ADH1 activity are immature at days past
conception (DPC)8 (Table 4-23). If fetal ADH1 is more mature than fetal CAT, it is conceivable
that the decrease in the open neural tube response observed for methanol combined with
fomepizole (Table 4-22) may be due to fomepizole having a greater effect on the metabolism of
fetal methanol to formaldehyde than is observed in adult rats. Unfortunately, the toxicity studies
were carried out during a period of development where ADH1 expression and activity are just
starting to develop (Table 4-23); therefore, it is uncertain whether any ADH1 was present in the
fetus to be inhibited by fomepizole.
4-79
-------
Table 4-23 Summary of ontogeny of relevant enzymes in CD-I mice and humans.
CD-1 Mouse
Days Past Conception (DPC)
6.5 7.5 8.5 9.5
Somites (8-12) (13-20) (21-29)
Human
Trimesters
1 2 3
CAT
mRNA
activity3
embryo 1 10 20
VYS 10 15 20
N/A N/A N/A
ADH1
mRNA - -
activity3
embryo 320 460 450
VYS 240 280 290
+ + +
ADH3
mRNA + + +
activity3
embryo 300 490 550
VYS 500 500 550
- - +
3Activity of CAT and ADH1 are expressed as nmol/minute/mg and pmol/minute/mg, respectively.
Source: Adapted with permission of Elsevier; Harris et al. (2003).
Dorman et al. (1995) provide additional support for their hypothesis that methanol's
developmental effects in CD-I mice are not caused by formate in an in vitro study involving the
incubation of GD8 whole CD-I mouse embryos with increasing concentrations of methanol or
formate. Developmental anomalies were observed on GD9, including cephalic dysraphism,
asymmetry and hypoplasia of the prosencephalon, reductions of brachial arches I and II,
scoliosis, vesicles on the walls of the mesencephalon, and hydropericardium (Table 4-24). The
concentrations of methanol used for embryo incubation (0-375 mM or 0-12,000 mg/L) were
chosen to be broadly equivalent to the peak methanol levels in plasma that have been observed
(approximately 100 mM or 3,200 mg/L) after a single 6-hour inhalation exposure to 10,000 ppm
(13,104 mg/m3). As discussed above, these exposure conditions induced an increased incidence
of open neural tubes on GD10 embryos when pregnant female CD-I mice were exposed on GD8.
(Table 4-22). Embryonic lesions such as cephalic dysraphism, prosencephalic lesions, and
brachial arch hypoplasia were observed with 250 mM (8,000 mg/L) methanol and 40 mM
(1,840 mg/L) formate. The study authors noted that a formate concentration of 40 mM
4-80
-------
(1,840 mg/L) greatly exceeds blood formate levels in mice inhaling 15,000 ppm methanol
(resulting in 0.75 mM blood formate concentration = 35 mg/L), a teratogenic dose.
Table 4-24 Dysmorphogenic effect of methanol and formate in neurulating CD-I mouse
embryos in culture (GD8).
Live embryos
Cephalic
dysraphism
Concentration No.
Treatment (mg/L) Total abnormal Severe
Vehicle
Methanol
Formate
1,984
4,000
5,984
8,000
12,000
184
368
552
920
1,840
20
13
14
13
15
12
12
13
9
16
16
3
1
5
7
7
7
2
5
5
7
14a
0
0
1
2
2
6a
0
1
0
2
10a
Prosencephalic lesions
Mode-
rate Total Hypoplasia
2
0
0
4
5
5
0
5
5
5
4
2
0
2
6
7
11a
0
6
5
7
14a
2
1
2
3
7a
9a
2
4
1
2
3
Asymmetry Total h
0
0
2
1
1
1
0
2
2
1
5a
2
1
4
4
8
10a
2
6
3
3
8
Brachial
arch
lypoplasia
0
0
1
1
6a
8a
1
0
0
1
13a
Bp < 0.05, as calculated by the authors.
Source: Adapted with permission of John Wiley & Sons; Dorman et al. (1995).
As discussed in Section 4.3.3, a series of studies by Harris et al. (2004; 2003) also
provide evidence as to the moieties that may be responsible for methanol-induced developmental
toxicity. Harris et al. (2004) have shown that among methanol and its metabolites, viability of
cultured rodent embryos is most affected by formate. In contrast, teratogenic endpoints (of
interest to this assessment) in cultured rodent embryos are more sensitive to methanol and
formaldehyde than formate. Data from these studies indicate that developmental toxicity may be
more related to formaldehyde than methanol, as formaldehyde-induced teratogenicity occurs at
several orders of magnitude lower than methanol (Table 4-12) (Hansen et al., 2005; Harris et al.,
2004). It should also be noted that CAT, ADH1, and ADH3 activities are present in both the rat
embryo and VYS at stages as early as 6-12 somites (Harris et al., 2003): thus, it is presumable
that in these ex vivo studies methanol is metabolized to formaldehyde and formaldehyde is
subsequently metabolized to S-formylglutathione.
Studies involving GSH depletion have been offered as support for the hypothesis that
formaldehyde is a key proximal teratogen, and for the role of ROS (see Section 4.7.3). Inhibition
of GSH synthesis with butathione sulfoximine (BSO) has little effect on developmental toxicity
endpoints, yet treatment with BSO and methanol or formaldehyde increases developmental
4-81
-------
toxicity (Harris et al., 2004). Among the enzymes involved in methanol metabolism, only ADH3-
mediated metabolism of formaldehyde is GSH dependent. While "depletion of GSH, as the
major cellular antioxidant, will also increase the accumulation of reactive oxygen species
(ROS)." This hypothesis that ADH3-mediated metabolism of formaldehyde is important for the
amelioration of methanol's developmental toxicity is also supported by the diminished ADH3
activity in the mouse versus rat embryos, which is consistent with the greater sensitivity of the
mouse to methanol developmental toxicity (Harris et al., 2003) (Section 4.3.3).
Without positive identification of the actual moiety responsible for methanol-induced
teratogenicity, MOA remains unclear. If the moiety is methanol, then it is possible that
generation of NADH during methanol oxidation creates an imbalance in other enzymatic
reactions. Studies have shown that ethanol intake leads to a >100-fold increase in cellular
NADH, presumably due to ADH1-mediated reduction of the cofactor NAD+ to NADH
(Cronholm, 1987; Smith and Newman, 1959). This is of potential importance because, for
example, ethanol intake has been shown to increase the in vivo and in vitro enzymatic reduction
of other endogenous compounds (e.g., serotonin) in humans (Svensson et al., 1999; Davis et al.,
1967). In rodents, CAT-mediated methanol metabolism may obviate this effect; in humans,
however, methanol is primarily metabolized by ADH1.
If the teratogenic moiety of methanol is formaldehyde, then reactivity with protein
sulfhydryls and nonprotein sulfhydryls (e.g., GSH) or DNA protein cross-links may be involved.
Metabolic roles ascribed to ADH3, particularly regulation of S-nitrosothiol biology (Foster and
Stamler, 2004), could also be involved in the MOA. Recently, Staab et al. (2008) have shown
that formaldehyde alters other ADH3-mediated reactions through cofactor recycling and that
formaldehyde alters levels of cellular S-nitrosothiol, which plays a key role in cellular signaling
and many cellular functions and pathways (Hess et al., 2005).
Studies such as those by Harris et al. (2004; 2003) and Dorman et al. (1995) suggest that
formate is not the metabolite responsible for methanol's teratogenic effects. The former
researchers suggest that formaldehyde is the proximate teratogen, and provide evidence in
support of that hypothesis. However, questions remain. As has been discussed, the capacity for
the metabolism of methanol to formaldehyde is likely lower in the fetus and neonate versus
adults (Section 3.3). Further, researchers in this area have not yet reported using a sufficient array
of enzyme inhibitors to conclusively identify formaldehyde as the proximate teratogen. Studies
involving other inhibitors or toxicity studies carried out in genetically engineered mice, while not
devoid of confounders, might further inform regarding the methanol MOA for developmental
toxicity.
Even if formaldehyde is ultimately identified as the proximate teratogen, methanol would
likely play a prominent role, at least in terms of transport to the target tissue. The high reactivity
4-82
-------
of formaldehyde would limit its unbound and unaltered transport as free formaldehyde from
maternal to fetal blood (Thrasher and Kilburn, 2001). However, methanol can be metabolized to
formaldehye in situ by multiple organ systems (Jelski et al., 2006; Motavkin et al., 1988; Biihler
et al., 1983) and dose-dependent increases of formaldehyde DNA adducts derived from
exogenous methanol exposure have been observed in multiple tissues such as liver, lung, spleen,
thymus, bone marrow, kidney, and WBC (exogenous adduct levels were less than 10% of
endogenous adduct levels for most organ systems; embryonic tissue was not examined) of rats
(Luetal., 2012).
4.7.2. Role of Folate Deficiency in the Developmental Toxicity of Methanol
As discussed in Sections 3.1 and 4.1, humans and other primates are susceptible to the
effects of methanol exposure associated with formate accumulation because they have lower
levels of hepatic tetrahydrofolate-dependent enzymes that help in formate oxidation.
Tetrahydrofolate-dependent enzymes and critical pathways that depend on folate, such as purine
and pyrimidine synthesis, may also play a role in the developmental toxicity of methanol. Studies
of rats and mice fed folate-deficient diets have identified adverse effects on reproductive
performance, implantation, fetal growth and developmental defects, and the inhibition of folate
cellular transport has been associated with several developmental abnormalities, ranging from
neural tube defects to neurocristopathies such as cleft-lip and cleft-palate, cardiac septal defects,
and eye defects (Antony, 2007). Folate deficiency has been shown to exacerbate some aspects of
the developmental toxicity of methanol in mice (see discussion of (Fu et al., 1996), and
(Sakanashi et al., 1996), in Section 4.3.1) and rats (see discussion of (Aziz etal., 2002), in
Section 4.4.1).
The studies in mice focused on the influence of FAD on the reproductive and skeletal
malformation effects of methanol. Sakanashi et al. (1996) showed that dams exposed to
5 g/kg-day methanol on GD6-GD15 experienced a threefold increase in the percentage of litters
affected by cleft palate and a 10-fold increase in the percentage of litters affected by exencephaly
when fed a FAD (resulting in a 50% decrease in liver folate) versus a FAS diet. They speculated
that the increased methanol effect from FAD diet could have been due to an increase in tissue
formate or a critical reduction in conceptus folate concentration immediately following the
methanol exposure. The latter appears more likely, given the high levels of formate needed to
cause embryotoxicity (Section 4.3.3) and the decrease in conceptus folate that is observed within
2 hours of GD8 methanol exposure (Dorman et al., 1995). Fu et al. (1996) confirmed the findings
of Sakanashi et al. (1996) and also determined that the maternal FAD diet had a much greater
impact on fetal liver folate than maternal liver folate levels.
4-83
-------
The rat study of Aziz et al. (2002) focused on the influence of FAD on the developmental
neurotoxicity of methanol. Experiments by Aziz et al. (2002) involving Wistar rat dams and pups
exposed to methanol during lactation provide evidence that methanol exposure during this
postnatal period affects the developing brain. These effects (increased spontaneous locomotor
activity, decreased conditioned avoidance response, disturbances in dopaminergic and
cholinergic receptors and increased expression of GAP-43 in the hippocampal region) were more
pronounced in FAD as compared to FAS rats. This suggests that folic acid may play a role in
methanol-induced neurotoxicity. These results do not implicate any particular proximate
teratogen, as folate deficiency can increase levels of both methanol, formaldehyde and formate
(Medinsky et al., 1997). Further, folic acid is used in a number of critical pathways such as
purine and pyrimidine synthesis. Thus, alterations in available folic acid, particularly to the
conceptus, could have significant impacts on the developing fetus apart from the influence it is
presumed to have on formate removal.
Another problem with the hypothesized folate deficiency MOA is that an explanation for
this greater mouse sensitivity is not readily apparent. Mouse livers actually have considerably
higher hepatic tetrahydrofolate and total folate than rat or monkey liver (Johlin et al., 1987).
4.7.3. Methanol-induced Formation of Free Radicals, Lipid Peroxidation, and
Protein Modifications
Oxidative stress in mother and offspring has been suggested to be part of the teratogenic
mechanism of a related alcohol, ethanol. Certain reproductive and developmental effects (e.g.,
resorptions and malformation rates) observed in Sprague-Dawley rats following ethanol
exposure were reported to be ameliorated by antioxidant (Vitamin E) treatment (Wentzel et al.,
2006; Wentzel and Eriksson, 2006). A number of studies have examined markers of oxidative
stress associated with methanol exposure.
McCallum et al. (2011 a: 201 Ib) treated adult male CD-I mice, DNA repair deficient
oxoguanine glycosylase (Oggl) knockout mice, NZW rabbits and cynomolgus monkeys
(Macacafascicularis) with a single i.p. injection of 2 g/kg methanol and measured 8-hydroxy-2'-
deoxyguanosine (8-oxodG), as an indicator of tissue oxidative DNA damage, 6 hours post-
injection in the lung, liver, kidney, bone marrow and spleen. They also examined these organs for
8-oxodG in adult male CD-I mice injected daily for 15 days with 2 g/kg methanol. They reported
no evidence of methanol-dependent increases in 8-oxodG in any of the species and organ
systems tested.
Miller and Wells (2011) exposed mouse embryos expressing human catalase (hCat) or
their wild-type controls, and acatalasemic (aCat)-expressing mouse embryos or their wild-type
4-84
-------
controls for 24 hours to 4 mg/mL methanol or vehicle on gestational day 9. They observed higher
methanol-induced teratogenicity in catalase deficient embryos, and interpreted this as an
indication that ROS is involved in the embryopathic mechanism of methanol. However,
contradictory results were obtained from subsequent in-vivo studies performed by the same
laboratory using the same mouse strains. Siu et al. (2013) treated pregnant hCat and aCat mice
and their wild-type (WT) controls with 4 g/kg of methanol (i.p.) or saline on GD 8. Although
catalase activities were confirmed to be substantially increased in the hCat maternal livers and
embryos, increases in fetal ophthalmic abnormalities and cleft palate, similar to those reported
for C57BL/6J mice by Rogers et al. (2004), were observed in methanol-exposed hCat mice and
their WT controls but not in methanol-exposed aCat mice or their WT controls. The authors
indicated that the relative resistance of aCat mice to the embryotoxic effects of methanol could
not be explained by differences in methanol metabolism because similar peak and AUC levels of
methanol and its formic acid metabolite were observed for male aCat and hCat mice and their
WT controls, but this would need to be verified with pharmacokinetic data for the female mice
and their affected embryos. Siu et al. (2013) suggest that the apparent discrepancy between their
in-vivo results and the Miller and Wells (2011) in-vitro results could be due to yet to be
determined maternal factors associated with metabolism and membrane transport and/or a
requirement for high catalase activity in the hCat mice, but acknowledge that it may also be an
indication that ROS does not play an important embryopathic role in vivo.
Skrzydlewska et al. (2005) provided inferential evidence for the effects of methanol on
free radical formation, lipid peroxidation, and protein modifications, by studying the protective
effects of N-acetyl cysteine and the Vitamin E derivative, U83836E, in the liver of male Wistar
rats exposed to the compound via gavage. Forty-two rats/group received a single oral gavage
dose of either saline or 50% methanol. This provided a dose of approximately 6,000 mg/kg, as
calculated by the authors. Other groups of rats received the same concentration of methanol, but
were also injected intraperitoneally with either N-acetylcysteine or U-83836E. N-acetylcysteine
and U-83836E controls were also included in the study design. Animals in each group were
sacrificed after 6, 14, and 24 hours or after 2, 5, or 7 days. Livers were rapidly excised for
electron spin resonance (ESR) analysis, and 10,000 x g supernatants were used to measure GSH,
malondialdehyde, a range of protein parameters, including free amino and sulfhydryl groups,
protein carbonyls, tryptophan, tyrosine, and bityrosine, and the activity of cathepsin B. They
reported (1) an ESR signal (thought to be indicative of free radical formation) at g = 2.003 in
livers harvested 6 and 12 hours after methanol exposure, (2) a significant decrease in GSH levels
that was most evident in rats sacrificed 12 and 24 hours after exposure; (3) increased
concentrations in the lipid peroxidation product, malondialdehyde (by a maximum of 44% in the
livers of animals sacrificed 2 days after exposure); (4) increased specific concentrations of
4-85
-------
protein carbonyl groups and bityrosine; but (5) reductions in the specific level of tryptophan.
Given the ability of N-acetylcysteine and U83836E to oppose these changes, at least in part, the
authors speculated that methanol-induced free radical formation and lipid peroxidation are
involved. However, it is unclear whether or not the metabolites of methanol, formaldehyde,
and/or formate, were involved in any of these changes.
Rajamani et al. (Rajamani et al., 2006) examined several oxidative stress parameters in
male Wistar rats following methotrexate-induced folate deficiency. Compared to controls, the
levels of free radical scavengers SOD, CAT, GSH peroxidase, oxidized GSH, protein carbonyls,
and lipid peroxidation were elevated in several regions of the brain, with greater increases
observed in the MTX-methanol-treated animals than in the MTX-alone group. The level of GSH
and protein thiols was decreased in all regions of the brain, with a greater decrease observed in
the MTX-methanol-treated animals than MTX-treated animals.
Dudka (2006) measured the total antioxidant status (TAS) in the brain of male Wistar rats
exposed to a single oral gavage dose of methanol at 3 g/kg. The animals were kept in a nitrous
oxide atmosphere (N2O/O2) throughout the experiment to reduce intrinsic folate levels, and
various levels of ethanol and/or fomepizole (as ADH antidotes) were administered i.p. after
4 hours. Animals were sacrificed after 16 hours, the brains homogenized, and the TAS
determined spectrophotometrically. As illustrated graphically by the author, methanol
administration reduced TAS in brain irrespective of the presence of ADH antidotes. The author
speculated that, while most of the methanol is metabolized in the liver, some may also reach the
brain. Metabolism to formate might then alter the NADH/NAD+ ratio resulting in an increase in
xanthine oxidase activity and the formation of the superoxide anion.
Parthasarathy et al. (2006b) investigated the extent of methanol-induced oxidative stress
in rat lymphoid organs. Six male Wistar rats/group received 2,370 mg/kg methanol (mixed 1:1
with saline) injected i.p. for 1, 15 or 30 days. A control group received a daily i.p. injection of
saline for 30 days. At term, lymphoid organs such as the spleen, thymus, lymph nodes, and bone
marrow were excised, perfused with saline, then homogenized to obtain supernatants in which
such indices of lipid peroxidation as malondialdehyde, and the activities of CAT, SOD, and GSH
peroxidase were measured. Parthasarathy et al. (2006b) also measured the concentrations of GSH
and ascorbic acid (nonenzymatic antioxidants) and the serum concentrations of a number of
indicators of liver and kidney function, such as ALT, AST, blood urea nitrogen (BUN), and
creatinine.
Table 4-25 shows time-dependent changes in serum liver and kidney function indicators,
which resulted from methanol administration. Treatment with methanol for increasing durations
resulted in increased serum ALT and AST activities and the concentrations of BUN and
creatinine.
4-86
-------
Table 4-25 Time-dependent effects of methanol administration on serum liver and kidney
function, serum ALT, AST, BUN, and creatinine in control and experimental
groups of male Wistar rats.
Parameters
ALT (umoles pyruvate
liberated /minute/mg protein)
AST (umoles pyruvate
liberated /minute/mg protein)
Urea (mg/L)
Creatinine (mg/L)
Control
29.0 ±2.5
5.8 ±0.4
301 ± 36
4.6 ±0.3
Methanol administration
Single dose
31.4 ±3.3
6.4 ±0.3
332 ± 29
4.8 ±0.3
(2,370 mg/kg)
15 days
53.1 ±2.3a
9.0 ± 1.2a
436 ± 35a
5.6±0.2a
30 days
60.4 ± 2.8a
13.7 ± 1.2a
513±32a
7.0±0.4a
ap < 0.05 versus controls.
Values are means ± SD of 6 animals.
Source: Adapted with permission of Japan Society for Occupational Health; Parthasarathy et al. (2006b).
Table 4-26 Effect of methanol administration on male Wistar rats on malondialdehyde
concentration in the lymphoid organs of experimental and control groups and
the effect of methanol on antioxidants in spleen.
Methanol administration (2,370 mg/kg)
Parameters
Malondialdehyde in lymphoid
Spleen
Thymus
Lymph nodes
Bone marrow
Control
organs
2.62 ±0.19
3.58 ±0.35
3.15 ±0.25
3.14 ±0.33
Single dose
4.14±0.25a
5.76 ± 0.36a
5.08 ± 0.24a
4.47±0.18a
15 days
7.22±0.31a
9.23 ± 0.57a
8.77 ± 0.57a
7.20 ± 0.42a
30 days
9.72 ± 0.52a
11.6±0.33a
9.17±0.67a
9.75 ± 0.56a
Antioxidant levels in spleen
SOD (units/mg protein)
CAT (umoles H2O2
consumed/min-mg protein
GPx (ug GSH consumed/min-
mg protein)
GSH (ug/mg protein)
Vitamin C (ug/mg protein)
2.40 ±0.16
35.8 ±2.77
11.2±0.60
2.11 ±0.11
0.45 ±0.04
4.06±0.19a
52.5±3.86a
20.0 ± 1.0a
3.75±0.15a
0.73 ± 0.05a
1.76±0.09a
19.1 ± 1.55a
7.07 ± 0.83a
1.66±0.09a
0.34±0.18a
1.00±0.07a
10.8± 1.10a
5.18±0.45a
0.89 ± 0.04a
0.11 ±0.03a
ap < 0.05, versus controls.
Values are means ± SD of six animals.
Source: Adapted with permission of Japan Society for Occupational Health; Parthasarathy et al. (2006b) (adapted).
Table 4-26 gives the concentration of malondialdehyde in the lymphoid organs of control
and experimental groups, and, as an example of all tissue sites examined, the levels of enzymatic
4-87
-------
and nonenzymatic antioxidants in spleen. The results show that malondialdehyde concentrations
were time-dependently increased at each tissue site and that, in spleen as an example of all the
lymphoid tissues examined, increasing methanol administration resulted in lower levels of all
antioxidants examined compared to controls. Parthasarathy et al. (2006b) concluded that
exposure to methanol may cause oxidative stress by altering the oxidant/antioxidant balance in
lymphoid organs in the rat.
4.7.4. Exogenous Formate Dehydrogenase as a Means of Detoxifying the
Formic Acid that Results from Methanol Exposure
In companion reports, Muthuvel et al. (2006a; 2006b) used 6 male Wistar rats/group to
test the ability of exogenously-administered formate dehydrogenase (FD) to reduce the serum
levels of formate that were formed when 3 g/kg methanol was administered i.p. to rats in saline.
In the first experiment, purified FD (from Candida boitinii) was administered by i.v. conjugated
to the N-hydroxysuccinimidyl ester of monomethoxy polyethylene glycol propionic acid
(PEG-FD) (Muthuvel et al., 2006b). In the second, rats were administered FD-loaded
erythrocytes (Muthuvel et al., 2006a). In the former case, some groups of rats were made folate
deficient by means of a folate-depleted diet; in the latter, folate deficiency was brought about by
i.p. administration of methotrexate. In some groups, the rats received an infusion of an equimolar
mixture of carbonate and bicarbonate (each at 0.33 mol/L) to correct the formate-induced
acidosis. As illustrated by the authors, methanol-exposed rats receiving a folate-deficient diet
showed significantly higher levels of serum formate than those receiving a folate-sufficient diet.
However, administration of native or PEG-FD reduced serum formate in methanol-receiving
folate-deficient rats to levels seen in animals receiving methanol and the folate-sufficient diet.
In the second report, Muthuvel et al. (2006a) carried out some preliminary experiments to
show that hematological parameters of normal, reconstituted but unloaded, and reconstituted and
FD-loaded erythrocytes, were similar. In addition, they showed that formate levels of serum were
reduced in vitro in the presence of FD-loaded erythrocytes. Expressing blood formate
concentration in mmol/L at the 1-hour time point after carbonate/bicarbonate and enzyme-loaded
erythrocyte infusion via the tail vein, the concentration was reduced from 10.63 ±1.3
(mean ± SD) in methanol and methotrexate-receiving controls to 5.83 ± 0.97 (n = 6). This
difference was statistically significant at the/? < 0.05 level. However, FD-loaded erythrocytes
were less efficient at removing formate in the absence of carbonate/bicarbonate. Effective
elimination of formate appears to require an optimum pH for the FD activity in the enzyme-
loaded erythrocytes.
4-88
-------
4.7.5. Summary and Conclusions Regarding MOA for Developmental Toxicity
Data from experiments carried out by Dorman et al. (1995) indicate that formate is not
the probable proximate teratogen in pregnant CD-I mice exposed to high concentrations of
methanol vapor. This conclusion is based on the observation that there appeared to be little, if
any, accumulation of formate in the blood of methanol-exposed mice, and exencephaly did not
occur until formate levels were grossly elevated. In addition, treatment of pregnant mice with a
high oral dose of formate did not induce neural tube closure defects at media concentrations
comparable to those observed in uterine decidual swelling after maternal exposure to methanol.
Lastly, methanol- but not formate- induced neural tube closure defects in mouse embryos in vitro
at media concentrations comparable to the levels of methanol detected in blood after a
teratogenic exposure.
Harris and colleagues (Hansen et al., 2005; Harris et al., 2004; Harris et al., 2003) carried
out a series of physiological and biochemical experiments on mouse and rat embryos exposed to
methanol, formaldehyde and formate, concluding that the etiologically important substance for
embryo dysmorphogenesis and embryolethality was likely to be formaldehyde rather than the
parent compound or formate. Specific activities for enzymes involved in methanol metabolism
were determined in rat and mouse embryos during the organogenesis period of 8-25 somites
(Harris et al., 2003). The experiment was based on the concept that differences in the metabolism
of methanol to formaldehyde and formic acid by the enzymes ADH1, ADH3, and CAT may
contribute to hypothesized differences in species sensitivity that were apparent in toxicological
studies. A key finding was that the activity of ADH3 (converting formaldehyde to formate) was
lower in mouse VYS than that of rats throughout organogenesis, consistent with the greater
sensitivity of the mouse to the developmental effects of methanol exposure. Another study
(Harris et al., 2004) which showed that the inhibition of GSH synthesis increases the
developmental toxicity of methanol also lends support to this hypothesis because ADH3-
mediated metabolism of formaldehyde is the only enzyme involved in methanol metabolism that
is GSH-dependent. These findings provide inferential evidence for the proposition that
formaldehyde may be the ultimate teratogen through diminished ADH3 activity. This concept is
further supported by the demonstration that the LOAELs for the embryotoxic effects of
formaldehyde in rat and mouse embryos were much lower than those for formate and methanol
(Hansen et al., 2005). The findings from both sets of experiments (Hansen et al., 2005; Harris et
al., 2004; Harris et al., 2003) suggest that the lower capacity of mouse embryos to transform
formaldehyde to formate (by ADH3) could explain the increased susceptibility of mouse versus
rat embryos to the toxic effects of methanol.
4-89
-------
Recent studies suggest that mouse embryo tissue may have a high sensitivity to oxidative
damage relative to other species following methanol exposure (Miller and Wells, 2011; Sweeting
et al., 2011). Sweeting et al. (2011) postulated that one possible explanation for this sensitivity
may be a strong reliance of mice on catalase over ADH to metabolize embryonic methanol. A
low ADH activity in mouse embryo relative to rats [(Harris et al., 2003), Section 4.3.3],
combined with a preference of catalase to metabolize methanol over hydrogen peroxide
(Sweeting et al., 2011), could lead to a reduction in catalase activity and a higher level of ROS in
mouse versus rat embryos, partially explaining the higher sensitivity of mice to the embryotoxic
effects of methanol. If an appreciable portion of methanol's teratogenicity in sensitive mouse
strains can be explained by this mode of action, and if this mode of action is not applicable to
human fetuses, then sensitive mouse strains may not adequately reflect human risk. However, the
evidence for this mode of action remains limited. Further, as discussed in Section 3.3, there is
reason to believe that human infants can metabolize methanol via a mechanism other than ADH,
and that this alternative mechanism could involve catalase (Iran et al., 2007).
While studies such as those by Harris et al. (2004; 2003) and Dorman and colleagues
(Dorman and Welsch, 1996; Dorman et al., 1995) strongly suggest that formate is not the
metabolite responsible for methanol's teratogenic effects, there are still questions regarding the
relative involvement of parent methanol, formaldehyde and ROS. However, both the proposed
formaldehyde and ROS MO As require methanol to be present at the target site. Methanol can be
metabolized to formaldehye in situ by multiple organ systems and the high reactivity of
formaldehyde would limit its unbound and unaltered transport as free formaldehyde (see
discussion in Section 4.7.1), and the ROS MOA would require the presence of methanol to alter
embryonic catalase activity.
4.8. Evaluation of Carcinogenicity
Carcinogenicity was not evaluated in this assessment.
4.9. Susceptible Populations and Life Stages
4.9.1. Possible Childhood Susceptibility
Studies in animals have identified the fetus as being more sensitive than adults to the
toxic effects of methanol; the greatest susceptibility occurs during gastrulation and early
organogenesis (NTP-CERHR, 2004). Table 4-23 summarizes some of the data regarding the
4-90
-------
relative ontogeny of CAT, ADH1, and ADH3 in humans and mice. Human fetuses have limited
ability to metabolize methanol as ADH1 activity in 2-month-old and 4-5 month-old fetuses is
3-4% and 10% of adult activity, respectively (Pikkarainen and Raiha, 1967). ADH1 activity in
9-22 week old fetal livers was found to be 30% of adult activity (Smith etal., 1971). Likewise,
ADH1 activity is -20-50% of adult activity during infancy (Smith et al., 1971; Pikkarainen and
Raiha, 1967). Activity continues to increase until reaching adult levels at 5 years of age
(Pikkarainen and Raiha, 1967). However, no difference between blood methanol levels in 1-year-
old infants and adults was observed following ingesting the same doses of aspartame, which
releases 10% methanol by weight during metabolism (Stegink et al., 1983). Given that the
exposure was aspartame as opposed to methanol, it is difficult to draw any conclusions from this
study vis-a-vis ontogeny data and potential influences of age differences in aspartame
disposition. With regard to inhalation exposure, increased breathing rates relative to adults may
result in higher blood methanol levels in children compared to adults (NTP-CERHR, 2004). It is
also possible that metabolic variations resulting in increased methanol blood levels in pregnant
women could increase the fetus' risk from exposure to methanol. In all, unresolved issues
regarding the identification of the toxic moiety increase the uncertainty with regards to the extent
and pathologic basis for early life susceptibility to methanol exposure.
The prevalence of folic acid deficiency has decreased since the United States and Canada
introduced a mandatory folic acid food fortification program in November 1998. However, folate
deficiency is still a concern among pregnant and lactating women, and factors such as smoking, a
poor quality diet, alcohol intake, and folic antagonist medications can enhance deficiency (NTP-
CERHR, 2004). Folate deficiency could affect a pregnant woman's ability to clear formate,
which has also been demonstrated to produce developmental toxicity in rodent in in vitro studies
at high-doses (Dorman et al., 1995). It is not known if folate-deficient humans have higher levels
of blood formate than individuals with adequate folate levels. A limited study in folate-deficient
monkeys demonstrated no formate accumulation following an endotracheal exposure of
anesthetized monkeys to 900 ppm methanol for 2 hours (Dorman et al., 1994). The situation is
obscured by noting that folic acid deficiency during pregnancy by itself is thought to contribute
to the development of severe congenital malformations (Pitkin, 2007).
4.9.2. Possible Gender Differences
There is limited information on potential differences in susceptibility to the toxic effects
of methanol according to gender. One study (n=12) reported a higher background blood
methanol level in human females versus males (Batterman and Franzblau, 1997), but a larger
study (n=35) did not observe gender differences (Sarkola and Eriksson, 2001). In rodents, fetuses
4-91
-------
exposed in utero were found to be the most sensitive subpopulation. One study suggested a
possible increased sensitivity of male versus female rat fetuses and pups. When rats were
exposed to methanol pre- and postnatally, 6- and 8-week-old male progeny had significantly
lower brain weights at 1,000 ppm, compared to those in females that demonstrated the same
effect only at 2,000 ppm (NEDO, 1987). In general, there is little evidence for substantial
disparity in the level or degree of toxic response to methanol in male versus female experimental
animals or humans. However, it is possible that the compound-related deficits in fetal brain
weight that were evident in the pups of FI generation Sprague-Dawley rats exposed to methanol
in the NEDO (1987) study may reflect a threshold neurotoxicological response to methanol. It is
currently unknown whether higher levels of exposure would result in brain sequelae comparable
to those observed in acutely exposed humans.
4.9.3. Genetic Susceptibility
Polymorphisms in enzymes involved in methanol metabolism may affect the sensitivity
of some individuals to methanol. For example, as discussed in Section 3, data summarized in
reviews by Agarwal (2001), Burnell et al. (1989), Bosron and Li (1986), and Pietruszko (1980)
discuss genetic polymorphisms for ADH. Class IADH, the primary ADH in human liver, is a
dimer composed of randomly associated polypeptide units encoded by three genetic loci
(ADH1 A, ADH1B, and ADH1C). Polymorphisms are observed at the ADH1B (ADH1B*2,
ADH1B*3) and ADH1C (ADH1C*2) loci. The ADH1B*2 phenotype is estimated to occur in
-15% of Caucasians of European descent, 85% of Asians, and less that 5% of African
Americans. ADH1C*1 is also highly prevalent in Asians, but has only been examined in a few
studies of Chinese and Korean samples (Eng et al., 2007). Fifteen percent of African Americans
have the ADH1B*3 phenotype, while it is found in less than 5% of Caucasian Europeans and
Asians. The only reported polymorphisms in ADH3 occur in the promoter region, one of which
reduces the transcriptional activity in vitro nearly twofold (Hedberg et al., 2001). While
polymorphisms in ADH3 are described in more than one report (Cichoz-Lach et al., 2007;
Hedberg et al., 2001), the functional consequence(s) for these polymorphisms remains unclear.
Although racial and ethnical differences in the frequency of the occurrence of ADH
alleles in different populations have been reported, ADH enzyme kinetics (Vmax and Km) have not
been reported for methanol. There is an abundance of information pertaining to the kinetic
characteristics of the ADH dimers to metabolize ethanol in vitro. Methanol blood concentrations
of 2.62 ± 1.33 mg/L (Table 3-1) in 18 Korean males (Woo et al., 2005) were considerably higher
than the sample U.S. background distribution estimated of 1.36 mg/L and 0.77 mg/L estimated in
Section 5.3.6. However, the functional and biological significance is not well understood due to
4-92
-------
the lack of data documenting metabolism and disposition of methanol or ethanol in individuals of
known genotype. Thus, while potentially significant, the contribution of ethnic and genetic
polymorphisms of ADH to the interindividual variability in methanol disposition and metabolism
cannot be reliably quantified at this time.
4-93
-------
5.DOSE-RESPONSE ASSESSMENTS
5.1. Inhalation Reference Concentration (RfC)48
In general, the RfC is an estimate (with uncertainty spanning perhaps an order of
magnitude) of a continuous inhalation exposure to the human population (including sensitive
subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime.
It is derived from a POD, generally an estimated 95 percent lower confidence limit on the BMD
(i.e., BMDL), with uncertainty factors applied to reflect limitations of the data used. The
inhalation RfC considers toxic effects for both the respiratory system (portal-of-entry) and
systems peripheral to the respiratory system (extra-respiratory or systemic effects). It is generally
expressed in mg/m3.
This assessment uses BMD modeling to identify the POD.49 The suitability of these
methods to derive a POD is dependent on the nature of the toxicity database for a specific
chemical. Details of the BMD analyses are found in Appendix D. The use of the BMD approach
for identifying the POD is preferred over the NOAEL/LOAEL approach because the BMD
approach includes consideration of the shape of the dose-response curve, is less dependent on
experimental dose selection, and estimates uncertainty pertaining to the modeled dose response.
Other limitations and uncertainties associated with the methanol database that influence
derivation of the RfC, such as uncertainties associated with human variability, animal-to-human
differences, and limitations in the database, are addressed through the use of rat and human
PBPK models and uncertainty factors.
5.1.1. Choice of Principal Study and Critical Effect(s)
5.1.1.1. Key Inhalation Studies
While a substantial body of information exists on the toxicological effects in humans
exposed to high concentrations of methanol for short durations, none of these studies are suitable
for quantification of subchronic, chronic, or in utero effects of methanol exposure. Table 4-21 in
the previous section summarizes the available experimental animal inhalation studies of
48 The RfC discussion precedes the RfD discussion in this assessment because the inhalation database ultimately
serves as the basis for the RfD. The RfD development would be difficult to follow without prior discussion of the
inhalation database and PK models used for the route-to-route extrapolation.
49 Use of BMD modeling involves fitting mathematical models to dose-response data and using the results to
estimate a POD that is associated with a selected benchmark response (BMR), such as a percentage increase in the
incidence of a particular lesion or a percentage decrease in body weight gain (see Section 5.1.2.2).
5-1
-------
methanol. Several of these studies, including monkey chronic (NEDO, 1987) and developmental
(Burbacher et al., 2004a: 2004b: 1999a: 1999b) studies, male rat reproductive studies (Lee et al.,
1991; Cameron et al., 1985; Cameron et al., 1984), and 4-week rat studies (Poonetal., 1994), are
lacking in key attributes (e.g., existence of a dose response, documented controls, and adequate
duration of exposure) necessary for use in the derivation of a chronic RfC. As discussed in
Section 4.6, the most appropriate endpoints for use in the derivation of an inhalation RfC for
methanol are associated with developmental toxicity. From the studies listed in Table 4-21, the
reproductive/developmental studies that were adequately documented and of the appropriate size
and design for use in the derivation of an RfC were selected and are described in Table 5-1.
5.1.1.2. Selection of Critical Effect(s)
5.1.1.2.1. Developmental Skeletal Abnormalities
Skeletal defects following methanol exposure have been observed in developmental
studies of rats (Weiss etal., 1996; NEDO, 1987; Nelson et al., 1985) and mice (Rogers and Mole,
1997: Bolonetal., 1993: Rogers etal., 1993b). The findings of Bolon et al. (1993) and Rogers
and Mole (1997) indicate that methanol is toxic to mouse embryos in the early stages of
organogenesis, on or around GD7. In the study by Rogers et al. (1993b), in which pregnant
female CD-I mice were exposed to methanol vapors (at concentrations of 1,000, 2,000, and
5,000 ppm) on GD6-GD15, reproductive and fetal effects included an increase in the number of
resorbed litters, a reduction in the number of live pups, and increased incidences of exencephaly,
cleft palate, and extra cervical ribs. The authors identified a NOAEL of 1,000 ppm (1,310
mg/m3) and a LOAEL of 2,000 ppm (2,620 mg/m3) based on the increased incidence of extra
cervical ribs, with 49.6% per litter in the 2,000 ppm dose group versus 28.0% per litter in the
control group. An increased incidence of extra cervical ribs was also observed in the rat
organogenesis study by NEDO (1987), with the 5,000 ppm dose group exhibiting an incidence of
65.2% per litter versus 0% in the control group, indicating that this endpoint is consistent across
species.
The biological significance of the cervical rib endpoint has been the subject of much
debate (Chernoff and Rogers, 2004). Previous studies have classified this endpoint as either a
malformation (birth defect of major importance) or a variation (morphological alteration of
minor significance). Evidence exists that incidence of supernumerary ribs (including cervical
ribs) is not just the addition of extraneous, single ribs, but rather is related to a general alteration
in the development and architecture of the axial skeleton as a whole. For example, in CD-I mice
exposed during gestation to various types of stress, food and water deprivation, and the herbicide
5-2
-------
dinoseb, supernumerary ribs were consistently associated with increases in the length of the 13th
rib (Branch et al., 1996). This relationship was present in all fetal ages examined in the study.
The authors concluded that these findings are consistent with supernumerary ribs being one
manifestation of a basic alteration in the differentiation of the thoraco-lumbar border of the axial
skeleton. The biological significance of this endpoint is further strengthened by the association of
supernumerary ribs with adverse health effects in humans. The most common effect associated
with the presence of extra cervical ribs is thoracic outlet disease (Nguyen et al., 1997; Fernandez
Noda et al., 1996; Henderson, 1914). Thoracic outlet disease is characterized by numbness
and/or pain in the shoulder, arm, or hands. Vascular effects associated with this syndrome include
cerebral and distal embolism (Beam et al., 1993; Connell et al., 1980; Short, 1975), while
neurological symptoms include extreme pain, migraine, and symptoms similar to Parkinson's
disease (Evans, 1999; Saxton etal., 1999; Fernandez Noda et al., 1996). Furthermore,
Schumacher et al. (1992) observed 242 rib anomalies in 218 children with tumors (21.8%) and
11 (5.5%) in children without malignancy, a statistically significant (p < 0.001) difference that
suggests an association between the presence of extra cervical ribs and childhood cancers. In
conclusion, the mouse cervical rib endpoint is biologically significant and potentially relevant to
humans, and thus appropriate for use in the derivation of a human health toxicity value (RfC or
RfD).
5.1.1.2.2. Developmental Neurotoxicity
NEDO (1987) reported reduced brain, pituitary, and thymus weights in FI and F2
generation Sprague-Dawley rats exposed to 1,000 ppm methanol. In a follow-up study of the FI
generation brain weight effects, NEDO (1987) reported decreased brain, cerebellum, and
cerebrum weights in FI males exposed to 1,000 ppm methanol from GDO through the FI
generation.50 The methanol exposure levels used in these studies are difficult to interpret because
dams were exposed prior to gestation, as well as during gestation and lactation, while pups were
exposed during gestation (in utero) and lactation. However, the results from NEDO (1987)
clearly show that postnatal methanol exposure increases the magnitude of brain weight
reduction. In another experiment by NEDO (1987) referenced in the previous section, rats
exposed to methanol only during organogenesis (GD7-GD17) exhibited decreases in brain
weights in offspring at 8 weeks of age that were less severe than in rat pups in the studies in
which methanol exposure was continued postnatally. This finding is not unexpected, given that
the brain undergoes tremendous growth beginning early in gestation and continuing into the
50 For the interpretation of the dose-response data, EPA did not rely on the statistics reported by NEDO (1987)
which were based on inappropriate t-test methods but, instead, relied on the results of the benchmark dose analyses
described in Appendix D.
5-3
-------
postnatal period. Rats are considered altricial (i.e., born at a relatively underdeveloped stage),
and thus many of their neurogenic events occur postnatally (Clancy et al., 2007). Brain effects
from postnatal exposure are also relevant to humans given that, in humans, gross measures of
brain growth increase for at least 2-3 years after birth, with the growth rate peaking
approximately 4 months after birth (Rice and Barone, 2000).
Change in brain weight is considered to be a biologically significant effect (U.S. EPA,
1998a). This holds true regardless of whether changes in body weight occur simultaneously
because brain weight is generally conserved even during malnutrition or weight loss, unlike
many other organs or tissues (U.S. EPA, 1998a). Thus, change in absolute brain weight is an
appropriate measure of effects on this critical organ. Decreases in brain weight have been
associated with simultaneous deficits in neurobehavioral and cognitive parameters in animals
exposed during gestation to various solvents, including toluene and ethanol (Gibson et al., 2000;
Coleman et al.. 1999: Hassetal., 1995). NEDO (1987) reported that brain, cerebellum, and
cerebrum weights decreased in a dose-dependent manner in male rats exposed to methanol
throughout gestation and the FI generation. While brain weight reduction has been observed in
adult rats exposed to methanol (TRL, 1986), it has not been observed in other developmental
bioassays of methanol. This lack of consistency across developmental studies may be due to the
fact that brain weight is not an endpoint that has been extensively measured in other
developmental studies of methanol [e.g., Rogers et al. (1993b)1.
Developmental neurobehavioral effects associated with methanol inhalation exposure
have also been investigated in monkeys. Burbacher et al. (2004a; 2004b: 1999a: 1999b) exposed
M fascicularis monkeys to 0, 200, 600, or 1,800 ppm (0, 262, 786, and 2,359 mg/m3) methanol,
2.5 hours/day, 7 days/week during premating/mating and throughout gestation (approximately
168 days). There appeared to be neurotoxicological deficits in methanol-exposed offspring. VDR
was significantly reduced in the 600 ppm (786 mg/m3) methanol group for males and in the
1,800 ppm (2,359 mg/m3) methanol group for both sexes. However, a dose-response trend for
this endpoint was only exhibited for females. In fact, the VDR response in females is the only
effect reported in the Burbacher et al. (2004a; 2004b: 1999a: 1999b) studies for which a
significant dose-response trend is evident. As discussed in Section 4.4.2, confidence in these
results may have been increased by statistical analyses that adjusted for multiple comparisons
(NTP-CERHR, 2004). However, the dose-response trend for VDR in females remained
significant both with (p = 0.009) and without (p = 0.0265) an adjustment for the shortened
gestational periods. In addition, VDR is a measure of functional deficits in sensorimotor
development that is consistent with other early developmental CNS effects (i.e., brain weight
changes discussed above) that have been observed in rats exposed to methanol.
5-4
-------
Another test performed by Burbacher et al. (2004a; 2004b: 1999a: 1999bX the Pagan test
of infant intelligence, indicated small, but non-significant, deficits of performance (decreased
time spent looking at novel faces versus familiar faces) in treated monkeys. Although not
statistically significant nor quantifiable, the results of this test should be considered, in
conjunction with the VDR test results and brain weight changes noted in the NEDO (1987) rat
study, as a possible indication of CNS effects. As discussed in Section 4.6.1.2, the monkey data
are not conclusive, and there is insufficient evidence to determine if the primate fetus is more or
less sensitive than rodents to methanol-induced teratogenesis. Taken together, however, the
results of the NEDO (1987) rat study and the Burbacher et al. (2004a; 2004b: 1999a: 1999b)
monkey studies suggest that prenatal exposure to methanol can result in adverse effects on
developmental neurology and function, which can be exacerbated by continued postnatal
exposure to methanol.
5.1.1.2.3. Reproductive Effects
In the Burbacher et al. (2004a; 2004b: 1999a: 1999b) studies, exposure of monkeys to
methanol levels ranging from 200 ppm (263 mg/m3) to 1,800 ppm (2,359 mg/m3) during
premating, mating, and throughout gestation resulted in no changes in reproductive parameters
other than a shorter period of gestation in all exposure groups that did not appear to be dose
related. As discussed in Section 4.6.1.2, although statistically significant relative to controls, the
biological significance of this shortened gestation finding is uncertain given the absence of a
dose-response relationship. Other developmental parameters, such as fetal crown-rump length
and head circumference, were unaffected by methanol exposure.
A number of studies described in Section 4.3.2 and summarized in Section 4.6.1.2 have
examined the potential toxicity of methanol to the male reproductive system (Lee et al., 1991;
Cameron et al., 1985; Cameron et al., 1984). Some of the observed effects, including a transient
decrease in testosterone levels, could be the result of chemically related strain on the rat
hormonal system. However, the data are insufficient to definitively characterize methanol as a
toxicant to the male reproductive system.
5.1.1.2.4. Selected Critical Effects
The studies considered for use in the derivation of an RfC for methanol are summarized
in Table 5-1. As discussed in Sections 5.1.3.1 and 5.3.1, there is uncertainty associated with the
selection of a critical effect from the methanol database for use in the derivation of an RfC.
Although monkeys may represent the more relevant species, the available monkey studies are not
5-5
-------
adequate for dose-response analysis. Taking into account the advantages and limitations of the
studies available for quantification purposes and the relative sensitivities for the effects observed,
two developmental effects were chosen as candidate critical effects for the purpose of dose-
response assessment, cervical rib anomalies in fetal CD-I mice (Rogers et al., 1993b) and
decreased brain weight in male Sprague-Dawley rats exposed throughout gestation and lactation
(NEDO, 1987). These endpoints can be reliably quantified and represent adverse effects in two
separate sensitive organ systems at key periods of development. RfC derivations based on these
two endpoints using different dose-response options are described in Appendix D and
summarized below.
5-6
-------
Table 5-1 Summary of studies considered most appropriate for use in derivation of an
RfC.
Reference
NEDO (1987)
Teratology
study
NEDO (1987)
Two-
generation
study
NEDO (1987)
Follow-up
study, F-i
generation
Rogers et al.
(1993b)
Burbacher et
al. (2004a;
2004b;
1999a;
1999b)
Number/
Species (strain) dose group
Rat 1 0-1 2/sex/ group
Sprague-Dawley
Rat Not specified - F-i
Sprague-Dawley anc' ^2
generation
10-147 sex/
group- F-i
generation
Mouse 30-114 pregnant
CD-1 dams/
group
Monkey 12 pregnant
M. fascicularis monkeys/group
Exposure
Duration
GD7-GD17
20 hr/day, F-r
Birth to end of
mating (M) or
weaning (F);
F2-birth to 8 wk
20 hr/day, GDO
through F-i
generation
GD6-GD15
2.5 hr/day,
7 days/wk,
during
premating,
mating and
gestation
NOAEL
Critical Effect (ppm)
Prenatal: increase in late- 1,000
term resorptions,
decrease in live fetuses,
reduced fetal weight, and
increased frequency of
litters with fetal
malformations,
variations, and delayed
ossifications
Postnatal: Reduced
brain, pituitary, thyroid,
thymus, and testis
weights at 8 wk
Reduced weight of brain, 1 00
pituitary, and thymus at
8, 16, and 24 wk
postnatal in F-i and at 8
wk in F2
Reduced brain weight at 500
3 wk and 6 wk (males
only). Reduced brain and
cerebrum weight at 8 wk
(males only)
Increased incidence of 1,000
extra cervical ribs, cleft
palate, exencephaly;
reduced fetal weight and
pup survival, delayed
ossification
Shortened period of
gestation; may be related
to exposure (no dose
response),
neurotoxicological.
deficits including reduced
performance in the VDR
test
LOAEL
(ppm)
5,000
1,000
1,000
2,000
_a
"Gestational exposure resulted in a shorter period of gestation in dams exposed to as low as 200 ppm (263 mg/m3). However,
because of uncertainties associated with these results, including the lack of a clear dose-response, EPA was not able to identify a
definitive NOAEL or LOAEL from this study.
5.1.2. Methods of Analysis for Identifying the POD—Application of PBPK and
BMD Models
Potential PODs for use in deriving the RfC, as described in Appendix D, have been
identified via the use of PBPK models, summarized in Section 3.4 and further described in
Appendix B. The administered doses used in the experimental animal studies were converted to
5-7
-------
an internal dose metric that was deemed most appropriate for the endpoint being considered. The
PBPK models are capable of estimating several internal dose metrics for methanol, including the
following:
• Cmax - The peak concentration of methanol in the blood during the exposure period;
• AUC - Area under the curve, which represents the cumulative product of
concentration and time for methanol in the blood; and
• Total metabolism - The production of metabolites of methanol, namely formaldehyde
and formate.
Because uncertainty exists surrounding identification of the proximate teratogen of
importance (methanol, formaldehyde, formate or ROS), the dose metric chosen for derivation of
an RfC was based on blood methanol levels, either Cmax or AUC, rather than total metabolism.
As discussed in Section 4.7, this decision was primarily based on a determination that (1) the
toxic moiety for developmental effects from methanol exposure is not likely to be the formate
metabolite and (2) methanol is an adequate dose metric, even if formaldehyde or ROS are
determined to have a significant role in the teratogenicity of methanol. The former determination
has been endorsed by other organizations (NTP-CERHR, 2004) and is supported by evidence
that formate blood levels do not correlate well with the developmental toxicity observed
following methanol exposure. The latter determination is based on evidence that (1) methanol
can be metabolized to formaldehye in situ by multiple organ systems, (2) the high reactivity of
formaldehyde would limit its unbound and unaltered transport as free formaldehyde and (3) the
hypothesized ROS MO A would require the presence of methanol to alter embryonic catalase
activity (see further discussion in Sections 4.7.1, 4.7.3 and 4.7.5).
Employing BMD modeling, a BMDL was then estimated using the selected internal dose
metric. Finally, after application of UFs (see Section 5.1.3.2) to this BMDL, the adjusted BMDL
was converted to a HEC via the use of a PBPK model parameterized for humans. The next
section describes the BMD modeling approach that was used to estimate the BMDL.
5.1.2.1. Application of the BMD/BMDL Approach
Several developments over the past few years impact the derivation of the RfC: (1) EPA
has developed BMD assessment methods (U.S. EPA, 2012a, 1995) and supporting software
(U.S. EPA, 2011 a) to improve upon the previous NOAEL/LOAEL approach; (2) MOA studies
have been carried out that can give more insight into methanol toxicity; and (3) EPA has refined
PBPK models for methanol on the basis of the work of Ward et al. (1997) (see Appendix B for a
description of the EPA models). The EPA PBPK models provide estimates of HECs from test
5-8
-------
animal exposures that are supported by pharmacokinetic information available for rodents,
monkeys and humans. The following sections describe how the BMD/BMDL approach, along
with the EPA PBPK models, are used to obtain PODs for use in the derivation of an RfC and
RfD for methanol consistent with current BMD technical guidance (U.S. EPA, 2012a).
The BMD approach attempts to fit models to the dose-response data for a given endpoint.
It has the advantage over the NOAEL/LOAEL approach of taking more of the dose-response
data into account when determining the POD, as well as estimating the dose at which an effect
may have a specific probability of occurring. The BMD approach also accounts, in part, for the
quality of the study by estimating a BMDL, the 95% lower confidence limit on the BMD. Larger
studies (i.e., those with more test subjects) and studies with a low background response (i.e., with
more test subjects for which a relationship between dose and response can be evaluated)
generally yield narrower confidence intervals (BMDL estimates closer to their corresponding
BMD estimates) than smaller studies and studies with a high background response. For this
reason and because the BMDL approach will take into account, in part, a study's power, dose
spacing, and the steepness of the dose-response curve, it is generally preferred over the
NOAEL/LOAEL approach.
Even though the BMD approach is preferred over the NOAEL/LOAEL approach,
uncertainties are still associated with its use. As indicated above, the BMD approach fits several
models to the dose-response data to determine which model exhibits the best fit.51 In the absence
of an established MOA or a biological basis for why one model should be preferred, model
selection is based on which exhibits the best fit to the experimental data. Model fit is evaluated
through use of goodness-of-fit diagnostics (i.e., overall %2, AIC, and $ residuals for individual
dose groups), as well as visual inspection, consistent with EPA guidance (U.S. EPA, 2012a).52
When performing a BMD analysis, it is important to choose a reliably measured or
estimated dose metric that has a close relationship to the health effect under consideration. For
the BMD analysis of the mouse cervical rib endpoint, peak (Cmax) internal methanol blood
concentrations were used as the dose metric [from the dams in each dose group at GD6, reported
by Rogers et al.(1993b)]. For the BMD analysis of the rat brain weight endpoint following
gestational exposure only (GD7-GD17), PBPK model estimates of Cmax methanol in blood for
51USEPA's BMDS 2.2 (U.S. EPA. 201 la) was used for this assessment as it provides data management tools for
running multiple models on the same dose-response data set. At this time, BMDS offers over 30 different models
that are appropriate for the analysis of dichotomous, continuous, nested dichotomous and time-dependent response
data. Results from all models include a reiteration of the model formula and model run options chosen by the user,
goodness-of-fit information, the BMD, and an estimate of the 95 percent lower-bound on the BMD (i.e., the
BMDL).
52Akaike's Information Criterion (AIC) (Akaike. 1973) is used for model selection and is defined as -2L + 2P where
L is the log-likelihood at the maximum likelihood estimates for the parameters and P is the number of model degrees
of freedom.
5-9
-------
the dams in each dose group were used as the dose metric. Cmax of methanol in blood (in mg/L)
was chosen as the appropriate internal dose metric for these two gestational exposure studies
because the magnitude of exposure is believed to be more important for these effects under these
study conditions than the duration of exposure, particularly for the cervical rib endpoint, which
has been shown to have a short gestational window of susceptibility (Rogers and Mole, 1997;
Bolonetal.. 1993)
For the BMD analysis of the rat brain weight endpoint following both gestational and
lactational exposure, PBPK model estimates of AUC methanol in blood for the dams in each
dose group were used as the dose metric. The results of NEDO (1987), described in Section 4.4.2
and shown in Table 4-13, indicate that there is no obvious cumulative effect from ongoing
methanol exposure on brain-weight decrements in rats exposed postnatally. That is, the dose
response in terms of percent of control is about the same at 3 weeks postnatal as at 8 weeks
postnatal in rats exposed throughout gestation and the FI generation. However, there does appear
to be a greater brain-weight effect in rats exposed postnatally versus rats exposed only during
organogenesis (GD7-GD17), suggesting a cumulative effect of methanol exposure. Specifically,
in male rats exposed during organogenesis only, there is no statistically significant decrease in
brain weight at 8 weeks after birth at the 1,000 ppm exposure level. Conversely, in male rats
exposed to the same concentration of methanol throughout gestation and the FI generation, there
was an approximately 5% decrease in brain weights (statistically significant at the/? < 0.01
level). Also, male rats exposed to 5,000 ppm methanol only during organogenesis experienced a
smaller decrease in brain weight at 8 weeks postnatal than male rats exposed to 2,000 ppm
methanol throughout gestation and the 8 week postnatal period (10% versus 13%). Further, brain
weight reductions have been observed in adult rats that were exposed to methanol for 90 days
with exposure beginning no earlier than 30 days of age (TRL, 1986). These results demonstrate
that brain weight is susceptible to both the magnitude and duration of exposure, and thus suggest
that a dose metric that incorporates a time component would be most appropriate. For these
reasons, and because AUC is more typically used in internal-dose-based assessments as well as
better reflecting total exposure within a given day, daily AUC (measured for 22 hours
exposure/day) was chosen as the most appropriate dose metric for modeling the effects of
methanol exposure on brain weights in rats exposed throughout gestation and continuing into the
FI generation.
5.1.2.2. BMD Approach Applied to Brain Weight Data in Rats
The NEDO (1987) teratology study reported decreases in brain weights in fetal rats and
rat pups exposed during gestation only (GD7-GD17) and the developmental study performed as
5-10
-------
a follow-up to the NEDO (1987) two generation rat study reported decreases in brain weights in
rat pups exposed during gestation and up to 8 weeks postnatally (see Section 4.4.2). Because of
the biological significance of decreases in absolute brain weight in the developing rat and
because this endpoint was not evaluated in other peer-reviewed studies, BMD analysis was
performed using dose-response data from both of these studies (see Appendix D for details).
Decreased brain weights observed in male rats at 8 weeks of age after gestation-only exposure
were not used for RfC derivation because the decreases seen were lower in magnitude at the
same dose level (1,000 ppm) compared to the decreases observed in rats exposed during both
gestation and postnatally. For the purpose of deriving an RfC for methanol, decreases in rat brain
weight at 6 weeks of age in the more sensitive gender, males, exposed throughout gestation and
continuing into the FI generation (both through lactation and inhalation routes) were used.
Decreases in brain weight observed at 6 weeks, rather than those seen at 3 and 8 weeks, were
chosen as the basis for the RfC derivation because they resulted in lower estimated BMDs and
BMDLs.
The first step in the BMD analysis is to convert the administered inhalation doses, given
as concentrations in air in ppm, to an internal dose metric using the EPA PBPK model (see
Appendix B for a detailed description of the PBPK models developed for methanol). Application
of the EPA methanol PBPK model is complicated by the exposure regimen used in the NEDO
(1987) developmental studies. The neonatal rats in the developmental study performed as a
supplement to the NEDO (1987) two-generation rat study were exposed to methanol in utero
before parturition (as well as via lactation and inhalation after parturition). Because data on
lactational transfer and early postnatal inhalation exposures to methanol are limited, the PBPK
model developed by EPA only estimates internal dose metrics for methanol exposure in non-
pregnant adult rats. Experimental data indicate that blood methanol kinetics following inhalation
exposures to non-pregnant (NP) mice and pregnant mice on GD6-GD10 are similar (Dorman et
al.. 1995: Perkins et al.. 1995b: Rogers etal.. 1993a: Rogers etal.. 1993b). In addition,
experimental data indicate that the maternal blood:fetal partition coefficient for mice and rats is
approximately 1 up to GD 20 (see Sections 3.2 and 3.4.1.2). Assuming that these findings also
apply for rats later in pregnancy, the data indicate that PBPK estimates of PK and blood dose
metrics for NP rats are better predictors of fetal exposure during gestation than would be
obtained from default extrapolations from external exposure concentrations. However, as is
discussed in Section 5.1.3.2.2, the additional routes of exposure to the pups in this study (via
both lactation and inhalation) present uncertainties in that the average blood levels in pups are
likely to be greater than those of their dams. The assumption made in this assessment is that, if
such differences exist between human mothers and their offspring, they are not significantly
greater than that which has been postulated for rats. Assuming this is true, the PBPK model-
5-11
-------
estimated adult blood methanol level is considered to be an appropriate dose metric for the
purpose of this analysis and the estimation of a human equivalent concentration (HEC).
The predicted AUC blood methanol values (adjusted for background) for rat dams
exposed to methanol in air at 0, 500, 1,000, or 2,000 ppm are presented in the second column of
Table 5-2. These AUC values are then used as the internal dose metric for the BMD analysis of
the response data (i.e., mean male brain weight of pups at 6 weeks of age) shown in the third
column of Table 5-2.53 The full details of this BMD analysis are provided in Appendix D.
Further details concerning the PBPK modeling are presented in Section 3.4 and Appendix B.
Table 5-2 The EPA PBPK model estimates of methanol blood levels (AUC) adjusted for
background (control) levels in rat dams following methanol inhalation
exposures and reported mean brain weights of 6-week-old male pups.
Exposure level
(ppm)
0
Blood methanol AUCa - control AUCa
(mg-hr/L)a in rat dams
0
Mean male rat (Fi generation)
brain weight at 6 weeksb
1.78 ±0.07
N
12
500 547 1.74 ±0.09 12
1,000 2,310 1.69±0.06C 11
2,000 17,500 1.52±0.07d 14
aAUCs were obtained by simulating 22 hr/day exposures for 5 days and calculated for the last 24 hours of that period; AUCs above
background were obtained by subtracting the estimated AUC for controls of 72 mg-hr/L.
bExposed throughout gestation and FI generation. Values are means ± SD
°p < 0.01
dp < 0.001, as calculated by the authors.
Data from NEDO (1987).
The EPA's BMD technical guidance (U.S. EPA, 2012a) suggests that, in the absence of
knowledge as to what level of response to consider adverse, a change in the mean equal to one
standard deviation (SD) from the control mean can be used as a benchmark response (BMR) for
continuous endpoints. However, it has been suggested that other BMRs, such as 5% change
relative to the estimated control mean, also be considered when performing BMD analyses on
developmental endpoints, such as fetal weight change (Kavlock et al., 1995). Therefore, both a
one SD change from the control mean and a 5% change relative to the estimated control mean
were considered as BMRs in the current analysis (see Appendix D for RfC derivations using
alternative BMRs).
As described in Appendix D and consistent with the EPA's BMD Technical Guidance
(U.S. EPA. 2012aX the BMDL from the Hill model was selected as the most appropriate POD
from which to derive an RfC derivation because this model yields the lowest BMDL from among
53
'All BMD assessments in this review were performed using BMDS version 2.2 (U.S. EPA. 201 la).
5-12
-------
a broad range of BMDLs and also provides a superior fit in the low dose region of the dose-
response nearest the BMD. The fit of the Hill model to the dose-response data for mean brain
weight in male rats is presented in Figure 5-1, with the response plotted against the chosen
internal dose metric of methanol AUC in blood of rats, adjusted for background. The BMDLiso
was estimated to be 858 mg-hr/L and the BMDLos was estimated to be 1,183 mg-hr/L.
Hill Model with 0.95 Confidence Level
8.
a:
c
(C
1.85
1.8
1.75
1.7
1.65
1.6
1.55
1.5
1.45
Hill
\
: BMDL
BMD
2000 4000
6000
8000 10000
dose
12000 14000 16000 18000
10:25 10/052011
Data points obtained from NEDO (1987).
Figure 5-1 Fit of the Hill model to decreased mean brain weight in male rats at 6 weeks
age using estimated AUC of methanol in blood (adjusted for background) as
the dose metric. The BMD is estimated based on a BMR of one SD change from
the control mean.
5.1.2.3. BMD Approach Applied to Cervical Rib Data in Mice
For the purpose of deriving an RfC for methanol based on a developmental endpoint in
mice, extra cervical rib incidence data were evaluated from Rogers et al. (1993b). Although the
teratology portion of the NEDO study (1987) also reported fetal malformations, including
increases in extra cervical rib incidence, in Sprague-Dawley rats, the Rogers et al. (1993b) study
was chosen for dose-response modeling because these effects were seen at lower doses, the study
5-13
-------
was peer-reviewed and published in the open literature, and data on individual animals were
available, yielding a more statistically robust analysis utilizing nested models in BMDS 2.2 (U.S.
EPA. 2011 a).
For cervical rib anomalies, Cmax of methanol in blood (in mg/L) was chosen as the
appropriate internal dose metric because studies that have identified a small gestational window
of susceptibility (Rogers and Mole, 1997; Bolonetal., 1993) suggest that the magnitude of
exposure is more important than the duration of exposure. Because the critical window for
methanol induction of cervical rib malformations in CD-I mice is thought to be between GD6
and GD7 (Rogers and Mole, 1997; Rogers etal., 1993a), the measured Cmax plasma methanol
levels at GD6 from the Rogers et al. (1993b) study are used after adjusting for background (i.e.,
1.6 mg/L).54 Cmax values for methanol in the blood of mice at GD6 from Rogers et al. (1993 a) are
summarized in Table 5-3. These Cmax values are then used as the internal dose metric for the
BMD analysis of the litter-specific incidence of extra cervical ribs. The average incidence of
extra cervical ribs/litter (expressed as %) reported by Rogers et al. (1993b) is shown in Table 5-3,
but litter-specific response data from this study of 170 litters obtained from John Rogers (via
personal communication) was used for the nested BMD analysis described in Appendix D. Due
to high mortality, the high-dose (15,000 ppm) group consisting of 5 litters was excluded from the
analysis. The individual animal response data for the four dose groups shown in Table 5-3 are
displayed in the Appendix D BMD model output file.
Table 5-3 Maximum methanol blood levels (Cmax) adjusted for background, in mice
following inhalation exposures to methanol along with the corresponding
incidence of extra cervical ribs observed.
Exposure (ppm)
0
1,000
2,000
5,000
Blood methanol Cmax - control Cmax
(mg/L)a in mouse dams
0
61.4
485
2,120
Mean Incidence of Extra Cervical
Ribs/Litter (%)
28
33.6
49.6
74.4
aCmax was adjusted for background by subtracting the Cmax for controls reported by Rogers et al. (|993b) of 1.6 mg/L.
Data from Rogers et al. (1993b)
54 Given that methanol inhalation dosimetry appears to be not significantly affected by the stage of pregnancy, data
from the later gestation days could be viewed simply as additional measurements in female GDI mice. Therefore the
BMD modeling results of using weighted concentration averages for all three gestation days measured were
compared with EPA's primary approach (using only the GD6 data). The results are not substantially different, and
the model fits were not as good as the model fits to the data using the GD6 blood levels. Thus, EPA has decided that
the use of the GD6 data as the dose metric is appropriate for this analysis.
5-14
-------
Both 10% and 5% extra risk BMRs were considered for this endpoint. A 10% extra risk
BMR is adequate for most traditional bioassays using 50 animals per dose group. A smaller BMR
of 5% extra risk is sometimes justified for developmental studies such as Rogers et al. (Rogers et
al., 1993b) depending on the size of the study and the severity of the effects observed. As
described in Appendix D, the best model fit to these data (from visual inspection and comparison
of AIC values) was obtained using the NLogistic model. The fit of the NLogistic model to the
dose-response data for increased incidence of extra cervical ribs in fetal mice is presented in
Figure 5-2. From this model, the BMDLos was estimated to be 43.10 mg/L and the BMDLio was
estimated to be 90.9 mg/L, expressed in terms of the Cmax above background for methanol in
blood.
Nested Logistic Model with 0.95 Confidence Level
0.8
0.7
T3
O>
"O
•S
0.6
0.5
0.4
0.3
0.2
^MDLBMP
2000
dose
10:56 12/162011
Data points obtained from Rogers et al. (1993b).
Figure 5-2 Fit of the nested logistic (NLogistic) model- to the incidence of extra cervical
rib in fetal mice versus Cmax adjusted for background of methanol in blood
from a GD6-GD15 inhalation study in mice. The BMD is estimated based on a
BMR of 0.05 extra risk.
5-15
-------
5.1.3. RfC Derivation - Including Application of Uncertainty Factors
5.1.3.1. Derivation of Candidate RfCs
Four potential PODs based on two developmental endpoints, cervical rib abnormalities in
mice and decreased brain weight in rats, each with two possible BMRs were considered for the
derivation of an RfC for methanol (see Appendix D for BMD modeling details). These PODs,
along with applied UFs (see Section 5.1.3.2 for details) and the estimated candidate RfCs
(obtained from PBPK models described in Appendix B) are presented in Table 5-4. This
information is presented so that comparisons can be made across the different endpoints (i.e.,
cervical rib abnormalities and decreased brain weight) and BMRs employed. Each approach for
RfC derivation has its strengths and limitations, but when considered together, this comparison
allows for a more informed determination of the RfC for methanol.
Table 5-4 Summary of PODs for critical endpoints, application of UFs and conversion to
candidate RfCs using PBPK modeling.
BMDL = PODinternal
RfCinternal = PODinternal/UFSa
RfC (mg/m3)b
Rogers et al. (1993b)
mouse cervical rib Cmax
10% BMR
90.9 mg/L
0.909 mg/L
41.8
5% BMR
43.1 mg/L
0.43 mg/L
20.0
NEDO (1987)
rat brain weight ADC
5% BMR 1 SD BMR
1 , 1 83 mg-hr/L 858 mg-hr/L
1 1 .85 mg-hr/L 8.58 mg-hr/L
24.5 17.8
aUFA=3; UFD = 3; UFH = 10; UFS = 1; UFL = 1; product of all UFs= 100; see Section 5.1.3.2 below for details.
bEach candidate RfC is the inhalation exposure concentration predicted to yield a blood concentration equal to its corresponding
RfCintemai, using the human PBPK model with an background blood concentration of 2.5 mg/L, which corresponds to the estimated
maximum background exposure rate for a diet including fruits and vegetables of 1,600 mg/day (COT. 2011) in a 70-kg person (see
discussion in Section 5.3.6); the final RfC is rounded to one significant figure.
As described in Section 5.1.3.2 and shown in Table 5-4, the internal BMDL (PODinternai)
values were divided by a total UF of 100 (UFH of 10, UFA of 3 and a UFD of 3) to yield an
RfCintemai, which was converted to a candidate RfC using the human PBPK model described in
Appendix B.55 Candidate RfCs estimated from the Rogers et al. (1993b) study based on extra
cervical rib incidence in mice employing Cmaxas the dose metric were 41.8 and 20.0 mg/m3 using
BMRs of 10% and 5%, respectively. Candidate RfCs estimated from the NEDO (1987) study
55 An algebraic equation provided near the end of Appendix B approximates the PBPK model predicted relationship
between methanol AUC and C^ blood levels above background and the HEC in ppm.
5-16
-------
based on decreases in brain weight at 6 weeks of age in male rats exposed during gestation and
throughout the FI generation employing AUC as the dose metric were 24.5 and 17.8 mg/m3 using
BMRs of 5% change relative to control mean and one SD from the control mean, respectively.
Because a one SD decrease in brain weight in male rats at 6 weeks (postnatal) resulted in the
lowest of the candidate RfC estimates and, therefore, the most likely to be protective against
other effects of methanol exposure, it was chosen as the critical endpoint for use in the RfC
derivation.
RfC = 858 mg-hr/L H- 100 = 8.58 mg-hr/L ^>PBPK^> 2X101 mg/m3
(rounded to 1 significant figure)
5.1.3.2. Application of UFs
UFs are applied to PODs to account for recognized uncertainties in extrapolation from
experimental conditions to the assumed human scenario (i.e., chronic exposure over a lifetime).
According to EPA guidance (U.S. EPA, 2002, 1994b), UFs used in deriving reference values are
generally applied to FEC or FED estimates. However, as described in Appendix B (Section
B.2.7, Table B-6), the human PBPK model developed for methanol is considered uncertain above
inhalation concentrations of 500 ppm (655 mg/m3) or oral ingestions of 50 mg/kg-day, since the
blood levels predicted rise above those for which there are model calibration data. The FEC
values (1,042 to 1,604 mg/m3) and FED values (133 to 220 mg/kg-day) predicted by the human
PBPK model for BMDLs from the candidate principal studies are well above these exposure
levels. Consequently, the standard EPA practice of applying a human PBPK model to derive
FEC or FED values prior to dividing by UFs (U.S. EPA. 2002. 1994b) would add considerable
model uncertainty. In order to avoid the uncertainty associated with applying the model to
exposure levels that are above the levels for which the model was calibrated and to account for
possible non-linearities in the external versus internal dose relationships at high doses, EPA has
applied the UFs to the internal BMDL (PODinternai) prior to FEC (and FED) derivation to obtain
an RfCintemai (and RfDintemai). This approach results in more scientifically reliable model
predictions by lowering the BMDLs to within the more linear, calibrated range of the human
PBPK model.
5.1.3.2.1. Interindividual variation UFH
A factor of 10 was applied to account for variation in sensitivity within the human
population (UFH). The UFH of 10 is commonly considered to be appropriate in the absence of
convincing data to the contrary. The data from which to determine the potential extent of
5-17
-------
variation in how humans respond to chronic exposure to methanol are limited, given the complex
nature of the developmental endpoint employed and uncertainties surrounding the importance of
metabolism to the observed teratogenic effects. Susceptibility to methanol is likely to involve
intrinsic and extrinsic factors. Some factors may include alteration of the body burden of
methanol or its metabolites, sensitization of an individual to methanol effects, or augmentation of
underlying conditions or changes in processes that share common features with methanol effects.
Additionally, inherent differences in an individual's genetic make-up, diet, gender, age, or
disease state may affect the pharmacokinetics and pharmacodynamics of methanol, influencing
susceptibility intrinsically (see Sections 3.3 and 4.9). Co-exposure to a pollutant that alters
metabolism or other clearance processes, or that adds to background levels of metabolites may
also affect the pharmacokinetics and pharmacodynamics of methanol, influencing susceptibility
extrinsically. The determination of the UF for human variation is supported by several types of
information, including information concerning background levels of methanol in humans,
variation in pharmacokinetics revealed through human studies and from PBPK modeling,
variation of methanol metabolism in human tissues, and information on physiologic factors
(including gender and age), or acquired factors (including diet and environment) that may affect
methanol exposure and toxicity.
Sensitivity analyses of the human PBPK models were performed (see Appendix B), and
the results suggest that parameter variability is not likely to result in methanol blood level
estimates that vary more than 3-fold, the toxicokinetic portion of the 10-fold UFH. However, one
needs to also consider the variation in background levels of methanol (Table 3-1), because that
can be a factor governing the impact of an exogenous methanol exposure. From the data in Table
3-1, it can be seen that the reported background levels of methanol in blood have ranged
considerably, from 0.25 to 5.2 mg/L. Overall, the extent of human interindividual variation in
(endogenous and exogenous) methanol toxicokinetics and toxicodynamics would be very
difficult to quantify given the significant uncertainties that exist regarding background levels and
methanol's mode of action.
The candidate effects for RfC derivation have been observed in a potentially susceptible
and sensitive fetal/neonatal subpopulation. However, there is also variability across fetuses and
neonates that need to be taken into account. Children vary in their ability to metabolize and
eliminate methanol and in their sensitivity to methanol's teratogenic effects. There is information
on PK and pharmacodynamic factors suggesting that children can have differential susceptibility
to methanol toxicity (see Section 4.9.1). Thus, there is uncertainty in children's responses to
methanol that should be taken into consideration for derivation of the UF for human variation
that is not available from either measured human data or PBPK modeling analyses. The enzyme
primarily responsible for metabolism of methanol in humans, ADH, has been reported to be
5-18
-------
reduced in activity in newborns. Differences in pharmacokinetics include potentially greater
pollutant intake due to greater ventilation rates, activity, and greater intake of liquids in children.
In terms of differences in susceptibility to methanol due to pharmacodynamic considerations, the
substantial anatomical, physiologic, and biochemical changes that occur during infancy,
childhood, and puberty suggest that there are developmental periods in which the endocrine,
reproductive, immune, audiovisual, nervous, and other organ systems may be especially
sensitive.
There are limited data from short-term exposure studies in humans and animal
experiments that suggest differential susceptibility to methanol on the basis of gender. Gender
can provide not only different potential targets for methanol toxicity, but also differences in
methanol pharmacokinetics and pharmacodynamics. NEDO (1987) reported that in rats exposed
to methanol pre- and postnatally, 6- and 8-week-old male progeny had significantly lower brain
weights at 1,000 ppm, whereas females only showed decreases at 2,000 ppm. In general, gender-
related differences in distribution and clearance of methanol may result from the greater muscle
mass, larger body size, decreased body fat, and increased volumes of distribution in males
compared to females.
5.1.3.2.2. Animal-to-human Extrapolation UFA
A factor of 3 was applied to account for uncertainties in extrapolating from rodents to
humans (UFA). Application of a full UF of 10 would depend on two areas of uncertainty:
toxicokinetic and toxicodynamic. The rodent-to-human toxicodynamic uncertainty is addressed
by a factor of 3, as is the practice for deriving RfCs (U.S. EPA, 1994b). In this assessment, the
toxicokinetic component of uncertainty is addressed by the determination of a HEC through the
use of PBPK modeling. Use of PBPK-estimated maternal blood methanol levels for the
estimation of HECs allows for the use of data-derived extrapolations rather than standard
methods for extrapolations from external exposure levels. Although PBPK model uncertainties
exist, for reasons discussed below, the toxicokinetic uncertainty is reduced to a value of 1 for
both of the candidate principal studies.
There is uncertainty surrounding the identification of the proximate teratogen of
importance (methanol, formaldehyde, or formate) for PBPK modeling, but it is not considered to
be substantial enough to warrant a higher uncertainty factor. A review of the reproductive and
developmental toxicity of methanol by a panel of experts concluded that methanol, not its
metabolite formate, is likely to be the proximate teratogen and that blood methanol level is a
useful biomarker of exposure (NTP-CERHR. 2004: Dormanetal.. 1995). The NTP-CERHR
Expert Panel based their assessment of potential methanol toxicity on an assessment of
5-19
-------
circulating blood levels (NTP-CERHR, 2004). EPA has chosen to use blood methanol levels as
the dose metric for RfC derivation primarily based on evidence that the toxic moiety is not likely
to be the formate metabolite of methanol (NTP-CERHR, 2004). While in vitro evidence indicates
that formaldehyde is more embryotoxic than methanol and formate (Harris et al., 2004; 2003),
the high reactivity of formaldehyde would limit its unbound and unaltered transport as free
formaldehyde from maternal to fetal blood (Thrasher and Kilburn, 2001) (see discussion in
Section 3.3). Thus, even if formaldehyde is ultimately identified as the proximate teratogen,
methanol would likely play a prominent role, at least in terms of transport to the target tissue.
Further discussions of methanol metabolism, dose metric selection, and MOA issues are in
Sections 3.3 and 4.7.
There is uncertainty regarding whether the rat and human PBPK models adequately
characterize species differences. However, given the chosen dose metrics (AUC or Cmax for
methanol in blood), uncertainties in the PBPK modeling of methanol are not expected to be
substantially greater for one species than another. Specifically, the analysis of parameter
sensitivity and uncertainty for the PBPK modeling performed with human and rat data gave
similar results as to how well the model fit the available data (Appendix B). Thus, the human and
rat PBPK model performed similarly using these dose metrics for comparisons between species.
HEC predictions from the models can vary depending on the dose metric selected (e.g.,
AUC or Cmax), but this is not a major source of uncertainty for the following reasons. In the case
of the mouse cervical rib endpoint, the choice of the Cmax dose metric was well justified based on
studies that show a narrow gestational window of susceptibility for this endpoint (Rogers and
Mole, 1997; Bolon et al., 1993). In the case of the rat brain weight endpoint, the choice of the
AUC dose metric was well justified based on studies which show an exacerbation of the effect
from cumulative exposure (NEDO, 1987; TRL, 1986). Study conditions that involved nearly 24
hours of exposure, resulted in an HEC estimate that was not significantly different (-10% lower)
than the HEC estimate that would be obtained using Cmax as the dose metric.
For estimation of an HEC from the NEDO (1987) rat study, uncertainty that could result
in the underestimation of toxicity exists regarding the use of maternal blood levels because of
possible species differences in the relation of maternal blood levels estimated by the model to
fetal and neonatal blood levels that would be obtained under the gestational, postnatal and
lactational exposure scenario. Young animals have different metabolic and physiological profiles
than adults. This fact, coupled with multiple routes of exposure, complicate the prediction of
5-20
-------
internal dose to the offspring.56 Further, methanol dosimetry data are not available for rat pups,
human infants, lactating rat dams, nor lactating human mothers (particularly, amounts in breast
milk). Given the high aqueous solubility of methanol, it may be reasonable to assume that
concentrations expressed in breast milk equal those in maternal blood. However dosimetry in the
developing infant would depend on when and to what extent metabolic capacity develops in rat
pups versus human infants. So while it would be possible to extrapolate the existing adult models
to those life-stages, such extrapolations, for the infant in particular, would be quite speculative
and uncertain. However, it is reasonable to assume that the ratio of the difference in blood
concentrations between a human infant and mother would be similar to and not significantly
greater than the difference in blood concentrations between a rat pup and their rat dam. This
assumption is based largely on the fact that key parameters and factors which determine the ratio
of fetal or neonatal versus maternal methanol blood levels in humans either do not change
significantly with age (partition coefficients, relative blood flows) or scale in a way that is
common across species (allometrically). While there is uncertainty surrounding this assumption,
it is not likely to have a major impact given that health-effects data indicate that most of the
effects of concern are due to in utero exposure, with a relatively small influence due to postnatal
exposures.
5.1.3.2.3. Database UFD
For determining the application of the database UF (UFo), EPA's guidance (U.S. EPA,
2002) states, "In addition to identifying toxicity information that is lacking, review of existing
data may also suggest that a lower reference value might result if additional data were
available.'" Based upon this guidance, a UFD of 3 was applied to account for deficiencies in the
toxicity database that may result in a lower reference value. While the database for methanol
toxicity is extensive in terms of the laboratory species and study design coverage, consisting of
chronic and developmental toxicity studies in rats, mice, and monkeys, a two-generation
reproductive toxicity study in rats, and neurotoxicity and immunotoxicity studies, there still
remains some uncertainty with respect to the potency, importance and relevance of reproductive,
developmental and chronic effects observed in monkeys. With regard to potency, uncertainty is
warranted given evidence that these effects have been observed in monkeys with average blood
levels that are close to, and in one case as little as 0.5 mg/L higher than, the range of
uncontaminated background levels in humans (see Figure 5-4 and discussion in Sections 5.3.6
56Stern et al. (1996) reported that when rat pups and dams were exposed together during lactation to 4,500 ppm
methanol in air, methanol blood levels in pups from GD6-PND21 were approximately 2.25 times greater than those
of dams. It is reasonable to assume that similar differences in blood methanol levels would be observed in the
NEDO (1987) FI study, as the exposure scenario is similar to that of Stern et al. (1996).
5-21
-------
and 5.3.7). Also, uncertainty regarding the potency, importance and relevance of these effects has
been expressed in the external peer review of this assessment (e.g., Appendix A external review
comments in response to Charge D3) and in three independent peer reviews of the individual
monkey studies (discussed below).
As discussed in Section 5.1.1.1, the available monkey studies are considered inadequate
or inferior to the candidate principal rodent studies for the purposes of RfC/D derivation. As has
been suggested by one of the peer reviewers who questioned the need for a 3-fold UFD
(e.g., Appendix A external review comments in response to Charge B4), this deficiency in the
dose-response data would not normally warrant a UFD given the scope of the existing database
and the qualitative value of the chronic and developmental monkey studies for hazard
identification. However, this deficiency is of particular concern for methanol given (1) metabolic
similarities that suggest monkeys should most closely represent the potential for effects in
humans (see Section 3.1) and (2) uncertainties regarding the importance and relevance of the
monkey effects mentioned above and discussed further below.
The reproductive effects (discussed in Section 4.3.2) and the developmental
neurobehavioral effects (discussed Section 4.4.2) reported by Burbacher et al. (2004a; 2004b:
1999a: 1999b) were independently peer reviewed by HEI (Burbacher et al., 2004a: 2004b:
1999a: 1999b) and NTP-CERHR (2004. 2003). The NEDO (1987) acute and chronic studies
were also independently peer reviewed (ERG, 2009). All three of these independent peer reviews
concluded that these studies identified effects of potential relevance but uncertain adversity that
warrant further research. For example, with respect to the Burbacher et al. (2004a; 2004b: 1999a:
1999b) CERHR (2004) stated that "the Panel could not determine whether or not the possible
effects observed in late gestation were treatment-related" and that the positive findings in DNT
tests provide "... evidence of subtle, but not definitive, adverse effects that are ... important from
a qualitative perspective" and suggested specific additional research topics to help resolve these
uncertainties. With respect to NEDO (1987) monkey studies, peer reviewers (ERG, 2009) noted
the small dose group sizes and profound data gaps in the report (e.g., materials and methods,
statistical methods, data), and also suggested additional research to improve both the qualitative
and quantitative interpretation of the NEDO (1987) findings.
In contrast to the data on chronic and reproductive toxicity, the developmental
neurotoxicity data are comparable across the two species and, of the uncertain effects observed in
monkeys, the results of the visually directed reaching (VDR) test are likely to be the most
reliable, discernible and relevant (see discussion in Section 4.4.2 and the BMD analysis in
Appendix D). Also, EPA's guidance (U.S. EPA, 2002) places particular emphasis on database
deficiencies in the area of developmental toxicity, stating that "If data from the available
toxicology studies raise suspicions of developmental toxicity and signal the need for
5-22
-------
developmental data on specific organ systems (e.g., detailed nervous system, immune system,
carcinogenesis, or endocrine system), then the database factor should take into account whether
or not these data are available and used in the assessment and their potential to affect the POD
for the particular duration RfD or RfC under development." Table 5-5 provides a comparison of
the lowest methanol blood LOAELs (excluding background) observed in rodent and monkey
developmental neurotoxicity studies. Even after using EPA's rat PBPK and monkey PK models
to convert external doses to internal blood levels (to account for toxicokinetic differences), the
rodent LOAEL blood level is 12-fold higher than the monkey LOEL blood level. Some of this
12-fold difference may be due to differences in species sensitivity, for which the UFA of 3-fold is
intended to account, but some of the difference may be due to other factors, including whether
appropriate and comparable endpoints were examined and whether appropriate study designs and
quality control measures were used. To account for these additional factors, a 3-fold UFD is
applied.
Table 5-5 Comparison of the lowest rodent and monkey methanol blood LOAELs
(excluding background) observed in developmental neurotoxicity studies.
Reference
NEDO (1987) Follow-up
study, F-i generation
Burbacheret al. (2004a;
2004b: 1999a: 1999b)
Rodent: Monkey Methanol
Strain/ Exposure
Rat, S-D
20 hr/day, GDO through F-i
generation
Monkey, M. fascicularis
2.5 hr/day, 7 days/wk;
premating, mating and
gestation
Blood Level Ratio
Effect
Reduced brain
weight
Neurotoxicology.
deficits (reduced
VDR test results)
LOAEL3 (ppm; mg/L bloodb)
1,000 ppm;
115 mg/L
600 ppm;
1 0 mg/L
12
5.1.3.2.4. Extrapolation from subchronic to chronic UFs
A UF of 1 was used for extrapolation from less than chronic results because
developmental toxicity (extra cervical rib incidence and decreased brain weight) was used as the
critical effect. The developmental period is recognized as a susceptible lifestage where exposure
during certain time windows is more relevant to the induction of developmental effects than
lifetime exposure (U.S. EPA. 1991).
5-23
-------
5.1.3.2.5. LOAEL-to-NOAEL extrapolation UFs
A UF of 1 was used for LOAEL to NOAEL (UPi,) because the current approach is to
address this extrapolation as one of the considerations in selecting a benchmark response (BMR)
for BMD modeling. In this case, the endpoint and benchmark response level employed for the
RfD/C derivation is appropriate for use in deriving the RfD/C under the assumption that it
represents a minimal biologically significant change.
5.1.3.3. Confidence in the RfC
The confidence in the RfC is medium to high. Confidence in the Rogers et al. (1993b)
study is high and confidence in the NEDO (1987) developmental studies is medium. The Rogers
et al. (1993b) study was well designed, as it included large sample sizes, and was well
documented, peer reviewed and published. While there are issues with the lack of detail
regarding methods and results in the NEDO (1987) report, the observed effect (brain weight
reduction) is a relevant endpoint that has been reproduced in an oral study of adult rats (TRL,
1986), and the exposure regimen involving pre- and postnatal exposures addresses a potentially
sensitive human subpopulation. Thus, the overall confidence in the two critical studies is
medium to high. Confidence in the database is medium. Though skeletal and brain effects have
been demonstrated and corroborated in multiple animal studies in rats, mice, and monkeys, some
study results were not quantifiable, thus there is uncertainty regarding which is the most relevant
test species, and there is limited data regarding reproductive or developmental toxicity of
methanol in humans. There is also uncertainty regarding the potential active agent — the parent
compound, methanol, formaldehyde, formate or some other (e.g., reactive oxygen species) agent.
There are deficiencies in the knowledge of the metabolic pathways of methanol in the human
fetus during early organogenesis, when the critical effects can be induced in animals. Thus, the
medium-to-high confidence in the critical studies and the medium confidence in the database
together warrant an overall confidence descriptor of medium to high.
5.1.4. Previous RfC Assessment
The health effects data for methanol were assessed for the IRIS database in 1991 and
were determined to be inadequate for derivation of an RfC.
5-24
-------
5.2. Oral Reference Dose (RFD)
In general, the RfD is an estimate of a daily exposure to the human population (including
susceptible subgroups) that is likely to be without an appreciable risk of adverse health effects
over a lifetime. It is derived from a POD, generally the 95 percent lower bound confidence limit
on the BMD, with uncertainty factors applied to reflect limitations of the data used. The RfD is
expressed in terms of mg/kg-day of exposure to a substance and is derived by a similar
methodology as is the RfC. Ideally, studies with the greatest duration of exposure and conducted
via the oral route of exposure give the most confidence for derivation of an RfD. For methanol,
the oral database is currently more limited than the inhalation database. With the development of
PBPK models for methanol, the inhalation database has been used to help bridge gaps in the oral
database to derive an RfD. As for the RfC, limitations and uncertainties associated with the
derivation of the RfD are addressed through the use of the BMD modeling approach, rat and
human PBPK models and uncertainty factors.
5.2.1. Choice of Principal Study and Critical Effect-with Rationale and
Justification
No studies have been reported in which humans have been exposed subchronically or
chronically to methanol by the oral route of exposure and thus, would be suitable for derivation
of an oral RfD. Data exist regarding effects from oral exposure in experimental animals, but they
are more limited than data from the inhalation route of exposure (see Sections 4.2, 4.3, and 4.4).
Only two oral studies of 90 days duration or longer in animals have been reported
(Soffritti et al.. 2002: TRL. 1986) for methanol. U.S. EPA CTRL. 1986) reported that there were
no differences in body weight gain, food consumption, or gross or microscopic evaluations in
Sprague-Dawley rats gavaged with 100, 500, or 2,500 mg/kg-day methanol versus control
animals. Liver weights in both male and female rats were increased, although not significantly, at
the 2,500 mg/kg-day dose level, suggesting a treatment-related response despite the absence of
histopathologic lesions in the liver. Brain weights of high-dose group males and females were
significantly less than control animals at terminal (90-day) sacrifice. The data were not reported
in adequate detail for dose-response modeling and subsequent BMD estimation. Based primarily
on the qualitative findings presented in this study, the 500 mg/kg-day dose was deemed to be a
NOAEL.57
57 U.S. EPA [TRL (1986)1 did not report details required for a BMD analysis such as standard deviations for mean
responses.
5-25
-------
The only lifetime oral study available was conducted by Soffritti et al. (2002) in Sprague-
Dawley rats exposed to 0, 500, 5,000, 20,000 ppm (v/v) methanol, provided ad libitum in
drinking water. Based on default, time-weighted average body weight estimates for Sprague-
Dawley rats (U.S. EPA. 1988). average daily doses of 0, 46.6, 466, and 1,872 mg/kg-day for
males and 0, 52.9, 529, 2,101 mg/kg-day for females were reported by the study authors. All rats
were exposed for up to 104 weeks, and then maintained until natural death. The authors report no
substantial changes in survival nor was there any pattern of compound-related clinical signs of
toxicity. The authors did not report noncancer lesions, and there were no reported compound-
related signs of gross pathology or histopathologic lesions indicative of noncancer toxicological
effects in response to methanol.
Five oral studies investigated the reproductive and developmental effects of methanol in
rodents (Aziz et al.. 2002: Fuetal.. 1996: Sakanashi et al.. 1996: Rogers etal.. 1993b: Infurna
and Weiss, 1986), including three studies that investigated the influence of folic acid diets on the
effects of methanol exposures (Aziz et al., 2002: Fuetal., 1996: Sakanashi et al., 1996). Infurna
and Weiss (1986) exposed pregnant Long-Evans rats to methanol at 2,500 mg/kg-day in drinking
water on either GDIS-GDI? or GD17-GD19. Litter size, pup birth weight, pup postnatal weight
gain, postnatal mortality, and day of eye opening were no different in treated animals versus
controls. Mean latency for nipple attachment and homing behavior (ability to detect home
nesting material) were different in both methanol treated groups, and these differences were
significantly different from controls. Rogers et al. (1993b) exposed pregnant CD-I mice via
gavage to 4 g/kg-day methanol, given in 2 equal daily doses. Incidence of cleft palate and
exencephaly was increased following this maternal exposure to methanol. Also, an increase in
totally resorbed litters and a decrease in the number of live fetuses per litter were observed.
Aziz et al. (2002). Fu et al. (1996). and Sakanashi et al. (1996) investigated the role of
folic acid in methanol-induced developmental neurotoxicity. Like Rogers et al. (1993b), the first
2 studies observed that an oral gavage dose of 4-5 g/kg-day methanol during GD6-GD15 or
GD6-GD10 resulted in an increase in cleft palate in mice fed sufficient folic acid diets, as well as
an increase in resorptions and a decrease in live fetuses per litter. Fu et al. (1996) also observed
an increase in exencephaly in the folic acid sufficient (FAS) group. Both studies found that an
approximately 50% reduction in maternal liver folate concentration resulted in an increase in the
percentage of litters affected by cleft palate (as much as threefold) and an increase in the
percentage of litters affected by exencephaly (as much as 10-fold). Aziz et al. (2002) exposed rat
dams throughout their lactation period to 0, 1, 2, or 4% v/v methanol via drinking water,
equivalent to doses of approximately 480, 960 and 1,920 mg/kg-day.58 Pups were exposed to
58 Assuming that Wistar rat drinking water consumption is 60 mL/kg-day (Rogers et al.. 2002X 1% methanol in
drinking water would be equivalent to 1% x 0.8 g/mL x 60 mL/kg-day = 0.48 g/kg-day = 480 mg/kg-day.
5-26
-------
methanol via lactation from PND1-PND21. Methanol treatment at 2% and 4% was associated
with significant increases in activity (measured as distance traveled in a spontaneous locomotor
activity test) in the FAS group (13 and 39%, respectively) and most notably, in the folic acid
deficient (FAD) group (33 and 66%, respectively) when compared to their respective controls. At
PND45, the condition avoidance response (CAR) in FAD rats exposed to 2% and 4% methanol
was significantly decreased by 48% and 52%, respectively, relative to nonexposed controls. In
the FAS group, the CAR was only significantly decreased in the 4% methanol-exposed animals
and only by 22% as compared to their respective controls.
5.2.1.1. Route-to-Route Extrapolation
Developmental effects are considered the most sensitive effects of methanol exposure
(see Section 5.1.1). EPA has derived an RfD by using developmental response data from the
candidate principal inhalation studies through route-to-route extrapolation employing the EPA
PBPK model (see Sections 3.4 and 5.1). Several factors support use of route-to-route
extrapolation for methanol. The oral database has significant limitations, including the limited
reporting of noncancer findings in the subchronic (TRL, 1986) and chronic studies (Soffritti et
al., 2002) of rats, and the use of high dose levels in the rodent oral developmental studies. In
addition, the limited data from oral studies indicate similar effects as reported via inhalation
exposure (e.g., the brain and fetal skeletal system are targets of toxicity). Further, methanol has
been shown to be rapidly and well-absorbed by both the oral and inhalation routes of exposure
(NTP-CERHR. 2004: Kavet and Nauss. 1990). Once absorbed, methanol distributes rapidly to all
organs and tissues according to water content, regardless of route of exposure.
As with the species-to-species extrapolation used in the development of the RfC, the dose
metric used for species-to-species and route-to-route extrapolation of inhalation data to oral data
is the Cmax (in the case of the mouse cervical rib endpoint) or AUC (in the case of the rat brain
weight endpoint) of methanol in blood. Simulations of human oral methanol exposure were
conducted using the model parameters as previously described for human inhalation exposures,
with human oral kinetic/absorption parameters from Sultatos et al. (2004) (i.e., kas = 0.2, ks; =
3.17, and ka; = 3.28). Human oral exposures were assumed to occur during six drinking episodes
during the day, at times 0, 3, 5, 8, 11, and 15 hours from the first ingestion of the day. For
example, if first ingestion occurred at 7 a.m., the six episodes would be at 7 a.m., 10 a.m.,
12 noon, 3 p.m., 6 p.m., and 10 p.m. Each ingestion event was treated as occurring over 3
minutes, during which the corresponding fraction of the daily dose was infused into the stomach
lumen compartment. The fraction of the total ingested methanol simulated at each of these times
was 25%, 10%, 25%, 10%, 25%, and 5%, respectively. Six days of exposure were simulated to
5-27
-------
allow for any accumulation (visual inspection of plots showed this to be finished by the 2nd or
3rd day), and the results for the last 24 hours were used. Dividing the exposure into more and
smaller episodes would decrease the estimated peak concentration, but have little effect on AUG.
This dose metric was used for dose-response modeling to estimate the BMDL or PODinternai.
5.2.2. RfD Derivation-Including Application of Uncertainty Factors
5.2.2.1. Derivation of Candidate RfDs
Inhalation studies considered in the derivation of the RfC are used to supplement the oral
database through use of route-to-route extrapolation, as previously described. As for the RfC,
BMD approaches were applied to the existing inhalation database, and the EPAPBPK model was
used for species-to-species extrapolations. Table 5-6 presents the candidate RfDs based on the
selected developmental endpoints, the BMD modeling approaches employed (see Appendix D
for details), and the applied UFs (see Section 5.2.2.2) using route-to-route extrapolation
(obtained from PBPK models described in Appendix B). Like the RfC derivation, the internal
BMDL (PODinternai) values are divided by a total UF of 100 (UFH of 10, UFA of 3 and a UFD of 3)
to yield an RfDintemai, which is converted to a candidate RfD using the human PBPK model
described in Appendix B.59 Candidate RfDs estimated from the Rogers et al. (1993b) study for
extra cervical rib incidence in mice employing Cmaxas the dose metric were 4.1 and 1.9 mg/kg-
day using BMRs of 10% and 5%, respectively. Candidate RfDs estimated from the NEDO
(1987) study based on decreases in brain weight at 6 weeks of age in male rats exposed during
gestation and throughout the FI generation employing AUC as the dose metric were 5.4 and
4.0 mg/kg-day for BMRs of 5% change relative to control mean and one SD from the control
mean, respectively. Because the cervical rib endpoint resulted in the lowest of the candidate RfD
estimates, it was chosen as the critical endpoint for use in the RfD derivation.
RfD = 43.1 mg/L H- 100 = 0.43 mg/L =^>PBPK^> 2 mg/kg-day
(rounded to 1 significant figure)
59 An algebraic equation is provided near the end of Appendix B that approximates the PBPK model predicted
relationship between methanol AUC above background and the HED in mg/kg-day.
5-28
-------
Table 5-6 Summary of PODs for critical endpoints, application of UFs and conversion to
candidate RfDs using PBPK modeling.
Rogers et al. (1993b)
(mouse cervical rib Cmax)
10% BMR 5% BMR
BMDL = PODmternai 90.9 mg/L 43.1 mg/L
RfDinternai = PODinternai/UFsa 0.909 mg/L 0.43 mg/L
|RfD (mg/kg/day)b 3.9 1.9
NEDO (1987)
(rat brain wt. AUC)
5% BMR 1 SD BMR
1,183mg-hr/L 858 mg-hr/L
11.83mg-hr/L 8.58 mg-hr/L
7.1 5.2
aUFA=3; UFD = 3; UFH = 10; UFS = 1; UFL = 1; product of all UFs= 100; see Section 5.2.2.2 below for details.
bEach candidate RfD is the oral dose predicted to yield a blood concentration equal to its corresponding RfDintemai, using the human
PBPK model described in Appendix B, with a background blood concentration of 2.5 mg/L, which corresponds to an estimated
maximum background exposure rate for a diet including fruits and vegetables of 1,600 mg/day in a 70-kg person (see discussion in
Section 5.3.6); the final RfC is rounded to one significant figure.
5.2.2.2. Application of UFs
Because the same studies, endpoints, BMD methods and PBPK models used to derive the
candidate RfCs were also used to calculate the candidate RfDs, the RfD derivation uses the same
values for uncertainty factors as are described for the RfC derivation (Section 5.1.3.2).
Consistent with the RfC derivation, in order to avoid the uncertainty associated with applying the
human PBPK model to exposure levels that are above the levels for which the model was
calibrated and to account for possible non-linearities in the external versus internal dose
relationships at high doses, EPA applied the UFs to the internal BMDL (PODinternai) prior to HED
derivation to obtain an RfDinternai (see Table 5-6).
5.2.2.3. Confidence in the RfD
The confidence in the RfD is medium to high. Despite the relatively high confidence in
the critical studies, all limitations to confidence as presented for the RfC also apply to the RfD.
Confidence in the RfD is slightly lower than for the RfC due to the lack of adequate oral studies
for the RfD derivation, necessitating a route-to-route extrapolation.
5.2.3. Previous RfD Assessment
The previous IRIS assessment for methanol included an RfD of 0.5 mg/kg-day that was
derived from a U.S. EPA (TRL, 1986) subchronic oral study in which Sprague-Dawley rats
5-29
-------
(30/sex/dose) were gavaged daily with 0, 100, 500, or 2,500 mg/kg-day of methanol. Elevated
levels of serum glutamic pyruvic transaminase (SGPT), serum alkaline phosphatase (SAP), and
increased, but not statistically significant, liver weights in both male and female rats suggest
possible treatment-related effects in rats dosed with 2,500 mg methanol/kg-day, despite the
absence of supportive histopathologic lesions in the liver. Brain weights of both high-dose group
males and females were significantly less than those of the control group. Based on these
findings, 500 mg/kg-day of methanol was considered a NOAEL in this rat study. Application of a
1,000-fold UF (interspecies extrapolation, susceptible human subpopulations, and subchronic to
chronic extrapolation) yielded an RfD of 0.5 mg/kg-day.
5.3. Uncertainties in the Inhalation RfC and Oral RfD
The following is a more extensive discussion of the uncertainties associated with the RfC
and RfD for methanol beyond that which is addressed quantitatively in Sections 5.1.2, 5.1.3, and
5.2.2. A summary of these uncertainties is presented in Table 5-7.
5-30
-------
Table 5-7 Summary of uncertainties in methanol noncancer assessment.
Consideration
Potential Impact
Decision
Justification
Choice of
study/endpoint
Minimal impact. RfD
and RfC estimates
from candidate
principal studies were
extremely close to one
another.
RfC is based on brain
weight reduction in rats
NEDO (1987); RfD is
based on cervical rib
anomalies in mice
Rogers et al. Q993b)
The chosen endpoints were observed in
adequate studies, have been observed in
other rodent studies, are considered
biological significant and relevant to
humans, and were the most sensitive of the
quantifiable endpoints for their respective
route of exposure.
Choice of model for
BMDL estimation
BMDLs from
adequately fitting
models differed by 5-
foldfortheRfC,
indicating high model
dependence, and were
within 25% of each
other for the RfD,
indicating little model
dependence.
Hill model was chosen
for derivation of the
POD for the RfC and
NLogistic model was
chosen for derivation of
the POD for the RfD.
Hill model was chosen because it resulted
in the lowest BMDL from among a wide
range (>3-fold) of BMDL estimates from
adequately fitting models. NLogistic model
was the best fitting model in accordance
with U.S. EPA (2012a) criteria.
Route-to-Route
Extrapolation method
Raises the RfD 7-fold
above 1988 methanol
RfD of 0.5 mg/kg-day
based on oral study by
TRL (1986)
Human PBPK model
was used to estimate
HED from blood levels
reported in Roger et al.
(1993b) study.
Rogers et al. (1993b) study was a high
quality study, measured a sensitive and
relevant endpoint, provided measured blood
concentrations that could be converted to
oral doses with the EPA human PBPK
model.
Statistical uncertainty
at POD (sampling
variability due to
bioassay size)
Choice of
species/gender
Relationship of the RfC
and RfD to
Background Blood
Levels and Blood
Levels In Monkeys
Associated with
Unquantifiable Effects
POD would be -50%
higher if BMD were
used
PODsforthe RfC and
RfD estimates based
on rat and mouse data
are similar; POD
estimates based on
monkey data would be
-30-50% lower
Adverse effects may
result if RfC and RfD
are too large.
A BMDL was used as
the POD
RfC and RfD were
based on the most
sensitive of relevant
and quantifiable
endpoints in the most
sensitive species and,
in the case of the RfC,
also in the most
sensitive gender.
RfD and RfCs are
deemed adequately
protective and
reasonable.
The BMDL is the lower limit of a one-sided
95% confidence interval on the BMD.
Mouse and rat studies gave similar results
for RfC/D. Qualitative evidence from NEDO
(1987). Burbacher et al. (2004b) and
Burbacher et al. (2004a) suggest that
monkeys may be a sensitive species, but
data are not as reliable for quantification.
No gender differences were noted by
Rogers et al. (1993b), but NEDO (1987)
reported slightly greater brain weight
changes in male offspring.
RfD and RfC would cause an appreciable
increase in the number of individuals with
blood levels above 2.5 mg/L, the high end
of the range of background methanol blood
levels associated with a diet that includes
fruits and vegetables.
5.3.1. Choice of Study/Endpoint
As discussed in Sections 5.1.1 and 5.2.1, developmental effects observed in two
candidate principal studies were considered for the purposes of RfC/D derivation. Brain weight
5-31
-------
reduction in rats (NEDO, 1987) and cervical rib anomalies in mice (Rogers et al., 1993b) were
selected as the endpoints reported in these studies that are most appropriate for RfC/D derivation.
Candidate RfCs derived based on these endpoints ranged from 17.8 to 41.8 mg/m3 (Table 5-4).
Candidate RfDs derived using route-to-route extrapolation and based on the same endpoints as
the candidate RfCs ranged from 1.9 to 5.4 mg/kg-day (Table 5-6).
Uncertainties associated with the Rogers et al. (1993b) study results are primarily with
respect to the relevance of developmental studies in rodents to humans, which is discussed in
Sections 5.1.1.2.1 and 5.3.5. There is less uncertainty associated with the Rogers et al. (1993b)
study methods and reporting because it has undergone independent peer review, is well
documented, used robust group sizes, and reports effects that have been observed by other
laboratories. In addition individual animal data were made available by the authors (see
Appendix D).
Uncertainties in the NEDO (1987) developmental study are primarily associated with the
reproducibility of the brain weight endpoint and the level and quality of study documentation.
Neonatal reduction in brain weight is not as well documented across laboratories and across
species and strains of test animals as is the fetal cervical rib endpoint. However, this is not a
major concern given that reduced brain weight following methanol gavage exposure was
reported in adult S-D rats by another laboratory CTRL. 1986). and in two other NEDO (1987)
S-D rat developmental inhalation studies, including in another teratogenicity study and in both
generations of a two generation study. In addition, CNS effects have been reported in inhalation
studies of monkeys, including abnormal brain histopathology following chronic methanol
exposure (NEDO, 1987) and delayed neurological development following gestational methanol
exposure (Burbacher et al., 2004a: 2004b: 1999a: 1999b). Further, the primary reason that the
developmental brain weight effect has not been identified in other species could be that it has not
been the focus of other laboratory research. More important is the uncertainty associated with the
lack of documentation in the NEDO (1987) supplementary developmental study that formed the
basis for EPA's benchmark dose analysis. The three primary reporting deficiencies in this study,
identified during external peer review (ERG, 2009), were: (1) lack of information on the number
and health of pregnant dams, (2) not reporting the body weight of the offspring, and (3) lack of a
statistical analysis of response data. While the methods employed in this supplementary study
were not adequately described, the methods used in the parent two-generation study were
adequately described. Because these two studies were conducted in the same laboratory, it is
reasonable to assume that the supplementary study was performed under the same protocol as the
two-generation study, starting with a number of F0 parents appropriate for a one-generation
5-32
-------
developmental study.60 While data related to maternal or gestational outcomes in the
supplementary study were not provided, signs of overt maternal toxicity were not reported in the
two-generation study at similar exposure levels, and thus it is reasonable to assume that they did
not occur, and would have been reported had they been observed, in the supplementary study.
With respect to the second source of documentation-related uncertainty, the NEDO
supplementary study only reported means and standard deviations for absolute brain weight
change and did not report body weight data for the offspring. However, body weight data
reported in the parent two-generation study did not indicate a body weight effect in the exposed
FI or F2 generation pups. Further, EPA neurotoxicity guidelines (U.S. EPA, 1998a) state that a
"change in brain weight is considered to be a biologically significant effect," and further states
that "it is inappropriate to express brain weight changes as a ratio of body weight and thereby
dismiss changes in absolute brain weight." The third source of documentation-related uncertainty
noted by the external peer reviewers of the NEDO supplementary study, was that the
investigators did not report the results of a more appropriate (e.g., ANOVA) test for statistical
significance. This is not a significant source of uncertainty because EPA did not rely on the
NEDO statistical tests, but instead performed its own more definitive trend test benchmark dose
analysis of the data (see Appendix D). In summary, while there are uncertainties concerning the
NEDO (1987) supplementary study that forms the basis of the RfC, particularly with respect to
reporting deficiencies, there is sufficient ancillary evidence to offset these concerns and allow for
the consideration the this study as a basis for RfC or RfD derivation.
The use of reproductive and neurotoxicity endpoints reported in developmental
(Burbacher et al., 2004a: 2004b: 1999a: 1999b) and chronic (NEDO, 1987) monkey studies
would potentially result in lower reference values, but significant uncertainties associated with
the reported dose-response data (e.g., an absent or questionable dose-response relationship)
preclude their use as the basis for an RfC. Burbacher et al. (2004a; 2004b: 1999a: 1999b)
exposedM. fascicularis monkeys to 0, 200, 600, or 1,800 ppm (0, 262, 786, and 2,359 mg/m3)
methanol 2.5 hours/day, 7 days/week during premating/mating and throughout gestation
(approximately 168 days). They observed a slight, but statistically significant, shortening of
gestation period in all exposure groups. As discussed in Sections 4.3.2 and 5.1.1.2, there are
questions concerning this effect and its relationship to methanol exposure. In these studies,
60 The number of F0 parents in the supplemental experiment was not reported, but the number of pups per dose
group was presented, and it is reasonable to assume that, consistent with the culling protocol used for the two-
generation study (NEDO. 1987 pages 185 and 189 ), each dose group pup came from a different litter (to avoid
"litter correlation" issues). EPA developmental neurotoxicity guidelines (U.S. EPA. 1998b) require that "on
postnatal day 11, either 1 male or 1 female pup from each litter (total of 10 males and 10 females per dose group)
should be sacrificed." Hence, by examining more than 10 male and 10 female litter-specific pups per dose group at
three time points (3, 6 and 8 wks), the NEDO supplementary study would exceed EPA recommendations for this
type of study.
5-33
-------
neurobehavioral function was assessed in infants during the first 9 months of life. Two tests out
of nine returned positive results that were possibly related to methanol exposure. The Fagan test
of infant intelligence indicated small, but non significant deficits of performance (decreased time
spent looking at novel faces versus familiar faces) in treated infants. Also, VDR performance
was reduced in all treated male infants, and was significantly reduced in the 1,800 ppm
(2,359 mg/m3) group for both sexes and in the 600 ppm (786 mg/m3) group for males. However,
as discussed in Appendix D, an overall dose-response trend for this endpoint was not apparent in
males and was only marginally significant in females, which had a larger overall sample size
across dose groups than males (21 females versus 13 males). A benchmark dose analysis was
done for the VDR effect in female monkeys using Cmax (adjusted for background blood
methanol) as the dose metric (results detailed in Appendix D.4). The BMDL was estimated to be
19.6 mg/L. While there are significant concerns regarding the dose-response relationship for this
VDR endpoint, this BMDL (based on Cmax) is consistent with the BMDLs (based on Cmax and
AUC) estimated from the rodent studies and represents a measure of functional deficit in
sensorimotor development that is possibly consistent with developmental CNS effects (i.e., brain
weight changes) that have been observed in rats (NEDO, 1987). Although the VDR test results
suggest that prenatal exposure to methanol can result in neurotoxicity to the offspring, the use of
such statistically borderline dose-response data is not warranted in the derivation of the RfC or
RfD, given the availability of better dose-response data in other species.
NEDO (1987) examined the chronic neurotoxicity of methanol in monkeys
(M. fascicularis) exposed to 0, 10, 100, or 1,000 ppm (13.1, 131, or 1,310 mg/m3) methanol for
up to 29 months. Multiple effects were noted at 131 mg/ m3, including slight myocardial effects
(negative changes in the T wave on an EKG), degeneration of the inside nucleus of the thalamus,
and abnormal pathology within the cerebral white tissue in the brain. The results support the
identification of 10 ppm (13.1 mg/m3) as the NOAEL for neurotoxic effects in monkeys exposed
chronically to inhaled methanol. However, as discussed in Section 4.2.2.3, there exists
significant uncertainty in the interpretation of these results and their utility in deriving an RfC for
methanol. These uncertainties include lack of appropriate control group data and the limited
nature of the reporting of the neurotoxic effects observed. Thus, while the NEDO (1987) study
suggests that monkeys may be a more sensitive species to the neurotoxic effects of chronic
methanol exposure than rodents, the deficiencies in the reporting of data preclude the use of this
study for the derivation of an RfC.
5-34
-------
5.3.2. Choice of Model for BMDL Estimation
As discussed in Section 5.1.2.1, in the absence of an established MOA or a biological
basis for why one model should be preferred, the choice of model for a dose-response analysis is
based on which model exhibits the best mathematical fit to the experimental data. There is
uncertainty inherent in this approach with respect to how well the selected model reflects the true
dose-response relationship. This uncertainty is increased when there is no biologically basis for
model selection and the dose-response data demonstrate a high degree of model dependence
(e.g., BMDL results vary widely for all models employed). This is the case for the BMD analysis
of the NEDO (1987) rat brain weight endpoint. As described in Appendix D, the BMDL from the
Hill model was selected as the most appropriate model for derivation of an RfC from this
endpoint, in accordance with EPA BMD Technical Guidance (U.S. EPA, 2012a), because it
yielded the lowest BMDL from among a broad range of BMDLs and provided a superior fit in
the low dose region nearest the BMD. If another adequately fitting model had been chosen, the
BMDL would have been as much as 5-fold higher. With respect to the mouse cervical rib
endpoint, model uncertainty is somewhat reduced because the nested Logistic (NLogistic) model
employed has some biological basis, in that it accounts for possible litter-specific covariates and
correlations, and BMDL results from the three nested models employed were within a relatively
small, 1.3-fold range. Therefore, in accordance with EPA BMD Technical Guidance (U.S. EPA,
2012a), the NLogistic model was selected a as the most appropriate model for derivation of an
RfC because it yielded the lowest AIC and exhibited a superior fit in the low dose region nearest
the BMD.
5.3.3. Route-to-Route Extrapolation
To identify a POD from which to derive an RfD based on cervical rib anomalies in mice,
a route-to-route extrapolation was performed using the POD from which the RfC was derived.
One way to characterize the uncertainty associated with this route-to-route extrapolation is to
compare the responses observed in the critical inhalation study to responses observed in similar
oral developmental studies. As discussed in Section 5.2.1, Rogers et al. (1993b) conducted both
an inhalation and oral developmental studies of methanol in CD-I mice. The oral study involved
a single dose of 4,000 mg/kg-day methanol and resulted in effects (i.e., cleft palate and
exencephaly) consistent with the skeletal abnormalities observed in the inhalation developmental
studies in CD-I mice (Rogers and Mole, 1997; Rogers et al., 1993b). Brain weight reductions
observed in rats in the other candidate principal developmental study (NEDO, 1987) have been
observed in an oral study in adult rats exposed to 2,500 mg/kg-day methanol (TRL, 1986). While
5-35
-------
the oral doses associated with adverse effect in the Rogers et al. (1993b) and TRL (1986) oral
studies were 11- to 30-fold higher than the 133-220 mg/kg-day human equivalent oral doses
estimated for BMDLs from the candidate inhalation studies (see Section 5.1.3.2), the observation
of similar effects in the same species following oral and inhalation exposure to methanol reduces
some of the uncertainty associated with a route-to-route extrapolation.
5.3.4. Statistical Uncertainty at the POD
Uncertainty in the BMD used to derive the POD for the RfC/D can be assessed through
confidence intervals. For the Hill and NLogistic models applied to the data for decreased brain
weight in rats and cervical rib anomalies in mice, respectively, there is a degree of uncertainty in
the BMD estimate at the BMR reflected by a 40-50% difference between the 95% one-sided
lower confidence limit (BMDL) and the maximum likelihood estimate of the BMD. Thus, the
POD for the RfC and RfD would be approximately 50% higher if the BMD were used instead of
the BMDL.
5.3.5. Choice of Species/Gender
The RfC derivation was based on decreased brain weight at 6 weeks (postnatal) in male
(the gender most sensitive to this effect) S-D rats (NEDO, 1987) and the RfD was based on
cervical rib anomalies in male and female CD-I mice (Rogers et al., 1993b). If the decreased
brain weight in female rats had been used instead, the RfC would have been approximately 66%
higher than the RfC based on the male data. As discussed in Section 5.3.1, while existing
developmental and chronic studies suggest that monkeys may be the more sensitive and relevant
species, these studies were not chosen for RfC or RfD derivation due to substantial deficiencies
in the NEDO (1987) monkey study and uncertainties in the dose-response data reported in the
Burbacher et al. (2004b: 1999b) study.
Researchers at the University of Toronto (Miller and Wells, 2011; Sweeting et al., 2011)
have suggested that developmental studies in rodents may not be suitable for assessing human
developmental toxicity. Their hypothesis that mouse studies are not relevant to humans is based
on a series of assumptions, as follows:
1) mouse embryos have a higher reliance on catalase over ADH to metabolize
embryonic methanol,
2) catalase has a higher affinity for methanol than reactive oxygen species,
3) due to this affinity, embryonic methanol competitively inhibits catalase
antioxidant activity,
5-36
-------
4) this competitive inhibition results in an increase in embryonic ROS activity, and
5) this increased embryonic ROS activity is the primary MOA responsible for the
teratogenic effects observed in mice following methanol exposure.
The validity of the first of these assumptions is uncertain given the complexity of enzyme
kinetics in humans, the limited knowledge of how a human fetus/infant metabolizes methanol,
existing evidence that a human fetus/infant can metabolize methanol via a mechanism(s) other
than ADH, and the possibility that this alternative mechanism could involve catalase (Tran et al.,
2007). The second assumption has greater validity as it is based on published reports of catalase
affinity (Km) for methanol (Perkins et al., 1995a: Ward et al., 1995) and hydrogen peroxide
(Vetrano et al., 2005). However, there is limited and conflicting evidence on the validity of
assumptions 3 and 4 (i.e., that catalase affinity for methanol can lead to an increase in embryonic
ROS). In order for assumptions 3 and 4 to be true, catalase affinity for methanol would need to
be strong enough to overcome the extremely high reaction rate between catalase and ROS61, and
other enzymes (e.g., glutathione and superoxide dismutase) can also protect against ROS. Miller
and Wells (2011) point out that methanol radicals have been detected by electron spin resonance
spectrometry in methanol intoxicated rats (Skrzydlewska et al., 2000), and methanol derived
adducts have been observed in the bile and urine of rats exposed to methanol and a free radical
spin trapping agent (Mason and Kadi ska, 2003). However, these observations do not answer the
question of whether methanol's impact on catalase activity can cause an overall increase in
embryonic ROS, and evidence to the contrary exists for adult organ systems. For instance, no
increase in a general indicator of tissue oxidative DNA damage [8-hydroxy-2'-deoxyguanosine
(8-oxodG)] was observed in the lungs, livers, bone marrow and spleen of male CD-I mice, DNA
repair deficient knockout mice, NZW rabbits and cynomolgus monkeys given a single i.p.
injection of 2 g/kg methanol and male CD-I mice injected daily for 15 days with 2 g/kg
methanol (McCallum et al., 2011 a: 2011b). With respect to the validity of the fifth assumption, it
has been suggested that in vitro studies that report an enhancement of methanol-induced
embryopathies in glutathione-depleted rat embryos (Harris et al., 2004) provide support for a
ROS-mediated mode of action for methanol developmental toxicity. However, as discussed in
Section 4.7.1, the impact of glutathione depletion on the methanol-induced embryopathies has
also been attributed to a decreased ability to metabolize formaldehyde (Harris et al., 2004). It has
also been suggested that the enhancement of methanol-induced embryopathies in acatalasemic
(aCat; low catalase activity) mouse embryos supports a ROS-mediated mode of action (Miller
and Wells, 2011). However, in vivo studies from the same laboratory using the same strains of
mice as the Miller and Wells (2011) study observed enhanced fetal effects in the hCat mice
61 The interaction rate of catalase with hydrogen peroxide (Kcat) is roughly 40,000,000/second (Garrett and Grisham.
2010).
5-37
-------
similar to those observed in mice by Rogers et al. (2004) and no enhancement of fetal effects in
aCat mice (Siu et al., 2013). Siu et al. (2013) acknowledge that their in vivo results imply no
ROS involvement in the embryopathology of methanol-induced fetal effects in mice. While ROS
may yet be determined to play a role in the pathological progression of methanol-induced fetal
effects in rodents, available information is not consistent or adequate to conclude this or that the
rodent developmental studies are not relevant in the assessment of human developmental toxicity
from methanol exposure.
Sweeting et al. (2011) have suggested that rabbits would be a more appropriate test
species than mice because rabbits may largely metabolize methanol via alcohol dehydrogenase
and more accurately reflect primate methanol and formic acid pharmacokinetic profiles. They
further state that rabbits are resistant to methanol teratogenicity. A developmental study in rabbits
via an appropriate route of exposure would be of interest, particularly if it involved an
investigation of effects over a broad set of gestational days. However, more research is needed
before it can be stated that a rabbit developmental study would be more relevant to humans than
rodent developmental studies and that rabbits are resistant to methanol teratogenicity. The
Sweeting et al. (2011) study assumes that the gestational window of susceptibility for
developmental effects in rabbits following methanol exposure is at or close to that for mice.
While the gestational window of susceptibility for developmental effects in mice is well studied
and documented (Degitz et al., 2004a: Degitz et al., 2004b: Rogers et al., 2004; Rogers and
Mole. 1997: Dorman and Welsch. 1996: FuetaL 1996: DormanetaL 1995: Andrews et al..
1993: Bolon et al., 1993: Rogers et al., 1993a: Rogers et al., 1993b), no studies have been done
to identify the gestational window of susceptibility for methanol exposures in rabbits. As mouse
studies have shown, missing the true gestational window of susceptibility for a species/strain can
make a marked difference in the developmental effects observed (Rogers and Mole, 1997: Bolon
etal.. 1993).
5.3.6. Relationship of the RfC and RfD to Background Levels of Methanol in
Blood
The available data on methanol blood levels (small numbers of studies and individuals,
differing results by study) do not support a precise estimate of a population distribution of blood
methanol levels. However, for the purpose of examining the relationship of the RfC and RfD to a
representative distribution of background of methanol blood levels, EPA has derived a sample
lognormal distribution that is consistent with data from relevant study groups in Table 3-1. The
arithmetic means and standard deviations reported in Table 3-1 across six study groups that did
not involve substantial dietary restrictions other than alcohol, Batterman and Franzblau (1997),
5-38
-------
Batterman et al. (1998). Lee et al. (1992). Sarkola and Eriksson (2001). Turner et al.(2006) and
Woo et al. (2005), were converted to log geometric means (|i) and log geometric standard
deviation (o) using the formulas given by Johnson and Kotz (1970). Then simulated methanol
values derived for the studies were used to fit an overall lognormal distribution.62 The mean and
SD for this sample background distribution are 1.36 mg/L and 0.77 mg/L, respectively. The U.K.
Food Standards Agency estimates that endogenous methanol production ranges from 300 to 600
mg/day (Lindinger et al., 1997) (4.3 to 8.6 mg/kg-day) and that diet can contribute up to an
additional 1,000 mg/day (14.3 mg/kg-day), principally from fruits and vegetables (COT, 2011).
Thus the upper bound of the combined endogenous and dietary exposures estimated in the U.K.
is 23 mg/kg-day. The methanol blood level predicted by EPA's PBPK model for this 23 mg/kg-
day maximum exposure rate is 2.5[4] mg/L, which is slightly below EPA's sample background
distribution mean + 2xSD of 2.9 mg/L. A small percentage (-7%) of the EPA sample background
population is predicted to have methanol blood levels above 2.5 mg/L.
Figure 5-3 illustrates the shift in EPA's sample background methanol blood level
distribution that would be associated with exposing every individual to methanol at the RfC or
the RfD. This analysis assumes that a RfC or RfD exposure would increase each individual's
methanol blood level by the same amount.63 According to this analysis, an RfC or RfD exposure
would increase the percentage of individuals with peak methanol blood levels at or above 2.5
mg/L from -7% to -14%. These estimates are not precise and do not account for interindividual
variability. However, they illustrate that the increase in individuals with higher than 2.5 mg/L
methanol blood levels (i.e., higher than the upper range of background methanol blood levels
associated with a diet that includes fruits and vegetables) following a RfD or RfC exposure
would not be negligible.
62Studies were weighted according to the extent to which they represent the U.S. population. The Sarkola and
Eriksson (2001) restricted alcohol consumption for one week prior to blood testing and was therefore given a weight
of 0.48, commensurate with the percentage of US population that are not regular drinkers (CDC. 2011). Woo et al.
(2005) studied Korean subjects, a unique population prone to having more than one variant of the gene coding for
alcohol dehydrogenase, which causes them to metabolize alcohol at a much higher efficiency than other gene
variants (EngetaL 2007). and was therefore given a weight of 0.036, commensurate with the Asian fraction of the
US population (SSDAN CensusScope. 2010). The other four studies were assigned a weight of one. Using these
weights and assuming that the distribution of each individual study and the overall distribution are both
log-normally distributed, the following simulation was performed: (1) studies were picked at random according to
the weights assigned to form a new combination of six studies; (2) methanol blood levels were randomly generated
from the lognormal distribution characterized by the \i and o estimated for each of the six randomly picked studies;
(3) the n and o for the overall lognormal distribution associated with these blood values were estimated by the
maximum likelihood estimation method; (4) the steps (1) to (3) were repeated 2,000 times and the mean of the 2,000
H and o estimates were used as the final parameters of the overall lognormal distribution.
63 In actuality this quantity will have relevant variability due to interindividual differences. However, we do not have
PK results to predict variability for blood levels for methanol exposures at the RfC/D. If the model predictions are
central estimates of blood levels from exposures, then adding in population variability can be expected to lead to
higher upper percentile estimates for the blood methanol levels in the environmentally exposed population.
5-39
-------
EPA's PBPK model predicts that a continuous daily methanol inhalation exposure
equivalent to the RfC would raise the methanol blood level of an individual with a high end
background methanol blood level of 2.5 mg/L by 0.43 mg/L. The model also predicts that a daily
oral exposure to methanol at the RfD (distributed as six bolus doses) would raise this potentially
susceptible individual's methanol blood level by a daily peak value of 0.46 mg/L and by a daily
average of approximately 0.19 mg/L. Section 5.3.7 discusses the relationship of these increases
to levels that have been reported in monkey studies to be associated with effects of uncertain, but
potential adversity.
Sample Background Blood Level Distribution
Sample Background + Peak RfD
Sample Background + RfC
0.5
1.5 2 2.5 3 3.5 4
mg Methanol/LiterBlood (mg/L)
4.5
5.5
*For the exposure regimen assumed (Section B.2.7), daily increases for an RfD vary between 0.01 and 0.44 mg/L (Appendix B,
Figure B-17).
Figure 5-3 Projected impact of daily peak RfD and RfC exposures on sample background
methanol blood levels (mg MeOH/Liter [mg/L] blood) in humans.
5.3.7. Relationship of the RfC and RfD to Methanol Blood Levels In Monkeys
Associated with Unquantifiable Effects of Uncertain Adversity
As discussed in Section 5.1.3.2.3, EPA believes that the existing methanol database
provides an incomplete characterization of methanol's potential to cause reproductive effects,
developmental neurotoxicity, and chronic neurotoxicity. NTP-CERHR (2003) reported "minimal
5-40
-------
concern" that up to 10 mg/L of methanol in blood would result in reproductive or developmental
toxicity in humans. However, based on an evaluation of the methanol blood levels corresponding
to effects observed in the Burbacher et al. (2004b: 1999b) reproductive and developmental
monkey study using the EPA monkey PK model, EPA believes there may be some uncertainty
associated with the NTP-CERHR (2003) conclusion. Further, the NTP-CERHR (2003) report
focused on the reproductive and developmental effects of methanol, and did not assess the
potential for effects from chronic exposure.
In the Burbacher et al. (2004b; 1999b) study, statistically significant shortened pregnancy
duration was observed in monkeys exposed to 200 ppm methanol, and statistically significant
VDR delay was observed in male monkey infants exposed to 600 ppm methanol for 2 hours
per day (see Section 4.3.2 and 4.4.2). EPA estimates that these two exposures resulted in peak
methanol blood levels in these monkeys of approximately 5 and 12 mg/L, respectively (including
a background level of 2 mg/L; see Appendix D, Table D-10). Also, NEDO (1987) observed
potential signs of CNS effects (minimal fibrosis of "responsive stellate cells" of the brain, likely
astrocytes) in histopathology in monkeys exposed chronically to 100 ppm for 21 hours per day
(see Section 4.4.2), which is estimated by EPA's monkey PK model to be associated with an
increase in methanol blood levels over background levels of approximately 1 mg/L,
corresponding to total methanol blood levels of roughly 3 mg/L (assuming a background in these
monkeys of 2 mg/L; see Appendix B, Table B-8). The significance of this stellate cell/astrocyte
response is uncertain at this time. However, the slight neurological effects (Chuwers et al., 1995)
and increased subclinical biomarkers for inflammation (Mann et al., 2002) in humans acutely
exposed to 200 ppm (262 mg/m3) methanol were associated with just over twice this methanol
blood level, 6.5 mg/L. Further, stellate astroglia are believed to play a key role in the
pathogenesis of CNS disorders and in response to tissue injury and inflammation. Thus, with
further research this endpoint could prove to be an important CNS effects associated with
methanol exposure.
As discussed in Sections 4.3.2 and 4.4.2, EPA could not derive a NOAEL or LOAEL
from the monkey studies which reported these reproductive and neurotoxicity endpoints.
However, as discussed in Section 5.1.3.2.3, these effects were important considerations with
respect to the determination of the database uncertainty factors. Figure 5-4 illustrates how
methanol blood level distributions for RfD and RfC exposures to the EPA sample background
distribution compares with the blood levels that have been associated with these uncertain, but
potentially adverse effects in monkeys. As discussed in the previous section, a RfC or RfD
exposure is expected to raise the methanol blood level of an individual with a high end
background methanol blood level of 2.5 mg/L to just under 3 mg/L, the lowest methanol blood
level that has been associated with these uncertain, but potentially adverse effects.
5-41
-------
0.5
-Sample Background Blood Level Distribution
• Sample Background + Peak RfD
•Sample Background + RfC
Estimated high end of
methanol background
blood levels
NEDO(1987)
Chronic monkey study; 100 ppm;
21 hrs/day; Minimal fibrosis of
"responsive stellate cells;"
%possiblyastrocytes(ERG 2009)
X.
1.5
Burbacheretal. (2004)
Pregnant monkeys;
200 ppm; 2 hrs/day;
Shortened pregnancy
duration
2 2.5 3 3.5 4
mg Methanol/Liter Blood (mg/L)
Figure 5-4 Relationship of monkey blood levels associated with effects of uncertain
adversity with projected impact of daily peak RfC and RfD exposures on
sample background methanol blood levels (mg MeOH/Liter [mg/L] blood) in
humans.
5.4. Cancer Assessment
A cancer assessment was not conducted for this document.
5-42
-------
6.REFERENCES
ACGIH (American Conference of Governmental Industrial Hygienists). (2000). 2000 TLVs and BEIs: based on
the documentations of the threshold limit values for chemical substances and physical agents and biological
exposure indices. Cincinnati, OH.
Adanir. J: Ozkalkanti. MY: Aksun. M. (2005). Percutaneous methanol intoxication: Case report [Abstract]. Eur J
Anaesthesiol 22: 560-561.
Agarwal DP. (2001). Genetic polymorphisms of alcohol metabolizing enzymes. Pathol Biol 49: 703-709.
Akaike. H. (1973). Information theory and an extension of the maximum likelihood principle. In BN Petrov; F
Csaki (Eds.), 2nd International Symposium on Information Theory (pp. 267-281). Budapest, Hungary:
Akademiai Kiado.
Albin. RL: Greenamyre. JT. (1992). Alternative excitotoxic hypotheses [Review]. Neurology 42: 733-738.
Andrews. JE: Ebron-Mccov. M: Logsdon. TR: Mole. LM: Kavlock. RJ: Rogers. JM. (1993). Developmental
toxicity of methanol in whole embryo culture: a comparative study with mouse and rat embryos. Toxicology
81:205-215.
Andrews. JE: Ebron-Mccov. M: Kavlock. RJ: Rogers. JM. (1995). Developmental toxicity of formate and formic
acid in whole embryo culture: a comparative study with mouse and rat embryos. Teratology 51: 243-251.
http://dx.doi.org/10.1002/tera. 1420510409.
Andrews. JE: Ebron-Mccov. M: Schmid. JE: Svensgaard. D. (1998). Effects of combinations of methanol and
formic acid on rat embryos in culture. Birth Defects Res A Clin Mol Teratol 58: 54-61.
http://dx.doi.org/10.1002/(SICI)1096-9926(199808)58:2<54::AID-TERA6>:3.0.CO:2-0.
Andrews. LS: Clary. JJ: Terrill. JB: Bolte. HF. (1987). Subchronic inhalation toxicity of methanol. J Toxicol
Environ Health 20: 117-124. http://dx.doi.org/10.1080/15287398709530965.
Ang. HL: Deltour. L: Havamizu. TF: Zgombic-Knight M: Duester. G. (1996). Retinoic acid synthesis in mouse
embryos during gastrulation and craniofacial development linked to class IV alcohol dehydrogenase gene
expression. J Biol Chem 271: 9526-9534.
Antony. AC. (2007). In utero physiology: role of folic acid in nutrient delivery and fetal development. Am J Clin
Nutr 85:5988-6038.
Apaja. M. (1980). Evaluation of toxicity and carcinogenicity of malonaldehyde: An experimental study in Swiss
mice. Acta Universitatis Ouluensis, Series D, Medica 55. Finland: Anat Pathol Microbiol.
Aschner. M: Kimelberg. HK. (1996). The role of glia in neurotoxicity. In M Aschner; HK Kimelberg (Eds.).
Boca Raton, FL: CRC Press.
Aziz. MH: Agrawal. AK: Adhami. VM: Ali. MM: Baig. MA: Seth. PK. (2002). Methanol-induced neurotoxicity
in pups exposed during lactation through mother; role of folic acid. Neurotoxicol Teratol 24: 519-527.
Barceloux. DG: Bond. GR: Krenzelok. EP: Cooper. H: Vale. JA. (2002). American academy of clinical
toxicology practice guidelines on the treatment of methanol poisoning [Review]. Clin Toxicol 40: 415-446.
Batterman. SA: Franzblau. A. (1997). Time-resolved cutaneous absorption and permeation rates of methanol in
human volunteers. Int Arch Occup Environ Health 70: 341-351.
Batterman. SA: Franzblau. A: D'Arcv. JB: Sargent. NE: Gross. KB: Schreck. RM. (1998). Breath, urine, and
blood measurements as biological exposure indices of short-term inhalation exposure to methanol. Int Arch
Occup Environ Health 71: 325-335.
Beam. P: Patel J: O'Flynn. WR. (1993). Cervical ribs: A cause of distal and cerebral embolism. Postgrad Med J
69: 65-68.
6-1
-------
Bennett. IL. Jr: Gary. FH: Mitchell. GL. Jr: Cooper. MN. (1953). Acute methyl alcohol poisoning: A review
based on experiences in an outbreak of 323 cases [Review]. Medicine (Baltimore) 32: 431-463.
Berkow. R: Fletcher. AJ. (1992). The Merck manual of diagnosis and therapy. In R Berkow; AJ Fletcher (Eds.),
(16th ed.). Rahway, NJ: Merck & Co.
Black. KA: Eells. JT: Noker. PE: Hawtrev. CA: Tephly. TR. (1985). Role of hepatic tetrahydrofolate in the
species difference in methanol toxicity. PNAS 82: 3854-3858.
Bolon. B: Dorman. DC: Janszen. D: Morgan. KT: Welsch. F. (1993). Phase-specific developmental toxicity in
mice following maternal methanol inhalation. Toxicol Sci 21: 508-516.
Bolon. B: Welsch. F: Morgan. KT. (1994). Methanol-induced neural tube defects in mice: Pathogenesis during
neurulation. Teratology 49: 497-517. http://dx.doi.org/10.1002/tera.1420490610.
Bosron. WF: Li. TK. (1986). Genetic polymorphism of human liver alcohol and aldehyde dehydrogenases, and
their relationship to alcohol metabolism and alcoholism. Hepatology 6: 502-510.
Bouchard. M: Brunei RC: Droz. PO: Carrier. G. (2001). A biologically based dynamic model for predicting the
disposition of methanol and its metabolites in animals and humans. Toxicol Sci 64: 169-184.
http://dx.doi.0rg/10.1093/toxsci/64.2.169.
Branch. S: Rogers. JM: Brownie. CF: Chernoff. N. (1996). Supernumerary lumbar rib: Manifestation of basic
alteration in embryonic development of ribs. J Appl Toxicol 16: 115-119.
http://dx.doi.org/10.1002/(SICI)1099-1263(199603)16:2<:115::AID-JAT309>:3.0.CO:2-H.
Brien. JF: Clarke. DW: Richardson. B: Patrick. J. (1985). Disposition of ethanol in maternal blood, fetal blood,
and amniotic fluid of third-trimester pregnant ewes. Am J Obstet Gynecol 152: 583-590.
Bucher. JR. (2002). The National Toxicology Program rodent bioassay: Designs, interpretations, and scientific
contributions. AnnN Y Acad Sci 982: 198-207. http://dx.doi.org/10.1111/i. 1749-6632.2002.tb04934.x.
Buhler. R: Pestalozzi. D: Hess. M: Von Wartburg. JP. (1983). Immunohistochemical localization of alcohol
dehydrogenase in human kidney, endocrine organs and brain. Pharmacol Biochem Behav 18: 55-59.
http://dx.doi.org/10.1016/0091-3057(83)90147-8.
Burbacher. TM: Shen. D: Grant. K: Sheppard. L: Damian. D: Ellis. S: Liberate. N. (1999a). Reproductive and
offspring developmental effects following maternal inhalation exposure to methanol in nonhuman primates
Part I: methanol disposition and reproductive toxicity in adult females. Cambridge, MA: Health Effects
Institute.
Burbacher. TM: Grant. K: Shen. D: Damian. D: Ellis. S: Liberato. N. (1999b). Reproductive and offspring
developmental effects following maternal inhalation exposure to methanol in nonhuman primates Part II:
developmental effects in infants exposed prenatally to methanol. Cambridge, MA: Health Effects Institute.
Burbacher. TM: Shen. DP: Lalovic. B: Grant KS: Sheppard. L: Damian. D: Ellis. S: Liberato. N. (2004a).
Chronic maternal methanol inhalation in nonhuman primates (Macaca fascicularis): exposure and
toxicokinetics prior to and during pregnancy. Neurotoxicol Teratol 26: 201-221.
http://dx.doi.0rg/10.1016/i.ntt.2003.10.003.
Burbacher. TM: Grant. KS: Shen. DP: Sheppard. L: Damian. D: Ellis. S: Liberato. N. (2004b). Chronic maternal
methanol inhalation in nonhuman primates (Macaca fascicularis): reproductive performance and birth
outcome. Neurotoxicol Teratol 26: 639-650. http://dx.doi.0rg/10.1016/j.ntt.2004.06.001.
Burnell JC: Li. TK: Bosron. WF. (1989). Purification and steady-state kinetic characterization of human liver
b3b3 alcohol dehydrogenase. Biochemistry 28: 6810-6815.
Butchko. HH: Stargel WW. (2001). Aspartame: scientific evaluation in the postmarketing period. Regul Toxicol
Pharmacol 34: 221-233.
Butchko. HH: Stargel WW: Comer. CP: Mayhew. DA: Benninger. C: Blackburn. GL: De Sonneville. LMJ:
Geha. RS: Hertelendv. Z: Koestner. A: Leon. AS: Liepa. GU: Mcmartin. KE: Mendenhall CL: Munro. 1C:
Novotnv. EJ: Renwick. AG: Schiffman. SS: Schomer. PL: Shavwitz. BA: Spiers. PA: Tephly. TR: Thomas.
JA: Trefz. FK. (2002). Aspartame: Review of safety [Review]. Regul Toxicol Pharmacol 35: S1-S93.
6-2
-------
Cal/EPA (California Environmental Protection Agency). (2012). Proposition 65: Interpretive guideline no. 2012-
01: Consumption of methanol resulting from pectin that occurs naturally in fruits and vegetables.
Cameron. AM: Nilsen. OG: Haug. E: Eik-Nes. KB. (1984). Circulating concentrations of testosterone,
luteinizing hormone and follicle stimulating hormone in male rats after inhalation of methanol. Arch Toxicol
7: 441-443.
Cameron. AM: Zahlsen. K: Haug. E: Nilsen. OG: Eik-Nes. KB. (1985). Circulating steroids in male rats
following inhalation of n-alcohols. In PL Chambers; E Cholnoky; CM Chambers (Eds.), Archives of
Toxicology Supplement: Receptors and Other Tagets for Toxic Substances (pp. 422-424). Berlin: Springer-
Verlag.
Carson. BL: McCann. JL: Ellis. HV. Ill: Herndon. BL: Baker. LH. (1981). Methanol health effects [EPA
Report]. Ann Arbor, MI: U.S. Environmental Protection Agency.
Caspi R: Foerster. H: Fulcher. CA: Hopkinson. R: Ingraham. J: Kaipa. P: Krummenacker. M: Palev. S: Pick. J:
Rhee. SY: Tissier. C: Zhang. P: Karp. PP. (2006). MetaCyc: a multiorganism database of metabolic
pathways and enzymes. Nucleic Acids Res 34: D511-D516.
Chen-Tsi C. (1959). [Materials on the hygienic standardization of the maximally permissible concentration of
methanol vapors in the atmosphere]. Gig Sanit 24: 7-12.
Chernoff. N: Rogers. JM. (2004). Supernumerary ribs in developmental toxicity bioassays and in human
populations: Incidence and biological significance. J Toxicol Environ Health B Crit Rev 7: 437-449.
Chuwers. P: Osterloh. J: Kelly. T: D'Alessandro. A: Quinlan. P: Becker. C. (1995). Neurobehavioral effects of
low-level methanol vapor exposure in healthy human volunteers. Environ Res 71: 141-150.
Cichoz-Lach. H: Partvcka. J: Nesina. I: Wojcierowski. J: Slomka. M: Celinski. K. (2007). Genetic
polymorphism of alcohol dehydrogenase 3 in digestive tract alcohol damage. Hepatogastroenterology 54:
1222-1227.
Clancy. B: Finlav. BL: Darlington. RB: Anand. KJ. (2007). Extrapolating brain development from experimental
species to humans [Review]. Neurotoxicology 28: 931-937. http://dx.doi.0rg/10.1016/j.neuro.2007.01.014.
Clarke. DW: Steenaart. NAE: Brien. JF. (1986). Disposition of ethanol and activity of hepatic and placental
alcohol dehydrogenase and aldehyde dehydrogenases in the third-trimester pregnant guinea pig for single and
short-term oral ethanol administration. Alcohol ClinExp Res 10: 330-336.
Clary. JJ. (2003). Methanol, is it a developmental risk to humans? [Review]. Regul Toxicol Pharmacol 37: 83-
91.http://dx.doi.org/10.1016/S0273-2300(02)00031-4.
Coleman. CN: Mason. T: Hooker. EP: Robinson. SE. (1999). Developmental effects of intermittent prenatal
exposure to 1,1,1-trichloroethane in the rat. Neurotoxicol Teratol 21: 699-708.
http://dx.doi.org/10.1016/S0892-0362(99)00035-5.
Connell. JL: Doyle. JC: Gurry. JF. (1980). The vascular complications of cervical ribs. ANZ J Surg 50: 125-130.
http://dx.doi.org/10.1111/i. 1445-2197.198Q.tbQ6648.x.
Cook. MR: Bergman. FJ: Cohen. HP: Gerkovich. MM: Graham. C: Harris. RK: Siemann. LG. (1991). Effects of
methanol vapor on human neurobehavioral measures. (ISSN 1041-5505) Research Report Number 42).
Boston, MA: Health Effects Institute.
Cook. RJ: Champion. KM: Giometti CS. (2001). Methanol toxicity and formate oxidation in NEUT2 mice. Arch
Biochem Biophys 393: 192-198. http://dx.doi.org/10.1006/abbi.2001.2485.
Cooper. RL: Mole. ML: Rehnberg. GL: Goldman. JM: Mcelroy. WK: Hein. J: Stoker. TE. (1992). Effect of
inhaled methanol on pituitary and testicular hormones in chamber acclimated and non-acclimated rats.
Toxicology 71: 69-81.
COT (Committee on Toxicity). (2011). COT Statement on the effects of chronic dietary exposure to methanol.
http://cot.food.gov.uk/cotstatements/cotstatementsvrs/cotstatements2011/cot201102.
6-3
-------
Cronholm. T. (1987). Effect of ethanol on the redox state of the coenzyme bound to alcohol dehydrogenase
studied in isolated hepatocytes. Biochem J 248: 567-572.
Cruzan. G. (2009). Assessment of the cancer potential of methanol [Review]. Crit Rev Toxicol 39: 347-363.
http://dx.doi.org/10.1080/10408440802475199.
Gumming. ME: Ong. BY: Wade. JG: Sitar. PS. (1984). Maternal and fetal ethanol pharmacokinetics and
cardiovascular responses in near-term pregnant sheep. Can J Physiol Pharmacol 62: 1435-1439.
D'Alessandro. A: Osterloh. JD: Chuwers. P: Quinlan. PJ: Kelly. TJ: Becker. CE. (1994). Formate in serum and
urine after controlled methanol exposure at the threshold limit value. Environ Health Perspect 102: 178-181.
Davis. VE: Brown. H: Huff. JA: Cashaw. JL. (1967). The alteration of serotonin metabolism to 5-
hydroxytryptophol by ethanol ingestion in man. J Lab Clin Med 69: 132-140.
Davoli. E: Cappellini. L: Airoldi L: Fanelli R. (1986). Serum methanol concentrations in rats and in men after a
single dose of aspartame. Food Chem Toxicol 24: 187-189.
De Kevser. J: Mostert JP: Koch. MW. (2008). Dysfunctional astrocytes as key players in the pathogenesis of
central nervous system disorders [Review]. J Neurol Sci 267: 3-16.
http://dx.doi.0rg/10.1016/i.ins.2007.08.044.
Degitz. SJ: Rogers. JM: Zucker. RM: Hunter. ES. III. (2004a). Developmental toxicity of methanol:
pathogenesis in CD-I and C57BL/6J mice exposed in whole embryo culture. Birth Defects Res A Clin Mol
Teratol 70: 179-184. http://dx.doi.org/10.1002/bdra.20009.
Degitz. SJ: Zucker. RM: Kawanishi. CY: Massenburg. GS: Rogers. JM. (2004b). Pathogenesis of methanol-
induced craniofacial defects in C57BL/6J mice. Birth Defects Res A Clin Mol Teratol 70: 172-178.
http://dx.doi.org/10.1002/bdra.20010.
Deltour. L: Foglio. MH: Duester. G. (1999). Metabolic deficiencies in alcohol dehydrogenase Adhl, Adh3, and
Adh4 null mutant mice. Overlapping roles of Adhl and Adh4 in ethanol clearance and metabolism of retinol
to retinoic acid. J Biol Chem 274: 16796-16801. http://dx.doi.org/10.1074/ibc.274.24.16796.
Dicker. E: Cedebaum. AI. (1986). Inhibition of the low-Km mitachondrial aldehyde dehydrogenase by diethyl
maleate and phorone in vivo and in vitro: implications for formaldehyde metabolism. Biochem J 240: 821-
827.
Dikalova. AE: Kadiiska. MB: Mason. RP. (2001). An in vivo ESR spin-trapping study: Free radical generation
in rats from formate intoxication-role of the Fenton reaction. PNAS 98: 13549-13553.
http://dx.doi.org/10.1073/pnas.251091098.
Dorman. DC: Moss. OR: F arris. GM: Janszen. D: Bond. JA: Medinsky. MA. (1994). Pharmacokinetics of
inhaled [14C]methanol and methanol-derived [14C]formate in normal and folate-deficient cynomolgus
monkeys. Toxicol Appl Pharmacol 128: 229-238. http://dx.doi.org/10.1006/taap.1994.1202.
Dorman. DC: Bolon. B: Starve. MF: Laperle. KMD: Wong. BA: Elswick. B: Welsch. F. (1995). Role of formate
in methanol-induced exencephaly in CD-I mice. Teratology 52: 30-40.
http://dx.doi.org/10.1002/tera.1420520105.
Dorman. DC: Welsch. F. (1996). Developmental toxicity of methanol in rodents. CUT Activities 16: 40360.
Dudka. J. (2006). The total antioxidant status in the brain after ethanol or 4-methylpyrazole administration to rats
intoxicated with methanol. Exp Toxicol Pathol 57: 445-448. http://dx.doi.0rg/10.1016/i.etp.2006.01.004.
Eng. MY: Luczak. SE: Wall TL. (2007). ALDH2, ADH1B, and ADH1C genotypes in Asians: a literature
review [Review]. Alcohol Res Health 30: 22-27.
ERG (Eastern Research Group Inc.). (2009). External letter peer review of reports documenting methanol studies
in monkeys, rats and mice performed by the New Energy Development Organization (NEDO). Lexington,
MA.
Ernstgard. L: Shibata. E: Johanson. G. (2005). Uptake and disposition of inhaled methanol vapor in humans.
Toxicol Sci 88: 30-38. http://dx.doi.org/10.1093/toxsci/kfi281.
6-4
-------
Estonius. M: Svensson. S: Hoog. JO. (1996). Alcohol dehydrogenase in human tissues: Localization of
transcripts coding for five classes of the enzyme. FEES Lett 397: 338-342. http://dx.doi.org/10.1016/S0014-
5793(96)01204-5.
Evans. AL. (1999). Pseudoseizures as a complication of painful cervical ribs. Dev Med Child Neurol 41: 840-
842. http://dx.doi.org/10.1017/S0012162299001668.
Pagan. JF: Singer. LT. (1983). Infant recognition memory as a measure of intelligence. In LP Lipsitt (Ed.),
Advances in infancy research (pp. 31-78). New York, NY: Ablex.
Fallang. B: Saugstad. OP: Gregaard. J: Hadders-Algra. M. (2003). Kinematic quality of reaching movements in
preterm infants. Pediatr Res 53: 836-842. http://dx.doi.org/10.1203/01.PDR.0000058925.94994.BC.
Fernandez Noda. El: Nunez-Arguelles. J: Perez Fernandez. J: Castillo. J: Perez Izquierdo. M: Rivera Luna. H.
(1996). Neck and brain transitory vascular compression causing neurological complications, results of
surgical treatment on 1300 patients. J Cardiovasc Surg (Torino) 37: 155-166.
Fisher. JW: Dorman. DC: Medinsky. MA: Welsch. F: Conolly. RE. (2000). Analysis of respiratory exchange of
methanol in the lung of the monkey using a physiological model. Toxicol Sci 53: 185-193.
Foster. MW: Stamler. J. (2004). New insights into protein s-nitrosylation-mitochondria as a model system. J Biol
Chem 279: 25891-25897. http://dx.doi.org/10.1074/ibc.M313853200.
Frederick. LJ: Schulte. PA: Apol A. (1984). Investigation and control of occupational hazards associated with
the use of spirit duplicators. Am Ind Hyg Assoc J 45: 51-55. http://dx.doi.org/10.1080/15298668491399361.
Fu. SS: Sakanashj TM: Rogers. JM: Hong. (1996). Influence of dietary folic acid on the developmental toxicity
of methanol and the frequency of chromosomal breakage in the CD-I mouse. Reprod Toxicol 10: 455-463.
Garrett R: Grisham. CM. (2010). Biochemistry. Belmont, CA: Cengage Learning.
Gibson. MAS: Butters. NS: Reynolds. JN: Brien. JF. (2000). Effects of chronic prenatal ethanol exposure on
locomotor activity, and hippocampal weight, neurons, and nitric oxide synthase activity of the young
postnatal guinea pig. Neurotoxicol Teratol 22: 183-192.
Gonzalez-Quevado. A: Obregon FUrbina. M: Rousso. T: Lima. L. (2002). Effect of chronic methanol
administration on amino-acids and monoamines in retina, optic nerve, and brain of the rat. Toxicol Appl
Pharmacol 185: 77-84.
Guerri. C: Sanchis. R. (1985). Acetaldehyde and alcohol levels in pregnant rats and their fetuses. Alcohol 2: 267-
270.
Haffner. HT: Wehner. HP: Schevtt KD: Besserer. K. (1992). The elimination kinetics of methanol and the
influence of ethanol. Int J Legal Med 105: 111-114.
Hansen. JM: Contreras. KM: Harris. C. (2005). Methanol, formaldehyde, and sodium formate exposure in rat
and mouse conceptuses: A potential role of the visceral yolk sac in embryotoxicity. Birth Defects Res A Clin
Mol Teratol 73: 72-82. http://dx.doi.org/10.1002/bdra.20094.
Hanzlik. RP: Fowler. SC: Eells. JT. (2005). Absorption and elimination of formate following oral administration
of calcium formate in female human subjects. Drug Metab Dispos 33: 282-286.
http://dx.doi.org/10.1124/dmd.104.001289.
Harris. C: Wang. SW: Lauchu. JJ: Hansen. JM. (2003). Methanol metabolism and embryotoxicity in rat and
mouse conceptuses: Comparisons of alcohol dehydrogenase (ADH1), formaldehyde dehydrogenase (ADH3),
and catalase. Reprod Toxicol 17: 349-357. http://dx.doi.org/10.1016/S0890-6238(03)00013-3.
Harris. C: Dixon. M: Hansen. JM. (2004). Glutathione depletion modulates methanol, formaldehyde and formate
toxicity in cultured rat conceptuses. Cell Biol Toxicol 20: 133-145.
http://dx.doi.Org/10.1023/B:CBTO.0000029466.08607.86.
6-5
-------
Hashimoto. M. (2008). Report letter from Masahiro Hashimoto, Director General of the Policy Planning and
Coordination Department of NEDO, to John Lynn, CEO of Methanol Institute, certifying the English
translations of the 1985 NEDO (New Energy and Industrial Technology Development Organization) original
Japanese methanol test reports are accurate and complete. Kawasaki City, Japan: New Energy and Industrial
Technology Development Organization.
Hass. U: Lund. SP: Simonsen. L: Fries. AS. (1995). Effects of prenatal exposure to xylene on postnatal
development and behavior in rats. Neurotoxicol Teratol 17: 341-349. http://dx.doi.org/10.1016/0892-
0362(94)00093-8.
Havasaka. Y: Havasaka. S: Nagakj Y. (2001). Ocular changes after intravitreal injection of methanol,
formaldehyde, or formate in rabbits. Pharmacol Toxicol 89: 74-78. http://dx.doi.org/10.1034/U600-
0773.2001.d01-138.x.
Hedberg. JJ: Backhand. M: Stromberg. P: Lonn. S: Dahl. ML: Ingelman-Sundberg. M: Hoog. JO. (2001).
Functional polymorphism in the alcohol dehy drogenase 3 (ADH3) promoter. Pharmacogenetics 11:815-824.
HEI (Health Effects Institute). (1987). Automotive methanol vapors and human health: An evaluation of existing
scientific information and issues for future research. Boston, MA.
Henderson. MS. (1914). Cervical rib: Report of thirty-one cases. J Bone Joint Surg Am 11: 408-430.
Hess. DT: Matsumoto. A: Kim. SO: Marshall HE: Stamler. JS. (2005). Protein S-nitrosylation: Purview and
parameters [Review]. Nat Rev Mol Cell Biol 6: 150-166. http://dx.doi.org/10.1038/nrml569.
Hines. RN: McCarver. DG. (2002). The ontogeny of human drug-metabolizing enzymes: Phase I oxidative
enzymes [Review]. J Pharmacol Exp Ther 300: 355-360. http://dx.doi.0rg/10.1124/ipet.300.2.355
Hines. RN: Mccarver. DG. (2002). The ontogeny of human drug-metabolizing enzymes: Phase I oxidative
enzymes [Review]. J Pharmacol Exp Ther 300: 355-360. http://dx.doi.0rg/10.1124/ipet.300.2.355.
Horton. VL: Higuchi. MA: Rickert. DE. (1992). Physiologically based pharmacokinetic model for methanol in
rats, monkeys, and humans. Toxicol Appl Pharmacol 117: 26-36.
HSDB (Hazardous Substances Data Bank). (2009). Methanol: Human health effects [Database]. Bethesda, MD:
National Library of Medicine.
Huang. OF: Gebrewold. A: Zhang. A: Altura. BT: Altura. BM. (1994). Role of excitatory amino acids in
regulation of rat pial microvasculature. Am J Physiol 266: R158-R163.
Infurna. R: Weiss. B. (1986). Neonatal behavioral toxicity in rats following prenatal exposure to methanol.
Teratology 33: 259-265. http://dx.doi.org/10.1002/tera.1420330302.
IPCS (International Programme on Chemical Safety). (1997). Methanol (Vol. 196). Geneva, Switzerland: World
Health Organization, http://www.inchem.org/documents/ehc/ehc/ehcl96.htm.
Ivvaswamy. A: Rathinasamy. S. (2012). Effect of chronic exposure to aspartame on oxidative stress in the brain
of albino rats. JBiosci 37: 679-688. http://dx.doi.org/10.1007/sl2038-012-9236-0.
Jelski. W: Chrostek. L: Markiewicz. W: Smitkowski. M. (2006). Activity of alcohol dehy drogenase (ADH)
isoenzymes and aldehyde dehy drogenase (ALDH) in the sera of patients with breast cancer. J ClinLab Anal
20: 105-108. http://dx.doi.org/10.1002/jcla.20109.
Johlin. FC: Fortman. CS: Nghiem. DP: Tephly. TR. (1987). Studies on the role of folic acid and folate-
dependent enzymes in human methanol poisoning. Mol Pharmacol 31: 557-561.
Johnson. NL: Kotz. S. (1970). Continuous univariate distributions: Distributions in statistics. New York, NY:
John Wiley & Sons.
Kavet R: Nauss. KM. (1990). The toxicity of inhaled methanol vapors [Review]. Crit Rev Toxicol 21: 21-50.
http://dx.doi.org/10.3109/10408449009089872.
Kavlock. RJ: Allen. BC: Faustman. EM: Kimmel CA. (1995). Dose-response assessments for developmental
toxicity. IV. Benchmark doses for fetal weight changes. Toxicol Sci 26: 211-222.
6-6
-------
Kawai T: Yasugi. T: Mizunuma. K: Horiguchi. S: Hirase. Y: Uchida. Y: Ikeda. M. (1991). Methanol in urine as
a biological indicator of occupational exposure to methanol vapor. Int Arch Occup Environ Health 63:311-
318.
Kerns. W: Tomaszewskj C: Mcmartin. K: Ford. M: Brent J. (2002). Formate kinetics in methanol poisoning.
Clin Toxicol 40: 137-143.
Kim. SW: Jang. YJ: Chang. JW: Hwang. O. (2003). Degeneration of the nigrostriatal pathway and induction of
motor deficit by tetrahydrobiopterin: An in vivo model relevant to Parkinson's disease. Neurobiol Dis 13:
167-176.
Kraut JA: Kurtz. I. (2008). Toxic alcohol ingestions: Clinical features, diagnosis, and management [Review].
Clin J Am Soc Nephrol 3: 208-225. http://dx.doi.org/10.2215/CJN.03220807.
Lee. E: Brady. AN: Brabec. MJ: Fabel T. (1991). Effects of methanol vapors on testosterone production and
testis morphology in rats. Toxicol Ind Health 7: 261-275.
Lee. EW: Terzo. TS: D'Arcv. JB: Gross. KB: Schreck. RM. (1992). Lack of blood formate accumulation in
humans following exposure to methanol vapor at the current permissible exposure limit of 200 ppm. Am Ind
Hyg Assoc J 53: 99-104. http://dx.doi.org/10.1080/15298669291359357.
Lee. EW: Garner. CD: Terzo. TS. (1994). Animal model for the study of methanol toxicity: Comparison of
folate-reduced rat responses with published monkey data. J Toxicol Environ Health 41: 71-82.
http://dx.doi.org/10.1080/15287399409531827.
Lewis. RJ. Sr. (1992). Sax's dangerous properties of industrial materials: v III (8th ed.). New York, NY: Van
Nostrand Reinhold.
Lindinger. W: Taucher. J: Jordan. A: Hansel A: Vogel. W. (1997). Endogenous production of methanol after the
consumption of fruit. Alcohol Clin Exp Res 21: 939-943. http://dx.doi.org/10.1111/i. 1530-
0277.1997.tb03862.x.
Lorente. C: Cordier. S: Bergeret A: De Walle. HEK: Goujard. J: Ayme. S: Knill-Jones. R: Calzolari. E: Bianchi.
E (2000). Maternal occupational risk factors for oral clefts. Scand J Work Environ Health 26: 137-145.
Lu. K: Gul H: Upton. PB: Moeller. BC: Swenberg. JA. (2012). Formation of hydroxymethyl DNA adducts in
rats orally exposed to stable isotope labeled methanol. Toxicol Sci 126: 28-38.
http://dx.doi.org/10.1093/toxsci/kfr328.
Magnuson. BA: Burdock. GA: Doull J: Kroes. RM: Marsh. GM: Pariza. MW: Spencer. PS: Waddell WJ:
Walker. R: Williams. GM. (2007). Aspartame: A safety evaluation based on current use levels, regulations,
and lexicological and epidemiological studies [Review]. Crit Rev Toxicol 37: 629-727'.
http://dx.doi.org/10.1080/10408440701516184.
Mann. WJ: Muttrav. A: Schaefer. D: Klimek. L: Faas. M: Konietzko. J. (2002). Exposure to 200 ppm of
methanol increases the concentrations of interleukin-lbeta and interleukin-8 in nasal secretions of healthy
volunteers. Ann Otol Rhinol Laryngol 111: 633-638.
Mannering. GJ: VanHarkea PR: Makar. AB: Tephly. TR: Watkins. WD: Goodman. JI. (1969). Role of the
intracellular distribution of hepatic catalase in the peroxidative oxidation of methanol. Ann N Y Acad Sci
168: 265-280.
Mason. RP: Kadiska. MB. (2003). Ex vivo detection of free radical metabolites of toxic chemicals and drugs by
spin trapping. In LJ Berliner (Ed.), In vivo EPR (ESR): Theory and applications (pp. 309-324). New York,
NY: Kluwer Academic/Plenum Publishers.
McCallum. GP: Siu. M: Ondovcik. SL: Sweeting. JN: Wells. PG. (201 la). Methanol exposure does not lead to
accumulation of oxidative DNA damage in bone marrow and spleen of mice, rabbits or primates. Mol
Carcinog 50: 163-172. http://dx.doi.org/10.1002/mc.20701.
McCallum. GP: Siu. M: Sweeting. JN: Wells. PG. (20 lib). Methanol exposure does not produce oxidatively
damaged DNA in lung, liver or kidney of adult mice, rabbits or primates. Toxicol Appl Pharmacol 250: 147-
153. http://dx.doi.0rg/10.1016/i.taap.2010.10.004.
6-7
-------
McMartin. KE: Martin-Amat G: Noker. PE: Tephly. TR. (1979). Lack of a role for formaldehyde in methanol
poisoning in the monkey. Biochem Pharmacol 28: 645-649. http://dx.doi.org/10.1016/0006-2952(79)90149-7
Medinsky. MA: Dorman. DC: Bond. JA: Moss. OR: Janszen. DB: Everitt JI. (1997). Pharmacokinetics of
methanol and formate in female cynomolgus monkeys exposed to methanol vapors. (HEI-RFA-89-1).
Boston, MA: Health Effects Institute.
Meister. A: Anderson. ME. (1983). Glutathione [Review]. Annu Rev Biochem 52: 711-760.
http://dx.doi.org/10.1146/annurev.bi.52.070183.003431.
Methanol Institute. (2009a). Biodiesel: A growing market for methanol. Arlington, VA.
Methanol Institute. (2009b). Frequently asked questions. Arlington, VA: Methanol Insitute.
http://www.methanol.org/Health-And-Safetv/Frequentlv-Asked-Questions.aspx.
Miller. L: Wells. PG. (2011). Altered methanol embryopathies in embryo culture with mutant catalase-deficient
mice and transgenic mice expressing human catalase. Toxicol Appl Pharmacol 252: 55-61.
http://dx.doi.0rg/10.1016/i.taap.2011.01.019.
Montserrat CA: Field. MS: Perry. C: Ghandour. H: Chiang. E: Selhub. J: Shane. B: Stover. PJ. (2006).
Regulation of folate-mediated one-carbon metabolism by 10-formyltetrahydrofolate dehydrogenase. JBiol
Chem281: 18335-18342.
Motavkin. PA: Okhotin. VE: Konovko. OO: Zimatkin. SM. (1988). Localization of alcohol and aldehyde
dehydrogenase in the human spinal cord and brain. Neurosci Behav Physiol 20: 79-84.
http://dx.doi.org/10.1007/BF01268118.
Muthuvel A: Raiamani. R: Manikandan. S: Sheeladevi R. (2006a). Detoxification of formate by formate
dehydrogenase-loaded erythrocytes and carbicarb in folate-deficient methanol-intoxicated rats. Clin Chirn
Acta 367: 162-169. http://dx.doi.0rg/10.1016/i.cca.2005.12.007.
Muthuvel A: Raiamani. R: Senthilvelan. M: Manikandan. S: Sheeladevi. R. (2006b). Modification of
allergenicity and immunogenicity of formate dehydrogenase by conjugation with linear mono methoxy poly
ethylene glycol: Improvement in detoxification of formate in methanol poisoning. Clin Chim Acta 374: 122-
128. http://dx.doi.0rg/10.1016/i.cca.2006.06.003.
NEDO (New Energy Development Organization). (1985a). Test report: 18-month inhalation carcinogenicity
study on methanol inB6C3Fl mice (test no. 4A-223). Tokyo, Japan: Mitsubishi Kasei Institute of
Toxicology and Environmental Sciences.
NEDO (New Energy Development Organization). (1985b). Test report: 24-month inhalation carcinogenicity
study on methanol in Fischer rats (Test No. 5A-268). Toyko, Japan: Mitsubishi Kasei Institute of Toxicology
and Environmental Sciences.
NEDO (New Energy Development Organization). (1987). Toxicological research of methanol as a fuel for
power station: summary report on tests with monkeys, rats and mice. Tokyo, Japan.
Nelson. BK: Brightwell WS: Mackenzie. PR: Khan. A: Burg. J. R.: Weigel WW: Goad. PT. (1985).
Teratological assessment of methanol and ethanol at high inhalation levels in rats. Toxicol Sci 5: 727-736.
Nguyen. T: Baumgartner. F: Nelems. B. (1997). Bilateral rudimentary first ribs as a cause of thoracic outlet
syndrome. J Natl Med Assoc 89: 69-73.
NRC (National Research Council). (1983). Risk assessment in the federal government: Managing the process.
Washington, DC: National Academies Press, http://www.nap.edu/openbook.php7record id=366&page=Rl.
NRC (National Research Council). (2011). Review of the Environmental Protection Agency's draft IRIS
assessment of formaldehyde. Washington, DC: National Academies Press.
http://www.nap.edu/catalog/13142.html.
NTP-CERHR (NTP Center for the Evaluation of Risks to Human Reproduction). (2003). NTP-CERHR
monograph on the potential human reproductive and developmental effects of methanol.
6-8
-------
NTP-CERHR (NTP Center for the Evaluation of Risks to Human Reproduction). (2004). NTP-CERHR expert
panel report on the reproductive and developmental toxicity of methanol [Review]. Reprod Toxicol 18: 303-
390. http://dx.doi.0rg/10.1016/i.reprotox.2003.10.013.
O'Callaghan. JP: Sriram. K. (2005). Glial fibrillary acidic protein and related glial proteins as biomarkers of
neurotoxicity [Review]. Expert Opin Drug Saf 4: 433-442. http://dx.doi.0rg/10.1517/14740338.4.3.433
Osterloh. JD: D'Alessandro. A: Chuwers. P: Mogadeddi H: Kelly. TJ. (1996). Serum concentrations of methanol
after inhalation at 200 ppm. J Occup Environ Med 38: 571-576.
Parthasarathv. NJ: Kumar. RS: Devi. RS. (2005a). Effect of methanol intoxication on rat neutrophil functions. J
Immunotoxicol 2: 115-211. http://dx.doi.org/10.1080/15476910500187425.
Parthasarathv. NJ: Kumar. RS: Karthikevan. P: SheelaDevi R. (2005b). In vitro and in vivo study of neutrophil
functions after acute methanol intoxication in albino rats. Toxicol Environ Chem 87: 559-566.
http://dx.doi.org/10.1080/02772240500382308.
Parthasarathv. NJ: Kumar. RS: Manikandan. S: Narayanan. GS: Kumar. RV: Devi. RS. (2006a). Effect of
methanol-induced oxidative stress on the neuroimmune system of experimental rats. Chem Biol Interact 161:
14-25. http://dx.doi.0rg/10.1016/i.cbi.2006.02.005.
Parthasarathv. NJ: Kumar. RS: Manikandan. S: Devi. RS. (2006b). Methanol-induced oxidative stress in rat
lymphoid organs. J Occup Health 48: 20-27. http://dx.doi.org/10.1539/ioh.48.20.
Parthasarathv. NJ: Srikumar. R: Manikandan. S: Narayanan. GS: Devi. RS. (2007). Effect of methanol
intoxication on specific immune functions of albino rats. Cell Biol Toxicol 23: 177-187.
http://dx.doi.org/10.1007/sl0565-006-0151-8.
Perkins. RA: Ward. KW: Pollack. GM. (1995a). Comparative toxicokinetics of inhaled methanol in the female
CD-I mouse and Sprague-Dawley rat. Toxicol Sci 28: 245-254.
Perkins. RA: Ward. KW: Pollack. GM. (1995b). A pharmacokinetic model of inhaled methanol in humans and
comparison to methanol disposition in mice and rats. Environ Health Perspect 103: 726-733.
Perkins. RA: Ward. KW: Pollack. GM. (1996). Methanol inhalation: site and other factors influencing
absorption, and an inhalation toxicokinetic model for the rat. Pharm Res 13: 749-755.
http://dx.doi.Org/10.1023/A:1016055701736.
Pietruszko. R. (1980). Alcohol and aldehyde dehydrogenase isozymes from mammalian liver-their structural
and functional differences. Isozymes Current Top Biology Med Res 4: 107-130.
Pikkarainen. PH: Raiha. NCR. (1967). Development of alcohol dehydrogenase activity in the human liver.
PediatrRes 1: 165-168. http://dx.doi.org/10.1203/00006450-196705000-00001.
Pitkin. RM. (2007). Folate and neural tube defects [Review]. Am J Clin Nutr 85: 285S-288S.
Plantinga. Y: Perdock. J: de Groot L. (1997). Hand function in low-risk preterm infants: Its relation to muscle
power regulation. Dev Med Child Neurol 39:6-11.
Pollack. GM: Kawagoe. JL. (1991). Determination of methanol in whole blood by capillary gas chromatography
with direct on-column injection. J Chromatogr A 570: 406-411. http://dx.doi.org/10.1016/0378-
4347(91)80546-0.
Pollack. GM: Brouwer. KLR: Kawagoe. JL. (1993). Toxicokinetics of intravenous methanol in the female rat.
Toxicol Sci 21: 105-110. http://dx.doi.0rg/10.1093/toxsci/21.l.105.
Pollack. GM: Brouwer. KLR. (1996). Maternal-fetal pharmacokinetics of methanol (Vol. NTIS). (74). Boston,
MA: Health Effects Institute.
Poon. R: Chu. I: Bjarnason. S: Potvin. M: Vincent. R: Miller. RB: Valli VE. (1994). Inhalation toxicity study of
methanol, toluene, and methanol/toluene mixtures in rats: effects of 28-day exposure. Toxicol Ind Health 10:
231-245.
Poon. R: Chu. I: Bjarnason. S: Vincent. R: Potvin. M: Miller. RB: Valli. VE. (1995). Short-term inhalation
toxicity of methanol, gasoline, and methanol/gasoline in the rat. Toxicol Ind Health 11: 343-361.
6-9
-------
Raiamani. R: Muthuvel A: Senthilvelen. M: Sheeladevi R. (2006). Oxidative stress induced by methotrexate
alone and in the presence of methanol in discrete regions of the rodent brain, retina and optic nerve. Toxicol
Lett 165: 265-273. http://dx.doi.0rg/10.1016/j.toxlet.2006.05.005.
Ramsey. JC: Andersen. ME. (1984). A physiologically based description of the inhalation pharmacokinetics of
styrene in rats and humans. Toxicol Appl Pharmacol 73: 159-175. http://dx.doi.org/10.1016/0041-
008X(84)90064-4.
Rice. D: Barone. S. (2000). Critical periods of vulnerability for the developing nervous system: Evidence from
humans and animal models [Review]. Environ Health Perspect 108: 511-533.
http://dx.doi.org/10.1289/ehp.00108s3511.
Rogers. JM: Barbee. BD: Rehnberg. BF. (1993a). Critical periods of sensitivity for the developmental toxicity of
inhaled methanol [Abstract]. Teratology 47: 395.
Rogers. JM: Mole. ML: Chernoff. N: Barbee. BD: Turner. CI: Logsdon. TR: Kavlock. RJ. (1993b). The
developmental toxicity of inhaled methanol in the CD-I mouse, with quantitative dose-response modeling for
estimation of benchmark doses. Teratology 47: 175-188. http://dx.doi.org/10.1002/tera.1420470302.
Rogers. JM: Mole. ML. (1997). Critical periods of sensitivity to the developmental toxicity of inhaled methanol
in the CD-I mouse. Teratology 55: 364-372. http://dx.doi.org/10.1002/(SICI) 1096-
9926(199706)55:6<:364::AID-TERA2>:3.0.CO:2-Y.
Rogers. JM: Brannen. KC: Barbee. BD: Zucker. RM: Degitz. SJ. (2004). Methanol exposure during gastrulation
causes holoprosencephaly, facial dysgenesis, and cervical vertebral malformations in C57BL/6J mice. Birth
Defects Res B Dev Reprod Toxicol 71: 80-88. http://dx.doi.org/10.1002/bdrb.200Q3.
Rogers. W: Wickstrom. M: Liber. K: MacKinnon. MD. (2002). Acute and subchronic mammalian toxicity of
naphthalenic acids from oil sands tailings. Toxicol Sci 66: 347-355.
Rubinstein. D: Escott E: Kelly. JP. (1995). Methanol intoxication with putaminal and white matter necrosis: MR
and CT findings. AJNR Am J Neuroradiol 16: 1492-1494.
Sakanashi. TM: Rogers. JM: Fu. SS: Connelly. LE: Keen. CL. (1996). Influence of maternal folate status on the
developmental toxicity of methanol in the CD-I mouse. Teratology 54: 198-206.
http://dx.doi.org/10.1002/(SICI)1096-9926(199610)54:4<l98::AID-TERA4>:3.0.CO:2-Y.
Salzman. M. (2006). Methanol neurotoxicity. Clin Toxicol 44: 89-90.
Sarkola. T: Eriksson. CJP. (2001). Effect of 4-methylpyrazole on endogenous plasma ethanol and methanol
levels in humans. Alcohol Clin Exp Res 25: 513-516.
Saxton. EH: Miller. TQ: Collins. JD. (1999). Migraine complicated by brachial plexopathy as displayed by MRI
and MRA: Aberrant subclavian artery and cervical ribs. J Natl Med Assoc 91: 333-341.
Savers. RR: Yant WP: Schrenk. HH: Chornyak. J: Pearce. SJ: Patty. FA: Linn. JG. (1944). Methanol poisoning
II Exposure of dogs for brief periods eight times daily to high concentrations of methanol vapor in air. J Ind
Hyg Toxicol 26: 255-259.
Schmutte. P: Bilzer. N: Penners. BM. (1988). Zur nuchternkinetik der begleitalkohole methanol und propanol-1.
Blutalkohol25: 137-142.
Schumacher. R: Mai. A: Gutjahr. P. (1992). Association of rib abnomalies and malignancy in childhood. Eur J
Pediatr 151: 432-434. http://dx.doi.org/10.1007/BF01959357.
Sedivec. V: Mraz. M: Flek. J. (1981). Biological monitoring of persons exposed to methanol vapours. Int Arch
Occup Environ Health 48: 257-271.
Short. DW. (1975). The subclavian artery in 16 patients with complete cervical ribs. J Cardiovasc Surg (Torino)
16: 135-141.
Simintzi. I: Schulpis. KH: Angelogianni. P: Liapi C: Tsakiris. S. (2007). The effect of aspartame metabolites on
the suckling rat frontal cortex acetylcholinesterase An in vitro study. Food Chem Toxicol 45: 2397-2401.
http://dx.doi.0rg/10.1016/i.fct.2007.06.016.
6-10
-------
Siragusa. RJ: Cerda. JJ: Baig. MM: Burgin. CW: Robbins. FL. (1988). Methanol production from the
degradation of pectin by human colonic bacteria. Am J ClinNutr47: 848-851.
Siu. MT: Wiley. MJ: Wells. PG. (2013). Methanol teratogenicity in mutant mice with deficient catalase activity
and transgenic mice expressing human catalase. Reprod Toxicol 36: 33-39.
http://dx.doi.0rg/10.1016/i.reprotox.2012.ll.006.
Skrzydlewska. E: Elas. M: Farbiszewskj R: Roszkowska. A. (2000). Effect of methanol intoxication on free-
radical induced protein oxidation. J Appl Toxicol 20: 239-243. http://dx.doi.org/10.1002/(SICI) 1099-
1263(200005/06)20:3<239::AID-JAT654>3.0.CO:2-2.
Skrzydlewska. E: Elas. M: Ostrowska. J. (2005). Protective effects of N-acetylcysteine and vitamin E derivative
U83836E on proteins modifications induced by methanol intoxication. Toxicol MechMeth 15: 263-270.
http://dx.doi.org/10.1080/15376520590968815.
Smith. M: Hopkinson. DA: Harris. H. (1971). Developmental changes and polymorphism in human alcohol
dehydrogenase. Ann Hum Genet 34: 251-271.
Smith. ME: Newman. HW. (1959). The rate of ethanol metabolism in fed and fasting animals. J Biol Chem 234:
1544-1549.
Soffritti. M: Belpoggi F: Cevolani. D: Guarino. M: Padovani. M: Maltoni. C. (2002). Results of long-term
experimental studies on the carcinogenicity of methyl alcohol and ethyl alcohol in rats. In MA Mehlman
(Ed.), Carcinogenesis bioassays and protecting public health: commemorating the lifework of Cesare Maltoni
and colleaques (pp. 46-69). Bologna, Italy: Ann. N. Y. Acad. Sci.
Sofroniew MV. Vinters HV. 2010. Astrocytes: biology and pathology. Acta Neuropathol 119(l):7-35.
SSDAN CensusScope (Social Science Data Analysis Network, CensusScope). (2010). United States: Population
by race [Database]. Ann Arbor, Michigan. Retrieved from http://www.censusscope.org/us/chart_race.html
Staab. CA: Lander. J: Brandt. M: Lengqvist J: Morgenstern. R: Grafstrom. RC: Hoog. JO. (2008). Reduction of
S-nitrosoglutathione by alcohol dehydrogenase 3 is facilitated by substrate alcohols via direct cofactor
recycling and leads to GSH-controlled formation of glutathione transferase inhibitors. Biochem J 413:
493504. http://dx.doi.org/10.1042/BJ20071666.
Stanton. ME: Crofton. KM: Gray. LE: Gordon. CJ: Boyes. WK: Mole. ML: Peele. DB: Bushnell PJ. (1995).
Assessment of offspring development and behavior following gestational exposure to inhaled methanol in the
rat. Toxicol Sci 28: 100-110.
Stegink. LD: Brummel MC: Mcmartin. K: Martin-Amat G: Filer. LJ. Jr: Baker. GL: Tephly. TR. (1981). Blood
methanol concentrations in normal adult subjects administered abuse doses of aspartame. J Toxicol Environ
Health 7: 281-290. http://dx.doi.org/10.1080/15287398109529979.
Stegink. LD: Brummel MC: Filer. LJ. Jr: Baker. GL. (1983). Blood methanol concentrations in one-year-old
infants administered graded doses of aspartame. JNutr 113: 1600-1606.
Stegink. LD: Filer. LJ: Bell EF: Ziegler. EE: Tephly. TR. (1989). Effect of repeated ingestion of aspartame-
sweetened beverage on plasma amino acid, blood methanol, and blood formate concentrations in normal
adults. Metabolism 38: 357-363. http://dx.doi.org/10.1016/0026-0495(89)90125-X.
Stern. S: Reuhl. K: Soderholm. S: Cox. C: Sharma. A: Balys. M: Gelein. R: Yin. C: Weiss. B. (1996). Perinatal
methanol exposure in the rat I Blood methanol concentration and neural cell adhesion molecules. Toxicol Sci
34: 36-46.
Sultatos. LG: Pasting. GM: Rosenfeld. CA: Flynn. EJ. (2004). Incorporation of the genetic control of alcohol
dehydrogenase into a physiologically based pharmacokinetic model for ethanol in humans. Toxicol Sci 78:
20-31. http://dx.doi.org/10.1093/toxsci/kfh057.
Svensson. S: Some. M: Lundsjo. A: Helander. A: Cronholm. T: Hoog. JO. (1999). Activities of human alcohol
dehydrogenases in the metabolic pathways of ethanol and serotonin. Eur J Biochem 262: 324-329.
6-11
-------
Sweeting. JN: Siu. M: Mccallum. GP: Miller. L: Wells. PG. (2010). Species differences in methanol and formic
acid pharmacokinetics in mice, rabbits and primates. Toxicol Appl Pharmacol 247: 28-35.
http://dx.doi.0rg/10.1016/i.taap.2010.05.009.
Sweeting. JN: Siu. M: Wiley. MJ: Wells. PG. (2011). Species- and strain-dependent teratogenicity of methanol
in rabbits and mice. Reprod Toxicol 31: 50-58. http://dx.doi.0rg/10.1016/j.reprotox.2010.09.014.
Tanner. CM. (1992). Occupational and environmental causes of parkinsonism [Review]. Occup Med (Lond) 7:
503-513.
Teng. S: Beard. K: Pourahmad. J: Moridanj M: Easson. E: Poon. R: O'Brien. PJ. (2001). The formaldehyde
metabolic detoxification enzyme systems and molecular cytotoxic mechanism in isolated rat hepatocytes.
ChemBiol Interact 130-132: 285-296. http://dx.doi.org/10.1016/80009-2797(00)00272-6.
Tephlv. TR: Mcmartin. KE. (1984). Methanol metabolism and toxicity. Food Sci Technol Bull Func Food 12:
111-140.
Thrasher. JD: Kilburn. KH. (2001). Embryo toxicity and teratogenicity of formaldehyde [Review]. Arch Environ
Health 56: 300-311. http://dx.doi.org/10.1080/00039890109604460.
Toth. BA: Wallcave. L: Patil K: Schmeltz. I: Hoffmann. D. (1977). Induction of tumors in mice with the
herbicide succinic acid 2,2-dimethylhydrazide. Cancer Res 37: 3497-3500.
Tran. MN: Wu. AH: Hill. DW. (2007). Alcohol dehydrogenase and catalase content in perinatal infant and adult
livers: potential influence on neonatal alcohol metabolism. Toxicol Lett 169: 245-252.
http://dx.doi.0rg/10.1016/i.toxlet.2007.01.012.
TRL (Toxicity Research Laboratories). (1986). Rat oral subchronic toxicity study with methanol. (TRL No. 032-
005). Muskegon, MI: Research Triangle Institute.
Tsakiris. S: Giannoulia-Karantana. A: Simintzj I: Schulpis. KH. (2006). The effect of aspartame metabolites on
human erythrocyte membrane acetylcholinesterase activity. Pharmacol Res 53: 1-5.
http://dx.doi.0rg/10.1016/i.phrs.2005.07.006.
Turner. C: Spanel P: Smith. D. (2006). A longitudinal study of methanol in the exhaled breath of 30 healthy
volunteers using selected ion flow tube mass spectrometry, SIFT-MS. Physiol Meas 27: 637-648.
http://dx.doi.Org/10.1088/0967-3334/27/7/007.
U.S. Congress. (2011). Consolidated Appropriations Act, 2012. (Pub. L. No. 112-74; 125 STAT. 786). 112th
U.S. Congress. http://www.gpo.gov/fdsYS/pkg/PLAW-l 12publ74/pdf/PLAW-l 12publ74.pdf
U.S. EPA (U.S. Environmental Protection Agency). (1986a). Guidelines for mutagenicity risk assessment [EPA
Report]. (EPA/630/R-98/003). Washington, DC. http://www.epa.gov/iris/backgrd.html
U.S. EPA (U.S. Environmental Protection Agency). (1986b). Guidelines for the health risk assessment of
chemical mixtures [EPA Report]. (EPA/630/R-98/002). Washington, DC.
http ://cfpub. epa. gov/ncea/cfm/recordisplav. cfm?deid=22567
U.S. EPA (U.S. Environmental Protection Agency). (1988). Recommendations for and documentation of
biological values for use in risk assessment [EPA Report]. (EPA/600/6-87/008). Cincinnati, OH.
http ://cfpub. epa. gov/ncea/cfm/recordisplav. cfm?deid=3 485 5.
U.S. EPA (U.S. Environmental Protection Agency). (1991). Guidelines for developmental toxicity risk
assessment [EPA Report]. (EPA/600/FR-91/001). Washington, DC: U.S. Environmental Protection Agency,
Risk Assessment Forum, http://www.epa.gov/raf/publications/guidelines-dev-toxicitv-risk-assessment.htm.
U.S. EPA (U.S. Environmental Protection Agency). (1994a). Interim policy for particle size and limit
concentration issues in inhalation toxicity studies [EPA Report]. Washington, DC.
http ://cfpub. epa. gov/ncea/cfm/recordisplav. cfm?deid= 186068.
U.S. EPA (U.S. Environmental Protection Agency). (1994b). Methods for derivation of inhalation reference
concentrations and application of inhalation dosimetry [EPA Report]. (EPA/600/8-90/066F). Research
Triangle Park, NC. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=71993.
6-12
-------
U.S. EPA (U.S. Environmental Protection Agency). (1995). The use of the benchmark dose approach in health
risk assessment [EPA Report]. (EPA/630/R-94/007). Washington, DC.
http://www.epa.gov/raf/publications/useof-bda-healthrisk.htm.
U.S. EPA (U.S. Environmental Protection Agency). (1996). Guidelines for reproductive toxicity risk assessment
[EPA Report]. (EPA/630/R-96/009). Washington, DC.
http://www.epa.gov/raf/publications/pdfs/REPRO51 .PDF.
U.S. EPA (U.S. Environmental Protection Agency). (1998a). Guidelines for neurotoxicity risk assessment [EPA
Report]. (EPA/630/R-95/001F). Washington, DC.
http://www.epa.gov/raf/publications/pdfs/NEUROTOX.PDF.
U.S. EPA (U.S. Environmental Protection Agency). (1998b). Health effects test guidelines OPPTS 870.6300
developmental neurotoxicity study [EPA Report]. (EPA 712C98239). Washington, DC: U.S. Environmentall
Protection Agency, http://www.regulations.gov/#!documentDetail:D=EPA-HQ-OPPT-2009-0156-0042.
U.S. EPA (U.S. Environmental Protection Agency). (2000a). Science policy council handbook: Risk
characterization [EPA Report]. (EPA 100-B-00-002). Washington, D.C.: Office of Science Policy, Office of
Research and Development, http://www.epa.gov/osa/spc/pdfs/rchandbk.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (2000b). Supplementary guidance for conducting health risk
assessment of chemical mixtures [EPA Report]. (EPA/630/R-00/002). Washington, DC.
http ://cfpub. epa. gov/ncea/cfm/recordisplav. cfm?deid=205 3 3.
U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose and reference
concentration processes [EPA Report]. (EPA/630/P-02/002F). Washington, DC: Risk Assessment Forum,
U.S. Environmental Protection Agency. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=51717.
U.S. EPA (U.S. Environmental Protection Agency). (2005a). Guidelines for carcinogen risk assessment [EPA
Report]. (EPA/630/P-03/001F). Washington, DC: Risk Assessment Forum.
http://www.epa.gov/cancerguidelines/
U.S. EPA (U.S. Environmental Protection Agency). (2005b). Supplemental guidance for assessing susceptibility
from early-life exposure to carcinogens [EPA Report] (pp. 1125-1133). (EPA/630/R-03/003F). Washington,
DC. http://www.epa.gov/cancerguidelines/guidelines-carcinogen-supplement.htm
U.S. EPA (U.S. Environmental Protection Agency). (2006a). A framework for assessing health risk of
environmental exposures to children [EPA Report]. (EPA/600/R-05/093F). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=158363.
U.S. EPA (U.S. Environmental Protection Agency). (2006b). Peer review handbook (3rd edition) [EPA Report].
(EPA/100/B-06/002). Washington, DC. http://www.epa.gov/peerreview/.
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Section E - Quantities of TRI chemicals in waste,
2008 Retrieved December 18, 2009, from http://www.epa.gov/tri/NationalAnalysis/archive/index.html
U.S. EPA (U.S. Environmental Protection Agency). (2009b). The Toxics Release Inventory (TRI) and factors to
consider when using TRI data Retrieved December 18, 2009, from
http://www.epa.gov/tri/triprogram/FactorsToConPDF.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2009c). TRI on-site and off-site reported disposed of or
otherwise released (in pounds), for facilities in all industries, for hazardous air pollutant chemicals, U.S.,
2008 (Version 5.1) [Database]. Washington, DC. Retrieved from http://www.epa.gov/triexplorer
U.S. EPA (U.S. Environmental Protection Agency). (201 la). Benchmark Dose Software (BMDS) (Version 2.2
R65 [Build: 04/13/2011]) [Computer Program]. Research Triangle Park, NC: National Center for
Environmental Assessment. Retrieved from http://www.epa. gov/NCEA/bmds/index.html
U.S. EPA (U.S. Environmental Protection Agency). (201 Ib). Recommended use of body weight 3/4 as the
default method in derivation of the oral reference dose [EPA Report]. (EPA/100/R11/0001). Washington,
DC. http://www.epa.gov/raf/publications/interspecies-extrapolation.htm
6-13
-------
U.S. EPA (U.S. Environmental Protection Agency). (2012a). Benchmark dose technical guidance. (EPA/100/R-
12/001). Washington, DC: Risk Assessment Forum.
http://www.epa.gov/raf/pubrications/pdfs/benchmarkdose guidance.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2012b). EPA announces NAS' review of IRIS Assessment
development process. Available online at
http://vosemite.epa.gOv/opa/admpress.nsf/0/lce2a7875daf093485257a000054df547OpenDocument
USDA (U.S. Department of Agriculture). (2010). Dietary guidelines for Americans, 2010 (7th ed.). Washington,
DC: U.S. Department of Health and Human Services. U.S. Government Printing Office.
http://www.health.gov/dietaryguidelines/2010.asp.
Vetrano. AM: Heck. DE: Mariano. TM: Mishin. V: Laskin. PL: Laskin. JD. (2005). Characterization of the
oxidase activity in mammalian catalase. J Biol Chem 280: 35372-35381.
http://dx.doi.org/10.1074/ibc.M503991200.
Ward. KW: Perkins. RA: Kawagoe. JL: Pollack. GM. (1995). Comparative toxicokinetics of methanol in the
female mouse and rat. Toxicol Sci 26: 258-264.
Ward. KW: Pollack. GM. (1996). Comparative toxicokinetics of methanol in pregnant and nonpregnant rodents.
Drug Metab Dispos 24: 1062-1070.
Ward. KW: Blumenthal GM: Welsch. F: Pollack. GM. (1997). Development of a physiologically based
pharmacokinetic model to describe the disposition of methanol in pregnant rats and mice. Toxicol Appl
Pharmacol 145: 311-322. http://dx.doi.org/10.1006/taap.1997.8170.
Weiss. B: Stern. S: Soderholm. SC: Cox. C: Sharma. A: Inglis. GB: Preston. R: Balys. M: Reuhl KR: Gelein. R.
(1996). Developmental neurotoxicity of methanol exposure by inhalation in rats (Vol. NTIS). (HEI Research
Report Number 73). Boston, MA: Health Effects Institute.
Wentzel P: Rydberg. U: Eriksson. UJ. (2006). Antioxidative treatment diminishes ethanol-induced congenital
malformations in the rat. Alcohol ClinExp Res 30: 1752-1760. http://dx.doi.org/10.llll/U530-
0277.2006.00208.x.
Wentzel P: Eriksson. UJ. (2006). Ethanol-induced fetal dysmorphogenesis in the mouse is diminished by high
antioxidative capacity of the mother. Toxicol Sci 92: 416-422. http://dx.doi.org/10.1093/toxsci/kfl024.
White. LR: Marthinsen. ABL: Richards. RJ: Eik-Nes. KB: Nilsen. OG. (1983). Biochemical and cytological
studies of rat lung after inhalation of methanol vapour. Toxicol Lett 17: 1-5. http://dx.doi.org/10.1016/0378-
4274(83)90027-9.
Wilson. SL: Cradock. MM. (2004). Review: Accounting for prematurity in developmental assessment and the
use of age-adjusted scores [Review]. J Pediatr Psychol 29: 641-649. http://dx.doi.org/10.1093/jpepsv/jsh067.
Woo. YS: Yoon. SJ: Lee. HK: Lee. CU: Chae. JH: Lee. CT: Kim. DJ. (2005). Concentration changes of
methanol in blood samples during an experimentally induced alcohol hangover state. Addict Biol 10: 351-
355. http://dx.doi.org/10.1080/13556210500352543.
Zorzano. A: Herrera. E. (1989). Disposition of ethanol and acetaldehyde in late pregnant rats and their fetuses.
Pediatr Res 25: 102-106.
6-14
------- |