&EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-81 -135 July 1982
Project Summary
Level 1 Bioassay
Sensitivity
D. J. Brusick and R. R. Young
O
Environmental assessment bioassays
are conducted to detect toxic constitu-
ents In complex emissions from Industrial
sites. Unlike instrumentation used in
analytical chemistry, the power (limit of
detectabllity) of the biological techniques
to detect specific chemicals or even
classes of chemicals is seldom known.
In this report, the published literature
is surveyed and used to establish a set
of sensitivity estimates for these tests.
These estimates wlH permit a comparison
of the bioassays and will also give an
estimate of the concentrations of toxic
materials that could be in a mixture
which registers negative in a particular
test (i.e.. What does a negative test
response indicate?).
Three tests, the Ames Salmonellal
microsoma mutaganesls assay, the In
vitro rodent cell (CHO) clonal toxtoity
assay, and the in vivo rodent toxicity
assay all have substantial published
data bases using study designs similar
to those employed in Level 1 environ-
mental assessment. The sensitivity limits
for these three tests are summarized in
this review. Methods developed to
assess these data will be applicable to
other Level 1 tests such as the rabbit
alveolar macrophage (RAM) assay,
aquatic tests, and other ecological
assays. However, these evaluations
await the development of a sufficient
data base on a wide variety of pure
compounds. Most of the other Level 1
bioassays (e.g., fish toxicity, RAM) have
extensive data bases for responses with
complex environmental samples, but
not with pure chemicals. The insect
toxicity test (Drosophila LD50) has a
large base of published data but the test
conditions are extremely variable, mak-
ing an interpretation of the sensitivity
difficult.
This Project Summary was developed
by EPA's Industrial Environmental Re-
search Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that Is fully docu-
mented in a separata report of the same
title (see Project Report ordering infor-
mation at back).
Introduction
The primary objective of EPA/IERL-RTP
Level 1 Environmental Assessment is to
reliably evaluate stationary-source emis-
sions for their toxic potential. With this
information, emission sources and pro-
cess streams can be ranked according
to their estimated toxic potency and
further testing can be applied efficiently.
One feature of Level 1 Environmental
Assessment which needs to be addressed
is the level of sensitivity for the chemical
and biological methods. There is little
problem in defining the limits of sensi-
tivity in chemical analysis; however, the
same type of definition for bioassays is
extremely difficult. This is especially true
for some of the bioassays because the
data are both quantitative and qualitative.
Qualitative responses are defined as
either positive or negative, toxic signs
or descriptions of growth. Quantitative
effects are those which are specifically
measured by .counting colonies, animal
deaths, or growth-medium turbidity.
This report discusses the problem of
determining the levels of sensitivity in
biological assays recommended for
-------
Level 1 Environmental Assessment as
defined in the IERL-RTP procedures
manual (1) and summarizes the sensi-
tivity data where sufficient data exist.
Data and sensitivity estimates have been
developed for the Ames Salmonella/
microsome mutagenesis assay, the ro-
dent cell (CHO) clonal toxicity assay,
and the in vivo rodent acute toxicity
assay.
Approach
The proposed data formatting and
evaluation scheme in "Level 1 Biological
Testing Assessment and Data Format-
ting" defines the concept of "no detect-
able toxicity" for each of the bioassays
as no significant response at a maximum
applied dose (2). This definition implies
that the test sample may contain toxi-
cants but that they are below the degree
of resolution inherent in the particular
assay. The following list contains sev-
eral factors that are important in influ-
encing the resolution of a test:
1. The maximum dose/concentration
that can be applied to the assay
system.
2. The inherent response of the assay
system to specific classes of
chemicals.
3. The number and site of critical tar-
gets in the test organism that must
be affected to produce a lethal or
toxicological response.
4. The validity of extrapolating dose
response data for lethality and
mutagenesis linearly down to low
dose levels.
5. The chemical interactions fsyner-
gistic and antagonistic) in complex
mixtures.
6. The ability of the assay system to
detoxify or eliminate the toxicant
will impact on the limit ofbioassay
resolution and detectability.
7. The ability of the assay system to
alter the toxicant metabolically to a
more toxic or mutagenic form will
affect the limit ofbioassay resolu-
tion and detectability.
8. Nature of the test response.
Review of Data in General
Defining the lower limits of assay sensi-
tivity is difficult. The critical factor in
developing a data base of comparable
information is protocol standardization.
Data from bioassays using standardized
procedures, such as the Ames Salmo-
ne//a/microsome mutagenesis assay and
the in vivo rodent acute toxicity assay,
can be easily collected and compared.
Problems arise when data from different
protocols are compared. The rather sim-
plistic approach of comparing Level 1
data collected under similar conditions
does not tell the entire story about the
true level of assay sensitivity or rele-
vance to environmental assessment.
Considerations such as the slope of the
response curve, the potential for bio-
accumulation or biotransf ormation, and
the effect of chronic versus acute
exposure -ideally should not be ignored.
Additional factors (e.g., the nature of
the test response) must be considered
when evaluating Level 1 data for levels
of sensitivity. Most Level 1 bioassays
produce continuous data sets over the
range of measured responses. A basic
difference exists, however, between
Ames mutagenicity data and toxicity
data produced by the other assays. In
the toxicity assays, any chemical may
be expected to exhibit toxicity if tested
at a high enough dose. Assay sensitivity
is limited by the physical capacity of the
system to accept test material and the
length of time during which the material
is applied. In the Ames assay, on the
other hand, a nonmutagenic compound
will not produce a positive response
regardless of applied dose. Ames data is
therefore discontinuous (e.g., a chemi-
cal is either mutagenic or nonmuta-
genic) and only positive (mutagenic)
chemicals exhibit a continuous data set.
If a sample is positive, it can be further
categorized as having high, moderate,
or low mutagenicity based on the ob-
served minimum effective concentration.
A modicum of uncertainty exists in dif-
ferentiating between nonmutagenic
chemicals and weakly positive chemi-
cals with activity below the threshold of
Ames assay sensitivity. Chemicals in
both situations are designated as having
nondetectable mutagenicity.
Tests with clearly dichotomous re-
sponses (e.g., Ames test with data
evaluated as mutagenic or nonmuta-
genic) are more amenable to sensitivity
evaluation than tests with continuous
responses (e.g., EC50 in the CHO clonal
toxicity test). For example, it appears
that mutation induction is a "single hit"
phenomenon and that the level of re-
sponse is a function of total dose (con-
centration x exposure time). This means
that one should obtain an equal response
by elevating the concentration over a
short exposure period or by extending the
exposure period at a low concentration.
Lethality, however, is generally not a
"single hit" phenomenon, and distinctly
different patterns of lethality will be ob-
tained from experiments when the dose
is held constant but the concentration
and time parameters are varied. Thus, to
define the lower limit effect as a function
of dose is meaningful for the Ames test
but not for bioassays measuring lethality
unless the other parameters are specified.
A Review of Ames Test Data
The Ames Salmonella/microsome
mutagenicity assay has a well-accepted
standardized protocol which has been
adapted to Level 1 Environmental Assess-
ment testing. A large amount of Ames
mutagenicity data is available that may
be used to evaluate the levels of sensi-
tivity and detectability of the Ames
assay. For example, McCann et al. (3)
have compiled a list of data from which
this evaluation is drawn.
The minimum effective concentration
(MEC expressed as amount-per-plate)
can be calculated for these various
compounds and compared. Table 1
summarizes the range of reported MEC
values for different chemical classes.
The chemicals reported in this section
are organized by chemical family using
the Multimedia Environmental Goals
(MEG) (4) classification scheme.
A Review of Rodent Toxicity Data
Level 1 acute rodent toxicity testing
is a valuable test method for toxicological
assessment of complex effluents. The
advantages of the in vivo toxicity assays
lie mainly in the fact that the testing is
performed in whole animals. Also there
is a significant background of rodent
test data on a wide range of toxicants
using standard test protocols, thus sup-
plying needed information for analysis
of levels of assay sensitivity and for
reliable interpretation of results with
complex effluents. The primary disad-
vantage of the assay is its inability to
predict the toxicity induced by long-term/
low-level exposures.
Table 2 summarizes the range of LD5C
values encountered for each chemical
(MEG) category in the review of rodent
toxicity data. The dose required to kill
50 percent of test animals (LD50) is
reported as milligrams of chemical pei
kilogram of animal body weight. Using
the approach discussed above, the sen-
sitivity of the assay to each chemical is
determined. This assumes that only the
chemical in question exerts toxicity and
that no antagonistic or synergistic effects
occur.
-------
Table 1. Level 1 Ames Assay: Range of Minimum Effective Concentration IMEC) Summarized by Chemical Class
MEG
Group*
6
8
10
14
16
17
21, 22
23
24
25
26
Chemical
Class
Glycols, Epoxides
Carboxylic Acids
and Derivatives
Amines
Sulfonic Acid,
Sulfoxides
Halogenated Aromatic
Compounds
Aromatic Nitroso
Compounds
Fused Polycyclic
Hydrocarbons
Heterocyclic Nitrogen
Heterocyclic Oxygen
Compounds
Heterocyclic Sulfur
Compounds
Organophosphon/s
Compounds
No. of
Entries
4
6
30
2
7
6
17
10
6
1
2
MECD Range,
Low and High
Values,
u.g/plate
7x 10-2
400
3x 10-3
64
6x 10-3
210
49
217
7.8 x 10-2
2
1.4x 10-2
73.5
1.3x 10-1
120
1 x 10-2
38
1.2x 10-2
42
11
1.1
60
Chemical
Benzo(a>pyrene-4, 5-oxide
1,2, 7,8-diepoxyoctane
2-(2-furyll-3-(nitro-2-furyl)-acrylamide (AF-2)
Melphalan
2-aminofluorene
N,N-dimethyl-4-lphenylazo) benzeneamine
Methyl methanesulfonate
Ethyl methanesulfonate
9-10-dichloro-methyl anthracene
1 0-bromoanthracene
2-nitrosofluorene
2-nitrosonaphthalene
D aunorubicin • HCI
Benzofeipyrene
4-nitroquinoline- 1 -oxide
U racil mustard
Aflatoxin B-, and aflatoxicol
Aflatoxin B2
Hycanthone methanesulfonate
Cyclophosphamide
isophosphamide
•MEG
Multimedia Environmental Goals (4); chemical classification scheme developed as part of £PA Level 1 Environmental Assessment testing program. Some
chemicals may be placet! in more than one group.
Minimum Effective Concentration. The minimum amount of test material required to give a positive response in the most sensitive tester strain.
A Review of In Vitro Mammalian
Clonal Toxicity Assays
Mammalian in vitro clonal toxicity
assays provide a sensitive and reliable
method to measure and compare the
cytotoxicity of test agents. The Chinese
hamster ovary (CHO) cell clonal toxicity
assay is routinely used to measure the
toxicity of environmental samples sub-
mitted under EPA Level 1 testing of
point source emissions. The measured
end point is the inhibition of colony for-
mation as a function of dose. The stan-
dard parameter for comparison is the
dose necessary to reduce the colony-
forming ability of quantitatively plated
mammalian cells by 50 percent (the
EC50 value). These survival data are
continuous over the doses tested, and
the EC50 may be determined statistically
or graphically from the data.
Quantitative comparisons of toxicity
data can only be made from assays con-
ducted under comparable conditions.
Since a wide variation in CHO toxicity
assay protocols was encountered in the
literature review, criteria were developed
for selecting cytotoxicity data for
comparison. Parameters for which stan-
dard ranges were developed included
attachment time, exposure period, cell
type, serum concentration, and cell
density. The amount of comparable
CHO clonal toxicity data was greatly
reduced when the requirements of a
standardized assay design were imposed.
Table 3 summarizes the range of EC50
values by chemical (MEG) category
from experiments meeting the CHO
clonal toxicity assay study design criteria.
Level 1 Bioassays Not Reviewed
The limits of resolution and detect-
ability of the remaining Level 1 bioassays
are not addressed in this report. These
assays, for the most part, have not been
applied on a sufficiently large scale or
performed under standardized conditions
for any valid comparison of the data to
be made. A complete review of the sen-
sitivity of these assays can be made in
tfie future once a larger data base and
standard study designs have been
developed.
-------
Table 2.
MEG
Groupb
1
2
3
6
7
8
9
10
11
12
13
14
15
16
17
18
20
21, 22
23
26
26
28
Level 1 Rodent Toxicity Assay: Range
Chemical
Class
Aliphatic Hydrocarbons
AlkylHalides
Ethers
Glycols, Epoxides
Aldehydes, Ketones
Carboxylic Acids
and Derivatives
Nitrites
Amines
Azo Compounds;
Hydrazine Derivatives
Nitrosamines
Thiols, Sulfides,
Disulfides
Sulfonic Acids,
Sulfoxides
Benzene, Substituted
Benzene Hydrocarbons
Halogenated Aromatics
Aromatic Nitroso Compounds
Phenols
Nitrophenols
Fused Polycyclic
Hydrocarbons
Heterocyclic Nitrogen
Compounds
Organophosphorus
Compounds
Metals and Organometallic
Compounds^
ofLDso*
No. of
Entries
2
- 7
1
2
5
9
2
9
5
1
2
2
8
7
3
4
1
2
4
2
18
Values Summarized by
LD50 Range,
Low and High
Values,
g/kg
800
1440
81
3160
170
238
4200
21
1300
52
5040
65
78
20
1625
33
265
3850
38
3700
200O
2100
16
3900
50
4000
135
812
300
1600
45
438
700
125
3340
135
292
4
4000
Chemical Class
Chemical
1 -nitropropane
nitromethane
2-chloroethanol
trichloro vin yl-silicone
Bis(2,3-epoxy-propyl) ether
9-epichlorohydrin
2-ethyl- 1,3-hexanediol
Ac role in
Cyclohexanone
2-fluoroacetamide
Citric acid
Phthalonitrile
Phen ylace tonitrile
2,2-dichloro-N-methyl-diethylamine HCL
2, 4-diaminotoluene
1 -methylhydrazine
1 , 1 -dimethylhydrazine
N-nitrosodiphenylamine
Methylsulfide
Tris ( 1 -aziridinyl) phosphine sulfide
Sodium lauryl sulfonate
Alky/ aryl sulfonate
n-cumenol methyl carbamate
Anthranilic acid methylester
Aldrin
Hexachlorobenzene
1 -chloro-2-nitrobenzene
p-nitroaniline
Phenol
Propylgallate
2, 4-dinitrophenol
Carbonyl
Phenanthrene
Diquat
Nicotine
Dichlorvos (DDVP)
Metepa
Sodium dimethylarsinate
Sodium chloride
aLDg0 = Dose lethal to 50 percent of animals.
bM£G = Mult/media Environmental Goals classification scheme 141.
cMetals and organometallic compounds are summarized as a group and not by individual elements.
4
-------
Table 3. Level 1 CHO Clonal Toxicity Assay: Range of EC5tf> Values Summarized by Chemical Class
EC50 Range,
Low and High
MEG
Group*
8
12
14
23
82
68
Chemical
Class
Carboxylic Acids
and Derivatives
Nitrosamines
Sulfonic Acids,
Sulfoxides
Heterocyclic Nitrogen
Compounds
Metals and Organometallic
Compounds^
No. of
Entries
1
7
5
1
17
Values,
lug/ml)
5,700
0.5
10,000
4.5
317.9
1
0.062
852.7
Chemical
Caprolactam
N-nitrosomethylurethane (NMUTI
Dimethylnitros-amine
Dimethylsulfate IDMS)
Propylmethane sulfonate (i-PMS)
ICR-191
Cadmium chloride
Chromic chloride-hexahydrate
*ECSQ = Effective Concentration of chemical that reduces colony development by SO percent relative to control levels.
&MEG - Multimedia Environmental Goal chemical classification scheme (41.
cMetals and organometallic compounds are summarized as a group and not as individual elements.
Health Effects Assays
The rabbit alevolar macrophage (RAM)
assay is the only health effects assay
without a suitable data base for evalua-
tion. The prospects are good that, as
this assay is incorporated into routine
testing programs, additional validation
work with pure chemicals will be under-
taken and reported.
Aquatic Ecological Assays
Aquatic toxicology has had a long
record of investigation into the physio-
logic response of aquatic organisms to
specific pure and complex samples.
Unfortunately, there is not a large data
base for work conducted exclusively
with Level 1 protocols using the recom-
mended indicator organisms. It is beyond
the scope of this project to compare and
interpolate data collected from similar,
but basically different, study designs.
Terrestrial Ecological Assays
The three Level 1 terrestrial ecological
assays have not been validated as well
as have the other Level 1 assays. This
lack of published work, both on proce-
dures and assay validation, has prompted
EPA's Industrial Environmental Research
Laboratory—RTF to direct the produc-
tion of a laboratory workbook for terres-
trial assays (5). No attempt was made
to collect and evaluate data from these
assays.
Conclusions
It is possible to estimate the level of
sensitivity of bioassays used in Level 1
Environmental Assessment. However,
with the exception of only a limited
number of tests currently proposed, a
data base collected under conditions
similar to those recommended for Level
1 testing is not available. For those
tests it is advised that the data base be
developed during ongoing Level 1
analyses. Where the data bases were
available, the sensitivity levels were
estimated. The mammalian in vitro
clonal toxicity test and the Ames Sal-
monella test appeared to be quite sensi-
tive compared to the in vivo rodent
toxicity test. The two in vitro tests also
approach the sensitivity required of the
chemical analyses performed in Level 1
assessments.
Existing data resulting from EPA/IERL-
RTP's ongoing Level 1 Environmental
Assessment programs may also provide
sufficient chemical and biological data
to evaluate the sensitivities of the
remaining bioassays.
The information in this report might
be useful, not only in developing an
appreciation of the intrinsic sensitivity
of the bioassays, but also in modifying
assays to increase their sensitivities.
References
1. Brusick, D.J., and R.R. Young, IERL-
RTP Procedures Manual: Environ-
mental Assessment, Biological
Tests. EPA-600/8-81-024, Litton
Bionetics, Inc., Kensington, MD,
October 1981, 177 pp.
2. Brusick, D.J. Level 1 Biological
Testing Assessment and Data For-
matting. EPA-600/7-80-079 (NTIS
PB80-184914), Litton Bionetics,
Inc., Kensington, MO, April 1980,
100 pp.
3. McCann, J., E. Choi, E. Yamasaki,
and B.N. Ames: Detection of carci-
nogens as mutagens in the Salmo-
ne//a/microsome test: Assay of 300
chemicals. Proc. Nat. Acad. Sci.
(USA), 72:5135-5139, 1975.
4. Cleland, J.G., and G.L. Kingsbury.
Multimedia Environmental Goals for
Environmental Assessment, Vol. 1.
EPA-600/7-77-136a (NTIS PB
276919), Research Triangle Insti-
tute, Research Triangle Park, NC,
November 1977, 353 pp.
5. Brusick, D.J., and T.A. Gezo: IEPL-
RTP Procedures Manual: Level 1 En-
vironmen tal A ssessmen t, Terres trial
Tests. EPA Contract No. 68-02-
2681, Technical Directive No. 401,
Litton Bionetics, Inc., Kensington,
MD. In preparation.
•US.QOVERNMENT PRINTING OFFICE:1M2-559-092-440
5
-------
D. J. Brusick and R. R. Young are with Litton Bionetics, Inc., Kensington, MD
20895.
Raymond G. Merrill is the EPA Project Officer (see below).
The complete report, entitled "Level 1 Bioassay Sensitivity," (Order No. PB
82-221 201; Cost: $9.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park. NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
------- |