Vaccari D.A., Strom P.F., Alleman J.E. Environmental Biology for Engineers and Scientists

Подождите немного. Документ загружается.

FIELD AND LABORATORY TOXICOLOGY

In the ﬁrst part of the chapter, we discuss toxicity testing in detail, including descriptions

of the types of tests commonly used, the variety of organisms used in testing, and extra-

polation of results from animals to humans. Then we describe epidemiology and how it is

used in the study of human disease.

20.1 TOXICITY TESTING

The laboratory measurement of toxicity is one of the largest activities in toxicology.

We can distinguish three ways to detect toxic effects: in vivo, in vitro, and in situ. An

in vivo test means a laboratory experiment with whole living organisms. Most of the

discussion here concerns this type of testing. In vitro tests (literally ‘‘in-glass’’) are per-

formed with living cells extracted from plants or, more often, from animals. These can be

faster and less expensive, but less precise. Therefore, they are best used for screening

purposes. In situ measurements are performed on organisms in their natural habitat.

These can be the most revealing, but in practice are limited by economic and logistical

constraints.

Here we are concerned with toxicity testing to determine the probability of particular

adverse effects, as distinguished from laboratory research experiments that seek to under-

stand toxic mechanisms. Toxicity tests are often called bioassays. This term is more gen-

erally used to describe experiments in which organisms are used as a kind of chemical

detector. For example, the growth rate of microorganisms is used to determine the potency

of vitamins.

Environmental Biology for Engineers and Scientists, by David A. Vaccari, Peter F. Strom, and James E. Alleman

788

Consider the task of understanding the toxicity of even a single chemical substance. All

of the following parameters must be selected:

 Species and strain

 Age or life stage of organisms

 Sex

 Toxic endpoint (mortality, tumor formation, etc.)

 Route of administration

 Duration of test

 Number of organisms per test

 Dosage levels

In addition, there are numerous decisio ns to be made about experimental protocol and

data handling. Clearly, the reporting of the toxici ty of a substance in terms of a single

number is very incomplete. There is little assurance that the effect of ingestion of a che-

mical on the growth of adult rats of both sexes has any relation to the 96-hour LD

inhalation in juvenile hamsters, let alone to the NOAEL for human males exposed der-

mally in an occupational setting.

Although the possible variety may seem daunting, toxicologists have created a variety

of standardized approaches and test protocols that make it easier to compare results

obtained by different experimenters. However, one should keep the following in mind:

The inherent variability of living things in their response to toxins, and their sensitivity

to numerous environmental factors, can produce large differences in results from different

laboratories, or even between tests conducted at the same lab at different times. For exam-

ple, duplicate tests of endosulfan toxicity to ﬁsh conducted at four labs produced LC

values with a geometric mean of 1.34, but which ranged from 0.68 to 3.30, a factor of

almost 5.

20.1.1 Design of Conventional Toxicity Tests

As in any project, the ﬁrst decisi on to be made in selecting or developing a toxicity test is

to carefully state the objectives. Furthermore, given the limitations of extrapolating from

laboratory to ﬁeld, the objectives often must be circumscribed in light of what can be dis-

covered by laboratory tests. Examples of uses for toxicity data include support for deci-

sions on product development or use, permitting activities for waste discharges to the

environment, risk assessment, or in support of environmental litigation. Considerations

in the design of conventional toxicity tests are described here . These exclude specialized

tests, such as for carcinogenicity, mutagenicity, or teratogenicity. Additional considera-

tions that apply to those are described in the following sections.

20.1.2 Test Duration

The time scale of the effect commonly falls into three categories. Acute toxicity tests are

usually complete within 24 to 96 hours. Short-term (also called subacute or subchronic)

toxicity tests take longer, up to about 10% of the life span of the organism. Long-term or

chronic toxicity tests cover most of the life span of the organisms. The life span of mice is

about 18 months, 24 months for rats, and 7 to 10 years for dogs and monkeys.

TOX ICITY TESTING 789

20.1.3 Selecting Organisms

The goal of toxicity testing may be generalized into two categories according to the target

of interest: humans or all other living things. Humans themselves may be used as subjects

in toxicological studies under carefully proscribed conditions. For example, in human

clinical trials, subjects may be administered dosages below the expected NOAEL and

then have their blood and urine sampled periodically to determine rates and mechanisms

of biotransformation and elimination.

However, for the most part, other mammals are used as surrogates for humans in the

determination of adverse effects. The mouse or rat is generally used for determination of

the LD

because they are fairly economical and easy to handle. They are also preferred

because a large number of toxicological data have been accumulated using them, which

makes it possible to compare results between different toxic substances. The pig is better

for purposes of extrapolation to humans because it is phylogenetically closer and more

similar to humans in diet as well. Of course, nonhuman primates such as monkeys are

closest of all. However, expense often precludes their use.

For environmental toxicity testing, a wider variety of organisms may be used. It may

be necessary to consider toxic effects on vertebrates other than mammals (ﬁsh, birds,

amphibians), on invertebrates, and on plants. Test organisms may be collected from the

wild. Whenever possible, organisms should be selected for environmental toxicity test-

ing that:

 Represent a broad range of sensitivities

 Are abundant and easily available species

 Are indigenous to any area that will be affected

 Include species that are important recreatio nally, commercially, or ecologically

 Are easy to culture and maintain in the laboratory

 Have adequate background information available, such as physiology, genetics, and

behavior

Table 20.1 lists some organisms commonly used for acute aquatic toxicity tests. Of

these, the most commonly used are the water ﬂea (Daphnia sp.), fathead minnow, bluegill,

rainbow trout, mysid shrimp, and sheepshea d minnow. Daphnia magna and mysid shrimp

have been found useful for predicting the MATC for ﬁsh.

Among terrestrial animals, the mammals used include several species of rats and mice,

as well as hamsters, guinea pigs, rabbits, and dogs. Common avian species are Northern

bobwhite (Colinus virginianus), Japanese quail (Coturnix japonica), mallard (Anas platyr-

hynchos), ring-necked pheasant (Phasianus colchicus), American kestrel, and screech owl

(Otus asio). A teratogenicity test has been developed using the South African clawed frog,

Xenopus laevis. Earthworms have been used to assess the toxicity of soil samples con-

taminated with hazardous wastes.

Finally, some toxicity tests use a combination of species in a single test. A small-scale

ecosystem with at lea st two interacting organisms is called a microcosm. At a medium

scale, they are calle d mesocosms. These are useful for evaluating ecosystem effects such

as biomagniﬁcation or predation. They may range from two organisms cultured together

in a test-tube microcosm, to a farm pond mesocosm used to evaluate agricultural pesticide

effects. Terrestrial microcosms may consist of soil cores with associated microorganisms

and invertebrates.

790 FIELD AND LABORATORY TOXICOLOGY

TABLE 20.1 Common Organisms Used in Aquatic Acute Toxicity Tests

Freshwater

Vertebrates

Rainbow trout, Salmo gairdneri

Brook trout, Salvelinus fontinalis

Fathead minnow, Pimephales promelas

Channel catﬁsh, Ictalurus punctatus

Bluegill, Lepomis macrochirus

Frog, Rana sp.; toad, Bufo sp.

Invertebrates

Daphnids: D. magna, D. pulex, D. pulicaria, Ceriodaphnia dubia

Amphipods: Gammarus lacustris, G. fasciatus, G. pseudolimnaeus

Crayﬁsh: Orconectes sp., Cambarus sp., Procambarus sp., or Pacifastacus leniusculus

Midges: Chironomus sp., Stoneﬂies, Pteronarcys sp.

Mayﬂies: Baetis sp., Ephemerella sp., Hexagenia limbata, H. bilineata

Snails: Physa integra; planaria, Dugesia tigrina

Algae

Chlorphyta (green algae), Selenastrum capricornutum

Cyanophyta (blue-green bacteria), Anabaena ﬂos-aquae, Microcystis aeruginosa

Chrysophyta (brown algae and diatoms), Navicula pelliculosa, Cyclotella sp., Synura petersenii

Saltwater

Vertebrates

Sheepshead minnow, Cyprinodon variegatus

Mummichog, Fundulus heteroclitus

Longnose killiﬁsh, Fundulus similis

Silverside, Menidia sp.

Threespine stickleback, Gasterosteus aculeatus

Pinﬁsh, Lagodon rhomboides

Spot, Leiostomus xanthurus

Sand dab, Citharichthys stigmaeus

Invertebrates

Copepods: Acartia tonsa, A. clausi

Shrimp: Penaeus setiferus, P. duorarum, P. aztecus

Grass shrimp: Palaemonetes pugio, P. vulgaris

Sand shrimp: Crangon septemspinosa

Mysid shrimp: Mysidopsis bahia

Blue crab: Callinextes sapidus

Fiddler crab: Uca sp.

Oyster: Crassostrea virginica, C. gigas

Polychaetes: Capitella capitata, Neanthes sp.

Algae

Chlorophyta (green algae): Chlorella sp., Chlorococcum sp.,

Dunaliella tertiolecta

Chrysophyta (brown algae and diatoms): Isochrysis galbana, Nitzschia closterium, Pyrmnesium

parvum, Skeletonema costatum, Thalassiosira pseudonana

Rhodophyta (red algae): Pophyridium cruentum

Source: Based on Rand and Petrucelli (1985); Landis and Yu (1995).

TOX ICITY TESTING

791

Several standard microcosms have been developed. For example, the standar dized

aquatic microcosm (SAM) is conducted in 4-L jars and provides a highly controlled

environment. Four species of invertebrates, 10 of algae, and one of bacteria are added

to a sterilized aquatic medium. Temperature, illumination, and chemical and biological

sampling schedule are speciﬁed over the 63-day test duration. Another example is the

soil core microcosm (SCM). It consists of 60-cm-long by 17-cm-diameter soil columns

obtained from the ﬁeld, containing naturally occurring organisms, and is tested over a per-

iod of at least 12 wee ks.

Testing should preferably be done with animals of both sexes, and both adults and

juveniles. Sometimes tests are done with equal numbers of both sexes, but they are

lumped together for measurement purposes.

20.1.4 Toxic Endpoint and Other Observations

All of the effects described in Table 17.1 are candidate endpoints for toxicity testing.

These include biochemical, histological, and individual effects. Population effects can

be revealed by measurement of growth, death, and fertility rates. Even ecological inter-

actions may be measured by using microcosms and mesocosms. Although the test objec-

tive may require a single endpoint, good use of an experiment calls for the reco rding of

numerous observations. Mortality is commonly the primary effect to be observed. How-

ever, other changes can reveal toxic effects that may resu lt in mortality beyond the test

period. These may include body weight and food consumption, which may be needed for

computating dosage. Autopsies should be performed on all the dead animals as well as

some of the survivors. This practice can identify the organs or tissues that have been

affected. General observations include changes in color or consistency of stool or

urine, and appearance of eyes, skin, and hair coat.

In aquat ic toxicity tests, a variety of chemical and physical measurements should be

performed regularly during the test. These may include pH, alkalinity, hardness, tempera-

ture, oxygen concentration, and salinity or conductivity of the medium.

In plant toxicity tests, endpoints may include growth rates, rate of seed germination,

and root elongation. The response of algae is typically measured in terms of growth. This,

in turn, is measured in cell mass, by extracting chlo rophyll and measuring spectrophoto-

metrically, or by cell counts (microscopic or electronic).

20.1.5 Route of Administration and Dosage

All of the natural routes of exposure (oral, dermal, or by inhalation) may be used experi-

mentally.The oral route may be administered with the food or by tube directly into the

gastrointestinal tract. The latter method is called gavage. In addition, chemicals may

be administered by injection intravenously (into the circulatory system), intraperitone-

ally (into the abdominal body cavity), intramuscularly (into the muscle), or subcuta-

neously (below the skin). These injection methods, however, run the risk of causing

local effects.

It is preferable to apply dosages of toxin normalized to the body weight of the animal.

However, it is often more practical to dose by a ﬁxed concentration in the food, air, or

water. Because in this case some animals will obtain a relatively higher or lower

dose than others, the standard deviation of the dose–response tolerance distribution

will be greater. The toxicant in an aquatic toxicity test may be an aqueous mixture

792 FIELD AND LABORATORY TOXICOLOGY

such as wastewater treatment plant efﬂuent. In this case, dilutions are used in place of

concentration, expressed as a percentage of the whole efﬂuent.

Concentration levels are chosen to brac ket the ED

expected value, based on experi-

ence with comparable substances. Initially, it may be necessary to perform a range-ﬁnding

test with dosage levels spaced out approximately by factors of 10. For example, three

levels might be chosen with concentrations of 0.1, 1.0, and 10 mg/L. Efﬂuents will be

diluted successively by factors of 10; thus, one might use 100%, 10%, and 1%. Subse-

quently, more closely spaced dosage levels may be tested to obtain deﬁnitive information.

The ratio between successive dosage levels will typically be between 1.2 and 2.0. For

example, one might use nominal dosages of 10.0, 15.0, 25.0, 35.0, and 50.0 (the ratio

of one value to the next need not be constant but should be in a narrow range). The

goal is to have several dosage levels show some toxic effect, with at least one dosage

with less than 35% of the organisms showing toxic effect, and at least one showing

more than 65%.

When the number of treatment levels is small, duplicates of each concentration may be

called for. The purpose of duplicates is to help determine conﬁdence intervals. Since this

determination can also be made with regression techniques in ﬁtting the dose–response

curve, similar information can be obtained by doubling the number of treatment levels

within the same range.

One test group, with at least as many organisms as receive each treatment level, is left

untreated as a control. If a signiﬁcant number of control organisms suffer toxic effects

(say, more than 10%), the entire test may be invalidated. If a chemical toxin is applied

by mixin g with a solvent such as methanol or DMSO, another control group should be

given a treatment cont aining the solvent only.

20.1.6 Number of Organisms Per Test

The number of organisms exposed to each treatment level varies greatly with the require-

ments of the test and other practical factors such as cost. Fewer than 10 organisms limits

the resolution of the response. For example, if there were only ﬁve organisms per test,

mortality could only have values in multiples of 20%. Nevertheless, for large organisms

such as dogs, small numbers are commonly used because of cost. In aquatic toxicity tests

with ﬁsh, 10 organisms per treatment are recommended for static tests, and 20 for ﬂow-

through.

20.1.7 Other Experimental Variables

There are numerous seemingly minor issues, such as the design of organism containmen t

structures, which nevertheless can signiﬁcantly affect the response of organisms. Of great

importance is the amount of space allotted to each organism. Too small a space is a cause

of stress, which can interact with the toxicity. Sometimes the design is speciﬁed comple-

tely as to materials and dimensions. For example, one tes t speciﬁes that cages for rats have

a ﬂoor area from 110 cm

for a 100-g animal to 450 cm

for one that is 500 g. Handling,

lighting, and temperature may be speciﬁed. A test may require an acclimation period

before initiating the toxic exposure. In acute and subchronic tests, feeding is sometimes

withheld.

Aquatic tests may be static, static renewal, or ﬂow-through. In static tests, the water

is not changed after the beginning of the test. In static renewal tests, it is changed

TOX ICITY TESTING 793

periodically, whereas ﬂow-through tests have a continuous change . Aeration is some-

times required and sometimes not, but in either case the oxygen levels may be speciﬁed

as a percentage of the saturation value. The volume of solution needed is based on the

mass of organisms in the test. Recommended values are 0.5 to 0.8 g of biomass per

liter for static tests, 1 to 10 g/L day for ﬂow-through tests. The temperature for many

species is 20 to 22



C, but some organisms must be tested in colder water. For example,

some crayﬁsh and mayﬂies, threespine stickleback, sand shrimp and bay shrimp are tested

at 17



C. Salmon, trout, stoneﬂies, ﬂounder, herring, and marine copepods are tested at



C. On the other hand, mysid shrimp should be cultured at 27



If the test involves plants, lighting must be speciﬁed not only as to intensity, but also as

to the photoperiod.

Concern about the care and well-being of terrestrial vertebrates (e.g., rats and mice)

has led to strict guidelines for their husbandry and humane treatment. Research organiza-

tions have animal use committees that review protocols for proposed tests. The guidelines

seek to ensure that any pain and suffering experienced by these animals is justiﬁed by the

potential gain in information. They have forced investigators to design their studies more

carefully in order to maximize the information gained, reducing the number of animals

used. It has also spurred the development of alternative tests that do not use live animals,

such as the in vitro tests described below.

20.1.8 Conventional Toxicity Tests

Table 20.2 give references to a number of toxicity tests, to give an idea of the variety

available. Daphnia species, water ﬂeas, are one of the most important organisms for aqua-

tic toxicity testing because of their small size and ease of cultivation. They are crustaceans

that measure about 0.2 to 3 mm, appearing to the unaided eye as a swimming speck.

D. magna requires relatively hard water, about 80 to 100 mg/L as CaCO

, which is

about double the hardness requirement for D. pulex. Testing is usually done with organ-

isms less than 24 hours old. A single test with 10 organisms can be conducted in a 125-mL

container. Th ey can be fed algae or ﬁshmeal. An acute toxicity test conducted with

Daphnia has 48 hours’ duration. Mortality is difﬁcult to determine, so the endpoint chosen

is usually immobilization, deﬁned as not swimming even after gentle prodding with a

glass rod. The chronic toxicity or partial life cycle test runs 21 days and measures growth

in terms of length or mass, and numbers of offspring produced per organism. Ceriodaph-

nia nubia is much smaller than the other species, and it reproduces faster. A faster, less

expensive chron ic toxicity test has been developed with it, called the three-brood renewal

toxicity test. Ten organisms are placed in 15 mL of test solution, one to a container.

Reproduction is measured after 7 days.

The most deﬁnitive chronic toxicity test, called an embryo-to-embryo test, would

encompass the entire life-cycle of an organism. However, this can be costly and time con-

suming. As an alternative, research has found that the MATC for ﬁsh can be established

by testing embryos, larvae, and juveniles. This is called early life stage (ELS) toxicity

testing. Numerous endpoints have been used, such as mortality, motility reduction, repro-

duction reduction, and so on.

Toxicity to bacteria has been assayed by various means, such as the effect on plate

counts, cell growth rates, or oxygen uptake rate. Particular reactions such as nitriﬁcation

rate may also be tested, as this tends to be particularly sensitive.

794 FIELD AND LABORATORY TOXICOLOGY

TABLE 20.2 References on Standard Toxicity Tests and Protocols

Method

Number Date Title

D 4229-84 1993 Conducting static acute toxicity tests on wastewaters with Daphnia.

E 724-89 1993 Standard guides for conducting static acute toxicity tests starting with

embryos of four species of bivalve molluscs.

E 729-88a 1993 Standard guide for conducting acute toxicity tests with ﬁshes, macroinverte-

brates and amphibians.

E 1191-90 1993 Standard guide for conducting the renewal life-cycle toxicity tests with

saltwater mysids.

E 1197-89 1993 Standard uide for conducting renewal life-cycle toxicity tests with Daphnia

magna.

E 1197-87 1993 Standard guide for conducting a terrestrial soil-core microcosm test.

E 1218-90 1993 Standard guide for conducting static 96-h toxicity tests with microalgae.

E 1241-88 1993 Standard guide for conducting early life stage toxicity tests with ﬁshes.

E 1295-89 1993 Standard guide for conducting three-brood, renewal toxicity tests with

Ceriodaphnia dubia.

E 1366-91 1993 Standard practice for standardized aquatic microcosms: freshwater.

E 1367-90 1993 Standard guide for conducting 10-day static sediment toxicity tests with

marine and estuarine amphipods.

E 1383-90 1993 Standard guide for conducting sediment toxicity tests with freshwater

invertebrates.

E 1391-90 1993 Standard guide for collection, storage, characterization, and manipulation of

sediments for toxicological testing.

E 1415-91 1993 Standard guide for conducting the static toxicity tests with Lemna gibba G3.

E 1439-91 1993 Standard guide for conducting the frog embryo teratogenesis assay—

Xenopus (FETAX).

E 1463-92 1993 Standard guide for conducting static and ﬂow-through acute toxicity tests

with mysids from the west coast of the United States.

D 4229-84 1984 Conducting static acute toxicity tests on wastewaters with Daphnia.

E 729-88a 1991 Standard guide for conducting acute toxicity tests with ﬁshes, macroinverte-

brates, and amphibians.

E 1193-87 1987 Standard guide for conducting renewal life-cycle toxicity tests with Daphnia

magna.

E 1197-87 1987 Standard guide for conducting a terrestrial soil-core microcosm test.

E 1218-90 1990 Standard guide for conducting static 96-h toxicity tests with microalgae.

E 1241-88 1991 Standard guide for conducting early life stage toxicity tests with ﬁshes.

E 1367-90 1990 Standard guide for conducting 10-day static sediment toxicity tests with

marine and estuarine amphipods.

1383-90 1990 Standard guide for conducting sediment toxicity tests with freshwater

invertebrates.

U.S. Environmental Protection Agency references

EPA/600/ 1983 Protocols for bioassessment of hazardous waste sites

2-83/054

EPA/600/ 1987 Methods for measuring the acute toxicity of efﬂuents to aquatic

4-78/012 organisms

EPA/600/2 1988 Protocols for short term toxicity screening of hazardous waste sites

Source: Based on Landis and Yu (1995). Methods from the Annual Book of ASTM Standards, American Society

of Testing and Materials, Philadelphia.

TOX ICITY TESTING

795

The frog embryo teratogenesis assay (FETAX) is a 96-hour in vitro test using

embryos of the South African clawed frog, Xenopus laevis. It has wide acceptance as

an alternative to mammalian teratogenesis testing for several reasons. The cost is reason-

able because the frog is easy to raise and produc es large amounts of eggs and sperm. A

large database of results is available. The test is very repeatable and the results correlate

well with known mammalian teratogens. The concentration of toxin that produces 50%

mortality (the LC

) and that produces abnormalities in half the embryos [50% terato-

genic concentration (TC

)] are recorded. The ratio LC

/TC

, called the teratogenic

index (TI), is an indication of the developmental hazard posed by the toxin. Values of

TI below 1.5 are considered to pose little hazard.

Mammalian tests are conducted primarily for extrapolation to humans. Rats are pop-

ular test subjects. They should be selected to be less than six weeks old and to have a

variation in weight of less than 20% for each sex. A typical test duration is 90 days. Sub-

chronic inhalation toxicity is tested by introducing the toxin directly into the chamber air.

Oral toxicity may be administered in the food or water or by gavage (injection into the

stomach by a tube). Mortality is the most common endpoint, but examinations are also

made by urinalysis, blood chemistry, and so on.

20.1.9 In Situ Measurement of Conventional Toxicity

A criticism of laboratory toxicity tests is that the artiﬁcial environment, especially for ver-

tebrates, can create stresses different from what the organisms experience in their habitat.

Furthermore, the modes of exposure in the ﬁeld will be mixed, and this may be difﬁcult to

simulate experimentally. In situ measurements overcome these problems, although intro-

ducing others. It is harder to get reproduci ble results since there is no control over the

culturing and handling of the organisms. A negative control cannot be assured to differ

only in its exposure to a toxicant. Therefore, a greater number of samples may be needed

to establish statistical signiﬁcance than for laboratory experiments. On the other hand,

whatever unknown variables there may be, at least one can be fairly certain that they

are representative of a natural setting. Finally, although in situ measurements may indicate

that a disturbance has occurred, the cause cannot always be attributed to toxic exposure.

Biochemical tests on organisms in their habitats that can indicate toxic effects are

called biomarkers. Biomarkers are of particular interest for compounds that do not bioac-

cumulate, and that are thus difﬁcult to measure in tissues directly. Any of the biochemical

effects described above could be used. Examples include activity of enzymes such as

cytochrome P450 (a positive indicator of toxicity) or acetylcholinesterase (a negative indi-

cation). Increased metallothionein is a sensitive biomarker for metal exposure. Despite

their potential, they have not been widely found useful in environmental applications.

Some have proven useful in clinical application for cancer detection.

Indicator species are species that are naturally present and that are sensit ive to the

toxin. They may be used in one of two ways: (1) to indicate the health of an ecosystem

by their presence or absence; or (2) to bioaccumulate toxins, making the toxins more

easily detectible than in the environment. For example, bivalves are often used as indica-

tor species for bacterial pollution in coastal and estuarine ecosystems since, as ﬁlter fee-

ders, they tend to accumulate them.

Sentinel organisms are organisms deliberately introduced for later recovery and test-

ing. The classic case is the canary in the coal mine. Mussels have been used as sentinels

by suspending them in the water in plastic trays. Organisms that are sessile or otherwise

796 FIELD AND LABORATORY TOXICOLOGY

easy to recover are required for this type of study. Sentinels can provide early warning to

take necessary action in time. For example, sentinels may be used in wells at the boundary

of contaminated sites.

20.1.10 Occupationa l Monitorin g

Several tests can detect the effect of exposure to organophosphate or carbamate pesticides

on acetylcholinesterase. Red blood cell cholinesterase can be measured but is thought to

be overly sensitive. Surface electromyography (EMG) indicates more directly the pre-

sence of any dysfunction in the neuromuscular junction.

20.1.11 Population and Community Parameters

Above the individual level, toxic effects can be detected by changes within a population or

in the structure or function of a community of populations. Since toxic effects often target

the young, or reduce reproduction, changes in age distribution may result. In particular, an

exposed population may shift to older organisms.

Community structure can be described with more or less detail using food webs or

population diversity indices . A diversity index is meaningful only by comparison with

similar ecosystems with different exposure histories or by tracking changes in an index

with time or distance from exposure. Other things being equal, increased diversity is

thought to represent more stability in an ecosystem. However, similar normal ecosystems

can vary greatly in their index values due to variations in developmental stage, the proxi-

mity to boundaries or ecotomes, and othe r effects.

A number of different measures of community function exist. One is the ratio of

respiration to photosynthesis. However, this indicates nutrient enrichment pollution

more directly than does the presence of toxic substances.

20.1.12 Testing for Carcinogenicity and Teratogenicity

Carcinogenic potential can be detected by three types of tests: long-term carcinogenicity

studies, rapid screening tests, and biomarkers. Long-term tests are the most deﬁnitive.

These generally use mice or rats and last the lifetime of the animals (18 and 24 months,

respectively). Two or three dose levels are usually used, the highest being the ‘‘maximum

tolerated dose’’ (MTD). The MTD is estimated from 90-day studies and is chosen so as

not to produce severe noncarcinogenic toxicological effects or to reduce signiﬁcantly the

life of any organisms that do not develop tumors. The other doses are typically one-fourth

to one-h alf of the MTD. The number of tumors is recorded by animal (location, whether

benign or malignant , the presence of any unusual tumors), the number of animals with

tumors, the number of tumors per animal, the number of tumor sites, and the time to onset.

In the United States, a set of minimum test standards for carcinogenicity has been

developed by the National Toxicology Program. These standards require that at least

two species of rodents be used, at least two doses plus a control must be administered,

and at least 50 males and 50 females of each species must be tested at each dose.

Thus, the minimum number of animals in a study is 600. Such testing costs from

$500,000 to $1.5 million for each chemical, yet can only detect excess tumor rates

above 5 or 10%. The largest bioassay ever p erformed used over 24,000 mice, yet still

was not sensitive enough to measure excess risk of 1%. Despite these difﬁculties, the

TOX ICITY TESTING 797