5.2.3 Gene–environment interactions

[Guest guest]

This report contains the collective views of an international group

of experts and does not necessarily represent the decisions or the

stated policy of the United Nations Environment Programme, the

International Labour Organization or the World Health Organization.

Environmental Health Criteria 225

Principles For Evaluating Health Risks To Reproduction Associated

With Exposure To Chemicals

http://www.inchem.org/documents/ehc/ehc/ehc225.htm#6.2.2

4.4 Summary

An enormous number of potential target sites and processes exist

that could be perturbed by a toxicant and produce adverse

reproductive findings. In conceptual terms, an examination of the

reproductive cycle indicates many of the processes that may be

targets for toxicant action. It is quite possible that one agent may

have more than one potential site or mechanism of action. It should

also be noted from a simple examination of the cycle that adverse

effects due to exposure to a toxicant may not be immediate. Exposure

in utero may result in latent reproductive deficits when the

individual reaches adulthood and attempts to reproduce.

Investigational animal studies are usually carried out to explore

the mode of action of a toxicant suspected of having an adverse

reproductive effect. Frequently, such potential reproductive

toxicants are identified during standard regulatory testing

protocols, including subacute and subchronic toxicity studies. Thus,

it is reasonable to first characterize the adverse reproductive

finding in terms of dose–response or pathogenesis and, if possible,

link this to a functional as well as a morphological deficit.

Metabolic/pharmacokinetic studies can be undertaken to analyse

target tissue dosimetry, kinetics and metabolism (activation or

detoxification) at relevant dose levels (i.e., those that cause

adverse effects and that occur during environmental exposure). Armed

with these data, specific in vitro (and perhaps in vivo) experiments

can be designed to investigate biochemical perturbations in target

tissues with the compound (or appropriate metabolite) at the

appropriate concentration. Reasonable models should be developed and

then thoroughly investigated in a sensitive animal species; finally,

results should be compared in different species. In vitro assays

could be carried out in some species that were non-responsive in

vivo to the action of the agent. Investigation of the response of

human target tissue in vitro may also be appropriate, when such

studies are feasible.

5.2.2 Pharmacokinetics and pharmacodynamics

If the rate of development is an important criterion for the

pharmacokinetic pattern that best matches an exposure-related

outcome, this becomes especially important when data are

extrapolated from animals to humans, because the rate of development

is much slower in humans than in most laboratory animal species.

There is little information on the relative importance of Cmax

versus AUC for many environmentally relevant chemicals. This is

obviously an important factor that should be considered in risk

assessment. It is especially important when results are extrapolated

from one species to another or when results are compared that

involve different patterns of exposure. Therefore, test guidelines

for the study of environmental chemicals may need to ensure adequate

kinetic analysis.

5.2.3 Gene–environment interactions

The outcome of developmental exposures is influenced significantly

by the genetics of the organism. This concept was demonstrated

empirically in experimental teratology almost 50 years ago by

showing that the teratogenic response to identical dosages of a

corticosteroid was dependent on the strain of mouse used. Such

differences occur frequently in assessments of chemicals that are

conducted in multiple strains or species of laboratory animals. The

basis for these differences may be in the pharmacokinetics and

metabolism of the compound or may be pharmacodynamic in nature;

therefore, the underlying reason for interstrain/interspecific

differences may not be obvious from the reproductive toxicity study

results.

Gene–environment interactions are likely to be a critically

important factor accounting for the variability in human response to

a toxic insult (Autrup, 2000; Bobrow & Grimbaldeston, 2000). It has

been estimated that at least 25% of human structural abnormalities

have a multifactorial cause; this value could actually be higher,

given that the etiology for about half of all malformations is

completely unknown. Studies combining molecular biology with

classical epidemiological approaches have demonstrated the existence

of allelic variants for developmentally important genes that may

enhance the susceptibility of the embryo. For example, the

association between heavy maternal cigarette smoking (>10

cigarettes/day) and cleft lip and/or palate in the offspring is

marginally significant until an allelic variant for TGF-alpha is

considered. The combination of smoking and the uncommon variant for

the gene raises the odds ratio to a highly significant level (Hwang

et al., 1995; Shaw et al., 1996).

5.3.1.2 Postnatal manifestations

As discussed above, organ systems and the entire organism mature

slowly over a long period that extends well into the postnatal

period and up to puberty. Thus, organisms are at risk for exposure-

induced functional defects for a longer period than they are at risk

for structural malformation. Functional abnormalities have been

linked to exposure during the prenatal or early postnatal period.

However, functional abnormalities can be difficult to detect, and it

may be necessary to use specially designed functional or behavioural

tests for such changes. Pre- and postnatal chemical exposure can

affect neurological function, simple and complex behaviour,

reproduction, endocrine function, immune competence, xenobiotic

metabolism and the function of hepatic, renal, respiratory and

cardiovascular organ systems. A recent workshop evaluated critical

periods of vulnerability for various organ systems in the developing

organism (Selevan et al., 2000).

It is beyond the scope of this monograph to discuss all these

subjects; however, the interested reader is referred to major

textbooks on developmental (Kimmel & Buelke-Sam, 1994) and general

toxicology (Klaassen et al., 1996) and previous IPCS monographs

(IPCS, 1984, 1986a, 1986b). The methods used to assess the function

of these systems in prenatally exposed progeny are similar to those

conventionally used in toxicology. Specific methods that have been

developed to assess physical, neurological and behavioural

development are discussed in more detail below. Methods for the

study of reproductive function were discussed in chapter 4.

1) Methods for assessing behaviour

The term " behavioural teratology " was first used by Werboff &

Gottlieb (1963), who studied rats exposed prenatally to

tranquillizer drugs. They described the effects of these drugs in

treated rats as interference with " the behavioural or functional

adaptation of the offspring to its environment. " The term

behavioural teratology has since been used to describe postnatal

deficits induced by prenatal exposure to chemicals (Barlow &

Sullivan, 1975). As discussed below, there is now a wider focus on

broadly defined developmental toxicology and neurotoxicology,

including structural and functional effects detected both pre- and

postnatally. Permanent functional deficits may be caused by macro-

or microstructural defects or by changes in neurochemical and

neurotransmitter synthesis, storage and release or receptor

function. Behavioural changes may precede neuropathological changes

and provide a more sensitive indication of a chemical's toxicity

(Spyker, 1975; Weiss, 1975; Tilson & , 1980; IPCS, 1986b;

Kimmel & Gaylor, 1988; Tilson, 1990, 1998; Landrigan et al., 2000).

Exposure to chemicals during development can result in a plethora of

effects, ranging from gross structural abnormalities and altered

growth to more subtle effects (Spyker, 1975). The qualitative

measures of some injuries during development may differ from those

seen in the adult, such as changes in tissue volume, misplaced or

misoriented neurons, altered connectivity or delays/acceleration of

the appearance of functional or structural end-points (Rodier,

1986). In some cases, the results of early injuries become evident

only as the nervous system matures and ages (Vorhees, 1986; Rodier,

1990; Harry, 1994; Kimmel & Buelke-Sam, 1994). The specificity of

the damage may be a function of the timing of cell proliferation or

differentiation at the time when effects are expressed.

2) Methods of assessing development and function

Regulatory acceptance of behavioural tests to assess function became

evident when behavioural end-points were included in OECD and US EPA

toxicity test guidelines. These tests have been reviewed

extensively, and a number of them have been standardized and

validated (Buelke-Sam et al., 1985; Kimmel et al., 1990; Tilson et

al., 1997). The different classes of tests to assess function are

briefly summarized below:

•

Physical development: Growth and survival are important indicators

of normal function. From the point of view of screening, body weight

gain and deviations from a normal range of body weight at a given

time in development may be significant indications of developmental

toxicity ( & Buelke-Sam, 1981). Most physical landmarks

correlate so well with body weight that it may be unnecessary to

record the timing of physical landmarks such as pinna detachment,

hair growth, incisor eruption and ear and eye opening ( &

Palmer, 1980).

•

Reflex development: The timing of acquisition of different reflexes

is frequently measured (Smart & Dobbing, 1971). These include static

and dynamic righting reflexes, negative geotaxic response, auditory

startle reflex and grasping and placing reflexes. It is important to

ensure not only that all pups acquire the reflexes but that they do

so within a reasonable range of time, which has to be determined for

the individual species, strain and housing and nutritional

conditions, which all influence the rate of development of reflexes.

•

Sensory development: Several screening tests that detect overall

sensory deficits rely on orientation or the response of an animal to

a stimulus. Responses are recorded as present, absent or changed in

magnitude (Moser & MacPhail, 1989). Another approach to the

characterization of sensory function involves the use of reflex

modification techniques (Crofton, 1990). Changes in stimulus

frequency or threshold required to elicit a reflex or to induce

habituation indicate possible changes in sensory function.

•

Motor functions: The timing of the normal development of motor

functions has been described by Alder & Zbinden (1977). Spontaneous

activity can be assessed in a familiar home cage environment and

also in the more unfamiliar open field, which is also used to

provide much more than just locomotor information. Spontaneous

activity can be assessed for short periods or over longer periods by

automatic activity-measuring equipment. This approach can provide an

integrated assessment of activity during the night when rodents are

most active. Specific aspects of motor and sensorimotor coordination

performance are studied by elicited motor activity tests, such as

crossing a narrow path, climbing a rope or balancing on a rotating

rod. Analysis of swimming movements has also been useful. Standard

neurobehavioural methods are available for such tests (Holson et

al., 1990; ECETOC, 1992; IPCS, 2001a).

•

Cognitive development: Cognitive development is essentially defined

as the ability to learn or respond appropriately to environmental

change. Numerous methods are available for evaluating cognitive

function in laboratory animals. Many reviews of these methods have

been published, along with examples of chemicals that affect

cognitive development (IPCS, 1986a, 1986b; ECETOC, 1992; Tilson et

al., 1997; US NRC 2000).

5.3.3.2 Outcomes measured in infancy and childhood

Humans mature slowly, so they are at risk for functional defects for

an extended period after birth. Outcomes observed in humans include

changes in growth, behaviour and organ or system function and

development. Cognitive, neurological, motor and sensory evaluations

are used, and reproductive function is evaluated. All of these are

vulnerable to the effects of toxicants. Childhood cancer is a

specific end-point that is rare but possible. The critical exposure

window for an adverse outcome will vary depending on the chemical

exposure (Rogan et al., 1986; son & son, 1996). There is

limited evidence in humans that exposure of one of the parents prior

to conception of the progeny could also result in an adverse outcome

(Aschengrau & Monson, 1989; Jarrell et al., 1996).

The lack of data on environmental exposure and postnatal effects

reflects the enormous complexity of documenting such changes in

children. Methods in developmental toxicity assessment must reflect

this diversity of postnatal functions. The studies are expensive

because they are generally prospective and longitudinal; that is, a

group is recruited and then followed over time to observe its

development. son & son (1996) have reviewed methodological

issues associated with the design of prospective, longitudinal

developmental studies. Standardized developmental scales must be

adapted for specific countries and cultures. The selection of

appropriate testing methods and conditions is very important when

assessing children because of shorter attention spans and increased

dependence on parental and environmental supports. The end-points

frequently used to assess developmental neurotoxicity in exposed

children have been reviewed by Winneke (1995); this is an important

area, because the brain is very vulnerable to insult over a long

period of time (Weiss, 2000; IPCS, 2001a). In addition, because of

the increasing complexity of functional capabilities during early

development, only a few tests appropriate for infants can be

readminstered to older children.

Exposure patterns as well as developmental characteristics change as

the child matures, and this must be taken into account. Both

biological and behavioural changes affect the potential for exposure

(reviewed in Cohen-Hubal et al., 2000). For example, small children

mouth toys that may contain harmful chemicals. Children's diets are

different, including liquid intake. Because of differences in

metabolism, they may reach higher levels with a given exposure than

those for adults. This combination of exposure and outcome

complexities makes assessment of childhood developmental toxicity an

extremely difficult endeavour.

6. RISK ASSESSMENT STRATEGIES FOR

REPRODUCTIVE TOXICITY

6.1 Introduction

Risk assessment is an empirically based process that estimates the

risk of adverse effects from exposure of an individual or population

to a chemical, physical or biological agent. The OECD test

guidelines, the US EPA risk assessment guidelines and additional

risk assessment procedures for new and existing chemicals have been

published and put into use by many different countries in Europe,

the Americas and Asia (United Kingdom Department of Health, 1991,

1995; EC, 1994, 1996; Health Canada, 1994; IPCS, 1994; Hertel,

1996). A list of assessments produced by various national and

international agencies on specific chemicals is included in

ECETOC/UNEP (1996).

Risk is defined as the probability of adverse effects in an

organism, a population or an ecological system caused under

specified conditions by a chemical, physical or biological agent

(OECD/IPCS, 2001). The risk assessment process usually involves four

steps: hazard identification, dose–response assessment, exposure

assessment and risk characterization (US NRC, 1983; WHO, 1999). The

first two components of the risk assessment process, hazard

identification and dose–response assessment, constitute the basic

toxicological evaluation. This evaluation is aimed at characterizing

the sufficiency and strength of the available toxicity data and may

indicate some level of confidence in the data. Each source of

information has its advantages and limitations, which determine

the " weight of the evidence. " Dose–response modelling may be

included, if data are available. The third component, exposure

assessment, estimates potential human exposure based on various

environmental and/or occupational scenarios. The integration of

human exposure and animal testing data with exposure assessment is

termed risk characterization and constitutes the final step in the

risk assessment process (Kimmel et al., 1986).

Risk management is the process that applies information obtained

through the risk assessment process to determine whether the

assessed risk should be reduced and, if so, to what extent. In some

cases, risk is the only factor considered in a decision to regulate

exposure to a substance. Alternatively, the risk posed by a

substance is weighed against social, ethical and medical benefits

and economic and technological factors in weighing alternative

regulatory options and making regulatory and public health

decisions. Risk management is purposely separated from the

scientific evaluation (i.e., risk assessment) for the following

reason: the scientific data should be fully evaluated in a context

free from the influence of non-scientific issues and pressures.

Relevant social, economic, political, public health or other issues

should be considered independently. The risk-balancing approach is

used by some agencies to consider the benefits as well as the risks

associated with use of the chemical. Additional sources of

information on reproductive toxicity risk assessment include IPCS

(1984, 1986a), ECETOC (1989), et al. (1995), EC (1996) and

(1997).

Regulatory agencies around the world have set standards over the

last three decades for limiting exposure to hazardous agents and

preventing reproductive toxicity. The regulations issued by these

agencies are based largely on experimental data on reproductive

toxicity. The approaches to assess the risk to reproductive health

include testing protocols in animals exposed to chemicals during

critical windows of the reproductive cycle. As described previously,

tests were initially designed to detect chemically induced

structural anomalies, but more recent strategies have been developed

to evaluate risk of functional deficiencies as well as structural

anomalies.

This chapter describes current strategies and approaches to

assessing developmental and reproductive toxicity and identifies

research needs to improve the scientific basis for risk assessment.

It is intended to provide the reader with an appreciation of the

complexity of reproductive toxicity risk assessment.

6.2 Testing strategies and protocols

Strategies and protocols for detecting reproductive toxicity differ

for different substances. Background information relevant to the

proposed tests and the purpose of the tests can also influence the

strategy or protocol used. Acceptable protocols also differ in

different geographic locations with different regulatory

authorities. OECD testing guidelines for chemicals were developed

and adopted by international agreement, and this has greatly

advanced the international acceptance of data produced in different

countries and laboratories (see section 6.3).