NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
National Research Council (US) Committee on Chemical Toxicology and Aging. Aging In Today's Environment. Washington (DC): National Academies Press (US); 1987.
The science of toxicology has a dual nature. It is the study of mechanisms by which environmental agents exert their toxic effects, and it is the empirical definition of the magnitude of toxicity of such agents and the risk they present to exposed human populations.* Toxicology, more than most disciplines of biology, depends on comparison of experimental results with large data bases acquired through the application of standard test protocols to large numbers of chemicals and types of radiation.
To study both the possible exacerbation of aging processes by environmental agents and the potential of increased toxicity for elderly people, mechanisms of aging must be considered with respect to their similarity to mechanisms of toxicity of known toxic agents. In addition, tests that will permit the screening of environmental agents for their potential to interact with aging processes or to present an increased hazard to the aged will need to be identified. Once such potentially interacting agents are identified, they must be evaluated in whole-animal systems that are appropriate surrogates for human response. The species selected must not only be appropriate with respect to aging, but must also be representative of human response for any particular chemical tested. Identification of common mechanisms and creation of test systems must both take place in the context of the existing data bases.
Some people view aging as a toxic process, and indeed some age-related functional changes mimic toxic processes. Aging is clearly associated with alterations in homeostasis and in organ and cellular integrity; many changes of aging resemble those induced by toxicants. Thus, aging might well be associated with alterations in metabolic states that lead to the body's formation of toxic molecules or to alterations in the normal regulation of concentrations of various natural substances in the body. Toxicologic research has found that even the most unlikely substances can be implicated in tissue damage if their concentrations deviate enough from normal.
Nutrition underscores the usefulness of considering aging as a toxic end point. For example, both deficiency and excess of vitamin B6 can lead to neurologic damage (and these changes superficially resemble those accompanying aging).
Another reason that aging can be usefully explored as a toxic process in its own right is that aging itself is probably the primary cause of or a cofactor in all age-associated disease. Furthermore, aging cannot be readily arrested and is present during any whole-animal experiment. Thus, all toxic processes induced by external agents can interact with aging processes, and the possible impact of this interaction would be greater with chronic exposures or toxic effects whose latency includes a substantial portion of the life span.
If aging is a toxic process, can it be mimicked by chemical or physical agents? The full answer to the question occupies much of this discussion, but the simple answer is that many agents present in the environment can, at high doses, induce some pathologic changes and physiologic deteriorations similar to those observed in the aged. This is a limited form of mimicry, because a given agent generally produces only a few aspects of aging in only one tissue (or at most several). Only a few agents, such as ionizing radiation, cause damage throughout the body.
The absence of universal mimicry of aging by toxic agents, even in a given tissue, is an important observation that must not be overinterpreted. A toxic agent might produce molecular lesions similar to aging-induced lesions. But the experience of toxicology is that the manifestation of molecular or cellular damage at the tissue level is not rigorously specific to the toxic agent. Diverse agents that are believed to produce the same kinds of damage at the cellular or molecular level sometimes induce different kinds of damage at the tissue level. Conversely, diverse toxic agents sometimes produce similar pathologic effects; that is, the processes that manifest some effects can be triggered by different agents.
All tissues deteriorate during aging. The observation of the breadth of the effects of aging requires a similarly broad toxicologic approach. Special attention must be directed toward evaluating data bases for long-term whole-animal bioassays, because these toxicologic evaluations have been used to evaluate the results of prolonged exposure to toxic agents at low concentrations, including environmentally important agents, and they constitute a major tool for quantitative risk assessment.
Neurotoxicology and immunotoxicology should also be emphasized. Most environmental agents have not been screened for neurotoxic and immunotoxic effects, and some neurologic and immunologic deficits of unknown etiology are observed in a large proportion of the aged.
From the toxicologic perspective, two general problems present the greatest concern in the consideration of environmental relationships with aging. First, the mechanisms of aging are unknown, so toxicologic methods of detecting perturbations in the essential biology of the aging processes are not readily feasible. Second, the empirical approach is thwarted because there are no comparative data to validate any test proposed to detect agents that specifically affect aging or the aged.
One approach to the two problems is to discern the mechanisms of aging and search for agents that specifically affect aging or the aged. Much effort has been expended on the former; great resources would be needed for the latter. One purpose of this report is to propose the most economical and scientifically sound plan for the overall approach to the two problems.
In addition to aspects of conventional toxicology that appear most relevant to the relationship of chemical toxicity and aging, three new types of questions must be addressed:
- How can agents that might interact with the aging processes to increase or hasten the appearance of pathologic effects or reduced resilience in the aged be identified? Certainly, the standard long-term bioassay could detect large age-associated pathologic changes, but it is not clear whether it can mimic exposures that are inherent in the human environment and absent from the pristine laboratory environment of the surrogate animal.
- How cam acute or subacute exposures in utero, during development, or during maturity that sensitize the exposed to other agents be identified? For example, in multistage carcinogenesis, exposure to a cancer initiator can sensitize animals to later exposure to a cancer promoter. The problem of whether similar mechanisms induce any of the wide spectrum of pathologic effects observed in the aged is extremely difficult to approach with standard toxicologic tests.
- How can environmental agents, particularly pharmaceuticals, that hold increased hazard for the aged be identified? Most toxicity is measured in young animals, so a more specific question is whether testing performed in young animals can predict the response of old animals. If the responses of young and old animals to a given agent are qualitatively similar, but quantitatively different, the application of a safety factor might permit the data from young animals to be modified for predicting risk in old animals. If the mechanism of action differs, such extrapolation would not be possible; testing with old animals, and the attendant expense and absence of comparative data, would then be necessary.
The remainder of this chapter summarizes the general concepts of toxicology (i.e., absorption, distribution, metabolism, and elimination), the effects of chemicals in the body as a function of age, mechanisms of toxicity at various levels of physiologic organization, and the importance of pharmacogenetics, biomarkers, and toxicity testing.
CHEMICAL FATE AND EFFECT
Chemicals enter the body by inhalation, ingestion, and contact with the skin. They can act at the local site of contact, or they can be absorbed, enter the bloodstream, and be transported to act at other sites. Toxic agents are eliminated from the blood by biological transformation and by excretion or accumulation at various sites. The liver is active in biotransforming toxic substances; however, enzymes in the kidneys, lungs, gastrointestinal tract, skin, and other tissues can also metabolize toxicants. Some toxicants accumulate in organs or tissues, where they might or might not produce toxic effects. Most toxicants are eliminated in the urine via the kidney or in the bile via the liver, although some volatile toxicants can be eliminated in the exhaled air via the lungs.
Most of the research on the effects of old age on the absorption, distribution, metabolism, and elimination of chemicals has focused on drugs, rather than on toxic environmental substances. Many toxic chemicals are converted to less toxic, or in some cases more toxic, chemicals by the same pathways that are responsible for the biotransformation of drugs, so data on drugs have relevance for toxicologic concerns. For information on aging and drug disposition, several comprehensive reviews of geriatric pharmacology are available (Greenblatt et al., 1982; Schmucker, 1978, 1985; Vestal and Dawson, 1985).
In contrast with the data on pharmaceutical agents, which are based on studies of both animals and humans, virtually all the data on toxicants have been obtained from experimental animals. Furthermore, as documented in a monograph by Calabrese (1986), the human data and many of the animal data are related to neonatal and young subjects. Of particular interest is the possible influence of age on the detoxification of carcinogens, on the conversion of procarcinogens to carcinogens, and on the interaction of carcinogens with potential inducing agents.
Age-related changes in carcinogen metabolism in young and old animals have recently been reviewed by Birnbaum (1987), who concluded that the available data were both conflicting and sparse. The results of metabolic studies depend on substrate, species, gender, and even strain. Age-related changes in the monoxygenase components of nonhepatic tissues have hardly been examined. Thus, although it has been tempting to hypothesize that age-dependent changes in carcinogen metabolism might contribute to the increased incidence of cancer with aging, this is an oversimplification. For specific compounds, biotransformation might differ with age in a given tissue, but the differences might not result in a greater concentration or longer duration of action of carcinogenic chemicals or reactive metabolites at vulnerable target sites.
Although many, but not all, animal studies have suggested that aging is not associated with a loss of the capacity of oxidative metabolism to respond to inducers—such as barbiturates, polycyclic hydrocarbons, and steroids—the question has received less attention from clinical investigators. In addition to metabolism, such toxicokinetic measures as absorption, distribution, and renal and biliary excretion and toxicodynamic determinants of tissue sensitivity can change with age. Thus, an increased carcinogen sensitivity in older persons might result from complex interactions of many processes that have not been investigated systematically in either humans or animals.
Awareness of the effects of age on the absorption, distribution, metabolism, and elimination of drugs can provide insight into the mechanism of altered response to chemicals in general. Studies of age differences in pharmacodynamics (biochemical and physiologic effects of drugs and their mechanisms of action) must take into account possible age differences in pharmacokinetics (absorption, distribution, metabolism, and elimination). For example, studies generally show that the elderly are more sensitive to the depressant effects of neuroactive drugs (such as diazepam) (Giles et al., 1978; Reidenberg et al., 1978) and the analgesic effects of narcotics (such as morphine) (Bellville et al., 1971; Kaiko, 1980); in contrast, the in vivo sensitivity of the heart to isoproterenol and its antagonist propranolol appears to decline with age (London et al., 1976; Van Brummelen et al., 1981; Vestal et al., 1979). There is also evidence that cellular biochemical responses to some drugs are altered with aging.
Absorption
Absorption is the major process by which toxicants are transported across body membranes. The main sites of absorption of toxic agents are the skin, lungs, and gastrointestinal tract. Many toxicants can be absorbed through the skin and enter the bloodstream. Chemical or physical injury and other circumstances can increase the skin's permeability. Toxicants that are absorbed by the lung are in the form of gases or solid or liquid aerosols. Absorption can be rapid and complete because the lungs have a large surface area and a blood supply that is close to inhaled air in the alveolae. A variety of environmental toxicants enter the food chain and are absorbed from the gastrointestinal tract. Many factors alter the gastrointestinal absorption of toxicants, including gastrointestinal motility, the physical and chemical properties of the toxicant, and gastrointestinal content.
A number of physiologic alterations associated with old age might be expected to affect absorption from the gastrointestinal tract. They include an increase in gastric pH, a reduction in intestinal blood flow, a reduction in the number of absorbing cells, a decrease in gastrointestinal motility, and a slowing of gastric emptying. An increase in gastric pH might affect the ionization and solubility of some substances, but there are few specific data on the range of pH values that can be encountered in the elderly population and the effects of increased pH on bioavailability. Older people have reduced gastrointestinal motility and slower emptying, which might be expected to decrease the rate of absorption.
Some studies have shown an increase in the time to peak plasma concentration after oral drug administration. This has only minor clinical importance because the extent of absorption did not differ between young and old subjects. Although the data are sparse, most studies on drug absorption in the elderly do not demonstrate a marked effect of age on the rate and extent of absorption.
Distribution
Once a toxicant enters the bloodstream, it is available for distribution throughout the body. Only free toxicants—those not bound to plasma proteins—are able to enter other sites. Such binding is of particular concern to toxicologists and medical scientists, because toxicants bound to those proteins can be displaced by other chemical agents and, once released, go to target organs and produce injury there.
The distribution of toxicants depends on their ability to cross cell membranes and on their affinity for various body components. Toxicants vary widely in these two characteristics. Some do not readily cross cell membranes and therefore have restricted distribution. Others bind to various sites in the body, such as fat, liver, kidneys, or bone. The major toxic action of a toxicant might take place where it binds, but often it does not. In fact, binding sites often serve as storage depots whose existence helps to protect the body from the toxic action.
A variety of age-related changes can alter the volume of distribution of substances throughout the body. Body composition is one of the most important. Total body water (both in absolute terms and as a percentage of body weight) is reduced by 10–15% between the ages of 20 and 80 years; lean body mass in proportion to body weight is also diminished with age, and body fat is increased. These changes can be predicted to cause higher blood concentrations of substances that are distributed mainly in body water or lean body mass. Alterations in body fat can result in the accumulation and prolongation of action of highly lipid-soluble substances.
In addition, a decrease in serum albumin in the aged means that greater amounts of substances that bind to serum albumin, such as the anticonvulsant phenytoin, will be free to diffuse into body tissue. In contrast, serum a1-acid glycoprotein is increased in the elderly, and that reduces plasma-protein binding of weak bases, such as the antidepressant imipramine and the antiarrhythmic drug lidocaine. Thus, because free-drug concentration is an important determinant of drug distribution and elimination, altered plasma-protein binding might be one cause of altered pharmacokinetics in the aged. Available data suggest, however, that disease and immobility have greater effects on albumin concentration than age itself.
Metabolism and Elimination
Some chemical agents that enter the body can remain as intact molecules, but many are biologically transformed by metabolic processes. Metabolic processes might involve simple and reversible chemical or physical interactions that primarily affect transport throughout the body and across membranes. In other cases, metabolic processes can substantially alter the chemical nature of the toxicant and create a more toxic or less toxic agent. The metabolic processes can facilitate elimination from the body.
It has been useful to consider the metabolic processes as being of two types. The first includes processes of oxidation, reduction, or hydrolysis that primarily alter or add functional (reactive) moieties to the molecule in question. The second includes chemical reactions of pre-existing or newly formed functional groups on the molecule with various endogenous chemicals (such as amino acids, sulfate, and glucuronic acid) to form conjugates, or new chemicals. The biosynthesis of these products often alters lipid or water solubility and ionization characteristics in ways that promote their secretion and excretion.
The major routes for elimination of chemical agents from the body are from the kidneys to urine, from the liver to bile to feces, and from the lungs to exhaled air. Minor routes include secretions from the body—such as sweat, tears, saliva, mucus, digestive juices, and milk—and hair, nails, and desquamated epithelial tissue. As mentioned above, such factors as age and disease state that interfere with kidney function or biliary excretion in the liver can affect the toxic potential of chemicals in the body.
The kidney's excretory mechanisms include filtration in the glomeruli and secretion and reabsorption in the renal tubules. Elimination via the kidneys is thus a function of blood flow to the kidneys, molecular volume relative to pore size of the glomerular filter, physicochemical characteristics of the molecule that affect membrane transport, and enzymatic or other systems that might activate or facilitate secretion and reabsorption. Chemicals that bind to large molecules, such as plasma proteins, might not be eliminated by filtration and might be retained in the body for long periods.
The liver is especially important as a route of elimination of chemicals that are ingested, because most of the blood from the gastrointestinal tract goes through the liver on its way to the general circulation. The liver is in a unique position to metabolize a chemical through its enzymatic systems and to secrete the metabolites into the bile. Bile empties into the intestines, where the chemical can either be further altered and reabsorbed or be eliminated in the feces. Injury to the liver often affects biliary function and impairs this route of elimination.
Elimination of chemicals from the body can be studied by pharmacokinetic measurements, which are often based on the remaining concentration of a chemical or its metabolites in the blood in relation to time. Such information usually provides a good estimate of the amount of the chemical available for toxic action. However, storage of the chemical in tissue depots or the toxicity of unmeasured, activated, intermediate chemical forms is sometimes more important.
Processes of metabolism and elimination can be altered in the elderly, but the evidence of altered hepatic drug metabolism in humans is indirect. Autopsy studies have demonstrated that liver mass in proportion to body weight declines after middle age and that liver blood flow decreases with increasing age. For some drugs, mainly those that undergo conjugation in the liver, there is no clear effect of age on metabolism. However, age appears to have a variable influence on the rates of metabolism of drugs that are oxidized in the liver; most of the wide interindividual variation in drug metabolism is more likely due to a variety of genetic and environmental factors. Although the available data indicate no effect of age on the inhibition of drug metabolism (Divoll et al., 1982; Vestal et al., 1987), the equally limited data on the susceptibility of the elderly to induction of hepatic drug metabolism conflict, some studies showing a decrease in the extent of induction (Salem et al., 1978; Twom-Barina et al., 1984) and others no age differences (Crowley et al., 1986; Pearson and Roberts, 1984). The effect of age on the induction and inhibition of drug metabolism, as well as other drug interactions, requires further investigation.
Cigarette smoke contains polycyclic hydrocarbons, which are potent inducers of some isozymes of cytochrome P-450. Most cross-sectional studies have indicated that cigarette smoking is associated with less induction of biotransformation in elderly than in young people. Whether this is intrinsic to aging or is the result of selective mortality could be established only by longitudinal studies, which have not been done. Nevertheless, the possibility of the greatly reduced capacity of some elderly patients to metabolize and eliminate drugs should be taken into consideration when prescribing drugs for the elderly. This can be done either by slightly reducing the dosage of potent drugs with low therapeutic indexes or by watching the patients very carefully, to ensure therapeutic efficacy of prescribed medications and to detect undesirable drug-related side effects early.
Studies in senescent experimental animals have shown reduced hepatic enzyme activity, with resulting reduced capacity to metabolize drugs and reduced hepatic enzyme induction. Most of the data have been acquired in rats and mice, and the apparent age-related changes might not be universal; species, strain, and sex differences have been important variables in rodent studies. Studies with liver tissue from nonhuman primates have not shown a significant decline in the content of cytochrome P-450 or in the specific activity of NADPH cytochrome c (P-450) reductase (Schmucker and Wang, 1987). Similarly, in vitro studies with human liver tissue (also limited) have shown no effect of age on microsomal drug-metabolizing activity (Woodhouse et al., 1984). The results conflict with those obtained in rodents, and to some extent they conflict with in vivo studies in humans. (In vivo studies in nonhuman primates have not been performed.) They do emphasize, however, the difficulties in extrapolating observations made in experimental animals to humans and the need for clinical investigation to evaluate the metabolism of drugs and chemicals in humans.
One important chemical that is widely used as a drug by persons of almost all ages is alcohol. Alcohol is distributed into total body water. In both rodents and humans, its volume of distribution decreases with age (Vestal et al., 1977; Wiberg et al., 1971). That decrease results in higher blood concentrations after equivalent doses in the elderly than in the young. Acute alcohol exposure inhibits and chronic exposure induces oxidative drug metabolism in the normal liver. Alcohol itself is oxidized predominantly by alcohol dehydrogenase, a cytoplasmic hepatic enzyme, and to a lesser extent by the hepatic microsomal enzyme system. Although the elderly are more sensitive to the behavioral and cognitive effects of alcohol, studies have not demonstrated age differences in alcohol metabolism in humans (Vestal et al., 1977). Age has been shown to influence alcohol metabolism in rats (Wiberg et al., 1970).
Dietary composition is an important environmental determinant of drug metabolism and drug toxicity (Alvares et al., 1979; Campbell and Hayes, 1974). Most studies have been conducted in experimental animals (Campbell and Hayes, 1974). Studies in healthy human volunteers have shown that a low-carbohydrate, high-protein diet (Kappas et al., 1976) and charcoal-broiled beef (Kappas et al., 1978) increase the metabolism of antipyrine and theophylline, and dietary brussels sprouts and cabbage (Pantuck et al., 1979) increase the metabolism of antipyrine and phenacetin. Charcoal-broiled beef contains benzo[a]pyrene and other polycyclic hydrocarbons (Lijinsky and Shubik, 1964), which stimulate the metabolism of benzo[a]pyrene in rat liver and placenta (Harrison and West, 1971). Brussels sprouts, cabbage, turnips, broccoli, cauliflower, and spinach induce benzo[a]pyrene hydroxylase in the rat (Wattenberg, 1971). Indol compounds in cabbage and brussels sprouts stimulate the metabolism of phenacetin, hexobarbital, and 7-ethoxycoumarin by rat intestine (Loub et al., 1975; Pantuck et al., 1976).
The extent to which the elderly might differ from younger adults in their response to manipulations of carbohydrate, protein, polycyclic hydrocarbons, cruciferous vegetables, and other dietary constituents requires investigation. Some attention has been given to clinical micronutrient-deficiency states. Treatment of ascorbic acid deficiency was associated with an increase in the plasma clearance of antipyrine in elderly patients admitted to a geriatric ward, but there was no difference in basal values between deficient and nondeficient patients, and there was no effect of ascorbic acid treatment in the nondeficient group (Smithard and Langman, 1978). In another study, changes in dietary ascorbic acid did not affect caffeine metabolism in the elderly (Trang et al., 1982). Although poor nutrition and dietary habit might contribute to the complex changes seen in hepatic drug metabolism in old age, the available data are inadequate to substantiate such a conclusion (Cusack and Denham, 1984).
In contrast with hepatic function, diminished renal function is common and easily measured in the elderly. Studies have indicated that both glomerular function and tubular function are affected by aging. Glomerular filtration rate, as measured by inulin or creatinine clearance, can fall by as much as 50%. Renal plasma flow declines by approximately 2% each year. The decline in renal function decreases the rate of elimination of drugs that are excreted unchanged by the kidney. In clinical medicine, reduction in the maintenance doses of drugs, such as the aminoglycoside antibiotics and the cardiac glycoside digoxin, is often necessary to prevent toxicity.
The decline in renal function is quite variable; some people exhibit little or no apparent decline (Lindeman et al., 1985). Therefore, in clinical practice it is essential to measure plasma concentrations of potentially toxic drugs and adjust doses accordingly to achieve therapeutic concentrations.
Not only geriatric patients, but fetuses, neonates, and children show rates of drug elimination that differ from those in normal adults and that can vary greatly with drug and person. For example, antipyrine and phenylbutazone elimination was approximately twice as fast in children 1–8 years old as in normal adults (Alvares et al., 1975). Although eight children without clinical symptoms of acute or chronic lead poisoning, but with biochemical manifestations, had the same rates of antipyrine and phenylbutazone elimination as normal children, two children with both clinical and biochemical signs exhibited increased plasma antipyrine half-lives. The half-lives returned to the normal range for children when chelation therapy was instituted. Similar results have been obtained in other studies that disclosed more rapid elimination of diazoxide (Pruitt et al., 1973), phenobarbital (Garretson and Dayton, 1970), and clindamycin (Kauffman et al., 1972) in children than in adults.
Human fetuses and neonates have much lower capacity than children or adults to eliminate drugs, although human fetuses and neonates, in contrast with those of some rodents, exhibit measurable hepatic microsomal drug-metabolizing activity (Pelkonen et al., 1973; Rane and Sjoqvist, 1972). In premature infants, the rates of elimination of some drugs appear to be even lower than those in full-term healthy newborns. For example, plasma indomethacin half-lives were 2 and 24 hours in two premature infants (Friedman et al., 1978), whereas the mean value was 4.7 hours in full-term healthy newborns (Traeger et al., 1973) and 2–3 or 7.2 hours in healthy adults (Duggan et al., 1972; Hucker et al., 1966; Palmer et al., 1974). Obviously, the marked differences in elimination rates among premature infants, full-term newborns, children, and adults have practical significance in the calculation of appropriate dosages. They also indicate the need to investigate pharmacokinetic and pharmacodynamic processes for many more drugs with respect to various age groups.
All the characteristics described here are important in studying the effects of environmental chemicals on aging. If the effects of chemicals on aging mimic other human health effects, the members of an exposed human population will not all react in the same way. The wide genetic heterogeneity of the human population will ensure differences in responses to environmental chemicals even among members with comparable exposures.
MECHANISMS OF TOXICITY AT THE MOLECULAR, CELLULAR, AND TISSUE LEVEL
The mechanisms by which toxic agents exert their actions are extremely diverse. Such mechanisms vary between specific agents and between doses of a given agent. Mechanisms of toxic action can be characterized by the dose at which the essential toxic effect is induced and by the specificity of that effect. Such a characterization is probably most relevant for the consideration of aging, in that the age-associated functional decline in physiologic systems and the lack of specificity as to cells and tissues have been characterized for elderly people, even though the exact mechanism of aging is still unknown.
Molecular Action
The molecular action of toxic agents can result from the creation of damaging chemical species that attack specific moieties of biologic molecules. The search for toxic mechanisms that are shared by aging and by specific toxic agents should include agents whose action is very broad, because aging broadly affects all tissues. Breadth of action can result from different types of mechanisms, however. Equally desirable would be the discovery of agents whose molecular mode of action influenced a specific aspect of the senescent phenotype, inasmuch as this would allow a finer dissection of mechanisms.
Damage can be induced in a specific molecule whose altered function can result in a broad range of toxic action. Damage to DNA with pleiotropic effects is an example. General toxic effects can also take place through agents whose molecular damage is by the production of a general damaging species. For example, ionizing radiation and many chemical agents produce free radicals either directly or by the action of intrinsic metabolic processes. Such free radicals, themselves or through a cascade of processes, can interact with many biomolecules to produce types of damage having different potential for biologic effect.
Furthermore, one tissue can be susceptible at one magnitude of exposure and other tissues at other magnitudes. Using ionizing radiation as an illustration, x rays produce a variety of ions and radicals in irradiated tissue. In exposed humans, high doses of x rays (over 100 Gy) can induce central nervous system collapse within minutes or hours. At lower doses (10 Gy), the dominant toxic effect occurs by inhibiting proliferation of the stem cells of gut crypts; the inhibition denudes the intestinal mucosa and leads to death within days. At still lower doses (1 Gy), x rays not only derange the hematopoietic system, but also increase cancer in irradiated populations. At very low doses (below 1 Gy), x rays produce both excess cancer and heritable mutation; cancer is considered the greater risk. The responses to the lower exposures have been observed in both epidemiologic and animal studies.
As discussed (Lindop and Rotblat, 1961; National Research Council, 1972; Sasaki and Kasuga, 1981), x rays produce life-shortening, as well as the other toxic effects. Thus, they provide an example of an agent that produces a nonspecific toxic free radical, which in turn attacks susceptible biomolecules. The biologic response, however, can be specific to various tissues, the dominant effect depending on the exposure dose.
Another principle of toxicology, that of limiting toxicity, is involved in the preceding example. Animals that succumb to very high doses of x rays would also suffer other toxic effects if not limited by the more severe and immediate effects. In considering the possible effects of any toxic agent on aging and the aged, we must view such an induced response as part of a series of toxic actions and as depending on dose, mode of exposure, duration of exposure, species exposed, and period of observation.
Other toxic agents are probably more specific in their initial actions and in their effects. For example, carbon monoxide binds to hemoglobin, reduces the oxygen-carrying capacity, and produces toxicity.
A final consideration of damage at the molecular level that could result in the pleiotropic manifestations observed in aging is whether a specific toxic action can be manifested in more general ways. It has been proposed that a specific action of a specific tissue can result in multiple deteriorative processes in various tissues. Although alterations in molecules that control regulatory processes can have multiple manifestations or can affect multiple tissues (e.g., in diabetes), no example of induced response equal in breadth to the manifestations of aging has been observed.
Cellular Effects
Toxic mechanisms can act at the level of the cell in two general ways: specific tissue functions can fail if enough cells of a specific type in that tissue are altered by the exposure, and toxic stress of a single cell can produce some toxic effects (as in cancer), heritable mutation, and teratogenesis. Aging must be considered a phenomenon in which most cells in most tissues can be discerned to be altered, so it seems logical to consider aging as similar to the first cellular effect mentioned, that is, as acting through deterioration in most cells of a given tissue.
The results of some aging studies have indicated that the cell is the essential unit of aging, with manifestations at the tissue, organ, and organism levels being sequelae of cellular deterioration. Although aging research has been directed toward determining the molecular damage or change that underlies cellular aging, it has been unsuccessful in establishing a molecular etiology. Thus, in a search for toxic agents that mimic aging, comparisons at the level of the cell and tissue are appropriate.
Effects at the Tissue Level
Toxic action in intact animals is most often characterized at the tissue level because altered tissue structure or function is commonly observed. Such specificity might be due to toxic stress at the site of exposure (e.g., lung, skin, or gastrointestinal tract), at the site of metabolic action (e.g., liver, brain, or kidney), or in susceptible target cells. It is important, however, in understanding the toxic mechanism acting at the tissue level to determine its etiology at the cellular and molecular levels.
To return to the example of ionizing radiation, the induced failure of the lining of the gut to perform its barrier function is the direct result of failure of stem cells to proliferate, which causes sloughing of the intestinal mucosa. The failure of the stem cells to proliferate is believed to proceed from the action of x-ray-induced free radicals in their chromatin through mitotic inhibition. Thus, when considering toxic effects related mechanistically to aging, the basic goal must be to identify the sequence of events from molecular changes to their sequelae at the level of the cell, the tissue, and the organism.
PHARMACOGENETICS
In the course of investigations of drug metabolism and drug disposition in humans, striking individual differences in response to drugs and in ability to metabolize and dispose of drugs have been noted. Some differences are due to environmental factors, some to age-dependent (developmental) factors, others to genetic factors, and many to complex interactions among those factors. The scientific study of genetic factors that account for individual differences in drug metabolism and drug response is called pharmacogenetics. Pharmacogenetics is of toxicologic importance in that it reveals how some people, because of their genetic constitutions, suffer toxic effects on exposure to xenobiotics at doses well tolerated by other people.
Of particular interest is the relationship between genetic differences in rates of metabolism among subjects and the synthesis of potentially toxic biotransformation products of a parent drug or other chemical. In the past, the hepatic drug-metabolizing enzyme system has been regarded as a detoxification system, because it converts lipid-soluble compounds that could otherwise remain in the body indefinitely to more polar metabolites that are readily excreted in urine. More recently, however, it has been recognized that this enzyme system can produce potentially toxic, highly reactive metabolites that combine with tissue macromolecules, including DNA, to produce necrosis, immunologic reactivity, and mutations (Sipes and Gandolfi, 1986).
Qualitative differences among subjects in pathways of drug metabolism and quantitative differences in the activities of the enzymes that catalyze those reactions and pathways could be involved in the regulation and control of such tissue damage. Thus, genetic differences among subjects can render some more and others less sensitive to the toxicity of different reactive metabolites.
As genetic entities are investigated in detail, the effects of age on their expression often become apparent. Expression of a phenotype, and hence genetically modified (increased or decreased) susceptibility to chemical agents, might occur only when a person bearing genes that predispose to a particular kind of toxicity is first exposed to the offending environmental chemical, and that might not happen until late in life. Drugs are prescribed more commonly to old than to young people, so the incidence of pharmacogenetically related drug toxicity might be expected to increase with age. In addition, some genetically determined conditions are expressed only relatively late in life, such as Huntington's chorea and some forms of neuromuscular disease and diabetes mellitus.
Many factors have been systematically investigated and identified as contributing to the large interindividual variations that characterize disposition and response to xenobiotics in humans (Figure 4–1). The factors include sex, time of day or season of drug administration, presence of disease, hormonal and nutritional status, stress, exposure to activators or inhibitors of the hepatic microsomal drug-metabolizing enzymes (including chronic administration of any of several hundred drugs), the status of the heart, liver, kidneys, and endocrine organs—and age (Conney et al., 1971; Gillette, 1971; Vesell, 1982b).
FIGURE 4–1
Schematic depiction of established or suspected environmental factors that can alter genetically controlled rates of drug elimination. Lines from each environmental factor to a central circle are wavy, to suggest that modification of genetically controlled (more...)
In the past 20 years, genetic factors that directly affect xenobiotic response and disposition in humans have been discovered (La Du, 1972; Omenn and Gelboin, 1984; Omenn and Motulsky, 1978; Vesell, 1969, 1971, 1973, 1984; Weinshilboum, 1984). Some 60 factors have been identified, and many more probably exist. In some people under some conditions, genetic factors are the major or even sole cause of such interindividual differences. For example, when age, sex, diet, and exposure to environmental chemicals that activate or inhibit the hepatic drug-metabolizing enzyme system remain constant among human subjects, large interindividual variations in response to and disposition of xenobiotics remain. Many of these variations have a genetic basis.
About 20% of patients in teaching hospitals in the United States are there for treatment of adverse drug reactions, and 5– 30% of the patients in these hospitals have at least one such reaction (Cluff et al., 1965). Adverse reactions to drugs occur more frequently and with greater severity in old than in young patients (Hall, 1982). The normal wide disparity in individual patients' responses to drugs is only one cause of adverse reactions, including drug toxicity, but it constitutes an important contribution to this major medical problem.
Table 4–1 lists 10 of the best-known and most intensively investigated pharmacogenetic conditions. In almost every one, a toxic response ensues owing to drug accumulation secondary to a marked reduction in the enzymatic conversion of the parent drug to pharmacologically inactive metabolites. The function of the enzyme is aberrant because of a point mutation in the gene that controls its synthesis. Like most other inborn errors of metabolism, the conditions listed in Table 4–1 are generally transmitted as autosomal recessive traits. Thus, affected subjects inherit a mutant allele from each of two phenotypically normal parents.
TABLE 4–1
Monogenic Metabolic Pharmacogenetic Conditions and Putative Aberrant Enzymes.
Geographic differences exist in the gene frequencies of several pharmacogenetic conditions. Age effects have also been identified for the acetylase polymorphism, but additional studies of other pharmacogenetic entities in Table 4–1 need to be performed with respect to age.
The history of the discovery of an age effect on the gene frequency of the acetylation polymorphism is particularly instructive, because it illustrates how careful the search for such age effects must be. In the original observations of Evans et al. (1960), age was not recognized as an important factor. Later, study of a small number of subjects seemed to confirm that observation (Farah et al., 1977). But in 1983, Evans re-evaluated his original data and reported an effect of both age and sex on plasma isoniazid concentrations (Iselius and Evans, 1983). In 1984, the proportion of slow acetylators was found to be significantly higher in older people, but no sex distributions were reported (Gachalyi et al., 1984). In 1985, the age effect on the acetylator phenotype was confirmed, but stated to occur only in males (Paulsen and Nilsson, 1985).
The increased frequency of the slow-acetylator phenotype with age is of special interest, because the phenotype is also associated with a markedly increased susceptibility to the development of bladder cancer on chronic industrial exposure to arylamines and hydrazines (Cartwright et al., 1982). Another association between susceptibility to cancer and a pharmacogenetic phenotype has been claimed: extensive metabolizers of debrisoquin were stated to be more common in patients with bronchogenic carcinoma than in age- and sex-matched controls (Ayesh et al., 1984).
One implication of those variations among normal people is that a given dose of a drug administered by a given route can be toxic in one subject, therapeutic in another, and without pharmacologic effect in a third. The existence of these differences presents a formidable challenge to a physician who must individualize therapy (especially for drugs with low therapeutic indexes).
The therapeutic ramifications of large interindividual variations in the disposition of many drugs make it necessary to identify the mechanism responsible. Such identification is beset with difficulties rooted in the extreme genetic and environmental heterogeneity of human beings. In laboratory animals such as rats and mice, however, heterogeneity can be controlled. Each variable can be manipulated independently, the quantitative contribution of each to interindividual variations in drug disposition can be studied, and dose-response curves can be constructed. In the last decade, such studies have revealed many factors that can affect drug disposition. In humans, pharmacogenetic conditions can be categorized into those that affect how the body acts on drugs (pharmacokinetic conditions) and those that affect how drugs act on the body (pharmacodynamic conditions) (Vesell, 1973).
Because most of the monogenetic (simple, single-gene) conditions mentioned above are rare and make only a few drugs toxic, they probably contribute in only a minor way to the major medical problem of adverse drug reactions. However, another development in pharmacogenetics suggests that genetic differences that directly affect xenobiotic disposition play a prominent role in commonly encountered forms of drug toxicity. Large interindividual variations that existed among unrelated people in response to phenylbutazone, bishydroxycoumarin, antipyrine, halothane, ethanol, phenytoin, nortriptyline, or salicylate were absent in pairs of monozygotic twins, but present in most, but not all, pairs of dizygotic twins (Vesell, 1973, in press). The magnitude of interindividual variations in rates of drug elimination among unrelated people was a factor of 30 for nortriptyline, 10 for bishydroxycoumarin, 6 for phenylbutazone and antipyrine, 3 for halothane, and 2 for ethanol.
The existence and operation of many environmental factors— each with a different capability of altering the basal, genetically controlled rate of drug disposition—make it difficult to attribute portions of the total interindividual variation to specific environmental factors. The task of partitioning the total variation in drug elimination among large heterogeneous populations is further complicated by the close association of such seemingly pure environmental factors as smoking and diet with other environmental factors, as well as with genetic factors.
Many environmental, developmental, nutritional, or endocrine factors can influence the rate at which a person eliminates a drug. In Figure 4–1 such factors are connected because they are often associated with each other in a given subject rather than being independent. In fact, they often interact dynamically to change a subject's characteristic basal rates of drug absorption, distribution, metabolism, excretion, or receptor interaction. Accordingly, the effects of each of these factors on drug response can be complex and can change with time even in the same subject (Vesell, 1980, 1982a,b).
BIOLOGIC MARKERS
A biologic marker is a biochemical, cellular, structural, or functional indicator of an event in a biologic system or sample. Biologic markers in humans, animals, or other biota can serve as measures of exposure to or injury by a xenobiotic by indicating internal or circulating dose, stored body burden, dose at a target tissue, or the early onset of a pathologic effect.
The concept of biologic markers grew out of cancer research that sought to identify the role of exogenous agents or host factors as causes of human cancer. Perera and Weinstein (1982) defined molecular cancer epidemiology as an approach that combined analytic epidemiology and molecular techniques to identify carcinogens in human tissues, cells, or fluids and to measure early morphologic, biochemical, or functional responses to carcinogens. Lower and Kanarek (1982) described molecular epidemiology as the measurement of molecular characteristics related to neoplastic disease.
Since those early papers, several symposia and workshops have been conducted to examine the use of biologic markers in disease prevention. The ultimate goal of marker research is to improve the predictive relationship between exposure, dose, and response. A more thorough understanding of the role of markers will help to prevent disease by more precisely assessing the magnitude of risk, identifying high-risk groups or individuals, and providing early warning of disease.
The development of cellular and molecular markers extending from exposure to the development of disease should provide a powerful tool for the environmental health sciences. Markers that indicate the presence of internal or biologically effective doses or of incipient disease can be useful in hazard identification, that is, as the qualitative step that causally associates an environmental agent with an adverse effect. Markers can also be used to determine dose-response relationships, particularly at the low doses relevant to most environmental chemicals. A major role of markers is to clarify the extent of human exposure in populations, the extent of individual exposure, and the proportion of high responders or outliers among the human population (Fowle, 1984).
The use of biologic markers hats raised a number of important ethical issues (Ashford, 1986). Among these is the concern that biologic screening could encourage a shift from environmental monitoring to human monitoring in the workplace. There is concern that detection of a susceptibility marker, for example, could be used to exclude a person from employment, and that focusing on detecting susceptible populations and excluding them from the workplace could replace efforts to remove toxic chemicals from the workplace.
Markers can be distinguished on a continuum of time as markers of susceptibility, exposure, circulating internal dose, biologically effective dose (or dose at receptor site), and potential or actual health impairment.
Markers of susceptibility indicate individual or population physiologic differences that affect response to environmental agents, regardless of exposure. They include differences in receptors, in metabolism, in immunoglobulins, or in organ reserve capacity or other variations that lead to altered response to environmental agents, including sex, age, physiologic state, and even diet. For example, the absence of the enzyme α-1-antitrypsin is a marker of susceptibility to chronic obstructive pulmonary disease.
Markers of exposure are biologic events or conditions that reveal information about external exposure, internal absorbed dose, or dose at the receptor site or site of toxic action. Markers of internal dose—indicating the amount of a material that is absorbed into the organism—include such pharmacokinetic characteristics as blood flow, capillary permeability, transport mechanisms, number of receptor sites, metabolism of the material, and route of administration. Additional factors include data on structure and stability of the material, peak or cumulative circulating dose, and half-life (Gibaldi and Perrier, 1982). Markers of biologically effective dose include such target-organ characteristics as rates of metabolic activation and detoxification, pre-existing susceptibilities, and reserve capacity.
Markers of potential health impairment include early biologic responses, such as alterations in the functions of target or nontarget tissue shortly after exposure. Later in the course of response to a toxicant or after accumulation of high doses, markers of health impairment include altered function of the affected tissue that could be considered a preclinical state of disease.
Exposure to environmental pollutants can lead to uptake of a biologically effective dose and ultimately to reversible or irreversible injury. Different types of markers can be used differently to determine exposure, dose, or health impairment. However, markers of these types are not always distinct from each other. Thus, markers of exposures and markers of effects are often difficult to differentiate.
Because the body responds to injury with only a limited number of biochemical or cellular changes, an effect marker might not always be specific for an individual pollutant.
TOXICITY TESTING
Toxicity testing is undertaken for two general purposes: to characterize a particular chemical or physical agent so as to determine its general or specific toxic properties, and to screen a large number of agents for their likelihood of producing particular toxic effects. In both cases, practical and scientific considerations influence decisions about the nature and extent of testing. Each decision entails a scientific judgment that is based both on known toxic mechanisms and on data from tests of many chemicals with a series of defined protocols. Practical restraints on resources and time limit the testing of any particular agent. The complete testing of an agent in the wide range of available standard protocols could cost tens of millions of dollars. Few agents are theoretically or practically important enough to justify such expenditures.
Toxicity-testing protocols can be divided into screening or auxiliary tests and whole-animal tests. Screening or auxiliary tests are usually short-term tests designed to identify a potential for inducing specific toxic effects. Whole-animal tests can be short term or long term and are designed to confirm or deny potential toxicity and to assay risk quantitatively.
Two characteristics of screening tests determine their utility. The first is whether the biologic process assayed by a test is identical with the process involved in the toxic effect of concern. For instance, is mutation as observed in the Salmonella typhimurium test a necessary and sufficient process in some or all types of carcinogenesis for which the test is used as a screen? The answer to this specific question about the Ames assay is probably no; other processes not detectable by this test are also relevant to the induction of cancer. Regardless, the Ames assay is a useful test because it has the second characteristic: predictive value.
Predictive value is the ability of a test to predict the outcome of another test or human response. It is a probabilistic measure of correlation between the outcome of the simple test and the “truth” as defined through whole-animal testing. Predictive value is independent of essential biology. If, however, a short-term test has power both in mimicking essential biology and in predicting whole-animal response, it approaches the ideal.
In designing screening tests from the toxicologic perspective of detecting agents that accelerate aging, the same criteria apply:
- Does the proposed test mimic the essential biology of aging?
- Does the proposed test detect agents that accelerate aging when tested in whole animals?
These questions accent the problems of designing screening tests for agents that accelerate or interact with aging.
Whole-animal experiments, particularly long-term experiments, are most predictive of human response and quantitative risk. The species tested should represent human populations in exhibiting similar toxic responses and in mimicking human absorption, metabolism, and biodistribution of the chemical agent being tested. Although all animals age, a particular species might be inappropriate as a surrogate for human metabolic response to a particular agent. The selection of species for whole-animal, long-term testing related to aging must serve the dual demands of the gerontologist and the toxicologist, and for the toxicologist that species will change, depending on the specific agent that is to be tested.
Laboratory tests must, for practical reasons, be performed with relatively small numbers of animals. Tests for environmental agents are generally performed at dosages much higher than those to which human populations are exposed. Data from such tests must be extrapolated to predict human responses to dosages experienced in the environment. This extrapolation requires measurement (or at least prediction or assumption) of the shape of the dose-response curve. Toxic effects are often classified according to whether there is a threshold dose (below which no toxic effects will be observed), or whether at low dosages toxic effects will be produced in only a small proportion of exposed animals. This important distinction must also be considered for toxic interactions with aging.
The measurement of dose-response relationships for reduction in life span is difficult. Dose-response curves for life-shortening are complex and not amenable to simple extrapolation to low doses. One test of the hypothesis that radiation exposure induces premature aging would be the demonstration that the onset of all diseases is advanced to the same extent and by a factor related to the degree of the shortening of the life span (National Research Council, 1980b).
Another extrapolation that must be considered is that from the laboratory to the real environment. Laboratory tests have most commonly been performed with protocols wherein a single agent is tested for its toxic effect; however, the human environment contains multiple agents that can interact. In addition, there is little evidence that interaction occurs between agents at the low concentrations often found in the environment, but it is important to consider the possibilities of such interactions. Interactions of multiple agents are now being considered for their toxic effects, and the results might guide similar efforts to study age-related effects.
Footnotes
- *
For detailed descriptions of general toxicology, see Hayes (1982) and Klaassen et al. (1986). More specifically, Williams et al. (1987) have discussed the toxicologic perspective of the relationships between aging and the environment.
- Principles of Toxicology in the Context of Aging - Aging In Today's EnvironmentPrinciples of Toxicology in the Context of Aging - Aging In Today's Environment
Your browsing activity is empty.
Activity recording is turned off.
See more...