Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
3 Systemic Exposures to Volatile Organic Compounds and Factors Influencing Susceptibility to Their Effects When evaluating health effects of chemicals, it is important to understand how they enter and are distributed in the body (systemic exposure) and how the body handles them and their metabolites. This chapter reviews general issues related to evaluation of exposure to volatile organic compounds (VOCs) in that context. It considers characterization of differences between laboratory animals and humans and im- plications for the interpretation of the animal-toxicology literature that is presented in Chapter 4, identifi- cation of human populations that might be more susceptible than others to the effects of the primary con- taminants of concern, and interactions that might result from exposure to mixtures of chemicals. VOCs are the focus of this chapter because the primary water contaminants at Camp Lejeune and specified in the study charge are in this class of compounds. As noted in Chapter 2, other contaminants have been detected in the water supplies, so exposures were more complex than just VOC mixtures. However, for the purposes of this report, the review has been restricted to the primary VOC contaminants of concern. ENVIRONMENTAL CONTAMINATION The major drinking-water contaminants of interest at Camp Lejeune are volatile organic chemi- cals (VOCs), mainly trichloroethylene (TCE) and tetrachloroethylene (perchloroethylene, PCE) but also vinyl chloride, methylene chloride, benzene, toluene, cis- and trans-1,2-dichloroethylene (DCE), and 1,1- DCE (see Chapter 2). All those except benzene are halogenated, short-chain aliphatic hydrocarbons (halocarbons); benzene is an aromatic hydrocarbon. The water solubility of these compounds increases with decreasing numbers of carbon or halogen atoms. The maximum water solubilities of PCE and TCE at 25Â°C, for example, are 150 and 1,366 mg/L, respectively. Volatility increases with decreasing molecu- lar weight, varying from 18.5 mm Hg for PCE to 74 mm Hg for TCE at 25Â°C (ATSDR 1997b,c). Widespread use of TCE and other VOCs has resulted in their frequent escape into the environ- ment (Wu and Schaum 2000). Figure 3-1 illustrates the pathways by which environmental media are con- taminated and how people may be exposed. Most VOCs that enter the environment do so by evaporation during their use or discharge. Concentrations in air in the immediate vicinity of point sources may be high, but winds rapidly dilute and disperse the vapors (from nondetectable to nanograms per cubic meter of air). Migration of VOCs from subsurface soil or groundwater into the air in basements (vapor intru- sion) also occurs. There does not appear to be a wide-scale assessment of the importance of the soil vapor intrusion pathway for human exposure to VOCs. The contribution of different variables to TCE permea- tion is described in a laboratory simulation by Fischer and Uchrin (1996), and another tracer gas was used to develop a mathematical model for the phenomenon (Olson and Corsi 2001). 67
68 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects FIGURE 3-1 Environmental contamination from solvents and exposure pathways. Source: EPA 1989. Contamination of drinking-water supplies is of greater health concern. In past years, halocarbons were generally regarded as water-insoluble. It is now recognized that they are soluble in water to a small extent. Maximum solubilities, for example, range from 150 mg/L (or parts per million) for PCE to 4,800 mg/L for methylene chloride. Concentrations typically found in finished drinking water in the United States range from parts per trillion to parts per billion (Moran et al. 2007). VOCs are found as contaminants of surface water and groundwater. Concentrations diminish rap- idly after VOCs enter bodies of water, primarily because of dilution and evaporation. Halocarbons rise to the surface or sink to the bottom, depending on their density. Halocarbons on the surface largely evapo- rate. The movement of halocarbons on the bottom depends on their solubilization in water and their mix- ing by currents or wave action; mixing causes them to reach the surface. Hydrocarbon solvents spilled onto the ground largely evaporate, although some can permeate soil and migrate through it until reaching groundwater or an impermeable layer. Migration of solvents through sandy soil of low organic content is most rapid and extensive (Munnecke and Van Gundy 1979). Solvents in groundwater tend to remain trapped until the water reaches the surface, although some are subject to microbial modification. PCE and TCE, for example, undergo reductive dehalogenation by microorganisms to a small extent to cis- and trans-1,2-DCE, vinyl chloride, and other products (Smith and Dragun 1984; McCarty 1993). Thus, halo- carbon-contaminated groundwater usually contains a relatively high proportion of parent compounds and small amounts of microbial degradation products. EXTERNAL EXPOSURE People are exposed to halocarbons and other VOCs in water by three major routes: inhalation, skin contact, and ingestion. A number of studies have looked at the relative importance of those routes. Weisel and Jo (1996) based estimates of internal doses of TCE and chloroform received from showering on results of experiments with human subjects. They concluded that inhalation and dermal exposure re- sulted in an internal dose of each chemical comparable with the dose ingested in 2 L of water. Gordon et
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 69 al. (2006) conducted a detailed investigation of the contribution of household water use to internal doses of chloroform and other trihalomethanes. Showering and bathing resulted in the highest blood and ex- haled-breath concentrations of chloroform in human subjects in household settings; inhalation and percu- taneous absorption were also found to be important routes of exposure. Giardino and Andelman (1996) reported that the temperature of water had a dominant effect on volatilization of TCE and chloroform dur- ing showering. Some 80% of TCE and 60% of chloroform were released from hot shower water. Haddad et al. (2006) used a physiologic model to assess different home exposure scenarios and concluded that ingestion contributed less than 50% of the total absorbed dose of TCE. Thus, systemic absorption from the lungs, skin, and gastrointestinal tract should all be taken into account in estimating internal doses that result from use of water supplies contaminated with VOCs. INTERNAL EXPOSURE The concept of dose has been refined during the last 15-20 years. The amount of a chemical to which a person is exposed is now termed the external exposure or administered dose. Absorption into the blood may be partial or complete, depending on the chemical and route of exposure. The amount of a chemical absorbed systemically from the lungs, gastrointestinal tract, and skin is termed the absorbed dose or internal dose. The amount of a chemical that reaches an organ or tissue where a toxic effect oc- curs is termed the target-organ dose. It is necessary here to specify the amount of the toxicologically ac- tive forms of the chemical. In the case of TCE, both the parent compound and trichloroethanol, a major metabolite, cause depression of the central nervous system (CNS) when present at sufficient concentra- tions. PCE can also produce CNS depression. Trichloroacetic acid, a major metabolite of both TCE and PCE, is believed to be primarily responsible for liver tumors in B6C3F1 mice (Bull 2000; Lash et al. 2000a). Thus, it is important to know or to be able to estimate the quantity of the bioactive moiety and how long it remains in the target organ if one wants to predict the magnitude and duration of toxic action (Andersen 1987). Pharmacokinetics, or toxicokinetics, and physiologically based toxicokinetic (PBTK) models are increasingly important in reducing uncertainties inherent in health risk assessments of TCE, PCE, methyl- ene chloride, and other VOCs (Andersen et al. 1987; Andersen 2003; Clewell and Andersen 2004; Cle- well et al. 2005; Krishnan and Johanson 2005). Toxicokinetics may be defined as the systemic uptake, distribution, metabolism, interaction with plasma and cellular components, and elimination of toxic chemicals and their metabolites. Kinetic processes determine how much of an external dose is absorbed into the blood; reaches the arterial circulation; binds to plasma proteins or other inactive sites; enters spe- cific organs; is biotransformed to toxicologically active and inactive forms; interacts with target mole- cules, cells, and tissues; and is eliminated from the target tissue and the body (Bruckner et al. 2008). One or more of those processes can vary widely from one route of exposure to another, from high to low doses, from one species to another, and from one individual to another. Gaining an understanding of how kinetic processes differ can substantially reduce the number of assumptions made in assessing toxicity and cancer risks posed by VOCs. Volatility and lipophilicity are two of the most important properties of VOCs that govern their toxicokinetics. The volatility of the compounds varies inversely with their molecular weight. TCE, for example, is of lower molecular weight and evaporates more readily than PCE. PCE is more lipid-soluble than TCE. Cell membranes are made up largely of lipids. Halocarbons pass freely through membranes from areas of high to low concentration by passive diffusion. Absorption Halocarbons and other VOCs are absorbed through intact human skin to a limited extent. The bar- rier to penetration is the stratum corneum, the skinâs outermost layer. The stratum corneum is composed
70 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects of very tightly adhering, keratinized epithelial cells, which present a much more substantial barrier to halocarbons than do living cell membranes. Important determinants of the rate and extent of percutaneous absorption of a chemical include the integrity and thickness of the stratum corneum, the surface area ex- posed and duration of contact, and the chemicalâs concentration, molecular size, and lipophilicity (Stewart and Dodd 1964; EPA 1992). Percutaneous absorption of VOCs is more extensive through rodent than through human skin, owing largely to the rodentsâ thinner stratum corneum and higher dermal blood flow rate (McDougal et al. 1990; Monteiro-Riviere et al. 1990). Poet et al. (2000) reported that the dermal permeability constant for absorption of 1,1,1,-trichloroethane from water into humans was one-fortieth that into rats. Those researchers concluded that people will not absorb substantial amounts of VOCs through their skin from contaminated water regardless of the duration of exposure. That conclusion con- flicts with that of Weisel and Jo (1996) and Gordon et al. (2006), who found percutaneous absorption to be an important route of human exposure. Halocarbons and most other VOCs are absorbed from the lungs rapidly and extensively. TCE and PCE, for example, appear in the arterial blood of rats within 1 min after initiation of inhalation exposure (Dallas et al. 1991, 1994). Most of the systemic absorption of inhaled VOCs occurs in the alveoli. The small lipophilic molecules readily diffuse bidirectionally through the thin capillary and alveolar type I cells. Such gases in the alveoli are believed to equilibrate almost instantaneously with blood in the pul- monary capillaries (Goldstein et al. 1974). The ratio of the concentration of a VOC in blood to its concen- tration in air at equilibrium is the blood:air partition coefficient. Partition coefficients have been measured in vitro with human and rat blood for a large number of VOCs (Gargas et al. 1989). Respiratory, or alveo- lar, ventilation rate and the ratio of cardiac output to pulmonary perfusion rate are two other important determinants of pulmonary uptake of VOCs. VOCs diffuse from areas of high to areas of low concentra- tion, so increases in respiratory rate (to maintain a high alveolar concentration) and increases in pulmo- nary blood flow rate (to maintain a large concentration gradient by removing capillary blood that contains a VOC) enhance systemic absorption. The higher those factors are, the greater the systemic uptake. The TCE blood:air partition coefficient of the rat is 2.7 times greater than that of the human (Gargas et al. 1989). Resting alveolar ventilation rates of rats and mice are as much as 11 and 23 times higher, respec- tively, than that of humans. Cardiac outputs of rats and mice are about 6 and 10 times greater than that of humans (Brown et al. 1997). Thus, for equivalent inhalation exposures to TCE and other VOCs, internal doses are substantially higher in rodents than in humans (Bruckner et al. 2008). Systemic absorption of VOCs during inhalation exposures depends on metabolism and tissue loading, in addition to the factors described above. The percentage uptake of inhaled TCE is initially high in experimental animals. Uptake progressively declines during exposure as a chemical accumulates in tissues, and its concentration in venous blood returning to the lungs increases, reducing the air:blood con- centration gradient (Dallas et al. 1989). A near steady state, or equilibrium, in uptake and in blood con- centrations is usually reached within an hour and maintained despite continued inhalation of a fixed air concentration of TCE. The same phenomenon was reported recently in human subjects inhaling TCE at 1 ppm for 6 h (Chiu et al. 2007). Blood concentrations of PCE, in contrast, slowly rose in the subjects dur- ing the last 4 h of a 6-h exposure to PCE at 1 ppm. That difference is due to PCEâs higher lipid solubility, which results in its greater and more prolonged uptake into body fat. Persons using contaminated water at Hadnot Point and Tarawa Terrace probably had intermittent PCE or TCE exposures during the day when they drank water and used heated water. Day-to-day exposures were also intermittent because the indi- vidual water-supply wells operated on a cycle schedule (see Chapter 2). Halocarbons and other VOCs are well absorbed after their ingestion. More than 90% of TCE given in water as an oral bolus (by gavage to rats that have been fasting) is absorbed systemically (DâSouza et al. 1985). Peak blood concentrations are observed within 5-10 min of dosing. The presence of food, particularly fatty foods, in the gut delays absorption of TCE and other organic solvents. Kim et al. (1990a) describe the time course of carbon tetrachloride in the venous blood of rats given an equiva- lent oral bolus dose of the halocarbon in water and in corn oil. The peak blood concentration of carbon tetrachloride is about 10 times higher in the water-vehicle group than in the oil-vehicle group, but the re- lationships between blood concentrations of carbon tetrachloride and time in the two groups are essen-
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 71 tially the same. Liver injury is more pronounced in the group that ingested carbon tetrachloride in water, apparently because of the liverâs markedly higher exposure to the hepatotoxin during the initial minutes after dosing. Inhalation results in substantially higher arterial blood and target-organ concentrations of VOCs than does ingestion of comparable doses. A number of factors are responsible for that phenomenon. As described above, fatty foods serve as a reservoir in the gut to prolong the absorption of lipophilic chemi- cals. All the cardiac output passes through the pulmonary circulation compared with about 20% through the gastrointestinal tract. More rapid blood flow in the lungs creates a greater concentration gradient, which results in more rapid diffusion into the blood. The distance that VOCs must diffuse from their ab- sorption surface to capillaries is considerably shorter in the alveoli than in the gastrointestinal mucosal epithelium. The most important route-dependent difference for well-metabolized VOCs is presystemic elimination after their ingestion (Bruckner et al. 2008). Presystemic Elimination of Oral Volatile Organic Compounds A substantial proportion of TCE and other well-metabolized VOCs that are ingested does not reach the arterial circulation or extrahepatic organs. It has not been established whether a significant pro- portion of low doses of VOCs undergo gastrointestinal metabolic clearance, though researchers have es- tablished the presence of several CYP3A isoforms in the small intestines of humans (Obach et al. 2001) and mice and rats (Martignoni et al. 2006). Chemicals absorbed into venous mesenteric blood vessels are conveyed via the portal vein through the liver before entering the mixed venous circulation. The liver contains the highest concentrations of CYP2E1 and other enzymes and is the major site of xenobiotic me- tabolism in the body. The efficiency of first-pass hepatic metabolism and clearance depends on the ad- ministered dose of the chemical, the rate at which it is ingested, and its propensity to be metabolized. White et al. (unpublished data) recently observed that bioavailability of 1,1,1,-trichloroethane, a poorly metabolized halocarbon, was markedly higher in orally dosed rats than was TCE, a well-metabolized halocarbon.1 The bioavailability of TCE was substantially higher when it was given as a single oral bolus (that is, all at one time) than when it was given slowly over several hours. Administration of the quickly absorbed chemical as a bolus resulted in its rapid arrival in amounts that exceeded (or saturated) the liverâs metabolic capacity. In contrast, neither the dose nor the rate of oral administration of 1,1,1,- trichloroethane affected its first-pass hepatic elimination or bioavailability, because it was poorly metabo- lized. The bioavailability of TCE, however, was significantly lower at lower doses because of its more efficient metabolic clearance. Lee et al. (1996) also found that hepatic first-pass elimination of oral TCE was inversely related to dose in rats. VOCs are exhaled during their first pass through the lungs. Lee et al. (1996) confirmed that pulmonary elimination of TCE was not dose-dependent. Andersen (NRC 1986) had proposed that pulmonary elimination of VOCs was governed instead by a VOCâs blood:air partition coef- ficient. In summary, VOCs that are extensively metabolized and quite volatile are most efficiently elimi- nated before they reach the arterial circulation. First-pass, or presystemic elimination, may have major implications for cancer and noncancer risks posed by ingestion of very low concentrations of VOCs in drinking water. Over 25 years ago, An- dersen (1981) proposed that the liver was capable of removing âvirtually allâ of a well-metabolized VOC after its ingestion if the amount in the portal blood was not high enough to saturate hepatic metabolism. As described below, most of the VOCs of interest at Camp Lejeune are extensively metabolized. Metabo- lism is required for their conversion to potentially cytotoxic or mutagenic substances. The liver should bear the brunt of metabolizing ingested VOCs. However, first-pass hepatic metabolic clearance and exha- lation will protect most extrahepatic organs by reducing the amount of parent compounds reaching them. 1 White, C.A., S. Muralidhara, C. Hines, and J.V. Bruckner. Effect of oral dosage level and rate on the bioavail- ability and metabolism of trichloroethylene and 1,1,1-trichloroethane. Submitted to Toxicol. Sci. Manuscript being prepared for submission for publication.
72 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects Parent halocarbons, as described previously, can depress CNS functions if they reach the brain in suffi- cient amounts. A number of extrahepatic tissuesâincluding brain, lung, renal, testicular, and breast tissue and bone marrowâcontain CYP2E1, other P-450s, and other enzymes that metabolize xenobiotics (de Waziers et al. 1990; Ding and Kaminsky 2003). The amounts of enzymes are usually considerably lower in those tissues than in the liver but can be high enough in some cell types to form quantities of reactive metabolites adequate to harm the cells. Hepatic halocarbon metabolites stable enough to be transported to other organs can potentially injure those organs. It is widely recognized, for example, that derivatives of glutathione conjugates of TCE and PCE formed in the liver are taken up and metabolized further by the kidneys to substabnces that may be nephrotoxic or carcinogenic (Lash et al. 2000b; Lash and Parker 2001; Lash et al. 2007). VOCs absorbed from the lungs and skin are not subject to presystemic elimina- tion. The efficiency of presystemic elimination of ingested halocarbons in humans remains to be estab- lished. Sufficiently sensitive analytic methods for quantifying VOCs in biologic specimens that allow di- rect testing of Andersenâs (1981) aforementioned hypothesis have not been available until very recently. Lee et al. (1996) used a gas-chromatography-electron-capture headspace technique to measure blood con- centrations in assessing presystemic elimination of TCE in rats. Their experimental approach required monitoring complete blood-TCE time courses. The lowest oral dose for which a complete time course could be delineated was 170 Âµg/kg. Some 60% of the dose was eliminated before reaching the ratsâ arte- rial circulation. More recently, a much more sensitive analytic method has been used; it involves VOC extraction and concentration on a solid fiber and measurement with gas chromatography-mass spectrome- try. Using that technique, Blount et al. (2006) measured 31 VOCs in the blood of the general U.S. popula- tion. Liu et al. (2008) have also used the technique to obtain blood time-course data on rats given TCE orally at as low as 1 ng/kg. Bioavailability was about 10% at the lowest doses. The analytic methodâs limit of quantification was 25 pg/mL (ppt). Rats have a greater capacity to metabolize TCE and other VOCs than humans, so first-pass hepatic elimination should be somewhat less efficient in humans. Weisel and Jo (1996), however, were able to detect TCE in exhaled breath for only seconds to a few minutes af- ter humans ingested water contaminated with TCE. Chloroform was undetectable in breath samples of persons who consumed chlorinated municipal water; this implies complete first-pass hepatic elimination. The efficiency of human presystemic elimination of TCE and other VOCs at environmental concentra- tions can be determined by extrapolation from animal data or by direct measurement. In summary, presys- temic elimination should protect most extrahepatic tissues from harm after ingestion of TCE, PCE, and other VOCs at environmental concentrations. Solvent or Vehicle Effects on VOC Toxicity Oral and dermal administration of VOCs in toxicology studies usually require that the lipophilic chemicals be dissolved or diluted in a suitable solvent. Corn oil and other digestible oils have been the most commonly-used vehicles, though aqueous emulsions, suspensions, and gelatin-encapsulated prepa- rations have been employed in toxicity and carcinogenicity investigations. Considerable effort has been devoted to assessing adverse health effects of VOCs in drinking water. A number of studies have been conducted to determine whether experiments in which VOCs were given to animals in corn oil were rele- vant to assessing risks from ingestion of VOCs in water. Kim et al. (1990a,b), for example, found that corn oil served as a reservoir in the gut to delay systemic absorption of carbon tetrachloride in rats. Al- though bioavailability of carbon tetrachloride given in corn oil and in an aqueous Emulphor emulsion was the same, peak blood concentrations of carbon tetrachloride and acute hepatotoxicity were much lower in the corn oil group. Raymond and Plaa (1997) found aqueous preparations of carbon tetrachloride were more acutely hepatotoxic to rats than when it was administered in corn oil, though the converse was true for nephrotoxicity of chloroform. Dissimilar findings have been reported in subacute studies. Condie et al. (1986), for example, observed that carbon tetrachloride was more hepatotoxic to mice after 90 days of oral dosing in corn oil than in an aqueous Tween emulsion. Koporec et al. (1995), however, found no dif-
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 73 ference in rats when given carbon tetrachloride in either corn oil or aqueous solution for 13 weeks. As described below, chloroform and other VOCs have been found to be hepatocarcinogens in mice when given chronically by gavage in corn oil, but not when delivered in drinking water. Under these circum- stances, interpretation requires consideration of the confounders introduced by both the vehicle and dose regimen. There is concern that vehicles may not only affect the absorption of VOCs, but may influence VOC metabolism and disposition and may have biological actions of their own. Oils in the gastrointesti- nal tract largely retain VOCs until the oil is emulsified and digested (Kim et al. 1990b). The lipids thus delay VOC absorption into the blood and can carry some of the VOC along into the lymphatics. Common surfactants used as emulsifying agents are known to modify drug absorption by altering the physical properties of membranes, as well as certain transport mechanisms (Xia and Onyuksel 2000). Feeding rats a diet supplemented with corn oil enhanced the induction of hepatic cytochrome P4502B1 by phenobarbi- tal (Kim et al. 1990c). This is one isozyme that metabolically activates high doses of TCE and several other VOCs in rats. Feeding animals a high-fat diet containing corn oil increases lipoperoxidation and susceptibility to oxidative stress by reducing antioxidant enzyme defenses (Domitrovic et al. 2006; Slim et al. 1996). A number of investigations have shown increased incidences of breast, colorectal, and pros- tate cancer in rodents maintained on high-fat diets, but recent human epidemiological studies have largely been inconclusive (Kushi and Giovannucci 2002; Thiebaut et al. 2007; Kobayashi et al. 2008). Pattern of Water Ingestion A personâs pattern of consumption of VOC-contaminated water can have a marked effect on halogenated chemicalsâ toxicokinetics and toxic or carcinogenic potential. For convenience in chronic oral-carcinogenicity studies, TCE, PCE, methylene chloride, and chloroform have usually been given daily by gavage. In each instance, an increased incidence of liver tumors in B6C3F1 mice was observed. No such increase was seen when the mice received tumorigenic doses of chloroform and other VOCs in drinking water (Jorgenson et al. 1985; Klaunig et al. 1986). Larson et al. (1994) saw marked necrosis and ensuing proliferation of hepatocytes in B6C3F1 mice given chloroform by gavage, but no such effects in mice that consumed the same daily doses in their water. La et al. (1996) reported greater DNA-adduct formation and hepatocellular proliferation in mice given 1,2,3-trichloropropane by gavage than in those receiving the chemical in drinking water. Sanzgiri et al. (1995) administered the same doses of carbon tetrachloride to rats by gavage and over 2 h by constant gastric infusion. Arterial blood concentrations of carbon tetrachloride and the extent of acute hepatic damage were greater in the gavage groups. Carbon tetrachloride and other halocarbons are quickly absorbed from the gastrointestinal tract, and the rapid de- livery of large quantities of carbon tetrachloride to the liver via the portal blood inhibited metabolism and killed hepatocytes. Both effects reduced hepatic metabolic clearance of the chemical. Such findings raise questions about the relevance of gavage toxicity and cancer-study results to real-life human exposures, in which people typically ingest contaminated water in divided doses over the course of the day. Systemic Distribution VOCs are transported by the arterial blood to tissues throughout the body. The lipophilic com- pounds do not bind appreciably to plasma proteins or hemoglobin but partition into their hydrophobic regions and into phospholipids, lipoproteins, and cholesterol present in the blood (Lam et al. 1990). Initial uptake into tissues depends primarily on their rate of blood flow and tissue:blood partition coefficient. The brain is a prime example of an organ with a high perfusion rate and high lipid content, hence a high brain:blood partition coefficient. Lipophilic VOCs quickly accumulate in the brain and can rapidly de- press its functions on initiation of sufficiently high external exposures (Warren et al. 2000). Inhalation of a few hundred ppm of TCE and PCE can inhibit psychophysiological functions in humans, while inhala-
74 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects tion of several thousand ppm will rapidly produce marked CNS depression. Then, redistribution to poorly perfused lipid-rich tissues (such as bone marrow, skin, and fat) with even higher tissue:blood partition coefficients occurs. Adipose tissue gradually accumulates large amounts of VOCs and slowly releases them back into the bloodstream because of its high tissue:blood partition coefficient and low blood perfu- sion rate. That prolongs exposure of other tissues to the chemicals (Bruckner et al. 2008). Metabolic Activation and Inactivation of Trichloroethylene and Perchloroethylene Metabolism, or biotransformation, plays a key role in modulating the toxicokinetics and the ensu- ing toxicity or carcinogenicity potential of the VOCs of interest at Camp Lejeune. As described previ- ously, most VOC metabolism occurs in the liver. Biotransformation in other tissues is quantitatively in- significant but can be toxicologically significant if CYP2E1 and some other enzymes are present. Specific hepatic and extrahepatic enzymes convert the VOCs to relatively water-soluble metabolites that can be eliminated more readily in the largely aqueous urine and bile. Conversion of the parent compounds and their reactive metabolites to less active or inactive metabolites that are more water-soluble and therefore more efficiently eliminated is termed metabolic inactivation or detoxification. The relative extent of acti- vation and inactivation of VOCs can vary substantially from one species to another and from one individ- ual to another. It is well established that the metabolic activation of the VOCs of interest in Camp Lejeune water, in decreasing order of magnitude, is as follows: mice > rats > humans (Elfarra et al. 1998; Lipscomb et al. 1998; Volkel et al. 1998; Lash and Parker 2001). Mice express very low concentrations of epoxide hydrolase (Lorenz et al. 1984), the enzyme that catalyzes the hydrolytic degradation (detoxifica- tion) of highly reactive epoxide metabolites of TCE and PCE. Many other factors or variables may also influence the metabolism and toxicokinetics of VOCs (Lof and Johanson 1998). The metabolic activation and inactivation of TCE has been described in detail elsewhere (ATSDR 1997b; Lash et al. 2000a; NRC 2006). TCE is metabolized primarily via an oxidative pathway involving sequential formation of a series of metabolites. The second, relatively minor pathway involves glutathione (GST) conjugation (Figure 3-2). The key metabolic pathways and metabolites of toxicologic interest are described briefly below. The initial step in the oxidative pathway is catalyzed by microsomal cytochrome P-450s. CYP2E1, as noted previously, is the primary P-450 isozyme responsible for oxidation of low concentra- tions of TCE (Lipscomb et al. 1997; Ramdhan et al. 2008). P-450-catalyzed oxidation of TCE in rodents and humans, in decreasing order of magnitude, is as follows: mice>rats>humans (Lash et al. 2000a). Whether TCE is initially converted to TCE oxide is controversial. Cai and Guengerich (2001) were able to detect formation of trace amounts of the epoxide by phenobarbital-induced rat liver P-450s but not by human liver P-450s. The majority of TCE is apparently converted to an oxygenated TCE-P-450 interme- diate, which rearranges to form chloral, a major metabolic intermediate. Chloral is oxidized to chloral hydrate, a sedative widely used in medical and dental procedures in infants and children (Vade et al. 1995; Keengwe et al. 1999). Chloral hydrate is both oxidized to trichloroacetic acid and reduced to tri- chloroethanol. Much trichloroethanol is conjugated with glucuronic acid and excreted in the urine. Tri- chloroethanol glucuronide that is excreted in the bile is hydrolyzed, reabsorbed, and oxidized in part to trichloroacetic acid. Chiu et al. (2007) recently observed that concentrations of trichloroacetic acid were significantly lower than trichlorethanol and trichloroethanol glucuronide concentrations in the blood of humans who had inhaled TCE at 1 ppm for 6 h. Modest amounts of dichloroacetic acid apparently are produced from trichloroacetic acid and trichloroethanol in mice, but relatively little dichloroacetic acid is formed in rats. Trace amounts of dichloroacetic acid were detected in one study of TCE-exposed humans (Fisher et al. 1998) but not in other studies (Lash et al. 2000b; Bloemen et al. 2001). Both trichloroacetic acid and dichloroacetic acid have been shown to be hepatic carcinogens in mice at high doses (Bull 2000). It is generally accepted that trichloroacetic acid is a nongenotoxic liver carcinogen in B6C3F1 mice, al- though its ability to cause liver cancer in humans has been discounted by findings in a number of labo-
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 75 FIGURE 3-2 Metabolism of trichloroethylene. Metabolites marked with are known urinary metabolites. Arrows with broken lines indicate other possible steps in forming DCA. CYP, cytochrome P-450; DCA, dichloroacetic acid; DCVC, S-(1,2-dichlorovinyl)-L-cysteine; DCVG, S-(1,2-dichlorovinyl)glutathione; DCVT, S-(1,2-dichlorovinyl)thiol; GST, glutathione S-transferase; NAcDCVC, N-acetyl-S-(1,2-dichlorovinyl)-L-cysteine; TCA, trichloroacetic acid; TCE, trichloroethylene; TCE-O-CYP, trichloroethylene-oxide-cytochrome P-450 complex; TCOG, trichloroethanol glucuronide; TCOH, trichloroethanol. Source: NRC 2006. ratory investigations (Bull 2000; Moore and Harrington-Brock 2000). The possible causative role of di- chloroacetic acid in human liver cancer is even more controversial (Walgren et al. 2005; Caldwell and Keshava 2006; Keshava and Caldwell 2006; Klaunig et al. 2007). The glutathione conjugation pathway is quite similar qualitatively, but not quantitatively, in rats and humans. The initial step in this second, minor pathway involves conjugation of TCE with glutathione to form S-(1,2-dichlorovinyl)glutathione (DCVG). DCVG formation occurs primarily in the liver at a rate about 10 times greater in rats than in humans (Green et al. 1997a). Much of the DCVG is excreted via the bile into the intestines and converted to S-(1,2-dichlorovinyl)-L-cysteine (DCVC). That metabolite is re- absorbed and taken up by the liver, where a portion is detoxified by N-acetylation. Bernauer et al. (1996) exposed rats and humans to TCE vapor at up to 160 ppm for 6 h. The rats excreted 8 times more N-acetyl- DCVC in their urine than did the human volunteers at each exposure level. Some DCVC is taken up by the kidneys and further metabolized by the enzyme Î²-lyase to S-(1,2-dichlorovinyl)thiol (DCVSH). DCVSH is then converted to unstable, highly reactive products, including chlorothioketene and thionoa- cylchloride (Lash et al. 2000a). Metabolic activation of DCVC to chlorothioketene was shown to occur 11 times more rapidly in rats than in humans (Green et al. 1997a). Lash et al. (2001b) also demonstrated that cultured rat renal cells are more sensitive to DCVC than are human renal cells. Chlorothioketene and
76 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects similarly unstable congeners are capable of covalently binding to renal cellular proteins and DNA, and this results in genotoxicity and cytotoxicity with ensuing regenerative hyperplasia and potentially renal- cell carcinoma. PCE, like TCE, is metabolized through cytochrome P-450-catalyzed oxidation and glutathione conjugation (Figures 3-3 and 3-4). CYP2E1 is not thought to play a major role. PCE is believed to be oxi- dized primarily by the CYP2B family in the rat (Hanioka et al. 1995). In humans, CYP2B6 is the primary isoform responsible for PCE metabolism, and there are minor contributions by CYP1A1 and CYP2C8 (White et al. 2001). The initial metabolite is the epoxide PCE-oxide. That metabolic intermediate can be biotransformed to several products (Lash and Parker 2001). The primary one is trichloroacetyl chloride, which reacts with water to form trichloroacetic acid, the predominant PCE metabolite found in the urine of rodents and humans (Birner et al. 1996; Volkel et al. 1998). Some trichloroacetic acid is converted to dichloroacetic acid. PCE is a much poorer substrate for CYPs than TCE (that is, PCE is much less FIGURE 3-3 Metabolism of PCE by P-450 pathway. *Identified urinary metabolites: l, PCE; 2, PCE epoxide; 3, trichloroacetyl chloride; 4, trichloroacetate; 5, trichloroethanol; 6, trichloethanol glucuronide; 7, oxalate dichloride; 8, trichloroacetyl aminoethanol; 9, oxalate; 10, dichloroacetate; 11, monochloroacetate; 12, chloral. Source: Lash and Parker 2001. Reprinted with permission; copyright 2001, Pharmacological Reviews.
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 77 FIGURE 3-4 Metabolism of PCE by glutathione conjugation pathway. *Identified urinary metabolites: 1, PCE; 2, TCVG; 3, TCVC; 4, NAcTCVC; 5, NAcTCVC sulfoxide; 6, 1,2,2-trichlorovinylthiol; 7, TCVCSO; 8, 2,2- dichlorothioketene; 9, dichloroacetate. Enzyrmes: GST, GGT, dipeptidase (DP), Î²-lyase, FMO3, CCNAT, CYP3A1/2, and CYP3A4. Unstable, reactive metabolites are shown in brackets. Source: Lash and Parker 2001. Reprinted with per- mission; copyright 2001, Pharmacological Reviews. extensively metabolized than TCE) (ATSDR 1997c; Chiu et al. 2007). Saturation of PCE oxidative me- tabolism occurs at a lower exposure concentration in humans than in rats. Rats metabolize substantially more PCE to trichloroacetic acid than do humans (Volkel et al. 1998). Only traces of dichloroacetic acid were detected in the urine of persons who inhaled PCE at 40 ppm for 6 h. Rats subjected to an equivalent exposure excreted relatively large amounts of dichloroacetic acid, a rodent hepatic carcinogen. A small proportion of absorbed PCE undergoes conjugation with glutathione to form S-(1,2,2- trichlorovinyl) glutathione (TCVG). That initial metabolic step is catalyzed by glutathione S-transferases
78 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects and occurs primarily in the liver. TCVG is converted to S-(1,2,2-trichlorovinyl)-L-cysteine (TCVC). TCVC, like the DCVC formed from TCE, is both detoxified in the liver by N-acetylation and metaboli- cally activated by Î²-lyase in the kidneys to cytotoxic, mutagenic thioketenes (Lash and Parker 2001). TCVC is also oxidized by flavin-containing monooxygenase 3 (FMO3) to TCVC sulfoxide (Ripp et al. 1997), which can rearrange spontaneously to form 2,2-dichlorothioketene. Thus, potent alkylating agents are formed via two subpathways in the kidneys. 2,2-Dichlorothioketene can also decompose to di- chloroacetic acid. Hence, dichloroacetic acid is derived from both glutathione- and CYP-dependent bio- transformation of PCE. PCE is conjugated with glutathione more extensively by rats (1-2% of the dose) (Dekant et al. 1986) than is TCE (less than 0.005% of the dose) (Green et al. 1997a). The extent of glu- tathione conjugation of PCE increases when the oxidative pathways begin to become saturated at high PCE exposure concentrations. Metabolic products of glutathione conjugates of TCE and PCE are primary contributors to the halocarbonsâ nephrotoxicity (Lash et al. 2007). Humans are likely to be less susceptible than rodents to the toxic or carcinogenic actions of PCE, as they are to those of TCE. Humans absorb less inhaled PCE and TCE, attain lower target-organ doses, and metabolically activate a smaller proportion of their internal dose. As noted above, rats exhibit a higher capacity for oxidation of PCE. Volkel et al. (1998) report finding substantially higher urinary ex- cretion of trichloroacetic acid, N-acetyl TCVC, and dichloroacetic acid in the urine of rats than in the urine of humans subjected to identical PCE inhalation regimens. Lash et al. (1990) and Cooper (1994) report 10 times higher activity of cysteine conjugate Î²-lyase activity in rat than in human kidney. Green et al. (1990) report that Î²-lyase-dependent metabolism of TCVC in rat kidney cystosol is more rapid and efficient than in either mice or humans. TCVC metabolism is also greater in male than in female rats. The male rat is more susceptible to PCE-induced nephrotoxicity. A low incidence of kidney cancer is seen in male but not female rats in TCE and PCE cancer bioassays (ATSDR 1997b,c). It is evident from the foregoing there is a great deal of information about the toxicokinetics, me- tabolism, and toxicology of TCE, PCE, and other VOCs in laboratory animals and humans. That knowl- edge allows scientists to judge the relevance of VOCsâ adverse effects in animals to humans with a rea- sonable degree of certainty. Mice and rats absorb more inhaled TCE and PCE, metabolically activate more of their absorbed dose, and inactivate epoxide metabolites less efficiently than do humans. Such interspecies toxicokinetic differences contribute to the greater susceptibility of rodents than of humans to TCE- and PCE-induced hepatic, lung, and renal tumors. Toxicodynamic species differences that predis- pose B6C3F1 mice to liver cancer are also recognized (see Chapter 4). POTENTIALLY SENSITIVE POPULATIONS Many factors or variables may alter the toxicokinetics of and the responses of a person to TCE, PCE, and other VOCs (Lof and Johanson 1998; Bruckner et al. 2008). Some variables that are character- istic of a particular group may increase susceptibility, others reduce it, and still others have no influence. The net effect of circumstances involving multiple variables can be difficult to predict. There are scenar- ios in which the toxicity or carcinogenicity of moderate exposure (for example, occupational exposure) or high exposure (for example, in animal cancer studies) to some VOCs may be significantly affected by variables like age, sex, genetics, physiologic condition, or lifestyle. As discussed below, that appears not to be the case for the concentrations of most of the VOC contaminants identified in water at Camp Le- jeune. Children There is concern that infants and children may be more vulnerable than adults to some adverse ef- fects of chemicals (Dourson et al. 2002; Daston et al. 2004). A National Research Council report (NRC 1993) emphasized that there were âwindows of vulnerabilityâ or short periods of early human develop-
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 79 ment when chemical exposures may significantly alter organ function or structure. Potentially vulnerable targets in infants and young children include the endocrine, reproductive, immune, visual, and nervous systems. Little information is available on the effects of TCE, PCE, and other solvents on the develop- ment of those organ systems in laboratory animals or humans. There is considerably more knowledge of consequences of exposure of adults, as discussed in Chapter 4. It is not clear whether organ-system development of young children or animals is influenced by exposure to VOCs. A number of chemicalsâsuch as lead, mercury, thalidomide, chloramphenicol, and organophosphorus insecticidesâare known to have more pronounced adverse effects in infants and young children than in adults (Bruckner 2000). Children are not necessarily more susceptible to toxicants. The most definitive human data on age-dependence available to the 1993 National Research Council committee were maximum tolerated doses of a variety of anticancer agents. Clinical trials in pediatric and adult patients revealed that children could tolerate higher doses of most of the antitumor drugs (Glaubiger et al. 1982; Marsoni et al. 1985). Susceptibility can vary markedly with a childâs age. The youngest (pre- mature and full-term newborns) are generally the most sensitive to drugs and other chemicals. Toxicodynamic and toxicokinetic factors are responsible for age-dependent differences in the tox- icity of VOCs and other chemicals. Toxicokinetic processes determine the amount of the active form of a chemical that reaches its target tissue or cell and how long it remains there. Toxicodynamics refers to the sequence of events that occur in a target tissue or cell on arrival of the bioactive form of a chemical. The events culminate in adverse effects that, in turn, dictate the magnitude and duration of toxic action. Major anatomic, biochemical, and physiologic changes occur during the neonatal period, infancy, childhood, and adolescence. Maturation can markedly affect the absorption, distribution, metabolism, and elimina- tion of many chemicals (Bruckner and Weil 1999; Bruckner 2000; Ginsberg et al. 2004). The systemic absorption of VOCs may be somewhat higher in infants in connection with some routes of exposure. Infantsâ and young childrenâs respiratory rates and cardiac outputs are relatively high and favor uptake of inhaled VOCs. That is counteracted to some extent by their smaller alveolar surface area for absorption (Snodgrass 1992). The rate of dermal absorption is comparable in full-term newborns and adults, although the ratio of skin surface area to body weight is about 2.7 times greater in infants than in adults. TCE, PCE, and other solvents are well absorbed from the gastrointestinal tract of all age groups. The low plasma binding capacity of neonates should result in an increased rate of excretion of di- chloroacetic acid and trichloroacetic acid, carcinogenic metabolites of TCE and PCE in mice, but it may be offset by neonatesâ larger extracellular water content, from which the metabolites have to be cleared. The net effect of immaturity on toxicokinetics can be quite difficult to predict (Bruckner 2000; Pastino et al. 2000). Age-dependent changes in biotransformation have been reasonably well characterized in humans and may have the greatest impact on VOC toxicokinetics and health risks (Hines and McCarver 2002). Concentrations of metabolic enzymes are quite low in newborns and develop asynchronously during the initial months and years. Concentrations of CYP2E1, the P-450 isozyme primarily responsible for oxida- tion of low doses of TCE (Guengerich et al. 1991), are very low at birth and increase steadily during the first year of life (Johnsrud et al. 2003). Because infants lack the enzymes that convert TCE, PCE, and other VOCs to toxic or mutagenic metabolites, they should be less susceptible to the chemicals than adults. Concentrations of CYP2E1 and additional enzymes that catalyze other steps in VOC metabolic pathways generally attain adult values within 6 months to 3 years. Reimche et al. (1989) determined the half-lives of chloral hydrate, an obligate oxidative metabolite of TCE, in premature newborns, full-term newborns, and young children to be 39.8, 27.8, and 9.7 h, respectively. That finding shows how the abil- ity to eliminate chloral hydrate metabolically increases with maturity. The greater metabolic clearance in children 1-6 years old is apparently due to their larger liver volume and higher blood flow (Murry et al. 2000) rather than higher CYP2E1 activity (Blanco et al. 2000). Greater metabolic capacity may result in increased formation of reactive metabolites of TCE and PCE, although they should also be more rapidly eliminated. Xenobiotic metabolism is similar in older children, adolescents, and adults (Alcorn and McNamara 2002).
80 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects Age-related changes in one toxicokinetic process may be offset or augmented by concurrent changes in other processes. Validated PBTK models are useful for predicting target-organ doses of bio- logically active parent compounds or metabolites under such circumstances. Sarangapani et al. (2003) constructed a PBTK model that integrated age-specific respiratory measures so that the disposition of four VOCs (PCE, vinyl chloride, isopropanol, and styrene) could be predicted; blood concentrations of the parent compounds in infants and adults were comparable or differed by a factor of less than 2 during the first year of life. Nong et al. (2006) recently incorporated age-specific liver volumes and CYP2E1 content into a PBTK model for toluene; combined interindividual and interage variability in blood toluene con- centrations over the periods of monitoring were within a factor of 2 except in neonates, whose concentra- tions were higher. Clewell et al. (2004) developed a âlife-stage modelâ to simulate blood concentrations of VOCs (PCE, methylene chloride, vinyl chloride, and isopropanol); the predicted internal concentra- tions at different life stages were within a factor of 2 except during the neonatal period, when the largest differences were manifested. A recent model by Rodriguez et al. (2007) similarly yielded predictions of relatively high blood concentrations of TCE, PCE, methylene chloride, benzene, chloroform, and methyl ethyl ketone in neonatal rats; the increases were due largely to pronounced metabolic immaturity in neo- nates. In summary, there is cause for concern that infants and young children will be more susceptible to adverse effects of chemicals. Anatomic and physiologic immaturity can predispose younger people to higher target-organ concentrations of some classes of chemicals. Heavy metals, such as lead and mercury, are known to be absorbed from the gastrointestinal tract and deposited in the brain in greater quantities in infants and young children. Cells in some developing organs (such as neurons in the brain) are more sen- sitive to injury because they must undergo highly ordered division, differentiation, and migration to func- tion effectively in later life; relatively low concentrations of lead inhibit those processes and affect neuro- development and cognitive ability. Conversely, clinical experience has shown that children tolerate higher doses of a number of anticancer drugs than do adults before exhibiting toxicity. Thus, susceptibility is both chemical-dependent and age-dependent. The youngest (premature and newborn infants) are usually the most different from adults and the most likely to be more sensitive to chemical injury. The net effect of anatomic and physiologic immaturity on sensitivity is difficult to predict for chemicals on which there have been few or no studies or data. Although few data are available for TCE, PCE, and other VOCs, PBTK models predict a difference of no more than a factor of 2 in blood concentrations of VOCs after equivalent exposures of infants and adults. Newborns are predicted to have the highest blood concentra- tions and would be expected to be the most sensitive to any neurologic effects caused by high doses of the parent compounds. Newborns should be less susceptible to adverse effects caused by metabolites formed from lower doses of VOCs due to their immature xenobiotic metabolic systems. The Elderly The elderly, like infants and children, may be more or less susceptible than young adults to VOC toxicity. The net effect of pharmacodynamic and pharmacokinetic changes with aging determines the sen- sitivity of geriatric populations. The aging CNS, for example, undergoes pharmacodynamic changes (such as neuronal loss, alteration in neurotransmitter and receptor numbers, and reduction in adaptability to ef- fects of toxicants) that may predispose to neurotoxicity (Ginsberg et al. 2005). Kiesswetter et al. (1997) observed more pronounced neurobehavioral effects of single or mixed solvents in occupational settings in older workers. Data are sorely lacking, however, on susceptibility to most other adverse effects. Toxicokinetic changes during aging have been of interest primarily with respect to therapeutics, although the environmental-health arena is now also focusing attention on geriatric populations (Geller and Zenick 2005). Despite some reduction in pulmonary capacity, inhalation PBTK-model predictions of steady-state blood concentrations of PCE, vinyl chloride, styrene, and isopropanol differ little among 10-, 15-, 25-, 50-, and 75-year-old people (Sarangapani et al. 2003). Systemic clearance of many drugs is typi- cally slower after the age of 60 years, particularly in those more than 80 years old (Ginsberg et al. 2005).
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 81 Slowing of clearance is due largely to diminution in cardiac output, which in turn reduces hepatic blood flow and metabolism and renal blood flow and excretion (McLean and LeCouteur 2004). Clewell et al. (2004) predicted that, for a given magnitude of exposure, blood concentrations of PCE and trichloroacetic acid, its major metabolite, would progressively rise during old age. That was attributed to reduction in pulmonary and metabolic clearance of PCE coupled with its accumulation in relatively large amounts of adipose tissue. Much work remains to be done to refine geriatric PBTK models and to integrate them with age-dependent pharmacodynamic changes. There are sources of variability other than pharmacodynamic and pharmacokinetic changes in re- sponses of geriatric populations to chemicals. They include the common use of multiple medications, in- adequate nutrition, and the prevalence of pre-existing disease states (Schmucker 1985). Compromised organ function can be exacerbated by toxicants in such a way that a modest degree of damage may result in marked dysfunction. In addition, normal aging processes can be accentuated by chemical stressors. Sex Differences It does not appear that women will differ substantially from men in most respects in their re- sponses to TCE and most other VOCs. Uptake and disposition of these lipophilic chemicals, however, can differ because of the higher proportion of body fat in many females. Absorbed doses of inhaled VOCs are usually higher and internal exposure longer in females. Nomiyama and Nomiyama (1974), for example, measured lower TCE concentrations in the exhaled breath of women volunteers after controlled inhalation exposure. Clewell et al. (2004) used a PBTK model to simulate concentrations of PCE and trichloroacetic acid in men and women over a lifetime of daily ingestion of PCE at 1 Âµg/kg. The women were predicted to attain higher blood PCE and trichloroacetic acid concentrations. The major sex differences in cyto- chrome P-450-mediated hepatic metabolism and drug kinetics observed in rats have not been found in humans and other mammals (Schwartz 2003; Bebia et al. 2004). Sex-specific biotransformation data are lacking, however, on most VOCs. Activity of CYP2E1, the major catalyst of oxidation of low concentra- tions of many VOCs, does not differ significantly between men and women (Snawder and Lipscomb 2000). Nevertheless, a sex-specific PBTK model predicts that women will exhibit higher blood benzene concentrations and 23-26% higher benzene metabolism, which might place them at greater risk than men after equivalent exposures (Brown et al. 1998); higher female body fat content was the major factor in this instance. Another PBTK modelâs predictions of steady-state blood concentrations of PCE, vinyl chloride, and styrene were largely sex-independent (Sarangapani et al. 2003). Relatively little is known about po- tential influences of contraceptives or hormone-replacement therapy on the metabolism and disposition of chemicals. Pregnancy Relatively little is known about the potential influence of pregnancy on the absorption, distribu- tion, metabolism, and elimination of VOCs. Physiologic changes that occur during pregnancy may protect against or enhance vulnerability to xenobiotic toxicity. Physiologic changes in gastrointestinal, cardiovas- cular, pulmonary, and renal systems may also affect xenobiotic absorption and elimination (Mattison et al. 1991). Fisher et al. (1989) developed a PBTK model for TCE and its primary metabolite, trichloroace- tic acid, in the pregnant rat. Pregnant rats were exposed to TCE by inhalation, as a single oral bolus, or in drinking water. The PBTK model predicted that fetal exposure to TCE and TCA would be over 60% of the maternal exposure regardless of the exposure route. The results suggested that a developing fetus is at risk of TCE and TCA exposure, but such modeling has not been completed for humans. Biochemical changes during pregnancy may also influence xenobiotic metabolism. Placental and fetal tissues, termed the fetoplacental unit, contain a variety of cytochrome P-450s, the enzyme super- family responsible for much of phase I xenobiotic metabolism (Raucy and Carpenter 1993; Pasanen and
82 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects Pelkonen 1994). Nakajima et al. (1992) found decreased cytochrome P-450 concentrations in the liver of pregnant Wistar rats. Pregnancy also decreased the metabolism of both TCE and toluene by maternal he- patic microsomes. Active CYP2E1 is believed to be present in human placenta at very low or negligible concentrations, although some evidence suggests that placental CYP2E1 may be induced by high expo- sure to ethanol (Rasheed et al. 1997; Hakkola et al. 1996; Botto et al. 1994). In general, those findings imply that the mother and fetus would be less exposed to the toxic metabolites formed via the oxidative metabolic pathway. Conversely, they would be more exposed to the parent compound. Because the pla- centa has little CYP2E1 activity, some amount of oxidative metabolites could be released into fetal circu- lation. It is not clear whether CYP2E1 is present in the human fetus. Vieira et al. (1996) found no evi- dence of human fetal hepatic CYP2E1 before birth, although concentrations of the isozyme rapidly in- crease after birth. In contrast, Carpenter et al. (1996) detected CYP2E1 in human fetal liver during weeks 16-24 of gestation. In addition, CYP2E1 protein concentrations increased in human fetal hepatocytes ex- posed to ethanol or clofibrate. There is no evidence of CYP2B6 mRNA expression or protein in the fetoplacental unit during any stage of pregnancy. Nonetheless, CYP2B6 is believed to be active in the oxidative metabolism of high doses of TCE, PCE, and other VOCs. Further study is needed to clarify those discrepancies in the presence and activity of fetoplacental CYP2E1 and CYP2B6. A new subject of research is the effect of pregnancy on peroxisome-proliferator-activated recep- tors (PPAR). PPARs are transcription factors that belong to the nuclear hormone receptor superfamily. PPARs regulate genes involved in cell differentiation, development, and metabolism. The three identified and described PPAR isoforms are PPARÎ±, PPARÎ²/Î´, and PPARÎ³. Among the isoforms, PPARÎ³ has the greatest influence on cellular homeostasis and carcinogenicity. However, all three PPAR isoforms play essential roles in physiologic change and development in the fetoplacental unit. Abnormalities in PPAR- regulated pathways may be implicated in reproductive and gestational disease (Toth et al. 2007; Borel et al. 2008). Two TCE metabolites, TCA and DCA, can induce PPARÎ± activation in humans. The combined effect of pregnancy and TCE-metabolite-induced PPAR activation is unknown. Genetics A variety of genetic polymorphisms can affect the quantity and quality of enzymes and the out- comes of exposure to solvents (Raunio et al. 1995; Wormhoudt et al. 1999). Such polymorphisms occur with different frequencies in different ethnic groups. It is often difficult to disentangle the influence of genetic traits from those of lifestyle and socioeconomic status. Shimada et al. (1994) report that Cauca- sians have higher total cytochrome P-450 and CYP2E1 concentrations than Japanese. Stephens et al. (1994) describe ethnic differences in the CYP2E1 gene among American blacks, European-Americans, and Taiwanese. Pronounced interethnic differences in rates of ethanol metabolism are associated with alcohol dehydrogenase and aldehyde dehydrogenase polymorphisms. Alcohol dehydrogenase and alde- hyde dehydrogenase catalyze secondary reactions in the TCE oxidative pathway. Inasmuch as CYP2E1 catalyzes the bioactivation of a number of VOCs to cytotoxic or mutagenic products (Guengerich et al. 1991), substantial differences in CYP2E1 concentrations in groups might be expected to result in different susceptibilities to injury. Lipscomb et al. (1997) found that hepatic CYP2E1 activity varied by a factor of about 10 in humans. PBTK model simulations of an 8-h inhalation exposure to TCE at 50 ppm and of consumption of 2 L of water containing TCE at 5 ppb revealed that the amount of VOC oxidized in the liver differed by only 2% in persons with the lowest and highest CYP2E1 content (Lipscomb et al. 2003). That blood delivery of TCE to the liver is much slower than CYP2E1-mediated bioactivation limits the influence of individual variability in CYP2E1. That phenomenon is addressed again below in connection with ethanol induction of TCE metabolism. Results of epidemiologic studies of possible relationships be- tween CYP2E1 concentrations and cancer incidence in VOC-exposed groups have been contradictory, and studies of larger populations and having greater statistical power are needed.
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 83 Other polymorphisms have been examined for their possible role in tumor induction in solvent- exposed populations. Bruning et al. (1997), for example, investigated the prevalence of glutathione S- transferase (GST) isozyme polymorphisms in TCE-exposed workers who had renal-cell carcinoma. The glutathione conjugation pathway appears to be responsible for formation of cytotoxic or genotoxic me- tabolites of TCE and PCE (see earlier section âMetabolic Activation and Inactivation of Trichloroethyl- ene and Perchloroethyleneâ). Bruning et al. (1997) noted that workers who had renal-cell carcinoma were more likely to carry functional GST1 and GSTM1 genes. High percentages of Caucasians and other eth- nic groups lack GSTM1 and GSTT1 (Bolt and Their 2006) and thus might be at reduced risk of renal cell carcinoma from TCE or PCE (Vermeulen and Bladeren 2001). Wiesenhutter et al. (2007), however, found no evidence that GSTM1, GSTP1, or NAT2 deletion polymorphisms affected development of renal cell carcinoma in persons with high occupational exposure to TCE. In conclusion, genetic differences in metabolic activation of TCE by the oxidative pathway do not appear likely to influence toxic or carcinogenic risks posed by the chemical at the concentrations meas- ured in mixed water supplies at Camp Lejeune. Polymorphisms that dictate the presence or absence of genes that code for isozymes that initiate metabolic activation of TCE via the glutathione conjugation pathway are more likely to influence susceptibility to TCE-induced kidney cancer. Lifestyle Dietary habits can influence the absorption, metabolism, and toxicity of VOCs in several ways. VOCs are rapidly absorbed by passive diffusion from all parts of the gastrointestinal tract. On ingestion with dietary fat, the chemicals partition into the lipids, and they remain there until they are emulsified and absorbed. That delays systemic uptake of VOCs, such as carbon tetrachloride, and results in reduced blood concentrations and reduced hepatic damage in rats (Kim et al. 1990a, b). Conversely, consumption of a high-fat diet increases hepatic CYP2E1 activity in rats, which can enhance the bioactivation of car- bon tetrachloride and other VOCs (Raucy et al. 1991). Carbohydrate deficiency also enhances the me- tabolism of solvents. An increasing number of dietary supplements, fruit juices, and vegetable compo- nents are being identified as inducers or inhibitors of cytochrome P-450s (Huang and Lesko 2004). Flavonoids in grapefruit juice were one of the first documented classes of naturally occurring cytochrome P-450 inhibitors. Other potent inhibitors are bergamottin, echinacea, and some constituents of Ginkgo biloba (Chang et al. 2006). Fasting for 1-3 days can significantly enhance the hepatotoxicity of medium to high doses of VOCs that undergo metabolic activation. Fasting results in decreased hepatic concentrations of glu- tathione because of cessation of intake of amino acids required for its synthesis. Glutathione plays a key role in detoxifying electrophilic metabolites of a number of VOCs, such as 1,1-DCE (Jaeger et al. 1974). Conversely, conjugation of glutathione with TCE or PCE can lead to limited formation of cytotoxic, mutagenic metabolites (see section âMetabolic Activation and Inactivationâ). Withholding food for 12-24 h also results in induction of CYP2E1, the major catalyst of activation of many VOCs. Bruckner et al. (2002) found that lack of food intake during sleep results in lipolysis and formation of acetone, an effec- tive CYP2E1 inducer, in rats. The animals were thus more susceptible to acute carbon tetrachloride hepa- totoxicity during their initial waking hours. Long-term food deprivation (starvation), however, results in reduced synthesis of CYP2E1 and other cytochrome P-450s and decreased metabolic activation of VOCs. Physical activity can significantly influence the toxicokinetics of solvents. Exercise increases two of the key determinants of uptake of inhaled VOCs: (1) respiratory and alveolar ventilation rate and (2) cardiac output and pulmonary blood flow. Exercise can double pulmonary uptake of VOCs (Astrand 1983), although this is often not considered in setting occupational exposure standards. Blood flow to the liver and kidneys is diminished with exercise, so biotransformation of well-metabolized solvents de- creases. A PBTK model for methylene chloride predicted that light exercise would result in a doubling of blood concentrations of methylene chloride and of metabolite formation via cytochrome P-450- and glu- tathione-dependent pathways (Dankovic and Bailer 1994).
84 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects Ethanol is an effective CYP2E1 inducer when ingested repeatedly in substantial amounts (Lieber 1997). There are numerous reports of marked potentiation of hepatic or renal damage by ethanol or other alcohols in persons occupationally exposed to potent hepatorenal toxicants, such as carbon tetrachloride (Folland et al. 1976; Manno et al. 1996). A group of moderate drinkers exposed to 1,1,1-trichloroethane vapor at 175 ppm showed a significant increase in metabolism and metabolic clearance of the chemical (Johns et al. 2006). 1,1,1-Trichloroethane is a relatively nontoxic solvent. Kaneko et al. (1994) exposed ethanol-pretreated rats by inhalation to TCE or 1,1,1-trichloroethane at 50-1,000 ppm. 1,1,1- Trichloroethane metabolism was enhanced at all vapor concentrations, but TCE metabolism was en- hanced by ethanol only at the highest concentration (1,000 ppm). The researchers concluded that altera- tions in the rate of biotransformation of low doses of well-metabolized VOCs, such as TCE, are of little consequence toxicologically because their biotransformation is perfusion-limited (limited by hepatic blood flow); most of the TCE entering the liver is metabolized, even in nondrinkers who still have CYP2E1 in excess for the small amounts of TCE arriving in the blood. Kedderis (1997) used a PBTK model to predict that a 10-fold increase in CYP2E1 activity in humans inhaling TCE at 10 ppm would result in only a 2% increase in TCE metabolism by the liver. Thus, increased bioactivation capacity due to ethanol or other factors should not increase risks of toxicity or cancer in Camp Lejeune residents because of their low exposures to TCE, 1,1-DCE, methylene chloride, vinyl chloride, benzene, or other exten- sively metabolized VOCs. As previously described in this chapter, PCE is poorly metabolized, although some of its metabolites are cytotoxic or mutagenic. Kedderis (1997) predicted that a 10-fold increase in CYP2E1 activity in humans inhaling PCE, as opposed to TCE, at 10 ppm would result in a 3.8-fold in- crease in formation of PCE metabolites in the liver. Enzyme induction would result in increased health risks posed by PCE. It should be recognized that the timing of ethanol consumption and VOC exposure is important. Prior repeated exposure to ethanol is necessary for substantial CYP2E1 synthesis to occur. Concurrent exposure to ethanol and a VOC, however, may sometimes be protective against both well-metabolized and poorly metabolized solvents. VOCs and ethanol are both metabolized by CYP2E1, so the two xeno- biotics compete for the available isozyme. That situation is known as competitive metabolic inhibition. Muller et al. (1975) observed that concurrent intake of ethanol and inhalation of TCE at 50 ppm by hu- man subjects resulted in a marked decrease in urinary excretion of TCEâs major metabolites, trichloroace- tic acid and trichloroethanol. In this instance, ethanol would afford protection against TCEâs oxidative metabolites. Metabolism of ethanol produces an excess of nicotinamide adenine dinucleotide, a cofactor that favors formation of trichloroethanol from chloral hydrate, at the expense of trichloroacetic acid. Re- duced formation of trichloroacetic acid would be protective against trichloroacetic acid-induced hepatic tumors. Larson and Bull (1989), however, observed that interaction in rats only with very high doses of TCE and ethanol. Medications and drugs of abuse that induce or inhibit CYP2E1 and other enzymes involved in the metabolism of VOCs can potentially alter the chemicalsâ toxicity or carcinogenicity. Phenobarbital and other barbiturates were among the first recognized cytochrome P-450 inducers. Notable inducers of CYP2E1 include, in addition to alcohols and acetone (Gonzalez 2007), acetaminophen, salicylates, phenytoin, chlorpromazine, isoniazid, and diazepam. Nakajima et al. (1992) showed that pretreatment of rats with phenobarbital, ethanol, or 3-methylcholanthrene significantly increased TCE oxidation. The same would be expected to occur in humans at high TCE doses. Again, cytochrome P-450 induction will probably not be of consequence at the concentrations found in the water supplies at Camp Lejeune. Some drugs (such as cycloheximide, disulfiram, and chloramphenicol) and the aforementioned natural constitu- ents of plants inhibit CYP2E1. Those compounds, in sufficient doses, would be protective against high doses of TCE and other VOCs that are bioactivated by CYP2E1. Tobacco smoke contains a number of compounds that are strong cytochrome P-450 inducers. Polycyclic hydrocarbons, such as 3-methylcholanthrene, are potent inducing agents. The polycyclic hy- drocarbons primarily stimulate synthesis of CYP1A1 and CYP1A2, cytochrome P-450 isozymes that play a modest role in catalyzing the biotransformation of TCE (Nakajima et al. 1992). Nicotine, however, is a
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 85 strong CYP2E1 inducer in rats (Micu et al. 2003). Cigarette smoke is known to induce CYP2E1 in both rodents and humans. Diseases Illness can be a major source of variability in a personâs response to VOCs. Impaired metabolism and systemic clearance of xenobiotics are commonly seen in persons with hepatitis or cirrhosis. Reduc- tion in metabolic capacity results from decrease in liver mass, reduced enzymatic activity, or diminution in liver blood flow. Lower concentrations of CYP2E1, CYP1A2, and glutathione are found in cirrhotic livers (Murray 1992). Lower cytochrome P-450-mediated bioactivation of VOCs can be protective, but reduced capacity to conjugate their electrophilic metabolites would have the opposite effect. Chronic renal disease has become more prevalent in the United States over the last decade (Coresh et al. 2007). Progressive loss of renal function will lead to impaired renal excretion of some po- tentially toxic or carcinogenic metabolites, such as trichloroacetic acid. Trichloroacetic acid is highly bound to albumin and other plasma proteins. Plasma-protein binding is reduced in patients with compro- mised renal function, apparently because of renal retention of substances that compete with trichloroacetic acid for protein-binding sites and because of reduced albumin synthesis. Thus, decreased formation of trichloroacetic acid from TCE and PCE and reduced plasma-protein binding would increase systemic clearance. That may be offset, however, by a decrease in renal excretion (Yuan and Venitz 2000). Im- pairment of renal bioactivation of glutathione metabolic intermediates of TCE and PCE by oxidation or Î²- lyase (see section âMetabolic Activation and Inactivationâ) would be protective (Bruckner et al. 2008). Diabetes mellitus is a metabolic disease characterized by hyperglycemia as a result of insulin de- ficiency (type I) or insulin resistance (type II). Type II diabetes accounts for 90% of cases in the United States. Animal experiments show that type II diabetes increases susceptibility to the toxicity of certain solvents apparently because of inhibition of tissue repair (Sawant et al. 2004). The human relevance of these animal findings is uncertain. CYP2E1 induction is a prominent effect of type I diabetes in rats but not in humans. Type II diabetes results in CYP2E1 induction in humans (Lucas et al. 1998; Wang et al. 2003). Obesity has been shown to result in induction of CYP2E1 in both rats and humans. Rats made obese by the feeding of an energy-rich diet were found to have higher hepatic catalytic activities for a number of CYP2E1 substrates (Raucy et al. 1991). The systemic clearance of chlorzoxazone, a CYP2E1 substrate, was recently shown to be more rapid in rats on a high-fat diet than in normal rats and more rapid in obese rats than in those on the high-fat diet (Khemawoot et al. 2007). CYP2E1 activity in hepatic and adipose-tissue microsomes of the animals followed the same order. Ketone bodies were increased in obese rats, as they were in diabetic animals that had fasted. Two ketone bodies, acetone and Î²- hydroxybutyrate, are CYP2E1 inducers. OâShea et al. (1994) observed that ketone bodies were also in- creased in the blood of volunteers who had fasted. They found that obesity in people was associated with increased 6-hydroxylation of chlorzoxazone. Lucas et al. (1998) similarly observed higher CYP2E1- mediated hydroxylation of chlorzoxazone in 17 obese patients; such people may be at increased risk for cytotoxicity and tumorigenicity from moderate to high, but not very low, VOC exposure. In summary, a number of factors may influence the toxicokinetics and, in turn, the adverse effects of TCE, PCE, and other VOCs. Much research has focused on factors that alter the metabolic activation or inactivation of those chemicals. Consumption of a high-fat diet and obesity can induce (increase the activity of) CYP2E1. Fasting, smoking, ethanol ingestion, acetone exposure, and several drugs induce CYP2E1 activity in the liver and other tissues. CYP2E1 induction can increase the toxic or carcinogenic potency of very high doses of some VOCs (such as TCE and PCE). That does not occur after low expo- sures to TCE and other well-metabolized VOCs (such as benzene, vinyl chloride, and methylene chlo- ride). CYP2E1 induction, however, may increase the potency of slowly metabolized VOCs, such as PCE. Some drugs and the constituents of some foods inhibit CYP2E1 and would be protective against oxidative metabolites of most VOCs.
86 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects INTERACTIONS Many occupational and environmental exposures to VOCs involve multiple chemicals. That is particularly true of contaminated environmental media, in that widespread use of solvents leads to their volatilization and their entry into surface waters and groundwater. A major portion of VOCs spilled onto the ground evaporates. Some, however, leaches through soil into groundwater and remains trapped there. The groundwater at about 90% of 1,608 hazardous-waste sites on the U.S. National Priorities List con- tains VOCs. TCE is the most frequently found of all chemicals, followed by lead, PCE, and benzene (Fay and Mumtaz 1998). The most common four-component VOC mixture is TCE, PCE, 1,1,1- trichloroethene, and 1,1-dichloroethane. ATSDR (2004) published a toxicologic profile addressing poten- tial health risks posed by that four-component mixture. Many U.S. citiesâ drinking-water supplies also contain complex mixtures of VOCs. Total concentrations range from parts per trillion to parts per billion (Moran et al. 2007). Trace amounts (less than 1 ppb) of a variety of VOCs are present in the blood of many nonoccupationally exposed members of the general population (Churchill et al. 2001; Blount et al. 2006). Exposure to multiple VOCs and possibly other chemicals raises the question of the consequences of chemical interactions for human health. Most studies have involved experiments with binary or ternary mixtures. One chemical may have no effect on, potentiate (enhance), or antagonize (inhibit) adverse ac- tions of a second or third chemical. Knowledge of mechanisms of VOC interactions involves largely the influence of one VOC on the metabolic activation or inactivation of another. Koizumi et al. (1982) pub- lished the results of one of the first such studies. They found that coexposure of rats to PCE and 1,1,1,- trichloroethane resulted in significant suppression of 1,1,1,-trichloroethane metabolism. Workers exposed to TCE and PCE were found to have lower urinary concentrations of TCE metabolites than workers ex- posed to TCE alone (Seiji et al. 1989). Such an interaction resulted from competitive metabolic inhibition, wherein the amounts of the combined chemicals exceeded the metabolic capacity of the study subjects. Such an interaction is protective against cytotoxicity and carcinogenicity in that the bioactivation of both TCE and PCE is reduced. Conversely, systemic concentrations of the parent compounds would be in- creased, and this might increase neurologic effects. PBTK modeling has been used by several research groups to predict the metabolic and toxi- cologic consequences of exposure to VOC mixtures. Competitive metabolic inhibition was evident in a PBTK-model approach to studying TCE and 1,1-DCE (El-Masri et al. 1996) and TCE and vinyl chloride (Barton et al. 1995). Later PBTK modeling efforts predicted interaction thresholds below which competi- tive metabolic inhibition would not occur. Dobrev et al. (2001), for example, reported that the thresholds for interaction of TCE with PCE and 1,1,1,-trichloroethane vapor in rats were 25 and 135 ppm, respec- tively, when the TCE concentration was 50 ppm. Those findings imply that protection from adverse ef- fects would occur in occupational settings when vapor concentrations were relatively high. An increase in blood TCE concentrations under these exposure conditions was predicted to result in a disproportionate increase in formation of nephrotoxic glutathione conjugation products in humans (Dobrev et al. 2002). Other PBTK modeling approaches are being developed to simulate the metabolic outcome of human ex- posures to up to four common VOC water pollutants (for example, TCE, PCE, chloroform, and 1,1,1- trichloroethane) (Mayeno et al. 2005). Competitive metabolic inhibition, with potentiation or protection from adverse effects of VOCs, would not occur at much lower exposure concentrations. Competitive metabolic inhibition and antagonism of (protection from) adverse effects of the VOCs would not occur at much lower exposures, such as those at Camp Lejeune. Additivity of toxic effects of chemicals that act by similar mechanisms is typically assumed in the absence of experimental evidence to the contrary. There does not appear to be experimental evidence of greater than additive interactions of VOCs (ATSDR 2004). One possible mechanism of potentiation is induction of CYP2E1 by one or more members of a VOC mixture. Experiments in rats dosed with single VOCs have shown that most of the compounds are not effective inducers of CYP2E1 or other cytochrome P-450 isozymes. Competitive metabolic inhibition, as described above, would result in antagonism of (that is, less than additive) adverse effects if metabolites are the bioactive moieties. Goldsworthy and
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 87 Popp (1987) found that the joint effect of TCE and PCE on peroxisome proliferation in the liver and kid- neys of mice and rats was less than additive. Stacey (1989) studied the joint action of TCE and PCE on the liver and kidneys of rats. Combined administration of near-toxic-threshold doses of the two solvents produced modest hepatorenal toxicity. Jonker et al. (1996) provided evidence that TCE and PCE in com- bination with two other similarly acting solvents affected kidney weight in rats given subtoxic doses of each chemical by gavage for 32 days. Competitive metabolic inhibition at relatively high exposure levels of toluene, ethylbenzene, and xylene has been predicted by PBTK modeling to result in higher internal exposures (and CNS depressant effects) than would occur with simple additivity (Dennison et al. 2005). Although experimental data are limited, the assumption of additivity of potential risks posed by VOC wa- ter contaminants at Camp Lejeune seems to be a reasonable, prudent approach. A few toxicity or carcinogenicity studies of complex chemical mixtures, including VOCs, have been conducted. The National Toxicology Program (NTP 1993) supplied F-344 rats and B6C3F1 mice with drinking water containing 25 contaminants for up to 26 weeks. The mixture contained TCE, PCE, methylene chloride, 1,1,1-trichloroethane, 1,1-DCE, 1,1-dichloroacetic acid, other solvents, heavy metals, polychlorinated biphenyls, and a phthalate. The total no-observed-adverse-effect levels for histologic changes in organs were 11 ppm in rats and 378 ppm in mice. Suppression of immune function occurred in female mice that consumed the mixture at 756 ppm for 2 weeks or 378 ppm for 13 weeks. Constan et al. (1996) saw centrilobular hyperplasia and apoptosis in the livers of rats after 1 mo. A followup study in chemically tumor-initiated rats showed that the contaminant mixture did not promote preneoplastic foci in the liver (Benjamin et al. 1999). Wang et al. (2002) supplied ICR mice with water containing chloroform, 1,1-dichloroacetic acid, 1,1-DCE, 1,1,1-trichloroethane, TCE, and PCE for 16 and 18 mo. There was a trend of increasing frequency of hepatocellular neoplasms in the male mice and increasing incidence of mammary adenocarcinomas in the high-dose female mice. The total concentration of VOCs in the drink- ing water of females was about 1,555 ppb. Most of the mixture was TCE (471 ppb) and PCE (606 ppb). Those concentrations are far lower than have previously been reported to produce tumors. The results must be regarded as preliminary in that the study design had a number of limitations, and the results have not been replicated. In addition, male B6C3F1 mice are particularly susceptible to hepatic tumors, and mice metabolically activate a substantially greater proportion of solvent doses than do humans. Multiple VOCs and other chemicals are commonly present in trace amounts (parts per trillion to parts per billion) in water from contaminated wells in the United States. The Environmental Protection Agency, in the absence of information to the contrary, assumes that any adverse effects of chemicals that act by the same mechanism are additive. Several VOCs act on some organs by similar mechanisms. Ani- mal experiments with high doses of combined VOCs have shown that one VOC inhibits the metabolic activation (that is, protects against adverse effects) of the other. That would not occur at the lower con- centrations that were found in the water supplies at Camp Lejeune. SUMMARY Residents of homes supplied with contaminated water can be exposed orally by drinking the wa- ter, as well as by inhalation and dermal exposure when using heated water for bathing, showering, and washing clothes and dishes. Experiments with TCE and chloroform have shown that ingestion and inhala- tion make comparable contributions to systemically absorbed doses, and the contribution from skin ab- sorption is minor. The concept of dose has been refined to three components: administered, or external dose; sys- temically absorbed, or internal dose; and target organ and tissue dose. It is most important to specify the dose of the bioactive moiety, whether it is the parent compound or one or more metabolites. Concurrent pharmacokinetic processes, including absorption, tissue distribution, binding, metabolism, and elimina- tion, determine tissue doses. One or more of these processes can vary significantly from one route of ex- posure to another, from one species to another, and from one person to another. Understanding how these processes differ can factor into predicting toxicity and cancer risks for various exposure scenarios.
88 Contaminated Water Supplies at Camp LejeuneâAssessing Potential Health Effects PCE, TCE, and other VOCs are quickly and extensively absorbed from the gastrointestinal tract. These small, uncharged, lipophilic molecules rapidly diffuse through membranes from areas of higher to lower concentration. It is typically assumed that 100% of doses of orally administered VOCs are ab- sorbed. A portion of VOCs reaching the pulmonary blood are exhaled before reaching the arterial circula- tion. Pulmonary and hepatic first-pass elimination acting in concert are responsible for removing almost 90% of very low doses of TCE, thereby affording extrahepatic organs protection from noncancer and can- cer effects from trace concentrations of such chemicals in drinking water. Less protection from poorly- metabolized VOCs (for example, PCE) is afforded. The pattern of consumption of contaminated water can substantially influence the toxicologic outcome. Differences in the type of controlled exposure used in animal studies compared with intermittent exposures in humans raises the question of the relevance of such cancer bioassay results to real-life human exposures. TCE, PCE, and other VOC vapors are also very well absorbed from the lungs. Pulmonary absorp- tion is largely determined by the chemicalâs blood:air partition coefficient, the animalâs alveolar ventila- tion rate, and its cardiac output. The ratsâ TCE blood:air partition coefficient is almost three times that of humans. Resting alveolar ventilation rates and cardiac outputs are markedly higher in mice than in rats and significantly higher in rats than in humans. Metabolism plays a key role in modulating the kinetics, and in turn the injury potential of VOCs. These chemicals can be biotransformed to more toxic or less toxic derivatives. The majority of metabo- lism occurs in the liver. TCE and PCE are metabolized by two primary metabolic pathways: cytochrome P-450s-catalyzed oxidation and glutathione S-transferase-mediated conjugation. The oxidation pathway accounts for the majority of metabolism of low-to-moderate doses of TCE and PCE. Oxidative metabo- lites are largely responsible for liver and lung toxicity and carcinogenicity. GSH conjugation becomes more prominent when high doses begin to saturate oxidation. TCE and PCE are metabolized quite simi- larly, although PCE is somewhat more potent because of formation of additional toxic products. Oxida- tive activation of TCE and PCE is much greater in mice and rats than in humans. Metabolic activation by the GSH pathway is substantially greater in rats than in humans. It is well-established that rodents absorb more inhaled TCE and PCE, metabolically activate a greater proportion, and detoxify epoxide metabolites less efficiently than humans. It is not clear whether infants and children are more susceptible to adverse effects of VOCs. Age- dependent changes in pharmacokinetics and pharmacodynamics may make an immature human more or less sensitive, depending upon the individualâs age, the chemical, and the organ system. Low concentra- tions of CYP2E1 in neonates and infants will result in increased TCE concentrations but low concentra- tions of oxidative metabolites. Conversely, children have a relatively large liver and high liver blood flow, placing them at greater risk than adults from effects of oxidative metabolites. Age-related changes in one toxicokinetic process may be augmented or offset by concurrent changes in other processes. Cells in de- veloping organs (for example, neurons in the brain) are more sensitive to injury. Thus, toxicant exposure during such a âwindow of susceptibilityâ can have serious, long-lasting consequences. The net effect of anatomical and physiologic immaturities is difficult to predict, particularly for classes of chemicals (for example, VOCs) for which there is very little information from animal or human studies. The net effect of toxicokinetic and toxicodynamic changes during aging is the major determinant of susceptibility of geriatric populations. It has been predicted with a PBTK model that PCE exposure will result in increased PCE concentrations in the elderly. Unfortunately, there are even fewer experimen- tal data from geriatric humans or animals with which to verify outcomes than there are data from pediatric populations. Additional compounding factors in the elderly include use of multiple medications, poor nu- trition, and preexisting disease states. Women do not appear to differ substantially from men in their responses to TCE, PCE, and other VOCs. Metabolism of solvents is not sex-dependent, but higher female body-fat content results in accu- mulation of higher body burdens of the lipophilic chemicals and increased formation of their metabolites. Relatively little is known about the influence of pregnancy on maternal and fetal disposition of VOCs and their metabolites. Animal models, however, show lower maternal TCE metabolism during pregnancy and limited fetal exposure to oxidative metabolites.
Syxtemic Exposures to VOCs and Factors Influencing Susceptability to their Effects 89 A variety of genetic polymorphisms in human populations can affect the quantity and quality of CYP450 and glutathione S-transferase enzymes and, in turn, the outcomes of exposure to solvents. There are marked interindividual differences in activity of hepatic CYP2E1, the primary isozyme responsible for metabolic oxidation of TCE. This interindividual difference is not believed to be toxicologically signifi- cant, however, for persons exposed to very low concentrations of TCE and other well-metabolized VOCs. The interindividual difference in oxidative capacity may be important, however, in the extent of metabolic activation and response to poorly-metabolized VOCs, such as PCE. Lifestyle can potentially influence an individualâs responses to VOCs in a number of ways. Die- tary habits and components, physical activity, ethanol intake, and certain drugs can affect metabolism and deposition of solvents. Serious illness, impaired metabolism and systemic clearance of parent compounds, and obesity are some additional factors that can affect the way the body handles exposure to TCE and PCE. Many occupational and environmental exposures to VOCs involve multiple chemicals. Knowl- edge of mechanisms of chemical interactions largely involves the effect of one VOC on the metabolic activation of a second. Concurrent exposures to sufficiently high doses typically involve competitive metabolic inhibition, which results in increased concentrations of parent compounds and lower production of metabolites. Such interactions will not occur at very low exposure concentrations.