Incorporating human interindividual biotransformation variance in health risk assessment

IVIVE: Facilitating the Use of In Vitro Toxicity Data in Risk Assessment and Decision Making

During the past few decades, the science of toxicology has been undergoing a transformation from observational to predictive science. New approach methodologies (NAMs), including in vitro assays, in silico models, read-across, and in vitro to in vivo extrapolation (IVIVE), are being developed to reduce, refine, or replace whole animal testing, encouraging the judicious use of time and resources. Some of these methods have advanced past the exploratory research stage and are beginning to gain acceptance for the risk assessment of chemicals. A review of the recent literature reveals a burst of IVIVE publications over the past decade. In this review, we propose operational definitions for IVIVE, present literature examples for several common toxicity endpoints, and highlight their implications in decision-making processes across various federal agencies, as well as international organizations, including those in the European Union (EU). The current challenges and future needs are also summarized for IVIVE. In addition to refining and reducing the number of animals in traditional toxicity testing protocols and being used for prioritizing chemical testing, the goal to use IVIVE to facilitate the replacement of animal models can be achieved through their continued evolution and development, including a strategic plan to qualify IVIVE methods for regulatory acceptance.

Identifying populations sensitive to environmental chemicals by simulating toxicokinetic variability

The thousands of chemicals present in the environment (USGAO, 2013) must be triaged to identify priority chemicals for human health risk research. Most chemicals have little of the toxicokinetic (TK) data that are necessary for relating exposures to tissue concentrations that are believed to be toxic. Ongoing efforts have collected limited, in vitro TK data for a few hundred chemicals. These data have been combined with biomonitoring data to estimate an approximate margin between potential hazard and exposure. The most "at risk" 95th percentile of adults have been identified from simulated populations that are generated either using standard "average" adult human parameters or very specific cohorts such as Northern Europeans. To better reflect the modern U.S. population, we developed a population simulation using physiologies based on distributions of demographic and anthropometric quantities from the most recent U.S. Centers for Disease Control and Prevention National Health and Nutrition Examination Survey (NHANES) data. This allowed incorporation of inter-individual variability, including variability across relevant demographic subgroups. Variability was analyzed with a Monte Carlo approach that accounted for the correlation structure in physiological parameters. To identify portions of the U.S. population that are more at risk for specific chemicals, physiologic variability was incorporated within an open-source high-throughput (HT) TK modeling framework. We prioritized 50 chemicals based on estimates of both potential hazard and exposure. Potential hazard was estimated from in vitro HT screening assays (i.e., the Tox21 and ToxCast programs). Bioactive in vitro concentrations were extrapolated to doses that produce equivalent concentrations in body tissues using a reverse dosimetry approach in which generic TK models are parameterized with: 1) chemical-specific parameters derived from in vitro measurements and predicted from chemical structure; and 2) with physiological parameters for a virtual population. For risk-based prioritization of chemicals, predicted bioactive equivalent doses were compared to demographic-specific inferences of exposure rates that were based on NHANES urinary analyte biomonitoring data. The inclusion of NHANES-derived inter-individual variability decreased predicted bioactive equivalent doses by 12% on average for the total population when compared to previous methods. However, for some combinations of chemical and demographic groups the margin was reduced by as much as three quarters. This TK modeling framework allows targeted risk prioritization of chemicals for demographic groups of interest, including potentially sensitive life stages and subpopulations.

Interspecies uncertainty in molecular responses and toxicity of mixtures

Most of the experimental toxicity testing data for chemicals are generated through the use of laboratory animals, namely, rodents such as rats and mice or other species. Interspecies extrapolation is needed to nullify the differences between species so as to use such data for human health/risk assessment. Thus, understanding of interspecies differences is important in extrapolating the laboratory results to humans and conducting human risk assessments based on current credible scientific knowledge. Major causes of interspecies differences in anatomy and physiology, toxicokinetics, injury repair, molecular receptors, and signal transduction pathways responsible for variations in responses to toxic chemicals are outlined. In the risk assessment process, uncertainty associated with data gaps in our knowledge is reflected by application of uncertainty factors for interspecies differences. Refinement of the risk assessment methods is the ultimate goal as we strive to realistically evaluate the impact of toxic chemicals on human populations. Using specific examples from current risk assessment practice, this chapter illustrates the integration of interspecies differences in evaluation of individual chemicals and chemical mixtures.

Interspecies extrapolation

Interspecies extrapolation encompasses two related but distinct topic areas that are germane to quantitative extrapolation and hence computational toxicology-dose scaling and parameter scaling. Dose scaling is the process of converting a dose determined in an experimental animal to a toxicologically equivalent dose in humans using simple allometric assumptions and equations. In a hierarchy of quantitative extrapolation approaches, this option is used when minimal information is available for a chemical of interest. Parameter scaling refers to cross-species extrapolation of specific biological processes describing rates associated with pharmacokinetic (PK) or pharmacodynamic (PD) events on the basis of allometric relationships. These parameters are used in biologically based models of various types that are designed for not only cross-species extrapolation but also for exposure route (e.g., inhalation to oral) and exposure scenario (duration) extrapolation. This area also encompasses in vivo scale-up of physiological rates determined in various experimental systems. Results from in vitro metabolism studies are generally most useful for interspecies extrapolation purposes when integrated into a physiologically based pharmacokinetic (PBPK) modeling framework. This is because PBPK models allow consideration and quantitative evaluation of other physiological factors, such as binding to plasma proteins and blood flow to the liver, which may be as or more influential than metabolism in determining relevant dose metrics for risk assessment.

Genetic polymorphism in metabolism and host defense enzymes: implications for human health risk assessment

Genetic polymorphisms in xenobiotic metabolizing enzymes can have profound influence on enzyme function, with implications for chemical clearance and internal dose. The effects of polymorphisms have been evaluated for certain therapeutic drugs but there has been relatively little investigation with environmental toxicants. Polymorphisms can also affect the function of host defense mechanisms and thus modify the pharmacodynamic response. This review and analysis explores the feasibility of using polymorphism data in human health risk assessment for four enzymes, two involved in conjugation (uridine diphosphoglucuronosyltransferases [UGTs], sulfotransferases [SULTs]), and two involved in detoxification (microsomal epoxide hydrolase [EPHX1], NADPH quinone oxidoreductase I [NQO1]). This set of evaluations complements our previous analyses with oxidative and conjugating enzymes. Of the numerous UGT and SULT enzymes, the greatest likelihood for polymorphism effect on conjugation function are for SULT1A1 (*2 polymorphism), UGT1A1 (*6, *7, *28 polymorphisms), UGT1A7 (*3 polymorphism), UGT2B15 (*2 polymorphism), and UGT2B17 (null polymorphism). The null polymorphism in NQO1 has the potential to impair host defense. These highlighted polymorphisms are of sufficient frequency to be prioritized for consideration in chemical risk assessments. In contrast, SNPs in EPHX1 are not sufficiently influential or defined for inclusion in risk models. The current analysis is an important first step in bringing the highlighted polymorphisms into a physiologically based pharmacokinetic (PBPK) modeling framework.

The impact of CYP2E1 genetic variability on risk assessment of VOC mixtures

Humans are simultaneously exposed to multiple chemicals in the environment. Many of the chemicals use the same enzymes in their metabolic pathways. Competitive inhibition may occur as one of the possible interactions between the xenobiotics in human body. For example, many volatile organic compounds (VOCs) are metabolized using P450 enzymes, specifically CYP2E1. Inheritable gene alterations may result in changes of function of the enzymes in different human subpopulations. Variations in quantity and/or quality of particular isoenzymes may cause differences in the metabolism of VOCs. These variations may cause higher sensitivity in certain populations. Using examples of three different mixtures, this review paper outlines the variances in CYP2E1 isoenzymes, effect of exposure to such mixtures on sensitive populations, and approaches to mixtures risk assessment.

The influence of genetic polymorphisms on population variability in six xenobiotic-metabolizing enzymes

This review provides variability statistics for polymorphic enzymes that are involved in the metabolism of xenobiotics. Six enzymes were evaluated: cytochrome P-450 (CYP) 2D6, CYP2E1, aldehyde dehydrogenase-2 (ALDH2), paraoxonase (PON1), glutathione transferases (GSTM1, GSTT1, and GSTP1), and N-acetyltransferases (NAT1 and NAT2). The polymorphisms were characterized with respect to (1) number and type of variants, (2) effects of polymorphisms on enzyme function, and (3) frequency of genotypes within specified human populations. This information was incorporated into Monte Carlo simulations to predict the population distribution and describe interindividual variability in enzyme activity. The results were assessed in terms of (1) role of these enzymes in toxicant activation and clearance, (2) molecular epidemiology evidence of health risk, and (3) comparing enzyme variability to that commonly assumed for pharmacokinetics. Overall, the Monte Carlo simulations indicated a large degree of interindividual variability in enzyme function, in some cases characterized by multimodal distributions. This study illustrates that polymorphic metabolizing systems are potentially important sources of pharmacokinetic variability, but there are a number of other factors including blood flow to liver and compensating pathways for clearance that affect how a specific polymorphism will alter internal dose and toxicity. This is best evaluated with the aid of physiologically based pharmacokinetic (PBPK) modeling. The population distribution of enzyme activity presented in this series of articles serves as inputs to such PBPK modeling analyses.

Genetic polymorphism in CYP2E1: Population distribution of CYP2E1 activity

Cytochrome P-450 2E1 (CYP2E1) is a key enzyme in the metabolic activation of a variety of toxicants including nitrosamines, benzene, vinyl chloride, and halogenated solvents such as trichloroethylene. CYP2E1 is also one of the enzymes that metabolizes ethanol to acetaldehyde, and is induced by recent ethanol ingestion. There is evidence that interindividual variability in the expression and functional activity of this cytochrome (CYP) may be considerable. Genetic polymorphisms in CYP2E1 were identified and linked to altered susceptibility to hepatic cirrhosis induced by ethanol and esophageal and other cancers in some epidemiological studies. Therefore, it is important to evaluate how such polymorphisms affect CYP2E1 function and whether it is possible to construct a population distribution of CYP2E1 activity based upon the known effects of these polymorphisms and their frequency in the population. This analysis is part of the genetic polymorphism database project described in the lead article in this series and followed the approach described in that article (Ginsberg et al., 2009, this issue). Review of the literature found that there are a variety of CYP2E1 variant alleles but the functional significance of these variants is still unclear. Some, but not all, studies suggest that several upstream 5' flanking mutations affect gene expression and response to inducers such as ethanol or obesity. None of the coding-region variants consistently affects enzyme function. Part of the reason for conflicting evidence regarding genotype effect on phenotype may be due to the wide variety of exposures such as ethanol or dietary factors and physiological factors including body weight or diabetes that modulate CYP2E1 expression. In conclusion, evidence is too limited to support the development of a population distribution of CYP2E1 enzyme activity based upon genotypes. Health risk assessments may best rely upon data reporting interindividual variability in CYP2E1 function for input into physiologically based pharmacokinetic (PBPK) models involving CYP2E1 substrates.

Cross-species comparison of conazole fungicide metabolites using rat and rainbow trout (Onchorhynchus mykiss) hepatic microsomes and purified human CYP 3A4

Ecological risk assessment frequently relies on cross-species extrapolation to predict acute toxicity from chemical exposures. A major concern for environmental risk characterization is the degree of uncertainty in assessing xenobiotic biotransformation processes. Although inherently complex, metabolite identification is critical to risk assessment since the product(s) formed may pose a greater toxicological threat than the parent molecule. This issue is further complicated by differences observed in metabolic transformation pathways among species. Conazoles represent an important class of azole fungicides that are widely used in both pharmaceutical and agricultural applications. The antifungal property of conazoles occurs via complexation with the cytochrome P450 monooxygenases (CYP) responsible for mediating fungal cell wall synthesis. This mode of action has cause for concern regarding the potential adverse impact of conazoles on the broad spectrum of CYP-based processes within mammalian and aquatic species. In this study, in vitro metabolic profiles were determined for thirteen conazole fungicides using rat and rainbow trout (Oncorhynchus mykiss) liver microsomes and purified human CYP 3A4. Results showed that 10 out of the 13 conazoles tested demonstrated identical metabolite profiles among rat and trout microsomes, and these transformations were well conserved via both aromatic and aliphatic hydroxylation and carbonyl reduction processes. Furthermore, nearly all metabolites detected in the rat and trout microsomal assays were detected within the human CYP 3A4 assays. These results indicate a high degree of metabolic conservation among species with an equivalent isozyme activity of human CYP 3A4 being present in both the rat and trout, and provides insight into xenobiotic biotransformations needed for accurate risk assessment.

Evaluation of physiologically based pharmacokinetic models for use in risk assessment

Physiologically based pharmacokinetic (PBPK) models are sophisticated dosimetry models that offer great flexibility in modeling exposure scenarios for which there are limited data. This is particularly of relevance to assessing human exposure to environmental toxicants, which often requires a number of extrapolations across species, route, or dose levels. The continued development of PBPK models ensures that regulatory agencies will increasingly experience the need to evaluate available models for their application in risk assessment. To date, there are few published criteria or well-defined standards for evaluating these models. Herein, important considerations for evaluating such models are described. The evaluation of PBPK models intended for risk assessment applications should include a consideration of: model purpose, model structure, mathematical representation, parameter estimation, computer implementation, predictive capacity and statistical analyses. Model purpose and structure require qualitative checks on the biological plausibility of a model. Mathematical representation, parameter estimation, computer implementation involve an assessment of the coding of the model, as well as the selection and justification of the physical, physicochemical and biochemical parameters chosen to represent a biological organism. Finally, the predictive capacity and sensitivity, variability and uncertainty of the model are analysed so that the applicability of a model for risk assessment can be determined. Published in 2007 by John Wiley & Sons, Ltd.

The impact of cytochrome P450 2E1-dependent metabolic variance on a risk-relevant pharmacokinetic outcome in humans

Risk assessments include assumptions about sensitive subpopulations, such as the fraction of the general population that is sensitive and the extent that biochemical or physiological attributes influence sensitivity. Uncertainty factors (UF) account for both pharmacokinetic (PK) and pharmacodynamic (PD) components, allowing the inclusion of risk-relevant information to replace default assumptions about PK and PD variance (uncertainty). Large numbers of human organ donor samples and recent advances in methods to extrapolate in vitro enzyme expression and activity data to the intact human enable the investigation of the impact of PK variability on human susceptibility. The hepatotoxicity of trichloroethylene (TCE) is mediated by acid metabolites formed by cytochrome P450 2E1 (CYP2E1) oxidation, and differences in the CYP2E1 expression are hypothesized to affect susceptibility to TCE's liver injury. This study was designed specifically to examine the contribution of statistically quantified variance in enzyme content and activity on the risk of hepatotoxic injury among adult humans. We combined data sets describing (1) the microsomal protein content of human liver, (2) the CYP2E1 content of human liver microsomal protein, and (3) the in vitro Vmax for TCE oxidation by humans. The 5th and 95th percentiles of the resulting distribution (TCE oxidized per minute per gram liver) differed by approximately sixfold. These values were converted to mg TCE oxidized/h/kg body mass and incorporated in a human PBPK model. Simulations of 8-hour inhalation exposure to 50 ppm and oral exposure to 5 micro g TCE/L in 2 L drinking water showed that the amount of TCE oxidized in the liver differs by 2% or less under extreme values of CYP2E1 expression and activity (here, selected as the 5th and 95th percentiles of the resulting distribution). This indicates that differences in enzyme expression and TCE oxidation among the central 90% of the adult human population account for approximately 2% of the difference in production of the risk-relevant PK outcome for TCE-mediated liver injury. Integration of in vitro metabolism information into physiological models may reduce the uncertainties associated with risk contributions of differences in enzyme expression and the UF that represent PK variability.

Inhibition of aflatoxin B1 mutagenicity by cyclopiazonic acid in the presence of human liver preparations

Co-occurrence of cyclopiazonic acid (CPA) and aflatoxin B(1) (AFB(1)) has been reported in different food commodities. Recently, we have shown that CPA reduces AFB(1) mutagenicity in the standard Salmonella-Microsome-Assay using rat S9-mix for metabolic activation (Environ. Toxicol. Pharmacol. 11 (2002) 207). When using S9-mix prepared from individual liver fractions of human patients, CPA was found to be non-mutagenic, but exerted a significant reduction of the mutagenicity of AFB(1). Moreover, CPA was shown to inhibit testosterone hydroxylation, but not methoxyresorufin dealkylation (MROD), in human S9. Thus, the reduction of the AFB(1) mutagenicity by CPA may be attributed to the inhibitory effect of CPA on cytochrome P450 (CYP450) 3A4 activity. These findings might be of relevance to the epidemiology of food-borne mycotoxicosis as similar molar ratios to those investigated here have been reported in food commodities.

Application of in vitro biotransformation data and pharmacokinetic modeling to risk assessment

The adverse biological effects of toxic substances are dependent upon the exposure concentration and the duration of exposure. Pharmacokinetic models can quantitatively relate the external concentration of a toxicant in the environment to the internal dose of the toxicant in the target tissues of an exposed organism. The exposure concentration of a toxic substance is usually not the same as the concentration of the active form of the toxicant that reaches the target tissues following absorption, distribution, and biotransformation of the parent toxicant. Biotransformation modulates the biological activity of chemicals through bioactivation and detoxication pathways. Many toxicants require biotransformation to exert their adverse biological effects. Considerable species differences in biotransformation and other pharmacokinetic processes can make extrapolation of toxicity data from laboratory animals to humans problematic. Additionally, interindividual differences in biotransformation among human populations with diverse genetics and lifestyles can lead to considerable variability in the bioactivation of toxic chemicals. Compartmental pharmacokinetic models of animals and humans are needed to understand the quantitative relationships between chemical exposure and target tissue dose as well as animal to human differences and interindividual differences in human populations. The data-based compartmental pharmacokinetic models widely used in clinical pharmacology have little utility for human health risk assessment because they cannot extrapolate across dose route or species. Physiologically based pharmacokinetic (PBPK) models allow such extrapolations because they are based on anatomy, physiology, and biochemistry. In PBPK models, the compartments represent organs or groups of organs and the flows between compartments are actual blood flows. The concentration of a toxicant in a target tissue is a function of the solubility of the toxicant in blood and tissues (partition coefficients), blood flow into the tissue, metabolism of the toxicant in the tissue, and blood flow out of the tissue. The appropriate degree of biochemical detail can be added to the PBPK models as needed. Comparison of model simulations with experimental data provides a means of hypothesis testing and model refinement. In vitro biotransformation data from studies with isolated liver cells or subcellular fractions from animals or humans can be extrapolated to the intact organism based upon protein content or cell number. In vitro biotransformation studies with human liver preparations can provide quantitative data on human interindividual differences in chemical bioactivation. These in vitro data must be integrated into physiological models to understand the true impact of interindividual differences in chemical biotransformation on the target organ bioactivation of chemical contaminants in air and drinking water.