Combinations of susceptible genotypes and individual responses to toxicants.

The variation in individual responses to exogenous agents has been shown to be exceptionally wide. It is because of this large diversity of responsiveness that risk factors to environmentally induced diseases have been difficult to pinpoint, particularly at low exposure levels. Opportunities now exist for studies of host factors in environmentally induced cancer or other diseases in which an environmental component can be presumed. Many of the studies have shown an elevated disease proneness for individuals carrying the potential at-risk alleles of metabolic genes, but a number of controversial results have also been reported. One possible explanation for the divergent findings is lack of knowledge of the other potentially relevant genotypes for a given exposure. This paper gives an overview of the published data on combinations of metabolic genotypes in relation to individual susceptibility to environmental toxicants.


Introduction
People are exposed to agents (chemical, physical, and biological) that may contribute to the prevalence of genetic damage and chronic diseases in the population. Many of the diseases are affected by both host factors and the external environment. For instance, although carcinogenesis in humans is known to be significantly influenced by external factors, the metabolic activation or inactivation of the procarcinogens may modulate the process. This is best exemplified by tobacco smoking; cigarette smoking is the main cause of lung cancer but only a minority of smokers develop pulmonary cancers.
Recent findings suggest that inherited differences in metabolic capacity may in fact play a primary role in susceptibility to environmentally induced diseases (1)(2)(3)(4). Genetic polymorphisms exist in a number of phase I (activating) and phase II (inactivating) enzymes. It is conceivable that individuals with genotypes associated with a more efficient activating enzyme and a less efficient inactivating enzyme might be at particularly high risk of adverse health effects, if exposed to toxicants. Phase I enzymes cytochrome P450 lAl (CYPlA1) and 2D6 (CYP2D6) are the most studied candidates as modifiers of individual responses to environmental agents (1)(2)(3)(4). The rare Val and m2 alleles of CYPIAJ gene (frequencies 1-2% in Caucasians) may increase individual cancer risk by heightening aryl hydrocarbon hydroxylase (AHH) enzyme inducibility. In contrast, the segment of Caucasians (approximately 7%) who possess two deficient CYP2D6 alleles (poor metabolizer [PM] genotype) may be at decreased risk of cancer (1-4), but at increased risk of several central nervous system disorders (5)(6)(7)(8).
Among phase II enzymes, N-acetyltransferases (NATs) and glutathione Stransferases (GSTs) have attracted most of the recent interest. The GST MI (GSTM1) and TI (GSTTI) genes are polymorphic so that the absence of enzyme activity results from a homozygous deletion of the respective gene, called the null genotype (9,10). About 50% of Caucasians have the GSTMI null genotype, which may pose an increased risk of various environmentally induced cancers (11)(12)(13)(14). The GSTTI null genotype, which putatively has a frequency of 15 to 25% in Caucasians, has been associated with enhanced susceptibility to primary brain tumors (astrocytoma and meningioma) (15), and to myelodysplastic syndromes, which are clonal proliferative disorders of bone marrow that often progress to acute myeloid leukemia (16).
The human N-acetylation polymorphism is demonstrable by individual variations in metabolism of several substrates, including sulfamethazine and isoniazid (17). The differences in the metabolism of these compounds distinguish phenotypically slow and fast acetylators. The slow acetylators, whose frequency worldwide ranges from about 10% to more than 90% (18), may be at increased risk of arylamineinduced cancers (19)(20)(21)(22).
Previous studies on acetylation polymorphisms have focused on the NAT2 gene locus, and allelic variants of NA T2 gene correlate with differences in acetylation capacity (23)(24)(25)(26). However, recent evidence shows structural heterogeneity in the NATI gene locus also, correlating with N-acetylation activity distinct from that typified by isoniazid and sulfamethazine (27,28). It has been suggested that NATI is primarily responsible for NAT activity in human uroepithelium (29), and correlation between NATI polyadenylation signal polymorphism and risk of colorectal cancer has been reported (30).
Given the number and variability in expression of carcinogen-metabolizing enzymes and the complexity of chemical exposures, assessment of a single polymorphic genotype cannot be expected to be sufficient for evaluating individual susceptibility to environmental agents. Establishment of a broader risk profile for each individual or subgroup is required. The relatively scarce data presently available on the combinations of susceptible genotypes in individual responses to toxicants is reviewed below.

Combined Genotypes and Disease Susceptibility
The first observation of the combined effect of metabolic genes in disease proneness was by Hayashi and co-workers (31) who described a 5.8-fold relative risk (95% CI 2.3-13.3) for all lung cancer types and a 9.1-fold relative risk (95% CI 3.4-24.4) for squamous cell carcinoma of the lung in Japanese individuals who were homozygous for the CYPIAI Val allele and concurrently lacked the GSTM1 gene. A similar, although less pronounced risk (OR 3.0, 95% CI 1.2-7.2) for developing this histological Environmental Health Perspectives -Vol 105, Supplement 4 * June 1997 type of lung cancer was attributed to Caucasians carrying the CYPIAI allele m2, and having the GSTMI null genotype (32). Subsequent evaluation of the effect of smoking in the Japanese study revealed that this genotype composition may pose an especially remarkable risk for squamous cell carcinoma (OR 41.0, 95% CI 8.7-193.6) at low-dose cigarette smoking (33). These findings are consistent with the notion that some procarcinogens in cigarette smoke are activated by CYPlAI and inactivated by GSTM1.
The major importance of the high AHH inducibility in combination with homozygous GSTMI null genotype in environmentally induced lung cancer is further underlined by the observation that the expressing GSTM1 gene appears to have a protective effect against bronchial cancer among individuals with inducible CYPlAI enzyme (34). In contrast, no multiplicative effect was found for CYPIAI and GSTM1 genotypes in a study which found a somewhat higher breast cancer risk (OR 1.6, 95% CI 1.2-23.4) for postmenopausal, light smokers with the CYPIAI Val allele compared to those without the allele (35). Interestingly, the presence of the GSTMI gene was recently correlated with induction of only low levels of CYPlAl mRNA (36).
The protecting role of GSTMI gene in environmental exposures was supported by a study among Finnish asbestos workers, where the GSTMJ null genotype was associated with a 2-fold risk (OR 1.8, 95% CI 1.0-3.5) of asbestos-associated malignant mesothelioma (37). Rather surprisingly, the NAT2 slow acetylator genotype appeared as a similarly important modifier of the mesothelioma risk (OR 2.1, 95% CI 1.1-4.1). Moreover, a striking interaction was observed between these genotypes; highly asbestos-exposed workers without the GSTMI gene and with the NA T2 slow acetylator genotype were 7.4fold (95% CI 1.6-34.0) more prone to develop the neoplasm than fast acetylators with GSTMI gene. N-acetylation may be a comparably important detoxification step in environmental exposures, as is glutathione conjugation. Thus, the combination of NA Ti and NA T2 susceptible genotypes may be a particularly unfavorable genotype composition in arylamine exposures. The recently observed association between increased risk (OR 1.9, 95% CI 1.2-3.6) of colorectal cancer and the fast NA Ti acetylator allele (NA TI *10) was most apparent (OR 2.8, 95% CI 1.4-5.7) among the fast NAT2 acetylators (30).
Given the substrate specificities of the GSTs, individuals with null genotypes at both GSTMI and GSTTI may be at particular risk of environmentally induced cancers. To date, only a few studies have addressed this issue comprehensively. Warwick and co-workers (38) studied the role of combinations of CYP2D6, GSTMI, and GSTTI genotypes in susceptibility to cervical intraepithelial neoplasia and squamous cell carcinoma of the lung. In this study, no combined effect was detected for the two GST genes. However, the CYP2D6 extensive metabolizer (EM) genotype and the GSTTI null genotype were shown to be important risk factors for these pulmonary disorders, both individually and in combination. The GSTTI null and CYP2D6 PM genotypes were shown recently to also increase the risk for two common pituitary brain tumors, astrocytoma and meningioma, but no interactive effects between the genotypes were identified (15). Moreover, in another study, no association between the CYP2D6 or GSTMJ genotypes and susceptibility to the pituitary tumors was detected (39).
Since the recently observed association between the GSTTI null genotype and total ulcerative colitis was also shown to be uninfluenced by the GSTMI genotype (40), the yet published data does not indicate concurrent deficiency of the GSTMI and GSTTI genes playing an important role in individual disease proneness. However, our preliminary findings that this genotype composition poses about a 2.5-fold risk of lung cancer compared to the presence of both of the GST genes (Saarikoski et al., unpublished data) warrant more thorough evaluation of this issue. Finally, the combined effect of CYP2EI and GSTMI genotypes was examined in a study suggesting a 2.5-fold risk of hepatocellular carcinoma for individuals with wild-type CYP2E1 gene compared to those with at least one variant allele in the transcription regulatory area of the gene (41). Again, no modulating effect was found for the GSTMI genotype.

Exposure Markers and Combined Susceptibility Genotypes
Efforts to relate metabolic phenotype or genotype to risk of environmentally induced diseases are now extending to studies on various markers of exposure such as DNA adduct formation and indicators of cytogenetic damage, e.g., sister chromatid exchanges and micronuclei.
Genotoxic agents can form DNA adducts and cytogenetic changes via a complex metabolic pathway that includes CYPlAl; intermediates can be detoxified by conjugation through pathways including GSTs and NATs. A recent study failed to support the importance of this metabolic pathway by reporting lack of any significant association between DNA adduct levels and CYPIAI and GSTMI genotypes among nonsmoking fire fighters (42). No relation was found between these genes and the Band T-micronuclei formation in PAHexposed chimney sweeps (43,44); the only statistically significant deviation was a 60% increase in aromatic DNA-adduct levels in GSTMI null chimney sweeps without any CYPIAI m2 alleles compared to controls with the same genotype composition (44).
However, the GSTMI null genotype has been associated with significantly higher aromatic DNA adduct levels in bus maintenance workers with the NA T2 slow acetylator genotype compared to those with the GSTMI gene (45). In a similar study, nonsmoking bus drivers with NA T2 slow acetylator genotype and GSTMI null genotype had the highest levels of both DNA adducts and cytogenetic damage (46). Moreover, aminobiphenyl-hemoglobin adduct levels are most elevated in smokers possessing this combination of genotypes compared to smokers with other combinations (47). Carcinogenic DNA adduct levels in the mucosa of the urinary bladder were highest in arylamine-exposed individuals who had inherited both the slow NA T2 acetylator genotype and the rapid NAT1 acetylationassociated allele NAT] *10 (48), further addressing the potential importance of individual acetylation capacity.

Conclusion
Knowledge of the genetic basis for variations in human metabolic capacity has opened new possibilities for studies focusing on the role of host factors in susceptibility to environmentally induced diseases. Rapid advances in methodology that determines potential metabolic at-risk genotypes, in combination with exposure markers such as DNA adduct levels in target (surrogate) tissues, may soon allow us to identify susceptible individuals and subgroups in environmentally exposed populations. However, the establishment of combined impact .of all relevant genes for a given exposure is anticipated to be a prerequisite for this.