Commentary: From phenotype, to genotype, to gene–environment interaction and risk for complex diseases
E-mail: olden{at}niehs.nih.gov
‘No one supposes that all the individuals of the same species are cast in the same actual mould. These individual differences are of the highest importance to us, for they are often inherited.’ (The Origin of Species, Charles Darwin, 1859)
The 1979 publication of the article by Lower et al.1 on ‘N-acetyltransferase phenotype and risk in urinary bladder cancer: approaches in molecular epidemiology’ generated considerable interest and enthusiasm for research to understand gene–gene and gene–environment interactions in human health and disease. This seminal publication in the Environmental Health Perspectives transformed population health research and provided the foundation for the massive sequencing efforts to identify genetic variations involved in modulating human response to drugs and other environmental xenobiotics.2–4 The study of Lower et al.1 was prompted by their interest in understanding the relationship between N-acetyltransferase phenotype and susceptibility to the development of bladder cancer from human exposure to arylamines, as a result of cigarette smoking or working in the chemical dye industry. Previous studies had suggested that (i) a significant portion of bladder cancer could be attributed to such exposures, (ii) the distribution of N-acetyltransferase activity in the liver was highly variable among individuals and (iii) that individuals with low enzyme activity (the so-called ‘slow acetylator phenotype’) were most susceptible to the development of bladder cancer.
Based on these observations, Lower et al.1 hypothesized that the slow acetylator phenotype would be over-represented in a population of bladder cancer patients occupationally exposed to arylamines. Their finding of an excess of individuals of the slow acetylator phenotype, within such a population from Denmark, confirmed both the earlier suggestion that arylamines play a role in bladder carcinogenesis and their hypothesis that slow acetylators are at increased risk. Also, this landmark study demonstrated the power of hypothesis-driven, population-based studies for assessing disease risk associated with specific interactions between genes and the environment. Because of this pioneering publication, epidemiological research is now being pursued to validate the biological significance of human genetic variants by correlating polymorphisms, affecting various metabolic pathways to disease risk. In some cases, population-based studies can provide mechanistic insight before functional genomics is informative. The extent of genetic polymorphism in the human genome is becoming increasingly clear with the advent of molecular cloning and gene sequencing, and this clarity has enhanced understanding of their involvement in disease susceptibility.
Over the past 27 years, numerous genetically determined phenotypes have been convincingly associated with change in susceptibility to various diseases. However, the fact that most are only weakly associated with risk suggests that multiple, rather than single, phenotypes contribute to increased or decreased risk. In fact, most associations are not strong enough to be by themselves diagnostic or predictive, and interactions between genes and metabolic pathways may obscure functional relationships, making association studies difficult to replicate or validate.
Interindividual pharmacokinetic variation in rates of drug elimination can vary dramatically. Multinational twin studies were conducted on the kinetics of several drugs to compare relative contribution of genetics and environmental factors with respect to interindividual variations.5 Results were remarkably similar for the various populations, and pharmacokinetic variation virtually disappeared within monozygotic twins, but was preserved within most dizygotic twins; meaning that variation in drug metabolism is primarily under genetic control. Altering the environment of the monozygotic twin had no effect on kinetics. If reported pharmacogenetic differences of 10- to 200-fold can be extrapolated directly to risk of human diseases, one could conclude that an individual can be 10–200-fold more sensitive to a given drug or environmental chemical due to differences in expression and activity of metabolic enzymes.
While Lower et al.1 did not coin the phrase ‘molecular epidemiology’, the application of such a molecular level approach to examine causality and relative risk in the human population had a significant impact on the development of environmental genomics. Furthermore, such studies resulted in increased awareness of the role of genetics as a factor that can drastically alter susceptibility to most chronic diseases, especially those triggered by environmental exposures.
Using many of the tools developed over the past 50 years (e.g. polymerase chain reaction techniques, high-throughput DNA -sequencing and oligonucleotide arrays), environmental health researchers are poised to make remarkable advances in identifying functional polymorphisms that increase or decrease risk from exposure to environmental xenobiotics. To promote research in this field, the National Institute of Environmental Health Sciences of the National Institutes of Health initiated the Environmental Genome Project in 1997.2,3,6,7 It is a comprehensive effort to re-sequence 544 candidate genes to identify polymorphisms that influence susceptibility to environmental exposures. In addition to polymorphisms discovery and characterization, the initiative supports population-based epidemiological and clinical studies, technology development and efforts to understand the social, legal and ethical implications of such research. The epidemiological and clinical-based population studies builds on the work of Lower et al.,1 using more accurate measures of exposure and gene sequencing technologies to correlate single nucleotide polymorphisms (SNPs) or haplotype with disease risk. Because the candidate genes have been characterized with respect to function, elucidating their role in disease development has a high probability of immediate success.
It has become clear that addressing the role of gene–environment interactions in the aetiology of complex diseases will require the development of a robust framework to account for ‘the environment’. In spite of current understanding of the multifactoral nature of the aetiological mechanisms of chronic diseases, biomedical researchers still tend to focus on limited or circumscribed components of disease. While the various disciplinary approaches have led to new insights into disease causation, they are unlikely to be able to provide a coherent explanation for disease aetiology. Progress will require large-scale, multi-institutional collaboration and interdisciplinary expertise to design, collect and analyse the appropriate experiments.
Epidemiologists have made significant contributions to our understanding of gene–environment interactions.8,9 However, the Achilles’ heel in the conduct of such studies is the lack of accurate measures of exposure. The most common approaches of estimating exposure using indirect surrogates such as questionnaires, toxic release and production inventories, and environmental monitoring do not take into account individual differences in uptake, excretion and metabolism of environmental xenobiotics and dose–response relationships. Not only is it difficult to measure exposure accurately, but also the available metrics are too costly for use in large-scale cohort studies required to detect significant associations between genes, environment and specific health outcomes.
In addition to impediments imposed by sample size and exposure assessment tools, population-based studies are fraught with other challenges related to the need for better statistical models. To define complex exposure–disease relations on a high variable genetic background, a comprehensive portrait of biological response to a specific exposure is needed. The achievement of this goal will require improved modelling capabilities to explain how multiple interactive components of the disease process respond when perturbed by exposure to environmental factors. I am hopeful that the newly emerging field of toxicogenomics10 will provide the critical databases on gene, protein and metabolite expression profiles necessary to sort out the respective roles of genes, physical environment and behaviour in the aetiology of chronic diseases that plague mankind.
In conclusion, the relationship between the environment and human illness has been well-established. The strength of these associations indicates that a large portion of the variation in disease incidence is due to genetic variation and differences in environmental exposures. However, despite several decades of experimental and epidemiological studies, the specific interactions that contribute to disease pathogenesis have been difficult to decipher. Current efforts to collect environmental data and analyse polymorphisms in genes that control xenobiotic metabolism and cell growth and repair mechanisms will be useful in achieving this objective.
 |
Notes |
|---|
Laboratory of Molecular Carcinogenesis, National Institute of Environmental Health Sciences, National Institutes of Health, DHHS, RTP, NC 27709, USA.
|
References |
|---|
 Top References |
|---|
1 Lower GM, Nilsson T, Nelson CE, Wolf H, Gamsky TE, Bryan GT. N-acetyltransferase phenotype and risk in urinary bladder cancer: approaches in molecular epidemiology. Preliminary results in Sweden and Denmark. Environl Health Perspect (1979) 29:71–79. (Reprinted Int J Epidemiol 2007;36:11–18.)[CrossRef]
2 Kaiser J. Environmental institute lays plan for gene hunt. Science (1997) 298:569–70.
3 Brown PO, Hartwell L. Genomics and human diseases-variations on variation. Nat Genet (1998) 18:91093.
4 Olden K, Wilson SH. Environmental health and genomics: visions and implications. Nat Rev Genet (2000) 1:149–53.[ISI][Medline]
5 Vessel ES. Genetic and environmental factors causing variation in drug response. Mutat Res (1991) 247:241–57.[ISI][Medline]
6 Wilson SH, Olden K. The environmental genome project, phase I and beyond. Molecular Prevention (2004) 4:147–56.
7 Olden K. Use of omic approaches in unraveling mechanisms of gene–environment interactions. Curr Genomics (2004) 5:1–6.[CrossRef]
8 Hunter D. Gene–environment interactions in human diseases. Nat Rev Genet (2005) 6:287–98.[ISI][Medline]
9 Olden K. Gene–gene and gene–environment interactions. In: Handbook on Genomic Medicine—Ginsburg G, Willard H, eds. New York: Elsevier. (In press).
10 Kaiser J. Tying genetics to the risk of environmental diseasse. Science (2003) 300:563.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||