Gene-Environment Interaction

In the field of epidemiology, there has been increasing interest in documenting the results of interaction between specific genes or genotypes and well-understood environmental exposures. Gene-environment interactions are situations in which the combination of an environmental factor, such as exposure to cigarette smoking, with a genetic factor, such as a diseasepredisposing mutation, results in a greater risk or severity of the disease in question than either the genetic or environmental factor acting alone. These interactions have been studied in plants and animals for many years. While they have been hypothesized also to occur in humans, they have been documented in humans only fairly recently. In recent years, many examples of gene-environment interaction have been reported in the literature.

There is now increasing acceptance of the idea that it is likely that all human disease is the result of the interaction between the genetic susceptibility to disease and environmental exposures during the course of life. It is important to study gene-environment interactions for several reasons. One is the expectation that it will improve our understanding of the etiology of specific diseases. Second, we might be able to identify populations that are at high risk due to their possessing higher frequencies of genotypes that denote susceptibility to the disease in question. We might also be able to identify modifiable risk factors in the form of environmental exposures. Finally, there is the underlying goal of being able to prevent disease through improved understanding of the risk factors that are involved and the knowledge of the underlying disease mechanisms.

Before we discuss what an interaction is—particularly an interaction between a gene and some environmental exposure—we need some background thinking about what an effect of a genetic factor is and how we should conceptualize the environment. Genetic variation in human populations fits the label given to it by some as the ‘ultimate public health problem.’ With the ever mounting discoveries of disease-related genetic variation and the identification of specific genes related to disease, the statement that everyone is at genetic risk for some disease has become an acceptable assessment. In genetic epidemiology, the search for disease-associated genes has been continuing for many years, first through family studies and later through specific analyses involving estimation of heritability and segregation and linkage analysis. More recently, the focus has been on largescale association studies in populations.

In research on the genetics of human diseases, one might ask what are the measurements for the genotype or, for that matter, for the genes that might [Page 407]increase one's risk of a disease. Family history, that is, the presence or absence of a family history for a disease in question, is one measure of genotype. However, it is crude at best and susceptible to the misclassification of significant proportions of individuals in an analysis. If a phenotype has been adequately described and the specific features of the disease are well known and easily observed and measured, then the specific phenotype for the presence of the disease may be indicative of an underlying gene or complex of genes for that disease. But phenotypic descriptions of diseases are subject to analytical problems, especially where those diseases are heterogeneous in their expression or represent a spectrum of disorders, such as autism. In such cases, the so-called behavioral phenotype becomes extremely important in studies of the genetic factors in the disease entity in question.

...