| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||
Department of Genes and Environment, Division of Epidemiology, Norwegian Institute of Public Health, Oslo, Norway; and Behavioral and Social Research Program, National Institute on Aging, National Institutes of Health, Bethesda, Maryland.
Address correspondence to Jennifer R. Harris, Department of Genes and Environment, Division of Epidemiology, Norwegian Institute of Public Health, PO Box 4404, Nydalen, N-0403 Oslo, Norway. E-mail: jennifer.harris{at}fhi.no
"What is inherited is DNA, everything else is developed." (Tanner, 1978, p. 117)
IN HIS book on development, James M. Tanner (1978) eloquently summarizes the well-known principle that gene expression requires environments. Such environmental influences may range from intracellular conditions to larger sociocultural effects that co-act and interact with genes in developmentally dynamic ways throughout life. Although recent methodologic advances have turned intense focus toward gene finding, the environmental theme is not new for geneticists, and renewed interest in gene–environment interaction is certainly upon us. The mounting literature demonstrating that environmental influences affect gene expression (McClearn, 2003) underscores that understanding such interactions will solve critical pieces of the puzzle behind complex traits. We are already witnessing this in pharmacogenetics (Crow & Johnson), and parallels extend to behavioral medicine where genetic differences may affect both the ease with which individuals can change their behavior and the outcomes produced by those changes. A good example is the differential response to physical training dependent on the angiotensin-converting enzyme insertion/deletion genotype (Montgomery et al., 1999). If we take these findings one step further, we could ask about the implications for compliance behavior based on genotypic differences.
Why has the environmental quest remained notably divorced from genetic studies? Certainly, one reason is that environment is notoriously hard to specify and quantitate, especially in humans. However, other explanations may be that gene detection efforts will not necessarily be boosted, statistically speaking, by incorporating interactive effects (Purcell, 2004). Furthermore, the extent to which seemingly environmental effects will be explained by epigenetic regulation or stochastic factors remains unknown. The discordant identical twin design offers a unique approach to investigate these issues to determine whether within-pair phenotypic differences reflect molecular events that lead to differences in the expressed genotypes. But the environment remains elusive for other reasons. Despite widespread evidence from quantitative genetic studies revealing the importance of environmental effects for nearly every phenotype studied, little is known regarding which specific features of the environment to measure and when in developmental time assessments should be conducted, etc. With the goal of understanding why we age so differently, an obvious question asks whether the large literature on key environmental measures known to affect aging can be brought to bear upon genetic inquiry. This issue was explored at a workshop entitled "The NIA Environmental Workshop for Genetically Informative Studies on Aging," which was held in February 2003 by the Behavioral and Social Research (BSR) Program at the National Institute on Aging. The workshop was part of an ongoing initiative at BSR to integrate genetics and genomics with behavioral and social science research on aging. The workshop congregated researchers with expertise in an array of fields and with diverse perspectives to explore the prevailing questions.
Several critical themes emerged as represented by articles appearing in this issue, including the importance of studying social environments, the need for methodologically rigorous approaches to investigate gene-by-environment (also known as G by E or G x E) interactions and covariation, and the importance of starting with known genetic influences, intermediary phenotypes, or biomarkers. Progress will rely on interdisciplinary work and using animal and human research to mutually inform the next steps. We hope that this special issue will serve as a catalyst for promoting interdisciplinary studies integrating genetics more explicitly with the behavioral and social sciences. This collection of papers is a start.
Social environment is central to the work by Ryff and Singer. They describe how features of the social environment are linked to morbidity, mortality, and biomarkers of aging and call on results from human and animal studies to demonstrate that social environment may modify genetic risk. The report by Deater-Deckard and Mayr draws upon research conducted to explore childhood and adolescent development. They emphasize studying gene–environment correlations and nonshared environmental mechanisms using a "bottom-up" approach that investigates candidate environmental factors while estimating genetic effects (which may be anonymous). Statistical methods for studying gene–environment interactions and correlations are continually progressing (Purcell, 2002) and being applied to questions of development and aging (Eaves, 2003). The aging environment is rich in transitions, and aging-relevant questions raise new issues that will require novel modeling approaches. For example, do gene–environment correlations show age effects contingent upon reductions in one's ability to select environments due to becoming a chronic caregiver or experiencing reduced physical functioning?
The synthesis by McClintock and colleagues outlines a strategy for identifying environments at higher levels of organization that may regulate specific genes and emphasizes the necessity of starting with traits for which some of the genes have been identified. They build upon results from rat models that demonstrate relationships between social isolation, hypervigilance, mammary tumors, and accelerated aging. The authors propose studying similar environmental regulation of gene expression that may explain the disproportionately high mortality among African Americans associated with premenopausal breast cancer. Johnson and Krueger build upon findings demonstrating that income and perceived control moderate genetic variance in physical health. They use twin data to illustrate how multivariate analyses of genetic and environmental influences may help elucidate causal processes involved in social class differences in health. Of particular importance is to test expectations regarding environmental buffering of genetic effects that are generated in the data. The authors highlight that further work is needed to identify circumstances under which genetic effects are environmentally malleable and discuss implications for gerontologic research. The article by Grigorenko helps frame gene–environment interactions with regard to what is known, what can be done, and implications for prevention. She addresses complexities of gene–environment interactions to help evaluate their value in statistical models. Information regarding interpretation of gene–environment interactions and the types of designs and methodologies that can be applied to investigate these interactions is systematically brought together. Examples are presented, and statistical power issues are discussed. Shanahan and Hofer focus on ways in which social context can alter gene expression. They develop a typology describing four processes by which social context could moderate gene expression. These include triggering, compensation, social control, and enhancement. Their work suggests new ways of studying genes, context, and behavioral development.
As articulated in the articles herein, gene–environment dynamics are complex, exceptionally difficult to study, and important for how we age. The task, indeed, seems daunting. Yet it was not that long ago when the thought of mapping human genetic variation down to single-nucleotide differences seemed unthinkable. Now, insights gained and advances in high-throughput methods coupled with large-scale epidemiologic studies are igniting renewed optimism for detecting genes influencing complex traits and understanding how these effects vary with age. We do not know if the "environmental" mother load will be as fruitful, but given inroads already made, forward-thinking initiatives are needed to tackle this challenge. Close integration of social sciences with genetics and genomics research will open new avenues for meeting this challenge. For example, the National Institute on Aging–sponsored National Research Council volume on adding biological indicators to social science research (Finch, Vaupel, & Kinsella, 2001) evolved from work undertaken to direct new research on the trajectories of human longevity, which recognized that advances require analysis of gene–environment–behavior interactions. Taking stock in the social science surveys that already exist with the goal of extending these data to include genetic and biological indicators can provide unique resources for exploring gene–environment dynamics. The value of large, nonascertained population samples is known for resolving questions of genetic heterogeneity, identifying predisposing gene variants, and verifying their significance in specific populations. More recently, such study designs have also become recognized for their promise in helping to elucidate gene-by-environment interactions including the role of lifestyles and behaviors and how these effects operate throughout life (e.g., National Institutes of Health Notice NOT-OD-04-041, Request for Information: Design and Implementation of a Large-Scale Prospective Cohort Study of Genetic and Environmental Influences on Common Diseases, May 2004).
The works presented in this special issue highlight why Tanner placed understanding genetic effects squarely within the context of development. Methodologic advances now poise us to explore these issues more closely. Along these lines, we hope this special issue will nurture and promote scientific interest in genetics among social science researchers in gerontology and recruit a new cast of investigators to study environments in the genetics of aging.
References
| ||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
