Sci. Aging Knowl. Environ., 20 July 2005
Vol. 2005, Issue 29, p. pe23
[DOI: 10.1126/sageke.2005.29.pe23]

PERSPECTIVES

Allelic Variation and Human Longevity

Almut Nebel, and Stefan Schreiber

The authors are at the Institute for Clinical Molecular Biology, University Hospital of the Christian-Albrechts-University, 24105 Kiel, Germany, E-mail: s.schreiber{at}mucosa.de (S.S.)

http://sageke.sciencemag.org/cgi/content/full/2005/29/pe23

Key Words: human longevity • healthy aging • allelic variation • linkage analysis • case-control association studies

Introduction

Longevity in humans is a complex trait influenced by both genetic and environmental factors (see "Aging Research Grows Up"). Twin studies have shown that allelic variation (DNA sequence variation in a particular gene) accounts for approximately 25% of the observed differences in life expectancy (1, 2). Similarly, family studies of centenarians from the United States and Iceland indicate that longevity has a strong genetic component (3, 4) (see Miller Viewpoint, "Mission (Not) Impossible," and Wollscheid-Lengeling Perspective). However, as has been observed for other complex conditions, genetic control of aging and longevity is also likely to be determined by subtle variations in many genes that participate in multiple functional pathways, resulting in weak to moderate effects.

Approaches to Identifying Longevity Genes

For the identification of genes predisposing to healthy aging and longevity in humans, two types of approaches are commonly applied: linkage analysis using affected sibling pairs (ASP) and population-based association studies. The advantage of ASP analysis is that it yields a comprehensive and systematic genome-wide overview of susceptibility regions involved in the longevity phenotype without relying on a priori hypotheses about the contribution of specific genes. In addition, linkage analysis is not subject to confounding due to population stratification (differences in allele frequencies between subpopulations due to differences in ethnicity, geography, etc.). On the other hand, sibpair studies usually only detect loci with relatively strong effects (5). To achieve the statistical power needed for the identification of the weak or moderate susceptibility factors (6) likely to play a role in longevity, DNA samples from thousands of long-lived sibpairs are required. Such a monumental task can only be undertaken by large collaborations such as the Genetics of Healthy Aging (GEHA) consortium, which is currently recruiting 2800 long-lived sibpairs (both aged 90+ years) from across Europe (7). Ideally, the children of the ASPs should also be included in linkage studies, because they provide additional valuable haplotype information. However, this would substantially increase the study recruitment effort. Another drawback of linkage analysis is that the identified chromosomal candidate regions are usually large and have to be narrowed by subsequent association and linkage disequilibrium experiments to enable localization of specific candidate genes.

Recent advances in ultra-high-throughput genotyping technology and statistical analysis have made population-based association studies of long-lived individuals (LLI) and controls an efficient alternative (see Kaeberlein Perspective). Future investigations are likely to involve large-scale case-control studies in which entire genomes or extensive selections of candidate genes representing complete functional pathways will be assessed for all possible sequence variation and for association with the longevity phenotype. Because case-control studies require that test cases and controls be sampled from the same population, the availability of large and appropriate study samples remains a limiting factor in conducting these studies.

Study Design

Long-lived individuals as a model system

Centenarians and nonagenarians with good cognitive and physical function who have not suffered from major age-related diseases have been proposed as a model system to study the genetics of longevity and healthy aging in humans (8) (see "Chasing 100"). Particularly valuable information about the heritability of the longevity trait is thought to be contributed by those LLI who represent the top percentiles of their respective birth cohort-specific age distributions (e.g., the 95th percentile and beyond) (9, 10). Exceptional longevity appears to have an especially strong genetic basis (4), which explains why centenarians and near-centenarians tend to cluster in families (see "Century Mark", "Centenarian Advantage", and "Hints of a 'Master Gene' for Extreme Old Age"). Given the frequency of LLI in developed countries [e.g., elderly people aged 80 and older made up 4% of the German population in 2003 (11)], collecting DNA samples from the hundreds or thousands of unrelated individuals required for association mapping with sufficient statistical power now seems feasible (10) (see "Rising Expectations").

Choosing appropriate controls

An inherent dilemma in genetic longevity studies is the lack of appropriate controls. The case-control design demands that case and control individuals only differ with regard to the studied phenotype and that all other important variables, such as gender, ethnicity, environmental factors and usually age, be equal or as similar as possible between the two groups to avoid false-positive or -negative findings. Thus, the perfect strategy in longevity research would be a prospective cohort study spanning the next 100 years or more, in which allelic variation in the surviving LLI would be compared with that of their peers who died earlier. An alternative could be to perform a comparative analysis of DNA from contemporary LLI and genetic material extracted from the remains of their peers who died decades ago. For various reasons, both of these recruitment strategies are extremely impractical, if not impossible. A realistic and widely applied, but only second-best, approach is one in which allele frequency data of LLI are compared with those of younger living controls. Although the genetic analysis of populations across time periods is presently the only feasible study design for the identification of longevity genes in humans, it is hampered by major pitfalls. One question that arises in this context is how many of the controls will become long-lived themselves? Since improvements in the standard of living even at an advanced age can lead to a substantial increase in life expectancy (12), the proportion of future LLI is difficult to predict. Another methodological problem is that the cases and controls used in these experiments were usually born several generations apart. Instead of an effect on longevity, frequency differences between LLI and younger controls could also reflect a change in population structure over time (i.e., gene flow marking recent migrations and admixture) or gene-environment interactions impacting differently on the gene pools across generations.

Problems of age matching

A younger control sample may not provide valid estimates of initial allele frequencies in the corresponding LLI population (13). Migration and population stratification are known confounding factors. Movements of people and subsequent admixture (mixing of two or more genetically distinct populations) can substantially influence relative allele frequencies, especially when the source and recipient populations differ to a considerable extent. Large changes in allele frequency, which in the short term are a function of the difference in frequency and migration rate, can be seen in just one generation (14). This is a major concern, particularly when typical "immigrant" populations are studied, for example, in the United States. Caucasians in the United States represent a considerably admixed population to which immigrants from a number of different European countries have contributed over the past 400 years. Because their origin has changed with time, the ethnic and therefore genetic composition of a U.S. LLI sample is likely to differ from that of a younger control population (10, 15). Inappropriate matching of cases and controls under even modest levels of population structure can cause both false-positive and false-negative findings (16). Concerns about population stratification can be addressed by sampling control individuals who are as close as possible to the LLI in terms of their age (born only one or two generations apart), ethnicity, and geographic origin. In addition, genome analysis using methods such as genomic control and structured association may compensate for some of the genetic differences between cases and controls that can arise from the lack of appropriate matching (16). Genomic control is a procedure in which an apparent association is adjusted for population stratification in the study sample, whereas association tests investigate population association between a phenotype and a particular allele.

Effects of gene-environment interactions

The gene pool of a population changes in response to environmental factors. So how realistic is it to assume that the genetic makeup of elderly people alive today is identical to that of past or future LLI? The risk of mortality conferred by genotypes may depend on interactions with environmental or behavioral risk factors that in turn vary by time and place (13). It is worth remembering that the LLI recruited today were born at the end of the 19th or the beginning of the 20th century. They have survived periods of deprivation and/or starvation (e.g., the Great Depression in the United States and two World Wars in Europe), as well as various infectious diseases in the pre-antibiotic and prevaccination era. The control individuals, on the other hand, were mostly born and brought up after World War II and experienced novel medical, social, and technological developments. Allelic variation associated with longevity is therefore context-dependent; a genetic variant that has allowed a centenarian to live to the present day may not necessarily increase the life expectancy of a younger generation.

Furthermore, in large populations, considerable changes in allele frequencies can also be caused within a few generations by small differences in selective pressure on genes that provide a survival advantage or disadvantage before and during the reproductive phase (17). This is a particularly important consideration in case-control longevity studies that investigate candidate genes shown to be targets of natural selection (e.g., those involved in the immune response).

Conclusion

In summary, case-control association studies hold much promise as a tool to investigate the genetic basis of longevity but also demand considerations about study design. Sufficient replication and validation of the initial association findings in very large samples from different populations should be mandatory. Moreover, the putative role of the identified candidate genes as modifiers of human life span has to be evaluated and rigorously tested through a series of functional analyses before they can be regarded as genuine longevity genes.


July 20, 2005
  1. M. McGue, J. W. Vaupel, N. V. Holm, B. Harvald, Longevity is moderately heritable in a sample of Danish twins born 1870-1880. J. Gerontol. 48, B237-B244 (1993).[Abstract]
  2. A. M. Herskind, M. McGue, N. V. Holm, T. I. Sorensen, B. Harvald, J. W. Vaupel, The heritability of human longevity: A population-based study of 2872 Danish twin pairs born 1870-1900. Hum. Genet. 97, 319-323 (1996).[CrossRef][Medline]
  3. H. Gudmundsson, D. F. Gudbjartsson, M. Frigge, J. R. Gulcher, K. Stefansson, Inheritance of human longevity in Iceland. Eur. J. Hum. Genet. 8, 743-749 (2000).[CrossRef][Medline]
  4. T. T. Perls, J. Wilmoth, R. Levenson, M. Drinkwater, M. Cohen, H. Bogan, E. Joyce, S. Brewster, L. Kunkel, A. Puca, Life-long sustained mortality advantage of siblings of centenarians. Proc. Natl. Acad. Sci. U.S.A. 99, 8442-8447 (2002).[Abstract/Free Full Text]
  5. N. Risch, K. Merikangas, The future of genetic studies of complex human diseases. Science 273, 1516-1517 (1996).[Abstract/Free Full Text]
  6. D. Botstein, N. Risch, Discovering genotypes underlying human phenotypes: Past successes for mendelian disease, future approaches for complex disease. Nat. Genet. 33, 228-237 (2003).
  7. A. Nebel, S. Schreiber, GEHA--the pan-European "Genetics of Healthy Aging" project. Sci. Aging Knowledge Environ. 2004 (21), pe23 (2004). [Abstract/Free Full Text]
  8. C. Franceschi, M. Bonaf�, Centenarians as a model for healthy aging. Biochem. Soc. Trans. 31, 457-461 (2003).[CrossRef][Medline]
  9. A. A. Puca, M. J. Daly, S. J. Brewster, T. C. Matise, J. Barrett, M. Shea-Drinkwater, S. Kang, E. Joyce, J. Nicoli, E. Benson, L. M. Kunkel, T. T. Perls, A genome-wide scan for linkage to human exceptional longevity identifies a locus on chromosome 4. Proc. Natl. Acad. Sci. U.S.A. 98, 10505-10508 (2001).[Abstract/Free Full Text]
  10. A. Nebel, P. J. P. Croucher, R. Stiegeler, S. Nikolaus, M. Krawczak, S. Schreiber, No association between microsomal triglyceride transfer protein (MTP) haplotype and longevity in humans. Proc. Natl. Acad. Sci. U.S.A. 102, 7906-7909 (2005).[Abstract/Free Full Text]
  11. Statistisches Bundesamt, Bev�lkerung Deutschlands bis 2050: 10. koordinierte Bev�lkerungsvorausberechnung (Statistisches Bundesamt, Wiesbaden, 2003).
  12. J. W. Vaupel, J. R. Carey, K. Christensen, Aging: It's never too late. Science 301, 1679-1681 (2003).[Abstract/Free Full Text]
  13. S. J. Lewis, E. J. Brunner, Methodological problems in genetic association studies of longevity--the apolipoprotein E gene as an example. Int. J. Epidemiol. 33, 962-970 (2004).[Abstract/Free Full Text]
  14. J. Wang, M. C. Whitlock, Estimating effective population size and migration rates from genetic samples over space and time. Genetics 163, 429-446 (2003).[Abstract/Free Full Text]
  15. B. J. Geesaman, E. Benson, S. J. Brewster, L. M. Kunkel, H. Blanch�, G. Thomas, T. T. Perls, M. J. Daly, A. Puca, Haplotype-based identification of a microsomal transfer protein marker associated with the human lifespan. Proc. Natl. Acad. Sci. U.S.A. 100, 14115-14120 (2003).[Abstract/Free Full Text]
  16. J. Marchini, L. R. Cardon, M. S. Phillips, P. Donnelly, The effects of human population structure on large genetic association studies. Nat. Genet. 36, 512-517 (2004).[CrossRef][Medline]
  17. J. H. Gillespie, Population Genetics: A Concise Guide (Johns Hopkins Univ. Press, Baltimore, 1998).
Citation: A. Nebel, S. Schreiber, Allelic Variation and Human Longevity. Sci. Aging Knowl. Environ. 2005 (29), pe23 (2005).




THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES:
Association of FOXO3A variation with human longevity confirmed in German centenarians.
F. Flachsbart, A. Caliebe, R. Kleindorp, H. Blanche, H. von Eller-Eberstein, S. Nikolaus, S. Schreiber, and A. Nebel (2009)
PNAS 106, 2700-2705
   Abstract »    Full Text »    PDF »




Science of Aging Knowledge Environment. ISSN 1539-6150