CELLULAR AGING: DNA POLYMORPHISMS

Many, if not the majority, of genetic loci in individuals in outbred, or wild, populations (including human) can have alternative versions of the gene, called alleles, that may or may not specify different genetic information. The term genetic polymorphism is used to describe a Mendelian trait that is present in at least two phenotypes (the observable physical characteristics of an organism) that are specified by different alleles and are present at a frequency of greater than 1 to 2 percent in the population. In contrast to a polymorphism, a rare genetic variant is one that is present in the population at a frequency of less than 1 percent, and most commonly at very low frequencies. Most, but not all, of the mutations that cause genetic diseases are in this category. For example, the alleles associated with Werner syndrome, a genetic disorder displaying a number of features of accelerated aging, are rare variants (see below).

The ABO blood group was the first human genetic polymorphism to be described—a discovery of immense theoretical and practical importance. Demonstrating that multiple alleles can occur at a specific genetic locus, often at comparable frequencies in the population, ultimately led to an understanding of the genetic basis of phenotypic variation in animal populations. The practical significance of this pioneering discovery is that it led to the routine use of whole-blood transfusions in the practice of medicine. Since this seminal observation, many polymorphic traits have been described and, although the precise number is not known at this time, it is now believed that the majority of the genetic loci in human genome are polymorphic. The number of alleles at any polymorphic locus is extremely variable. At some loci, such as the human leukocyte antigen (HLA) genes in the major histocompatibility complex (MHC), literally dozens of alleles have been identified. Thus, with this degree of variation throughout the human genome, it is highly likely that every individual on earth, with the exception of identical twins, possesses a unique genotype.

DNA polymorphisms are defined as any alternative DNA sequence that is present in 1 to 2 percent or more of the population. The extent of genetic variation at the DNA level greatly exceeds that which is present in gene products (i.e., proteins). DNA polymorphisms are more frequent in sequences that are not involved in the regulation or specification of gene products. This part of the genome does not seem to affect the phenotype of the organism, and, therefore, mutations in these sequences are very likely to be selectively neutral and could accumulate more rapidly than in genetically active areas.

DNA polymorphic variants can be deletions, duplications, or inversions of segments of DNA. Two types of DNA sequences that are highly polymorphic are minisatellites that are composed of tandemly repeated 10 to 60 base-pair (bp) sequences and microsatellites that are segments composed of tandem repeats of 1 to 3 bp sequences. These sequences have proven to be very useful for gene mapping because: (1) they are distributed throughout the genome; (2) they are highly polymorphic in that the number of repeats is extremely variable; and (3) individual alleles can be identified by amplification of sequences by the polymerase chain reaction (PCR) and the size difference of these sequences determined by gel electrophoresis.

The most common type of DNA polymorphism, accounting for most genetic variation among human populations, is the single nucleotide change (single nucleotide polymorphism, or SNP). Since the completion of the first drafts of the human genome sequence, the identification and location of SNPs has progressed very rapidly. A working draft of the sequence assembled by the International Human Genome Sequencing Consortium is in a public database on the World Wide Web and can be easily accessed at http://genome.ucsc.edu. As of mid-2001, 1.42 million SNPs throughout the genome had been identified, with estimates that 60,000 SNPs are within regions that are transcribed into RNA. SNPs in these sequences could cause an amino acid change in the protein product specified by the gene, which, in turn, could have biological consequences. Moreover, it is currently estimated that 85 percent of the coding regions (exons) in genes are within 5,000 bp of a SNP; thus, it is almost certain that polymorphisms air in the regulatory elements of some genes and could, therefore, have an affect on the level of activity of these loci. In addition, SNPs are of sufficient density throughout the genome to serve as markers for the identification and mapping of specific combinations of alleles (haplotypes) that are associated with specific phenotypes. To accomplish this, informative SNPs (in or near genetic loci) will have to be identified and their frequencies in various populations determined. This will require a huge number of SNPs to be identified in many thousands of individuals. This endeavor will require the development of very efficient assays to screen such large populations, and a number of laboratories are developing highly efficient screening for such studies.

A database designed to serve as a central repository for SNPs, and for short deletion and insertion polymorphisms, has been established by the National Center for Biotechnology Information (NCBI) in collaboration with the National Genome Research Institute (see its website at www.ncbi.nlm.nih.gov/SNP/index.html). This repository also contains data derived from other species, both mammalian and nonmammalian, and is linked to national databases that contain other biological information. This repository will contain a large amount of information; for example, the update of 1 July 2001 indicated that 2,985,822 SNPs had been submitted (but not all fully characterized) to this database. Other public databases of human SNPs that are accessible on the Web have been established. One is located at the University of Utah in Salt Lake City (www.genome.utah.edu/genesnps), and another has been established in Europe sponsored by a consortium of major institutions (http://hgbase.cgr.ki.se).

DNA polymorphisms and aging

The wide variation in the maximum life span, even among mammalian species, is well established and is related to the genetic endowment of each species. There are also qualitative and quantitative differences in the phenotype of aging among mammalian species. For example, the extensive atherosclerotic involvement of the arterial system associated with extensive morbidity (illness) and mortality with advancing age is virtually unique to humans. On the other hand, the extent to which genetic differences are responsible for the variation in longevity between individuals within a species is unknown. The variation of individual life spans is evident in populations of inbred organisms, indicating that environmental factors and, very likely, chance events contribute to this variability. However, the extent of the genetic contribution to this interindividual variation in the manifestation of aging and maximum survivability remains to be established.

It is generally accepted that aging and life span are regulated by multiple genes. Although the precise number is not known, it has been speculated that relatively few genes may be directly involved in this process. The most direct approach to the identification of aging and longevity genes would be to search for quantitative trait loci (QTL). These loci contain genes that regulate traits, such as blood pressure, that can be defined in specific units of measurement (in the case of blood pressure, the units would be millimeters of mercury) and are, in most cases, regulated by multiple genes. Longevity, which can be measured in units of time (e.g., days or years), is another example of a quantitative trait. A QTL study of aging might, for example, involve strains of laboratory mice that exhibit significantly different maximum life spans. Genetic loci that effect the phenotype (in this case, longevity) and their relative contribution, can be identified by a sophisticated analysis of data derived from the segregation pattern of polymorphic markers in relation to the phenotype (longevity) of the offspring from crosses between the strains, and from back crosses.

This type of genetic analysis is not realistically feasible with human subjects. However, association studies, another experimental approach to the identification of "aging genes," can be carried out in human subjects. Such studies involve the search for linkage between a specific polymorphic allele(s) or DNA polymorphism(s) and a specific trait. For example, one could compare the frequency of polymorphic alleles of a gene in an exceptionally long-lived population (e.g., centenarians) and a well-defined control population. A number of studies have been carried out with the human leukocyte antigen (HLA) loci in the major histocompatibility complex (MHC), one of the most polymorphic class of genes in the mammalian genome. These studies have yielded conflicting results, probably due to a number of methodological problems, including inaccurate identification of specific alleles in the pregenomic era. Similar studies designed to establish linkage between the incidence of a specific age-associated disease and specific alleles of a polymorphic locus have provided new information of considerable interest. For example, the association of the e4 allele of the apolipoprotein E gene, a gene that codes for a protein involved in lipid transport in the vascular system, with an elevated risk of developing Alzheimer's disease is now well established.

Clearly demonstrable associations, such as that between apolipoprotein B and Alzheimer's disease, are infrequent. More subtle associations are difficult to detect because of the relative paucity of genetic markers (e.g., SNPs) that, up to this time, have been identified in the human genome. Moreover, aging and the regulation of life span are multifactorial phenotypic traits, regulated by multiple genes interacting with the environment. Therefore, it is unlikely that polymorphic variants at a single locus will have a profound effect on the aging process or longevity; what is more likely is that combinations of alleles (haplotypes) will be associated with specific aging phenotypes. As indicated above, highly efficient methods to determine the frequency of SNPs in human populations are being developed, which will make the haplotyping of large numbers of individuals within a population feasible. The task may be simplified by the emerging observation that human genetic diversity is surprisingly limited. Theoretically there could be hundreds, even thousands, of variants at each locus, but in reality the number of alleles at most loci appears to be small, only two or three in many cases.

Aging at the cellular level

All DNA polymorphisms that affect the phenotype of the organism are expressed in some fashion at the cellular level, even if only in the secretion of an abnormal gene product that acts at a site distant from the secretory cell. Alleles of genes involved in basic cellular functions could affect the phenotype of multiple cell types, or in some cases all cell types. For example, polymorphisms in genes involved in DNA synthesis or cell division could alter the function of cells that are actively proliferating or are potentially capable of proliferation. On the other hand, a variant gene involved in an essential function such as aerobic respiration could potentially alter the phenotype of every cell in an organism.

The replication and repair of nuclear (genomic) DNA involves a variety of functions that are essential for cell survival. These metabolic processes are essential for the accurate transmission of genetic information from one generation of a cell or organism to the next, and for the maintenance of normal gene function. Diminished fidelity of DNA replication and/or repair will result in an increased mutation rate, which can lead to decrements of cell function, cell death and/or increased risk of transformation to a malignant (cancerous) cell type. There have been a number of experimental observations that suggest (but do not prove) that the accumulation of genomic mutations in somatic cells is a causal mechanism of aging both at the cellular and organismal levels. This hypothesis, generally attributed to Szilard, implies that the efficiency and fidelity of DNA replication and repair affect the rate of aging and maximum life span. Some of the observations that are consistent with this hypothesis are:

1. The efficiency and extent of the repair of DNA damage induced by ultraviolet light (UV) is directly related to the maximum life span of the species. This result correlates with the observation that lower levels of DNA damage and mutations are present in experimental animals that are on a diet that restricts caloric intake. Dietary restriction has been shown to extend maximal life span in multiple species and has been extensively exploited as an experimental model in aging research.
2. Normal human cells in tissue culture can divide only a limited number of times. This is often referred to as replicative senescence and is associated with changes in cell structure and gene activity. It now appears that one mechanism that determines the replicative potential of some cell types in cultures and in tissues is the extent of loss of specialized hexanucleotide repeat sequences of DNA (telomeres) at the ends of the chromosomes. Approximately fifty to one hundred bp are lost from this region with each cell division. This occurs because telomerase, the enzyme that synthesizes these repeat sequences, is functionally inactive in most human somatic cells. It has been postulated that after a sufficient number of replications these structures become so short that the cells perceive them as damaged DNA and irreversibly cease cell division.
3. The Werner syndrome, a rare genetic disease that is associated with decreased longevity and many features (but not all) of premature aging, is caused by a mutation in a helicase gene. This class of genes catalyzes the unwinding of the double helix of DNA, which is necessary for a number of essential functions, such as DNA replication and repair and messenger RNA transcription. Cultured cells derived from individuals with Werner syndrome display numerous abnormalities in their chromosomes and complete fewer cell divisions before the onset of replicative senescence than cultures derived from normal (non-Werner) individuals.
4. Some investigators have recently reported that the frequency of gene mutations in cell populations in the body increases with age. These observations are consistent with a diminution of the fidelity of DNA replication and/or repair with advancing age. Alternatively, this age-associated increase of mutation frequency could merely reflect a steady accumulation through time. At the present time there is no definitive evidence for the existence of decrements in the efficiency and/or fidelity of DNA replication or repair with advancing age.

There are a large number of proteins involved in the replication and repair of DNA. At this time there are 125 genes known to be directly involved in DNA repair. The products of these genes perform many specific functions in the repair process, including: recognition of damaged sites (DNA binding proteins); excision of the damaged region (exonucleases and endonucleases); replication of a new strand following excision of the damaged area (polymerases); and ligation of the newly synthesized segment of the strand (ligases).

Following completion of the first draft of the human genome, the identification of polymorphic alleles in these loci has proceeded very rapidly. For example, as of July 2001, 252 SNPs had been identified in genes that are associated with DNA repair. Moreover, more genes that code for proteins that are involved in DNA repair are being discovered. Even allowing for some inaccuracies at this time in the current Utah database, the frequency of this class of DNA polymorphisms is such that the existence of alleles with differing functional activities is almost a certainty.

The identification of specific alleles or groups of alleles (haplotypes) that effect the aging phenotype and/or longevity will involve multiple experimental approaches, as described above. If specific allelic associations are shown to be associated with some aspect of aging or longevity, the next step will be to establish a causal relationship between the alleles and the aging process. These studies will include a biochemical characterization of each allele to identify functional alterations, such as increased enzymatic activity. Transgenic technology—the insertion or deletion of genes into the germ line of experimental animals—will certainly play a pivotal role in establishing a cause-and-effect relationship between specific alleles or haplotypes and the aging phenotype.

Potential significance

Whether studies of the effect of DNA polymorphisms in genes involved in the replication and repair of DNA lead to significant new insights into the causes of aging will not be known until such studies have been completed. The potential significance is high, however. A convincing demonstration that genes involved in DNA replication and repair effect the life span and the phenotype of aging would provide strong support for the hypothesis that mutations in somatic cells contribute to the aging process.

Will these studies be of importance in other areas of biomedical research? The answer to this question is most certainly "yes." All aspects of DNA metabolism are essential for the survival of virtually all organisms. The identification, mapping, and functional studies of polymorphic loci will provide important new information about the mechanisms of DNA replication and repair. At a more applied level, these studies will almost certainly increase our understanding of many diseases, including cancer. It is now generally accepted that this disease is caused by gene mutations in somatic cells, and studies of the nature and extent of variation in genes involved in DNA repair and replication will be a very important area of cancer research.

Whether studies of genetic variation in aging will have practical applications is, again, dependent on the outcome of research in this area. An obvious possibility is that information derived from these studies will permit the identification of individuals who are at increased risk for the development for specific age-related conditions and diseases. Such developments would raise some ethical concerns, mainly concerning the potential for discrimination. On the other hand, with the development of effective therapies and preventive programs, the ability to accurately predict the risk of developing specific diseases years, or even decades, in advance of their onset would be of immense practical value in the treatment of elderly patients.

Thomas H. Norwood

See also Accelerated Aging: Human Progeroid Syndromes; Alzheimer'S Disease; Cancer, Biology; Cellular Aging: Telomeres; DNA Damage and Repair; Genetics: Gene Expressions; Genetics: Gene-Environment Interaction; Stress.

BIBLIOGRAPHY

Baumforth, K. R. N.; Nelson, P. N.; Digby, J. E.; O'Neil, J. D.; and Murray, P. G. "The Polymerase Chain Reaction." Journal of Clinical Pathology: Molecular Pathology 52 (1999): 1–10.

Blacker, D., and Tanzi, R. E. "The Genetics of Alzheimer's Disease: Current Status and Future Prospects." Archives of Neurology 55, no. 3 (1998): 294–296.

Caruso, C.; Candore, G.; Romano, G. C.; Lio, D.; Bonafe, M.; Valensin, S.; and Franceschi, C. "Immunogenetics of Longevity. Is Major Histocompatibility Complex Polymorphism Relevant to the Control of Human Longevity? A Review of Literature Data." Mechanisms of Ageing and Development 122 (2001): 445–462.

Finch, C., and Kirkwood, T. B. L. Chance, Development, and Aging. New York: Oxford University Press, 2000.

Flint, J., and Mott, R. "Finding the Molecular Basis of Quantitative Traits: Successes and Pitfalls." Nature Reviews Genetics 2, no. 6 (2001): 437–445.

International Human Genome Sequencing Consortium. "Initial Sequencing and Analysis of the Human Genome." Nature 409 (2001): 860–921.

The International SNP Working Group. "A Map of Human Genome Sequence Variation Containing 1.42 Million Single Nucleotide Polymorphisms." Nature 409 (2001): 928–933.

Kent, J. "Human Genome Browser." In Human Genome Project Working Draft. World Wide Web document. http://genome.ucsc.edu

Lander, E. S. "The New Genomics: Global Views of Biology." Science 274 (2001): 536–546.

Martin, G. M. "Genetics and Pathobiology of Ageing." Philosophical Transactions of the Royal Society London: Biological Sciences 352 (1997): 1773–1780.

Martin, G. M.; Oshima, J.; Gray, M. D.; and Poot, M. "What Geriatricians Should Know about the Werner Syndrome." Journal of the American Geriatric Society 47 (1999): 1136–1144.

Ronen, A., and Glickman, B. W. "Human DNA Repair Genes." Environmental and Molecular Mutagenesis 37 (2001): 241–283.

Szilard, L. "On the Nature of the Aging Process." Proceedings of the National Academy of Sciences (USA). 45 (1959): 30–45.

Vijg, J.; DollÉ M. E. T.; Martus, H.-J.; and Boerrigter, M. E. T. A. "Transgenic Mouse Models for Studying Mutations In Vivo: Applications in Aging Research." Mechanisms of Ageing and Development 98 (1997): 189–202.

Vijg, J. "Somatic Mutations and Aging: A Reevaluation." Mutation Research 447 (2000): 117–135.

Ventner, J. C., et al. "The Sequence of the Human Genome." Science 291 (2001): 1304–1351.

Vogel, F., and Motulsky, A. G. "Population Genetics: Description and Dynamics." In Human Genetics. Problems and Approaches, 3d ed. Edited by F. Vogel and A. G. Motulsky. New York: Springer, 1997. Pages 497–508.

Wright, W. E., and Shay, J. A. "Cellular Senescence as a Tumor-Protection Mechanism: The Essential Role of Counting." Current Opinion in Genetic Development 11 (2001): 98–103.