1. GENETICS IN AGING
1.1. GENETIC APPROACHES TO AGING RESEARCH
A very large volume of descriptive data on age-related changes in biochemical, physiological and pathological parameters has been accumulated during the past 50 years, particularly for the case of mice and man. It has not been possible, however, to determine, from such information, a sequence of events that can elucidate the initial, primary factors responsible for the cascade of complex events that accompany aging. Genetic approaches offer that potential as, by definition, one is dealing with alternative forms of information present at the time of conception. Our purpose is to point out some of the many opportunities to carry out genetic research on the biology and pathobiology of aging in species other than Drosophila melanogaster and Caenorhabditis elegans.
1.1.1. GENETIC LOCI IN HUMAN OF POTENTIAL RELEVANCE TO THE PATHOBIOLOGY OF AGING
Except for the recent interest in allelic variants associated with centenarians (1), research in this area has been confined to the study of genetic loci uncovered by physicians and medical geneticists confronted with pedigrees with accelerated on- sets and/or severe expressions of phenotypes that overlap with what can be observed in "usual" senescence. These were phenotypes generally regarded as diseases of aging, in that their expression may be coupled to some underlying process or processes of aging. By focusing upon gene action in the expressed gene, the hope is that one can "work backwards", as it were, in order to arrive at abetter understanding of mechanisms of aging. There is also the expectation that, for every such locus discovered on the basis of a strong mutational effect, there could, in principle, exist alleles that could confer unusual resistance to these phenotypes.
Such alleles would elucidate mechanisms that maintain structure and function during the latter course of the life course. Based upon a survey of such loci, which was conducted in 1978 (2). It was concluded that several thousand genetic loci of human could play a role in our varying susceptibilities to different aspects of the pathobiology of aging. It also suggested that a smaller subset might be of exceptional interest. These were of two categories. The first was categorised as "segmental progeroid syndromes", as they appeared to impact upon many aspects of the senescent phenotype of man. The most striking of these was the Werner syndrome, sometimes referred to as "Progeria of the adult", to differentiate it from "Progeria of childhood" (also known simply as "Progeria", or, more correctly, as the Hutchinson-Gilford syndrome).
These disorders differ from each other in many ways, including the fact that WS is now well established as an autosomal recessive disorder, whereas the mode of inheritance of the Hutchinson-Gilford syndrome is less well established, but is likely to be an autosomal dominant (3). The second category of genetic loci of special interest involves a more restricted phenotype ("unimodal progeroid syndromes") (4). Some of these are of exceptional medical importance. Examples would be loci that modulate susceptibility to dementia of the Alzheimer type (DAT), a disorder that may involve almost 50% of individuals over the age of 85 years in some populations (5).
The key concept involved in the evolutionary theory of why we age is the attenuation of the force of natural selection with respect to the age of gene effects (6). Two classes of gene action are thought to form the basis for the emergence of senescence. The first postulates a role for late-acting constitutional mutations (i.e. mutations one is born with, as opposed to acquired mutations of subsets of somatic cells). The second class of gene action involves alleles selected for beneficial effects early in the life course, but which have deleterious effects late in the life course, when they escape the force of natural selection. This type of gene action has been referred to as "antagonistic pleiotropy" (6). Once selected, such alleles are likely to spread throughout a population. Moreover, they may well have evolved in a number of independent populations and species. One might therefore refer to such alleles as leading to "public" modulations of aging (7).
In brief, these arguments emphasise a number of clinical and pathological discordances between what one observes in WS and in "usual" aging. To cite just one example, whereas the ratio of epithelial to mesenchymal neoplasms in usual aging is about 10: 1, that ratio decreases to about I: 1 in WS (9). Moreover, there are relatively high frequencies of unusual cancers in WS. An example is the approximately 1000- fold increases in the prevalence of acrallentiginous melanoma, a special form of melanoma involving the soles of the feet, nail beds and nasal mucosa (9). There are also cell biological lines of evidence that argue that the lesion in WS may result in a unique pathway to somatic cell dysfunction. Cultures of WS somatic cells have markedly reduced replicative potentials compared to cultures derived from age-matched control (10). The exits from the mitotic cell cycle of WS cells may be different from what obtains for the case of control cultures.
Unlike controls, senescent WS cultures are still capable of robust expression of cFOS (II). WS cells have elongated S phases and cells may become arrested in S (12). While the loss of telomerase enzyme activity, with subsequent reduction in telomere repeats, appears to explain the replicative senescence of normal somatic cells, the telomere restriction fragment lengths of senescent WS cells are not unusually short (13). Nevertheless, the fact that WS cells exhibit genome instability (8) raises the following question: Do other pathways leading to genome instability provide more faithful modulations of senescent phenotypes? As we shall see below, mutations do accumulate in aging mammalian tissues and must surely be related to the cancers of old age, which are observed, so far as we know, in all aging mammals, and which appear to scale to lifespan despite environmental differences and differences in tumor spectrum (14).
WS is now known to be caused by null mutations at a locus (WRN) coding for a member of the RecQ family of helicases (15-19). As such, the gene product could playa role in such processes as DNA replication, DNA repair, DNA recombination, gene transcription and chromosome segregation. Biochemical studies from the laboratories of Judy Campisi and Larry Loeb now indicate the presence of a 3'-5' exonuclease domain in the Werner. This would be consistent with a role in DNA repair. The gene product is localised to the nucleoli of actively replicating human cells (20, 21), raising the possibility that ribosomal DNA may serve as a substrate. The Werner protein may, however, shuttle between the nucleoplasm and the nucleoli (21). It will take some time before we understand the specific molecular functions of the Werner protein.
While WS is extremely rare (1-22 per million) (22), the prevalence of heterozygous individuals may be as high as 1/150 to 1/200 individuals, at least in Japan (15). Somatic cells from such individuals have been shown to be hypersensitive to a powerful carcinogen and genotoxin, 4-nitroquinoline-1-oxide (23). This laboratory observation supports a preliminary epidemiological study suggesting tha1 family members of WS pedigrees may be at higher risks for malignancy (24).
Another important issue is the degree to which "leaky" mutations or polymorphism at the WRN locus may influence susceptibility to common age-related disorders. A single publication indicates that one such polymorphism, involving the substitution of an argenine for cysteine three amino acids proximal to the C-terminal nuclear localisation signal, may protect Japanese subjects from myocardial infarction (25). Like all such genetic association studies, however, that observation requires confirmation in a variety of populations.
1.1.2. COMPARATIVE GERONTOLOGIC APPROACH
We wished to determine the frequencies of somatic cell mutation in homologous epithelial cell types (renal tubular epithelial cells) from mammals with short (C57BL/6 mice), long (human) and intermediate (beagle dogs) lifespans. We chose an endogenous X-linked recessive locus (HPR7) for which selective methods existed for their detection. Endogenous targets have certain advantages over systems, that utilise transfected reporter genes, as results for the latter could be influenced by position effects. (On the other hand, the great merit of the LA C Z plasmid reporter system is that one can determine mutation frequencies in all cell types). Gerontologists continue to be interested in Leslie Orgel's protein synthesis error catastrophe theory of cell senescence (26, 27). That theory predicted an exponential cascade of somatic mutations towards the end of the lifespan.
This was in contrast to the predictions of the intrinsic mutation theory of the late Sir Macfarlane Burnet for which linear increases in accumulations would be envisioned (28). Finally, there was the possibility that the rates of accumulation of mutations would simply be functions of chronological time. With that scenario, more mutations would eventually accumulate in human subjects simply because they live so long. The results of these still incomplete investigations have been summarised elsewhere (29). For the case of humans there was a highly significant increase in the frequencies of mutation with increasing donor age (30). The best fits for the kinetics for both canine and human cells were exponential. The frequencies of mutations in these human epithelial cells were about 10-fold greater than those reported for human peripheral blood T lymphocytes (31).
Identical assays on aging mice failed to provide evidence for accumulation of HPRT mutations (32). By contrast, however, previous and unpublished studies indicate that large-scale chromosomal mutations are very common in aging mice, much more common than in canines or humans (33). We can tentatively conclude that the patterns of somatic mutations are variable among mammalian species. Such results are consistent with one of the predictions of the evolutionary biological theory of aging -that one can anticipate substantial interspecific variations in patterns of aging (34).
18.104.22.168. HOW DO MANY SPECIES OF BIRDS MANAGE TO ACHIEVE THEIR UNUSUAL LONGEVITIES?
Holmes & Austad (35) have written a valuable review of avian species as models for the comparative biology of aging They pointed out that the exceptional longevity of birds is consistent with theories on the evolution of life history parameters. Species that evolve under high hazard conditions, such as high predation, infectious disease, accidents, malnutrition and drought, evolve life history strategies that com~ine rapid development, high fecundity at sexual maturity and short lifespans. With the invention of flight, many species of birds have effectively escaped predators, thus permitting the evolution of different life history strategies, including longer periods of reproduction and prolonged lifespans. They do this despite exceptionally high metabolic rates and high blood glucose levels, thus defying two of the major postulated mechanisms of aging, macromolecular oxidative damage (36) and glycation/glycoxidation (37).
1.1.3. PROMISING EXPERIMENTAL APPROACHES
The various avenues of research briefly outlined above can be complemented by a variety of interesting experimental approaches. For example, one can ask, using hybrids and heterokaryons, if the relative resistance of avian cells is a function of nuclear genes, mitochondrial genes, or both. If nuclear, one can, in principle, attempt to complement the sensitive murine cells with avian chromosomes.
The unusual vulnerability of mice to chromosomal lesions might be related to their relatively high sensitivity to oxidative stress. One might therefore attempt to select cell lines on the basis of unusual resistance to oxidative stress and evaluate their genome stability in comparison to parental lines. We have done this for the case of murine embryonic stem cells; lines selected for growth in 95% oxygen did indeed exhibit greater genomic stability, even in the presence of ambient concentrations of oxygen (39). Attempts to create embryonic stem cell-derived transgenic mice resistant to oxidative stress have so far failed because of chromosomal lesions in the stem cells used to generate chimeras. A less stringent protocol for the isolation of moderate resistance to oxidative stress might, however, prove more successful.
Human genetics, comparative gerontology and experimental murine genetics offer powerful approaches to the dissection of mechanisms of aging and age-related disease. Among these mechanisms, oxidative damage to macromolecules may be of particular significance (7). These macromolecular targets include DNA.