Causes of Aging

PDFlib PLOP: PDF Linearization, Optimization, Protection

Page inserted by evaluation versionwww.pdflib.com – [email protected]

61

Causes of AgingROBIN HOLLIDAYa

CSIRO Molecular Science, Sydney Laboratory, P.O. Box 184,North Ryde, NSW 2113, Australia

ABSTRACT: A broad biological approach makes it possible to understand whyaging exists and also why different mammalian species have very different maximumlongevities. The adult organism is maintained in a functional state by at least tenmajor mechanisms, which together constitute a substantial proportion of all biologi-cal processes. These maintenance mechanisms eventually fail, because the evolvedphysiological and anatomical design of higher animals is incompatible with continualsurvival. The life span of each mammalian species depends on the efficiency of main-tenance of their cells, tissues, and organs, and there is much evidence that such main-tenance is more effective in long-lived species, such as humans, than in short-livedsmall mammals. It is also evident that there is an inverse relationship between repro-ductive potential and longevity, which would be expected if available metabolicresources are shared between investment in reproduction and investment in the pres-ervation of the adult body. It is proposed that the eventual failure of maintenanceleads to the pathological changes seen in age-associated disease. Although we nowhave a biological understanding of the aging process, much future research will beneeded to uncover the cellular and molecular changes that give rise to age-associateddiseases. The major aim of such research is to devise procedures to delay or preventthe onset of these diseases.

Those who maintain that aging is an unsolved problem in biology tend to take a narrowview, believing that a single cause of aging exists, or that it is controlled by a few “geron-togenes.” A broad view, which encompasses a considerable proportion of the whole ofbiological knowledge, makes it clear why aging exists. This knowledge provides answersto three basic questions, at least at the biological level: Why do we age? Why do we live aslong as we do? and Why do different mammalian species have very different maximumlife spans? In answering these questions, a great deal is revealed about the mechanismsthat underpin eventual senescence and death. Almost all the material in the following dis-cussion will be found in my book Understanding Ageing,1 which is fully referenced, andother recent reviews are available.2 4

EARLY EVOLUTIONARY ORIGINS OF AGING

The aging of somatic cells must have occurred quite early in the evolution of multicel-lular animals. Initially, primitive animals probably had considerable powers of regenera-tion and renewal, as do the coelenterates and flatworms today. Such organisms may bepotentially immortal, although in natural environments their life would be ended by one ofmany environmental hazards. As more complex animals evolved the distinction betweenthe germline and the soma, or body, became much more clear-cut, and in particular organ-isms evolved where all the cells of the body are postmitotic, except the germline cells.

aTel: 61 2 9490 5156; fax: 61 2 9490 5010; e-mail: [email protected]

62 ANNALS NEW YORK ACADEMY OF SCIENCES

This is the case in nematodes and many insects. Such animals, when kept under good envi-ronmental conditions, clearly have a finite life span. This can be attributed to the fact thatnondividing cells, active in metabolism, cannot be expected to survive indefinitely.

At first sight, it seems that aging is nonadaptive, inasmuch as an organism that can sur-vive and reproduce indefinitely is fitter in Darwinian terms than one that reproduces for agiven period of time and then dies. Why then did aging evolve in the first place? Theanswer to this question, oddly enough, lies in the Darwinian realization that organismsnormally produce far more offspring than can possibly survive and reproduce themselves.The environment is hostile, and individuals are competing for limited resources. Thiscompetition results in the natural selection of the fittest. In these circumstances the proba-bility of an organism surviving and reproducing for a long period become very small, sopotential immortality confers very little, if any, adaptive advantage. In other words, suchorganisms are not necessarily the fittest because resources are used to maintain the somafor a long period of time. It is a better strategy for the survival of an organism’s lineage toinvest resources into growth to adulthood and reproduction, rather than in long-term main-tenance of the soma. Thus, the organism that evolves a soma with a limited survival timeis at an advantage over one that attempts to maintain the soma indefinitely. This disposablesoma theory neatly explains the early origins of aging in animals.5,6

Subsequently, as evolution proceeded, there arose many variations in the pattern ofaging. Many adult vertebrates grew continuously, and these tended to have very long lifespans. Although the signs of senescence were less obvious than in species that had con-stant adult body size, their survival for a century or so was still a minute fraction of evolu-tionary time. Life span variability is seen particularly in fish, where small species maysurvive for a year and very large ones for several decades.2,7

Mammals and birds clearly evolved from cold blooded vertebrates, which had a finitelife span, so in a sense mammals and birds merely inherited the life style strategy thatincluded senescence and aging. The next section briefly reviews some features of themammalian body plan that make aging inevitable.

THE EVOLVED DESIGN OF MAMMALS

A vast amount of information is available about the cells, tissues, and organs of mam-mals. Much of this comes from the biomedical investigations of the human body encom-passing many disciplines. These show that many organ systems have very limited capacityfor regeneration and renewal, and it is these features of our anatomy that make senescenceand aging inevitable. The neurons of the brain are postmitotic and very active in metabo-lism. Although DNA can be repaired and proteins turned over, cells that are lost cannot bereplaced. There are many reasons why one individual cell cannot survive indefinitely.Some DNA lesions are not repaired, and some altered or abnormal proteins cannot bedegraded by proteases and therefore accumulate. The brain is very definitely a nonrenew-able structure. The same applies to the retina (an extension of the brain). The rods andcones continually synthesize photoreceptors, and the oldest are removed. This processdoes not achieve a steady state, and remnants of partially degraded photoreceptor elementsaccumulate in the cells themselves, or in the underlying epithelial layer. Eventually thedegenerative process of retinopathy occurs. The crystallin proteins of the lens of the eyeare laid down at an early stage and cannot be replaced. Lens transparency depends on their

HOLLIDAY: CAUSES OF AGING 63

molecular homogeneity. Unfortunately proteins are subject to many chemical changes,including the processes of oxidation, glycation, racemization of amino acids, and deami-dation. Because these cannot be prevented or reversed, the molecules gradually lose theirinitial properties, and cataracts may occur.

Collagen and elastin are also very long-lived proteins that are subject to chemicalchange. It is well established that collagen becomes progressively cross-linked with age,thereby losing its initial elasticity. The heart is a highly efficient pump, but like the brain, ithas very limited capacity for repair or renewal. The muscle cells are postmitotic andunlike most skeletal muscle they cannot be replaced by the division of myoblasts. Theanatomy of the major blood vessels is also incompatible with efficient repair. The cross-linking of elastin and collagen results in hardening of arteries, and the inner wall is subjectto damage, including the buildup of atherosclerotic plaques. The basic anatomical problemis that there is only one vascular system, and it cannot be shut down for repair. It is, in fact,very difficult to repair a machine while it is operating, and the same is true of the vascularsystem. A potentially immortal organism would need to have two vascular systems, one ofwhich could be shut down and repaired, while the other kept operating. We did not evolvein that way.

Teeth provide an instructive example of the way components of the body have evolved“to last a lifetime.” Clearly the shape and size of adult teeth are genetically determined,but they are also subject to wear and tear, as well as decay. This is one of many examplesthat demonstrate the artificiality of the distinction that is often made between “wear andtear” theories of aging (or the stochastic accumulation of various defects) and the “pro-gram” theories. Both, in fact, are interrelated and important. Some herbivores that contin-ually crop plants have incisors that keep growing at the base, which is clearly a secondaryadaptation to produce “immortal” teeth. Many other herbivores, however, do not have thisability, and it is well known that an estimate of a horse’s age can be made by examinationof the wear on its teeth.

MAINTENANCE OF THE ORGANISM

Although the evolved design of many body components is incompatible with indefinitesurvival, this does not mean that maintenance mechanisms are unsuccessful. The life his-tory of a mammalian organism comprises development and growth to the adult, a fairlylong period of reproduction, followed by loss of fertility, and the senescence, and death.Maintenance of cell, tissue, and organ function is essential during development and repro-duction. The total resources available to a mammalian organism are allocated to three majorfunctions: first, ongoing metabolism, second, all aspects of reproduction, and third, a set ofmaintenance mechanisms. These three functions consume all available metabolic energy,and although there may be some overlap between them, it is possible to itemize their mainfeatures, as shown in TABLE 1. The major maintenance mechanisms are as follows.

Wound Healing

Damage to skin and muscle can be effectively repaired, and broken bones can rejoin.Loss of blood is prevented by clotting, and the smaller arteries and veins can be replaced.


Nevertheless, mammals do not have strong regenerative capacity. Severed limbs or digitsare not replaced, and major nerves that are cut cannot be rejoined. In this respect, somelower vertebrates have greater regenerative capacity, inasmuch as lost limbs can beregrown.

Immunity

All organisms are subject to attack by pathogens and parasites, and a complex immunesystem has evolved to protect the organism. Immunology is, of course, a science in its ownright, and leaving aside the “immunologic” theory of aging,8 the immune response is notthought to have any relationship to the study of longevity or aging. Nevertheless, it is avital maintenance mechanism, and without it an organism does not survive very long.

DNA Repair

Although DNA is a stable molecule, it is continually subject to intrinsic and extrinsicdamage. It is highly likely that oxygen free radicals are an important source of damage.4 Abattery of repair enzymes exists that continually monitors DNA for abnormalities in struc-ture, excises or removes such damage, and then fills in any gaps by repair synthesis andrejoining. One of the commonest defects in DNA is the loss of purine residues. Indeed, ithas been estimated that up to 10,000 of these lesions occur in each cell per day.9 All thisdamage is effectively repaired. There may be lesions, however, that are not repaired, per-haps because they are less common, and the necessary enzymes have never evolved todeal with them.10 Also, there may be adjacent lesions on both strands of a DNA moleculethat are difficult to repair and that can lead to chromosome breaks.

Synthesis of Macromolecules

DNA repair overlaps with mechanisms to ensure that DNA is synthesized with extremeaccuracy. The insertion of an incorrect base by the replicating polymerase is usually cor-

TABLE 1. The Allocation of All Available Energy Resources in Mammals

Normal functions Reproduction Maintenance

Biochemical synthesisMetabolismRespirationCell turnoverMovementFeeding and digestionExcretion

Gonads, gametes, and sexDevelopmentGestationSucklingCare of offspringGrowth to adult

Wound healingImmunityProtein turnoverDefense against free radicalsProofreadingDNA repairDetoxificationEpigenetic stabilityApoptosisFat storageHomeostasis


rected by an editing excision/replacement mechanism. However, if this fails, there is abackup mismatch repair system. The removal of errors in DNA synthesis depends onmany enzymes and accessory proteins. RNA and proteins are made with less accuracy;nevertheless it would be wasteful, as well as harmful, to synthesize defective molecules,so it is not surprising that proofreading mechanisms exist to detect and remove errors. Allthese proofreading mechanisms consume energy. The question of the optimum accuracyof synthesis of macromolecules is an interesting one. In general, it seems to be the casethat rapid synthesis results in more errors, and slower synthesis allows time for more effi-cient editing. There must be an optimum or some balance between the two, which maywell not be the same for all mammalian species (see below).

Protein Turnover

As has been mentioned, protein molecules are subject to many postsynthetic modifica-tions. Some modifications are, of course, a normal part of the maturation of proteins andplay essential roles in their function, but there are many others that are abnormal, withpotentially deleterious effects on the cell. These molecules are usually recognized andremoved by proteases and the proteosome. This is a very important ongoing process,essential for the normal function of cells. Amino acids that cannot be reused are brokendown and the nitrogen excreted in the form of urea. The removal of abnormal proteins isnot completely successful, particularly if the protein is inaccessible or is part of a nonre-placeable structure (such as the walls of major arteries). Also, altered proteins may formhigh molecular weight aggregates that are resistant to proteolytic digestion such as AGEs(advanced glycation end products) or the amyloid plaques in Alzheimer’s disease. Thegradual accumulation of these high molecular weight protein or peptide aggregates are animportant part of the aging process.

Detoxification

Animals have complex diets, and toxic chemicals are a common component. In partic-ular, plants often defend themselves against animals by synthesizing such compounds. Inresponse, mammals have evolved a large set of detoxifying enzymes, collectively knownas the P450 cytochromes. These comprise a very complex family of enzymes located inthe liver, but also in other tissues that can degrade a very wide range of chemicals. Nowa-days, these include many man-made chemicals that would never have been encounteredduring evolution. Thus, the detoxification system has an inbuilt “overkill” capacity to dealwith any new chemicals that may arise in the diet or environment.

Defences against Free Radicals

Oxygen free radicals are continually generated by respiration and some other meta-bolic processes. Although very short lived, they are highly reactive and can damage DNA,proteins, and membranes. Organisms have developed major defences against free radicalattack. There are enzymes that break down free radicals, such as superoxide dismutase,


catalase, glutathione peroxidase, and reductase. Metabolites exist that react with free radi-cals, acting as free-radical “sinks,” such as carotenoids or other antioxidants. It is likelythat the evolution of the respiratory organelle, the mitochondrion, protects the chromo-somal DNA in the nucleus from free-radical attack. It is well known that the small mito-chondria DNA genome mutates at a much higher rate than chromosomal DNA.

Epigenetic Controls

Differentiated cells stably maintain their particular biochemical and morphologicalcharacteristics. This depends on the activities of genes responsible for the cells’ special-ized functions, together with the inactivity of all the genes needed for all other specializedcells. These controls of gene activity are generally referred to as epigenetic, and they aresuperimposed on the information in DNA, which is present in all cells. Many believe thatepigenetic controls are entirely due to proteins that bind to specific DNA sequences, butthere are now many indications that chemical modification of DNA is an essential compo-nent. The major modified base in mammals is 5-methylcytosine, and it is known that thepattern of this methylation is inherited through mitotic division, and therefore stably main-tained in those specialized cells that are capable of division, as well as in postmitotic cells.Obviously, it is extremely important to maintain epigenetic controls, because if normalregulation is lost, then a cell can adopt an abnormal phenotype and become, for example, aneoplastic cell. This can occur through mutation, but epigenetic defects are also likely tobe involved.11

Apoptosis

The suicide mechanism known as apoptosis is triggered in a variety of contexts. Itremoves unwanted cells during development, or in the immune system, but it also comesinto play when damaged or abnormal cells arise. Otherwise such cells would have harmfuleffects on the organism. Although it has been suggested that aging and apoptosis may belinked, the relationship is not at all simple. Apoptosis is, at least in part, a maintenanceprocess to prevent deleterious changes. If apoptosis does not come into play, for whateverreason, then an abnormal cell will survive, and this may contribute to senescence andaging.

Homeostatic Mechanisms

These comprise a large set of physiological or regulatory processes that maintain cells,tissues, and organs in a normal functional state. The most important homeostatic mecha-nism in mammals and birds is the control of body temperature. This produces a muchmore uniform internal environment, with less dependence on the external one. Therefore,mammals and birds can colonize a wider range of environments than cold-blooded verte-brates. It also allows many biochemical processes to be optimized, with the activity ofmany proteins adapted to body temperature. Many other homeostatic mechanisms dependon hormones or growth factors, which ensure that potential variables (such as blood-sugar


levels) are controlled. Too many examples exist to review here, but it is worth mentioningthat a stress response, such as the heat-shock response, can be regarded as a cellularhomeostatic maintenance mechanism, which protects cells from an even greater rise intemperature.

Fat Storage

In natural environments it is common for the availability of food to vary considerably.Thus, periods of glut may alternate with periods of scarcity. To ensure survival in theabsence of food, mammals have evolved an efficient energy storage mechanism that cantide them over periods in a harsh environment. The laying down and reuse of fat can there-fore be regarded as a maintenance mechanism.

There are several features of the totality of maintenance mechanisms that should beemphasized. First, the study of all these processes comprises a major part of all biologicalresearch. Inasmuch as aging and death are ultimately the result of failure of maintenance,it is not unreasonable to propose that all this research is in one way or another related tothe study of aging itself. Second, it is fashionable to invoke specific “gerontogenes” that insome way control longevity and aging, but there are innumerable genes that specify thecomponents of all maintenance mechanisms. All of these relate in one way or another tothe efficiency and also the eventual failure of maintenance. We know of many examples ofsingle gene mutations that have pleiotropic effects on the phenotype, and some of theseclearly relate to aging. Third, the various theories of aging that have been proposed usuallyrelate quite closely to failure of maintenance. Thus, the oxygen free radical theory of agingis directly related to the failure to nullify their dangerous effects. This, in turn, overlapswith the somatic mutation theory, which is clearly related to the failure of DNA repair. Theprotein error theory proposes that abnormal proteins can cause escalating damage byreducing the accuracy of synthesis. Clearly this is related to the failure to remove abnor-mal molecules by proteolysis. This failure also results in the accumulation of abnormalprotein molecules, which constitutes another theory of aging. The immunologic theory ofaging suggests that the immune system eventually loses its ability to distinguish self fromnonself antigens and therefore inflicts pathological damage on cells and tissues. The dys-differentiation theory of aging proposes that ectopic protein synthesis (i.e., the synthesis ofa specialized protein in an inappropriate cell) is an important feature of senescence. This isrelated to the loss of epigenetic controls. It is very likely that there is some truth in all ofthese theories of aging, because aging is multicausal.1

Finally, new information about the importance and complexity of maintenance is con-tinually being obtained, and a recent example is the discovery of peptide antibiotics inhuman skin.12 Clearly this is an important defense mechanism against bacterial infection,which is rather distinct from the more familiar immune responses to infection.

REPRODUCTION, MAINTENANCE, AND LONGEVITY

The “disposable soma” theory of the evolution of aging and longevity predicts thatthere should be some trade-off between resources invested in rapid growth and reproduc-


tion and resources invested in maintenance of the soma. The balance between reproduc-tion and maintenance depends on the level of environmental hazard. In a high-riskenvironment, where annual mortality is very high, it would then be expected that develop-ment and reproduction would be rapid; fewer resources would be devoted to maintenance,and therefore a shorter life span would be seen. In a low-risk environment, the annual mor-tality is much lower, so we would expect the evolution of slow-breeding long-lived spe-cies. In the adaptive radiation of mammals, there have been evolutionary trends thatincrease reproduction and reduce longevity. The carnivores provide one example, becausethe highly specialized stoats and weasels need a continual supply of food (i.e., they live ina high-risk environment), produce large litters that mature rapidly, and have a short lifespan in captivity. The other trend is an increase in longevity, exemplified by the primates.Small monkeys reproduce rapidly and have short life spans; larger monkeys, small apes,the great apes, and humans have progressively longer life spans. This is associated withfewer offspring and a much lower annual mortality. If lower mortality is regarded as a“successful” adaptation to the environment, then it is also clearly associated with naturalselection for longer life spans.

The foregoing use of the terms life span and longevity refers to the documented lengthof life of mammalian species kept in captivity, that is, the protected environment of a zooor a laboratory cage. There is some relationship of this measured life span with the likelysurvival of that species in a natural environment, which is usually hard to determine. How-ever, if it is assumed that the population size is constant, it is possible to calculate the aver-age expectation of life at birth, provided the various reproductive parameters are known.These are the gestation period, the litter size, the time to develop to a fertile adult, and theinterlitter interval. For early human-hunter gatherers, assuming a constant population size,the expectation of life at birth was only about 16 years; for females that reached reproduc-tive age, it was about 28 years.13

The best available reproductive and life span-in-captivity data for 47 mammalian spe-cies demonstrates a clear inverse relationship between maximum life span and reproduc-

TABLE 2. Correlation between Maintenance Parameters and Maximum Life Span of Mammalian Speciesa

Positive CorrelationsLongevity of fibroblasts in vitroLongevity of erythrocytes in vivoDNA repairPoly-ADP ribose polymerase-Ray-induced ADP-ribose transferase

Carotenoids in serum

Negative CorrelationsCross-linking of collagenProduction of oxygen free radicalsAuto-oxidation of tissuesMetabolic rate and oxidized DNA basesDNA methylation declineCarcinogen binding to DNAMutagenicity of activated carcinogenIncidence of cancer

aFor sources, see refs. 1, 11, and 15.


tive potential.1,14 The fecundity/life span ratio is highest in small living rodents andrabbits, then decreases through small carnivores, small primates, large carnivores, largerherbivores, pachyderms, the great apes, and humans. Many attempts have been made overthe years to relate maximum life span to metabolic rate, weight, and brain size, or anycombination of these in mammalian species. It is often found that bats (Chiroptera) with ahigh metabolic rate provide an exception to any general rule. It is striking that bats havelong life spans and low rates of reproduction, as expected from their low-risk lifestyle. Theanalysis of reproductive potential and maximum life span strongly confirms a predictionof the disposable soma theory.

Another prediction is that the efficiency of maintenance should relate to maximum lon-gevity. A number of comparative studies have been carried out, although more are needed.In almost every case there is the expected relationship between efficiency of the mainte-nance parameter studied and the maximum life span of the species. In other cases, the rela-tionship is inverse, but this is also in the expected direction. The studies that have beenpublished are listed in TABLE 2 (for sources, see refs. 1, 11, and 15).

AGE-RELATED DISEASES

It is commonly assumed by clinicians that the pathological conditions commonly seenin the elderly are distinct from “natural aging.” Dementia in the sixth and seventh decadeis Alzheimer’s disease, but dementia in the tenth decade, or thereafter, is natural aging.Also, it is said that an age-related disease should be considered as such, because many eld-erly people never exhibit the symptoms. Thus, many centenarians have no obvious sign ofheart disease.

From a biological point of view, what should we expect? It is clear that aging com-prises a deterioration of many organ systems that are not obviously related to each other.The loss of neurons in the brain is not obviously related to the accumulation of atheroscle-rotic plaques in aortas or the loss of transparency of the lens. The cross-linking of collagenand skin wrinkling are not obviously related to retinopathy, and so on. There is a degree ofsynchrony in all known age-related changes, but the synchrony is certainly not exact. Wewould therefore expect that in a given individual, one tissue or organ system deterioratesin advance of others. This is then usually diagnosed as a disease or pathological state. Inanother individual a different degenerative condition may appear. Although aging itselfcannot be regarded as a disease, the various diseases associated with old age can certainlybe regarded as part of the overall processes of aging. This applies to dementia, cardiovas-cular and cerebrovascular disease, osteoarthritis, osteoporosis, late-onset diabetes, renalfailure, loss of sight or hearing, carcinomas, as well as many other less well-known condi-tions. These diseases can be broadly related to the failure of various maintenance mecha-nisms, as shown in TABLE 3.

It cannot be overemphasized that all the research devoted to the study of age-relateddiseases is in fact related to gerontology itself. This research has three aims: (1) bettertreatment of the disease in question, (2) the elucidation of the cause, or causes, of the dis-ease, and (3) the development of procedures to postpone or prevent the onset of the dis-ease. Aims two and three can certainly be regarded as being within the province ofgerontology, and therefore more research on aging itself is very likely to throw much lighton the origins and development of age-related disease. This was clearly recognized by the


geriatrician Steiglitz more than 55 years ago in his article “The social urgency of researchon aging.”16 Unfortunately, much persuasion is necessary to convince the present commu-nity of clinical and biomedical research scientists.

CONCLUSIONS

We now have answers to the three basic questions posed at the beginning of this article.We age because we evolved from organisms that also age. We age because our evolvedbody structure is incompatible with continual survival. We age because our various main-tenance mechanisms eventually fail to preserve the normal structure and function of cellsand tissues. We live as long as we do because we have evolved a lifestyle with low annualmortality.13 This has allowed for more resources to be invested in maintenance and less inreproduction. By contrast, species that live in a high-risk environment can only survive byinvesting much more heavily in reproduction, with correspondingly fewer resources allo-cated to maintenance. Thus, the adaptive radiation of mammals to many ecological nicheshas also resulted in the evolution of longevities over an approximately 50-fold range.Thus, when considered at the level of the organism, aging is no longer an unsolved prob-lem in biology.

Nevertheless, at the level of fine detail, the actual molecular and cellular changes thatproduce the aging phenotype, there is a great deal to learn. An understanding of thesechanges will come from further studies of maintenance mechanisms, and more important,the reasons why maintenance eventually fails. This new knowledge will greatly increaseour understanding of the origins of age-associated disease and will concomitantly make itpossible to prevent or delay the onset of these diseases. The aim of all this research is notto increase the overall life span but to significantly extend the “health span” so that thequality of the elderly is greatly improved and the costs of health care for the aged aregreatly reduced.

TABLE 3. General Relationships between Cell or Tissue Maintenance and Some Major Human Age-associated Diseases

Failure of maintenance Major pathologies

Neurons Dementias

Retina, lens Blindness

Insulin metabolism Type II diabetes

Blood vessels and heart Cardiovascular and cerebrovascular disease

Bone structure Osteoporosis

Immune system Autoimmune disorders

Epigenetic controls Cancer

Joints Osteoarthritis

Glomeruli Renal failure


REFERENCES

1. HOLLIDAY, R. 1995. Understanding Ageing. Cambridge University Press. Cambridge.2. FINCH, C.E. 1990. Longevity, Senescence and the Genome. University Press. Chicago.3. GRIMLEY EVANS, J. & T. FRANKLIN WILLIAMS, Eds. 1992. Oxford Textbook of Geriatric Medicine.

Oxford University Press. Oxford.4. MARTIN, G.M., S.N. AUSTAD & T.K. JOHNSON. 1996. Genetic analysis of aging: Role of oxidative

damage and environmental stress. Nat. Genet. 13: 25 34.5. KIRKWOOD, T.B.L. & R. HOLLIDAY. 1979. The evolution of ageing and longevity. Proc. R. Soc.

Lond. B 205: 531 546.6. KIRKWOOD, T.B.L. 1985. Comparative and evolutionary aspects of longevity. In Handbook of the

Biology of Aging. C.E. Finch & E.L. Schneider, Eds.: 27 44. Van Nostrand Reinhold. NewYork.

7. COMFORT, A. 1979. The Biology of Senescence, 3rd Ed. Churchill Livingstone. London.8. WALFORD, R.L. 1969. The Immunologic Theory of Ageing. Munksgaard. Copenhagen.9. LINDAHL, T. 1979. DNA glycosylases, endonucleases for apurinic/pyrimidinic sites and base

excision-repair. Prog. Nucleic Acid Res. Mol. Biol. 22: 135 192.10. LINDAHL, T. 1993. Instability and decay of the primary structure of DNA. Nature 362: 709 715.11. HOLLIDAY, R. 1996a. Neoplastic transformation: The contrasting stability of human and mouse

cells. In Genetic Instability in Cancer. T. Lindahl, Ed.: 103 115. Cancer Surveys 18: ColdSpring Harbor Laboratory Press. New York.

12. HARDER, J., J. BARTELS, E. CHRISTOPHERS & J-M. SCHRODER. 1997. A peptide antibiotic fromhuman skin. Nature 387: 861.

13. HOLLIDAY, R. 1996. The evolution of human longevity. Perspect. Biol. Med. 40: 100 107.14. HOLLIDAY, R. 1994. Longevity and fecundity in eutherian mammals. In Genetics and Evolution

of Aging. M.R. Rose & C.E. Finch, Eds.: 217 225. Kluwar Academic Publishers. The Neth-erlands.

15. HOLLIDAY, R. 1997. Understanding Aging. Philos. Trans. R. Soc. Lond. B 352: 1793 1797.16. STEIGLITZ, E.J. 1942. The social urgency of research in ageing. In Problems of Ageing: Biologi-

cal & Medical Aspects, 2nd Ed. E.V. Cowdray, Ed.: 890 907. Williams & Williams. Balti-more.

Documents

Causes of Aging