Inflamm-Aging an Evolutionary Perspective on Immunosenescence

Inflamm-aging An Evolutionary Perspective on Immunosenescence

CLAUDIO FRANCESCHI,a,b,e MASSIMILIANO BONAFÈ,a SILVANA VALENSIN,a FABIOLA OLIVIERI,b MARIA DE LUCA,d ENZO OTTAVIANI,c AND GIOVANNA DE BENEDICTISd aDepartment of Experimental Pathology, University of Bologna, Bologna, Italy bDepartment of Gerontological Research, Italian National Research Center on Aging (INRCA), Ancona, Italy cDepartment of Animal Biology, University of Modena and Reggio Emilia, Modena, Italy dDepartment of Cell Biology, University of Calabria, Calabria, Italy

ABSTRACT: In this paper we extend the “network theory of aging,” and we argue that a global reduction in the capacity to cope with a variety of stressors and a concomitant progressive increase in proinflammatory status are major characteristics of the aging process. This phenomenon, which we will refer to as “inflamm-aging,” is provoked by a continuous antigenic load and stress. On the basis of evolutionary studies, we also argue that the immune and the stress responses are equivalent and that antigens are nothing other than particular types of stressors. We also propose to return macrophage to its rightful place as central actor not only in the inflammatory response and immunity, but also in the stress response. The rate of reaching the threshold of proinflammatory status over which diseases/disabilities ensue and the individual capacity to cope with and adapt to stressors are assumed to be complex traits with a genetic component. Finally, we argue that the persistence of inflammatory stimuli over time represents the biologic background (first hit) favoring the susceptibility to age-related diseases/disabilities. A second hit (absence of robust gene variants and/or presence of frail gene variants) is likely necessary to develop overt organ-specific age-related diseases having an inflammatory pathogenesis, such as atherosclerosis, Alzheimer’s disease, osteoporosis, and diabetes. Following this perspective, several paradoxes of healthy centenarians (increase of plasma levels of inflammatory cytokines, acute phase proteins, and coagulation factors) are illustrated and explained. In conclusion, the beneficial effects of inflammation devoted to the neutralization of dangerous/harmful agents early in life and in adulthood become detrimental late in life in a period largely not foreseen by evolution, according to the antagonistic pleiotropy theory of aging.

INTRODUCTION: THE NETWORK HYPOTHESIS OF AGING

According to the “network theory,”1,2 the aging process is indirectly controlled by a variety of defense functions or anti-stress responses, globally acting as anti- aging mechanisms (FIG. 1). The stressors are very diverse and include different phys-

eAddress for correspondence: Claudio Franceschi, Department of Experimental Pathology, University of Bologna, Via s. Giacomo 12, 40126 Bologna, Italy. Phone: 0039 051 209 4730 or 0039 051 209 4747. [email protected]

244 FRANCESCHI et al.: INFLAMM-AGING 245

FIGURE 1. The “anti-aging” cellular defense network (inspired by Kirkwood and Franceschi, 1992, with modifications). Each part of the system is controlled by a variety of genes, which can serve multiple pathways. Only a few relevant examples of genes affecting life span throughout evolution are reported in the figure. The activity exerted by each specific defense mechanism is expected to contribute to the overall individual longevity. How- ever, the capacity to reach the extreme limits of human life span in good shape is likely to depend on the global optimization of this network of defense mechanisms, which in turn allows for the age-related persistence of appropriate efficient integrated functions (DNA rep- lication, gene regulation, protein folding, and energy production). ical (UV and gamma radiation, heat), chemical (components of the body and products of metabolism such as oxygen-free radicals and reducing sugars), and biological (viruses, bacteria) agents.1 The efficiency of this network, which includes DNA repair enzymes and poly(ADP-ribosyl)polymerase (PARP), antioxidants (superoxide dismutase, SOD; catalase, CAT), heat shock proteins (HSP) and other stress proteins (H-Ras, RAS), and the immune system, among others, is genetically controlled and differs in various species and individuals, accounting for the observed differences in life span.2 On average, the longevity assured by each specific defense function is expected to be similar within a species. However, the genetic variability within the population is expected to result in variation in the individual extent to which the organism is predisposed to age and die from specific causes. For example, some individuals are likely to be less well protected against oxygen radicals than others, and these individuals will therefore experience a greater toll of oxidative damage. In- stances of extreme longevity, such as human centenarians, are of special interest, for they are likely to be endowed with unusually high levels of each of the important in- 246 ANNALS NEW YORK ACADEMY OF SCIENCES gredients of the anti-aging network. 2 In this paper we extend the network theory of aging, and we argue that a global reduction in the capability to cope with a variety of stressors and a concomitant progressive increase in the proinflammatory status, which we will call “inflamm-aging,” is a major characteristic of the aging process and that it is provoked by a continuous antigenic load and stress. We also argue that the immune and the stress responses are equivalent and that antigens are nothing other than particular types of stressors, on the basis of evolutionary studies strongly sup- porting this hypothesis. Following this perspective, several paradoxes of healthy centenarians are illustrated and explained. Finally, current studies on the genetics of human longevity (centenarians) are starting to identify the possible variants of genes, which lifelong modulate and optimize the immune/stress response.

THE CAPABILITY TO COPE WITH STRESS AND THE EXTENSION OF LONGEVITY IN LOWER EUKARYOTES: EVIDENCE OF A CONSERVED ANTI-AGING NETWORK

In recent years a number of lower eukaryotic organisms have been identified as useful models for studying the genetic mechanisms involved in aging and longevity. In particular, studies on yeast (Saccharomyces cerevisiae), worms (Caenorhabditis elegans), and insects (Drosophila melanogaster) indicate that molecular variants of the genes involved in the response to oxidative, radiation-induced, and thermal stress, among others, markedly influence organismal longevity.3–5 Surprisingly, most of the molecular pathways acting in these organisms are similar to those set- tling the stress defense network of mammals, including humans.3–5 Some of the most interesting examples are briefly summarized: (1) the response to oxidative stress requires the induction of superoxide dismutase and catalase (free radical scav- enger enzymes) to protect D. melanogaster as well as human cells, a superinduction of these enzymes resulting in extended longevity;5 (2) heat shock induces the transcription of heat shock proteins (HSP) in all eukaryotes, in which they are responsible for the refolding of heat-denatured proteins;4 (3) PARP (a DNA repair gene) has a specific enzymatic activity proportional to life span in a variety of species, from insects to primates;6 (4) H-Ras is an intracellular signaling molecule involved in the recovery from radiation-induced stress (UV) of S. cerevisiae, very similar to what occurs in human fibroblasts. Moreover, the human H-Ras gene can modulate yeast longevity.3 In conclusion, a number of ancestral, evolutionary-conserved molecular pathways responsible for resistance to a variety of stressors are involved in longevity from yeast to humans (FIG.1).

AN ADDITIONAL, POWERFUL ANTI-STRESS MECHANISM: THE IMMUNE SYSTEM

As the evolutionary key to understanding the aging process has been successfully identified in the mechanisms of stress resistance, we propose to go further in the equation stress resistance = ability to survive. In particular, we refer to the acquisi- tion of specialized functions for the defense of the organism, which occurred in invertebrates, where some cells (immunocytes, responsible for chemotaxis and FRANCESCHI et al.: INFLAMM-AGING 247

FIGURE 2. The equivalence between antigens and stressors (inspired by Ottaviani et al., 1991, with modifications). From invertebrates to humans, the cellular response to a number of stressors appears to be highly maintained and involves the upregulation of a variety of evolutionary-conserved mediators, such as oxygen-free radicals, nitric oxide (NO), proinflammatory cytokines (IL-1, IL-6, TNFα), propiomelanocortin-derived peptides (ACTH, β- endorphin, α-MSH), steroids (cortisol), biogenic amines (noradrenaline, adrenaline, dopamine), and neuropeptides (CRH). From this evolutionary perspective the equivalence between stressors and antigens emerges, and the stimuli that activate the immune system are nothing other than a particular type of stressor. phagocytosis) became able to fight against a variety of external stimuli and gave rise to a complex, stereotyped, and molecularly conserved set of mechanisms, termed “prototype stress response” (FIG. 2).7,8 Thus, the response to stress became more complex in invertebrates and involved other mechanisms, such as inflammation and innate immunity, sharing ancestral molecular pathways with the response to non- antigenic stressors.9,10 A peculiar cell type, the macrophage, emerged with a central role in this scenario.11,12 As illustrated in FIGURE 2, macrophages, that is, cells with phagocytic activity, can be stimulated and/or activated by bacterial products, neuropeptides, neurohormones, cytokines, and other stimuli and can release large amounts of proinflammatory cytokines, nitric oxide, biogenic amines, neuropeptides, hormones, among others.13 This occurs in both invertebrates and vertebrates. Thus, we propose to return macrophage to its rightful place as central actor, at the 248 ANNALS NEW YORK ACADEMY OF SCIENCES

FIGURE 3. The central role of the macrophage (inspired by Ottaviani and Franceschi, 1997, with modifications) in innate immunity, stress response, and inflammation is illustrated. In invertebrates, a single cell is responsible for all these integrated responses, collective- ly called “prototype stress response.” In higher vertebrates, such as humans, a complex “immuno-neuro-endocrine suprasystem” is the player for this role. cellular level, not only of the inflammatory response and immunity, as originally suggested by Metchnikoff, but also of the stress response.11 The march of complex- ity of the immune system reaches greater sophistication in vertebrates, particularly mammals, with the emergence of clonotypic immunity. It is conceivable that the immune system (or better the immuno-neuro-endocrine system) of humans is the ultimate result of the selection for fitness impinging upon an integrated network of mediators and cells, which allows the individual to respond to environmental changes and stressors.14 In this perspective, most of the biomolecular features of aging can be considered the result of an adaptive capacity at the organismal level, due to a co- operating mixture of old, conserved (macrophage-centered) and new, evolutionary recent (lymphocyte-centered) immune and anti-stress mechanisms (FIG. 3).14

PROINFLAMMATORY STATUS OF AGING (INFLAMM-AGING) AS A RESULT OF CHRONIC ACTIVATION OF THE MACROPHAGE WITH AGE (MACROPH-AGING)

Current data are compatible with the conceptualization of aging as the result of chronic stress impinging upon the macrophage as one of the major target cells in this process.14,15 This hypothesis, illustrated in FIGURE 4, suggests the existence of a di- rect relation between age and macrophage activation, mostly responsible for the presence of a subclinical chronic inflammatory process in the elderly. This phenom- FRANCESCHI et al.: INFLAMM-AGING 249

FIGURE 4. Inflamm-aging as a consequence of macroph-aging. The increase in proinflammatory status at an organismal level, caused by chronic age-related stimulation of the macrophage, called “macroph-aging,” is referred to as “inflamm-aging.” enon is only part of the whole spectrum of change characteristic of immunosenescence, and indeed the macrophage is not the only cell involved in the aging process. Lymphocytes are also largely affected during immunosenescence, and the continuous age-related, largely inescapable, antigenic stress provokes a variety of changes even in the most evolutionary recent, clonotypical immune system. The results were illustrated elsewhere14 and include the expansion of memory cells, the decrease and even the exhaustion of naive cells, the shrinkage of the T-cell repertoire, and the global reduction of the “immunological space.”14 On the whole, data on immunosenescence indicate that changes occurring over time might be considered the result of global reshaping, where the immune system continuously looks for possible stable points for optimal functioning.16 Conceivably, this phenomenon is more general, likely involving the stress response in every tissue and organ. To explain a variety of experimental data concerning the pathophysiology of immunosenescence, we propose an oversimplified but heuristically powerful framework, according to which the effects of stress (including antigenic stress) accumulate, the capability of reshaping decreases, and the inflammatory status becomes pervasive over time. As Hans Selye had suggested since the beginning of his pioneering research, the biological effects of stress, defined as a general adaptation syndrome without any connotative (positive or negative) value, depend on the intensity of the stressor and the individual capacity for coping.10 Usually, the effects of continuous exposure of living organisms to minimal stress have been neglected, whereas the effects of strong stress have been implicitly assumed as the only ones deserving attention and having biological consequences. We can therefore understand the unexpected finding that stress, particularly minimal stress, can increase survival in a variety of organisms and cell types according to the general theory of hormesis, that is, the beneficial effects of extremely low doses of agents that are otherwise toxic at higher doses.17,18 In the usual environment, hormetic and detrimental effects of stress occur concomi- tantly in the same individual, and the outcome is the result of a balance between these two components. On the basis of these considerations, it is possible to hypothesize that a high individual capability of coping can systematically shift the effect of stress towards hormesis. On the contrary, a low individual capability of coping is a driving force towards the detrimental effects of stress. These general considerations 250 ANNALS NEW YORK ACADEMY OF SCIENCES

FIGURE 5. The individual thresholds of inflamm-aging. An individual threshold of the capability to cope with stress is hypothesized. If inflamm-aging trespasses on this level, the tran- sition between successful (continuous line) and unsuccessful aging (dotted line) will occur. In accordance with epidemiologic data, the period of life during unsuccessful aging (disability) is maximal in the elderly (60–80 years) and minimal in young people and centenarians. on stress and stress response can help in understanding a major paradox of aging and longevity. A comparison between data on inflammatory parameters in elderly affected by a variety of diseases and in healthy centenarians showed that an ever-increas- ing inflammatory status is shared by both groups.19–22 Thus, the proinflammatory status is seemingly a characteristic of both successful and unsuccessful aging. To explain this apparent contradiction, we hypothesize that a threshold exists beyond which the adverse effects of stress become evident and drive the organism towards unsuccessful aging and death. This threshold varies among individuals and with age. FIGURE 5 suggests that to reach the pathological threshold, the proinflammatory process in young adults must be characterized by a high rate, but the period of disability before death will be short. In the elderly, this rate is slower and the period in which they will suffer disability before death will be quite long. In centenarians, we envis- age a situation that is similar to that of young adults, but the time needed to reach the pathological threshold is extremely long. These hypotheses fit with recent data in the literature indicating that centenarians have lived in good shape and without disability until very old age,24 despite the fact that in most of them the biochemical parameters related to inflammation can reach values that can be high.19,25

THE TWO HITS THEORY OF INFLAMM-AGING: AN INTERPLAY BETWEEN GENETIC AND ENVIRONMENTAL COMPONENTS

The individual capacity to cope with and adapt to stressors can be assumed as a complex trait having a genetic component. From this point of view, people can be subdivided into “robust” or “frail” according to the combination of gene variants that are involved in the variance of the trait.26,27 We also assume that the rate of reaching the threshold of proinflammatory status over which disease/disability ensues has a genetic component, likely partially distinct from that responsible for frailty or ro- FRANCESCHI et al.: INFLAMM-AGING 251

FIGURE 6. The two hits hypothesis of inflamm-aging. The persistence of inflamm-aging over time represents the first hit (inflammatory background) favoring the susceptibility to age- related diseases. A second hit is necessary in order to develop overt age-related diseases and disabilities, such as atherosclerosis, Alzheimer's dementia, osteoarthritis, and diabetes, currently considered organ-specific inflammatory diseases. The second hit can be identified in the absence of robust gene variants and/or the presence of frail gene variants. bustness. In other words, we postulate that in each individual the rate of reaching the threshold of proinflammatory status can vary within a genetically constrained range, depending on the hormetic or detrimental intensity of the different lifelong (environmental) stressors to which he will be exposed. Environment, and particularly hormetic stressors, can therefore play a major role in extending longevity by increas- ing/inducing the efficiency of the anti-stress network. In any case, the interaction of these environmental factors with the genetic makeup of a given individual (or species) represents a biological constraint for the extension of longevity and for the capacity of stress to modulate the aging rate. The general situation that characterizes aging is illustrated in FIGURE 6, which tries to explain the progressive shift from successful to unsuccessful aging, assuming that the proinflammatory status is largely physiologic. Indeed, as first suggested by Metchnikoff,10,11 inflammation is an evolutionary-conserved “positive” phenomenon, enabling the body to react to and neu- tralize foreign damaging agents. It can be speculated that the capability to mount a strong inflammatory process can contribute to fitness and survival and that people characterized by such a capacity have been positively selected for. In our two hits theory, we argue that inflammatory stimuli persist over time and inflammatory reac- tions add up, eventually representing a biological background (first hit) favoring the susceptibility to diseases. A second hit is necessary to develop overt age-related diseases and disabilities, which can be identified in the absence of robust gene variants and/or the presence of frail gene variants, both accounting for different disability/ mortality thresholds in different individuals. The development of most age-related diseases can have two different types of genetic components besides environmental effects. The first component refers to the possible heritability of the mechanisms re- 252 ANNALS NEW YORK ACADEMY OF SCIENCES sponsible for the first hit (inflammatory background), while the second concerns the genes conferring robustness or frailty at either a local or an organismal level. In particular, we predict that the concomitant presence of a high inflammatory status plus the presence of inherited variants of frailty genes or the absence of robust genes can explain why some individuals, but not others, are more susceptible to a certain type of age-related disease. Because in most cases such genes express their pathological effects in specific cells and organs, this hypothesis could help to explain the onset of a particular pathological condition, such as atherosclerosis, Alzheimer’s dementia, and diabetes, currently considered organ-specific inflammatory diseases.28,29

INFLAMM-AGING AND THE CHALLENGE OF THE IL-6 PUZZLE AND OTHER PARADOXES IN CENTENARIANS

Most of the foregoing arguments apply to what has been defined as the “cytokine for gerontologists,” that is, interleukin-6 (IL-6).30 The plasma levels of IL-6 are low or undetectable in most young people and start to increase in healthy people at about 50–60 years of age. Accordingly, stimulated peripheral blood lymphocytes (PBL) from aged people produce higher levels of IL-6 than do PBL from young subjects.31 The well established increase with age of IL-6 plasma levels appears to be unexpect- edly present in both persons who enjoyed successful aging and those who suffered pathological aging. This increase continues with age, until the extreme limit of human life, and high levels of IL-6 are found in a high percentage of centenarians in good shape.19 In these subjects other proteins, such as acute phase proteins, lipoprotein a (Lp(a)), fibrinogen and other coagulation factors, and proinflammatory cytokines are similarly increased (FIG. 7).19,21,23,25 In sharp contrast, high levels of IL-6 have been referred to as the most powerful predictors of morbidity and mortality in the elderly.32,33 This situation is either a challenging paradox or more simply what should be expected in the last phase of life in extremely old persons, whose proba- bility of dying is very high. Indeed, we can assume that the increase in IL-6 with age is the consequence of the successful adaptation to a number of stresses, including infections, which unceasingly occur throughout life. This process occurs continuously, and in centenarians inflamm-aging eventually reaches levels very close to the threshold of morbidity and mortality, which indeed occur within months or a few years in

FIGURE 7. The paradox of the proinflammatory status of healthy centenarians. FRANCESCHI et al.: INFLAMM-AGING 253 these subjects. Incidentally, despite the high sensitivity of IL-6 plasma levels to acute and chronic infections, as well as to other environmental conditions, strong genetic control of IL-6 plasma levels is emerging.34 This latter finding, while confirm- ing our hypothesis of a genetic component of the first hit, strongly indicates that future breakthroughs depend on genetic research concerning inflamm-aging. Thus, inflamm-aging could be the ultimate proof that the beneficial effect of the defense system network (innate immunity, stress, and inflammation), devoted to the neutralization of dangerous/harmful agents early in life and in adulthood, turns out to be detrimental late in life, in a period largely not foreseen by evolution. Thus, a tradeoff between early beneficial effects and late negative outcomes can occur at the genetic and molecular level. Similar phenomena have been predicted by evolutionary genet- icists who proposed the theory of antagonistic pleiotropy.

ACKNOWLEDGMENTS

We acknowledge E.U. “GENAGE,” M.U.R.S.T. 40%, M.U.R.S.T. Project “Genetic determinants of human longevity,” and the Ministry of Health Projects project, “The prevention of chronic age-related diseases: the model of centenarians,” for support.

REFERENCES

1. FRANCESCHI, C. 1989. Cell proliferation, cell death and aging. Aging. Clin. Exp. Res. 1: 3–15. 2. KIRKWOOD, T.B.L. & C. FRANCESCHI. 1992. Is ageing as complex as it would appear? New perspectives in gerontological research. Ann. N.Y. Acad. Sci. 663: 412–417. 3. JAZWINSKI, S.M. 1999. Molecular mechanisms of yeast longevity. Trends Microbiol. 7: 247–252. 4. MURAKAMI, S. & T.E. JOHNSON. 1998. Life extension and stress resistance in Cae- norhabditis elegans modulated by the tkr-1 gene. Curr. Biol. 8: 1091–1094. 5. SOHAL, R.S. & R. WEINDRUCH. 1996. Oxidative stress, caloric restriction, and aging. Science 273: 59–63. 6. BURKLE, A.1998. Poly(ADP-ribose) polymerase and aging. Exp. Gerontol. 33: 519–523. 7. OTTAVIANI, E., E. CASELGRANDI, M. BONDI et al. 1991. The “immuno-mobile” brain: evolutionary evidence. Adv. Neuroimmunol. 1: 27–39. 8. OTTAVIANI, E., A. COSSARIZZA, C. ORTOLANI et al. 1991. ACTH-like molecules in gas- tropod molluscs: a possible role in ancestral immune response and stress. Proc. Roy. Soc. Ser. B 245: 215–218. 9. OTTAVIANI, E., S. VALENSIN & C. FRANCESCHI. 1998. The neuro-immunological inter- face in an evolutionary perspective: the dynamic relationship between effector and recognition systems. Front. Biosci. 15: D431–D438. 10. OTTAVIANI, E. & C. FRANCESCHI. 1996. The neuroimmunology of stress from invertebrates to man. Progr. Neurobiol. 48: 421–440. 11. OTTAVIANI, E. & C. FRANCESCHI. 1997. The invertebrate phagocytic immunocyte: clues to a common evolution of the immune and the endocrine systems. Immunol. Today. 18: 169–174. 12. OTTAVIANI, E. & C. FRANCESCHI. 1998. A new theory on the common evolutionary ori- gin of natural immunity, inflammation and stress response: the invertebrate phagocytic immunocyte as an eye-witness. Domest. Animal Endocrinol. 15: 291–296. 13. OTTAVIANI, E., A. FRANCHINI & C. FRANCESCHI. 1997. Pro-opiomelanocortin-derived peptides, cytokines and nitric oxide in immune responses and stress: an evolutionary approach. Int. Rev. Cytol. 170: 79–142. 254 ANNALS NEW YORK ACADEMY OF SCIENCES

14. FRANCESCHI, C., S. VALENSIN, F. FAGNONI et al. 1999. Biomarkers of immunosenescence: the challenge of heterogeneity and the role of antigenic load. Exp Gerontol. 34: 911–921. 15. FRANCESCHI, C., M. BONAFÈ & S. VALENSIN. 2000. Human immunosenescence: the prevailing of innate immunity, the failing of clonotypic immunity, and the filling of immunological space. Vaccine 18: 1717–1720. 16. FRANCESCHI, C., D. MONTI, P. SANSONI & A. COSSARIZZA. 1995. The immunology of exceptional individuals: the lesson of centenarians. Immunol. Today 16: 12–16. 17. JOHNSON, T.E. & H. BRUUNSGAARD. 1998. Implications of hormesis for biomedical aging research. Hum. Exp. Toxicol. 17: 263–265. 18. CALABRESE, E.J. & L.A. BALDWIN. 1999. The marginalization of hormesis. Toxicol. Pathol. 27: 187–194. 19. BAGGIO, G., S. DONAZZAN, D. MONTI et al. 1998. Lipoprotein(a) and lipoprotein pro- file in healthy centenarians: a reappraisal of vascular risk factors. FASEB J. 12: 433– 437. 20. CHIRICOLO, M., G. BARTOLINI, M. ORLANDI et al. 1986. Prostaglandin and tromboxane biosynthesis in resting and activated platelet-free monocytes from aged subjects. Gerontology 32: 69–73. 21. MARI, D., P.M. MANNUCCI, R. COPPOLA et al. 1995. Hypercoagulability in centenarians: the paradox of successful aging. Blood 85: 3144–3149. 22. MARI, D., P.M. MANNUCCI, F. DUCA et al. 1996. Mutant factor V (Arg506Gln) in healthy centenarians. Lancet 347: 1044. 23. MANNUCCI, P.M., D. MARI, G. MERATI et al. 1997. Gene polymorphisms predicting high plasma levels of coagulation and fibrinolysis proteins. A study in centenarians. Arterioscler. Thromb. Vasc. Biol. 17: 755–759. 24. HITT, R., Y. YOUNG-XU, M. SILVER & T. PERLS. 1999. Centenarians: the older you get, the healthier you have been. Lancet 354: 652. 25. BRUUNSGAARD, H., K. ANDERSEN-RANBERG, B. JEUNE et al. 1999. A high plasma con- centration of TNF-alpha is associated with dementia in centenarians. J. Gerontol. A. Biol. Sci. Med. Sci. 54: M357–M364. 26. YASHIN, A., J.W. VAUPEL, K.F. ANDREEV et al. 1998. Combining genetic and demographic information in population studies of aging and longevity. J. Epidemiol. Bio- statist. 3: 289–294. 27. YASHIN, A.I., G. DE BENEDICTIS, J.W. VAUPEL et al. 1999. Genes, demography and life span: the contribution of demographic data in genetic studies on aging and longevity. Am. J. Human Genet. 65: 1178–1193. 28. WEISSBERG, P.L. & M.R. BENNETT. 1999. Atherosclerosis--an inflammatory disease. N. Engl. J. Med. 340: 1928–1929. 29. EIKELENBOOM, P. & R. VEERHUIS. 1999. The importance of inflammatory mechanisms for the development of Alzheimer's disease. Exp. Gerontol. 34: 453–461. 30. ERSHLER, W.B. 1993. Interleukin-6: a cytokine for gerontologists. J. Am. Geriatr. Soc. 41: 176–181. 31. FAGIOLO, U., A. COSSARIZZA, E. SCALA et al. 1993. Increased cytokine production in mononuclear cells of healthy elderly people. Eur. J. Immunol. 23: 2375–2378. 32. HARRIS, T.B., L. FERRUCCI, R.P. TRACY et al. 1999. Associations of elevated interleukin-6 and C-reactive protein levels with mortality in the elderly. Am. J. Med. 106: 506–512. 33. FERRUCCI, L., T.B. HARRIS, J.M. GURALNIK et al. 1999. Serum IL-6 level and the development of disability in older persons. J. Am. Geriatr. Soc. 47: 639–646. 34. FISHMAN, D., G. FAULDS, R. JEFFERY et al. 1998. The effect of novel polymorphisms in the interleukin-6 (IL-6) gene on IL-6 transcription and plasma IL-6 levels, and an asso- ciation with systemic-onset juvenile chronic arthritis. J. Clin. Invest. 102: 1369–1376.