PDFlib PLOP: PDF Linearization, Optimization, Protection Page inserted by evaluation version www.pdflib.com – [email protected]

PART I . BASIC AGING PROCESSES

Human Longevity and Aging: Possible Role of Reactive Oxygen Species RICHARD G. CUTLER Gerontology Research Center National Institute on Aging 4940 Eastern Avenue Baltimore, Maryland 21224

INTRODUCTION The number of elderly people in the developed nations of the world is growing at an ever increasing rate. This is evident both in absolute number as well as percentage with reference to total population. The major concern, however, appears to be the negative economic impact such a change in population composition would have on a country. This is because of the steadily increasing cost of medical care of an aging population, particularly for those individuals over 80 years of age, which is the most rapidly growing fraction. There is consequently much interest in gerontology and in exploring a wide range of strategies that might be effective in reducing the cost of a growing aged population.’S2 Although the economic burden of the elderly population is clearly evident and is usually most emphasized when discussing the subject of aging and geriatric medicine, I would like to point out that an often overlooked and perhaps even greater cost to society of human aging is the steady decline in many physiological and mental processes that appear to occur shortly after the age of 30 years in the apparent absence of major disease or expensive medical care needs. In this regard, a decrease of productivity and creativity may occur in most people over the some 30 to 40 years of steadily declining health over the normal human life span. We often forget that aging occurs in most individuals without the necessary accompaniment of serious disease, which usually comes much later in life. Thus, the economic as well as the humanist benefits in reducing aging rate may go far beyond the predicted savings in medical cost of just the elderly p ~ p u l a t i o n . ~ In this light, a long-term goal in biogerontology should not only be to reduce the medical costs of the growing elderly population but also to increase the healthy and productive years at all ages of our life span.3 In this regard, we frequently hear from medically-oriented gerontologists that they seek to add “more life to our years and not more years to our life.” I would only partially agree with this statement and believe our goal should be more ambitious. That is, to not only add more life to our years but also more years to our life. Indeed, it is probably impossible to simply add more life to our years without also adding more years to our life. It is likely to become more evident in the near future that the only long-term satisfactory method of reducing the medical costs of the elderly will be through a direct intervention of the aging process itself. That is, the development of intervention methods aimed more directly towards treatment of the causes of aging and not simply the effects of aging, as has been the traditional medical approach in the past. This goal of reducing the overall aging rate in humans runs immediately into the problem of the vast complexity of normal human biology and what little 1

2

ANNALS NEW YORK ACADEMY OF SCIENCES

knowledge we presently have in this area of science. In fact, it is usually argued that the mechanisms, whatever they might be, that cause aging are likely to be so complex that we can only hope to begin to understand how to control aging rate after much more is learned of normal biology. Another negative argument is that aging is likely to have many complex and multiple causes and so it is foolish to believe that relatively few genetic or biochemical means of intervention would be effective or could have any significant impact on decreasing human aging rate. This all may be true, but the arguments given are not too convincing. The fact is that the problem of the complexity of the aging process and of the possibility of developing intervention strategies to increase the healthy years of life span and productivity by reducing human aging rate has simply not received serious scientific attention or evaluation. My objective in this paper is to briefly present an alternative argument indicating that, in spite of the vast complexity of aging, the processes governing aging rate or life span may be much less complex and therefore more subject to intervention in the near future. The basis of this prediction comes largely from evolutionary and comparative studies of mammalian species closely related to one another biologically and evolutionarily but having substantial differences in their life span and aging rate. These studies led to the formulation of the ‘longevity determinant gene hypothesis’&’ which predicts that aging is a result of normal biological processes necessary for life but also having long-term negative or aging effects on the organism. These normal biological processes have been divided into two major categories: (a) developmentally-linked biosenescent processes and (b) the continually-acting biosenescent processes. According to this hypothesis, longevity of a species is related to how efficiently the aging effects of these common developmental and metabolic processes they all share have been reduced. I shall not deal with the processes in the first category but shall note that they include many of the hormones associated with development. For example, most pituitary, thyroid, adrenocorticoid, ovarian and testicular hormones have been found to have long-term aging effects.8.8aAn important exception may be growth hormone, which does appear to have many rejuvenative effects when administered to old experimental animals.R However, long-term studies of growth hormone administration have not yet been conducted to determine possible side effects o r degree of life span extension, if any.9 Our research has focused on the second category, the continuously-acting biosenescent processes, in testing the longevity determinant gene hypothesis. Here, we have examined the possible aging effects of oxygen metabolism, which is essential for life but is also known to produce what are called ‘reactive oxygen species’ during normal energy production. Such reactive oxygen species can interact with the cell’s genetic apparatus and alter its proper state of differentiation. These changes in differentiation, as they have been predicted to occur during normal aging, have been called dysdifferentiation and could explain many aspects of the normal aging process. The unique feature of the longevity determinant gene hypothesis is that it suggests that all animal species share common aging causes and common mechanisms of regulating aging rate. Although the processes causing aging may be very complex, it is predicted that less complex key antiaging processes exist that govern aging rate or life pan.^.^?''' For example, human and chimpanzee are closely related to one another, both evolutionarily (about 3-5 million years from a common ancestor) and biologically (DNA sequences are about 98% common). Iod Because of the remarkable biological similarities between human and chimpanzee, we would expect the aging processes to be also similar qualitatively and, as far as we know, they are. Yet, in spite of these remarkable biological similarities,

CUTLER: REACTIVE OXYGEN SPECIES

3

the chimpanzee has a life span about one half that of human (50 yrs vs 100 yrs) and accordingly appears to age about twice as fast in all biological aspects and exhibits the same age-dependent diseases but occurring in half the time. So the central question we ask is what is so unique about human biology as compared to the chimpanzee or other shorter-lived animals that allows human to be so long-lived? Because of the strikingly similar biology between human and chimpanzee as well as their close evolutionary relation to one another, it is reasonable to predict that the genetic and biochemical differences governing their life span differences are also not likely to be very great. Thus, we ask, how complex biologically speaking are the processes governing human aging rate? And is it possible from a comparative biological study of animal species closely related to human, but having different life spans, to learn of new mechanisms of how to further enhance human health, vigor and productivity for longer periods of time? According to this approach, we should place equal effort on understanding the mechanisms governing aging rate and health maintenance as well as on understanding the mechanisms causing aging and disease.

FIGURE 1. Percent survival curve for hu-

mans under different environmental hazard conditions. Note different 50% mean values but constant maximum life span potential I (MLSP) values of about 100 yrs. Data taken in part from Comfort,” Acsidi and Ne- g meresk&ri,!*and Strehler.I3 (From Cutler.’ Reprinted by permission from Plenum

8o

~

Press.) 20

60

Biological Nature of Human Aging and Longevity Human life expectancy (average survival time) for most of the time Homo supiens has been on this earth (-l00,000 yrs) is generally believed to have been about 20 to 30 y e a r ~ ~ *(FIG. . l ~ . 1). ~ ~Only recently over the last 400 years or so in developed nations has life span expectancy increased, and this increase has been rather dramatic (about 40 to 50 years).’* The reasons causing this increase certainly include reducing infant mortality and environmental hazards, accompanied by better nutrition and healthcare, but these traditional explanations may not be the complete answer.14 The important point to be made here, however, is that in primitive cultures where life expectancy was about 30 years, few people lived long enough to suffer appreciatively from the processes of senescence or aging. Death was almost entirely due to environmental hazards and infectious diseases not aging. Today in developed countries, although people have life expectancies in the range of 70 to 80 years, they are not living younger longer but are instead actually living older longer. Now most people die of problems largely related to the processes of aging and not to environmental hazards. This is because the increased life expectancy occurred with aging rate remaining essentially unchanged. Thus, the increase of life expectancy did not occur as a result of slowing

4

ANNALS NEW YORK ACADEMY OF SCIENCES

down normal aging processes but instead by reducing the major external environmental hazards to life. This means that the present day problem of an older population in developed nations is actually an artifact of our civilization and not a natural state of human populations. Some theoretical models of aging predict that normal human aging would result in a steady decline in functional capacity of essentially every physiological and mental process, beginning shortly after the age of sexual m a t u r a t i ~ n " , ' ~ * ~ ~ , ' ~ (FIG.2). If this is true, then the slope of the decrease in maximum capacity to function with age could be taken to generally reflect an average or overall aging rate of the body. A fundamental question of biological research in gerontology today is to understand the molecular and biochemical mechanisms leading to this age-dependent decrease in function. What is unique in our laboratory research program is that we have also asked what determines the slope or rate of loss of these physiological functions. With increasing age there also occurs an increased incidence of disease, which finally results in the death of the individual. An example of such a disease is cancer, which is shown in FIG. 3, indicating the well-known age-dependent increased incidence of cancer.I5 Since we do not yet understand even one mechanism causing aging, it is of course difficult to clearly define what aging actually is, and as a result there is some confusion in separating decline in physiological function as being due to a primary aging process or due to disease both related to and not related to primary aging processes. That is, many diseases could be a result of aging or even a direct manifestation of aging or not related to aging at all but simply related to chronological age. To illustrate this, there is some evidence indicating that the causes of aging and cancer are similar, being based on genetic instability mechanisms. On the other hand, there is much evidence indicating that causes of cardiovascular disease are related to nutrition, which may have little effect in accelerating aging rate. There is also likely to be a positive feedback interaction between aging and diseases of aging, and so both phenomena are likely

Age in Years

FIGURE 2. Change in organ function with age. (From Bafitis & Sargent." Reprinted by permission from the Journal of Gerontology.)

CUTLER. REACTlVE OXYGEN SPECIES

5

L

2x0-

FIGURE 3. Age-specific death rates from several neoplasms: colon, lung, and leukernia per 106 people and for bone and kidney per lo7people. (Taken in part from K o h n . 9

B

6 W

im

-

8oo

-

4w

-

$

to play an important role in determining the overall health status of individuals as they grow older. It is also expected theoretically that there should be great uniformity in both the qualitative and quantitative nature of aging between different individuals and in the biological functions that are affected. With increasing age, essentially all body functions in every person are found to decline uniformly with increasing chronological age. Persons that are unusually long lived and are able to live past 100 years do so not apparently because they age more slowly but instead because they age more uniformly. That is, they appear not to suffer from any particular .'~~ weak link in their body functions such as from heart disease or d i a b e t e ~ . ' ~ In addition, the well-known rectangularization of the human survival curve over the past 400 years or so in developed countries is the result of reducing the random components of survival, leaving largely the nonrandom factor. These nonrandom factors are thought to consist essentially of aging processes. Jffound to be correct, such information could be important in the development of effective means to increase the healthy years of life span, as is illustrated in TABLE 1. In this table, the theoretical point is made that, even if the major diseases causing death today could somehow be completely eliminated, the impact on increased life expectancy over the entire country is not nearly so great as one might believe. For example, elimination of all cancer results in an increase of only about 2 years of life expectancy. The reason for this result may be that most people suffering from the major killer diseases today are 65 years of age or older. Thus, the removal of a major disease causing death of the elderly would simply result in uncovering a new disease or health problem since essentially every aspect of body function is being reduced through the aging process. Such studies indicate that significant increase in the healthy years of life span of the general population of a nation is only likely to be achieved by actually reducing uniformly the rate of aging of the entire body, not by a piecemeal approach of reducing or eliminating specific disease processes. 15b Thus, efforts being made in the field of cancer or even Alzheimer's disease may not achieve dramatic reduction in medical costs if effective therapy simply uncovers new disease processes characteristic of all people as they grow older.

6

ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 1. Gain in Expectancy of Life at Birth and at Age 65 Due to Elimination of Various Causes of Death"

Gain in Expectancy of Life (Yrs) if Cause Was Eliminated Cause of Death

at Birth

at Age 65

Major cardiovascular-renal Heart disease Vascular diseases affecting central nervous system Malignant neoplasms Accidents other than by motor vehicle Motor vehicle accidents Influenza and pneumonia Infectious diseases (excluding tuberculosis) Diabetes mellitus Tuberculosis

10.9 5.9

10.0 4.9 1.2

I .3

1.2 0.1

2.3 0.6

0.6 0.5 0.2

0.1

0.2 0.1

0.2

0.2

0.1

0.0

From life tables published by the National Center of Health Statistics.Is

The Longevity Determinant Gene Hypothesis If significant gain to reduce the high costs of human aging to society does indeed require a uniform reduction in human aging rate, then how might this objective be achieved? To begin with, it is first necessary to estimate how complex the problem is that needs to be solved. That is, in-depth studies are seriously needed, for example, to determine the processes governing aging rate and how many genes might be involved. There is unfortunately very little data in this area but a few studies based on an evolutionary comparison of species closely related to one another but having differences in aging rate have suggested that key longevity determinant processes may exista5 The basis of such studies is the fact that different mammalian species do indeed have different aging rates and that they age qualitatively in a similar way (TABLE2). In addition, estimates have been made as to how fast life span has Most Confident Relative Maximum Life Span Potential and Life Span Energy Potential Estimates for Primatesa

TABLE 2.

Species (Common Name)

MLSP (Yrs)

Human Great apes: chimp, orangutan, gorilla Macaca, baboon Capuchin Marmoset, tamarin, squirrel monkey Tree shrew

100

850

50

450 500 800

40 40 20 15

LEP (kcal/g)

600 500

Rounded-off estimates of MLSP and LEP, showing relative MLSP and LEP values for species where this data is most reliable. There appears to be a 5-fold difference in MLSP and a 2-fold difference in LEP among the primate species.

LEP

421 SW so I

3%

25

SIC

I3

taken from literature values for MLSP and specific metabolic rate (SMR). (From Cutler.I9Reprinted by permission from the American Physiological Society.)

FIGURE 4. Phylogenetic relationship of MLSP and life span energy potential (LEP) estimates for the primate species. Data represent estimates

311

so0 .I*

I7

SS?

IS IS

tl

I5 20

la

IS

20

30 30 10

*2

3aa

2s 22 30

3s. 331

Sl?

35

3s 13

3.

Ihc/d

~

8

ANNALS NEW YORK ACADEMY OF SCIENCES

evolved along the hominid ancestral-descendant sequence leading to the Homo supiens In these studies life span was found to increase at a maximum rate of about 14 years per 100,000 years about 100,000 years ago5 (FIGS.4-6). Such a high rate suggests that alterations in gene sequence are not likely to be involved in this rapid evolution of life span. Instead, these data are more consistent with the concept that changes in gene regulatory processes occurred during hominid evolution, resulting in a uniform decrease in aging rate.5J0 These evolutionary comparative studies have led to the proposal that key longevity determinant genes of a regulatory nature might exist that are capable of governing the aging rate of the entire organism. This concept is in contrast to what has been generally believed,19bwhere aging is thought to be a result of biological

Divergence of Human Races

[

i

9 4 H. sopiens modern

85 M sopiens Aunt. obor.

78 M erecius pertinensis 6 9 M erectus jovonicus 60 H. hobilis Gorillo Pon Po1190

-

R punjobicus Hylobatidae CERCOPITHECOIDEA ( I8 -40) 22 CEBOIDEA (15-40)

6.5 TUPAIOIDEA (5.5 -7) 0

7: homuncutus 1

60

1

1

40

1

1

36

l

1

24

1

1

1

12

i 1

1

0

Millions of years before present

FIGURE 5. Evolution of maximum life span potential for the Anthropoidea. 0, fossil data; living species. (From C ~ t l e r Reprinted .~ by permission from the Journal of Human

*,

Evolution.)

functions at least as equally complex as the organism itself and that life span or aging rate is determined not by key longevity determinant genes but by thousands of genes operating by highly complex mechanisms unique for each cell and/or tissue of the organism. Another deduction of the longevity determinant gene hypothesis is that aging is not genetically programmed for some evohtionary survival benefit but instead is the result of normal biological processes essential for life to exist.19cTABLE3 shows the major classes of normal metabolic processes thought to be important as primary causes of aging. Since all mammalian species have remarkably similar biological processes (particularly chimpanzee and human), the question then arises of what normal biological processes are most responsible for causing aging

9

CUTLER: REACTIVE OXYGEN SPECIES

I " " " " ' " " " '

-

,g -.-

90-

H. europoeus pre-Wurm -

Y

0

EW

c 0

n

BO-

70-

r a

c -

0

-

f

60-

.-E x

= 0

30-

I

I I.4

I

I

,

1.2

1.0

0.8

0.6

0.4

0.2

0

Millions of years before present

FIGURE 6. Rate of increase in maximum life span potential along the hominid ancestraldescendant sequence leading to modern man. 0, MLSP; 0 , rate of change in MLSP per million years. (From C ~ t l e r Reprinted .~ by permission from the Journal of Human Evolution.)

and what key mechanisms evolved to control the rate of their expression and consequently the longevity of the organism? To answer this question, we began a comparison of human biology with that of other closely related species as a function of their life span with the hope of discovering small differences that could help explain why humans are so longlived. One such comparative study is shown in FIG.7, showing a plot of life span vs specific metabolic rate. These data suggest that aging rate of different species may be related to their metabolic rate? Thus, it appears that by-products of oxygen metabolism may have a role in causing aging,20aThis figure also indicates that most animals consume a constant amount of energy over their life span but that humans are exceptional in consuming about four times the energy over their life span. Thus, we humans are already exceptional in not only having more years of life but also in getting more life out of our years as compared to other mammalian species.

Possible Sources of Products Contributing to the Age-Dependent Destabilization of the Proper Differentiated State of Cells

TABLE 3.

Basic concept: Aging is u reJrrlt of normul developmental ond metabolic processes: 1. By-products of development (growth and sexual hormones) 2. By-products of stress (adrenocorticoids)

3. By-products of metabolism (oxygen metabolim)

ANNALS NEW YORK ACADEMY OF SCIENCES

10

Alricon bhphont

0 Prlrnoles A Arliodoctylo 0 Cornivoro

0 Periswdoclylo A Proboscidao Hydrocoidla 0 Loiomorvho

A Rdaniia lnnctivoro 0 Marsupilia

40

80

120

240

280

SMR (cal/g x day)

FIGURE 7. Life span energy potential of mammalian species. Data are based on specific metabolic rate (cal/g/day) and life span (yrs) where 90% mortality occurs (LS-90) taken from previously published data. Plots indicate three major classes of life span energy potentials (LEP). (From Cutler.zo Reprinted by permission from Gerontology.)

Oxidative Stress as a Primary Aging Process It is well known now that utilization of oxygen represents an efficient mechanism for aerobic organisms to generate energy but, during this process, by-products called reactive oxygen species are also created that could damage a cell.*Ob Indeed, all aerobic organisms require a vast complex network of defense mechaSome Endogenous Biological Systems That Generate the Superoxide Radical 02-

TABLE 4.

1. Enzymes:

Tryptophan dioxygenase Intoleamine dioxygenase Xanthine oxidase Cytochrorne P-450 Peroxidase ( e . g . , during NADP oxidation) Aldehyde oxidase 2. Small molecules: Reduced riboflavin, FMNHz, FADHz Diphenols ( e . g . , adrenalin) Melanin Thiols 3. Cellular organelles: Mitochondnal electron transport chain Microsomal electron transport chain ( e . g . , P-450 detoxification system)

CUTLER REACTIVE OXYGEN SPECIES

11

Some Antioxidant Defense Mechanisms That May Be Longevity Determinants

TABLE 5.

1. Nonenzymatic:

Alpha-tocopherol (membrane-bound) Ascorbate (water-soluble) Beta-carotene (singlet oxygen quencher) Urate (singlet oxygen quencher) Ceruloplasmin (plasma protein) Ubiquinol-10 (membrane-bound) Bilirubin (albumin-bound fatty acids) Ergothioneine (muscle tissue) 2. Enzymatic: Superoxide dismutase (Cu/Zn and Mn types) Glutathione peroxidase (Se and non-Se types) Catalase (peroxisomal matrix) 3. Auxillary enzymes: NADPH-quinone oxidoreductase (two-electron reduction) Epoxide hydrolase (two-electron reduction) UDP-glucoronyltransferase (conjugation enzyme) Sulfotransferase (conjugation enzyme) GSH S-transferase (conjugation enzyme) GSSG reductase Glucose-6-phosphate dehydrogenase (NADPH supply) GSSG export enzymes

nisms to reduce the toxic effects of these by-products of oxidative energy metabolism.*Ob There are many other sources of reactive oxygen species in addition to energy metabolism (TABLE4), and already a number of defense mechanisms have been identified (TABLE5 ) . Some strategies that could be used to reduce the oxidative stress state of a cell and/or organism to increase life span are shown in TABLE6. Many different diseases, some related to aging, appear to be reduced by dietary antioxidant levels, indicating the biological significance of antioxidants in governing the rate of disease incidence (TABLE7). There are also data not widely known indicating that the onset frequency of cancer incidence leading to death These data imply that appears to be related to the aging rate of a species.2k.20d*z1 not only does cancer incidence increase exponentially with chronological age but TABLE 6.

Different Strategies to Lower the Toxic Effects of Active Oxygen

Species 1. Lower average rate of total body oxygen utilization. Increase body size, lower body temperature, torpor, sleep, hibernation. 2. Increase tissue concentration of antioxidants (superoxide dismutase, alpha tocopherol, carotenoids, urate, ceruloplasmin). 3. Decrease tissue concentration of some antioxidants that may have prooxidant properties (ascorbate, glutathione, catalase (Fe++). 4. Decrease in the intensity of metabolic reactions that produce active oxygen species (cytochrome P-450INADPH detoxification reaction). 5 . Increase intrinsic resistance of cellular constituents to damage and peroxidation by active oxygen species (lowering the peroxidative potential of membranes by possible decrease in percent unsaturated fatty acids).

12

ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 7. Chronic Diseases Found to be Related to Deficiencies in Dietary Antioxidants30a 1. Cardiovascular disease 2. Myocardial ischemia/reperfusion injury 3. Cataracts 4. Rheumatoid arthritis 5. Exercise-induced hypoxia reperfusion injury of joints 6. Parkinson’s disease 7. Alzheimer’s disease 8. Cancer (lung, esophagus, gastric/colon, cervical dysplasia)

that the rate of increased incidence is also related to the aging rate of the species19 (FIG.8). This important result suggests that whatever mechanism protects human from cancer so efficiently as compared to other species may also be the same one responsible for the slow human aging rate.2’ Much evidence indicates that cancer is caused by genetic alterations, resulting in an improper differentiated cell. Such genetic alterations could be caused by

mouse 3.5 yrr

squirrel monkey 25 yrs

Rhesus 40 y r r

I

I

I

I

I I I

I

I

I

I

I

I I I

I

I I

I I

I I

I I I

I

I

I I I I

I I

I I

human 100 y r s

I I I I

f

I

I

I

I

chimpanzee 5 5 yrs

I

I I t

f

/

/ /

Age ( y r s )

FIGURE 8. Cumulative net risk of death from cancer as a function of age and maximum life span potential. This family of curves represents typical data describing the onset frequency of all types of cancer with age in different mammalian species. The data are plotted as probability being proportional to age.‘ Curves for mouse and human are most reliable (solid curues) based on the species’ life span and data indicating relative rank order of longer-lived mammals having less cancer incidence with age. (From Cutler and Semsei.*I Reprinted by permission from the Journal of Gerontology.)

CUTLER: REACTIVE OXYGEN SPECIES

13

TABLE 8. Evidence for the Importance of Oxygen Radicals in Genetic Damage,

Disease, and Cancer 1. Dietary carcinogens and anticarcinogens, oxyradicals, and degenerative diseases Many mutagens and carcinogens act through generation of oxygen radicals. Oxygen radicals also play a major role as endogenous initiators of degenerative diseases (cancer, heart disease) through DNA damage, mutation, and promotion effects. B. N. Ames, Science 221: 1256, 1983.22 2. Prooxidant state and tumor promotion There is convincing evidence that active oxygen, peroxides and radicals can promote or initiate cells to neoplastic growth. Prooxidant state can be caused by xenobiotics, metabolites, inhibitors of antioxidant state, and membrane-active agents. P. A. Cerutti, Science 227: 375, 1985.23

reactive oxygen specie^^^*^^ (TABLE8). Thus, a common mechanism which could cause both cancer and aging is reactive oxygen species causing a genetic instability. Such an idea is consistent with the dysdifferentiation hypothesis of aging, where aging is predicted to be a result of cells drifting slowly away from their proper state of differentiati~n.~~ There is considerable evidence supporting dysdifferentiation as a primary cause of aging25(TABLE9) and of reactive oxygen species to be a primary cause of aging (TABLE10). Taking these data together provides more support that aging may be a result of genetic instability leading to dysdifferentiated cells, which finally causes a decline in physiological function and an increase in disease frequency. An important aspect of this hypothesis is that it predicts that life span is a result of factors acting to stabilize the proper differentiated state of cells. That is, with human being the longest-lived mammalian species, this hypothesis would predict that human cells should be the most stable in maintaining their proper state of differentiation. If this is true, then it would be important to know what mechanism acts to stabilize states of differentiation. Essentially nothing is known of such mechanisms acting to stabilize the proper differentiated state of cells once the genetic program of development has been completed. Thus, in light of the longevity determinant hypothesis, this area of research should certainly receive increased attention. If reactive oxygen species do play a role in destabilizing the genetic apparatus of cells, then antioxidants and other defense mechanisms acting to lower cellular oxidative stress may be an important component determining genetic stability. TABLE 9. Some Evidence supporting Dysdifferentiation as a Possible Causative

Factor in Aging I . Cells in tissue culture from longer-lived species appear more stable in terms of spontaneous mutation, greater cell-doubling number to reach a crisis state (ROhmz5)and lower sensitivity to transformation by chemical mutagens. 2. Rate of accumulation of chromosomal aberrations is related to aging rate of the species. 3. Age-dependent accumulation of abnormal cell types (metaplasias). 4. Qualitative and quantitative age-dependent changes in gene expression, structural proteins, and enzyme levels have been found. 5 . Basic housekeeping functions of a cell appear to be adequate in older organisms.

ANNALS NEW YORK ACADEMY OF SCIENCES

14

TABLE 10. Evidence Supporting Active Oxygen Species as a Possible Causative

Factor in Dysdifferentiation 1 . Low concentration of mutagenicicarcinogenic agents induce improper gene regulation, some by induction of transposable elements involved in control of gene expression. 2. Active oxygen species react with chromatin, oxidize specific DNA bases, and cause single-strand DNA breaks. 3. Active oxygen species induce chromosomal aberrations, and longer-lived species have a slower rate of accumulation of chromosomal aberrations. 4. Longer-lived species show a slower rate in accumulation of lipofuscin age pigments. 5 . Aging rate is proportional to metabolic rate (cal/g/day) for many different mammalian species.

There is some evidence supporting the prediction that cells from longer-lived species are more stable in maintaining their proper state of differentiation. As shown in FIG.9, cells from longer-lived species divide more times before reaching a crisis period in cell c ~ l t u r e . These 2~ and other data suggest the simple hypothesis shown in FIGURE10 of how oxygen radicals may cause aging. This model can, however, be expanded into what we have called the multistep model of cancer and 11. The important feature of this model is the possibilaging, as shown in FIGURE ity of taking advantage of much of what has already been learned in cancer research (such as the multistep model of carcinogenesis) and applying it towards an advancement of an integrated model for research in aging and cancer mechanisms. For example, factors governing aging rate could act at the initiation stage (mutagens) or the progressive stage (epigenetic). Thus, dietary factors known to

I

humon

r

= 0.955

mouse I

I

20

I

1

40

I

I

60

1

1

80

I

I

100

MLSP (yrs)

FIGURE 9. Population doubling of cells in culture before a crisis state is reached. (Adapted from Rohme.zs)

15

CUTLER REACTIVE OXYGEN SPECIES

-

utilization of oxygen

-

active oxygen species O;, .OH. ROOH

-

genetic apparatus of cells

-

changes in proper regulation and expression

-

aging pro€8ss

FIGURE 10. Model of how oxygen radicals may cause aging.

be possible causes of cancer (initiators or promotors) may also be acting to accelerate aging rate, and in turn those agents found to protect against cancer (dietary antioxidants) may also be of some benefit in protecting against abnormally high aging rate. A more general model indicating in detail the various areas where reactive oxygen species might play a role in aging is shown in FIG.12. These data briefly reviewed here have suggested the working hypothesis shown in TABLE11 that has guided much of our recent experimental work. Most recently, an extension of the dysdifferentiation model has been proposed. It is now recognized that small changes in the proper differentiated state of cells may have their greatest impact in certain areas of the brain involved in regulating homeostatic Such tissues are represented by the various areas of the pituitary gland and the hypothalamus. Specific areas of the brain such as those noted could be under high oxidative stress and often consist of a relatively few number of cells. Thus, alterations of regulatory genes in regulatory tissues are predicted to be the most important targets in the dysdifferention model of aging. Many tissues of the body are in a constant cell renewal process. Tissues made up of nondividing cells are constantly renewing themselves through endogenous degradation-resynthesis processes and tissues made up of dividing cells are constantly being renewed through cell turnover. Such renewal processes are called

Cancer Initiation

Promotion

Cell proliferation immortality

Propagation Aging

Cell death

FIGURE 11. Multistep model of aging and cancer. This model represents the proposed multistep phenomena of both cancer and aging in their progress to increasing stages of dysdifFerentiation. Ultimate end result of transformed cancer cell is cell proliferation immortality and of the aging process (other than cancer) is cell death. (From Cutler and Semsei.*' Reprinted by permission from the Journal of Gerontology.)

TABLE 11. The Working Hypothesis Being Tested in Recent Experimental Work 1. Aging is in part a result of changes in proper gene expression and regulation. This

process has been called dysdifferentiation. 2. Active oxygen species contribute to the dysdifferentiation process. 3. Mechanisms that act to reduce the dysdifferentiative effects of active oxygen species prolong the proper state of differentiation and thus the longevity of the animal. 4. Human longevity is a result of unusually high efficiency of these stabilizing mechanisms. 5 . Aging and cancer are a result of the same dysdifferentiation process.

Source of production: mitochondria cytochrome P-450 metabolic pathways

ENDOGENOUS GENERATION OF ACTIVE OXYGEN SPECIES

Modulatingfactors: caloric intake drugs exercise

PROPAGATION) Antioxidants

I

I I

1

DIFFERENTIATION/ GENE REGULATION

Cell morphology Hormone receptor Specific RNA species Oncogene expression

Steady state oxidation Mutational load DNA repair

It

CHANGE IN ENZYMATIC ACTIVITY, SPEClFlClTy

(DNA/PROTEIN COMPLW

FIGURE 12. Model of oxidative initiation of aging.

Modulating factors: diet (fatty acids, cholesterol) repair enzymes metal chelators chain-terminating antioxidants fatly acid composition

Antioxidants

1 , -

I

1 1

I

I

I

NON-NUCLEAR PROTEIN (MICROSOMES, MEMBRANES)

Protein turnover Protease activity

17

CUTLER: REACTIVE OXYGEN SPECIES

Fenilizatwn

Development

Reproduction

Aging

Death

FIGURE 13. Dysdifferentialive and remodeling nature of aging.

remodeling, and it is proposed that dysdifferentiation of cells might influence proper remodeling kinetics as well as the qualitative aspects of tissue remodeling. This model is shown in FIGURE 13 and could explain some of the qualitative agedependent changes found in bone and skin and how small changes of dysdifferentiation can be greatly amplified through remodeling kinetics.

Testing the Dysdifferentiative Hypothesis of Aging Details will not be presented in this paper of our work testing the dysdifferentiation hypothesis of aging.6 Briefly, we have conducted some experiments to determine if increased improper gene expression does occur with age.26 These results are summarized in TABLE12. More recent experiments in our laboratory have shown an age-dependent increase in the c-myc proto-oncogene in apparently normal tissuesz7(FIG.14). Because of evidence that an increase of c-myc expression may be caused by genetic alterations (perhaps via reactive oxygen species), it appears possible that such genetic alterations may also be accumulating with age in normal-appearing tissue, increasing the probability of transformation of every cell in the organ.27

Testing the Role Reactive Oxygen Species May Have in Aging Our first approach in testing the concept that aging may in part be caused by reactive oxygen species was carried out by determining if longer-lived species had superior defense mechanisms (antioxidants). According to our hypothesis, TABLE 12. Summary of Investigation of Possible Age-Dependent Changes of

Specific Genes No change with age (brain, liver, kidney): 1. Alpha-fetoprotein (mouse) 2. Casein (mouse) 3. Alpha and beta hemoglobin (human WI-38 cells) Increase with age (brain, liver, kidney): 1. Alpha and beta hemoglobin, quantitative (mouse and human) 2. Mouse leukemia virus (MuLV), qualitative (mouse) 3. Mouse mammary tumor virus (MMTV), qualitative (mouse) 4. c-myc oncogene, quantitative (3rd Extron, mouse)

18

ANNALS NEW YORK ACADEMY OF SCIENCES

r = 0842 P = 0.001 3

FIGURE 14. Relative c-myc proto-oncogene RNA levels in liver of mice as a function of age as determined by both Northern and slot blot data. The points are from three independent experiments (0,slot blot; On, Northern). Newborn RNA is used as internal standard in all experiments and assigned 100%. (From Semsei et a/.27Reprinted by permission from Oncogene.)

longer-lived species would not be predicted to have different types of protective mechanisms (a qualitative difference) but rather more of the same type of protective mechanisms (a quantitative difference) resulting from genetic changes occurring in regulatory genes.19,28 13 and shown Results of some of these experiments are summarized in TABLE 15.29These data generally indispecifically for superoxide dismutase in FIGURE cate that longer-lived species, particularly human, do indeed have higher levels of antioxidants per amount of reactive oxygen species produced e n d o g e n o ~ s l yA .~~

20-

FIGURE 15. Superoxide dismutase (SOD) concentration per specific metabolic rate (SMR) in liver of mammals as a function of maximum lifespan potential (MLSP). (From C ~ t l e r . ~ ' Reprinted by permission from Plenum Press.)

CUTLER: REACTIVE OXYGEN SPECIES

19

TABLE l3. Summary of Antioxidant Comparison Results Positive Correlation

No Correlation

1 . Cu/Zn SOD 2. Mn SOD 3. Carotenoids 4. Alpha tocopherol 5. Urate 6. Ceruloplasmin

1. Ascorbate 2. Retinol

Negative Correlation 1. Catalase

2. Glutathione 3. Glutathione peroxidase

simple experiment supporting this conclusion was carried out by comparing the spontaneous autoxidation rate in air of whole tissue homogenates taken from animals having different life spans. Rate of autoxidation was determined by measuring the amount of peroxides being produced (TBA assay). A typical result is shown in FIGURE16, indicating an increased resistance of autoxidation with increasing life span. Thus, in this experiment, human brain tissue was found to be the most resistant to spontaneous autoxidation. The mechanism for this resistance is not yet understood but could involve less peroxidative substances (unsaturated fatty acids) as well as higher levels of antioxidants. These data are supported by the recent publication of a series of papers indicating positive results of antioxidant vitamins and beta carogene in disease prevention.3oa

10 Mus musculus (3.5 yrs)

8

-

CI

0 X

P 0

Peromyscus leucopus

6

4

squirrel monkey (18 yrs) Rhesus (34 yrs)

2

human

4

8

12

(90 yrs)

16

Time (hrs) FIGURE 16. Kinetics of autoxidation of brain homogenate. (From Cutler.3oReprinted by permission from the National Academy of Sciences.)

20

ANNALS NEW YORK ACADEMY OF SCIENCES

If antioxidant protection is indeed higher in tissues of longer-lived species, then it would be expected that these same tissues should have a lower steady state level of oxidative damage. Since our model predicts the genetic apparatus of cells to be a key target to oxidative damage, we have compared the steady state oxidative damage level in DNA as a function of age and aging rate in different species. FIGURE 17 shows the chemical structures of some of the more common oxidative products of nucleic acid bases that might be expected to be found in v ~ v o . ~ ~ The experiments to be described were made possible by taking advantage of the high sensitivity of the electrochemical detector (ECD) in detecting the oxidative products of deoxyguanosine (dG). One product is 8-OHdG and can be measured at femtomole levels using an HPLC/ECD instrument. Typical results for liver 18, indicating that DNA from longer-lived species DNA are shown in FIGURE does appear to have a lower steady state level of oxidative damage, at least for 8-OHdG, as compared to short-lived species. After the repair of oxidative damage by excision of the oxidized nucleotide, the nucleoside appears later in the urine of an organism. By this pathway, a 24-hr urine sample would be expected to contain the total body level of excised products of 8-OHdG. This procedure allows the assessment of total body level of repair of oxidative DNA damage. Furthermore, if we assume that all oxidative products are repaired, then the amount of 8-OHdG detected could reflect rate of DNA damage as well as rate of repair.

Adenine

w

n

Thymini

Thymine Glycol

8 Hydrorvnuaninc

FIGURE 17. Mqjor modifications of nucleic acid bases formed by the superoxide radical. (From Jackson et ~ l .Reprinted ~ * by permission from the Journal of Clinical Inuestigation.)

21

CUTLER. REACTIVE OXYGEN SPECIES

10

y 0

-

M

FIGURE 18. Steady state level of 8-OH& per dG in liver DNA in species having different life spans (MLSPs).

g 2 u

'

: I

-

squirrel monkey

5 -

-

I

I

20

I

I

60

I

I

I

100

Preliminary results of such experiments are shown in FIGURE 19, where the amount of 8-OHdG per creatinine or lean body mass was found to decrease with increasing life span. These results suggest that the low levels of 8-OHdG found in human urine are likely to be the result of less 8-OHdG in the DNA needed to be repaired. The reason for the low level of oxidative damage in human DNA may in turn be the result of the unusually high levels of antoxidants that have been found in human tissue. Thus, the comparative results of antioxidant levels in tissues, DNA damage in tissues and urine levels of oxidized nucleosides are mutually complementary and together support the importance of oxidative damage as a cause of aging and of antioxidants in governing aging rate. These results are summarized in FIGURE 20.

Experimental Results from Other Laboratories Supporting the Longevity Determinant Gene Hypothesis An important argument demonstrating the potential feasibility of effective intervention in human aging processes is the demonstration in laboratory animals that relatively simple processes can significantly increase life span. Below is a brief list of such experiments. 1. It is now well established that food or calorie restriction can dramatically decrease aging rate in experimental animals (mice and rats) as compared to ad libitum-fed controls.33The mechanism of how food restriction works is now being actively explored, but it does not appear to be a result of food restriction itself. Instead, a popular hypothesis is that food restriction stimulates through a few key regulatory mechanisms a change in hormonal status and/or a neuroendocrine response that in turn acts to slow down the aging rate of most physiological systems of the organism. Thus, food restriction experiments support the view that aging rate is coordinately governed by a few key regulatory p r o c e ~ s e s . ~ ~ , ~ ~

22

ANNALS NEW YORK ACADEMY OF SCIENCES

Rhesus

(n.5,

I

=

5)

chimpanzee

-

20

( n = 10, I = 5 )

40

Maximum

human ( n = IS , I = 6.) \

80

60

Iifespan potential

(yrs)

FIGURE 19. Relative amount of 8-OHdG per creatinine in urine samples taken from species having different life spans (MLSPs).

CUTLER: REACTIVE OXYGEN SPECIES

23

2. Experiments designed to select for long-lived strains of Drosophila and nematodes have been remarkably successful. Preliminary results indicate that some of the long-lived Drosophila strains have enhanced levels of antioxidant prote~tion.3~"~ More impressive are the results in nematodes, where substantial increase of life span was found to be due to a change in activity of a single gene called AGE-1 .39,40 These experiments demonstrate that life span can be significantly increased by changes in one or a few genes. 3. Transgenic Drosophila strains carrying extra gene dosages of the protein system elongation factor EL-1 have been found to have a life span about twice as long as controls.41 This result again demonstrates that life span can be increased by changing the expression of a single gene. 4. Hormone replacement therapy using human growth hormone, both in human and experimental animals, has been found to have remarkable rejuvenative results in terms of thickness of skin and the mass ratio of lean muscle to body 5. Topical application of retinoic acid to skin has been found to have antiaging effects and appears to reverse cellular changes in the skin that seem to be dysdifferentiated. Retinoic acid, known to effect states of differentiation of a cell, may actually be capable of reversing many of the characteristics of aging.42

SUMMARY AND CONCLUSION A brief overview has been given of the biological nature of human aging processes, where it has been emphasized that, in addition to the diseases of aging, there is also great economic loss as a result of human aging processes that began many years before medical costs related to aging begin to escalate. Because of the ubiquitous nature of aging, reducing the function of essentially all physiological processes, it appears that the only long-term solution to human aging problems is to decrease uniformly the aging rate of the entire body. Although the uniform decrease of aging rate has usually been considered impossible, where emphasis has consequently been placed on diseases of aging by the medically-orientated investigator, there is now at least one theoretical argument, accompanied by some experimental data, that suggests that progress can be made in achieving this goal. This progress has been based on the longevity determinant gene hypothesis predicting the existence of a relatively few key regulatory factors governing aging rate of the entire organism. If this hypothesis is not true, then indeed the prospect for significant intervention into human aging would appear impossible in the near future. Experiments have been briefly reviewed testing the longevity determinant gene hypothesis, the possibility that aging may be a result of dysdifferentiation and if aging rate is determined by mechanisms acting to stabilize the differentiated state of cells. In testing the dysdifferentiation hypothesis of aging, there is not yet much data one way or the other. It is evident, however, that changes in gene expression do occur with age, sometimes involving endogenous retroviruses or oncogenes. Other morphological evidence shows an increase with age in unusual cell type such as metaplasia cells. However, there is considerably more evidence indicating that aging may be a result of genetic instability (as it is in cancer) and that longer-lived species appear to have a more stable genetic apparatus and superior protective mechanisms against reactive oxygen species. There is a striking similarity in this model of

I I _ I I

input

DNAdamage

Antioxidant defense processes

+

Steady state equilibrium mutational load of genome

output

DNA damage

*

Urine 8-OHdG/creatinine

%

s

m

E b

4 0

m

hfe span potential (yrs)

I

Maximum

20

I

60

I

I

too

~~~

life span potential ( y r s )

I

Ll

Steady state mutation load

II

D N A data

Rate of damage output

Urine data

fPnO.,.

nVnmPCCP..

FIGURE 20. Model showing the relationshipof oxidative damage in DNA with that of oxidative damage found in the urine. The conclusion is thal antioxidant defense mechanisms have played a major role in accounting for the decrease in DNA damage in longer-lived species rather than DNA

Rate of damage input

Maximum

DNA data predicted

ANNALS NEW YORK ACADEMY OF SCIENCES

26 TABLE 14.

Summary and Conclusion

1. Evidence indicates aging to be a result of genetic instability (dysdifferentiation) and

2. 3.

4. 5.

aging rate to be governed by processes acting to stabilize the differentiated state of cells. Evidence also indicates cancer to be a result of genetic instability leading to altered states of differentiation. Aging rate and onset frequency of cancer appear to be correlated in different mammalian species. Age-dependent increase in expression of endogenous viruses (MuLV, MMTV, 4-1 human endogenous virus) and oncogenes (c-myc) are found to occur in apparently normal noncancerous tissues. Methods known to decrease aging rate (food restriction) also are effective in decreasing cancer incidence and those increasing aging rate (ionizing radiation) increase cancer rate.

TABLE 15.

Knowledge and Knowledge Gaps Knowledge

Knowledge Gaps

Aging is associated with dysdifferen tiation. Oxyradicals can cause dysdifferentiation (genetic alterations). Aging is associated with oxyradical generation and defense/ protective mechanisms.

But to what extent does this association occur? Can dysdifferentiation cause aging? Rut, do oxyradicals actually cause sufficient genetic alterations in vivo to contribute to aging? But, is the association of aging, oxyradicals and dysdifferentiation a cause-and-effect relationship?

aging and models of cancer, and much might be gained in bringing together these two fields of research. Taking all of these data together, as summarized in TABLE14, it appears we may be on the right track and that mechanisms acting to protect DNA against oxidative damage may be one class of longevity determinant mechanisms. There is of course much work remaining to be done, some of which is listed in TABLE15 in terms of our knowledge and our gaps of knowledge in this field. REFERENCES SCHNEIDER, E. L. & J. M. GURALNIK. 1990. The aging of America. Impact on health care costs. J. Am. Med. Assoc. 263: 233-2340, 2. FRIES,J. F. & L. M. CRAPO.1981. Vitality and Aging. W. H. Freeman & C o . San Francisco, CA. 3. MARSH,R. P. 1989. Ethical implications of life extension. Age 12: 103-106. 4. CUTLER,R. G. 1976. Nature of aging and life maintenance processes. In Interdisciplinary Topics in Gerontology. R. G. Cutler, Ed. Vol. 9: 83-133. S . Karger. Easel. 5. CUTLER,R. G. 1976. Evolution of longevity in primates. J . Hum. Evol. 5: 169-202. 6. CUTLER, K.G . 1982. Longevity is determined by specific genes: Testing the hypothesis. I n Testing the Theories of Aging. R. Adelman & G. Roth, Eds. 25-114. CRC 1.

Press. Boca Raton, FL.

CUTLER: REACTIVE OXYGEN SPECIES

27

7. CUTLER,R. G. 1984. Evolutionary biology of aging and longevity. I n Aging and Cell Structure. J. E. Johnson, Ed. Vol. 2: 371-428. Plenum Press. New York, NY. 8. EVERITT,A. & J. MEITES.1989. Aging and anti-aging effects of hormones. J. Gerontol. 44: B139-Bl41. 8a. MEITES,J. 1990. Aging: hypothalamic catecholamines, neuroendocrine-immune interactions, and dietary restriction. Proc. SOC.Exp. Biol. Med. 195: 304-311. D., A. G. FELLER, H . S. NAGRAJ, G. A. GERGANS, P. Y. LALITHA,A. F. 9. RUDMAN, & D. !?. MATTSON.1990. GOLDBERG, R. A. SCHLENKEK, L. COHN, I. w.RUDMAN Effects of human growth hormone in men over 60 years old. N. Engl. J. Med. 323: 1-6. 10. CUTLER,R. G. 1975. Evolution of human longevity and the genetic complexity governing aging rate. Proc. Natl. Acad. Sci. USA 72: 4664-4668. 10a. WILSON,A. C. 1976. Gene regulation in evolution. In Molecular Evolution. F. J. Ayala, Ed. 225-234. Sinauer Associates, Inc. 11. COMFORT,A. 1978. The Biology of Senescence. Elsevier. New York, NY. 1970. History of Human Lifespan and Mortality, Akadt12. ACSADI,G. & J. NEMESK~RI. miai Kiad6. Budapest. B. L.. 1978. Time, Cells and Aging. Academic Press. New York, NY. 13. STREHLER, 14. SAGAN,L. A. 1987. The Health of Nations. Basic Books, Inc. New York, NY. 15. KOHN,R. R. 1978. Principles of Mammalian Aging. Prentice-Hall, Inc. Englewood Cliffs, NJ. 15a. KOHN,R. R. 1982. Cause of death in very old people. J. Am. Med. Assoc. 247: 27932797. S. J., B. A. CARNES& C. CASSEL.1990. In search of Methuselah: esti15b. OLSHANSKY, mating the upper limits to human longevity. Science 250: 634-640. E . L. & J. W. ROWE,EDS. 1990. Handbook of the Biology of Aging. 3rd 16. SCHNEIDER, edit. Academic Press. New York, NY. 17. BAFITIS,H. & F. SARGENT.1977. Human physiological adaptability through the life sequence. J. Gerontol. 3 2 402-410. 18. National Center of Health Statistics, USPHS & U.S. Bureau of the Census, “Some Demographic Aspects of Aging in the United States.” Feb. 1973. 19. CUTLER,R. G. 1986. Aging and oxygen radicals. I n Physiology of Oxygen Radicals. A. E. Taylor, S. Matlon & P. Ward, Eds. Clinical Monograph Series. 251-285. Am. Physiol. SOC.Bethesda, MD. G. A. 1975. Maturation and longevity in relation to cranial capacity in homi19a. SACHER, nid evolution. In Primate Functional Morphology and Evolution. R. Tuttle, Ed. 417441. Mouton. The Hague. E. L. & J. D. REED. 1985. Modulations of aging processes. In Handbook 19b. SCHNEIDER, of the Biology of Aging. C . E . Finch & E. L. Schneider, Eds. 45-76. Van Nostrand Reinhold. New York. 19c. KIRKLAND,J. L. 1989. Evolution and ageing. Genome 31: 398-405. 20. CUTLER.R. G. 1983. Superoxide dismutase, longevity . and specific metabolic rate. Gerontology 29: 113-120. D. 1981. The aging- .process. Proc. Natl. Acad. Sci. USA 7 8 7124-7128. 2Oa. HARMAN. B. & J. M. C. GUTTERIDGE. 1989. Free Radicals in Biology and Medi20b. HALLIWELL, cine. Clarendon Press. Oxford. 20c. SAUL,R. L., P. GEL & B. N . AMES.1987. Free radicals, DNA damage, and aging. In Modern Biological Theories of Aging. H. R. Warner, R. N. Butler, R. L . Sprott & E. L . Schneider, Eds. 113-129. Raven Press. New York. 20d. ANISIMOV,V. N. 1989. Dependence of susceptibility t o carcinogenesis on species life span. Arch. Geschwulstforsch. 59: 205-213. 1989. Development, cancer and aging: Possible com21. CUTLER,R. G. A N D 1. SEMSEI. mon mechanisms of action and regulation. J . Gerontol. 44: 25-34. 22. AMES,B. N. 1983. Dietary carcinogens and anticarcinogens. Science 221: 1256-1264. 23. CEnuTri, P. A. 1985. Prooxidant states and tumor promotion. Science 227: 375-381. 24. CUTLER,R. G. 1985. Dysditrerentiation and aging. I n Molecular Biology of Aging: Gene Stability and Gene Expression. R. S. Sohal, L. Birnbaum & R. G. Cutler, Eds. 307-340. Raven Press. New York. NY. I

28

25.

25a. 26. 27. 28. 29. 30. 30a. 31. 32. 33. 34. 35. 36. 37.

38. 39. 40. 41. 42.

ANNALS NEW YORK ACADEMY OF SCIENCES ROHME,D. 1981. Evidence for a relationship between longevity of mammalian species and life spans of normal fibroblasts in vitro and erythrocytes in vivo. Proc. Natl. Acad. Sci. USA 7 8 5009-5013. FINCH,C. E. 1984. Recent findings in the neurobiology of aging. In Frontiers in Medicine. Implications for the Future. R. J. Morin & R. J . Bing, Eds. 264-283. Human Sciences Press, lnc. New York. ONO,T. & R. G. CUTLER.1978. Age-dependent relaxation of gene repression; increase of globin and endogenous murine leukemia virus related RNA in brain and liver of mouse. Proc. Natl. Acad. Sci. USA 75: 4431-4435. SEMSEI,I., S. MA & R. G. CUTLER.1989. Tissue and age specific expression of the myc protooncogene family throughout the life span of the C57BL/6J mouse strain. Oncogene 4 465-470. CUTLER,R. G. 1984. Antioxidants, aging, and longevity. I n Free Radicals in Biology. W. A. Pryor, Ed. Vol. 6: 371-428. Academic Press. New York, NY. J. M., T. ONO& R. G. CUTLER.1980. Superoxide dismutase: correlation TOLMASOFF, with life span and specific metabolic rate in primate species. R o c . Natl. Acad. Sci. USA 72 2777-2781. CUTLER,R. G. 1985. Peroxide-producing potential of tissues: correlation with the longevity of mammalian species. Proc. Natl. Acad. Sci. USA 82: 4798-4802. SLATER,T. F. & G. BLOCK. 1991. Antioxidant vitamins and 0-carotene in disease prevention. Am. J . Clin. Nutr. 53(Suppl.): 189s-396s. CUTLER,R. G. 1985. Antioxidants and longevity of mammalian species. In Molecular Biology of Aging. A. D. Woodhead, A. D. Blackett & A. Hollaender, Eds. 15-73. Plenum Press. New York, NY. JACKSON,J . H., E. GAJEWSKI,I. U. SCHRAUFSTATTER, P. A. HYSLOP,A. F. FuCIARELLI, C. G. COCHRANE 8 M.DIZDAROGLU. 1989. Damage to the bases in DNA induced by stimulated human neutrophils. J. Clin. Invest. 84: 1644-1649. WEINDRUCH, R. & R. L. WALFORD.1988. The Retardation of Aging and Disease by Dietary Restriction. C. C. Thomas, Pub. Springfield, IL. Yu, B. P., D. W. LEE, C. G. MARLER& J.-H. CHOI. 1990. Mechanism of food restriction: protection of cellular homeostasis. Proc. SOC.Exp. Biol. Med. 193: 1315. LUCKINBILL, L. S., J. L. GRAVES,A. H. REED& S. KOETSAWANG. 1988. Localizing genes that defer senescence in Drosophila melanogaster. Heredity 60:367-374. ARKING,R. & S. P. DUDAS. 1989. Review of genetic investigations into the aging processes of Drosophila. J. Am. Geriatr. SOC.37: 757-773. D. MICHAUD,M. CHARBONNEAU & A. J. HILLIKER. PHILLIPS,J. P., S. D. CAMPBELL, 1989. Null mutation of copperlzinc superoxide dismutase in Drosophila confers hypersensitivity to paraquat and reduced longevity. Proc. Natl. Acad. Sci. USA 86: 2761-2765. SETO,N. 0. L., S. HAYASHI& G. M. TENER.1990. Over expression of Cu-Zn superoxide dismutase in Drosophila does not affect life-span. Proc. Natl. Acad. Sci. USA 87: 4270-4274. FRIEDMAN, D. B. & T. E. JOHNSON.1987. A mutation in the age-1 gene in Caenorhabditis elegnns lengthens life and reduces hermaphrodite fertility. Genetics 118: 75-86. JOHNSON,T. E. 1988. Genetic specification of life span: processes, problems, and potentials. J. Gerontol. 43: B87-B92. SHEPHERD,J. C., U. WALLDORF,P. HUG & W. J. GEHRING.1989. Fruit flies with additional expression of the elongation factor EF-la live longer. Proc. Natl. Acad. Sci. USA 86: 7520-7521. KLIGMAN,A. M., G. L. GROVE,R. HIROSE& J. J. LEYDEN.1986. Topical tretionin for photoaged skin. J. Am. Acad. Dermatol. 15: 836-859.