NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Gilbert SF. Developmental Biology. 6th edition. Sunderland (MA): Sinauer Associates; 2000.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of Developmental Biology

Developmental Biology. 6th edition.

Show details

Aging: The Biology of Senescence

Entropy always wins. Each multicellular organism, using energy from the sun, is able to develop and maintain its identity for only so long. Then deterioration prevails over synthesis, and the organism ages. Aging can be defined as the time-related deterioration of the physiological functions necessary for survival and fertility. The characteristics of aging—as distinguished from diseases of aging (such as cancer and heart disease)—affect all the individuals of a species.

Many evolutionary biologists (Medawar 1952; Kirkwood 1977) would deny that aging is part of the genetic repertoire of an animal. Rather, they would consider aging to be the default state occurring after the animal has fulfilled the requirements of natural selection. After its offspring are born and raised, the animal can die. Indeed, in many organisms, from moths to salmon, this is exactly what happens. As soon as the eggs are fertilized and laid, the adults die. However, recent studies have indicated that there are genetic components to senescence, and that the genetically determined life span characteristic of a species can be modulated by altering genes or diet.

Maximum Life Span and Life Expectancy

The maximum life span is a characteristic of the species. It is the maximum number of years a member of that species has been known to survive. The maximum human life span is estimated to be 121 years (Arking 1998). The life spans of tortoises and lake trout are both unknown, but are estimated to be more than 150 years. The maximum life span of a domestic dog is about 20 years, and that of a laboratory mouse is 4.5 years. If a Drosophila fruit fly survives to eclose (in the wild, over 90% die as larvae), it has a maximum life span of 3 months.

However, a person cannot expect to live 121 years, and most mice in the wild do not live to celebrate their first birthday. The life expectancy, the amount of time a member of a species can expect to live, is not characteristic of species, but of populations. It is usually defined as the age at which half the population still survives. A baby born in England in the 1780s could expect to live to be 35 years old. In Massachusetts during that same time, the life expectancy was 28 years. This was the normal range of human life expectancy for most of the human race in most times. Even today, the life expectancy in some areas of the world (Cambodia, Togo, Afghanistan, and several other countries) is less than 40 years. In the United States, a child born in 1986 can expect to live 71 years if male and 78 years if female.*

Given that in most times and places, humans did not live much past 40 years, our awareness of human aging is relatively new. A 65-year-old person was rare in colonial America, but is a common sight today. Some survival curves for female Homo sapiens in the United States are plotted in Figure 18.35. In 1900, 50% of American women were dead by age 58. In 1980, 50% of American women were dead by age 81. Thus, the phenomena of senescence and the diseases of aging are much more common today than they were a century ago. In 1900, people did not have the “luxury” of dying from heart attacks or cancers. These diseases generally occur in people over the age of 50 years. Rather, people died (as they are still dying in many parts of the world) from infectious diseases and parasites (Arking 1998). Similarly, until recently, relatively few people exhibited the more general human sensecent phenotype: graying hair, sagging and wrinkling skin, joint stiffness, osteoporosis (loss of bone calcium), loss of muscle fibers and muscular strength, memory loss, eyesight deterioration, and the slowing of sexual responsiveness. As Shakespeare noted in As You Like It, those who did survive to senescence left the world “sans teeth, sans eyes, sans taste, sans everything.”

Figure 18.35. Survival curves for U.

Figure 18.35

Survival curves for U.S. females in 1900, 1960, and 1980. M50 represents the age at which 50% of the individuals of each population survived. (After Arking 1998.)

Causes of Aging

The general senescent phenotype is characteristic of each species. But what causes it? This question can be asked at many levels. We will be looking primarily at the cellular level of organization. Even here, there is evidence for many different theories, and there is not yet a consensus on what causes aging.

Oxidative damage

One major theory sees our metabolism as the cause of our aging. According to this theory, aging is a by-product of normal metabolism; no mutations are required. About 2–3% of the oxygen atoms taken up by the mitochondria are reduced insufficiently to reactive oxygen species (ROS). These ROS include the superoxide ion, the hydroxyl radical, and hydrogen peroxide. ROS can oxidize and damage cell membranes, proteins, and nucleic acids. Evidence for this theory includes the observation that Drosophila that overexpress enzymes that destroy ROS (catalase, which degrades peroxide, and superoxide dismutase) live 30–40% longer than do controls (Orr and Sohal 1994; Parkes et al. 1998). Moreover, flies with mutations in the methuselah gene (named after the Biblical fellow said to have lived 969 years) live 35% longer than wild-type flies. The methusaleh mutants have enhanced resistance to paraquat, a poison that works by generating ROS within cells (Lin et al. 1998). These findings not only suggest that aging is under genetic control, but also provide evidence for the role of ROS in the aging process. In C. elegans, too, individuals with mutations that increase the synthesis of ROS-degrading enzymes live much longer than wild-type nematodes (Larsen 1993; Vanfleteren and De Vreese 1996).

The evidence for ROS involvement in mammalian aging is not as clear. Mutations in mice that result in the lack of certain ROS-degrading enzymes do not cause premature aging (Ho et al. 1997; Melov et al. 1998). However, there may be more genetic redundancy in mammals than in invertebrates, and other genes may be up-regulated to produce related ROS-degrading enzymes. Migliaccio and colleagues (1999) have observed mutant mice that live one-third longer than their wild-type littermates. These mice lack a particular protein, p66shc. They develop normally, but the lack of p66shc apparently gives them cellular resistance to ROS, and thus higher resistance to oxygen-induced stress on membranes and proteins. The p66shc protein may be a component of a signal transduction pathway that leads to apoptosis upon oxygen stress, and it may be involved in mediating the life spans of mammals.

Another type of evidence does suggest that ROS may be important in mammalian aging: aging in mammals can be slowed by caloric restriction (Lee et al. 1999). However, caloric restriction can also have other effects, so it is not certain if it works by preventing ROS synthesis. Also, vitamins E and C are both ROS inhibitors, and vitamin E increases the longevity of flies and nematodes when it is added to their diet (Balin et al. 1993; Kakkar et al. 1996). However, results in mammals are not as easy to interpret, and there is no clear evidence that ROS inhibitors work as well as in invertebrates (Arking 1998).

General wear-and-tear and genetic instability

“Wear-and-tear” theories of aging are among the oldest hypotheses proposed to account for the general scenescent phenotype (Weismann 1891; Szilard 1959). As one gets older, small traumas to the body build up. Point mutations increase in number, and the efficiencies of the enzymes encoded by our genes decrease. Moreover, if a mutation occured in a part of the protein synthetic apparatus, the cell would make a large percentage of faulty proteins (Orgel 1963). If mutations arose in the DNA-synthesizing enzymes, the rate of mutations would be expected to increase markedly, and Murray and Holliday (1981) have documented such faulty DNA polymerases in senescent cells. Likewise, DNA repair may be important in preventing senescence, and species whose members' cells have more efficient DNA repair enzymes live longer (Figure 18.36; Hart and Setlow 1974). Moreover, genetic defects in DNA repair enzymes can produce premature aging syndromes in humans (Yu et al. 1996; Sun et al. 1998).

Figure 18.36. Correlation between life span and the ability of fibroblasts from various mammalian species to repair DNA.

Figure 18.36

Correlation between life span and the ability of fibroblasts from various mammalian species to repair DNA. Capacity for repair is represented in autoradiography by the number of grains from radioactive thymidine per cell nucleus. Note that the y axis (more...)

Mitochondrial genome damage

The mutation rate in mitochondria is 10–20 times faster than the nuclear DNA mutation rate (Johnson et al. 1999). It is thought that mutations in mitochondria could (1) lead to defects in energy production, (2) lead to the production of ROS by faulty electron transport, and/or (3) induce apoptosis. Age-dependent declines in mitochondrial function are seen in many animals, including humans (Boffoli et al. 1994). A recent report (Michikawa et al. 1999) shows that there are “hot spots” for age-related mutations in the mitochondrial genome, and that mitochondria with these mutations have a higher replication frequency than wild-type mitochondria. Thus, the mutants are able to outcompete the wild-type mitochondria and eventually dominate the cell and its progeny. Moreover, the mutations may not only allow more ROS to be made, but may make the mitochondrial DNA more susceptible to ROS-mediated damage.

Telomere shortening

Telomeres are repeated DNA sequences at the ends of chromosomes. They are not replicated by DNA polymerase, and they will shorten at each cell division unless maintained by telomerase. Telomerase adds the telomere onto the chromosome at each cell division. Most mammalian somatic tissues lack telomerase, so it has been proposed (Salk 1982; Harley et al. 1990) that telomere shortening could be a “clock” that eventually prohibits the cells from dividing any more. When human fibroblasts are cultured, they can divide only a certain number of times, and their telomeres shorten. If these cells are made to express telomerase, they can continue dividing (Bodnar et al. 1998; Vaziri and Benchimol 1998).

However, there is no correlation between telomere length and the life span of an animal (humans have much shorter telomeres than mice), nor is there a correlation between human telomere length and a person's age (Cristofalo et al. 1998). Telomerase-deficient mice do not show profound aging defects, which we would expect if telomerase were the major factor in determining the rate of aging (Rudolph et al. 1999). It has been suggested that telomere-dependent inhibition of cell division might serve primarily as a defense against cancer rather than as a kind of “aging clock.”

Genetic aging programs

Several genes have been shown to affect aging. In humans, Hutchinson-Gilford progeria syndrome causes children to age rapidly and to die (usually of heart failure) as early as 12 years (Figure 18.37). It is caused by a dominant mutant gene, and its symptoms include thin skin with age spots, resorbed bone mass, hair loss, and arteriosclerosis. A similar syndrome is caused by mutations of the klotho gene in mice (Kuro-o et al. 1997). The functions of the products of these genes are not known, but they are thought to be involved in suppressing the aging phenotypes. These proteins may be extremely important in determining the timing of senescence.

Figure 18.37. Children with progeria.

Figure 18.37

Children with progeria. Although less than 8 years old, the child on the right has a phenotype similar to that of an aged person. The hair loss, fat distribution, and transparency of the skin are characteristic of the normal human aging pattern seen in (more...)

In C. elegans, there appear to be at least two genetic pathways that affect aging. The first pathway involves the decision to remain a larva or to continue growth. After hatching, the C. elegans larva proceeds through four instar stages, after which it can become an adult or (if the nematodes are overcrowded or if there is insufficient food) can enter a nonfeeding, metabolically dormant dauer stage. It can remain a dauer larva for up to 6 months, rather than becoming an adult that lives only a few weeks. When it comes out of the dauer stage, it will live as long as if it had never been a dauer larva. In the dauer stage, adult development is suppressed, and extra defenses against ROS are synthesized. If some of the genes involved in this pathway are mutated, adult development is allowed, but the ROS defenses are still made. The resulting adults live twice to four times as long as wild-type adults (Figure 18.38; Friedman and Johnson 1988). The second pathway involves the gonads. Germ cells appear to inhibit longevity, while the somatic cells of the gonads act to prolong the life of the nematode (Hsin and Kenyon 1999).

Figure 18.38. Proposed mechanism for extending the life span of C.

Figure 18.38

Proposed mechanism for extending the life span of C. elegans through the dauer larva pathway. (A) Wild-type animal in a favorable environment makes a ligand that activates a pathway that inhibits the DAF-16 protein. This allows metamorphosis to the adult (more...)

As human life expectancy increases due to our increased ability to prevent and cure disease, we are still left with a general aging syndrome that is characteristic of our species. Unless attention is paid to the general aging syndrome, we risk ending up like Tithonios, the miserable wretch of Greek mythology to whom the gods awarded eternal life, but not eternal youth.

Snapshot Summary: Metamorphosis, Regeneration, and Aging


Amphibian metamorphosis includes both morphological and biochemical changes. Some structures are remodeled, some are replaced, and some new structures are formed.


Many changes during amphibian metamorphosis are regionally specific. The tail epidermis dies, the head epidermis does not. An eye will persist even if transplanted into a degenerating tail.


The hormones responsible for amphibian metamorphosis are the thyroid hormones thyroxine (T4) and triiodothyronine (T3). The coordination of metamorphic changes appears to be due to early changes that occur at low concentrations of the thyroid hormones. This is called the threshold concept. The molecular basis for the autoinduction of thyroid hormones may be the ability of thyroid hormones to induce production of more thyroid hormone receptor protein. Thyroid hormones act predominantly at the transcriptional level.


Heterochrony involves changing the relative rate of development in different parts of the animal. In animals with direct development, the tadpole stage has been lost. Some frogs, for instance, form limbs while in the egg.


In neoteny, the juvenile (larval) form is slowed down, while the gonads and germ cells mature at their normal rate. In progenesis, the gonads and germ cells mature rapidly, while the rest of the body matures normally. In both instances, the animal can mate while in its larval form.


In ametabolous insects, there is direct development. In hemimetabolous insects, there is a nymph stage wherein the immature organism is usually a smaller version of the adult. In holometabolous insects, there is a dramatic metamorphosis from larva to pupa to sexually mature adult.


In the period between larval molts, the larva is called an instar. After the last instar stage, the larva undergoes a metamorphic molt to become a pupa. The pupa will undergo an instar molt to become an adult.


During the pupal stage, the imaginal discs and histoblasts grow and differentiate to produce the structures of the adult body.


The anterior-posterior, dorsal-ventral, and proximal-distal axes are sequentially specified and involve interactions between different compartments in the imaginal discs.


Molting is caused by the hormone hydroxyecdysone. In the presence of high titres of juvenile hormone, the molt is an instar molt. In low concentrations of juvenile hormone, the molt produces a pupa; and if no juvenile hormone is present, the molt is an imaginal molt.


The ecdysone receptor gene can produce nRNA that can form at least three different proteins. The types of ecdysone receptors in a cell may influence the response of that cell to hydroxyecdysone. The ecdysone receptors bind to DNA to activate or repress transcription.


There are three major types of regeneration. In epimorphosis (such as regenerating limbs), tissues dedifferentiate into a blastema, divide, and re-differentiate into the new structure. In morphallaxis (characteristic of hydra), there is a repatterning of existing tissue with little or no growth. In compensatory regeneration (such as in the liver), cells divide but retain their differentiated state.


In the regenerating salamander limb, the epidermis forms an apical ectodermal cap. The cells beneath it dedifferentiate to form a blastema. The differentiated cells lose their adhesions and re-enter the cell cycle. This does not happen in mammals.


In hydras, there appear to be head activation gradients, head inhibition gradients, foot activation gradients, and foot inhibition gradients. Hydra budding occurs where these gradients are minimal.


In mammals, medical researchers are testing whether paracrine factors may permit local regeneration. Bone and neural cells are being returned to embryonic conditions in the hopes that they will regrow. Natural inhibitors of neural regeneration have recently been discovered, and their circumvention may allow spinal cord regeneration.


The maximum life span of a species is how long its longest observed member has lived. It is largely characteristic of a given species. Life expectancy is the time at which approximately 50 percent of the members of a given population of a species still survive.


There are several levels at which we can study aging, including cellular, biochemical, and genetic studies. Reactive oxygen species (ROS) can damage cell membranes, inactivate proteins, and mutate DNA. Mutations that alter the ability to make or degrade ROS can change the life span of the mutants.


Mitochondria may be a target for proteins that regulate aging.


Aging is the time-related deterioration of the physiological functions necessary for survival and reproduction. The phenotypic changes of senescence (which affect all members of the species) are not to be confused with diseases of senescence, such as cancer and heart disease (which affect individuals).



You can see why the funding of Social Security is problematic in the United States. When it was created in 1935, the average working citizen died before age 65. Thus, he (and it usually was a he) was not expected to get back what he had paid into the system. Similarly, marriage “until death do us part” was an easier feat when death occurred in the third or fourth decade of life. The death rate of young women due to infections associated with childbirth was high throughout the world before antibiotics.

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2000, Sinauer Associates.
Bookshelf ID: NBK10041


  • Cite this Page

Recent Activity

    Your browsing activity is empty.

    Activity recording is turned off.

    Turn recording back on

    See more...