No treatment on the market has been proved to slow human aging--the buildup of molecular and cellular damage that increases vulnerability to infirmity as we grow older. But one intervention, consumption of a low-calorie yet nutritionally balanced diet, works incredibly well in a broad range of animals, increasing longevity and prolonging good health. Those findings suggest that calorie restriction could delay aging in humans, too.

Unfortunately, for maximum benefit, people would probably have to reduce their calorie intake by roughly 30 percent, equivalent to dropping from 2,500 calories a day to 1,750. Although a few hardy souls are currently attempting to do this, most mortals could not stick to that harsh a regimen, especially for years on end. But what if someone could create a pill that mimicked the physiological effects of eating less without actually forcing people to go hungry? Could such a calorie-restriction mimetic, as we call it, enable people to stay healthy longer, postponing age-related disorders (such as diabetes, atherosclerosis, heart disease and cancer) until very late in life?

We first posed this question in the mid-1990s, after we came upon a chemical agent that, in rodents, seemed to reproduce many of calorie restriction's benefits. Since then, we and others have been searching for a compound that would safely achieve the same feat in people. We have not succeeded yet, but our failures have been informative and have fanned hope that calorie-restriction, or CR, mimetics can indeed be developed eventually.

Our hunt for CR mimetics grew out of our desire to better understand calorie restriction's many effects on the body. Scientists first recognized the value of the practice more than 60 years ago, when they found that rats fed a low-calorie diet lived longer on average than free-feeding rats and had a reduced incidence of conditions that become increasingly common in old age. What is more, some of the treated animals survived longer than the oldest-living animals in the control group, which means that the maximum life span (the oldest attainable age), not merely the average life span, increased.

The rat findings have been replicated many times and extended to creatures ranging from yeast to fruit flies, worms, fish, spiders, mice and hamsters. Until fairly recently, the studies were limited to short-lived creatures genetically distant from humans. But a long-term study in dogs was published a couple of years ago to show that CR could be effective for our pets as well. A few long-term projects under way in a species more closely related to humans--the rhesus monkey--suggest that primates respond to calorie restriction almost identically to rodents, which makes us more optimistic than ever that CR mimetics could help people.

The monkey projects--initiated by our group at the National Institute on Aging in the late 1980s and by our colleagues at the University of WisconsinMadison in the early 1990s--demonstrate that, compared with control animals that eat normally, calorie-restricted monkeys have lower body temperatures and levels of the pancreatic hormone insulin, and as young adults they retain more youthful levels of certain hormones (such as DHEAS, or dehydroepiandrosterone sulfate) that tend to fall with age.

The animals also look better on indicators of risk for age-related diseases. For example, they have lower blood pressure and triglyceride levels (signifying a decreased likelihood of heart disease), and they have more normal blood glucose levels (pointing to a reduced risk for diabetes, which is marked by unusually high blood glucose levels). They and the other monkeys must be followed still longer, however, before we will know whether low food intake can increase both average and maximum life spans in monkeys: rhesus monkeys typically live about 25 years and sometimes up to 40. Findings in primates bode well for the possibility that CR will have beneficial effects in humans. Indeed, we have demonstrated that biological hallmarks of CR, such as decreased insulin and body temperature and a slowed rate of decline in serum DHEAS levels, are associated with better survival in men from the Baltimore Longitudinal Study of Aging.

The Journey Starts
BY 1995 WE WANTED to know how the many physiological and biochemical changes induced by calorie restriction actually delayed aging in mammals. We suspected that changes in cellular metabolism would be key. By "metabolism" we mean the uptake of nutrients from the blood and their conversion to energy usable for cellular activities. We focused on metabolism in part because the benefits of calorie restriction clearly depend on reducing the overall amount or temporal pattern of fuel coming into the body for processing. Also, calorie restriction affects aging in a wide variety of tissues, which implies that it alters biological processes present in all cells. Few processes are more fundamental than metabolism.

We specifically wondered whether changes related to metabolism of the sugar glucose would account for CR's benefits. Glucose, which forms when the body digests carbohydrates, is the primary source of energy in the body.We also wanted to know whether alterations in the secretion and activity of insulin, which influences glucose use by cells, would be important. Insulin is secreted as glucose levels in the blood rise after a meal, and it serves as the key that opens cell "doors" to the sugar. We concentrated on glucose and insulin because reductions in their levels and increases in cellular sensitivity to insulin are among the most consistent hallmarks of calorie restriction in both rodents and primates, occurring very soon after restriction is begun.

Others began publishing data showing that metabolic processes involving glucose and insulin influence life span. For instance, several investigations achieved remarkable extensions of life span in nematode worms by mutating genes similar to those involved in molecular responses to insulin in mammals. More recently, researchers have found that lowered intake of glucose or disruption of glucose processing can extend life span in yeast.

An "Aha!" Moment
AROUND THE TIME the nematode work came out, we began to scour the scientific literature for ways to manipulate insulin secretion and sensitivity without causing diabetes or its opposite, hypoglycemia. Our search turned up studies from the 1940s and 1950s mentioning a compound called 2-deoxy-D-glucose (2DG) that was being tested in rodents for treating cancer but that also reportedly lowered insulin levels in the blood. As we perused the literature further, we had a true "aha!" moment.

The compound apparently reproduced many classic responses to calorie restriction--among them reduced tumor growth, lowered temperature and elevated levels of glucocorticoid hormones. If 2DG really could mimic many aspects of calorie restriction in animals, we thought, perhaps it would do the same for people.

While we were planning our first studies of 2DG, we learned that 2DG worked by disrupting a key enzyme involved in processing glucose in cells. The compound structurally resembles glucose, so it enters cells readily. Initial metabolism of 2DG resembles that of glucose, but subsequent metabolism is inhibited such that glucose processing essentially chokes on the intermediate compound produced from 2DG [see box on page 64].

The net result is that cells make smaller amounts of glucose's by-products, just as occurs when calorie restriction limits the amount of glucose going into cells. In essence, 2DG tricks the cell into a metabolic state similar to that seen during calorie restriction, even though the body is taking in normal amounts of food. The "metabolic stress" produced by CR or a CR mimetic forces the cellular machinery to work harder to restore ATP to required levels. Studies have shown that the mitochondria actually proliferate in response to this demand and also operate more efficiently while producing fewer dangerous by-products of metabolism.

Why might more efficient functioning of the ATP-producing machinery help combat aging? We cannot say with certainty, but we have some ideas. A long-standing theory of aging blames the production of molecules called free radicals. The lion's share of free radicals in the body are emitted as the ATP-making machinery operates. Over time these highly reactive molecules are thought to cause permanent damage to various parts of cells, including the protein complexes that generate ATP. Thus, the metabolic stress produced in the cell by 2DG and calorie restriction slow the rate at which free radicals form and disrupt cells. 2DG may alter metabolism in another way by limiting insulin secretion and thereby minimizing insulin's unwanted actions in the body.

We also suspect that cells interpret reduced levels of raw materials for the ATP-making machinery as a signal that food supplies are scarce. Cells may well respond to that message by switching to a self-protective mode, inhibiting activities not needed for cell maintenance and repair--such as reproduction--and pouring most of their energy into preserving the integrity of their parts. If that idea is correct, it could explain why calorie restriction has been shown to increase production of substances that protect cells from excess heat and other stresses.

This adoption of a self-preservation mode would mirror changes that have been proposed to occur on an organismic level in times of food scarcity. In the generally accepted "disposable soma" theory of aging, Thomas Kirkwood of Newcastle University in England has proposed that organisms balance the need to procreate against the need to maintain the body, or soma. When resources are plentiful, organisms can afford both to maintain themselves and to grow and reproduce. But when food is limited, the body invokes processes that inhibit growth and reproduction and takes extra care to preserve the soma.

Recent research has indicated another potential pathway for mimicking CR. A National Institute on Aging study led by R. Michael Anson showed that a regimen of intermittent fasting--in which mice were allowed free access to food on alternating days--resulted in beneficial effects similar to those of calorie restriction, including reduced blood glucose and insulin levels and increased resistance of brain cells to toxic stress. Surprisingly, the food intake and body weight of these mice did not diverge substantially from control mice that had unlimited access to food. These data suggest that an absolute reduction in calorie intake may not underlie all of CR's effects; rather hormonal changes related to the stress of intermittent fasting may play an important role. CR and 2DG both induce this mild stress, as indicated by higher circulating levels of the stress hormone corticosterone. Many investigators now think that this mild stress conditions the organism to better withstand even more extreme stress.

Testing Begins
IN OUR FIRST experiments on 2DG's effectiveness, we delivered low doses to rats by adding it to their feed for six months. The treatment moderately reduced fasting blood glucose levels, body weight and temperature and robustly reduced fasting insulin levels--findings consistent with the actions of calorie restriction itself. Interestingly, after an initial adjustment to the novel diet, the 2DG group did not eat significantly less food than the controls. Thus, these exciting preliminary analyses revealed that it was possible to mimic at least some sequelae of calorie restriction without reducing food intake.

Shortly after we published these results, in 1998, other groups began identifying more ways that 2DG imitates calorie restriction, including reduced heart rate and increased resistance to stress and toxins. Subsequently we initiated a long-term study of rats on a diet supplemented with 2DG. The results confirmed our previous findings that 2DG slightly reduces blood glucose and body temperature. Contrary to our expectations, however, the 2DG diet in this experiment did not extend maximum life span. Animals treated with 2DG showed better survival for the first half of the life span, but maximum life span was not extended, because of cardiac toxicity.

The work so far clearly provides a "proof of concept" that inhibiting glucose metabolism can re-create many effects of calorie restriction. Regrettably, 2DG has a fatal flaw preventing it from being a magic pill. Though safe at certain low levels, it apparently becomes toxic for some animals when the amount delivered is raised just a bit or given over long periods. The narrowness of the safety zone separating helpful and toxic doses would bar it from human use. We hope this is not a general feature of CR mimetics.

Moving On
ASSUMING OUR long-term studies confirm that inhibiting metabolism can retard aging, the task becomes finding other substances that yield 2DG's benefits but are safer over a broader range of doses and delivery schedules. Several candidates seem promising in early studies, including iodoacetate, which inhibits cellular metabolism, as 2DG does, but through a different mechanism. Treatment with antidiabetic medications that enhance cellular sensitivity to insulin might be helpful as well, as long as the amounts given do not cause blood glucose levels to fall too low. Metformin (Glucophage), which has resulted in moderate life-span extension in preliminary animal experiments, has been suggested as a possible candidate in this category.

A great deal of research implicates glucose metabolism in regulating life span, yet other aspects of metabolism can also change in reaction to calorie restriction. When the body cannot extract enough energy from glucose in food, it may shift to breaking down protein and fat. Pharmaceuticals that target these processes might serve as CR mimetics, either alone or in combination with drugs that intervene in glucose metabolism. Some compounds that act in those pathways have already been identified, although researchers have not yet assessed their potential as CR mimetics.

Drugs that replicate only selected effects of calorie restriction could have a role to play as well. In theory, antioxidant vitamins might fit that bill. Research conducted to date, however, indicates that this particular intervention probably will not extend longevity. Resveratrol, an antioxidant found in grapes and red wine, affects certain genes (the sirtuins) that may be involved in CR, at least in lower animal models. Lipoic acid, another antioxidant, is currently being used in combination with acetyl-carnitine, a metabolic efficiency enhancer, to produce some antiaging effects. In fact, this "cocktail" is now commercially available. Several companies, including GeroScience, are pursuing various CR mimetic strategies.

Unlike the multitude of elixirs being touted as the latest antiaging cure, CR mimetics would alter fundamental processes that underlie aging. Many candidate mimetics, such as resveratrol, appear to work downstream in the sequence of events that elicit the antiaging effects of CR, although glycolytic inhibitors such as 2DG come closest to targeting the "causes," rather than the "symptoms" or pathologies, of aging.

To illustrate this important point, consider that average life span has increased from about 40 years in the early 1900s to about 80 years today while maximum life span has remained unchanged at about 122 years. Much, if not all, of the increase in average life span has resulted from improved health care and nutrition. Thus, to truly "mimic" CR and alter the fundamental biology of aging, candidate compounds must be shown to increase both maximum and average life span. The goal is to devise compounds that fool cells into activating maintenance and repair activities that lead to greater health and longevity of the organism. If scientists can develop agents that offer the benefits of 2DG without its drawbacks, they will finally enable people to have their cake--a longer, healthier life--and eat it, too.

THE AUTHORS
MARK A. LANE, DONALD K. INGRAM and GEORGE S. ROTH researched calorie restriction for many years at the National Institute on Aging of the National Institutes of Health. Lane is now director of project management at Wyeth in Collegeville, Pa., and continues to collaborate with Ingram and Roth. After 26 years at the NIA, Ingram retired as chief of the laboratory of experimental gerontology and is now professor and head of the Nutritional Neuroscience and Aging Laboratory at the Pennington Biomedical Research Center of the Louisiana State University System in Baton Rouge, where he continues to conduct research in CR mimetics. Roth, who spent nearly 30 years as a full-time researcher at the NIA, is now chief executive officer of GeroScience, a biotechnology venture devoted to antiaging strategies.