Throughout most of human history, a wide girth has been viewed as a sign of health and prosperity. It seems both ironic and fitting, then, that corpulence now poses a growing threat to the health of many inhabitants of the richest nations. The measure of the hazard in the U.S. is well known: 59 percent of the adult population meets the current definition of clinical obesity, according to a 1995 report by the Institute of Medicine, easily qualifying the disease for epidemic status. Epidemiologists at Harvard University conservatively estimate that treating obesity and the diabetes, heart disease, high blood pressure and gall stones caused by it rang up $45.8 billion in health care cos ts in 1990, the latest year studied. Indirect costs because of missed work pitched another $23 billion onto the pile. That year, a congressional committee calculated, Americans spent about $33 billion on weight-loss products and services. Yet roughly 300,000 men and women were sent early to their graves by the damaging effects of eating too much and moving too little.
The problem is as frustrating as it is serious. Quick and easy solutions--liquid diets, support groups, acupressure, appetite-suppressing "aroma sticks" and even the best-intentioned attempts to eat less and exercise more--have all failed in well-controlled trials to reduce the weight of more than a small fraction of their obese adherents by at least 10 percent for five years--an achievement shown to increase life expectancy sharply.
The discovery last summer of leptin, a natural hormone that cures gross obesity when injected into mutant mice that lack it, raised hopes of a better quick fix. Those hopes have faded as subsequent studies have found no fat people who share the leptin-related mutations seen in mice. But the identification of leptin is only one of many important advances over the past several years that have opened a new chapter in the understanding of obesity.
Armed with powerful new tools in molecular biology and genetic engineering, scientists are seeking physiological explanations for some of the most puzzling aspects of the fattening of industrial society. Why is obesity on the rise, not just in the U.S. but in nearly all affluent countries? How is it that some individuals remain fat despite constant diets, whereas others eat what they want without gaining a pound? Why is it so hard to lose a significant amount of weight and nearly impossible to keep it off? Perhaps most important, what can be done to slow and eventually reverse this snowballing trend? The traditional notion that obesity is simply the well-deserved consequence of sloth and gluttony has led to unhelpful and sometimes incorrect answers to these questions. Science may at last offer better.
What Makes the World Go Round
Contrary to conventional wisdom, the U.S. is not the fattest nation on earth. Obesity is far more common on Western Samoa and several other Pacific islands. On Nauru, a mere dot of eight square miles once covered to overflowing with seabird guano, the 7,500 islanders have traded that valuable source of phosphate to fertilizer companies in exchange for one of the highest per capita incomes in the world. Many also traded their plows for lounge chairs and their traditional diet of fish and vegetables for Western staples such as canned meats, potato chips and beer. Within the course of a generation, the change has taken its toll on their bodies. By 1987 well over 65 percent of men and 70 percent of women on Nauru were obese, and one third suffered from diabetes.
Many countries, developed and developing, are heading in the same direction at an alarming pace. Changes in diet alone do not explain the trend. Surveys--some of which admittedly are of dubious accuracy--show that the proportion of calories Americans get from fat has dropped about eight points since the 1980s, to 34 percent. Yet the prevalence of obesity has risen by a similar amount in nearly the same period. Britons ate 10 percent fewer calories overall in 1991 than in 1980, according to government estimates, while the number of heavyweights doubled. Polls that show gasoline consumption and hours spent watching television rising about as quickly as the rate of obesity in some countries seem to explain part of the disparity. Evolutionary biology may provide a deeper explanation, however. In 1962 James V. Neel of the University of Michigan proposed that natural selection pressured our distant ancestors to acquire "thrifty genes", which boosted the ability to store fat from each feast in order to sustain people through the next famine. In today's relative surfeit, Neel reasoned, this adaptation has become a liability. The theory is supported by the Nauruans' plight and also by studies of the Pima Indians, a tribe whose progenitors split into two groups sometime during the Middle Ages. One group settled in southern Arizona; the other moved into the Sierra Madre Mountains in Mexico. By the 1970s most of the Indians in Arizona had been forced out of farming and had switched to an American diet with 40 percent of its calories from fat. They now endure the highest incidence of obesity reported anywhere in the world--far higher than among their white neighbors. About half develop diabetes by age 35.
Eric Ravussin, a researcher with the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), has compared Pimas in Arizona with their distant relatives in Maycoba, Mexico, who still live on subsistence farming and ranching. Although the groups share most of the same genes, Pimas in Maycoba are on average 57 pounds (26 kilograms) lighter and about one inch (2.5 centimeters) shorter. Few have diabetes. Maycobans also eat about half as much fat as their counterparts to the north, and they spend more than 40 hours a week engaged in physical work. The fact that Mexican Pimas remain lean provides strong evidence that the high rate of obesity among American Pimas is the result not of a genetic defect alone but of a genetic susceptibility--exceptionally thrifty genes--turned loose in an environment that offers easy access to high-energy food while requiring little hard labor.
Because all human populations seem to share this genetic susceptibility to varying degrees, "we are going to see a continuing increase in obesity over the next 25 years" as standards of living continue to rise, predicts F. Xavier Pi-Sunyer, director of the obesity research center at St. Luke's-Roosevelt Hospital in New York City. He warns that "some less developed countries are particularly at risk. It is projected that by 2025, more than 20 percent of the population of Mexico will have diabetes."
Studies of Pimas, islanders and migrants "all seem to indicate that among different populations, the prevalence of obesity is largely determined by environmental conditions," Ravussin concludes. A few doctors have proposed changing those conditions by levying a "fat tax" on high-calorie foods or raising insurance rates for those who fail to show up at a gym regularly.
But economic and legal punishments are unlikely to garner much popular support, and no one knows whether they would effectively combat obesity. So most researchers are turning back to factors they think they can control: the genetic and biological variables that make one person gain weight while others in the same circumstances stay lean.
Finding Genes That Fit
Doctors have long known that the tendency to gain weight runs in families--how strongly is still under debate. Numerous analyses of identical twins reared apart have shown that genetic factors alone control a large part of one's body mass index, an estimate of body fat commonly used to define obesity [see the sidebar "A Shifting Scale"]. A few have found weight to be as dependent on genes as height: about 80 percent. But the majority have concluded that genetic influences are only about half that potent.
Investigators at the National Institutes of Health who examined more than 400 twins over a period of 43 years concluded that "cumulative genetic effects explain most of the tracking in obesity over time," including potbellies sprouting in middle age. Interestingly, the researchers also determined that "shared environmental effects were not significant" in influencing the twins' weight gain. That result is bolstered by five studies that compared the body mass indexes of adopted children with their biological and adoptive parents. All found that the family environment--the food in the refrigerator, the frequency of meals, the type of activities the family shares--plays little or no role in determining which children will grow fat. Apparently, only dramatic environmental differences, such as those between the mountains of Mexico and the plains of Arizona, have much effect on the mass of a people.
Just which genes influence our eating, metabolism and physical activity, and how they exert their power, remains a mystery. But geneticists do have some encouraging leads. Five genes that can cause rodents to balloon have now been pinpointed. Tubby Mice Obese, cloned by Jeffrey M. Friedman and others at the Rockefeller University, encodes a blueprint for leptin, a hormone produced by fat cells. Mice with a mutation in this gene produce either no leptin or a malformed version and quickly grow to three times normal weight. Diabetes, cloned last December by a team at Millennium Pharmaceuticals in Cambridge, Mass., codes for a receptor protein that responds to leptin by reducing appetite and turning up metabolism. Mice with a bad copy of this gene do not receive the leptin signal, and they, too, get very fat from infancy.
Within the past year scientists at the Jackson Laboratory in Bar Harbor, Me., have cloned two other fat genes, named fat and tubby. Mice with a mutation in either of these genes put on weight gradually--more like humans do. The fat gene gets translated into an enzyme that processes insulin, the hormone that signals the body that it has been fed. But the protein produced by the tubby gene is unlike any ever seen. Researchers do not yet know why mice with errors in fat, tubby or agouti yellow, a fifth obesity gene discovered several years ago, put on extra ounces.
Although geneticists have located versions of all five genes within human DNA, "so far, when we have looked for human mutations on these genes, we haven't found them," reports L. Arthur Campfield, a research leader at Hoffmann-La Roche, the drug company that has bought the rights to Millennium's work on the leptin receptor. In fact, clinical studies by Friedman and others have shown that unlike obese and diabetes mice, heavy humans generally produce a normal amount of leptin given the amount of fat they are carrying. At least at first glance, there seems to be nothing wrong with their leptin systems. All of which is no surprise to most obesity researchers, who have long maintained that there must be multiple genes that interact with one another and with economic and psychological pressures to set an individual's susceptibility to weight gain. Although identifying clusters of interrelated genes is considerably trickier than finding single mutations, some labs have made headway in mice. David West of the Pennington Biomedical Research Center in Baton Rouge, La., has been crossing one strain that fattens dramatically on a high-fat diet with a closely related strain that remains relatively lean on the same menu. By tracking the way the trait is passed from one generation to the next, West has proved that the fat sensitivity is carried by one to four dominant genes, and he has narrowed down the chromosome segments on which they could lie. Interestingly, the tubby gene happens to rest within one of these segments.
Eventually the genes involved in human weight regulation should be found. But that is the simple part. To make a dent in obesity, physiologists will then have to figure out how all these genes work in real bodies outside the lab. The first step will be to resolve once and for all an old dieters' debate: Do we or do we not have set points--predetermined weights at which our bodies are happiest--and can they be changed?
Set up for Failure
A typical American adult gains about 20 pounds between the ages of 25 and 55. "If you figure that an adult ingests 900,000 to one million calories a year and you calculate the energy cost of those additional 20 pounds," observes Rudolph L. Leibel, co-director of the human metabolism laboratory at Rockefeller, you find that "just a few tenths of 1 percent of the calories ingested are in fact being stored. That degree of control or balance is extraordinary." [A complete transcript of Scientific American's interview with Leibel is available.]
Multiple feedback loops maintain the body at a stable weight by shunting messages through the bloodstream and the autonomic nervous system between the brain, the digestive tract, muscle--and, it turns out, fat. Until recently, fat was generally considered just a passive storage tissue. In fact, says Ronald M. Evans of the Salk Institute in La Jolla, Calif., "it is a type of endocrine tissue. Fat secretes signals--hormones such as leptin--and also monitors and responds to signals from other cells."
Last December, Evans reported his discovery of a new hormone, with the catchy name of 15d-PGJ2, that is produced inside fat cells and seems to trigger the formation of new ones, at least in children. Any drug that tried to interfere with the hormone to prevent new fat from forming would probably work only in children, Evans says, because fat cells in adults usually inflate in size rather than increase in number. But a synthetic molecule that mimics 15d-PGJ2, called troglitazone, does appear to be an effective drug for the type II diabetes associated with obesity, because it also signals muscle cells to respond normally to insulin.
In mapping the maze of intertwined pathways that control short-term appetite as well as factors (such as fat and carbohydrate levels) that change over days or weeks, researchers are slowly working out how all these signals combine to hold weight steady. Two major theories vie for acceptance: set point and settling point.
The set-point hypothesis is the older and more deterministic. It asserts that the brain continuously adjusts our metabolism and subconsciously manipulates our behavior to maintain a target weight. Although the set point may change with age, it does so according to a fixed genetic program; diet or exercise can move you away from your set point, at least for a time, but the target itself cannot change--or so the theory goes. Last year Leibel and his colleagues Michael Rosenbaum and Jules Hirsch, who are three of the strongest proponents of the set-point theory, completed a study that seems to support their hypothesis. The physicians admitted 66 people to the Rockefeller hospital. Some of the patients were obese, and some had never been overweight, but all had been at the same weight for at least six months. Over the next three months the subjects ate only precisely measured liquid meals. The doctors ran an extensive battery of tests on the volunteers and then increased the calories that some were fed and put the others on restricted diets. When the subjects had gained 10 percent or lost either 10 or 20 percent of their original weight, the tests were run again to see what had changed.
The investigation disproved some tidbits of weight-gain folklore, such as that thin people do not digest as much of their food as heavyweights. The study also found that "the idea that you will be fatter--or will require fewer calories to maintain your starting body weight--as a result of having yo-yoed down and back up again is wrong," Rosenbaum adds. Moreover, the research showed that obese people, when their weight is stable, do not eat significantly more than lean people with the same amount of muscle but less fat.
But the trial's real purpose was to determine how much of a fight the body puts up when people attempt to change the weight they have maintained for a long time--why, in other words, dieters tend to bounce back to where they started. When both lean and obese subjects dropped weight, "it seemed to set off a bunch of metabolic alarms," Leibel recalls. The subjects' bodies quickly started burning fewer calories--15 percent fewer, on average, than one would expect given their new weight. Surprisingly, the converse also seems to be true for weight gain. Even rotund people have to eat about 15 percent more than one would expect to stay very far above their set point.
That fact raises a major problem for set-point theory: How does it explain the rapid increase in the prevalence of obesity? "Clearly, set points have to be rising, just as we are getting taller in every generation," Rosenbaum says. "But set points are not changeable in adulthood, as far as we can tell. So there must be a window of opportunity sometime in childhood where the environment influences the set point," he speculates. "If you could figure out when and how that occurs, maybe you could modify the environment then, and you wouldn't have to worry about your kids getting fat 20 years down the line." That will remain wishful thinking until set-point advocates demonstrate how weight is centrally controlled. Their best guess now, explains Louis A. Tartaglia, a scientist at Millennium, is that "the body's set point is something like a thermostat"--a lipostat, some have called it--and leptin acts like the thermometer.
As you gain weight, Friedman elaborates, "you make more leptin. That shuts off appetite, increases energy expenditure and undoubtedly does other things to restore body weight to the set point. Conversely, if you get too thin, levels of leptin fall, and now you eat more, burn less, and again your weight returns to where it started. Now that we know what the gene and its product are, we can test that simpleminded theory."
Amgen, a biotechnology firm in Thousand Oaks, Calif., that has reportedly promised Rockefeller up to $100 million for the right to produce leptin, has begun injecting the hormone into obese people in clinical trials. "The goal," Rosenbaum says, "is to co-opt your body into working with you rather than against you to maintain an altered body weight" by tricking it into believing it is fatter than it is.
But the body may not be easily fooled. In May, scientists at the University of Washington reported that they had engineered mice that lack the gene for neuropeptide Y (NPY), the most powerful appetite stimulant known. Leptin curtails NPY production; this, it was thought, is how it quells hunger. But mice lacking NPY do not lose weight--something else compensates.
Critics of the set-point hypothesis also protest that it fails to explain the high rates of obesity seen in Nauruans and American Pimas. Moreover, if body fat is centrally controlled, they argue, the amount of fat in your diet should have little impact on your weight. Numerous studies have found the contrary. One recent survey of some 11,600 Scotsmen observed that obesity was up to three times more common among groups that ate the most fat than among those who relied on sugars for most of their energy.
Fat in the Balance
At a conference last year, researchers reviewed the evidence and judged that although the set-point hypothesis has not been disproved, there is more "biological merit" to the idea of a "settling point." This newer theory posits that we maintain weight when our various metabolic feedback loops, tuned by whatever susceptibility genes we carry, settle into a happy equilibrium with our environment. Economic and cultural changes are upsetting this equilibrium and propelling more people--those with more genetic risk factors--into obesity.
The prime culprit suspected in this trend is hardly surprising: it is the fat dripping off hamburgers, smoothing out ice cream and frying every meat imaginable. But biochemists are at last working out precisely why fat is bad. For years, they have known that people fed a high-fat meal will consume about the same amount as those given a high-carbohydrate meal. Because fat has more calories per bite, however, the subjects with greasy grins tend to ingest more energy than they can burn, a phenomenon known as passive overconsumption.
One reason for this, according to biopsychologist John E. Blundell of the University of Leeds, seems to be that the systems controlling hunger and satiety respond quickly to protein and carbohydrates but slowly to fat--too slowly to stop a high-fat meal before the body has had too much. Metabolic systems seem to favor carbohydrates (which include sugars and starches) as well. Knock down a soda or a plate of pasta, and your body will soon speed up its carbohydrate combustion. Polish off a bag of pork rinds, however, and your fat oxidation rate hardly budges, points out Jean-Pierre Flatt, a biochemist at the University of Massachusetts Medical School. Most incoming fat is shipped directly to storage, then burned later only if carbohydrate reserves dip below some threshold, which varies from person to person.
There is another way to increase the rate at which fat is burned for energy: pack on the pounds. More fat on the body yields more fatty acids circulating in the bloodstream. That in turn boosts fat oxidation, so that eventually a "fat balance" is reached where all the fat that is eaten is combusted, and weight stabilizes. Many genetic and biological factors can influence the fat oxidation rate and thus affect your settling point in a particular environment.
Olestra, an artificial fat approved earlier this year by the Food and Drug Administration, may change that rate as well. Olestra tastes more or less like an ordinary fat, but it flows undigested through the body. A preliminary study by George A. Bray, Pennington's executive director, suggests that the ingredient may short-circuit passive overconsumption. For two weeks, Bray replaced the natural fat in his subjects' meals with olestra. "They did not compensate at all by eating more food," he reports, adding that "it remains to be seen whether that holds up in longer-term studies."
The fat balance explains in part why settling points vary among people who overeat fat: some oxidize fat efficiently at normal weights; others burn too little until excess pounds force the oxidation rate up. But the model does not by itself explain why some do not overeat at all. To answer that, Flatt has proposed a "glycogen hypothesis."
The human body can store about a day's supply of carbohydrates in the form of glycogen, a simple starch. Glycogen reserves function somewhat like fuel tanks; we partially refill the stores with each meal but rarely top them off. In fact, the range between "empty" and "full" appears to be a matter of individual preference, influenced by such factors as the diversity and palatability of food at hand, social pressures and meal habits. People who are content with lower glycogen levels or who frequently deplete them through exercise burn fat more readily than those who like to keep their tanks full, Flatt suggests. But he concedes that the "crucial link from glycogen stores to appetite remains to be proven."
Researchers need more evidence before they can pronounce either set point or settling point--or neither--correct. James O. Hill of the University of Colorado Health Sciences Center has begun collecting some of those critical data. He is assembling a registry of the most precious resource in obesity research: the people who have lost a large amount of weight and kept it off for several years without relapse. Hill has already identified about 1,000 such individuals and has begun examining a handful for biochemical clues to their success.
Unfortunately, no current explanation of weight regulation leaves much room for voluntary control; all the metabolic cycles involved are governed subconsciously. Settling-point theory does at least suggest that sufficiently drastic changes in lifestyle might prod the body to resettle at a new weight. But without assistance, changes radical enough to make a difference are evidently uncomfortable enough to be infeasible--for millions of dieters have tried this strategy and failed.
Getting over the Hump
Increasingly, obesity researchers argue that the most effective assistance they can provide their patients will probably be pharmacological. "The treatment philosophy of the past 40 years, which has been to train patients to eat differently, is simply not going to cure the epidemic of obesity that we see worldwide," asserts Barbara C. Hansen, director of the obesity research center at the University of Maryland School of Medicine.
Untangling the biology beneath body fat has created a plethora of new drug targets that has drawn dozens of pharmaceutical firms off the sidelines [see sidebar "A Spoonful of Medicine: Obesity Drugs under Development"]. The potential market is enormous, not only because obesity is common and growing but also because even an ideal drug will have to be taken indefinitely, according to Hansen and others. "Obesity isn't curable," Bray says. "It's like high blood pressure. If you don't take the medication, your blood pressure won't stay down. And if you don't take drugs--or do something--to treat obesity, your weight won't stay down."
Part of the reason for the resurgence of commercial interest is a shift in policy at the FDA, which decided in April to allow the appetite suppressant dexfenfluramine to be prescribed for obesity in the U.S., as it already is in 65 other countries. It is the first weight-loss drug approved in the U.S. in 23 years, and nearly all obesity researchers agree it has been too long coming. The FDA also recently relaxed its guidelines for obesity-drug applications. "As our compromise right now, we're suggesting that a company can present us with two years of data--in some cases, one year if the data look good enough and the company gives us a firm commitment to do follow-up studies under tight controls," says Leo Lutwak, a medical officer with the FDA's Center for Drug Evaluation & Research.
Lutwak admits that with only two years of information, the FDA may approve drugs that turn out to have serious long-term side effects. "The best we can hope for is something like insulin for the treatment of diabetes," Leibel says. Insulin rescues a type I diabetic by replacing a hormone that is missing. "But after 15 years, you begin to have complications of our inability to perfectly mimic the biology," Leibel continues. "If we're lucky, that's the kind of problem we'll face in the treatment of obesity." Lutwak responds that "when that happens, the public will be informed, and they will have to make a decision about whether it is worth it."
If the long-term cost of treatment is unknown, the benefits are becoming clearer, thanks to studies on people who have an operation, called gastroplasty, that reduces the size of the stomach. Although infrequently used in the U.S., the procedure has proved remarkably effective in Sweden. A long-term study there of 1,150 obese patients who underwent gastric surgery found that they typically dropped 66 pounds over two years--88 pounds if a more severe procedure was used--whereas control subjects given standard dietary treatment lost nothing. The surgery cured more than two thirds of those with diabetes, compared with 16 percent cured in the control group. Likewise, twice as many (43 percent) of the hypertension cases were cured by the operation.
Gastroplasty has drawbacks in addition to the risks that always accompany major surgery--principally a high rate of digestive complications. Drug treatments might be better, but Hansen's work with rhesus monkeys suggests that prevention would be best. A decade ago her team began a trial on young adult monkeys, equivalent in maturity to 20-year-old men. The researchers adjusted the animals' food supply so that they neither gained nor lost weight. "In the past 10 years we have had 100 percent success preventing both obesity and type II diabetes," Hansen asserts. "In the control group, which was simply allowed to feed freely on the same diet, half are diabetic. Because everything we know about human obesity is also true of nonhuman primate obesity, that shows you the power of weight control."
It does not, unfortunately, demonstrate a feasible way to achieve it. The NIDDK has launched a program to educate Americans about ways to avoid weight gain, but Susan Z. Yanovski, the program's director, admits that so far it has had little perceptible impact. There is no major lobbying organization for the disease, notes Pi-Sunyer, and the NIH directs less than 1 percent of its research funding at obesity. "Many people seem to be unaware of how big a health problem this is now and how big it is going to grow, particularly when you look at the increasing obesity of children," Yanovski says. Because obese adolescents usually become fat adults, "we're really heading for trouble in another 20 to 30 years," she adds.
At least one grade school intervention has had modest success, knocking a
few percentage points off the number of children who turn into overweight adolescents
by taking fat out of the children's lunches, giving them more strenuous recreation
and educating their parents about weight control. "We have to be very careful
about putting children on restrictive diets," Yanovski warns. "That is inappropriate.
But we can be more proactive in getting our kids away from the television set,
more physically active, riding their bikes instead of being driven everywhere.
If people recognize that this is a serious public health problem affecting their
children, then maybe they will start taking some action." If not, economists
should start adjusting their models now to account for the tremendous health
care cost increases that lie ahead.