The Enormity of Obesity

PAST AND PRESENT

There is no question that profound genetic influences affect weight regulation. Humans who lack leptin, an appetite-suppressing hormone produced by fat cells, or melanocortin-4 receptor, which is downstream of leptin in the hypothalamus, have insatiable appetites. "Their phenotype is dominated by their hyperphagia," says O'Rahilly. "I mean these kids just eat like crazy."

While such debilitating mutations are very rare, more subtle predispositions to plumpness are not. The problem is that we owe our genetic make-up to the selection pressures exerted by a once kill-or-be-killed environment, very different from the one that operates now. Humans are designed to put on weight rather than lose it, and the thrifty nature of our genes encourages us to take full advantage of modern life's tasty trappings. It's "an obesogenic environment," says Bloom. "I think it's a tribute to our self-control that we're only this obese."

This evolutionary necessity of maintaining body weight also might shed some light on the apparent complexity of the regulatory system. "It is very important to eat to survive," says O'Rahilly, "so it's likely that we'll have put in multiple fail-safe mechanisms to try and ensure that we do continue to eat." In other words, the pathways involved in appetite regulation watch out for each other; if one is cut off, another takes over.

Tschöp cites the gut hormone cholecystokinin as an example of appetite regulation. It can be given to induce satiety in the short term, but in the long-run, caloric intake returns to where it was. Tschöp says that other specific examples of such physiological redundancy have proved elusive. "Everybody is talking about redundancy," he says, "but where are the actual data?"

Trying to integrate the plethora of weight regulatory pathways is "still premature," says Seckl, "but you can begin to come up with some principles." Indeed, one pathway has emerged as being of central importance.

FAT OF THE GLAND

©2004 Elsevier Science

Leptin acts directly on arcuate nucleus neurons coexpressing NPY and AgRP, and POMC and CART, via the ObRb form of the leptin receptor expressed on these cells. The former neurons stimulate anabolic and orexigenic effects and are suppressed by leptin; the latter neurons stimulate catabolic and anorexic actions, promoting weight loss. Downstream, these neurons target neurons expressing melanocortin 4 receptors that are activated by the POMC product α MSH, and inhibited by the neuropeptide AgRP. Activation of these neurons promotes catabolism by reducing food intake and increasing energy expenditure. (Reprinted from Cell, 116:337–50, 2004.)

Leptin, an appetite-suppressing hormone secreted by fat cells, is "a truly critical player in the field," says Cummings. Its discovery in 1994⁴ is widely credited with starting this golden age of research. "If you asked people about adipose tissue 10 years ago, everyone said, well, it's a tedious insulator and store for fat. And that was the sum of world knowledge," says Seckl. "Now we have a dozen new hormones which come out of it."

Leptin receptors were soon identified in two populations of neurons in the arcuate nucleus of the hypothalamus. One population produces the appetite-stimulating neuropeptides NPY (neuropeptide Y) and AgRP (agouti-related protein), which are suppressed by leptin, while POMC (proopiomelanocortin) and CART (cocaine- and amphetamine-related transcript) neurons suppress appetite and are activated by leptin. As fat stores are reduced, so is the amount of circulating leptin, causing appetite to increase.

Hopes that leptin could be used to suppress appetite pharmacologically were dashed, however, when it was found that most obese people have high levels of the hormone; they are resistant, rather than deficient. The mechanism for leptin resistance remains unresolved, but a new study from William Banks at the Veteran Affairs Medical Center, St. Louis shows that triglyceride fats, which are elevated in obesity, impede the transport of leptin across the blood-brain barrier, preventing it from reaching its receptors in the hypothalamus.⁵⁶ All the known monogenic mutations (leptin, leptin receptor, POMC, prohormone convertase 1, melanocortin 4 receptor) that lead to hyperphagia in humans are in the leptin-hypothalamic system (see diagram at left).

The melanocortin 4 receptor is a particularly hot target because it is locally expressed in the brain. Cummings explains that current anti-obesity drugs are relatively nonspecific. Sibutramine (Meridia), for example, tinkers with serotonin levels. "Manipulating them is fraught with side effects," he says, "and those limit the doses you can use."

The identification of another system involving the hypothalamus has led to the development of one of only two potential anti-obesity drugs currently in Phase III clinical trails. The positive effects of marijuana on appetite have long been known, and in the 1990s, a receptor for endogenous cannabinoids was identified in the hypothalamus. Sanofi-Synthelabo has developed a blocker for the cannabinoid receptor CB1, called Rimonabant, and trials have revealed spectacular effects on reducing body weight, waist size, and normalizing lipid metabolism.⁷

O'Rahilly points out, however, that more people withdrew from treatment than from placebo groups, so "there were clearly some side effects." An obvious concern might be that blocking a receptor linked to a mood-altering drug might have negative psychological consequences. Yet Marie-Camille Borde, who heads Sanofi-Synthelabo's Rimonabant research group, says that depression indices reveal no difference between treatment and placebo groups.

BLOOD AND GUTS

Although the current emphasis is on appetite regulation, it is theoretically possible to target the energy equation's other side, by manipulating energy expenditure. Drugs that raise metabolism, such as beta 3 agonists and modified forms of thyroxin, have been tried before but were only marginally effective and/or toxic. Dinitrophenol, for example, worked by disrupting oxidative phosphorylation in mitochondria. "Very effective," says O'Rahilly, "but it killed people."

Less distinction exists between metabolic and appetite targets than was once thought, says O'Rahilly. Many hypothalamic targets, he says, seem to affect both. Indeed, "It's hard to find one that doesn't." Low levels of leptin induce other physiologic systems such as reproductive and immune functions to conserve calories when energy reserves are low.

Another approach might be to prevent fat formation in the first place. Angiogenesis inhibitors prevent blood vessel formation; they have been developed with a view to treating cancer by starving tumors. But as adipose volume increases, new blood vessels are required to supply it. Maria Rupnick of Boston's Brigham and Women's Hospital has shown that not only can leptin-deficient ob/ob mice be prevented from putting on weight through treatment with angiogenesis inhibitors, but they also can be induced to lose fat-mass.⁸

There are concerns, however, about the long-term effects of preventing fat formation. "If you do not have adipose tissue in which to put those extra calories, where are you going to put them?" asks O'Rahilly. "You are probably going to put them where they shouldn't be, in the liver, muscle, pancreatic beta cell. And we know that's very bad." Rupnick points out that no adverse health consequences have been seen so far in experimental animals, and she suggests that the excess calories might be accounted for by an increase in metabolic rate.

BETWEEN MEALS

While the leptin system is regarded as a long-term regulator of energy balance, other pathways are under scrutiny for their effects on short-term, meal-to-meal appetite regulation. The intestine produces PYY after a meal, inhibiting NPY in the arcuate nucleus. It is a particularly attractive target because it appears to have a causal role in obesity. Bloom's group has found that it reduces appetite in mice⁹ and humans.¹⁰ Furthermore, PYY is underexpressed in obese patients, and this can be corrected, in the short-term at least, by treatment with PYY. "A peripheral injection of gut hormone produces a blood level rise and acts exactly as if you'd eaten a meal," says Bloom, "So it's not going to be at a risk of massive side effects, because that's what happens every time you eat a meal." However, some controversy surrounds the rodent work owing to a "widespread inability of people to replicate the results," says Cummings. Bloom says that he is about to start longer-term trials on human subjects.

Further up the digestive tract, ghrelin, secreted in the stomach, has been dubbed the hunger hormone for its stimulatory effect on appetite. Cummings' group has shown that in humans, circulating ghrelin levels rise before meals and fall rapidly after eating.¹¹ Ghrelin also increases after dieting, suggesting that it is one way the body fights against weight loss.

"In the beginning, we thought, well, it's a hunger hormone from the stomach, [so] it does something in the brain. Let's block it and we have an obesity drug," says Tschöp, who was part of the Eli Lilly group that first identified the peptide's appetite-stimulating effects. But, just as leptin levels are high in obesity, ghrelin levels turned out to be low. Furthermore, ghrelin knockout animals are of normal weight. Blocking the remaining ghrelin in obese patients might still prove to help weight loss, but progress is hampered by the lack of suitable ghrelin receptor antagonists.

LEPTIN, UNPLUGGED

Meanwhile, leptin continues to surprise. Two recent papers show that the hormone can rewire neuronal circuitry in the hypothalamus.¹²¹³ "You know those old movies where somebody at a telephone exchange center connects people by unplugging and plugging cables? That's pretty much how that works," explains Tschöp.

Leptin exposure suppresses excitatory inputs to NPY neurons and promotes them in POMC neurons. Ghrelin also alters these inputs, but in a direction opposite to leptin. Disruptions to the wiring in ob/ob can be corrected in mice treated with leptin.¹³ Neuronal connections between the arcuate nucleus and other hypothalamic centers also are influenced by exposure of the developing brain to leptin.¹² According to Rockefeller University's Jeffrey Friedman, who discovered leptin in 1994 and was involved in this work,¹³ "The wiring diagrams of this system ... might be subtly, or not so subtly, different between the lean and the obese." Jeffrey Flier of the Beth Israel Deaconess Medical Center, Boston, says it might be possible "to forcefully rewire in an appropriate manner, but that's much further down the line."

Until then, there's always surgery. Gastric bypass might seem a drastic solution to the battle of the bulge, but it is effective: Weight comes off, and it usually stays off. Appetite is suppressed, which is likely linked to observations that postoperative levels of PYY and ghrelin are high and low, respectively.

©2004 Elsevier Science

involves long-term afferent signals from fat (leptin) and pancreatic β cells (insulin) and short-term, meal-related afferent signals from the gut, including inhibitors of feeding (PYY, GLP-1, and CCK) and the stimulator of feeding (ghrelin). Efferent outputs regulate appetite, energy expenditure, hormonal milieu, energy partitioning, and the status of reproduction and growth. (Reprinted from Cell, 116:337–50, 2004.)

It's an expensive procedure, although Cummings says it pays for itself in the long run because it removes the need to take many of the costly medicines that obese people require. For example, many obese people have diabetes, and 83% of those are cured wholesale, often while still in the recovery room, says Cummings. "It's a kind of a mystical thing."

Cummings predicts that surgery will be outmoded as more effective medicines are developed. He suggests that in the same way that hypertension is treated with multiple classes of drugs working on many different sites, the same will evolve for excess body weight over the course of the next couple of decades.

Tschöp prefers a parallel with interventions in reproductive physiology. He points to the use of progesterone to trick the brain into believing that a woman is pregnant. "I'm sure that ... we'll be able to target the exact, essential mechanisms and trick the brain into believing the stomach is already full." He adds, "Maybe we don't yet know the progesterone of energy balance."

Stuart Blackman stuart.blackman@talk21.com is a freelance writer in Edinburgh, UK.