Would a rat risk dying just to satisfy its desire for chocolate?
A few years ago I found out. In my laboratory, we gave rats unlimited access to standard chow as well as to a mini cafeteria full of appetizing, high-calorie foods: sausage, cheesecake, chocolate. The rats decreased their intake of the healthy but bland items and switched to eating the cafeteria food almost exclusively. They gained weight. They became obese.
We then warned the rats as they were eating—by flashing a light—that they would receive a nasty foot shock. Rats eating the bland chow would quickly stop and scramble away, but time and again the obese rats continued to devour the rich food, ignoring the warning that they had been trained to fear. Their hedonic desire overruled their basic sense of self-preservation.
Our finding mirrored a previous trial conducted by Barry Everitt of the University of Cambridge—only his rats were hooked on cocaine.
So are the fat rats addicted to food? An inability to suppress a behavior, despite the negative consequences, is common in addiction. Scientists are finding similar compulsiveness in certain people. Almost all obese individuals say they want to consume less, yet they continue to overeat even though they know that doing so can have shockingly negative health or social consequences. Studies show that overeating juices up the reward systems in our brain—so much so in some people that it overpowers the brain's ability to tell them to stop eating when they have had enough. As with alcoholics and drug addicts, the more they eat, the more they want. Whether or not overeating is technically an addiction, if it stimulates the same brain circuits as drug use, in the same way, then medications that dial down the reward system could help obese people to eat less.
Until the early 1990s, society viewed obesity solely as a behavioral disorder: overweight individuals lacked willpower and self-control. Since then, the view has changed dramatically, in the scientific community at least.
The first change in opinion arose from pioneering work by the late Douglas Coleman of Jackson Laboratory in Bar Harbor, Me., and by Jeffrey Friedman of the Rockefeller University. Experiments with two strains of mice, both genetically prone to obesity and diabetes, determined what drove the mice to overeat. The researchers discovered that one strain had a genetic defect in fat cells that secrete a hormone called leptin. Mice, like humans, normally secrete leptin after a meal to suppress appetite and prevent overeating. The obese mice had a leptin deficiency—and an insatiable appetite. Later the researchers found that obesity in the second strain of mice was caused by a genetic defect in their ability to respond to leptin and regulate its actions. The findings seemed to make it clear that hormones regulate appetite and therefore body weight. A hormonal imbalance could lead to overeating; indeed, obesity runs rampant in certain human families that have a genetic deficiency in leptin.
Two observations suggest that viewing obesity as a hormone disorder is too simplistic, however. First, only a small number of obese people in the U.S. and elsewhere have a genetic deficiency in appetite-related hormones. Second, if obesity is driven purely by hormonal imbalances, then blood tests of obese people ought to show either a lower level of hormones that suppress appetite or a higher level of hormones that increase appetite. Yet the reverse is true. Obese individuals generally have a paradoxically high level of appetite-suppressing hormones, including leptin and insulin.
This is where the concept of food addiction comes into play. Appetite-controlling hormones affect certain pathways of neurons—we call them feeding circuits—in the hypothalamus. These hormones also affect systems in the brain that control feelings of reward, which makes sense. If you have not eaten for many hours, you will spend a lot of time, effort and money to obtain food—and it will taste very good! As the old adage says, “Hunger is the best sauce.”
During periods of hunger, hormones heighten the reactivity of food-related reward circuits in the brain, particularly in the striatum. The striatum contains high concentrations of endorphins—chemicals that enhance feelings of pleasure and reward.
As you eat, your stomach and gut release appetite-suppressing hormones that decrease pleasure signals that are triggered by the striatum and other components of the reward system. This process makes food seem less attractive, and you may switch your activity away from eating and toward other pursuits. The suite of appetite-regulating hormones controls feeding, in part by modulating the pleasurable experience of consuming a meal.
Yet some modern, appetizing foods—especially those that are fatty, sugary andvisually appealing—affect reward systems strongly enough to override the appetite-suppressing hormones, thus prompting us to eat when we would not otherwise feel hungry. These foods activate our reward circuits more powerfully than leptin's ability to shut them down. All of us have experienced this effect: you have just finished a big dinner and could not possibly eat another bite. Yet when the chocolate cake appears, you can miraculously “find room” for one last morsel—one that happens to be the most calorie-laden of the day.
Therein lies the rub. We have evolved an efficient brain system to help maintain a healthy and consistent body weight by signaling when it is time to eat and when it is time to stop. But highly appetizing foods can often veto these signals and drive weight gain.
Our body responds to the override by elevating the blood levels of the appetite-suppressing hormones higher and higher as body weight increases, yet the hormones become progressively less effective as the body build ups a tolerance to their actions. Moreover, brain-imaging studies by researchers at Brookhaven National Laboratory and the Oregon Research Institute show that the brain's reward systems in overweight individuals may actually respond less strongly to food—even to junk food—than they do in people of healthy weight. These muffled reward circuits depress mood. How does an individual overcome this funk? By eating more delectable food to gain a temporary boost, thereby perpetuating the cycle. The obese, in other words, may have to overeat just to experience the same degree of pleasure that lean people enjoy from less food.
Obesity, it seems, is not caused by a lack of willpower. And it is not always caused by an imbalance in hormones. In some cases at least, obesity may be caused by hedonic overeating that hijacks the brain's reward networks. Like addictive drugs, overeating creates a feedback loop in the brain: the more you consume, the more you crave, and the harder it is to satisfy that craving.
But does that make hedonic eating an addiction?
Tolerance and relapse
Drugs of abuse, such as morphine, stimulate the brain's reward systems in the same way that food does. Yet the similarities do not end there. When morphine is injected into the striatum of rats, it triggers bingelike overeating, even in rats that have been fed to satiety. This response shows that morphine and other opiates mimic the effects of neurotransmitters (brain chemicals) such as endorphins that are naturally produced in the brain to stimulate feeding behaviors.
We might expect, then, that drugs that block the action of endorphins could reduce hedonic overeating. Recent studies have shown that endorphin blockers do lessen the activation of reward circuits in humans and rodents that are presented with appetizing food, and the subjects eat less. The blockers can also reduce heroin, alcohol and cocaine use in human drug addicts, supporting the idea that common mechanisms regulate hedonic overeating and addictive drug use. Strikingly, rats that are allowed to binge on food every day and then are treated with endorphin blockers display behaviors that closely resemble withdrawal, a symptom of drug addiction. That result suggests that hedonic overeating can induce a brain state akin to drug dependence.
These discoveries add credence to the idea that overeating in some circumstances may share core features of drug addiction. So do similarities that we see with another basic neurotransmitter: dopamine. All known addictive drugs lead to the release of dopamine into the striatum. Dopamine is central to motivation; it spurs people to seek drugs. Most experts maintain that this action drives the development of addiction, although the precise mechanisms are hotly debated. It turns out that appetizing food also stimulates the release of dopamine into the striatum, motivating people to focus on obtaining and consuming food.
Paradoxically, obese rats and mice produce very low levels of dopamine, probably because their brain reward systems adapt to constant overstimulation by appetizing food. Lower dopamine levels in obese individuals may drive overeating just to obtain the same level of pleasure from food as that derived by lean individuals. In exciting work from a group led by Ivan de Araujo of Yale University, mice fed a high-fat diet were found to have lower circulating levels of a fat called oleoylethanolamine, which is produced by the gut. Remarkably, when this compound was injected into overweight mice, levels of dopamine returned to normal, and the mice ate more low-fat food and less high-fat food.
Imaging studies have illuminated other connections between eating and addiction. The striatum of obese individuals shows low levels of a receptor that responds to dopamine, the dopamine D2 receptor (D2R). Those suffering from alcoholism or from opiate, cocaine or methamphetamine addiction similarly tend to have low levels of D2R. And we now know that people who are born with a D2R deficiency are at greater genetic risk of developing both obesity and drug addiction. The condition results in lower levels of activity in the brain's reward systems, suggesting that overeating is just a coping behavior to obtain normal pleasure from eating. These people also tend to have trouble learning to avoid actions that have negative consequences; brain systems involved in suppressing risky yet rewarding behaviors, such as consuming high-calorie food or using drugs, may not work as effectively.
Our lab study of rats backs up this idea. The obese rats that ate the cafeteria food, ignoring warnings about being shocked, had low levels of D2R in the striatum. These experiments and others demonstrate that drug use in addicted rats and hedonic eating in overweight rats persist even when the animals face negative consequences. Many obese individuals struggle so badly with their poor food choices that they voluntarily undergo risky procedures, such as gastric bypass surgery, to help them control their eating. Yet very often they relapse to overeating and gain weight.
This cycle of engaging in a bad habit that gives short-term pleasure and then attempting to abstain from it and eventually relapsing sounds disturbingly like drug addiction. Given the latest research, it seems that obesity is caused by an overpowering motivation to satisfy the reward centers—the pleasure centers—of the brain. The hormonal and metabolic disturbances in obese individuals may be a consequence of weight gain rather than a cause.
New treatments possible
The similarities between obesity and addiction have led some experts to say that the two conditions should be treated in the same manner—in particular, that obesity be included in the most recent update to the Diagnostic and Statistical Manual of Mental Disorders—the bible of psychiatry that provides guidelines for diagnosing mental illnesses, known as the DSM-5. This proposal sparked lively debate among neuroscientists and psychiatrists, but arbiters for the DSM-5 ultimately dropped the idea, largely to avoid obese people being labeled as mentally ill.
Caution may have been warranted because despite the parallels, obesity and addiction differ in important ways. For example, if food is addictive, then surely it must contain some unique component that drives the addiction—the nicotine of junk food, if you will. Work by Nicole Avena, now at the Icahn School of Medicine at Mount Sinai, the late Bartley Hoebel of Princeton University and others lends some credence to the idea that particular fats or sugars may be responsible. A small study by David Ludwig of Boston Children's Hospital suggests that highly processed, quickly digested carbohydrates could trigger cravings. But research overall indicates that no one ingredient stokes addictionlike behaviors. Rather the combination of fats and sugars, together with calorie content, seems to maximize food's “hedonic impact.”
Other experts, including Hisham Ziauddeen, I. Sadaf Farooqi and Paul C. Fletcher of the University of Cambridge, have argued that tolerance and withdrawal do not occur in obese people in the same way that they do in drug addicts—obesity and drug addiction, they say, are fundamentally different.
This view is debatable. If obese individuals must eat more and more to overcome reduced activation of reward networks in the brain, that sounds a lot like tolerance. And weight loss can trigger negative mood and depression, much like that experienced by former addicts who try to practice abstinence, suggesting that withdrawal may be in effect.
Another objection experts have raised is that it is preposterous to apply the concept of addiction to eating because we are all, in a sense, addicted to food. If we were not, we would not survive.
The difference in obesity, I would suggest, is that modern high-calorie foods can overwhelm our biological feedback networks in a way that other foods do not. During millions of years of evolution, the major concern of humans was not suppressing appetite but hunting, collecting or growing enough food to persist in lean times. Perhaps our feeding circuits are better at motivating food intake when we are hungry than they are at suppressing food intake when we are full. It is easy to imagine that the brain would regard overeating of high-calorie food as tremendously beneficial if it is uncertain when food will again be available. But this behavior is no longer adaptive and can be counterproductive in a world where food is bountiful.
Scientists have made fair points in arguing against an addiction model of obesity. I, too, fear that the term “addiction” comes loaded with unhelpful preconceptions. Still, compulsive eating and compulsive drug use seem to share obvious features, most notably an inability to control consumption. It is up to scientists to determine whether these similarities are superficial or stem from common alterations in the brain. More important, we must determine whether the addiction model is useful. Unless it helps us design new treatment approaches, the debate is simply an academic exercise.
For an addiction model to have value, it should make accurate predictions about treatment options, including new medications. One example comes from Arena Pharmaceuticals, which obtained approval from the U.S. Food and Drug Administration in 2012 to market a drug called Belviq (lorcaserin) for weight loss in obese or overweight adults. The drug stimulates a brain protein, the serotonin 2C receptor, that is involved in satiety.
Another drug is rimonabant, which had been approved in Europe to help curb appetite in obese individuals. The drug exploits the well-known property of cannabis to increase desire for food—the so-called munchies. Cannabis activates a brain protein called the cannabinoid receptor 1, so researchers reasoned that inhibiting that receptor would decrease desire for food. Rimonabant does exactly that. A notable side effect is its ability to decrease tobacco users' desire to smoke. In rats, the drug also decreases the desire to use alcohol, opiates and stimulants such as cocaine.
As with all new drugs, however, caution is required. Both Belviq and rimonabant have triggered depression in some individuals. Rimonabant was linked to an increased risk of suicidal thoughts. This finding led European authorities to remove rimonabant from the market and prompted U.S. officials to not approve it. Why depression emerged is still unclear. Although an addiction model of obesity may yield new treatments, they must be thoroughly scrutinized.
Before scientists can declare that overeating is or is not an addiction, they will have to identify precisely which networks and cellular adaptations in the brain drive compulsive drug use and then determine if the same mechanisms also motivate compulsive food intake. It is likely that addiction networks for cocaine and for food operate in different parts of the brain yet use similar mechanisms. For example, recent work on compulsive sugar eating has implicated projections that connect the lateral hypothalamus to the midbrain neurons that synthesize dopamine. But it is not clear yet whether this circuit plays a similar role in drug seeking.
Scientists are also stuying whether common genetic variations, such as those that affect D2R, contribute to drug addiction and obesity. Identifying such genes may reveal new targets for medications to treat both disorders.
Even if researchers eventually prove that obesity sometimes arises from an addiction to food, and we find that antiaddiction medications can help people lose weight, obese individuals will continue to struggle with one factor that seems now to be endemic in America: they will probably be surrounded by overweight family members, friends and co-workers who are still overeating, putting them in the same difficult environment they were in before. As we know from recovering drug addicts and alcoholics, environmental cues are a major cause of craving and relapse. Western society, saturated in fat and temptation, will make it hard for any obese person to quit.