In the 1960s, prisoner feeding experiments by University of Vermont physician–researcher Ethan Sims also pointed toward metabolic homeostasis. In 1967, Sims fed inmates at the Vermont State Prison upwards of 10,000 kcal per day. Over 200 days on this overfeeding regimen, 20 inmates gained an average of 20 to 25 lb.3 The metabolic rates of these previously normal-weight subjects sped up in response to their increased caloric consumption, as if to defend their initial, lower weights. The men had difficulty maintaining weight gain, and most shed all the weight they had gained relatively easily once their calorie intake returned to normal. The exceptions were two inmates who gained weight swiftly and effortlessly but then struggled to lose that weight even after caloric consumption was reduced. That both these men had family histories of obesity added empirical support to the notion that overweight could be heritable.

In 1986, the University of Pennsylvania’s Albert Stunkard offered the most compelling evidence yet that one’s weight could be largely determined by one’s parentage (1986). Stunkard and colleagues used a Danish adoption registry of 540 adults, the majority of whom had been adopted by the age of 1 between 1927 and 1947. The adoption records included the heights and weights of the adoptees’ biologic and adoptive parents. Stunkard et al. used those data to compare the body-mass indexes of both sets of parents with those of the adoptees, most of whom had reached middle age by the time of their study. They found that, despite having shared an environment with their adoptive parents, the adoptees’ body-mass indexes approximated those of their biologic parents rather than their adoptive parents. Accordingly, most adoptees inherited their biologic parents’ obesity: four fifths of those with two obese biologic parents were obese, as compared with one seventh of those with normal-weight biologic parents.

Four years later, Stunkard and another team of researchers used another twin registry, this time from Sweden, to find more support for the genetics of weight regulation (1990b). The Swedish twin registry included 247 pairs of identical twins — 154 pairs that had been raised together and 93 pairs that had been adopted by different parents. The identical twins, it turned out, had virtually the same weight regardless of whether they had grown up together or separately. As reported in another article in the same issue of the Journal, Claude Bouchard and colleagues at Laval University in Quebec had followed the effects of overfeeding on 12 pairs of adult, male identical twins over a period of 100 days (1990a). All the twins consumed the same number of calories — a total of 84,000 excess kilocalories over the course of the experiment. The subjects’ resulting weight gain ranged from 4.3 kg to 13.3 kg, with considerable variation in body-fat percentage, fat mass, fat distribution, and deposition of both subcutaneous and visceral fat among the pairs of twins. But although the responses to overfeeding varied widely among the twin pairs, within each pair of twins there was little difference in weight gain and even less difference in body-fat distribution and visceral-fat accumulation.

While twin studies and feeding experiments continued, obesity research also took a decidedly molecular turn with the discovery of the peptide hormone and satiety factor leptin in 1994. Building on the work that Douglas Coleman had been conducting at the Jackson Laboratory since the 1960s, as well as the mapping of obesity-gene mutations in mice performed by Leibel et al. in the 1980s and early 1990s, Jeffrey Friedman and colleagues at Rockefeller University cloned the gene ob that encodes leptin.4 In the years since, tens of thousands of articles have been published on leptin and related subjects, such as the hunger-stimulating hormone ghrelin, interactions between these two hormones and the neurotransmitter neuropeptide Y, and the signaling pathways of molecules involved in appetite and the genetic mutations that might interfere with these pathways.

Today, molecular genetics is central to obesity research. In 2007, Mark McCarthy, Andrew Hattersley, and their colleagues in the United Kingdom identified a common variant in FTO, the fat-mass and obesity–associated gene, and gene hunters aided by the use of next-generation–sequencing technology continue to identify gene variants or mutations such as the DYRK1Bdiscovery presented in this issue of the Journal (Keramati et al., pages 1909–1919). These studies, of course, reinforce what some physician–researchers have been insisting for more than a century: that obesity is innate, that weight regulation is not governed by a uniform tally of “calories in–calories out,” and to quote Jules Hirsch, that “there is a biochemical or basic biological element in what it is that we call `willpower.’”5 The views of many Americans notwithstanding, weight is clearly far from being entirely within an individual’s control. Genetic predispositions, in tandem with the development of food environments that facilitate overeating and built environments requiring minimal energy expenditure, may help explain why so many Americans are obese today.