Barry E. Levin, MD, professor, Department of Neurology and Neurosciences, UMDNJ-New Jersey Medical School (NJMS), and assistant chief of neurology, East Orange VA Medical Center (right) and Ambrose Dunn-Meynell, PhD, adjunct assistant professor, Department of Neurology and Neurosciences, NJMS
Obesity: Brain Over Body
here is an obesity epidemic in the developed world that is associated with a number of other life-shortening medical illnesses. Obesity places individuals at risk for type 2 diabetes mellitus, hyperlipidemia, hypertension, heart disease, stroke, cancer, and respiratory and musculoskeletal disorders. From 1990 to 2005, its U.S. prevalence increased from 9.9 to 21.9%. In 2005, 37% of New Jerseyans were overweight and 22% were obese. More distressing is the doubling of the number of 6-to-12-year-olds and the tripling of the number of 13-to-19-year-olds who were overweight over this period. Even worse, New Jersey has the highest incidence in the nation of obesity in low-income children age 2 to 5 years. Treatment of obesity-related diseases cost New Jersey $2.3 billion in 2003 and these costs are increasing by 36% every year. Yet despite these alarming statistics, very little is known about the underlying causes of obesity or about the effective means to treat it.
Over the past 25 years, we have searched for the biological factors that predispose individuals to become obese and for the underlying reasons that so few can maintain weight loss after they become obese. Using rats bred for their genetic predisposition to become obese or remain lean on high calorie diets, we have found that the obesity-prone rats are born with a raised threshold for sensing hormonal signals from the body’s fat stores (leptin) and pancreas (insulin), which normally act on specialized neurons in the brain to reduce food intake and increase energy expenditure. Thus, they overeat until their increased obesity results in high enough leptin and insulin levels to activate these neurons. This raised threshold for responding to such signals may also underlie the inability of such individuals to maintain weight loss once they become obese.
Most importantly, raising either obesity-prone or obesity-resistant pups with genetically obese mothers causes the pups to become even more obese as adults and to alter the development and function of neural pathways which regulate food intake and energy expenditure. Given the discouragingly low rate of success in treating obesity by non-surgical means, our work suggests that identification of the perinatal factors which promote obesity in developing youngsters may offer the best long-term solution for stemming the increasing obesity epidemic.
Why do some of us get fat?
A genetically inherited predisposition underlies about two thirds of obesity in humans. However, most will not become obese without help from their environment. Our society has conveniently provided such a fertile environment — a plethora of cheap, highly palatable, calorically dense foods obtainable with little physical activity. Lean individuals appear to regulate energy intake, expenditure and storage (energy homeostasis), and body weight very well over long periods. However, some individuals have a “thrifty genotype” which allows them to slip beyond the bounds of this tightly regulated system and to eat and store more calories as fat when food is readily available. This is a useful strategy in feral animals during the summer and fall in anticipation of oncoming winter or in hunter-gather societies where food is only intermittently available. It should confer a competitive survival advantage under those circumstances. However, those conditions no longer exist for most humans in developed countries.
Over the last 25 years, we have searched for the mechanisms by which this hypothetical thrifty genotype affects body weight regulation. We originally started with a strain of outbred rats in which individuals shared many but not all genes. When fed a calorically dense diet, approximately half overate and became obese. The rest regulated their intake appropriately and remained lean. We selectively bred the highest and lowest weight gainers from this outbred population to produce genetically distinct substrains of obesity-prone and obesity-resistant rats. The obesity-prone rats and some obese humans share several common characteristics. Their obesity is inherited in a polygenic fashion and they become obese and develop insulin resistance, hypertension and hyperlipidemia when fed a calorically dense diet. Most importantly, once obesity develops, caloric restriction causes weight loss but the vast majority regain all of their lost weight once food is again readily available.
There is no simple explanation for this “permanent” upward resetting of the body weight set-point in obese individuals. Our work and that of many others suggests that this set-point is a function of the constant dialogue between the brain and body. The brain is the master controller of overall energy homeostasis. It utilizes a distributed network of specialized “metabolic sensing” neurons to monitor neural, hormonal and metabolic signals from the body. These neurons differ from others in the brain in that they alter their firing rate in proportion to changes in local brain levels of metabolites such as glucose, amino acids, ketone bodies or fatty acids, or hormones such as leptin and insulin. The integrated output of these metabolic sensors activates behavioral, metabolic and neurohumoral systems involved in the regulation of energy homeostasis.
When we overeat, adipose stores increase and produce more of the hormone leptin. This hormone enters the brain and acts on metabolic sensing neurons to reduce food intake and increase energy expenditure. Similarly, the pancreas produces more insulin, which has a similar effect on these sensing neurons. We found that obesity-prone rats are born with a raised threshold for sensing both of these hormones. Thus, when fed a calorically dense diet, they continue to overeat and increase their fat stores until their leptin and insulin levels rise to a point that exceeds this raised threshold to put a break on intake. Thus, this reduced central leptin and insulin sensitivity may be the manifestation of the thrifty genotype in our rats. It might also be responsible for obesity in some humans, but only when they are provided with an excess of calories, a situation which is all too common in our modern world.
|Figure 1. A double-label in situ hybridization of metabolic sensing neuropeptide Y (NPY) neurons (red) in the hypothalamic arcuate nucleus. These neurons express glucokinase (GK; green), an enzyme which confers the ability to sense changes in ambient glucose levels. White arrows point to examples of neurons which express both NPY and GK. These same neurons are responsive to leptin and insulin.
This requirement of environmental input for full manifestation of the obese phenotype in adults has more sinister implications for developing fetuses and infants. When we made obesity-prone rat dams obese during gestation and lactation, their offspring became more obese as adults than they would have if their mothers had remained lean. Even more disturbing, genetically lean pups cross-fostered at birth to such obese dams also became obese and insulin resistant as adults. Furthermore, these early interventions were associated with permanent changes in the development and function of the distributed network of metabolic sensing neurons that control energy homeostasis.
What can we do about obesity?
Once individuals become obese, fewer than 10% can lose and maintain their weight loss for more that 1-2 years using currently available non-surgical therapies. This is not simply a matter of poor will power; our studies have shown that there are genetic, neurological and metabolic causes for this resistance to permanent weight loss. Thus, we believe that the best way to treat obesity is to prevent its development. This requires us to focus our research efforts on identifying critical factors that either encourage or ameliorate the development of obesity during the perinatal and early childhood period. All is not doom and gloom. Our unpublished studies suggest that exercise during the immediate post-weaning period not only prevents the development of obesity during exercise, but prevents obesity from developing for long periods after exercise is stopped. Such findings provide hope that there will be ways to halt the obesity epidemic and prevent its long term adverse health outcomes.
Barry Levin received his MD degree from Emory University School of Medicine and did his neurology residency at Cornell Affiliated Hospitals. He is currently professor of neurology and neurosciences at New Jersey Medical School and assistant chief of neurology at the East Orange VA Medical Center.