The prevalence of obesity and associated metabolic disorders such as type 2 diabetes is increasing in developed countries. The mechanisms controlling appetite and body weight are complex, making treatment a significant clinical challenge. Nonetheless, the recent explosion of new information concerning the roles of neural circuits and neuroendocrine signaling pathways to control body weight regulation holds promise for new and better therapies. In this TriPoint, a basic scientist discusses research in cell and animal models that are defining novel processes by which peripheral signals are transmitted to and interpreted by the central nervous system to modulate appetite and ingestive behaviors. A clinical researcher describes how the information from these models is being translated into humans and provides additional evidence for these pathways gained from research in humans. Finally, a clinician discusses recently approved agents as well as those in the pipeline and how they will be used to treat obesity in patients.
BASIC RESEARCHER PERSPECTIVE
The past two decades have been a wild ride for investigators in the fields of neuroendocrine control of food intake, body weight, and glucose homeostasis. The discovery of leptin and the elucidation of the role of the central melanocortin system in regulating energy balance were catalysts that propelled the fields forward. These and other discoveries were key as the rates of obesity and diabetes are still a major challenge for patients and clinicians around the globe. Leptin is a key hormone secreted by adipocytes that is required for the regulation of energy balance, glucose homeostasis, and nearly every neuroendocrine axis.
The melanocortin system is made up of neurons whose cell bodies reside in the arcuate nucleus of the hypothalamus. The central melanocortin system is comprised of pro-opiomelanocortin (POMC) neurons and agouti related protein (AgRP) neurons. POMC neurons produce α-melanocyte-stimulating hormone (α-MSH), an endogenous agonist of melanocortin 4 receptors (MC4Rs). α-MSH and other MC4-R agonists regulate food intake, body weight, and glucose homeostasis. Conversely, AgRP neurons secrete AgRP, the endogenous MC4R antagonist.
Complex Brain Circuits
One of the most important lessons we have learned concerns the remarkable division of labor among these brain pathways. Not surprisingly, the neural circuits controlling energy balance are exceptionally complex, with a segregation of duties between neurons that control energy expenditure and food intake. Moreover, sets of neurons in the hypothalamus respond to metabolic cues and regulate glucose homeostasis (including hepatic glucose production) in a manner that is dissociable from changes in food intake and body weight.
These concepts have been highlighted by genetic loss and restoration (rescue) studies. For example, we now know that leptin acts in distinct populations of neurons to regulate energy expenditure, food intake, and glucose homeostasis. Contrary to early predictions, direct leptin action on POMC neurons regulates hepatic insulin sensitivity and glucose production but does not significantly regulate food intake and body weight. Instead, leptin acts on other neurons including those in the lateral hypothalamus and brainstem to regulate food intake.
Another example of this complexity of these regulatory pathways is in the heterogeneity in seemingly similar classes of neurons. Electrophysiological studies indicate that the responses of POMC neurons are remarkably segregated. In particular, metabolic signals (e.g., leptin, insulin, serotonin, etc.) act acutely on nonoverlapping populations of POMC neurons to change their membrane potential and firing rates. Thus, after more than 15 years, the old adage “the more we know the less we know” seems ever more appropriate and much remains to be done if we are to develop more rationale strategies to combat the growing epidemic of obesity and diabetes.
Another key concept is the remarkable plasticity of the circuits controlling energy balance, especially during development. For example, developmental ablation of key orexigenic (i.e., appetite stimulating) neurons (AgRP/NPY) in mice produces very mild effects on food intake and body weight. In marked contrast, ablation of the very same neurons in adult mice produces frank anorexia and eventually starvation. Complex developmental patterns of gene expression may underlie or enable this flexibility. For example, some neurons that express POMC during development do not express POMC in adult neurons. These and other observations need to be considered when assessing the results of studies using genetic modified mice. They also suggest that events during development may irreversibly alter the neural circuitry underlying energy balance and related processes.
New Technologies Allow Selective Targeting
An appreciation of the complexity of the system demanded the field raise its game to the next level. Fortunately, new tools exist that allow acute manipulation of neuronal activity. These techniques have allowed investigators to directly test the role of identified neurons in regulating particular functions. The use of optogenetics and designer channels (and drugs targeting them) have led the way. Briefly, these approaches allow researchers to express light or chemical-sensitive channels into identified groups of neurons.
These engineered channels enable scientists to acutely activate or inhibit the activity (i.e., firing rates) of these neurons in a reversible fashion. For example, activation of AgRP neurons by light or designer drugs has clearly established their roles in regulating feeding. Currently, these types of approaches are being applied to several CNS circuits to help identify the role of distinct classes of neurons in regulating energy balance. Going forward it will be important to develop additional genetic tools that allow manipulation of key genes in chemically identified neurons in adult animals, thereby circumventing the issues of developmental complexity, plasticity, and compensation.
These basic science discoveries paved the way to the approval of the first anti-obesity drugs in years. Hopefully, these advances will give physicians new tools in their arsenal to treat patients struggling with obesity, diabetes, and related conditions. Currently, bariatric surgery represents a treatment option commonly chosen by doctors and patients. It is noteworthy that the field is now investigating the molecular and physiological basis for the effectiveness of the different bariatric surgeries in causing sustained weight loss and the associated improvement in glucose homeostasis. Hopefully, mechanistic insights drawn from these investigations will enable the pharmaceutical and biotech industries to develop targeted therapies that circumvent the need for invasive surgeries to induce sustained weight loss and diabetes remission.
CLINICAL RESEARCHER PERSPECTIVE
Body Weight Regulation Enters the Scientific Mainstream
Much has changed in the past 20 years with regard to how we approach research into the physiology of weight regulation. Prior to this, obesity was viewed primarily as a behavioral disorder and clinical trials were dominated by innumerable approaches to lower calorie intake using personal and group motivational behavior techniques. The obesity research landscape shifted dramatically with the discovery of mutations in the leptin gene in obese mice in 1994, catapulting to the forefront the concept of physiological regulation of a body weight “set point.” The first reports of leptin-deficient humans soon followed. In an important test of the relevance of leptin signaling and proof of concept for the existence of a body weight set point in humans, leptin replacement in children with this deficit completely reversed their severe, early-onset obesity.
Since then, using improved techniques to discover and study novel neurotransmitters and neuroelectrical modulators in animal models, the basic research literature began to build the neural architecture whereby the brain governs body weight. In one of the more remarkable realizations that emerged from these animal studies, disruptions in the homeostasis of this system result in both unwanted weight gain and weight loss. Obesity and cachexia, therefore, represent opposite ends of the disease spectrum involving the same weight regulation systems.
Challenges of Translational Human Research
Following Willie Sutton’s lead, when it comes to understanding the control of body weight, the “money,” so to speak, is in the brain. Not unexpectedly, determining which of the findings in animals are relevant to humans has been challenging.
At the cellular level, neuronal modulation leading to appetite regulation may take place by transynaptic delivery of neurotransmitters (e.g., dopamine, serotonin) or neuroendocrine factors (e.g., α-melanocyte stimulating hormone), by propagation of a depolarization potential from one neuron to another, or both. Several advances in brain imaging have allowed for indirect measures of brain activity in humans. For example, functional magnetic resonance imaging (fMRI), arterial spin labeling (ASL), and positron emission tomography (PET) measure physiological markers of neuronal activity (e.g., increased blood flow to, and glucose uptake by, active neurons and supporting cells) or neurotransmitter binding (e.g., dopamine) to receptors. However, these techniques remain surrogates of actual neuronal activity, and the spatial and temporal resolutions currently available do not provide the granularity to determine functional communication within or between nuclei of the hypothalamus and brainstem.
In addition, the list of peripheral hormonal and other modulators of body weight is expanding rapidly, making it important to integrate these new findings with existing physiological constructs. For example, studying the brain response to drinking a glucose beverage would seem a straightforward proposition. However, alterations in brain activity measured during the simple act of ingesting glucose involve far more than responses to rising blood glucose levels. As soon as the drink is started, taste receptors in the mouth and then stretch receptors in the esophagus and stomach immediately send signals to the brain. Shortly thereafter, secretion of several gut hormones (e.g., insulin, cholecystokinin, glucagon-like peptide 1) and inhibition of others (e.g., ghrelin) that have central brain receptors occur in anticipation of or in response to the rise in glucose.
Metabolism of glucose by bacteria in the gut microbiome, as well as in the liver (e.g., lactate), results in generation of additional centrally acting metabolic substrates. Finally, a generally poor understanding of the blood-brain-barrier transport and pharmacologic properties of many hormones and substrates often makes guesswork of whether measured changes in brain activity correspond with binding of a hormone or nutrient with its receptor or are the result of neural signaling downstream from the original activation site. Therefore, the simple act of swallowing and digesting glucose elicits a cascade of neurological and hormonal signals that converge, temporally overlap, and make interpretation of subsequent brain responses diffi cult. In other words, it’s complicated.
We are still just beginning to understand the fundamentals of the physiology underlying neuroendocrine control of body weight regulation and the pathophysiology that results in expression of obesity and cachexia in humans. Like most chronic diseases, the reality is that the complexity of this system defies our original hope for a simple model that lends itself easily to experimentation and points to an obvious and efficacious treatment. As we move forward with these studies, though, one of the largest areas of research growth has been in revisiting behavioral models of food intake using the brain imaging techniques mentioned above. Game theory, impulsivity control, and “reward” centers are all being explored as regulators of increased food intake in obese patients. Since the vastly expanded cortex in humans, compared to animals, provides complexity of thought as well as behavior, this leaves a large, uncharted territory to study.
CLINICAL PRACTITIONER PERSPECTIVE
Several signals produced by the adipose tissue and gastrointestinal tract (GIT) are involved in the regulation of energy homeostasis and body weight. The central nervous system (CNS) is responsible for integrating these peripheral signals with other information that arises from the external and internal milieus, including the dopaminergic, adrenergic, serotonergic, opioid, and endocannabinoid (EC) regulatory pathways. All these input signals trigger several compensatory responses to maintain a balance between energy intake and expenditure. The malfunctioning of one or more components of this complex machinery might result in energy imbalance and significant changes in body weight. The outstanding scientific advances in our understanding of this machinery have opened new opportunities for the development of novel pharmacologic strategies for the treatment of obesity.
Weight Loss Drugs: A History of Failures and Withdrawals
The landmark discovery of leptin in 1994 unleashed a wave of expectations regarding its potential to solve the obesity pandemic. Unfortunately, leptin therapy has proved to be effective only for the few patients who have congenital leptin deficiency. The vast majority of obese individuals are leptin-resistant and refractory to therapy. Similarly, drugs targeting different hormone systems within the GIT have not yet met their promise as anti-obesity agents. In the past decade, the clinical use of synthetic compounds acting as pharmacological blockers of the
EC receptor CB1 (CB1R) attracted special attention for the treatment of human obesity and its co-morbidities. Nevertheless, the initial enthusiasm was replaced by disappointment with the withdrawal of Rimonabant from the market due to reports of serious psychiatric adverse events. In the same direction, old and not so old compounds that modulate dopamine, norepinephrine, and serotonin availability in the CNS, such as amphetamine derivatives and selective serotonin reuptake inhibitors, have been either banned or approved only for short-term use because of their toxicity and/or cardiovascular safety. As a consequence, the lipase inhibitor Orlistat is currently the only anti-obesity drug approved for long-term therapy in most countries around the world.
Novel Medications, New Hopes, Old Fears
In 2012, after an interval of 13 years, the FDA approved two new drugs to battle obesity. The first one was lorcaserin (Belviq, Arena Pharmaceuticals), a selective serotonin 5HT2c receptor agonist that regulates appetite and reduces food intake, with no effect on energy expenditure. In clinical trials, lorcaserin promoted a significant, but still modest, weight loss compared to placebo, with a mean body weight change of 4.5% to 5.8% after one year. Its mechanism of action is similar to that of fenfluramine, which was withdrawn from the market due to heart valve damage. In rats, lorcaserin increased the incidence of mammary and brain tumors, but current evidence shows a large margin of safety in humans. Nevertheless, an old fear is back, as Arena recently notified the Committee for Medicinal Products for Human Use (CHMP) of its withdrawal of its application for marketing authorization for Belviq in Europe due to the lack of time to address all of the CHMP’s safety concerns.
The second approved drug is a combination of the long-established anorectic agent phentermine with the antiepileptic drug topiramate in an extended-release formulation (PHEN+TOP ER; Qsymia, Vivus). In clinical trials, the mean percentage change of body weight at one year with PHEN+TOP ER was 7.8% to 10.9%, values significantly greater than in the placebo groups. Again, an old fear has surfaced, as PHEN+TOP ER therapy is associated with elevations in resting heart rate (similarly to sibutramine), which may increase the risk for fatal arrhythmias. Another concern includes topiramate-related teratogenicity, and for this reason, the FDA’s approval required a risk evaluation and mitigation strategy (REMS).
Summary: Old Targets, New Concepts
The next round of the fight against obesity will likely rely on old targets. We will continue to bear witness of the endless search for more selective, effective, and safe compounds acting on specific neurotransmitters and receptors within the CNS. Despite the problems with the first CB1R antagonists, there is a hope that peripherally restricted CB1R antagonists might avoid the
known undesirable side effects of the central blockers.
Pharmacological interventions on GIT-brain signaling pathways are now a reality for the treatment of type 2 diabetes and might prove valuable for weight loss in obesity. For instance, the GLP-1 analogues exenatide and liraglutide can result in weight loss and are promising alternatives, alone or in combination with compounds targeting other GIT and metabolic receptors, for future use in non-diabetic obese population.
The novel concept of combining lower doses of two drugs that act simultaneously in different regulatory mechanisms within the brain has attracted considerable attention. Besides Qsymia, ongoing trials in this area include the association of bupropion with the opioid antagonist naltrexone in a sustained release (Contrave) or with the anti-epileptic zonisamide (Empatic). And, there are more to come, based on the premise that the combined therapies might result in higher effi cacy with fewer side effects. Only time will tell if this is true. Meanwhile, I would not be surprised if a new anti-obesity agent is discovered by serendipity among the various preparations in the pipeline primarily aimed for the treatment of diabetes, hypertension, depression, and other diseases.
— This article was reviewed by Daniel J. Bernard, PhD, and Margaret E. Wierman, MD, of the Endocrine Society’s Research Affairs Core Committee.