Meal-related satiety signals arise from the gastrointestinal tract in response to incoming nutrients and promote meal termination and continued “fullness” after the meal. We know that many satiety hormones, including cholecystokinin (CCK) and glucagon-like peptide 1 (GLP-1) released from the small intestine, signal to the brain through the sensory fibers of the vagus nerve. These vagal fibers synapse in the hindbrain and directly activate neurons in the area postrema (AP) and nucleus tractus solitarius (NTS). Excitation of these hindbrain neurons correlates with and is necessary for the anorexic response to these hormones, but the conscious decision to stop eating involves additional processing in other brain regions. Our research has focused on connections between the NTS and a number of rostral forebrain areas that are known for their involvement in reward-motivated behavior. The working hypothesis is that these connections are a mechanism by which satiety signals rising over the course of a meal reduce the reward value of food and motivation to continue eating, ultimately leading to meal termination. In our recent paper (AJP Endo Metab 2018), we showed that indeed, gastric nutrient infusions, which bypass the mouth and allow complete experimenter control over nutrient delivery to the gut, significantly reduce motivation to obtain palatable food.  

Much of our work on this topic has investigated NTS neurons containing GLP-1 (functioning here as a neurotransmitter, not a gut peptide) and the neurons in other brain areas that express GLP-1 receptors. GLP-1 neurons are activated by vagal afferent stimulation, and a subset of these cells project to the nucleus accumbens (NAc), an area most well known for its role in reward in the context of drugs of abuse. We hypothesized that this GLP-1 projection plays a role in satiation, and we showed that endogenous release of GLP-1 at this site suppresses meal size largely by reducing the palatability of food, and that these receptors mediate some of the satiating effects of food in the gastrointestinal tract (J Neurosci 2011, AJP: Endocrinol Metab 2013). The availability of transgenic mouse models that facilitate visualization and manipulation of GLP-1 and GLP-1 receptor-expressing neurons next led us to investigate the lateral septum (LS) and bed nucleus of the stria terminalis (BNST) as important sites for endogenous GLP-1 effects on feeding. These forebrain nuclei receive strong GLP-1 neuron projections and densely express GLP-1 receptors, and also have roles in reward, learning, memory and stress responses. In a recent series of papers (AJP Regul Integr Comp Physiol 2016; Physiology & Behavior 2018; Physiology & Behavior 2019; Neuropharmacology 2018), we demonstrated that GLP-1 action in these locations influences food intake, meal patterns, motivation for food, and mediates the anorexic response to acute stress in both rats and mice.

We are currently funded to continue our investigation into the brain GLP-1 system, with a focus on elucidating the connectivity among GLP-1 receptor-expressing neurons and determining how they mediate these diverse effects on feeding behavior. Using modern neuroanatomical tracing methods that confer unprecedented cell-type specificity, we found that in several brain areas, GLP-1-receptive neurons project to other nuclei where GLP-1 receptors are also expressed, the first evidence of such an arrangement in a neuropeptide system. With our collaborator at University College London, Dr. Stefan Trapp, we have begun to investigate the electrophysiological responses of these GLP-1-responsive neurons (e.g., Neuropharmacology 2018). We are applying novel cell-type-specific transsynaptic tracing methods to characterize and confirm synaptic connections among these neurons, and will also identify other brain regions and cell types that provide synaptic inputs to GLP-1-receptive neurons. Because the goal is to understand how activity within this system influences feeding behavior, we are also using chemogenetic techniques to selectively excite or silence specific GLP-1-receptor neuron  projections, and will determine how these manipulations affect feeding and food-motivated behavior in a variety of test paradigms.

It is clear that we do not eat only when we are metabolically depleted. Although we may feel full after a meal, the sight or aroma of desirable food can dramatically alter our sense of satiety. We hypothesize that this is the result of the brain’s anticipation of a highly rewarding food, and that this process suppresses the brain’s perception of and/or response to satiety signals arising from the gut. Gastrointestinal satiety signals do not  disappear upon the ingestion of a sweet high-fat food, rather they are further stimulated. The fact that we eat more nonetheless suggests that the brain can override these normally potent signals. Numerous findings suggest that orexin neurons in the lateral hypothalamus are involved in anticipation of rewarding food. In rats and mice, these neurons are activated by cues that predict strong positive reinforcement, including chocolate or sugar treats. Orexin neurons project too many brain nuclei, including the hindbrain NTS and the AP, another region that receives feedback from the gastrointestinal tract. We hypothesized that this projection serves as a direct link between food anticipation and satiety signaling and published a series of studies (AJP Regul Integr Comp Physiol 2011; Psychopharmacology 2014) demonstrating that hindbrain orexin receptor signaling increases food intake by reducing satiation, increasing motivation for sugar, and increasing food-seeking behavior. Next, we hypothesized that orexin may work to suppress satiation signals in other brain areas, as well. The ventral  tegmental area (VTA) is best known as the origin of the dopamine neurons that form the heart of the mesolimbic reward pathway, which plays a key role in natural and drug reward. We and others have shown that the VTA receives input from the NTS, potentially bringing gastrointestinal satiation signal information to this site, and in addition, orexin receptors are expressed there. In a series of experiments (AJP Regul Integr Comp Physiol 2016), we found that orexin works in the VTA to counteract the satiating effects of gastrointestinal nutrients. This work provided some of the first functional evidence that VTA neurons are indeed integrating satiety and food reward information. Future directions for this project include determining the neurochemical phenotypes of relevant orexin-receptive neurons, as well as their anatomical projections.

In humans, binge eating is characterized by the consumption of an objectively large amount of food in a short period of time accompanied by a loss of control, and is a core feature of multiple eating disorders, including binge eating disorder, bulimia nervosa, and the binge/purge subtype of anorexia nervosa. Clinical studies suggest that impaired satiation is likely to play a role in the development and/or maintenance of binge eating behavior, though the mechanisms are not yet well understood. Studies of individuals with eating disorders have been unable to determine whether these differences in satiation responses are pre-existing risk factors that contribute to the development of binge eating behavior or occur as a consequence of binge eating behavior and contribute to illness maintenance, but rodent models can help address these questions. Our lab uses a model in which rats or mice are given ad libitum access to standard rodent chow and are also presented with 20-hour time-limited access to palatable high-fat diet (HFD) on an intermittent schedule (e.g., every 4th day). While no rodent model can fully reproduce the complex features of human eating disorders, the intermittent HFD access model reliably reproduces the excessive intake phenotype that is characteristic of binge eating. In a recent series of experiments (Obesity, 2020), we demonstrated that repeated intermittent HFD access does impair the satiety response to gastrointestinal nutrients, and we identified an altered response to the pancreatic satiety hormone amylin that likely contributes to this impairment. We now aim to identify neural mechanisms underlying this change in sensitivity to gut signals.

Our rodent work in this direction has been informed in part by our collaboration with Dr. Pam Keel in our Psychology department at FSU, and Dr. Williams is currently a co-investigator on Dr. Keel’s R01-funded project focusing on related questions in human subjects. This study examines the biological mechanisms through which weight suppression (the difference between an individual’s highest and current weight) may contribute to binge eating by increasing motivation for food and decreasing satiation (Physiol Behav, 2019). As we continue to obtain new data in humans, we are well positioned to use the complementary rodent model to perform experiments in which we can experimentally manipulate these factors and directly assess alterations in brain tissue.