Animals and humans are exposed to a variety of foods throughout their lives and must learn which foods are energy-rich in order to survive. How do we learn to associate flavors with caloric value, and how does information regarding caloric intake travel from the gut to the brain? Hypothalamic agouti-related protein (AgRP) neurons, essential for feeding behavior, evoke feeding in rodent models within minutes when stimulated (Belgardt et al., 2009; Krashes, 2011). As infusions of nutrients provoke a drop in AgRP activity (Goldstein et al., 2021), we hypothesized that this reduction in AgRP firing is critical to the formation of learned associations between nutrients and flavors and designed an experiment to test this hypothesis. 

       Flavor-nutrient learning is a common paradigm for investigating how mice learn to associate specific flavors with caloric intake (Myers, 2018). Infusions of glucose or fat paired with consumption of a particular flavor can condition a preference for that flavor over a flavor paired with an infusion of water (Sclafani, 2011). Over time, mice learn to associate calorie-free flavors with nutritive infusions. There are two main phases of a flavor-nutrient paradigm: training and testing. During training, each mouse is separately presented with the conditioned stimulus (CS+, the fat-paired or glucose-paired flavor) and its control (CS-, the water-paired flavor), with the necessary infusion administered within each session. During testing, each mouse is presented with both flavors (CS+ and CS-) and a lickometer records the number of licks of each flavor. This data is then used to calculate a preference index (PI), a numerical representation of the mouse’s preference. 

       We predicted that optogenetic (light-induced) stimulation of AgRP neurons, counteracting nutrient-induced reductions in AgRP neuron activity, would reduce mice’s ability to form flavor-nutrient associations. During both CS+ and CS- training sessions, the experimental cohort received optogenetic (light-induced) AgRP neuron stimulation, while the control cohort did not. Our original results contradicted our hypothesis; AgRP stimulation during testing potentiated flavor-nutrient preferences in the experimental cohort in both fat and glucose trials. Further analysis of the data indicated that mice in the experimental cohort licked more, or consumed more of the flavors, during training. When intake during training was controlled for, there was no longer a significant difference in the flavor preferences established between the experimental and control groups. However, even after controlling for consumption differences within the glucose trials, mice in the experimental cohort retained their flavor-nutrient preferences for longer than mice in the control cohort. Consequently, we are currently investigating whether AgRP stimulation strengthens the memory, or retention, of flavor-nutrient preferences. 

       Mapping flavor-nutrient learning pathways could potentially allow us to manipulate them, increasing or decreasing the palatability of different foods and flavors. This capability could assist individuals in maintaining healthier eating habits. Additionally, furthering understanding of AgRP’s role in feeding behavior could lead to the development of additional pharmacological advancements and interventions for weight loss. Deciphering these neuronal pathways is becoming increasingly important in the context of the worldwide obesity epidemic.