How is gustatory information represented in the cortex? The answer to this question has been one of the mostly intensely debated topics in the field of gustatory neuroscience1-3. Two theories have been proposed and discussed over the past decades. According to the labeled-line theory the physiochemical identity of a gustatory stimulus is encoded by specific subsets of neurons, each of which is narrowly tuned to a single taste quality (i.e. salty, sweet, sour, bitter or umami)4-7. An alternative theory, the across-neuron pattern theory, postulates that taste coding relies on the combined activity of large ensembles of cortical neurons8. According to this theory, neurons do not need to be selective for specific taste qualities, instead each neuron can densely represent information by encoding multiple qualities3,9. Both theories are supported by experimental evidence. While a recent 2-photon calcium imaging study suggested the exclusive presence of narrowly tuned neurons in GC of anesthetized mice7; years of electrophysiological recordings in anesthetized and alert rodents demonstrate the presence of both narrowly tuned and densely coding neurons in GC10-13. Despite evidence that neurons using different coding strategies exist in GC, the debate is still polarized and no unifying view of taste coding in the cortex has emerged. The overarching goal of this proposal is to test the hypothesis that narrowly tuned and densely coding neurons reflect different stages of cortical processing. Specifically, we propose that the coding scheme varies depending on the cortical layers, with superficial layers (i.e., layers 2/3) featuring narroly tuned neurons and deep layers (i.e., layers 4 and 5) containing densely coding neurons. In addition, the experiments in this proposal aim at providing a mechanistic explanation for these differences, linking coding properties with laminar-dependent variations in inhibitory drive. The experiments will rely on a combination of sophisticated electrophysiological approaches to directly relate coding properties, balance between excitation and inhibition, and properties of local microcircuits. The framework of this grant is firmly grounded in the literature on cortical coding of sensory information14-17, and supported by our preliminary results, showing layer-specific differences in taste response properties and local connectivity. This research will help the field go beyond a dated debate and will move the discussion on taste coding toward a more circuit-oriented perspective. If successful this research will provide, for the first time, a unifid view of taste coding in cortical circuits of alert rodents.
The ability to perceive gustatory information is essential for survival. Animals decide to ingest or reject foods based on taste. Despite the importance of gustation, to date there is no consensus on how information about taste is represented in the gustatory cortex (GC) - the primary cortical area devoted to represent taste. Over the past decades two theories have been advanced. According to the first theory (known as 'labeled- line theory'), gustatory stimuli are represented by the activity of few and very specialized neurons. For instance, the labeled-line theory postulates that the sensation of sweetness experienced when an animal consumes sugar is associated with the activation of a selective group of 'sweet-sensing' neurons. According to another theory (the 'across neuro pattern theory'), taste is represented by the combined activation of ensembles of neurons. In this view, a neuron can participate to the representation of multiple taste quality. Recordings of taste evoked activity in GC have provided evidence for both theories, with some neurons appearing selective and others encoding multiple stimuli. This mixed evidence suggests that neither one of the two theories provides a complete picture of taste coding in GC. Instead, it is likely that a ne framework is needed to account for the experimental results. The goal of this proposal is to formulate and test such new framework. Specifically, we postulate that different neural responses arise from the organization of GC circuits and reflect different processing stages of gustatory information. We hypothesize that selective neurons dominate in the superficial layers and that neurons encoding multiple tastants occur in deep layers of GC. This layer difference is likely the results of laminar-specific inhibitory circuits, which are stronger in the superficial layers. To test this view, the experiments in this proposal rely on an array of sophisticated electrophysiological techniques in brain slices and in behaving animals. If successful, the results of this grant will provide novel and important insights into the cortical mechanisms underlying taste processing. Understanding how gustatory information is encoded in cortical circuits will provide potential pharmacological targets for shaping the perception of taste quality and influencing taste learning in physiological and pathological conditions.