Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Biomedical Sciences



Research Advisor

John D. Boughter, Ph.D.


William Earl Armstrong, Ph.D. Matthew Ennis, Ph.D. Kristin Marie Hamre, Ph.D. Wen Lin Sun, M.D., Ph.D.


We eat what tastes good. We also eat because it is necessary for our health. In fact, some of the most nutritious foods (e.g., vegetables) are often less appetizing, and the tastiest (e.g., fast food, ice cream) may be the least healthy. Despite the former, we may also have a lower limit of what we accept at which point nutrition becomes irrelevant (e.g., “spinach is just too yucky”). Further, we may eat unhealthily because of overwhelming urges. We investigated the complex interactions of taste and feeding at the neurobiological level using the experiments described.

In one sense, this neurobiology begins at the periphery with information about ingested substances (i.e., presumably food) being sent to central nuclei. The taste pathways provide one of these routes to the central nervous system. In terms of regulating feeding, we have the neurobiological substrates for urge, pleasure, and displeasure. The relationship of the dopamine (DA) system with reward is well-known, and indeed, studies have shown taste nuclei project to these areas.

Since earlier studies and data collected in our lab showed that the neurons of the parabrachial nucleus (PBN) projected to the ventral tegmental area (VTA), and lesioning the PBN attenuates taste-elicited release of DA in the nucleus accumbens, we hypothesized this connection plays a crucial role in the control of feeding, especially with regard to the processing of both appetitive and aversive stimuli, and the relationship of this processing to classical reward circuitry. We therefore utilized a number of neuroanatomical and behavioral techniques to probe taste and intake-related activity in the PBN, VTA, and the PBN-to-VTA circuit. The overarching goal was to contribute to a comprehensive understanding of the taste and reward neural mechanisms that mediate feeding.

We used a variety of immunohistochemical methods to test our hypotheses, including one measuring c-Fos-like immunoreactivity (FLI) in neurons (a measure that correlates with neuronal activation in some systems such as taste). Intraoral stimuli increased FLI in the PBN across a number of subnuclei, and in this case, we used a diaminobenzidine stain (DAB) with brightfield microscopy. Comparing C57BL6/J (B6) with mice lacking TRPM5 (KO) showed that some of this increase is driven by taste receptor input, but this effect is predominantly for quinine hydrochloride (QHCl). On the other hand, increases in FLI to sucrose (relative to water) in the lateral PBN were the same for both B6 and KO mice, leading to the conclusion that this FLI may be visceral in nature. Sucrose-elicited FLI in the external lateral subnucleus (el) was probably visceral, whereas QHCl-elicited FLI there was taste-related. We also combined measurement of FLI with retrograde tracing under fluorescent microscopy to compare activity in PBN projections to the VTA and gustatory thalamus (VPMpc). Retrograde tracing revealed two largely independent projections, with VTA-projecting neurons found more contralaterally, and VPMpc-projecting neurons found ipsilaterally. However, both types of cells are found in the caudal, gustatory “waist” portion of the PBN. Interestingly, there is a lack of VTA-projecting cells in the el. Patterns of FLI were consistent with the DAB

experiment, except with higher expression as compared to water in this fluorescent experiment in a few subnuclei. This may have been due to methodological differences. As for double-labeled cells, more VTA-projecting cells expressed FLI in response to sucrose or QHCl than to water; this numbered to only about 5% of cells, however, and did not differ according to side. This was compared to double-labeling in VPMpc- projecting cells, where the percent of tracer was around 10% for both QHCl and sucrose on the ipsilateral side and 5% on the contralateral side.

We looked at FLI throughout the VTA as well to see if the activity indicated there was a differential response to stimuli with varying taste valence. First, using the same intraorally-stimulated mice with DAB-stained sections, we observed FLI in the VTA. It did not occur in a stimulus-specific fashion and apparently not in a taste-dependent fashion (no significant differences between B6 and KO). In another experiment using fluorescent stains and confocal microscopy, we looked at the FLI in the VTA while delineating it by subnuclei, counting section by section, and identifying DA and GABA cell types. There were many more DA cells in the VTA than GABA cells, and they had distinct patterns of expression across subnuclei and section levels (i.e., within the anteroposterior [AP] dimension). The rostromedial tegmental area was located as a region with higher GABA cell expression. More DA cells were double-labeled with FLI for QHCl than for water or sucrose in the caudal linear nucleus of the raphe. Few GABA cells were double-labeled with FLI.

To show the PBN-to-VTA circuit’s role in taste-mediated feeding, we attempted a procedure that would selectively activate VTA-projecting PBN neurons using designer receptors exclusively activated by designer drugs (DREADDs). However, we were unable to verify the efficacy of clozapine-N-oxide to activate the circuit and opted for an alternative manipulation. We instead inhibited the VTA with direct injections of the GABA agonist, muscimol. This resulted in mice reducing their licking (relative to baseline) of sucralose, but not QHCl or water (i.e., an arrangement of non-caloric stimuli with palatable, aversive, and neutral valence). Muscimol also reduced licking of sucrose and QHCl-adulterated sucrose (i.e., caloric stimuli). The reduction in licking to caloric stimuli was accompanied by a decrease in the rate of intake, i.e., muscimol-inhibited mice slowed their lick rate and possibly stopped licking sooner compared to vehicle- injected controls.

Overall, this project confirmed that both the PBN and VTA function to communicate taste and reward information. Although the PBN-to-VTA circuit’s function remained elusive, the evidence of the direct path connecting these two nuclei was fortified. Further, to our knowledge, this was the first time evidence was found of its existence as a PBN projection pathway that is mostly separate from the projection to the gustatory thalamus. Combined with the knowledge of this circuit, the activity in these nuclei and the ability to affect consumption by inactivating the VTA suggest the PBN and VTA work together to influence feeding by detecting and integrating information about palatability and calories.




Available for download on Wednesday, July 08, 2020