The lateral hypothalamus (LH) and the central nucleus of the amygdala (CeA) exert an influence on ingestive behavior and are reciprocally connected to gustatory and viscerosensory areas, including the nucleus of the solitary tract (NST) and the parabrachial nuclei (PbN). We investigated the effects of LH and CeA stimulation on the activity of 101 taste-responsive neurons in the hamster PbN. Eighty three of these neurons were antidromically activated by stimulation of these sites; 57 were antidromically driven by both. Of these 83 neurons, 21 were also orthodromically activated-8 by the CeA and 3 by the LH. Additional neurons were excited (n ϭ 5) or inhibited (n ϭ 8) by these forebrain nuclei but not antidromically activated. Taste stimuli were: 0.032 M sucrose, 0.032 M sodium chloride (NaCl), 0.032 M quinine hydrochloride (QHCl), and 0.0032 M citric acid. Among the 34 orthodromically activated neurons, more sucrose-best neurons were excited than inhibited, whereas the opposite occurred for citric-acidand QHCl-best cells. Neurons inhibited by the forebrain responded significantly more strongly to citric acid and QHCl than cells excited by these sites. The effects of electrical stimulation were mimicked by microinjection of DL-homocysteic acid, indicating that cells at these forebrain sites were responsible for these effects. These data demonstrate that many individual PbN gustatory neurons project to both the LH and CeA and that these areas modulate the gustatory activity of a subset of PbN neurons. This neural substrate is likely involved in the modulation of taste activity by physiological and experiential factors.

Previous studies have shown a modulatory influence of forebrain gustatory areas, such as the gustatory cortex and lateral hypothalamus, on the activity of taste-responsive cells in the nucleus of the solitary tract (NST). The central nucleus of the amygdala (CeA), which receives gustatory afferent information, also exerts descending control over taste neurons in the parabrachial nuclei (PbN) of the pons. The present studies were designed to investigate the role of descending amgydaloid projections to the NST in the modulation of gustatory activity. Extracellular action potentials were recorded from 109 taste-responsive cells in the NST of urethan-anesthetized hamsters and analyzed for a change in excitability following electrical and chemical stimulation of the CeA. Electrical stimulation of the CeA orthodromically modulated 36 of 109 (33.0%) taste-responsive NST cells. An excitatory response was observed in 33 (30.28%) cells. An initial decrease in excitability to electrical stimulation of the CeA, suggestive of postsynaptic inhibition, was observed in three (2.75%) NST taste cells. NST cells modulated by the CeA were significantly less responsive to taste stimuli than cells that were not. Many of these cells were under the modulatory influence of the contralateral CeA (28/36 = 77.8%) as well as the ipsilateral (22/36 = 61.1%); 14 (38.9%) were excited bilaterally. Latencies for excitation were longer after ipsilateral than after contralateral CeA stimulation. Microinjection of DL-homocysteic acid (DLH) into the CeA mimicked the effect of electrical stimulation on each of the nine cells tested: DLH excited eight and inhibited one of these electrically activated NST cells. Application of subthreshold electrical stimulation to the CeA during taste trials increased the taste responses of every CeA-responsive NST cell (n = 7) tested with this protocol. These effects would enhance taste discriminability by increasing the signal-to-noise ratio of taste-evoked activity.

Although the reciprocal projections between the bed nucleus of the stria terminalis (BNST) and the gustatory parabrachial nuclei (PbN) have been demonstrated neuroanatomically, there is no direct evidence showing that the projections from the PbN to the BNST carry taste information or that descending inputs from the BNST to the PbN modulate the activity of PbN gustatory neurons. A recent electrophysiological study has demonstrated that the BNST exerts modulatory influence on taste neurons in the nucleus of the solitary tract (NST), suggesting that the BNST may also modulate the activity of taste neurons in the PbN. In the present study, we recorded from 117 taste-responsive neurons in the PbN and examined their responsiveness to electrical stimulation of the BNST bilaterally. Thirteen neurons (11.1%) were antidromically invaded from the BNST, mostly from the ipsilateral side (12 cells), indicating that a subset of taste neurons in the PbN project their axons to the BNST. The BNST stimulation induced orthodromic responses on most of the PbN neurons: 115 out of 117 (98.3%), including all BNST projection units. This descending modulation on the PbN gustatory neurons was exclusively inhibitory. We also confirmed that activation of this efferent inhibitory projection from the BNST reduces taste responses of PbN neurons in all units tested. The BNST is part of the neural circuits that involve stress-associated feeding behavior. It is also known that brain stem gustatory nuclei, including the PbN, are associated with feeding behavior. Therefore, this neural substrate may be important in the stress-elicited alteration in ingestive behavior.

Taste-responsive cells in the nucleus of the solitary tract (NST) either project to the parabrachial nuclei (PbN) of the pons, through which taste information is transmitted to forebrain gustatory nuclei, or give rise to axons terminating locally within the medulla. Numerous anatomical studies clearly demonstrate a substantial projection from the rostral NST, where most taste-responsive cells are found, to the PbN. In contrast, previous electrophysiological studies in the rat have shown that only a small proportion (21-45%) of taste-responsive NST cells are antidromically activated from the PbN, suggesting that less than half the cells recorded from the NST are actually involved in forebrain processing of gustatory information. In the present experiment we investigated the projections from the NST to the PbN electrophysiologically in urethane anesthetized hamsters. Responses of 101 single neurons in the rostral NST were recorded extracellularly following lingual stimulation with 32 mM NaCl, sucrose and quinine hydrochloride (QHCl) and 3.2 mM citric acid. The taste-responsive region of the PbN was identified electrophysiologically and stimulated with a concentric bipolar electrode to antidromically activate each NST cell. Of the 101 taste-responsive NST cells, 81 (80.2%) were antidromically activated from the ipsilateral PbN. The mean firing rates to taste stimulation and the spontaneous activity of these projection neurons were significantly greater than those of non-projecting cells. Every sucrose-best neuron in the sample projected to the PbN. The mean conduction velocity of the 23 QHCl-best neurons was significantly lower than that of the other 58 PbN projection neurons, suggesting that the most QHCl-responsive cells are a subset of smaller neurons. These data show that a large majority of NST cells responsive to taste stimulation of the anterior tongue project to the gustatory subdivisions of the PbN and that these cells have the most robust responses to gustatory stimulation.

The G protein subunit alpha-gustducin is expressed in a subset of light (Type II) but not in dark (Type I) cells in rat vallate taste buds. The thymidine analogue 5-bromo-2'-deoxyuridine (BrdU) is incorporated into DNA during the S-phase of the cell cycle and can be used to determine the time of origin of a cell. In this study, 31 rats were injected with BrdU (50 mg/kg i.p.) and perfused at various times, from 2.5 to 10.5 days, following BrdU administration. Vallate papillae were embedded in polyester wax, cut into 4 microm transverse sections, and characterized with antibodies to BrdU and alpha-gustducin. Sections were processed for indirect immunofluorescence or with an immunoperoxidase procedure. From immunoperoxidase material on 21 rats, counts of alpha-gustducin- and BrdU-labeled cells were obtained from 300-800 taste bud profiles at each survival time; a total of 4122 taste bud profiles were examined. Cells with nuclei immunoreactive for BrdU occurred within the taste buds at 2.5 days and double-labeled cells were clearly evident at 3.5 days; a small number of double-labeled cells were seen as early as 2.5 days. Double-labeled cells reached a peak at 6.5 days and did not decline significantly by 10.5 days. Cells labeled for BrdU but not alpha-gustducin peaked at 5.5 days and showed a significant decline by 8.5 days. These latter cells included light cells not expressing alpha-gustducin and dark cells, which have previously been shown to have a shorter life span than light cells. These data suggest that expression of alpha-gustducin appears very early in a cell's life span and that these cells are longer lived than many of the cells that do not express this G protein.