Tsuyoshi Shimura

Neural substrates for conditioned taste aversion in the rat

Conditioned taste aversions (CTAs) are well known to be robust and long-lasting instances of learning induced by a single CS (taste)-US (malaise) pairing. CTA can be taken as a general model to search for neural mechanisms of learning and memory. In spite of extensive research on CTAs using a variety of approaches during the last three decades, the neural mechanisms of taste aversion learning still remain unsolved. In this article we propose a model of neural substrates of CTAs on the basis of our recent studies incorporating previous findings by other workers. Our studies mainly included experiments using ibotenic acid injections into various parts of the rat brain as a lesion technique, and c-fos immunohistochemistry in naive and CTA trained rats. CTAs were established by pairing the ingestion of saccharin (CS) with an ip injection of LiCl (US). Behavioral studies have shown that the parabrachial nucleus (PBN), medial thalamus, and basolateral nucleus of the amygdala are essential for both acquisition and retention of CTAs. C-fos studies suggested that association between gustatory CS and visceral US takes place in the PBN. The gustatory cortex (GC) may modify the strength of this association depending on the nature of the CS, viz., novel or familiar. The amygdala is indispensable for the expressions of CTAs. Tastes with hedonic values are stored in the GC in a long-term manner.

Conditioned taste aversion in rats with excitotoxic brain lesions

Conditioned taste aversion (CTA) is well known to be a robust and long-lasting learning after a single conditioned stimulus (CS) (taste)--unconditioned stimulus (US) (malaise) pairing. The neural mechanisms of this taste aversion learning still remain to be resolved. To elucidate the basic brain mechanisms of the taste aversion learning, we examined the effects of lesions of various sites of the rat brain on the acquisition and retention of CTAs. Confined brain lesions were made by injections of a small amount of excitotoxic drug, ibotenic acid. CTAs were established to saccharin (CS) by pairing its ingestion with an i.p. injection of LiCl (US). Rats lacking the parabrachial nucleus (PBN) almost completely failed to acquire CTAs. The second most effective lesion was in the medial thalamus including the parvocellular part of the ventral posteromedial nucleus of the thalamus (VPMpc) and the midline part, followed by the damage of the lateral nuclear group of the amygdala including the basolateral amygdaloid nucleus. Lesions of the gustatory cortex (GC) and hippocampus induced moderate effects, but lesions in the other subnuclei of the amygdala, such as the medial and central amygdaloid nuclei, entorhinal cortex, lateral hypothalamic area, and ventromedial hypothalamic nucleus induced slight or no effects. On the other hand, paired lesions among the amygdala, medial thalamus and GC caused severe impairment of CTAs; in particular, lesions of amygdala and VPMpc completely disrupted acquisition of CTAs. These results suggest that the PBN, medial thalamus and the lateral nuclear group of the amygdala play an essential role in the formation of taste aversion learning.

Representation of hedonics and quality of taste stimuli in the parabrachial nucleus of the rat

The distribution of evoked expression of the proto-oncogene c-fos was immunohistochemically examined in the parabrachial nucleus (PBN) of water-deprived rats after free ingestion of palatable liquids, after intra-oral infusion of aversive taste solutions including various bitter substances, and after an intraperitoneal injection of LiCl. C-fos immunoreactive neurons (c-fos neurons) were densely packed in the external lateral subnucleus (els), external medial subnucleus (ems), dorsal lateral subnucleus (dls), central lateral subnucleus (cls), and the central medial subnucleus (cms) depending on the kind of stimulation. The rostral part of the els may be related to general visceral inputs; the caudal part of the els, negative hedonics or aversive behavior; the dls, positive hedonics or ingestive behaviour. Both the ems and the els may be related to taste information for bitter tasting compounds and HCl; the cls, for sucrose and saccharin; and the cms, for NaCl.

C-fos expression in the rat brain after intraperitoneal injection of lithium chloride.

The distribution of evoked expression of the proto-oncogene c-fos was immunohistochemically examined in the rat brain after intraperitoneal injection of isotonic LiCl, which is commonly used to induce internal malaise in the conditioned taste aversion paradigm. C-fos-like immunoreactive neurones (c-fos neurones) were most densely observed in the central amygdaloid nucleus, external lateral subnucleus of the parabrachial nucleus (PBN), posteromedial and commissural parts of the nucleus of the tractus solitarius (NTS) and area postrema (AP). Experiments including vagotomy, intravenous injection of LiCl and lesions of the area postrema suggest that NTS neurones are activated via both sides of the vagus nerves, while AP neurones, humorally as well as neurally via the vagal nerve with a right side predominance. The activated NTS and AP neurones project mainly to the external lateral subnucleus of the PBN and lightly to the central lateral subnucleus of the PBN. These results are discussed in terms of the role of LiCl in the formation of conditioned taste aversion.

The medial preoptic area is involved in both sexual arousal and performance in male rats: re-evaluation of neuron activity in freely moving animals

A total of 74 single unit activities was recorded from the medial preoptic-anterior hypothalamic continuum (MPOA) during free copulatory behavior of male rats. Forty-six units (62.2%) showed changed activities during at least one phase of male copulatory movements; 26 units (35.1%) increased in the firing rate during pelvic thrusting; 32 units (43.2%) increased in activity during backward jumping immediately after intromission; only one unit (1.4%) showed decreased firing rate during thrusting and backward jumping; 12 units (16.2%) increased in activity during pursuit of a female; 19 units (25.7%) were suppressed during genital grooming. Furthermore, 67 units (90.5%) showed a significant change in activity throughout a series of copulatory behavior. From the introduction of a female up to ejaculation, relatively large number of units increased in the firing rate above the value during pre-introduction adaptation period. During postejaculatory interval, however, most units decreased in activity below the level during copulation. These results strongly suggest that the MPOA is involved in both sexual arousal and performance in male rats.

Ibotenic acid lesions of the parabrachial nucleus and conditioned taste aversion : Further evidence for an associative deficit in rats

Rats with extensive ibotenic acid lesions centered in the gustatory zone of the pontine parabrachial nucleus (PBN) failed to acquire a conditioned taste aversion (CTA) induced by lithium chloride (LiCl) toxicosis (Experiments 1 and 4). This deficit cannot be explained as an inability to either perceive or process gustatory information because lesioned rats that failed to acquire a CTA readily acquired a conditioned flavor preference (Experiment 2). Similarly, the CTA deficit cannot be attributed to an inability to experience or process visceral input because PBN-lesioned rats that failed to acquire a CTA successfully learned an aversion to a trigeminal stimulus, capsaicin, when paired with LiCl-induced illness (Experiment 3). This pattern of results supports the view that cell bodies within the PBN are essential for the associative processes that govern CTA learning.

Neurochemical modulation of ingestive behavior in the ventral pallidum

The nucleus accumbens and its related circuitry have been shown to play an important role in promoting the intake of hedonically desirable food. A previous report has demonstrated that the blockade of GABAA receptors in the ventral pallidum (VP), a target of GABAergic projection from the nucleus accumbens, greatly increases food, but not water, intake in satiated rats [ Stratford et al. (1999)Brain Research, 825, 199–203]. The present study examined which neurotransmission in the VP is specifically involved in the intake of normally preferred taste stimuli. Microinjections of the GABAA antagonist bicuculline selectively increased the intake of saccharin solution but not that of water and quinine solution in water‐deprived rats. In contrast, the facilitation of GABAA receptors by microinjections of muscimol in the VP generally suppressed the intake of saccharin, water and quinine. The same injections induced strong aversive taste reactivity responses to oral stimulation with not only quinine but also water and saccharin. The local administration of d‐Ala2,N‐Me‐Phe4,Glyol5‐enkephalin, a selective µ‐opioid receptor agonist, into the VP had time‐dependent effects, decreasing saccharine intake early and increasing intake late. Microinjections of SCH‐23390, a dopamine D1 receptor antagonist, in the VP suppressed the intake of saccharin but not water or quinine. Microinjections of sulpiride, the dopamine D2 receptor antagonist, and 6‐cyano‐7‐nitroquinoxaline‐2,3‐dione, the AMPA/kainate glutamate receptor antagonist, had no effect on fluid intake. These results reveal that GABA, opioid and D1 receptors in the VP are involved in the consumption of hedonically positive taste stimuli.

Ventral tegmental lesions reduce overconsumption of normally preferred taste fluid in rats

Previous studies have suggested that the brain regions along the taste pathway and its anatomical interfacing with the brain reward system are concerned with palatability-induced consumption. To clarify whether the ventral tegmental area (VTA) is involved in the behavioral expression induced by taste pleasantness, we examined the effects of lesions to the VTA on the consumption of taste stimuli in rats. (1) Bilateral extensive electrolytic lesions to the VTA selectively reduced the consumption of a normally preferred taste fluid (0.1 M sucrose) compared to that of sham-operated animals during a 24-h two-bottle choice test. The consumption of other fluids, including non-preferred taste fluids (HCl and quinine hydrochloride) was not different between the lesioned and sham animals. (2) The injection of midazolam (3 mg/kg), a benzodiazepine agonist, or morphine (2 mg/kg) significantly increased the consumption of 0.1 M sucrose fluids in the sham animals. The same injections, however, failed to increase intake of the 0.1 M sucrose in rats with 6-hydroxydopamine lesions of the VTA. Neither midazolam nor morphine modified the intake of non-preferred quinine (0.0003 M) solution in both the lesioned and sham animals. These results suggest that dopaminergic mediation in the VTA is required to enhance the consumption of normally preferred fluids exclusively.

Brainstem lesions and gustatory function: III. The role of the nucleus of the solitary tract and the parabrachial nucleus in retention of a conditioned taste aversion in rats.

Bilateral electrolytic lesions of the nucleus of the solitary tract (NST) or ibotenic acid lesions of the pontine parabrachial nuclei (PBN) failed to disrupt retention of a preoperatively acquired conditioned taste aversion (CTA) to 0.3 M alanine. For both sham- and NST-lesioned rats, the CTA persisted following 3 nonreinforced conditioned stimulus (CS) presentations. For PBN-lesioned rats, retention was more labile. The preoperatively acquired CTA was extinguished by the 3rd nonreinforced CS exposure. When assessed postoperatively using a novel CS, NST-lesioned rats acquired a new CTA, although they were rendered anosmic with zinc sulfate (P. S. Grigson, T. Shimura, & R. Norgren, 1997). Rats with PBN lesions, however, failed to acquire a second CTA postoperatively. Thus, the PBN is essential for the acquisition of a CTA, but neither of the brainstem gustatory nuclei need be intact for the retention of a preoperatively acquired CTA.

Single unit responses of the amygdala after conditioned taste aversion in conscious rats

Amygdalar neuronal responses to sodium saccharin used as the conditioned stimulus (CS) and to other taste stimuli including sucrose, NaCl, HCl and quinine hydrochloride were recorded before and after the acquisition of conditioned taste aversion (CTA) in freely behaving rats. Of 73 units recorded from the basolateral nucleus of the amygdala (BLA), 17 (23%) and 1 (1%) exhibited facilitatory and inhibitory responses, respectively, to both the CS and sucrose after aversive conditioning to the CS. On the other hand, 3 (5%) and 11 (17%) of 64 units recorded from the central nucleus of the amygdala (Ce) exhibited facilitatory and inhibitory responses, respectively. The responsiveness of these BLA and Ce units to other taste stimuli did not change significantly. These findings that the facilitatory effect was dominant in the BLA, while the inhibitory effect was more frequent in the Ce suggest that the BLA and Ce are differentially involved in CTA.