Faculty Bibliography

The conditioning of fear responses to a simple acoustic stimulus (pure tone) paired with footshock can be mediated by the transmission of auditory information to the lateral nucleus of the amygdala from either the auditory thalamus or the auditory cortex. We examined the processing capacity of the thalamo-amygdala pathway by making lesions of the auditory cortex and testing the extent to which conditioned fear responses generalized to tones other than the one paired with footshock. Two studies were performed, one in an anatomically constrained computational model of the fear conditioning network and the other in rats. Stimulus generalization was unaffected in both. These findings support the validity of the model as an approach to studying the neural basis of conditioned fear learning, and in addition suggest that the thalamo-amygdala pathway, possibly by the use of population coding, is capable of performing at least crude stimulus discriminations

Previous studies suggest that the left and right amygdalae are interconnected in rodents. The origin and topography of these connections have, however, remained obscure. In the present study, we investigated the interamygdaloid projections originating in the different divisions of the basal and accessory basal nuclei of the rat amygdala by using the Phaseolus vulgaris leucoagglutinin anterograde tract-tracing technique. The basal nucleus gave rise to substantial interamygdaloid projections. However, the density of the projections depended on the location of Phaseolus vulgaris leucoagglutinin injection in the basal nucleus. The magnocellular and intermediate divisions projected heavily to the homonymous regions on the contralateral side, as well as to the nucleus of the lateral olfactory tract. The parvicellular division projected lightly to the homonymous region on the contralateral side, to the contralateral anterior amygdaloid area and to the medial division of the central nucleus. The contralateral projections originating in the accessory basal nucleus were light compared to those of the basal nucleus. These data indicate that interamygdaloid connections in the rat brain are extensive and topographically organized. Via these connections, one amygdala may rapidly activate the contralateral side. This may explain, for example, why the epileptic seizures in one amygdala spread contralaterally and cause the development of independent seizure activity in kindling model of temporal lobe epilepsy

The amygdaloid complex plays an important role in the detection of emotional stimuli, the generation of emotional responses, the formation of emotional memories, and perhaps other complex associational processes. These functions depend upon the flow of information through intricate and poorly understood circuitries within the amygdala. As part of an ongoing project aimed at further elucidating these circuits, we examined the intra-amygdaloid connections of the accessory basal nucleus in the rat. In addition, we examined connections of the anterior cortical nucleus and amygdalahippocampal area to determine whether portions of these nuclei should be included in the accessory basal nucleus (as some earlier studies suggest). Phaseolus vulgaris leucogglutinin was injected into different rostrocaudal levels of the accessory basal nucleus (n = 12) or into the anterior cortical nucleus (n = 3) or amygdalahippocampal area (n = 2). The major intra-amygdaloid projections from the accessory basal nucleus were directed to the medial and capsular divisions of the central nucleus, the medial division of the amygdalohippocampal area, the medial division of the lateral nucleus, the central division of the medial nucleus, and the posterior cortical nucleus. The projections originating in the anterior cortical nucleus and the lateral division of the amygdalohippocampal area differed from those originating in the accessory basal nucleus, which suggests that these areas are not part of the accessory basal nucleus. The present findings and our previous data suggest that each of the deep amygdaloid nuclei have different intra-amygdaloid connections. The pattern of these various connections suggests that information entering the amygdala from different sources can be integrated only in certain amygdaloid regions

Conditioned fear in rats was assessed for the effects of pretraining amygdala lesions (unilateral vs. bilateral) across unconditioned stimulus (US) modalities (white noise vs. shock). In contrast to sham controls, unilateral amygdala lesions significantly reduced conditioned freezing responses, whereas bilateral amygdala lesions resulted in a nearly complete lack of freezing to both the conditioned stimulus (CS) and the context. The lesion effects were more pronounced for CS conditioning but were consistent across US modalities. It was concluded that white noise can serve as an effective US and that unilateral amygdala lesions attenuate but do not eliminate conditioned fear in rats. The results support our interpretation of a recent fear conditioning study in humans (K. S. LaBar, J. E. LeDoux, D. D. Spencer, & E. A. Phelps, 1995)

Stimulation of the medical geniculate body elicits extracellular single unit responses in the lateral nucleus of the amygdala that are dependent upon glutamatergic neurotransmission [Li et al. (1995) Exp. Brain Res., 105-87-100]. In the present study, we examined the contribution of inhibitory amino acid transmission to these excitatory responses. Antagonists of GABAA or GABAB receptors were delivered microiontophoretically to cells activated by stimulation of the medial geniculate body. Blockade of GABAA receptors with bicuculline resulted in a pronounced increase in evoked short latency unit responses (4-8 ms). In some cases, cells that were not responsive to the stimulation became responsive in the presence of bicuculline. In contrast, delivery of GABAB antagonists, Phaclofen or 2-OH-saclofen, did not affect these short-latency responses. Using paired-pulse stimulation, both short (< 30 ms) and longer (> 50 ms) latency inhibitory processes were revealed. GABAA blockade eliminated the short latency inhibition and GABAB blockade eliminated the longer latency inhibition in most cells. These results suggest that the activation of GABAA and GABAB receptors differentially regulate glutamatergic synaptic transmission in the auditory thalamo-amygdala pathway. Moreover, our findings suggest that at least part of this regulation is via a feedforward mechanism. We tested the sufficiency of feedforward inhibition to account for the data using a simple computational model that incorporates the results presented here

A classical fear conditioning paradigm was used to examine the effect of acute ethanol on the acquisition of context conditioning, a hippocampal-dependent associative task, and tone conditioning, a hippocampal-independent task. Administration of ethanol before the presentation of seven tone-shock pairings severely disrupted the acquisition of context conditioning, but had only a slight effect on tone conditioning, when conditioned fear was measured 48 h later. This effect was dose dependent: a dose of 0.5 g/kg had no effect on either context or tone conditioning, while doses of 1.0 and 1.5 g/kg disrupted context conditioning by 78-86%, and tone conditioning by 9-17%. Subsequent experiments indicated that ethanol's preferential effect on context conditioning could not be attributed to the fact that context conditioning is weaker than tone conditioning, ethanol-induced changes in motivational state or state-dependent learning. The effect of ethanol on stimulus-induced increases in hippocampal and neocortical expression of c-fos mRNA, a marker for changes in metabolic neuronal activity, was also examined. Ethanol completely blocked the induction of hippocampal c-fos mRNA by exposure to the conditioning context alone or seven tone-shock pairings, but only attenuated neocortical responses to these stimuli. Together, these results suggest that ethanol disrupts hippocampal-dependent learning by preferentially impairing stimulus processing at the level of the hippocampus

The lateral nucleus of the amygdala (LA), a key component of the fear conditioning circuitry, receives a rapid but relatively impoverished auditory input from the auditory thalamus and a slower but richer input from the auditory cortex. We examined in urethane anesthetized rats whether individual cells in the LA receive convergent inputs from these two areas, and whether different postsynaptic receptors contribute to the temporally separated excitations over the two pathways. With both extracellular and intracellular recordings, individual cells could be activated by stimulation of each pathway. In extracellular recordings iontophoretic application of the N-methyl-D-aspartate (NMDA) receptor antagonist APV and the L-alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) receptor antagonist CNQX demonstrated that synaptic transmission in both pathways depends on AMPA receptors, whereas transmission in the thalamic pathway also depends on the involvement of NMDA receptors. The involvement of NMDA receptors in synaptic activation of the LA from the thalamus but not the cortex was confirmed in intracellular recordings using systemic injections of the NMDA antagonist MK-801. The slow time course of NMDA currents could provide LA cells with a mechanism to integrate the inputs arriving rapidly from the thalamus and somewhat later from the cortex, thus allowing the LA to integrate signals in the two pathways during the acquisition and expression of conditioned fear reactions

Information from most of the sensory modalities enters the amygdala via the lateral nucleus. The olfactory information, however, arrives at the amygdala through the superficial nuclei, including the periamygdaloid cortex. To find out whether the olfactory information can modulate the processing of sensory information in the lateral nucleus we injected Phaseolus vulgaris leucoagglutinin, an anterograde axonal tracer, into the different divisions of the periamygdaloid cortex. We found that the PAC division of the periamygdaloid cortex projects to the ventrolateral and medial divisions, but not the dorsolateral division, of the lateral amygdaloid nucleus. Therefore, the projection from the PAC to the lateral nucleus provides a route, by which the olfactory information may become associated with other sensory modalities. Also, together with our previous finding that the lateral nucleus projects to the periamygdaloid cortex, the present data demonstrate that the lateral nucleus and the PAC are reciprocally connected