Faculty Bibliography

Anatomical and physiological studies indicate that the amino acid L-glutamate is the excitatory transmitter in sensory afferent pathways to the amygdala and in intraamygdala circuits involving the lateral and basal nuclei. The regional, cellular, and subcellular immunocytochemical localizations of N-methyl-D-aspartate (NMDA) and L-alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA), two major classes of glutamate receptors, were examined in these areas of the amygdala. A monoclonal antibody and a polyclonal antiserum directed against the R1 subunit of the NMDA receptor were used. Each immunoreagent produced distinct distributions of perikaryal and neuropilar staining. Dendritic immunoreactivity was localized primarily to asymmetric (excitatory) synaptic junctions, mostly on spines, consistent with the conventional view of the organization and function of NMDA receptors. Whereas the anti-NMDAR1 antiserum produced sparse presynaptic axon terminal labeling and extensive glial labeling, the anti-NMDAR1 antibody labeled considerably fewer glia and many more presynaptic axon terminals. Labeled presynaptic terminals formed asymmetric and symmetric synapses, suggesting presynaptic regulation of both excitatory and inhibitory transmission. Immunoreactivity for different subunits of the AMPA receptor (GluR1, GluR2/3, and GluR4) was uniquely distributed across neuronal populations, and some receptor subunits were specific to certain cell types. Immunoreactivity for GluR1 and Glu2/3 was predominantely localized to dendritic shafts and was more extensive than that of GluR4 due to heavy labeling of proximal portions of dendrites. The distribution of GluR4 immunoreactivity was similar to NMDAR1: GluR4 was seen in presynaptic terminals, glia, and dendrites and was primarily localized to spines. The presynaptic localization of GluR4 in the absence of GluR2 suggests glutamate-mediated modulation of presynaptic Ca++ concentrations. These data add to our understanding of the morphological basis of pre- and postsynaptic transmission mechanisms and synaptic plasticity in the amygdala

The lateral nucleus of the amygdala (LA) is the first site in the amygdala where the plasticity underlying fear conditioning could occur. We simultaneously recorded from multiple LA neurons in freely moving rats during fear conditioning trials in which tones were paired with foot shocks. Conditioning significantly increased the magnitude of tone-elicited responses (often within the first several trials), converted unresponsive cells into tone-responsive ones, and altered functional couplings between LA neurons. The effects of conditioning were greatest on the shortest latency (less than 15 ms) components of the tone-elicited responses, consistent with the hypothesis that direct projections from the auditory thalamus to LA are an important link in the circuitry through which rapid behavioral responses are controlled in the presence of conditioned fear stimuli

The amygdaloid complex is involved in associational processes, such as the formation of emotional memories about sensory stimuli. However, the anatomical connections through which the different amygdaloid nuclei process incoming information and communicate with the other amygdaloid nuclei, is poorly understood. As part of an ongoing project aimed at elucidating the intrinsic connections of the rat amygdaloid complex, we injected the anterograde tracer PHA-L (Phaseolus vulgaris-leucoagglutinin) into different rostrocaudal levels of the basal nucleus of the amygdala in 21 rats and analyzed the distribution of labeled fibers and terminals throughout the amygdaloid complex. The connectional analysis, together with cytoarchitectonic observations, suggested that contrary to previous notions the basal nucleus in the rat has three divisions: magnocellular, intermediate, and parvicellular. The magnocellular division has heavy reciprocal connections with the lateral portion of the parvicellular division and the intermediate division projects weakly to the parvicellular division, whereas the projection from the medial portion of the parvicellular division to the intermediate division is heavy and the lateral and medial portions of the parvicellular division are only weakly interconnected, as are the magnocellular and intermediate divisions. The main intraamygdaloid targets of the basal nucleus projections are the nucleus of the lateral olfactory tract, the anterior amygdaloid area, the medial and capsular divisions of the central nucleus, the anterior cortical nucleus, and the amygdalohippocampal area. Our findings provide the most detailed understanding of the intra-amygdala connections of the basal nucleus to date and show that the connections within the basal nucleus and between the basal nucleus and other amygdaloid areas are more widespread and topographically organized than previously recognized

Classical fear conditioning was used in the present study as a model for investigating emotional learning and memory in human subjects with lesions to the medial temporal lobe. Animal studies have revealed a critical role for medial temporal lobe structures, particularly the amygdala, in simple and complex associative emotional responding. Whether these structures perform similar functions in humans is unknown. On both simple and conditional discrimination tasks, unilateral temporal lobectomy subjects showed impaired conditioned response acquisition relative to control subjects. This impairment could not be accounted for by deficits in nonassociative sensory or autonomic performance factors, or by differences in declarative memory for the experimental parameters. These results show that temporal lobe structures in humans, as in other mammals, are important components in an emotional memory network

Five healthy male subjects participated in a classical conditioning experiment, and positron emission tomography (PET) was used to compare regional cerebral blood flow before and after conditioning. The subjects participated in three different experimental phases. In the first (habituation) phase they listened to 24 repetitions of a tone with random intervals. In the second (acquisition) phase, the tone was paired with a brief shock to the wrist. In the third (extinction) phase, the tone was presented alone again. 15OPET scans were taken during the habituation and extinction phases. Because the habituation and extinction phases were similar, any difference in blood flow to the tones presented during extinction probably reflected conditioning that occurred during the acquisition phase. Statistical parametric mapping (SPM) analysis of the PET data showed significantly increased activation in the right hemisphere in the orbito-frontal cortex, dorsolateral prefrontal cortex, inferior and superior frontal cortices, and inferior and middle temporal corticies. The only activated areas in the left hemisphere were area 19 and the superior frontal cortex. The results are interpreted as evidence for the involvement of cortical areas in human classical conditioning

The emotional reactivity of rats with lesions of the dorsal portion of medial prefrontal cortex (mPFC) was examined using a classical fear conditioning paradigm. Conditioned fear behavior (freezing responses) was measured during both the acquisition and extinction phases of the task. Lesions enhanced fear reactivity to both the conditioned stimulus (CS) and contextual stimuli during both phases, suggesting that dorsal mPFC lesions produce a general increase in fear reactivity in response to fear conditioning. M. A. Morgan, L. M. Romanski, and J. E. LeDoux (1993) found that lesions just ventral to the present lesions had no effect during acquisition of the same task and prolonged the fear response to the CS (but not the context) during extinction. Thus, both dorsal and ventral regions of mPFC are involved in the fear system, but each modulates different aspects of fear responsivity

Lesions placed in the rostral perirhinal cortex (rPRh) after fear conditioning interfere with the expression of conditioned fear responses elicited by auditory and visual conditioned stimuli when these stimuli are presented in a context that differs from the conditioning context. The present study examined whether lesions of the rPRh have similar effects when animals are tested in the conditioning context. Two days after male rats received classical fear conditioning, involving the pairing of an auditory conditioned stimulus (CS) with footshock, bilateral electrolytic lesions were produced in the rPRh. Five days later conditioned freezing behavior was measured during a 60-s exposure to the CS in a novel context and then 1 hr later in the conditioning context. There were 3 major findings. First, rPRh-lesioned animals froze significantly less than controls to the CS in the novel context, thus confirming previously reported findings. Second, rPRh-lesioned animals also froze less than controls to the CS in the conditioning context, but froze significantly more to the CS in the conditioning than in the novel context, suggesting that at least part of the deficit in the novel context is due to the absence of contextual cues. Third, animals with rPRh lesions froze significantly less than controls to the conditioning context itself. This latter finding suggests that rPRh lesions interfere with contextual processing and that the improvement of performance in the conditioning context might have been even greater had lesioned animals been able to fully process contextual cues.(ABSTRACT TRUNCATED AT 250 WORDS)