Faculty Bibliography

The bed nucleus of the stria terminalis (BNST) is believed to be a critical relay between the central nucleus of the amygdala (CE) and the paraventricular nucleus of the hypothalamus in the control of hypothalamic-pituitary-adrenal (HPA) responses elicited by conditioned fear stimuli. If correct, lesions of CE or BNST should block expression of HPA responses elicited by either a specific conditioned fear cue or a conditioned context. To test this, rats were subjected to cued (tone) or contextual classical fear conditioning. Two days later, electrolytic or sham lesions were placed in CE or BNST. After 5 days, the rats were tested for both behavioral (freezing) and neuroendocrine (corticosterone) responses to tone or contextual cues. CE lesions attenuated conditioned freezing and corticosterone responses to both tone and context. In contrast, BNST lesions attenuated these responses to contextual but not tone stimuli. These results suggest CE is indeed an essential output of the amygdala for the expression of conditioned fear responses, including HPA responses, regardless of the nature of the conditioned stimulus. However, because lesions of BNST only affected behavioral and endocrine responses to contextual stimuli, the results do not support the notion that BNST is critical for HPA responses elicited by conditioned fear stimuli in general. Instead, the BNST may be essential specifically for contextual conditioned fear responses, including both behavioral and HPA responses, by virtue of its connections with the hippocampus, a structure essential to contextual conditioning. The results are also not consistent with the hypothesis that BNST is only involved in unconditioned aspects of fear and anxiety

Plasticity in the lateral nucleus of the amygdala is thought to be critical for the acquisition of Pavlovian fear conditioning. The pathways that transmit auditory conditioned stimulus information originate in auditory processing regions of the thalamus and cortex, but the pathways mediating transmission of unconditioned stimuli to the amygdala are poorly understood. Recent studies suggest that somatosensory (footshock) unconditioned stimulus information is also relayed in parallel to the lateral nucleus of the amygdala from the thalamus (the posterior intralaminar thalamic complex, PIT) and the cortex (parietal insular cortex). In the present study we reexamined this issue. Our results showed that bilateral electrolytic lesions of the PIT alone blocked fear conditioning, whereas bilateral excitotoxic PIT lesions had no effect. These electrolytic PIT lesions did not affect fear conditioning using a loud noise as unconditioned stimulus, defining the effects of PIT lesions as a disruption of somatosensory as opposed to auditory processing. Finally, we performed combined bilateral excitotoxic lesions of the PIT nuclei and electrolytic lesions of the parietal insular cortex. These, like excitotoxic lesions of PIT alone, had no effect on the acquisition of fear conditioning. Thus, somatosensory regions of the thalamus and cortex may well be important routes of unconditioned stimulus transmission to the amygdala in fear conditioning, but information about the unconditioned somatosensory stimulus is also transmitted from other sources that send fibers through, but do not form essential synapses in, the thalamus en route to the amygdala

Several years ago, we found that lesions of ventral medial prefrontal cortex (mPFCv) disrupted performance during the extinction component of a classical fear conditioning task without affecting acquisition performance. We called this emotional perseveration, hypothesizing that mPFCv may normally act to inhibit fear responses to a conditioned stimulus (CS) when the CS no longer signals danger. Subsequent studies have supported this hypothesis, showing that mPFCv is crucial for the memory of prior extinction training. The present study examined the effects of mPFCv lesions made after training. Such lesions resulted in reduced freezing to contextual stimuli and normal responding to the CS presented alone during a retention test. Rats were then subjected to extinction trials (CS without US) over multiple days. In contrast to pre-training lesions, post-training lesions had little effect on extinction rate. All rats were given additional training. Lesioned rats expressed greater fear reactions than controls, indicating that prior extinction was less effective in them. Lesioned rats also showed resistance to extinction during reextinction trials, confirming our earlier finding that lesions made before training weaken the effectiveness of extinction trials. These results suggest three conclusions. First, an intact mPFCv during acquisition may protect the animal from prolonged responding during extinction trials following brain insult. Second, changes in mPFCv may predispose subjects toward enhanced fear reactions that are difficult to extinguish when reexposed to fearful stimuli, due to a diminished capacity to benefit from the fear-reducing impact of prior extinction experience. Third, contextual cues processed by mPFCv may influence extinction performance