While a number of early studies demonstrated that hippocampal damage attenuates the expression of recent, but not remotely trained tasks, an emerging body of evidence has shown that damage to, or inactivation of, the hippocampus often impairs recall across a wide range of training–testing intervals. Collectively, these data suggest that the time course of hippocampal involvement in the storage or recall of previously-acquired memories may differ according to hippocampal subregion and the particular learning task under consideration. The present study examined the contributions of dorsal (DH) and ventral (VH) hippocampus to the expression of previously-acquired trace fear conditioning, a form of Pavlovian conditioning in which the offset of an initially neutral cue or cues and the onset of an aversive stimulus is separated by a temporal (trace) interval. Specifically, either saline or the GABA-A agonist muscimol was infused into DH or VH prior to testing either 1, 7, 28, or 42days after trace fear conditioning. The results revealed a marked dissociation: pre-testing inactivation of DH failed to impair performance at any time-point, while pre-testing inactivation of VH impaired performance at all time-points. Importantly, pre-testing inactivation of VH had no effect on the performance of previously-acquired delay conditioning, suggesting that the deficits observed in trace conditioning cannot be attributed to a deficit in performance of the freezing response. Collectively, these data suggest that VH, but not DH, remains a neuroanatomical locus critical to the recall or expression of trace fear conditioning over an extended period of time.

The dorsal and ventral subregions of the hippocampus likely play dissociable roles in some forms of learning. For example, we have previously demonstrated that temporary inactivation of ventral, but not dorsal, hippocampus dramatically impaired the acquisition of trace fear conditioning, while temporary inactivation of dorsal, but not ventral, hippocampus impaired spatially guided reinforced alternation (Czerniawski et al. (2009) Hippocampus 19:20–32). Importantly, emerging data suggest that lesions, temporary inactivation, and NMDA receptor antagonism within these subregions can produce quite different patterns of behavioral effects when administered into the same region. Specifically, while neither lesions nor temporary inactivation of dorsal hippocampus impair the acquisition of trace fear conditioning, learning in this paradigm is severely impaired by pre‐training administration of the NMDA receptor antagonist dl ‐2‐phosphonovaleric acid (APV) in dorsal hippocampus; the effect of NMDA receptor antagonism within ventral hippocampus on the acquisition and expression of trace conditioning, or on learning in general, has not yet been systematically explored. The present study extends our previous work examining the differential effect of lesions or inactivation of the dorsal and ventral hippocampal subregions by systematically examining the effect of regionally selective pre‐training or pre‐testing administration of APV on the acquisition and expression of trace and contextual fear conditioning. The results of these studies demonstrate that while pre‐training NMDA receptor antagonism within either the dorsal or ventral subregion of the hippocampus impaired the acquisition of both trace and contextual conditioning, pre‐testing NMDA receptor antagonism within ventral, but not dorsal, hippocampus impaired the expression of previously‐acquired trace and contextual fear conditioning. These data suggest that selectively manipulating the integrity of individual subregions may result in compensatory mechanisms that can support learning, and that NMDA‐dependent plasticity within both dorsal and ventral hippocampus is normally required for the acquisition and maintenance of memory in trace and contextual fear conditioning. © 2011 Wiley Periodicals, Inc.

Long-lasting, experience-dependent changes in synaptic strength are widely thought to underlie the formation of memories. Many forms of learning-related plasticity are likely mediated by NMDA receptor activation and plasticity-related gene expression in brain areas thought to be important for learning and memory, including the hippocampus. Here, we examined the putative role of activity-regulated cytoskeletal-associated protein (Arc), an immediate-early gene (IEG) whose expression is tightly linked to the induction and maintenance of some forms of neuronal plasticity, in hippocampus-dependent and hippocampus-independent forms of learning. The extent to which learning-induced Arc expression may depend on NMDA receptor activation was also assessed. First, we observed an increase in Arc gene and protein products in both dorsal hippocampus (DH) and ventral hippocampus (VH) of male Sprague Dawley rats after hippocampus-dependent trace and contextual fear conditioning, but not after hippocampus-independent delay fear conditioning. Specific knockdown of Arc using antisense oligodeoxynucleotides (ODNs) in DH or VH attenuated the learning-related expression of Arc protein, and resulted in a dramatic impairment in trace and contextual, but not delay, fear conditioning. Finally, pretraining infusions of the NMDA receptor antagonist APV into the DH or VH blocked the learning-induced enhancement of Arc in a regionally selective manner, suggesting that NMDA receptor activation and Arc translation are functionally coupled to support hippocampus-dependent memory for fear conditioning. Collectively these results provide the first evidence suggesting that NMDA receptor-dependent expression of the IEG Arc in both DH and VH likely underlies the consolidation of a variety of forms of hippocampus-dependent learning.