Search

Epileptiform EEG activity associated with ischemia can contribute to early damage of hippocampal neurons, and seizure activity may also lead to dysfunction in extrahippocampal regions. In this study, seizure activity associated with the four-vessel occlusion model of cerebral ischemia was monitored using chronically implanted electrodes in the CA1/subicular region, the perirhinal cortex, and the prefrontal cortex of the rat. Background EEG amplitude was reduced in all recording sites during occlusion, but spiking and bursting activity was also observed. Seizure activity occurred in most animals during the first several hours of reperfusion, but was not observed on subsequent days. Epileptiform spikes and bursts were often synchronous between two or three recording sites, and spikes in the CA1 region also often occurred just prior to spikes in other sites. These results demonstrate that the four-vessel occlusion model can lead to patterns of seizure activity in the hippocampus, prefrontal and perirhinal cortices.

One of the routine memory abilities impaired in amnesic patients with temporal-lobe damage is object-recognition memory--the ability to discriminate the familiarity of previously encountered objects. Reproducing this impairment has played a central role in animal models of amnesia during the past two decades, and until recent years most of the emphasis was on describing how hippocampal damage could impair object recognition. Today most investigators are looking outside the hippocampus to explain the impairment. This paper reviews studies of object-recognition memory in rats with hippocampal damage produced by ablation, fornix transection, or forebrain ischemia. Some new perspectives on previous findings reinforce the conclusion that damage to the hippocampus has little if any impact on the ability to recognize objects, while damage in some areas outside the hippocampus is far more effective. The few circumstances in which hippocampal damage can impair performance on object-recognition tasks are situations where ancillary abilities are likely to play a significant role in supporting task performance. Some of the factors that contributed to the origins and persistence of the hippocampalcentric view of object-recognition are considered, including lesion confounds, failure to distinguish between impaired task performance and impairment of a memory ability, and disproportionate attention to a few lesion studies in monkeys, even though the hypothesis was tested far more times in rats, under a greater variety of conditions, and rejected on nearly every occasion.

We review evidence from experiments conducted in our laboratory on retrograde amnesia in rats with damage to the hippocampal formation. In a new experiment reported here, we show that N-methyl-D-aspartate (NMDA)-induced hippocampal damage produced retrograde amnesia for both hidden platform and two-choice visible platform discriminations in the Morris water task. For both problems there was a significant trend for longer training-surgery intervals to be associated with worse retention performance. Little support is offered by our work for the concept that there is a process involving hippocampal-dependent consolidation of memories in extrahippocampal permanent storage sites. Long-term memory consolidation may take place within the hippocampus. The hippocampus may be involved permanently in storage and/or retrieval of a variety of relational and nonrelational memories if it was intact at the time of learning, even involving information which is definitely not affected in anterograde amnesia after hippocampal damage.

Three experiments examined the effects of perirhinal cortex lesions on rats' retrograde and anterograde memory for object-discriminations and water-maze place-memory problems. In Experiment 1, rats learned two object-discriminations - the first was learned 2 weeks before surgery and the second 24 h before surgery. Rats with perirhinal cortex lesions displayed mildly impaired retention of both object discriminations, with no evidence of a temporal gradient. They also learned a new discrimination at a normal rate, but were impaired on a retention test 24 h later. In Experiment 2, rats learned two water-maze place problems, conducted in different rooms - the first was learned 4 weeks before surgery and the second during the week immediately before surgery. Rats with perirhinal cortex lesions displayed deficits on the early retention trials of both place problems, but they quickly relearned both problems. In Experiment 3, rats with perirhinal cortex lesions learned a new place problem at a normal rate and performed as well as control rats on a retention test 3 weeks later. Although some of the results are consistent with the conclusion that perirhinal damage disrupts storage or retrieval of place information acquired before surgery, additional considerations suggest instead a role for perirhinal cortex in the representation of nonspatial information that makes a useful but nonessential contribution to water-maze performance.

Using a within-subjects design, rats were trained on two place-memory problems and five object-discrimination problems at different intervals prior to receiving either ibotenate lesions of the hippocampal formation or sham surgery. Places # 1 and 2 were fixed-platform water-maze tasks that were run in different rooms and they were learned during the 14th and 2nd week before surgery, respectively. Object-discrimination problems # 1-5 were learned during the 13th, 10th, 7th, 4th, and 1st week before surgery, respectively. Rats with hippocampal lesions displayed impaired retention of both Place problems with no evidence of a temporal gradient to the impairment. In contrast to their retrograde place-memory deficits, the hippocampal rats displayed normal retention of the five object-discriminations that were learned before surgery. Hippocampal lesions had similar consequences for anterograde learning, as the lesioned rats were impaired in acquisition of a new water-maze problem that was run in a third room (Place #3), whereas they showed normal acquisition of two new object-discriminations. The findings indicate that the hippocampal formation is not required for long-term consolidation of information underlying accurate performance of object-discriminations, and that its critical role in memory for places persists for at least 14 weeks, and probably for as long as those memories exist.

The literature on the role of the hippocampus in object-recognition contains a paradox: Transient forebrain ischemia (ISC) produces hippocampal damage and severe deficits on the delayed nonmatching-to-sample (DNMS) task, yet hippocampal ablation (ABL) produces milder deficits. Experiment 1 confirmed that pretrained rats display severe DNMS deficits following ISC, but not ABL. Ischemia produced loss of CA1 neurons, but no obvious extrahippocampal damage. In Experiments 2 and 3, ISC rats from Experiment 1 received ABL, and ABL rats received ISC; neither treatment affected DNMS performance. In Experiment 4, rats that received ISC followed 1 hr later by ABL displayed only mild deficits. It is hypothesized that ISC-induced DNMS deficits are due to extrahippocampal damage produced by pathogenic processes that involve the hippocampus.

Transient cerebral ischemia can produce irreversible neuronal damage and permanent learning and memory impairments in humans. This study examined whether ischemia-induced brain damage in rats results in impairments on the delayed nonmatching-to-sample (DNMS) task, a nonspatial recognition task analogous to tests on which amnesic patients display impairments. Male Wistar rats received either sham surgery or 20-min forebrain ischemia induced by bilateral carotid occlusion and hypotension. Four weeks after surgery, ischemic rats were significantly impaired in both learning and performing the DNMS task at retention intervals up to 5 min. Extensive presurgical training did not reduce this impairment. Observable cell loss in ischemic rats was limited to CA1 pyramidal neurons and a subset of cells in the dentate gyrus. The results indicate that ischemic damage to the hippocampus in rats results in recognition memory deficits similar to those produced by ischemic damage in humans.