Encoding versus retrieval of spatial memory: Double dissociation between the dentate gyrus and the perforant path inputs into CA3 in the dorsal hippocampus

Abstract

The hippocampus is an essential neural structure for spatial memory. Computational models suggest that the CA3 subregion of the hippocampus plays an essential role in encoding and retrieval of spatial memory. The perforant path (PPCA3) and dentate gyrus (DG)-mediated mossy fibers (MFs) compose major afferent inputs into CA3. A possible functional dissociation between these afferent inputs was attempted using a simple navigation test (i.e., the modified Hebb-Williams maze). Behavioral testing was combined with electrolytic lesions of PPCA3 or neurotoxic lesions of the DG, to eliminate each afferent input into CA3. Lesions in either afferent input into CA3 affected learning of an effective navigational path on the maze. The contributions of the two CA3 afferent inputs, however, were different regarding encoding and retrieval of memory measured based on indices operationally defined for the behavioral paradigm (i.e., encoding, the number of errors reduced within a day; retrieval, the number of errors reduced between days). The DG-lesioned animals exhibited deficits regarding the encoding index, but not the retrieval index, whereas the PPCA3-lesioned rats displayed deficits regarding the retrieval index, but not the encoding index. The results suggest that the two major afferent inputs of CA3 may contribute differentially to encoding and retrieval of spatial memory.

The role of the direct perforant path input to the CA1 subregion of the dorsal hippocampus in memory retention and retrieval

“…Given the observed selective deficits in retrieval of spatial cues necessary for navigation of a maze and previous evidence implicating CA1 in retrieval (Hasselmo, 1995;Hasselmo et al, 1995;Sybirska et al, 2000;Kesner et al, 2004;Lee and Kesner, 2004b;Jerman et al, 2006;Rolls and Kesner, 2006), we reasoned that similar selective deficits may be apparent in additional CA1-dependent tasks. It has been shown previously that CA3-lesioned rats are significantly impaired on the acquisition, but not retention of contextual fear-conditioning whereas, CA1 lesioned rats are significantly impaired on retention but not acquisition (Lee and Kesner, 2004a).…”

“…Encoding of information acquired in a Hebb-Williams maze or in contextual fear conditioning, using a within-days analysis, is impaired by targeted lesions to the CA3 and dentate gyrus (DG) subregions, but not from targeted lesions to the CA1 subregion. In contrast, using a between-days analysis, retention and retrieval is disrupted following CA1, but not following CA3 or DG lesions (Lee and Kesner, 2004a;Lee and Kesner, 2004b;Jerman et al, 2006). Impairments have also been demonstrated in delay-dependent retrieval without impairing immediate recall or encoding of spatial information after infusing glutamatergic antagonists (Lee and Kesner, 2002), or cyclooxygenase-2 inhibitors (Sharifzadeh et al, 2006) into the CA1 subregion, but not when infused into the CA3 subregion.…”

“…The present version was modified to facilitate spatial and/or place learning with a moderate level of difficulty using a fixed configuration of inner barriers (i.e., Problem 4, described previously) (Rabinovitch and Rosvold, 1951;Lee and Kesner, 2004b;Rogers and Kesner, 2004). Two boxes, diagonally positioned on opposite sides of the maze, were designated as the start and goal boxes.…”

“…Errors were then averaged across blocks of the first and last 3 trials for each day to produce a general learning index. It was assumed that encoding and retrieval interact on a trial-by-trial basis (one reason it makes it difficult to functionally differentiate them behaviorally) and thus indices were developed (Lee and Kesner, 2004b;Rogers and Kesner, 2004) to dissociate the two processes. A within-day (W/D) encoding score was calculated by subtracting B (the average number of errors across the last three trials of Day 1) from A (the average number of errors across the first three trials of Day 1), and D (the average number of errors across the last three trials on Day 2) from C (the average number of errors across the first three trials on Day 2).…”

Disconnection analysis of CA3 and DG in mediating encoding but not retrieval in a spatial maze learning task

“…Recent experimental evidence supports this theoretical hypothesis. Lee and Kesner (2002) manipulated the N-methyl-Daspartate (NMDA) receptors in CA3 by injecting APV (amino-5-phosphonovaleric acid, a pharmacological blocker for the NMDA receptors) selectively into CA3 in a delayed non-matching-toplace (DNMP) task. They found that rats injected with APV were not impaired in performing the DNMP task in a familiar environment in which they had been trained.…”