The hippocampus is composed of anatomically heterogeneous subregions, including CA1, CA3, dentate gyrus and subiculum. A major feature of CA3 is its recurrent collateral circuitry, by which the CA3 pyramidal cells make excitatory synaptic contacts with each other. This feature is unique to this subregion as CA1 pyramidal cells are not extensively interconnected. These anatomical differences inspired David Marr (1971) to conceptualize the hippocampus, and in particular, area CA3, as an autoassociator network that performs pattern storage and retrieval. Because of the unique geometry of the hippocampus, it is difficult to effectively manipulate this region with traditional pharmacological techniques. By utilizing a transgenic model we developed to target this subregion selectively, we can investigate the role of CA3 activity and structure in the development and processing of hippocampal-dependent memory. Last year, we showed that CA3-NMDA R1 receptor KO mice (CA3-NR1 KO mice) are impaired in rapid contextual acquisition (Cravens et al., 2006). To investigate potential mechanisms underlying this behavioral deficit, Juan Belforte has been conducting in vivo multi-electrode recording from area CA3 as well as area CA1. Preliminary results showed a decrease in the neural activity of pyramidal cells in the CA3 region, as evidenced by a lower mean firing rate in the mutants. This decrease seemed to be independent of the familiarity with the environment since it was manifested in novel as well as in familiar linear tracks. Concurrently, the spatial tuning of dorsal CA3 place cell in the linear tracks was impaired in familiar environments. These results suggest that NRs in dorsal CA3 modulate basal as well as the formation of spatial representations. In contrast, CA1 pyramidal cells in these mutants show nearly normal spatial tuning of place cells without changes in basal activity. ? In addition to its role in pattern recognition and the rapid acquisition of contextual information, CA3 is known to be particularly vulnerable to the effects of stress. Chronic stress leads to the development of a number of neurostructural, neuroendocrine and behavioral changes that also appear to be characteristic of patients with clinical depression, and may function to precipitate the onset of major depressive disorders. In animal models, one of the most dramatic effects of chronic stress in the brain is dendritic atrophy in the CA3 region of the hippocampus. It is not yet clear whether and how CA3 atrophy affects hippocampal processing. By examining the role of CA3 under conditions of chronic stress, we can gain additional insight into how hippocampal subregions interact under both normal and pathological conditions. To address this question, it is necessary to manipulate the CA3 stress response independently of other potential contributing variables. In this respect, it is known that dendritic atrophy in CA3 following chronic stress requires NRs. Using the CA3-NR1 KO mice, it is possible to dissect hippocampal subregions genetically with respect to chronic stress. To demonstrate that stress-induced atrophy of CA3 apical dendrites requires functional NRs, Kimberly Christian and Angela Miracle have analyzed Golgi-impregnated hippocampal neurons from CA3-NR1-KO mice and their floxed control littermates following chronic immobilization stress (CIS). Although significant atrophy of apical dendrites in CA3 was observed in floxed control mice, mutant mice were immune to the decrease in both total length and dendritic branch number following CIS. These data suggest a critical involvement of glutamatergic pathways in mediating the hippocampal stress response. Further, this result confirms our initial hypothesis that the CA3 stress response can be manipulated via the NR. Analysis of the dendritic morphology in area CA1 suggests that preventing CA3 atrophy via a functional NR knockout further prevents stress-induced atrophy in CA1, demonstrating a feedforward effect of CA3 atrophy on the CA1 subregion of the hippocampus. The modulation of CA1 stress-induced atrophy by the CA3 NR suggests that CA3 does play a critical role in shaping the hippocampal response to stress. These data indicate that the vulnerability of CA3 neurons to chronic stress may be pivotal in regulating the output of the hippocampus by altering morphological features of CA1 neurons. Furthermore, these correlated structural changes are likely to dramatically alter the integration of synaptic inputs in CA3 and CA1 neurons, an effect it will be possible to isolate following completion of the physiological characterization of these mice outlined in the previous study. ? The third project is focused on the seizure susceptibility of CA3-NR1 KO mice. There has been considerable interest in the selective cell loss in different hippocampal subfields after traumatic injury, hypoxia-ischemia and epileptic insults. This interest arises not only for selective cellular mechanisms but also because of an interest in how the different fields contribute to learning and memory. While the excitotoxicity for selective cell loss is mainly mediated by NRs, the receptor distribution may not be able to determine the subfield selectivity since the expression is relatively ubiquitous among the hippocampal subfields. To evaluate the subfield selective roles of NRs in excitotoxicity, we subjected CA3-NR1 KO mice to kainic acid (KA)-induced excitotoxic insults. We found massive CA1 pyramidal cell loss and dysregulation of GABAergic interneurons following seizures induced by systemic KA administration. These results uncovered a role for CA3 NRs in neurodegeneration and control of excitatory input in area CA1.? The fourth research focus of this CA3 project is to clarify the role of hippocampal oscillations originating from area CA3 (i.e. sharp wave/ripples) in memory consolidation. While accumulating evidence suggests that memory consolidation processes occur during sleep that contribute to a reorganization of hippocampal-dependent memory traces that increasingly depend on cortical regions, the underlying cellular mechanisms of this phenomenon are unknown. One attractive hypothesis would be that the memory trace is transferred or established in other brain regions due to hippocampus-generated oscillatory activity. In particular, sharp wave activity is known to be generated intermittently in area CA3 during slow-wave sleep and immobility periods during waking hours, and to drive propagating 100-200Hz ripple activity from hippocampal to cortical structures. To address the role of CA3-dependentoscillatory activity, we have generated CA3 cell-specific inducible cell ablation mice by crossing floxed-diphtheria toxin receptor mice with the G32-4 Cre line. Preliminary results showed that over 80% of CA3 pyramidal cells were ablated following i.p. administration of diphtheria toxin to Cre/fDTR double transgenic mice. Immunocytochemical, electrophysiological and behavioral analyses are underway. In summary, we hope that the three studies described above will advance our understanding of how the CA3 subregion contributes to memory formation and consolidation in normal and pathological conditions.
| Kishimoto, Yasushi; Nakazawa, Kazu; Tonegawa, Susumu et al. (2006) Hippocampal CA3 NMDA receptors are crucial for adaptive timing of trace eyeblink conditioned response. J Neurosci 26:1562-70 |
| Nakazawa, Kazu; McHugh, Thomas J; Wilson, Matthew A et al. (2004) NMDA receptors, place cells and hippocampal spatial memory. Nat Rev Neurosci 5:361-72 |
