A critical element of achieving the project aims involves establishing an in-house animal resource to yield a reliable supply of behaviorally characterized young and aged Long-Evans rats, replicating the model of normal cognitive aging we and others have exploited in over two decades of research. Unfortunately, NIA IRP policies instituted shortly after my group joined the institute have barred the importation of Long-Evans rats from Charles River Laboratories, despite their approved vendor status at other biomedical research institutions and intramural research programs of the NIH. We are hopeful that the pending recruitment of new veterinary leadership of the NIA IRP animal program (a position that has remained vacant since 2008) will lead to progress on this long-standing issue. In the interim, taking advantage of collaborative support from colleagues at Johns Hopkins University, we recently completed a study demonstrating that neuronal histone deacetylase 2 protein in the hippocampus is up-regulated in response to training in a spatial learning and memory task in young adult rats, and that this response is blunted in aged subjects with deficits in cognitive capacities that require the hippocampus. A preliminary report of these findings was presented at the 2009 Society for Neuroscience meeting, and a full-length manuscript is nearing completion for submission. Normal cognitive aging is frequently characterized as a decline in hippocampal-dependent memory alongside a deterioration of executive function. In a second recent study, we assessed the consequences of these impairments at the level of the interaction between memory systems. For this purpose, we developed a water-based T-maze procedure in which escape was contingent on learning a place (e.g., go East) or response strategy (go Right). Prior to testing, male Long-Evans rats were evaluated using a standard water maze task and classified on the basis of spatial learning ability as Aged Unimpaired (AU) or Aged Impaired (AI), relative to Young adults (Y). Subsequent T-maze training required learning each of the four potential task contingencies successively (Place/East;Response/Right;Place/West;Response/Left). Key results include the observation that learning scores from the initial behavioral characterization selectively correlated with the number of trials required to master the Place component of the T-maze. Results for Response learning, in contrast, were unrelated to performance in the standard water maze. Further analysis revealed an interaction between cognitive status and strategy learning such that, while Y animals acquired Place more quickly than Response strategies, AI rats exhibited the opposite pattern. Performance among AU subjects was intermediate. These findings point to a competitive interaction between place and response learning systems, and consistent with this interpretation, error analysis revealed that Y rats made incorrect choices predominantly by following a Place strategy, whereas errors among AI rats disproportionately involved use of a response strategy. In a final test, animals were challenged for their ability to switch between Response and Place strategies within a session. Whereas all but one of the Y and AU rats succeeded, only half the AI rats achieved criterion performance at this phase. Together our results suggest that normal cognitive aging involves a disruption in the competition between hippocampus- and striatum-dependent learning systems for the control of behavior. Findings from this study are scheduled for presentation at the 2010 annual Society for Neuroscience meeting. Converging evidence has linked histone acetylation to hippocampus-dependent memory, including reports that administration of histone deacetylase inhibitors (HDACis) can enhance cognitive function. In a related investigation aimed at the development of potential interventions for cognitive aging, we recently evaluated the pharmacological and behavioral effects of the novel HDACi, EVX001688, under an MCRACA arrangement with EnVivo Pharmaceuticals, Inc.. This drug inhibits multiple HDACs and is highly brain penetrant. Thus, systemic administration in young adult, male Long-Evans rats produced a robust, dose-related increase in histone H3 and H4 across all subfields of the hippocampus at 90 minutes post-injection. In separate groups of rats, next we tested the cognitive effects of the same doses of EVX001688 using a novel, one-session redundant place/cue water maze task in which rats swam to a cued escape platform maintained in a constant location across trials. Pretraining drug administration yielded a dose-dependent benefit on retention at 24 hrs;a mid-range dose group displayed significantly stronger spatial memory than either the low or high dose groups. In addition, while vehicle treated rats displayed a modest, numerical preference for the target quadrant of the maze, only the middle dose group scored reliably above chance. The findings are consistent with the proposal that chromatin modification mediated by histone acetylation can regulate hippocampal memory, and that targeted HDAC inhibition might benefit age-related cognitive decline and other conditions in which memory is compromised. These findings will be presented in abstract form at the 2010 Society for Neuroscience convention. Although we have not been in a position to test the efficacy of treatment for age-related cognitive impairment in our established model due to animal resource limitations (described above), we are exploring alternative approaches. Key evidence that hippocampal memory depends on dynamic epigenetic modification includes the observation that contextual fear conditioning induces robust histone acetylation in the hippocampus. Ongoing work in NAS aimed to extend this work to another exemplar hippocampal-dependent learning and memory paradigm, the Morris water maze. Main goals were to document the temporal dynamics of hippocampal histone modification induced by water maze training, and to map the neuronal distribution of these modifications in the hippocampus. Male Long-Evans rats learned procedural aspects of the Morris water maze task first and were later re-trained on a new task variant. Animals were sacrificed at various post-training intervals (1h, 4h, 6h, and 24h, n = 9/group), and histone modifications were analyzed via quantitative immunoblotting in microdissected hippocampal subfields, and by immunohistochemistry in histological sections. Home cage and non-spatial cue training groups served as controls. Overall, behavioral training in the water maze elicited relatively subtle histone modifications in comparison with the changes reported following contextual fear conditioning. Immunoblotting revealed that these changes are temporally dynamic and return towards baseline in all hippocampal subfields by 24h. Stereological estimates revealed that behavioral training potently increased the number of immunopositive H3 serine 10 phosphorylated neurons in the upper blade of the dentate gyrus, involving a small subpopulation of granule cells. Together the findings indicate that water maze training - a prototypical procedure for examining hippocampal learning and memory - engages a subtle pattern of temporally, regionally, cell-type, and site-specific histone modifications. These results are schedule for presentation at the 2010 Society for Neuroscience meeting in November and will be prepared for full-length publication in the next reporting period.
Showing the most recent 10 out of 20 publications
