A ROLE FOR COLONY STIMULATING FACTOR 1 RECEPTOR SIGNALING AND MICROGLIOSIS DURING EPILEPTOGENESIS

Abstract

Evidence from experimental models of epilepsy support that prolonged seizures (status epilepticus, SE) promote pathological hippocampal synaptodendritic remodeling which contributes to the development of seizures and cognitive decline. One potential mechanism underlying the SE-induced sequelae is microgliosis. Evidence from models of experimental epilepsy supports a significant spatiotemporal correlation between SE-induced decreases in the microtubule associated protein 2 (Map2) loss and microgliosis in the hippocampus. In addition, pharmacological suppression of microgliosis after SE with the drug rapamycin attenuated the losses of Map2 and the dendritic ion channels Kv.4.2 and HCN1 in the hippocampus. This microglia suppression paralleled a recovery of the SE-induced recognition and spatial memory deficits. Based on these studies, we hypothesized that the inhibition of microgliosis during epileptogenesis will attenuate the SE-induced hippocampal dendritic and cognitive pathology. To further investigate the role of microgliosis in the SE-induced dendritic pathology, we tested the efficacy of a more selective inhibitor of the survival and proliferation of microglia, PLX3397, using the pilocarpine model of SE and acquired epilepsy. PLX3397 binds to colony stimulating factor 1 receptor (CSF1R) on microglia and inhibits the downstream signaling responsible for survival and proliferation of these cells. To test this hypothesis, we induced SE in male rats with pilocarpine (280-300mg/kg) while and controls (Ctrl) received saline. Rats were randomly assigned to a diet of either chow alone (vehicle; Veh) or chow with PLX3397 (50mg/kg) for 20 days post-SE. At two weeks post-SE, rats were subjected to novel object recognition (NOR) and Barnes maze (BM) to evaluate hippocampal-dependent recognition memory, and spatial learning and memory, respectively. Following the behavioral assessments, rats were sacrificed for brain analysis at 20 days post-SE. We used histological analysis to determine the amount of microgliosis with IBA1 and dendritic stability with Map2. We used western blotting to measure the protein levels of molecules involved in the crosstalk between microglia and astrocytes: GFAP, IL-6, C3, and iC3b. We also measured the protein levels of the dendritic ion channels Kv4.2 and HCN1, and the synaptic marker PSD95. NOR showed that the Ctrl+Veh and Ctrl+PLX3397 groups spent significantly more time exploring the novel object (p < .05), while the SE+Veh and SE+PLX3397 did not. Similar results were observed in the BM test, Ctrl+Veh and Ctrl+PLX3397 groups had a faster latency to find the target compared to the SE+Veh and SE+PLX3397 groups (p < .05). These data suggest that recognition and spatial memory deficits induced by SE were not attenuated by treatment with PLX3397. We found that the PLX3397 treatment significantly decreased microgliosis in Ctrl+PLX3397 rats compared to Ctrl+Veh rats (p < .05). As expected, we found a significant increase in the number of microglial cells in hippocampi of SE+Veh rats compared to Ctrl+Veh rats (p < .05). Interestingly, in the PLX3397-treated SE group, we observed two distinctive groups which we categorized as responders and non-responders when compared to the SE+Veh group. The SE+PLX responders had significantly decreased microgliosis compared to the SE+Veh group (p < .05). The SE+PLX non-responders had higher levels of microgliosis compared to the SE+Veh group (p < .05). We found levels of GFAP were increased in the SE+Veh group compared to the Ctrl+Veh group (p < .05). Treatment with PLX3397 in the SE group reduced these levels compared to the vehicle treated SE group (p < .05). We also found increases in C3 and iC3b following the induction of SE compared to Ctrl+Veh group (p < .05), and these levels remained similar in the SE+PLX3397 group compare to the SE+Veh group (p > .05). There was a reduction in Map2 immunoreactivity as well as the protein levels of Kv4.2 and PSD95 in the SE+Veh group compared to the Ctrl+Veh group (p < .05). We found that treatment with PLX3397 recovered the SE-induced loss of Map2 labeled dendrites compared to SE+Veh group (p < .05). However, treatment with PLX3397 did not recover the SE-induced reduction Kv4.2 and PSD95 (p > .05). In parallel, we found that a group of SE+PLX3397 animals did not have reduced microgliosis compared to the SE+Veh group (p < .05), and therefore was categorized as a non-responder group. Our findings are the first to show that blocking CSF1R signaling with PLX3397 suppressed microgliosis in the hippocampus, partially recovered the SE-induced decline of Map2 immunoreactivity in the hippocampal CA1 region but had no effect in the recognition or spatial memory deficits. These data suggest that while hippocampal microgliosis may play a role in the disruption of dendritic structural stability in the hippocampus it does not seem to critically contribute to the memory decline that occurs during epileptogenesis.