THE ROLE OF COMPLEMENT C3 IN THE HIPPOCAMPAL PATHOLOGY OF STATUS EPILEPTICUS
Epilepsy is comorbid with cognitive and psychiatric dysfunctions. This pathophysiology, associated with hippocampal synaptodendritic structural and functional changes, is exacerbated by prolonged seizures (status epilepticus; SE). We found a correlation between hippocampal dendritic loss and microgliosis after SE, along with hyperactivation of the classical complement pathway (C1q-C3). These paralleled increased seizure frequency and memory deficits in a rat model of SE and acquired epilepsy. C1q leads to C3 cleavage into biologically active fragments C3a and C3b. Evidence suggests that C1q and C3b contribute to synaptic stripping by microglia in the developing brain and neurodegenerative disorders. Thus, we hypothesized that SE-induced C3 activation may alter hippocampal synaptic protein levels thereby promoting memory deficits.
To test the hypothesis, different groups of wild type (WT) or C3 deficient (C3KO) mice were injected with pilocarpine (350mg/kg) to induce SE or saline (controls): WT-C, WT-SE, C3KO-C, and C3KOSE. At two weeks after SE, mice were subjected to novel object recognition (NOR) to evaluate recognition memory, and Barnes maze (BM) to measure hippocampal-dependent spatial learning and memory. Following behavioral testing, mice were sacrificed and hippocampi collected at either 2 or 5 weeks after SE to measure changes in C3 protein levels and levels of synaptic proteins including PSD95, Vglut1, and Vgat. As a method of verifying our findings, we used a second model of pilocarpine-induced SE in male Sprague Dawley rats. Starting at 7 days after SE, rats were treated with cobra venom factor (CVF; 100ng/g, i.p.) or vehicle (veh) every third day. On days 14-15 rats were subjected to open field and NOR to measure anxiety and recognition memory. On day 16, rats were sacrificed and hippocampi collected for western blotting.
WT and C3KO mice were able to reach stage 4.5-6 seizures after pilocarpine injections. In NOR trial 1, exploration time for both objects was similar in all groups (p > .05). In trial 2, WT-C and C3KO-C mice spent more time exploring the novel object than the familiar one (p < .05) while WT-SE mice explored both objects equally (p > .05). Interestingly, C3KO-SE mice spent more time with the novel object similar to controls (p > .05), suggesting that the deficit in object recognition memory induced by SE was attenuated in C3KO mice. Similarly, veh- and CVF-treated control rats spent more time exploring the novel object during trial 2 (p < .05). The veh-treated SE rats did not show significant preference for the novel object versus familiar (p > .05), whereas the CVF-treated SE rats explored the novel object significantly more than the familiar (p < .05). These findings support that C3 inhibition after SE prevents recognition memory deficits. Furthermore, there was a reduction in synaptic proteins PSD95 and Vgat in the SE-veh group compared to the C-veh group. This difference was not observed in the C-CVF and SE-CVF groups, suggesting that blocking C3 after SE is neuroprotective against hippocampal synaptic loss.
Taken together, these findings are the first to show an association between C3 activation and hippocampal and cognitive deficits in two rodent models of SE and acquired TLE. We found that depletion of C3 is sufficient to attenuate SE-induced deficits in NOR-evaluated recognition memory and changes in the levels of an inhibitory synaptic protein. In conclusion, our data suggest that SE-induced complement C3 activation contributes to hippocampal synaptic remodeling and impairments in recognition memory, and that the complement C3 may be a potential therapeutic target for the memory comorbidities associated with SE. Future studies will determine the effect of C3 inhibition on spontaneous recurrent seizures, and whether C3-guided and microglial-dependent phagocytosis is an underlying mechanism for the SE-induced epileptogenic synaptic remodeling.