The Role Of DNA Methylation In The Process Of Refrectory Epileptogenesis And Its Possible Mechanisms

PART ONE: STUDIES ON THE EXPRESSION ANDPOSSIBLE MECHANISMS OF DNAMETHYTRANSFERASE1AND3A IN THE BRAINTISSUE OF LITHIUM-PILOCARPRINE-KINDLED RATSObjective: DNA methylation is a key epigenetic modification of DNAthat is catalyzed by DNA methyltransferase (DNMT). Increasing evidencessuggest that DNA methylation in the central nervous system (CNS)regulates synaptic plasticity as well as neuronal network activity. Ourprevious study demonstrated that DNMT1and DNMT3a were significantlyup-regulated in the brains of the patients with intractable temporal lobeepilepsy (TLE). It is impossible to learn the expression pattern andspatiotemporal change of DNMT1and DNMT3a by Human TLE tissuesamples and to obtain completely normal brain tissue as control, due to thelimitations of Ethics and Morals. The rat model oflithium-pilocarprine-induced seizures can mimic the human TLE andpeople can understand the process of epileptogenesis with this model.Therefore, we evaluated dynamic expression of DNMT1and DNMT3a in the brain tissues of kindled rats (lithium chloride-pilocarprine model).Meanwhile, we observed the behavior and electroencephalogram (EEG)changes of kindled rats after the interference of5-Aza-2'-deoxycytidine(5-AZA), inhibitors of DNMT activity, to explore the mechanism ofDNMTs in the onset of epileptogenesis.Methods: There are two parts in this experiment. To detect theexpression of DNMT1and DNMT3a, seventy-two Sprague-Dawley ratswere randomly divided into control group and kindled group in theproportion of1to8. The kindled group was injected intraperitoneally withpilocarpine hydrochloride and the control group was injected with normalsaline. The kindled group was again divided into seven subgroups,including6h,1day,3days,7days,14days,30days and60days post-seizure,each group contained eight rats(Eight kindling failure and dead rats wererejected). Subsequently, rats were sacrificed at different times and thehippocampus of the kindled rats and control rats were removed. Thetemporal spatial localization and expression of DNMT1and DNMT3a weredetected through immunohistochemistry, immunofluorescence and Westernblotting analysis. To evaluate the impact of5-AZA in the onset of theepilepsy, twenty-four same SD rats were randomly divided into threegroups:pilocarpine, DMSO+pilocarpine and5-AZA+DMSO+pilocarpinegroup, each group contained eight rats. Following injection of DMSO or5-AZA+DMSO into lateral ventricle of DMSO+pilocarpine or5-AZA+ DMSO+pilocarpine group rats, three groups were induced by pilocarpinerespectively, and then the latency and dosage of pilocarpine required forinducing SE as well as seizure duration and electroencephalogram (EEG)changes were recorded in kindled rats after initial insult.Results: In the brain tissue of the control group, both DNMT1andDNMT3a-stained cells were principally expressed in the nucleus ofneurons in the dentate gyrus (DG) and CA3-CA1. In the kindled group,however, they were not only expressed in neuronal nuclei but in partialcytoplasm of neurons as well. Meanwhile, more DNMT1-positive neuronsand DNMT3a-positive neurons were evident in kindled group whencompared to the control group, and DNMTs-positive cells were stronglystained after seizure. Compared to the control group, DNMT1expressionhad no obvious change at1h post-ssizure, but its expression began to beremarkably up-regulated from1day to14days post-seizure, and it was thenmaintained at a relatively high level until30days and60days post-seizure,DNMT3a expression was increased at6h post-seizure, reached a stable highlevel at1day and3days, and then slightly dropped but still maintained at arelatively high level during the latent and chronic period. In interventionexperiments, we observed that sixteen rats were successfully kindled inpilocarpine group and DMSO+pilocarpine group as well as five rats wereinduced in5-AZA+DMSO+pilocarpine group. Compared to pilocarpinegroup and DMSO+pilocarpine group,5-AZA+DMSO+pilocarpine group rats displayed a higher seizure threshold as well as seizure duration weredecreased and the latency-induced seizure was delayed. Moreover, theepileptic discharges of in5-AZA+DMSO+pilocarpine group wereweakened when compared to those in the saline+Pilocarpine group.Conclusion:The changes of expression location and expression levelsof DNMT1and DNMT3a in the hippocampus of kindled rats wereconsistent with the expression pattern in the brain tissue of patients withTLE.5-AZA decreased seizure susceptibility in response to convulasntdrugs in rats and5-AZA may inhibit seizure in the onset of epileptogenesisby inhibiting the function of DNMTs, which supported a role forup-regulation of DNMTs in the pathological mechanism of epilepsy. PART TWO: DOWN-REGULATION OF RASGRF1EXPRESSION IN PATIENTS WITH REFRACTORYEPILEPSY AND KINDLED ANIMALSObjective: Ras-guanine nucleotide-releasing factor1(RasGRF1), aRas-specific activator, is selectively expressed in neurons and enriched atsynaptic sites. The gene encoding RasGRF1is an imprinted gene that isregulated through epigenetic methylation mechanism, and its expressionlevel after birth increases in parallel with the development of a neuronal circuitry. Evidences suggested that RasGRF1can mediate forms ofsynaptic plasticity and may participate in the regulation of neuronalexcitability and neurite outgrowth though various signal transductionpathway. In a previous study, we analysised the cerebrospinal fluid (CSF)proteome of patients suffering from TLE and the control usingpolyacrylamide gel electrophoresis and found that Ras-GRF1weredecreased in the CSF of TLE patients compared with the CSF of controls.To further learn the relationship between RasGRF1and epilepsy, wemeasured the concentrations of RasGRF1in CSF with larger sample size.Meanwhile we investigated RasGRF1expression in brain tissue of patientswith drug-refractory temporal lobe epilepsy (TLE) and lithiumchloride-pilocarprine kindled rats to discuss its possible role in TLE.Methods:1The detection of CSF-RasGRF1in epilepsy patients:40intractable epilepsy patients were recruited from the Epilepsy Clinic ofFirst Affiliated Hospital of Chongqing Medical University. All CSF ofpatients were collected by lumbar puncture, RasGRF1concentration wasdetected using sandwich enzyme-linked immunosorbent assay (ELISA),compared with30subjects that had no evidence of any neurologicaldiseases. The expression of RasGRF1in the brain tissue of epilepsypatients and kindled rats: We randomly selected30temporal neocortexfrom patients with TLE in our epilepsy brain bank as the experiment groupand9temporal neocortex from head trauma as the control group. The expression of RasGRF1protein was examined by double-labelimmunofluorescence, immunohistochemistry and Western blotting.Meanwhile,64SD rats were used for kindled model and were randomlydivided into the control group and kindled group in the proportion of1to7.Following the onset of SE, kindled rats were randomized into six groups:1day,7days,14days,30days and60days post-seizure,each groupcontained8rats (The kindling failure and dead rats were rejected). Theepileptic group were injected with pilocarpine hydrochloride to induce thesustained seizures, and control rats were treated identically saline. Therats were sacrificed at different time and the hippocampus and adjacentcortex of the kindled rats and control rats were removed. Double-labelimmunofluorescence, immunohistochemistry and Western blotting wereused to detect the distribution and level of RasGRF1protein expressionduring the different phases(the acute period, the latent period and the latentperiod)in the epileptic and control rats.Results: The mean concentrations of CSF-RasGRF1in epilepticpatients and control group were respectively0.965±0.057mg/L and1.408±0.092mg/L (p<0.05). Immunostainning demonstrated that RasGRF1was mainly expressed in the plasma membrane and axon of neurons,co-localizing with MAP2, but was not co-expressed with GFAP-positiveastrocytes in brain tissue of patients with TLE and kindled rats. RasGRF1immunoreactivity in the rats was observed mainly in the CA1to CA3 regions and GD as well as the cortex of hippocampus. In TLE patients andkindled rats, neurons exhibiting weakened Ras-GRF1signal were detectedand RasGRF1-positve neurons displayed the shorter axons of neurons.Compared to the controls, western blotting showed decreased level ofRasGRF1protein in the brain tissue from patients with TLE; westernblotting indicated that RasGRF1expression in kindled rats already reducedat1day post-seizures, then gradually decreased during the latent periodand reached a minimum level during the latent period.Conclusions: Level of RasGRF1expression was down-regulated inthe brain tissue of patients with TLE and kindled rats, and the axons ofRasGRF1-stained neurons became shorter. These results demonstrated thatRasGRF1, a synaptic protein, could be involved in the pathophysiology ofepilepsy by influencing cytoskeletal and signaling transductant pathways.Up-regulation of CSF-RasGRF1in the intractable epilepsy patients wasconsistent with the screening of CSF proteomics, and its variation was inaccordance with the expression of RasGRF1in the brain tissue of patientswith TLE and epileptic rats. This result suggested that RasGRF1might be abiomarker for drug-refractory epilepsy.