CIC-2Contributes To Tonic Inhibition Mediated By α5Subunit-containing GABA_A Receptor In Experimental Temporal Lobe Epilepsy

Temporal lobe epilepsy, characterized by spontaneous recurrent seizures, learning and memory impairments is associated with neurodegeneration, abnormal reorganization of the circuitry, and loss of functional inhibition in hippocampus. In adult hippocampus, the GABAergic cells mediate the major inhibitory function of the principal neurons, promoting the Cl-entry through the GABAA receptor, whether through phasic (synaptic) or tonic (extrasynaptic) conductance. Aside from classical synaptic component, tonic GABAergic inhibition mediated by extrasynaptic GABAA receptor received increasing attention over the past years. There is growing evidence that tonic inhibition plays an important role in epilepsy, memory and cognition.Since GABAA receptor-mediated inhibition depends on the maintenance of intracellular Cl" concentration at low levels in mature neurons, a shift in Ecl is likely to participate in the generation and not merely a consequence of TLE. As we known, chloride homeostasis is regulated by cation-chloride cotransporters and chloride channels. The transmembrane distribution of chloride determines the direction of the chloride flux gated by GABAA receptor-mediated response in neurons. ClC-2is a member of the supergene family of voltage-gated chloride channels. It is proved to be inwardly rectifying, and plays an important role in setting the intracellular chloride concentration in neurons expressing inhibitory GABAA receptors (GABAARs).Several observations have shown that repetitive activation of GABAARs leads to a high intracellular chloride load that is generated by a GABAAR-mediated Cl-influx driven by HCO3-efflux via GABAARs, which dose not fade because of rapid replenishment of HCO3-by intracellular carbonic anhydrase (CA) activity. Furthermore, as a main chloride extruder in mature neurons, KCC2has been reported to be downregulated in experimental TLE models. Therefore, the chloride extrusion capacity seems to play a key role in setting the susceptibility of neurons to epileptiform activity. Since the conductance of ClC-2is large and does not display time-dependent inactivation, it is well suited for stabilize ECl. However, whether ClC-2contributes to epilepsy or not is controversial. The studies in human established the link between mutations in ClC-2and epilepsy, while ClC-2KO mice did not lead to higher seizure susceptibility. In addition, Rinke and coworkers used KO mice to demonstrate that ClC-2helps to extrude chloride from neurons under condition of high chloride load and contributes substantially to the background conductance. However, the amplitude and frequency of mIPSC, sIPSC, and PPR (a measure for presynaptic release) show no difference between WT and KO mice. Therefore, the possibility that ClC-2is involved in phasic inhibition by changing the number of postsynaptic GABAARs, inhibitory synapse number, and probability of release could be excluded.It posed the question whether ClC-2is related to tonic inhibition mediated by extrasynaptic GABAARs. As is known, a5subunit-containing GABAARs have a restricted distribution in dendritic areas of hippocampal CA1and CA3regions, and play a predominant role in tonic inhibition of CA1pyramidal cells. Here we investigated whether the ClC-2expression is altered in CA1pyramidal cells in pilocarpine-treated rats, and whether a5subunit-containing GABAARs could affect ClC-2currents by pharmacological intervention.Part Ⅰ Behavioural features and the pattern of neuronal loss in the hippocampus of pilocarpine-treated ratsObjective:To observe the behavioural features of pilocarpine-induced rats and neuronal loss in the CA1pyramidal cell layer in pilocarpine-treated rats.Methods:Male Sprague-Dawley rats weighing180-200g were used. Status epilepticus (SE) was induced by pilocarpine hydrochloride (340mg/kg, i.p.). To lessen peripheral cholinergic effects, atropine methylbromide (5mg/kg, i.p.) was administered30min before pilocarpine. If seizure activity was not initiated within1hour after the initial pilocarpine hydrochloride dose, an additional dose of170mg/kg was given. The onset of SE was defined as the appearance of stage5seizures, followed by continuous and convulsive behaviorally detectable seizure activity. Diazepam (10mg/kg, i.p.) was injected90min after its onset. Rats were hand fed after SE until they were able to drink and eat on their own (2～3d). After SE, rats were killed at different time-point, representative of the different phases of the natural history of the disease:24h immediately followed by the SE (acute phase),5d after SE (latency),14d after the first occurrence of spontaneous recurrent seizure (SRS, chronic epilepsy).30d after the first occurrence of SRS (late chronic epilepsy). Rats were continuously video monitored for the appearance of SRSs, only rats that displayed multiple spontaneous stage3-4seizures in the subsequent month were included into the14d and30d groups. Control rats (con) were treated identically except that saline was substituted for pilocarpine.Immunofluorescence was adopted to confirm the pattern of neuronal loss in this model. Primary antibody was mouse anti-neuronal nuclei (NeuN, neuron marker,1:500, Millipore). All the sections were treated by a mixture of FITC-conjugated secondary antibodies (1:400, Jackson ImmunoResearch). The quantification of the immunofluorescence staining in CA1area was performed by counting the number of NeuN positive cells per section. In each rat, every fourth section was picked from a series of consecutive hippocampus sections (20μm), and five sections were counted for each rat. An average number of neurons were obtained for each rat across five sections, and then the Mean±SEM across rats was determined. Statistical analysis was performed with SPSS10.0(SPSS Inc, USA). All data was presented as Mean±SEM. Differences in changes of values over times or drugs were tested using one-way ANOVA followed by individual post hoc comparisons (Tukey’s post hoc tests).Results:Convulsive status epilepticus (SE) were observed24±3min (Mean±SEM, n=51) after pilocarpine injection, which was intervened after90min by a low dose of diazepam. It did not stop the SE, but decreased its severity and mortality (7SE rats were dead). SE spontaneously alleviated5～7h after diazepam administration, and then rats entered postictal status, lasting2～3days, with self-limiting and generalized seizures (duration:10±3min; mean frequency:5±2seizures per day, n=38), before undergoing a latent period in which they were apparently normal. The first spontaneous recurrent seizure (SRS) occurred at12±1days after SE (n=32). Rats that did not display any SRS in one month after SE were deleted (n=4), while the others kept experiencing SRSs (duration:4±2min; mean frequency:4±1seizures per day, n=28). According to Racine’s classification criteria, partial seizures (stage3-4) were more frequent compared to secondarily generalized seizures (stage5-6), and tended to recur in a cluster way.Pilocarpine-treated rats displayed marked neuronal loss in the hilus of DG and CA3pyramidal cell layer. However, CA1pyramidal cell layer was well preserved. Compared with control group, the number of NeuN+cells in the CA1pyramidal cell layers in pilocarpine-treated rats did not decreased significantly at different phases. Summary:1. The first spontaneous recurrent seizure (SRS) occurred at12±1days after SE.2. Rats kept experiencing SRSs with a mean frequency of4±1seizures per day and a mean duration of4±2min in the subsequent month.3. No significant neuronal loss in the CA1pyramidal cell layer in pilocarpine-treated rats.Part Ⅱ The location of CIC-2IR and the quantification of ClC-2protein level in the CA1pyramidal cells layer in pilocarpine-treated ratObjective:To investigate the location of ClC-2IR in hippocampus and the change of protein level of ClC-2in CA1region in pilocarpine-induced rats.Methods:Double immunofluorescence staining was performed to localize ClC-2in hippocampus. Rats from epileptic and control groups were anesthetized with urethane (1.5g/Kg, i.p.) and transcardially perfused with heparinized saline, followed by4%paraformaldehyde (PFA) in0.1M phosphate buffer, PH7.4. The intact brains were removed and post-fixed for3h in the same fixative, then transferred into30%sucrose solution. Transverse brain sections (20μm) were cut using a cryostat. All sections were blocked with3%donkey serum in0.3%Triton X-100for1h at room temperature and incubated with primary antibodies over one night at4℃, then the sections were incubated for2h at room temperature with secondary antibodies. The stained sections were examined with a fluorescence microscope and images were captured with a CCD spot camera.Western blot was performed to confirm the expression of ClC-2after SE in protein level. Rats were decapitated, brains were immediately removed and frozen on dry ice, then stored at-80℃until processing. The CA1regions were microdissected on dry ice. The tissue samples were homogenated and sonicated in15mmol/L Tris buffer, then centrifuged at13,000g for15min at4℃to isolate the supernatant containing protein samples. The samples were separated by gel electrophoresis (SDS-PAGE) and transferred onto a PVDF membrane. The membrane was blocked for1h at room temperature in blocking buffer and then incubated with anti-ClC-2antibody (1:200, Alomone Labs) overnight at4℃, followed by incubation in anti-rabbit secondary antibody (1:1000, Cell Signaling Technology). The immune bands were visualized by enhanced chemiluminescence (ECL, Amersham, USA) and film exposure (Kodak, Rochester, NY). The membrane was stripped with stripping buffer for30min at50℃and re-probed with anti-α-actin (1:1000, Santa Cruz), followed by incubation in anti-mouse secondary antibody at a dilution of1:4000(Cell Signaling Technology) as an endogenous control protein to ensure equal loading. Densitometric analysis of bands on the film was conducted with a computer-assisted imaging analysis system, and normalized to β-actin immunoreactivity.Results:In control rats, ClC-2immunoreactivity (ClC-2IR) was present with high levels of labeling in the cell membrane and perinuclear cytoplasm of pyramidal cells (PCs), and relatively low levels of labeling in the proximal apical dendrites of PCs in CA1regions. Double immunofluorescence staining showed that ClC-2only co-localized with NeuN, neither with GFAP nor with Ibal. In pilocarpine-treated rats, an increased ClC-2IR was present, especially in CA1PCs dendrites (strata radiatum). Meanwhile, ClC-2IR maintained high levels in the pyramidal cell layer.Immunoblot was performed to quantify this change, ClC-2protein level was significantly higher than that in control group.Summary:1. ClC-2IR was present with high levels of labeling in the cell membrane and perinuclear cytoplasm of pyramidal cells (PCs), and relatively low levels of labeling in the proximal apical dendrites of PCs in CA1region.2. ClC-2upregulated in CA1pyramidal cells (PCs) in the chronic post-SE rats.3. The obvious enhancement of ClC-2IR during the chronic period is in the apical dendrites of PCs in CA1region.Part Ⅲ CIC-2alteration is involved in tonic inhibition mediated by a5subunit-containing GABAARs in CA1regionObjective:To investigate whether the upregulation of ClC-2in CA1PCs is functional and whether ClC-2is involved in tonic inhibition mediated by a5subunit-containing GABAARs. Reducing the tonic inhibition would aggravate or improve the generation and propagation of SRS in post-SE rats.Methods:Whole-cell voltage-clamp recordings in hippocampal slices prepared from epileptic and control rats. Following decapitation, brains were quickly submerged in ice-cold sucrose-ACSF. Transverse slices (400μm) were cut with a Vibratome. Before recording, slices were incubated for30min at32℃in ACSF. Then slices were transferred to the recording chamber and continuously perfused (5ml/min) with recirculating extracellular solution (oxygenated with95%O2-5%CO2). Visualized patch-clamp recordings from the CA1pyramidal neurons were performed by IR-DIC videomicroscopy with an Axopatch200B amplifier, filtered at2kHz and digitized at20kHz. C1C-2currents were evoked at test potentials between+40and-20mV (increment:-20mV; holding potential:-10mV).Itonic were recorded as the change in mean holding currents (ΔIhold) after applying SR-95531(100μM) while voltage clamped at-60mV. The Ihold was sampled and averaged in recordings as follows:100s before (t1) and200s after L-655,708application (t2), and100s after SR-95531application (t3). Mean ΔIhold before and after drugs application were calculated for each neuron. Before and after values for an individual neuron were compared using the Kolmogorov-Smirnov (KS) test. Differences between groups were analyzed with one-way ANOVA followed by individual post hoc comparisons (Tukey’s post hoc tests).A stainless-steel cannula (22gauge,0.7mm outer diameter) was implanted above the right lateral ventricle (0.5mm posterior to the bregma,1.2mm lateral to midline and4mm below the surface of the dura). The cannula was secured by dental cement. A stainless-steel obturator was inserted into the cannula to prevent clogging and infection. Rats were allowed to recover for7days before inducing SE. Intracerebroventricular (i.c.v.) injection was made from the latency till14d after the first occurrence of SRS via an internal cannula (28gauge,0.36mm outer diameter).Results:C1C-2currents increased by53.2%(control,453.1±47.4pA, n=9; post-SE,694.5±56.2pA, n=12; P<0.05) in polocarpine-treated rats, consistent with the upregulation of C1C-2. And a decrease in L-655,708treatment slices by-26.9%was observed (vehicle,702.3±63.1pA, n=10; L-655,708,513.6±50.2pA, n=15; P<0.05). A significant increase in GABAAR-mediated tonic inhibition (Itonic) in CA1PCs, compared to neurons from control (control:132.1±17.9pA, n=12; post-SE:207.4±13.6pA, n=15; P<0.05). L-655,708(100nM) decreased Itonic in control neurons by-21.6±9.2%(n=7), but produced a significantly greater decrease in post-SE neurons (-30.3±7.9%, n=11).L-655,708treatment starting from the latency till14d after the first occurrence of SRS in vivo resulted in an earlier manifest of the first occurrence of SRS at9±1days after SE in the8μM and16μM L-655,708groups (n=7each group, P<0.05), but did not show significant effect on the duration and mean frequency of SRSs. Summary:1. C1C-2currents increased in CA1PCs in pilocarpine-treated rats.2. A significant increase in GABAAR-mediated tonic inhibition (Itonic) in CA1PCs, compared to neurons from control. It is probably due to the compensatory increased tonic inhibition for replenishing the decreased synaptic GABAergic innervation at dendrites in TLE model.3. L-655,708attenuated the increased Itonic in vitro and produces an earlier manifest of the first occurrence of SRS in vivo in pilocarpine-treated rats. We suggest that the chronically raised tonic inhibition mediated by a5subunit-containing GABAARs is associated with the generation of SRS.4. L-655,708reversed the increase in C1C-2currents in vitro. We suggest that C1C-2modification in CA1PCs is probably to assist the chloride extrusion and maintain the inwardly directed driving force for chloride ions, which is a prerequisite for hyperpolarizing inhibition of extrasynaptic GABAARs.Part Ⅳ L-655,708attenuates the increased LTD, but produces no significant change in decreased LTP in post-SE rats in vivo.Objective:To observe what the change of synaptic plasticity of post-SE rats in CA1area, and whether a5subunit-containing GABAARs could affect amplitude of field potentials by pharmacological intervention.Methods:Experiments were performed on male Sprague-Dawley rats. Urethane (1.5g/kg, ip) was used to induce and maintain anesthesia. Additional doses (0.5g/kg) were given if needed. Surgical level of anesthesia was verified by the stable mean arterial blood pressure during noxious stimulation. The trachea was cannulated, and the animal breathed spontaneously. The left femoral vein and artery were cannulated for intravenous injection and continuous monitoring of blood pressure respectively. Colorectal temperature was kept constant (37-38℃) by means of a feedback-controlled heating blanket. Rats were placed in a stereotaxic frame for all recordings. A small craniotomy (1mm×1.5mm) was performed, and then the dura and arachnoid were carefully removed. A stainless steel guide cannula was implanted in the right lateral ventricle (1mm lateral to midline,0.5mm posterior to bregma and4mm below the surface of dura), an internal cannula was used for intracerebroventricular injections (i.c.v.). The craniotomy hole was sealed with dental cement after implantation procedure. Field excitatory postsynaptic potentials (fEPSPs) were recorded in the stratum radiatum of the dorsal hippocampus in response to stimulation of the ipsilateral Scaffer collateral-commissural pathway. The recording site was located3.4mm posterior to bregma and2.5mm lateral to midline, and the stimulating site was located4.2mm posterior to bregma and3.8mm lateral to midline. The final depth of both electrodes was adjusted to optimize the electrically evoked fEPSPs. The intensity of test stimuli (0.033Hz,0.1ms duration) was adjusted to evoke a fEPSP that was50%of maximum. Conditioning high frequency stimulation (HFS:100Hz,50stimuli per train,4trains at interval of15s) was induced LTP, and LTD was induced by low frequency stimulation (LFS:1Hz,900stimuli15min). Baseline was recorded for>30min prior to injection of drug/vehicle to ensure a steady response. All values were expressed as mean±SEM. The magnitude of LTP/LTD was measured as the percentage of the baseline fEPSP amplitude during30min periods, just prior to HFS/LFS. Before and after values for an individual rat were compared using the Wilcoxon signed ranks test. Differences between groups were analyzed with Kruskal-Wallis test.Western blot was performed to compare the protein levels of NR1, NR2A, NR2B, GluRl, GluR2and GluR3in CA1area in post-SE rats with that in control ones. Rats were decapitated, brains were immediately removed and frozen on dry ice, then stored at-80℃until processing. The CA1regions were microdissected on dry ice. The tissue samples were homogenated and sonicated in Tris buffer, then centrifuged to isolate the supernatant containing protein samples. The samples were separated by gel electrophoresis (SDS-PAGE) and transferred onto a PVDF membrane. The membrane was blocked for1h at room temperature in blocking buffer and then incubated with primary antibody overnight at4℃, followed by incubation in secondary antibody. The immune bands were visualized by enhanced chemiluminescence (ECL, Amersham, USA) and film exposure (Kodak, Rochester, NY). The membrane was stripped with stripping buffer for30min at50℃and re-probed with anti-β-actin, followed by incubation in secondary antibody as an endogenous control protein to ensure equal loading. Densitometric analysis of bands on the film was conducted with a computer-assisted imaging analysis system, and normalized to P-actin immunoreactivity.Results:The amplitude of fEPSPs induced by HFS decreased by14.84%(control:150.9±7.57%, n=5; post-SE:128.5±9.43%, n=5; P<0.05) in post-SE rats, and an increase in the amplitude of fEPSPs induced by LFS was detected (control:68.3±7.31%, n=5; post-SE:117.2±6.22%, n=5; P<0.05) compared to rats from control. L-655,708attenuates the increased LTD (vehicle:113.7±6.91%, n=4; L-655,708:91.6±6.57%, n=5; P<0.05), but produces no significant change in decreased LTP in post-SE rats in vivo (vehicle:130.3±7.36%, n=4; L-655,708:132.4±7.01%,n=5;P>0.05). Immunoblot was performed to quantify the protein levels of NR1, NR2A, NR2B, GluR1, GluR2and GluR3in CA1area. NR2B and GluR2downregulated in CA1pyramidal cells (PCs) in the chronic post-SE rats, and other ones were higher than that in control group.Summary:1. The amplitude of fEPSPs induced by HFS decreased significantly in post-SE rats, and an increase in the amplitude of fEPSPs induced by LFS was detected compared to rats from control.2. L-655,708attenuates the increased LTD, but produces no significant change in decreased LTP in post-SE rats in vivo.ConclusionThe major findings of this study are that ClC-2upregulated functionally in CA1pyramidal cells (PCs) in the chronic post-SE rats, with a corresponding increase in tonic GABAergic inhibition, and that dampening this tonic inhibition by L-655,708reverses the increase in ClC-2currents in CA1PCs in vitro and in amplitude of fEPSPs induced by LFS in vivo.