Facilitating Factors And Mechanisms Of Post-traumatic Epileptogenesis

Objective:The exact molecular and cellular mechanisms of post-traumatic epileptogenesis are still unknown.The challenge is if we can identify treatable factors that perhaps currently being overlooked,which facilitate post-traumatic epileptogenesis.Peripheral infection has been demonstrated to worsen the consequences of a variety of central nervous system injuries.However,the effect of peripheral infection on post-traumatic epileptogenesis remains unclear.We hypothesized that peripheral infection at a chronic time point after TBI can act as 2^nd hit to the brain previously primed by TBI,increasing neuronal excitability in the perilesional cortex and hippocampus and reactivate astrocytes,facilitating post-traumatic epileptogenesis and other comorbidities.Astrocytes play a critical role in neuroinflammation and also an important part of peripheral infection to promote post-traumatic epilepsy,but the specific mechanism is unknown.The second part of this study investigated the effects of sigma-1 receptor（Sig-1R）activation on lipopolysaccharide（LPS）-induced inflammatory reactions and oxidative/nitrosative stress in cultured astrocytes.On another side,it is demonstrated that post-traumatic epileptogenesis may be associated with loss of inhibitory interneurons and the impairment of the inhibitory neuronal network.However the alterations of a specific interneuronal subclass,calretinin（CR）positive neurons and CR-related gene networks in the brain after TBI,and their relationship with post-traumatic epileptogenesis remain unknown.Our another objective was to investigate the changes of CR network in the cerebral cortex and thalamus at a chronic time point after TBI,which may contribute to elucidation of the mechanisms of post-traumatic epileptogenesis.Methods:The first part,adult rats were subjected to lateral fluid-percussion injury（FPI）.Lesion endophenotype[focal inflammatory(TBI_FI)vs.cavity-forming(TBI_CF)]was assessed with T2-weighted magnetic resonance imaging（MRI）at 6 wk post-TBI.Then,rats were randomized into 4 groups:TBI_FI-Veh,TBI_FI-LPS,TBI_CF-Veh,TBI_CF-LPS.At 8wk post-TBI,rats received a single injection of lipopolysaccharide（LPS;5mg/kg,i.p.）or vehicle,rats in sham group only received vehicle injection.At 12 wk post-TBI and 4 wk post-LPS,rats underwent open-field（OF）,elevated plus-maze（EPM）,sucrose preference,and forced swimming tests.At 16 wk post-TBI,a 4-wk long video-EEG（vEEG）monitoring was performed,followed by Morris water-maze（MWM）test.In the end,rats were subjected to pentylenetetrazol（PTZ）seizure-susceptibility test and sacrificed 2 hours after PTZ injection.The extent and progression of cortical lesion was assessed with unfolded maps prepared from MRI at 6 wk and Nissl-stained sections at 23 wk post-TBI.The spatial pattern of neuronal activation 2h after PTZ administration was assessed by analyzing c-Fos immunohistochemistry expression.The second part,the culture of astrocytes was performed,then the levels of NO,reactive oxygen species（ROS）and glutathione（GSH）were measured to determine whether activation of sig-1R has an inhibitory effect on LPS-induced inflammatory reaction in cultured astrocytes.Next,the expression of Nrf2,a key factor in cell defense against oxidative stress,was detected and correlated with the release of inflammatory mediators from astrocytes.In addition,the expression of heme oxygenase-1（HO-1）,a redox-sensitive enzyme with antioxidant,antiapoptotic and anti-inflammatory effects that can be activated by Nrf2 was also measured.Finally,the expression of nuclear factor-kappa B（NF-κB）was detected,which stimulates proinflammatory cytokine release and whose transcriptional activity can be regulated by Nrf2,to investigate the possible mechanisms underlying the antioxidative/nitrosative stress and antiinflammatory effects of Sig-1R activation on astrocytes.The third part,adult rats were injured with lateral FPI.Their brains were removed and processed for thionin staining and NeuN,CR immunohistochemistry at 1month post-TBI.For the cerebral cortex,we first generated 2 dimensional（2D）unfolded maps of the cortical lesion from thionin-stained preparations.Then quantitative analyses of CR-ir neurons in the ipsilateral cerebral cortex were transferred into the unfolded maps.We also investigated the alterations of expression of CR-related genes in the perilesional cortex at 3 months post-TBI using bioinformatics analysis,and in particular,further quantified two genes（Tubb3 and Reln）using RT-qPCR assays.For the thalamus,the number of CR-ir neurons in the reticular nucleus（RT）was quantified using unbiased stereology and the volume of this nucleus was estimated based on Cavalieri principle.The density of CR-ir terminals in the ventral anterior nucleus（VA）-ventral lateral nucleus（VL）and the area of this complex at the AP level-1.80 from bregma was assessed.We also studied the expression of neurotransmitter receptors and regulators in the thalamus of a separate group of rats at 6 months post-TBI using quantitative reverse-transcriptase polymerase chain reaction（RT-qPCR）,and assessed further the calretinin-related network alterations with bioinformatics analysis using existing RNA-seq data-set obtained at 3months post-TBI in our previous study.Results:The first part,T2-weighted MRI at 6 wk post-TBI indicated that 53%of the rats with TBI developed a TBI_FI endophenotype and 47%a TBI_CF endophenotype.In the OF,no difference was found between the TBI-Veh and sham groups.TBI-LPS rats,however,spent less time in the middle（p<0.01）and more in the outer zone(TBI_CF-LPS vs.sham,p<0.01;TBI_FI-LPS vs.sham,p<0.05)as compared to the sham group,regardless of the lesion endophenotype.Moreover,TBI-LPS rats spent more time in the outer zone than TBI-Veh animals（p<0.05）.In the EPM,TBI-Veh and TBI-LPS rats spent less time in open arms than sham-controls（p<0.05）.In the MWM,the latency to find the hidden platform was prolonged both in TBI-Veh and TBI-LPS rats,and they spent less time in the target quadrant than sham-controls（p<0.05）.v-EEG revealed that 2 of 7 rats in the TBI_FI-Veh group developed spontaneous seizures.No difference in the occurrence of seizures was found between the TBI-LPS and TBI-Veh groups,but the percent of animals with seizures in the PTZ test was higher in TBI_FI-LPS than in TBI_FI-Veh group（p<0.05）.Cumulative duration of PTZ-induced seizures was longer in TBI-LPS than TBI-Veh rats（p<0.01）,particularly in rats with a TBI_FI endophenotype（p<0.05）.The lesion area in unfolded histologic maps was 111%of that in unfolded MRI maps（p<0.05）in the TBI-LPS group,which related to lesion progression in the TBI_FI endophenotype（p<0.05）.At 2h following PTZ stimulation,TBI-Veh rats showed more c-Fos expression than sham-operated controls ipsilaterally[regardless of rostral（p<0.01）or caudal area（p<0.05）];TBI-LPS rats revealed more c-Fos expression in the contralateral cortex rostrally（p<0.01）and in the perilesional cortex caudally compared to TBI-Veh group（p<0.05）.The increase of c-Fos expression bilaterally in TBI-LPS group was more prominent in layer II-IV than layer V-VI,and more pronounced rostrally than caudally.TBI_FII endophenotype showed a trend of higher neuronal excitability in the perilesional and contralateral cortex compared to TBI_CFF endophenotype.The percentage of rats with neuronal activation in bilateral dentate gyrus was higher in TBI-LPS group than TBI-Veh group（p<0.05）,which resulted from TBI_FII endophenotype（p<0.05）.The second part,LPS significantly increased the expression levels of iNOS and TNF-αin astrocytes,while the selective Sig-1R agonist SA4503 reversed this effect of LPS mainly by downregulating the expression of iNOS and tumor necrosis factorα（TNF-α）and upregulating glutathione（GSH）in cultured astrocytes.To investigate the mechanism by which SA4503 caused these effects,the expression of nuclear factor erythroid-derived class 2（Nrf2）and heme oxygenase 1（HO-1）were then examined through western blotting.The results showed that SA4503 treatment significantly increased the expression of Nrf2 and HO-1,indicating the antioxidant/nitrosative stress and anti-inflammatory effects of Sig-1R activation in astrocytes were partially mediated by Nrf2 and HO-1 activation.The third part,for the cerebral cortex,at 1 month post-TBI,visual analysis showed morphologically CR-ir neurons were darker stained with smaller or irregular cell somas and varicose,segmented,shorter processes in the areas within or bordering the lesion.From the unfolded map,the location and distribution of the cortical lesion was consistent with our previous study,covering the auditory,visual,somatosensory and parietal cortices.However,the size and extent of the lesion varied largely.Severe loss of CR-ir neurons was observed in the core of the cortical lesion,while comparable or more cells were present in the areas close or distal to the lesion.Sometimes the neuronal loss was extensive and widespread throughout the cerebral cortex.In addition,at 3 months after TBI,expression of calretinin-related genes in the perilesional cortex was downregulated,and gene network generated from these CR-related genes regulating several functions such as synaptic transmission,perikaryon,learning and memory,apoptotic process and cognition was altered.Among these genes,Tubb3 showed a trend of downregulation and Reln was upregulated in the perilesional cortex at 2 months post-TBI in RT-qPCR analysis.For the thalamus,at 1 month post-TBI,there was a remarkable reduction in the RT volume ipsilaterally（p<0.001）and a severe loss of CR-ir neurons within RT bilaterally（ipsilaterally,p<0.001;contralaterally,p<0.05）.No correlation was found between the volume reduction and severity of loss of CR-ir neurons in the RT.In addition,there was a pronounced decrease in the density of CR-ir terminals in bilateral VA-VL（ipsilaterally,p<0.01;contralaterally,p<0.05）,particularly ipsilaterally and in the lateral aspect of the nucleus.The decreased CR-ir terminals in the VA-VL was correlated with the loss of CR-ir neurons in the RT ipsilaterally after TBI（r=0.7000,p<0.05,n=9）.At 6 months post-TBI,the gene expression profile of neurotransmitter receptors and regulators in the thalamus was downregulated.Further bioinformatics analysis of CR-related genes suggested alterations in networks of neurological diseases and organismal injury controlled by 2 possible regulators:neuregulin（family）and nicotine.Conclusion:Exposure to peripheral infection at the chronic time point post-TBI can act as facilitating factor,increased seizure susceptibility,augmented anxiety-like behavior and facilitated the progression of cortcal lesion.The mechanisms were related to enhanced neuronal excitability in the perilesional cortex and bilateral dentate gyrus and reactivation of astrocytes.The anti-oxidative/nitrosative stress and anti-inflammatory effects of Sig-1R activation in astrocytes are partially mediated by Nrf2 and HO-1 activation.On another side,our data provided the first quantitative evidence of the remarkable degeneration of the CR immunoreactivity in the cerebral cortex and thalamus and simultaneous CR-releted gene network alterations at a chronic time point after TBI,which may contribute to elucidation of the secondary damage of TBI and mechanisms of post-traumatic epileptogenesis.