Effect Of Specific Silence Of Cx43on Cerebral Vasospasm After Subarachnoid Hemorrhage

Background:Spontaneous subarachnoid hemorrhage(SAH) is a severe and common cerebrovascular disease. The World Health Organization has reported that the incidence of SAH among different countries is various from2.0/10ten thousand people in China to22.5/10ten thousand people in Finland. And nearly50%of SAH patients died from delayed cerebral vascular vasospasm(CVS). Therefore, CVS is the main course of serious disability or death after SAH. Although intensive and deep researches has been made in the past years on the pathogenesis of CVS after SAH, the exact mechanism has not been fully understood yet. This fact has prevented the development of a specific treatment.Gap junction(GJ) is a direct channel of information transmission between adjacent cells. It is anchored by two connexon on adjacent membrane, which is composed of six subunit connexins. At present at least twenty different subunits were isolated in gap junction protein family. It has been reported that Cx40, Cx43, cx45and Cx37are the mainly component connexins in the vascular wall. The center of gap junctions is a small hydrophilic tube with about2nm diameter, allowing the molecular weight of less than1.5KD ion and small molecules(such as Ca2+, cAMP, cGMP, IP3etc) passing through. The material exchange between neighboring cells is called the gap junctional intercellular communication, which play an-important physiological function in the coordination contraction in heart and smooth muscle cells, information transfer, cell growth control, tumorigenesis, nerve conduction and embryonic development.Previously we found the upregulation of gap junctional protein43(Cx43) expression in the development of CVS after experimental SAH. In this study, further experimental research on the potential pathological role of Cx43in cerebral vasospasm was investigated with specific RNA interfering technique. Firstly we screened the optimal siRNA strand in cultured smooth muscle cells from basilar artery. Secondly recombinant adenovirus carrying the optimal siRNA screened from above in vitro was constructed. Thirdly the specific silence of Cx43expression was detected after intracisternal injection of adenovirus-mediated shRNA. Finally, the inhibitory effect of Cx43RNAi on vasospasm after SAH was investigated in a rabbit vivo model to certify the pathophysiological role of GJ in CVS.Materials and methods:The vascular smooth muscle cells (VSMCs)taken from the media of rabbits’basilar arteries(BAs) were cultivated with the explant attachment method and identified with light phase contrast microscopy and immunocytochemical staining. Semiquantitative RT-PCR was performed to detect the relative content of mRNA for connexins in those cells. The P-actin primers were used as internal control for semiquantitative RT-PCR. The products of PCR were electrophoresed in agarose gel stained with EB and then photographed and semiquantitated with derisitometer scanning of the ethidium bromide-stained agarose gels. The density ratio of connexins RT-PCR band to the β-actin RT-PCR band was used as the relative content of connexin mRNA. The results of semiquantitative RT-PCR were taken down as means±SE and analyzed with t-test. According to Elbashir SM and Tuschl T RNAi guidelines, three double-stranded siRNAs targeting different coding regions of Cx43gene of rabbit and one non-specific siRNA were designed and chemically synthesized. The synthesized siRNAs were transfected into the cultured VSMCs by siPORT NeoFX. Subsequently Cx43mRNA expression was evaluated by semiquantitative RT-PCR after transfection for24h,48h,72h. Then optimal siRNA among the three was screened based on their silencing effect on Cx43gene ecpression. Scrape loading dye transfer (SLDT) was used to determine gap junctional intercellular communication(GJIC) function after transfected with the optimal siRNA.Recombinant adenovirus shuttle plasmid was constructed with inserting the rabbit Cx43specific shRNA scfeened from above optimal siRNA in empty vector. Adenoviral vector system was used to produce recombinant adenovirus Ad5-H1-connexin43-shRNA-GFP through co-transfection adenovirus vector plasmid into293cells. Recombinant adenovirus carrying shRNA was injected into the rabbit cisterna magna. Two rabbits in each groups were euthanized after Id,3d,5d,7d and14d. And the transfection efficacy of recombinant adenovirus on the basilar arteries was detected with fluorescence microscope, light microscope and RT-PCP.After established a model of CVS following SAH in rabbit, digital subtraction angiography was performed to detect the change of the BA diameter before and after treatment or pretreatment with intracisternal injection of Ad5-Hl-connexin43-shRNA-GFP. Western blotting was used to analyze the expression change of Cx43protein in basilar artery tissues of different groups. The morphological changes of the BA were observed under a light microscope. Results:The cell phenotype and anti-α-actin antibody immunocytochemical staining indicated that the cultured cells were smooth muscle cells and the purity of the cells was higher. The sequence of three pairs double-stranded siRNAs targeting different coding regions of Cx43gene and one non-specific siRNA are as following:①378siRNA:CCTGA-AGCAGATTGAGATA;②454siRNA:CTGCG AACCTACATCATCA;③654siRNA:GGTGTCTCTTGCCTTGGAAT;④non-siRNA:TTCTCCGAACGTGT-CACGT. The result of semiquantitative RT-PCR analysis demonstrated that the gray scale of the cells transfected with the three pairs double-stranded siRNAs was significantly lower than that of the control group and non-specific siRNA transfected group with the dose dependant manner. The relative inhibitory rates of378siRNA with the concentration of50nmol·L-1,100nmol·L-1and150nmol·L-1were62.12±2.43%,73.23±3.11%and82.31±5.14%respectively compared to the control (P<0.01). The mean inhibitory rates of the different concentration by454siRNA and654siRNA is52.61±1.92%and31.84±2.82%respectively (p<0.01). No significant change of Cx43mRNA expression was detected in non-specific siRNA group (p>0.05. Obviously, the inhibitory effect of378siRNA on Cx43expression was the best. After transfection of150nmol·L-1378siRNA for8h,24h,48h,72h and96h. The inhibitory ratios on Cx43expression were28.71±0.83%,38.7±0.81%,75.76±3.61%,85.93±0.87%and29.32±2.31%respectively (P<0.05). The inhibitory effect of378siRNA could be detected as early as8h, and as late as1w with time-dependent manner. The GJIC function of the VSMCs was reduced by150nmol·L-1378siRNA, with the total number of communicating cells on either side of the scrape line in the group less than that of the control group (p<0.05. Only1～2layers in378siRNA group were observed compared with3～4layers in control group.After identification, recombinant adenovirus carrying shRNA was constructed successfully, and the purification was more than95%. One day after the intracisternal injection, the BA adventitia was slightly transfected by Ad5-Hl-connexin43-shRNA-GFP. The maximum transfection through the whole vessel wall was observed after5days, and the transfection was still significant until after7days. Even after14days, a slight transfection could be detected in the BA daventitia. The rabbit double-hemorrhage model of SAH was successfully established. Cerebral angiograms on day7showed severe narrowing of the BAs. There were no difference in arterial narrowing between the treatment group and SAH-only group on day7. In contrast, arterial narrowing in the pretreatment group on day7was significantly less than that in the SAH-only group. In the SAH-only group, the expression of Cx43mRNA and protein at1d,3d,5d,7d and14d was significantly higher compared with that before SAH, and exhibited a time-dependent change. The maximum expression was at7d, and decreased at14d. After the intracisternal injection of Ad5-H1-connexin43-shRNA-GFP without SAH, the expression of Cx43mRNA and protein was decreased significantly especially at5d,7d and14d. The expression in the pretreatment group at1d,3d,5d,7d and14d was significantly less than that in the SAH-only group. There were no difference between the treatment group at Id,3d and SAH-only group. In contrast, the expression in the treatment group at7d,14d was significantly less than that in the SAH-only group.Conclusion:Gap junctions play an important role in the pathogenesis of cerebral vasospasm. With specific silences Cx43in vascular wall, cerebral vasospasm after experimental SAH are significantly attenuated. The result demonstrated that over expression of Cx43in the smooth muscle cells is a potential target for the treatment of CVS. Further investigation are needed for the mollecular mechanism of the effect of silencing Cx43on CVS.