Roles of astrocytic connexin-43, hemichannels, and gap junctions in oxygen-glucose deprivation/reperfusion injury induced neuroinflammation and the possible regulatory mechanisms of salvianolic acid B and carbenoxolone

Glia-mediated neuroinflammation is related to brain injury exacerbation after cerebral ischemia/reperfusion (I/R) injury. Astrocytic hemichannels or gap junctions, which were mainly formed by connexin-43, have been implicated in I/R damage. However, the exact roles of astrocytic hemichannels and gap junction in neuroinflammatory responses induced by I/R injury remain unknown.

We propose a vicious cycle exists between astrocytic hemichannel and microglial activation after OGD/R injury, which would aggravate neuroinflammatory responses and neuronal damage. Astrocytic Cx43, hemichannels, and GJIC play critical roles in OGD/R injury-induced neuroinflammatory responses; treatment differentially targeting astrocytic Cx43, hemichannels, and GJIC may provide novel avenues for therapeutics during cerebral I/R injury.

Primary cultured astrocytes were subjected to OGD/R injury, an in vitro model of I/R injury. Salvianolic acid B (SalB) or carbenoxolone (CBX) were applied for those astrocytes. Besides, Cx43 mimetic peptides Gap19 or Gap26 were also applied during OGD/R injury; Cx43 protein levels were determined by western blot and cytoimmunofluorescene staining, hemichannel activities by Ethidium bromide uptake and ATP concentration detection, and gap junction intercellular communication (GJIC) permeability by parachute assay. Further, astrocyte-conditioned medium (ACM) was collected and incubated with microglia. Meanwhile, ATP or apyrase were applied to explore the role of ATP during OGD/R injury. Microglial activation, M1/M2 phenotypes, and M1/M2-related cytokines were detected. Also, microglia-conditioned medium (MEM) was collected and incubated with astrocytes to further investigate its influence on astrocytic hemichannel activity and GJIC permeability. Lastly, effects of ACM and MCM on neuronal viability were detected by flow cytometry.

We found that OGD/R induced abnormally opened hemichannels with increased ATP release and EtBr uptake but reduced GJIC permeability. WB tests showed decreased astrocytic plasma membrane's Cx43, while showing an increase in cytoplasma. Treating OGD/R-injured microglia with ATP or OGD/R-ACM induced further microglial activation and secondary pro-inflammatory cytokine release, with the M1 phenotype predominating. Conversely, astrocytes incubated with OGD/R-MCM exhibited increased hemichannel opening but reduced GJIC coupling. Both SalB and CBX inhibited abnormal astrocytic hemichannel opening and ATP release and switched the activated microglial phenotype from M1 to M2, thus providing effective neuroprotection. Application of Gap19 or Gap26 showed similar results with CBX. We also found that OGD/R injury caused both plasma membrane p-Cx43(Ser265) and p-Src(Tyr416) significantly upregulated; application of SalB may be inhibiting Src kinase and attenuating Cx43 internalization. Meanwhile, CBX treatment induced obviously downregulation of p-Cx43(Ser368) and p-PKC(Ser729) protein levels in plasma membrane. Conclusions: We propose a vicious cycle exists between astrocytic hemichannel and microglial activation after OGD/R injury, which would aggravate neuroinflammatory responses and neuronal damage. Astrocytic Cx43, hemichannels, and GJIC play critical roles in OGD/R injury-induced neuroinflammatory responses; treatment differentially targeting astrocytic Cx43, hemichannels, and GJIC may provide novel avenues for therapeutics during cerebral I/R injury.