Neuroinflammation and astrocytic reaction in the course of Phoneutria nigriventer (armed-spider) blood-brain barrier (BBB) opening

Phoneutria nigriventer spider venom (PNV) causes uneven BBB permeability throughout different cerebral regions. Little is known about cellular and molecular responses which course with the PNV-induced BBB opening. We investigate by immunohistochemistry (IHC) and Western blotting (WB), the GFAP, S100, IFN-gamma and TNF-alpha proteins expression in hippocampus and cerebellum after different time-points from venom or saline intravenous injection. All proteins variably altered its expression temporally and regionally. WB showed increased GFAP content at 15-45 min followed by a shift below the control level which was less pronounced in hippocampus. IHC showed reactive gliosis during all the trial period. In cerebellum, GFAP was mostly immunodetected in astrocytes of the molecular layer (Bergmann glia), as was S100 protein. The maximum S100 immunolabeling was achieved at 5 h. IFN-gamma and TNF-alpha, expressed mostly by hippocampal neurons, increased along the trial period, suggesting a role in BBB permeability. In envenomed animals, closer contacts astrocyte-astrocyte, granule cells-granule cells and astrocytes-Purkinje cells were observed in cerebellum. Closer contacts between neurons-neurons astrocytes-astrocytes were also seen in hippocampus. PNV contains serotonin, histamine, Ca2+ channels-blocking toxins, some of which affect glutamate release. The hypothesis that such substances plus the cytokines generated, could have a role in BBB permeability, and that calcium homeostasis loss and disturbance of glutamate release are associated with the marked GFAP/S100 reaction in Bergmann glia is discussed. The existence of a CNS mechanism of defense modulated differentially for fast synthesis and turnover of GFAP, S100, IFN-gamma and TNF-alpha proteins was evident. A clear explanation for this differential modulation is unclear, but likely result from regional differences in astrocytic/neuronal populations, BBB tightness, and/or extent/distribution of microvasculature and/or ion channels density/distribution. Such differences would respond for transient characteristics of BBB disruption. This in vivo model is useful for studies on drug delivery throughout the CNS and experimental manipulation of the BBB. (C) 2009 Elsevier Inc. All rights reserved.