S100B Impairs Oligodendrogenesis and Myelin Repair Following Demyelination Through RAGE Engagement

Increased expression of S100B and its specific receptor for advanced glycation end products (RAGE) has been described in patients with multiple sclerosis (MS), being associated with an active demyelinating process. We previously showed that a direct neutralization of S100B reduces lysophosphatidylcholine (LPC)-induced demyelination and inflammation using an ex vivo demyelinating model. However, whether S100B actions occur through RAGE and how oligodendrogenesis and remyelination are affected are not clarified. To evaluate the role of the S100B–RAGE axis in the course of a demyelinating insult, organotypic cerebellar slice cultures (OCSC) were demyelinated with LPC in the presence or absence of RAGE antagonist FPS-ZM1. Then, we explored the effects of the S100B–RAGE axis inhibition on glia reactivity and inflammation, myelination and neuronal integrity, and on oligodendrogenesis and remyelination. In the present study, we confirmed that LPC-induced demyelination increased S100B and RAGE expression, while RAGE antagonist FPS-ZM1 markedly reduced their content and altered RAGE cellular localization. Furthermore, FPS-ZM1 prevented LPC-induced microgliosis and astrogliosis, as well as NF-κB activation and pro-inflammatory cytokine gene expression. In addition, RAGE antagonist reduced LPC-induced demyelination having a beneficial effect on axonal and synaptic protein preservation. We have also observed that RAGE engagement is needed for LPC-induced oligodendrocyte (OL) maturation arrest and loss of mature myelinating OL, with these phenomena being prevented by FPS-ZM1. Our data suggest that increased levels of mature OL in the presence of FPS-ZM1 are related to increased expression of microRNAs (miRs) associated with OL differentiation and remyelination, such as miR-23a, miR-219a, and miR-338, which are defective upon LPC incubation. Finally, our electron microscopy data show that inhibition of the S100B–RAGE axis prevents axonal damage and myelin loss, in parallel with enhanced functional remyelination, as observed by the presence of thinner myelin sheaths when compared with Control. Overall, our data implicate the S100B–RAGE axis in the extent of myelin and neuronal damage, as well as in the inflammatory response that follows a demyelinating insult. Thus, prevention of RAGE engagement may represent a novel strategy for promoting not only inflammatory reduction but also neuronal and myelin preservation and/or remyelination, improving recovery in a demyelinating condition as MS.