Genetics

This post asks what MS susceptibility genes, almost all of which point at immune cells, are doing in neural tissue, and whether anyone can show it rather than assume it. It covers the bioinformatics case for KIF21B as a shared MS and schizophrenia risk gene, expressed in both brain and immune tissues, placed inside microtubule-based axonal transport machinery by protein interaction network analysis yet correlating with T cell activation and lymphocyte differentiation by co-expression. It covers the preprint that instruments plasma proteins with pQTLs from UK Biobank and deCODE, tests them by Mendelian randomization and colocalization against brain imaging-derived phenotypes, reports 120 cis and 20 trans associations, validates them in brain tissue proteomic and transcriptomic data, and finds eleven proteins with direct effects on risk across five brain disorders including MS. And it covers the astrocyte-specific oestrogen receptor beta deletion that reproduced the effects of ovariectomy on the hippocampus, with raised astrocyte Enolase 1 tracking worse cognition, built on the earlier result that oral estriol reduced cerebral cortex atrophy in women with MS.