Bioinformatics analysis of myelin-microbe interactions suggests multiple types of molecular mimicry in the pathogenesis of multiple sclerosis

Multiple sclerosis (MS) is a devastating autoimmune disease that leads to the destruction of the myelin sheath in the human central nervous system (CNS). Infection by viruses and bacteria has been found to be strongly associated with the onset of MS or its severity. We postulated that the immune system's attack on the myelin sheath could be triggered by viruses and bacteria antigens that resemble myelin sheath components. An in-silico bioinformatics approach was undertaken in order to identify viral and bacterial antigens that resemble myelin oligodendrocyte glycoprotein (MOG) and myelin basic protein (MBP). To this end, we simultaneously analyzed both protein structures and amino acid sequences from viral and bacterial proteins and compared them to MOG and MBP. Possible associations between MBP and human parvovirus B19 (HPV-B19) and adeno-associated virus 4 (AAV-4) capsid protein structures were identified. MBP and MOG were associated with antigens from different viruses and bacteria, including Aspergillus species, Lactobacillus, Burkholderia, Clostridium, Schizosaccharomyces, SARS-CoV-2, and some gut flora metabolites. We also identified similarities between MBP and MOG proteins and bile salt hydrolase (BSH), glycosyltransferase (WcfQ), and Wzy enzymes. Identical amino acids between MBP and BSH at the active site, and protected amino acids in MOG aligning with WcfQ and Wzy enzymes were observed. Overall, our results offer valuable insights into the role of different viral and bacterial protein antigens in MS pathogenesis and suggest the possibility of identifying new therapeutic targets using in silico bioinformatics approaches. Our proposed approach could also likely be adapted for other CNS diseases in order to develop new biological insights and treatments.