Computer-Optimized Immunization Strategy to Boost Highly Cross-Reactive Antibodies to RSV

Abstract

Abstract Respiratory Syncytial Virus (RSV) is a tremendous burden on public health. RSV infects humans within two years of life and causes continual infections throughout their lifetime. RSV is estimated to cause approximately 30 million episodes of acute lower respiratory infection resulting in over 100,000 deaths annually (in children ages 0 – 5 alone). Despite this burden, no vaccine is currently available for RSV. RSV has been classified into two subtypes (A and B) based on genomic and antigenic properties. Evolutionary rates are similar to other RNA viruses with estimates of 1.83–1.95 x 10−3 nucleotide substitutions/site/year. The high mutation rates of RSV produce an evolutionary landscape that has led to genetic and antigenic variation across and within subtypes. The attachment glycoprotein (G-protein) is must abundant on the surface of the virus and contains an almost completely conserved region, the central conserved domain (CCD). We developed a computational model of the humoral immune system that simulates B cell responses to G-protein antigen. We use a sequence-based approach to approximate antigenic differences across (>1000) RSV strains. We then use computer simulations to identify an immunization strategy that can boost antibody levels to the CCD region. Last, we demonstrate that the immunization strategy results in a highly cross-reactive antibody response.