Using comparative genomics to identify virulence traits and vaccine candidates in Mannheimia haemolytica

Abstract

Bovine respiratory disease (BRD) is the principal cause of morbidity and mortality among feedlot cattle. Mannheimia haemolytica is consistently implicated in this condition, but treatment options are diminishing with the rise of antimicrobial resistance and intensifying consumer pressure to reduce reliance on conventional therapies. Thus, sustainable alternatives like vaccination are required. In this study, the phenotypic and genotypic diversity of BRD pathogens were examined with the objective to identify vaccine targets using reverse vaccinology, an innovative approach to identify antigens via genomic sequence. Preliminary surveillance confirmed M. haemolytica serotype 2 isolates were predominant in healthy animals (75.5%) while serotypes 1 (70.7%) and 6 (19.5%) were common in diseased animals. Pathogens of BRD, including M. haemolytica, Pasteurella multocida and Histophilus somni were also isolated from North American BRD mortalities, and compared using pulsed-field gel electrophoresis and antimicrobial susceptibility. Concurrently, polymerase chain reaction detection of bacterial and viral agents confirmed that M. haemolytica with bovine viral diarrhea virus were the most prevalent. Whereas isolates from live cattle were found to have a relatively low level of resistance, several pathogens from the mortalities were found to contain integrative conjugative elements (ICE) conferring resistance to seven antimicrobial classes. These ICEs were transferred via conjugation to other bacterial species, emphasizing the need for alternative antimicrobial therapies. Collectively, data from these investigations informed the selection of 11 diverse M. haemolytica strains for whole genome sequencing and comparative analyses. Several bacteriophage associated genes and CRISPR-Cas regulated gene expression systems were identified and are likely contributing to virulence in M. haemolytica. Coding sequences across all genomes were screened using pan-genome analysis, identifying 291 candidates with cell-surface associated signatures. Using a cell-free translation system and enzyme-linked immunosorbent assay the candidates were screened against serum from cattle challenged with serovar 1, 2 or 6 of M. haemolytica, and ranked according to immunogenicity. The top five vaccine candidates included Ssa1, ComE, a solute binding protein, an outer membrane protein, and the periplasmic component of an ABC transporter. With further characterization, these unique antigenic candidates could be developed into a vaccine to effectively reduce the dependence on antimicrobial therapies.