Phylo-epidemiological and pathogenic diversity of Mycobacterium tuberculosis strains in London with implications for vaccine develpoment

Abstract

Approximately one-third of the global population is infected with tuberculosis causing approximately 1.7 million deaths. Currently, the BCG vaccine is used to protect against TB, but it cannot prevent primary infection or reactivation of latent infection. Ideally a vaccine should protect against a diverse array of Mycobacterium tuberculosis strains and promote a strong, long-lasting TH1 cell-mediated immune response. Whilst evaluating the efficiency of novel vaccines using laboratory control strains (M. tuberculosis H37Rv, H37Ra and M. bovis-BCG), it is important to test efficacy against a representative panel of wild-type circulating strains. In England 42.2% of TB cases are reported in London and the diversity of nationalities generates a diverse pool of strains consisting of globally representative TB strains. The aim of the study was to construct a representative panel of strains for vaccine evaluation studies and general TB research. Common M. tuberculosis strains were identified by performing molecular MIRUVNTR and spoligotyping on 2363 isolates from TB cases reported in London during a one-year period. Epidemiological analysis demonstrated there were representatives from 13 global regions, including high TB burden countries. An algorithm was designed to select strains for a preliminary panel based on associations between MTBC families in clusters of more common strains, the country of birth and VNTR sub-clusters. The preliminary panel contained 42 MTBC strains belonging to 10 MTBC families from patients born in 17 countries. Results of phylogenetic analysis of all 2363 isolates was used to select a smaller panel of strains from the preliminary panel to represent MTBC lineages to investigate if wild-type strains were phenotypically similar. The final panel included five strains from each of the Baker et al., 2004 M. tuberculosis lineages (M. tuberculosis Beijing, LAM10, two CAS, EAI5 strains representing lineage I, II, III, IV, respectively) and an M. africanum strain. In vitro tissue culture experiments demonstrated significantly higher growth of the Beijing strain compared to the other wild-type and laboratory strains. Higher growth rates of this strain were also observed in a cell-free culture system. Aerosol challenge of guinea pigs with wild-type strains showed a quicker dissemination of the EAI5 strain from the lung to the spleen 16 days post-challenge, but significantly higher c.f.u. count of the Beijing strain in the spleen 56 days post-challenge. Collectively, the data demonstrated that there are phenotypic differences between wild-type circulating MTBC strains.