The current tuberculosis vaccine is a live vaccine derived from Mycobacterium bovis and attenuated by serial in vitro passaging. All vaccine substrains in use stem from one source, strain Bacille Calmette-Guérin. However, they differ in regions of genomic deletions, antigen expression levels, immunogenicity, and protective efficacy.
As a RecA phenotype increases genetic stability and may contribute restricting the ongoing evolution of the various BCG substrains while maintaining their protective efficacy, we aimed to inactivate recA by allelic replacement in BCG vaccine strains representing different phylogenetic lineages (Pasteur, Frappier, Denmark, Russia). Homologous gene replacement was achieved successfully in three out of four strains. However, only illegitimate recombination was observed in BCG substrain Russia. Sequence analyses of recA revealed that a single nucleotide insertion in the 5' part of recA led to a translational frameshift with an early stop codon making BCG Russia a natural recA mutant. At the protein level BCG Russia failed to express RecA.
According to phylogenetic analyses BCG Russia is an ancient vaccine strain most closely related to the parental M. bovis. We hypothesize that recA inactivation in BCG Russia occurred early and is in part responsible for its high degree of genomic stability, resulting in a substrain that has less genetic alterations than other vaccine substrains with respect to M. bovis AF2122/97 wild-type.