Beyond phenotype: The genomic heterogeneity of co-infecting Mycobacterium abscessus smooth and rough colony variants in cystic fibrosis patients.

Two unrelated cystic fibrosis patients were co-infected with Mycobacterium abscessus smooth and rough phenotypes. Smooth M. abscessus is proposed as the infecting form, and the subsequent loss of glycopeptidolipids in the host leads to a rough phenotype. Whole-genome sequencing (WGS) diagnosed two different M. abscessus strains in patient N°1 but only one strain in patient N°2. In patient N°1, rough isolate had novel mutations potentially involved in smooth-to-rough morphology changes. In patient N°2, four genes were present in only the smooth isolate. In addition, we obtained different susceptibility profiles in the four clinical isolates. We revealed a new paradigm describing a cystic fibrosis patient infected with two different clones, including a rough isolate, and identifying a rough M. abscessus clone that did not lose glycopeptidolipids. We propose WGS for the identification of heterogenic isolates and genetic determinants of antimicrobial resistance, which we believe will positively influence treatment prognosis.

A Simple and Rapid Gene Disruption Strategy in Mycobacterium abscessus: On the Design and Application of Glycopeptidolipid Mutants.

Little is known about the disease-causing genetic determinants that are used by Mycobacterium abscessus, increasingly acknowledged as an important emerging pathogen, notably in cystic fibrosis. The presence or absence of surface exposed glycopeptidolipids (GPL) conditions the smooth (S) or rough (R) M. abscessus subsp. abscessus (M. abscessus) variants, respectively, which are characterized by distinct infective programs. However, only a handful of successful gene knock-out and conditional mutants have been reported in M. abscessus, testifying that genetic manipulation of this mycobacterium is difficult. To facilitate gene disruption and generation of conditional mutants in M. abscessus, we have designed a one-step single cross-over system that allows the rapid and simple generation of such mutants. Cloning of as small as 300 bp of the target gene allows for efficient homologous recombination to occur without additional exogenous recombination-promoting factors. The presence of tdTomato on the plasmids allows easily sifting out the large background of mutants spontaneously resistant to antibiotics. Using this strategy in the S genetic background and the target gene mmpL4a, necessary for GPL synthesis and transport, nearly 100% of red fluorescent clones exhibited a rough morphotype and lost GPL on the surface, suggesting that most red fluorescent colonies obtained after transformation incorporated the plasmid through homologous recombination into the chromosome. This system was further exploited to generate another strain with reduced GPL levels to explore how the presence of these cell wall-associated glycolipids influences M. abscessus hydrophobicity as well as virulence in the zebrafish model of infection. This mutant exhibited a more pronounced killing phenotype in zebrafish embryos compared to its S progenitor and this effect correlated with the production of abscesses in the central nervous system. Overall, these results suggest that the near-complete absence of GPL on the bacterial surface is a necessary condition for optimal pathogenesis of this mycobacterium. They also suggest that GPL content affects hydrophobicity of M. abscessus, potentially altering the aerosol transmission, which is of particular importance from an epidemiological and clinical perspective.