Laleh Majlessi, Wafa Frigui, Christophe Guilhot, Mickael Orgeur, Philip Supply, Alessandro Cascioferro, Alexandre Pawlik, Timothy P. Stinear, Christiane Bouchier, Roland Brosch, Mamadou Daffé, Gilles Etienne, Laurence Ma, Françoise Laval, Eva C. Boritsch, Fabien Le Chevalier, Wladimir Malaga, Pathogénomique mycobactérienne intégrée, Institut Pasteur [Paris], Institut de pharmacologie et de biologie structurale (IPBS), Centre National de la Recherche Scientifique (CNRS)-Université Toulouse III - Paul Sabatier (UT3), Université Fédérale Toulouse Midi-Pyrénées-Université Fédérale Toulouse Midi-Pyrénées, Université Sorbonne Paris Cité (USPC), Génomique (Plate-Forme) - Genomics Platform, University of Melbourne, Centre d’Infection et d’Immunité de Lille - INSERM U 1019 - UMR 9017 - UMR 8204 (CIIL), Centre National de la Recherche Scientifique (CNRS)-Centre Hospitalier Régional Universitaire [Lille] (CHRU Lille)-Université de Lille-Institut National de la Santé et de la Recherche Médicale (INSERM)-Institut Pasteur de Lille, Réseau International des Instituts Pasteur (RIIP)-Réseau International des Instituts Pasteur (RIIP), The authors acknowledge support from a European Community grant (no. 260872), the EU-EFPIA Innovative Medicines Initiative (grant no. 115337), the Agence National de Recherche (ANR-14-JAMR-001-02) and the Fondation pour la Recherche Médicale FRM (DEQ20090515399 and DEQ20130326471). High-throughput sequencing was performed on the Genomics Platform, a member of the ‘France Génomique’ consortium (ANR10-INBS-09-08). R.B. is a member of the LabEx consortium IBEID at the Institut Pasteur. F.L.-C. was supported by the French Region Ile-de-France (Domaine d’Intérêt Majeur Maladies Infectieuses et Emergentes) PhD programme. E.C.B. was supported by a stipend from the Pasteur–Paris University (PPU) International PhD programme and the Institut Carnot Pasteur Maladies Infectieuses., ANR-14-JAMR-0001,noTBsec,New intervention strategy for tuberculosis: blocking multiple essential targets(2014), ANR-10-INBS-0009,France-Génomique,Organisation et montée en puissance d'une Infrastructure Nationale de Génomique(2010), European Project: 260872,EC:FP7:HEALTH,FP7-HEALTH-2010-single-stage,MM4TB(2011), European Project: 115337,EC:FP7:SP1-JTI,IMI-JU-03-2010,PreDiCT-TB(2012), Pathogénomique mycobactérienne intégrée - Integrated Mycobacterial Pathogenomics, Institut Pasteur [Paris]-Centre National de la Recherche Scientifique (CNRS), Centre National de la Recherche Scientifique (CNRS)-Université Toulouse III - Paul Sabatier (UPS), Génomique (Plate-Forme), Department of Microbiology and Immunology, Centre d’Infection et d’Immunité de Lille (CIIL) - U1019 - UMR 8204 (CIIL), Institut Pasteur de Lille, Réseau International des Instituts Pasteur (RIIP)-Réseau International des Instituts Pasteur (RIIP)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Institut Pasteur [Paris] (IP), Université Toulouse III - Paul Sabatier (UT3), Université de Toulouse (UT)-Université de Toulouse (UT)-Centre National de la Recherche Scientifique (CNRS), Réseau International des Instituts Pasteur (RIIP)-Réseau International des Instituts Pasteur (RIIP)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université de Lille-Centre Hospitalier Régional Universitaire [Lille] (CHRU Lille)-Centre National de la Recherche Scientifique (CNRS), Lassailly-Bondaz, Anne, programmation conjointe européenne sur la résistance antimicrobienne - New intervention strategy for tuberculosis: blocking multiple essential targets - - noTBsec2014 - ANR-14-JAMR-0001 - JPI AMR - VALID, Organisation et montée en puissance d'une Infrastructure Nationale de Génomique - - France-Génomique2010 - ANR-10-INBS-0009 - INBS - VALID, More Medicines for Tuberculosis - MM4TB - - EC:FP7:HEALTH2011-02-01 - 2016-01-31 - 260872 - VALID, Model-based preclinical development of anti-tuberculosis drug combinations - PreDiCT-TB - - EC:FP7:SP1-JTI2012-05-01 - 2017-04-30 - 115337 - VALID, and Université Fédérale Toulouse Midi-Pyrénées-Université Fédérale Toulouse Midi-Pyrénées-Centre National de la Recherche Scientifique (CNRS)
International audience; Mycobacterium tuberculosis is a major, globally spread, aerosol-transmitted human pathogen, thought to have evolved by clonal expansion from a Mycobacterium canettii-like progenitor. In contrast, extant M. canettii strains are rare, genetically diverse, and geographically restricted mycobacteria of only marginal epidemiological importance. Here, we show that the contrasting evolutionary success of these two groups is linked to loss of lipooligosaccharide biosynthesis and subsequent morphotype changes. Spontaneous smooth-to-rough M. canettii variants were found to be mutated in the polyketide-synthase-encoding pks5 locus and deficient in lipooligosaccharide synthesis, a phenotype restored by complementation. Importantly, these rough variants showed an altered host–pathogen interaction and increased virulence in cellular- and animal-infection models. In one variant, lipooligosaccharide deficiency occurred via homologous recombination between two pks5 genes and removal of the intervening acyltransferase-encoding gene. The resulting single pks5 configuration is similar to that fixed in M. tuberculosis, which is known to lack lipooligosaccharides. Our results suggest that pks5-recombination-mediated bacterial surface remodelling increased virulence, driving evolution from putative generalist mycobacteria towards professional pathogens of mammalian hosts.Tuberculosis is a major human infectious disease. Although many aspects of the disease-causing potential of its aetiological agent M. tuberculosis are known1, our understanding of the molecular events that favoured its evolutionary success as one of the most widely distributed human pathogens remains scant. New insights into this question are important for uncovering the mechanisms of pathogenesis and new drug targets2. Strains of the closely related and phylogenetically early branching M. canettii, also named ‘smooth tubercle bacilli’ (STB)3, are powerful resources to investigate the evolution of M. tuberculosis and the M. tuberculosis complex (MTBC)3. The first strain of M. canettii was isolated by Georges Canetti in 1969, and since then, fewer than 100 isolates have been described, most of which have been isolated from tuberculosis patients with a connection to the Horn of Africa4,5,6,7. Despite their geographic restriction, M. canettii strains show much greater genetic variability and are less virulent/persistent than M. tuberculosis. Genome comparisons suggest that M. tuberculosis evolved by clonal expansion from a pool of M. canettii-like tubercle bacilli through the gain of virulence and persistence mechanisms3,8,9. Although some genomic differences have been found to be specific for a single M. canettii strain, apparently due to isolated horizontal gene transfer (for example, the mce5 operon3 or the eptABCD operon10, present exclusively in strain STB-J)3, others are conserved throughout all M. canettii strains as a result of phylogenetic ancestry (for example, cobF, which is present in M. canettii and deleted from the MTBC)3.Here, we investigate phenotypic differences between M. canettii and M. tuberculosis and focus on the unique, conserved, smooth (S) colony morphotype of M. canettii that contrasts with the rough (R) morphotype of MTBC members. Previously, S morphotypes of non-tuberculous mycobacterial species such as Mycobacterium avium11, Mycobacterium abscessus12, Mycobacterium kansasii13 or Mycobacterium marinum14 have been found to be less virulent than R morphotypes, raising the question of whether the highly conserved M. tuberculosis R morphotype might have been the result of evolutionary selection, based on host–pathogen interactions favouring a more virulent or persistent phenotype. In other mycobacterial species, S/R variation is often attributed to different kinds of cell surface glycolipid, such as glycopeptidolipid (GPL) for M. avium15 and M. abscessus16,17 or lipooligosaccharide (LOS) for M. kansasii13 and M. marinum14,18,19. Insights from earlier studies of M. canettii have remained abstruse, as no specific lipid exclusively present in the S morphotype has been identified20, nor has the genetic basis for morphotype variation been determined4.Building on recent data from several M. canettii genomes3, we studied smooth and spontaneously converted R variants of two different M. canettii strains, STB-K (CIPT 140070010) and STB-I (CIPT 140070007)3, hereafter termed KS/R and IS/R, respectively. We used whole genome sequencing (WGS) and identified differences in the genes of the pks5 locus, which in Mycobacterium smegmatis, M. marinum or M. kansasii are implicated in LOS biosynthesis14,21,22,23. In this Article, we uncover the mechanisms underlying the S-to-R morphology change of tubercle bacilli and explore the biological consequences with emphasis on the patho-evolution of M. tuberculosis.