No abstract

The molecular processes underlying epidemic waves of methicillin-resistant Staphylococcus aureus (MRSA) are poorly understood1. While a major role has been attributed to the acquisition of virulence determinants by horizontal gene transfer2, there are insufficient epidemiological and functional data supporting that concept. We here report the spread of clones containing a previously extremely rare3,4 mobile genetic element-encoded gene, sasX. We demonstrate that sasX has a key role in MRSA colonization and pathogenesis, significantly enhancing nasal colonization, lung disease, and abscess formation, and promoting mechanisms of immune evasion. Moreover, we observed the recent spread of sasX from sequence type (ST) 239 to invasive clones belonging to other STs. Our study identifies sasX as a quickly spreading critical determinant of MRSA pathogenic success and a promising target for therapeutic interference. Importantly, our results provide proof-of-principle that horizontal gene transfer of key virulence determinants drives MRSA epidemic waves.

The current pandemic of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) skin infections is caused by several genetically unrelated clones. Here, we analyzed virulence of globally occurring CA-MRSA strains in a rabbit skin infection model. We used rabbits because neutrophils from this animal species have relatively high sensitivity to Panton-Valentine leukocidin (PVL), a toxin epidemiologically correlated with many CA-MRSA infections. Virulence in the rabbit model correlated with in-vitro neutrophil lysis and transcript levels of phenol-soluble modulin α and α-toxin, but not PVL genes. Furthermore, abscesses caused by USA300 and its PVL-negative progenitor USA500 were comparatively large and similar in size, suggesting PVL has had a limited role in the evolution of USA300 virulence in the context of skin infections. Our study indicates a major but not exclusive impact of virulence on the epidemiological success of USA300 and other CA-MRSA strains and emphasizes the importance of core genome-encoded toxins in CA-MRSA skin infections.

c Phthiocerol dimycocerosates (PDIMs) and structurally related phenolic glycolipids (PGLs) are complex cell wall lipids unique to pathogenic mycobacteria. While these lipids have been extensively studied in recent years, there are conflicting reports on some aspects of their biosynthesis and on the role of PDIMs and especially PGLs in virulence of Mycobacterium tuberculosis. This has been complicated by the natural deficiency of PGLs in many clinical strains of M. tuberculosis and the frequent loss of PDIMs in laboratory M. tuberculosis strains. In this study, we isolated seven mutants of Mycobacterium marinum deficient in PDIMs and/or PGLs in which multiple genes of the PDIM/PGL biosynthetic locus were disrupted by transposon insertion. Zebrafish infection experiments showed that M. marinum strains lacking one or both of these lipids were avirulent, suggesting that both PDIMs and PGLs are required for virulence. We also found that these strains were hypersensitive to antibiotics and exhibited increased cell wall permeability. Our studies provide new insights into the biosynthesis of PDIMs/PGLs and may help us to understand the role of PDIMs and PGLs in M. tuberculosis virulence. P athogenic mycobacteria produce two structurally related, methyl-branched fatty acid-containing lipids called phthiocerol dimycocerosates (PDIMs) and phenolic glycolipids (PGLs). PDIMs and PGLs have long-chain fatty acid backbones consisting of 3-methoxy (or 3-keto, 3-hydroxy), 4-methyl, 9,11-dihydroxy glycols (phthiocerols) and p-glycosylated phenylglycols (glycosyl phenolphthiocerols), respectively, that are diesterified with di-, tri-, and tetramethyl-branched acyl chains (mycocerosates) (reviewed in reference 28). PDIMs have been identified in Mycobacterium tuberculosis, M. africanum, M. bovis, M. leprae, M. marinum, M. ulcerans, M. kansasii, M. haemophilum, M. microti, and M. gastri, all of which are pathogenic for humans or animals. PGLs are produced by the same set of pathogenic mycobacterial species, except that in M. tuberculosis only a subset of clinical isolates produces PGLs.The role of PDIMs in virulence was first suggested by two independent studies using signature-tagged transposon mutagenesis, which identified mutants of M. tuberculosis that were unable to either produce or properly localize PDIMs to the cell wall and demonstrated that these mutants were attenuated in animal models of infection (8,12,33). Since then, circumstantial evidence supporting a role for PDIMs in M. tuberculosis virulence has accumulated. The role of PGLs in M. tuberculosis virulence is less clear and is confounded by the fact that laboratory strains (H37Rv, Erdman) and many clinical isolates, including CDC1551 and MT103, are naturally deficient in PGL production due to a 7-basepair deletion in pks15/1, while some clinical isolates of the East Asian lineage have an intact pks15/1 gene and produce PGLs (31). Mutations of pks15/1 in M. tuberculosis HN878, a strain that produces both PDIMs and PGLs and exhibits a hypervirulent phenotype in infect...

The proline-glutamic acid (PE) and proline-proline-glutamic acid (PPE) family proteins are prevalent in pathogenic mycobacteria and play a diverse role in mycobacterial pathogenesis. While some members have been studied, the function of most PE/PPE proteins remains unknown. In this study, we isolated a transposon-inactivated PPE38 mutant of Mycobacterium marinum and characterized its phenotype. We found that the PPE38 protein is associated with the cell wall and exposed on the cell surface. The inactivation of PPE38 altered the bacterial cell surface properties and led to deficiencies in cord formation, sliding motility, and biofilm formation. The PPE38 mutant was defective in phagocytosis by macrophages and exhibited reduced virulence in adult zebrafish. We also found that PPE38 is involved in the induction of proinflammatory cytokines in infected macrophages. Together, our results indicate that PPE38, a previously uncharacterized protein, plays a role in mycobacterial virulence, presumably by modulating the host innate immune response. Mycobacterium tuberculosis is a highly successful human pathogen that latently infects one-third of the world's population, causing 10 million new infections and 2 million deaths annually. One reason for the success of M. tuberculosis lies in its ability to evade host immune defense mechanisms and to create a niche within host cells, enabling the bacterium to persist for long periods. M. tuberculosis infects and survives within macrophages by evading macrophage killing mechanisms through a variety of strategies (reviewed in references 51 and 56). During persistent infection, M. tuberculosis is thought to reside within a granuloma, a cellular accumulation around the bacilli that is comprised mainly of macrophages, dendritic cells, T cells, B cells, and fibroblasts (reviewed in reference 31). Granulomas are generally thought to contain the infection by limiting bacterial growth and spread (21, 61). However, recent studies by Ramakrishnan and coworkers using zebrafish embryos demonstrated that the granuloma is a dynamic structure and that Mycobacterium marinum, an aquatic mycobacterium closely related to M. tuberculosis, uses the granuloma as a niche to recruit uninfected macrophages, raising the possibility that granulomas are exploited by the pathogen for expansion and dissemination in early infection (23,25,26).PE and PPE family proteins, named for the conserved N-terminal-domain-containing proline-glutamic acid (PE) motif or proline-proline-glutamine (PPE) motif, are unique to mycobacteria and particularly prevalent in pathogenic mycobacteria (35). They account for a significant fraction (ϳ10%) of the coding capacity of pathogenic mycobacteria. There are 168 PE/PPE proteins in M. tuberculosis (22) and 281 in M. marinum (64). The relatively conserved N termini are approximately 110 and 180 amino acids in the PE and PPE families, respectively. The C-terminal domains of both the PE and PPE protein families are highly variable in both size and sequence and often contain repetitive ...