
Full text loading...
22 The Mycolactones: Biologically Active Polyketides Produced by Mycobacterium ulcerans and Related Aquatic Mycobacteria, Page 1 of 2
< Previous page | Next page > /docserver/preview/fulltext/10.1128/9781555815783/9781555814687_Chap22-1.gif /docserver/preview/fulltext/10.1128/9781555815783/9781555814687_Chap22-2.gifAbstract:
Mycolactones are a family of lipophilic small molecules and virulence factors produced as secondary metabolites by Mycobacterium ulcerans and some highly related aquatic mycobacteria. This chapter describes what is known about mycolactones and their unique role in the pathogenesis of Buruli ulcer, explains their unusual biosynthetic locus, and highlights the key questions that remain to be answered. A recent report describes an abundant extracellular matrix produced by M. ulcerans that harbors vesicles that are rich in mycolactones. Mycolactones are related to antibacterial macrolides such as erythromycin and immunosuppressants such as FK506 because they all share a polyketide-derived macrolactone core. The apoptotic ability of mycolactone F is significantly less than that of the other mycolactones; a feature that may be associated with its shorter acyl side chain. All of these mycolactone-producing mycobacteria, including M. ulcerans, form a distinct lineage within a genetically more diverse assemblage of Mycobacterium marinum that have clearly evolved from a common M. marinum-like progenitor. All mycolactone-producing mycobacteria found so far have been discovered through their ability to cause disease in vertebrates, but they may be acting as “indicator” species and represent only a proportion of the mycolactones that exist in nature.
Full text loading...
Characteristic histopathology of WT M. ulcerans strain 1615 infection in a guinea pig (A and B) compared with M. ulcerans mycolactone negative transposon mutant MU 1615::Tn119 (C and D). (A) M. ulcerans 1615 infection, showing fatty necrosis, sparse cells, and microhemorrhage characteristic of acute M. ulcerans infection in humans (hematoxylin and eosin [H&E] stain); inset, apoptotic cells in an area devoid of bacteria. (B) Extracellular clusters of M. ulcerans 1615 (ZiehlNeelsen stain). (C) M. ulcerans mycolactone-negative mutant 1615::Tn119 showing granulomatous response with large influx of inflammatory cells (H&E stain). (D) Ziehl-Neelsen stain showing the intracellular location of the mycolactone-negative mutant.
Structure of the mycolactones A to F. All structural variations occur in the side chain and are highlighted by gray shading. Minor cometabolites are denoted by *.
Proposed pathways for the biosynthesis of mycolactone A/B as deduced from the complete DNA sequence of pMUM001 and transposon mutagenesis ( Stinear et al., 2004 ). (A) Module and domain arrangement for synthesis of the core. (B) Module and domain arrangement for synthesis of the side chain. (C) Map of the pMUM001 plasmid showing the relative positions of the mls genes and accessory genes. (D) Key for domain function. Identical color indicates both same function and sequence (97 to 100% aa identity). (See the color insert for color version of this figure.)
Alignment of the 16 Mls KS domains, showing the 11 variable aa residues and a phylogenetic reconstruction. The scale bar indicates 1 aa substitution per 1,000 sites.
chapter 6, Schematic representation of mycobacterial LAM. Araf, arabinofuranose; Ins, myo-inositol; Manp, mannopyranose; MPI, mannosyl-phosphatidyl-myo-inositol; Rn, fatty acyl residue. The mean molecular mass of M. tuberculosis and M. bovis BCG ManLAM is around 17 kDa, with a heterogeneity estimated at 6 kDa (Venisse et al., 1993). We estimate that these ManLAM contain approximately 60 Araf and 50 Manp units. Manp units are distributed between the mannose caps and the mannan core (30 to 35 Manp units) (Nigou et al., 2000). The mannan domain of the M. kansasii ManLAM contains a low proportion of disaccharide side chains (Guerardel et al., 2003). 5-Methylthiopentose (MTP) was identified as 5-deoxy-5-methylthio-xylo-furanose (Joe et al., 2006) and has been described on the ManLAM of M. tuberculosis strains (Ludwiczak et al., 2002; Treumann et al., 2002) and ManLAM and LM of a M. kansasii clinical isolate (Guerardel et al., 2003). Succ indicates succinyl residues located on the arabinan domain of ManLAM of M. bovis BCG (Delmas et al., 1997) and of a M. kansasii clinical isolate (Guerardel et al., 2003). One to four succinyl groups, depending on the M. bovis BCG strain, esterify the 3,5-α-Araf units at position O-2 (Delmas et al., 1997), and an average of two succinic acids per LAM was found in the case of M. kansasii ManLAM (Guerardel et al., 2003).
chapter 6, LAM biosynthesis schema. The biosynthesis of the triacylated forms of PIM and lipoglycans is shown. PimA is essential in M. smegmatis (Kordulakova et al., 2002). Rv2611c appears to be essential in M. tuberculosis, but not in M. smegmatis (Kordulakova et al., 2003; G. Stadthagen, M. Jackson, and B. Gicquel, unpublished results). Sugar donors are in blue and identified glycosyl-transferases in red. AcylT, acyl-transferase; ManT(s), mannosyl-transferase(s); AraT(s), arabinosyl-transferase(s); C35/C50, polyprenol; C35/C50-P-Man, polyprenol-monophosphorylmannose; C35/C50-P-Araf, polyprenol-monophosphoryl-β-D-Araf.
chapter 6, Cell signaling pathways triggered by PI-based lipoglycans. (A) LM (Quesniaux et al., 2004; Vignal et al., 2003), and to a lesser extent PIM (Gilleron et al., 2003), activate macrophages and DCs through a TLR2/TLR1-dependent but TLR6-independent pathway that requires MyD88 (Quesniaux et al., 2004). Only Ac3 LM and Ac4LM are active (Gilleron et al., 2006), whereas the residual PIM activity is independent of the acylation degree (from one to four fatty acids) (Gilleron et al., 2003). It is not known whether lipoglycans are presented to the receptor in a monomeric or multimeric form. ManLAM and AraLAM do not signal through TLR2 as a consequence of steric hindrance: the arabinan domain masks the lipomannan moiety of the molecule (Guerardel et al., 2003). The molecular bases of PILAM activity are not clear yet. (B) ManLAM inhibits IL-12 and TNF-α (Nigou et al., 2001) and induces IL-10 production by LPS-stimulated DCs through DC-SIGN ligation (Geijtenbeek et al., 2003). The signaling pathway involves activation of PI3K and ERK1/2 (Caparros et al., 2006). In macrophages, ManLAM inhibits the LPS-induced production of TNF-α and IL-12 (Knutson et al., 1998), independently of IL-10 production, through IRAK-M activation (Pathak et al., 2005). ManLAM exerts other inhibitory activities on macrophages including inhibition of IFN-γ-mediated activation (Sibley et al., 1988), M. tuberculosis-induced apoptosis (Rojas et al., 2000), and phagolysosome biogenesis (Fratti et al., 2003). Phagolysosome biogenesis is associated with ManLAM binding to MR (Kang et al., 2005) and requires inhibition of both the cytosolic Ca2+ rise/calmodulin pathway and PI3K signaling (Vergne et al., 2003). Inhibition of apoptosis (Rojas et al., 2000) and possibly IFN-γ-mediated activation (Briken et al., 2004) are also dependent on the alteration of Ca2+-dependent intracellular events, suggesting that they could be also both mediated by MR. LM and PIM also bind MR and DC-SIGN (Pitarque et al., 2005; Torrelles et al., 2006); however, little is known about the functional consequences. LM induces a TLR2-dependent production of proinflammatory cytokines but concomitantly inhibits, most probably through C-type lectin binding, TLR4-mediated cytokine production (Quesniaux et al., 2004). The net cytokine response is dependent on the receptor equipment of the cells as well as the LM used and their acylation degree (Quesniaux et al., 2004).
Chapter 8, Crystal structure of the M.tb. PE/PPE protein complex. (A) Surface representation of the PE/PPE protein complex. The PE protein Rv2431c is shown in red, and the PPE protein Rv2430c is in blue. (B) The PE/PPE protein complex viewed down its longitudinal axis. (C) Ribbon diagram of the PE/PPE protein complex. The complex is composed of seven helices. Two helices of the PE protein interact with two helices of the PPE protein to form a four-helix bundle. Regions of high sequence conservation are indicated by arrows and discussed in the text. (D) Interface hydrophobicity of the PPE and PE proteins. The hydrophobicity of the interaction interface between the PPE and PE protein is color coded: the most apolar regions are indicated in red, orange, and yellow, and the most polar regions are indicated in blue. Notice the extensive apolar regions that are shielded from solvent as the complex forms. Adapted from Strong et al. (2006).
chapter 09, MspA, a general porin of M. smegmatis. (A) Side view of MspA integrated into a lipid bilayer. (B) Electrostatic potential of MspA in top view. The electrostatic potential is represented by the Gasteiger charges for the atoms in the surface of MspA. Negative charges are shown in red, positive charges are shown in blue. These figures are based on the crystal structure of MspA (Faller et al., 2004). Panels A and B were created with the visualization software PyMol (DeLano Scientific LLC) and ViewerLight (Accelrys), respectively.
Chapter 10, (A) Mycobactin/carboxymycobactin-mediated iron acquisition by intraphagosomal M. tuberculosis. The scheme shows alternatives for (ferri-) MBT/(ferri-)CMBT trafficking and delivery of iron to the bacterial cytoplasm. Under iron-limiting conditions, IdeR repression is decreased, and MBTs and CMBTs are synthesized by the siderophore biosynthesis machinery (SBM) and exported outside the cell by a yet unknown mechanism (a). Lipophilic MBTs have been shown to localize to the bacterial surface, perhaps at the membrane, where they can acquire Fe3+ from ferri-CMBTs (b). MBTs at the cell surface have been suggested to function as ionophores to facilitate Fe3+ transport across the membrane and/or act as transient stores of Fe3+ (c). MBTs can also diffuse throughout the intracellular milieu of the macrophage and acquire Fe3+ from cytoplasmic iron sources to form ferri-MBTs (d). Ferri-MBTs can diffuse in the intracellular milieu and accumulate in lipid droplets in contact with phagosomes (e). MBTs can sequester Fe3+ from transferrin in the macrophage (f). CMBTs can acquire Fe3+ from transferrin in vitro and are likely to do so in vivo as well (g). Porins may facilitate inward trafficking of ferri-CMBTs through the waxy cell envelope (h). Porins may also facilitate inward trafficking of ferri-MBTs. The IrtAB system has been proposed to transport ferri-CMBTs to the bacterial cytoplasm (i), but it is possible that the system transports ferri-MBTs as well. In the cytoplasm, an iron reductase (R) would release the iron from the chelates as Fe2+ (j). It is also possible that a membrane reductase coupled with an iron transport system (FeT) removes Fe3+ from ferrisiderophores at the extracellular side of the membrane and transports the Fe2+ to the cytoplasm (k). Regardless of how Fe2+ is delivered to the cytoplasm, it will be directed to synthesis of iron-containing compounds or temporarily stored (l). (B) Exochelin MS-mediated iron acquisition in M. smegmatis. The MBT/CMBT system of M. smegmatis, which is comparable to that of M. tuberculosis, is not shown. Under iron-limiting conditions, IdeR repression is decreased, and EXCs are synthesized and mobilized outside the cell, possibly through a mechanism involving ExiT (a). Secreted EXCs chelate Fe3+ from environmental sources (b). Ferri-EXCs have been suggested to be bound at the cell envelope by the putative ferrisiderophore receptor FxuD, and possibly by a 29-kDa cell envelope protein not shown (c). Ferri-EXCs are likely to be transported to the cytoplasm by the FxuABC system (d), where an iron reductase would release the iron as Fe2+ (e). Released Fe2+ is directed to synthesis of iron-containing compounds or temporarily stored (f).
chapter 11, Ribbon diagram of the E. coli vitamin B12BtuCDF, ABC permease protein structure. The transporter is assembled from two membrane-spanning BtuC subunits (red and yellow) and two ABC cassettes BtuD (green and blue). At the ATP binding sites, cyclotetravanadate molecules are bound to the transporter (ball-and-stick models at the BtuD interface). Vitamin B12 is delivered to the periplasmic side of the transporter by a binding protein (BtuF, light blue), then translocated through a pathway provided at the interface of the two membrane-spanning BtuC subunits. It finally exits into the cytoplasm at the large gap between the four subunits. This transport cycle is powered by the hydrolysis of ATP by the ABC cassettes BtuD. Reprinted from Locher and Borths (2004), with permission of the authors.
chapter 12, Model of MmpL secretion. The proposed models for PDIM and SL-1 secretion through MmpL7 and MmpL8, respectively, are shown. PpsA-E and Mas are polyketide synthases that extend straight chain fatty acids to phthiocerol and mycocerosic acid, respectively (Azad et al., 1997; Azad et al., 1996; Trivedi et al., 2005). The enzymes FadD26 and FadD28 are thought to be AMP ligases that activate straight-chain fatty acids for transfer to the Pps and Mas enzymes (Trivedi et al., 2004). The thioesterase TesA is also required for the synthesis of PDIM and interacts with PpsE (Rao and Ranganathan, 2004). PapA5 is able to catalyze the esterification of mycocerosic acids to phthiocerol to form PDIM (Onwueme et al., 2004). MmpL7 and DrrABC are required to transport PDIM across the cell membrane (Camacho et al., 1999; Cox et al., 1999). Finally, LppX is a signal sequence containing protein that is thought to transport PDIM across the periplasm to the cell wall (Sulzenbacher et al., 2006). Two models for SL-1 secretion are shown. Pks2 is required for the synthesis of SL1278 (Sirakova et al., 2001). In model A, MmpL8 recruits an as yet unidentified biosynthetic factor to complete the synthesis of SL-1 from SL1278, before transport across the cell membrane (Jain and Cox, 2005). In model B, MmpL8 exports SL1278 across the cell membrane, after which it is converted to SL-1 by an as yet unidentified enzyme (Converse et al., 2003). CM, cytoplasmic membrane; PG, peptidoglycan; mAG, mycolyl-arabinogalactan.
chapter 13, Working model for biogenesis and export of ESAT-6 proteins in M. tuberculosis H37Rv plus the positions of various genes and deletions (Brodin et al., 2004b). The upper part presents a possible functional model indicating predicted subcellular localization and known or potential protein-protein interactions. Rosetta stone analysis indicates direct interaction between proteins Rv3870 and Rv3871; Rv3868 is an AAA-ATPase that is likely to act as a chaperone. ESAT-6 and CFP-10 are predicted to be exported through a transmembrane channel, consisting of at least Rv3870, Rv3871, and Rv3877 and possibly Rv3869 in a process catalyzed by ATP hydrolysis. The lower part shows key genes, the various proteins from the RD1 region, their sizes (number of amino acid residues), and the protein families.
chapter 15, Pictorial depiction of domains predicted in PpsA-E genes and Mas, by PKSDB and ITERDB. The ketide units in the final PDIM structure are indicated by colors similar to the synthesizing modules.
chapter 15, (A) Biosynthesis of PDIM demonstrates thiotemplate-based assembly line enzymology carried out by PpsA-E, FadD26, Mas, and PapA5 proteins. (B) Mycobactin synthesis by mbt-1 and -2 clusters. mbt-1 cluster proteins synthesize the mycobactin core, and mbt-2 cluster proteins along with MbtG modify the core structure of mycobactin.
Chapter 19, Structure of the hbhA locus in different bacterial species. The hbhA gene of the different species indicated is highlighted as the blue arrow in bold. Conserved open reading frames between the different species are shown by the same color. The corresponding putative proteins encoded by the genes are indicated on the top. Chromosomal deletions or absence of genes are indicated by the dotted lines, and pseudogenes in M. leprae are depicted by the crosses.
Chapter 21, Genetic organization of the GPL locus in various mycobacterial species. The ORFs are depicted as arrows and have been drawn proportionally to their size (Ripoll et al., 2007). Color code: light blue, mmpL family; black, unknown; purple, sugar biosynthesis, activation, transfer and modifications; red, lipid biosynthesis, activation, transfer and modifications; green, pseudopeptide biosynthesis; yellow, required for GPL transport to the surface; gray, regulation.
Chapter 21, (A) Transmission electron microscopy analysis of the M. smegmatis wild-type strain mc2155 and the gap mutant strain after ruthenium red staining (Sonden et al., 2005). (B) Topology of Gap protein predicted by Sosui program. (C) Amino acid alignment of various Gap orthologues found in sequenced mycobacterial genomes: M. smegmatis (MSMEG), M. tuberculosis (Rv), M. avium subsp. paratuberculosis (MAP) and M. leprae (ML). Color code: red, high concensus (>90%); blue, low consensus (>50%). The central parts bordered in orange correspond to the predicted cytoplasmic portion and may be the region of selectivity of Gap proteins.
chapter 22, Proposed pathways for the biosynthesis of mycolactone A/B as deduced from the complete DNA sequence of pMUM001 and transposon mutagenesis ( Stinear et al., 2004 ). (A) Module and domain arrangement for synthesis of the core. (B) Module and domain arrangement for synthesis of the side chain. (C) Map of the pMUM001 plasmid showing the relative positions of the mls genes and accessory genes. (D) Key for domain function. Identical color indicates both same function and sequence (97% to 100% aa identity).