SGM Journals
MICROBIAL PATHOGENICITY

Paradoxical conservation of a set of three cellulose-targeting genes in Mycobacterium tuberculosis complex organisms

Felix Mba Medie, Iskandar Ben Salah, Michel Drancourt, Bernard Henrissat

  • 1Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes, UMR CNRS 6236, IRD 198, IFR 48, Faculté de Médecine, Université de la Méditerranée, Marseille, France
  • 2Architecture et Fonction des Macromolécules Biologiques, UMR 6098, IFR 48, CNRS, Université de la Méditerranée et Université de Provence, Marseille, France
  • Correspondence
    Bernard Henrissat
    bernard.henrissat{at}afmb.univ-mrs.fr

    • Microbiology 2010; 156(5):1468–1475 · https://doi.org/10.1099/mic.0.037812-0

    View at publisher PubMed

    Abstract

    The genome of the tuberculosis agent Mycobacterium tuberculosis encodes a putative cellulose-binding protein (CBD2), one candidate cellulase (Cel12), and one fully active cellulase (Cel6). This observation is puzzling, because cellulose is a major component of plant cell walls, whereas M. tuberculosis is a human pathogen without known contact with plants. In order to investigate the biological role of such cellulose-targeting genes in M. tuberculosis we report here the search for and transcription analysis of this set of genes in the genus Mycobacterium. An in silico search for cellulose-targeting orthologues found that only 2.5 % of the sequenced bacterial genomes encode the Cel6, Cel12 and CBD2 gene set simultaneously, including those of the M. tuberculosis complex (MTC) members. PCR amplification and sequencing further demonstrated the presence of these three genes in five non-sequenced MTC bacteria. Among mycobacteria, the combination of Cel6, Cel12 and CBD2 was unique to MTC members, with the exception of Mycobacterium bovis BCG Pasteur, which lacked CBD2. RT-PCR in M. tuberculosis H37Rv indicated that the three cellulose-targeting genes were transcribed into mRNA. The present work shows that MTC organisms are the sole mycobacteria among very few organisms to encode the three cellulose-targeting genes CBD2, Cel6 and Cel12. Our data point toward a unique, yet unknown, relationship with non-plant cellulose-producing hosts such as amoebae.

    • The GenBank/EMBL/DDBJ accession numbers for the Cel6, Cel12 and CBD2 nucleotide sequences of non-sequenced MTC members determined in this study are shown in Table 3⇑.

    • A supplementary table, showing a list of all sequenced bacteria in the CAZy database with one or more cellulose-targeting gene(s) in their genome, and two supplementary figures, showing a phylogenetic tree of Cel6 gene sequences in 17 bacteria using the neighbour-joining method with K2P distance correction model, and a phylogenetic tree of CBD2 gene sequences in 16 bacteria using the neighbour-joining method with K2P distance correction model, are available with the online version of this paper.

    Edited by: D. W. Ussery

    INTRODUCTION

    Cellulose, a linear polymer of glucose residues connected by β-1,4 linkages, makes up about half of the cell walls of plants and is both the most abundant polysaccharide and the major source of photosynthetically fixed carbon on Earth (Doi and Kosugi, 2004; Gilbert et al., 2008; Lynd et al., 2002). The natural resistance of cellulose to microbial digestion has probably guided plants to use it to build thick cell walls that can last up to several thousand years. Indeed, only a few saprophytic micro-organisms of fungal or bacterial origin have evolved cellulases, i.e. enzymes able to bind to cellulose and hydrolyse it to glucose (Lynd et al., 2002). A noteworthy feature of cellulases is that their structure is frequently modular, so that the catalytic domain carries one or more cellulose-binding modules.

    The recent discovery that Mycobacterium tuberculosis, the pathogen responsible for human tuberculosis, encodes a functional cellulase, Cel6 (Rv0062), was therefore totally unexpected (Varrot et al., 2005). The analysis of the M. tuberculosis genome shows that gene Rv1090 potentially encodes a second cellulase (Cel12) and that gene Rv1987 encodes a candidate cellulose-binding protein (CBD2). These three cellulose-targeting genes are completed by the presence of a putative β-glucosidase gene, Rv0186.

    Cellulases are found in the Carbohydrate-Active EnZymes database () (Cantarel et al., 2009) in nine sequence-based families of glycoside hydrolases (GHs), namely GH5, GH6, GH7, GH8, GH9, GH12, GH44, GH45 and GH48. Of these, the two mycobacterial cellulases (Cel6 and Cel12) belong to families GH6 and GH12, respectively.

    In plant cell walls, cellulose is found embedded in various complex polysaccharides (hemicelluloses and pectins) and lignin (Lynd et al., 2002; Varner & Lin, 1989), and the cellulolytic genes are usually accompanied by genes that encode a range of enzymes active against xylans and pectins.

    Orthologues of the M. tuberculosis genes Cel6, Cel12 and CBD2 have been found in all currently available mycobacterial genomes, including three of the seven species of the M. tuberculosis complex (MTC), i.e. M. tuberculosis H37Rv (AL123456), M. tuberculosis H37Ra (ABQ72837), M. tuberculosis CDC1551 (AE000516), M. tuberculosis F11 (ABR05454), M. tuberculosis KZN 1435 (CP001658), Mycobacterium bovis (BX248337) and M. bovis BCG Tokyo (AP010918). It is currently unknown whether other MTC members or non-tuberculous groups such as the Mycobacterium avium complex (MAC) or Mycobacterium abscessus complex also have the same putative cellulose-targeting genes. Likewise, there is no information regarding the transcription of these genes, hampering further investigation of the potential role of cellulose-targeting genes in Mycobacterium species. In an effort towards understanding the role of cellulose-targeting genes in Mycobacterium, we have searched for the presence of the Cel6, Cel12 and CBD2 gene set in representatives of the Mycobacterium complexes and have examined their transcription.

    METHODS

    Bioinformatics analyses.

    The presence or absence of the Cel6, Cel12 and CBD2 genes in 907 bacterial genomes was extracted from the CAZy database (), which is continuously updated in our laboratory based on day-to-day examination of the daily releases from GenBank (Cantarel et al., 2009). The survey was then completed by blast (Altschul et al., 1990) searches of additional genomes of the MTC, MAC and other non-tuberculosis Mycobacterium (NTM) organisms that are available as reference sequences at the NCBI (). These genomes included three MTC members, i.e. M. tuberculosis H37Rv (AL123456) (Cole et al., 1998), M. bovis (BX248333) (Garnier et al., 2003) and M. bovis BCG (Pasteur and Tokyo strains) (NC_008769, AP010918) (Brosch et al., 2007; Seki et al., 2009), two MAC members, i.e. M. avium 104 (CP00479) and M. avium subsp. paratuberculosis K-10 (AE016958) (Li et al., 2005), Mycobacterium leprae Br4923 (FM211192), M. leprae TN (AL450380), M. abscessus (CU458896), Mycobacterium smegmatis (CP000480), Mycobacterium ulcerans (CP000325) (Stinear et al., 2007, 2008), Mycobacterium marinum (CP000854) (Stinear et al., 2008), Mycobacterium vanbaalenii (CP000511), Mycobacterium sp. MCS (CP000384), Mycobacterium sp. JLS (CP000580), Mycobacterium sp. KMS (CP000518) and Mycobacterium gilvum (CP000656). The genomic organization of the Cel6, Cel12 and CBD2 genes was derived from inspection of the above-mentioned Mycobacterium genomes. Signal peptide prediction was done using Phobius ().

    PCR amplification and sequencing.

    In order to avoid artefacts, we controlled the quality of the genomic DNA and of any PCR primer pair used by including negative controls in all our PCR mixes. Also we took care to include all pathogenic Mycobacterium species, including the MTC organisms, the MAC organisms and the NTM organisms. Genomic DNA isolated from the MTC members (M. tuberculosis H37Rv CIP103471, M. bovis CIP105050, Mycobacterium africanum CIP105147T (type 1), M. bovis BCG vaccine strain type 105060, Mycobacterium microti CIP104256, Mycobacterium canettii CIP140060001T, Mycobacterium pinnipedii ATCC BAA-688 and Mycobacterium caprae CIP105776T), from the MAC members (M. avium subsp. hominissuis IWGMT49, Mycobacterium intracellulare CIP104243, Mycobacterium chimaera CIP107892T and Mycobacterium colombiense CIP108962), and from M. abscessus CIP104536, a Mycobacterium marinum clinical isolate and Mycobacterium fortuitum ATCC 49404 was used as a template for the PCR amplification and sequencing of M. tuberculosis Cel6, Cel12 and CBD2 gene orthologues. PCR primer pairs were designed after alignment of Mycobacterium genome sequences from GenBank (Table 1⇓) by using Primer3 Input 0.4.0 (). PCRs were carried out in a 2720 Thermal Cycler (Applied Biosystems) in a 50 μl final volume containing 25 μl H2O, 5 μl 10× buffer (Qiagen), 25 μM MgCl2, 100 μM of each dNTP, 5 μM each primer (Eurogentec), 2.5 U Taq DNA polymerase (Invitrogen) and 5 μl mycobacterial DNA, using the following program: 5 min at 95 °C, followed by 35 cycles consisting of 95 °C for 30 s, 60 °C for 30 s and 72 °C for 90 s, and a 10 min elongation step at 72 °C. For each experiment, a negative control consisting of PCR mix without target DNA was included. PCR products were resolved by electrophoresis on a 1.5 % (w/v) agarose gel and stained with ethidium bromide. PCR products were then purified by adding 50 μl distilled H2O to a purification plate (Millipore) that was agitated for 10 min. Forward and reverse sequencing mixtures contained 3 μl buffer (BigDye v1, Applied Biosystems), 10 μl distilled H2O and 1 μl of 3.2 pmol primer μl−1 in a final volume of 16 μl. The sequencing reaction comprised an initial denaturation step of 1 min at 95 °C, followed by 25 cycles of denaturation at 96 °C for 10 s, annealing at 50 °C for 5 s and elongation at 60 °C for 3 min. Sequencing products were purified using a Sephadex plate (Amersham Biosciences) that was centrifuged at 720 g for 3 min and deposited on a MicroAmp Optical 96-well reaction plate (Applied Biosystems). Sequencing electrophoresis was performed on a 3100 Genetic Analyzer (Applied Biosystems).

    Table 1.

    Number of Cel6, Cel12 and CBD2 genes in different bacterial species derived from the examination of their complete genome

    RT-PCR.

    Total RNA was extracted from 40 ml of exponential-phase cultures of Mycobacterium organisms. Mycobacteria were harvested by centrifugation at 1000 g for 5 min and the supernatant was discarded. The pellet was resuspended in 1 ml RLT buffer (Qiagen) containing 0.1 % 2-mercaptoethanol and transferred to a 2 ml screw-capped tube containing silica beads (IEPSA Medical Diagnostics) and the mixture was homogenized in a FastPrep FP120 instrument for 45 s at speed 6.5 (Qbiogene). After centrifugation at 8000 g for 1 min at room temperature, 700 μl ethanol was added and the solution was divided and applied into two RNeasy spin columns (Qiagen). Centrifugation was carried out at 8000 g for 15 s and the sample was then washed once with 700 μl RW1 buffer and twice with 500 μl RPE buffer (Qiagen). The RNA was eluted with 50 μl RNase-free water per column, and the samples were further treated with DNase I to eliminate any genomic DNA contamination. The RNA analysis was done by formaldehyde 1.5 % agarose gel electrophoresis. For RT-PCR, 10 μl RNA was mixed with 25 μl 2× Master Mix (Invitrogen), 1 μl (10 μM) of forward and reverse gene-specific primers (Table 2⇓) and 2.5 U Platinum Pfx DNA polymerase (Invitrogen), and made up to 50 μl with distilled water. The reaction was carried out in a 2720 Thermal Cycler under the following conditions: 45 °C for 30 min, 95 °C for 2 min, followed by 25 cycles consisting of 95 °C for 30 s, 58 °C for 45 s and 72 °C for 1 min, and a 5 min elongation step at 72 °C. RT-PCR products were resolved by 1.5 % agarose gel electrophoresis and stained with ethidium bromide.

    Table 2.

    PCR primers used in this study

    RESULTS

    Bioinformatics analyses

    Out of the 907 bacterial genomes analysed in the CAZy database (November 2009), only 2.5 % harbour the Cel6, Cel12 and CBD2 genes simultaneously (Table 1⇑). These 23 genomes include a group of 16 saprophytic organisms, and seven MTC members (Fig. 1⇓). It is worth mentioning that most plant cell wall degraders do not appear in this list, as they rarely harbour a Cel12 orthologue. Among the MTC members, only M. bovis BCG Pasteur strain 1173P2 lacks the CBD2 gene (Mahairas et al., 1996). The Cel12 gene was found to be present in all the MTC members and absent in all other Mycobacterium species under study. The structure of this gene is unusual, being composed of two overlapping fragments (CelA2a and CelA2b). Although the sequence similarity with bona fide Cel12 proteins such as that of Streptomyces lividans (Sulzenbacher et al., 1997) is significant, the MTC Cel12 proteins all appear to have a truncation which has eliminated the signal peptide and the 40 first amino acids. The significance and consequences of this truncation are not clear, since they do not affect the catalytic region of the protein (data not shown). Non-MTC mycobacteria, such as M. avium 104, M. avium subsp. paratuberculosis K-10, M. gilvum, M. marinum and M. vanbaalenii, encode only two of the three genes (Table 1⇑). A few other mycobacteria harbour only one of the three genes (for instance the strictly environmental M. smegmatis species, Mycobacterium sp. MCS, Mycobacterium sp. JLS, Mycobacterium sp. KMS and the Buruli ulcer agent M. ulcerans). Finally, the intracellular human pathogen M. leprae lacks the three cellulose-targeting genes altogether (Table 1⇑).

    Figure image not available in archive
    Fig. 1.

    Clustering of bacteria according to their content of cellulose-targeting genes. Only bacteria with one or more cellulose-targeting gene(s) are shown. Black arrows and numbers indicate the position of reference bacteria for identification of species in Supplementary Table S1.

    The disposition of the three cellulose-targeting genes within the genomes of various MTC members was similar (Fig. 2⇓). This similarity in organization was also true as for the size and the intergenic spacers between these genes. In M. avium 104 and M. avium subsp. paratuberculosis K10, the intergenic spacer between Cel6 and CDB2 was 2.4 Mb, similar to that of M. vanbaalenii, while in M. marinum, it was two times larger. The orientation of these genes in the genome was similar in all MTC, MAC and some NTM members, except for M. vanbaalenii and M. gilvum (Fig. 2⇓).

    Figure image not available in archive
    Fig. 2.

    Disposition of Cel6, Cel12 and CBD2 genes within the genome of MTC, MAC and NTM members. Arrows indicate the 5′→3′ orientation, and distances between adjacent genes are indicated (Mb).

    Cel6 and CBD2 putative gene products were found to possess a signal peptide across the Mycobacterium genus, while the Cel12 gene product lacked a signal peptide in all MTC members. We observed that the 31.27 kb region of the M. marinum genome that contains the Cel6 gene was deleted in M. ulcerans; a similar observation was made when aligning the Cel12 genomic region in M. tuberculosis H37Rv with the corresponding region in M. marinum and M. avium genomes, which lack this gene. Within MTC members, the Cel6 gene interspecies nucleotide sequence similarity ranged from 97.6 % between M. microti and M. canettii to 99.7 % between M. microti and M. caprae. We found no deletion or insertion within the Cel6 gene in the MTC reference strains. The CBD2 gene sequence of M. canettii exhibited 99.4 % similarity to the MTC reference M. tuberculosis H37Rv sequence, while M. caprae, M. microti, M. pinnipedii and M. africanum yielded 100 % CBD2 gene sequence similarity to the MTC reference sequence; M. africanum, M. caprae, M. microti, M. pinnipedii and M. canettii yielded identical sequences for the Cel12 gene. These data show that the three cellulose-targeting genes are highly conserved among MTC members.

    PCR amplification, sequencing and RT-PCR

    PCR amplification and sequencing showed that Cel6, Cel12 and CBD2 genes were present in the genome of M. canettii, M. africanum, M. pinnipedii, M. microti, M. caprae, M. tuberculosis H37Rv and M. bovis, whereas the M. bovis BCG Pasteur strain lacked the CBD2 gene. All original Cel6, Cel12 and CBD2 nucleotide sequences of non-sequenced MTC members have been deposited in GenBank (Table 3⇓). In MAC strains, Cel6 and CBD2 were detected in M. intracellulare, M. chimaera, M. colombiense and M. avium subsp. hominissuis, while in NTM strains, the Cel6 gene was present in M. smegmatis and M. fortuitum. Experimental PCR amplification and sequencing data obtained in MTC members correlated with in silico data using a blast search with the M. tuberculosis H37Rv nucleotide reference sequences of the Cel6, Cel12 and CBD2 genes (GenBank accession no. NC 000962) (Cole et al., 1998) as query sequences against the non-redundant database of NCBI.

    Table 3.

    GenBank accession numbers of original sequences determined in this study

    The absence of genomic DNA in our RNA samples was further confirmed by a lack of PCR amplification (data not shown). RT-PCR amplification using RNA isolated from M. tuberculosis H37Rv CIP103471 showed that Cel6, Cel12 and CBD2 were transcribed into mRNA. A single band of the expected (200 bp) size was detected (data not shown). The RT-PCR product was sequenced and the sequence yielded 100 % similarity to the M. tuberculosis H37Rv reference sequence. No PCR band was detected in DNase I-treated negative controls. This indicates that the Cel6, Cel12 and CBD2 genes present in the genome of M. tuberculosis H37Rv are transcribed into mRNA.

    DISCUSSION

    Using bioinformatics analyses we have observed the presence of a set of three cellulose-targeting genes in Mycobacterium species with a completely sequenced genome. The study was completed by PCR-based experiments to identify, amplify and determine the DNA and mRNA sequences of the three cellulose-targeting genes in 14 other Mycobacterium species. According to our database search, the set of three cellulose-targeting genes, Cel6, Cel12 and CBD2, is present in only 2.5 % of the bacteria with a complete genome sequence (Table 1⇑ and Supplementary Table S1). These bacteria include soil inhabitants and marine organisms in contact with decaying plants. Because cellulose almost never occurs as a pure component in plant cell walls, saprophytes have a complete spectrum of enzymes for all constituents of plant cell walls, and cellulolytic microbes usually produce a cohort of hemicellulase and pectinase enzymes along with their cellulases. In contrast, the MTC members stand out because the putative cellulose-targeting genes are not accompanied by other plant cell wall-digesting enzymes. The fact that in MTC members the cellulolytic machinery is so focused suggests that the plant cell wall may not be the actual target for cellulolytic activity in MTC organisms.

    The lack of Cel6, Cel12 and CBD2 genes in the M. leprae genome is congruent with the severe genome reduction that accompanied the evolution of this species towards a strictly parasitic lifestyle (Cole et al., 2001). Five Mycobacterium species (Table 1⇑) encoded only one of the three cellulolytic-targeting genes, suggesting a limited activity towards cellulose, if any. The vast majority of Mycobacterium species (Table 1⇑ and Supplementary Table S1) encoded Cel6 and a putative CBD2 gene. The fact that the MTC organisms (mammal-associated species responsible for tuberculosis) were the sole Mycobacterium organisms to exhibit the complete set of Cel6, Cel12 and CBD2 genes was unexpected. M. bovis has been shown to be transmitted from bovine to bovine and from bovine to humans (Ashford et al., 2001) by the oral route, thus being potentially in close contact with cellulose-containing foodstuffs in the digestive tract of vertebrates such as mice (Boulahrouf et al., 1990) and pigs (Varel et al., 1984). Likewise, we have recently shown that living M. tuberculosis organisms can be detected in the stools of patients with pulmonary tuberculosis (El Khechine et al., 2009). M. bovis, however, is also efficiently transmitted directly by the aerosol route, being responsible for outbreaks of tuberculosis (Rodwell et al., 2008) and bovine tuberculosis (Wilkins et al., 2008). It has also been demonstrated that M. bovis can survive ingestion by amoebae, which suggests that protozoa could significantly enhance the survival of M. bovis in the soil and hence may be instrumental in the transmission of bovine tuberculosis (Taylor et al., 2003). Likewise, M. tuberculosis is mainly transmitted directly by the aerosol route and has no known environmental reservoir (Riley et al., 1995). However, the fact that these three cellulose-targeting genes are highly conserved in MTC organisms argues against the idea that these genes would be in a degenerate state.

    A recent report has identified the Legionella pneumophila strain 130b CelA gene, which has been shown to degrade carboxymethyl cellulose and microcrystalline cellulose (Pearce & Cianciotto, 2009). Because L. pneumophila resides in amoebae which produce cellulose, it has been suggested that CelA could be involved in intra-amoeba growth of Legionella. However, it has been observed that an L. pneumophila CelA-deficient mutant and the wild-type strain both exhibit comparable growth in amoebae trophozoites and macrophages and comparable persistence in the lungs, indicating that celA is not required for intracellular life nor for lung infection (Pearce & Cianciotto, 2009). Based on these data, it has been suggested that the CelA protein could promote the growth of this organism in natural aquatic environments that contain higher levels and/or different types of polysaccharides, including cellulose produced by plants, amoebae or other bacteria.

    Our observation that MTC organisms also possess cellulose-targeting genes in their genome is somewhat similar to this recent report. In M. avium 104, the M. tuberculosis 7.8 kb region that includes Cel12 was deleted, leaving a region of 2.1 kb covered by two hypothetical proteins and one spacer (Fig. 3⇓), neither of them matching the M. tuberculosis Cel12a and Cel12b genes, implying that these proteins were not Cel12 orthologues. This observation suggests that the Cel12 gene has been uniquely retained in the MTC strains from a Mycobacterium early ancestor. Phylogenetic trees based on Mycobacterium Cel6 and CBD2 sequences show that the eight MTC members cluster together (Supplementary Figs S1 and S2). This observation suggests that these genes are typical of MTC organisms and Mycobacterium and that no recent lateral gene transfer has occurred.

    Figure image not available in archive
    Fig. 3.

    Schematic diagram of the Cel12 genomic region in M. tuberculosis H37Rv and M. avium 104, which lack Cel12. The region in M. tuberculosis H37Rv contains six genes in tandem, including Cel12. The homologous region in M. avium 104 contains only two genes, MAV_1212 and MAR_1213, separated by a spacer of 246 bp, pointing to the complete lack of the M. tuberculosis Cel12 genome region in M. avium 104.

    The capacity of a gene to respond to internal/external cues can be determined by its ability to be transcribed into mRNA. We have studied the transcription of the Cel6, Cel12 and CBD2 genes, and all three were transcribed into mRNA. The transcription of Cel6 was predictable, as this gene has been cloned and the encoded protein is expressed in Escherichia coli in an active form (Varrot et al., 2005). Moreover, the Cel6 gene has been shown to be downregulated in an M. tuberculosis mutant lacking SigF (Geiman et al., 2004), indicating that the regulation of this gene is SigF-dependent. The Cel12 gene lacks a peptide signal and exhibits a potentially fragmented structure which could suggest gene decay. It has been shown that the second part of the Cel12 gene, namely CelA2b, is upregulated upon treatment of M. tuberculosis H37Rv with mefloquine (a derivative of 4-quinoline methanol) for 24 h (Danelishvili et al., 2005). This observation suggests that even though the M. tuberculosis Cel12 gene presents an unusual structure, it is implicated in the mechanisms of resistance to mefloquine in M. tuberculosis. Many cellulases share a common basic architecture that comprises a catalytic domain linked to a cellulose binding domain (CBD), which determines the efficiency of degradation of insoluble cellulose. In our study we found that M. tuberculosis CBD2 was transcribed into mRNA, and this gene also possesses a signal peptide, which certainly points to an extracellular role.

    Bioinformatics searches combined with PCR-based analyses have revealed that a set of three cellulose-targeting genes (Cel6, Cel12 and CBD2) is present in the genome of MTC members. The only exception is the M. bovis BCG Pasteur strain, in which the CBD2 gene has been lost along with several other portions of DNA. We have verified that the cellulose-targeting genes are transcribed into mRNA and it has been previously shown that the Cel6-encoded protein can be expressed as an active protein (Varrot et al., 2005). Preliminary observations from this group suggest that Cel12 and CBD2 can also be cloned and the proteins expressed in E. coli (F. Mba Medie and others, unpublished observations).

    It has been shown that examination of the complete genome of a fastidious micro-organism can help to design a suitable culture medium (Renesto et al., 2003). Here we show that analysis of the genome sequence can also reveal unexpected metabolic traits or abilities. Indeed, the presence of cellulose-targeting proteins reported here raises the question of a possible, yet unknown, environmental stage in MTC bacteria, perhaps as soil saprophytes or as hosts of cellulose-producing organisms such as amoebae, as suggested for M. bovis (Hagedorn et al., 2009; Taylor et al., 2003).

    Acknowledgments

    F. M. M. is the recipient of an Infectiopole Sud doctoral fellowship. The enthusiastic support of Didier Raoult is gratefully acknowledged.

    References

    1. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol 215, 403–410.
    2. Ashford, D. A., Whitney, E., Raghunathan, P. & Cosivi, O. (2001). Epidemiology of selected mycobacteria that infect humans and other animals. Rev Sci Tech 20, 325–337.
    3. Boulahrouf, A., Fonty, G. & Gouet, P. (1990). Establishment of cellulolytic bacteria in the digestive tract of conventionally reared young mice: effect of the dietary cellulose content in the adult. FEMS Microbiol Lett 57, 87–90.
    4. Brosch, R., Gordon, S. V., Garnier, T., Eiglmeier, K., Frigui, W., Valenti, P., Dos Santos, S., Duthoy, S., Lacroix, C. & other authors (2007). Genome plasticity of BCG and impact on vaccine efficacy. Proc Natl Acad Sci U S A 104, 5596–5601.
    5. Cantarel, B. L., Coutinho, P. M., Rancurel, C., Bernard, T., Lombard, V. & Henrissat, B. (2009). The Carbohydrate-Active EnZymes database (CAZy): an expert resource for Glycogenomics. Nucleic Acids Res 37, D233–D238.
    6. Cole, S. T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., Gordon, S. V., Eiglmeier, K., Gas, S. & other authors (1998). Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393, 537–544.
    7. Cole, S. T., Eiglmeier, K., Parkhill, J., James, K. D., Thomson, N. R., Wheeler, P. R., Honoré, N., Garnier, T., Churcher, C. & other authors (2001). Massive gene decay in the leprosy bacillus. Nature 409, 1007–1011.
    8. Danelishvili, L., Wu, M., Young, L. S. & Bermudez, L. E. (2005). Genomic approach to identifying the putative target of and mechanisms of resistance to mefloquine in mycobacteria. Antimicrob Agents Chemother 49, 3707–3714.
    9. Doi, R. H. & Kosugi, A. (2004). Cellulosomes: plant-cell-wall-degrading enzyme complexes. Nat Rev Microbiol 2, 541–551.
    10. El Khechine, A., Henry, M., Raoult, D. & Drancourt, M. (2009). Detection of Mycobacterium tuberculosis complex organisms in the stools of patients with pulmonary tuberculosis. Microbiology 155, 2384–2389.
    11. Garnier, T., Eiglmeier, K., Camus, J. C., Medina, N., Mansoor, H., Pryor, M., Duthoy, S., Grondin, S., Lacroix, C. & other authors (2003). The complete genome sequence of Mycobacterium bovis. Proc Natl Acad Sci U S A 100, 7877–7882.
    12. Geiman, D. E., Kaushal, D., Ko, C., Tyagi, S., Manabe, Y. C., Schroeder, B. G., Fleischmann, R. D., Morrison, N. E., Converse, P. J. & other authors (2004). Attenuation of late-stage disease in mice infected by the Mycobacterium tuberculosis mutant lacking the SigF alternate sigma factor and identification of SigF-dependent genes by microarray analysis. Infect Immun 72, 1733–1745.
    13. Gilbert, H. J., Stalbrand, H. & Brumer, H. (2008). How the walls come crumbling down: recent structural biochemistry of plant polysaccharide degradation. Curr Opin Plant Biol 11, 338–348.
    14. Hagedorn, M., Rohde, K. H., Russell, D. G. & Soldati, T. (2009). Infection by tubercular mycobacteria is spread by nonlytic ejection from their amoeba hosts. Science 323, 1729–1733.
    15. Li, L., Bannantine, J. P., Zhang, Q., Amonsin, A., May, B. J., Alt, D., Banerji, N., Kanjilal, S. & Kapur, V. (2005). The complete genome sequence of Mycobacterium avium subspecies paratuberculosis. Proc Natl Acad Sci U S A 102, 12344–12349.
    16. Lynd, L. R., Weimer, P. J., van Zyl, W. H. & Pretorius, I. S. (2002). Microbial cellulose utilization: fundamentals and biotechnology. Microbiol Mol Biol Rev 66, 506–577.
    17. Mahairas, G. G., Sabo, P. J., Hickey, M. J., Singh, D. C. & Stover, C. K. (1996). Molecular analysis of genetic differences between Mycobacterium bovis BCG and virulent M. bovis. J Bacteriol 178, 1274–1282.
    18. Pearce, M. M. & Cianciotto, N. P. (2009). Legionella pneumophila secretes an endoglucanase that belongs to the family-5 of glycosyl hydrolases and is dependent upon type II secretion. FEMS Microbiol Lett 300, 256–264.
    19. Renesto, P., Crapoulet, N., Ogata, H., La, S. B., Vestris, G., Claverie, J. M. & Raoult, D. (2003). Genome-based design of a cell-free culture medium for Tropheryma whipplei. Lancet 362, 447–449.
    20. Riley, R. L., Mills, C. C., Nyka, W., Weinstock, N., Storey, P. B., Sultan, L. U., Riley, M. C. & Wells, W. F. (1995). Aerial dissemination of pulmonary tuberculosis. A two-year study of contagion in a tuberculosis ward. 1959. Am J Epidemiol 142, 3–14.
    21. Rodwell, T. C., Moore, M., Moser, K. S., Brodine, S. K. & Strathdee, S. A. (2008). Tuberculosis from Mycobacterium bovis in binational communities, United States. Emerg Infect Dis 14, 909–916.
    22. Seki, M., Honda, I., Fujita, I., Yano, I., Yamamoto, S. & Koyama, A. (2009). Whole genome sequence analysis of Mycobacterium bovis bacillus Calmette-Guerin (BCG) Tokyo 172: a comparative study of BCG vaccine substrains. Vaccine 27, 1710–1716.
    23. Stinear, T. P., Seemann, T., Pidot, S., Frigui, W., Reysset, G., Garnier, T., Meurice, G., Simon, D., Bouchier, C. & other authors (2007). Reductive evolution and niche adaptation inferred from the genome of Mycobacterium ulcerans, the causative agent of Buruli ulcer. Genome Res 17, 192–200.
    24. Stinear, T. P., Seemann, T., Harrison, P. F., Jenkin, G. A., Davies, J. K., Johnson, P. D., Abdellah, Z., Arrowsmith, C., Chillingworth, T. & other authors (2008). Insights from the complete genome sequence of Mycobacterium marinum on the evolution of Mycobacterium tuberculosis. Genome Res 18, 729–741.
    25. Sulzenbacher, G., Shareck, F., Morosoli, R., Dupont, C. & Davies, G. J. (1997). The Streptomyces lividans family 12 endoglucanase: construction of the catalytic core, expression, and X-ray structure at 1.75 Å resolution. Biochemistry 36, 16032–16039.
    26. Taylor, S. J., Ahonen, L. J., de Leij, F. A. & Dale, J. W. (2003). Infection of Acanthamoeba castellanii with Mycobacterium bovis and M. bovis BCG and survival of M. bovis within the amoebae. Appl Environ Microbiol 69, 4316–4319.
    27. Varel, V. H., Fryda, S. J. & Robinson, I. M. (1984). Cellulolytic bacteria from pig large intestine. Appl Environ Microbiol 47, 219–221.
    28. Varner, J. E. & Lin, L. S. (1989). Plant cell wall architecture. Cell 56, 231–239.
    29. Varrot, A., Leydier, S., Pell, G., Macdonald, J. M., Stick, R. V., Henrissat, B., Gilbert, H. J. & Davies, G. J. (2005). Mycobacterium tuberculosis strains possess functional cellulases. J Biol Chem 280, 20181–20184.
    30. Wilkins, M. J., Meyerson, J., Bartlett, P. C., Spieldenner, S. L., Berry, D. E., Mosher, L. B., Kaneene, J. B., Robinson-Dunn, B., Stobierski, M. G. & Boulton, M. L. (2008). Human Mycobacterium bovis infection and bovine tuberculosis outbreak, Michigan, 1994–2007. Emerg Infect Dis 14, 657–660.