Figures
Abstract
Mycobacterium abscessus group subsp., such as M. massiliense, M. abscessus sensu stricto and M. bolletii, are an environmental organism found in soil, water and other ecological niches, and have been isolated from respiratory tract infection, skin and soft tissue infection, postoperative infection of cosmetic surgery. To determine the unique genetic feature of M. massiliense, we sequenced the complete genome of M. massiliense type strain JCM 15300 (corresponding to CCUG 48898). Comparative genomic analysis was performed among Mycobacterium spp. and among M. abscessus group subspp., showing that additional ß-oxidation-related genes and, notably, the mammalian cell entry (mce) operon were located on a genomic island, M. massiliense Genomic Island 1 (MmGI-1), in M. massiliense. In addition, putative anaerobic respiration system-related genes and additional mycolic acid cyclopropane synthetase-related genes were found uniquely in M. massiliense. Japanese isolates of M. massiliense also frequently possess the MmGI-1 (14/44, approximately 32%) and three unique conserved regions (26/44; approximately 60%, 34/44; approximately 77% and 40/44; approximately 91%), as well as isolates of other countries (Malaysia, France, United Kingdom and United States). The well-conserved genomic island MmGI-1 may play an important role in high growth potential with additional lipid metabolism, extra factors for survival in the environment or synthesis of complex membrane-associated lipids. ORFs on MmGI-1 showed similarities to ORFs of phylogenetically distant M. avium complex (MAC), suggesting that horizontal gene transfer or genetic recombination events might have occurred within MmGI-1 among M. massiliense and MAC.
Citation: Sekizuka T, Kai M, Nakanaga K, Nakata N, Kazumi Y, Maeda S, et al. (2014) Complete Genome Sequence and Comparative Genomic Analysis of Mycobacterium massiliense JCM 15300 in the Mycobacterium abscessus Group Reveal a Conserved Genomic Island MmGI-1 Related to Putative Lipid Metabolism. PLoS ONE 9(12): e114848. https://doi.org/10.1371/journal.pone.0114848
Editor: Jean Louis Herrmann, Hopital Raymond Poincare - Universite Versailles St. Quentin, France
Received: February 27, 2014; Accepted: November 14, 2014; Published: December 11, 2014
Copyright: © 2014 Sekizuka et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported in part by a Grant-in-Aid (25461178) for Scientific Research (C) from the Japan Society for the Promotion of Science (http://www.jsps.go.jp/english/index.html), by a grant from the Ohyama Health Foundation (http://www.disclo-koeki.org/10a/01044/index.html) and by a Grant-in-Aid (H25-Shinko-Ippan-015) from the Ministry of Health, Labour, and Welfare, Japan (http://www.jsps.go.jp/english/e-grants/grants.html). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Yoshihiko Hoshino is a PLOS ONE Editorial Board member. This does not alter the authors' adherence to all PLOS ONE policies on sharing data and materials, as detailed online in the guide for authors.
Introduction
Nontuberculous mycobacteria (NTM) are classified into slowly growing mycobacterium (SGM) and rapidly growing mycobacterium (RGM) species; some of these bacteria cause pulmonary diseases [1]. Among RGM, the Mycobacterium abscessus group has been shown to be an emerging respiratory pathogen in cystic fibrosis, non-cystic-fibrosis bronchiectasis and chronic obstructive pulmonary disease [2], [3], [4], [5], [6], and is also an environmental organism found in soil, water and other ecological niches [7], [8]. The M. abscessus group consists of three subspecies, M. abscessus subsp. abscessus (M. abscessus sensu stricto), M. abscessus subsp. massiliense (M. massiliense) and M. abscessus subsp. bolletii (M. bolletii) [9], [10]. The three subspecies can generally be distinguished by phylogenetic analysis of the housekeeping gene, rpoB, and the macrolide resistance-related gene, erythromycin ribosome methyltransferase (erm) (41). Bryant et al. and Nakanaga et al. have recently reported more detailed classification methods, including, respectively, a whole-genome single nucleotide polymorphism (SNP) approach and a multiplex PCR method using insertion/deletion regions identified by whole-genome sequencing alignment analysis [4], [11]. Several subcutaneous infections following surgery, other medical treatments or traumatic injury have recently been found to be caused by M. massiliense [12], [13], [14], [15]. It was also recently reported that M. massiliense caused cutaneous infections that could not be attributed to a prior invasive procedure [16]. Phylogenetic analyses of the M. abscessus group have been performed, putative virulence factors of M. abscessus sensu stricto have been identified and studied, and the comparative whole-genome analysis of M. abscessus group isolated from patients of wide geographical origin have been performed [4], [17], [18], [19]; however, a detailed comparative analysis of M. abscessus group subspp. to determine M. massiliense unique genetic feature is lacking. Thus, in the current study, we sequenced the complete M. massiliense JCM 15300 (CCUG 48898) genome and compared it with that of M. abscessus group subspecies.
Results and Discussion
Genomic sequence of M. massiliense JCM 15300
The complete chromosomal sequence of M. massiliense JCM 15300 was obtained by de novo assembly of short reads followed by gap-closing using directed PCR. The genome consisted of 4,978,382 base pairs (bps) with a GC content of 64.1% and 4,950 predicted coding sequences (CDSs), 46 tRNA genes, one rRNA operon and two prophages (Fig. 1A). The chromosomal sequence corresponded to the predicted restriction fragment profiles obtained by PFGE analysis (data not shown). A draft genomic sequence of CCUG 48898 corresponding to JCM 15300 has been previously deposited in GenBank (NZ_AHAR01000000) by another research group. Thus, we performed a comparative pair-wise sequence alignment, revealing highly conserved synteny to the complete genomic sequence of JCM 15300 (S1 Figure and S1 Table). There were 188 mutations within 33 CDSs and 7 non-coding sites, suggesting that the differences between type strains may be due to frequent passaging and cultivation in various laboratories and bioresource centers. JCM15300 strain is smooth colony morphotype, and then there are no nonsense or frameshift mutations and in mps1-mps2-gap (MMASJCM_4183, MMASJCM_4184 and MMASJCM_4185) or mmpl4b (MMASJCM_4202) (data not shown), these data is consistent with a previous report [20].
A. BLAST atlas of M. massiliense JCM 15300. The coding region of strain JCM 15300 was aligned against those of 14 other Mycobacterium genomes using BLASTP. The results are displayed as colored circles with increasing color intensity signifying increased similarity. It was estimated that the number of conserved proteins was 1,516 among all 14 Mycobacterium genomes. B. Box plot of identity percentage of conserved proteins between M. massiliense JCM 15300 and 14 other Mycobacterium spp. The top of each box in the box plot indicates the 75th percentile, the bottom of each box indicates the 25th percentile and the center bar represents the median. C. Neighbor-joining phylogenetic tree based on 16S rRNA gene sequencing of Mycobacterium with 1,000-fold bootstrapping. Scale bar indicates number of substitutions per site. The number at each branch node represents the bootstrapping value. Nocardia abscessus JCM 6043 (GenBank: AF430018) and Gordonia aichiensis DSM43978T (X80633) were used as outgroups.
Comparative genomic analysis within the Mycobacterium genus
To characterize the genomic features of M. massiliense JCM 15300, a BLAST atlas analysis was performed; corresponding orthologs in complete and draft genomic sequences of other Mycobacterium spp. were compared with those of M. massiliense JCM 15300 as a reference (M. bolletii BD is a draft genomic sequence, but it is closely related to M. massiliense) (Fig. 1A). The BLAST atlas identified the conserved proteins in the core genome, which was represented by 973 CDSs (19.7%) shared among all 15 Mycobacterium spp. genomes. M. massiliense JCM 15300 was highly similar to M. abscessus ATCC 19977 and M. bolletii BD in the M. abscessus group (Fig. 1B). In contrast, M. massiliense JCM 15300 showed a low similarity (∼73% of mean identity) to SGM and other RGM (Fig. 1B). The 16S rRNA phylogenetic analysis suggested complete identity of M. massiliense JCM 15300 to M. abscessus ATCC 19977 and M. bolletii BD (Fig. 1C). These results indicate that M. massiliense is difficult to distinguish among the three M. abscessus subspecies using 16S rRNA gene phylogeny and that the three subspecies belong to the M. abscessus group as suggested by many reports.
The above analysis demonstrated that there were several highly variable gene clusters and notable differences in GC content (64.1%) among the 14 Mycobacterium spp. One prophage, located in the region from 1,816 to 1,880 kbs, had a lower GC content (59.64%) and partially shared some conserved CDSs with M. abscessus ATCC 19977 (gray bar in the lower right of Fig. 1A). The average GC content of all 14 Mycobacterium spp. and 620 mycobacteriophages [21] was approximately 66% and 64%, respectively, suggesting that the low-GC content prophage was recently acquired. In contrast, another prophage, located in the region from 3,964,186 to 4,013,302 bps, had an average GC content (64%), indicating that it could be specific to M. massiliense JCM 15300 (gray bar in the upper left of Fig. 1A).
Intriguingly, a notable genomic island from 946,561 to 1,057,603 bps, designated M. massiliense genomic island 1 (MmGI-1; indicated by the blue bar in the upper right of Fig. 1A), appeared to be conserved among M. massiliense JCM 15300, M. bolletii BD and M. avium 104. The genomic island contained gene clusters associated with lipid metabolism and lipid-related transporters (Fig. 2 and Table 1). ß-oxidation-related genes were also identified, such as long-chain fatty acid-CoA ligase (MMASJCM_1018, MMASJCM_1019, MMASJCM_1028), acyl-CoA dehydrogenase (MMASJCM_1023, MMASJCM_1030, MMASJCM_1035, MMASJCM_1038), enoyl-CoA hydratase (MMASJCM_1008, MMASJCM_1009, MMASJCM_1010, MMASJCM_1022), 3-hydroxyacyl-CoA dehydrogenase (MMASJCM_1006, MMASJCM_1034), acyl-CoA thiolase (MMASJCM_1016, MMASJCM_1036) and acetyl-CoA acetyltransferase (MMASJCM_1014) (Table 1).
Phylogenetic tree based on 203,267 core genome SNPs in the whole-genome-sequenced M. abscessus group by the maximum-likelihood method with 1,000-fold bootstrapping. The scale indicates that a branch with a length of 0.1 is 10 times as long as one that would show a 1% difference between the nucleotide sequences at the beginning and end of the branch. The number at each branch node represents the bootstrapping value. The ORFs of M. massiliense strain JCM 15300 were aligned against the genomic sequences of 56 other M. abscessus group strains and M. avium 104 using TBLASTN (E-value cutoff, 1.00E-10; identity cutoff, 70%). A heatmap was constructed from amino acid identity.
An ortholog of the mammalian cell entry (mce) operon (MMASJCM_0985 to _0992) was found in the genomic island (Fig. 2 and Table 1). The mce operon of Actinomycetales species has been suggested to encode a subfamily of ATP-binding cassette (ABC) transporters that have a possible role in remodeling the cell envelope [22] and entry of the pathogen into non-phagocytic cells [23]. Although the function of the Mce protein family has not been clearly established, its members are believed to be membrane lipid transporters. For example, it has been demonstrated that the mce4 operon is required for cholesterol utilization and uptake by M. tuberculosis [24] and M. smegmatis [25]. M. massiliense JCM 15300 contained 8 loci from the mce operon, and one mce operon on the MmGI-1 genomic island demonstrated approximately 99% similarity to that of M. bolletii BD and approximately 80% similarity to that of M. avium 104.
To characterize a provenance of MmGI-1 regions, the regions were subjected to BLASTN/BLASTP search against NCBI nt/nr databases excluding M. abscesses group sequences. Although the nucleotide search with BLASTN did not show notable homology to MmGI-1 region, the protein search with BLASTP showed that 105 ORFs on MmGI-1 showed significant similarity to ORFs of Actinomycetales with 32 to 95% identity. Of 105 ORFs, forty-two ORFs showed similarities to ORFs of phylogenetically distant M. avium complex (MAC) (Fig. 3), suggesting that the MmGI-1 region might have been acquired through horizontal gene transfer or genetic recombination events with MAC.
Phylogenetic tree based on the 16S rRNA was constructed by Neighbor-joining method with 1,000-fold bootstrapping. Scale bar indicates number of substitutions per site. Species of black characters indicate that complete or draft genome sequences have been deposited at DDBJ/EMBL/GenBank. M. abscessus group is labeled by a yellow box. The number of BLASTP top hit orthologous genes against MmGI-1 genes are shown with a right bar chart.
Using 55 draft genomic sequences from the M. abscessus group [17] and one complete genomic sequence from M. massiliense JCM 15300, variation among the genomic islands was investigated. The phylogeny of M. abscessus group strains was further characterized by identifying 203,267 SNPs in the commonly shared genomic sequence (Fig. 2). The SNP phylogenetic analysis identified three clusters (i.e., massiliense, bolletii and abscessus clusters) from the M. abscessus group, consistent with a previous report [17]. Phylogenetic and heatmap analyses suggested that MmGI-1 was partially shared among M. massiliense-related strains (Fig. 2). Notably, the ß-oxidation-related loci (MMASJCM_0982 to _1042) were also well conserved in M. bolletii BD and M24. These additional lipid-related metabolic genes may be important for high growth potential with additional lipid metabolism such as putative ß-oxidation pathway, extra factors for survival in the environment (as suggested by the presence of MCE family protein) or synthesis of complex membrane-associated lipids (as suggested by the presence of a long-chain-fatty-acid-CoA ligase).
Comparative genomic analysis within the M. abscessus group
To characterize the genomes of the previously described three clusters, we performed further comparative and BLAST atlas analyses based on the nucleotide sequences of two complete genomes and the predicted amino acid sequences of CDSs, respectively (S2 Figure and S2 and S3 Table), and then also performed pan-genomic analysis with 30 M. massiliense, 2 M, bolletii and 25 M. abscessus genome sequences because of a validation (S3 Figure). The pan-genomic analysis data is consistent with a previous report [19]. The comparative analysis yielded the following four results: i) as a massiliense cluster-specific feature, there were six unique regions (†1–6 in S2 Figure and Table 2) that contained an average GC content of 64%; ii) as a JCM 15300-specific feature, there were 10 unique regions (• in S2 Figure and S2 Table) that had relatively low GC content; iii) the MmGI-1 genomic island (Fig. 3 and ¶ in S2 Figure) was shared with M. bolletii and showed partial similarity to M. avium 104; iv) there were two common deletions (†7–8 in S2 Figure and S3 Table) in the massiliense cluster and one conserved region in the abscessus group (§ in S2 Figure and S3 Table).
In addition to the MmGI-1 genomic island described above, the massiliense cluster contained three notable conserved loci: i) a molybdopterin oxidoreductase (Fig. 2, Fig. 4A and Table 2); ii) universal stress proteins, an alcohol dehydrogenase and a xylulose-5-phosphate phosphoketolase (Fig. 2, Fig. 4B and Table 2); iii) a cyclopropane fatty acyl-phospholipid synthase and an S-adenosyl-L-methionine-dependent methyltransferase (Fig. 2, Fig. 4C and Table 2). In contrast to MmGI-1, these three regions were well conserved within the massiliense cluster.
GenBank accession numbers are given in parentheses. The orange arrows indicate the unique genes in the massiliense cluster. BLASTN match scores less than 200 are not shown.
Choo et al. previously reported that a high proportion of accessory strain-specific genes indicating an open, non-conservative pan-genome structure, and clear evidence of rapid phage-mediated evolution [19]. In fact, specific genes in M. massiliense JCM15300 contained phage-related genes, i.e. putative prophage integrase (S2 Table). On the other hand, in adjacent gene loci of three conserved regions, i.e. MMASJCM-2099..2100, MMASJCM-2507..2524 and MMASJCM-4337..4346, there are no phage-related genes (Fig. 4 and Table 2). These data suggest that these conserved regions might be core-genome regions in ancestral M. abscessus group, and then have been deleted from genomes of M. abscessus and M. bolletii.
Prevalence of MmGI-1 and massiliense cluster unique regions in Japanese M. massiliense and M. abscessus isolates
We examined the prevalence of MmGI-1 and three massiliense cluster unique regions in Japanese M. massiliense and M. abscessus isolates using conventional PCR methods (S4 Table), because of in silico analysis using only isolates of Malaysia, France, United Kingdom and United States. The ratio of MmGI-1 positive M. massiliense and M. abscessus was 31.8% (14/44) and 1.4% (1/70), respectively (Fig. 5A and S5 Table). Applying Fisher's exact test, the proportion of MmGI-1 positive M. massiliense is significantly higher than that of M. abscessus (P = 0.0001). M. massiliense frequently possesses three massiliense cluster unique regions in not only Japanese but also other countries (Malaysia, France and United States) isolates (Fig. 5A and S5 Table), suggesting that MmGI-1 and the massiliense cluster unique regions are highly conserved in M. massiliense isolated from various countries.
A bar chart showing the prevalence of MmGI-1 and three massiliense cluster unique regions in Japanese M. massiliense and M. abscessus isolates (A). The curves represent in vitro growth (OD at 530 nm) over a period of 21 days at 37°C in aerobic (B) and microaerobic (C) conditions. Data represent the means ± SE from 6 MmGI-1 positive M. massiliense, 8 MmGI-1 negative M. massiliense and 12 M. abscessus isolates. M. mas and M. abs shows M. massiliense and M. abscessus, respectively. Key: +, positive; -, negative. * P<0.05; ** P<0.01 (Student's t-test).
Growth ability of MmGI-1 positive M. massiliense
The massiliense cluster contained a conserved molybdopterin oxidoreductase as described above, and an ortholog was also identified in the strictly anaerobic bacterium, Desulfitobacterium hafniense. It has been reported that molybdopterin oxidoreductase may provide the ability for anaerobic energy metabolism [26]. The xylulose-5-phosphate phosphoketolase may play a role in heterolactic fermentation in anaerobic heterolactic acid bacteria, including Lactobacillus and Leuconostoc organisms [27]. Moreover, the universal stress protein in Pseudomonas aeruginosa has been reported to have a crucial role in survival under anaerobic conditions [28]. These studies suggest that M. massiliense may grow or survive under anaerobic or hypoxic conditions. Indeed, the oxygen partial pressure in various tissues is approximately 20–50 mm Hg (3–7% oxygen) [29], [30], [31], [32]. To determine growth ability under hypoxic conditions, 27 smooth colony morphology isolates (12 M. abscessus, 8 MmGI-1 positive M. massiliense and 7 MmGI-1 negative M. massiliense isolates) were subjected to aerobic and microaerobic (approximately 6% O2) conditions (Fig. 5B and 5C), because the aggregation of rough colony morphology isolates were hard to measure the degree of turbidity in the broth culture. In aerobic condition, MmGI-1 positive M. massiliense isolates show well growth than MmGI-1 negative isolates including M. abscessus (Fig. 5B). On the other hand, in microaerobic condition, the growth didn't show significant differences between M. massiliense and M. abscessus (Fig. 5C). MMASJCM-2099..2100 and MMASJCM-2057..2524 regions highly conserved in M. massiliense isolated from Japan, Malaysia, France, United Kingdom and United States, as well as MmGI-1. Although functions of these regions are still unclear, the importance of MmGI-1 might be supported by the existence on these conserved regions in M. massiliense, and MmGI-1 might relate to high growth potential with additional lipid metabolism such as putative ß-oxidation pathway.
Phylogenetic analysis of mycolic acid synthase-related genes
The comparative genomic analysis indicated that M. massiliense including Japanese isolates possessed two extra CDSs that are possibly involved in the cyclopropanation of mycolic acid. A cyclopropane fatty acyl-phospholipid synthase (MMASJCM_4340) and an S-adenosyl-L-methionine-dependent methyltransferase (MMASJCM_4343) were detected only in the massiliense cluster (Fig. 4C). Both putative proteins encoded by these CDSs possessed the mycolic acid cyclopropane synthetase (CMAS) domain (pfam02353). Mycobacterium spp. possess 3 to 10 paralogs with a CMAS domain; for example, CmaA (cyclopropane mycolic acid synthase) and MmaA (methyl mycolic acid synthase) have been well characterized [33]. A phylogenetic analysis of CMAS domain-related proteins has indicated that one of the two extra proteins, MMASJCM_4340, is orthologous to MSMEG_1351 of M. smegmatis and MycrhN_0769/MycrhN_3064 of M. rhodesiae (S4 Figure). The other protein, MMASJCM_4343, is orthologous to UfaA1 (cyclopropane fatty acid synthase), which is present in a part of RGM and SGM species. The function of UfaA1 in mycolate biosynthesis is not clear [34]. The massiliense cluster has two unique mycolic acid synthesis-associated proteins that are not present in the abscessus or bolletii clusters.
Conclusions
The M. abscessus group is classified as RGM species and consists of three closely related organisms, M. abscessus, M. bolletii and M. massiliense. A comparative analysis based on three clusters in the M. abscessus group revealed that a genomic island MmGI-1 of M. massiliense may be involved in high growth potential with additional lipid metabolism such as putative ß-oxidation pathway. Moreover, MmGI-1 is conserved in Actinomycetales, especially Mycobacterium, and horizontal gene transfer or genetic recombination events might have occurred within MmGI-1 among M. massiliense and MAC. Although M. abscessus subspp. is an environmental organism found in soil, water and other ecological niches, the difference of detail ecological niches is still unclear among subspecies-level. Our data suggests that the massiliense cluster unique regions including MmGI-1 might be linked to differences in ecological niches, such as lipid rich environment, of M. massiliense and M. abscessus. Further studies are required to understand the specific genetic features identified in this study.
Materials and Methods
Bacterial strains
We sequenced Mycobacterium massiliense type strain JCM 15300 (CCUG 48898), which was originally isolated from the sputum of a 50-year-old woman with an 8-year history of bronchiectasis and hemoptysis [35]. This strain was obtained from the Japan Collection of Microorganisms at the Riken BioResource Center (BRC-JCM; Saitama, Japan) on September 18, 2009.
Short-read DNA sequencing
An M. massiliense strain DNA library (insert size of ∼600 bp) was prepared using the Nextera DNA Sample Prep Kit (Illumina-compatible) (EPICENTRE Biotechnologies, Madison, WI). DNA clusters were generated on a slide using the Cluster Generation Kit (ver. 4) on an Illumina Cluster Station (Illumina, San Diego, CA), according to the manufacturer's instructions. A paired-end sequencing run for 83 mers was performed using an Illumina Genome Analyzer IIx (GA IIx) with the TruSeq SBS Kit v5. Fluorescent images were analyzed using the Illumina RTA1.8/SCS2.8 base-calling pipeline to obtain FASTQ-formatted sequence data.
De novo assembly of short DNA reads and gap-closing
Prior to de novo assembly, the obtained 80-mer reads were assembled using ABySS-pe v1.2.5 [36] with the following parameters: k60, n60, c68.4, t10, e10 and q20. Predicted gaps were amplified with specific PCR primer pairs followed by Sanger DNA sequencing with the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA).
Validation of gap closing and sequencing errors by short-read mapping
To determine whether mis-assembled sequences and incorrect gap-closing remained after reference-assisted gap-closing, 40-mer short reads were aligned to the tentative complete chromosomal DNA sequence using Maq software (ver. 0.7.1) with the easyrun Perl command [37]. We then performed a read alignment to validate possible errors using the MapView graphical alignment viewer [38].
Annotation
Gene prediction was performed for the complete genomic sequence with the RAST annotation server [39], followed by InterProScan [40] search and BLASTP search using nr database for validation. Genomic information, such as nucleotide variations and circular representations, was analyzed with gview software [41].
Pairwise alignment of chromosomal sequences
Pairwise alignment was performed by BLASTN and TBLASTN homology searches [42] followed by visualization of the aligned images with the ACT [43] or EMBOSS dottup program [44].
BLAST atlas
A BLAST atlas was generated by a BLASTP homology search [42] using the gview program [41]. The atlas displays BLASTP comparison results. The visualized area shows that the length of similar genes covers at least 80% between M. massiliense JCM 15300 and other Mycobacterium spp.
SNP analysis
To construct simulated paired-end reads from the available genomic sequences of M. abscessus group strains, SimSeq software [45] was used with “SimSeq.jar” and “SamToFastq.jar” commands with the following default parameter modifications: number of pairs of reads, “—read_number 2000000”; mean library insert size, “—insert_size 150”; and paired-end reads length of 120 mer, “−1 120 −2 120”. These parameters indicated that 4 million hypothetical 120-mer reads were generated without mutations or indels from the genomic sequences used for SNP identification. To generate short-read mapping data of all M. abscessus group strains compared with the reference chromosomal sequence of M. massiliense JCM 15300, bwasw [46] and samtools [47] software was used with the default parameters. All SNPs were extracted by VarScan v2.3.4 [48] with the default parameters. All SNPs were concatenated to generate a pseudo sequence for phylogenetic analysis. The DNA maximum-likelihood program (RAxML v7.25) [49] was used for phylogenetic analysis with 1,000-fold bootstrapping. FigTree v. 1.2.3 software was used to display the generated tree.
Phylogenetic analysis
Nucleotide and amino acid sequences were aligned with mafft v6.86 [50] followed by phylogenetic analysis using the neighbor-joining method or maximum-likelihood method with 1,000-fold bootstrapping in clustalW2 [51] or RAxML v7.25 software [49]. FigTree v. 1.2.3 software was used to display the generated tree.
PCR amplification
The PCR mixture contained approximately 1 ng of template DNA, 1× PrimeSTAR GXL Buffer (Takara Biochem. Shiga, Japan), 200 µM of each dNTP, 200 nM of each primer, and a total of 2.5 unit of PrimeSTAR GXL DNA polymerase (Takara Biochem.). The primer sequences for PCR amplification are shown in S4 Table. PCR was performed in 25 µl volumes under the following conditions: at 98°C for 20 sec followed by 30 cycles at 98°C for 15 sec, 65°C for 15 sec and 68°C for 1 min (for below 1.5 kb amplicons) or 5 min (for over 1.5 kb amplicons). Amplified PCR products were electrophoresed in 1.0% (w/v) agarose gel at 100 V and detected by staining with GelRed (Biotium Inc. Hayward, CA).
Bacterial culture
The M. abscessus and M. massiliense type strains were cultured at 37°C in Middlebrook 7H9 broth (Difco) supplemented with 10% OADC (BD) and 0.05% Tween 80 under aerobic or microaerobic (6% aerobic O2 tension) conditions with AnaeroPack (Mitsubishi Gas Chemical Company, Inc., Tokyo, Japan). Growth was monitored by removing aliquots at the indicated time points and measuring the OD at 530 nm.
Statistical analysis
The statistical test between MmGI-1 positive M. massiliense and M. abscessus was calculated by Fisher's Exact Test. Data of bacterial culture are expressed as mean ± standard error (SE) from 7 MmGI-1 positive M. massiliense, 8 MmGI-1 negative M. massiliense and 12 M. abscessus isolates. Statistical analysis was performed using the student's t-test. The t-test was used to investigate whether the means of two groups are statistically different from each other. Differences were considered significant with a p-value of <0.05 and 0.01.
Supporting Information
S1 Figure.
Comparative analysis between the complete genomic sequence of the M. massiliense JCM 15300 strain and draft genomic sequences of M. massiliense CCUG 48898. The upper dot plot represents synteny between JCM 15300 and CCUG 48898, and the yellow vertical bars indicate gap regions in the draft genome of CCUG 48898. The bottom table shows gaps between contigs in CCUG 48898.
https://doi.org/10.1371/journal.pone.0114848.s001
(TIF)
S2 Figure.
Genomic comparison and BLAST atlas of 3 clusters in the M. abscessus group. Comparative analysis of M. massiliense JCM 15300 and M. abscessus ATCC 19977 using a BLASTN homology search visualized by the ACT program (middle) and a BLAST atlas of M. massiliense JCM 15300 and M. abscessus ATCC 19977. In the comparative analysis, the red and blue bars between chromosomal DNA sequences represent nucleotide matches in the forward and reverse directions, respectively. BLASTN match scores less than 999 are not shown. In the BLAST atlas, the coding regions of JCM 15300 or ATCC 19977 were aligned against those of other M. abscessus group strains using BLASTP, and the results are displayed as colored bars (as in Fig. 1A). The three yellow boxes represent prophages on each chromosome. Specific features are represented by characters: †, unique region in the massiliense cluster; •, unique region in JCM 15300; §, unique region in the abscessus cluster; ¶, MmGI-1 (also see blue bars in Fig. 1A).
https://doi.org/10.1371/journal.pone.0114848.s002
(TIF)
S3 Figure.
Visualization for M. abscessus group pan-genomes and core genomes. A. Curve for pan-genomes and core genomes of M. abscessus group. The box plots indicate the pan- or core genome size for each genome comparison. The median values were connected to represent the relationship between genome number and gene cluster number. B. Curve for the new gene cluster number observed with every increase in the number of M. abscessus group genomes.
https://doi.org/10.1371/journal.pone.0114848.s003
(TIF)
S4 Figure.
Phylogenetic tree of mycolic acid cyclopropane synthetase domain (CMAS, pfam02353) proteins in Mycobacterium using the maximum-likelihood method with 1,000-fold bootstrapping. The scale indicates that a branch length of 0.3 is 30 times as long as one that would show a 1% difference between the amino acid sequences at the beginning and end of the branch. The number at each branch node represents the bootstrapping value. The proteins in red indicate proteins that are conserved only in the massiliense cluster.
https://doi.org/10.1371/journal.pone.0114848.s004
(TIF)
S1 Table.
Mutation sites in the complete genomic sequence of M. massiliense JCM 15300 compared with those in draft genomic sequences of M. massiliense CCUG 48898.
https://doi.org/10.1371/journal.pone.0114848.s005
(PDF)
S2 Table.
The unique gene loci in M. massiliense JCM15300.
https://doi.org/10.1371/journal.pone.0114848.s006
(PDF)
S3 Table.
The deleted genes of massiliense and bolletii clusters among M. abscessus group.
https://doi.org/10.1371/journal.pone.0114848.s007
(PDF)
S4 Table.
Oligonucleotide primer sequences used in PCR assays and the judging method for presence of MmGI-1 and other M. massiliense unique regions.
https://doi.org/10.1371/journal.pone.0114848.s008
(PDF)
S5 Table.
Isolates analyzed in the present study and results of conventional PCR based detection against MmGI-1 and other M. massiliense unique regions.
https://doi.org/10.1371/journal.pone.0114848.s009
(PDF)
Author Contributions
Conceived and designed the experiments: TS M. Kai YH M. Kuroda. Performed the experiments: TS M. Kai KN NN YK SM YH M. Kuroda. Analyzed the data: TS M. Kai M. Kuroda. Contributed reagents/materials/analysis tools: TS M. Kai MM YH M. Kuroda. Wrote the paper: TS M. Kuroda. Performed genomic sequencing: TS M. Kai M. Kuroda.
References
- 1. Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, et al. (2007) An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. American journal of respiratory and critical care medicine 175:367–416.
- 2. Brown-Elliott BA, Nash KA, Wallace RJ Jr (2012) Antimicrobial susceptibility testing, drug resistance mechanisms, and therapy of infections with nontuberculous mycobacteria. Clinical microbiology reviews 25:545–582.
- 3. Olivier KN, Weber DJ, Wallace RJ Jr, Faiz AR, Lee JH, et al. (2003) Nontuberculous mycobacteria. I: multicenter prevalence study in cystic fibrosis. American journal of respiratory and critical care medicine 167:828–834.
- 4. Bryant JM, Grogono DM, Greaves D, Foweraker J, Roddick I, et al. (2013) Whole-genome sequencing to identify transmission of Mycobacterium abscessus between patients with cystic fibrosis: a retrospective cohort study. Lancet 381:1551–1560.
- 5. Iseman MD, Marras TK (2008) The importance of nontuberculous mycobacterial lung disease. American journal of respiratory and critical care medicine 178:999–1000.
- 6. Chan ED, Bai X, Kartalija M, Orme IM, Ordway DJ (2010) Host immune response to rapidly growing mycobacteria, an emerging cause of chronic lung disease. American journal of respiratory cell and molecular biology 43:387–393.
- 7. Falkinham JO 3rd (1996) Epidemiology of infection by nontuberculous mycobacteria. Clinical microbiology reviews 9:177–215.
- 8. Primm TP, Lucero CA, Falkinham JO 3rd (2004) Health impacts of environmental mycobacteria. Clinical microbiology reviews 17:98–106.
- 9. Bastian S, Veziris N, Roux AL, Brossier F, Gaillard JL, et al. (2011) Assessment of clarithromycin susceptibility in strains belonging to the Mycobacterium abscessus group by erm(41) and rrl sequencing. Antimicrobial agents and chemotherapy 55:775–781.
- 10. Macheras E, Roux AL, Bastian S, Leao SC, Palaci M, et al. (2011) Multilocus sequence analysis and rpoB sequencing of Mycobacterium abscessus (sensu lato) strains. Journal of clinical microbiology 49:491–499.
- 11. Nakanaga K, Sekizuka T, Fukano H, Sakakibara Y, Takeuchi F, et al. (2014) Discrimination of Mycobacterium abscessus subsp. massiliense from Mycobacterium abscessus subsp. abscessus in Clinical Isolates by Multiplex PCR. Journal of clinical microbiology 52:251–259.
- 12. Furuya EY, Paez A, Srinivasan A, Cooksey R, Augenbraun M, et al. (2008) Outbreak of Mycobacterium abscessus wound infections among "lipotourists" from the United States who underwent abdominoplasty in the Dominican Republic. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America 46:1181–1188.
- 13. Medjahed H, Gaillard JL, Reyrat JM (2010) Mycobacterium abscessus: a new player in the mycobacterial field. Trends in microbiology 18:117–123.
- 14. Villanueva A, Calderon RV, Vargas BA, Ruiz F, Aguero S, et al. (1997) Report on an outbreak of postinjection abscesses due to Mycobacterium abscessus, including management with surgery and clarithromycin therapy and comparison of strains by random amplified polymorphic DNA polymerase chain reaction. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America 24:1147–1153.
- 15. Otsuki T, Izaki S, Nakanaga K, Hoshino Y, Ishii N, et al. (2012) Cutaneous Mycobacterium massiliense infection: a sporadic case in Japan. The Journal of dermatology 39:569–572.
- 16. Nakanaga K, Hoshino Y, Era Y, Matsumoto K, Kanazawa Y, et al. (2011) Multiple cases of cutaneous Mycobacterium massiliense infection in a "hot spa" in Japan. Journal of clinical microbiology 49:613–617.
- 17. Cho YJ, Yi H, Chun J, Cho SN, Daley CL, et al. (2013) The Genome Sequence of 'Mycobacterium massiliense' Strain CIP 108297 Suggests the Independent Taxonomic Status of the Mycobacterium abscessus Complex at the Subspecies Level. PloS one 8:e81560.
- 18. Ripoll F, Pasek S, Schenowitz C, Dossat C, Barbe V, et al. (2009) Non mycobacterial virulence genes in the genome of the emerging pathogen Mycobacterium abscessus. PloS one 4:e5660.
- 19. Choo SW, Wee WY, Ngeow YF, Mitchell W, Tan JL, et al. (2014) Genomic reconnaissance of clinical isolates of emerging human pathogen Mycobacterium abscessus reveals high evolutionary potential. Sci Rep 4:4061.
- 20. Pawlik A, Garnier G, Orgeur M, Tong P, Lohan A, et al. (2013) Identification and characterization of the genetic changes responsible for the characteristic smooth-to-rough morphotype alterations of clinically persistent Mycobacterium abscessus. Mol Microbiol 90:612–629.
- 21. Joseph J, Rajendran V, Hassan S, Kumar V (2011) Mycobacteriophage genome database. Bioinformation 6:393–394.
- 22. Casali N, Riley LW (2007) A phylogenomic analysis of the Actinomycetales mce operons. BMC genomics 8:60.
- 23. Arruda S, Bomfim G, Knights R, Huima-Byron T, Riley LW (1993) Cloning of an M. tuberculosis DNA fragment associated with entry and survival inside cells. Science 261:1454–1457.
- 24. Pandey AK, Sassetti CM (2008) Mycobacterial persistence requires the utilization of host cholesterol. Proceedings of the National Academy of Sciences of the United States of America 105:4376–4380.
- 25. Klepp LI, Forrellad MA, Osella AV, Blanco FC, Stella EJ, et al. (2012) Impact of the deletion of the six mce operons in Mycobacterium smegmatis. Microbes and infection/Institut Pasteur 14:590–599.
- 26. Kim SH, Harzman C, Davis JK, Hutcheson R, Broderick JB, et al. (2012) Genome sequence of Desulfitobacterium hafniense DCB-2, a Gram-positive anaerobe capable of dehalogenation and metal reduction. BMC microbiology 12:21.
- 27. Suzuki R, Katayama T, Kim BJ, Wakagi T, Shoun H, et al. (2010) Crystal structures of phosphoketolase: thiamine diphosphate-dependent dehydration mechanism. The Journal of biological chemistry 285:34279–34287.
- 28. Boes N, Schreiber K, Hartig E, Jaensch L, Schobert M (2006) The Pseudomonas aeruginosa universal stress protein PA4352 is essential for surviving anaerobic energy stress. Journal of bacteriology 188:6529–6538.
- 29. Klotz T, Vorreuther R, Heidenreich A, Zumbe J, Engelmann U (1996) Testicular tissue oxygen pressure. The Journal of urology 155:1488–1491.
- 30. Shahidi M, Wanek J, Blair NP, Little DM, Wu T (2010) Retinal tissue oxygen tension imaging in the rat. Investigative ophthalmology & visual science 51:4766–4770.
- 31. Wang W, Vadgama P (2004) O2 microsensors for minimally invasive tissue monitoring. Journal of the Royal Society, Interface/the Royal Society 1:109–117.
- 32. Ponce LL, Pillai S, Cruz J, Li X, Julia H, et al. (2012) Position of probe determines prognostic information of brain tissue PO2 in severe traumatic brain injury. Neurosurgery 70:1492–1502 discussion 1502–1493.
- 33. Barkan D, Rao V, Sukenick GD, Glickman MS (2010) Redundant function of cmaA2 and mmaA2 in Mycobacterium tuberculosis cis cyclopropanation of oxygenated mycolates. Journal of bacteriology 192:3661–3668.
- 34. Banerjee R, Vats P, Dahale S, Kasibhatla SM, Joshi R (2011) Comparative genomics of cell envelope components in mycobacteria. PloS one 6:e19280.
- 35. Adekambi T, Reynaud-Gaubert M, Greub G, Gevaudan MJ, La Scola B, et al. (2004) Amoebal coculture of "Mycobacterium massiliense" sp. nov. from the sputum of a patient with hemoptoic pneumonia. Journal of clinical microbiology 42:5493–5501.
- 36. Simpson JT, Wong K, Jackman SD, Schein JE, Jones SJ, et al. (2009) ABySS: a parallel assembler for short read sequence data. Genome research 19:1117–1123.
- 37. Li H, Ruan J, Durbin R (2008) Mapping short DNA sequencing reads and calling variants using mapping quality scores. Genome research 18:1851–1858.
- 38. Bao H, Guo H, Wang J, Zhou R, Lu X, et al. (2009) MapView: visualization of short reads alignment on a desktop computer. Bioinformatics 25:1554–1555.
- 39. Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, et al. (2008) The RAST Server: rapid annotations using subsystems technology. BMC genomics 9:75.
- 40. Jones P, Binns D, Chang HY, Fraser M, Li W, et al. (2014) InterProScan 5: genome-scale protein function classification. Bioinformatics 30:1236–1240.
- 41. Petkau A, Stuart-Edwards M, Stothard P, Van Domselaar G (2010) Interactive microbial genome visualization with GView. Bioinformatics 26:3125–3126.
- 42. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. Journal of molecular biology 215:403–410.
- 43. Carver TJ, Rutherford KM, Berriman M, Rajandream MA, Barrell BG, et al. (2005) ACT: the Artemis Comparison Tool. Bioinformatics 21:3422–3423.
- 44. Rice P, Longden I, Bleasby A (2000) EMBOSS: the European Molecular Biology Open Software Suite. Trends in genetics: TIG 16:276–277.
- 45. Earl D, Bradnam K, St John J, Darling A, Lin D, et al. (2011) Assemblathon 1: a competitive assessment of de novo short read assembly methods. Genome research 21:2224–2241.
- 46. Li H, Durbin R (2010) Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics 26:589–595.
- 47. Li H (2011) A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data. Bioinformatics 27:2987–2993.
- 48. Koboldt DC, Chen K, Wylie T, Larson DE, McLellan MD, et al. (2009) VarScan: variant detection in massively parallel sequencing of individual and pooled samples. Bioinformatics 25:2283–2285.
- 49. Stamatakis A (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22:2688–2690.
- 50. Katoh K, Toh H (2010) Parallelization of the MAFFT multiple sequence alignment program. Bioinformatics 26:1899–1900.
- 51. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, et al. (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23:2947–2948.