| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Meningitis and Special Pathogens Branch, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, GA 30333, USA
2 Washington University, School of Medicine, Barnes-Jewish Hospital, St Louis, MO 63110, USA
3 Center for Pulmonary and Infectious Disease Control, University of Texas Health Center at Tyler, 11937 US Hwy 271, Tyler, TX 75708-3154, USA
4 Tuberculosis/Mycobacteriology Branch, Division of AIDS, STD and TB Laboratory Research, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, GA 30333, USA
5 Department of Microbiology, University of Texas Health Center at Tyler, 11937 US Hwy 271, Tyler, TX 75708-3154, USA
Correspondence
June M. Brown
jmb6{at}cdc.gov
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Published online ahead of print on 12 March 2004 as DOI 10.1099/ijs.0.02743-0.
The GenBank/EMBL/DDBJ accession numbers for the 16S rRNA gene sequences determined in this work are as follows: Mycobacterium porcinum, AY012582, AY012574, AY012580 and AY012581; Mycobacterium boenickei, AY012573; Mycobacterium houstonense, AY012579; Mycobacterium neworleansense, AY012575; Mycobacterium septicum, AY012576; Mycobacterium brisbanense, AY012577.
Differences in the RT patterns of the type and reference strains of M. fortuitum, signature nucleotides of hypervariable region A and DNA relatedness among strains in the species and their relatedness to other strains of the M. fortuitum third biovariant complex and related species are available as supplementary material in IJSEM Online.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 1991). Involvement of the central nervous system is rare, but meningitis may develop after trauma or surgery. The immunocompromised patient is at special risk for developing severe diseases, especially catheter-related infection with bacteraemia. Disseminated cutaneous disease with skin abscesses, cellulitis and subcutaneous nodules is common with Mycobacterium chelonae and Mycobacterium abscessus but rare with members of the M. fortuitum third biovariant complex (Ingram et al., 1993). It is important for practitioners to be aware of these organisms as possible aetiological agents, as they are resistant to most first-line anti-tuberculous agents.
Wallace et al. (1991) were the first to describe the M. fortuitum unnamed third biovariant complex, containing two unnamed taxa (sorbitol-positive and sorbitol-negative). Recent work has shown that a strain previously identified phenotypically as a member of the M. fortuitum third biovariant complex (sorbitol-negative) represented a novel species, Mycobacterium septicum (Schinsky et al., 2000). In this report, we demonstrate that, despite the common conception that the M. fortuitum third biovariant complex comprises two phenotypically homogeneous taxa, significant heterogeneity exists. After detailed analysis of 21 clinical isolates formerly classified as the M. fortuitum third biovariant complex (sorbitol-negative), three clinical isolates formerly classified as the M. fortuitum third biovariant complex (sorbitol-positive), the type strain and two clinical isolates of M. fortuitum, the type strain of Mycobacterium peregrinum (pipemidic acid-susceptible), the type strain of Mycobacterium porcinum, the type strain and one reference strain of Mycobacterium senegalense and the type strain of M. septicum, we propose four novel species: Mycobacterium boenickei sp. nov., Mycobacterium houstonense sp. nov., Mycobacterium neworleansense sp. nov. and Mycobacterium brisbanense sp. nov. In addition, 13 of the M. fortuitum third biovariant complex (sorbitol-negative) human clinical isolates were recognized as M. porcinum, previously identified by Tsukamura et al. (1983) from submandibular lymph nodes from swine, and one clinical isolate was recognized as M. septicum.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The following strains were obtained from the American Type Culture Collection (ATCC), Manassas, VA, USA: M. senegalense ATCC 35796T, M. peregrinum ATCC 14467T, and M. fortuitum group ATCC 49404. M. septicum ATCC 700731T was obtained from the collection of the Actinomycetes Reference Laboratory, Centers for Disease Control and Prevention, Atlanta, GA, USA. The remaining strains were provided by the Mycobacteria/Nocardia Laboratory, University of Texas Health Center at Tyler, TX, USA. Type and reference strains of M. fortuitum, M. peregrinum (pipemidic acid-susceptible), M. porcinum, M. senegalense and M. septicum were used as controls for the isolates studied. Strains were stored as suspensions in trypticase soy broth (Becton Dickinson) with 20 % glycerol (Sigma) at 70 °C and maintained on Löwenstein–Jensen medium slants (Remel) until analysis.
|
). Conventional biochemical tests were performed as used routinely in the Actinomycetes Reference Laboratory of the Centers for Disease Control and Prevention (Berd, 1973), and classical tests for species differentiation of the mycobacteria were performed (Silcox et al., 1981). Carbon-utilization tests were performed (Yassin et al., 1995). Semi-quantitative catalase analysis and assessment of the thermostability of catalase at 68 °C were performed (Kirschner et al., 1992; Vestal, 1975). Additionally, we tested for the utilization of acetamide (BBL) and citrate (BBL) as sole carbon and nitrogen sources, and for growth at 42 °C using the methods described by Wallace et al. (1991).
A broth microdilution method using cation-supplemented Mueller–Hinton broth, as described by Wallace et al. (1991), was used for antimicrobial-susceptibility tests. Plates were incubated at 30 °C for 72 h. MIC values were determined for amikacin, amoxycillin/clavulanate, ampicillin, cefotaxime, ceftriaxone, ciprofloxacin, doxycycline, erythromycin, imipenem, minocycline, sulfamethoxazole, trimethoprim/sulfamethoxazole and vancomycin. Break points were derived from aerobic dilution interpretative criteria found in National Committee for Clinical Laboratory Standards document M7-A4 (NCCLS, 1997).
DNA purification.
Strains were subcultured from Löwenstein–Jensen medium slants into 200 ml Mueller–Hinton broth containing 1·0 ml Tween 80 (Sigma) and 3·0 g glycine (Sigma) and then grown for 1–2 days at 35 °C before being harvested by centrifugation at 3795 g for 30 min. DNA was purified from lysed protoplasts as previously described by Lasker et al. (1992). Repeat extractions were performed with a 20 % (w/v) SDS (Roche) solution to improve the DNA yield, a method adapted from Loeffelholz & Scholl (1989).
DNA relatedness and G+C content determination.
The hydroxyapatite method used for DNA labelling and DNA relatedness was described previously by Brenner et al. (1982). Using a nick translation kit (Gibco), DNA was labelled in vitro with [32P]dCTP. The optimal hybridization temperature used was 75 °C; the percentage divergence was calculated to the nearest 0·5 % (Brenner et al., 1983). Relative binding ratios expressed as percentages were the means of at least two hybridization experiments.
The G+C content of DNA from representative isolates was determined spectrophotometrically by using the thermal denaturation method as described by Mandel et al. (1970).
Ribotyping.
Genomic DNA (1–3 mg ml–1) was digested for 8 h at 35 °C with 20 U of either of the two restriction enzymes PvuII or SalI (Roche) in the buffers recommended by the manufacturer. DNA fragments were separated by electrophoresis in 0·85 % (w/v) agarose gel (Gibco). The DNA fragments were then transferred to a nylon membrane (Nytran; Schleicher & Schuell) and were probed at 37 °C with digoxigenin-labelled cDNA derived from purified 16S and 23S rRNA of Escherichia coli by reverse transcription as described previously (Popovic et al., 1993). The rDNA-containing fragments were visualized using the ‘Genius' kit (Roche) protocol.
PCR amplification and restriction enzyme analysis.
DNA from bacterial cells harvested from trypticase soy agar (Remel) plates was prepared for PCR amplification according to methods previously described by Steingrube et al. (1995), Telenti et al. (1993) and Wilson et al. (1998). A 439 bp segment of the hsp65 gene was amplified from ground-cell supernatants by PCR with 1·0 U Taq DNA polymerase (Perkin Elmer) in optimized buffer E (1·5 mM MgCl2, pH 9·0; Invitrogen) containing 83 µM each dNTP, 9 % (v/v) DMSO and 1 µM of each primer, i.e. TB11 (5'-ACCAACGATGGTGTGTCCAT-3') and TB12 (5'-CTTGTCGAACCGCATACCCT-3') (Midland Certified Reagent Co.), together with the appropriate positive and negative controls according to a modification of the method of Telenti et al. (1993). The amplification parameters were 45 cycles of 1 min each at successive temperatures of 94, 55 and 72 °C, followed by a 10 min extension at 72 °C.
Two commercially available restriction endonucleases, BstEII and HaeIII (Promega), were used to produce PCR restriction analysis (PRA) band patterns, as described previously by Steingrube et al. (1995, 1997). Restriction digests were incubated for the appropriate time periods at the appropriate temperatures, using buffers recommended by the manufacturer.
Restriction fragments were electrophoresed on 3 % (w/v) Metaphor agarose, 4 bp resolution (FMC Bioproducts), containing ethidium bromide (0·625 µg ml–1), in a Mini-SubCell electrophoresis system (Bio-Rad) at 95 V for 1·5–2·0 h. PRA band sizes (in bp) were estimated on a computerized Bio Image system (Millipore), with a 100 bp ladder (Life Technologies) and a pGem-bp ladder (Promega) as molecular size standards. Restriction fragment sizes were rounded to the nearest 5 bp, as recommended by Telenti et al. (1993).
16S rRNA gene sequencing.
Purified genomic DNA was amplified using the Expand High Fidelity PCR System (Roche), which includes a low concentration of proof-reading enzyme Pfu (Barnes, 1994) to increase the fidelity of products. The protocol was modified with the addition of 5 % (v/v) DMSO to increase the yield of high-G+C templates (Scheidl et al., 1995). Each 50 µl reaction contained approximately 10 ng DNA, 2·5 U polymerase, 1·5 mM MgCl2, 5 % DMSO, 200 µM dNTPs, 100 nM each of primers FD1 5'-AGAGTTTGATCCTGGCTCAG and RD1 5'-AAGGAGGTGATCCAGCC. Amplification was performed on an ABI 9700 (Applied Biosystems) thermocycler at 94 °C for 5 min, followed by 35 cycles of 94 °C for 15 s, 50 °C for 15 s and 72 °C for 90 s, and finalized by a single extension of 72 °C for 5 min followed by maintenance at 4 °C.
Amplification products (5 µl) were electrophoresed on a 1·2 % (w/v) agarose gel with a 500 bp ladder for 30 min at 85 V. Excess dNTPs and primers were removed from products with the QIAquick 8 PCR Purification kit (Qiagen), a vacuum-based, chaotropic, salt–silica binding method. Cycle sequencing was performed using standard protocols with the addition of 5 % DMSO (Scheidl et al., 1995), utilizing 9 pM each primer: FD1, 5'-AGAGTTTGATCCTGGCTCAG, positions 7–25; 5'-AGTTGATCCTGGCTCAG, 9–25; 5'-TACGGGAGGCAGCAG, 342–356; 5'-CAGCAGCCGCGGTAATAC, 517–535; 5'-CTTGAGTGCACGAAGGAGTGG, 650–671; 5'-ATTAGATACCCTGGTAG, 786–802; 5'-CCCGCAACGAGCGCAACCC, 1094–1113; 5'-GGGCCTTGTACACACCG, 1384–1400; 5'-AAGTCGTAACAAGGTAACC, 1491–1509; RD1, 5'-GGCTGGATCACCTCCTT, 1524–1540 (positions correspond to positions on the Escherichia coli 16S rRNA gene; Brosius et al., 1978). Excess dyes were removed with magnetic carboxylate beads (Agencourt Bioscience) and reactions were sequenced on an ABI 3100 sequencer (Applied Biosystems).
Phylogenetic analysis.
The 16S rRNA gene sequences were assembled in SEQMERGE (Accelrys) and trimmed to a minimum of two confirming reads. The sequences (>1400 bp) generated in the present study were compared with the GenBank database gene sequences derived from related Mycobacterium species and submitted by other investigators. When representative isolates [BAA-328, RT-A (n=10); ATCC 49939, RT-D; W6236, RT-G; and W6243, RT-F] of one cluster (n=13) of M. fortuitum third biovariant sorbitol-negative isolates were aligned in PILEUP, the sequences of the representative isolates corresponded to the sequence of M. porcinum (GenBank accession no. AF480588) that was deposited in the GenBank database in March 2002 (Turenne et al., 2001). Phylogenetic trees were created in TREECON using the Jukes–Cantor model (Jukes & Cantor, 1969) with 1000 bootstrapping resamples, topology was inferred by using the neighbour-joining method, and sequences of 21 Mycobacterium species were rooted with outgroup Nocardia otitidiscaviarum (GenBank accession no. M59056). The trimmed 16S rRNA gene sequences were aligned using CLUSTAL W in MegAlign (DNASTAR) to produce similarity values (Thompson et al., 1994).
The 37 bp hypervariable region A, beginning with and corresponding to position 175 in the E. coli numbering system designated by Kirschner et al. (1993), was extracted from the 16S rRNA gene sequences and aligned with the BioEdit program (Hall, 1999).
HPLC.
Mycolic acid analysis of representative isolates was performed according to the established procedure for mycobacteria (Butler et al., 1991, 1992). The chromatographic patterns generated by these isolates were matched to standard HPLC patterns of control species by using relative retention time (RRT) ratios for peaks, as calculated by the chromatographic software (Beckman Instruments). The ratios were derived by using a high-molecular-weight standard (Ribi ImmunoChem Research) as the reference peak.
Cellular fatty acid analysis.
Cells were grown for 3 days, saponified and the liberated fatty acids were methylated and analysed by GLC (Weyant et al., 1996). Identification of fatty acids was performed using a commercially available software package (MIDI) utilized with CDC library created using LGS software (Weyant et al., 1996). The presence of a trans isomer of hexadecenoic (16 : 1
9t) acid (reported by MIDI software as summed feature 4, which comprises i-2-OH-15 : 0, 16 : 19t or both) was confirmed by acetylation as described by Weyant et al. (1996).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Most strains studied were isolated either in the USA (66 %) or in Australia (19 %) (Table 1).
DNA hybridization and G+C content determination
DNA-relatedness studies divided the 24 M. fortuitum third biovariant complex (sorbitol-positive and sorbitol-negative) isolates into six genetic groups according to the phylogenetic definition of a species as ‘strains with approximately 70 % or greater DNA–DNA relatedness and with 5 % or less 
Tm’ (Wayne et al., 1987), and according to the criteria described by Brenner et al. (1993) (Supplementary Table available in IJSEM Online). One cluster (n=13) represented by ATCC BAA-328, formerly referred to as strains of the M. fortuitum third biovariant (sorbitol-negative), was recognized as M. porcinum, while one strain (ATCC BAA-329) was recognized as M. septicum. In addition to Mycobacterium genomospecies 1, consisting of six strains (including ATCC 49935T), formerly referred to as strains of the M. fortuitum third biovariant (sorbitol-negative), three other genomospecies are described. These include Mycobacterium genomospecies 2, consisting of two strains (including ATCC 49403T) and formerly referred to as strains of the M. fortuitum third biovariant (sorbitol-positive), Mycobacterium genomospecies 3, consisting of one strain (ATCC 49404T) and formerly referred to as the reference strain of the M. fortuitum third biovariant (sorbitol-negative), and Mycobacterium genomospecies 4, consisting of one strain (ATCC 49938T) and formerly referred to as the M. fortuitum third biovariant (sorbitol-positive).
The G+C contents for representative isolates are given in the species descriptions. Except for Mycobacterium genomospecies 3 (ATCC 49404T), which had a G+C content of 60 mol%, all values were within the 61–71 mol% range for Mycobacterium species (Takeuchi & Hatano, 1998).
Ribotyping
Ribotype (RT) analysis with the enzyme SalI revealed seven different RT patterns for the 21 M. fortuitum third biovariant sorbitol-negative isolates: RT groups A (n=10), B (n=6), C, D, E, F and G (n=1 isolate each). Two different patterns for the three M. fortuitum third biovariant sorbitol-positive isolates were revealed, RT groups H (n=2) and I (n=1), which differed from those of the type strains of M. fortuitum, M. peregrinum, M. senegalense and M. septicum (Fig. 1a). In addition, different RT patterns were found between the type and the reference strains of M. fortuitum (RT-Mf1 and RT-Mf2) with the enzyme SalI (data not shown; results summarized in supplementary data available in IJSEM Online, Table 2). Since the initial RT groups were determined with SalI and these RT groups were used as the basis for DNA–DNA relatedness, chemotaxonomic, 16S rRNA gene sequence and PRA analyses, the patterns are retained in Table 2 as references. As SalI did not digest DNA from the type strain of M. senegalense, further RT analysis with the enzyme PvuII was performed to compare representative isolates of the SalI RT groups and the studied type strains, including the type strain of M. senegalense. As before, nine different RT patterns that differed from those of the type strains of M. fortuitum, M. peregrinum, M. senegalense and M. septicum were found among these 24 isolates of the M. fortuitum third biovariant complex (Fig. 1a).
|
|
, lanes 6–14), differed from the PRA patterns of control strains M. senegalense ATCC 35796T, M. peregrinum ATCC 14467T and M. fortuitum ATCC 6841T (Fig. 1b, c, lanes 2–4) but not from M. septicum ATCC 700731T (Fig. 1b, c, lanes 5). Patterns of HaeIII- and BstEII-derived digests of the PCR amplicons of M. septicum ATCC 700731T, the four isolates of M. porcinum, i.e. ATCC BAA-328 (RT-A), ATCC 49939 (RT-D), W6243 (RT-F) and W6236 (RT-G), and M. septicum BAA-329 (RT-E) were identical to each other, as shown in Fig. 1(b, c, lanes 5, 6 and 9–12, respectively), and summarized in Table 2, but differed from all other patterns. RFLP with PRA of Mycobacterium genomospecies 3 (ATCC 49404T) (Fig. 1b, lane 8) with the enzyme HaeIII was similar to the RFLP patterns of Mycobacterium genomospecies 1 (ATCC 49935T) (RT-B) (Fig. 1b, lane 7) and Mycobacterium genomospecies 2 (ATCC 49403T) (RT-H) (Fig. 1b, lane 13). Although the RFLP pattern of Mycobacterium genomospecies 3 (ATCC 49404T) with enzyme BstEII (Fig. 1c, lane 8) differed from the pattern of Mycobacterium genomospecies 1 (ATCC 49935T) (RT-B) (Fig. 1c, lane 7), the PRA pattern of Mycobacterium genomospecies 3 (ATCC 49404T) (Fig. 1c, lane 8) with this enzyme was identical to that of Mycobacterium genomospecies 2 (ATCC 49403T) (RT-H) (Fig. 1c, lane 13). The combined patterns of HaeIII- and BstEII-derived digests of Mycobacterium genomospecies 4 (ATCC 49938T) (RT-I) (Fig. 1b, c, lanes 14) were unique (Table 2). The PRA pattern of the type strain of M. porcinum ATCC 33776T is given in Table 2. Although the PRA pattern of the type strain is not shown in Fig. 1(b, c), the PRA pattern was identical to those of M. porcinum isolates ATCC BAA-328 (RT-A), ATCC 49939 (RT-D), W6243 (RT-F) and W6236 (RT-G).
16S rRNA gene analysis
The 1401 bp region of a 16S rRNA gene, corresponding to E. coli positions 84–1494 (GenBank accession no. J01859), was determined and analysed for the Mycobacterium isolates shown in Fig. 2. The multiple alignment was used as the input to a phylogeny inference software package (TREECON) employing the neighbour-joining method of Saitou & Nei (1987) to generate a phylogenetic tree (Fig. 2). The GenBank accession numbers and strain designations for the sequences used in this alignment are given in Fig. 2.
|
By 16S rRNA gene sequence analysis, M. septicum ATCC BAA-329, formerly referred to as the M. fortuitum third biovariant (sorbitol-negative), had a similarity of 100 %, over the 1401 bp region, to M. septicum ATCC 700731T. These two strains had the same PRA pattern, but differed phenotypically and genetically when compared using ribotyping.
Mycobacterium genomospecies 1 (ATCC 49935T) was 99·9 % similar, by pairwise comparisons of 16S rRNA gene sequences, to M. porcinum ATCC 33776T and to the M. porcinum isolates (ATCC BAA-328, W6243, ATCC 49939 and W6236) and was 100 % related to Mycobacterium genomospecies 3 (ATCC 49404T).
Mycobacterium genomospecies 2 (ATCC 49403T), formerly referred to as the M. fortuitum third biovariant (sorbitol-positive), was 99·7 % similar, by pairwise comparisons of 16S rRNA gene sequences, to M. senegalense ATCC 35796T and was 99·1 % related to M. fortuitum ATCC 6841T.
The close sequence similarity between Mycobacterium genomospecies 3 (ATCC 49404T) and Mycobacterium genomospecies 1 (ATCC 49935T) was not supported phenotypically or genetically by DNA relatedness studies, by ribotyping or by PRA.
Mycobacterium genomospecies 4 (ATCC 49938T) was not phylogenetically related by 16S rRNA gene sequence analysis to any of our studied group of isolates of the M. fortuitum third biovariant complex (Fig. 2).
The signature nucleotides of hypervariable region A are available as supplementary material in IJSEM Online. As Kirschner et al. (1992) have previously pointed out, a highly variable region between positions 175 and 212 (E. coli numbering) contains possible species-specific regions. Unique sequence stretches were found for Mycobacterium genomospecies 4 (ATCC 49938T). However, there were no nucleotide differences among the type strains of M. peregrinum and M. septicum, none among M. porcinum ATCC 33776T, M. porcinum isolates, M. septicum (ATCC BAA-329), Mycobacterium genomospecies 1 (ATCC 49935T) and Mycobacterium genomospecies 3 (ATCC 49404T), and none between Mycobacterium genomospecies 2 (ATCC 49403T) and M. senegalense ATCC 35796T.
Phenotypic analysis
Microscopic morphological studies showed that all isolates studied were Gram-positive, pleomorphic bacilli lacking spores and capsules, but frequently appearing as long filamentous forms with rudimentary branches. No right-angled branched forms were observed. Each isolate, under low-power stereomicroscopic examination, showed white to slightly beige, small-diameter (approx. 1 mm) colonies after incubation on heart infusion agar with 5 % (v/v) rabbit blood for 2 days at 35 °C. No aerial hyphae were seen. Colonial variations in surface texture and edges were seen: all M. porcinum strains were mucoid, convex and entire-edged, except isolate W6236, which demonstrated slightly irregular translucent edges; M. septicum (ATCC BAA-329) showed both rough, wrinkled, irregularly edged colonies and mucoid, convex, round, entire-edged colonies; all Mycobacterium genomospecies 1 strains were matt, domed and scalloped-edged; both Mycobacterium genomospecies 2 strains were mucoid, convex, round and entire-edged; Mycobacterium genomospecies 3 was rough, wrinkled and irregularly edged; and Mycobacterium genomospecies 4 was mucoid, convex, round and entire-edged. All strains were acid-fast when tested using the modified Kinyoun method. All M. fortuitum third biovariant complex (sorbitol-positive and sorbitol-negative) isolates under study grew on Löwenstein–Jensen medium at 35 °C in less than 7 days and were negative for both growth in lysozyme and utilization of citrate. All were positive for utilization of acetamide and were positive for 3 day arylsulfatase production, except for two isolates (W6241 and W6229) that were positive in 14 days for arylsulfatase production. All produced acid from i-myo-inositol, D-mannitol, mannose and trehalose, were resistant to ampicillin, cefotaxime and ceftriaxone and were susceptible to ciprofloxacin, imipenem and trimethoprim/sulfamethoxazole. Antimicrobial susceptibility and biochemical differences among these isolates and type strains are summarized in Table 3.
|
). The mycolates for the isolates under study appeared within the expected range for the rapidly growing mycobacteria, which elute after 5 min when compared by the HPLC method. Comparison of HPLC chromatograms and RRT ratios for representative isolates with those of strains of M. fortuitum, M. peregrinum, M. senegalense and M. septicum revealed that separation of these species was problematic when the current HPLC method was used (results not shown). The RRT ratios for initial eluting peaks for each of the species studied were the same, but M. senegalense exhibited some differences in the RRT ratios for the late-eluting peaks, as reported previously (Schinsky et al., 2000). The late-eluting peaks for the studied isolates, M. fortuitum, M. peregrinum and M. septicum exhibited the same RRT, with slight differences in relative heights. Visual recognition when the current HPLC method was used clearly distinguished these isolates from M. abscessus and M. chelonae, which were previously included in the M. fortuitum complex.
The cellular fatty acid compositions of saponified whole cells of the type strains of M. fortuitum, M. peregrinum, M. septicum and representative studied isolates from M. porcinum, M. septicum (ATCC BAA-329), Mycobacterium genomospecies 1, Mycobacterium genomospecies 2, Mycobacterium genomospecies 3 and Mycobacterium genomospecies 4 are characterized by large amounts (18–36 %) of 16 : 0 and 18 : 19c and the presence of 14 : 0, 16 : 19c, 16 : 17c, 16 : 19t, 18 : 2, 18 : 17c, 18 : 0, 10-methyl-18 : 0 (tuberculostearic acid) and 20 : 4. The overall cellular fatty acid profile of M. senegalense is similar to those of the other isolates tested, except for the presence of small amounts of 19 : 0cyc (2 %), the absence of 18 : 2 and 20 : 4, and larger amounts of 18 : 17c (7 % vs T-2 %) and 10-methyl-18 : 0 (18 % versus 4–10 %). The studied isolates of the M. fortuitum third biovariant complex all have virtually the same cellular fatty acid content and thus must be identified by additional methods such as DNA relatedness studies.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Kiehlbauch et al., 1991; Popovic et al., 1993), we initially ribotyped 24 phenotypically identified isolates of the M. fortuitum third biovariant complex (sorbitol-negative and sorbitol-positive) and compared the RT patterns with type and reference strains of M. fortuitum, M. peregrinum, M. senegalense and M. septicum. RT patterns of SalI- and PvuII-derived digests divided the 24 isolates studied into nine groups: seven RT groups of the M. fortuitum third biovariant complex (sorbitol-negative) [RT groups A (n=10), B (n=6) and C, D, E, F and G (n=1 isolate each)] and two RT groups of the M. fortuitum third biovariant complex (sorbitol-positive) [RT groups H (n=2) and I (n=1)]. These results suggested that this group of isolates of the M. fortuitum third biovariant complex is more heterogeneous than has previously been appreciated, and provided a genetic framework for other taxonomic studies. For this purpose, a polyphasic approach was used, assessing an expanded battery of phenotypic characteristics, chemotaxonomic profiles and genotypic analysis that included a PCR-based identification method analysing patterns obtained from a 439 bp fragment of the hsp-65 kDa gene (PRA), 16S rRNA gene sequencing and DNA–DNA hybridization studies. Decision priority was given to the results of DNA–DNA hybridization in conjunction with 16S rRNA gene sequence similarities, provided that phenotypic properties were in agreement with genetic characteristics (Wayne et al., 1987; Stackebrandt & Goebel, 1994).
Relatedness values obtained using the hydroxyapatite method reflect degrees of overall similarity between genomic DNAs and thus define bacterial species in phylogenetic terms (Wayne et al., 1987). There were two exceptions to the definition for a species designation among the M. porcinum strains. Labelled M. porcinum ATCC BAA-328 was 69 % related, with 1·0 % divergence, to W6236 (RT-G); however, labelled W6236 (RT-G) was 83 % related, with 0·0 % divergence, to ATCC BAA-328. In addition, M. porcinum labelled ATCC BAA-328 was 62 % related, with 0·5 % divergence, to W6243 (RT-F). Strain W6243 (RT-F) was not labelled; however, another labelled M. porcinum strain, ATCC 49939, was 73 % related, with 1·5 % divergence, to W6243 (RT-F). Furthermore, M. porcinum ATCC BAA-328 was found to be 100 % similar, in pairwise comparisons of 16S rRNA gene sequence, to ATCC 49939, W6236 and W6243 (Fig. 2). Thus, these three isolates were included in the taxon M. porcinum. The labelled type strain of M. porcinum, ATCC 33776T, was 91 % related to ATCC BAA-328, with 1·5 % divergence, and was 58 % related to Mycobacterium genomospecies 1 (ATCC 49935T), with 5·0 % divergence. This close DNA relationship between M. porcinum ATCC 33776T and ATCC BAA-328 confirmed the identity of M. porcinum from human clinical sources. M. porcinum ATCC 33776T showed relatively low DNA–DNA relatedness to Mycobacterium genomospecies 2 (ATCC 49403T) (39 %, 10 % divergence), Mycobacterium genomospecies 3 (ATCC 49404T) (43 %, 7·0 % divergence), Mycobacterium genomospecies 4 (ATCC 49938T) (16 %, 11·5 % divergence), M. septicum ATCC 700731T (40 %, 7·5 % divergence) and M. septicum ATCC BAA-329 (57 %, 8·0 % divergence).
By 16S rRNA gene sequence analysis, ATCC BAA-329 (RT-E), formerly referred to as a M. fortuitum third biovariant (sorbitol-negative), was found to have a similarity of 100 %, over the 1401 bp region, to M. septicum ATCC 700731T and had a PRA profile that was identical to that of M. septicum ATCC 700731T. Labelled genomic DNA of ATCC BAA-329 was 61 % related to M. septicum ATCC 700731T, with 2·0 % divergence, in DNA–DNA hybridization studies. Labelled M. septicum ATCC 700731T was 78 % related to ATCC BAA-329, with 4·5 % divergence, meeting the criteria defined by Wayne et al. (1987) for the same species. Although ATCC BAA-329 differed phenotypically from all the other isolates studied (Table 3), genetically, this strain, except for a unique RT pattern, was most similar to M. septicum ATCC 700731T.
The range of relatedness between labelled M. porcinum ATCC BAA-328 and the six Mycobacterium genomospecies 1 strains was 48–59 %, with divergences ranging from 2·5 to 3·5 %. In addition, the range of relatedness between labelled Mycobacterium genomospecies 1 (ATCC 49935T) and representative M. porcinum strains was 47–65 %, with 3·0–4·0 % divergence. Both of these DNA relatedness groups met the criteria for unique taxa, but the strains comprising Mycobacterium genomospecies 1 were not sufficiently distinct phenotypically from M. porcinum to meet these criteria. The only significant difference was the resistance of 83 % of Mycobacterium genomospecies 1 strains to minocycline. Mycobacterium genomospecies 1 showed low relatedness values (41 and 42 %) to Mycobacterium genomospecies 2 and Mycobacterium genomospecies 3. The relatedness of Mycobacterium genomospecies 1 to Mycobacterium genomospecies 4 was not determined, but the reciprocal level of relatedness was low (3 and 7 %) with Mycobacterium genomospecies 1 (ATCC 49935T and ATCC 49937). Low levels of relatedness (20–47 %) to the type strains of M. fortuitum, M. peregrinum, M. senegalense and M. septicum were found.
The labelled genomic DNA of Mycobacterium genomospecies 2 (ATCC 49403T) (RT-H), previously referred to as a reference isolate for the M. fortuitum third biovariant complex (sorbitol-positive), was found to be 96 % related, with 0 % divergence, to Mycobacterium genomospecies 2 (ATCC 49934) (RT-H), another reference isolate for the M. fortuitum third biovariant complex (sorbitol-positive). Mycobacterium genomospecies 2 (ATCC 49403T) showed relatively low DNA–DNA relatedness to Mycobacterium genomospecies 1 (ATCC 49935T) (41 %, 7·5 % divergence) and W6229 (21 %, 8·5 % divergence), Mycobacterium genomospecies 3 (ATCC 49404T) (38 %, 7·5 % divergence) and Mycobacterium genomospecies 4 (ATCC 49938T) (15 %, 9·0 % divergence). Low values of relatedness (35–65 %) to the type or reference strains of M. fortuitum, M. peregrinum, M. porcinum, M. senegalense and M. septicum were found.
One strain, ATCC 49404T (RT-C), formerly the reference strain for the M. fortuitum third biovariant (sorbitol-negative) and now referred to as Mycobacterium genomospecies 3, was 100 % similar in a pairwise comparison of 1401 bp of the 16S rRNA gene sequence with that of Mycobacterium genomospecies 1 (ATCC 49935T) (RT-B). This similarity was not supported phenotypically, by DNA relatedness studies, by ribotyping or by PRA analysis. Phenotypically, Mycobacterium genomospecies 3 (ATCC 49404T) could be distinguished from Mycobacterium genomospecies 1 by its production of acid from i-erythritol and D-xylose, and by its susceptibility to doxycycline and minocycline. Genetically, labelled genomic DNA of Mycobacterium genomospecies 3 (ATCC 49404T) was 46 % related to that of Mycobacterium genomospecies 1 (ATCC 49935T) (RT-B); conversely, labelled genomic DNA of Mycobacterium genomospecies 1 (ATCC 49935T) was 42 % related to that of Mycobacterium genomospecies 3 (ATCC 49404T), with 6·0 % divergence. These values for relatedness do not meet the criteria recommended by Wayne et al. (1987) for strains to be considered members of the same species. Mycobacterium genomospecies 3 had relatively low values for DNA relatedness to Mycobacterium genomospecies 2 (36 %) and with the reciprocal value of Mycobacterium genomospecies 4 (1 %, 8 % divergence). Low values of relatedness (31–44 %) to the type or reference strains of M. fortuitum, M. peregrinum, M. porcinum, M. senegalense and M. septicum were found. RT patterns of Mycobacterium genomospecies 3 (ATCC 49404T) with both SalI and PvuII differed from the RT pattern of Mycobacterium genomospecies 1 (ATCC 49935T) and the RT patterns of each of the other isolates studied (Fig. 1a). The PRA pattern of Mycobacterium genomospecies 3 (ATCC 49404T) with the enzyme HaeIII was similar to the PRA patterns of Mycobacterium genomospecies 1 (ATCC 49935T) (RT-B) and Mycobacterium genomospecies 2 (ATCC 49403T) (RT-H) (Fig. 1b). Although the PRA pattern of Mycobacterium genomospecies 3 (ATCC 49404T) with the enzyme BstEII differed from that of Mycobacterium genomospecies 1 (ATCC 49935T) (RT-B) (Fig. 1c), its PRA pattern was similar to that of Mycobacterium genomospecies 2 (ATCC 49403T) (RT-H) (Fig. 1c).
The labelled genomic DNA of Mycobacterium genomospecies 4 (ATCC 49938T) (RT-I) was related to the other isolates of the M. fortuitum third biovariant complex that were studied by DNA–DNA hybridization analysis, with values ranging from 1 to 13 % and divergences ranging from 7·0 to 9·5 %. Low values of relatedness (1–13 %) to type or reference strains of M. fortuitum, M. peregrinum, M. porcinum, M. senegalense and M. septicum were found. Phenotypically, this isolate differed from all of the other isolates studied (Table 3). Furthermore, the 16S rRNA gene sequence of Mycobacterium genomospecies 4 (ATCC 49938T) was not phylogenetically related to any of our group of studied isolates of the M. fortuitum third biovariant complex (Fig. 2). Thus, Mycobacterium genomospecies 4 (ATCC 49938T) met the criteria of Wayne et al. (1987) for a novel species.
Application of a PCR-based identification method (PRA) to our group of organisms showed that it had a lower discriminatory ability than ribotyping with either PvuII (Fig. 1a) or SalI (Table 2). This discrepancy could be due to the low degree of heterogeneity existing in a DNA fragment of 439 bp. All M. porcinum isolates tested [W6228 (RT A), W6230 (RT A), ATCC 33776T, ATCC BAA-328 (RT-A), W6233 (RT A), W6235 (RT A), W6237 (RT A), W6239 (RT A), W6241 (RT A), W6242 (RT A), W6244 (RT A), ATCC 49939 (RT-D), W6243 (RT-F) and W6236 (RT-G)] yielded the same PRA restriction pattern: two fragments of 235 and 205–210 bp with BstEII and three fragments of 140/125/100 bp with HaeIII. In addition to the similar phenotypic characteristics (Table 3), DNA reassociation and 16S rRNA gene sequences (Fig. 2), this PRA grouping further supported the status of these strains as M. porcinum according to taxonomic norms. The species M. porcinum was studied by Tsukamura et al. (1983) and recently recognized by Turenne et al. (2001).
Except for their identical HaeIII and BstEII PRA patterns, both M. septicum ATCC 700731T and ATCC BAA-329 (RT-E) differed from the M. porcinum isolates in terms of other phenotypic and genetic characteristics (Tables 2 and 3, Fig. 2).
When the PRA pattern of the type strain of Mycobacterium genomospecies 1 was compared with the PRA patterns of other isolates of Mycobacterium genomospecies 1 (ATCC 49937, W6229, W6231, W6240 and W6245), both the type strain and the five other isolates of Mycobacterium genomospecies 1 gave the same PRA patterns as M. porcinum with the enzymes BstEII and HaeIII, except that they lacked the 100 bp fragment with HaeIII. These two restriction patterns for Mycobacterium genomospecies 1 and M. porcinum, BstEII-derived 235/205 and HaeIII-derived 140/125 and BstEII-derived 235/205 and HaeIII-derived 140/125/100, respectively, were the recognized patterns of M. fortuitum third biovariant complex sorbitol-negative isolates previously observed by Steingrube et al. (1995).
Although the PRA patterns of the HaeIII- and BstEII-derived digests of the PCR amplicon of Mycobacterium genomospecies 3 (ATCC 49404T) (RT-C) (Fig. 1b, c, lanes 8) were identical to those of Mycobacterium genomospecies 2 (ATCC 49403T) (RT-H) (Fig. 1b, c, lanes 13), these two strains were not similar phenotypically or genetically (Table 3, Fig. 2). This restriction pattern, 235/115/85 bp with BstEII and 140/125 bp with HaeIII, was the commonly recognized pattern of the M. fortuitum third biovariant complex sorbitol-positive isolates (Steingrube et al., 1995). The other sorbitol-positive strain tested, Mycobacterium genomospecies 4 (ATCC 49938T), displayed a restriction pattern different from that of all other isolates tested.
Further studies on this PCR-based identification method with more isolates of the third biovariant complex and additional enzymes are needed to establish the frequency of the individual taxa among clinical isolates and to establish whether each of these new genetic groups will yield specific, unique PRA patterns.
Phenotypically, all of the M. fortuitum third biovariant complex sorbitol-negative and sorbitol-positive isolates under study differed from the type strains of M. fortuitum ATCC 6841T, M. peregrinum ATCC 14467T (pipemidic acid-susceptible), M. senegalense ATCC 35796T and M. septicum ATCC 700731T. They differed from M. fortuitum ATCC 6841T in their ability to produce acid from both D-mannitol and i-myo-inositol, they differed from M. peregrinum ATCC 14467T (pipemidic acid-susceptible) in their ability to produce acid from i-myo-inositol, they differed from M. peregrinum ATCC 14467T and M. septicum ATCC 700731T in their ability to utilize acetamide and they differed from M. senegalense ATCC 35796T in both their ability to produce acid from i-myo-inositol and their inability to utilize citrate. Phenotypic differences among the strains studied can be found in Table 3.
Chemotaxonomically, each of these new taxonomic groups exhibited typical HPLC profiles consistent with rapidly growing Mycobacterium species. However, in our present study no distinction could be made either among M. porcinum, M. septicum, Mycobacterium genomospecies 1, Mycobacterium genomospecies 2, Mycobacterium genomospecies 3 and Mycobacterium genomospecies 4 or between these groups and the type strains of M. fortuitum ATCC 6841T, M. peregrinum ATCC 14467T and M. septicum ATCC 700731T, except for M. senegalense ATCC 35796T, as Schinsky et al. (2000) also found in their earlier study.
Consistent with a report by Yassin et al. (1993), these mycobacterial taxonomic groups showed similar cellular fatty acid profiles that could not be differentiated from each other, from the type strains of M. fortuitum ATCC 6841T, M. peregrinum ATCC 14467T, M. senegalense ATCC 35796T and M. septicum ATCC 700731T or from related genera, such as Corynebacterium, Nocardia, Rhodococcus or Tsukamurella. These new taxonomic groups had a DNA G+C content that ranged from 60 to 64 mol%, within the limits of the genus Mycobacterium (Takeuchi & Hatano, 1998).
On the basis of genotypic data, we propose that Mycobacterium genomospecies 1 be classified as a novel species, Mycobacterium bonikei sp. nov. Mycobacterium genomospecies 1 is very similar phenotypically to M. porcinum, except in terms of the nitrate-reducing ability of the type strain of M. porcinum; however, since this group is very different from all M. porcinum isolates in genetic studies, including DNA–DNA reassociation, 16S rRNA gene sequence analysis, ribotyping and PRA, a decision was made to give this group nomenspecies status represented by ATCC 49935T.
On the basis of phenotypic and genotypic data, we propose that Mycobacterium genomospecies 2 be classified as a novel species, Mycobacterium houstonense sp. nov.
We also propose that Mycobacterium genomospecies 3 and 4, each containing one strain, be formally classified as Mycobacterium neworleansense sp. nov. and Mycobacterium brisbanense sp. nov., respectively. The decision to give nomenspecies status rather than genomospecies status was based on the determination that each of these proposed novel species exhibited a clear-cut difference from every other species on the basis of at least two phenotypic properties, not including antimicrobial susceptibility differences.
Description of Mycobacterium boenickei sp. nov.
Mycobacterium boenickei (boe.nick'e.i. N.L. gen. masc. n. boenickei of Bönicke, in honour of the contributions of Rudolf Bönicke, a German mycobacteriologist, who first recognized the heterogeneity within the M. fortuitum complex).
The organisms are acid-fast, Gram-positive, pleomorphic bacilli. Long filamentous forms are often observed, but spores and capsules are absent. Each isolate, under low-power stereomicroscopic examination, shows white to slightly beige, small-diameter (approx. 1 mm) colonies after incubation on heart infusion agar with 5 % (v/v) rabbit blood for 2 days at 35 °C. Colonies are matt, domed, scalloped-edged and do not demonstrate aerial hyphae. Growth occurs on Löwenstein–Jensen medium at 35 °C in less than 7 days, but no growth occurs at 42 °C; growth occurs on 5 % NaCl and on MacConkey's agar without crystal violet at 28 °C. None of the isolates grow in lysozyme or utilize citrate, and five of six (83 %) isolates produce arylsulfatase in 3 days. The semi-quantitative catalase activity of all isolates is reactive (>45 mm). All isolates utilize acetamide, reduce nitrate, exhibit iron uptake and produce urease and thermostable catalase. D-Fructose, D-glucose, i-myo-inositol, D-mannitol, D-mannose, salicin and D-trehalose are utilized as sole carbon sources, but adonitol, L-arabinose, cellobiose, dulcitol, i-erythritol, D-galactose, glycerol, lactose, maltose, melibiose, raffinose, L-rhamnose, D-sorbitol, starch, sucrose and D-xylose are not utilized. Acid is produced oxidatively from D-fructose, D-glucose, i-myo-inositol, D-mannitol, D-mannose, salicin and D-trehalose, but not from adonitol, L-arabinose, cellobiose, dulcitol, i-erythritol, D-galactose, glycerol, lactose, maltose, melibiose, raffinose, L-rhamnose, D-sorbitol, starch, sucrose or D-xylose. All of the isolates are resistant to ampicillin, cefotaxime, ceftriaxone and erythromycin, five of six (83 %) are resistant to doxycycline, minocycline and vancomycin and all are susceptible to amikacin, amoxycillin/clavulanate, ciprofloxacin, gentamicin, imipenem, sulfamethoxazole and trimethoprim/sulfamethoxazole. The type strain has a DNA G+C content of 64 mol%. The nearest phylogenetic neighbours, according to 16S rRNA gene sequence similarity, are M. neworleansense and all M. porcinum isolates studied (all 99·9 %).
The type strain, which was isolated from a leg wound, is W5998T (=ATCC 49935T=DSM 44677T). DNA relatedness values among the strains in the species and their relatedness to other strains of the M. fortuitum third biovariant complex and related species are available in IJSEM Online.
Description of Mycobacterium houstonense sp. nov.
Mycobacterium houstonense [hous'ton.en.se. N.L. neut. adj. houstonense pertaining to Houston, TX, USA, where the first isolate of the M. fortuitum third biovariant (sorbitol-positive) was identified].
The organisms are acid-fast, Gram-positive, pleomorphic bacilli. Long filamentous forms are often observed, but the organisms have no spores or capsules. Both isolates, under low-power stereomicroscopic examination, show white to slightly beige, small-diameter (approx. 1 mm) colonies after incubation on heart infusion agar with 5 % (v/v) rabbit blood for 2 days at 35 °C. Colonies are mucoid, convex, round and entire-edged and do not demonstrate aerial hyphae. Both grow on Löwenstein–Jensen medium at 35 and 42 °C in less than 7 days and grow on 5 % NaCl and on MacConkey's agar without crystal violet at 28 °C. Semi-quantitative catalase of both isolates is weakly reactive (<45 mm). Both have arylsulfatase activity by 3 days, utilize acetamide, reduce nitrate, exhibit iron uptake and produce urease and thermostable catalase. Both are unable to utilize citrate or grow in lysozyme. D-Fructose, D-glucose, i-myo-inositol, D-mannitol, D-mannose, D-sorbitol and D-trehalose are utilized as sole carbon sources, but adonitol, L-arabinose, cellobiose, dulcitol, i-erythritol, D-galactose, glycerol, lactose, maltose, melibiose, raffinose, L-rhamnose, salicin, starch, sucrose and D-xylose are not utilized. Acid is produced oxidatively from D-fructose, D-glucose, i-myo-inositol, D-mannitol, D-mannose, D-sorbitol and D-trehalose, but not from adonitol, L-arabinose, cellobiose, dulcitol, i-erythritol, D-galactose, glycerol, lactose, maltose, melibiose, raffinose, L-rhamnose, salicin, starch, sucrose or D-xylose. Both isolates are resistant to ampicillin, cefotaxime, ceftriaxone, doxycycline, erythromycin, minocycline and vancomycin, and both isolates are susceptible to amikacin, amoxycillin/clavulanate, ciprofloxacin, imipenem, sulfamethoxazole and trimethoprim/sulfamethoxazole. The type strain has a DNA G+C content of 64 mol%. The nearest phylogenetic neighbours, according to 16S rRNA gene sequence similarity, are M. fortuitum ATCC 6841T and M. senegalense ATCC 35796T.
The type strain, which was isolated from a face wound, is W5198T (=ATCC 49403T=DSM 44676T). DNA relatedness among the strains in the species and their relatedness to other strains of M. fortuitum third biovariant complex and related species are available in IJSEM Online.
Description of Mycobacterium neworleansense sp. nov.
Mycobacterium neworleansense (new.or.le.an.sen'se. N.L. neut. adj. neworleansense pertaining to New Orleans, LA, USA, the source of the type strain).
The organisms are acid-fast, Gram-positive, pleomorphic bacilli. Long filamentous forms are often observed, but spores and capsules are absent. Colonies are white to slightly beige and small in diameter (approx. 1 mm) after incubation on heart infusion agar with 5 % (v/v) rabbit blood for 2 days at 35 °C. Colonies are rough, wrinkled, irregular-edged and do not demonstrate aerial hyphae. Growth occurs on Löwenstein–Jensen medium at 35 °C in less than 7 days, but no growth occurs at 42 °C. Growth occurs on 5 % NaCl and on MacConkey's agar without crystal violet at 28 °C. Semi-quantitative catalase activity is positive (>45 mm). The isolate has arylsulfatase activity by 3 days, utilizes acetamide, reduces nitrate, exhibits iron uptake and produces urease and thermostable catalase. It does not utilize citrate or grow in lysozyme. Acid is produced oxidatively from i-erythritol, D-fructose, D-glucose, i-myo-inositol, D-mannitol, D-mannose, salicin, D-trehalose and D-xylose but not from adonitol, L-arabinose, cellobiose, dulcitol, D-galactose, glycerol, lactose, maltose, melibiose, raffinose, L-rhamnose, D-sorbitol, starch or sucrose. Resistant to ampicillin, cefotaxime, ceftriaxone, erythromycin and vancomycin, but susceptible to amikacin, amoxycillin/clavulanate, ciprofloxacin, doxycycline, imipenem, minocycline, sulfamethoxazole, trimethoprim/sulfamethoxazole and vancomycin. The type strain has a DNA G+C content of 60 mol%. The nearest phylogenetic neighbours, according to 16S rRNA gene sequence similarity, are Mycobacterium boenickei ATCC 49935T and all M. porcinum isolates studied.
The type strain, which was recovered from a scalp wound, is W6705T (=ATCC 49404T=DSM 44679T). DNA relatedness among the strains in the species and their relatedness to other strains of the M. fortuitum third biovariant complex and related species are available in IJSEM Online.
Description of Mycobacterium brisbanense sp. nov.
Mycobacterium brisbanense (bris.ban.en'se. N.L. neut. adj. brisbanense pertaining to Brisbane, Queensland, Australia, the source of the type strain).
The organisms are acid-fast, Gram-positive, pleomorphic bacilli. Long filamentous forms are often observed, but spores and capsules are absent. Colonies are white to slightly beige and small in diameter (approx. 1 mm) after incubation on heart infusion agar with 5 % (v/v) rabbit blood for 2 days at 35 °C. Colonies are mucoid, convex, round, entire-edged and do not demonstrate aerial hyphae. Growth occurs on Löwenstein–Jensen medium at 35 °C in less than 7 days, but no growth occurs at 42 °C. Growth occurs on 5 % NaCl and on MacConkey's agar without crystal violet at 28 °C. The isolate has arylsulfatase activity by 3 days, utilizes acetamide, reduces nitrate, produces urease and exhibits iron uptake. It does not utilize citrate or grow in lysozyme. Semi-quantitative catalase activity is weakly positive (<45 mm). It does not utilize citrate, grow in lysozyme or produce thermostable catalase. Adonitol, L-arabinose, i-erythritol, D-fructose, D-glucose, i-myo-inositol, D-mannitol, D-mannose, L-rhamnose, salicin, D-sorbitol and D-trehalose are utilized as sole carbon sources, but cellobiose, dulcitol, D-galactose, glycerol, lactose, maltose, melibiose, raffinose, starch, sucrose and D-xylose are not utilized. Produces acid oxidatively from adonitol, L-arabinose, i-erythritol, D-fructose, D-glucose, i-myo-inositol, D-mannitol, D-mannose, L-rhamnose, salicin, D-sorbitol and D-trehalose, but not from cellobiose, dulcitol, D-galactose, glycerol, lactose, maltose, melibiose, raffinose, starch, sucrose or D-xylose. Resistant to ampicillin, cefotaxime, ceftriaxone, doxycycline, erythromycin, minocycline and vancomycin, but susceptible to amikacin, amoxycillin/clavulanate, ciprofloxacin, imipenem, sulfamethoxazole and trimethoprim/sulfamethoxazole. The type strain has a DNA G+C content of 62 mol%. The nearest phylogenetic neighbour, according to 16S rRNA gene sequence similarity, is Mycobacterium diernhoferi ATCC 19340T.
The type strain, which was isolated from an antral sinus, is W6743T (=ATCC 49938T=DSM 44680T). DNA relatedness to other strains of the M. fortuitum third biovariant complex and related species are available in IJSEM Online.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Berd, D. (1973). Laboratory identification of clinically important aerobic actinomycetes. Appl Microbiol 25, 665–681.[Medline]
Brenner, D. J., McWhorter, A. C., Knutson, J. K. & Steigerwalt, A. G. (1982). Escherichia vulneris: a new species of Enterobacteriaceae associated with human wounds. J Clin Microbiol 15, 1133–1140.
Brenner, D. J., Hickman-Brenner, F. W., Lee, J. V. & 7 other authors (1983). Vibrio furnissii (formerly aerogenic biogroup of Vibrio fluvialis), a new species isolated from human feces and the environment. J Clin Microbiol 18, 816–824.
Brenner, D. J., Grimont, P. A., Steigerwalt, A. G., Fanning, G. R., Ageron, E. & Riddle, C. F. (1993). Classification of citrobacteria by DNA hybridization: designation of Citrobacter farmeri sp. nov., Citrobacter youngae sp. nov., Citrobacter braakii sp. nov., Citrobacter werkmanii sp. nov., Citrobacter sedlakii sp. nov., and three unnamed Citrobacter genomospecies. Int J Syst Bacteriol 43, 645–658.
Brosius, J., Palmer, M. L., Kennedy, P. J. & Noller, H. F. (1978). Complete nucleotide sequence of a 16S ribosomal RNA gene from Escherichia coli. Proc Natl Acad Sci U S A 75, 4801–4805.
Butler, W. R., Jost, K. C., Jr & Kilburn, J. O. (1991). Identification of mycobacteria by high-performance liquid chromatography. J Clin Microbiol 29, 2468–2472.
-DNA molecular size marker digested with EcoRI and HindIII; lanes 1 and 15 (b, c), 100 bp and pGem markers, respectively; lane 2, M. senegalense ATCC 35796T; 3, M. peregrinum ATCC 14467T; 4, M. fortuitum ATCC 6841T; 5, M. septicum ATCC 700731T; 6, M. porcinum ATCC BAA-328 (RT-A); 7, Mycobacterium genomospecies 1 (ATCC 49935T) (RT-B); 8, genomospecies 3 (ATCC 49404T) (RT-C); 9, M. porcinum ATCC 49939 (RT-D); 10, M. septicum ATCC BAA-329 (RT-E); 11, M. porcinum W6243 (RT-F); 12, M. porcinum W6236 (RT-G); 13, Mycobacterium genomospecies 2 (ATCC 69403T) (RT-H); and 14, Mycobacterium genomospecies 4 (ATCC 49938T) (RT-I). Molecular sizes for ribotyping are given in kb. Molecular sizes for PCR-RFLP analysis are given in bp.