- Email: [email protected]
Use of PCR-restriction fragment length polymorphism analysis of the hsp65 gene for rapid identification of mycobacteria in Brazil
Journal of Microbiological Methods 37 (1999) 223–229
Journal of Microbiological Methods
Use of PCR-restriction fragment length polymorphism analysis of the hsp65 gene for rapid identification of mycobacteria in Brazil ´ Magarinos Torres b , Adalgiza da Silva Rocha a , Cassiana da Costa Leite b , Helio a ´ Antonio Basilio de Miranda , Marcia Quinhones Pires Lopes a , Wim Maurits Degrave a , a, Philip Noel Suffys * a
Laboratory of Molecular Biology and Diagnosis of Infectious Diseases, Department for Biochemistry and Molecular Biology, Oswaldo Cruz Institute, Oswaldo Cruz Foundation, Av. Brasil 4365, Manguinhos 21045 -900, Rio de Janeiro, Brazil b Richet Medical Center, Rio de Janeiro, Brazil Received 30 November 1998; received in revised form 16 April 1999; accepted 19 April 1999
Abstract Polymerase chain reaction amplification of part of the gene coding for the heat shock protein hsp65 followed by restriction enzyme analysis (PRA) is a recently described tool for rapid identification of mycobacteria. In this study, the speed and simplicity of PRA for identification of isolates of mycobacteria from patients with clinical symptoms of tuberculosis was evaluated and compared with identification results obtained by commercially available methods. Established PRA patterns were observed for nineteen isolates of Mycobacterium tuberculosis, eleven belonging to the complex M. avium-intracellulare, four of M. kansasii, one of M. fortuitum, one of M. abscessus, three of M. gordonae and one of the recently described species M. lentiflavum, as identified by commercially available methods. Two isolates of M. fortuitum and one of M. gordonae had unique and so far undescribed PRA patterns, suggesting geographically-related intra-species variation within the hsp65 sequence. We propose the inclusion of these new patterns in the PRA identification algorithm and have defined more accurately the molecular weight values of the restriction fragments. This is the first report on the isolation of M. lentiflavum in Brazil suggesting that identification by means of PRA could be useful for detection of mycobacterial species that are usually unnoticed. Where the use of several commercial techniques in combination was necessary for correct identification, PRA demonstrated to be a simple technique with good cost–benefit for characterization of all mycobacterial isolates in this study. 1999 Elsevier Science B.V. All rights reserved. Keywords: Identification; Mycobacteria; PRA
1. Introduction Identification of mycobacteria up to the species level is necessary for application of adequate drug *Corresponding author. Tel.: 1 55-21-598-4289; fax: 1 55-21270-9997. E-mail address: [email protected] (P.N. Suffys)
therapy and to address epidemiological questions. Most laboratories identify isolates from the genus Mycobacterium to the species level by analysis of the phenotypic and biochemical characteristics of the organism using culture in solid media, which is a time-consuming process. Use of recent culture procedures such as the BACTEC and biphasic MBCheck systems speeds up the identification process
0167-7012 / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0167-7012( 99 )00062-7
224
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
but a significant growth of the microorganisms is still needed (Abe et al., 1992). Alternative methods based on chromatography of lipid components using TLC, HPLC and GLC are either not sufficiently discriminatory or require sophisticated equipment (Glickman et al., 1994). Several identification methods based on the detection of specific DNA patterns or sequences through hybridization with commercially available probes, commercial or ‘in house’ species-specific PCR or genus-specific PCR followed by differential hybridization steps have been developed but these methods generally only differentiate a limited number of mycobacterial species or are difficult to use on a routine basis (De Beenhouwer et al., 1995; Kox et al., 1997). The most sensitive technique for identification of a large number of mycobacterial species is sequencing of a fragment of conserved genes such as the 16S rDNA (Rogall et al., 1990), the 32 kD antigen (Soini et al., 1994), superoxide dismutase (Zolg and Philippi-Schultz, 1994) and hsp65 (Kapur et al., 1995); sequencing however is expensive and demands technical expertise. An attractive alternative is Mycobacterium genus specific amplification, followed by restriction enzyme analysis of the amplicon, using either the hsp65 (PRA; Plikaytis et al., 1992; Telenti et al., 1993) or the 16S rDNA region (ARDRA; Dobner et al., 1996; Vaneechoutte et al., 1993) as target sequence. The latter technique is technically less demanding and identification by PRA has been compared with identification by conventional biochemical tests for diagnostic (Devallois et al., 1997b; Taylor et al., 1997) and taxonomic purposes (Plikaytis et al., 1992; Telenti et al., 1993), for characterization of new species (Domenech et al., 1997; Springer et al., 1996) or for better characterization of already defined mycobacterial species (Picardeau et al., 1997). Here we report the use of PRA as an alternative for commercially available identification methods of mycobacteria.
2. Materials and methods
al., 1992). After verification of PRA patterns of reference strains, the extraction procedure from solid or liquid cultures was simplified: 50 ml of a liquid culture or a loopfull of mycobacterial mass from solid cultures was resuspended in 0.5 ml of 10 mM Tris–HCl, 1 mM EDTA, 1% Triton X-100 and submitted to three cycles of freeze–boiling (5 min, 708C; 10 min, 1008C). The supernatant was stored at 2 208C and 2 ml of this mixture was tested in the PCR assay.
2.2. Patients and clinical samples Different kinds of clinical samples from patients with suspicion of tuberculosis or with mycobacterial infection were used including sputum (17), induced sputum (10), lung biopsy (1), liver biopsy (1), broncoalveolar lavage (9), cerebrospinal fluid (1), pleural liquid (1), blood (1), cervical ganglio (1) and bone marrow aspirate (1). Non-sterile samples were liquefied and decontaminated with N-acetyl-L-cysteine–sodium hydroxide (Kubica et al., 1963); sterile samples were inoculated directly in MB / Bact Processbottles (Organon Teknica), Lowenstein-Jensen or 7H11 agar for detection and identification, except for skin biopsy and cervical ganglion which were triturated and homogenized, and for blood and bone marrow aspirate, which were lysed and centrifuged in Isolator tubes (Wampole Laboratories, Cranbury, NJ, USA) before inoculation.
2.3. Commercial methods for identification of mycobacteria Mycobacterial isolates were identified either as M. tuberculosis by the Roche Amplicor M. tuberculosis PCR test (Roche Diagnostics, Basel, Switzerland) or by the Accuprobe system (Gen-Probe, San Diego, CA, USA) or as MAC, M. kansasii or M. gordonae by the Accuprobe system. Mycobacterial samples that were not identified by any of these systems were sent to Specialty Laboratories (Santa Monica, CA, USA) for analysis of mycolic acids using a Beckman System Gold HPLC instrument and patterns recognition with the Infometrix PIROUETTE software.
2.1. Culture of mycobacterial strains and DNA extraction
2.4. PCR conditions and restriction enzyme analysis
High quality DNA from mycobacterial reference cultures was prepared as described earlier (Santos et
A 10-ng amount of purified mycobacterial DNA or 2 ml of supernatant were added to 50 ml PCR
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
reaction mixture containing 50 mM KCl (pH 8.3); 1.5 mM MgCl 2 ; 10% glycerol; 200 mM of each dNTP; 0.5 mM of the primers Tb11 and Tb12 (24) and 1.25 U of Taq polymerase (Amersham, UK). The amplification procedure was modified from an earlier protocol (Telenti et al., 1993) and consisted of 45 cycles of 1 min at 948C; 1 min at 658C and 1 min at 728C, followed by a final extension step at 728C for 7 min. A 10-ml volume of the PCR reaction mixture were analyzed by gel electrophoresis and staining with ethidium bromide and when positive, 15 ml was digested with 10 U BspI (shizomer of HaeIII, produced in our laboratory) or 6 U BstEII (New England Biolabs, Beverly, MA, USA) under mineral oil (Sigma, St. Louis, MO, USA). Analysis of restriction enzyme products was done by gel electrophoreses on a 5% NuSieve GTG agarose gel (FMC BioProducts, Rockland, ME, USA), using a commercial 50 or 25 bp DNA ladder (Gibco BRL, Gaithersburg, MD, USA) as molecular weight marker.
2.5. Computer analysis Besides visual analysis of the digest, DNA patterns were submitted to computer analysis. Electrophoresis and photograph conditions were standardized to assure reproducibility within and between gels. Photographs were scanned with a HPScanJet scanner and the images introduced into the GELCOMPAR software (Version 4.0; Applied Maths, Kortrijk, Belgium). Normalization was performed using two external reference markers per gel and a library of DNA patterns of reference strains was constructed.
2.6. Analysis of sequences in GenBank For determination of exact sizes of restriction fragments generated by the PRA system, restriction maps were constructed in the following way: the 441 bp fragment localized between the primers Tb11 and Tb12 from the sequence MbaA from M.bovis BCG P3 (GenBank accession nr. M17705) was compared with the GenBank database (release 105.0) by FastA (Wisconsin Package, Version 9.1, Genetics Computer Group). The 240 sequences with highest similarity were extracted and verified for the presence of Tb11 and Tb12 using Findpatterns, permitting a mismatch
225
of maximum three nucleotides per primer. Twenty sequences were retrieved, fifteen being part of the 65 kDa antigen of nine different mycobacterial species, three sequences of unknown origin, part of an open reading frame of Streptomyces albus and part of the Tsukamurella tyrosinosolvenshsp60 gene. Most sequences had perfect primer matching except for M. leprae and S. albus having one mismatch in each primer and for T. tyrosinosolvens, showing two mismatches in each primer. Restriction maps were constructed for HaeIII and BstEII using Map and restriction fragment sizes determined manually.
3. Results and discussion Diagnosis of mycobacterial diseases by PCR followed by restriction analysis was first described for differentiation of species belonging to the M. tuberculosis complex from those belonging to the M. avium complex (Bollet et al., 1992; Rodrigo et al., 1992), and later on adapted for identification of medically relevant and frequently encountered laboratory isolates (Plikaytis et al., 1992; Telenti et al., 1993; Vaneechoutte et al., 1993). Recently, 34 different mycobacterial species were identified performing PRA of the hsp65 gene (Zolg and Philippi-Schultz, 1994) and the method was shown to be reliable for rapid identification of mycobacteria growing in BACTEC media (Taylor et al., 1997). To get familiar with the technique and to pinpoint steps that could increase reproducibility, PRA was firstly performed on purified DNA from mycobacterial reference strains. The most frequent observed problem was partial restriction digestion with BstEII, an enzyme that has optimal activity at 658C, a drawback that was overcome by addition of a small amount of mineral oil. The PRA patterns obtained with reference strains were identical as those observed by Telenti et al. (Telenti et al., 1993) except for an extra band of 60 bp in the HaeIII digest of M. gordonae, M. chelonae and M. fortuitum (Table 1), a band that was also observed in clinical / ambiental isolates of these species (Fig. 1). Identification algorithms that were constructed recently report the presence of an 60 bp band in the HaeIII digest of M. gordonae but not of M. fortuitum and M. chelonae (Devallois et al., 1997b; Taylor et al., 1997) and low-molecular-weight bands have been reported in a
226
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
Table1 Molecular weight values of the restriction patterns generated by PRA of reference strains of mycobacteria and proposal of new PRA patterns as obtained from clinical isolates Species (origin)
PRA pattern
M. tuberculosis (ATCC 27294)
M. tuberculosis
M. intracellulare (ATCC 13950)
M. intracellulare
M. fortuitum (ATCC 6841)
M. fortuitum
M. avium (ATCC 25291)
M. avium
M. kansasii (ATCC 12478)
M. kansasii I
M. gordonae (ATCC 14470)
M. gordonae I
M. gordonae ureolyticum
M. gordonae III
M. chelonae (NCTC 946)
M. chelonae I
M. lentiflavum
M. lentiflavum
M. gordonae VII d M. fortuitum III d
New New
Molecular weight values BstEII
HaeIII
235 / 115 / 85 233 / 114 / 79 245 / 125 / 80 235 / 115 / 100 233 / 114 / 94 245 / 125 / 100 235 / 115 / 85 233 / 208 245 / 125 / 80 235 / 210 233 / 208 245 / 220 235 / 210 245 / 220 235 / 115 / 85 245 / 125 / 80 235 / 115 / 100 245 / 125 / 100 320 / 130 325 / 140 440 Undigested 235 / 115 / 100 235 / 115 / 85
150 / 125 / 70 152 / 127 / 69 a 160 / 140 / 70 b 140 / 125 / 60 145 / 127 / 59 a 155 / 140 / 60 b 140 / 120 / 60 139 / 123 / 58 a 155 / 135 b 125 / 100 127 / 103 a 140 / 105 b 130 / 105 / 80 140 / 105 / 70 b 155 / 110 / 60 170 / 115 b 125 / 110 140 / 120 b 195 / 60 210 b 140 / 120 150 / 135 c 155 / 110 135 / 90 / 85
a As obtained by construction of a restriction map from the GenBank sequences. GenBank accession numbers are M15467 (M. tuberculosis), U55829 (M. intracellulare), U55834 (M. fortuitum) and U55826 (M. avium). b As described by Telenti et al. (1993). c As described by Springer et al. (1996). d Identity of strains was confirmed by evaluation of growth rate, pigmentation, nitrate reduction, arylsulfatase, iron uptake and growth on NaCl.
Fig. 1. PRA patterns of M. lentiflavum (1), M. abcessus (2), M. fortuitum (3 and 4) and M. gordonae (5) after digestion with BstEII (A) or BspRI (B). The molecular weight marker (M) is a 50 bp ladder.
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
study concentrating on rapidly growing mycobacteria (Steingrube et al., 1995). Consideration of such bands is not only academical because the presence of a single band in this molecular-weight range differentiates M. kansasii from M. avium (Devallois et al., 1997a) and M. smegmatis from M. margaritense (Domenech et al., 1997). As observed in other laboratories (Devallois et al., 1997b; Taylor et al., 1997; Bollet et al., 1992), generally smaller molecular-weight values were dedicated to the bands when compared to those in the algorithm presented by Telenti et al. (1993), a fact that could be due to differences in running conditions, agarose types and computer algorithms used for calculation of molecular weight values. In this study, molecular-weight values were calculated by position regression by experimental curve fitting upon comparison with 50 and 25 bp ladders and compared with theoretical values obtained after construction of HaeIII and BstEII restriction maps of sequences obtained from GenBank (Table 1). Although definition of exact molecular values is not important for identification of an unknown pattern by comparison with reference patterns or with a local database, it could be important for comparison of identification data generated in different laboratories. In a second stage, PRA was performed on bacterial isolates from patients with suspicion of mycobacterial infection. In total, 43 clinical isolates from 42 different individuals were identified, with species belonging to the M. avium-intracellulare complex being the most frequently isolated atypical mycobacteria, confirming earlier observations (Barreto et al., 1993). When comparing identification results obtained with commercial methods with those obtained by PRA, all but three isolates had corresponding patterns. All 19 isolates of M. tuberculosis and nine isolates of M. avium were represented by an easily recognizable unique PRA pattern, confirming that also in Brazil, PRA is a simple tool for rapid differentiation between the two most frequently encountered mycobacterial species. Strain heterogeneity within the complex MAIS is responsible for at least five additional PRA patterns (Devallois et al., 1997a) but the two isolates identified here as MAIS by the GenProbe system demonstrated the PRA pattern that was originally described for M. intracellulare by Telenti et al. (1993). We have preliminary
227
data however that, for correct identification of M.intracellulare in Brazil, just as is the case in the Caribbean Isles and the Indian subcontinent, a larger number of PRA patterns will have to be considered [unpublished]. From the three isolates that were identified by HPLC as M. fortuitum, two were from the same patient taken with a 5-month interval and had a so far undescribed PRA pattern (Fig. 1); the other was characterised by the M. fortuitum I pattern. Two of the four strains identified as M. gordonae also had a so far undescribed PRA pattern (Fig. 1), while two other strains had respectively the M. gordonae I and III pattern. Having in mind that we analyzed a relative small number of strains while in earlier studies, PRA was performed on a large number of strains, we feel that a predominance of certain strains with a particular PRA patterns could exist in some geographic areas. We propose the inclusion of these new patterns that we would like to call M. fortuitum III and M. gordonae VII in the PRA identification algorithm that was recently constructed by Devallois et al. (1997b) (Table 1). Although genetic variation has also been reported for M. kansasii (Alcaide et al., 1997), we encountered the M. kansasii I pattern in four different isolates. One isolate that was identified as M. abscessus by HPLC had a pattern similar to the recently described M. abscessus II by Devallois et al. (1997b), except for an extra band of 70 bp (Fig. 1). Double bands of 60 and 70 bp have consistently been observed in isolates of the M. chelonae complex, collected in another setting (our unpublished observations) and we therefore feel that this band should be added in the algorithm. Presence of an 70 bp fragment has also been observed in some isolates of M. gordonae (pattern IV), normally characterized by a 60 bp band (Taylor et al., 1997). Whether these differences are geographic peculiarities has to be determined. Finally, one isolate was characterized as the recently described species M. lentiflavum, an organism with biochemical characteristics of M. avium and a 16S rRNA sequence closely related to that of M. simiae and M. genavense (Springer et al., 1996). This species was described for the first time in 1996 and has so far been reported in Italy, Switzerland and Germany only (Springer et al., 1996; Tortoli et al., 1997; Haase et al., 1997). The PRA patterns of the strain detected here was identical to one of the three
228
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
earlier published patterns (Rodrigo et al., 1992). Although isolates reported in literature were mostly fortuitous or traced to contaminated bronchoscopes, some strains were isolated from sterile tissue (Springer et al., 1996; Haase et al., 1997) and association between transition from colonization to disease and decreasing CD4-level of immonocompromised patients was observed (Tortoli et al., 1997). Whether the isolation of M. lentiflavum in this case was associated with disease is not known because the organism was isolated from sputum of a patient with suspicion of pulmonary tuberculosis. In conclusion, PRA has proven to be a fast and relatively simple method for identification of mycobacteria, to an extent which is normally not obtained by routine procedures. The fact that three isolates were characterized by new pattern suggests that particular strains could be circulating in certain geographic areas, meaning that a sufficient number of isolates have to be analyzed before using PRA as a single identification method. After submission of the manuscript, we became aware of a compilation of PRA patterns available through the internet at the following address: http:\\www.hospvd.ch:8005.
Acknowledgements This work received financial support from the CNPq and CBAB. We thank the National Reference ´ Center of Tuberculosis Prof. Helio Fraga (Curicica, Rio de Janeiro, Brazil) for providing us with some of the mycobacterial reference strains and Marlei Gomes da Silva and Leila de Souza Fonseca from the Federal University of Rio de Janeiro for helping us performing the conventionel identification of M. fortuitum and M. gordonae.
References Abe, C., Hosojima, S., Fukasawa, Y., Kazumi, Y., Takahashi, M., Hirano, K., Mori, T., 1992. Comparison of MB-Check, BACTEC, and egg-based media for recovery of mycobacteria. J. Clin. Microbiol. 30, 878–881. Alcaide, F., Richter, I., Bernasconi, C., Springer, B., Hagenau, C., Schultze-Robbecke, R., Tortoli, E., Martin, R., Bottger, E.C., Telenti, A., 1997. Heterogeneity and clonality among isolates
of Mycobacterium kansasii: implications for epidemiological and pathogenicity studies. J. Clin. Microbiol. 35, 1959–1964. Barreto, J.A., Palaci, M., Ferrazoli, L., Martins, M.C., Suleiman, J., Lorenco, J.R., Ferreira, O.C., Riley, Jr.L. W., Johnsonn, W.D. Jr., Galvao, P.A., 1993. Isolation of Mycobacterium avium complex bone marrow aspirates of AIDS patients in Brazil. J. Infect. Dis. 168, 777–779. Bollet, C., De Lamballerie, X., Zandotti, C., Vignoli, C., Gevaudan, M.J., De Micco, P., 1992. Detection and identification of Mycobacterium tuberculosis, M. bovis BCG, and M. avium by two-step polymerase chain reaction. Comparison with ELISA using A60 antigen. Microbiologica 15, 345–350. De Beenhouwer, H., Liang, Z., DeRijk, P., VanEekeren, C., Portaels, F., 1995. Detection and identification of mycobacteria by DNA amplification and oligonucleotide-specific capture plate hybridization. J. Clin. Microbiol. 33, 2994–2998. Devallois, A., Picardeau, M., Paramasivan, C.N., Vincent, V., Rastogi, N., 1997a. Molecular characterization of Mycobacterium avium complex isolates giving disconcordant results in AccuProbe tests by PCR-restriction enzyme analysis, 16S rRNA gene sequencing, and DT1-DT6 PCR. J. Clin. Microbiol. 35, 2767–2772. Devallois, A., SengGoh, K., Rastogi, N., 1997b. Rapid identification of mycobacteria to species level by PCR-restriction fragment length polymorphism analysis of the hsp65 gene and proposition of an algorithm to differentiate 34 mycobacterial species. J. Clin. Microbiol. 35, 2969–2973. Dobner, P., Feldman, K., Rifai, M., Loschner, T., Rinder, H., 1996. Rapid identification of mycobacterial species by PCR amplification of hypervariable 16 rRNA gene promoter region. J. Clin. Microbiol. 34, 866–869. Domenech, P., Jimenez, M.S., Menendez, M.C., Bull, T.J., Samper, S., Manrique, A., Garcia, M.J., 1997. Mycobacterium mageritense sp. nov. Int. J. Syst. Bacteriol. 47, 535–540. Glickman, S.E., Kilburn, J.O., Butler, W.R., Ramos, L.C., 1994. Rapid identification of mycolic acid patterns of mycobacteria by high-performance liquid chromatography using pattern recognition software and a Mycobacterium library. J. Clin. Microbiol. 32, 740–745. Haase, G., Kentrup, H., Skopnik, H., Springer, B., Bottger, E.C., 1997. Mycobacterium lentiflavum: an etiologic agent of cervical lymphadenitis. Clin. Infect. Dis. 25, 1245–1246. Kapur, V., Li, L.-L., Hamrick, M.R., Plikyatis, B.B., Shinnick, T.M., Telenti, A., Jacobs, W.R., Banerjee, A., Cole, S., Yuen, K.Y., Clarridge, J.E., Kreiswirth, B.N., Musser, J.M., 1995. Rapid Mycobacterium species assignment and unambiguous identification of mutations associated with antimicrobial resistance in Mycobacterium tuberculosis by automated sequencing. Arch. Pathol. Lab. Med. 119, 131–138. Kox, L.F.F., Jansen, H.M., Kuijper, S., Kolk, A.H.J., 1997. Multiplex PCR assay for immediate identification of the infecting species in patients with mycobacterial disease. J. Clin. Microbiol. 35, 1492–1498. Kubica, G.P., Dye, W.E., Cohen, M.L., Middlebrook, G., 1963. Sputum digestion and concentration with N-acetyl-L-cysteinesodium hydroxide for culture of mycobacteria. Am. Rev. Respir. Dis. 87, 775. Picardeau, M., Prod’hom, G., Raskine, L., LePennec, M.P.,
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229 Vincent, V., 1997. Genotypic differentiation of five subspecies of Mycobacterium kansasii. J. Clin. Microbiol. 35, 25–32. Plikaytis, B.B., Plikaytis, B.D., Yakrus, M.A., Butler, W.R., Woodley, C.L., Silcox, V.A., Shinnick, T.M., 1992. Differentiation of slowly growing Mycobacterium species including Mycobacterium tuberculosis, by gene amplification and restriction fragment length polymorphism analysis. J. Clin. Microbiol. 30, 1815–1822. Rodrigo, G., Kallenius, G., Hoffmann, E., Svenson, S.B., 1992. Diagnosis of mycobacterial infections by PCR and restriction enzyme analysis. Lett. Appl. Microbiol. 15, 41–44. Rogall, T., Flohr, T., Bottger, E.C., 1990. Differentiation of Mycobacterium species by direct sequencing of amplified DNA. J. Gen. Microbiol. 136, 1915–1920. Santos, A.R., De Miranda, A.B., Lima, L.M., Suffys, P.N., Degrave, W.M., 1992. Method for high yield preparation in large and small scale of nucleic acids from mycobacteria. J. Microbiol. Methods 28, 1236–1243. Soini, H., Bottger, E., Viljanen, M.K., 1994. Identification of mycobacteria by PCR-based sequence determination of the 32-kilodalton protein gene. J. Clin. Microbiol. 32, 2944–2947. Springer, B., Whei-Kuo, W., Bodmer, T., Haase, G., Pfyffer, G.E., Kroppenstedt, R.M., Schroder, K.-H., Elmer, S.J., Kilburn, O., Kirchner, P., Telenti, A., Coyle, M.B., Bottger, E.C., 1996. Isolation and characterization of a unique group of slowly growing mycobacteria: description of Mycobacterium lentiflavum sp. nov. J. Clin. Microbiol. 34, 1100–1107.
229
Steingrube, V.A., Gibson, J.L., Brown, B.A., Zhang, Y., Wilson, R.W., Rajagopalan, M., WallaceJr, J., 1995. PCR amplification and restriction endonuclease analysis of a 65-kilodalton heat shock protein gene sequence for taxonomic separation of rapidly growing mycobacteria. J. Clin. Microbiol. 33, 149–153. Taylor, T.B., Patterson, C., Hale, Y., Safranek, W.W., 1997. Routine use of PCR-restriction fragment length polymorphism analysis for identification of mycobacteria in liquid media. J. Clin. Microbiol. 35, 79–85. Telenti, A., Marchesi, F., Balz, M., Bally, F., Bottger, E., Bodmer, T., 1993. Rapid identification of mycobacteria to the species level by polymerase chain reaction and restriction enzyme analysis. J. Clin. Microbiol. 31, 175–178. Tortoli, E., Piersimoni, C., Kirchner, P., Bartoloni, A., Burrini, C., Lachini, C., Mantella, A., Muzzi, G., PasseriniTosi, C., Penati, V., Scarparo, C., TuliaSimonetti, M., Bottger, E.C., 1997. Characterization of mycobacterial isolates phylogenetically related to, but different from, Mycobacterium simiae. J. Clin. Microbiol. 35, 697–702. Vaneechoutte, M., de Beenhouwer, H., Claeys, G., Verschraegen, G., de Rouck, A., Paepe, N., Elaichouni, A., Portaels, F., 1993. Identification of Mycobacterium species using amplified ribosomal DNA restriction analysis. J. Clin. Microbiol. 31, 2061–2065. Zolg, J.W., Philippi-Schultz, S., 1994. The superoxide dismutase gene, a target for detection and identification of mycobacteria by PCR. J. Clin. Microbiol. 32, 2801–2812.
Journal of Microbiological Methods
Use of PCR-restriction fragment length polymorphism analysis of the hsp65 gene for rapid identification of mycobacteria in Brazil ´ Magarinos Torres b , Adalgiza da Silva Rocha a , Cassiana da Costa Leite b , Helio a ´ Antonio Basilio de Miranda , Marcia Quinhones Pires Lopes a , Wim Maurits Degrave a , a, Philip Noel Suffys * a
Laboratory of Molecular Biology and Diagnosis of Infectious Diseases, Department for Biochemistry and Molecular Biology, Oswaldo Cruz Institute, Oswaldo Cruz Foundation, Av. Brasil 4365, Manguinhos 21045 -900, Rio de Janeiro, Brazil b Richet Medical Center, Rio de Janeiro, Brazil Received 30 November 1998; received in revised form 16 April 1999; accepted 19 April 1999
Abstract Polymerase chain reaction amplification of part of the gene coding for the heat shock protein hsp65 followed by restriction enzyme analysis (PRA) is a recently described tool for rapid identification of mycobacteria. In this study, the speed and simplicity of PRA for identification of isolates of mycobacteria from patients with clinical symptoms of tuberculosis was evaluated and compared with identification results obtained by commercially available methods. Established PRA patterns were observed for nineteen isolates of Mycobacterium tuberculosis, eleven belonging to the complex M. avium-intracellulare, four of M. kansasii, one of M. fortuitum, one of M. abscessus, three of M. gordonae and one of the recently described species M. lentiflavum, as identified by commercially available methods. Two isolates of M. fortuitum and one of M. gordonae had unique and so far undescribed PRA patterns, suggesting geographically-related intra-species variation within the hsp65 sequence. We propose the inclusion of these new patterns in the PRA identification algorithm and have defined more accurately the molecular weight values of the restriction fragments. This is the first report on the isolation of M. lentiflavum in Brazil suggesting that identification by means of PRA could be useful for detection of mycobacterial species that are usually unnoticed. Where the use of several commercial techniques in combination was necessary for correct identification, PRA demonstrated to be a simple technique with good cost–benefit for characterization of all mycobacterial isolates in this study. 1999 Elsevier Science B.V. All rights reserved. Keywords: Identification; Mycobacteria; PRA
1. Introduction Identification of mycobacteria up to the species level is necessary for application of adequate drug *Corresponding author. Tel.: 1 55-21-598-4289; fax: 1 55-21270-9997. E-mail address: [email protected] (P.N. Suffys)
therapy and to address epidemiological questions. Most laboratories identify isolates from the genus Mycobacterium to the species level by analysis of the phenotypic and biochemical characteristics of the organism using culture in solid media, which is a time-consuming process. Use of recent culture procedures such as the BACTEC and biphasic MBCheck systems speeds up the identification process
0167-7012 / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0167-7012( 99 )00062-7
224
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
but a significant growth of the microorganisms is still needed (Abe et al., 1992). Alternative methods based on chromatography of lipid components using TLC, HPLC and GLC are either not sufficiently discriminatory or require sophisticated equipment (Glickman et al., 1994). Several identification methods based on the detection of specific DNA patterns or sequences through hybridization with commercially available probes, commercial or ‘in house’ species-specific PCR or genus-specific PCR followed by differential hybridization steps have been developed but these methods generally only differentiate a limited number of mycobacterial species or are difficult to use on a routine basis (De Beenhouwer et al., 1995; Kox et al., 1997). The most sensitive technique for identification of a large number of mycobacterial species is sequencing of a fragment of conserved genes such as the 16S rDNA (Rogall et al., 1990), the 32 kD antigen (Soini et al., 1994), superoxide dismutase (Zolg and Philippi-Schultz, 1994) and hsp65 (Kapur et al., 1995); sequencing however is expensive and demands technical expertise. An attractive alternative is Mycobacterium genus specific amplification, followed by restriction enzyme analysis of the amplicon, using either the hsp65 (PRA; Plikaytis et al., 1992; Telenti et al., 1993) or the 16S rDNA region (ARDRA; Dobner et al., 1996; Vaneechoutte et al., 1993) as target sequence. The latter technique is technically less demanding and identification by PRA has been compared with identification by conventional biochemical tests for diagnostic (Devallois et al., 1997b; Taylor et al., 1997) and taxonomic purposes (Plikaytis et al., 1992; Telenti et al., 1993), for characterization of new species (Domenech et al., 1997; Springer et al., 1996) or for better characterization of already defined mycobacterial species (Picardeau et al., 1997). Here we report the use of PRA as an alternative for commercially available identification methods of mycobacteria.
2. Materials and methods
al., 1992). After verification of PRA patterns of reference strains, the extraction procedure from solid or liquid cultures was simplified: 50 ml of a liquid culture or a loopfull of mycobacterial mass from solid cultures was resuspended in 0.5 ml of 10 mM Tris–HCl, 1 mM EDTA, 1% Triton X-100 and submitted to three cycles of freeze–boiling (5 min, 708C; 10 min, 1008C). The supernatant was stored at 2 208C and 2 ml of this mixture was tested in the PCR assay.
2.2. Patients and clinical samples Different kinds of clinical samples from patients with suspicion of tuberculosis or with mycobacterial infection were used including sputum (17), induced sputum (10), lung biopsy (1), liver biopsy (1), broncoalveolar lavage (9), cerebrospinal fluid (1), pleural liquid (1), blood (1), cervical ganglio (1) and bone marrow aspirate (1). Non-sterile samples were liquefied and decontaminated with N-acetyl-L-cysteine–sodium hydroxide (Kubica et al., 1963); sterile samples were inoculated directly in MB / Bact Processbottles (Organon Teknica), Lowenstein-Jensen or 7H11 agar for detection and identification, except for skin biopsy and cervical ganglion which were triturated and homogenized, and for blood and bone marrow aspirate, which were lysed and centrifuged in Isolator tubes (Wampole Laboratories, Cranbury, NJ, USA) before inoculation.
2.3. Commercial methods for identification of mycobacteria Mycobacterial isolates were identified either as M. tuberculosis by the Roche Amplicor M. tuberculosis PCR test (Roche Diagnostics, Basel, Switzerland) or by the Accuprobe system (Gen-Probe, San Diego, CA, USA) or as MAC, M. kansasii or M. gordonae by the Accuprobe system. Mycobacterial samples that were not identified by any of these systems were sent to Specialty Laboratories (Santa Monica, CA, USA) for analysis of mycolic acids using a Beckman System Gold HPLC instrument and patterns recognition with the Infometrix PIROUETTE software.
2.1. Culture of mycobacterial strains and DNA extraction
2.4. PCR conditions and restriction enzyme analysis
High quality DNA from mycobacterial reference cultures was prepared as described earlier (Santos et
A 10-ng amount of purified mycobacterial DNA or 2 ml of supernatant were added to 50 ml PCR
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
reaction mixture containing 50 mM KCl (pH 8.3); 1.5 mM MgCl 2 ; 10% glycerol; 200 mM of each dNTP; 0.5 mM of the primers Tb11 and Tb12 (24) and 1.25 U of Taq polymerase (Amersham, UK). The amplification procedure was modified from an earlier protocol (Telenti et al., 1993) and consisted of 45 cycles of 1 min at 948C; 1 min at 658C and 1 min at 728C, followed by a final extension step at 728C for 7 min. A 10-ml volume of the PCR reaction mixture were analyzed by gel electrophoresis and staining with ethidium bromide and when positive, 15 ml was digested with 10 U BspI (shizomer of HaeIII, produced in our laboratory) or 6 U BstEII (New England Biolabs, Beverly, MA, USA) under mineral oil (Sigma, St. Louis, MO, USA). Analysis of restriction enzyme products was done by gel electrophoreses on a 5% NuSieve GTG agarose gel (FMC BioProducts, Rockland, ME, USA), using a commercial 50 or 25 bp DNA ladder (Gibco BRL, Gaithersburg, MD, USA) as molecular weight marker.
2.5. Computer analysis Besides visual analysis of the digest, DNA patterns were submitted to computer analysis. Electrophoresis and photograph conditions were standardized to assure reproducibility within and between gels. Photographs were scanned with a HPScanJet scanner and the images introduced into the GELCOMPAR software (Version 4.0; Applied Maths, Kortrijk, Belgium). Normalization was performed using two external reference markers per gel and a library of DNA patterns of reference strains was constructed.
2.6. Analysis of sequences in GenBank For determination of exact sizes of restriction fragments generated by the PRA system, restriction maps were constructed in the following way: the 441 bp fragment localized between the primers Tb11 and Tb12 from the sequence MbaA from M.bovis BCG P3 (GenBank accession nr. M17705) was compared with the GenBank database (release 105.0) by FastA (Wisconsin Package, Version 9.1, Genetics Computer Group). The 240 sequences with highest similarity were extracted and verified for the presence of Tb11 and Tb12 using Findpatterns, permitting a mismatch
225
of maximum three nucleotides per primer. Twenty sequences were retrieved, fifteen being part of the 65 kDa antigen of nine different mycobacterial species, three sequences of unknown origin, part of an open reading frame of Streptomyces albus and part of the Tsukamurella tyrosinosolvenshsp60 gene. Most sequences had perfect primer matching except for M. leprae and S. albus having one mismatch in each primer and for T. tyrosinosolvens, showing two mismatches in each primer. Restriction maps were constructed for HaeIII and BstEII using Map and restriction fragment sizes determined manually.
3. Results and discussion Diagnosis of mycobacterial diseases by PCR followed by restriction analysis was first described for differentiation of species belonging to the M. tuberculosis complex from those belonging to the M. avium complex (Bollet et al., 1992; Rodrigo et al., 1992), and later on adapted for identification of medically relevant and frequently encountered laboratory isolates (Plikaytis et al., 1992; Telenti et al., 1993; Vaneechoutte et al., 1993). Recently, 34 different mycobacterial species were identified performing PRA of the hsp65 gene (Zolg and Philippi-Schultz, 1994) and the method was shown to be reliable for rapid identification of mycobacteria growing in BACTEC media (Taylor et al., 1997). To get familiar with the technique and to pinpoint steps that could increase reproducibility, PRA was firstly performed on purified DNA from mycobacterial reference strains. The most frequent observed problem was partial restriction digestion with BstEII, an enzyme that has optimal activity at 658C, a drawback that was overcome by addition of a small amount of mineral oil. The PRA patterns obtained with reference strains were identical as those observed by Telenti et al. (Telenti et al., 1993) except for an extra band of 60 bp in the HaeIII digest of M. gordonae, M. chelonae and M. fortuitum (Table 1), a band that was also observed in clinical / ambiental isolates of these species (Fig. 1). Identification algorithms that were constructed recently report the presence of an 60 bp band in the HaeIII digest of M. gordonae but not of M. fortuitum and M. chelonae (Devallois et al., 1997b; Taylor et al., 1997) and low-molecular-weight bands have been reported in a
226
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
Table1 Molecular weight values of the restriction patterns generated by PRA of reference strains of mycobacteria and proposal of new PRA patterns as obtained from clinical isolates Species (origin)
PRA pattern
M. tuberculosis (ATCC 27294)
M. tuberculosis
M. intracellulare (ATCC 13950)
M. intracellulare
M. fortuitum (ATCC 6841)
M. fortuitum
M. avium (ATCC 25291)
M. avium
M. kansasii (ATCC 12478)
M. kansasii I
M. gordonae (ATCC 14470)
M. gordonae I
M. gordonae ureolyticum
M. gordonae III
M. chelonae (NCTC 946)
M. chelonae I
M. lentiflavum
M. lentiflavum
M. gordonae VII d M. fortuitum III d
New New
Molecular weight values BstEII
HaeIII
235 / 115 / 85 233 / 114 / 79 245 / 125 / 80 235 / 115 / 100 233 / 114 / 94 245 / 125 / 100 235 / 115 / 85 233 / 208 245 / 125 / 80 235 / 210 233 / 208 245 / 220 235 / 210 245 / 220 235 / 115 / 85 245 / 125 / 80 235 / 115 / 100 245 / 125 / 100 320 / 130 325 / 140 440 Undigested 235 / 115 / 100 235 / 115 / 85
150 / 125 / 70 152 / 127 / 69 a 160 / 140 / 70 b 140 / 125 / 60 145 / 127 / 59 a 155 / 140 / 60 b 140 / 120 / 60 139 / 123 / 58 a 155 / 135 b 125 / 100 127 / 103 a 140 / 105 b 130 / 105 / 80 140 / 105 / 70 b 155 / 110 / 60 170 / 115 b 125 / 110 140 / 120 b 195 / 60 210 b 140 / 120 150 / 135 c 155 / 110 135 / 90 / 85
a As obtained by construction of a restriction map from the GenBank sequences. GenBank accession numbers are M15467 (M. tuberculosis), U55829 (M. intracellulare), U55834 (M. fortuitum) and U55826 (M. avium). b As described by Telenti et al. (1993). c As described by Springer et al. (1996). d Identity of strains was confirmed by evaluation of growth rate, pigmentation, nitrate reduction, arylsulfatase, iron uptake and growth on NaCl.
Fig. 1. PRA patterns of M. lentiflavum (1), M. abcessus (2), M. fortuitum (3 and 4) and M. gordonae (5) after digestion with BstEII (A) or BspRI (B). The molecular weight marker (M) is a 50 bp ladder.
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
study concentrating on rapidly growing mycobacteria (Steingrube et al., 1995). Consideration of such bands is not only academical because the presence of a single band in this molecular-weight range differentiates M. kansasii from M. avium (Devallois et al., 1997a) and M. smegmatis from M. margaritense (Domenech et al., 1997). As observed in other laboratories (Devallois et al., 1997b; Taylor et al., 1997; Bollet et al., 1992), generally smaller molecular-weight values were dedicated to the bands when compared to those in the algorithm presented by Telenti et al. (1993), a fact that could be due to differences in running conditions, agarose types and computer algorithms used for calculation of molecular weight values. In this study, molecular-weight values were calculated by position regression by experimental curve fitting upon comparison with 50 and 25 bp ladders and compared with theoretical values obtained after construction of HaeIII and BstEII restriction maps of sequences obtained from GenBank (Table 1). Although definition of exact molecular values is not important for identification of an unknown pattern by comparison with reference patterns or with a local database, it could be important for comparison of identification data generated in different laboratories. In a second stage, PRA was performed on bacterial isolates from patients with suspicion of mycobacterial infection. In total, 43 clinical isolates from 42 different individuals were identified, with species belonging to the M. avium-intracellulare complex being the most frequently isolated atypical mycobacteria, confirming earlier observations (Barreto et al., 1993). When comparing identification results obtained with commercial methods with those obtained by PRA, all but three isolates had corresponding patterns. All 19 isolates of M. tuberculosis and nine isolates of M. avium were represented by an easily recognizable unique PRA pattern, confirming that also in Brazil, PRA is a simple tool for rapid differentiation between the two most frequently encountered mycobacterial species. Strain heterogeneity within the complex MAIS is responsible for at least five additional PRA patterns (Devallois et al., 1997a) but the two isolates identified here as MAIS by the GenProbe system demonstrated the PRA pattern that was originally described for M. intracellulare by Telenti et al. (1993). We have preliminary
227
data however that, for correct identification of M.intracellulare in Brazil, just as is the case in the Caribbean Isles and the Indian subcontinent, a larger number of PRA patterns will have to be considered [unpublished]. From the three isolates that were identified by HPLC as M. fortuitum, two were from the same patient taken with a 5-month interval and had a so far undescribed PRA pattern (Fig. 1); the other was characterised by the M. fortuitum I pattern. Two of the four strains identified as M. gordonae also had a so far undescribed PRA pattern (Fig. 1), while two other strains had respectively the M. gordonae I and III pattern. Having in mind that we analyzed a relative small number of strains while in earlier studies, PRA was performed on a large number of strains, we feel that a predominance of certain strains with a particular PRA patterns could exist in some geographic areas. We propose the inclusion of these new patterns that we would like to call M. fortuitum III and M. gordonae VII in the PRA identification algorithm that was recently constructed by Devallois et al. (1997b) (Table 1). Although genetic variation has also been reported for M. kansasii (Alcaide et al., 1997), we encountered the M. kansasii I pattern in four different isolates. One isolate that was identified as M. abscessus by HPLC had a pattern similar to the recently described M. abscessus II by Devallois et al. (1997b), except for an extra band of 70 bp (Fig. 1). Double bands of 60 and 70 bp have consistently been observed in isolates of the M. chelonae complex, collected in another setting (our unpublished observations) and we therefore feel that this band should be added in the algorithm. Presence of an 70 bp fragment has also been observed in some isolates of M. gordonae (pattern IV), normally characterized by a 60 bp band (Taylor et al., 1997). Whether these differences are geographic peculiarities has to be determined. Finally, one isolate was characterized as the recently described species M. lentiflavum, an organism with biochemical characteristics of M. avium and a 16S rRNA sequence closely related to that of M. simiae and M. genavense (Springer et al., 1996). This species was described for the first time in 1996 and has so far been reported in Italy, Switzerland and Germany only (Springer et al., 1996; Tortoli et al., 1997; Haase et al., 1997). The PRA patterns of the strain detected here was identical to one of the three
228
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229
earlier published patterns (Rodrigo et al., 1992). Although isolates reported in literature were mostly fortuitous or traced to contaminated bronchoscopes, some strains were isolated from sterile tissue (Springer et al., 1996; Haase et al., 1997) and association between transition from colonization to disease and decreasing CD4-level of immonocompromised patients was observed (Tortoli et al., 1997). Whether the isolation of M. lentiflavum in this case was associated with disease is not known because the organism was isolated from sputum of a patient with suspicion of pulmonary tuberculosis. In conclusion, PRA has proven to be a fast and relatively simple method for identification of mycobacteria, to an extent which is normally not obtained by routine procedures. The fact that three isolates were characterized by new pattern suggests that particular strains could be circulating in certain geographic areas, meaning that a sufficient number of isolates have to be analyzed before using PRA as a single identification method. After submission of the manuscript, we became aware of a compilation of PRA patterns available through the internet at the following address: http:\\www.hospvd.ch:8005.
Acknowledgements This work received financial support from the CNPq and CBAB. We thank the National Reference ´ Center of Tuberculosis Prof. Helio Fraga (Curicica, Rio de Janeiro, Brazil) for providing us with some of the mycobacterial reference strains and Marlei Gomes da Silva and Leila de Souza Fonseca from the Federal University of Rio de Janeiro for helping us performing the conventionel identification of M. fortuitum and M. gordonae.
References Abe, C., Hosojima, S., Fukasawa, Y., Kazumi, Y., Takahashi, M., Hirano, K., Mori, T., 1992. Comparison of MB-Check, BACTEC, and egg-based media for recovery of mycobacteria. J. Clin. Microbiol. 30, 878–881. Alcaide, F., Richter, I., Bernasconi, C., Springer, B., Hagenau, C., Schultze-Robbecke, R., Tortoli, E., Martin, R., Bottger, E.C., Telenti, A., 1997. Heterogeneity and clonality among isolates
of Mycobacterium kansasii: implications for epidemiological and pathogenicity studies. J. Clin. Microbiol. 35, 1959–1964. Barreto, J.A., Palaci, M., Ferrazoli, L., Martins, M.C., Suleiman, J., Lorenco, J.R., Ferreira, O.C., Riley, Jr.L. W., Johnsonn, W.D. Jr., Galvao, P.A., 1993. Isolation of Mycobacterium avium complex bone marrow aspirates of AIDS patients in Brazil. J. Infect. Dis. 168, 777–779. Bollet, C., De Lamballerie, X., Zandotti, C., Vignoli, C., Gevaudan, M.J., De Micco, P., 1992. Detection and identification of Mycobacterium tuberculosis, M. bovis BCG, and M. avium by two-step polymerase chain reaction. Comparison with ELISA using A60 antigen. Microbiologica 15, 345–350. De Beenhouwer, H., Liang, Z., DeRijk, P., VanEekeren, C., Portaels, F., 1995. Detection and identification of mycobacteria by DNA amplification and oligonucleotide-specific capture plate hybridization. J. Clin. Microbiol. 33, 2994–2998. Devallois, A., Picardeau, M., Paramasivan, C.N., Vincent, V., Rastogi, N., 1997a. Molecular characterization of Mycobacterium avium complex isolates giving disconcordant results in AccuProbe tests by PCR-restriction enzyme analysis, 16S rRNA gene sequencing, and DT1-DT6 PCR. J. Clin. Microbiol. 35, 2767–2772. Devallois, A., SengGoh, K., Rastogi, N., 1997b. Rapid identification of mycobacteria to species level by PCR-restriction fragment length polymorphism analysis of the hsp65 gene and proposition of an algorithm to differentiate 34 mycobacterial species. J. Clin. Microbiol. 35, 2969–2973. Dobner, P., Feldman, K., Rifai, M., Loschner, T., Rinder, H., 1996. Rapid identification of mycobacterial species by PCR amplification of hypervariable 16 rRNA gene promoter region. J. Clin. Microbiol. 34, 866–869. Domenech, P., Jimenez, M.S., Menendez, M.C., Bull, T.J., Samper, S., Manrique, A., Garcia, M.J., 1997. Mycobacterium mageritense sp. nov. Int. J. Syst. Bacteriol. 47, 535–540. Glickman, S.E., Kilburn, J.O., Butler, W.R., Ramos, L.C., 1994. Rapid identification of mycolic acid patterns of mycobacteria by high-performance liquid chromatography using pattern recognition software and a Mycobacterium library. J. Clin. Microbiol. 32, 740–745. Haase, G., Kentrup, H., Skopnik, H., Springer, B., Bottger, E.C., 1997. Mycobacterium lentiflavum: an etiologic agent of cervical lymphadenitis. Clin. Infect. Dis. 25, 1245–1246. Kapur, V., Li, L.-L., Hamrick, M.R., Plikyatis, B.B., Shinnick, T.M., Telenti, A., Jacobs, W.R., Banerjee, A., Cole, S., Yuen, K.Y., Clarridge, J.E., Kreiswirth, B.N., Musser, J.M., 1995. Rapid Mycobacterium species assignment and unambiguous identification of mutations associated with antimicrobial resistance in Mycobacterium tuberculosis by automated sequencing. Arch. Pathol. Lab. Med. 119, 131–138. Kox, L.F.F., Jansen, H.M., Kuijper, S., Kolk, A.H.J., 1997. Multiplex PCR assay for immediate identification of the infecting species in patients with mycobacterial disease. J. Clin. Microbiol. 35, 1492–1498. Kubica, G.P., Dye, W.E., Cohen, M.L., Middlebrook, G., 1963. Sputum digestion and concentration with N-acetyl-L-cysteinesodium hydroxide for culture of mycobacteria. Am. Rev. Respir. Dis. 87, 775. Picardeau, M., Prod’hom, G., Raskine, L., LePennec, M.P.,
A. da Silva Rocha et al. / Journal of Microbiological Methods 37 (1999) 223 – 229 Vincent, V., 1997. Genotypic differentiation of five subspecies of Mycobacterium kansasii. J. Clin. Microbiol. 35, 25–32. Plikaytis, B.B., Plikaytis, B.D., Yakrus, M.A., Butler, W.R., Woodley, C.L., Silcox, V.A., Shinnick, T.M., 1992. Differentiation of slowly growing Mycobacterium species including Mycobacterium tuberculosis, by gene amplification and restriction fragment length polymorphism analysis. J. Clin. Microbiol. 30, 1815–1822. Rodrigo, G., Kallenius, G., Hoffmann, E., Svenson, S.B., 1992. Diagnosis of mycobacterial infections by PCR and restriction enzyme analysis. Lett. Appl. Microbiol. 15, 41–44. Rogall, T., Flohr, T., Bottger, E.C., 1990. Differentiation of Mycobacterium species by direct sequencing of amplified DNA. J. Gen. Microbiol. 136, 1915–1920. Santos, A.R., De Miranda, A.B., Lima, L.M., Suffys, P.N., Degrave, W.M., 1992. Method for high yield preparation in large and small scale of nucleic acids from mycobacteria. J. Microbiol. Methods 28, 1236–1243. Soini, H., Bottger, E., Viljanen, M.K., 1994. Identification of mycobacteria by PCR-based sequence determination of the 32-kilodalton protein gene. J. Clin. Microbiol. 32, 2944–2947. Springer, B., Whei-Kuo, W., Bodmer, T., Haase, G., Pfyffer, G.E., Kroppenstedt, R.M., Schroder, K.-H., Elmer, S.J., Kilburn, O., Kirchner, P., Telenti, A., Coyle, M.B., Bottger, E.C., 1996. Isolation and characterization of a unique group of slowly growing mycobacteria: description of Mycobacterium lentiflavum sp. nov. J. Clin. Microbiol. 34, 1100–1107.
229
Steingrube, V.A., Gibson, J.L., Brown, B.A., Zhang, Y., Wilson, R.W., Rajagopalan, M., WallaceJr, J., 1995. PCR amplification and restriction endonuclease analysis of a 65-kilodalton heat shock protein gene sequence for taxonomic separation of rapidly growing mycobacteria. J. Clin. Microbiol. 33, 149–153. Taylor, T.B., Patterson, C., Hale, Y., Safranek, W.W., 1997. Routine use of PCR-restriction fragment length polymorphism analysis for identification of mycobacteria in liquid media. J. Clin. Microbiol. 35, 79–85. Telenti, A., Marchesi, F., Balz, M., Bally, F., Bottger, E., Bodmer, T., 1993. Rapid identification of mycobacteria to the species level by polymerase chain reaction and restriction enzyme analysis. J. Clin. Microbiol. 31, 175–178. Tortoli, E., Piersimoni, C., Kirchner, P., Bartoloni, A., Burrini, C., Lachini, C., Mantella, A., Muzzi, G., PasseriniTosi, C., Penati, V., Scarparo, C., TuliaSimonetti, M., Bottger, E.C., 1997. Characterization of mycobacterial isolates phylogenetically related to, but different from, Mycobacterium simiae. J. Clin. Microbiol. 35, 697–702. Vaneechoutte, M., de Beenhouwer, H., Claeys, G., Verschraegen, G., de Rouck, A., Paepe, N., Elaichouni, A., Portaels, F., 1993. Identification of Mycobacterium species using amplified ribosomal DNA restriction analysis. J. Clin. Microbiol. 31, 2061–2065. Zolg, J.W., Philippi-Schultz, S., 1994. The superoxide dismutase gene, a target for detection and identification of mycobacteria by PCR. J. Clin. Microbiol. 32, 2801–2812.