IS1626, a new IS900-related Mycobacterium avium insertion sequence

Xiaoling Puyang1, Karen Lee1, Corey Pawlichuk1 and Dennis Y. Kunimoto1

Department of Medical Microbiology and Immunology, University of Alberta, Edmonton, Canada T6G 2H71

Author for correspondence: Dennis Y. Kunimoto. Tel: +1 780 407 1418. Fax: +1 780 4927521. e-mail: Dennis.Kunimoto{at}Ualberta.Ca


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
An insertion sequence designated IS1626 was isolated and characterized from a Mycobacterium avium clinical strain. IS1626 was detected by high-stringency hybridization with the pMB22/S12 probe from IS900 of Mycobacterium paratuberculosis. IS1626 is 1418 bp in size and has a G+C content of 65 mol%. It has neither terminal inverted repeats nor flanking direct repeats. Analysis of three IS1626 insertion sites in the M. avium strain and the corresponding potential insertion sites in two IS1626-free M. avium strains indicated a consensus sequence of CATGCN(4–5)TCCTN(2)G for IS1626 insertion. In the three clones examined, IS1626 has the same orientation with respect to this target site. IS1626 has two major ORFs. ORF1179 encodes a predicted protein of 393 amino acids. ORF930, on the complementary strand of ORF1179, encodes a protein of 310 amino acids. The Shine–Dalgarno sequence for ORF930 is partially located in the flanking region, similar to other IS900-related elements. Analysis of the comparable features of insertion sequences and their variable occurrence in related organisms is useful for studying the evolution of these elements and their hosts.

Keywords: mobile genetic elements, transposition, IS900, direct repeats, inverted repeats

Abbreviations: IS, insertion sequence; MAC, Mycobacterium avium complex

The GenBank accession number for the IS1626 sequence determined in this work is AF071067.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Insertion elements and transposons are mobile genetic elements that are found integrated into the chromosome of most organisms. The most basic are known as insertion sequences (IS) and are typically characterized by terminal inverted repeats and flanking direct repeats, together with a single major ORF that contains a gene for transposition (Galas & Chandler, 1989 ). Transposons are larger elements which contain genes, in addition to a transposition gene, often encoding antibiotic resistance or virulence factors (Berg et al., 1988 ; Galas & Chandler, 1989 ). Insertion of these elements can either be random, directed or even site-specific. Insertion elements have been used as genetic tools for the manipulation of DNA. For example, if insertion interrupts a functional gene, it can cause a mutation and the phenotype can be selected for, as was done to select virulence genes from Bordetella (Weiss et al., 1983 ). Insertion sequences such as IS6110 that insert randomly and are present in multiple copies can be used as markers for strain typing (Thierry et al., 1990 ). Insertion elements can be used to carry reporter genes in searches for promoters (Labes et al., 1997 ) or to carry regulatable promoters to manipulate the expression of selected genes (Murray et al., 1992 ).

Insertion elements often show species specificity or a narrow host range. Only recently have insertion elements been described in mycobacteria and only five have been described from the Mycobacterium avium complex (MAC).

IS900, the first insertion sequence characterized in mycobacteria, was identified from the clone pMB22, which was derived from a genomic library prepared from a human Crohn’s disease isolate of Mycobacterium paratuberculosis (Green et al., 1989 ). IS900 was shown to be highly specific to M. paratuberculosis (Vary et al., 1990 ; Moss et al., 1991 , 1992b ).

Some strains of M. avium were found to produce a complex banding pattern after low-stringency screening of Southern blots of M. avium strains using the clone pMB22 as a probe. Subsequently, IS901 was identified and found in most M. avium animal strains (Kunze et al., 1991 , 1992 ; Nishimori et al., 1995 ) and IS902 was identified from M. avium subsp. silvaticum (Moss et al., 1992a ). However, IS901 and IS902 are virtually identical.

Recently, IS1110 was discovered during a study of plasmid incidence in AIDS-derived M. avium strains. IS1110 is a 1457 bp element lacking terminal inverted repeats and is related to IS900 and IS901. IS1110 was detected in some M. avium isolates showing polymorphism and was considered to be a highly mobile genetic element since the transposition events of IS1110 can be detected in random colonies without any selection pressure (Hernandez Perez et al., 1994 ).

IS1245 was identified from an M. avium genomic fragment. It is a 1313 bp element with two imperfect inverted repeats and one ORF. IS1245 can be detected in a variety M. avium isolates, but human isolates characteristically show a high number of copies and a greater diversity of restriction fragment length patterns (Guerrero et al., 1995 ).

In Mycobacterium intracellulare, an insertion sequence, IS1141, has been identified (GenBank accession no. L10239). IS1141 is 1588 bp long, containing 23 bp inverted repeats at its ends. Although IS1141 only appears in some M. intracellulare strains, its transposition may be associated with colonial variation.

In this report, we describe another insertion sequence isolated from M. avium, IS1626, which is the most closely related insertion element to IS900 described to date.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Preparation of mycobacterial DNA.
Mycobacterial cells were grown on Middlebrook 7H10 agar slants at 37 °C in 10% CO2 for 4–6 weeks. Genomic DNA was isolated by a glass bead homogenization method (Jacobs et al., 1991 ).

DNA probes and colony screening.
The 300 bp HindIII/SacI fragment from pMB22/S12 of IS900 (a generous gift from Dr J. J. McFadden) was labelled with [{alpha}-32P]dCTP by the random primer labelling method (Feinberg & Vogelstein, 1984 ) and used as a probe for Southern blot analysis (Southern, 1975 ) and for colony screening (see below). To clone the desired fragments, DNA fragments of the appropriate size were cut from a gel and isolated using a Geneclean II kit (Bio101) as described by the supplier. Fragments were then ligated into pBluescript (Stratagene) and recombinant plasmids were obtained after transformation of competent Escherichia coli DH5{alpha} cells. Colony screening (Grunstein & Hogness, 1975 ) was used to select the IS900-related fragments.

DNA sequencing.
Plasmid DNA was prepared by a modified alkaline denaturation mini-preparation procedure (Sambrook et al., 1989 ). DNA was sequenced using the dideoxy chain-termination method of Sanger et al. (1977) with either with the Sequenase version 2.0 DNA Sequencing Kit, (United States Biochemical) or by Taq DNA polymerase cycle sequencing as developed by Murray (1989) .

T3/T7 primers were used to sequence the insert of pBluescript recombinants. Some primers were selected from newly determined sequences in order to sequence adjacent sections.

Computer analysis of sequence data.
DNA Strider 1.2 (Douglas, 1995 ) was used to check the repeats of IS1626 and determine the ORF. The Wisconsin GCG software (Genetics Computer Group) was used for pairwise comparison and multiple sequence alignment of the nucleotide sequence and predicted protein sequence of IS1626 with members of IS900 family


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Screening for IS900-related elements
Sixty-six BamHI-digested clinical MAC isolates were screened by Southern blotting with the subclone of IS900, pMB22/S12. Two of these MAC strains hybridized with this probe (Fig. 1). Strain 21878 showed three positive hybridizing bands at 3·8 kb, 8 kb and 9 kb, whereas strain 13219 showed one hybridizing band at 10 kb. As pMB22/S12 is reported to be specific for M. paratuberculosis, it was suspected either that these MAC strains contained a new IS900-related element or that they had been wrongly identified. It was therefore critical to confirm the species to which they belonged.



View larger version (34K):
[in this window]
[in a new window]
 
Fig. 1. BamHI-digested genomic DNA of M. avium strain 21878 and M. intracellulare strain 13219 probed with radiolabelled pMB22/S12. Strain 21878 shows three hybridizing bands at 3·8 kb, 8 kb and 9 kb (lane 1) and strain 13219 shows one hybridizing band at 10 kb (lane 2). Size markers (Life Technologies) are shown on the left.

 
Strain 21878 was identified as M. avium by Gen-Probe testing (GEN-PROBE), by 16S rRNA sequence determination (Rogall et al., 1990 ) and by restriction analysis of the 65 kDa heat-shock protein gene (Telenti et al., 1993 ). Primers P1 and P2 (Guerrero et al., 1995 ), and AV6 and AV7 (Thierry et al., 1993 ), were used in PCR to detect the presence of the M. avium-specific sequences IS1245 and DT6, respectively, in strain 21878. Results were as described in the literature for M. avium and were the same as the ATCC M. avium strain 25291, which was used as a positive control.

Strain 13219 was identified as M. intracellulare by Gen-Probe testing (GEN-PROBE), by 16S rRNA sequence determination (Rogall et al., 1990 ) and by restriction analysis of the 65 kDa heat-shock protein gene (Telenti et al., 1993 ). It was negative by PCR for IS1245 and DT6, and was positive by PCR for the M. intracellulare specific fragment DT1 using the primers IN38 and IN41 (Thierry et al., 1993 ). Results were as described in the literature for M. intracellulare and were the same as the ATCC M. intracellulare strain 13950. The IS in strain 13219 was not studied in detail. It is believed to be related, but not identical, to the IS in strain 21878.

As M. avium strain 21878 had three positive hybridizing bands, it was chosen for further study. From this strain, several subgenomic libraries were constructed with the desired size DNA corresponding to the three positive hybridizing bands, and then screened with a radiolabelled pMB22/S12 probe. Three corresponding clones, pB3.8, pB8.0 and pEH5.0, were recovered. Restriction mapping and Southern blotting confirmed that these three clones contained the same 1·3 kb fragment.

Identification of IS1626
The region containing the hybridizing fragment in pB3.8 was sequenced. Computer analysis of the sequence demonstrated a high degree of homology with IS900, indicating an IS900-related element. No inverted repeats or direct repeats were found. Therefore, from pB3.8 alone it was not possible to identify the exact end points of the IS900-related element. Further subcloning and sequencing of pB8.0 and pEH5.0 were subsequently performed. Comparison of the sequences of the three clones indicated that they contain the same fragment, which was designated IS1626. The divergence points of IS1626 for the three clones was determined.

To further confirm the termini of IS1626, DNA primers generated from sequences flanking IS1626 in pEH5.0 were used to directly sequence the equivalent insertion locus from two IS1626-free M. avium strains. By comparing the sequences at the divergence points from the three clones with the sequences from IS1626-free M. avium strains, the exact ends of IS1626 were determined. IS1626 is 1418 bp long and its G+C content is 65 mol%, which is similar to that found in the host genomic DNA (about 62–70%). The G+C content of IS1626 is similar to that of IS elements IS900, IS901 and IS110, which also have G+C contents comparable to that of their hosts, M. paratuberculosis, M. avium and Streptomyces, respectively (Bruton & Chater, 1987 ; Green et al., 1989 ; Kunze et al., 1991 ).

In contrast to most other insertional elements, IS1626 does not possess either terminal repeats or flanking direct repeats. This absence is characteristic of previously characterized IS900-related elements. In most IS elements, the inverted repeats are believed to provide the recognition and binding sites for the transposase protein (Grindley, 1985 ; Berg et al., 1988 ; Galas & Chandler, 1989 ) and the direct repeats are most likely derived from the staggered cut of the targeted sequence followed by repair upon transposition. However, the structure of these IS900-related elements suggests that the inverted repeat is not essential for the binding of all transposases and that the target staggered cut is not the only model for transposition.

IS1626 insertion site
Sequence analysis of the flanking regions of clones pB3.8, pB8.0 and pEH5.0 and of the potential insertion sites in two IS1626-free M. avium isolates indicated an insertion site specificity. The consensus insertion site for IS1626 is 5'-CATGCN4–5*TCCTN2G-3' (the asterisk denotes site of insertion) and in three clones the element inserts in the same orientation with respect to this target sequence, which is similar to the insertion site sequences described for IS900-related elements. This specific insertion suggests that the flanking sequence plays an important role during transposition. Examination of 13 insertion site sequences from IS1626 and other IS900-related elements (Table 1) revealed that they all have CAT at the left insertion site and CCT at the right insertion site. These two 3 nt-elements were separated by 6–9 nt in the consensus insertion sequences. It was also noted that the same 3 nt elements also appear in their IS sequences with the internal CAT at the right side, 7–9 nt away from the CCT in the right flanking region, and the internal CCT at the left side, 5–11 nt away from the CAT in the left flanking region. These short sequences might be important for transposase recognition. The presence of nearby regions of limited homology with the ends of insertion sequences has been suggested to play a role in site selection (Galas & Chandler, 1989 ). Morisato & Kleckner (1984) found that the Tn10 transposase itself can make double-stranded cuts. Their study supports the cut-and-paste process. Therefore, it is conceivable that, like class I restriction endonucleases, the transposase in IS1626 and IS900-related IS elements may recognize and cut the CAT–CCT asymmetric sequences at both ends of transposable elements in the donor and the consensus insertion site in the recipient, and then ligate them in the same order. If these cleavages are made by blunt-end cutting followed by the cut-and-paste process, then simple insertion occurs. This transposition will not generate target duplication and the IS will always be in the same orientation with respect to their consensus site. At present, the mechanisms for transposition of IS900 family elements are far from completely understood. However, the insertion specificity of these IS elements may reflect base sequence recognition by their transposition proteins, and this recognition specificity may depend not only on the specific nucleotides but also the local DNA conformation.


View this table:
[in this window]
[in a new window]
 
Table 1. Comparison of the putative insertion sites of IS900-related elements

 
Comparison of IS1626 with other IS elements by analysis of the two major ORFs
There are two major ORFs in IS1626 (Fig. 2). There is one major ORF starting at position 172 and ending with a TGA codon at position 1398. In this same reading frame several possible GTG and ATG start codons were found. Among these various possibilities, initiation is only likely to take place with a GTG codon at nt 220 on the basis that it is adjacent to a recognizable Shine–Dalgarno sequence (GGAGG) from nucleotides 211 to 215. This ORF of 1179 nucleotides (ORF1179) would encode a protein of 393 amino acids with an expected molecular mass of 44 kDa.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 2. Arrows depicting the ORFs of IS1626 which are at least 200 bp long. ORF1179 and ORF930 are indicated by the shaded regions.

 
It was noted that within the first 180 nucleotides of IS1626 and upstream of ORF1179 there is a -35-like region with the sequence TTGAGA, and a -10-like region with the sequence TGTAAG. These two sequences are in good agreement with the E. coli promoter consensus sequences TTGACA and TATAAT, respectively (Dale & Patki, 1990 ). It is expected that this region could promote the transcription of ORF1179, although its significance has not yet been determined.

Sequence homologies between IS1626 and IS900, IS901, IS1110, IS110 and IS116 were determined by sequence alignment with the Wisconsin GCG software (Table 2). The DNA sequence comparisons indicated that IS1626 shows 82% homology with IS900, and a variable homology of 53–62% with the other related IS elements. No significant homologies were found with IS elements from E. coli or other mycobacterial species. Comparison of the predicted amino acid sequences of the major ORFs from IS900, IS901, IS1110, IS110 and IS116 indicated that ORF1179 shows 63% identity with IS900, and 30–42% identity with the others.


View this table:
[in this window]
[in a new window]
 
Table 2. Homologies (%) between IS1626 and IS900-related sequences

 
CLUSTAL alignment of the major ORFs of these elements indicated that the 3' carboxy termini of the predicted proteins show higher similarities than the 5' ends. Of note was the observation that all IS elements contain the conserved motif K-DAKDA, which has been found in reverse transcriptase enzymes (Xiong & Eickbush, 1988 ).

Similar to IS900 and IS902, IS1626 has another long ORF, ORF930, on the complementary strand and in the reverse direction of ORF1179 (Fig. 2). ORF930 starts at the very 3' end of IS1626 at nt 1410 and ends at nt 481. Its ATG start codon is preceded by a Shine–Dalgarno sequence, AGGAGA, which is partially located in the flanking region. This ORF930 encodes a predicted protein of 310 amino acids. Its translational signals are formed from IS1626 and its flanking region, but it does not seem to have any upstream transcriptional signals. Furthermore, analysis of the putative insertion sites for IS900-related elements (Table 1) showed that the complementary strands contain a Shine–Dalgarno sequence and an AUG-initiation-codon-like structure. It seems that these IS elements are ‘in search’ of the translation and transcription signals needed to obtain an active element or to express genes encoded by the IS themselves. This may also explain their unusual insertion site specificity and the orientation of the element with respect to its target site. Comparison of the amino acid sequences indicated that ORF930 of IS1626 shows 53% identity to ORF2 of IS900 and 31% identity with ORF2 of IS902. However, computer searches of the GenBank database failed to reveal any significant similarity between ORF930 and other known sequences.

Distribution of IS1626 and evolution of IS900-related elements
Unlike other related IS elements, IS1626 seems to be uncommon in M. avium. Of 66 MAC isolates screened, only one M. avium strain contained the IS1626 element and only three copies were present. The source of IS1626 is unknown. As indicated by Kunze et al. (1991) , IS900 and IS901 might have been acquired by separate events after the speciation of their hosts, M. paratuberculosis and M. avium, respectively. Sequence analysis showed that there is 18% difference at the DNA level between IS1626 and IS900. This level of diversity is much higher than the genome sequence divergence estimated between M. avium and M. paratuberculosis (<2%) (McFadden et al., 1992 ). The DNA diversity among the IS elements of M. avium is even higher. The sequence comparison between IS1626 and other IS900-related elements suggests that the acquisition of these elements was subsequent to the divergence of the host species. Since several related IS elements such as IS110, IS1110 and IS116 are carried on either a phage or a plasmid, phages and plasmids are probably the vehicles for these elements. With an association of these elements with such vectors, there may be high potential for horizontal transfer of IS900-like elements both within and among strains of MAC and other distantly related organisms. The distribution of IS900 family elements in Mycobacterium, Streptomyces and other bacteria, and the variable occurrence in M. avium strains may be evidence of horizontal transfer after the divergence of these species. This implies that horizontal transfer is common, but that the rate of transfer may be relatively low so that sufficient time passes for the IS element to accumulate substantial genetic divergence. The relatively few copies of IS1626 and its low frequency of occurrence in M. avium probably reflects the more recent invasion of this element in this species.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Berg, D. E., Kazic, T., Phadnis, S., Dodson, K. W. & Lodge, J. K. (1988). Mechanism and regulation of transposition. In Transposition, pp. 107-129. Edited by A. J. Kingsman. New York: Cambridge University Press.

Bruton, C. J. & Chater, K. F. (1987). Nucleotide sequence of IS110, an insertion sequence of Streptomyces coelicolor A3(2). Nucleic Acids Res 15, 7053-7065.[Abstract]

Dale, J. W. & Patki, A. (1990). Mycobacterial gene expression and regulation. In Molecular Biology of the Mycobacteria, pp. 173-198. Edited by J. J. McFadden. London: Academic Press.

Douglas, S. E. (1995). DNA Strider. An inexpensive sequence analysis package for the Macintosh. Mol Biotechnol 3, 37-45.[Medline]

Feinberg, A. P. & Vogelstein, B. (1984). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Addendum. Anal Biochem 137, 266-267.[Medline]

Galas, D. J. & Chandler, M. (1989). Bacterial insertion sequences. In Mobile DNA, pp. 109-162. Edited by D. E. Berg. Washington, DC: American Society for Microbiology.

Green, E. P., Tizard, M. L. V., Moss, M. T., Thompson, J., Winterbourne, D. J., McFadden, J. J. & Hermon-Taylor, J. (1989). Sequence and characteristics of IS900, an insertion element identified in a human Crohn’s disease isolate of Mycobacterium paratuberculosis. Nucleic Acids Res 17, 9063-9073.[Abstract]

Grindley, N. D. F. (1985). Transpositional recombination in prokaryotes. Annu Rev Biochem 54, 863-896.[Medline]

Grunstein, M. & Hogness, D. S. (1975). Colony hybridization: a method for the isolation of cloned DNAs that contain a specific gene. Proc Natl Acad Sci USA 72, 3961-3965.[Abstract]

Guerrero, C., Bernasconi, C., Burki, D., Bodmer, T. & Telenti, A. (1995). A novel insertion element from Mycobacterium avium, IS1245, is a specific target for analysis of strain relatedness. J Clin Microbiol 33, 304-307.[Abstract]

Hernandez Perez, M., Fomukong, N. G., Hellyer, T., Brown, I. N. & Dale, J. W. (1994). Characterization of IS1110, a highly mobile genetic element from Mycobacterium avium. Mol Microbiol 12, 717-724.[Medline]

Jacobs, W. R., Kalpana, G. V., Cirillo, J. D., Pascopella, L., Snapper, S. B., Udani, R. A., Jones, W., Barletta, R. G. & Bloom, B. R. (1991). Genetic systems for mycobacteria. Methods Enzymol 204, 537-555.[Medline]

Kunze, Z. M., Wall, S., Appelberg, R., Silva, M. T., Portaels, F. & McFadden, J. J. (1991). IS901, a new member of a widespread class of atypical insertion sequences, is associated with pathogenicity in Mycobacterium avium. Mol Microbiol 5, 2265-2272.[Medline]

Kunze, Z. M., Portales, F. & McFadden, J. J. (1992). Biologically distinct subtypes of Mycobacterium avium differ in possession of insertion sequence IS901. J Clin Microbiol 30, 2366-2372.[Abstract]

Labes, G., Bibb, M. & Wohlleben, W. (1997). Isolation and characterization of a strong promoter element from the Streptomyces ghanaensis phage I19 using the gentamicin resistance gene (aacC1) of Tn1696 as reporter. Microbiology 143, 1503-1512.[Abstract]

Leskiw, B. K., Mevarech, M., Barritt, L. S., Jensen, S. E., Henderson, D. J., Hopwood, D. A., Bruton, C. J. & Chater, K. F. (1990). Discovery of an insertion sequence, IS116, from Streptomyces clavuligerus and its relatedness to other transposable elements from actinomycetes. J Gen Microbiol 136, 1251-1258.[Medline]

McFadden, J. J., Kunze, Z. M., Portaels, F., Labrousse, V. & Rastogi, N. (1992). Epidemiological and genetic markers, virulence factors and intracellular growth of Mycobacterium avium in AIDS. Res Microbiol 143, 423-430.[Medline]

Morisato, D. & Kleckner, N. (1984). Transposase promotes double strand breaks and single strand joints at Tn10 termini in vivo. Cell 39, 181-190.[Medline]

Moss, M. T., Green, E. P., Tizard, M. L., Malik, Z. P. & Hermon-Taylor, J. (1991). Specific detection of Mycobacterium paratuberculosis by DNA hybridization with a fragment of the insertion element IS900. Gut 32, 395-398.[Abstract]

Moss, M. T., Malik, Z. P., Tizard, M. L. V., Green, E. P., Sanderson, J. D. & Hermon-Taylor, J. (1992a). IS902, an insertion element of the chronic-enteritis-causing Mycobacterium avium subsp. silvaticum. J Gen Microbiol 138, 139-145.[Medline]

Moss, M. T., Sanderson, J. D., Tizard, M. L. V., Hermon-Taylor, J., El-Zaatari, F. A. K., Markesich, D. C. & Graham, D. Y. (1992b). Polymerase chain reaction detection of Mycobacterium paratuberculosis and Mycobacterium avium subsp. silvaticum in long term cultures from Crohn’s disease and control tissues. Gut 33, 1209-1213.[Abstract]

Murray, A., Winter, N., Lagranderie, M. & 7 other authors (1992). Expression of Escherichia coli ß-galactosidase in Mycobacterium bovis BCG using an expression system isolated from Mycobacterium paratuberculosis which induced humoral and cellular immune responses. Mol Microbiol 6, 3331–3342.[Medline]

Murray, V. (1989). Improved double-stranded DNA sequencing using the linear polymerase chain reaction. Nucleic Acids Res 17, 8889.[Medline]

Nishimori, K., Eguchi, M., Nakaoka, Y., Onodera, Y., Ito, T. & Tanaka, K. (1995). Distribution of IS901 in strains of Mycobacterium avium complex from swine by using IS901-detecting primers that discriminate between M. avium and Mycobacterium intracellulare. J Clin Microbiol 33, 2102-2106.[Abstract]

Rogall, T., Flohr, T. & Böttger, E. C. (1990). Differentiation of Mycobacterium species by direct sequencing of amplified DNA. J Gen Microbiol 136, 1915-1920.[Medline]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sanger, F., Nicklen, S. & Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74, 5463-5467.[Abstract]

Southern, E. M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Biol 98, 503-517.[Medline]

Telenti, A., Marchesi, F., Balz, M., Bally, F., Böttger, E. C. & 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.[Abstract]

Thierry, D., Brisson-Noel, A. & Mekalanos, J. J. (1990). Characterization of a Mycobacterium tuberculosis insertion sequence, IS6110, and its application in diagnosis. J Clin Microbiol 28, 2668-2673.[Medline]

Thierry, D., Vincent, V., Clément, F. & Guesdon, J.-L. (1993). Isolation of specific DNA fragments of Mycobacterium avium and their possible use in diagnosis. J Clin Microbiol 31, 1048-1054.[Abstract]

Vary, P. H., Andersen, P. R., Green, E., Hermon-Taylor, J. & McFadden, J. J. (1990). Use of highly specific DNA probes and the polymerase chain reaction to detect Mycobacterium paratuberculosis in Johnes disease. J Clin Microbiol 28, 933-937.[Medline]

Weiss, A. A., Hewlett, E. L., Myers, G. D & Falkow, S. (1983). Pertussis toxin and extracytoplasmic adenylate cyclase as virulence factors. Infect Immun 42, 33-41.[Medline]

Xiong, Y. & Eickbush, T. H. (1988). Similarity of reverse transcriptase-like sequences of virus, transposable elements and mitochondrial introns. Mol Biol Evol 5, 675-690.[Abstract]

Received 8 June 1999; accepted 9 July 1999.