- Email: [email protected]
Transposition of Tn4560 of Streptomyces fradiae in Mycobacterium smegmatis
FEMS Microbiology Letters 206 (2002) 241^246
www.fems-microbiology.org
Transposition of Tn4560 of Streptomyces fradiae in Mycobacterium smegmatis Apoorva Bhatt b
a;
*, Graham R. Stewart b , Tobias Kieser
a
a Department of Genetics, John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, UK Centre for Molecular Microbiology and Infection, Imperial College, South Kensington, London SW7 2AZ, UK
Received 4 October 2001; received in revised form 15 November 2001; accepted 21 November 2001 First published online 17 December 2001
Abstract Tn4560 (8.6 kb) was derived from Tn4556, a Tn3-like element from Streptomyces fradiae. It contains a viomycin resistance gene that has not been used previously for selection in mycobacteria. Tn4560, cloned in a Streptomyces plasmid, was introduced by electroporation into Mycobacterium smegmatis mc2 155. Tn4560 transposed into the host genome: there was no obvious target sequence preference, and insertions were in or near several conserved open reading frames. The insertions were located far apart on different AseI macrorestriction fragments. Unexpectedly, the transposon delivery plasmid, pUC1169, derived from the Streptomyces multicopy plasmid pIJ101, replicated partially in M. smegmatis, but was lost spontaneously during subculture. Replication of pUC1169 probably contributed to the relatively high efficiency of Tn4560 delivery: up to 28% of the potential M. smegmatis transformants acquired a stable transposon insertion. The data indicated that Tn4560 may be useful for random mutagenesis of M. smegmatis. ß 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Mycobacterium smegmatis mc2 155; Tn4556; Viomycin resistance; pIJ101; pMT660 ; Transposon mutagenesis
1. Introduction Transposable elements derived from Streptomyces spp. might also function in the taxonomically related mycobacteria and could be useful additions to the small list of transposable elements available for use in genetic manipulation of mycobacteria [1^4]. Indeed, one such element, IS117, has been shown to transpose into speci¢c sites in Mycobacterium smegmatis [5]. Tn4560, a viomycin-resistant derivative of the Streptomyces fradiae transposable element Tn4556 [6], was tested for random transposition in M. smegmatis. Tn4556 resembles Tn3 which transposes replicatively via a co-integrate between the donor and recipient replicons [7,8]. Co-integrates are converted into the end products of transposition by site-speci¢c recombination (resolution) mediated by a transposon-encoded resolvase and a recombination site, res, within the transposon. Tn4556 derivatives transposed into random sites (one copy
* Corresponding author. Present address: Department of Biochemistry, University of Cambridge, 80, Tennis Court Road, Cambridge CB2 1GA, UK. Tel.: +44 (1223) 333657; Fax: +44 (1223) 766002. E-mail address : [email protected] (A. Bhatt).
per replicon) in many streptomycetes and 5-bp duplications of the target sequence were created [9]. Here we show that Tn4560 transposed in M. smegmatis mc2 155. No target sequence speci¢city was observed and insertions were found inside or near many conserved open reading frames (ORFs), indicating that Tn4560 has potential for mutagenesis of M. smegmatis. 2. Materials and methods 2.1. Bacterial strains, transformation, media and DNA extraction For the preparation of electrocompetent cells, M. smegmatis mc2 155 [10] was grown in 7H9 broth (Difco) containing 0.05% Tween 80 and 10% ODAC supplement (oleic acid, dextrose, albumin, catalase; Difco). For all other procedures the strain was grown in Lennox broth+0.05% Tween 80 [11]. Genomic DNA was extracted from M. smegmatis mc2 155 by the salting-out method [11]. The Tn4560 delivery plasmid pUC1169 (Fig. 1; [6]) was propagated and isolated from Streptomyces as described earlier [11]. pYUB12, a kanamycin-resistant derivative of
0378-1097 / 02 / $22.00 ß 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 0 1 ) 0 0 5 5 0 - X
FEMSLE 10292 21-1-02
242
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
the autonomously replicating plasmid pAL5000 [14], was used as a control for determining electroporation e¤ciency. Electroporation of M. smegmatis mc2 155 was performed as described before [10]. Antibiotic-resistant transformants were selected on 7H11 agar (Difco) containing 10% ODAC, 0.5% glycerol and, as appropriate, 10 Wg ml31 kanamycin sulfate (Sigma) or 30 Wg ml31 viomycin (gift from P¢zer to D.A. Hopwood). 2.2. PCR methods The oligonucleotide primers TSR1 (5P-TCGACTCCTCGATCGTCAACC-3P) and TSR2 (5P-GAGCGTCATCAGCTGCATACC-3P) were used to amplify a part of the tsr gene of pUC1169 using 30 cycles of 96³C for 1 min, 57³C for 2 min, and 72³C for 1 min. Ligationmediated polymerase chain reaction (LM-PCR) of the Tn4560^chromosome junctions was based on a previously described method [15] using the oligonucleotides SalI-AD (5P-TCGACATTAATG-3P) and End primer (5-GGATAAGAATTCATTAATG-3P) to form a partially double-stranded adapter (SalI adapter) for ligation to SalI cohesive ends. Genomic DNA (0.4^0.5 Wg) was digested with 5 U SalI and then ligated to 25 nmol of the SalI adapter in 20 Wl, and 0.5^1 Wl of the ligation mix was then used as template for PCR ampli¢cation (in single block PTC-1001, MJ Research) using primers Tn4560L1 (5P-GCACTAGTCGGCAGCCTGCTCATCCG-3P) or Tn4560-R1 (5P-GCACTAGTTTCACGTACGGCCCGGAG-3P), and the above End primer. The PCR conditions were : hot start of 95³C for 9 min, denaturation at 95³C for 30 s, annealing at 55³C for 30 s, extension at 72³C for 90 s, total number of cycles 35. For transposon^chromosome junctions that could not be ampli¢ed from SalI-digested DNA, NgoAIV (Gibco BRL) was used in combination with the complementary oligonucleotides UNIV5AD1 (5P-GACTCGCGAATTCCGACAGTTGA-3P) and NgoAIV-AD2 (5P-CCGGTCAACTGTCG-3P). UNIV5AD1, and the transposon-speci¢c primers Tn4560-L2 (5PCGTCAAAAACTCGACCGGCA-3P) or Tn4560-R2 (5PCAGAATTCCCCCTTGCCACAGATAACAG-3P) were used for PCR as above.
3. Results and discussion 3.1. Introduction of Tn4560 into M. smegmatis mc2 155 by electroporation Tn4560 was delivered on pUC1169, a Streptomyces plasmid that was not expected to replicate in mycobacteria (Fig. 1). pUC1169 (1 Wg) was isolated from Streptomyces lividans or from Streptomyces coelicolor and introduced by electroporation into M. smegmatis mc2 155. Selection was for viomycin resistance encoded by the transposon. No viomycin-resistant (VioR ) transformants were obtained when incubation was at 37³C, the normal growth temperature for M. smegmatis. This was surprising because in Streptomyces Tn4560 transposes e¤ciently at 37³C [18]. However, reproducibly (four experiments) after 5 days incubation at 30³C between V103 and 5U103 very small VioR colonies of irregular shape were observed (Fig. 2a). The apparent transformation frequency was approximately 10% of the frequency obtained with the autonomously replicating pYUB12. There was no VioR growth on the no-DNA control. Using 30, 50 or 100 Wg ml31 viomycin gave the same number of transformants, and the number of colonies did not decrease with increasing viomycin concentration. This indicated that the small col-
2.3. DNA sequencing Puri¢ed PCR products (QIAquick gel extraction kit) were sequenced using the Amersham Thermo Sequenase1 dye terminator or PE Applied Biosystems BigDye1 terminator sequencing kits and Tn4560-speci¢c primers. Sequences were compared with the M. smegmatis database at http://www.tigr.org, and with the Mycobacterium tuberculosis and Mycobacterium bovis databases at http:// www.sanger.ac.uk. BLAST searches [16] were performed at http://www.ncbi.nlm.nih.gov/BLAST/, and ORFs were predicted using the FRAME program [17] at http:// watson.nih.go.jp/ujun/cgi-bin/frameplot.pl.
Fig. 1. Map of pUC1169. Tn4560 (shaded region) was cloned as a BamHI fragment into the BglII site of pMT660 [12], a temperature-sensitive derivative of the multicopy Streptomyces plasmid pIJ101 which replicates by a rolling circle mechanism. vph, viomycin phosphotransferase conferring viomycin resistance [13]; tnpA, transposase; tnpR, putative resolvase ; res, putative co-integrate resolution site ; IR, 38-bp terminal inverted repeats; rep, pIJ101 replicase; ori, pIJ101 ds origin of replication; tsr, thiostrepton resistance gene useful for selection in streptomycetes but not in mycobacteria.
FEMSLE 10292 21-1-02
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
243
propagated individually for 3 days at 30³C in 10 ml L-broth+Tween 80 without antibiotic, producing dense cultures. Then, 5-Wl samples from each of these cultures were spotted onto 7H11 agar plates containing viomycin and incubated for 5 days at 30³C. Of the 90 spots, 65 contained no growth, indicating complete loss of the resistance determinant. In the remaining 25 spots there was either fast and con£uent growth, or 5^20 single colonies of normal size and appearance. (Streaking out the con£uent growth also produced normal-sized colonies; Fig. 2b.) The viomycin resistance of cultures obtained from these normal-sized colonies was stable in the absence of selection and the cultures grew normally also at 37³C. In a parallel experiment, the stability of the primary small VioR transformants at 37³C was tested by patching them on viomycin agar and incubating at 37³C (instead of 30³C). Con£uent or normal-size single-colony growth was observed in 6% of the patches (nine out of 150) and there was no growth in the rest. These results suggested that unstable primary transformants produced stable VioR segregants. 3.3. Southern hybridisation to detect Tn4560 in the stable M. smegmatis transformants
Fig. 2. M. smegmatis pUC1169 transformants growing on supplemented 7H11 agar containing 30 Wg ml31 viomycin after incubation for 1 week at 30³C. a: Primary unstable VioR transformants : small and of irregular shape. b: Colonies of cells containing Tn4560 integrated in the chromosome : large colonies, `normal' morphology. The scale bar indicates 2 mm.
ony phenotype was not caused by low level expression of viomycin resistance, and it was suspected that the viomycin resistance was highly unstable. (Using less than 30 Wg ml31 viomycin resulted in background growth of non-resistant bacteria.) 3.2. Segregation of stable VioR colonies from the unstable primary transformants Re-streaking the primary transformants on viomycin agar and incubation at 30³C resulted in slow con£uent growth, indicating that the viomycin resistance was inherited. To test the stability of the viomycin resistance in the absence of selection, 90 primary transformants were
Genomic DNA from independent normal-sized colonies was tested for the presence of Tn4560 using Southern hybridisation and probing with labelled pUC1169. Among 18 independent samples, 16 showed the expected Tn4560-internal BclI fragments and two bands which were di¡erent in each sample, presumably representing Tn4560^chromosome junction fragments. Each strain contained only one copy of Tn4560, and vector-speci¢c bands were missing. One of the 18 samples lacked the 1.01-kb transposon-internal BclI fragment and had presumably su¡ered a deletion. Another sample contained no transposon and was thus a spontaneous VioR mutant. Ten transformants containing intact Tn4560 are shown in Fig. 3. These results suggested that Tn4560 had inserted into di¡erent sites in independent VioR M. smegmatis isolates. 3.4. Tn4560 target sequences in M. smegmatis The sequences £anking 10 transposon insertions were ampli¢ed using LM-PCR, and sequenced using the appropriate transposon primer. As in Streptomyces, Tn4560 insertion generated 5-bp direct repeat duplications. The average G+C content of the duplications was 48% (low for mycobacteria), but the average G+C content of the entire sequences shown in Table 1 was 65% which was typical for mycobacteria. Aligning the sequences at the insertion sites revealed no consensus sequence (Table 1). BLAST searches of the M. smegmatis database (http:// www.tigr.org) gave a perfect match for each of the target sequences, and provided further surrounding sequences.
FEMSLE 10292 21-1-02
244
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
were annotated with a functional prediction, suggesting that they may be important for the bacteria at least under certain conditions. The two remaining ORFs were similar to hypothetical proteins (Nos. 6 and 9). 3.5. PFGE analysis of M. smegmatis strains containing Tn4560
Fig. 3. Southern blot of BclI-digested total DNA from 10 stable M. smegmatis VioR transformants probed with labelled pUC1169. The two internal Tn4560 bands (*) are present in all 10 transformants and the pMT660-speci¢c bands (b) are missing. The left (R) and right (S) end Tn4560^pMT660 junction bands are replaced by two bands of sizes. s 1.3 kb and s 1.9 kb, respectively, containing the transposon end and £anking chromosomal DNA. In transformant 2, the second £anking band is present slightly above the 4.36-kb internal band, clearly visible in shorter exposures.
Nine out of 10 inserts were in coding regions, suggesting that there was no marked bias for intergenic regions which make up about 10% of bacterial genomes and tend to have a lower G+C content than the coding sequences. Only one of the inserts (No. 6 in Table 1) was between ORFs. All but two of the ORFs were similar to database entries that
Pulsed ¢eld gel electrophoresis (PFGE) of AseI-digested DNA was used to assess the genomic distribution of the 10 Tn4560 inserts (analysed earlier in Fig. 3). Tn4560 has no AseI site, and transposon-carrying AseI fragments were identi¢ed by the 8.6-kb size increase, and by hybridisation to labelled pUC1169 (Fig. 4). Transposon insertions were found in eight di¡erent AseI fragments (the M. smegmatis mc2 155 chromosome produces 17 AseI macrorestriction fragments; K. Dharmalingam, personal communication). Two fragments of V550 kb and V155 kb were targeted twice (at di¡erent sites) by Tn4560. This suggested that the transposon insertions were scattered around the M. smegmatis chromosome. 3.6. Replication of pUC1169 in M. smegmatis The Streptomyces plasmid pUC1169 was not expected to replicate in M. smegmatis. Several observations suggested, however, that pUC1169 replicated in M. smegmatis, and that this replication was important for obtaining transposon mutants: (1) insertion of an Escherichia coli plasmid into the BamHI site in the pUC1169 rep gene
Table 1 Sequence analysis of Tn4560 insertion sites in M. smegmatis Target sequence 1
GGCCGCCCAT CCCAAGCTGG 2 CCCGCTACCG CGACGGACGC 3 CGAGCGCGAC GGGCAACTGC 4 GTTCTGCGGC GAGGATCCCT 5 GCGACGCTGG CCCGCTCGGG 6 GCGAGCGACC CTGCGACAAA 7 CCAGGTGAAG CCACGGCCGC 8 AGGACCGAAA GATGATCGAA 9 TGCGCGCCAA CATGCTCGAT 10 CATCGCGTCC CCTCAAGGTC
CTCTACAAGG GGATCACCAG CGTGCACCGC ACCGCGTCGG TGATCAACAG TCGCCGAGTT ATGTCGATCG CGGTGGACTA GCTCGGTGGG TGTCCAGGTT GGCAGTGGCG GCGCCGAACT CTCGTCGAAG AGACCGTCGT CACCCACTGC GACGCCTCGG CGCCGGGGCC GGGGTCGCGC GGCACGGCTC GTCGGGGTCA
Putative ORF disruption by or in the vicinity of Tn4560 insertion
Closest homologue
TGTTT
LuxR-type transcriptional regulator
Streptomyces coelicolor LipR
GTCTA
Acyl CoA thioesterase
Streptomyces coelicolor TesB
ATCCT
Hydrogenase expression/formation protein
Rhizobium leguminosarum HypE
GCTAC
Oxidoreductase
Streptomyces coelicolor oxidoreductase SCJ4.25c
GTATC
Long-chain fatty-acid CoA ligase
Mycobacterium tuberculosis FadD8
TTCAC TTCCC
Upstream of two divergent ORFs, an acyl CoA dehydrogenase and a hypothetical protein Dehalogenase
Mycobacterium tuberculosis FadE32 and Mesorhizobium loti hypothetical protein Pseudomonas aeruginosa haloacid dehalogenase
TATCC
Aldose
Mesorhizobium loti aldolase
GGCGT
Hypothetical protein
GACTA
FAD-binding monooxygenases
Streptomyces coelicolor hypothetical protein SC5C7.24c Mycobacterium tuberculosis Rv1260
Five-bp insertion sites are shown in bold. Tn4560^chromosome junction sequences were compared to the incomplete M. smegmatis genome sequence using the TIGR (The Institute for Genomic Research) BLAST site (http://www. tigr.org/cgi-bin/BlastSearch/blast.cgi ?organism = m_smegmatis). The contig giving a perfect match was analysed for the presence of ORFs using FRAME analysis. Protein database was then searched for homologues of the disrupted and neighbouring ORFs using NCBI (National Centre for Biotechnology Information) BLAST.
FEMSLE 10292 21-1-02
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
245
From the intensities of the tsr-speci¢c PCR bands and the pUC1169-speci¢c bands from the Southern hybridisation, it was estimated that at least 105 copies of pUC1169 were present in the above samples, each of which was derived from a single transformant. It is worth noting that pUC1169 could not be detected by PCR in any of the strains that contained Tn4560 integrated in the chromosome. The high e¤ciency of Tn4560 delivery was probably helped by this unexpected replication of pUC1169 (at 30³C) in M. smegmatis: up to 28% of the cells that received pUC1169 by electroporation (small VioR colonies) produced a stable transposon insertion. 3.7. Utility of Tn4560 in mycobacteria
Fig. 4. PFGE analysis of AseI-digested genomic DNA of the 10 stable VioR transformants. a: Ethidium bromide-stained 1% agarose gel. b: Southern blot probed with labelled pUC1169. Lane C, untransformed M. smegmatis mc2 155; lanes 1^10, VioR transformants ; lane M1, V DNA concatemers ; lane M2, Saccharomyces cerevisiae chromosomes. Solid arrows indicate new bands and broken arrows show regions where a band is missing. PFGE was done as described earlier [5]. The 0.5U Tris-borate-EDTA electrophoresis bu¡er was supplemented with 50 WM thiourea (Sigma) to prevent oxidative degradation of DNA. Electrophoresis was for 24 h at 6 V cm31 and 14³C, using pulse times from 5 to 25 s (linear ramp) and an angle of 120³.
abolished replication in Streptomyces, and no VioR colonies were obtained on electroporation of M. smegmatis. (2) Transposition was not observed at 37³C, a non-permissive temperature for replication of the temperature-sensitive pUC1169 replicon. Tn4560 has been shown to transpose e¤ciently at 37³C in Streptomyces [18]. (3) Southern analysis of total DNA that was isolated from non-selective liquid cultures that did not give rise to VioR colonies (see above) produced very faint pUC1169-speci¢c bands after 1 week exposure in seven of 18 samples (data not shown). (4) PCR ampli¢cation (30 cycles) of part of the tsr gene, which was speci¢c for pUC1169, gave bands of the expected size from the samples that were positive in the above Southern blot, but not from samples that did not show pUC1169-speci¢c bands. (5) Re-streaking primary VioR transformants on 7H11 agar containing viomycin and incubation at 30³C for 7 days produced many small VioR colonies in which the pUC1169-speci¢c tsr gene was detected using PCR.
Our results suggested that Tn4560 was suitable for producing a great variety of gene-inactivating mutations in M. smegmatis. In addition, because of its Streptomyces origin, Tn4560 may be less likely to be subject to interference from resident transposons than Mycobacterium elements. A transposon derived from the insect element Himar1 has been shown to function in M. smegmatis, but the preference for AT-centred target sequences probably limits their randomness in high-G+C organisms like the mycobacteria [4]. This study has also demonstrated the e¡ective use of a Streptomyces viomycin resistance gene in mycobacteria. There is no known cross-resistance between viomycin and the commonly used kanamycin and hygromycin resistance markers, allowing the use of Tn4560 in conjunction with most currently used mycobacterial vectors. Spontaneous occurrence of VioR mutants has been described [19] and was observed in one out of 18 strains that were analysed. Lack of pUC1169 replication at 37³C, the normal incubation temperature for M. smegmatis, provided a simple selection for strains containing Tn4560 insertions. This selection needed to be done separately on independent primary transformants because pooling of multiple transformants produced many sibling colonies. For the construction of representative mutant libraries, it would be advantageous to use a temperature-sensitive phage [20] or a conditionally replicating plasmid with counterselectable markers [3]. Tn4560 was also shown to transpose into seemingly random locations in the genomes of the slow-growing M. tuberculosis H37Rv and M. bovis BCG (Bhatt, A., Stewart, R., Young, D.B. and Kieser, T., unpublished data). As in M. smegmatis, large VioR colonies containing Tn4560 insertions segregated from small primary VioR transformants. The e¤cient transposition of IS117 [5] and of Tn4560 made it likely that other Streptomyces transposable elements may also function in mycobacteria, thus opening up a totally new `tool box' for the genetic manipulation of mycobacteria.
FEMSLE 10292 21-1-02
246
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
Acknowledgements We thank Kay Fowler for providing pUC1169 cloned into an E. coli vector, and Douglas Young for his comments on a draft of the manuscript. A.B. was supported by a PhD studentship from the John Innes Foundation, Norwich, UK. References [1] Guilhot, C., Otal, I., Van Rompaey, I., Martin, C. and Gicquel, B. (1994) E¤cient transposition in mycobacteria : construction of Mycobacterium smegmatis insertional mutant libraries. J. Bacteriol. 176, 535^539. [2] Harris, N.B., Feng, Z., Liu, X., Cirillo, S.L., Cirillo, J.D. and Barletta, R.G. (1999) Development of a transposon mutagenesis system for Mycobacterium avium subsp. paratuberculosis. FEMS Microbiol. Lett. 175, 21^26. [3] Pelicic, V., Jackson, M., Reyrat, J.M., Jacobs Jr., W.R., Gicquel, B. and Guilhot, C. (1997) E¤cient allelic exchange and transposon mutagenesis in Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 94, 10955^10960. [4] Rubin, E.J., Akerley, B.J., Novik, V.N., Lampe, D.J., Husson, R.N. and Mekalanos, J.J. (1999) In vivo transposition of mariner-based elements in enteric bacteria and mycobacteria. Proc. Natl. Acad. Sci. USA 96, 1645^1650. [5] Bhatt, A. and Kieser, T. (1999) Transposition of IS117 of Streptomyces coelicolor A3 in Mycobacterium smegmatis. Microbiology 145, 1201^1207. [6] Chung, S.T. (1987) Tn4556, a 6.8 kb transposable element of Streptomyces fradiae. J. Bacteriol. 169, 4436^4441. [7] Siemieniak, D.R., Slightom, J.L. and Chung, S.T. (1990) Nucleotide sequence of Streptomyces fradiae transposable element Tn4556 : a class-II transposon related to Tn3. Gene 86, 1^9. [8] Sherratt, D. (1989) Tn3 and related transposable elements: Site-speci¢c recombination and transposition. In: Mobile DNA (Berg, D. and Howe, M., Eds.), pp. 163^184. American Society for Microbiology, Washington, DC.
[9] Olson, E.R. and Chung, S.T. (1988) Transposon Tn4556 of Streptomyces fradiae: nucleotide sequence of the ends and the target sites. J. Bacteriol. 170, 1955^1957. [10] Snapper, S.B., Melton, R.E., Kieser, T., Mustafa, S. and Jacobs Jr., W.R. (1990) Isolation and characterization of e¤cient plasmid transformation mutants of Mycobacterium smegmatis. Mol. Microbiol. 4, 1911^1919. [11] Kieser, T., Bibb, M.J., Buttner, M.J., Chater, K.F. and Hopwood, D.A. (2000) Practical Streptomyces Genetics. The John Innes Foundation, Norwich. [12] Birch, A.W. and Cullum, J. (1985) Temperature-sensitive mutants of the Streptomyces plasmid pIJ702. J. Gen. Microbiol. 131, 1299^1303. [13] Bibb, M.J., Ward, J.M. and Cohen, S.N. (1985) Nucleotide sequences encoding and promoting expression of three antibiotic resistance genes indigenous to Streptomyces. Mol. Gen. Genet. 199, 26^36. [14] Snapper, S.B., Lugosi, L., Jekkel, A., Melton, R.E., Kieser, T., Bloom, B.R. and Jacobs Jr., W.R. (1988) Lysogeny and transformation in mycobacteria : stable expression of foreign genes. Proc. Natl. Acad. Sci. USA 85, 6987^6991. [15] Prod'hom, G., Lagier, B., Pelicic, V., Hance, A.J., Gicquel, B. and Guilhot, C. (1998) A reliable ampli¢cation technique for the characterization of genomic DNA sequences £anking insertion sequences. FEMS Microbiol. Lett. 158, 75^81. [16] Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) Basic local alignment search tool. J. Mol. Biol. 215, 403^410. [17] Bibb, M.J., Findlay, P.R. and Johnson, M.W. (1984) The relationship between base composition and codon usage in bacterial genes and its use for the simple and reliable identi¢cation of protein-coding sequences. Gene 30, 157^166. [18] Ikeda, H., Takada, Y., Chang-Hong, P., Tanaka, H. and Omura, S. (1993) Transposon mutagenesis by Tn4560 and applications with avermectin-producing Streptomyces avermitilis. J. Bacteriol. 175, 2077^2082. [19] Yamada, T., Masuda, K., Shoji, K. and Hori, M. (1972) Analysis of ribosomes from viomycin sensitive and resistant strains of Mycobacterium smegmatis. J. Bacteriol. 112, 1^6. [20] Bardarov, S., Kriakov, J., Carriere, C., Yu, S., Vaamonde, C., McAdam, R.A., Bloom, B.R., Hatfull, G.F. and Jacobs Jr., W.R. (1997) Conditionally replicating mycobacteriophages : a system for transposon delivery to Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 94, 10961^10966.
FEMSLE 10292 21-1-02
www.fems-microbiology.org
Transposition of Tn4560 of Streptomyces fradiae in Mycobacterium smegmatis Apoorva Bhatt b
a;
*, Graham R. Stewart b , Tobias Kieser
a
a Department of Genetics, John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, UK Centre for Molecular Microbiology and Infection, Imperial College, South Kensington, London SW7 2AZ, UK
Received 4 October 2001; received in revised form 15 November 2001; accepted 21 November 2001 First published online 17 December 2001
Abstract Tn4560 (8.6 kb) was derived from Tn4556, a Tn3-like element from Streptomyces fradiae. It contains a viomycin resistance gene that has not been used previously for selection in mycobacteria. Tn4560, cloned in a Streptomyces plasmid, was introduced by electroporation into Mycobacterium smegmatis mc2 155. Tn4560 transposed into the host genome: there was no obvious target sequence preference, and insertions were in or near several conserved open reading frames. The insertions were located far apart on different AseI macrorestriction fragments. Unexpectedly, the transposon delivery plasmid, pUC1169, derived from the Streptomyces multicopy plasmid pIJ101, replicated partially in M. smegmatis, but was lost spontaneously during subculture. Replication of pUC1169 probably contributed to the relatively high efficiency of Tn4560 delivery: up to 28% of the potential M. smegmatis transformants acquired a stable transposon insertion. The data indicated that Tn4560 may be useful for random mutagenesis of M. smegmatis. ß 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Mycobacterium smegmatis mc2 155; Tn4556; Viomycin resistance; pIJ101; pMT660 ; Transposon mutagenesis
1. Introduction Transposable elements derived from Streptomyces spp. might also function in the taxonomically related mycobacteria and could be useful additions to the small list of transposable elements available for use in genetic manipulation of mycobacteria [1^4]. Indeed, one such element, IS117, has been shown to transpose into speci¢c sites in Mycobacterium smegmatis [5]. Tn4560, a viomycin-resistant derivative of the Streptomyces fradiae transposable element Tn4556 [6], was tested for random transposition in M. smegmatis. Tn4556 resembles Tn3 which transposes replicatively via a co-integrate between the donor and recipient replicons [7,8]. Co-integrates are converted into the end products of transposition by site-speci¢c recombination (resolution) mediated by a transposon-encoded resolvase and a recombination site, res, within the transposon. Tn4556 derivatives transposed into random sites (one copy
* Corresponding author. Present address: Department of Biochemistry, University of Cambridge, 80, Tennis Court Road, Cambridge CB2 1GA, UK. Tel.: +44 (1223) 333657; Fax: +44 (1223) 766002. E-mail address : [email protected] (A. Bhatt).
per replicon) in many streptomycetes and 5-bp duplications of the target sequence were created [9]. Here we show that Tn4560 transposed in M. smegmatis mc2 155. No target sequence speci¢city was observed and insertions were found inside or near many conserved open reading frames (ORFs), indicating that Tn4560 has potential for mutagenesis of M. smegmatis. 2. Materials and methods 2.1. Bacterial strains, transformation, media and DNA extraction For the preparation of electrocompetent cells, M. smegmatis mc2 155 [10] was grown in 7H9 broth (Difco) containing 0.05% Tween 80 and 10% ODAC supplement (oleic acid, dextrose, albumin, catalase; Difco). For all other procedures the strain was grown in Lennox broth+0.05% Tween 80 [11]. Genomic DNA was extracted from M. smegmatis mc2 155 by the salting-out method [11]. The Tn4560 delivery plasmid pUC1169 (Fig. 1; [6]) was propagated and isolated from Streptomyces as described earlier [11]. pYUB12, a kanamycin-resistant derivative of
0378-1097 / 02 / $22.00 ß 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 0 1 ) 0 0 5 5 0 - X
FEMSLE 10292 21-1-02
242
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
the autonomously replicating plasmid pAL5000 [14], was used as a control for determining electroporation e¤ciency. Electroporation of M. smegmatis mc2 155 was performed as described before [10]. Antibiotic-resistant transformants were selected on 7H11 agar (Difco) containing 10% ODAC, 0.5% glycerol and, as appropriate, 10 Wg ml31 kanamycin sulfate (Sigma) or 30 Wg ml31 viomycin (gift from P¢zer to D.A. Hopwood). 2.2. PCR methods The oligonucleotide primers TSR1 (5P-TCGACTCCTCGATCGTCAACC-3P) and TSR2 (5P-GAGCGTCATCAGCTGCATACC-3P) were used to amplify a part of the tsr gene of pUC1169 using 30 cycles of 96³C for 1 min, 57³C for 2 min, and 72³C for 1 min. Ligationmediated polymerase chain reaction (LM-PCR) of the Tn4560^chromosome junctions was based on a previously described method [15] using the oligonucleotides SalI-AD (5P-TCGACATTAATG-3P) and End primer (5-GGATAAGAATTCATTAATG-3P) to form a partially double-stranded adapter (SalI adapter) for ligation to SalI cohesive ends. Genomic DNA (0.4^0.5 Wg) was digested with 5 U SalI and then ligated to 25 nmol of the SalI adapter in 20 Wl, and 0.5^1 Wl of the ligation mix was then used as template for PCR ampli¢cation (in single block PTC-1001, MJ Research) using primers Tn4560L1 (5P-GCACTAGTCGGCAGCCTGCTCATCCG-3P) or Tn4560-R1 (5P-GCACTAGTTTCACGTACGGCCCGGAG-3P), and the above End primer. The PCR conditions were : hot start of 95³C for 9 min, denaturation at 95³C for 30 s, annealing at 55³C for 30 s, extension at 72³C for 90 s, total number of cycles 35. For transposon^chromosome junctions that could not be ampli¢ed from SalI-digested DNA, NgoAIV (Gibco BRL) was used in combination with the complementary oligonucleotides UNIV5AD1 (5P-GACTCGCGAATTCCGACAGTTGA-3P) and NgoAIV-AD2 (5P-CCGGTCAACTGTCG-3P). UNIV5AD1, and the transposon-speci¢c primers Tn4560-L2 (5PCGTCAAAAACTCGACCGGCA-3P) or Tn4560-R2 (5PCAGAATTCCCCCTTGCCACAGATAACAG-3P) were used for PCR as above.
3. Results and discussion 3.1. Introduction of Tn4560 into M. smegmatis mc2 155 by electroporation Tn4560 was delivered on pUC1169, a Streptomyces plasmid that was not expected to replicate in mycobacteria (Fig. 1). pUC1169 (1 Wg) was isolated from Streptomyces lividans or from Streptomyces coelicolor and introduced by electroporation into M. smegmatis mc2 155. Selection was for viomycin resistance encoded by the transposon. No viomycin-resistant (VioR ) transformants were obtained when incubation was at 37³C, the normal growth temperature for M. smegmatis. This was surprising because in Streptomyces Tn4560 transposes e¤ciently at 37³C [18]. However, reproducibly (four experiments) after 5 days incubation at 30³C between V103 and 5U103 very small VioR colonies of irregular shape were observed (Fig. 2a). The apparent transformation frequency was approximately 10% of the frequency obtained with the autonomously replicating pYUB12. There was no VioR growth on the no-DNA control. Using 30, 50 or 100 Wg ml31 viomycin gave the same number of transformants, and the number of colonies did not decrease with increasing viomycin concentration. This indicated that the small col-
2.3. DNA sequencing Puri¢ed PCR products (QIAquick gel extraction kit) were sequenced using the Amersham Thermo Sequenase1 dye terminator or PE Applied Biosystems BigDye1 terminator sequencing kits and Tn4560-speci¢c primers. Sequences were compared with the M. smegmatis database at http://www.tigr.org, and with the Mycobacterium tuberculosis and Mycobacterium bovis databases at http:// www.sanger.ac.uk. BLAST searches [16] were performed at http://www.ncbi.nlm.nih.gov/BLAST/, and ORFs were predicted using the FRAME program [17] at http:// watson.nih.go.jp/ujun/cgi-bin/frameplot.pl.
Fig. 1. Map of pUC1169. Tn4560 (shaded region) was cloned as a BamHI fragment into the BglII site of pMT660 [12], a temperature-sensitive derivative of the multicopy Streptomyces plasmid pIJ101 which replicates by a rolling circle mechanism. vph, viomycin phosphotransferase conferring viomycin resistance [13]; tnpA, transposase; tnpR, putative resolvase ; res, putative co-integrate resolution site ; IR, 38-bp terminal inverted repeats; rep, pIJ101 replicase; ori, pIJ101 ds origin of replication; tsr, thiostrepton resistance gene useful for selection in streptomycetes but not in mycobacteria.
FEMSLE 10292 21-1-02
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
243
propagated individually for 3 days at 30³C in 10 ml L-broth+Tween 80 without antibiotic, producing dense cultures. Then, 5-Wl samples from each of these cultures were spotted onto 7H11 agar plates containing viomycin and incubated for 5 days at 30³C. Of the 90 spots, 65 contained no growth, indicating complete loss of the resistance determinant. In the remaining 25 spots there was either fast and con£uent growth, or 5^20 single colonies of normal size and appearance. (Streaking out the con£uent growth also produced normal-sized colonies; Fig. 2b.) The viomycin resistance of cultures obtained from these normal-sized colonies was stable in the absence of selection and the cultures grew normally also at 37³C. In a parallel experiment, the stability of the primary small VioR transformants at 37³C was tested by patching them on viomycin agar and incubating at 37³C (instead of 30³C). Con£uent or normal-size single-colony growth was observed in 6% of the patches (nine out of 150) and there was no growth in the rest. These results suggested that unstable primary transformants produced stable VioR segregants. 3.3. Southern hybridisation to detect Tn4560 in the stable M. smegmatis transformants
Fig. 2. M. smegmatis pUC1169 transformants growing on supplemented 7H11 agar containing 30 Wg ml31 viomycin after incubation for 1 week at 30³C. a: Primary unstable VioR transformants : small and of irregular shape. b: Colonies of cells containing Tn4560 integrated in the chromosome : large colonies, `normal' morphology. The scale bar indicates 2 mm.
ony phenotype was not caused by low level expression of viomycin resistance, and it was suspected that the viomycin resistance was highly unstable. (Using less than 30 Wg ml31 viomycin resulted in background growth of non-resistant bacteria.) 3.2. Segregation of stable VioR colonies from the unstable primary transformants Re-streaking the primary transformants on viomycin agar and incubation at 30³C resulted in slow con£uent growth, indicating that the viomycin resistance was inherited. To test the stability of the viomycin resistance in the absence of selection, 90 primary transformants were
Genomic DNA from independent normal-sized colonies was tested for the presence of Tn4560 using Southern hybridisation and probing with labelled pUC1169. Among 18 independent samples, 16 showed the expected Tn4560-internal BclI fragments and two bands which were di¡erent in each sample, presumably representing Tn4560^chromosome junction fragments. Each strain contained only one copy of Tn4560, and vector-speci¢c bands were missing. One of the 18 samples lacked the 1.01-kb transposon-internal BclI fragment and had presumably su¡ered a deletion. Another sample contained no transposon and was thus a spontaneous VioR mutant. Ten transformants containing intact Tn4560 are shown in Fig. 3. These results suggested that Tn4560 had inserted into di¡erent sites in independent VioR M. smegmatis isolates. 3.4. Tn4560 target sequences in M. smegmatis The sequences £anking 10 transposon insertions were ampli¢ed using LM-PCR, and sequenced using the appropriate transposon primer. As in Streptomyces, Tn4560 insertion generated 5-bp direct repeat duplications. The average G+C content of the duplications was 48% (low for mycobacteria), but the average G+C content of the entire sequences shown in Table 1 was 65% which was typical for mycobacteria. Aligning the sequences at the insertion sites revealed no consensus sequence (Table 1). BLAST searches of the M. smegmatis database (http:// www.tigr.org) gave a perfect match for each of the target sequences, and provided further surrounding sequences.
FEMSLE 10292 21-1-02
244
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
were annotated with a functional prediction, suggesting that they may be important for the bacteria at least under certain conditions. The two remaining ORFs were similar to hypothetical proteins (Nos. 6 and 9). 3.5. PFGE analysis of M. smegmatis strains containing Tn4560
Fig. 3. Southern blot of BclI-digested total DNA from 10 stable M. smegmatis VioR transformants probed with labelled pUC1169. The two internal Tn4560 bands (*) are present in all 10 transformants and the pMT660-speci¢c bands (b) are missing. The left (R) and right (S) end Tn4560^pMT660 junction bands are replaced by two bands of sizes. s 1.3 kb and s 1.9 kb, respectively, containing the transposon end and £anking chromosomal DNA. In transformant 2, the second £anking band is present slightly above the 4.36-kb internal band, clearly visible in shorter exposures.
Nine out of 10 inserts were in coding regions, suggesting that there was no marked bias for intergenic regions which make up about 10% of bacterial genomes and tend to have a lower G+C content than the coding sequences. Only one of the inserts (No. 6 in Table 1) was between ORFs. All but two of the ORFs were similar to database entries that
Pulsed ¢eld gel electrophoresis (PFGE) of AseI-digested DNA was used to assess the genomic distribution of the 10 Tn4560 inserts (analysed earlier in Fig. 3). Tn4560 has no AseI site, and transposon-carrying AseI fragments were identi¢ed by the 8.6-kb size increase, and by hybridisation to labelled pUC1169 (Fig. 4). Transposon insertions were found in eight di¡erent AseI fragments (the M. smegmatis mc2 155 chromosome produces 17 AseI macrorestriction fragments; K. Dharmalingam, personal communication). Two fragments of V550 kb and V155 kb were targeted twice (at di¡erent sites) by Tn4560. This suggested that the transposon insertions were scattered around the M. smegmatis chromosome. 3.6. Replication of pUC1169 in M. smegmatis The Streptomyces plasmid pUC1169 was not expected to replicate in M. smegmatis. Several observations suggested, however, that pUC1169 replicated in M. smegmatis, and that this replication was important for obtaining transposon mutants: (1) insertion of an Escherichia coli plasmid into the BamHI site in the pUC1169 rep gene
Table 1 Sequence analysis of Tn4560 insertion sites in M. smegmatis Target sequence 1
GGCCGCCCAT CCCAAGCTGG 2 CCCGCTACCG CGACGGACGC 3 CGAGCGCGAC GGGCAACTGC 4 GTTCTGCGGC GAGGATCCCT 5 GCGACGCTGG CCCGCTCGGG 6 GCGAGCGACC CTGCGACAAA 7 CCAGGTGAAG CCACGGCCGC 8 AGGACCGAAA GATGATCGAA 9 TGCGCGCCAA CATGCTCGAT 10 CATCGCGTCC CCTCAAGGTC
CTCTACAAGG GGATCACCAG CGTGCACCGC ACCGCGTCGG TGATCAACAG TCGCCGAGTT ATGTCGATCG CGGTGGACTA GCTCGGTGGG TGTCCAGGTT GGCAGTGGCG GCGCCGAACT CTCGTCGAAG AGACCGTCGT CACCCACTGC GACGCCTCGG CGCCGGGGCC GGGGTCGCGC GGCACGGCTC GTCGGGGTCA
Putative ORF disruption by or in the vicinity of Tn4560 insertion
Closest homologue
TGTTT
LuxR-type transcriptional regulator
Streptomyces coelicolor LipR
GTCTA
Acyl CoA thioesterase
Streptomyces coelicolor TesB
ATCCT
Hydrogenase expression/formation protein
Rhizobium leguminosarum HypE
GCTAC
Oxidoreductase
Streptomyces coelicolor oxidoreductase SCJ4.25c
GTATC
Long-chain fatty-acid CoA ligase
Mycobacterium tuberculosis FadD8
TTCAC TTCCC
Upstream of two divergent ORFs, an acyl CoA dehydrogenase and a hypothetical protein Dehalogenase
Mycobacterium tuberculosis FadE32 and Mesorhizobium loti hypothetical protein Pseudomonas aeruginosa haloacid dehalogenase
TATCC
Aldose
Mesorhizobium loti aldolase
GGCGT
Hypothetical protein
GACTA
FAD-binding monooxygenases
Streptomyces coelicolor hypothetical protein SC5C7.24c Mycobacterium tuberculosis Rv1260
Five-bp insertion sites are shown in bold. Tn4560^chromosome junction sequences were compared to the incomplete M. smegmatis genome sequence using the TIGR (The Institute for Genomic Research) BLAST site (http://www. tigr.org/cgi-bin/BlastSearch/blast.cgi ?organism = m_smegmatis). The contig giving a perfect match was analysed for the presence of ORFs using FRAME analysis. Protein database was then searched for homologues of the disrupted and neighbouring ORFs using NCBI (National Centre for Biotechnology Information) BLAST.
FEMSLE 10292 21-1-02
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
245
From the intensities of the tsr-speci¢c PCR bands and the pUC1169-speci¢c bands from the Southern hybridisation, it was estimated that at least 105 copies of pUC1169 were present in the above samples, each of which was derived from a single transformant. It is worth noting that pUC1169 could not be detected by PCR in any of the strains that contained Tn4560 integrated in the chromosome. The high e¤ciency of Tn4560 delivery was probably helped by this unexpected replication of pUC1169 (at 30³C) in M. smegmatis: up to 28% of the cells that received pUC1169 by electroporation (small VioR colonies) produced a stable transposon insertion. 3.7. Utility of Tn4560 in mycobacteria
Fig. 4. PFGE analysis of AseI-digested genomic DNA of the 10 stable VioR transformants. a: Ethidium bromide-stained 1% agarose gel. b: Southern blot probed with labelled pUC1169. Lane C, untransformed M. smegmatis mc2 155; lanes 1^10, VioR transformants ; lane M1, V DNA concatemers ; lane M2, Saccharomyces cerevisiae chromosomes. Solid arrows indicate new bands and broken arrows show regions where a band is missing. PFGE was done as described earlier [5]. The 0.5U Tris-borate-EDTA electrophoresis bu¡er was supplemented with 50 WM thiourea (Sigma) to prevent oxidative degradation of DNA. Electrophoresis was for 24 h at 6 V cm31 and 14³C, using pulse times from 5 to 25 s (linear ramp) and an angle of 120³.
abolished replication in Streptomyces, and no VioR colonies were obtained on electroporation of M. smegmatis. (2) Transposition was not observed at 37³C, a non-permissive temperature for replication of the temperature-sensitive pUC1169 replicon. Tn4560 has been shown to transpose e¤ciently at 37³C in Streptomyces [18]. (3) Southern analysis of total DNA that was isolated from non-selective liquid cultures that did not give rise to VioR colonies (see above) produced very faint pUC1169-speci¢c bands after 1 week exposure in seven of 18 samples (data not shown). (4) PCR ampli¢cation (30 cycles) of part of the tsr gene, which was speci¢c for pUC1169, gave bands of the expected size from the samples that were positive in the above Southern blot, but not from samples that did not show pUC1169-speci¢c bands. (5) Re-streaking primary VioR transformants on 7H11 agar containing viomycin and incubation at 30³C for 7 days produced many small VioR colonies in which the pUC1169-speci¢c tsr gene was detected using PCR.
Our results suggested that Tn4560 was suitable for producing a great variety of gene-inactivating mutations in M. smegmatis. In addition, because of its Streptomyces origin, Tn4560 may be less likely to be subject to interference from resident transposons than Mycobacterium elements. A transposon derived from the insect element Himar1 has been shown to function in M. smegmatis, but the preference for AT-centred target sequences probably limits their randomness in high-G+C organisms like the mycobacteria [4]. This study has also demonstrated the e¡ective use of a Streptomyces viomycin resistance gene in mycobacteria. There is no known cross-resistance between viomycin and the commonly used kanamycin and hygromycin resistance markers, allowing the use of Tn4560 in conjunction with most currently used mycobacterial vectors. Spontaneous occurrence of VioR mutants has been described [19] and was observed in one out of 18 strains that were analysed. Lack of pUC1169 replication at 37³C, the normal incubation temperature for M. smegmatis, provided a simple selection for strains containing Tn4560 insertions. This selection needed to be done separately on independent primary transformants because pooling of multiple transformants produced many sibling colonies. For the construction of representative mutant libraries, it would be advantageous to use a temperature-sensitive phage [20] or a conditionally replicating plasmid with counterselectable markers [3]. Tn4560 was also shown to transpose into seemingly random locations in the genomes of the slow-growing M. tuberculosis H37Rv and M. bovis BCG (Bhatt, A., Stewart, R., Young, D.B. and Kieser, T., unpublished data). As in M. smegmatis, large VioR colonies containing Tn4560 insertions segregated from small primary VioR transformants. The e¤cient transposition of IS117 [5] and of Tn4560 made it likely that other Streptomyces transposable elements may also function in mycobacteria, thus opening up a totally new `tool box' for the genetic manipulation of mycobacteria.
FEMSLE 10292 21-1-02
246
A. Bhatt et al. / FEMS Microbiology Letters 206 (2002) 241^246
Acknowledgements We thank Kay Fowler for providing pUC1169 cloned into an E. coli vector, and Douglas Young for his comments on a draft of the manuscript. A.B. was supported by a PhD studentship from the John Innes Foundation, Norwich, UK. References [1] Guilhot, C., Otal, I., Van Rompaey, I., Martin, C. and Gicquel, B. (1994) E¤cient transposition in mycobacteria : construction of Mycobacterium smegmatis insertional mutant libraries. J. Bacteriol. 176, 535^539. [2] Harris, N.B., Feng, Z., Liu, X., Cirillo, S.L., Cirillo, J.D. and Barletta, R.G. (1999) Development of a transposon mutagenesis system for Mycobacterium avium subsp. paratuberculosis. FEMS Microbiol. Lett. 175, 21^26. [3] Pelicic, V., Jackson, M., Reyrat, J.M., Jacobs Jr., W.R., Gicquel, B. and Guilhot, C. (1997) E¤cient allelic exchange and transposon mutagenesis in Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 94, 10955^10960. [4] Rubin, E.J., Akerley, B.J., Novik, V.N., Lampe, D.J., Husson, R.N. and Mekalanos, J.J. (1999) In vivo transposition of mariner-based elements in enteric bacteria and mycobacteria. Proc. Natl. Acad. Sci. USA 96, 1645^1650. [5] Bhatt, A. and Kieser, T. (1999) Transposition of IS117 of Streptomyces coelicolor A3 in Mycobacterium smegmatis. Microbiology 145, 1201^1207. [6] Chung, S.T. (1987) Tn4556, a 6.8 kb transposable element of Streptomyces fradiae. J. Bacteriol. 169, 4436^4441. [7] Siemieniak, D.R., Slightom, J.L. and Chung, S.T. (1990) Nucleotide sequence of Streptomyces fradiae transposable element Tn4556 : a class-II transposon related to Tn3. Gene 86, 1^9. [8] Sherratt, D. (1989) Tn3 and related transposable elements: Site-speci¢c recombination and transposition. In: Mobile DNA (Berg, D. and Howe, M., Eds.), pp. 163^184. American Society for Microbiology, Washington, DC.
[9] Olson, E.R. and Chung, S.T. (1988) Transposon Tn4556 of Streptomyces fradiae: nucleotide sequence of the ends and the target sites. J. Bacteriol. 170, 1955^1957. [10] Snapper, S.B., Melton, R.E., Kieser, T., Mustafa, S. and Jacobs Jr., W.R. (1990) Isolation and characterization of e¤cient plasmid transformation mutants of Mycobacterium smegmatis. Mol. Microbiol. 4, 1911^1919. [11] Kieser, T., Bibb, M.J., Buttner, M.J., Chater, K.F. and Hopwood, D.A. (2000) Practical Streptomyces Genetics. The John Innes Foundation, Norwich. [12] Birch, A.W. and Cullum, J. (1985) Temperature-sensitive mutants of the Streptomyces plasmid pIJ702. J. Gen. Microbiol. 131, 1299^1303. [13] Bibb, M.J., Ward, J.M. and Cohen, S.N. (1985) Nucleotide sequences encoding and promoting expression of three antibiotic resistance genes indigenous to Streptomyces. Mol. Gen. Genet. 199, 26^36. [14] Snapper, S.B., Lugosi, L., Jekkel, A., Melton, R.E., Kieser, T., Bloom, B.R. and Jacobs Jr., W.R. (1988) Lysogeny and transformation in mycobacteria : stable expression of foreign genes. Proc. Natl. Acad. Sci. USA 85, 6987^6991. [15] Prod'hom, G., Lagier, B., Pelicic, V., Hance, A.J., Gicquel, B. and Guilhot, C. (1998) A reliable ampli¢cation technique for the characterization of genomic DNA sequences £anking insertion sequences. FEMS Microbiol. Lett. 158, 75^81. [16] Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) Basic local alignment search tool. J. Mol. Biol. 215, 403^410. [17] Bibb, M.J., Findlay, P.R. and Johnson, M.W. (1984) The relationship between base composition and codon usage in bacterial genes and its use for the simple and reliable identi¢cation of protein-coding sequences. Gene 30, 157^166. [18] Ikeda, H., Takada, Y., Chang-Hong, P., Tanaka, H. and Omura, S. (1993) Transposon mutagenesis by Tn4560 and applications with avermectin-producing Streptomyces avermitilis. J. Bacteriol. 175, 2077^2082. [19] Yamada, T., Masuda, K., Shoji, K. and Hori, M. (1972) Analysis of ribosomes from viomycin sensitive and resistant strains of Mycobacterium smegmatis. J. Bacteriol. 112, 1^6. [20] Bardarov, S., Kriakov, J., Carriere, C., Yu, S., Vaamonde, C., McAdam, R.A., Bloom, B.R., Hatfull, G.F. and Jacobs Jr., W.R. (1997) Conditionally replicating mycobacteriophages : a system for transposon delivery to Mycobacterium tuberculosis. Proc. Natl. Acad. Sci. USA 94, 10961^10966.
FEMSLE 10292 21-1-02