Previous Article | Next Article ![]()
Journal of Bacteriology, June 2004, p. 3656-3659, Vol. 186, No. 11
0021-9193/04/$08.00+0 DOI: 10.1128/JB.186.11.3656-3659.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Romain Brunel, Harry J. Flint, and Karen P. Scott*
Rowett Research Institute, Bucksburn, Aberdeen AB21 9SB, United Kingdom
Received 11 November 2003/ Accepted 13 February 2004
|
|
|---|
|
|
|---|
Copies of tet(W) found in environmentally and phylogenetically distinct bacterial isolates (2, 20) have remarkable sequence conservation, which argues for the extensive rapid transfer of this gene in nature. Previous work showed that tet(W) could be transferred at high frequencies (102 to 105) in vitro between strains of the rumen anaerobe Butyrivibrio fibrisolvens (19) and that resistance transfer was accompanied by acquisition of a large chromosomal fragment of 40 to 50 kb (19). The mobile chromosomal DNA integrated at a preferred site in the recipient genome, and Southern blots of pulsed-field gels showed that the new band hybridized to a tet(W) probe (2). The object of the present study was to investigate the organization and distribution of the putative element involved in the transfer of tet(W) between B. fibrisolvens strains, TnB1230.
TnB1230 has sequence similarity to Tn1549. Genomic DNA was purified from B. fibrisolvens 1.230 grown overnight in anaerobic M2GSC medium (14) containing 10 µg of tetracycline/ml. Tetracycline-resistant clones (resistant to a 5-µg ml1 concentration) containing tet(W) were identified in Lambda ZAPII and SuperCos 1 libraries (Stratagene Europe, Amsterdam, The Netherlands). Assembly of 12 kb of the possible 40- to 50-kb TnB1230 sequence revealed the presence of eight unidirectional open reading frames (ORFs) (Fig. 1). The tet(W) gene has a higher G+C content (53%) than the other genes (Fig. 1), due to the preference of tet(W) for a G or C as the third nucleotide base in the codon (2). B. fibrisolvens itself has a low DNA G+C content (36 to 41%) (5).
![]() View larger version (16K): [in a new window] |
FIG. 1. Diagram showing the genetic organization of 12 kb of the sequenced region of TnB1230. The arrows indicate the location and direction of transcription of each ORF. The positions of the DRs (DR1 and DR2) are indicated by hatched boxes. The number of amino acids (aa) encoded by each ORF is shown, and the DNA G+C content is also indicated. The sequence identity to ORFs encoded by Tn1549 is indicated as appropriate. This figure is not drawn to scale.
|
tet(W) is flanked by identical DNA sequences. Two identical, 707-bp DNA direct repeat (DR) sequences flank tet(W): DR1 finishes 657 bp upstream of the tet(W) start codon, and DR2 starts 35 bp after the tet(W) stop codon (Fig. 1). These sequences contain ORFs capable of encoding proteins of 98 amino acids. An alternative start codon is present upstream of DR1, outside the repeated DNA sequence, potentially encoding a longer 168-amino-acid protein. The ORF contained in DR2 is in the same frame as tet(W).
Although DR sequences flanking antibiotic resistance genes often correspond to insertion sequence (IS) elements (18), we did not obtain evidence to suggest that the DR sequences in TnB1230 were ISs. They have no sequence similarity to previously characterized IS elements, and they lack terminal inverted repeats and the DDE triad characteristic of the catalytic core of many IS transposases (9, 11), although there is a possible alternative DDD triad. On the other hand, database searches showed that the proteins encoded by the DRs had significant identity to bacterial nitroreductases (38% identity to the nitroreductase [Nrd] from Clostridium acetobutylicum [GenBank accession number AE007749]). The longer upstream ORF-reading into DR1 encodes an oxygen-insensitive type I nitroreductase (25), Nrd1, while DR2 encodes an identical nitroreductase (Nrd2) that is truncated at the N terminus (Fig. 1). Nitroreductase enzymes are involved in the reduction of nitrogen-containing aromatic compounds, often releasing toxic, carcinogenic, or mutagenic by-products. The antimicrobial agent metronidazole is activated following reduction by nitroreductase enzymes. All of the B. fibrisolvens strains (1.230, 2221, or 2221R transconjugants) were sensitive to metronidazole (in concentrations ranging from 20 to 80 µg ml1), indicating that they contain active Nrd enzymes.
Sequences upstream of tet(W) are required for full expression of resistance. The sequence immediately upstream from the tet(W) gene is highly conserved in six diverse, obligately anaerobic bacterial isolates harboring tet(W) (Table 1). In fact, only two isolates had any sequence divergence in the 340 bp upstream of the ATG codon: Clostridium sp. strain K10 had 18 random base pair changes, while Mitsuokella multiacidus P208 contained an inserted 23-bp DR (giving >93% sequence conservation) (Fig. 2). Nucleotides upstream of the tet(W) gene carried by the opportunistic livestock pathogen Arcanobacterium pyogenes (GenBank accession number AY049983) (4) are only 64% identical to those of the commensal anaerobes described here (data not shown).
|
View this table: [in a new window] |
TABLE 1. General characteristics of anaerobic bacteria used in this studya
|
![]() View larger version (75K): [in a new window] |
FIG. 2. The CLUSTALW program was used to align sequences upstream of tet(W) genes from B. fibrisolvens 1.230 (GenBank accession number AJ222769), M. multiacidus P208 (GenBank accession number AY603069), and Clostridium sp. strain K10 (GenBank accession number AY601650) with tet(M) from E. faecalis (GenBank accession number M85225) and tet(O) from Campylobacter jejuni (GenBank accession number M18896). Solid arrows indicate the inverted repeats predicted to form secondary stem-loop structures. The DRs present only in M. multiacidus P208 are indicated by dashed arrows above the sequence. The putative poly(U) attenuation terminator (TTTTT), ribosome binding sites (GGAXG), and the ATG start codon are boxed. Nucleotides conserved in all five sequences are underscored with asterisks. The positions of the upstream primers used to amplify full-length (PCRB1) or truncated [tet(W)FF] sequences for expression cloning (indicated by [1] and [2], respectively) are shown by dashed overlining. The sequences in tet(W) and tet(M) encoding the putative leader peptides are shown in bold.
|
G (at 25°C) of 23.2 kcal/mol (2, 22) (Fig. 2), culminating in a poly(U) sequence. This organization is characteristic of attenuation terminator sequences (http://www.bork.embl-heidelberg.de/Docu/attenuation/index.html). Similar inverted repeat sequences are located upstream of the ribosome protection genes tet(M), tet(O), tet(S), tet(T), and tet(32), and they are important for regulating gene expression in tet(O) (24) and tet(M) (21). The expression of tet(M) depends on the synthesis of a 28-amino-acid leader peptide encoded immediately upstream of the tet(M) ribosome binding site (21). A similar sequence potentially encoding a 14-amino-acid peptide (MLYMCMSATYNPQW) is present upstream of the tet(W) gene in the commensal anaerobes (Fig. 2). Each leader peptide contains a large proportion of rare amino acids; 50% of the amino acids in the tet(W)-encoded leader peptide correspond to the five rarest amino acids, which normally comprise <14% of the total residues in a protein (10). Sequences downstream of tet(W) diverge within 120 bp of the stop codon in the different isolates (data not shown). Escherichia coli cells transformed with pGEM-T vector (Promega) clones containing amplified fragments of the tet(W) gene plus either 27 or 340 bp of the upstream sequence (Fig. 2) were grown in concentrations of tetracycline ranging from 0 to 80 µg ml1. Shorter clones containing only the tet(W) gene grew poorly in concentrations of tetracycline exceeding 5 µg ml1 (optical density at 650 nm was 0.1 after 24 h of growth in a concentration of 15 µg ml1), whereas those containing the tet(W) gene and additional upstream sequences were able to grow well in the presence of up to 60 µg ml1 (optical density at 650 nm was 0.35 after 24 h of growth in a concentration of 60 µg ml1). This finding confirms that the region between 27 and 340 bp upstream of the tet(W) start codon is required for full expression of Tcr, at least in E. coli, presumably because it contains promoter sequences.
Possible evolution of TnB1230. Primer sets designed for use against specific TnB1230 ORFs flanking tet(W) failed to give products with any of the other commensal bacteria containing tet(W) (Table 1), indicating the presence of different genetic supports for tet(W) in other host bacteria. The A. pyogenes tet(W) gene is sometimes associated with a mob gene related to the tnpZ genes found on the clostridial mobilizable transposons (4). The A. pyogenes tet(W) gene transferred at very low frequencies (109 to 1011) (4) compared to those of TnB1230 (102 to 105) (19), presumably because the element relies on mobilization rather than self-conjugation.
The arrangement of TnB1230 ORFs encoding transfer proteins suggests that this element arose from a precursor lacking both nrd and tet(W) genes. The nrd gene may have integrated into this conjugative transposon, thereby interrupting, but not inactivating, an operon of genes involved in conjugative transfer. Transmissible, plasmid-borne nitroreductases from other groups of bacteria have been reported (16). This transposable element harboring a nitroreductase gene may have been the immediate precursor for TnB1230, with a small incoming element carrying tet(W) inserting into the nrd gene. The partial duplication of the nrd gene, found in TnB1230, may have occurred during the initial insertion event or subsequently by homologous recombination between an excised tet(W) fragment with nrd sequences downstream and an intact nrd present on another copy of the precursor transposon. This series of events explains why the sequences of the nitroreductase genes flanking tet(W) are identical and have a lower DNA G+C content than the surrounding ORFs and also why the DNA G+C content of the tet(W) gene is different from that of other TnB1230 or B. fibrisolvens genes. We noted previously that during conjugative mating experiments, B. fibrisolvens transconjugants occasionally gained more than one copy of TnB1230, illustrated by additional band shifts in the pulsed-field gel electrophoresis profile (19). Not all these bands hybridized to a tet(W) probe, and they may thus be the result of movement of shorter derivatives of TnB1230. It remains possible that TnB1230 is a complex, composite transposon carrying more than one set of transfer genes, but this possibility can be resolved only when the complete sequence of TnB1230 is available. The work reported here establishes for the first time the close linkage of a tet(W) gene with a cluster of transfer genes.
Nucleotide sequence accession number. The DNA sequence of TnB1230 discussed here has been deposited in the GenBank database under accession number AJ222769.
We acknowledge the financial support of SEERAD (Scottish Executive Environment and Rural Affairs Department) and FSA (Food Standards Agency). R.B. was a summer student from the I.U.T., Clermont-Ferrand, France, on placement at RRI.
Present address: Moredun Research Institute, Pentlands Science Park, Bush Loan, Midlothian EH26 0PZ, United Kingdom. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»