Identification and Characterization of IS1411, a New Insertion Sequence Which Causes Transcriptional Activation of the Phenol Degradation Genes in Pseudomonas putida

Abstract

A new insertion sequence (IS element), IS1411, was identified downstream of the phenol degradation genes pheBA that originated from plasmid DNA of Pseudomonas sp. strain EST1001. According to sequence analysis, IS1411 belongs to a new family of IS elements that has recently been named the ISL3 family (J. Mahillon and M. Chandler, Microbiol. Mol. Biol. Rev. 62:725–774, 1998). IS1411 generates 8-bp duplication of the target DNA and carries 24-bp inverted repeats (IRs), highly homologous to the IRs of other IS elements belonging to this family. IS1411 was discovered as a result of insertional activation of promoterless pheBA genes in Pseudomonas putida due to the presence of outward-directed promoters at the left end of IS1411. Both promoters located on the IS element have sequences that are similar to the consensus sequence of Escherichia coli ς⁷⁰. IS1411 can produce IS circles, and the circle formation is enhanced when two copies of the element are present in the same plasmid.

Insertion sequences (IS elements) are the simplest transposable DNA elements and generally encode one protein required for transposition. They range in length from 800 to 2,500 bp and can be found in the genomes of a wide range of bacteria (reviewed in references 11 and 25). IS elements were originally identified as the causative agents of highly polar mutations because they reduced expression of genes downstream of the insertion points (26, 34). A number of IS elements, however, have been implicated in the transcriptional activation of silent genes (2, 6, 7, 10, 12, 23, 28, 32, 40, 47 [see also the references in reference 25]). Transposable DNA elements can move through bacterial populations horizontally, via transmission of genetic material from one bacterium to another, and play an important role in the dissemination and acquisition of accessory genes involved in antibiotic resistance, virulence, pathogenicity, and catabolic pathways (reviewed in references 11 and 25). Soil bacteria, e.g., Pseudomonas spp., are known to metabolize a broad range of aromatic compounds and are therefore ideal agents for environmental detoxification (15). Molecular characterization of plasmids carrying the genes for catabolism of aromatics has revealed the modular structure of these plasmids: the catabolic genes are usually parts of composite transposons or they are found to be flanked by genes having similarity to transposase genes of IS elements (41). This indicates that IS elements could play an important role in the evolution of catabolic pathways in soil bacteria and in the regulation of gene expression. However, little is known about the mechanism of transposition of these DNA elements in soil bacteria.

We have previously shown that introduction of a plasmid carrying the pheBA genes encoding catechol 1,2-dioxygenase and phenol monooxygenase, respectively, into Pseudomonas putida PaW85 enables the bacterium to use the hybrid plasmid-chromosome-encoded pathway for phenol degradation (18). Here we characterize a novel IS element—IS1411 of Pseudomonas sp.—which is located downstream of the pheBA operon (Fig. 1) and has the potential to activate these genes due to outward-directed promoters on its left end. Sequence analysis of IS1411 has revealed that this DNA element belongs to the ISL3 family of IS elements (25). IS1411 produces IS circles, and as a result of transposition of IS1411 upstream to the promoterless pheBA operon, two copies of the element are present in the same plasmid. The possible mechanisms of transposition of this element will be discussed.

FIG. 1.

Organization of the pheBA operon in plasmid pAT1140 (18). The pheB and pheA genes are flanked by two IS elements, IS1472 and IS1411 (GenBank accession no. M57500). The black boxes show the locations of the pheBA genes and the transposase genes (tnpA) of IS1472 and IS1411. The open boxes represent the intergenic regions. The promoter of the pheBA operon (designated p_i) is located upstream of IS1472. The arrow indicates the direction of transcription of the genes. The right-end sequences of the transposon Tn4652 (42, 43) are shown by shaded boxes. IRR indicates the 46-bp terminal IR of the right end of Tn4652. (B) Organization of the pheBA operon and IS1411 in plasmids pEST1414 (19) and pINS113 (present study). The promoterless pheBA operon in pEST1414 is present, starting from the ClaI site. Only restriction sites relevant to the experiments presented in this paper are shown. C, ClaI; H, HindIII; K, KpnI. The left and right IRs of IS1411 are designated IRL and IRR, respectively. The arrow below the map of pINS113 indicates the direction of transcription of the pheBA genes from outward-directed promoters at the left end of the inserted IS1411.

MATERIALS AND METHODS

Bacterial strains, plasmids, and media.

The bacterial strains and plasmids used in this study are described in Table 1. The DNA fragments containing the left end of IS1411 were initially subcloned into the pBluescript SK(+) vector to obtain appropriate cloning sites for construction of pL1411END and pKTtnpA. To construct pL1411END, the 454-bp HindIII-RsaI fragment (containing the left end of IS1411 from the RsaI site and the sequence that flanked the left end in pINS113 up to the HindIII site in the pheB gene) was cloned into pBluescript SK. Oligonucleotide 113 (5′-AAGGGTGTAGAAAAAAT-3′), complementary to nucleotides (nt) 14 to 31 relative to the left end of IS1411 (Fig. 2), and a reverse primer complementary to a pBluescript SK sequence were used to amplify the 246-bp fragment containing the outward-oriented promoters of IS1411. The PCR-generated fragment was cut with BamHI and cloned into pKTlacZ cleaved with BamHI and SmaI. To construct pKTtnpA, the 665-bp HindIII fragment (Fig. 1B), cloned initially from pEST1414 into pBluescript, was inserted with BamHI and XhoI ends into pKTlacZ. Bacteria were grown on Luria-Bertani medium (27). Antibiotics were added at the following final concentrations: for Escherichia coli, ampicillin at 100 μg/ml; for P. putida, carbenicillin at 1,500 μg/ml. E. coli was incubated at 37°C (for enzyme assays, at 30°C), and P. putida was incubated at 30°C. Early-stationary-phase cultures were used for enzyme assays. E. coli was transformed with plasmid DNA as described by Hanahan (14). P. putida was electrotransformed by using the protocol of Sharma and Schimke (35).

TABLE 1.

Bacterial strains and plasmids used

Strain or plasmid	Genotype or construction	Source or reference
E. coli HB101	subE44 subF58 hsdS3 (rB− mB−) recA13 ara-14 proA2 lacY1 galK2 rpsL20 xyl-5 mtl-1	4
P. putida PaW85	Tn4652	1
Plasmids
pBluescript SK(+)	Cloning vector (Apr)	Stratagene
pEST1414	Plasmid pAYC32 carrying promoterless pheBA operon	19
pINS113	pEST1414 carrying insertion of IS1411 upstream of the pheBA genes	This work
pKTlacZ	Promoter probe vector containing lacZ gene in pKT240	16
pL1411END	PCR-generated 246-bp IS1411 left-end fragment cloned into pKTlacZ by left terminus towards lacZ	This work
PKTtnpA	665-bp HindIII fragment from pEST1414 cloned into pKTlacZ by left terminus of IS1411 in opposite direction to lacZ	This work

FIG. 2.

Nucleotide sequence of the left end of IS1411. The 8-bp target sequence that was duplicated during transposition of IS1411 upstream of the pheBA genes is underlined with a bold line. The 24-bp IR of the element is in boldface italics. The translation start sites of pheB and tnpA of IS1411 are marked by double lines. The outward-directed promoters of IS1411 are outlined with solid lines (−10 hexamers) and dashed lines (−35 hexamers). The transcription start sites for these promoters are indicated by arrows at the coding strand of pheB. The putative −10 and −35 hexamers of the tnpA promoter are shown by solid and dashed lines, respectively. Three 5′ ends of the tnpA mRNA, mapped by reverse transcriptase, are indicated by bent arrows at the coding strand of the tnpA gene. The location of oligonucleotide (oligo) 113, used for construction of plasmid pL1411END, is indicated by the dotted arrow.

DNA sequencing and mRNA mapping.

DNA sequencing was performed with a Sequenase version 2.0 DNA-sequencing kit (Amersham). A reverse transcriptase reaction was carried out to identify the 5′ ends of mRNA initiated from the outward-directed promoters and from the promoter of tnpA of IS1411. Total RNA (10 μg), purified from P. putida PaW85 and E. coli HB101 as described by Blomberg et al. (3), was used as the template in primer extension reactions. Primer 113 (described above) was used to map the outward-directed promoters. To map the transcription initiation from the tnpA promoter, the primer ORF2 (5′-CGAGGTTATTCAGTT-3′), complementary to nt 47 to 61 relative to the start codon GTG of the tnpA gene, was used in the primer extension analysis. Dideoxy sequencing standards of the double-stranded template (4 μg) were prepared by using the same primers.

Enzyme assays.

The β-galactosidase (β-Gal) assay was performed as described by Miller (27). Protein concentration in cell lysates was measured by the Bradford method (5).

Detection of IS1411 circles.

Plasmid DNA was isolated by the alkaline lysis protocol (31). Southern blot analysis of DNA preparations of pINS113 and pEST1414 was carried out as described previously (31). One microgram of DNA was loaded onto an agarose gel. The radioactive DNA probe used in filter hybridization was an [α-³²P]dCTP-labeled 630-bp HindIII-Eco47III restriction fragment derived from the IS1411 circle DNA containing both ends of IS1411. The hybridization signals at IS circles were quantitated by PhosphorImager (ImageQuant 4.2a software; Molecular Dynamics).

RESULTS

Insertional activation of the pheBA genes.

We have previously shown that when plasmid pEST1414 carrying the promoterless phenol degradation genes pheBA was introduced into P. putida PaW85 and bacteria were selected for growth on phenol minimal plates, promoters for the transcription of these genes were created as a result of base substitutions, deletions, and the transposition of transposon Tn4652 (19). Additionally, in one case (plasmid pINS113) activation of the pheBA genes was observed as a consequence of the insertion of a 1.4-kb-long DNA segment upstream of these genes. The nucleotide sequence of the inserted DNA revealed that a 1,419-bp element, bounded by 24-bp inverted repeats (IRs), had been inserted 35 nt upstream from the ATG start codon of the pheB gene. During this insertion, the 8-nt-long target sequence GGAATACA had been duplicated. The element was designated IS1411. The nucleotide sequence of the left end of IS1411 is shown in Fig. 2.

IS1411 originates downstream from the pheBA operon.

The nucleotide sequence of the inserted element IS1411 in plasmid pINS113 was identical to the sequence that was located downstream from the pheBA operon in the plasmid pEST1414 (Fig. 1). Interestingly, in its original location, the left IR of IS1411 overlapped the pheA gene by 21 nt and the element lacked direct repeats of the target DNA. Both copies of IS1411 were present in the plasmid pINS113. Because we observed no copies of IS1411 in the chromosome of P. putida PaW85 (data not shown), we suppose that the IS element originated from the same plasmid, pEST1414 (Fig. 1).

Besides pINS113, we have described another plasmid carrying two copies of IS1411 (data not shown). In that case (plasmid pM13) the insertion of IS1411 (which also generated 8-bp direct repeats of the target DNA) led to the inactivation of the pheBA genes due to transposition into the constitutively expressed operon.

Comparison of the sequence of IS1411 with the sequences of other IS elements.

IS1411 contains a 433-amino-acid-encoding open reading frame, designated ORFA, from nt 97 to 1405. The sequence of ORFA was compared with other sequences present in the GenBank and EMBL databases by using the TFASTA and FASTA programs. The predicted amino acid sequence of ORFA exhibited 60% identity with that of the transposase of IS1096 from Mycobacterium smegmatis and approximately 40% identity with those of the transposases of IS31831 from Corynebacterium glutamicum and IS13869 from Brevibacterium lactofermentum. This analysis clearly indicated that ORFA of IS1411 encodes transposase, and the gene was designated tnpA. TnpA of IS1411 also displayed motifs similar to the transposase sequences of several other IS elements, i.e., IS204, IS1476, IS1181, IS1165, IS1167, IS1396, IS1001, IS1193, ISL3, ISAE1, and IST1. The identity of the amino acid sequence of IS1411 TnpA with the sequences of the TnpAs of these IS elements ranged from 20 to 24%. The phylogenetic tree of all of the TnpA sequences was constructed via the CBRG (Computational Biochemistry Research Group) server (http://cbrg.inf.ethz.ch/) by using the Darwin program. The tree-fitting index (1.16) when all of the sequences were compared was too high (above 1.0). Therefore, two TnpA sequences, those of ISAE1 and IST1, which exhibited the lowest level of similarity to the other TnpAs compared, were eliminated in the course of tree construction. The unrooted phylogenetic tree constructed without these sequences, shown in Fig. 3, had a tree-fitting index of 0.96. The tree demonstrated that the TnpAs of IS1411, IS1096, IS31831, and IS13869 were more closely related to each other than to the rest of the TnpAs.

FIG. 3.

Unrooted phylogenetic tree of the transposases of IS1411 and its related elements. Multiple alignment of transposase sequences and construction of the phylogenetic tree were carried out via the CBRG server as described in the text. PAM distances are indicated at the branches of the tree. The tree-fitting index is 0.96. DNA sequence accession numbers and hosts (in parentheses) are as follows: IS1411 (Pseudomonas sp.), M57500; IS1096 (M. smegmatis), M76495; IS31831 (C. glutamicum), D17429; IS13869 (B. lactofermentum), Z66534; IS1396 (Serratia marcescens), U13612; IS1181 (Staphylococcus aureus), L14544; IS1193 (Streptococcus thermophilus), Y13713; IS1167 (Streptococcus pneumoniae), M36180; IS1476 (Enterococcus faecium), U63997; IS1165 (Leuconostoc mesenteroides), X62617; ISL3 (Lactobacillus delbrueckii), X79114; IS204 (Nocardia asteroides), U10634; IS1001 (Bordetella parapertussis), X66858.

Multiple sequence alignment of the TnpAs of IS1411, IS1096, IS31831, and IS13869 revealed strongly conserved amino acids over the entire protein (Fig. 4). The central region of the protein (the sequence from amino acids 209 to 273) contained seven residues (Fig. 4) that were conserved in all 13 TnpA sequences used for construction of the phylogenetic tree shown in Fig. 3. The IRs of the four IS elements compared in Fig. 4 also had a high degree of identity (Fig. 5). Interestingly, although the amino acid sequence of the TnpA of IS204 exhibited similarity only in the central region of the protein, the IRs of IS204 had a remarkable degree of homology to the sequences of the IRs of IS1411 (Fig. 5). Moreover, the terminal 14 nt of the IRs of IS204 were identical to the terminal sequences of the 24-bp IRs of IS1096.

FIG. 4.

Alignment of the deduced amino acid sequence of the transposase of IS1411 with transposases of IS1096, IS13869, and IS31831. Gaps introduced to optimize the alignment are shown by lines. Identical amino acids are marked by asterisks, and similar amino acids are indicated by dots. The alignments were generated via the CBRG server by using the Darwin program. Identical amino acids that were conserved in all 13 transposases analyzed in Fig. 3 are shown by shaded boxes.

FIG. 5.

Sequence alignment of IRs of IS1411, IS1096, IS204, IS13869, and IS31831. The asterisks indicate the nucleotides conserved in all of the IS elements compared. IRR and IRL, right and left IRs, respectively.

IS1411 activates transcription of the pheBA operon by outward-directed promoters at the left terminus of the element.

Many transposons and IS elements carry promoters that can activate transcription of flanking genes (reviewed in references 11 and 25). Therefore, we mapped the transcriptional start site of the pheBA operon in the plasmid pINS113, constitutively expressing the pheBA genes and carrying an insertion of IS1411 upstream of these genes, by using primer extension analysis. The 5′ ends of the mRNA were localized at C and G nucleotides, 51 and 105 nt inside the left end of IS1411, respectively (Fig. 2 and 6). The upper band was located just 8 nt downstream from the putative promoter sequence that resembled the ς⁷⁰-type promoter consensus TTGACAN_16–18TATAAT. The sequences TGGAAA, similar to the −35 hexamer, and TAAGAT, similar to the −10 hexamer, of this promoter were separated by the 18-nt-long spacer sequence. The proximal putative transcription start point was located 7 nt downstream from the sequence TAAGAT and was separated by 17 bp from the sequence TTGGTG, which resembles the −35 hexamer (Fig. 6).

FIG. 6.

Mapping of transcription initiation from outward-directed promoters of IS1411 by reverse transcriptase. Lanes G, A, T, and C show DNA-sequencing reactions of the left end of IS1411. Lanes 1 to 3 represent primer extension reactions carried out with total RNA isolated from the following bacteria: lane 1, P. putida PaW85 carrying pINS113; lane 2, E. coli HB101 carrying pINS113; lane 3, P. putida PaW85 carrying pEST1414 (negative control). The primer extension products are indicated on the right by arrows. The interrupted sequence of the left end of IS1411, including the −10 sequences of the promoters (boxed) and the transcription start points (indicated by asterisks), are shown on the left.

The presence of the outward-directed promoters at the left end of IS1411 was verified by subcloning the DNA fragment containing the left-end sequence of IS1411 to the promoter-probe-plasmid pKTlacZ upstream from the promoterless lacZ gene (construct pL1411END). A high level of β-Gal expression, about 400 times higher than that of pKTlacZ, was observed in the cells of P. putida PaW85 carrying pL1411END (not shown).

Expression of the promoter of the IS1411 transposase gene.

We failed to map the transcription start site of IS1411 tnpA both in the plasmid pEST1414 and in plasmids where the genes pheB and pheA upstream from the element were actively transcribed (not shown). Because the potential promoter region of the tnpA gene can be located only within a narrow 100-bp DNA region in the left end of the element (the putative translation initiator codon GTG of the tnpA gene is located 105 nt inside the left end of the IS [Fig. 2]), it is possible that the IS element-encoded transposase could repress transcription from its own promoter due to the overlap of the transposase binding site and the tnpA promoter. To avoid that possibility, we cloned the DNA fragment containing the potential promoter sequence upstream of the lacZ gene into the plasmid pKTlacZ. Expression of the resulting plasmid, pKTtnpA, was investigated in the cells of both E. coli and P. putida (Fig. 7), and a modest level of expression of β-Gal was observed in the cells carrying pKTtnpA. In comparison with results obtained in the E. coli background the level of expression of β-Gal detected in the cells of P. putida was approximately three times lower.

FIG. 7.

β-Gal activity measured in E. coli HB101 and P. putida PaW85 carrying either the promoter-probe-vector pKTlacZ or pKTtnpA containing the tnpA promoter region. The data (means ± standard deviations) from at least four independent experiments are presented.

To map the transcriptional start site of the lacZ transcription fusion in the cells carrying pKTtnpA, primer extension analysis of RNA was performed by using avian myeloblastosis virus reverse transcriptase. Three specific bands at positions −31, −29, and −11 from the tnpA translation initiation codon were revealed (these sites are indicated in Fig. 2). Due to the weak activity of the putative tnpA promoter, we did not expect to find promoter sequences exhibiting similarity to the well-known promoter consensus. The only candidate exhibiting similarity to the −10 consensus sequence for ς⁷⁰ was the sequence GAAAAT located upstream from the primer extension products identified at nt −31 and −29 from the tnpA gene (Fig. 2). The sequence TGGAAA was found 17 bp upstream of this hexamer.

IS1411 can form circular DNA molecules.

If DNA was isolated from E. coli cells by an alkaline lysis procedure and electrophoresed in an agarose gel, a small weak band accompanied the band of plasmid pINS113. Analysis of HindIII digestion of the pINS113 preparation revealed that this band moved in the gel as a 1.4-kb DNA fragment and that it was weaker than the other DNA fragments derived from the HindIII digestion of pINS113. Formation of IS minicircles has been shown for several IS elements (24, 29, 33, 37, 44, 46). To study whether the 1.4-kb HindIII DNA fragment represents the linearized IS1411 circle, we cloned this fragment into pBluescript SK and sequenced the insert. The DNA-sequencing data clearly demonstrated that the 1.4-kb HindIII fragment was derived from the IS1411 circle. The IS1411 circle was composed of the complete IS1411, with 5 bp separating the two IS ends. The 5-bp sequence AAACC was derived from the 3′ end of the pheA gene just flanking the left IR of IS1411 at the original location of this element in the plasmid pINS113.

No visible band corresponding to the IS circle was observed in the agarose gel either by gel electrophoresis of the plasmid pEST1414 isolated from E. coli carrying the single copy of IS1411 or by gel electrophoresis of pINS113 or pEST1414 isolated from P. putida (not shown). In order to reveal whether circularization of IS1411 requires more than one copy of the element in the plasmid and whether IS1411 could also circularize in the cells of P. putida, we performed Southern blot analysis of pINS113 and pEST1414 preparations from P. putida PaW85 and E. coli HB101. Both the uncut DNA and DNA digested with KpnI were gel electrophoresed and transferred onto a nitrocellulose filter. The filter was hybridized with the radioactive probe derived from the IS1411 circle (see Materials and Methods). The results of the hybridization are shown in Fig. 8. In the case of undigested DNA prepared from E. coli, the strong hybridization signal at the location of the IS1411 circle was detected if pINS113 was isolated (Fig. 8, lane 3). This signal was weaker in the case of pEST1414 (Fig. 8, lane 4). The strength of the hybridization signal in the pEST1414 preparation was 4% of that of the pINS113 preparation as quantitated by using the PhosphorImager. When DNA was isolated from P. putida, the IS1411 circle became detectable in the pINS113 preparation only after overexposure of the hybridized filter (Fig. 8, lane 2).

FIG. 8.

IS1411 forms circular DNA molecules. An autoradiograph of the Southern blot of the preparations of pINS113 and pEST1414 is shown. The radioactive DNA probe was prepared from IS1411. Marker sizes (lane 9) are indicated on the right of the autoradiograph. Lanes: 1 and 5, pEST1414 prepared from P. putida; 2 and 6, pINS113 prepared from P. putida; 3 and 7, pINS113 prepared from E. coli; 4 and 8, pEST1414 prepared from E. coli. Lanes 1 to 4 contain uncut DNA, and lanes 5 to 8 contain DNA cut with KpnI. The additional weak bands on lane 3 are of unknown origin and are not discussed in this report.

Plasmid pEST1414 contains one KpnI site within the tnpA gene of IS1411 (Fig. 1B). Thus, the KpnI digestion of the pEST1414 preparation opens the plasmid molecule and 1.4-kb IS circle. pINS113, carrying two IS elements, is cut into two fragments: an 11.5-kb fragment containing mostly the vector DNA and a 5.4-kb fragment covering sequences of the pheBA operon flanked at both sides by IS1411 sequences up to the KpnI sites (Fig. 1B). According to hybridization data, the 1.4-kb restriction fragment derived from the IS1411 circle hybridized with the radioactive probe was visible in the case of pINS113 and pEST1414 prepared from E. coli (Fig. 8, lanes 7 and 8) and when pINS113 was prepared from P. putida (Fig. 8, lane 6). Again, the hybridization signal was strongest in analysis of pINS113 isolated from E. coli (Fig. 8, lane 7). If DNA was isolated from P. putida, the radioactive band corresponding to the IS circle after gel electrophoresis of the pINS113 preparation became visible after prolonged exposure of the filter (Fig. 8, lane 6), and in addition, an almost undetectable band also appeared in lane 5, containing the pEST1414 preparation.

DISCUSSION

Here we report on a novel IS element, IS1411, that was discovered by activation of the transcription of the phenol degradation genes pheBA in P. putida. We have used the plasmid pEST1414 carrying the promoterless pheBA operon as a reporter plasmid to study mutation processes in starving cells of P. putida (19). In contrast to the other promoter-creating genetic events (i.e., base substitutions, deletions, and generation of fusion promoters by Tn4652) we detected, the insertional activation of the pheBA genes by IS1411 was rare. It has only been discovered once. The IS1411 that activated the pheBA genes originated from the same plasmid, pEST1414, downstream from the activated genes (Fig. 1). The infrequency of the IS1411 transposition event could be explained by a phenomenon called transposition immunity, i.e., the mobile DNA elements transpose much less frequently into a plasmid replicon that already contains a copy of the element than into a replicon lacking the element (reviewed in references 22 and 25). In addition, the level of expression of the tnpA gene of IS1411 in P. putida was very low (Fig. 7).

Vertes et al. (45) suggested that IS31831 and IS1096 belong to a new family of IS elements. The homology search for IS1411 revealed, in addition to IS31831 and IS1096, several IS elements that encode transposases exhibiting similarities to the deduced amino acid sequence of the TnpA of IS1411 (Fig. 3 and 4). The remarkable degree of similarity of the transposase sequences of IS elements analyzed in Fig. 3 indicates that these IS elements belong to the same family. According to the grouping of 443 IS elements in 17 families (25), this is the ISL3 family. The DNA elements having higher degrees of similarity to IS1411 tnpA are distributed among soil bacteria (Fig. 4). These DNA elements also contain similar IRs (Fig. 5) and might constitute a distinct subgroup within the ISL3 family.

The DNA segment containing the pheBA operon and IS1411 has been cloned from plasmid DNA of Pseudomonas sp. strain EST1001, which is a derivative of Pseudomonas sp. strain S13 (20, 21). Strain S13 was isolated in 1976 by P. A. Williams and was not, therefore, of local origin. The tnpA gene identical to the tnpA of IS1411 has been identified downstream of the chlorobiphenyl degradation genes bph in hybrid strain JHR22 of Burkholderia capacia (37a). Traces of IS1411 have also been found in the 4-nitrotoluene-degrading Pseudomonas sp. strain TW3: the sequence of xylB homologue ntnB* was interrupted by a piece of DNA identical to the left end of IS1411 (17). However, we disagree with the authors’ interpretation (17) that the disrupting sequence was derived from the Tn4652-encoded transposase.

Bacterial transposable elements utilize two major modes of transposition. In nonreplicative transposition (the cut-and-paste mechanism), transposase cuts both DNA strands at the element’s two ends, and the element is transferred to the target (e.g., Tn10 and Tn7 [reviewed in reference 13]). The replicative transposition involves cointegrate formation, in which the donor and target sequences are fused, the element is duplicated, and then the cointegrate is resolved by recombination between the two copies of the element. This restores the initial donor molecule with one copy of the element and restores the target molecule with the second copy (e.g., Tn3 family transposons and Mu phage [reviewed in reference 35]). In some cases the resolution step is carried out by element-encoded resolvase. There is no published data about mechanisms of transposition of IS elements similar to IS1411. For IS1096, it has been shown that this element encodes a putative resolvase (8). However, this is the only indication that IS1096-like elements could transpose replicatively. The other related IS elements are smaller, and they have not been shown to contain sequences encoding putative resolvases. The plasmids pINS113 and pM13 with insertions of IS1411 that we isolated also retained the original copy of the element. Thus, one should discuss the fact that during insertion IS1411 had duplicated in these plasmids, which indicates that the mechanism of transposition of IS1411 might be replicative. However, the fact that the plasmids pINS113 and pM13 carry two copies of IS1411 could just as well be the result of a conservative event in which the IS copy moved from one plasmid to a sibling plasmid.

One of the interesting features of IS1411 is its formation of IS circles (Fig. 8). Circularization of the transposable element occurs during the transfer of conjugative transposons (30). The circular forms have also been observed among members of the IS3 family (e.g., IS3 itself [33], IS911 [29], IS2 [24], and IS150 [46]) and also for other DNA elements (e.g., IS1 [44], IS117 [37], and Tn4451 [9]). It has been supposed that circle formation could be an intermediate step of transposition (9, 37, 44). An unconventional pathway of transposition, i.e., transposition through a circular intermediate, has been experimentally confirmed for IS2 and IS911 (24, 38, 39). The formation of a figure eight molecule (in which only one of the IS strands has undergone cleavage and transfer to the opposite end, resulting in circularization of a single strand) as a precursor to the circle has been observed for these elements (24, 38). There is also data indicating that a circular transposition intermediate could arise replicatively (37, 44). Therefore, it is tempting to speculate that IS1411, which belongs to the distinct ISL3 family, can follow (at least in some cases) a transposition pathway that utilizes an IS circle as an intermediate.

The frequency of circularization of IS1411 was higher in E. coli than in P. putida (Fig. 8). The fact that formation of the IS1411 circle was enhanced in the E. coli background indicates that the expression of the IS1411 transposase may be downregulated in P. putida. Our attempts to map the IS1411 tnpA promoter revealed that the level of expression of the reporter gene lacZ under the tnpA promoter was also higher in E. coli than in P. putida (Fig. 7). The basis for these differences is at present unclear. It is possible that transcription of the tnpA of IS1411 is more tightly controlled in P. putida than in heterologous hosts. In comparison with plasmid pEST1414 carrying a single copy of IS1411, the frequency of IS1411 circle formation was higher in the case of plasmid pINS113 carrying two copies of the element (Fig. 8). The IS circles can also be easily detected in analysis of the pM13 preparation (not shown). Thus, it is possible that a difference in circularization frequency can be a simple gene dosage effect. The fact that elevated levels of transposase stimulated circle formation has been demonstrated for IS2 (24) and for IS911 (38).

Transposable elements isolated from different soil bacteria can be of importance in regulating gene expression due to silencing or activating certain genes. They take part in genomic rearrangements and can be involved in the evolution of new catabolic operons. So far, little is known about the mechanisms and regulation of the transposition of DNA elements in soil bacteria. As discussed above, the IS element IS1411 is distributed in different aromatic-compound-degrading bacteria. This DNA element is also capable of driving expression of promoterless genes due to the presence of the outward-directed promoters in the left end. Therefore, the study of the transposition processes of IS1411 would extend our awareness of genetic processes in soil bacteria.