Transposase-induced excision and circularization of the bacterial insertion sequence IS911

The insertion sequence excision enhancer: A PrimPol-based primer invasion system for immobilizing transposon-transmitted antibiotic resistance genes

Evolutionary studies often identify genes that have been exchanged between different organisms and the phrase Lateral or Horizontal Gene Transfer is often used in this context. However, they rarely provide any mechanistic information concerning how these gene transfers might have occurred. With the astonishing increase in the number of sequences in public databases over the past two or three decades, identical antibiotic resistance genes have been identified in many different sequence contexts. One explanation for this would be that genes are initially transmitted by transposons which have subsequently decayed and can no longer be detected. Here, we provide an overview of a protein, IEE (Insertion Sequence Excision Enhancer) observed to facilitate high-frequency excision of IS629 from clinically important Escherichia coli O157:H7 and subsequently shown to affect a large class of bacterial insertion sequences which all transpose using the copy-out-paste-in transposition mechanism. Excision depends on both IEE and transposase indicating association with the transposition process itself. We review genetic and biochemical data and propose that IEE immobilizes genes carried by compound transposons by removing the flanking insertion sequence (IS) copies. The biochemical activities of IEE as a primase with the capacity to recognize DNA microhomologies and the observation that its effect appears restricted to IS families which use copy-out-paste-in transposition, suggests IS deletion occurs by abortive transposition involving strand switching (primer invasion) during the copy-out step. This reinforces the proposal made for understanding the widespread phenomenon loss of ISApl1 flanking mcr-1 in the compound transposon Tn6330 which we illustrate with a detailed model. This model also provides a convincing way to explain the high levels of IEE-induced precise IS excision.

Structures of ISCth4 transpososomes reveal the role of asymmetry in copy-out/paste-in DNA transposition

Copy-out/paste-in transposition is a major bacterial DNA mobility pathway. It contributes significantly to the emergence of antibiotic resistance, often by upregulating expression of downstream genes upon integration. Unlike other transposition pathways, it requires both asymmetric and symmetric strand transfer steps. Here, we report the first structural study of a copy-out/paste-in transposase and demonstrate its ability to catalyze all pathway steps in vitro. X-ray structures of ISCth4 transposase, a member of the IS256 family of insertion sequences, bound to DNA substrates corresponding to three sequential steps in the reaction reveal an unusual asymmetric dimeric transpososome. During transposition, an array of N-terminal domains binds a single transposon end while the catalytic domain moves to accommodate the varying substrates. These conformational changes control the path of DNA flanking the transposon end and the generation of DNA-binding sites. Our results explain the asymmetric outcome of the initial strand transfer and show how DNA binding is modulated by the asymmetric transposase to allow the capture of a second transposon end and to integrate a circular intermediate.

Detection and Characterization of Transposons in Bacteria

Bacterial transposons, through their ability to transfer DNA sequences from one position in the genome to another, play a central role in the shape and the evolution of genomes. Extensive studies have been performed during the last five decades to understand the molecular mechanisms involved in the transposition of a variety of elements. Among the methods used, the papillation and the mating out coupled to arbitrary primed PCR assays described in this chapter are widely used as very powerful approaches to detect and characterize transposition events in vivo.

Transposition of insertion sequence IS256Bsu1 in Bacillus subtilis 168 is strictly dependent on recA

We developed an insertion sequence transposition detection system called the "jumping cat assay" and applied it to the Bacillus subtilis chromosome using IS256Bsu1 derived from B. subtilis natto. The high frequency of transposition enabled us to explore host factors; combining the assay and genetic analyses revealed that recA is essential for the transposition of IS256Bsu1. Detailed analyses using various domain mutants of recA demonstrated that this essentiality is not related to the function of recA in homologous recombination. Instead, the ATP binding and hydrolysis function seemed to be crucial for IS transposition. To elucidate the role of recA, we focused on the muB gene of the enterobacteriophage Mu. Based on information from the NCBI Conserved Domain Database, both MuB and RecA belong to the P-loop dNTPase superfamily. Further experiments revealed that muB complements the transposition-defective phenotype of a recA deletant, although it could not rescue UV sensitivity. These results suggest that recA shares a common function with muB that helps the transposition of IS256Bsu1 in B. subtilis.

Copy-out-Paste-in Transposition of IS911: A Major Transposition Pathway

IS911 has provided a powerful model for studying the transposition of members of a large class of transposable element: the IS3 family of bacterial Insertion Sequences (IS). These transpose by a Copy-out-Paste-in mechanism in which a double-strand IS circle transposition intermediate is generated from the donor site by replication and proceeds to integrate into a suitable double strand DNA target. This is perhaps one of the most common transposition mechanisms known to date. Copy-out-Paste-in transposition has been adopted by members of at least eight large IS families. This chapter details the different steps of the Copy-out-Paste-in mechanism involved in IS911 transposition. At a more biological level it also describes various aspects of regulation of the transposition process. These include transposase production by programmed translational frameshifting, transposase expression from the circular intermediate using a specialized promoter assembled at the circle junction and binding of the nascent transposase while it remains attached to the ribosome during translation (co-translational binding). This co-translational binding of the transposase to neighboring IS ends provides an explanation for the longstanding observation that transposases show a cis-preference for their activities.

The diversity of prokaryotic DDE transposases of the mutator superfamily, insertion specificity, and association with conjugation machineries

Transposable elements (TEs) are major components of both prokaryotic and eukaryotic genomes and play a significant role in their evolution. In this study, we have identified new prokaryotic DDE transposase families related to the eukaryotic Mutator-like transposases. These genes were retrieved by cascade PSI-Blast using as initial query the transposase of the streptococcal integrative and conjugative element (ICE) TnGBS2. By combining secondary structure predictions and protein sequence alignments, we predicted the DDE catalytic triad and the DNA-binding domain recognizing the terminal inverted repeats. Furthermore, we systematically characterized the organization and the insertion specificity of the TEs relying on these prokaryotic Mutator-like transposases (p-MULT) for their mobility. Strikingly, two distant TE families target their integration upstream σ_A dependent promoters. This allowed us to identify a transposase sequence signature associated with this unique insertion specificity and to show that the dissymmetry between the two inverted repeats is responsible for the orientation of the insertion. Surprisingly, while DDE transposases are generally associated with small and simple transposons such as insertion sequences (ISs), p-MULT encoding TEs show an unprecedented diversity with several families of IS, transposons, and ICEs ranging in size from 1.1 to 52 kb.

Cotranslational control of DNA transposition: a window of opportunity

Transposable elements are important in genome dynamics and evolution. Bacterial insertion sequences (IS) constitute a major group in number and impact. Understanding their role in shaping genomes requires knowledge of how their transposition activity is regulated and interfaced with the host cell. One IS regulatory phenomenon is a preference of their transposases (Tpases) for action on the element from which they are expressed (cis) rather than on other copies of the same element (trans). Using IS911, we show in vivo that activity in cis was ~200 fold higher than in trans. We also demonstrate that a translational frameshifting pause signal influences cis preference presumably by facilitating sequential folding and cotranslational binding of the Tpase. In vitro, IS911 Tpase bound IS ends during translation but not after complete translation. Cotranslational binding of nascent Tpase permits tight control of IS proliferation providing a mechanistic explanation for cis regulation of transposition involving an unexpected partner, the ribosome.

Genomic Distribution and Divergence of Levansucrase-Coding Genes in Pseudomonas syringae

In the plant pathogenic bacterium, Pseudomonas syringae, the exopolysaccharide levan is synthesized by extracellular levansucrase (Lsc), which is encoded by two conserved 1,296-bp genes termed lscB and lscC in P. syringae strain PG4180. A third gene, lscA, is homologous to the 1,248-bp lsc gene of the bacterium Erwinia amylovora, causing fire blight. However, lscA is not expressed in P. syringae strain PG4180. Herein, PG4180 lscA was shown to be expressed from its native promoter in the Lsc-deficient E. amylovora mutant, Ea7/74-LS6, suggesting that lscA might be closely related to the E. amylovora lsc gene. Nucleotide sequence analysis revealed that lscB and lscC homologs in several P. syringae strains are part of a highly conserved 1.8-kb region containing the ORF, flanked by 450-452-bp and 49-51-bp up- and downstream sequences, respectively. Interestingly, the 450-452-bp upstream sequence, along with the initial 48-bp ORF sequence encoding for the N-terminal 16 amino acid residues of Lsc, were found to be highly similar to the respective sequence of a putatively prophage-borne glycosyl hydrolase-encoding gene in several P. syringae genomes. Minimal promoter regions of lscB and lscC were mapped in PG4180 by deletion analysis and were found to be located in similar positions upstream of lsc genes in three P. syringae genomes. Thus, a putative 498-500-bp promoter element was identified, which possesses the prophage-associated com gene and DNA encoding common N-terminal sequences of all 1,296-bp Lsc and two glycosyl hydrolases. Since the gene product of the non-expressed 1,248-bp lscA is lacking this conserved N-terminal region but is otherwise highly homologous to those of lscB and lscC, it was concluded that lscA might have been the ancestral lsc gene in E. amylovora and P. syringae. Our data indicated that its highly expressed paralogs in P. syringae are probably derived from subsequent recombination events initiated by insertion of the 498-500-bp promoter element, described herein, containing a translational start site.

The interplay of mRNA stimulatory signals required for AUU-mediated initiation and programmed -1 ribosomal frameshifting in decoding of transposable element IS911

The IS911 bacterial transposable element uses -1 programmed translational frameshifting to generate the protein required for its mobility: translation initiated in one gene (orfA) shifts to the -1 frame and continues in a second overlapping gene (orfB), thus generating the OrfAB transposase. The A-AAA-AAG frameshift site of IS911 is flanked by two stimulatory elements, an upstream Shine-Dalgarno sequence and a downstream stem-loop. We show here that, while they can act independently, these stimulators have a synergistic effect when combined. Mutagenic analyses revealed features of the complex stem-loop that make it a low-efficiency stimulator. They also revealed the dual role of the upstream Shine-Dalgarno sequence as (i) a stimulator of frameshifting, by itself more potent than the stem-loop, and (ii) a mandatory determinant of initiation of OrfB protein synthesis on an AUU codon directly preceding the A6G motif. Both roles rely on transient base pairing of the Shine-Dalgarno sequence with the 3' end of 16S rRNA. Because of its effect on frameshifting, the Shine-Dalgarno sequence is an important determinant of the level of transposase in IS911-containing cells, and hence of the frequency of transposition.

Formation of an inverted repeat junction in the transposition of insertion sequence ISLC3 isolated from Lactobacillus casei

An insertion sequence, ISLC3, of 1351 bp has been isolated from Lactobacillus casei. Formation of IS circles containing a 3 bp spacer (complete junction) or deletion of 25 bp at the left inverted repeat (IRL) between the abutted IS ends of the ISLC3 junction region (deleted junction) was also discovered in the lactobacilli and Escherichia coli system studied. We found that the promoter formed by the complete junction P(jun) was more active than that formed by the 25 bp deleted junction P(djun) or the indigenous promoter P(IRL). The corresponding transcription start sites for both promoter P(jun) and P(IRL) as well as P(djun) were subsequently determined using a primer extension assay. The activity of transposase OrfAB of ISLC3 was also assayed using an in vitro system. It was found that this transposase preferred to cleave a single DNA strand at the IRR over the IRL end in the transposition process, suggesting that attack of one end by the other was oriented from IRR to IRL.

Truncated forms of IS911 transposase downregulate transposition

IS911 naturally produces transposase (OrfAB) derivatives truncated at the C-terminal end (OrfAB-CTF) and devoid of the catalytic domain. A majority species, OrfAB*, was produced at higher levels at 42 degrees C than at 30 degrees C suggesting that it is at least partly responsible for the innate reduction in IS911 transposition activity at higher temperatures. An engineered equivalent of similar length, OrfAB[1-149], inhibited transposition activity in vivo or in vitro when produced along with full-length transposase. We isolated several point mutants showing higher activity than the wild-type IS911 at 42 degrees C. These fall into two regions of the transposase. One, located in the N-terminal segment of OrfAB, lies between or within two regions involved in protein multimerization. The other is located within the C-terminal catalytic domain. The N-terminal mutations resulted in reduced levels of OrfAB* while the C-terminal mutation alone appeared not to affect OrfAB* levels. Combination of N- and C-terminal mutations greatly reduced OrfAB* levels and transposition was concomitantly high even at 42 degrees C. The mechanism by which truncated transposase species are generated and how they intervene to reduce transposition activity is discussed. While transposition activity of these multiply mutated derivatives in vivo was resistant to temperature, the purified OrfAB derivatives retained an inherent temperature-sensitive phenotype in vitro. This clearly demonstrates that temperature sensitivity of IS911 transposition is a complex phenomenon with several mechanistic components. These results have important implications for the several other transposons and insertion sequences whose transposition has also been shown to be temperature-sensitive.

IS911 transpososome assembly as analysed by tethered particle motion

Initiation of transposition requires formation of a synaptic complex between both transposon ends and the transposase (Tpase), the enzyme which catalyses DNA cleavage and strand transfer and which ensures transposon mobility. We have used a single-molecule approach, tethered particle motion (TPM), to observe binding of a Tpase derivative, OrfAB[149], amputated for its C-terminal catalytic domain, to DNA molecules carrying one or two IS911 ends. Binding of OrfAB[149] to a single IS911 end provoked a small shortening of the DNA. This is consistent with a DNA bend introduced by protein binding to a single end. This was confirmed using a classic gel retardation assay with circularly permuted DNA substrates. When two ends were present on the tethered DNA in their natural, inverted, configuration, Tpase not only provoked the short reduction in length but also generated species with greatly reduce effective length consistent with DNA looping between the ends. Once formed, this 'looped' species was very stable. Kinetic analysis in real-time suggested that passage from the bound unlooped to the looped state could involve another species of intermediate length in which both transposon ends are bound. DNA carrying directly repeated ends also gave rise to the looped species but the level of the intermediate species was significantly enhanced. Its accumulation could reflect a less favourable synapse formation from this configuration than for the inverted ends. This is compatible with a model in which Tpase binds separately to and bends each end (the intermediate species) and protein-protein interactions then lead to synapsis (the looped species).

Genome comparison in silico in Neisseria suggests integration of filamentous bacteriophages by their own transposase

We have identified filamentous prophages, Nf (Neisserial filamentous phages), during an in silico genome comparison in Neisseria. Comparison of three genomes of Neisseria meningitidis and one of Neisseria gonorrhoeae revealed four subtypes of Nf. Eleven intact copies are located at different loci in the four genomes. Each intact copy of Nf is flanked by duplication of 5'-CT and, at its right end, carries a transposase homologue (pivNM/irg) of RNaseH/Retroviral integrase superfamily. The phylogeny of these putative transposases and that of phage-related proteins on Nfs are congruent. Following circularization of Nfs, a promoter-like sequence forms. The sequence at the junction of these predicted circular forms (5'-atCTtatat) was found in a related plasmid (pMU1) at a corresponding locus. Several structural variants of Nfs--partially inverted, internally deleted and truncated--were also identified. The partial inversion seems to be a product of site-specific recombination between two 5'-CTtat sequences that are in inverse orientation, one at its end and the other upstream of pivNM/irg. Formation of internally deleted variants probably proceeded through replicative transposition that also involved two 5'-CTtat sequences. We concluded that the PivNM/Irg transposase on Nfs integrated their circular forms into the chromosomal 5'-CT-containing sequences and probably mediated the above rearrangements.

Transposition of ISHp608, member of an unusual family of bacterial insertion sequences

ISHp608 from Helicobacter pylori is active in Escherichia coli and represents a recently recognised group of insertion sequences. Its transposase and organisation suggest that it transposes using a different mechanism to that of other known transposons. The IS was shown to excise as a circular form, which is accompanied by the formation of a resealed donor plasmid backbone. We also demonstrate that TnpA, which is less than half the length of other transposases, is responsible for this and for ISHp608 transposition. Transposition was shown to be site specific: both insertion and transposon excision require a conserved target, 5'TTAC. Deletion analysis suggested that potential secondary structures at the left and right ends are important for transposition. In vitro TnpA bound both ends, showed a strong preference for a specific single-stranded DNA and introduced a single-strand break on the same strand at each end. Although many of the characteristics of ISHp608 appear similar to rolling-circle transposons, there are differences suggesting that, overall, transposition occurs by a different mechanism. The results have permitted the formulation of several related models.

Functional domains of the IS1 transposase: analysis in vivo and in vitro

The IS1 bacterial insertion sequence family, considered to be restricted to Enterobacteria, has now been extended to other Eubacteria and to Archaebacteria, reviving interest in its study. To analyse the functional domains of the InsAB' transposase of IS1A, a representative of this family, we used an in vivo system which measures IS1-promoted rescue of a temperature-sensitive pSC101 plasmid by fusion with a pBR322::IS1 derivative. We also describe the partial purification of the IS1 transposase and the development of several in vitro assays for transposase activity. These included a DNA band shift assay, a transposase-mediated cleavage assay and an integration assay. Alignments of IS family members (http://www-is.biotoul.fr) not only confirmed the presence of an N-terminal helix-turn-helix and a C-terminal DDE motif in InsAB', but also revealed a putative N-terminal zinc finger. We have combined the in vitro and in vivo tests to carry out a functional analysis of InsAB' using a series of site-directed InsAB' mutants based on these alignments. The results demonstrate that appropriate mutations in the zinc finger and helix-turn-helix motifs result in loss of binding activity to the ends of IS1 whereas mutations in the DDE domain are affected in subsequent transposition steps but not in end binding.

Isolation and characterization of Tn-Dha1, a transposon containing the tetrachloroethene reductive dehalogenase of Desulfitobacterium hafniense strain TCE1

A new 9.9 kb catabolic transposon, Tn-Dha1, containing the gene responsible for tetrachloroethene (PCE) reductive dechlorination activity, was isolated from Desulfitobacterium hafniense strain TCE1. Two fully identical copies of the insertion sequence ISDha1, a new member of the IS256 family, surround the gene cluster pceABCT, a truncated gene for another transposase and a short open reading frame with homology to a member of the twin-arginine transport system (tatA). Evidence was obtained by Southern blot for an alternative form of the transposon element as a circular molecule containing only one copy of ISDha1. This latter structure most probably represents a dead-end product of the transposition of Tn-Dha1. Strong indications for the transposition activity of ISDha1 were given by polymerase chain reaction (PCR) amplification and sequencing of the intervening sequence located between both inverted repeats (IR) of ISDha1 (IR junction). A stable genomic ISDha1 tandem was excluded by quantitative real-time PCR. Promoter mapping of the pceA gene, encoding the reductive dehalogenase, revealed the presence of a strong promoter partially encoded in the right inverted repeat of ISDha1. A sequence comparison with pce gene clusters from Desulfitobacterium sp. strains PCE-S and Y51 and from Dehalobacter restrictus, all of which show 100% identity for the pceAB genes, indicated that both Desulfitobacterium strains seem to possess the same transposon structure, whereas only the pceABCT gene cluster is conserved in D. restrictus.

IS911 partial transposition products and their processing by the Escherichia coli RecG helicase

Insertion of bacterial insertion sequence IS911 can often be directed to sequences resembling its ends. We have investigated this type of transposition and shown that it can occur via cleavage of a single end and its targeted transfer next to another end. The single end transfer (SET) events generate branched DNA molecules that contain a nicked Holliday junction and can be considered as partial transposition products. Our results indicate that these can be processed by the Escherichia coli host independently of IS911-encoded proteins. Such resolution depends on the presence of homologous DNA regions neighbouring the cross-over point in the SET molecule. Processing is often accompanied by sequence conversion between donor and target sequences, suggesting that branch migration is involved. We show that resolution is greatly reduced in a recG host. Thus, the branched DNA-specific helicase, RecG, involved in processing of potentially lethal DNA structures such as stalled replication forks, also intervenes in the resolution of partial IS911 transposition products.

-1 frameshifting at a CGA AAG hexanucleotide site is required for transposition of insertion sequence IS1222

The discovery of programmed -1 frameshifting at the hexanucleotide shift site CGA_AAG, in addition to the classical X_XXY_YYZ heptanucleotide shift sequences, prompted a search for instances among eubacterial insertion sequence elements. IS1222 has a CGA_AAG shift site. A genetic analysis revealed that frameshifting at this site is required for transposition.

Host processing of branched DNA intermediates is involved in targeted transposition of IS911

A simplified system using bacterial insertion sequence IS911 has been developed to investigate targeted insertion next to DNA sequences resembling IS ends. We show here that these IR-targeted events occur by an unusual mechanism. In the circular IS911 transposition intermediate the two IRs are abutted to form an IR/IR junction. IR-targeted insertion involves transfer of a single end of the junction to the target IR to generate a branched DNA structure. The single-end transfer (SET) intermediate, but not the final insertion product, can be detected in an in vitro reaction. SET intermediates must be processed by the bacterial host to obtain the final insertion products. Sequence analysis of these IR-targeted insertion products and of those obtained in vivo revealed high levels of DNA sequence conversion in which mutations from one IR were transferred to another. These sequence changes cannot be explained by the classic transposition pathway. A model is presented in which the four-way Holliday-like junction created by SET is processed by host-mediated branch migration, resolution, repair and replication. This pathway resembles those described for processing other branched DNA structures such as stalled replication forks.

Gene conversion in transposition of Escherichia coli element IS30

The mobile element IS30 has dual target specificity, since it can integrate adjacent to the inverted repeat (IR) of another IS30 copy or into hot-spot sequences characterized by a well-defined consensus showing no similarity to the ends of the element. The result of such integrations into these targets is different, as gene conversion events take place frequently during insertion next to an IR end, while this phenomenon has never been observed in targeting hot-spot sequences. Conversion events in IR-targeting cannot be explained exclusively by the activity of the transposase, but suggest the involvement of the homologous recombination and repair machinery of the host cell. Here, we show that the homology between the donor and target sequences is required for conversion and the starting point of the process is the site of integration. The frequency of conversion depends on the distance of mutations from the end of the targeted element. Remarkable bias is found in the role of donor and target DNA, since generally the donor sequence is converted depending on the target. Conversion was shown to occur also without formation of transposition products. All these data are consistent with the idea of the establishment, migration and resolution of a Holliday-like cruciform structure, which can be responsible for conversion events. To explain the variety of conversion products in IR-targeting, a molecular model has been proposed and discussed.

Detection and analysis of transpositionally active head-to-tail dimers in three additional Escherichia coli IS elements

This study demonstrates that Escherichia coli insertion elements IS3, IS150 and IS186 are able to form transpositionally active head-to-tail dimers which show similar structure and transpositional activity to the dimers of IS2, IS21 and IS30. These structures arise by joining of the left and right inverted repeats (IRs) of two elements. The resulting junction includes a spacer region (SR) of a few base pairs derived from the flanking sequence of one of the reacting IRs. Head-to-tail dimers of IS3, IS150 and IS186 are unstable due to their transpositional activity. They can be resolved in two ways that seem to form a general rule for those elements reported to form dimers. One way is a site-specific process (dimer dissolution) which is accompanied by the loss of one IS copy along with the SR. The other is 'classical' transposition where the joined ends integrate into the target DNA. In intramolecular transposition this often gives rise to deletion formation, whereas in intermolecular transposition it gives rise to replicon fusion. The results presented for IS3, IS150 and IS186 are in accordance with the IS dimer model, which is in turn consistent with models based on covalently closed minicircles.

Genetic variability of the frameshift region in IS911 transposable elements from Escherichia coli clinical isolates

The IS911 bacterial transposable element has been analyzed for its mechanism of transposition and for the way it controls the expression of its genes by programmed -1 translational frameshifting. In the present study the prevalence of IS911 has been determined in the Enterobacteriaceae family and in other Gram-negative bacilli. Three variants, found in Escherichia coli clinical isolates and having mutations in the region implicated in frameshifting, were functionally characterized. All three were altered in their frameshifting and transposition abilities, suggesting that the frameshift region of IS911 may constitute a target for mutations reducing the transposition frequency of this mobile element in natural populations of E. coli.

Escherichia coli insertion sequence IS150: transposition via circular and linear intermediates

IS150, a member of the widespread IS3 family, contains two consecutive out-of-phase open reading frames, orfA and orfB, that partially overlap. These open reading frames encode three proteins, InsA, InsB, and the InsAB protein, which is jointly encoded by both open reading frames by means of programmed translational frameshifting. We demonstrate that the InsAB protein represents the IS150 element's transposase. In vivo, the wild-type IS150 element generates circular excision products and linear IS150 molecules. Circular and linear species have previously been detected with mutant derivatives of other members of the IS3 family. Our finding supports the assumption that these products represent true transposition intermediates of members of this family. Analysis of the molecular nature of these two species suggested that the circular forms are precursors of the linear molecules. Elimination of InsA synthesis within the otherwise intact element led to accumulation of large amounts of the linear species, indicating that the primary role of InsA may be to prevent abortive production of the linear species and to couple generation of these species to productive insertion events.

Palindromic unit-independent transposition of IS1397 in Yersinia pestis

Palindromic units (PUs) are intergenic repeated sequences scattered over the chromosomes of Escherichia coli and several other enterobacteria. In the latter, IS1397, an E. coli insertion sequence specific to PUs, transposes into PUs with sequences close to the E. coli consensus. Reasons for this insertion specificity can relate to either a direct recognition of the target (by its sequence or its structure) by the transposase or an interaction between a specific host protein and the PU target DNA sequence. In this study, we show that for Yersinia pestis, a species deprived of PUs, IS1397 can transpose onto its chromosome, with transpositional hot spots. Our results are in favor of a direct recognition of target DNA by IS1397 transposase.

Timing and targeting of P-element local transposition in the male germline cells of Drosophila melanogaster

The P element in Drosophila melanogaster preferentially transposes into nearby sites. The local insertions display a preferential orientation toward the starting element. We investigated the mechanism of the P-element local transposition by isolating and characterizing local insertions in the male germline. We designed a genetic screen employing a marker gene that is carried in the P element and is dose sensitive. This dose effect allows isolation of flies containing newly transposed P elements in the presence of the starting element. A rapid molecular screen with PCR was used to identify 45 local insertions located within an approximately 40-kb genomic region on both sides of the starting element. Our system permits the isolation of the cluster progeny derived from a single insertion event, but none was isolated. The data suggest that local transposition occurs in the meiotic cell cycle. Nearly all of the local insertions were located within the promoter regions of the genes that were active in the male germline cells, suggesting that local insertions target predominantly active promoters. Our analysis shows that local transposition of the P element is highly regulated, displaying a cell-type specificity and a target specificity.

The basis of asymmetry in IS2 transposition

In the first step of IS2 transposition, the formation of an IS2 minicircle, the roles of the two IS ends differ. Terminal cleavage initiates exclusively at the right inverted repeat (IRR) - the donor end - whereas IRL is always the target. At the resulting minicircle junction, the two abutted ends are separated by a spacer of 1 or 2 basepairs. In this study, we have identified the determinants of donor and target function. The inability of IRL to act as a donor results largely from two sequence differences between IRL and IRR - an extra basepair between the conserved transposase binding sequences and the end of the element, and a change of the terminal dinucleotide from CA-3' to TA-3'. These two changes also impose a characteristic size on the minicircle junction spacer. The only sequences required for the efficient target function of IRL appear to be contained within the segment from position 11-42. Although IRR can function as a target, its shorter length and additional contacts with transposase (positions 1-7) result in minicircles with longer, and inappropriate, spacers. We propose a model for the synaptic complex in which the terminus of IRL makes different contacts with the transposase for the initial and final strand transfer steps. The sequence differences between IRR and IRL, and the behavioural characteristics of IRL that result from them, have probably been selected because they optimize expression of transposase from the minicircle junction promoter, Pjunc.

The terminal inverted repeats of IS911: requirements for synaptic complex assembly and activity

The bacterial insertion sequence IS911 transposes via a covalently closed circular intermediate. Circle formation involves transposase-mediated pairing of both insertion sequence ends. While full-length transposase, OrfAB, binds poorly in vitro to IS911 DNA fragments carrying a copy of the IS911 end, truncated protein derivatives carrying the first 135 (OrfAB[1-135]) or 149 (OrfAB[1-149]) amino acid residues bind efficiently. They generate a paired-end complex containing two such fragments which resembles that expected for the first synaptic complex. Shorter protein derivatives lacking a region involved in multimerisation do not form these complexes but modify the binding of OrfAB[1-135] and OrfAB[1-149]. DNaseI footprinting demonstrated that OrfAB[1-149] protects a sub-terminal (internal) region of the inverted repeats which includes two blocks of sequence (beta and gamma) conserved between the left (IRL) and right (IRR) ends. DNA binding assays in vitro and measurement of recombination activity in vivo of sequential deletion derivatives of the two inverted repeats suggested a model in which the N-terminal region of OrfAB binds the conserved boxes beta and gamma in a sequence-specific manner and anchors the two insertion sequence ends into a paired-end complex. The external region of the inverted repeat is proposed to contact the C-terminal transposase domain carrying the catalytic site.

IS1294, a DNA element that transposes by RC transposition

IS1294, found on the ColD-like resistance plasmid pUB2380, is IS91-like. It is an active 1.7-kb insertion sequence that lacks terminal inverted repeats, displays insertion-site specificity, and does not generate direct repeats of the target site. The element has one large open reading frame, tnp(1294), encoding a transposase of 351 amino acids, related to members of the REP family of replication proteins used by RC-plasmids of gram-positive bacteria. IS1294 transposes using rolling-circle replication, initiated at one end of the element, oriIS, and terminated at the other, terIS. oriIS and terIS are highly conserved among like IS elements. oriIS resembles the leading strand replication origins of RC-plasmids; terIS resembles a rho-independent transcription terminator. IS1294 mediates not only its own transposition, but also sequences adjacent to terIS. A transposition model for IS1294 and related elements, involving rolling-circle replication and single-strand DNA intermediates, is presented.

Transposition in Lactobacillus sakei: inactivation of a second lactocin S operon by the insertion of IS1520, a new member of the IS3 family of insertion sequences

The analysis of spontaneous bacteriocin-negative mutants has led to the identification and characterization of a new, transpositionally active, insertion sequence of the IS3 family in the lactocin-S-producing Lactobacillus sakei strain L45. The element, which has been designated IS1520, is 1302 bp long with 10 bp perfect inverted repeat ends and generates direct repeats of a trinucleotide of target sequence upon transposition to the lactocin S locus. IS1520 encodes two consecutive, partially overlapping, major ORFs, which are frameshifted in a manner typical of the IS3 family. Despite a low overall DNA sequence similarity, the putative frameshifting region of IS1520 is highly similar to the corresponding region of IS1163, which is a related element previously shown to be active in L. sakei L45.

A minimal system for Tn7 transposition: the transposon-encoded proteins TnsA and TnsB can execute DNA breakage and joining reactions that generate circularized Tn7 species

In the presence of ATP and Mg(2+), the bacterial transposon Tn7 translocates via a cut and paste mechanism executed by the transposon-encoded proteins TnsA+TnsB+TnsC+TnsD. We report here that in the presence of Mn(2+), TnsA+TnsB alone can execute the DNA breakage and joining reactions of Tn7 recombination. ATP is not essential in this minimal system, revealing that this cofactor is not directly involved in the chemical steps of recombination. In both the TnsAB and TnsABC+D systems, recombination initiates with double-strand breaks at each transposon end that cut Tn7 away from flanking donor DNA. In the minimal system, breakage occurs predominantly at a single transposon end and the subsequent end-joining reactions are intramolecular, with the exposed 3' termini of a broken transposon end joining near the other end of the Tn7 element in the same donor molecule to form circular transposon species. In contrast, in TnsABC+D recombination, breaks occur at both ends of Tn7 and the two ends join to a target site on a different DNA molecule to form an intermolecular simple insertion. This demonstration of the capacity of TnsAB to execute breakage and joining reactions supports the view that these proteins form the Tn7 transposase.

The role of tandem IS dimers in IS911 transposition

Using a combined in vivo and in vitro approach, we demonstrated that the transposition products generated by IS911 from a dimeric donor plasmid are different from those generated from a plasmid monomer. When carried by a monomeric plasmid donor, free IS911 transposon circles are generated by intra-IS recombination in which one IS end undergoes attack by the other. These represent transposition intermediates that undergo integration using the abutted left (IRL) and right (IRR) ends of the element, the active IRR-IRL junction, to generate simple insertions. In contrast, the two IS911 copies carried by a dimeric donor plasmid not only underwent intra-IS recombination to generate transposon circles but additionally participated in inter-IS recombination. This also creates an active IRR-IRL junction by generating a head-to-tail IS tandem dimer ([IS]2) in which one of the original plasmid backbone copies is eliminated in the formation of the junction. Both transposon circles and IS tandem dimers are generated from an intermediate in which two transposon ends are retained by a single strand joint to generate a figure 8 molecule. Inter-IS figure 8 molecules generated in vitro could be resolved into the [IS]2 form following introduction into a host strain by transformation. Resolution did not require IS911 transposase. The [IS]2 structure was stable in the absence of transposase but was highly unstable in its presence both in vivo and in vitro. Previous studies had demonstrated that the IRR-IRL junction promotes efficient intermolecular integration and intramolecular deletions both in vivo and in vitro. Integration of the [IS]2 derivative would result in a product that resembles a co-integrate structure. It is also shown here that the IRR-IRL junction of the [IS]2 form and derivative structures can specifically target one of the other ends in an intramolecular transposition reaction to generate transposon circles in vitro. These results not only demonstrate that IS911 (and presumably other members of the IS3 family) is capable of generating a range of transposition products, it also provides a mechanistic framework which explains the formation and activity of such structures previously observed for several other unrelated IS elements. This behaviour is probably characteristic of a large number of IS elements.

Reactivation of insertionally inactivated Shiga toxin 2 genes of Escherichia coli O157:H7 caused by nonreplicative transposition of the insertion sequence

IS1203v is an insertion sequence which has been found in inactivated Shiga toxin 2 genes of Escherichia coli O157:H7. We analyzed the transpositional mechanism of IS1203v in order to investigate whether the Shiga toxin 2 genes inactivated by IS1203v could revert to the wild type. When the transposase activity of IS1203v was enhanced by artificial frameshifting, IS1203v was obviously excised from the Shiga toxin 2 gene in a circular form. The IS1203v circle consisted of the entire IS1203v, but an extra 3-bp sequence (ATC) intervened between the 5' and 3' ends of IS1203v. The extra 3-bp sequence was identical to a direct repeat which was probably generated upon insertion. Moreover, we detected the Shiga toxin 2 gene with a precise excision of IS1203v. In the wild-type situation, the transposition products of IS1203v could be observed by PCR amplification. These results show that IS1203v can transpose in a nonreplicative manner and that the Shiga toxin gene inactivated by this insertion sequence can revert to the wild type.

Timing of developmentally programmed excision and circularization of Paramecium internal eliminated sequences

Paramecium internal eliminated sequences (IESs) are short AT-rich DNA elements that are precisely eliminated from the germ line genome during development of the somatic macronucleus. They are flanked by one 5'-TA-3' dinucleotide on each side, a single copy of which remains at the donor site after excision. The timing of their excision was examined in synchronized conjugating cells by quantitative PCR. Significant amplification of the germ line genome was observed prior to IES excision, which starts 12 to 14 h after initiation of conjugation and extends over a 2- to 4-h period. Following excision, two IESs were shown to form extrachromosomal circles that can be readily detected on Southern blots of genomic DNA from cells undergoing macronuclear development. On these circular molecules, covalently joined IES ends are separated by one copy of the flanking 5'-TA-3' repeat. The similar structures of the junctions formed on the excised and donor molecules point to a central role for this dinucleotide in IES excision.

IS911 transposition is regulated by protein-protein interactions via a leucine zipper motif

Efficient intermolecular transposition of bacterial insertion sequence IS911 involves the activities of two element-encoded proteins: the transposase, OrfAB, and a regulatory factor, OrfA. OrfA shares the majority of its amino acid sequence with the N-terminal part of OrfAB. This includes a putative helix-turn-helix and three of four heptads of a leucine zipper motif. OrfA strongly stimulates OrfAB-mediated intermolecular transposition both in vivo and in vitro. The present results support the notion that this is accomplished by direct interaction between these two proteins via the leucine zipper. We used both a genetic approach, based on gene fusions with phage lambda repressor, and a physical approach, involving co-immunoprecipitation, to show that OrfA not only undergoes oligomerisation but is capable of engaging with OrfAB to form heteromultimers, and that the leucine zipper is necessary for both types of interaction. Furthermore, mutation of the leucine zipper in OrfA inactivated its regulatory function. Previous observations demonstrated that the integrity of the leucine zipper motif was also important for OrfAB binding to the IS911 terminal inverted repeats. Here, we show, in gel shift experiments, using a derivative of OrfAB deleted for the C-terminal catalytic domain, OrfAB[1-149], that the protein is capable of pairing two inverted repeats to generate a species resembling a "synaptic complex". Preincubation of OrfAB[1-149] with OrfA dramatically reduced formation of this complex and favored formation of an alternative complex devoid of OrfA. Together these results suggest that OrfA exerts its regulatory effect by interacting transiently with OrfAB via the leucine zipper and modifying OrfAB binding to the inverted repeats.

Tn5: A molecular window on transposition

DNA transposition is an underlying process involved in the remodeling of genomes in all types of organisms. We analyze the multiple steps in cut-and-paste transposition using the bacterial transposon Tn5 as a model. This system is particularly illuminating because of the existence of structural, genetic, and biochemical information regarding the two participating specific macromolecules: the transposase and the 19-bp sequences that define the ends of the transposon. However, most of the insights should be of general interest because of similarities to other transposition-like systems such as HIV-1 DNA integration into the host genome.

Linearization and transposition of circular molecules of insertion sequence IS3

IS3 transposase has been shown to promote production of characteristic circular and linear IS3 molecules from the IS3-carrying plasmid; IS3 circles have the entire IS3 sequence with terminal inverted repeats, IRL and IRR, which are separated by a three base-pair sequence originally flanking either end in the parental plasmid, whereas linear IS3 molecules have three nucleotide overhangs at their 5' ends. Here, we showed that a plasmid carrying an IS3 derivative, which is flanked by different sequences at both ends, generated IS3 circles and linear IS3 molecules owing to the action of transposase. Cloning and sequencing analyses of the linear molecules showed that each had the same 5'-protruding three nucleotide overhanging sequences at both ends, suggesting that the linear molecules were not generated from the parental plasmid by the two double-strand breaks at both end regions of IS3. The plasmid carrying IS3 with a two base-pair mutation in the terminal dinucleotide, which would be required for transposase to cleave the 3' end of IS3, could still generate linear molecules as well as circles. Plasmids bearing an IS3 circle were cleaved by transposase and gave linear molecules with the same 5'-protruding three nucleotide overhanging sequences. These show that the linear molecules are generated from IS3 circles via a double-strand break at the three base-pair intervening sequence. Plasmids carrying an IS3 circle with the two base-pair end mutation still were cleaved by transposase, though with reduced efficiencies, suggesting that IS3 transposase has the ability to cleave not only the 3' end of IS3, but a site three nucleotides from the 5' end of IS3. IS3 circles also were shown to transpose to the target plasmids. The end mutation almost completely inhibited this transposition, showing that the terminal dinucleotides are important for the transfer of the 3' end of IS3 to the target as well as for the end cleavage.

Importance of illegitimate recombination and transposition in IS30-associated excision events

In the present study we report on the excision of IS30 elements and IS30-derived composite transposons. Frequent loss of IS30 was observed during dissolution of dimeric IS30 structures, containing IR-IR junctions, leading to resealed donor molecules. In contrast, unambiguous transpositional excision resulting in resealed remainder products could not be identified in the case of a monomeric element. The bias in the excision of monomeric and dimeric IS30 structures indicates a difference in the molecular mechanism of transposition of IS30 monomers and dimers. Sequence data on the rarely detected plasmids missing full IS or Tn copies rather suggest that all products were derived from illegitimate recombination. The reaction occurred between short homologies and was independent of the transposase activity. Similar IS30 excision events accompanied by multiple plasmid or genome rearrangements were detected in Pseudomonas putida and Rhizobium meliloti, yielding stable replicons that retained the selective marker gene of the transposon. We provide evidence that both transposition and illegitimate recombination can contribute to the stabilization of replicons through the elimination of IS elements, which emphasizes the evolutionary significance of these events.

Formation and transposition of the covalently closed IS30 circle: the relation between tandem dimers and monomeric circles

In the present study, we demonstrate that a circular IS30 element acts as an intermediate for simple insertion. Covalently closed IS and Tn circles constructed in vitro are suitable for integration into the host genome. Minicircle integration displays all the characteristics of transpositional fusion mediated by the (IS30 )2 dimer regarding target selection and target duplication. Evidence is provided for in vivo circularization of the element located either on plasmids or on the genome. It is shown that circle formation can occur through alternative pathways. One of them is excision of IS30 from a hot spot via joining the IRs. This reaction resembles the site-specific dimerization that leads to (IS30 )2 establishment. The other process is the dissolution of (IS30 )2 dimer, when the element is excised from an IR-IR joint. These pathways differ basically in the fate of the donor replicon: only dimer dissolution gives rise to resealed donor backbone. Analysis of minicircles and the rearranged donor replicons led us to propose a molecular model that can account for differences between the circle-generating processes. Our focus was to the dissolution of IR-IR joints located on the host genome, because these events promoted extensive genomic rearrangements and accompanied minicircle formation. The results present the possibility of host genome reorganization by IS30-like transposition.

Transposition of IS1 circles

BACKGROUND

IS1, the smallest active transposable element in bacteria, encodes transposase. IS1 transposase promotes transposition as well as production of miniplasmids from a plasmid carrying IS1 by deletion of the region adjacent to IS1. The IS1 transposase also promotes production of IS1 circles consisting of the entire IS1 sequence and a sequence, 6-9 bp in length, as a spacer between terminal inverted repeats of IS1. The biological significance of the generation of IS1 circles is not known.

RESULTS

Plasmids carrying an IS1 circle with a spacer sequence 6-9 bp long transposed to target plasmids at a very high frequency when transposase was produced from a co-resident plasmid. The products were target plasmids with the donor plasmid inserted at the ends of IS1 in the IS1 circle. This insertion accompanied the removal of the spacer sequence and duplication of the sequence at the target site. IS1 circles with a much longer spacer sequence transposed less frequently. The SOS response was induced in cells harbouring a plasmid with an IS1 circle owing to transposase. IS1 circles could transpose in the strain deficient in H-NS, a nucleoid-associated DNA-binding protein known to be required for the transposition of IS1.

CONCLUSIONS

IS1 circles appear to act as intermediates for simple insertion into the target DNA via cleavage of the circles which induces the SOS response. H-NS may function in promoting the assembly of an active IS1 DNA-transposase complex at the terminal inverted repeats.

Excision of IS492 requires flanking target sequences and results in circle formation in Pseudoalteromonas atlantica

The gram-negative marine bacterium Pseudoalteromonas atlantica produces extracellular polysaccharide (EPS) that is important in biofilm formation by this bacterium. Insertion and precise excision of IS492 at a locus essential for extracellular polysaccharide production (eps) controls phase variation of EPS production in P. atlantica. Examination of IS492 transposition in P. atlantica by using a PCR-based assay revealed a circular form of IS492 that may be an intermediate in transposition or a terminal product of excision. The DNA sequence of the IS492 circle junction indicates that the ends of the element are juxtaposed with a 5-bp spacer sequence. This spacer sequence corresponds to the 5-bp duplication of the chromosomal target sequence found at all IS492 insertion sites on the P. atlantica chromosome that we identified by using inverse PCR. IS492 circle formation correlated with precise excision of IS492 from the P. atlantica eps target sequence when introduced into Escherichia coli on a plasmid. Deletion analyses of the flanking host sequences at the eps insertion site for IS492 demonstrated that the 5-bp duplicated target sequence is essential for precise excision of IS492 and circle formation in E. coli. Excision of IS492 in E. coli also depends on the level of expression of the putative transposase, MooV. A regulatory role for the circular form of IS492 is suggested by the creation of a new strong promoter for expression of mooV by the joining of the ends of the insertion sequence element at the circle junction.

Tdd-4, a DNA transposon of Dictyostelium that encodes proteins similar to LTR retroelement integrases

Tdd-4 is the first DNA transposon to be isolated from Dictyostelium discoideum. This element was isolated by insertion into a target plasmid. Two classes of elements were identified which include a 3.8 kb version and a 3.4 kb deleted version. Sequence analysis reveals that the 145 bp inverted terminal repeats contain the 5'-TGellipsisCA-3' conserved terminal dinucleotides found in prokaryotic transposons and integrated LTR retroelement DNA sequences. Tdd-4 open reading frames are assembled by removal of six introns. Introns 1-5 conform to the GT-AG rule, whereas intron 6 appears to be an AT-AA intron. Also, intron 6 undergoes an alternative 5' splicing reaction. The alternatively spliced region encodes 15 tandem SPXX repeats that are proposed to function as a DNA binding motif. By analogy to other transposons that encode two proteins from the same gene, the full-length Tdd-4 protein is the putative transposase and the truncated Tdd-4 protein is the putative transposition inhibitor. Protein database searches demonstrate Tdd-4 encoded proteins are unique for a DNA element by containing similarities to retroviral/retrotransposon integrases. The putative Tdd-4 transposase contains the same structural relationship as integrases by possessing an N-terminal HHCC motif, a central DDE motif and a C-terminal DNA-binding domain composed of the SPXX motif.

IS911 transposon circles give rise to linear forms that can undergo integration in vitro

High levels of expression of the transposase OrfAB of bacterial insertion sequence IS911 leads to the formation of excised transposon circles, in which the two abutted ends are separated by 3 bp. Initially, OrfAB catalyses only single-strand cleavage at one 3' transposon end and strand transfer of that end to the other. It is believed that this molecule, in which both transposon ends are held together in a single-strand bridge, is then converted to the circular form by the action of host factors. The transposon circles can be integrated efficiently into an appropriate target in vivo and in vitro in the presence of OrfAB and a second IS911 protein OrfA. In the results reported here, we have identified linear transposon forms in vivo from a transposon present in a plasmid, raising the possibility that IS911 can also transpose using a cut-and-paste mechanism. However, the linear species appeared not to be derived directly from the plasmid-based copy by direct double-strand cleavages at both ends, but from preformed excised transposon circles. This was confirmed further by the observation that OrfAB can cleave a cloned circle junction both in vivo and in vitro by two single-strand cleavages at the 3' transposon ends to generate a linear transposon form with a 3'-OH and a three-nucleotide 5' overhang at the ends. Moreover, while significantly less efficient than the transposon circle, a precleaved linear transposon underwent detectable levels of integration in vitro. The possible role of such molecules in the IS911 transposition pathway is discussed.

Multiple oligomerisation domains in the IS911 transposase: a leucine zipper motif is essential for activity

Structure-function relationships involved in oligomerisation of the transposase OrfAB of the bacterial insertion sequence IS911 have been investigated. Site-directed mutagenesis and sequential deletion coupled with immunoprecipitation have led to the definition of three regions of the protein capable of promoting multimerisation. These include a region predicted to assume a coiled-coil conformation, which is shown to be essential for activity, promoting correct multimerisation of the N-terminal domain of OrfAB and sequence-specific binding to the IS911 terminal inverted repeats mediated by this domain. This region presents the structural and functional characteristics of the leucine zipper motif described in eukaryotic proteins. The two other regions are located further towards the C-terminal end of the protein, adjacent to the leucine zipper and in the region that carries the conserved catalytic DD(35)E motif.

Tn10 transposition via a DNA hairpin intermediate

We present evidence that excision of the nonreplicative transposon Tn10 involves three distinct chemical steps, first-strand nicking, hairpin formation, and hairpin resolution. This three-step mechanism makes it possible for a single protein-active site to cleave two DNA strands of opposite polarity, as appears to be the case in this reaction. We infer the existence of alternating bifunctionality within the active site with suitable modulation of substrate components between steps. DNA double-strand breaks are also made by a "hairpin mechanism" in V(D)J recombination, possibly reflecting the same basic constraints faced in the Tn10 system. Similarities in the basic chemical steps in Tn10 transposition and V(D)J recombination suggest that the V(D)J mechanism may have evolved from a bacterial transposition system.

Identification and characterization of IS1411, a new insertion sequence which causes transcriptional activation of the phenol degradation genes in Pseudomonas putida

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 sigma70. 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

Insertion sequences (ISs) constitute an important component of most bacterial genomes. Over 500 individual ISs have been described in the literature to date, and many more are being discovered in the ongoing prokaryotic and eukaryotic genome-sequencing projects. The last 10 years have also seen some striking advances in our understanding of the transposition process itself. Not least of these has been the development of various in vitro transposition systems for both prokaryotic and eukaryotic elements and, for several of these, a detailed understanding of the transposition process at the chemical level. This review presents a general overview of the organization and function of insertion sequences of eubacterial, archaebacterial, and eukaryotic origins with particular emphasis on bacterial elements and on different aspects of the transposition mechanism. It also attempts to provide a framework for classification of these elements by assigning them to various families or groups. A total of 443 members of the collection have been grouped in 17 families based on combinations of the following criteria: (i) similarities in genetic organization (arrangement of open reading frames); (ii) marked identities or similarities in the enzymes which mediate the transposition reactions, the recombinases/transposases (Tpases); (iii) similar features of their ends (terminal IRs); and (iv) fate of the nucleotide sequence of their target sites (generation of a direct target duplication of determined length). A brief description of the mechanism(s) involved in the mobility of individual ISs in each family and of the structure-function relationships of the individual Tpases is included where available.

ISD1, an insertion element from the sulfate-reducing bacterium Desulfovibrio vulgaris Hildenborough: structure, transposition, and distribution

Insertion element ISD1, discovered when its transposition caused the insertional inactivation of an introduced sacB gene, is present in two copies in the genome of Desulfovibrio vulgaris Hildenborough. Southern blot analysis indicated at least two insertion sites in the sacB gene. Cloning and sequencing of a transposed copy of ISD1 indicated a length of 1,200 bp with a pair of 44-bp imperfect inverted repeats at the ends, flanked by a direct repeat of the 4-bp target sequence. AAGG and AATT were found to function as target sequences. ISD1 encodes a transposase from two overlapping open reading frames by programmed translational frameshifting at an A6G shifty codon motif. Sequence comparison showed that ISD1 belongs to the IS3 family. Isolation and analysis of the chromosomal copies, ISD1-A and ISD1-B, by PCR and sequencing indicated that these are not flanked by direct repeats. ISD1-A is inserted in a region of the chromosome containing the gapdh-pgk genes (encoding glyceraldehyde-3-phosphate dehydrogenase and phosphoglycerate kinase). Active transposition to other loci in the genome was demonstrated, offering the potential of a new tool for gene cloning and mutagenesis. ISD1 is the first transposable element described for the sulfate reducers, a large and environmentally important group of bacteria. The distribution of ISD1 in genomes of sulfate-reducing bacteria is limited. A single copy is present in the genome of D. desulfuricans Norway.

Terminal inverted repeats of insertion sequence IS30 serve as targets for transposition

In the present study, we demonstrate that the terminal inverted repeats of the Escherichia coli insertion sequence IS30 are functional target sites for the transposition of the (IS30)2 dimer, which represents an intermediate structure in the transposition of IS30. Comparative analysis of various target regions revealed that the left and right ends differ in their "attractivity." In our experiments, the joined left and right ends, i.e., the (IS30)2 intermediate structure, was found to be the most preferred target. It was also shown that flanking sequences can influence the target activity of the terminal repeats. The functional part of the target region was localized in the inverted repeats by means of mutational analysis, and it corresponds to the binding site of IS30 transposase. Insertion of 1 bp into the right inverted repeat resulted in unusual target duplication accompanied by gene conversion. The choice of the terminal inverted repeats as targets in transposition leads to the reconstruction of the (IS30)2 structure, which may induce a cascade of further rearrangements. Therefore, this process can play a role in the evolution of the genome.

Isolation and characterization of IS1 circles

Transposase encoded by insertion sequence IS1 is produced from two out-of-phase reading frames by translational frameshifting that occurs in a run of adenines. An IS1 mutant with a single adenine insertion in the run of adenines efficiently produces transposase, resulting in generation of miniplasmids by deletion for a region adjacent to IS1 from a plasmid carrying the IS1 mutant. Here, we found that besides miniplasmids, cells harboring the plasmid contained minicircles without the region required for replication. Cloning and DNA sequencing of the minicircles revealed that most of them were IS1 circles consisting of the entire IS1 sequence and a sequence, 5-9 bp in length, which intervenes between terminal inverted repeats, IRL and IRR, of IS1. Analysis of more IS1 circles isolated by polymerase chain reaction revealed that the intervening sequence was derived from the region flanking either IRL or IRR in the parental plasmid, suggesting that IS1 circles are generated by an excision event from the parental plasmid. The IS1 circles may be formed due to the cointegration reaction occurring within the parental plasmid carrying IS1.

Mobilization of a Minos transposon in Drosophila melanogaster chromosomes and chromatid repair by heteroduplex formation

Transposase-mediated mobilization of the element Minos has been studied in the Drosophila melanogaster genome. Excision and transposition of a nonautonomous Minos transposon in the presence of a Minos transposase gene was detected with a dominant eye color marker carried by the transposon. Frequencies of excision in somatic tissues and in the germ line were higher in flies heterozygous for the transposon than in homozygotes or hemizygotes. Transposition of a X chromosome-linked insertion of Minos into new autosomal sites occurred in 1-12% of males expressing transposase, suggesting that this system is usable for gene tagging and enhancer trapping in Drosophila. Sequence analysis of PCR-amplified donor sites after excision showed precise restoration of the original target sequence in approximately 75% of events in heterozygotes and the presence of footprints or partially deleted elements in the remaining events. Most footprints consisted of the four terminal bases of the transposon, flanked by the TA target duplication. Sequencing of a chromosomal donor site that was directly cloned after excision showed a characteristic two-base mismatch heteroduplex in the center of the 6-bp footprint. Circular extrachromosomal forms of the transposon, presumably representing excised Minos elements, could be detected only in the presence of transposase. A model for chromatid repair after Minos excision is discussed in which staggered cuts are first produced at the ends of the inverted repeats, the broken chromatid ends are joined, and the resulting heteroduplex is subsequently repaired. The model also suggests a simple mechanism for the production of the target site duplication and for regeneration of the transposon ends during reintegration.