Structural and regulatory divergence among site-specific recombination genes of lambdoid phage

Characterization of the relationship between integrase, excisionase and antirepressor activities associated with a superinfecting Shiga toxin encoding bacteriophage

Shigatoxigenic Escherichia coli emerged as new food borne pathogens in the early 1980s, primarily driven by the dispersal of Shiga toxin-encoding lambdoid bacteriophages. At least some of these Stx phages display superinfection phenotypes, which differ significantly from lambda phage itself, driving through in situ recombination further phage evolution, increasing host range and potentially increasing the host's pathogenic profile. Here, increasing levels of Stx phage Φ24(B) integrase expression in multiple lysogen cultures are demonstrated along with apparently negligible repression of integrase expression by the cognate CI repressor. The Φ24(B) int transcription start site and promoter region were identified and found to differ from in silico predictions. The unidirectional activity of this integrase was determined in an in situ, inducible tri-partite reaction. This indicated that Φ24(B) must encode a novel directionality factor that is controlling excision events during prophage induction. This excisionase was subsequently identified and characterized through complementation experiments. In addition, the previous proposal that a putative antirepressor was responsible for the lack of immunity to superinfection through inactivation of CI has been revisited and a new hypothesis involving the role of this protein in promoting efficient induction of the Φ24(B) prophage is proposed.

Purification and characterization of bacteriophage P22 Xis protein

The temperate bacteriophages lambda and P22 share similarities in their site-specific recombination reactions. Both require phage-encoded integrase (Int) proteins for integrative recombination and excisionase (Xis) proteins for excision. These proteins bind to core-type, arm-type, and Xis binding sites to facilitate the reaction. lambda and P22 Xis proteins are both small proteins (lambda Xis, 72 amino acids; P22 Xis, 116 amino acids) and have basic isoelectric points (for P22 Xis, 9.42; for lambda Xis, 11.16). However, the P22 Xis and lambda Xis primary sequences lack significant similarity at the amino acid level, and the linear organizations of the P22 phage attachment site DNA-binding sites have differences that could be important in quaternary intasome structure. We purified P22 Xis and studied the protein in vitro by means of electrophoretic mobility shift assays and footprinting, cross-linking, gel filtration stoichiometry, and DNA bending assays. We identified one protected site that is bent approximately 137 degrees when bound by P22 Xis. The protein binds cooperatively and at high protein concentrations protects secondary sites that may be important for function. Finally, we aligned the attP arms containing the major Xis binding sites from bacteriophages lambda, P22, L5, HP1, and P2 and the conjugative transposon Tn916. The similarity in alignments among the sites suggests that Xis-containing bacteriophage arms may form similar structures.

A new site-specific integration system for mycobacteria

Site-specific integration into the mycobacterial chromosome can produce stable transformants useful for understanding pathogenesis. However, gene expression can be problematic at certain sites of integration. We have used the Streptomyces phiC31 integration system to integrate vector DNA into Mycobacterium smegmatis, M. bovis BCG, and M. tuberculosis through site-specific recombination. A single dominant insertion site was found in M. smegmatis, as previously reported. Three different insertion sites were found in M. bovis BCG. In M. smegmatis, integrated vectors appear to be far more stable than episomal plasmids during unselected passage in vitro, although excision products are detectable. Plasmids based on the phiC31 integration system could make useful tools for the study of mycobacterial genetics.

Corrected sequence of the bacteriophage p22 genome

We report the first accurate genome sequence for bacteriophage P22, correcting a 0.14% error rate in previously determined sequences. DNA sequencing technology is now good enough that genomes of important model systems like P22 can be sequenced with essentially 100% accuracy with minimal investment of time and resources.

R391: a conjugative integrating mosaic comprised of phage, plasmid, and transposon elements

The conjugative, chromosomally integrating element R391 is the archetype of the IncJ class of mobile genetic elements. Originally found in a South African Providencia rettgeri strain, R391 carries antibiotic and mercury resistance traits, as well as genes involved in mutagenic DNA repair. While initially described as a plasmid, R391 has subsequently been shown to be integrated into the bacterial chromosome, employing a phage-like integration mechanism closely related to that of the SXT element from Vibrio cholerae O139. Analysis of the complete 89-kb nucleotide sequence of R391 has revealed a mosaic structure consisting of elements originating in bacteriophages and plasmids and of transposable elements. A total of 96 open reading frames were identified; of these, 30 could not be assigned a function. Sequence similarity suggests a relationship of large sections of R391 to sequences from Salmonella, in particular those corresponding to the putative conjugative transfer proteins, which are related to the IncHI1 plasmid R27. A composite transposon carrying the kanamycin resistance gene and a novel insertion element were identified. Challenging the previous assumption that IncJ elements are plasmids, no plasmid replicon was identified on R391, suggesting that they cannot replicate autonomously.

Control of directionality in integrase-mediated recombination: examination of recombination directionality factors (RDFs) including Xis and Cox proteins

Similarity between the DNA substrates and products of integrase-mediated site-specific recombination reactions results in a single recombinase enzyme being able to catalyze both the integration and excision reactions. The control of directionality in these reactions is achieved through a class of small accessory factors that favor one reaction while interfering with the other. These proteins, which we will refer to collectively as recombination directionality factors (RDFs), play architectural roles in reactions catalyzed by their cognate recombinases and have been identified in conjunction with both tyrosine and serine integrases. Previously identified RDFs are typically small, basic and have diverse amino acid sequences. A subset of RDFs, the cox genes, also function as transcriptional regulators. We present here a compilation of all the known RDF proteins as well as those identified through database mining that we predict to be involved in conferring recombination directionality. Analysis of this group of proteins shows that they can be grouped into distinct sub-groups based on their sequence similarities and that they are likely to have arisen from several independent evolutionary lineages. This compilation will prove useful in recognizing new proteins that confer directionality upon site-specific recombination reactions encoded by plasmids, transposons, phages and prophages.

Site-specific recombination of bacteriophage P22 does not require integration host factor

Site-specific recombination by phages lambda and P22 is carried out by multiprotein-DNA complexes. Integration host factor (IHF) facilitates lambda site-specific recombination by inducing DNA bends necessary to form an active recombinogenic complex. Mutants lacking IHF are over 1,000-fold less proficient in supporting lambda site-specific recombination than wild-type cells. Although the attP region of P22 contains strong IHF binding sites, in vivo measurements of integration and excision frequencies showed that infecting P22 phages can perform site-specific recombination to its maximum efficiency in the absence of IHF. In addition, a plasmid integration assay showed that integrative recombination occurs equally well in wild-type and ihfA mutant cells. P22 integrative recombination is also efficient in Escherichia coli in the absence of functional IHF. These results suggest that nucleoprotein structures proficient for recombination can form in the absence of IHF or that another factor(s) can substitute for IHF in the formation of complexes.

Inducible prophages contribute to Salmonella virulence in mice

We show that Salmonella typhimurium harbours two fully functional prophages, Gifsy-1 and Gifsy-2, that can be induced by standard treatments or, more effectively, by exposing bacteria to hydrogen peroxide. Curing bacteria for the Gifsy-2 prophage significantly reduces Salmonella's ability to establish a systemic infection in mice. Cured strains recover their virulence properties upon relysogenization. Phage Gifsy-2 carries the sodC gene for a periplasmic [Cu,Zn]-superoxide dismutase previously implicated in the bacterial defences against killing by macrophages. The contribution of the Gifsy-1 prophage to virulence - undetectable in the presence of Gifsy-2 as prophage - becomes significant in cells that lack Gifsy-2 but carry the sodC gene integrated in the chromosome. This confirms the involvement of Gifsy-2-encoded SodC protein in Salmonella pathogenicity and suggests that the Gifsy-1 prophage carries one or more additional virulence genes that have a functional equivalent on the Gifsy-2 genome.

Identification of site-specific recombination genes int and xis of the Rhizobium temperate phage 16-3

Phage 16-3 is a temperate phage of Rhizobium meliloti 41 which integrates its genome with high efficiency into the host chromosome by site-specific recombination through DNA sequences of attB and attP. Here we report the identification of two phage-encoded genes required for recombinations at these sites: int (phage integration) and xis (prophage excision). We concluded that Int protein of phage 16-3 belongs to the integrase family of tyrosine recombinases. Despite similarities to the cognate systems of the lambdoid phages, the 16-3 int xis att system is not active in Escherichia coli, probably due to requirements for host factors that differ in Rhizobium meliloti and E. coli. The application of the 16-3 site-specific recombination system in biotechnology is discussed.

Site-specific recombination of temperate Myxococcus xanthus phage Mx8: regulation of integrase activity by reversible, covalent modification

Temperate Myxococcus xanthus phage Mx8 integrates into the attB locus of the M. xanthus genome. The phage attachment site, attP, is required in cis for integration and lies within the int (integrase) coding sequence. Site-specific integration of Mx8 alters the 3' end of int to generate the modified intX gene, which encodes a less active form of integrase with a different C terminus. The phage-encoded (Int) form of integrase promotes attP x attB recombination more efficiently than attR x attB, attL x attB, or attB x attB recombination. The attP and attB sites share a common core. Sequences flanking both sides of the attP core within the int gene are necessary for attP function. This information shows that the directionality of the integration reaction depends on arm sequences flanking both sides of the attP core. Expression of the uoi gene immediately upstream of int inhibits integrative (attP x attB) recombination, supporting the idea that uoi encodes the Mx8 excisionase. Integrase catalyzes a reaction that alters the primary sequence of its gene; the change in the primary amino acid sequence of Mx8 integrase resulting from the reaction that it catalyzes is a novel mechanism by which the reversible, covalent modification of an enzyme is used to regulate its specific activity. The lower specific activity of the prophage-encoded IntX integrase acts to limit excisive site-specific recombination in lysogens carrying a single Mx8 prophage, which are less immune to superinfection than lysogens carrying multiple, tandem prophages. Thus, this mechanism serves to regulate Mx8 site-specific recombination and superinfection immunity coordinately and thereby to preserve the integrity of the lysogenic state.

Site-specific recombination of temperate Myxococcus xanthus phage Mx8: genetic elements required for integration

Like most temperate bacteriophages, phage Mx8 integrates into a preferred locus on the genome of its host, Myxococcus xanthus, by a mechanism of site-specific recombination. The Mx8 int-attP genes required for integration map within a 2.2-kilobase-pair (kb) fragment of the phage genome. When this fragment is subcloned into a plasmid vector, it facilitates the site-specific integration of the plasmid into the 3' ends of either of two tandem tRNAAsp genes, trnD1 and trnD2, located within the attB locus of the M. xanthus genome. Although Int-mediated site-specific recombination occurs between attP and either attB1 (within trnD1) or attB2 (within trnD2), the attP x attB1 reaction is highly favored and often is accompanied by a deletion between attB1 and attB2. The int gene is the only Mx8 gene required in trans for attP x attB recombination. The int promoter lies within the 106-bp region immediately upstream of one of two alternate GTG start codons, GTG-5208 (GTG at bp 5208) and GTG-5085, for integrase and likely is repressed in the prophage state. All but the C-terminal 30 amino acid residues of the Int protein are required for its ability to mediate attP x attB recombination efficiently. The attP core lies within the int coding sequence, and the product of integration is a prophage in which the 3' end of int is replaced by host sequences. The prophage intX gene is predicted to encode an integrase with a different C terminus.

Site-specific integration of the conjugal Vibrio cholerae SXT element into prfC

Vibrio cholerae O139, the first non-O1 serogroup of V. cholerae to give rise to epidemic cholera, is characteristically resistant to the antibiotics sulphamethoxazole, trimethoprim, chloramphenicol and streptomycin. Resistances to these antibiotics are encoded by a 62 kb self-transmissible, conjugative, chromosomally integrating element designated the 'SXT element'. We found that the SXT element integrates site specifically into both V. cholerae and Escherichia coli K-12 into the 5' end of prfC, the gene encoding peptide chain release factor 3. Integration of the SXT element interrupts the chromosomal prfC gene, but the element encodes a new 5' end of prfC that restores the reading frame of this gene. The recombinant of prfC allele created upon element integration is functional. The integration and excision mechanism of the SXT element shares many features with site-specific recombination found in lambdoid phages. First, like lambda, the SXT element forms a circular extrachromosomal intermediate through specific recombination of the left and right ends of the integrated element. Second, chromosomal integration of the element occurs via site-specific recombination in a 17 bp sequence found in the circular form of the SXT element and a similar 17 bp sequence in prfC. Third, both chromosomal integration and excision of the SXT element were found to require an element-encoded int gene with strong similarities to the lambda integrase family. Based on the properties of the SXT element, we propose to classify this element as a CONSTIN, an acronym for a conjugative, self-transmissible, integrating element.

Tn4371: A modular structure encoding a phage-like integrase, a Pseudomonas-like catabolic pathway, and RP4/Ti-like transfer functions

Tn4371 is a 55-kb catabolic transposon originally isolated from Ralstonia eutropha A5 that encodes enzymes catalyzing the complete degradation of biphenyl. Unlike previously described transposons encoding similar genes for aromatic compound degradation. Tn4371 carries a phage-like degradation, Tn4371 integrase gene and RP4/Ti-like transfer genes. Tn4371 transposition involves an excision/integration process and, consistent with this site-specific recombination mechanism, the ends of the element are transiently covalently bound. Transposition is targeted to a limited number of sites on the CH34 chromosome and pMOL30 plasmid as well as on RP4. One of these sites consists of a 5'-TTTTTCAT-3' sequence which is also present between the covalently joined ends of the transposon. Conjugative transfer of Tn4371 could not yet be demonstrated although the functionality of its transfer machinery could be established through the identification of a second transposable element, Tn-bph, which contains the right half of Tn4371, including the bph catabolic gene cluster and the identified transfer genes. Tn-bph transfers by conjugation and integrates in a new host genome independently of the larger element. Tn4371 thus appears as composite transposon combining an enteric phage-like integration system, RP4/Ti-like conjugation genes, and Pseudomonas-like catabolic genes.

The site-specific recombination locus of mycobacteriophage Ms6 determines DNA integration at the tRNA(Ala) gene of Mycobacterium spp

Genetic determinants of the temperate mycobacteriophage Ms6 required for chromosomal integration were identified. DNA sequence analysis of an attP-containing fragment revealed an ORF encoding a protein of 372 amino acid residues with a C-terminus similar to other conserved C-terminal regions typical of the phage integrase family. Comparison of the sequences of attP, attB and bacteria-prophage junctions attL and attR showed a 26 bp common core sequence, where recombination takes place, near the 5' end of the integrase gene. Nucleotide sequence analysis of the attB chromosomal region showed that the core site overlaps the 3' end of the tRNA(Ala) gene. An integration-proficient plasmid vector was constructed and efficiently inserted at the tRNA(Ala) gene of Mycobacterium smegmatis, Mycobacterium vaccae, Mycobacterium bovis BCG and Mycobacterium tuberculosis H37Ra. It was demonstrated that Ms6 and D29 integrative systems can be used in conjunction for inserting genes at multiple loci. The site-specific integration system of mycobacteriophage Ms6 is a new tool for mycobacterial genetic analysis and is poorly related to those of the L5 bacteriophage family.

Lysogenic bacteriophage M1 from Selenomonas ruminantium: isolation, characterization and DNA sequence analysis of the integration site

Bacteriophage M1 from the ruminal bacterium Selenomonas ruminantium strain ML12 comprises a 30 nm icosahedral capsid, a 25 nm tail and 48 kb of linear dsDNA with cohesive ends. A restriction map of the phage genome has been constructed. The presence of bacteriophage M1 in the rumen has been demonstrated by PCR amplification and Southern blot analysis of DNA from rumen bacterial samples obtained from ten different sheep. Lysogeny was demonstrated by hybridization of M1 DNA to host chromosomal DNA and by identification and cloning of a 2.3 kb region of the phage containing the predicted attP domain which promotes chromosomal integration. DNA sequencing of the attP region demonstrated two major ORFs surrounding the predicted attP site and structural analysis of this region revealed a motif comprising three different inverted repeats surrounding a 12 bp palindrome. Analysis of the translated amino acid sequence upstream of the attP site demonstrated the presence of conserved residues found within integrase proteins of several temperate phages of different bacterial species.

Integration of the temperate phage phiU into the putative tRNA gene on the chromosome of its host Rhizobium leguminosarum biovar trifolii

The plasmid pCI6, carrying the attP site of the temperate phage phiU, integrates into the attB site on the chromosome of Rhizobium leguminosarum biovar trifolii strain 4S. The 4 kb EcoRI-HindIII region of pCI6 involved in site-specific integration was subcloned as the attP fragment of phage phiU and sequenced. The attL fragment, one of the new DNA junctions generated from the insertion of pCI6 into the chromosome of the host Rhizobium, was used as a hybridization probe for isolation of the attB fragment of strain 4S. The nucleotide sequence of the 2 kb PstI fragment of strain 4S, which hybridized with the attL fragment, was decided and compared with that of the attP fragment. A 53 bp common sequence was expected to be the core sequence of site-specific integration between phage phiU and strain 4S. One of the ORFs on the attP fragment, which was located adjacent to the core sequence, had structural homology to the integrase family. However, the attB fragment showed high homology with the tRNA genes of Agrobacterium tumefaciens and E. coli. A 47 bp sequence of the 53 bp core sequence overlapped with this tRNA-like sequence. This indicates that the target site of phage phiU integration is the putative tRNA gene on the chromosome of the Rhizobium host.

Genetic determinants of immunity and integration of temperate Myxococcus xanthus phage Mx8

An 8.1-kb fragment of the temperate Myxococcus xanthus phage Mx8 genome, when cloned into a plasmid vector, permits site-specific integration of the plasmid and confers superinfection immunity. Sequence analysis of a 9.5-kb region of Mx8 DNA containing this fragment reveals 19 densely packed open reading frames, four of which have predicted products with known or suspected activities. The Mx8 imm gene, required for superinfection immunity, has a sequence similar to that of Arabidopsis thaliana G-box-binding factor 1. Mx8 makes a DNA adenine methylase, Mox, and integrase, Int, related to other methylases and integrases. The int gene has two alternate translation initiation codons within the extensively overlapping uoi (upstream of int) gene. Comparison of the predicted product of the uoi gene with Salmonella phage P22 and Streptomyces plasmid Xis proteins shows that temperate phage excisionases may use variations of a helix-turn-helix motif to recognize specific DNA sequences.

Similarities and differences among 105 members of the Int family of site-specific recombinases

Alignments of 105 site-specific recombinases belonging to the Int family of proteins identified extended areas of similarity and three types of structural differences. In addition to the previously recognized conservation of the tetrad R-H-R-Y, located in boxes I and II, several newly identified sequence patches include charged amino acids that are highly conserved and a specific pattern of buried residues contributing to the overall protein fold. With some notable exceptions, unconserved regions correspond to loops in the crystal structures of the catalytic domains of lambda Int (Int c170) and HP1 Int (HPC) and of the recombinases XerD and Cre. Two structured regions also harbor some pronounced differences. The first comprises beta-sheets 4 and 5, alpha-helix D and the adjacent loop connecting it to alpha-helix E: two Ints of phages infecting thermophilic bacteria are missing this region altogether; the crystal structures of HPC, XerD and Cre reveal a lack of beta-sheets 4 and 5; Cre displays two additional beta-sheets following alpha-helix D; five recombinases carry large insertions. The second involves the catalytic tyrosine and is seen in a comparison of the four crystal structures. The yeast recombinases can theoretically be fitted to the Int fold, but the overall differences, involving changes in spacing as well as in motif structure, are more substantial than seen in most other proteins. The phenotypes of mutations compiled from several proteins are correlated with the available structural information and structure-function relationships are discussed. In addition, a few prokaryotic and eukaryotic enzymes with partial homology with the Int family of recombinases may be distantly related, either through divergent or convergent evolution. These include a restriction enzyme and a subgroup of eukaryotic RNA helicases (D-E-A-D proteins).

Purification and DNA-binding properties of the integrase protein Int encoded by Lactobacillus plantarum phage

The Lactobacillus plantarum temperate phage phi g1e (42,259 bp) encodes an integrase gene int linked to a phage attachment site attP (Kakikawa et al., 1997). To investigate phi g1e recombination, the integrase protein Int was overproduced in Escherichia coli under the T7 promoter, and purified. The Int protein had an apparent molecular mass of 42.0 kDa, corresponding well with that (45.5 kDa) predicted from the DNA sequence. Amino-acid sequencing revealed that the N-terminal 20 amino-acids of the purified Int protein completely coincided with those deduced from the DNA sequence, although deficient in the first methionine. Gel mobility-shift assays demonstrated that Int bound specifically to the attP region. In addition, footprinting analysis showed that Int protected about 35 bases, containing the 24-bp core domain at attP, from DNase I attack. These results are indicative of site-specific interaction of Int with the attP site, the reaction prerequisite for integration and excision of the phi g1e genome into and/or out of the host chromosome.

Sequence analysis and characterization of phi O1205, a temperate bacteriophage infecting Streptococcus thermophilus CNRZ1205

The complete nucleotide sequence of phi O1205, a temperate bacteriophage infecting Streptococcus thermophilus strain CNRZ1205, was determined. The phage genome has a unit length of 43,075 bp and appears to be packaged by the so-called headful mechanism. The genomic organization and structure of phi O1205 resemble those of several temperate lactococcal phages that display a life-cycle-specific organization, where ORFs believed to be involved in the lysogenic life-cycle are clustered and arranged in an orientation opposite to the ORFs supposedly involved in the lytic life-cycle. Database searches revealed putative functions for several identified ORFs and further indicated that phi O1205 is genetically related to a particular group of lactococcal phages. Three genes encoding the major structural proteins were identified on the phi O1205 genome. The phage attachment site attP, the bacterial attachment site attB, and the two phage/chromosome junctions attL and attR were identified and found to contain a 40 bp common core sequence.

Molecular characterization of the genes involved in O-antigen modification, attachment, integration and excision in Shigella flexneri bacteriophage SfV

Bacteriophage SfV is a temperate phage of Shigella flexneri responsible for converting serotype Y (3,4) to serotype 5a (V; 3,4) through its glucosyl transferase gene. The glucosyl transferase (gtr) gene of SfV has been cloned and shown to partially convert S. flexneri serotype Y to serotype 5a. In this study, we found that the serotype-converting region of SfV was approximately 2.5 kb in length containing three continuous ORFs. The recombinant strain carrying the three complete ORFs expressed the type V and group antigen 3,4, both indistinguishable from that of S. flexneri 5a wild-type strain. The interruption of orf5 or orf6 gave partial conversion in the S. flexneri recombinant strain indicated by the incomplete replacement of group antigen 3,4. The region adjacent to the serotype-conversion genes was found to be identical to the attP-int-xis region of phage P22. Altogether, an approximately 2.2-kb sequence covering a portion of the serotype-conversion (approximately 500 nt)-attP-int-xis regions of SfV was remarkably similar to that of P22.

Characterization of the genes encoding integrative and excisive functions of Lactobacillus phage øg1e: cloning, sequence analysis, and expression in Escherichia coli

øg1e is a temperate phage of the Lactobacillus strain G1e. The phage-host junctions attR and attL cloned from the lysogen have a 24-bp common (core) sequence implicated in recombination. DNA sequencing analysis of a 5.2-kbp SacI fragment of the øg1e phage genome (42.5 kbp) revealed two possible open reading frames (ORF), xis and int, and the phage attachment (recombination) site (attP), whose 24-bp sequence is identical to the core sequence detected in attR and attL. The deduced int product (Int) is a basic protein of 391 amino acids with an estimated pI of 9.70, and significantly resembles other presumed integrases encoded by the Lactobacillus and Lactococcus phages including øadh and øLC3, as well as the Escherichia coli phages such as lambda. The predicted øg1e xis protein (Xis) is small and very acidic (66 amino acids; pI 4.55), and shows a resemblance (32% overall identity) with a putative excisionase encoded by the Staphylococcus phage ø11. The øg1e Int with a deduced molecular mass of 45.5 kDa was overproduced in E. coli cells, and electrophoretically analyzed.

Genetic characterization of site-specific integration functions of phi AAU2 infecting "Arthrobacter aureus" C70

All the essential genetic determinants for site-specific integration of corynephage phi AAU2 are contained within a 1,756-bp DNA fragment, carried on the integrative plasmid p5510, and are shown to be functional in Escherichia coli. One open reading frame, ORF4, encoding a protein of 266 amino acids was shown to represent the phi AAU2 integrase. The nucleotide sequence of the phi AAU2 attachment site, attP, and the attB, attL, and attR sequences in the host "Arthrobacter aureus" C70 were determined. Identical nucleotide sequences were shown to be responsible for the integration of p5510 in the chromosomes of Corynebacterium glutamicum, Brevibacterium divaricatum, and B. lactofermentum, and a sequence almost identical to attB was found to be present in these three strains. In contrast to other phage site-specific recombination systems, a plasmid encompassing only int-attP failed to integrate into the host chromosome. This led to the identification of an 800-bp noncoding region, immediately upstream of int, absolutely required for site-specific integration of p5510.

Circuit simulation of genetic networks

Genetic networks with tens to hundreds of genes are difficult to analyze with currently available techniques. Because of the many parallels in the function of these biochemically based genetic circuits and electrical circuits, a hybrid modeling approach is proposed that integrates conventional biochemical kinetic modeling within the framework of a circuit simulation. The circuit diagram of the bacteriophage lambda lysislysogeny decision circuit represents connectivity in signal paths of the biochemical components. A key feature of the lambda genetic circuit is that operons function as active integrated logic components and introduce signal time delays essential for the in vivo behavior of phage lambda.

Analysis of the mechanism of DNA recombination using tangles

The DNA of all organisms has a complex and essential topology. The three topological properties of naturally occurring DNA are supercoiling, catenation, and knotting. Although these properties are denned rigorously only for closed circular DNA, even linear DNAin vivocan have topological properties because it is divided into topologically separate subdomains (Drlica 1987; Roberge & Gasser, 1992). The essentiality of topological properties is demonstrated by the lethal consequence of interfering with topoisomerases, the enzymes that regulate the level of DNA supercoiling and that unlink DNA during its replication (reviewed in Wang, 1991; Bjornsti, 1991; Drlica, 1992; Ullspergeret al. 1995).

Novel organization of the site-specific integration and excision recombination functions of the Staphylococcus aureus serotype F virulence-converting phages phi 13 and phi 42

Functions required for site-specific integration and excision of the Staphylococcus aureus serotype F virulence-converting phages phi 13 and phi 42 were localized and characterized. Like other temperate phages, integration of phi 13 and phi 42 sequences was found to require the product of an int gene located close to the phage attP site. Both int genes are almost identical, express proteins possessing characteristic features of the Int (integrase) family of recombinases, but share very little homology with previously described int genes, including those of the serotype B S. aureus phages L54a and phi 11. Nevertheless, all four S. aureus phages share an almost identical short sequence located immediately 5' to these distinct int genes, suggesting a common mechanism of int gene regulation. Upstream from these common sequences, the sequences of phi 13 and phi 42 are quite distinct from each other, and from the corresponding regions of phi 11 and L54a which encode the Xis proteins that are required with Int to mediate site-specific excision of the latter phages. Surprisingly, phi 13 and phi 42 sequences encompassing the attP sites and int genes, but lacking either an adjacent or more distant phage excision protein gene, were sufficient to mediate site-specific excision of integrated phage DNA sequences.

Characterization of an open reading frame involved in site-specific integration of filamentous phage Cf1t from Xanthomonas campestris pv. citri

Cf1t is a single-stranded DNA filamentous phage; a 1.9-kb segment of DNA from Cf1t was found to be responsible for site-specific integration into Xanthomonas campestris pv. citri (XW47), in the absence of any Xanthomonas origin of replication. Deletion analysis and introduction of amber stop codons into this fragment from Cf1t revealed an open reading frame (ORF344) which was involved in the integration function. The predicted amino-acid sequence of ORF344 bears no homology with conserved sequences of the integrase family.

Molecular genetics of the chloramphenicol-resistance transposon Tn4451 from Clostridium perfringens: the TnpX site-specific recombinase excises a circular transposon molecule

The chloramphenicol-resistance transposon Tn4451 undergoes precise conjugative deletion from its parent plasmid plP401 in Clostridium perfringens and precise spontaneous excision from multicopy plasmids in Escherichia coli. The complete nucleotide sequence of the 6338 bp transposon was determined and it was found to encode six genes. Genetic analysis demonstrated that the largest Tn4451-encoded gene, tnpX, was required for the spontaneous excision of the transposon in both E. coli and C. perfringens, since a Tn4451 derivative that lacked a functional tnpX gene was completely stable in both organisms. Because the ability of this derivative to excise was restored by providing the tnpX gene on a compatible plasmid, it was concluded that this gene encoded a trans-acting site-specific recombinase. Allelic exchange was used to introduce the tnpX delta 1 allele onto plP401 and it was shown that TnpX was also required for the conjugative excision of Tn4451 in C. perfringens. It was also shown by hybridization and polymerase chain reaction (PCR) studies that TnpX-mediated transposon excision resulted in the formation of a circular form of the transposon. The TnpX recombinase was unique because it potentially contained the motifs of two independent site-specific recombinase families, namely the resolvase/invertase and integrase families. Sequence analysis indicated that the resolvase/invertase domain of TnpX was likely to be involved in the excision process by catalysing the formation of a 2 bp staggered nick on either side of the GA dinucleotide located at the ends of the transposon and at the junction of the circular form. The other Tn4451-encoded genes include tnpZ, which appears to encode a second potential site-specific recombinase. This protein has similarity to plasmid-encoded Mob/Pre proteins, which are involved in plasmid mobilization and multimer formation. Located upstream of the tnpZ gene was a region with similarity to the site of interaction of these mobilization proteins.

Characterization of genetic elements required for site-specific integration of Lactobacillus delbrueckii subsp. bulgaricus bacteriophage mv4 and construction of an integration-proficient vector for Lactobacillus plantarum

Temperate phage mv4 integrates its DNA into the chromosome of Lactobacillus delbrueckii subsp. bulgaricus strains via site-specific recombination. Nucleotide sequencing of a 2.2-kb attP-containing phage fragment revealed the presence of four open reading frames. The larger open reading frame, close to the attP site, encoded a 427-amino-acid polypeptide with similarity in its C-terminal domain to site-specific recombinases of the integrase family. Comparison of the sequences of attP, bacterial attachment site attB, and host-phage junctions attL and attR identified a 17-bp common core sequence, where strand exchange occurs during recombination. Analysis of the attB sequence indicated that the core region overlaps the 3' end of a tRNA(Ser) gene. Phage mv4 DNA integration into the tRNA(Ser) gene preserved an intact tRNA(Ser) gene at the attL site. An integration vector based on the mv4 attP site and int gene was constructed. This vector transforms a heterologous host, L. plantarum, through site-specific integration into the tRNA(Ser) gene of the genome and will be useful for development of an efficient integration system for a number of additional bacterial species in which an identical tRNA gene is present.

The actinophage RP3 DNA integrates site-specifically into the putative tRNA(Arg)(AGG) gene of Streptomyces rimosus

The temperate actinophage RP3 integrates site-specifically into the chromosome of Streptomyces rimosus R6-554. The phage attachment site attP and the hybrid attachment sites of the integrated prophage--attL and attR--were cloned and sequenced. The 54nt core sequence, common to all RP3 related attachment sites, comprises the 3' terminal end of a putative tRNA(Arg)(AGG) gene. AttB bears the complete tRNA gene which is restored in attL after integration. A 7.5kb HindIII fragment, bearing attP, was used to construct an integrative plasmid to simulate the integration process in vivo and to localize the phage genes necessary for site specific integration. The int and xis genes were sequenced and compared to other recombination genes.

Identification of an HP1 phage protein required for site-specific excision

Transposon insertion mutagenesis and transformation were used to locate genes responsible for excision in the temperature phage HP1 of Haemophilus influenzae. A 6.5 kb segment of DNA near the left end of the phage genome was sequenced, and 11 new open reading frames were identified. Two face-to-face overlapping promoter sequences organized these open reading frames into two operons transcribed in opposite directions. Interruption of the first open reading frame in the rightward operon created lysogens unable to produce phages. Provision of the uninterrupted open reading frame in trans restored phage production. The gene identified by this procedure, cox, was cloned and the protein product was expressed at high levels in Escherichia coli. The Cox protein is a 79-residue basic protein with a predicted strong helix-turn-helix DNA-binding motif. Extracts induced to express high levels of Cox contained a 9 kDa protein. These extracts inhibited integrative recombination and were required for excisive recombination mediated by HP1 integrase. The HP1 cox gene location is similar to that of the homologous excisive and regulatory genes from coliphages P2 and 186. These phages appear to share a distinctive organization of recombination proteins and transcriptional domains differing markedly from phage lambda and its relatives.

Identification and characterization of genes involved in excision of the Lactococcus lactis conjugative transposon Tn5276

The 70-kb transposon Tn5276, originally detected in Lactococcus lactis NIZO R5 and carrying the genes for nisin production and sucrose fermentation, can be conjugally transferred to other L. lactis strains. Sequence analysis and complementation studies showed that the right end of Tn5276 contains two genes, designated xis and int, which are involved in excision. The 379-amino-acid int gene product shows high (up to 50%) similarity with various integrases, including that of the Tn916-related conjugative transposons. The xis gene product, like almost all known excisionase (Xis) proteins, is a small (68-residue), basic protein. Expression of both the Tn5276 int and xis genes is required for efficient excision of the ends of Tn5276 in Escherichia coli that appeared to be circularized in the excision process. Mutational analysis of the xis and int genes showed that excision efficiency is dependent on the integrity of the int gene but that an intact xis gene is also required for efficient excision.

Characterization of the lactococcal temperate phage TP901-1 and its site-specific integration

The temperate lactococcal phage TP901-1, induced by UV light from Lactococcus lactis subsp. cremoris 901-1, was characterized. The restriction map was found to be circular, and the packaging of TP901-1 DNA was concluded to occur by a headful mechanism. The pac region was localized on the 38.4-kb phage genome. TP901-1 belongs to the class of P335 phages (V. Braun, S. Hertwig, H. Neve, A. Geis, and M. Teuber, J. Gen. Microbiol. 135:2551-2560, 1989). Evidence is presented that the phages TP936-1 (V. Braun, S. Hertwig, H. Neve, A. Geis, and M. Teuber, J. Gen. Microbiol. 135:2551-2560, 1989) and C3-T1 (A. W. Jarvis, V. R. Parker, and M. B. Bianchin, Can. J. Microbiol. 38:398-404, 1992) are very closely related to or are identical to TP901-1. The lytically propagated TP901-1 phages were able to lysogenize both indicator strains Lactococcus cremoris 3107 and Wg2. Lysogenization resulted in site-specific integration of the phage genome into the bacterial chromosome. Only one chromosomal attB site was found in 20 independent lysogens. The attP region of TP901-1 and the attL and attR regions were cloned and sequenced. The results showed a core region of only 5 bp, in which the recombination occurs, followed after a 1-bp mismatch by a 7-bp identical region, TCAAT(T/C)AAGGTAA. This result was further verified by sequencing of the attB region obtained by PCR. An integration vector was constructed with the 6.5-kb EcoRI fragment from TP901-1 containing attP. This vector also functions in the plasmid-free strains, MG1363 and LM0230 with only one specific attB site, strongly indicating a more general use of the TP901-1-based integration vector in lactococci.

Genetic selection for mutations that impair the co-operative binding of lambda repressor

Bacteriophage lambda repressor binds co-operatively to adjacent pairs of DNA target sites. A novel combination of positive genetic selections, involving two different operon fusions derived from P22 challenge phages, was used to isolate mutant lambda repressors that have lost the ability to bind co-operatively to tandem sites yet retain the ability to bind a strong, single site. These cb (co-operative binding) mutations result in 10 different amino acid changes, which define eight residues in the carboxyl-terminus of repressor. Because challenge phage derivatives may be applied to study essentially any specific protein-DNA interaction, analogous combinations of genetic selections may be used to explore the ways that a variety of proteins interact to assemble regulatory complexes.

The int genes of bacteriophages P22 and lambda are regulated by different mechanisms

Bacteriophage P22 and lambda are related bacteriophages with similar gene organizations. In lambda the cll-dependent Pl promoter is responsible for lambda int gene expression. The only apparent counterpart to pl in P22 is oriented in the opposite direction, and cannot transcribe the P22 int gene. We show that this promoter, called P(al), is active both in vivo and in vitro, and is dependent upon the P22 cll-like gene, called c1. We have also determined the DNA sequence of a 3.3 kb segment that closes the gap between previously reported sequences to give a continuous sequence between the P22 pL promoter and the int gene. The newly determined sequence is densely packed with genes from the pL direction, and the proteins predicted by the sequence show excellent correlation with the proteins mapped by Youderian and Susskind in 1980. However, the sequence contains no apparent genes in the opposite (p(al)) direction, and no additional binding motifs for the P22 c1 protein. We conclude that int gene expression in P22 is regulated by a different mechanism than in lambda.

Localization and nucleotide sequences of genes mediating site-specific recombination of the SLP1 element in Streptomyces lividans

SLP1 is a 17.2-kbp genetic element indigenous to the Streptomyces coelicolor chromosome. During conjugation, SLP1 can undergo excision and subsequent site-specific integration into the chromosomes of recipient cells. We report here the localization, nucleotide sequences, and initial characterization of the genes mediating these recombination events. A region of SLP1 adjacent to the previously identified site of integration, attP, was found to be sufficient to promote site-specific integration of an unrelated Streptomyces plasmid. Nucleotide sequence analysis of a 2.2-kb segment of this region reveals two open reading frames that are adjacent to and transcribed toward the attP site. One of these, the 1,365-bp int gene of SLP1, encodes a predicted 50.6-kDa basic protein having substantial amino acid sequence similarity to a family of site-specific recombinases that includes the Escherichia coli bacteriophage lambda integrase. A linker insertion in the 5' end of the cloned int gene prevents integration, indicating that Int is essential for promoting integration. An open reading frame (orf61) lying immediately 5' to int encodes a predicted 7.1-kDa basic peptide showing limited sequence similarity to the excisionase (xis) genes of other site-specific recombination systems.

Excisive recombination of the SLP1 element in Streptomyces lividans is mediated by Int and enhanced by Xis

The functions mediating site-specific recombination of the SLP1 element have been mapped to a 2.2-kb region that includes the site of integration (attP), a gene (int) that specifies a function both necessary and sufficient for integration of SLP1, and an open reading frame, orf61, suspected of encoding a protein, Xis, that shows limited similarity to the excisionases of other site-specific recombination systems. Here we describe experiments that investigate the respective roles of orf61 and int in the excision of SLP1. We constructed derivatives of the high-copy-number Streptomyces plasmid pIJ101 that express orf61, int, or both orf61 and int from transcriptional fusions to the Tn5 aph gene and tested the ability of these constructs to promote excision of an adventitious attP-containing plasmid that had been integrated site-specifically into the attB site of the Streptomyces lividans chromosome. Expression of the int gene product alone from an exogenous promoter was sufficient for excision of the integrated plasmid. This result indicates that the SLP1 int-encoded protein can carry out excisive, as well as integrative, recombination. The orf61 gene product, when expressed from an exogenous promoter, inhibited int-mediated integration at the chromosomal attB site. Moreover, under conditions in which excision and transfer normally occur, precise excision of SLP1 was enhanced by the orf61-encoded protein. On the basis of these findings, we here designate the orf61 gene as xis.

Genetic organization and sequence of the region encoding integrative functions from Lactobacillus gasseri temperate bacteriophage phi adh

A 2.0-kb fragment from the Lactobacillus gasseri temperate bacteriophage phi adh contained the essential genetic determinants for site-specific integration. The nucleotide sequence of this fragment was determined. An open reading frame (intG), which adjoined the phage attachment site (attP), encoded a deduced protein related to the integrase family. The organization of this region was comparable to other phage site-specific recombination systems.

Characterization of genetic elements required for site-specific integration of the temperate lactococcal bacteriophage phi LC3 and construction of integration-negative phi LC3 mutants

The genetic elements required for the integration of the temperate lactococcal bacteriophage phi LC3 into the chromosome of its bacterial host, Lactococcus lactis subsp. cremoris, were identified and characterized. The phi LC3 phage attachment site, attP, was mapped and sequenced. DNA sequence analysis of attP and of the bacterial attachment site, attB, as well as the two phage-host junctions, attR and attL, in the chromosome of a phi LC3 lysogen, identified a 9-bp common core region, 5'-TTCTTCATG'-3, within which the strand exchange reaction takes place during integration. The attB core sequence is located within the C-terminal part of an open reading frame of unknown function. The phi LC3 integrase gene (int), encoding the phi LC3 site-specific recombinase, was identified and is located adjacent to attP. The phi LC3 Int protein, as deduced from the nucleotide sequence, is a basic protein of 374 amino acids that shares significant sequence similarity with other site-specific recombinases of the integrase family. Phage phi LC3 int- and int-attP-defective mutants, conferring an abortive lysogenic phenotype, were constructed.

The oac gene encoding a lipopolysaccharide O-antigen acetylase maps adjacent to the integrase-encoding gene on the genome of Shigella flexneri bacteriophage Sf6

Lysogens of Shigella flexneri harbouring the temperate bacteriophage, Sf6, have been previously shown to undergo a serotype conversion due to O-acetylation of the O-antigen of the lipopolysaccharide. A partial physical map of the phage genome has been constructed. Analysis of the phage DNA suggests that the phage packages by a headful mechanism and that the mature DNA molecules are terminally redundant. Cloning of the PstI fragments of Sf6 enabled the region encoding the serotype conversion to be localized, showing that this was clearly phage-encoded. The gene was further localized by mutagenesis with Tn5 and the nucleotide sequence of the entire 2693-bp PstI fragment was determined. Two major open reading frames (ORFs) were found capable of encoding proteins of 44.1 and 37.2 kDa. The latter corresponds to the O-antigen acetylase and its gene has been designated oac. The oac gene is capable of converting Sh. flexneri serotypes X, Y, 1a and 4a to 3a, 3b, 1b and 4b, respectively. The Oac protein bears a high degree of homology to the NodX protein of Rhizobium leguminosarum suggesting that it, too, may be a sugar acetylase. The second ORF immediately upstream from oac corresponds to the bacteriophage Sf6 integrase responsible for chromosomal integration and is highly homologous to the integrases of Escherichia coli bacteriophages P4 and phi 80, but less closely related to those of P1, P2, P22, 186 and lambda.

Insertional inactivation of the Staphylococcus aureus beta-toxin by bacteriophage phi 13 occurs by site- and orientation-specific integration of the phi 13 genome

Lysogenization of Staphylococcus aureus by the serotype F converting bacteriophage phi 13 results in loss of beta-toxin expression. Sequence analysis of the S. aureus beta-toxin gene (hlb), the attachment site (attP)-containing region of phi 13 DNA and the chromosome/bacteriophage DNA junctions of a phi 13 lysogen, revealed that the molecular mechanism of loss of beta-toxin expression was due to insertion of the phi 13 genome into the 5' end of hlb. The insertion site (attB) within hlb contained a 14 base pair core sequence in common with attP and both ends of the integrated linear prophage genome of a phi 13 lysogen. These findings indicate that integration of the phi 13 genome into hlb is site- and orientation-specific.

Interactions between lambda Int molecules bound to sites in the region of strand exchange are required for efficient Holliday junction resolution

lambda Site-specific recombination proceeds via two sequential single-strand exchanges that first generate and then resolve a Holliday recombination intermediate. The resolution of artificial Holliday junctions (chi-forms) is well suited to studying the mechanisms involved in reciprocal strand exchange because the linear products of this reaction are stable and easily quantitated. To study the interactions between Int molecules bound at the sites of strand exchange, artificial Holliday junctions containing only the seven base-pair overlap region and the four core-type Int binding sites were used as a model system. In vitro resolution of these structures yields products of both top- and bottom-strand exchange. An abortive product resulting from simultaneous cleavage of the top and bottom strands also occurs at low frequency. Inactivation of one of the four Int binding sites by multiple base substitutions does not significantly affect the efficiency of resolution but has a dramatic effect on the directionality, i.e. the choice of top- or bottom-strand exchange. When any two of the four core-type sites are similarly inactivated, strand exchange is very inefficient and the amount of aberrant cleavage is somewhat greater than for the Holliday junction with four intact Int binding sites. Analysis of the resolution products of Holliday junctions with various combinations of defective Int binding sites leads to the following conclusions: (1) three functional core-type Int binding sites are necessary and sufficient for a strand exchange; (2) the Int molecules that are partners in a strand exchange interact with Int bound to a "cross-core" site that is not directly involved in carrying out the reaction; (3) Int molecules bound to the core-type sites interact in a way that reduces the occurrence of abortive double-strand cleavage events.

Interaction of integration host factor from Escherichia coli with the integration region of the Haemophilus influenzae bacteriophage HP1

The specific DNA-binding protein integration host factor (IHF) of Escherichia coli stimulates the site-specific recombination reaction between the attP site of bacteriophage HP1 and the attB site of its host, Haemophilus influenzae, in vitro and also appears to regulate the expression of HP1 integrase. IHF interacts specifically with DNA segments containing the att sites and the integrase regulatory region, as judged by IHF-dependent retardation of relevant DNA fragments during gel electrophoresis. The locations of the protein-binding sites were identified by DNase I protection experiments. Three sites in the HP1 attP region bound IHF, two binding sites were present in the vicinity of the attB region, and one region containing three partially overlapping sites was present in the HP1 integrase regulatory segment. The binding sites defined in these experiments all contained sequences which matched the consensus IHF binding sequences first identified in the lambda attP region. An activity which stimulated the HP1 site-specific integration reaction was found in extracts of H. influenzae, suggesting that an IHF-like protein is present in this organism.

Sequence analysis and comparison of int and xis genes from staphylococcal bacteriophages L54a and phi 11

The DNA fragment encoding the integrase and excisionase genes involved in site-specific recombination of staphylococcal bacteriophage phi 11 was cloned and sequenced. The int and xis genes and the recombination site, attP, were highly clustered in a 1.7-kilobase DNA fragment with the gene order attP-int-xis. The int and xis genes were transcribed divergently, with the int gene transcribed toward the attp site and the xis gene transcribed away from the attP site. The deduced Int is a basic protein of 348 residues with an estimated molecular weight of 41,357. In contrast, the deduced Xis is an acidic protein containing 66 amino acids with an estimated molecular weight of 7,621. The site-specific recombination system of phi 11 was compared with that of a closely related bacteriophage, L54a.

Characterization of the genetic elements required for site-specific integration of plasmid pSE211 in Saccharopolyspora erythraea

The 18.1-kilobase plasmid pSE211 integrates into the chromosome of Saccharopolyspora erythraea at a specific attB site. Restriction analysis of the integrated plasmid, pSE211int, and adjacent chromosomal sequences allowed identification of attP, the plasmid attachment site. Nucleotide sequencing of attP, attB, attL, and attR revealed a 57-base-pair sequence common to all sites with no duplications of adjacent plasmid or chromosomal sequences in the integrated state, indicating that integration takes place through conservative, reciprocal strand exchange. An analysis of the sequences indicated the presence of a putative gene for Phe-tRNA at attB which is preserved at attL after integration has occurred. A comparison of the attB site for a number of actinomycete plasmids is presented. Integration at attB was also observed when a 2.4-kilobase segment of pSE211 containing attP and the adjacent plasmid sequence was used to transform a pSE211- host. Nucleotide sequencing of this segment revealed the presence of two complete open reading frames (ORFs) and a segment of a third ORF. The ORF adjacent to attP encodes a putative polypeptide 437 amino acids in length that shows similarity, at its C-terminal domain, to sequences of site-specific recombinases of the integrase family. The adjacent ORF encodes a putative 98-amino-acid basic polypeptide that contains a helix-turn-helix motif at its N terminus which corresponds to domains in the Xis proteins of a number of bacteriophages. A proposal for the function of this polypeptide is presented. The deduced amino acid sequence of the third ORF did not reveal similarities to polypeptide sequences in the current data banks.

Nucleotide sequence of the region encompassing the int gene of a cryptic prophage and the dna Y gene flanked by a curved DNA sequence of Escherichia coli K12

The nucleotide sequence of a 2.5 kb region encompassing a curved DNA segment (BENT-9) randomly cloned from the total Escherichia coli chromosome was determined. This region was found to contain the dna Y gene encoding a transfer RNA. The curved DNA structure was demonstrated to be located just upstream of the dna Y promoter. The results of sequencing further revealed that the int gene of a cryptic prophage, qsr', which has been shown to be present in the E. coli genome, is located next to the dna Y gene.

A novel family of potentially mobile DNA elements encoding site-specific gene-integration functions: integrons

A family of novel mobile DNA elements is described, examples of which are found at several independent locations and encode a variety of antibiotic resistance genes. The complete elements consist of two conserved segments separated by a segment of variable length and sequence which includes inserted antibiotic resistance genes. The conserved segment located 3' to the inserted resistance genes was sequenced from Tn21 and R46, and the sequences are identical over a region of 2026 bases, which includes the sulphonamide resistance gene sull, and two further open reading frames of unknown function. The complete sequences of both the 3' and 5' conserved regions of the DNA element have been determined. A 59-base sequence element, found at the junctions of inserted DNA sequences and the conserved 3' segment, is also present at this location in the R46 sequence. A copy of one half of this 59-base element is found at the end of the sull gene, suggesting that sull, though part of the conserved region, was also originally inserted into an ancestral element by site-specific integration. Inverted or direct terminal repeats or short target site duplications, both of which are characteristics of class I and class II transposons, are not found at the outer boundaries of the elements described here. Furthermore, the conserved regions do not encode any proteins related to known transposition proteins, except the DNA integrase encoded by the 5' conserved region which is implicated in the gene insertion process. Mobilization of this element has not been observed experimentally; mobility is implied from the identification of the element in at least four independent locations, in Tn21, R46 (IncN), R388 (IncW) and Tn1696. The definitive features of these novel elements are (i) that they include site-specific integration functions (the integrase and the insertion site); (ii) that they are able to acquire various gene units and act as an expression cassette by supplying the promoter for the inserted genes. As a consequence of acquiring different inserted genes, the element exists in a variety of forms which differ in the number and nature of the inserted genes. This family of elements appears formally distinct from other known mobile DNA elements and we propose the name DNA integration elements, or integrons.