SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. Supplementary material
  8. References
  9. Supporting Information

Summary The Yersinia high-pathogenicity island (HPI) encodes the siderophore yersiniabactin-mediated iron uptake system. The HPI of Yersinia pseudotuberculosis I has previously been shown to be able to excise precisely from the bacterial chromosome by recombination between the attB-R and attB-L sites flanking the island. However, the nature of the Y. pseudotuberculosis HPI excision machinery remained unknown. We show here that, upon excision, the HPI forms an episomal circular molecule. The island thus has the ability to excise from the chromosome, circularize and reintegrate itself, either in the same location or in another asn tRNA copy. We also demonstrate that the HPI-encoded bacteriophage P4-like integrase (Int) plays a critical role in HPI excision and that, like phage integrases, it acts as a site-specific recombinase that catalyses both excision and integration reactions. However, Int alone cannot efficiently promote recombination between the attB-R and attB-L sites, and we demonstrate that a newly identified HPI-borne factor, designated Hef (for HPI excision factor) is also required for this activity. Hef belongs to a family of recombination directionality factors. Like the other members of this family, Hef probably plays an architectural rather than a catalytic role and promotes HPI excision from the chromosome by driving the function of Int towards an excisionase activity. The fact that the HPI, and probably several other pathogenicity islands, carry a machinery of integration/excision highly similar to those of bacteriophages argues for a phage-mediated acquisition and transfer of these elements.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. Supplementary material
  8. References
  9. Supporting Information

The genus Yersinia is composed of 11 species that can be divided into pathogenic and non-pathogenic groups. Pathogenic strains may be further separated into two levels of pathogenicity: low pathogenicity (Yersinia enterocolitica biotypes 2–5) and high pathogenicity (Yersinia pestis, Yersinia pseudotuberculosis and Y. enterocolitica 1B). Strains with a moderate level of pathogenicity are responsible for mild enteric infections in humans and are not lethal for mice at low doses (Cornelis, 1987). In contrast, high-pathogenicity strains cause disseminated infections in humans and are lethal for mice at low doses.

A pathogenicity island (PAI) was found to be specifically associated with the latter group (De Almeida et al., 1993) and was subsequently termed high-pathogenicity island (HPI) (Carniel et al., 1996). A direct role for this genetic element in Yersinia pathogenesis has been demonstrated by showing that deletion of some HPI-borne genes resulted in decreased bacterial virulence (Carniel et al., 1992; Rakin et al., 1995; Bearden et al., 1997). Moreover, the transfer of the functional part of the HPI to a low-pathogenicity Y. enterocolitica strain of biotype 2 was sufficient to increase the virulence of the host strain for mice, suggesting that the presence of this island may be sufficient to differentiate high- from low-pathogenicity strains (Pelludat et al., 2002). Thus, although the HPI might not be the sole factor accounting for the high level of virulence in yersiniae, it clearly contributes to this phenotype.

The HPI confers to the bacteria the ability to capture iron bound to eukaryotic molecules via the synthesis of a siderophore called yersiniabactin (Heesemann, 1987; Gehring et al., 1998; Perry et al., 1999). The island is composed of two functionally distinct parts (Fig. 1A). The left-hand part is variable in size and is not conserved in the three highly pathogenic species, while the ≈ 30 kb right-hand part, termed the yersiniabactin locus, is highly conserved and corresponds to the functional part of the HPI (yersiniabactin biosynthesis, transport and regulation). All criteria defining a pathogenicity island are fulfilled by the HPI: it is a large chromosomal DNA fragment that carries genes important for virulence, as well as mobility loci [insertion sequence (IS) elements and a P4-like integrase gene], it is bordered on one side by an asn tRNA locus and is flanked by two 17 bp repeats (attB-R and attB-L), its G+C content (56%) is much higher than that of the core genome (47%), and its distribution is restricted to a subset of species and strains within the genus (Fetherston and Perry, 1994; Carniel et al., 1996; Buchrieser et al., 1998; Rakin et al., 1999).

image

Figure 1. Genetic map of the Y. pseudotuberculosis HPI indicating the location of some of the primers used in this study and the gene inactivations. A. Genetic map of the Y. pseudotuberculosis HPI. B. Insertion of pPsnSac (in grey) into the HPI-borne psn locus (in black), following a single cross-over event. C. Mutagenesis of int by replacing most of its coding sequence with a kan cassette. D. Mutagenesis of hef by replacing part of its coding sequence with a kan cassette. Arrows below the maps represent the location and direction of the primers used to look for the presence of the HPI, to check the mutants or to clone the genes.0

Download figure to PowerPoint

PAIs are considered to be mobile ‘foreign’ genetic elements that have been acquired by horizontal transfer. Integration of the HPI into any asn tRNA site on the host chromosome requires only the presence of an attP site and a functional int gene (Rakin et al., 2001). The mechanisms of HPI excision remained to be elucidated. Some PAIs have kept the capacity to excise precisely from the chromosome (Hacker et al., 1990; Blum et al., 1994; Buchrieser et al., 1998; Ruzin et al., 2001; Turner et al., 2001). This is not the case for the Y. enterocolitica 1B HPI, which has undergone a process of genetic stabilization in the host chromosome (Bach et al., 1999). In Y. pestis, excision of the HPI is not precise but occurs at high frequencies (2 × 10−3), as part of a much larger chromosomal deletion of a 102 kb region designated ‘pgm locus’ (Fetherston et al., 1992; Hare et al., 1999). This deletion results from homologous recombination between two IS100 copies flanking the pgm locus (Fetherston et al., 1992; 1999; Fetherston and Perry, 1994). Until now, the only Yersinia species in which the HPI has been shown to excise precisely from the chromosome is Y. pseudotuberculosis (Buchrieser et al., 1998). Excision of this HPI occurs at a rate of ≈ 10−4 by recombination between the two flanking attB sites homologous to the attP site of bacteriophage P4 (Buchrieser et al., 1998; Rakin et al., 2001).

The aim of this study was to get better insights into the excision machinery of the Y. pseudotuberculosis HPI. For this purpose, the HPI excision capacity was investigated in different strains of Y. pseudotuberculosis and, in those with an HPI that was non-excisable, the mechanisms responsible for the lack of HPI excision were identified. We show here that the HPI forms a circular episomal molecule upon excision from the bacterial chromosome. A functional integrase as well as a novel HPI-encoded factor, designated Hef, are essential for HPI excision. We provide evidence that Hef, a new member of the family of recombination directionality factors, plays a critical role in the HPI excision process, probably by driving the function of Int towards an excisionase activity.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. Supplementary material
  8. References
  9. Supporting Information

The HPI forms a circular episome upon excision from the chromosome

We have shown previously that the HPI of Y. pseudotuberculosis IP32637 has the ability to excise precisely from the chromosome (Buchrieser et al., 1998). Circularization upon excision has been described recently for two PAIs: the SaPIbov2 of Staphylococcus aureus (Ubeda et al., 2003) and the VPI of Vibrio cholerae (Rajanna et al., 2003), as well as for the ICE-like element carrying the HPI in Escherichia coli ECOR31 (Schubert et al., 2004). In order to determine whether the excised Y. pseudotuberculosis HPI could also form a circular episome, outward primers located at the extremities of the island (primers 144A/143A in Fig. 1A) were used to search for a polymerase chain reaction (PCR) product generated upon circularization of the HPI. No product was observed, suggesting either that no circular form of the HPI was generated upon excision, or that the number of episomal molecules was too low to be detected by this procedure. To test the latter hypothesis, a second round of amplification was performed using the initial PCR mixture as template and primers internal to the amplified sequence (primers 229A/229B in Fig. 1A). This nested PCR generated a product of the expected size (Supplementary material, Fig.S1A), the sequence of which was composed of the regions flanking the left and right ends of the HPI, separated by the attB sequence (Fig.S1B). Therefore, once excised from the chromosome, the HPI can form an episomal structure in the bacterial cytosol. This process is highly reminiscent of the mechanisms of phage excision. However, circular forms of the HPI seem to be very rare (much less frequent than one copy per bacterial chromosome) and probably transient, because of either their inability to replicate in the episomal form or their capacity to reinsert into the host genome. Although the former hypothesis is likely, the latter may also be true. This is suggested by the observation that, in individual colonies of strain IP32637, the HPI could be found inserted into either of two different asn tRNA (Buchrieser et al., 1998).

Re-evaluation of the rate of spontaneous HPI excision in strain IP32637

In order to study the mechanisms of excision of the HPI in Y. pseudotuberculosis, a marker that positively selects the clones deleted from the island was required as an alternative to the previously used time- and labour-consuming method based on pesticin sensitivity (Buchrieser et al., 1998). For this purpose, the counterselectable marker sacB cloned into the recombinant plasmid pPsnSac (Table 1) was inserted in the HPI of strain IP32637 by homologous recombination with the resident psn locus, following a single cross-over event (Fig. 1B). The resulting merodiploid strain designated IP32637sacB was subsequently used to determine the rate of spontaneous and precise excision (RΔ) of its HPI. This rate was found to be 7.5 × 10−5, a value close to that of ≈ 10 × 10−5 obtained previously using the pesticin sensitivity selection (Buchrieser et al., 1998).

Table 1. E. coli strains and plasmids used in this study.
 CharacteristicsSource or reference
Strains
Sm10 λpir thi, leu, thr, tonA, lacY, supE, recA[(RP4-2 Tc::Mu)] (λpir) Miller and Mekalanos (1988)
DH5αλpir supE44 ΔlacU169 (φ80 lacZΔM15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1[(RP4-2 Tc::Mu)] (λpir) Hanahan (1983)
XL1-bluesupE44 hsdR17 recA1 endA1 gyrA46 thi relA1 lac F′[proAB +lacI qlacZΔM15 Tn10 (tet r)]Invitrogen
DT-91F′, traD36, dam::tn9, lacIq, Δ(lacZ) M15, proA+ B+/endA1, glnV, sbcBC, thi-1, rpsL (StrR) (lac-pro) (P1) Skouloubris et al. (2001)
Plasmids
pBluescriptAmpRStratagene
pBluescript-psn psn cloned into the SacI and EcoRI sitesThis study
pBluescript-psn::sac sacB introduced into the BclI and SphI sites of psnThis study
pGP704Suicide vector, AmpR Miller and Mekalanos (1988)
pPsnSac psn::sacB cloned into the SacI and EcoRI sites of pGP704This study
pKOBEGTs, low copy number, CmR Chaveroche et al. (2000)
pUC4KHigh copy number, KmRAmersham
pAM239Low copy number, pLac, spectinomycinRGift from J. P. Bouché
pAM239-int int and its promoter cloned into the EcoRI and XbaI sitesThis study
pBAD33Low copy number, pArab, CmR Guzman et al. (1995)
pBADhef hef cloned into the HindIII and KpnI sites under the pArab promoter, CmRThis study
pASK75High copy number, inducible tetAp/o, AmpR Skerra (1994)
pASK75-int int gene cloned under the inducible tet promoter into pASK75This study

To evaluate the accuracy of RΔ, its median value and interdecile range were calculated on the results of 13 independent experiments. The median value of RΔ was found to be 6.6 × 10−6 for the IP32637 HPI, with an interdecile range (interval containing 80% of the observed RΔ) of 1 × 10−6−66 × 10−6. This variability in RΔ suggested that some variations in growth conditions influenced the rate of HPI deletion. The role of growth phase, temperature, type of medium (solid or liquid), iron availability and pH on HPI excisability was investigated. Some variations in RΔ were observed within each set of experiments; however, no specific growth condition was found significantly to increase or decrease the HPI excision rate (Supplementary material, TableS1). The possibility of a bacterium-to-bacterium variability in HPI excisability was investigated on three independent colonies of strain IP32637sacB, but no significant difference in RΔ was observed among these colonies (TableS1). Therefore, although a variation (usually around fivefold) in RΔ may be observed between experiments, no growth parameter influencing the excision capacity of the island could be identified.

The rate of HPI excision differs in various strains of Y. pseudotuberculosis I

In order to determine whether the rate of spontaneous HPI excision found in strain IP32637 was similar in other Y. pseudotuberculosis I strains, seven epidemiologically unrelated Y. pseudotuberculosis isolates were selected (Table 2), and the counterselectable marker sacB was introduced into their psn gene, as for IP32637. Two groups of strains were observed. One group (designated Exc+) contained, in addition to IP32637, two strains with an HPI that had the capacity to excise from the chromosome, although at a lower rate than in IP32637 (Table 2). The other group (designated Exc–) was composed of five strains in which no precise excision of the HPI could be detected (RΔ < 3 × 10−8). Therefore, in some Y. pseudotuberculosis I strains, the HPI does not have the ability to excise spontaneously from the chromosome.

Table 2. Rate of HPI excision in various Y. pseudotuberculosis I strains.
StrainIP32637 sacBIP32781 sacBIP32533 sacBIP32953 sacBIP32941 sacBIP32790 sacBIP33005 sacBIP32777 sacB
  1. RΔ, HPI excision rate. This value was determined by checking 100 SucrR colonies with primers internal to and flanking the HPI. When no precise excision of the island was detected, the HPI excisability was investigated further by screening large numbers of colonies (between 700 and 900) by colony blot hybridization with an irp2 probe, followed by PCR analysis.

  2. ND, not done.

Geographical originFranceFranceNew ZealandFranceFranceItalyGermanyFrance
HostHumanHumanDeerHumanHumanPigMonkeyHuman
6.6 × 10−52.5 × 10−68.5 × 10−7< 3 × 10−8< 5.4 × 10−9< 6 × 10−8< 6.2 × 10−9< 5.6 × 10−8
ExcisabilityExc+Exc+Exc+Exc–Exc–Exc–Exc–Exc–
asn tRNA insertion33222223
5 bp deletion in intNoNDNDYesYesYesYesNo

Lack of HPI excision results from neither the chromosomal location of the HPI nor a defect in the attB sites

In various strains of Y. pseudotuberculosis I, the HPI can be found inserted into any of the three copies of the asn tRNA locus present on the bacterial chromosome (Buchrieser et al., 1998). To determine whether the location of the HPI in a specific asn tRNA locus may be linked to its excision potential, the position of the island on the chromosome of the eight Y. pseudotuberculosis strains studied was investigated by hybridization of their DNA with an int and asn tRNA probe (TableS2). The HPI was found to be inserted at two asn tRNA positions in these strains (Fig. 2): in asn2 (five strains) or asn3 (three strains). However, there was no correlation between the site of insertion of the HPI and its ability to excise from the chromosome (Table 2).

image

Figure 2. Hybridization of the EcoRI-digested DNA of strains IP32637 ΔHPI, IP32953 HPI+ and IP32637 HPI+ with an asn tRNA probe. The IP32953 HPI+ and the IP32637 HPI+ profiles are representative of strains that have the HPI inserted into asn 2 and asn 3 respectively.

Download figure to PowerPoint

We showed previously that the HPI of strain IP32637 is flanked by a 17 bp repeat homologous to the attP site of bacteriophage P4, and that HPI excision results from the recombination between these two attB sites, leaving a single attB copy on the junction fragment (Buchrieser et al., 1998). To determine whether a mutation in the attB flanking sites may explain the lack of precise excision of the HPI in some Y. pseudotuberculosis strains, the partial genome sequence of the Exc– strain IP32953 (available at http://greengenes.llnl.gov/bbrp/html/y.pseudo.html) was examined. The two attB sites of IP32953 were intact and identical to those of the Exc+ strain IP32637, indicating that the lack of excisability of the HPI in IP32953 is not attributable to an alteration in the attB sites.

Most Exc– strains exhibit a 5 bp deletion in int

The precise excision of the HPI by recombination between two short-length repeats suggests a mechanism of site-specific recombination that requires the action of a site-specific recombinase. A likely candidate for this function is the P4-like integrase encoded by the HPI-borne int gene, located next to the asn tRNA locus (Fig. 1A). The role of a PAI-encoded integrase in the excision of its cognate island has been demonstrated recently for the integrase genes encoded by the 66 kb SRL PAI of Shigella flexneri 2a (Turner et al., 2001) and by the SaPIbov2 of S. aureus (Ubeda et al., 2003). Also consistent with a role for int in HPI excision, the Y. enterocolitica 1B int gene is naturally interrupted by a premature stop codon, and the HPI is incapable of precise excision from the chromosome (Bach et al., 1999). Comparison of the nucleotide sequences of the HPI-borne int genes in strains IP32953 (Exc–) and IP32637 (Exc+) showed that the two sequences are identical, except at position 602, where 5 bp is missing in the IP32953 sequence (Fig. 3A), leading to a premature stop codon at position 630 and to a putative product truncated of the last 211 amino acids. To determine whether a similar mutation is present in the int gene of the four other Exc– strains, a 130 bp fragment encompassing the 5 bp deletion identified in strain IP32953 was sequenced in each Exc– strain. One strain (IP32777) had an intact gene fragment, whereas the three others exhibited the same 5 bp deletion in the int sequence (Table 2). This mutation is different from the G to T substitution that inactivates the int gene of Y. enterocolitica Ye8081 (Buchrieser et al., 1998; Bach et al., 1999; Rakin et al., 1999). The fact that most Exc– strains had a non-functional P4-like integrase also argued for a role for this protein in the excision of the Y. pseudotuberculosis HPI.

image

Figure 3. A. Region of the int gene surrounding the 5 nucleotide deletion in the IP32953 sequence. B. Alignment of the amino acid (aa) sequences of Hef, AlpA (CP4-57 prophage) and Vis (P4 bacteriophage). C. Alignment of the amino acid (aa) sequences of the seven RDFs constituting the SLP1 subfamily. The helix–turn–helix (HTH) domain is indicated by bold characters and by the line above the sequences. clustal v software was used for alignment. (*) identical nucleotides or aa, (⋅) similar aa.

Download figure to PowerPoint

The HPI-encoded integrase is required for the excision of its cognate island

To confirm the role of the HPI-borne integrase in the excision of the Y. pseudotuberculosis HPI, the int gene of strain IP32637sacB was inactivated by allelic exchange between the chromosomal locus and a linear DNA fragment carrying an int copy disrupted by a kan cassette (Fig. 1C), using the rapid mutagenesis procedure LFH-PCR (Derbise et al., 2003). When int was inactivated, HPI excision became undetectable (RΔ < 8 × 10−8; Table 3) in the mutant strain (Int3). To ensure that loss of HPI excision was the consequence of int disruption and not of a polar effect of the mutation, the functional int gene of IP32637 cloned into the low-copy-number plasmid pAM239 (yielding pAM239-int) and the control plasmid pAM239 were introduced into the Int3 mutant. As shown in Table 3, the presence of pAM239 alone did not modify RΔ, whereas trans-complementation of Int3 with a functional int allele restored the excision capacity of the HPI to a level comparable to that of the parental strain. Therefore, a functional P4-like integrase is an essential component of the HPI excision machinery.

Table 3. Role of int in HPI excision.
 IP32637sacBInt3aInt3 (pAM239)Int3 (pAM239-int)IP32953sacB (pAM239)IP32953sacB (pAM239-int)
  • RΔ, HPI excision rate. This value was determined by checking 100 SucrR Int3 colonies for kanamycin resistance.

  • a

    . Int3, IP32637sacB derivatives with a deletion in the HPI-borne int gene. To construct this mutant, a 500 bp region upstream (primer pair 132A/132B-kan) and downstream (primer pair 133A-kan/133B) of int was amplified. In each pair, one primer contained at its 5′ end a 17 bp region homologous to the 5′ extremity of the kan gene. Simultaneously, the NdeI-digested plasmid pUC4K was used as template to amplify the kan cassette (primer pair 136A/B). Upstream and downstream int PCR products were mixed with the kan PCR fragment and with primer pair 132A/133B. The resulting product was used as template for a PCR with primer pair 132A/133B. Each PCR was performed using a 3:1 mixture of AmpliTaq DNA polymerase (Applied Biosystems) and Pfu polymerase (Stratagene). To allow allelic exchange between the chromosomal target gene and the mutated int sequence generated by PCR, the linear products were introduced by electroporation into strain IP32637sacB harbouring the pKOBEG plasmid (Chaveroche et al., 2000). Selection of transformants was achieved on Km agar plates. The deletion of int was checked by PCR with primer pairs 266A/266B, 266A/167, 266B/166 and by hybridization with the int and kan probes. Primer pairs 266A/266B and 183A/213 were designed to hybridize outside the region of recombination (Fig. 1 and TableS2).

7.5 × 10−5< 8 × 10−8< 9.4 × 10−83 × 10−6< 6.4 × 10−89 × 10−6

To determine whether, in the Exc– strain IP32953, the 5 bp deletion in int was the only defect accounting for the lack of excisability of the island, plasmids pAM239 and pAM239-int were introduced into this strain. Although pAM239 alone had no effect on the excision capacity of the HPI, the introduction of a functional int gene into IP32953sacB restored the ability of the HPI to excise from the chromosome, to a level comparable to that of strain IP32637sacB (Table 3). This demonstrates that the 5 bp deletion in int was the sole determinant responsible for the lack of HPI excision in strain IP32953. This inactivation of int results in the fixation of the island on the bacterial chromosome and may represent a selective advantage for the host bacterium, by providing a supplementary and powerful means of acquiring the essential iron molecules necessary for bacterial survival.

Int is expressed constitutively but at a very low level

To evaluate the basal level of int expression, RNA dot-blots using twofold dilutions (5 µg to 0.15 µg) of RNA extracted from strains IP32637sacB and IP32637 ΔHPI (negative control) were performed, using an int probe and an rpsL probe (a single-copy, constitutively expressed chromosomal gene; Keener and Nomura, 1996). A hybridization signal, positive down to 0.31 µg of RNA, was observed with the rpsL probe, whereas no signal was detected with the int probe, even with amounts of RNA as high as 5 µg (data not shown). This suggested that the level of int expression is very low under basal conditions, because of either a low rate of int transcription or a rapid decay of int mRNAs. We thus looked at the production of the integrase protein in strain IP32637sacB by Western blot analysis. Although in the positive control (E. coli XL1 overexpressing int cloned into plasmid pASK75), the expected band of 48 kDa was detected, the anti-Int rabbit polyclonal antiserum did not recognize any band of the expected size in the protein extract of IP32637sacB (data not shown). As int transcription may be transient, we investigated int expression during the various bacterial growth phases. Reverse transcription PCR (RT-PCR, a method more sensitive than RNA dot-blot) on 10-fold dilutions of RNA templates (1 µg, 100 ng and 10 ng) harvested at early exponential, stationary and late stationary growth phases from strain IP32637sacB yielded a product of the expected size whatever the growth phase. Therefore, int is constitutively expressed, but its level of transcription is very low. As excision of the HPI is readily observed (≈ 6 × 10−6) under the same growth conditions, these results indicate that very low levels of int transcripts are sufficient to promote HPI excision.

Identification of Hef, a novel factor involved in HPI excision

One Exc– strain (IP32777) did not exhibit the 5 bp deletion in int found in other Exc– strains (Table 2). Furthermore, trans-complementation of IP32777sacB with pAM239-int did not confer to the HPI the capacity to excise from the bacterial chromosome, indicating that the non-excisability of the island is not attributable to a defect in int. To determine whether degenerate attB sites may be responsible for the lack of HPI excision, the right and left borders of the HPI were sequenced. The attB-R and attB-L sites of strain IP32777 were found to be intact and identical to those of the Exc+ strain IP32637. Thus, absence of HPI excision in strain IP32777 could not be explained by either a defect in int or degenerate attB sites. This suggested that, in addition to int and the two attB sites, another factor is required for HPI excision.

It was reported previously that, in E. coli K-12, excision of the cryptic prophage CP4-57 from the bacterial chromosome requires the presence of a factor designated AlpA (Kirby et al., 1994; Trempy et al., 1994). Overexpression of AlpA resulted in a dramatic increase in CP4-57 excision (Trempy et al., 1994). AlpA was also found to be homologous to Vis, a protein encoded by bacteriophage P4. A search for an alpA homologue on the HPI in the partial genome sequence of strain IP32953 identified a coding sequence that we designated hef (for HPI excision factor), which has the capacity to encode a small basic protein (calculated pI of 9.69) of 61 amino acids. Hef shares 37% identity and 64% similarity with AlpA (Kirby et al., 1994) and 36% identity and 60% similarity with Vis (Polo et al., 1996). Vis and AlpA are as distantly related between themselves (33% identity and 56% similarity) as with Hef (Fig. 3B). The three proteins contain a conserved helix–turn–helix (HTH) motif that may represent a DNA-binding domain (Trempy et al., 1994; Lewis and Hatfull, 2001). hef is located on the left-hand part of the HPI, in a region containing several short phage-like cryptic genes (Fig. 1A). A hef homologue (designated yp43 in strains 6/69M and KIM10, and ypo1904 in CO92) is present at the same position on the Y. pestis HPI, and the two genes are 100% identical in the two species. In contrast, the DNA region where hef is located is absent from the Y. enterocolitica HPI (Carniel et al., 1996; Rakin et al., 1999). A search for a Hef homologue encoded by the HPI of Y. enterocolitica Ye8081 (sequence available at http://www.sanger.ac.uk/Projects/Y_enterocolitica/) and WA (Rakin et al., 1999) remained negative, suggesting that there is no hef gene on the HPI of this species. Similarly, analysis of the HPI sequence of the uropathogenic E. coli strain CFT073 (Welch et al., 2002) did not identify any gene that may encode a product similar to Hef.

Recently, Lewis and Hatfull (2001) identified by database mining a large group of proteins that are predicted to be involved in the control of directionality in integrase-mediated site-specific recombination. This group of proteins was designated ‘recombination directionality factors’ (RDFs). Most RDFs are typically small (<100 amino acids) basic proteins, with a putative HTH DNA-binding motif. The Y. pestis Hef homologue (named yp43) was found to belong to the RDF family and, more specifically, to a subgroup designated SLP1 (Lewis and Hatfull, 2001). Within the same subgroup, RDFs may be distantly related in terms of amino acid sequences. This is indeed the case for Hef and the other members of the SLP1 subfamily (Fig. 3C). However, all members of this subfamily share one common characteristic, which is that they have a P4-like cognate integrase.

A functional hef gene is essential for HPI excision

To determine whether hef does indeed play a role in the excision of the Y. pseudotuberculosis HPI, this gene was disrupted in strain IP32637sacB by allelic exchange between the chromosomal hef gene and a kan cassette, using the LFH-PCR technique (Fig. 1D). One mutant, designated Hef1, was used to determine the rate of HPI excision. No Hef1 colony with a precise deletion of the island could be detected (Table 4). To confirm that loss of HPI excision was the direct consequence of hef inactivation, this gene was amplified and cloned into the moderate-copy-number plasmid pBAD33 (Guzman et al., 1995), downstream of the arabinose promoter, yielding pBADhef. In the absence of arabinose induction, some Hef1(pBADhef) clones in which the HPI had excised precisely from the chromosome were detected, although at a low rate (Table 4), suggesting a slight leak of the pBAD promoter on pBADhef. In contrast, overexpression of hef resulted in a marked increase in HPI excision, with an RΔ approximately 100 times higher than that of the parental strain IP32637sacB (Table 4). The enhancing effect of the overexpression of hef on HPI excision was also examined in the presence of an intact HPI-borne hef gene. As shown in Table 4, overexpression of hef in IP32637sacB resulted in a 200-fold increase in RΔ. Therefore, a functional hef gene is essential for HPI excision, and its level of expression determines the rate of excision.

Table 4. Role of hef in HPI excision.
 IP32637sacBHef1aHef1 (pBADhef) No inductionHef1 (pBADhef) InductionIP32637sacB (pBADhef) No inductionIP32637sacB (pBADhef) InductionInt3Int3 (pBADhef) InductionIP32777sacBIIP32777sacB (pBADhef) Induction
  • RΔ, HPI excision rate.

  • a

    . Hef1, IP32637sacB derivatives with a deletion of the HPI-borne hef gene. The same technique as that described in the footnote to Table 3, using primer pairs 199A/199B-kan and 200A-kan/200B, was used to amplify a PCR product containing the kan cassette flanked by regions homologous to the extremities and surrounding regions of hef. The deletion of hef was checked with primer pairs 183A/213, 183A/167, 213/166 (TableS2).

  • Int3, IP32637sacB derivatives with a deletion in the HPI-borne int gene.

6 × 10−6< 6.3 × 10−83.9 × 10−84.5 × 10−40.8 × 10−61.6 × 10−4< 8 × 10−8< 2.5 × 10−8< 5.6 × 10−84 × 10−5

We showed above that no defect in the attB sites or the int gene could account for the Exc– phenotype of strain IP32777. To test whether a defect in hef expression might explain the lack of HPI excisability, pBADhef was introduced into strain IP32777sacB. When hef was overexpressed, the HPI became able to excise from the chromosome at a rate higher than that of the parental strain (Table 4). Therefore, absence of HPI excision in strain IP32777 results from a defect in hef expression.

To determine whether hef overexpression could circumvent the effect of an inactivation of int, pBADhef was introduced into the Int3 mutant (Δint). Overexpression of hef did not allow the detection of HPI-deleted clones (Table 4), indicating that both Hef and Int are required for the precise excision of the HPI.

Hef is not a transcriptional regulator of int

In E. coli K-12, AlpA was shown previously to upregulate the transcription of the CP4-57 integrase gene, which in turn mediates the excision of the prophage from the bacterial chromosome (Trempy et al., 1994). To determine whether, in a manner similar to AlpA, Hef acts on int transcription, RT-PCRs were performed on serial dilutions of RNA templates extracted from strain IP32637sacB and from the hef mutant Hef1. The threshold (0.25 ng of RNA template) of detection of an RT-PCR product was the same for the two strains (Fig.S2). When RT-PCRs were performed on RNA extracts of strains IP32637sacB and Hef1(pBADhef), overexpression of hef did not lead to a detectable augmentation of int transcription (Fig.S2). Therefore, hef inactivation or overexpression had no effect on int transcription. We also looked at the synthesis of Int in strain IP32637sacB(pBADhef) overexpressing hef, but were still unable to detect the protein by Western blot (data not shown). Therefore, although the level of hef expression severely affects the rate of HPI excision, it has little or no effect on int transcription.

Hef is predicted to drive the function of Int towards an excisionase activity

With the exception of the HP1-P2 subfamily, which is known also to act as a transcriptional regulator, the members of the other RDF subfamilies (including SLP1) are devoid of this activity (Lewis and Hatfull, 2001). This substantiates our observation that Hef does not have any significant effect on int transcription. The best studied RDF is the phage lambda-encoded Xis protein. Xis determines the directionality of recombination by binding to specific sequences and introducing DNA bends that induce the formation of a complex in which attB-R and attB-L are bridged by bivalent DNA-binding Int proteins (Kim and Landy, 1992). In addition to functioning as a DNA architectural factor, Xis also co-operatively recruits the integrase through interactions at its C-terminus (Numrych et al., 1992; Wu et al., 1998; Sam et al., 2002). The DNA conformational modifications introduced by Xis promote the excision activity of Int, while preventing the integration reactions.

Because of the homology of Hef with a family of recombination directionality factors, and as one-third of the RDFs identified by Lewis and Hatfull (2001) have been experimentally confirmed as excisionases, we can predict that Hef, like the other members of this family, has an architectural rather than a catalytic role and promotes HPI excision from the chromosome by driving the activity of Int towards an excisionase. Although the combined roles of Int and RDFs have been shown previously to be important for the excision of some phage genomes from the bacterial host chromosome, this is the first demonstration that similar factors, encoded by pathogenicity islands, are also required for their excision.

Conclusion

PAIs have defined characteristics that clearly differentiate them from the backbone genome (Hacker et al., 1997). They are thought to be mobile elements that have been acquired by horizontal transfer. The HPI has retained the ability to excise from the chromosome, to form a circular episomal DNA molecule and to reintegrate, either in the same location or in another asn tRNA copy, within the genome of Y. pseudotuberculosis. The fact that the HPI is present in a wide variety of bacterial genera such as E. coli, Enterobacter, Citrobacter, Klebsiella and Salmonella (Schubert et al., 1998; 2000; Karch et al., 1999; Bach et al., 2000; Oelschlaeger et al., 2003), where it displays a highly conserved genetic organization and a very high level of nucleotide identity (≈ 99%), suggests that this island has been acquired recently by these various species. This suggests that the HPI has retained a high potential of mobility. However, whether the HPI still has the ability to be horizontally transferred remains to be determined and, if so, the nature of the molecular mechanisms mediating this transfer will need to be elucidated.

Very recently, the HPI of one particular E. coli strain (ECOR31) was found structurally to resemble an integrative and conjugative element termed ICE, and it was proposed that this element was a mobilizable progenitor of the HPI (Schubert et al., 2004). However, this element is restricted to a single isolate, is inserted in an asn tRNA different from that of other HPI-positive enterobacteria, contains a mobilization segment with a GC content different from that of the HPI and is inserted in a region of the HPI known to be a hot-spot of IS integration. This rather suggests that this mosaic structure results from the accidental insertion of the mobilization segment in the integration hot-spot of this particular strain.

Another hypothesis is that PAIs are of phage origin. A few PAIs have been shown to have the capacity to be transferred by a helper phage. This is the case for SaPI1 of S. aureus, which could be transduced by the helper phage 80α (Lindsay et al., 1998; Ruzin et al., 2001), and for V. cholerae VPI, which could use phage CP-T1 for transfer (O’Shea and Boyd, 2002). This process is reminiscent of the P2–P4 phage interactions in which the defective bacteriophage P4 uses P2 as a helper phage for transduction. Although the number of PAIs identified in members of the Enterobacteriaceae family is growing steadily, there has been no demonstration until now that these PAIs may be mobilized by a helper phage. Similarly, very few or no sequences homologous to genes involved in phage replication, regulation and capsid assembly have been identified on these PAIs. However, the HPI possesses attB sites as well as hef and int genes homologous to bacteriophage P4 genes. The island, like phages, excises precisely from the host chromosome at the att sites and forms circular molecules that can subsequently reinsert into chromosomal attB sites. Therefore, the fact that the HPI, and probably several other PAIs, carries a machinery of integration/excision genetically and functionally highly similar to those of bacteriophages argues strongly for phage-mediated acquisition and transfer of these elements.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. Supplementary material
  8. References
  9. Supporting Information

Bacterial strains and growth conditions

The E. coli strains and the plasmids used in this study are listed in Table 1. Bacteria were grown in Luria–Bertani (LB) broth, on LB agar plates or on LB agar plates without NaCl and supplemented with 10% sucrose (LB-Sucr). Y. pseudotuberculosis strains were taken from the collection of the Yersinia Research Unit, Institut Pasteur. They were grown at 28°C for 24 h with shaking (liquid media) or for 48 h (solid media). E. coli strains were grown at 37°C for 24 h. When necessary, ampicillin (100 µg ml−1), kanamycin (30 µg ml−1), chloramphenicol (25 µg ml−1), irgasan (1 µg ml−1) or spectinomycin (50 µg ml−1) was added to the media.

DNA techniques

Genomic and plasmid DNA was extracted with the Isoquick nucleic acid extraction kit and the Qiagen plasmid maxi kit respectively. Approximately 5 µg of DNA from the various Yersinia strains was digested with EcoRI and subjected to Southern hybridization. Colony blot hybridizations were done on bacterial colonies spotted onto nylon filters (Maniatis et al., 1989) and hybridized with a digoxigenin (DIG)-labelled irp2 probe. Sequencing of DNA fragments was performed by ESGS-Group Qbiogene or Genome express, using the Big Dye Terminator technique and an AB3730 XL capillary sequencer.

PCR primers listed in TableS2 were purchased from Genset. For probe preparations, PCRs were performed with 1 unit of AmpliTaq DNA polymerase (Applied Biosystems). For cloning experiments, PCRs were done with 1 unit of Pfu polymerase (Stratagene). PCR amplification mixtures contained 1 µM each primer and 200 µM each of the four deoxynucleoside triphosphates, except for the DIG probe in which dTTP was used at 26 µM and Dig-11dUTP at 13 µM. Each assay involved three steps of 25 cycles of amplification as follows: 94°C for 30 s, 55°C for 30 s, 72°C for 60 s kb−1 synthesized.

Construction of recombinant psn::sacB Y. pseudotuberculosis strains

To insert a sacB cassette into the HPI-borne psn gene of various Y. pseudotuberculosis strains, the psn locus deleted of its 3′ part was PCR amplified with primer pair F1/F2 and cloned into the SacI–EcoRI sites of pBluescript, resulting in plasmid pBluescript-psn, which was introduced in E. coli DT91 (dam–) to allow BclI digestion. The sacB gene with its own promoter was PCR amplified with primer pair 121G/D and inserted into the SphI and BclI sites of the cloned psn gene, removing 400 bp of psn and resulting in plasmid pBluescript-psn::sac. pBluescript-psn::sac was digested with EcoRI and BclI, and the psn::sacB fragment was subcloned into the SacI and EcoRI sites of the suicide vector pGP704, resulting in plasmid pPsnSac. This plasmid was introduced by electroporation, first into E. coli DH5αλpir, and then into E. coli Sm10λpir. E. coli Sm10λpir(pPsnSac) was mated on filters with the various strains of Y. pseudotuberculosis I listed in Table 2. Y. pseudotuberculosis transconjugants were selected on LB plates containing irgasan and ampicillin. Integration of the plasmid into the chromosomal psn gene was checked by PCR amplification with primer pairs 113C/121F, 121I/pGP704fw and pGP704rv/113F (Fig. 1B and TableS2).

Determination of the HPI excision rate

Yersinia pseudotuberculosis strains harbouring a sacB-labelled HPI were grown in LB broth at 28°C for 24 h, and 10-fold serial dilutions were streaked in parallel on LB and LB-Sucr plates. As a portion of the colonies able to grow on LB-Sucr could result from either a mutation in sacB or a recombinational event in psn, leading to the excision of pPsnSac, the actual number of ΔHPI mutants was determined by performing either: (i) PCR amplification of irp2 (primer pair 18/19) and ybtPQ (primer pair 123B/124A) on 100–500 SucrR colonies (Fig. 1A); or (ii) colony blot hybridization with a DIG-labelled irp2 probe when a higher number (700–900) of colonies was analysed. ΔHPI mutants were taken into account only if both irp2 and ybtPQ loci were absent. To determine the number of clones with precise HPI excision, i.e. starting and ending in the attB-R and attB-L sites, three PCRs were performed with primers encompassing the extremities of the HPI and the junction fragment generated upon HPI excision. As the HPI is known to insert into different copies of the asn tRNA locus (Buchrieser et al., 1998), two different sets of primers for the extremities and the junction fragment were used (Fig. 1A). Set 1 included primer pairs A10/143A (right-hand end), 144A/A9 (left-hand end) and A9/A10 (junction fragment) for the HPI inserted into asn3. Set 2 included primer pairs 143A/143B (right-hand end), 144A/144B (left-hand end) and 143B/144B (junction fragment) for the HPI inserted into asn2. True excisants were clones entirely and precisely deleted of their HPI.

Construction and complementation of int and hef mutants of IP32637sacB

To construct mutants, PCR fragments formed of a kanamycin resistance cassette flanked by long portions (i.e. ≈ 500 bp) of the extremities and surrounding regions of int or hef were generated, using the LFH-PCR technique (Derbise et al., 2003), as described in the footnotes to Tables 3 and 4. Cells were rendered electrocompetent as described previously (Conchas and Carniel, 1990). To ensure that no mutation in the regions surrounding the target sequence was created in Int3 during the LFH-PCR, the 500 bp chromosomal regions flanking the integrated kan cassette were sequenced.

To complement int, a 1521 bp product containing the entire int ORF and upstream region (predicted to contain the int promoter) was amplified by PCR with primer pair 215A/215B. This product was cloned into the EcoRI–XbaI sites of plasmid pAM239, resulting in pAM239-int. The int orientation did not allow its expression from the plac plasmid promoter. pAM239-int was introduced by electroporation, first into E. coli XL1-Blue and then into IP32637sacB Δint, IP32953sacB and IP3777sacB isolates. The Y. pseudotuberculosis hef gene was amplified with primer pair 214A/214B and inserted into the HindIII–KpnI sites of plasmid pBAD33 under the pArab promoter, resulting in pBADhef. This plasmid was introduced by electroporation into E. coli XL1-Blue and then into Y. pseudotuberculosis IP32637sacB, Hef1 and IP32777sacB isolates. Sequencing of the cloned hef gene confirmed that the amplification process did not introduce any error in the sequence.

Extraction of RNA, RNA dot-blot and RT-PCR

Overnight bacterial cultures were harvested at early exponential (OD600 = 0.2), stationary (OD600 = 1) and late stationary (OD600 = 1.6 and OD600 = 2) phases. Bacterial RNAs were extracted with the RNAwiz and DNA-free kits (Ambion), according to the manufacturer's recommendations. The quality of the RNA was checked by agarose gel electrophoresis and PCR amplification with primer pair 289A/289B to ensure that the preparation contained no trace of DNA. For RNA dot-blot, twofold serial dilutions (5 µg to 0.18 µg) of the RNA samples were spotted onto a nylon membrane (Hybond N+ Amersham). As a control for hybridization, 1.5 µg of IP32637sacB denatured DNA was also spotted. The two probes used corresponded to an ≈ 400 bp internal portion of rpsL and int obtained by PCR amplification with primer pairs 301A/301B and 116B/8A respectively (TableS2). The PCR products were purified with the PCR purification kit (Qiagen) and radioactively labelled with 33P using the Prime it II kit (Stratagene). Prehybridization and hybridization were performed at 52°C. The membrane was washed in 1× SSC, 0.1% SDS for 15 min and then in 0.1× SSC, 0.1% SDS for 10 min before being subjected to autoradiography. RT-PCR was performed using the SuperScript One Step RT-PCR with Platinum Taq kit (Invitrogen). The reaction contained the supplier buffer (1×) mixed with 0.2 µM primer pair 289A/289B (corresponding to an internal region of int), 1 µl of RT/Platinum Taq mix and 0.5 µg to 0.01 ng of RNA template. Each reaction involved one step for 30 min at 50°C, one step for 2 min at 94°C and 30 cycles of three steps as follows: 94°C for 15 s, 55°C for 30 s, 68°C for 45 s.

SDS–PAGE and Western blot

Total bacterial proteins obtained after sonication were subjected to SDS–PAGE on 8% SDS-polyacrylamide gels. A rabbit polyclonal antisera was raised against a mixture of two peptides: CTLKPSDKPFKVSDSH (amino acids 9–24) and CRESYAPPYTIGKNK (amino acids 403–417) of the HPI-borne integrase. Western blots were performed according to the protocol of Burnette (1981), using a 1:500 dilution of the Int-specific antiserum.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. Supplementary material
  8. References
  9. Supporting Information

The advice of Didier Guillemot and Claire Bernede in statistical analyses has been extremely useful. B.L. is the recipient of a grant from the French Ministry of Research and Technology. The work of U.D and J.H. was supported by the DFG (grant SFB479, A1).

Supplementary material

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. Supplementary material
  8. References
  9. Supporting Information

Fig. S1. A. Amplification of the junction fragment of the HPI circular form by nested PCR using primer pair 229A/229B. PCR was performed on thermolysates of strain IP32637sacB and IP32637ΔHPI.

B. Sequence of the junction fragment of the circular form obtained by nested PCR.

Fig. S2. RT-PCR amplification of int mRNA from different hef recombinant strains of IP32637, using various amounts of RNA templates (numbers above the figure).

Table S1. Influence of different growth conditions on HPI excision in strain IP32637sacB.

Table S2. Primers used in this study.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. Supplementary material
  8. References
  9. Supporting Information
  • Bach, S., Buchrieser, C., Prentice, M., Guiyoule, A., Msadek, T., and Carniel, E. (1999) The high-pathogenicity island of Yersinia enterocolitica Ye8081 undergoes low-frequency deletion but not precise excision, suggesting recent stabilization in the genome. Infect Immun 0: 50915099.
  • Bach, S., De Almeida, A., and Carniel, E. (2000) The Yersinia high-pathogenicity island is present in different members of the family Enterobacteriaceae. FEMS Microbiol Lett 183: 289294.
  • Bearden, S.W., Fetherston, J.D., and Perry, R.D. (1997) Genetic organization of the yersiniabactin biosynthetic region and construction of avirulent mutants in Yersinia pestis. Infect Immun 65: 16591668.
  • Blum, G., Ott, M., Lischewski, A., Ritter, A., Imrich, H., Tschäpe, H., and Hacker, J. (1994) Excision of large DNA regions termed pathogenicity islands from tRNA-specific loci in the chromosome of an Escherichia coli wild-type pathogen. Infect Immun 62: 606614.
  • Buchrieser, C., Brosch, R., Bach, S., Guiyoule, A., and Carniel, E. (1998) The high-pathogenicity island of Yersinia pseudotuberculosis can be inserted into any of the three chromosomal asn tRNA genes. Mol Microbiol 30: 965978.
  • Burnette, W.N. (1981) ‘Western blotting’: electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein. Anal Biochem 112: 195203.
  • Carniel, E., Guiyoule, A., Guilvout, I., and Mercereau-Puijalon, O. (1992) Molecular cloning, iron-regulation and mutagenesis of the irp2 gene encoding HMWP2, a protein specific for the highly pathogenic Yersinia. Mol Microbiol 6: 379388.
  • Carniel, E., Guilvout, I., and Prentice, M. (1996) Characterization of a large chromosomal ‘high-pathogenicity island’ in biotype 1B Yersinia enterocolitica. J Bacteriol 178: 67436751.
  • Chaveroche, M.K., Ghigo, J.M., and D’Enfert, C. (2000) A rapid method for efficient gene replacement in the filamentous fungus Aspergillus nidulans. Nucleic Acids Res 28: E97.
  • Conchas, R.F., and Carniel, E. (1990) A highly efficient electroporation system for transformation of Yersinia. Gene 87: 133137.
  • Cornelis, G. (1987) [Yersinia enterocolitica, an excellent model for the molecular study of the pathogenicity of invasive bacteria]. Bull Mem Acad R Med Belg 142: 126136.
  • De Almeida, A., Guiyoule, A., Guilvout, I., Iteman, I., Baranton, G., and Carniel, E. (1993) Chromosomal irp2 gene in Yersinia: distribution, expression, deletion and impact on virulence. Microb Pathog 14: 921.
  • Derbise, A., Lesic, B., Dacheux, D., Ghigo, J.M., and Carniel, E. (2003) A rapid and simple method for inactivating chromosomal genes in Yersinia. FEMS Immunol Med Microbiol 38: 113116.
  • Fetherston, J.D., and Perry, R.D. (1994) The pigmentation locus of Yersinia pestis KIM6+ is flanked by an insertion sequence and includes the structural genes for pesticin sensitivity and HMWP2. Mol Microbiol 13: 697708.
  • Fetherston, J.D., Schuetze, P., and Perry, R.D. (1992) Loss of the pigmentation phenotype in Yersinia pestis is due to the spontaneous deletion of 102 kb of chromosomal DNA which is flanked by a repetitive element. Mol Microbiol 6: 26932704.
  • Fetherston, J.D., Bertolino, V.J., and Perry, R.D. (1999) YbtP and YbtQ: two ABC transporters required for iron uptake in Yersinia pestis. Mol Microbiol 32: 289299.
  • Gehring, A.M., Mori, I.I., Perry, R.D., and Walsh, C.T. (1998) The nonribosomal peptide synthetase HMWP2 forms a thiazoline ring during biogenesis of yersiniabactin, an iron-chelating virulence factor of Yersinia pestis. Biochemistry 37: 17104.
  • Guzman, L.M., Belin, D., Carson, M.J., and Beckwith, J. (1995) Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J Bacteriol 177: 41214130.
  • Hacker, J., Bender, L., Ott, M., Wingender, J., Lund, B., Marre, R., and Goebel, W. (1990) Deletions of chromosomal regions coding for fimbriae and hemolysins occur in vitro and in vivo in various extraintestinal Escherichia coli isolates. Microb Pathog 8: 213225.
  • Hacker, J., Blum-Oehler, G., Muhldorfer, I., and Tschape, H. (1997) Pathogenicity islands of virulent bacteria: structure, function and impact on microbial evolution. Mol Microbiol 23: 10891097.
  • Hanahan, D. (1983) Studies on transformation of Escherichia coli with plasmids. J Mol Biol 166: 557580.
  • Hare, J.M., Wagner, A.K., and McDonough, K.A. (1999) Independent acquisition and insertion into different chromosomal locations of the same pathogenicity island in Yersinia pestis and Yersinia pseudotuberculosis. Mol Microbiol 31: 291303.
  • Heesemann, J. (1987) Chromosomal-encoded siderophores are required for mouse virulence of enteropathogenic Yersinia species. FEMS Microbiol Lett 48: 229233.
  • Karch, H., Schubert, S., Zhang, D., Zhang, W., Schmidt, H., Olschlager, T., and Hacker, J. (1999) A genomic island, termed high-pathogenicity island, is present in certain non-O157 Shiga toxin-producing Escherichia coli clonal lineages. Infect Immun 67: 59946001.
  • Keener, J., and Nomura, M. (1996) Regulation of ribosome synthesis. In Escherichia coli and Salmonella. Cellular and Molecular Biology, 2nd edn. Neidhardt, F.C. (ed.). Washington, DC: American Society for Microbiology Press, pp. 14171431.
  • Kim, S., and Landy, A. (1992) Lambda Int protein bridges between higher order complexes at two distant chromosomal loci attL and attR. Science 256: 198203.
  • Kirby, J.E., Trempy, J.E., and Gottesman, S. (1994) Excision of a P4-like cryptic prophage leads to Alp protease expression in Escherichia coli. J Bacteriol 176: 20682081.
  • Lewis, J.A., and Hatfull, G.F. (2001) Control of directionality in integrase-mediated recombination: examination of recombination directionality factors (RDFs) including Xis and Cox proteins. Nucleic Acids Res 29: 22052216.
  • Lindsay, J.A., Ruzin, A., Ross, H.F., Kurepina, N., and Novick, R.P. (1998) The gene for toxic shock toxin is carried by a family of mobile pathogenicity islands in Staphylococcus aureus. Mol Microbiol 29: 527543.
  • Maniatis, T., Fritsch, E.F., and Sambrook, J. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor University.
  • Miller, V.L., and Mekalanos, J.J. (1988) A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J Bacteriol 170: 25752583.
  • Numrych, T.E., Gumport, R.I., and Gardner, J.F. (1992) Characterization of the bacteriophage lambda excisionase (Xis) protein: the C-terminus is required for Xis-integrase cooperativity but not for DNA binding. EMBO J 11: 37973806.
  • O'Shea, Y.A., and Boyd, E.F. (2002) Mobilization of the Vibrio pathogenicity island between Vibrio cholerae isolated mediated by CP-T1 generalized transduction. FEMS Microbiol Lett 214: 153157.
  • Oelschlaeger, T.A., Zhang, D., Schubert, S., Carniel, E., Rabsch, W., Karch, H., and Hacker, J. (2003) The high-pathogenicity island is absent in human pathogens of Salmonella enterica subspecies I but present in isolates of subspecies III and VI. J Bacteriol 185: 11071111.
  • Pelludat, C., Hogardt, M., and Heesemann, J. (2002) Transfer of the core region genes of the Yersinia enterocolitica WA-C serotype O:8 high-pathogenicity island to Y. enterocolitica MRS40, a strain with low levels of pathogenicity, confers a yersiniabactin biosynthesis phenotype and enhanced mouse virulence. Infect Immun 70: 18321841.
  • Perry, R.D., Balbo, P.B., Jones, H.A., Fetherston, J.D., and DeMoll, E. (1999) Yersiniabactin from Yersinia pestis: biochemical characterization of the siderophore and its role in iron transport and regulation. Microbiology 145: 11811190.
  • Polo, S., Sturniolo, T., Deho, G., and Ghisotti, D. (1996) Identification of a phage-coded DNA-binding protein that regulates transcription from late promoters in bacteriophage P4. J Mol Biol 257: 745755.
  • Rajanna, C., Wang, J., Zhang, D., Xu, Z., Ali, A., Hou, Y.M., and Karaolis, D.K. (2003) The vibrio pathogenicity island of epidemic Vibrio cholerae forms precise extrachromosomal circular excision products. J Bacteriol 185: 68936901.
  • Rakin, A., Urbitsch, P., and Heesemann, J. (1995) Evidence for two evolutionary lineages of highly pathogenic Yersinia species. J Bacteriol 177: 22922298.
  • Rakin, A., Noelting, C., Schubert, S., and Heesemann, J. (1999) Common and specific characteristics of the high-pathogenicity island of Yersinia enterocolitica. Infect Immun 67: 52655274.
  • Rakin, A., Noelting, C., Schropp, P., and Heesemann, J. (2001) Integrative module of the high-pathogenicity island of Yersinia. Mol Microbiol 39: 407415.
  • Ruzin, A., Lindsay, J., and Novick, R.P. (2001) Molecular genetics of SaPI1 – a mobile pathogenicity island in Staphylococcus aureus. Mol Microbiol 41: 365377.
  • Sam, M.D., Papagiannis, C.V., Connolly, K.M., Corselli, L., Iwahara, J., Lee, J., et al. (2002) Regulation of directionality in bacteriophage lambda site-specific recombination: structure of the Xis protein. J Mol Biol 324: 791805.
  • Schubert, S., Rakin, A., Karch, H., Carniel, E., and Heesemann, J. (1998) Prevalence of the ‘high-pathogenicity island’ of Yersinia species among Escherichia coli strains that are pathogenic to humans. Infect Immun 66: 480485.
  • Schubert, S., Cuenca, S., Fischer, D., and Heesemann, J. (2000) High-pathogenicity island of Yersinia pestis in enterobacteriaceae isolated from blood cultures and urine samples: prevalence and functional expression. J Infect Dis 182: 12681271.
  • Schubert, S., Dufke, S., Sorsa, J., and Heesemann, J. (2004) A novel integrative and conjugative element (ICE) of Escherichia coli: the putative progenitor of the Yersinia high-pathogenicity island. Mol Microbiol 51: 837848.
  • Skerra, A. (1994) Use of the tetracycline promoter for the tightly regulated production of a murine antibody fragment in Escherichia coli. Gene 151: 131135.
  • Skouloubris, S., Labigne, A., and De Reuse, H. (2001) The AmiE aliphatic amidase and AmiF formamidase of Helicobacter pylori: natural evolution of two enzyme paralogs. Mol Microbiol 40: 596609.
  • Trempy, J.E., Kirby, J.E., and Gottesman, S. (1994) Alp suppression of Lon: dependence on the slpA gene. J Bacteriol 176: 20612067.
  • Turner, S.A., Luck, S.N., Sakellaris, H., Rajakumar, K., and Adler, B. (2001) Nested deletions of the SRL pathogenicity island of Shigella flexneri 2a. J Bacteriol 183: 55355543.
  • Ubeda, C., Tormo, M.A., Cucarella, C., Trontonda, P., Foster, T.J., Lasa, I., and Penades, J.R. (2003) Sip, an integrase protein with excision, circularization and integration activities, defines a new family of mobile Staphylococcus aureus pathogenicity islands. Mol Microbiol 49: 193210.
  • Welch, R.A., Burland, V., Plunkett, G., III, Redford, P., Roesch, P., Rasko, D., et al. (2002) Extensive mosaic structure revealed by the complete genome sequence of uropathogenic Escherichia coli. Proc Natl Acad Sci USA 99: 1702017024.
  • Wu, Z., Gumport, R.I., and Gardner, J.F. (1998) Defining the structural and functional roles of the carboxyl region of the bacteriophage lambda excisionase (Xis) protein. J Mol Biol 281: 651661.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. Supplementary material
  8. References
  9. Supporting Information

Table S1. Influence of different growth conditions on HPI excision in strain IP32637sacB. Table S2. Primers used in this study. Fig. S1. A. Amplification of the junction fragment of the HPI circular form by nested PCR using primer pair 229A/229lB. PCR was performed on thermolysates of strain IP32637sacB and IP32637_HPI. B. Sequence of the junction fragment of the circular form obtained by nested PCR. Fig. S2. RT-PCR amplification of int mRNA from different hef recombinant strains of IP32637, using various amounts of RNA templates (numbers above the figure). The strains that served for RNA extraction are indicated on the right of the figure.

FilenameFormatSizeDescription
MMI_4073_sm_tableS1.doc25KSupporting info item
MMI_4073_sm_tableS2.doc34KSupporting info item
MMI_4073_sm_figS1.tif160KSupporting info item
MMI_4073_sm_figS2.tif168KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.