• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of iaiPermissionsJournals.ASM.orgJournalIAI ArticleJournal InfoAuthorsReviewers
Infect Immun. Oct 1999; 67(10): 5091–5099.
PMCID: PMC96857

The High-Pathogenicity Island of Yersinia enterocolitica Ye8081 Undergoes Low-Frequency Deletion but Not Precise Excision, Suggesting Recent Stabilization in the Genome

Editor: P. E. Orndorff

Abstract

Highly pathogenic strains of Yersinia pestis, Y. pseudotuberculosis, and Y. enterocolitica are characterized by the possession of a pathogenicity island designated the high-pathogenicity island (HPI). This 35- to 45-kb island carries an iron uptake system named the yersiniabactin locus. While the HPIs of Y. pestis and Y. pseudotuberculosis are subject to high-frequency spontaneous deletion from the chromosome, we were initially unable to obtain HPI-deleted Y. enterocolitica 1B isolates. In the present study, using a positive selection strategy, we identified three HPI-deleted mutants of Y. enterocolitica strain Ye8081. In these three independent clones, the chromosomal deletion was not limited to the HPI but encompassed a larger DNA fragment of approximately 140 kb. Loss of this fragment, which occurred at a frequency of approximately 5 × 10−7, resulted in the disappearance of several phenotypic traits, such as growth in a minimal medium, hydrolysis of o-nitrophenyl-β-d-thiogalactopyranoside, Tween esterase activity, and motility, and in a decreased virulence for mice. However, no precise excision of the Ye8081 HPI was observed. To gain more insight into the molecular basis for this phenomenon, the putative machinery of HPI excision in Y. enterocolitica was analyzed and compared to that in Y. pseudotuberculosis. We show that the probable reasons for failure of precise excision of the HPI of Y. enterocolitica Ye8081 are (i) the interruption of the P4-like integrase gene located close to its right-hand boundary by a premature stop codon and (ii) lack of conservation of 17-bp att-like sequences at both extremities of the HPI. These mutations may represent a process of HPI stabilization in the species Y. enterocolitica.

Yersinia spp. are gram-negative bacteria belonging to the family Enterobacteriaceae (3). The genus is composed of 11 species which can be divided into three pathogenicity groups. Strains nonpathogenic for humans are found mainly in the environment and sometimes also as transient intestinal saprophytes. These include Yersinia intermedia, Y. frederiksenii, Y. kristensenii, Y. mollaretii, Y. bercovieri, Y. aldovae, Y. rohdei, Y. ruckeri, and Y. enterocolitica biotype 1A (38). Strains with a moderate level of pathogenicity are widespread in cold and temperate countries and are responsible for mild enteric infections in humans (38). They are not lethal for mice at low doses and correspond to Y. enterocolitica strains of biotypes 2 to 5 (14). The group of high-pathogenicity strains is composed of Y. pestis, Y. pseudotuberculosis, and Y. enterocolitica biotype 1B (46), which are all lethal for mice at low doses and cause disseminated infections in humans.

This high level of pathogenicity correlates with the presence of a pathogenicity island (PAI) called the high-pathogenicity island (HPI) in Yersinia, as it differentiates low- and high-pathogenicity strains (10). PAIs are large pieces of chromosomal DNA that carry virulence genes, possibly acquired by phage-mediated horizontal transfer (24). The HPI of Y. enterocolitica 1B (10) is a chromosomal fragment of 45 kb that carries virulence genes known as the yersiniabactin locus that are involved in iron uptake (10, 39). Yersiniabactin is a siderophore (26) that endows the bacterium with the ability to acquire the iron molecules necessary for its in vivo growth and dissemination. Most of the length of the HPI is occupied by the yersiniabactin locus (Fig. (Fig.1),1), which comprises the genes irp1 to irp5, involved in nonribosomal biosynthesis of the siderophore (22, 39), and the outer membrane siderophore receptor gene fyuA (27, 42). The G+C content of these genes is higher than the chromosomal background. The yersiniabactin locus is highly conserved among Y. enterocolitica 1B strains (10) and in the two other high-pathogenicity species, Y. pseudotuberculosis (7) and Y. pestis (2, 8, 20). The region located on the right side of this locus, between irp2 and the right-hand boundary of the Y. enterocolitica HPI (Fig. (Fig.1),1), has not yet been sequenced. However, based on hybridization experiments with Y. pestis probes (8), it is most likely that this region carries genes recently described as composing another part of the yersiniabactin locus in Y. pestis (17). The right-hand border of the Y. enterocolitica HPI is defined by an asn-tRNA gene (10). In contrast, the ≈15-kb region of the Y. enterocolitica HPI extending on the left side of the yersiniabactin locus (Fig. (Fig.1)1) is not well conserved among various isolates of this species (10) and is different from the regions found on the HPIs of the two other high-pathogenicity species (7, 8). The only genes identified until now on this 15-kb region correspond to a cluster of three repeated sequences (IS1400 [10], IS1328 [41], and RS3 [10]) (Fig. (Fig.1).1). The left-hand boundary of the island has not been precisely identified.

FIG. 1
Genetic organization of the HPI of Y. enterocolitica Ye8081. The HPI is represented by the gray line, and the identified genes are represented by arrows. Dashed lines below the diagram represent probes, named on the basis of their size in kilobases and ...

The HPI of Y. enterocolitica 1B thus shows nearly all of the criteria of a typical PAI as defined by Hacker and colleagues (24): (i) it is a large chromosomal DNA fragment, (ii) it carries virulence genes essential for the expression of the high-virulence phenotype, (iii) it harbors several repeated sequences, (iv) it is bordered on one side by a tRNA gene, and (v) its G+C content is different from that of the rest of the chromosome. However, while PAIs are often described as mobile elements, and while the HPIs of Y. pestis (18, 19) and Y. pseudotuberculosis (7) are able to undergo spontaneous deletion from the host chromosome, the Y. enterocolitica HPI seems to be stable. Our previous attempts to identify spontaneous HPI deletion mutants among natural isolates of Y. enterocolitica 1B or following repeated subcultures of high-pathogenicity strain Ye8081 were unsuccessful (16). To date, no precise excision of this island has been reported. The aim of this study was to confirm whether the HPI of Y. enterocolitica was a stable feature of this organism’s genome and, if it was, determine the species-specific mechanism of stabilization.

MATERIALS AND METHODS

Bacterial strains, media, and plasmids.

Study of HPI stability was performed on strain Ye8081, a Y. enterocolitica biotype 1B serotype O:8 strain that was previously used to characterize the PAI of this species (10). Forty-one other strains of Yersinia, including 19 Y. enterocolitica strains of different bioserotypes, 11 Y. pseudotuberculosis strains of serotypes I to V, 3 Y. pestis, 2 Y. bercovieri, 1 Y. frederiksenii, 1 Y. kristensenii, 2 Y. intermedia, 2 Y. mollaretii strains, and 1 Y. aldovae strain were also studied. The Escherichia coli strains used were K-12 strain TG1 [Δ(lac-proAB) supE thi hsdD5/F′ traD36 proA+ proB+ lacIq lacZΔM15] as host for M13 bacteriophage, TOP10F′ [F′{lacIq Tn10 (Tetr)} mcrA Δ(mrr-hsdRMS-mcrBC) [var phi]80lacZΔM15 ΔlacX74 recA1 deoR araD139 Δ(ara-leu)7697 galU galK rpsL (Strr) endA1 nupG] for transformation with plasmid pCRII-TOPO (Invitrogen), and Epicurian Coli XL1-Blue [recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac (F′ proAB lacIq ZΔM15 Tn10 (Tetr)] (Stratagene) for electroporation of pBluescript and pUC19. Yersinia strains were grown at 28°C for 24 h in peptone broth (Difco) or in M9 minimal medium supplemented with 0.4% glycerol, 10 mM CaCl2, 2 mM MgSO4, and 1 μg vitamin B1 per liter or for 48 h on Trypticase soy agar (Diagnostics Pasteur) plates. E. coli strains were grown at 37°C for 24 h in Luria broth (Difco) or on Luria agar plates. When necessary, ampicillin at a concentration of 50 μg/ml was added to the media. The plasmids used were pCRII-TOPO (Invitrogen), pUC19 (Biolabs), and pBluescript II KS+ (Stratagene).

DNA techniques.

Isolation of genomic DNA was performed with an IsoQuick nucleic acid extraction kit (MicroProbe). Plasmid extraction was carried out according to the method of Birnboim and Doly (4). For Southern hybridizations, approximately 10-μg aliquots of DNAs from the various Yersinia strains were digested with EcoRI and subjected to overnight electrophoresis at 50 V in 0.8% agarose gels. DNAs were transferred to nylon membranes (Hybond N+; Amersham) by using a vacuum blotting system (Vacugene XL; Pharmacia) and hybridized with the corresponding probes. Colony blot hybridizations were done on bacterial colonies spotted onto nylon filters and treated as previously described (35). Double-stranded DNA labeling was performed either radioactively with [α-33P]ATP (NEN), using a Megaprime DNA labeling systems kit (Amersham), or nonradioactively with digoxigenin (DIG), either by incorporation of DIG-11-dUTP during PCRs or with the random primed labeling system (Boehringer Mannheim). Plasmid or cosmid insert probes were purified by elution from an agarose gel with a QIAquick gel extraction kit (QIAGEN). Single-stranded oligonucleotides were labeled with DIG by using a DIG-oligonucleotide tailing kit (Boehringer Mannheim). The oligonucleotides used as probes or for probe amplification are listed in Table Table1.1. Hybridizations were carried out at 65°C (DIG-labeled probes) or 68°C (radiolabeled probes) for random primed probes and at 54°C (BG1), 59°C (BG2), or 56°C (att-like) for DIG-labeled oligonucleotides.

TABLE 1
Primers used for PCR amplification or sequencing and oligonucleotides used as probes

PCR.

PCRs were carried out in a 50-μl reaction volume with an OmniGene thermal cycler (Hybaid). Fifty to 100 ng of target DNA, 200 μM each deoxynucleoside triphosphate, 1 μM each primer, and 1 U of Taq polymerase (Perkin-Elmer) or Pfu polymerase (Stratagene) were mixed in 1× polymerase buffer. Amplification involved 30 cycles, each consisting of a denaturation step of 30 s at 92°C, an annealing step of 1 min 30 s at 55°C, and a polymerization step of 1 min 30 s at 72°C, ending with a final cycle at 72°C for 10 min. Primers used for PCR (Table (Table1)1) were designed with the PC-Rare software (21).

Sequencing and sequence analysis.

To sequence the int gene, a 1,872-bp fragment was amplified by PCR with primers int1 and int2 (Table (Table1),1), using the Ye8081 genomic DNA as template, and was subsequently cloned into the EcoRI and BamHI sites of plasmid pUC19. The insert was then ligated into the replicative form of bacteriophages M13mp18 and M13mp19 and sequenced by the dideoxy-chain termination method (47), using a Thermo Sequenase core sequencing kit with 7-deaza-dGTP from Vistra (Amersham) and a Vistra DNA Sequencer 725 (Amersham). A portion of the int genes of Y. enterocolitica strains IP845 and IP24764 was sequenced by amplification of a 336-bp fragment of their genomic DNAs by nested PCR with primers int1 and int2 for the first round and primers intC and intD for the second round (Table (Table1).1). The amplified fragments were then cloned into vector pCRII-TOPO (Invitrogen) and sequenced with a Cy5-AutoRead sequencing kit (Pharmacia), using the pgemD forward and M13 reverse primers and an annealing temperature of 65°C for 10 min, followed by incubation at 37°C for 10 min before cooling at room temperature for 10 min. An ALF express automatic sequencer (Pharmacia) was used for sequencing reactions. To sequence the left border of the Ye8081 HPI, the 0.7-kb BglII-XhoI fragment (BgX0.7 [Fig. 1]) that has previously been shown to correspond to the limit of the HPI (10) was cloned into the corresponding sites of pBluescript. Sequencing was performed by the dideoxy-chain termination method with Sequenase (United States Biochemical Corp.). Sequence analysis was performed with Assembly Lign, BLAST N, BLAST P (1), and Clustal V (28).

Pulsed-field gel electrophoresis.

Genomic DNA was prepared in agarose plugs as previously described (23). After digestion with SpeI, NotI, XbaI, or SrfI, the macrorestriction fragments were resolved with a CHEF-DRIII apparatus (Bio-Rad Laboratories), using an electric field of 6 V/cm and an angle of 120°. Migration of the DNA fragments was performed in 0.5× Tris-borate-EDTA buffer and in 0.9% agarose gels at 17°C. Pulse times varied according to the size of the fragments to be resolved.

Pesticin production and selection of pesticin-resistant colonies.

Pesticin was prepared according to the method of Toora et al. (51). Induction of pesticin production by Y. pestis EV76 was obtained by incubating the strains for 19 h in the presence of 0.75 μg of mitomycin C (Sigma) per ml. Y. enterocolitica pesticin-resistant colonies were detected as previously described (7).

Bacterial motility testing.

Motility of the parental strain Y. enterocolitica Ye8081 and of the three isogenic mutants Ye8081H-.1, Ye8081H-.2, and Ye8081H-.3 was tested on tryptone swarm plates (1% Bacto Tryptone, 0.5% NaCl, 0.3% Bacto Agar). Approximately 5 × 105 bacteria were spotted onto these plates and incubated at 25 or 37°C for 24 h. Staining of flagella was performed as described elsewhere (45).

Biochemical tests.

Phenotypic characterization of Yersinia strains was performed with API 20E and API 50CH strips (bioMérieux, Marcy l’Etoile, France). Hydrolysis of o-nitrophenyl-β-galactopyranoside (ONPG) was monitored by suspending bacteria grown on agar plates in a tube containing distilled water and an ONPG disc (Sanofi Diagnostics Pasteur, Paris, France) and incubating them for up to 24 h at 37°C in a water bath (33). Presence of pyrazinamidase was evaluated as previously described (29). DNase activity was determined by spotting the bacteria on DNase agar media (Diagnostics Pasteur) and incubating them at 28°C for 48 h (6). Presence of DNase was characterized by a translucent zone after washing with 1 N HCl solution. To test for Tween esterase activity, bacterial cells were spotted onto Tween 80 plates (containing, per liter, 10 g of pancreatic peptone [Prolabo], 5 g of NaCl, 0.1 g of CaCl2, 4.63 ml of Tween 80 [Merck], and 12 g of Bacto Agar [Difco] [pH 7.4]). After incubation at 28°C for 24 h, a cloudy zone indicated the presence of the enzyme.

Animal infection and estimation of the LD50.

To measure the median lethal dose (LD50) of the wild-type strain Ye8081 and of the Ye8081H-.1 mutant derivative, 0.5-ml aliquots of 10-fold serial dilutions of bacterial suspensions were inoculated intravenously into 16-day-old OF1 female mice (Iffa Credo, Lyon, France). Groups of five mice were used for each dilution, and five dilutions were used each time. In some experiments, mice were iron overloaded by intraperitoneal injection of 0.5 ml of saline containing 1 mg of ferric chloride prior to the bacterial challenge. LD50s were calculated by the method of Reed and Muench (43).

Nucleotide sequence accession numbers.

The sequences of the int gene and of the left border of the HPI have been deposited in the EMBL databank under accession no. AJ010281 and AJ238284, respectively.

RESULTS

Identification of deletion derivatives of Y. enterocolitica Ye8081.

The absence of HPI deletion strains among natural isolates of Y. enterocolitica 1B (16) and our failure to obtain deletion of the island upon serial subcultures in vitro suggested that if the excision of the HPI does occur in Y. enterocolitica, it should be at a much lower frequency than in Y. pseudotuberculosis (7). Thus, in order to observe rare deletion mutants, we designed a positive selection procedure based on the fact that the yersiniabactin receptor FyuA which is located on the HPI is also the receptor for the bacteriocin pesticin (42). Pesticin acts like lysozyme in hydrolyzing murein lipoproteins on the bacterial cell wall (17). Colonies deleted of the pesticin receptor should become resistant to the action of this molecule. Upon treatment of strain Ye8081 with crude extracts of pesticin, 800 colonies that grew in the presence of the bacteriocin were collected, spotted on nylon membranes, and hybridized with two PCR-amplified probes (fyuA and irp2) encompassing the HPI-borne fyuA or irp2 gene (Fig. (Fig.1).1). Of the 800 pesticin-resistant colonies tested, 797 hybridized with both probes, indicating the presence of at least a portion of the HPI in these strains and suggesting that a mutation in the pesticin receptor may have been responsible for their resistance to the bacteriocin. The three remaining colonies, designated Ye8081H-.1, Ye8081H-.2, and Ye8081H-.3 and obtained from three independent platings, failed to hybridize with either the fyuA or the irp2 probe. Southern hybridization of the EcoRI-digested DNA of the three mutants and of the parental strain Ye8081 with the Cl8 and E3.5 probes (Fig. (Fig.1)1) derived from cosmid clones pHC25 and pHC15 (10) confirmed the absence of these two DNA regions in the three mutant strains. Based on the estimated number of bacteria per bacteriocin-induced lysis plaque, the frequency of deleted clones of strain Ye8081 was estimated to be 5 × 10−7.

To determine the extent of the deletion that encompasses the fyuA and irp2 genes, eight cosmid-derived probes (BgX0.7, Bg7, EBa1.2, E3.5, E8.8, E6, E9, and ClBa1.2 [Fig. 1]) covering the HPI with the exclusion of the repeated elements (IS1400, RS3, and IS1328), were used to hybridize with the EcoRI-digested DNA of strain Ye8081 and of the three mutants. None of these eight probes hybridized with Ye8081H-.1, Ye8081H-.2, or Ye8081H-.3 DNA. Moreover, the two probes located at each extremity of the HPI (HB12.5 and E18 [Fig. 1]) and encompassing the flanking chromosomal regions did not hybridize with the DNAs of the mutant strains, indicating that the unstable regions do not correspond precisely to the HPI but extend beyond the island on both sides and cover a region at least 67 kb long.

Analysis of the genomic profiles of strain Ye8081 by pulsed-field gel electrophoresis after digestion with SpeI, NotI, XbaI, or SrfI showed that several restriction fragments differed between the parental strain and the three mutants. In contrast, the genomic profiles of these three independent deletion mutants were identical, whatever the enzyme used (data not shown), suggesting that the same chromosomal fragment was removed in the three clones. Moreover, with the SrfI enzyme, which gives a small number of restriction fragments, it was possible to show that the genomic profile of the wild-type strain Ye8081 differed from that of the three isogenic deletion derivatives by the presence of two fragments of approximately 220 and 390 kb and by the absence of a ≈470-kb fragment (Fig. (Fig.2A).2A). Both the HB12.5 and E18 probes located at each extremity of the island hybridized with the 390-kb SrfI fragment in strain Ye8081 but with no fragment in the deletion derivatives (Fig. (Fig.2B).2B). By subtracting the sizes of these fragments, it was estimated that the region deleted in Ye8081H-.1, Ye8081H-.2, and Ye8081H-.3 encompassed approximately 140 kb of chromosomal DNA.

FIG. 2
Comparison of genomic restriction profiles of the parental strain Ye8081 and of the three deletion derivatives (Ye8081H-). (A) Pulsed-field gel electrophoresis of SrfI-digested genomic DNA. M, molecular weight markers, whose sizes (in kilobases; values ...

Deletion of the 140-kb unstable fragment is accompanied by the loss of several phenotypic traits.

Deletion of the 140-kb unstable fragment resulted in complete removal of the HPI and therefore the inability of the mutants to synthesize yersiniabactin (2, 39). Analysis of the impact of the HPI deletion on other phenotypic and metabolic properties showed that the three deletion derivatives differed from the parental strain Ye8081 by the alteration of four characters. First, while the parental strain and the three mutants grew equally well in a rich medium (peptone broth), the deletion derivatives exhibited defective growth in a minimal medium (M9). Second, the ability to hydrolyze ONPG was abolished in the three mutants. In the genus Yersinia, hydrolysis of ONPG is not due to a β-galactosidase sensu stricto but to an ONPG-hydrolase (38). Third, all three deletion mutants were impaired in the ability to hydrolyze Tween 80. This Tween esterase activity is due to a phospholipase A enzyme whose genes (yplA and ylpB) have been recently identified in strain Ye8081 (48). To determine whether this locus was present in the deletion clones, primers YplA1 and YplB2, located at the 5′ extremity of ylpA and at the 3′ end of ylpB, respectively, were used to amplify the entire locus by PCR. Fragments of the same size were obtained in the parental strain and the three mutants, indicating that the deletion did not remove this locus and suggesting that other elements necessary for the expression of a Tween esterase activity were altered. The fourth character lost by the mutants was motility. The genus Yersinia is characterized by the ability of the bacteria to be motile at 28°C but not at 37°C (except Y. pestis, which is always nonmotile) (38). As expected, both the wild-type strain and the mutant derivatives were motionless on swarm plates incubated at 37°C, but only the wild-type strain gained motility upon incubation at 25°C (Fig. (Fig.3A).3A). Flagellum staining of 25°C-grown bacteria demonstrated that the deletion strains had lost the capacity to produce these motility elements (Fig. (Fig.3B).3B).

FIG. 3
Motility of the parental strain Ye8081 and of the deletion derivatives (Ye8081H-). (A) Swarm plates inoculated with the parental strain and one isogenic deletion mutant and incubated at 37 or 25°C. (B) Flagellum staining of the parental strain ...

Therefore, our data demonstrate that the 140-kb unstable fragment carries, in addition to the HPI, genes involved directly or indirectly in growth in a minimal medium, in ONPG and Tween 80 hydrolysis, and in expression of flagella at the surface of the bacteria.

Impact of deletion of the 140-kb unstable fragment on the virulence of the mutant strain.

To evaluate the effect of the 140-kb chromosomal deletion on the virulence of Y. enterocolitica, the LD50s of Ye8081 and Ye8081H-.1 were estimated after intravenous injection into mice. The presence of the virulence plasmid pYV was checked prior to animal infection. The LD50 of the mutant strain (>106 CFU) was more than 7,000 times greater than that of the parental strain (1.4 × 102 CFU). The decreased virulence of the mutant strain can be at least partly attributed to the deletion of the HPI, which is known to be important for Yersinia pathogenicity (2, 11, 42). A decrease in virulence was also observed when the Y. enterocolitica yersiniabactin receptor gene fyuA was mutagenized (42). However, the participation of the adjacent unstable chromosomal regions in this process could not be ruled out. High-pathogenicity Y. enterocolitica strains of biotype 1B are inherently lethal for laboratory animals, while low-pathogenicity strains of biotypes 2 to 5 are naturally devoid of the HPI and do not kill mice at low doses. Robins-Browne and Kaya Prpic (46) demonstrated that injection of the animals with iron and/or an exogenous siderophore (Desferal) prior to the bacterial challenge did not significantly change the virulence of the high-pathogenicity strains but increased the virulence of the strains of biotypes 2 to 5, by providing them with the iron molecules necessary for their in vivo growth and dissemination. To determine whether the decreased virulence of the mutant strains was essentially due to the loss of the iron-chelating system carried by the HPI, or whether the adjacent unstable regions could also be involved, mice were treated with ferric chloride before bacterial infection. The virulence of the high-pathogenicity strain Ye8081 (LD50 of 3.1 × 102 CFU) was not modified by the iron treatment. In contrast, the pathogenicity of the mutant strain Ye8081H-.1 increased (to an LD50 of 2 × 105 CFU) upon iron treatment but remained significantly lower than that of the parental strain. Although the detrimental effect of iron on the mouse immune response may have contributed to the decreased LD50, these results suggest that the reduced virulence observed upon deletion of the 140-kb chromosomal region is partly attributable to the loss of the HPI but that other genes located on this unstable region also play a role in the virulence of strain Ye8081. It has recently been shown that mutagenesis of the phospholipase A locus of strain Ye8081 alters the virulence of the mutant strains for mice (48). It thus likely that the abolition of the Tween 80 activity observed in our deletion mutants may have contributed to their reduced virulence.

Altogether our data demonstrate that a spontaneous 140-kb deletion which carries the HPI and genes involved in ONPG hydrolysis, Tween 80 esterase activity, motility, and pathogenicity can occur in the chromosome of strain Ye8081. However, our data also indicate that, in contrast to the HPI of Y. pseudotuberculosis (7), the homologous island of Y. enterocolitica 1B does not undergo precise excision.

The P4-like integrase gene is mutated in Y. enterocolitica 1B strains.

The HPI of Y. pseudotuberculosis, which is capable of precise excision from the host chromosome (7), contains a gene homologous to the bacteriophage P4 integrase gene (int) (7, 25). An integrase gene almost identical to that of Y. pseudotuberculosis was identified at the same position on the HPI of Y. pestis (9). To determine whether the HPI of Y. enterocolitica 1B carries a similar int gene, the segment of the HPI located immediately downstream of the asn-tRNA locus of strain Ye8081 (Fig. (Fig.1)1) was amplified by PCR, cloned into pUC19 and bacteriophage M13, and sequenced. A 1,260-bp-long sequence homologous to the bacteriophage P4 integrase (40) gene (57.5% identity over 1,260 bp) and highly similar to the int gene of Y. pseudotuberculosis (98.8% identity over 1,260 bp) was identified. The G+C content of this sequence was 52.9% and therefore, as is the case for other HPI-borne genes, much higher than the average G+C content of the host chromosome (46 to 48%). However, while the Y. pseudotuberculosis int sequence had the capacity to code for a 420-amino-acid (aa) protein (7), the Ye8081 sequence could encode only a 138-aa-long peptide because a T residue instead of a G at position 415 of the nucleotide sequence introduced a premature stop codon. To ensure that this stop codon was not an artifact generated during PCR amplification and to determine whether a truncated int gene was characteristic of all Y. enterocolitica 1B strains, the integrase region surrounding the premature stop codon was amplified and sequenced for two other Y. enterocolitica strains of biotype 1B: IP845 (serotype O:20) and IP24764 (serotype O:13 [20]). For this purpose, nested PCR with primer pairs int1-int2 and intC-intD followed by cloning into pCRII-TOPO was carried out. The 336-bp amplified sequences of the two strains were identical to the Ye8081 sequence and exhibited the same stop codon interrupting the open reading frame (ORF). These results suggest that the int gene is commonly mutated in Y. enterocolitica 1B strains. Without this premature stop codon, the Y. enterocolitica int gene would have had the potential to code for a 420-aa integrase protein with 98.6% identity and 99.3% similarity to its Y. pseudotuberculosis homologue and with 49.5% identity and 77.6% similarity to the bacteriophage P4 integrase.

To analyze the distribution of the int gene within the genus Yersinia, a 712-bp probe generated by PCR with primers BP3 and BP4 (Table (Table1)1) was used to hybridize with the EcoRI-digested DNAs of various pathogenic and nonpathogenic species of Yersinia (see Materials and Methods). A perfect correlation was observed between the presence of the HPI and the int gene; i.e., the int gene was found in high-pathogenicity strains only (data not shown). Furthermore, under the stringent hybridization conditions used in our study, only one fragment hybridized with the probe, indicating that a single copy of the P4-like integrase gene is present on the chromosome of high-pathogenicity strains.

Our results thus demonstrate that although a well-conserved integrase gene is present on the HPI of Y. enterocolitica 1B, this gene is interrupted and therefore probably nonfunctional in this group of bacteria.

The 17-bp repeat located at the left-hand border of the HPI of Y. enterocolitica Ye8081 is degenerate.

In Y. pseudotuberculosis (7, 25) and in Y. pestis (9), a 17-bp-long sequence resembling the attP site of bacteriophage P4 was identified in the same orientation at the two extremities of the island. Excision of the Y. pseudotuberculosis HPI probably occurs by recombination between the two copies of the 17-bp att-like repeats and results in the generation of a fusion fragment containing the 17-bp sequence at the junction site (7). Similar att-like repeats at the borders of the Ye8081 HPI were sought. Analysis of the previously published sequence of the asn-tRNA locus bordering the right-hand extremity of the HPI of strain Ye8081 (10) revealed the presence of a 17-bp sequence (5′-CCAGTCAGAGGAGCCAA-3′) identical to the HPI flanking sequence in Y. pseudotuberculosis and Y. pestis. To determine whether this sequence was also present at the other extremity of the Ye8081 HPI, the DNAs of cosmid clones pHC25 and pHC15 (Fig. (Fig.1)1) were subjected to single and/or double digestion with EcoRI, BamHI, BglII, and ClaI and hybridized with a 12-nucleotide (nt)-long oligonucleotide (Table (Table1)1) covering a portion of the att-like site. In cosmid pHC15, one hybridizing fragment having the expected size and corresponding to the cosmid segment carrying the asn-tRNA gene was recognized for each digestion (data not shown). In contrast, no restriction fragment from cosmid pHC25 hybridized with the probe, indicating that the portion of the 17-bp att-like site present on the probe was either absent or degenerate at the left-hand extremity of the HPI.

To further investigate this point, the 0.7-kb BglII-XhoI region previously defined by hybridization as containing the left-hand border of the HPI (10) was cloned into plasmid pBluescript and sequenced. Two partial ORFs designated ORF1 and ORF2 were identified (Fig. (Fig.4).4). The predicted 66-aa product of ORF1 was similar (35% identity and 59% similarity over 62 aa) to the C-terminal part of the fructose-specific IIA/FPR component of the phosphoenolpyruvate-dependent sugar phosphotransferase system of Haemophilus influenzae (44) and of other bacterial species such as Salmonella typhimurium and E. coli. ORF2 had the capacity to encode the first 121 aa of a protein highly similar (87.6% identity and 97.5% similarity over the entire 121 aa) to the N-terminal portion of a hypothetical 19.9-kDa protein of E. coli, whose gene is located in the mhpT-adhC intergenic region (5). When the 0.7-kb BglII-XhoI fragment which contains these two partial ORFs was hybridized with the genomic DNAs of Y. pestis and Y. pseudotuberculosis, no fragment was recognized, indicating that this region is not conserved in the two other pathogenic species of Yersinia. A FASTA search of a nucleotide motif corresponding to the 17-bp att-like site on the 707-bp BglII-XhoI sequenced fragment identified a region located between ORF1 and ORF2 and containing a short stretch of 6 contiguous nt (5′-AGTCAG-3′) identical to those of the att-like site and a total of 9 identical nt over the 17 nt (Fig. (Fig.4).4). This moderate level of identity may indicate either that this region of partial identity is not the border of the HPI or that this border contains a degenerate att-like site. To determine whether this region was indeed the left-hand boundary of the HPI, two oligonucleotides (BG1 and BG2) flanking the putative att-like site (Fig. (Fig.4)4) were hybridized with the EcoRI-digested genomic DNA of the high-pathogenicity strain Ye8081 and of the low-pathogenicity strains Y. enterocolitica IP864 (biotype 4 and serotype O:3) and IP383 (biotype 2 and serotype O:9). The BG1 probe hybridized with the DNA of both the high- and low-pathogenicity strains, while BG2 hybridized only with the DNA of the high-pathogenicity strain (Fig. (Fig.4).4). Since the HPI has been defined as the portion of DNA encompassing the yersiniabactin iron uptake machinery which is found uniquely in high-pathogenicity strains (10), it can be concluded from these results that the boundary of the Y. enterocolitica HPI is located between BG1 and BG2 and therefore that it most likely corresponds to the degenerate att-like site identified (as HPI excision is disabled in Y. enterocolitica, experimental evidence for this is not available). Our data also indicate that a protein highly similar to an as yet uncharacterized protein of E. coli is encoded by the left-hand portion of the HPI.

FIG. 4
(A) Schematic representation of the 707-bp BglII-XhoI region encompassing the left-hand boundary of the Ye8081 HPI and nucleotide sequence of the region where the degenerate att-like site was identified. (B) Hybridization of the EcoRI-digested genomic ...

These results thus demonstrate that in contrast to the HPI of Y. pseudotuberculosis and Y. pestis, the HPI of Y. enterocolitica Ye8081 contains only one conserved 17-bp att-like site at its right-hand boundary and no identical sequence at its left-hand boundary.

DISCUSSION

A previous analysis of various Yersinia species demonstrated that the HPI-borne irp2 gene was unique to high-pathogenicity strains (16). Within this group of bacteria, we showed that some isolates of Y. pestis and Y. pseudotuberculosis had lost this gene and had a reduced virulence in the mouse model. This was not the case for Y. enterocolitica 1B strains, for which no irp2-deleted natural isolates were observed (16). Equally, attempts to obtain irp2-deletion mutants of strain Ye8081 by repeated subcultures, UV irradiation, or treatment with mitomycin C in search of irp2-non-hybridizing clones were unsuccessful (9a). This led us to conclude that the HPI of Y. enterocolitica is more stable than the islands of the two other high-pathogenicity species. The possibility of a low-frequency HPI deletion in Ye8081 could nonetheless not be eliminated. Using another strain of Y. enterocolitica 1B (WA), Rakin and Heesemann (41) identified some colonies with deletions of both fyuA and irp2 or of fyuA alone (but not irp2 alone). The former mutants may have represented HPI deletions, but they were not characterized. To screen a high number of Ye8081 colonies for the loss of the HPI, we developed a positive selection strategy based on the dual function of FyuA as a receptor for both yersiniabactin and the bacteriocin pesticin (27, 42). Using this strategy, we were able to obtain three Ye8081 mutants lacking an approximately 140-kb fragment encompassing the whole of the HPI and adjacent chromosomal sequences. The fact that the genomic profiles of the three independently obtained clones were identical suggests that they shared a common mechanism of deletion. Characterization of this mechanism was beyond the scope of this study but deserves further investigation.

Surprisingly, deletion of the HPI appears to result from different mechanisms in the three highly pathogenic Yersinia species. In Y. pestis, this deletion occurs at a frequency of 10−5 by homologous recombination between two IS100 copies flanking a large 102-kb chromosomal region named the pgm locus (18, 19). In Y. pseudotuberculosis, the HPI excises precisely from the chromosome at a frequency of ≈10−4, probably by site-specific recombination (7). In Y. enterocolitica 1B, we found that the HPI is lost a frequency of ≈5 × 10−7 by deletion of a 140-kb unstable chromosomal fragment. Despite these apparent differences, the three HPIs superficially appear to have the same genetic apparatus for site-specific recombination, i.e., a site-specific recombinase (the P4-like integrase) encoded within the excisable DNA downstream of the asn-tRNA locus and short flanking direct repeat sequences (the 17-bp P4 att-like sites), representing the target sites for the recombinase. However, while the int genes of Y. pseudotuberculosis (7, 25) and Y. pestis (9) are intact and presumably have the capacity to code for a potentially functional integrase protein, the int gene of Y. enterocolitica 1B is prematurely interrupted and one of the 17-bp flanking regions is incompletely conserved. Whether a functional integrase is required for PAI excision is not clear. In the case of uropathogenic E. coli, both PAI I and PAI II have the capacity to excise from the chromosome upon recombination between direct flanking repeats (6) despite the fact that the integrase genes present on both PAIs are cryptic (24). On the other hand, the mobile PAI of Staphylococcus aureus (34) and the symbiosis island of rhizobia (50) both carry an intact integrase gene adjacent to one boundary of the island, closely resembling the Y. pseudotuberculosis HPI, which is also mobile (7). Furthermore, it has been shown that increased expression of slpA, a gene closely related to P4 int, leads to the excision of the cryptic P4-like prophage in E. coli (31). The participation of the integrase gene in the PAI excision process thus appears likely but remains to be demonstrated.

Site-specific recombination also requires a target DNA sequence specifically recognized by the recombinase. In the PAIs of several bacterial species, short direct repeats, often homologous to phage attachment sites, are present at each extremity. This is the case for the PAIs of Helicobacter pylori (12), H. influenzae (36, 37), Vibrio cholerae (30, 32), S. aureus (34), Dichelobacter nodosus (13), and rhizobia (50) among others. The sizes of these repeats vary between 135 and 16 nt. Interestingly, PAIs that have retained the capacity to excise precisely from the chromosome have similar-size repeats at their boundaries: 17 nt for the PAI of S. aureus, 17 nt for the symbiosis island of rhizobia, 16 nt for PAI I of E. coli, 18 nt for PAI II of E. coli, and 17 nt for the HPI of Y. pseudotuberculosis. Therefore, the conservation and the size of these repeats may be crucial for precise excision of the intervening DNA. In the case of Y. enterocolitica 1B, the putative att-like site situated at the left-hand boundary is degenerate and shares only 6-nt perfect identity with the right-hand repeat. This region of perfect identity is probably too short to function efficiently as the target site for a recombinase.

The capacities of precise excision of the HPI of Y. pseudotuberculosis and Y. pestis may not differ drastically. In Y. pseudotuberculosis, the HPI can precisely excise at the att-like sites and possess a potentially functional int gene. In Y. pestis, two perfectly conserved 17-nt flanking att-like sites and an intact P4-like integrase gene are also present on the HPI (9), suggesting that this island has retained the potential for precise excision. Absence of detection of Y. pestis strains deleted precisely of the 35-kb HPI could be attributable to a higher frequency of deletion of the entire 102-kb unstable pgm locus compared to the frequency of excision of the HPI alone. In contrast to Y. pestis and Y. pseudotuberculosis, the HPI of Y. enterocolitica 1B carries mutations in the excision apparatus that probably account for its inability to excise precisely from the chromosome.

Therefore, this study demonstrates that the Y. enterocolitica HPI is located on a 140-kb unstable chromosomal fragment that can delete at a low frequency. However, this HPI has lost the capacity to excise precisely from the chromosome, probably because both the excision protein and its target site are now defective. To our knowledge, the Yersinia HPI is the only PAI found in various species within the same genus, and in other bacterial genera such as Klebsiella (15), Citrobacter (15), and E. coli (49), suggesting horizontal dissemination. We hypothesize that the differences we observe between the HPI in different species of Yersinia represent early steps in the evolution of this PAI. Progressive mutations in the excision machinery of the HPI are the first prerequisite for permanent acquisition of the PAI-specific virulence factors by the host genome. These mutations may be advantageous for the bacteria by facilitating their dissemination into mammalian hosts, but they are disadvantageous for the horizontal spread of the element among different bacterial groups. They would be expected to occur long before progressive adjustment of the island’s G+C content to that of the chromosome and the disappearance of the scars that signal its probable acquisition by a bacteriophage.

ACKNOWLEDGMENTS

We thank Françoise Guinet for fruitful discussion and critical reading of the manuscript, Alexander Pym and Viviane Chenal for advice, Roland Brosch for help with sequence analysis, and Martine Kiredjian, Adina Candrea, and François Bimet for sharing their experience in flagellum staining.

S. Bach received a grant from the Ministère de l’Education Nationale, de la Recherche et de la Technologie. M. Prentice acknowledges support from the Trustees of the Joint Research Board of St. Bartholomew’s Hospital. C. Buchreiser received a grant from the Austrian Program for Advanced Research and Technology.

REFERENCES

1. Altschul S F, Gish W, Miller W, Myers E W, Lipman D J. Basic local alignment search tool. J Mol Biol. 1990;215:403–410. [PubMed]
2. Bearden S W, Fetherston J D, Perry R D. Genetic organization of the yersiniabactin biosynthetic region and construction of avirulent mutants in Yersinia pestis. Infect Immun. 1997;65:1659–1668. [PMC free article] [PubMed]
3. Bercovier H, Mollaret H H. Genus XIV: Yersinia Van Loghem 1944, 15AL. In: Krieg N R, Holt J G, editors. Bergey’s manual of systematic bacteriology. Vol. 1. Baltimore, Md: Williams & Wilkins; 1984. pp. 498–506.
4. Birnboim H C, Doly J. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 1979;7:1513–1523. [PMC free article] [PubMed]
5. Blattner F R, Plunkett III G, Bloch C A, Perna N T, Burland V, Riley M, Collado-Vides J, Glasner J D, Rode C K, Mayhew G F, Gregor J, Davis N W, Kirkpatrick H A, Goeden M A, Rose D J, Mau B, Shao Y. The complete genome sequence of Escherichia coli K-12. Science. 1997;277:1453–1474. [PubMed]
6. Blum G, Ott M, Lischewski A, Ritter A, Imrich H, Tschäpe H, Hacker J. Excision of large DNA regions termed pathogenicity islands from tRNA-specific loci in the chromosome of an Escherichia coli wild-type pathogen. Infect Immun. 1994;62:606–614. [PMC free article] [PubMed]
7. Buchrieser C, Brosch R, Bach S, Guiyoule A, Carniel E. The high-pathogenicity island of Yersinia pseudotuberculosis can be inserted into any of the three chromosomal asn tRNA genes. Mol Microbiol. 1998;30:965–978. [PubMed]
8. Buchrieser C, Prentice M, Carniel E. The 102-kilobase unstable region of Yersinia pestis comprises a high-pathogenicity island linked to a pigmentation segment which undergoes internal rearrangement. J Bacteriol. 1998;180:2321–2329. [PMC free article] [PubMed]
9. Buchrieser C, Rusniok C, Frangeul L, Couve E, Billault A, Kunst F, Carniel E, Glaser P. The 102-kb pgm locus of Yersinia pestis: sequence analyses and comparison of selected regions among different Y. pestis and Yersinia pseudotuberculosis strains. Infect Immun. 1999;67:4851–4861. [PMC free article] [PubMed]
9a. Carniel, E. Unpublished data.
10. Carniel E, Guilvout I, Prentice M. Characterization of a large chromosomal “high-pathogenicity island” in biotype 1B Yersinia enterocolitica. J Bacteriol. 1996;178:6743–6751. [PMC free article] [PubMed]
11. Carniel E, Guiyoule A, Guilvout I, Mercereau-Puijalon O. Molecular cloning, iron-regulation and mutagenesis of the irp2 gene encoding HMWP2, a protein specific for the highly pathogenic Yersinia. Mol Microbiol. 1992;6:379–388. [PubMed]
12. Censini S, Lange C, Xiang Z, Crabtree J E, Ghiara P, Borodovsky M, Rappuoli R, Covacci A. cag, a pathogenicity island of Helicobacter pylori, encodes type I-specific and disease-associated virulence factors. Proc Natl Acad Sci USA. 1996;93:14648–14653. [PMC free article] [PubMed]
13. Cheetham B F, Katz M E. A role for bacteriophages in the evolution and transfer of bacterial virulence determinants. Mol Microbiol. 1995;18:201–208. [PubMed]
14. Cornelis G, Laroche Y, Balligand G, Sory M-P, Wauters G. Yersinia enterocolitica, a primary model for bacterial invasiveness. Rev Infect Dis. 1987;9:64–87. [PubMed]
15. De Almeida A. Etude de différents marqueurs, et plus particulièrement du gène irp2, chez Yersinia pestis. Ph.D. thesis. Paris, France: Paris VII University; 1993.
16. De Almeida A M P, Guiyoule A, Guilvout I, Iteman I, Baranton G, Carniel E. Chromosomal irp2 gene in Yersinia: distribution, expression, deletion and impact on virulence. Microb Pathog. 1993;14:9–21. [PubMed]
17. Ferber D M, Brubaker R R. Mode of action of pesticin: N-acetylglucosaminidase activity. J Bacteriol. 1979;139:495–501. [PMC free article] [PubMed]
18. Fetherston J D, Perry R D. 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. 1994;13:697–708. [PubMed]
19. Fetherston J D, Schuetze P, Perry R D. 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. 1992;6:2693–2704. [PubMed]
20. Gehring A M, Demoll E, Fetherston J D, Mori I, Mayhew G F, Blattner F R, Walsh C T, Perry R D. Iron acquisition in plague—modular logic in enzymatic biogenesis of yersiniabactin by Yersinia pestis. Chem Biol. 1998;5:573–586. [PubMed]
21. Griffais R, André P M, Thibon M. K-tuple frequency in the human genome and polymerase chain reaction. Nucleic Acids Res. 1991;14:3887–3891. [PMC free article] [PubMed]
22. Guilvout I, Mercereau-Puijalon O, Bonnefoy S, Pugsley A P, Carniel E. High-molecular-weight protein 2 of Yersinia enterocolitica is homologous to AngR of Vibrio anguillarum and belongs to a family of proteins involved in nonribosomal peptide synthesis. J Bacteriol. 1993;175:5488–5504. [PMC free article] [PubMed]
23. Guiyoule A, Rasoamanana B, Buchrieser C, Michel P, Chanteau S, Carniel E. Recent emergence of new variants of Yersinia pestis in Madagascar. J Clin Microbiol. 1997;35:2826–2833. [PMC free article] [PubMed]
24. Hacker J, Blum-Oehler G, Mühldorfer I, Tschäpe H. Pathogenicity islands of virulent bacteria: structure, function and impact on microbial evolution. Mol Microbiol. 1997;23:1089–1097. [PubMed]
25. Hare J M, Wagner A K, McDonough K A. Independent acquisition and insertion into different chromosomal locations of the same pathogenicity island in Yersinia pestis and Yersinia pseudotuberculosis. Mol Microbiol. 1999;31:291–303. [PubMed]
26. Heesemann J. Chromosomal-encoded siderophores are required for mouse virulence of enteropathogenic Yersinia species. FEMS Microbiol Lett. 1987;48:229–233.
27. Heesemann J, Hantke K, Vocke T, Saken E, Rakin A, Stojiljkovic I, Berner R. Virulence of Yersinia enterocolitica is closely associated with siderophore production, expression of an iron-repressible outer membrane polypeptide of 65000 Da and pesticin sensitivity. Mol Microbiol. 1993;8:397–408. [PubMed]
28. Higgins D G, Sharp P M. CLUSTAL: a package for performing multiple sequence alignment on a microcomputer. Gene. 1988;73:237–244. [PubMed]
29. Kandolo K, Wauters G. Pyrazinamidase activity in Yersinia enterocolitica and related organisms. J Clin Microbiol. 1985;21:980–982. [PMC free article] [PubMed]
30. Karaolis D K, Johnson J A, Bailey C C, Boedeker E C, Kaper J B, Reeves P R. A Vibrio cholerae pathogenicity island associated with epidemic and pandemic strains. Proc Natl Acad Sci USA. 1998;95:3134–3139. [PMC free article] [PubMed]
31. Kirby J E, Trempy J E, Gottesman S. Excision of a P4-like cryptic prophage leads to Alp protease expression in Escherichia coli. J Bacteriol. 1994;176:2068–2081. [PMC free article] [PubMed]
32. Kovach M E, Shaffer M D, Peterson K M. A putative integrase gene defines the distal end of a large cluster of ToxR-regulated colonization genes in Vibrio cholerae. Microbiology. 1996;142:2165–2174. [PubMed]
33. Le Minor L, Ben Hamida F. Avantages de la recherche de la bêta-galactosidase sur celle de la fermentation du lactose en milieu complexe dans le diagnostic bactériologique, en particulier des Enterobacteriaceae. Ann Inst Pasteur. 1962;102:267–277.
34. Lindsay J A, Ruzin A, Ross H F, Kurepina N, Novick R P. The gene for toxic shock toxin is carried by a family of mobile pathogenicity islands in Staphylococcus aureus. Mol Microbiol. 1998;29:527–543. [PubMed]
35. Maniatis T, Fritsch E F, Sambrook J. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory; 1989.
36. Martin K, Morlin G, Smith A, Nordyke A, Eisenstark A, Golomb M. The tryptophanase gene cluster of Haemophilus influenzae type b: evidence for horizontal gene transfer. J Bacteriol. 1998;180:107–118. [PMC free article] [PubMed]
37. Mhlanga-Mutangadura T, Morlin G, Smith A L, Eisenstark A, Golomb M. Evolution of the major pilus gene cluster of Haemophilus influenzae. J Bacteriol. 1998;180:4693–703. [PMC free article] [PubMed]
38. Mollaret H H, Bercovier H, Alonso J M. Yersinia. In: Le Minor L, Véron M, editors. Bactériologie médicale. Paris, France: Flammarion; 1989. pp. 451–472.
39. Pelludat C, Rakin A, Jacobi C A, Schubert S, Heesemann J. The yersiniabactin biosynthetic gene cluster of Yersinia enterocolitica: organization and siderophore-dependent regulation. J Bacteriol. 1998;180:538–546. [PMC free article] [PubMed]
40. Pierson L S, III, Kahn M L. Integration of satellite bacteriophage P4 in Escherichia coli. DNA sequences of the phage and host regions involved in site-specific recombination. J Mol Biol. 1987;196:487–496. [PubMed]
41. Rakin A, Heesemann J. Virulence-associated fyuA/irp2 gene cluster of Yersinia enterocolitica biotype 1B carries a novel insertion sequence IS1328. FEMS Microbiol Lett. 1995;129:287–292. [PubMed]
42. Rakin A, Saken E, Harmsen D, Heesemann J. The pesticin receptor of Yersinia enterocolitica: a novel virulence factor with dual function. Mol Microbiol. 1994;13:253–263. [PubMed]
43. Reed L J, Muench H. A simple method of estimation fifty per cent endpoints. Am J Hyg. 1935;27:493–497.
44. Reizer J, Reizer A, Saier M H. Novel PTS proteins revealed by bacterial genome sequencing: a unique fructose-specific phosphoryl transfer protein with two HPr-like domains in Haemophilus influenzae. Res Microbiol. 1996;147:209–215. [PubMed]
45. Richard C, Kiredjian M, editors. Méthodes de laboratoires pour l’identification des bacilles a Gram négatif aérobies stricts. 2nd ed. Paris, France: Institut Pasteur; 1995.
46. Robins-Browne R M, Kaya Prpic J. Effects of iron and desferrioxamine on infections with Yersinia enterocolitica. Infect Immun. 1985;47:774–779. [PMC free article] [PubMed]
47. Sanger F, Nicklen S, Coulson A R. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA. 1977;74:5463–5467. [PMC free article] [PubMed]
48. Schmiel D H, Wagar E, Karamanou L, Weeks D, Miller V L. Phospholipase A of Yersinia enterocolitica contributes to pathogenesis in a mouse model. Infect Immun. 1998;66:3941–3951. [PMC free article] [PubMed]
49. Schubert S, Rakin A, Karch H, Carniel E, Heesemann J. Prevalence of the “high-pathogenicity island” of Yersinia species among Escherichia coli strains that are pathogenic to humans. Infect Immun. 1998;66:480–485. [PMC free article] [PubMed]
50. Sullivan J T, Ronson C W. Evolution of rhizobia by acquisition of a 500-kb symbiosis island that integrates into a phe-tRNA gene. Proc Natl Acad Sci USA. 1998;95:5145–5149. [PMC free article] [PubMed]
51. Toora S, Buduamoako E, Ablett R F, Smith J. Inhibition and inactivation of pathogenic serogroups of Yersinia enterocolitica by a bacteriocin produced by Yersinia kristensenii. Lett Appl Microbiol. 1994;19:40–43.

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...