SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Gene transfer between separate lineages of a bacterial pathogen can promote recombinational divergence and the emergence of new pathogenic variants. Temperate bacteriophages, by virtue of their ability to carry foreign DNA, are potential key players in this process. Our previous work has shown that representative strains of Salmonella typhimurium (LT2, ATCC14028 and SL1344) are lysogenic for two temperate bacteriophages: Gifsy-1 and Gifsy-2. Several lines of evidence suggested that both elements carry genes that contribute to Salmonella virulence. One such gene, on the Gifsy-2 prophage, codes for the [Cu, Zn] superoxide dismutase SodCI. Other putative pathogenicity determinants were uncovered more recently. These include genes for known or presumptive type III-translocated proteins and a locus, duplicated on both prophages, showing sequence similarity to a gene involved in Salmonella enteropathogenesis (pipA). In addition to Gifsy-1 and Gifsy-2, each of the above strains was found to harbour a specific set of prophages also carrying putative pathogenicity determinants. A phage released from strain LT2 and identified as phage Fels-1 carries the nanH gene and a novel sodC gene, which was named sodCIII. Strain ATCC14028 releases a lambdoid phage, named Gifsy-3, which contains the phoP/phoQ-activated pagJ gene and the gene for the secreted leucine-rich repeat protein SspH1. Finally, a phage specifically released from strain SL1344 was identified as SopEΦ. Most phage-associated loci transferred efficiently between Salmonella strains of the same or different serovars. Overall, these results suggest that lysogenic conversion is a major mechanism driving the evolution of Salmonella bacteria.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Salmonellae are enteric pathogenic bacteria capable of infecting a wide variety of animal species. Some serotypes (or serovars) are adapted to specific hosts; others show a wide host range (Kingsley and Bäumler, 2000; Uzzau et al., 2000). Prototypical of the latter class, Salmonella enterica serovar Typhimurium (S. typhimurium) can infect humans, rodents and all kinds of livestock including poultry. Depending on the nature and immunological state of the host, the pathology of S. typhimurium infection can vary from a local, self-limiting gastroenteritis to a systemic illness (Miller and Pegues, 2000). As with other animal and plant pathogens, Salmonella–host interactions are largely mediated by the activity of specialized type III secretion systems, which enable bacterial effector proteins to translocate into eukaryotic cells (Hueck, 1998). Two such systems, encoded in separate pathogenicity islands, SPI-1 and SPI-2, have been specifically associated with the enteric and systemic phases of S. typhimurium infection respectively (Galán, 1999; Hensel, 2000; Schechter and Lee, 2000). The genetic determinants of Salmonella host range remain largely unknown. Recent efforts to tackle this point have involved comparative screening of mutants in two animal models (Miao et al., 1999; Tsolis et al., 1999). One such study identified a gene differentially required for S. typhimurium virulence in mice and calves. This gene codes for a protein containing leucine-rich repeats, termed SlrP, bearing similarities to the ipaH protein family of Shigella and yopM of Yersinia (Tsolis et al., 1999). In a separate study, two S. typhimurium genes with homology to slrP were identified and shown to encode type III-secreted proteins named SspH1 and SspH2. Simultaneous inactivation of these two proteins resulted in attenuation of S. typhimurium virulence in calves but not in mice (Miao et al., 1999). The heterogeneous occurrence of slrP, sspH1 and sspH2 genes in Salmonella serovars (Tsolis et al., 1999) suggests extensive shuffling of these sequences during Salmonella evolution.

Horizontal gene transfer is widely regarded as the main mechanism driving the evolution of Salmonella pathogenicity (Ochman et al., 2000). Evidence from recent work suggests that a significant contribution to this process results from lysogenic conversion by temperate phage. The sopE gene, which codes for a type III-secreted protein that promotes internalization of bacteria in host cells, lies within a P2-like prophage (Hardt et al., 1998). Mirold et al. (1999) were able to induce this prophage from an epidemic strain and showed it to be capable of transferring the sopE sequence to a susceptible recipient strain. This strongly suggests the horizontal acquisition of the gene within the S. typhimurium complex. The phenotypic changes accompanying the acquisition are unknown and probably subtle; the sopE gene is not required for S. typhimurium virulence in mice (Hardt et al., 1998), and it is absent from a large group of S. typhimurium strains responsible for recent epidemic outbreaks in humans and animals (e.g. strain DT104; Hardt et al., 1998). A separate line of work from our laboratory led to the identification of two lambdoid phages, Gifsy-1 and Gifsy-2, present in the lysogenic state in most S. typhimurium strains (Figueroa-Bossi et al., 1997; Figueroa-Bossi and Bossi, 1999) including strain DT104 (our unpublished results). Both prophages contribute to S. typhimurium virulence in mice, albeit to vastly different extents. Removal of the Gifsy-2 prophage results in over 100-fold attenuation, whereas curing for the Gifsy-1 reduces virulence slightly, and this effect can only be detected in strains already cured for the Gifsy-2 prophage (Figueroa-Bossi and Bossi, 1999). We determined that the Gifsy-2 prophage carries the sodCI gene, which encodes a periplasmic [Cu, Zn] superoxide dismutase previously implicated in bacterial defence against macrophage oxidative burst (De Groote et al., 1997; Farrant et al., 1997; Fang et al., 1999). The sodCI gene alone could partially complement the absence of the Gifsy-2 prophage in strains that contained Gifsy-1 but not in strains lacking the latter (Figueroa-Bossi and Bossi, 1999). These findings provided independent confirmation of the sodCI gene role in S. typhimurium virulence and predicted the existence of additional, possibly redundant pathogenicity determinants on both prophages. Recently, some genes linked to bacterial virulence were found to lie within the Gifsy prophages' genomes. They include a Gifsy-1 gene (gipA) involved in the bacterial colonization of small intestine (Stanley et al., 2000) and a Gifsy-2 gene of unknown function, named srfH (Worley et al., 2000) or sseI (Miao and Miller, 2000), induced in macrophages via the SsrB activator protein and encoding a type III-secreted protein (Miao and Miller, 2000; Worley et al., 2000). Finally, the grvA gene of Gifsy-2 naturally attenuates pathogenicity through a sodCI-dependent mechanism that remains elusive (T. Ho and J. Slauch, personal communication).

We show in this study that both Gifsy-1 and Gifsy-2 prophages contain sequences with various degrees of similarity to known or presumptive virulence genes, including an open reading frame (ORF) similar to the pipA gene of Salmonella enterica serovar Dublin (S. dublin;Wood et al., 1998) and a Gifsy-1 ORF specifying a hypothetical leucine-rich protein similar to the YopM protein of Yersinia species. Furthermore, we have isolated and characterized other S. typhimurium bacteriophages. A novel phage specifically released by strain ATCC14028, morphologically and structurally related to the Gifsy prophages, was named Gifsy-3. This phage's genome was found to include recently characterized pagJ and sspH1 loci (Gunn et al., 1998; Miao et al., 1999). A phage specifically released by strain LT2 and identified as the Fels-1 phage was found to include a new sodC gene, sodCIII. Finally, strain SL1344 released a phage related to SopEΦ. Each of the three phages efficiently lysogenized S. typhimurium strains that did not carry them as prophages. Furthermore, phages Gifsy-1, Gifsy-2 and Gifsy-3 proved competent to infect and lysogenize strains from serovars other than Typhimurium. Overall, these results indicate that lysogeny constitutes a major source of genomic diversity among Salmonella bacteria.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Phages Gifsy-1 and Gifsy-2 contain multiple putative pathogenicity determinants

DNA sequence analysis and electron microscope examination indicate Gifsy-1 and Gifsy-2 bacteriophages to be related to bacteriophage lambda in terms of overall genome organization, primary structure and morphology. In addition, the sequence work reveals that both elements contain ORFs with a high degree of sequence similarity to virulence determinants of Salmonella or other enteric pathogens. Some of these ORFs are clustered at the right end of the prophage map, within the interval delimited by putative tail fibre assembly protein genes (tfaO and tfaT) and the attR site (Fig. 1). This interval corresponds to the dispensable ‘b’ region of bacteriophage lambda (Daniels et al., 1983).

image

Figure 1. Schematic representation of the ‘b’ regions of Gifsy-1 and Gifsy-2 prophages. The diagram compiles data from the Genome Sequence Center (S. typhimurium project), Washington University, St Louis, MO, USA, and data from our laboratory (GenBank files AF147699, AF254761 and AF254763).

Gifsy-1. From left to right: ORF tfaO, presumed tail fibre assembly protein gene similar to the tfa gene (formerly ORF 194) of bacteriophage lambda and related phages (Montag et al., 1989; Haggård-Ljungquist et al., 1992). ORF gogD shows over 90% identity to the pagJ/pagK loci of S. typhimurium (Gunn et al., 1998). Elements e1 and e2 are similar to the promoter-distal segments of invertase family genes (Plasterk and van de Putte, 1985) and of the S. typhimurium umuC gene (Smith et al., 1990) respectively. Element e3 is similar to the N-terminal-coding portion of Shigella sonnei IS630 transposase gene (Matsutani et al., 1987). ORF gogC does not exhibit significant similarity to any known gene. Element e4 shares identity with a non-coding segment of pathogenicity island SPI-5 (Wood et al., 1998). ORF gogB specifies a hypothetical, leucine-rich protein related to the YopM protein of Yersinia pestis (Leung and Straley, 1989).

Gifsy-2. From left to right; presumptive tail fibre assembly protein gene tfaT. ORF gtgB corresponds to the sfrH gene described by Worley et al. (2000) and the sseI gene described by Miao and Miller (2000). ORF gtgC is similar to ORFs found within a pathogenicity island-associated prophage and in the adherence factor plasmid of enteropathogenic E. coli (Perna et al., 1998; Tobe et al., 1999). ORF gtgD shows identity to the yedK locus of E. coli (Raha et al., 1993; Blattner et al., 1997). Element e1 is a 35 bp sequence identical to S. typhimurium tRNA2Pro gene. ORF gtgF specifies a hypothetical protein similar to the MsgA virulence protein of S. typhimurium. (Gunn et al., 1995). ORF gtgE does not show similarity to any entry in DNA or protein sequence databases.

Download figure to PowerPoint

Gifsy-1.  The right-end portion of the Gifsy-1 prophage genome includes a sequence > 90% identical to S. typhimurium phoP/phoQ-activated loci pagJ and pagK (Gunn et al., 1998). (We show below that the pagJ gene is part of yet another lambdoid prophage, Gifsy-3.) The pagJ/K homologous segment is followed by a series of truncated stretches with sequence identity to various genes or elements, including genes involved in DNA transposition and inversion, suggestive of extensive recombinational scrambling (Fig. 1). Near the Gifsy-1 attR site lies a 1467 bp ORF, denoted gogB (Gifsyone gene; GenBank AF254761), whose predicted product scores significant similarity to members of the leucine-rich repeat family (LRR), such as yopM of Yersinia pestis (Leung and Straley, 1989), and to a recently described class of type III-secreted proteins of S. typhimurium (Miao and Miller, 2000). To assess the role of the gogB locus in Salmonella virulence, we constructed a gogB deletion by allelic exchange (Δ[gogB]2::Kn; see Experimental procedures) and orally challenged mice with the null mutant mixed with its wild-type sibling. We saw no major differences in the abilities of the two strains to colonize the small intestine and spleen of infected animals, suggesting that the gogB locus does not play an essential role in murine salmonellosis. Additional candidates for pathogenicity determinants are found outside the Gifsy-1 ‘b’ region. In particular, the prophage genome includes an ORF, gogA (GenBank AF254760) whose hypothetical product shows 72% identity with the PipA protein of S. dublin (Wood et al., 1998). A nearly identical copy of the gogA locus is present within the Gifsy-2 genome.

Gifsy-2.  The ‘b’ region of the Gifsy-2 prophage includes five major ORFs, denoted gtgB to gtgF (Gifsytwo gene; Fig. 1). ORF gtgB (GenBank AF254763) specifies a hypothetical protein of 322 amino acids. While this work was in progress, two independent reports identified this ORF as a locus under the control of SPI-2-encoded SsrB activator protein (srfH;Worley et al., 2000) and encoding a type III-translocated protein (seeI;Miao and Miller, 2000). Neither of these studies explicitly assessed the contribution of gtgB (sfrH, seeI) to virulence. We performed mouse virulence assays with a strain in which the gtgB sequence is replaced by a KnR cassette (see Experimental procedures). We found strain MA6915 (Gifsy-1[–]Δ[gtgB]1::Kn) to compete equally with strain MA5973 (Gifsy-1[–]) for growth in splenic and intestinal tissues of mice after oral challenge (data not shown). The gtgB locus is followed by an ORF (gtgC) showing similarity to sequences found in prophage and plasmid of enterohaemorrhagic Escherichia coli (Perna et al., 1998; Tobe et al., 1999). The ORF named gtgD specifies a putative protein that exhibits 85% identity to the YedK protein of E. coli (Blattner et al., 1997). Intriguingly, the yedK locus lies within a segment of the E. coli chromosome that is absent from the corresponding location in S. typhimurium (Raha et al., 1993). At the far end of the Gifsy-2 prophage map, the ORF denoted gtgF specifies a hypothetical 63-amino-acid protein showing 76% identity to Salmonella virulence protein MsgA (Gunn et al., 1995). The Gifsy-2 genome also includes a pipA gene homologue. The nucleotide sequence of this ORF, designated gtgA (GenBank AF254762), is virtually identical (98% identity) to that of the gogA locus of Gifsy-1 (see above). Not surprisingly, a DNA fragment carrying the gtgA sequence disrupted by a CmR cassette was found to recombine with gogA and gtgA loci at comparable frequencies. Interestingly, the 1.3 kb insertion completely abolished the ability of the Gifsy-2 prophage to form active phage, but it did not so when present in Gifsy-1. This corroborated evidence accumulated in the course of this work indicating a restricted capacity of Gifsy-2 to accommodate excess DNA in the virion. The gogA::Cm and gtgA::Cm insertions were separately moved into Gifsy-2[–] and Gifsy-1[–] backgrounds, respectively, and the resulting strains were tested for mouse virulence in competition with the respective parental strains. No significant differences were observed (data not shown).

Gifsy-3, a novel lambdoid phage, carries S. typhimurium pagJ and sspH1 loci

The presence of regions of sequence homology between Gifsy-1 and Gifsy-2 phage genomes allows genetic markers isolated in one prophage to be transferable onto the other during generalized transduction (Figueroa et al., 1997). This is the case for mini-Tn10 insertions linked to sbcE21, a small deletion disrupting a regulatory element on Gifsy-1 and causing derepression of the xis gene and loss of the prophage by abortive excision (Figueroa et al., 1997). While measuring recombination of zfh-8157::Tn10dTc and sbcE21 markers onto the Gifsy-2 prophage, we noticed that, whenever the strain used as recipient was derived from S. typhimurium strain ATCC14028, recombination would occasionally take place at a site other than Gifsy-1 or Gifsy-2. This suggested that this specific strain background contained a third region homologous to the segment being studied, probably a third prophage. The Tn10dTc insertion was mapped at about 120 kb from the XbaI restriction endonuclease site at 29.7 centisomes of the S. typhimurium chromosome (Fig. 2). The presence of the sbcE21 mutation caused an ≈ 50 kb segment, including the mutant site and the nearby Tn10dTc marker, to become unstable and lost at high frequency. The resulting strain (MA6051; Fig. 2A, lane 3) became sensitive to a phage spontaneously released from cultures of the parental strain. These data indicated that the excisable fragment was indeed a prophage, fully competent to undergo an infection cycle. The new phage – morphologically indistinguishable from the Gifsy-1 and Gifsy-2 phages (data not shown) – was named Gifsy-3.

image

Figure 2. A. Pulsed-field gel electrophoretic separation and Southern hybridization analysis of chromosomal DNA fragments from strains carrying or lacking physically marked Gifsy-3 prophage. A segment of the Gifsy-1 prophage carrying insertion elements sfh-8157::Tn10dTc and zfh-8165::MudF(lac) was transferred by homologous recombination (transduction) into the corresponding region of the Gifsy-3 prophage of S. typhimurium strain ATCC14028. In the resulting strain (MA6020), two XbaI restriction endonuclease recognition sites are introduced in Gifsy-3 prophage DNA: one within the Tn10dTc element and the other within the acquired Gifsy-1 segment. The presence of the sbcE21 allele within this segment (Figueroa-Bossi et al., 1997) causes the entire Gifsy-3 prophage to become excised (strain MA6051). Chromosomal DNA, digested with XbaI nuclease, was separated by PFGE and subjected to Southern hybridization analysis. Lane 1, strain ATCC14028 (wild type); lane 2, strain MA6020 (zch-8165::MudF zch-8157::Tn10dTc zbcE21); lane 3, strain MA6051 (TcS Lac segregant of strain MA6020). Lanes 4, 5 and 6, Southern hybridization of samples in lanes 1, 2 and 3, respectively, to 32P-labelled DNA from plasmid pK4114. The latter carries a fragment from the Gifsy-1 prophage (including the zfh-8157::Tn10dTc insertion), which shares sequence identity with the Gifsy-2 prophage. Normally, this probe hybridizes to 850 kb and 35 kb XbaI fragments (Figueroa-Bossi et al., 1997). In strain MA6020 (lane 2), a third hybridization signal is present (band b) reflecting the transfer of the Tn10dTc element and nearby sequence to the Gifsy-3 prophage.

B. Position of Gifsy-2 and Gifsy-3 prophages in S. typhimurium ATCC14028 chromosome. The open box within the Gifsy-3 diagram represents Gifsy-1 material (MudF element not shown; note the size of the b fragment is normally ≈ 9 kb; it increases to ≈ 20 kb in the presence of zch-8165::MudF). The orientation of the Gifsy-3 prophage was determined upon analysing the cleavage and hybridization patterns of strains resulting from the transfer of additional Gifsy-1 markers (data not shown). The position and orientation of the Gifsy-2 prophage were determined previously (Figueroa-Bossi et al., 1997).

Download figure to PowerPoint

The location of the Gifsy-3 prophage at 27–28 centisomes in strain ATCC14028 made this element a candidate for carrying the pagJ–sspH1 loci (Gunn et al., 1998; Miao et al., 1999). Southern analysis showed a polymerase chain reaction (PCR)-amplified DNA fragment from the region between pagJ and sspH1 to hybridize strongly to purified Gifsy-3 phage DNA (data not shown; see below). Moreover, a fragment spanning the sspH1 sequence could be amplified by PCR from strain ATCC14028 and from purified Gifsy-3 DNA, but not from strain MA6051 (ATCC14028 Gifsy-3[–]). Thus, these results confirmed the presence of the pagJ–sspH1 region on the Gifsy-3 prophage. Bacteria cured for this prophage (strain MA6051) showed no detectable impairment in their ability to cause systemic infection and death in mice after oral inoculation (data not shown).

The work above suggested that the Gifsy-3 prophage is absent from strain LT2. To test whether strain LT2 was susceptible to Gifsy-3 phage infection, a supernatant from a culture of strain MA5975 (ATCC14028 Gifsy-1[–]Gifsy-2[–]) was spotted on a lawn of cells from strain MA4587 (LT2 Gifsy-1[–]Gifsy-2[–]). Several plaques were observed. DNA from phage recovered from a number of such plaques had a restriction enzyme cleavage pattern indistinguishable from the Gifsy-3 DNA pattern.

Strain LT2-specific Fels-1 prophage carries the nanH gene and a novel sodC gene, sodCIII

The plaquing assay described in the paragraph above was also performed after switching donor and recipient strains. Several plaques were also observed in this case, suggesting that strain MA4587 (LT2 Gifsy-1[–]Gifsy-2[–]) spontaneously released a phage against which strain MA5975 (ATCC14028 Gifsy-1[–]Gifsy-2[–]) was non-immune. Phage suspensions from 16 independent plaques were spotted against a set of tester strains derived from S. typhimurium strain Q1 carrying or lacking the Fels-1 or Fels-2 prophages (Affolter et al., 1983). These tests identified the phage in all 16 samples as the Fels-1 phage. We found Fels-1 phage to be capable of forming plaques on Gifsy-1-cured derivatives from a number of different S. typhimurium isolates. Several clues suggested that the Fels-1 prophage might carry the nanH gene, which codes for neuramidinase and was reported to lie near a phage gene in the strain LT2 chromosome (Hoyer et al., 1992). This was confirmed by finding that a nanH-specific DNA fragment, amplified by PCR from strain LT2, hybridized strongly to Fels-1 phage DNA (data not shown; see also below).

Examination of the S. typhimurium strain LT2 genome sequencing project database (Genome Sequence Center, Washington University, St Louis, MO, USA), revealed an ORF whose hypothetical product showed sequence identity to the sodCI gene for periplasmic superoxide dismutase (De Groote et al., 1997; Farrant et al., 1997). The presence of phage homology in the neighbourhood of the putative gene suggested that, like the sodCI gene, this sodCI homologue lay within the genome of a prophage. We tested whether the ORF sequence could be amplified from any of the phages analysed in this study. A PCR product of the expected length was obtained with Fels-1 DNA as the template. Sequence analysis confirmed this fragment to include the presumptive sodC gene, which was named sodCIII (GenBank AF254764). The location of the gene within the Fels-1 genome is different from, although close to, that occupied by the sodCI gene in the Gifsy-2 prophage (Fig. 3). The deduced SodCIII polypeptide shows 60% identity to the SodCI protein precursor and 65% identity to the S. typhimurium SodCII precursor (Fang et al., 1999). The sequence alignment in Fig. 4 tentatively identifies amino acid residues that differentiate phage-encoded and chromosomally encoded SodC proteins. It is intriguing to consider that strain LT2, which is normally lysogenic for Fels-1 and Gifsy-2 prophages, contains three different sodC genes.

image

Figure 3. Gene organization within the putative tail regions of Gifsy-2 and Fels-1 prophages. The diagram compiles sequence data from De Groote et al. (1997), the Genome Sequence Center (S. typhimurium project), Washington University, St Louis, MO, USA, and our laboratory (GenBank AF254764). Putative tail genes are named according to their similarity to the corresponding genes in bacteriophage lambda (Daniels et al., 1983). ORFs named ailT and ailF show sequence identity to the adherence/invasion (ail) locus of Yersinia enterocolitica (Miller et al., 1990).

Download figure to PowerPoint

image

Figure 4. Alignment of the deduced amino acid sequences of S. typhimurium SodCI, SodCII and SodCIII protein precursors and of the E. coli SodC protein precursor. Sequence data are from the following references: S. typhimurium SodCI, De Groote et al. (1997), Farrant et al. (1997); S. typhimurium SodCII, Fang et al. (1999); S. typhimurium SodCIII, this work (GenBank AAF82484); E. coli SodC, Imlay and Imlay (1996). White lettering specifies residues that are different between phage and bacterial proteins but are conserved within each group. Black lettering on a dark-shaded background specifies highly conserved residues; black lettering on a light-shaded background specifies mildly conserved residues.

Download figure to PowerPoint

In a previous study, we found the sodCI gene alone sufficient to restore virulence partially in a Gifsy-2-cured strain (Figueroa-Bossi and Bossi, 1999). To test whether the Fels-1-borne sodCIII gene could substitute for sodCI in the above assay, we isolated a Gifsy-2-cured derivative of strain ATCC14028 lysogenized by phage Fels-1 (see below). Three groups of five mice were separately injected intraperitoneally with strains MA6275 (Gifsy-2[–]), MA6729 (Gifsy-2[–]Fels-1[+]) and MA6601 (Gifsy-2[–]sodCI[+]) at a dose of ≈ 1000 bacteria mouse−1. All mice challenged with the MA6275 and MA6729 strains survived the treatment, whereas only one animal in the group exposed to strain MA6601 survived. These results suggest that the sodCI and sodCIII genes are not interchangeable with regard to their participation in mouse virulence. This might reflect an inefficient expression of the sodCIII gene or a poorer activity of the gene product compared with SodCI. The latter was recently shown to have one of the highest catalytic rates ever measured for superoxide dismutases (Pesce et al., 2000).

Strain SL1344 releases a functional SopEΦ

Although initially identified as prophage in strain SL1344, the ‘SopE’ phage (SopEΦ; Mirold et al., 1999) could not be recovered as a virus from this strain and was inferred to be defective (Hardt et al., 1998). In the course of the present work, we found that strain MA6247 (SL1344 Gifsy-1[–]Gifsy-2[–]) spontaneously released a phage forming plaques on lawns of strain MA5975 (ATCC14028 Gifsy-1[–]Gifsy-2[–]). Southern hybridization analysis confirmed the presence of the sopE gene in the SopEΦ genome. As already observed with Fels-1, the plaquing efficiency of SopEΦ (from SL1344) was significantly lower on recipient strains harbouring Gifsy-1 as prophage. These findings, which suggest a general interference by the Gifsy-1 prophage on superinfecting phages, might account for the difficulties encountered in the initial identification of SopEΦ as phage (Hardt et al., 1998).

Lysogeny allows gene transfer between strains of the same or different Salmonella serovars

All phages described here were found to be fully competent for lysogeny of non-immune strains. We were able to interchange the prophage complements of strains LT2, ATCC14028 and SL1344 and to generate variants with new combinations. In all cases, lysogeny was accompanied by the acquisition of the genes associated with that particular phage (Fig. 5). For some of the phages, the ability to lysogenize strains from different serovars was also assessed. We obtained derivatives of Salmonella enterica serovar Typhi (S. typhi) and serovar Abortusovis (Salmonella abortusovis) lysogenic for phages Gifsy-1, Gifsy-2 and Gifsy-3, and derivatives of Salmonella enterica serovar Gallinarum (S. gallinarum) lysogenic for Gifsy-1. Lysogeny of S. typhi by Gifsy-2 required that a resident prophage present at the Gifsy-2 attachment site in the S. typhi chromosome be removed first. Although the Gifsy-2 prophages of S. typhi and S. typhimurium are highly divergent throughout most of their genomes (the S. typhi Gifsy-2 genome lacks the sodCI, gtgA and gtgB genes; data from the Sanger Sequence Centre, UK), S. typhi Gifsy-2 prophage confers immunity to infection by Gifsy-2 phage from S. typhimurium.

image

Figure 5. Pulsed-field gel electrophoretic separation and Southern hybridization analysis of chromosomal DNA fragments from experimentally induced lysogens. Strain lysogeny by different phages was carried out as described in Experimental procedures. Chromosomal DNA XbaI nuclease digests were separated by PFGE and analysed by hybridization to 32P-labelled PCR products. Probe ‘pagJ/K*’ covers a segment immediately adjacent to the coding sequences of the pagJ and pagK genes (positions 1136–1949 of GenBank AF013776 and positions 1123–1938 of GenBank AF013775; Gunn et al., 1998). Probe ‘nanH’ spans the entire nanH gene (co-ordinates 338–1382 of GenBank M55342; Hoyer et al., 1992). Probe ‘sopE’ spans the sopE gene (co-ordinates 1336–2378 of GenBank AF043239, Hardt et al., 1998).

A. Lane 1, Strain LT2 (wild type); lane 2, strain MA6495 (LT2 Gifsy-3[+]); lane 3, strain MA6502 (LT2 Gifsy-1[–]Gifsy-2[–]Gifsy-3[+]); lanes 4–6, hybridization of samples in lanes 1–3, respectively, to the pagJ/K* probe; lane 7, strain 4587 (LT2 Gifsy-1[–]Gifsy-2[–]); lane 8, strain MA6595 (LT2 Gifsy-1[–]Gifsy-2[–] SopEΦ[+]; lanes 9–12, hybridization of samples in lanes 7 and 8 to probes sopE (lanes 9 and 10) and nanH (lanes 11 and 12).

B. Lane 1, strain MA6052 (ATCC14028 Gifsy-1[–]Gifsy-2[–]Gifsy-3[–]); lane 2, strain MA6556 (ATCC14028 Gifsy-1[–]Gifsy-2[–]Gifsy-3[–]Fels-1[+]); lanes 3–6, hybridization of samples in lanes 1 and 2 to probes nanH (lanes 3 and 4) and pagJ/K* (lanes 5 and 6).

C. Lane 1, strain Q1 (wild type); lane 2, strain Q1-F1 (Q1 Fels-1[+]); lanes 3 and 4, hybridization of samples in lanes 1 and 2, respectively, to the nanH probe.

D. Lane 1, strain MA6498 (SL1344 Gifsy-1[–]Gifsy-2[–]Gifsy-3[+]); lanes 2–4, hybridization of sample in lane 1 to the probes sopE (lane 2), pagJ/K* (lane 3) and nanH (lane 4).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The data presented here indicate that lysogeny by temperate phage provides S. typhimurium with a variable assortment of genetic loci that may extend the pathogenic arsenal of the organism. Many such loci appear to be fully integrated in the regulatory network of virulence determinants. In particular, four out of the five phages analysed here encode proteins known or inferred to be substrates of SPI-1- and/or SPI-2–encoded type III secretion systems. Of these, the SseI protein of Gifsy-2 (GtgB) and the SspH1 protein of phage Gifsy-3 belong to a recently identified family of effector proteins that share translocation signals and some of which contain leucine-rich repeats (Miao and Miller, 2000). The putative GogB protein of Gifsy-1 might be considered as another, albeit distantly related, member of this family. In addition, ORFs with similarity to virulence genes pipA, pagJ/K and msgA were identified on phages Gifsy-1 and/or Gifsy-2. Phage Fels-1 was found to encode a novel sodC gene, sodCIII. Experiments assessing the involvement of some of the above loci in S. typhimurium virulence in mice proved inconclusive. This suggests that none of the loci that were tested corresponds to any of the determinants that make up the contribution of the Gifsy-1 and Gifsy-2 prophages to mouse virulence (Figueroa-Bossi and Bossi, 1999; see Introduction). Genes needed in the enteric phase of Salmonella infection are often dispensable for murine salmonellosis. This was shown to be the case for the pipA gene (Wood et al., 1998) and for the Gifsy-3-borne sspH1 gene (Miao et al., 1999). Of course, the possibility remains that not all loci studied here are actually involved in pathogenicity. In particular, failure of the sodCIII gene to complement, even partially, a sodCI deletion mutant during mouse infection suggests that the Fels-1-borne gene is poorly expressed or defective. Clearly, further work is needed to clarify this issue.

Overall, our data suggest that phages play an active role in the spreading of genetic determinants in the Salmonella genus. The ability of phage-borne loci to ‘plug in’ the regulatory circuitry of the bacterium after lysogeny may allow the latter to sample different combinations of effector molecules rapidly. Any combination improving fitness in the host or allowing colonization of new niches would be positively selected. As selection may vary as a function of the host, one might anticipate that independent isolates from broad-host-range Salmonella serovars have different prophage complements. These differences might provide clues regarding the phylogenetic history of individual strains. That is the case for two representative strains used in this study. As shown above, the Gifsy-3 prophage is naturally found in S. typhimurium strain ATCC14028 and is absent from strains LT2 and SL1344. A recent survey of the distribution of the Gifsy-3-borne sspH1 gene in the Salmonella genus found this locus to be present in only a tiny fraction of isolates from Salmonella enterica subspecies I (Tsolis et al., 1999). In contrast, the gene occurs at a significantly higher frequency in other S. enterica subspecies and is present in both Salmonella bongori strains that were tested (Tsolis et al., 1999). It is tempting to speculate that incorporation of the sspH1 gene in strain the ATCC14028 chromosome, via Gifsy-3 or related phage, reflects a recent excursion of this particular strain in the host reservoir of other Salmonella species or subspecies. Similar considerations can be made for strain LT2. Hoyer et al. (1992) analysed the presence of the nanH gene in over 200 S. typhimurium isolates, including the 22 strains of the original LT collection (Lilleengen, 1948), and found that only strain LT2 contained the gene. In contrast, the gene was present in about 60% of isolates from S. enterica subspecies III (formerly Salmonella arizonae). Vimr and coworkers proposed that strain LT2 obtained the nanH gene by lateral transfer from a separate Salmonella subspecies (Hoyer et al., 1992; Roggentin et al., 1993). The association of the nanH gene with phage Fels-1 demonstrated in this work lends support to this hypothesis.

Recombination is likely to constitute an additional, important source of variability in the repertoires of phage-associated virulence loci. Evidence for both legitimate and illegitimate events accumulated throughout the present study. In one case, the partners involved were phage Gifsy-2 and an as yet unidentified prophage present in the strain ATCC14028 chromosome. The hybrid phage produced has Gifsy-2 immunity but carries a section of this phage genome replaced by material from the unknown prophage (our unpublished results). This suggests that recombination can allow rescue and mobilization of virulence loci associated with defective prophages. Two candidates for such elements – present in most, if not all, S. typhimurium strains – are at the site of mig-3/pagK loci (40–42 centisomes; Valdivia and Falkow, 1997; Gunn et al., 1998) and at the site of the sspH2 gene (66.8–69 centisomes; Miao et al., 1999). It is interesting to notice that many of the phage-associated virulence loci described here tend to be clustered immediately downstream from putative distal tail fibre genes, which, in the best-studied phages (e.g. lambda, T4 and P2), specify receptor binding and determine host range (Haggård-Ljungquist et al., 1992; Juhala et al., 2000). High-level recombination between these genes of otherwise unrelated bacteriophages has been proposed to be an important mechanism generating phage variants with altered host range (Montag et al., 1989; Haggård-Ljungquist et al., 1992). Sequence comparison near the end of the presumptive tail-encoding region of phages Gifsy-1, Gifsy-2, Gifsy-3 and of the mig-3/pagK element reveals a mosaic structure indicative of multiple genetic exchanges (Fig. 6). Significantly, such DNA scrambling produces alternative forms of a putative tail assembly protein (Fig. 6). This raises the attractive possibility that ‘pathogenic’ phages integrate the shuffling of virulence genes into the mechanism responsible for variations in host range.

image

Figure 6. Schematic diagram of the region near putative tail assembly protein genes of Gifsy-1, Gifsy-2 and Gifsy-3 prophages and in prophage-type element at 40–42 centisomes of the S. typhimurium chromosome. The diagram compiles data from Valdivia and Falkow (1997; GenBank AF020804), Gunn et al. (1998; GenBank AF013775 and AF013776) and data from the Genome Sequence Center (S. typhimurium project) Washington University, St Louis, MO, USA. Boxes of the same colour specify regions sharing more than 85% DNA sequence identity. Sequence comparisons were made using the blast algorithm of Altschul et al. (1997).

Download figure to PowerPoint

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains, plasmids and growth conditions

Bacterial strains used in this study are listed in Table 1. Wild-type S. typhimurium strains were obtained from the sources indicated by Figueroa-Bossi and Bossi (1999). Strain Q1 and its Fels-lysogenic derivatives were obtained from Diana Downs and Alan Anderson respectively. Plasmid pK4114 carries a 15 kb DNA segment that spans the left portion of the Gifsy-1 prophage genome (Figueroa-Bossi et al., 1997). Plasmid pNK2884 (Kleckner et al., 1991) was the source of the CmR cassette used for gene disruption. Plasmids pKD4 and pKD46 were used in gene replacement experiments (Datsenko and Wanner, 2000). Unless specified otherwise, bacteria were cultured at 37°C in liquid media or in media solidified with 1.5% (w/v) Difco agar. Luria–Bertani (LB) broth [1% bacto tryptone (w/v), 0.5% Difco yeast extract (w/v), 0.5% NaCl (w/v)] was used as complex medium. When required, antibiotics were included at the following concentrations: tetracycline hydrochloride, 25 µg ml−1; chloramphenicol, 10 µg ml−1; kanamycin monosulphate, 50 µg ml−1 Na ampicillin 50 µg ml−1. Phage-induced bacterial lysis on agar plates was monitored using ‘Green’ medium [(Chan et al., 1972) 0.8% bacto tryptone, 0.1% yeast extract (w/v), 0.5% NaCl (w/v), 6.7 g l−1 glucose, 630 mg l−1 Alizarin yellow (GG), 66 mg l−1 Methyl blue] or EBU medium [(Bochner, 1984) 1% bacto tryptone (w/v), 0.5% Difco yeast extract (w/v), 0.5% NaCl (w/v) 2.5 g l−1 glucose, 0.5% K2HPO4 (w/v), 12.5 mg l−1 Evans blue, 25 mg l−1 uranine (sodium fluorescein)]. Liquid cultures were grown in New Brunswick gyratory shakers, and growth was monitored by measuring the optical density (OD) at 600 nm with a Milton–Roy Spectronic 301 spectrophotometer.

Table 1. Bacterial strains.
StrainRelevant genotypeaReferenceb
  • a . Only the changes from the native prophage complement of the strain are indicated. Tn10dTc is a transpositionally inert derivative of transposon Tn10 ( Kleckner et al., 1991). MudF is a transpositionally inert Mu-derived element with Mu ends flanking the E. coli lac operon and a kanamycin resistance gene (Galitski and Roth, 1996).

  • b

    . Where not specified, the strain was constructed in the course of this work.

  • c . The initial ‘zfh’ designation of the Tn10dTc and MudF insertions in this strain ( Figueroa et al., 1997) was changed to ‘zch’ to comply with their new location in the Gifsy-3 prophage.

S. typhimurium ATCC14028s
 ATCC14028sWild type Fields et al. (1986)
 MA5973 Gifsy-1[–] Figueroa-Bossi and Bossi (1999)
 MA5975 Gifsy-1[–]Gifsy-2[–] Figueroa-Bossi and Bossi (1999)
 MA6020c zch-8157::Tn10dTc zch-8165::MudF sbcE21 
 MA6051 Gifsy-3[–] 
 MA6052 Gifsy-1[–]Gifsy-2[–]Gifsy-3[–] 
 MA6275 Gifsy-2[–] Figueroa-Bossi and Bossi (1999)
 MA6556 Gifsy-1[–]Gifsy-2[–]Gifsy-3[–]Fels-1[+] 
 MA6601 Gifsy-2[–]Δ[ilvI-leuO]3308::sodC1 Figueroa-Bossi and Bossi (1999)
 MA6729 Gifsy-2[–]Fels-1[+] 
 MA6877Δ[recBD-argA]zfh-8157::Tn10dTc sbcE21 
 MA6909 Gifsy-2[–]gogA1::Cm 
 MA6914 Gifsy-2[–]Δ[gogB]2::Kn 
 MA6915 Gifsy-1[–]Δ[gtgB]1::Kn 
 MA6922 Gifsy-1[–]gtgA1::Cm 
 MA6927Δ[gtgB]1::Kn 
S. typhimurium LT2
 LT2Wild type Lilleengen (1948)
 MA4587 Gifsy-1[–]Gifsy-2[–] Figueroa-Bossi et al. (1997)
 MA6495 Gifsy-3[+] 
 MA6502 Gifsy-1[–]Gifsy-2[–]Gifsy-3[+] 
 MA6595 Gifsy-1[–]Gifsy-2[–] SopEΦ[+] 
S. typhimurium SL1344
 SL1344Wild type Hoiseth and Stocker (1981)
 MA6244 Gifsy-1[–] Figueroa-Bossi and Bossi (1999)
 MA6247 Gifsy-1[–]Gifsy-2[–] Figueroa-Bossi and Bossi (1999)
 MA6498 Gifsy-1[–]Gifsy-2[–]Gifsy-3[+] 
S. typhimurium Q1
 Q1Wild type Boyd and Bidwell (1957)
 Q1 (F1) Fels-1[+] Affolter et al. (1983)
 Q1 (F1 F2) Fels-1[+]Fels-2[+] Affolter et al. (1983)
E. coli K-12
 BW25113 lacI q rrnB T14ΔlacZWJ16hsdR514ΔaraBA-DAH33ΔrhaBADLD78 Datsenko and Wanner (2000)
 BW25141 lacI q rrnB T14ΔlacZWJ16ΔphoBR580 hsdR514ΔaraBADAH33 ΔrhaBADLD78galU95 endABT333uidAMluI)::pir+recA1 Datsenko and Wanner (2000)

Transformation and transduction

Transformation of S. typhimurium or E. coli with plasmid DNA was accomplished by electroporation using a Bio-Rad gene pulser under the conditions specified by the manufacturer. Generalized transduction was carried out using a high-frequency transducing mutant of phage P22, HT 105/1 int-201 (Schmieger, 1972), as described previously (Maloy, 1990).

Preparation of phage and isolation of lysogenic bacteria

Individual patches of phage-infected bacteria (visualized on Green or EBU indicator plates) were spread onto LB plates and incubated at 37°C for several hours. Plate contents were resuspended in LB, and the bulk of bacterial cells was removed by centrifugation. Aliquots (0.5–2 ml) from the supernatants were either used to inoculate cultures for larger scale preparation of phage DNA as described previously (Figueroa-Bossi and Bossi, 1999) or incubated at 37°C with shaking for several more hours. The latter step (growth of the residual bacteria in the presence of excess phage) creates conditions favouring the accumulation of lysogens. After reaching stationary phase, cultures were diluted in order to obtain individual colonies upon plating. Colonies were picked, resuspended in 0.5 ml of LB and aliquots from the suspensions streaked immediately across a trail of phage particles on Green or EBU indicator plates. Lack of a colour change at the intersection of bacteria and phage identified lysogenic bacteria. Typically, these represented between 10% and 60% of the clones tested.

Disruption and deletion/replacement of prophage loci

The disruption of ORFs gogA and gtgA was achieved according to a procedure described previously from our laboratory (Toro et al., 1998). The procedure stems from the observation that S. typhimurium recB mutants carrying sbcE suppressor mutations (Figueroa-Bossi et al., 1997) are particularly proficient in recombination between DNA fragments and the chromosome (Toro et al., 1998). A PCR-amplified fragment spanning the gtgA sequence was cloned into plasmid pGEMT (Promega). The resulting plasmid (pGEM-gtgA) has a unique BamH1 nuclease recognition site, approximately in the middle of the gtgA coding sequence. A 1.3 kb DNA fragment carrying a CmR cassette was obtained upon cleaving plasmid pNK2884 DNA (Kleckner et al., 1991) with BamH1 enzyme. This fragment was purified and ligated to BamH1-cleaved pGEM-gtgA yielding plasmid pGEM-gtgA1::Cm. Treatment with EagI restriction endonuclease released the gtgA::Cm insert; this was gel purified and used for transformation of S. typhimurium strain MA6877, a recBD sbcE mutant derivative of strain ATCC14028. In the vast majority of CmR transformants obtained, the CmR cassette was found to be inserted, with comparable frequencies, at either the gogA (Gifsy-1) or the gtgA (Gifsy-2) loci. Deletion/replacement of ORFs gogB and gtgB was achieved according to the protocol developed recently by Datsenko and Wanner (2000). Template plasmid pKD4 was amplified with appropriately designed primers, and the resulting PCR products were introduced by electroporation into an S. typhimurium ATCC14028 strain expressing phage lambda Red functions (from plasmid pKD46). KnR transformants were confirmed to carry the KnR cassette at the expected locations: replacing the segment between co-ordinates 179 and 1702 of the gogB sequence (GenBank AF254761) and the segment between co-ordinates 40 and 1219 of the gtgB sequence (GenBank AF254763).

Pulsed-field gel electrophoresis (PFGE) and Southern hybridization

Bacterial cells were embedded in agarose plugs, gently lysed and chromosomal DNA digested with XbaI restriction endonuclease in situ as described previously (Figueroa-Bossi et al., 1997). PFGE was carried out at 6 V cm−1 with a pulse frequency of 6 s for 15 h followed by 5 h 30 min at a pulse frequency of 60 s. DNA fragments were stained by soaking gel in 1 µg ml−1 ethidium bromide. Fragments were then vacuum blotted onto nylon membranes and hybridized to DNA probes (plasmid or PCR products) 32P-labelled with the Prime-a-Gene system (Promega), according to the manufacturer's instructions.

PCR amplification and DNA sequencing

PCR amplification from bacterial colonies or from purified DNA was carried out using Taq DNA polymerase (Appligene-Oncor) according to standard protocols. In gene replacement experiments (above), a TaqPfu polymerase mix (4:1) was used. The oligonucleotides used as primers for the PCR were chemically synthesized by Isoprim or ESGS. DNA sequence analysis was carried out by Cybergene. Sequence comparisons were made using the blast algorithm (Altschul et al., 1997).

Virulence assays

Female BALB/c mice (7–8 weeks old) were inoculated orally or intraperitoneally as described previously (Figueroa-Bossi and Bossi, 1999).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We are grateful to Jean Lepault for help with the electron microscopy examinations. Curing S. typhi for the resident Gifsy-2 prophage and isolation of a derivative lysogenized by the Gifsy-2 phage from S. typhimurium was carried out during a summer stay by N.F.-B. and L.B. in the laboratory of Stanley Maloy, University of Illinois, Urbana-Champaign, IL, USA; we are grateful to Stan for his hospitality and support, and to the Philippe Foundation for support to N.F.-B. We thank Jeff Gardner and Stan Maloy for comments on the manuscript, and Theresa Ho and Jim Slauch for sharing unpublished data. We are grateful to Diana Downs and Alan Anderson for the gift of strains. This work was supported by a grant from the French Ministry of Public Education.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Affolter, M., Parent-Vaugeois, C. & Anderson, A. (1983) Curing and induction of the Fels 1 and Fels 2 prophages in the Ames mutagen tester strains of Salmonella typhimurium. Mutat Res 110: 243262.
  • Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W. & Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 33893402.DOI: 10.1093/nar/25.17.3389
  • Blattner, F.R., Plunkett, G., III, Bloch, C.A., Perna, N.T., Burland, V. & Riley, M., et al. (1997) The complete genome sequence of Escherichia coli K-12. Science 277: 14531474.DOI: 10.1126/science.277.5331.1453
  • Bochner, B.R. (1984) Curing bacterial cells of lysogenic viruses by using UCB indicator plates. Biotechniques 2: 234240.
  • Boyd, J.S.K. & Bidwell, D.E. (1957) The type A phages of Salmonella typhimurium: identification by a standardized cross-immunity test. J Gen Microbiol 16: 217228.
  • Chan, R.K., Botstein, D., Watanabe, T. & Ogata, Y. (1972) Specialized transduction of tetracycline resistance by phage P22 in Salmonella typhimurium. II. Properties of a high-frequency-transducing lysate. Virology 50: 883898.
  • Daniels, D.L., Schroeder, J.L., Szybalski, W., Sanger, F. & Blattner, F.R. (1983) A molecular map of coliphage lambda. In Lambda II. Hendrix, R.W., Roberts, J.W., Stahl, F.W., and Weisberg, R.A. (eds). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, pp. 469517.
  • Datsenko, K.A. & Wanner, B.L. (2000) One-step inactivation of chromosomal genes in Escherichia coli K12 using PCR products. Proc Natl Acad Sci USA 97: 66406645.DOI: 10.1073/pnas.120163297
  • De Groote, M.A., Ochsner, U.A., Shiloh, M.U., Nathan, C., McCord, J.M. & Dinauer, M.C., et al. (1997) Periplasmic superoxide dismutase protects Salmonella from products of phagocyte NADPH-oxidase and nitric oxide synthase. Proc Natl Acad Sci USA 94: 1399714001.DOI: 10.1073/pnas.94.25.13997
  • Fang, F.C., DeGroote, M.A., Foster, J.W., Bäumler, A.J., Ochsner, U. & Testerman, T., et al. (1999) Virulent Salmonella typhimurium has two periplasmic Cu, Zn-superoxide dismutases. Proc Natl Acad Sci USA 96: 75027507.DOI: 10.1073/pnas.96.13.7502
  • Farrant, J.L., Sansone, A., Canvin, J.R., Pallen, M.J., Langford, P.R. & Wallis, T.S., et al. (1997) Bacterial copper- and zinc-cofactored superoxide dismutase contributes to the pathogenesis of systemic salmonellosis. Mol Microbiol 25: 785796.
  • Fields, P.I., Swanson, R.V., Haidaris, C.G. & Heffron, F. (1986) Mutants of Salmonella typhimurium that cannot survive within the macrophage are avirulent. Proc Natl Acad Sci USA 83: 51895193.
  • Figueroa-Bossi, N. & Bossi, L. (1999) Inducible prophages contribute to Salmonella virulence in mice. Mol Microbiol 33: 167176.
  • Figueroa-Bossi, N., Coissac, E., Netter, P. & Bossi, L. (1997) Unsuspected prophage-like elements in Salmonella typhimurium. Mol Microbiol 25: 161173.
  • Galán, J.E. (1999) Interaction of Salmonella with host cells through the centisome 63 type III secretion system. Curr Opin Microbiol 2: 4650.DOI: 10.1016/s1369-5274(99)80008-3
  • Galitski, T. & Roth, J.R. (1996) A search for a general phenomenon of adaptive mutability. Genetics 143: 645659.
  • Gunn, J.S., Alpuche-Aranda, C.M., Loomis, W.P., Belden, W.J. & Miller, S.I. (1995) Characterization of the Salmonella typhimurium pagC/pagD chromosomal region. J Bacteriol 177: 50405047.
  • Gunn, J.S., Belden, W.J. & Miller, S.I. (1998) Identification of PhoP–PhoQ activated genes within a duplicated region of the Salmonella typhimurium chromosome. Microb Pathog 25: 7790.DOI: 10.1006/mpat.1998.0217
  • Haggård-Ljungquist, E., Halling, C. & Calender, R. (1992) DNA sequences of the tail fiber genes of bacteriophage P2: evidence for horizontal transfer of tail fiber genes among unrelated bacteriophages. J Bacteriol 174: 14621477.
  • Hardt, W.D., Urlaub, H. & Galán, J.E. (1998) A substrate of the centisome 63 type III protein secretion system of Salmonella typhimurium is encoded by a cryptic bacteriophage. Proc Natl Acad Sci USA 95: 25742579.DOI: 10.1073/pnas.95.5.2574
  • Hensel, M. (2000) Salmonella pathogenicity island 2. Mol Microbiol 36: 10151023.DOI: 10.1046/j.1365-2958.2000.01935.x
  • Hoiseth, S.K. & Stocker, B.A.D. (1981) Aromatic-dependent Salmonella typhimurium are non-virulent and effective as live vaccines. Nature 291: 238239.
  • Hoyer, L.L., Hamilton, A.C., Steenbergen, S.M. & Vimr, E.R. (1992) Cloning, sequencing and distribution of the Salmonella typhimurium LT2 sialidase gene, nanH, provides evidence for interspecies gene transfer. Mol Microbiol 6: 873884.
  • Hueck, C.J. (1998) Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol Mol Biol Rev 62: 379433.
  • Imlay, K.R. & Imlay, J.A. (1996) Cloning and analysis of sodC, encoding the copper–zinc superoxide dismutase of Escherichia coli. J Bacteriol 178: 25642571.
  • Juhala, R.J., Ford, M.E., Duda, R.L., Youlton, A., Hatfull, G.F. & Hendrix, R.W. (2000) Genomic sequences of bacteriophages HK97 and HK022: pervasive genetic mosaicism in the lambdoid bacteriophages. J Mol Biol 299: 2751.
  • Kingsley, R.A. & Bäumler, A.J. (2000) Host adaptation and the emergence of infectious disease: the Salmonella paradigm. Mol Microbiol 36: 10061014.DOI: 10.1046/j.1365-2958.2000.01907.x
  • Kleckner, N., Bender, J. & Gottesman, S. (1991) Uses of transposons with emphasis on Tn10. Methods Enzymol 204: 139180.
  • Leung, K.Y. & Straley, S.C. (1989) The yopM gene of Yersinia pestis encodes a released protein having homology with the human platelet surface protein GPIb alpha. J Bacteriol 171: 46234632.
  • Lilleengen, K. (1948) Typing Salmonella typhimurium by means of bacteriophage. Acta Pathol Microbiol Scand Suppl 77: 11125.
  • Maloy, S.R. (1990) Experimental Techniques in Bacterial Genetics. Boston: Jones and Bartlett.
  • Matsutani, S., Ohtsubo, H., Maeda, Y. & Ohtsubo, E. (1987) Isolation and characterization of IS elements repeated in the bacterial chromosome. J Mol Biol 196: 445455.
  • Miao, E.A. & Miller, S.I. (2000) A conserved amino acid sequence directing intracellular type III secretion by Salmonella typhimurium. Proc Natl Acad Sci USA 97: 75397544.DOI: 10.1073/pnas.97.13.7539
  • Miao, E.A., Scherer, C.A., Tsolis, R.M., Kingsley, R.A., Adams, L.G., Bäumler, A.J. & Miller, S.I. (1999) Salmonella typhimurium leucine-rich repeat proteins are targeted to the SPI1 and SPI2 type III secretion systems. Mol Microbiol 34: 850864.
  • Miller, S.I. & Pegues, D.A. (2000) Salmonella species, including Salmonella typhi. In Principles and Practice of Infectious Diseases. Mandell, G.L., Bennett, J.E., and Dolin, R. (eds). Philadelphia, PA: Churchill Livingstone, pp. 23442363.
  • Miller, V.L., Bliska, J.B. & Falkow, S. (1990) Nucleotide sequence of the Yersinia enterocolitica ail gene and characterization of the Ail protein product. J Bacteriol 172: 10621069.
  • Mirold, S., Rabsch, W., Rohde, M., Stender, S., Tschape, H. & Russmann, H., et al. (1999) Isolation of a temperate bacteriophage encoding the type III effector protein SopE from an epidemic Salmonella typhimurium strain. Proc Natl Acad Sci USA 96: 98459850.DOI: 10.1073/pnas.96.17.9845
  • Montag, D., Schwarz, H. & Henning, U. (1989) A component of the side tail fiber of Escherichia coli bacteriophage λ can functionally replace the receptor-recognizing part of a long tail fiber protein of the unrelated bacteriophage T4. J Bacteriol 171: 43784384.
  • Ochman, H., Lawrence, J.G. & Groisman, E.A. (2000) Lateral gene transfer and the nature of bacterial innovation. Nature 405: 299304.
  • Pesce, A., Battistoni, A., Stroppolo, M.E., Polizio, F., Nardini, M. & Kroll, J.S., et al. (2000) Functional and crystallographic characterization of Salmonella typhimurium Cu,Zn superoxide dismutase coded by the sodCI virulence gene. J Mol Biol 302: 465478.DOI: 10.1006/jmbi.2000.4074
  • Plasterk, R.H. & Van De Putte, P. (1985) The invertible P-DNA segment in the chromosome of Escherichia coli. EMBO J 4: 237242.
  • Perna, N.T., Mayhew, G.F., Posfay, G., Elliott, S., Donnenberg, M.S., Kaper, J.B. & Blattner, F.R. (1998) Molecular evolution of a pathogenicity island from enterohemorrhagic Escherichia coli O157:H7. Infect Immun 66: 38103817.
  • Raha, M., Kihara, M., Kawagishi, I. & Macnab, R.M. (1993) Organization of the Escherichia coli and Salmonella typhimurium chromosomes between flagellar regions IIIa and IIIb, including a large non-coding region. J Gen Microbiol 139: 14011407.
  • Roggentin, P., Schauer, R., Hoyer, L.L. & Vimr, E.R. (1993) The sialidase superfamily and its spread by horizontal gene transfer. Mol Microbiol 9: 915921.
  • Schechter, L.M. & Lee, C.A. (2000) Salmonella invasion of non-phagocytic cells. In Subcellular Biochemistry, Vol. 33. Oelschlaeger, T., and Hacker, J. (eds). New York: Kluwer Academic/Plenum Publishers, pp. 289320.
  • Schmieger, H. (1972) Phage P22 mutants with increased or decreased transduction activities. Mol Gen Genet 119: 7588.
  • Smith, C.M., Koch, W.H., Franklin, S.B., Foster, P.L., Cebula, T.A. & Eisenstadt, E. (1990) Sequence analysis and mapping of the Salmonella typhimurium LT2 umuDC operon. J Bacteriol 172: 49644978.
  • Stanley, T.L., Ellermeier, C.D. & Slauch, J.M. (2000) Tissue-specific gene expression identifies a gene in the lysogenic phage Gifsy-1 that affects Salmonella enterica serovar Typhimurium survival in Peyer patches. J Bacteriol 182: 44064413 .
  • Tobe, T., Hayashi, T., Han, C.G., Schoolnik, G.K., Ohtsubo, E. & Sasakawa, C. (1999) Complete DNA sequence and structural analysis of the enteropathogenic Escherichia coli adherence factor plasmid. Infect Immun 67: 54555462.
  • Toro, C.S., Mora, G.C. & Figueroa-Bossi, N. (1998) Gene transfer between related bacteria by electrotransformation: mapping Salmonella typhi genes in Salmonella typhimurium. J Bacteriol 180: 47504752.
  • Tsolis, R.M., Townsend, S.M., Miao, E.A., Miller, S.I., Ficht, T.A., Adams, L.G. & Bäumler, A.J. (1999) Identification of a putative Salmonella enterica serotype Typhimurium host range factor with homology to IpaH and YopM by signature-tagged mutagenesis. Infect Immun 67: 63856393.
  • Uzzau, S., Brown, D.J., Wallis, T., Rubino, S., Leori, G. & Bernard, S., et al. (2000) Host adapted serotypes of Salmonella enterica. Epidemiol Infect 125: (in press).
  • Valdivia, R.H. & Falkow, S. (1997) Fluorescence-based isolation of bacterial genes expressed within host cells. Science 277: 20072011.DOI: 10.1126/science.277.5334.2007
  • Wood, M.W., Jones, M.A., Watson, P.R., Hedges, S., Wallis, T.S. & Galyov, E.E. (1998) Identification of a pathogenicity island required for Salmonella enteropathogenicity. Mol Microbiol 29: 883891.
  • Worley, M.J., Ching, K.H.L. & Heffron, F. (2000) Salmonella SsrB activates a global regulon of horizontally acquired genes. Mol Microbiol 36: 749761.DOI: 10.1046/j.1365-2958.2000.01902.x
Footnotes
  1. Present address: Dipartimento di Scienze Biomediche, Università di Sassari, 07100 Sassari, Italy.