Genomic analysis and growth-phase-dependent regulation of the SEF14 fimbriae of Salmonella enterica serovar Enteritidis

Robert A. Edwardsa,1, Brian C. Matlock1, Brian J. Heffernan1 and Stanley R. Maloy1

Department of Microbiology, University of Illinois, Urbana, IL 61801, USA1

Author for correspondence: Robert A. Edwards. Tel: +1 901 448 8101. Fax: +1 901 448 8462. e-mail: redwards{at}utmem.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Salmonella enterica serovar Enteritidis is a leading cause of food poisoning in the USA and Europe. Although Salmonella serovars share many fimbrial operons, a few fimbriae are limited to specific Samonella serovars. SEF14 fimbriae are restricted to group D Salmonella and the genes encoding this virulence factor were acquired relatively recently. Genomic, genetic and gene expression studies have been integrated to investigate the ancestry, regulation and expression of the sef genes. Genomic comparisons of the Salmonella serovars sequenced revealed that the sef operon is inserted in leuX in Salmonella Enteritidis, Salmonella Paratyphi and Salmonella Typhi, and revealed the presence of a previously unidentified 25 kb pathogenicity island in Salmonella Typhimurium at this location. Salmonella Enteritidis contains a region of homology between the Salmonella virulence plasmid and the chromosome downstream of the sef operon. The sef operon itself consists of four co-transcribed genes, sefABCD, and adjacent to sefD there is an AraC-like transcriptional activator that is required for expression of the sef genes. Expression of the sef genes was optimal during growth in late exponential phase and was repressed during stationary phase. The regulation was coordinated by the RpoS sigma factor.

Keywords: pili, salmonellosis, regulation, genomics, pathogenicity islands

The GenBank accession number for the sequence reported in this paper is AF239978.

a Present address: Department of Molecular Sciences, University of Tennessee Health Sciences Center, MSB101 858 Madison Ave, Memphis, TN 38163, USA.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Fimbrial adhesins mediate interactions between bacteria and solid surfaces, including attachment to host cells. Among the pathogenic enteric bacteria multiple types of fimbriae are essential for a virulent lifestyle. For example, Salmonella possess fimbriae that bind to different host tissues, either intestinal epithelial tissues, specific intestinal cells involved in the immune response (M-cells) or macrophages (Bäumler et al., 1996a , b ; Edwards et al., 2000 ). Fimbriae with limited distribution and specific roles in virulence are thought to contribute to some of the unique aspects of virulence not seen in closely related bacteria.

Most fimbriae (with the notable exception of the type-IV pili) have a similar mechanism of translocation and biogenesis. The fimbrial shaft is composed of a major subunit that is translocated across the cytoplasmic membrane via the general secretory (sec-dependent) system. In the periplasm the subunits are prevented from premature aggregation by binding to a chaperone. Resolution of the crystal structure of two fimbrial chaperone/subunit complexes elegantly revealed the interactions involved in regulating the periplasmic embrace between these two partners, simultaneously preventing premature subunit–subunit interactions but allowing subunit translocation across the outer membrane (Choudhury et al., 1999 ; Sauer et al., 1999 ). A specific usher protein mediates export across the outer membrane, with a distinct chaperone–usher pair encoded by each fimbrial operon. Although chaperones and ushers from different fimbrial systems are very homologous, the usher and chaperone work together and are normally specific for the fimbrial subunits – a chaperone and usher pair will not usually mediate translocation of fimbrial subunits from another operon. Many fimbrial systems also encode one or more minor components (so called because they are much less abundant than the major subunit) that are exported as part of the fimbrial structure (Edwards et al., 1996 ; Edwards & Puente, 1998 ). These subunits often form the adhesin that mediates the interaction between the fimbriae and the solid surface (Khan & Schifferli, 1994 ).

A variety of fimbriae have previously been characterized in Salmonella enterica, including the plasmid-encoded (PEF), long polar (LPF), thin aggregative (AGF) and type I (FIM) fimbriae (Bäumler et al., 1997 ; Townsend et al., 2001 ). Many of the fimbriae from Salmonella enterica subsp. enterica serovar Enteritidis are similar to fimbriae from Salmonella Typhimurium, although a number of these fimbriae were initially given alternate names (for example the Salmonella Enteritidis AGF homologue was initially called SEF17 and the FIM homologue was initially called SEF21). In addition to the fimbriae that are shared between serovars, Salmonella Enteritidis contains fimbriae such as the Salmonella Enteritidis SEF14 fimbriae that are not found in Salmonella Typhimurium. The SEF14 fimbriae require four proteins for biogenesis: the major subunit (SefA), chaperone (SefB) and usher (SefC) have previously been characterized (Clouthier et al., 1993 ; Thorns et al., 1990 ; Turcotte & Woodward, 1993 ). The minor subunit (SefD) was also previously identified but not thought to be part of the SEF14 fimbriae (Clouthier et al., 1994 ).

We recently reported the role of the SEF14 fimbriae in virulence and showed that SEF14 fimbriae are essential for binding to macrophages (Edwards et al., 2000 ). In this report we describe a multi-faceted approach to dissecting the sef operon, utilizing genomics and genetics to define and inactivate genes. We identify the principal regulatory element that controls sef gene expression and characterize environmental regulation of sef gene expression.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
DNA analysis.
The Salmonella Typhi genomic DNA sequence was obtained from the Sanger Centre (http://www.sanger.ac.uk; J. Parkhill, unpublished), the Salmonella Typhimurium and Salmonella Paratyphi genomic DNA sequences were obtained from the Washington University St Louis Sequencing Center (http://genome.wustl.edu/; M. McClelland, unpublished) and the Salmonella Enteritidis genome is currently being sequenced at the University of Illinois (R. A. Edwards, D. N. Green, G. J. Olsen & S. R. Maloy, unpublished). DNA sequence from the Salmonella Enteritidis project and further information about the genome sequencing projects is available from http://www.salmonella.org. BLAST searches were performed either via the NCBI website (http://www.ncbi.nlm.nih.gov) or locally using the WU-BLAST executables (Altschul et al., 1990 , 1994 ; Altschul & Gish, 1996 ; http://blast.wustl.edu). Salmonella Enteritidis DNA was sequenced either from clones (described below) or directly from PCR products. Both types of samples were sequenced using ABI big-dye termination sequencing by the University of Illinois Biotechnology Center (http://www.life.uiuc.edu/biotech). To generate PCR products, primers were designed to be identical (except for the introduction of restriction endonuclease sites) to either the known Salmonella Enteritidis sequence (where available) or to the corresponding sequence in Salmonella Typhi or Salmonella Typhimurium as appropriate. PCR was performed as described elsewhere (Maloy et al., 1996 ) using annealing temperatures calculated from the G+C content of the primers. PCR products were analysed by gel electrophoresis and purified using the Wizard PCR purification kit (Promega) prior to sequencing.

Bacterial strains, plasmids, phage and growth conditions.
All bacterial strains, plasmids and phage used in this study are given in Table 1. All strains were grown in Luria–Bertani (LB) medium, supplemented with the following antibiotics where necessary (final concentrations, µg ml-1): kanamycin, 50; chloramphenicol, 30; tetracycline, 10; ampicillin, 90. All transductions were performed using P22 HT int. Transductants were purified on Evans Blue Uranine (EBU) plates and checked for lysogens by cross-streaking against phage P22 H5 as described previously (Maloy et al., 1996 ).


View this table:
[in this window]
[in a new window]
 
Table 1. Strains and plasmids used in this study

 
Strain construction.
To facilitate genetic manipulation sefD was amplified by PCR and cloned into the {pi}-dependent suicide vector pDMS197. The {pi}-dependent plasmids were routinely maintained in MST4718, a {pi}+ derivative of wild-type Salmonella Typhimurium strain LT2 (A. E. Stanley & S. R. Maloy, unpublished). To construct lacZ operon fusions between sefD and MudJ, MST4718(pRE204) was used as a recipient in transductions with phage P22 grown on TT10289 (Maloy et al., 1996 ). Transitory cis-complementation of the transposase-deficient MudJ allows random insertion events, a fraction of which will be in the plasmid. Potential fusions were identified by the co-localization of TetR from pDMS197 and KanR from MudJ on the same plasmid. Transposon insertions with MudJ fused to sefD were identified by PCR using one primer that hybridizes in sefD (5'-GGTCAGAATTCCTCAGCCCATAA-3') and a second primer that hybridized in the trpAB region of MudJ (5'-CAGGAAGATCGCACTCCAGCCAGCT-3'). The PCR products were sequenced to confirm the exact point of insertion of the MudJ element.

To introduce the mutations onto the chromosome, the suicide plasmid was electroporated either into TYT3362 (a sef+ derivative of Salmonella Typhimurium LT2 whose construction is described below) or wild-type LK5 with selection for the MudJ-encoded KanR marker. Enrichment for second recombination events that eliminated the suicide vector was performed by growth on sucrose (Edwards et al., 1998 ) and confirmed by checking for loss of the plasmid-encoded TetR marker and by PCR to confirm inheritance of the mutant allele.

To construct both the constitutively expressed sefR construct (pRE234) and the fusion between sefR and the arabinose-inducible promoter (pRE235), a sefR PCR product was cloned into either pUC19 or pBAD18, respectively, using primers 5'-GGTCAGAATTCCTCAGCCCATAA-3' and 5'-CGGAGTCTGCAGTGAAGCGTAAAAA-3'. This 1078 bp product encodes the 3' end of the sefD gene, all of the sefR gene and 38 bp upstream of sefR. The expression of sefR in the pUC19 clone is controlled by the lac promoter in this plasmid, and the pBAD18 construct contains the arabinose-inducible promoter and araC, the gene encoding the transcriptional activator for expression from that promoter (Guzman et al., 1995 ).

Identification of linked markers and genetic manipulations.
To identify markers linked to the sef island, plasmid pBSL142 (a delivery vehicle for Tn5dGen; Alexeyev et al., 1995 ) was electroporated into Salmonella Enteritidis sefA1::Kan that had been heated to inactivate the restriction modification system (Edwards et al., 1999 ). Because pBSL142 is a suicide plasmid, any GenR colonies should result from transposition events. Colonies were pooled and used as donors in a second transduction with wild-type Salmonella Enteritidis as the recipient, selecting for both GenR and KanR simultaneously. The resulting colonies contain both sefA1::Kan and a linked Tn5dGen. The GenR marker was subsequently separated from the sefA1::Kan allele by transduction. Several GenR alleles were transduced from Salmonella Enteritidis to Salmonella Typhimurium LT2. The transductants were screened for the presence of the sef locus by a PCR assay for each of the sef genes.

To transduce the sefD::MudJ fusion into a strain harbouring sefA1::Kan, the fusion was linked to a GenR allele that lies outside of the sef island and has no effect on the regulation of expression of the sefD::MudJ fusion. The markers were moved into Salmonella Enteritidis LK19 by transduction followed by selection for GenR and screening for both sefA1::Kan and sefD::MudJ by PCR. The presence of sefB and lacZ were confirmed by PCR both before and after expression studies to confirm that the sefA1::Kan and MudJ elements did not recombine. If the sefA1::Kan and sefD::MudJ elements recombined both sefB and lacZ would be deleted. Although this may happen at a very low frequency, such recombination events were never observed.

One of the Tn5dGen mutations located very close to the sef operon, but not within the operon itself co-transduces with sefA::Kan approximately 80% of the time. This insertion does not affect virulence and lies outside the sef insertion in the genome. This transposon insertion was used to backcross the {Delta}sefA mutation into an otherwise wild-type background by isolation of a strain with the transposon linked to the in-frame deletion, followed by P22 transductions of the transposon into LK5 and screening for co-inheritance of the {Delta}sefA allele.

ß-Galactosidase assays.
Assays were performed as described elsewhere and the activities are reported in Slauch units (Maloy, 1990 ; Slauch & Silhavy, 1991 ).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Analysis of the sef operon
The sef fimbrial operon (Fig. 1) contains all the genes required for biogenesis of the SEF14 fimbriae. sefA encodes the most abundant subunit of the fimbrial shaft and sefD possibly encodes the tip-located adhesin (R. Edwards, unpublished). Based on the high level of sequence homology with previously characterized fimbrial proteins, the proteins encoded by sefB and sefC are the periplasmic-located chaperone and outer-membrane-located usher, respectively. We show below that the fifth sef gene, sefR, encodes a transcriptional regulator of the sef operon. The dlp gene located downstream of sefR encodes a thiol:disulphide oxidoreductase that is a homologue of dsbA. This gene was initially identified as a homologous region between the Salmonella Enteritidis virulence plasmid and the Salmonella Typhi genome (Rodriguez-Pena et al., 1997 ). As Salmonella Typhi does not contain the virulence plasmid, the Salmonella Typhi dlp homologue (called dlt) must be located on the chromosome in this serovar. Analysis of the Salmonella Typhimurium genome sequence suggests that the only dlp homologue present in this serovar is located on the virulence plasmid adjacent to a different fimbrial operon (pef). A region that extends downstream from Salmonella Enteritidis sefD was PCR-amplified and sequenced, and showed that both sef and dlp are located on the chromosome in Salmonella Enteritidis. Furthermore, our genome sequencing has confirmed the chromosomal location of sef in Salmonella Enteritidis. We have sequenced the entire virulence plasmid, including the Salmonella Enteritidis pef region and the dlp homologue adjacent to this, and have not identified any virulence plasmid-encoded sef sequences. In Salmonella Enteritidis alone, dlp is on both the chromosome and the virulence plasmid.



View larger version (18K):
[in this window]
[in a new window]
 
Fig. 1. Organization of the sef gene cluster and adjacent DNA. The sef gene cluster lies between leuX and trpS. In Salmonella Enteritidis the five sef genes are separated from leuX by a single insertion element. In Salmonella Typhi a complete P4 phage (approx. 11·5 kb) is inserted at leuX. This phage is absent from Salmonella Enteritidis. There are no sef genes in Salmonella Typhimurium, but this position on the chromosome is occupied with a previously unidentified and uncharacterized pathogenicity island. A single ORF on this island has been identified by random mutagenesis to be induced when bacteria are internalized by macrophages (Pfeifer et al., 1999 ). (a) Schematic organization of the region surrounding the sef insertion. (b) Organization of the sef operon.

 
Genomic comparisons
The sef island was initially mapped to 97·6 centisomes (cs) on the Salmonella Enteritidis chromosome (Collinson et al., 1996 ). The order of genes surrounding the sef island – valS (97·3 cs), pepA (97·4 cs), leuX (previously unmapped), sef, dlp (previously unmapped), mcrD (previously unmapped) and trpRS (99·9 cs) – confirmed this observation (Sanderson et al., 1996 ). From the comparison of several closely related sequences, the sef insertion is located within the leuX gene, which encodes a leucine-tRNA. The sef island has a mean G+C content of 35·4 mol%, significantly less than the mean for the Salmonella chromosome of 51·9 mol%. By comparing the G+C content profile the junctions of the sef operon with the ancestral chromosome could be identified. The junction adjacent to sefA occurs at an IS1230 element; however, there does not appear to be any ancestral insertion scars in the genomic sequence at the sefR end of the island. The sef insertion per se is 6751 bp, extended from 779 bases upstream of sefA to 926 bases downstream of sefR (data not shown). This analysis suggests that dlp is not part of the sef island and may have been acquired independently.

Three group D Salmonella serovars have been shown to contain the sef operon inserted at leuX: Salmonella Enteritidis, Salmonella Typhi and Salmonella Paratyphi. Phage genomes separate leuX and the IS1230 element/sef island in the latter two serovars, but not in Salmonella Enteritidis (Fig. 1). Sequencing data suggests that the Salmonella Typhi sef operon contains multiple frameshift mutations. These have been confirmed by independent sequencing, suggesting that the sef genes are not expressed in Salmonella Typhi (Townsend et al., 2001 ).

Unlike Salmonella Typhi and Salmonella Typhimurium, there is no phage genome inserted between the IS1230 element and leuX in Salmonella Enteritidis. However, the entire leuX gene has been substituted by a putative transposon containing a different leuX gene. The transposon contains 16 bp direct inverted repeats that delimit the foreign DNA inserted into the Salmonella Enteritidis genome and a transposase gene immediately adjacent to one inverted repeat (Fig. 1).

Salmonella Typhimurium, which is not a member of the group D Salmonella, does not contain any of the sef island, IS1230 element, transposon or phage genome at leuX. Based upon analysis of the Salmonella Typhimurium genome sequence, we have identified a previously uncharacterized 25337 bp pathogenicity island inserted at this site (data not shown). This island contains very limited homology to sequences in the databases; however, one ORF is similar to the Shigella flexneri virulence gene, shiB, and another was identified as being specifically induced inside macrophages (Pfeifer et al., 1999 ). This island has not been characterized further; however, it indicates that as in Escherichia coli there have been multiple independent acquisitions of pathogenicity islands at leuX in the Salmonella chromosome: the sef island was probably inherited after group D Salmonella diverged from other groups and Salmonella Typhimurium inherited a separate island.

The sequence of the Salmonella Enteritidis sef region from more than 1 kb upstream of sefA to a few base pairs downstream of sefD was available from GenBank as several different sequences (accession nos L11008, L11009, L11010, U07129 and X98516); however, sequence information in the region downstream of sefD was not available. Using primers designed to be compatible with Salmonella Enteritidis sefD and Salmonella Typhi mcrD regions, the DNA downstream of sefD was PCR-amplified and sequenced. The sequence adjacent to sefD in Salmonella Enteritidis was similar to the corresponding region from the Salmonella Typhi genome and contains both sefR, encoding the transcriptional activator, and dlp, encoding the thiol:disulphide oxidoreductase. The sequence was compiled together with the previously published sequences and the entire contig was deposited in the GenBank database with the accession number AF239978. The sequence of adjacent regions in the Salmonella Enteritidis genome have been released as part of the genome sequencing project and can be retrieved from http://www.salmonella.org.

Transduction of the sef island to Salmonella Typhimurium LT2
A genetic approach was taken to characterize the islands and genes that were identified from the genomic analysis. Salmonella Enteritidis contains a potent restriction system that limits genetic manipulations in this strain. The restriction system was circumvented by moving the entire sef island from Salmonella Enteritidis to Salmonella Typhimurium. Several Tn5dGen transposon insertions linked to the sef operon were identified. Following transduction from Salmonella Enteritidis to Salmonella Typhimurium two classes of insertion were found. The first were 100% linked to the sefA operon in transductions between Salmonella Enteritidis and Salmonella Typhimurium, although there was variable linkage in transductions between Salmonella Enteritidis and Salmonella Enteritidis. This class comprised insertions located in the sef island (Fig. 2a). The Salmonella Enteritidis–Salmonella Enteritidis transductions reflect the expected genetic linkage based upon physical distance between the two markers on the chromosome, but because Salmonella Typhimurium does not contain the sef island, transduction of the Tn5dGen from Salmonella Enteritidis to Salmonella Typhimurium always resulted in inheritance of the sef island. The second class of insertions had a greatly reduced linkage with the sef operon in transductions between Salmonella Enteritidis and Salmonella Typhimurium compared to transductions between Salmonella Enteritidis and Salmonella Enteritidis. This class comprised insertions near the sef island but in chromosomal DNA that is common between Salmonella Enteritidis and Salmonella Typhimurium. Transduction of these markers did not demand inheritance of the sef genes (Fig. 2b).



View larger version (9K):
[in this window]
[in a new window]
 
Fig. 2. Transduction of sef-linked markers. The sef genes are on an island not found in Salmonella Typhimurium. Linked transposons can either lie within the island (a) or outside the island, on a region also found in Salmonella Typhimurium (b). Crosses selecting for the transposon will result in different genotypes depending on the location of the TnGen. If the Tn lies within the sef island it will be 100% linked with the sef genes in crosses between Salmonella Enteritidis and Salmonella Typhimurium (a). All recipients must receive the sef island. If the Tn lies outside the island (b) recombination can occur between the island and the Tn (i), resulting in inheritance of only the Tn, or distal to the island (ii), resulting in inheritance of the island and the Tn. In this case not all recipients receive the sef island. In crosses between Salmonella Enteritidis and Salmonella Enteritidis normal genetic linkage will be observed.

 
SefR regulates sef gene expression
The sequence downstream of sefD contained a potential regulator of sef gene expression. This putative gene, which we called sefR, encodes a protein with high homology to members of the AraC-like family of transcriptional activators, with highest homology to the fimbrial regulator subclass of activators (Gallegos et al., 1997 ) (Fig. 3). Furthermore sefR is located immediately downstream of the fimbrial operon and is convergently transcribed with this operon, a common location for activators of gene expression (Gaastra & Svennerholm, 1996 ). To determine whether SefR regulates sef gene expression, we tested the effect of SefR on a sefD::MudJ fusion that places the lacZ gene of MudJ under the control of the sefD promoter.



View larger version (104K):
[in this window]
[in a new window]
 
Fig. 3. Alignment of SefR predicted sequence with representative AraC-like transcriptional activators. Salmonella Enteritidis SefR (SEFR_SENT) shows a high degree of similarity to other members of the AraC family of transcriptional activators, particularly those that regulate fimbrial gene expression. Similarity is most conserved over the C-terminal DNA binding region. Identical residues are shown in inverse type. BfpT (BFPT_ECOLI), FasH (FASH_ECOLI), CfaD (CFAD_ECOLI) and AggR (AGGR_ECOLI) regulate expression of fimbrial operons in E. coli. VirF (VIRF_SHIDY) regulates expression of virulence genes in Shigella spp. (Caron et al., 1989 ; Edwards & Schifferli, 1997 ; Jost & Adler, 1993 ; Nataro et al., 1994 ; Tobe et al., 1996 ).

 
Like other Salmonella, wild-type Salmonella Enteritidis is naturally lac-, thus no ß-galactosidase activity was detected in this strain. Low levels of ß-galactosidase activity were measured from the sefD::MudJ fusion in an otherwise wild-type cell. However, when sefR expressed from the arabinose-inducible BAD promoter was introduced into the sefD::MudJ strain, ß-galactosidase expression was increased even in the absence of arabinose (Fig. 4). When sefR expression was induced by adding 0·2% arabinose to the medium very high levels of ß-galactosidase expression were attained (Fig. 4). Together these results indicate that SefR is indeed the activator of sef gene expression.



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 4. Expression of the sef operon. Low background levels of expression are seen without exogenous sefR. In the absence of arabinose (open bars) pBAD-sefR increased expression of sefD but expression of sefR was unaffected. When 0·2% arabinose was added (solid bars) expression of the sefD::lacZ fusion increased further. When sefA was inactivated by the kanamycin cassette, expression of the lacZ fusion was never detected, confirming the polarity of this insertion. When sefA is inactivated by the non-polar in-frame deletion, expression of the lacZ fusion was unaffected. Solid bars represent cultures induced with 0·2% arabinose, open bars represent uninduced cultures. Data are representative of numerous experiments.

 
From the sequences of the sef island, no potential promoter elements between sefA and sefD were identified that could control expression of sefB, sefC or sefD, suggesting that the sef genes are transcribed as a single operon from a promoter upstream of sefA. The sefD::MudJ fusion was transduced into the various sef mutants to test this hypothesis. ß-Galactosidase assays were used to measure sefR-dependent expression of the sef operon in these mutants by comparing expression levels before and after the addition of arabinose to strains containing the AraC-dependent sefR expression plasmid (Fig. 4). Strains which do not have any other mutations in the sef operon except the sefD::MudJ fusion showed approximately sixfold induction of gene expression following the addition of arabinose. Insertion of a polar kanamycin cassette in the first gene in the operon, sefA, abolished sefR-dependent expression of sefD, suggesting that a single transcript encodes the entire operon. In contrast, an in-frame deletion of sefA ({Delta}sefA) did not reduce sefR induction of sefD gene expression.

Growth-phase regulation of sef gene expression
To investigate the regulation of sef gene expression, ß-galactosidase activities were compared at different time points during growth. Expression of both sefR and sefD was optimal during late exponential growth and declined during stationary phase (Fig. 5). Growth-phase-dependent regulation of gene expression appears to occur at the sefR promoter as expression of both sefR and sefD were regulated by growth phase and we have shown that sefR regulates sefD expression. We had previously noted that sef gene expression was higher in a Salmonella Typhimurium LT2 background than in a wild-type Salmonella Enteritidis background. Salmonella Typhimurium has several genetic defects accumulated over years of maintenance in the laboratory. One such defect is a mutation that reduces rpoS expression; thus we reasoned that RpoS may be responsible for both the difference in expression between Salmonella Enteritidis and Salmonella Typhimurium and the growth-phase-dependent regulation of sef gene expression. To investigate whether the difference in expression levels observed between Salmonella Enteritidis and Salmonella Typhimurium could be explained by rpoS effects, an rpoS::Pen mutation was transduced from Salmonella Typhimurium to Salmonella Enteritidis. Because there was no detectable difference in growth rate, ß-galactosidase activity was measured after 16 h of growth, when sef expression is normally repressed (Fig. 5). Inactivation of rpoS relieved this repression and the sef genes were expressed at a high level at this time. In contrast, when rpoS was restored on a plasmid the repression returned (Fig. 6). Together these results suggest that expression of the sef operon is repressed, either directly or indirectly, by the stationary phase sigma factor {sigma}s in Salmonella Enteritidis.



View larger version (20K):
[in this window]
[in a new window]
 
Fig. 5. Expression of sef genes is regulated by growth phase. The expression of sefD and sefR were measured and were optimal during late exponential growth. Expression of both the sefD::lacZ fusion (solid squares) and the sefR::lacZ fusion (open squares) is maximal during late exponential growth and declines rapidly as the culture enters stationary phase. A650 for the wild-type strain is shown on the right axis and represented as open triangles and a dashed line. No difference in growth rate was detected between any of the strains used. No expression was ever observed in LK5, the wild-type strain (solid circles).

 


View larger version (16K):
[in this window]
[in a new window]
 
Fig. 6. Expression of the sef genes is regulated by rpoS. In the absence of rpoS (rpoS::Pen) expression of both sefD::lacZ and sefR::lacZ fusions is increased (note that sefR is not added exogenously). In contrast when rpoS is complemented from a plasmid, expression of sefR is reduced to wild-type levels in an rpoS background.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
An integrated approach was taken to dissect the sef operon of Salmonella Enteritidis. Genomic analysis delimited the island and identified the potential ORFs in this island. Genetic analysis was used to mutate the genes on the sef island and expression studies were used to characterize the regulation of fimbrial gene expression. We showed that the sef operon of Salmonella Enteritidis contains four structural genes (sefABCD) and an adjacent gene (sefR) that encodes an AraC-like regulator of sef gene expression. A single transcript probably encodes the products of sefA through to sefD. SefR, an AraC-like transcriptional activator, regulates transcription of this operon. Expression of the operon is inhibited by the stationary phase sigma factor, {sigma}s, encoded by rpoS in response to growth-phase-dependent signals.

Analysis of the available genome sequences from various Salmonella serovars revealed a novel large pathogenicity island in all four serovars. From hybridization and PCR studies it was known that the sef operon was located on an island which is absent from many other Salmonella but the proximity of phage genomes to this operon had not been appreciated. It remains to be determined whether the phage genomes adjacent to the sef operon are responsible for the transmission of this island between strains. The Salmonella Typhimurium genome sequence revealed a large insertion in which only a single ORF had been previously identified. Our analysis suggests that there may be other, as yet uncharacterized, virulence genes in this region.

A region of homology between the virulence plasmid and the chromosome was identified downstream of the sef operon. Salmonella Typhi does not contain the virulence plasmid but contains the chromosomal homologue of this region and Salmonella Typhimurium contains the region on the virulence plasmid but does not contain the chromosomal homologue. The homology between plasmid and chromosome is potentially only found in a few Salmonella serovars, including Salmonella Enteritidis, which contain both virulence plasmid and chromosomal homologues. Such regions of homology may allow the plasmid to integrate into the chromosome by homologous recombination. As the virulence plasmid is self-transmissible this would result in the formation of an Hfr capable of transferring virulence factors and other genetic elements between serovars (Ahmer et al., 1999 ). The formation of an Hfr in this way could have profound impact on the emergence of infectious disease and the spread of antibiotic resistance genes.

After this manuscript was first submitted, another group also described the chromosomal region surrounding the sef island (Collighan & Woodward, 2001 ). There appears to have been a rearrangement around the sef island when their description of the island is compared to Fig. 1. In the strain of Salmonella Enteritidis used in their study, leuX is located downstream of sefD and the sef island is flanked by two insertion elements. They also identify an AraC-like transcriptional activator that they designate SefE. However, when this manuscript was prepared this sequence was not available in the DNA or protein databases and therefore could not be compared to the sequence described here. A third group has also reported the sequence of an AraC-like regulator that is involved in the regulation of the sef operon. This has been deposited in GenBank with the accession number 7330248 (J. A. Botten, I. Kotlarski & R. Morona, unpublished). The two available SefR sequences only differ at two of their 271 residues.

A previous study indicated that sefABC were probably encoded in a single operon (Clouthier et al., 1993 ). This study also noted that accessory proteins in addition to SefA, SefB and SefC must be required for export of SEF14 fimbriae as no fimbriae were detected on the surface of a smaller clone of the island (Clouthier et al., 1993 ). The clone that was used in this study was a HindIII subclone from a cosmid. There is a HindIII site in sefD which would produce a clone of the size observed (5·3 kb), suggesting that clones lacked both sefD and sefR, and providing a likely explanation why SEF proteins were not expressed. Another report described a unique 18 kDa fimbria, ascribed to sefD. In this report, antibodies were prepared against the fimbrial protein and used to probe for the distribution of the fimbriae amongst enteric bacteria. Genomic DNA was also probed with a sefD-specific PCR product. Both these techniques suggested that sefD, unlike sefABC, had a widespread distribution among enteric bacteria (Clouthier et al., 1994 ). Several lines of evidence refute this distribution. (i) There is no homology between sefD or its predicted product and the DNA and protein sequences in the E. coli genomic databases even though it was reported that the 18 kDa fimbria can be found in E. coli. (ii) More exhaustive studies using either PCR or Southern hybridization were unable to identify a sefD homologue outside the group D Salmonella (data not shown). (iii) Our studies have shown that a single transcript initiated upstream of sefA and extending through sefD encodes these fimbrial proteins. Moreover, the close coupling of the UGA stop codon for sefC and the AUG initiation codon for sefD indicates that SefC and SefD proteins are co-expressed as expected for components of the same fimbriae. There are no known cases where subunits that form different fimbriae are co-expressed from the same operon, so this close coupling suggests that SefD is part of the SEF14 fimbriae.

Comparison of growth-phase-dependent regulation of gene expression, and the expression of the sef genes in wild-type Salmonella Enteritidis and Salmonella Typhimurium LT2, a natural rpoS mutant, indicated that the sef genes are repressed by {sigma}s, the stationary phase sigma factor. These results do not identify whether rpoS affects sef gene expression directly or indirectly and therefore the precise role of rpoS in regulation of sef gene expression, as well as the role of other genetic elements outside of the sef operon remains to be determined.

A previous report showed that the sef genes require a rare arginine tRNA (tRNA-UCU) for expression (Clouthier et al., 1998 ). It is not known why this tRNA is required. However, genomic analysis of the sef operon showed that the mean G+C content of this region is 35·4 mol%. In comparison, the entire Salmonella Enteritidis genome has a mean G+C content of 52 mol%. The arginine codon recognized by the rare tRNA is AGA. Of the arginine codons in the sef island, 58% are AGA and only 2% are CGC, the preferred Salmonella arginine codon. The limited distribution of the sef island suggested that it is a recent acquisition on the chromosome and this is supported by its unusual codon usage profile. The regulation of the sef operon by tRNA-UCU may simply reflect this different codon usage which presumably betrays some ancestral host with an A+T-rich genome, rather than a specific regulatory mechanism for sensing arginine concentrations.

Although many other fimbrial systems from Salmonella serovars have been identified and characterized, type I fimbriae and the aggregative fimbriae encoded by agfBAC were the only examples for which a transcriptional regulator has been identified (Romling et al., 2000 ; Tinker & Clegg, 2000 , 2001 ; Tinker et al., 2001 ). It remains to be seen which elements are required for these systems. Many other fimbrial systems use AraC-like transcriptional activators to regulate gene expression and often the regulator is immediately adjacent to the fimbrial operon, as is found for the sef operon. For example, the 987P fimbria in enterotoxigenic E. coli utilizes a homologous regulator (FasH) to activate gene transcription (Edwards & Schifferli, 1997 ). The similarity and location of these regulators and the similar organization of the fimbrial operons as a whole [in general they are organized major subunit-chaperone-usher-(optional minor subunits)-regulator] suggests that many fimbrial operons have evolved from some common ancestor.

All of the AraC-like family of transcriptional activators involved in regulation of fimbrial gene expression have similar C-terminal domains. This domain with conserved motifs is the DNA-binding domain. In contrast the N-terminal domains are extremely different. This domain with no conserved motifs is presumably the sensor domain that regulates DNA binding in response to environmental cues. This suggests – as practical experience has shown (Edwards & Puente, 1998 ; Gallegos et al., 1997 ; Jordi, 1992 ; Martinez-Laguna et al., 1999 ) – that each regulator may respond to different environmental stimuli, and hence each fimbria is expressed in different conditions and perhaps in a different location in the host.


   ACKNOWLEDGEMENTS
 
We thank the sequencing centre at the Sanger Centre and the Washington University St Louis Genome Sequencing Center for communication of DNA sequence data prior to publication. We thank Steve Libby for the rpoS::Pen strain, Ferric Fang for the pBAD-rpoS clone and Gary Olsen and Jim Slauch for helpful discussions. We are indebted to Dieter Schifferli for strains, discussions and advice with this work. This work was supported by grants from the National Institutes of Health (NIH award number PHS GM34715), the University of Illinois Critical Research Initiative and the Illinois Council for Food and Agriculture Research (C-FAR).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Ahmer, B. M., Tran, M. & Heffron, F. (1999). The virulence plasmid of Salmonella typhimurium is self-transmissible. J Bacteriol 181, 1364-1368.[Abstract/Free Full Text]

Alexeyev, M. F., Shokolenko, I. N. & Croughan, T. P. (1995). New mini-Tn5 derivatives for insertion mutagenesis and genetic engineering in gram-negative bacteria. Can J Microbiol 41, 1053-1055.[Medline]

Altschul, S. F. & Gish, W. (1996). Local alignment statistics. Methods Enzymol 266, 460-480.[Medline]

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol 215, 403-410.[Medline]

Altschul, S. F., Boguski, M. S., Gish, W. & Wootton, J. C. (1994). Issues in searching molecular sequence databases. Nat Genet 6, 119-129.[Medline]

Bäumler, A. J., Tsolis, R. M., Bowe, F. A., Kusters, J. G., Hoffmann, S. & Heffron, F. (1996a). The pef fimbrial operon of Salmonella typhimurium mediates adhesion to murine small intestine and is necessary for fluid accumulation in the infant mouse. Infect Immun 64, 61-68.[Abstract]

Bäumler, A. J., Tsolis, R. M. & Heffron, F. (1996b). The lpf fimbrial operon mediates adhesion of Salmonella typhimurium to murine Peyer’s patches. Proc Natl Acad Sci USA 93, 279-283.[Abstract/Free Full Text]

Bäumler, A. J., Gilde, A. J., Tsolis, R. M., van der Velden, A. W., Ahmer, B. M. & Heffron, F. (1997). Contribution of horizontal gene transfer and deletion events to development of distinctive patterns of fimbrial operons during evolution of Salmonella serotypes. J Bacteriol 179, 317-322.[Abstract]

Caron, J., Coffield, L. M. & Scott, J. R. (1989). A plasmid-encoded regulatory gene, rns, required for expression of the CS1 and CS2 adhesins of enterotoxigenic Escherichia coli. Proc Natl Acad Sci USA 86, 963-967.[Abstract]

Choudhury, D., Thompson, A., Stojanoff, V., Langermann, S., Pinkner, J., Hultgren, S. J. & Knight, S. D. (1999). X-ray structure of the FimC–FimH chaperone–adhesin complex from uropathogenic Escherichia coli. Science 285, 1061-1066.[Abstract/Free Full Text]

Clouthier, S. C., Muller, K. H., Doran, J. L., Collinson, S. K. & Kay, W. W. (1993). Characterization of three fimbrial genes, sefABC, of Salmonella enteritidis. J Bacteriol 175, 2523-2533.[Abstract]

Clouthier, S. C., Collinson, S. K. & Kay, W. W. (1994). Unique fimbriae-like structures encoded by sefD of the SEF14 fimbrial gene cluster of Salmonella enteritidis. Mol Microbiol 12, 893-903.[Medline]

Clouthier, S. C., Collinson, S. K., White, A. P., Banser, P. A. & Kay, W. W. (1998). tRNA(Arg) (fimU) and expression of SEF14 and SEF21 in Salmonella enteritidis. J Bacteriol 180, 840-845.[Abstract/Free Full Text]

Collighan, R. J. & Woodward, M. J. (2001). The SEF14 fimbrial antigen of Salmonella enterica serovar Enteritidis is encoded within a pathogenicity islet. Vet Microbiol 80, 235-245.[Medline]

Collinson, S. K., Liu, S. L., Clouthier, S. C., Banser, P. A., Doran, J. L., Sanderson, K. E. & Kay, W. W. (1996). The location of four fimbrin-encoding genes, agfA, fimA, sefA and sefD, on the Salmonella enteritidis and/or S. typhimurium XbaI–BlnI genomic restriction maps. Gene 169, 75-80.[Medline]

Edwards, R. A. & Puente, J. L. (1998). Fimbrial expression in enteric bacteria: a critical step in intestinal pathogenesis. Trends Microbiol 6, 282-287.[Medline]

Edwards, R. A. & Schifferli, D. M. (1997). Differential regulation of fasA and fasH expression of Escherichia coli 987P fimbriae by environmental cues. Mol Microbiol 25, 797-809.[Medline]

Edwards, R. A., Cao, J. & Schifferli, D. M. (1996). Identification of major and minor chaperone proteins involved in the export of 987P fimbriae. J Bacteriol 178, 3426-3433.[Abstract]

Edwards, R. A., Keller, L. H. & Schifferli, D. M. (1998). Improved allelic exchange vectors and their use to analyze 987P fimbria gene expression. Gene 207, 149-157.[Medline]

Edwards, R., Helm, R. & Maloy, S. (1999). Increasing DNA transfer efficiency by temporary inactivation of host restriction. BioTechniques 26, 892-900.[Medline]

Edwards, R. A., Schifferli, D. M. & Maloy, S. R. (2000). A role for Salmonella fimbriae in intraperitoneal infections. Proc Natl Acad Sci USA 97, 1258-1262.[Abstract/Free Full Text]

Fang, F. C., Libby, S. J., Buchmeier, N. A., Loewen, P. C., Switala, J., Harwood, J. & Guiney, D. G. (1992). The alternative sigma factor katF (rpoS) regulates Salmonella virulence. Proc Natl Acad Sci USA 89, 11978-11982.[Abstract]

Gaastra, W. & Svennerholm, A.-M. (1996). Colonization factors of human enterotoxigenic Escherichia coli (ETEC). Trends Microbiol 4, 444-452.[Medline]

Gallegos, M. T., Schleif, R., Bairoch, A., Hofmann, K. & Ramos, J. L. (1997). Arac/XylS family of transcriptional regulators. Microbiol Mol Biol Rev 61, 393-410.[Abstract]

Guzman, L. M., Belin, D., Carson, M. J. & Beckwith, J. (1995). Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J Bacteriol 177, 4121-4130.[Abstract]

Jordi, B. J. (1992). The mode of action of CfaD of Escherichia coli and VirF of Shigella flexneri and other members of the AraC family of positive regulators. Mol Microbiol 6, 3451.

Jost, B. H. & Adler, B. (1993). Site of transcriptional activation of virB on the large plasmid of Shigella flexneri 2a by VirF, a member of the AraC family of transcriptional activators. Microb Pathog 14, 481-488.[Medline]

Keller, L. H., Schifferli, D. M., Benson, C. E., Aslam, S. & Eckroade, R. J. (1997). Invasion of chicken reproductive tissues and forming eggs is not unique to Salmonella enteritidis. Avian Dis 41, 535-539.[Medline]

Khan, A. S. & Schifferli, D. M. (1994). A minor 987P protein different from the structural fimbrial subunit is the adhesin. Infect Immun 62, 4223-4243.

Maloy, S. (1990). Experimental Techniques in Bacterial Genetics, 1st edn. Boston, MA: Jones and Bartlett.

Maloy, S. R., Stewart, V. J. & Taylor, R. K. (1996). Genetic Analysis of Pathogenic Bacteria. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Martinez-Laguna, Y., Calva, E. & Puente, J. L. (1999). Autoactivation and environmental regulation of bfpT expression, the gene coding for the transcriptional activator of bfpA in enteropathogenic Escherichia coli. Mol Microbiol 33, 153-166.[Medline]

Nataro, J. P., Yikang, D., Yingkang, D. & Walker, K. (1994). AggR, a transcriptional activator of aggregative adherence fimbria I expression in enteroaggregative Escherichia coli. J Bacteriol 176, 4691-4699.[Abstract]

Pfeifer, C. G., Marcus, S. L., Steele-Mortimer, O., Knodler, L. A. & Finlay, B. B. (1999). Salmonella typhimurium virulence genes are induced upon bacterial invasion into phagocytic and nonphagocytic cells. Infect Immun 67, 5690-5698.[Abstract/Free Full Text]

Rodriguez-Pena, J. M., Alvarez, I., Ibanez, M. & Rotger, R. (1997). Homologous regions of the Salmonella enteritidis virulence plasmid and the chromosome of Salmonella typhi encode thiol:disulphide oxidoreductases belonging to the DsbA thioredoxin family. Microbiology 143, 1405-1413.[Abstract]

Romling, U., Rohde, M., Olsen, A., Normark, S. & Reinkoster, J. (2000). AgfD, the checkpoint of multicellular and aggregative behaviour in Salmonella typhimurium regulates at least two independent pathways. Mol Microbiol 36, 10-23.[Medline]

Sanderson, K., Hessel, A., Liu, S. & Rudd, K. E. (1996). The genetic map of Salmonella typhimurium, edition VIII. In Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology , pp. 1903-1999. Edited by F. C. Neidhardt. Washington, DC:American Society for Microbiology.

Sauer, F. G., Fütterer, K., Pinkner, J. S., Dodson, K. W., Hultgren, S. J. & Waksman, G. (1999). Structural basis of chaperone function and pilus biogenesis. Science 285, 1058-1061.[Abstract/Free Full Text]

Slauch, J. M. & Silhavy, T. J. (1991). cis-acting ompF mutations that result in OmpR-dependent constitutive expression. J Bacteriol 173, 4039-4048.[Medline]

Swords, W. E., Cannon, B. M. & Benjamin, W. H.Jr (1997). Avirulence of LT2 strains of Salmonella typhimurium results from a defective rpoS gene. Infect Immun 65, 2451-2453.[Abstract]

Thorns, C. J., Sojka, M. G. & Chasey, D. (1990). Detection of a novel fimbrial structure on the surface of Salmonella enteritidis by using a monoclonal antibody. J Clin Microbiol 28, 2409-2414.[Medline]

Tinker, J. K. & Clegg, S. (2000). Characterization of FimY as a coactivator of type 1 fimbrial expression in Salmonella enterica serovar Typhimurium. Infect Immun 68, 3305-3313.[Abstract/Free Full Text]

Tinker, J. K. & Clegg, S. (2001). Control of FimY translation and type 1 fimbrial production by the arginine tRNA encoded by fimU in Salmonella enterica serovar Typhimurium. Mol Microbiol 40, 757-768.[Medline]

Tinker, J. K., Hancox, L. S. & Clegg, S. (2001). FimW is a negative regulator affecting type 1 fimbrial expression in Salmonella enterica serovar typhimurium. J Bacteriol 183, 435-442.[Abstract/Free Full Text]

Tobe, T., Schoolnick, G. K., Sohel, I., Bustamante, V. H. & Puente, J. L. (1996). Cloning and characterization of bfpTVW, genes required for the transcriptional activation of bfpA in enteropathogenic Escherichia coli. Mol Microbiol 21, 963-975.[Medline]

Townsend, S. M., Kramer, N. E., Edwards, R. A. & 7 other authors (2001). Salmonella enterica serotype Typhi possesses a unique repertoire of fimbrial genes. Infect Immun 69, 2894–2901.[Abstract/Free Full Text]

Turcotte, C. & Woodward, M. J. (1993). Cloning, DNA nucleotide sequence and distribution of the gene encoding the SEF14 fimbrial antigen of Salmonella enteritidis. J Gen Microbiol 139, 1477-1485.[Medline]

Received 9 March 2001; revised 5 June 2001; accepted 12 June 2001.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Edwards, R. A.
Articles by Maloy, S. R.
Articles citing this Article
PubMed
PubMed Citation
Articles by Edwards, R. A.
Articles by Maloy, S. R.
Agricola
Articles by Edwards, R. A.
Articles by Maloy, S. R.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2001 Society for General Microbiology.