Eavesdropping by Bacteria: The Role of SdiA in Escherichia coli and Salmonella enterica Serovar Typhimurium Quorum Sensing

Abstract

Many gram-negative bacteria utilize N-acyl-L-homoserine lactones (AHLs) to bind to transcriptional regulators leading to activation or repression of target genes. Escherichia coli and Salmonella enterica do not synthesize AHLs but do contain the AHL receptor, SdiA. Studies reveal that SdiA can bind AHLs produced by other bacterial species and thereby allow E. coli and S. enterica to regulate gene transcription. The Salmonella sdiA gene regulates the rck gene, which mediates Salmonella adhesion and invasion of epithelial cells and the resistance of the organism to complement. In E. coli, there is some evidence that SdiA may regulate genes associated with acid resistance, virulence, motility, biofilm formation, and autoinducer-2 transport and processing. However, there is a lack of information concerning the role of SdiA in regulating growth and survival of E. coli and Salmonella in food environments, and therefore studies in this area are needed.

Introduction

M any Gram-negative bacteria utilize N-acyl-L-homoserine lactones (AHLs) as signal molecules to sense their local population densities and in turn regulate gene transcription. This process is known as quorum sensing (QS) (Williams, 2007; Williams et al., 2007). The AHLs are produced within the cell by bacterial AHL synthases (LuxI-like enzymes) and are transported to the external milieu. As the bacterial numbers increase, the external concentration of AHLs also increases. When the AHLs reach a threshold level, the molecules re-enter the cells and bind to their receptor proteins (LuxR-like proteins). The AHL-LuxR complex forms dimers or multimers and binds DNA at specific gene promoter sequences known as lux-boxes followed by activation or repression of target gene transcription (Williams, 2007; Williams et al., 2007). Genes that may be affected include those involved with production of virulence factors, motility, antibiotic production, bioluminescence, biofilm formation, as well as other functions (Smith et al., 2004).

Both Gram-negative and Gram-positive bacteria have a common QS system mediated by autoinducer 2 (AI-2), which is produced by the enzyme LuxS (encoded by luxS) (Smith et al., 2004; Vendeville et al., 2005; Gobbetti et al., 2007). In addition, LuxS is also a key enzyme of the activated methyl cycle resulting in the production of the major methyl donor, S-adenosylmethionine, with 4,5-dihydroxy-2,3-pentanedione, the precursor of AI-2, as a by-product (Vendeville et al., 2005; McDougald et al., 2007). LuxS, therefore, is involved in both QS and cellular metabolism. An additional QS system is the autoinducer-3/epinephrine/norepinephrine (AI-3/epi/norepi) signaling system found in a number of Gram-negative bacterial species (Walters and Sperandio, 2006). The AI-3/epi/norepi system has been shown to have an important role in regulating the virulence of Escherichia coli O157:H7 (Sperandio et al., 2003; Walters and Sperandio, 2006). Indole may behave as a QS molecule in the stationary phase of bacterial growth (Hirakawa et al., 2005; Lee et al., 2007). Other types of signaling molecules have also been described (Smith et al., 2004).

Species in the genera Escherichia, Klebsiella, Salmonella (Smith and Ahmer, 2003), and Shigella (Lerat and Moran, 2004; Yao et al., 2006) produce a LuxR homolog known as SdiA (suppressor of cell division inhibitor), but these organisms lack AHL synthases and therefore do not produce AHLs. However, the SdiA molecules can bind AHLs produced by other microorganisms and affect transcription in these non-AHL-producing bacteria (Yao et al., 2006). Fuqua (2006) called the luxR homolog genes not linked to or associated with AHL synthase genes, “orphan” genes. However, Subramoni and Venturi (2009) prefer to call these luxR genes “solo” genes since their LuxR proteins can act on their own without a cognate LuxI AHL molecule. “Solo” is preferred to “orphan” since, in bacterial genetics, the term “orphan” is used to describe a putative gene with unknown function that does not appear to have any homologs.

Case et al. (2008) described the phenomenon of non-AHL-producing microorganisms binding and utilizing AHLs produced by other organisms as eavesdropping. Thus, these microorganisms can modulate their behavior by responding to signals that they do not synthesize. Using a database comprised of 512 completed bacterial genomes, Case et al. (2008) did a phylogenetic survey and identified a number of bacterial genomes that had LuxR homologs but did not have corresponding AHL synthases. Many of these bacteria, however, had one or more LuxI/LuxR homologs. Thus, some bacteria contain more luxR genes than luxI genes. For example, Pseudomonas aeruginosa has the QscR solo homolog in addition to LasI/LasR and RhlI/RhlR homologs (Lequette et al., 2006; Subramoni and Venturi, 2009). Malott et al. (2009) have described the CepR2 solo homolog in Burkholderia cenocepacia, which also has CepI/CepR and CciI/CciR homologs. These solo LuxR homologs, QscR and CepR2, can use endogenously synthesized AHLs as well as exogenous AHLs. A recent review by Patankar and González (2009) gives a detailed discussion of solo (orphan) LuxR regulators found in various bacterial species.

Other bacteria that have luxR but do not have a gene for AHL synthesis include Xanthomonas, Pseudomonas, and Brucella species (Case et al., 2008). While there appears to be a number of bacteria lacking AHL synthases but having LuxR-like AHL receptors, only the SdiA proteins produced by E. coli and Salmonella Typhimurium have been studied in detail.

The Role of SdiA in E. coli

In E. coli, the sdiA open reading frame encodes a 241 amino acid protein (SdiA) with a predicted molecular mass of 28 kDa (Wang et al., 1991). Utilizing a neighbor-joining tree (a method for reconstruction of phylogenetic trees from evolutionary distance data; Saitou and Nei, 1987) of LuxR family members, Gray and Garey (2001) demonstrated that the SdiAs of E. coli and Salmonella Typhimurium did not group with the LuxR homologs found in other enterobacterial species. These SdiAs are more closely related to the RhlR AHL receptor of P. aeruginosa, thus suggesting horizontal transfer of the pseudomonad homolog to E. coli and Salmonella Typhimurium (Gray and Garey, 2001; Lerat and Moran, 2004) and possibly to other species that have the SdiA homolog.

Soulére et al. (2007) found that the AHL-binding site of the AHL-dependent transcription regulator, TraR, of Agrobacterium tumefaciens has a high structural similarity to that of E. coli SdiA. Key amino acid residues that associate with AHLs are strictly conserved in both TraR and SdiA, as well as in LuxR (Vibrio fischeri) and LasR (P. aeruginosa). The data of Soulére et al. (2007) indicate a high degree of homology for the active sites of LuxR-type proteins.

Yao et al. (2006) have determined the three-dimensional structure of E. coli SdiA (residues 1–171) complexed with C₈-homoserine lactone (C₈-HSL) (structures of a number of AHLs are given in Atkinson and Williams, 2009; Ng and Bassler, 2009). When bound to AHLs, the LuxR homologs, demonstrate “folding switch” behavior leading to a stable folded, soluble tertiary structure resistant to proteolysis (Zhu and Winans, 2001). SdiA demonstrates a similar behavior in the presence of AHLs. SdiA produced in E. coli is found in insoluble inclusion bodies. However, when bound to C₈-HSL, the protein has a folded three-dimensional soluble structure (Yao et al., 2006). In addition to C₈-HSL, Yao et al. (2006) found that C₆-, 3-oxo-C₆-, and 3-oxo-C₈-HSLs activated the folding switch of SdiA.

Van Houdt et al. (2006) screened a promoter trap library from E. coli MG1655 with C₆-HSL and found six genes that were upregulated and nine that were downregulated in response to the AHL, and the same responses were also obtained with 3-oxo-C₆-HSL. In an sdiA knockout mutant, there was no effect on the genes when AHLs were added. The response to AHLs by E. coli was seen at 30°C but weakly or not at all at 37°C (Van Houdt et al., 2006).

SdiA and cell division

When expressed from a multicopy plasmid, the sdiA gene product suppresses expression of a number of chromosomally encoded cell division inhibitors in E. coli, leading to an increase in cell division (Wang et al., 1991). SdiA acts as a positive regulator of transcription of the ftsQAZ gene cluster and increases the cellular levels of the FtsZ protein. The ftsQAZ gene cluster in E. coli is a cell division activator and it has been shown that overexpression of the ftsQAZ gene cluster increases bacterial cell septation and blocks the action of endogenous cell division inhibitors (Wang et al., 1991). However, bacterial cells with deleted sdiA did not show a defect in cell division (Wang et al., 1991). There is no evidence that the chromosomal sdiA gene in the absence or presence of AHLs has an effect on cell division in E. coli. The effect seen with cell division in the presence of the overexpressed sdiA gene appears to be an artifact of overexpression. Sitnikov et al. (1996) found that a plasmid-encoded ftsQA-lacZY fusion was activated by multicopy plasmid-encoded sdiA. When the AHLs, 3-oxo-C₆-HSL, 3-OH-C₄-HSL, or C₁₂-HSL, were added, there was a further increase in ftsQA transcription by overexpressed sdiA (Sitnikov et al., 1996). Lee et al. (2008) found that the addition of 1 mM indole to E. coli K-12 led to a decrease in ftsQ expression, whereas overexpression of sdiA induced expression of ftsQ. Indole had no effect on ftsQ expression in an sdiA mutant.

SdiA and antibiotic resistance

The presence of sdiA on a multicopy vector induced resistance to mitomycin C and nalidixic acid in E. coli K-12 (Wei et al., 2001b). An sdiA null mutant did not show hypersensitivity to these antibiotics. Wei et al. (2001a) demonstrated that overexpression of sdiA increased the resistance of E. coli K-12 to both tetracycline and nalidixic acid by ∼8-fold as compared to the control E. coli containing only a chromosomal sdiA gene. Overexpression did not lead to an increase in resistance to rifampin, chloramphenicol, spectinomycin, or kanamycin (Wei et al., 2001a). When the sdiA gene was cloned into a multicopy vector and transformed into wild-type E. coli (strain W4573), there was a several fold increase in quinolone and chloramphenicol resistance, as well as a lesser increase in resistance to kanamycin and tetracycline (Rahmati et al., 2002). An sdiA null mutant of strain W4573 was more sensitive to fluoroquinolones than the wild type but was not more sensitive to chloramphenicol, nalidixic acid, or tetracycline. A constitutive multidrug efflux pump, AcrAB, confers antibiotic resistance in E. coli (Nikaido, 1998). Rahmati et al. (2002) demonstrated that sdiA overexpression led to an ∼4-fold increase in AcrAB levels, which correlates with the increase in antibiotic resistance shown by the E. coli strain. The hypersensitivity of an sdiA null mutant of E. coli strain W4573 toward fluoroquinolones was linked to a 50% decrease in the level of AcrB (Rahmati et al., 2002).

SdiA and virulence

In E. coli O157:H7, overexpression of sdiA (on a multicopy plasmid) led to impaired adherence of the organism to Caco-2 cells (epithelial cells from human colon), inhibited expression of messenger RNA of the virulence genes espD (involved in actin-based pedestal formation) and eae (encodes the intimin protein required for adhesion), as well as that of fliC (involved in motility) (Kanamaru et al., 2000). Due to the repression of fliC by overexpression of sdiA, the E. coli O157:H7 strain was not motile. A diarrheic enteropathogenic E. coli strain that had many of the virulence genes of enterohemorrhagic E. coli O157:H7 did not show inhibition of the production of EspA, EspB, EspD, intimin, or adherence to Caco-2 cells when sdiA was overexpressed (Kanamaru et al., 2000).

A medium in which human fecal microbiota had been grown, followed by removal of the microbiota by filtration (the medium was termed “conditioned medium”) was utilized by de Sablet et al. (2009) to determine the effect of compounds synthesized by fecal microbiota on the production of Shiga toxin 2 (Stx2) by E. coli O157:H7 (strain EDL933). Growth of E. coli O157:H7 in the conditioned medium led to an ∼4-fold decrease in Stx2 production at 24 h. In addition, de Sablet et al. (2009) found that the conditioned medium also repressed the production of Stx1. Using mutants and QS-reporter systems, the repression of stx2 by the human fecal microbiota was shown to be independent of known quorum-sensing pathways such as SdiA, QseA, QseC, or AI-3 (de Sablet et al., 2009). Thus, SdiA does not appear to have a role in the synthesis of Stx2 by E. coli O157:H7. Stx2 production was approximately the same in the wild-type O157:H7 (strain EDL933) and the isogenic EDL933 ΔsdiA strain (de Sablet et al., 2009). It would be interesting to know if the addition of exogenous AHLs to the wild-type E. coli would have an effect on Stx2 production.

The locus of enterocyte effacement (LEE) pathogenicity island is responsible for the A/E (attaching and effacing lesions) phenotype in enteropathogenic and enterohemorrhagic E. coli (Law, 2000). At the right end of the LEE locus are the genes espA, espB, and espD, which encode proteins necessary for signaling and A/E activity. At the left end are the genes encoding a type III secretion system necessary for secretion of proteins, including the products of espA, espB, and espC. Between these two regions is the eae gene encoding intimin (binds to enterocytes) and the tir gene, which encodes the receptor for intimin (Law, 2000). Utilizing E. coli O157:H7, Hughes et al. (2008) found that when 3-oxo-C₆-HSL was added, SdiA repressed the transcription of LEE genes. In addition, Hughes et al. (2008) demonstrated, by electrophoretic motility shift assays, that SdiA binds to the promoter of ler (LEE-encoded regulator). Ler (encoded by ler) is the master regulator of the LEE locus (Haack et al., 2003; Mellies et al., 2007). Thus, SdiA in the presence of AHL appears to repress the LEE virulence locus in E. coli O157:H7 by regulating the transcription of ler (Hughes et al., 2008). However, Sharma et al. (2010) demonstrated that the deletion of sdiA in E. coli O157:H7 strain 86-24 Δstx2 Δlac had an insignificant effect on expression ler, espA, and eae, indicating that SdiA has little effect on expression of LEE genes. Sharma and coworkers (2010) did not examine the wild-type strain with added AHLs. The results reported by Hughes et al. (2008) are in conflict with those obtained by Sharma et al. (2010) probably due to bacterial strain differences.

Dziva et al. (2004) found that the sdiA gene was necessary for the intestinal colonization of 10–14-day-old calves by E. coli O157:H7 EDL933 by comparing wild-type and sdiA mutant strains. A mutant lacking the sdiA gene did not colonize calves but did adhere to HeLa cells (indicates intimin activity), secreted EspD (a LEE-encoded type III secretion protein), and demonstrated fluorescent staining for F-actin at the site of bacterial attachment (indicates A/E activity). The sdiA-negative mutant of E. coli O157:H7 has virulence properties that should lead to increased colonization of calves, but Dziva et al. (2004) have shown that an intact sdiA gene is necessary for colonization.

Ruminants, in particular cattle, constitute a large reservoir of E. coli O157:H7 and other Shiga-toxin-producing E. coli (Gyles, 2007). The study by Dziva et al. (2004) indicating that the sdiA gene plays a role in colonization of the intestine of calves explains in part why cattle act as reservoirs of Shiga-toxin-producing E. coli.

SdiA and biofilms

In an E. coli strain (MG1655) that produced biofilms, Suzuki et al. (2002) found that overexpression of sdiA (on a multicopy plasmid) led to a 1.4-fold increase in biofilm formation on polystyrene microtiter plates as compared to the wild-type control strain. An isogenic sdiA null mutant showed a threefold (significant, p < 0.01) decrease in biofilm formation as compared to the wild-type strain. However, Lee et al. (2007), in contrast to the results of Suzuki et al. (2002), found that an sdiA mutant of E. coli K-12 BW25113 demonstrated an ∼51-fold increase in biofilm formation as compared to the wild type. One difference between the two studies is that Suzuki et al. (2002) performed their experiments at 37°C, whereas Lee et al. (2007) used 30°C.

Lee et al. (2009) studied the effect of reconfiguring the amino acid sequence of SdiA on biofilm formation by E. coli K-12. An isogenic sdiA-negative mutant of wild-type E. coli BW25113 increased biofilm formation 18-fold as compared to the wild type. There was a 3.5-fold decrease in biofilm formation with overexpressed sdiA as compared to the sdiA mutant in Luria-Bertani medium at 8 h; however, sdiA1E11 (encodes a truncated SdiA protein lacking the carboxy DNA-binding domain) decreased biofilm formation comparable to the wild type (Lee et al., 2009). Thus, wild-type sdiA, overexpressed sdiA, and sdiA1E11 strains show reduced biofilm formation as compared to the sdiA mutant. Testing a number of truncated mutants indicated that elimination of the carboxy DNA-binding domain of SdiA led to a decrease in biofilm formation (Lee et al., 2009). Truncated sdiA mutants such as sdiA1E11 produce increased levels of indole as compared to the wild type, and Lee et al. (2007) have demonatrated that indole decreased biofilm formation in E. coli.

The wild-type E. coli BW25113 and sdiA-negative mutant produced similar levels of extracellular indole. Overexpressed sdiA decreased indole production by 6.3-fold and sdiA1E11 increased the extracellular indole level by 1.9-fold as compared to the wild type. In addition, the intracellular level of indole was increased in the sdiA1E11 mutant (Lee et al., 2009). Lee et al. (2009) found that motility in the wild-type strain and the overexpressed sdiA strain were similar; however, motility was increased 1.8-fold in the sdiA-negative strain and was decreased 1.8-fold in the sdiA1E11 strain as compared to the wild-type strain (Lee et al., 2009). Indole at 500 μM inhibited motility by 50%; indole production by sdiA1E11 was >500 μM at 24 h, which explains why that particular mutation inhibits motility (Lee et al., 2009). The increased level of indole and decreased motility explains why the sdiA1E11 mutant shows a decrease in biofilm formation (Lee et al., 2009)

Lee et al. (2009) also created a mutant known as sdiA2D10, which did not encode a truncated molecule but had mutations at 4 positions. Unlike the truncated sdiA1E11 mutation, the nontruncated sdiA2D10 mutant was sensitive to AHLs. In the presence of C₄-, C₆-, C₈-, 3-oxo-C₈-, C₁₀-, C₁₂-, 3-oxo-C₁₂-, and 3-oxo-C ₁₄-HSL, sdiA2D10 increased biofilm formation four to sevenfold, whereas addition of these AHLs to the wild type decreased biofilm formation by 1.3-fold as compared to no addition of AHLs. The AHLs had no effect on biofilm formation by the truncated mutant, sdiA1E11.

The interesting study by Lee et al. (2009) indicated that the truncated SdiA mutants (lacking the carboxy DNA-binding domain) were strongly reduced in their ability to form biofilms primarily because of their ability to produce increased levels of indole. In contrast, the nontruncated mutant, sdiA2D10, increased biofilm formation ∼7-fold in the presence of C₈- or 3-oxo-C₁₂-HSLs as compared to no addition of AHLs (Lee et al., 2009). The binding of AHLs by wild-type E. coli decreased biofilm formation (Lee et al., 2007). Depending on the type of mutation in the sdiA gene, biofilm production can be either increased or decreased (Lee et al. 2009).

In E. coli, there is a close relationship between indole and sdiA. Lee et al. (2007) demonstrated that the addition of 600 μM indole increased the induction of sdiA transcription by ∼3-fold in a E. coli K-12 yceP mutant. The addition of indole (500 μM) led to an ∼2-fold decrease in motility and inhibited biofilm formation by 50%. A mutation in sdiA increased motility by 2-fold and biofilm formation by 51-fold (Lee et al., 2007). In addition, Lee et al. (2007) found that the addition of C₄-, C₆-, or C₈-HSLs to wild-type E. coli K-12 decreased biofilm formation on polystyrene plates but 3-oxo-C₈-, C₁₀-, or C₁₂-HSLs had no effect on biofilm formation. Thus, the binding of AHLs by wild-type E. coli led to a decrease in biofilm formation. It is clear that SdiA inhibits both motility and biofilm formation. Indole induces the production of SdiA leading to the repression of motility and biofilm production. Lee et al. (2007) hypothesized that indole controls biofilm formation in E. coli through SdiA. Lee et al. (2008) found that the addition of indole had no effect on biofilm formation by the sdiA mutant of E. coli K-12 BW25113, indicating that in the absence of the sdiA gene, indole does not inhibit biofilm formation.

Utilizing E. coli O157:H7 strain 86-24 Δstx2 Δlac, Sharma et al. (2010) found that a deletion in the sdiA gene enhanced the adherence of bacterial cells to HEp-2 (human laryngeal epithelial) cells, indicating that SdiA inhibits adherence to epithelial cells. Expression of fliC (encodes flagellin) and csgA (encodes curlin protein of curli) in the E. coli O157:H7 mutant train was increased 2.7-fold and 3.4-fold, respectively; the increase in expression of the fliC and csgA genes indicates that SdiA has a negative effect on expression of these genes, leading to decreased motility and adherence (Sharma et al. 2010).

Deletion of sdiA in E. coli O157:H7 strain 86-24 Δstx2 Δlac led to an increase in motility as compared to the parent strain, thus indicating that SdiA has a negative effect on the motility of E. coli O157:H7 (Sharma et al., 2010). Biofilm formation on polystyrene plates increased ∼1.6-fold when sdiA was deleted from the O157:H7 strain, indicating that SdiA inhibited biofilm formation (Sharma et al., 2010). Wood et al. (2006) studied eight E. coli strains (none were O157:H7 strains) and found that highly motile strains were the best biofilm producers. The decreased motility of the parent O157:H7 strain as compared to the sdiA mutant demonstrated by Sharma and coworkers (2010) may explain the decreased biofilm production by the parent strain.

Role of sdiA examined using DNA microarrays

Wei et al. (2001a) did a microarray analysis on E. coli K-12 to determine the effect of overexpression of sdiA (on a multicopy plasmid) on gene transcription. Among genes that were upregulated by sdiA overexpression were the ftsAQZ genes and the acr operon, which encodes drug efflux systems (Wei et al., 2001a). Genes in which transcription was downregulated by overexpression of sdiA included genes associated with motility. Motility tests indicated that overexpression of sdiA did lead to a decrease in motility (Wei et al., 2001a). Lee et al. (2008) compared gene induction and repression in wild-type E. coli K-12 (BW25113) to that of an isogenic ΔsdiA strain and found that the presence of the chromosomal sdiA gene led to the repression (3- to 10-fold) of genes encoding proteins involved in uridine monophosphate synthesis and uracil transport and also repressed (three to sevenfold) the genes whose products are involved in cell motility (including flagellar biosynthesis), chemotaxis, and fimbrae ( fimA and fimC). There was upregulation (3- to 6-fold) of genes involved with purine biosynthesis, as well as upregulation (3- to 17-fold) of expression of genes involved in curli formation (Lee et al., 2008). Lee et al. (2008) did not examine the effect of AHL addition on gene expression. In contrast to Lee et al. (2008), Sharma et al. (2010), using E. coli O157:H7 strain 86-24 Δstx2 Δlac, found that the presence of the sdiA gene had a negative effect on curli formation.

Comparing an E. coli luxS mutant (derived from strain K-12 W3110) plus AI-2 (present in culture supernatant from wild-type W3110) with the luxS mutant minus AI-2 (culture supernatant from the mutant), DeLisa et al. (2001) demonstrated that the presence of AI-2 resulted in a twofold increase in sdiA transcription using microarrays. Thus, the microarray data indicated that AI-2 had only a weak effect of sdiA transcription. However, utilizing a luxS-positive (AI-2 producing) E. coli O157:H7 strain, Sperandio et al. (2001) found that there was ∼11-fold increased induction in sdiA transcription when compared to an isogenic luxS-negative strain. It is not clear whether the upregulation of the sdiA gene is due to AI-2 acting as a quorum sensor or to LuxS acting in its metabolic role in the methyl cycle. If the addition of purified AI-2 or its precursor, 4,5-dihydroxy-2,3-pentanedione, to the sdiA mutant leads to the induction of sdiA transcription, then AI-2 is acting as a quorum sensor.

A microarray analysis of E. coli K-12 BW25113 by Lee et al. (2009) indicated that overexpression of sdiA (on a multicopy plasmid) repressed expression of indole-related genes (tnaC, tnaB, and tnaA) 7.5- to 22.6-fold, repressed curli genes (csgDEFG, 17.1- to 36.8-fold; csgBAC, 10.6- to 29.9-fold), and repressed genes involved in acid resistance (4.0- to 32-fold) as compared to an sdiA-deleted mutant. Overexpression of the truncated sdiA mutant, sdiA1E11, encoding an SdiA molecule lacking the carboxy DNA-binding domain induced the indole-related genes 6.5- to 16-fold, induced curli genes 2.6- to 4.6-fold, and induced acid resistance genes 2.5- to 3.7-fold as compared to overexpressed sdiA (Lee et al., 2009). The results comparing overexpression of sdiA1E11 to overexpressed sdiA indicates that sdiA1E11 regulates gene expression differently from wild-type sdiA (Lee et al., 2009).

SdiA and regulation of AI-2 transport and processing

SdiA plays a role in regulating AI-2 transport into the bacterial cell and the further processing of AI-2. In E. coli, the lsrACDBFG operon encodes proteins involved in the transport (from the exterior milieu to the interior of the cell) and further modification of AI-2. In the absence of intracellular AI-2, LsrR represses expression of the lsr operon. The AI-2 transport system is encoded by lsrA, lsrC, lsrD, and lsrB. Low level internalization of AI-2 takes place and once inside the cell, AI-2 is phosphorylated by a kinase encoded by lsrK. Phosphorylation of AI-2 leads to derepression of lsr expression, assembly of the lsr transporter, and rapid uptake of AI-2. It is possible that phospho-AI-2 binds to the LsrR repressor and antagonizes its repression of the lsr transport system. Further processing of phospho-AI-2 is carried out by lsrF and lsrG (Wang et al., 2005; Xavier et al., 2007). Zhou et al. (2008) demonstrated that a double mutant, ΔsdiA-ydiV, repressed expression of lsrR and lsrACDBFG, leading to slower internal transport of AI-2. YdiV (a hypothetical protein encoded by ydiV) was identified by microarrays as being upregulated by sdiA overexpression. In addition, there was an ∼2-fold decrease in the intracellular level of cyclic adenosine monophosphate (cAMP) in the double mutant. Addition of cAMP to the double mutant increased expression of the lsr operon, leading to an increase in AI-2 internalization (Zhou et al., 2005). Thus, there is a complex relationship between YdiV, cAMP, and SdiA, which affects the internalization and processing of AI-2.

SdiA and acid resistance

In E. coli K-12 MG1655 (with chromosomal sdiA), the gadA gene was upregulated when C₆-HSL was added; upregulation was noted at 30°C but not at 37°C (Van Houdt et al., 2006). gadA encodes glutamate decarboxylase A, involved in acid resistance in E. coli BW25113 (i.e., Acid Resistance System 2 [AR2]), which allows survival of the bacteria at pH values of 2 to 3. The addition of C₆-HSL increased the tolerance of the wild-type E. coli strain to pH 4, whereas the addition of AHL to an isogenic strain with a deleted sdiA did not lead to acid tolerance and resulted in cell death (Van Houdt et al., 2006). Thus, SdiA is involved in acid resistance. The authors did not complement the mutant E. coli strain with an active sdiA gene followed by addition of C₆-HSL. Lee et al. (2009) showed that overexpression of SdiA downregulated genes associated with acid resistance, including gadA.

When wild-type E. coli O157:H7 was compared to an isogenic sdiA mutant, Hughes et al. (2008) found that there was a 20–50-fold decrease in expression of AR2 genes in the mutant strain. Addition of 3-oxo-C₆-HSL to the wild-type strain gave a 10–40-fold increase in AR2 gene expression. The stimulatory effect was not seen with addition of the AHL to the sdiA mutant (Hughes et al., 2008).

Dyszel et al. (2010a) screened random transposon-based luciferase fusions in E. coli K-12 and E. coli O157:H7 (ATCC 700927) to identify genes that responded to 3-oxo-C₆-HSL. The genes that gave a luciferase response were tested for sdiA dependence. In E. coli K-12, gadW was upregulated and fliE was downregulated, whereas in E. coli O157:H7, gadE, yhiD, and hdeA were upregulated and fliE was downregulated (Dyszel et al., 2010a). The genes gadE, gadW, yhiD, and hdeA are involved in glutamate-dependent acid resistance and are located within the acid fitness island. gadW was induced sixfold by 3-oxo-C₆-HSL at 37°C and 25-fold at 30°C; gadE was induced 2.2-fold by the AHL at 37°C and 16-fold at 30°C; and fliE was repressed fivefold by the AHL at 37°C and 18-fold at 30°C. SdiA-AHL-regulated gene expression was more responsive at 30°C than at 37°C (Dyszel et al., 2010a). In a glucose/glutamate medium at pH 2.0, sdiA ⁺ E. coli K-12 at 30°C showed a ninefold increase in cell survival at 2 h as compared to the sdiA ⁻ strain; survival at 37°C was only increased two- to threefold. In the sdiA ⁺ E. coli O157:H7 strain, survival at pH 2.0 was increased two- to threefold at both 30°C and 37°C (Dyszel et al., 2010a). The addition of 3-oxo-C₆-HSL did not significantly increase survival at pH 2.0, suggesting that the basal level of SdiA activity was sufficient for acid resistance.

Indole at 600 μM repressed expression of the glutamate decarboxylase acid resistance gene two- to fourfold (Lee et al., 2007). Thus, indole decreases acid resistance in E. coli. The trpE mutant (a tryptophan mutant) that produces ∼5-fold less indole was ∼50-fold less sensitive to pH 2.5 as compared to the wild type. An sdiA mutant showed 17-fold decreased survival to pH 2.5 (Lee et al., 2007). The addition of 2 mM indole decreased the survival of E. coli at pH 2.5 by 350- to 650-fold. The addition of 2 mM indole to the sdiA mutant only decreased survival of the organism by ∼4-fold (Lee et al., 2007). It appears that indole represses acid resistance, whereas the sdiA gene (or its product) is necessary for resistance to low pH environments. Lee et al. (2007) suggested that indole controls acid resistance in E. coli through SdiA.

Other potential functions of SdiA

Ghosh et al. (2009), utilizing a strain of E. coli BW25113 lysogenic for λ phage, demonstrated that the addition of a mixture of AHLs to the E. coli strain led to ∼9-fold increase in spontaneous prophage induction as compared to the strain lacking AHL addition. In an isogenic strain lacking the sdiA gene, there was ∼7-fold decrease in phage induction as compared to the wild type in the presence of AHLs. The data of Ghosh et al. (2009) indicate that spontaneous prophage induction in E. coli may be controlled by QS.

When E. coli strains isolated from urinary tract patients were grown in the presence of lomefloxacin (a fluoroquinolone) and ceftazidime (a third-generation cephalosporin), two mutants (one to each drug) were isolated that showed increased resistance to fluoroquinolones, chloramphenicol, and β-lactams (Tavio et al., 2010). Both mutants demonstrated elevated levels of sdiA gene transcripts, indicating that sdiA was overexpressed. The sdiA amplification results in overexpression of the AcrAB-TolC drug-efflux pump, resulting in increased resistance to antibiotics (Tavio et al., 2010). It is not clear why lomefloxacin and ceftazidime led to an increase in sdiA gene transcripts but growth in the presence of these antibiotics may decrease genome stability and favors spontaneous large tandem duplications, leading to overexpressed genes (Tavio et al., 2010)

The binding and responding to AHLs by SdiA interferes with the use of E. coli as a detector/biosensor of AHLs. In the construction of E. coli AHL biosensors by insertion of plasmids containing the LuxR-like biosensors from other organisms, Lindsay and Ahmer (2005) found that the presence of the sdiA gene interferes with the function of the inserted biosensor plasmids, particularly, the Rh1R- and AhyR-biosensors. Lindsay and Ahmer (2005) recommended that sdiA be eliminated from the E. coli strains before their use as biosensors to detect AHLs.

The Role of SdiA in Salmonella enterica Serovar Typhimurium

Sequence analysis of the Salmonella Typhimurium SdiA indicated a 69% amino acid identity to the E. coli SdiA (Ahmer et al., 1998). The sdiA gene was present in all 114 Salmonella enterica strains isolated from Malaysian vegetables and poultry meat belonging to 38 different serotypes, suggesting that the sdiA gene is conserved in S. enterica (Khoo et al., 2009). The gene is probably conserved in E. coli also.

A genetic screen was performed using Salmonella Typhimurium 14028 to identify those genes that might be regulated by sdiA (Ahmer et al., 1998). The screening technique involved cloning the sdiA gene on a multicopy plasmid containing the araBAD promoter and inserting it into an sdiA-negative strain of wild-type Salmonella Typhimurium. Thus, a strain was produced in which expression of sdiA depended on the presence of arabinose (Ahmer et al., 1998). MudJ transposon insertion mutagenesis led to the identification of 10 lacZY transcriptional fusions that responded to arabinose (i.e., sdiA expression). These fusions responded to plasmid sdiA overexpression; however, none of the fusions demonstrated activity in the wild-type or sdiA-negative mutant strains (Ahmer et al., 1998). The failure of the wild-type Salmonella Typhimurium to activate the fusions indicated that the organism does not synthesize AHLs or that the genes regulated by SdiA are also under the control of another regulatory protein (s). In the wild-type strain, the genes regulated by chromosomally encoded SdiA may not be able to activate the promoters, whereas overexpressed SdiA is able to overcome the need for these other regulators and activate transcription in the absence of AHLs (Ahmer et al., 1998).

Seven of the MudJ fusions responding to sdiA overexpression were situated in the rck operon located on the Salmonella Typhimurium virulence plasmid (Ahmer et al., 1998). The rck (resistance to complement killing) operon encodes the Rck protein, an outer membrane protein that induces resistance to the bactericidal action of complement by preventing the polymerization of the complement component C9 on the bacterial surface (C9 is a component of the membrane attack complex of complement) as well as conferring the ability to adhere to and invade Chinese hamster ovary epithelial cells (Cirillo et al., 1996).

An sdiA-positive Salmonella Typhimurium 14028 strain containing a plasmid with the rck promoter fused to luxCDABE showed activation of rck when exogenous AHLs were added, whereas an isogenic strain lacking sdiA did not respond to the AHLs. The most active AHLs were 3-oxo-C₆-HSL and 3-oxo-C₈-HSL (Michael et al., 2001). The binding of 3-oxo-C₆-HSL was demonstrated to be reversible by Janssens et al. (2007). AHLs produced by Yersinia enterocolitica, Chromobacterium violaceum, V. fischeri, and Hafnia alvei activated the Salmonella rck operon (Michael et al., 2001). Michael et al. (2001) found that a Y. enterocolitica mutant (lacking the AHL synthase gene, yenI) did not elicit a response by the Salmonella serovar. YenI is the synthase for C₆-HSL and 3-oxo-C₆-HSL in Y. enterocolitica (Atkinson et al., 2006). Thus, Salmonella Typhimurium can detect the AHLs produced by other bacteria in an SdiA-dependent manner. Expression of sdiA from a multicopy vector resulted in a greater activation of Prck::luxCDABE as compared to the wild-type sdiA-expressing strain of Salmonella Typhimurium. The addition of 3-oxo-C₈-HSL increased the activity of rck in both strains (wild-type and overexpressed sdiA) of Salmonella (Michael et al., 2001).

In swarming cells of Salmonella Typhimurium 14028 expression of sdiA was upregulated as compared to swimming cells (Kim and Surette, 2006). Utilizing a wild-type Salmonella Typhimurium containing an rck::luxCDABE transcription fusion and an sdiA-deficient mutant (also containing the transcription fusion) under swarming conditions, Kim and Surette (2006) found that rck expression was activated by the addition of C₆-HSL or C₈-HSL in the wild type, whereas there was no expression of rck in the AHL-supplemented sdiA-negative mutant.

One of the MudJ fusions outside of the rck operon that responded to overexpression of sdiA was the gene srgE (sdiA-regulated gene E). srgE is a horizontally acquired gene with unknown function located within a 39-kb island (Smith and Ahmer, 2003). The gene product has a predicted molecular mass of 55.7 kDa, and the coding region has a G + C content of 36% as compared to overall Salmonella G + C content of 53% (Smith and Ahmer, 2003). Chromosomal sdiA-dependent activation of psrgE-luxCDABE (electroporated into Salmonella Typhimurium) occurred in the presence of C₆-HSL or 3-oxo-C₈-HSL (Smith and Ahmer, 2003). Thus, Michael et al. (2001) and Smith and Ahmer (2003) have shown that expression of rck and srgE genes is sdiA-dependent in the presence of exogenous AHLs.

Volf et al. (2002) demonstrated that sdiA transcription in the Salmonella Typhimurium F98 strain is suppressed by iron and that the deletion of the fur box or the addition of dipyridyl (an iron chelator) increased transcription of sdiA twofold. The fur gene controls iron uptake and storage (Harvie et al., 2005).

Expression of the srgE gene in Salmonella Typhimurium is SdiA-AHL dependent (Smith and Ahmer, 2003). Smith et al. (2008) investigated whether the gut microbiota of various animal species produce AHLs of the correct type or in the proper concentrations to activate Salmonella SdiA. The Salmonella strain with the srgE::luxCDABE fusion was passaged through various animals, including mice, guinea pigs, rabbits, pigs, calves, chicks, and turtles, to determine if the srg gene was expressed in the salmonellae isolated from animal feces. The AHL-producing microorganisms that allowed expression of srgE by activating SdiA were present only in turtles. The turtles were found to be colonized with Aeromonas hydrophila. Thus, SdiA was activated during transit of Salmonella Typhimurium through the turtle intestinal tract, and the source of AHLs was probably A. hydrophila (Smith et al., 2008). The major AHL produced by A. hydrophila is C₄-HSL, and small amounts of C₆-HSL are also found (Swift et al., 1997). It is surprising that microbially produced AHLs in the other animals tested did not activate Salmonella Typhimurium SdiA.

During the transit of intragastrically inoculated Salmonella Typhimurium 14028 through mice previously infected with Y. enterocolitica, there was activation of SdiA-dependent expression of srgE in Salmonella. However, SdiA is not activated in the absence of Yersinia infection or when the mice were infected with Y. enterocolitica with a mutated yenI gene, indicating that SdiA utilized the Y. enterocolitica–induced AHLs (Dyszel et al., 2010b). In Y. enterocolitica, YenI (encoded by yenI) directs the synthesis of C₆- and 3-oxo-C₆-HSLs (Atkinson et al., 2006). When the yenI gene was inserted into sdiA-positive and sdiA-negative strains of Salmonella Typhimurium 14028, AHLs were produced by both strains of Salmonella. The sdiA⁺/yenI⁺ strain outcompeted the sdiA ⁻/yenI⁺ strain in mice, indicating that the sdiA-positve strain had a competitive edge over the negative strain in the mouse when AHLs were present (increase in fitness); however, both strains grew equally well in Luria-Bertani broth (Dyszel et al., 2010b). Deletion of the rck operon or individual genes located in the operon as well as the srgE gene eliminated the increased fitness of the sdiA⁺/yenI⁺ strain over the sdiA ⁻/yenI⁺ in the mouse (Dyszel et al., 2010b).

In a mixed culture (Luria-Bertani broth at 30°C) of S. enterica and Pectobacterium carotovorum (responsible for soft rot in tomatoes), sdiA-dependent expression of the srgE gene was observed indicating that the AHLs produced by P. carotovorum activated SdiA (Noel et al., 2010). However, the phenomenon was not observed when both organisms were inoculated into tomatoes. Under a variety of storage conditions, Salmonella serovars, in the presence or absence of P. carotovorum, grew in tomatoes; however, there was no expression of sdiA by the Salmonella strains even in advanced stage of soft rot. The acidic conditions, pH 4.2 to 4.4, may be responsible for lack of sdiA expression, but expression was strongly inhibited in tomato pulp buffered to pH 5.4 to 6.0, suggesting that the tomato environment is inhibitory to sdiA expression. Inoculation of tomatoes with a strain of Salmonella Typhimurium carrying a high-copy-number plasmid that overexpressed sdiA led to expression of srgE even in the absence of AHLs (Noel et al., 2010). The failure to detect sdiA expression in tomatoes coinfected with Salmonella and P. carotovorum appears to be due to the lack of expression of the gene rather than due to inhibition of its activity (Noel et al., 2010).

While Rck induces resistance to the bactericidal action of serum complement and affects adherence to and invasion of Chinese hamster ovary cells, the role of the sdiA gene in the virulence of Salmonella Typhimurium is not clear. A mutation in the sdiA gene in which the carboxy terminal helix-turn-helix DNA-binding motif was deleted in the SdiA protein led to increased virulence of the mutant S. Typhimurim in the mouse model (20/40 mice died) as compared to the wild type (12/40 mice died) (Volf et al., 2002). However, Ahmer (2004) stated that Salmonella sdiA mutants were as virulent as the wild type in the mouse, chicken, and bovine models of Salmonella infection. However, it is not clear that Volf et al. (2002) and Ahmer (2004) used similar mutant strains. The use of different mutants may explain the discrepancy of the results.

Taga et al. (2001, 2003) have demonstrated that the transport, uptake, and processing of AI-2 is lsr-mediated in Salmonella Typhimurium in a manner similar to that in E. coli. Since sdiA has a role in AI-2 transport and processing in E. coli, it would be interesting to know if the sdiA gene has a similar role in Salmonella.

The sdiA gene can be used as the basis of a polymerase chain reaction (PCR) detection system for Salmonella. Halatsi et al. (2006) demonstrated that 81 non-Salmonella strains (24 species, including E. coli, Klebsiella pneumoniae, and the 4 species of Shigella) were negative in the Salmonella sdiA PCR assay, whereas 155 strains (101 serotypes) of Salmonella were positive. Using artificially inoculated human feces, Halatsi et al. (2006) and Oikonomou et al. (2008) were able to identify Salmonella Enteritidis by the PCR method with a detection limit of 10³ CFU/g but when pre-enrichment and enrichment were used, the detection limit of the PCR method was 10² CFU/g. Thus, PCR assays utilizing primers based on the sdiA gene sequences from Salmonella can be used to specifically detect the organism.

Production of AHLs in Food

Bacteria have been shown to produce AHLs in food products. Gram et al. (1999) demonstrated that AHLs were produced in cold-smoked salmon stored under N₂ at 5°C when the cell densities of psychrotrophic strains of Enterobacteriaceae were ≥10⁶ CFU/g. Similar results in terms of AHL production and Enterobacteriaceae levels were obtained with vacuum-packed beef stored for 28 days at 5°C (Bruhn et al., 2004). The presence of Enterobacteriaceae at ≥10⁶ cfu/g and AHL-mediated gene regulation may play a role in bacterial induction of off-flavors and food spoilage. The production of AHLs were examined in A. hydrophila, Y. enterocolitica, and P. aeruginosa strains grown on agars containing vegetable, meat, and fish extracts (Medina-Martínez et al., 2006a, b). In general, AHLs were produced by the bacterial species on meat and fish extracts, but production was more variable on vegetable extracts since several vegetable extracts appeared to inhibit AHL production.

Several bacteria produce enzymes, lactonases and acylases, that degrade AHLs (Dong and Zhang, 2005; Czajkowski and Jafra, 2009). Dong et al. (2001) demonstrated that insertion of a bacterial AHL lactonase gene into potato plants prevented the soft-rotting of potato tuber tissues induced by Erwinia carotovora. The capacity of E. carotovora to cause soft-rotting depends on the ability to secrete proteins and enzymes, which is AHL-dependent (Cui et al., 2005). The seaweed, Delisea pulchra, produces halogenated furanones (structurally similar to AHLs), which inhibit AHL action by binding to the LuxR receptor (Givskov et al., 1996). Additional studies are needed to adequately address the role of AHLs in foods and the possibility of inhibiting AHL action in foods.

Conclusions

A note of caution is necessary concerning the interpretation of the effects obtained with the overexpression of sdiA. Many of the studies on the role of SdiA in Salmonella Typhimurium and E. coli involved overexpression of the sdiA gene by cloning the gene into multicopy plasmids followed by transformation of the plasmid into the bacterial cell. Overexpression of the sdiA gene does not represent the normal physiological state; therefore, the results of these studies must be viewed with caution. Expression of specific genes must be shown to be regulated significantly by chromosomal sdiA in the presence versus the absence of exogenous AHLs to determine that the gene is a component of the SdiA regulon (Dyszel et al., 2010a).

The Salmonella chromosomal sdiA gene is involved in the activation of the rck gene when exogenous AHLs are present. The Rck protein induces resistance to complement activity and allows adhesion and invasion of epithelial cells by Salmonella; thus, Rck is involved in virulence of Salmonella Typhimurium.

E. coli with a chromosomal sdiA gene upregulated gadA in the presence of C₆-HSL, and the bacteria demonstrated greater resistance to a pH 4.0 environment. Wild-type E. coli was several-fold more resistant to pH 2.5 as compared to an isogenic strain with a mutated sdiA gene. The addition of AHLs to wild-type E. coli inhibited motility and biofilm formation. Indole appears to control acid resistance and biofilm formation through SdiA. In E. coli O157:H7, SdiA, in the presence of AHL, repressed transcription of LEE genes; therefore, sdiA may play a role in limiting the virulence of E. coli O157:H7. A double mutant ΔsdiAΔydiV (ydiV encodes a hypothetical protein) of E. coli showed a decrease in the level of intracellular cAMP. In addition, there was a decrease in expression of the lsr operon (involved in transport and internalization of AI-2) in the double mutant. Addition of cAMP to the double mutant increased lsr expression. Thus, cAMP is involved in the transport of AI-2 into the bacterial cell, and SdiA has a role in regulating expression of the lsr operon and accumulation of cAMP. Studies indicate that chromosomally expressed sdiA appears to be involved in virulence, acid resistance, motility, and biofilm formation in E. coli. In addition, SdiA, in a complex system involving cAMP, has a role in regulating the internalization and further processing of AI-2.

It is probable that the regulation of additional genes in E. coli and Salmonella is under the control of the chromosomal sdiA gene. The published literature indicates that only a few studies have been performed examining the role of SdiA-AHL in Salmonella Typhimurium and E. coli, and further research is necessary to discover other activities influenced by the sdiA gene and SdiA-AHL in these and other organisms. Information on the activation of SdiA in food products contaminated with AHL-producing bacteria is lacking. Since Salmonella species and diarrheagenic E. coli are important food-borne pathogens, the role of SdiA in regulating growth and survival of these pathogens in foods should be investigated. Use of proteomic and genomic technologies, either in foods or in model food systems, will help elucidate the genes, proteins, and metabolic pathways that are affected by QS via SdiA. The construction of mutant genes that are up- or downregulated by sdiA should provide information on how those genes affect the behavior of the organism in the presence or absence of SdiA-AHLs. Further, a number of compounds have been isolated or synthesized that antagonize/interfere with QS systems, and the application of these antagonists may potentially be useful in inhibiting growth or virulence mechanisms of bacteria in different environments, including foods. An increased understanding of the role of SdiA and other QS systems may make it possible for foods to be formulated that will interrupt QS and inhibit the growth of spoilage or pathogenic organisms, leading to improved food quality and safety.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Ahmer

BMM

. Cell-to-cell signaling in Escherichia coli and Salmonella enterica. Mol Microbiol, 2004; 52:933–945.

Ahmer

BMM

, van Reeuwijk

, Timmers

, Valentine

, Heffron

. Salmonella Typhimurium encodes an SdiA homolog, a putative quorum sensor of the LuxR family, that regulates genes on the virulence plasmid. J Bacteriol, 1998; 180:1185–1193.

Atkinson

, Chang

C-Y

, Sockett

, Cámara

, Williams

. Quorum sensing in Yersinia enterocolitica controls swimming and swarming motility. J Bacteriol, 2006; 188:1451–1461.

Atkinson

, Williams

. Quorum sensing and social networking in the microbial world. J R Soc Interface, 2009; 6:959–978.

Bruhn

, Christensen

, Flodgaard

, Nielsen

, Larsen

, Givskov

, Gram

. Presence of acylated homoserine lactones (AHLs) and AHL-producing bacteria in meat and potential role of AHL in spoilage of meat. Appl Environ Microbiol, 2004; 70:4293–4303.

Case

, Labbate

, Kjelleberg

. AHL-driven quorum sensing circuits: their frequency and function among the proteobacteria. ISME J, 2008; 2:345–349.

Cirillo

, Heffernan

, Wu

, Harwood

, Fierer

, Guiney

. Identification of a domain in Rck, a product of the Salmonella Typhimurium virulence plasmid, required for both serum resistance and cell invasion. Infect Immun, 1996; 64:2019–2023.

Cui

, Chatterjee

, Hasegawa

, Dixit

, Leigh

, Chatterjee

. ExpR, a LuxR homolog of Erwinia carotovora subsp. Carotovora, which specifies a global regulatory RNA-binding protein. J Bacteriol, 2005; 187:4792–4803.

Czajkowski

, Jafra

. Quenching of acyl-homoserine lactone-dependent quorum sensing by enzymatic disruption of signal molecules. Acta Biochim Pol, 2009; 56:1–16.

10.

DeLisa

, Wu

C-F

, Wang

, Valdes

, Bentley

. DNA microarray-based identification of genes controlled by autoinducer 2-stimulated quorum sensing in Escherichia coli. J Bacteriol, 2001; 183:5239–5247.

11.

de Sablet

, Chassard

, Bernalier-Donadille

, Vareille

, Gobert

, Martin

. Human microbiota-secreted factors inhibit Shiga-toxin synthesis by enterohemorrhagic Escherichia coli O157:H7. Infect Immun, 2009; 77:783–790.

12.

Dong

Y-H

, Wang

L-H

, Xu

J-L

, Zhang

H-B

, Zhang

X-F

, Zhang

L-H

. Quenching quorum-sensing-dependent bacterial infection by an N-acyl homoserine lactonase. Nature, 2001; 411:813–817.

13.

Dong

Y-H

, Zhang

L-H

. Quorum sensing and quorum-quenching enzymes. J Microbiol, 2005; 43,No. S:101–109.

14.

Dyszel

, Soares

, Swearingen

, Lindsay

, Smith

, Ahmer

BMM

. E. coli K-12 and EHEC genes regulated by SdiA. PLoS ONE, 2010a. 5:e8946.

15.

Dyszel

, Smith

, Lucas

, Soares

, Swearingen

, Vross

, Young

, Ahmer

BMM

. Salmonella enterica serovar Typhimurium can detect acyl homoserine lactone production by Yersinia enterocolitica in mice. J Bacteriol, 2010b. 192:29–37.

16.

Dziva

, van Diemen

, Stevens

, Smith

, Wallis

. Identification of Escherichia coli O157:H7 genes influencing colonization of the bovine gastrointestinal tract using signature-tagged mutagenesis. Microbiology, 2004; 150:3631–3646.

17.

Fuqua

. The QscR quorum-sensing regulon of Pseudomonas aeruginosa: an orphan claims its identity. J Bacteriol, 2006; 188:3169–3171.

18.

Ghosh

, Roy

, Williamson

, Srinivasiah

, Wommack

, Radosevich

. Acyl-homonserine lactones can induce virus production in lysogenic bacteria: an alternative paradigm for prophage induction. Appl Environ Microbiol, 2009; 75:7142–7152.

19.

Givskov

, de Nys

, Manefield

, Gram

, Maximilien

, Eberl

, Molin

, Steinberg

, Kjelleberg

. Eukaryotic interference with homoserine lactone-mediated prokaryotic signaling. J Bacteriol, 1996; 178:6618–6622.

20.

Gobbetti

, De Angelis

, Di Cagno

, Minervini

, Limitone

. Cell-cell communication in food related bacteria. Int J Food Microbiol, 2007; 120:34–45.

21.

Gram

, Christensen

, Ravn

, Molin

, Givskov

. Production of acylated homoserine lactones by psychrotrophic members of the Enterobacteriaceae isolated from foods. Appl Eviron Microbiol, 1999; 65:3458–3463.

22.

Gray

, Garey

. The evolution of bacterial LuxI and LuxR quorum sensing regulators. Microbiology, 2001; 147:2379–2387.

23.

Gyles

. Shiga toxin-producing Escherichia coli: an overview. J Anim Sci, 2007; 85:E45–E62.

24.

Haack

, Robinson

, Miller

, Fowlkes

, Mellies

. Interaction of Ler at the LEE5 (tir) operon of enteropathogenic Escherichia coli. Infect Immun, 2003; 71:384–392.

25.

Halatsi

, Oikonomou

, Lambiri

, Mandilara

, Vatopoulos

, Kyriacou

. PCR detection of Salmonella spp. using primers targeting the quorum sensing gene sdiA. FEMS Microbiol Lett, 2006; 259:201–207.

26.

Harvie

, Vílchez

, Steggles

, Ellar

. Bacillus cereus Fur regulates iron metabolism and is required for full virulence. Microbiology, 2005; 151:569–577.

27.

Hirakawa

, Inazumi

, Masaki

, Hirata

, Yamaguchi

. Indole induces the expression of multidrug exporter genes in Escherichia coli. Mol Microbiol, 2005; 55:1113–1126.

28.

Hughes

, Gonzalez

, Edrington

, Sperandio

. Enterohemorrhagic Escherichia coli (EHEC) O157:H7 regulates ler, the master regulator of the LEE genes and the glutamate-dependent acid resistance system through the LuxR homologue, SdiA. Abstract B-143 presented at the 108th General Meeting, American Society of Microbiology, Boston, MA, 2008.

29.

Janssens

JCA

, Metzger

, Daniels

, Ptacek

, Verhoeven

, Habel

, Vanderlelyden

, de Vos

, de Keersmaecker

SCJ

. Synthesis of N-acyl homoserine lactone analogues reveals strong activators of SdiA, the Salmonella enterica serovar Typhimurium LuxR homologue. Appl Environ Microbiol, 2007; 73:535–544.

30.

Kanamaru

, Kanamaru

, Tatsuno

, Tobe

, Sasakawa

. SdiA, an Escherichia coli homologue of quorum-sensing regulators, controls the expression of virulence factors in enterohaemorrhagic Escherichia coli O157:H7. Mol Microbiol, 2000; 38:805–816.

31.

Kim

, Surette

. Coordinated regulation of two independent celll-cell signaling systems and swarmer differentiation in Salmonella enterica serovar Typhimurium. J Bacteriol, 2006; 188:431–440.

32.

Khoo

C-H

, Cheah

Y-K

, Lee

L-H

, Sim

J-H

, Salleh

, Sidik

, Radu

, Sukardi

. Virulotyping of Salmonella enterica subsp. Enterica isolated from indigenous vegetables and poultry meat in Malaysia using multiplex-PCR. Antonie van Leeuwenhoek, 2009; 96:441–457.

33.

Law

. Virulelnce factors of Escherichia coli O157 and other Shiga toxin-producing E. coli. J Appl Microbiol, 2000; 88:729–745.

34.

Lee

, Jayaraman

, Wood

. Indole is an inter-species biofilm signal mediated by SdiA. BMC Microbiol, 2007; 7:42.

35.

Lee

, Maeda

, Hong

, Wood

. Reconfiguring the quorum–sensing regulator SdiA of Escherichia coli to control biofilm formation via indole and N-acylhomoserine lactones. Appl Environ Microbiol, 2009; 75:1703–1716.

36.

Lee

, Zhang

X-S

, Hegde

, Bentley

, Jayaraman

, Wood

. Indole cell signaling occurs primarily at low temperatures in Escherichia coli. ISME J, 2008; 2:1007–1023.

37.

Lequette

, Lee

J-H

, Ledgham

, Lazdunski

, Greenberg

. A distinct QscR regulon in the Pseudomonas aeruginosa quorum-sensing circuit. J Bacteriol, 2006; 188:3365–3370.

38.

Lerat

, Moran

. The evolutionary history of quorum-sensing systems in bacteria. Mol Biol Evol, 2004; 21:903–913.

39.

Lindsay

, Ahmer

BMM

. Effect of sdiA on biosensors of N-acylhomoserine lactones. J Bacteriol, 2005; 187:5054–5058.

40.

Malott

, O'Grady

, Toller

, Inhülsen

, Eberl

, Sokol

. A Burkholderia cenocepacia orphan LuxR homolog is involved in quorum sensing regulation. J Bacteriol, 2009; 191:2447–2460.

41.

McDougald

, Rice

, Kjelleberg

. Bacterial quorum sensing and interference by naturally occurring biomimics. Anal Bioanal Chem, 2007; 387:445–453.

42.

Medina-Martínez

, Uyttendaele

, Demolder

, Debrevere

. Influence of food system conditions on N-acyl-L-homoserine lactones production by Aeromonas species. Int J Food Microbiol, 2006a. 112:244–252.

43.

Medina-Martínez

, Uyttendaele

, Meireman

, Debrevere

. Relevance of N-acyl-L-homoserine lactones production by Yersinia enterocolitica in fresh foods. J Appl Microbiol, 2006b. 102:1150–1158.

44.

Mellies

, Barron

AMS

, Carmona

. Enteropathogneic and enterohemorrhagic Escherichia coli virulence gene regulation. Infect Immun, 2007; 75:4199–4210.

45.

Michael

, Smith

, Swift

, Heffron

, Ahmer

BMM

. SdiA of Salmonella enterica is a LuxR homolog that detects mixed microbial communities. J Bacteriol, 2001; 183:5733–5742.

46.

W-L

, Bassler

. Bacterial quorum-sensing network architectures. Annu Rev Genet, 2009; 43:197–222.

47.

Nikaido

. Multiple antibiotic resistance and efflux. Curr Opin Microbiol, 1998; 1:516–523.

48.

Noel

, Joy

, Smith

, Fatica

, Schneider

, Ahmer

BMM

, Teplitski

. Salmonella SdiA recongnizes N-acyl homoserine lactone signals from Pectobacterium carotovorum in vitro, but not in a bacterial soft rot. Mol Plant Microbe Interact, 2010; 23:273–282.

49.

Oikonomou

, Halatsi

, Kyriacou

. Selective PCR: a novel internal amplification control strategy for enhanced sensitivity in Salmonella diagnosis. Lett Appl Microbiol, 2008; 46:456–461.

50.

Patankar

, González

. Orphan LuxR regulators of quorum sensing. FEMS Microbiol Rev, 2009; 33:739–756.

51.

Rahmati

, Yang

, Davidson

, Zechiedrich

. Control of the AcrAB multidrug efflux pump by quorum-sensing regulator SdiA. Mol Microbiol, 2002; 43:677–685.

52.

Saitou

, Nei

. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol, 1987; 4:406–425.

53.

Sharma

, Bearson

. Evaluation of the effects of sdiA, a luxR homologue, on adherence and motility of Escherichia coli O157:H7. Microbiology, 2010; 156:1303–1312.

54.

Sitnikov

, Schineller

, Baldwin

. Control of cell division in Escherichia coli: regulation of transcription of ftsQA involves both rpoS and SdiA-mediated autoinduction. Proc Natl Acad Sci USA, 1996; 93:336–341.

55.

Smith

, Fratamico

, Novak

. Quorum sensing: a primer for food microbiologists. J Food Prot, 2004; 67:1053–1070.

56.

Smith

, Ahmer

BMM

. Detection of other microbial species by Salmonella: expression of the SdiA regulon. J Bacteriol, 2003; 185:1357–1366.

57.

Smith

, Dyszel

, Soares

, Ellermeier

, Altier

, Lalwhon

, Adams

, Konjufca

, Curtiss

III , Slauch

, Ahmer

BMM

. SdiA, an N-acylhomoserine lactone receptor, becomes active during the transit of Salmonella enterica through the gastrointestinal tract of turtles. PLoS ONE, 2008; 3:e2826.

58.

Soulére

, Frezza

, Queneau

, Doutheau

. Exploring the active site of acyl homoserine lactone-dependent transcriptional regulators with bacterial quorum sensing modulators using molecular mechanics and docking studies. J Mol Graph Model, 2007; 26:581–590.

59.

Sperandio

, Torres

, Giron

, Kaper

. Quorum sensing is a global regulatory mechanism in enterohemorrhagic Escherichia coli O157:H7. J Bacteriol, 2001; 183:5187–5197.

60.

Sperandio

, Torres

, Jarvis

, Nataro

, Kaper

. Bacteria-host communication: the language of hormones. Proc Natl Acad Sci USA, 2003; 100:8951–8956.

61.

Subramoni

, Venturi

. LuxR-family “solos”: bachelor sensors/regulators of signaling molecules. Microbiology, 2009; 155,Pt 5:1377–1385.

62.

Suzuki

, Wang

, Weilbacher

, Pernestig

A-K

, Melefors

, Georgellis

, Babitzke

, Romeo

. Regulatory circuitry of the CsrA/CsrB and BarA/UvrY systems of Escherichia coli. J Bacteriol, 2002; 184:5130–5140.

63.

Swift

, Karlyshev

, Fish

, Durant

, Winson

, Chhabra

, Williams

, MacIntyre

, Stewart

GSAB

. Quorum sensing in Aeromonas hydrophila and Aeromonas salmonicida: identification of the LuxRI homologs AhyRI and AsaRI and their cognate N-acylhomoserine lactone signal molecules. J Bacteriol, 1997; 179:5271–5281.

64.

Taga

, Miller

, Bassler

. Lsr-mediated transport and processing of AI-2 in Salmonella typhimurium. Mol Microbiol, 2003; 50:1411–1427.

65.

Taga

, Semmelhak

, Bassler

. The LuxS-dependent autoinducer AI-2 controls the expression of an ABC transporter that functions in AI-2 uptake in Salmonella typhimurium. Mol Microbiol, 2001; 42:777–793.

66.

Tavio

, Aquili

, Poveda

, Antunes

, Sánchez-Céspedes

, Vila

. Quorum-sensing regulator sdiA and marA overexpression is involved in in vitro-selected multidrug resistance of Escherichia coli. J Antimicrob Chemother, 2010; 65:1178–1186.

67.

Van Houdt

, Aertsen

, Moons

, Vanoirbeek

, Michiels

. N-acyl-homoserine lactone signal interception by Escherichia coli. FEMS Microbiol Lett, 2006; 256:83–89.

68.

Vendeville

, Winzer

, Heurlier

, Tang

, Hardie

. Making “sense” of metabolism: autoinducer-2, LuxS, and pathogenic bacteria. Nat Rev Microbiol, 2005; 3:383–396.

69.

Volf

, Seveik

, Havlickova

, Sisak

, Damborsky

, Rychlik

. Role of SdiA in Salmonella enterica serovar Typhimurium physiology and virulence. Arch Microbiol, 2002; 178:94–101.

70.

Walters

, Sperandio

. Quorum sensing in Escherichia coli and Salmonella. Int J Med Microbiol, 2006; 296:125–131.

71.

Wang

, Hashimoto

, Tsao

C-Y

, Valdes

, Bentley

. Cyclic AMP (cAMP) and cAMP receptor protein influence both synthesis and uptake of extracellular autoinducer 2 in Escherichia coli. J Bacteriol, 2005; 187:2066–2076.

72.

Wang

, de Boer

PAJ

, Rothfield

. A factor that positively regulates cell division by activating transcription of the major cluster of essential cell division genes of Escherichia coli. EMBO J, 1991; 10:3363–3372.

73.

Wei

, Lee

J-M

, Smulski

, LaRossa

. Global impact of sdiA amplification revealed by the comprehensive gene expression profiling of Escherichia coli. J Bacteriol, 2001a. 183:2265–2272.

74.

Wei

, Vollmer

, LaRossa

. In vivo titration of mitomycin C action by four Escherichia coli genomic regions on multicopy plasmids. J Bacteriol, 2001b. 183:2259–2264.

75.

Williams

. Quorum sensing, communication and cross-kingdom signalling in the bacterial world. Microbiology, 2007; 153:3923–3938.

76.

Williams

, Winzer

, Chan

, Cámara

. Look who's talking: communication and quorum sensing in the bacterial world. Phil Trans R Soc B, 2007; 362:1119–1134.

77.

Wood

, González Barrios

, Herzberg

, Lee

. Motility influences biofilm architecture in Escherichia coli. Appl Microbiol Biotechnol, 2006; 72:361–367.

78.

Xavier

, Miller

, Lu

, Kim

, Rabinowitz

, Pelczer

, Semmelhack

, Bassler

. Phosphorylation and processing of the quorum-sensing molecule autoinducer-2 in enteric bacteria. ACS Chem Biol, 2007; 2:128–136.

79.

Yao

, Martinez-Yamout

, Dickerson

, Brogan

, Wright

, Dyson

. Structure of Escherichia coli quorum sensing protein SdiA: activation of the folding switch by acyl homoserine lactones. J Mol Biol, 2006; 355:262–273.

80.

Zhou

, Meng

, Sun

. An EAL domain protein and cyclic AMP contribute to the interaction between the two quorum sensing systems in Escherichia coli. Cell Res, 2008; 18:939–948.

81.

Zhu

, Winans

. The quorum-sensing transcriptional regulator TraR requires its cognate signaling ligand for protein folding, protease resistance, and dimerization. Proc Natl Acad Sci USA, 2001; 98:1507–1512.