Transcriptome Analysis of Listeria Monocytogenes Identifies Three Groups of Genes Differently Regulated by Prfa

Blackwell Science, LtdOxford, UKMMIMolecular Microbiology0950-382XBlackwell Publishing Ltd, 200347Original ArticleAnalysis of the PrfA regulon of L. monocytogenesE. Milohanic et al.

Molecular Microbiology (2003) 47(6), 1613–1625

Transcriptome analysis of Listeria monocytogenes identiﬁes three groups of genes differently regulated by PrfA

Eliane Milohanic,1,2 Philippe Glaser,2 regulation by PrfA of most group III genes. In the Jean-Yves Coppée,3 Lionel Frangeul,2 Yolanda Vega,4 presence of cellobiose, all the group I genes were José A. Vázquez-Boland,4† Frank Kunst,2 downregulated, whereas group III genes remained Pascale Cossart1 and Carmen Buchrieser2* fully activated. Group II genes were repressed in all 1Unité des Interactions Bactéries-Cellules, 2Laboratoire de conditions tested. A comparison of the expression Génomique des Microorganismes Pathogènes, and profiles between a second L. monocytogenes strain 3Plateau Technique Puces à ADN, Institut Pasteur, 28, (P14), its spontaneous mutant expressing a constitu- Rue du Docteur Roux, 75724 Paris Cedex 15, France. tively active PrfA variant (P14prfA*) and its corre- 4Grupo de Patogénesis Molecular y Genómica sponding prfA-deleted mutant (P14DprfA) and the Bacteriana, Facultad de Veterinaria, Universidad de Leon, EGDe strain revealed interesting strain-specific differ- 24071, Spain. ences. Sequences strongly similar to a sigma B- dependent promoter were identified upstream of 22 group III genes. These results suggest that PrfA Summary positively regulates a core set of 12 genes preceded PrfA is the major regulator of Listeria virulence gene by a PrfA box and probably expressed from a sigma expression. This protein is a member of the Crp/Fnr A-dependent promoter. In contrast, a second set family of transcription regulators. To gain a deeper of PrfA-regulated genes lack a PrfA box and are understanding of the PrfA regulon, we constructed a expressed from a sigma B-dependent promoter. This whole-genome array based on the complete genome study reveals that PrfA can act as an activator or a sequence of Listeria monocytogenes strain EGDe and repressor and suggests that PrfA may directly or indi- evaluated the expression profiles of the wild-type rectly activate different sets of genes in association EGDe and a prfA-deleted mutant (EGDe DprfA). Both with different sigma factors. strains were grown at 37∞C in brain–heart infusion broth (BHI) and BHI supplemented with either acti- Introduction vated charcoal, a compound known to enhance Listeria monocytogenes is a Gram-positive, facultative virulence gene expression, or cellobiose, a sugar intracellular bacterial pathogen that causes severe food- reported to downregulate virulence gene expression borne infections in humans and animals. The main in spite of full expression of PrfA. We identified three symptoms include meningoencephalitis, septicaemia groups of genes that are regulated differently. Group and abortions and, as shown recently, gastroenteritis I comprises, in addition to the 10 already known (Vazquez-Boland et al., 2001). Several L. monocytogenes genes, two new genes, lmo2219 and lmo0788, both virulence genes have been identified. They include genes positively regulated and preceded by a putative PrfA involved in adherence and uptake by the host cell, escape box. Group II comprises eight negatively regulated from the phagocytic vacuoles and intracellular replication genes: lmo0278 is preceded by a putative PrfA box, or in intra- and intercellular movement (Cossart and and the remaining seven genes (lmo0178–lmo0184) Lecuit, 1998; Vazquez-Boland et al., 2001). The only are organized in an operon. Group III comprises 53 regulatory factor identified to date, necessary for the reg- genes, of which only two (lmo0596 and lmo2067) are ulation of the expression of most of these virulence genes, preceded by a putative PrfA box. Charcoal addition is PrfA. PrfA activates all genes of the so-called virulence induced upregulation of group I genes but abolished gene cluster of L. monocytogenes (prfA, plcA, hly, mpl, actA and plcB) as well as the expression of inlA and inlB, Accepted 2 December, 2002. *For correspondence. E-mail which encode two invasion proteins (InlA and InlB) [email protected]; Tel. (+33) 1 45 68 83 72; Fax (+33) 1 45 68 87 86. (Dramsi et al., 1996), inlC, which encodes a small †Present address: Molecular Microbiology Section, Department of Clinical Veterinary Science, University of Bristol, Langford, BS40 secreted internalin-like protein (InlC) (Engelbrecht et al., 5DU, UK 1996; Lingnau et al., 1996), and hpt, a gene encoding a

1614 E. Milohanic et al. UhpT-related sugar phosphate transporter that mediates Results rapid intracellular proliferation (Chico-Calero et al., 2002). Listeria monocytogenes DNA macroarray and PrfA is structurally and functionally related to the Crp/ experimental design Fnr family of transcription regulators (Lampidis et al., 1994). Most members of the Crp/Fnr family have been Whole-genome arrays were constructed by amplifying studied in Gram-negative bacteria, except the Flp pro- ª500-bp-long PCR products specific for each gene using tein of Lactobacillus casei (Irvine and Guest, 1993), Fnr open reading frame (ORF)-specific primers. Ninety-nine of Bacillus subtilis, a regulator involved in anaerobic per cent (2816 ORFs) of the 2853 predicted ORFs of the respiration (Cruz Ramos et al., 1995), and PrfA, one of L. monocytogenes EGDe genome were amplified suc- the best-studied representatives of this family in Gram- cessfully. The PCR products were spotted onto nylon positive bacteria. Crp/Fnr-like regulators are site-specific membranes and then hybridized with 33P-labelled single- DNA-binding proteins possessing a C-terminal helix– stranded cDNA derived from total cellular RNA. The turn–helix (HTH) DNA-binding motif and a b-roll expression profiles of the different L. monocytogenes wild- structure in their N-terminal domains. PrfA binds to a type and mutant strains were studied after growth at 37∞C palindromic PrfA recognition sequence (PrfA box) in BHI and BHI supplemented with activated charcoal located at position -40 from the transcription start site (BHIC) or cellobiose (BHICel). Charcoal increases the in PrfA-dependent promoters (Mengaud et al., 1989; transcription levels of prfA (Ripio et al., 1996) and genes Sheehan et al., 1996). PrfA-dependent transcription of under its control (Geoffroy et al., 1989), whereas cellobi- virulence genes is weak below 30∞C but becomes ose downregulates virulence gene expression (Park and induced at 37∞C because of a switch of the non-trans- Kroll, 1993; Brehm et al., 1996; Renzoni et al., 1997). lated leader RNA of PrfA between an inactive structure Total RNA was purified from a culture grown to OD600 = 0.6 at low temperatures and an active structure at high as it has been shown previously that the overall amount temperatures (Johansson et al., 2002). The medium of prfA transcript drops in stationary phase (Mengaud composition is critical for full expression of PrfA- et al., 1991). For each growth condition, two different controlled genes in L. monocytogenes. Starvation con- mRNA preparations from independent cultures were used ditions (incubation in minimal essential medium; Bohne for cDNA synthesis and subsequent hybridization to two et al., 1994) or incubation in brain–heart infusion (BHI) sets of arrays. To identify statistically significant differ- medium supplemented with activated charcoal (Ripio ences in gene expression, we used the Statistical Analysis et al., 1996) induce the PrfA regulon. In contrast, for Microarrays (SAM) program (Tusher et al., 2001). All metabolizable unphosphorylated sugars, such as glu- genes with statistically significant changes in the level of cose, maltose, fructose, mannose, trehalose and expression and with at least a twofold change were con- cellobiose, inhibit the expression of PrfA-dependent sidered in this analysis. All primary data from the tran- virulence genes (Park and Kroll, 1993; Milenbachs scriptome experiments and the statistical analysis are et al., 1997; Ripio et al., 1997a). However, in the available as Supplementary material. presence of cellobiose, PrfA is fully expressed, suggesting that PrfA is post-transcriptionally modified (Renzoni et al., 1997) and can switch between a transcriptionally Expression profiles of wild-type L. monocytogenes EGDe active and inactive form upon interaction with a compared with its isogenic prfA mutant in BHI hypothetical activating factor (Renzoni et al., 1997; Ripio et al., 1997b; Vega et al., 1998). Analysis of the expression profiles in BHI revealed 70 Using the extensive data set obtained during the genes, organized in 47 predicted transcriptional units (13 genome sequencing project of L. monocytogenes EGDe predicted operons and 34 single genes), differentially tran- (Glaser et al., 2001) to construct whole-genome macro- scribed in the DprfA mutant relative to the wild-type EGDe arrays, we have been able to study the global regulatory strain. Sixty-two of these genes were upregulated, capacity of PrfA. Gene expression patterns of two L. whereas eight (organized in two transcriptional units) were monocytogenes wild-type strains (EGDe and P14) and downregulated in the wild-type EGDe strain when com- their isogenic prfA-deleted mutants as well as a prfA* pared with the prfA mutant strain (Table 1). These genes mutant (P14prfA*) constitutively overexpressing all PrfA- encode primarily known or predicted transport proteins dependent virulence genes were studied and compared (20), proteins involved in stress response (15) and pro- in different conditions. Three groups of genes differently teins of unknown function (21) (Table 1). From the 10 regulated by PrfA were identified. Real-time quantitative previously known PrfA-dependent virulence genes, seven polymerase chain reaction (PCR) was used to corrobo- (prfA, plcA, hly, mpl, actA, plcB and inlA), were detected rate the gene expression patterns revealed by the as clearly upregulated by PrfA, whereas three (inlB, inlC macroarrays. and hpt) were slightly or not PrfA activated in BHI. Sixty-

Analysis of the PrfA regulon of L. monocytogenes 1615

Table 1. Genes differentially regulated in wild-type L. monocytogenes EGDe and P14prfA* relative to their isogenic prfA deleted mutants.

© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 47, 1613–1625 1616 E. Milohanic et al. three PrfA-regulated genes are newly identified in this regulated by PrfA were not activated in BHI (see above). study. It has been reported that activated charcoal in the culture As PrfA binds to a specific palindromic sequence, the medium increases the synthesis of listeriolysin (LLO) PrfA box (consensus TTAACAnnTGTTAA), we searched encoded by the hly gene and also that of the lecithinase for putative PrfA boxes upstream of all the newly identified encoded by the plcB gene (Geoffroy et al., 1989; 1991; genes. Five of these genes (lmo2219, lmo0788, lmo0278, Ripio et al., 1996), suggesting that charcoal somehow lmo0596, lmo2067) are preceded by a putative PrfA box exerts an effect on the L. monocytogenes virulence gene (Table 2). Gene lmo2219 encodes a protein similar to regulation mechanisms. Analysis of the expression levels PrsA of Bacillus subtilis, a post-translocation molecular showed that, after growth in BHIC, all previously chaperon (Kontinen et al., 1991; Kontinen and Sarvas, described PrfA-regulated genes, including inlB, inlC and 1993), genes lmo0788 and lmo0596 code for probable hpt, were indeed regulated by PrfA in the EGDe wild membrane proteins of unknown function, lmo0278 codes type. The level of PrfA activation was increased between for a sugar ABC transporter and lmo2067 is a gene two- and fourfold compared with growth without charcoal coding for a bile salt hydrolase recently shown to be PrfA (Table 1). The upregulation of two of the newly identified regulated and involved in the intestinal and hepatic genes, lmo2219 and lmo0788, was similar to that of the phases of listeriosis (Dussurget et al., 2002). In contrast, known virulence genes. The eight genes identified as for the remaining 37 genes (single genes or first genes being negatively regulated by PrfA (lmo0278, lmo0178– of a predicted operon), no sequences similar to PrfA lmo0184) in BHI remained repressed at a similar level in boxes were identified. Using MEME (Bailey and Elkan, BHIC (Table 1). Surprisingly, 52 of the remaining 53 PrfA- 1994), a tool for discovering motifs in a group of regulated genes were no longer activated by PrfA in related DNA sequences, and BIOPROSPECTOR (http:// BHIC (Table 1). They showed expression levels similar to bioprospector. standford.edu/) allowing for the modelling that of EGDeDprfA grown in BHI (Supplementary of gapped motifs and motifs with palindromic patterns (Liu material). In summary, the addition of charcoal to the et al., 2001), we searched for an alternative PrfA binding medium allowed us to define three groups of genes. site upstream of these newly identified genes. No other Group I comprises 12 genes, including the previously putative PrfA binding sites were found. Interestingly, many known virulence genes, the activation of which by PrfA is of these genes have homologues, which belong to the increased upon charcoal treatment. Group II comprises stress-induced sigma B regulon in B. subtilis (Petersohn eight negatively regulated genes, the regulation of which et al., 2001; Price et al., 2001). Therefore, a search for is not altered by the presence of charcoal, and group III putative sigma B promoters using a matrix based on comprises the 53 remaining genes, the activation of 47 B. subtilis sigma B promoter sequences extracted from which by PrfA is abolished in growth in BHIC (Table 1). the B. subtilis regulatory/promoter database DBTB Furthermore, 12 additional upregulated and four down- (http://elmo.ims.u-tokyo.ac.jp/dbtbs) was undertaken in regulated genes were identified in BHIC compared with the L. monocytogenes EGDe genome sequence (F. growth in BHI. The primary data and the statistical analy- Chetouani and M. S. Gelfand, unpublished data). We also sis are available as Supplementary material (Tables S2A conducted a motif search using the B. subtilis sigma B and S2B). consensus sequence AGGTTT-N17-GGGTAT with a maximum of two mismatches and a maximum distance of 300 Transcriptome of wild-type L. monocytogenes P14, nucleotides upstream from the start codon of the gene as its isogenic DprfA and P14prfA* mutants in BHI criteria. These searches allowed us to identify a sequence and BHIC similar to a sigma B-dependent promoter upstream of 22 of the 63 newly identified genes. Six of these are the first Listeria monocytogenes strain EGDe has been chosen gene of putative co-transcribed units, accounting for 33 for transcriptome analysis, as the complete genome genes in total (Table 3). The primary data and the statis- sequence for strain EGDe is available. Moreover, EGDe tical analysis are available as Supplementary material is an intensively studied strain in numerous laboratories. (Tables S1A and S1B). However, virulence heterogeneity among L. monocytogenes strains is well documented (Brosch et al., 1993), DNA–DNA hybridization studies indicate considerable Expression profiles of wild-type L. monocytogenes genetic differences among different strains of L. monocy- EGDe compared with its isogenic prfA mutant in togenes (M. Doumith et al., unpublished data), and strain charcoal-supplemented BHI (BHIC) differences in the levels of virulence gene expression have Transcriptome analysis of the EGDe wild-type strain com- been reported (Ripio et al., 1996). To date, all known PrfA- pared with its isogenic mutant grown in BHI indicated that regulated genes are implicated in virulence. Analysis and three genes (inlB, inlC and hpt) previously shown to be comparison of the PrfA regulon of different Listeria iso-

Table 2. Sequence and position of putative PrfA boxes in the ﬁve new PrfA box-containing genes.

PrfA box

Gene Function Present in L. innocua Reg. +/- Distance from start codon Sequence lmo2219 Similar to PrsA from B. subtilis Ye s +-206 TTTACAcaTATTAA lmo0788 Unknown Yes +-79 TAAACAacTATTTA lmo0278 Similar to ATP-binding protein Yes – -30 TGAACAcaAGTTAA lmo0596 Unknown Yes +-106 TTAAAAggTTTTAA lmo 2067 Bile salt hydrolase No +-147 TTAAAAatTTTTAA

Small capital letters indicate mismatches compared with the consensus PrfA box (TTAACAnnTGTTAA). Gene names correspond to the L. monocytogenes EGDe gene names on the ListiList server http://genolist.pasteur.fr/ListiList/. Reg. +/-, regulated positively or negatively respectively. Numbers indicate the position from the start codon of the gene. lates might shed light on the molecular basis of virulence Gly145Ser, which ‘freezes’ the regulatory protein in its differences. In order to address this question, we investi- active conformation (Ripio et al., 1996). A prfA deleted gated a second L. monocytogenes strain, L. monocytoge- mutant of P14 (P14DprfA) was constructed by double nes P14, a serovar 4b clinical isolate from a human cross-over as described previously (Chico-Calero et al., listeriosis outbreak in Spain in 1989 (Vazquez-Boland 2002). et al., 1991). Like EGDe and all wild-type strains of L. Virulence gene expression was very low in wild-type monocytogenes, P14 shows weak haemolytic and lecithi- P14 after growth in BHI. After growth in BHIC, the viru- nase activities (Ripio et al., 1996; 1997b). The main rea- lence genes and one of the newly identiﬁed group I genes son for choosing this strain was the availability of a prfA* (lmo0778) were activated. However, the expression levels mutant (P14prfA*), expressing a constitutive PrfA* form of all other genes remained low (data not shown). We resulting from a single amino acid substitution in PrfA, therefore took advantage of strain P14prfA* and com-

Table 3. Putative sigma B promoter regions of group III genes.

Gene Description Distance from start codon Putative sigma B promoter sequence lmo0596 Unknown -103 GTTTTA-N13-GGCTAT lmo 2067 Similar to conjugated bile salt hydrolase -65 GTTTTA-N13-GGGTAC opuCA Similar to betaine/carnitine/choline/ABC transporter -83 GTTTAA-N14-GGGAAA opuCB Similar to betaine/carnitine/choline/ABC transporter opuCC Similar to betaine/carnitine/choline/ABC transporter opuCD Similar to betaine/carnitine/choline/ABC transporter lmo1602 Similar to general stress protein -53 GTTTTA-N14-GGGTAT lmo1601 Similar to general stress protein lmo2748 Similar to B. subtilis stress protein YdaG -58 GTTTGA-N14-TGGAAA lmo2230 Similar to arsenate reductase -144 GTTTCT-N13-GGGTAG lmo2231 Similar to cation efflux system lmo0913 Similar to succinate semi-aldehyde dehydrogenase -82 GATTAA-N13-TGGAAA lmo0669 Similar to oxidoreductase -175 GTTTTA-N13-GGGAAG lmo0670 Unknown lmo2695 Similar to dihydroxyacetone kinase -66 GTTTTG-N13-GGGAAA lmo2696 Similar to hypothetical dihydroxyacetone kinase lmo2697 Unknown lmo1694 Similar to CDP-abequose synthase -58 GTTTTA-N13-GGGAAT lmo0539 Similar to tagatose 1,6-diphosphate aldolase -86 GTTTTA-N14-TGGTAT lmo0784 Similar to mannose specific PTS component IIA -230 GTTTTC-N14-GGGTAA lmo0783 Similar to mannose specific PTS component IIB lmo0782 Similar to mannose specific PTS component IIC lmo0781 Similar to mannose specific PTS component IID lmo0602 Weakly similar to transcriptional regulator (PaiA) -37 GTTTCA-N13-GTGAAA lmo2391 Similar to B. subtilis YhfK protein -61 GTTTTA-N13-GGGAAA lmo0937 Unknown -69 GTTTAA-N13-GGGAAT lmo0994 Unknown -63 GTTTAT-N15-GGGAAT lmo0794 Unknown -80 GTTTCC-N14-GGGAAT lmo2213 Unknown -79 GTTTCA-N13-TGGAAA sepA Unknown -72 GTTTTG-N13-AGGTAT lmo1261 Unknown -67 GTTTAA-N14-GGGAAT lmo0439 Unknown -65 GTTTCA-N14-GGGAAA rsbV Anti-antisigma factor (antagonist of RsbW) -63 GTTTTA-N15-GGGTAA

© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 47, 1613–1625 1618 E. Milohanic et al. pared its expression profile with that of P14DprfA. Expres- cellobiose, in contrast to charcoal, abolished the regula- sion of prfA in strain P14prfA* after growth in BHI is tion of group I genes, except for prfA, but not that of ninefold higher than in the P14 wild-type strain, thus rep- groups II and III, further suggesting two different mecha- resenting a significant activation of virulence gene expres- nisms of regulation by PrfA. Analysis of the response of sion (Table 1). Only one (lmo0788) of the two newly strain P14prfA* relative to its DprfA counterpart showed identified group I genes of EGDe was activated in that the presence of cellobiose did not alter the activation P14prfA* relative to P14DprfA, and the negatively regu- of group I genes, but that all except 17 of the group III lated group II genes of strain EGDe were not repressed genes remained activated as observed for strain EGDe in P14prfA* relative to P14DprfA. (Table 1). Forty-six of the (Table 1). The primary data and the statistical analysis are 53 positively regulated group III genes identified in strain available as Supplementary material (Tables S5A, S5B, EGDe after growth in BHI corresponded to the positively S6A and S6B). regulated genes in strain P14prfA*. One hundred and five additional PrfA activated genes, specific to strain Confirmation of the macroarray results using real-time P14prfA*, were also identified (Table S4B). Growth of quantitative PCR strain P14prfA* in BHIC did not result in increased activation of group I genes, except for inlA, in agreement with Real-time quantitative PCR analysis of 18 genes, five previous results obtained for the hly and plc genes (Ripio representative of the group I genes (hly, actA, inlA, et al., 1996; 1997b). The negatively regulated group II lmo2219, lmo0788), three representative of the group II genes of strain EGDe remained uninduced by PrfA. Acti- genes (lmo0178, lmo0184, lmo0278) and 10 representa- vation of all but five (lmo0596, lmo2230, lmo2231, tive of the group III genes (lmo0596, lmo2067, opuCD, lmo0669, lmo0670) of the group III genes was abolished lmo1694, lmo2231, lmo0913, lmo0670, lmo2570, after growth in BHIC, as in strain EGDe. The primary data lmo2697, lmo0781), was conducted to verify the macroar- for the transcriptome experiments for L. monocytogenes ray transcription profiling data. For group III, one gene P14prfA* and P14DprfA are available as Supplementary representative of a predicted operon was chosen for material (Tables S3A, S3B, S4A and S4B). quantitative PCR. The average quantity of DNA molecules present in the EGDe wild-type strain grown at 37∞C in BHI relative to the EGDeDprfA strain and in P14prfA* relative Effect of cellobiose on PrfA-regulated genes to its DprfA counterpart grown in BHICel was determined. Several studies have shown that the presence and utiliza- Figure 1 shows that there was a very strong positive cor- tion of different carbohydrates have a remarkable impact relation (r = 0.82 for EGDe and r = 0.60 for P14) between on virulence gene expression in L. monocytogenes (for a the data obtained by the two different techniques. Primers review, see Kreft and Vazquez-Boland, 2001). For exam- used for the PCR are available as Supplementary material ple, metabolizable unphosphorylated sugars have been (Table S5). shown to inhibit the expression of PrfA-dependent virulence genes. However, when L. monocytogenes is grown Analysis of the differentially regulated genes in the presence of cellobiose, transcription of prfA is not inhibited, and the amount of PrfA protein is only slightly The majority of the newly identified genes code for trans- decreased (Renzoni et al., 1997). We were interested in port proteins, proteins involved in stress response and the impact of cellobiose on the expression of all the genes proteins of unknown function. Twenty genes encode pro- regulated by PrfA identified in our study. The four L. mono- teins constituting different transport systems. Sixteen of cytogenes strains EGDe, EGDeDprfA, P14prfA* and these are organized in three operons, two of which are P14DprfA were grown in BHI supplemented with 25 mM dedicated to carbohydrate transport. Operon lmo0784– cellobiose. Analysis of the expression profiles again lmo0781 codes for proteins of a phosphoenolpyruvate- revealed the existence of three groups of genes differently dependent phosphotransferase system (PTS) probably regulated by PrfA in the EGDe strain and two groups in devoted to mannose transport, and operon lmo178– the P14prfA* strain. As reported, prfA itself was still tran- lmo0184 constitutes a sugar ABC transport system scribed in both strains in the presence of cellobiose. In L. belonging to the OSP family of ABC systems specific for monocytogenes EGDe, all the previously known virulence di- and oligosaccharides and polyols (Dassa and Bouige, genes as well as the two newly identified (lmo2219 and 2001). lmo0278, coding for the only ABC ATPase present lmo0788) group I genes were not activated in the pres- in the genome of L. monocytogenes belonging to the OSP ence of cellobiose. The group II genes belonging to the family, is co-ordinately regulated with lmo178–lmo0184. sugar transport operon (lmo0178–0184, lmo278) were still The enzymes encoded by this operon are probably repressed. Surprisingly, all but nine of the group III genes involved in the transport of oligo alpha-1,6 or oligo alpha- remained activated by PrfA in BHICel (Table 1). Therefore, 1,4 saccharides (E. Dassa, personal communication). The

© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 47, 1613–1625 Analysis of the PrfA regulon of L. monocytogenes 1619 of the redox balance of the cell. Twenty-one genes encode proteins of unknown function (Table 1). Genes lmo2219 and lmo0178 were analysed further. Gene lmo2219 is preceded by a putative PrfA box (Table 2) and belongs to group I genes, PrfA activated in BHI and BHIC but not in the presence of cellobiose. Its deduced protein sequence shows high homology to the lipoprotein PrsA of B. subtilis (63% protein similarity over the entire length) and to the Lactococcus lactis PrtM lipoprotein (52% protein similarity over the entire length; Vos et al., 1989). In B. subtilis, PrsA is located at the outer side of the membrane and is important for the refolding of several mature proteins after their translocation through the membrane (Jacobs et al., 1993; Kontinen and Sarvas, 1993). We attempted to construct an in frame deletion mutation of lmo2219 in the chromo- some of the EGDe wild-type strain. However, several attempts to construct this mutant were unsuccessful. This might result from the fact that, as in B. subtilis (Kontinen and Sarvas, 1993), PrsA has an essential role in L. monocytogenes. Gene lmo0178 belongs to group II genes, which are repressed by PrfA in the wild-type EGDe. It is the first gene of a putative operon of seven genes, which are all repressed in EGDe. Its deduced amino acid sequence shows high sequence similarity to different xylose repressor proteins. The highest homology is to XylR of Anaero- cellum thermophilum (54% amino acid similarity over the entire length of the protein) and to XylR of B. subtilis (36% amino acid similarity over the entire length of the protein), where it acts as a transcriptional repressor of xylose-using Fig. 1. Correlation of macroarray and quantitative real-time PCR enzymes (Kreuzer et al., 1989). To investigate the effect results. The fold changes in transcripts present of 18 genes in the EGDe wild-type strain grown at 37∞C in BHI relative to the EGDeDprfA of xylose on gene lmo0178, we performed reverse tran- strain (A) and in the P14prfA* relative to its DprfA mutant grown in scription (RT)-PCR assays. Equal amounts of total RNA BHICel (B) were log transformed, and values were plotted against were extracted from EGDe and the EGDe prfA mutant each other to evaluate their correlation. The points on the graph D represent genes that were analysed by both methods. strain grown in BHI with and without 2% xylose. Expres- sion of ami, which is not regulated by PrfA but is expressed at a very high level in BHI, was used as a positive control. Fluorescent analysis of the PCR products third operon identified is the opuCA–opuCD operon, revealed a five- to 10-fold higher level of the xylR mRNA which has been shown to be an ATP-driven, osmoregu- in EGDeDprfA compared with EGDe, whatever the xylose lated glycine transport system implicated in the ability of concentration (data not shown). Thus, xylR is indeed L. monocytogenes to tolerate environments of elevated repressed by PrfA, and xylose is not the inducing sugar osmolarity and reduced temperature (Ko and Smith, of this operon, as is the case for B. subtilis (Kreuzer et al., 1999). Fifteen genes encode proteins probably involved 1989). in stress response. Among these are gene lmo1601 coding a protein 59% similar to the general stress protein Discussion YtxH from B. subtilis and lmo2748, which is 67% similar to the B. subtilis stress protein YdaG. Gene lmo2230 The primary goal of this study was to analyse the PrfA encodes a putative arsenate reductase, which might be regulon at the level of the complete genome. PrfA is the involved in detoxification. lmo0913 and lmo0669 encode master regulator of L. monocytogenes virulence genes a potential succinate semi-aldehyde dehydrogenase and (prfA, plcA, hly, mpl, actA, plcB, inlA, inlB, inlC and hpt), a hypothetical oxidoreductase respectively. One would which are all absent from L. innocua, the closest non- expect these enzymes to be involved in the maintenance pathogenic Listeria species. Thus, we reasoned that the

© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 47, 1613–1625 1620 E. Milohanic et al. study of gene expression of all L. monocytogenes genes Some of these may have important consequences for dependent on the presence of PrfA might identify novel global virulence. virulence genes or elucidate whether this regulon contains The major findings of our study are that PrfA can act as additional genes. This approach identified three groups of an activator and as a repressor and that PrfA regulates genes regulated differently by PrfA in strain L. monocyto- different groups of genes depending on the growth condi- genes EGDe. Group I comprises the 10 known virulence tions. PrfA is a DNA-binding protein that recognizes a so- genes and two newly identified genes (lmo0788 and called PrfA box to which the PrfA–RNA polymerase lmo2219). lmo0788 encodes a protein of unknown func- (RNAP) complex binds under appropriate conditions to tion, and lmo2219 codes for a protein similar to a post- initiate transcription (for a review, see Kreft and Vazquez- translocation molecular chaperone. In contrast to the Boland, 2001). We identified putative PrfA boxes known virulence genes, which are all absent from the L. upstream of all group I genes and one group II gene innocua genome, these genes are present in L. innocua. (lmo0278). The putative PrfA box identified for the ABC The group I genes are PrfA activated in BHI. Activation is ATPase lmo0278 is situated at position -30 from the start increased in BHIC and abolished in BHICel. Group II codon (Table 2), in agreement with a possible direct role comprises eight genes, seven (lmo0178–lmo0184) of for PrfA in repression. In contrast, a putative PrfA box is which are organized in an operon, constituting a predicted present upstream of only two group III genes, lmo0596, ABC transport system involved in sugar transport and encoding a protein of unknown function, and lmo2067, metabolism, and are repressed in the EGDe wild-type coding for a bile salt hydrolase recently shown to be strain relative to its prfA deleted mutant independently of involved in the intestinal and hepatic phases of listeriosis, the conditions tested (BHI, BHIC or BHICel). Group III and indeed PrfA regulated (Dussurget et al., 2002) comprises 53 genes, of which 44 are also PrfA activated (Table 2). This finding is puzzling as, up to now, genes in BHICel but not in BHIC, and only two are absent from shown to be regulated by PrfA contain a PrfA box. The L. innocua (lmo2067 and sepA) (Table 1). Quantitative absence of a PrfA box and other conserved motifs sug- PCR for a subset of genes representative of each of the gests indirect regulation of the group III genes by PrfA. three groups (after growth in BHI) showed a strong posi- Interestingly, the group III genes mainly encode proteins tive correlation with the macroarray data (Fig. 1A). involved in different stress responses, including a trans- Comparison of these results with the expression pro- port system involved in cold shock response and osmo- files of a second L. monocytogenes strain expressing a larity stress (opuCA–opuCD) (Ko and Smith, 1999), two constitutively active form of PrfA (P14prfA*) identified all proteins (lmo1602 and lmo1601) similar to general stress but one (lmo2219) of the group I genes and all but seven response proteins or proteins that might be involved in the of the group III genes compared with EGDe (Table 1). maintenance of the redox balance of the cell. Some homo- DNA–DNA hybridization of the EGDe macroarray with logues of these proteins in B. subtilis belong to the stress- genomic DNA from strain P14 indicated that the genes induced sigma B regulon of B. subtilis (Petersohn et al., expressed differently between these two strains are 2001; Price et al., 2001). indeed present in both isolates. Interestingly, the putative In bacteria, alternative sigma factors of RNAP are sugar transport operon lmo0178–lmo0184 and the OSP known to play a crucial role in regulating gene expression ABC ATPase lmo0278, which are repressed in EGDe, upon major changes in the environment. The alternative were not repressed in the three conditions tested. This sigma factor, sigma B, was shown to play an important might result from the replacement of the conserved ade- role in stress-induced gene regulation, and its targets are nine at position 13 of the consensus PrfA box sequence particularly well studied in B. subtilis, where sigma B is by a guanine in the P14prfA* strain, which might prevent activated when cells encounter growth-limiting energy and PrfA from acting as a repressor. Group I genes were PrfA environmental stresses (Akbar et al., 2001). Sigma B regulated in the P14prfA* strain constitutively overex- binds to a specific promoter sequence, which has been pressing PrfA, independently of the conditions studied, identified for many genes of the B. subtilis sigma B regu- thus confirming that PrfA in its active conformation lon. In B. subtilis, sigma B itself is under a complex regu- bypasses the repressor mechanism triggered by ferment- latory network (Hecker et al., 1996). In L. monocytogenes, able sugars (Ripio et al., 1997a; Brehm et al., 1999). The sigma B contributes to survival in conditions of oxidative quantitative PCR results for the P14 strain also showed a stress, starvation and reduced osmolarity (Ferreira et al., strong positive correlation with the macroarray data 2001) and enables L. monocytogenes in stationary phase (Fig. 1B).The comparative analysis of the expression pro- to adapt and to resume growth at reduced temperature files of two different L. monocytogenes strains reveals that (Becker et al., 2000). A search for a probable sigma B- differences in gene expression and gene regulation exist dependent promoter among all the genes identified in this among different isolates of the same bacterial species. study using a matrix constructed from 47 sigma B pro-

© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 47, 1613–1625 Analysis of the PrfA regulon of L. monocytogenes 1621 moter sequences of B. subtilis and a motif search using merase (Finnzymes), 0.6 mM each primer and 0.2 mM the sigma B consensus sequence of B. subtilis identified dNTPs (Perkin-Elmer). Reactions were cycled 35 times 22 promoter regions accounting for 33 genes in total, as (94∞C for 15 s; 55∞C for 30 s; 72∞C for 60 s) with one final cycle of 72 C for 7 min in a thermocycler. Successful ampli- some are organized in operons (Table 3). Genes pre- ∞ fication of each PCR product was verified on agarose gels, ceded by a putative sigma B promoter belonged to group and negative PCRs were repeated, resulting in the final III and were also included. In contrast, none of the group amplification of 2816 ORFs of the 2853 ORFs (99%) identi- I or group II genes displayed a sigma B-like promoter, fied in the L. monocytogenes EGDe genome. For macroarray except prfA itself. PrfA can be transcribed from two differ- preparation, nylon membranes (Qfilter, Genetix) were soaked ent promoters. The P1 and P2 promoters of prfA are not in 10 mM TE, pH 7.6. Spot blots of ORF-specific PCR prod- transcribed in the same way. Depending on the conditions, ucts and controls were printed using a Qpix robot (Genetix). Immediately after spot deposition, membranes were neutral- prfA has been shown to be transcribed from either P1prfA ized for 15 min in 0.5 M NaOH, 1.5 M NaCl, washed three or P2prfA resulting in either a 0.9 or a 0.8 kb transcript times with distilled water and stored wet at -20∞C until use. (Freitag et al., 1993; Renzoni et al., 1997). Recent genetic evidence suggests that P2prfA is a sigma B-dependent promoter (Nadon et al., 2002). This situation is reminis- Validation of the macroarrays cent of that of the global virulence gene regulator SarA in To ensure that DNA samples were successfully deposited on Staphylococcus aureus, the transcription of which is at the nylon membranes and to assess differential hybridization least partially controlled by sigma B (Deora et al., 1997). to the target genes 33P-labelled genomic DNA from L. mono- It can be transcribed from three different promoters, of cytogenes EGDe was hybridized before transcriptome anal- which P1 and P2 have been reported to be sigma A ysis to the macroarray. The total intensity of all pixels within dependent and P3 sigma B dependent (Deora et al., each spot was determined after scanning of exposed phos- 1997). Our results strongly suggest an interplay between phor screens. Fifty-six spots scored as undetectable and were therefore not taken into account in our study. In order the PrfA regulon and the sigma B regulon. It will be inter- to establish optimal experimental conditions, RNA from the esting to decipher at what stage of the infectious process EGDe wild-type strain and its prfA-deleted mutant grown in the PrfA-regulated genes are expressed. Whether other BHI or BHIC was tested. To determine the best labelling genes are regulated by PrfA in other growth conditions method, random hexamers (Boehringer Mannheim) or a mix- also deserves further investigation. ture of all 3¢ oligonucleotides (Metabion), each at a concentration of 0033 mM, specific for each amplified gene, were used. Similar results were obtained using the two different Experimental procedures labelling methods, suggesting that the observed distribution reflected the true distribution of transcripts (data not shown). Strains and growth conditions For all further experiments, a mixture of specific oligonucle- Listeria monocytogenes EGDe and its derivative DprfA42 otides was chosen. The intensity distribution patterns showed (Bockmann et al., 1996), L. monocytogenes P14, P14prfA* good reproducibility (r = 0.959) between independent exper- (also named P14-A in other studies) (Ripio et al., 1996) and iments. In contrast, the correlation between the intensity of P14DprfA (this study) were used. For all experiments, a single spots generated with cDNAs from wild-type cells and those colony of each Listeria strain was grown in BHI (Difco), BHI generated with cDNAs from prfA mutant cells showed that supplemented with 0.2% activated charcoal (BHIC; Merck) or the mutation caused significant changes in the expression BHI supplemented with 25 mM cellobiose (BHICel; Sigma) profile (r = 0.905). overnight at 37∞C with shaking. Sample collection, cell lysis and total RNA isolation Primer design, amplification of PCR products and construction of whole-genome macroarrays BHI, BHIC or BHICel was inoculated with 100 ml of the overnight culture and incubated at 37∞C with shaking. Exponen- Pairs of specific oligonucleotide primers were designed for tially growing cells (OD600 0.6–0.8) were harvested for 2 min each of the 2853 ORFs of the L. monocytogenes EGDe at 6000 g at 4∞C. The pellets were flash frozen on dry ice– genome. For primer design, a modified version of primer 3, ethanol and stored at -80∞C. For extraction, cells were resus- which allows testing of the specificity of the obtained PCR pended by vortexing in 400 ml of resuspension buffer products against the complete genome sequence to avoid (12.5 mM Tris, 5 mM EDTA and 10% glucose). Then, 500 ml cross-hybridization, was used (CAAT-Box; L. Frangeul et al., of acid phenol (pH 4.6) and 0.4 g of glass beads (0.2–0.3 mm unpublished). Primers (Metabion) were chosen in order to diameter; Sigma) were added. The cells were sheared amplify a fragment of 300–600 bp specific for each ORF with mechanically using a Fastprep apparatus (Bio101). After cen- a melting temperature of 55–65∞C. Amplification reactions trifugation at 13 000 g for 5 min, the supernatant was trans- were performed in 96-well plates (Perkin-Elmer) in a 100 ml ferred to a fresh tube, and 1 ml of Trizol reagent (Gibco BRL) reaction volume containing 20 ng of chromosomal DNA from was added. The sample was incubated for 5 min at room L. monocytogenes EGDe, 2 U of DyNazyme EXT DNA poly- temperature. Total RNA was extracted twice with chloroform–

© 2003 Blackwell Publishing Ltd, Molecular Microbiology, 47, 1613–1625 1622 E. Milohanic et al. isoamyl alcohol (24:1, v/v) and precipitated in 0.7 volumes of SAM/). We chose a delta value corresponding to a false isopropanol. After a washing step with 70% ethanol, the RNA discovery rate (FDR) <10%. The FDR was computed as the pellet was dissolved in sterile DNase- and RNase-free water ratio of the estimated number of ‘false significant’ genes to (ICN Biomedicals) and quantified by absorbance at 260 and the total number of ‘significant’ genes. Genes with a twofold 280 nm. Purity and integrity of RNA were controlled on aga- expression change that were significant according to this rose gels, and RNA was stored at -80∞C until use. analysis were taken into account. The complete data set of the statistical analysis is available online as Supplementary material. Genomic DNA labelling and cDNA labelling

Chromosomal DNA (30 ng) isolated from L. monocytogenes Real-time quantitative PCR strains EGDe or P14 was 33P-labelled using a random-primed DNA labelling kit (Boehringer Mannheim). Labelled genomic Real-time quantitative PCR was conducted on the same total DNAs were purified before hybridization using QIAquick col- cellular RNAs as those used for transcriptome experiments. umns (Qiagen). Primers (Table S7) were designed with PRIMEREXPRESS soft- For cDNA synthesis, random hexamers or a set of 3¢- ware (Applied Biosystems) and purchased from Eurogentec. specific oligonucleotide primers were used in reverse RNA (20 mg) was treated with 50 U of RNase-free DNase I transcription reactions in the presence of [a-33P]-dCTP (Roche) for 30 min at 37∞C. DNase was inactivated by phenol (2000–3000 Ci mmol-1; Amersham). Total RNA (1 mg) and chloroform extraction, and RNA was precipitated. cDNA syn- 50 U of AMV reverse transcriptase (Roche) with RNase H thesis was performed for 60 min at 42∞C using 10 mg of RNA, 1 activity were used. Labelled cDNA was purified to remove 2 ml of AMV reverse transcriptase (25 U ml- ; Roche) and 2 ml unincorporated nucleotides before hybridization using a of the primer mix. To determine the primer concentration with QIAquick column (Qiagen). the best efficiency, concentrations of 100 nM, 200 nM and 400 nM in a 25 ml reaction volume were tested for each primer. The optimal concentration of 200 nM was used in the Hybridization final experiments. Real-time quantitative PCR was performed in a 25 ml reaction volume containing 200 ng of cDNA, 12.5 ml Hybridization and washing steps were carried out using of SYBR PCR master mix (Applied Biosystems) and 2 ml of SSPE buffer (0.18 M NaCl, 10 mM NaH2PO4, 1 mM EDTA, gene-specific primers (200 nM). Amplification and detection pH 7.7). Macroarrays were prewet in 2¥ SSC and prehybrid- of specific products were performed with the ABI Prism 7700 ized for at least 2 h in 10 ml of hybridization solution (5¥ sequence detection system (PE Applied Biosystems) with the SSPE, 2% SDS, 1 Denhardt’s reagent, 100 g of sheared ¥ m following cycle profile: one cycle at 50∞C for 2 min, one cycle salmon sperm DNA ml-1) at 65 C in roller bottles. Hybridiza- ∞ at 95∞C for 10 min, 40 cycles at 95∞C for 15 s and 60∞C for tion was carried out for 20 h at 65∞C with 5 ml of hybridization 1 min. Each product was deposited on agarose gels to verify solution and the entire spin-purified cDNA probe. After the presence of a single band. The quantity of cDNA for each hybridization, membranes were washed twice at room tem- experimental gene was normalized to the quantity of rpoB perature and twice at 65∞C in 0.5¥ SSC, 0.2% SDS. Arrays cDNA in each sample. For each gene, triplicate assays were were then sealed in thin bags and exposed to a phosphor done. To check whether contaminating chromosomal DNA screen (Molecular Dynamics) for 24–72 h. For each strain, was present, each sample was tested in control reactions that two independent RNA preparations were tested, and two did not contain reverse transcriptase. cDNAs from each of the RNA preparations were hybridized to two sets of arrays and analysed. Acknowledgements Data analysis This work received financial support from the European Membranes were scanned using a 445SI PhosphorImager Commission contract QLG2-CT-1999-00932, the Spanish (Molecular Dynamics). The ARRAYVISION software (Imaging Ministerio de Ciencia y Technología (BMC-2000-0553), the Research) was used for quantification of the hybridization Ministere de l’Education Nationale de la Recherche et de la intensities and for normalization. The intensity of each spot Technologie (PFRMMIP) and the Pasteur Institut (PTR 6). was normalized according to the median value of the total E.M. holds a post-doctoral fellowship from SOREDAB SAS. intensities of all spots on each array, which allowed direct F. Chetouani we thank for access to unpublished results. We comparison of the two strains. The global background was thank the following persons for help in preparing macroar- calculated from the average intensity of 610 ‘no-DNA’ spots rays: E. Zalachas, C Lacroix, A. Schiavo and E. Couvé. homogeneously distributed throughout the membrane. Only Special thanks to A. Bourdet, who took time to help us with spots for which the sum of the normalized intensity values of the quantitative PCR experiments, Tim Stinear, Claire Poyart the wild type and the mutant strain was higher than the sum and Chester Price for critical reading of the manuscript, and of the global background of the wild type and the mutant A. Maitournam and M. Zouine for help with statistical analy- membrane were taken into account. sis. We also thank the reviewers for their constructive com- For identification of genes with statistically significant ments. P Cossart is an international investigator from the changes in expression SAM, a statistical technique for finding Howard Hughes Medical Institute. significant genes in a set of microarray experiments was used (Tusher et al., 2001) (http://www-stat.stanford.edu/~tibs/

Supplementary material sequences requires additional factor(s) and is influenced by iron. Mol Microbiol 22: 643–653. The following material is available from http://www.blackwell Bohne, J., Sokolovic, Z., and Goebel, W. (1994) Transcrip- publishing.com/products/journals/suppmat/mole/mole3413/ tional regulation of prfA and PrfA-regulated virulence mmi3413sm.htm genes in Listeria monocytogenes. Mol Microbiol 11: 1141– Table S1A. Primary expression data obtained for wild-type L. 1150. monocytogenes EGDe and its isogenic prfA deleted mutant Brehm, K., Kreft, J., Ripio, M.T., and Vazquez-Boland, J.A. after growth in BHI. (1996) Regulation of virulence gene expression in patho- Table S1B. Genes whose expression was significantly genic Listeria. Microbiologia 12: 219–236. different between L. monocytogenes EGDe and EGD- Brehm, K., Ripio, M.T., Kreft, J., and Vazquez-Boland, J.A. eDprfA after growth in BHI. (1999) The bvr locus of Listeria monocytogenes mediates Table S2A. Primary expression data obtained for wild-type L. virulence gene repression by beta-glucosides. J Bacteriol monocytogenes EGDe and its isogenic prfA deleted mutant 181: 5024–5032. after growth in BHIC. Brosch, R., Catimel, B., Milon, G., Buchrieser, C., Vindel, E., Table S2B. Genes whose expression was significantly and Rocourt, J. (1993) Virulence heterogeneity of Listeria different between L. monocytogenes EGDe and EGD- monocytogenes strains from various sources (food, eDprfA after growth in BHIC. human, animal) in immunocompetent mice and its associ- Table S3A. Primary expression data obtained for wild-type L. ation with typing characteristics. J Food Prot 56: 296– monocytogenes P14prfA* and its isogenic prfA deleted 301. mutant after growth in BHI. Chico-Calero, I., Suarez, M., Gonzalez-Zorn, B., Scortti, M., Table S3B. Genes whose expression was significantly Slaghuis, J., Goebel, W., and Vazquez-Boland, J.A. (2002) different between L. monocytogenes P14prfA* and P14DprfA Hpt, a bacterial homolog of the microsomal glucose-6- after growth in BHI. phosphate translocase, mediates rapid intracellular prolif- Table S4A. Primary expression data obtained for wild-type L. eration in Listeria. Proc Natl Acad Sci USA 99: 431–436. monocytogenes P14prfA* and its isogenic prfA deleted Cossart, P., and Lecuit, M. (1998) Interactions of Listeria mutant after growth in BHIC. monocytogenes with mammalian cells during entry and Table S4B. Genes whose expression was significantly actin-based movement: bacterial factors, cellular ligands different between L. monocytogenes P14prfA* and and signaling. EMBO J 17: 3797–3806. P14DprfA after growth in BHIC. Cruz Ramos, H., Boursier, L., Moszer, I., Kunst, F., Danchin, Table S5A. Primary expression data obtained for wild-type L. A., and Glaser, P. (1995) Anaerobic transcription activation monocytogenes EGDe and its isogenic prfA deleted mutant in Bacillus subtilis: identification of distinct FNR-dependent after growth in BHICel. and -independent regulatory mechanisms. EMBO J 14 Table S5B. Genes whose expression was significantly differ- (23): 5984–5994. ent between L. monocytogenes EGDe and EGDeDprfA Dassa, E., and Bouige, P. (2001) The ABC of ABCS: a after growth in BHICel. phylogenetic and functional classification of ABC systems Table S6A. Primary expression data obtained for wild-type L. in living organisms. Res Microbiol 152: 211–229. monocytogenes P14prfA* and its isogenic prfA deleted Deora, R., Tseng, T., and Misra, T.K. (1997) Alternative mutant after growth in BHICel. transcription factor sigmaSB of Staphylococcus aureus: Table S6B. Genes whose expression was significantly differ- characterization and role in transcription of the global ent between L. monocytogenes P14prfA* and P14DprfA after regulatory locus sar. J Bacteriol 179: 6355–6359. growth in BHICel. Dramsi, S., Lebrun, M., and Cossart, P. (1996) Molecular and Table S7. Genes and corresponding primers used for genetic determinants involved in invasion of mammalian quantitative PCR. cells by Listeria monocytogenes. Curr Topics Microbiol Immunol 209: 61–77. References Dussurget, O., Cabanes, D., Dehoux, P., Lecuit, M., Consortium, E.L.G., Buchrieser, C., et al. (2002) Listeria Akbar, S., Gaidenko, T.A., Kang, C.M., O’Reilly, M., Devine, monocytogenes bile salt hydrolase is a virulence factor K.M., and Price, C.W. (2001) New family of regulators in involved in the intestinal and hepatic phase of listeriosis. the environmental signaling pathway which activates the Mol Microbiol 45: 1095–1106. general transcription factor sigma (B) of Bacillus subtilis. J Engelbrecht, F., Chun, S.-K., Ochs, C., Hess, J., Lottspeich, Bacteriol 183: 1329–1338. F., Goebel, W., and Sokolovic, Z. (1996) A new PrfA- Bailey, T.L., and Elkan, C. (1994) Fitting a mixture model by regulated gene of Listeria monocytogenes encoding a expectation maximization to discover motifs in biopoly- small, secreted protein which belongs to the family of inter- mers. Proc Int Conf Intell Syst Mol Biol 2: 28–36. nalins. Mol Microbiol 21: 823–837. Becker, L.A., Evans, S.N., Hutkins, R.W., and Benson, A.K. Ferreira, A., O’Byrne, C.P., and Boor, K.J. (2001) Role of (2000) Role of sigma (B) in adaptation of Listeria monocy- sigma (B) in heat, ethanol, acid, and oxidative stress resis- togenes to growth at low temperature. J Bacteriol 182: tance and during carbon starvation in Listeria monocyto- 7083–7087. genes. Appl Environ Microbiol 67: 4454–4457. Bockmann, R., Bickneite, C., Middendorf, B., Goebel, W., Freitag, N.E., Rong, L., and Portnoy, D.A. (1993) Regulation and Sokolovic, Z. (1996) Specific binding of the Listeria of the prfA transcriptional activator of Listeria monocytoge- monocytogenes transcriptional regulator PrfA to target nes: multiple promoter elements contribute to intracellular

growth and cell-to-cell spread. Infect Immun 61: 2537– that may be involved in regulation. Infect Immun 57: 3695– 2544. 3701. Geoffroy, C., Gaillard, J.L., Alouf, J., and Berche, P. (1989) Mengaud, J., Dramsi, S., Gouin, E., Vazquez-Boland, J.A., Production of thiol-dependent hemolysins by Listeria Milon, G., and Cossart, P. (1991) Pleiotropic control of monocytogenes and related species. J Gen Microbiol 135: Listeria monocytogenes virulence factors by a gene which 481–487. is autoregulated. Mol Microbiol 5: 2273–2283. Geoffroy, C., Raveneau, J., Beretti, J.L., Lecroisey, A., Milenbachs, A., Brown, D.P., Moors, M., and Youngman, P. Vazquez-Boland, J.A., Alouf, J.E., and Berche, P. (1991) (1997) Carbon-source regulation of virulence gene expres- Purification and characterization of an extracellular 29- sion in Listeria monocytogenes. Mol Microbiol 23: 1075– kilodalton phospholipase C from Listeria monocytogenes. 1085. Infect Immun 59: 2382–2388. Nadon, C.A., Bowen, B.M., Wiedmann, M., and Boor, Glaser, P., Frangeul, L., Buchrieser, C., Rusniok, C., Amend, K.J. (2002) Sigma B Contributes to PrfA-mediated viru- A., Baquero, F., et al. (2001) Comparative genomics of lence in Listeria monocytogenes. Infect Immun 70: 3948– Listeria species. Science 294: 849–852. 3952. Hecker, M., Schumann, W., and Volker, U. (1996) Heat-shock Park, S.F., and Kroll, R.G. (1993) Expression of listeriolysin and general stress response in Bacillus subtilis. Mol Micro- and phosphatidylinositol-specific phospholipase C is biol 19: 417–428. repressed by the plant-derived molecule cellobiose in List- Irvine, A.S., and Guest, J.R. (1993) Lactobacillus casei con- eria monocytogenes. Mol Microbiol 8: 653–661. tains a member of the CRP-FNR family. Nucleic Acids Res Petersohn, A., Brigulla, M., Haas, S., Hoheisel, J.D., Volker, 21: 753. U., and Hecker, M. (2001) Global analysis of the general Jacobs, M., Andersen, J.B., Kontinen, V., and Sarvas, M. stress response of Bacillus subtilis. J Bacteriol 183: 5617– (1993) Bacillus subtilis PrsA is required in vivo as an extra- 5631. cytoplasmic chaperone for secretion of active enzymes Price, C.W., Fawcett, P., Céremonie, H., Su, N., Murphy, synthesized either with or without pro-sequences. Mol C.K., and Youngman, P. (2001) Genome-wide analysis of Microbiol 1993: 957–966. the general stress response in Bacillus subtilis. Mol Micro- Johansson, J., Mandin, P., Renzoni, A., Chiaruttini, C., biol 41: 757–774. Springer, M., and Cossart, P. (2002) An RNA thermosen- Renzoni, A., Klarsfeld, A., Dramsi, S., and Cossart, P. (1997) sor controls expression of virulence genes in Listeria Evidence that PrfA, the pleiotropic activator of virulence monocytogenes. Cell 110: 551–561. genes in Listeria monocytogenes, can be present but inac- Ko, R., and Smith, L.T. (1999) Identification of an ATP-driven, tive. Infect Immun 65: 1515–1518. osmoregulated glycine betaine transport system in Ripio, M.T., Dominguez-Bernal, G., Suarez, M., Brehm, K., Listeria monocytogenes. Appl Environ Microbiol 65: 4040– Berche, P., and Vazquez-Boland, J. (1996) Transcriptional 4048. activation of virulence genes in wild type strains of Listeria Kontinen, V.P., and Sarvas, M. (1993) The PrsA lipoprotein monocytogenes. response to a change in extracellular is essential for protein secretion in Bacillus subtilis and sets medium composition. Res Microbiol 147: 371–384. a limit for high-level secretion. Mol Microbiol 4: 727–737. Ripio, M., Brehm, K., Lara, M., Suarez, M., and Vazquez- Kontinen, V.P., Saris, P., and Sarvas, M. (1991) A gene Boland, J.A. (1997a) Glucose-1-phosphate utilization by (prsA) of Bacillus subtilis involved in a novel, late stage of Listeria monocytogenes is PrfA dependent and coordi- protein export. Mol Microbiol 5: 1273–1283. nately expressed with virulence factors. J Bacteriol 179: Kreft, J., and Vazquez-Boland, J.A. (2001) Regulation of vir- 7174–7180. ulence genes in Listeria. Int J Med Microbiol 291: 849–852. Ripio, M.T., Dominguez-Bernal, G., Lara, M., Suarez, M., and Kreuzer, P., Gartner, D., Allmansberger, R., and Hillen, W. Vazquez-Boland, J.A. (1997b) A Gly145Ser substitution in (1989) Identification and sequence analysis of the Bacillus the transcriptional activator PrfA causes constitutive over- subtilis W23 xylR gene and xyl operator. J Bacteriol 1: expression of virulence factors in Listeria monocytogenes. 3840–3845. J Bacteriol 179: 1533–1540. Lampidis, R., Gross, R., Sokolovic, Z., Goebel, W., and Kreft, Sheehan, B., Klarsfeld, A., Ebright, R., and Cossart, P. J. (1994) The virulence regulator protein of Listeria ivanovii (1996) A single substitution in the putative helix-turn-helix is highly homologous to PrfA from Listeria monocytogenes motif of the pleiotropic activator PrfA attenuates Listeria and both belong to the Crp-Fnr family of transcription reg- monocytogenes virulence. Mol Microbiol 20: 785–797. ulators. Mol Microbiol 13: 141–151. Tusher, V., Tibshirani, R., and Chu, C. (2001) Significance Lingnau, A., Chakraborty, T., Niebuhr, K., Domann, E., and analysis of microarrays applied to ionizing radiation Wehland, J. (1996) Identification and purification of novel response. Proc Natl Acad Sci USA 98: 5116–5121. internalin-related proteins in Listeria monocytogenes and Vazquez-Boland, J.A., Ferrer, D., and Rocourt, J. (1991) Listeria ivanovii. Infect Immun 64: 1002–1006. Heterogeneity of strains of Listeria monocytogenes iso- Liu, X., Brutlag, D.L., and Liu, J.S. (2001) BioProspector: lated during an outbreak of listeriosis among adults in discovering conserved DNA motifs in upstream regulatory Valencia in 1989. Enferm Infect Microbiol Clin 7: 442– regions of co-expressed genes. Pac Symp Biocomput 444. 127–138. Vazquez-Boland, J.-A., Kuhn, M., Berche, P., Chakraborty, Mengaud, J., Vicente, M.F., and Cossart, P. (1989) Tran- T., Dominguez-Bernal, G., Goebel, W., et al. (2001) Liste- scriptional mapping and nucleotide sequence of the Liste- ria pathogenesis and molecular virulence determinants. ria monocytogenes hlyA region reveal structural features Clin Microbiol Rev 14: 1–57.

Vega, Y., Dickneite, C., Ripio, M.T., Bockmann, R., Vos, P., van Asseldonk, M., van Jeveren, F., Siezen, R., Gonzalez-Zorn, B., Novella, S., et al. (1998) Functional Simons, G., and de Vos, W.M. (1989) A maturation pro- similarities between the Listeria monocytogenes virulence tein is essential for production of active forms of Lacto- regulator PrfA and cyclic AMP receptor protein: the PrfA* coccus lactis SK11 serine proteinase located in or (Gly145Ser) mutation increases binding afﬁnity for target secreted from the cell envelope. J Bacteriol 171: 2795– DNA. J Bacteriol 180: 6655–6660. 2802.