Studies on the production of quorum-sensing signal molecules in Mannheimia haemolytica A1 and other Pasteurellaceae species

Transcription

1 FEMS Microbiology Letters 206 (2002) 25^30 Studies on the production of quorum-sensing signal molecules in Mannheimia haemolytica A1 and other Pasteurellaceae species Rebecca J. Malott, Reggie Y.C. Lo * Department of Microbiology, University of Guelph, Guelph, ON, Canada, N1G 2W1 Received 6 August 2001; received in revised form 30 September 2001; accepted 19 October 2001 First published online 26 November 2001 Abstract The bioluminescence assay system using Vibrio harveyi reporter strains were used to examine quorum-sensing autoinducer (AI) activity from Mannheimia haemolytica A1 cell-free culture supernatant. We showed that M. haemolytica A1 cell-free culture supernatant contains molecules that can stimulate the quorum-sensing system that regulates the expression of the luciferase operon in V. harveyi. Specifically, M. haemolytica A1 can stimulate only the quorum system 2 but not system 1, suggesting that the culture supernatant only contains molecules similar to AI-2 of V. harveyi. The bioluminescence assay was also used to show that culture supernatants from related Pasteurellaceae organisms, Pasteurella multocida, Pasteurella trehalosi, Actinobacillus suis and Actinobacillus pleuropneumoniae, also contain AI-2-like molecules. This is consistent with the presence of a luxs homolog in the genomes of P. multocida and A. pleuropneumoniae. AluxS homolog was cloned by PCR from M. haemolytica A1 using sequencing data from the ongoing genome sequencing project. The cloned luxs M:h: was able to complement AI-2 production in the Escherichia coli DH5K luxs mutant. This is the first report of a quorum-sensing activity in M. haemolytica A1 and suggests that this bacterium utilizes this mechanism to regulate expression of genes under specific conditions. ß 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords: Pasteurellaceae, quorum-sensing system, autoinducer 2 activity, bioluminescence assay 1. Introduction Bacteria depend on numerous forms and levels of genetic mechanisms to regulate the dynamic expression of the genes. One such form of genetic regulation is quorum sensing, where the bacteria requires a minimal population or `quorum' in its environment to in uence the expression of speci c groups of genes. To accomplish this task, the bacteria synthesize, release and respond to a speci c extracellular substance referred to as the autoinducer (AI). Current research indicates that there are various types of quorum-sensing systems employed by the microbial population for density-dependent genetic regulation [1]. The density-dependent expression of luminescence in Vibrio harveyi is regulated by two quorum-sensing systems [2]. System 1 has been reported to be the more sensitive and speci c of the two and is hypothesized to be involved * Corresponding author. Tel: +1 (519) , ext. 3363; Fax: +1 (519) address: rlo@micro.uoguelph.ca (R.Y.C. Lo). in intraspecies communication [3]. System 2 has been reported to be less sensitive and less speci c and is thought to be involved in interspecies communication [3]. The V. harveyi interspecies quorum-sensing system (system 2) has been shown to respond to the cellular density of foreign populations including Salmonella typhimurium, Escherichia coli and Helicobacter pylori to regulate the expression of its luciferase (lux) operon, resulting in the production of bioluminescence [4,5]. One of the key components in quorum-sensing system 2 of V. harveyi is the LuxS protein, which has been designated as the AI-2 synthase. Recently, it has been shown that LuxS is associated with a metabolic pathway involved in the conversion of S-adenosylmethionine into a furanone derivative [6]. S-ribosylhomocysteine is cleaved by LuxS to produce homocysteine and 4,5-dihydroxy-2,3-pentanedione. The latter molecule spontaneously cyclizes into 4-hydroxy-5-methyl-furanone or other derivatives in V. harveyi [6]. A similar pathway is present in organisms that exhibit the system 2 quorum-sensing system such as E. coli, S. typhimurium and V. cholerae [6]. A BLAST search analysis using V. harveyi LuxS revealed the presence of homologs in the genomes of Gramnegative and Gram-positive bacteria including the afore / 02 / $22.00 ß 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S (01)

2 26 R.J. Malott, R.Y.C. Lo / FEMS Microbiology Letters 206 (2002) 25^30 mentioned species. Some notables included were Haemophilus in uenzae and Pasteurella multocida with a sequence similarity of 88 and 86 percent respectively. Both of these bacteria are members of the family Pasteurellaceae. Mannheimia haemolytica is also a member of the Pasteurellaceae family. This Gram-negative bacterium is commonly found as a commensal organism on the mucous membranes of the upper respiratory tracts of healthy cattle, sheep and goats. However, when the animal is under stress, M. haemolytica serotype A1 may become the principal microorganism involved with the development of pneumonic pasteurellosis in cattle, and pneumonia and septicemia in sheep [7]. With the close phylogenetic relationship between H. in- uenzeae, P. multocida and M. haemolytica, it is likely that M. haemolytica also possesses a luxs homolog, and thus, the ability to produce a signaling molecule that could induce the luminescence activity of V. harveyi. Since the regulation of virulence genes in M. haemolytica is of importance to its pathogenesis, the identi cation and characterization of a possible quorum-sensing system may further the understanding of its role in bovine pneumonic pasteurellosis. 2. Materials and methods 2.1. Bacteria and culture conditions The bacterial strains used in this study are listed in Table 1. M. haemolytica A1, P. multocida USDA, P. trehalosi T3, Actinobacillus pleuropneumoniae 05 and Actinobacillus suis were stored as glycerol stocks at 380³C and grown at 37³C on blood (sheep) agar plates or in brain heart infusion (BHI) broth. A. pleuropneumoniae was grown in BHI supplemented with 0.02% NAD. V. harveyi strains were grown at 30³C on Luria-marine (LM) agar plates [1] or in AI bioassay (AB) medium [9]. E. coli DH5K was grown in Luria^Bertani broth supplemented with 1 mg l 31 of dextrose and 50 mg l 31 of thymine (LT) at 37³C or on solid LT agar plates. The LT media was supplemented with 100 mg l 31 of ampicillin when necessary Enzymes and chemicals The DNA restriction endonucleases and DNA modifying enzymes used in these experiments were purchased from Amersham Pharmacia Biotech (Baie d'urfe, QC, Canada) or Canadian Life Technologies (Burlington, ON, Canada). PLATINUM Taq DNA polymerase was also purchased from Canadian Life Technologies Preparation of cell-free culture supernatants The test culture cell-free supernatants were prepared essentially as described [4]. Brie y, a saturated M. haemolytica A1 over-night culture was diluted 1:100 into 50 ml of fresh BHI broth, grown at 37³C, and the cell density monitored by optical density at 600 nm. A 10-ml sample was removed at timed intervals. The sample was centrifuged at 3000Ug for 10 min, the supernatant was recovered, passed through a 0.2-Wm pore size lter (Nalgene, Rochester, NY, USA) and stored at 320³C until use. The V. harveyi and E. coli culture supernatants were prepared in the same manner except the bacteria were grown in AB and LT media, respectively Assay for the production of signaling molecules The procedure for the V. harveyi bioluminescence assay was adopted from Surette and Bassler [4,10]. The V. harveyi reporter strain was grown overnight in AB medium and diluted 1:5000 into 18 ml of fresh AB medium. Two milliliters of the test supernatant was added to the diluted culture resulting in a nal concentration of 10% (v/v). The assay culture was grown at 30³C with aeration for 7 h. At hourly intervals, both the production of luminescence and cell density were measured by sampling 400 Wl of the culture and kept on ice. Three 100-Wl aliquots were added to three separate microtiter plate wells. The microtiter plates were white, at bottomed, 96-well plates supplied by Dymex Technologies Inc. (Franklin, MA, USA). The samples were applied to every other well to minimize luminescent interference between wells. The microtiter plate was placed into a Wallac Model 1420 liquid scintillation counter to record luminescence (Wallac, Gaithersburg, MD, USA). Table 1 Bacterial strains used in this study Strains Relevent genotype/phenotype Source/reference M. haemolytica A1 clinical isolate Dr. P.E. Shewen, University of Guelph, Canada P. trehalosi T3 clinical isolate Dr. P.E. Shewen, University of Guelph, Canada P. multocida USDA clinical isolate Dr. P.E. Shewen, University of Guelph, Canada A. pleuropneumoniae 05 clinical isolate Dr. J. MacInnes, University of Guelph, Canada A. suis clinical isolate Dr. S. Rosendal, University of Guelph, Canada V. harveyi BB152 luxm: :Tn5 (AI-1 3,2 ) Dr. B.L. Bassler, Princeton University, USA V. harveyi BB170 luxn: :Tn5 (sensor 1 3,2 ) Dr. B.L. Bassler, Princeton University, USA V. harveyi BB886 luxq: :Tn5 (sensor 1,2 3 ) Dr. B.L. Bassler, Princeton University, USA V. harveyi BB120 wild-type (1,2 ) Dr. B.L. Bassler, Princeton University, USA E. coli DH5K DH1, vlacu169 (x80laczvm15) [8]

3 R.J. Malott, R.Y.C. Lo / FEMS Microbiology Letters 206 (2002) 25^30 27 The cell density was measured by making serial dilutions with the remaining 100 Wl of the sample. Duplicates of each dilution were spread onto solid LM medium and the plates were incubated at 30³C overnight. The resulting colonies were counted the following morning. The relative light units (RLUs) for each 100-Wl aliquot at each time point were calculated as follows: (light counts per second3background) 60 s 0.1 ml/(colony forming units ml 31 ), resulting in cpm cell 31. The cpm cell 31 for each 100-Wl aliquot were averaged and the standard deviation was calculated. The RLU were graphed as a function of time over the assay time course Cloning of luxs from M. haemolytica A1 PCR primers were designed based on the nucleotide sequence of the ongoing genome sequencing project of M. haemolytica A1 being carried out at Baylor College of Medicine Human Genome Sequencing Center (BCM- HGSC; The sequencing project was supported by an USDA grant to Dr. S. Highlander at BCM-HGSC. The primers were forward, 5P-TGTTCTTGCAGGTACCGATA and reverse, 5P-AAT- GCGTCTAGACAAAGCG. PCR was carried out in thinwalled microcentrifuge tubes (Gordon Technologies, Toronto, ON, Canada) in a Cetus DNA 480 Thermocycler (Perkin-Elmer, Foster City, CA, USA). A typical 50-Wl reaction consisting of: 10 ng of template, 100 bmol of each primer, 0.2 mm of each deoxynucleoside triphosphate, 2 mm MgSO 4 and 5 U of PLATINUM Taq polymerase in the PCR bu er supplied by the manufacturer. The reaction consists of a hot start of 2 min at 95³C, then 30 cycles of 1 min each at 94³C, 57³C and 72³C. The 2-kbp PCR product was digested with KpnI and XbaI and cloned into the E. coli^m. haemolytica shuttle vector pnf2176 [11], also digested with KpnI and XbaI. Recombinant plasmids were screened in E. coli DH5K transformants after selection with ampicillin. The recombinant plasmids were mapped to con rm insertion of the PCR product into the vector. A plasmid that had cloned the luxs-containing fragment, named prmluxs, was recovered. The expression of luxs in this plasmid was con rmed by its ability to complement E. coli DH5K to produce AI-2 activity in the V. harveyi bioluminescence assay as described above. 3. Results and discussion 3.1. M. haemolytica A1 produces a signaling molecule similar to AI-2 of V. harveyi V. harveyi BB170 (sensor 1 3, sensor 2 ) was used in this bioluminescence assay. The characteristic quorum-sensing behavior of V. harveyi BB170 is shown in Fig. 1 with the addition of BHI as a negative control. After the initial 1:5000 dilution, the RLU decreased in the rst 4 h due Fig. 1. Activity pro les of the V. harveyi BB170 bioluminescence assay. Typical activity pro les are shown for the reporter strain with the addition of BHI (b); the addition of M. haemolytica A1 culture supernatant from a test culture at early exponential phase (R); the addition of M. haemolytica A1 culture supernatant from a test culture at early stationary phase (8); and the addition of V. harveyi BB152 culture supernatant from a test culture at early stationary phase (F). Error bars indicate standard deviation of three independent sample measurements. to the decrease in cell density and dilution of AI-2 in the culture. As the cell density increased, the concentration of AI-2 increased and the reporter strain responded by an exponential increase in RLU to reach the predilution RLU level by 7 h of the assay. When the culture supernatant from V. harveyi BB152 (AI-1 3, AI-2 ) grown to late exponential phase (OD ) was used an exogenous source of AI-2, the reporter strain maintained a high level of RLU throughout the 7-h time course (Fig. 1). This is due to V. harveyi BB170 sensing and responding to the exogenous AI-2 present in the V. harveyi BB152 culture supernatant, thus eliminating the dilution e ect. M. haemolytica A1 culture supernatants from various growth stages were tested in the assay. Upon the addition of the M. haemolytica A1 culture supernatant from early exponential phase (OD ) to V. harveyi BB170, the reporter strain behaved in a similar fashion to that of the characteristic quorum-sensing activity pro le (Fig. 1), suggesting that at the early exponential phase of growth, M. haemolytica A1 does not produce a signaling molecule that is able to induce the luminescent property of V. harveyi BB170. However, when M. haemolytica A1 culture supernatant from late exponential phase (OD ) was used in the assay, the reporter strain produces an activity pro- le similar to that with the addition of known exogenous AI-2. This indicates that there is a signaling molecule similar to the AI-2 of V. harveyi produced by M. haemolytica A1 at this phase of growth that is able to induce the luminescence property V. harveyi BB M. haemolytica A1 does not produce a signaling molecule similar to AI-1 of V. harveyi The V. harveyi strain BB886 was used to examine if

4 28 R.J. Malott, R.Y.C. Lo / FEMS Microbiology Letters 206 (2002) 25^30 M. haemolytica A1 could induce the luminescence property through the quorum-sensing system 1. The characteristic quorum-sensing activity pro le for V. harveyi BB886 is shown with the addition of BHI broth (Fig. 2). As in the above experiment with BB170, dilution of the culture (hence the concentration of AI-1) resulted in a decrease of RLU in the rst 4 h. As the cell density increased, the induced RLUs returned to the predilution values due to the increase in AI-1 concentration. On the other hand, the addition of culture supernatant from V. harveyi BB120 (1,2 ) to the BB886 assay resulted in a high level of luminescence throughout the assay period (data not shown). When M. haemolytica A1 culture supernatant from late exponential phase of growth (OD ) was used in the assay, a similar curve to the characteristic quorum-sensing pro le was observed (Fig. 2). This suggests there is no production of an AI-1-like molecule from M. haemolytica A1 that could induce the luminescence property of V. harveyi BB886 at this stage of growth The production of the M. haemolytica A1 AI-2-like molecule is proportional to cell density Fig. 3. Comparison of the V. harveyi BB170 bioluminescence assays at the 5 h time point along the time course of the assay. Black bars: addition of V. harveyi culture supernatant; white bars: addition of M. haemolytica A1 culture supernatants; gray bar: addition of BHI control. The OD of each culture when the supernatants were collected is indicated on the X-axis. Error bars indicate standard deviation of three independent sample measurements. Test culture supernatants of M. haemolytica A1 from di erent growth stages were examined for luminescence induction in V. harveyi BB170. The RLU values for each of the assays were compared at the 5-h time point. This time point has been used as an appropriate stage to monitor the e ects of exogenous AI-2 in the bioluminescence assay [4,5]. The RLU at this time point is compared to the RLU value obtained for V. harveyi BB170 produced by the addition of BHI broth alone. The results in Fig. 3 show that there is a gradual increase in RLU production in the V. harveyi BB170 reporter strain as the OD of the M. haemolytica A1 test culture increases. This provides further evidence that M. haemolytica A1 produces a signaling molecule similar to V. harveyi AI-2 that is able to induce the reporter strain V. harveyi BB170. It also con- rms that the production of the signaling molecule is proportional to cell density. V. harveyi BB170 assays using culture supernatants from the reporter strain V. harveyi BB152 grown to early^mid-exponential phase and early stationary phase were also compared at the 5-h time point (Fig. 3). As seen with the M. haemolytica A1 culture supernatants, the RLU production increased as the cell density of the test culture increased Production of AI-2-like molecules from other Pasteurellaceae species Culture supernatant was collected from several Pasteurellaceae to determine if they can induce quorum-sensing system 2 in V. harveyi. The results from the V. harveyi BB170 bioluminescence assay show that the culture supernatants from P. multocida, A. pleuropneumoniae, P. trehalosi and A. suis induced an increase of 27-fold, 6-fold, 5- fold and 34-fold (respectively) in bioluminescence compared to media control. This data suggests that these microorganisms also produce AI-2-like quorum-sensing molecules and likely exhibit the system 2 quorum-sensing mechanism luxs M:h: is functionally homologous to luxs E:c: Fig. 2. Activity pro les of the V. harveyi BB886 bioluminescence assay. Typical activity pro les for the reporter strain with the addition of BHI (b); and the addition of M. haemolytica A1 culture supernatant from a test culture at late exponential phase (F). Error bars indicate standard deviation of three independent sample measurements. The E. coli strain DH5K has been shown to carry a frame-shift mutation in its luxs gene, and is incapable of producing AI-2 [12]. A BLAST analysis of the M. haemolytica A1 sequencing project identi ed a LuxS homolog within Contig 870 that is 71% identical and 83% similar to LuxS of V. harveyi. PCR primers were designed to amplify a 2-kbp product that encompass the putative luxs locus. The PCR product was cloned into pnf2176 to produce prmluxs. This plasmid was transformed into

5 R.J. Malott, R.Y.C. Lo / FEMS Microbiology Letters 206 (2002) 25^30 29 Fig. 4. Comparison of bioluminescence produced by V. harveyi BB170 in response to the addition of culture supernatants from E. coli DH5K (pnf2176 and prmluxs), M. haemolytica A1 as well as a LT-medium control. The RLUs were compared at the 5 h time point along the assay. The cell-free supernatants of the test strains were collected at late exponential phase of growth. Error bars indicate standard deviation of three independent sample measurements. E. coli DH5K, and the resulting transformant examined for the ability to induce bioluminescence in V. harveyi BB170. The results in Fig. 4 show that prmluxs restores the production of AI-2 in DH5K, in support of the presence of a functional homolog of luxs in M. haemolytica A1. 4. Conclusion We have demonstrated that several members of the Pasteurellaceae exhibit a density-dependent genetic regulatory mechanism that is similar to the system 2 quorum-sensing mechanism in V. harveyi. In V. harveyi, LuxS is involved in the synthesis of AI-2 molecules that accumulate as the cell population density increases. The increasing concentration of AI-2 interacts with the periplasmic binding LuxP protein and activates the system 2 sensor LuxQ protein, resulting in either positive or negative regulation of gene expression [2]. It has been hypothesized that quorum-sensing system 2 is a widely distributed system for interspecies communication [2]. In addition to identifying the presence of a LuxS homolog in the genome, there have been studies that showed the presence of AI-2-like activity in the cell-free supernatants of Helicobacter pylori [5], Shigella exneri [13], Neisseria meningitidis [14], Actinobacillus actinomycetemcomitans [15,16], Bacillus anthracis [17], Bacillus subtilis [18] as well as in a number of peridontal pathogens [19,20]. The present study showed that these Pasteurellaceae species also produce AI-2 quorum-sensing molecules as hypothesized. Numerous research are being carried out in many laboratories to examine the regulation of virulence genes by quorum-sensing system 2 with the objective of linking virulence expression to density-dependent gene regulation [21]. The expression of a type III secretion system in E. coli O157:H7 has been shown to be regulated by AI- 2 [22]. This is one of the rst reports of potential virulence genes being regulated by this quorum-sensing mechanism. The expression of the leukotoxin, the sialoglycoprotease, the capsule and other potential virulence factors of M. haemolytica A1 [23] are being examined in light of the present data. Recently, it has been suggested that in A. actinomycetemcomitans, the expression of the leukotoxin is increased three-fold in response to increased level of AI-2 [16]. If some of these virulence genes are regulated by a quorum-sensing mechanism, e orts can be made to short circuit the system through AI analogues or other means to attenuate expression of virulent factor(s). To fully support the hypothesis that M. haemolytica A1 and the other members of the Pasteurellaceae exhibit a system 2 quorum-sensing mechanism, it would be necessary to identify the other components of the quorum-sensing pathway. Since the LuxS protein has been suggested to be associated with a metabolic pathway involved in the conversion SAM into a furanone derivative [6], it is possible that the AI-2-like molecule is a secondary metabolite of LuxS-mediated metabolism of SAM in M. haemolytica A1 (and these Pasteurellaceae organisms) and is not directly involved in a quorum-sensing mechanism. A preliminary BLAST analysis showed the presence of a LuxP homolog in the genome of P. multocida and a LuxQ homolog in the genome of H. in uenzae. This is in consistent with the presence of a system 2 quorum-sensing mechanism in these microorganisms. Experiments are in progress to characterize other components such as the LuxP, LuxQ in M. haemolytica A1. Acknowledgements We thank Dr. Bonnie Bassler for the gift of the V. harveyi reporter strains. We also thank Dr. Sarah Highlander for supplying us sequence data from the ongoing M. haemolytica genome sequencing project. This research is supported by a grant from the Natural Sciences and Engineering Research Council of Canada. References [1] Bassler, B.L. and Wright, M. (1994) Multiple signalling systems controlling expression of luminescence in Vibrio harveyi: sequence and function of genes encoding a second sensory pathway. Mol. Microbiol. 13, 273^286. [2] Lilley, B.N. and Bassler, B.L. (2000) Regulation of quorum sensing in Vibrio harveyi by LuxO and Sigma-54. Mol. Microbiol. 36, 940^954. [3] Bassler, B.L., Greenberg, E.P. and Stevens, A.M. (1997) Cross-species induction of luminescence in the quorum sensing bacterium Vibrio harveyi. J. Bacteriol. 179, 4043^4045. [4] Surette, M.G. and Bassler, B.L. (1998) Quorum sensing in Escherichia coli, Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 95, 7046^7050. [5] Forsyth, M.H. and Cover, T.L. (2000) Intercellular communication

6 30 R.J. Malott, R.Y.C. Lo / FEMS Microbiology Letters 206 (2002) 25^30 in Helicobacter pylori: luxs is essential for the production of an extracellular signaling molecule. Infect. Immun. 68, 3193^3199. [6] Schauder, S., Shokat, K., Surette, M.G. and Bassler, B.L. (2001) The LuxS family of bacterial autoinducers: biosynthesis of a novel quorum-sensing signaling molecule. Mol. Microbiol. 41, 463^476. [7] Lo, R.Y.C. and Shewen, P.E. (1981) The Genus Pasteurella. In: The Prokaryotes, 2 edn. (Balows, A., Truper, H.G., Dworkin, M., Harder, W. and Schleifer, K.-H., Eds.), pp. 3331^3338. Springer-Verlag, New York. [8] Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. [9] Greenberg, E.P., Hastings, J.W. and Ulitzur, S. (1979) Induction of luciferase synthesis in Beneckea harveyi by other marine bacteria. Arch. Microbiol. 120, 87^91. [10] Surette, M.G. and Bassler, B.L. (1999) Regulation of autoinducer production in Salmonella typhimurium. Mol. Microbiol. 31, 585^595. [11] Fedorova, N.D. and Highlander, S.K. (1977) Plasmids for heterologous expression in Pasteurella haemolytica. Gene 186, 207^211. [12] Surette, M.G., Miller, M.B. and Bassler, B.L. (1999) Quorum sensing in Escherichia coli, Salmonella typhimurium, and Vibrio harveyi: a new family of genes responsible for autoinducer production. Proc. Natl. Acad. Sci. USA 96, 1639^1644. [13] Day Jr., W.A. and Maurelli, A.T. (2001) Shigella exneri LuxS quorum-sensing system modulates virb expression but is not essential for virulence. Infect. Immun. 69, 15^23. [14] Green, A.D., Winzer, K., Delaroy, M., Sun, Y., Blackley, D., Hardie, K.R., Tang, C. and Baldwin, T. (2001) Synthesis of autoinducer 2 depends on LuxS in Neisseria meningitidis. In: ASM Conference on Cell^Cell Communication in Bacteria, Snowbird, Utah, abstract 64. [15] Barbieri, B., Spratt, D., Wilson, M. and McNab, R. (2001) Quorum sensing in the peridontal pathogen Actinobacillus actinomycetemcomitans. In: ASM Conference on Cell^Cell Communication in Bacteria, Snowbird, Utah, abstract 52. [16] Fong, K.P., Chung, W.O., Lamont, R.J. and Demuth, D.R. (2001) Intra- and interspecies regulation of gene expression by Actinobacillus actinomycetemcomitans luxs. In: ASM Conference on Cell^Cell Communication in Bacteria, Snowbird, Utah, abstract 56. [17] Bourgogne, A.D. and Koehler, T.M. (2001) Bacillus anthracis harbors a functional orthologue of luxs, a gene essential for autoinducer-2 activity. In: 101st ASM General Meeting, Orlando, FL, abstract B-385. [18] Das, S.K., Sedelnikove, S.E., Baker, P.J., Ruzheinikov, S.N., Foster, S., Hartley, A., Horsburgh, M.J. and Rice, D.W. (2001) Cloning, puri cation, crystallization and preliminary crystallographic analysis of Bacillus subtilis LuxS. Acta Crystallogr. D. Biol. Crystallogr. 57, 1324^1325. [19] Chung, W.O., Park, Y., Lamont, R.J., McNab, R., Barbieri, B. and Demuth, D.R. (2001) Signaling system in Porphyromonas gingivalis based on a LuxS protein. J. Bacteriol. 183, 3903^3909. [20] Frias, J., Ollie, E. and Alsina, M. (2001) Peridontal pathogens produce quorum sensing signal molecules. Infect. Immun. 69, 3431^3434. [21] Robson, N.D., Cox, A.R.J., McGowan, S.J., Bycroft, B.W. and Salmond, G.P.C. (1997) Bacterial N-acyl-homoserine-lactone-dependent signalling and its potential biotechnological applications. TIBTCH 15, 458^464. [22] Sperandio, V., Mellies, J.L., Nguyen, W., Shin, S. and Kaper, J.B. (1999) Quorum sensing controls expression of the type III secretion gene transcription and protein secretion in enterohemorragic and enteropathogenic Escherichia coli. Proc. Natl. Acad. Sci. USA 96, 15196^ [23] Lo, R.Y.C. (2001) Genetic analysis of virulence factors of Mannheimia (Pasteurella) haemolytica A1. Vet. Microbiol. 83, 23^35.