(Actinobacillus) actinomycetemcomitans LsrB and RbsB Proteins with ...

JOURNAL OF BACTERIOLOGY, Aug. 2007, p. 5559–5565 0021-9193/07/$08.00⫹0 doi:10.1128/JB.00387-07 Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Vol. 189, No. 15

Differential Interaction of Aggregatibacter (Actinobacillus) actinomycetemcomitans LsrB and RbsB Proteins with Autoinducer 2䌤 Hanjuan Shao,1 Deanna James,1 Richard J. Lamont,2 and Donald R. Demuth1* Department of Periodontics, Endodontics and Dental Hygiene, University of Louisville School of Dentistry, Louisville, Kentucky,1 and Department of Oral Biology, University of Florida School of Dentistry, Gainesville, Florida2 Received 15 March 2007/Accepted 11 May 2007

Our previous studies showed that the Aggregatibacter actinomycetemcomitans RbsB protein interacts with cognate and heterologous autoinducer 2 (AI-2) signals and suggested that the rbsDABCK operon encodes a transporter that may internalize AI-2 (D. James et al., Infect. Immun. 74:4021–4029, 2006.). However, A. actinomycetemcomitans also possesses genes related to the lsr operon of Salmonella enterica serovar Typhimurium which function to import AI-2. Here, we show that A. actinomycetemcomitans LsrB protein competitively inhibits the interaction of the Vibrio harveyi AI-2 receptor (LuxP) with AI-2 from either A. actinomycetemcomitans or V. harveyi. Interestingly, LsrB was a more potent inhibitor of LuxP interaction with AI-2 from V. harveyi whereas RbsB competed more effectively with LuxP for A. actinomycetemcomitans AI-2. Inactivation of lsrB in wild-type A. actinomycetemcomitans or in an isogenic RbsB-deficient strain reduced the rate by which intact bacteria depleted A. actinomycetemcomitans AI-2 from solution. Consistent with the results from the LuxP competition experiments, the LsrB-deficient strain depleted AI-2 to a lesser extent than the RbsB-deficient organism. Inactivation of both lsrB and rbsB virtually eliminated the ability of the organism to remove AI-2 from the extracellular environment. These results suggest that A. actinomycetemcomitans possesses two proteins that differentially interact with AI-2 and may function to inactivate or facilitate internalization of AI-2. and gram-negative bacteria, and many, if not all, of these organisms produce an AI-2-like signal that is recognized by LuxP and is capable of inducing the expression of the lux operon of V. harveyi. As a result, AI-2 has been suggested to represent a universal quorum-sensing signal that is recognized by many organisms (37) and has been reported to influence a wide variety of cellular processes including type III secretion (33), cell motility (17, 32), the development of biofilms (3, 7, 12, 20, 21, 39), the expression of virulence factors (9, 14, 19), and iron uptake (8). However, many of these organisms lack components of the dedicated response circuits that mediate the quorum-sensing response of Vibrio subsp. Therefore, the mechanisms that govern the response to AI-2 in these organisms are not well understood. Interestingly, S. enterica serovar Typhimurium appears to actively internalize AI-2 via an ABC transporter encoded by the lsrACDBFGE operon (35), and after uptake, AI-2 is phosphorylated by an AI-2 kinase (LsrK) and induces the expression of the lsr operon (38). Our studies have focused on the oral pathogen Aggregatibacter (Actinobacillus) actinomycetemcomitans, a gram-negative organism that is associated with aggressive forms of earlyonset periodontitis and other systemic infections (4, 24, 31, 40). A. actinomycetemcomitans expresses luxS and produces an AI2-like signal that induces V. harveyi bioluminescence (9) and regulates the growth of the organism under iron limitation by controlling the expression of a variety of iron storage and uptake genes (8). A. actinomycetemcomitans lacks the sensor kinase/phosphatase of V. harveyi (LuxQ) but expresses two genes that encode putative periplasmic proteins (RbsB and LsrB) that are related to LuxP, the receptor for AI-2 in V. harveyi (13). The RbsB protein is similar to the Escherichia coli periplasmic ribose binding protein encoded by the rbs operon

Autoinducer 2 (AI-2) is a quorum-sensing signal that was initially identified in Vibrio harveyi (2) and is produced by the luxS gene (28, 29). AI-2 is produced by LuxS-catalyzed cleavage of S-ribosylhomocysteine to produce homocysteine and 4,5-dihydroxy-2,3-pentanedione (36), which in turn undergoes further rearrangement to produce AI-2. However, two distinct structural forms of AI-2 have been identified. Salmonella enterica serovar Typhimurium produces 2R,4S-2,3,3,4-methyltetrahydroxytetrahydrofuran (R-THMF) whereas Vibrio subsp. produce the borate diester form of S-THMF (6, 22). AI-2dependent quorum sensing is quite complex and is involved in the regulation of a variety of genes in Vibrio species, including the lux operon of V. harveyi. Detection of AI-2 by V. harveyi is mediated by LuxP, a periplasmic AI-2 receptor (6) that associates with the LuxQ sensor kinase-phosphatase (23) and initiates a phosphotransfer cascade involving LuxU (11) and the response regulator LuxO (10). LuxO in turn influences the expression of multiple small regulatory RNAs (Qrr regulators) that influence the expression of LuxR (15, 16), the master regulator of the lux operon (16). Recently, an additional twocomponent system encoded by varSA in Vibrio cholerae has been shown to converge on the quorum-sensing circuit by regulating the expression of the small RNAs encoded by csrBCD which control the expression of Qrr via CsrA (15). LuxS is highly conserved in a wide range of gram-positive

* Corresponding author. Mailing address: Department of Periodontics, Endodontics and Dental Hygiene, University of Louisville School of Dentistry, 501 South Preston Street, Room 209, Louisville, KY 40292. Phone: (502) 852-3807. Fax: (502) 852-4052. E-mail: drdemu01 @louisville.edu. 䌤 Published ahead of print on 25 May 2007. 5559

5560

SHAO ET AL.

(rbsDABCK) that functions to transport ribose into the cell. However, we have shown that A. actinomycetemcomitans RbsB competes with LuxP for AI-2 and inhibits AI-2-mediated induction of V. harveyi bioluminescence. An A. actinomycetemcomitans mutant lacking RbsB failed to attain high cell density under iron-limiting growth conditions and exhibited reduced expression of a ferric ABC transporter that is regulated by luxS (13). In addition, inactivation of rbsB reduced but did not completely eliminate the ability of the organism to deplete AI-2 from solution, suggesting that redundant mechanisms may exist in A. actinomycetemcomitans for interacting with AI-2. In this report, we show that LsrB may also contribute to the interaction of A. actinomycetemcomitans with AI-2. Purified LsrB protein competitively inhibited the interaction of AI-2 with the V. harveyi AI-2 receptor LuxP. Interestingly, LsrB was a more potent inhibitor of LuxP interaction with AI-2 from V. harveyi whereas RbsB competed more effectively with LuxP for A. actinomycetemcomitans AI-2. Inactivation of lsrB reduced the rate at which intact bacteria depleted AI-2 from solution, and a double knockout of lsrB and rbsB virtually eliminated AI-2 depletion. Thus, A. actinomycetemcomitans possesses two proteins that differentially interact with AI-2, suggesting that the organism may be capable of detecting and responding to a wide range of AI-2 concentrations in the oral biofilm.

MATERIALS AND METHODS Bacterial strains and growth conditions. A. actinomycetemcomitans was cultured in brain heart infusion medium (Becton Dickinson) supplemented with 40 mg of NaHCO3 per liter. Cultures were maintained at 37°C. Growth of the lsrB and lsrB rbsB mutant strains was carried out in the medium described above supplemented with 80 mg of spectinomycin per liter or with 25 mg of kanamycin and 80 mg of spectinomycin per liter, respectively. E. coli was cultured in LuriaBertani (LB) medium with aeration at 37°C. E. coli strains containing the pBAD/ gIIIA expression plasmid were cultured in LB medium supplemented with 50 mg of ampicillin per liter. V. harveyi BB170 (sensor 1⫺, sensor 2⫹) was a gift from B. Bassler (Princeton University) and was grown overnight in AB medium (1) with aeration at 30°C. AB medium consists of 0.3 M NaCl, 50 mM MgSO4, 0.2% Casamino Acids, 10 mM potassium phosphate (pH 7.0), 1 mM L-arginine, 2% glycerol, 1 ␮g/ml thiamine, and 10 ng/ml riboflavin. Expression and purification of A. actinomycetemcomitans LsrB. A. actinomycetemcomitans lsrB was identified from the complete genome sequence of A. actinomycetemcomitans HK1651 and amplified using genomic DNA from strain JP2 (5) using primers P1 (5⬘-GCGCCATGGGCAAAGCCAACGTTCAAAAA ACG-3⬘) and P2 (5⬘-CGTCTAGATTGAAATTGTAGTTGTCGAT-3⬘) (see Fig. 1). The underlined sequences in the primers above represent NcoI and XbaI restriction sites to facilitate cloning into pBAD/gIIIA (Invitrogen). Amplification was performed using the following PCR program: 95°C for 5 min and then 30 cycles of 95°C for 1 min, 60°C for 2 min, and 72°C for 3 min. The resulting PCR product was ligated with pGEMT-Easy (Promega) and transformed into E. coli DH5␣ to produce strain GEMT-lsrB. Recombinant clones were confirmed with EcoRI digestion to release the lsrB insert. Plasmid pGEMT-lsrB was then digested with NcoI and XbaI, and the resulting fragment was ligated with pBAD/ gIIIA that was also digested with NcoI and XbaI. After transformation into E. coli TOP10, recombinant clones were confirmed by restriction analysis. The resulting E. coli strain, designated pBAD/gIIIA-lsrB, was used to express the LsrB protein containing a C-terminal six-His tag. E. coli pBAD/gIIIA-lsrB was grown to early exponential phase in 100 ml of LB medium supplemented with 50 ␮g/ml ampicillin, and expression of LsrB was induced with arabinose at a final concentration of 0.02%. Growth with aeration continued for 3 to 4 h, after which cells were harvested, lysed in 0.2% sodium dodecyl sulfate, and centrifuged to 10,000 ⫻ g to remove cell debris. LsrB was purified from the cell extract by affinity chromatography using HisTrap nickel resin (GE Healthcare). The column was washed with 10 ml of distilled H2O and equilibrated with the same volume of loading buffer (20 mM sodium phosphate, pH 7.5, 0.5 M NaCl, and 10 mM imidazole). The crude cell extract was loaded

J. BACTERIOL.

FIG. 1. The lsrACDBFG and lsrRK operons of A. actinomycetemcomitans. (A) Comparison of the lsr operons of S. enterica serovar Typhimurium (top) and A. actinomycetemcomitans (bottom). The numbers represent percent amino acid sequence identity between the corresponding S. enterica and A. actinomycetemcomitans genes. The A. actinomycetemcomitans lsr operon lacks the lsrE gene present in S. enterica. (B) Expanded view of the A. actinomycetemcomitans lsrDB region showing the locations of oligonucleotide primers (see Materials and Methods) used to clone lsrB gene fragments for expression in E. coli and gene inactivation and verification in A. actinomycetemcomitans.

onto the column by syringe, and the column was washed with 20 ml of loading buffer. Bound LsrB was eluted with elution buffer (20 mM sodium phosphate, pH 7.5, 0.5 M NaCl, and 0.5 M imidazole). The eluted protein was dialyzed overnight against 10 mM sodium phosphate, pH 7.5, and lyophilized. The purified LsrB protein was suspended in distilled H2O, and protein concentration was determined using a MicroBCA assay (Pierce) according to the manufacturer’s instructions. Purity was confirmed by visualizing LsrB protein on 4 to 15% gradient NuPAGE gels (Invitrogen) after Coomassie blue staining. Determination of V. harveyi bioluminescence. Preparations of V. harveyi AI-2 were obtained from conditioned medium of V. harveyi BB170 cultures grown in AB broth. After cells were harvested, the conditioned medium was filtered through a 0.22-um-pore-size filter and used immediately. AI-2 from A. actinomycetemcomitans was partially purified from an exponential phase culture using a Sep-Pak C18 column (Waters) as previously described by Sperandio et al. (32). Chromatography was carried out according to the instructions of the manufacturer, and AI-2 activity was present in the column flowthrough fraction. For determination of V. harveyi bioluminescence, an aliquot of an overnight V. harveyi BB170 culture was diluted 1:25,000 into fresh sterile AB medium, and 90 ␮l of cells was added to each well of a microtiter plate. Positive control reactions then received 10 ␮l of V. harveyi AI-2 or 10 ␮l of the partially purified A. actinomycetemcomitans AI-2 that was previously diluted 1:25 with sterile AB medium. Negative control reactions received 10 ␮l of sterile AB medium. Experimental wells received 10 ␮l of V. harveyi or A. actinomycetemcomitans AI-2 preparations that were preincubated for 30 min at 30°C with various concentrations of the purified LsrB protein (0 to 5,000 ng/ml). The microtiter plates were then shaken at 500 rpm at 30°C. Bioluminescence was measured in hourly intervals using a Wallac Victor3 multilabel counter (Perkin Elmer) and expressed as a percentage of the bioluminescence measured in the positive control reaction. For direct competition experiments, V. harveyi and A. actinomycetemcomitans AI-2 preparations were added to diluted V. harveyi BB170 cells (in the absence of LsrB), and the reaction mixtures were incubated at 30°C as described above until the induction of bioluminescence occurred in the positive control reaction. Routinely, bioluminescence of the positive control reaction began to increase at the 3-h time point. At that time, all reactions were removed from the shaker, and experimental wells received purified LsrB protein to a final concentration of 0 to 5 ␮g per ml. Plates were then incubated with shaking for an additional 3 to 5 h to allow the induction of V. harveyi bioluminescence to continue, and during this period, light production was monitored at hourly intervals as described above. Routinely, the bioluminescence in the positive control reactions increased to ⬎400-fold over the negative control (cells without A. actinomycetemcomitans AI-2) at the 6-h time point and ⬎1,200-fold at the 7-h time point. Inactivation of A. actinomycetemcomitans LsrB. Two fragments encompassing the entire A. actinomycetemcomitans lsrB gene were amplified from A. actinomycetemcomitans genomic DNA using primer pairs P3 (5⬘-GGTACCATGAAA ACGCGTGTAAAATTA-3⬘) and P4 (5⬘-GGATCCTTCCGGATGATCTTTG GCAA-3⬘), and P5 (5⬘-GGATCCAGCCCGACGGTCACCGAC-3⬘) and P6 (5⬘TCTAGAGAAATTGTAGTTGTCGATATTG-3⬘) as shown in Fig. 1. The underlined sequences in the primers above represent KpnI (P3), BamHI (P4 and

A. ACTINOMYCETEMCOMITANS LsrB INTERACTION WITH AI-2

VOL. 189, 2007

P5), and XbaI (P6) restriction sites to facilitate cloning. Amplification was performed using the following PCR program: 95°C for 5 min and then 30 cycles of 95°C for 1 min, 60°C for 2 min, and 72°C for 3 min. The resulting PCR products were initially ligated with pGEMT-Easy (Promega) and transformed into E. coli DH5␣. Recombinant clones were confirmed by EcoRI restriction digest to release the appropriate insert fragments. Each fragment was then cleaved from pGEM-T by digesting with KpnI/BamHI or BamHI/XbaI and ligated successively into pUC19 to reconstitute the intact lsrB gene. The resulting plasmid, pUC19lsrB, was then cleaved with BamHI for insertion of a spectinomycin resistance cassette. The spectinomycin resistance cassette of plasmid pVT1461 (kindly supplied by K. Mintz, University of Vermont) was obtained by restriction with SphI and ligated with pUC19lsrB to create pUC19lsrB-spec. This plasmid was then transformed into E. coli DH5␣, and recombinant clones were confirmed by restriction analysis. Purified pUC19lsrB-spec was introduced into wild-type A. actinomycetemcomitans (strain JP2) or an isogenic RbsB-deficient A. actinomycetemcomitans strain (13) by electroporation as previously described (9) and Ampr Specr clones were chosen for further study. To confirm integration of the plasmid into the A. actinomycetemcomitans genome, DNA was isolated and used as template for PCR using primers P7 (5⬘-GGTACCGATCACCTTAGGCAC TATG-3⬘) and Spe3 (5⬘-AACATGTATTCACGAACGAAAATCGAT-3⬘) (restriction site is underlined). Primer P7 anneals to the 3⬘ end of lsrD (upstream of primer P3), and primer Spe3 anneals within the spectinomycin resistance marker. Clones producing the appropriate PCR product of 1.3 kb confirmed insertion of the suicide construct into A. actinomycetemcomitans genomic DNA. The identity of the amplicon was also determined by sequencing from each end using oligonucleotides P7 and Spe3 as primers. Finally, the lack of lsrB transcripts in the mutant organism was also confirmed using real-time PCR. To complement the LsrB-deficient strain, lsrB was amplified from A. actinomycetemcomitans genomic DNA using primers lsrBF (5⬘-GCGGGATCCGAAA ACGCGTGTAAAA-3⬘) and lsrBR (5⬘-GCGTCTAGATGAGAAATTGTAGT TGTCG-3⬘), and the resulting amplicon was ligated with pGEM-T Easy (Promega). After transformation of E. coli DH5␣, clones were confirmed by isolation of plasmid and the release of the appropriate lsrB fragment after digestion with BamHI and XbaI (underlined sites in the PCR primers above). The released fragment was then ligated into pYGK-ltxP that was cleaved with the same restriction enzymes. PYGK-ltxP contains the promoter of the A. actinomycetemcomitans leukotoxin operon introduced into the KpnI and BamHI sites of the shuttle vector pYGK (6) and positioned to drive lsrB expression. The resulting plasmid was confirmed by restriction digestion and then introduced into the lsrB mutant strain by electroporation. Depletion of AI-2 by wild-type and LsrB-deficient strains of A. actinomycetemcomitans. Overnight cultures of A. actinomycetemcomitans were harvested by centrifugation, washed twice in sterile AB medium, and suspended in 0.5 ml of partially purified A. actinomycetemcomitans AI-2 at a cell density of approximately 4 ⫻ 108 cells per ml. Cell suspensions were incubated at 30°C for 0 to 15 min. Bacteria were then removed by centrifugation, and the supernatant was filtered through a 0.22-␮m-pore-size filter. The filtered supernatants were analyzed in triplicate for AI-2 activity using the V. harveyi BB170 reporter strain. Bioluminescence was measured and analyzed as described above.

RESULTS LsrB protein interacts with AI-2 from A. actinomycetemcomitans and V. harveyi. Our previous studies showed that the RbsB protein of A. actinomycetemcomitans competed with V. harveyi LuxP for AI-2, suggesting that RbsB may function as an A. actinomycetemcomitans receptor for AI-2 (13). However, as shown in Fig. 1A, A. actinomycetemcomitans also possesses genes that are closely related to the lsrACDBFG operon of S. enterica serovar Typhimurium that functions to transport and internalize AI-2 (34, 35, 38). To determine if the lsr operon of A. actinomycetemcomitans contributes to interaction with AI-2, the putative periplasmic AI-2-interacting protein LsrB was purified and used in competition assays to determine if LsrB inhibited the interaction of AI-2 with LuxP and reduced AI-2-dependent induction of V. harveyi bioluminescence. As shown in Fig. 2A, preincubation of LsrB with partially purified AI-2 from A. actinomycetemcomitans reduced bioluminescence of the V. harveyi BB170 reporter

5561

in a dose-dependent manner. Similar results were obtained using a direct competition assay where LsrB was added to V. harveyi BB170 cells 3 h after they were stimulated with A. actinomycetemcomitans AI-2. Further incubation of the prestimulated cells for an additional 3 h induced V. harveyi bioluminescence by 420-fold in the control culture which did not receive LsrB. In contrast, the addition of 1, 2, or 5 ␮g of LsrB to AI-2-stimulated cells completely inhibited the production of light by the reporter strain, as shown in Fig. 2B. Together, these results show that LsrB inhibits the interaction of LuxP with AI-2 and suggest that LsrB may function as a receptor for AI-2 in A. actinomycetemcomitans. The structure of LsrB of S. enterica serovar Typhimurium with bound AI-2 has been determined by Miller et al. (22). This study showed that LsrB interacts with a form of AI-2 that lacks boron, and, presumably, the structure of A. actinomycetemcomitans AI-2 is similar to that of S. enterica. Therefore, to determine if A. actinomycetemcomitans LsrB is capable of interacting with the borate diester form of AI-2, competition assays were carried out with cells that were stimulated with AI-2 obtained from V. harveyi BB170 (i.e., conditioned medium from an overnight culture of V. harveyi BB170). As shown in Fig. 2C, LsrB also competitively inhibited the interaction of LuxP with V. harveyi AI-2, resulting in a dose-dependent inhibition of bioluminescence. The results above and our previous work (13) suggest that A. actinomycetemcomitans expresses two proteins, LsrB and RbsB, that are capable of interacting with AI-2. To further characterize the activity of these proteins, we compared their interaction with AI-2 derived from A. actinomycetemcomitans and from V. harveyi. Interestingly, as shown in Fig. 3A, RbsB was a significantly more potent inhibitor of bioluminescence stimulated by A. actinomycetemcomitans AI-2 than LsrB. In contrast, LsrB was a more potent inhibitor of bioluminescence stimulated by V. harveyi AI-2 (Fig. 3B). The calculated 50% inhibitory concentration (IC50) for LsrB inhibition of bioluminescence stimulated by V. harveyi AI-2 was 0.5 nM whereas the calculated IC50 for inhibition of bioluminescence stimulated by A. actinomycetemcomitans AI-2 was 2.5 nM. In contrast, for RbsB inhibition of bioluminescence the IC50 was 12.5 nM with V. harveyi AI-2 and 0.3 nM for A. actinomycetemcomitans AI-2. Thus, A. actinomycetemcomitans expresses two AI-2-interacting proteins that appear to exhibit different binding affinity and specificity for AI-2. Inactivation of lsrB reduces A. actinomycetemcomitans elimination of AI-2 from the extracellular environment. Taga et al. (35) previously showed that inactivation of the lsr operon of S. enterica serovar Typhimurium reduced but did not eliminate AI-2 transport, suggesting that alternative transport mechanisms may exist. Likewise, our previous studies of rbsB in A. actinomycetemcomitans showed that its inactivation reduced but did not completely eliminate AI-2 depletion by intact cells (13). The results above suggest that the lsr operon of A. actinomycetemcomitans may account for that residual activity. Therefore, to determine if the lsr operon functions in concert with rbsDABCK in the interaction of A. actinomycetemcomitans with AI-2, an LsrB-deficient strain and an LsrB- and RbsBdeficient strain of A. actinomycetemcomitans were constructed by insertional mutagenesis. Insertional inactivation of lsrB was confirmed by detection of the appropriate PCR amplicon gen-

5562

SHAO ET AL.

FIG. 2. A. actinomycetemcomitans LsrB competes with V. harveyi LuxP for AI-2 derived from A. actinomycetemcomitans JP2 (A and B) or from V. harveyi BB170 (C). In panels A and C, LsrB (0 to 200 ng) was incubated for 30 min at 30°C with AI-2 that was partially purified from a mid-exponential culture of A. actinomycetemcomitans JP2. Purification of A. actinomycetemcomitans AI-2 was carried out as described by James et al. (13). Samples were then added to V. harveyi BB170 cells that were diluted 1:25,000 from an overnight culture in a microtiter plate and incubated at 30°C for up to 7 h. Bioluminescence was determined at hourly intervals by luminometry using a PerkinsElmer Wallac Victor multilabel counter. The data shown represent the 6-h time point and are expressed as a percentage of the positive control. The positive control reaction was comprised of V. harveyi BB170 cells that were stimulated with A. actinomycetemcomitans AI-2 in the absence of LsrB. Light production in the positive control reaction at the time point shown was ⬎400-fold higher than that measured from V. harveyi BB170 cells in the absence of the A. actinomycetemcomitans AI-2. Panel B shows a direct competition of LsrB and LuxP for A. actinomycetemcomitans AI-2. V. harveyi BB170 cells were stimulated with A. actinomycetemcomitans AI-2 and incubated for 3 h at 30°C; then 1 ␮g (), 2 ␮g (f), or 5 ␮g (}) of LsrB was added to the reaction mixtures, and incubation was continued for an additional 4 h.

J. BACTERIOL.

FIG. 3. Comparison of LsrB- (F) and RbsB-mediated () inhibition of V. harveyi BB170 bioluminescence stimulated by A. actinomycetemcomitans AI-2 (A) and V. harveyi AI-2 (B). V. harveyi bioluminescence was determined as described in Materials and Methods. Percent inhibition was calculated by relating the bioluminescence of experimental samples with the bioluminescence of the positive control, which did not receive inhibitor, using the following formula: [1 ⫺ (experimental bioluminescence/control bioluminescence)] ⫻ 100. Bioluminescence of the positive control was ⬎400-fold higher than that observed in the absence of AI-2.

erated from primers that annealed in lsrD and in the spectinomycin marker gene, and the lack of lsrB transcription was confirmed by reverse transcription-PCR (data not shown). Similar results were obtained when lsrB mutagenesis was carried out in an RbsB-deficient strain of A. actinomycetemcomitans (13) to generate the lsrB rbsB double knockout. The interaction of wild-type, LsrB-deficient, RbsB-deficient, and the LsrB- and RbsB-deficient A. actinomycetemcomitans

Samples were assayed for induction of bioluminescence at hourly intervals as described in Materials and Methods. The positive control reaction that did not receive LsrB protein (F) exhibited a 420-fold stimulation of bioluminescence relative to V. harveyi BB170 cells that were incubated in the absence of AI-2.

VOL. 189, 2007


5563

results suggest that LsrB-mediated depletion of A. actinomycetemcomitans AI-2 is less efficient than RbsB, consistent with the results of the competition assays which showed that RbsB was the more potent competitive inhibitor of V. harveyi bioluminescence stimulated by A. actinomycetemcomitans AI-2. DISCUSSION

FIG. 4. LsrB- and RbsB-mediated depletion of A. actinomycetemcomitans AI-2. Intact cells of A. actinomycetemcomitans JP2 (F), isogenic LsrB-deficient (}), RbsB-deficient (f), or LsrB- and RbsBdeficient strains (Œ) and the lsrB mutant complemented with a plasmid borne copy of lsrB (E) were incubated with partially purified AI-2 from a mid-exponential culture of A. actinomycetemcomitans for 0 to 15 min. Bacterial cells were then removed by centrifugation, and the supernatant was filtered through a 0.22-␮m-pore-size filter. The AI-2 bioactivity at each time point was determined by incubating an aliquot of each filtered sample with V. harveyi BB170 for 6 h as described in Materials and Methods and measuring total bioluminescence. Data are expressed as a percentage of the light production induced by AI-2 in the positive control, in which V. harveyi BB170 was stimulated with A. actinomycetemcomitans AI-2 that was not preincubated with A. actinomycetemcomitans cells. In these reactions, the total bioluminescence in the positive control reactions was approximately 400-fold greater than background levels observed with V. harveyi BB170 in the absence of AI-2. All reactions were carried out in triplicate.

strains with AI-2 was followed by incubating bacteria with partially purified A. actinomycetemcomitans AI-2 for various times and then removing the bacterial cells and assaying for residual AI-2 by monitoring the induction of bioluminescence using the V. harveyi BB170 reporter. As shown in Fig. 4, wildtype A. actinomycetemcomitans rapidly and progressively depleted AI-2 from solution, resulting in ⬎95% inhibition of V. harveyi bioluminescence after exposure of bacterial cells to AI-2 for 15 min. Incubation of the LsrB-deficient strain with AI-2 for 15 min resulted in similar inhibition of V. harveyi bioluminescence. However, the LsrB mutant depleted AI-2 to a lesser but significant extent (P ⱕ 0.05) than wild type at the 5- and 10-min time points. Complementing the mutant strain with a plasmid-borne copy of lsrB restored the wild-type phenotype. The RbsB-deficient strain depleted AI-2 at a significantly reduced rate relative to wild type, exhibiting only 10% inhibition of V. harveyi bioluminescence at 10 min and approximately 50% inhibition after a 15-min incubation with AI-2. Finally, the strain lacking both LsrB and RbsB did not significantly deplete AI-2 after incubation for 10 min and only slightly depleted AI-2 activity, by approximately 20%, after 15 min (Fig. 4). Thus, inactivation of both lsrB and rbsB further reduced the ability of A. actinomycetemcomitans to deplete AI-2 from solution over either single mutation, confirming that both LsrB and RbsB participate in AI-2 interaction. The partially purified A. actinomycetemcomitans AI-2 preparation itself was stable and exhibited no loss of bioactivity upon incubation for 15 min in the absence of cells (not shown). These

Previous studies identified an operon in S. enterica, lsrACDBFGE, that encodes an ABC-type transporter of AI-2, but genetic inactivation of this transport system did not completely eliminate the ability of intact cells to deplete AI-2 from the extracellular environment (13, 35). In A. actinomycetemcomitans, we showed that inactivation of rbsDACBK also reduced but did not eliminate the depletion of AI-2 by bacterial cells. This suggests that there may be functional redundancy in the interactions of these organisms with AI-2. Examination of the A. actinomycetemcomitans HK1651 genome sequence (www.oralgen.lanl.gov) indicated that it also encodes a putative ABC-type transporter (lsrACDBFG) that is related to the lsr operon of S. enterica. Our results here show that the putative periplasmic protein encoded by this operon (LsrB) contributes to the elimination of AI-2 from the extracellular environment by A. actinomycetemcomitans. Thus, it is possible that the lsr operon of A. actinomycetemcomitans may function similarly to lsrACDBFGE of S. enterica and import AI-2. Interestingly, although insertional inactivation of lsrB has polar effects and alters the expression of the downstream lsrFG genes, depletion of AI-2 was restored by complementing the mutant with only lsrB. This is consistent with the idea that the lsr operon encodes an AI-2 transporter since the genes encoding the transport functions reside upstream of lsrB and should not be affected by inactivation of lsrB. However, the possibility that LsrB inactivates extracellular AI-2 cannot be excluded. LsrB is structurally related to the V. harveyi receptor for AI-2 (LuxP), and we showed that LsrB competes with LuxP for AI-2 from either A. actinomycetemcomitans or V. harveyi. Interestingly, LsrB was a better competitive inhibitor in the presence of V. harveyi AI-2 (IC50, 0.5 nM) than with AI-2 obtained from A. actinomycetemcomitans (IC50, 2.5 nM). This is in contrast to the activity of the RbsB protein which competed much more effectively with LuxP for A. actinomycetemcomitans AI-2 (IC50 values of 0.3 nM versus 12.5 nM). Inactivation of either rbsB or lsrB reduced the ability of the mutated strains to deplete A. actinomycetemcomitans AI-2 from the extracellular environment, suggesting that both genes encode putative periplasmic proteins that may be involved in internalization or inactivation of AI-2. Inactivation of both genes resulted in almost complete loss of AI-2 elimination. The residual depletion of AI-2 detected at longer incubation times could be explained by lowlevel interaction of AI-2 with other sugar transport systems. For example, the A. actinomycetemcomitans genome contains a second operon that is annotated as a ribose transporter, but this operon encodes transport proteins of the CUT1 family of carbohydrate uptake systems (30), whereas rbsDACBK encodes a CUT2 transporter. CUT1 transporters are usually involved in transporting di- and/or oligosaccharides (30), suggesting that the primary specificity of this ABC transporter may not be ribose or AI-2. In the context of the oral biofilm which is the natural habitat

5564

SHAO ET AL.

for A. actinomycetemcomitans, Rickard et al. (27) recently suggested that early colonizers of oral surfaces respond to a defined level of AI-2, above or below which the formation of the microbial community is suppressed. The expression by A. actinomycetemcomitans of two proteins that differentially interact with AI-2 may allow the organism to sample and respond to a broader range of AI-2 concentrations which may facilitate its colonization of both early (i.e., low biomass and low AI-2 concentration) and mature (high biomass and high AI-2 concentration) oral microbial communities. It is also possible that the presence of RbsB and LsrB may facilitate the detection of different forms of AI-2 that may exist in the oral biofilm. An interesting potential outcome of this hypothesis is that A. actinomycetemcomitans may differentially respond to AI-2 in a receptor-dependent manner. We are currently analyzing the A. actinomycetemcomitans receptor mutants using genomic microarrays to identify differential patterns of gene expression in these strains when exposed to AI-2. It is also possible, given the apparent higher affinity of LsrB for the borate diester form of AI-2, that the primary function of the lsr operon may be to acquire boron from the environment. Very little is known about boron requirements and homeostasis in microbes, but borate is known to be required in plants for sugar transport, flower retention, and pollen formation (18). Borate is also present in the human diet and has recently been shown to be important for growth and proliferation of human cells (26). Furthermore, a specific Na⫹-coupled borate transporter, NaBC1, has been identified in humans (25) and is highly expressed in salivary glands. Since borate forms stable diesters with cis-diols on furanoid rings (18), AI-2 is an ideal microbial scavenger for borate. It is possible that A. actinomycetemcomitans and other oral and enteric organisms have evolved to obtain borate in diester form with THMF from the host diet utilizing the Lsr transporter. The function of rbsDABCK may be to efficiently recycle noncomplexed AI-2. In summary, our results suggest that LsrB may function as a periplasmic AI-2 binding protein in A. actinomycetemcomitans. LsrB-mediated interaction with AI-2 is redundant with the RbsB protein but LsrB appears to interact more efficiently with the borate diester form of AI-2 than does RbsB. The similarity of the lsr operon with the known AI-2 transporter in S. enterica suggests that A. actinomycetemcomitans may import AI-2, but further study of the extracellular and intracellular functions of AI-2 will be required to determine how AI-2 secretion and/or uptake influences growth of A. actinomycetemcomitans in the oral biofilm. ACKNOWLEDGMENT This work was supported by Public Health Service Grant RO1 DE14605 from the National Institute of Dental and Craniofacial Research. REFERENCES 1. Bassler, B. L., M. Wright, R. E. Showalter, and M. R. Silverman. 1993. Intercellular signaling in Vibrio harveyi: sequence and function of genes regulating expression of luminescence. Mol. Microbiol. 9:773–786. 2. Bassler, B. L., M. Wright, and M. R. Silverman. 1994. Multiple signaling systems controlling expression of luminescence in Vibrio harveyi: sequence and function of genes encoding a second sensory pathway. Mol. Microbiol. 13:273–286. 3. Blehert, D. S., R. J. Palmer, J. B. Xavier, J. S. Ameida, and P. E. Kolenbrander. 2003. Autoinducer 2 production by Streptococcus gordonii DL1 and the biofilm phenotype of a luxS mutant are influenced by nutritional conditions. J. Bacteriol. 185:4851–4860.

J. BACTERIOL. 4. Block, P. J., A. C. Fox, C. Yoran, and A. J. Kaltman. 1973. Actinobacillus actinomycetemcomitans endocarditis: report of a case and review of the literature. Am. J. Med. Sci. 276:387–392. 5. Brogan, J. M., E. T. Lally, K. Poulsen, M. Kilian, and D. R. Demuth. 1994. Regulation of Actinobacillus actinomycetemcomitans expression: analysis of the promoter regions of leukotoxic and minimally leukotoxic strains. Infect. Immun. 62:501–508. 6. Chen, X., S. Schauder, N. Potier, A. Van Dorsselear, I. Pelczer, B. L. Bassler, and F. M. Hughson. 2002. Structural identification of a bacterial quorum sensing signal containing boron. Nature 415:545–549. 7. Cole, S. P., J. Harwood, R. Lee, R. She, and D. G. Guiney. 2004. Characterization of monospecies biofilm formation by Helicobacter pylori. J. Bacteriol. 186:3124–3132. 8. Fong, K. P., L. Gao, and D. R. Demuth. 2003. luxS and arcB control aerobic growth of Actinobacillus actinomycetemcomitans under iron limitation. Infect. Immun. 71:298–308. 9. Fong, K. P., W. O. Chung, R. J. Lamont, and D. R. Demuth. 2001. Intra- and interspecies regulation of gene expression by Actinobacillus actinomycetemcomitans LuxS. Infect. Immun. 69:7625–7634. 10. Freeman, J. A., and B. L. Bassler. 1999. A genetic analysis of the function of LuxO, a two-component response regulator involved in quorum sensing in Vibrio harveyi. Mol. Microbiol. 31:665–677. 11. Freeman, J. A., and B. L. Bassler. 1999. Sequence and function of LuxU: a two-component phosphorelay protein that regulates quorum sensing in Vibrio harveyi. J. Bacteriol. 181:899–906. 12. Hammer, B. K., and B. L. Bassler. 2003. Quorum sensing controls biofilm formation in Vibrio cholerae. Mol. Microbiol. 50:101–104. 13. James, D. E., H. Shao, R. J. Lamont, and D. R. Demuth. 2006. Actinobacillus actinomycetemcomitans ribose binding protein RbsB interacts with cognate and heterologous autoinducer-2 signals. Infect. Immun. 74:4021–4029. 14. Kim, S. J., S. E. Lee, Y. R. Kim, C. M. Kim, P. Y. Ryu, H. E. Choy, S. S. Chung, and J. H. Rhee. 2003. Regulation of Vibrio vulnificus virulence by the LuxS quorum-sensing system. Mol. Microbiol. 48:1647–1664. 15. Lenz, D. H., M. B. Miller, J. Zhu, R. V. Kulkarni, and B. L. Bassler. 2005. CsrA and three redundant small RNAs regulate quorum sensing in Vibrio cholerae. Mol. Microbiol. 58:1186–1202. 16. Lenz, D. H., K. C. Mok, B. N. Lilley, R. V. Kulkarni, N. S. Wingreen, and B. L. Bassler. 2004. The small RNA chaperone Hfq and multiple small RNAs control quorum sensing in Vibrio harveyi and Vibrio cholerae. Cell 118:69–82. 17. Loh, J. T., M. H. Forsyth, and T. L. Cover. 2004. Growth phase regulation of flaA expression in Helicobacter pylori is luxS dependent. Infect. Immun. 72:5506–5510. 18. Loomis, W. D., and R. W. Durst. 1992. Chemistry and biology of boron. Biofactors 3:229–239. 19. Lyon, W. R., J. C. Madden, J. C. Levin, J. L. Stein, and M. G. Caparon. 2001. Mutation of luxS affects growth and virulence factor expression in Streptococcus pyogenes. Mol. Microbiol. 42:145–157. 20. McNab, R., S. K. Ford, A. El-Sabaeny, B. Barbieri, G. S. Cook, and R. J. Lamont. 2003. LuxS-based signaling in Streptococcus gordonii: autoinducer 2 controls carbohydrate metabolism and biofilm formation with Porphyromonas gingivalis. J. Bacteriol. 185:274–284. 21. Merritt, J., F. Qi, S. D. Goodman, M. H. Anderson, and W. Shi. 2003. Mutation of luxS affects biofilm formation in Streptococcus mutans. Infect. Immun. 71:1972–1979. 22. Miller, S. T., K. B. Xavier, S. R. Campagna, M. E. Taga, M. F. Semmelhack, B. L. Bassler, and F. M. Hughson. 2004. Salmonella typhimurium recognizes a chemically distinct form of the bacterial quorum-sensing signal AI-2. Mol. Cell 15:677–687. 23. Neiditch, M. B., M. J. Federle, S. T. Miller, B. L. Bassler, and F. M. Hughson. 2005. Regulation of LuxPQ receptor activity by the quorum sensing signal autoinducer-2. Mol. Cell 18:507–518. 24. Page, M. I., and E. O. King. 1966. Infection due to Actinobacillus actinomycetemcomitans and Haemophilus aphrophilus. N. Engl. J. Med. 275:181–188. 25. Park, M., Q. Li, N. Shcheynikov, W. Zeng, and S. Muallem. 2004. NaBC1 is a ubiquitous electrogenic Na⫹-coupled borate transporter essential for cellular boron homeostasis and cell growth and proliferation. Mol. Cell 15:331– 341. 26. Park, M., Q. S. N. Li, S. Muallem, and W. Zeng. 2005. Borate transport and cell growth and proliferation. Not only in plants. Cell Cycle 4:24–26. 27. Rickard, R. H., R. J. Palmer, Jr., D. S. Blehert, S. R. Campagna, M. F. Semmelhack, P. G. Egland, B. L. Bassler, and P. E. Kolenbrander. 2006. Autoinducer 2: a concentration dependent signal for mutualistic bacterial biofilm growth. Mol. Microbiol. 60:1446–1456. 28. Schauder, S., and B. L. Bassler. 2001. The languages of bacteria. Genes Dev. 15:1468–1480. 29. Schauder, S., K. Showkat, M. G. Surette, and B. L. Bassler. 2001. The LuxS family of bacterial autoinducers: biosynthesis of a novel quorum-sensing signal molecule. Mol. Microbiol. 41:463–476. 30. Schneider, E. 2001. ABC transporters catalyzing carbohydrate uptake. Res. Microbiol. 152:303–310. 31. Slots, J., H. S. Reynolds, and R. J. Genco. 1980. Actinobacillus actinomyce-

VOL. 189, 2007

32. 33.

34.

35.


temcomitans in human periodontal disease: a cross-sectional microbiological investigation. Infect. Immun. 29:1013–1020. Sperandio, V., A. G. Torres, J. A. Giron, and J. B. Kaper. 2001. Quorum sensing is a global regulatory mechanism in enterohemorrhagic Escherichia coli O157:H7. J. Bacteriol. 183:5187–5197. Sperandio, V., J. L. Mellies, W. Nguyen, S. Shin, and J. B. Kaper. 1999. Quorum sensing controls expression of the type III secretion gene transcription and protein secretion on enterohemorrhagic and enteropathogenic Escherichia coli. Proc. Natl. Acad. Sci. USA 96:15196–15201. Taga, M. E., J. L. Semmelhack, and B. L. Bassler. 2001. The LuxS-dependent autoinducer AI-2 controls the expression of an ABC transporter that functions in AI-2 uptake in Salmonella typhimurium. Mol. Microbiol. 42:777– 793. Taga, M. E., S. T. Miller, and B. L. Bassler. 2003. Lsr-mediated transport and processing of AI-2 in Salmonella typhimurium. Mol. Microbiol. 50:1411– 1427.

5565

36. Winzer, K., K. R. Hardie, N. Burgess, N. Doherty, D. Kirke, M. T. G. Holden, R. Linforth, K. A. Cornell, A. J. Taylor, P. J. Hill, and P. Williams. 2002. LuxS: its role in central metabolism and the in vitro synthesis of 4-hydroxy5-methyl-3(2H)-furanone. Microbiology 148:909–922. 37. Xavier, K. B., and B. L. Bassler. 2003. LuxS quorum sensing: more than just a numbers game. Curr. Opin. Microbiol. 6:191–197. 38. Xavier, K. B., S. T. Miller, W. Lu, J. H. Kim, J. Rabinowitz, I. Pelczer, M. F. Semmelhack, and B. L. Bassler. 2007. Phosphorylation and processing of the quorum sensing molecule autoinducer-2 in enteric bacteria. ACS Chem. Biol. 2:128–136. 39. Yoshida, A., T. Ansai, T. Takehara, and H. K. Kuramitsu. 2005. LuxS-based signaling affects Streptococcus mutans biofilm formation. Appl. Environ. Microbiol. 71:2372–2380. 40. Zambon, J. J., J. Slots, and R. J. Genco. 1983. Serology of oral Actinobacillus actinomycetemcomitans and serotype distribution in human periodontal disease. Infect. Immun. 41:19–27.