The Actinobacillus actinomycetemcomitans ribose binding protein RbsB interacts with cognate and heterologous autoinducer 2 signals

Abstract

Autoinducer 2 (AI-2) produced by the oral pathogen Actinobacillus actinomycetemcomitans influences growth of the organism under iron limitation and regulates the expression of iron uptake genes. However, the cellular components that mediate the response of A. actinomycetemcomitans to AI-2 have not been fully characterized. Analysis of the complete genome sequence of A. actinomycetemcomitans (www.oralgen.lanl.gov) indicated that the RbsB protein was related to LuxP, the AI-2 receptor of Vibrio harveyi. To determine if RbsB interacts with AI-2, the bioluminescence of the reporter strain V. harveyi BB170 (sensor 1-, sensor 2+) was determined after stimulation with partially purified AI-2 from A. actinomycetemcomitans or conditioned medium from V. harveyi cultures in the presence and absence of purified six-His-tagged RbsB. RbsB efficiently inhibited V. harveyi bioluminescence induced by both A. actinomycetemcomitans AI-2 and V. harveyi AI-2 in a dose-dependent manner, suggesting that RbsB competes with LuxP for AI-2. Fifty percent inhibition occurred with approximately 0.3 nM RbsB for A. actinomycetemcomitans AI-2 and 15 nM RbsB for V. harveyi AI-2. RbsB-mediated inhibition of V. harveyi bioluminescence was reversed by the addition of 50 mM ribose, suggesting that A. actinomycetemcomitans AI-2 and ribose bind at the same site of RbsB. The RbsB/AI-2 complex was thermostable since A. actinomycetemcomitans AI-2 could not be recovered by heating. This was not due to heat inactivation of A. actinomycetemcomitans AI-2 since signal activity was unaffected by heating in the absence of RbsB. Furthermore, an isogenic A. actinomycetemcomitans mutant that was unable to express rbsB was deficient in depleting A. actinomycetemcomitans AI-2 from solution relative to the wild-type organism. Inactivation of rbsB also influenced the ability of the organism to grow under iron-limiting conditions. The mutant strain attained a cell density of approximately 30% that of the wild-type organism under iron limitation. In addition, real-time PCR showed that the expression of afuABC, encoding a major ferric ion transporter, was reduced by approximately eightfold in the rbsB mutant. This phenotype was similar to that of a LuxS-deficient mutant of A. actinomycetemcomitans that is unable to produce AI-2. Together, our results suggest that RbsB may play a role in the response of A. actinomycetemcomitans to AI-2.

FIG. 1.

Expression of A. actinomycetemcomitans RbsB polypeptide. The rbsB gene was amplified from A. actinomycetemcomitans genomic DNA by use of primers derived from the complete genome sequence of A. actinomycetemcomitans HK1651 (26). Cloning of the amplified product into the vector pQE60, expression, and affinity purification of RbsB protein were done as described in Materials and Methods. (A) Coomassie-stained gel subjected to PAGE. Lane 1, size markers; lane 2, affinity-purified RbsB. (B) Western blot of gel subjected to PAGE (from panel A). RbsB was visualized using anti-six-His monoclonal antibody. The doublet likely arises from incomplete processing of the RbsB signal sequence in the E. coli host. Molecular size markers are labeled on the left.

FIG. 2.

RbsB inhibits V. harveyi BB170 bioluminescence induced by A. actinomycetemcomitans AI-2 (A) and V. harveyi AI-2 (B) in a dose-dependent manner. Purified RbsB protein (0.3 to 5,000 ng per ml) was incubated with partially purified AI-2 from A. actinomycetemcomitans or with conditioned AI medium from an overnight V. harveyi culture for 30 min at 30°C. After subsequent addition of V. harveyi BB170 cells, bioluminescence was measured at hourly intervals by using a Wallac Victor³ microtiter plate reader. The data shown represent the 6-h time point after the addition of AI-2, at which time the positive control (V. harveyi BB170 cells exposed to conditioned AI medium from an overnight V. harveyi culture) exhibited approximately 1,100- to 1,300-fold induction of bioluminescence over that of the negative-control reaction mixture containing sterile AI medium. Reactions were performed in triplicate.

FIG. 3.

(A) Direct competition of RbsB with V. harveyi BB170 cells induced with V. harveyi AI-2. RbsB protein (2.5 μg per ml, •; 5 μg per ml, ▾; or 10 μg per ml, ▪) was added to V. harveyi BB170 cells at the onset of induction of AI-2-mediated bioluminescence (e.g., the 3-h time point in the experiment shown), and light production was monitored at various time points after the addition of RbsB. The positive-control reaction mixture (⧫) did not receive RbsB and exhibited ∼1,200-fold induction of bioluminescence at the 5.5-h time point. (B) RbsB-mediated inhibition of V. harveyi BB170 bioluminescence induced by A. actinomycetemcomitans AI-2 is reversed by the addition of ribose. V. harveyi BB170 cells were mixed with partially purified AI-2 from A. actinomycetemcomitans and incubated at 30°C with shaking. At the onset of the induction of bioluminescence (3 h after the addition of A. actinomycetemcomitans AI-2), RbsB protein alone (2.5 μg per ml, left bar), RbsB (2.5 μg per ml) in the presence of 10 mM ribose (middle bar), or 50 mM ribose (right bar) was added. Light production was subsequently determined at the 5.5-h time point. Positive-control reaction mixtures did not receive RbsB and exhibited approximately 1,000-fold induction of light over levels for the negative control (V. harveyi BB170 induced with sterile AI broth). (C) Influence of ribose on induction of V. harveyi BB170 bioluminescence in the absence of RbsB. V. harveyi BB170 cells were induced with partially purified A. actinomycetemcomitans AI-2 or with A. actinomycetemcomitans AI-2 in the presence of ribose at a final concentration of 0 to 1,000 mM. The induction of bioluminescence was determined for each sample as described in Materials and Methods. All reactions were performed in triplicate.

FIG. 4.

RbsB, AI-2, and the RbsB/AI-2 complex are heat stable. (A) A. actinomycetemcomitans AI-2 was incubated for 30 min at 30°C with purified RbsB (2.5 μg) and then heated for 10 min at 37°C, 55°C, or 65°C. Samples were subsequently tested for induction of V. harveyi BB170 bioluminescence. (B) To assess the heat stability of AI-2, partially purified A. actinomycetemcomitans AI-2 was heated for 10 min at 37°C, 55°C, 65°C, or 100°C and then added to V. harveyi BB170 cells. Bioluminescence was determined as described above. (C) For the RbsB protein, purified RbsB (2.5 μg) was heated for 10 min at 37°C, 50°C, or 65°C and then incubated with partially purified A. actinomycetemcomitans AI-2 for 30 min at 30°C. Induction of V. harveyi BB170 bioluminescence was determined as described in Materials and Methods.

FIG. 5.

Inactivation of the A. actinomycetemcomitans rbs operon influences the kinetics of depletion of AI-2 from solution by bacterial cells. Wild-type (•—•) and Rbs-deficient (○—○) strains of A. actinomycetemcomitans were incubated at 37°C with a solution of partially purified A. actinomycetemcomitans AI-2. At various times up to 30 min, bacterial cells were removed by centrifugation and the supernatant was analyzed in triplicate for the induction of V. harveyi bioluminescence as described in Materials and Methods.

FIG. 6.

Inactivation of A. actinomycetemcomitans rbsB influences aerobic growth under iron limitation and expression of afuA, encoding a ferric iron transporter. (A) Levels of growth of wild-type A. actinomycetemcomitans (wt), isogenic rbsB and luxS mutants (RbsB⁻ and LuxS⁻, respectively), and the rbsB mutant complemented with a plasmid-borne copy of rbsB (RC) were determined under iron-replete conditions (black bars) and in the presence of 100 μM EDDHA (gray bars). Cultures were analyzed at mid-exponential to late exponential phases of growth by determining optical density at 600 nm (O.D._600nm). The results presented are averages from three independent experiments. (B) Expression of afuA was monitored by real-time PCR using gene-specific primers and the fluorophore SYBR green. Fluorescence as a function of cycle number was plotted for reactions using RNA from wild-type A. actinomycetemcomitans (solid line) and an isogenic rbsB mutant (dotted line). Results from a negative-control reaction using RNA from the mutant strain but without reverse transcriptase (RT) are shown by the dashed line. The cycle threshold was calculated from the raw fluorescence data by using the onboard analysis software supplied with the SmartCycler system.