Biofilm Mode of Growth of Streptococcus intermedius Favored by a Competence-Stimulating Signaling Peptide

Abstract

Gram-positive and gram-negative bacteria use quorum sensing to coordinate population behavior. In several streptococci, quorum sensing mediated by competence-stimulating peptides (CSP) is associated with development of competence for transformation. We show here that a synthetic CSP favored the biofilm mode of growth of Streptococcus intermedius without affecting the rate of culture growth.

In natural ecosystems, microorganisms grow preferentially attached to surfaces, forming matrix-enclosed biofilms (3). In this mode of growth, microorganisms exhibit increased resistance to antimicrobials and to host immune defense mechanisms (15). Both gram-positive and gram-negative bacteria communicate through quorum-sensing (QS) signals to coordinate population behavior. Biofilm maturation and expression of virulence factors are some of several identified QS-controlled behaviors (4, 7, 13). Not surprisingly, QS has emerged as an attractive target for fighting biofilm infections.

The streptococci comprise a diverse group of gram-positive bacteria, ranging from pathogens to commensal organisms that may be involved in opportunistic infections. In the oral cavity, streptococci constitute approximately 60% of the initial biofilm microflora on the tooth surfaces (16). Members of the anginosus group, including Streptococcus intermedius, Streptococcus anginosus, and Streptococcus constellatus, are also found in the gastrointestinal and genitourinary tracts. The anginosus group may be associated with serious purulent infections, with S. intermedius exhibiting tropism for abscesses of the liver and brain (24).

While the QS signals employed by a large number of gram-negative bacteria are homoserine lactones, gram-positive bacteria typically use secreted peptides. The accumulated secreted peptides are generally detected via a sensor kinase receptor, inducing autophosphorylation of the kinase protein. The phosphoryl group is subsequently transferred to a response regulator that triggers, directly or indirectly, the required changes in gene expression.

Bacterial natural transformation is thought to provide a selective advantage by allowing competent cells to acquire new traits, for example, antibiotic resistance, by incorporation of DNA released from other cells. Streptococci have a QS system that regulates natural competence for genetic transformation. The peptide QS signal termed competence-stimulating peptide (CSP) has been identified in various strains of the mitis, anginosus (9, 10), and mutans groups of streptococci (9, 10). In the anginosus group, distinct species commonly encode and respond to identical CSPs. In other groups of streptococci, the CSPs are most often species specific and in several instances are specific to certain groups of strains (9).

The disruption of QS genes in gram-negative bacteria, including Pseudomonas aeruginosa (4), Burkholderia cepacia (7), and Aeromonas hydrophila (13), has been associated with altered biofilm formation. Similar findings have recently been reported for gram-positive bacteria, including the competence QS circuit in Streptococcus mutans (11) and Streptococcus gordonii (12), in which disruption of QS regulatory genes resulted in reduced biofilm formation or altered biofilm architecture. Initiation of a QS response during competence development is influenced by environmental conditions. For S. intermedius, conditions that support endogenous induction of competence development have been described (9, 19). It appears, however, that the growth-dependent requirements for spontaneous competence can in several instances be overcome by addition of synthetic CSP (8). To determine whether the CSP by itself may mediate a response affecting S. intermedius biofilm formation, we examined the role of synthetic CSP under conditions that prevented spontaneous competence. We found that CSP-treated cells formed more biofilm than untreated cells, whereas competence development occurred exclusively in the presence of synthetic CSP.

Environmental influence on S. intermedius biofilm formation and competence development.

We carried out various assays to establish conditions in which S. intermedius formed biofilm in the absence of spontaneous competence, while still responding to addition of synthetic CSP.

The biofilm assay was adapted from the method of O'Toole and Kolter (17), which is based on the ability of bacteria to form biofilms on solid polystyrene surfaces. Overnight cells of S. intermedius NCTC 11324 grown in Todd-Hewitt broth (THB) (Difco Laboratories, Detroit, Mich.) supplemented with yeast extract were diluted 1:200 or 1:20 in trypticase soy broth (TSB) (Becton Dickinson and Company, Sparks, Md.); 500 μl was inoculated into wells of polystyrene flat-bottom 24-well microtiter plates (Nunc, Copenhagen, Denmark). The plates were incubated at 37°C aerobically or in a 5%-CO₂ aerobic atmosphere for 24 h. The amount of biofilm formed in the wells was measured by staining the biofilm cells with 0.1% safranin. After wells were rinsed with distilled H₂O, bound safranin was released from the stained cells with 30% acetic acid, and the absorbance of the solution was measured at 530 nm. Alternatively, the biofilms were washed twice and suspended in 500 μl of fresh TSB by scraping the bottom and lateral walls of the wells. To disperse clumps and chains, the biofilm suspensions were gently sonicated for 10 s at 4°C, and the optical density was measured at 600 nm.

S. intermedius growth and biofilm formation in TSB was negligible under aerobic conditions. When grown in 5% CO₂, S. intermedius formed homogeneous biofilms. The largest amount of biofilm was formed with inoculums of cells diluted 1:200.

To examine whether competence could be initiated by endogenous CSP, the plasmid pVA838 expressing erythromycin resistance (14) was added at the time of inoculation (final concentration, 0.2 μg/ml). After a 24-h incubation, transformants were selected by growth on THB agar plates containing the selective antibiotic. Biofilm cells grown in THB supplemented with horse serum (9, 19) were included as positive controls. Cells grown in THB supplemented with horse serum showed spontaneous competence development, with a transformation yield of 40 transformants/ml, whereas growth in TSB resulted in no transformants. Competence development in TSB could, however, be induced by addition of synthetic CSP as described below. Based on these preliminary observations, growth of cells diluted 1:200 in TSB incubated at 37°C in a 5% CO₂ incubator was chosen to further examine the effect of CSP on biofilm formation.

Synthetic CSP in the nanomolar concentration range enhances S. intermedius biofilm formation and induces competence.

The peptide CSP 11325 (amino acid sequence: NH₂-DSRIRMGFDFSKLFGK-COOH), which induces competence in S. intermedius NCTC 11324, was synthesized (MedProbe) (9). CSP concentrations varying between 0.2 and 100 nM were tested for their effect on biofilm formation and competence development in TSB medium after 24 h. CSP was added immediately after dilution of the cells (1:200). Competence was assayed by exposure of the cells to pVA838 (final concentration, 1 μg/ml) during culture growth and selection of transformants on THB agar containing the selective antibiotic.

A relationship between the amount of CSP and the degree of CSP stimulating activity was observed (Fig. 1a and b). The steep increase in biofilm formation (Fig. 1a) and competence induction (Fig. 1b) in the concentration range between 0.2 and 1.6 nM is consistent with the CSP autoinducing response mechanism.

FIG. 1.

Titration of CSP stimulating activity. (a) Biofilm formation monitored after 24 h by measuring cell density (OD₆₀₀) of biofilm cells resuspended in TSB medium. (b) Competence-stimulating activity, determined as erythromycin-resistant transformants on THB agar containing the selective antibiotic after 24 h of exposure of 1 ml of culture cells to 1 μg of pVA838 and various CSP concentrations. The results express mean values and standard errors for two to four samples from three independent experiments.

We also investigated the proportion of competent cells in the biofilm and planktonic fractions, respectively, upon addition of 100 nM CSP. At 24 h, 0.032% of the cells in the biofilm fraction were transformants, in contrast to only 0.003% of the cells in the planktonic fraction.

The enhancement of biofilm formation by CSP was verified by scanning electron microscopy. Biofilms were grown in TSB or in minimal medium (MM) (6) containing 10 mM glucose as with the biofilm assay described above, except that a polystyrene disk (Nunc) was immersed in each well before inoculation. After 24 h the disks were removed, rinsed with distilled H₂O, and fixed with 2.5% glutaraldehyde in 0.1 M Sørensen buffer. Dehydrated samples were obtained through a series of ethanol rinses and dried at the critical point with liquid CO₂. CSP-treated cells formed more biofilm than untreated cells during growth in both TSB and MM (Fig. 2). Increased chain formation has been associated with the deletion of the CSP-encoding gene in S. mutans (11). It was not, however, apparent in the SEM images whether the CSP influenced chain formation in S. intermedius. Additionally, in liquid culture growth, long chains were observed in both the presence and the absence of CSP by phase-contrast microscopy, with no observable difference in chain size (data not shown).

FIG. 2.

Scanning electron microscopy of S. intermedius biofilms grown in the absence or presence of 100 nM synthetic CSP in TSB (a) or MM (b). Scale bar, 20 μm.

CSP promotes the biofilm mode of growth without affecting the growth rate.

To examine whether enhancement of biofilm formation was related to an effect on bacterial growth rate, the growth rate in liquid culture was monitored in the absence or presence of 100 nM synthetic CSP. Optical density measurements at 600 nm were taken at different time intervals until stationary phase. No effect on the rate of culture growth was observed.

We also determined the relative distributions of planktonic and biofilm cells during growth in the wells after 4, 6, 8, and 24 h (Fig. 3). The 500-μl planktonic fraction was transferred to a cuvette, and the optical density at 600 nm was measured. The biofilm fraction was resuspended in 500 μl of fresh TSB and sonicated as described above, and the optical density was measured at 600 nm. After 4 h, no difference in biofilm formation was detected between growth in the presence or absence of CSP. After 6 h, the biofilm formed in the presence of CSP was 22% higher than without the CSP, and after 8 h, an increase of 29% was observed. After 24 h, an increase of approximately 100% was observed. Notably, the total growth, assessed as the sum of planktonic and biofilm values, was not affected by addition of CSP at any time.

FIG. 3.

S. intermedius growth in the absence or presence of 100 nM synthetic CSP. Cell density of biofilm (gray bars) and planktonic cells (white bars) in the wells after 4, 6, 8, or 24 h of incubation, monitored by measuring the absorbance at 600 nm (OD₆₀₀). Biofilm cells were washed and resuspended to the same volume as the planktonic cells. The results correspond to mean values and standard errors for three independent experiments.

Specificity of the CSP response.

The specificity of the CSP effect on biofilm formation was tested with both S. intermedius and S. mutans. S. intermedius biofilm formation was unaffected by exposure to up to 400 nM S. mutans synthetic CSP 159 (synthesized by MedProbe; amino acid sequence, NH₂-SGSLSTFFRLFNRSFTQALGK-COOH). We also tested whether S. mutans biofilm formation was enhanced by CSP 159 and whether S. intermedius CSP 11325 affected S. mutans biofilm formation. S. mutans LT-11 biofilm formed in the presence of CSP 159 was enhanced by approximately 85% but was unaffected by S. intermedius CSP 11325. Thus, the CSP response was species specific for both S. intermedius and S. mutans.

DNA structural role in S. intermedius biofilm.

DNA has recently been reported as an important component of the P. aeruginosa biofilm structure (25). It was suggested that the origin of the DNA could be a result of active DNA transport from the P. aeruginosa cells into extracellular vesicles. Such a mechanism has not been reported for gram-positive microorganisms. Induction of lysis of a subfraction of the cells has been observed, however, during Streptococcus pneumoniae competence development (21).

To test whether DNA could be an important component of the biofilm formed by S. intermedius, DNase I (20 U/μl; Roche) was added to the initial inoculums at 40, 200, or 400 U/ml (final concentration). The inhibitory effect of DNase I on biofilm formation was observed after growth in both the presence and the absence of CSP (Fig. 4). With the maximum concentration of DNase I (400 U/ml), biofilm formed in the presence of CSP was reduced by 45% (standard error, 9.5) and without CSP by 52% (standard error, 3.8), a difference that was not statistically significant (t test; P = 0.419). The results indicate that DNA may play an important role in the structure of S. intermedius biofilm. We are currently investigating whether the expression of DNA-binding proteins up-regulated by CSP (2) may be involved in the observed increased biofilm formation in response to the CSP.

FIG. 4.

Effect of DNase I on biofilm formation in the absence or presence of 100 nM synthetic CSP. Biofilm formation was monitored by measuring the absorbance of biofilm resuspended cells at 600 nm (OD₆₀₀). DNase I was added at final concentrations of 40, 200, and 400 U/ml. The results correspond to mean values and standard errors for one to three samples from three independent experiments.

The QS-biofilm connection is clear for several bacteria (4, 5, 7, 13). Mutants deficient in QS, however, can still form biofilms, although in smaller amounts or lacking the typical architecture of mature biofilms. For S. gordonii (12) and S. mutans (11), for instance, inactivation of competence-regulatory genes does not abolish the ability of the cells to form biofilms, although the biofilms differ from the wild types. Deletion of regulatory genes involved in QS responses, however, may result in secondary mutations to compensate for disruption of QS regulons, thus confusing interpretation of results (22). In our study we investigated directly the role of the CSP signal in biofilm formation and competence development of the wild-type strain. We found that the respective synthetic CSPs favored the biofilm mode of growth for both S. intermedius and S. mutans, and we confirmed the role of CSP in S. intermedius competence development (9, 19).

For the mitis and anginosus groups of streptococci, the genes encoding the CSP, the kinase receptor, and the cognate response regulator are organized in an operon. For S. mutans, however, the promoter for the CSP-encoding gene is distinct from the promoter for the kinase receptor and the response regulator. Moreover, inactivation of the genes encoding the histidine kinase or the response regulator results in complete abolishment of competence for S. pneumoniae (18), while for S. mutans competence is reduced or not affected (1, 10). This indicates that the mechanisms of regulation may vary among different streptococcal species. Despite such differences, both S. intermedius and S. mutans responded to their respective CSPs by exhibiting increased biofilm formation.

The S. pneumoniae CSP response during competence development is well characterized. Genome-scale studies have recently shown, however, that among the CSP-inducible genes, more than half are dispensable for transformation, while several induced genes are stress related (20). Stress responses are also related to biofilm formation. Environmental conditions, such as nutritional limitation, appear to trigger biofilm formation by several bacterial species (23). We are currently investigating whether S. intermedius exposed to the synthetic CSP may sense stress conditions not actually present in the environment and whether this in turn may favor the biofilm mode of growth. In the biofilm, the cells are probably able to cope more efficiently with stress conditions. Elucidation of how cells communicate to regulate such group behavior may lead to novel strategies for control of bacterial infection.