Regulation of the gtfBC and ftf genes of Streptococcus mutans in biofilms in response to pH and carbohydrate

Yunghua Lia,1 and Robert A. Burneb,1

Center for Oral Biology and Department of Microbiology and Immunology, University of Rochester School of Medicine and Dentistry, 601 Elmwood Avenue, Rochester, NY 14642, USA1

Author for correspondence: Robert A. Burne. Tel: +1 352 392 0011. Fax: +1 352 392 7357. e-mail: rburne{at}dental.UFL.EDU


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Streptococcus mutans produces a number of extracellular sucrose-metabolizing enzymes that contribute to the ability of the organism to cause dental caries, including three glucosyltransferases, the products of the gtfB, gtfC and gtfD genes, and a fructosyltransferase, encoded by the ftf gene. To better understand the regulation of the expression of these genes under environmental conditions that more closely mimic those in dental plaque, two strains of S. mutans harbouring fusions of the gtfBC (SMS102) and ftf (SMS101) promoters to a chloramphenicol acetyltransferase (CAT) gene were examined in biofilms formed in vitro. The strains were grown in a Rototorque biofilm reactor in a tryptone-yeast extract-sucrose medium. CAT specific activity in biofilm cells was measured at quasi-steady state or following additions of 25 mM sucrose or glucose, with or without pH control. After approximately 10 generations of biofilm growth, the ftf and gtfBC genes of S. mutans were found to be expressed at levels different from those reported for planktonic cells growing under otherwise similar conditions. The expression of these genes was induced by the addition of sucrose to the quasi-steady-state cultures. Expression of the gtfBC genes was influenced by environmental pH, since CAT specific activities in quasi-steady-state biofilms of strain SMS102 grown without pH control were twice those produced by cells grown with pH control. Moreover, addition of glucose to quasi-steady-state biofilms resulted in increased expression of the gtfBC–cat fusion, although the magnitude of the induction was less than that seen with sucrose. The effect of pH on ftf expression was negligible. A modest and transient induction of ftf was observed in biofilms pulsed with excess glucose and the kinetics and level of induction of ftf by excess carbohydrate were dependent on the pH of the biofilms. This study demonstrates that the type and amount of carbohydrate and the environmental pH have a major influence on transcription of the gtfBC and ftf genes when the organisms are growing in biofilms, and provides evidence for previously undisclosed regulatory circuits for exopolysaccharide gene expression in S. mutans.

Keywords: exopolysaccharides, pathogenesis, glucans, dental caries, dental plaque

Abbreviations: CAT, chloramphenicol acetyltransferase

a Present address: Faculty of Dentistry, University of Toronto, 124 Edward Street, Toronto, Ontario, Canada M5G 1G6.

b Present address: Department of Oral Biology, PO Box 100424, Gainesville, FL 32610-0424, USA.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Streptococcus mutans has been described as an ‘obligate biofilm’ organism (Burne, 1998b ), and is considered to be a principal aetiological agent of dental caries (Hamada & Slade, 1980 ). The ability of S. mutans to initiate caries depends on several virulence-associated traits, including (i) initial adherence to the tooth through high-affinity adhesins, such as the SpaP protein, (ii) the ability to synthesize insoluble, extracellular polysaccharides that enhance accumulation and persistence of the organism on the tooth surfaces, (iii) a high capacity to catabolize carbohydrates and produce acids and (iv) the ability to grow and continue to metabolize carbohydrates at low pH (Burne, 1998a ; Kuramitsu, 1993 ). Among these virulence attributes, the ability of S. mutans to produce extracellular polysaccharides from dietary carbohydrates has been demonstrated to significantly enhance its cariogenicity (Yamashita et al., 1993 ).

S. mutans synthesizes glucan polymers from sucrose via the actions of three secreted glucosyltransferases (GTFs), encoded by the gtfB, gtfC and gtfD genes (Kuramitsu, 1993 ). The gtfB and gtfC genes are in an operon-like arrangement and encode enzymes that produce mainly water-insoluble glucans, whereas the gtfD gene, which is not linked to the gtfBC locus, encodes an enzyme that catalyses the formation of a water-soluble glucan. It is the water-insoluble glucans made by GtfBC that play important roles in adhesion and accumulation of the organisms on the tooth surfaces, and in establishing the extracellular polysaccharide matrix that is responsible for the structural integrity of dental biofilms. These polysaccharides are also thought to provide the organisms with a unique microenvironment for their growth, metabolism and survival (Bowden & Hamilton, 1998 ; Liljemark & Bloomquist, 1996 ; Nakano & Kuramitsu, 1992 ; Yamashita et al., 1993 ). S. mutans produces a single fructosyltransferase (FTF), the product of the ftf gene, which catalyses the synthesis of high-molecular-mass fructans from sucrose (Ebisu et al., 1975 ; Shiroza & Kuramitsu, 1988 ). Fructans produced by S. mutans are believed to function primarily as extracellular storage compounds that can be metabolized during periods of nutrient deprivation (Burne et al., 1996 ).

A number of studies indicate that the expression of the genes for the exopolysaccharide-synthesizing enzymes of S. mutans is dependent on environmental conditions, including growth rate, pH, carbon source, and whether the organisms are attached to the surfaces (Burne et al., 1997 ; Hudson & Curtiss, 1990 ; Kiska & Macrina, 1994 ; Wexler et al., 1993 ). Also, the use of gtf and ftf gene fusion strains in a continuous-flow biofilm fermenter has shown that organisms growing in thicker, mature (7-d-old) biofilms have higher levels of expression of the gtfBC genes compared with cells grown in suspension or in thinner (2-d-old) biofilms (Burne et al., 1997 ; Hudson & Curtiss, 1990 ; Kiska & Macrina, 1994 ; Wexler et al., 1993 ). In contrast, ftf was found to be dramatically down-regulated in 7 d biofilms compared with 2 d biofilms. The specific mechanisms governing regulation of exopolysaccharide synthesis in S. mutans biofilms have yet to be discovered. In this study, we test the hypothesis that pH or changes in carbohydrate source and concentration in biofilms influence expression of gtf and ftf in mature biofilms.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, growth media and biofilm reactor conditions.
The bacterial strains utilized in this study were S. mutans harbouring ftf–cat (SMS101) or gtfBC–cat (SMS102) gene fusions. The construction of these strains and their genotypic stabilities are detailed elsewhere (Burne et al., 1997 ; Hudson & Curtiss, 1990 ; Wexler et al., 1993 ). The stability of the gene fusions in biofilms was again confirmed in this study by quantifying biofilm bacteria on selective and non-selective media, as previously detailed (Burne et al., 1997 ; Li et al., 2000 ). The strains were routinely maintained on brain heart infusion (BHI) agar (Difco) supplemented with 10 µg erythromycin ml-1. For the development of biofilms, a tryptone-yeast extract base medium (TY; Burne et al., 1987 ) was diluted twofold and supplemented with 10 mM sucrose and 10 µg erythromycin ml-1, or with glucose (5 mM) and 10 µg erythromycin ml-1. In addition, the medium contained 50 mM potassium phosphate buffer (pH 7·8) for the experiments requiring pH control, or 90 mM KCl for those not requiring pH control, to ensure that the potassium ion concentration was similar in all cases. Mono-species biofilms were developed in a modified Rototorque biofilm reactor (Burne et al., 1997 ) with a working volume of 0·6 l. The Rototorque is a stirred tank biofilm reactor that can be run in either batch or continuous mode. For this study, continuous culture was used by pumping medium at a fixed dilution rate and the biofilms were formed on slides that were inserted on the inner wall of the vessel. As configured for this study (Li et al., 2000 ), the Rototorque reactor allowed biofilm formation on 12 polystyrene slides, each with a mean exposed surface area of 31 cm2 for biofilm accumulation. After inoculation with 10 ml of an overnight culture of S. mutans SMS101 or SMS102, medium was pumped into the vessel at a constant rate of 60 ml h-1 (‘dilution rate’, D=0·1 h-1). Samples were taken of quasi-steady-state biofilms and of biofilms pulsed with glucose or sucrose. Since bacteria in a biofilm are physiologically heterogeneous, a true steady state is not achieved, so the term quasi-steady state is used here instead. Cultures were designated to have reached quasi-steady state after the equivalent of 10 mean generation times using calculations detailed elsewhere (Burne & Chen, 1998 ). The speed of the rotating inner drum of the Rototorque was kept constant at 75 r.p.m. and the temperature was maintained by immersion of the vessel in a 37 °C circulating water bath.

Assessment of physical characteristics of S. mutans biofilms.
Biofilm dry weights were determined by mechanically dissociating the biofilms from the slides, collecting the material by centrifugation, washing once with dH2O, lyophilizing the sample and weighing. For measurement of total carbohydrate in biofilms, the lyophilized samples were resuspended in 5 ml dH2O and the anthrone method (Dubois et al., 1956 ) was used to measure total carbohydrate using glucose as the standard. Bacterial viability in biofilms was estimated by direct microscopic enumeration and plate counts with cells prepared as follows. Biofilms were collected, washed once, and resuspended in 5 ml of reduced transport fluid (Loesche & Syed, 1973 ) at pH 7·2. The samples were then subjected to a gentle sonication at a setting of 150 W for 20 s to break bacterial chains, and the samples were split into two parts. The cell suspensions for viable cell counts were decimally diluted and directly inoculated by a spiral plating system (Autoplate model 3000; Spiral Biotech) onto BHI agar supplemented with 10 µg erythromycin ml-1. The viable counts were made after the plates were cultivated at 37 °C for 24 h in a 5% CO2 aerobic atmosphere. Microscopic enumeration of bacteria was conducted as described by Koch (1994) using a Petroff–Hausser counting chamber. The ratio of viable cell counts to the counts obtained by visual enumeration was expressed as percentage viable cells. The formation and spatial distribution of biofilms on the surfaces were also assessed by using phase-contrast microscopy, as previously described (Burne et al., 1997 ; Li et al., 2000 ).

In situ measurement of biofilm pH.
The pH of the biofilms was measured as previously detailed (Li et al., 2000 ) using a superminiature, Beetrode pH electrode (model MEPH3L; World Precision Instruments). Briefly, after the cultures reached quasi-steady state (Burne & Chen, 1998 ), slides with biofilms were removed from the vessel and placed on end on a paper towel to allow excess medium to be absorbed from the end of the slides. The micro-reference electrode, which was connected through the Bee-Cal adapter and two cables to the pH probe and a standard pH meter, was positioned in the biofilm to be partially immersed in the biomass. In situ measurement of pH was conducted immediately by placing the tip of the pH probe into biofilms and a series of pH readings was recorded from a minimum of 30 different sites selected at random.

Carbohydrate pulsing and biofilm sampling.
Quasi-steady-state biofilms, or biofilms which had been pulsed with either glucose or sucrose (25 mM) immediately after initial biofilm sampling (T0), were used for measurements of chloramphenicol acetyltransferase (CAT) activity expressed from the gtf or ftf promoters. Subsequently, the biofilms were sampled by removing three slides at 15, 30 and 60 min after the carbohydrate pulse. The biofilms were mechanically dissociated from the slides by scraping with a sterile razor blade into 40 ml ice-cold Tris/HCl (10 mM, pH 7·8) and were centrifuged at 8000 g for 10 min at 4 °C. The cell pellets were washed twice and resuspended in 1 ml of the same buffer for the preparation of cell-free lysates. The pH profiles in the liquid phase at each sampling time were recorded by measuring pH in 10 ml of culture fluid.

Preparation of cell-free lysates and CAT assays.
Cell-free lysates were prepared for the analysis of CAT activity as previously described (Chen et al., 1998 ). Briefly, the cell pellets were resuspended in 1 ml Tris buffer in a 2 ml, screw-cap microcentrifuge tube (Sarstedt). One-third volume of pre-chilled (-20 °C) glass beads (0·2 mm mean diameter) was added to the sample, and the cells were homogenized using a Bead Beater (Biospec Products) in four 30 s intervals, with cooling on ice in the intervals. The lysates were centrifuged at 12000 g at 4 °C for 15 min and the supernatant material was collected for CAT assays. Protein concentrations of the cell-free lysates were determined by the method of Bradford (1976) using a commercially available reagent (Bio-Rad). CAT activity was measured by a spectrophotometric method (Shaw, 1979 ) with use of the colorimetric substrate 5,5'-dithio-bis-nitrobenzoic acid (DTNB; Boehringer Mannheim). All assays were performed in triplicate with internal standards containing all reagents except chloramphenicol to account for chloramphenicol-independent reduction of DTNB. One unit of CAT activity was defined as the amount of enzyme needed to catalyse the acetylation of 1 µmol chloramphenicol min-1 (mg protein)-1.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Physical characteristics of S. mutans biofilms at quasi-steady state
The biofilms at quasi-steady state were indistinguishable from biofilms of S. mutans and Streptococcus salivarius described previously (Burne et al., 1997 ; Li et al., 2000 ) (data not shown) and consisted of many microcolonies, aggregates and mats, with a considerable amount of extracellular polymer. As was observed with S. salivarius biofilm formation (Li et al., 2000 ), biofilms of S. mutans formed more efficiently at low pH and the proportion of cells in the planktonic phase was far lower in cultures grown at lower pH values. In the cultures with buffered medium (50 mM potassium phosphate buffer, pH 7·8), the dry weight of the biofilms was 180 µg cm-2 (Table 1). When the medium was not buffered, the biomass of the biofilms was 50% greater than that of the biofilms formed in buffered medium. Analysis of total carbohydrates showed that the biofilms formed without pH control contained a higher percentage (68%) of carbohydrates than those biofilms formed with pH control (49%), and the total number of cells in the biofilms grown with pH control was about half that of biofilms formed without pH control. S. mutans SMS102 showed no significant differences from strain SMS101 in its growth characteristics, final yields or appearance in the biofilm reactor when grown under the same culture conditions (data not shown).


View this table:
[in this window]
[in a new window]
 
Table 1. Physical characteristics of the biofilms of recombinant S. mutans SMS101 growing in sucrose-defined media at quasi-steady state

 
Similar to Rototorque biofilms of S. salivarius (Li et al., 2000 ), the S. mutans biofilms were heterogeneous and achieved thicknesses ranging from 10 to over 300 µm, and significant differences were noted between the pH values measured in the fluid phase and in the biofilms (Table 2). When pH control was imposed by buffering of the medium, the pH of the planktonic phase was near neutrality (pH 6·68±0·15), because the production of acid by the organisms always accompanies growth, and the pH measured within the biofilms was about 0·6 units lower. In the cultures without pH control, the in situ pH of the biofilms was slightly higher than the pH measured in the fluid phase, probably due to the cells acting as an effective buffer, and 0·8 pH units lower than in the biofilms formed with buffered medium. Therefore, the use of buffered medium was again an effective means to generate populations of biofilm cells exposed to a significant difference in pH, and the effects of pH on exopolysaccharide gene expression could be reliably examined.


View this table:
[in this window]
[in a new window]
 
Table 2. In situ pH profiles in the biofilms of recombinant S. mutans strains growing in sucrose-defined media at quasi-steady state

 
CAT specific activity in biofilm cells at ‘quasi-steady state’ and following a carbohydrate pulse
Biofilm cells were cultivated in a sucrose-containing medium to quasi-steady state at a dilution rate of 0·1 h-1 for a minimum of a calculated 10 generations, i.e. >69 h (Burne & Chen, 1998 ). CAT activities were measured immediately before addition of sucrose to a final concentration of 25 mM and at 15, 30 and 60 min after the addition of sucrose (Table 3). Addition of sucrose in all cases resulted in a steady decline in the pH of the planktonic phase (Table 3). At quasi-steady state, CAT activity in the SMS101 (ftf–cat) strain that was cultivated at the lower pH was always modestly lower than in biofilm cells grown with pH control, whereas SMS102 (gtfBC–cat) showed about a twofold increase of CAT when grown at lower pH values. Expression of the ftf and gtfBC genes was induced more than two- and fourfold, respectively, by the addition of sucrose to the steady-state cultures grown in buffered medium. An increase in CAT activity of about threefold in both SMS101 and 102 grown without pH control was seen after addition of sucrose. In all cases, there were differences in the kinetics of the induction between the gtf and ftf gene fusions and the induction pattern was different depending on whether the strains were cultivated with pH control or not. CAT activity in the gtf gene fusion strain increased steadily over a 1 h period following sucrose addition to biofilms grown in unbuffered medium. In contrast to the behaviour of SMS102, the induction of the ftf gene fusion in SMS101 by sucrose occurred within 15 min and was generally followed by a decline in CAT, the rate of which depended on the growth conditions. Notably, the rate of decline of CAT activity in SMS101 following a sucrose pulse was slower than that previously reported for strain SMS101 in biofilms (Burne et al., 1997 ), perhaps because of differences in the composition of the culture media or the time allowed for formation of biofilms in this study.


View this table:
[in this window]
[in a new window]
 
Table 3. CAT specific activity of recombinant S. mutans strains growing in in vitro biofilms under sucrose limitation and following a 25 mM sucrose pulse

 
To determine whether the limiting growth carbohydrate in steady-state biofilms could affect the expression of gtf or ftf, biofilms were formed in the presence of sucrose for 10 vessel volume changes, and then for an additional 10 vessel volume changes in the base medium containing 5 mM glucose. When biofilms were formed in buffered medium, the level of CAT activity expressed by these strains at quasi-steady state, before addition of sucrose, was essentially the same as that observed for biofilms growing with sucrose as the limiting carbohydrate. After addition of 25 mM sucrose to the glucose-limited biofilms, the patterns for induction of ftf and gtfBC genes were essentially similar to those patterns observed when quasi-steady-state, sucrose-limited biofilms were pulsed with sucrose (Table 4). Repeating this experimental strategy with glucose-limited biofilms formed in the absence of pH control (Table 4) yielded results essentially identical to those seen with biofilms formed in sucrose (Table 3). In all cases then, and in contrast to when cultures were exposed to excess sucrose, growth with sucrose as the limiting carbohydrate did not result in higher gtf or ftf expression when compared with glucose-limited biofilms. Moreover, growth in limiting concentrations of glucose did not change the ability of the strains to respond to induction by sucrose.


View this table:
[in this window]
[in a new window]
 
Table 4. CAT specific activity of recombinant S. mutans strains growing in in vitro biofilms with glucose limitation following a 25 mM sucrose pulse

 
Finally, quasi-steady-state biofilms were formed in a sucrose-containing medium followed by a 25 mM glucose pulse (Table 5). Interestingly, the CAT specific activity of strain SMS102 (gtfBC–cat) increased about fourfold when glucose was added to biofilms growing in buffered medium. CAT specific activities of strain SMS102 also increased about twofold when glucose was added to biofilms growing without pH control. In the case of S. mutans SMS101 growing with pH control, there was little increase in the level of CAT activity after addition of glucose to the vessel. However, in SMS101 biofilms grown without pH control, transcription of ftf increased a little more than twofold after glucose addition, and then rapidly declined to baseline levels within 1 h.


View this table:
[in this window]
[in a new window]
 
Table 5. CAT specific activity of recombinant S. mutans strains growing in in vitro biofilms under sucrose limitation and following a 25 mM glucose pulse

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
S. mutans in dental biofilms is subjected to a variety of different environmental conditions, most notably sudden changes in the type and amount of nutrients, accompanied by rapid and significant fluctuations in pH (Burne, 1998a ; Carlsson, 1984 ). The ability of this organism to sense and respond to changes in the environment is crucial to its survival and cariogenicity (Bowden & Hamilton, 1998 ). In a previous study, significant differences in expression of gtf and ftf were observed in mature biofilms versus relatively thinner biofilms or planktonic cells (Burne et al., 1997 ). One possible explanation for the differences in expression of the exopolysaccharide-synthesizing machinery of S. mutans in thicker biofilms is that, because of mass transport limitations, the pH and carbohydrate concentrations experienced by the organisms in the thick biofilms differ significantly from those to which planktonic cells or thinner biofilms are exposed. Results presented in this communication support the idea that expression of the ftf and gtfBC genes in biofilms is strongly influenced by carbohydrate availability, the type of carbohydrate, and the environmental pH.

A number of new findings have emerged from this study. The first is that environmental pH clearly influences the expression of the gtfBC genes of S. mutans growing in biofilms. Biofilms grown without pH control, which had biofilm pH values of around 5·3, consistently had twofold more CAT activity expressed from the gtfBC promoter than SMS102 cultivated with pH control. Moreover, following a carbohydrate pulse, the level of expression of gtfBC in the biofilms without pH control was significantly higher than in biofilms with pH control. Since increased production of GtfBC would likely result in a greater proportion of carbohydrate being incorporated into biofilm polysaccharides, this observation is consistent with the finding that the biofilms formed at low pH were composed of a much greater percentage of carbohydrate than those formed at a more neutral pH value. Also, a previous study (Wexler et al., 1993 ) showed that when S. mutans was grown in steady-state continuous chemostat culture, lowering of the pH from 7 to 6 resulted in about a 50% increase in the expression of the gtfBC genes. Interestingly, the level of expression of gtfBC achieved by suspended populations at steady state or after a sucrose pulse (Wexler et al., 1993 ) was lower than the level of expression seen in biofilms at quasi-steady state or after sucrose induction in this study. Thus, one possible explanation for the enhanced gtfBC expression in biofilms may be due to the relatively lower pH values achieved in biofilms (Table 2), compared to the continuous culture studies, or to the existence of low pH microenvironments in which gtf expression is markedly elevated. A second significant, and possibly related, finding is that gtfBC expression in biofilms was induced by addition of excess glucose to the culture, albeit the level of induction was not as great as that seen with sucrose. There is no logical reason for glucose, which is not a substrate for Gtfs, to act as a specific inducer of gtfBC. Instead, it is more likely that the gtfBC genes are induced in response to acidification of the biofilms or in response to the presence of an excess of a metabolizable carbohydrate. The latter could be signalled by directly sensing carbohydrate flow through the PTS or the glycolytic pathway, or could reflect a response to increased growth rate, although in continuous culture we found gtfBC expression was inversely correlated with how fast the cells were growing (Wexler et al., 1993 ). Of note, there does not seem to be a hierarchy of control of expression of gtfBC by carbohydrate and pH, since induction by excess carbohydrate occurs whether the cells are growing with or without pH control. Possibly, then, the effects of pH and carbohydrate are exerted through different pathways. Finally, we cannot exclude the idea that differences in phosphate concentrations in the buffered and unbuffered cultures also had an impact on the results.

Another important finding was that growth of the biofilms under quasi-steady-state conditions with glucose as the limiting carbohydrate did not result in down-regulation of gtf or ftf expression compared to cells growing with sucrose as the limiting carbohydrate. It has been suggested that sucrose is a specific inducer of exopolysaccharide-producing enzymes, but if that were the case, then one might expect higher levels of CAT activity expressed at T0 in cells growing in sucrose (Tables 3 and 5) versus CAT activities at T0 in biofilms being fed on glucose (Table 4), which was not the case. It could be that high levels of sucrose are required for induction, although this would not explain why both ftf and gtf can be induced by addition of either glucose or sucrose. These findings emphasize the possibility that lowering of the pH following addition of carbohydrates or that an increase in the availability of carbohydrate, perhaps sensed via the PTS or glycolytic intermediates, are mechanisms by which gtf and ftf gene expression could be regulated. In fact, we now have preliminary evidence (unpublished) from continuous chemostat culture of these strains at different pH values, under carbohydrate-limiting and carbohydrate-excess conditions, that not only pH, but the amount of carbohydrate available to cells is a factor that influences both ftf and gtf gene expression.

The expression pattern of ftf is more complex than that of gtfBC. As with gtf, the expression levels of ftf in quasi-steady-state biofilms are the same regardless of whether glucose or sucrose was the limiting carbohydrate. In contrast, ftf expression in the lower pH biofilms was not enhanced above that in biofilms formed with buffered medium. As seen previously with both suspended (Wexler et al., 1993 ) and adherent (Burne et al., 1997 ) populations of S. mutans, sucrose was an efficient inducer of ftf gene expression. In this study, we found that induction by sucrose occurred regardless of whether the cells were growing at the lower or higher pH value, or whether the steady-state biofilms were formed with glucose or sucrose as the limiting carbohydrate. However, addition of glucose to quasi-steady-state biofilms had only a modest effect on induction of ftf. In continuous chemostat culture, increasing the growth rate of strain SMS101, or lowering of the pH from 7 to 6, enhanced ftf transcription (Wexler et al., 1993 ), so these factors were likely to have had an effect on ftf transcription after addition of excess carbohydrate. Still, in all cases, the responses of ftf to added carbohydrate, i.e. the rate of induction and the subsequent decline in CAT activity, differed as a function of whether the culture medium was buffered or not. Thus, global factors regulating gene expression in relation to carbohydrate flow or pH may overlap with the regulatory networks that control ftf transcription.

Another notable finding is that expression from the gtfBC promoter in biofilms in this study was higher than that observed in cells grown in continuous chemostat culture (Wexler et al., 1993 ), although not as high as in mature (7 d) biofilms (Burne et al., 1997 ). For example, values in the chemostat were reported to range from 0·017 to 0·049 U (mg protein)-1 under steady-state conditions, and levels after sucrose induction were as high as 0·15 U (mg protein)-1. It can be inferred from the results obtained here that the existence of low pH microenvironments and perhaps the presence of extracellular storage polysaccharides that increase the amount of carbohydrate available to subpopulations of cells within the biofilm may account for the apparent stimulation of gtf expression in biofilms (Burne et al., 1997 ). Low growth rate microenvironments could also exist and impact gtfBC expression. We also have noted that ftf expression in mature (7 d) biofilms is almost completely repressed. Interestingly, the level of CAT expressed from the ftf gene fusion in quasi-steady-state biofilms (Tables 3–5) was about one-third of that in suspended cells in steady-state continuous chemostat culture (Wexler et al., 1993 ). Also, following induction of ftf in the chemostat using sucrose, levels of CAT peaked at about 0·45 U (Wexler et al., 1993 ), whereas the peak expression noted here was about 0·33 U. So again, the trend is that ftf expression in biofilms is partially repressed compared with suspended populations. Recently, we have also found evidence that growth at pH 5 leads to nearly complete repression of ftf (C. Browngardt & R. A. Burne, unpublished), so low pH in the biofilms may account in part for lowered expression of ftf in biofilms.

Clearly, pH and the source and amount of carbohydrate influence the transcription of the exopolysaccharide machinery of S. mutans in biofilms. However, a number of other factors could also account for altered gtf and ftf expression in mature biofilms. The first is that the availability of other nutrients and oxygen may be altered compared to young biofilms or planktonic cells. Goodman & Gao (2000) have found that gtf expression occurs at specific growth phases and have suggested a possible cell-density-dependent component regulating gtfBC, raising the possibility that intercellular signalling may be involved in regulating gtf expression. Although the results we have collected thus far do not fully support that quorum sensing is a regulator of exopolysaccharide synthesis in S. mutans, a luxS homologue is identifiable in the S. mutans chromosome, and peptide-mediated intercellular signalling is also established in S. mutans (Li et al., 2001 ). Therefore, the idea that exopolysaccharide production is regulated by quorum sensing mechanisms cannot yet be excluded. In summary, this study highlights the need for additional research with suspended and adherent oral streptococci to determine whether the changes in the expression patterns of genes of S. mutans growing in biofilms are restricted to spatially isolated subpopulations or to pathways that globally regulate biofilm gene expression in densely packed populations.


   ACKNOWLEDGEMENTS
 
We would like to thank Dr Margaret Chen and Dr José Lemos for a critical evaluation of this manuscript. This work was supported by grant number DE12236 and DE13239 from the National Institute of Dental Research.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bowden, G. H. & Hamilton, I. R. (1998). Survival of oral bacteria. Crit Rev Oral Biol Med 9, 54-85.[Abstract]

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248-254.[Medline]

Burne, R. A. (1998a). Oral streptococci: products of their environment. J Dent Res 77, 445-452.[Abstract]

Burne, R. A. (1998b). Regulation of gene expression in adherent populations of oral streptococci. In Microbial Pathogenesis: Current and Emerging Issues , pp. 55-70. Edited by D. J. LeBlanc, M. S. Lantz & L. M. Switalski. Indianapolis, IN:Indiana University Press.

Burne, R. A. & Chen, Y. M. (1998). The use of continuous flow bioreactors to explore gene expression and physiology of suspended and adherent populations of oral streptococci. Methods Cell Sci 20, 181-190.

Burne, R. A., Schilling, K., Bowen, W. H. & Yasbin, R. E. (1987). Expression, purification, and characterization of an exo-beta-D-fructosidase of Streptococcus mutans. J Bacteriol 169, 4507-4517.[Medline]

Burne, R. A., Chen, Y. M., Wexler, D. L., Kuramitsu, H. & Bowen, W. H. (1996). Cariogenicity of Streptococcus mutans strains with defects in fructan metabolism assessed in a program-fed specific-pathogen-free rat model. J Dent Res 75, 1572-1577.[Abstract]

Burne, R. A., Chen, Y. M. & Penders, J. E. (1997). Analysis of gene expression in Streptococcus mutans in biofilms in vitro. Adv Dent Res 11, 100-109.[Medline]

Carlsson, J. (1984). Regulation of sugar metabolism in relation to the ‘feast-and famine’ existence of plaque. In Cariology , pp. 205-211. Edited by B. Guggenheim. Basel:Karger.

Chen, Y. M., Weaver, C. A., Mendelsohn, D. R. & Burne, R. A. (1998). Transcriptional regulation of the Streptococcus salivarius 57.I urease operon. J Bacteriol 180, 5769-5775.[Abstract/Free Full Text]

Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A. & Smith, F. (1956). Colorimetric method for the determination of sugar and related substances. Anal Chem 28, 350-356.

Ebisu, S., Keijiro, K., Kotani, S. & Misaka, A. (1975). Structural differences in the fructans elaborated by Streptococcus mutans and Strep. salivarius. J Biochem 78, 879-887.[Abstract]

Goodman, S. D. & Gao, Q. (2000). Characterization of the gtfB and gtfC promoters from Streptococcus mutans GS-5. Plasmid 43, 85-98.[Medline]

Hamada, S. & Slade, H. D. (1980). Biology, immunology, and cariogenicity of Streptococcus mutans. Microbiol Rev 44, 331-384.

Hudson, M. C. & Curtiss, R.III (1990). Regulation of expression of Streptococcus mutans genes important to virulence. Infect Immun 58, 464-470.[Medline]

Kiska, D. L. & Macrina, F. L. (1994). Genetic regulation of fructosyltransferase in Streptococcus mutans. Infect Immun 62, 1241-1251.[Abstract]

Koch, A. L. (1994). Growth measurement. In Methods for General and Molecular Bacteriology , pp. 249-276. Edited by P. Gerhardt, R. G. E. Murray, W. A. Wood & N. R. Krieg. Washington, DC:American Society for Microbiology.

Kuramitsu, H. K. (1993). Virulence factors of mutans streptococci: role of molecular genetics. Crit Rev Oral Biol Med 4, 159-176.[Abstract]

Li, Y. H., Chen, Y. M. & Burne, R. A. (2000). Regulation of urease gene expression by Streptococcus salivarius growing in biofilms. Environ Microbiol 2, 169-177.[Medline]

Li, Y. H., Lau, P. C., Lee, J. H., Ellen, R. P. & Cvitkovitch, D. G. (2001). Natural genetic transformation of Streptococcus mutans growing in biofilms. J Bacteriol 183, 897-908.[Abstract/Free Full Text]

Liljemark, W. F. & Bloomquist, C. (1996). Human oral microbial ecology and dental caries and periodontal diseases. Crit Rev Oral Biol Med 7, 180-198.[Abstract]

Loesche, W. J. & Syed, S. A. (1973). The predominant cultivable flora of carious plaque and carious dentine. Caries Res 7, 201-216.[Medline]

Nakano, Y. J. & Kuramitsu, H. K. (1992). Mechanism of Streptococcus mutans glucosyltransferase: hybrid-enzyme analysis. J Bacteriol 174, 5639-5646.[Abstract]

Shaw, W. V. (1979). Chloramphenicol acetyltransferase activity from chloramphenicol-resistant bacteria. Methods Enzymol 43, 737-755.

Shiroza, T. & Kuramitsu, H. K. (1988). Sequence analysis of the Streptococcus mutans fructosyltransferase gene and flanking regions. J Bacteriol 170, 810-816.[Medline]

Wexler, D. L., Hudson, M. C. & Burne, R. A. (1993). Streptococcus mutans fructosyltransferase (ftf) and glucosyltransferase (gtfBC) operon fusion strains in continuous culture. Infect Immun 61, 1259-1267.[Abstract]

Yamashita, Y., Bowen, W. H., Burne, R. A. & Kuramitsu, H. K. (1993). Role of the Streptococcus mutans gtf genes in caries induction in the specific-pathogen-free rat model. Infect Immun 61, 3811-3817.[Abstract]

Received 19 April 2001; revised 4 June 2001; accepted 8 June 2001.