OUP user menu

Effects of mutating putative two-component systems on biofilm formation by Streptococcus mutans UA159

Samir P. Bhagwat , Julia Nary , Robert A. Burne
DOI: http://dx.doi.org/10.1111/j.1574-6968.2001.tb10952.x 225-230 First published online: 1 December 2001


Streptococcus mutans has at least six pairs of open reading frames that are homologous to bacterial two-component regulatory systems. Putative response regulators from five out of six of these pairs were successfully mutated by insertion of a kanamycin resistance marker and the effects of inactivation of the genes on the ability of the cells to form biofilms in an in vitro model were assessed. Disruption of the response regulators of four systems had no effect on biofilm formation, whereas disruption of one response regulator caused a substantial decrease in biofilm formation as compared to the wild-type S. mutans.

  • Biofilm
  • Dental caries
  • Gene regulation
  • Environmental sensing

1 Introduction

Two-component signal transduction systems play important roles in bacterial adaptation and virulence by sensing changes in the environment and modulating gene expression in response to a variety of stimuli [1]. Two-component systems consist of a sensor, or histidine protein kinase, which is anchored to the cytoplasmic membrane, and an effector, or response regulator (RR), which is a DNA binding protein that modulates the expression of certain target genes. Changes in the environment are sensed by the histidine kinase, the signal is transmitted to the RR in the form of phosphorylation–dephosphorylation of certain amino acid residue(s), and the RR modulates expression of specific genes to bring about the adaptive response. Two-component systems are involved in the regulation of diverse metabolic processes such as chemotaxis, sporulation, antibiotic/bacteriocin production, osmoregulation and pathogenicity [2]. It has been shown that some two-component systems may be essential for the survival of the cell [3,4]. Because of their importance in the regulation of cellular metabolism, it has been suggested that they be used as targets to develop novel antibiotics [3].

It has become clear in recent years that bacteria in nature usually exist not as free-floating cells, but as biofilms, which are communities of organisms adhering to surfaces [5,6]. Bacteria in biofilms are in a profoundly different state than the bacteria in the planktonic state (free-living) [5,6]. Bacterial cells often change their gene expression profiles in response to becoming a member of a biofilm community [7,8], leading to the idea that specific genetic pathways are instrumental in differentiation of bacteria from the planktonic to the biofilm state, and vice versa. Consequently, there has been considerable effort in recent years to dissect the genetic pathways leading to either the biofilm or planktonic stage of life [5,6]. Infections mediated by biofilms are of great medical and industrial importance [6], and one such biofilm infection is dental plaque that causes dental caries and involves Streptococcus mutans[7].

In many cases, two-component regulatory systems are either implicated in or have been directly shown to regulate biofilm development [9,10]. There have been systematic attempts to mine recently sequenced bacterial genomes for homologs of two-component bacterial systems [1,11,12] and study their phenotypes. We searched the nearly completed genome of S. mutans for homologs of two-component signal transduction systems and found at least six pairs of open reading frames (ORFs) homologous to known bacterial histidine kinases and RRs. This communication describes the effect of mutating putative RRs of S. mutans on biofilm formation.

2 Materials and methods

2.1 Bacterial strains and media

S. mutans UA159 [13] was used in all the experiments to test biofilm formation. Escherichia coli DH10B (Life Technologies, MD, USA) and JM107 were used for general cloning purposes. Streptococci were grown in brain heart infusion (BHI) broth or on agar (1.5% w/v). For biofilm formation, cells were grown in BM medium [9]. E. coli strains were grown in L broth or L agar. The following antibiotics were used as needed: ampicillin (Ap) at 100 μg ml−1 or kanamycin (Km) at 50 μg ml−1 for E. coli and 1 mg ml−1 for S. mutans.

2.2 Molecular biology techniques

Standard molecular biology techniques were used for cloning, plasmid DNA preparation, restriction analysis and testing of the mutants by Southern blot hybridization. The Qiagen mini plasmid DNA kit (Qiagen, Chatsworth, CA, USA) was used whenever DNA of higher purity was desired. Individual genes were amplified with a pair of gene-specific primers, and in some cases, with sequences at the end such that specific restriction sites were added to the PCR products or by recombinant PCR. PCR products were purified by Qiagen PCR purification kit (Qiagen), digested with appropriate enzymes and ligated to similarly-digested pGEM7 (Promega, Madison, WI, USA). E. coli JM107 or DH10B were transformed with the ligation mix and plated on L medium containing Ap and X-Gal. White colonies were screened for inserted DNA and plasmid DNA from all positive clones was used for further cloning after confirming the identity of the clones by sequencing (ACGT, IL, USA). The RR gene in each clone was opened by a restriction site unique to the clone (HpaI for tcaR, Bsu36I for tcbR, EcoNI for tcdR, PstI for tceR and BglII for tcfR; Table 1), the ends were blunted by treatment with either Klenow fragment (Life Technologies) or T4 DNA polymerase (Life Technologies), and the fragment was ligated to a blunt-ended, gel-purified Km resistance determinant flanked by transcription terminators (ΩKm) [14]. E. coli was transformed with the ligation mix and plated on L agar containing Km. The presence of the disrupted RR gene was confirmed by restriction analysis and DNA from positive clones was selected for transformation into S. mutans UA159.

View this table:
Table 1

Putative two-component response regulators of S. mutans UA159

MutantHomologous toAccession number
tcaRCiaR of S. pneumoniaeCAB54564
tcbRSpiR2 of S. pneumoniaeCAC10170
tccR (unable to obtain mutant)VicR of S. pyogenesCAB65451
tcdRScnR (a response regulator homolog) of S. pyogenesT09003
tceRA putative two-component response regulator of S. pyogenesAAK34393
tcfRLlrG of Lactococcus lactis subsp. lactisAAK05845

Streptococcal transformations were performed as described elsewhere [7]. After the recovery phase, the cells were plated on media containing Km. Chromosomal DNA was prepared from selected clones and verification of gene inactivation was done by Southern blot hybridization. The blots were hybridized to α-32P-labeled ΩKm probe or a gene-specific probe. The bands were visualized by autoradiography.

2.3 Growth and biofilm assays

Biofilm assays were performed as follows. Overnight cultures of S. mutans UA159 and its corresponding mutants were inoculated into the BHI medium from a single colony and grown overnight at 37°C in a 5% CO2 atmosphere. In the morning, the cultures were diluted 1:25 in BHI and the growth was monitored at 600 nm. When the cultures reached OD600 nm of 0.5–0.6 these cultures were diluted 1:100 in BM medium and the diluted culture was distributed in several wells of a 96-well polystyrene, flat-bottom microtiter plate (Corning, New York, NY, USA). Microtiter plates were incubated for 20–24 h and then processed to quantitate biofilm formation. Briefly, to remove planktonic and loosely bound cells, the plates were rotated at 200 rpm for 5 min on a Shaker20 (National Labnet Company, NJ, USA). The BM medium was removed by inverting the plate and shaking gently. The wells were washed once with 200 μl of water, then 0.1% w/v of crystal violet (50 μl) was added to each well and the plates were incubated for 15 min at room temperature. After removing the crystal violet, the wells were washed two times with 200 μl of water and air-dried. Biofilm formation was quantitated by extracting the remaining crystal violet two times with 200 μl of ethanol. The samples were diluted to a final volume of 1 ml with 99% ethanol and the absorbance was measured at 575 nm. The wells with uninoculated BM medium were processed the same way and were used as negative controls.

2.4 Determination of growth of bacteria in broth

Overnight cultures of bacteria in BHI medium were diluted 1:100 in the BM medium and incubated at 37°C in a 5% CO2 atmosphere. The growth was monitored at 600 nm using a Spectronic 20 spectrophotometer (Milton Roy Company, PA, USA).

3 Results and discussion

The S. mutans genome sequence, which is nearly complete (http://www.genome.ou.edu/smutans.html), was searched for putative members of two-component regulatory systems and six such pairs, designated tcaRKtcfRK (for two-component system A, etc.) (see Table 1 and Fig. 1), were found. The RR (R) and the histidine kinase (K) encoded by tcaRK were most similar to the RR CiaR and the cognate histidine kinase CiaH, which are involved in the development of competence, as well as susceptibility to penicillin, in Streptococcus pneumoniae[15]. The RR and histidine kinase encoded by tcbRK were most similar to the products of the genes iR2 and iH of S. pneumoniae, respectively [16]. TccR and TccK were similar to the VicR and VicH proteins of Streptococcus pyogenes, as well as to YycF and YycG of Bacillus subtilis and Staphylococcus aureus[3,4], which are essential for growth. TcdR and TcdK were most similar to the ScnR and ScnH of S. pyogenes, which may be regulators of lantibiotic production in that bacterium [17]. TceR and TceK were most similar to the members of a putative two-component system of unknown function in S. pyogenes and those of the system encoded by tcfRK were most similar to members of the llr two-component system of Lactococcus lactis subsp. lactis[17]. Genes flanking the two-component system genes were annotated as detailed in Fig. 1.

Figure 1

Genetic maps of six putative two-component systems of S. mutans UA159. Genetic maps of the six putative two-component systems are shown. Different regions of the S. mutans genome were analyzed by the MacVector package for the presence of ORFs equal to or greater than 100 amino acids. Protein sequences specified by those ORFs were searched for similarities with known two-component systems using the BLAST program. The arrows reflect the direction of transcription. The upstream and downstream ORFs are designated by numbers which are shown above each ORF and their homologies to known proteins are listed below. (A) System tca. 1: putative phosphate uptake regulatory protein of S. pyogenes, phoU (AAK34095). 2: Aminopeptidase N (Lys-AP) from Lactococcus lactis subsp. lactis (Q48656). tcaR: putative RR. tcaK: putative histidine kinase. 3: Conserved hypothetical protein from S. pyogenes (AAK34089). 4: pyrimidine nucleoside phosphorylase from L. lactis subsp. lactis, pdp (AAK05540). 5: deoxyribose phosphate aldolase of S. mutans (AAK28415). (B) System tcb. 1: putative phosphoglycerate dehydrogenase (BAA88823). 2: sakacin A production RR (BAA88824). 3: DedA protein of E. coli (P09548). tcbR: putative RR. tcbK: putative histidine kinase. 4: no homolog found. 5: voltage-dependent L-type calcium channel α15 subunit of chicken (O42398). 6: no homolog found. (C) System tcc. tccR: putative RR. tccK: putative histidine kinase. 1: VicX protein of S. pneumoniae (CAB65440). 2: putative ribonuclease III of S. pyogenes (AAK33526). 3: putative chromosomal segregation SMC protein of S. pyogenes (AAK33527). 4: putative phenylalanyl tRNA synthetase (α subunit) of S. pyogenes (AAK33709). 5: hypothetical 19.8-kDa protein from Lactococcus delbrueckii subsp. lactis (P46543) and protease synthase- and sporulation-negative regulatory protein PAI of B. subtilis (P21340). 6: phenylalanyl tRNA synthetase (β subunit) of S. pyogenes (AAK33710). (D) System tcd. 1: putative pyrazinamidase/nicotinamidase from S. pyogenes (AAK34512). 2: aspartyl tRNA synthetase from L. lactis (AAK06063). 3: putative Glu-tRNA Gln amidotransferase subunit C of S. pyogenes (AAK34511). 4: putative Glu-tRNA Gln amidotransferase subunit A of S. pyogenes (AAK34510). 5: putative Glu-tRNA Gln amidotransferase subunit B of S. pyogenes (AAK34509). tcdR: putative RR. tcdK: putative histidine kinase. (E) System tce. 1: putative primosomal replication factor V of S. pyogenes (AAK34400). 2: putative methionyl tRNA transferase (fmt) from S. pyogenes (AAK34399). 3: putative RNA binding protein (SunL) of S. pyogenes (AAK34398). 4: putative phosphoprotein phosphatase of S. pyogenes (AAK34397). 5: putative protein kinase of S. pyogenes (AAK34396). 6: conserved hypothetical protein of S. pyogenes (AAK34395). t ceK: putative histidine kinase. tceR: putative RR. 7: conserved hypothetical protein of L. lactis subsp. lactis (AAK04990). 8: putative polynucleotide polynucleotidyl transferase (general stress protein) of S. pyogenes (AAK34392). 9: putative transcriptional regulator of S. pyogenes (AAK34719). (F) System tcf. 1: glucosyltransferase-S (GtfD) of S. mutans (BAA26115). 2: glucosyltransferase-SI (GtfC) of S. mutans (BAA26102). 3: ABC transporter ATP binding protein L. lactis subsp. lactis (AAK05848). 4: ABC transporter permease and substrate binding protein (YsaB) L. lactis subsp. lactis (AAK05847). tcfR: putative RR. tcfK: putative histidine kinase.

The strategy for generating RR mutants was similar for all six systems. First, either whole or part of the ORFs were cloned into pGEM7 after amplification by PCR. The ΩKm element was cloned into the middle of the cloned ORF of the pGEM7-based clone using an appropriate restriction enzyme, with or without subsequent modifying enzyme treatment, and the resulting clones were used to transform S. mutans UA159. We were successful in isolating RR mutants of five of the six putative two-component systems, but were unable to isolate mutants in system tcc. This is not surprising since system tcc is similar to the two-component systems essential for growth of B. subtilis and S. aureus[3,4].

To compare the biofilm-forming capacity of the mutants with the parent, we modified an established microtiter-based assay [9]. The results of those assays are summarized in Fig. 3A and visualized in Fig. 3B–D. Out of five clones tested, mutation in the RR of system B (tcbR) (Fig. 2) led to an almost 10-fold reduction in biofilm formation. The other four mutants had no significant defect in biofilm formation as compared to the wild-type strain. It is possible that the biofilm-forming potential of a strain would depend on the extent of its growth, i.e., the better a strain grows, the better it forms biofilms. To rule out this possibility, the growth of strain UA159 and the tcbR mutant was assessed in independent growth curve experiments as shown in Fig. 4. It can be seen that the wild-type strain and the tcbR mutant grew equally well over a period of 26 h and their growth rates were similar, effectively ruling out the possibility that differences in biofilm formation were due to differences in growth. We also measured the growth of the strains in microtiter plates and found that the tcbR mutant grew slightly better than the wild-type strain over a 44-h period. Also, the morphologies of UA159 and the tcbR mutant were indistinguishable on plates or by phase contrast microscopy (data not shown).

Figure 3

A: Biofilm formation by different mutants as compared to the wild-type strain. In vitro biofilm assays were performed as described in Section 2. Crystal violet in the wells was extracted with ethanol and biofilm formation was quantitated by measuring absorbance at 575 nm and giving an indirect measure of the number of cells attached to the surface of the wells in the form of biofilms. Results represent the average of three independent experiments and error bars represent standard deviations. A: A typical in vitro biofilm assay. Microtiter plates were processed as described in Section 2, air-dried and visualized and photographed before quantitation. B: Wells containing uninoculated medium only; C: UA159; D: tcbR mutant. The growth from the wells marked with ‘x’ was removed for measuring absorbance at 600 nm (to measure growth) before processing.

Figure 2

Southern blot hybridization to confirm the identity of the mutants. Chromosomal DNA from putative tcbR mutants and UA159 were digested with HincII. The genetic map of the response regulator (tcbR) and the histidine kinase (tcbK) of the putative two-component regulatory system B is shown in panel A. The HincII sites are shown as letter H. Southern blots were performed as described in Section 2 and the membranes were probed with radioactively labeled ΩKm-specific probe. In this gel picture, the mutant shows a ∼3.2-kb band (0.97 kb+2.2 kb ΩKm) whereas the wild-type does not show any band. B: Here, a tcb-RR gene-specific probe was utilized. In this gel picture, the mutant shows a ∼3.2-kb band (as explained above), whereas the wild-type shows a 0.97-kb band (which is the length of the fragment between the two HincII sites). C: M: tcbR mutant, WT: wild-type.

Figure 4

Growth of S. mutans UA159 wild-type strain as compared to the mutants. Bacteria were grown in the BM medium and the absorbance of the culture was monitored at 600 nm periodically over 26 h. ◯: wild-type strain; ▿: tcbR mutant; ▵: F mutant.

The mechanism by which inactivation of tcbR affects biofilm formation is not known. Of note, none of the mutants is defective in the ability to be transformed with plasmid or chromosomal DNA, so the effect is not exerted through the competence-development pathway. As mentioned earlier, the TcbRK system is most similar to the spi two-component system in S. pneumoniae, which is induced by a peptide SpiP [16]. This two-component system is thought to regulate expression of putative bacteriocins [16]. It is possible, then, that the TcbRK system could regulate gene expression in response to a peptide, perhaps one that acts as a signal to stimulate stable biofilm formation. Another important question that remains to be answered is which structural genes are regulated by the TcbRK system. A recent study on the biofilm formation of Streptococcus parasanguis revealed that the ability of this organism to form biofilms is abolished in a fap1 mutant, which specifies an adhesin protein required for the assembly of fimbriae [18]. It is possible that the TcbRK system regulates the expression of adhesins in S. mutans in response to an environmental signal. Ongoing studies are devoted to developing more information about how expression of this system is regulated, and to which environmental signals this system responds, in order to shed light on modulation of biofilm formation by the human pathogen S. mutans.


We thank Drs. Margaret Chen and Thomas Wen for critical evaluation of the manuscript and Chris Browngardt and Jana Penders for technical assistance. This study was supported by DE13239.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
View Abstract