DnaK expression in response to heat shock of Streptococcus mutans

ELSEVIER

FEMS Microbiology

Letters 131(1995)

255-261

DnaK expression in response to heat shock of Streptococcus mutans Gayatri C. Jayaraman a, Robert A. Burne ayb~ * ’ Department of Microbiology and Immunology, University of Rochester Medical Center, 601 Ebnwood Avenue, Rochester, NY 14642, USA b Department of Dental Research, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, NY 14642, USA Received 21 June 1995; accepted

10 July 1995

Abstract The oral pathogen, Streptococcus mutans, persistently colonizes human hosts and initiates oral disease despite extreme variations in environmental conditions. To begin to investigate the role of the stress protein, DnaK (Hsp70), in environmental stress responses by S. mutuns, pulse-chase experiments were initially used to establish that a functional heat shock response existed in this organism. A C-terminal fragment of the S. mutuns dnaK gene was cloned and engineered to be expressed with a histidine tag. Using the recombinant DnaK protein that had been purified by nickel affinity chromatography, an antibody specific for the S. mutans DnaK protein was generated to analyse DnaK expression under homeostatic and heat shock conditions. Western blot analysis indicated that the anti-recombinant DnaK antibody specifically recognized a protein (molecular mass approx. 68 kDa) which was induced in response to thermal stress. Elucidating the role of DnaK in responses by S. mutuns to various environmental stressors will provide a better understanding of how DnaK is involved in survival of extreme environments and the contribution of the DnaK protein to the virulence of S. mutuns. Keywords:

Streptococcus

mutans; DnaK, Molecular

chaperone;

Heat shock

1. Introduction The heat shock response is a homeostatic mechanism that is presumed to be required for cell survival. In response to heat and a variety of physical and chemical stressors, organisms increase the synthesis of a set of evolutionarily conserved proteins known as heat shock proteins (HSPs) [l-4]. Heat shock proteins of the Hsp70/DnaK family are known to play a central role in the responses of eukaryotes and prokaryotes to stresses [5]. Unlike eukaryotes,

* Corresponding author. Tel.: + 1 (716) 275 0381; Fax: (716) 473 2679; E-mail: [email protected]. Federation of European Microbiological SSDf 0378-1097(95)00265-O

Societies

+ 1

which synthesize several forms of Hsp70, prokaryotes generally synthesize one form, the DnaK protein. The gene encoding DnaK (&IL&) was originally identified in Escherichia coli as being essential for bacteriophage A DNA replication [6]. Later, it was shown that DnaK is almost indispensable for E. coli viability at high and low temperatures [7]. DnaK is now known to be induced in response to multiple stresses, including acid shock [8,9] and carbon starvation [lo]. The DnaK protein is an ATP-dependent molecular chaperone which functions in physical association with the heat shock proteins GrpE and DnaJ, facilitating the folding and/or assembly of nascent proteins, and of proteins that have been denatured by heat shock [ll], oxidative insults, or

256

G.C. Jayaraman, R.A. Burne/

FEMS Microbiology

exposure to other environmental stressors [12]. Additionally, DnaK plays critical roles in regulation of the response of bacteria to stress through modulation of the stability of transciptional factors controlling stress gene expression [13,141. Currently, no information is available on the heat shock response of the oral pathogen, Streptococcus mutans, about the role that DnaK could play in enabling this organism to respond rapidly to sudden changes in its environmental conditions, or about how known stress response proteins contribute to the tolerance of extreme environments by this organism. Notably, the ability to withstand environmental extremes in dental plaque, in particular acidification, is a major contributor to the cariogenic potential of this organism. This study describes initial experiments to identify the DnaK protein in S. mutuns and to demonstrate a functional heat shock response when the organism encounters environmental stress. Determining the role that DnaK plays in protecting oral streptococci and other lactic acid bacteria from environmental insults may provide a better understanding of mechanisms used by these organisms to tolerate adverse conditions. In a broader sense, it will extend the current knowledge about the major heat shock protein, DnaK, in streptococci, and the mechanisms by which it is regulated in response to environmental stress.

2. Materials and methods

2.1. Bacterial strains and plasmids S. mutans GS-5 and E. coli DHlOB were maintained and grown as previously described [I5]. The E. coli strain MlS[pREP4] was maintained and grown as recommended by Qiagen, the supplier of the QIAexpress vector kit. Plasmid-bearing E. coli strains were selected and maintained on L-agar supplemented with ampicillin (100 pg ml-’ 1 and/or kanamycin (25 pg ml-’ 1. 2.2. DNA manipulations DNA was prepared from S. mutans as previously described [151. Plasmid DNA was isolated from E. coli by a rapid boiling method 1161, followed by

Letters 131 (1995) 255-261

phenol and chloroform extractions, or by CsCl buoyant-density-gradient centrifugation [17]. Nucleotide sequence was determined by the Sanger chaintermination method [18] using the Sequenase version 2.0 kit (US Biochemicals), with [ cr35-S]dATP (Amersham) as the labelled nucleotide. DNA sequencing reactions were primed using an ml3/pUC universal forward primer (US Biochemicals), an ml3/pUC universal reverse primer (Promega) or an oligonucleotide (21-mer) complementary to pQE31 as suggested by Qiagen. The sequences were analysed using the University of Wisconsin Genetics Computer Group programs 1191. 2.3. Heat shock and metabolic labelling of S. mutans S. mutans cells were grown in a chemically defined minimal medium (FMC) [20] supplemented with 1% glucose, without methionine, at 37”C, to mid-exponential phase (OD,, = 0.7). Aliquots were heat shocked at 40°C or 42°C while controls were left at 37°C. After 10 or 15 min of incubation at the higher temperature, the cells were labelled with [ 35Slmethionine (New England Nuclear) followed by a chase with cold methionine. The cells were collected by centrifugation, washed in 10 mM Tris . HCl buffer and incubated at 37°C for 1 h with lysozyme (10 mg ml-‘) and mutanolysin (500 U ml-‘). Cells were lysed by boiling in one volume of sample buffer [21], and proteins were separated by electrophoresis on a 12% SDS-polyacrylamide gel (SDS-PAGE) with 14C-protein high molecular mass standards (Life Technologies Gibco BRL). The gel was dried and exposed to XAR-5 film (Kodak) (Fig. 11. 2.4. Protein purification

and antibody generation

To prepare the purified recombinant S. mutans DnaK protein (rDnaK) a 1.14-kb fragment from the plasmid pGJ7 (Fig. 21, encoding approximately 320 amino acids from the C-terminal portion of the S. mutuns DnaK protein, was cloned into the plasmid vector pQE31 (Qiagen). This resulted in the in-frame fusion of six consecutive histidine residues to the N-terminus of rDnaK (Fig. 3). rDnaK protein synthesis was induced by the addition of isopropyl-@thiogalactoside (IPTG) to a final concentration of 5

G.C. Jayaraman, R.A. Burne/ FEMS Microbiology Letters 131 (1995) 255-261

mM. Protein lysates were prepared by sonication and the rDnaK protein was purified by nickel affinity chromatography 1221. Individual fractions (1 ml) eluted from the affinity column were subjected to electrophoresis on a 12% SDS-PAGE, and proteins were visualized by staining with silver nitrate [23]. Amino acid sequence analysis was performed on fractions containing the peak amounts of rDnaK by using the Edman degradation process, as previously described [24], with an automated microsequencer

CQ u)

37% 10’

l!f

40% 10’

15’

42% 10’

15’

EcoRl P( fW)

251

UGA 1

WI

GORl

P(ogk)

1

ECORl

P(dnaK)

SaCl

lEgIll

* fruA (-4.3kbp)

or139 9pE ----Gz-+ (,.OSLbp) (SS7kbp) (1.Skbp) I

P4J-5

I ’

t

P(fi

’ p6m

pw7

’

’ I

Fig. 2. A schematic diagram of the recombinant bacteriophage AJECP-5 and selected plasmid derivatives. The recombinant bacteriophage AJECP-5, harboring the fi& gene of 8. mutans GS-5. Plasmid subclones, p6h5, p6hQ and p4h5, were derived by ligating the corresponding EcoRI fragment into the plasmid vector pBGS8. Plasmid subclone pGJ7 was derived by ligating the corresponding EcoRI/SacI fragment into the plasmid vector pUC19. The schematic also shows the location and size of S. mutans orfj9, grpE and dnaK gene homologues and their respective promoters (PI. The arrows depict the direction of transcription.

(Applied Biosystems, Model 473A). Protein concentrations were measured by Bradford protein assay

Fig. 1. Pulse-chase of S. mutans cells subjected to heat shock. S. mutans GS-5 cells were grown at 37°C to mid-exponential phase (OD,, approx. 0.7) in chemically defined minimal media (FMC, 20) with 1% glucose without methionine, at the conditions described in Materials and methods. Ahquots (2 ml) were transferred to a water bath at 40°C or 42°C respectively. Controls were left at 37°C. After 10 or 15 min of incubation at the higher temperature, the cells were labelled with 2 ~1 [35S]methionine (1002 Ci mmol-‘; Amersham) for 5 min. This was followed by the addition of 10 ~1 unlabelled L-methionine (2 mg ml-‘). Following a 5-min incubation, the cells were collected, washed and lysed as described in Materials and methods. The protein lysates were separated on a 12% SDS-PAGE gel with i4C-labelled high molecular mass protein markers and analysed by autoradiography. The arrows indicate the 68-kDa and the 61-kDa protein hands, respectively. Lanes: 1, 14C-labeled high molecular mass protein standards; 2, 4, 6, protein lysates from cells incubated at 37°C 40°C and 42°C respectively, labelled 10 min after incubation at the higher temperature; 3, 5, 7, protein lysates from cells incubated at 37°C 40°C and 42”C, respectively, lahelled 15 min after incubation at the higher temperature.

[251. The purified protein (210 pg) was separated by SDS-PAGE gel electrophoresis and a portion of the gel containing the 43-kDa rDnaK protein was excised. The gel fragment was emulsified in complete Freund’s adjuvant and the suspension was used to immunize a female New Zealand White rabbit. Following two subsequent injections with 210 pg of antigen each, serum was prepared from the animal and stored as aliquots at -20°C. 2.5. Western

blot analysis

S. mutans cells that had been heat shocked at 40°C or 42°C for 1 h were collected by centrifugation, washed and lysed as previously described [26]. The protein lysates were collected by centrifugation and subjected to SDS-PAGE electrophoresis along with high molecular mass protein markers (Life Technologies Gibco BRL). The protein samples were blotted onto an Immobilon-P filter membrane (Millipore) and subjected to Western Blot analysis as previously described 1161. DnaK induction as a result of heat shock was quantitated by densitometric analysis using the IS1000 Digital Imaging System (SUN BIOscience Inc.).

258

G.C. Jayaraman, R.A. Burne / FEMS Microbiology

Letters 131 (1995) 255-261

ELUTION

3. Results

FRACTIONS

3.1. Heat shock response of S. mutans Pulse-labelling of S. mutans GS-5 cells was used to examine protein synthesis in response to heat shock from 37°C to 40°C or to 42°C. The heat shock response resulted in the induction of at least two proteins with molecular masses of 68 kDa and 61 kDa (Fig. l), corresponding to those of the known heat shock proteins DnaK and GroEL, respectively. 3.2. Overexpression and purification nant DnaK protein (rDnaK)

of the recombi-

The C-terminal two-thirds of the dnaK gene was cloned from plasmid pGJ7 into the plasmid vector

Fig. 4. Silver stain of elution fractions containing rDnaK. A l-l culture of transformed E. coli MlS[pREP4] cells, grown to midexponential phase COD,, = 0.7) in LB containing kanamycin (25 Fg ml-‘) and ampicillin (100 pg ml-‘) was induced for rDnaK expression with the addition of IPTG to a final concentration of 5 mM. 3 h after induction, the cells were harvested and lysed using the procedure recommended Qiagen. Protein lysates were collected and loaded onto a 4-ml Ni-NTA column matrix which had been washed and equilibrated as recommended by Qiagen. rDnaK was eluted using an imidazole buffer gradient. Fractions (1 ml) were collected and proteins in 10 ~1 of each fraction were separated by electrophoresis through a 12% SDS-PAGE gel. Fractions 34-50 were visualized by staining with Silver nitrate as described in Materials and methods. The arrow depicts the location of the 43-kDa recombinant protein band.

pQE31 (Figs. 2 and 3) as described in Materials and methods; creating a recombinant DnaK protein (rDnaK) fused to six histidine residues and the Nterminus. rDnaK protein expression in E coli was induced, protein lysates were prepared, and proteins were fractionated by nickel affinity chromatography. The primary protein eluting after extensive washing and elution with imidazole buffer had an apparent molecular mass of 43 kDa in SDS-PAGE, consistent with that predicted for rDnaK (Fig. 4). The results of Edman degradation confirmed that the sequence of the purified protein, fused to six consecutive histidine residues, was identical to that predicted for the nucleotide sequence; [HHHHHH] TDLSGVTSTQISLPFIT. Fig. 3. Construction of a His, fusion to the S. mutans DnaK protein. The plasmid pGJ7 (Fig. 2) was digested with the restriction enzymes BglII and St1 to liberate a 1.14-kb fragment encoding approximately 320 amino acids from the C-terminal portion of the S. mutans DnaK protein. The fragment was purified using the Elu-Quik DNA purification kit (Schleicher and Schuell) and cloned into a gel-purified, BamHI/HincII-digested pQE31 vector (Qiagen). This allowed for the in-frame fusion of six consecutive histidine residues to the N-terminus of the C-terminal portion of the S. mutans DnaK protein to allow for one-step purification of the recombinant DnaK protein, rDnaK. ‘P/O’ denotes the T5 promoter/Lac operator region and ‘RBS’ denotes the ribosome binding site.

3.3. S. mutans DnaK is induced in response to heat shock Western Blot analysis of S. mutans protein lysates separated on a 12% SDS-PAGE gel demonstrated that the anti-rDnaK antibody, but not pre-immune sera, recognized a protein of a molecular mass consistent with that of known DnaK proteins (68 kDa) (Fig. 5). The lower molecular mass species that are recognized by the anti-rDnak antibody, and not by pre-immune sera, are likely the result of degradation

G.C. Jayaraman, RA. Burne / FEMS Microbiology Letters 131 (1995) 255-261

42°C

40°C

37°C

PI

68 KDa

Fig. 5. Western blot analysis of S. murans subjected to heat shock. S. mutans cells were grown in brain heart infusion broth (Difco) to mid-exponential phase, at 37°C in 5% CO, atmosphere. Aliquots (100 ml) were transferred to a water bath at 40°C and 42°C respectively, for 1 h. Controls were left at 37°C. The cells were collected by centrifugation, washed and lysed as described in Materials and methods. The protein lysates were collected by centrifugation and 5 pg of each protein were separated by electrophoresis on a 12% SDS-PAGE gel along with high molecular mass protein markers (Life Technologies Gibco BIU). The protein samples were transblotted onto Immobilon-P filter paper and subjected to Western blot analysis as described in Materials and methods. Western blot analysis was done using a 1:lOOO dilution of the generated rDnaK antibody that had been adsorbed against E. coli. A 1:1500 dilution of a goat anti-rabbit antibody conjugated to horseradish peroxidase (Kirkegaard and Perry Lab. Inc.) was used as the secondary antibody. Diaminobenzidine (Pierce) was used as the peroxidasc substrate. Lanes: 1, 2, protein lysate from S. mufans cells, heat shocked for 1 h at 42°C and 4o”C, respectively; 3, 4, protein lysate from S. mutans cells grown at 37°C; 1, 2, 3, probed with 1:lOOO dilution of the generated rDnaK antibody, adsorbed against E. coli cells; 4, probed with 1:lOOO dilution of rabbit pre-immune sera adsorbed against E. coli cells.

of DnaK following incubation at elevated temperatures. The results also demonstrated that this immunoreactive protein was induced in response to heat shock, at least two-fold (measured by densitometric analysis), as a consequence of heat shock from 37°C to 40°C and 42”C, respectively.

4. Discussion The data presented here demonstrate that a functional heat shock response exists in S. mutans. After

259

heat shocking S. mutans from 37°C to 40°C and 42°C respectively, there was an apparent induction of at least two proteins with the molecular masses of 68 kDa and 61 kDa (Fig. 11, similar to those of DnaK and GroEL proteins, respectively. In order to definitively identify the S. mutans DnaK protein and quantitate its induction in response to heat shock, monospecific antiserum to S. mutans DnaK was generated. Results from Western blot analysis (Fig. 4) demonstrated that the anti-rDnaK antibody specifically recognized the S. mutans DnaK protein. That this immunoreactive protein species represents the DnaK protein is supported by nucleotide sequence data which demonstrated that this gene lies in a cluster with the grpE gene and other genes involved in heat shock response. The S. mutans GS-5 dnaK gene product had > 75% identity with DnaK proteins of Gram-positive bacteria [27]. The induction of DnaK in response to heat shock of S. mutans lends credence to the hypothesis that DnaK could be important in the ability of S. mutans to survive temperature changes, and thus may also play important roles in stress responses, as in other many other organisms. The role of DnaK in survival of shock responses appears to be principally related to its ability to function as a molecular chaperone. It allows nascent polypeptides and denatured proteins to assume their correct configurations [28], although the protein clearly is involved in regulation of particular stress responses [1,8,3,4]. McCarthy and co-workers [29] have proposed that DnaK can be considered a cellular thermometer, since its up-regulation generally occurs as a result of increased stresses on the cells. DnaK levels in cells could, therefore, provide an indirect picture of the state of the cell in response to many stressors to which bacteria in the human oral cavity may be exposed. Thus, we would propose that DnaK expression may be a highly relevant marker to begin to understand stress responses and tolerance of environmental extremes in oral streptococci. Given the importance of DnaK in regulating cellular responses to environmental stressors, it is possible that the protein plays a significant role in survival of S. mutans, an organism that is capable of persisting and initiating disease despite extreme variations in its environmental conditions. The S. mutans dnaK gene and its gene product can serve as a model

260

G.C. Jayuraman,

R.A. Burnr / FEMS Microbzology Letters 131 (1995) 255-261

to study S. mutans responses to other stressors. particularly environmental acidification, which is a principal ecological determinant exploited by this bacterium. The role that DnaK could play in enabling S. mutuns to survive low pH is important in terms of the pathogenicity of the organism, since studies demonstrate that low pH confers a selective advantage for the growth of S. mutans over that of other less acid-tolerant bacteria [30,31]. Studies are in progress to determine the transcriptional organization and regulatory mechanisms governing S. mutuns dnaK gene expression under a variety of environmental conditions. These studies will provide insight into molecular control of stress gene expression in oral streptococci and other lactic acid bacteria, and add to our current knowledge of control of streptococcal gene expression.

Acknowledgements We thank Jana Penders for expert technical assistance and Robert E. Marquis for critical review of the manuscript. We are grateful for the assistance of Dan Frank in antibody preparation and Brian vanWuyckhuyse for protein sequence analysis. This study was supported in part by PHS Grant DE09878 to R.A.B.

References [l] Taglicht, D., Padan, A., Oppenheim, B. and Schuldincr. S. (1987) An alkaline shift induces the heat shock response in Escherichia coli. J. Bacterial. 169, 885-887. [2j Heyde, M., CoH, J.L. and Portalier, R. (1991) Identification of Escherichia coli genes whose expression increases as a function of external pH. Mol. Gen. Genet. 229, 197-205. [3] Volker, U., Mach, H., Schmid, R. and Hecker, M. (1992) Stress proteins and cross-protection by heat shock and salt stress in Bacillus subtifis. J. Gen. Microbial. 138, 2125-35. [4] Chuang, S.E. and Blattner, F.R. (1993) Characterization ot twenty-six new heat shock genes of Escherichiu co/i. J. Bacterial. 175, 5242-5252. [5] Lindquist, S. and Craig, E.A. (1988) The heat-shock proteins. Annu. Rev. Genet. 22, 631-677. [6] Georgopoulos, C. (1977) A new bacterial gene (gropcf which affects A DNA replication. Mol. Gen. Genet. 151, 35-39.

[7] Bukau, B. and Walker, G.C. (1990) Mutations altering heat shock specific subunit of RNA polymerase suppress major cellular defects of E. cdi mutants lacking the DnaK chaperone. EMBO .I. Y, 4027-4036. [81 Heyde, M. and Portalier, R. (1990) Acid shock proteins of Escherichia coli. FEMS Microbial. Lett. 57, 19-26. [9] Foster, J.W. (1991) Salmonella acid shock proteins arc required for the adaptive acid tolerance response. J. Bacterial. I73,6896-6902. 101 Spence, J.A.C. and Georgopoulos, C. (1990) Role of Escherichiu coli heat shock proteins DnaK and HtpG tC62.5) in response to nutritional deprivation. J. Bacterial. 172. 7157-7166. I 111 Skowyra, D. and Wickner, S. (1993) The interplay of the GrpE heat shock protein and Mg*+ in RepA monomerization by DnaJ and DnaK. J. Biol. Chem. 268, 25296-301. 121 Gething, M.J. and Sambrook, J. (1992) Protein folding in the cell. Nature 355, 33-45. il.31 Gamer, J., Bujard, H. and Bukau, B. t lYY2) Physical intcraction between heat shock proteins DnaK, DnaJ, and GrpE and the bacterial heat shock transcription factor sigma 32. Cell 69, 833-842. [ 141 Liberek, K., Gahtski, T.P., Zylicz, M. and Georgopoulos, C. (1992) The DnaK chaperone modulates the heat shock rcsponse of Escherichia coli by binding to the sigma 32 transcription factor. Proc. Natl. Acad. Sci. USA 89, 35163520. [15] Burne, R.A.. Rubinfeld, B., Bowen, W.H. and Yasbin, R.E. (1986) Tight genetic linkage of a glucosyltransferase and dextranase of Streptococcus mums GS-5. J. Dent. Res. 65. 1392-1401. [ 161 Maniatis, T., Fritsch, E.F. and Sambrook, J. (1989) Molecular Cloning: A Laboratory Manual. 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. [17] Godson, G.N., and Vapne, D. (1973) A simple method of preparing large amounts of &X174 RF1 supercoiled DNA. Biochim. Biophys. Acta 229, 516-520. [ 181Sanger, F.. Nicklen, S. and Coulsen, A.R. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5436-5467. [19] Devereaux. J.. Haeberli, P. and Smithies, 0. (1984) A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12. 3X7-395. [Ztt] Terleckyj, B., Willett, N.P. and Shockman, G.D. (lY75) Growth of several cariogenic strains of oral streptococci in a chemically defined medium. Infect. Immun. 11,649-655. [21] Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of bacteriophage T4. Nature 227, 680-685. [22] Hoffman, A. and Roeder, R.G. (1991) Purification of Histagged proteins in non-denaturing conditions suggests a convenient method for protein interaction studies. Nucleic Acids Res. 19, 6337-6338. [23] Sammons, D.W., Adams, L.D. and Nishizawa, E.E. (lY81) Ultrasensitive silver-based color staining of polypeptides in polyacrylamide gels. Electrophoresis 2, 135. [24] Edman, P. and Begg, G. (1967) A protein sequenator. Eur. J. Biochem. 1, 80-91.

G.C. Jayaraman, Rd. Burne/FEMS [25] Bradford, 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, 248254. [26] Wexler, D.L., Hudson, M.C. and Bume, R.A. (1993) Srreprococcu.r mutans fructosyltransferase (frf) and glucosyltransferase (gtfBC1 operon fusions in continuous culture. Infect. Immun. 61, 1259-1267. 1271 Jayaraman, G., Penders, J.E.C. and Burne, R.A. (1993) Isolation of the dnaK gene of Streptococcus mutans GS-5. Abstracts of the Annual Meeting: American Society for Microbiology, Washington, D.C., Abstr. #D-124.

Microbiology Letters 131 (1995) 255-261

261

[28] Craig, E.A. and Gross, C.A. (1991) Is Hsp70 the cellular thermometer? Trends Biochem. Sci. 16, 135-140. [29] McCarty, J.S. and Walker, G.C. (1991) DnaK as a thermometer: threonine-199 is site of autophosphorylation and is critical for ATPase activity. Proc. Natl. Acad. Sci. USA 88, 9513-9517. [30] Bender, G.R., Sutton, S.V.W. and Marquis, R.E. (1986) Acid tolerance, proton permeabilities, and membrane ATPases of oral streptococci. Infect. Immun. 53, 331-338. [31] Belli, W.A. and Marquis, R.E. (1991) Adaptation of Streptococcus mutans and Enterococcus hirae to acid stress in continuous culture. Appl. Environ. Microbial. 57, 1134-1138.