Adherence of Streptococcus sanguis clinical isolates to smooth surfaces and interactions of the isolates with Streptococcus mutans glucosyltransferase.

Streptococcus sanguis isolated from human dental plaque were grown in Todd-Hewitt broth. Cells were collected by centrifugation and lyophilized after extensive washing with water. The cell-associated glucosyltransferase (GTase) activities of S. sanguis strains were assayed with [14C]sucrose. Strain differences in GTase activity were significant within the same serotype or biotype or both. The ability of S. sanguis cells to adhere to smooth glass surfaces was generally weak, irrespective of significant cell-associated GTase activity synthesizing water-insoluble, gel-like glucans. Resting cells of most S. sanguis strains bound extracellular GTase from Streptococcus mutans strain B13 (serotype d), resulting in the strong adherence of the S. sanguis cells to smooth glass surfaces in the presence of sucrose. Conversely, S. mutans B13 cells also could bind extracellular GTase from some strains of S. sanguis examined. The sucrose-dependent adherence of S. mutans cells was not altered, although S. sanguis strains from which the extracellular GTases were obtained did not produce significant adherence in the presence of sucrose. In view of these findings, it was suggested that S. mutans GTase could affect the adherence of S. sanguis to smooth tooth surfaces in the oral cavity.     

Interaction of inflammatory cells and oral microorganisms. III. Modulation of rabbit polymorphonuclear leukocyte hydrolase release response to Actinomyces viscosus and Streptococcus mutans by immunoglobulins and complement.

In the absence of antiserum, rabbit polymorphonuclear leukocytes (PMNs) released lysosomal enzymes in response to Actinomyces viscosus (19246) but not to Streptococcus mutans (6715). Antibodies had a marked modulating influence on these reactions. PMN hydrolase release was significantly enhanced to both organisms when specific rabbit antiserum and isolated immunoglobulin G (IgG) were included in the incubations. Immune complex F(ab')2 fragments of IgG directed against S. mutans agglutinated bacteria. Immune complexes consisting of S. mutans and F(ab')2 fragments of IgG directed against this organism were not effective as bacteria-IgG complexes in stimulating PMN release. The intensity of the release response to bacteria-IgG complexes was also diminished when PMNs were preincubated with isolated Fc fragments derived from IgG. Fresh serum as a source of complement components had no demonstrable effect on PMN release either alone or in conjuction with antiserum in these experiments. These data may be relevant to the mechanisms and consequences of the interaction of PMNs and plaque bacteria in the pathogenesis of periodontal disease.     

Salivary-agglutinin-mediated adherence of Streptococcus mutans to early plaque bacteria.

Interspecies binding is important in the colonization of the oral cavity by bacteria. Streptococcus mutans can adhere to other plaque bacteria, such as Streptococcus sanguis and Actinomyces viscosus, and this adherence is enhanced by saliva. The salivary and bacterial molecules that mediate this interaction were investigated. Salivary agglutinin, a mucinlike glycoprotein known to mediate the aggregation of many oral streptococci in vitro, was found to mediate the adherence of S. mutans to S. sanguis or A. viscosus. Adherence of S. mutans to saliva- or agglutinin-coated S. sanguis and A. viscosus was inhibited by antibodies to the bacterial agglutinin receptor. Expression of the S. sanguis receptor (SSP-5) gene in Enterococcus faecalis increased adhesion of this organism to saliva- or agglutinin-coated S. sanguis and A. viscosus. This interaction could be inhibited by antibodies to the agglutinin receptor. The results suggest that salivary agglutinin can promote adherence of S. mutans to S. sanguis and A. viscosus through interactions with the agglutinin receptor on S. mutans.     

Effect of dextranase on the extracellular polysaccharide synthesis of Streptococcus mutans; chemical and scanning electron microscopy studies.

A dextranase preparation (AD17) partially purified from a culture liquor of Spicaria violacea strain IFO 6120 significantly inhibited the formation of artifcial dental plaque on a steel wire or on an extracted tooth surface. Changes in the surface morphology of Streptococcus mutans cells due to AD17 action were studied using scanning electron microscopy. S. mutans cells grown in 5% sucrose-containing broth were coated with sticky amorphous capsule-like material, whereas cells grown in sucrose in the presence of AD17 or in glucose instead of sucrose did not synthesize such capsular material. AK17 degraded commercially available dextrans of molecular weight 7 X 1(04) and 2 X 10(6) to liberate glucose and various oligosaccharides, including isomaltose. On the other hand, AD17 hydrolyzed the extracellular polysaccharides (mainly glucan in nature) of some strains of S. mutans to a limited degree. Only 15 to 36% of the total polysaccharides were hydrolyzed by AD1M with little release of isomaltose. Prolonged incubation of the polysaccharides from S. mutans with AD17 did not release additional reducing sugars, which indicates that AD17 did not contain alpha-1,3-glucanase activity. These results suggest that glucosidic linkages which are susceptible to AD17 may play an important role in the adherence of S. mutans cells to smooth surfaces.     

Effects of molecular weight of dextran on the adherence of Streptococcus sanguis to damaged heart valves.

Dextran-producing streptococci such as Streptococcus sanguis are the organisms most frequently associated with infective endocarditis in humans. A series of experiments was designed to study how the molecular weight of dextrans affects the adherence of an endocarditis strain of S. sanguis to canine heart valves covered with platelets and fibrin. The data indicated that this adherence was dependent on dextrans of high molecular weight, such as dextran T-2000 or glucans isolated from S. sanguis or S. mutans. The adherence properties of the strain studied were not modified by prior exposure of the bacterial cells of valve leaflets to high-molecular-weight dextrans. Preexposure of bacterial cells or valve leaflets to low-molecular-weight dextrans decreased their adherence. Low-molecular-weight dextrans interfered with adherence of dextran-positive strains to damaged heart valves.     

Interaction of inflammatory cells and oral microorganisms. IV. In vitro release of lysosomal constituents from polymorphonuclear leukocytes exposed to supragingival and subgingival bacterial plaque.

The deposition of bacterial plaques on tooth surfaces appears to be responsible for the initiation and progression of periodontal disease. In this study, human peripheral blood polymorphonuclear leukocytes (PMNs) actively released lysosomal constituents upon in vitro exposure to either viable or irradiated, supragingival or subgingival dental plaque. Plaques were obtained from the PMN donors (autologous plaque) or from pooled samples (homologous plaque) secured from patients with periodontal lesions. Fresh sera from PMN donors amplified the release reactions to supragingival and subgingival plaques. Heated (56 degrees C, 30 min) sera also enhanced release reactions, but not as consistently as fresh serum. It was postulated that modulation of PMN release by serum is mediated by complement components and/or antibodies to plaque bacteria. Electron microscopic observations indicated that degranulation and discharge of PMN lysosomal enzymes may be associated with phagocytosis of gram-positive and gram-negative plaque bacteria and with reverse endocytosis of lysosomes from cells contacting relatively large masses of aggregated plaque bacteria. These data suggest that PMN lysosome release in response to plaque may serve as a potential mechanism of tissue injury in the pathogenesis of gingival and periodontal inflammation.     

Expression of Streptococcus mutans gtf genes in Streptococcus milleri.

The Streptococcus mutans glucosyltransferase (GTF) genes gtfB and gtfC were ligated into Escherichia coli-streptococcus shuttle plasmids and introduced into Streptococcus milleri. gtfB transformant KSB8 formed an S. mutans-like rough colony on mitis salivarius agar and expressed an extracellular GTF-I, of 158 kDa, and two cell-bound GTF-Is, of 158 and 135 kDa. gtfC transformant KSC43 formed a semirough colony on mitis salivarius agar and expressed primarily an extracellular GTF-SI, of 146 kDa, and two cell-bound GTF-SIs, of 146 and 152 kDa. The extracellular GTFs from KSB8 and KSC43 were purified and characterized. The two types of GTF also reacted specifically with monoclonal antibodies directed against each enzyme. Both enzymes synthesized significant amounts of oligosaccharides, consisting primarily of alpha-1,6-glucosidic linkages, as well as water-insoluble glucans, containing alpha-1,3-glucosidic linkages. Insoluble-glucan-synthesizing activities of both enzymes were stimulated (three- to sixfold) by the addition of dextran T10 and were inhibited in the presence of 1.5 M ammonium sulfate. The Km(s) for sucrose and the optimal pHs were also similar for both enzymes. However, when the transformants were grown in Todd-Hewitt broth supplemented with sucrose, KSC43 cells, expressing GTF-SI activity, adhered to glass surfaces in vitro, while KSB8 cells, expressing GTF-I activity, did not. These results are discussed relative to the potential role of the gtfB and gftC genes in S. mutans cariogenicity.     

Effect of sucrose in culture media on the location of glucosyltransferase of Streptococcus mutans and cell adherence to glass surfaces

Streptococcus mutans strain B13 (serotype D) almost exclusively produced free glucosyltransferase (GTase) in the culture supernatant when grown in sucrose-free TTY broth medium, which was composed of Trypticase (Baltimore Biological Laboratory [BBL] Cockeysville, Md.), tryptose (Difco Laboratories, Detroit, Mich.), yeast extract (BBL), salts, and 1% glucose. Organisms grown in sucrose-free TTY broth retained very weak cell-associated GTase activity and did not adhere significantly to glass surfaces in the presence of exogenous sucrose. If sucrose was added to TTY broth, however, GTase was found on the cell surface where cell-bound, water-insoluble glucans were synthesized. Most commercially available products of Todd-Hewitt broth were found to contain trace amounts of sucrose, as did Trypticase soy broth (BBL), whereas brain heart infusion broth (Difco and BBL) was found to be essentially free of sucrose. Almost all detectable GTase activity was cell associated when S. mutans B13 was grown in Todd-Hewitt or trypticase soy broth. Heat-treated B13 cells grown in Todd-Hewitt broth and cell-free, water-insoluble glucans bound free GTase and produced marked adherence in the presence of sucrose. Experiments strongly suggest that the binding sites for free GTase are the surface glucans, and cell-associated and extracellular GTases are most likely alternate states of the same enzyme protein.     

Binding of lectins to Streptococcus mutans cells and type-specific polysaccharides, and effect on adherence.

The lectin concanavalin A (Con A) agglutinated the cells of 13 of 15 strains of the seven serotypes of Streptococcus mutans in an 18-h incubation period. Strains of types a, d, f, and g agglutinated within 2 h. Strains of a, d, and f were also agglutinated in 2 h by the castor bean lectin RCA. S. sanguis, S. salivarius, S. bovis, Actinomyces viscosus, A. naeslundii, and Lactobacillus plantarum were agglutinated within 2 h. The S. mutans type f polysaccharide was precipitated by Con A. The a, b, c, d, and e polysaccharides were not precipitated. Glucan from d and e strains of S. mutans and dextran T2000 were also precipitated by Con A. D-glucose inhibited the agglutination of type f cells by Con A and the agglutination of type d cells by D-galactose. The quantity of [acetyl-3H]Con A bound was not proportional to the degree of agglutination. Cells grown in sucrose medium bound more Con A than those grown in glucose medium. After treatment with dextranase, the sucrose-grown cells bound two- to fourfold more Con A. The binding of Con A to the type-specific polysaccharide or to teichoic acid could not be determined by the use of specific antibody due to the binding of Con A to the antibody globulin on the cell surface. Con A bound to S. mutans cells did not inhibit the activity of cell-bound glucosyltransferase, glucan synthesis, and in vitro adherence. Bound Con A also did not inhibit the ability of heat-treated cells to bind glucosyltransferase, synthesize glucan, and produce in vitro adherence.     

Antibody-mediated inhibition of dextran-sucrose-induced agglutination of Streptococcus mutans

Immune sera to strains of Streptococcus mutans were found to inhibit agglutination of bacterial suspensions to which either high-molecular-weight dextran or sucrose was added. This inhibitory activity was shown to be mediated by antibody of the immunoglobulin G class. A semiquantitative assay was developed which demonstrated cross-inhibition of dextran/sucrose-induced agglutination among several strains of S. mutans. Antiserum to a partially purified glucosyltransferase was found to lack agglutination inhibition activity, consistent with the hypothesis that the dextran-binding antigen detected by the assay is immunologically distinct from the glucan-synthetic enzyme. A model for glucan synthesis and binding consistent with the reported data is described.     

Activity of two Streptococcus mutans bacteriocins in the presence of saliva, levan, and dextran.

The extracellular dextrans produced from sucrose by Streptococcus mutans strains BHT and GS-5 did not prevent the synthesis or release of active bacteriocins by these two strains. In addition, several streptococci that were genetically sensitive to these bacteriocins, and that could synthesize a variety of extracellular dextrans and levans from sucrose, remained phenotypically sensitive when grown in the presence of sucrose. Bacteriocin activity was not altered by treatment with high-molecular-weight dextran or by human saliva. The bacteriocins produced by, and active against, S. mutans thus appear to be capable of acting in vivo and may play a role in regulating the bacterial ecology of the oral cavity.     

Effects of local immunization of hamsters with glucosyltransferase antigens on infection with Streptococcus sanguis.

The effects of immunization with antigens of the Streptococcus mutans glucosyltransferase (GTF) complex on oral challenge with two Streptococcus sanguis strains (H7PR3 and 34) in hamsters were studied. Antisera to S. mutans GTF complex were able to inhibit one-third (strain H7PR3) to one-half (strain 34) of the S. sanguis GTF activity which could be inhibited when these S. sanguis GTFs were incubated with antisera to S. sanguis GTF. Washed, intact cells of strains H7PR3 and 34 were able to remove a significant amount of enzyme inhibitory activity and immunoglobulin G antibody activity from antisera to S. mutans GTF. These results established the existence of an antigenic relationship between S. sanguis and S. mutans GTFs. The effect of injection of S. mutans strain 6715 GTF or phosphate-buffered saline, incorporated into complete Freund adjuvant, on oral challenge with S. sanguis was compared in 243 hamsters in nine experiments. Salivary and serum GTF inhibitory activity was present in all GTF-injected animals before challenge. After a 2- or 3-day challenge with S. sanguis H7PR3 (seven experiments) or 34 (two experiments), fewer bacteria were recovered from GTF-injected hamsters in every experiment. Significant differences were observed in six of the nine experiments. During the 7- to 21-day period after challenge, 33% of the phosphate-buffered saline-injected sham group (group I) still had S. sanguis recoverable from the molar surfaces, whereas only 19% of the S. mutans GTF-injected group (group II) remained infected with S. sanguis (P less than 0.01). These results suggest that immunization with GTF from S. mutans may influence the colonization potential of S. sanguis in the oral cavity.     

Cloned gtfA gene of Streptococcus mutans LM7 alters glucan synthesis in Streptococcus sanguis.

Streptococcus mutans LM7 (Bratthall serotype e) chromosomal DNA was partially digested with EcoRI and ligated into the positive-selection plasmid vector pOP203(A2+). The ligation mixture was transformed into Escherichia coli, and transformants were selected for tetracycline resistance. Recombinant-bearing clones were screened for their ability to ferment raffinose, using the procedure of Robeson et al. (J. Bacteriol. 153:211-221, 1983). One raffinose-fermenting clone was isolated and found to contain a plasmid with an insert consisting of four EcoRI fragments totalling approximately 10.3 kilobases (kb). This strain was capable of growth on defined medium plus raffinose or sucrose and generated reducing sugars from a sucrose substrate. Southern hybridization analysis of the four EcoRI fragments revealed homology not only to S. mutans LM7 chromosomal DNA but also to S. mutans serotypes b, c, and f. Subcloning of this fragment array into a streptococcal E. coli shuttle vector indicated that a 2.4-kb EcoRI fragment was essential for sucrase activity. E. coli minicell experiments revealed a gene product of 55 kilodaltons. These data along with restriction endonuclease analysis and Southern hybridizations suggested that the cloned S. mutans LM7 gene was closely related to the gtfA gene cloned by Robeson et al. from S. mutans PS13 (Bratthall serotype c). The shuttle plasmid containing the 2.4-kb fragment was transformed into Streptococcus sanguis, which subsequently displayed increased sucrase activity in both intracellular and extracellular fractions. Elevated levels of synthesis of alcohol-insoluble and water-insoluble glucans were observed with crude extracellular fractions of the S. sanguis strain bearing the 2.4-kb fragment. An isolate cured of the shuttle plasmid plus the 2.4-kb fragment displayed wild-type S. sanguis glucan synthesis. In S. sanguis, this gtfA allele may play a role in glucan synthesis by interacting with extant high-molecular-weight glucosyltransferases.     

Effect of glucose and sucrose on survival in batch culture of Streptococcus mutans C67-1 and a noncariogenic mutant, C67-25.

The growth and survival of two strains of Streptococcus mutans in 5% (wt/vol) glucose or sucrose broth was investigated. S. mutans strain C67-1 showed little loss of viability after 30 h of incubation in batch culture in the presence of either sugar. S. mutans strain C67-25, a noncariogenic mutant of C67-1 that has lost the ability of the latter to produce sticky, insoluble extracellular polysaccharide when grown in sucrose broth, showed a dramatic loss of viability after 30 h of incubation in either glucose or sucrose broth, the effect being most marked in the presence of glucose. The loss of viability was shown to be due to acid production. Insoluble extracellular polysaccharide production appears to be a phenomenon favoring the survival of organisms subjected to high sucrose levels. Other factors must be involved, however, since there are differences between the two strains as regards their survival in glucose broth.     

Effects of oxygen on pyruvate formate-lyase in situ and sugar metabolism of Streptococcus mutans and Streptococcus sanguis.

The strictly anaerobic metabolism of sugar in strains of Streptococcus mutans and Streptococcus sanguis was studied because deep layers of dental plaque are strictly anaerobic. Galactose-grown cells of these streptococcal strains had higher pyruvate formate-lyase activity than did glucose-grown cells. Among these strains, two strains of S. mutans had a significantly higher pyruvate formate-lyase activity than did the others. This enzyme is extremely sensitive to oxygen, and even in situ the enzyme was inactivated by exposure of the cells to air. Lactate was less than 50% of the total end product of the strictly anaerobic incubation of the galactose-grown cells of S. mutans with excess glucose, and a significant amount of formate, acetate, and ethanol was produced through the catalysis of pyruvate formate-lyase. But the cells exclusively produced lactate when exposed to air for 2 min before the anaerobic incubation. The metabolism of sorbitol by S. mutans was seriously impaired by the exposure of the cells to oxygen, and the metabolic rate was reduced to less than 1/20 of that found under strictly anaerobic conditions because of the inactivation of pyruvate formate-lyase. S. sanguis produced a smaller amount of the volatile products from glucose than did S. mutans because of the low level of pyruvate formate-lyase. However, the pyruvate formate-lyase in situ in S. sanguis was less sensitive to oxygen than was that in S. mutans. Because of this low sensitivity, S. sanguis metabolized glucose more rapidly under aerobic conditions, whereas the rates of the aerobic and anaerobic metabolism of glucose by S. mutans were similar, which suggests that S. mutans rather than S. sanguis can sustain the rapid sugar metabolism in the deep layers of dental plaque.     

Properties of Streptococcus mutans Grown in a Synthetic Medium: Binding of Glucosyltransferase and In Vitro Adherence, and Binding of Dextran/Glucan and Glycoprotein and Agglutination

The influence of culture media on various properties of Streptococcus mutans was investigated. Strains of S. mutans (serotypes c, d, f, and g) were grown in a complex medium (Todd-Hewitt broth [THB]) or a synthetic medium (SYN). The SYN cells, in contrast to THB cells, did not bind extracellular glucosyltransferase and did not produce in vitro adherence. Both types of cells possessed constitutive levels of glucosyltransferase. B13 cells grown in SYN plus invertase-treated glucose possessed the same level of constitutive enzyme as THB cells. In contrast to THB cells, the SYN cells of seven serotype strains did not agglutinate upon the addition of high-molecular-weight dextran/glucan. Significant quantities of lower-molecular-weight (2 x 10(4) or 7 x 10(4)) dextran and B13 glucan were bound by SYN cells. SYN cells agglutinated weakly in anti-glucan serum (titers, 0 to 16), whereas THB cells possessed titers of 32 to 256. Evidence for the existence of a second binding site in agglutination which does not possess a glucan-like polymer has been obtained. B13 cells grown in invertase-treated THB agglutinated to the same degree as normal THB cells. The nature of this site is unknown. SYN cells possess the type-specific polysaccharide antigen. B13 cells did not bind from THB a glycoprotein which reacts with antisera to the A, B, or T blood group antigens or which allows agglutination upon the addition of dextran. The results demonstrate that S. mutans grown in a chemically defined medium possesse markedly different biochemical and biological activities than cells grown in a complex organic medium.     

Binding of Todd-Hewitt broth antigens by Streptococcus mutans.

Streptococcus mutans 10449, grown in chemically defined culture medium, was tested for its ability to bind 3H-labeled Todd-Hewitt broth components (greater than 12,000 Mr). Maximum adsorption of radioactivity occurred within 5 min at room temperature, and cell-bound material was not completely removed by extended washing with buffer. Heat-killed, arsenate-inhibited, and viable bacteria bound similar quantities. Only 0.09% of the radioactivity in the preparation of high Mr Todd-Hewitt broth components was removed by absorption with excess numbers of S. mutans 10449 cells. Binding followed saturation kinetics and was competitively inhibited by unlabeled medium components, both the dialyzable and nondialyzable fractions. Other oral streptococci were also found to bind these complex medium components. Rabbit antiserum elicited to the high-molecular-weight Todd-Hewitt broth components reacted with monkey cardiac muscle and with S. mutans coated with medium components. Absorption of the anti-Todd-Hewitt broth serum with homogenized heart removed antibodies that reacted with Todd-Hewitt broth-coated S. mutans. Therefore, the tissue-specific antigens of this beef heart infusion medium that adsorb to S. mutans can interfere with the detection and characterization of antigens shared by these bacteria and animal tissues.     

Dextran-mediated interbacterial aggregation between dextran-synthesizing streptococci and Actinomyces viscosus.

Streptococcus sanguis and Streptococcus mutans bind to the surface of Actinomyces viscosus, producing large microbial aggregates. Aggregates form rapidly and are not easily dissociated by vigorous mixing. The binding is mediated by dextran. Glucose-grown streptococci will not aggregate unless they are first mixed with high-molecular-weight dextran. Aggregation is induced with dextrans isolated from Leuconostoc, S. sanguis, or S. mutans. Sucrose-grown streptococci will adhere to A. viscosus without the addition of an exogenous source of dextran. A. viscosus will bind dextran and then bind glucose-grown streptococci. Aggregation occurs over a wide pH range and is dependent on cations. The aggregating activity of A. viscosus is both protease and heat sensitive. The aggregating activity of S. sanguis is heat stable but sensitive to dextranase.     

Binding of Glucosyltransferase and Glucan Synthesis by Streptococcus mutans and Other Bacteria

Lyophilized and heat-treated cells from the seven serotypes of Streptococcus mutans were examined for their ability to bind added insoluble-product glucosyl-transferase (GTase) and to synthesize cell-associated glucan from [(14)C]sucrose. Lyophilized cells of serotypes a and g did not synthesize any more additional glucan than did the controls after exposure to GTase. These cells, however, synthesized four- to eightfold-greater quantities of glucan than did the cells of the remaining serotypes. Lyophilized cells of serotypes b, c, d, e, and f synthesized two- to threefold-greater quantities of glucan after exposure to GTase than did the controls without added enzyme. Lyophilized cells of serotypes a and g synthesized 6- to 10-fold-greater quantities of glucan than did heat-treated cells of the same strain after binding of GTase. Lyophilized cells of the remaining serotypes synthesized only 1.6- to 3.3-fold-greater quantities of glucan than did the heat-treated cells. These results demonstrate that heat treatment to inactivate cell-associated GTase does not create additional GTase binding sites in S. mutans and that serotypes a and g are considerably more active in cell-associated glucan synthesis than cells of the other five serotypes. Ten species of gram-positive and gram-negative bacteria from five genera which do not produce in vitro plaque synthesized 10- to 100-fold-less glucan than did the S. mutans strains after exposure to GTase. Of these species, S. sanguis, Actinomyces viscosus, and A. naeslundii synthesized the largest quantities of glucan. Three mutant strains of S. mutans which possess a reduced ability for in vitro adherence but do agglutinate with glucan or dextran synthesized only one-third as much glucan after binding of GTase as the control. These results are discussed in relation to in vitro and in vivo plaque development and the agglutination of S. mutans. The results support earlier findings which indicate that the presence of bacterial species other than S. mutans in smooth-surface dental plaque is due in part to contact of the cells with glucan in the developing plaque and not to the binding of cell-free GTase and the in situ synthesis of glucan. The results obtained with these representative strains of the seven serotypes of S. mutans may not apply to the same extent to other strains within the serotypes.     

Strains of Streptococcus mutans and Streptococcus sobrinus attach to different pellicle receptors.

We compared the levels of adsorption of Streptococcus mutans JBP and Streptococcus sobrinus 6715 to experimental pellicles formed from unsupplemented and glucosyltransferase (GTF)-supplemented saliva. Pellicles formed on hydroxyapatite beads from GTF or from saliva-GTF mixtures possessed detectable GTF activity. Low levels of GTF activity were also detected in clarified whole human saliva, but not in samples of submandibular saliva. The adsorptive behavior of S. mutans JBP to pellicles formed from saliva or saliva-GTF mixtures was strikingly different from that of S. sobrinus 6715. S. mutans JBP adsorbed in higher numbers to pellicles formed from whole or submandibular saliva than to buffer-treated hydroxyapatite under the assay conditions used, in which blocking with albumin was used. In contrast, S. sobrinus 6715 attached in lower numbers and did not show enhanced adsorption to pellicles prepared from saliva. Pellicles prepared from the high-molecular-weight mucin fraction of submandibular saliva effectively promoted adsorption of S. mutans JBP, but none of the saliva fractions tested enhanced the attachment of S. sobrinus 6715 above the levels of buffer controls. Exposure of pellicles which contained GTF to sucrose to permit in situ synthesis of glucan markedly enhanced attachment of S. sobrinus 6715 but not attachment of S. mutans JBP. Also, the presence of sucrose throughout the adsorption period did not enhance attachment of S. mutans JBP. Both organisms possessed cell-associated GTF, and GTF preparations derived from S. sobrinus 6715 and Streptococcus sanguis FC-1 behaved like GTF derived from S. mutans JBP. S. sobrinus 6715 attached in high numbers to dextran-treated hydroxyapatite, whereas S. mutans JBP did not. These observations suggest that S. mutans JBP cells possess an adhesin which binds to salivary components in the pellicles. In contrast, S. sobrinus 6715 cells appear to possess an adhesin which binds to glucan in the pellicles. Four additional strains of S. mutans and four additional strains of S. sobrinus behaved qualitatively like strains JBP and 6715, respectively, and thus the differences observed appear to be representative of these species. Collectively, our data indicate that S. mutans and S. sobrinus attach to different receptors in experimental pellicles.