Inhibitory Effect of Bovine Milk Lactoferrin on the Interaction between a Streptococcal Surface Protein Antigen and Human Salivary Agglutinin*

Human whole saliva induces aggregation ofStreptococcus mutans cells via an interaction between a surface protein antigen (PAc) of the organism and salivary agglutinin. Bovine milk inhibits the saliva-induced aggregation of S. mutans. In this study, the milk component that possesses inhibitory activity against this aggregation was isolated and found to be lactoferrin. Surface plasmon resonance analysis indicated that bovine lactoferrin binds more strongly to salivary agglutinin, especially to high molecular mass glycoprotein, which is a component of the agglutinin, than to recombinant PAc. The binding of bovine lactoferrin to salivary agglutinin was thermostable, and the optimal pH for binding was 4.0. To identify the saliva-binding region of bovine lactoferrin, 11 truncated bovine lactoferrin fragments were constructed. A fragment corresponding to the C-terminal half of the lactoferrin molecule had a strong inhibitory effect on the saliva-induced aggregation of S. mutans, whereas a fragment corresponding to the N-terminal half had a weak inhibitory effect. Seven shorter fragments corresponding to lactoferrin residues 473–538 also showed a high ability to inhibit the aggregation of S. mutans. These results suggest that residues 473–538 of bovine lactoferrin are important in the inhibition of saliva-induced aggregation of S. mutans.

Streptococcus mutans has been strongly implicated in causation of dental caries, a common human disease (1,2). Colonization of the tooth surface by S. mutans is initiated by binding of the organism to salivary components on tooth surfaces (3). This binding is mediated by a 190-kDa surface protein antigen (PAc) 1 of S. mutans, variously designated as antigen I/II, B, IF, P1, SR, and MSL-1 (1,(3)(4)(5). Various salivary components have been reported to bind to S. mutans or to induce its aggregation (6 -8). We have recently shown that the PAc of S. mutans binds to a complex of high molecular mass salivary glycoprotein and secretory immunoglobulin A (sIgA) (9).
Bovine milk is commonly found in the human diet. Since bovine milk is produced on a large scale at low cost, and is easily delivered to the oral cavity, it has been used for passive immunization in prevention measures targeting several pathogens (10 -13). Bovine milk contains several protein components, including caseins, immunoglobulins, lactalbumin, lactoferrin, lactoglobulin, lactoperoxidase, and lysozyme (14). Casein and lactoperoxidase have been reported to inhibit the adherence of S. mutans to saliva-coated hydroxyapatite (15,16). -Casein reduces the glucosyltransferase activity of S. mutans, which in turn reduces glucan formation (17), and lactoferrin has a bactericidal effect on S. mutans (18).
In this study, we examined the effects of bovine milk on the saliva-induced aggregation of S. mutans cells. We purified and characterized the aggregation inhibitory activity present in milk and determined that this activity is due to lactoferrin. The interaction between lactoferrin and salivary agglutinin was further examined by surface plasmon resonance. Finally, deletion analysis of lactoferrin was used to identify the region of lactoferrin responsible for its interaction with saliva.

EXPERIMENTAL PROCEDURES
Bacterial Strains-S. mutans strains MT8148 (3) and Xc (19) were used as representative strains of S. mutans serotype c. S. mutans TK18 is a recombinant strain that produces a large amount of PAc (3). Streptococcus sanguinis ATCC 10556, Streptococcus oralis ATCC 10557, and Streptococcus gordonii ATCC 10558 were used as type strains. Escherichia coli M15[pREP4] was obtained from Qiagen. The culture media used were 2 ϫ TY broth (20) for Escherichia coli and brain heart infusion (BHI, Difco) broth for streptococci.
Saliva-Unstimulated whole saliva was collected from a single donor (male, 42 years of age) in an ice-chilled tube and clarified by centrifugation at 12,000 ϫ g for 10 min.
Salivary Agglutinin-Salivary agglutinin was isolated by the method of Oho et al. (9). Briefly, clarified whole saliva diluted 1/2 with aggregation buffer (1.5 mM KH 2 PO 4 (pH 7.2), 6.5 mM Na 2 HPO 4 , 2.7 mM KCl, 137 mM NaCl) was incubated with an equal volume of a cell suspension of S. mutans MT8148 at 37°C for 30 min. Cells were collected by centrifugation and washed with aggregation buffer, and adsorbed salivary agglutinin was eluted with the same buffer supplemented with 1 mM EDTA. The eluate was filtered (0.2-m pore size) and subjected to gel filtration chromatography on a Superdex 200 HR (Amersham Pharmacia Biotech) equilibrated with aggregation buffer. The eluate at the void volume was collected and used as salivary agglutinin. For the surface plasmon resonance analysis to examine the binding of lactoferrin, salivary agglutinin was dissociated into its components of high molecular mass glycoprotein and sIgA by electrophoretic fractionation (9).
Milk Components-Bovine ␣-casein, ␤-casein, -casein, lactalbumin, lactoferrin, and lactoperoxidase were purchased from Sigma. Bovine ␥-casein was purchased from Research Organics, and bovine lactoglobulin from ICN Biomedicals. Bovine immunoglobulin G was prepared from bovine milk, using affinity chromatography on a HiTrap protein G column (5 ml) (Amersham Pharmacia Biotech) according to the method of Oho et al. (21). Iron-saturated bovine lactoferrin and iron-free lactoferrin (apolactoferrin) were prepared from bovine lactoferrin according to the methods of Kawasaki et al. (22) and Shimazaki et al. (23), respectively. The degree of iron saturation of lactoferrin was determined by the Wako Fe-B test (Wako, Osaka, Japan). Bovine lactoferrin (Sigma) was determined to be 19.3% iron-saturated. Lactoferricin B was a gift from the Nutrition Science Laboratory, Morinaga Milk Industry Co., Zama, Japan. Protein content was determined according to the method of Lowry et al. (24), with bovine serum albumin as a standard.
Fractionation of Bovine Milk-The milk component responsible for inhibiting aggregation was isolated by subjecting bovine milk to fast protein liquid chromatography (FPLC). First, the milk fat was removed by centrifugation at 12,000 ϫ g for 15 min, and the skimmed milk was dialyzed against 10 mM imidazole HCl buffer (pH 7.0). Then, the milk sample was passed through a 0.2-m filter and applied to a Mono S HR 5/5 column (Amersham Pharmacia Biotech) that had been equilibrated with 10 mM imidazole HCl buffer (pH 7.0). After sample application, the column was washed with 5 volumes of the same buffer, and the bound material was eluted with a linear gradient (0 -1 M) of NaCl in the same buffer. Each fraction was analyzed for protein by monitoring the absorbance at 280 nm (A 280 ) and was assayed for aggregation inhibitory activity.
Sequence Determination-The N-terminal amino acid sequence of the isolated aggregation inhibitory bovine milk component was determined by Edman degradation using a Shimadzu PSSQ-21 gas-phase sequencer (Shimadzu, Kyoto, Japan).
Aggregation Assay-Streptococcal cells were suspended in aggregation buffer at an A 550 of ϳ1.5. Either 25 l of whole saliva or 10 l of salivary agglutinin (0.5 mg/ml) was mixed with 1 ml of the cell suspension and various amounts of bovine milk component, and the total volume of the reaction mixture was adjusted to 1.5 ml with aggregation buffer. CaCl 2 was added to the mixture of salivary agglutinin at a final concentration of 1 mM. Bacterial aggregation was determined by monitoring the change in A 550 at 37°C for 2 h with a UV-visible recording spectrophotometer (Ultrospec 3000, Amersham Pharmacia Biotech).
Binding of Bovine Lactoferrin to rPAc or Salivary Agglutinin-Surface plasmon resonance, which permits real-time analysis of macromolecular interactions (25), was used to examine the binding of bovine lactoferrin to rPAc, salivary agglutinin, or to components of salivary agglutinin. Binding assays were carried out with a BIAcore 2000 surface plasmon resonance biosensor (Amersham Pharmacia Biotech). First, rPAc, salivary agglutinin, high molecular mass glycoprotein separated by electrophoretic fractionation, and sIgA separated by electrophoretic fractionation were immobilized on carboxymethylated, dextran-coated, gold-surfaced CM5 sensor chips via primary amino group linkages according to the method of Johnsson et al. (26). For immobilization of each protein, 35 l of a 300 g/ml solution in 10 mM sodium acetate buffer (pH 4.5) was passed over the activated chip surface, while phosphate-buffered saline (pH 7.0) was maintained at 5 l/min throughout the immobilizing process. Binding of rPAc, salivary agglutinin, high molecular mass glycoprotein, and sIgA to the chip surfaces occurred at 5.8, 7.4, 7.1, and 10.9 ng/mm 2 , respectively. Each milk component, diluted in an appropriate running buffer, was then passed over the immobilized surface at a flow rate of 10 l/min. The effect of pH on the binding of bovine lactoferrin to salivary agglutinin was assayed in 10 mM potassium phosphate buffer (pH 2-8) containing 0.15 M NaCl. Divalent cation specificity was examined in phosphate-buffered saline (pH 7.0) containing 0 -2 mM CaCl 2 , MgCl 2 , or MnCl 2 . The dissociation phase of binding was initiated by the injection of the diluent buffer at 10 l/min. All binding experiments were performed at 25°C. The surface resonance signal in each binding cycle was expressed in resonance units (RU). A resonance of 1,000 RU corresponds to a shift of 0.1°in the resonance angle, which corresponds to a change in surface protein concentration of ϳ1 ng/mm 2 (27).
Heat Treatment-In thermal stability studies, lactoferrin was heated at 40 -100°C for 15 min and was then subjected to the surface plasmon resonance binding assay.
Bovine Lactoferrin Fragments-Truncated bovine lactoferrin fragments were prepared as 6 ϫ His-tagged fusion proteins by cloning of polymerase chain reaction-amplified lactoferrin gene fragments into expression vector pQE-30 (Qiagen). The following sets of primers were used for amplification: LfN-F, 5Ј-TATAGAGCTCATGAAGCTCTTCGT- Lactoferrin and DHFR fusion proteins were extracted from whole cell extracts of E. coli M15[pREP4] cells containing the recombinant plasmids. Cells were cultured in 2 ϫ TY broth containing 100 g/ml ampicillin and 25 g/ml kanamycin at 37°C until an A 550 of 1.0 was attained. Expression was induced by addition of isopropyl-␤-D-thiogalactopyranoside to the cultures at a final concentration of 1 mM, and the cultures were grown for 3 h. Cells were harvested by centrifugation at 5,000 ϫ g for 20 min, and one-step purification of the fusion proteins was performed with Ni 2ϩ -HiTrap chelating columns (1 ml) (Amersham Pharmacia Biotech) according to the manufacture's instructions. In brief, the cell pellet was solubilized in 10 mM Tris-HCl (pH 8.0), 0.1 M sodium phosphate, 6 M guanidine HCl (buffer A) at 5 ml/g and mixed by inversion for 1 h at 4°C. The lysate was centrifuged at 10,000 ϫ g for 20 min at 4°C, and the cleared supernatant was applied to a Ni 2ϩ -HiTrap chelating column that had been equilibrated with buffer A. The column was extensively washed with buffer A and then with 5 or more volumes of 10 mM Tris-HCl (pH 8.0), 0.1 M sodium phosphate, 8 M urea (buffer B) containing 10 mM imidazole until the A 280 of eluate was less than 0.01. The fusion proteins were eluted with buffer B containing 250 mM imidazole.
The eluted proteins were refolded by sequential dialysis against buffers containing decreasing urea concentrations for 18 h in each buffer at 4°C (28). The gradient buffers contained 4, 2, and 1 M urea in 0.1 M Tris-HCl (pH 8.0), 0.1 M sodium phosphate, and 2 mM dithiothreitol. After dialysis against 1 M urea, fusion proteins were dialyzed against 50 mM sodium phosphate (pH 8.0) containing 0.3 M NaCl for 18 h at 4°C. Each fusion protein was analyzed by SDS-PAGE.
SDS-PAGE and Western Blotting-SDS-PAGE was performed using 12.5 and 15% polyacrylamide gels according to the method of Laemmli (29). After electrophoresis, the gels were stained with Coomassie Brilliant Blue R-250. Electrophoresis calibration kits (Amersham Pharmacia Biotech) were used as molecular mass markers. For Western blotting, samples were subjected to SDS-PAGE and transferred electrophoretically to nitrocellulose membranes according to the method of Burnette (30). After blocking with 1% bovine serum albumin in Tris-buffered saline (20 mM Tris-HCl (pH 7.5), 150 mM NaCl) containing 1% Triton X-100, the membranes were treated with alkaline phosphatase-conjugated goat anti-bovine lactoferrin antiserum (Betchyl Laboratories).
Statistical Analysis-Differences between the control and the test samples in aggregation were determined by Student's t test.

Isolation and Characterization of the Milk Component That
Inhibits Aggregation-The FPLC fraction of bovine milk eluted at 0.64 M NaCl inhibited the saliva-induced aggregation of S. mutans cells (Fig. 1). Coomassie staining of the SDS gel revealed a single 80-kDa band in this fraction (Fig. 2A, lane 2). In Western blot, rabbit anti-bovine lactoferrin antiserum reacted with this band (Fig. 2B, lane 1). The N-terminal amino acid sequence of this component was Ala-Pro-Arg-Lys-Asn-Val-Arg-Trp-Cys-Thr, which corresponds to the N terminus of bovine lactoferrin (31). These results indicated that the aggregation inhibitory component is lactoferrin.
Aggregation of Streptococcal Cells-Aggregation of the typical S. mutans strain MT8148 (serotype c) in the presence of whole saliva or salivary agglutinin was examined by a spectrophotometric assay. Both whole saliva and salivary agglutinin induced strong aggregation. Testing of various bovine milk components revealed that lactoferrin inhibited this saliva-induced aggregation in a dose-dependent manner (Fig. 3). Of the milk components tested, bovine lactoferrin had the strongest inhibitory activity, whereas other components, such as lactoperoxidase, ␣-casein, and -casein, showed weak inhibitory activity (Table I). Other oral streptococci, such as S. mutans Xc, S. sanguinis ATCC 10556, S. oralis ATCC 10557, and S. gordonii ATCC 10558, were also tested for their ability to aggre-gate in the presence of whole saliva with or without bovine lactoferrin. Bovine lactoferrin showed the same inhibitory effect on the aggregation of these strains that it did on the aggregation of S. mutans MT8148 (Table II).
Binding of Bovine Lactoferrin to rPAc or Salivary Agglutinin-The binding of bovine lactoferrin to rPAc, salivary agglutinin, or to components of salivary agglutinin separated by electrophoretic fractionation was analyzed by surface plasmon resonance. Lactoferrin (50 g/ml) in phosphate-buffered saline (pH 7.0) was allowed to react with immobilized ligands on a sensor chip. The biosensor response of bovine lactoferrin to rPAc, salivary agglutinin, high molecular mass glycoprotein, and sIgA was 149 Ϯ 16, 470 Ϯ 13, 718 Ϯ 47, and 34 Ϯ 1 RU/ng of immobilized ligand, respectively (mean Ϯ S.D. of triplicate assays).
Binding of bovine lactoferrin to immobilized salivary agglutinin was enhanced by the addition of CaCl 2 to the running buffer, with an optimum concentration of 0.5 mM CaCl 2 . MgCl 2 FIG. 1. Fractionation of bovine milk by FPLC. Ten ml of defatted milk were dialyzed against 10 mM imidazole HCl buffer (pH 7.0) and then applied to a Mono S HR 5/5 column. The bound material was eluted with a linear gradient of NaCl (0 -1.0 M) in 10 mM imidazole HCl buffer (pH 7.0). Fractions were monitored for protein by their absorbance at 280 nm (q) and for their inhibitory effect on the aggregation of S. mutans cells (E).

FIG. 3. Dose-dependent inhibition of the saliva-induced aggregation of S. mutans cells by bovine lactoferrin.
S. mutans MT8148 cells grown in BHI broth were harvested and resuspended in aggregation buffer. The suspensions were adjusted to an A 550 of ϳ1.5 with aggregation buffer. The cell suspensions (1 ml) were mixed with 25 l of whole saliva and various amounts of lactoferrin, and the total volume of the reaction mixture was adjusted to 1.5 ml. Aggregation was measured by the reduction in A 550 after 2 h. Percent inhibition was calculated as 100 ϫ [(a Ϫ b)/a], where a is the mean value without lactoferrin (control), and b is the mean value with lactoferrin. Values are given as the means Ϯ S.D. of triplicate assays. and MnCl 2 did not enhance binding (data not shown). In thermal stability studies, the biosensor response induced by binding of bovine lactoferrin to immobilized salivary agglutinin gradually decreased as the temperature used to heat the lactoferrin was raised. However, lactoferrin still bound to salivary agglutinin even after heating at 100°C (Fig. 4A). The pH maximum for binding of bovine lactoferrin to salivary agglutinin was pH 4.0, and no detectable binding occurred at pH 2.0 (Fig. 4B).
Effects of Lactoferrin Fragments on the Aggregation of S. mutans Cells-To identify the saliva-binding region of the bovine lactoferrin molecule, 11 6 ϫ His-tagged lactoferrin fragments were cloned and expressed in E. coli. These fusion proteins were purified and used in spectrophotometric aggregation assays. SDS-PAGE analysis of each lactoferrin fragment showed a single band (data not shown). The N-terminally truncated lactoferrin fragment, LfC (residues 345-689), strongly inhibited saliva-induced aggregation of S. mutans cells, whereas the C-terminally truncated fragment LfN (residues 1-344) weakly inhibited the aggregation (Fig. 5). Fragments Lf4 (residues 345-571), Lf41 (residues 345-538), Lf45 (residues 366 -571), Lf46 (residues 399 -571), and Lf47 (residues 432-571) also exhibited strong inhibition of saliva-induced aggregation of S. mutans, as did the shorter fragment Lf411 (residues 473-538). In contrast, fragments Lf43 (residues 345-472) and Lf44 (residues 345-439) exhibited only weak inhibitory activity. The 6 ϫ His-tagged DHFR, which was used as control, also weakly inhibited aggregation. DISCUSSION Human saliva induces aggregation of S. mutans via an interaction between PAc of the organism and salivary agglutinin, which is a complex of high molecular mass glycoprotein and sIgA (3,9). Gong et al. (32) also showed that salivary film on hydroxyapatite contains a complex of macromolecular protein enriched in sIgA and ␣-amylase, which forms a S. sanguinisbinding site. In this study, we showed that bovine milk lacto-   ferrin inhibited the saliva-induced aggregation of S. mutans cells. The binding of bovine lactoferrin to rPAc, salivary agglutinin, and components of salivary agglutinin was examined using surface plasmon resonance. Bovine lactoferrin bound more strongly to salivary agglutinin, especially to high molecular mass glycoprotein, than to rPAc, suggesting that bovine lactoferrin may inhibit the interaction between PAc and salivary agglutinin by binding to high molecular mass glycoprotein of salivary agglutinin. Aggregation of other streptococcal cells induced by whole saliva was also inhibited by bovine lactoferrin, indicating that the inhibitory effect of lactoferrin is not specific for S. mutans.
The optimal pH for the binding of bovine lactoferrin to salivary agglutinin was 4.0, and the stability of lactoferrin to bind to salivary agglutinin was not affected by previous heat treatment. The isoelectric point of bovine lactoferrin is ϳ8.0 (33). It can be sterilized at high temperatures at pH 4.0 without any significant loss of bactericidal activity, suggesting that it is thermally stable at pH 4.0 (34). Bovine lactoferrin may adopt a conformation suitable for interaction with salivary agglutinin at this pH as well.
Lactoferrin is an iron-binding glycoprotein, and its iron-binding capacity is associated with many biological functions (35,36). The lactoferrin preparation used in this study was 19.3% iron-saturated. To examine the role of iron binding in inhibition of S. mutans aggregation, we also prepared apolactoferrin and iron-saturated lactoferrin and assayed them for their ability to inhibit the saliva-induced aggregation. No significant differences were observed among the inhibitory properties of these three types of lactoferrin (data not shown). These results are consistent with those of Soukka et al. (37), who observed that these three types of lactoferrin cause no difference in the binding of S. mutans, although the assay was performed using saliva-coated hydroxyapatite. These results suggest that iron ion in lactoferrin does not play a significant role in the binding of bovine lactoferrin to salivary agglutinin. In another experiment, Soukka et al. (38) showed that apolactoferrin effectively agglutinates S. mutans cells but not the other bacteria. However, our preliminary studies have shown that all of the three types of lactoferrin did not induce the aggregation of S. mutans cells. 2 The cause of this discrepancy may be ascribed to differences in strain of S. mutans used or the experimental condition.
To identify the saliva-binding region of the lactoferrin mole-cule, we prepared a series of truncated lactoferrin fragments and assayed their effects on the saliva-induced aggregation of S. mutans cells. Our results suggest that lactoferrin residues 473-538 play an important role in the inhibition of salivainduced aggregation of S. mutans. Other fragments lacking these residues, such as LfN (residues 1-344), Lf43 (residues 345-472), and Lf44 (residues 345-439), exhibited only weak inhibitory activity. The lactoferrin molecule is proposed to consist of two lobes (N-lobe and C-lobe) (40). The N-lobe contains the active domains for bactericidal action and heparin binding (31,41), whereas the C-lobe contains a functional domain for hepatocyte binding and internalization (42). In these previous studies, lactoferrin fragments were prepared by tryptic cleavage of lactoferrin and isolated by high performance liquid chromatography. Here, we prepared truncated lactoferrin fragments using recombinant DNA technology. Our results indicate that the lactoferrin domain responsible for binding to salivary agglutinin is within the C-lobe of the protein.
The mechanism of binding of lactoferrin to salivary agglutinin remains unclear. The predicted pI value and secondary structure of each lactoferrin fragment were obtained using the DNA software package, DNASIS (Hitachi Software Engineering, Tokyo, Japan). Secondary structure was predicted according to the method of Chou and Fasman (43). Although all the active fragments containing residues 473-538 had acidic pI values, the inactive fragment Lf44 also had an acidic pI value (pI ϭ 5.2). Therefore, electrostatic interactions do not seem to be involved in agglutinin binding. Furthermore, the inhibitory fragments of lactoferrin did not retain characteristic secondary structures. Lactoferricin B, a 25-amino acid peptide derived from the N-lobe of bovine lactoferrin, has bactericidal activity (44). The antibacterial properties of lactoferricin B are attributed to the disruption of target cell membranes by the basic residues arrayed along the outside of the lactoferricin B molecule (45). We found that lactoferricin B had no inhibitory effects on the saliva-induced aggregation of S. mutans cells (data not shown). Further studies are necessary to elucidate the mechanism by which active lactoferrin fragments inhibit the salivainduced aggregation of S. mutans.
There are two types of bacterial interaction with salivary components; saliva-induced bacterial aggregation in solution phase and bacterial adherence to salivary components adsorbed on the tooth surface. Gibbons and Hay (46) and Raj et al. (47) reported that proline-rich proteins and statherin serve as pellicle receptors for some of streptococcal strains, but do not 2 M. Mitoma, T. Oho, Y. Shimazaki, and T. Koga, unpublished data.

FIG. 5. Inhibition of saliva-induced aggregation of S. mutans cells by lactoferrin fragments. S. mutans
MT8148 cells grown in BHI broth were harvested and resuspended in aggregation buffer. The suspensions were adjusted to an A 550 of ϳ1.5 with aggregation buffer. The cell suspensions (1 ml) were mixed with 25 l of whole saliva and 1 nM lactoferrin or lactoferrin fragment, and the total volume of the reaction mixture was adjusted to 1.5 ml. Aggregation was measured by the reduction in A 550 after 2 h. Percent inhibition was calculated as 100 ϫ [(a Ϫ b)/a], where a is the mean value without lactoferrin preparation (control), and b is the mean value with lactoferrin preparation. Values are given as the means Ϯ S.D. of triplicate assays. *, p Ͻ 0.05; **, p Ͻ 0.01; ***, p Ͻ 0.001 (compared with DHFR).
induce aggregation of the organisms in suspension. On the basis of these findings, Gibbons (48) proposed a model that an apparent conformational change occurs when salivary components bind to hydroxyapatite, which exposes the binding sites for bacterial adhesin. This explains the difference between bacterial aggregation and adherence. In the present study, we found that lactoferrin in bovine milk possessed inhibitory activity against saliva-induced aggregation of S. mutans in solution phase. Therefore, we are unable to exclude the possibility that milk components other than lactoferrin may possess inhibitory effect on the binding of bacterial cells to a salivary film. Further studies are necessary to clarify effects of milk components on the adherence of bacterial cells to a salivary film.
Lactoferrin attracted a great deal of attention for its wide variety of functions (39). Lactoferrin is viewed as a potential contributor to dental caries prevention by virtue of its inhibitory effect on the binding of S. mutans to acquired pellicles on the tooth surface and its bactericidal action on S. mutans (18).
We have now demonstrated that bovine lactoferrin inhibits the interaction between PAc of S. mutans and salivary agglutinin by binding strongly to salivary agglutinin. Residues 473-538 of bovine lactoferrin play an important role in the interaction of lactoferrin with salivary agglutinin.