Isolation and characterization of monoclonal antibodies specific for ...

3 downloads 54 Views 3MB Size Report
May 18, 1987 - WILLIAM P. McARTHUR,1 AND A. S. BLEIWEISl* .... 1987, K106, p. 220; ...... Nesbitt, W. E., R. H. Staat, B. Rosan, K. G. Taylor, and R. J. Doyle.
INFECTION AND IMMUNITY, Nov. 1987, p. 2759-2767 0019-9567/87/112759-09$02.00/0 Copyright © 1987, American Society for Microbiology

Vol. 55, No. 11

Isolation and Characterization of Monoclonal Antibodies Specific for Antigen P1, a Major Surface Protein of Mutans Streptococci GREGG Y. AYAKAWA,1 LEE W. BOUSHELL,' PAULA J. CROWLEY,' GREGORY W. ERDOS,2 WILLIAM P. McARTHUR,1 AND A. S. BLEIWEISl* Departments of Oral Biologyl and Microbiology and Cell Science,2 University of Florida, Gainesville, Florida 32610 Received 18 May 1987/Accepted 4 August 1987

A panel of 15 murine monoclonal antibodies (MAbs; 14 immunoglobulin Gl, 1 immunoglobulin G2a) directed against antigen P1, a major surface protein of mutans streptococci, was prepared. All of these MAbs reacted by the enzyme-linked immunosorbent assay with solubilized wall material from Streptococcus mutans Ingbritt 175 (a serotype c strain which retains significant amounts of P1 in its cell wall), culture supernatant fluid from Ingbritt 162 (a strain which excretes large amounts of P1 into the culture medium), and purified P1. By Western immunoblotting, these MAbs were observed to react with a high-molecular-weight polypeptide which comigrated with antigen P1. None of these MAbs cross-reacted with human heart tissue or with various eucaryotic proteins. When whole cells of various strains of mutans streptococci were screened against the panel of MAbs, the strongest reactivities were noted with strains of serotype c and e S. mutans, while a serotype f strain of S. mutans, along with S. sobrinus and S. cricetus strains, reacted somewhat more weakly. S. rattus strains were completely negative. Results obtained with bacterial culture supernatants were qualitatively similar. The surface localization of antigen P1 was confirmed by electron microscopy with an indirect immunogold technique. In sectioned S. mutans cells, labeling appeared to be associated with a fibrillar "fuzzy coat" layer, which was far more prominent on cells of Ingbritt 175 than on those of Ingbritt 162.

Investigations in several laboratories (11, 23, 25-27, 33) have revealed the existence of a number of proteins closely associated with the cell wall of mutans streptococci. Most prominent among these is a high-molecular-weight component variously designated antigen B (26, 27), IF (16), P1 (11), and I/II (25, 33). Forester et al. (11) demonstrated that these proteins are, in fact, immunologically and biochemically identical. For the sake of convenience, we will use the designation P1 in this communication. Antigen P1 is a glycoprotein of molecular weight 185,000 (11, 26, 27) which is apparently covalently linked to the streptococcal cell wall, since treatment of walls with sodium dodecyl sulfate (SDS) fails to remove it completely (26, 29). However, P1 is also readily detectable in spent culture fluids (11, 25, 26, 33), indicating that a proportion of the protein is released into the medium during growth. Although P1 was first described for serotype c strains of Streptococcus mutans, analogous proteins have since been found in strains representing all serotypes of mutans streptococci with the exception of serotype b (S. rattus) (27). Thus, three of the four species of mutans streptococci (S. mutans, S. sobrinus, and S. cricetus) possess P1-like proteins. Both crossreactive and species-specific antigenic determinants are present in this family of high-molecular-weight proteins (13, 25, 27, 30). Several studies (18, 19, 28) have demonstrated that P1 is an effective vaccine antigen in animal models of dental caries, but there is evidence, albeit controversial, that this moiety may cross-react immunologically with human heart tissue (11, 16, 26, 27). At present, the issue of heart crossreactivity remains unresolved, as polyclonal non-affinitypurified antisera were used in previous studies of this phenomenon. Smith et al. (30), using monoclonal antibodies (MAbs) to their antigen I/II, failed to detect any crossreactivity with heart tissue. *

The function of antigen P1 also remains unclear. Douglas and Russell (7, 8) have shown that rabbit antisera to this protein inhibit adherence of S. mutans to glass surfaces and to saliva-coated hydroxyapatite. Curtiss and his colleagues (4-6, 14) cloned this protein from S. sobrinus 6715 (serotype g) in Escherichia coli and found that antisera prepared against this cloned product (which they designated SpaA) inhibit sucrose-induced aggregation of this strain. In addition, SpaA- mutants of S. sobrinus are defective in sucrosedependent aggregation and also exhibit a very low level of cariogenicity in gnotobiotic rats (5, 6). There is also evidence that SpaA is a glucan-binding protein (4). These studies clearly demonstrate the crucial role played by SpaA in the cariogenicity of S. sobrinus strains. Although the close relationship between SpaA in S. sobrinus and P1 in S. mutans would strongly support similar functions for both proteins, definitive evidence concerning P1 is lacking at present. As a first step towards answering some of the unresolved questions about this major surface protein, we have prepared a panel of MAbs directed against antigen P1. In this report, we describe the production of these MAbs and our initial characterization studies with them. We also discuss the application of these MAbs in the screening of various strains of mutans streptococci for P1-cross-reactive epitopes and in the ultrastructural localization of these determinants. (This work was presented in part previously [G. Y. Ayakawa, P. J. Crowley, G. W. Erdos, and A. S. Bleiweis, Abstr. Annu. Meet. Am. Soc. Microbiol. 1987, K106, p. 220; G. Y. Ayakawa, L. W. Boushell, W. P. McArthur, and A. S. Bleiweis, Abstr. Annu. Meet. Am. Soc. Microbiol. 1987, K107, p. 220].) MATERIALS AND METHODS Bacterial strains and growth conditions. The bacterial strains used in this study are listed in Table 1. Five of the

Corresponding author. 2759

2760

AYAKAWA ET AL.

INFECT. IMMUN.

TABLE 1. Bacterial strains used in this study Species (serotype) and strain

S. cricetus (a) AHT ..............

Source" Little

Bleiweis E49 .............. HS-1 .............. Bleiweis Bleiweis HS-6 ..............

OMZ61 ..............

Little

S. rattus (b) BHT .............. FA-1 ..............

Bleiweis Bleiweis

S. mutans (c/e/f)

P1 retainers Ingbritt 175 (c) .............. NG7 (c) .............. NG8 (c) .............. P1 nonretainers Ingbritt 162 (c) .............. NG5 (c) .............. Others B66 (c) .............. GS-5 (c) .............. NCTC 10449 (c) .............. B2 (e) .............. B14 (e) .............. V100 (e) .............. OMZ175 (f) ..............

Knox Knox Knox Knox Knox

Bleiweis Bleiweis Bleiweis Bleiweis Bleiweis Bleiweis Bleiweis

S. sobrinus (d/g/h) B13 (d) .............. Bleiweis

SL-1 (d) .............. 6715 (g) .............. KlR (g) .............. GV266 SC-4 (g)b .............. ATCC 33748 (h) ..............

Bleiweis Bleiweis Bleiweis

Fitzgerald

ATCC

a Abbreviations: ATCC, American Type Culture Collection, Rockville, Md.; Bleiweis, culture collection of A. S. Bleiweis, University of Florida; Fitzgerald, R. J. Fitzgerald, Veterans' Administration Hospital, Miami, Fla.; Knox, K. W. Knox, Institute of Dental Research, Sydney, Australia; Little, W. Little, National Institute of Dental Research, Bethesda, Md. b This strain is a derivative of 6715 which excretes large amounts of P1 and was originally isolated by J. Tanzer, University of Connecticut, Farmington.

serotype c strains were categorized as P1 retainers or P1 nonretainers by K. W. Knox, Institute of Dental Research, Sydney, Australia. The former group comprises strains which retain significant amounts of P1 in their cell walls, while the latter designation refers to isolates which excrete virtually all of their P1 into the culture medium. Cells were grown aerobically overnight in the chemically defined medium of Terleckyj et al. (32) at 37°C as previously described (1). Preparation of mouse immunogen. SDS-washed cell walls were prepared from S. mutans Ingbritt 175 essentially as described by Russell et al. (29). Briefly, this procedure involves breaking the cells with glass beads, followed by extensive washing with distilled water, 1 M NaCl, 1% (wt/vol) SDS, and again with distilled water. Walls were stored as a suspension in distilled water (containing 0.1% sodium azide as a preservative) at 4°C. Solubilized cell wall material was prepared by mutanolysin digestion of the SDS-washed walls by the procedure of Knox et al. (17). Mutanolysin was purchased from Miles Scientific, Div. Miles Laboratories, Inc., Naperville, Ill. This solubilized wall material was stored in small portions at -200C. Immunization of mice. Female BALB/c mice (Charles

River Breeding Laboratories, Wilmington, Mass.) were immunized intraperitoneally with SDS-washed cell walls (50 ,ug of protein in complete Freund adjuvant). Two weeks later, a second identical dose was administered to the mice. The animals were test bled from the retroorbital plexus 1 week following the second immunization. These sera were screened at a 1:100 dilution by the enzyme-linked immunosorbent assay (ELISA) against glutaraldehyde-fixed whole cells of Ingbritt 175 (3) and antigen P1 (obtained from K. W. Knox). Mice exhibiting the highest titers of antibody were given intravenous boosters with either solubilized wall material or antigen P1 (10 [Lg of protein). Fusion and screening of hybridomas. Three days after the booster immunization, mice were sacrificed by cervical dislocation. Spleens were removed aseptically, and spleen cells were fused with cells of the Sp2/O-Agl4 nonsecreting mouse myeloma line by the method of Oi and Herzenberg (24), with polyethylene glycol as the fusing agent. Hybrids were selected in RPMI 1640 medium containing 15% fetal calf serum (Hyclone Laboratories, Inc., Logan, Utah) and supplemented with hypoxanthine, aminopterin, and thymidine. Hybridomas were screened for antibody production by ELISA with either glutaraldehyde-fixed whole cells or solubilized wall material from Ingbritt 175. The former screening procedure would identify clones secreting MAbs specific for P1 epitopes exposed at the cell surface, while the latter method would enable one to detect antibodies binding to both exposed and buried determinants. Positive cultures were expanded in RPMI 1640 containing 15% fetal calf serum and cloned twice in semisolid agarose (SeaPlaque; FMC Corp., Rockland, Maine). Stable antibody-producing hybridoma lines were stored as frozen suspensions (in 10% dimethyl sulfoxide) in liquid N2. Initial characterization of MAbs. MAbs secreted by cloned hybridoma lines were screened by ELISA and Western blotting (immunoblotting) against solubilized wall proteins of Ingbritt 175, ammonium sulfate-precipitated culture supernatant proteins of Ingbritt 162 (26), and antigen P1. In addition, MAbs were screened by ELISA against an SDS extract of human heart tissue (1), rabbit heart actin, rabbit skeletal muscle myosin, human fibrinogen, fibronectin, and keratin (all from Sigma Chemical Co., St. Louis, Mo.), cloned M protein from type 6 Streptococcus pyogenes (courtesy of V. Fischetti, Rockefeller University, New York), and a pepsin extract (2) of S. pyogenes Manfredo (M type 5). The last strain was obtained from M. W. Cunningham, University of Oklahoma Health Sciences Center, Oklahoma City. The ELISA was performed as outlined by Ayakawa et al. (1), except that phosphate-buffered saline (PBS) containing 1% (wt/vol) bovine serum albumin (Sigma) was used as a blocking and antibody dilution reagent. Hybridoma culture fluids served as a source of MAbs. Western blotting was carried out as described previously (1), with some modifications. A 7.5% acrylamide resolving gel was used for the separation of SDS-solubilized polypeptides. Blots were stained for total protein with a commercially available colloidal gold reagent (AuroDye; Janssen Life Sciences Products, Piscataway, N.J.). Lastly, blots were developed with a plastic incubation tray (Bio-Rad Laboratories, Richmond, Calif.), which allowed the use of smaller volumes of reagents for washing and development. Screening of bacterial culture supernatants against MAbs. Concentrated culture supernatants were prepared from overnight cultures of streptococci in defined medium. Cells were harvested by centrifugation, and the supernatants were

MAbs TO S. MUTANS SURFACE PROTEIN P1

VOL. 55, 1987

passed through a filter (0.45-,um pore size) to remove any remaining cells. Sodium azide (0.1%) was added to the spent medium, which was then concentrated 10- to 20-fold through a YM-10 ultrafiltration membrane (Amicon Corp., Danvers, Mass.). The concentrated supernatant was then dialyzed against several changes of 0.1% sodium azide at 4°C to remove amino acids which would interfere with the protein assay. Supernatants were stored at 4°C. Polypeptides in culture supernatants were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to nitrocellulose sheets as described elsewhere (1). The sheets were incubated with MAbs in a Miniblotter II apparatus (Immunetics, Cambridge, Mass.), which allowed us to use 200 jil or less of each hybridoma culture fluid. The sheets were then washed and developed as previously described (1). Protein assay. Protein contents of samples were determined by a modification of the method of Lowry et al. (20), with bicinchoninic acid as the colorimetric detection reagent (BCA Protein Assay Reagent; Pierce Chemical Co., Rockford, Ill.). Immunoelectron microscopy. Streptococcal cells from lateexponential- or stationary-phase cultures were harvested and washed three times in PBS. They were then incubated in hybridoma culture fluid (diluted 1:10 in PBS) at 4°C for 24 h. RPMI 1640 containing normal mouse immunoglobulin G (IgG) (Cappel Laboratories, Malvern, Pa.) served as a negative control. After three washes in PBS, the cells were incubated in affinity-purified goat anti-mouse IgG (Cappel) coupled to colloidal gold (ca. 15-nm diameter) by the method of Horisberger and Rosset (15). This second antibody incubation was for 1 h at room temperature or overnight at 4°C. For whole-mount electron microscopy, the cells were then washed, fixed briefly in 4% paraformaldehyde in PBS, mounted on Formvar-coated grids, and observed with a JEOL JEM 100 C-X electron microscope. For thin sectioning, cells were prepared as described above except that they were fixed in 2.5% glutaraldehyde for 1 h, followed by 1%

2761

FIG. 1. Western blot analysis of a representative anti-Pl MAb with three antigenic preparations: ammonium sulfate-precipitated culture supernatant proteins from S. mutans Ingbritt 162 (162 S, 10 ,ug of protein per lane), antigen P1, (P1, 5 ,ug of protein per lane), and mutanolysin-solubilized SDS-washed cell walls from S. mutans Ingbritt 175 (175 W, 20 jig of protein per lane). Lanes: S, AuroDye protein stain; M, anti-Pl MAb 4-1OA; P, IgG fraction from a polyclonal rabbit anti-Pl serum (courtesy of K. Knox) diluted 1:200; C, control MAb (3-4D) specific for S. rattus BHT membranes. Immunoblots were developed with peroxidase-conjugated secondary antibody as described in the text. Labels to the left of the blots denote the positions of antigen P1 and the Mrs (103) of lower-

molecular-weight polypeptides. TABLE 2. Anti-Pl MAbs ELISA reactivity (A492)

MAba

1-6F 4-2H

6-11Ab 2-9E 3-3B 3-8A 3-8D 3-9B 3-10E 4-9D

2-8Gb 4-1OAb 5-3Eb 5-5Db 6-8Cb 3 4Dc

Conjugate control

Ingbritt 175 (P1 retainer)

Ingbritt 162 (P1 nonretainer) culture

wall proteins

supernatant proteins

0.058 0.102 0.153 0.258 0.535 0.132 0.260

0.116 0.170 0.173 0.422 0.572 0.289 0.262 0.216 0.241 0.331 0.562 0.469 0.485 0.314 0.117 0.007 0.002

0.111 0.176 0.180 0.415 0.327 0.269 0.391 0.175 0.019 0.019

P1

0.161 0.250 0.141 0.494 1.064 0.388 0.342 0.379 0.246 0.473 0.653 0.564 0.690 0.541 0.245 0.002 0.004

All MAbs were of the IgGl subclass except 3-3B (IgG2a). b These MAbs were isolated from mice given boosters of antigen P1 (see Materials and Methods). All remaining anti-Pl MAbs were obtained from mice given boosters with Ingbritt 175 cell walls. c Negative control MAb specific for S. rattus BHT membranes. a

OS04, dehydrated in ethanol, and embedded in Spurr resin (31). To enhance staining of the surface "fuzzy coat" of the cells (see Fig. 5), 0.5% tannic acid was sometimes included in the primary fixative of embedded cells. TABLE 3. Reactivity of anti-Pl MAbs against concentrated culture supernatants of mutans streptococci by Western blotting No. of positive Species (serotype) and strain

MAbs/no. tested

S. mutans (c/e/f) Ingbritt 162 (c) ........................................ 15/15 V100 (e) ......................................... 15/15 OMZ175 (f) ......................................... 3/13

S. cricetus (a) OMZ61 ......................................... HS-6 .........................................

1/13 0/13

S. sobrinus (d/g/h) B13 (d) ......................................... SL-1 (d) .......................................... GV266 SC-4 (g) ....................................... ATCC 33748 (h) .......................................

0/15 3/15 0/15 0/15

2762

AYAKAWA ET AL.

INFECT. IMMUN.

TABLE 4. Reactivity of MAbs with whole cells of S. mutans strains by ELISA Serotype and strain

Serotype c P1 nonretainers Ingbritt 162 NG5 P1 retainers Ingbritt 175 NG7 NG8 Others NCTC 10449 GS-5 B66

No. of positivea MAbs/no. tested

Mean A492 of positive MAbs

0/15

0/15 7/15 12/15 8/15

0.260 0.188 0.220

0/15 0/15 8/13

0.274

0/14 7/15 9/15

0.196 0.187

6/13

0.156

Serotype e B14 B2 V100

Serotype f, OMZ175

" An A492 of 0.08 or greater was used as an indication of a positive reaction.

RESULTS

Production and characterization of MAbs. From a total of four separate fusions involving six mice, 15 stable anti-Plproducing hybridoma lines were isolated. The reactivities of these MAbs with ammonium sulfate-precipitated culture supernatant proteins from S. mutans Ingbritt 162, solubilized cell wall material from S. mutans Ingbritt 175, and purified P1 were tested by ELISA. The first three MAbs listed in Table 2 were obtained from fusions in which the hybrids were screened with whole Ingbritt 175 cells as the antigen, which resulted in a disappointing yield of anti-Pl clones with relatively low reactivities. Subsequent fusions were screened with solubilized cell wall material from Ingbritt 175, which allowed us to detect MAbs binding to both surface-exposed and buried determinants of the P1 molecule. These fusions yielded a greater number of anti-Pl clones with a higher level of reactivity against the test antigens (Table 2). In general, it appeared that the MAbs obtained from mice boosted with antigen P1 (Table 2) displayed stronger reactivities with the test antigens than did the remaining MAbs. These MAbs were also tested by Western blotting against the same three antigens. Results with a representative MAb (4-1OA) are presented in Fig. 1; the other MAbs yielded essentially identical results. The MAbs reacted strongly with a polypeptide which comigrated with antigen P1. Several lower-molecular-weight polypeptides also reacted with the MAbs; these may represent proteolytic breakdown products of the parent molecule. P1 has been reported to be very sensitive to such degradation (11). Alternatively, some of these polypeptides may represent unrelated molecules which possess epitopes cross-reactive with P1 determinants. The negative control MAb 3-4D did not bind to any of the polypeptides in these preparations. Screening of culture supernatants of mutans streptococci against MAbs. Concentrated culture supernatants from several strains of mutans streptococci were screened by Western blotting against the panel of MAbs (Table 3). The serotype c and e strains of S. mlutans reacted with all of the MAbs, while the serotype f strain only reacted with three of

the antibodies tested. This may reflect a fundamental difference between the serotype f protein and those of serotypes c and e. Strains representing other species of mutans streptococci only reacted with an occasional MAb. These culture supernatants were also screened with a polyclonal rabbit anti-Pl IgG preparation (obtained from K. Knox), and all were found to contain P1-like proteins (data not shown). Screening of whole cells of mutans streptococci against MAbs. Glutaraldehyde-fixed whole cells of 24 strains of mutans streptococci were screened by ELISA against the anti-Pl MAbs. Results with S. mutans strains are shown in Table 4, while the data obtained with other species of mutans streptococci are summarized in Table 5. None of the MAbs bound to either of the Pl-nonretaining serotype c strains, while the three P1 retainers reacted with 47 to 75% of the antibodies tested (Table 4). This variability in reactivity among the P1 retainers may be due to different degrees of P1 retention. Of the remaining serotype c strains tested, two (NCTC 10449 and GS-5) exhibited no reactivity with any of the MAbs. These long-established laboratory strains may have lost the ability to retain significant quantities of P1 because of prolonged in vitro cultivation, a phenomenon which has been observed previously (21). Binding of the MAbs to serotype e and f S. mutans strains was also strain variable (Table 4). Reactivity of the MAbs with whole cells of other species of mutans streptococci was generally weaker than with those of S. mutans (Table 5). As expected, strains of S. rattus were negative, since this species lacks any wall protein analogous to P1. Isolates of S. cricetus and S. sobrinus reacted to various degrees with the MAbs. These data confirm the existence of shared epitopes between the P1-like proteins found in the various species of mutans streptococci. The fact that more MAbs demonstrated positive reactivity with the streptococcal strains by the whole-cell ELISA than by the Western blotting screen may reflect differences in techniques. Unfortunately, we found that the ELISA was unsuitable for screening the concentrated culture supernatants (unpublished observations). Alternatively, a subtle conformational difference may exist between the secreted and cell-bound forms of antigen P1 which results in differing patterns of reactivity, or the secreted antigen might undergo a greater degree of proteolysis, resulting in loss of reactive

epitopes. TABLE 5. Reactivity of MAbs with whole cells of other species of mutans streptococci by ELISA Species (serotype)

No. of positivea MAbs/no. tested

Mean A492 of positive MAbs

S. cricetus (a) E49 OMZ61 AHT HS-6 HS-1

8/15 7/13 2/13 0/15 0/15

0.097 0.177 0.106

S. rattus (b) BHT FA-1

0/15 0/15

S. sobrinus (d/g/h) B13 (d) SL-1 (d) 6715 (g) KlR (g) ATCC 33748 (h)

7/14 9/14 0/14 5/15 1/13

and strain

" See Table 4. footnote a.

0.093 0.127 0.125 0.106

MAbs TO S. MUTANS SURFACE PROTEIN P1

VOL. 55, 1987

2763

4 .

Vs

'.I as

A FIG. 2. Whole-mount immunoelectron microscopy of S. mutans Ingbritt 175 cells reacted with MAbs 6-11A (A) and 1-6F (B). MAb binding was detected by using a colloidal gold reagent (see text). Arrowheads in A indicate the apparent binding of the MAb to extracellular structures. Bars, 0.1 pum (A) and 0.2 p.m (B).

These strains were also screened with the polyclonal anti-Pl rabbit antibody. The results obtained by this procedure essentially paralleled those seen with the MAbs (data not shown). The S. mutans strains reacted somewhat more strongly than did strains of other species, with the exception of GS-5 and NCTC 10449, which were negative with the polyclonal antibody as well as with the MAbs. Interestingly, the two Pl-nonretaining serotype c strains (Ingbritt 162 and NG5) did react with the polyclonal anti-Pl, although to a significantly lesser extent than did the P1-retaining strains. This indicates that the antigen is indeed present on these nonretainers and that the differences between the two categories are probably quantitative in nature. Reactivity of MAbs with heterologous antigens. Because of the reported cross-reactivity of antigen P1 with human heart tissue (11, 16, 26, 27) and the possible functional relationship between this moiety and the M protein of S. pyogenes (12), we screened our panel of MAbs by ELISA against several heterologous antigens. These included detergent extracts of human heart tissue (1), rabbit heart actin, rabbit skeletal muscle myosin, human fibrinogen, fibronectin, keratin, cloned type 6 M protein (10), and a pepsin extract from a type 5 strain of S. pyogenes (2). In addition, the MAbs were screened against S. rattus BHT membranes, since our previous work (1, 9) has indicated the presence of heart-crossreactive polypeptides in these membranes. None of these antigens demonstrated reactivity with any of the 15 anti-Pl MAbs (data not shown). Electron microscopy. To localize antigen P1 on the surface of S. mutans cells at the ultrastructural level, an indirect immunogold labeling technique was used. To select appropriate MAbs for these studies, all 15 of the MAbs were screened against whole cells of Ingbritt 175. Surprisingly, many of the MAbs which reacted strongly with the antigenic preparations listed in Table 2 did not react at all with the whole cells by the immunogold technique. These antibodies may bind to epitopes of P1 which are not exposed at the cell surface in this strain. We chose two MAbs, 6-11A and 1-6F, for further study. Whole cells of Ingbritt 175 (P1 retainer) were reacted with MAbs 6-11A and 1-6F (Fig. 2). Labeling of the cell surface by 6-11A was extensive (Fig. 2A), and a significant fraction of the gold particles appeared to be associated with structures exterior to the cell surface (arrowheads). With MAb 1-6F, the gold label on Ingbritt 175 was more closely bound to the cell surface than it was with 6-11A (Fig. 2B). These differences in localization suggest that the two MAbs may

react with different epitopes on the P1 molecule. Surprisingly, the Ingbritt 162 (P1 nonretainer) cells also appeared to be labeled by both MAbs to a considerable extent (data not shown). This labeling may reflect leakage of P1 from the cells, since much of it was localized to the septal region between cells, where wall synthesis is most active. Also, a high percentage of the colloidal gold particles appeared to be nonspecifically trapped between cells in large clumps. Alternatively, these MAbs may bind to a portion of the P1 molecule which remains cell associated even in a nonretaining strain, with the lack of ELISA reactivity (Table 4) attributable to the glutaraldehyde treatment of the cells used to coat the ELISA plates. Whole cells of four other strains of mutans streptococciS. mutans OMZ175 (serotype f) and V100 (serotype e), S. rattus BHT (serotype b), and S. sobrinus B13 (serotype d)were treated by the same method with MAb 1-6F. Both S. mutans strains demonstrated significant binding of the MAb (Fig. 3), while the remaining two strains were negative (data not shown). To more precisely localize the reactive epitopes on the surface of S. mutans cells, thin sections of previously labeled organisms were prepared and examined by electron microscopy (Fig. 4). MAb 1-6F labeled both Ingbritt 162 and Ingbritt 175, confirming the results obtained with unsectioned cells (Fig. 4A and B). Interestingly, MAb 6-11A did not label the Pl-nonretainer Ingbritt 162 cells (Fig. 4C) but did react with the P1-retaining Ingbritt 175 cells (Fig. 4D). Thus, the labeling of the unsectioned cells of Ingbritt 162 by 6-11A (data not shown) appears to be a nonspecific phenomenon, although we cannot rule out the possibility that processing of the cells for thin sectioning may have resulted in a loss of the cognate epitope from this strain. These results indicate that these two MAbs recognize different epitopes on the Ph molecule, with the 1-6F determinant being retained by a strain which excretes most of its Ph into the culture medium and the 6-11A epitope being lost in such strains. Closer inspection of Fig. 4 reveals that labeling of the cells was associated with a fuzzy coat layer on the surface. A portion of Fig. 4D was enlarged to demonstrate this labeling in greater detail (Fig. 4E). Figure 5 shows tannic acid-stained sections of both Ingbritt 175 (Fig. 5A) and Ingbritt 162 (Fig. 5B). The fuzzy coat was far more prominent on the surfaces of the P1-retaining Ingbritt 175 cells than on those of its Pl-nonretaining counterpart, confirming an association between this layer and antigen Ph.

2764

AYAKAWA ET AL.

IN FECT . I MMUN .

FIG. 3. Whole-mount immunoelectron microscopy of S. mutans OMZ175 (A) and V100 (B) cells reacted with MAb 1-6F. Bar, 0.2 ,um.

DISCUSSION In this paper, we have described the preparation of a panel of 15 MAbs directed against antigen P1. Our characterization studies indicated that these MAbs react most strongly with strains of S. mutans serotypes c and e, while reactivity with a serotype f strain of S. mutans and with S. sobrinus and S. cricetus isolates was weaker (Tables 4 and 5). Smith et al. (30) tested a group of five MAbs directed against their antigens I/II (P1), I, and II with both ammonium sulfateprecipitated culture supernatants and whole cells as the antigen. They found the strongest reactivities associated with serotype c S. mutans, while culture supernatant from a serotype e strain and whole cells from serotypes e and f reacted somewhat more weakly (30). S. cricetus AHT also reacted weakly with an MAb directed against antigen I. Since these researchers only examined seven strains of mutans streptococci (30), it is difficult to compare their results with ours, but in general their data would support the results we obtained with our panel of MAbs, although the antibodies prepared in our laboratory appear to have a somewhat broader specificity. DeNardin et al. (E. DeNardin, C. Sprowl, R. T. Evans, M. Stinson, and R. J. Genco, Proc. Int. Conf. Cell. Mol. Clin. Aspects Streptococcus mutans, 1985, p. 469) screened two anti-Pl MAbs against several strains of mutans streptococci and found that reactivity was limited to strains of serotypes c and e. Thus, although the P1-like proteins of mutans streptococci crossreact between serotypes when polyclonal sera are used (13, 27), the isolation of an anti-Pl MAb which reacts equally well with strains of S. mutans, S. sobrinus, and S. cricetus has so far proven elusive. This suggests that the common determinants may not be strongly immunogenic or that they are quantitatively minor epitopes. Our immunoelectron microscopy studies (Fig. 2 to 4) confirm the surface localization of antigen P1 and demonstrate that this protein is apparently a constituent of the fuzzy coat surrounding the streptococcal cell. Moro and

Russell (22), using polyclonal monospecific anti-Pi rabbit sera and horseradish peroxidase-conjugated goat anti-rabbit IgG, also localized this protein to the cell surface. However, the resolution of the peroxidase technique would not allow one to visualize the fuzzy coat layer which is easily discernible in immunogold-labeled preparations (Fig. 4). These data represent the first demonstration of an association between this major surface protein and the fuzzy coat layer. Experiments are in progress in an attempt to remove this layer with proteolytic enzymes. Immunogold labeling studies of such denuded cells would allow us to ascertain whether any P1 is associated with the peptidoglycan layer itself or whether the protein is exclusively localized to the fuzzy coat. In support of the results of Smith et al. (30), we were unable to demonstrate any reactivity of our MAbs with human heart tissue. However, the very specificity of MAbs which makes them such useful reagents may render them unsuitable for the detection of tissue-cross-reactive epitopes, particularly if these are relatively minor or weakly immunogenic determinants. We plan to use our MAbs as tools for the purification of P1 by immunoaffinity chromatography. By immunizing rabbits with this purified immunogen and subsequently passing the resulting antisera through a column of immobilized P1, we hope to generate a polyclonal affinity-purified anti-Pl immunoglobulin preparation. Since this polyclonal reagent would contain antibodies binding to a large proportion, if not all, of the immunogenic determinants on the P1 molecule, detection of heart-crossreactive epitopes, if they exist, would be facilitated. Our battery of MAbs, as well as the polyclonal anti-Pl preparation and the purified P1, should also be useful probes in the study of potential functions for this major surface protein. It is clear from the work of Curtiss and his colleagues (4-6, 14) that the SpaA protein in S. sobrinus is a crucial factor in determining the cariogenic potential of this microorganism. Similar studies of the analogous polypeptide in S. mutans should provide valuable information in elucidating the events, at the molecular level, which result in the development of dental caries.

VOL. 55, 1987

0*

0. _ _ .~~~~~~~I 9

_~ B

tt

MAbs TO S. MUTANS SURFACE PROTEIN P1 .

2765

p

Kr

C.

S.D~~~~0

,

V&

S.

--------

X

,

_

00

.4'

S.

at, **.-

' ~ ~ ~ ~.~~~~~~~~~~~~~~ ~t

, , . . ..see,

4

D 06 sk '0I 0

I. a

.'It

40

4

A,

**E

FIG. 4. Immunoelectron microscopy of thin sections of S. mutans Ingbritt 162 (A and C) and Ingbritt 175 (B, D, and E) reacted with MAbs 1-6F (A and B) and 6-11A (C to E). Note the association of colloidal gold particles with the surface fuzzy coat layer of the Ingbritt 175 cells. The cell labeled with double arrows in panel D is depicted in greater detail in panel E to demonstrate this surface reactivity more clearly. Bars, 0.2 Lm (A to D) and 0.1 ,um (E).

2766

INFECT. IMMUN.

AYAKAWA ET AL.

Ud

Sr:,

FIG. 5. Electron microscopy of thin sections of Ingbritt 175 (A) and Ingbritt 162 (B). These cells were treated with 0.5% tannic acid (see text) to enhance the staining of the surface fuzzy-coat layer. Note the relative abundance of this material on the Pl-retaining Ingbritt 175 cells. Bar, 0.1 ,um. ACKNOWLEDGMENT This work was supported by Public Health Service grants DE-08007 and DE-06231 from the National Institute of Dental Research. LITERATURE CITED

1. Ayakawa, G. Y., J. L. Siegel, P. J. Crowley, and A. S. Bleiweis. 1985. Immunochemistry of the Streptococcus mutans BHT cell membrane: detection of determinants cross-reactive with human heart tissue. Infect. Immun. 48:280-286. 2. Beachey, E. H., G. L. Campbell, and I. Ofek. 1974. Peptic digestion of streptococcal M protein. II. Extraction of M antigen from group A streptococci with pepsin. Infect. Immun. 9:891-896. 3. Cunningham, M. W., and S. M. Russell. 1983. Study of heartreactive antibody in antisera and hybridoma culture fluids against group A streptococci. Infect. Immun. 42:531-538. 4. Curtiss, R., III. 1985. Genetic analysis of Streptococcus mutans virulence. Curr. Top. Microbiol. Immunol. 118:253-277. 5. Curtiss, R., III. 1986. Genetic analysis of Streptococcus mutans virulence and prospects for an anticaries vaccine. J. Dent. Res. 65:1034-1045. 6. Curtiss, R., III, S. A. Larrimore, R. G. Holt, J. F. Barrett, R. Barletta, H. H. Murchison, S. M. Michalek, and S. Saito. 1983. Analysis of Streptococcus mutans virulence attributes using recombinant DNA and immunological techniques, p. 95-104. In R. J. Doyle and J. E. Ciardi (ed.), Glucosyltransferases, glucans, sucrose and dental caries. IRL Press, Washington, D.C. 7. Douglas, C. W. I., and R. R. B. Russell. 1982. Effect of specific antisera on adherence properties of the oral bacterium Streptococcus mutans. Arch. Oral Biol. 27:1039-1045. 8. Douglas, C. W. I., and R. R. B. Russell. 1984. Effect of specific antisera upon Streptococcus mutans adherence to saliva-coated hydroxyapatite. FEMS Microbiol. Lett. 25:211-214. 9. Doyle, G., D. Everhart, C. Mallett, G. Ayakawa, and A. S. Bleiweis. 1986. Demonstration of shared antigenic determinants between Streptococcus mutans BHT cell membrane, human heart tissue and myosin using monoclonal antibodies to S. mutans. J. Gen. Microbiol. 132:2885-2892. 10. Fischetti, V. A., K. F. Jones, B. N. Mainjula, and J. R. Scott. 1984. Streptococcal M6 protein expressed in Escherichia coli. Localization, purification, and comparison with streptococcalderived M protein. J. Exp. Med. 159:1083-1095. 11. Forester, H., N. Hunter, and K. W. Knox. 1983. Characteristics of a high molecular weight extracellular protein of Streptococcus mutans. J. Gen. Microbiol. 129:2779-2788. 12. Forester, H., N. Hunter, J. A. Loudon, K. M. Weston, and

13.

14.

15. 16.

17. 18. 19.

20. 21.

22. 23. 24.

K. W. Knox. 1985. Protein P1 of Streptococcus mutans: an M-protein-like virulence factor of oral streptococci, p. 107. In Y. Kimura, S. Kotami, and Y. Shiokawa (ed.), Recent advances in streptococci and streptococcal diseases. Reedbooks, Ltd., Berkshire, England. Hardy, L. N., K. W. Knox, R. A. Brown, A. J. Wicken, and R. J. Fitzgerald. 1986. Comparison of extracellular protein profiles of seven serotypes of mutans streptococci grown under controlled conditions. J. Gen. Microbiol. 132:1389-1400. Holt, R. G., Y. Abiko, S. Saito, M. Smorawinska, J. B. Hansen, and R. Curtiss III. 1982. Streptococcus mutans genes that code for extracellular proteins in Escherichia coli K-12. Infect. Immun. 38:147-156. Horisberger, M., and J. Rosset. 1977. Colloidal gold, a useful marker for transmission and scanning electron microscopy. J. Histochem. Cytochem. 25:295-305. Hughes, M., S. M. MacHardy, A. J. Sheppard, and N. C. Woods. 1980. Evidence for an immunological relationship between Streptococcus mutans and human cardiac tissue. Infect. Immun. 27:576-588. Knox, K. W., L. K. Campbell, and D. Bratthall. 1983. Detection of antigens in enzymic lysates of cell wall from Streptococcus mutans strains. J. Dent. Res. 62:1033-1037. Lehner, T., M. W. Russell, and J. Caldwell. 1980. Immunisation with a purified protein from Streptococcus mutans against dental caries in rhesus monkeys. Lancet i:995-996. Lehner, T., M. W. Russell, J. Caldwell, and R. Smith. 1981. Immunization with purified protein antigens from Streptococcus mutans against dental caries in rhesus monkeys. Infect. Immun. 34:407-415. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. McBride, B. C., M. Song, B. Krasse, and J. Olsson. 1984. Biochemical and immunological differences between hydrophobic and hydrophilic strains of Streptococcus mutans. Infect. Immun. 44:68-75. Moro, I., and M. W. Russell. 1983. Ultrastructural localization of protein antigens I/II and III in Streptococcus mutans. Infect. Immun. 41:410-413. Nesbitt, W. E., R. H. Staat, B. Rosan, K. G. Taylor, and R. J. Doyle. 1980. Association of protein with the cell wall of Streptococcus mutans. Infect. Immun. 28:118-126. Oi, V. T., and L. A. Herzenberg. 1980. Immunoglobulinproducing hybrid cell lines, p. 351-365. In B. B. Mishell and S. M. Shiigi (ed.), Selected methods in cellular immunology. W. H. Freeman & Co., San Francisco.

MAbs TO S. MUTANS SURFACE PROTEIN P1

VOL. 55, 1987 25. Russell, M. W., L. A. Bergmeier, E. D. Zanders, and T. Lehner. 1980. Protein antigens of Streptococcus mutans: purification and properties of a double antigen and its protease-resistant component. Infect. Immun. 28:486-493. 26. Russell, R. R. B. 1979. Wall-associated protein antigens of Streptococcus mutans. J. Gen. Microbiol. 114:109-115. 27. Russell, R. R. B. 1980. Distribution of cross-reactive antigens A and B in Streptococcus mutans and other oral streptococci. J. Gen. Microbiol. 118:383-388. 28. Russell, R. R. B., D. Beighton, and B. Cohen. 1982. Immunisation of monkeys (Macaca fascicularis) with antigens purified from Streptococcus mutans. Br. Dent. J. 152:81-84. 29. Russell, R. R. B., S. L. Peach, G. Colman, and B. Cohen. 1983. Antibody responses to antigens of Streptococcus mutans in

30.

31.

32. 33.

2767

monkeys (Macaca fascicularis) immunized against dental caries. J. Gen. Microbiol. 129:865-875. Smith, R., T. Lehner, and P. C. L. Beverley. 1984. Characterization of monoclonal antibodies to Streptococcus mutans antigenic determinants I/II, I, II, and III and their serotype specificities. Infect. Immun. 46:168-175. Spurr, A. R. 1969. A low viscosity resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26:31-43. Terleckyj, B., N. P. Willett, and G. D. Shockman. 1975. Growth of several cariogenic strains of oral streptococci in a chemically defined medium. Infect. Immun. 11:649-655. Zanders, E. D., and T. Lehner. 1981. Separation and characterization of a protein antigen from cells of Streptococcus mutans. J. Gen. Microbiol. 122:217-225.