JOURNAL OF CLINICAL MICROBIOLOGY, Dec. 1987, p. 2285-2288 Vol. 25, No. 12 0095-1137/87/122285-04$02.00/0 Copyright © 1987, American Society for Microbiology Identification of sobrinus with Monoclonal Antibodies

JOHANNESJ. DE SOET,l* PHILIPPUS J. VAN DALEN,' BEN J. APPELMELK,2 AND JOHANNES DE GRAAFF' Department of Oral Microbiology, ACTA,' and Department of Medical Microbiology, Research Group for Commensal Infections,2 Vrije Universiteit, 1081 BT Amsterdam, The Netherlands Received 17 April 1987/Accepted 13 August 1987

Identification of Streptococcus sobrinus is often difficult to perform because of the great resemblance of the organism to other oral streptococcal . Therefore, monoclonal antibodies were prepared which were shown to be highly specific for S. sobrinus. Cross-reactivity with other oral microorganisms has not been observed in an -linked immunosorbent assay and an immunofluorescence assay. These monoclonal antibodies belonged to the subclass immunoglobulin G2b. To be certain that the strains used in cross-reactivity tests were S. sobrinus, their DNA base composition was measured as a golden standard. Additional tests like colony morphology and sugar fermentation with the API 20 Strep system (Analytab Products, Montalieu- Vercieu, France) were performed. These additional tests turned out to be necessary because 100% correct identification could not be obtained by separate tests. Immunological characterization with the clones OMVU10 and OMVU11 proved to be discriminative between S. sobrinus and other streptococcal species.

Streptococcus sobrinus, previously called Streptococcus immunization schedule consisted of four injections of 0.1 ml mutans serotype d/g, is known to be cariogenic in animals of bacterial suspension given at 2-week intervals. The first (18). It has been found in higher numbers in injection was given intravenously, the second intraperitone- from humans with active caries than from caries-free indi- ally with complete Freund adjuvant, and the third intraperi- viduals (11-13). These results suggest that S. sobrinus may toneally with incomplete Freund adjuvant. The last injection be more cariogenic than other members of the S. mutans (intravenous) was given 4 days before fusion. group. Hybridomas were made by fusing the splenocytes with It is evident that for an evaluation of the role of S. sobrinus P3-X63-Ag8.653 myeloma cells at a 1:5 ratio. The fusion in dental caries, a reliable method for identification and procedure was that described by De Haan et al. (7). Hybrids quantification is required (26). Mitis-salivarius bacitracin were screened for the production of immunoglobulin by agar (MSB) is often used for the isolation of streptococci of testing the hybridoma culture supernatant (HCS) in an the S. mutans group. However, this selective medium is not enzyme-linked immunosorbent assay (ELISA). suitable for direct identification of S. sobrinus and, more- ELISA. Bacterial cell suspension (75 pil) was incubated at over, strongly inhibits its growth (8, 14, 15, 21, 25, 27, 29). 4°C for 16 h in 96-well microdilution plates (655001; Greiner). Therefore, S. sobrinus may be easily missed when this Free binding places were blocked by incubation with 1% medium is used. A recently developed selective Trypticase- bovine serum albumin and 0.05% (wt/vol) Tween 80 in PBS yeast-cystine--bacitracin agar (TYCSB) has been re- at room temperature for 1 h. Between every incubation, the ported to be more suitable for the isolation and identification plates were washed three times with PBS supplemented with of S. sobrinus colonies which exhibit a "halo" on it (21, 27). 0.05% Tween 80 (PBST). HCS (75 ,ul) diluted in PBST was On the other hand, antisera have proved to be useful for the identification of pathogenic in dental plaque samples added. The plates were incubated for 2 h at room tempera- (2, 9, 19, 20). Monoclonal antibodies (MAbs) have also been ture. After washing, 75 ,ul of peroxidase-conjugated goat used for this purpose (3, 23). anti-mouse immunoglobulin was added (1:1,000; affinity- The aims of this study were (i) to produce MAbs against S. purified; Bio Makor, Israel) and incubated for 1 h at room sobrinus that were highly specific and therefore suitable for temperature. Peroxidase activity was measured by adding 75 identification purposes, and (ii) to compare the results with ,ul of o-phenylenediamine (8 mg) with 8 ,ul of H202 in 20 ml commonly used identification methods, such as sugar fer- of citrate buffer (0.01 M, pH 9.5). The reaction was stopped mentation, colony morphology on selective media, and with 50 p.l of H2SO4 (10%) after 15 min. The OD of each well guanine plus cytosine content determination of the DNA. was measured with a Micro plate reader (MR 600; Dynatech Industries, Inc., Torrance, Calif.) at 490 nm. MAbs against Escherichia coli lipopolysaccharide served as a negative MATERIALS AND METHODS control (B. J. Appelmelk, A. M. J. J. Verweij-van Vught, Bacterial strains. The microorganisms used in this study J. J. Maaskant, W. F. Schouten, A. J. R. De Jonge, L. G. are listed in Table 1. Bacterial cell suspensions were pre- Thijs, and D. M. Maclaren, J. Med. Microbiol., in press). pared by growing the cells anaerobically (85% N2, 5% C02, The titer of HCS was defined as the dilution by which the and 10% H2) overnight in Todd-Hewitt broth at 37°C. The OD of the ELISA was two times the OD of the negative cells were centrifuged, washed, and diluted in phosphate- control. Clones which were specific for S. sobrinus were buffered saline (PBS) to an optical density (OD) at 650 nm of detected by their positive reactions in ELISA with S. 1.0. sobrinus and by their negative reactions with other strepto- Hybridoma production. The bacterial cell suspension of cocci. strain HG459 was used to immunize BALB/c mice. The Screening of the microorganisms was performed by ELISA. A positive ELISA was defined as an OD of 0.5 or * Corresponding author. greater at an HCS dilution of 1:1,000. 2285 2286 DE SOET ET AL. J. CLIN. MICROBIOL.

TABLE 1. Microorganisms used for cross-reactivity studies and their determination Acid production from": Species and strain (origin') G + C%b ELISA' Halo" Fluorescent cells Inulin Raffinose Streptococcus sobrinus HG456 43.5 +l- v + HG459 6715DP (6) 44.3 + + v + HG594 (7) 45.0 + v + HG596 OMZ176 (4) 44.4 +l- + HG598 50B4 (4) 44.7 + + HG599 OMZ65 (4) 44.4 + +l1- V v + HG688 OMZ65 (3) 43.9 + HG689 B276 (3) 44.6 + HG690 B13 (3) 43.7 + +l- + HG702 (7) 44.2 + + HG707 ATCC 33478 44.3 + HG713 OlHl (1) 43.7 + HG716 5OB4SS (1) 44.9 + HG718 50B4 (1) 44.1 + HG732 (7) 44.2 + HG738 OMZ176 (2) 44.2 + HG739 KIR (2) 43.4 + HG743 6715 (2) 43.0 + HG724 SE1l (1) ND + HG210 4540 (5) 38.8 + HG221 999 (5) 38.0 + HG222 1000 (5) 38.2 + + HG236 (7) ND + HG237 (7) 38.3 + HG240 (7) 38.4 + HG801 JC2 (4) 37.0 HG712 16Ba (1) ND + HG714 LM7 (1) 38.0 + HG730 AT10 (1) ND + Streptococcus cricetus + HG48 E-49 N 42.0 ND v +D HG709 370 (1) ND ND + HG773 E-49 (2) ND ND V HG744 HS6 (2) 42.1 ND ND ND Streptococcus rattus HG725 Z-BHT (1) ND HG726 JH145 (1) ND ND ND HG751 FAl (3) ND + + HG745 BHT (3) 41.7 Various microorganisms ND ND ND a Origins were as follow: 1, C. Van Looveren, The Netherlands; 2, A. L. Coykendall, United States; 3, R. Whiley, United Kingdom; 4, W. H. Van Palenstein Helderman, The Netherlands; 5, J. S. Van Der Hoeven, The Netherlands; 6, S Hamada, Japan; 7, our isolate. b Guanine and cytosine content of DNA. ND, Not determined. C An OD of 0.5 or higher in an ELISA with clone OMVU10 (diluted 1:1,000). d Existence of a halo around the colony on TYCSB and TSY20B agar plates. When differences between the two media were observed, both results are given as TYCSB/TSY20B. +, a halo of at least 1 mm was observed around the colony; -, no halo was observed; ND, not determined. e Acid production from inulin and raffinose by the API 20 Strep system. V, Variable; ND, not determined. f Fluorescent cells by immunofluorescence technique. " The various oral microorganisms tested were S. ferus 8S1 (3) and HD3 (2); S. macacae H1569 (3); Streptococcus sanguis I and II HG168, HG169, HG310, HG311, HG533, HG536, HG857, and HG865 (7); HG170, HG178, and HG467 (7); Streptococcus milleri HG293, HG294, and HG587 (7); Streptococcus sdlivarius HG668 (7); HG228 and HG336 (7); Streptococcus morbillorum HG858 and HG874 (7); Actinomyces viscosus HG484 (NY334; 5); case HG498 (7); Bacteroides gingivalis HG762 (7); Bacteroides melaninogenicus HiG753 (7); Bacteroides intermed us HG763 (7); Capnocytophaga ochracea HG347 (7); and Candida albicans HG428 (7). These strains were tested for reactivity on ELISA and immunofluorescence only.

Immunofluorescence assay. The immunofluorescence as- slides were incubated for 1 h at room temperature with 40 ,ul say used in this study was performed with pure and mixed of an antibody dilution (clone OMVUiO; 1:500 in PBST). bacterial cultures. The method used was a modification of After the slides were washed, 40 ,uI of fluorescein isothiocy- that described by Mouton et al. (19). Briefly, bacterial anate-conjugated, affinity-purified goat anti-mouse immuno- suspensions were diluted to an OD at 650 nm of 0.1. The globulin G (IgG; Bio Makor) was added and incubated for 1 suspensions wère placed on objective slides (XBM-302- h at room temperature. The slides were examined with a 89912; Bio Merieux) and dried at 70'C for 1 h. Then the fluorescence microscope at 490 nm. VOL. 25, 1987 IDENTIFICATION OF S. SOBRINUS 2287

Guanine plus cytosine content determination. To identify OD strains as S. sobrinus, the guanine plus cytosine content of their DNA was examined. This method was used, together 1.5 - with determination of sugar fermentation and colony mor- phology, to obtain a reliable identification. DNA was isolated as described by Meyer and Schleifer (17). After the cells were harvested they were incubated with 1.0 - lysozyme. Lysis was achieved by shaking the cells with glass beads (diameter, 0.13 mm) for 15 min. After extractions with phenol and chloroform, DNA was precipitated by adding cethyltrimethylammonium bromide and isopropanol. The purity and concentration was measured by determination of 0.5 the OD at 260 and 280 nm. The DNA was considered to be pure when a ratio of the OD at 260 nm/OD at 280 nm of 1.8 4- -- or higher was obtained. Buoyant density centrifugation was performed essentially as described by Mandel et al. (16) and ! ~~~~~~~~~~~2 3 41 Van Steenbergen et al. (28). 103 dilution Other identification methods. Determination of acid pro- FIG. 1. Titration of HCS of clone OMVU10 against bacterial duction from sugars was performed with the API 20 Strep suspensions of S. sobrinus HG459 (0) and S. mutans HG714 (*) by system (Analytab Products, Montalieu-Vercieu, France), ELISA (OD at 490 nm against antibody dilution). according to the recommendation of the manufacturer. The identification of the strains was based on the criteria de- scribed by Beighton et al. (1). DISCUSSION Determination of colony morphology was performed on two selective media, TYCSB and TSY20B. On these media S. mutans was previously considered to be one species. the presence of a halo has been reported to be characteristic Later, the species proved to be heterogeneous and to consist for S. sobrinus (21, 27). of at least eight serotypes (2, 20). On the basis of DNA homology studies, however, the S. mutans group was di- vided into six different species (1, 4-6, 22). Because three of RESULTS these species, S. mutans, Streptococcus ferus, and Strepto- coccus macacae, have serotype c antigens, classification of Isolation of MAbs. MAbs were isolated by limiting dilu- serotypes is insufficient. Routine identification methods with tion. About 250 wells which contained hybridomas were selective media and sugar fermentation are, however, insuf- screened for antibody activity against strain HG459. Two ficient to distinguish among the different species (20, 29). wells turned out to be positive in an ELISA. After subclon- The aim of the present study was to develop a method to ing, two clones producing MAbs were isolated: OMVU1O identify S. sobrinus strains from dental plaque by nonselec- and OMVU11. The subclass for both clones was IgG2b, as tive cultural procedures. In this study we have reported the determined in an ELISA with subclass-specific antisera. For production of two MAbs, clones OMVU10 and OMVU11, identification of oral bacteria, the HCS method was used. which can be used for the determination of S. sobrinus. The HCS titer was 1:3,000 for all S. sobrinus strains tested Isolation of S. sobrinus can be achieved by using MSB or (Fig. 1). Cross-reactivity with strains other than S. sobrinus TYCSB agar. For accurate bacterial counts, however, these was not observed (Table 1). media are not trustworthy, because the recovery of S. To verify the accuracy of previous identification methods, sobrinus on MSB agar has been reported to be poor (21, 29), the DNA base composition was determined (Table 1). All and some S. sobrinus strains do not grow on TYCSB and strains which fitted within a G+C range of 43.5 to 46% TSY20B (Table 1). Moreover, the existence of a halo around proved to be positive by ELISA for our MAbs. S. sobrinus colonies is not an invariable characteristic of S. The results of the characterization by sugar fermentation sobrinus, because we have shown that some strains did not with the API 20 Strep system are listed in Table 1. Because produce a halo, while some produced haloes which were so acid production from inulin and raffinose was variable both small that they might easily have been missed. for some S. sobrinus and for some S. mutans strains, S. The most reliable method for identification of streptococci sobrinus strains could not always be distinguished from S. is the study of DNA homologies, but this technique is too mutans strains by sugar fermentation alone. cumbersome for routine use (6). In combination with sugar Discrimination between S. sobrinus and Streptococcus fermentation reactions, S. sobrinus can be distinguished rattus was possible by means of the API 20 Strep system. All from S. mutans by means of the DNA base composition S. rattus strains were positive for arginine hydrolysis, while analyses. the other strains tested were not. The results of colony Identification with the API 20 Strep system was based on morphology on TYCSB and TSY20B plates are listed in the criteria of Perch et al. (20), Facklam (10), and Beighton Table 1. Strains HG456, HG596, and HG690 did not present et al. (1). S. mutans should produce acid from inulin and a halo on TSY20B. They did, however, present a halo on raffinose, while S. sobrinus should not. On this basis, 10% of TYCSB, but only after 4 days of incubation. Strains HG688, our S. mutans strains and 15% of our S. sobrinus strains HG689, and HG707 did not present a halo on either medium, would have been incorrectly identified, and 20% of both S. while strain HG716 did not grow on either medium. In mutans and S. sobrinus strains would not have been identi- double-blind experiments two observers could not reliably fied because they failed to produce acid from one of the two identify S. sobrinus colonies by detection of a halo, because sugars. the haloes were too small (results not shown). We conclude These problems with the identification of S. sobrinus that this method is unsuitable for unknown samples. strains can be avoided by the use of the MAbs described in 2288 DE SOET ET AL. J. CLIN. MICROBIOL. this report. The MAbs OMVU10 and OMVU11 were raised man, P. Sampaio Camargo, and D. Backer. 1978. Antibodies against S. sobrinus serotype g, but cross-reactivity with against Streptococcus mutans and glucosyltransferases in car- strains of S. sobrinus serotype d was observed (Table 1). ies-free and caries-active military recruits. Adv. Exp. Med. This is in accordance with results of other studies (24), in Biol. 107:369-383. which it has been shown 14. Jordan, H. V. 1986. Cultural methods for the identification and that serotypes d and g are cross- quantitation of Streptococcus mutans and Lactobacilli in oral reactive (24). Even strain HG707, which is identical to strain samples. Oral Microbiol. Immunol. 1:23-27. SL-1 and which has been reported to be nonserotypable (5), 15. Little, W. A., D. C. Korts, L. A. Thomson, and W. H. Bowen. reacted positively in our ELISA with clones OMVU10 and 1977. Comparative recovery of Streptococcus mutans on ten OMVU11. Cross-reactivity with other oral species could not isolation media. J. Clin. Microbiol. 5:578-583. be detected by ELISA and an immunofluorescence assay. 16. Mandel, M., C. L. Schildkraut, and J. Marmur. 1968. Use of The use of the clones OMVU10 and OMVU11 in an CsCl density gradient analysis to determine the guanine plus ELISA or in an immunofluorescence assay has, to date, cytosine content of DNA. Methods Enzymol. 12:184-195. allowed us to identify all S. sobrinus strains within 2 days. 17. Meyer, S. A., and K. H. Schleifer. 1975. Rapid procedure for the Therefore, identification methods based on immunological approximate determination of the deoxyribonucleic acid base composition of micrococci, staphylococci, and other bacteria. staining with clone OMVU10 and OMVU11 enable quick Int. J. Syst. Bacteriol. 25:383-385. and reliable identification of S. sobrinus. At present we are 18. Michalek, S. M., J. R. McGhee, and J. M. Navia. 1975. Viru- developing those methods to make large-scale epidemiolog- lence of S. mutans. A sensitive method for evaluating ical studies possible. cariogenicity in young gnotobiotic . Infect. Immun. 12:69- 75. LITERATURE CITED 19. Mouton, C., P. G. Hammond, J. Slots, M. J. Reed, and R. J. 1. Beighton, D., H. Hayday, R. R. B. Russell, and R. A. Whiley. Genco. 1981. Identification of Bacteroides gingivalis by fluores- 1984. Streptococcus macacae sp. nov. from dental plaque of cent antibody staining. Ann. Microbiol. (Paris) 132B:69-83. monkeys (Macacafascicularis). Int. J. Syst. Bacteriol. 34:332- 20. Perch, B., E. Kjems, and T. Ravn. 1974. Biochemical and 335. serological properties of Streptococcus mutans from various 2. Bratthall, D., and B. Kohler. 1977. Streptococcus mutans human and animal sources. Acta Pathol. Microbiol. Scand. serotypes: some aspects of their identification, antigen shift and Sect. B 82:357-370. relationship to caries. J. Dent. Res. 55:C15-C21. 21. Schaeken, M. J. M., J. S. V. D. Hoeven, and H. C. M. Franken. 3. Chen, P., V. Bochacki, H. S. Reynolds, J. Beanan, D. N. Tatakis, 1986. Comparative recovery of Streptococcus mutans on five J. J. Zambon, and R. J. Genco. 1986. The use of monoclonal isolation media, including a new simple medium. J. Dent. Res. antibodies to detect Bacteroides gingivalis in biological sam- 65:906-908. ples. Infect. Immun. 54:798-803. 22. Schleifer, K. M., R. Kilppert-balz, J. Kraus, and F. Gehring. 4. Coykendall, A. L. 1974. Four types of Streptococcus mutans 1984. Relatedness and classification of Streptococcus mutans based on their genetic, antigenic and biochemical characteris- and mutans-like streptococci. J. Dent. Res. 63:1047-1050. tics. J. Gen. Microbiol. 83:327-338. 23. Simonson, L. G., B. R. Merrell, R. F. Rouse, and 1. L. Shklair. 5. Coykendall, A. L. 1977. Proposal to elevate the subspecies of 1986. Production and characterization of monoclonal antibodies Streptococcus mutans to species status, based on their molec- to Bacteroides gingivalis. J. Dent. Res. 65:95-97. ular composition. Int. J. Syst. Bacteriol. 27:26-30. 24. Smith, R., and T. Lehner. 1986. Preparation and characteriza- 6. Coykendall, A. L., O. P. Daily, M. J. Kramer, and M. E. Beath. tion of monoclonal antibodies to antigenic determinants of 1971. DNA-DNA hybridization studies of Streptococcus Streptococcus mutans and sobrinus, p. 357-365. In T. Lehner mutans. J. Dent. Res. 50(Suppl.):1131-1139. and G. Cimasoni (ed.), The borderland between caries and 7. De Haan, P., A. J. R. De Jonge, T. Verbrugge, and D. M. periodontal disease III. Editions Medicine et Hygiène, Geneva. Boorsma. 1985. Epitope-specific monoclonal antibodies against 25. Tanzer, J. M., A.-C. Borjesson, L. Laskowski, A. B. Kurasz, and the hapten penicillin. Int. Arch. Allerg. Appl. Immunol. 76:42- M. Testa. 1984. -sucrose-potassium tellurite-bacitracin 46. agar, an alternative to mitis salivarius-bacitracin agar for enu- 8. Emilson, C. G., and D. Bratthall. 1976. Growth of Streptococcus meration of Streptococcus mutans. J. Clin. Microbiol. 20:653- mutans on various selective media. J. Clin. Microbiol. 4:95-98. 659. 9. Emilson, C. G., B. Kohler, and D. Bratthall. 1974. Immune 26. Tanzer, J. M., and J. Clive. 1986. Quantitative considerations in fluorescence determination of the relative proportions of Strep- microbiological evaluations for caries: risk for type Il errors tococcus mutans in human dental plaque: a comparison with resulting from use of MSB agar. Oral Microbiol. Immunol. cultural techniques. Arch. Oral Biol. 20:81-86. 1:28-30. 10. Facklam, R. R. 1974. Characteristics of Streptococcus mutans 27. Van Palenstein Helderman, W. H., M. Ijseldijk, and J. H. J. A. isolated from human dental plaque and blood. Int. J. Syst. Huis in't Veld. 1983. A selective medium for the two major Bacteriol. 24:313-319. subgroups of the bacterium Streptococcus mutans, isolated 11. Holbrook, W. P., and D. Beighton. 1986. Streptococcus mutans from human dental plaque and saliva. Arch. Oral Biol. 28:599- levels in saliva and distribution of serotypes among 9-year old 603. Icelandic children. Scand. J. Dent. Res. 95:37-42. 28. Van Steenbergen, T. J. M., J. J. De Soet, and J. De Graaff. 1979. 12. Hoover, C. I., E. Newbrun, G. Mettraux, and H. Graf. 1980. DNA base composition of various strains of Bacteroides Microflora and chemical composition of dental plaque from melaninogenicus. FEMS Microbiol. Lett. 5:127-130. subjects with hereditary fructose intolerance. Infect. Immun. 29. Wade, W. G., M. J. Aldred, and W. M. Walker. 1986. An 28:853-859. improved medium for isolation of Streptococcus mutans. J. 13. Huis in 't Veld, J. H. J., D. Bannet, W. H. van Palenstein Helder- Med. Microbiol. 22:319-323.