- Email: [email protected]
The adsorption of human salivary components to strains of the bacterium Streptococcus mutans
0003-9969/8483.00+ 0.00 Copyright 0 1984Pergamon Press Ltd
Archs oral Bid. Vol. 29, No. 10, pp. 751-757, 1984 Printed in Great Britain. ,411rights reserved
THE ADSORPTION OF HUMAN SALIVARY COMPONENTS TO STRAINS OF THE BACTERIUM STREPTOCOCCUS
MUTANS
C. W. I. DOUGLAS* and R. R. B. RUSSELL Royal College of Surgeons of England, Dental Research Unit, Downe, Orpington, Kent, England, U.K. Summary-U:sing strains of Streptococcus mutans suspended in human saliva, the salivary proteins capable of binding to the surface of the bacteria were identified by immunological and electrophoretic techniques. Six binding components were recognized: IgA, lysozyme, some high molecular weight material (> 400,000), probably a glycoprotein, a low molecular weight component (1 l-l 3,000) a 150,000 mol. wt protein, and one major component, mol. wt 2&25,000 which did not resolve fully on SDS-polyacrylamide electrophoresis. All these salivary components could be desorbed from the bacteria with 1 M NaCl, and subsequent extraction of the same cells with 6M guanidineHC1 did not release any more salivary material. The significance of the binding of these salivary components is unknown but some may modify the behaviour of the organisms in t&o.
INTRODUCTION
The oral bacterium Streptococcus mutans is thought to be a major aetiological agent in dental caries (Hamada and Sladie, 1980). Considerable attention has therefore been paid to factors affecting the distribution of Strep. mutans in the mouth and particularly to the mechanisms whereby the organism adheres to and colonizes the tooth surface. Teeth in vivo are covered with an integument called the acquired pellicle, which is largely composed of salivary proteins and glycoproteins (Lenz and Miihlemann, 1963; Mayhall, 1970; Mrstavik and Kraus, 1973); it is now recognized that the initial deposition of bacteria on to the tooth must involve some interaction between bacterial surface components and salivary constituents of the acquired pellicle (Orstavik, 1980). Oral bacteria bind a variety of salivary components including immunoglobulins (Brandtzaeg, Fjellanger and Gjeruldsen, 1968; Sirisinha, 1970), salivaryaggregating factors (Hay, Gibbons and Spinell, 1971; Kashket and Donaldson, 1972; Ericson, Pruitt and Wedel, 1975; Mrstavik, 1978; Eggert, 1980), lysozyme (Pollock et al., 1976; IZlrstavik, 1978), blood groupreactive substances (Williams and Gibbons, 1975; Rijlla and Kilian, 1!)76; Gibbons and Qureshi, 1978) aggregated /I-2-microglobulin (Ericson et al., 1979) and a-amylase (Douglas, 1983). Some of these are present in the acquired pellicle, and so it is possible that such molecules could act as receptor sites on the tooth surface to which bacteria could bind. It is also possible that binding of salivary molecules to bacteria could have the opposite effect of reducing attachment to teeth, either by aggregating the bacteria and so enhancing their susceptibility to clearance by normal salivary flow and swallowing or by altering the properties of their surface (Magnusson and Ericson, 1976; Mrstavik, 1978).
*Present address: Department of Oral Pathology, School of Clinical Dentistry, University of Sheffield, Sheffield SIO 2TA, England, U.K. 751
Our aim was to define the components of saliva which adsorb to the surface of Strep. mutans. A preliminary report of some of the information has been presented elsewhere (Douglas and Russell, 1982). MATERIALS AND METHODS Bacteria and culture conditions
Nine strains of Strep. mutans were used, representing a range of serotypes as follows: AHT (a), FAl (b), 3209 (c), Ingbritt (c), 1649 (c), B13 (d), P4 (e), 85 (e), and Kl (g). Strains 85 and 1649 were isolated in these laboratories from dental plaque of macaque monkey (Macaca fascicularis) and strain 3209 was obtained from Dr J. A. Cole, Department of Biochemistry, University of Birmingham, U.K. The others were from our laboratory stock-collection. All strains were maintained on Todd-Hewitt agar at 4°C and when required were subcultured directly into a semidefined medium (Cas MM) described by Russell (1979) but with 0.5 per cent (w/v) casein hydrolysate (Oxoid) instead of individual amino acids. Saliva
Whole saliva from several people was stimulated by chewing Parafilm (American Can Co., Greenwich, U.S.A.) and immediately cleared by centrifugation at 14,OOOg in a Beckman model 52-21 centrifuge for 10min at 8°C. Samples were stored at 4°C for a maximum of 8 h. A large proportion of the work was carried out using saliva from one of us (C.W.1.D) but the phenomena described here applied to all saliva samples tested. of salivary proteins to Strep. mutans Each of the Strep. mutans strains was cultured overnight in 20ml Cas MM, harvested by centrifugation and washed twice in l/lOth strength Dulbecco phosphate-buffered saline (PBS, Oxoid). The cells were then mixed with 10 ml of saliva for 30 min on
Aakorption
C. W. I. DOUGLAS and R. R. B. RUSSELL
152
a spiromixer (Denley Ltd, Billingshurst, Sussex, U.K.), washed a further three times in l/10 PBS and extracted by agitation in 1 ml 1 M NaCl for 30min. After removing cells by centrifugation, the supernatants were dialysed (610,000 mol. wt cut-off) overnight against distilled water. The cell pellet was then resuspended in 1 ml 6 M guanidine-HCl (pH 3.7) and further extracted by agitation for 30 min to solubilize material that may be tightly bound. The supernatant extract was also dialysed against distilled water. To obtain larger quantities of extract, cells of Strep. mutans strain 3209 collected from 400 ml Cas MM 24-h cultures were used and treated in a manner similar to that just described, with the exception that 25 ml of saliva was used to resuspend the washed cells and these were subsequently extracted with 5 ml 1 M NaCl. The 6M guanidine step was omitted. All dialysed extracts were concentrated by lyophilization. SDS-polyacrylamide PAGE)
gel
electrophoresis
(SDS-
Electrophoresis in the presence of SDS was performed as described by Russell (1979) using 12 per cent (w/v) acrylamide gels. Gels were fixed and stained in one step by immersion of the gel in 0.2 per cent (w/v) Coomassie brilliant blue [Page Blue 83 (BDS)] in acetic acid-ethanol-water (10:45:45 by vol.). De-staining was effected in a similar solvent mixture using methanol instead of ethanol. Proteins for molecular weight standards were obtained from Sigma. Gels were also stained for glycoprotein using the periodic acid-Schiff method described by Segrest and Jackson (1972). Western blotting
Electrophoretic transfer of proteins from SDS-polyacrylamide gels (i.e. western-blotting) was performed essentially according to the method of Burnette (198 1) but with the following modifications; (1) a trans-blot apparatus (Bio-Rad Laboratories Ltd, Watford, England) was used; (2) 5 per cent (w/v) haemoglobin solution in tris-saline (10 mM tris-HCl, 0.15 M NaCl, pH 7.4) was used for blocking freeprotein binding sites on the nitrocellulose and for incubations in antiserum and enzyme-antibody conjugates; (3) antigen bands on the nitrocellulose were detected by anti-immunoglobulin peroxidase conjugate (Dako and Sigma) at a dilution of l/200 for 2 h at room temperature. Staining was with 1 ml 0.6 per cent (w/v) cl-chloro-1-naphthol (Sigma) in methanol, 9 ml 50 mM tris pH 7.4 and 10 ~1 hydrogen peroxide (30 vol). Bands appeared dark grey on a white background.
Strep. mutans antigens by CIE, immunodiffusion and western-blotting. None was found. Goat antiserum to human salivary proline-rich protein A (PRP-A) was a kind gift from Dr A. Bennick, Department of Biochemistry, University of Toronto, Canada. Goat anti-IgA serum was obtained from Meloy Laboratories, Springfield, U.S.A. and rabbit anti-human albumin serum from Dako. Crossed-immunoelectrophoresis
(CIE)
The method employed was similar to that described by Eckersall and Beeley (1980). In some experiments, antigen samples were labelled with 12?odine using iodobeads (Pierce Chemical Co., Rockford, U.S.A.) before running in the CIE system as follows: 5 ~1 samples of radiolabelled material were mixed with 10 pl unlabelled saliva and, after electrophoresis and staining, non-overlapping regions of precipitation peaks were carefully cut out using the method of Norrild, Bjerrum and Vestergaard (1977). Each sample was then boiled in SDS sample buffer and examined by SDS-PAGE. Gels were dried on Whatman 3MM filter paper using a Bio-Rad Model 224 drier (Bio-Rad). Autoradiography
Radioactively-labelled protein bands on dried SDS-PAGE gels and peaks in crossed-immunoelectrophoretograms were detected by exposure to X-omat AR film (Kodak) in conjunction with a Cronex intensifying screen (Du Pont de Numours & Co., Wilmington, U.S.A.) at -20°C (Swanstrom and Shank, 1978). Lysozyme
Lysozyme activity was detected in agarose gels by the method of Osserman and Lawlor (1966) but with the modification that 7.5 x 5cm glass microscope slides were covered with 8 ml of 1 per cent agarose (Sigma) containing 4 mg of a freeze-dried preparation of Micrococcus lysodeiktcus (Sigma) in 0.05 M phosphate buffer pH 6.5.
RESULTS
Antisera
of saliva with Strep. mutans Approximately 20 distinct protein bands in the native saliva samples could be resolved using SDS-PAGE. Adsorption of saliva with an excess of Strep. mutans (3209) cells markedly reduced the intensity of one band on SDS-PAGE which resolved as a zone rather than a sharp band having a mol. wt of 2&25,000 (see Fig. 1). This material was present in all the saliva samples tested and was absorbable by Strep. mutans cells in all cases. No other changes in protein band pattern could be seen after the adsorption of saliva with Strep. mutans.
Anti-saliva serum was raised in rabbits by three intramuscular injections at 21-day intervals of 0.9 ml cleared pooled saliva + 0.1 ml Alu Gel S (Serva) as adjuvant. The animals were bled 7 days after the third injection and the IgG fraction of the serum prepared by precipitation with caprylic acid according to the method of Steinbuch and Audran (1969). The resulting antibody was tested for reactivity with
to Strep. mutans To investigate binding to Strep. mutans, bacteria that had been mixed with saliva were extracted with 1 M NaCl and 6 M guanidine-HCl. Both extraction procedures solubilized Strep. mutans proteins and it was found that extracts yielded a large number of bands visible on SDS-PAGE. To identify the pro-
Adsorption
Binding of salivary proteins
Binding of salivary components to Strep. mutans
--
-
--
-
iiiiiiim ----- -- -
&K._
== ---
-
--
Fig. 1. Diagram of a 12 per cent SDS-polyacrylamide gel stained with Coomassie brilliant blue showing the proteins of (A) whole saliva and (B) saliva that has been absorbed with Strep. muians (3209) cells.
153
Western-blots of extracts taken from a range of Strep. mutans strains that had been treated with saliva showed that the major component binding to all of them was the 20-25,000 mol. wt material. There was some variability in the binding of some of the other components. The strains of serotypes c and e (85, 1649, 3209, Ingbritt and P4) appeared to bind more IgA than the other strains, as detected by intensity of the band on nitrocellulose. Low levels were detected in extracts of strain FAl, whereas IgA was not found in extracts of strains B13, AHT and Kl. The high mol. wt material at the top of the stacking section of the gel and the 150,000 mol. wt band were present in all extracts. Only the extracts of strain FAl consistently contained the 13,000 mol. wt protein, but this or another component of similar molecular weight was often seen in extracts of serotype c strains. Crossed-immunoelectrophoresis. The antiserum to saliva precipitated 10 distinct antigens under standard CIE conditions (Fig. 3). Analysis by CIE of saliva that had been adsorbed with Strep. mutans showed that peaks 3, 4, 5 and 6 were considerably lower and the area under each peak was reduced by approx. 35, 17, 5 and 14 per cent respectively. Extracts of Strep. mutans strain 3209 that had been coated with saliva contained two antigen peaks, corresponding to the positions of 3 and 4, that were visible by staining with Coomassie blue. However, when the sensitivity of antigen detection was increased by labelling the extracts with lz5iodine, there
teins of salivary origin in this mixture, the following techniques were used: Western-blotting. Protein bands on SDS-PAGE
were transferred to nitrocellulose and the salivary components present identified by probing with antisaliva serum. A maximum of five bands were detected in NaCl extracts (Fig. 2) and no more salivary proteins could be seen in 6 M guanidine-HCl extracts. Two of the bands, the low mol. wt (13,000) component and the high mol. wt material in the stacking gel, were not always observed. The major staining component resolved as a zone with mol. wt between 20 and 25,000, which corresponded to the band described above that was adsorbable by treatment of saliva with. Strep. mutuns. cells. One of the human salivary acidic proline-rich proteins (A) isolated by Bennick and Connell (1971) migrates anomalously in the presence of SDS and although described as having a mol. wt of 11,145 behaves as if its mol. wt were 24,000 (Rajan and Bennick, 1983). Probing western-blots of the salivary binding components with anti-PRP-A serum failed to identify the 20-25,000 mol. wt region as the prolinerich protein. The salivary protein band on western-blots having a mol. wt of 59,000 was shown to be IgA heavy chain by specific reaction with anti-IgA serum. The two other salivary components binding to Strep. mutans cells were a protein with mol. wt approx. 150,000 and some high mol. WI: material that did not enter the stacking section of the gel and therefore could not be resolved by the SDS-PAGE system we used.
UK,-
Fig. 2. Diagram of all the salivary proteins that bind to separated by SDS-PAGE and identified by probing western-blots of the gel with antisaliva serum. Track A shows the salivary proteins present in 1 M NaCl extracts of Strep. murans cells that had been coated with saliva and B, bands detectable in whole saliva.
Strep. mutans strains
154
C. W. I. DOUGLAS and R. R. B. RUSSELL
salivary protein binding to Strep. mutans (3209) was approx. 11,000 and appeared to be smaller than the band detected by western-blotting (i3,OOO). It is not certain whether these components are identical; they may have migrated differently in the two systems employed. Lysozyme. Zones of clearing in the M. lysodeikticuslagarose gel were visible around wells containing samples of extract from Strep. mutans that had been mixed with saliva. Saliva also showed activity whereas an extract of untreated Strep. mutans did not. Lysozyme activity was not detected in the CIE system used here, because its isoelectric point is higher than the pH of the electrophoresis buffer and it would therefore migrate in the opposite direction to that of the other proteins present. Also no activity was detected on SDS-PAGE western-blots, probably because of the effect of the SDS. Fig. 3. Diagram of a crossed-immunoelectrophoretogram (CIE) of whole saliva against rabbit antiserum to whole saliva. Precipitin peaks were stained with Coomassie brilliant blue.
were five distinct peaks on autoradiographs of CIE slides. These were antigens 2, 3, 4, 5 and 6 (see Fig. 3); peak 3 had a spur on it indicating that it may be two separate antigens. identification of CIE antigen peaks and correlation with SDS-PAGE. Saliva antigen peak 1 was identified as cc-amylase enzymically by performing CIE of a sample of saliva in agarose incorporating 0.5 per cent starch (BDH), incubating at 37°C for 1 h afterwards, then staining with iodine solution. Antigen peak 2 was shown to be IgA by CIE with anti-IgA serum in an intermediate gel. Peak 2 was extracts by also precipitated from ‘*‘iodine-labelled anti-IgA serum. Peak 8 was shown to be albumin by CIE into anti-human albumin serum. To identify the molecular weight of the salivary proteins binding to Strep. mutans (3209), the “‘iodine-1abelled peaks were cut from the CIE slide, taking care to keep contamination of each by neighbouring antigen material to a minimum. SDS-PAGE and autoradiography of these excised peaks revealed that, despite caution, considerable overlap of peaks had occurred but it was still possible to deduce the approximate mol. wt of the antigens. Figure 4 shows a diagram of an autoradiogram of the ‘2siodine-labelled proteins on SDSSPAGE present in each excised antigen peak. Peaks 2, 3 and 4 were similar but peak 2 had a band of mol. wt 59,000 that was absent from the others. This lines up with the position that IgA heavy chain runs to on our SDS-PAGE and confirms the identification of this antigen obtained by CIE into specific anti-IgA serum. Peak 4 had the strongest low moi. wt band and therefore peak 3 may be a mixture of the high mol. wt material in the stacking gel and the strong 150,000 mol. wt band. Peaks 5 and 6 were inseparable by the cutting-out technique and so were run together on the SDS-PAGE. The 150,000 and the 20-25,000 mol. wt region were the only visible components of peaks 5 and 6; the 20-25,000 mol. wt being the strongest region. The mol. wt of the smallest
DISCUSSION
We have shown that six distinct salivary components bind to the surface of Strep. mutans serotype c strain 3209 (Table I). They were identified as (I) IgA, (2) lysozyme, (3) a low mol. wt protein (1 l-13,000), (4) a component which does not resolve sharply on SDS-PAGE but which has an apparent molecular weight of 2&25,000, (5) a 150,000 mol. wt component and (6) some high mol. wt material which does not enter the stacking section of the SDS-PAGE probably, therefore, having a mol. wt in excess of 400,000. The binding of some of these salivary components to Strep. mutans, e.g. the 20-25,000 mol. wt material,
2
3
IMWW-
-
llKb-
-
4
-
5P6
Sal. -
-
Fig. 4. Diagram of an auto-radiogram of ‘25iodine-labelled salivary proteins separated by 12 per cent SDS-PAGE that are present in each of 5 of the antigen peaks shown in Fig. 3. Peaks were excised from the CIE and boiled in SDS sample buffer before being run on the polyacrylamide gel.
Binding of salivary components to Strep. mutuns Table 1. Identification af salivary components recognized by
755
Lysozymes are reported as having isoelectric points of approx. pH 11 (Salton, 1957) and would not therefore be detected here. The low molecular-weight Binding to Molecular Identity Strep. mutans bands detected by western-blotting may well contain weight CIE peak lysozyme but no enzyme activity was detected in these amylase 1 55,000 blots. The 1l-13,000 mol. wt component is also of a 2 59,000 + similar molecular size to certain other known salivary (glycJp?tein?) + 3 (high) proteins, for example µglobulin (Evrin et al., 4 1l-13,000 + 1971) which binds to some oral streptococci (Ericson 5 150,000 f et al., 1979) and a phosphoprotein (Eggen and R61la, 6 20-25,000 + _ 7 ? 1983) found in acquired pellicle. Further work is albumin 8 68,000 required to determine if the material we detected is 9 ? identical or related to any of these components. 10 ? Mutinous glycoproteins and blood-group-reactive lysozyme + substance are salivary macromolecules that are large enough to behave in a manner on SDS-PAGE similar to the high mol. wt material described here; both these have been shown to adsorb to oral bacteria has been reported previously (Douglas and Russell, 1982) and the observation that IgA adsorbs to cells (Riilla and Kilian, 1976; Gibbons and Qureshi, 1978). Some of the factors in saliva which aggregate Strep. is in agreement with results obtained by Brandtzaeg et al. (1968). Also, ‘we confirm the observation by mutans are also considered to be high mol. wt glycoproteins (Levine ef al., 1978; Eggert, 1980) and Orstavik (1978) that lysozyme binds to some strains aggregating activity can be eluted from Strep. mutuns of streptococci. However, unlike these and other previous workers who have mainly investigated the cells that have been treated with saliva (Rundegren binding of those sa.livary components which had and Ericson, 198 1) which presents the possibility that already been charac:terized, the methods we used the high molecular-weight material is a Strep. mutans salivary agglutinin. enabled us to identify molecules for which no funcIt is possible that salivary macromolecules interact tion is yet known. Our findings show that when western-blots were used to analyse extracts, the major directly with the surfaces of oral microorganisms, and salivary component binding to all the strains of Strep. that these interactions could result in a modification mutans tested was the 2&25,000 mol. wt protein, The of the behaviour of the organisms in vivo. For identity of this material is not yet known. A prolineexample, the salivary aggregating factors just menrich protein-A (Bennick and Connell, 1971) and the tioned, may contribute to the removal of bacteria light chain of immunoglobulins migrate to a similar from the mouth. Pollock et al. (1976, 1981) have position on SDS-PAlGE as the salivary component in shown that IgA and lysozyme agglutinate cells and question, but the material did not react with specific that lysozyme has bacteriolytic activity. Both of these anti-PRP serum and it was separable from IgA, the and certain bacterial-aggregating factors have been main immunogiobulin class in saliva, by CIE. shown to be present in the acquired pellicle and may therefore contribute to tooth surface colonization by Crossed immuno-electrophoresis of whole saliva oral bacteria (Magnusson and Ericson, 1976). Howinto rabbit anti-saliva serum consistently produced ever, the ecological significance of adsorption of the 10 peaks. Eckersall and Beeley (1980) found 7 peaks and Ioneja et al. (1982) found 11 peaks in their other salivary components to Strep. mutans is as yet unknown, but a salivary component similar to the antigen analyses of h.uman saliva. This indicates that 2O-25,000 mol. wt binding component described here detection of salivary Icomponents with immunological methods varies with the antigen source and with the has been shown to bind to hydroxyapatite in vitro immunization procedures; therefore there may be (Douglas and Russell, 1982). If specific interactions are involved in the adherence of Strep. mutans to some binding components which cannot be detected pellicle, this binding component may conceivably by the methods employed here. Also, animals other play a role. than rabbits may be required for producing antibodies to some salivary proteins. It is known, however, that Strep. mutans is not an The use of CIE to analyse salivary components, initial colonizer of teeth (Carlsson, 1965) and therethat became bound to Strep. mutans in these experi- fore in vivo the organism probably does not bind directly to the acquired pellicle. It may well adhere to ments, consistently showed two main antigen peaks other already resident organisms. Clark and Gibbons in Coomassie blue-stained preparations. These corresponded to peaks 3 and 4 (Fig. 3) but surprisingly the (1977) noted that, when Strep. mutans is suspended in saliva, mimicking conditions in the mouth, the capac2O-25,OOOmol. wt component (peak S/6) did not produce an intense antigen peak in CIE. We believe ity of the bacteria to adhere to saliva-coated hydroxyapatite is almost abolished. This inhibitory phenompeaks 3 and 4 are the high mol. wt material posienon may be due either to salivary aggregation of the tioned at the top of the SDS-PAGE and the 13,OOOmol. wt protein respectively. The former of bacteria (Liljemark, Bloomquist and Germaine, 1981) or alternatively to the binding of certain salithese salivary components is probably glycoprotein in nature because of its positive reaction with periodic vary factors to specific receptors on the bacterial acid-Schiff reagent. The 13,000 mol. wt protein is of surface. This may, therefore, block the cell-surface sites which would otherwise be involved in adherence a similar molecular size to lysozyme but its isoelectric point must be below pH 8.6 to have migrated toward to the saliva-coated hydroxyapatite. Caution is therethe anode in the CIE buffer system used here. fore advisable in suggesting a role in adherence for CIE
C. W. I.
756
DOUGLAS and R. R. B. RUSSELL
this or any other salivary factor just because it binds to both the tooth surface and bacterial ce!ls. It would be desirable, however, to know which, if any, of the salivary components are involved in the colonization of the mouth by Strep. mutans. Also, identification of the bacterial surface receptors for these salivary components could conceivably provide a new approach to the development of methods for controlling Strep. mutans oral populations. Cowman, Fitzgerald and Schaefer (1976) demonstrated specific utilization of certain salivary proteins as growth substrates by strains of Strep. mutans and Strep. sanguis by observing changes in protein profiles of saliva on isoelectric focusing gels before and after incubation in the presence of the bacteria. It seems possible that some protein bands which they said disappeared from the saliva upon incubation became bound to the surface of the bacteria rather than degraded by them, although we did not detect complete disappearance of any salivary protein by SDS-PAGE. Strep. mutans strains do produce proteolytic enzymes (Cowman, Perrella and Fitzgerald, 1976; Rosengren and Winblad, 1976) and therefore the binding of some of the salivary components to Strep. mutans described here may represent an interaction between surface bound enzyme and substrate.
REFERENCES
Bennick A. and Connell G. E. (1971) Purification and partial characterization of four proteins from human parotid saliva. Biochem. J. 123, 455464. Brandtzaeg P., Fjellanger E. and Gjeruldsen S. T. (1968) Adsorption of immunoglobulin A onto oral bacteria in vivo. J. Bact. 96, 242-249.
Burnette W. N. (1981) “Western blotting”. Electrophoretic transfer of proteins from sodium dodecyl sulphatepolyacrylamide gels to unmodified nitro-cellulose and radiographic detection with antibody and radioiodinated protein A. Analyt. Biochem. 112, 1955203. Carlsson J. (1965) Zooglea-forming streptococci resembling Streptococcus sanguis isolated from dental plaque in man. Odont. Reuy 16, 348. Clark W. B. and Gibbons R. J. (1977) Influence of salivary components and extracellular polysaccharide synthesis from sucrose on the attachment of Streptococcus mutans 6715 to hydroxyapatite surfaces. Infect. Immun. 18, 514523. Cowman R. A., Fitzgerald R. J. and Schaefer S. J. (1976) Role of salivary factors in the nitrogen metabolism of plaque-forming oral Streptococci. In: Microbial Aspects Dental Caries (Edited by Stiles H. M., Loesche W. J. and O’Brien T. C.) pp. 465475. Information Retrieval, Washington. Cowman R. A., Perrella M. M. and Fitzgerald R. J. (1976) Caseinolytic and glycoprotein hydrolase activity of Strep-
of
oral Biol. 28, 567-573.
Douglas C. W. I. and Russell R. R. B. (1982) The interaction of salivary components with oral streptococci. In: and Streptococcal
Reu. 44, 331-384.
Hay D. I., Gibbons R. J. and Spine11D. M. (1971) Characteristics of some high molecular weight constituents with bacterial aggregating activity from whole saliva and dental plaque.-Caries I&. 5, li l-l 13. Joneia J. M.. Greenstein P. M.. Revis G. J. and Crowler A. J. - (1982) Identification of transcalciferin as a major component of human parotid saliva by crossed immunoelectrophoretic mapping. Archs oral Biol. 27, 51-58. Kashket S. and Donaldson C. G. (1972) Saliva-induced aggregation of oral streptococci. J. ‘Bact.‘112, 1127-l 133. Lenz H. and Miihlemann H. R. (1963) Repair of etched enamel exposed to the oral environment. Helv. odont. Acta 7, 4749.
Levine M. J., Herzberg M. C., Levine M. S., Ellison S. A., Stinson M. W., Li H. C. and van Dyke T. (1978) Specificity of salivary-bacterial interactions: Role of terminal sialic acid residues in the interaction of salivary glycoproteins with Streptococcus sanguis and Streptococcus mutans. Infect. Immun. 19, 107-l 15. Liljemark W. F., Bloomquist C. G. and Germaine G. R. (1981) Effect of bacterial aggregation on the adherence of oral streptococci to hydroxyapatite. Infect. Immun. 31, 935-941. Magnusson I. and Ericson Th. (1976) Effect of salivary agglutinins on reactions between hydroxyapatite and a serotype c strain of Streptococcus mutans. Caries Res. 10, 273-286. Mayhall C. W. (1970) Concerning the composition and source of the acquired enamel pellicle of human teeth. Archs oral Biol. 15, 1327-1341. Norrild B., Bjerrum 0. J. and Vestergaard B. F. (1977) Polypeptide analysis of individual immunoprecipitates from crossed immunoelectrophoresis. Analyt. Biochem. 81, 43241. Orstavik D. (1978) The in vitro attachment of an oral streptococcus sp. to the acquired tooth enamel pellicle. Archs oral Biol. 23, 167-174.
tococcus mutans. J. dent. Res. 55, 391-399.
Douglas C. W. I. (1983) The binding of human salivary cc-amylase by oral strains of streptococcal bacteria. Archs
Basic Concepts of Streptococci
Eggen K. H. and Rolla G. (1983) Further studies on the composition of the acquired enamel pellicle. &and. J. dent. Res. 91, 439446. Eggert F. M. (1980) The nature of secretory agglutinins and aggregating factors I. Secretory conglutinin-like factor, secretory bacterial aggregating factors and secretory IgA antibody in human saliva and amniotic fluid. Int. Arehs Allergy appl. Immun. 61, 192-202. Ericson D., Bratthall D., Bjijrck L., Myhre E. and Kronvall G. (1979) Interactions between human serum proteins and oral streptococci reveal occurrence of receptors and aggregated BZmicroglobulin. Infect. Immun. 25,279-283. Ericson Th., Pruitt K. and Wedel H. (1975) The reaction of salivary substances with bacteria. J. oral Path. 4,307-323. Evrin P.-E., Peterson P. A., Wide L. and Berggard I. (1971) Radioimmunoassay of p-microglobulin in human biological fluids. &and. J. ciin. Lab. Invest. 28, 439-443. Gibbons R. J. and Qureshi J. B. (1978) Selective binding of blood group reactive salivary mucins by Streptococcus mutans and other oral organisms. Ifect. Immun. 22, 665-67 1. Hamada S. and Slade H: D. (1980) Biology, immunology and cariogenicity of Streptococcus mutans. Microbial.
Diseases
(Edited by Holm S. E. and Christensen P.) pp. 116-l 18. Reedbooks, Chertsey, Surrey. Eckersall P. D. and Beeley J. (1980) A comparison of crossed immunoelectrophoresis and crossed immunoelectrofocusing for the investigation of human salivary proteins. Electrophoresis 1, 62-67.
Orstavik D. (1980) Salivary factors in initial plaque formation. In: Microbial Adhesion to Surfaces (Edited bv Berkley R. C. W., Lynch J. M., Melling-J., Rutter P. R. and Vincent B.), _. DO.407-424. Ellis Horwood. Chichester. Orstavik D. and Kraus F. W. (1973) The acquired pellicle immunofluorescent demonstration of specific proteins. J. oral Path. 2, 68-76.
Ossennan E. F. and Lawlor D. P. (1966) Serum and urinary lysozyme (muramidase) in monocytic and monomyclocytic leukemia. J. exp. Med. 124, 921-952. Pollock J. J., Iacono V. J., Goodman Bicker H., Mackay B. M., Katona L. I., Taichman L. B. and Thomas E.
Binding of salivary components to Strep. mutans (1976) The binding, aggregation and lytic properties of lysozyme. In: Microbial Aspects of Dental Caries (Edited by Stiles H. M., Loesche W. J. and O’Brien T. C.) pp. 325-352. Information Retrieval, Washington. Pollock J. J., Katona L. I., Goodman H., Cho M. I. and Iacono V. J. (1981) Bacteriolysis of Streptococcus mutans BHT by lysozyme and. inorganic anions normally present in human saliva. Archs oral Biol. 26, 711-716. Rajan A. I. and Bennick A. (1983) Demonstration of proline-rich proteins in rabbit parotid saliva and partial characterization of some of the proteins. Archs oral Biol. 28, 431439.
Riilla G. and Kilian M. (1976) Haemagglutination activity of plaque forming bacteria. Caries Res. 11, 85-89. Rosengren L. and Winblad B. (1976) Proteolytic activity of Streptococcus mutans. Oral Surg. 42, 810-809.
Rundegren reactions strain of Russell R.
J. and Ericson T. (1981) Effect of calcium on between salivary agglutinin and a serotype c Streptococcus mutans. J. oral Path. 10, 269-275. R. B. (1979) Glucan-binding proteins of Streptococcus muzans serotype c. J. gen. Microbial. 112, 197-201.
757
Sahon M. R. J. (1957) The properties of lysozyme and its action on microorganisms. Bad. Rev. 21, 82-99. Segrest J. P. and Jackson R. L. (1972) Molecular weight determination of glycoproteins by polyacrylamide gel electrophoresis in sodium dodecyl sulphate. Meth. Enzym. XXVIII,
54-63.
Sirisinha S. (1970) Reactions of human salivary immunoglobulins with indigenous bacteria. Archs oral Biol. 15, 551-554. Steinbuch M. and Audran R. (1969) The isolation of IgG from mammalian sera with the aid of caprylic acid. Archs Biochem. Biophys. 134, 279-284.
Swanstrom R. and Shank P. R. (1978) X-ray intensifying screens greatly enhance the detection by autoradiography of the radioactive isotopes 32P and 1251.Analyt. Biochem. 86, 184-192.
Williams R. C. and Gibbons R. J. (1975) Inhibition of streptococcal attachment to receptors on human buccal epithelial cells by antigenically similar salivary glycoproteins. Infect. Immun. 11, 71 I-718.
Archs oral Bid. Vol. 29, No. 10, pp. 751-757, 1984 Printed in Great Britain. ,411rights reserved
THE ADSORPTION OF HUMAN SALIVARY COMPONENTS TO STRAINS OF THE BACTERIUM STREPTOCOCCUS
MUTANS
C. W. I. DOUGLAS* and R. R. B. RUSSELL Royal College of Surgeons of England, Dental Research Unit, Downe, Orpington, Kent, England, U.K. Summary-U:sing strains of Streptococcus mutans suspended in human saliva, the salivary proteins capable of binding to the surface of the bacteria were identified by immunological and electrophoretic techniques. Six binding components were recognized: IgA, lysozyme, some high molecular weight material (> 400,000), probably a glycoprotein, a low molecular weight component (1 l-l 3,000) a 150,000 mol. wt protein, and one major component, mol. wt 2&25,000 which did not resolve fully on SDS-polyacrylamide electrophoresis. All these salivary components could be desorbed from the bacteria with 1 M NaCl, and subsequent extraction of the same cells with 6M guanidineHC1 did not release any more salivary material. The significance of the binding of these salivary components is unknown but some may modify the behaviour of the organisms in t&o.
INTRODUCTION
The oral bacterium Streptococcus mutans is thought to be a major aetiological agent in dental caries (Hamada and Sladie, 1980). Considerable attention has therefore been paid to factors affecting the distribution of Strep. mutans in the mouth and particularly to the mechanisms whereby the organism adheres to and colonizes the tooth surface. Teeth in vivo are covered with an integument called the acquired pellicle, which is largely composed of salivary proteins and glycoproteins (Lenz and Miihlemann, 1963; Mayhall, 1970; Mrstavik and Kraus, 1973); it is now recognized that the initial deposition of bacteria on to the tooth must involve some interaction between bacterial surface components and salivary constituents of the acquired pellicle (Orstavik, 1980). Oral bacteria bind a variety of salivary components including immunoglobulins (Brandtzaeg, Fjellanger and Gjeruldsen, 1968; Sirisinha, 1970), salivaryaggregating factors (Hay, Gibbons and Spinell, 1971; Kashket and Donaldson, 1972; Ericson, Pruitt and Wedel, 1975; Mrstavik, 1978; Eggert, 1980), lysozyme (Pollock et al., 1976; IZlrstavik, 1978), blood groupreactive substances (Williams and Gibbons, 1975; Rijlla and Kilian, 1!)76; Gibbons and Qureshi, 1978) aggregated /I-2-microglobulin (Ericson et al., 1979) and a-amylase (Douglas, 1983). Some of these are present in the acquired pellicle, and so it is possible that such molecules could act as receptor sites on the tooth surface to which bacteria could bind. It is also possible that binding of salivary molecules to bacteria could have the opposite effect of reducing attachment to teeth, either by aggregating the bacteria and so enhancing their susceptibility to clearance by normal salivary flow and swallowing or by altering the properties of their surface (Magnusson and Ericson, 1976; Mrstavik, 1978).
*Present address: Department of Oral Pathology, School of Clinical Dentistry, University of Sheffield, Sheffield SIO 2TA, England, U.K. 751
Our aim was to define the components of saliva which adsorb to the surface of Strep. mutans. A preliminary report of some of the information has been presented elsewhere (Douglas and Russell, 1982). MATERIALS AND METHODS Bacteria and culture conditions
Nine strains of Strep. mutans were used, representing a range of serotypes as follows: AHT (a), FAl (b), 3209 (c), Ingbritt (c), 1649 (c), B13 (d), P4 (e), 85 (e), and Kl (g). Strains 85 and 1649 were isolated in these laboratories from dental plaque of macaque monkey (Macaca fascicularis) and strain 3209 was obtained from Dr J. A. Cole, Department of Biochemistry, University of Birmingham, U.K. The others were from our laboratory stock-collection. All strains were maintained on Todd-Hewitt agar at 4°C and when required were subcultured directly into a semidefined medium (Cas MM) described by Russell (1979) but with 0.5 per cent (w/v) casein hydrolysate (Oxoid) instead of individual amino acids. Saliva
Whole saliva from several people was stimulated by chewing Parafilm (American Can Co., Greenwich, U.S.A.) and immediately cleared by centrifugation at 14,OOOg in a Beckman model 52-21 centrifuge for 10min at 8°C. Samples were stored at 4°C for a maximum of 8 h. A large proportion of the work was carried out using saliva from one of us (C.W.1.D) but the phenomena described here applied to all saliva samples tested. of salivary proteins to Strep. mutans Each of the Strep. mutans strains was cultured overnight in 20ml Cas MM, harvested by centrifugation and washed twice in l/lOth strength Dulbecco phosphate-buffered saline (PBS, Oxoid). The cells were then mixed with 10 ml of saliva for 30 min on
Aakorption
C. W. I. DOUGLAS and R. R. B. RUSSELL
152
a spiromixer (Denley Ltd, Billingshurst, Sussex, U.K.), washed a further three times in l/10 PBS and extracted by agitation in 1 ml 1 M NaCl for 30min. After removing cells by centrifugation, the supernatants were dialysed (610,000 mol. wt cut-off) overnight against distilled water. The cell pellet was then resuspended in 1 ml 6 M guanidine-HCl (pH 3.7) and further extracted by agitation for 30 min to solubilize material that may be tightly bound. The supernatant extract was also dialysed against distilled water. To obtain larger quantities of extract, cells of Strep. mutans strain 3209 collected from 400 ml Cas MM 24-h cultures were used and treated in a manner similar to that just described, with the exception that 25 ml of saliva was used to resuspend the washed cells and these were subsequently extracted with 5 ml 1 M NaCl. The 6M guanidine step was omitted. All dialysed extracts were concentrated by lyophilization. SDS-polyacrylamide PAGE)
gel
electrophoresis
(SDS-
Electrophoresis in the presence of SDS was performed as described by Russell (1979) using 12 per cent (w/v) acrylamide gels. Gels were fixed and stained in one step by immersion of the gel in 0.2 per cent (w/v) Coomassie brilliant blue [Page Blue 83 (BDS)] in acetic acid-ethanol-water (10:45:45 by vol.). De-staining was effected in a similar solvent mixture using methanol instead of ethanol. Proteins for molecular weight standards were obtained from Sigma. Gels were also stained for glycoprotein using the periodic acid-Schiff method described by Segrest and Jackson (1972). Western blotting
Electrophoretic transfer of proteins from SDS-polyacrylamide gels (i.e. western-blotting) was performed essentially according to the method of Burnette (198 1) but with the following modifications; (1) a trans-blot apparatus (Bio-Rad Laboratories Ltd, Watford, England) was used; (2) 5 per cent (w/v) haemoglobin solution in tris-saline (10 mM tris-HCl, 0.15 M NaCl, pH 7.4) was used for blocking freeprotein binding sites on the nitrocellulose and for incubations in antiserum and enzyme-antibody conjugates; (3) antigen bands on the nitrocellulose were detected by anti-immunoglobulin peroxidase conjugate (Dako and Sigma) at a dilution of l/200 for 2 h at room temperature. Staining was with 1 ml 0.6 per cent (w/v) cl-chloro-1-naphthol (Sigma) in methanol, 9 ml 50 mM tris pH 7.4 and 10 ~1 hydrogen peroxide (30 vol). Bands appeared dark grey on a white background.
Strep. mutans antigens by CIE, immunodiffusion and western-blotting. None was found. Goat antiserum to human salivary proline-rich protein A (PRP-A) was a kind gift from Dr A. Bennick, Department of Biochemistry, University of Toronto, Canada. Goat anti-IgA serum was obtained from Meloy Laboratories, Springfield, U.S.A. and rabbit anti-human albumin serum from Dako. Crossed-immunoelectrophoresis
(CIE)
The method employed was similar to that described by Eckersall and Beeley (1980). In some experiments, antigen samples were labelled with 12?odine using iodobeads (Pierce Chemical Co., Rockford, U.S.A.) before running in the CIE system as follows: 5 ~1 samples of radiolabelled material were mixed with 10 pl unlabelled saliva and, after electrophoresis and staining, non-overlapping regions of precipitation peaks were carefully cut out using the method of Norrild, Bjerrum and Vestergaard (1977). Each sample was then boiled in SDS sample buffer and examined by SDS-PAGE. Gels were dried on Whatman 3MM filter paper using a Bio-Rad Model 224 drier (Bio-Rad). Autoradiography
Radioactively-labelled protein bands on dried SDS-PAGE gels and peaks in crossed-immunoelectrophoretograms were detected by exposure to X-omat AR film (Kodak) in conjunction with a Cronex intensifying screen (Du Pont de Numours & Co., Wilmington, U.S.A.) at -20°C (Swanstrom and Shank, 1978). Lysozyme
Lysozyme activity was detected in agarose gels by the method of Osserman and Lawlor (1966) but with the modification that 7.5 x 5cm glass microscope slides were covered with 8 ml of 1 per cent agarose (Sigma) containing 4 mg of a freeze-dried preparation of Micrococcus lysodeiktcus (Sigma) in 0.05 M phosphate buffer pH 6.5.
RESULTS
Antisera
of saliva with Strep. mutans Approximately 20 distinct protein bands in the native saliva samples could be resolved using SDS-PAGE. Adsorption of saliva with an excess of Strep. mutans (3209) cells markedly reduced the intensity of one band on SDS-PAGE which resolved as a zone rather than a sharp band having a mol. wt of 2&25,000 (see Fig. 1). This material was present in all the saliva samples tested and was absorbable by Strep. mutans cells in all cases. No other changes in protein band pattern could be seen after the adsorption of saliva with Strep. mutans.
Anti-saliva serum was raised in rabbits by three intramuscular injections at 21-day intervals of 0.9 ml cleared pooled saliva + 0.1 ml Alu Gel S (Serva) as adjuvant. The animals were bled 7 days after the third injection and the IgG fraction of the serum prepared by precipitation with caprylic acid according to the method of Steinbuch and Audran (1969). The resulting antibody was tested for reactivity with
to Strep. mutans To investigate binding to Strep. mutans, bacteria that had been mixed with saliva were extracted with 1 M NaCl and 6 M guanidine-HCl. Both extraction procedures solubilized Strep. mutans proteins and it was found that extracts yielded a large number of bands visible on SDS-PAGE. To identify the pro-
Adsorption
Binding of salivary proteins
Binding of salivary components to Strep. mutans
--
-
--
-
iiiiiiim ----- -- -
&K._
== ---
-
--
Fig. 1. Diagram of a 12 per cent SDS-polyacrylamide gel stained with Coomassie brilliant blue showing the proteins of (A) whole saliva and (B) saliva that has been absorbed with Strep. muians (3209) cells.
153
Western-blots of extracts taken from a range of Strep. mutans strains that had been treated with saliva showed that the major component binding to all of them was the 20-25,000 mol. wt material. There was some variability in the binding of some of the other components. The strains of serotypes c and e (85, 1649, 3209, Ingbritt and P4) appeared to bind more IgA than the other strains, as detected by intensity of the band on nitrocellulose. Low levels were detected in extracts of strain FAl, whereas IgA was not found in extracts of strains B13, AHT and Kl. The high mol. wt material at the top of the stacking section of the gel and the 150,000 mol. wt band were present in all extracts. Only the extracts of strain FAl consistently contained the 13,000 mol. wt protein, but this or another component of similar molecular weight was often seen in extracts of serotype c strains. Crossed-immunoelectrophoresis. The antiserum to saliva precipitated 10 distinct antigens under standard CIE conditions (Fig. 3). Analysis by CIE of saliva that had been adsorbed with Strep. mutans showed that peaks 3, 4, 5 and 6 were considerably lower and the area under each peak was reduced by approx. 35, 17, 5 and 14 per cent respectively. Extracts of Strep. mutans strain 3209 that had been coated with saliva contained two antigen peaks, corresponding to the positions of 3 and 4, that were visible by staining with Coomassie blue. However, when the sensitivity of antigen detection was increased by labelling the extracts with lz5iodine, there
teins of salivary origin in this mixture, the following techniques were used: Western-blotting. Protein bands on SDS-PAGE
were transferred to nitrocellulose and the salivary components present identified by probing with antisaliva serum. A maximum of five bands were detected in NaCl extracts (Fig. 2) and no more salivary proteins could be seen in 6 M guanidine-HCl extracts. Two of the bands, the low mol. wt (13,000) component and the high mol. wt material in the stacking gel, were not always observed. The major staining component resolved as a zone with mol. wt between 20 and 25,000, which corresponded to the band described above that was adsorbable by treatment of saliva with. Strep. mutuns. cells. One of the human salivary acidic proline-rich proteins (A) isolated by Bennick and Connell (1971) migrates anomalously in the presence of SDS and although described as having a mol. wt of 11,145 behaves as if its mol. wt were 24,000 (Rajan and Bennick, 1983). Probing western-blots of the salivary binding components with anti-PRP-A serum failed to identify the 20-25,000 mol. wt region as the prolinerich protein. The salivary protein band on western-blots having a mol. wt of 59,000 was shown to be IgA heavy chain by specific reaction with anti-IgA serum. The two other salivary components binding to Strep. mutans cells were a protein with mol. wt approx. 150,000 and some high mol. WI: material that did not enter the stacking section of the gel and therefore could not be resolved by the SDS-PAGE system we used.
UK,-
Fig. 2. Diagram of all the salivary proteins that bind to separated by SDS-PAGE and identified by probing western-blots of the gel with antisaliva serum. Track A shows the salivary proteins present in 1 M NaCl extracts of Strep. murans cells that had been coated with saliva and B, bands detectable in whole saliva.
Strep. mutans strains
154
C. W. I. DOUGLAS and R. R. B. RUSSELL
salivary protein binding to Strep. mutans (3209) was approx. 11,000 and appeared to be smaller than the band detected by western-blotting (i3,OOO). It is not certain whether these components are identical; they may have migrated differently in the two systems employed. Lysozyme. Zones of clearing in the M. lysodeikticuslagarose gel were visible around wells containing samples of extract from Strep. mutans that had been mixed with saliva. Saliva also showed activity whereas an extract of untreated Strep. mutans did not. Lysozyme activity was not detected in the CIE system used here, because its isoelectric point is higher than the pH of the electrophoresis buffer and it would therefore migrate in the opposite direction to that of the other proteins present. Also no activity was detected on SDS-PAGE western-blots, probably because of the effect of the SDS. Fig. 3. Diagram of a crossed-immunoelectrophoretogram (CIE) of whole saliva against rabbit antiserum to whole saliva. Precipitin peaks were stained with Coomassie brilliant blue.
were five distinct peaks on autoradiographs of CIE slides. These were antigens 2, 3, 4, 5 and 6 (see Fig. 3); peak 3 had a spur on it indicating that it may be two separate antigens. identification of CIE antigen peaks and correlation with SDS-PAGE. Saliva antigen peak 1 was identified as cc-amylase enzymically by performing CIE of a sample of saliva in agarose incorporating 0.5 per cent starch (BDH), incubating at 37°C for 1 h afterwards, then staining with iodine solution. Antigen peak 2 was shown to be IgA by CIE with anti-IgA serum in an intermediate gel. Peak 2 was extracts by also precipitated from ‘*‘iodine-labelled anti-IgA serum. Peak 8 was shown to be albumin by CIE into anti-human albumin serum. To identify the molecular weight of the salivary proteins binding to Strep. mutans (3209), the “‘iodine-1abelled peaks were cut from the CIE slide, taking care to keep contamination of each by neighbouring antigen material to a minimum. SDS-PAGE and autoradiography of these excised peaks revealed that, despite caution, considerable overlap of peaks had occurred but it was still possible to deduce the approximate mol. wt of the antigens. Figure 4 shows a diagram of an autoradiogram of the ‘2siodine-labelled proteins on SDSSPAGE present in each excised antigen peak. Peaks 2, 3 and 4 were similar but peak 2 had a band of mol. wt 59,000 that was absent from the others. This lines up with the position that IgA heavy chain runs to on our SDS-PAGE and confirms the identification of this antigen obtained by CIE into specific anti-IgA serum. Peak 4 had the strongest low moi. wt band and therefore peak 3 may be a mixture of the high mol. wt material in the stacking gel and the strong 150,000 mol. wt band. Peaks 5 and 6 were inseparable by the cutting-out technique and so were run together on the SDS-PAGE. The 150,000 and the 20-25,000 mol. wt region were the only visible components of peaks 5 and 6; the 20-25,000 mol. wt being the strongest region. The mol. wt of the smallest
DISCUSSION
We have shown that six distinct salivary components bind to the surface of Strep. mutans serotype c strain 3209 (Table I). They were identified as (I) IgA, (2) lysozyme, (3) a low mol. wt protein (1 l-13,000), (4) a component which does not resolve sharply on SDS-PAGE but which has an apparent molecular weight of 2&25,000, (5) a 150,000 mol. wt component and (6) some high mol. wt material which does not enter the stacking section of the SDS-PAGE probably, therefore, having a mol. wt in excess of 400,000. The binding of some of these salivary components to Strep. mutans, e.g. the 20-25,000 mol. wt material,
2
3
IMWW-
-
llKb-
-
4
-
5P6
Sal. -
-
Fig. 4. Diagram of an auto-radiogram of ‘25iodine-labelled salivary proteins separated by 12 per cent SDS-PAGE that are present in each of 5 of the antigen peaks shown in Fig. 3. Peaks were excised from the CIE and boiled in SDS sample buffer before being run on the polyacrylamide gel.
Binding of salivary components to Strep. mutuns Table 1. Identification af salivary components recognized by
755
Lysozymes are reported as having isoelectric points of approx. pH 11 (Salton, 1957) and would not therefore be detected here. The low molecular-weight Binding to Molecular Identity Strep. mutans bands detected by western-blotting may well contain weight CIE peak lysozyme but no enzyme activity was detected in these amylase 1 55,000 blots. The 1l-13,000 mol. wt component is also of a 2 59,000 + similar molecular size to certain other known salivary (glycJp?tein?) + 3 (high) proteins, for example µglobulin (Evrin et al., 4 1l-13,000 + 1971) which binds to some oral streptococci (Ericson 5 150,000 f et al., 1979) and a phosphoprotein (Eggen and R61la, 6 20-25,000 + _ 7 ? 1983) found in acquired pellicle. Further work is albumin 8 68,000 required to determine if the material we detected is 9 ? identical or related to any of these components. 10 ? Mutinous glycoproteins and blood-group-reactive lysozyme + substance are salivary macromolecules that are large enough to behave in a manner on SDS-PAGE similar to the high mol. wt material described here; both these have been shown to adsorb to oral bacteria has been reported previously (Douglas and Russell, 1982) and the observation that IgA adsorbs to cells (Riilla and Kilian, 1976; Gibbons and Qureshi, 1978). Some of the factors in saliva which aggregate Strep. is in agreement with results obtained by Brandtzaeg et al. (1968). Also, ‘we confirm the observation by mutans are also considered to be high mol. wt glycoproteins (Levine ef al., 1978; Eggert, 1980) and Orstavik (1978) that lysozyme binds to some strains aggregating activity can be eluted from Strep. mutuns of streptococci. However, unlike these and other previous workers who have mainly investigated the cells that have been treated with saliva (Rundegren binding of those sa.livary components which had and Ericson, 198 1) which presents the possibility that already been charac:terized, the methods we used the high molecular-weight material is a Strep. mutans salivary agglutinin. enabled us to identify molecules for which no funcIt is possible that salivary macromolecules interact tion is yet known. Our findings show that when western-blots were used to analyse extracts, the major directly with the surfaces of oral microorganisms, and salivary component binding to all the strains of Strep. that these interactions could result in a modification mutans tested was the 2&25,000 mol. wt protein, The of the behaviour of the organisms in vivo. For identity of this material is not yet known. A prolineexample, the salivary aggregating factors just menrich protein-A (Bennick and Connell, 1971) and the tioned, may contribute to the removal of bacteria light chain of immunoglobulins migrate to a similar from the mouth. Pollock et al. (1976, 1981) have position on SDS-PAlGE as the salivary component in shown that IgA and lysozyme agglutinate cells and question, but the material did not react with specific that lysozyme has bacteriolytic activity. Both of these anti-PRP serum and it was separable from IgA, the and certain bacterial-aggregating factors have been main immunogiobulin class in saliva, by CIE. shown to be present in the acquired pellicle and may therefore contribute to tooth surface colonization by Crossed immuno-electrophoresis of whole saliva oral bacteria (Magnusson and Ericson, 1976). Howinto rabbit anti-saliva serum consistently produced ever, the ecological significance of adsorption of the 10 peaks. Eckersall and Beeley (1980) found 7 peaks and Ioneja et al. (1982) found 11 peaks in their other salivary components to Strep. mutans is as yet unknown, but a salivary component similar to the antigen analyses of h.uman saliva. This indicates that 2O-25,000 mol. wt binding component described here detection of salivary Icomponents with immunological methods varies with the antigen source and with the has been shown to bind to hydroxyapatite in vitro immunization procedures; therefore there may be (Douglas and Russell, 1982). If specific interactions are involved in the adherence of Strep. mutans to some binding components which cannot be detected pellicle, this binding component may conceivably by the methods employed here. Also, animals other play a role. than rabbits may be required for producing antibodies to some salivary proteins. It is known, however, that Strep. mutans is not an The use of CIE to analyse salivary components, initial colonizer of teeth (Carlsson, 1965) and therethat became bound to Strep. mutans in these experi- fore in vivo the organism probably does not bind directly to the acquired pellicle. It may well adhere to ments, consistently showed two main antigen peaks other already resident organisms. Clark and Gibbons in Coomassie blue-stained preparations. These corresponded to peaks 3 and 4 (Fig. 3) but surprisingly the (1977) noted that, when Strep. mutans is suspended in saliva, mimicking conditions in the mouth, the capac2O-25,OOOmol. wt component (peak S/6) did not produce an intense antigen peak in CIE. We believe ity of the bacteria to adhere to saliva-coated hydroxyapatite is almost abolished. This inhibitory phenompeaks 3 and 4 are the high mol. wt material posienon may be due either to salivary aggregation of the tioned at the top of the SDS-PAGE and the 13,OOOmol. wt protein respectively. The former of bacteria (Liljemark, Bloomquist and Germaine, 1981) or alternatively to the binding of certain salithese salivary components is probably glycoprotein in nature because of its positive reaction with periodic vary factors to specific receptors on the bacterial acid-Schiff reagent. The 13,000 mol. wt protein is of surface. This may, therefore, block the cell-surface sites which would otherwise be involved in adherence a similar molecular size to lysozyme but its isoelectric point must be below pH 8.6 to have migrated toward to the saliva-coated hydroxyapatite. Caution is therethe anode in the CIE buffer system used here. fore advisable in suggesting a role in adherence for CIE
C. W. I.
756
DOUGLAS and R. R. B. RUSSELL
this or any other salivary factor just because it binds to both the tooth surface and bacterial ce!ls. It would be desirable, however, to know which, if any, of the salivary components are involved in the colonization of the mouth by Strep. mutans. Also, identification of the bacterial surface receptors for these salivary components could conceivably provide a new approach to the development of methods for controlling Strep. mutans oral populations. Cowman, Fitzgerald and Schaefer (1976) demonstrated specific utilization of certain salivary proteins as growth substrates by strains of Strep. mutans and Strep. sanguis by observing changes in protein profiles of saliva on isoelectric focusing gels before and after incubation in the presence of the bacteria. It seems possible that some protein bands which they said disappeared from the saliva upon incubation became bound to the surface of the bacteria rather than degraded by them, although we did not detect complete disappearance of any salivary protein by SDS-PAGE. Strep. mutans strains do produce proteolytic enzymes (Cowman, Perrella and Fitzgerald, 1976; Rosengren and Winblad, 1976) and therefore the binding of some of the salivary components to Strep. mutans described here may represent an interaction between surface bound enzyme and substrate.
REFERENCES
Bennick A. and Connell G. E. (1971) Purification and partial characterization of four proteins from human parotid saliva. Biochem. J. 123, 455464. Brandtzaeg P., Fjellanger E. and Gjeruldsen S. T. (1968) Adsorption of immunoglobulin A onto oral bacteria in vivo. J. Bact. 96, 242-249.
Burnette W. N. (1981) “Western blotting”. Electrophoretic transfer of proteins from sodium dodecyl sulphatepolyacrylamide gels to unmodified nitro-cellulose and radiographic detection with antibody and radioiodinated protein A. Analyt. Biochem. 112, 1955203. Carlsson J. (1965) Zooglea-forming streptococci resembling Streptococcus sanguis isolated from dental plaque in man. Odont. Reuy 16, 348. Clark W. B. and Gibbons R. J. (1977) Influence of salivary components and extracellular polysaccharide synthesis from sucrose on the attachment of Streptococcus mutans 6715 to hydroxyapatite surfaces. Infect. Immun. 18, 514523. Cowman R. A., Fitzgerald R. J. and Schaefer S. J. (1976) Role of salivary factors in the nitrogen metabolism of plaque-forming oral Streptococci. In: Microbial Aspects Dental Caries (Edited by Stiles H. M., Loesche W. J. and O’Brien T. C.) pp. 465475. Information Retrieval, Washington. Cowman R. A., Perrella M. M. and Fitzgerald R. J. (1976) Caseinolytic and glycoprotein hydrolase activity of Strep-
of
oral Biol. 28, 567-573.
Douglas C. W. I. and Russell R. R. B. (1982) The interaction of salivary components with oral streptococci. In: and Streptococcal
Reu. 44, 331-384.
Hay D. I., Gibbons R. J. and Spine11D. M. (1971) Characteristics of some high molecular weight constituents with bacterial aggregating activity from whole saliva and dental plaque.-Caries I&. 5, li l-l 13. Joneia J. M.. Greenstein P. M.. Revis G. J. and Crowler A. J. - (1982) Identification of transcalciferin as a major component of human parotid saliva by crossed immunoelectrophoretic mapping. Archs oral Biol. 27, 51-58. Kashket S. and Donaldson C. G. (1972) Saliva-induced aggregation of oral streptococci. J. ‘Bact.‘112, 1127-l 133. Lenz H. and Miihlemann H. R. (1963) Repair of etched enamel exposed to the oral environment. Helv. odont. Acta 7, 4749.
Levine M. J., Herzberg M. C., Levine M. S., Ellison S. A., Stinson M. W., Li H. C. and van Dyke T. (1978) Specificity of salivary-bacterial interactions: Role of terminal sialic acid residues in the interaction of salivary glycoproteins with Streptococcus sanguis and Streptococcus mutans. Infect. Immun. 19, 107-l 15. Liljemark W. F., Bloomquist C. G. and Germaine G. R. (1981) Effect of bacterial aggregation on the adherence of oral streptococci to hydroxyapatite. Infect. Immun. 31, 935-941. Magnusson I. and Ericson Th. (1976) Effect of salivary agglutinins on reactions between hydroxyapatite and a serotype c strain of Streptococcus mutans. Caries Res. 10, 273-286. Mayhall C. W. (1970) Concerning the composition and source of the acquired enamel pellicle of human teeth. Archs oral Biol. 15, 1327-1341. Norrild B., Bjerrum 0. J. and Vestergaard B. F. (1977) Polypeptide analysis of individual immunoprecipitates from crossed immunoelectrophoresis. Analyt. Biochem. 81, 43241. Orstavik D. (1978) The in vitro attachment of an oral streptococcus sp. to the acquired tooth enamel pellicle. Archs oral Biol. 23, 167-174.
tococcus mutans. J. dent. Res. 55, 391-399.
Douglas C. W. I. (1983) The binding of human salivary cc-amylase by oral strains of streptococcal bacteria. Archs
Basic Concepts of Streptococci
Eggen K. H. and Rolla G. (1983) Further studies on the composition of the acquired enamel pellicle. &and. J. dent. Res. 91, 439446. Eggert F. M. (1980) The nature of secretory agglutinins and aggregating factors I. Secretory conglutinin-like factor, secretory bacterial aggregating factors and secretory IgA antibody in human saliva and amniotic fluid. Int. Arehs Allergy appl. Immun. 61, 192-202. Ericson D., Bratthall D., Bjijrck L., Myhre E. and Kronvall G. (1979) Interactions between human serum proteins and oral streptococci reveal occurrence of receptors and aggregated BZmicroglobulin. Infect. Immun. 25,279-283. Ericson Th., Pruitt K. and Wedel H. (1975) The reaction of salivary substances with bacteria. J. oral Path. 4,307-323. Evrin P.-E., Peterson P. A., Wide L. and Berggard I. (1971) Radioimmunoassay of p-microglobulin in human biological fluids. &and. J. ciin. Lab. Invest. 28, 439-443. Gibbons R. J. and Qureshi J. B. (1978) Selective binding of blood group reactive salivary mucins by Streptococcus mutans and other oral organisms. Ifect. Immun. 22, 665-67 1. Hamada S. and Slade H: D. (1980) Biology, immunology and cariogenicity of Streptococcus mutans. Microbial.
Diseases
(Edited by Holm S. E. and Christensen P.) pp. 116-l 18. Reedbooks, Chertsey, Surrey. Eckersall P. D. and Beeley J. (1980) A comparison of crossed immunoelectrophoresis and crossed immunoelectrofocusing for the investigation of human salivary proteins. Electrophoresis 1, 62-67.
Orstavik D. (1980) Salivary factors in initial plaque formation. In: Microbial Adhesion to Surfaces (Edited bv Berkley R. C. W., Lynch J. M., Melling-J., Rutter P. R. and Vincent B.), _. DO.407-424. Ellis Horwood. Chichester. Orstavik D. and Kraus F. W. (1973) The acquired pellicle immunofluorescent demonstration of specific proteins. J. oral Path. 2, 68-76.
Ossennan E. F. and Lawlor D. P. (1966) Serum and urinary lysozyme (muramidase) in monocytic and monomyclocytic leukemia. J. exp. Med. 124, 921-952. Pollock J. J., Iacono V. J., Goodman Bicker H., Mackay B. M., Katona L. I., Taichman L. B. and Thomas E.
Binding of salivary components to Strep. mutans (1976) The binding, aggregation and lytic properties of lysozyme. In: Microbial Aspects of Dental Caries (Edited by Stiles H. M., Loesche W. J. and O’Brien T. C.) pp. 325-352. Information Retrieval, Washington. Pollock J. J., Katona L. I., Goodman H., Cho M. I. and Iacono V. J. (1981) Bacteriolysis of Streptococcus mutans BHT by lysozyme and. inorganic anions normally present in human saliva. Archs oral Biol. 26, 711-716. Rajan A. I. and Bennick A. (1983) Demonstration of proline-rich proteins in rabbit parotid saliva and partial characterization of some of the proteins. Archs oral Biol. 28, 431439.
Riilla G. and Kilian M. (1976) Haemagglutination activity of plaque forming bacteria. Caries Res. 11, 85-89. Rosengren L. and Winblad B. (1976) Proteolytic activity of Streptococcus mutans. Oral Surg. 42, 810-809.
Rundegren reactions strain of Russell R.
J. and Ericson T. (1981) Effect of calcium on between salivary agglutinin and a serotype c Streptococcus mutans. J. oral Path. 10, 269-275. R. B. (1979) Glucan-binding proteins of Streptococcus muzans serotype c. J. gen. Microbial. 112, 197-201.
757
Sahon M. R. J. (1957) The properties of lysozyme and its action on microorganisms. Bad. Rev. 21, 82-99. Segrest J. P. and Jackson R. L. (1972) Molecular weight determination of glycoproteins by polyacrylamide gel electrophoresis in sodium dodecyl sulphate. Meth. Enzym. XXVIII,
54-63.
Sirisinha S. (1970) Reactions of human salivary immunoglobulins with indigenous bacteria. Archs oral Biol. 15, 551-554. Steinbuch M. and Audran R. (1969) The isolation of IgG from mammalian sera with the aid of caprylic acid. Archs Biochem. Biophys. 134, 279-284.
Swanstrom R. and Shank P. R. (1978) X-ray intensifying screens greatly enhance the detection by autoradiography of the radioactive isotopes 32P and 1251.Analyt. Biochem. 86, 184-192.
Williams R. C. and Gibbons R. J. (1975) Inhibition of streptococcal attachment to receptors on human buccal epithelial cells by antigenically similar salivary glycoproteins. Infect. Immun. 11, 71 I-718.