Archs oral Eiol. Vol. 36, No. 2, pp. 155-I 60, 1991 Printed in Great Britain. All rightsreserved
Copyright
0
0003-9969/91 S3.00 + 0.00 1991 PergamonPressplc
INHIBITION OF THE GROWTH OF STREPTOCOCCUS MUTANS, STREPTOCOCCUS SOBRINUS AND LACTOBACILLUS CASEI BY ORAL PEROXIDASE SYSTEMS IN HUMAN SALIVA M. LUMIKARI,T. SOUKKA,S. NUEMIO and J. TENOVUO* Department of Cariology, Institute of Dentistry, University of Turku, LemminkHisenkatu 2, SF-20520 Turku, Finland (Accepted 22 August 1990) Summary-Streptococcus murans, Strep. sobrinus and L.actobacillus casei were grown in glucose-supplemented, sterilized, human whole saliva, adjusted to pH 5, 6 or 7. Components of the antibacterial peroxidase system-hypothiocyanous acid (HOSCN) and hypothiocyanite ions (OSCN-)-were generated by adding exogenous H,O, to sterilized saliva containing endogenous peroxidases and thiocyanate (SCN-) ions. HOSCN/OSCN- generation was proportional to the amount of H,O, added, and more HOSCN/OSCN- was detected in saliva at pH 7 than at pH 5. However, the growth of mutans streptococci and L. casei was inhibited at pH 5 by HOSCN/OSCN-, whereas no inhibition was found at pH 7. The findings show that (a) sufficient amount of HOSCN/OSCN- will inhibit the growth of cariogenic bacteria in human saliva at pH 5; (b) this amount of HOSCN/OSCN- can be generated in saliva by exogenously added H,O,; and (c) peroxidase systems have stronger antistreptococcal effects in human whole saliva than in phosphate buffer.
Key words: cariogenic bacteria, peroxidases, antimicrobial, whole saliva, streptococci, hypothiocyanite.
INTRODUCTION Human whole saliva contains two peroxidase enzymes which are considered important mucosal defence factors. Major salivary glands secrete salivary peroxidase (Mansson-Rahemtulla et al., 1988), which is structurally somewhat different but catalytically quite similar to bovine milk lactoperoxidase (Pruitt et al., 1988). Oral polymorphonuclear leucocytes release myeloperoxidase into gingival crevicular fluid and whole saliva (Mansson-Rahemtulla et al., 1986; Cao and Smith, 1989), in amounts proportional to the degree of gingival inflammation (Cao and Smith, 1989). Both salivary peroxidase and myeloperoxidase catalyse the oxidation of thiocyanate (SCN-) ions by hydrogen peroxide to form hypothiocyanite ions (OSCN-) or hypothiocyanous acid (HOSCN), which are the actual antimicrobial agents of oral peroxidase systems (Pruitt and Reiter, 1985). Most studies on the antibacterial activity of oral peroxidases have been done in vitro, using purified lactoperoxidase as the oxidizing enzyme and synthetic media for pure cultures of oral bacteria (Pruitt and Reiter, 1985). Under these circumstances, glucose-stimulated metabolic events (glucose and oxygen uptake, acid production) of Streptococcus mutans (Germaine and Tellefson, 1981; Carlsson, Iwami and Yamada, 1983; Thomas et of.; 1983) and *To whom correspondence should be addressed. Abbreviations: BHI, brain-heart infusion broth; DTT, dithiothreitol: NBS. 5-thio-2-nitrobenzoic acid: NBS,. 5,5-dithiobis-(2-nitrobenzoic acid); PBS, phosphatk buffered saline. AOB 36,2--D
Strep. rattus (Donoghue, Hudson and Perrons, 1987) are easily inhibited. Similar anti-streptococcal bacteriostatic inhibition has been observed with purified human salivary peroxidase (Mansson-Rahemtulla et al., 1987) and myeloperoxidase (Tenovuo et al., 1988), suggesting that the peroxidase systems may also affect the in viuo growth of oral streptococci. In order to study the antimicrobial action of peroxidase systems in saliva itself, without synthetic media and added lactoperoxidase, we have now grown Strep. mutans, Strep. sobrinus and Lactobacilhs casei in sterilized human whole saliva with variations in pH and HOSCN/OSCNconcentrations. With this procedure, we expected to get more detailed information than before as to how peroxidase systems affect the growth of cariogenic bacteria in vivo. Glucose-supplemented sterilized saliva is, however, only a model system which is not identical with the in vivo situation where there are also bacterial interactions, and where bacteria obtain nutrients by, for example, degrading salivary glycoproteins (Williams and Powlen, 1959; van der Hoeven, de Jong and van Nieuw Amerongen, 1989).
MATERIALSAND METHODS Bacterial strains and cell preparations Strep. mutans ATCC 25 175, Strep. sobrinus ATCC 33478 and L. casei ATCC 393, were used. Strep. mutans and Strep. sobrinus were grown in BHI (Oxoid Ltd, Basingstoke, England) and L. casei in Lactobacilli broth AOAC (Difco Labs, Detroit, MI, U.S.A.) at 37°C overnight. The cells were harvested 155
M. LUMIKARI et al.
156
by centrifugation (MSE Super Minor Centrifuge, Fisons, England) at 1800 g for 10 min, washed twice with PBS (Orion Diagnostica, Espoo, Finland) and resuspended in ice-chilled PBS (AeZO= 0.8). Collection and treatment of saliva
Paraffin-stimulated whole saliva (5&60 ml for each experiment) was collected from one healthy, nonsmoking, male donor in the morning before each experiment. He did not eat or drink anything 1 h prior to the collection. The saliva was sterilized by centrifugation at 40,OOOg (MSE Hi-Spin 21, Fisons, England), for 30min at 4°C and by a subsequent filtration of the supernatant through Millex@-GS filters (0.45 p, Millipore S.A., Molsheim, France). The desired pH for the different experiments was adjusted by adding sterilized lactic acid (Fluka AG, Buchs, Switzerland) to the saliva. Fresh saliva was used for all experiments. Chemical assays
The following factors were analysed before and after sterilizing the saliva samples. Lysozyme activity was estimated with Micrococcus diffusion plates (Lysozyme Kit@, Kallestad Laboratories, Chaska, MN, U.S.A.), using lyophilized human urine lysozyme as a standard. The lactoferrin concentration was analysed by a non-competitive avidin-biotin enzyme immunoassay (Vilja, Krohn and Tuohimaa, as a 1985), using human colostral lactoferrin reference. Total concentrations of salivary IgA and IgG were determined by a ‘trapping antibody’-type enzyme immunoassay (Lehtonen et al., 1984). The changes in salivary peroxidase, thiocyanate (SCN- ) and hypothiocyanite (OSCN- ) were followed before and during the experiments as described below. Salivary peroxidase activity was measured by following the rate of oxidation of NBS to (NBS),, by the generation of OSCN- ions during the oxidation of SCN- by salivary peroxidase and H,Oz (Wever et al., 1982). Peroxidase activity was expressed as milliunits (mu), according to the calculation described by Mansson-Rahemtulla et al. (1986). Hypothiocyanite ions were assayed by reaction of the coloured anionic monomer of (NBS), (Tenovuo, Pruitt and Thomas, 1982a) and SCNions were quantitated by the ferric nitrate method of Betts and Dainton (1953). Experimental
procedures
In every experiment the initial salivary concentration of HOSCN/OSCNwas determined. After centrifugation the levels of SCN- and peroxidase were measured and when the saliva sample was sterilized and the desired pH adjusted (pH 5, 6 or 7), the amounts of HOSCN/OSCNand peroxidase were again measured. To sterile tubes containing 2.5 ml of sterilized saliva, 25 ~1 of hydrogen peroxide (E. Merck AG, Darmstadt, Germany) were added to a final concentration of 50 p M, and the concomitant generation of OSCN- was followed. A portion (50 ~1) of the cell suspension of Strep. mutans, Strep. sobrinus or L. casei was added together with 25 ~1 of 1 M glucose (BDH Chemicals Ltd, Poole, England) (final
concentration 1OmM). Control tubes contained no H,O,. In every experiment a sterility control was present; tubes with only saliva and glucose but no added bacteria or H, OZ. The tubes were incubated aerobically on a shaking water bath, at 37°C for 20 h and the growth was followed spectrophotometrically (Hitachi model 101, Hitachi Ltd, Japan) at 620nm, before and after the 20-h incubation period. After the incubation, saliva-Strep. mutans or saliva-Strep. sobrinus mixtures were sonicated with an ultrasonic disintegrator (100 W, Measuring & Scientific Equipment Ltd, London, England) for 1 min at an amplitude of 7 pm. Bacterial mixtures were then transferred to Mitis Salivarius agar plates, supplemented with 20% glucose and 0.5 pg/ml bacitracin (Sigma Chemical Co., St Louis, MO, U.S.A.). The plates were incubated for 3 days, at 37”C, in candle jars. Saliva-L. casei mixtures were transferred to Rogosa SL agar plates (Difco Lab, Detroit, MI, U.S.A.) and incubated anaerobically for 3 days at 37°C. All saliva-bacteria mixtures were also cultivated on blood agar plates (Oxoid, 5% bovine blood) and incubated for 2 days in a CO,-atmosphere at 37°C. RESULTS
The sterilization of whole saliva by centrifugation and ultrafiltration, followed by pH adjustment, almost entirely removed HOSCN/OSCNfrom saliva samples, whereas SCN- levels remained practically unaffected (Table 1). The activity of salivary peroxidase and myeloperoxidase decreased in various experiments by IO-30%, but was always > 0.3 mU in the final, sterilized sample. Small concentrations of lactoferrin and lysozyme were also found in sterilized samples (Table 1). HOSCN/OSCNcould be easily regenerated in sterilized whole saliva by adding small amounts of H,O, (Fig. IA). The generation of HOSCN/OSCNwas directly proportional to the added concentration of H,O, both in saliva and in phosphate buffer (Fig. 1). There was a trend for slightly more HOSCN/OSCN- to be formed in saliva at pH 7 than at pH 5 (Fig. IA) and for HOSCN/OSCNto decompose faster at lower pH (Fig. 2). In preliminary experiments at pH 6.0 the growth of Strep. mutans was inhibited by HOSCN/OSCNin a dose-dependent way. Based on these experiments and the known in vivo levels of HOSCN/OSCN(Tenovuo et al., Table 1. Some salivary defence factors in paraffin-stimulated whole saliva before and after sterilization of the samples Agent
N’
Fresh sample
Sterilized sample
Salivary pcroxidase (mu) Thiocyanate (mM) Hypothiocyanite bM) Lactoferrin @g/ml) Lysozyme @g/ml) Total IgA (mg/l) Total IgG (mg/l)
8 9 10 3 13 7 7
0.88 f 0.26’ l.lOkO.39 18.5 f 8.6 4.17 f 0.74 9.03 f 1.33 41.0 * 20.5 11.5k4.8
0.66 f 0.28 1.11 kO.38 1.31 f 2.17 1.27 f 0.21 1.64 f 1.08 36.2 * 17.9 8.98 f 3.67
’ Number of experiments. 2Mean + SD.
Peroxidases and cariogenic bacteria
0
157
3
6
12
20
TIME Iht
0
50 100 H202 lpMI
25
250
Fig. 1. Generation of HOSCN/OSCN- in sterilized human whole saliva (A) and in 0.1 M phosphate buffer (B), both adjusted to pH 5 (m), 6 (m) or 7 (O), after addition of the indicated concentrations of H,O,. Phosphate buffer was supplemented also with lactoperoxidase (5 pg/ml) and SCN- (1 mM).
1982a), a concentration of 50 PM of H,O, was for further experiments at various pH values. The growth curve of Strep. mutans in glucosesupplemented, sterilized saliva (adjusted to pH 5.0), with and without HOSCN/OSCNis shown in Fig. 3. The number of cells decreased rapidly after addition into the saliva, followed by a relatively linear growth up to 20 h, except when HOSCN/OSCN(40-45 PM) was present. The peroxidase systems blocked the streptococcal growth completely in a few hours. Based on these time-course experiments, a 20-h growth period was selected for further studies. In some experiments we plated the bacteria 48 h after the generation of HOSCN/OSCNand still no growth was observed. Growth and HOSCN/OSCN--mediated inhibition of the bacteria were both strongly dependent on pH. chosen
Fig. 3. Growth curve of Strep. mutansin sterilized human whole saliva (adjusted to pH 5.0) with (a---0) and without (O--O) HOSCN/OSCN- (4045 PM). Before aerobic incubation of saliva or saliva-HOSCN/OSCNmixtures at 37°C the tubes were supplemented with 10 mM glucose and 50~1 of Strep. mutanscell suspension.
At pH 5.0 the growth of Strep. sobrinus (control cells) was less abundant than that of Strep. mutans (p < O.OOl), and in all experiments at this low pH HOSCN/OSCNcompletely inhibited the growth of Strep. sobrinus but not of Strep. mutans (Fig. 4). At pH 6 and 7, only slight and statistically not significant inhibition of streptococcal growth was found with HOSCN/OSCN(Fig. 4). In all these experiments, no growth at all was observed in the purity controls, i.e. in the samples of sterilized saliva with no added bacteria. The results were similar whether counted on Mitis Salivarius or blood agar plates. Also, the addition of H,O, (50 PM) without peroxidase and/or SCN- ions into BHI did not inhibit the growth of mutans streptococci at pH 5 (data not shown), suggesting that the growth inhibition was not due to H,Or but rather to HOSCN/OSCN-. Incorporation
Strep. z \
sobrinus
ao
24.0 0 g
s= 10
20
SALIVA
7
I L. cord 6.0 6D 4.0
20
ma
5n
‘0
-
BUFFER
0 0
2
4
6 6 TIME(min)
10
12
Fig. 2. Decomposition of HOSCN/OSCNin sterilized human whole saliva and in 0.1 M phosphate buffer at pH 5, 6 or 7. The concentration of added H,O, was 50/1M.
60
10
PI4
Fig. 4. Growth of Strep. mutans,Strep.sobrinusand L.casei in sterilized, glucose-supplemented human whole saliva at pH 5, 6 or 7. The saliva-bacteria mixtures were supplemented with 50pM HrO, (hatched columns), which produced 40-45 /IM HOSCN/OSCN-. The controls (open columns) contained neither H,O, nor HOSCN/OSCN-. The values are mean log counts f SD from 8-10 different experiments. *** =p -z 0.001 (Student’s r-test).
M. LUMIKARI et al.
158
of a reducing agent, DTT (final concentration 1 mM), into the reaction mixture completely abolished the growth inhibitory effect of HOSCN/OSCN- at pH 5. The addition of lactoperoxidase (5 pg/ml) together with H,O, into the saliva did not enhance the growth inhibition when compared to the addition of H,Oz alone. In contrast to the mutans streptococci, at pH 5 a smaller but still highly significant (p < 0.001) growth inhibition of L. casei by HOSCN/OSCNwas observed (Fig. 4). No inhibition occurred at pH 6 and 7. At all pH values, the growth of L. casei in glucose-supplemented human saliva was more abundant (p < 0.001) than that of mutans streptococci (Fig. 4). Possible changes in the pH of the saliva medium during incubation were also measured. In experiments with initial pH 5, HOSCN/OSCNgrowth inhibition kept the pH values stable (range &-0.2 pH units) over the entire 20-h incubation whereas the pH in the control tubes slightly decreased (range 0.2-0.5 pH units). DISCUSSION
Various oral microorganisms, including oral streptococci (de Jong e? al., 1984; van der Hoeven ef al., 1989), can grow in human saliva (Williams and Powlen, 1959; Samaranayake er al., 1986). Streptococcal cell production in saliva is, however, carbohydrate limited but, with appropriate glucose supplementation, doubling times of 1.6-3 h have been reported for oral streptococci (de Jong et al., 1984). We followed the growth in saliva by measuring the turbidity (A,,,) and by plating bacterial samples on solid media. With both methods there was some variation between different experiments, perhaps due to the aggregation of bacterial cells. When the same bacterial culture was tested in saliva samples from 5 persons, there was some inter-individual variation in the growth rates but, in all cases, the growth was inhibited at pH 5 by HOSCN/OSCN(data not shown). In order to minimize these sources of variation, in all experiments we used whole saliva from the same donor and, before growth measurements, ultrasonic dispersion to disaggregate the cells. Because AeZOreadings and plate counts gave qualitatively similar results, for simplicity we have here reported only the bacterial plate counts. Studies with centrifuged human saliva have shown that glucose uptake of saliva-suspended Strep. rattus (Strep. mutans BHT, serotype b) could be completely abolished at neutral pH when supplemented with H202 (Germaine and Tellefson, 1981, 1982). Glucosestimulated acid production in plaque-saliva mixtures (Tenovuo ef al., 1981), and in centrifuged saliva supplemented with Strep. mutans (ManssonRahemtulla et al., 1982), was also inhibited by HOSCN/OSCNin a dose-dependent way. Our results now extend these findings by showing that HOSCN/OSCNcan also prevent the growth of Strep. mutans, Strep. sobrinus and L. casei in human whole saliva. The inhibition was strongest at low pH and no inhibition at all was found at neutral pH. These findings agree with studies showing that the antimetabolic effects of the lactoperoxidase/salivary
peroxidase-myeloperoxidase SCN--H,O, systems are stronger at low pH than at neutral pH (Kersten, Moorer and Wever, 1981; Thomas et al., 1983; Mansson-Rahemtulla er al., 1987; Tenovuo et al. 1988). At low pH, the major oxidation product is HOSCN (Thomas, 1981), which is soluble in organic solvents and thus possibly penetrates the lipophilic bacterial cell envelopes more easily than the OSCNions, which dominate at neutral pH (Thomas, 1981). As expected, a reducing agent (DTT) completely abolished the growth inhibition, confirming that the HOSCN/OSCNpair was indeed responsible for it. Of the three bacterial strains, L. cnsei was most resistant to HOSCN/OSCNmediated inhibition. Although at pH 5 Strep. sobrinus was totally inhibited by HOSCNJOSCN- unlike Strep. rnutuns, this difference may be due to the slower overall growth rate of Strep. sobrinus than of Strep. mutans in saliva at low pH. Also, possible differences in endogenous H,O, production by the bacteria could have affected their susceptibility to inhibition (Pruitt and Reiter, 1985). Interestingly, with the same Strep. mutuns strain as now used, we could not find any loss of viability by the lactoperoxidase-SCN--H,O, (HOSCN/ OSCN- > 200 PM) system, even at pH 4.5, if the cells were exposed to HOSCN/OSCNin PBS for 90 min (Tenovuo et al., 1988). However, streptococci the now exposed to salivary peroxidase/ myeloperoxidase-SCN--H,O, system in saliva at low pH did not recover, although the HOSCN/OSCNconcentration was as low as 45-50pM. In both studies, H,O, was added to the incubation medium (PBS or saliva) 1 min before the addition of bacteria. Therefore, it is unlikely that any short-lived, powerful oxidation products (Tenovuo et al., 1981; Pruitt et al., 1982) could have influenced the results. The Strep. mutans strain used in these experiments is able to produce some extracellular H,Oz in the presence of glucose (Carlsson et al., 1983). This bacterial HzOz may further enhance the antibacterial activity of peroxidases (Carlsson et al., 1983; Thomas et al., 1983), but in the present study it was not enough to inhibit the cell division and therefore exogenous H,O, was added. Thus, our results suggest that in human saliva the antimicrobial effects of oral peroxidase systems may be stronger than many in vitro studies have indicated. One possibility is that some other salivary constituents enhance the antimicrobial action of peroxidases synergistically. Such interactions have been observed in vitro, e.g. secretory IgA, lactoferrin and lysozyme can enhance the antimetabolic activity of the lactoperoxidase-SCN--H,02 system (Tenovuo et al., 1982b; Moldoveanu et al., 1983; Arnold et al., 1984). Because sterilized saliva contains IgA, lactoferrin and some lysozyme, these interactions with oral peroxidase systems could have contributed to the inhibitory effects. Salivary HOSCN/OSCN- levels can be elevated in vitro and also in vivo by appropriate additions of H,Oz (Mansson-Rahemtulla er al., 1982, 1983; Pruitt, Mansson-Rahemtulia and Tenovuo, 1983). HOSCN/OSCNlevels up to 150-200 PM are easily generated in vivo at pH 5.5 (Mansson-Rahemtulla et 1983). According to our present study, al., HOSCN/OSCNin human saliva at low pH has the
Peroxidases and cariogenic bacteria potential to inhibit significantly, presumably by bactericidal activity, those bacteria which are cariogenic in man. Although clinical studies have failed to establish a correlation between salivary HOSCN/OSCN- levels and dental caries (Mandel er al., 1983; Lamberts et al., 1984; Grihn et al., 1988), activation of peroxidase systems in uiuo by appropriate amounts of exogenous H,O, may be of clinical importance (Hoogendoom, 1985). Further studies of the synergism between antimicrobial systems in human mouth may also reveal clinically applicable methods to combat pathogenic bacteria. investigation was supported by the Academy of Finland, the Finnish Dental Society, Emil Aaltonen Foundation, Niilo Helander Foundation and Svenska Vetenskapliga CentralraHdet.
Acknowledgements-This
REFERENCES
Arnold R. R., Russell J. E., Devine S. M., Adamson M. and Pruitt K. M. (1984) Antimicrobial activity of the secretory innate defense factors lactoferrin, lactoperoxidase and lysozyme. In: Carioiogy Today (Edited by Guggenheim B.) pp. 75-88. Karger, Base]. Betts R. H. and Dainton F. S. (1953) Electron transfer and other processes involved in the spontaneous bleaching of acidified solutions of ferric thiocyanate. J. Am. Chem. Sot. 75, 5721-5727.
Cao C. F. and Smith Q. T. (1989) Crevicular fluid myeloperoxidase at healthy, gingivitis and periodontitis sites. J. ciin. Periodont. 16, 17-20. Carlsson J., Iwami Y. and Yamada, T. (1983) Hydrogen peroxide excretion by oral streptococci and effect of lactoperoxidase-thiocyanate-hydrogen peroxide. Z&t. Immun. 40, 70-80.
Donoghue H. D., Hudson D. E. and Perrons C. J. (1987) Effect of the lactoperoxidase system on streptococcal acid production and growth. J. dent. Res. 66, 616618. Germaine G. R. and Tellefson L. M. (1981) Effect of human saliva on glucose uptake by Streptococcus mutans and other oral microorganisms. Inject. Immun. 31, 598-607. Germaine G. R. and Tellefson L. M. (1982) Glucose uptake by Streptococcus mutans, Streptococcus mitis, and Actinomyces viscosus in the presence of human saliva. Inject. Immun. 38, 1060-1067. GrHhn E., Tenovuo J., Lehtonen O-P., Eerola E. and Vilja P. (I 988) Antimicrobial systems of human whole saliva in relation to dental caries, cariogenic bacteria, and gingival inflammation in young adults. Acta odonr. stand. 46, 67-74.
van der Hoeven J. S., de Jong M. H. and van Nieuw Amerongen A. (1989) Growth of oral microflora on saliva from different glands. Microbial Ecol. Hlth Dis. 2, 171-180. Hoogendoorn H. (1985) Activation of the salivary peroxidase antimicrobial system: clinical studies. In: The Lactoperoxidase System : Chemistry and Clinical Significance (Edited by Pruitt K. M. and Tenovuo J.)
pp. 217-227. Marcel Dekker, New York. de Jong M. H., van der Hoeven J. S., van OS J. H. and Olijve J. H. (1984) Growth of oral Streptococcus species and Actinomyces viscosus in human saliva. Appl. Environ. Microbial.
47, 901-904.
Kersten H. W., Moorer W. R. and Wever R. (1981) Thiocyanate as a cofactor in myeloperoxidase activity against Streptococcus mutans. J. dent. Res. 60, 831-837. Lamberts B. L., Pruitt K. M., Pederson E. D. and Golding M. P. (1984) Comparison of salivary peroxidase system components in caries-free and caries-active naval recruits. Caries Res. 18, 488494.
159
Lehtonen O-P., GrAhn E., Stihlberg T. and Laitinen L. A. (1984) Amount and avidity of salivary and serum antibodies against Streptococcus mutans in two groups of human subjects with different dental caries susceptibility. Infect. Immun. 43, 308-313.
Mandel I. D., Behrman J., Levy R. and Weinstein D. (1983) The salivary lactoperoxidase system in caries-resistant and -susceptible adults. J. dent. Res. 62, 922-925. Mansson-Rahemtulla B., Pruitt K. M., Tenovuo J. and Adamson M. (1982) Interaction of oral streptococci with the salivary peroxidase system. In: Basic Concepts of Streptococci and Streptococcal Diseases (Edited by Holm S. E. and Christensen P.) pp. 118-119. Reedbooks, Chertsev. Ennland. Mansson-Rahet%ulla B., Pruitt K. M., Tenovuo J. and Le T. (1983) A mouthrinse which optimizes in oivo generation of hypothiocyanite. J. dent. Res. 62, 1062-1066. Mansson-Rahemtulla B., Baldone D. C., Pruitt K. M. and Rahemtulla F. (1986) Specific assays for peroxidases in human saliva. Archs oral Biol. 31, 661-668. Mansson-Rahemtulla B., Baldone D. C., Pruitt K. M. and Rahemtulla F. (1987) Effects of variations in DH and hypothiocyanite ~concentrations on S. mutans -glucose metabolism. J. dent. Res. 66, 486-491. Mansson-Rahemtulla B., Rahemtulla F., Baldone D. C., Pruitt K. M. and Hjerpe A. (1988) Purification and characterization of human salivary peroxidase. Biochemistry 27, 233-239.
Moldoveanu Z, Tenovuo J., Pruitt K. M., ManssonRahemtulla B. and Mestecky J. (1983) Antibacterial properties of milk: IgA-peroxidase-lactoferrin interactions. Ann. N. Y. Acad. Sci. 489, 848-850. Pruitt K. M. and Reiter B. (1985) Biochemistry of peroxidase system: antimicrobial effects. In: The Lactoperoxidase
System
: Chemistry and Biological Significance
(Edited bv Pruitt K. M. and Tenovuo J.), DD. . . 143-178. Marcel dekker, New York. Pruitt K. M., Tenovuo J., Andrews R. W. and McKane T. (1982) Lactoperoxidase-catalyzed oxidation of thiocyanate: polarographic study of the oxidation products. Biochemistry
21, 562-567.
Pruitt K. M, Mansson-Rahemtulla B. and Tenovuo J. (1983) Detection of the hvpothiocvanite (OSCN-) ion in human parotid saliva an; the et&t of pH on GSCNgeneration in the salivary peroxidase antimicrobial system. Archs oral Biol. 28, 517-525. Pruitt K. M., Mansson-Rahemtulla B., Baldone D. C. and Rahemtulla B. (1988) Steady-state kinetics of thiocyanate oxidation catalyzed by human salivary peroxidase. Biochemistry
27, 240-245.
Samaranayake L. P., Hughes A., Wectman D. A. and MacFarlane T. W. (1986) Growth and acid production of Candida species in human saliva supplemented with glucose. J. oral Path. 15, 251-254. Tenovuo J., Mansson-Rahemtulla B., Pruitt K. M. and Arnold R. R. (1981) Inhibition of dental plaque acid production by the salivary lactoperoxidase antimicrobial system. Infect. Immun. 34, 208-214. Tenovuo J., Pruitt K. M. and Thomas E. L. (1982a) Peroxidase antimicrobial system of human saliva: hypothiocyanite levels in resting and stimulated saliva. J. dent. Res. 61, 982-985.
Tenovuo J., Moldoveanu Z., Mestecky J., Pruitt K. M. and Mansson-Rahemtulla B. (1982b) Interaction of snecific and innate factors of immunity: IgA enhan& the antimicrobial effect of the lactoperoxidase system against Streptococcus mutans. J. Immun. 128, 726-731. Tenovuo J., Anttila O., Lumikari M. and Sievers G. (1988) Antibacterial effect of myeloperoxidase against Streptococcus mutans. Oral Microbial.
Immun. 3, 68-71.
Thomas E. L. (1981) Lactoperoxidase-catalyzed oxidation of thiocyanate: the equilibria between oxidized forms of thiocyanate. Biochemistry 20, 3273-3280.
160
M. LLJMIKARI er al.
Thomas E. L., Pera K. A., Smith K. W. and Chwang A. K. (1983) Inhibition of Streprococcus mutans by the lactoperoxidase antimicrobial system. Infecr. Immun. 39, 761-718.
Vilja P., Krohn K. and Tuohimaa P. (1985) A rapid and sensitive non-competitive avidin-biotin assay for lactofertin. J. Immun. Meth. 76, 73-83.
Wever R., Kast W. M., Kasinoedin J. H. and Boylens R. (1982) The peroxidation of thiocyanate catalysed by myeloperoxidase and lactoperoxidase. Biochim. Biophys. Acra 709, 212-219. Williams N. B. and Powlen D. 0. (1959) Human parotid saliva as a sole source of nutrient for oral micro-organisms. Archs oral Biol. 1, 4861.
Copyright
0
0003-9969/91 S3.00 + 0.00 1991 PergamonPressplc
INHIBITION OF THE GROWTH OF STREPTOCOCCUS MUTANS, STREPTOCOCCUS SOBRINUS AND LACTOBACILLUS CASEI BY ORAL PEROXIDASE SYSTEMS IN HUMAN SALIVA M. LUMIKARI,T. SOUKKA,S. NUEMIO and J. TENOVUO* Department of Cariology, Institute of Dentistry, University of Turku, LemminkHisenkatu 2, SF-20520 Turku, Finland (Accepted 22 August 1990) Summary-Streptococcus murans, Strep. sobrinus and L.actobacillus casei were grown in glucose-supplemented, sterilized, human whole saliva, adjusted to pH 5, 6 or 7. Components of the antibacterial peroxidase system-hypothiocyanous acid (HOSCN) and hypothiocyanite ions (OSCN-)-were generated by adding exogenous H,O, to sterilized saliva containing endogenous peroxidases and thiocyanate (SCN-) ions. HOSCN/OSCN- generation was proportional to the amount of H,O, added, and more HOSCN/OSCN- was detected in saliva at pH 7 than at pH 5. However, the growth of mutans streptococci and L. casei was inhibited at pH 5 by HOSCN/OSCN-, whereas no inhibition was found at pH 7. The findings show that (a) sufficient amount of HOSCN/OSCN- will inhibit the growth of cariogenic bacteria in human saliva at pH 5; (b) this amount of HOSCN/OSCN- can be generated in saliva by exogenously added H,O,; and (c) peroxidase systems have stronger antistreptococcal effects in human whole saliva than in phosphate buffer.
Key words: cariogenic bacteria, peroxidases, antimicrobial, whole saliva, streptococci, hypothiocyanite.
INTRODUCTION Human whole saliva contains two peroxidase enzymes which are considered important mucosal defence factors. Major salivary glands secrete salivary peroxidase (Mansson-Rahemtulla et al., 1988), which is structurally somewhat different but catalytically quite similar to bovine milk lactoperoxidase (Pruitt et al., 1988). Oral polymorphonuclear leucocytes release myeloperoxidase into gingival crevicular fluid and whole saliva (Mansson-Rahemtulla et al., 1986; Cao and Smith, 1989), in amounts proportional to the degree of gingival inflammation (Cao and Smith, 1989). Both salivary peroxidase and myeloperoxidase catalyse the oxidation of thiocyanate (SCN-) ions by hydrogen peroxide to form hypothiocyanite ions (OSCN-) or hypothiocyanous acid (HOSCN), which are the actual antimicrobial agents of oral peroxidase systems (Pruitt and Reiter, 1985). Most studies on the antibacterial activity of oral peroxidases have been done in vitro, using purified lactoperoxidase as the oxidizing enzyme and synthetic media for pure cultures of oral bacteria (Pruitt and Reiter, 1985). Under these circumstances, glucose-stimulated metabolic events (glucose and oxygen uptake, acid production) of Streptococcus mutans (Germaine and Tellefson, 1981; Carlsson, Iwami and Yamada, 1983; Thomas et of.; 1983) and *To whom correspondence should be addressed. Abbreviations: BHI, brain-heart infusion broth; DTT, dithiothreitol: NBS. 5-thio-2-nitrobenzoic acid: NBS,. 5,5-dithiobis-(2-nitrobenzoic acid); PBS, phosphatk buffered saline. AOB 36,2--D
Strep. rattus (Donoghue, Hudson and Perrons, 1987) are easily inhibited. Similar anti-streptococcal bacteriostatic inhibition has been observed with purified human salivary peroxidase (Mansson-Rahemtulla et al., 1987) and myeloperoxidase (Tenovuo et al., 1988), suggesting that the peroxidase systems may also affect the in viuo growth of oral streptococci. In order to study the antimicrobial action of peroxidase systems in saliva itself, without synthetic media and added lactoperoxidase, we have now grown Strep. mutans, Strep. sobrinus and Lactobacilhs casei in sterilized human whole saliva with variations in pH and HOSCN/OSCNconcentrations. With this procedure, we expected to get more detailed information than before as to how peroxidase systems affect the growth of cariogenic bacteria in vivo. Glucose-supplemented sterilized saliva is, however, only a model system which is not identical with the in vivo situation where there are also bacterial interactions, and where bacteria obtain nutrients by, for example, degrading salivary glycoproteins (Williams and Powlen, 1959; van der Hoeven, de Jong and van Nieuw Amerongen, 1989).
MATERIALSAND METHODS Bacterial strains and cell preparations Strep. mutans ATCC 25 175, Strep. sobrinus ATCC 33478 and L. casei ATCC 393, were used. Strep. mutans and Strep. sobrinus were grown in BHI (Oxoid Ltd, Basingstoke, England) and L. casei in Lactobacilli broth AOAC (Difco Labs, Detroit, MI, U.S.A.) at 37°C overnight. The cells were harvested 155
M. LUMIKARI et al.
156
by centrifugation (MSE Super Minor Centrifuge, Fisons, England) at 1800 g for 10 min, washed twice with PBS (Orion Diagnostica, Espoo, Finland) and resuspended in ice-chilled PBS (AeZO= 0.8). Collection and treatment of saliva
Paraffin-stimulated whole saliva (5&60 ml for each experiment) was collected from one healthy, nonsmoking, male donor in the morning before each experiment. He did not eat or drink anything 1 h prior to the collection. The saliva was sterilized by centrifugation at 40,OOOg (MSE Hi-Spin 21, Fisons, England), for 30min at 4°C and by a subsequent filtration of the supernatant through Millex@-GS filters (0.45 p, Millipore S.A., Molsheim, France). The desired pH for the different experiments was adjusted by adding sterilized lactic acid (Fluka AG, Buchs, Switzerland) to the saliva. Fresh saliva was used for all experiments. Chemical assays
The following factors were analysed before and after sterilizing the saliva samples. Lysozyme activity was estimated with Micrococcus diffusion plates (Lysozyme Kit@, Kallestad Laboratories, Chaska, MN, U.S.A.), using lyophilized human urine lysozyme as a standard. The lactoferrin concentration was analysed by a non-competitive avidin-biotin enzyme immunoassay (Vilja, Krohn and Tuohimaa, as a 1985), using human colostral lactoferrin reference. Total concentrations of salivary IgA and IgG were determined by a ‘trapping antibody’-type enzyme immunoassay (Lehtonen et al., 1984). The changes in salivary peroxidase, thiocyanate (SCN- ) and hypothiocyanite (OSCN- ) were followed before and during the experiments as described below. Salivary peroxidase activity was measured by following the rate of oxidation of NBS to (NBS),, by the generation of OSCN- ions during the oxidation of SCN- by salivary peroxidase and H,Oz (Wever et al., 1982). Peroxidase activity was expressed as milliunits (mu), according to the calculation described by Mansson-Rahemtulla et al. (1986). Hypothiocyanite ions were assayed by reaction of the coloured anionic monomer of (NBS), (Tenovuo, Pruitt and Thomas, 1982a) and SCNions were quantitated by the ferric nitrate method of Betts and Dainton (1953). Experimental
procedures
In every experiment the initial salivary concentration of HOSCN/OSCNwas determined. After centrifugation the levels of SCN- and peroxidase were measured and when the saliva sample was sterilized and the desired pH adjusted (pH 5, 6 or 7), the amounts of HOSCN/OSCNand peroxidase were again measured. To sterile tubes containing 2.5 ml of sterilized saliva, 25 ~1 of hydrogen peroxide (E. Merck AG, Darmstadt, Germany) were added to a final concentration of 50 p M, and the concomitant generation of OSCN- was followed. A portion (50 ~1) of the cell suspension of Strep. mutans, Strep. sobrinus or L. casei was added together with 25 ~1 of 1 M glucose (BDH Chemicals Ltd, Poole, England) (final
concentration 1OmM). Control tubes contained no H,O,. In every experiment a sterility control was present; tubes with only saliva and glucose but no added bacteria or H, OZ. The tubes were incubated aerobically on a shaking water bath, at 37°C for 20 h and the growth was followed spectrophotometrically (Hitachi model 101, Hitachi Ltd, Japan) at 620nm, before and after the 20-h incubation period. After the incubation, saliva-Strep. mutans or saliva-Strep. sobrinus mixtures were sonicated with an ultrasonic disintegrator (100 W, Measuring & Scientific Equipment Ltd, London, England) for 1 min at an amplitude of 7 pm. Bacterial mixtures were then transferred to Mitis Salivarius agar plates, supplemented with 20% glucose and 0.5 pg/ml bacitracin (Sigma Chemical Co., St Louis, MO, U.S.A.). The plates were incubated for 3 days, at 37”C, in candle jars. Saliva-L. casei mixtures were transferred to Rogosa SL agar plates (Difco Lab, Detroit, MI, U.S.A.) and incubated anaerobically for 3 days at 37°C. All saliva-bacteria mixtures were also cultivated on blood agar plates (Oxoid, 5% bovine blood) and incubated for 2 days in a CO,-atmosphere at 37°C. RESULTS
The sterilization of whole saliva by centrifugation and ultrafiltration, followed by pH adjustment, almost entirely removed HOSCN/OSCNfrom saliva samples, whereas SCN- levels remained practically unaffected (Table 1). The activity of salivary peroxidase and myeloperoxidase decreased in various experiments by IO-30%, but was always > 0.3 mU in the final, sterilized sample. Small concentrations of lactoferrin and lysozyme were also found in sterilized samples (Table 1). HOSCN/OSCNcould be easily regenerated in sterilized whole saliva by adding small amounts of H,O, (Fig. IA). The generation of HOSCN/OSCNwas directly proportional to the added concentration of H,O, both in saliva and in phosphate buffer (Fig. 1). There was a trend for slightly more HOSCN/OSCN- to be formed in saliva at pH 7 than at pH 5 (Fig. IA) and for HOSCN/OSCNto decompose faster at lower pH (Fig. 2). In preliminary experiments at pH 6.0 the growth of Strep. mutans was inhibited by HOSCN/OSCNin a dose-dependent way. Based on these experiments and the known in vivo levels of HOSCN/OSCN(Tenovuo et al., Table 1. Some salivary defence factors in paraffin-stimulated whole saliva before and after sterilization of the samples Agent
N’
Fresh sample
Sterilized sample
Salivary pcroxidase (mu) Thiocyanate (mM) Hypothiocyanite bM) Lactoferrin @g/ml) Lysozyme @g/ml) Total IgA (mg/l) Total IgG (mg/l)
8 9 10 3 13 7 7
0.88 f 0.26’ l.lOkO.39 18.5 f 8.6 4.17 f 0.74 9.03 f 1.33 41.0 * 20.5 11.5k4.8
0.66 f 0.28 1.11 kO.38 1.31 f 2.17 1.27 f 0.21 1.64 f 1.08 36.2 * 17.9 8.98 f 3.67
’ Number of experiments. 2Mean + SD.
Peroxidases and cariogenic bacteria
0
157
3
6
12
20
TIME Iht
0
50 100 H202 lpMI
25
250
Fig. 1. Generation of HOSCN/OSCN- in sterilized human whole saliva (A) and in 0.1 M phosphate buffer (B), both adjusted to pH 5 (m), 6 (m) or 7 (O), after addition of the indicated concentrations of H,O,. Phosphate buffer was supplemented also with lactoperoxidase (5 pg/ml) and SCN- (1 mM).
1982a), a concentration of 50 PM of H,O, was for further experiments at various pH values. The growth curve of Strep. mutans in glucosesupplemented, sterilized saliva (adjusted to pH 5.0), with and without HOSCN/OSCNis shown in Fig. 3. The number of cells decreased rapidly after addition into the saliva, followed by a relatively linear growth up to 20 h, except when HOSCN/OSCN(40-45 PM) was present. The peroxidase systems blocked the streptococcal growth completely in a few hours. Based on these time-course experiments, a 20-h growth period was selected for further studies. In some experiments we plated the bacteria 48 h after the generation of HOSCN/OSCNand still no growth was observed. Growth and HOSCN/OSCN--mediated inhibition of the bacteria were both strongly dependent on pH. chosen
Fig. 3. Growth curve of Strep. mutansin sterilized human whole saliva (adjusted to pH 5.0) with (a---0) and without (O--O) HOSCN/OSCN- (4045 PM). Before aerobic incubation of saliva or saliva-HOSCN/OSCNmixtures at 37°C the tubes were supplemented with 10 mM glucose and 50~1 of Strep. mutanscell suspension.
At pH 5.0 the growth of Strep. sobrinus (control cells) was less abundant than that of Strep. mutans (p < O.OOl), and in all experiments at this low pH HOSCN/OSCNcompletely inhibited the growth of Strep. sobrinus but not of Strep. mutans (Fig. 4). At pH 6 and 7, only slight and statistically not significant inhibition of streptococcal growth was found with HOSCN/OSCN(Fig. 4). In all these experiments, no growth at all was observed in the purity controls, i.e. in the samples of sterilized saliva with no added bacteria. The results were similar whether counted on Mitis Salivarius or blood agar plates. Also, the addition of H,O, (50 PM) without peroxidase and/or SCN- ions into BHI did not inhibit the growth of mutans streptococci at pH 5 (data not shown), suggesting that the growth inhibition was not due to H,Or but rather to HOSCN/OSCN-. Incorporation
Strep. z \
sobrinus
ao
24.0 0 g
s= 10
20
SALIVA
7
I L. cord 6.0 6D 4.0
20
ma
5n
‘0
-
BUFFER
0 0
2
4
6 6 TIME(min)
10
12
Fig. 2. Decomposition of HOSCN/OSCNin sterilized human whole saliva and in 0.1 M phosphate buffer at pH 5, 6 or 7. The concentration of added H,O, was 50/1M.
60
10
PI4
Fig. 4. Growth of Strep. mutans,Strep.sobrinusand L.casei in sterilized, glucose-supplemented human whole saliva at pH 5, 6 or 7. The saliva-bacteria mixtures were supplemented with 50pM HrO, (hatched columns), which produced 40-45 /IM HOSCN/OSCN-. The controls (open columns) contained neither H,O, nor HOSCN/OSCN-. The values are mean log counts f SD from 8-10 different experiments. *** =p -z 0.001 (Student’s r-test).
M. LUMIKARI et al.
158
of a reducing agent, DTT (final concentration 1 mM), into the reaction mixture completely abolished the growth inhibitory effect of HOSCN/OSCN- at pH 5. The addition of lactoperoxidase (5 pg/ml) together with H,O, into the saliva did not enhance the growth inhibition when compared to the addition of H,Oz alone. In contrast to the mutans streptococci, at pH 5 a smaller but still highly significant (p < 0.001) growth inhibition of L. casei by HOSCN/OSCNwas observed (Fig. 4). No inhibition occurred at pH 6 and 7. At all pH values, the growth of L. casei in glucose-supplemented human saliva was more abundant (p < 0.001) than that of mutans streptococci (Fig. 4). Possible changes in the pH of the saliva medium during incubation were also measured. In experiments with initial pH 5, HOSCN/OSCNgrowth inhibition kept the pH values stable (range &-0.2 pH units) over the entire 20-h incubation whereas the pH in the control tubes slightly decreased (range 0.2-0.5 pH units). DISCUSSION
Various oral microorganisms, including oral streptococci (de Jong e? al., 1984; van der Hoeven ef al., 1989), can grow in human saliva (Williams and Powlen, 1959; Samaranayake er al., 1986). Streptococcal cell production in saliva is, however, carbohydrate limited but, with appropriate glucose supplementation, doubling times of 1.6-3 h have been reported for oral streptococci (de Jong et al., 1984). We followed the growth in saliva by measuring the turbidity (A,,,) and by plating bacterial samples on solid media. With both methods there was some variation between different experiments, perhaps due to the aggregation of bacterial cells. When the same bacterial culture was tested in saliva samples from 5 persons, there was some inter-individual variation in the growth rates but, in all cases, the growth was inhibited at pH 5 by HOSCN/OSCN(data not shown). In order to minimize these sources of variation, in all experiments we used whole saliva from the same donor and, before growth measurements, ultrasonic dispersion to disaggregate the cells. Because AeZOreadings and plate counts gave qualitatively similar results, for simplicity we have here reported only the bacterial plate counts. Studies with centrifuged human saliva have shown that glucose uptake of saliva-suspended Strep. rattus (Strep. mutans BHT, serotype b) could be completely abolished at neutral pH when supplemented with H202 (Germaine and Tellefson, 1981, 1982). Glucosestimulated acid production in plaque-saliva mixtures (Tenovuo ef al., 1981), and in centrifuged saliva supplemented with Strep. mutans (ManssonRahemtulla et al., 1982), was also inhibited by HOSCN/OSCNin a dose-dependent way. Our results now extend these findings by showing that HOSCN/OSCNcan also prevent the growth of Strep. mutans, Strep. sobrinus and L. casei in human whole saliva. The inhibition was strongest at low pH and no inhibition at all was found at neutral pH. These findings agree with studies showing that the antimetabolic effects of the lactoperoxidase/salivary
peroxidase-myeloperoxidase SCN--H,O, systems are stronger at low pH than at neutral pH (Kersten, Moorer and Wever, 1981; Thomas et al., 1983; Mansson-Rahemtulla er al., 1987; Tenovuo et al. 1988). At low pH, the major oxidation product is HOSCN (Thomas, 1981), which is soluble in organic solvents and thus possibly penetrates the lipophilic bacterial cell envelopes more easily than the OSCNions, which dominate at neutral pH (Thomas, 1981). As expected, a reducing agent (DTT) completely abolished the growth inhibition, confirming that the HOSCN/OSCNpair was indeed responsible for it. Of the three bacterial strains, L. cnsei was most resistant to HOSCN/OSCNmediated inhibition. Although at pH 5 Strep. sobrinus was totally inhibited by HOSCNJOSCN- unlike Strep. rnutuns, this difference may be due to the slower overall growth rate of Strep. sobrinus than of Strep. mutans in saliva at low pH. Also, possible differences in endogenous H,O, production by the bacteria could have affected their susceptibility to inhibition (Pruitt and Reiter, 1985). Interestingly, with the same Strep. mutuns strain as now used, we could not find any loss of viability by the lactoperoxidase-SCN--H,O, (HOSCN/ OSCN- > 200 PM) system, even at pH 4.5, if the cells were exposed to HOSCN/OSCNin PBS for 90 min (Tenovuo et al., 1988). However, streptococci the now exposed to salivary peroxidase/ myeloperoxidase-SCN--H,O, system in saliva at low pH did not recover, although the HOSCN/OSCNconcentration was as low as 45-50pM. In both studies, H,O, was added to the incubation medium (PBS or saliva) 1 min before the addition of bacteria. Therefore, it is unlikely that any short-lived, powerful oxidation products (Tenovuo et al., 1981; Pruitt et al., 1982) could have influenced the results. The Strep. mutans strain used in these experiments is able to produce some extracellular H,Oz in the presence of glucose (Carlsson et al., 1983). This bacterial HzOz may further enhance the antibacterial activity of peroxidases (Carlsson et al., 1983; Thomas et al., 1983), but in the present study it was not enough to inhibit the cell division and therefore exogenous H,O, was added. Thus, our results suggest that in human saliva the antimicrobial effects of oral peroxidase systems may be stronger than many in vitro studies have indicated. One possibility is that some other salivary constituents enhance the antimicrobial action of peroxidases synergistically. Such interactions have been observed in vitro, e.g. secretory IgA, lactoferrin and lysozyme can enhance the antimetabolic activity of the lactoperoxidase-SCN--H,02 system (Tenovuo et al., 1982b; Moldoveanu et al., 1983; Arnold et al., 1984). Because sterilized saliva contains IgA, lactoferrin and some lysozyme, these interactions with oral peroxidase systems could have contributed to the inhibitory effects. Salivary HOSCN/OSCN- levels can be elevated in vitro and also in vivo by appropriate additions of H,Oz (Mansson-Rahemtulla er al., 1982, 1983; Pruitt, Mansson-Rahemtulia and Tenovuo, 1983). HOSCN/OSCNlevels up to 150-200 PM are easily generated in vivo at pH 5.5 (Mansson-Rahemtulla et 1983). According to our present study, al., HOSCN/OSCNin human saliva at low pH has the
Peroxidases and cariogenic bacteria potential to inhibit significantly, presumably by bactericidal activity, those bacteria which are cariogenic in man. Although clinical studies have failed to establish a correlation between salivary HOSCN/OSCN- levels and dental caries (Mandel er al., 1983; Lamberts et al., 1984; Grihn et al., 1988), activation of peroxidase systems in uiuo by appropriate amounts of exogenous H,O, may be of clinical importance (Hoogendoom, 1985). Further studies of the synergism between antimicrobial systems in human mouth may also reveal clinically applicable methods to combat pathogenic bacteria. investigation was supported by the Academy of Finland, the Finnish Dental Society, Emil Aaltonen Foundation, Niilo Helander Foundation and Svenska Vetenskapliga CentralraHdet.
Acknowledgements-This
REFERENCES
Arnold R. R., Russell J. E., Devine S. M., Adamson M. and Pruitt K. M. (1984) Antimicrobial activity of the secretory innate defense factors lactoferrin, lactoperoxidase and lysozyme. In: Carioiogy Today (Edited by Guggenheim B.) pp. 75-88. Karger, Base]. Betts R. H. and Dainton F. S. (1953) Electron transfer and other processes involved in the spontaneous bleaching of acidified solutions of ferric thiocyanate. J. Am. Chem. Sot. 75, 5721-5727.
Cao C. F. and Smith Q. T. (1989) Crevicular fluid myeloperoxidase at healthy, gingivitis and periodontitis sites. J. ciin. Periodont. 16, 17-20. Carlsson J., Iwami Y. and Yamada, T. (1983) Hydrogen peroxide excretion by oral streptococci and effect of lactoperoxidase-thiocyanate-hydrogen peroxide. Z&t. Immun. 40, 70-80.
Donoghue H. D., Hudson D. E. and Perrons C. J. (1987) Effect of the lactoperoxidase system on streptococcal acid production and growth. J. dent. Res. 66, 616618. Germaine G. R. and Tellefson L. M. (1981) Effect of human saliva on glucose uptake by Streptococcus mutans and other oral microorganisms. Inject. Immun. 31, 598-607. Germaine G. R. and Tellefson L. M. (1982) Glucose uptake by Streptococcus mutans, Streptococcus mitis, and Actinomyces viscosus in the presence of human saliva. Inject. Immun. 38, 1060-1067. GrHhn E., Tenovuo J., Lehtonen O-P., Eerola E. and Vilja P. (I 988) Antimicrobial systems of human whole saliva in relation to dental caries, cariogenic bacteria, and gingival inflammation in young adults. Acta odonr. stand. 46, 67-74.
van der Hoeven J. S., de Jong M. H. and van Nieuw Amerongen A. (1989) Growth of oral microflora on saliva from different glands. Microbial Ecol. Hlth Dis. 2, 171-180. Hoogendoorn H. (1985) Activation of the salivary peroxidase antimicrobial system: clinical studies. In: The Lactoperoxidase System : Chemistry and Clinical Significance (Edited by Pruitt K. M. and Tenovuo J.)
pp. 217-227. Marcel Dekker, New York. de Jong M. H., van der Hoeven J. S., van OS J. H. and Olijve J. H. (1984) Growth of oral Streptococcus species and Actinomyces viscosus in human saliva. Appl. Environ. Microbial.
47, 901-904.
Kersten H. W., Moorer W. R. and Wever R. (1981) Thiocyanate as a cofactor in myeloperoxidase activity against Streptococcus mutans. J. dent. Res. 60, 831-837. Lamberts B. L., Pruitt K. M., Pederson E. D. and Golding M. P. (1984) Comparison of salivary peroxidase system components in caries-free and caries-active naval recruits. Caries Res. 18, 488494.
159
Lehtonen O-P., GrAhn E., Stihlberg T. and Laitinen L. A. (1984) Amount and avidity of salivary and serum antibodies against Streptococcus mutans in two groups of human subjects with different dental caries susceptibility. Infect. Immun. 43, 308-313.
Mandel I. D., Behrman J., Levy R. and Weinstein D. (1983) The salivary lactoperoxidase system in caries-resistant and -susceptible adults. J. dent. Res. 62, 922-925. Mansson-Rahemtulla B., Pruitt K. M., Tenovuo J. and Adamson M. (1982) Interaction of oral streptococci with the salivary peroxidase system. In: Basic Concepts of Streptococci and Streptococcal Diseases (Edited by Holm S. E. and Christensen P.) pp. 118-119. Reedbooks, Chertsev. Ennland. Mansson-Rahet%ulla B., Pruitt K. M., Tenovuo J. and Le T. (1983) A mouthrinse which optimizes in oivo generation of hypothiocyanite. J. dent. Res. 62, 1062-1066. Mansson-Rahemtulla B., Baldone D. C., Pruitt K. M. and Rahemtulla F. (1986) Specific assays for peroxidases in human saliva. Archs oral Biol. 31, 661-668. Mansson-Rahemtulla B., Baldone D. C., Pruitt K. M. and Rahemtulla F. (1987) Effects of variations in DH and hypothiocyanite ~concentrations on S. mutans -glucose metabolism. J. dent. Res. 66, 486-491. Mansson-Rahemtulla B., Rahemtulla F., Baldone D. C., Pruitt K. M. and Hjerpe A. (1988) Purification and characterization of human salivary peroxidase. Biochemistry 27, 233-239.
Moldoveanu Z, Tenovuo J., Pruitt K. M., ManssonRahemtulla B. and Mestecky J. (1983) Antibacterial properties of milk: IgA-peroxidase-lactoferrin interactions. Ann. N. Y. Acad. Sci. 489, 848-850. Pruitt K. M. and Reiter B. (1985) Biochemistry of peroxidase system: antimicrobial effects. In: The Lactoperoxidase
System
: Chemistry and Biological Significance
(Edited bv Pruitt K. M. and Tenovuo J.), DD. . . 143-178. Marcel dekker, New York. Pruitt K. M., Tenovuo J., Andrews R. W. and McKane T. (1982) Lactoperoxidase-catalyzed oxidation of thiocyanate: polarographic study of the oxidation products. Biochemistry
21, 562-567.
Pruitt K. M, Mansson-Rahemtulla B. and Tenovuo J. (1983) Detection of the hvpothiocvanite (OSCN-) ion in human parotid saliva an; the et&t of pH on GSCNgeneration in the salivary peroxidase antimicrobial system. Archs oral Biol. 28, 517-525. Pruitt K. M., Mansson-Rahemtulla B., Baldone D. C. and Rahemtulla B. (1988) Steady-state kinetics of thiocyanate oxidation catalyzed by human salivary peroxidase. Biochemistry
27, 240-245.
Samaranayake L. P., Hughes A., Wectman D. A. and MacFarlane T. W. (1986) Growth and acid production of Candida species in human saliva supplemented with glucose. J. oral Path. 15, 251-254. Tenovuo J., Mansson-Rahemtulla B., Pruitt K. M. and Arnold R. R. (1981) Inhibition of dental plaque acid production by the salivary lactoperoxidase antimicrobial system. Infect. Immun. 34, 208-214. Tenovuo J., Pruitt K. M. and Thomas E. L. (1982a) Peroxidase antimicrobial system of human saliva: hypothiocyanite levels in resting and stimulated saliva. J. dent. Res. 61, 982-985.
Tenovuo J., Moldoveanu Z., Mestecky J., Pruitt K. M. and Mansson-Rahemtulla B. (1982b) Interaction of snecific and innate factors of immunity: IgA enhan& the antimicrobial effect of the lactoperoxidase system against Streptococcus mutans. J. Immun. 128, 726-731. Tenovuo J., Anttila O., Lumikari M. and Sievers G. (1988) Antibacterial effect of myeloperoxidase against Streptococcus mutans. Oral Microbial.
Immun. 3, 68-71.
Thomas E. L. (1981) Lactoperoxidase-catalyzed oxidation of thiocyanate: the equilibria between oxidized forms of thiocyanate. Biochemistry 20, 3273-3280.
160
M. LLJMIKARI er al.
Thomas E. L., Pera K. A., Smith K. W. and Chwang A. K. (1983) Inhibition of Streprococcus mutans by the lactoperoxidase antimicrobial system. Infecr. Immun. 39, 761-718.
Vilja P., Krohn K. and Tuohimaa P. (1985) A rapid and sensitive non-competitive avidin-biotin assay for lactofertin. J. Immun. Meth. 76, 73-83.
Wever R., Kast W. M., Kasinoedin J. H. and Boylens R. (1982) The peroxidation of thiocyanate catalysed by myeloperoxidase and lactoperoxidase. Biochim. Biophys. Acra 709, 212-219. Williams N. B. and Powlen D. 0. (1959) Human parotid saliva as a sole source of nutrient for oral micro-organisms. Archs oral Biol. 1, 4861.
