Combined inhibitory effect of lactoferrin and lactoperoxidase system on the viability of Streptococcus mutans, serotype c
TERO SOUKKA, MARIANNE LUMIKARI AND JORMA TENOVUO Department of Cariology, Institute of Dentistry, University of Turku, Titrku, Finland
Soukka T, Lumikari M, Tenovuo J; Combined inhibitory effect of lactoferrin and lactoperoxidase system on the viability oi Streptococcus mulans, serotype c. ScandJ Dent Res 1991; 99: 390-6. Abstract - We have studied the effects of iron-free lactoferrin (apo LF) and lactoperoxidase .system (lactoperoxidase, LP/SCN'/HgO,), separately and together, on ihe viability o[Streptococcus mulans (serotype c) in vitro. The bacteria were incubated in buffered KCl (pH 5.5) with and without the above components which were used at concentrations normally present in human saliva. Both apo LF and LP-system had a bactericidal efTect against S. mitlans at low pH. Together they showed an additive, but not a synergistic, antibacterial eflect against S. mutans. Apo LF enhanced the LP enzyme activity but decreased the yield of the antimicrobial component, hypothiocyanitc (HOSCN/OSCN~), when incorporated into the reaction mixtures. This decrease, which was most pronounced at low pH, was due to an LP-independent reaction between apo LF and HOSCN/OSCN". Our study indicates that the LP-system and apo LF can be combined to combat oral S. mutans. Key words: lactoferrin; peroxidascs; Streptococcus mutans. J. Tenovuo, Institute of Dentistry, University of Turkti, Lemminkaisenkatu 2, SF-20520 Turku, Finland. Accepted for publication 7 November 1990.
Several antimicrobial agents, both innate and acquired, control the microbial metabolism and growth in human whole saliva (1). Although extensive studies have been done of the antimicrobial properties of many of these defense factors as isolated molecules, relatively little is known of their interactions with eaeh other. In vitro studies have shown, however, that lysozyme may bind to secretory IgA, which may direct it to specific baeterial targets (2), and hypothiocyanite
(OSCN ) ions, the antimicrobial products of the .salivary peroxidase systems (peroxidase/ SCN'/HjOj), have been proposed to enhance the antibacterial effects of lysozyme (3,4). Laetoperoxidase (LP) .system, the bovine homolog of human salivary peroxidase antimicrobial system, is potentiated by secretory IgA (5). Interaction (both synergistic and antagonistic) has also been reported between lactoferrin and sIgA (6,7). Very little is known of any interactions
LACTOFERRIN-LACTOPEROXIDASE EFFECTS ON S. MUTANS between peroxidase systems and lactoferrin. Lactoferrin, apo LF, and Fe^"^-saturated LF all have a stabilizing effect on the enzymatic activity of the lactoperoxidase molecule (5) and LF has enhanced the antimetabolic effect ofthe LP system against S. mutans (8). LASSITER et al. (9) observed that LP enzyme had an inhibitory effect on the bactericidal aetivity of LF against S. mutans but this was concluded to be a non-specific protein effect rather than an enzymatie effeet. At low eoncentrations ( < 5 mM) thioeyanate ions (SCN"), whieh are essential for the oral peroxidase systems, enhanced the killing of S. mutans by LF whereas higher concentrations were inhibitory (9). The mechanism of SCN" involvement is not known. Because of this limited knowledge of laetoferrin-peroxidase interactions, we have now studied how these defense systems inhibit the viabihty of S. mutans, serotype c, both separately and together. The antibacterial factors were used in concentrations normally found in human saliva or plaque fluid (1). S. mutans, serotype c, was chosen as a target microorganism because its response to both peroxidase systems (10-13) and LF (9) as separate antibacterial agents is already well characterized. Furthermore, S. mutans serotype c strains are highly cariogenic (14) and their inhibition by the combined action of these two oral defense systems is interesting because both systems have been incorporated in some commercial oral health products, such as toothpastes and mouthsprays. Material and methods Lactoferrin and peroxidase preparations • Lactofcrrin,
which was a kind gift from Synrma-Oleofma Company, Brussels, Belgium, was used in its iron-free form (apo LF). The purity and hotnogeneity of this preparation was analyzed with FPLC chromatography (LKB, 2138 Uvicord S, Bromtna, Sweden) and sodium dodecyl sulfate (SDS) - polyacryiamide gel electrophoresis (PhastGel, PhastSystem, Pharmacia, Sweden), using 10-15% gradient gels. With both systems only one peak/
391
band was observed. The iron-free state of apo LF was controlled by ferrozine method (15). Apo LF was found to be free also from lysozyme (Lysozyme Kit, Kallestad Laboratories, Chaska MN, USA), peroxidase (16), and IgA (17). Lactoperoxidase, purified from bovine milk, was a product of Sigma Chemical Co., St. Louis, MO, USA, and had a purity index (A.(|.)/A9go) of 0.81. Polyclonal antibodies raised against this lactoperoxidase preparation did not reveal any cross-reactions with lysozyme or lactoferrin when analyzed by immunoblotting. Potassium thioeyanate and hydrogen peroxide (30% solution) were products of E. Merck AG, Darmstadt, FRG. (Nbs), (5,5'-dithiobis (2-nitrobenzoic acid)) was purchased by Aldrich Chemical Co., Milwaukee, WI, USA. Before use it was reduced to Nbs (5-thio-2-nitrobenzoic acid) with 2-mercaptoethanol. Culture conditions and cell preparations - S. mulans
ATCC 25175, serotype c, was grown aerobically at 37°C in Brain Heart Infusion broth (BHI, Oxoid Ltd., Basingstoke, England). After 16-17 h the bacteria were in early stationary phase after which they were transferred into a fresh BHI medium and grown to an optical density of 0.3 (mid-log phase) at 620 nm (Hitachi model 101, Hitachi Ltd., Japan). The cells were then harvested by centrifugation (Martin Christ, Piccolo, Osterode/ arz) at 2000 ^ for 10 min and washed twice with ice-cold 0.05 mM KCl buffered to pH 5.5 with HCl. After centrifugation, the bacteria were suspended in the buffered KCl to give a final concentration equivalent to an optical density of 0.5-0.6 at 620 nm. Effecl of iacloferrin ctndlor peroxidase system on the
viability of S. mutans - To sterilized incubation tubes containing 800 |il of the cell suspension above, apo LF and/or LP/SCN^/H,O, system were added to a final volume of 1.0 ml. Apo LF and LP were dissolved in the buffered KCl (pH 5.5) to the concentrations desired; control tubes did not contain any apo LF or peroxidase system components. Apo LF was used in thefinalconcentrations of 10 and 50 mg/1. To getierate the antimicrobial products of the peroxidase system, HOSCN (hypothiocyanous acid) and OSCN" (hypothiocyanite ions), the following components were used: lactoperoxidase (5 mg/1), KSCN (1 mM) and HjO, (25-100 |iM to yield various amounts of HOSCN/OSCN"). The amount of HOSCN/ OSCN" was assayed from identical tubes (1 ml) in which bacterial cells were replaced by an equiv-
SOUKKA ET AL.
392
alent volume of 0.05 mM KCl. An incubation time of 75 min was chosen because ofthe previous data (9) and of our preliminary kinetic experiments: this time was needed for maximum inhibition of S. mutans by apo LF. After incubation in a shaking water bath (37°C), serial dilutions ofthe cell suspensions were plated on Mitis Salivarius (MS) agar plates supplemented with 20% glucose (BDH Chemicals Ltd., Poole, England) and 0.5 lig/ml bacitracin (Sigma Chemical Co., St. Louis, MO, USA). Before plating the samples were vortexed vigorously (each tube 20 times) to avoid possible aggregation ofthe cells. The plates were incubated anaerobically for 3 days at 37°C before counting. Prevention of aggregation of streptococci by ultrasonic treatment (10 s; MSE Ultrasonic Disintegrator, Poolc, England) slightly increased the bacterial cotints. Also cultivation oiS. mutans on blood agar plates gave somewhat higher counts than on MSB agar. However, neither sonication nor use of a non-selective medium changed the results qualitatively. Effecl of apo LF on LP enzyme activity
T o study
the possible effect of apo LF on LP enzyme activity, 5 mg/1 of LP was incubated with various concentrations (10, 25, 50, and 100 mg/1) of apo LF in 0.05 mM KCl, pH 5.5, for 75 min at 22°C. The
control tubes did not contain any apo LF. After the incubation the LP enzyme activity was determined by the Nbs-SCN" method (16). Effecl of apo LF on HOSCNjOSCN-
generation by
Ihe LP-system - The generation of HOSCN/OSCN" by LP/SCN"/H,,O2 in 0.05 mM KCl, pH 5.5, was studied in the presence and absence of various (10, 25, 50 and 100 mg/1) concentrations of apo LF. Apo LF was mixed with LP and SCN" and the reaction was started by adding 50 or 100 \lM H.,0.2 (final concentration). After an incubation of 75 min at 22°C, HOSCN/OSCN" was assayed using the Nbs assay described by AUNE & THOMAS (18), as modified by PRUITT el at. (19). To study the
possible direct interaction of HOSCN/OSCN" with apo LF, we used the following procedure. Lactoperoxidase (200 |ig) was bound to a small hydroxyapatite (Bio-Rad Laboratories, Richmond, CA, USA) column (1.0x0.5 cm), which was brought into equilibrium with 0.05 mM KCl, pH 5.5 or 7.0. LP-frec HOSCN/OSCN" was prepared by eluting the column with the above KCl solutions supplemented with KSCN (1 mM) and HjO, (200 |iM). At pH 7 and 5.5 the average recovery of HOSCN/OSCN" was 135 and 65 \iM, respectively. Ten, 25, 50, and 100 mg/1 of apo LF were then added to the enzyme-free HOSCN/
Table I F.ffect of irort-free lacloferrtn (apo LF), lactoperoxidase system (LPfSCN' jH.fi.,) and their combinations on inability (A4SB agar plates, CFUs) ofS. mutans, serolype c, after 75 min incubation in 0.05 niM KCt, pH 5.5. All agents, except H.,0., alone, inhibited the streptococci slalislically .ngnificanlly (V 50 |i.M), the inhibition was statistically stronger (/' 60 min) incubation times. We ehose these optimal conditions for our in vitro studies because in this way standardized conditions became possible, and because these conditions are suitable also for the peroxidasemediated inhibition of 5. mulans (20). As expected, apo LF inhibited signifiicantly and dose-dependently the viability of
SOUKKA ET AL.
394
20 40 60 80 apo LF (mg/l)
100
Fig. 2. Effect of iron-free lactoferrin (apo LF) on recovery of HOSCN/OSCN" after 75-min incubation in lactoperoxidase-frec KCl solution, buffered either to pH 5.5 or pH 7.0. Results are given as percentages (mcan±SD, n = 6) of controls with no apo LF.
S. mutans at low pH (4, 9, 21). Because no such bactericidal activity has been observed with Fe'"^-containing LF (22), other forms of LF except the iron-free (apo) were not included in the present study. In contrast to that of apo LF, the baetericidal effect of HOSCN/OSCN" against S. mutans at low pH is a new observation. Studies of the inhibitory action of HOSCN/ OSCN" against S. mutans, serotype c, have so far indicated only bacteriostatic, antimetabolic activity against this microorganism (10, 11, 13, 20, 23). However, when myeloperoxidase was used instead of LP at low pH together with SCN" ions and H2O2 against
S. mutans, bactericidal activity was also reported (24). We have also recently demonstrated that salivary peroxidase systems can effectively inhibit the growth of both S. mulans and S. sobrinus at low pH when these bacteria are inoculated in sterilized, glucosesupplemented human whole sahva (25). It seems that the composition ofthe incubation medium and its pH are of vital importance for the mode of inhibition (20). Peroxidase systems are mueh more effective at acidic than at neutral pH (12, 24). At low pH, the major oxidation product is the hypothioeyanous acid, HOSCN, with a pK of 5.3 (26). This means that in our experiments at pH 5.5 the uncharged form ofthe HOSCN/OSCN" pair dominates and it has been proposed to penetrate the bacterial cells walls more readily than the charged form, OSCN" (11). This is obviously the main reason why oral peroxidase systems are more effective at acidic pH than at higher pH-values. We did our experiments in KCl since CI" and K^ ions interfere very little with the bactericidal activity of apo LF, whereas phosphate, citrate, glycine and sulfate reduce effectively the bactericidal activity of apo LF (27). All those anions which can occupy the anionic binding sites on the cell surfaces of S. mutans inhibit the binding of LF and subsequent killing of these bacteria (9). As with LF, phosphate ions inhibit also the binding of LP to S. mutans (28) and may, in this way, inhibit the antibacterial activity of the LP.system by reducing the enzyme activity in the target area of HOSCN/OSCN" formation and subsequent bacterial inhibition. Indeed, we were not able to see any killing of S. mutans cells (serotype e) which were incubated in phosphate-buffered saline at pH 4.5 for 90 min in the presence ofthe complete LP-system (13). Thus high phosphate eoneentration seems to block effeetively the antistreptococcal activity of both apo LF and the LP system. The present study clearly shows that apo LF and LP-mediated antibacterial systems
LACTOFERRIN-LACTOPEROXIDASE EFFECTS ON S. MUTANS (an act together and that they have an additive, but not a synergistic, inhibitory effect against cariogenie S. mutans. This was evident with physiologic salivary concentrations of all inhibitory components (1). In accordance with our previous studies (5), LP enzyme activity was stabilized by the presenee of apo LF. However, surprisingly, apo LF inhibited in a dose-dependent way the concentration of HOSCN/OSCN" generated by the LP.system. Our present observations strongly suggest that this was due to a pH-dependent direct reaction between apo LF and HOSCN/OSCN". Interestingly the reduced amount of HOSCN/OSCN" was still high enough to kill S. mutans cells at low pH together with apo LF. In summary, although apo LF interferes with the HOSCN/OSCN" formation, peroxidase system and apo LF can be combined to produee an additive bacterieidal effeet against S. mutans.
395
PRUITT KM, MANSSON-RAHEMTULLA B . Inter-
action of specific and innate factors of immunity: IgA enhances the antimicrobial effect of the lactoperoxidase system against Streptococcus mutans. J Im?nunot 1982; 128: 726-31. 6. COLE MF, ARNOLD R , MESTECKY J , KULHAVY
R, MCGHEE J R . Studies with human lactoferrin and Streptococcus mutans. In: STILES H , LOESCHE W, O'BRIEN T , eds. Microbial aspects of
denial caries II. Washington, D.C: Information Retrieval, 1976; 359-74. 7. ROGERS HJ, SYNGE C . Bacteriostatic effect of
human milk on Escherichia coli: the role of IgA. Immunology 1978; 34: 19-24. 8. MOLDOVEANU
Z , TENOVUO J ,
PRUITT
KM,
MANSSON-RAHEMTULLA B , MESTECKY J. Anti-
bacterial properties of milk: IgA-peroxidaselactolerrin interactions. Ann .NY Acad Sci 1983; 409: 848-50. 9. LASSITER MO, NEWSOME AL, SAMS LD, A R -
NOLD RR. Characterization of lactoferrin interaction with Streptococcus mulans. J Denl Res 1987; 66: 480-5. 10. CARLSSON J, IWAMI Y, YAMADA T . Hydrogen
Acknowledgments - This study was supported by the Academy of Finland, The Finnish Dental Society and Niilo Helander Foundation.
peroxide excretion by oral streptococci and effect of lactoperoxidase-thiocyanate-hydrogen peroxide. Infecl Immun 1983; 40: 70-80.
11. THOMAS EL, PERA KA, SMITH KW, CHWANG
AK. Inhibition of Streptococcus tnutans by the lactoperoxidase antimicrobial system. Irifect /mmK« 1983; 39: 767-78.
References L TENOVUO J. Non-immunoglobulin defense factors in human saliva. In: TENOVUO J, ed. Hutnan saliva: clinical chemistry and microbiology. Vol. II. Boca Raton; FL: CRC Press, 1989; 55-9L 2. GERMAINE
G , TELLEFSON
LM. Simple
and
rapid procedure for the selective removal of lysozyme from human saliva. Infect Imniitn 1979; 26: 991-5. 3. POLLOCK JJ, GOODMAN BICKER G , KATONA LI,
CHO M I , IACONO VJ. Lysozyme bacteriolysis. In: KLEINBERG I, ELLISON SA, MANDEL ID,
eds. Saliva and dental caries. New York: Spec. Supp. Microbiology Abstracts, Information Retrieval, 1979; 429 47. 4. ARNOLD RR, RUSSELL J E , DEVINE SM, ADAM-
12. MANSSON-RAHEMTULLA B , BALDONE DC, PRU-
ITT KM, RAHEMTULLA F. Effects of variations in pH and hypothiocyanite concentrations on S. mulans glucose metabolism. J Dent Res 1987; 66:486-91. 13. TENOVUO J, ANTTILA O , LUMIKARI M , SIEVERS
G. /Antibacterial effect of myeloperoxidase against Slreptococcns mulans. Oral Microbiol Imniitvol 1988; 3: 68-71. 14. LOESCHE WJ. Role of Streptococcus mulans in human dental decay. Microbiol Rev 1986; 50: 353-80. 15. FELDKAMP C S , WATKINS R , BAGINSKI ES, ZAK
B. Es.sential serum trace metals. Determination of iron. Microchem J 1977; 22: 335-46.
SON M, PRUITT KM. Antimicrobial activity of 16. MANSSON-RAHEMTULLA the secretory innate defense factors lactoferrin, ITT KM, RAHEMTULLA lactoperoxidase and lysozyme. In: GUGGENperoxidases in human HEIM B, ed. Cariology today. Basel: Karger AG, 1986; 31: 661-8. 1984; 75-88. 17. LEHTONEN O - P , GRAHN '). TENOVUO J,
MOLDOVEANU
Z , MESTECKY J,
B , BALDONE DC, PRU-
F. Specific assays for saliva. Arch Oral Biol E , STAHLBERG T , LAI-
TiNEN LA. Amount and avidity of salivary and
SOUKKA ET AL.
396
serum antibodies against Streptococcus mutans in two groups of human subjects with different dental caries susceptibility. Infect Immun 1984; 43: 308-13. 18. AUNE T M , THOMAS EL. Accumulation of hy-
pothiocyanite ion during peroxidase-catalyzed oxidation of thioeyanate. Eur J Biochem 1977; 80: 209-14. 19. PRUITT KM, TENOVUO J, MANSSON-RAHEMTULLA B, HARRINGTON P, BALDONE DC. Is
thioeyanate peroxidation at equilibrium in vivo? Bioehim Biophys Acta 1986; 870: 385-91.
lactoperoxidase system on Streptococcus mutam and other microorganisms. In: STILES HS, LOESCHE WJ, O'BRIEN TC. Microbial aspects of
dental caries. Washington, D . C : Information Retrieval, 1976; 353-7. 24. KERSTEN H W , MOORER WR, WEVER R . Thio-
eyanate as a cofactor in myeloperoxidase activity against Streptococcus mutans. J Dent Res
1981; 60: 831-7. 25. LUMIKARI M , SOUKKA T , NURMIO S, TENOVUO
J. Inhibition of the growth of Streptococcus mutans. Streptococcus sobrinus a n d Lactobacillus
idase system: antimicrobial effects. In: PRUITT
casei by oral peroxidase systems in human saliva. Arch Oral Biol: in press.
K M , TENOVUO J , eds. The lactoperoxidase system:
26. THOMAS EL. Lactoperoxidase-catalyzed oxi-
chemistry and biological significance. N e w York:
dation of thioeyanate: equilibria between oxidized forms of thioeyanate. Biochemistry 1981; 20: 3273-80. 27. ARNOLD RR. Innate immunity and Streptococ-
20. PRUITT KM, REITER B. Biochemistry of perox-
Marcel Dekker, 1985; 143-78. 21. ARNOLD RR, BREWER M , GAUTHIER J J . Bac-
tericidal activity of Iiuman lactoferrin: sensitivity of a variety of microorganisms. Infect Immun 1980; 28: 893-8. 22. ARNOLD RR, RUSSELL J E , CHAMPION WJ, BREWER M , G.^UTHIER J J . Bactericidal activi-
ty of human lactoferrin: differentiation from the stasis of iron deprivation. Infect Immun 1982; 35: 792-9. 23. HooGENDOORN H. Thc inhibitory action ofthe
cus mutans. In: HAMADA S, MICHALEK SM, KIYONO H , MENAKER L , M C G H E E J R , eds. Molecular microbiology and immunobiology of Strep-
tococcus mutans. Amsterdam: Elsevier Science Publishers, 1986; 421-32. 28. PRUITT KM, ADAMSON M , ARNOLD R . Lacto-
peroxidase binding to streptococci. Infect Immun 1979; 25: 304-9.
TERO SOUKKA, MARIANNE LUMIKARI AND JORMA TENOVUO Department of Cariology, Institute of Dentistry, University of Turku, Titrku, Finland
Soukka T, Lumikari M, Tenovuo J; Combined inhibitory effect of lactoferrin and lactoperoxidase system on the viability oi Streptococcus mulans, serotype c. ScandJ Dent Res 1991; 99: 390-6. Abstract - We have studied the effects of iron-free lactoferrin (apo LF) and lactoperoxidase .system (lactoperoxidase, LP/SCN'/HgO,), separately and together, on ihe viability o[Streptococcus mulans (serotype c) in vitro. The bacteria were incubated in buffered KCl (pH 5.5) with and without the above components which were used at concentrations normally present in human saliva. Both apo LF and LP-system had a bactericidal efTect against S. mitlans at low pH. Together they showed an additive, but not a synergistic, antibacterial eflect against S. mutans. Apo LF enhanced the LP enzyme activity but decreased the yield of the antimicrobial component, hypothiocyanitc (HOSCN/OSCN~), when incorporated into the reaction mixtures. This decrease, which was most pronounced at low pH, was due to an LP-independent reaction between apo LF and HOSCN/OSCN". Our study indicates that the LP-system and apo LF can be combined to combat oral S. mutans. Key words: lactoferrin; peroxidascs; Streptococcus mutans. J. Tenovuo, Institute of Dentistry, University of Turkti, Lemminkaisenkatu 2, SF-20520 Turku, Finland. Accepted for publication 7 November 1990.
Several antimicrobial agents, both innate and acquired, control the microbial metabolism and growth in human whole saliva (1). Although extensive studies have been done of the antimicrobial properties of many of these defense factors as isolated molecules, relatively little is known of their interactions with eaeh other. In vitro studies have shown, however, that lysozyme may bind to secretory IgA, which may direct it to specific baeterial targets (2), and hypothiocyanite
(OSCN ) ions, the antimicrobial products of the .salivary peroxidase systems (peroxidase/ SCN'/HjOj), have been proposed to enhance the antibacterial effects of lysozyme (3,4). Laetoperoxidase (LP) .system, the bovine homolog of human salivary peroxidase antimicrobial system, is potentiated by secretory IgA (5). Interaction (both synergistic and antagonistic) has also been reported between lactoferrin and sIgA (6,7). Very little is known of any interactions
LACTOFERRIN-LACTOPEROXIDASE EFFECTS ON S. MUTANS between peroxidase systems and lactoferrin. Lactoferrin, apo LF, and Fe^"^-saturated LF all have a stabilizing effect on the enzymatic activity of the lactoperoxidase molecule (5) and LF has enhanced the antimetabolic effect ofthe LP system against S. mutans (8). LASSITER et al. (9) observed that LP enzyme had an inhibitory effect on the bactericidal aetivity of LF against S. mutans but this was concluded to be a non-specific protein effect rather than an enzymatie effeet. At low eoncentrations ( < 5 mM) thioeyanate ions (SCN"), whieh are essential for the oral peroxidase systems, enhanced the killing of S. mutans by LF whereas higher concentrations were inhibitory (9). The mechanism of SCN" involvement is not known. Because of this limited knowledge of laetoferrin-peroxidase interactions, we have now studied how these defense systems inhibit the viabihty of S. mutans, serotype c, both separately and together. The antibacterial factors were used in concentrations normally found in human saliva or plaque fluid (1). S. mutans, serotype c, was chosen as a target microorganism because its response to both peroxidase systems (10-13) and LF (9) as separate antibacterial agents is already well characterized. Furthermore, S. mutans serotype c strains are highly cariogenic (14) and their inhibition by the combined action of these two oral defense systems is interesting because both systems have been incorporated in some commercial oral health products, such as toothpastes and mouthsprays. Material and methods Lactoferrin and peroxidase preparations • Lactofcrrin,
which was a kind gift from Synrma-Oleofma Company, Brussels, Belgium, was used in its iron-free form (apo LF). The purity and hotnogeneity of this preparation was analyzed with FPLC chromatography (LKB, 2138 Uvicord S, Bromtna, Sweden) and sodium dodecyl sulfate (SDS) - polyacryiamide gel electrophoresis (PhastGel, PhastSystem, Pharmacia, Sweden), using 10-15% gradient gels. With both systems only one peak/
391
band was observed. The iron-free state of apo LF was controlled by ferrozine method (15). Apo LF was found to be free also from lysozyme (Lysozyme Kit, Kallestad Laboratories, Chaska MN, USA), peroxidase (16), and IgA (17). Lactoperoxidase, purified from bovine milk, was a product of Sigma Chemical Co., St. Louis, MO, USA, and had a purity index (A.(|.)/A9go) of 0.81. Polyclonal antibodies raised against this lactoperoxidase preparation did not reveal any cross-reactions with lysozyme or lactoferrin when analyzed by immunoblotting. Potassium thioeyanate and hydrogen peroxide (30% solution) were products of E. Merck AG, Darmstadt, FRG. (Nbs), (5,5'-dithiobis (2-nitrobenzoic acid)) was purchased by Aldrich Chemical Co., Milwaukee, WI, USA. Before use it was reduced to Nbs (5-thio-2-nitrobenzoic acid) with 2-mercaptoethanol. Culture conditions and cell preparations - S. mulans
ATCC 25175, serotype c, was grown aerobically at 37°C in Brain Heart Infusion broth (BHI, Oxoid Ltd., Basingstoke, England). After 16-17 h the bacteria were in early stationary phase after which they were transferred into a fresh BHI medium and grown to an optical density of 0.3 (mid-log phase) at 620 nm (Hitachi model 101, Hitachi Ltd., Japan). The cells were then harvested by centrifugation (Martin Christ, Piccolo, Osterode/ arz) at 2000 ^ for 10 min and washed twice with ice-cold 0.05 mM KCl buffered to pH 5.5 with HCl. After centrifugation, the bacteria were suspended in the buffered KCl to give a final concentration equivalent to an optical density of 0.5-0.6 at 620 nm. Effecl of iacloferrin ctndlor peroxidase system on the
viability of S. mutans - To sterilized incubation tubes containing 800 |il of the cell suspension above, apo LF and/or LP/SCN^/H,O, system were added to a final volume of 1.0 ml. Apo LF and LP were dissolved in the buffered KCl (pH 5.5) to the concentrations desired; control tubes did not contain any apo LF or peroxidase system components. Apo LF was used in thefinalconcentrations of 10 and 50 mg/1. To getierate the antimicrobial products of the peroxidase system, HOSCN (hypothiocyanous acid) and OSCN" (hypothiocyanite ions), the following components were used: lactoperoxidase (5 mg/1), KSCN (1 mM) and HjO, (25-100 |iM to yield various amounts of HOSCN/OSCN"). The amount of HOSCN/ OSCN" was assayed from identical tubes (1 ml) in which bacterial cells were replaced by an equiv-
SOUKKA ET AL.
392
alent volume of 0.05 mM KCl. An incubation time of 75 min was chosen because ofthe previous data (9) and of our preliminary kinetic experiments: this time was needed for maximum inhibition of S. mutans by apo LF. After incubation in a shaking water bath (37°C), serial dilutions ofthe cell suspensions were plated on Mitis Salivarius (MS) agar plates supplemented with 20% glucose (BDH Chemicals Ltd., Poole, England) and 0.5 lig/ml bacitracin (Sigma Chemical Co., St. Louis, MO, USA). Before plating the samples were vortexed vigorously (each tube 20 times) to avoid possible aggregation ofthe cells. The plates were incubated anaerobically for 3 days at 37°C before counting. Prevention of aggregation of streptococci by ultrasonic treatment (10 s; MSE Ultrasonic Disintegrator, Poolc, England) slightly increased the bacterial cotints. Also cultivation oiS. mutans on blood agar plates gave somewhat higher counts than on MSB agar. However, neither sonication nor use of a non-selective medium changed the results qualitatively. Effecl of apo LF on LP enzyme activity
T o study
the possible effect of apo LF on LP enzyme activity, 5 mg/1 of LP was incubated with various concentrations (10, 25, 50, and 100 mg/1) of apo LF in 0.05 mM KCl, pH 5.5, for 75 min at 22°C. The
control tubes did not contain any apo LF. After the incubation the LP enzyme activity was determined by the Nbs-SCN" method (16). Effecl of apo LF on HOSCNjOSCN-
generation by
Ihe LP-system - The generation of HOSCN/OSCN" by LP/SCN"/H,,O2 in 0.05 mM KCl, pH 5.5, was studied in the presence and absence of various (10, 25, 50 and 100 mg/1) concentrations of apo LF. Apo LF was mixed with LP and SCN" and the reaction was started by adding 50 or 100 \lM H.,0.2 (final concentration). After an incubation of 75 min at 22°C, HOSCN/OSCN" was assayed using the Nbs assay described by AUNE & THOMAS (18), as modified by PRUITT el at. (19). To study the
possible direct interaction of HOSCN/OSCN" with apo LF, we used the following procedure. Lactoperoxidase (200 |ig) was bound to a small hydroxyapatite (Bio-Rad Laboratories, Richmond, CA, USA) column (1.0x0.5 cm), which was brought into equilibrium with 0.05 mM KCl, pH 5.5 or 7.0. LP-frec HOSCN/OSCN" was prepared by eluting the column with the above KCl solutions supplemented with KSCN (1 mM) and HjO, (200 |iM). At pH 7 and 5.5 the average recovery of HOSCN/OSCN" was 135 and 65 \iM, respectively. Ten, 25, 50, and 100 mg/1 of apo LF were then added to the enzyme-free HOSCN/
Table I F.ffect of irort-free lacloferrtn (apo LF), lactoperoxidase system (LPfSCN' jH.fi.,) and their combinations on inability (A4SB agar plates, CFUs) ofS. mutans, serolype c, after 75 min incubation in 0.05 niM KCt, pH 5.5. All agents, except H.,0., alone, inhibited the streptococci slalislically .ngnificanlly (V 50 |i.M), the inhibition was statistically stronger (/' 60 min) incubation times. We ehose these optimal conditions for our in vitro studies because in this way standardized conditions became possible, and because these conditions are suitable also for the peroxidasemediated inhibition of 5. mulans (20). As expected, apo LF inhibited signifiicantly and dose-dependently the viability of
SOUKKA ET AL.
394
20 40 60 80 apo LF (mg/l)
100
Fig. 2. Effect of iron-free lactoferrin (apo LF) on recovery of HOSCN/OSCN" after 75-min incubation in lactoperoxidase-frec KCl solution, buffered either to pH 5.5 or pH 7.0. Results are given as percentages (mcan±SD, n = 6) of controls with no apo LF.
S. mutans at low pH (4, 9, 21). Because no such bactericidal activity has been observed with Fe'"^-containing LF (22), other forms of LF except the iron-free (apo) were not included in the present study. In contrast to that of apo LF, the baetericidal effect of HOSCN/OSCN" against S. mutans at low pH is a new observation. Studies of the inhibitory action of HOSCN/ OSCN" against S. mutans, serotype c, have so far indicated only bacteriostatic, antimetabolic activity against this microorganism (10, 11, 13, 20, 23). However, when myeloperoxidase was used instead of LP at low pH together with SCN" ions and H2O2 against
S. mutans, bactericidal activity was also reported (24). We have also recently demonstrated that salivary peroxidase systems can effectively inhibit the growth of both S. mulans and S. sobrinus at low pH when these bacteria are inoculated in sterilized, glucosesupplemented human whole sahva (25). It seems that the composition ofthe incubation medium and its pH are of vital importance for the mode of inhibition (20). Peroxidase systems are mueh more effective at acidic than at neutral pH (12, 24). At low pH, the major oxidation product is the hypothioeyanous acid, HOSCN, with a pK of 5.3 (26). This means that in our experiments at pH 5.5 the uncharged form ofthe HOSCN/OSCN" pair dominates and it has been proposed to penetrate the bacterial cells walls more readily than the charged form, OSCN" (11). This is obviously the main reason why oral peroxidase systems are more effective at acidic pH than at higher pH-values. We did our experiments in KCl since CI" and K^ ions interfere very little with the bactericidal activity of apo LF, whereas phosphate, citrate, glycine and sulfate reduce effectively the bactericidal activity of apo LF (27). All those anions which can occupy the anionic binding sites on the cell surfaces of S. mutans inhibit the binding of LF and subsequent killing of these bacteria (9). As with LF, phosphate ions inhibit also the binding of LP to S. mutans (28) and may, in this way, inhibit the antibacterial activity of the LP.system by reducing the enzyme activity in the target area of HOSCN/OSCN" formation and subsequent bacterial inhibition. Indeed, we were not able to see any killing of S. mutans cells (serotype e) which were incubated in phosphate-buffered saline at pH 4.5 for 90 min in the presence ofthe complete LP-system (13). Thus high phosphate eoneentration seems to block effeetively the antistreptococcal activity of both apo LF and the LP system. The present study clearly shows that apo LF and LP-mediated antibacterial systems
LACTOFERRIN-LACTOPEROXIDASE EFFECTS ON S. MUTANS (an act together and that they have an additive, but not a synergistic, inhibitory effect against cariogenie S. mutans. This was evident with physiologic salivary concentrations of all inhibitory components (1). In accordance with our previous studies (5), LP enzyme activity was stabilized by the presenee of apo LF. However, surprisingly, apo LF inhibited in a dose-dependent way the concentration of HOSCN/OSCN" generated by the LP.system. Our present observations strongly suggest that this was due to a pH-dependent direct reaction between apo LF and HOSCN/OSCN". Interestingly the reduced amount of HOSCN/OSCN" was still high enough to kill S. mutans cells at low pH together with apo LF. In summary, although apo LF interferes with the HOSCN/OSCN" formation, peroxidase system and apo LF can be combined to produee an additive bacterieidal effeet against S. mutans.
395
PRUITT KM, MANSSON-RAHEMTULLA B . Inter-
action of specific and innate factors of immunity: IgA enhances the antimicrobial effect of the lactoperoxidase system against Streptococcus mutans. J Im?nunot 1982; 128: 726-31. 6. COLE MF, ARNOLD R , MESTECKY J , KULHAVY
R, MCGHEE J R . Studies with human lactoferrin and Streptococcus mutans. In: STILES H , LOESCHE W, O'BRIEN T , eds. Microbial aspects of
denial caries II. Washington, D.C: Information Retrieval, 1976; 359-74. 7. ROGERS HJ, SYNGE C . Bacteriostatic effect of
human milk on Escherichia coli: the role of IgA. Immunology 1978; 34: 19-24. 8. MOLDOVEANU
Z , TENOVUO J ,
PRUITT
KM,
MANSSON-RAHEMTULLA B , MESTECKY J. Anti-
bacterial properties of milk: IgA-peroxidaselactolerrin interactions. Ann .NY Acad Sci 1983; 409: 848-50. 9. LASSITER MO, NEWSOME AL, SAMS LD, A R -
NOLD RR. Characterization of lactoferrin interaction with Streptococcus mulans. J Denl Res 1987; 66: 480-5. 10. CARLSSON J, IWAMI Y, YAMADA T . Hydrogen
Acknowledgments - This study was supported by the Academy of Finland, The Finnish Dental Society and Niilo Helander Foundation.
peroxide excretion by oral streptococci and effect of lactoperoxidase-thiocyanate-hydrogen peroxide. Infecl Immun 1983; 40: 70-80.
11. THOMAS EL, PERA KA, SMITH KW, CHWANG
AK. Inhibition of Streptococcus tnutans by the lactoperoxidase antimicrobial system. Irifect /mmK« 1983; 39: 767-78.
References L TENOVUO J. Non-immunoglobulin defense factors in human saliva. In: TENOVUO J, ed. Hutnan saliva: clinical chemistry and microbiology. Vol. II. Boca Raton; FL: CRC Press, 1989; 55-9L 2. GERMAINE
G , TELLEFSON
LM. Simple
and
rapid procedure for the selective removal of lysozyme from human saliva. Infect Imniitn 1979; 26: 991-5. 3. POLLOCK JJ, GOODMAN BICKER G , KATONA LI,
CHO M I , IACONO VJ. Lysozyme bacteriolysis. In: KLEINBERG I, ELLISON SA, MANDEL ID,
eds. Saliva and dental caries. New York: Spec. Supp. Microbiology Abstracts, Information Retrieval, 1979; 429 47. 4. ARNOLD RR, RUSSELL J E , DEVINE SM, ADAM-
12. MANSSON-RAHEMTULLA B , BALDONE DC, PRU-
ITT KM, RAHEMTULLA F. Effects of variations in pH and hypothiocyanite concentrations on S. mulans glucose metabolism. J Dent Res 1987; 66:486-91. 13. TENOVUO J, ANTTILA O , LUMIKARI M , SIEVERS
G. /Antibacterial effect of myeloperoxidase against Slreptococcns mulans. Oral Microbiol Imniitvol 1988; 3: 68-71. 14. LOESCHE WJ. Role of Streptococcus mulans in human dental decay. Microbiol Rev 1986; 50: 353-80. 15. FELDKAMP C S , WATKINS R , BAGINSKI ES, ZAK
B. Es.sential serum trace metals. Determination of iron. Microchem J 1977; 22: 335-46.
SON M, PRUITT KM. Antimicrobial activity of 16. MANSSON-RAHEMTULLA the secretory innate defense factors lactoferrin, ITT KM, RAHEMTULLA lactoperoxidase and lysozyme. In: GUGGENperoxidases in human HEIM B, ed. Cariology today. Basel: Karger AG, 1986; 31: 661-8. 1984; 75-88. 17. LEHTONEN O - P , GRAHN '). TENOVUO J,
MOLDOVEANU
Z , MESTECKY J,
B , BALDONE DC, PRU-
F. Specific assays for saliva. Arch Oral Biol E , STAHLBERG T , LAI-
TiNEN LA. Amount and avidity of salivary and
SOUKKA ET AL.
396
serum antibodies against Streptococcus mutans in two groups of human subjects with different dental caries susceptibility. Infect Immun 1984; 43: 308-13. 18. AUNE T M , THOMAS EL. Accumulation of hy-
pothiocyanite ion during peroxidase-catalyzed oxidation of thioeyanate. Eur J Biochem 1977; 80: 209-14. 19. PRUITT KM, TENOVUO J, MANSSON-RAHEMTULLA B, HARRINGTON P, BALDONE DC. Is
thioeyanate peroxidation at equilibrium in vivo? Bioehim Biophys Acta 1986; 870: 385-91.
lactoperoxidase system on Streptococcus mutam and other microorganisms. In: STILES HS, LOESCHE WJ, O'BRIEN TC. Microbial aspects of
dental caries. Washington, D . C : Information Retrieval, 1976; 353-7. 24. KERSTEN H W , MOORER WR, WEVER R . Thio-
eyanate as a cofactor in myeloperoxidase activity against Streptococcus mutans. J Dent Res
1981; 60: 831-7. 25. LUMIKARI M , SOUKKA T , NURMIO S, TENOVUO
J. Inhibition of the growth of Streptococcus mutans. Streptococcus sobrinus a n d Lactobacillus
idase system: antimicrobial effects. In: PRUITT
casei by oral peroxidase systems in human saliva. Arch Oral Biol: in press.
K M , TENOVUO J , eds. The lactoperoxidase system:
26. THOMAS EL. Lactoperoxidase-catalyzed oxi-
chemistry and biological significance. N e w York:
dation of thioeyanate: equilibria between oxidized forms of thioeyanate. Biochemistry 1981; 20: 3273-80. 27. ARNOLD RR. Innate immunity and Streptococ-
20. PRUITT KM, REITER B. Biochemistry of perox-
Marcel Dekker, 1985; 143-78. 21. ARNOLD RR, BREWER M , GAUTHIER J J . Bac-
tericidal activity of Iiuman lactoferrin: sensitivity of a variety of microorganisms. Infect Immun 1980; 28: 893-8. 22. ARNOLD RR, RUSSELL J E , CHAMPION WJ, BREWER M , G.^UTHIER J J . Bactericidal activi-
ty of human lactoferrin: differentiation from the stasis of iron deprivation. Infect Immun 1982; 35: 792-9. 23. HooGENDOORN H. Thc inhibitory action ofthe
cus mutans. In: HAMADA S, MICHALEK SM, KIYONO H , MENAKER L , M C G H E E J R , eds. Molecular microbiology and immunobiology of Strep-
tococcus mutans. Amsterdam: Elsevier Science Publishers, 1986; 421-32. 28. PRUITT KM, ADAMSON M , ARNOLD R . Lacto-
peroxidase binding to streptococci. Infect Immun 1979; 25: 304-9.
