INFECTION AND IMMUNITY, Jan. 1982, p. 20-24 Vol. 35, No. 1 0019-9567/82/010020-05$02.00/0

Lactoperoxidase and Protect Bacteria from Hydrogen Peroxide MICHAEL ADAMSON AND JAN CARLSSON* Department of Oral Microbiology, University of Umea', S-901 87 Umead, Sweden Received 22 June 1981/Accepted 2 September 1981 and thiocyanate were shown to protect and three oral streptococcal species from the bactericidal effect of hydrogen peroxide under aerobic conditions. Lactoperoxidase in the absence of thiocyanate was also protective for two of the bacterial species in a dilution medium but potentiated hydrogen peroxide toxicity for the other two under the same conditions. The products of the reaction between hydrogen peroxide and thiocyanate in the presence of lactoperoxidase were not bactericidal except in the case of E. coli, and then only under special conditions. The results demonstrate the effectiveness of lactoperoxidase and thiocyanate in protecting living cells from hydrogen peroxide toxicity. Although the effect on human cells was not examined in this study, extrapolation of these results to the cells of the oral mucosa would suggest an important protective role of lactoperoxidase and thiocyanate against the toxic effects of hydrogen peroxide in the oral cavity.

Hydrogen peroxide is capable of killing a MATERIALS AND METHODS variety of life forms, including bacteria and Microorganisms. strain mammalian cells (1, 6, 7, 12, 14, 20, 26), even NCTC 10449, S. salivarius strain NCTC 9759, S. though these cells may have highly developed sanguis strain ATCC 10556, and E. coli K-12 strain defense mechanisms against hydrogen peroxide AB1157 (3) were used as test strains. (8). For this reason, free hydrogen peroxide may Preparation of media. Trypticase-soy-yeast extract pose a threat to both human and bacterial cells (TSY) broth (pH 7.3) contained, per liter, 17 g of living in the human body. Among the putative Trypticase peptone (BBL Microbiology Systems, sites of hydrogen peroxide production in hu- Cockeysville, Md.), 3 g ofphytone peptone (BBL), 5 g mans are the surfaces of the oral cavity, where of yeast extract (Difco Laboratories, Detroit, Mich.), 5 hydrogen peroxide-forming bacteria constitute g of NaCI, 2.5 g of K2HPO4, and 2.5 g of glucose. The broth was autoclaved at 121°C for 20 min. The dilution up to 60% of the facultatively anaerobic flora solution was prepared by dissolving 4.3 g ofNaCI, 0.42 (18). Free hydrogen peroxide has not been de- g of Na2HPO4, 1.0 g of K2HPO4, and 10 g of sodium 1B- tected in the saliva, however (18), presumably glycerophosphate * 5H20 in 500 ml of water, pH 7.0 because it is quickly destroyed by the salivary (solution 1), and 0.24 g of CaCl2 and 0.1 g of enzymes lactoperoxidase and thiocyanate. MgC92 6H2O in 500 ml of water (solution 2) (25). Hydrogen peroxide is a substrate of lactoper- Solutions 1 and 2 were autoclaved separately, oxidase in oxidizing thiocyanate to hypothio- cooled, and pooled. Peptone-yeast extract-glucose cyanate (OSCN) (2, 15, 16). This product of the (PYG) agar medium was prepared from two solutions. reaction oxidizes bacterial sulfhydryl groups One solution contained, per liter, 20 g of tryptone (Difco), 10 g of yeast extract (Difco), and 20 g of NaCl. (23, 29), damages bacterial cell membranes (19, The other solution contained 28 g of agar (Difco) and 21), and causes a rapid bacteriostatic effect in 2 g of glucose. The two solutions were autoclaved both gram-positive and gram-negative bacteria separately, cooled to 50°C, mixed, and poured into (9; 17, 22, 24, 27; B. Reiter, A. Pickering, J. D. plates. Oram, and G. S. Pope, J. Gen. Microbiol. 33: The blood agar medium was prepared as described xii, 1963). A slower bactericidal effect has been by Holdeman et al. (13) and contained defibrinated reported for some gram-negative bacteria, in- horse blood (GIBCO, Bio-Cult Ltd., Paisley, Scot- cluding Escherichia coli (5, 28). Lactoperoxi- land) hemolyzed by freeze-thawing. dase and thiocyanate have also been shown to Chemicals. Hydrogen peroxide (30%o, wt/wt; Perhy- drol) was from E. Merck AG, Darmstadt, Germany. protect some oral bacteria from the effects of Lactoperoxidase (L 2005; from milk) was from Sigma bactericidal concentrations of hydrogen perox- Chemical Co., St. Louis, Mo. Potassium thiocyanate ide under anaerobic conditions (6). This study was from Riedel-De Haen AG, Seelze-Hannover, Ger- examines the effects oflactoperoxidase and thio- many. cyanate on the survival of bacteria treated with Exposure to lactoperoxidase-thiocyanate-hydrogen hydrogen peroxide under aerobic conditions. peroxide. The strains were grown at 37°C in the broth 20 VOL. 35, 1982 PROTECTION BY LACTOPEROXIDASE AND THIOCYANATE 21 under aerobic conditions. In exponential growth, they 4 were diluted in aerated dilution solution to a density of E. coli K-12 AB1157 about 3 x 104 organisms per ml. At this dilution of the growing culture, no components of the broth inter- fered with hydrogen peroxide toxicity or substituted for thiocyanate in its reaction with lactoperoxidase. A 0.2-ml sample ofbacterial suspension was added to 1.8 ml ofthe dilution solution containing various combina- *_4--_-4_44 4 tions of lactoperoxidase and thiocyanate. Hydrogen peroxide was added to the bacteria 2 min after the start 90 0 of the dilution procedure. Samples (0.1 ml) were taken Co v 2 at regular time intervals from the reaction mixtures _-J and were spread over the surface of agar plates. When 0 -J3 the plates were incubated anaerobically, they were w stored at 37°C in an anaerobic box with an atmosphere 0 of 1OTo H2 and 5% CO2 in nitrogen (31). The numbers w of viable cells were determined. -J In one series of experiments, a 0.2-ml sample of strain AB1157 in exponential growth in broth was co LL added to 1.8 ml of aerated dilution solution giving 0 about 107 organisms per ml. All other conditions were a: the same as above. Only under these conditions could w a bactericidal effect of the hydrogen peroxide-lacto- peroxidase-thiocyanate combination be observed. z 1 RESULTS 2 When E. coli at a density ofabout 3 x 103 cells per ml of dilution solution was exposed to 1 mM hydrogen peroxide, lactoperoxidase protected I I I Il from the bactericidal effect of hydrogen perox- 0 1 2 3 4 ide in both the presence and absence of thiocya- TIME (h) nate (Fig. 1). Similar results were obtained with FIG. 1. Killing ofE. coli in dilution solution at 37°C S. salivarius. When S. mutans was tested under when exposed under aerobic conditions to various similar conditions, lactoperoxidase protected combinations of lactoperoxidase, thiocyanate, and hy- only in the presence of thiocyanate. In the drogen peroxide. The number of viable cells deter- absence ofthiocyanate, lactoperoxidase potenti- mined on blood agar incubated anaerobically after ated the bactericidal effect of hydrogen peroxide various time intervals in the presence of (1) 1 mM on S. mutans (Fig. 2). Results with S. sanguis H202; (2)1 mM H202 plus lactoperoxidase (25 ,ug) plus 2 mM thiocyanate; (3) 1 mM H202 plus lactoperoxi- were similar to those obtained with S. mutans. dase (25 ,ug/ml); (4) no addition. Similar viable counts A bactericidal effect of lactoperoxidase-thio- were obtained on PYG agar incubated aerobically. cyanate-hydrogen peroxide greater than that of hydrogen peroxide itself was demonstrated only in E. coli and then only under very specific conditions (Fig. 3). An exponentially growing culture of E. coli K-12 strain AB1157 in broth nate under these conditions (Fig. 3) did not was diluted 10-fold in dilution solution giving a protect from hydrogen peroxide toxicity, as in density of about 1 organisms per ml and ex- the experiments illustrated in Fig. 1. However, posed for more than 1 h to lactoperoxidase- cells exposed to hydrogen peroxide in the pres- thiocyanate-hydrogen peroxide. Many of the ence of lactoperoxidase did recover much faster cells were damaged to such an extent that they than cells exposed to hydrogen peroxide in the could not recover when incubated on PYG agar absence of lactoperoxidase. The former cells or blood agar aerobically. When incubated an- started to grow rapidly within 1 h, whereas the aerobically on blood agar, however, most of the others did not (Fig. 3). cells recovered even after a 3-h exposure. A pronounced bacteriostatic effect of lactoperoxi- DISCUSSION dase-thiocyanate-hydrogen peroxide on E. coli In the presence of thiocyanate, lactoperoxi- was found under these conditions (Fig. 3), and dase has been shown to protect microorganisms readily visible colonies did not usually develop from the bactericidal effect of hydrogen perox- on the agar plates until after more than 24 h of ide under anaerobic conditions (6). The present incubation. The surviving cells under all other study demonstrated that this protection against conditions (Fig. 3) formed large colonies within hydrogen peroxide toxicity was also effective 24 h. Lactoperoxidase in the absence of thiocya- under aerobic conditions. For a variety oforgan- 22 ADAMSON AND CARLSSON INFECT. IMMUN. 4 9[ S. mutans NCTC 10449 E. coli K-12 AB1157

9 , 4 v 2 90 cm 8 -J 7 co cn -J w 0 0 w -J C') -J -J U- w 0 3 w wm-) CID -J 1 z LA. 0 2 w o 2 1 z TIME (h) 4 FIG. 2. Killing of S. mutans in dilution solution at 37°C when exposed under aerobic conditions to vari- ous combinations oflactoperoxidase, thiocyanate, and hydrogen peroxide. See legend to Fig. 1. Similar viable counts were obtained on blood agar incubated anaero- bically and on PYG agar incubated aerobically. isms over a wide range of hydrogen peroxide 5 concentrations, lactoperoxidase and thiocyanate offered protection (6). Even E. coli, for which a 6 bactericidal effect of lactoperoxidase-thiocya- nate-hydrogen peroxide has been demonstrated 3 ~ (4, 5, 21, 28; Reiter et al., J. Gen. Microbiol. 0 1 2 3 33:xii, 1963), was protected. The present study TIME (h) showed that this bactericidal effect was ex- pressed only when the cells were cultured under FIG. 3. Killing of E. coli K-12 when an exponen- tially growing culture in broth was diluted 10-fold in aerobic conditions after exposure to lactoperoxi- aerobic dilution solution and exposed at 37°C to vari- dase-thiocyanate-hydrogen peroxide. Most of ous combinations of 25 ,ug of lactoperoxidase (LPO) the cells incubated under anaerobic conditions per ml, 2 mM thiocyanate (SCN-), and 1 mM H202. on blood agar after the exposure proved to be After various time intervals, O.1-ml samples of the viable and apparently had the capacity to repair reaction mixtures were diluted and subcultured on the any lesions that may have been formed during surface of blood agar (BA) or PYG agar plates under aerobic exposure to the products of the lactoper- aerobic or anaerobic conditions. The number of viable oxidase reaction. cells was determined after various time intervals in the In the absence of thiocyanate, lactoperoxi- presence of the following additions and incubation conditions: (1) H202, BA, anaerobic; (2) H202, PYG, dase potentiated the toxic effect of hydrogen aerobic; (3) H202 plus LPO plus SCN-, BA, anaero- peroxide in some strains and prevented it in bic; (4) H202 plus LPO plus SCN-, PYG, anaerobic; others. This effect was strain specific, but was (5) H202plus LPO plus SCN-, BA, aerobic; (6) H202 also influenced by other conditions during the plus LPO plus SCN-, PYG, aerobic; (7) H202 plus exposure to hydrogen peroxide in the presence LPO, BA, anaerobic; (8) H202 plus LPO, PYG, aero- of lactoperoxidase. In dilution solution, lacto- bic; (9) no addition, BA, anaerobic. VOL. 35, 1982 PROTECTION BY LACTOPEROXIDASE AND THIOCYANATE 23 peroxidase protected E. coli K-12 strain AB1157 because it is quickly converted to OSCN- by (Fig. 1), but it did not significantly influence cell salivary lactoperoxidase and thiocyanate. How- killing when a growing culture of this strain was ever, it is possible to estimate the concentration exposed to hydrogen peroxide in dilution solu- of hydrogen peroxide which would exist in the tion containing 10% TSY broth (Fig. 3). Lacto- absence oflactoperoxidase. In human saliva, the peroxidase potentiated the peroxide-mediated average measured OSCN- concentration is ap- killing of S. sanguis strain ATCC 10556 in dilu- proximately 40 ,uM (K. M. Pruitt, J. Tenovuo, tion solution, but did not influence the killing of W. Fleming, and M. Adamson, Caries Res., in a growing culture in TSY broth (6). It is possible press). Therefore, in the absence of lactoperoxi- that various components of the cell and the dase, the average hydrogen peroxide concentra- broth could function as a substrate in reactions tion should be at least 40 ,uM. Catalase in whole catalyzed by lactoperoxidase in the presence of saliva is not efficient in destroying hydrogen hydrogen peroxide. In some organisms this peroxide in this concentration range (18). Since could be lethal for the cell, whereas in others it OSCN- may inhibit hydrogen peroxide-produc- protected the cell from the toxic effect ofhydro- ing bacteria (24), the rate ofproduction ofhydro- gen peroxide. gen peroxide might increase in the absence of The toxicity of hydrogen peroxide is not limit- lactoperoxidase. In addition, because a bacteri- ed to bacteria. Human cells can be much more um is essentially a point source of hydrogen sensitive to hydrogen peroxide toxicity than peroxide, the local hydrogen peroxide concen- many bacteria. Human cells in culture can be tration on the colonized epithelial cell surfaces killed at concentrations of hydrogen peroxide could be much higher than that extrapolated lower than 10 FjM (12, 26). Primary lesions in from whole-saliva assays of OSCN-. both human cells and bacteria are single-strand- ed breaks in DNA (1, 12). Most mammalian cells ACKNOWLEDGMENT and aerobic bacteria have highly developed de- This study was supported by Swedish Medical Research fense mechanisms against hydrogen peroxide Council project 4977. and other intermediates of oxygen reduction, such as superoxide and hydroxyl radicals (8, 11). The concerted action of the enzymes superoxide LITERATURE CITED 1. Ananthswamy, H. N., and A. Ess. 1977. Repair of dismutase, catalase, and peroxidase keep the hydrogen peroxide-induced single-strand breaks in Es- intracellular concentrations of the oxygen inter- cherichia coli deoxyribonucleic acid. J. Bacteriol. mediates at nontoxic levels. If a cell actually 130:187-191. becomes injured, there are systems for the re- 2. Anne, T. M., and E. L. Thomas. 1977. Accumulation of hypothiocyanite ion during peroxidase-catalysed oxida- pair of the lesions (1, 7, 12). The cells of the tion of thiocyanate ion. Eur. J. Biochem. 80:209-214. mucous membranes of the oral cavity may be 3. Rchman, B. J. 1972. Pedigrees of some mutant strains of particularly exposed to oxygen intermediates Escherichia coli K-12. Bacteriol. Rev. 36:525-557. because they are colonized by bacteria in the 4. BJ6rck, L., and 0. Cla_on. 1980. Correlation between concentration of hypothiocyanite and antibacterial effect presence of significant levels of oxygen (10). of the lactoperoxidase system against Escherichia coli. J. Many oral bacteria, especially streptococci, Dairy Sci. 63:919-922. form hydrogen peroxide (14, 18, 29), and an 5. Bj3rck, L., C.-G. Ros6n, V. Marshll, and B. Reiter. 1975. obvious function of lactoperoxidase and thiocy- Antibacterial activity of the lactoperoxidase system in milk against pseudomonads and other gram-negative bac- anate in salivary secretions could be to protect teria. Appi. Microbiol. 30:199-204. the cells of the mucous membrane against hy- 6. Carbson, J. 1980. Bactericidal effect ofhydrogen peroxide drogen peroxide toxicity. Lactoperoxidase con- is prevented by the lactoperoxidase-thiocyanate system verts the highly toxic hydrogen peroxide formed under anaerobic conditions. Infect. Immun. 29:1190-1192. 7. Carson, J., and V. S. Carpenter. 1980. The recA' gene by oral streptococci into the less toxic product product is more important than catalase and superoxide OSCN-, and this product actually inhibits the dismutase in protecting Escherichia coli against hydrogen formation of hydrogen peroxide by the strepto- peroxide toxicity. J. Bacteriol. 142:319-321. cocci (24). 8. Chance, B., H. Slei, and A. Boveris. 1979. Hydroperoxide metabolism in mammalian organs. Physiol. Rev. 59:527- Verification of such a protective role on the 605. part of salivary lactoperoxidase and thiocyanate 9. Dogma, I. L., A. C. Kerr, and B. H. Amdur. 1962. for the human oral mucosa awaits the direct Characterization of an antibacterial factor in human parot- demonstration of protection in vitro of human id secretions, active against Lactobacillus casei. Arch. Oral Biol. 7:81-90. epithelial cells and studies on the oral conditions 10. Eskow, R. N., and W. J. Loesche. 1971. Oxygen tensions in persons who may be lacking salivary lactoper- in the human oral cavity. Arch. Oral Biol. 16:1127-1128. oxidase. In the absence of lactoperoxidase, the 11. Fridovch, I. 1978. The biology of oxygen radicals. Sci- hydrogen peroxide concentration in the oral ence 201:875-880. 12. Hoffman, M. E., and R. Meneg_hn. 1979. Action of cavity may reach significant levels. The exis- hydrogen peroxide on human fibroblast in culture. Photo- tence of hydrogen peroxide in saliva has not chem. Photobiol. 30:151-155. been directly demonstrated (18), presumably 13. Holdeman, L. V., P. Cato, and W. E. C. Moore (ed.). 1977. 24 ADAMSON AND CARLSSON INFECT. IMMUN.

Anaerobic laboratory manual, 4th ed. Virginia Polytech- and in a chemically defined nic Institute and State University, Blacksburg. culture medium. J. Gen. Microbiol 43:31-43. 14. Holmberg, K., and H. 0. Hollander. 1973. Production of 23. Mhlckeebo, M. N. 1979. Antibacterial action of lactoperox- bactericidal concentrations of hydrogen peroxide by idase-thiocyanate-hydrogen peroxide on Streptococcus Streptococcus sanguis. Arch. Oral Biol. 18:423-434. agalactiae. Appl. Environ. Microbiol. 38:821-826. 15. Hoogendoorn, H. 1974. The effect of lactoperoxidase- 24. Morison, M., and W. F. Steele. 1968. Lactoperoxidase, thiocyanate-hydrogen peroxide on the metabolism of car- the peroxidase in the salivary gland, p. 89-110. In P. iogenic micro-organisms in vitro and in the oral cavity. Persson (ed.), Biology of the mouth. American Associa- Mouton, Publishers, Den Haag, Netherlands. tion for the Advancement of Science, Washington, D.C. 16. Hoogendoorn, H., J. P. Plessens, W. Schdoes, and L. A. 25. Mler, A. J. R. 1966. Microbiological examination ofroot Stoddard. 1977. Hypothiocyanite ion: the inhibitorformed canals and periapical tissues of human teeth. Odontol. by the system lactoperoxidase-thiocyanate-hydrogen per- Tidskr. 74:Suppl. 1-380. oxide. I. Identification of the inhibiting compound. Caries 26. Nathan, C. F., L. H. Brukner, S. C. Siverstein, and Z. A. Res. 11:77-84. Cohn. 1979. Extacellular cytolysis by activated macro- 17. Klenoff, S. J., W. H. Clem, and R. G. Luebke. 1966. The phages and granulocytes. J. Exp. Med. 149:84-99. peroxidase-thiocyanate-hydrogen peroxide antimicrobial 27. Prutt, K. M., M. Adamson, and R. Arnold. 1979. Lacto- system. Biochim. Biophys. Acta 117:63-72. peroxidase binding to streptococci. Infect. Immun. 18. Kraus, F. W., J. F. Nickerson, W. I. Perry, and A. P. 25:304-309. Walker. 1957. Peroxide and peroxidogenic bacteria in 28. Relter, B., V. M. E. Marshall, L. BjOrck, and C.-G. Rosin. human saliva. J. Bacteriol. 73:727-735. 1976. Nonspecific bactericidal activity of the lactoperoxi- 19. Law, B. A., and P. John. 1981. Effect of the lactoperoxi- dase-thiocyanate-hydrogen peroxide system of milk dase bactericidal system on the formation of the electro- aginst Escherichia coli and some gram-negative patho- chemical proton gradient in E. coli. FEMS Microbiol. gens. Infect. Immun. 13:800-807. Lett. 10:67-70. 29. Tbomas, E. L., and T. M. Aune. 1978. Lactoperoxidase, peroxide, thiocyanate antimicrobial system: correlation of 20. LeBlen, T. W., and M. C. Bromel. 1975. Antibacterial sulflhydryl oxidation with antimicrobial action. Infect. properties of a peroxidogenic strain of Streptococcus Immun. 20:456-463. mitior (mitis). Can. J. Microbiol. 21:101-103. 30. Thompson, R., and A. Johmon. 1950. The inhibitory 21. Marshall, V. M. E., and B. Reiter. 1980. Comparison of action of saliva on the diphtheria bacillus: hydrogen the antibacterial activity of the hypothiocyanite anion peroxide, the inhibitory agent produced by salivary strep- towards Streptococcus lactis and Escherichia coli. J. Gen. tococci. J. Infect. Dis. 88:81-85. Microbiol. 120:513-516. 31. Yamada, T., and J. Carlson. 1975. Regulation of lactate 22. Mlckelson, M. N. 1966. Effect of lactoperoxidase and dehydrogenase and change of fermentation products in thiocyanate on the growth of Streptococcus pyrogenes streptococci. J. Bacteriol. 124:55-61.