Lactoperoxidase Binding to Streptococci - Infection and Immunity

Vol. 25, No. 1

INFECTION AND IMMUNITY, July 1979, 304-309 0019-9567/79/07-0304/06$02.00/0

Lactoperoxidase Binding to Streptococci KENNETH M. PRUITT,' * MICHAEL ADAMSON,' AND ROLAND ARNOLD2 Laboratory of Molecular Biology' and Department of Microbiology,2 University ofAlabama in Birmingham, Birmingham, Alabama 35294 Received for publication 26 April 1979

There have been conflicting reports regarding the binding of lactoperoxidase to bacterial cell surfaces. We describe here the effects of cell-bound lactoperoxidase on acid production by suspensions of Streptococcus mutans (NCTC 10449) in the presence of hydrogen peroxide and thiocyanate. Saline suspensions of log-phase bacteria were treated with 0.1 mg of lactoperoxidase per ml and were then washed thoroughly. The addition of hydrogen peroxide and thiocyanate markedly reduced the acid production of these lactoperoxidase-treated bacteria but had no effect on the acid production of untreated controls. After a 3-h incubation in saline, the lactoperoxidase-treated bacteria produced acid in the presence of hydrogen peroxide and thiocyanate at the same rate as untreated bacteria. These observations suggest that lactoperoxidase is initially bound to the cell surface in an enzymatically active form at a concentration sufficient to inhibit acid production. The lactoperoxidase is slowly degraded or desorbed as the bacteria stand in saline suspension.

Lactoperoxidase (LPO) catalyzes the oxidation of thiocyanate ions by hydrogen peroxide to produce an antimicrobial agent, apparently the hypothiocyanite ion (-OSCN), which oxidizes susceptible protein sulffihydryl groups and interferes with the metabolism and growth of many different types of microorganisms. Since both LPO and the thiocyanate ion are present in mucosal secretions, and since extracellular hydrogen peroxide is generated by many commensal bacteria, the LPO system probably serves an important regulatory function on mucosal surfaces. [For reviews and discussions of the earlier literature, see Aune and Thomas (1); Thomas and Aune (13); Hamon and Klebanoff (2); Hoogendoorn et al. (5); Mickelson (7); Rieter et al. (9); Tenovuo (11); and Wood (14).] The enzyme has high surface affinity and binds to human enamel (8) and to salivary sediment (12) in a catalytically active form. However, controversy exists as to whether or not LPO binds to the surface of target microorganisms and functions on the cell membrane. Steele and Morrison (10) reported that the growth of Streptococcus cremoris 972 cells which had been treated with LPO and then resuspended in LPO-free medium was inhibited by the addition of thiocyanate and hydrogen peroxide. Control cultures treated identically but not exposed to LPO grew normally. They concluded that LPO bound in an enzymatically active form to the cell surface was responsible for the growth inhibition. Hoogendoorn (4) was

unable to reproduce the observations of Steele and Morrison (10). He suggested that cell binding of LPO might be reversible, and that by thoroughly washing his cells, he could have removed the bound enzyme. In the present report we describe experiments which we have carried out with LPO-treated Streptococcus mutans 10449 (serotype c). The microorganisms were exposed to a solution of LPO and washed free of detectable LPO activity. The addition of thiocyanate and hydrogen peroxide to the treated bacteria inhibited glucosestimulated acid production. Inhibition was not observed in saline suspensions of the LPOtreated and washed microorganisms which had stood at room temperature for over 3 h. We conclude that LPO enzyme molecules are bound initially to the cell surface in a catalytically active form. However, upon standing at room temperature the enzyme activity of the LPO is

lost. MATERIALS AND METHODS The water used in all experiments was deionized by treatment with a commerical mixed-bed resin and subsequently distilled from a two-stage quartz still. Unless otherwise noted, all chemicals were of reagent grade. To remove adsorbed proteins, all surfaces which contacted the mixtures used in the acid production assays were carefully washed with the commercial detergent Micro (obtained from International Products Corp., Trenton, N.J.) just before the assay. Lactoperoxidase. Bovine lactoperoxidase purified from the milk of cows was obtained as a lyophilized

304

VOL. 25, 1979

LACTOPEROXIDASE BINDING TO STREPTOCOCCI

powder from Sigma Chemical Co. LPO activity was calculated by the method of Pruitt and Adamson (8), but the assay conditions were modified by the method of S. Marklund (personal communication) as follows. The final concentration of hydrogen peroxide in the assay mixture was 1.5 mM, and the final pyrogallol concentration was 25 mM. The pyrogallol stock solution was prepared in 20 mM HCl to increase pyrogallol stability. The assay buffer was 0.2 M sodium phosphate (pH 6.0). The A4w was digitized at 1-s intervals, using a model 2200A Datalogger (Fluke Instrument Co.). LPO activity calculations were based on the first 30 s of the reaction. Thiocyanate. Solutions of thiocyanate were prepared by dissolving potassium thiocyanate (J. T. Baker Chemical Co.) in 0.9% sodium chloride. Concentrations of thiocyanate ion were determined by using the ferric nitrate method (5). Ferric nitrate [10 g of Fe(NO3)3.9H20] was dissolved in 20 ml of concentrated nitric acid and diluted to a final volume of 200 ml with water. The thiocyanate sample to be analyzed (0.1 ml) was added to 2 ml of the ferric nitrate stock solution. Absorbance was measured at 460 nm, and thiocyanate concentration was calculated by comparison with a thiocyanante standard. Peroxide. Hydrogen peroxide was purchased as a 30% solution (J. T. Baker Chemical Co.) and stored at 4VC. Peroxide was assayed by measuring the absorbance at 460 nm of O-dianisidine oxidized by hydrogen peroxide in the presence of horseradish peroxidase. The O-dianisidine was obtained as the hydrochloride salt from Sigma Chemical Co. Horseradish peroxidase (880 U/mg) was purchased from Worthington Biochemicals Corporation. For the assay, O-dianisidine (10 mg) was dissolved in 10 ml of methanol, and 0.2 ml of this stock solution was added to 2 ml of 0.2 M sodium phosphate (pH 6.0) together with 0.1 ml of horseradish peroxidase stock solution (50 ,ug/ml in 0.2 M phosphate, pH 6). The hydrogen peroxide solution to be assayed (0.1 ml) was added to the assay mixture, and the A4ws was measured. The hydrogen peroxide concentration was calculated assuming that each mole of oxidized O-dianisidine corresponded to one mole of hydrogen peroxide in the sample. The molar extinction coefficient of oxidized O-dianisidine at 460 nm was assumed to be 11,300 cm-'. Absolute determinations of peroxide concentrations were necessary because the 30% peroxide solution was unstable even when stored at 4°C. Over a 3-year period, the concentration declined from 30 to 19%. Hypothiocyanite. The -OSCN ion was assayed by reaction with the colored anionic monomer of DTNB as described by Aune and Thomas (1). Bacteria. A stock of Streptococcus mutans 10449 was lyophilized and stored in ampoules at 4°C. For the experiments each day, lyophiles were reconstituted and inoculated into brain heart infusion broth (BHI) and incubated at 37°C for 18 h. A 1-ml portion of this culture was used to inoculate 100 ml of BHI. The second culture was grown to late log phase (absorbance at 660 nrm [A6e] was 0.5 to 1.0) at 370C and harvested by centrifugation. The bacterial pellets were washed twice with saline at room temperature (21 to 23°C) and resuspended to an A6s of about 10. The cells were maintained in this thick suspension at room

305

temperature, and portions were used for the experiments that day. Sodium hydroxide. Carbonate-free sodium hydroxide was prepared by adding 10 ml of water to 10 g of reagent-grade NaOH pellets, stirring for 10 min, and centrifuging the solution while still hot. One-half of the supernatant was then made up to 1 liter with saline. This gave an approximately 0.1 M solution of base. The solution was standardized by titration against potassium hydrogen phthalate (acidimetric standard; Fisher Scientific Co.) and stored at room temperature in a bottle fitted with an Ascarite-Drierite air intake tube to exclude carbon dioxide. For routine acid production assays, portions of the stock base solution were diluted to 1.00 mM with saline and were prepared fresh daily. Acid production assay. Portions (100 pl each) of the twice-washed bacteria suspended in saline were added to 2 ml of saline in the titration cell. The mixture was stirred constantly with a magnetic stirring bar. The pH was controlled with a Radiometer pHstat titrator type TTT1c combined with an SBR2 Titragraph and an ABU1 Autoburette. The mixture was maintained at pH 6.5 by automatic addition of 1.00 mM NaOH. Acid production was stimulated by the addition of 100 A1 of glucose solution to give a final glucose concentration in the titration cell of 1%. The rate of base addition required to maintain the pH at 6.5 was constant during the first 10 min after glucose addition. The rate of acid production was measured as the slope of the volume versus the time curve taken from the recorder. To be certain that the rate of base addition was, in fact, directly proportional to the rate of acid production by the bacteria, the following test was made. A precision peristaltic pump (Minipuls II; Gilson Medical Electronics, Inc.) was calibrated gravimetrically, using distilled water. A solution of lactic acid in saline was standardized by titration against the previously standardized NaOH solution. To mimic acid production by bacteria, the calibrated pump was used to deliver the lactic acid solution (0.982 mM) at fixed rates to the titration cell. The experiment was carried out exactly as though bacteria were being analyzed, except that 100 pi of saline was added instead of the bacterial suspension. The rate of base addition required to maintain pH at 6.5 for several fixed rates of lactic acid addition was then measured. Adsorption of lactoperoxidase to the glass electrode. Bacteria which normally responded to glucose stimulation in the presence of thiocyanate (3.7 mM) and hydrogen peroxide (130,uM) by producing acid at a rate of 100 nmol/min per absorbance unit were completely inhibited when pH was monitored by using a combination glass electrode previously exposed to a saline solution of LPO (0.02 mg/ml). This inhibitory effect was not removed by repeated washing of the electrode with water and 70% ethanol. However, when the electrode was washed with undiluted laboratory detergent (Micro), the inhibitory effect was abolished and the bacteria responded normally to the addition of glucose in the presence of thiocyanate and hydrogen peroxide. An important practical aspect of titration experiments utilizing a combination glass electrode is the

306

INFECT. IMMUN.

PRUITT, ADAMSON, AND ARNOLD

diffusion of glucose into the saturated KCl salt bridge in the electrode. During the course of a typical experiment, a significant quantity of glucose diffuses into the porous plug of the electrode. Even though the electrode is carefully washed between experiments, this glucose may not be entirely removed. The result is that in subsequent experiments it will diffuse back out into the titration cell and produce a significant rate of acid production when bacteria are added, even though no further additon of glucose is made. We avoid this problem by washing the electrode thoroughly with water and 70% ethanol and then soaking it for at least 5 min, with stirring, in 5 ml of saline to allow back diffusion of the glucose. The washing procedure is then repeated. With this technique, we find that the acid production of S. mutans 10449 in the absence of added glucose is less than 1% of the rate observed when excess glucose is added. Adsorption of LPO to bacteria. The washed bacteria were suspended in saline (final A6we = 7) containing LPO (0.1 mg/ml). The mixture (final volume, 2 ml) was incubated for 5 min at room temperature and centrifuged, and the pellets washed twice with 3 ml of saline to remove unbound LPO. The washing procedure required approximately 30 min. No LPO activity could be detected in the supernatant from the second wash. The washed, LPO-treated bacteria were suspended in saline and allowed to stand at room temperature. For acid production analysis, 100 Ill of the suspension was added to 2 ml of saline in the titration cell (A6we = 0.50), and acid production was stimulated by addition of 100 td of glucose solution. Inhibition of acid production by the LPO system. After addition of bacteria to the saline in the titration cell, portions of saline solutions of potassium thiocyanate and hydrogen peroxide were added in sufficient quantity to give final concentrations of 130 AIM hydrogen peroxide and 3.7 mM thiocyanate. LPO dissolved in saline was added in sufficient quantity to give a final concentration of 0.03 mg/nil. The resulting inhibition of acid production was reversed by addition of saline-diluted 2-mercaptoethanol (final concentration, 630,uM). In some control experiments, the LPOtreated bacteria were replaced by a portion of the supernatant from the second wash. This supernantant, which presumably contained the same concentration of unbound LPO as the interstitial fluid in the washed bacterial pellet, was added to the pH stat along with untreated bacteria to insure that the LPO-catalyzed inhibition of acid production in treated bacteria did not result from liquid-phase LPO in the interstitial fluid of the cell pellet from the second wash.

S .I ,30

^,/

00 M 0 . 0210

nanomoles of sodium hydroxide added per minute

FIG. 1. (A) Titration of lactic acid. The acid (0.982 mM in saline) was added at fixed rates to 2.0 ml of normal saline in the titration cell. The pH was maintained at 6.5 by the addition of base (1.00 mM NaOH in saline) with an automatic burette. The points (0) indicate observations, and the line is the theoretical line, with slope 1.00 and intercept 0,0. (B) Titration of acid produced by S. mutans 10449. Various quantities of bacteria suspended in saline were added to 2 ml of saline. Acid production was stimulated by the addition of 100 ,ul of 22% glucose. The pH was maintained at 6.5 by the addition of NaOH. Points indicate observations. The slope of the line, determined by least-squares analysis, is 245.7 nmol of NaOH per min per absorbance unit. The intercept is 1. 79 nmol/ min.

was stimulated by the addition of glucose, the rate of base addition required to maintain the pH at 6.5 was directly proportional to the quantity (measured as A66o) of bacteria added (Fig.

1B). Rate of acid production by S. mutans 10449. When stimulated by the addition of glucose and maintained at pH 6.5, the bacteria produced acid at a significant rate for periods up to 30 min (Fig. 2). The rate of acid production during the first 10 min was constant, but decreased to approximately 80% of the initial value some 15 min after glucose addition. The absolute ranges ofthe control values in Table 1 are typical of the variability observed for a given concentration of bacteria taken from the same culture over a 6-h period. The day-to-day variation of different cultures was much greater. The effect of LPO treatment on acid proThe inset in Fig. 2 shows that the duction. RESULTS addition of the complete LPO system to an Test of the acid production assay. When active suspension of bacteria terminated acid lactic acid was added to saline at a constant production in less than 1 min. Subsequent adknown rate with a calibrated precision pump dition of mercaptoethanol regenerated acid proand the pH was maintained at 6.5 by the addi- duction. tion of sodium hydroxide, the rate of addition of Acid production by LPO-treated and the base was directly proportional to the rate of washed S. mutans 10449. Suspensions of the addition of the lactic acid (Fig. 1A). bacteria which had been treated with LPO, Titration of acid produced by S. mutans washed, and mixed with thiocyanate, glucose 10449. When various amounts of S. mutans and peroxide were able to produce acid at initial 10449 were added to saline and acid production rates (Fig. 2, curves 1, 4, and 6) which were


VOL. 25, 1979 12000

3

XI

0

1

M

z

tetmnnlcsesiuLPO

FIG. 2.Efcto E0 400

40

co~~~~in fe lcoeadto E

Min. 20

I0

min. FIG. 2.

Effect

after glucose addition

30

after glucose addition

of LPO

treatment

on

glucose stimu-

lated acid production by S. mutans 10449. Suspensions of bacteria were exposed to LPO or to saline and washed twice as described in the text. Pellets from the second wash were suspended in saline, and portions were added to saline in the titration cell. The curves, which are the actual recorder tracings, were generated by adding glucose to the individual suspensions and recording the volume of base added to maintain the pH at 6.5. For each curve, thiocyanate (final concentration, 3.7 mM) was added before glucose addition, and hydrogen peroxide (final concentration, 130 AM) was added 4 min after glucose. Curves 1, 4, and 6 were generated by LPO-treated and washed bacteria (44, 134, and 201 min, respectively, after exposure to LPO). Curves 2 and 5 are controls. For curve 3, 100Al of an appropriate dilution of the supernatant from the second wash of the LPOtreated bacteria was added to the control suspension before addition of glucose as a control for possible LPO activity in interstitial fluid. Inset, Inhibition of glucose stimulated acid production by S. mutans 10449 and reversal of the inhibition by mercaptoethanol. Twice-washed bacteria were suspended in saline in the titration cell as described in the text. Thiocyanate was added, followed by glucose, hydrogen peroxide, lactoperoxidase, and mercaptoethanol at the indicated times. The curve is the actual recorder tracing of the volume of base which was added to maintain the pH at 6.5.

comparable to controls (Fig. 2, curves 2 and 5) similarly treated but not exposed to LPO (Table 1). However, after 15 min, there was a substantial decrease in the acid production rate of the

307

LPO-treated bacteria compared to the appropriate controls. The LPO-treated bacteria showed a greater drop in acid production from 3 to 15 min than did the appropriate controls (Table 1). The differences in LPO-treated bacteria and controls were time dependent. The acid production curve for the LPO-treated bacteria 201 min after exposure to LPO (Fig. 2, curve 6) was not significantly different in any respect from the control curves over the 30-min interval monitored. Interstitial fluid from the bacterial pellet. When bacteria were treated with LPO and washed as described above, the supernatant from the second wash contained no measurable LPO activity in the pyrogallol assay. Nevertheless, to insure that bound LPO and not liquidphase interstitial LPO was responsible for the inhibition of acid production, a portion of the supernatant was added to the pH stat apparatus along with untreated bacteria, peroxide, and thiocyanate. The volume of supernatant used was equivalent to the volume of interstitial fluid transferred to the pH stat along with the LPOtreated bacteria. The untreated bacteria in the presence of supernatant (Fig. 2, curve 3) from the second wash produced acid at the same rate and for the same length of time as control bacteria alone. LPO activity of LPO-treated, washed bacteria. LPO-treated bacteria responded with normal acid production when stimulated with glucose. However, when incubated for 15 min with thiocyanate and peroxide, these LPOtreated bacteria showed a 50% reduction in acid production compared to controls (Table 1). To obtain an estimate of the LPO activity associated with these cells, dilutions of an LPO enzyme preparation were added to untreated bacteria together with thiocyanate and peroxide. The time required for a given dilution of LPO to reduce acid production to 50% of the control level was taken as the measure of the LPO effect. This number was related to the amount of LPO added (Fig. 3). Acid production terminated within 1 min after the addition of 0.1 mg of LPO to the titration cell containing the cofactors. The addition of as little as 10 ng of LPO to the titration cell resulted in a significant inhibition of acid production after 15 min. Thus, the inhibition of acid production provides a more sensitive measure of LPO activity than does the oxidation of pyrogallol. (In our hands, the minimum quantity of LPO which can be accurately measured by the pyrogallol assay is 100 ng.) When the complete LPO system was added to a glucose-stimulated suspension of S. mutans 10449, acid production terminated within 1 min

308

INFECT. IMMUN. PRUITT, ADAMSON, AND ARNOLD TABLE 1. Effects of the LPO system on glucose-stimulated acid production by S. mutans 10449 % Decrease Min after exposure to LpOa

Prepn

Acid production rate (3 min after glucose

addition)'

Acid production rate (15 min after glucose addition)b

3 -. 15 min'

Compared to control; 15 min

Control 107 ± 11d 83 ± 6d 22 ± 6d 44 100 35 65 LPO treated 58 134 118 LPO treated 48 59 42 201 115 74 36 LPO treated 11 a Bacteria were exposed to LPO, centrifuged, and washed twice as described in the text. to maintain pH at 6.5. bEach value indicates nanomoles of NaOH added per minute per c Each value indicates the 15-min acid production rate for each preparation compared to its acid production rate at 3 min. d Mean ± absolute range for three separate experiments.

A6ee

c 24

'iv c

-4

w

ap8 S 0

20

40

60

8

I0

nanograms lactoperoxidase added

FIG. 3. Effect of variations in LPO ca oncentration inhibition of acid production by glucose-stimulated S. mutans 10449. The bacteria wer-e suspended in 2 ml of saline (A6eo = 0.5) containing thiocyanate (3.7mM). Acidproduction was stimulated by addition ofglucose (final concentration, 1%) and m described in the text. The points (0) indicate the times at which the rate of acid productiion, after the addition ofperoxide, decreased to 50% of the control value for the respective quantities of adc led lactoperoxidase. Data from Fig. 2 (curves 1 and 4) were used (A) to estimate the inhibition of acid pr,oduction for LPO-treated and washed bacteria. on

(Fig. 2, inset). Neither the enzyme n or either of the cofactors added singly or in pajirs had any effect on glucose-stimulated acid prioduction at the concentrations employed in thLese experments. Addition of mercaptoethan ol restored acid production, presumably by re du. cng the protein sulfenyl thiocyanate derivatives produced by LPO-generated hypothiot cyanite (1). DISCUSSION Our results show that LPO is adso)rbed to the cell surface of S. mutans 10449. Thee enzyme is

apparently catalytically active on t]he cell surface, since addition of peroxide and lthiocyanate at concentrations which would produ ce no effect on untreated control bacteria result;ed in a sig-

nificant decrease in acid production of the LPOtreated and washed microorganisms. The amount of LPO enzyme activity associated with the LPO-treated cells can be determined from Fig. 3. The time required for acid production to decrease to 50% of the control value for curves 1 and 4 (Fig. 2) corresponded to what was found for the addition of approximately 0.01 Ag of LPO together with cofactors to cell suspensions not previously exposed to the enzyme. Assuming a molecular weight of 76,000, this amount corresponded to 1.32 x 10-'3 mol of LPO or 7.92 x 1010 molecules. The cell suspension contained 1 A6wo unit of bacteria. We typically found approximately 1.2 x 109 colony-forming units per absorbance unit for this microorganism. Thus, there were approximately 66 LPO molecules per colony-forming unit. Due to chaining, the actual number of cells present in a given suspension is, of course, usually greater than the number of colony-forming units found. Thus the actual number of LPO molecules present per cell was probably significantly less than 66. In addition, this calculation neglects any free LPO present in the suspension and thus overestimates the actual number of bound enzyme molecules. Apparently the cell-bound enzyme was very efficient, since so few molecules were required per cell to inhibit acid production. After standing at room temperature, the LPOtreated bacteria recovered their normal response to glucose stimulation, suggesting a loss of LPO activity on the bacterial surface. Similar studies of the activity of LPO bound to enamel (8), to apatite (12) and to salivary sediment (12) indicate that the surface binding per se does not make the enzyme unstable. There remain at least two possible explanations for the instability of cell-bound LPO: (1) the adsorbed enzyme may actually be degraded or inactivated by the bacteria or (2) the adsorbed molecules may be desorbed on standing and the redistribution of LPO molecules through the liquid phase may


VOL. 25, 1979

lead to less effective generation of -OSCN in the immediate vicinity of the cells. Since the experimental protocol was designed to determine whether LPO binds in an enzymatically active conformation to the bacterial cell surface, it was necessary to wash the bacteria free of interstitial LPO. The artificially low concentration of LPO resulting from this procedure would not be representative of the LPO levels expected in the developing plaque matrix. In this plaque matrix, losses of cell-bound LPO activity due to desorption or degradation could be replaced from the interstitial pools of LPO. However, as the plaque develops, access of salivary LPO to deep interior regions may become limited and degradation or desorption of cellbound enzyme molecules could lead to significant reduction of LPO activity. Therefore, although our demonstration of the binding of LPO and subsequent loss of enzymatic activity may explain the contradictory results of Steele and Morrison (10) and Hoogendoorn (4), the biological significance of such degradation or desorption remains an open question. The binding of LPO to bacteria could enhance the antimicrobial effects of the enzyme. S. sanguis, one of the early colonizers of the tooth surface, has been shown (3), along with other oral bacteria (6), to produce high levels of hydrogen peroxide. The binding of LPO to such bacteria would place the enzyme at the site of highest peroxide concentration, thus kinetically favoring the generation of -OSCN. Similarly, the -OSCN generated by cell-bound LPO would be at its greatest concentration in the immediate vicinity of its target. ACKNOWLEDGMENTS This work was supported by Public Health Service contract DE-52456 and grant DE-04801 from the National Institute of Dental Research.

309

LITERATURE CITED 1. Aune, T. M., and E. L. Thomas. 1978. Oxidation of protein sulfhydryls by products of peroxidase-catalyzed oxidation of thiocyanate ion. Biochemistry 17:10051010. 2. Hamon, C. B., and S. J. Klebanoff. 1973. A peroxidasemediated Streptococcus mitis-dependent antimicrobial system in saliva. J. Exp. Med. 137:438-450. 3. Holmberg, K., and H. 0. Hallander. 1973. Production of bactericidal concentrations of hydrogen peroxide by Streptococcus sanguis. Arch. Oral Biol. 18:423-434. 4. Hoogendoorn, H. 1974. The effect of lactoperoxidasethiocyanate-hydrogen peroxide on the metabolism of cariogenic micro-organisms in vitro and in the oral cavity. Mouton, Publishers, Den Haag, The Netherlands. 5. Hoogendoorn, H., J. P. Piessens, W. Scholtes, and L. A. Stoddard. 1977. Hypothiocyanite ion; the inhibitor formed by the system lactoperoxidase-thiocyanate-hydrogen peroxide. Caries Res. 11:77-84. 6. Kraus, F. W., J. F. Nickerson, W. L. Perry, and A. P. Walker. 1957. Peroxide and peroxidogenic bacteria in human saliva. J. Bacteriol. 73:727-735. 7. Mickelson, M. N. 1977. Glucose transport in Streptococcus agalactiae and its inhibition by lactoperoxidasethiocyanate-hydrogen peroxide. J. Bacteriol. 132:541548. 8. Pruitt, K. M., and M. Adamson. 1977. Enzyme activity of salivary lactoperoxidase adsorbed to human enamel. Infect. Immun. 17:112-116. 9. Reiter, B., V. M. E. Marshall, L. Bjorck, and C.-G. Rosen. 1976. Nonspecific bacterial activity of the lactoperoxidase-thiocyanate-hydrogen peroxide system of milk against Escherichia coli and some gram-negative pathogens. Infect. Immun. 13:800-807. 10. Steele, W. F., and M. Morrison. 1969. Antistreptococcal activity of lactoperoxidase. J. Bacteriol. 97:635-639. 11. Tenovuo, J. 1978. Inhibition by thiocyanate of lactoperoxidase-catalyzed oxidation and iodination reactions. Arch. Oral Biol. 23:899-903. 12. Tenovuo, J., J. Valtakoski, and M. L. E. Knuuttila. 1977. Antibacterial activity of lactoperoxidase adsorbed by human salivary sediment and hydroxyapatite. Carieg Res. 11:257-262. 13. Thomas, E. L., and T. M. Aune. 1978. Lactoperoxidase, peroxide, thiocyanate antimicrobial system: correlation of sulfhydryl oxidation with antimicrobial action. Infect. Immun. 20:456-463. 14. Wood, J. L. 1975. Biochemistry, p. 156-221. In A. A. Newman (ed.), Chemistry and biochemistry of thiocyanic acid and its derivatives. Academic Press, Inc., New York.