Further in vivo studies on the plaque-inhibiting effect of chlorhexidine and its binding mechanisms SONNI METTE WAALER Department of Pedodontics and Caries Prophylaxis, Department of Prosthetic Dentistry and Stomatognathic Physiology, Dental Faculty, University ef Oslo, Oslo, Morway
Waaler SM: Further in vivo studies on the plaque-Inhibiting effect of chlorhexidine and its binding mechanisms, Scand J Dent Res 1990; 98: 422-7. Abstract — The aim of the present study was to investigate the relative importance of chemical groups in the oral cavity which bind chlorhexidine of clinical signilicance. This was performed by climcal experiments where a test panel of 11 individuals rinsed with chlorhexidine at pH 7, pH 5.5, and pH 3. Rinsing with 0.02 M EDTA was performed prior to the chlorhexidine rinse, because this procedure has been shown to enhance the antibacterial effect of chlorhexidine in vitro. However, pre-dnses with EDTA did not increase the clinical effect of chlorhexidine. Bacteriologic tests showed comparable andbacterial effect of chlorhexidine at pH 7 and pH 3. Mouthrinses of pH 7 and pH 5.5 had comparable plaque-inhibiting effects, whereas the effect of a rinse of pH 3 was not different from that of the placebo. This finding is iiiteipreted to suggest that primary phosphate (pK about 2) is an important receptor site for chlorhexidine, whereas secondary phosphate (pK = 7.0) is of negligible importance. Previous studies have clearly shown that phosphate binds chlorhexidine. The lack of clinical effect at pH 3 is probably due to precipitation of salivary proteins by hydrogen ions, making primary phosphate groups unavailable for interaction with chlorhexidine. Key words: chlorhexidine; clinical study; plaque inhibition. Department of Pedodontics and Garies Prophylaxis, Dental Faculty, Geitmyrsveien 71, 0455 Oslo 4, Norway. Accepted for publication 20 February 1990.
The bis-biguanides, and in particular chlorhexidine, are unique in their antiplaque effeet among all the antibacterial substances that have been tested (1). No correlation between andbacterial acdvity and plaqueinhibiting effect was evident for the substances tested (I). It has recently been detnoostrated that combinations of chlorhexidine
and other antibacterial agents which enhanced the antibacterial activity, did not necessarily itnprove their clinical effect (2, 3). Much research has been perfortned concerning the mechanisms by which cblorhexidine exerts its effect, but many aspects are still not well understood, The retention (the substantivity) was first
IN VIVO MECHANISMS OT CHLORHEXIDINE assumed to be the crucial property of chlorhexidine (4). However, it was later observed that quaternary ammonium compounds were retained in the mouth to a similar and even. higher degree than chlorhexidine, but that they exhibited scarcely any antiplaque effect (5). It was later suggested that an ion exchange mechanism was involved. Chlorhexidine was thought to be retained and to displace calcium from its binding sites in the mouth during rinsing procedures, and was subsequently displaced by calcium when the relative concentration of this ion increased (6). The sites aivailable for binding have been suggested to be mainly carboxyl groups (7). This was based on the fitiding that both clinical effect and retention were markedly reduced when chlorhexidine mouthrinses at pH 3 was used, this value coinciding approximately with the pH at which carboxyl groups lose their charge (8). However, results from previous experiments (9, 10) indicated that phosphate groups (and probably also sulfate groups) provide clinically significant binding sites for chlorhexidine in the mouth. Mucins and bacterial surfaces provide bound phosphate and sulfate groups (11,12). This suggestion was based on the observations that lanthanum ions, with their high a.ffmity for phosphate groups (13) were able to block receptor sites for chlorhexidine in vivo. Sitnilar effects were observed for Sn^ * and Zn"*^ *, which also have higher affinity for phosphate than for carboxyl (9, 10). The aim of the present study was to further investigate the significance of bound phosphate as receptor sites for chlorhexidine in the mouth. Furthermore, in vitro studies (14) have shown that the chelating agent EDTA enhances the antimicrobial actiwty of chlorhexidine, probably by displacing cations on the bacterial surfaces, making these sites more easily available for chlorhexidine molecules. Thus, the second aim of the study was to investigate whether pre-rinses with EDTA increased the plaque-inhibiting effect exerted by chlorhexidine.
423
Material and methods The present study comprised two experiments.
EXPERIME_NT 1 A plaque inhibition study, the effect of pre-rinses with
EDTA - Eleven individuals volunteered for the experiment and informed consent was obtained. The group was mixed with respect to dental health, and consisted of clinical and preclinical dental students, dental staff, dental hygienists, and persons with no professional relation to dentistry'. The age range was between 19 and 56 yr. Prior to each test period their teeth were scaled and polished. Each participant was given written instructions on how to perform the test regimens. For each test period of 4 days, no mechanical or other oral hygiene was allowed. Sucrose-contasnitig chewing gum was used by each participant for 5 min six times daily. The respective test substances were used as aqueous solutions and a 10 ml mouthrinse "w^as used for 1 min twice daily. At the end of each test period, the plaque score was recorded, using the Plaque Index (Pl.l) of SILNESS & LOE (15).
The study was carried out according to a double-blind cross-over design (1'6—18). During the study, all participants tested the following rinsing regimens: water (control), 0.02 M EDTA pH 7, pre-rinses with water or EDTA followed by rinses with 0.075% chlorhexidine acetate pH 7. The relatively low concentration of chlorhexidine was chosen to make any improvement in effect easily demonstrated. Statistical analyses was carried out using Student's ^-test for paired data.
EXPERIMENT 2 Bacteriologic test: Antibacterial activity of chlorhexidine acetate at pH 7 or pH 3 — Streptococcus sanguis ATCC
10556 (NGTC 7863) grown in Brain Heart Infusion broth (BHI, Difco Laboratories, Detroit, Michigan, USA) was collected in exponential phase by centrifugation (9000^, 4''C). The bacteria were washed once in 10' mM Hepes buffer (Sigma), pH 7, and the optical density was adjusted to 1.5 (680 nm, Uvicon 810 Kontron Spectro-
424
WAALER Table 1
Mean Plague Index (PLI) values obtained after rinsing with cklorkexidine (CH) with and without pre-rinses with EDTA. Both water and EDTA alone were included as controls (n = ll)
Mean PLI SD
H,O
EDTA GHpH = 7
H,O G H p H == 7
1.25 (0.34)
0.34 (0.22)
0.39 (0.19)
EDTA 1.14 (0.37)
photometer) corresponding to 0.4 mg protein/ml Results cell SHspension (BCA Protein Reagent, Pierce EXPERIMENT 1 Chemical Company, Rockford, IL, USA). Susceptibility tests were performed by exposing the bacte- The results from the plaque inhibition study ria to serial dilutions of chlorhexidine acetate employing EDTA pre-rinses are shown in (GH) in 10 mM Hepes buffer at pH 7 or pH 3. Titrations were performed in sterile microdter Table 1. The use of a 0.02 M EDTA solution prior trays (Dynatech, Essex, England/Greiner, Niirtingen. West Germany). The fmal concentration to the chlorhexidine solution did not enhance of CH in the first well was 0.1% (1.47 mM) and the clinical effect of chlorhexidine signifithe dilution coefficient was 0.625. The bacterial candy (P = 0.42). The effect of using EDTA inoculum was 75 |il. After 5 min exposure to GH, alone was not statistically different frotn the 10 jxl aiiqnots from each well were transferred to water control (P=0.40). Chlorhexidine used separate wells containing 150 ^il fresh BHI broth after an EDTA or a water pre-rinse gave a in a second microtiter tray. Erom each of these marked plaque itihibition compared with the wells 10 jil aliquots. were transferred to blood agar control, as could be expected (F=0.00). plates (Golumbia Agar Base with 7% human blood). The second microdter trays and the blood agar plates were incubated at 37°C for 20 h. Andbacterial activity was assessed by comparEXPERIMENT 2 ing growth of CH-exposed and non-exposed bacteBacteriologic test — CH had the same antibacria. Growth was assessed by OD measurements of each well in the microtiter tray (MR 710 Mi- terial effect at pH 7 and pH 3. At both pH 7 and pH 3, exposure of Strep, sanguis 10556, croplate Reader, Dynatech.) and by scoring bacteto GH for 5 min reduced subsequent bacteririal growth on the blood agar plates. Gells exposed to 10 mM Hepes buffer pH 7 or al growth in BHI broth as well as on blood pH 3 without addition of GH were included as agar plates at concentrations down to controls. The assay was performed in duplicate 0.0037%. Exposure of the cells to Hepes bufand repeated twice. fer without CH at pH 3 for 5 min had no Plague inhibition study. The effect ef chlorhexidine effect on subseqtient bacterial growth, mouthrinses al pH 7, 5.5, and 3 — Chlorhexidine solu- whereas prolonged incubation markedly retions at a concentration of 0.075% chlorhexidine duced subsequent growth in BHI broth and were adjusted to pH 3 and pH 5.5 with acetic on blood agar plates. acid. These acidified chiorhexidine solutions were Plaque inhibition study - The results of the used as mouthrinses. The study was otherwise perstudy employing chlorhexidine rinses of pH formed as described above, involving the same test subjects as in Experiment 1. Statistical analyses 3, 5.5, and 7 are shown in Table 2. The was carried out using Student's t-test for paired chlorhexidine solution of pH 3 showed no data. antiplaque effect, the PLI values were not
IN VIVO MECHANISMS OF CHLORHEXIDINE
425
Table 2 Mean Plaque Index (PLI) values after cldorkexidine (CH) mouthrinses at differentpH (a
Mean Pl.I SD
CH pH = 3
CHpH = 5.5
CHpH = V
1.29 (0.38)
0.37 (0,.25)
0.39 (0.19)
1.25 (0,34)
tibacterial acdvity was tnaintained at pH 3, and the loss of clinical effect thus seemed to be dependent on factors other than antibacterial activity, confirming the findings of GJERMO et al. (7). These authors reported that the clinical effect was almost eliminated at pH 3, and they suggested that this was due Discussion to carboxyl groups being the major receptor Sucrose-induced plaque (19) was used as a sites, because these groups had lost their negmodel ill the present study, since this was adve charge at this pH. judged to be relevant for the clinical situaTbe present study demonstrated a total loss tion. The standardized sucrose exposure was of effect when the chlorhexidine at pH 3 also thought to eliminate the effect of any was used. However, in the light of previous major dietary differences in the test panel. studies in our laboratory, wbich showed that The study showed that the use of EDTA •J to -J of the clinically significant binding prior to a chlorbexidine moutbrinse did not sites in the oral cavity may be phosphate increase the plaque-inbibiting effect in vivo, groups (9, 10), another interpretation than although EDTA has been found to enhance that suggested by GJERMO et al. (7) may seem the antibacterial effect of chlorhexidine in justified. The suggestion that phosphate vitro (14). It was thus confirmed that the groups were important was based on the antibacterial effect does not necessarily coin- observation that Zn++ and in particular cide with the clinical effect. However, it may ,Sn'^* and La*"*^ were able to block the be that the lack of effect by EDTA was due receptor sites for chlorhexidine and reduce to the low concetitration of EDTA used. The its clinical effect. When rinsing with chloruse of a 0.2 M EDTA solution was tried in hexidine at pH 3, it appears conceivable that a pilot experiment by the author, but the the hydrogen ions cause precipitation of the experiment was discontinued due to develop- macrotnolecules which bind chlorhexidine at ment of hypersensitivity of the teeth. The neutral pH. These molecules contain pripart of tbe study concerning chlorhexidine mary phosphate (and probably sulfate) solutions at difTerent pH values showed that groups, and the loss of chnical effect of chlorthe aotiplaque efTect of chlorhexidine was hexidine at pH 3 may not necessarily be not decreased when lowering the pH of the caused by lack of charged binding sites, but solution from 7.0 to 5.5. This may be inter- rather by the precipitation of salivary preted as lack of significance of secondary proteins by hydrogen ions,, a reaction which phosphate (HPO4') as a binding site for the renders the potential binding sites for chlorclinical effect of chlorhexidine. Chlorhexi- hexidine unavailable. Such reactions have dine at pH 3 did not show an}' andplaque previously been discussed by NEBMAN & efTect. Bacteriologic tests sbowed that full an- NEUMAN (20). It can easily be demonstrated statistically different from the placebo [P = 0.81). The solution of pH 5.5 ga\'e no significantly different results when compared to the solution of pH 7 (F=0.80). Both of tbese coocentrations thus gave full clinical effect.
426
WAALER
in vitro that water acidified by acetic acid to pH 3 causes an immediate precipitation of salivary proteins when applied in clinically relevant proportions to whole .saliva (about 10:1). It may also he that the acid-precipitated proteins could be rapidly lost hy expectorations, thus giving a more rapid turnover of the mucin layer on the oral mucosa, as suggested by GIERTSEN et al. (3). Experiments described by GJERMO et al. (21) showed that
after-rinses with acidified water of pH 3 subsequent to a chlorhexidine rinse of neutral pH only eliminated about 3 of the clitiical effect.. This supports the contention that primary phosphate {and probably also sulfate) are significant binding sites for chlorhexidine. The residual effect must be assumed to be due to receptor sites which were still charged at pH 3. The same conclusion can be drawn from the experiments showing that an after-rinse with an aquotjs solution of SnFj (pH 3) did not reduce the effect of the initial chlorhexidine rinse at neutral pH (9). Furthermore, the high antibacterial effect of chlorhexidine at pH 3 indicates that primary phosphate groups are important binding sites for chlorhexidine on bacterial surfaces. The observed reduced clinical effect of chlorhexidine at pH 3 (applied as an afterrinse) described in previous papers, could indicate significance of both carboxyl and secondary phosphate as essential binding sites for chlorhexidine in the oral cavity. The present study eliminates secondary phosphate as a potential binding site for chlorhexidine, since the cHnical effects of chlorhexidine applied as solutions of pH 7 and pH 5.5 were comparable. An agent which is known to increase the antibacterial effect of chlorhexidine in vitro did not change the clinical efFect of chlorhexidine in vivo, demonstrating again that the antibacterial effect alone does not explain the clinical effect of chlorhexidine.
References 1. GJERMO P, BAASTAD KL, ROLLA G . The
plaque inhibiting capacity of 11 antibacterial compounds. J Periodont Res 1970; 5: 102-9. 2. GIERTSEN E , SOHEIE AA,, ROLLA G . Inhibition
of plaque formation and plaque acidogenicity by zinc and chlorhexidme combinations. Scand J Dent Res 1988; 96: 541-50. 3. GIERTSEN E , SCHEIE AA, ROLLA G . Antimi-
crobial and antiplaque effects of a chlorhexidine and Triton X-100 combination. Scand J Dent Res 1989; 97: 233-^-1. 4. ROLLA G , LOE H , SCHIOTT CR. Retention of
chlorhexidine in the human oral cavity. Arch Oral Biol 1971; 16: 1109-16. 5. Bo'NEsvoLL P, GJERMO P. A comparison between chlorhexidine and some quaternary ammonium compounds with regard to retention, salivary concentration and plaque-inhibiting efFect in the buman oral cavity after mouchrinses. Arch Oral Biol 1978; 23: 289-94. 6. ROLLA G , MELSEN B . On the mechanism of
the plaque inhibition by chlorhexidine. J Dent Res 1975; 54: Spec Iss B: 57-62. 7. GJERMO
P, BONESVOLL
P, HJELJOKD
LG,
ROLLA G . Influence of variation of pH of chlorhexidioe mouth rinses on oral retention and plaque-inhibiting effect. Caries Res 1975; 9: 74-^2. 8. LoEWY AG, SiEKEVrrz P. Cell structure and function. 2 ed. New York: Holt, Rinehart and Winston, 1969; 113. 9. WAALER SM, ROLLA G . Plaque inhibiting ef-
fecl of combinations ofchlorhexidine and the metal ions zinc and tin. Acta Odontol Scand 1980; 38: 213-7. 10. WAALER SM, ROLLA G . Effect of cblorbexi-
dine and lanthanum on plaque formation. Scand J Dent Res 1983; 91: 260-2. 11. EMBERY G . Tbe role of anionic glyco-conjugateSj particularly sulfated glycoproteins in relation to the oral cavity. In: KLEINBERG I, ELLISON SA, MANDEL ID, eds. Saliva and dental
caries. New York; IRL Press, 1979; 105-11. 12. DOYLE RJ. How cell walls of Gram-positive bacteria interact with metal ions.. In: BEVERIDGE T J , DOYLE RJ, eds. Metal ions and bacteria. New York: John Wiley and Sons, 1989; 275-94. 13. WINTER M R G , ROLLA G , WHITE AJ. The
IN VIVO MEGHANISMS OF CHLORHEXIDINE eiTect of inorganic cadons on dental plaque. In: LEACH SA, ed. Dental plaque and surface interactions in the oral cavity. London: I R L Press, 1980; 211-23. 14. BROWN M R W , RtcHAUBS RME. Effect of ethylenediamine tetraacetate on the resistance of Pseudomonas aeruginosa lo antibacterial agents. Mature 1965; 207: 1391-3. 15. SILNESS J, LOE H . Periodontal disease in pregnancy. II. Correlation between oral hygiene and periodontal conditions. Acta Odontol Scand 1964; 22: 121-35. 16. AINAMO J, SJOBLOM M , AINAMO A, TIANINEN
L. Growth of plaque while chewing sucrose and sorbitol flavored gum. J Clin Periodontol 1977; 4: 151-60. 17. MouTON C, ScHEiNiN A, MAKINEN KK. Effect
427
on plaqne of a xylitoi-containing chewinggtim. A clinical and biochemical study. Ada Odontol Scand 1975; 33: 33-40. 18. SjOBLOM M, AiNAMO' A, AiNAMO J. Antimicrobial effect of four different toothpastes. Scand 3 Dent Res 1976; 84: 377-80. 19. GARLSSON J , EGELBEEG J. Effect of diet on early plaque formation in man. Odontol Revy 1965; 16: 112-25. 20. NEUMAN WF,
NEUMAK M W . The chemical dy-
rtamics of bone mineral. Chicago: University of Chicago Press, 1958; II. 21. GJERMO P, BONESVOLL P, ROLLA G .
Rela-
tionship between plaqtie-inhibiting effect and retention of cblorhexidine in the human oral cavity. Arch Oral Biol 1974; 19: 1031-4.
Waaler SM: Further in vivo studies on the plaque-Inhibiting effect of chlorhexidine and its binding mechanisms, Scand J Dent Res 1990; 98: 422-7. Abstract — The aim of the present study was to investigate the relative importance of chemical groups in the oral cavity which bind chlorhexidine of clinical signilicance. This was performed by climcal experiments where a test panel of 11 individuals rinsed with chlorhexidine at pH 7, pH 5.5, and pH 3. Rinsing with 0.02 M EDTA was performed prior to the chlorhexidine rinse, because this procedure has been shown to enhance the antibacterial effect of chlorhexidine in vitro. However, pre-dnses with EDTA did not increase the clinical effect of chlorhexidine. Bacteriologic tests showed comparable andbacterial effect of chlorhexidine at pH 7 and pH 3. Mouthrinses of pH 7 and pH 5.5 had comparable plaque-inhibiting effects, whereas the effect of a rinse of pH 3 was not different from that of the placebo. This finding is iiiteipreted to suggest that primary phosphate (pK about 2) is an important receptor site for chlorhexidine, whereas secondary phosphate (pK = 7.0) is of negligible importance. Previous studies have clearly shown that phosphate binds chlorhexidine. The lack of clinical effect at pH 3 is probably due to precipitation of salivary proteins by hydrogen ions, making primary phosphate groups unavailable for interaction with chlorhexidine. Key words: chlorhexidine; clinical study; plaque inhibition. Department of Pedodontics and Garies Prophylaxis, Dental Faculty, Geitmyrsveien 71, 0455 Oslo 4, Norway. Accepted for publication 20 February 1990.
The bis-biguanides, and in particular chlorhexidine, are unique in their antiplaque effeet among all the antibacterial substances that have been tested (1). No correlation between andbacterial acdvity and plaqueinhibiting effect was evident for the substances tested (I). It has recently been detnoostrated that combinations of chlorhexidine
and other antibacterial agents which enhanced the antibacterial activity, did not necessarily itnprove their clinical effect (2, 3). Much research has been perfortned concerning the mechanisms by which cblorhexidine exerts its effect, but many aspects are still not well understood, The retention (the substantivity) was first
IN VIVO MECHANISMS OT CHLORHEXIDINE assumed to be the crucial property of chlorhexidine (4). However, it was later observed that quaternary ammonium compounds were retained in the mouth to a similar and even. higher degree than chlorhexidine, but that they exhibited scarcely any antiplaque effect (5). It was later suggested that an ion exchange mechanism was involved. Chlorhexidine was thought to be retained and to displace calcium from its binding sites in the mouth during rinsing procedures, and was subsequently displaced by calcium when the relative concentration of this ion increased (6). The sites aivailable for binding have been suggested to be mainly carboxyl groups (7). This was based on the fitiding that both clinical effect and retention were markedly reduced when chlorhexidine mouthrinses at pH 3 was used, this value coinciding approximately with the pH at which carboxyl groups lose their charge (8). However, results from previous experiments (9, 10) indicated that phosphate groups (and probably also sulfate groups) provide clinically significant binding sites for chlorhexidine in the mouth. Mucins and bacterial surfaces provide bound phosphate and sulfate groups (11,12). This suggestion was based on the observations that lanthanum ions, with their high a.ffmity for phosphate groups (13) were able to block receptor sites for chlorhexidine in vivo. Sitnilar effects were observed for Sn^ * and Zn"*^ *, which also have higher affinity for phosphate than for carboxyl (9, 10). The aim of the present study was to further investigate the significance of bound phosphate as receptor sites for chlorhexidine in the mouth. Furthermore, in vitro studies (14) have shown that the chelating agent EDTA enhances the antimicrobial actiwty of chlorhexidine, probably by displacing cations on the bacterial surfaces, making these sites more easily available for chlorhexidine molecules. Thus, the second aim of the study was to investigate whether pre-rinses with EDTA increased the plaque-inhibiting effect exerted by chlorhexidine.
423
Material and methods The present study comprised two experiments.
EXPERIME_NT 1 A plaque inhibition study, the effect of pre-rinses with
EDTA - Eleven individuals volunteered for the experiment and informed consent was obtained. The group was mixed with respect to dental health, and consisted of clinical and preclinical dental students, dental staff, dental hygienists, and persons with no professional relation to dentistry'. The age range was between 19 and 56 yr. Prior to each test period their teeth were scaled and polished. Each participant was given written instructions on how to perform the test regimens. For each test period of 4 days, no mechanical or other oral hygiene was allowed. Sucrose-contasnitig chewing gum was used by each participant for 5 min six times daily. The respective test substances were used as aqueous solutions and a 10 ml mouthrinse "w^as used for 1 min twice daily. At the end of each test period, the plaque score was recorded, using the Plaque Index (Pl.l) of SILNESS & LOE (15).
The study was carried out according to a double-blind cross-over design (1'6—18). During the study, all participants tested the following rinsing regimens: water (control), 0.02 M EDTA pH 7, pre-rinses with water or EDTA followed by rinses with 0.075% chlorhexidine acetate pH 7. The relatively low concentration of chlorhexidine was chosen to make any improvement in effect easily demonstrated. Statistical analyses was carried out using Student's ^-test for paired data.
EXPERIMENT 2 Bacteriologic test: Antibacterial activity of chlorhexidine acetate at pH 7 or pH 3 — Streptococcus sanguis ATCC
10556 (NGTC 7863) grown in Brain Heart Infusion broth (BHI, Difco Laboratories, Detroit, Michigan, USA) was collected in exponential phase by centrifugation (9000^, 4''C). The bacteria were washed once in 10' mM Hepes buffer (Sigma), pH 7, and the optical density was adjusted to 1.5 (680 nm, Uvicon 810 Kontron Spectro-
424
WAALER Table 1
Mean Plague Index (PLI) values obtained after rinsing with cklorkexidine (CH) with and without pre-rinses with EDTA. Both water and EDTA alone were included as controls (n = ll)
Mean PLI SD
H,O
EDTA GHpH = 7
H,O G H p H == 7
1.25 (0.34)
0.34 (0.22)
0.39 (0.19)
EDTA 1.14 (0.37)
photometer) corresponding to 0.4 mg protein/ml Results cell SHspension (BCA Protein Reagent, Pierce EXPERIMENT 1 Chemical Company, Rockford, IL, USA). Susceptibility tests were performed by exposing the bacte- The results from the plaque inhibition study ria to serial dilutions of chlorhexidine acetate employing EDTA pre-rinses are shown in (GH) in 10 mM Hepes buffer at pH 7 or pH 3. Titrations were performed in sterile microdter Table 1. The use of a 0.02 M EDTA solution prior trays (Dynatech, Essex, England/Greiner, Niirtingen. West Germany). The fmal concentration to the chlorhexidine solution did not enhance of CH in the first well was 0.1% (1.47 mM) and the clinical effect of chlorhexidine signifithe dilution coefficient was 0.625. The bacterial candy (P = 0.42). The effect of using EDTA inoculum was 75 |il. After 5 min exposure to GH, alone was not statistically different frotn the 10 jxl aiiqnots from each well were transferred to water control (P=0.40). Chlorhexidine used separate wells containing 150 ^il fresh BHI broth after an EDTA or a water pre-rinse gave a in a second microtiter tray. Erom each of these marked plaque itihibition compared with the wells 10 jil aliquots. were transferred to blood agar control, as could be expected (F=0.00). plates (Golumbia Agar Base with 7% human blood). The second microdter trays and the blood agar plates were incubated at 37°C for 20 h. Andbacterial activity was assessed by comparEXPERIMENT 2 ing growth of CH-exposed and non-exposed bacteBacteriologic test — CH had the same antibacria. Growth was assessed by OD measurements of each well in the microtiter tray (MR 710 Mi- terial effect at pH 7 and pH 3. At both pH 7 and pH 3, exposure of Strep, sanguis 10556, croplate Reader, Dynatech.) and by scoring bacteto GH for 5 min reduced subsequent bacteririal growth on the blood agar plates. Gells exposed to 10 mM Hepes buffer pH 7 or al growth in BHI broth as well as on blood pH 3 without addition of GH were included as agar plates at concentrations down to controls. The assay was performed in duplicate 0.0037%. Exposure of the cells to Hepes bufand repeated twice. fer without CH at pH 3 for 5 min had no Plague inhibition study. The effect ef chlorhexidine effect on subseqtient bacterial growth, mouthrinses al pH 7, 5.5, and 3 — Chlorhexidine solu- whereas prolonged incubation markedly retions at a concentration of 0.075% chlorhexidine duced subsequent growth in BHI broth and were adjusted to pH 3 and pH 5.5 with acetic on blood agar plates. acid. These acidified chiorhexidine solutions were Plaque inhibition study - The results of the used as mouthrinses. The study was otherwise perstudy employing chlorhexidine rinses of pH formed as described above, involving the same test subjects as in Experiment 1. Statistical analyses 3, 5.5, and 7 are shown in Table 2. The was carried out using Student's t-test for paired chlorhexidine solution of pH 3 showed no data. antiplaque effect, the PLI values were not
IN VIVO MECHANISMS OF CHLORHEXIDINE
425
Table 2 Mean Plaque Index (PLI) values after cldorkexidine (CH) mouthrinses at differentpH (a
Mean Pl.I SD
CH pH = 3
CHpH = 5.5
CHpH = V
1.29 (0.38)
0.37 (0,.25)
0.39 (0.19)
1.25 (0,34)
tibacterial acdvity was tnaintained at pH 3, and the loss of clinical effect thus seemed to be dependent on factors other than antibacterial activity, confirming the findings of GJERMO et al. (7). These authors reported that the clinical effect was almost eliminated at pH 3, and they suggested that this was due Discussion to carboxyl groups being the major receptor Sucrose-induced plaque (19) was used as a sites, because these groups had lost their negmodel ill the present study, since this was adve charge at this pH. judged to be relevant for the clinical situaTbe present study demonstrated a total loss tion. The standardized sucrose exposure was of effect when the chlorhexidine at pH 3 also thought to eliminate the effect of any was used. However, in the light of previous major dietary differences in the test panel. studies in our laboratory, wbich showed that The study showed that the use of EDTA •J to -J of the clinically significant binding prior to a chlorbexidine moutbrinse did not sites in the oral cavity may be phosphate increase the plaque-inbibiting effect in vivo, groups (9, 10), another interpretation than although EDTA has been found to enhance that suggested by GJERMO et al. (7) may seem the antibacterial effect of chlorhexidine in justified. The suggestion that phosphate vitro (14). It was thus confirmed that the groups were important was based on the antibacterial effect does not necessarily coin- observation that Zn++ and in particular cide with the clinical effect. However, it may ,Sn'^* and La*"*^ were able to block the be that the lack of effect by EDTA was due receptor sites for chlorhexidine and reduce to the low concetitration of EDTA used. The its clinical effect. When rinsing with chloruse of a 0.2 M EDTA solution was tried in hexidine at pH 3, it appears conceivable that a pilot experiment by the author, but the the hydrogen ions cause precipitation of the experiment was discontinued due to develop- macrotnolecules which bind chlorhexidine at ment of hypersensitivity of the teeth. The neutral pH. These molecules contain pripart of tbe study concerning chlorhexidine mary phosphate (and probably sulfate) solutions at difTerent pH values showed that groups, and the loss of chnical effect of chlorthe aotiplaque efTect of chlorhexidine was hexidine at pH 3 may not necessarily be not decreased when lowering the pH of the caused by lack of charged binding sites, but solution from 7.0 to 5.5. This may be inter- rather by the precipitation of salivary preted as lack of significance of secondary proteins by hydrogen ions,, a reaction which phosphate (HPO4') as a binding site for the renders the potential binding sites for chlorclinical effect of chlorhexidine. Chlorhexi- hexidine unavailable. Such reactions have dine at pH 3 did not show an}' andplaque previously been discussed by NEBMAN & efTect. Bacteriologic tests sbowed that full an- NEUMAN (20). It can easily be demonstrated statistically different from the placebo [P = 0.81). The solution of pH 5.5 ga\'e no significantly different results when compared to the solution of pH 7 (F=0.80). Both of tbese coocentrations thus gave full clinical effect.
426
WAALER
in vitro that water acidified by acetic acid to pH 3 causes an immediate precipitation of salivary proteins when applied in clinically relevant proportions to whole .saliva (about 10:1). It may also he that the acid-precipitated proteins could be rapidly lost hy expectorations, thus giving a more rapid turnover of the mucin layer on the oral mucosa, as suggested by GIERTSEN et al. (3). Experiments described by GJERMO et al. (21) showed that
after-rinses with acidified water of pH 3 subsequent to a chlorhexidine rinse of neutral pH only eliminated about 3 of the clitiical effect.. This supports the contention that primary phosphate {and probably also sulfate) are significant binding sites for chlorhexidine. The residual effect must be assumed to be due to receptor sites which were still charged at pH 3. The same conclusion can be drawn from the experiments showing that an after-rinse with an aquotjs solution of SnFj (pH 3) did not reduce the effect of the initial chlorhexidine rinse at neutral pH (9). Furthermore, the high antibacterial effect of chlorhexidine at pH 3 indicates that primary phosphate groups are important binding sites for chlorhexidine on bacterial surfaces. The observed reduced clinical effect of chlorhexidine at pH 3 (applied as an afterrinse) described in previous papers, could indicate significance of both carboxyl and secondary phosphate as essential binding sites for chlorhexidine in the oral cavity. The present study eliminates secondary phosphate as a potential binding site for chlorhexidine, since the cHnical effects of chlorhexidine applied as solutions of pH 7 and pH 5.5 were comparable. An agent which is known to increase the antibacterial effect of chlorhexidine in vitro did not change the clinical efFect of chlorhexidine in vivo, demonstrating again that the antibacterial effect alone does not explain the clinical effect of chlorhexidine.
References 1. GJERMO P, BAASTAD KL, ROLLA G . The
plaque inhibiting capacity of 11 antibacterial compounds. J Periodont Res 1970; 5: 102-9. 2. GIERTSEN E , SOHEIE AA,, ROLLA G . Inhibition
of plaque formation and plaque acidogenicity by zinc and chlorhexidme combinations. Scand J Dent Res 1988; 96: 541-50. 3. GIERTSEN E , SCHEIE AA, ROLLA G . Antimi-
crobial and antiplaque effects of a chlorhexidine and Triton X-100 combination. Scand J Dent Res 1989; 97: 233-^-1. 4. ROLLA G , LOE H , SCHIOTT CR. Retention of
chlorhexidine in the human oral cavity. Arch Oral Biol 1971; 16: 1109-16. 5. Bo'NEsvoLL P, GJERMO P. A comparison between chlorhexidine and some quaternary ammonium compounds with regard to retention, salivary concentration and plaque-inhibiting efFect in the buman oral cavity after mouchrinses. Arch Oral Biol 1978; 23: 289-94. 6. ROLLA G , MELSEN B . On the mechanism of
the plaque inhibition by chlorhexidine. J Dent Res 1975; 54: Spec Iss B: 57-62. 7. GJERMO
P, BONESVOLL
P, HJELJOKD
LG,
ROLLA G . Influence of variation of pH of chlorhexidioe mouth rinses on oral retention and plaque-inhibiting effect. Caries Res 1975; 9: 74-^2. 8. LoEWY AG, SiEKEVrrz P. Cell structure and function. 2 ed. New York: Holt, Rinehart and Winston, 1969; 113. 9. WAALER SM, ROLLA G . Plaque inhibiting ef-
fecl of combinations ofchlorhexidine and the metal ions zinc and tin. Acta Odontol Scand 1980; 38: 213-7. 10. WAALER SM, ROLLA G . Effect of cblorbexi-
dine and lanthanum on plaque formation. Scand J Dent Res 1983; 91: 260-2. 11. EMBERY G . Tbe role of anionic glyco-conjugateSj particularly sulfated glycoproteins in relation to the oral cavity. In: KLEINBERG I, ELLISON SA, MANDEL ID, eds. Saliva and dental
caries. New York; IRL Press, 1979; 105-11. 12. DOYLE RJ. How cell walls of Gram-positive bacteria interact with metal ions.. In: BEVERIDGE T J , DOYLE RJ, eds. Metal ions and bacteria. New York: John Wiley and Sons, 1989; 275-94. 13. WINTER M R G , ROLLA G , WHITE AJ. The
IN VIVO MEGHANISMS OF CHLORHEXIDINE eiTect of inorganic cadons on dental plaque. In: LEACH SA, ed. Dental plaque and surface interactions in the oral cavity. London: I R L Press, 1980; 211-23. 14. BROWN M R W , RtcHAUBS RME. Effect of ethylenediamine tetraacetate on the resistance of Pseudomonas aeruginosa lo antibacterial agents. Mature 1965; 207: 1391-3. 15. SILNESS J, LOE H . Periodontal disease in pregnancy. II. Correlation between oral hygiene and periodontal conditions. Acta Odontol Scand 1964; 22: 121-35. 16. AINAMO J, SJOBLOM M , AINAMO A, TIANINEN
L. Growth of plaque while chewing sucrose and sorbitol flavored gum. J Clin Periodontol 1977; 4: 151-60. 17. MouTON C, ScHEiNiN A, MAKINEN KK. Effect
427
on plaqne of a xylitoi-containing chewinggtim. A clinical and biochemical study. Ada Odontol Scand 1975; 33: 33-40. 18. SjOBLOM M, AiNAMO' A, AiNAMO J. Antimicrobial effect of four different toothpastes. Scand 3 Dent Res 1976; 84: 377-80. 19. GARLSSON J , EGELBEEG J. Effect of diet on early plaque formation in man. Odontol Revy 1965; 16: 112-25. 20. NEUMAN WF,
NEUMAK M W . The chemical dy-
rtamics of bone mineral. Chicago: University of Chicago Press, 1958; II. 21. GJERMO P, BONESVOLL P, ROLLA G .
Rela-
tionship between plaqtie-inhibiting effect and retention of cblorhexidine in the human oral cavity. Arch Oral Biol 1974; 19: 1031-4.
