Short Communication

Ultracytochemical localization of hydrogen peroxideproduction by dental plaque bacteria

Alejandro Oyarzu n a, *, Patricio Smithb

aBiochemistry and Oral Biology Unit, Faculty of Odontology, University of Chile, ChilebDepartment of Periodontics, Faculty of Odontology, University of Chile, Casilla 1903, Santiago, Chile

Accepted 27 April 1998

Abstract

This study looked for evidence of in vitro hydrogen peroxide (HP) synthesis in human dental plaque, using anultracytochemical technique that included incubation in a CeCl3-rich medium. Supragingival dental plaque was

obtained from periodontally healthy individuals and subgingival dental plaque from human periodontal diseasesites. Speci®city of the cytochemical reaction was demonstrated using catalase (as HP scavenger). HP productionwas indicated by an electron-dense precipitate localized at the cell envelope of unidenti®ed Gram-positive andGram-negative bacteria in both supra- and subgingival samples. The ultracytochemical reaction localized the HP

production primarily to the plasma membrane and periplasmic space. # 1998 Elsevier Science Ltd. All rightsreserved.

Keywords: Hydrogen peroxide; Periodontal disease; Bacteria

1. Introduction

The production of hydrogen peroxide by bacteria

was ®rst reported by McLeod and Gordon in pneumo-

coccus (1923). b-haemolytic streptococci (Hadley et al.,

1941) and lactobacilli (Wheather et al., 1952; Dahiya

and Speck, 1968) also produce the same compound.

Within the mouth, the production of hydrogen per-

oxide appears to be restricted to streptococci and

occurs when glucose is catabolized in the presence of

oxygen (Kraus et al., 1957; Holmberg and Hallander,

1973). Streptococcus sanguis (I and II) and Strep.

mitior are strong HP producers, but neither has cata-

lase or peroxidase activity (Ryan and Kleinberg, 1995).

Aside from Gram-positive bacteria, many Gram-nega-

tive micro-organisms are involved in oxygen-derived

free-radical synthesis. Bacteria such as Campylobacter,

Neisseria, Pseudomonaa, Eikenella and Actinobacillus

have a positive oxidase activity (Baron et al., 1995).Porphyromonas gingivalis also expresses superoxide dis-

mutase in both aerated and anaerobic conditions(Amano et al., 1990). Most knowledge of antioxidantenzymes and oxygen-derived free-radical synthesis by

oral micro-organisms comes from in vitro microbiolo-gical and biochemical studies (Ryan and Kleinberg,1995). Our purpose now was to demonstrate synthesisof hydrogen peroxide by human dental-plaque bacteria

under in vitro conditions using an ultracytochemicalmethod.

2. Materials and methods

Eight supragingival plaque samples were obtained

from four periodontally healthy individuals whose agesranged from 40 to 50 years. Sixteen subgingivalsamples were obtained from eight patients with adult

periodontal disease. After careful elimination of supra-gingival plaque, subgingival samples were obtainedwith a sterile Gracey curette at the bottom of the peri-

Archives of Oral Biology 43 (1998) 907±910

0003-9969/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved.

PII: S0003-9969(98 )00062-4

ARCHIVESOFORALBIOLOGY

PERGAMON

* Corresponding author. Fax: 0056-2-777-6062; E-mail:

aoyarzun@abello.dic.uchile.cl.

Abbreviations: ATZ, 3-amino-1, 2,4-triazol.

odontal pocket (5±7 mm) on the vestibular or lingualsurfaces of upper or lower incisors. Probing depths

were measured with a Michigan periodontal probeafter the samples had been taken. The criteria forpatient selection were: no history of relevant systemic

disease, absence of pharmacological or periodontaltherapy during the last 6 months, presence of advancedadult periodontal disease (diagnosed by clinical and

radiographic examination), pocket depth greater than5 mm at the site of sample collection, and the presenceof bleeding on probing.

Samples were immediately placed in an Eppendorftube with 2 ml of a stock bu�er solution that con-tained 140 mM NaCl, 1 mM CaCl2, 5 mM KCl,15 mM Hepes and 5.6 mM glucose, pH 7.4, at 48C for

15 min (Ho�stein et al., 1988), and then centrifuged at1000 G for 5 min. Pellets were then resuspended infresh stock bu�er for 5 min at room temperature and

centrifuged again. Pellets obtained from the second 5-min wash were resuspended in 2 ml of an incubationmedium consisting of the above stock bu�er with

10 mM ATZ, 10 mM NaN3 and 1 mM CeCl3 (Hirai etal., 1991). The pellets were incubated for 30 min at378C in this medium and then washed in stock bu�er.

Pellets were ®xed in 2% glutaraldehyde in 0.12%Milloning phosphate bu�er, pH 7.4, at 48C for 1 h andpost®xed in 1% osmium tetroxide. Fixed pellets wereincubated in a cold 0.5% uranyl acetate solution for

20 min, dehydrated in a graded series of ethanols andembedded in Epon 812. For light microscopy examin-ation, semithin 1-mm sections were stained with tolui-

dine blue. Ultrathin (60 nm) sections mounted oncopper grids were observed in a Zeiss AM-109 electronmicroscope operated at 80 kV. Counterstaining with

uranyl acetate and lead citrate was not done. CeCl3was omitted during the processing of negative controlsamples. The speci®city of the cytochemical reactionwas tested by adding 0.2 mg/ml catalase (bovine liver,

C10; Sigma) to bu�er stock washes and incubationmedia in six samples. ATZ and NaN3 were omitted(catalase inhibitors).

3. Results

Supragingival plaque samples showed a mixed ¯oracomposed predominantly of Gram-positive cocci and

rods. The cytochemical reaction was frequentlydetected at the plasma membrane and cell wall ofGram-positive cocci. Depending on the plane of sec-

tion, some bacilli showed electron-dense cytochemicalreaction all over (Fig. 1). We also noted a spot-likereaction localized in the cytoplasm of several Gram-

positive cocci. This reaction was associated with fold-ing, trilaminar-membrane units resembling a bacterialmesosome (Fig. 2).

All subgingival samples contained a large number ofcocci, rods, ®laments and spirochaetes scattered in a

®ne granular matrix (Fig. 3). The cytochemical reac-tion was observed in some cells as a continuous elec-

tron-dense material distributed over the entire cellenvelope (Fig. 4) and in others as an interrupted pre-

cipitate. At higher magni®cations, the HP productionwas detected in the internal plasma membrane and

periplasmic space of Gram-negative bacteria (Fig. 5).When catalase (0.2 mg/ml) was added to bu�er stock

washes and incubation media, the cytochemical reac-tion was completely inhibited (Fig. 6).

Fig. 2. Supragingival plaque sample showing a short rod with

hydrogen peroxide deposition at the periphery of the cell

envelope (arrows). Cocci with hydrogen peroxide deposited in

a cytoplasmic spot-like structure (arrowhead). �32,288;bar=0.3 mm.

Fig. 1. Supragingival plaque sample showing a mixed ¯ora

principally composed of Gram-positive cocci and short rods.

Arrowheads show hydrogen peroxide-producing bacteria.

A. OyarzuÂn, P. Smith / Archives of Oral Biology 43 (1998) 907±910908

4. Discussion

Several investigations have detected structural evi-dence for the production of hydrogen peroxide in

eukaryotic cell populations (Briggs et al., 1975; Hiraiet al., 1991; Warren et al., 1989; Vissers et al., 1985;

Ohno et al., 1982; Ho�stein et al. 1988). In those stu-dies, production of hydrogen peroxide was visualized

as an electron-dense cerium perhydroxide precipitatedistributed along the plasma membrane of the cells.

The morphological appearance and distribution ofhydrogen peroxide-producing sites in our work are inaccordance with the above-mentioned studies, as the

electron-dense precipitate was observed in associationwith plasma membrane units in both Gram-negative

and Gram-positive bacteria. The speci®city of the reac-tion was con®rmed when cerium perhydroxide depo-

sition was inhibited by adding catalase to theincubation media. Our study demonstrates the applica-bility of this ultracytochemical method using cerium

ions for the morphological analysis of oxygen-derivedfree radical synthesis in human dental plaque.

References

Amano, A., Shizukuishi, S., Tamagawa, H., Iwakura, K.,

Tsunasawa, S., Tsunemitsu, A., 1990. Characterization of

superoxide dismutases puri®ed from either anaerobically

maintained or aerated Bacteroides gingivalis. J Bacteriol.

172, 1457±1463.

Baron, E., Weiisfeld, A., Fuselier, P., Brenner, D., 1995.

Classi®cation and identi®cation of bacteria. In: P. Murray,

E. Baron, F. Tenover, R. Yolken (Ed.). Manual of Clinical

Microbiology. ASM Press, Washington, DC, pp. 249±264.

Fig. 5. Gram-negative bacteria phagocytosed by a polymor-

phonuclear neutrophil (N). The cytochemical reaction covers

the inner plasma membrane and periplasmic space. The outer

plasma membrane is free of hydrogen peroxide (arrows).

Fig. 4. Higher magni®cation of inset shown in Fig. 3. The

hydrogen peroxide-producing bacteria show an irregular elec-

tron-dense precipitate that partially hides the cell envelope

(arrowheads). �35,055; bar=0.3 mm.

Fig. 3. Pleomorphic group of micro-organisms in which

hydrogen peroxide-producing bacteria are demonstrated by

an electron-dense precipitate located at the cell envelope

(inset). BP, bacterial plaque. �5,244; bar=2 mm.

Fig. 6. Note the inner and outer plasma membrane and peri-

plasmic space free of the electron-dense precipitate (arrow-

A. OyarzuÂn, P. Smith / Archives of Oral Biology 43 (1998) 907±910 909

Briggs, R., Drath, D., Karnovsky, M., 1975. Localization of

NADH oxidase on the surface of human polymorphonuc-

lear leukocytes by a new cytochemical method. J Cell Biol.

67, 566±586.

Dahiya, R.S., Speck, M.L., 1968. Hydrogen peroxide for-

mation by lactobacilli and its e�ect on Staphylococcus aur-

eus. J Dairy Sci. 58, 1568±1572.

Hadley, F.P., Hadley, P., Leather, W.W., 1941. Variation in

peroxide production by beta hemolytic streptococci. J infect

Dis. 68, 264±277.

Hirai, K., Moriguchi, K., Wang, G., 1991. Human neutro-

phils produce free radicals from the cell-zymosan interface

during phagocytosis and from the whole cell plasma mem-

brane when stimulated with calcium ionophore A23187.

Exp Cell Res. 194, 19±27.

Ho�stein, S.T., Gennaro, D.E., Meunier, P.C., 1988.

Cytochemical demonstration of constitutive hydrogen per-

oxide production by macrophages in synovial tissue from

rats with adjuvant arthritis. Am J Pathol. 130, 120±125.

Holmberg, K., Hallander, H.O., 1973. Production of bacteri-

cidal concentrations of hydrogen peroxide by Streptococcus

sanguis. Archs oral Biol. 18, 423±434.

Kraus, F.W., Nicherson, J.F., Perry, W.I., Walker, A.P.,

1957. Peroxide and peroxidogenic bacteria in human saliva.

J Bacteriol. 73, 727±735.

McLeod, J.W., Gordon, J., 1923. Production of hydrogen per-

oxide by bacteria. J Path Bact. 26, 499±506.

Ohno, Y., Hirai, K., Kanoh, Uchino, H., Ogawa, K., 1982.

Subcellular localization of H2O2 production in human neu-

trophils stimulated with particles and an e�ect of cytochala-

sin-B on the cells. Blood 60, 253±260.

Ryan, C.S., Kleinberg, I., 1995. Bacteria in human mouths

involved in the production and utilization of hydrogen per-

oxide. Archs oral Biol. 40, 753±763.

Vissers, M., Day, W., Winterbourn, C., 1985. Neutrophils

adherent to nonphagocytable surface (glomerular basement

membrane) produce oxidants only at the site of attachment.

Blood 66, 161±166.

Warren, J., Kunkel, R., Simon, R., Johnson, K., Ward, P.,

1989. Ultrastructural cytochemical analysis of oxygen rad-

ical-mediated Immunoglobulin A immune complex induced

lung injury in the rat. Lab Invest. 60, 651±658.

Wheather, D.M., Hirsch, A., Mattick, A.T.R., 1952. Possible

identity of ``lactobacillin'' with hydrogen peroxide produced

by lactobacilli. Nature 170, 623±624.

A. OyarzuÂn, P. Smith / Archives of Oral Biology 43 (1998) 907±910910