Otat Microbiol Ittmmtiol 1992: 1: 280-284
Demonstration of a bimodal coaggregation reaction between Porphyromonas gingivalis and Treponema denticola
D. Grenier Departement de Sante Buccale, Faculte de Medecine Dentaire, Universite de Montreal, Montreal, Quebec, Canada
Grenier D. Demonstration of a bimodal coaggregation reaction between Porphyromonas gingivalis and Treponema denticola. Oral Microbiol Itntmtttol 1992: 7: 280-284, This study detnonstrates a strong coaggregation reaction between 2 suspected periodontopathogens; Porpliyrotnotias gingivalis and Trcponettia dentieola. Other black-pigmented oral bacterial species tested did not coaggregate with T denticola. This specific interbacterial aggregation was bitnodal, since heating of both cell Key words: coaggregation; adherence: Tretypes was required to completely eliminate the reaction. The coaggregation reacponema denticoia: Porpbyromonas gingition occurred between pH 4 and 9. Under sotne conditions, arginine and Dvatis galactosamine were effective in preventing the coaggregation. The heat-sensitive Daniel Grenier, Groupe de Recherche en receptor on P. gingivalis was found to be loosely bound and could be released Ecologie Buccale, Ecole de Medecine Dentaiby a light ultrasonic treatment of the cells. It is suggested that the bacterial re, Universite Laval, Sainte-Foy (Quebec), Canada, G1K 7P4 interaction described may participate in the establishtnent of a potentially pathogenic subgingival plaque. Accepted for publication March 2, 1992
Nutnerous studies have shown an association between some forms of periodontal disease and increased numbers of spirochetes and gratn-negative anaerobic bacteria in affected sites (16, 17, 19). Several types of bacterial interactions, including coaggregation and nutritional relationships, are thought to be critical for the initial development and the establishment of a potentially pathogenic subgingival plaque (3, 5, 11, 18, 21). The evidence coticerning coaggregation of oral bacteria was first reported by Gibbons & Nygaard (4). Further studies have shown possible bacterial aggregation between gratn-positive and gratn-positive (2, 20), gratn-positive and gratn-negative (3, 13) as well as gramnegative and gram-negative bacteria (10, 12, 14). Different types of surface cotnponents including carbohydrates, fimbriae and outer tnetnbrane proteins have the ability to participate in coaggregation .reactions between oral bacteria (11). A second tnechanistn of bacterial interactions involves the production of growth-stitnulating factors which may contribute to bacterial successions iti periodontal sites (5, 18, 21). I recently
observed a tnutual, synergistic nu- viously described (15). All cultures were tritional relationship between Porphyro- incubated in an anaerobic chatnber (Nr monas gingivalis atid Treponetna dentieo- H2-CO2, 80;10;10) at 37^C. Black-pigla (D. Grenier, subtnitted). This ben- mented bacterial species were grown for eficial relationship appears to involve 1 day whereas T. denticola was cultiintitnate contact between cells of both vated for 7 days. bacterial species. The aim of this study was to characterize the nature of the Coaggregation assay coaggregation reaction between P. gingiBacterial cultures were harvested by valis and T. denticola. centrifugation at 10,000 xg for 15 tnin at 4 C. The cells were washed twice in Material and methods distilled water, and the final pellet was Bacterial strains and growth conditions suspended in distilled water to an abThe following bacterial strains were sorbance, at 660 nm, of 0.75 for the used in this study; T. denticola ATCC black-pigtnented bacterial species and 35405; P gittgivalis ATCC 33277, 381, 0.4 for T. denticola. The bacterial susHWllD-5; Porphyromonas asaccharo- petisions were stored at 4 C for up to 5 lytica ATCC 25260, BM4; Porpliyro- days. Bacterial cells showed no aptnotias etidodontalis ATCC 35406; Prevo- parent self-aggregation when grown and tella intertnedia 5W2; Prevotella tnel- prepared as above. The coaggregation aninogenica ATCC 25845; Prevotella mixture, in 12- by 75-mtn round-bottom loescheii ATCC 15930; Prevotella dentic- test tubes, consisted of 100 //I of T detiola ATCC 33185; and Bacteroides levii ticola, 100 /(I of the black-pigmented ATCC 29147. Growth of black-pig- bacterial species and 200 /d of phosmented bacterial species was carried out phate-buffered saline (PBS; 25 mM soin brain heart infusion broth (BBL diutn phosphate buffer [pH 6.8] conMicrobiology Systetns, Cockeysville, taining 75 tnM NaCl). The tubes were MD) containing hemin (10 //g/ml) and shaken for 5 min at room tetnperature vitatnin K (1 /zg/tnl). T. denticola was on a rocking platfortn mixer (Red-Rogrown in fiuid NOS medium as pre- tor, Hoefer Scientific Instruments, San
Coaggregation of P. gingivalis atid T. denticola Francisco, CA) operating at 150 RPM, and then scored according to a visual rating scale of 0 through 4-1- as previously described by Cisar et al. (2); 0, no visible aggregates in the cell suspension; 1+, small uniform aggregates in the cell suspension; 2-I-, definite aggregates easily seen but suspension retnained turbid; 3-I-, large aggregates which settled rapidly, leaving sotne turbidity in the supernatant fluid; 4-f, clear supernatant fiuid and large aggregates which settled immediately. Except where noted, the coaggregation experiments were done with P. gitigivalis 33277 and T. denticola 35405.
281
cose, D-tnannitol, L-rhatnnose, D-sorbitol, D-sucrose; D-galactosamine, Dglucosatnine; N-acetyl-D-galactosatnine, N-acetyl-D-glucosatnine, N-acetylneuratnin-lactose; EDTA.
Results Of the 8 black-pigmented bacterial species tested, only the 3 strains of P. gittgivalis possessed the ability to fortn large coaggregates with T. denticola cells (Fig. 1). The coaggregation reaction between P. gingivalis and T. dettticola gave Ultrasonic treatment of bacterial cells a tnaxiniutn score of 4-f over a pH P gingivalis and T. denticola cells were range of 5 to 8. Weaker coaggregations subtnitted to a light ultrasonic treat- (2-f) were observed at pH 4 and 9, ment in order to detach any loosely whereas no visible coaggregation ocbound structures or proteins. Briefiy, curred at pH 3 and 10. The coaggregathe bacterial suspensions were sonicated tion reaction was cotnpletely elitninated three times for 10 s (20% duty cycle. when both P. gitigivalis and T. denticola Sonic distneinbrator model 150, Artek cells were heated at 75 C for 30 tnin. Systems, Farmingdale, NY) and the The interbacterial aggregation still occells washed twice by centrifugation. curred wheti otily one cell type was subEffect of pH on the coaggregation reaction Cell suspensions were prepared in dis- tnitted to such heating (Fig. 1, Table 1). The coaggregation reaction between P. tilled water and the coaggregation assay A tnilder heat treatment (55°C for 30 tnin) of both bacterial suspensions did gitigivalis and T denticola was per- was perfortned as described earlier. not interfere with the coaggregation. fortned using the following buffers inCotnplete inhibition of coaggregation stead of PBS; 25 tnM citrate buffer (pH pH sensitivity of the binding components was also obtained by a proteolytic treat3, 4, 5), 25 tnM phosphate buffer (pH 6, 7), 25 tnM Tris hydrochloride buffer P gingivalis and T. dcttticola cells were ment of both cell types with proteinase (pH 8, 9) and 25 mM carbonate buffer suspended in 100 tnM citrate buffer at K (Table 1). With the exception of L-arginitie and (pH 10, 11). All buffers were supple- pH 3, as well as in 100 mM carbonate buffer at pH 12. After iticubation at D-galactosamine, other putative inhibimented with 75 mM NaCl. room temperature for 30 tnin, the cells tors tested did not affect the coaggregawere harvested by centrifugation, sus- tion between P. gingivalis and T. denticoHeat treatment and proteoiytic treatment pended in distilled water and the coag- la (Table 2). Arginine, at a final concenof bacterial celis gregation assay was carried out as de- tration of 150 tnM, cotnpletely eliminated the coaggregation between P. Cell suspensions of P. gittgivalis and T. scribed above. denticola were heated at 75°C for 30 min and used in the coaggregation assay. Samples of the cell suspensions were also incubated at 37 C for 3 h with an equal volume of proteinase K. (Sigtna Chetnical Co., St. Louis, MO; 0.5 tng/ml in PBS). At the end of the incubation period, the cells were harvested by centrifugation, and the pellets were washed twice in distilled water. Cell suspensions were prepared in distilled water and the coaggregation assay was then carried out as described above. Effect of putative inhibitors on the coaggregation reaction
Putative inhibitors of coaggregation reactions were incorporated in PBS and assayed for their ability to inhibit the coaggregation between P. gitigivalis atid T. denticola. Heat-treated and nontreated bacterial cells were tested. The following compounds, at final concentrations (in the coaggregation mixture) ranging frotn 50 to 250 tnM, were used; alanine, arginine, asparagine, cysteine, histiditie, leucine, lysine, proline, serine and threonine (all in the L-fortn); Dfructose, D-fucose, D-galactose, D-glu-
Fig. J. Visual coaggregation reactions between P. gingivalis and T denticola. A. P. gingivalis and T. denticola. B. Heat-treated P. gingivalis and T. denticola. C. P. gittgivalis and heat-treated T denlicola. D. Heat-treated P. gingivalis and heat-treated T. denlieola
Grenier
282
Table I. Effect of heat treatment and protease treatment on the coaggregation reaction between P. gingivalis and T denticola Bacteria Heat-treated P. gitigivalis
Heat-treated T. denlieola Coaggregation'
No No Yes Yes
No Yes No Yes
Proteasetreated P. gitigivalis
Proteasetreated T denticola
No No Yes Yes
No Yes No Yes
•'Scored as described methods
4-h 3-14-10
4-H 3-t3-f 0 in
Material
and
Incubation of T denticola cells at pH 12 resulted in a loss of their ability to coaggregate with heat-treated P. gitigivalis cells. However, these treated spirochetes could still react with nontreated P. gingivalis cells, indicating that the heat-sensitive binding cotnponent present on T. denticola was setisitive to high pH. A similar treattnent of P. gingivalis did not change the coaggregation profile of the bacteria. Treattnent of P. gingivalis at pH 3 elitninated all coaggregation reactions. Finally, treattnent of T. denticola cells at this low pH allowed coaggregation with non-treated P. gitigivalis but not with the heattreated cells.
Table 3. Effect of ultrasonic treatment of P. gingivalis on the coaggregation reaction with T. denticola Treatment of bacteria P. gingivalis No No Sonicated Sonicated
T dctiticota Coaggregation" No Heat-treated No Heat-treated
4+ 34440
Assay in the presence of 150 mM arginine** No No 0 Heat-treated No 3 4Sonicated No 3 4•'Scored as described in Material and methods. *" Final concentration in the coaggregation mixture
suggests that physical constraints due to the arginine may prevent the coaggregaThis study demonstrated a strong coag- tion tnediated by the second binding gingivalis and heat-treated or non- gregatioti reaction between 2 suspected cotnpotient on P. gingivalis. The absence treated T. denticola. No significant inhi- periodontopathogens; P. gitigivalis and of inhibition, caused by arginine, when bition by arginine was observed when T. denticola. The coaggregation was bi- heat-treated or sonicated P. gitigivalis heat-treated P. gitigivalis cells were modal, since heating of both bacterial cells were used, suggests that the heatmixed with non-treated T. dentieola. D- parttiers was required to eliminate the sensitive receptor is denaturated or regalactosatnine, at a high concentration reaction. The coaggregation of P. gin- tnoved, and that the resulting coaggre(250 tnM), was able to reduce but not givalis and T. denticola appears to be gation with T. denticola can thus be cotnpletely inhibit the coaggregation of highly specific, since a similar reaction mediated by the second binding compoP. gitigivalis with heat-treated or non- was not observed with other black-pig- nent present on the cell surface of P. treated T denticola. When heat-treated mented bacterial species, including P. gingivalis. The coaggregation between P. P gitigivalis cells were used in the coag- asaccharolytica and P. etidodontalis. Atigingivalis atid T. denticola was shown to gregation assay, no inhibition by D-gal- additional strain of T. denticola showed be reduced in the presence of D-galactoactosatnine was observed. also the ability to coaggregate with P. samine. The lack of inhibition produced A light ultrasonic treattnent of the gitigivalis (data not shown). by EDTA suggests that cations are not bacterial suspensions was done to deterThe schematic representation of the involved in the interaction. mine whether the binding components proposed bitnodal coaggregation reacThis is the first report describing the were firtnly or loosely attached to the tion between P. gitigivalis and T. dentico-coaggregation properties of T. denticola surface of the bacteria. No difference la is presented in Fig. 2. Two binding with P. gingivalis. Previous studies have was observed in the binding property components, one heat-sensitive and one already shown the ability of T denticola of sonicated T. denticola cells. However, heat-stable, are thought to be present to attach to epithelial cells (22, 23), sonication of P. gingivalis cells appeared on both bacterial species. The heat-sen- basetnent tnembrane cotnponents (7) to remove its heat-sensitive receptor, as sitive receptor on P. gitigivalis is loosely and to tnodified hydroxyapatite sursuggested by; i) the cotnplete loss of bound and can be released by a light faces (1). Extracellular deposits of acidic aggregation with heat-treated T. dentic- ultrasotiic treatment of the bacterial mucopolysaccharides on the surface of ola cells (as noted with heat-treated P cells. The attachment mediated by this T. denticola have been thought to pargitigivalis) and ii) the absence of inhi- receptor is inhibited by L-argitiitie. Sur- ticipate in the attachtnent to epithelial bition by arginitie for the coaggregatioti prisingly, no coaggregation between cells (22). However, prelitninary experireaction with non-treated T. dentieola non-treated P. gitigivalis and non- ments in my laboratory have indicated cells (as noted with heat-treated P. giti- treated T denticola cells occurred in the that the coaggregation property of T. givalis) (Table 3).' , . . . • presence of arginine. This observation denticola is not affected by treattnent of the cells with hyaluronidase, suggesting Tahle 2. Effect of L-arginine and D-galactosamine on the coaggregation t'eaction between P. that tnucopolysaccharides are not ingingivalis and T. denticola volved. Bac teria Recently, P. gingivatis was showti to Coaggregation'' in presence of b coaggregate with various strains of EuArginine (mM) D-galactosamine (mM)*" Heat-treated Heat-treated .sobacteriutn tiucleatutn (10, 12) and ActiP. gitigivalis T. denticola 0 50 150 0 125 250 notnyces viscosus (3). In the first case, No No 0 44242+ 44the reaction was prevented by lactose No Yes 34240 34• 3-f 14and related sugars or by heat treattnent Yes No 444434444434of the fusobacterium, whereas in the second case, the attachtnent was not in"Scored as described in Material and methods. ''Final concentration in the coaggregation mixture hibited by carbohydrates atid was otily Discussion
Coaggregation of P. gingivalis and T. denticola Heat sensitive
Heat stable 10.
11.
12. Fig. 2. Schematic representation of the proposed bimodal coaggregation reaction between P. gingivalis and T denlieola
partially reduced by heating the P. gin- cally reading the manuscript and T.-T. givalis cells. Hetnagglutinating activity Huynh for excellent technical assistof P. gitigivalis has been previously re- ance. This work was supported by the ported to be inhibited by arginine (6, 8). Medical Research Council of Canada This hemagglutinin and the P. gingivalis (grant no; DG-383-384-385). heat-sensitive binding cotnponent, detnonstrated in the present teport, could represent the satne tnolecule, since both are inhibited in a sitnilar fashion by arginine. Isogai et al. (9) have suggested References that P. gingivalisfimbriae,which cati be 1. Cimasoni G, McBride BC. Adherence of removed by sonication, tnay be responTreponema denticola to modified hysible for the adherence to epithelial cells. droxyapatite. J Dent Res 1987: 66: This adhesin tnay represent the P. gin1727-1729. givalis loosely bound, heat-sensitive 2. Cisar JO, Kolenbrander PE, Mclntire binding cotnponent observed in the EC. Specificity of coaggregation reacpresent study. Another black-pigtnented tions between human oral streptococci bacterial species, P. loescheii, has been and strains of Actinomyces viscosus or Actinotnvces naeslundii. Infect Immun previously shown to possess loosely at1979: 24: 742-752. tached adhesin molecules on its surface 3. Ellen RP. Schwarz-Faulkner S, Grove (25). It was also detnonstrated that 2 DA. Coaggregation among periodontal distinct types of fitnbriae-associated pathogens, emphasizing Bacteroides ginproteins on P. loescheii were responsible givctlis - Actinomyces viscosus cohesion for the coaggregation with Streptococon a saliva-coated mineral surface. Can cus sanguis and Actitiotnyces israelii. J Microbiol 1988: 34: 299-306. The coaggregation phenomenon de4. Gibbons RJ, Nygaard M. Interbacterial aggregation of plaque bacteria. Arch scribed in this study tnay be important Oral Biol 1970: 15: 1397-1400. for the establishment of a potentially 5. Grenier D, Mayrand D. Nutritional repathogenic subgingival plaque. The rellationships between oral bacteria. Infect evance of this bacterial interaction is Immun 1986: 53: 616-620. also based on the fact that a tnutual 6. Grenier D, Mayrand D. Functional charsytnbiotic relatiotiship has beeti obacterization of e.xtracellular vesicles proserved between P. gitigivalis and T. dett- duced by Bacteroides gingivalis. Infect ticola (D. Grenier, submitted). By proImmun 1987: 55: 111-117. viding a tneans of attachtnent, the inter7. Haapa.salo M. Singh U, McBride BC. Uitto VJ. Sulfliydryl-dependent attachaction could favor growth enhancement ment of Treponema denticola to laminin of the microbial consortium. A sitnilar and other proteins. Infect Immun 1991: concept has been suggested for the 59: 4230-4237. coaggregation of Bacteriotietna matru8. Inoshita E, Amano A, Haiiioka T, Tatnachotii and S. .sanguis in which B. tnatrugawa H, Shizukuishi S, Tsunemitsu A. chotii could derive energy by metabIsolation and some properties of exoheolizing lactate produced by the streptomagglutinin from the culture medium of cocci (24). Bacteroides gingivalis 381. Infect Immun Acknowledgements
I thank Dr. H. Scott (University of British Columbia, Vancouver) for criti-
1986: 52: 421-427. 9. Isogai H, Isogai E, Yoshimura F, Suzuki T, Kagota W, Takano K. SpeciHc inhibition of adherence of an oral strain of Bacteroides gingivalis 381 to epithelial
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
283
cells by monoclonal antibodies against the bacterial funbriae. Arch Oral Biol 1988: 33: 479-485. Kinder SA, Holt SC. Characterization of coaggregation between Bacteroides gingivalis T22 and Fusobaeteriutn nucleatum T18. Infect Immun 1989: 57: 3425-3433. Kolenbrander PE. Intergeneric coaggregation among human oral bacteria and ecology of dental plaque. Annu Rev Microbiol 1988: 42: 627-656. Kolenbrander PE, Andersen RN. Inhibition of coaggregation between Fusohacterium nuclealutii and Porphyromonas (Bacleroide.s) gitigivalis by lactose and related sugars. Infect Immun 1989: 57: 3204-3209. Kolenbrander PE, Celesk RA. Coaggregation of human oral Cytophaga species and Actinomyces i.sraelii. Infect Immun 1983: 40: 1178-1185. Kolenbrander PE, Andersen RN, Moore LVH. Coaggregation of Fusohacteriutn nucleatutn, Selenotnonas flueggei. Selenomotias infcli.w Selenomonas noxia and 5clenomonas spuiigetia with strains from 11 genera of oral bacteria. Infect Immun 1989: 57: 3194-3203. Leschine SB, Canale-Parola E. Rifampin as a selective agent for isolation of oral spirochetes. J Clin Microbiol 1980: 12: 792-795. Loesche WJ, Syed SA, Schmidt E, Morrison EC. Bacterial profiles of subgingival plaques in periodontitis. J Periodontol 1985: 56: 447-456. Loesche WJ, Syed SA, Morrison EC, Kerry GA, Higgins T, Stoll J. Metronidazole in periodontitis. I. Clinical and bacteriological results after 15 to 30 weeks. J Periodontol 1984: 55: 325-335. Mayrand D. Virulence promotion by mixed bacterial infections. In: BayerSymposium VIII. The pathogenesis of bacterial infections. Berlin: Springer-Verlag, 1985: 281-291. Moore WEC, Holdeman LV, Cato EP et al. Comparative bacteriology of juvenile periodontitis. Infect Immun 1985: 48: 507-519. Mouton C. Reynolds HS. Gasiecki EA. Genco RJ. /" vitro adhesion of tufted oral streptococci to Bacterionema matruchotii. Curr Microbiol 1979: 3: 181-186. Ohta H. Kato K, Fukui K, Gottschal JC. Microbial interactions and the development of periodontal disease. J Periodont Res 1991: 26: 255-257. Olsen I. Attachment of Trepotietna detilieola to cultured human epithelial cells. Scand J Dent Res 1984: 92: 55-63. Reijntjens FMJ. Mikx FHM, WoltersLutgerhorst JML, Maltha JC. Adherence of oral treponemes and their effect on morphological damage and detachment of epithelial cells //; viiro. Infect Immun 1986: 51: 642-647. Weerkamp AH. Coaggregation of Strep-
284
Grenier toeoecus salivarius with gram-negative oral bacteria: mechanism and ecological significance. In: Mergenhagen SE, Rosan B, ed. Molecular basis of oral mi-
crobial adhesion. Washington, DC: Am Soc Microbiol, 1985: 177-183. 25. Weiss El, Kolenbrander PE, London J, Hand AR, Andersen RN. Fimbria-as-
sociated proteins of Bacteroides loeseheii PK1295 mediate intergeneric coaggregations. J Bacteriol 1987: 169: 4215-4222.
Demonstration of a bimodal coaggregation reaction between Porphyromonas gingivalis and Treponema denticola
D. Grenier Departement de Sante Buccale, Faculte de Medecine Dentaire, Universite de Montreal, Montreal, Quebec, Canada
Grenier D. Demonstration of a bimodal coaggregation reaction between Porphyromonas gingivalis and Treponema denticola. Oral Microbiol Itntmtttol 1992: 7: 280-284, This study detnonstrates a strong coaggregation reaction between 2 suspected periodontopathogens; Porpliyrotnotias gingivalis and Trcponettia dentieola. Other black-pigmented oral bacterial species tested did not coaggregate with T denticola. This specific interbacterial aggregation was bitnodal, since heating of both cell Key words: coaggregation; adherence: Tretypes was required to completely eliminate the reaction. The coaggregation reacponema denticoia: Porpbyromonas gingition occurred between pH 4 and 9. Under sotne conditions, arginine and Dvatis galactosamine were effective in preventing the coaggregation. The heat-sensitive Daniel Grenier, Groupe de Recherche en receptor on P. gingivalis was found to be loosely bound and could be released Ecologie Buccale, Ecole de Medecine Dentaiby a light ultrasonic treatment of the cells. It is suggested that the bacterial re, Universite Laval, Sainte-Foy (Quebec), Canada, G1K 7P4 interaction described may participate in the establishtnent of a potentially pathogenic subgingival plaque. Accepted for publication March 2, 1992
Nutnerous studies have shown an association between some forms of periodontal disease and increased numbers of spirochetes and gratn-negative anaerobic bacteria in affected sites (16, 17, 19). Several types of bacterial interactions, including coaggregation and nutritional relationships, are thought to be critical for the initial development and the establishment of a potentially pathogenic subgingival plaque (3, 5, 11, 18, 21). The evidence coticerning coaggregation of oral bacteria was first reported by Gibbons & Nygaard (4). Further studies have shown possible bacterial aggregation between gratn-positive and gratn-positive (2, 20), gratn-positive and gratn-negative (3, 13) as well as gramnegative and gram-negative bacteria (10, 12, 14). Different types of surface cotnponents including carbohydrates, fimbriae and outer tnetnbrane proteins have the ability to participate in coaggregation .reactions between oral bacteria (11). A second tnechanistn of bacterial interactions involves the production of growth-stitnulating factors which may contribute to bacterial successions iti periodontal sites (5, 18, 21). I recently
observed a tnutual, synergistic nu- viously described (15). All cultures were tritional relationship between Porphyro- incubated in an anaerobic chatnber (Nr monas gingivalis atid Treponetna dentieo- H2-CO2, 80;10;10) at 37^C. Black-pigla (D. Grenier, subtnitted). This ben- mented bacterial species were grown for eficial relationship appears to involve 1 day whereas T. denticola was cultiintitnate contact between cells of both vated for 7 days. bacterial species. The aim of this study was to characterize the nature of the Coaggregation assay coaggregation reaction between P. gingiBacterial cultures were harvested by valis and T. denticola. centrifugation at 10,000 xg for 15 tnin at 4 C. The cells were washed twice in Material and methods distilled water, and the final pellet was Bacterial strains and growth conditions suspended in distilled water to an abThe following bacterial strains were sorbance, at 660 nm, of 0.75 for the used in this study; T. denticola ATCC black-pigtnented bacterial species and 35405; P gittgivalis ATCC 33277, 381, 0.4 for T. denticola. The bacterial susHWllD-5; Porphyromonas asaccharo- petisions were stored at 4 C for up to 5 lytica ATCC 25260, BM4; Porpliyro- days. Bacterial cells showed no aptnotias etidodontalis ATCC 35406; Prevo- parent self-aggregation when grown and tella intertnedia 5W2; Prevotella tnel- prepared as above. The coaggregation aninogenica ATCC 25845; Prevotella mixture, in 12- by 75-mtn round-bottom loescheii ATCC 15930; Prevotella dentic- test tubes, consisted of 100 //I of T detiola ATCC 33185; and Bacteroides levii ticola, 100 /(I of the black-pigmented ATCC 29147. Growth of black-pig- bacterial species and 200 /d of phosmented bacterial species was carried out phate-buffered saline (PBS; 25 mM soin brain heart infusion broth (BBL diutn phosphate buffer [pH 6.8] conMicrobiology Systetns, Cockeysville, taining 75 tnM NaCl). The tubes were MD) containing hemin (10 //g/ml) and shaken for 5 min at room tetnperature vitatnin K (1 /zg/tnl). T. denticola was on a rocking platfortn mixer (Red-Rogrown in fiuid NOS medium as pre- tor, Hoefer Scientific Instruments, San
Coaggregation of P. gingivalis atid T. denticola Francisco, CA) operating at 150 RPM, and then scored according to a visual rating scale of 0 through 4-1- as previously described by Cisar et al. (2); 0, no visible aggregates in the cell suspension; 1+, small uniform aggregates in the cell suspension; 2-I-, definite aggregates easily seen but suspension retnained turbid; 3-I-, large aggregates which settled rapidly, leaving sotne turbidity in the supernatant fluid; 4-f, clear supernatant fiuid and large aggregates which settled immediately. Except where noted, the coaggregation experiments were done with P. gitigivalis 33277 and T. denticola 35405.
281
cose, D-tnannitol, L-rhatnnose, D-sorbitol, D-sucrose; D-galactosamine, Dglucosatnine; N-acetyl-D-galactosatnine, N-acetyl-D-glucosatnine, N-acetylneuratnin-lactose; EDTA.
Results Of the 8 black-pigmented bacterial species tested, only the 3 strains of P. gittgivalis possessed the ability to fortn large coaggregates with T. denticola cells (Fig. 1). The coaggregation reaction between P. gingivalis and T. dettticola gave Ultrasonic treatment of bacterial cells a tnaxiniutn score of 4-f over a pH P gingivalis and T. denticola cells were range of 5 to 8. Weaker coaggregations subtnitted to a light ultrasonic treat- (2-f) were observed at pH 4 and 9, ment in order to detach any loosely whereas no visible coaggregation ocbound structures or proteins. Briefiy, curred at pH 3 and 10. The coaggregathe bacterial suspensions were sonicated tion reaction was cotnpletely elitninated three times for 10 s (20% duty cycle. when both P. gitigivalis and T. denticola Sonic distneinbrator model 150, Artek cells were heated at 75 C for 30 tnin. Systems, Farmingdale, NY) and the The interbacterial aggregation still occells washed twice by centrifugation. curred wheti otily one cell type was subEffect of pH on the coaggregation reaction Cell suspensions were prepared in dis- tnitted to such heating (Fig. 1, Table 1). The coaggregation reaction between P. tilled water and the coaggregation assay A tnilder heat treatment (55°C for 30 tnin) of both bacterial suspensions did gitigivalis and T denticola was per- was perfortned as described earlier. not interfere with the coaggregation. fortned using the following buffers inCotnplete inhibition of coaggregation stead of PBS; 25 tnM citrate buffer (pH pH sensitivity of the binding components was also obtained by a proteolytic treat3, 4, 5), 25 tnM phosphate buffer (pH 6, 7), 25 tnM Tris hydrochloride buffer P gingivalis and T. dcttticola cells were ment of both cell types with proteinase (pH 8, 9) and 25 mM carbonate buffer suspended in 100 tnM citrate buffer at K (Table 1). With the exception of L-arginitie and (pH 10, 11). All buffers were supple- pH 3, as well as in 100 mM carbonate buffer at pH 12. After iticubation at D-galactosamine, other putative inhibimented with 75 mM NaCl. room temperature for 30 tnin, the cells tors tested did not affect the coaggregawere harvested by centrifugation, sus- tion between P. gingivalis and T. denticoHeat treatment and proteoiytic treatment pended in distilled water and the coag- la (Table 2). Arginine, at a final concenof bacterial celis gregation assay was carried out as de- tration of 150 tnM, cotnpletely eliminated the coaggregation between P. Cell suspensions of P. gittgivalis and T. scribed above. denticola were heated at 75°C for 30 min and used in the coaggregation assay. Samples of the cell suspensions were also incubated at 37 C for 3 h with an equal volume of proteinase K. (Sigtna Chetnical Co., St. Louis, MO; 0.5 tng/ml in PBS). At the end of the incubation period, the cells were harvested by centrifugation, and the pellets were washed twice in distilled water. Cell suspensions were prepared in distilled water and the coaggregation assay was then carried out as described above. Effect of putative inhibitors on the coaggregation reaction
Putative inhibitors of coaggregation reactions were incorporated in PBS and assayed for their ability to inhibit the coaggregation between P. gitigivalis atid T. denticola. Heat-treated and nontreated bacterial cells were tested. The following compounds, at final concentrations (in the coaggregation mixture) ranging frotn 50 to 250 tnM, were used; alanine, arginine, asparagine, cysteine, histiditie, leucine, lysine, proline, serine and threonine (all in the L-fortn); Dfructose, D-fucose, D-galactose, D-glu-
Fig. J. Visual coaggregation reactions between P. gingivalis and T denticola. A. P. gingivalis and T. denticola. B. Heat-treated P. gingivalis and T. denticola. C. P. gittgivalis and heat-treated T denlicola. D. Heat-treated P. gingivalis and heat-treated T. denlieola
Grenier
282
Table I. Effect of heat treatment and protease treatment on the coaggregation reaction between P. gingivalis and T denticola Bacteria Heat-treated P. gitigivalis
Heat-treated T. denlieola Coaggregation'
No No Yes Yes
No Yes No Yes
Proteasetreated P. gitigivalis
Proteasetreated T denticola
No No Yes Yes
No Yes No Yes
•'Scored as described methods
4-h 3-14-10
4-H 3-t3-f 0 in
Material
and
Incubation of T denticola cells at pH 12 resulted in a loss of their ability to coaggregate with heat-treated P. gitigivalis cells. However, these treated spirochetes could still react with nontreated P. gingivalis cells, indicating that the heat-sensitive binding cotnponent present on T. denticola was setisitive to high pH. A similar treattnent of P. gingivalis did not change the coaggregation profile of the bacteria. Treattnent of P. gingivalis at pH 3 elitninated all coaggregation reactions. Finally, treattnent of T. denticola cells at this low pH allowed coaggregation with non-treated P. gitigivalis but not with the heattreated cells.
Table 3. Effect of ultrasonic treatment of P. gingivalis on the coaggregation reaction with T. denticola Treatment of bacteria P. gingivalis No No Sonicated Sonicated
T dctiticota Coaggregation" No Heat-treated No Heat-treated
4+ 34440
Assay in the presence of 150 mM arginine** No No 0 Heat-treated No 3 4Sonicated No 3 4•'Scored as described in Material and methods. *" Final concentration in the coaggregation mixture
suggests that physical constraints due to the arginine may prevent the coaggregaThis study demonstrated a strong coag- tion tnediated by the second binding gingivalis and heat-treated or non- gregatioti reaction between 2 suspected cotnpotient on P. gingivalis. The absence treated T. denticola. No significant inhi- periodontopathogens; P. gitigivalis and of inhibition, caused by arginine, when bition by arginine was observed when T. denticola. The coaggregation was bi- heat-treated or sonicated P. gitigivalis heat-treated P. gitigivalis cells were modal, since heating of both bacterial cells were used, suggests that the heatmixed with non-treated T. dentieola. D- parttiers was required to eliminate the sensitive receptor is denaturated or regalactosatnine, at a high concentration reaction. The coaggregation of P. gin- tnoved, and that the resulting coaggre(250 tnM), was able to reduce but not givalis and T. denticola appears to be gation with T. denticola can thus be cotnpletely inhibit the coaggregation of highly specific, since a similar reaction mediated by the second binding compoP. gitigivalis with heat-treated or non- was not observed with other black-pig- nent present on the cell surface of P. treated T denticola. When heat-treated mented bacterial species, including P. gingivalis. The coaggregation between P. P gitigivalis cells were used in the coag- asaccharolytica and P. etidodontalis. Atigingivalis atid T. denticola was shown to gregation assay, no inhibition by D-gal- additional strain of T. denticola showed be reduced in the presence of D-galactoactosatnine was observed. also the ability to coaggregate with P. samine. The lack of inhibition produced A light ultrasonic treattnent of the gitigivalis (data not shown). by EDTA suggests that cations are not bacterial suspensions was done to deterThe schematic representation of the involved in the interaction. mine whether the binding components proposed bitnodal coaggregation reacThis is the first report describing the were firtnly or loosely attached to the tion between P. gitigivalis and T. dentico-coaggregation properties of T. denticola surface of the bacteria. No difference la is presented in Fig. 2. Two binding with P. gingivalis. Previous studies have was observed in the binding property components, one heat-sensitive and one already shown the ability of T denticola of sonicated T. denticola cells. However, heat-stable, are thought to be present to attach to epithelial cells (22, 23), sonication of P. gingivalis cells appeared on both bacterial species. The heat-sen- basetnent tnembrane cotnponents (7) to remove its heat-sensitive receptor, as sitive receptor on P. gitigivalis is loosely and to tnodified hydroxyapatite sursuggested by; i) the cotnplete loss of bound and can be released by a light faces (1). Extracellular deposits of acidic aggregation with heat-treated T. dentic- ultrasotiic treatment of the bacterial mucopolysaccharides on the surface of ola cells (as noted with heat-treated P cells. The attachment mediated by this T. denticola have been thought to pargitigivalis) and ii) the absence of inhi- receptor is inhibited by L-argitiitie. Sur- ticipate in the attachtnent to epithelial bition by arginitie for the coaggregatioti prisingly, no coaggregation between cells (22). However, prelitninary experireaction with non-treated T. dentieola non-treated P. gitigivalis and non- ments in my laboratory have indicated cells (as noted with heat-treated P. giti- treated T denticola cells occurred in the that the coaggregation property of T. givalis) (Table 3).' , . . . • presence of arginine. This observation denticola is not affected by treattnent of the cells with hyaluronidase, suggesting Tahle 2. Effect of L-arginine and D-galactosamine on the coaggregation t'eaction between P. that tnucopolysaccharides are not ingingivalis and T. denticola volved. Bac teria Recently, P. gingivatis was showti to Coaggregation'' in presence of b coaggregate with various strains of EuArginine (mM) D-galactosamine (mM)*" Heat-treated Heat-treated .sobacteriutn tiucleatutn (10, 12) and ActiP. gitigivalis T. denticola 0 50 150 0 125 250 notnyces viscosus (3). In the first case, No No 0 44242+ 44the reaction was prevented by lactose No Yes 34240 34• 3-f 14and related sugars or by heat treattnent Yes No 444434444434of the fusobacterium, whereas in the second case, the attachtnent was not in"Scored as described in Material and methods. ''Final concentration in the coaggregation mixture hibited by carbohydrates atid was otily Discussion
Coaggregation of P. gingivalis and T. denticola Heat sensitive
Heat stable 10.
11.
12. Fig. 2. Schematic representation of the proposed bimodal coaggregation reaction between P. gingivalis and T denlieola
partially reduced by heating the P. gin- cally reading the manuscript and T.-T. givalis cells. Hetnagglutinating activity Huynh for excellent technical assistof P. gitigivalis has been previously re- ance. This work was supported by the ported to be inhibited by arginine (6, 8). Medical Research Council of Canada This hemagglutinin and the P. gingivalis (grant no; DG-383-384-385). heat-sensitive binding cotnponent, detnonstrated in the present teport, could represent the satne tnolecule, since both are inhibited in a sitnilar fashion by arginine. Isogai et al. (9) have suggested References that P. gingivalisfimbriae,which cati be 1. Cimasoni G, McBride BC. Adherence of removed by sonication, tnay be responTreponema denticola to modified hysible for the adherence to epithelial cells. droxyapatite. J Dent Res 1987: 66: This adhesin tnay represent the P. gin1727-1729. givalis loosely bound, heat-sensitive 2. Cisar JO, Kolenbrander PE, Mclntire binding cotnponent observed in the EC. Specificity of coaggregation reacpresent study. Another black-pigtnented tions between human oral streptococci bacterial species, P. loescheii, has been and strains of Actinomyces viscosus or Actinotnvces naeslundii. Infect Immun previously shown to possess loosely at1979: 24: 742-752. tached adhesin molecules on its surface 3. Ellen RP. Schwarz-Faulkner S, Grove (25). It was also detnonstrated that 2 DA. Coaggregation among periodontal distinct types of fitnbriae-associated pathogens, emphasizing Bacteroides ginproteins on P. loescheii were responsible givctlis - Actinomyces viscosus cohesion for the coaggregation with Streptococon a saliva-coated mineral surface. Can cus sanguis and Actitiotnyces israelii. J Microbiol 1988: 34: 299-306. The coaggregation phenomenon de4. Gibbons RJ, Nygaard M. Interbacterial aggregation of plaque bacteria. Arch scribed in this study tnay be important Oral Biol 1970: 15: 1397-1400. for the establishment of a potentially 5. Grenier D, Mayrand D. Nutritional repathogenic subgingival plaque. The rellationships between oral bacteria. Infect evance of this bacterial interaction is Immun 1986: 53: 616-620. also based on the fact that a tnutual 6. Grenier D, Mayrand D. Functional charsytnbiotic relatiotiship has beeti obacterization of e.xtracellular vesicles proserved between P. gitigivalis and T. dett- duced by Bacteroides gingivalis. Infect ticola (D. Grenier, submitted). By proImmun 1987: 55: 111-117. viding a tneans of attachtnent, the inter7. Haapa.salo M. Singh U, McBride BC. Uitto VJ. Sulfliydryl-dependent attachaction could favor growth enhancement ment of Treponema denticola to laminin of the microbial consortium. A sitnilar and other proteins. Infect Immun 1991: concept has been suggested for the 59: 4230-4237. coaggregation of Bacteriotietna matru8. Inoshita E, Amano A, Haiiioka T, Tatnachotii and S. .sanguis in which B. tnatrugawa H, Shizukuishi S, Tsunemitsu A. chotii could derive energy by metabIsolation and some properties of exoheolizing lactate produced by the streptomagglutinin from the culture medium of cocci (24). Bacteroides gingivalis 381. Infect Immun Acknowledgements
I thank Dr. H. Scott (University of British Columbia, Vancouver) for criti-
1986: 52: 421-427. 9. Isogai H, Isogai E, Yoshimura F, Suzuki T, Kagota W, Takano K. SpeciHc inhibition of adherence of an oral strain of Bacteroides gingivalis 381 to epithelial
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
283
cells by monoclonal antibodies against the bacterial funbriae. Arch Oral Biol 1988: 33: 479-485. Kinder SA, Holt SC. Characterization of coaggregation between Bacteroides gingivalis T22 and Fusobaeteriutn nucleatum T18. Infect Immun 1989: 57: 3425-3433. Kolenbrander PE. Intergeneric coaggregation among human oral bacteria and ecology of dental plaque. Annu Rev Microbiol 1988: 42: 627-656. Kolenbrander PE, Andersen RN. Inhibition of coaggregation between Fusohacterium nuclealutii and Porphyromonas (Bacleroide.s) gitigivalis by lactose and related sugars. Infect Immun 1989: 57: 3204-3209. Kolenbrander PE, Celesk RA. Coaggregation of human oral Cytophaga species and Actinomyces i.sraelii. Infect Immun 1983: 40: 1178-1185. Kolenbrander PE, Andersen RN, Moore LVH. Coaggregation of Fusohacteriutn nucleatutn, Selenotnonas flueggei. Selenomotias infcli.w Selenomonas noxia and 5clenomonas spuiigetia with strains from 11 genera of oral bacteria. Infect Immun 1989: 57: 3194-3203. Leschine SB, Canale-Parola E. Rifampin as a selective agent for isolation of oral spirochetes. J Clin Microbiol 1980: 12: 792-795. Loesche WJ, Syed SA, Schmidt E, Morrison EC. Bacterial profiles of subgingival plaques in periodontitis. J Periodontol 1985: 56: 447-456. Loesche WJ, Syed SA, Morrison EC, Kerry GA, Higgins T, Stoll J. Metronidazole in periodontitis. I. Clinical and bacteriological results after 15 to 30 weeks. J Periodontol 1984: 55: 325-335. Mayrand D. Virulence promotion by mixed bacterial infections. In: BayerSymposium VIII. The pathogenesis of bacterial infections. Berlin: Springer-Verlag, 1985: 281-291. Moore WEC, Holdeman LV, Cato EP et al. Comparative bacteriology of juvenile periodontitis. Infect Immun 1985: 48: 507-519. Mouton C. Reynolds HS. Gasiecki EA. Genco RJ. /" vitro adhesion of tufted oral streptococci to Bacterionema matruchotii. Curr Microbiol 1979: 3: 181-186. Ohta H. Kato K, Fukui K, Gottschal JC. Microbial interactions and the development of periodontal disease. J Periodont Res 1991: 26: 255-257. Olsen I. Attachment of Trepotietna detilieola to cultured human epithelial cells. Scand J Dent Res 1984: 92: 55-63. Reijntjens FMJ. Mikx FHM, WoltersLutgerhorst JML, Maltha JC. Adherence of oral treponemes and their effect on morphological damage and detachment of epithelial cells //; viiro. Infect Immun 1986: 51: 642-647. Weerkamp AH. Coaggregation of Strep-
284
Grenier toeoecus salivarius with gram-negative oral bacteria: mechanism and ecological significance. In: Mergenhagen SE, Rosan B, ed. Molecular basis of oral mi-
crobial adhesion. Washington, DC: Am Soc Microbiol, 1985: 177-183. 25. Weiss El, Kolenbrander PE, London J, Hand AR, Andersen RN. Fimbria-as-
sociated proteins of Bacteroides loeseheii PK1295 mediate intergeneric coaggregations. J Bacteriol 1987: 169: 4215-4222.
