- Email: [email protected]
Synergistic degradation of bovine serum albumin by mutans streptococci and other dental plaque bacteria
FEMS MicrobiologyLetterst)0(1992)259-262 © 1992Federationof European MicrobiologicalSocieties0378-1(197/t;2/$05.00 Publishedby Elsevier
259
FEMSLE 04762
Synergistic degradation of bovine serum albumin by mutans streptococci and other dental plaque bacteria Karen A. H o m e r and David Beighton Hunterian Dental Research Unit, London Hospital Medical College, TurnerStreet. Whitechapel, London, U.K.
Received 17 October 1991 Accepted 12November1991 Key words: Dental plaque bacteria; Mutans streptococci; Proteolytic activity; Synergism 1. SUMMARY Mutans streptococci (Streptococcus mutans and Streptococcus sobrinus) exhibited low levels of proteolytic activity against the model protein substrate, FITC-labelled bovine serum albumin, when incubated alone. Inclusion of other members of the dental plaque micro~iora in the assay usually resulted in marked increases in the degree of proteolysis and a high level of synergy. Interactions between mutans streptococci and either Streptococcus oralis or Fusobacterium nucleatum gave rise to the greatest degree of synergistic proteolytic degradation.
2. INTRODUCTION Studies in rats [I] and macaque monkeys [2,3] have shown that the growth rates of dental plaque bacteria are independent of the presence of host diet. Bacteria adhering to the teeth must therefore be capable of degrading, and obtaining nutriCorrespondence to: K.A. Homer, Hunterian Dental Research Unit, London Hospital Medical College, -turner Street, Whitechapel,London,El 2AD, U.K.
ents necessary for growth from host-derived macromolecules, including proteins and glycoproteins present in salivary secretions and gingival crevicular fluid exudates. The degradation of these macromolecules will be mediated by the concerted action of a combination of glycosidic and proteolytic enzyme activities, which have been reported for a wide range of dental plaque bacteria [3-7]. The ability of mutans streptococci (Streptococcus mutans and Streptococcus sobrinus ), the principle causative agents of dental caries [8], to interact with other bacteria and to degrade proteins has not been reported previously. These two species each pessess low, but measurable, levels of proteolytic activity [4] and here we report the interaction of these mutans streptococci with principle members of the supragingival plaque microflora to degrade synergistically the model, non-glycosylated protein, bovine serum albumin. 3. MATERIALS AND METHODS 3.1. Growth o f microorganisms and preparation o f cell suspensions Each strain of bacterium was plated onto Fastidious Anaerobe Agar (Lab M., Salford, Lanes.,
2611
U,K.) and incubated at 37°C for 2 days. Str~7~tococcus sobrinus (strains KI and SLI ), Streptococcus mutans (strains Inghritt and NCTC 11,14491, Actmomyces t'i.~'cosus (strain WVU627) and StrtT~tococcus oralis (strain EFI861 wcrc grown under microacrophilic conditions in a candle jar. Porphyrontonas gingit'alis (strain W831 and Fusobacterium mtcleatttttt (strain 14461 were grown in an anaerobic cabinet (Don Whitlcy, Shipley, W. Yorks., U.K.). Colonies of each strain were removed into 51,) mM N-Tris-(hydroxymethyl)methyl-2-aminoethane sultbnic acid (Sigma Chemical Company, Poolc, Dorset, U.K.) buffer, pH 7.5 (TES buffer) and small aliquols of the suspension were removed for protein estimation. Concentrations of the ecI[ suspensions were adjusted with TES buffer to give 20/¢g/ml bacterial protein.
3.2. Estimation of plvtein concentrations of bacteria/ sttspe/tsions Aliquots (1 ml) of bacterial suspensions from each stlain vvcre dispensed into Eppcndorf tubes and were subjected to centrifugation 113111.)0x g, 5 min) to pellet cells. The supcrnatant was removed and the cells wcrc rcsuspcnded in 1,,I,5ml of 1,).5 M sodium hydroxide, Bacterial protein was extracted by boiling according to the method of Herbert el al. [9] and, following remowd of cell debris by centrifugation and neutralization of the supcrnatant, protein concentrations wcre dctcrmined using the Coomassie bluc dye-binding assay (Pierce, Rockford, 1L) and by comparison with a standard curve constructed using bovine serum albumin (Sigma). 3.3. Assto' of protein degradation hy bacteria Fluorcscein isothiocyanate-labelled (FITClabelled) bovine serum albumin was purchased from Sigma and diluted with TES buffer to give a concentration of 11,111~.g/ml, Degradation of this substrate was monitored using the method described by Homer and Beighton [10]. Briefly, assays contained: 50 ~1 of bacterial suspensions of strains K|, SLI, lngbritt or 10449; 511 #1 of strains W83, 1446, EFI86, or WVU627; 511 #1 o!" FITClabelled protein solution; 51,1 ~1 of TES buffer, pH 7.5. The degradation of the substrate by each strain individually was also monitored, adding 50
#1 of TES buffer to maintain the total assay wflume at 21,10 #1, Control assays contained no bacterial suspension to allow for the non-biological breakdown of substratc and the intrinsic background fluorescence of the substrate. Each assay was replicated 12 times in %-well microtitre trays (Perkin-Elmer) and incubated fl~r up to 24 h prior to the determination of fluorescence at excitation 4~5 and emission 525 nm (Perkin-Elmer LS-3B fluorescence spectrometer with a plate-reading attachment). Proteasc activity is given in terms of mean relative fluorescence after the deduction of control values.
3,4. Statistical anal.vsis The mean and standard deviations of the actual fluorescence values, with the appropriate control fluorescence values subtracted, were calculated for each type of assay (individual bacterial species and combinations of bacterial specics). In order to determine the predicted interaction of combinations of bacteria, and so their predicted ability to degrade BSA. the 12 assay results for each of two bacterial species tested in combination were randomly paired to obtain 12 predicted values. The means of these 12 predicted fluorescence values wcrc compared, using a t-test for independent samples, with the actual fluorescence values tk~r each pair of bacterial species (each strain of mutans streptococci with F, nuch'atum strain 1446, S. oralis strain EF186, A. ciscosus strain WVU627 and !: gingicalis W83) tested for their ability to interact synergistically. This was considered to have occurred if the actual fluorescence wducs wcrc significantly (P < (1.(1111) greater than the calculated predicted values,
4, RESULTS Each bacterial species when incubated alone with FITC-BSA, with the exception of P. ght,qit'alis, exhibited similar low proteolytic activities reflected by the small increases in relative fluorescence (Table 1). The protcolytic interactions between S. mutans strains and the E nuch'amm,
261 100
Table I Degradation of FITC-BSA by individual strains of bacteria. ;is indicated by an increase in relative fluorescente ( ± SD) Bacterial strain
Relative fluorescence
S. mutam" strain NC'TC 10449 S. mutans strain Ingbrin S. sohrinus strain SL-! S. sobrinu.~- strain K! E mtrleatum strain 1448 S. ora/is strain EFISfi A. viscosus strain WVU627 P. ghzgh'alis strain W83
23.1 +_ 3.9 32.7 + 4.8 t8.2_+ 2.(I 37.9_+ 9.4 9.7_+ 1.2 I 1.4+ 3.3 14.2+ 5.3 373.2 + 12.1
+~
.
+p
80
*
z
so *
~
4o
~,,-I, 20
Bacterial strains were incubated with FITC-BSA in TES buffer (50 mM, pH 7.5) at 37°C for 24 h, as in MATFRIAI.S AND MI!rlI()DS.
0
1
2 K1
3
!
2 $L-1
3
Fig. 2. Degradation of bovine serum albumin by $. sohrim~s
S oralis and A. riscosus strains are shown in Fig. 1. Significant synergistic interactions occurred between each of the two S. mutans strains and F. nuck, atum and S. oralis; with A. riscosus, significant interactions were apparent only with S. metans strain lngbritt. A similar pattern was observed for the interactions between S. sobrinus strains and F. nucleatum and S. oralis (Fig. 2) but
100
a 80
~ 8o 0 ~
40
strains in combination with other species of dental plaque bacteria. Degradation of FITC-BSA (shown as increase in relative fluorescence) was measured after 24 h incubation at 37°C. Filled bars indicate actual relative fluorescence values, open bars indicate calculated relative fluorescence values (assuming no synergistic interactions) for S. sobrinus strains KI and SL-I in combination with (1) F. nucleatum, (21 S. ora/is, and (3) A. t'iscosu.~.
again interactions with A. i'iscosus were inconsistent. Analysis of the interactions with P. gingiralis and the mutans streptococci showed that significant interactions occurred with P. gingivalis and each of S. mutans strain NCTC 10449 and S. sobrinus strain SL-I, but not with the other two strains (data not shown). The actual and predicted relative fluorescence values for the P. gingiralis-S, mtttans interactions were 499__. 31.6 and 396 +_ 10.5 ( P < 0.01 ) and for P. gingicalis-S. sobritms, 508 + 37.0 and 391 +_ 11.6 ( P < 0.01), respectively.
I---
,,-I, n-. 0
5. DISCUSSION 1
2
3
1
2
3
INGBRITT NCTC 1 0 4 4 9 Fig. I. Degradation of b~winc serum albumin by S. mutans strains in combination wilh other species of dental plaque bacteria. Degradation of FITC-BSA (shown as increase in relative fluorescence) wa,~ measured after 24 h incubation ,,t 37°C. Filled bars indicate actual relative fluorescence values, open bars indicate calculated relative fluorescence v;,lues (assuming no synergistic interactions) Ior S. mutans strains lngbrin and NCTC 1(1449 in combinafitm with (1) F. mwleaturn, (2) S. oralis, and (31 A. risco.~'u.~.
We have used BSA as a model protein to study the proteolytic interactions of supragingivai dental plaque bacteria. Although this protein will be encountered only rarely by these bacteria in vivo, the mechanisms of proteolysis required for the degradation of BSA are likely to be universal and, given the non-glycosylated nature of BSA, permits a study of the proteolytic interactions of
262 organisms without the steric hindrance afforded by oligosaccharide sidechains, Mutans streptococci as a group have amongst the lowest levels of proteolytic activity found in the oral streptococci, when measured using FITC-labelled BSA and a range of synthetic 7amido-4-mcthylcoumarin-linked endopeptidase substrates [4]. Here we have shown that the proteolytic activity of mutans streptococci can be effectively increased when they are incubated together with other species of dental plaque bacteria. The increase in proteolytic activity was due to the synergistic interactions between the individual strains of mutans streptococci and the other dental plaque bacteria. The synergistic interactions may be a consequence of the ability of different bacterial species to produce a range of extracellular proteolytic activities that has different specificities and affinities for particular peptide bonds of amino acid sequences. This is apparent for oral streptococci [4] and is likely to explain the interaction between subgingival plaque pathogens F. nucleatum and P. gingit'alis, reported by Garbia et al. [11] for the synergistic degradation of casein. In vivo the levels of proteolytic activity produced by dental plaque bacteria are greatly influenced by the availability of the host's diet: withdrawal of the diet significantly increases a large number of different peptidase, carboxypeptidase and endopeptidase activities [12]. These activities are produced to enable dental plaque bacteria to grow, in the absence of host diet, by degrading host-derived glycoproteins and proteins. It has
previously been proposed that this degradative activity is a cooperative process with enzyme activities being contributed by different members of the dental plaque flora [13], In this paper we have shown clearly that dental plaque bacterial proteolytic activities can interact to degrade protein and that this interaction is usually synergistic.
REFERENCES [I] Beckers,}I.J.A. and Van Der lloeven, J,S. (t~.'82) Infect. Immun 35 583-587. [2] Beighlon.D. and Hayday. H. {1986) Arch. Oral Biol. 31, 449-454. 13] Beighton, D., Smith. K. and Hayday, It. 11986) Arch. Oral Biol, 31,829-835. [4] Homer, K.A., Whiley, R.A, and Beighton, D. (1990] FEMS Microbiol. Left. 67, 257-261]. 151 Slots, J. (1981)J. Clin. Microbiol. 14, 288-2'44. [6] Tanner, A.C.R., Strzempko. M.N., getsky, C.A. and McKinley,G.A. (19851J. Clin. Microbiol, 22, 333-335. [7] Van Winkelhoff,A,J., V~m Steenbergen, J,M., Kippuw, N. and De Graaff. J. (1986) Anton. Van Leeuwenhoek 52, 163-171. [8] Loesche,W,J. (1986) Microbiol. Rev. 50, 353-381). [9] Herbert, D., Phipps, P.J. and Strange, R.E. (1971) In: Methods in Microbiology(Navis, R.J, and Ribbons. D.W., Eds.), Vol. 5B, pp. 2111-344.AcademicPress, London. [16] Homer, K.A, and Beighton, D. (19911)Anal. Biochem. 191, 133-137. [11] Gharbia, S.E., Shah, H.N. and Welch. S.G. {1989) Curt, Microbiol. 19, 231-235. [121 Smith, K. and Beighton, D. (1986) J. Dent. Res. 65, 1340-1352. [13] Beighton, D., Smith, K., Gtenister. D.A.. Salamon, K. and Keevil, C,W. (1988) Microb. Ecol. Health Dis. I, 85-94.
259
FEMSLE 04762
Synergistic degradation of bovine serum albumin by mutans streptococci and other dental plaque bacteria Karen A. H o m e r and David Beighton Hunterian Dental Research Unit, London Hospital Medical College, TurnerStreet. Whitechapel, London, U.K.
Received 17 October 1991 Accepted 12November1991 Key words: Dental plaque bacteria; Mutans streptococci; Proteolytic activity; Synergism 1. SUMMARY Mutans streptococci (Streptococcus mutans and Streptococcus sobrinus) exhibited low levels of proteolytic activity against the model protein substrate, FITC-labelled bovine serum albumin, when incubated alone. Inclusion of other members of the dental plaque micro~iora in the assay usually resulted in marked increases in the degree of proteolysis and a high level of synergy. Interactions between mutans streptococci and either Streptococcus oralis or Fusobacterium nucleatum gave rise to the greatest degree of synergistic proteolytic degradation.
2. INTRODUCTION Studies in rats [I] and macaque monkeys [2,3] have shown that the growth rates of dental plaque bacteria are independent of the presence of host diet. Bacteria adhering to the teeth must therefore be capable of degrading, and obtaining nutriCorrespondence to: K.A. Homer, Hunterian Dental Research Unit, London Hospital Medical College, -turner Street, Whitechapel,London,El 2AD, U.K.
ents necessary for growth from host-derived macromolecules, including proteins and glycoproteins present in salivary secretions and gingival crevicular fluid exudates. The degradation of these macromolecules will be mediated by the concerted action of a combination of glycosidic and proteolytic enzyme activities, which have been reported for a wide range of dental plaque bacteria [3-7]. The ability of mutans streptococci (Streptococcus mutans and Streptococcus sobrinus ), the principle causative agents of dental caries [8], to interact with other bacteria and to degrade proteins has not been reported previously. These two species each pessess low, but measurable, levels of proteolytic activity [4] and here we report the interaction of these mutans streptococci with principle members of the supragingival plaque microflora to degrade synergistically the model, non-glycosylated protein, bovine serum albumin. 3. MATERIALS AND METHODS 3.1. Growth o f microorganisms and preparation o f cell suspensions Each strain of bacterium was plated onto Fastidious Anaerobe Agar (Lab M., Salford, Lanes.,
2611
U,K.) and incubated at 37°C for 2 days. Str~7~tococcus sobrinus (strains KI and SLI ), Streptococcus mutans (strains Inghritt and NCTC 11,14491, Actmomyces t'i.~'cosus (strain WVU627) and StrtT~tococcus oralis (strain EFI861 wcrc grown under microacrophilic conditions in a candle jar. Porphyrontonas gingit'alis (strain W831 and Fusobacterium mtcleatttttt (strain 14461 were grown in an anaerobic cabinet (Don Whitlcy, Shipley, W. Yorks., U.K.). Colonies of each strain were removed into 51,) mM N-Tris-(hydroxymethyl)methyl-2-aminoethane sultbnic acid (Sigma Chemical Company, Poolc, Dorset, U.K.) buffer, pH 7.5 (TES buffer) and small aliquols of the suspension were removed for protein estimation. Concentrations of the ecI[ suspensions were adjusted with TES buffer to give 20/¢g/ml bacterial protein.
3.2. Estimation of plvtein concentrations of bacteria/ sttspe/tsions Aliquots (1 ml) of bacterial suspensions from each stlain vvcre dispensed into Eppcndorf tubes and were subjected to centrifugation 113111.)0x g, 5 min) to pellet cells. The supcrnatant was removed and the cells wcrc rcsuspcnded in 1,,I,5ml of 1,).5 M sodium hydroxide, Bacterial protein was extracted by boiling according to the method of Herbert el al. [9] and, following remowd of cell debris by centrifugation and neutralization of the supcrnatant, protein concentrations wcre dctcrmined using the Coomassie bluc dye-binding assay (Pierce, Rockford, 1L) and by comparison with a standard curve constructed using bovine serum albumin (Sigma). 3.3. Assto' of protein degradation hy bacteria Fluorcscein isothiocyanate-labelled (FITClabelled) bovine serum albumin was purchased from Sigma and diluted with TES buffer to give a concentration of 11,111~.g/ml, Degradation of this substrate was monitored using the method described by Homer and Beighton [10]. Briefly, assays contained: 50 ~1 of bacterial suspensions of strains K|, SLI, lngbritt or 10449; 511 #1 of strains W83, 1446, EFI86, or WVU627; 511 #1 o!" FITClabelled protein solution; 51,1 ~1 of TES buffer, pH 7.5. The degradation of the substrate by each strain individually was also monitored, adding 50
#1 of TES buffer to maintain the total assay wflume at 21,10 #1, Control assays contained no bacterial suspension to allow for the non-biological breakdown of substratc and the intrinsic background fluorescence of the substrate. Each assay was replicated 12 times in %-well microtitre trays (Perkin-Elmer) and incubated fl~r up to 24 h prior to the determination of fluorescence at excitation 4~5 and emission 525 nm (Perkin-Elmer LS-3B fluorescence spectrometer with a plate-reading attachment). Proteasc activity is given in terms of mean relative fluorescence after the deduction of control values.
3,4. Statistical anal.vsis The mean and standard deviations of the actual fluorescence values, with the appropriate control fluorescence values subtracted, were calculated for each type of assay (individual bacterial species and combinations of bacterial specics). In order to determine the predicted interaction of combinations of bacteria, and so their predicted ability to degrade BSA. the 12 assay results for each of two bacterial species tested in combination were randomly paired to obtain 12 predicted values. The means of these 12 predicted fluorescence values wcrc compared, using a t-test for independent samples, with the actual fluorescence values tk~r each pair of bacterial species (each strain of mutans streptococci with F, nuch'atum strain 1446, S. oralis strain EF186, A. ciscosus strain WVU627 and !: gingicalis W83) tested for their ability to interact synergistically. This was considered to have occurred if the actual fluorescence wducs wcrc significantly (P < (1.(1111) greater than the calculated predicted values,
4, RESULTS Each bacterial species when incubated alone with FITC-BSA, with the exception of P. ght,qit'alis, exhibited similar low proteolytic activities reflected by the small increases in relative fluorescence (Table 1). The protcolytic interactions between S. mutans strains and the E nuch'amm,
261 100
Table I Degradation of FITC-BSA by individual strains of bacteria. ;is indicated by an increase in relative fluorescente ( ± SD) Bacterial strain
Relative fluorescence
S. mutam" strain NC'TC 10449 S. mutans strain Ingbrin S. sohrinus strain SL-! S. sobrinu.~- strain K! E mtrleatum strain 1448 S. ora/is strain EFISfi A. viscosus strain WVU627 P. ghzgh'alis strain W83
23.1 +_ 3.9 32.7 + 4.8 t8.2_+ 2.(I 37.9_+ 9.4 9.7_+ 1.2 I 1.4+ 3.3 14.2+ 5.3 373.2 + 12.1
+~
.
+p
80
*
z
so *
~
4o
~,,-I, 20
Bacterial strains were incubated with FITC-BSA in TES buffer (50 mM, pH 7.5) at 37°C for 24 h, as in MATFRIAI.S AND MI!rlI()DS.
0
1
2 K1
3
!
2 $L-1
3
Fig. 2. Degradation of bovine serum albumin by $. sohrim~s
S oralis and A. riscosus strains are shown in Fig. 1. Significant synergistic interactions occurred between each of the two S. mutans strains and F. nuck, atum and S. oralis; with A. riscosus, significant interactions were apparent only with S. metans strain lngbritt. A similar pattern was observed for the interactions between S. sobrinus strains and F. nucleatum and S. oralis (Fig. 2) but
100
a 80
~ 8o 0 ~
40
strains in combination with other species of dental plaque bacteria. Degradation of FITC-BSA (shown as increase in relative fluorescence) was measured after 24 h incubation at 37°C. Filled bars indicate actual relative fluorescence values, open bars indicate calculated relative fluorescence values (assuming no synergistic interactions) for S. sobrinus strains KI and SL-I in combination with (1) F. nucleatum, (21 S. ora/is, and (3) A. t'iscosu.~.
again interactions with A. i'iscosus were inconsistent. Analysis of the interactions with P. gingiralis and the mutans streptococci showed that significant interactions occurred with P. gingivalis and each of S. mutans strain NCTC 10449 and S. sobrinus strain SL-I, but not with the other two strains (data not shown). The actual and predicted relative fluorescence values for the P. gingiralis-S, mtttans interactions were 499__. 31.6 and 396 +_ 10.5 ( P < 0.01 ) and for P. gingicalis-S. sobritms, 508 + 37.0 and 391 +_ 11.6 ( P < 0.01), respectively.
I---
,,-I, n-. 0
5. DISCUSSION 1
2
3
1
2
3
INGBRITT NCTC 1 0 4 4 9 Fig. I. Degradation of b~winc serum albumin by S. mutans strains in combination wilh other species of dental plaque bacteria. Degradation of FITC-BSA (shown as increase in relative fluorescence) wa,~ measured after 24 h incubation ,,t 37°C. Filled bars indicate actual relative fluorescence values, open bars indicate calculated relative fluorescence v;,lues (assuming no synergistic interactions) Ior S. mutans strains lngbrin and NCTC 1(1449 in combinafitm with (1) F. mwleaturn, (2) S. oralis, and (31 A. risco.~'u.~.
We have used BSA as a model protein to study the proteolytic interactions of supragingivai dental plaque bacteria. Although this protein will be encountered only rarely by these bacteria in vivo, the mechanisms of proteolysis required for the degradation of BSA are likely to be universal and, given the non-glycosylated nature of BSA, permits a study of the proteolytic interactions of
262 organisms without the steric hindrance afforded by oligosaccharide sidechains, Mutans streptococci as a group have amongst the lowest levels of proteolytic activity found in the oral streptococci, when measured using FITC-labelled BSA and a range of synthetic 7amido-4-mcthylcoumarin-linked endopeptidase substrates [4]. Here we have shown that the proteolytic activity of mutans streptococci can be effectively increased when they are incubated together with other species of dental plaque bacteria. The increase in proteolytic activity was due to the synergistic interactions between the individual strains of mutans streptococci and the other dental plaque bacteria. The synergistic interactions may be a consequence of the ability of different bacterial species to produce a range of extracellular proteolytic activities that has different specificities and affinities for particular peptide bonds of amino acid sequences. This is apparent for oral streptococci [4] and is likely to explain the interaction between subgingival plaque pathogens F. nucleatum and P. gingit'alis, reported by Garbia et al. [11] for the synergistic degradation of casein. In vivo the levels of proteolytic activity produced by dental plaque bacteria are greatly influenced by the availability of the host's diet: withdrawal of the diet significantly increases a large number of different peptidase, carboxypeptidase and endopeptidase activities [12]. These activities are produced to enable dental plaque bacteria to grow, in the absence of host diet, by degrading host-derived glycoproteins and proteins. It has
previously been proposed that this degradative activity is a cooperative process with enzyme activities being contributed by different members of the dental plaque flora [13], In this paper we have shown clearly that dental plaque bacterial proteolytic activities can interact to degrade protein and that this interaction is usually synergistic.
REFERENCES [I] Beckers,}I.J.A. and Van Der lloeven, J,S. (t~.'82) Infect. Immun 35 583-587. [2] Beighlon.D. and Hayday. H. {1986) Arch. Oral Biol. 31, 449-454. 13] Beighton, D., Smith. K. and Hayday, It. 11986) Arch. Oral Biol, 31,829-835. [4] Homer, K.A., Whiley, R.A, and Beighton, D. (1990] FEMS Microbiol. Left. 67, 257-261]. 151 Slots, J. (1981)J. Clin. Microbiol. 14, 288-2'44. [6] Tanner, A.C.R., Strzempko. M.N., getsky, C.A. and McKinley,G.A. (19851J. Clin. Microbiol, 22, 333-335. [7] Van Winkelhoff,A,J., V~m Steenbergen, J,M., Kippuw, N. and De Graaff. J. (1986) Anton. Van Leeuwenhoek 52, 163-171. [8] Loesche,W,J. (1986) Microbiol. Rev. 50, 353-381). [9] Herbert, D., Phipps, P.J. and Strange, R.E. (1971) In: Methods in Microbiology(Navis, R.J, and Ribbons. D.W., Eds.), Vol. 5B, pp. 2111-344.AcademicPress, London. [16] Homer, K.A, and Beighton, D. (19911)Anal. Biochem. 191, 133-137. [11] Gharbia, S.E., Shah, H.N. and Welch. S.G. {1989) Curt, Microbiol. 19, 231-235. [121 Smith, K. and Beighton, D. (1986) J. Dent. Res. 65, 1340-1352. [13] Beighton, D., Smith, K., Gtenister. D.A.. Salamon, K. and Keevil, C,W. (1988) Microb. Ecol. Health Dis. I, 85-94.