- Email: [email protected]
Coaggregation profiles of the microflora from root surface caries lesions
Archives of Oral Biology (2005) 50, 23—32
www.intl.elsevierhealth.com/journals/arob
Coaggregation profiles of the microflora from root surface caries lesions S. Shen, L.P. Samaranayake*, H.-K. Yip Oral Bio-sciences, Faculty of Dentistry, The University of Hong Kong, Prince Philip Dental Hospital, 34 Hospital Road, Hong Kong SAR, China Accepted 1 July 2004
KEYWORDS Microflora; Bacterial coaggregation; Dental plaque; Biofilms
Summary Bacterial coaggregation reactions between different species and the auto aggregation of the same species are associated with the initiation and development of dental plaque and biofilms. As no such data is available on isolates from root caries lesions, we evaluated, by a visual, semi-quantitative scoring system and a spectrophotometric, quantitative assay, the coaggregation of 22 different wild-type microbial species comprising ten bacterial genera and a single Candida spp. The quantitative coaggregation assay we used proved to be a more sensitive method than the semi-quantitative, visual evaluation as the results yielded the percent coaggregation. Fusobacterium nucleatum, Lactobacillus acidophilus, Streptococcus bovis II/2 and Gemella morbillorum were observed having higher degrees of autoaggregation than the other examined strains. Significant levels of inter-species coaggregation was seen between: (1) Actinomyces spp. and Veillonella spp.; (2) Actinomyces israelii and Peptostreptococcus prevotii; (3) Campylobacter gracilis and Actinomyces spp.; (4) Prevotella intermedia and nine different species; and (5) Fusobacterium nucleatum and six other species. The single Candida albicans isolate did not coaggregate to a significant extent with any of the 21 bacterial isolates studied. Scanning electron microscopy observation of the coaggregation interactions between bacterial pairs having strong coaggregation reactions revealed varying adhesive patterns. Our findings on coaggregation amongst these isolates imply existence of multiple interactions between the coaggregation-inducing bacterial species in root caries. In particular, Actinomyces spp., Veillonella spp., Prevotella spp. and Fusobacterium spp. appear to play a significant role in this context. # 2004 Elsevier Ltd. All rights reserved.
Introduction * Corresponding author. Tel.: +852 2859 0480; fax: +852 2547 6133. E-mail address: [email protected] (L.P. Samaranayake).
Cariogenic bacteria have a strong tendency to adhere to tooth surfaces. Having colonized the tooth surface, the bacteria multiply, aggregate with
0003–9969/$ — see front matter # 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.archoralbio.2004.07.002
24
additional bacterial species and embed themselves in a matrix of extracellular polymers of both salivary and bacterial origin. This process leads to the formation of a biofilm on the tooth surface–—the dental plaque.1,2 Bacterial coaggregation, which is defined as cellto-cell adherence of different bacterial species or strains, is an integral accompaniment to such plaque formation.3 It differs from autoaggregation, which is defined as the adherence of bacteria belonging to the same strain.4 Both intra- and inter-species aggregation interactions play a major role in the formation of dental plaque on host tooth surfaces. The coaggregation interaction is a highly specific process mediated by the recognition of complementary lectin—carbohydrate molecules between the aggregating partners.5 Although the lectin—carbohydrate-mediated coaggregation can be reversed by the addition of simple sugars and chelating agents, these mediators of coaggregation demonstrate a genetic stability that is not affected by either the cell age of the partners or the culture media used.3,5,6 Studies on coaggregation interactions of plaque bacteria were pioneered by Gibbons and Nygaard7 and since then many workers have investigated the phenomenon using a number of techniques .4 In early studies, positive coaggregation reactions of bacteria isolated from initial plaque were found when Actinomyces naeslundii, including former A. viscosus strains, were paired with Streptococcus sanguis or S. mitis.6 Some strains of plaque bacteria belonging to other genera, such as Actinobacillus spp., Prevotella spp. (including most strains belonging to Bacteroides spp. prior to change of nomenclature), Capnocytophaga spp., Fusobacterium spp., Rothia spp. and Veillonella spp., were also found possessing coaggregation properties.3 Accumulating evidences has indicated that several microorganisms are involved in root surface caries lesions.8—11 Although different groups have found varying proportions of common oral bacteria in root caries, there is consensus that Actinomyces spp., Lactobacillus spp. and Streptococcus spp. play a major role in this disease process, somewhat akin to occlusal caries.8—14 On perusal of the literature we were unable to find data on the autoaggregation or coaggregation interactions of bacteria isolated from root caries lesions. Hence the main objectives of this study were: (1) to assess the intra- and interspecies coaggregation interactions of 22 wild-type root caries isolates belonging to 11 different genera; (2) to evaluate and compare the coaggregation interactions of these isolates using a visual, semiquantitative assay and a spectrophotometric, quantitative assay; and (3) to observe the ultrastructural
S. Shen et al.
features of coaggregation reactions between bacterial pairs which demonstrate strong coaggregation, using scanning electron microscopy (SEM).
Materials and methods Microorganisms used In total 22 different microbial isolates, obtained from root surface caries lesions in elderly Chinese subjects as described previously,11 were used. In brief, the organisms were isolated from mixed cultures of bacteria derived from altered dentine of root caries lesions in elderly ethnic Chinese as described previously.11 The purity and the identity of the organisms were reconfirmed using API 20 Strep Identification System, API 20 C AUX Yeast Identification System (bioMe ´rieux sa, Marcyl’Etoile, France) and RapID ANA II Identification System (Innovative Diagnostic Systems, L.P., Norcross, GA, USA). The fresh, wild-type bacteria were then aliquoted in 1.5 mL trypticase soy broth (TSB) (Difco Laboratories, Detroit, Michigan, USA) vials stored at 70 8C, as the yeast isolates were stored in 1.5 mL distilled water at 30 8C, until the experiments were performed. In total the isolates, which were chosen randomly from the pool of identified wild-type strains, belonged to 11 different genera including Actinomyces spp. (2 species), Campylobacter spp. (1 species), Capnocytophaga spp. (1 species), Fusobacterium spp. (1 species), Gemella spp. (1 species), Lactobacillus spp. (2 species), Peptostreptococcus spp. (2 species), Prevotella spp. (1 species), Streptococcus spp. (9 species), Veillonella spp. (1 species) and Candida spp. (1 species). Two stock cultures of reference bacterial strains, Fusobacterium nucleatum ATCC 10953 and S. sanguinis (previous S. sanguis) ATCC 10558, which exhibited strong coaggregation reactions in previous studies of ours,15 served as positive controls (Table 1).
Pre-culture of microorganisms Recovery of the stock microorganisms 1.5 mL stock microorganisms were thawed and transferred onto agar media. Streptococci were inoculated on Columbia blood agar (Oxoid, Unipath Ltd., Wade Road, Basingstoke, Hampshire, UK) and incubated aerobically at 37 8C for 18—24 h. Actinomyces spp., Campylobacter spp., Capnocytophaga spp., Fusobacterium spp., Gemella spp., Peptostreptococcus spp., Prevotella spp. and Veillonella spp. were inoculated on Columbia blood agar
Coaggregation profiles of the microflora from root surface caries lesions
Table 1 Microorganisms used in bacterial coaggregation assays. Genus
Species and strain numbera
Streptococcus spp.
S. acidominimus ND2-3 S. bovis II/2 ND2-2 S. equinus ND1-2 S. constellatus ND10-13A S. milleri II ND9-11 S. mitis ND10-2 S. mutans ND13-1A S. sanguinis I ND4-7 S. sanguinis II ND7-3 Gemella spp. G. morbillorum ND2-8 Peptostreptococcus spp. P. magnus NT1-2A P. prevotii ND1-6A Veillonella spp. V. parvula ND8-6A Lactobacillus spp. L. acidophilus ND7-2A L. minutis ND2-1A Actinomyces spp. A. israelii NT6-2A A. odontolyticus ND2-2A Capnocytophaga spp. C. sputigena ND2-12A Campylobacter spp. C. gracilis ND9-8A Prevotella spp. P. intermedia ND8-9A Fusobacterium spp. F. nucleatum NT6-6A Candida spp. C. albicans ND2-1S Positive Control S. sanguinis ATCC 10558 F. nucleatum ATCC 10953 a All strains were fresh, wild-type isolates from human root caries lesions except for the positive controls derived from the American Type Culture Collection (ATCC).
(Oxoid) supplemented with 5% sheep blood and incubated anaerobically at 37 8C for 18—24 h. Lactobacilli were inoculated on Rogosa agar (Difco) and incubated anaerobically at 37 8C for 18—24 h. Candida albicans was inoculated on Sabouraud’s dextrose agar (Oxoid) and incubated aerobically at 37 8C for 24—48 h. Preparation of inocula for the coaggregation assays One to two loopfuls of each recovered culture was transferred to a complex medium containing 5 g/L tryptone (Difco), 5 g/L yeast extract (Difco), 0.05% v/v Tween 80 (0.5 mL/L) (Sigma Chemical Co., St. Louis, MO, USA) and glucose (0.2%) (Difco) buffered to pH 7.5 with K2HPO46,16 and incubated at 37 8C. Candida spp. were incubated aerobically for 24 h. Actinomyces spp., Campylobacter spp., Capnocytophaga spp., Fusobacterium spp., Gemella spp., Lactobacillus spp., Peptostreptococcus spp., Prevotella spp., Streptococcus spp. and Veillonella spp. were incubated anaerobically for two to three days.11,15 After incubation, cultures were centrifuged for 10 min at 10,000 rpm (Centrifuge GS-15R, Beckman Instruments, Inc., Palo Alto, CA, USA). The deposit
25
was washed twice in phosphate buffered saline (PBS) (0.1 M, pH 7.3), then dispersed in coaggregation buffer consisting CaCl2 (1 104 M), MgCl2 (1 104 M), NaN3 (0.02%) and NaCl (0.15 M) (dissolved in 0.001 M Tris adjusted to pH 8.0).6 The microbial suspensions were adjusted to a concentration of 5 109 microorganisms per mL, using a spectrophotometer (Pharmacia LKB, Ultraspec III, Pharmacia Inc., USA) at a wavelength of 660 nm for bacteria and 520 nm for yeasts. Correlation between viable counts and optical density measurements was ensured by quantification of colony forming units (CFUs) spiral plated (Spiral System, Model DU Inc., Cincinnati, Ohio, USA) on to appropriate solid media.15
Coaggregation assays Visual, semi-quantitative assay All coaggregation assays were conducted according to Cisar et al.6 and Kolenbrander et al.17 as follows. Autoaggregation reactions were measured by visual observation of the degree of aggregation of the bacterial inocula over a period of 60 min. Interspecies coaggregation was evaluated by pairing the bacteria as shown in Table 1 using the following technique. Equal volumes (0.2 mL) of each of two microbial suspensions were mixed in sterile test tubes and vortexed for 10 s on a vortex mixer (Autovortex mixer SA2, Stuart Scientific Co. Ltd., UK). Tubes containing: (a) coaggregation buffer solution alone; and (b) each microbial test suspension alone, served as negative controls. Type cultures of F. nucleatum ATCC 10953 and S. sanguinis ATCC 10558 that previously demonstrated strong coaggregation reactions and served as a positive control.15 After mixing, the suspensions were scored immediately for coaggregation. The suspensions were then kept at room temperature for 1 h, mixed for 10 s on the vortex mixer, and scored again.15 The scoring system of Cisar et al.6 was used to evaluate the degree of coaggregation in the mixed suspensions by naked eye visualization. In brief, this system has a scoring range from ‘‘’’ to ‘‘4+’’ as follows: ‘‘’’ shows no change in turbidity and no evidence of coaggregates in the mixed suspensions; ‘‘1+’’ represents turbid supernatant with finely dispersed coaggregates; ‘‘2+’’ demonstrates definite coaggregates which do not precipitate immediately; ‘‘3+’’ manifests slightly turbid supernatant with formation of large precipitating coaggregates; and ‘‘4+’’ exhibits clear supernatant and large coaggregates which precipitate immediately. All assays were conducted in duplicate on two different occasions.
26
S. Shen et al.
Spectrophotometric, quantitative assay Measurement of the autoaggregation percentage of each organism was evaluated as follows: 1 mL bacterial inocula was transferred into a sterile cuvette (Elkay Ultra-VU1 Disposable cuvettes, Micro Square, Elkay Products Inc., Shrewsburg, MA, USA) and the optical density (OD) was recorded over a period of 60 min using the spectrophotometer at a wavelength of 660 nm (520 nm for Candida albicans). The autoaggregation degree of each of the 22 organisms was calculated as the percent decrease of OD after 60 min using Eq. (1): % Autoaggregation ¼
OD0 OD60 100 OD0
trifuged at 2000 rpm for 2 min and the supernatant of the mixture (about 0.6 mL) was gently transferred to a new sterile cuvette. The OD of the supernatant was then recorded using the spectrophotometer at a wavelength of 660 nm. The quantitative coaggregation rate of each pair of the coaggregation partners were calculated using Eq. (2):5 % Coaggregation ¼
ðODA þ ODB Þ 2 ODðAþBÞ ODA þ ODB 100 (2)
(1)
where OD0 is the optical density of the microorganism measured immediately after the inocula was transferred into the cuvette; and OD60 is the optical density of the microorganism measured 60 min later after the inocula was transferred. Evaluation of the degree of coaggregation of each pair of the organisms was performed as–—equal volumes (0.5 mL) of each of two paired microbial inocula were mixed in a sterile cuvette (Elkay), vortexed for 10 s and kept at room temperature for a period of 60 min. The mixture was then cen-
where ODA and ODB stand for the optical densities of bacterium A and bacterium B, respectively; and OD(A+B) stands for the optical density of the mixture of inocula of bacterium A and bacterium B.
Observation of the bacterial pairs with strong coaggregation using SEM The ultrastructural features of four bacterial pairs showing strong coaggregation interactions were observed by scanning electron microscopy. These were F. nucleatum ATCC 10953 with S. sanguinis ATCC 10558, which served as a positive control; A.
Figure 1 The correlation of data derived from coaggregating profiles of 231 pairs of organisms using the semiquantitative, visual assay and quantitative, spectrophotometric assay (r = 0.955, P < 0.001).
Coaggregation profiles of the microflora from root surface caries lesions
israelii NT6-2A with P. prevotii ND1-6A; A. israelii NT6-2A with V. parvula ND8-6A; and A. odontolyticus ND2-2A with V. parvula ND8-6A–—all of which showed ‘‘4+’’ degree coaggregations. For this purpose the unmixed bacterial inocula prepared as described above were mixed, allowed to coaggregate, and one drop of the mixed suspension was transferred onto a cover glass and air-dried at room temperature. The specimen was then fixed in 1% osmium tetraoxide (OsO4) vapor, dehydrated thoroughly in a freeze drying system (Labconco Freeze Dry System/Freezone1 4.5, Kansas City, Missouri 64132, USA), sputter coated with palladium—gold to a thickness of around 20 nm (Jeol Fine Coat, Ion Sputter JFC-1100, Jeol, Tokyo, Japan) and observed using a scanning electron microscope (Philips XL 30CP, Philips Electronics N.V., Eindhoven, the Netherlands).
Statistical analysis The correlation coefficient between the results from visual, semi-quantitative assay (graded from 0 to 4) and spectrophotometric, quantitative assay (ranged from 0% to 78.4%) was calculated using SPSS statistical program (SPSS version 8.0, SPSS Inc., USA).
Results Comparison of the visual and optical assays The optical assay showed a much higher sensitivity than the visual assay. However, despite the varying sensitivities of the two assays used, there was highly significant concordance in the results between the semi-quantitative, visual method and the quantitative, spectrophotometric assay (r = 0.955, P < 0.001) (Fig. 1).
Inter-species coaggregation–—visual, semi-quantitative assay Coaggregation of the 21 species of bacteria was evaluated using visual, semi-quantitative assay. Of the 210 bacterial pairs studied, 29 pairs (13.8%) had varying degrees of positive coaggregation. These data are shown in the upper right half of Table 2 P. intermedia ND8-9A and F. nucleatum NT6-6A coaggregated with 9 and 6 other species, respectively. The ‘‘4+’’ degree of coaggregation reactions occurred in the mixtures of A. israelii NT6-2A with V. parvula ND8-6A, A. odontolyticus ND2-2A with V. parvula ND8-6A and A. israelii NT62A with P. prevotii ND1-6A. The ‘‘3+’’, ‘‘2+’’ and
27
‘‘1+’’ degrees of coaggregation were recorded in three, seven and sixteen bacterial pairs, respectively (Table 2).
Inter-species coaggregation–— spectrophotometric, quantitative assay Percentages of coaggregation interactions examined using the quantitative coaggregation assay are shown in lower left half of Table 2. The percent coaggregation interactions ranged from 0% to 78.4%. Overall, 48 of 210 (22.9%) bacterial pairs showed coaggregation percentiles higher than 5.0%. The strongest degree of coaggregation reactions were observed in mixtures of A. israelii NT62A with V. parvula ND8-6A (78.4% coaggregation), A. odontolyticus ND2-2A with V. parvula ND8-6A (61.1%), A. israelii NT6-2A with P. prevotii ND16A (66.7%), C. gracilis ND9-8A with A. odontolyticus ND2-2A (49.7%), P. intermedia ND8-9A with L. minutis ND2-1A (47.1%) and C. gracilis ND9-8A with A. israelii NT6-2A (46.6%).
Degree of autoaggregation of bacterial isolates The autoaggregation reactions of the bacteria in percent are also shown in the diagonal column of Table 2. F. nucleatum NT6-6A (55.0%), L. acidophilus ND7-2A (35.1%), S. bovis II/2 ND2-2 (27.8%), G. morbillorum ND2-8 (21.6%), P. magnus NT1-2A (12.5%), P. intermedia ND8-9A (10.1%) and S. mutans ND13-1A (10.0%) all demonstrated relatively higher autoaggregation compared with their counterparts during the 60-min observation period.
Coaggregation of Candida albicans and bacteria The single isolate of C. albicans ND2-1S that were used in the study, did not coaggregate to a significant extent with any of the bacteria. However, S. acidominimus ND2-3 (8.7%), S. equinus ND1-2 (6.2%) and S. constellatus ND10-13A (6.0%) showed some degree of avidity to candidal blastospores as they demonstrated more than 5.0% coaggregation in the spectrophotometric assay, but not in the visual assay (Table 2). Candida blastospores did not demonstrate significant autoaggregation (7.6%) (Table 2).
SEM findings When the coaggregation patterns of bacterial pairs having strong coaggregation reactions were examined using scanning electron microscopy, several
28
S. Shen et al.
Table 2 A catacomb illustrating the results of quantitative (Y-axis) and semi-quantitative (X-axis) assessment of coaggregation reactions of 22 microorganisms isolated from root caries lesions.
(a) The emboldened and italicized figures in the diagonal grille indicate the percentage of autoaggregation interactions in the quantitative, spectrophotometric assay; (b) Figures with the ‘+’ marks in the upper right half of the catacomb indicate the coaggregation degree measured by the semi-quantitative, visual assay; and (c) Emboldened figures in the lower left half of the table indicate the percentages of coaggregation interactions greater than 5.0% as measured by the quantitative, spectrophotometric assay.
patterns, ranging from simple cell-to-cell adhesion to the formation of a complex coaggregation network, were found (Fig. 2). One of the most common observations was a single rod coaggregating with multiple cocci (Fig. 2b and c). However, no typical ‘‘corn— cob’’ structure was observed between ‘‘rod and coccal form’’ coaggregation partners. A detailed observation of the adhesion patterns (Fig. 2h) demonstrated tight adherence between some coaggregating bacteria that led to the virtual loss of discernible cell walls of the bacterial partners.
Discussion All 22 microorganisms used in the present study were isolated from 30 different root surface caries lesions in 18 different elderly, ethnic Chinese and were randomly chosen from the pool of identified wild-type isolates belonging to 11 different genera.11 Amongst 210 bacterial pairs evaluated here, 29 (13.8%) were found to demonstrate positive coaggregation reactions. This rate is slightly higher than the findings of Gibbons and Nygaard7 who
Coaggregation profiles of the microflora from root surface caries lesions
29
Figure 2 Several patterns of coaggregation reactions between bacterial pairs having strong coaggregation (scored ‘4+’) reactions are shown in this composite figure. (a) The simplest pattern of coaggregation wherein a single F. nucleatum ATCC 10953 adheres to a single S. sanguinis ATCC 10558 (6000); (b) A single F. nucleatum ATCC 10953 coaggregating with multiple, paired and chained S. sanguinis ATCC 10558 cells (6000); (c) An A. israelii NT6-2A coaggregating with multiple, unchained P. prevotii ND1-6A cells (12,000). Multiple reacting sites appear to exist on the surface of A. israelii; (d) Formation of a complex network with a large sheaf of aggregated F. nucleatum ATCC 10953 and multiple S. sanguinis ATCC 10558 (6,000); (e) Formation of a complex coaggregation network of A. israelii NT6-2A and P. prevotii ND1-6A (6,000); (f) Formation of a complex coaggregation network A. odontolyticus ND2-2A and V. parvula ND8-6A (6,000); (g) Formation of a complex coaggregation network A. israelii NT6-2A and V. parvula ND8-6A (6,000); (h) A higher magnification of a coaggregating site of F. nucleatum ATCC 10953 and S. sanguinis ATCC 10558 (60,000). The cell walls of both coaggregation partners appear to merge so that the confines of the two cells are not clearly visible.
30
tested 23 bacterial strains including several prominent plaque species and observed 23 coaggregating pairs in a total of 253 (9.1%) bacterial pairs. It has been reported that the bacteria isolated from one single site show higher degree of positive coaggregation reactions than those from different sites.18 Perhaps the low frequency of positive coaggregating pairs seen in the current study is likely to be due to their varying origin, as the organisms were mainly from different root lesions in many individuals. Kolenbrander et al.19 evaluated 122 human oral bacteria comprising 11 genera and observed intrageneric coaggregation reactions in S. oralis, S. sanguis, S. mitis and S. gordonii species and no such reactions amongst species belonging to ‘‘secondary" plaque colonizers such as Fusobacterium spp., Prevotella spp. and Lactobacillus spp. The unique intrageneric coaggregation property possessed by the pioneer colonizers, i.e. Streptococcus spp. and Actinomyces spp., appear to be a major advantage for these bacteria to initiate colonization at clean tooth surfaces. In the present study, a relatively higher degree of intra-generic coaggregation was recorded in S. bovis II/2 ND2-2 (27.8%) and S. mutans ND13-1A (10.0%) and not the other six species of streptococci. It appears therefore that streptococci isolated from well-established root caries lesions have a lesser propensity for intra-generic coaggregation hence their establishment on root surface carious niches may be due to mechanisms unrelated to intra-generic coaggregation. On the contrary, several other wildtype species, such as F. nucleatum NT-6-6A (55.0%), L. acidophilus ND7-2A (35.1%), S. bovis II/2 ND2-2 (27.8%) and G. morbillorum ND2-8 (21.6%) also showed strong autoaggregation interactions implying their ability for autonomous colonization and accretion on root caries lesions. With respect to interspecies interactions the ‘‘4+’’ or the greatest degree of coaggregation was recorded with the following combinations–—A. israelii NT6-2A with V. parvula ND8-6A (78.4% coaggregation), A. odontolyticus ND2-2A with V. parvula ND86A (61.1%) and A. israelii NT6-2A with P. prevotii ND1-6A (66.7%). The results are in agreement with previous studies which showed that Actinomyces spp. plays major role in coaggregation interactions.6,7,18,20 Interestingly though the strong coaggregation bond between A. odontolyticus or A. israelii with strains of either V. parvula or P. prevotii has not been reported hitherto fore. Previous detailed studies on the coaggregation interactions between Streptococcus spp. and Actinomyces spp. have revealed the complementary adhesion-receptor mechanisms amongst the latter organisms.6,18,21 It is also known that fimbriae are important in adherence of some strains A. naeslundii (including former
S. Shen et al. A. viscosus strains) to other cells or surfaces.22 However, little data are available on the mechanisms mediating coaggregation between A. odontolyticus and V. parvula or between A. israelii and P. prevotii. As Actinomyces spp. are predominantly isolated from root caries lesions and are considered key pathogens in this process,8,10,11,13,23,24 our results point to possible existence of ligand—receptor interactions between A. odontolyticus and V. parvula and between A. israelii and P. prevotii. Further biochemical studies on the latter interactions are warranted to clarify these mechanisms. P. intermedia ND8-9A, F. nucleatum NT6-6A and C. gracilis ND9-8A were the ‘‘top three’’ bacteria that coaggregated with multiple bacterial species in the present study. It is known that cell surface appendages on Bacteroides spp. serve as coaggregation bridges mediating their adhesion to other bacteria in dental plaque.17,25 Due to taxonomic reclassification of Bacteroides spp. most of the species belonging to this genus are now designated in the genera Prevotella and Campylobacter. 26 We evaluated the coaggregation between either Campylobacter species (C. gracilis ND9-8A, previous Bacteroides gracilis) or Prevotella species (P. intermedia ND8-9A, previous Bacteroides intermedia) with 20 species of other bacteria. A large proportion of positive coaggregating pairs between either P. intermedia ND8-9A or C. gracilis ND9-8A with strains of Streptococcus spp., Gemella spp., Peptostreptococcus spp., Lactobacillus spp. and Actinomyces spp. indicate that the Gram-negative obligate anaerobic rods may play an important role in the interactions leading to root caries formation. Furthermore our findings of the coaggregation between F. nucleatum NT6-6A and six other bacterial species, i.e. S. bovis II/2 ND2-2, S. constellatus ND10-13A, S. sanguinis (previous S. sanguis) II ND73, L. acidophilus ND7-2A, C. sputigena ND2-12A and P. intermedia ND8-9A, reconfirm the reports that these obligate, anaerobic, fusiform organisms have the ability to coaggregate with a large representation of oral bacteria acting as a key organism in the dental plaque formation possibly during latter stages of plaque maturation and modulation of the climax community.2,27 Candida spp. have been reported to be associated with root caries.28—30 Previous workers have demonstrated that C. albicans may coaggregate with a range of oral bacteria including S. sanguis, S. salivarius, S. mutans, S. mitis,31 S. gordonii, S. oralis, S. anginosus, S. pyogenes,32 L. amylovorus,33 F. nucleatum31,33 and A. viscosus.31 However, in the present study, we failed to record significant coaggregation reactions between the single C. albicans isolate and other 21 bacteria, except for the
Coaggregation profiles of the microflora from root surface caries lesions
weak coaggregation reactions with S. acidominimus ND2-3 (8.7%), S. equinus ND1-2 (6.2%) and S. constellatus ND10-13A (6.0%). It seems that the coaggregation mechanisms play an insignificant role in C. albicans colonization associated with root caries. Indeed this may be one reason why Candida species comprise a very low percent of flora in root caries.11,13,29 The SEM observation of interactions between bacterial pairs that showed strong coaggregation revealed interesting adhesive characteristic between the coaggregating partners. F. nucleatum ATCC 10953, A. israelii NT6-2A and A. odontolyticus ND2-2A demonstrated multiple receptor sites on each cell surface that appeared to react with several cocci. The rods in particular served as a framework that allowed the coccal cells belonging to S. sanguinis ATCC 10558, V. parvula ND8-6A and P. prevotii ND1-6A to adhere. The formation of cross-links between F. nucleatum ATCC 10953 and S. sanguinis ATCC 10558, A. israelii NT6-2A and V. parvula ND8-6A, A. odontolyticus ND2-2A and V. parvula ND8-6A as well as A. israelii NT6-2A and P. prevotii ND1-6A cells, appears to be the foundation of a fairly intricate structure that sustains the complex architectural features of the biomass of the root lesions.34 Detailed observation of the coaggregation foci between partners demonstrated very tight interactions, wherein the cell walls merged resulting in seamless confluence of the organisms (Fig. 2h). However, we were unable to detect ‘‘corn—cob’’ formation in the present study. It has been reported that this special structure was usually found at periphery of developing plaque35,36 and replication of this structure could be accomplished in vitro by mixing rods and cocci at a ratio of 1:10.37,38 To conclude, further work with varying groups of wild-type bacteria isolated from root caries lesions is required to elucidate the mechanisms underlying intra- and inter-generic coaggregation reactions that may be unique to this ecological niche, which has been sparsely studied. However, the current data which appear to be the first detailed study of coaggregation reactions of root caries bacteria should serve as basic information for future workers.
Acknowledgement The authors would like to thank Miss Joyce Y.Y. Yau and Mr. Simon Lee for the assistance during the experiment. This study was supported by a grant (a/c 10201963) from the Committee on Research
31
and Conference Grants (CRCG), the University of Hong Kong, Hong Kong SAR, China.
References 1. Marsh P, Martin M. Oral Microbiology. 4th edn. Oxford: Wright; 1999. 2. Kolenbrander PE. Oral microbial communities: biofilms, interactions, and genetic systems. Ann Rev Microbiol 2000;54:413—37. 3. Kolenbrander PE. Intergeneric coaggregation among human oral bacteria and ecology of dental plaque. Ann Rev Microbiol 1988;42:627—56. 4. Kinder S, Holt SC. Coaggregation between bacterial species. Methods Enzymol 1994;236:254—70. 5. McIntire FC, Vatter AE, Baros J, Arnold J. Mechanism of coaggregation between Actinomyces viscosus T14V and Streptococcus sanguis 34. Infect Immun 1978;21(3):978—88. 6. Cisar JO, Kolenbrander PE, McIntyre FC. Specificity of coaggregation reactions between human oral streptococci and strains of Actinomyces viscosus or Actinomyces naeslundii. Infect Immun 1979;24(3):742—52. 7. Gibbons RJ, Nygaard M. Interbacterial aggregation of plaque bacteria. Arch Oral Biol 1970;15(12):1397—400. 8. van Houte J, Lopman J, Kent R. The predominant cultivable flora of sound and carious human root surfaces. J Dent Res 1994;73(11):1727—34. 9. Schupbach P, Osterwalder V, Guggenheim B. Human root caries: microbiota in plaque covering sound, carious and arrested carious root surfaces. Caries Res 1995;29(5):382— 95. 10. Schupbach P, Osterwalder V, Guggenheim B. Human root caries: microbiota of a limited number of root caries lesions. Caries Res 1996;30(1):52—64. 11. Shen S, Samaranayake LP, Yip H-K, Dyson JE. Bacterial and yeast flora of root surface caries in elderly, ethnic Chinese. Oral Dis 2002;8(4):207—17. 12. Ellen RP, Banting DW, Fillery ED. Longitudinal microbiological investigation of hospitalized population of older adults with a high root surface caries risk. J Dent Res 1985;64(12):1377—81. 13. Brown LR, Billings RJ, Kaster AG. Quantitative comparisons of potentially cariogenic microorganisms cultured from noncarious and carious root and coronal tooth. Infect Immun 1986;51(3):765—70. 14. Keltjens HMA, Schaeken MJM, van der Hoeven JS, Hendriks JC. Microflora of plaque from sound and carious root surfaces. Caries Res 1987;21(3):193—9. 15. Sedgley CM. Enterobacteriaceae in the human oral cavity. (Ph. D. Thesis). The University of Hong Kong: Hong Kong. 1995; 214—225. 16. Maryanski JH, Wittenberger CL. Mannitol transport in Streptococcus mutans. J Bacteriol 1975;124(3):1475—81. 17. Kolenbrander PE, Andersen RN, Holdeman LV. Coaggregation of oral Bacteroides species with other bacteria: central role in coaggregation bridges and competitions. Infect Immun 1985;48(3):741—6. 18. Kolenbrander PE, Inouye Y, Holdeman LV. New Actinomyces and Streptococcus coaggregation groups among human oral isolates from the same site. Infect Immun 1983;41(2):501—6. 19. Kolenbrander PE, Andersen RN, Moore LV. Intra-generic coaggregation among strains of human oral bacteria: potential role in primary colonization of the tooth surface. Appl Environ Microbiol 1990;56(12):3890—4.
32
20. Komiyama K, Gibbons RJ. Interbacterial adherence between Actinomyces viscosus and strains of Streptococcus pyogenes, Streptococcus agalactiae, and Pseudomonas aeruginosa. Infect Immun 1984;44(1):86—90. 21. Kolenbrander PE, Williams BL. Prevalence of viridans streptococci exhibiting lactose-inhibitable coaggregation with oral actinomycetes. Infect Immun 1983;41(2):449—52. 22. Gibbons RJ, Hay DI, Cisar JO, Clark WB. Adsorbed salivary proline-rich protein-1 and statherin: receptors for type 1 fimbriae of Actinomyces viscosus T14V-J1 on apatitic surfaces. Infect Immun 1988;56(11):2990—3. 23. Jordan HV, Hammond BF. Filamentous bacteria isolated from human root surface caries. Arch Oral Biol 1972;17 (9):1333—42. 24. Nyvad B, Kilian M. Microbiology of the early colonization of human enamel and root surfaces in vivo. Scand J Dent Res 1987;95(5):369—80. 25. Stinson MW, Safulko K, Levine MJ. Adherence of Porphyromonas (Bacteroides) gingivalis to Streptococcus sanguis in vitro. Infect Immun 1991;59(1):102—8. 26. Shah HN, Collins MD. Prevotella, a new genus to include Bacteroides melaninogenicus and related species formerly classified in the genus Bacteroides. Int J Syst Bacteriol 1990;40(2):205—8. 27. Kolenbrander PE. Coaggregation of human oral bacteria: potential role in the accretion of dental plaque. J Appl Bacteriol 1993;74(Suppl.):79S—86S. 28. Scheinin A, Pienihakkinen K, Tiekso J, Holmberg S. Multifactorial modeling for root caries prediction. Community Dent Oral Epidemiol 1992;20(1):35—7.
S. Shen et al.
29. Beighton D, Lynch E. Relationships between yeasts and primary root-caries lesions. Gerodontology 1993;10(2):105—8. 30. Narhi TO, Vehkalahti MM, Siukosaari P, Ainamo A. Salivary findings, daily medication and root caries in the old elderly. Caries Res 1998;32(1):5—9. 31. Bagg J, Silverwood RW. Coagglutination reactions between Candida albicans and oral bacteria. J Med Microbiol 1986;22(2):165—9. 32. Jenkinson HF, Lala HC, Shepherd MG. Coaggregation of Streptococcus sanguis and other streptococci with Candida albicans. Infect Immun 1990;58(5):1429—36. 33. Hsu LY, Minah GE, Peterson DE, Wingard JR, Merz WG, Altomonte V, et al. Coaggregation of oral Candida isolates with bacteria from bone marrow transplant recipients. J Clin Microbiol 1990;28(12):2621—6. 34. Kolenbrander PE, London J. Adhere today, here tomorrow: oral bacterial adherence. J Bacteriol 1993;175(11):3247— 52. 35. Jones SJ. A special relationship between spherical and filamentous microorganisms in mature human dental plaque. Arch Oral Biol 1972;17(3):613—6. 36. Listgarten MA. Structure of the microbial flora associated with periodontal health and disease in man. J Periodontol 1976;47(1):1—18. 37. Kolenbrander PE, Andersen RN. Multigeneric aggregations among oral bacteria: a network of independent cell-to-cell interactions. J Bacteriol 1986;168(2):851—9. 38. Kolenbrander PE, Andersen RN. Intergeneric rosettes: sequestered surface recognition among human periodontal bacteria. Appl Environ Microbiol 1988;54(4):1046—50.
www.intl.elsevierhealth.com/journals/arob
Coaggregation profiles of the microflora from root surface caries lesions S. Shen, L.P. Samaranayake*, H.-K. Yip Oral Bio-sciences, Faculty of Dentistry, The University of Hong Kong, Prince Philip Dental Hospital, 34 Hospital Road, Hong Kong SAR, China Accepted 1 July 2004
KEYWORDS Microflora; Bacterial coaggregation; Dental plaque; Biofilms
Summary Bacterial coaggregation reactions between different species and the auto aggregation of the same species are associated with the initiation and development of dental plaque and biofilms. As no such data is available on isolates from root caries lesions, we evaluated, by a visual, semi-quantitative scoring system and a spectrophotometric, quantitative assay, the coaggregation of 22 different wild-type microbial species comprising ten bacterial genera and a single Candida spp. The quantitative coaggregation assay we used proved to be a more sensitive method than the semi-quantitative, visual evaluation as the results yielded the percent coaggregation. Fusobacterium nucleatum, Lactobacillus acidophilus, Streptococcus bovis II/2 and Gemella morbillorum were observed having higher degrees of autoaggregation than the other examined strains. Significant levels of inter-species coaggregation was seen between: (1) Actinomyces spp. and Veillonella spp.; (2) Actinomyces israelii and Peptostreptococcus prevotii; (3) Campylobacter gracilis and Actinomyces spp.; (4) Prevotella intermedia and nine different species; and (5) Fusobacterium nucleatum and six other species. The single Candida albicans isolate did not coaggregate to a significant extent with any of the 21 bacterial isolates studied. Scanning electron microscopy observation of the coaggregation interactions between bacterial pairs having strong coaggregation reactions revealed varying adhesive patterns. Our findings on coaggregation amongst these isolates imply existence of multiple interactions between the coaggregation-inducing bacterial species in root caries. In particular, Actinomyces spp., Veillonella spp., Prevotella spp. and Fusobacterium spp. appear to play a significant role in this context. # 2004 Elsevier Ltd. All rights reserved.
Introduction * Corresponding author. Tel.: +852 2859 0480; fax: +852 2547 6133. E-mail address: [email protected] (L.P. Samaranayake).
Cariogenic bacteria have a strong tendency to adhere to tooth surfaces. Having colonized the tooth surface, the bacteria multiply, aggregate with
0003–9969/$ — see front matter # 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.archoralbio.2004.07.002
24
additional bacterial species and embed themselves in a matrix of extracellular polymers of both salivary and bacterial origin. This process leads to the formation of a biofilm on the tooth surface–—the dental plaque.1,2 Bacterial coaggregation, which is defined as cellto-cell adherence of different bacterial species or strains, is an integral accompaniment to such plaque formation.3 It differs from autoaggregation, which is defined as the adherence of bacteria belonging to the same strain.4 Both intra- and inter-species aggregation interactions play a major role in the formation of dental plaque on host tooth surfaces. The coaggregation interaction is a highly specific process mediated by the recognition of complementary lectin—carbohydrate molecules between the aggregating partners.5 Although the lectin—carbohydrate-mediated coaggregation can be reversed by the addition of simple sugars and chelating agents, these mediators of coaggregation demonstrate a genetic stability that is not affected by either the cell age of the partners or the culture media used.3,5,6 Studies on coaggregation interactions of plaque bacteria were pioneered by Gibbons and Nygaard7 and since then many workers have investigated the phenomenon using a number of techniques .4 In early studies, positive coaggregation reactions of bacteria isolated from initial plaque were found when Actinomyces naeslundii, including former A. viscosus strains, were paired with Streptococcus sanguis or S. mitis.6 Some strains of plaque bacteria belonging to other genera, such as Actinobacillus spp., Prevotella spp. (including most strains belonging to Bacteroides spp. prior to change of nomenclature), Capnocytophaga spp., Fusobacterium spp., Rothia spp. and Veillonella spp., were also found possessing coaggregation properties.3 Accumulating evidences has indicated that several microorganisms are involved in root surface caries lesions.8—11 Although different groups have found varying proportions of common oral bacteria in root caries, there is consensus that Actinomyces spp., Lactobacillus spp. and Streptococcus spp. play a major role in this disease process, somewhat akin to occlusal caries.8—14 On perusal of the literature we were unable to find data on the autoaggregation or coaggregation interactions of bacteria isolated from root caries lesions. Hence the main objectives of this study were: (1) to assess the intra- and interspecies coaggregation interactions of 22 wild-type root caries isolates belonging to 11 different genera; (2) to evaluate and compare the coaggregation interactions of these isolates using a visual, semiquantitative assay and a spectrophotometric, quantitative assay; and (3) to observe the ultrastructural
S. Shen et al.
features of coaggregation reactions between bacterial pairs which demonstrate strong coaggregation, using scanning electron microscopy (SEM).
Materials and methods Microorganisms used In total 22 different microbial isolates, obtained from root surface caries lesions in elderly Chinese subjects as described previously,11 were used. In brief, the organisms were isolated from mixed cultures of bacteria derived from altered dentine of root caries lesions in elderly ethnic Chinese as described previously.11 The purity and the identity of the organisms were reconfirmed using API 20 Strep Identification System, API 20 C AUX Yeast Identification System (bioMe ´rieux sa, Marcyl’Etoile, France) and RapID ANA II Identification System (Innovative Diagnostic Systems, L.P., Norcross, GA, USA). The fresh, wild-type bacteria were then aliquoted in 1.5 mL trypticase soy broth (TSB) (Difco Laboratories, Detroit, Michigan, USA) vials stored at 70 8C, as the yeast isolates were stored in 1.5 mL distilled water at 30 8C, until the experiments were performed. In total the isolates, which were chosen randomly from the pool of identified wild-type strains, belonged to 11 different genera including Actinomyces spp. (2 species), Campylobacter spp. (1 species), Capnocytophaga spp. (1 species), Fusobacterium spp. (1 species), Gemella spp. (1 species), Lactobacillus spp. (2 species), Peptostreptococcus spp. (2 species), Prevotella spp. (1 species), Streptococcus spp. (9 species), Veillonella spp. (1 species) and Candida spp. (1 species). Two stock cultures of reference bacterial strains, Fusobacterium nucleatum ATCC 10953 and S. sanguinis (previous S. sanguis) ATCC 10558, which exhibited strong coaggregation reactions in previous studies of ours,15 served as positive controls (Table 1).
Pre-culture of microorganisms Recovery of the stock microorganisms 1.5 mL stock microorganisms were thawed and transferred onto agar media. Streptococci were inoculated on Columbia blood agar (Oxoid, Unipath Ltd., Wade Road, Basingstoke, Hampshire, UK) and incubated aerobically at 37 8C for 18—24 h. Actinomyces spp., Campylobacter spp., Capnocytophaga spp., Fusobacterium spp., Gemella spp., Peptostreptococcus spp., Prevotella spp. and Veillonella spp. were inoculated on Columbia blood agar
Coaggregation profiles of the microflora from root surface caries lesions
Table 1 Microorganisms used in bacterial coaggregation assays. Genus
Species and strain numbera
Streptococcus spp.
S. acidominimus ND2-3 S. bovis II/2 ND2-2 S. equinus ND1-2 S. constellatus ND10-13A S. milleri II ND9-11 S. mitis ND10-2 S. mutans ND13-1A S. sanguinis I ND4-7 S. sanguinis II ND7-3 Gemella spp. G. morbillorum ND2-8 Peptostreptococcus spp. P. magnus NT1-2A P. prevotii ND1-6A Veillonella spp. V. parvula ND8-6A Lactobacillus spp. L. acidophilus ND7-2A L. minutis ND2-1A Actinomyces spp. A. israelii NT6-2A A. odontolyticus ND2-2A Capnocytophaga spp. C. sputigena ND2-12A Campylobacter spp. C. gracilis ND9-8A Prevotella spp. P. intermedia ND8-9A Fusobacterium spp. F. nucleatum NT6-6A Candida spp. C. albicans ND2-1S Positive Control S. sanguinis ATCC 10558 F. nucleatum ATCC 10953 a All strains were fresh, wild-type isolates from human root caries lesions except for the positive controls derived from the American Type Culture Collection (ATCC).
(Oxoid) supplemented with 5% sheep blood and incubated anaerobically at 37 8C for 18—24 h. Lactobacilli were inoculated on Rogosa agar (Difco) and incubated anaerobically at 37 8C for 18—24 h. Candida albicans was inoculated on Sabouraud’s dextrose agar (Oxoid) and incubated aerobically at 37 8C for 24—48 h. Preparation of inocula for the coaggregation assays One to two loopfuls of each recovered culture was transferred to a complex medium containing 5 g/L tryptone (Difco), 5 g/L yeast extract (Difco), 0.05% v/v Tween 80 (0.5 mL/L) (Sigma Chemical Co., St. Louis, MO, USA) and glucose (0.2%) (Difco) buffered to pH 7.5 with K2HPO46,16 and incubated at 37 8C. Candida spp. were incubated aerobically for 24 h. Actinomyces spp., Campylobacter spp., Capnocytophaga spp., Fusobacterium spp., Gemella spp., Lactobacillus spp., Peptostreptococcus spp., Prevotella spp., Streptococcus spp. and Veillonella spp. were incubated anaerobically for two to three days.11,15 After incubation, cultures were centrifuged for 10 min at 10,000 rpm (Centrifuge GS-15R, Beckman Instruments, Inc., Palo Alto, CA, USA). The deposit
25
was washed twice in phosphate buffered saline (PBS) (0.1 M, pH 7.3), then dispersed in coaggregation buffer consisting CaCl2 (1 104 M), MgCl2 (1 104 M), NaN3 (0.02%) and NaCl (0.15 M) (dissolved in 0.001 M Tris adjusted to pH 8.0).6 The microbial suspensions were adjusted to a concentration of 5 109 microorganisms per mL, using a spectrophotometer (Pharmacia LKB, Ultraspec III, Pharmacia Inc., USA) at a wavelength of 660 nm for bacteria and 520 nm for yeasts. Correlation between viable counts and optical density measurements was ensured by quantification of colony forming units (CFUs) spiral plated (Spiral System, Model DU Inc., Cincinnati, Ohio, USA) on to appropriate solid media.15
Coaggregation assays Visual, semi-quantitative assay All coaggregation assays were conducted according to Cisar et al.6 and Kolenbrander et al.17 as follows. Autoaggregation reactions were measured by visual observation of the degree of aggregation of the bacterial inocula over a period of 60 min. Interspecies coaggregation was evaluated by pairing the bacteria as shown in Table 1 using the following technique. Equal volumes (0.2 mL) of each of two microbial suspensions were mixed in sterile test tubes and vortexed for 10 s on a vortex mixer (Autovortex mixer SA2, Stuart Scientific Co. Ltd., UK). Tubes containing: (a) coaggregation buffer solution alone; and (b) each microbial test suspension alone, served as negative controls. Type cultures of F. nucleatum ATCC 10953 and S. sanguinis ATCC 10558 that previously demonstrated strong coaggregation reactions and served as a positive control.15 After mixing, the suspensions were scored immediately for coaggregation. The suspensions were then kept at room temperature for 1 h, mixed for 10 s on the vortex mixer, and scored again.15 The scoring system of Cisar et al.6 was used to evaluate the degree of coaggregation in the mixed suspensions by naked eye visualization. In brief, this system has a scoring range from ‘‘’’ to ‘‘4+’’ as follows: ‘‘’’ shows no change in turbidity and no evidence of coaggregates in the mixed suspensions; ‘‘1+’’ represents turbid supernatant with finely dispersed coaggregates; ‘‘2+’’ demonstrates definite coaggregates which do not precipitate immediately; ‘‘3+’’ manifests slightly turbid supernatant with formation of large precipitating coaggregates; and ‘‘4+’’ exhibits clear supernatant and large coaggregates which precipitate immediately. All assays were conducted in duplicate on two different occasions.
26
S. Shen et al.
Spectrophotometric, quantitative assay Measurement of the autoaggregation percentage of each organism was evaluated as follows: 1 mL bacterial inocula was transferred into a sterile cuvette (Elkay Ultra-VU1 Disposable cuvettes, Micro Square, Elkay Products Inc., Shrewsburg, MA, USA) and the optical density (OD) was recorded over a period of 60 min using the spectrophotometer at a wavelength of 660 nm (520 nm for Candida albicans). The autoaggregation degree of each of the 22 organisms was calculated as the percent decrease of OD after 60 min using Eq. (1): % Autoaggregation ¼
OD0 OD60 100 OD0
trifuged at 2000 rpm for 2 min and the supernatant of the mixture (about 0.6 mL) was gently transferred to a new sterile cuvette. The OD of the supernatant was then recorded using the spectrophotometer at a wavelength of 660 nm. The quantitative coaggregation rate of each pair of the coaggregation partners were calculated using Eq. (2):5 % Coaggregation ¼
ðODA þ ODB Þ 2 ODðAþBÞ ODA þ ODB 100 (2)
(1)
where OD0 is the optical density of the microorganism measured immediately after the inocula was transferred into the cuvette; and OD60 is the optical density of the microorganism measured 60 min later after the inocula was transferred. Evaluation of the degree of coaggregation of each pair of the organisms was performed as–—equal volumes (0.5 mL) of each of two paired microbial inocula were mixed in a sterile cuvette (Elkay), vortexed for 10 s and kept at room temperature for a period of 60 min. The mixture was then cen-
where ODA and ODB stand for the optical densities of bacterium A and bacterium B, respectively; and OD(A+B) stands for the optical density of the mixture of inocula of bacterium A and bacterium B.
Observation of the bacterial pairs with strong coaggregation using SEM The ultrastructural features of four bacterial pairs showing strong coaggregation interactions were observed by scanning electron microscopy. These were F. nucleatum ATCC 10953 with S. sanguinis ATCC 10558, which served as a positive control; A.
Figure 1 The correlation of data derived from coaggregating profiles of 231 pairs of organisms using the semiquantitative, visual assay and quantitative, spectrophotometric assay (r = 0.955, P < 0.001).
Coaggregation profiles of the microflora from root surface caries lesions
israelii NT6-2A with P. prevotii ND1-6A; A. israelii NT6-2A with V. parvula ND8-6A; and A. odontolyticus ND2-2A with V. parvula ND8-6A–—all of which showed ‘‘4+’’ degree coaggregations. For this purpose the unmixed bacterial inocula prepared as described above were mixed, allowed to coaggregate, and one drop of the mixed suspension was transferred onto a cover glass and air-dried at room temperature. The specimen was then fixed in 1% osmium tetraoxide (OsO4) vapor, dehydrated thoroughly in a freeze drying system (Labconco Freeze Dry System/Freezone1 4.5, Kansas City, Missouri 64132, USA), sputter coated with palladium—gold to a thickness of around 20 nm (Jeol Fine Coat, Ion Sputter JFC-1100, Jeol, Tokyo, Japan) and observed using a scanning electron microscope (Philips XL 30CP, Philips Electronics N.V., Eindhoven, the Netherlands).
Statistical analysis The correlation coefficient between the results from visual, semi-quantitative assay (graded from 0 to 4) and spectrophotometric, quantitative assay (ranged from 0% to 78.4%) was calculated using SPSS statistical program (SPSS version 8.0, SPSS Inc., USA).
Results Comparison of the visual and optical assays The optical assay showed a much higher sensitivity than the visual assay. However, despite the varying sensitivities of the two assays used, there was highly significant concordance in the results between the semi-quantitative, visual method and the quantitative, spectrophotometric assay (r = 0.955, P < 0.001) (Fig. 1).
Inter-species coaggregation–—visual, semi-quantitative assay Coaggregation of the 21 species of bacteria was evaluated using visual, semi-quantitative assay. Of the 210 bacterial pairs studied, 29 pairs (13.8%) had varying degrees of positive coaggregation. These data are shown in the upper right half of Table 2 P. intermedia ND8-9A and F. nucleatum NT6-6A coaggregated with 9 and 6 other species, respectively. The ‘‘4+’’ degree of coaggregation reactions occurred in the mixtures of A. israelii NT6-2A with V. parvula ND8-6A, A. odontolyticus ND2-2A with V. parvula ND8-6A and A. israelii NT62A with P. prevotii ND1-6A. The ‘‘3+’’, ‘‘2+’’ and
27
‘‘1+’’ degrees of coaggregation were recorded in three, seven and sixteen bacterial pairs, respectively (Table 2).
Inter-species coaggregation–— spectrophotometric, quantitative assay Percentages of coaggregation interactions examined using the quantitative coaggregation assay are shown in lower left half of Table 2. The percent coaggregation interactions ranged from 0% to 78.4%. Overall, 48 of 210 (22.9%) bacterial pairs showed coaggregation percentiles higher than 5.0%. The strongest degree of coaggregation reactions were observed in mixtures of A. israelii NT62A with V. parvula ND8-6A (78.4% coaggregation), A. odontolyticus ND2-2A with V. parvula ND8-6A (61.1%), A. israelii NT6-2A with P. prevotii ND16A (66.7%), C. gracilis ND9-8A with A. odontolyticus ND2-2A (49.7%), P. intermedia ND8-9A with L. minutis ND2-1A (47.1%) and C. gracilis ND9-8A with A. israelii NT6-2A (46.6%).
Degree of autoaggregation of bacterial isolates The autoaggregation reactions of the bacteria in percent are also shown in the diagonal column of Table 2. F. nucleatum NT6-6A (55.0%), L. acidophilus ND7-2A (35.1%), S. bovis II/2 ND2-2 (27.8%), G. morbillorum ND2-8 (21.6%), P. magnus NT1-2A (12.5%), P. intermedia ND8-9A (10.1%) and S. mutans ND13-1A (10.0%) all demonstrated relatively higher autoaggregation compared with their counterparts during the 60-min observation period.
Coaggregation of Candida albicans and bacteria The single isolate of C. albicans ND2-1S that were used in the study, did not coaggregate to a significant extent with any of the bacteria. However, S. acidominimus ND2-3 (8.7%), S. equinus ND1-2 (6.2%) and S. constellatus ND10-13A (6.0%) showed some degree of avidity to candidal blastospores as they demonstrated more than 5.0% coaggregation in the spectrophotometric assay, but not in the visual assay (Table 2). Candida blastospores did not demonstrate significant autoaggregation (7.6%) (Table 2).
SEM findings When the coaggregation patterns of bacterial pairs having strong coaggregation reactions were examined using scanning electron microscopy, several
28
S. Shen et al.
Table 2 A catacomb illustrating the results of quantitative (Y-axis) and semi-quantitative (X-axis) assessment of coaggregation reactions of 22 microorganisms isolated from root caries lesions.
(a) The emboldened and italicized figures in the diagonal grille indicate the percentage of autoaggregation interactions in the quantitative, spectrophotometric assay; (b) Figures with the ‘+’ marks in the upper right half of the catacomb indicate the coaggregation degree measured by the semi-quantitative, visual assay; and (c) Emboldened figures in the lower left half of the table indicate the percentages of coaggregation interactions greater than 5.0% as measured by the quantitative, spectrophotometric assay.
patterns, ranging from simple cell-to-cell adhesion to the formation of a complex coaggregation network, were found (Fig. 2). One of the most common observations was a single rod coaggregating with multiple cocci (Fig. 2b and c). However, no typical ‘‘corn— cob’’ structure was observed between ‘‘rod and coccal form’’ coaggregation partners. A detailed observation of the adhesion patterns (Fig. 2h) demonstrated tight adherence between some coaggregating bacteria that led to the virtual loss of discernible cell walls of the bacterial partners.
Discussion All 22 microorganisms used in the present study were isolated from 30 different root surface caries lesions in 18 different elderly, ethnic Chinese and were randomly chosen from the pool of identified wild-type isolates belonging to 11 different genera.11 Amongst 210 bacterial pairs evaluated here, 29 (13.8%) were found to demonstrate positive coaggregation reactions. This rate is slightly higher than the findings of Gibbons and Nygaard7 who
Coaggregation profiles of the microflora from root surface caries lesions
29
Figure 2 Several patterns of coaggregation reactions between bacterial pairs having strong coaggregation (scored ‘4+’) reactions are shown in this composite figure. (a) The simplest pattern of coaggregation wherein a single F. nucleatum ATCC 10953 adheres to a single S. sanguinis ATCC 10558 (6000); (b) A single F. nucleatum ATCC 10953 coaggregating with multiple, paired and chained S. sanguinis ATCC 10558 cells (6000); (c) An A. israelii NT6-2A coaggregating with multiple, unchained P. prevotii ND1-6A cells (12,000). Multiple reacting sites appear to exist on the surface of A. israelii; (d) Formation of a complex network with a large sheaf of aggregated F. nucleatum ATCC 10953 and multiple S. sanguinis ATCC 10558 (6,000); (e) Formation of a complex coaggregation network of A. israelii NT6-2A and P. prevotii ND1-6A (6,000); (f) Formation of a complex coaggregation network A. odontolyticus ND2-2A and V. parvula ND8-6A (6,000); (g) Formation of a complex coaggregation network A. israelii NT6-2A and V. parvula ND8-6A (6,000); (h) A higher magnification of a coaggregating site of F. nucleatum ATCC 10953 and S. sanguinis ATCC 10558 (60,000). The cell walls of both coaggregation partners appear to merge so that the confines of the two cells are not clearly visible.
30
tested 23 bacterial strains including several prominent plaque species and observed 23 coaggregating pairs in a total of 253 (9.1%) bacterial pairs. It has been reported that the bacteria isolated from one single site show higher degree of positive coaggregation reactions than those from different sites.18 Perhaps the low frequency of positive coaggregating pairs seen in the current study is likely to be due to their varying origin, as the organisms were mainly from different root lesions in many individuals. Kolenbrander et al.19 evaluated 122 human oral bacteria comprising 11 genera and observed intrageneric coaggregation reactions in S. oralis, S. sanguis, S. mitis and S. gordonii species and no such reactions amongst species belonging to ‘‘secondary" plaque colonizers such as Fusobacterium spp., Prevotella spp. and Lactobacillus spp. The unique intrageneric coaggregation property possessed by the pioneer colonizers, i.e. Streptococcus spp. and Actinomyces spp., appear to be a major advantage for these bacteria to initiate colonization at clean tooth surfaces. In the present study, a relatively higher degree of intra-generic coaggregation was recorded in S. bovis II/2 ND2-2 (27.8%) and S. mutans ND13-1A (10.0%) and not the other six species of streptococci. It appears therefore that streptococci isolated from well-established root caries lesions have a lesser propensity for intra-generic coaggregation hence their establishment on root surface carious niches may be due to mechanisms unrelated to intra-generic coaggregation. On the contrary, several other wildtype species, such as F. nucleatum NT-6-6A (55.0%), L. acidophilus ND7-2A (35.1%), S. bovis II/2 ND2-2 (27.8%) and G. morbillorum ND2-8 (21.6%) also showed strong autoaggregation interactions implying their ability for autonomous colonization and accretion on root caries lesions. With respect to interspecies interactions the ‘‘4+’’ or the greatest degree of coaggregation was recorded with the following combinations–—A. israelii NT6-2A with V. parvula ND8-6A (78.4% coaggregation), A. odontolyticus ND2-2A with V. parvula ND86A (61.1%) and A. israelii NT6-2A with P. prevotii ND1-6A (66.7%). The results are in agreement with previous studies which showed that Actinomyces spp. plays major role in coaggregation interactions.6,7,18,20 Interestingly though the strong coaggregation bond between A. odontolyticus or A. israelii with strains of either V. parvula or P. prevotii has not been reported hitherto fore. Previous detailed studies on the coaggregation interactions between Streptococcus spp. and Actinomyces spp. have revealed the complementary adhesion-receptor mechanisms amongst the latter organisms.6,18,21 It is also known that fimbriae are important in adherence of some strains A. naeslundii (including former
S. Shen et al. A. viscosus strains) to other cells or surfaces.22 However, little data are available on the mechanisms mediating coaggregation between A. odontolyticus and V. parvula or between A. israelii and P. prevotii. As Actinomyces spp. are predominantly isolated from root caries lesions and are considered key pathogens in this process,8,10,11,13,23,24 our results point to possible existence of ligand—receptor interactions between A. odontolyticus and V. parvula and between A. israelii and P. prevotii. Further biochemical studies on the latter interactions are warranted to clarify these mechanisms. P. intermedia ND8-9A, F. nucleatum NT6-6A and C. gracilis ND9-8A were the ‘‘top three’’ bacteria that coaggregated with multiple bacterial species in the present study. It is known that cell surface appendages on Bacteroides spp. serve as coaggregation bridges mediating their adhesion to other bacteria in dental plaque.17,25 Due to taxonomic reclassification of Bacteroides spp. most of the species belonging to this genus are now designated in the genera Prevotella and Campylobacter. 26 We evaluated the coaggregation between either Campylobacter species (C. gracilis ND9-8A, previous Bacteroides gracilis) or Prevotella species (P. intermedia ND8-9A, previous Bacteroides intermedia) with 20 species of other bacteria. A large proportion of positive coaggregating pairs between either P. intermedia ND8-9A or C. gracilis ND9-8A with strains of Streptococcus spp., Gemella spp., Peptostreptococcus spp., Lactobacillus spp. and Actinomyces spp. indicate that the Gram-negative obligate anaerobic rods may play an important role in the interactions leading to root caries formation. Furthermore our findings of the coaggregation between F. nucleatum NT6-6A and six other bacterial species, i.e. S. bovis II/2 ND2-2, S. constellatus ND10-13A, S. sanguinis (previous S. sanguis) II ND73, L. acidophilus ND7-2A, C. sputigena ND2-12A and P. intermedia ND8-9A, reconfirm the reports that these obligate, anaerobic, fusiform organisms have the ability to coaggregate with a large representation of oral bacteria acting as a key organism in the dental plaque formation possibly during latter stages of plaque maturation and modulation of the climax community.2,27 Candida spp. have been reported to be associated with root caries.28—30 Previous workers have demonstrated that C. albicans may coaggregate with a range of oral bacteria including S. sanguis, S. salivarius, S. mutans, S. mitis,31 S. gordonii, S. oralis, S. anginosus, S. pyogenes,32 L. amylovorus,33 F. nucleatum31,33 and A. viscosus.31 However, in the present study, we failed to record significant coaggregation reactions between the single C. albicans isolate and other 21 bacteria, except for the
Coaggregation profiles of the microflora from root surface caries lesions
weak coaggregation reactions with S. acidominimus ND2-3 (8.7%), S. equinus ND1-2 (6.2%) and S. constellatus ND10-13A (6.0%). It seems that the coaggregation mechanisms play an insignificant role in C. albicans colonization associated with root caries. Indeed this may be one reason why Candida species comprise a very low percent of flora in root caries.11,13,29 The SEM observation of interactions between bacterial pairs that showed strong coaggregation revealed interesting adhesive characteristic between the coaggregating partners. F. nucleatum ATCC 10953, A. israelii NT6-2A and A. odontolyticus ND2-2A demonstrated multiple receptor sites on each cell surface that appeared to react with several cocci. The rods in particular served as a framework that allowed the coccal cells belonging to S. sanguinis ATCC 10558, V. parvula ND8-6A and P. prevotii ND1-6A to adhere. The formation of cross-links between F. nucleatum ATCC 10953 and S. sanguinis ATCC 10558, A. israelii NT6-2A and V. parvula ND8-6A, A. odontolyticus ND2-2A and V. parvula ND8-6A as well as A. israelii NT6-2A and P. prevotii ND1-6A cells, appears to be the foundation of a fairly intricate structure that sustains the complex architectural features of the biomass of the root lesions.34 Detailed observation of the coaggregation foci between partners demonstrated very tight interactions, wherein the cell walls merged resulting in seamless confluence of the organisms (Fig. 2h). However, we were unable to detect ‘‘corn—cob’’ formation in the present study. It has been reported that this special structure was usually found at periphery of developing plaque35,36 and replication of this structure could be accomplished in vitro by mixing rods and cocci at a ratio of 1:10.37,38 To conclude, further work with varying groups of wild-type bacteria isolated from root caries lesions is required to elucidate the mechanisms underlying intra- and inter-generic coaggregation reactions that may be unique to this ecological niche, which has been sparsely studied. However, the current data which appear to be the first detailed study of coaggregation reactions of root caries bacteria should serve as basic information for future workers.
Acknowledgement The authors would like to thank Miss Joyce Y.Y. Yau and Mr. Simon Lee for the assistance during the experiment. This study was supported by a grant (a/c 10201963) from the Committee on Research
31
and Conference Grants (CRCG), the University of Hong Kong, Hong Kong SAR, China.
References 1. Marsh P, Martin M. Oral Microbiology. 4th edn. Oxford: Wright; 1999. 2. Kolenbrander PE. Oral microbial communities: biofilms, interactions, and genetic systems. Ann Rev Microbiol 2000;54:413—37. 3. Kolenbrander PE. Intergeneric coaggregation among human oral bacteria and ecology of dental plaque. Ann Rev Microbiol 1988;42:627—56. 4. Kinder S, Holt SC. Coaggregation between bacterial species. Methods Enzymol 1994;236:254—70. 5. McIntire FC, Vatter AE, Baros J, Arnold J. Mechanism of coaggregation between Actinomyces viscosus T14V and Streptococcus sanguis 34. Infect Immun 1978;21(3):978—88. 6. Cisar JO, Kolenbrander PE, McIntyre FC. Specificity of coaggregation reactions between human oral streptococci and strains of Actinomyces viscosus or Actinomyces naeslundii. Infect Immun 1979;24(3):742—52. 7. Gibbons RJ, Nygaard M. Interbacterial aggregation of plaque bacteria. Arch Oral Biol 1970;15(12):1397—400. 8. van Houte J, Lopman J, Kent R. The predominant cultivable flora of sound and carious human root surfaces. J Dent Res 1994;73(11):1727—34. 9. Schupbach P, Osterwalder V, Guggenheim B. Human root caries: microbiota in plaque covering sound, carious and arrested carious root surfaces. Caries Res 1995;29(5):382— 95. 10. Schupbach P, Osterwalder V, Guggenheim B. Human root caries: microbiota of a limited number of root caries lesions. Caries Res 1996;30(1):52—64. 11. Shen S, Samaranayake LP, Yip H-K, Dyson JE. Bacterial and yeast flora of root surface caries in elderly, ethnic Chinese. Oral Dis 2002;8(4):207—17. 12. Ellen RP, Banting DW, Fillery ED. Longitudinal microbiological investigation of hospitalized population of older adults with a high root surface caries risk. J Dent Res 1985;64(12):1377—81. 13. Brown LR, Billings RJ, Kaster AG. Quantitative comparisons of potentially cariogenic microorganisms cultured from noncarious and carious root and coronal tooth. Infect Immun 1986;51(3):765—70. 14. Keltjens HMA, Schaeken MJM, van der Hoeven JS, Hendriks JC. Microflora of plaque from sound and carious root surfaces. Caries Res 1987;21(3):193—9. 15. Sedgley CM. Enterobacteriaceae in the human oral cavity. (Ph. D. Thesis). The University of Hong Kong: Hong Kong. 1995; 214—225. 16. Maryanski JH, Wittenberger CL. Mannitol transport in Streptococcus mutans. J Bacteriol 1975;124(3):1475—81. 17. Kolenbrander PE, Andersen RN, Holdeman LV. Coaggregation of oral Bacteroides species with other bacteria: central role in coaggregation bridges and competitions. Infect Immun 1985;48(3):741—6. 18. Kolenbrander PE, Inouye Y, Holdeman LV. New Actinomyces and Streptococcus coaggregation groups among human oral isolates from the same site. Infect Immun 1983;41(2):501—6. 19. Kolenbrander PE, Andersen RN, Moore LV. Intra-generic coaggregation among strains of human oral bacteria: potential role in primary colonization of the tooth surface. Appl Environ Microbiol 1990;56(12):3890—4.
32
20. Komiyama K, Gibbons RJ. Interbacterial adherence between Actinomyces viscosus and strains of Streptococcus pyogenes, Streptococcus agalactiae, and Pseudomonas aeruginosa. Infect Immun 1984;44(1):86—90. 21. Kolenbrander PE, Williams BL. Prevalence of viridans streptococci exhibiting lactose-inhibitable coaggregation with oral actinomycetes. Infect Immun 1983;41(2):449—52. 22. Gibbons RJ, Hay DI, Cisar JO, Clark WB. Adsorbed salivary proline-rich protein-1 and statherin: receptors for type 1 fimbriae of Actinomyces viscosus T14V-J1 on apatitic surfaces. Infect Immun 1988;56(11):2990—3. 23. Jordan HV, Hammond BF. Filamentous bacteria isolated from human root surface caries. Arch Oral Biol 1972;17 (9):1333—42. 24. Nyvad B, Kilian M. Microbiology of the early colonization of human enamel and root surfaces in vivo. Scand J Dent Res 1987;95(5):369—80. 25. Stinson MW, Safulko K, Levine MJ. Adherence of Porphyromonas (Bacteroides) gingivalis to Streptococcus sanguis in vitro. Infect Immun 1991;59(1):102—8. 26. Shah HN, Collins MD. Prevotella, a new genus to include Bacteroides melaninogenicus and related species formerly classified in the genus Bacteroides. Int J Syst Bacteriol 1990;40(2):205—8. 27. Kolenbrander PE. Coaggregation of human oral bacteria: potential role in the accretion of dental plaque. J Appl Bacteriol 1993;74(Suppl.):79S—86S. 28. Scheinin A, Pienihakkinen K, Tiekso J, Holmberg S. Multifactorial modeling for root caries prediction. Community Dent Oral Epidemiol 1992;20(1):35—7.
S. Shen et al.
29. Beighton D, Lynch E. Relationships between yeasts and primary root-caries lesions. Gerodontology 1993;10(2):105—8. 30. Narhi TO, Vehkalahti MM, Siukosaari P, Ainamo A. Salivary findings, daily medication and root caries in the old elderly. Caries Res 1998;32(1):5—9. 31. Bagg J, Silverwood RW. Coagglutination reactions between Candida albicans and oral bacteria. J Med Microbiol 1986;22(2):165—9. 32. Jenkinson HF, Lala HC, Shepherd MG. Coaggregation of Streptococcus sanguis and other streptococci with Candida albicans. Infect Immun 1990;58(5):1429—36. 33. Hsu LY, Minah GE, Peterson DE, Wingard JR, Merz WG, Altomonte V, et al. Coaggregation of oral Candida isolates with bacteria from bone marrow transplant recipients. J Clin Microbiol 1990;28(12):2621—6. 34. Kolenbrander PE, London J. Adhere today, here tomorrow: oral bacterial adherence. J Bacteriol 1993;175(11):3247— 52. 35. Jones SJ. A special relationship between spherical and filamentous microorganisms in mature human dental plaque. Arch Oral Biol 1972;17(3):613—6. 36. Listgarten MA. Structure of the microbial flora associated with periodontal health and disease in man. J Periodontol 1976;47(1):1—18. 37. Kolenbrander PE, Andersen RN. Multigeneric aggregations among oral bacteria: a network of independent cell-to-cell interactions. J Bacteriol 1986;168(2):851—9. 38. Kolenbrander PE, Andersen RN. Intergeneric rosettes: sequestered surface recognition among human periodontal bacteria. Appl Environ Microbiol 1988;54(4):1046—50.