A new technique is presented for analyzing subgingival bacterial plaque. Different materials (polytetrafluoroethylene, gold, dentin) kept for several days in periodontal pockets of patients suffering from periodontitis were analyzed by electron microscopy and fluorescence in situ hybridization (FISH). Those parts of the carriers extending into the deepest zone of the pockets were predominantly colonized by spirochetes and Gram-negative bacteria whereas those segments in contact with a shallower region were colonized by streptococci. Independent of the material used, the bacterial colonization of the carriers appears to be similar. FISH using eubacteria- and species-specific oligonucleotides on semi-thin cross-sections of the carriers in combination with confocal laser scanning microscopy allowed detailed analysis of the architecture of biofilms and identification of putative periodontal pathogens with single cell resolution.
Though more than 400 different bacterial species have been isolated from periodontal pockets, only a limited number has been associated with periodontitis . These species are thought to participate in the formation of a biofilm on subgingival tooth surfaces . Such biofilm forming bacteria are obviously able to reveal a different pattern of gene expression than planktonic cells . It is known that certain bacterial species frequently appear in close relationship to each other in periodontal pockets corresponding to the pocket probing depths . So far, an analysis of the subgingival microbiota relied on sampling of bacteria either by paper points or by mechanical debridement. Both sampling procedures, however, disrupt the organization of biofilms. Biofilm formation has also been extensively studied using in vitro models, like flow chambers or chemostats. However, these studies might not necessarily reflect the situation in a periodontal pocket and clearly have limitations regarding fastidious and so far uncultured microorganisms. The only method to study subgingival plaque available so far required the extraction of teeth .
Since extraction of teeth is often not possible, the development of alternative methods for obtaining intact bacterial biofilms was desirable. While a special device for collecting supragingival plaque has recently been described , no procedure for collecting undisturbed subgingival plaque has yet been described. The aim of this study was to present a new method for sampling and monitoring the subgingival bacterial flora as a biofilm.
While electron microscopy by definition gives information on different bacterial morphotypes, the information on the spatial distribution has to be interpreted with caution. Fluorescence microscopy after in situ hybridization with oligonucleotide probes (FISH), however, not only allows for the identification of bacteria [7,8] but also provides valuable information on their spatial distribution [9,10].
From a previous study, it was known that extended polytetrafluoroethylene (e-PTFE) membranes used to cover superficial defects after surgery were colonized by plaque bacteria . Therefore a further aim of this study was to compare bacterial colonization on PTFE membranes with those on different materials to get an optimized carrier for plaque bacteria. Dentin was chosen as a control material comparable to tooth surface, while gold foil has been selected because of its suitability in scanning electron microscopy. The carriers were inserted into periodontal pockets from patients suffering from rapidly progressive periodontitis (RPP). Biofilm formation was monitored by electron microscopy and FISH after different incubation periods.
Materials and methods
A total of 12 patients scheduled for initial periodontal therapy participated in this study. Individuals who had been periodontally treated within the last 6 months or receiving any antimicrobial therapy during this period were excluded. RPP was diagnosed according to clinical and radiological criteria . Distal and mesial sites were selected because of their pronounced probing depth measurements (mean 8.07±1.63 mm).
Preparation of carriers and sampling of bacterial biofilm
Carriers consisting of gold foil or e-PTFE membranes and constructed as shown in Fig. 1 were carefully inserted to reach the bottom of the pocket. The carrier was fixed supragingivally to the tooth surface by using cyanoacrylic glue (Octyldent™ Closure Medical Corp., Raleigh, NC, USA). The construction was guided by the assumption that the carrier positioned in the pocket might be colonized from both the tooth and the soft tissue side. To stabilize soft flexible materials like e-PTFE or gold foil, the carriers were mounted on commercially available ‘plast-o-probe’ sticks (Maillefer, Ballaigues, Switzerland). Alternatively, 3-mm wide strips of e-PTFE or gold foil were attached to the plastic surfaces by four dots of cyanoacrylate (Tesa®, Beiersdorf, Hamburg, Germany). The respective material was wrapped around the tips of plast-o-probes exceeding them at one side. Additionally dentin slices, not exceeding 3 mm in width each, were mounted on strips of e-PTFE. To get a better adaptation to the curvature of the dental root, it became necessary to divide the solid dentin into several segments not exceeding 3 mm in length each. Carriers were then positioned in the pockets with the longer end of the foils facing the teeth (Fig. 1) and attached to tooth surfaces. After 3 or 6 days of exposure, carriers were removed from the periodontal pockets. Only those carriers that kept their stable position during the exposure period were used for further investigations. Altogether 52 carriers (31 e-PTFE, 12 dentin, nine gold) were analyzed.
For electron microscopy, specimens were fixed with either 2.5% (v/v) glutaraldehyde (Sigma, Munich, Germany), diluted in 0.1 M cacodylate buffer (pH 7) or a buffered formaldehyde–glutaraldehyde fixative at 4°C overnight . All the other treatments for transmission or scanning electron microscopy were done as previously described .
For FISH, four carriers were fixed with 3.7% (v/v) formaldehyde in phosphate-buffered saline (pH 7.4) for at least 3 h. Embedding in cold polymerizing resin Technovit 8100 (Kulzer, Wehrheim, Germany) and sectioning of the specimens were performed as described earlier for tissue biopsies .
Group-specific probe TRE I (5′-ACGCAAGCTCATCCTCAAG-3′) has been published earlier and has been shown to specifically detect group I of oral treponemes, most of which are as yet uncultured . EUB-338 (5′-GCTGCCTCCCGTAGGAGT-3′) specific for the domain Bacteria was used to visualize the entire bacterial population in the specimens . Probes were labeled with Cy3 or FITC, respectively, and obtained from Interactiva (Ulm, Germany).
Hybridization buffer containing 0.9 M NaCl, 20 mM Tris–HCl, pH 7.3, 0.01% (w/v) sodium dodecyl sulfate, 20% (v/v) formamide and 1 μM probe each was applied on the sections. To assess the specificity of the probes of fixed cells, the following Treponema strains served as controls: Treponema vincentii (ATCC 33580 and RITZ A), both Treponema species belonging to phylogenetic group I of oral treponemes, thus serving as positive controls. Treponema denticola (ATCC 33521), Treponema pectinovorum (ATCC 33768), Treponema socranskii subsp. buccale (ATCC 35534), Treponema socranskii subsp. socranskii (ATCC 35536), Treponema maltophilum (ATCC 51939), Treponema lecithinolyticum (ATCC 700332) and Treponema phagedenis subsp. reiteri (kindly provided by B. Wilske, Munich, Germany) serving as negative controls. Control slides with fixed treponeme cultures were included in every hybridization experiment (data not shown). After 3.5 h of incubation in a dark humid chamber at 46°C, sections were rinsed with distilled water, air-dried in the dark and mounted with Citifluor AF 1 (The Chemical Laboratory of the University of Kent, UK).
A confocal laser scanning microscope (CLSM) model LSM 510 (Carl Zeiss, Oberkochen, Germany) equipped with an Ar-ion laser (488 nm) and two HeNe lasers (543 and 633 nm) was used to record optical sections. Image processing was performed with a standard software package delivered with the instrument (Zeiss LSM version 1.6).
The ultrastructure of subgingival plaque
The formation of subgingival plaque was analyzed on different carriers that were kept in periodontal pockets of RPP patients for 3–6 days. After short incubation (3 days), islets-like colonization could be observed by scanning electron microscopy (Fig. 2). The loose colonization could also be seen in cross cut ultrathin sections (Fig. 3). Independent of the three materials used (e-PTFE, dentin, gold), a homogeneous electron dense pellicle could be demonstrated on the different carriers; also the bacterial colonization was similar. Analyzing more than 4000 electron micrographs, we can state that different Gram-negative rods and treponemes were found on those parts of the carriers corresponding to the deepest part of the pockets (Figs. 3 and 5). Gram-positive cocci could preferentially be observed on the upper segments of the carriers (Fig. 4). Early colonizing bacteria were characterized by an extracellular matrix especially shown in ultrathin sections parallel to the carrier surfaces (Fig. 5). Treponemes were found in close contact to this matrix (Fig. 5). When carriers were kept in the periodontal pockets for 6 days, a massive and dense colonization of bacteria could be shown (Fig. 6). After this time, a confluent biofilm resulted. Besides Gram-negative bacteria, some segments revealed a massive colonization of treponemes (Fig. 6). In such areas, the treponemes seemed to be arranged in a regular pattern.
The use of carriers inserted in periodontal pockets offered the opportunity to follow the development of subgingival bacterial flora as biofilm and to consider the spatial arrangement of different bacterial morphotypes. Additionally the specific coaggregation of different bacterial species, which is visible in Figs. 5 and 6, could be analyzed. While electron microscopical methods demonstrated the organization of subgingival biofilm build up with different morphotypes such as cocci, rods or spirochetes, other methods were needed to carry out a detailed identification of bacteria.
Identification of bacteria by FISH
Since the direct CLSM analysis of biofilm carriers such as gold, dentin or e-PTFE was hampered by light reflection and autofluorescence, semi-thin sections of Technovit-embedded material were used. The EUB-FITC probe allowed exact orientation within the sample and revealed different bacterial morphotypes in cross-sectioned subgingival biofilm of a RPP patient (Fig. 7). The thickness of the biofilm could be measured with 40–45 μm.
At higher magnification, different morphotypes of e.g. cocci or rods – some organized as microcolonies – could clearly be differentiated (Fig. 8). Moreover, simultaneous hybridization with probe TRE I-Cy3 revealed some large spirochetes interspersed between other bacteria. The higher magnification (Fig. 8) of this biofilm illustrates the opportunity to obtain information about the length of the treponemes or their undulation. To our knowledge, this is the first time that the numbers and the spatial arrangement of spirochetes within the periodontal biofilm could be demonstrated. In this part of the biofilm, treponemes were rather separately localized, whereas in other areas high numbers of group I treponemes were spread in the biofilm as also shown by electron microscopy (Fig. 6).
Here we present a method that allows the removal of subgingival plaque as widely undisturbed biofilm. The advantage of this method is that neither insertion nor removal of carriers requires teeth or crown removal. The precise fixed localization of the carriers allows the association of distinct bacterial morphotypes or species with the depth of periodontal pockets. In addition, information about the time-dependent formation of biofilm can be monitored.
Electron microscopical analysis of different carriers revealed that the nature of the carrier material does not seem to have a great influence on the colonization. This is in agreement with a previous observation emphasizing that as compared to supragingival conditions, the surface characteristics were less important for subgingival plaque , showing almost no differences in subgingival plaque composition on hydrophilic or hydrophobic polymers . A possible explanation for this similar colonization pattern may be the salivary pellicle on the surface of each carrier rendering the colonization conditions alike [17–19]. Formation of dental plaque on epoxy resin crowns has been described earlier [20,21]. However, this epoxy resin crown model is only of limited use for studying periodontitis, as any information on deep periodontal pockets is missing. High numbers of spirochetes and the presence of Gram-negative rods and coccoid cells were observed only on those parts of the carriers localized in the deepest zone of the periodontal pockets. In another ultrastructural study, Gram-positive cocci were shown as part of the apical border plaque . In our study, we found Gram-positive cocci only on the most coronal segments of the plaque carriers obviously exposed to a higher oxygen tension.
Modern determination of bacterial flora via cluster analysis examining the relationship among different taxa revealed that some species do not exist separately in the periodontal pockets but rather form complexes . Electron microscopy allowed to study the complexity of human plaque biofilm showing structural or spatial arrangements of different bacterial morphotypes (Fig. 6), described as ‘corncob’ or ‘bristle-brush’ formations [5,22]. The coaggregation and coadhesion of oral bacteria was summarized recently , however, there is doubt whether this in vitro study reflects the situation in vivo. In contrast, the ultrastructural biofilm analysis enables the characterization of different bacterial morphotypes and their spatial arrangements. Further taxonomic characterizations are only possible with additional methods such as immuno-electron microscopy and/or in situ hybridization using specific oligonucleotide probes. The advantage of using 16S rRNA-directed oligonucleotide probes for the identification and visualization of spatial distribution of different bacterial species was shown convincingly for environmental biofilms [7,8]. While the direct analysis of carriers with CLSM was hindered by reflection and autofluorescence of the carriers, the use of Technovit sections overcame these problems and allowed both the localization and identification of microorganisms.
Detailed information on biofilm structure in habitats such as subgingival plaque is still lacking. FISH using different probes simultaneously may reveal the exact composition and architecture of particular biofilm. Thus, this novel technique presented here may contribute to an understanding of biofilm formation and perhaps the role of single bacterial species in the etiopathogenesis of periodontal infections.