| Sign In to gain access to subscriptions and/or personal tools. |
Novel Bacterial Phylotypes in Endodontic Infections
1 Department of Endodontics, Estácio de Sá University, Rio de Janeiro, Brazil; and Correspondence: * corresponding author’s address, R. Herotides de Oliveira 61/601, Icaraí, Niterói, RJ, Brazil 24230-230; siqueira{at}estacio.br
Although molecular studies have revealed potential oral pathogens among the phyla Spirochaetes and Deferribacteres, their occurrence in endodontic infections has not been consistently investigated. In this study, we devised a nested PCR-DGGE approach to survey samples from infected root canals for the presence of members of these two phyla, and to examine their diversity. The primers used also amplified DNA from Atopobium species. Eight of 10 cases showed bands representative of the target bacterial groups. DGGE profiles revealed a mean number of 6.5 intense and faint bands. No single band occurred in all profiles. Sequences from intense bands excised from the gel showed similarities to species/phylotypes of all target groups—Flexistipes species (Deferribacteres phylum), uncharacterized spirochetes, and Atopobium species. Analysis of these data indicates that uncultivated Spirochaetes and Deferribacteres phylotypes are frequent members of the endodontic microbiota and may be potential pathogens involved with the etiology of periradicular diseases.
Key Words: endodontic microbiology • 16S rDNA • molecular biology
Although bacterial species or phylotypes detected in the oral cavity fall within 11 different phyla (Paster et al., 2001, 2002), representatives from only 6 of those have been reported in endodontic infections to date, namely, Bacteroidetes, Spirochaetes, Firmicutes, Actinobacteria, Fusobacteria, and Proteobacteria (Siqueira, 2003). Even so, the possibility exists that the diversity within some of these phyla has not been thoroughly explored in endodontic infections. For instance, 8 of the 10 cultivable oral Treponema species have been detected in endodontic infections by molecular methods (Baumgartner et al., 2003; Rôças et al., 2003; Siqueira and Rôças, 2003 Siqueira and Rôças, 2004). However, the diversity of spirochetes in the oral cavity has been reported to be far greater than expected, and estimates suggest that about 80% of the oral treponemes remain uncultivable (Dewhirst et al., 2000; Paster et al., 2001). Consequently, it is possible that other spirochetes may occur in infected root canals. Moreover, it is possible that species belonging to the other phyla occurring in the oral cavity can also infect root canals. Members of the Deferribacteres phylum have been recently suggested to be involved with periodontal diseases (Paster et al., 2001; Kumar et al., 2003). These bacteria have not been found in endodontic infections, and this may be because they are actually absent, or they are difficult to isolate and/or identify by cultivation procedures. It has been postulated that a comprehensive description of microbial communities inhabiting different habitats requires the use of techniques that sidestep cultivation (Hugenholtz et al., 1998). Genetic fingerprinting techniques represent a powerful tool for the investigation of the structure of microbial communities in diverse ecosystems and can be used for microbial identification. A commonly used strategy for the fingerprinting of complex bacterial communities consists of 16S rDNA-based PCR, followed by product analysis by denaturing gradient gel electrophoresis (DGGE) (Muyzer, 1999). The DGGE approach is based on electrophoresis of PCR products in polyacrylamide gels containing a linearly increasing gradient of DNA denaturants (Muyzer, 1999). In DGGE, DNA fragments of the same length, but with different base-pair sequences, can be separated. A single-base change in a sequence can be resolved (Fischer and Lerman, 1983), which provides PCR-DGGE with a great potential to identify closely related species based on 16S rDNA sequence divergence. In addition to being applied to the broad-range analysis of complex microbial communities (Muyzer and Smalla, 1998), the method can also be used to detect specific bacterial groups by altering the primer targets or by using a nested PCR (nPCR) approach (Ercolini, 2004). We undertook the present study to investigate the occurrence of members of the phyla Spirochaetes and Deferribacteres in endodontic infections, and to examine their diversity through a culture-independent approach. Representatives of these phyla have been associated with periodontal diseases, and the possibility exists that these phyla also contain potential endodontic pathogens.
Study Population and Sample Collection The present study included ten consecutive adult patients (ages ranging from 19 to 65 yrs) who had been referred for root canal treatment to the Department of Endodontics, Estácio de Sá University. Samples were taken from the root canals of teeth having caries lesions, necrotic pulps, and radiographic evidence of periradicular diseases. Selected teeth showed no periodontal pockets deeper than 4 mm. The study protocol was institutionally approved, and informed consent was obtained from the patients. Protocols for sampling procedures and DNA extraction were as described previously (Siqueira and Rôças, 2003).
nPCR-DGGE Assay In the second round of amplification, a fragment located in the V6-to-V8 regions of the 16S rDNA (Tung et al., 2002) was amplified from the first PCR products with the following universal bacterial primers: 968f-GC (5'-CGCCCGCCGCGCGCGGCGGG CGGGGCGGGGGCACGGGGGGAACGCGAAGAACCTTAC-3') and 1401r (5'-CGGTGTGTACAAGACCC-3') (Nübel et al., 1996). DNA extracted from clinical samples was used as target in the first PCR reaction, which was performed in a 50-µL mixture containing 40 pmol of each primer, 5 µL of 10X PCR buffer (Biotools, Madrid, Spain), 1.5 mM MgCl2, 1.25 U Tth DNA polymerase (Biotools), and 0.2 mM of each deoxyribonucleoside triphosphate (Biotools). Afterward, 1 µL of the PCR products generated in the first round of amplification was used as a template for the nested reaction in a PCR mixture comprised of 25 pmol of universal primers, 5 µL of 10X PCR buffer, 3.8 mM MgCl2, 2.5 U of Tth DNA polymerase, 0.2 mM concentration of each deoxyribonucleoside triphosphate, and sterile filtered milliQ water, to a final volume of 50 µL. Negative controls consisting of sterile milliQ water instead of sample were included with each batch of samples analyzed. PCR amplification was performed in a DNA thermocycler (Primus 25/96, MWG-Biotech, Ebersberg, Germany). Cycling conditions were as follows: (first round) 30 cycles of denaturation at 94°C for 45 sec, annealing at 60°C for 45 sec, and extension at 72°C for 90 sec, and a final elongation step at 72°C for 15 min; and (second round) initial denaturation at 94°C for 2 min, followed by 28 cycles of denaturation at 94°C for 1 min, annealing at 55°C for 1 min, extension at 72°C for 2 min, and a final step of 72°C for 10 min. Prior to DGGE analysis, the presence of PCR products was confirmed by electrophoresis in an agarose gel stained with ethidium bromide and viewed under UV transillumination. DGGE was performed with use of the Dcode Universal Mutation Detection System (Bio-Rad Dcode, Richmond, VA, USA) at 75V and 60°C for 16 hrs in 0.5X TAE buffer [20 mM Tris-acetate (pH 7.4), 10 mM sodium acetate, 0.5 mM disodium EDTA]. The nPCR products (30 µL) were loaded onto 6% (w/v) polyacrylamide gels containing a linear gradient ranging from 20% to 70% denaturant [100% denaturant corresponded to 7 M urea and 40% (v/v) formamide]. Afterward, the gel was stained with SYBR green I nucleic acid gel stain (Molecular Probes, Leiden, The Netherlands) for 40 min, and then scanned in a Storm PhosphorImager (Amersham Biosciences, Little Chalfont, Buckinghamshire, UK).
Sequence Analysis
PCR amplicons generated in the first reaction were detected in 8 of 10 samples examined. The 2 samples that yielded negative results after the nPCR reaction gave positive results when subjected to a single amplification reaction with primers 968f-GC/1401r. These 2 samples were not analyzed by DGGE. These results served to indicate that bacterial DNA was available in all clinical samples, and that the PCR reactions were conducted without significant amounts of inhibitors in samples. Negative controls yielded no bands.
The DGGE profiles generated by nPCR amplification of endodontic samples revealed distinct banding patterns from different clinical samples (Fig. 1
Analysis of the DNA sequences of the excised bands 1 and 3 showed low-scoring homology (82% and 85%, respectively) to the sequence of an uncultured bacterium (SSE7) detected in anaerobic bioreactors. Bands 11, 13, and 14 had sequences with 98–100% similarity to band 1. Sequences of bands 2 and 4 showed 95% and 99% similarity to Flexistipes-like sp. oral clone BA121, respectively. This phylotype was first detected in subgingival samples from refractory periodontitis (Paster et al., 2001). The sequence of band 5 had a match value of 99% to both Synergistes sp. P1 P4G-18, also detected in subgingival samples, and Flexistipes oral clone BA121. Bands 6 and 8 showed sequences with 97% and 100% similarity to Flexistipes E1 E3-33, a clone detected in an infected root canal (Munson et al., 2002). The sequence of band 7 was related (85% similarity) to Atopobium rimae. The sequence of band 12 showed 85% similarity to the last 177 nucleotides of band 7. The last 201 nucleotides of sequences 9 and 10 differed by only 1 nucleotide, and showed sequence similarity of 87% to Deferribacteres oral clone D084. Band 15 was a chimera and was discarded from the study. Sequences from bands 1 to 8 were aligned with the sequences of close relatives and then used to construct a phylogenetic tree (Fig. 2  ).
Our initial purpose of this study was to analyze the spirochetal diversity in infected root canals using primer C90, which is reported to be selective for spirochetes, but also amplifies the 16S rDNA from Deferribacteres and Atopobium, particularly when spirochetes are in low numbers in, or absent from, samples (Paster et al., 2001). Surprisingly, our findings revealed a high dominance and frequency of Deferribacteres bacteria, particularly Flexistipes species, which had never been consistently found to occur in endodontic infections. Deferribacteres species/phylotypes have been only recently detected in the oral cavity (Paster et al., 2001; Hutter et al., 2003). Of the 8 Deferribacteres clones detected in a study by Paster et al.(2001), 7 formed a coherent cluster (oral cluster) distantly related to Synergistes jonesii, a cultivable, bovine ruminal bacterium. The other clone, named Flexistipes-like sp. oral clone BA121, was detected in one subject with refractory periodontitis, and fell within the S. jonesii cluster rather than the oral cluster. Three of the sequences obtained in the present study showed some level of similarity to clone BA121. Our results indicate that Flexistipes species may be common members of the endodontic microbiota, since at least one-half of the samples surveyed harbored these bacteria. Three sequences showed high similarity (> 99%) to Flexistipes oral clone BA121 or Flexistipes sp. E3-33. The other 2 sequences showed 95% and 97% similarities to these 2 oral clones, respectively, suggesting that they are different phylotypes of the Flexistipes group. The fact that Flexistipes species have never been consistently found in endodontic infections by culture suggests either that most phylotypes can be uncultivable, that they occur in numbers below the detection limits of culture approaches, or that they can present ambiguous phenotypic behaviors which can lead culture procedures to misidentification. The latter speculation is based on the fact that, in the only study that isolated a Flexistipes species from an infected root canal, identification was based on sequencing of the 16S rDNA and not by phenotype-based methods (Munson et al., 2002). Because some Flexistipes species are cultivable (Munson et al., 2002), efforts should be made to cultivate the most prevalent Flexistipes phylotypes found in endodontic infections, in an attempt to determine their pathogenicity and susceptibility to endodontic medicaments and procedures. The occurrence of spirochetes in endodontic infections has long been obscured by difficulties in culturing these micro-organisms. Recent studies using species-specific nPCR approaches revealed a high prevalence of oral treponemes in those infections (Baumgartner et al., 2003; Rôças et al., 2003; Siqueira and Rôças, 2003 Siqueira and Rôças, 2004). In the present study, 5 of the excised bands yielded sequences with similarities to members of the Spirochaetes phylum, but none of them clustered with Treponema species. This was rather unexpected, given the high prevalence of treponemes in endodontic infections, as previously reported by species-specific nPCR. Differences may have been due to the higher sensitivity of species-specific nPCR when compared with nPCR-DGGE (Rôças et al., 2003; Ercolini, 2004). Moreover, it is possible that the faint bands observed in DGGE profiles, which were not sequenced, might have been from treponemes, suggesting that they were in low numbers in the samples examined. It has been reported that all oral spirochetes belong to the genus Treponema (Dewhirst et al., 2000). However, our findings suggest that other spirochetes may be present as part of the oral microbiota. Nevertheless, a disadvantage of the DGGE assay is that the sequences generated are too short to disclose phylogenetic positions if no close match is present in the databases. Further studies should confirm or refute these findings. DGGE is a reliable molecular tool that permits the construction of a ‘picture’ of the structure of the microbiota (Muyzer, 1999). An additional advantage is that it allows for the identification of community members by the sequencing of excised bands (Muyzer and Smalla, 1998). Our findings pointed to the applicability of the PCR-DGGE method for identification of members of the endodontic microbiota, including previously uncharacterized bacteria. The fingerprint generated also allowed for the analysis of the diversity of these bacterial groups in endodontic samples, with a mean number of 6.5 intense and faint bands per profile. Intensity of bands in the DGGE gel is supposedly related to the density of corresponding bacterial phylotypes within the sample (Murray et al., 1996). In this study, dominant (intense) bands excised from the gel were sequenced, and comparison with database sequences revealed phylotypes of all target bacterial groups. Overall, the most intense bands belonged to Flexistipes phylotypes, which indicates their dominance among the bacteria targeted herein. Findings from the present study indicate that uncultivated species of the phyla Spirochaetes and Deferribacteres, as well as from the Atopobium genus, can be part of the endodontic microbiota, and therefore may participate in the etiology of periradicular diseases. Analysis of these data also expands the list of phyla represented in endodontic infections to include Deferribacteres. Ongoing studies from our laboratory, using specific probes for Deferribacteres phylotypes, are under way to investigate the prevalence of these bacteria in different types of endodontic infections.
This study was supported by grants from CNPq and PRONEX, Brazilian Governmental Institutions. Received for publication April 15, 2004. Revision received January 14, 2005. Accepted for publication February 24, 2005.
Journal of Dental Research, Vol. 84, No. 6,
565-569 (2005) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
