JOURNAL OF CLINICAL MICROBIOLOGY, July 2005, p. 33143319 Vol. 43, No. 70095-1137/05/$08.000 doi:10.1128/JCM.43.7.33143319.2005Copyright 2005, American Society for Microbiology. All Rights Reserved.

Uncultivated Phylotypes and Newly Named Species Associated withPrimary and Persistent Endodontic Infections

J. F. Siqueira, Jr.* and I. N. RocasDepartment of Endodontics, Estacio de Sa University, Rio de Janeiro RJ, Brazil

Received 21 January 2005/Returned for modification 12 March 2005/Accepted 16 March 2005

Endodontic infections have been traditionally studied by culture methods, but recent reports showing thatover 50% of the oral microbiota is still uncultivable (B. J. Paster et al., J. Bacteriol. 183:37703783, 2001) raisethe possibility that many endodontic pathogens remain unknown. This study intended to investigate theprevalence of several uncultivated oral phylotypes, as well as newly named species in primary or persistentendodontic infections associated with chronic periradicular diseases. Samples were taken from the root canalsof 21 untreated teeth and 22 root-filled teeth, all of them with radiographic evidence of periradicular bonedestruction. Genomic DNA was isolated directly from each sample, and 16S rRNA gene-based nested orheminested PCR assays were used to determine the presence of 13 species or phylotypes of bacteria. Species-specific primers had already been validated in the literature or were developed by aligning closely related 16SrRNA gene sequences. Species specificity for each primer pair was confirmed by running PCRs against a panelof several oral bacteria and by sequencing DNA from representative positive samples. All species or phylotypeswere detected in at least one case of primary infections. The most prevalent species or phylotypes found inprimary infections were Dialister invisus (81%), Synergistes oral clone BA121 (33%), and Olsenella uli (33%). Ofthe target bacteria, only these three species were detected in persistent infections. Detection of uncultivatedphylotypes and newly named species in infected root canals suggests that there are previously unrecognizedbacteria that may play a role in the pathogenesis of periradicular diseases.

Periradicular diseases are arguably among the most commoninflammatory diseases that affect humans (10). Overwhelmingevidence indicates that microorganisms infecting the root canalof the teeth are the major causative agents of these diseases(27). Data from culture and molecular studies have demon-strated that the microbiota associated with primary endodonticinfections is conspicuously dominated by anaerobic bacteriaand that an infected root canal can harbor from 10 to 30bacterial species (22, 33, 34, 37). On the other hand, the mi-crobiota of persistent endodontic infections associated withtreatment failure has been shown to be composed of fewerspecies, with dominance of facultative bacteria (32, 35). Eventhough over 300 bacterial species have already been isolatedfrom or detected in infected root canals, no single species hasbeen consistently found to be the major endodontic pathogen.

Application of molecular genetic methods to the analysis ofthe bacterial diversity in the oral cavity has revealed a stillbroader spectrum of extant bacteria than previously reportedby cultivation approaches (23). Overall, over 700 different spe-cies belonging to 11 divisions (or phyla) of the domain Bacteriahave been detected in the oral cavity of humans (17, 23, 24).About 50% of these bacteria are known only by 16S rRNAgene sequences (phylotypes) (23). This raises the interestingpossibility that uncultivated and as-yet-uncharacterized speciesthat have remained undetected in studies by traditional iden-tification methods may make up a large fraction of the livingoral microbiota and may participate in the etiology of oraldiseases, including periradicular diseases.

PCR amplification of conserved regions of the 16S rRNAgene, followed by cloning and sequencing of PCR products,has been widely applied to study of the bacterial diversity indifferent human healthy and diseased sites. However, the clon-ing approach is usually time consuming, labor intensive, andexpensive, being virtually impractical for analysis of multiplesamples in epidemiological investigations. Several studies havedevised PCR primers based on sequences from oral clonelibraries to survey larger numbers of samples from periodon-tally diseased subjects for the presence of uncultivated bacteriain an attempt to establish correlations with disease etiology(15, 19, 20, 26). Most of these studies revealed that severalphylotypes may play an important role in periodontal disease.To the best of our knowledge, no study has as yet used suchapproaches to investigate the prevalence of uncultivated oralphylotypes in endodontic infections.

The applicability of molecular genetic methods does not relyuniquely on the detection of uncultivable bacteria but also ona more reliable identification of diverse bacterial species, par-ticularly some nutrient-demanding and therefore difficult-to-grow oral anaerobic bacteria. Several fastidious bacterial spe-cies have only been recently reported to occur in endodonticinfections by molecular methods. These include Tannerella for-sythia (6, 34), several Treponema species (16, 25, 29), Prevotellatannerae (39), Filifactor alocis (28), Dialister pneumosintes (30),Haemophilus aphrophilus (34), Eubacterium infirmum (12), andCentipeda periodontii (31), all of them recognized periodontalpathogens. Members of the family Coriobacteriaceae, particu-larly from the genera Olsenella and Atopobium, have beencommonly detected in association with some oral diseases,including caries and marginal periodontitis (21, 23), but theyhave never been consistently found in endodontic infections,

* Corresponding author. Mailing address: Department of Endodon-tics, Estacio de Sa University, Av. Almte Ary Parreiras 311/1001 Icara,Niteroi, RJ Brazil 24230-322. Phone: 55 21 8874-1022. Fax: 55 212503-7289, ext. 223. E-mail: siqueira@estacio.br.

3314

and difficulties in identifying these species by phenotype-basedmethods can be one of the reasons for that.

It has been revealed that the main putative endodonticpathogens are also the main periodontal pathogens, with few(if any) exceptions (27, 37). Several new species or phylotypeshave been suggested to be involved with the pathogenesis ofperiodontal diseases (19, 23), but it is still uncertain whethermost of these phylotypes are present in infected root canalsassociated with periradicular diseases. Therefore, the presentinvestigation intended to use a devised nested or heminestedPCR assay to survey samples from primary or persistent end-odontic infections for the presence of some newly named bac-terial species and uncultivated phylotypes that have been re-cently detected in association with periodontal diseases. Mostof the target bacteria had never been previously found inendodontic infections. For the species already detected in end-odontic infections, prevalence had not been previously inferredbecause of the small number of samples examined (11, 22).

MATERIALS AND METHODS

Subjects. Root canal samples collected for previous investigations (25, 29, 32)and stored in TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8) at 20C wereavailable for reanalysis in this study. Samples were taken from patients who hadbeen seeking root canal treatment or retreatment at the Department of End-odontics, Estacio de Sa University, Rio de Janeiro RJ, Brazil. Only teeth fromadult patients (ages ranging from 18 to 80 years), all of them having radiographicevidences of periradicular disease, were included in this study. Overall, 43 rootcanal samples were obtained and grouped as follows, according to the clinicaldiagnoses: (i) 21 cases of primary endodontic infections (untreated teeth) asso-ciated with asymptomatic chronic periradicular lesions and (ii) 22 cases of per-sistent endodontic infections (root-filled teeth) associated with asymptomaticchronic periradicular lesions, which had been selected for retreatment.

All the teeth with primary endodontic infections showed carious lesions andnecrotic pulps. All the teeth with persistent infections had endodontic therapycompleted 2 years earlier. Root-filled teeth were coronally restored and nodirect exposure of the filling material to the oral cavity was evident. Termini ofthe root canal fillings ranged from 0 to 4 mm short of the radiographic apex. Allselected teeth showed no significant gingival recession and an absence of peri-odontal pockets of 4 mm deep.

Sampling procedures and DNA extraction. Root canal samples were takenfrom untreated or root-filled teeth under strict aseptic conditions as previouslydescribed (25, 29, 32). Endodontic files with the handle cut off and paper pointsused for sampling the canals were transferred to cryotubes containing 1 ml of 5%dimethyl sulfoxide in trypticase-soy broth (Difco, Detroit, MI) and immediatelyfrozen at20C. Further, samples were brought to room temperature, and DNAwas extracted using the protocol described previously (25, 29, 32).

The following strains were used in the present study: Actinobacillus actinomy-cetemcomitans ATCC 43718, Actinomyces radicidentis CCUG 42377, Atopobiumparvulum ATCC 33793, Atopobium rimae ATCC 49626, Enterococcus faecalisATCC 29212, F. alocis ATCC 35896, Fusobacterium nucleatum subsp. nucleatumATCC 25586, Fusobacterium nucleatum subsp. polymorphum ATCC 10953,Olsenella uli ATCC 49627, Olsenella profusa DSM 13989, Porphyromonas end-odontalis ATCC 35406, Porphyromonas gingivalis ATCC 33277, Prevotella inter-media ATCC 25611, Prevotella pallens ATCC 700821, Propionibacterium propi-onicum U13a-a, Pseudoramibacter alactolyticus C11b-d, Streptococcus intermediusATCC 27335, T. forsythia ATCC 43037, Treponema denticola B1 (Forsyth DentalInstitute), and Treponema lecithinolyticum ATCC 700332. All bacteria weregrown under appropriate culture conditions. Identification and purity of thecultures were confirmed by Gram staining, microscopic morphotyping, visualiza-tion of colony features on agar plates under magnification, growth on selectivemedium, and biochemical tests. Bacterial DNA was prepared as described pre-viously (25, 29, 32).

Design of specific PCR primers. 16S rRNA gene species- or phylotype-specificPCR primers were used. Primer sequences were as previously described orslightly modified therefrom (4, 19), except for those specific for Synergistes oralclone BA121 (formerly Flexistipes BA121), Synergistes oral clone E3_33 (formerlyFlexistipes E3_33), O. uli, O. profusa, and P. pallens, which were designed for thisstudy. Briefly, 16S rRNA gene sequences of each of these bacteria were retrieved

from the GenBank at the National Center for Biotechnology Information web-site and aligned with the sequences of their nearest neighbors in the phylogenetictree using the CLUSTAL W program (36) to identify variable areas betweenspecies. Potential primers were designed from these areas and BLAST (1) wasused to verify their specificity by comparing primer sequences with all availablesequences in the GenBank database. Based on BLAST searches, one or twoprimer sequences for each of the target species/phylotypes were selected andfurther tested for specificity against DNA from the bank of reference strains usedas controls and by sequencing of PCR products obtained from representativepositive clinical samples. The PCR species- or phylotype-specific primers areshown in Table 1.

Nested or heminested PCR identification. The whole-genome DNA extractsfrom clinical samples were used as templates in a 16S rRNA gene-based nestedor heminested PCR method devised to detect the target species/phylotypes inendodontic samples. In the first PCR, a practically full-length 16S rRNA genefragment was amplified using a pair of universal 16S rRNA gene primers, whichconsisted of the forward universal primer 27f (9) and the reverse universal primer1,492r (38). Aliquots of 5 l (each) of the supernatant from clinical samples wereused as targets in the first PCR. PCR amplification was performed with 25 l ofreaction mixture containing a 0.2 M concentration of forward and reverseuniversal primers, 2.5 l of 10 PCR buffer (Biotools, Madrid, Spain), 2 mMMgCl2, 1.25 U of Tth DNA polymerase (Biotools), and 25 M each deoxyribo-nucleoside triphosphate (Biotools).

Afterwards, 1 l of the universal reaction mixture was then used as a templatefor the nested or heminested specific reaction. The second PCR used to assessthe occurrence of the target species/phylotypes was performed with 50 l ofreaction mixture containing 1 M concentration of each primer, 5 l of 10PCR buffer (Biotools), 2 mM MgCl2, 1.25 U of Tth DNA polymerase (Biotools),and 0.2 mM each deoxyribonucleoside triphosphate (Biotools). PCRs were per-formed in 25-well microtiter plates. Negative controls consisting of sterile ultra-pure water instead of sample were included with each batch of samples analyzed.

Preparations were amplified in a DNA thermocycler (Mastercycler personal;Eppendorff, Hamburg, Germany). The PCR temperature profile for the univer-sal reaction included an initial denaturation step at 97C for 1 min; followed by26 cycles, each of a denaturation step at 97C for 45 s, a primer-annealing stepat 55C for 45 s, and an extension step at 72C for 1 min; and a final step of 72Cfor 4 min. PCR cycling conditions for the second round of amplification specificfor D. invisus comprised an initial denaturation step of 95C for 10 min; followedby 26 cycles, each of denaturation at 95C for 30s, primer annealing at 68C for1 min, and an extension at 72C for 1 min; and a final extension step of 72C for2 min. For TM7 oral clone I025, cycling conditions consisted of an initial dena-turation step of 95C for 10 min; followed by 28 cycles, each of denaturation at95C for 30s, primer annealing at 60C for 1 min, and an extension at 72C for 1min; and then a final extension step of 72C for 3 min. For Olsenella species andSynergistes oral clone BA121, the temperature profile included an initial dena-turation step at 95C for 2 min, and a touchdown PCR was performed as follows:the denaturing temperature of each cycle was carried out at 95C for 30 s. Theannealing temperature was initially set at 68C and was then lowered 0.5C everyother cycle until it reached 63C. Seventeen additional cycles were carried out at63C. Primer annealing was performed using this scheme for 30 s, and primerextension was carried out at 72C for 1 min. The final extension step was at 72Cfor 5 min. The temperature profile for the other species/phylotypes encompassedan initial denaturation step at 95C for 2 min, and touchdown PCR as follows: adenaturing temperature of each cycle at 95C for 30 s amd an annealing tem-perature initially set at 64C and then lowered 0.5C every other cycle until itreached 61C. Twenty-one additional cycles were carried out at 61C. Primerannealing was performed using this scheme for 30 s, and primer extension wascarried out at 72C for 1 min. The final extension step was at 72C for 5 min. PCRamplicons were separated by electrophoresis in a 1.5% agarose gel, which wasstained with 0.5 g/ml ethidium bromide and viewed under UV transillumina-tion. A 100-bp DNA ladder digest (Biotools) served as the molecular size stan-dard.

Sequencing. To confirm the specificity of the primers, randomly selectedrepresentative PCR products for each target species or phylotypes were purifiedwith a PCR purification system (Wizard PCR Preps; Promega, Madison, WI) andthen sequenced directly on the ABI 377 automated DNA sequencer using dyeterminator chemistry (Amersham Biosciences, Little Chalfont, Buckingham-shire, United Kingdom). Sequence data and chromatograms were inspected andedited by using BioEdit software (http://www.mbio.ncsu.edu/BioEdit/bioedit.html) (14). Sequences were then analyzed using the BLAST algorithm (1).

VOL. 43, 2005 NEW BACTERIA IN ENDODONTIC INFECTIONS 3315

RESULTS

After the first round of amplification using universal bacte-rial primers, all samples yielded an amplicon of approximately1,500 bp. This indicated that bacteria were present in all casesexamined, demonstrating the suitability of the DNA for PCRanalysis and indicating the absence of inhibitors in the reactionmixture. Negative controls using sterile ultrapure water insteadof sample yielded no amplicon.

The sequences of primers specific for each species or phy-lotype are shown in Table 1. The specificity of each primer wastested against a panel of representative oral bacteria. The useof each primer set resulted in no PCR product of the expectedsize from nontargeted species. Primers specific for O. uli, O.profusa, A. parvulum, and P. pallens resulted in one band of theexpected size when respective reference DNA was used. Toconfirm the specificity of each species- or phylotype-specificPCR primer, representative PCR products of the expected sizeobtained from clinical samples were sequenced and comparedto the original sequences in GenBank database. The primersequence for D. invisus was originally designed to target Diali-

ster oral clone GBA27 (19) but it also annealed without anymismatches to the sequences of D. invisus and Dialister oralclones FY011 and BS095. However, sequences of the eightrandomly selected representative PCR products from clinicalsamples all showed higher levels of similarity to D. invisus (99.4to 100%). Nonetheless, even though our discussion hereafterrelies on this species, the possibility exists that some of thenonsequenced PCR products could have been from other Di-alister phylotypes. For O. uli, the levels of similarity among thesequences were 98.7 to 100%. Two amplicons generated by theO. uli primer yielded sequences of a new Olsenella phylotype,EI15. For O. profusa, the levels of similarity among the se-quences were 99.1 to 100%. For P. pallens, the levels of simi-larity among the sequences were 99.8 to 100%. For A. parvu-lum, the level of similarity between the sequences was 99.3%.For Synergistes clones BH017/D084 (formerly DeferribacteresBH017/D084), the levels of similarity among the sequenceswere 98.6 to 100%. All sequences showed higher similarities toclone BH017. For Synergistes BA121, the levels of similarityamong the sequences were 99.3 to 100%. For Synergistes

TABLE 1. PCR primer pairs used for detection of novel species/phylotypes in endodontic infections

Target Primer pairs (53) Base position(amplicon size in bp)Referenceor source

Actinobaculum oral clone EL030 AGA GTT TGA TCC TGG CTC AGa 827 19CGC AGA ATC CGT GGA AAG A 825843 (844)

Atopobium parvulum AGA GTT TGA TCC TGG CTC AG 827 19TGC GGC ACG GAA GAA ATA CTC CCC 784807 (827)

Synergistes oral clones BH017/D084 AGA GTT TGA TCC TGG CTC AG 827 19CGT CAA TGT TTC CAT CTC CTA C 968989 (988)

Synergistes oral clone W090 AGA GTT TGA TCC TGG CTC AG 827 19GAA AGT ACG TCG TCG CCC TTT CAG 954977 (997)

Desulfobulbus oral clone R004 AGA GTT TGA TCC TGG CTC AG 827 19GAA GGC ACC ACC CAC TTT CAT GGG 9931016 (1,035)

Dialister invisus CAG AAA TGC GGA GTT CTT CTT CG 10071029 19CCC GGG AAC GTA TTC ACC Gb 13691387 (381)

Synergistes oral clone BA121 AGA GTT TGA TCC TGG CTC AG 827 This studyTGC GAA AGG GTC GAT CCG C 9821000 (999)

Synergistes oral clone E3_33 AGA GTT TGA TCC TGG CTC AG 827 This studyACA CTT GTA CGT CTC CAT ACA C 959980 (1,000)

Olsenella profusa AGA GTT TGA TCC TGG CTC AG 827 This studyTGC GGC ACG GAC GGA CAA TCC G 788809 (829)

Olsenella uli AGA GTT TGA TCC TGG CTC AG 827 This studyTGC GGC ACG GAG GGA TCG TCC C 788809 (829)

Prevotella pallens TGT GCG TTA TTG CAT GTA TCG TAT 442465 This studyCCC CGA AGG GCA TAT TTA TCT C 9821003 (562)

TM7 oral clone I025 CCC TGC AGT GAG GGA TAA GA 135154 4GTT TTC ATC GCT CGC TAA CTT G 590611 (477)

Universal 16S rRNA gene AGA GTT TGA TCC TGG CTC AG 827 9ACG GCT ACC TTG TTA CGA CTT 14921512 (1,505) 38

a Universal 16S rRNA gene forward primer (base position relative to E. coli 16S rRNA gene).b Universal 16S rRNA gene reverse primer (base position relative to E. coli 16S rRNA gene).

3316 SIQUEIRA AND ROCAS J. CLIN. MICROBIOL.

E3_33, the level of similarity between the sequences was99.7%. For Synergistes clone W090 (formerly DeferribacteresW090), the levels of similarity among the sequences were 98.4to 99.2%. For Desulfobulbus R004, the levels of similarityamong the sequences of the expected size were 98 to 100%. Asmaller PCR product (about 175-bp long) was also found inmany specimens amplified with the Desulfobulbus R004 primerand showed 100% similarity to the D. invisus sequence. ForActinobaculum EL030, the levels of similarity among the se-quences were 98 to 99.1%. For TM7 oral clone I025, the levelsof similarity among the sequences were 98.7 to 100%.

All target species/phylotypes were found in at least onesample from primary endodontic infections (untreated teeth).Only the three most prevalent species in primary infectionswere detected in persistent endodontic infections (root-filledteeth), but at lower frequencies. Prevalence data for the targetspecies/phylotypes in primary or persistent endodontic infec-tions are represented in Table 2.

DISCUSSION

In this study, we investigated the occurrence of some namedspecies and uncultivated phylotypes in endodontic infections.Named species comprised newly proposed species or speciesthat have undergone recent taxonomic reclassification (7, 8,18). The target species are usually difficult to cultivate in arti-ficial media and/or difficult to identify by phenotypic features(7, 8, 18), and this may have resulted in underestimation oftheir presence in endodontic infections. For instance, Dialisterspecies have been consistently detected in endodontic infec-tions only after the advent of molecular genetic approaches(22, 30). Dialister pneumosintes has been found at high preva-lence in root canals from teeth associated with different formsof periradicular diseases (30). The newly named D. invisus is ananaerobic gram-negative coccobacillus closely related to Diali-ster pneumosintes, with 93% sequence similarity between thetwo taxa (8). Munson et al. (22) first isolated this species frominfected root canals in a study using cultivation and 16S rRNA

gene sequencing. This was the only species detected in all fivecases analyzed in their study (22). In the present investigation,D. invisus was detected in 81% of the primarily infected rootcanals. This considerably high prevalence lends additional cre-dence to the assumption that this species can be an importantendodontic pathogen. Although in lower prevalence values, D.invisus was also detected in cases of failed endodontic treat-ment (14% of the teeth). This suggests that this species canalso be involved with persistent endodontic infections.

Although Olsenella and Atopobium species have been re-cently isolated/detected in endodontic infections (5, 11, 22), nostudy had consistently investigated their prevalence in theseinfections. Cultural and molecular analyses of samples takenfrom five infected root canals revealed the occurrence of O. uliin three cases, A. parvulum in two cases, and O. profusa in onecase (22). O. uli has been found to predominate over othergram-positive rods in root canal samples taken after chemo-mechanical preparation and intracanal medication, suggestingthat this species can resist intracanal disinfection measures andthereby may be involved in persistent infections (5). Herein, O.uli was present in one-third of the samples from primary in-fections and in one root-filled tooth. O. profusa and A. parvu-lum were detected in 9.5% and 5% of the untreated teeth,respectively, but in no retreatment case. These findings re-vealed that Olsenella species, particularly O. uli, are commonmembers of the microbiota associated with primary endodon-tic infections.

Prevotella species have been frequently found in differenttypes of endodontic infections (2, 13, 34). The most commonlyisolated/detected species include Prevotella intermedia, Pre-votella nigrescens, and P. tannerae (2, 39), but others, includingnonpigmented species, have also been found (13). In this study,we devised a pair of PCR primers to identify P. pallens, ablack-pigmented Prevotella species, directly in clinical samplesand detected this species in 9.5% of the primarily infectedcanals. This is the first report of the occurrence of P. pallens inendodontic infections.

In addition to newly named species, this study also focusedthe occurrence of as-yet-uncultivated bacteria in endodonticinfections. All phylotypes targeted in this study were detectedin at least one sample from primary endodontic infections; onephylotype was found in one case of persistent infection. Clonesfrom the Synergistes division were found to be common mem-bers of the endodontic microbiota in untreated teeth. One-third of these samples harbored Synergistes oral clone BA121,which was first detected in one subject with refractory peri-odontitis (23). This phylotype was also detected in one root-filled tooth. Although firstly detected in endodontic samples(22), Synergistes clone E3_33 was detected in only one speci-men. Clone BH017 (or D084) from the Synergistes group wasalso detected at a relatively high prevalence value, i.e., 29%, ofthe primary infections. Although clones BH017 and D084 wereindistinguishable by our assay, all sequences from representa-tive amplicons yielded higher similarities to the former. Thesetwo clones have also been among the most strongly associatedwith marginal periodontitis (19). Synergistes clone W090 wasdetected in 24% of the primarily infected root canals. Thisclone has been reported to occur in 73% of samples taken fromsubjects with periodontitis (19). Detection of these unculti-vated Synergistes phylotypes at high prevalence in infected root

TABLE 2. Frequency of detection of novel species/phylotypes inprimary or persistent endodontic infections as revealed by 16S

rRNA gene-based nested or heminested PCR

Species or clone(s)Primary endodontic

infections(no. of untreated teeth)a

Persistent endodonticinfections

(no. of root-filled teeth)a

Dialister invisus 17/21 (81) 3/22 (14)Synergistes oral clone

BA1217/21 (33) 1/22 (4.5)

Olsenella uli 7/21 (33) 1/22 (4.5)Synergistes oral clones

BH017/D0846/21 (29) 0/22 (0)

Synergistes oral cloneW090

5/21 (24) 0/22 (0)

Actinobaculum oralclone EL030

3/21 (14) 0/22 (0)

Desulfobulbus oralclone R004

3/21 (14) 0/22 (0)

TM7 oral clone I025 2/21 (9.5) 0/22 (0)Olsenella profusa 2/21 (9.5) 0/22 (0)Prevotella pallens 2/21 (9.5) 0/22 (0)Atopobium parvulum 1/21 (5) 0/22 (0)Synergistes oral clone

E3_331/21 (5) 0/22 (0)

a Number of positive cases for the target species/number of samples examined(percentage).

VOL. 43, 2005 NEW BACTERIA IN ENDODONTIC INFECTIONS 3317

canals suggests that they can be previously unrecognized bac-teria that play a role in the pathogenesis of periradicular le-sions.

Other uncultivated phylotypes targeted in this study werealso detected for the first time in primary endodontic infec-tions. Desulfobulbus oral clone R004 and Actinobaculum oralclone EL030 were both detected in 14% of the samples. Thesephylotypes have been previously detected at high frequenciesin subjects with marginal periodontitis (82% and 44%, respec-tively) (19). TM7 oral clone I025 was detected in 9.5% of thecases. This finding expands the list of phyla represented inendodontic infections to include TM7 bacteria. Candidate di-vision TM7 is one of the several divisions described for thedomain Bacteria, and because it has no cultivated representa-tive, members can be identified only by molecular methods.Recent studies have demonstrated a high prevalence of TM7bacteria, particularly clone I025, in association with periodon-tal diseases (4, 19). Based on our findings, it appears that cloneI025 can be found in infected root canals, but it is not soprevalent in endodontic infections as it is in periodontal infec-tions. Such a difference may be a result of different environ-mental conditions between the root canal and the gingivalcrevice, the latter being more favorable to the establishment ofTM7 bacteria. However, differences may have been due tomethodological reasons or geographical influence (3, 4).

As expected, samples from root-filled teeth yielded few pos-itive results. Because of the restricted room for bacterial es-tablishment and the low availability of nutrients within treatedroot canals, the number of bacterial species composing themicrobiota associated with persistent endodontic infections issignificantly lower than that occurring in primary endodonticinfections (32, 35). Thus, only the bacterial species that areable to adapt to the bleak environmental conditions withinfilled root canals will succeed in causing persistent endodonticinfections. E. faecalis has been shown to be the most frequentlydetected species in root-filled teeth, and 17 of the 22 casesexamined herein had been positive for this facultative species(32). Even so, this is the first study to reveal the occurrence ofD. invisus, Synergistes BA121, and O. uli in the canals of root-filled teeth; their possible association with recalcitrant perira-dicular diseases merits further elucidation.

The microbial etiology of periradicular diseases has beendemonstrated to be more complex than previously anticipatedby cultivation studies. In addition to detecting some cultivablespecies in increased prevalence, molecular methods have alsoexpanded the list of putative endodontic pathogens by inclu-sion of some fastidious bacterial species or even uncultivatedbacteria that had never been previously found in endodonticinfections by cultivation procedures (27). As a consequence, alarger and ever-expanding number of species have been sus-pected to be involved with causation of periradicular diseases.Since endodontic infections develop in a previously sterileplace which as such does not contain a normal microbiota,every bacterial species present in the mixed consortium has thepotential to play a role in the infectious process. Although thePCR assay used in the present study does not allow quantita-tion, identification of the taxa present in endodontic samplesand determination of their prevalence are still of extreme valueto help unravel the bacterial diversity in infected root canals.This information will make it possible to use other techniques

such as real-time PCR, DNA microarrays, or reverse-captureDNA-DNA hybridization assays to investigate the numbers ofthese taxa in root canal samples.

In conclusion, the present investigation has expanded the listof endodontic bacteria by including several newly named oralspecies, as well as novel phylotypes recently identified by 16SrRNA gene sequence analysis. These findings point to an ur-gent need for a steady and sustained effort to obtain thesebacteria in culture so that their phenotypic traits, includingvirulence and susceptibility to antimicrobials, can be properlyassessed.

ACKNOWLEDGMENTS

This study was supported by grants from CNPq, a Brazilian Gov-ernmental Institution.

We express our gratitude to Jari Jalava, Bruce Paster, GoranSundqvist, Lucia Teixeira, and Arie van Winkelhoff for providing someof the bacterial strains used in this study.

REFERENCES

1. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990.Basic local alignment search tool. J. Mol. Biol. 215:403410.

2. Baumgartner, J. C., B. J. Watkins, K. S. Bae, and T. Xia. 1999. Associationof black-pigmented bacteria with endodontic infections. J. Endod. 25:413415.

3. Baumgartner, J. C., J. F. Siqueira, Jr., T. Xia, and I. N. Rocas. 2004.Geographical differences in bacteria detected in endodontic infections usingpolymerase chain reaction. J. Endod. 30:141144.

4. Brinig, M. M., P. W. Lepp, C. C. Ouverney, G. C. Armitage, and D. A.Relman. 2003. Prevalence of bacteria of division TM7 in human subgingivalplaque and their association with disease. Appl. Environ. Microbiol. 69:16871694.

5. Chavez de Paz, L. E., A. Molander, and G. Dahlen. 2004. Gram-positive rodsprevailing in teeth with apical periodontitis undergoing root canal treatment.Int. Endod. J. 37:579587.

6. Conrads, G., S. E. Gharbia, K. Gulabivala, F. Lampert, and H. N. Shah.1997. The use of a 16S rDNA PCR for the detection of endodontopatho-genic bacteria. J. Endod. 23:433438.

7. Dewhirst, F. E., B. J. Paster, N. Tzellas, B. Coleman, J. Downes, D. A. Spratt,and W. G. Wade. 2001. Characterization of novel human oral isolates andcloned 16S rDNA sequences that fall in the family Coriobacteriaceae: de-scription of Olsenella gen. nov., reclassification of Lactobacillus uli asOlsenella uli comb. nov. and description of Olsenella profusa sp. nov. Int. J.Syst. Evol. Microbiol. 51:17971804.

8. Downes, J., M. Munson, and W. G. Wade. 2003. Dialister invisus sp. nov.,isolated from the human oral cavity. Int. J. Syst. Evol. Microbiol. 53:19371940.

9. Edwards, U., T. Rogall, H. Blocker, M. Emde, and E. C. Botter. 1989.Isolation and direct complete nucleotide determination of entire genes.Characterization of a gene coding for 16S ribosomal RNA. Nucleic AcidsRes. 17:78437853.

10. Figdor, D. 2002. Microbial aetiology of endodontic treatment failure andpathogenic properties of selected species. Umea University odontologicaldissertations no. 79. Umea University, Umea, Sweden.

11. Fouad, A. F., J. Barry, M. Caimano, M. Clawson, Q. Zhu, R. Carver, K.Hazlett, and J. D. Radolf. 2002. PCR-based identification of bacteria asso-ciated with endodontic infections. J. Clin. Microbiol. 40:32233231.

12. Fouad, A. F., K.-Y. Kum, M. L. Clawson, J. Barry, C. Abenoja, Q. Zhu, M.Caimano, and J. D. Radolf. 2003. Molecular characterization of the presenceof Eubacterium spp. and Streptococcus spp. in endodontic infections. OralMicrobiol. Immunol. 18:249255.

13. Haapasalo, M. 1989. Bacteroides spp. in dental root canal infections. Endod.Dent. Traumatol. 5:110.

14. Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignmenteditor and analysis program for Windows 95/98/NT. Nucleic Acids Symp.Ser. 41:9598.

15. Harper-Owen, R., D. Dymock, V. Booth, A. J. Weightman, and W. G. Wade.1999. Detection of unculturable bacteria in periodontal health and disease byPCR. J. Clin. Microbiol. 37:14691473.

16. Jung, I.-Y., B.-K. Choi, K.-Y. Kum, Y.-J. Yoo, T.-C. Yoon, S.-J. Lee, and C.-Y.Lee. 2001. Identification of oral spirochetes at the species level and theirassociation with other bacteria in endodontic infections. Oral Surg. OralMed. Oral Pathol. Oral Radiol. Endod. 92:329334.

17. Kazor, C. E., P. M. Mitchell, A. M. Lee, L. N. Stokes, W. J. Loesche, F. E.Dewhirst, and B. J. Paster. 2003. Diversity of bacterial populations on the

3318 SIQUEIRA AND ROCAS J. CLIN. MICROBIOL.

tongue dorsa of patients with halitosis and healthy patients. J. Clin. Micro-biol. 41:558563.

18. Kononen, E., E. Eerola, E. V. G. Frandsen, J. Jalava, J. Matto, S. Salmen-linna, and H. Jousimies-Somer. 1998. Phylogenetic characterization andproposal of a new pigmented species to the genus Prevotella: Prevotellapallens sp. nov. Int. J. Syst. Bacteriol. 48:4751.

19. Kumar, P. S., A. L. Griffen, J. A. Barton, B. J. Paster, M. L. Moeschberger,and E. J. Leys. 2003. New bacterial species associated with chronic peri-odontitis. J. Dent. Res. 82:338344.

20. Leys, E. J., S. R. Lyons, M. L. Moeschberger, R. W. Rumpf, and A. L. Griffen.2002. Association of Bacteroides forsythus and a novel Bacteroides phylotypewith periodontitis. J. Clin. Microbiol. 40:821825.

21. Munson, M. A., A. Banerjee, T. F. Watson, and W. G. Wade. 2004. Molecularanalysis of the microflora associated with dental caries. J. Clin. Microbiol.42:30233029.

22. Munson, M. A., T. Pitt-Ford, B. Chong, A. Weightman, and W. G. Wade.2002. Molecular and cultural analysis of the microflora associated with end-odontic infections. J. Dent. Res. 81:761766.

23. Paster, B. J., S. K. Boches, J. L. Galvin, R. E. Ericson, C. N. Lau, V. A.Levanos, A. Sahasrabudhe, and F. E. Dewhirst. 2001. Bacterial diversity inhuman subgingival plaque. J. Bacteriol. 183:37703783.

24. Paster, B. J., W. A. Falkler, Jr., C. O. Enwonwu, E. O. Idigbe, K. O. Savage,V. A. Levanos, M. A. Tamer, R. L. Ericson, C. N. Lau, and F. E. Dewhirst.2002. Prevalent bacterial species and novel phylotypes in advanced nomalesions. J. Clin. Microbiol. 40:21872191.

25. Rocas, I. N., J. F. Siqueira, Jr., A. F. B. Andrade, and M. Uzeda. 2003. Oraltreponemes in primary root canal infections as detected by nested PCR. Int.Endod. J. 36:2026.

26. Sakamoto, M., Y. Huang, M. Umeda, I. Ishikawa, and Y. Benno. 2002.Detection of novel oral phylotypes associated with periodontitis. FEMSMicrobiol. Lett. 217:6569.

27. Siqueira, J. F., Jr. 2002. Endodontic infections: concepts, paradigms andperspectives. Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod. 94:281293.

28. Siqueira, J. F., Jr., and I. N. Rocas. 2003. Detection of Filifactor alocis in

endodontic infections associated with different forms of periradicular dis-eases. Oral Microbiol. Immunol. 18:263265.

29. Siqueira, J. F., Jr., and I. N. Rocas. 2003. PCR-based identification ofTreponema maltophilum, T. amylovorum, T. medium, and T. lecithinolyticumin primary root canal infections. Arch. Oral Biol. 48:495502.

30. Siqueira, J. F., Jr., and I. N. Rocas. 2003. Positive and negative bacterialassociations involving Dialister pneumosintes in primary endodontic infec-tions. J. Endod. 29:438441.

31. Siqueira, J. F., Jr., and I. N. Rocas. 2004. Nested PCR detection of Centipedaperiodontii in primary endodontic infections. J. Endod. 30:135137.

32. Siqueira, J. F., Jr., and I. N. Rocas. 2004. Polymerase chain reaction-basedanalysis of microorganisms associated with failed endodontic treatment. OralSurg. Oral Med. Oral Pathol. Oral Radiol. Endod. 97:8594.

33. Siqueira, J. F., Jr., I. N. Rocas, and A. S. Rosado. 2004. Investigation ofbacterial communities associated with asymptomatic and symptomatic end-odontic infections by DGGE fingerprinting approach. Oral Microbiol. Im-munol. 19:363370.

34. Siqueira, J. F., Jr., I. N. Rocas, R. Souto, M. Uzeda, and A. P. Colombo. 2000.Checkerboard DNA-DNA hybridization analysis of endodontic infections.Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod. 89:744748.

35. Sundqvist, G., D. Figdor, S. Persson, and U. Sjogren. 1998. Microbiologicanalysis of teeth with failed endodontic treatment and the outcome of con-servative re-treatment. Oral Surg. Oral Med. Oral Pathol. Oral Radiol.Endod. 85:8693.

36. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W:improving the sensitivity of progressive multiple sequence alignment throughsequence weighting, position-specific gap penalties and weight matrix choice.Nucleic Acids Res. 22:46734680.

37. Tronstad, L., and P. T. Sunde. 2003. The evolving new understanding ofendodontic infections. Endod. Top. 6:5777.

38. Weisburg, W. G., S. M. Barns, D. A. Pelletier, and D. J. Lane. 1991. 16Sribosomal DNA amplification for phylogenetic study. J. Bacteriol. 173:697703.

39. Xia, T., J. C. Baumgartner, and L. L. David. 2000. Isolation and identifica-tion of Prevotella tannerae from endodontic infections. Oral Microbiol. Im-munol. 15:273275.

VOL. 43, 2005 NEW BACTERIA IN ENDODONTIC INFECTIONS 3319