Detection and Clonal Analysisof Anaerobic Bacteria Associatedto Endodontic-Periodontal LesionsCassio V. Pereira,* Rafael N. Stipp,† Douglas C. Fonseca,‡ Luciano J. Pereira,§ and Jose F. Hofling†

Background: Microbial agents in root canal systems can in-duce periodontal inflammation. The aims of this study are todetect anaerobic microorganisms in endodontic-periodontallesions, determine the genetic diversity among them, and as-sess the simultaneous colonization of the pulp and periodontalmicroenvironments by a single clone.

Methods: Twenty-seven teeth of patients with endodontic-periodontal lesions were selected. Samples were spread on anagar-blood medium, the detection of each species was per-formed using a polymerase chain reaction, and the determina-tion of the simultaneous presence of the same species in themicroenvironments by one or more clones was determined usingarbitrarily primed PCR.

Results: Prevotella intermedia (Pi) was the most prevalentspecies of the colonies in periodontal pockets, whereas Porphy-romonas gingivalis (Pg) and Pi were the more prevalent in rootcanals. Isolates of Pi and Pg were simultaneously identified inroot canals and periodontal pockets. Eighteen percent of teethexhibited the simultaneous colonization by Pg, Tannerella for-sythia (previously T. forsythensis), and Porphyromonas endo-dontalis in the pulp and periodontal microenvironments. Thepresence of these species was noted even in niches from whichno colonies were isolated. Seventeen different genotypes werefound in periodontal and pulp sites, with the majority of sites col-onized by one or two different genotypes. A high degree of geno-type similarity was found for samples of Pg isolated from onlyone site as well as for those isolated from both microenviron-ments.

Conclusion: Different clones of Pi and Pg with a high intraspe-cific genotype similarity were found to colonize the same ana-tomic sites in endodontic-periodontal infections. J Periodontol2011;82:1767-1775.

KEY WORDS

Bacteria, anaerobic; endodontics; periapical diseases;periodontics; random amplified polymorphic DNA technique.

Bacteria can harm the oral tissuethrough the release of toxins, en-zymes, and residual metabolic

products. Such microorganisms are inevi-tably correlated to the etiology of pulp andperiapical lesions, particularly protease-producing anaerobic bacteria, which breakdown diverse polypeptides involved in thehost defense mechanism.1-3

There is a close relationship betweenroot canal systems (root canal treat-ment) and the periodontal environment.Lateral and accessory canals can trans-port toxic substances and facilitate thebacterial colonization of dental tissues.Toxins and microbial agents in the pulpspace can induce periodontal inflamma-tion. However, the effect of these irritantsfrom the periodontal space to the pulp isless clearly defined.4 Experiments com-paring polymorphisms in genomic fin-gerprints were conducted to clarify thisissue.5

The evidence of a microbial involve-ment, especially of pigmented speciessuch as Porphyromonas gingivalis (Pg),Prevotella intermedia (Pi), and Prevotellanigrescens (Pn), in the pathogenesis ofperiodontal inflammation and periapicallesions accumulated in the literature.6,7

However, a microbial species consists ofa large number of clones, and the major-ity of infectious bacterial diseases can becaused by some clones of pathogenicspecies.8-10

A number of studies5,11,12 that aimed tocharacterize the microbial communities of

* Department of Microbiology and Immunology, Lavras University Center, Lavras, MG,Brazil.

† Department of Microbiology and Immunology, Piracicaba Dental School, University ofCampinas, Piracicaba, SP, Brazil.

‡ Department of Periodontics, Lavras University Center.§ Department of Physiology and Pharmacology, Federal University of Lavras, Lavras, MG,

Brazil.

doi: 10.1902/jop.2011.110063

J Periodontol • December 2011

1767

pulp and periradicular infections on the basis ofDNA fingerprints revealed an unexpected diversityof microbial species. However, there are few stud-ies using polymerase chain reaction (PCR) and ar-bitrarily primed PCR (AP-PCR) for the detection ofvirulent clones of species that cause endodontic-periodontal infections.

The hypothesis of the present study is that there isa genetic diversity among anaerobic microorganismsin endodontic-periodontal lesions and that there isa simultaneous colonization of the pulp and periodon-tal microenvironments by a single clone. This investi-gation is important because different strains from thesame bacteria species can present variations in theirvirulence and resistance influencing the treatmentoutcome. Thus, the aim of the present study is to per-form molecular analyses for the detection of Pg, Pi,Pn, Porphyromonas endodontalis(Pe), and Tanner-ella forsythia (Tf ) in patients with endodontic-peri-odontal lesions. After isolation in an anaerobicculture medium and molecular characterization, thesamples were submitted to AP-PCR for the determina-tion of genetic diversity among the microorganismsand the assessment of the possibility of simultaneouscolonization of the pulp and periodontal microenvi-ronments by a single clone.

MATERIALS AND METHODS

This study received approval from the Ethics Commit-tee, Sacred Heart University, Bauru, SP, Brazil, underprocess number 013/2003, and data collection wasdone between 2004 and 2005 considering the diffi-culty in finding elective patients. The participants weremade aware of the objectives and procedures of thestudy and agreed to participate by signing terms of in-formed consent.

Twenty-seven teeth were selected from male(40%) and female (60%) adults (8 males and 13 fe-males; age range: 27 to 51 years) with endodontic-periodontal lesions diagnosed through clinical andradiographic examinations at the dental clinics ofthe Sacred Heart University. None of the patients,as reported in their files, were smokers or under theuse of any medication (including contraceptive pillsfor females). Inclusion criteria included intact teethwith necrotic pulp tissue, periapical lesions, chronicperiodontal disease, and the absence of systemicconditions. For the diagnosis of chronic periodontaldisease, the following clinical-radiographic factorswere considered: clinical attachment levels, the pres-ence of inflammation, probing depths (periodontalpockets >5 mm), and bone loss.2,13 Exclusion criteriawere the use of antibiotic therapy (6 months before thesample collection) and the clinical diagnosis of fistulabecause these factors could have influenced the for-mation of the microbiota in these microenvironments.

All volunteers had intact teeth or restorations on theenamel level only.

After the removal of the supragingival biofilm, pre-viously sterilized paper cones were used to collectsamples from the interior of the root canal system (af-ter the aseptic opening of the crown) as well as fromthe periodontal pocket of the tooth affected by end-odontic-periodontal lesions. Samples were also col-lected from the gingival sulcus of the same tooth inthe opposing hemiarch as the control for the determi-nation of the absence of microorganisms in regionsthat did not exhibit periodontal disease. In cases ofatresic and/or curved canals that hindered the collec-tion with paper cones, the procedure was carried outwith the aid of endodontic files.

Samples were transported to the Laboratory ofMicrobiology and Immunology, Piracicaba DentalSchool, University of Campinas, Piracicaba, SP, Brazilin reduced transport fluid in £1 hour from the time ofcollection and dispersed and agitated for 20 seconds.For each sample, 100 mL of 10-1 to 10-4 dilutions werespread on dishes with brucella agar medium enrichedwith defibrinated sheep blood, hemin (10 mg/mL),and vitamin K1 (1 mg/mL). The dishes were incubatedat 37�C in anaerobic jars containing two anaerobiosisgenerators,i one anaerobiosis indicator, and silica toabsorb moisture for £10 days. After this period, thelargest possible number of colonies with dark-brownto black pigmentation were isolated from primary cul-ture dishes and cut for the obtainment of a pure cul-ture.

Pure cultures were also identified through Gramstaining for the confirmation of the microscopic char-acteristics with Gram-negative rods. Pairs of specificprimers were used for Pg, Pi, Pn, Pe, and Tf for thePCR amplification of a specific sequence of DNA(Table 1).10,14,15

PCR-amplification reactions were performed in areaction mixture containing 1· PCR buffer (10 mMTris-HCl, pH 9.0, 50 mM KCl, and15 mM MgCl2),200 mM of four deoxynucleotide triphosphates,¶

12.5 pmol of each primer, 2.5 U Taq DNA polymer-ase,# and 5 mL supernatant from the samples submit-ted for DNA extraction at a final volume of 50 mL. Thereaction mixture was incubated in a thermocycler.Samples were initially denatured at 94�C for 5 minutesand then submitted to 35 cycles of 60 seconds at94�C, 45 seconds at 67�C, and 30 seconds at 72�C.A negative control (absence of DNA template), a pos-itive control, and molecular mass marker (100–basepair [bp] DNA ladder) were included in each amplifi-cation set. PCR-amplification products were ana-lyzed through electrophoresis in a 1.5% agarose gel

i Anaerogen, Oxoid Brazil, Sao Paulo, SP, Brazil.¶ Invitrogen Life Technology of Brazil, Sao Paulo, SP, Brazil.# Invitrogen Life Technology of Brazil.

Anaerobic Bacteria and Endodontic-Periodontal Lesions Volume 82 • Number 12

1768

run in Tris-borate-EDTA buffer, stained with ethid-ium bromide, and video documented under ultravio-let light.

AP-PCR was used for the determination of virulentclones among species of anaerobic filamentous bac-teria collected from root canals and periodontalpockets.5 AP-PCR was also used for samples of spe-cies isolated from only one of the microenvironments(root canal systems or the periodontium). This prac-tice did not allow the determination of the simulta-neous colonization by clones of these species inpatients with this occurrence but contributed towardthe knowledge of the genetic variability of these mi-croorganisms in each anatomic site.

A random-amplification polymor-phic DNA (RAPD) analysis was usedto determine the existence of the ge-netic affinity between the same spe-cies isolated from the root canalsand gingival sulcus simultaneouslyor from only one anatomic site in asingle patient.

The specification of samples as Pg,Pi, and Pn was carried out using PCRDNA amplification of a specific se-quent of bp for each of the speciesas described. In complement to theclinical isolates, standard strains wereused as a reference: Pg AmericanType Culture Collection 33277, PiAmerican Type Culture Collection25611, and Pn National Collectionof Type Cultures 9336.

Total genomic DNA was obtainedusing additional RNase treatment.16

The concentration of DNA was calcu-lated from the measurement (260nm) in a spectrophotometer at A260,and the quality was estimated by theA260:A280 ratio, electrophoresis inan agarose gel, and comparison withstandard DNA.

The AP-PCR amplification was per-formed in a volume of 25 mL containing200 mM deoxynucleotide triphos-phate,** 1.2 mM primer, 50 ng DNAtemplate, and 2.5 U Taq DNA polymer-ase†† in 1· PCR buffer, 10 mM Tris-HCl,pH 9.0, 50 mM KCl, and 1.5 mM MgCl2.A negative control without DNA tem-plate was included in each experiment.Arbitrary primers 970 to 11 (59-GTAAGGCCG-39) and 13 (59- CAG-CACCCAC-39) were selected.16 Thereaction mixture was submitted toamplification in a thermocycler pro-

grammed for one cycle of denaturation at 94�C for12 minutes followed by 30 cycles at 94�C for 1 minute,32�C for 1 minute, and 72�C for 2 minutes and a finalelongation at 72�C for 7 minutes. Samples werecooled to 4�C, and the PCR-amplification productswere analyzed through electrophoresis in a 1.5% aga-rose gel run in Tris-borate-EDTA buffer. A standardmolecular mass of 100 bp‡‡ was included in eachgel. DNA was stained with ethidium bromide, and eachgel was video documented under ultraviolet light.

Table 1.

Primers for PCR Identification of Pg, Pi, and Pn

Anaerobic Bacteria Primers bp (n)

Pg 59-AGGCAGCTTGCCATACTGCG-39 44359-ACTGTTAGCAACTACCGATGT-39

Pi 59-CGGTCTGTTAAGCGTGTTGTG-39 9959-CACCATGAATTCCGCATACG-39

Pn 59-CAGCCAAACACGATACCTGTTG-39 15059-TTCCATTGGACACATCAGCATT-39

Pe 59-GCTGCAGCTCAACTGTAGTC-39 67259-CCGCTTCATGTCACCATGTC-39

Tf 59-GGGTGAGTAACGCGTATGTAACCT-39 12759-ACCCATCCGCAACCAATAAA-39

Table 2.

Number of Colonies Identified as Pg, Pi, Pn, Pe, and TfIsolated From Periodontal Pockets and Root Canals

Black-Pigmented

Microbial Species

Periodontal Pocket Root Canal

n % n %

Pg 7 9.6 5 10.9

Pn 4 5.5 3 6.5

Pi 18 24.7 5 10.9

Pe 1 1.4 0 0.0

Tf 0 0.0 0 0.0

Other black-pigmentedspecies

43 58.9 33 71.7

Total 73 100 46 100

Number of colonies identified as Pg, Pn, Pe, and Tf isolated from periodontal pockets and root canalsin teeth with endodontic-periodontal lesions.

** Invitrogen Life Technology of Brazil.†† Invitrogen Life Technology of Brazil.‡‡ Invitrogen Life Technology of Brazil.

J Periodontol • December 2011 Pereira, Stipp, Fonseca, Pereira, Hofling

1769

The determination of the colonization of the peri-odontium or root canal alone and the simultaneouspresence of the same species in both microenviron-ments by one or more infecting clones was performedthrough a comparison of the RAPD fingerprinting ofthe multiple isolates.

The patterns of DNA amplified through AP-PCR(arbitrary primers 970-11) were expressed numeri-cally by their mobility values in the gel (in millimeters).Mobility values were converted into binary values, therepresentations of which corresponded to the pres-ence and absence of bands (1 and 0, respectively).Isolates were characterized by the combinations ofbands of high and medium intensity and were repro-ducible after two independent assays such thatdistinct combinations of polymorphic bands weredesignated electrophoretic types.17

The genetic relationship among isolates was deter-mined by the similarity coefficient (simple matching),which allowed for the use of data based on binaryvalues. Thus, the genetic similarity matrix was de-signed and treated by the sequential hierarchical ag-glomerative non-overlapping clustering technique ofthe unweighted pair group method of arithmeticmean to generate a tree with a two-dimensional clas-sification denominated a trellis diagram.18 Theseanalyses were performed with the aid of the pro-gram.§§ In all analyses, samples with electrophoreticprofiles revealing a 95% to 100% degree of similaritywere considered to be the same genotype.19

In statistical analyses of data, the x2 test was usedto compare the prevalence of species identified in dif-ferent anatomic sites (the periodontal pocket and rootcanal). Data were entered in a statistical software pro-gram.ii

Table 3.

Molecular Detection of Pg, Tf, and PePeriodontal Pockets and Root Canals

Periodontal Pocket Root Canal

Tooth Pg Tf Pe Pg Tf Pe

1 + + + + - +

2 - - - + - -

3 + - - - - -

4 + - - + + -

5 + + + + + -

6 + - - - - -

7 + + + + - -

8 + + + + - -

9 + + + + + +

10 + + + + + +

11 + + + + + +

12 + + + + - +

13 + + - - - +

14 + + - + - +

15 + + + + + +

16 + + - - - +

17 + + - + + +

18 + + - - - -

19 + + + + + +

20 + + - - - +

21 + - - - - +

22 + - - + + +

23 - - + + + -

24 + + + + - +

25 + + + + - +

26 + + + + - +

27 + + + + - ++ = molecular detection of the species in the root canal or periodontalpocket microenvironments.- = absence of molecular detection of the species in the root canal orperiodontal pocket microenvironments.

Table 4.

Prevalence of Pg, Pe and Tf in SamplesFrom Periodontal Pockets and Root Canals

Molecular

Detection

Pg Pe Tf*

Microenvironment n % n % n %

Periodontal pocket + 25 92.6 16 59.3 20 74.1- 2 7.4 11 40.7 07 25.9

Root canal + 20 74.1 19 70.4 09 33.3- 07 25.9 08 29.6 18 66.7

Prevalence of Pg, Pe and Tf in samples from periodontal pockets and rootcanals in teeth with endodontic-periodontal lesions.* P = 0.029.+ = molecular detection of the species in the root canal or periodontalpocket microenvironments.- = absence of molecular detection of the species in the root canal orperiodontal pocket microenvironments.

§§ NTSYSpc v.1.70, Applied Biostatistics, Port Jefferson, NY.ii SPSS, v.14.0, IBM, Chicago, IL.

Anaerobic Bacteria and Endodontic-Periodontal Lesions Volume 82 • Number 12

1770

RESULTS

Table 2 displays the number of colonies identified asPg, Pi, Pn, Pe, and Tf by PCR based on pure cultures.Pi was the most prevalent among the clinical isolatesfrom periodontal pockets, with 18 colonies identified(24.6%). Among the samples from a root canal, fivecolonies were identified as Pg, and five colonies wereidentified as Pi, with each species accounting for10.8% of the total number of isolates in this anatomicsite. Pi was identified in root canals and periodontalpockets of two teeth (patients 16 and 17), and Pg

was identified in both anatomic sites in one tooth (pa-tient 17). All samples identified as Pn (n = 7) colonizedonly one microenvironment in the same tooth. Amongthe 119 samples isolated, only 42 samples were iden-tified as Pg, Pi, or Pn, indicating a greater variability ofblack-pigmented species as colonizers of teeth withendodontic-periodontal lesions. None of the strainstested was characterized as being from the speciesTf, and only one isolate from a periodontal pocket(tooth 6) was identified as Pe. Additionally, none ofthe control samples were positive for any anaerobicbacteria tested.

Tables 3 and 4 display the results of the genomicDNA extraction of possible bacteria on the papercones used for the collection of samples from micro-environments and the confirmation of the presence ofPg, Pe, and Tf by PCR, even if the culture had beennegative for black-pigmented colonies. The electro-phoretic profiles indicated the simultaneous coloniza-tion by the three species in periodontal pockets androot canals in 18.5% of teeth (n = 5). There was a prev-alence of92.6%(n = 25), 59.3% (n = 16) and 74.1% (n =20) in the periodontal pocket and 74.1% (n = 20),70.4% (n = 19) and 33.3% (n = 9) in the root canalfor Pg, Pe, and Tf, respectively.

Figure 1.Representative gel and trellis diagram Pi strains obtained with OperonPrimer Arbitrary (OPA-970)¶¶ in samples isolated simultaneously fromboth sites: periodontol pocket and root canal (teeth 16 and 17), or fromonly one of the microenvironments (teeth 7, 8, 15 and 18). The firstnumber denotes the tooth number in accordance to table 3, the letterdenotes the isolated site (periodontal pocket (p) or root canal (c) and thesecond number denotes the sample number from each microenviroment.MM = molecular mass.

Figure 2.Representative gel and trellis diagram of Pg strains obtained with OPA-970 ## in samples isolated simultaneously from both sites: periodontalpocket and root canal (tooth17), or fromonly one of the microenvironments(tooth 3). The first number denotes the tooth number in accordance totable 3, the letter denotes the isolated site (periodontal pocket (p) orroot canal (c) and the second number denotes the sample number fromeach microenviroment. MM = molecular mass.

¶¶ Invitrogen Life Technology of Brazil.## Invitrogen Life Technology of Brazil.

J Periodontol • December 2011 Pereira, Stipp, Fonseca, Pereira, Hofling

1771

The image in the gel revealed the simultaneouspresence of Pg in the samples collected from peri-odontal pockets and root canals of 18 of the 27 teethsubmitted for analysis. Only two samples obtainedfrom periodontal pockets and seven from root canalsdid not exhibit this species. Only Tf had a significantlygreater prevalence (P = 0.029) in periodontal pockets incomparison to the colonization in root canals (Table 4).

PCR results for the detection of Pg, Pe, and Tf basedon the DNA extraction of the bacteria present on thepaper cones used for the sampling also indicatedthe presence of these species in microenvironmentsin which there was no isolation of colonies.

Figures 1 through 3 display the electrophoretic pro-files and biologic trellis diagrams resulting from thegenetic similarity analysis of species isolated simulta-neously from both microenvironments and specieswith >1 isolate from the same anatomic site, respec-tively. Trellis diagrams revealed the occurrence of17 different genotypes established in periodontaland pulp sites, with the majority of individuals colo-nized by one to two distinct genotypes.

The intra-individual analysis revealed the coloni-zation by strains with a high degree of genetic similar-ity in microenvironments of periodontal pockets androot canals in the samples identified in teeth 16 and17. The results also indicate a slight coincident occur-rence between genotypes from samples 1 and 4 iso-lated from the periodontium of tooth 15 and samples1 and 2 isolated from the root canal of tooth 18 (Fig. 1).

Trellis diagrams of Pg samples revealed a high de-gree of genotype similarity for samples isolated froma single site as well as those isolated from both micro-environments (Fig. 2). However, colonies identified asPn were only isolated from the root canal, and the trel-lis diagrams displayed in Figure 3 indicates the colo-nization by three distinct genotypes in this patient, two

of which (samples 2 and 3 fromtooth 9) exhibited a greatersimilarity to one another.

DISCUSSION

Pathologic alterations in thecervical periodontium demon-strated that the mechanismsinvolved in periodontal dis-ease were similar to those in-volved in the pathogenesis ofperiapical lesions. In the pres-ent study, all volunteers hadintact teeth or restorations onthe enamel level only, with noindication of the contamina-tion of the pulp tissue throughthe crown pathway. Thus, thelesions were endodontic-peri-

odontal lesions, in which the primary infection waslikely of a periodontal origin because the directcommunication between the pulp and periodontiumwas limited to the apical foramen and lateral ca-nals.5

A number of microorganisms, mostly anaerobicbacteria, may be responsible for the onset and progres-sion of endodontic-periodontal disease.4,12 Episodesof disease may represent a change in the ecologicalbalance between the bacterium and host factors asa result, for instance, of an alteration in the relativeand absolute number of these microorganisms ora modulation in specific factors of the host.20

The culture method is considered the referencemethod compared to other techniques used to deter-mine the presence of microorganisms. Advantages ofthe culture method include the possibility of quanti-fying isolated species, obtaining strains in a pure cul-ture for subsequent studies (such as genotyping), andthe determination of virulence factors. Moreover, theculture method is the only manner to obtain an antibio-gram. In contrast, the culture consists of exhaustive,burdensome laboratory steps and also produces lowerdetection values compared to methods based on mo-lecular biology in which, theoretically, much smallernumbers of microorganisms can be detected.14,19

The lower detection values for culture methods com-pared to molecular biology methods was demonstratedin the comparison of results of the identification of Pi, Pg,Pn, Pe, and Tf among isolated colonies and the de-tection of Pg, Pe, and Tf in periodontal pockets and nec-rotized rootcanals (Tables2 through4).Results indicatethe detection of these species through PCR analysis inanatomic sites not detected by cultures.

Figure 3.Representative gel and trellis diagram of Pn strains obtained with OPA-970 *** in samples isolatedfrom only the pulpar microenvironment (tooth 9). The first number denotes the tooth number in accordanceto table 3, the letter denotes the isolated site (periodontal pocket (p) or root canal (c) and the secondnumber denotes the sample number from each microenviroment. MM = molecular mass.

*** Invitrogen Life Technology of Brazil.

Anaerobic Bacteria and Endodontic-Periodontal Lesions Volume 82 • Number 12

1772

Some sites had Tf and Pe, as detected by PCR, butthere was no growth of Tf and Pe on agar plates. Thisresult implies that Tf and Pe were under the culture de-tection levels. Thus, the non-detection of certain spe-cies using the culture method did not necessarilymean the absence of these species but, rather, indi-cated that the number of microorganisms was belowthe detection threshold of the method. The absenceof pure cultures justified the non-use of PCR for thedetection of Pi and Pn. In contrast, the clinical signifi-cance of the detection of pathogens that representeda tiny portion of the oral microbiota in terms of the col-onization consistency seemed questionable, as didwhether such low levels were capable of offering suffi-cient virulence to cause the destruction of dental sup-port tissues.16 These statements appeared to explainthe results of the present study and complementedthose of previous studies1,10,21 that demonstrated thefrequent isolation of black-pigmented anaerobic bac-teria of the genera Porphyromonas and Prevotella inroot canal infections, whereas the culture of these mi-croorganisms in the laboratory was variable. Accord-ing to the authors cited,1,10,21 these variations weredue to differences in the selection of cases and the col-lection, transportation, cultivation, and identificationmethodologies.

Values displayed in Table 2 indicate the greaterprevalence of Pi among the clinical isolates from peri-odontal pockets (24.6%) compared to the other spe-cies identified. Previous studies1,10,16 that involvedthe identification of pathogens of endodontic and peri-odontal infections using a similar methodology re-ported values close to those of the present study.The greater prevalence of Pi in these microenviron-ments was directly related to its virulence becauseits high degree of hemolytic activity determined itspathogenicity and the progression of the disease.The hemoglobin and hemin that resulted from thehemolytic action on red blood cells was a strong exog-enous source of iron that favored the growth of black-pigmented bacteria, including Pi.10 However, therewas no significant difference between the different mi-croenvironments in the prevalence of Pg, Pn, and Pi.

The low prevalence of Pn in the present study iscompatible with that reported in previous studies,7,22

which indicated that that this species accounted forthe majority of the genus Prevotella in healthy individ-uals, whereas active, healthy sites in patients withperiodontal infection were mainly colonized by Pi.The colonization of Pi appeared to be more stable insites with periodontal destruction compared to inhealthy sites.23 In contrast, one study24 reported thatboth species were often found in sites with and withoutperiodontal disease.

Among clinical isolates, a significant number ofsamples were not identified as belonging to any of

the species tested. This result suggested that end-odontic-periodontal lesions were multicolonized bya considerable variety of microbial agents.4,9,12,19

The greater prevalence of species from the genusPrevotella in the results of the isolation and identifi-cation of black-pigmented bacteria may have beenexplained by their greater ease of culture becausethese species seemed to tolerate minimal levels ofoxygen. Similar behavior seems not to occur withspecies of the genus Porphyromonas, which are con-sidered more fastidious and therefore, may beunderestimated by culture methods such as thoseapplied in the present study.1,25

Tables 3 and 4 indicate that there was the simulta-neous colonization in the pulp and periodontal ana-tomic sites by all species tested. Pg was the mostprevalent, followed by Pe, which is the same preva-lence order reported in the literature,4 with values of72% and 56%, respectively. Different bacterial pat-terns were evidenced between the periodontium androot canals, which thereby confirmed the need for ad-ditional studies that can demonstrate the degrees ofcolonization of each species and their importance tothe etiology of the disease and the colonizing commu-nities of these microenvironments.12,26,27

There was a significant difference in the detection ofTf between the root canal (33%) and periodontium(74.1%). Previous studies10,28 that performed PCRsreport similar values (79%) regarding the prevalenceof this species in periodontal disease. The comparisonof the results in the different sites suggested the exis-tence of modulating factors, such as immunologicmechanisms of the host, genetic predisposition, sys-temic alterations, and smoking habits, as determi-nants of the microbial colonization and successionin oral microenvironments, favoring the colonizationand growth of this microorganism in the periodontalenvironment.4

The analysis of the genetic diversity of the Pi iso-lates (Figs. 1 through 3) revealed that the majorityof individuals had one to two different clones, whichcorroborated the findings of previous studies7,23,29-31

that used similar methodologies. In two teeth (16and 17), colonies of Pi were simultaneously isolatedfrom the periodontal pocket and root canal. The gen-otyping of these microorganisms indicated an elec-trophoretic profile that was similar to the 80% levelbetween these colonies (Fig. 1), thereby confirmingthat the pulp-infection pathway originated from a pri-mary periodontal infection.32,33 One study30 on theclonality of other bacterial species involved in the etio-pathogenesis of periodontal disease also reveals thedetection of some clone types per patient or site.

Pg was identified in colonies isolated from the pulpand periodontal microenvironments of tooth 17. Thetrellis diagram in Figure 2 demonstrates a nearly

J Periodontol • December 2011 Pereira, Stipp, Fonseca, Pereira, Hofling

1773

90% similarity between isolates and suggested thatthe colonization of both anatomic sites was performedby the same clone of this species, as demonstratedelsewhere.16

Only one tooth exhibited >1 colony as Pn fromthe same microenvironment (9C1 through 9C3).The genotyping indicated a high degree of similaritybetween two of the samples (Fig. 3). Previous studiesdetected multiple clone types of Pn in 43% of individ-uals,16 whereas other studies reported only one clonetype of the species colonizing periodontal and end-odontic sites in the same oral cavity.30 Studies on thegenetic diversity of periodontal pathogens revealed abroad heterogeneity within the species7,16,23,24,29,30

and the absence of predominant clone types associ-ated with the disease and demonstrated the stabilityof clones even after periodontal treatment.23,24,30

These findings suggested that all clones of a speciesare equally effective at colonizing periodontal sitesbecause the potential for virulence is not restrictedto a particular clone, in contrast with other truly path-ogenic species.

AP-PCR was selected for this study because nestedPCR may be useful for distinguishing alleles but not forgenotyping. Furthermore, multiplex PCR is a modifi-cation of PCR intended to rapidly detect deletions orduplications in a large gene. Although it can be usefulfor genetic discrimination, the number of producedPCR bands is frequently lower than that obtained withAP-PCR. Thus, multiplex PCR can be considered lessefficient for genetic discrimination. AP-PCR (alsoknown as RAPD) is the technique of choice for rapid,cheap, and confident genetic discrimination or fortracing microorganisms in the environment. Becausemany amplicons are generated, differences, if theyexist, are easier detected.5,11

CONCLUSIONS

The present study demonstrates that Pi and Pg werethe most frequently identified species among the col-onies isolated from periodontal pockets and root ca-nals, respectively. Pn was less prevalent in patientswith endodontic-periodontal lesions, whereas Pg hadhigh rates of simultaneous detection in the pulp andperiodontal microenvironments. Different clones ofPi and Pg colonized the same anatomic sites, the pe-riodontium and/or pulp, in the endodontic-periodontalinfection, with these clones exhibiting a high degree ofintraspecific genotype similarity.

ACKNOWLEDGMENTS

The authors are grateful to the Brazilian fosteringagency Research Support Foundation of the Stateof Sao Paulo, Sao Paulo, SP, Brazil, for supporting thisstudy (grant process number 2003/10423-4) and re-searchers Vanessa Pardi, PhD, from Sacred Heart

University, and Daniel Saito, PhD, from State Univer-sity of Campinas, Piracicaba, SP, Brazil, for statisticalanalyses and technical assistance, respectively. Dr.Pereira acknowledges a scholarship from the Na-tional Council for Scientific and Technological De-velopment (Brasilia, Federal District, Brazil). Theauthors report no conflicts of interest related to thisstudy.

REFERENCES1. Gomes BP, Jacinto RC, Pinheiro ET, et al. Porphy-

romonas gingivalis, Porphyromonas endodontalis, Pre-votella intermedia and Prevotella nigrescens inendodontic lesions detected by culture and by PCR.Oral Microbiol Immunol 2005;20:211-215.

2. Lindhe J, Karing T, Lang NP. Clinical Periodontologyand Implant Dentistry, 4th ed. Oxford, United King-dom: Blackwell Munksgard; 2003:1448.

3. Slots J. Enzymatic characterization of some oral andnonoral gram-negative bacteria with the API ZYMsystem. J Clin Microbiol 1981;14:288-294.

4. Baumgartner JC, Siqueira JF Jr., Xia T, Rocxas IN.Geographical differences in bacteria detected in end-odontic infections using polymerase chain reaction. JEndod 2004;30:141-144.

5. Welsh J, McClelland M. Fingerprinting genomes usingPCR with arbitrary primers. Nucleic Acids Res 1990;18:7213-7218.

6. Sundqvist G, Johansson E, Sjogren U. Prevalence ofblack-pigmented bacteroides species in root canalinfections. J Endod 1989;15:13-19.

7. van Steenbergen TJ, Van der Velden U, Abbas F, deGraaff J. Microflora and bacterial DNA restrictionenzyme analysis in young adults with periodontitis. JPeriodontol 1991;62:235-241.

8. Finlay BB, Falkow S. Common themes in microbialpathogenicity. Microbiol Rev 1989;53:210-230.

9. Petit MD, van Winkelhoff AJ, van Steenbergen TJ, deGraaff J. Porphyromonas endodontalis: Prevalenceand distribution of restriction enzyme patterns infamilies. Oral Microbiol Immunol 1993;8:219-224.

10. Boutaga K, van Winkelhoff AJ, Vandenbroucke-Grauls CM, Savelkoul PH. Periodontal pathogens: Aquantitative comparison of anaerobic culture and real-time PCR. FEMS Immunol Med Microbiol 2005;45:191-199.

11. Williams JG, Kubelik AR, Livak KJ, Rafalski JA,Tingey SV. DNA polymorphisms amplified by arbitraryprimers are useful as genetic markers. Nucleic AcidsRes 1990;18:6531-6535.

12. Montagner F, Gomes BP, Kumar PS. Molecularfingerprinting reveals the presence of unique com-munities associated with paired samples of rootcanals and acute apical abscesses. J Endod 2010;36:1475-1479.

13. Savage A, Eaton KA, Moles DR, Neeleman I. Asystematic review of definitions of periodontitis andmethods that have been used to identify this disease.J Clin Periodontol 2009;36:458-467.

14. Nagashima S, Yoshida A, Suzuki N, Ansai T, TakeharaT. Use of the genomic subtractive hybridization tech-nique to develop a real-time PCR assay for quanti-tative detection of Prevotella spp. in oral biofilmsamples. J Clin Microbiol 2005;43:2948-2951.

Anaerobic Bacteria and Endodontic-Periodontal Lesions Volume 82 • Number 12

1774

15. Fouad AF, Barry J, Caimano M, et al. PCR-basedidentification of bacteria associated with endodonticinfections. J Clin Microbiol 2002;40:3223-3231.

16. Goncxalves RB, Robitaille M, Mouton C. Identical clonaltypes of Porphyromonas gingivalis or Prevotella ni-grescens recovered from infected root canals and sub-gingival plaque. Oral Microbiol Immunol 1999;14:197-200.

17. Selander RK, Caugant DA, Ochman H, Musser JM,Gilmour MN, Whittam TS. Methods of multilocus en-zyme electrophoresis for bacterial population geneticsand systematics. Appl Environ Microbiol 1986;51:873-884.

18. Sneath PHA, Sokal RR. Numerical Taxonomy. SanFrancisco: Freeman and Company; 1973:573.

19. Alves AC, Napimoga MH, Klein MI, Hofling JF,Goncxalves RB. Increase in probing depth is correlatedwith a higher number of Prevotella intermedia geno-types. J Periodontol 2006;77:61-66.

20. Lamont RJ, Jenkinson HF. Life below the gum line:Pathogenic mechanisms of Porphyromonas gingivalis.Microbiol Mol Biol Rev 1998;62:1244-1263.

21. Guillot E, Mouton C. PCR-DNA probe assays foridentification and detection of Prevotella intermediasensu stricto and Prevotella nigrescens. J Clin Micro-biol 1997;35:1876-1882.

22. Maeda N, Okamoto M, Kondo K, et al. Incidence ofPrevotella intermedia and Prevotella nigrescens inperiodontal health and disease. Microbiol Immunol1998;42:583-589.

23. Matto J, Saarela M, von Troil-Linden B, et al. Distri-bution and genetic analysis of oral Prevotella intermediaand Prevotella nigrescens. Oral Microbiol Immunol1996;11:96-102.

24. Teanpaisan R, Douglas CW, Walsh TF. Character-isation of black-pigmented anaerobes isolated fromdiseased and healthy periodontal sites. J PeriodontalRes 1995;30:245-251.

25. Menard C, Mouton C. Clonal diversity of the taxonPorphyromonas gingivalis assessed by random am-plified polymorphic DNA fingerprinting. Infect Immun1995;63:2522-2531.

26. Siqueira JF Jr., Rocxas IN, Debelian GJ, et al. Profilingof root canal bacterial communities associated with

chronic apical periodontitis from Brazilian and Norwe-gian subjects. J Endod 2008;34:1457-1461.

27. Saito D, Marsh TL, de Souza Cannavan F, Hofling JF,Goncxalves RB. Assessment of intraradicular bacterialcomposition by terminal restriction fragment length poly-morphism analysis. Oral Microbiol Immunol 2009;24:369-376.

28. Tanner A, Bouldin H. The microbiota of early peri-odontitis lesions in adults. J Clin Periodontol 1989;16:467-471.

29. Matto J, Saarela M, von Troil-Linden B, Alaluusua S,Jousimies-Somer H, Asikainen S. Similarity of salivaryand subgingival Prevotella intermedia and Prevotellanigrescens isolates by arbitrarily primed polymerasechain reaction. Oral Microbiol Immunol 1996;11:395-401.

30. Teanpaisan R, Douglas CW, Eley AR, Walsh TF.Clonality of Porphyromonas gingivalis, Prevotella in-termedia and Prevotella nigrescens isolated from peri-odontally diseased and healthy sites. J Periodontal Res1996;31:423-432.

31. Teanpaisan R, Douglas CW, Nittayananta W. Isolationand genotyping of black-pigmented anaerobes fromperiodontal sites of HIV-positive and non-infectedsubjects in Thailand. J Clin Periodontol 2001;28:311-318.

32. Kipioti A, Nakou M, Legakis N, Mitsis F. Microbiolog-ical findings of infected root canals and adjacentperiodontal pockets in teeth with advanced peri-odontitis. Oral Surg Oral Med Oral Pathol 1984;58:213-220.

33. Kobayashi T, Hayashi A, Yoshikawa R, Okuda K,Hara K. The microbial flora from root canals andperiodontal pockets of non-vital teeth associated withadvanced periodontitis. Int Endod J 1990;23:100-106.

Correspondence: Dr. Luciano Jose Pereira, Department ofPhysiology and Pharmacology, Federal University ofLavras, Caixa Postal 3037, CEP 37200-000, Lavras, MG,Brazil. E-mail: lucianojosepereira@dmv.ufla.br.

Submitted January 31, 2011; accepted for publicationMarch 11, 2011.

J Periodontol • December 2011 Pereira, Stipp, Fonseca, Pereira, Hofling

1775