exploiting molecular methods to explore endodontic ...means of their phenotypic features, this...

Exploiting Molecular Methods to Explore EndodonticInfections: Part 1—Current Molecular Technologies forMicrobiological DiagnosisJ. F. Siqueira, Jr, DDS, MSc, PhD, and I. N. Rocas, DDS, MSc, PhD

AbstractEndodontic infections have been traditionally studiedby culture-dependent methods. However, as with otherareas of clinical microbiology, culture-based investiga-tions are plagued by significant problems, including theprobable involvement of viable but uncultivable micro-organisms with disease causation and inaccurate mi-crobial identification. Innumerous molecular technolo-gies have been used for microbiological diagnosis inclinical microbiology, but only recently some of thesetechniques have been applied in endodontic microbiol-ogy research. This paper intended to review the mainmolecular methods that have been used or have thepotential to be used in the study of endodontic infec-tions. Moreover, advantages and limitations of currentmolecular techniques when compared to conventionalmethods for microbial identification are also discussed.

Key WordsApical periodontitis, endodontic microbiology, molecu-lar technology, microbiological diagnosis

From the Department of Endodontics, Estacio de Sa Uni-versity, Rio de Janeiro, Rio de Janeiro, Brazil.

Address requests for reprint to Jose F. Siqueira, Jr, R.Herotides de Oliveira 61 / 601, Icaraı, Niteroi, RJ, Brazil 24230-230. E-mail address: [email protected].

Copyright © 2005 by the American Association ofEndodontists

In essence, endodontic infection is the infection of the root canal of the tooth and is theprimary etiologic agent of the different forms of periradicular inflammatory diseases

(1, 2). The pathological process is started when the dental pulp becomes necrotic,usually as a sequel to caries, and then infected by micro-organisms that are usuallynormal inhabitants of the oral cavity. The root canal containing a necrotic pulp affordsmicro-organisms a moist, warm, nutritious, and anaerobic environment, which is, byand large, protected from host defenses. Such conditions are highly conducive tomicrobial colonization and multiplication. After the endodontic infection is established,micro-organisms enter in close contact with the periradicular tissues via the apicalforamen and occasional foraminas, inflict damage to these tissues, and give rise toinflammatory changes. Periradicular inflammatory diseases are arguably among themost common diseases that affect human beings (3, 4).

Traditionally, endodontic bacteria have been studied by means of cultivation-based techniques, which rely on isolation, growth, and laboratory identification bymorphology and biochemical tests. However, cultivation and other traditional identifi-cation methods have been demonstrated to have several limitations when it comes tomicrobiological diagnosis (5). The past decade has brought many advances in micro-bial molecular diagnostics, the most prolific being in DNA-DNA hybridization as well asin polymerase chain reaction (PCR) technology and its derivatives. Indeed, findingsfrom cultivation-based methods with regard to the microbiota living in diverse ecosys-tems have been supplemented and significantly expanded with molecular biology tech-niques, and the impact of these methods on the knowledge about the oral microbiota inhealthy and diseased conditions is astonishing and breathtaking.

The first part of this paper describes the state of the art of molecular methodologiesalready used or with potential to be applied to the study of the microbiota associatedwith endodontic infections. The significant contribution brought about by moleculartechnology and its impact on the redefinition of endodontic infections are discussed inthe second part of this paper.

Traditional Identification MethodsCulture

For more than a century, cultivation using artificial growth media has been thestandard diagnostic test in infectious diseases. The microbiota associated with differentsites in the human body has been extensively and frequently scrutinized by studies usingcultivation approaches. The success in cultivation of important pathogenic bacteriaprobably led microbiologists to feel satisfied with and optimistic about their results andto recognize that there is no dearth of known pathogens (6, 7). But should we be socomplacent with what we know about human pathogens?

Making micro-organisms grow under laboratory conditions presupposes someknowledge of their growth requirements. Nevertheless, very little is known about thespecific growth factors that are utilized by innumerous micro-organisms to survive invirtually all habitats, including within the human body (8). A huge proportion of themicrobial species in nature are difficult to be tamed in the laboratory. Certain bacteriaare fastidious or even impossible to cultivate. Some well-known human pathogens, suchas Mycobacterium leprae and Treponema pallidum continue to defy scientists regard-ing their cultivation under laboratory conditions (9).

Review

JOE — Volume 31, Number 6, June 2005 Molecular Methods in Endodontic Microbiology: Part 1 411

Of 36 bacterial divisions noted by Hugenholtz et al. (10), 13 divi-sions are exclusively comprised of as-yet uncultivated bacteria. Updatedanalyses have indicated that presently 52 phyla can be discerned, ofwhich 26 are candidate phyla, that is, they are uncultivable and knownonly by gene sequences (11). In fact, there has been a pronounced biastowards the study of representatives of four bacterial phyla, namelyProteobacteria, Firmicutes, Actinobacteria, and Bacteroidetes, outof 52 bacterial phyla (12). This bias is arguably related to the fact thatrepresentatives of these phyla are easier to cultivate with the methodsavailable nowadays. It is glaringly obvious that there is an urgent needfor this bias to be rectified.

Infectious disease professionals have long been aware of the factthat when cultivation is applied to diagnose the causative agents ofdiseases of suspected microbial etiology, including pneumoniae, en-cephalitis, meningitis, pericarditis, acute diarrhea, and sepsis, a largenumber of cases can not be explained microbiologically (13, 14).Moreover, the etiology of a number of chronic inflammatory diseaseswith features suggestive of infection, such as rheumatoid arthritis, sys-temic lupus erythematosis, atherosclerosis, and diabetes mellitus, re-main obscure (15). Taking into consideration that known bacterialpathogens fall within 7 out of the 52 candidate bacterial divisions andthat cultivation-independent approaches have shown that 40 to 75% ofthe human microbiota in different sites are composed of as-yet uncul-tivated bacteria (16 –18), it is fair to realize that there can be manypathogens which remain to be identified. Therefore, it is of concern thatclinical microbiology continues to rely on cultivation-based identifica-tion procedures (6).

Advantages and LimitationsThe main advantages of cultivation approaches are related to their

broad-range nature, which makes it possible to identify a great variety ofmicrobial species in a sample, including those that are not being soughtafter. Still, cultivation makes it possible to determine antimicrobial sus-ceptibilities of the isolates and to study their physiology and pathoge-nicity.

However, cultivation-based identification approaches have severallimitations: they are costly; they can take several days to weeks to identifysome fastidious anaerobic bacteria (that can delay antimicrobial treat-ment); they have a very low sensitivity (particularly for fastidious anaer-obic bacteria); their specificity may be also low and is dependent on theexperience of the microbiologist; they have strict dependence on themode of sample transport; they are time-consuming and laborious.Finally, the impossibility of cultivating a large number of bacterial spe-cies as well as the difficulties in identifying many cultivable speciesrepresent the major drawbacks of cultivation-based approaches.

Difficulties in CultivationThere are many possible reasons for bacterial unculturability (7,

19, 20) (Table 1). Obviously, if micro-organisms can not be cultivated,they can not be identified by phenotype-based methods. While we stayrelatively ignorant on the requirements of many bacteria to grow, iden-

tification methods that are not based on bacterial culturability are re-quired. This would avoid that many pathogens pass unnoticed when oneis microbiologically surveying clinical samples.

It is worth pointing out that the fact that a given species is uncul-tivable does not necessarily imply that the same species will remainindefinitely impossible to cultivate. For instance, a myriad of strict an-aerobic bacteria were uncultivable 100 yr ago, but further develop-ments in cultivation techniques have to a large extent helped solve thisproblem. There is a growing trend to develop approaches and culturemedia that allow cultivation of previously uncultivated bacteria. Strate-gies may rely on application of cultivation procedures that better mimicconditions existing in the natural habitat from which the samples wereobtained. Recent efforts to accomplish this have met with some successby including the following: the use of agar media with little or no addednutrients; relatively lengthy periods of incubation (more than 30 days);and inclusion of substances that are typical of the natural environmentin the artificial growth media (21, 22). However, it is not unreasonableto surmise that a huge number of species will remain uncultivable foryears to come.

Difficulties in IdentificationAccurate identification of microbial isolates is paramount in clin-

ical microbiology. For a given microbial species to be identified bymeans of their phenotypic features, this species has to be cultivated.However, one should be mindful that in some circumstances even thesuccessful cultivation of a given microorganism does not necessarilymean that this micro-organism can be successfully identified.

For slow-growing and fastidious bacteria, traditional phenotypicidentification is difficult and time-consuming. In addition, interpreta-tion of phenotypic test results can involve a substantial amount of sub-jective judgment and personnel’s expertise. Still, one major difficultyassociated with microbial identification based on phenotypic features isthat of divergence and convergence. Divergence occurs for strains of thesame species, which are genetically similar, but have evolved to bedifferent phenotypically. Convergence occurs for strains of differentspecies, which are genetically different, but have evolved to have similarphenotypic behavior (23). In both situations, phenotypically based di-agnostic tests would result in misidentification.

16S rRNA gene (rDNA) sequencing approach was first proposed toidentify uncultivable bacteria (see section on Broad-Range PCR in thisarticle), but it has also been recently used for identification of cultivablebacteria that shows atypical phenotypic behavior and cannot be accu-rately identified by culture. Unlike phenotypic identification, which canbe modified by the variability of expression of characters, 16S rDNAsequencing provides unambiguous data even for rare isolates. Bosshardet al. (24) reported that 14% of all aerobic gram-positive rods isolatedin their clinical microbiology laboratory required 16S rDNA sequence-based identification. Drancourt et al. (25) reported that about 0.5 to 1%of the bacterial isolates recovered in their laboratory were not possibleto be identified by phenotypic criteria. Using 16S rDNA sequence anal-ysis, they reported that several atypical isolates corresponded to 27 new

TABLE 1. Reasons for bacterial unculturability

a) Lack of essential nutrients or growth factors in the artificial culture medium.b) Toxicity of the culture medium itself, which can inhibit bacterial growth.c) Production of substances inhibitory to the target microorganism by other species present in a mixed consortium.d) Metabolic dependence on other species for growth.e) Disruption of bacterial intercommunication systems induced by separation of bacteria on solid culture media.f) Bacterial dormancy, which is a state of low metabolic activity that some bacteria develop under certain stressful environmental

conditions, such as starvation. Dormant bacterial cells can be unable to divide or to form colonies on agar plates without apreceding resuscitation phase.

Data according to references 7, 19, 20.

Review

412 Siqueira and Rocas JOE — Volume 31, Number 6, June 2005

bacterial species associated with human diseases, of which 11 isolateswere previously uncharacterized species and 16 isolates had never beenreported in humans, being previously regarded as environmental. Dran-court et al. (26) evaluated the utility of 16S rDNA sequencing as a meansto identify a collection of 177 isolates obtained from environmental,veterinary, and clinical sources. Conventional identification failed toproduce accurate results because of inappropriate biochemical profiledetermination in 76 isolates (58.7%), Gram staining in 16 isolates(11.6%), oxidase and catalase activity determination in five isolates(3.6%) and growth requirement determination in two isolates (1.5%).They reported that the overall performance of 16S rDNA sequence anal-ysis was excellent, because it succeeded in identifying 90% of pheno-typically unidentifiable bacterial isolates. Song et al. (27) evaluated theutility of 16S rDNA sequencing as a means of identifying clinically im-portant gram-positive anaerobic cocci. Based on the sequences of the13 type strains obtained in their study, 84% (131 of 156) of the clinicalisolates were accurately identified to species level, with the remaining25 clinical strains revealing nine unique sequences that could representeight novel species. This finding was in contrast to the phenotypic iden-tification results, by which only 56% of isolates were correctly identifiedto species level. These studies clearly shown that sequence analysis ofthe 16S rDNA can be used to overcome difficulties in identifying culti-vable species that are notoriously refractory to identification by pheno-typic means.

MicroscopyMicroscopy may suggest an etiologic agent, but it rarely provides

definitive evidence of infection by a particular species. Microscopicfindings regarding bacterial morphology may be misleading, becausemany species can be pleomorphic and conclusions can be influenced bysubjective interpretation of the investigator. In addition, microscopy haslimited sensitivity and specificity to detect micro-organisms in clinicalsamples. Limited sensitivity is because a relatively large number of mi-crobial cells are required before they are seen under microscopy (e.g.104 bacterial cells/ml of fluid) (9). Some micro-organisms can evenrequire appropriate stains and/or approaches to become visible. Lim-ited specificity is because our inability to speciate micro-organismsbased on their morphology and staining patterns.

Immunological MethodsImmunological methods are based on the specificity of antigen-

antibody reaction. It can detect micro-organisms directly or indirectly,the latter by detecting host immunoglobulins specific to the target mi-cro-organism. The enzyme-linked immunosorbent assay (ELISA) andthe direct or indirect immunofluorescence tests are the most commonlyused immunological methods for microbial identification. Advantagesof immunological methods for identification of micro-organisms in-

clude: (a) they take no more than a few hours to identify a microbialspecies; (b) they can detect dead micro-organisms; (c) they can beeasily standardized; and (d) they have low cost (28). However, theyhave also important limitations as they can detect only target species,they have low sensitivity (about 104 cells), their specificity is variableand depends on types of antibodies used, and they can detect deadmicro-organisms (28, 29).

Molecular Genetic MethodsInvestigations of many aquatic and terrestrial environments using

cultivation-independent methods have revealed a great deal of previ-ously unsuspected microbial diversity (30). In fact, the cultivable mem-bers of these living systems represent less than 1% of the total extantpopulation (31, 32). Novel culture-independent methods for microbialidentification that involve DNA amplification of 16S rDNA followed bycloning and sequencing have recently been used to determine the bac-terial diversity within diverse environments, such as deep-sea sediments(33), hot springs (34), and human diseased and healthy sites (16 –18,35–39). Perhaps not surprisingly, the number of recognized bacterialphyla has exploded from the original estimate of 11 in 1987 to 36 in1998 (10). The latest tally of bacterial phyla is now probably near 52, ofwhich one-half has only uncultivable representatives (11).

A significant contribution of molecular methods to medical micro-biology relates to the identification of previously unknown humanpathogens (40 – 44). In addition, molecular studies have revealed thatabout 40 to 50% of the bacterial clones in the oral cavity (17) and about70% of clones in the gut and colon (16, 35) represent unknown andas-yet uncultivable species. As a consequence, it is fair to realize thatthere can exist a number of uncharacterized pathogens in this unculti-vable proportion of the human microbiota.

Molecular diagnostic methods have several advantages over othermethods with regard to microbial identification (Table 2). Limitationsof molecular approaches will be discussed further ahead in this paper.

There are a plethora of molecular methods for the study of micro-organisms and the choice of a particular approach depends on thequestions being addressed (Fig. 1). Broad-range polymerase chain re-action (PCR) followed by sequencing can be used to investigate themicrobial diversity in a given environment. Microbial community struc-ture can be analyzed via fingerprinting techniques, such as denaturinggradient gel electrophoresis (DGGE) and terminal restriction fragmentlength polymorphism (T-RFLP). Fluorescence in situ hybridization(FISH) can measure abundance of particular species and provide in-formation on their spatial distribution in tissues. Among other applica-tions, DNA-DNA hybridization macroarrays and microarrays, species-specific PCR, nested PCR, multiplex PCR, and real-time PCR can be usedto survey a large number of clinical samples for the presence of target

TABLE 2. Advantages of molecular genetic methods over other methods for microbial identification

Advantages of Molecular Genetic Methodsa) Detection of not only cultivable species but also of uncultivable microbial species or strains.b) Higher specificity and accurate identification of microbial strains with ambiguous phenotypic behavior, including divergent or

convergent strains.c) Detection of microbial species directly in clinical samples, without the need for cultivation.d) Higher sensitivity.e) Faster and less time-consuming.f) They offer a rapid diagnosis, which is particularly advantageous in cases of life-threatening diseases or diseases caused by slow-

growing micro-organisms.g) They do not require carefully controlled anaerobic conditions during sampling and transportation, which is advantageous since

fastidious anaerobic bacteria and other fragile micro-organisms can lose viability during transit.h) They can be used during antimicrobial treatment.i) When a large number of samples are to be surveyed in epidemiological studies, samples can be stored and analyzed all at once.

Review


species. Variations in PCR technology can also be used to type microbialstrains. This review will restrict discussion to the most commonly usedapproaches applied to the research of the endodontic microbiota andsome with potential to be used with this intent.

Gene Targets for Microbial IdentificationMolecular approaches for microbial identification rely on the fact

that certain genes contain revealing information about the microbialidentity. Ideally, a gene to be used as target for microbial identificationshould contain regions that are unique to each species. Following thepioneer studies by Woese (45), the genes encoding rRNA molecules,which are present in all cellular forms of life, namely, the domainsBacteria, Archaea, and Eucarya, have been widely used for compre-hensive identification of virtually all living organisms and inference oftheir natural relationships.

Ribosomes are intracellular particles composed of rRNA and pro-teins. In Escherichia coli, about two-thirds of the ribosome is rRNA andthe remainder is protein (46). The sizes of ribosomes are given inSvedburg (S) units, which represent a measure of how rapidly particlesor molecules sediment in an ultracentrifuge. Bacterial and archaealcells have 70S ribosomes composed of 30S and 50S subunits. The 30Ssubunit contains a 16S rRNA molecule, having approximately 1540 nu-cleotides. The 50S subunit contains a 23S rRNA molecule, having ap-

proximately 2900 nucleotides, and a small 5S rRNA molecule havingonly about 120 nucleotides. Fungi have 80S ribosomes composed of 40Sand 60S subunits. The 40S subunit contains 18S rRNA and the larger 60Ssubunit has 25S rRNA and 5.8S rRNA (47).

Large subunit genes (23S and 25S rDNA) and small subunit genes(16S and 18S rDNA) have been widely used for microbial identification,characterization and classification. The small subunit rDNA is amongthe most evolutionary conserved macromolecules in all living systems(45). The advantages of using small subunit rDNA is that it is found in allorganisms, is long enough to be highly informative and short enough tobe easily sequenced, particularly with the advent of automated DNAsequencers (45). The small subunit rDNA contains some regions thatare virtually identical in all representative of a given domain (conservedregions) and other regions that vary in sequence from one species toanother (variable regions) (Fig. 2). Variable regions contain the most in-formation about the genus and species of the bacterium, with unique signa-tures that allow specific identification. The 16S rRNA of bacteria and archaeaand the 18S rDNA of fungi and other eukaryotes have been extensivelyexamined and sequenced and have been used to determine phylogeneticrelationships among living organisms. In addition, data from rDNA se-quences can also be used for accurate and rapid identification of knownbacterial species, using techniques that do not require microbial cultivation.

Figure 1. Examples of molecular techniques used or with potential to be used in the study of endodontic infections. The choice for a particular method will dependon the question being addressed.

Review


There are now more than 90,000 different bacterial 16S rDNAsequences in public databases, while the number of 23S rDNA se-quences is still considerably smaller (�1400), though it keeps increas-ing (48). Consequently, 16S rDNA is the most useful target for bacterialidentification by molecular approaches, and 23S rDNA is becoming asuitable alternative. Recently, the intergenic, noncoding spacer se-quences have also been used, because they are more likely to be vari-able than the rDNAs. The ones that have been used include the 16S-23Sintergenic spacer for bacteria (49) and the internal transcribed spacerfor fungi (50). These regions are useful for differentiating betweenclosely related species because of heterogeneity of its length and se-quence but are not adequate to establish phylogenetic relationships(51).

PCRThe PCR process was conceived by Kary Mullis in 1983 and ever

since has revolutionized the field of molecular biology by being able toamplify as few as one copy of a gene into millions to billions of copies ofthat gene (52). The impact of PCR on biological and medical researchhas been remarkable, dramatically speeding the rate of progress of thestudy of genes and genomes. Nowadays, it is possible to isolate essen-tially any gene from any organism using PCR, which made this techniquea cornerstone of genome sequencing projects (53). Since its introduc-tion, PCR has spawned an increasing number of associated technologiesfor diverse applications. Perhaps the most widespread advance in clin-ical diagnostic technology has come from the application of PCR fordetection of microbial pathogens (54 –56).

The PCR method is based on the in vitro replication of DNA throughrepetitive cycles of denaturation, primer annealing and extension steps(Fig. 3). The target DNA serving as template melts at temperatures highenough to break the hydrogen bonds holding the strands together, thusliberating single strands of DNA. Two short oligonucleotides (primers)are annealed to complementary sequences on opposite strands of thetarget DNA. Primers are selected to encompass the desired geneticmaterial, defining the two ends of the amplified stretch of DNA. Insequence, a complementary second strand of new DNA is synthesizedthrough the extension of each annealed primer by a thermostable DNApolymerase in the presence of excess deoxyribonucleoside triphos-phates. All previously synthesized products act as templates for newprimer-extension reactions in each ensuing cycle. The result is theexponential amplification of new DNA products, which confers extraor-dinary sensitivity in detecting the target DNA. In fact, PCR has unrivaledsensitivity—it is at least 10 to 100 times more sensitive than the othermore sensitive identification method (29, 57).

There are several methods to check if the intended PCR productwas generated. The most commonly used method for detecting PCRproducts is electrophoresis in an agarose gel. Aliquots of the PCR reac-

tion are loaded into the gel and an electrical gradient is applied througha buffer solution. The products migrate through the gel according tosize, with larger products running a shorter distance in the gel becausethey experience more resistance in the gel matrix. DNA ladder digestsrepresent DNA fragments of known size and are run in the same gel toserve as molecular size standard. This allows the size of the PCR prod-ucts to be estimated. The gel is usually visualized using ethidium bro-mide staining and ultraviolet transillumination. Designed primers areexpected to generate a PCR product of a given size and the observationof a PCR amplicon of the predicted size in the electrophoretic gel isconsistent with a positive PCR result. Identity of PCR products can still beconfirmed by sequencing of the PCR product, hybridization of a specificoligonucleotide probe to a region of the PCR product that is internal tothe priming sites, or restriction enzyme cleavage of the PCR productusing an enzyme that is known to cut a specific sequence within theproduct [restriction fragment length polymorphism (RFLP)].

Numerous derivatives in PCR technology have been developedsince its inception. The most used PCR-derived assays are described inthe following sections.

Nested PCRNested PCR uses the product of a primary PCR amplification as

template in a second PCR reaction and was devised mainly to haveincreased sensitivity (58). The first PCR products are subjected to asecond round of PCR amplification with a separate primer set, whichanneals internally to the first products. This approach shows increasedsensitivity allowing the detection of the target DNA several folds lowerthan conventional PCR. Increased sensitivity is because of the large totalnumber of cycles. In addition, target DNA is amplified in the first roundof amplification, with subsequent dilution of other DNA and inhibitorspresent in the sample. The set of primers used in the second round ofPCR results in additional specificity. The second reaction is performedwith reduced background of eukaryotic DNA and other regions of thebacterial DNA (57). Even if nonspecific DNA amplification occurs in thefirst round of amplification, the nonspecific PCR product does not serveas template in the second reaction, since it is highly unlikely to possessregions of DNA complementary to the second set of specific primers(59).

The major drawback of nested-PCR protocol is the high probabilityof contamination during transfer of the first-round amplification prod-ucts to a second reaction tube and special precautions should be takento avoid this.

Reverse Transcriptase PCR (RT-PCR)RT-PCR was developed to amplify RNA targets and exploits the use

of the enzyme reverse transcriptase, which can synthesize a strand ofcomplementary DNA (cDNA) from an RNA template. Most RT-PCR as-says employ a two-step approach. In the first step, reverse transcriptaseconverts RNA into single-stranded cDNA. In the second step, PCR prim-ers, DNA polymerase, and nucleotides are added to create the secondstrand of cDNA. Once the double-stranded DNA template is formed, itcan be used as template for amplification as in conventional PCR (60).The RT-PCR process may be modified into a one-step approach by usingit directly with RNA as the template. In this approach, an enzyme withboth reverse transcriptase and DNA polymerase activities is used, suchas that from the bacteria Thermus thermophilus.

Multiplex PCRMost PCR assays have concentrated on the detection of a single

microbial species by means of individual reactions. In multiplex PCR,two or more sets of primers specific for different targets are introducedin the same reaction tube. Since more than one unique target sequence

Figure 2. Schematic drawing of the 16S rRNA gene (rDNA). Orange areas cor-respond to variable regions, which contain information about the genus and thespecies. Red areas correspond to conserved regions of the gene.

Review


in a clinical specimen can be amplified at the same time, multiplex PCRassays permit the simultaneous detection of different microbial species.Multiplex PCR assays have been used to minimize the time and expen-diture needed for detection approaches. Primers used in multiplex as-says must be designed carefully to have similar annealing temperaturesand to lack complementarity (61, 62).

PCR-Based Microbial TypingPCR technology can also be used for clonal analysis of micro-

organisms. An example of the PCR techniques used for this purposeincludes the arbitrarily primed PCR (AP-PCR), also referred to as ran-dom amplified polymorphic DNA (RAPD) (63). AP-PCR is a relativelyrapid PCR-based genomic fingerprinting technique that can be used asa tool to determine whether two isolates of the same species are epide-miologically related. The method utilizes a single 10- to 20-base ran-dom sequence primer that anneals to unspecified DNA target sites underconditions that allow for mismatched base-pairing. The use of a ran-dom-sequence primer at low stringency allows for priming at sites withimperfect matches. Genetic variations between two DNA templates re-sult in discriminative DNA fingerprints because of the differences in thepriming sites. The amplicons generated form a pattern in the electro-phoretic gel that may be strain specific. The advantage of AP-PCR is itsability to furnish highly specific DNA profiles with no prerequisite forknowing the DNA sequences. Primers may also be designed to targetknown genetic elements, such as enterobacterial repetitive intergenicconsensus sequences (ERIC-PCR) (64, 65) and repetitive extragenicpalindromic sequences (REP-PCR) (66). Clonal analysis may help elu-cidate whether certain strains of a given species are more associatedwith signs or symptoms of a given disease. Clonal analysis also may helpto track the origin of micro-organisms infecting a given site. For in-stance, by comparing bacterial strains isolated from the root canal andother oral sites, one can have information as to where micro-organismspresent in the root canal system came from. Clonal analysis can alsotrack the origin of the micro-organisms present in a suspected focaldisease by comparing the isolates found in the infected site with others

present in the suspected focus of infection, including the oral cavity(and infected root canals).

Real-Time PCRConventional PCR assays are qualitative or can be adjusted to be

semi-quantitative. One exception is the real-time PCR, which is charac-terized by the continuous measurement of amplification productsthroughout the reaction (67).

There are several different real-time PCR approaches. The threegeneral real-time PCR chemistries for amplifying and detecting DNAtargets are SYBR-Green, TaqMan, and molecular beacon (68, 69).SYBR Green is the simplest and most affordable method, and consists ofa fluorescent dye that binds to double-stranded DNA. During extension,increasing amounts of dye bind to the increasing amount of newlyformed double-stranded DNA. Fluorescence is measured at the end ofthe extension step of every PCR cycle to monitor the increasing amountof amplified DNA. Dye that remains unbound exhibits little fluorescencein solution. The SYBR-Green assay is very sensitive but has diminishedspecificity, as the dye binds to all double-stranded DNA present, andprimer dimers can result in a false reading (70).

The TaqMan method can be more specific than the SYBR-Greenassay. Increased specificity of the TaqMan assay results from the utili-zation of a specific labeled oligonucleotide probe along with the primers(71, 72). The TaqMan probe is a 20- to 30-base-long oligonucleotidesequence that specifically anneals to a sequence flanked by the twoprimers. The TaqMan probe contains a reporter fluorescent dye at the5� end and a quencher dye at the 3� end that quenches the emissionspectrum of the reporter dye (Fig. 4). As long as the probe remainsunbound, it is intact and no signal is generated. During the extensionstep of real-time PCR, the Taq DNA polymerase enzyme cleaves theTaqMan probe, resulting in separation of the reporter from thequencher (Fig. 4). This results in increased fluorescence emission.

Another real-time PCR assay uses molecular beacons, which aresingle-stranded nucleic acid molecules with a stem-and-loop structure(69). The loop portion is complementary to a sequence in the target

Figure 3. Scheme for PCR.

Review


DNA. The stem is formed by the annealing of complementary arm se-quences located on either side of the probe. A fluorescent marker isattached to the end of one arm, and a quencher is attached to the end ofthe other arm. Free molecular beacons acquire a hairpin structurewhen in solution, and the stem keeps the arm sequences in close prox-imity. This results in efficient quenching of the fluorescent dye (Fig. 4).When hybridizing to their complementary target, molecular beacons areforced to undergo a conformational change, with consequent formationof probe that hybridizes to the template. The conformational changeforces the fluorescent dye and the quencher apart, generating fluores-cence (73) (Fig. 4).

By monitoring the release of fluorescence with each PCR cycle, theprogress of the reaction is recorded in real time and the amount of DNAin the sample is quantified. Thus, accumulation of PCR products ismeasured automatically during each cycle in a closed tube format usinga thermocycler combined with a fluorimeter. Direct measurement of theaccumulated PCR product allows the phases of the reaction to be mon-itored. The initial amount of target DNA in the reaction can be related toa cycle threshold defined as the cycle number at which there is a statis-tically significant increase in fluorescence. Target DNA can then bequantified by construction of a calibration curve that relates cyclethreshold to known amounts of template DNA (74).

Real-Time PCR technology is commercially available as the Taq-Man (Applied Biosystems, Perkin-Elmer Corp.) and LightCycler (RocheDiagnostics Corp.) systems. Real-time PCR assays allow the quantifica-tion of individual target species as well as total bacteria in clinicalsamples. The advantages of real-time PCR are the rapidity of the assay(30 – 40 min), the ability to quantify and identify PCR products directlywithout the use of agarose gels, and the fact that contamination of thenucleic acids can be limited because of avoidance of postamplificationmanipulation (75).

Broad-Range PCRPCR technology can also be used to investigate the whole microbial

diversity in a given environment. In broad-range PCR, primers are de-signed that are complementary to conserved regions of a particulargene that are shared by a group of micro-organisms. For instance,primers that are complementary to conserved regions of the 16S rDNAhave been used with the intention of exploiting the variable internalregions of the amplified sequence for sequencing and further identifi-cation (76). The strength of broad-range PCR lies in the relative absenceof selectivity, so that (in principle) any kind of bacteria present in asample can be detected and identified. This aspect is in analogy tocultivation and in contrast to species-specific molecular approaches(48). Thus, broad-range PCR can detect the unexpected and in thisregard it is far more effective and accurate than culture. Broad-rangePCR has allowed the identification of several novel, fastidious or uncul-tivable bacterial pathogens directly from diverse human sites (5, 7, 77).

Initially, bulk nucleic acids are extracted directly from samples.Afterwards, the 16S rDNA is isolated from the bulk DNA via PCR witholigonucleotide primers specific for conserved regions of the gene (uni-versal or broad-range primers). Longer distances between primer pairsgenerally result in less sensitivity, but this provides more variable se-quence for more accurate identification and may reduce the risks ofamplifying DNA from contaminants in PCR reagents, which appear to befragmented into smaller sizes (48). Amplification with universal prim-ers results in a mixture of all of the 16S rDNA amplified from the bacteriain the sample. In mixed infections, direct sequencing of the PCR prod-ucts cannot be performed because there are mixed products from thedifferent species composing the consortium. PCR products are thencloned into a plasmid vector, which is used to transform Escherichiacoli cells, establishing a library of 16S rDNA from the sample. Cloned

Figure 4. Probes for real-time PCR. (A) TaqMan probes. The probe in solution folds and bring the reporter fluorescent dye close enough to the quencher dye toquench the fluorescence. The probe bound to the PCR product is digested by the 5� nuclease activity of Taq polymerase during the extension step of the cycle. (B)Molecular beacons. Hybridization to its target induces a conformational change in the molecular beacon probe that forces the arm sequences apart and causes thefluorophore to move away from the quencher.

Review


genes are then sequenced individually and submitted for identificationto databases, usually via the World Wide Web. Preliminary identificationcan be done by using similarity searches in public databases. A �99%identity in 16S rDNA sequence has been the most accepted criterionused to identify a bacterium to the species level (25, 26). A 97 to 99%identity in 16S rDNA sequence has been the criterion used to identify abacterium at the genus level, and �97% identity in 16S rDNA genesequence has been the criterion used to define a potentially new bacte-rial species (25, 26). In addition to the similarity search, phylogeneticanalysis should be accomplished since it provides a much more accu-rate assessment (48, 51).

The analytical sensitivity of most broad-range PCR assays is inpractice above 10, if not 100, gene copies per PCR, which is significantlylower when compared to most species-specific PCR assays (48). Be-cause broad-range primers are used, there is a high risk for DNA frommicrobial contaminants to be amplified. A wide range of precautions isnecessary to avoid contamination, including separate room for pre- andpost-PCR work, UV decontamination of surface areas, use of high-qual-ity reagents, and adequate sampling techniques and vials for clinicalspecimens (48, 59, 78).

Denaturing Gradient Gel ElectrophoresisTechniques for genetic fingerprinting of microbial communities

can be used to determine the diversity of different micro-organismsliving in diverse ecosystems and to monitor microbial community be-havior over time. A commonly used strategy for genetic fingerprinting ofcomplex microbial communities encompasses the extraction of DNA,the amplification of the 16S rDNA using broad-range primers, and thenthe analysis of PCR products by denaturing gradient gel electrophoresis(DGGE).

In DGGE, DNA fragments of the same length but with differentbase-pair sequences can be separated (79, 80). The DGGE technique isbased on electrophoresis of PCR-amplified 16S rDNA (or other genes)fragments in polyacrylamide gels containing a linearly increasing gra-dient of DNA denaturants (a mixture of urea and formamide). As thePCR product migrates in the gel, it encounters increasing concentra-tions of denaturants and, at some position in the gel, it will becomepartially or fully denatured. Partial denaturation causes a significantdecrease in the electrophoretic mobility of the DNA molecule. Mole-cules with different sequences may have a different melting behaviorand will therefore stop migrating at different positions in the gel. Theposition in the gel at which the DNA melts is determined by its nucleotidesequence and composition (81). A GC-rich sequence (GC-clamp) isadded to the 5�-end of one of the PCR primers, which ensures that atleast part of the DNA remains double stranded (82). DNA bands inDGGE can be visualized using ethidium bromide, SYBR Green I, or silverstaining.

In DGGE, multiple samples can be analyzed concurrently, makingit possible to compare the structure of the microbial community ofdifferent samples and to follow changes in microbial populations overtime, including after antimicrobial treatment. Specific bands can also beexcised from the gels, re-amplified and sequenced to allow microbialidentification. The DGGE method and its application in endodontic mi-crobiology research have been recently reviewed (83). Temperaturegradient gel electrophoresis (TGGE) uses the same principle as DGGE,except for the fact that the gradient is temperature rather than chemicaldenaturants.

Terminal-RFLPT-RFLP is a recent molecular approach that can assess subtle genetic

differences between microbial strains as well as provide insight into thestructure and function of microbial communities (84). T-RFLP analysis

measures the size polymorphism of terminal restriction fragments from aPCR amplified marker. T-RFLP is a modification of the conventional RFLPapproach. In T-RFLP, rDNA from different species in a community is PCR-amplified using one of the PCR primers labeled with a fluorescent dye, suchas 4,7,2�,7�-tetrachloro-6-carboxyfluorescein (TET) or phosphoramiditefluorochrome 5-carboxyfluorescein (6-FAM) (85). PCR products are thendigested with restriction enzymes, generating different fragment lengths.Digestion of PCR products with judiciously selected restriction endonucle-ases produces terminal fragments appropriate for sizing on high resolutionsequencing gels. The latter step is performed on automated systemssuch as the ABI gel or capillary electrophoresis systems that providedigital output (85).

The use of a fluorescently labeled primer limits the analysis to onlythe terminal fragments of the enzymatic digestion. This simplifies thebanding pattern, thus allowing the analysis of complex communities aswell as providing information on diversity as each visible band repre-sents a single species. All terminal fragment sizes generated from diges-tion of a PCR product pool can be compared with the terminal fragmentsderived from sequence databases to derive phylogenetic inference.Through application of automated DNA sequencer technology, T-RFLPhas considerably greater resolution than gel-based community profilingtechniques, such as DGGE/TGGE (84, 85).

DNA-DNA HybridizationDNA-DNA hybridization methodology employs DNA probes, which

entail segments of single-stranded DNA, labeled with an enzyme, radio-active isotope or a chemiluminescence reporter, that can locate andbind to their complementary nucleic acid sequences. DNA-DNA hybrid-ization arrays on macroscopic matrices, such as nylon membranes,have been often referred to as “macroarrays.” DNA probe may targetwhole genomic DNA or individual genes. Whole genomic probes aremore likely to cross-react with nontarget micro-organisms because ofthe presence of homologous sequences between different microbialspecies. Oligonucleotide probes based on signature sequences of spe-cific genes may display limited or no cross-reactivity with nontargetmicro-organisms when under optimized conditions. In addition, oligo-nucleotide probes can differentiate between closely related species oreven subspecies and can be designed to detect uncultivable bacteria.

Socransky et al. (86) introduced a method for hybridizing largenumbers of DNA samples against large numbers of digoxigenin-labeledwhole genomic DNA or 16S rDNA-based oligonucleotide probes on asingle support membrane—the checkerboard DNA-DNA hybridizationmethod. Denatured DNA from clinical samples is placed in lanes on anylon membrane using a Minislot device. After fixation of the samples tothe membrane, the membrane is placed in a Miniblotter 45 with thelanes of samples at 90° to the lanes of the device. Digoxigenin-labeledwhole genomic DNA probes are then loaded in individual lanes ofthe Miniblotter. After hybridization, the membranes are washed athigh stringency and the DNA probes detected using antibody todigoxigenin conjugated with alkaline phosphatase and chemilumi-nescence detection.

The checkerboard method permits the simultaneous determina-tion of the presence of a multitude of bacterial species in single ormultiple clinical samples. Thus, it is particularly applicable in large-scale epidemiological research. In addition to the reported advantagesof molecular methods, DNA-DNA hybridization technology has the ad-ditional feature that microbial contaminants are not cultivated, nor istheir DNA amplified (87). Because the numbers of contaminating mi-cro-organisms are not increased, one may assume that, if present, theywould be in numbers below detection limits of the checkerboard DNA-DNA hybridization method, which have been reported to be by the orderof 103 to 104 cells (86, 88).

Review


A modification of the checkerboard method was proposed by Pas-ter et al. (89) and consists of a PCR based, reverse-capture checker-board hybridization methodology. The procedure circumvents the needfor in vitro bacterial cultivation, necessary for preparation of wholegenomic probes. Up to 30 reverse-capture oligonucleotide probes thattarget regions of the 16S rDNA are deposited on a nylon membrane inseparate horizontal lanes using a Minislot apparatus. Each probe issynthesized with a polythymidine tail, which is cross-linked to the mem-brane support via UV irradiation, leaving the probe available for hybrid-ization. The 16S rDNA from clinical samples are PCR amplified using adigoxigenin labeled primer. Hybridizations are performed in verticalchannels in a Miniblotter apparatus with digoxigenin-labeled PCR am-plicons from up to 45 samples. Hybridization signals are detected usingchemifluorescence procedures.

DNA MicroarraysDeveloping a high-density macroarray is a difficult task, as large

membranes represent a practical limitation when it comes to the vol-ume of probes that must be used, and particularly when small amountsof samples are available. In addition, because of the slow hybridizationkinetics involved, they usually require longer periods for analysis. Min-iaturization of these assays saves time while cutting costs in diagnosticapplications and research (90). Low volumes reduce reagent consump-tion and increase sample concentration, thus improving reaction kinet-ics. The use of nonporous supports allows faster hybridization andeasier washing steps. As a consequence, hundreds to thousands of re-sults can be obtained in the time needed for a single macroarray exper-iment.

DNA microarray methods were first described in 1995 (91) andessentially consist of many probes that are discretely located on a non-porous solid support, such as a glass slide (90, 92). Printed arrays andhigh-density oligonucleotide arrays are the most commonly used typesof microarrays. A printed array comprises probes that are spotted ontospecific locations on the solid support platform and range in size from500 to 2,000 base pairs in length. Probes can be DNA fragments such aslibrary clones or PCR products. By means of a robotic arrayer andcapillary printing tips, thousands of elements can be printed on a singlemicroscope slide. A high-density oligonucleotide microarray has a largenumber of probes attached to a solid support substrate. In this method,arrays are constructed by synthesizing single-stranded oligonucleotidesin situ by use of photolithographic techniques. These probes are typi-cally shorter gene sequences comprised of 15 to 30 nucleotides (93).Advantages of the former method include relatively low cost and sub-stantial flexibility. Moreover, sequence information is not needed toprint a probe. Advantages of the latter method include higher densityand elimination of the need to obtain and store cloned DNA or PCRproducts (94).

Once microarrays have been printed, targets are prepared forhybridization. Targets may be PCR products, genomic DNA, total RNA,cDNA and so on, and usually incorporate either a fluorescent label orsome other moiety, such as biotin, that permits subsequent detectionwith a secondary label. Targets are applied to the array and those thathybridize to complementary probes are detected using some type ofreporter molecule. Following hybridization, arrays are imaged using ahigh-resolution scanner. Spatial resolution of these systems can be aslow as of 3 to 5 �m, although 10 �m of resolution is sufficient. Sophis-ticated computer software programs can then be used to analyze theresults.

DNA microarrays can be used to enhance PCR product detectionand identification. When PCR is used to amplify microbial DNA fromclinical specimens, microarrays can then be used to identify the PCRproducts by hybridization to an array that is composed of species-

specific probes. Using broad-range primers, such as those that amplifythe 16S rDNA, a single PCR can be used to detect hundreds of bacterialspecies simultaneously. When coupled to PCR, microarrays have detec-tion sensitivity similar to conventional molecular methods with theadded ability to discriminate several species at a time.

The advent of DNA microarrays has already given a marked con-tribution to many fields, including cellular physiology (95, 96), cancerbiology (97–99), and pharmacology (100). In clinical microbiology, inaddition to being applied to microbial detection and identification usingrDNA information, microarrays have been used to analyze the geneticpolymorphisms of specific loci associated with resistance to antimicro-bial agents, to explore the distribution of genes among isolates from thesame and similar species, to analyze quorum-sensing systems, to un-derstand the evolutionary relationship between closely related speciesand to study host-pathogen interactions, particularly by identifyinggenes from pathogens that may be involved in pathogenicity and bysurveying the host response to infection (101–107).

Fluorescence In Situ HybridizationFluorescence in situ hybridization (FISH) with rRNA-targeted oli-

gonucleotide probes has been developed for in situ identification ofindividual microbial cells (108). This technique detects nucleic acidsequences by a fluorescently labeled probe that hybridizes specificallyto its complementary target sequence within the intact cell. In additionto provide identification, FISH gives information about presence, mor-phology, number, organization and spatial distribution of micro-organ-isms (109). FISH not only allows the detection of cultivable microbialspecies, but also of as-yet uncultivable micro-organisms (110).

rRNAs are the main target molecules for FISH. This is because theycan be found in all living organisms; they are relatively stable and occurin high copy numbers (usually several thousands per cell); and theyinclude both variable and highly conserved sequence domains (45).Signature sequences unique to a given group of micro-organisms, rang-ing from whole phyla to individual species, can be identified. In the vastmajority of applications for identification of bacteria or archaea, FISHprobes target 16S rRNA (109). The oligonucleotide probes used in FISHare generally between 15 and 30 nucleotides long and covalently linkedat the 5�-end to a single fluorescent dye molecule. Common fluorophorsinclude fluorescein, tetramethylrhodamine, Texas red and carbocya-nine dyes like Cy3 or Cy5 (108, 109).

A typical FISH protocol includes four steps: the fixation and per-meabilization of the sample; hybridization with the respective probes fordetecting the respective target sequences; washing steps to remove un-bound probe; and the detection of labeled cells by microscopy or flowcytometry (111).

Limitations of Molecular MethodsMolecular techniques have been used to overcome the limitations

of cultivation procedures. Nonetheless, as with all methods, they are notwithout their own limitations. The following discussion verses on themajor shortcomings associated with the most used molecular tech-niques.

DNA-DNA HybridizationAs with most molecular assays, DNA-DNA hybridization assays,

such as the checkerboard technique, have been used only to detecttarget microbial species and consequently fail to detect the unexpected.For whole genomic probe preparation, bacteria must be initially cul-tured. Consequently, DNA-DNA hybridization assays using whole ge-nome probes can only detect cultivable species. In addition, wholegenomic probes used in the DNA-DNA hybridization assays may showcross-reactivity with nontarget micro-organisms because of the pres-

Review


ence of homologous sequences between different bacterial species. Thiscan give rise to false positive results. However, Socransky et al. (86, 88)reported that most probes tested in the checkerboard assay did notexhibit cross-reactions under the experimental conditions. Someprobes, as those to Tannerella forsythia, Porphyromonas gingivalis,and Treponema denticola, did not exhibit cross-reactions with anyheterologous species tested. When cross-reactions were observed theywere always within genera and were quite limited (88).

Reverse-capture checkerboard assay using oligonucleotideprobes may show higher specificity than the conventional checkerboardtechnique and probes can be devised to target and detect uncultivablespecies (89).

PCR-Derived TechniquesTable 3 lists the main limitations of PCR technology. Many of them

can be gotten around by variations in technology. The issues related tothe ability of PCR to detect either an extremely low number of cells ordead cells are of special interest when one interprets the results of PCRidentification procedures in endodontic microbiology research. There-fore, these issues are worth a separate detailed discussion.

The “Too-High Sensitivity” IssueThe high detection rate of PCR may be a reason of concern, spe-

cifically when nonquantitative assays are employed. It has been com-monly claimed that because PCR can detect a very low number of cellsof a given microbial species, the results obtained by this method mayhave no significance with regard to disease causation. Nevertheless,some other factors should be also addressed in this discussion, whichrepresent great advantages of such highly sensitive identification proce-dure in endodontic research.

When taking samples from endodontic infections, difficultiesposed by the physical constraints of the root canal system and by thelimitations of the sampling techniques can make it difficult the attain-ment of a representative sample from the main canal (116). If cells of agiven species are sampled in a number below of the detection rate of thediagnostic test, species prevalence will be underestimated.

It is also important to take into consideration the analytical sensi-tivity needed for the specific clinical sample. For example, a sensitivity ofabout 104 microbial cells per ml is required for urine, while a sensitivityof one cell may be of extreme relevance for samples from blood orcerebrospinal fluid (117). There is no clear evidence as to the micro-bial load necessary for a periradicular lesion to be induced. Endodonticinfections are characterized by a mixed infection and individual speciesmay play different roles in the consortium or dominate various stages ofthe infection. At least theoretically, all bacterial species established inthe infected root canal have the potential to be considered endodontic

pathogens (118). Based on this, it would be glaringly prudent to use themethod with the highest sensitivity to detect all species colonizing theroot canal.

In fact, detection of very low numbers of cells in clinical samplesmay not be as common as anticipated. There are numerous factors thatcan influence PCR reactions, sometimes dramatically reducing sensitiv-ity for direct microbial detection in clinical samples. Thus, the analyticalsensitivity of the method does not always correspond to its “clinical”sensitivity. It is well known that the effects of inhibitors are magnified insamples with low number of target DNA and therefore can significantlydecrease the sensitivity of the method (62). Another impediment refersto the aliquots of the whole sample used in PCR reactions. Most of thePCR assays that have been used so far in endodontic research have ananalytical sensitivity of about 10 to 100 cells. If one takes into accountthe dilution factor dictated by the use of small aliquots (5–10%) of thewhole sample in each amplification reaction, the actual sensitivity of theassay is 200 to 400 to 2000 to 4000 cells in the whole sample, withoutdiscounting the effects of inhibitors (57). These numbers are still lowerthan the detection limits of other methods, but can represent moresignificance with regard to etiology.

Therefore, the use of highly sensitive techniques is welcome in thestudy of endodontic infections, decreasing the risks for potentially im-portant species to pass unnoticed during sample analysis. Althoughqualitative results do not lack significance, the use of quantitative mo-lecular assays, like the real-time PCR, can allow inference of the role ofa given species in the infectious process while maintaining high sensi-tivity and the ability to detect fastidious or uncultivable microbial spe-cies.

The “Dead-Cell” IssueDetection of dead cells by a given identification method can be

regarded as a double-edged sword. This can be at the same time anadvantage and a limitation of the method. On the one hand, this abilitycan allow detection of uncultivable bacteria or fastidious bacteria thatcan die during sampling, transportation or isolation procedures (57,119, 120). On the other hand, if the bacteria were already dead in theinfected site, they may also be detected and this might give rise to a falseassumption about their role in the infectious process (121, 122).

Several studies have addressed this issue. In a mouse model ofLyme disease, it was revealed by PCR that Borrelia burgdorferi DNAdisappeared from tissues immediately after a 5-day antibiotic treatmentcourse, and that the PCR results correlated well with culture results(123). Similarly, investigators using a chinchilla model of otitis mediafound a strong correlation between PCR results and culture results whenanimals were injected with viable Haemophilus influenza (124).

TABLE 3. Limitations of PCR-derived technologies

a) Most PCR assays used for identification purposes qualitatively detect the target microorganism but not its levels in the sample.Quantitative results can however be obtained in real-time PCR assays.

b) Most PCR assays only detect one species or a few different species (multiplex PCR) at a time. However, broad-range PCRanalysis can provide information about the identity of virtually all species in a community.

c) Like DNA-DNA hybridization, most PCR assays only detect target species and consequently fail to detect unexpected species.This can be overcome by broad-range PCR assays.

d) In addition to being laborious and costly, broad-range PCR analyses can be affected by some factors, such as biases inhomogenization procedures (76), preferential DNA amplification (112, 113) and differential DNA extraction (114, 115).

e) Microorganisms with thick cell walls, such as fungi, may be difficult to break open and may require additional steps for lysisand consequent DNA release to occur.

f) False positive results have the potential to occur because of PCR amplification of contaminant DNA. The most importantmeans of contamination is through carryover of amplification product and special precautions should be taken to avoid this.

g) False negatives may occur because of enzyme inhibitors or nucleases present in clinical samples, which may abort theamplification reaction and degrade the DNA template, respectively. Analysis of small sample volumes may also lead to falsenegative results, particularly if the target species is present in low numbers.

Review


When animals were injected with purified bacterial DNA or with killedbacteria, the amplifiable DNA rapidly disappeared. The authors con-cluded that DNA, and DNA from intact but nonviable bacteria, does notpersist in an amplifiable form for more than a day in the presence of aneffusion. In a rabbit model of syphilis in which live or heat-killed Trepo-nema pallidum was injected into the skin and testes, heat-killed trepo-nemes were no longer detectable by PCR after 15 to 30 days, whereasviable micro-organisms were detected by PCR for months (125). Astudy of pulmonary tuberculosis treatment showed that sputum smearsand cultures for Mycobacterium tuberculosis convert to negative be-fore PCR results, but that PCR results do correlate with clinical responseto antibiotics and also can predict relapse (126). It was not clear if thepersistent PCR signal seen in some of these patients was a result of lowlevels of viable micro-organisms, or because of amplifiable DNA fromnonviable micro-organisms. All these studies show that bacterial DNAcan be cleared from the host sites after bacterial death and that DNAfrom different species may differ as to the elimination kinetics at differ-ent body sites. It remains to be clarified for how long bacterial DNA fromdead cells can remain detectable in the infected root canal system.

Detection of microbial DNA sequences in clinical specimens doesnot indicate viability of the micro-organism. However, this issue shouldbe addressed with common sense and without any sort of biases. Thefact that some micro-organisms die during the course of an infectiousprocess does not necessarily implicate that in a determined momentthese micro-organisms did not participate in the pathogenesis of thedisease. In addition, the fate of DNA from micro-organisms that haveentered and not survived in root canals is unknown. DNA from deadcells might be adsorbed by dentine because of affinity of hydroxyapatiteto this molecule (127). However, it remains to be shown if DNA frommicrobial dead cells can really be adsorbed in dentinal walls and, ifeven, it can be retrieved during sampling with paper points. In fact, it ishighly unlikely that free microbial DNA can remain intact in an environ-ment colonized by living micro-organisms. Dnases released by someliving species as well as at cell death can degrade free DNA in theenvironment. It has been reported the presence of Dnase activity onwhole bacterial cells and vesicles thereof that can degrade linear DNA(128). These bacteria include common putative endodontic pathogens,such as Porphyromonas endodontalis, P. gingivalis, T. forsythia, Fu-sobacterium species, Prevotella intermedia, and Prevotella nigre-scens. Indeed, Dnases are of concern during sample storage, becausethey can be carried along with the sample and induce DNA degradation,with consequent false negative result after PCR amplification.

Therefore, although there is a possibility of detecting DNA fromdead cells in endodontic infections, this possibility is conceivably low. Inthe event DNA from dead cells is detected, the results by no means lacksignificance with regard to participation in disease causation. However,the ability to detect DNA from dead cells poses a major problem whenone is investigating the immediate effectiveness of antimicrobial treat-ment, because DNA from cells that have recently died can be detected.To circumvent or at least minimize this problem, one can use someadjustments in the PCR assay or take advantage of PCR technology de-rivatives, such as RT-PCR. McCarty and Atlas (129) investigated theeffect of amplicon size on the PCR detection of Legionella pneumophilaafter chlorine inactivation. Two amplicons specific to the L. pneumo-phila mip gene were used for the PCR analyses. A 168-bp amplicon wasdetectable longer after chlorine treatment than a 650-bp amplicon. Onthe basis of their data, they concluded that larger amplicons appear tocorrelate better with bacterial viability. Findings from their studyshowed that smaller fragments of DNA may persist for a longer time aftercell death than larger sequences. Therefore, designing primers to gen-erate large amplicons may reduce the risks of positive results because ofDNA from dead cells. Moreover, because RNAs are more labile and have

a shorter half-life when compared to DNA, they can be rapidly degradedafter cell death (122). Thus, assays directed towards the detection ofrRNA or mRNA through RT-PCR can be more reliable for detection ofliving cells.

Unraveling the Oral MicrobiotaThe oral cavity harbors one of the highest accumulations of micro-

organisms in the body. Even though viruses, yeasts and protozoa can befound as constituents of the oral microbiota, bacteria are by far the mostdominant inhabitants of the oral cavity. There are an estimated 1010

bacteria in the mouth (130). A high diversity of bacterial species hasbeen revealed in the oral cavity by cultivation and application of molec-ular methods to the analysis of the bacterial diversity has revealed a stillmore broad and diverse spectrum of extant oral bacteria. Overall, bac-teria detected from the oral cavity fall into 11 phyla that comprised over700 different species or phylotypes (17, 131, 132). About 40 to 50% ofthese bacteria are novel uncultivated species or phylotypes, which areknown only by 16S rDNA sequences (17). This raises the interestingpossibility that uncultivated and as-yet-uncharacterized species thathave remained invisible to studies using traditional identification meth-ods may make up a large fraction of the living oral microbiota, and mayparticipate in the etiology of oral diseases.

In fact, the introduction of molecular approaches in oral micro-biology research has brought about a significant body of new knowledgewith regard to the oral microbiota in health and disease. The develop-ment of molecular bacterial identification methods has made it possibleto study the role that uncultivable bacteria or even other micro-organ-isms such as archaea play in oral diseases (133–137). Consequently, asignificant revolution in the knowledge of the oral microbiota in healthand disease has taken place after the advent of molecular techniques formicrobial identification. In this context, endodontic infections are farfrom being an exception. The second part of this review deals with thesignificant impact of molecular genetic methods for microbial identifi-cation on the knowledge of endodontic infections.

AcknowledgmentThis study was supported in part by grants from CNPq, a Bra-

zilian Governmental Institution.

References1. Kakehashi S, Stanley HR, Fitzgerald RJ. The effects of surgical exposures of dental

pulps in germ-free and conventional laboratory rats. Oral Surg Oral Med Oral Pathol1965;18:340 – 8.

2. Sundqvist G. Bacteriological studies of necrotic dental pulps [Dissertation]. Umea,Sweden: University of Umea, 1976.

3. Eriksen HM, Kirkevang L-L, Petersson K. Endodontic epidemiology and treatmentoutcome: general considerations. Endod Top 2002;2:1–9.

4. Figdor D. Microbial aetiology of endodontic treatment failure and pathogenic prop-erties of selected species. Umea University Odontological Dissertations No. 79.Umea (Sweden): Umea University, 2002.

5. Relman DA. Emerging infections and newly-recognised pathogens. Neth J Med1997;50:216 –20.

6. Relman DA. New technologies, human-microbe interactions, and the search forpreviously unrecognized pathogens. J Infect Dis 2002;186:S254 – 8.

7. Wade W. Unculturable bacteria: the uncharacterized organisms that cause oralinfections. J R Soc Med 2002;95:81–3.

8. Relman DA, Falkow S. Identification of uncultured micro-organisms: expanding thespectrum of characterized microbial pathogens. Infect Agents Dis 1992;1:245–53.

9. Fredricks DN, Relman DA. Application of polymerase chain reaction to the diagnosisof infectious diseases. Clin Infect Dis 1999;29:475– 86.

10. Hugenholtz P, Goebel BM, Pace NR. Impact of culture-independent studies on theemerging phylogenetic view of bacterial diversity. J Bacteriol 1998;180:4765–74.

11. Rappe MS, Giovannoni SJ. The uncultured microbial majority. Annu Rev Microbiol2003;57:369 –94.

12. Hugenholtz P. Exploring prokaryotic diversity in the genomic era. Genome Biology2002;3:1– 8.

Review


13. Perkins BA, Relman D. Explaining the unexplained in clinical infectious diseases:looking forward. Emerg Infect Dis 1998;4:395–7.

14. Nikkari S, Lopez FA, Lepp PW, et al. Broad-range bacterial detection and the analysisof unexplained death and critical illness. Emerg Infect Dis 2002;8:188 –94.

15. Fredricks DN, Relman DA. Infectious agents and the etiology of chronic idiopathicdiseases. Curr Clin Top Infect Dis 1998;18:180 –200.

16. Suau A, Bonnet R, Sutren M, et al. Direct analysis of genes encoding 16S rRNA fromcomplex communities reveals many novel molecular species within the human gut.Appl Environ Microbiol 1999;65:4799 – 807.

17. Paster BJ, Boches SK, Galvin JL, et al. Bacterial diversity in human subgingivalplaque. J Bacteriol 2001;183:3770 – 83.

18. Hayashi H, Sakamoto M, Benno Y. Phylogenetic analysis of the human gut micro-biota using 16S rDNA clone libraries and strictly anaerobic culture-based methods.Microbiol Immunol 2002;46:535– 48.

19. Wade WG. Non-culturable bacteria in complex commensal populations. Adv ApplMicrobiol 2004;54:93–106.

20. Kell DB, Young M. Bacterial dormancy and culturability: the role of autocrinegrowth factors. Curr Opin Microbiol 2000;3:238 – 43.

21. Breznak JA. A need to retrieve the not-yet-cultured majority. Environ Microbiol2002;4:4 –5.

22. Stevenson BS, Eichorst SA, Wertz JT, Schmidt TM, Breznak JA. New strategies forcultivation and detection of previously uncultured microbes. Appl Environ Micro-biol 2004;70:4748 –55.

23. Tanner A, Lai C-H, Maiden M. Characteristics of oral gram-negative species. In: SlotsJ, Taubman MA, eds. Contemporary Oral Microbiology and Immunology. St Louis:Mosby, 1992;299 –341.

24. Bosshard PP, Abels S, Zbinden R, Bottger EC, Altwegg M. Ribosomal DNA sequenc-ing for identification of aerobic gram-positive rods in the clinical laboratory (an18-month evaluation). J Clin Microbiol 2003;41:4134 – 40.

25. Drancourt M, Berger P, Raoult D. Systematic 16S rRNA gene sequencing of atypicalclinical isolates identified 27 new bacterial species associated with humans. J ClinMicrobiol 2004;42:2197–202.

26. Drancourt M, Bollet C, Carlioz A, Martelin R, Gayral JP, Raoult D. 16S ribosomal DNAsequence analysis of a large collection of environmental and clinical unidentifiablebacterial isolates. J Clin Microbiol 2000;38:3623–30.

27. Song Y, Liu C, McTeague M, Finegold SM. 16S ribosomal DNA sequence-basedanalysis of clinically significant gram-positive anaerobic cocci. J Clin Microbiol2003;41:1363–9.

28. Sixou M. Diagnostic testing as a supportive measure of treatment strategy. Oral DisSuppl 2003;1:54 – 62.

29. Zambon JJ, Haraszthy VI. The laboratory diagnosis of periodontal infections. Peri-odontol 2000 1995;7:69 – 82.

30. Pace NR. A molecular view of microbial diversity and the biosphere. Science 1997;276:734 – 40.

31. Ward DM, Weller R, Bateson MM. 16S rRNA sequences reveal numerous unculti-vated micro-organisms in a natural environment. Nature 1990;345:63–5.

32. Amann RI, Ludwig W, Schleifer KH. Phylogenetic identification and in situ detectionof individual microbial cells without cultivation. Microbiol Rev 1995;59:143– 69.

33. Kato C, Li L, Tamaoka J, Horikoshi K. Molecular analyses of the sediment of the11,000-m deep Mariana Trench. Extremophiles 1997;1:117–23.

34. Hugenholtz P, Pitulle C, Hershberger KL, Pace NR. Novel division level bacterialdiversity in a Yellowstone hot spring. J Bacteriol 1998;180:366 –76.

35. Wilson KH, Blitchington RB. Human colonic biota studied by ribosomal DNA se-quence analysis. Appl Environ Microbiol 1996;62:2273– 8.

36. Sakamoto M, Umeda M, Ishikawa I, Benno Y. Comparison of the oral bacterial florain saliva from a healthy subject and two periodontitis patients by sequence analysisof 16S rDNA libraries. Microbiol Immunol 2000;44:643–52.

37. Hold GL, Pryde SE, Russel VJ, Furrie E, Flint HJ. Assessment of microbial diversity inhuman colonic samples by 16S rDNA sequence analysis. FEMS Microbiol Ecol2002;39:33–9.

38. Wang X, Heazlewood SP, Krause DO, Florin THJ. Molecular characterization of themicrobial species that colonize human ileal and colonic mucosa by using 16S rDNAsequence analysis. J Appl Microbiol 2003;95:508 –20.

39. Pei Z, Bini EJ, Yang L, Zhou M, Francois F, Blaser MJ. Bacterial biota in the humandistal esophagus. Proc Natl Acad Sci USA 2004;101:4250 –5.

40. Anderson BE, Dawson JE, Jones DC, Wilson KH. Ehrlichia chaffeensis, a new speciesassociated with human ehrlichiosis. J Clin Microbiol 1991;29:2838 – 42.

41. Relman DA, Schmidt TM, MacDermott RP, Falkow S. Using 16S rRNA to find thecause of Whipple’s disease. N Engl J Med 1992;327:293–301.

42. Relman DA. The identification of uncultured microbial pathogens. J Infect Dis 1993;168:1– 8.

43. Relman DA. Detection and identification of previously unrecognized microbialpathogens. Emerg Infect Dis 1998;4:382– 8.

44. Relman DA. The search for unrecognized pathogens. Science 1999;284:1308 –10.45. Woese CR. Bacterial evolution. Microbiol Rev 1987;51:221–71.

46. Atlas RM. Principles of microbiology, 2nd ed. Dubuque: WCB Publishers, 1997.47. Madigan MT, Martinko JM, Parker J. Brock biology of micro-organisms, 9th ed.

Upper Saddle River, NJ; Prentice-Hall, 2000.48. Maiwald M. Broad-range PCR for detection and identification of bacteria. In: Persing

DH, Tenover FC, Versalovic J, et al., eds. Molecular Microbiology. Diagnostic Prin-ciples and Practice. Washington, DC: ASM Press, 2004;379 –90.

49. Gonzalez N, Romero J, Espejo RT. Comprehensive detection of bacterial populationsby PCR amplification of the 16S–23S rRNA spacer region. J Microbiol Meth 2003;55:91–7.

50. Anderson IC, Cairney JWG. Diversity and ecology of soil fungal communities: in-creased understanding through the application of molecular techniques. EnvironMicrobiol 2004;6:769 –79.

51. Lepp PW, Relman DA. Molecular phylogenetic analysis. In: Persing DH, Tenover FC,Versalovic J, et al., eds. Molecular Microbiology. Diagnostic Principles and Practice.Washington, DC: ASM Press, 2004;161– 80.

52. Mullis KB, Ferre F, Gibbs RA. The Polymerase Chain Reaction. Boston: Birkhauser,1994.

53. Lee HC, Tirnady F. Blood Evidence. How DNA is Revolutionizing the Way We SolveCrimes. Cambridge, MA: Perseus Publishing, 2003.

54. Whelen AC, Persing DH. The role of nucleic acid amplification and detection in theclinical microbiology laboratory. Annu Rev Microbiol 1996;50:349 –73.

55. Tang Y-W, Procop GW, Persing DH. Molecular diagnostics of infectious diseases.Clin Chem 1997;43:2021–38.

56. Louie M, Louie L, Simor AE. The role of DNA amplification technology in the diag-nosis of infectious diseases. CMAJ 2000;163:301–9.

57. Siqueira JF Jr, Rocas IN. PCR methodology as a valuable tool for identification ofendodontic pathogens. J Dent 2003;31:333–9.

58. Haqqi TM, Sarkar G, David CS, Sommer SS. Specific amplification with PCR of arefractory segment of genomic DNA. Nucleic Acids Res 1988;16:11844

59. McPherson MJ, Moller SG. PCR. Oxford, UK: BIOS Scientific Publishers Ltd, 2000;67– 87.

60. Sambrook J, Russell DW. Molecular Cloning: A Laboratory Manual, 3rd ed. ColdSpring Harbor, NY: Cold Spring Harbor Laboratory Press, 2001;8.46 – 8.53.

61. Dieffenbach CW, Dveksler GS. PCR Primer. A Laboratory Manual. Plain View, NY:Cold Spring Harbor Laboratory Press, 1995.

62. Hayden RT. In vitro nucleic acid amplification techniques. In: Persing DH, TenoverFC, Versalovic J, et al., eds. Molecular Microbiology. Diagnostic Principles andPractice. Washington, DC: ASM Press, 2004;43– 69.

63. Welsh J, McClelland M. Fingerprinting genomes using PCR with arbitrary primers.Nucleic Acids Res 1990;18:7213– 8.

64. De Bruijn FJ. Use of repetitive (repetitive extragenic palindromic and enterobacte-rial repetitive intergenic consensus) sequences and the polymerase chain reactionto fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria.Appl Environ Microbiol 1992;58:2169 – 87.

65. Arora DK, Hirsch PR, Kerry BR. PCR-based molecular discrimination of Verticil-lium chlamydosporium isolates. Mycol Res 1996;7:801–9.

66. Higgins CF, Ames GF, Barnes WM, Clement JM, Hofung M. A novel inter-cistonicregulatory element of prokaryotic operons. Nature 1982;298:760 –2.

67. Belgrader P, Benett W, Hadley D, Richards J, Stratton P, Mariella R Jr, Milanovich F.PCR detection of bacteria in seven minutes. Science 1999;284:449 –50.

68. Cashion AK, Driscoll CJ, Sabek O. Emerging genetic technologies in clinical andresearch settings. Biol Res Nurs 2004;5:159 – 67.

69. Mackay IM. Real-time PCR in the microbiology laboratory. Clin Microbiol Infect2004;10:190 –212.

70. Bustin SA. Absolute quantification of mRNA using realtime reverse transcriptionpolymerase chain reaction assays. J Mol Endocrinol 2000;25:169 –93.

71. Holland PM, Abramson RD, Watson R, Gelfand DH. Detection of specific polymerasechain reaction product by utilizing the 5�-3� exonuclease activity of Thermusaquaticus DNA polymerase. Proc Natl Acad Sci USA 1991;88:7276 – 80.

72. Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative PCR. Genome Res1996;6:986 –94.

73. Mhlanga MM, Malmberg L. Using molecular beacons to detect single-nucleotidepolymorphisms with real-time PCR. Methods 2001;25:463–71.

74. Wittwer CT, Kusukawa N. Real-time PCR. In: Persing DH, Tenover FC, Versalovic J, etal., eds. Molecular Microbiology. Diagnostic Principles and Practice. Washington,DC: ASM Press, 2004;71– 84.

75. Raoult D, Fournier PE, Drancourt M. What does the future hold for clinical micro-biology? Nat Rev Microbiol 2004;2:151–9.

76. Gobel UB. Phylogenetic amplification for the detection of uncultured bacteria andthe analysis of complex microbiota. J Microbiol Meth 1995;23:117–28.

77. Pitt TL, Saunders NA. Molecular bacteriology: a diagnostic tool for the millennium.J Clin Pathol 2000;53:71–5.

78. Dragon E. Handling reagents in the PCR laboratory. PCR Methods Appl 1993;3:S8 –9.

79. Myers RM, Fischer SG, Lerman LS, Maniatis T. Nearly all single base substitutions in

Review


DNA fragments joined to a GC-clamp can be detected by denaturing gradient gelelectrophoresis. Nucleic Acids Res 1985;13:3131– 45.

80. Muyzer G, de Waal EC, Uitterlinden AG. Profiling of complex microbial populationsby denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 1993;59:695–700.

81. Gasser RB. What’s in that band? Int J Parasitol 1998;28:989 –96.82. Muyzer G, Smalla K. Application of denaturing gradient gel electrophoresis (DGGE)

and temperature gradient gel electrophoresis (TGGE) in microbial ecology. AntonieVan Leeuwenhoek 1998;73:127– 41.

83. Siqueira JF Jr, Rocas IN, Rosado AS. Application of denaturing gradient gel electro-phoresis (DGGE) to the analysis of endodontic infections. J Endod (in press).

84. Marsh TL. Terminal restriction fragment length polymorphism (T-RFLP): an emerg-ing method for characterizing diversity among homologous populations of ampli-fication products. Curr Opin Microbiol 1999;2:323–7.

85. Clement BG, Kehl LE, De Bord KL, Kitts CL. Terminal restriction fragment patterns(TRFPs), a rapid, PCR-based method for the comparison of complex bacterialcommunities. J Microbiol Meth 1998;31:135– 42.

86. Socransky SS, Smith C, Martin L, Paster BJ, Dewhirst FE, Levin AE. “Checkerboard”DNA-DNA hybridization. Biotechniques 1994;17:788 –92.

87. Papapanou PN, Madianos PN, Dahlen G, Sandros J. “Checkerboard” versus culture:a comparison between two methods for identification of subgingival microbiota. EurJ Oral Sci 1997;105:389 –96.

88. Socransky SS, Haffajee AD, Cugini MA, Smith C, Kent RL Jr. Microbial complexes insubgingival plaque. J Clin Periodontol 1998;25:134 – 44.

89. Paster BJ, Bartoszyk IM, Dewhirst FE. Identification of oral streptococci using PCR-based, reverse-capture, checkerboard hybridization. Meth Cell Sci 1998;20:223–31.

90. Stover AG, Jeffery E, Xu J, Persing DH. Hybridization array technology. In: PersingDH, Tenover FC, Versalovic J, et al., eds. Molecular Microbiology. Diagnostic Prin-ciples and Practice. Washington, DC: ASM Press, 2004;619 –39.

91. Schena M, Shalon D, Davis RW, Brown PO. Quantitative monitoring of gene expres-sion patterns with a complementary DNA microarray. Science 1995;270:467–70.

92. Cheung VG, Morley M, Aguilar F, Massimi A, Kucherlapati R, Childs G. Making andreading microarrays. Nature Genet 1999;21:15–9.

93. Shackel NA, Gorrell MD, McCaughan GW. Gene array analysis and the liver. Hepa-tology 2002;36:1313–25.

94. Afshari CA. Perspective: microarray technology, seeing more than spots. Endocri-nology 2002;143:1983–9.

95. DeRisi JL, Iyer VR, Brown PO. Exploring the metabolic and genetic control of geneexpression on a genomic scale. Science 1997;278:680 – 6.

96. Tao H, Bausch C, Richmond C, Blattner FR, Conway T. Functional genomics: ex-pression analysis of Escherichia coli growing on minimal and rich media. J Bacte-riol 1999;181:6425– 40.

97. Khan J, Simon R, Bittner M, et al. Gene expression profiling of alveolar rhabdomyo-sarcoma with cDNA microarrays. Cancer Res 1998;58:5009 –13.

98. Perou CM, Jeffrey SS, van de Rijn M, et al. Distinctive gene expression patterns inhuman mammary epithelial cells and breast cancers. Proc Natl Acad Sci USA 1999;96:9212–7.

99. Wang K, Gan L, Jeffery E, et al. Monitoring gene expression profile changes inovarian carcinomas using cDNA microarray. Gene 1999;229:101– 8.

100. Marton MJ, DeRisi JL, Bennett HA, et al. Drug target validation and identification ofsecondary drug target effects using DNA microarrays. Nat Med 1998;4:1293–301.

101. Cummings CA, Relman DA. Using DNA microarrays to study host-microbe interac-tions. Emerg Infect Dis 2000;6:513–25.

102. Diehn M, Relman DA. Comparing functional genomic datasets: lessons from DNAmicroarray analyses of host-pathogen interactions. Curr Opin Microbiol 2001;4:95–101.

103. Lucchini S, Thompson A, Hinton JC. Microarrays for microbiologists. Microbiology2001;147:1403–14.

104. Kato-Maeda M, Gao Q, Small PM. Microarray analysis of pathogens and their inter-action with hosts. Cell Microbiol 2001;3:713–9.

105. Call DR, Borucki MK, Loge FJ. Detection of bacterial pathogens in environmentalsamples using DNA microarrays. J Microbiol Meth 2003;53:235– 43.

106. Vasil ML. DNA microarrays in analysis of quorum sensing: strengths and limitations.J Bacteriol 2003;185:2061–5.

107. Bryant PA, Venter D, Robins-Browne R, Curtis N. Chips with everything: DNA mi-croarrays in infectious diseases. Lancet Infect Dis 2004;4:100 –11.

108. Moter A, Gobel UB. Fluorescence in situ hybridization (FISH) for direct visualizationof micro-organisms. J Microbiol Meth 2000;41:85–112.

109. Amann R, Fuchs BM, Behrens S. The identification of micro-organisms by fluores-cence in situ hybridisation. Curr Opin Biotechnol 2001;12:231– 6.

110. Moter A, Leist G, Rudolph R, et al. Fluorescence in situ hybridization shows spatialdistribution of as yet uncultured treponemes in biopsies from digital dermatitislesions. Microbiology 1998;144:2459 – 67.

111. Wagner M, Horny M, Daims H. Fluorescence in situ hybridisation for the identifi-cation and characterisation of prokaryotes. Curr Opin Microbiol 2003;6:302–9.

112. Reysenbach AL, Giver LJ, Wickham GS, Pace NR. Differential amplification of rRNAgenes by polymerase chain reaction. Appl Environ Microbiol 1992;58:3417– 8.

113. Wintzingerode FV, Gobel UB, Stackebrandt E. Determination of microbial diversityin environmental samples: pitfalls of PCR-based rRNA analysis. FEMS Microbiol Rev1997;21:213–29.

114. Rochelle PA, Cragg BA, Fry JC, Parkes RJ, Weightman AJ. Effect of sample handlingon estimation of bacterial diversity in marine sediments by 16S rRNA gene sequenceanalysis. FEMS Microbiol Ecol 1994;15:215–26.

115. Kirk JL, Beaudette LA, Hart M, et al. Methods of studying soil microbial diversity. JMicrobiol Meth 2004;58:169 – 88.

116. Siqueira JF Jr, Rocas IN. Polymerase chain reaction-based analysis of micro-organ-isms associated with failed endodontic treatment. Oral Surg Oral Med Oral PatholOral Radiol Endod 2004;97:85–94.

117. Boissinot M, Bergeron MG. Toward rapid real-time molecular diagnostic to guidesmart use of antimicrobials. Curr Opin Microbiol 2002;5:478 – 82.

118. Sundqvist G, Figdor D. Life as an endodontic pathogen. Ecological differences be-tween the untreated and root-filled root canals. Endod Top 2003;6:3–28.

119. Wang R-F, Cao W-W, Cerniglia CE. PCR detection and quantitation of predominantanaerobic bacteria in human and animal fecal samples. Appl Environ Microbiol1996;62:1242–7.

120. Rantakokko-Jalava K, Nikkari S, Jalava J, et al. Direct amplification of rRNA genes indiagnosis of bacterial infections. J Clin Microbiol 2000;38:32–9.

121. Josephson KL, Gerba CP, Pepper IL. Polymerase chain reaction detection of nonvi-able bacterial pathogens. Appl Environ Microbiol 1993;59:3513–5.

122. Keer JT, Birch L. Molecular methods for the assessment of bacterial viability. JMicrobiol Meth 2003;53:175– 83.

123. Malawista SE, Barthold SW, Persing DH. Fate of Borrelia burgdorferi DNA in tissuesof infected mice after antibiotic treatment. J Infect Dis 1994;170:1312– 6.

124. Aul JJ, Anderson KW, Wadowsky RM, et al. Comparative evaluation of culture andPCR for the detection and determination of persistence of bacterial strains and DNAsin the Chinchilla laniger model of otitis media. Ann Otol Rhinol Laryngol 1998;107:508 –13.

125. Wicher K, Abbruscato F, Wicher V, Collins DN, Auger I, Horowitz HW. Identificationof persistent infection in experimental syphilis by PCR. Infect Immun 1998;66:2509 –13.

126. Kennedy N, Gillespie SH, Saruni AO, et al. Polymerase chain reaction for assessing treat-ment response in patients with pulmonary tuberculosis. J Infect Dis 1994;170:713–6.

127. Bernardi G. Chromatography of nucleic acids on hydroxyapatite. Nature 1965;206:779 – 83.

128. Leduc A, Grenier D, Mayrand D. Outer membrane-associated deoxyribonucleaseactivity of Porphyromonas gingivalis. Anaerobe 1995;1:129 –34.

129. McCarty SC, Atlas RM. Effect of amplicon size on PCR detection of bacteria exposedto chlorine. PCR Methods Appl 1993;3:181–5.

130. Mims C, Nash A, Stephen J. Mims� Pathogenesis of Infectious Diseases, 5th ed. SanDiego: Academic Press, 2001;361–91.

131. Kazor CE, Mitchell PM, Lee AM, et al. Diversity of bacterial populations on the tonguedorsa of patients with halitosis and healthy patients. J Clin Microbiol 2003;41:558 –63.

132. Paster BJ, Falkler WA Jr, Enwonwu CO, et al. Prevalent bacterial species and novelphylotypes in advanced noma lesions. J Clin Microbiol 2002;40:2187–91.

133. Kumar PS, Griffen AL, Barton JA, Paster BJ, Moeschberger ML, Leys EJ. New bacterialspecies associated with chronic periodontitis. J Dent Res 2003;82:338 – 44.

134. Sakamoto M, Huang Y, Umeda M, Ishikawa I, Benno Y. Detection of novel oralphylotypes associated with periodontitis. FEMS Microbiol Lett 2002;217:65–9.

135. Griffen AL, Kumar PS, Leys EJ. A quantitative, molecular view of oral biofilm communitiesin health and disease suggests a role for uncultivated species. Polymicrobial diseases, AmSociety for Microbiology conferences, Lake Tahoe, Nevada, 2003; p. 13.

136. Brinig MM, Lepp PW, Ouverney CC, Armitage GC, Relman DA. Prevalence of bacteriaof division TM7 in human subgingival plaque and their association with disease.Appl Environ Microbiol 2003;69:1687–94.

137. Lepp PW, Brinig MM, Ouverney CC, Palm K, Armitage GC, Relman DA. MethanogenicArchaea and human periodontal disease. Proc Natl Acad Sci USA 2004;101:6176–81.

Review


exploiting molecular methods to explore endodontic ...means of their phenotypic features, this...

Documents