JOURNAL OF CLINICAL MICROBIOLOGY, Nov. 2005, p. 5721–5732 Vol. 43, No.

11
0095-1137/05/$08.00⫹0 doi:10.1128/JCM.43.11.5721–5732.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Defining the Normal Bacterial Flora of the Oral Cavity

Jørn A. Aas,1,2* Bruce J. Paster,1,3 Lauren N. Stokes,1 Ingar Olsen,2
and Floyd E. Dewhirst1,3
Department of Molecular Genetics, The Forsyth Institute,1 and Faculty of Dentistry,2 University of Oslo, Oslo, Norway,
and Department of Oral and Developmental Biology, Harvard School of Dental Medicine, Boston, Massachusetts3
Received 10 June 2005/Returned for modification 2 August 2005/Accepted 12 August 2005

More than 700 bacterial species or phylotypes, of which over 50% have not been cultivated, have been
detected in the oral cavity. Our purposes were (i) to utilize culture-independent molecular techniques to extend
our knowledge on the breadth of bacterial diversity in the healthy human oral cavity, including not-yet-
cultivated bacteria species, and (ii) to determine the site and subject specificity of bacterial colonization. Nine
sites from five clinically healthy subjects were analyzed. Sites included tongue dorsum, lateral sides of tongue,
buccal epithelium, hard palate, soft palate, supragingival plaque of tooth surfaces, subgingival plaque, max-
illary anterior vestibule, and tonsils. 16S rRNA genes from sample DNA were amplified, cloned, and trans-
formed into Escherichia coli. Sequences of 16S rRNA genes were used to determine species identity or closest
relatives. In 2,589 clones, 141 predominant species were detected, of which over 60% have not been cultivated.
Thirteen new phylotypes were identified. Species common to all sites belonged to the genera Gemella, Granu-
licatella, Streptococcus, and Veillonella. While some species were subject specific and detected in most sites, other

Downloaded from jcm.asm.org by on May 21, 2008

species were site specific. Most sites possessed 20 to 30 different predominant species, and the number of
predominant species from all nine sites per individual ranged from 34 to 72. Species typically associated with
periodontitis and caries were not detected. There is a distinctive predominant bacterial flora of the healthy oral
cavity that is highly diverse and site and subject specific. It is important to fully define the human microflora
of the healthy oral cavity before we can understand the role of bacteria in oral disease.

The oral cavity is comprised of many surfaces, each coated on soft and hard tissues in healthy subjects. They also found
with a plethora of bacteria, the proverbial bacterial biofilm. that the profiles of the soft tissues were more similar to each
Some of these bacteria have been implicated in oral diseases other than those of supragingival and subgingival plaques.
such as caries and periodontitis, which are among the most Our purposes were as follows: (i) to utilize culture-indepen-
common bacterial infections in humans. For example, it has dent molecular techniques to extend our knowledge on the
been estimated that at least 35% of dentate U.S. adults aged 30 breadth of bacterial diversity in the healthy human oral cavity,
to 90 years have periodontitis (1). In addition, specific oral including not-yet-cultivated phylotypes, and (ii) to determine
bacterial species have been implicated in several systemic dis- the site and subject specificity of bacterial colonization.
eases, such as bacterial endocarditis (4), aspiration pneumonia
(26), osteomyelitis in children (8), preterm low birth weight (6, MATERIALS AND METHODS
20), and cardiovascular disease (2, 34). Surprisingly, little is Subjects. Five subjects, representing both genders, ranging in age from 23 to
known about the microflora of the healthy oral cavity. 55 and with no clinical signs of oral mucosal disease were included in the study.
By using culture-independent molecular methods, we previ- Subjects did not suffer from severe halitosis. The periodontia were healthy in that
ously detected over 500 species or phylotypes in subgingival all periodontal pockets were less than 3 mm deep with no redness or inflamma-
tion of the gums. Subjects did not have active white spot lesions or caries on the
plaque of healthy subjects and subjects with periodontal dis- teeth. Our results were consistent with these clinical observations in that species
eases (21), necrotizing ulcerative periodontitis in human im- typically found in caries subjects were not detected (3). The subjects had not used
munodeficiency virus-positive subjects (23), dental plaque in antibiotics for the last 6 months.
children with rampant caries (3), noma (22), and on the tongue Sample collection. Samples from the following nine sites were analyzed for
each subject: dorsum of the tongue, lateral sides of the tongue, buccal fold, hard
dorsum of subjects with and without halitosis (15). Other in-
palate, soft palate, labial gingiva and tonsils of soft tissue surfaces, and supra-
vestigators have used similar techniques to determine the bac- gingival and subgingival plaques from tooth surfaces. Microbiological samples of
terial diversity of saliva (25), subgingival plaque of a subject supragingival and subgingival plaque samples were taken with a sterile Gracey
with gingivitis (16), and dentoalveolar abscesses (10, 29). Over curette. Other samples were collected with sterile swab brushes.
half of the species detected have not yet been cultivated. Data Sample lysis. Plaque samples were directly suspended in 50 ␮l of 50 mM Tris
buffer (pH 7.6), 1 mM EDTA, pH 8, and 0.5% Tween 20. Proteinase K
from these studies have implicated specific species or phylo- (200 ␮g/ml; Roche Applied Science, Indianapolis, IN) was added to the mixture.
types in a variety of diseases and oral infections, but still only The samples were then heated at 55°C for 2 h. Proteinase K was inactivated by
limited information is available on species associated with heating at 95°C for 5 min. Detection of species is dependent upon obtaining
health. Recently, Mager et al. (19) demonstrated significant DNA that can be amplified. Thus, a difficult-to-lyse bacterium may not be
detected. However, by using our lysis technique, we were able to detect many
differences in the bacterial profiles of 40 oral cultivable species
hard-to-lyse species, such as species of Actinomyces and Streptococcus.
Amplification of 16S rRNA genes by PCR and purification of PCR products.
The 16S rRNA genes were amplified under standardized conditions using a
* Corresponding author. Mailing address: Institute of Oral Biology, universal primer set (forward primer, 5⬘-GAG AGT TTG ATY MTG GCT
University of Oslo, Postbox 1052 Blindern, 0316 Oslo, Norway. Phone: CAG-3⬘; reverse primer, 5⬘-GAA GGA GGT GWT CCA RCC GCA-3⬘) (21).
(47) 22840343. Fax: (47) 22840305. E-mail: [email protected]. Primers were synthesized commercially (Operon Technologies, Alameda, CA).

5721
5722 AAS ET AL. J. CLIN. MICROBIOL.

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 1. Bacterial profiles of the buccal epithelium of healthy subjects. Distribution and levels of bacterial species/phylotypes among five subjects
are shown by the columns of boxes to the right of the tree as either not detected (clear box), ⬍15% of the total number of clones assayed (shaded
box), or ⱖ15% of the total number of clones assayed (darkened box). The 15% was chosen arbitrarily. GenBank accession numbers are provided.
Marker bar represents a 5% difference in nucleotide sequences.

The PCR primers do not necessarily include all bacterial species. Nevertheless, mM Tris. Correct sizes of the inserts were determined in a PCR with an M13
a wide range of phylogenetic types were obtained in this study and our previous (⫺20) forward primer and an M13 reverse primer (Invitrogen). Prior to sequenc-
studies by using this universal primer set (3, 15, 21–23). PCR was performed in ing of the fragments, the PCR-amplified 16S rRNA fragments were purified and
thin-walled tubes with a GeneAmp PCR system 9700 (ABI, Foster City, CA). concentrated with Microcon 100 (Amicon, Bedford, MA), followed by use of the
One microliter of the lysed sample was added to a reaction mixture (final volume, QIAquick PCR purification kit (QIAGEN, Valencia, CA).
50 ␮l) containing 20 pmol of each primer, 40 nmol of deoxynucleoside triphos- 16S rRNA gene sequencing. Purified PCR-amplified 16S rRNA inserts were
phates, and 1 U of Platinum Taq polymerase (Invitrogen, San Diego, CA). In a sequenced using an ABI Prism cycle sequencing kit (BigDye terminator cycle
hot-start protocol, the samples were preheated at 95°C for 4 min, followed by sequencing kit with AmpliTaq DNA polymerase FS, GeneAmp PCR system
amplification under the following conditions: denaturation at 95°C for 45 s, 9700; ABI). The primers used for sequencing have been described previously
annealing at 60°C for 45 s, and elongation at 72° for 1.5 min, with an additional (21). Quarter dye chemistry was used with 80 ␮M primers and 1.5 ␮l of PCR
15 s for each cycle. A total of 30 cycles were performed; this was followed by a product in a final volume of 20 ␮l. Cycle sequencing was performed with a
final elongation step at 72°C for 15 min. The results of the PCR amplification GeneAmp PCR system 9700 (ABI) with 25 cycles of denaturation at 96°C for
were examined by electrophoresis in a 1% agarose gel. DNA was stained with 10 s, annealing at 55° for 5 s, and extension at 60°C for 4 min. The sequencing
ethidium bromide and visualized under short-wavelength UV light. reactions were run on an ABI 3100 DNA sequencer (ABI).
Cloning procedures. Cloning of PCR-amplified DNA was performed with the 16S rRNA gene sequencing and data analysis of unrecognized inserts. A total
TOPO TA cloning kit (Invitrogen) according to the instructions of the manu- of 2,589 clones with an insert of the correct size of approximately 1,500 bases
facturer. Transformation was done with competent Escherichia coli TOP10 cells were analyzed. The number of clones per subject that were sequenced ranged
provided by the manufacturer. The transformed cells were then plated onto from 42 to 69, with an average of 57.5 and a standard deviation of 7.1. A
Luria-Bertani agar plates supplemented with kanamycin (50 ␮g/ml), and the sequence of approximately 500 bases was obtained first to determine identity or
plates were incubated overnight at 37°C. Each colony was placed into 40 ␮l of 10 approximate phylogenetic position. Full sequences of about 1,500 bases were
VOL. 43, 2005 NORMAL ORAL BACTERIAL FLORA 5723

FIG. 2. Bacterial profiles of the maxillary anterior vestibule of healthy subjects. Distribution and levels of bacterial species/phylotypes among
five subjects are as described for Fig. 1. Novel phylotypes identified in this study are indicated in bold. GenBank accession numbers are provided.
Marker bar represents a 5% difference in nucleotide sequences.

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 3. Bacterial profiles of the tongue dorsum of healthy subjects. Distribution and levels of bacterial species/phylotypes among five subjects
are as described for Fig. 1. Novel phylotypes identified in this study are indicated in bold. GenBank accession numbers are provided. Marker bar
represents a 5% difference in nucleotide sequences.
5724 AAS ET AL. J. CLIN. MICROBIOL.

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 4. Bacterial profiles of the lateral tongue surface of healthy subjects. Distribution and levels of bacterial species/phylotypes among five
subjects are as described for Fig. 1. Novel phylotypes identified in this study are indicated in bold. GenBank accession numbers are provided.
Marker bar represents a 5% difference in nucleotide sequences.

obtained by using five to six additional sequencing primers (15) for those species RESULTS AND DISCUSSION
deemed novel. For identification of closest relatives, the sequences of the un-
recognized inserts were compared to the 16S rRNA sequences of over 10,000 It has long been known that oral bacteria preferentially
microorganisms in our database and over 100,000 sequences in the Ribosomal
colonize different surfaces in the oral cavity as a result of
Database Project (7) and the GenBank databases. Our cutoff for species differ-
entiation was 2%, or approximately 30 bases for a full sequence. The similarity specific adhesins on the bacterial surface binding to comple-
matrices were corrected for multiple base changes at single positions by the mentary specific receptors on a given oral surface (11, 12).
method of Jukes and Cantor (13). Similarity matrices were constructed from the Indeed, the profiles of 40 cultivable bacterial species differed
aligned sequences by using only those sequence positions for which data were markedly on different oral soft tissue surfaces, saliva, and su-
available for 90% of the strains tested. Phylogenetic trees were constructed by
the neighbor-joining method of Saitou and Nei (24). TREECON, a software
pragingival and subgingival plaques from healthy subjects (19).
package for the Microsoft Windows environment, was used for the construction The purpose of this study was to define the predominant bac-
and drawing of evolutionary trees (28). We are aware of the potential creation of terial flora of the healthy oral cavity by identifying and com-
16S rRNA chimera molecules assembled during the PCR (18). The percentage paring the cultivable and the not-yet-cultivated bacterial spe-
of chimeric inserts in 16S rRNA gene libraries ranged from 1 to 15%. Chimeric
cies on different soft tissues and supra- and subgingival plaques.
sequences were identified by using the Chimera Check program in the Ribo-
somal Database Project, by treeing analysis, or by base signature analysis. Species Based on the analysis of 2,589 16S rRNA clones, the bacte-
identification of chimeras was obtained, but the sequences were not examined for rial diversity of the microflora from nine different sites of five
phylogenetic analysis. clinically healthy subjects was striking—a total of 141 different
Nucleotide sequence accession numbers. The complete 16S rRNA gene se- bacterial taxa representing six different bacterial phyla were
quences of clones representing novel phylotypes defined in this study, sequences
of known species not previously reported, and published sequences are available
detected. The six phyla included the Firmicutes (previously
for electronic retrieval from the EMBL, GenBank, and DDBJ nucleotide se- referred to as the low-G⫹C gram positives, such as species of
quence databases under the accession numbers shown in Fig. 1 to 9. Streptococcus, Gemella, Eubacterium, Selenomonas, Veillonella,
VOL. 43, 2005 NORMAL ORAL BACTERIAL FLORA 5725

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 5. Bacterial profiles of the hard palate of healthy subjects. Distribution and levels of bacterial species/phylotypes among five subjects are
as described for Fig. 1. Novel phylotypes identified in this study are indicated in bold. GenBank accession numbers are provided. Marker bar
represents a 5% difference in nucleotide sequences.

and related ones), the Actinobacteria (previously referred to as about 60% as not-yet-cultivable phylotypes (J. A. Aas, S. R. Dar-
the high-G⫹C gram positives, such as species of Actinomyces, dis, A. L. Griffen, L. N. Stokes, A. M. Lee, I. Olsen, F. E. Dew-
Atopobium, Rothia, and related ones), the Proteobacteria (e.g., hirst, E. J. Leys, and B. J. Paster. J. Dent. Res. 84:abstract 2805,
species of Neisseria, Eikenella, Campylobacter, and related 2005). Consequently, it is important to identify both the cultivable
ones), the Bacteroidetes (e.g., species of Porphyromonas, Pre- and not-yet-cultivated bacterial flora in a given environment be-
votella, Capnocytophaga, and related ones), the Fusobacteria fore we can ascribe association to health or disease status. There
(e.g., species of Fusobacterium and Leptotrichia), and the TM7 is no evidence to suggest that the not-yet-cultivated segment is
phylum, for which there are no cultivable representatives. As any less important than the cultivable segment.
determined in our previous studies using culture-independent Bacterial profiles for each site tested for each subject are
molecular techniques (15, 21, 22), over 60% of the bacterial shown in nine phylogenetic trees (see Fig. 1 through 9). In
flora was represented by not-yet-cultivated phylotypes. Thir- these dendrograms, the distribution of bacterial species or
teen new phylotypes (see Fig. 2 through 9) were identified for phylotypes in each subject can be observed at a glance. For
the first time in this study. In a comparable ongoing study of example, it is clear that Streptococcus mitis, S. mitis bv. 2, and
caries in permanent teeth, using the same techniques, we de- Gemella hemolysans are the predominant species of the buccal
tected 224 species in the analysis of 1,275 16S rRNA clones with epithelium (Fig. 1). Each of these species was detected in most
5726 AAS ET AL. J. CLIN. MICROBIOL.

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 6. Bacterial profiles of the soft palate of healthy subjects. Distribution and levels of bacterial species/phylotypes among five subjects are
as described for Fig. 1. Novel phylotypes identified in this study are indicated in bold. GenBank accession numbers are provided. Marker bar
represents a 5% difference in nucleotide sequences.

or all of the subjects and represented ⱖ15% of the total num- example, S. mitis bv. 2 was well represented at the lateral side
ber of assayed clones (Fig. 1). Similarly, the predominant bac- of the tongue but not detected on the tongue dorsum. Con-
terial flora for the other eight oral sites was examined. In the versely, S. parasanguinis strain 85-81 was present on the tongue
maxillary anterior vestibule, S. mitis, Granulicatella spp., and dorsum, but was absent on the lateral side. This was not sur-
Gemella spp. predominated (Fig. 2). On the tongue dorsum, prising, because these surfaces are known to be different in
several species of Streptococcus, such as S. mitis, Streptococcus ultrastructure and function. For instance, the lateral side of the
australis, Streptococcus parasanguinis, Streptococcus salivarius, tongue has a smooth nonkeratinized surface, in contrast to the
Streptococcus sp. clone FP015, and Streptococcus sp. clone dorsum of the tongue, which is a keratinized, highly papillated
FN051, Granulicatella adiacens, and Veillonella spp. were the surface with a large surface area and underlying serous glands.
predominant species (Fig. 3). On the lateral tongue surface These anatomic differences likely influence the ecology of
(Fig. 4), S. mitis, S. mitis bv. 2, Streptococcus sp. clone DP009, these habitats and create microbial environmental differences.
Streptococcus sp. clone FN051, S. australis, G. adiacens, G. On the hard palate, the predominant bacterial species included
hemolysans, and Veillonella spp. predominated. It is interesting S. mitis, S. mitis biovar 2, Streptococcus sp. clone FN051, Strepto-
that there were considerable differences in the bacterial pro- coccus infantis, Granulicatella elegans, G. hemolysans, and Neisse-
files of the tongue dorsum and the lateral tongue surface. For ria subflava (Fig. 5). On the soft palate, S. mitis, other cultivable
VOL. 43, 2005 NORMAL ORAL BACTERIAL FLORA 5727

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 7. Bacterial profiles of the tonsils of healthy subjects. Distribution and levels of bacterial species/phylotypes among five subjects are as
described for Fig. 1. Novel phylotypes identified in this study are indicated in bold. GenBank accession numbers are provided. Marker bar
represents a 5% difference in nucleotide sequences.
5728 AAS ET AL. J. CLIN. MICROBIOL.

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 8. Bacterial profiles of the tooth surfaces of healthy subjects. Distribution and levels of bacterial species/phylotypes among five subjects
are as described for Fig. 1. Novel phylotypes identified in this study are indicated in bold. GenBank accession numbers are provided. Marker bar
represents a 5% difference in nucleotide sequences.
VOL. 43, 2005 NORMAL ORAL BACTERIAL FLORA 5729

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 9. Bacterial profiles of the subgingival plaque of healthy subjects. Distribution and levels of bacterial species/phylotypes among five
subjects are as described for Fig. 1. Novel phylotypes identified in this study are indicated in bold. GenBank accession numbers are provided.
Marker bar represents a 5% difference in nucleotide sequences.

and not-yet-cultivable species of Streptococcus, G. adiacens and G. adiacens, Actinomyces sp. clone BL008, and Abiotrophia defectiva
hemolysans were predominant (Fig. 6). The tonsil bacterial flora were often detected (Fig. 8). Finally, in subgingival plaque, several
(Fig. 7) was rather diverse—some subjects had Prevotella and species of Streptococcus and Gemella were often found (Fig. 9).
Porphyromonas spp. (subjects 1 and 5), and others had Firmicutes The number of cultivable and not-yet-cultivable species de-
species (subjects 2, 3, and 4). On the tooth surface, several species tected in each subject for each site is also shown in Fig. 1
of Streptococcus, including Streptococcus sp. clone EK048, S. san- through 9. For example, in Fig. 1, only four species (S. mitis,
guinis, and S. gordonii, and Rothia dentocariosa, G. hemolysans, G. S. mitis bv. 2, A. defectiva, and G. hemolysans) were detected
5730 AAS ET AL. J. CLIN. MICROBIOL.

TABLE 1. Number of predominant bacterial species per site and subject

Total no. of species/site
Subject Total no. of
no. Maxillary Tongue Tongue Hard Soft Tooth species/subject
Buccal Tonsils Subgingival
vestibule dorsum lateral palate palate surface

1 20 5 23 14 21 16 28 27 14 66
2 12 6 13 18 18 20 14 12 6 45
3 11 9 10 9 15 14 22 21 22 72
4 5 7 10 8 4 6 11 12 4 34
5 4 3 17 20 6 13 18 16 21 64

Total 32 15 40 34 42 38 59 52 47

on the buccal epithelium in subject 5. For convenience, the 1, with 28 predominant species; collectively, 57 species were
number of bacterial species detected in each oral site for each detected on the tonsils (Table 1; Fig. 7). In contrast, the max-
subject is summarized in Table 1. Surprisingly, the highest illary anterior vestibule had the lowest diversity of the bacterial
number of different species was found on the tonsils in subject flora compared with the other sites, with as few as three to nine

Downloaded from jcm.asm.org by on May 21, 2008

FIG. 10. Site specificity of predominant bacterial species in the oral cavity. In general, bacterial species or phylotypes were selected on
the basis of their detection in multiple subjects for a given site. Distributions of bacterial species in oral sites among subjects are indicated by the
columns of boxes to the right of the tree as follows: not detected in any subject (clear box), ⬍15% of the total number of clones assayed (yellow
box), ⱖ15% of the total number of clones assayed (green box). The 15% cutoff for low and high abundance was chosen arbitrarily. Marker bar
represents a 10% difference in nucleotide sequences.
VOL. 43, 2005 NORMAL ORAL BACTERIAL FLORA 5731

species detected among the subjects (Table 1; Fig. 2). A com- forsythia, and Treponema denticola, were not detected in any
mon question asked is how many bacterial species are present sites tested. In addition, the bacterial flora commonly thought
in the oral cavity of a single individual. The last column of to be involved in dental caries and deep dentin cavities, rep-
Table 1 indicates the number of predominant species in each resented by Streptococcus mutans, Lactobacillus spp., Bifid-
healthy individual from all nine different sites; thus, the num- obacterium spp., and Atopobium spp., were not detected in
bers ranged from a low of 34 in subject 4 to a high of 72 in supra- and subgingival plaques from clinically healthy teeth. A
subject 3. more detailed description of bacterial species associated with
Overall, cultivable and not-yet-cultivable species of Gemella, oral disease is discussed elsewhere (17). As noted in this study
Granulicatella, Streptococcus, and Veillonella were commonly on healthy subjects, some species are site specific at one or
detected in most sites. S. mitis was the most commonly found multiple sites, while other species are subject specific. As much
species in essentially all sites and subjects (Fig. 1 through 9). In as 60% of the species detected are not presently cultivable.
one subject, 79% of the clones identified were S. mitis (data not Overall, there are still more species to be discovered, although
shown in figures). Note that in this study, S. mitis and Strepto- the number of new species is beginning to reach saturation.
coccus pneumoniae have been grouped together. Some mem- As we have previously asserted (21), to rigorously assess the
bers of the mitis group share more than 99% 16S rRNA se- association of specific species or phylotypes with oral health or
quence similarity, although DNA-DNA similarity values for disease, it is necessary to analyze larger numbers of clinical
the entire chromosomes are estimated to be less than 60% samples for the levels of essentially all oral bacteria in well-
(14). Both S. mitis and Streptococcus oralis have been associ- controlled clinical studies. The bacterial complexes involved in
ated with bacterial endocarditis, especially in patients with periodontal disease as defined by Socransky et al. (27) were
prosthetic valves (9). In addition, they are now recognized as based on the microbial analyses of 185 subjects representing
frequent causes of infection in immunocompromised patients, about 13,000 plaque samples using DNA probes to 40 bacterial

Downloaded from jcm.asm.org by on May 21, 2008

particularly immediately after tissue transplants and in neutro- cultivable species in checkerboard hybridization assays. We are
penic cancer patients (30). G. adiacens, often considered an currently developing DNA probes for approximately 500
opportunistic pathogen, was also detected at all sites in the known bacterial species and novel phylotypes for use in similar
healthy oral cavity. G. adiacens isolates have been reported studies. These DNA probes are being developed for use in the
from bacteremia/septicemia in patients with infective endocar- checkerboard hybridization assay (3, 22, 23) and DNA mi-
ditis/atheroma (33) and in primary bacteremia in patients with croarray formats (S. K. Boches, A. M. Lee, B. J. Paster, and
neutropenic fever (33). A high mortality rate for endocarditis F. E. Dewhirst, J. Dent. Res. 83:abstract 2263, 2004). Our
by G. adiacens has been reported (5). intent was to first establish the full bacterial diversity of the
Figure 10 represents the overall summary of this study— oral cavity and then to determine variation and reproducibility
namely, that there are emerging bacterial profiles that help using the microarrays. Consequently, it will be relatively easy
define the healthy oral cavity. As already discussed, several to compare the bacterial composition of a statistically signifi-
species, such as S. mitis and G. adiacens, were detected in most cant number of samples to more precisely identify those spe-
or all oral sites, whereas several species were quite site specific. cies that are associated with health and with oral disease from
For example, R. dentocariosa, Actinomyces spp., S. sanguinis, different anatomic sites. It is necessary to first define the bac-
S. gordonii, and A. defectiva appeared to preferentially colonize terial flora of the healthy oral cavity before we can determine
the teeth, while S. salivarius was found mostly on the tongue the role of oral bacteria in disease.
dorsum. Some species appeared to have a predilection for soft
tissue, e.g., S. sanguinis and S. australis did not colonize the ACKNOWLEDGMENTS
teeth or subgingival crevice. S. intermedius preferentially colo-
This study was supported by NIH grant DE-11443 from the National
nized the subgingival plaque in most of the subjects but was not Institute of Dental and Craniofacial Research and grants from the
detected in most other sites. On the other hand, Neisseria spp. Dental Faculty, University of Oslo, Norway.
were not found in subgingival plaque but were present in most
REFERENCES
other sites. Simonsiella muelleri colonized only the hard palate.
Indeed, S. muelleri was initially isolated from the human hard 1. Albandar, J. M., J. A. Brunelle, and A. Kingman. 1999. Destructive peri-
odontal disease in adults 30 years of age and older in the United States,
palate (32), although it has been isolated from a neonate with 1988–1994. J. Periodontol. 70:13–29.
dental cyst and early eruption of teeth (31). Several Prevotella 2. Beck, J., R. Garcia, G. Heiss, P. S. Vokonas, and S. Offenbacher. 1996.
Periodontal disease and cardiovascular disease. J. Periodontol. 67:1123–
species were detected in most sites, but only in one or two 1137.
subjects. For example, P. melaninogenica and Prevotella sp. 3. Becker, M. R., B. J. Paster, E. J. Leys, M. L. Moeschberger, S. G. Kenyon,
clone BE073 were abundant in seven out of nine sites of one J. L. Galvin, S. K. Boches, F. E. Dewhirst, and A. L. Griffen. 2002. Molecular
analysis of bacterial species associated with childhood caries. J. Clin. Micro-
subject and were detected sporadically in other subjects (Fig. 10). biol. 40:1001–1009.
Prevotella sp. clone HF050 was found in the maxillary anterior 4. Berbari, E. F., F. R. Cockerill III, and J. M. Steckelberg. 1997. Infective
vestibule of one subject, dominating the bacterial flora as 44% endocarditis due to unusual or fastidious microorganisms. Mayo Clin. Proc.
72:532–542.
of the clones (Fig. 2 and 10). This clone was also found in lower 5. Bouvet, A., and J. F. Acar. 1984. New bacteriological aspects of infective
proportions on the soft palate and tonsils of another subject endocarditis. Eur. Heart J. 5(Suppl. C):45–48.
6. Buduneli, N., H. Baylas, E. Buduneli, O. Turkoglu, T. Kose, and G. Dahlen.
(Fig. 6, 7, and 10). 2005. Periodontal infections and pre-term low birth weight: a case-control
In conclusion, there is a distinctive bacterial flora in the study. J. Clin. Periodontol. 32:174–181.
healthy oral cavity which is different from that of oral disease. 7. Cole, J. R., B. Chai, R. J. Farris, Q. Wang, S. A. Kulam, D. M. McGarrell,
G. M. Garrity, and J. M. Tiedje. 2005. The Ribosomal Database Project
For example, many species specifically associated with peri- (RDP-II): sequences and tools for high-throughput rRNA analysis. Nucleic
odontal disease, such as Porphyromonas gingivalis, Tannerella Acids Res. 33:D294–D296.
5732 AAS ET AL. J. CLIN. MICROBIOL.

8. Dodman, T., J. Robson, and D. Pincus. 2000. Kingella kingae infections in V. A. Levanos, M. A. Tamer, R. L. Ericson, C. N. Lau, and F. E. Dewhirst.
children. J. Paediatr. Child. Health 36:87–90. 2002. Prevalent bacterial species and novel phylotypes in advanced noma
9. Douglas, C. W., J. Heath, K. K. Hampton, and F. E. Preston. 1993. Identity lesions. J. Clin. Microbiol. 40:2187–2191.
of viridans streptococci isolated from cases of infective endocarditis. J. Med. 23. Paster, B. J., M. K. Russell, T. Alpagot, A. M. Lee, S. K. Boches, J. L. Galvin,
Microbiol. 39:179–182. and F. E. Dewhirst. 2002. Bacterial diversity in necrotizing ulcerative peri-
10. Dymock, D., A. J. Weightman, C. Scully, and W. G. Wade. 1996. Molecular odontitis in HIV-positive subjects. Ann. Periodontol. 7:8–16.
analysis of microflora associated with dentoalveolar abscesses. J. Clin. Mi- 24. Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method
crobiol. 34:537–542. for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406–425.
11. Gibbons, R. J. 1989. Bacterial adhesion to oral tissues: a model for infectious 25. Sakamoto, M., M. Umeda, I. Ishikawa, and Y. Benno. 2000. Comparison of
diseases. J. Dent. Res. 68:750–760. the oral bacterial flora in saliva from a healthy subject and two periodontitis
12. Gibbons, R. J., D. M. Spinell, and Z. Skobe. 1976. Selective adherence as a patients by sequence analysis of 16S rDNA libraries. Microbiol. Immunol.
determinant of the host tropisms of certain indigenous and pathogenic bac- 44:643–652.
teria. Infect. Immun. 13:238–246. 26. Scannapieco, F. A. 1999. Role of oral bacteria in respiratory infection. J.
13. Jukes, T. H., and C. R. Cantor. 1969. Evolution of protein molecules, vol. 3, Periodontol. 70:793–802.
p. 21–132. In H. N. Munro (ed.), Mammalian protein metabolism. Academic 27. Socransky, S. S., A. D. Haffajee, M. A. Cugini, C. Smith, and R. L. Kent, Jr.
Press, Inc., New York, N.Y. 1998. Microbial complexes in subgingival plaque. J. Clin. Periodontol. 25:
14. Kawamura, Y., X. G. Hou, F. Sultana, H. Miura, and T. Ezaki. 1995. De- 134–144.
termination of 16S rRNA sequences of Streptococcus mitis and Streptococcus
28. Van de Peer, Y., and R. De Wachter. 1994. TREECON for Windows: a
gordonii and phylogenetic relationships among members of the genus Strep-
software package for the construction and drawing of evolutionary trees for
tococcus. Int. J. Syst. Bacteriol. 45:406–408.
the Microsoft Windows environment. Comput. Appl. Biosci. 10:569–570.
15. Kazor, C. E., P. M. Mitchell, A. M. Lee, L. N. Stokes, W. J. Loesche, F. E.
Dewhirst, and B. J. Paster. 2003. Diversity of bacterial populations on the 29. Wade, W. G., D. A. Spratt, D. Dymock, and A. J. Weightman. 1997. Molec-
tongue dorsa of patients with halitosis and healthy patients. J. Clin. Micro- ular detection of novel anaerobic species in dentoalveolar abscesses. Clin.
biol. 41:558–563. Infect. Dis. 25(Suppl. 2):S235–S236.
16. Kroes, I., P. W. Lepp, and D. A. Relman. 1999. Bacterial diversity within the 30. Whatmore, A. M., A. Efstratiou, A. P. Pickerill, K. Broughton, G. Woodard,
human subgingival crevice. Proc. Natl. Acad. Sci. USA 96:14547–14552. D. Sturgeon, R. George, and C. G. Dowson. 2000. Genetic relationships
17. Kumar, P. S., A. L. Griffen, J. A. Barton, B. J. Paster, M. L. Moeschberger, between clinical isolates of Streptococcus pneumoniae, Streptococcus oralis,
and E. J. Leys. 2003. New bacterial species associated with chronic peri- and Streptococcus mitis: characterization of “atypical” pneumococci and or-
odontitis. J. Dent. Res. 82:338–344. ganisms allied to S. mitis harboring S. pneumoniae virulence factor-encoding
genes. Infect. Immun. 68:1374–1382.

Downloaded from jcm.asm.org by on May 21, 2008

18. Liesack, W., H. Weyland, and E. Stackebrandt. 1991. Potential risk of gene
amplification by PCR as determined by 16S rDNA analysis of a mixed- 31. Whitehouse, R. L., H. Jackson, M. C. Jackson, and M. M. Ramji. 1987.
culture of strict barophilic bacteria. Microb. Ecol. 21:191–198. Isolation of Simonsiella sp. from a neonate. J. Clin. Microbiol. 25:522–525.
19. Mager, D. L., L. A. Ximenez-Fyvie, A. D. Haffajee, and S. S. Socransky. 2003. 32. Whitehouse, R. L., H. Merrill, M. C. Jackson, and H. Jackson. 1990. A stable
Distribution of selected bacterial species on intraoral surfaces. J. Clin. Peri- variant of Simonsiella muelleri with unusual colonial and cellular morphol-
odontol. 30:644–654. ogy. J. Bacteriol. 172:1673–1675.
20. Offenbacher, S., H. L. Jared, P. G. O’Reilly, S. R. Wells, G. E. Salvi, H. P. 33. Woo, P. C., A. M. Fung, S. K. Lau, B. Y. Chan, S. K. Chiu, J. L. Teng, T. L.
Lawrence, S. S. Socransky, and J. D. Beck. 1998. Potential pathogenic Que, R. W. Yung, and K. Y. Yuen. 2003. Granulicatella adiacens and Abiotro-
mechanisms of periodontitis associated pregnancy complications. Ann. Peri- phia defectiva bacteraemia characterized by 16S rRNA gene sequencing.
odontol. 3:233–250. J. Med. Microbiol. 52:137–140.
21. Paster, B. J., S. K. Boches, J. L. Galvin, R. E. Ericson, C. N. Lau, V. A. 34. Wu, T., M. Trevisan, R. J. Genco, J. P. Dorn, K. L. Falkner, and C. T.
Levanos, A. Sahasrabudhe, and F. E. Dewhirst. 2001. Bacterial diversity in Sempos. 2000. Periodontal disease and risk of cerebrovascular disease: the
human subgingival plaque. J. Bacteriol. 183:3770–3783. first national health and nutrition examination survey and its follow-up study.
22. Paster, B. J., W. A. Falkler, Jr., C. O. Enwonwu, E. O. Idigbe, K. O. Savage, Arch. Intern. Med. 160:2749–2755.