Molecular Identification of Dialister pneumosintes in Subgingival Plaque of Humans

Abstract

Dialister pneumosintes is a nonfermentative, anaerobic, gram-negative rod that grows with small, circular, transparent, shiny, smooth colonies on blood agar. Even though D. pneumosintes has been recovered from deep periodontal pockets, little is known about the relationship between the organism and destructive periodontal disease. This study describes a rapid PCR method to identify D. pneumosintes in periodontal samples. The PCR identification method detected as little as 10 pg of D. pneumosintes DNA or about 1 to 10 cells without nonspecific amplification of various periodontopathic bacteria. Twelve of 22 subgingival samples from adult periodontitis lesions yielded D. pneumosintes either by culture or by PCR identification. In culture-positive samples, D. pneumosintes averaged 3.9% (0.001 to 10.8%) of total isolates. Studies are needed to delineate virulence factors of D. pneumosintes pertinent to periodontal disease.

Our laboratory has noticed the frequent presence of a previously unrecognized gram-negative anaerobic rod in adult periodontitis lesions and a virtual absence of the organism from healthy periodontal sites (anecdotal observations). The organism forms convex, clear, transparent, shiny, smooth colonies on blood agar. Conventional biochemical tests failed to identify the organism conclusively but were consistent with the diagnosis of Dialister pneumosintes. D. pneumosintes has previously been isolated from nasopharyngeal (13) and gingival (10) sites, but its occurrence in subgingival plaque is not commonly reported (12), possibly due to difficulties in culturing and identifying the species.

The aims of this study were (i) to provide a definitive identification of the D. pneumosintes-like organism found in primary cultures of subgingival plaque samples, (ii) to develop a PCR detection method for D. pneumosintes, and (iii) to compare the efficacies of cultural and PCR detection methods to identify D. pneumosintes in subgingival plaque samples of adult periodontitis lesions.

MATERIALS AND METHODS

Bacteria and culture conditions.

D. pneumosintes type strain ATCC 33048, nine clinical isolates of D. pneumosintes, and other test organisms were maintained on nonselective medium containing 4.3% brucella agar (BBL Microbiology Systems, Cockeysville, Md.), 0.3% Bacto Agar, 0.2% yeast extract, 5% defibrinated sheep blood, 0.2% hemolyzed sheep erythrocytes, 0.0005% hemin, and 0.00005% menadione. D. pneumosintes ATCC 33048 was purchased from the American Type Culture Collection (Manassas, Va.). Incubation took place at 37°C in an anaerobic chamber (Coy Laboratory Products, Ann Arbor, Mich.) containing 85% N₂–10% H₂–5% CO₂.

Bacterial cell suspensions were prepared from 5 to 7 days' growth on agar plates by washing the plates with dilution broth consisting of 0.5% NaCl (Sigma Chemical Co., St. Louis, Mo.), 0.25% Thiotone E-peptone (BBL Microbiology Systems), and 0.25% tryptose (Difco Laboratories, Detroit, Mich.). CFU were determined in 10-fold serially diluted bacterial suspensions plated on brucella blood agar.

Biochemical analysis.

An Api-20 Ana test kit (bioMerieux, Marcy l'Etoile, France) was employed according to the manufacturer's instructions. The kit tests for beta-galactosidase, arginine dihydrolase, lysine decarboxylase, ornithine decarboxylase, citrate utilization, H₂S production, urease, tryptophane deaminase, indole production, acetoin production, gelatinase, and glucose, mannose, inositol, sorbitol, and arabinose fermentation and oxidation. Briefly, bacterial cells grown on blood agar plates were transferred to test tubes containing sterile water; the bacterial cell concentration was then adjusted to an optical density equivalent to a 0.05 McFarland standard, and Api-20 Ana strips were inoculated and incubated anaerobically at 37°C for 24 to 48 h.

Cloning and DNA sequencing.

Cloning and DNA sequencing of the 16S rRNA gene were performed to confirm the identity of D. pneumosintes. Genomic DNA of oral strain 1022, presumptively identified as D. pneumosintes, was extracted as previously described (2). The genomic DNA was used as a template for PCR amplification using a pair of primers targeting universal sequences shared by most eubacteria (1).

The estimated ∼500-bp PCR product was gel purified using the QIAquick gel extraction kit according to the manufacturer's instructions (Qiagen Inc., Valencia, Calif.). The purified PCR product was ligated into a plasmid vector and transformed into Escherichia coli with a TA cloning kit dual promoter (Invitrogen Co., Carlsbad, Calif.). Transformed colonies were identified from Luria-Bertani medium plates (Difco Laboratories) containing ampicillin, X-Gal (5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside), and IPTG (isopropyl-β-d-thiogalactopyranoside), and the recombinant plasmid was extracted using the Wizard Plus Minipreps DNA purification system (Promega, Madison, Wis.). The cloned PCR product was sequenced by the USC Comprehensive Cancer Center, Los Angeles, Calif. The partial sequence of the 16S rRNA gene was compared to available 16S rRNA gene sequences by the sequence matching program of the Ribosomal Database Project (7) and was found to be highly similar to that of D. pneumosintes. Subsequently, a larger fragment of the 16S rRNA gene from D. pneumosintes strain 1022 was obtained by PCR with species-specific primers targeting the signature sequences of the 16S rRNA gene of D. pneumosintes ATCC 33048 (see “PCR” below). Sequencing of this larger 16S rRNA fragment also confirmed the D. pneumosintes species identity.

PCR.

A 16S rRNA-based PCR detection method was developed for rapid detection of D. pneumosintes, based on the 16S rRNA sequence of D. pneumosintes (7). PCR primers targeting signature sequences of D. pneumosintes were designed to be (5′ to 3′) TTC TAA GCA TCG CAT GGT GC and GAT TTC GCT TCT CTT TGT TG. The utility of these primers for specific detection of D. pneumosintes was verified with the Probe Match program (7).

Both bacterial genomic DNA and boiled bacterial cells were used as templates for PCR amplification. The bacterial genomic DNA was prepared by using the IsoQuick nucleic acid extraction kit (ORCA Research, Inc., Bothel, Wash.) by following the manufacturer's recommendation. The genomic DNA template was denatured in boiling water for 10 min and quickly chilled on ice, and 5 μl was used for PCR amplification. The template from boiled bacterial cells was prepared by suspending bacteria in boiling water for 10 min, followed by quick chilling on ice and centrifugation at 12,000 × g for 3 min to remove debris. The supernatant was used for PCR amplification.

PCR was performed as described previously (2). A total of 5 μl of template was added to 45 μl of reaction mixture containing 5 μl of 10× PCR buffer (Promega), 1.25 U of Taq DNA polymerase (Promega), a 0.2 mM concentration of each deoxyribonucleotide (Pharmacia LKB, Piscataway, N.J.), and 1.5 mM MgCl₂. Amplification was performed in a DNA thermal cycler (PTC-100; MJ Research, Boston, Mass.). The temperature profile included an initial step of 95°C for 2 min, followed by 36 cycles of 94°C for 30 s, 55°C for 1 min, and 72°C for 2 min and a final step of 72°C for 10 min. PCR products were analyzed by 1.5% agarose gel electrophoresis as described previously (1).

PCR specificity and detection limit.

The specificity of the PCR-based D. pneumosintes detection method was evaluated against a panel of common oral bacteria. The oral test microorganisms included Porphyromonas gingivalis, Prevotella intermedia, Campylobacter species, Eubacterium species, Fusobacterium species, and Peptostreptococcus micros. To test for possible cross-reactivity, the genomic DNA from nontarget organisms was extracted as described above and diluted to approximately 1 to 2 μg/ml, and aliquots of 5 μl were obtained for PCR amplification with primers for D. pneumosintes.

The detection limit of the PCR method was determined using templates prepared from (i) serially diluted genomic DNA of D. pneumosintes type strain ATCC 33048 and clinical strain 1022 and (ii) serially diluted cell suspensions of D. pneumosintes strain ATCC 33048 and clinical strain 1022. The lowest level of genomic DNA or bacterial cells detected by the PCR method was defined as the detection limit.

The efficacy of the PCR method was further evaluated using samples containing a constant number of nontarget cells spiked with a variable number of cells of D. pneumosintes strain ATCC 33048 or strain 1022 to simulate clinical plaque samples. Five microliters of bacterial suspensions containing 1 to 10,000 cells was added to 5 μl of a bacterial suspension containing 5 × 10⁵ nontarget cells. Ten microliters of the bacterial cell mixture was placed in boiling water for 10 min, quickly chilled on ice, and centrifuged, and the supernatant was used for PCR amplification as described above.

Correlation between cultural identification based on colony morphology and PCR identification of D. pneumosintes from primary cultures.

Twenty-five colonies from primary culture (five colonies from each of five periodontitis patients) presumptively identified as D. pneumosintes were subcultured to ensure purity. Templates were prepared from three to four colonies of each pure culture and subjected to PCR identification using D. pneumosintes-specific primers as described above. PCR products from the amplification of these 25 isolates were also analyzed by restriction enzyme digestion using EcoRI and ApaI (Sigma Chemical Co.), by following the manufacturer's instructions. Briefly, approximately 0.5 to 1 μg of PCR product was digested either with EcoRI at 37°C or with ApaI at 30°C for 1 h. Enzymatic reactions were arrested with 0.5 M EDTA and the digested products were analyzed by 3% agarose gel electrophoresis.

Detection of D. pneumosintes in subgingival plaque of adult periodontitis lesions.

Comparison between culture recovery and PCR detection of subgingival D. pneumosintes was carried out in 22 adult periodontitis samples submitted to the Oral Microbiology Testing Laboratory by extramural dentists. Details of the protocol for subgingival sample collection have been described elsewhere (1). Briefly, samples were collected by inserting sterile paper points to the depth of each of three periodontal pockets and retaining them therein for 15 s. The three paper points were then combined in one vial containing VMGA III transport medium (8). VMGA III samples were warmed to 37°C and subjected to vortex mixing to ensure even distribution of bacteria in the transport medium, and aliquots of 100 μl were used for either PCR or cultural identification.

For cultural identification of D. pneumosintes, a 100-μl sample was serially diluted in 10-fold steps in dilution broth; the dilutions were inoculated onto brucella blood agar plates and incubated in the Coy anaerobic chamber for 5 to 7 days. Proportional recovery of D. pneumosintes was determined by comparing the colony counts of the organism to total viable counts on the blood agar plates.

In the PCR assay, a 100-μl sample was washed three times with 0.3 ml of Tris-EDTA buffer. After the final wash, the bacterial pellet was resuspended in 0.3 ml of distilled water, boiled for 10 min, quickly chilled on ice, and centrifuged at 14,000 × g for 3 min to remove large debris. Five microliters of the supernatant was used for PCR analysis.

RESULTS

Colony morphology of D. pneumosintes in primary cultures.

Figure 1 shows the colony morphology of D. pneumosintes on a blood agar plate from a primary culture of subgingival plaque. D. pneumosintes colonies appeared flat to convex, transparent to pinkish, shiny, smooth, and approximately 0.5 mm in diameter, and each showed several white spots at the center. Colonies of Eubacterium species, which resembled those of D. pneumosintes, were more coarse and opaque, and each exhibited a single raised, white dot at the center. Preliminary characterization showed D. pneumosintes to be a gram-negative rod that was obligately anaerobic, nitrate negative, bile sensitive, kanamycin susceptible, vancomycin resistant, and colistin resistant.

FIG. 1.

D. pneumosintes colony morphology. The colony appears flat, transparent, shiny, and smooth on blood agar. The photograph was obtained by using a stereoscopic microscope at a ×4 magnification.

Biochemical analysis.

D. pneumosintes type strain ATCC 33048 and nine clinical isolates of the species yielded negative reactions in all Api-20 Ana tests.

Cloning and sequencing of the 16S rRNA gene.

Combining the two fragments of the 16S rRNA gene of strain 1022 yielded a contiguous 1.1-kb fragment. Sequencing of the 1.1-kb fragment yielded only 10 unknown nucleotides. Analysis of the 1.1-kb fragment by the Similarity Matrix Program from the Ribosomal Database Project showed a similarity value of 0.985 between D. pneumosintes clinical strain 1022 and type strain ATCC 33048 and a total of 12 nucleotide mismatches between the two strains. The analysis also showed high similarity values between D. pneumosintes strain 1022 and two deposited sequences of Eubacterium species: strains SC-5 and SC-3D (similarity values of 0.986 and 0.986, respectively). Upon further inquiry, it was revealed that these two strains actually belong to the D. pneumosintes species (personal communication from the depositor of the strains).

PCR detection of D. pneumosintes.

PCR amplification of templates prepared from genomic DNA of D. pneumosintes strain ATCC 33048 and clinical strain 1022 resulted in a single 1.1-kb band (Fig. 2). No amplification product was detected with the nontarget organisms tested (data not shown).

FIG. 2.

PCR amplification of D. pneumosintes genomic DNA shows a 1,105-bp product. Lane M, 1-kb ladder marker; lane 1, ATCC 33048; lane 2, clinical strain 1022.

The detection limit of the PCR method was 10 pg of genomic DNA of D. pneumosintes (Fig. 3) or one D. pneumosintes bacterial cell (Fig. 4). In spiked laboratory samples, the PCR identification method detected as few as one D. pneumosintes cell in a background of 5 × 10⁵ non-D. pneumosintes cells.

FIG. 3.

PCR sensitivity determined using D. pneumosintes genomic DNA as a template. Lane M, 1-kb ladder marker; lanes 1 to 5, 10,000, 1,000, 100, 10, and 1 pg of strain ATCC 33048 DNA as a template, respectively; lanes 7 to 11, 10,000, 1,000, 100, 10, and 1 pg of strain 1022 DNA as a template, respectively.

FIG. 4.

PCR sensitivity determined using D. pneumosintes cells. Lane M, 1-kb ladder marker; lanes 1 to 5, 10,000, 1,000, 100, 10, and 1 cell of strain ATCC 33048 as a template, respectively; lanes 7 to 11, 10,000, 1,000, 100, 10, and 1 cell of strain 1022 as a template, respectively.

Correlation between cultural identification and PCR identification of individual D. pneumosintes colonies.

All 25 colonies presumptively identified as D. pneumosintes based on colony morphology were confirmed by PCR to be correctly identified. Also, restriction enzyme digestion of the 1.1-kb PCR product produced fragments of 509 and 596 bp with EcoRI and 763 and 342 bp with ApaI, as would be expected for D. pneumosintes (Fig. 5).

FIG. 5.

PCR product of D. pneumosintes clinical isolates analyzed by restriction enzyme digestion. Lane M, 1-kb ladder marker. (A) Lane 1, undigested 1,105-bp PCR product of D. pneumosintes; lanes 2 to 11, EcoRI digests of the PCR product of 10 D. pneumosintes clinical isolates showing the predicted sizes of 596 and 509 bp. (B) Lane 1, undigested 1,105-bp PCR product; lanes 2 to 11, ApaI digests of the PCR product of 10 D. pneumosintes clinical isolates showing the predicted sizes of 763 and 342 bp.

Cultural and PCR detection of D. pneumosintes in adult periodontitis.

In 22 subgingival plaque samples from adult periodontitis lesions, D. pneumosintes was identified by culture in 8 samples and by PCR in 9 samples (Table 1). Five samples were positive by both culture and PCR, three were culture positive and PCR negative, and four were negative by culture and positive by PCR. In culture-positive samples, D. pneumosintes averaged 3.9% (0.001 to 10.8%) of total isolates.

TABLE 1.

Cultural and PCR detection of D. pneumosintes in adult periodontitis samples

Patient no.	Culture result (% of total isolates)	PCR result
1	7.6	Positive
2	5.5	Positive
3	10.3	Positive
4	7.1	Positive
5	0.2	Positive
6	0	Positive
7	0	Positive
8	0	Positive
9	0	Positive
10	0.2	Negative
11	0.2	Negative
12	0.01	Negative
13	0	Negative
14	0	Negative
15	0	Negative
16	0	Negative
17	0	Negative
18	0	Negative
19	0	Negative
20	0	Negative
21	0	Negative
22	0	Negative

DISCUSSION

D. pneumosintes is a small, nonfermentative, gram-negative rod that grows with punctiform, circular, convex, clear, transparent, shiny, smooth colonies on blood agar (6). The species was first isolated from nasopharyngeal secretions of patients during the flu epidemic of 1918 to 1921 (13). Moore et al. (10, 11) identified D. pneumosintes in subgingival sites of children and young adults with gingivitis and periodontitis. D. pneumosintes has also been recovered from pus and body fluids (4) and from human bite wounds (3). The organism has shown pathogenic potential in various body sites, including the lung, brain, and dental root canals (12).

Olitsky and Gates originally named the organism Bacterium pneumosintes (13), but it was placed in the genus Dialister, and in 1970 it was included in the genus Bacteroides (5). As a result of the taxonomic revision of the genus Bacteroides in the 1980s, Shah and Collins proposed removing the organism from this genus (14). In 1994, Moore and Moore transferred the species back to the genus Dialister (9).

Although the colony morphology of D. pneumosintes resembles that of Eubacterium species, the negative gram stain reaction excludes the Eubacterium species. Lack of reactivity in conventional biochemical tests complicated initial classification of D. pneumosintes. However, the colony morphology suggested the diagnosis of D. pneumosintes. The sequence of a 1.1-kb partial 16S rRNA gene, comprising 70% of the complete 16S rRNA gene, was compared to that of the D. pneumosintes type strain, ATCC 33048. The results showed a similarity value of 0.985 with and only 12 nucleotide mismatches from the sequence of D. pneumosintes strain ATCC 33048. Slight variations in the 16S rRNA gene sequences among strains of the same species are not unusual (7). The variability in the D. pneumosintes 16S rRNA gene warrants further examination.

Using signature sequences from the 16S rRNA gene of the D. pneumosintes type strain, ATCC 33048, we developed a PCR detection method for the bacterium. PCR-based identification was able to detect as little as 10 pg of DNA or 1 to 10 cells with no nonspecific amplification of other species. The PCR detection method also validated our presumptive identification of D. pneumosintes, since all D. pneumosintes-like isolates from primary culture showed the expected 1.1-kb amplicon. Restriction enzyme digestion of PCR products yielded the predicted fragment sizes, further validating our presumptive identification of the species.

This study compared culture and PCR detection of D. pneumosintes in human subgingival plaque samples. Of 22 samples, 3 were culture positive and PCR negative, and 4 were culture negative and PCR positive. The results for the remaining 15 samples were either positive (5 samples) or negative (10 samples) by both identification methods. Our 16S rRNA PCR detection method identified as few as one D. pneumosintes cell in artificially constructed test samples and was expected to perform better than culture. However, because the present PCR procedure employed as little as 5 of 300 μl of washed and reconstituted clinical sample, whereas culture used the entire 100 μl of unwashed sample, some D. pneumosintes cells may have escaped detection in the PCR assay. Increasing the working sample amounts or concentrating the clinical sample prior to PCR amplification might improve the sensitivity. However, detecting infinitesimal amounts of subgingival D. pneumosintes may be of little significance in determining the organism's importance in periodontal disease.

A few previous studies have reported the presence of D. pneumosintes in human periodontal disease. The organism has been isolated in low proportions from children and young adults with gingivitis (10, 11). Because of D. pneumosintes' slow growth and the difficulties in isolating the strictly anaerobic organism and distinguishing it from Eubacterium species in primary culture, it is possible that earlier cultural studies of the subgingival microbiota may have missed the bacterium. Moore and Moore (12) reported increased proportional recovery of D. pneumosintes with increased severity of periodontal disease, but the organism comprised less than 2% of subgingival isolates. This study showed four subgingival specimens containing more than 5% cultivable D. pneumosintes, suggesting that D. pneumosintes may be a numerically significant organism in the subgingival plaque of some adult periodontitis patients.

In summary, this study showed a frequent presence of D. pneumosintes in human subgingival plaque of periodontitis lesions. The PCR-based detection method for D. pneumosintes developed in this study may help delineate the relationship between the species and destructive periodontal disease.