search for


Composition and Diversity of Salivary Microbiome Affected by Sample Collection Method
J Oral Med Pain 2022;47:10-26
Published online March 30, 2022;
© 2022 Korean Academy of Orofacial Pain and Oral Medicine

Yeon-Hee Lee1, Ji-Youn Hong2, Gi-Ja Lee3

1Department of Orofacial Pain and Oral Medicine, Kyung Hee University Dental Hospital, Kyung Hee University School of Dentistry, Seoul, Korea
2Department of Periodontology, Periodontal-Implant Clinical Research Institute, Kyung Hee University School of Dentistry, Seoul, Korea
3Department of Biomedical Engineering, Kyung Hee University, Seoul, Korea
Correspondence to: Yeon-Hee Lee
Department of Orofacial Pain and Oral Medicine, Kyung Hee University Dental Hospital, 26 Kyungheedae-ro, Dongdaemun-gu, Seoul 02447, Korea
Tel: +82-2-958-9454
Fax: +82-2-968-0588

This work was supported by a National Research Foundation of Korea funded by the Korean government under Grant number NRF/2020R1F1A1070072 to (Y.H.L); and the Korea Medical Device Development Fund funded by the Korean government (the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, and the Ministry of Food and Drug Safety, Republic of Korea) under Project number KMDF_PR_20200901_0023, 9991006696.
Received December 29, 2021; Revised January 14, 2022; Accepted January 14, 2022.
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Purpose: The purpose of this study was to investigate whether various saliva collection methods affect the observed salivary microbiome and whether microbiomes of stimulated and unstimulated saliva and plaque differ in richness and diversity.
Methods: Seven sampling methods for unstimulated saliva, stimulated saliva, and plaque samples were applied to six orally and systemically healthy participants. Bacterial 16S ribosomal RNA genes of 10 major oral bacterial species, namely, Aggregatibacter actinomycetemcomitans, Porphyromonas gingivalis, Tannerella forsythia, Treponema denticola, Fusobacterium nucleatum, Prevotella intermedia, Prevotella nigrescens, Streptococcus mitis, Streptococcus sobrinus, and Lactobacillus casei, were analyzed by real-time polymerase chain reaction. We comprehensively examined the dependence of the amount of bacterial ribosomal DNA (rDNA), bacterial-community composition, and relative abundance of each species on sample collection methods.
Results: There were significant differences in the bacterial rDNA copy number depending on the collection method in three species: F. nucleatum, P. nigrescens, and S. mitis. The species with the highest richness was S. mitis, with the range from 89.31% to 100.00%, followed by F. nucleatum, P. nigrescens, T. denticola, T. forsythia, and P. intermedia, and the sum of the proportions of the remaining five species was less than 1%. The species with the lowest observed richness was P. gingivalis (<0.1%). The Shannon diversity index was the highest in unstimulated saliva collected with a funnel (4.449). The Shannon diversity index was higher in plaque samples (3.623) than in unstimulated (3.171) and stimulated (3.129) saliva and in mouthwash saliva samples (2.061).
Conclusions: The oral microbial profile of saliva samples can be affected by sample collection methods, and saliva differs from plaque in the microbiome. An easy and rapid technique for saliva collection is desirable; however, observed microbial-community composition may more accurately reflect the actual microbiome when unstimulated saliva is assayed.
Keywords : Bacteria; Methods; Microbiome; Oral cavity; Saliva; Shannon diversity index

The composition of oral microflora is highly complex, diverse, and habitat dependent. Interactions between the resident microbiota and the host are believed to be strongly involved in the maintenance of oral health [1]. Local alterations of the oral microbiota in relation to ecological perturbations have been regarded as a prerequisite for the development of dental caries, periodontitis, and oral cancer [2]. In addition, an imbalance and dysbiosis of oral microbial flora contribute to systemic and oral diseases. The oral microbiota is also closely related to systemic diseases, including rheumatoid arthritis, adverse pregnancy outcomes, several types of cancer, and cardiovascular disease [3]. Recently, it was widely recognized that good oral health is associated with good general health. Although good oral health is based on a good oral microbiota and microbial interactions, efforts were made only recently to establish a standard for sampling methods for the oral microbiota.

The oral cavity is one of the most complex microbial habitats in the human body. Approximately 700 species of microorganisms are present in the human oral cavity [2]. Nonetheless, only ~50% of the ~700 species of oral microorganisms have been cultivated and named [4]. The oral cavity contains a complex environment that encompasses small distinct habitats, such as teeth, the buccal mucosa, soft and hard palate, and tongue, which form a species-rich heterogeneous ecological system [5]. Therefore, the observed composition of the oral bacterial community may depend on the method of sampling of oral bacteria and on sampling sites in the oral cavity. Here, we wanted to evaluate the dependence of the observed oral microbiome on the sampling method in healthy individuals prior to a study on patients with oral or systemic diseases.

Saliva is an attractive medium for studies on biomarkers of oral health and disease for several reasons, e.g., saliva collection is noninvasive and rapid and saliva is safe to handle, easy to transport and store, and inexpensive. Notably, several cross-sectional studies involving stimulated-saliva samples have reported salivary bacterial profiles that distinguish patients with periodontitis, patients with dental caries, and orally healthy individuals [6,7], suggesting that salivary bacterial profiles may serve as a biomarker for the screening of a population for an oral disease at preclinical stages. To the best of our knowledge, no study has addressed similarities of the salivary microbiota by comparing oral microbiomes determined in unstimulated- and stimulated-saliva samples collected from the same individuals and by comparing the observed oral microbiomes between saliva and plaque samples obtained by various techniques.

On the other hand, very few studies have addressed the dependence of observed bacterial richness and diversity on the methods of collection of saliva samples. The collection of stimulated-saliva samples is significantly faster and more comfortable for the patient than the collection of an unstimulated-saliva sample; this observation may lend support to the use of stimulated-saliva samples for the screening of larger populations. Nevertheless, a recent study on two healthy individuals revealed higher bacterial diversity in unstimulated-saliva samples than in stimulated-saliva samples [8]. In contrast, some research articles indicate that stimulated-saliva samples may be more useful than unstimulated-saliva samples for the identification of specific oral bacterial taxa such as Streptococcus mutans and Aggregatibacter actinomycetemcomitans [9,10]. S. mutans, Streptococcus sobrinus, and Lactobacillus spp. are representative bacteria known to be associated with dental caries.

Technological advances have led to the development of many techniques for detecting and quantifying periodontal pathogens, including microbiological culture, enzymatic assays, DNA-DNA hybridization, immunoassays, and polymerase chain reaction (PCR) assays [11]. Nonetheless, most of these approaches are time-consuming and cannot accurately quantify periodontal pathogens. Some time ago, real-time PCR (i.e., quantitative PCR, qPCR) was developed to overcome these limitations, allowing for accurate quantification with higher sensitivity, specificity, simplicity, and rapidity [12]. Thus, highly sensitive microbiological detection techniques, such as real-time PCR, can be employed to identify and quantify oral bacteria in saliva samples [13].

The purposes of this study were i) to characterize the oral microbiome; ii) to determine ribosomal DNA (rDNA) copy numbers of 10 major oral microbes (A. actinomycetemcomitans, Porphyromonas gingivalis, Tannerella forsythia, Treponema denticola, Fusobacterium nucleatum, Prevotella intermedia, Prevotella nigrescens, Streptococcus mitis, S. sobrinus, and Lactobacillus casei) in saliva and plaque samples using real-time PCR assays in periodontally healthy young adults; and iii) to investigate whether there is a difference in bacterial profiles depending on the sampling method. In this work, we aimed to analyze the oral microbiome of saliva and plaque collected from six healthy individuals over 20 years of age. The richness of the bacterial community was analyzed by real-time PCR, and the Shannon diversity index was used to evaluate diversity, and the results can have crucial implications as to which method would be advantageous.


1. The Clinical Cohort and Patient Selection

This study was carried out in strict accordance with the recommendations with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans. All participants provided informed consent, and all protocols were approved by the Kyung Hee University Dental Hospital Institutional Review Board (IRB no. KH-DT20030).

Inclusion criteria were as follows: medically healthy adults with a healthy periodontal condition, non-smoking, having lost <2 teeth in permanent dentition, able to voluntarily read and judge the consent form, and able to participate. Individuals taking drugs that affect salivation, including psychiatric medications (such as antianxiety or sleeping pills) and antibiotics, were excluded. Pregnant or lactating women were excluded too, as were adults with a lack of compliance with clinical examination and sample collection. Furthermore, we excluded adults with systemic diseases or disabilities that influence the capacity for oral-health self-care or cause salivary-gland dysfunction as well as adults with a partial denture or a fixed orthodontic device(s) that can alter the oral microbiome and salivary flow rate.

Thus, this study was conducted on six healthy adults (two males and four females, aged 34.14±7.51 years (mean±standard deviation [SD]), with ages ranging from 25 to 42 years) aged 20 years or older who voluntarily participated in saliva and dental-plaque sampling between February 2021 and May 2021.

2. Saliva and Plaque Collection Methods

All saliva samples were collected between 8 and 11 AM. The saliva sampling was performed during only one visit within 30 minutes under the following four conditions: 1. Do not consume food (including beverages; only water is allowed) 30 minutes before sampling. 2. Do not brush your teeth within 2 hours before the sampling. 3. Limit activities that may affect salivation, such as chewing gum, within 1 hour. 4. Limit the use of drugs that may affect salivation and the oral microbiome. Plaque samples were taken by applying gentle pressure to the buccal region of the first molar and lingual lower anterior teeth with dental curettes. The saliva and plaque samples were immediately placed in an ice box having a temperature below 4°C, and experiments were promptly conducted on the day the samples were obtained.

Salivary flow rates were expressed in mL/min and were determined by examining the amount of saliva secreted within 3 minutes by the following seven methods, which were applied in the following order: Method 1. The stimulated whole-saliva flow rate (SFR) was measured by pouring all the saliva produced while chewing gum into a-50 mL conical tube for 3 minutes. Method 2. The unstimulated whole-saliva flow rate (UFR) was measured by naturally spitting out all the saliva obtained in a comfortable state into a conical tube for 3 minutes. Method 3. After intense mouth-washing with 10 mL of saline for 20 seconds, the liquid mixed with saliva was spit out into a conical tube. Method 4. Plaque samples were obtained with dental curettes from the lingual surface of the mandibular incisors and the buccal surface of the maxillary molars. Method 5. The UFR was quantified using the GeneFiX Saliva DNA Microbiome Kit (Isohelix, Kent, United Kingdom). Method 6. The UFR was determined using the OMNIgene · ORAL|OM-501 (DNA Genot) Kit (DNA Genotek Inc., Ottawa, ON, Canada). Method 7. The UFR was measured using a funnel and a 15-mL conical tube container. There was a 2 to 5-minute break between the methods. Both saliva and mouthwash samples were transported to our laboratory and stored at −80°C within 10 minutes after collection.

3. Oral Hygiene and Periodontal Health

The following clinical parameters were determined at baseline by an experienced periodontist: oral hygiene, probing depth, the clinical attachment level, plaque index, and gingival index. The probing depth and clinical attachment level were measured by means of a Williams14W probe (Hu-Friedy Mfg. Co., Chicago, IL, USA). The plaque index is an indicator of oral hygiene management and was determined via the O’Leary plaque index. The gingival index was scored according to the Löe and Silness criteria. All participants had good oral hygiene and periodontal condition. Four of the six had a Decayed, Missing or Filled Teeth Surfaces (DMFS) index of 0 (caries-free, with no history of the disease), and the remaining two had a DMFS index >0 (currently caries free, with a history of the disease). The average DMFS index of the six subjects was 1.3 (±1.5, SD).

4. Salivary Flow Rates

Based on the obtained data, the SFR within 1 minute was calculated. With the stimulation, an SFR of <0.7 mL/min was regarded as very low, 0.7 to 0.99 mL/min as low, and ≥1.0 mL/min as normal [14]. Without stimulation, it is generally accepted that a very low SFR is ≤0.1 mL/min [15]. Values between 0.1 and 0.2 mL/min are thought to be low values, while those higher than 0.2 mL/min should be regarded as normal [16]. It has been shown that subjective symptoms of dry mouth, i.e., xerostomia, are often present below a salivary flow rate of approximately 0.1 to 0.2 mL/min when measured without stimulation. All participants had normal UFR and SFR.

5. DNA Extraction and 16S Ribosomal RNA Gene Sequencing

The amount of bacterial DNA, bacterial-community composition, and individual taxonomic abundance of oral bacterial species were compared among all the seven conditions (methods). Bacterial genomic DNA was extracted the samples obtained by one method involving simulation, four methods without stimulation, and one method involving mouthwash as well as from one plaque sample. The seven sampling methods were applied in the following order: Method 1 (stimulated saliva secretion), Method 2 unstimulated saliva (secretion), Method 3 (involves mouthwash), Method 4 (collection of dental plaque samples), Method 5 (unstimulated saliva secretion) involving the GeneFiX Saliva DNA Microbiome Kit, Method 6 (unstimulated saliva secretion) based on the OMNIgene · ORAL|OM-501 (DNA Genot) Kit, and Method 7 (unstimulated saliva secretion) involving a funnel.

In this work, the salivary flow rate was defined as the amount of saliva secreted per minute, and UFRs measured by five sampling methods—including Method 2 (UFR), Method 5 (GeneFiX Saliva DNA Microbiome), Method 6 (OMNigen · ORAL|OM-501), and Method 7 (UFR with a funnel)—and the SFR determined by Method 1 (SFR) were compared (Fig. 1).

1) Oral bacteria

In the saliva and plaque samples, the absolute and relative amount and relative abundance of A. actinomycetemcomitans, P. gingivalis, T. forsythia, T. denticola, F. nucleatum, P. intermedia, P. nigrescens, S. mitis, S. sobrinus, and L. casei were assessed.

2) Relative fluorescence units, cycle threshold values, and a qPCR plot

The DNA copy number of each oral bacterium was confirmed by finding the cycle threshold (Ct) value obtained after qPCR for each oral bacteriome in the standard curve. Real-time PCR that is quantitative is also known as qPCR. Real-time PCR results can be either qualitative (presence or absence of a sequence) or quantitative (number of copies of DNA).

Unlike conventional semiquantitative PCR, real-time PCR used in our qPCR experiment can detect the amount of amplified DNA or RNA products in real time in relative fluorescence units (RFUs). The expression level of a specific gene was quantified as a “Ct value,” that is, the number of cycles required for RFU detection during rDNA amplification.

After completion of the PCR, one needs to convert the relative fluorescence reading to Ct, using the software provided with the PCR machine. RFUs refer to the use of specific dyes that either intercalate into DNA in a nonspecific manner or are primer-specific probes that are specifically integrated into PCR products as they are synthesized. From the data, we determined whether the reaction was optimized and could therefore be utilized for analysis. This is because a researcher will base any judgments about the upregulation or downregulation of the gene(s) of interest on the Ct values generated in relation to housekeeping genes. The Ct values are derived from the relative fluorescence and linked to the amount of starting material (a small amount of starting rDNA corresponds to a high Ct value), primer efficiency, and the magnitude of amplification that occurs per cycle. To ensure that the assay is optimized, the researcher needs to consider both linearity of the reaction (by constructing a standard curve) and good amplification efficiency (Fig. 2). Based on the Ct values, P. gingivalis, P. intermedia, and S. sobrinus were not detected.

6. Bacterial-DNA Isolation from Saliva Samples

Saliva samples were vortexed vigorously, and 500 μL of a sample was added to a tube containing 500 μL of lysis buffer (5 mM ethylene-diamine-tetraacetic acid [EDTA], 5 M guanidine hydrochloride, and 0.3 M sodium acetate). After vortexing to mix the sample with lysis buffer, the tubes were incubated at 65°C for 10 minutes. The S2 buffer made of 0.25 g/mL silicon dioxide (Merck KGaA, Darmstadt, Germany) was thoroughly mixed by vortexing, and 20 μL of this buffer was added to the mixture of the sample with lysis buffer. After vortexing, the tubes were incubated for 5 minutes at room temperature with intermittent inverting. The mixture was centrifuged at 5,000 rpm for 30 seconds, and the supernatant was carefully removed.

One milliliter of PureLink (Invitrogen Corporation, Carlsbad, CA, USA) PCR purification washing buffer 1 (50 mM 3-[N-morpholino] propane sulfonic acid buffer [pH 7.0] with 1 M sodium chloride) was activated by the addition of 160 mL of 100% ethanol and then was added into the tubes and mixed by vortexing until beads were resuspended completely. After centrifugation at 5,000 rpm for 30 seconds, the supernatant was removed carefully, and 1,000 μL of washing buffer 2 (ethanol) was added and vortexed to resuspend the beads completely. Finally, the tubes were centrifuged at 5,000 rpm for 30 seconds, and the supernatant was removed completely. One hundred microliters of elution buffer (100 mM Tris-HCl [pH 7.5], 1 M EDTA) was added into the tube and vortexed to resuspend the beads. The tubes were incubated at 65°C for 10 minutes to dissolve the DNA and separate it from the beads. After centrifugation at 13,000 rpm for 5 minutes, the supernatant was transferred to a new sterile microcentrifuge tube and used for PCR.

7. Real-Time PCR Amplification

Real-time PCR amplification reactions were performed on each sample with primers specific for the 10 species of oral bacteria (A. actinomycetemcomitans, P. gingivalis, T. forsythia, T. denticola, F. nucleatum, P. intermedia, P. nigrescens, S. mitis, S. sobrinus, and L. casei). Bacterial 16S ribosomal RNA (rRNA) primers were used to quantify total bacteria. The reaction mixture consisted of 10 μL of 2X Master-mix (GeNet Bio, Daejeon, Korea), 2.5 pM forward and reverse primers, and 5 μL of the DNA template. The total reaction volume (20 μL) was subjected to real-time PCR amplification under the following conditions: predenaturation at 95°C for 10 minutes, followed by 45 cycles of denaturation at 95°C for 15 seconds and annealing/extension at 60°C for 1 minute. The synthesized plasmid DNA of each bacterium served as a positive control, while DNase/RNase-free water was used as a negative control.

8. α-Diversity According to the Shannon Diversity Index

α-Diversity was calculated as the Shannon diversity index, and bacterial richness was measured as the total number of bacterial DNA copies. The α-diversity levels of microbial profiles were compared via Shannon index calculations by means of the following formula: H=–∑pi×ln(pi), where pi is the relative abundance, H is the Shannon diversity index, pi is the proportion of cells of the ith species in the whole community: pi=n/N, where n is the number of cells of a given taxon/species, and N is the total number of bacterial cells in a community. The minimum value that the Shannon diversity index can take is 0 and means that there is no diversity; the greater the value, the higher is the diversity. In real-world ecological data, the Shannon diversity index’s range is typically from 1.5 to 3.5 and rarely reaches 4.5 [17].

9. Data Analysis

Data were analyzed for descriptive statistics and presented as numbers (%) for categorical variables and as mean±SD for continuous variables. All bacterial data were transferred to an Excel spreadsheet (Microsoft Corp., Redmond, WA, USA) and analyzed based on Ct values. The Ct value is the number of cycles required for the fluorescent signal to cross the threshold and is inversely proportional to the amount of DNA in the sample. Ct values over 40 were regarded as “ND (not detected),” and those less than 40 were accepted as valid data. To calculate the bacterial rDNA copy number, we utilized a real-time PCR assay in which, by using a standard curve or by comparing the average Ct values with those of standard samples, we calculated actual or relative rDNA content. One-way analysis of variance (ANOVA) with Tukey’s post hoc test were used in comparison analysis for continuous variables. Comparisons of the α-diversity of saliva and plaque samples at the probe level were made by the Mann–Whitney test with Benjamini−Hochberg correction for multiple comparisons. Spearman’s correlation coefficients were computed to determine a correlation between a salivary flow rate and the copy number of each oral bacterium. In all analyses, data with a two-tailed p-value of less than 0.05 were considered statistically significant.


1. Bacterial Diversity According to the Shannon Index

The Shannon diversity index was used to evaluate the diversity of oral bacteria in samples obtained by each sample collection method (Table 1). This index takes into account both species abundance and species richness. Method 1 (stimulated salivation), Method 2 (unstimulated salivation), Method 3 (mouthwash with saline), Method 4 (dental plaque collection), Method 5 (GeneFiX Saliva DNA Microbiome Kit), Method 6 (OMNIgene · ORAL|OM-501 [DNA Genot Kit]), and Method 7 (unstimulated salivation measured with the funnel).

Among the seven sampling methods, Method 7 (unstimulated salivation measured with the funnel) (4.449) had the highest Shannon index, followed by Method 6 (OMNIgene · ORAL|OM-501 [DNA Genot Kit]) (4.197), Method 5 (GeneFiX Saliva DNA Microbiome Kit) (3.725), Method 4 (plaque collection) (3.623), Method 2 (3.171), Method 1 (stimulated salivation) (3.129), and Method 3 (mouthwash with saline) (2.061). That is, oral-microbiome diversity in terms of the Shannon diversity index was the highest when unstimulated-saliva samples were analyzed using the funnel, followed by two commercial containers, plaque analysis results, and methods based on stimulated and unstimulated salivation without the funnel. The Shannon diversity index of the plaque sample was higher than that of the unstimulated- and stimulated-saliva samples obtained using only conical tubes. The lowest Shannon diversity index was observed in the saliva obtained by Method 3 (mouthwash with saline).

2. The Richness According to Total DNA Copy Number of Oral Bacteria

Fig. 3 shows the total DNA copy numbers of oral bacteria in relation to the sample collection methods. There was no significant difference in total 16S rDNA copy number among Method 5 (GeneFiX Saliva DNA Microbiome Kit), Method 6 (OMNIgene · ORAL|OM-501 [DNA Genot Kit]), and Method 7 (unstimulated salivation measured with the funnel). The values obtained by Methods 5 and 6 were higher than those obtained by Method 1 (stimulated salivation), Method 2 (unstimulated salivation), Method 3 (mouthwash with saline), and Method 4 (plaque analysis). The value obtained by Method 7 (unstimulated salivation measured with the funnel) was significantly higher than that obtained by Method 1 (stimulated salivation), Method 3 (mouthwash with saline), and Method 4 (dental plaque) analysis (p<0.001).

That is, the total number of bacterial rDNA copies was significantly larger in Methods 5 to 7 than in Methods 1 to 4. The total numbers of bacterial DNA copies also differed significantly depending on the collection method, and the values obtained with the commercial kits and funnel were higher than the values obtained when only conical tubes were used. When commercial kits were used or a funnel was employed to prevent saliva from leaking out of the conical tube, the total copy number of bacterial rDNA was significantly higher than that obtained when saliva was collected using only a conical tube without a commercial kit and dental plaque was analyzed.

3. Salivary Flow Rates Compared among the Saliva Collection Methods

Fig. 4 shows the differences in the salivary flow rates among the sample collection methods. The SFR was significantly higher than the UFR (1.75±0.72 mL/min vs. 1.27±0.27 mL/min, p-value=0.003). The UFR (1.27±0.27 mL/min) did not differ significantly among the methods involving the funnel (1.19±0.28 mL/min), the GeneFiX Saliva DNA Microbiome Kit (1.03±0.43 mL/min), or the OMNigene · ORAL|OM-501 Kit (1.39±0.28 mL/min). The SFR (1.75±0.72 mL/min) was significantly higher than the UFR, UFR measured with the funnel, and UFR determined with the GeneFiX Saliva DNA Microbiome Kit (p<0.05) (Fig. 4). Namely, there was no significant difference in the UFR among the collection method.

In other words, the UFR was significantly smaller than the SFR (1.75±0.72 mL/min vs. 1.27±0.27 mL/min, p-value=0.003).

4. Correlations between the Salivary Flow Rate and Bacterial Richness

Table 2 presents the correlation between the salivary flow rate and the number of bacterial 16S rDNA copies. Among the 10 oral bacteria investigated, some bacteria, including T. forsythia, P. nigrescens, T. denticola, and L. casei, featured a negative correlation with the salivary flow rate; S. sobrinus manifested a strong positive correlation, whereas the 16S rDNA copy number of the other five oral bacterial species did not correlate with the salivary flow rate determined by any method.

The T. forsythia copy number (r=−0.399, p=0.016), P. nigrescens copy number (r=−0.362, p=0.030), and total bacterial DNA copy number (r=−0.331, p=0.049) negatively correlated with the salivary flow rate when statistical analysis was performed by synthesizing 7 conditions.

The S. sobrinus copy number (r=0.963, p=0.002) strongly positively correlated with the UFR. Regarding the salivary flow rate when the GeneFiX Saliva DNA Microbiome Kit was employed, the T. denticola copy number (r=−0.816, p=0.048) and L. casei copy number (r=−0.877, p=0.022) correlated negatively with this parameter. Regarding the SFR, the S. sobrinus copy number (r=0.963, p=0.002) positively correlated with this rate. That is, the amount of some species of oral bacteria correlated with the salivary flow rate, but other oral bacterial species analyzed did not.

5. Each Observed Oral-Bacterial-Community Composition Depending on Sampling Methods

Table 3 shows the comparison of each bacterial DNA copy number according to the sampling methods. Among the 10 tested oral bacteria, bacterial DNA copy numbers of three species—F. nucleatum, P. nigrescens, and S. mitis—were significantly different among the collection methods. For the other seven oral bacteria—A. actinomycetemcomitans, P. gingivalis, T. forsythia, T. denticola, P. intermedia, S. sobrinus, and L. casei—the difference in copy numbers among the sample collection methods was not statistically significant.

In the case of F. nucleatum, the copy number was significantly larger in Methods 5 to 7 than in Methods 1 to 3 methods. The copy number of F. nucleatum of was significantly larger in Method 3 than in Method 6. In the case of P. nigrescens, the copy number was significantly larger in Methods 5 to 7 than in Methods 1 to 4. For S. mitis, the DNA copy number was significantly larger in Method 2 than in Methods 1 and 3 to 7.

6. Relative Abundance (%) of Each Bacterial Species Depending on the Sampling Method within Each Study Participant

Although statistical analysis was not performed in this case, relative abundance of each bacterial species was visualized and analyzed in relation to the method of sampling of oral bacteria in each subject (Fig. 5). When the microbiomes of samples collectively among subjects or within the same subject were visualized, the patterns of the microbiomes could be clustered into three types. Cluster 1: Unstimulated- and stimulated-saliva samples collected using only conical tubes and mouthwash saliva samples; Cluster 2: saliva samples obtained using commercial kits or the addition of the funnel to conical tubes; Cluster 3: Unique patterns of plaque samples.

The oral microbiomes were similar among the saliva samples obtained by Methods 1 to 3 within a single individual. The species with the highest abundance was S. mitis, with the range from 89.31% to 100.00% in Methods 1 to 3 in all six subjects. With Methods 1 to 3 the second most common species after S. mitis was F. nucleatum, with a range of 0.66% to 33.58%, and the next most abundant species were P. nigrescens (0.0%-15.88%) and T. denticola (0.0%-11.91%). The relative abundance levels of the remaining species were negligible, and the abundance levels of A. actinomycetemcomitans, P. gingivalis, S. sobrinus, and L. casei were less than 1% with any sample collection method.

In the plaque sample (Method 5), the oral bacteria showed a pattern distinct from that of the saliva sample; in Subject 1, T. forsythia relative abundance (85.48%) was the highest, followed by F. nucleatum (9.94%). On the contrary, Subjects 2 and 6 featured F. nucleatum as the top species (81.79% and 85.27%, respectively), followed by T. forsythia (13.88% and 9.16%, respectively). Subject 3 had the highest relative abundance of F. nucleatum (70.67%), followed by S. mitis (18.37%). Subjects 4 and 5 had the highest relative abundance of S. mitis (76.40% and 84.87%, respectively), followed by F. nucleatum (12.02% and 6.16%). Among all the seven sampling methods in all six subjects, the lowest relative abundance was observed for P. gingivalis (<0.1%), followed by L. casei (<0.2%), S. sobrinus (<1%), and A. actinomycetemcomitans (<3%).

Within one individual, the species that showed the greatest variation among the seven sampling methods was S. mitis (0%-100%). Species with less wide ranges of relative abundance were F. nucleatum (0%-85.27%), P. nigrescens (0%-58.31%), P. intermedia (0%-48.27%), and T. denticola (0%-12%). Therefore, there was a difference in the profile of bacteria among the subjects and among sampling methods within each subject; therefore, it is necessary to improve the understanding of the sampling and to standardize the assay conditions.

7. Absolute DNA Copy Numbers of Each Oral Bacterium within an Individual

Method 4 showed a unique microbiome, whereas Methods 5 to 7 yielded microbiomes that were similar to one another. Differences in the pattern of absolute oral bacterial copy numbers according to sampling method were documented within one individual (Fig. 6). When we looked at the absolute bacterial DNA copy numbers, there were individual differences, and there were also differences among the sampling methods. Of note, the absolute bacterial copy numbers were high in Methods 5 to 7. Besides, S. mitis, F. nucleatum, T. denticola, and T. forsythia were observed at high levels. A high level of P. intermedia was observed only in Subjects 2 and 6, although this level was less than 0.3% in the other four subjects. In the plaque sample (Method 4), P. intermedia and S. mitis were fairly abundant.


The purposes of this study were to test whether various saliva collection methods affect the observed profile of the 10 major salivary microbes and to determine whether the observed microbiomes of unstimulated- and stimulated-saliva samples and plaque samples differ in diversity and richness. The Shannon diversity index was the highest in unstimulated saliva collected with the funnel, and this index was higher in the unstimulated- and stimulated-saliva samples and plaque samples than in the mouthwash sample. There was no significant difference in the total number of DNA copies, richness of the oral microbiome, among unstimulated- and stimulated-saliva samples and plaque samples. In three species, F. nucleatum, P. nigrescens, and S. mitis, there were significant differences in the bacterial DNA copy number among the collection method. Even if saliva and plaque are collected from the same person at the same time, the richness and diversity might differ depending on the sample collection method.

Firstly, saliva is a typical diagnostically important biological fluid that contains useful biomarkers. In addition, the main advantage of saliva for biomarker analysis is that saliva can be easily, safely, and noninvasively collected routinely in the dental office [18]. Saliva sampling is noninvasive, low-cost, and simple; additionally, these samples are convenient for the screening of patients for oral pathogens. A saliva-based assay has been proposed as an approach to population-based screening for oral health and disease, and alterations in salivary bacterial profiles have been suggested as candidate biomarkers of oral health and disease [1]. For practical reasons, stimulated-saliva samples may be preferred over unstimulated-saliva samples because the former can be collected in larger amounts and considerably faster than unstimulated-saliva samples can be [19]. In the present study, when commercial kits or a funnel were used, bacterial diversity was higher in unstimulated saliva than in stimulated saliva. One study is consistent with our findings: bacterial diversity measured via the 16S rRNA gene in the oral microbiome of saliva from two healthy individuals was higher in unstimulated-saliva samples than in stimulated-saliva samples [8]. Additionally, unstimulated saliva has been routinely regarded as a representative average environment of the entire ecosystem of the oral cavity [20]. In terms of S. mutans detection in the diagnosis of dental caries, stimulated-saliva samples manifested high sensitivity (94%) similar to that of plaque samples but showed lower specificity (11%) than that of plaque samples (17%) [9]. A more precise sampling method is required for proper determination of oral microbial profiles.

We studied 10 oral bacteria by real-time PCR. With the development of microbiome analysis technologies, more than half of the ~700 oral microorganisms have been identified [2]. The oral microbiota of healthy subjects, up to 101 species have been described [21]. Real-time PCR has been devised to overcome the limitations of traditional PCR, thereby allowing for accurate quantification with higher sensitivity, specificity, simplicity, and rapidity [12]. Real-time PCR has high sensitivity for oral microbiome characterization and can be used to identify and quantify oral bacteria in saliva samples [13]. Of course, by means of next-generation sequencing (NGS), which is a highly developed analytical approach, whole genomes of oral bacteria can be characterized [22]. NGS is considered the gold standard of oral-microbiome analysis. On the other hand, NGS is still expensive and technically difficult to implement [23]. Consequently, more research on quantification and characterization of the oral microbiome by real-time PCR is needed to determine whether unstimulated saliva is an appropriate proxy of microbial composition of the oral cavity.

Here, we focused on each of the 10 major oral bacterial species including A. actinomycetemcomitans, P. gingivalis, T. forsythia, T. denticola, F. nucleatum, P. intermedia, P. nigrescens, S. mitis, S. sobrinus, and L. casei. In this work, the highest microbial richness was documented for S. mitis, followed by F. nucleatum, P. nigrescens, T. denticola, T. forsythia, and P. intermedia. The sum of the proportions of the remaining five species was less than 1%. S. mitis is a gram-positive bacterium, a mesophilic α-hemolytic species of Streptococcus that inhabits the human mouth. S. mitis is most commonly found in the throat, nasopharynx, and mouth. It can cause infective endocarditis and primary bacteremia. S. mitis is a pioneer commensal bacterial species colonizing many surfaces of the oral cavity in healthy individuals. The commensal species S. mitis is a predominant pioneer colonizer of the oral cavity from early infancy throughout the lifespan and is thought to form the basis of oral biofilms by supplying adherence sites for secondary colonizers [24]. Among healthy subjects, Streptococcus is the most abundant genus, constituting 45% of the genera present [25]. The precise reasons for S. mitis commensalism are still unclear [26].

Furthermore, four bacterial species, T. forsythia, P. nigrescens, T. denticola, and L. casei, were negatively correlated here with the salivary flow rate. Saliva plays an essential role in shaping and maintenance of the ecological equilibrium within the resident oral microbiota. It also plays a major part in the maintenance of the relationship between the host and oral microbiota in a symbiotic state [27]. In addition, saliva comprises proteins such as mucins that block the adherence of certain microorganisms to oral surfaces through binding and aggregating mechanisms. Saliva exerts antimicrobial action through numerous proteins and peptides, including mucins, lysozyme, lactoferrin, statherin, histatins, and secretory immunoglobulin A [27]. Therefore, when the salivation rate is high, the amount of harmful oral bacteria should decrease. The reason why S. sobrinus relative abundance strongly positively correlated with the salivary flow rate requires further investigation.

Let us take a closer look at the representative oral bacteria that we investigated. A. actinomycetemcomitans is a gram-negative facultative anaerobe that is a non-motile bacterium that is associated with periodontitis and endocarditis [28]. T. denticola, T. forsythia, and P. gingivalis are members of the red complex of periodontal pathogens. P. gingivalis, T. denticola, and T. forsythia not only exist as a consortium that is associated with chronic periodontitis but also exhibit synergistic virulence, resulting in immunoinflammatory bone resorption [29]. T. denticola is a gram-negative obligate anaerobe that is a motile, and highly proteolytic spirochete bacterium. P. gingivalis is a key pathogen of periodontitis and is a gram-negative, rod-shaped, anaerobic, pathogenic bacterium that is implicated in infections of the upper gastrointestinal tract, respiratory tract, and colon; Alzheimer’s disease; and rheumatoid arthritis [30]. The odds ratio for infection with P. gingivalis is 11.2-fold greater in periodontitis patients than in the healthy population [31]. It is present at higher prevalence and abundance in disease-active areas. We expected to detect lower prevalence and abundance in our periodontally healthy study population. Among the 10 representative oral microbes in our study, the species with the lowest microbial richness was P. gingivalis (<0.1%). Consistently with our results, P. gingivalis is reported to be undetectable in the majority of subjects with age-matched, periodontally healthy controls [31]. By contrast, one published PCR assay detects P. gingivalis in over 90% of periodontally healthy subjects, whereas other red complex bacteria—T. forsythia and T. denticola—are detectable in over 50.0% of these subjects [11].

F. nucleatum, a gram-negative anaerobe, is an oral commensal and periodontal pathogen associated with a wide spectrum of human diseases [32]. F. nucleatum is ubiquitous in the oral cavity and is absent or only infrequently detected elsewhere in the body under normal conditions [33]. The abundance of F. nucleatum is known to be affected by environmental factors. In the presence of a subgingival biofilm, smoking increases the abundance of F. nucleatum in both periodontally healthy and diseased individuals [34]. During coinfection with F. nucleatum and other oral species, e.g., T. forsythia, P. gingivalis, and streptococci, synergy in virulence is observed, as evidenced by enhanced bone loss, abscess, or death [35,36]. P. intermedia, a gram-negative obligate anaerobe, is a pathogenic bacterium. P. intermedia is an oral bacterium frequently associated with periodontal diseases [37]. P. nigrescens is a gram-negative bacterium that is obtained from saliva and a supragingival biofilm [38]. S. sobrinus is a gram-positive nonmotile anaerobic member of the genus Streptococcus. S. sobrinus, in conjunction with the closely related species S. mutans, is pathogenic to humans and enhances the formation of caries within teeth [39]. L. casei is a gram-positive, nonmotile bacterium that has a wide pH and temperature range in terms of survival and complements the growth of Lactobacillus acidophilus, which is a lactic acid bacterium that is a producer of the enzyme amylase. L. casei is a probiotic species that may be effective in alleviating gastrointestinal pathogenic bacterial diseases [40]. A number of further investigations are needed, such as a comparison of sizes among all oral bacteria, research on the nature of the bacterial surface, interactions between oral bacteria, and the relationship between the salivary flow rate or pH and the bacterial profiles.

A limitation of this study is small sample size and omission of advanced techniques such as NGS. We did not comprehensively assess whole genomes of the oral bacteria; therefore, additional research is needed with a large study population and a more advanced analytical technology. Nonetheless, our study is the first to comprehensively investigate the salivary flow rate, oral bacterial richness, and diversity in relation to sample collection methods. Additionally, we analyzed for the first time 10 species of oral bacteria in unstimulated and stimulated saliva, mouthwash saliva, and dental plaque simultaneously. When a funnel was used for saliva collection, the Shannon diversity index of unstimulated saliva was the highest, but examination of stimulated saliva under the same conditions is necessary too. Although mouth-rinsing saliva collection and plaque sampling are noninvasive, simple, and useful for detecting a wide range of bacterial species, their limitation is the lack of clear diagnostic criteria. If a simple collection kit can quantify and characterize an oral microbiome in the future, it is expected to be helpful for predicting dysbiosis.

Analysis of the data uncovered a considerable dependence of the observed bacterial profile on the saliva collection method. The observed bacterial abundance levels in saliva are consistent among the sampling methods when specialized commercial kits or a funnel are used. Furthermore, dental plaque cannot be an adequate surrogate for saliva as a biological sample because the plaque has a unique microbial profile. Before our work, no studies have addressed differences between unstimulated versus stimulated saliva samples, between saliva and plaque samples, or among saliva samples obtained by different sampling-method–dependent microbial techniques comprehensively. Thus, additional research is needed to clarify our conclusions about the observed bacterial profile depending on the sample collection method.


No potential conflict of interest relevant to this article was reported.

Fig. 1. Sampling method. (A) Sampling with Method 2 (unstimulated whole-saliva flow rate), (B) sampling with Method 1 (stimulated whole-saliva flow rate), (C) the GeneFiX Saliva DNA Microbiome Kit of Method 5, (D) sampling with Method 5, (E) the OMNIgene · ORAL|OM-501 (DNA Genot) Kit of Method 6, (F) sampling with Method 6, (G) a funnel and a 15-mL conical tube container of Method 7, and (H) sampling with Method 7.
Fig. 2. Bacterial rDNA amplification for each bacterial species. (A) Aggregatibacter actinomycetemcomitans, (B) Tannerella forsythia, (C) Treponema denticola, (D) Fusobacterium nucleatum, (E) Prevotella nigrescens, (F) Streptococcus mitis, (G) Lactobacillus casei, (H) total bacterial copy numbers. Considering the cycle threshold values, Porphyromonas gingivalis, Prevotella intermedia, and Streptococcus sobrinus were not detected. RFU, relative fluorescence unit.
Fig. 3. The total DNA copy numbers (×108) of oral bacteria in relation to the sample collection methods.
Fig. 4. Comparison of salivary flow rates determined by the various sample collection methods. UFR, unstimulated whole-saliva flow rate; SFR, stimulated whole-saliva flow rate. *Statistically significant p<0.05.
Fig. 5. Intraindividual differences in the observed oral microbiome depending on the sampling method.
Fig. 6. Absolute bacterial DNA copy numbers (×104) of each oral bacterium by subjects.

Total bacterial 16S rDNA copy numbers (×108)

Sampling method No. Mean Standard deviation 95% CI Minimum Maximum p-value Post-hoc analysis Shannon diversity index

Lower Upper
1 6 8.056 9.703 –2.127 18.239 0.37 27.14 0.001** 1<7, 8, 9 3.129
2 6 13.561 12.792 0.137 26.985 2.00 35.69 2<7, 8 3.171
3 6 0.895 0.991 –0.145 1.935 0.24 2.87 5<7, 8, 9 2.061
4 6 3.568 2.984 0.436 6.699 1.39 9.41 6<7, 8, 9 3.623
5 6 53.717 40.672 11.034 96.399 0.72 106.18 1, 2, 5, 6<7 3.725
6 6 51.424 33.246 16.534 86.313 0.21 85.72 1, 2, 5, 6<8 4.197
7 6 41.673 38.803 0.952 82.394 0.75 113.57 1, 5, 6<9 4.449
Total 42 24.699 32.088 14.699 34.698 0.21 113.57 - - 3.479

The results were obtained via ANOVA and post hoc analysis. Method 1 (stimulated salivation), Method 2 (unstimulated salivation), Method 3 (mouthwash with saline), Method 4 (dental plaque collection), Method 5 (GeneFiX Saliva DNA Microbiome Kit), Method 6 (OMNIgene · ORAL|OM-501 [DNA Genot Kit]), and Method 7 (unstimulated salivation measured with the funnel).

**Statistically significant p<0.01.

Correlation between the salivary flow rate and the bacterial 16S rDNA copy number

Correlation Total salivary flow rate UFR with using the GeneFiX Saliva DNA Microbiome Kit UFR with using the OMNIgene · ORAL|OM-501 (DNA Genot) Kit UFR with funnel UFR SFR
Aggregatibacter actinomycetemcomitans
Spearman’s rho –0.164 0.515 –0.733 –0.119 –0.295 –0.039
p-value 0.340 0.296 0.097 0.822 0.570 0.941
Porphyromonas gingivalis
Spearman’s rho 0.144 NA NA NA 0.622 NA
p-value 0.401 NA NA NA 0.187 NA
Tannerella forsythia
Spearman’s rho –0.399* –0.640 –0.373 –0.119 –0.307 –0.603
p-value 0.016 0.171 0.467 0.822 0.554 0.205
Treponema denticola
Spearman’s rho –0.319 –0.816* –0.136 NA 0.029 –0.488
p-value 0.058 0.048 0.798 NA 0.956 0.326
Fusobacterium nucleatum
Spearman’s rho –0.328 0.470 –0.241 –0.124 –0.326 0.335
p-value 0.051 0.347 0.646 0.814 0.529 0.517
Prevotella intermedia
Spearman's rho –0.135 –0.151 0.549 NA –0.643 –0.347
p-value 0.433 0.775 0.259 NA 0.168 0.501
Prevotella nigrescens
Spearman’s rho –0.362* –0.285 –0.270 –0.118 –0.298 –0.379
p-value 0.030 0.584 0.604 0.824 0.566 0.458
Streptococcus mitis
Spearman’s rho –0.027 0.524 –0.546 –0.202 –0.167 –0.094
p-value 0.877 0.286 0.262 0.700 0.752 0.860
Streptococcus sobrinus
Spearman’s rho 0.115 0.384 NA 0.400 0.414 0.963**
p-value 0.503 0.453 NA 0.432 0.414 0.002
Lactobacillus casei
Spearman’s rho –0.253 –0.877* –0.243 0.117 0.298 –0.334
p-value 0.137 0.022 0.643 0.826 0.566 0.518
Spearman’s rho –0.331* 0.526 –0.007 –0.303 –0.126 –0.037
p-value 0.049 0.284 0.990 0.560 0.812 0.945

UFR, unstimulated salivary flow rate; SFR, stimulated salivary flow rate; NA, not applicable.

The results were obtained using Spearman correlation analysis.

Statistically significant *p<0.05, **p<0.01.

A comparison of bacterial rDNA copy numbers (×108) among the sampling methods

Sampling method Mean Standard deviation 95% confidence interval Minimum Maximum p-value Post hoc analysis

Lower Upper
Aggregatibacter actinomycetemcomitans
1 48,481.83 118,755.75 –76,144.69 173,108.35 0 290,891 0.803 NA
2 72,648.83 177,952.57 –114,100.94 259,398.60 0 435,893 NA
3 0.00 0.00 0.00 0.00 0 0 NA
4 0.00 0.000 0.00 0.00 0 0 NA
5 155,255.83 380,297.571 –243,841.99 554,353.66 0 931,535 NA
6 231,633.83 567,384.699 –363,799.89 827,067.56 0 1,389,803 NA
7 135,011.83 330,710.101 –212,047.13 482,070.80 0 810,071 NA
Porphyromonas gingivalis
1 0.00 0.000 0.00 0.00 0 0 0.482 NA
2 248.00 607.473 –389.50 885.50 0 1,488 NA
3 0.00 0.000 0.00 0.00 0 0 NA
4 1,224.83 3,000.217 –1,923.70 4,373.37 0 7,349 NA
5 0.00 0.000 0.00 0.00 0 0 NA
6 0.00 0.000 0.00 0.00 0 0 NA
7 0.00 0.000 0.00 0.00 0 0 NA
Tannerella forsythia
1 355,958.00 871,915.470 –559,061.17 1,270,977.17 0 2,135,748 0.222 NA
2 211,748.33 503,390.743 –316,527.85 740,024.52 0 1,238,968 NA
3 0.00 0.000 0.00 0.00 0 0 NA
4 2,099,318.17 3,277,437.316 –1,340,141.36 5,538,777.69 0 8,124,376 NA
5 1,502,990.50 1,644,647.735 –222,961.45 3,228,942.45 0 4,267,478 NA
6 1,252,173.50 1,240,246.501 –49,385.38 2,553,732.38 0 3,298,916 NA
7 1,241,259.00 1,149,639.690 34,786.14 2,447,731.86 0 2,808,641 NA
Treponema denticola
1 0.00 0.000 0.00 0.00 0 0 0.173 NA
2 2,930.67 4,642.936 –1,941.80 7,803.13 0 10,328 NA
3 0.00 0.000 0.00 0.00 0 0 NA
4 107,831.17 169,439.506 –69,984.69 285,647.02 0 416,793 NA
5 1,387,600.83 2,221,861.572 –944,099.89 3,719,301.55 0 5,351,755 NA
6 1,146,313.50 1,831,171.145 –775,382.76 3,068,009.76 0 4,624,036 NA
7 1,250,334.00 1,778,722.126 –616,320.39 3,116,988.39 0 3,857,556 NA
Fusobacterium nucleatum
1 2,583,035.17 6,255,387.314 –3,981,591.22 9,147,661.56 1,341 15,351,471 0.001** 1<7, 8, 9
2 848,921.33 1,905,446.388 –1,150,722.02 2,848,564.68 2,938 4,735,872 2<7, 8, 9
3 9,491.17 11,264.236 –2,329.92 21,312.26 0 31,266 5<7, 8, 9
4 6,949,199.33 13,487,558.600 –7,205,125.36 2,110,3524.02 26,193 34,268,887 6<8
5 1,797,1378.00 11,077,713.842 6,346,030.29 29,596,725.71 220,663 31,946,939 NA
6 23,469,532.17 16,176,389.816 6,493,452.17 40,445,612.16 72,678 48,801,958 NA
7 17,916,312.33 14,954,030.845 2,223,019.40 33,609,605.26 186,742 41,035,643 NA
Prevotella intermedia
1 0.00 0.000 0.00 0.00 0 0 0.574 NA
2 1,162.33 1,309.009 –211.39 2,536.05 0 2,864 NA
3 0.00 0.000 0.00 0.00 0 0 NA
4 167,200.17 398,235.223 –250,722.07 585,122.40 0 979,896 NA
5 1,424,853.67 3,023,476.213 –1,748,090.03 4,597,797.36 0 7,541,492 NA
6 3,050,499.33 6,758,569.824 –4,042,184.71 10,143,183.38 0 16,791,026 NA
7 3,474,653.33 8,196,610.191 –5,127,161.53 12,076,468.19 0 20,197,483 NA
Prevotella nigrescens
1 1,216,207.67 2,961,131.913 –1,891,309.70 4,323,725.03 0 7,260,558 0.001** 1<7, 8, 9
2 257,909.50 618,473.162 –391,138.28 906,957.28 0 1,520,267 2<7, 8, 9
3 1,181.33 2,893.664 –1,855.38 4,218.05 0 7,088 5<7, 8, 9
4 195,761.17 248,636.588 –65,166.93 456,689.26 0 6,48948 6<7, 8, 9
5 9,445,436.33 8,137,458.487 905,697.38 17,985,175.29 70,265 18,600,527 NA
6 8,430,044.33 6,304,841.930 1,813,518.51 15,046,570.16 10,520 15,919,347 NA
7 14,709,256.67 13,472,389.344 570,851.14 28,847,662.20 30,698 37,125,238 NA
Streptococcus mitis
1 4,640,524.00 7,936,513.959 –3,688,336.61 12,969,384.61 202,335 20,678,070 0.007** 1<2
2 18,641,779.33 21,376,243.689 –3,791,212.64 41,074,771.31 4,298,926 59,281,611 1, 5, 6, 7, 8, 9 <2
3 2,165,331.17 4,621,794.193 –2,684,944.38 7,015,606.71 121,333 11,596,582 NA
4 259,170.50 207,344.603 41,575.68 476,765.32 3,939 489,159 NA
5 3,568.50 8,266.907 –5,107.09 12,244.09 0 20,424 NA
6 2,269.17 3,545.361 –1,451.46 5,989.79 0 7,535 NA
7 5,245.00 10,135.053 –5,391.09 15,881.09 0 25,313 NA
Streptococcus sobrinus
1 2,175.67 5,329.273 –3,417.06 7,768.40 0 13,054 0.809 NA
2 1,592.00 3,899.588 –2,500.37 5,684.37 0 9,552 NA
3 0.00 0.000 0.00 0.00 0 0 NA
4 824.50 2,019.604 –1,294.94 2,943.94 0 4,947 NA
5 786.00 1,925.299 –1,234.48 2,806.48 0 4,716 NA
6 0.00 0.000 0.00 0.00 0 0 NA
7 646.83 1,584.412 –1,015.90 2,309.57 0 3,881 NA
Lactobacillus casei
1 645.17 892.945 –291.92 1,582.25 0 2,313 0.695 NA
2 451.67 522.072 –96.21 999.55 0 1,120 NA
3 21.83 53.481 –34.29 77.96 0 131 NA
4 41.33 64.186 –26.03 108.69 0 131 NA
5 2,149.00 5,263.953 –3,375.18 7,673.18 0 12,894 NA
6 492.17 1,205.557 –772.99 1,757.32 0 2,953 NA
7 749.17 1,835.076 –1,176.63 2,674.96 0 4,495 NA
1 8.0562 9.70339 –2.1269 18.2392 0.37 27.14 0.001** 1<7, 8, 9
2 13.5610 12.79201 0.1366 26.9854 2.00 35.69 2<7, 8, 9
3 0.8949 0.99110 –0.1452 1.9350 .24 2.87 5<7, 8, 9
4 3.5680 2.98421 0.4362 6.6997 1.39 9.41 6<7, 8, 9
5 53.7167 40.67216 11.0339 96.3995 0.72 106.18 NA
6 51.4238 33.24600 16.5342 86.3133 0.21 85.72 NA
7 41.6730 38.80262 0.9521 82.3938 0.75 113.57 NA

NA, not applicable.

The results were obtained via ANOVA and post hoc analysis.

Statistically significant **p<0.01.

  1. Lee YH, Chung SW, Auh QS, et al. Progress in oral microbiome related to oral and systemic diseases: an update. Diagnostics (Basel) 2021;11:1283.
    Pubmed KoreaMed CrossRef
  2. Lamont RJ, Koo H, Hajishengallis G. The oral microbiota: dynamic communities and host interactions. Nat Rev Microbiol 2018;16:745-759.
    Pubmed KoreaMed CrossRef
  3. Lu M, Xuan S, Wang Z. Oral microbiota: a new view of body health. Food Sci Hum Wellness 2019;8:8-15.
  4. Aas JA, Paster BJ, Stokes LN, Olsen I, Dewhirst FE. Defining the normal bacterial flora of the oral cavity. J Clin Microbiol 2005;43:5721-5732.
    Pubmed KoreaMed CrossRef
  5. Faran Ali SM, Tanwir F. Oral microbial habitat a dynamic entity. J Oral Biol Craniofac Res 2012;2:181-187.
    Pubmed KoreaMed CrossRef
  6. Belstrøm D, Fiehn NE, Nielsen CH, et al. Differences in bacterial saliva profile between periodontitis patients and a control cohort. J Clin Periodontol 2014;41:104-112.
    Pubmed CrossRef
  7. Belstrøm D, Fiehn NE, Nielsen CH, et al. Differentiation of salivary bacterial profiles of subjects with periodontitis and dental caries. J Oral Microbiol 2015;7:27429.
    Pubmed KoreaMed CrossRef
  8. Simón-Soro A, Tomás I, Cabrera-Rubio R, Catalan MD, Nyvad B, Mira A. Microbial geography of the oral cavity. J Dent Res 2013;92:616-621.
    Pubmed CrossRef
  9. Dasanayake AP, Caufield PW, Cutter GR, Roseman JM, Köhler B. Differences in the detection and enumeration of mutans streptococci due to differences in methods. Arch Oral Biol 1995;40:345-351.
    Pubmed CrossRef
  10. Asikainen S, Alaluusua S, Saxén L. Recovery of A. actinomycetemcomitans from teeth, tongue, and saliva. J Periodontol 1991;62:203-206.
    Pubmed CrossRef
  11. Choi H, Kim E, Kang J, et al. Real-time PCR quantification of 9 periodontal pathogens in saliva samples from periodontally healthy Korean young adults. J Periodontal Implant Sci 2018;48:261-271.
    Pubmed KoreaMed CrossRef
  12. Kralik P, Ricchi M. A basic guide to real time PCR in microbial diagnostics: definitions, parameters, and everything. Front Microbiol 2017;8:108.
    Pubmed KoreaMed CrossRef
  13. Jung JY, Yoon HK, An S, et al. Rapid oral bacteria detection based on real-time PCR for the forensic identification of saliva. Sci Rep 2018;8:10852.
    Pubmed KoreaMed CrossRef
  14. Flink H, Bergdahl M, Tegelberg A, Rosenblad A, Lagerlöf F. Prevalence of hyposalivation in relation to general health, body mass index and remaining teeth in different age groups of adults. Community Dent Oral Epidemiol 2008;36:523-531.
    Pubmed CrossRef
  15. Iorgulescu G. Saliva between normal and pathological. Important factors in determining systemic and oral health. J Med Life 2009;2:303-307.
    Pubmed KoreaMed
  16. Flink H, Tegelberg A, Lagerlöf F. Influence of the time of measurement of unstimulated human whole saliva on the diagnosis of hyposalivation. Arch Oral Biol 2005;50:553-559.
    Pubmed CrossRef
  17. Ifo SA, Moutsambote JM, Koubouana F, et al. Tree species diversity, richness, and similarity in intact and degraded forest in the tropical rainforest of the Congo basin: case of the forest of Likouala in the Republic of Congo. Int J For Res 2016;2016:7593681.
  18. Belstrøm D, Holmstrup P, Bardow A, Kokaras A, Fiehn NE, Paster BJ. Comparative analysis of bacterial profiles in unstimulated and stimulated saliva samples. J Oral Microbiol 2016;8:30112.
    Pubmed KoreaMed CrossRef
  19. Schafer CA, Schafer JJ, Yakob M, Lima P, Camargo P, Wong DT. Saliva diagnostics: utilizing oral fluids to determine health status. Monogr Oral Sci 2014;24:88-98.
    Pubmed CrossRef
  20. Gomar-Vercher S, Simón-Soro A, Montiel-Company JM, Almerich-Silla JM, Mira A. Stimulated and unstimulated saliva samples have significantly different bacterial profiles. PLoS One 2018;13:e0198021.
    Pubmed KoreaMed CrossRef
  21. Caselli E, Fabbri C, D'Accolti M, Soffritti I, Bassi C, Mazzacane S, Franchi M. Defining the oral microbiome by whole-genome sequencing and resistome analysis: the complexity of the healthy picture. BMC Microbiol 2020;20:120.
    Pubmed KoreaMed CrossRef
  22. Sembler-Møller ML, Belstrøm D, Locht H, Enevold C, Pedersen AML. Next-generation sequencing of whole saliva from patients with primary Sjögren's syndrome and non-Sjögren's sicca reveals comparable salivary microbiota. J Oral Microbiol 2019;11:1660566.
    Pubmed KoreaMed CrossRef
  23. Motro Y, Moran-Gilad J. Next-generation sequencing applications in clinical bacteriology. Biomol Detect Quantif 2017;14:1-6.
    Pubmed KoreaMed CrossRef
  24. Abranches J, Zeng L, Kajfasz JK, et al. Biology of oral streptococci. Microbiol Spectr 2018;6:10.1128/microbiolspec.GPP3-0042-2018.
    Pubmed KoreaMed CrossRef
  25. Rinninella E, Raoul P, Cintoni M, et al. What is the healthy gut microbiota composition? A changing ecosystem across age, environment, diet, and diseases. Microorganisms 2019;7:14.
    Pubmed KoreaMed CrossRef
  26. Engen SA, Rørvik GH, Schreurs O, Blix IJ, Schenck K. The oral commensal Streptococcus mitis activates the aryl hydrocarbon receptor in human oral epithelial cells. Int J Oral Sci 2017;9:145-150.
    Pubmed KoreaMed CrossRef
  27. Lynge Pedersen AM, Belstrøm D. The role of natural salivary defences in maintaining a healthy oral microbiota. J Dent 2019;80 Suppl 1:S3-S12.
    Pubmed CrossRef
  28. Oscarsson J, Claesson R, Lindholm M, Höglund Åberg C, Johansson A. Tools of Aggregatibacter actinomycetemcomitans to evade the host response. J Clin Med 2019;8:1079.
    Pubmed KoreaMed CrossRef
  29. Kesavalu L, Sathishkumar S, Bakthavatchalu V, et al. Rat model of polymicrobial infection, immunity, and alveolar bone resorption in periodontal disease. Infect Immun 2007;75:1704-1712.
    Pubmed KoreaMed CrossRef
  30. How KY, Song KP, Chan KG. Porphyromonas gingivalis: an overview of periodontopathic pathogen below the gum line. Front Microbiol 2016;7:53.
    Pubmed KoreaMed CrossRef
  31. Griffen AL, Becker MR, Lyons SR, Moeschberger ML, Leys EJ. Prevalence of Porphyromonas gingivalis and periodontal health status. J Clin Microbiol 1998;36:3239-3242.
    Pubmed KoreaMed CrossRef
  32. Guven DC, Dizdar O, Alp A, et al. Analysis of Fusobacterium nucleatum and Streptococcus gallolyticus in saliva of colorectal cancer patients. Biomark Med 2019;13:725-735.
    Pubmed CrossRef
  33. Segata N, Haake SK, Mannon P, et al. Composition of the adult digestive tract bacterial microbiome based on seven mouth surfaces, tonsils, throat and stool samples. Genome Biol 2012;13:R42.
    Pubmed KoreaMed CrossRef
  34. Moon JH, Lee JH, Lee JY. Subgingival microbiome in smokers and non-smokers in Korean chronic periodontitis patients. Mol Oral Microbiol 2015;30:227-241.
    Pubmed CrossRef
  35. Kuriyama T, Nakagawa K, Kawashiri S, Yamamoto E, Nakamura S, Karasawa T. The virulence of mixed infection with Streptococcus constellatus and Fusobacterium nucleatum in a murine orofacial infection model. Microbes Infect 2000;2:1425-1430.
    Pubmed CrossRef
  36. Settem RP, El-Hassan AT, Honma K, Stafford GP, Sharma A. Fusobacterium nucleatum and Tannerella forsythia induce synergistic alveolar bone loss in a mouse periodontitis model. Infect Immun 2012;80:2436-2443.
    Pubmed KoreaMed CrossRef
  37. Kamma JJ, Nakou M, Manti FA. Microbiota of rapidly progressive periodontitis lesions in association with clinical parameters. J Periodontol 1994;65:1073-1078.
    Pubmed CrossRef
  38. Moraes LC, Fatturi-Parolo CC, Ferreira MB, Só MV, Montagner F. Saliva, supragingival biofilm and root canals can harbor gene associated with resistance to lactamic agents. Braz Oral Res 2015;29:52.
    Pubmed CrossRef
  39. Conrads G, de Soet JJ, Song L, et al. Comparing the cariogenic species Streptococcus sobrinus and S. mutans on whole genome level. J Oral Microbiol 2014;6:26189.
    Pubmed KoreaMed CrossRef
  40. Ingrassia I, Leplingard A, Darfeuille-Michaud A. Lactobacillus casei DN-114 001 inhibits the ability of adherent-invasive Escherichia coli isolated from Crohn's disease patients to adhere to and to invade intestinal epithelial cells. Appl Environ Microbiol 2005;71:2880-2887.
    Pubmed KoreaMed CrossRef