The Associations of Periodontopathic Bacteria and Oral Candida with Periodontal Inflamed Surface Area in Older Adults Receiving Supportive Periodontal Therapy

The periodontal inflamed surface area (PISA) has been proposed for assessment of the total periodontal inflammatory status in people with periodontitis. This study was performed to investigate the associations of periodontopathic bacteria and candida with PISA in older people. We enrolled 100 patients aged ≥ 60 years who visited Hiroshima University Hospital. PISA and periodontal epithelial surface area (PESA) were calculated in each patient. Oral rinse samples were collected for DNA extraction. Periodontopathic bacteria and candida were detected by polymerase chain reaction. The mean values of PISA and PESA were significantly greater in T. forsythia-positive patients than in T. forsythia-negative patients. T. forsythia/C. albicans double-positive patients exhibited significantly greater PISA values than did non-double-positive patients. Additionally, PISA values were significantly greater in T. forsythia//T. denticola/C. albicans triple-positive patients than in T. forsythia//T. denticola/C. albicans non-triple-positive patients (p = 0.02). Propensity score-matching was performed between periodontopathic bacteria-positive and -negative patients using propensity scores generated from clinical factors. Importantly, T. forsythia/T. denticola double-positive patients exhibited significantly greater PISA values than non-double-positive patients among 72 propensity score-matched patients. Our preliminary results highlight the importance of the presence of T. forsythia and T. denticola for periodontal inflammation severity in older Japanese people.


Introduction
A dynamic balance between oral microorganisms and the host (involving genetics, host immune response, and circadian rhythm) is critical for the maintenance of oral health [1]. Periodontitis is a polymicrobial infectious disease mainly mediated by dysbiosis of the subgingival microbiota [2]. Specific complexes of bacteria are important for the onset of periodontitis [3]. So-called red complex bacteria such as Treponema denticola, Tannerella forsythia, and Porphyromonas gingivalis are associated with severe periodontitis [3]. Both P. gingivalis and T. denticola are present in subgingival plaque at deep periodontal pockets [4]. Microbial colonization with red complex bacteria is preceded by colonization with so-called orange complex bacteria (i.e., Fusobacterium subspecies, Prevotella intermedia, Prevotella nigrescens, Peptostreptococcus micros, Campylobacter showae, Campylobacter gracilis, Campylobacter rectus, Streptococcus constellatus, and Eubacterium nodatum) [3]. These red and orange complex bacteria have predominant roles in the etiology of periodontitis. Aggregatibacter actinomycetemcomitans is an important pathogenic species associated with localized aggressive periodontitis [5]. In addition, the presence of oral candida species is related to periodontitis [6,7]. Therefore, both periodontopathic bacteria and oral candida are presumed to initiate active inflammation in periodontal tissues.
The periodontal inflamed surface area (PISA) has been proposed for assessment of the total periodontal inflammatory status in people with periodontitis [8]. PISA is calculated by measuring clinical attachment level, recession, and bleeding on probing; this measurement reflects the surface area of bleeding pocket epithelium in square millimeters [8]. Bleeding on probing is regarded as a vital sign of the presence of local periodontal inflammation [8]. Therefore, PISA can be employed to quantitatively assess the inflammatory burden of periodontitis. However, the associations of periodontopathic bacteria and oral candida with PISA have not been fully elucidated in older Japanese people. The objective of this study was to investigate the associations of periodontopathic bacteria (e.g., P. gingivalis, T. forsythia, and T. denticola) and Candida albicans with PISA in older people.

Patients
In this study, we enrolled 100 patients aged ≥ 60 years who visited Hiroshima University Hospital from 2018 to 2019 and received supportive periodontal maintenance therapy. No patients had oral lesions, such as oral cancer or pre-malignant lesion (i.e., leukoplakia). In addition, no patients had oral pseudomembrane formation and erythematous or atrophic lesions, which are characteristic findings of oral candidiasis. No patients received antibiotics or anti-inflammatory drugs (e.g., steroids) at the time of examination. The study protocol was approved by the Ethical Committee of Hiroshima University (approval no. E-1115) and all patients provided written informed consent to participate. A schematic diagram of this study is shown in Figure 1. Medical comorbidities, including cardiovascular diseases (e.g., heart disease and arrhythmia), hypertension, diabetes, and dyslipidemia, were among the clinical factors assessed in this study. Regarding smoking status, we divided patients into three groups: non-smokers (people who had never smoked in their lifetime), current smokers (people who had smoked daily or occasionally), and former smokers (people who had smoked in their lifetime, but who had quit smoking before the examination).

Oral Rinse Sample Collection and DNA Extraction
Patients were asked to rinse their mouths with saline for 15 s, then expel the rinse samples into sterile plastic tubes. Collected samples were centrifuged and supernatants were decanted; the resulting pellets were stored at −80 • C. DNA was extracted using a PureLink™ Microbiome DNA Purification kit (Thermo Fisher Scientific, Waltham, MA, USA), in accordance with the manufacturer's protocol.

Statistical Analysis
The Mann-Whitney U test or Kruskal-Wallis test was used to compare continuous numerical variables between groups. The χ 2 test or Fisher's exact test was used to compare categorical variables between groups. Binomial logistic regression analysis was conducted using a forced entry method. Binomial logistic regression analysis was performed using the presence of periodontal bacteria/candida as the dependent variable and variables with a p-value of <0.20 in univariate analysis as independent variables. A propensity score-matched analysis was performed to eliminate the effects of clinical confounding factors. Propensity scores were calculated by logistic regression analysis of eight clinical parameters (i.e., age and sex; cardiovascular disease, hypertension, diabetes, dyslipidemia, and smoking statuses; and number of remaining teeth). IBM SPSS Statistics, version 24.0 (IBM Corp., Armonk, NY, USA) was used for statistical analysis. To calculate the sample size of propensity score-matched patients required for paired t-tests, G*Power (version 3.1.9.4, Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany) was used with an effect size of 0.5, 80% statistical power, and a significance level of 0.05. The minimum sample size required for paired t-tests was 34 patients per group. For all analyses, p < 0.05 was considered statistically significant.

Associations of PISA and PESA with Clinical Factors in Older People
Amplified PCR products indicating the presence of P. gingivalis, T. forsythia, T. denticola, and C. albicans are shown in Figure 2. Of the 100 patients in this study, P. gingivalis was detected in 56 (56.0%), T. forsythia was detected in 83 (83.0%), T. denticola was detected in 49 (49.0%), and C. albicans was detected in 39 (39.0%). Table 2 summarizes the associations of PISA and PESA with clinical factors. PESA values were significantly greater in smokers than in non-smokers (p = 0.04). Moreover, PISA values were greater in patients with diabetes or hypertension than in patients without those comorbidities. However, no significant associations were found between PISA and clinical factors. Polymerase chain reaction (PCR) detection of periodontopathic bacteria and candida. PCR was performed to detect DNA from P. gingivalis, T. forsythia, T. denticola, and C. albicans. PCR products were electrophoresed on 2% agarose gels, stained with ethidium bromide, and exposed to ultraviolet light using an ultraviolet transilluminator.  Figure 3 summarizes the associations of periodontopathic bacteria with PISA and PESA. The mean values of PISA and PESA tended to be greater in patients with P. gingivalis than in patients without P. gingivalis; however, these differences were not statistically significant. PISA values were significantly greater in patients with T. forsythia than in patients without T. forsythia (p = 0.01). The mean values of PISA tended to be greater in patients with T. denticola than in patients without T. denticola; however, this difference was not statistically significant. PISA values were significantly greater in T. forsythia/T. denticola double-positive patients than in T. forsythia/T. denticola non-double-positive patients (i.e., both double-negative and single-positive patients) (p = 0.01). In addition, PISA values were significantly greater in P. gingivalis/T. forsythia/T. denticola triple-positive patients than in non-P. gingivalis/T. forsythia/T. denticola triple-positive patients (i.e., triple-negative, single-positive, or double-positive patients) (p = 0.03). Significant differences in PESA were found between T. denticola-positive and -negative patients (p = 0.02), between P. gingivalis/T. denticola double-positive patients and non-P. gingivalis/T. denticola double-positive patients (p = 0.02), between T. forsythia/T. denticola double-positive patients and non-T. forsythia/T. denticola double-positive patients (p = 0.02), and between P. gingivalis/T. forsythia/T. denticola triple-positive patients and non-P. gingivalis/T. forsythia/T. denticola triple-positive patients (p = 0.03).

Associations of the Combined Presence of Periodontopathic Bacteria and C. albicans with PISA in Older People
No significant associations were found between C. albicans and PISA or PESA values ( Figure 4). Next, the associations of the combined presence of periodontopathic bacteria and C. albicans with PISA and PESA were investigated. PISA values were greater in periodontopathic bacteria/C. albicans double-positive patients than in periodontopathic bacteria/C. albicans non-double-positive patients ( Figure 4). PISA values were significantly greater in T. forsythia/C. albicans double-positive patients than in T. forsythia/C. albicans non-double-positive patients (p = 0.04) ( Figure 4). Notably, PISA values were significantly greater in T. forsythia//T. denticola/C. albicans triple-positive patients than in T. forsythia//T. denticola/C. albicans non-triple-positive patients (p = 0.02) (Figure 4).

Associations of the Presence of Periodontopathic Bacteria/Candida with Clinical Factors
The results of binomial logistic regression analysis with the presence of periodontal bacteria/candida as dependent valuable are shown in Supplementary Tables S1-S15. Binomial logistic regression analysis revealed that the presence of T. denticola was significantly associated with PESA (odds ratio = 1.00, p = 0.048). The presence of P. gingivalis/C. albicans was significantly associated with the number of remaining teeth (odds ratio = 0.91, p = 0.03). Additionally, the presence of P. gingivalis/T. forsythia/C. albicans was significantly associated with the number of remaining teeth (odds ratio = 0.91, p = 0.03). However, no statistically significant relationship was observed between the presence of periodontal bacteria/candida and PISA.
Finally, the associations of the presence of P. gingivalis, P. gingivalis/T. forsythia, and T. forsythia/T. denticola with PISA were investigated in propensity score-matched patients. None of the eight clinical variables were significantly associated with the presence of periodontopathic bacteria (Tables 3-5). Thirty-six P. gingivalis-positive patients exhibited greater PISA values than P. gingivalis-negative patients. However, the presence of P. gingivalis was not significantly associated with PISA (Table 3, Figure 5). In addition, P. gingivalis and T. forsythia double-positivity was not significantly associated with PISA in 72 propensity score-matched patients (Table 4, Figure 5). Importantly, 36 T. forsythia and T. denticola double-positive patients exhibited significantly greater PISA and PESA values than T. forsythia and T. denticola non-double-positive patients (p = 0.02 and p = 0.03, respectively) (Table 5, Figure 5).

Discussion
Significant correlations have been found between PISA and the severity of periodontitis or periodontal indexes (e.g., bleeding on probing and plaque accumulation) in systemically healthy patients with periodontitis [13]. To the best of our knowledge, no other studies have investigated the associations of periodontopathic bacteria and oral candida with PISA in Japanese older people. In this study, propensity score-matching analysis revealed that PISA values were greater in the T. forsythia/T. denticola double-positive group than in the T. forsythia/T. denticola non-double-positive group. Additionally, PISA values tended to be greater in the P. gingivalis/T. forsythia double-positive group than in the P. gingivalis/T. forsythia non-double-positive group, but this difference was not statistically significant. These results suggest that co-infection with T. forsythia and T. denticola is a critical factor associated with chronic periodontal inflammation in Japanese older people.
T. denticola is periodontal pathogen that contributes to the severity of periodontitis by enhancement of alveolar bone resorption [14]. T. denticola has several virulence factors such as dentilisin [15], the major surface protein [16], and factor H-binding protein B [17], which have been implicated in the progression of periodontitis. Importantly, the major surface protein of T. denticola plays a role in bacterial adherence to epithelial cells and components of the extracellular matrix [18]. Additionally, T. forsythia has a variety of notable virulence factors, including Bacteroides surface protein A [19], a trypsin-like protease [20], glucosidases [21], forsythia detaching factor [22], and hemagglutinin [23].
T. forsythia is frequently present along with P. gingivalis in patients who exhibit active periodontitis [24,25]. The growth of T. forsythia is accelerated when it is co-cultured with F. nucleatum [26]. Notably, the virulence of T. forsythia lipopolysaccharide was weakened upon co-culture with P. gingivalis [27]. These results imply that the virulence of T. forsythia is affected by co-infection with other periodontopathic bacteria. Lanza et al. reported that T. denticola was associated with greater gingival bleeding, while T. forsythia was associated with chronic periodontitis severity, suggesting that T. denticola and T. forsythia may be the main red complex members involved in periodontitis progression [28].
Virulence factors present within Candida species include extracellular hydrolytic enzymes (i.e., phospholipase, proteinases, lipases, and coagulase) [29]. Phospholipase enzyme digests membrane phospholipids of the host cell, thereby promoting cell lysis and subsequent infection [29]. C. albicans-produced metallopeptidase presumably aids in periodontitis progression by degrading the extracellular matrix [30]. In this study, PISA values were significantly greater in the T. forsythia/C. albicans double-positive group than in the T. forsythia/C. albicans non-double-positive group. Additionally, T. forsythia//T. denticola/C. albicans triple-positive patients showed significantly greater PISA values than non-triplepositive patients. Although the biological association between periodontopathic bacteria and C. albicans remains unknown, C. albicans may enhance periodontal inflammation in the presence of periodontopathic bacteria.
The presence of inflamed and ulcerated subgingival epithelium in the periodontal pocket provides the opportunity for bacterial endotoxins to disseminate into the bloodstream, which triggers inflammatory processes and the impairment of distant organs [31]. Thus far, multiple studies have shown associations between PISA and systemic diseases [32][33][34][35]. PISA was identified as a significant predictor of HbA1c level in healthy people without diabetes [32]. Diabetes is a known risk factor for periodontitis [36]. Although no significant association was observed between diabetes and PISA in this study, PISA values were greater in patients with diabetes than in patients without diabetes, indicating that diabetes is a major risk factor for periodontitis. Elevated PISA values and bleeding on probing have been associated with high blood pressure [33]. A 5-year longitudinal study revealed that greater PISA values were significantly associated with an elevated incidence of mild cognitive impairment in community-dwelling older people, suggesting that chronic periodontal inflammation contributes to the deterioration of cognitive function in older people [34]. Finally, elevated PISA values were independently predictive of poor prognosis in patients with lacunar infarcts [35]. These results suggest that elevated PISA values are associated with periodontitis severity and the presence of systemic disease; thus, PISA may be a valuable prognostic factor for some lifestyle-related diseases.
This study had an important limitation in terms of the sampling method; namely, oral rinse samples contain bacteria derived from various sites in the oral cavity. Thus, subgingival plaque and crevicular fluid samples may be more appropriate to determine the presence of periodontopathogens in the periodontal pocket. Further studies are needed regarding the relationship between periodontopathogens and periodontal inflammation to enable generalization of our findings.

Conclusions
T. forsythia/T. denticola co-infection was significantly associated with greater PISA values in older Japanese people after adjustment for potentially clinical confounding factors using a propensity score-matching method. T. forsythia and T. denticola are involved in chronic periodontal inflammation in older people. The preliminary results in this study highlight the important roles of several periodontal pathogens in severe periodontal inflammation in older people. However, the impact of oral dysbiosis on periodontal inflammation has not been fully elucidated. Therefore, further studies should focus on clarifying the associations of oral microbiota with PISA.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/diagnostics11081397/s1, Table S1: Multivariable analysis with the presence of P. gingivalis as dependent valuable, Table S2: Multivariable analysis with the presence of T. forsythia as dependent valuable, Table S3: Multivariable analysis with the presence of T. denticola s as dependent valuable, Table S4: Multivariable analysis with the presence of P. gingivalis/T. forsythia as dependent valuable, Table S5: Multivariable analysis with the presence of P. gingivalis/T. denticola as dependent valuable, Table S6: Multivariable analysis with the presence of T. forsythia/T. denticola as dependent valuable, Table S7: Multivariable analysis with the presence of P. gingivalis/T. forsythia/T. denticola as dependent valuable, Table S8: Multivariable analysis with the presence of C. albicans as dependent valuable, Table S9: Multivariable analysis with the presence of P. gingivalis/C. albicans as dependent valuable, Table S10: Multivariable analysis with the presence of T. forsythia/C. albicans as dependent valuable, Table S11: Multivariable analysis with the presence of T. denticola/C. albicans as dependent valuable, Table S12: Multivariable analysis with the presence of P. gingivalis/T. forsythia/C. albicans as dependent valuable, Table S13: Multivariable analysis with the presence of P. gingivalis/T. denticola/C. albicans as dependent valuable, Table S14: Multivariable analysis with the presence of T. forsythia/T. denticola/C. albicans as dependent valuable, Table S15: Multivariable analysis with the presence of P. gingivalis/T. forsythia/T. denticola/C. albicans as dependent valuable.