PPAR, RXR, and VDR Expressions in Diabetes

The present study protocol was approved by the ethical committee of the Medical Ethics Committee of Tokat Gaziosmanpasa University. The study was designed as a cross-sectional clinical study and conducted at Tokat Gaziosmanpaşa University Faculty of Dentistry Department of Periodontology. Informed consent was obtained from all participants, and detailed oral and radiological examination was performed by an experienced clinician (H.B.Y.).

Forty-two participants, 14 participants in each group, were enrolled in the study. The study groups were as follows.

Healthy controls (C, mean age 41.05±1.80, seven men, seven women) Periodontitis patients with stage 3 grade B, (P, mean age 42.35±1.92, seven men, seven women) Patients with diabetes and periodontitis with stage 3 grade C, (DP, mean age 41.09±1.28, seven men, seven women)

All patients and healthy individuals went through a detailed oral examination, and periodontal health and periodontitis diagnosis were based on the criteria defined by the recent classification of periodontal health and diseases in the 2017 International World Workshop for a Classification of Periodontal Diseases and Conditions [21].

All participants were never smokers. Exclusion criteria were conditions which might affect the inflammatory state, the existence of less than ten or over missing teeth, tobacco or any drug use, presence of pregnancy or lactation, and existence of antibiotic or periodontal therapy within previous six months. Diabetes patients in the present study had HbA1c values greater than 7.0 even though they were under diabetic treatment. And the diabetic patients who had HbA1c values below 7.0 were not included.

Clinical measurements Clinical parameters were plaque index (PI), gingival index (GI) and clinical attachment levels (CAL), and all analyses were performed before the biopsy procedure. Full mouth measurements and the measurements of the relevant teeth were recorded separately. CAL was measured from the cement-enamel junction to the bottom of the periodontal pocket and Williams's type periodontal probe was used for measurements (Hu-Friedy Co., Chicago, IL, USA). PI and GI measurements were performed via an explorer (Hu-Friedy Co., Chicago, IL, USA). All analyses were performed from six points, three from buccal aspect and three from the palatal aspect as mesial, middle, and distal sites for both teeth. A mean value of six measurements was recorded.

Collection of Gingival Samples Gingival biopsies were obtained according to a protocol by a previous study [22]. Two gingival biopsies were obtained from each participant. Gingival sampling in periodontally healthy individuals was performed by crown lengthening procedure in the routine treatment protocol or before orthodontically indicated tooth extraction. In the periodontitis groups, biopsies were obtained from the inflamed gingival tissue with minor surgical periodontal procedures. All biopsies were obtained from posterior maxillary teeth (tooth #4, #5, #6 or #7).

Firstly, local anesthesia was performed, and a gingival sample was dissected with a tissue scissor from an area with requiring either gingivectomy/crown lengthening procedures from patients with impaired gingival topography or before tooth extraction indicated for orthodontic treatment in healthy individuals. As for the periodontitis patients, a gingival sample was dissected using a tissue scissor during the scaling and root planning procedure from the gingival areas with severe gingival inflammation. One gingival sample from each participant was immediately placed in 10% neutral buffered formalin for 48 hours and underwent histological tissue processing. The other samples of each participant were immediately frozen at -80⁰C until the day of analysis.

Histopathological Evaluation All histological procedures were performed by an experienced blinded researcher (F.G.). Gingival samples which underwent histological tissue processing were first washed and cleansed from formalin and were dehydrated with alcohol series. Then all tissues were cleared with xylene and embedded in paraffin blocks. Each block sectioned throughout as serial sections and three sections from each block were selected for Hematoxylin-eosin (H&E) staining. Fibroblast and inflammatory cell infiltration were evaluated from H&E stained slides under 1000x magnification via a light microscope .

For fibroblast and inflammatory cell evaluation, the gingival area at the connective tissue border neighboring gingival epithelium was evaluated. A cell counting frame of 10.000 µm2 area was selected under 1000x magnification. The total inflammatory cells (neutrophil, lymphocyte, eosinophil, and macrophage cells) within the frame were counted as inflammatory cell counting. The fibroblasts were also counted likewise. The measurements were performed from three different points, and the mean of these three measurements was recorded [23].

PPAR-γ, RXR-α, and VDR Immunohistochemistry Immunohistochemistry was performed as reported in previous studies [23-26]. Three sections were chosen for each immunohistochemistry staining from each participant. Firstly, all slides were deparaffinized and dehydrated with alcohol series. After washing with distilled water, antigen retrieval was performed via sodium citrate buffer (pH 6.0) for two h at 70°C, and then endogenous peroxidase activity was suppressed with 3% hydrogen peroxide treatment. After hydrogen peroxide treatment, all slides were incubated with normal rabbit serum for 30 min. After normal serum incubation, primary antibody diluents were prepared and applied to the samples overnight at a humidified chamber at 4⁰C. The primary antibodies were goat polyclonal anti-PPAR-γ antibody (1:250), anti-RXR-α antibody (1:250), and anti-VDR antibody (1:250). After primary antibody incubation, all slides were washed with phosphate buffer solution (PBS) three times for five minutes (3x5) and biotinylated immunoglobulin G was applied for 30 min. Again a wash of 3x5 PBS, all samples were exposed to a streptavidin-horseradish peroxidase-conjugated reagent for another 30 min. After washed with 3x5 PBS, all sections were treated with AEC chromogen to visualize staining. Counterstaining was performed with Meyer's hematoxylin, and sections were mounted. After allowing dry for two days, samples were examined under 400x magnification using light microscopy [23, 24].

Immunohistochemical semi-quantitative H SCORE Analysis Immunohistochemistry evaluation was performed on three different areas in each section, and three sections were evaluated for each patient. Nine measurements were performed for each participant. All cells were marked according to their staining as no staining '0', low staining '1', mild staining '2', and dense staining '3'. Staining scores were converted to a numeric value as 'H score' through a formula (∑Pi(i+l)) which is a frequently used value for immunohistochemistry [23, 24]. In this formula, i shows the staining intensity score and Pi: indicates the percentage of the stained cells.

RT-PCR Analysis The primers to be used for the genes were shown in table 1. The RNA isolation and RT-PCR analysis performed according to a previous study [27]. Briefly, for RT-PCR analysis, first gingival tissues were homogenized and subjected to sonication to reveal RNA in the tissues and tissue supernatants were obtained. Then the supernatants were submitted to the total mRNA isolation step. The tissues treated with Qiazole, then allowed to incubate in the shaker for 5 minutes. At the end of 5 minutes, chloroform was added to the medium containing the cells and incubation continued for 3 minutes at room temperature and then the samples were centrifuged. At the end of the centrifuge, the supernatant at the top of the tube transferred to another tube. Afterward, isopropanol was added to the supernatant and incubated for 10 minutes at room temperature, and the centrifugation step was repeated. At the end of the centrifugation, the pellet deposited at the bottom was washed with 75% ethanol, remaining supernatant was centrifuged again and suspended in pure, sterile water without RNase. mRNAs were determined by spectrophotometric method. Then using the qRT-PCR kit, one µg mRNA were transformed into single-step cDNA using specific primers targeting the gene of interest, later the genes were identified with the help of qRT-PCR (Verso, Thermo Fisher Scientific, Massachusetts, USA). The results were normalized to the β-actin control gene.

Statistical Analysis The power of the study was over 85% based on a previous study with a similar study design [25]. IBM SPSS software was used for statistical analysis. One Sample K-S test was used to test the normality. For the fibroblast cell counts, inflammatory cell counts, and CAL, One Way ANOVA followed by Tukey test used. For other parameters, Mann Whitney U and Kruskal Wallis tests were used. Data were presented as mean ± SD or percentage as appropriate. p < 0.05 were considered statistically significant.