Disruption of cholesterol homeostasis triggers periodontal inflammation and alveolar bone loss

Human subjects

Human gingival tissue samples comprising both epithelial and connective tissue were collected from twelve subjects (34–65 years old; 49.17 ± 12.40 years), including six healthy individuals for non-inflamed gingival tissue and six patients with chronic periodontitis for inflamed gingival tissue. The research was accepted by the Institutional Review Board of Chonnam National University Dental Hospital (Gwangju, Republic of Korea, #DTMP-2022-005). Patients were explained fully about the procedures and confirmed in the consent. The tissue of gingiva was promptly preserved in liquid nitrogen and stocked at ‑80 °C until the experiments were performed.


C57BL/6 male and female mice were fed a high-cholesterol diet with or without cholesterol-lowering drugs. Homozygous Ch25h knockout (Ch25h−/−) mice30 were purchased from the Jackson Laboratory. Male C57BL/6 (wild-type and Ch25h−/−) mice were specifically used for the induction of experimental periodontitis. Male mice (five weeks old) were fed for 13 weeks with an AIN-76A diet (Feed Laboratory) as a regular diet. For the high-cholesterol diet, mice were fed for 13 weeks with a modified AIN-76A diet supplemented with 2% cholesterol. A group of mice fed the 2% cholesterol diet were injected intragingivally with 15 mg kg−1 DMSO (Sigma‒Aldrich, D2650) according to body weight as a vehicle or SR3335 (Cayman Chemical, 12072), a selective RORα inverse agonist, once per week for 13 weeks. All animals were housed in pathogen-free barrier facilities. The study protocols were approved by the Animal Care and Ethics Committee of Chonnam National University.

Experimental periodontitis (PD) induction in mice

An experimental PD model using male mice (10 weeks old) was established via silk ligature 5-0 (Ailee, SK521). The ligature encircled the right second molar in the maxilla, and the left side was used as the control31. Experimental PD was also induced via intragingival injection (once per day for 6 days) of adenoviruses Ad-Ch25h and Ad-Rorα (1 × 109 PFU in a total volume of 5 μl and 8 µl, respectively) between the first and the second molar of 12-week-old male mice using a Hamilton syringe (32 GA, 9.25 mm, 20°). The control side was injected with empty adenovirus (Ad-C). A group of mice with ligature-induced PD were injected intragingivally with 15 mg kg−1 according to body weight of SR3335 once per day for 7 days. The mice were sacrificed, and the maxilla was harvested for μCT and histological analyses after 9 days of ligature31. All adenoviral vectors were supplied by Vector Biolabs.

Primary cell culture

Primary human gingival fibroblasts (GFs) were collected from the marginal gingival tissue of clinically healthy individuals. Participants were notified about the purpose of this study, which was performed by the Institutional Review Board of Chonnam National University Hospital (#DTMP-2022-005). Gingival tissue was minced mechanically using a scalpel in phosphate-buffered saline solution (PBS) and instantly cultured in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, 12800-017) with 10% fetal bovine serum (FBS, Capricon; containing approximately <50 mg/dl cholesterol and 3.0 mg/dl LDL), 1% penicillin and 1% streptomycin. Human GFs were nurtured at 37 °C in a humidified atmosphere of 5% (v/v) CO2 and used for experiments at a maximum passage of 9. Cells were treated with cholesterol (cholesterol–methyl-β-cyclodextrin, Sigma‒Aldrich, C-4951) for 24 h or 36 h, 25-hydroxycholesterol (25-HC, Sigma‒Aldrich, H1015), Pg lipopolysaccharides (LPS) (InvivoGen, tlrl-pglps) and Ec LPS (Sigma‒Aldrich, L2630), IL1β (GenScript, 14 Z02922−10) or TNFα (GenScript, Z02774-20) for 24 h. Human GFs were infected with CH25H-, CYP7B1-, and RORα-overexpressing or control adenovirus (Ad-C) for 2 h, followed by culture for 48 h. Control or CH25H siRNA (Dharmacon RNA Technologies, L-008288-01) was transfected into human GFs seeded in six-well plates (2.5 × 105 cells per well) using Lipofectamine™ RNAiMAX (Invitrogen, 56532) according to the manufacturer’s protocol, in the presence of IL1β (2 ng ml−1) or TNFα (50 ng ml−1). Human GFs were treated with a cholesterol synthesis inhibitor [triparanol 1 μM or 5 μM (Cayman, 20918) or lovastatin 1 μM or 10 μM (Enzo, BML-G245)] or anti-CXCL16 blocking antibody 1 μg ml−1 or 2.5 μg ml−1 (R&D Systems, AF976) in the presence of IL1β (2 ng ml−1) or TNFα (50 ng ml−1).

Microcomputed tomography (μCT)

The maxillae were collected and fixed in 10% neutral buffered formalin (NBF) overnight. The structure of alveolar bone was evaluated by using the SkyScan 1172 μCT scanner (SkyScan) with the following parameters: 49 kV, 200 μA, 0.7 mm aluminum filter, and 11 μm resolution. μCT data were analyzed between the first and second molar regions on the buccal and lingual sides after drawing the region of interest (ROI) using CTAn (SkyScan) and Mimics 14.0 (Materialise).

Histology and immunohistochemistry

Gingival tissues from patients and mice with or without PD were fixed in 10% NBF for 24 h, embedded in paraffin, and sectioned at 5 μm thickness. Mouse maxillary tissues were decalcified with 0.5 M ethylenediamine tetra-acetic acid (EDTA, pH 8.0) for 2 weeks, embedded in paraffin, and sectioned at 5 μm thickness. Alveolar bone erosion was evaluated via hematoxylin and eosin (H&E) staining. For immunohistochemical staining, endogenous peroxidases were quenched in sections of human and mouse periodontal tissues, and epitope retrieval was induced by 3% H2O2 for 10 min and 0.1% trypsin for 30 min at 37 °C. The sections were incubated with mouse anti-MMP1 (Novus, NBP2-22123; 1:150), rabbit anti-IL6 (Novus, NB600−1131; 1:150), rabbit anti-IL8 (Biorbyt, orb229133; 1:150), rabbit anti-PTGS2 (Cayman, 160106; 1:150), rabbit anti-RORα (Sigma‒Aldrich, AV45608; 1:150), rabbit anti-CH25H (Bioss, bs-6480R; 1:150), and rabbit anti-CXCL16 (Genetex, GTX116706; 1:150), followed by staining with EnVision+System-HRP and AEC+ substrate (Dako, K4005, K4009) and counterstaining with hematoxylin (Dako, S3309). Images were captured by using a Zeiss Axio Scope A1 microscope and quantified by ImageJ software (National Institutes of Health, v.1.51a).

Cholesterol quantification and lipid profiling

For determination of cellular cholesterol levels, human GFs were treated with IL1β, TNFα, cholesterol synthesis inhibitors (triparanol (Cayman) and lovastatin (Enzo), or anti-CXCL16 blocking antibody (R&D Systems) in the presence of IL1β (2 ng ml−1), TNFα (50 ng ml−1) or cholesterol (200 μM). The quantity of intracellular cholesterol was determined by a cholesterol quantification kit (Biomax, BM-CHO−100). Briefly, cellular lipids were extracted with a mixture of chloroform, isopropanol, and NP-40 (7:11:0.1) in a microhomogenizer and centrifuged for 10 min at 15,000 × g. The organic phase was collected and dried under vacuum at 55 °C for 12 h. From the lipid extract obtained, the amounts of total and free cholesterol were determined using a total cholesterol quantification kit following the manufacturer’s instructions (Biomax). The cholesteryl ester content was calculated as the total cholesterol minus free cholesterol.

Cholesterol and lipoprotein imaging

The influx of cholesterol or oxidized low-density lipoprotein (OxLDL) into the cell was evaluated via microscopy. Briefly, human GFs were seeded on a poly-L-lysine-coated culture cover glass (Paul Marienfeld GmbH & Co. KG, 0111520) and stimulated with 5 μg ml−1 1,1-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate oxidized low-density lipoprotein (Dil-OxLDL, Kalen Biomedical, 770232-9) for 24 h. After washing three times with PBS and fixing with 4% ice-cold fresh paraformaldehyde (PFA, Sigma‒Aldrich, P6148), cell nuclei were stained with DAPI (l μg ml−1) for 10 min. For cholesterol staining in periodontal tissue, frozen sections of gingival tissues from human patients and experimental PD mice (10 μm) were washed in PBS and fixed in 4% fresh PFA. Gingival tissue sections were stained with filipin (250 mg ml−1; Sigma‒Aldrich, F9765) and washed twice with PBS. All procedures were performed in the dark and at room temperature. Human GFs and periodontal tissues were examined by using a Zeiss Axio Scope A1 microscope attached to the fluorescence part, and the acquired images were analyzed using ImageJ software.

Enzyme-linked immunosorbent assay (ELISA)

To quantify the secretion of MMP1, IL6, IL8 and the production of PGE2 from human GFs, ELISA kits for MMP1 (RayBiotech, #ELH-MMP1−1), IL6 (R&D Systems, D6050), IL8 (R&D Systems, D8000C) and PGE2 (R&D Systems, KGE004B) were used. Briefly, human GFs were cultured without serum under each experimental set of conditions. The culture medium was collected and centrifuged for 10 min at 1200 × g, and the supernatant fractions were stored at ‑80 °C until analysis. The MMP1, IL6, IL8, and PGE2 levels were determined following the manufacturer’s instructions.

RNA isolation and real-time quantitative polymerase chain reaction (RT‒qPCR)

Total RNA from primary cultures of human GFs and gingiva was isolated using TRIzol reagent (Sigma‒Aldrich, 93289). The synthesis of cDNA was performed from 0.25 to 0.5 μg of RNA via reverse transcription (Promega, A3803) and subjected to PCR (Geneall, 501-025). The Supplementary Table lists the PCR primers and experimental conditions. Quantitative RT‒PCR was executed with SYBR Premix Ex Tag (Takara Bio, RR420A) using a StepOnePlus Real-Time PCR system (Thermo Fisher Scientific). All qRT-PCR reactions were performed in duplicate, and fold changes were calculated using relative quantification methods with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) serving as an internal control.

Western blotting

Human GFs were lysed in RIPA buffer, a protease inhibitor cocktail (Roche, 04906845001), and a phosphatase inhibitor cocktail (Roche, 11697498001). Briefly, 30 µg of protein was separated by SDS‒PAGE using 8%, 10%, and 15% PAGE gels, and the blots were incubated with the indicated antibodies. As a loading control, mouse anti-β-actin (Sigma‒Aldrich, A3854) was used. The membranes were incubated with the following antibodies: rabbit anti-MMP1 (Abnova, PAB12708), rabbit anti-IL6 (Novus), rabbit anti-IL8 (Biorbyt), rabbit anti-PTGS2 (Cayman), rabbit anti-RORα (Santa Cruz, Sc-28612), and rabbit anti-CH25H (Bioss). After washing with Tris-buffered saline with 0.1% Tween (TBS-T), the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (anti-rabbit IgG; Sigma‒Aldrich, A6154) and detected using an ECL solution (Cytiva, RPN2235). Images were obtained using ImageSaver6 software via an EZ capture MG system (ATTO) and were quantified using ImageJ software.

Reporter gene assay

The RORE motif (50 ng per well, ActiveMotif, 32175) was transfected into human GFs using LipofectamineTM RNAiMAX (Invitrogen). After recovery from transfection, human GFs were stimulated with IL1β (1 ng ml−1), TNFα (25 ng ml−1), cholesterol (100 μM), and 25-HC (20 μM) or were infected with 800 MOI of Ad-C or Ad-CH25H for 36 h. Renilla luciferase activities were measured by the LightSwitchTM Luciferase assay kit (ActiveMotif, 32031).

Microarray analysis

Total RNA was extracted from human GFs treated with IL1β (2 ng ml−1) or TNFα (50 ng ml−1), and the concentration, purity, and integrity were verified by spectrophotometry. Three replicates for each experimental group were isolated and processed. RNA from human GFs was analyzed using the Affymetrix GeneChip Mouse Gene 2.0 ST Array following the Affymetrix protocol (Macrogen). Data were normalized using algorithms supplied with the feature extraction software. The cutoff values for the identification of differentially expressed genes were an adjusted P value below 0.05 (FDR < 0.05) and an absolute value of fold change over 1.5 (|FC|>1.5).

Statistical analysis

All experiments were repeated at least three times. Values are presented as the mean ± SEM. Statistical analyses were analyzed by GraphPad Prism version 8 (GraphPad Software, Inc.). Quantified data were initially tested for conformation to a normal distribution using the Kolmogorov‒Smirnov or Shapiro‒Wilk test, followed by analysis with two-tailed Student’s t-test for comparison of means between two groups or analysis of variance (ANOVA) followed by Tukey’s post hoc test (multi-comparison) for comparison of means among three or more independent groups, as appropriate. The n value represents the number of independent experiments or mice. Differences with P values < 0.05 were assigned statistically significant.

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *