Sirtuin 3 deficiency exacerbates age‐related periodontal disease

Abstract Background Sirtuin 3 (SIRT3), a mitochondrial NAD+‐dependent deacetylase, has received much attention for its effect on metabolism and aging. However, the role of SIRT3 in periodontal disease remains unknown. Objective This study aimed to investigate the functional role of SIRT3 in age‐related periodontal disease and underlying mechanisms. Methods Sixteen mice were randomly assigned into four groups: the young wild type (WT), the aged WT, the young SIRT3‐knockout (KO), and the aged SIRT3‐KO. SIRT3 and cyclophilin D (CypD) expression and protein lysine acetylation levels in alveolar bones were detected by western blot. The bone architecture and the distance of CEJ‐ABC were assessed using micro‐CT and HE staining. The osteoclast number was observed through tartrate‐resistant acid phosphatase (TRAP) staining. Mitochondrial morphology in SIRT3 knockdown MC3T3‐E1 osteoblastic cells was analyzed by Immunofluorescence staining. In gingival tissues, the NAD+/NADH ratio was measured, and oxidative stress was detected by MitoSOX staining, HO‐1 staining, and MnSOD expression. Mitochondrial biogenesis was measured by PGC‐1α expression and oxygen consumption rate (OCR). Results In parallel with the imbalanced NAD+/NADH ratio, the SIRT3 expression was significantly decreased in the alveolar bones of the aged mice, accompanied by a global elevation of protein acetylation levels. The aged SIRT3‐KO group showed the highest rate of bone resorption and the largest number of TRAP‐positive osteoclasts among the four groups. Moreover, the reactive oxygen species level was up‐regulated in the young and the aged SIRT3‐KO groups. SIRT3 deficiency promoted mitochondrial fission and increased the CypD expression. Furthermore, the lack of SIRT3 reduced the PGC‐1α expression in gingival tissues and exhibited a significant reduction in maximal OCR. Conclusion Reduced SIRT3 abundance contributes to aged‐related periodontal disease via the exacerbation of oxidative stress and mitochondrial dysfunction.


| INTRODUC TI ON
Aging, a common phenomenon to all multicellular organisms, is described as an endogenous and progressive decay during the whole life span. 1 Some chronic diseases, such as diabetes, cancer, cardiovascular disease, and neurodegenerative disease, are associated with chronological aging. Periodontal disease, also related to age, is an infectious disease resulting in inflammation within the toothsupporting tissues, progressively diminished tooth attachment and alveolar bone loss. 2,3 The "Mitochondrial Theory of Aging" posits that a decline in mitochondrial function over time is an underlying cause of aging. 4,5 Sirtuins (SIRT1-7) are a family of nicotinamide adenine dinucleotide (NAD + )-dependent protein deacetylases 6 that regulate multiple cellular processes. 7,8 SIRT3 is a major mitochondria-localized deacetylase and controls mitochondrial protein acetylation, 9,10 participating in the progress of human aging. 11 It initiates robust deacetylase activity toward a series of metabolic targets, including subunits of the electron transport chain (ETC), as well as enzymes involved in redox balance, tricarboxylic acid cycle, and ketone body production. [12][13][14][15] Reactive oxygen species (ROS) is a natural by-product of cellular respiration, 16 that increases dramatically with age. 17 Mitochondria are both the major source and principal attack target of ROS. 18 A previous study reported that SIRT3 can reduce ROS levels by modulating key antioxidant enzymes directly, acting as a shield against oxidative damage. 11 At higher ROS levels, longer mitochondrial permeability transition pore (mPTP) openings may release a ROS burst leading to mitochondria destruction. 19 Cyclophilin D (CypD), an undisputed regulator of the mPTP, is located in the mitochondrial matrix, which can be translocated to the inner mitochondrial membrane. 20 The evidence indicates that CypD can be deacetylated by SIRT3, resulting in an inhibited mPTP opening, 21-23 but whether SIRT3 regulates CypD involved in the pathogenesis of periodontal disease has not been evaluated.
Although the crucial role of SIRT3 has been confirmed, limited in vivo studies have explored its specific effects in periodontal tissues.
The current study was designed to investigate the functional role of SIRT3 in the pathogenesis of age-related periodontal disease and to explore underlying mechanisms. were used and kept on a 12:12-h light-dark cycle at 22°C with water and food ad libitum.
Immuno-reactive protein bands were visualized using a chemiluminescence machine (Bio-Rad,). Band relative optical density was detected by NIH Image J software and normalized with GAPDH levels.

| Micro-CT analysis and alveolar bone loss measurement
The harvested right mandibles were fixed in 10% formaldehyde for 72 h, and then scanned by μCT 50 (Scanco Medical) under medium resolution with a voxel size of 10 μm and an energy setting of 90 kV, 114 mA, and an integration time of 500 ms. Raw images were reconstructed and analyzed with SCANCO Medical Evaluation and Visualizer software. The region of interest (ROI) was defined as the buccal-lingual crossline of the first mandibular molar in the furcation zone (between the mesial and distal roots). Ten slices before and after the identified furcation slice were added to generate a ROI, and the bone volume per total volume (BV/TV, %) was calculated within it. After scanning, the degree of alveolar bone resorption was evaluated by the threedimensional image analysis software, in accordance with the previous study. 24 The distance from the cemento-enamel junction (CEJ) to the alveolar bone crest (ABC) of the first molar was measured, and the average of six sites (mesio-buccal, buccal, disto-buccal, disto-lingual, lingual, and mesio-lingual) was used as a measurement of bone loss.
The mandibles were then air dried. Each jaw was oriented and the distance from the CEJ to the ABC was observed, using a stereomicroscope (Leica EZ4HD, Leica Microsystems AG) equipped with a digital camera. Recordings were made in the long axis of lingual root surfaces of the three molars. The photos were captured by SPOT RT software (Spot Diagnostic Instruments,).

| Hematoxylin-eosin and TRAP staining
The mandibles were decalcified in a 10% EDTA solution for 4 weeks at 4°C, dehydrated with increasing concentrations of ethanol, and embedded in paraffin wax. Serial sagittal sections at a thickness of

| Quantitative real-time PCR
For the in vivo experiments, the left side of the mandibular bone was snap-frozen in liquid nitrogen and kept at 80°C until use. The total RNA was extracted from the mandibular bone via an RNA Extraction Kit (TAKARA). The concentration and purity of extracted RNA were evaluated by a spectrophotometer (Thermo Scientific NanoDrop).
After the cDNA was synthesized, real-time quantitative PCR was performed using a SYBR ® Premix Ex Taq ™ II (TAKARA) on an ABI QuantStudio 3 PCR System (Applied Biosystems). The primer sequences for Runx2, ALP, OPG, RANKL, IL-1β, IL-6 and TNF-ɑ, MnSOD, PGC-1α and GAPDH are listed in Table 1. The CT values of detected genes were calculated in relation to GAPDH by the 2 −∆∆CT method.
For the in vitro experiments, the total RNA was extracted from shCtrl and shSIRT3 group cells using Trizol (Life Technologies), and cDNA samples containing 1000 ng RNA were synthesized using the iScript TM reverse Transcription Supermix (Bio-rad). The PCR reaction was performed with iTaqTM Μniversal SYBR Green Supermix (Bio-rad) and run in a CFX Connect quantitative PCR instrument.
Gene expression results were expressed as arbitrary units relative to the expression of Actin. The primer sequences for SIRT3, OPG, Ndufs4, and Actin are listed in Table 1. The calculations were performed following the 2 −ΔΔCT method, where the Ctrl was calculated as the average CT value deriving from the Ctrl group.

| MitoSOX staining
The mitochondrial ROS level in gingival tissues was measured by MitoSOX Red, a specific fluorochrome that selectively reacts with superoxide in the mitochondria. Fresh gingivae around the mandibular first molars were harvested and immediately incubated with 5 μM MitoSOX Red (Molecular Probes, Thermo) at 37°C for 30 min and protected from light. Then, the gingivae were prepared for the frozen section procedure. Meanwhile, the nuclei were stained with DAPI (Invitrogen). Fluorescent images were obtained using a Laser Scanning Confocal Microscope (LSCM, OLYMPUS FV3000) and quantified using the NIH Image J software. The intensity was normalized to the young wild-type (WT) group.

| NAD + /NADH Measurements
Gingival levels of NAD + and NADH were measured using tissue homogenates with the EnzyChrom NAD + /NADH Assay Kit, according

Genes
Forward primer Reverse primer
Optical density at 565 nm was recorded at time zero and at 15 min using a 96-well plate reader spectrophotometer. The NAD + and NADH concentrations were determined by comparing the difference in absorbance with standard curves, and they were normalized to the WT-young group.

| Immunohistochemistry analysis
Gingival tissues were fixed, immersed in paraffin and processed for sectioning. The sections were dewaxed and rehydrated. After anti-

| Immunofluorescence microscopy and mitochondrial morphology analysis
After treatment as described above, MC3T3-E1 cells were seeded on glass coverslips in 6-well plates and incubated overnight at 37°C. This application determines the mean area, perimeter, major radius, and minor radius of cell mitochondria. Mitochondrial fusion/fission was evaluated using the form factor (FF = 4π × area/perimeter 2 ) and the aspect ratio (AR = major radius/minor radius) 26 after top hat filtering was applied to the raw images to remove noise and to obtain a precise definition of the mitochondrial morphology.

| Measurement of the oxygen consumption rate
The oxygen consumption rate (OCR) of MC3T3-E1 osteoblas-

| Statistics
Data are presented as mean ± SD. The statistical analysis was conducted with Student's t-test for pairwise comparison and a one-way ANOVA followed by Turkey's post hoc test for multiple comparisons. All analyses were performed using GraphPad Prism 7.0 software (Graphpad Software). Statistical significance was determined as p < .05.

| Decreased SIRT3 protein abundance with age was related to hyperacetylation and a reduced NAD + / NADH ratio in alveolar bone
Sirtuin 3 is a major mitochondrial deacetylase that targets many enzymes involved in central metabolism. 1 As age increased, we observed a decline of protein abundance of SIRT3 in alveolar bone ( Figure 1A1-A2), which was accompanied with an increased protein acetylation modification ( Figure 1B), suggesting that SIRT3 may play a role in the pathogenesis of age-related periodontal disease.
SIRT3-KO further enhanced the protein acetylation status of the alveolar bone after aged ( Figure 1B). Since nicotinamide adenine dinucleotide (NAD + ) is the cofactor of SIRT3 and serves as a key player in cellular metabolism, we sought to identify whether the reduced SIRT3 expression in maxillary bone was connected to dysregulated NAD + biology. Indeed, the NAD + level was reduced in the aged group, and SIRT3-KO exacerbated the depletion of NAD + in the alveolar bone at both ages ( Figure 1C1). As compared to the young WT mice, the NADH level was slightly increased in the young SIRT3-KO and attained a statistical difference in the aged SIRT3-KO, resulting in a downregulation of the NAD + /NADH ratio in both young and aged SIRT3-KO groups ( Figure 1C2-C3). Taken together, these data showed that SIRT3 deficiency resulting from age was related to hyperacetylation and a reduced NAD + /NADH ratio in alveolar bone.

| Sirtuin 3 deficiency exacerbated age-related periodontal bone resorption
To understand the effect of reduced SIRT3 expression on alveolar bone loss, we measured alveolar bone architecture and morphology.
The linear measurement of the three-dimensional images indicated that the CEJ-ABC distance was significantly increased in the aged WT mice in comparison with the young WT group ( Figure 2A1-A4).
The SIRT3-KO group exhibited high CEJ-ABC distance at young age, which was further increased as they aged, suggesting that SIRT3 deficiency exacerbates age-related periodontal bone loss. Similar findings were observed in the H&E staining images ( Figure 2A3). Moreover, the alveolar bone architecture was determined by the micro-CT evaluation of tomographic cross-sectional slices taken within the furcation zone ( Figure 2B1). The micro-CT images of the mandibles showed morphological abnormalities in the aged WT and SIRT3-KO mice, as compared to the young WT, including increases in bone marrow cavities and decreased trabecular connectivity ( Figure 2B2).
The calculated BV/TV of the trabecular bone was significantly lower under aged or SIRT3-KO or both conditions, indicating that SIRT3 deficiency decreases trabecular connectivity ( Figure 2B3).
Through TRAP staining, we found only a few osteoclasts in the Young WT group but many more in the aged and SIRT3-KO groups.
The highest osteoclast abundance was observed in the aged SIRT3-KO mice, indicating that aging and SIRT3-KO may synergistically contribute to osteoclastogenesis ( Figure 2C1-C2). In addition, we evaluated the expression of gene markers involved in osteogenic differentiation ( Figure 2D). The mRNA levels of Runx2 and ALP were decreased in aged WT, and further reduced when combined with SIRT3-KO. As regulators of osteoclastogenesis and bone inflammation, the gene expression of OPG was suppressed while the RANKL was elevated under aged or SIRT3-KO or both conditions, resulting in decreased OPG/RANKL ratio.
To confirm the impact of SIRT3 on osteoclastogenesis, we knocked down SIRT3 in vitro with SIRT3 −/− shRNA. Consistently, we found the mRNA level of OPG was significantly decreased in the shSIRT3 group ( Figure 2E). In order to investigate whether the inflammatory response occurred within the alveolar bone tissues, we measured the gene expression level of three proinflammatory cytokines-interleukin-1 (IL-1), IL-6, and tumor necrosis factorα (TNFα)-which have a central role in periodontal tissue destruction. 27 Also, it is well known that IL-1β stimulates bone loss and has an inhibitory effect to bone formation. 28 We observed an approximately 20% increase in the IL-1β expression in the young SIRT3-KO group compared with the young WT, which was further increased in aged mice ( Figure 2F). Both IL-6 and TNFα mRNA levels showed a significant increase in the aged SIRT3-KO group, compared to the young SIRT3-KO group. Previous studies showed that those proinflammatory cytokines elicit tissue destruction and bone resorption by stimulating collagenase and the RANKL, which induces osteoclast differentiation. 29,30 Together, these observations suggested that SIRT3 deficiency exacerbated age-related periodontal bone resorption.

| Sirtuin 3 deficiency induced oxidative damage in gingival tissues
To determine whether oxidative stress contributed to the pathogenesis in periodontal tissues due to SIRT3 deficiency,

| Sirtuin 3 deficiency influenced mitochondrial dynamics in periodontal tissues
To test whether the increased mitochondrial ROS has measurable effects, we quantified mitochondrial morphology with SIRT3 knockdown MC3T3-E1 osteoblastic cells. Mitochondrial fusion/fission was evaluated using the form factor (FF = 4π × area/perimeter 2 ) and the aspect ratio (AR = major radius/minor radius). We found both the AR and FF significantly decreased in the SIRT3 knockdown cells, indicating that SIRT3 deficiency promotes mitochondrial fission ( Figure 4A1-A2). Evidence suggests that SIRT3 regulates mitochondrial fission via deacetylation of CypD. 21 Also, the CypD-dependent mPTP opening is responsible for ROS-induced cellular damage. 33,34 In this study, we used IHC staining and observed increased CypD protein expression in the gingival tissues of either young SIRT3-KO or aged mice ( Figure 4B1-B2). This finding was further confirmed by western blot and indicated that the protein abundance of CypD

| Sirtuin 3 deficiency suppressed mitochondrial biogenesis in periodontal tissues
To test whether SIRT3 deficiency influences mitochondria function, we chose to test PGC-1α, which is considered to be a major regulator of mitochondrial biogenesis, and reduction in its levels has been related to mitochondrial dysfunction. 35 The IHC staining data showed that the PGC-1α protein expression was significantly decreased in the gingival tissues of the aged WT and both SIRT3-KO groups as compared to the young WT ( Figure 5A1-A2). Consistently, the mRNA level of PGC-1α also decreased in aged and/or SIRT3-KO mice ( Figure 5B). To further investigate the functional role of SIRT3 in mitochondrial metabolism, we measured OCR through Seahorse analysis ( Figure 5C1). SIRT3 knockdown MC3T3-E1 osteoblastic cells exhibited a significant reduction in maximal OCR in comparison with the shCtrl group ( Figure 5C2), indicating that SIRT3 deficiency impacts mitochondrial respiration and the oxidative phosphorylation (OXPHOS) system. The Ndufs4 gene encodes a nuclear-encoded accessory subunit of the mitochondrial membrane respiratory chain NADH dehydrogenase (complex I, or NADH:ubiquinone oxidoreductase), which removes electrons from NADH and passes them to the electron acceptor ubiquinone. 36,37 We found that SIRT3 knockdown significantly down-regulated the Ndufs4 mRNA level ( Figure 5D), implying that SIRT3 may decrease the NAD + /NADH ratio by regulating complex I function. Taken together, these data suggested that SIRT3 deficiency reduced the PGC-1α level and mitochondrial respiration, which may suppress mitochondrial biogenesis during aging in periodontal tissues.

| DISCUSS ION
The increased prevalence and severity of poor periodontal health have long been associated with advancing age. The present study assessed the functional role of SIRT3 in age-related periodontal disease and the underlying mechanisms. We found a reduced SIRT3 expression and down-regulated NAD + /NADH ratio in the periodontal mitochondria. 38 The regulation of ROS by SIRT3 and ROS-induced damage could be important mechanisms by which SIRT3 influences aging and the age-related periodontal disease. MnSOD is the enzyme that converts harmful superoxide radicals to hydrogen peroxide in mitochondria. 39,40 SIRT3 directly deacetylates MnSOD in mitochondria, significantly enhancing its ability to scavenge ROS. 41,42 PGC1α is identified as an inducible upstream regulator of mitochondrial biogenesis that determines mitochondrial mass and function through its downstream targets NRF1 and NRF2. 11 Moreover, PGC1α plays a vital role in controlling SIRT3 gene expression. SIRT3 may also upregulate PGC1α via a positive feedback mechanism. 43 The structure of the mPTP includes the voltage-dependent ion channel, adenine nucleotide translocator (ANT), and CypD. Binding of CypD to ANT initiates a tunnel-like structure that connects the mitochondrial matrix with the cytosol. 44 A previous study showed that SIRT3 suppressed excessive an mPTP opening and maintained mitochondrial membrane potential by activation of CypD via deacetylation. 45 In this context, we found that the CypD expression was up-regulated in aged mice or SIRT3-KO mice, which emphasized the significance of SIRT3 in these processes.
The mechanisms for decreased SIRT3 expression during aging remain unclear. One possible reason could be the depletion of NAD + . 46 Indeed, emerging evidence shows that the NAD + level declines during aging, which decreased SIRT3 as a feedback, thereby leading to mitochondrial dysfunction and metabolic abnormalities. 47,48 In this study, we found SIRT3 knockdown may decrease complex I function, which could impact the NAD + /NADH ratio. A previous study provided direct evidence that oxidative stress-induced physiological stem cell aging and tissue degeneration are rescued by upregulation of SIRT3. 49 Furthermore, supplementation with NAD + precursors like nicotinamide riboside, nicotinamide mononucleotide, and nicotinamide could ameliorate some aging-related diseases. 50,51 Additionally, the mitochondrial NAD + level and SIRT3 expression could also be increased by exercise, fasting, and caloric restriction. [52][53][54] There are some limitations in our study. The aged mice in this study were 16 months old, which is beyond the range of middleaged mice (14-15 months) according to the definition from the Jackson Laboratory. However, they are still younger than mice defined as "old," which ranges from 18 to 24 months of age. We did not perform rescue experiments to restore SIRT3 activity or expression to see whether that would ameliorate periodontal pathology. Also, most of our findings are based on observational data, more mechanistic studies are needed to uncover the underlying signaling pathways.

| CON CLUS IONS
(1) Sirtuin 3 expression was decreased in the alveolar bone of the aged mice, accompanied by a reduced NAD + /NADH ratio and an elevated proteins acetylation level.

CO N FLI C T O F I NTE R E S T
None.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.