Objectives Important roles for SIRT1 are implicated in ageing and age-related diseases. The role of SIRT1 in osteoarthritis (OA), however, remains partially unknown. To investigate the role of SIRT1 in chondrocytes in vivo, cartilage-specific Sirt1-conditional knockout (CKO) mice were analysed using an experimental OA model.
Methods OA was surgically induced in 8-week-old C57BL6/J (wild-type) mice and Sirt1-CKO (Sirt1flox/flox; Col2a1-Cre) mice generated using the Cre-loxP system. We examined changes in Sirt1 protein during the development of surgically-induced OA and during ageing in wild-type mice. OA progression in Sirt1-CKO mice was evaluated histologically at 2, 4 and 8 weeks after surgery, and at 1 year of age without surgery compared with control (Sirt1flox/flox) mice.
Results The number of Sirt1-positive chondrocytes decreased during ageing, and although it was increased at 2 weeks after surgery, then gradually decreased to the presurgical level during the progression of OA in wild-type mice. Sirt1-CKO mice showed no obvious skeletal abnormalities. The histological OA score was significantly higher in 1-year-old Sirt1-CKO mice than in control mice. Sirt1-CKO mice showed accelerated OA progression at 2 and 4 (but not 8) weeks compared with control mice. Immunohistochemical analysis revealed increases in type X collagen, matrix metalloproteinase 13, a disintegrin and metalloproteinase with thrombospondin motifs-5, apoptotic markers, and acetylated nuclear factor-κB p65 in Sirt1-CKO mice compared with control mice 2 weeks after surgery.
Conclusions Loss of Sirt1 in chondrocytes led to the accelerated development of OA in mice. Our observations suggest that SIRT1 has a preventive role against the development of OA.
- Knee Osteoarthritis
Statistics from Altmetric.com
Osteoarthritis (OA) is a degenerative disease of human articular cartilage often seen in aged individuals. Multiple factors contribute to OA development, including hereditary predisposition, mechanical stress and ageing.1 ,2 During OA development, a number of pathological changes are observed, including an imbalance between anabolic and catabolic activity of chondrocytes,1 ,2 increased production of cartilage-degrading enzymes such as matrix metalloproteinase (MMP) and a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS)3 ,4 and increased apoptosis.5 ,6
Sirt1 is a mammalian homologue of Sir2, which is a histone deacetylase that regulates gene expression and protein function by deacetylating lysine residues in histone7–9 and non-histone proteins, such as p53,10 forkhead box protein O,11 the RelA/p65 subunit of nuclear factor-κB (NF-κB),12 peroxisome proliferator-activated receptor-γ13 and peroxisome proliferator-activated receptor-γ coactivator 1α.14 SIRT1 targets are involved in essential biological processes such as stress responses, DNA repair and inflammation, which are important factors in ageing and age-related disease.15 ,16 Thus, SIRT1 is suggested to play key roles in the regulation of ageing and age-related diseases.17–19
SIRT1 promotes cartilage-specific gene expression20 and protects chondrocytes against radiation-induced senescence.21 SIRT1 inhibits apoptosis of in vitro human chondrocytes22 ,23 and SIRT1 inhibition in human chondrocytes leads to osteoarthritic gene expression changes,24 suggesting that SIRT1 has important roles in chondrocytes. SIRT1 has a protective role by suppressing interleukin-1β (IL-1β)- and tumour necrosis factor α-induced expression of cartilage-degrading enzymes by modulating the NF-κB pathway.25 ,26 Further, adult heterozygous Sirt1-knockout mice exhibit reduced safranin-O staining intensity and increased OA progression compared with wild-type mice,27 and mice carrying a point mutation that ablates Sirt1 enzymatic activity exhibit increased OA progression,28 strongly suggesting a critical role for SIRT1 in cartilage homeostasis. Heterozygous Sirt1-knockout mice and mice with a Sirt1 point mutation have smaller body sizes than wild-type mice,27 ,28 and Srt1-heterozygous knockout mice exhibit metabolic impairments and reduced bone mass.29 ,30 Therefore, indirect systemic effects of progressive OA on the phenotype of heterozygous Sirt1-knockout mice are possible and a direct role of Sirt1 in chondrocytes needs to be tested in vivo. The in vivo role of SIRT1 in the development of OA and the cartilage-specific role of SIRT1, however, remain unclear. We investigated the role of SIRT1 in chondrocytes and in the development of OA using cartilage-specific Sirt1-conditional knockout (CKO) mice.
Materials and methods
Details of additional data, methods, primers and reagents are available in the online supplementary information material.
C57BL6/J mice (wild-type; Charles River Japan, Yokohama, Japan) were used to examine Sirt1 expression and localisation in the development of OA. To generate cartilage-specific Sirt1-knockout mice, mice carrying a floxed Sirt1 allele (Sirt1flox/flox; Jackson Laboratory), in which exon 4 of Sirt1 was flanked by loxP sites,31 were crossed with Col2a1-Cre/+ transgenic mice (kindly provided by Dr S Murakami, Case Western Reserve University, Cleveland, Ohio, USA) expressing Cre recombinase under the control of the collagen type II promoter.32 All procedures were approved by the Institutional Animal Care and Use Committee of the University of Kobe. Mice were maintained under pathogen-free conditions and were freely allowed access to food, water and activity.
Generation of cartilage-specific Sirt1-CKO mice
To generate mice in which Sirt1 was inactivated in cartilage, heterozygous Sirt1-CKO (Sirt1flox/+; Col2a1-Cre) mice were crossed with Sirt1flox/flox mice to generate homozygous Sirt1flox/flox; Col2a1-Cre mice (Sirt1-CKO). Sirt1-CKO mice were obtained at the expected Mendelian ratio, survived normally and mated successfully. Sirt1flox/flox mice without the Cre transgene were used as control mice. DNA was extracted from mouse tails using DNeasy (Qiagen, Valencia, California, USA) according to the manufacturer's instructions. PCR was performed for genotyping.31 Primers for Cre and signal transducer and activator of transcription 1 (STAT1) were used. PCR for STAT1 was used to confirm the PCR reaction efficiency.
OA model and ageing model
Experimental OA was induced in the knee joint of 8-week-old wild-type, Sirt1-CKO and control mice by transecting the medial collateral ligament and resecting the medial meniscus under a microscope.33 To examine Sirt1 expression changes during ageing, we analysed 2-, 3-, 4-month-old and 1-year-old wild-type mice. To examine the effect of Sirt1 disruption in chondrocytes during ageing, we compared the Sirt1-CKO mice and control mice at 1 year of age.
RNA isolation and real-time PCR
Articular cartilage was collected from the medial femoral condyle and medial tibial plateau of six wild-type mice at 2, 4 and 8 weeks after surgery, and at time 0 (before surgically inducing OA), and the obtained cartilage was pooled, as described previously.34 For characterisation of Sirt1-CKO mice, the articular cartilage was collected from each knee joint of the postnatal day 3 Sirt1-CKO mice and control mice. Total RNA was extracted using TRIzol (Invitrogen, Burlington, Ontario), followed by RNeasy (Qiagen), according to the manufacturer’s instructions. Real-time PCR was performed on the ABI Prism V.7700 sequence detection system (Applied Biosystems) using the TaqMan probe. To confirm the efficiency of Sirt1 knockdown in the Sirt1-CKO mice, SYBR Green real-time PCR was performed.
Postnatal day 0 Sirt1-CKO mice and control littermates were skinned, eviscerated and fixed in 95% ethanol. Alcian Blue staining was performed, followed by placement into potassium hydroxide to remove the soft tissue. After visualising the cartilage staining, Alizarin Red staining was performed to evaluate mineralisation.
The Sirt1-CKO mice and control mice (both female and male mice) were killed at 2, 4 and 8 weeks after surgery (n=6 for each time point/group), and 1-year-old Sirt1-CKO mice and control mice (n=4/group) were killed, and the entire knee joints were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer solution overnight at 4°C, decalcified for 2 weeks with 10% EDTA and embedded in paraffin wax. Each specimen was cut into 5-μm slices along the sagittal plane and stained with safranin O–fast green. Three slices were selected from each medial femoral condyle and medial tibial plateau and pictures were obtained at a magnification of 40×. The histological OA grade for each field was evaluated using the Osteoarthritis Research Society International (OARSI) cartilage OA histopathology grading system (score 0–6).35 OA grading was assessed by a single observer who was blinded to the study.
After epitope retrieval, sections were incubated with primary antibody, followed by secondary antibody. The signal was developed as a brown reaction product using the peroxidase substrate 3,3′-diaminobenzidine with methyl green counterstaining or haematoxylin counterstaining. To detect Sirt1 with immunofluorescence, Alexa Fluor 488–conjugated goat secondary antibody against rabbit IgG (Cell Signaling Technology, Danvers, Massachusetts, USA) was used. Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining was performed with the In Situ Cell Death Detection kit (Fluorescein; Roche Applied Science, Indianapolis, Indiana, USA). All images were obtained under a microscope (Biozero, KEYENCE, Itasca, Ohio, USA). The following antibodies were used: Sirt1 (Millipore, Billerica, Massachusetts, USA), type X collagen (LSL, Tokyo, Japan), type II collagen (embryos; LSL; 1-year-old mice; Southern Biotechnology Associates Inc., Birmingham, Alabama, USA), MMP-13 (Abcam, Cambridge, MA), ADAMTS-5 (Santa Cruz Biotechnology, Santa Cruz, California, USA), cleaved caspase3 (Cell Signaling Technology), poly(ADP-ribose) polymerase (PARP) p85 (Promega, Madison, Wisconsin, USA) and acetylated NF-κB p65 (Sigma-Aldrich, St. Louis, Missouri, USA).
Cell culture and western blotting analysis
Chondrocytes were isolated from the femoral chondyle and tibial plateau of postnatal day 7 Sirt1-CKO and control mice. Chondrocytes were seeded on 6-well plates (2×105/well). After reaching 60%–70% confluency, cells were cultured with 10 ng/ml IL-1β (R&D Systems, Minneapolis, Minnesota, USA) for 24 h, followed by protein extraction and western blotting using the following antibodies: anti-α-tubulin (Sigma-Aldrich), NF-κB p65, acetyl-NF-κB p65 (Lys310; Cell Signaling Technology) antibodies and HRP-conjugated antirabbit and mouse IgG (GE Healthcare, Tokyo, Japan).
All values are expressed as mean±SD. An unpaired two-tailed Student t test was used to compare differences between two groups. One-way analysis of variance was used to compare multiple groups with the Bonferroni method as a post hoc test. p Values <0.05 were considered statistically significant.
Sirt1 protein first increased, then gradually decreased to the presurgical level in wild-type mice with experimental OA
We examined Sirt1 protein during the surgically-induced development of OA in knee joints of wild-type mice. Immunohistochemical analyses revealed Sirt1-positive chondrocytes distributed from the superficial to the deep zone of the cartilage. Time-course analysis revealed that Sirt1 staining intensity significantly increased at 2 weeks after surgery compared with time 0 (before inducing OA), and then gradually decreased with OA progression. The number of Sirt1-positive chondrocytes significantly increased at 2 weeks compared with time 0, and then significantly decreased at 8 weeks after surgery compared with that at 2 and 4 weeks (n=3; figure 1A,B). We extracted RNA from the articular cartilage of the knee joint and examined the Sirt1 mRNA expression changes at each time point. Real-time PCR analysis showed that the Sirt1 mRNA level increased approximately twofold at 2 weeks compared with time 0. Sirt1 mRNA expression was slightly decreased at 4 weeks and returned to the presurgical level at 8 weeks (figure 1C).
Sirt1 protein changes during ageing in wild-type mice
We examined changes in Sirt1 protein during ageing in the knee joints of wild-type mice. Sirt1 protein was mainly observed in the superficial and middle zone and Sirt1 staining in 2-, 3- and 4-month-old mice was stronger than that in 1-year-old mice (figure 1D). Sirt1-positive cells were significantly decreased in 1-year-old mice compared with 2-, 3- and 4-month-old mice (n=3; figure 1E).
Characterisation and efficiency of Sirt1 knockdown in cartilage-specific Sirt1-CKO mice
Because mice homozygous for the Sirt1-null allele do not survive for more than 1 month,31 ,36 we generated cartilage-specific Sirt1-CKO mice using the Cre-loxP system. The DNA recombination between two lox sites in the Sirt1 allele by Cre recombinase was confirmed by the presence of a recombined allele (figure 2A). The efficiency of the loss of Sirt1 expression in chondrocytes was confirmed by immunohistochemistry and real-time PCR. Immunohistochemistry for Sirt1 showed remarkably reduced Sirt1 protein expression in the femoral epiphyses of Sirt1-CKO embryos (embryonic day (E)15.5; figure 2B) and the articular cartilage of tibial plateaus of Sirt1-CKO mice (10-week-old; figure 2C), and the kidney was Sirt1-positive in both Sirt1-CKO and control mice (8-week-old; figure 2D). Real-time PCR analysis showed an approximately 90% reduction in Sirt1 mRNA in the articular cartilage of the knee joint of postnatal day 3 Sirt1-CKO mice compared with control mice (n=5; figure 2E).
Disruption of Sirt1 in chondrocytes did not cause obvious skeletal abnormalities during embryogenesis
We examined the effect of Sirt1 disruption in chondrocytes on skeletal growth, and observed no obvious skeletal abnormalities, joint formation or subchondral bone in Sirt1-CKO mice compared with littermate control mice (figure 3A,B and see online supplementary figure S1). Embryonic (E15.5) growth plate structure and immunohistochemistry for type II and type X collagen did not differ significantly between littermate control and Sirt1-CKO embryos (figure 3C,D).
Disruption of Sirt1 in chondrocytes caused accelerated cartilage degeneration in 1-year-old mice and accelerated the development of experimental OA
Although skeletal growth was not significantly different between Sirt1-CKO and control mice, at 1 year of age, safranin O staining of the articular cartilage of the medial femoral condyle and the tibial plateau was weaker in Sirt1-CKO mice than in control mice (figure 4A). In addition, the summed tibial and femoral OARSI score was significantly higher in Sirt1-CKO mice than in control mice, although the OARSI scores were low (figure 4B). These observations suggest the accelerated progression of cartilage degeneration in Sirt1-CKO mice during ageing. To further examine the effects of Sirt1 loss in chondrocytes during OA development, we created an experimental OA model using Sirt1-CKO mice. Histological OA scores in the medial femoral condyle and tibial plateau were significantly higher in Sirt1-CKO mice than in littermate control mice at 2 and 4 weeks (figure 4C,D), but not at 8 weeks. There were no sex-specific differences in either group.
Increased protein of chondrocyte hypertrophic markers, cartilage-degrading enzymes, apoptotic markers and acetylated NF-κB p65, and decreased chondrocyte anabolic marker protein in Sirt1-CKO mice
To further examine the mechanism underlying accelerated OA progression in Sirt1-CKO mice, we examined the expression of key factors in OA development, such as chondrocyte hypertrophic markers and cartilage-degrading enzymes, including type X collagen, MMP-13 and ADAMTS-5. All markers were significantly increased in Sirt1-CKO mice compared with control mice 2 weeks after surgery (figure 5A). We also examined the apoptotic marker proteins including cleaved caspase 3 and PARP p85 fragment, and the TUNEL-positive cells, which were both increased in Sirt1-CKO mice compared with control mice 2 weeks after surgery (figure 5A,B). The NF-κB pathway mediates inflammatory processes, such as MMPs;37 therefore, we further examined NF-κB acetylation using an antibody that reacts with acetylated NF-κB P65 at lysine 310. Two weeks after surgery, Sirt1-CKO mice showed stronger staining and significantly increased acetylated NF-κB P65-positive cells compared with control mice (figure 5A). In addition, an IL-1β induced significantly more acetylated NF-κB P65 protein in primary culture of Sirt1-CKO chondrocytes compared with control chondrocytes (figure 5C). On the other hand, type II collagen staining was weaker in Sirt1-CKO than in control mice at 1 year of age (figure 5D).
The present study examined Sirt1 expression and localisation in the cartilage and the role of Sirt1 in the development of OA in vivo by analysing cartilage-specific Sirt1-CKO mice. Sirt1 protein first increased in the early phase of OA development and then gradually decreased in the advanced OA stage in our mouse OA models. These observations are consistent with previous findings that SIRT1 is clearly expressed in the less damaged human articular cartilage and in normal cartilage, while it is decreased in severely degenerated cartilage.20 ,24 Mechanical stress induces Sirt1 mRNA and Sirt1 protein expression in skeletal muscle, with the protein transiently increasing and then gradually decreasing.38 These observations suggest that Sirt1 expression is induced by stress caused by surgically-induced OA and that the expression is decreased by a currently unknown mechanism. A number of studies demonstrated that Sir2/SIRT1 plays key roles in cell survival under cellular stress conditions, such as nutrient starvation, mechanical stress, oxidative stress and DNA damage.11 ,38–40 These previous studies and our observations suggest that SIRT1 is involved in cellular stress, especially in the early cartilage degeneration process. Supporting this hypothesis, we observed accelerated OA progression in Sirt1-CKO mice soon after surgically inducing OA. Taken together, our observations indicate that Sirt1 plays a protective role in chondrocytes under mechanical stress and prevents the development of OA.
SIRT1 is reported to regulate both anabolic pathways and catabolic pathways including apoptotic pathways in chondrocytes.20 ,22 ,23 ,41 ,42 The mechanism of the accelerated OA progression in Sirt1-CKO might be caused by alterations of these pathways.
SIRT1 inhibits the apoptosis of human chondrocytes through the modulation of various pathways.22 ,23 ,41 ,42 A recent study reported that increased apoptotic chondrocytes and advanced OA progression are observed in heterozygous Sirt1 knockout mice,27 consistent with our observations. Together, these observations suggest that Sirt1 plays a preventive role in the apoptotic process under mechanical stress conditions and that accelerated OA progression in Sirt1-CKO mice might be caused in part by increased apoptosis of chondrocytes.
We also observed increased cartilage-degrading enzymes, including MMP-13 and ADAMTS-5. One possible mechanism for these observations is the modulation of NF-κB signalling through the deacetylation of NF-κB p65. Notably, acetylation of the NF-κB p65 subunit was increased in Sirt1-CKO mice compared with control mice after surgery, and also increased in IL-1ß-stimulated primary chondrocytes of Sirt1-CKO mice. The NF-κB pathway mediates IL-1ß-induced expression of MMPs, including MMP-13.37 ,43 In addition, high-magnitude mechanical stress induced by cyclic tensile strain activates and translocates NF-κB p65 in chondrocytes.44 SIRT1 physically interacts with the NF-κB P65 subunit and deacetylates the lysine 310 residue of P65, thereby inactivating NF-κB and inhibiting expression of its target genes.12 ,26 We recently reported that SIRT1 overexpression in human chondrocytes reduces NF-κB p65 target gene expression by deacetylating its subunit.26 Moreover, resveratrol, a potent stimulator of SIRT1, inhibits IL-1β-induced inflammatory responses via SIRT1-dependent suppression of NF-κB in rat chondrocytes.45 Thus, SIRT1 may have a preventive role in the development of OA via modulation of the NF-κB pathway. On the other hand, our results revealed that ADAMTS-5 was increased in Sirt1-CKO mice compared with control mice after surgery. ADAMTS-5 is activated via syndecan-4 in an NF-κB dependent manner in nucleus pulpous cells.46 Increased ADAMTS-5 in Sirt1-CKO mice might be related to the NF-κB pathway, though the detailed mechanism requires further study.
SIRT1 is considered a longevity factor with roles in ageing and degenerative diseases.17–19 ,47 Therefore, we also investigated the involvement of Sirt1 in OA progression during ageing. Sirt1-positive chondrocytes were significantly decreased in 1-year-old mice compared with 2-, 3- and 4-month-old wild-type mice in association with decreased safranin-O staining. Similarly decreased SIRT1 expression during ageing is reported in other cells and tissues such as embryonic fibroblasts, liver, neurones and hippocampus.47–50 These observations suggest that decreased SIRT1 expression contributes to the degeneration of various tissues, including cartilage. Supporting this notion, we observed accelerated cartilage degeneration in relation to SIRT1 loss in chondrocytes. The accelerated cartilage degeneration was associated with reduced safranin-O and type II collagen staining. These observations are consistent with a previous in vitro study showing that SIRT1 stimulates COL2A1 expression through an interaction with SOX9.20 Sirt1 expression is significantly decreased in the articular cartilage of aged mice, and adult heterozygous Sirt1 knockout mice exhibit increased progression of OA compared with age-matched wild-type mice.27 These observations are consistent with the present study and the notion that Sirt1 plays a preventive role in the development of OA with ageing.
There are some limitations to the present study. Young mice (8-week-old mice) were used for the surgical induction of OA. Therefore, this model may not represent OA pathology in humans. At 8 weeks after surgery, OA scores were not significantly different between Sirt1-CKO and control mice, which might be dependent on age and/or the surgical method. Because OA was surgically induced by transecting the medial collateral ligament and resecting the medial meniscus, OA progressed quickly and a large part of the articular cartilage disappeared at 8 weeks, even in the control mice.
In conclusion, we demonstrated that cartilage-specific Sirt1-deficient mice developed accelerated OA progression under mechanical stress and ageing. Our observations suggest that SIRT1 plays a preventive role against the development of OA. Further studies regarding the role of SIRT1 in cartilage will provide novel insight into the pathomechanisms of OA that may lead to new therapeutic approaches.
We would like to thank Ms Kyoko Tanaka, Ms Minako Nagata and Ms Maya Yasuda for technical assistance.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online figure
Handling editor Tore K Kvien
Contributors The seven authors are justifiably credited with authorship, according to the authorship criteria. In detail: T Matsuzaki—conception, design, acquisition of data, analysis and interpretation of data, drafting of the manuscript, final approval given; T Matsushita—conception, design, acquisition of data, analysis and interpretation of data, drafting of the manuscript, final approval given; KT—conception, critical revision of manuscript; T Matsumoto—critical revision of manuscript, final approval given; KN—critical revision of manuscript, final approval given; RK—conception, design, critical revision of manuscript, final approval given; MK—critical revision of manuscript, final approval given.
Funding This study was supported by the Japan Society of the Promotion of Science, Grants-in-Aid for Scientific Research (Grant No 21791393), and the Japan Orthopaedics and Traumatology Foundation, Research Grant (Grant No 216) for T Matsushita.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.