Article Text
PDF
Abstract
Objective A growing body of evidence indicates that the protein deacetylase, SirT1, affects chondrocyte biology and survival. This report aims to evaluate in vivo attributes of SirT1 in cartilage biology of 129/J murine strains.
Methods Heterozygous haploinsufficient (SirT1+/−) and wild-type (WT; SirT1+/+) 129/J mice aged 1 or 9 months were systematically compared for musculoskeletal features, scored for osteoarthritis (OA) severity, and monitored for chondrocyte apoptosis in articular cartilage. Sections of femorotibial joints were stained for type II collagen and aggrecan. Protein extracts from articular chondrocytes were isolated and immunoblotted for SirT1 and active caspase 3.
Results Phenotypic observations show that, at 1 month of age, SirT1+/− mice were smaller than WT and showed a significant decrease in full-length SirT1 (FLSirT1; 110 kDa) protein levels. Levels of FLSirT1 were further decreased in both strains at 9 months. Immunoblot assays for 9-month-old strains revealed the presence of the inactive cleaved SirT1 variant (75 SirT1; 75 kDa) in WT mice, which was undetected in age-matched SirT1+/− mice. Nine-month-old SirT1+/− mice also showed increased OA and increased levels of apoptosis compared with age-matched WT mice.
Conclusion The data suggest that the presence of 75 SirT1 may prolong viability of articular chondrocytes in adult (9-month-old) mice.
Statistics from Altmetric.com
Introduction
SirT1 has recently been shown to affect cartilage biology and chondrocyte survival.1,–,4 Several reports support the notion that full-length SirT1 (FLSirT1) enhances chondrocyte survival through protein tyrosine phosphatase 1B inhibition or mitochondrial-related pathways.2 3 Enhanced chondrocyte apoptosis has been attributed to cartilage degeneration in osteoarthritis (OA) and rheumatoid arthritis.5,–,7 It appears that the presence of proinflammatory cytokines such as tumour necrosis factor α (TNFα) and interleukin 1 (IL-1β) may result in enhanced chondrocyte death and increased levels of secreted catabolic enzymes.5 7 Our recent data show that chondrocytes exposed to TNFα undergo cathepsin B-mediated cleavage of the nuclear FLSirT1 (110 kDa) to generate a stable but enzymatically inactive 75 kDa fragment of SirT1 (75 SirT1), which is partly exported to the cytoplasm.4 8 In the cytoplasm, 75 SirT1 colocalises with the mitochondrial membrane and associates with cytochrome C, preventing chondrocyte apoptosis.4
It is widely held that ageing is a known risk factor for OA development.9 Given our previous data, age-related systemic inflammation10 may induce the formation of 75 SirT1 from its wild-type (WT) FLSirT1 form. We therefore monitored age-related effects on FLSirT1 and their attributes in cartilage homoeostasis and chondrocyte viability.
It has been found that total body knockout mice for SirT1 (null; SirT1−/−) do not survive longer than 1 month postnatally,11 and are thus not appropriate for testing OA development in aging mice. McBurney and colleagues reported that SirT1 null embryos of the 129/J strain possess abnormal phenotypic features and are not likely to survive postnatally.11 During embryogenesis and postnatal stages, SirT1 null mice are significantly smaller than WT littermates, implying that SirT1 may be involved in early and late developmental stages. Here we test our hypothesis by a systematic comparison of adult (9-month-old) and young (1-month-old) heterozygous (SirT1+/−) and WT 129/J mice.
Methods
Animal protocols
Experimental procedures involving 129/J mice were carried out in accordance with NIH committees for animal use and care (ARAC guidelines). 129/J heterozygous (SirT1+/−, six for each experiment) and WT (SirT1+/+, six for each experiment) littermates were generated from SirT1+/− matings. Genomic DNA from mice tail fragment was periodically sampled for genotyping by PCR using the Extract-N-Amp Tissue PCR kit (Sigma-Aldrich, St Louis, Missouri, USA). PCR primer sequences for SirT1 amplification were (forward) 5′-TTCACATTGCATGTGTGTGTG-3′ and (reverse) 5′- TAGCCTGCGTAGTGTTGGTG -3′.
After the mice had been sacrificed, articular cartilage was isolated and either fixed for histology or further processed for chondrocyte isolation as described by Gosset et al (2008).12 Freshly isolated chondrocytes were processed to obtain protein extracts. Protein extracts (n=3) were supplemented with 10 µg/ml of the cathepsin B inhibitor, ALLN and the complete protease inhibitor cocktail (Roche Applied Science, Indianapolis, Indiana, USA). They were then run on a 4–12% polyacrylamide gradient gel (Bio-Rad Laboratories, Rishon Le Zion, Israel) and blotted on a poly(vinylidene difluoride) membrane. After blocking, membranes were incubated with primary and secondary antibodies as previously described.4 8
For in vitro assays, WT murine SirT1 (mSirT1) was stably expressed in human OA-derived chondrocytes (n=3), as described by Dvir-Ginzberg et al.1 Chondrocyte lines expressing mSirT1 were treated with 50 ng/ml TNFα (PeproTech, Rehovot, Israel) and 10 µg/ml ALLN for 24 h. After induction, protein extracts were obtained, and immunoblot assays were carried out, as previously indicated.
Immunohistochemistry
Knee joints were fixed in 4% paraformaldehyde, decalcified, dehydrated using a graded series of ethanol washes and embedded in paraffin. Frontal and lateral sections of 5 µm were predigested with hyaluronidase (Sigma-Aldrich, St. Louis, MO, USA) in 10 mM Tris/HCl, pH 7.5, for 30 min at 37°C, then incubated overnight at room temperature with primary antibodies to cleaved caspase 3 (Cell Signaling Technology, Boston, MA, USA), aggrecan or type II collagen (Santa Cruz, Santa Cruz, CA, USA) in Tris-buffered saline (pH 7.4) containing 0.1% bovine serum albumin. The presence of cleaved caspase 3 denotes end-stage apoptotic events. Early apoptotic cell death events were monitored by the M30 CytoDEATH in situ fluorescent kit (Roche Applied Science), according to the manufacturer's guidelines. Using this kit, cell nuclei were stained with 4′,6-diamidino-2-phenylindole (Sigma-Aldrich) at a final concentration of 1 µg/ml in phosphate-buffered saline. Briefly, the antibody used in the kit recognises an epitope of cytokeratin 18, which is present after cleavage by caspases during early apoptotic events.
Additional sections were stained with Safranin-O (Sigma-Aldrich) according to the manufacturer's instructions. After staining, slides were visualised under a Leica DMR Microscope (Leica Microsystems, Richmond, IL, USA).Representative sections were captured and analysed for intensity using Image J software, or scored with osteoarthritis research society international (OARSI) semiquantitative scoring for mice.13 After Image J software analyses, chondrocytes were considered positive for cleaved caspase 3 if they displayed an intensity 10-fold greater than the background.
Statistical analysis and OA scoring
The Student t test was carried out to determine the differences between two equivalent treatments within a group. Error bars indicate the SD around the mean value of the data point. The severity of OA was scored on the medial tibial plateau and medial femoral plateau of the knee joint, according to OARSI semiquantitative scoring for mice.13 Safranin-O-stained cartilage sections are ranked according to loss of Safranin-O staining intensity, level of cartilage fibrillation, and erosions extending into calcified cartilage. Two representative sections of six different mice (n=6) within a genotype were blindly assessed by three different readers.
Results and discussion
In 3-week-old 129/J strains, photographs reveal that both SirT1−/− and SirT1+/− mice are significantly smaller than WT mice (figure 1A). Examining SirT1+/− mice at 1 month reveals that they are smaller in size (left panel, figure 1B), while 9-month-old SirT1+/− mice appear more arched (right panel, figure 1B), possibly indicating the presence of musculoskeletal defects (figure 1B). Immunoblot analyses in figure 1B (left lower panel) show that, at 1 month, SirT1+/− mice possess less FLSirT1 (110 kDa) than WT mice, consistent with their genotype. However, 9-month-old SirT1+/− and WT mice exhibited significantly decreased levels of FLSirT1 protein compared with the equivalent 1-month-old mice (figure 1B, right lower panel). Overall, it appears that FLSirT1 levels significantly decrease in articular chondrocytes with age.
Adult (9-month-old) SirT1+/− mice show increased osteoarthritis (OA) compared with wild-type (WT) mice. (A) Photograph of 3-week-old littermates of the 129/J strain. (B) Photograph of young (1-month-old; 1M) and adult (9-month-old; 9M) SirT1+/− versus WT (SirT1+/+) mice. The lower panel of (B) shows immunoblots (IB) of freshly isolated articular chondrocytes from each genotype (n=3), probed with the N-SirT1 antibody. (C) Left panel depicts SirT1+/− versus WT (right panel) articular cartilage sections from the medial tibial plateau (MTP) stained with Safranin-O (upper, 1 month; lower, 9 months) at ×20 magnification. MTPs from adult (9-month-old) mice were blindly ranked for OA severity (graph in lower panel), based on OARSI semiquantitative scoring for mice.13 HET denotes haploinsufficient SirT1+/−, and WT denotes SirT1+/+ mice. Black arrows in the micrographs indicate MTP of each representative section. (D) Immunocytochemistry (IHC) for collagen 2α1 (Col2a1; upper panels) and aggrecan (lower panels) in adult (9-month-old) mice. Right upper graph exhibits densitometry for Col2a1 level in the extracellular matrix, while the lower right graph exhibits densitometry for aggrecan levels in murine articular chondrocytes. A.U, arbitrary units of intensity. Black arrow in SirT1+/− aggrecan staining points to irregular enlarged cell clusters in MTP articular cartilage. Black bars in (D) represent SirT1+/−, and white bars represent WT mice. Bars represent mean (SD) from six mice (with two repetitions per mouse). Arrows in the immunoblots indicate the FLSirT1 band (110 kDa).
Further inspection of cartilage morphology via Safranin-O staining did not reveal any significant differences in intensity between the genotypes at 1 month (figure 1C upper panel). However, at the age of 9 months, SirT1+/− mice exhibited decreased Safranin-O staining intensity, reflecting negatively charged glucosaminoglycans, such as aggrecan and other proteoglycans, compared with equivalent WT littermates (figure 1C lower panel). While both strains did not show any signs of OA at 1 month, overall 9-month-old mice were scored higher for OA severity (figure 1C lower panel). More specifically, 9-month-old SirT1+/− mice displayed a twofold increase in medial tibial plateau OA severity compared with WT mice of the same age, while OA severity of the medial femoral plateau showed no difference among the genotypes. Additional immunohistochemical analyses of tibial articular cartilage from 9-month-old WT mice (figure 1D) displayed reduced collagen 2α1 level in SirT1+/− compared with WT mice (figure 1D upper panel). Examining aggrecan levels showed decreased staining in SirT1+/− compared with WT mice (figure 1D lower panel). Increased aggrecan levels were mostly apparent in the superficial articular zone of 9-month-old WT mice. Interestingly, articular chondrocytes of the mid-zone in SirT1+/− mice were larger in size and formed irregular cell clusters compared with aged-matched WT mice.
As FLSirT1 was not detected in either WT or SirT1+/− mice at 9 months of age, we attempted to detect the presence of 75 SirT1 in those mice (figure 2A), by immunoblotting with an N-terminally reactive antibody to SirT1, which detects both FLSirT1 and cleaved 75 SirT1.8 Interestingly, 9-month-old SirT1+/− mice showed decreased levels of 75 SirT1 compared with WT mice. The 75 SirT1 fragment is known to be generated in the presence of TNFα and IL-1β.8 The lower panel of figure 2A presents the predicted cleavage site of mSirT1 based on the site previously characterised for human SirT1.8
Nine-month-old (9M) SirT1+/− mice show enhanced cleaved caspase 3 compared with wild-type (WT) mice. (A) Protein extracts (n=3) from duplicate SirT1+/− versus duplicate WT mice were obtained and immunoblotted (IB) with an N-SirT1 antibody. The lower panel shows the predicted cleavage site for human and mouse SirT1, which generate a cleaved SirT1 product of similar molecular mass. (B) Stably transfected human chondrocytes expressing mSirT1 were treated as indicated above the blot and expressed enhanced 75 SirT1 and active cathepsin B (aCathB) in the presence of tumour necrosis factor (TNF)α. (C) Immunohistochemistry (IHC) of the medial tibial plateau (MTP) at ×20 magnification (upper panel) and equivalent densitometry (lower graph) for chondrocytes positive for cleaved caspase 3 in adult SirT1+/− and WT mice. (D) Immunoblot analyses for cleaved caspase 3 in adult SirT1+/− versus WT mice. (E) Immunofluorescent micrographs show increased green fluorescence (twofold) for early-stage apoptotic chondrocytes (denoted as M30 cytoDEATH) in SirT1+/− versus WT mice. 4′,6-Diamidino-2-phenylindole (DAPI) was used to stain the nuclei in blue fluorescence. Dashed white bars boarder the MTP (left) from the medial femoral plateau (right). The dashed box in the merged image was magnified ×16. Bars represent mean (SD) (n=6 with two repetitions per mouse). Protein extracts were obtained from three different mice or three different stable cell lines expressing mSirT1.
Using mSirT1 stably expressing chondrocyte lines confirmed in vitro that mSirT1 is cleaved in the presence of TNFα, a process that is inhibited on the addition of the cathepsin B enzymatic inhibitor, ALLN (figure 2B). The protein levels of active cathepsin B, responsible for cleavage of FLSirT1, are also increased in TNFα-treated mSirT1 chondrocyte lines. These observations may indicate that, overall, inflammatory cytokines are systemically upregulated with age, as previously described by Chung et al.10
Since FLSirT1 is associated with enhanced viability in human chondrocytes,2 3 we initially postulated that lack of FLSirT1 may increase chondrocyte death in both strains. However, chondrocytes positive for cleaved caspase 3 were sixfold higher in adult SirT1+/− mice than WT mice, which also agrees with immunoblot analyses (figure 2C,D). In addition, early apoptotic events were more pronounced in SirT1+/− mice (twofold) than WT mice (figure 2E). The variable results obtained from these in situ assays may stem from their semiquantative nature. However, the overall findings reveal a marked increase in apoptosis in 9-month-old heterozygote mice compared with WT mice. As the majority of SirT1 in adult WT mice consists of the cleaved 75 SirT1 fragment, we postulate that it too may help to sustain chondrocyte viability in adult mice, which is consistent with our in vitro findings in human samples.4
The involvement of cathepsin B in OA pathology has been previously illustrated by Baici et al,14 who found increased secreted cathepsin B in the articular cartilage from a patient with OA compared with normal samples.14 Despite the destructive effects of cathepsin B on cartilage extracellular matrix (ECM), its intracellular activity may also affect cell biology and viability under proinflammatory stress.15 Our previous report shows that cathepsin B targets FLSirT1 for site-specific cleavage generating a stable undegraded 75 SirT1 form,8 which is transferred to the mitochondria and helps to prolong chondrocyte survival.4 Here we show that 75 SirT1 is raised in adult mice, possibly due to systemic autoimmunity apparent in the synovium,4 5 which may increase with age.10 As expected, cleavage and inactivation of FLSirT1 may inhibit the expression of various ECM genes, as seen for collagen 2α1 and aggrecan1; however, it appears from our preliminary data that this variant has a protective effect against apoptosis, similar to that of FLSirT1.2 3
In conclusion, 75 SirT1 is generated with age, possibly as a response to systemic inflammatory cues. The presence of 75 SirT1 correlates with enhanced cell survival in vivo, which establishes an alternative role for SirT1 as a longevity factor. However, generation of 75 SirT1 may occur at the cost of FLSirT1-mediated ECM expression, ultimately leading to ECM instability and cartilage breakdown.
Acknowledgments
This work was supported by the Marie Curie European IRG reintegration grant and Intramural Research Program of the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) of the National Institutes of Health. We thank Professor Michael W McBurney (Ottawa Hospital Research Institute) for his contribution of 129/J mice strains and for critically reviewing the manuscript. The retroviral expression plasmid, pHanPuro-mSirT1, was a gift from Dr Vittorio Sartorelli, NIAMS, NIH.
References
Footnotes
-
Funding NIH and EU.
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.