SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Objective

Articular chondrocyte senescence is responsible, at least in part, for the increased incidence of osteoarthritis (OA) with increased age. Recently, it was suggested that caveolin 1, a 21–24-kd membrane protein, participates in premature cellular senescence. Caveolin 1 is the principal structural component of caveolae, vesicular invaginations of the plasma membrane. This study was undertaken to investigate whether the catabolic factors oxidative stress and interleukin-1β (IL-1β) induce features of premature senescence of articular chondrocytes through up-regulation of caveolin 1 expression.

Methods

Caveolin 1 expression was investigated in human OA cartilage by real-time polymerase chain reaction and in rat OA cartilage by immunohistologic analysis. We studied whether IL-1β and H2O2 induce caveolin 1 expression in OA chondrocytes and analyzed the relationship between cellular senescent phenotypes and caveolin 1 expression in human chondrocytes.

Results

In human and rat OA articular cartilage, caveolin 1 positivity was associated with cartilage degeneration. Both IL-1β and H2O2 up-regulated caveolin 1 messenger RNA and protein levels, and both treatments induced marked expression of senescent phenotypes: altered cellular morphology, cell growth arrest, telomere erosion, and specific senescence-associated β-galactosidase activity. Caveolin 1 overexpression induced p38 MAPK activation and impaired the ability of chondrocytes to produce type II collagen and aggrecan. In contrast, down-regulation of caveolin 1 with antisense oligonucleotide significantly inhibited the features of chondrocyte senescence induced by catabolic factors. Caveolin 1 induction and stresses with both IL-1β and H2O2 up-regulated p53 and p21 and down-regulated phosphorylated retinoblastoma (Rb), suggesting that the p53/p21/Rb phosphorylation pathway, as well as prolonged p38 MAPK activation, may mediate the features of chondrocyte senescence induced by stress.

Conclusion

Our findings suggest that IL-1β and oxidative stress induce features of premature senescence in OA chondrocytes, mediated, at least in part, by stress-induced caveolin 1 expression. This indicates that caveolin 1 plays a role in the pathogenesis of OA via promotion of chondrocyte down-regulation.

Articular cartilage stability depends on the biosynthetic activities of chondrocytes, which counteract normal degradation of matrix macromolecules. Aging and the degeneration of articular cartilage in osteoarthritis (OA) are distinct processes; however, the incidence and prevalence of synovial joint degeneration increase dramatically in middle age (1) and the risk of posttraumatic OA following intraarticular fracture of the knee increases 3–4-fold after the age of 50 years (2), suggesting that age-related cartilage changes render the tissue incapable of adequately maintaining the extracellular matrix.

Previous investigations have indicated that replicative senescence of chondrocytes occurs in vivo (3–6). Studies of human articular chondrocytes from donors ranging in age from 1 year to 87 years showed that senescence-associated β-galactosidase (SA–β-gal) activity increased with advancing age, whereas both mitotic activity and mean telomere length (MTL) declined with age (5). This is indirect evidence in support of the hypothesis that age-related changes in articular cartilage increase vulnerability of the tissue to degeneration and that the association between OA and aging is due, at least in part, to features of chondrocyte senescence. Furthermore, increased SA–β-gal activity was observed in damaged OA cartilage, and MTL was shorter in cells near the lesion compared with distal sites in the same joint (6). These findings also suggest that features of chondrocyte senescence participate in the pathogenesis of OA. However, the exact mechanism of chondrocyte senescence and its implications with regard to OA pathogenesis remain unclear.

Cellular senescence is classified into 2 types: intrinsic senescence (telomere-dependent replicative senescence) and extrinsic premature senescence (telomere-independent senescence). It is thought that extrinsic senescence is induced by several types of stresses, such as oxidative stress, ultraviolet (UV) irradiation, or secretory factors (e.g., proinflammatory cytokines) (7–9). Also, mechanical and chemical stresses are well known to induce degeneration of articular cartilage. A variety of catabolic stresses involving mechanical loading, cytokines, and oxidative stress participate in the pathophysiology of OA. These catabolic factors may result in extrinsic stress–induced premature senescence of articular chondrocytes. If features of premature senescence in chondrocytes have an important role in the development and progression of OA, understanding of the mechanisms of chondrocyte down-regulation will be important for devising new approaches to the prevention and treatment of this disease.

Caveolae are vesicular invaginations of the plasma membrane. Caveolin 1 is the principal structural component of caveolae in vivo. Caveolin 2 has the same tissue distribution as caveolin 1 (10), while caveolin 3 is expressed only in striated muscle cell types (cardiac and skeletal) (11). It has been proposed that caveolins participate in vesicular trafficking events and signal transduction processes (12, 13), and several independent lines of evidence suggest that signaling molecules are sequestered, organized, and functionally regulated by caveolae microdomains (14, 15). Both caveolae and caveolin are expressed most abundantly in terminally differentiated cells, such as adipocytes, endothelial cells, and muscle cells. Interestingly, stress-induced cellular senescence is up-regulated by the expression of endogenous caveolin 1 (16, 17). These findings strongly indicate that caveolin 1 plays a central role in promoting stress-induced premature senescence. Recent work shows that caveolin 1 induces premature cellular senescence in response to various stress conditions, such as UV irradiation and oxidative stress, in primary cultures of fibroblasts, and that the senescent phenotype of fibroblasts can be reversed by reduction of caveolin 1 (17). These results suggest that caveolin 1 expression is closely involved in common stress-induced and age-related diseases.

Recently, caveolin 1 expression was demonstrated in normal human knee joint cartilage by immunohistochemical analysis (18). However, its role in joint cartilage remains unknown. Findings in the present study revealed, for the first time, that stress factors, i.e., interleukin-1β (IL-β) and oxygen free radicals, induce features of premature senescence of articular chondrocytes and that caveolin 1 mediates, at least in part, this process.

It has been reported that p38 MAPK is a senescence-executing molecule which is activated by both telomere-dependent and telomere-independent senescence-inducing stimuli (19), and that the ERK/MAPK pathway is responsible for Ras-induced senescence (20). The p53 tumor suppressor protein plays a critical role in regulating cell growth arrest (21). In addition, p21 mediates p53-dependent G1 arrest by inhibiting the activity of cyclin-dependent kinases (CDKs), which phosphorylate the retinoblastoma (Rb) gene product, as well as other substrates (22); unphosphorylated Rb does not release E2F and does not induce G1 entry into the cell cycle (23). We tried to elucidate the underlying signal transduction pathways involved in premature chondrocyte senescence induced by catabolic stresses. Features of chondrocyte senescence accelerated by catabolic stresses may contribute to the risk of cartilage degeneration by reducing chondrocyte viability.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Materials.

Recombinant human IL-1β was obtained from R&D Systems (Minneapolis, MN); collagenase, Pronase, and SB202190 from Sigma (St. Louis, MO); the Isogen RNA extraction kit from Nippon Gene (Toyama, Japan); Hybond N nylon membrane and enhanced chemiluminescence (ECL) Western blotting detection reagents from Amersham Biosciences (Piscataway, NJ); horseradish peroxidase–conjugated antibodies from Dako (Glostrup, Denmark); mouse monoclonal antiactin (C-2) IgG1 antibody from Santa Cruz Biotechnology (Santa Cruz, CA); rabbit anti–caveolin 1 polyclonal antibody from Chemicon (Temecula, CA); rabbit anti–ERK-1/2, goat anti–phospho–ERK-1/2, rabbit anti-p38, rabbit anti–phospho-p38, rabbit anti–phospho-Rb, rabbit anti-p53 polyclonal antibodies, and p21Waf1/Cip1 (DCS60) monoclonal antibody from Cell Signaling Technology (Beverly, MA); QIAprep Spin Miniprep Kits and EndoFree Plasmid Maxi Kits from Qiagen (Valencia, CA); the CGM BulletKit from Cambrex Bio Science (Walkersville, MD); Superscript II RNase H reverse transcriptase and Zero Blunt TOPO PCR Cloning Kit from Invitrogen (Carlsbad, CA); and LightCycler FastStart DNA Master SYBR Green I and LightCycler primer sets for human GAPDH, type II collagen (CII), aggrecan, and p21 from Roche (Heidelberg, Germany).

Animal OA model.

The study protocol was approved by the university ethics committee. Twelve-week-old male Sprague-Dawley rats (n = 20) weighing ∼250 gm (Charles River Japan, Yokohama, Japan) underwent unilateral anterior cruciate ligament (ACL) and medial collateral ligament (MCL) transection as described previously (24), with slight modification. Briefly, animals were anesthetized with intraperitoneal sodium pentobarbital, and ACL and MCL were performed on 1 hind limb. An incision on the medial side of the patellar tendon provided access to the joint space, after which the patella was dislocated laterally with the leg in extension, and the ACL and MCL transected. A positive finding on an anterior drawer test and lateral instability test ensured complete transection of the ligaments. After relocation of the patella, the joint capsule was closed. The contralateral knee capsule was incised only longitudinally, as a sham operation. The articular capsule and skin were washed with saline and sutured.

To elucidate the involvement of caveolin 1 in the progression of cartilage degeneration, we investigated the levels of caveolin 1 expression in chondrocytes prior to the development of cartilage degeneration in the OA rat model. Rats were killed by administration of diethyl ether 1, 2, 3, 4, or 6 weeks (n = 4 at each time point) after the operation. Both knees were harvested, and articular bone and cartilage were obtained and paraffin blocks prepared using standard histologic procedures. Serial sections of paraffin-embedded bone and cartilage tissue were cut and immunostained using antibody to caveolin 1. The sections were dewaxed and irradiated with microwaves in 0.01M sodium citrate buffer (pH 6.0) at 850W, twice for 5 minutes each time. After washing in phosphate buffered saline (PBS; pH 7.4), the sections were treated with 0.3% H2O2 for 30 minutes. After blocking with goat serum (1:70), the sections were incubated for 1 hour at 37°C with primary antibody against caveolin 1 (dilution 1:100). The primary antibody was detected with biotinylated secondary antibodies, followed by incubation with streptavidin–biotin–peroxidase complex. Peroxidase activity was visualized with 3,3′-diaminobenzidine. Sections were counterstained with Mayer's hematoxylin.

Isolation and culture of chondrocytes.

Full-thickness human articular cartilage samples were obtained from the knee joints of 8 female patients with OA (age range 60–74 years) who were undergoing knee arthroplasty. Informed consent was obtained in accordance with the local ethics commission.

The harvested cartilage samples were chopped, and digested overnight at 37°C in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum (FCS), 0.5 mg/ml Pronase E, and 0.5 mg/ml type II collagenase. The resulting single-cell suspension was filtered through a 70-μm nylon mesh strainer, and the cells were incubated at 37°C in a 5% CO2 humidified atmosphere in DMEM containing 10% FCS, 2 mML-glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin. We investigated the phenotypes of chondrocytes during the study period and found that the chondrocyte phenotype was retained during the period of incubation (results not shown).

When cells were subconfluent, the monolayer was subjected to enzymatic dissociation with trypsin and seeded onto new dishes. Twenty-four hours after subculture, the adherent cells were subjected to stress with 50 μM H2O2 or 10 ng/ml IL-1β for 24 hours, and then washed with DMEM and replenished with fresh medium containing 10% FCS.

Chondrocytes used in the transfection and antisense experiments were obtained from articular cartilage of a 66-year- old patient with OA, after informed consent was obtained. Chondrocytes were isolated and cultured by the methods described above.

Chondrocyte proliferative lifespan assay.

The cells were subcultured every 7–10 days. At each subculturing, the total number of cells was determined by counting with a hemocytometer, and cells were transferred to new dishes at a ratio of 1:3. Nonadherent cells were counted 6 hours after seeding, and that number was subtracted from the number of seeded cells. The increase in cumulative population doublings at each subculture was calculated based on the number of cells attached and the cell yield at the time of the next subcultivation. The end of the replicative lifespan was defined by failure of the population to double after 4 weeks in mass cultures. The in vitro lifespan (remaining replicative capacity) was expressed as population doubling (PD) up to cellular senescence (25). The population doublings were calculated with the equation PD = log10 (N/N0) × 3.33, where N is the number of cells at the end of the experiment and N0 is the number of cells at the beginning of the experiment (26). To retain the phenotypic characteristics of chondrocytes during serial passages to determine the lifespan, we used a CGM BulletKit (chondrocyte growth medium) instead of DMEM.

Flow cytometric analysis for cell cycle phase.

After trypsin treatment, detached cells were fixed with 70% ice-cold ethanol. The cells were digested with 100 μg/ml RNase at 37°C for 20 minutes and then stained with 12.5 μg/ml propidium iodide for 30 minutes in the dark. Filtered samples were then analyzed for cycle content on a FACSCalibur cell sorter, using CellQuest software (Becton Dickinson, Mountain View, CA), and the percentages of cells in the G1, S, and G2/M phases were determined using ModFit LT software (Verity Software House, Topsham, ME).

SA–β-gal activity assay.

SA–β-gal activity was detected as previously described (27), using the Senescence Detection Kit according to the recommendations of the manufacturer (BioVision, Mountain View, CA). Cells were photographed under reflected light using a digital high-fidelity microscope (VH-8000; Keyence, Osaka, Japan).

Assessment of chondrocyte telomere length by Southern blotting.

Genomic DNA was extracted from cells (∼1–5 × 106) using a DNeasy kit according to the recommendations of the manufacturer (Qiagen). The DNA concentration of each sample was determined by UV spectrophotometry, and 2 μg was digested with 20 units each of Hinf I and Rsa I in a 60-μl reaction. The reactions were electrophoresed on 0.8% agarose in parallel with digoxigenin-labeled λ Hind III size standards. The gel was transferred by capillary action onto a Hybond N+ nylon membrane in 20× saline–sodium citrate (SSC) and baked for 2 hours at 80°C. The membrane was prehybridized for 4–16 hours at 37°C in hybridization buffer (50% formamide, 5× SSC, 0.1% sodium lauryl sulfate, 0.02% sodium dodecyl sulfate [SDS], and 2% blocking agent). A synthetic oligonucleotide (ODN) complementary to human telomeric repeat sequences, (CCCTAA)3, labeled at the 3′ end with digoxigenin, was diluted to 50 pM in hybridization buffer, and the membrane was probed for 16–24 hours at 37°C. Excess probe was removed by washing the membrane in 2× SSC with 0.1% SDS at ambient temperature (twice for 15 minutes each time), then in 0.5% SSC with 0.1% SDS at 37°C (twice for 15 minutes each time). A goat antidigoxigenin alkaline phosphatase–conjugated antibody and a chemiluminescent substrate (CDP-Star; Roche, Mannheim, Germany) were used to detect the digoxigenin-labeled probe. Autoradiograms of the blots were developed by exposure to x-ray film. The terminal restriction fragment MTL was determined from densitometric analysis of the chemiluminescence signals, using the method of Harley et al (28).

Reverse transcriptase–polymerase chain reaction (RT-PCR) and quantitative real-time RT-PCR.

To investigate whether catabolic stresses induce expression of caveolin 1 in chondrocytes, caveolin 1 messenger RNA (mRNA) in cultured chondrocytes was analyzed by RT-PCR. OA cartilage samples were obtained from the knee joints of 8 OA patients, and control articular cartilage samples from the knee joints of 5 patients with traumatic knee injury.

Total RNA was extracted from the chondrocytes by acid guanidine–phenol–chloroform extraction using Isogen. First-strand complementary DNA (cDNA) was synthesized with Superscript II reverse transcriptase. PCR amplification was performed using specific primers (Table 1). The PCR products were analyzed by electrophoresis in 2% agarose gels stained with ethidium bromide, and bands were visualized and photographed under UV excitation.

Table 1. Sequences of polymerase chain reaction primers, length of polymerase chain reaction products, optimal annealing temperatures, and sequence accession numbers from the NCBI Entrez search system
 Accession no.Primer (forward, reverse)Product, bpAnnealing temperature, °CCycles
GAPDHNM_0020465′-GAAGGTGAAGGTCGGAGTC-3′2266030
  5′-GAAGATGGTGATGGGATTTC-3′   
Caveolin 1NM_0017535′-AAGGAGATCGACCTGG-3′3095835
  5′-GGAATAGACACGGCTG-3′   
p53NM_0005465′-AGCATCTTATCCGAGTGG-3′3006035
  5′-TCTTGCGGAGATTCTCTT-3′   
p16BC0219985′-GCCACTCTCACCCGAC-3′2036040
  5′-GCATGGTTACTGCCTCT-3′   
RetinoblastomaNM_0003215′-CCTCGAAGCCCTTACA-3′2606032
  5′-AGTGGTTTAGGAGGGTTG-3′   

Quantitative real-time RT-PCR was performed with a spectrofluorometric LightCycler, using FastStart DNA Master SYBR Green I. For each run, a standard curve was generated from purified DNA, ranging from 10 to 105 copies. Specificity of the expected products was demonstrated by melting curve analysis. To standardize mRNA levels, we amplified GAPDH as an internal control. Normalized gene expression was calculated as the ratio between the copy number of the gene of interest and that of GAPDH cDNA. The template source was either 5 ng first-strand cDNA or purified DNA standard.

Blunt PCR products of p53, p21, p16, and Rb were subcloned into a Blunt TOPO PCR vector, by cloning. The subcloned vectors were transformed into DH5α-T1–competent cells. After amplification, each insert was prepared by restriction enzyme digestion, checked with a sequencing analyzer, and used as a DNA standard.

Immunoblotting.

To determine whether catabolic stresses induce the expression of caveolin 1 in chondrocytes, protein expression of caveolin 1 in cultured chondrocytes was investigated by Western blotting. Cells were lysed in a buffer containing protease inhibitors, and the protein concentrations in the whole cell lysates were determined by Bradford assay (Bio-Rad, Hercules, CA). Equal amounts of protein were size-fractionated by SDS–polyacrylamide gel electrophoresis and transferred electrophoretically onto polyvinylidene difluoride membranes. The membranes were blocked with 5% nonfat dry milk in PBS–0.1% Tween 20, probed with primary antibody at appropriate dilutions, and reacted with the appropriate horseradish peroxidase–conjugated secondary antibody. The resultant complexes were processed for detection using an ECL Western blotting detection system. Densitometry of the signal bands was analyzed with Image Gauge version 4.0 (Fuji, Tokyo, Japan).

Down-regulation of caveolin 1 by antisense oligonucleotide.

Although application of antisense ODN may only partially block caveolin 1 expression, we investigated the inhibitory effect of down-regulation of caveolin 1 on chondrocyte biology, using an antisense ODN. The caveolin 1 antisense ODN was designed and synthesized by Biognostik (Gottingen, Germany). Randomized ODN was used as control. The cellular uptake efficiency was monitored using a fluorescein isothiocyanate–labeled ODN. The chondrocytes were incubated with 2 μM antisense or control ODN at 37°C for 4 hours.

Transfection of chondrocytes with caveolin 1 cDNA–expressing plasmid.

The full-length cDNA encoding human caveolin 1 was subcloned into a pcDNA3.1/NT-GFP-TOPO vector (Invitrogen), by cloning. The subcloned vectors were transformed into TOP10 chemically competent cells. After amplification, each insert was checked with a sequencing analyzer. Chondrocytes were transiently transfected with a cDNA-encoding caveolin 1 gene by electroporation using Human Chondrocyte Nucleofector kits according to the protocol recommended by the manufacturer (Amaxa Biosystems, Cologne, Germany). The vector with no insert was used as a control. The expression of Cycle 3 GFP (enhanced green fluorescent protein), which was also encoded by the plasmid, was checked for transfection efficiency by fluorescence microscopy. Caveolin 1 mRNA and protein were quantified by RT-PCR and Western blotting, respectively, as described above.

Enzyme-linked immunosorbent assay (ELISA).

To elucidate the significance of stress-induced senescence of chondrocytes in the development and/or progression of OA, we studied the chondrocytes' capability to produce the main components of articular cartilage, proteoglycan (aggrecan) and CII. The amount of CII released from cultured chondrocytes was determined by ELISA using a Native Type II Collagen Detection Kit (Chondrex, Redmond, WA). Aggrecan content in the chondrocyte culture medium was measured by ELISA according to the protocol of the kit manufacturer (BioSource International, Camarillo, CA).

Statistical analysis.

Results were expressed as the mean ± SD. Means were compared by analysis of variance. P values less than 0.05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Caveolin 1 mRNA expression in articular cartilage from patients with OA.

As assessed by real-time PCR analysis, the levels of caveolin 1 mRNA in 6 of the 8 human OA articular cartilage samples (75%) were higher in the central degenerated regions than in the peripheral regions with less degeneration in the same samples (Figure 1A).

thumbnail image

Figure 1. Expression of caveolin 1 in articular cartilage from osteoarthritis (OA) patients and in a rat OA model. A, In 6 of 8 patients with OA (75%), levels of caveolin 1 mRNA were significantly higher in the central degenerated regions than in the peripheral regions with less degeneration from the same sample (∗ = P < 0.05; ∗∗ = P < 0.01). B, Gradual increases in caveolin 1 expression and cartilage degeneration were observed in surface cartilage of the tibial plateaus of the ligament-transected knees with increasing time after the operation, whereas increased caveolin 1 expression was not seen even at 4 weeks after sham operation. Arrows indicate cartilage degeneration; asterisks indicate the depth of the articular cartilage layer. The mean depth of articular cartilage decreased gradually with time in the ligament-transected rats, but not in sham-operated animals. Bars = 50 μm.

Download figure to PowerPoint

Caveolin 1 expression during progression of articular cartilage degeneration in rats with OA.

Two weeks after MCL/ACL transection in the rat OA model, both caveolin 1 expression and cartilage degeneration were observed in superficial cartilage from 75% of the femoral condyles (3 of 4 rats) and 100% of the tibial plateaus (4 of 4 rats) from the ligament-transected knees. The mean depths of articular cartilage decreased gradually with time in the transected groups, whereas no significant changes in the depth of the cartilage layer were observed in sham-operated groups.

With increasing time after the operation, we observed a gradual increase in caveolin 1 expression in articular cartilage from tibial plateaus of the transected rats (Figure 1B). In contrast, in control right knees (sham operated), there was no significant difference in caveolin 1 expression in articular cartilage over 6 weeks, and no significant progression of cartilage degeneration was observed during the 6 weeks.

Induction of premature senescence of articular chondrocytes by IL-1β and H2O2.

The phenotype of cells with stress-induced senescence was analyzed based on increased cell size, SA–β-gal activity, and cell cycle arrest at the G0/G1 phase. When chondrocytes were incubated with 10 ng/ml IL-1β or 50 μM H2O2 for 24 hours, both treatments markedly increased the percentage of cells in the G0/G1 phase and reduced the percentage in the S phase (Figure 2A). After stimulation with IL-1β or H2O2 for 24 hours, the chondrocytes were cultured in fresh medium for another 5 days. Both stresses induced the chondrocytes to become large and flat (Figure 2B) and enhanced SA–β-gal activity in the chondrocytes (Figure 2C).

thumbnail image

Figure 2. Induction of features of cellular senescence in osteoarthritic chondrocytes by interleukin-1β (IL-1β) and H2O2. A, Chondrocytes were treated with 10 ng/ml IL-1β or 50 μM H2O2 for 24 hours and then were stained with propidium iodide for flow cytometric analysis. The percentages of cells in the G1, S, and G2/M phases of the cell cycle were determined using ModFit LT software. B and C, Cellular morphology (B) and senescence-associated β-galactosidase (SA–β-gal) staining (C) were determined after chondrocytes were cultured in fresh medium without stress for another 5 days after having been stressed with 10 ng/ml IL-1β or 50 μM H2O2 for 24 hours. Bars = 55 μm. D, Mean telomere length (MTL) of chondrocytes stressed with IL-1β or H2O2 at each passage. The primary culture of chondrocytes from the same specimen, designated as 0 population doublings (PD), was also investigated. Chondrocytes cultured in the absence of stress throughout the passages were used as controls. Results shown are representative of 4 independent experiments.

Download figure to PowerPoint

In each passage, the chondrocytes were stressed with IL-1β or H2O2 for 1 day. After 5 stresses, the MTL was significantly shortened (Figure 2D). The lifespan of chondrocytes in the absence of stress was 36.8 ± 4.6 PD (mean ± SD; n = 4), whereas chondrocyte lifespan was 24.3 ± 4.9 PD (n = 4) after repeated stress with IL-1β (P = 0.033 versus controls) and 20.5 ± 5.7 PD (n = 4) after repeated stress with H2O2 (P = 0.012 versus controls). In preliminary experiments (data not shown), we also confirmed the IL-1β–induced premature senescence at several concentrations of IL-1β (0.1, 1.0, 5.0, and 10.0 ng/ml) in vitro. The results indicated that premature senescence of chondrocytes was induced by IL-1β at lower concentrations as well. Data from 4 independent experiments were analyzed.

Up-regulation of caveolin 1 expression by IL-1β and H2O2.

Quantitative real-time RT-PCR analysis showed that both IL-1β and H2O2 enhanced caveolin 1 mRNA expression in a time-dependent manner, with the peak level occurring at 3 hours after stimulation (Figures 3A and B) (n = 3). By Western blot analysis, we confirmed that both IL-1β and H2O2 also induced prolonged up-regulation of caveolin 1 protein expression (Figures 3C and D) (n = 3).

thumbnail image

Figure 3. Up-regulation of caveolin 1 expression in osteoarthritic (OA) chondrocytes by IL-1β and H2O2 stimulation. AD, Chondrocytes were incubated with 10 ng/ml IL-1β or 50 μM H2O2 for the indicated times. Caveolin 1 mRNA expression was analyzed by real-time reverse transcriptase–polymerase chain reaction (A), and levels relative to GAPDH were quantitated (B) (n = 3). Caveolin 1 protein levels were analyzed by Western blotting (C), and the bands were subjected to densitometry analysis with standardization to actin (D) (n = 3). Results shown are representative of 3 independent experiments. ∗ = P < 0.05; ∗∗ = P < 0.01, compared with the initial phase (0 hours after stimulation). E, OA chondrocytes were pretreated with 2 μM antisense (AS) or control oligonucleotide (ODN) for 2 days, and were further stimulated with 10 ng/ml IL-1β or 50 μM H2O2 for 24 hours. Whole cell lysates were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis and Western blotting. Bands were subjected to densitometry analysis with standardization to actin (n = 4). Results shown are representative of 4 independent experiments. F, After pretreatment with 2 μM antisense or control ODN for 2 days and further stimulation with 10 ng/ml IL-1β or 50 μM H2O2 for 24 hours, the OA chondrocytes were incubated in fresh medium with 2 μM ODN for a further 3 days, and stained for SA–β-gal. Bars = 90 μm. Values in B, D, and E are the mean or the mean and SD. See Figure 2 for other definitions.

Download figure to PowerPoint

Down-regulation of caveolin 1 blocks stress-induced senescence of chondrocytes.

When chondrocytes were incubated with caveolin 1 antisense ODN, constitutive caveolin 1 expression was successfully down-regulated. Pretreatment with caveolin 1 antisense ODN significantly inhibited the up-regulation of caveolin 1 expression induced by IL-1β and H2O2 (Figure 3E) (n = 4) and also significantly blocked the increase in SA–β-gal activity induced by IL-1β and H2O2 (Figure 3F) (n = 4).

Prolonged activation of p38 MAPK is responsible for features of chondrocyte senescence.

Expression levels of phospho–p38 MAPK were markedly increased after IL-1β stimulation, with a peak level reached at 12 hours and sustained for more than 24 hours, whereas levels of phospho–ERK-1/2 were transiently down-regulated from 6 to 12 hours (Figure 4A). H2O2 also significantly and persistently up-regulated the level of phospho–p38 MAPK, but did not have a significant effect on ERK activation (Figure 4B). Furthermore, pretreatment with the specific p38 MAPK inhibitor SB202190 blocked the enhancement of SA–β-gal activity induced by IL-1β and H2O2 (Figure 4C). Results from 3 independent experiments were analyzed.

thumbnail image

Figure 4. Activation of p38 MAPK mediates features of senescence in chondrocytes. A and B, after incubation with 10 ng/ml IL-1β or 50 μM H2O2 for the indicated times, osteoarthritic (OA) chondrocytes were harvested and lysed. Whole cell lysate proteins were subjected to immunoblotting using specific antibodies against human p38 MAPK, phospho-p38, ERK-1/2, and phospho–ERK-1/2. Results shown are representative of 3 independent experiments. C, OA chondrocytes were pretreated with 10 μM SB202190, a specific p38 MAPK inhibitor, for 1 hour prior to the stress. The cells were subjected to SA–β-gal staining 5 days after stress. Also, in another experiment, cells were subjected to SA–β-gal staining 4 days after caveolin 1 (Cav-1) transfection. Bars = 90 μm. D, Caveolin 1 cDNA–expressing plasmids were transduced into OA chondrocytes by transfection. Three days after transfection, the cells were harvested and lysed. Whole cell lysate proteins were subjected to immunoblotting using specific antibodies, as described in Figure 3. E, Levels of p21, p53, retinoblastoma (Rb), and p16 mRNA were analyzed by quantitative reverse transcriptase–polymerase chain reaction (RT-PCR). Normalized gene expression levels were expressed as the ratio between the gene of interest and the GAPDH cDNA copy number. Values are the mean and SD from 3 independent experiments. F, Levels of p53, p21WAF1, and phosphorylated Rb protein were analyzed by Western blotting (p16 protein was undetectable even after stimulation). Results shown are representative of 3 independent experiments. See Figure 2 for other definitions.

Download figure to PowerPoint

Caveolin 1 overexpression induces p38 MAPK activation and cellular senescence in chondrocytes.

Compared with chondrocytes transfected with empty vectors, the chondrocytes transduced with caveolin 1 cDNA showed higher levels of phospho–p38 MAPK and lower levels of phospho–ERK-1/2 (Figure 4D). This further confirmed the occurrence of caveolin 1 overexpression–induced chondrocyte senescence that was indicated by the SA–β-gal staining results.

Up-regulation of p53 and p21 and dephosphorylation of Rb protein in chondrocytes.

Quantitative real-time RT-PCR analysis showed increased expression levels of p53 and p21 mRNA in chondrocytes 3 days after stress with IL-1β and H2O2 (Figure 4E). Caveolin 1 overexpression led to up-regulated levels of p53 and p21 mRNA (Figure 4E). Neither stress (IL-1β, H2O2) nor caveolin 1 overexpression significantly affected the level of Rb mRNA expression (Figure 4E). Western blot analysis demonstrated that caveolin 1 overexpression and both stresses (IL-1β, H2O2) up-regulated p53 and p21 protein levels and down-regulated phosphorylated Rb protein levels (Figure 4F). The constitutive expression level of p16 mRNA was marginal in chondrocytes, although both stresses, as well as caveolin 1 overexpression, increased its level by ∼2 fold (Figure 4E). However, p16 protein was undetectable by Western blotting, even after stimulation (data not shown).

Impaired production of matrix proteins by chondrocytes with stress-induced senescence.

Quantitative real-time RT-PCR revealed a marked reduction in levels of expression of mRNA for CII (Figure 5A) and aggrecan (Figure 5B) in chondrocytes 3 days after stress with IL-1β or H2O2. Decreased production of CII from senescent chondrocytes after administration of either stress factor was also confirmed by ELISA (Figure 5C) (n = 4). In addition, caveolin 1 overexpression impaired the ability of chondrocytes to express CII (Figure 5A) (n = 4) and aggrecan (Figure 5B) (n = 4).

thumbnail image

Figure 5. Impaired production of type II collagen (CII) and aggrecan (Agg) in osteoarthritic chondrocytes subjected to stress. A and B, RNA was extracted from chondrocytes 5 days after stimulation with interleukin-1β (IL-1β) (10 ng/ml) or H2O2 (100 μM) or 4 days after transfection with caveolin 1 cDNA. Expression levels of CII and aggrecan mRNA were analyzed by real-time reverse transcriptase–polymerase chain reaction (n = 4). C, Chondrocytes were seeded on 24-well plates 3 days after stimulation with IL-1β (10 ng/ml) or H2O2 (100 μM) and incubated for another 2 days. CII was digested from the plates and levels were determined by enzyme-linked immunosorbent assay. Values are the mean or the mean and SD. ∗∗ = P < 0.01 compared with chondrocytes cultured in the absence of stress, by analysis of variance.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Our results indicate that the catabolic stresses IL-1β and oxidative stress induce features of chondrocyte senescence through the overexpression of caveolin 1 in articular cartilage. Recently, attention has been focused on caveolin 1 overexpression–induced cellular senescence in human somatic cells (16, 17). Our findings of stress-induced expression of caveolin 1 and its involvement in features of chondrocyte senescence suggest that caveolin 1 has a role in the pathogenesis of articular cartilage degeneration in OA, by promoting stress-induced down-regulation of chondrocytes.

We observed expression of caveolin 1 in degenerated cartilage from rats with OA. Stronger expression of caveolin 1 in chondrocytes was found in central degenerated regions as compared with peripheral regions with less degeneration from the same articular cartilage samples. In samples (central degenerated region and peripheral region with less degeneration) from the same articular cartilage specimens from human OA patients, positivity for caveolin 1 was associated with histologic changes characteristic of OA. Furthermore, in rats with OA induced by transection of the ACL and MCL, our histologic analysis revealed an increase in caveolin 1 expression in chondrocytes, occurring prior to the progression of articular cartilage degeneration. Our findings in human articular cartilage tissue and in a rat OA model thus indicate for the first time that expression of caveolin 1 is involved in the progression of articular cartilage degeneration.

It has long been theorized that oxidative stress, resulting in the accumulation of proteins that have undergone oxidative damage, is the cause of aging-related changes in a number of tissues (29–31). Detection of nitrotyrosine, a marker of oxidative damage, in articular cartilage supports the notion of a role for oxidative damage to cartilage in the setting of both aging and OA (32). Articular chondrocytes actively produce endogenous reactive oxygen species, including nitric oxide (NO) (33), O2 (34), HO (35), and H2O2 (36), suggesting the occurrence of oxidative stress in articular cartilage. In the present study, we confirmed that, although H2O2 in nontoxic levels (<100 μM) did not affect the viability of chondrocytes, chondrocytes showed the characteristic phenotypes of cellular senescence after exposure to oxidative stress (37).

This is the first reported study to demonstrate that features of chondrocyte senescence are induced by IL-1β. IL-1 has been shown both to inhibit chondrocyte anabolic activity, including proteoglycan synthesis (38), and to stimulate catabolic activity, including production of metalloproteinases (39). It has also been shown to stimulate chondrocyte expression of reactive oxygen species such as NO, which results in increased oxidative damage (40). Our finding of features of stress-induced senescence further extends knowledge of the roles of IL-1β in cartilage degeneration.

Cellular senescence represents an arrested state in which the cells remain viable but are not stimulated to divide by serum or passage in culture. In the present study, cell cycle analysis demonstrated that treatment with either IL-1β or H2O2 significantly induced chondrocyte arrest at the G0/G1 phase. This is further supported by evidence that IL-1β inhibits 3H-thymidine incorporation in cultured articular chondrocytes (41, 42). Our data suggest that H2O2 also increases the percentage of chondrocytes in the G2/M phase of the cell cycle; this phenomenon has also been found in fibroblasts after H2O2 stimulation (43).

Unlike quiescent cells, senescent cells cannot replicate, but instead become enlarged and show multiple molecular changes when stimulated with any physiologic mitogen (44). Senescent cells exhibit enlarged cell volume, an increase in cell surface area, and altered morphology. In the present study, we demonstrated that both stresses induced the chondrocytes to become large and flat. In addition, studies have demonstrated that senescent cells elevate the activity of a unique neutral β-galactosidase, SA–β-gal (26). Because SA–β-gal activation is not related to growth arrest and can be determined in cells in situ, it serves as an ideal biomarker for qualitative and quantitative determination of senescence. Thus, senescence can be measured by activation of SA–β-gal in addition to cell enlargement and inability to replicate. In the present study, we also found that SA–β-gal activity was enhanced in chondrocytes 5 days after a single administration of stress with IL-1β or H2O2. Taken together, these results indicate that both catabolic stresses induce features of chondrocyte senescence.

Erosion beyond the minimum critical telomere length necessary for DNA replication (5–7.6 kbp) results in cell cycle arrest, a phenomenon referred to as replicative senescence (44, 45). Our data showed that stresses with IL-1β and H2O2 accelerated telomere erosion in chondrocytes, and consequently, their lifespans were significantly shortened. These results suggest that in vivo long-term exposure to stress factors may accelerate telomere erosion and induce replicative senescence in arthritic chondrocytes.

Down-regulation of caveolin 1 is sufficient to drive cell transformation to tumorigenicity (46). Previous reports suggest that caveolin 1 may modulate the lifespan of cells (47). Recent work showed that caveolin 1 overexpression arrests mouse embryonic fibroblasts in the G0/G1 phase (48) and that caveolin 1 plays a central role in promoting cellular senescence in fibroblasts (16). Our data showed that both IL-1β and H2O2 up-regulated caveolin 1 mRNA and protein levels. Furthermore, caveolin 1 antisense ODN blocked the features of chondrocyte senescence after stress with IL-1β and H2O2, and caveolin 1 overexpression was able to induce features of chondrocyte senescence. These results indicate that up-regulated expression of caveolin 1 mediates the features of chondrocyte senescence induced by stress with IL-1β and H2O2.

In this context, we also studied the signal transduction pathways involved in features of chondrocyte senescence. It has been reported that p38 MAPK is a senescence-executing molecule that is activated by both telomere-dependent and telomere-independent senescence-inducing stimuli (19), and that the ERK/MAPK pathway is responsible for Ras-induced senescence (20). We found that both IL-1β and H2O2 induced prolonged activation of p38 in chondrocytes. Caveolin 1 overexpression also induced p38 activation. The specific p38 MAPK inhibitor SB202190 blocked features of chondrocyte senescence induced by stress with IL-1β and H2O2, which further supports the notion that prolonged p38 MAPK activation is necessary for stress-induced chondrocyte activity. IL-1β transiently inhibited ERK activation 6–12 hours after stimulation, and H2O2 did not significantly affect ERK activation during the observation period. These findings suggest that ERK activation might not be a common pathway involved in features of chondrocyte senescence accelerated by IL-1β and H2O2. Furthermore, we also found that caveolin 1 overexpression inhibited ERK activation in human articular chondrocytes.

The p53 tumor suppressor protein plays a critical role in regulating cell growth arrest (21); p21 mediates p53-dependent G1 arrest by inhibiting the activity of CDKs, which phosphorylate the Rb gene product, as well as other substrates (22). Unphosphorylated Rb does not release E2F and does not induce G1 entry into the cell cycle (23). In this study, we found that caveolin 1 expression and stresses with both IL-1β and H2O2 up-regulated p53 and p21, but down-regulated phosphorylated Rb in articular chondrocytes. These results suggest that the p53/p21/Rb dephosphorylation pathway may mediate the stress-induced chondrocyte growth arrest.

Although loss of the ability of cells to divide is an accepted measure of senescence, cell function may begin to deteriorate before cell cycle arrest is reached. To demonstrate dysfunction of stress-induced chondrocytes, we confirmed their decreased production of articular cartilage matrix CII and aggrecan. In contrast to the results of previous investigations (3, 30), our studies demonstrated the expression of CII and aggrecan in chondrocytes 3–5 days after stress with IL-1β or H2O2, rather than during incubation with IL-1β or H2O2. An age-related decline in anabolic response to insulin-like growth factor 1 was found in articular cartilage chondrocytes, and chondrocytes from OA cartilage also have a reduced response to insulin-like growth factor 1 (1). These data therefore provide evidence that a cumulative increase of premature senescent chondrocytes in the articular cartilage may provide an alternative explanation for the degeneration of articular cartilage in OA.

In conclusion, our data strongly suggest that in OA, exposure to IL-1β and oxidative stress induces features of chondrocyte senescence, mediated, at least in part, by caveolin 1. Prolonged p38 MAPK activation and the p53/p21/Rb dephosphorylation pathway are necessary to mediate the features of stress-induced senescence. Further studies are needed to clarify whether blocking of caveolin 1 expression prevents cartilage degeneration. Features of chondrocyte senescence accelerated by catabolic factors contribute to the risk of cartilage degeneration by reducing the ability of cells to maintain and repair tissue.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We are grateful to Ms Yumiko Sugiyama for excellent technical assistance.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES