Perturbation of nuclear lamin A causes cell death in chondrocytes

Authors


Abstract

Objective

Mutations in LMNA encoding the A-type lamins cause several diseases, including those with features of premature aging and skeletal abnormalities. The aim of this study was to examine the expression of lamin A in cartilage from patients with osteoarthritis (OA) and the effects of its overexpression on chondrocyte senescence and apoptosis.

Methods

Human chondrocyte-like cells (SW-1353) were used. RNA isolated from human OA and non-OA cartilage was used for profiling messenger RNA expression, using Affymetrix microarray analysis. The effects of lamin A overexpression on mitochondrial function and apoptosis were examined by assessing mitochondrial membrane potential, ATP levels, and cytochrome c release, and with a TUNEL assay. Western blotting was performed to determine protein expression.

Results

Lamin A expression was markedly elevated in OA cartilage samples compared with non-OA control samples. Western blot analysis confirmed increased expression of lamin A in OA compared with non-OA cartilage. Interleukin-1β treatment inhibited lamin A accumulation, whereas treatment with prostaglandin E2 (PGE2) caused a marked increase in lamin A accumulation. These effects of exogenous PGE2 on lamin A expression were mediated via the EP2/EP4 receptors. Transfected chondrocytes that expressed lamin A displayed markers of early senescence/apoptosis.

Conclusion

The results of this study suggest that lamin A is up-regulated in OA chondrocytes, and that increased nuclear accumulation of lamin A in response to catabolic stress may account for the premature aging phenotype and apoptosis of OA chondrocytes.

Osteoarthritis (OA) is an age-related disease characterized by progressive loss of articular cartilage, new bone formation, and often, synovial proliferation. Over the past decade, there have been significant developments in the scientific understanding of OA, including the identification of a variety of mediators and signaling pathways that contribute to cartilage and bone remodeling (1, 2). However, the molecular mechanisms that play a role in the initiation and perpetuation of arthritis are not clear.

Central to OA is the altered, catabolic phenotype of the articular chondrocyte, the single cellular component in cartilage. Therefore, gaining an understanding of the dysregulation of chondrocyte function, change in phenotype, and altered extracellular matrix interactions in OA is the major focus of investigations. OA chondrocytes undergo a series of complex changes, including hypertrophy, proliferation, catabolic alteration, and, ultimately, death. These characteristic changes result in the loss of proteoglycans and collagens due to increased production of matrix metalloproteinases and aggrecanases.

Because age is the risk factor most strongly correlated with OA, understanding age-related changes in cartilage is essential. During the aging process, chondrocytes undergo a decrease in mitotic and synthetic activity, exhibit decreased responsiveness to anabolic growth factors, and synthesize smaller and less uniform large aggregating proteoglycans and fewer functional link proteins (3). Age also appears to be an independent factor that predisposes articular chondrocytes to apoptosis, because the expression levels of specific proapoptotic genes (Fas, FasL, caspase 8, and p53) are higher in “aged” cartilage (4, 5).

The relationship between the normal aging process and the molecular and biochemical events that lead to chondrocyte senescence in OA is, therefore, being researched. Several investigators (6–10) have suggested that inflammatory mediators associated with OA may promote “premature” chondrocyte senescence, leading to progressive degeneration of cartilage. For example, increasing evidence suggests that oxidative stress due to the production of reactive oxygen and nitrogen species is particularly implicated in chondrocyte senescence and cartilage aging (11).

In an effort to gain insights into the accelerated cartilage aging phenomena observed in OA, we sought to determine whether lamin A, a nuclear envelope protein that has been implicated in progeria, could play a role in the premature senescence of OA chondrocytes. Lamins are intermediate filament proteins that form the nuclear lamina, a meshwork of intermediate filaments on the inner nuclear envelope membrane. In humans, 3 genes encode nuclear lamins. LMNA encodes the A-type lamins, consisting of lamin A and lamin C, the major somatic cell isoforms. Lamins provide physical scaffolding and structural support for the nucleus and function as an anchor for various proteins, some of which interact with DNA. Both direct and indirect interactions between lamins and chromatin may affect gene transcription, nuclear organization, transport of material in and out of the nucleus, cell cycle regulation, and cell differentiation (12, 13).

Mutations in LMNA cause inherited diseases that are collectively known as laminopathies (14). One of these diseases is Hutchinson-Gilford progeria syndrome, in which the LMNA mutation leads to a defect in prelamin A processing, resulting in the accumulation of a truncated, permanently farnesylated lamin A variant. This leads to accelerated aging of mesenchymal tissues and the development of bone and joint abnormalities at young ages (15). Furthermore, the A-type lamins play an important role in cell responses to mechanical force (16).

For these reasons, we examined the potential role of lamin A in OA and observed that lamin A was up-regulated in OA cartilage. Here, we provide evidence that increased expression of lamin A causes mitochondrial dysfunction, ATP depletion, and chondrocyte apoptosis.

MATERIALS AND METHODS

Reagents.

All of the experimental media and fetal bovine serum (FBS) were purchased from Life Technologies. IL-1β was purchased from PeproTech, and enzyme-linked immunosorbent assay (ELISA) kits for the detection of caspase 3 and cytochrome c were purchased from R&D Systems and Active Motif, respectively. Other chemicals, EP2 receptor antagonist (AH6809), EP4 receptor antagonist (AH23848), and kits for chemiluminescence detection and ATP analysis were purchased from Sigma-Aldrich. Mitochondrial JC-1 dye was purchased from Molecular Probes. The antibodies for Western blot analysis were obtained from various sources; lamin A antibodies were obtained from Abcam; lamin B1, p16, and p21 antibodies were obtained from Santa Cruz Biotechnology; and β-actin/catalase antibodies were obtained from Sigma.

Vectors.

Complementary DNA constructs encoding lamin A and the R482Q lamin A variant have been described previously (17). The heterozygous LMNA mutation leading to the R482Q substitution in the C-terminal domain of lamin A and lamin C causes Dunnigan-type familial partial lipodystrophy. We used R482Q constructs as a positive control, because in OA cartilage or in isolated OA chondrocytes, DAPI staining did not reveal any gross change in nuclear morphology, and overexpression of R482Q does not cause nucleoplasmic foci, in contrast to other variants. However, overexpression of other variants of lamin A causes strong nuclear morphologic changes.

Procurement of human cartilage.

Human cartilage was obtained from the knees of patients with a diagnosis of advanced OA (patients were ages ∼50–85 years, and 85% were female) who were undergoing knee replacement surgery, and from nonarthritic (control) knees (subjects were ages 50–88 years, and 50% were female) under the guidelines of the Institutional Review Board of New York University (NYU) School of Medicine for use of surgically discarded human tissues. Nonarthritic knee cartilage was obtained from the National Disease Research Interchange (Philadelphia, PA). Patients with OA had not received steroidal/nonsteroidal antiinflammatory drugs for at least 2 weeks before surgery. All specimens were examined by the authors and were confirmed to have gross evidence of OA (i.e., thinning of cartilage, focal eburnation and erosion, and reduced proteoglycan content as indicated by Safranin O staining). Pathologists at the NYU Hospital for Joint Diseases confirmed that all specimens had histologic evidence of OA.

RNA isolation from OA cartilage.

Cartilage was milled into fine powder using a 6800 Freezer/Mill (CE), and total RNA was isolated as described previously (18).

Cells and cell culture.

The SW-1353 cell line (obtained from ATCC) was isolated from a primary grade II chondrosarcoma of the right humerus of a 72-year-old white female. SW-1353 cells were cultured in 10 ml of Dulbecco's modified Eagle's medium (DMEM) containing 10% FBS, 100 units/ml penicillin, and 100 mg/ml streptomycin at 37°C in a humidified atmosphere of 5% CO2 in air. The cells were subcultured at split ratios of 1:3 to 1:4, using trypsin–EDTA. The medium was changed every 3 days.

OA chondrocyte monolayer cultures.

Briefly, OA cartilage slices were minced finely and digested with collagenase for 12–16 hours in Ham's F-12 medium with 5% FBS, as described previously (19). The cell suspension was used to establish cultures in T175 flasks. Within 2–3 days of harvesting, primary chondrocytes were replated at 80% confluence in 100-mm tissue culture dishes or 24-well plates before being used in the experiments.

Interleukin-1β (IL-1β) and prostaglandin E2 (PGE2) treatment.

For all studies, the cultures were adapted to serum-free conditions by overnight incubation in Ham's F-12 medium containing 0.2% endotoxin-free human albumin. Cultures were then either left untreated or were incubated with IL-1β (10 ng/ml) or PGE2 (10 μM) for the remainder of the experiment. EP2 receptor antagonist (AH6809) and EP4 receptor antagonist (AH23848) were used at a 10-μM concentration to block PGE2 activity; this concentration was selected based on our previous studies (20).

Western blot analysis.

Proteins were extracted from monolayer chondrocytes using M-PER reagent (Pierce), and protein expression was estimated using a bicinchoninic acid reagent (Pierce). Total proteins (20–100 μg) were electrophoresed on sodium dodecyl sulfate–polyacrylamide gels, using a Bio-Rad electrophoresis system. Proteins were transferred to nitrocellulose blots for 1 hour at 100V and probed with a mouse anti–lamin A monoclonal antibody (Abcam) and α-tubulin/β-actin/catalase monoclonal antibodies to correct for variations in sample loading. Blots were developed with an enhanced chemiluminescence detection kit, and the results were quantitated by scanning and determining pixel counts using ImageJ 1.43 software (National Institutes of Health). Raw intensity data for lamin A were normalized to the internal protein control catalase/β-actin and plotted as bar graphs.

Transfection of human chondrocytes.

The cells were grown in DMEM complete medium (10% FBS) in 24-well plates (200,000 cells/well) for 4–6 hours and then transferred to Opti-MEM medium 1–3 hours before transfection. Transfection with TransIT-LT1 reagent was carried out according to the manufacturer's protocol (Mirus Bio).

Proliferation assay.

A bromodeoxyuridine (BrdU) cell proliferation kit (EMD Biosciences) was used to compare proliferation between control chondrocytes and chondrocytes overexpressing lamin A. Twenty-four or 48 hours after transfection, the cells were incubated at 37°C overnight with BrdU, according to the manufacturer's protocol. The cells were then washed with growth medium and fixed with formalin, and BrdU incorporation was measured spectrophotometrically at 450–540 nm. The extent of proliferation was expressed as the percentage of BrdU uptake compared with control.

Measurement of ATP levels.

The levels of intracellular ATP were determined using a Bioluminescent Somatic Cell Assay Kit (Sigma-Aldrich) according to the manufacturer's protocol and as reported previously (21). The amount of ATP released was expressed as nmoles per 106 cells.

Active caspase 3 assay.

Apoptosis of cells as a result of lamin A overexpression was determined quantitatively by measuring active caspase 3 in cell lysates, using an ELISA kit (R&D Systems) according to the manufacturer's protocol. Active caspase 3 was measured, with results expressed as nanograms per milligrams of total protein. Caspase 3–specific inhibitor (Ac-DEVD-CHO) and caspase-negative inhibitor (Z-FA-FMK) were obtained from BD Biosciences.

Mitochondrial staining.

Changes in the mitochondrial membrane potential were assessed using the fluorescent dye JC-1 (Molecular Probes), as described previously (21). Cells were visualized under an Olympus IX71 fluorescence microscope, and staining results were analyzed using IPLab software.

Measurement of cytochrome c levels.

Cells from monolayer cultures were harvested 24 hours after treatment, and the mitochondrial and cytosolic fractions of the cells were isolated using a Mitochondrial Fractionation Kit (Active Motif) according to the manufacturer's instructions. Levels of cytochrome c from the purified fractions were determined by ELISA (Active Motif).

TUNEL staining.

TUNEL staining was performed to evaluate apoptosis. SW-1353 cells were seeded at a density of 5 × 105/well into 6-well microplates. The following day, cells were transfected with lamin A and control vector (pCDNA4), using TransIT-LT1 reagent (Mirus Bio) as described above. The cells were fixed with 4% paraformaldehyde and processed for the TUNEL assay, using an in situ cell death detection kit (Click-iT) as recommended by the manufacturer (Invitrogen).

Immunofluorescence microscopy.

Cells were washed twice with phosphate buffered saline (PBS) and fixed with 500 μl of IC Fixation Buffer containing paraformaldehyde (eBioscience) for 15 minutes at room temperature. After fixation, the cells were serially diluted with PBS and washed 6 times. Cells were permeabilized using 0.25% Triton X-100 in PBS for 5 minutes at room temperature. After permeabilization, the cells were washed twice with 500 μl of PBS and incubated in blocking solution (10% BSA in PBS) for 30 minutes at 37°C. The cells were then incubated with primary antibody (either anti–FLAG M2 monoclonal antibody or anti–lamin A antibody; Abcam) at dilutions of 1:500 and 1:100, respectively, for 1 hour at 37°C. The cells were then washed 3 times with PBS (5 minutes each) and incubated further with secondary antibody (fluorescein isothiocyanate–labeled anti-mouse; Sigma-Aldrich) at a 1:100 dilution for 1 hour at 37°C. After secondary antibody incubation, the cells were washed 3 times with PBS and then viewed using an Olympus IX71 fluorescence microscope.

Labeling and hybridization of microarrays.

Five micrograms of total RNA was used for double-stranded complementary DNA (ds-cDNA) synthesis (SuperScript Choice System; Gibco BRL). Purified ds-cDNA was used for synthesis of biotin-labeled complementary RNA (cRNA), using an Enzo BioArray HighYield RNA Transcript Labeling Kit (Affymetrix). The cRNA was purified using a Qiagen RNeasy kit and was fragmented at 95°C for 35 minutes for target preparation. The fragmented cDNA (target) was hybridized against both human U95Av2 and human U133A arrays, as suggested by the manufacturer (Affymetrix), and expression was normalized as described previously (18).

Real-time quantitative polymerase chain reaction (qPCR).

Total RNA (1 μg) was primed using oligo(dT)18 primers, and cDNA was synthesized using a Clontech cDNA Synthesis Kit according to the manufacturer's directions. Predesigned TaqMan primer sets were purchased from Applied Biosystems. Real-time PCRs were run on an ABI Prism 7300 sequence detection system (Applied Biosystems), and relative expression levels were calculated using the comparative threshold (Ct) method (22). Briefly, according to the comparative Ct method (Ct value = the number of PCR cycles that elapse before the threshold is reached for the target nucleic acid), arithmetic formulas are used to calculate relative expression levels compared with normal (nondiseased) samples. The amount of target, normalized to an endogenous housekeeping gene (GAPDH) and relative to the non-OA control samples, is then calculated using the following formula: 2math image, where ΔΔCt = ΔCt (individual patient) − ΔCt (mean of the non-OA control), ΔCt (individual patient) = Ct of the target gene − Ct of the housekeeping gene, and ΔCt (mean of the non-OA control) = mean Ct of the target gene in non-OA control patients − mean Ct of the housekeeping gene in non-OA controls. The equation thus represents normalized expression of the target gene in the individual patient relative to normalized expression of the mean of the values for non-OA controls.

Statistical analysis.

All of the experiments were performed 3–6 times, and each culture condition was tested in triplicate. Results are expressed as the mean ± SD. Unpaired 2-tailed t-tests were performed to analyze significance, using GraphPad Prism software version 4.0.

RESULTS

Expression of lamins in cartilage.

We investigated the relative levels of lamin messenger RNA in normal and OA cartilage, using gene expression profiling. Comparison of RNA from pooled samples of OA and normal cartilage using U95Av2 and U133A microarrays revealed a significant ∼2-fold up-regulation of lamin A mRNA in diseased tissue (P = 0.009 and P = 0.0005, respectively) (Figure 1A). The expression levels of lamin B1 and lamin B2 mRNA in OA samples were equivalent to those of non-OA controls (P = 0.8 and P = 0.1, respectively) (data not shown). We further validated the elevated expression of lamin A mRNA in OA samples compared with non-OA samples (P = 0.03), using qPCR (Figure 1B). Lamin A up-regulation in OA cartilage was also confirmed at the protein level (Figure 1C). Immunoblotting of cartilage protein extracts using a lamin A–specific monoclonal antibody revealed a protein with an apparent molecular mass of 70–72 kd. The mean lamin A levels in 5 OA cartilage samples compared with 4 non-OA cartilage samples were determined by ImageJ analyses (with normalization to catalase) (Figure 1C). In all OA samples, the expression of lamin A was elevated compared with that in non-OA controls (P = 0.0012). Immunohistochemical staining of sections of OA cartilage revealed uniform staining of lamin A throughout the cartilage, from the articular surface to the deep zone. Staining was also observed in both lesional and nonlesional areas (results not shown).

Figure 1.

Elevated lamin A expression in osteoarthritis (OA) cartilage. A, Gene expression (microarray) analysis of lamins. Pooled RNA samples from non-OA cartilage specimens (n = 2 pools) and OA cartilage specimens (n = 5 pools) were used for microarray analysis, using U95Av2 and U133A microarray chips; each pool represents RNA material from at least 5–10 individual subjects. After normalization, lamin A expression was expressed as arbitrary units. B, Quantitative polymerase chain reaction analysis of 6 non-OA samples and 8 OA samples, represented as the relative fold change of lamin A expression. C, Lamin A expression in 5 OA cartilage samples and 4 non-OA (N) cartilage samples, as determined using ImageJ version 1.43 software. Protein extracts from non-OA and OA cartilage were analyzed by Western blotting, using lamin A–specific antibody and catalase antibody (internal control). Bars show the mean ± SD.

Regulation of lamin A expression.

There have been no reports of factors that induce lamin A expression in mammalian cells. In order to identify potential inducers of lamin A expression in human articular chondrocytes, we studied the effect of the inflammatory mediators PGE2 and IL-1β. OA chondrocytes cultured in the presence of PGE2 (10 μM) for 24 hours stained more strongly than unstimulated cells following immunofluorescence labeling of lamin A. Conversely, treatment with IL-1β (10 ng/ml) for 24 hours decreased lamin A expression (Figure 2A). We confirmed the increased expression of lamin A in PGE2- and IL-1–treated cells, using immunoblot analysis. Preincubation of IL-1–treated chondrocytes with the cyclooxygenase 2 selective inhibitor celecoxib had an additive inhibitory effect on lamin A protein expression (Figure 2B). To further elucidate the role of endogenous PGE2 in the regulation of lamin A expression by chondrocytes, primary cells were cultured in the presence of celecoxib. Chondrocytes treated with celecoxib expressed less lamin A compared with untreated cells (Figures 2B and C). In contrast, lamin B1 levels were not affected by the addition of PGE2, IL-1β, or celecoxib (Figures 2B and C).

Figure 2.

Increased lamin A expression in human osteoarthritis (OA) chondrocytes following prostaglandin E2 (PGE2) stimulation. Human OA chondrocyte monolayer cultures were incubated in serum-free Ham's F-12 medium with or without PGE2/interleukin-1β (IL-1β) treatment. A, OA chondrocytes were treated with PGE2 or IL-1β, and lamin A levels were assessed by immunofluorescence or Western blot analysis. Cell nuclei were counterstained with DAPI. B, OA chondrocytes were treated with IL-1β for 24 hours in the presence or absence of celecoxib (Cele; 2 μM), and lamin A and lamin B1 levels were assessed by Western blot analysis. C, OA chondrocytes were incubated with EP2 receptor antagonist (ant.) or EP4 receptor antagonist (10 μM) for 2 hours and then stimulated with PGE2 (10 μM) for 24 hours. The expression of lamin A and lamin B1 in total cell lysates was determined by Western blot analysis. Bars show the mean ± SD.

To further investigate the mechanism of PGE2-dependent induction of lamin A, we stimulated chondrocytes with PGE2 following blockade of the PGE2 receptors EP2 and EP4. Blocking the EP2/EP4 receptors with receptor-specific antagonists (AH6809 and AH23848) neutralized the effect of PGE2; this was accompanied by down-regulation of lamin A at the protein level, as detected by immunoblotting (Figure 2C). These findings suggest that PGE2 significantly increased lamin A expression through the EP2/EP4 receptors. However, blocking of the EP2/EP4 receptors in unstimulated conditions did not affect lamin A or lamin B1 expression (data not shown).

Effect of lamin A overexpression on chondrocyte proliferation and viability.

To elucidate the role of lamin A in cellular aging, we used a BrdU cell proliferation kit to compare proliferation between control chondrocytes and chondrocytes overexpressing lamin A. Chondrocytes were transfected with wild-type lamin A and its variant, R482Q, followed by incubation with BrdU 24 hours or 48 hours after transfection. Transient overexpression of lamin A in these cells increased mRNA expression by 6–10-fold compared with vector-transfected cells (Figure 3A). The increased expression of lamin A observed in these experiments was confirmed by immunoblotting (Figure 3A).

Figure 3.

Overexpression of lamin A induces expression of senescence markers p21 and p16 in chondrocytes. Chondrocytes were transfected with lamin A constructs (wild-type and variant R482Q) and analyzed 24 hours after transfection. A, Relative lamin A expression as measured by quantitative polymerase chain reaction and Western blot analysis, with vector used as a negative control. Bars show the mean. B, Western blot analysis of transfected chondrocytes probed with anti-p21WAF1 antibody, anti-p16INK4A antibody, and anti-catalase antibody as an internal loading control. GFP = green fluorescent protein.

We observed a significant decrease in the proliferation of cells overexpressing lamin A (by 52% at 24 hours and 47% at 48 hours, as compared with vector-transfected cells; P < 0.001) (data not shown). Overexpression of the lamin A R482Q variant had a similar inhibitory effect on the proliferation of chondrocytes (by 81% at 24 hours and 52% at 48 hours; P < 0.001), suggesting that both wild-type lamin A and the R482Q variant had similar effects on chondrocyte proliferation. Furthermore, overexpression of lamin A also decreased cell viability, as measured by decreased total lactate dehydrogenase activity, from a mean ± SD of 14.05 ± 1.9 to 7.35 ± 0.41 units/mg protein (P < 0.02), as compared with vector-transfected cells.

Increased expression of senescence markers.

To determine whether decreased cellular proliferation was accompanied by increased cellular senescence following lamin A overexpression, we measured the expression of regulatory proteins p21WAF1 and p16 INK4A. In senescent cells, expression of these inhibitors was induced. As shown in Figure 3B, cells transiently overexpressing lamin A and the R482Q variant showed increased expression of p21, while cells transfected with vector alone expressed barely detectable amounts of p21. Overexpression of lamin A resulted in a significant increase in protein expression of p21 in wild-type and lamin A mutants, by 2–3-fold. However, unlike p21, p16 levels were observed to decrease following lamin A overexpression in chondrocytes.

Effect of lamin A overexpression on cytosolic cytochrome c accumulation.

Mitochondrial depolarization leads to redistribution of cytochrome c within the cytosolic and mitochondrial cell fractions. We investigated the role of lamin A in this process by measuring cytochrome c levels in the cytosol and mitochondrial cell fractions. Transient expression of lamin A in chondrocytes (Figure 4A) increased the levels of cytochrome c in the cytosol fraction (from a mean ± SD of 3.76 ± 0.4 to 6.04 ± 1.4 ng/mg protein; P < 0.5), and this was accompanied by a concomitant decrease in the mitochondrial fraction (from 14.4 ± 7.9 to 10.6 ± 5.9 ng/mg protein). IL-1β, a known inducer of apoptosis in cultured chondrocytes that was included as a positive control, also caused a redistribution of cytochrome c levels. In the same experiment, we also assessed ATP levels and observed that overexpression of lamin A in chondrocytes resulted in a 40–50% decrease in cellular ATP levels, 24 hours posttransfection (for vector control, mean ± SD 11.3 ± 0.89 pmoles/106 cells; for lamin A, 5.99 ± 0.41 [P < 0.02]). This redistribution of cytochrome c indicated that lamin A overexpression leads to the breakdown of mitochondrial membrane integrity and reduced cellular energy levels.

Figure 4.

Dysregulated mitochondrial function in wild-type lamin A–overexpressing chondrocytes. Human chondrocytes (SW-1353) were transfected with lamin A or vector control, and mitochondrial functions and apoptosis (as assessed by TUNEL staining) were determined 24–48 hours after transfection. A, Estimated cytochrome c levels in mitochondrial and cytosolic cell fractions. Values represent the relative amount of cytochrome c in mitochondrial and cytosolic cell fractions. The interleukin-1β (IL-1β)–induced increase in cytochrome c in cytoplasm was used as a positive control. B, Active caspase 3 levels in chondrocytes transfected with lamin A in the presence or absence of caspase-specific peptide inhibitor (Ac-DEVD-CHO) or negative control (Z-FA-FMK). Values in A and B are the mean ± SD of 3 independent experiments. C, Representative images showing TUNEL-positive cells in human chondrocytes transfected with lamin A. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1529-0131.

Effect of lamin A overexpression on caspase 3 activation.

We evaluated whether overexpression of wild-type lamin A in chondrocytes can lead to increased levels of caspase 3, a known mediator of chondrocyte apoptosis (Figure 4B). Overexpression of lamin A increased caspase 3 levels by 2–3-fold (mean ± SD 2.7 ± 0.7 μg/mg protein) over basal levels (0.5 ± 0.09 μg/mg protein; P < 0.01) within 24 hours. This effect was inhibited by preincubation with Ac-DEVD-CHO, a caspase 3–specific inhibitor. In the same experiments, the compound Z-FA-FMK (a peptide that has no effect on caspases) did not inhibit caspase 3 activation and had no effect on lamin A–induced expression of caspase 3. Lamin A overexpression did not change the nuclear morphology in human chondrocytes.

Effect of overexpression of lamin A on apoptosis.

In parallel studies, we confirmed that overexpression of lamin A induced DNA fragmentation, as an additional marker of apoptosis. TUNEL assays detected apoptosis-mediated DNA fragmentation. Overexpression of lamin A induced significant DNA fragmentation at 48 hours posttransfection in comparison with control vector–transfected chondrocytes, as indicated by the green fluorescence in the TUNEL assay (Figure 4C). As expected, IL-1β (10 ng/ml) treatment promoted apoptosis, as indicated by positive TUNEL staining, in cultured chondrocytes.

Effect of protein farnesyl transferase inhibitor (FTI) on lamin A–mediated cellular defects.

Because overexpression of lamin A slowed the growth of chondrocytes, we evaluated the effect of FTI on the cellular defects observed in cells expressing wild-type lamin A. Culturing transfected chondrocytes overexpressing lamin A with FTI prevented lamin A–induced mitochondrial membrane depolarization, as indicated by increased red fluorescence of healthy mitochondria following staining with JC-1 dye (Figure 5A). This was accompanied by a moderate (7%) decrease in the leakage of cytochrome c from the mitochondria to the cytoplasm compared with that observed in untreated lamin A–overexpressing cells (data not shown).

Figure 5.

Dysregulated mitochondrial function induced in response to lamin A overexpression is mitigated by farnesyl transferase inhibitor (FTI) pretreatment of chondrocytes. Chondrocytes (SW-1353) were preincubated with FTI (10 μM) 24 hours prior to lamin A transfection. The cells were collected and analyzed for various mitochondrial functions. A, Representative photomicrographs showing mitochondrial depolarization as analyzed by JC-1 dye incorporation. Red color due to aggregation of JC-1 monomer indicates normal healthy mitochondria. Lamin A overexpression led to the accumulation of JC-1 monomer (green color), indicating dysfunctional/depolarized mitochondria. FTI treatment restored mitochondrial polarization. B, Active caspase 3 expression, as determined by enzyme-linked immunosorbent assay. C, Total cellular ATP production, as determined using chemiluminescence methods. Values in B and C are the mean ± SD of 3 similar experiments. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1529-0131.

Similarly, the addition of FTI to cells overexpressing lamin A decreased active caspase 3 levels (from a mean ± SD of 2,048.0 ± 732 to 349 ± 273 ng/mg protein) (Figure 5B) and increased cellular ATP levels to values similar to control (from 6.69 ± 0.55 to 9.32 ± 0.29 nmoles/106 cells; P = 0.02) (Figure 5C). These data suggest that the farnesylation step during lamin synthesis is in part necessary for lamin A–mediated induction of chondrocyte apoptosis.

DISCUSSION

Chondrocyte viability decreases in the setting of cartilage injury, aging, and disease (3, 6, 8). According to several reports, chondrocytes in OA cartilage exhibited classic signs of apoptosis (6, 8). Because chondrocytes are essential for maintaining the integrity of cartilage extracellular matrix, thus enabling normal joint function, identifying the cellular mechanisms that control cell survival could be an important step in developing treatments to prevent cartilage loss.

The results of the current study are the first to demonstrate that lamin A is up-regulated in OA chondrocytes, where it inhibits mitochondrial function, lowers cellular ATP levels, promotes senescence, and induces apoptosis. Lamin A is induced by PGE2, a mediator produced in significant amounts by OA chondrocytes (23). Furthermore, our results highlight a potential role for lamin A in premature aging and apoptosis/senescence of chondrocytes associated with OA. Overexpression of lamin A led to apoptosis, as evidenced by mitochondrial depolarization, caspase activation, decreased cellular ATP expression, and increased cytosolic cytochrome c expression. These effects were partially reversed following FTI treatment, suggesting that farneyslation of lamin A is necessary for induction of apoptosis. Recent studies have also shown that treatment with FTIs can reverse the nuclear morphology defects in cells expressing the truncated lamin A variant in Hutchinson-Gilford progeria syndrome (24–26).

We previously showed that increased production of PGE2 is associated with chondrocyte apoptosis (21). Our current study provides strong supporting evidence that the addition of PGE2 leads to increased lamin A expression, subsequently leading to a change in mitochondrial function and cell death. PGE2-induced expression of lamin A was confirmed at the gene expression level (using qPCR), by immunostaining (using confocal microscopy), and by immunoblot analysis. PGE2 mediates catabolic effects in OA chondrocytes via the EP4 receptor (20); the current study supports this observation, because we demonstrated that PGE2, via the EP2/EP4 receptor, increased lamin A expression but not lamin B1 expression. In contrast, IL-1 was shown to inhibit, rather than stimulate, lamin A expression in chondrocytes. This result was somewhat unexpected, because we previously reported that IL-1β is known to stimulate PGE2 in chondrocytes and promote apoptosis (21). These results suggest that lamin A expression is suppressed by downstream components of IL-1β signaling, independently of PGE2; however, the extent to which lamin A is required for IL-1β–induced apoptosis was not examined in the current study.

The decreased level of lamin A in IL-1–stimulated chondrocytes may occur due to activation of multiple pathways. For example, IL-1–induced mitochondrial dysfunction may lead to NLRP3 inflammasome–mediated caspase activation, which leads to lamin degradation (27); increased reactive oxygen or nitrogen species generated by IL-1 treatment activate caspases (28); lamins are substrate for IL-1–induced caspase activity; and IL-1 has also been shown to induce carboxymethylation of lamin and alter subnuclear distribution and degradation (29).

Mitochondria are involved in many cellular processes, and mitochondrial dysfunction has been associated with apoptosis, aging, and various pathologic conditions, including OA (8, 30, 31). Mitochondrial dysfunction due to disruption of mitochondrial membrane potential may lead to senescence and apoptosis of cells via a cascade of signals, including changes in ATP levels, release of cytochrome c into the cytosol, and activation of caspase 3 (21). Lamin A overexpression in chondrocytes effected the activation of these classical pathways, culminating in cell death by apoptosis. Similar observations have been reported in fibroblasts isolated from patients with LMNA mutations, which demonstrated mitochondrial respiratory chain protein obliteration and changes in mitochondrial membrane potential (32).

Similarly, it has been shown that human skin fibroblasts isolated from patients with lipodystrophy syndrome and LMNA mutations have increased p16 and p21 expression, leading to senescence with increasing cellular passages (32). Increased expression of cyclin-dependent kinase inhibitors p21 and p16 has been associated with inhibition of cell cycle arrest or senescence in many cell types (33–35). Lamin A–overexpressing chondrocytes had elevated levels of p21 but not p16 at 24 hours posttransfection as compared with control or vector-transfected cells. It was further noted that p21 expression increased at the onset of senescence; this was followed by the late onset of expression of p16, resulting in the inhibition of retinoblastoma protein kinases, leading to cell arrest (33–35). Furthermore, Zhou et al (36) previously reported that knockdown of p16 by small interfering RNA contributed to the recovery of chondrocytes, with increased proliferation and overall increased repair capacity. In our studies, cells transiently overexpressing lamin A showed increased expression of p21 compared with cells transfected with vector alone, and this was accompanied by decreased proliferative capacity as determined by BrdU incorporation.

Thus, our data show that minor perturbations in the expression of lamin A led to cellular senescence, decreased cellular energy stores, and apoptosis. These results further demonstrate that lamin A accumulation/metabolism is not limited to its role in the development of accelerated aging observed in laminopathy syndromes, and also that overexpression of normal lamin A protein in disease may play a role in the premature aging of chondrocytes in OA. It is of further interest to note that senescence (cell cycle arrest) may also lead to a hypertrophic cell phenotype, decreased proliferation, a decreased response to growth factors, dysregulated gene expression, and aging of cartilage tissue (8). Because the molecular pathology of several laminopathies resembles an accelerated form of normal aging, it is intriguing to draw a parallel between an alteration in lamin A function, senescence, and apoptosis in chondrocytes.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Abramson had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Attur, Ben-Artzi, Abramson.

Acquisition of data. Attur, Ben-Artzi, Al-Mussawir.

Analysis and interpretation of data. Attur, Ben-Artzi, Yang, Al-Mussawir, Worman, Palmer, Abramson.

Acknowledgements

We thank Ann Rupel for assistance in preparing the manuscript and figures. We also wish to thank Ms Jyoti Patel for processing, storing, and maintaining the primary chondrocytes. Finally, we thank Mandar Dave for performing ATP, caspase, and mitochondrial staining assays.

Ancillary