To determine whether statins induce apoptosis in rheumatoid arthritis (RA) synoviocytes.
To determine whether statins induce apoptosis in rheumatoid arthritis (RA) synoviocytes.
The effects of lipophilic and hydrophilic statins (fluvastatin and pravastatin, respectively) on the apoptosis of cultured RA synoviocytes were examined in vitro. Apoptosis was analyzed by flow cytometry after staining with JC-1 (to measure the mitochondrial transmembrane potential), active caspase 3, annexin V, and propidium iodide. Add-back experiments were conducted to determine which downstream products of the mevalonate pathway could suppress apoptosis. Modulation of various signaling pathways induced by statins, including protein prenylation, was also investigated.
Fluvastatin, but not pravastatin, induced apoptosis in RA synoviocytes in a concentration-dependent (1–10 μM) and time-dependent (48–96 hours) manner. Another lipophilic statin, pitavastatin, displayed almost the same effects as fluvastatin. In sharp contrast, lipophilic statins did not significantly increase apoptosis in synoviocytes from patients with osteoarthropathy. Apoptosis induced by fluvastatin was mitochondrial- and caspase 3–dependent and was abrogated by mevalonate and geranylgeranyl pyrophosphate, but not by farnesyl pyrophosphate. In addition, the geranylgeranyl transferase inhibitor GGTI-298 mimicked the effect of fluvastatin on RA synoviocytes. Treatment of RA synoviocytes with the RhoA kinase inhibitor Y-27632 caused apoptosis. Fluvastatin decreased the amount of RhoA protein in the membrane fraction, but increased the amount in the cytosolic fraction.
Fluvastatin induced apoptosis in RA synoviocytes through a mitochondrial- and caspase 3–dependent pathway and by the blockage of mevalonate pathways, particularly through the inhibition of protein geranylgeranylation and RhoA/RhoA kinase pathways. These findings suggest that lipophilic statins have potential as novel therapeutic agents for RA.
Statins, inhibitors of 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase, are effective lipid-lowering agents and are used extensively in medical practice. Recent experimental and clinical evidence indicates that statins gain some of their cholesterol-independent, or so-called pleiotropic, effects by improving endothelial function, decreasing oxidative stress and inflammation, and inhibiting the thrombogenic response in the vascular wall (1, 2). Many of these cholesterol-independent effects reflect the ability of statins to block the synthesis of important isoprenoid intermediates, such as farnesyl pyrophosphate (FPP) and geranylgeranyl pyrophosphate (GGPP), which serve as lipid attachments for the posttranslational modification of a variety of intracellular signaling molecules. In particular, the inhibition of the small GTP-binding proteins, including Rho, Ras, and Rac, the proper membrane localization and function of which are dependent upon isoprenylation, may play an important role in mediating the biologic effects of statins (1, 2).
In addition, recent observations suggest that statins have pleiotropic immunomodulatory properties (3). It was first reported in 1995 (4) that treatment of cardiac transplant recipients with pravastatin was associated with a reduction in rejection episodes and an increase in survival rates, independently of its cholesterol-lowering effects, although the precise mechanism remained to be determined. Subsequently, 3 important experimental studies demonstrated that certain statins are potentially involved in the immune response. First, statins were found to repress interferon-γ–induced class II major histocompatibility complex molecules by means of the class II transactivator in a variety of cells, including endothelial cells and monocyte/macrophages, thus inhibiting T cell activation (5). This effect was reversed by mevalonic acid, which directly implicates HMG-CoA reductase in this activity. Second, lovastatin and simvastatin were shown to inhibit the interactions between leukocyte function–associated antigen 1 (LFA-1) and intercellular adhesion molecule 1 by binding to a specific recognition site on LFA-1, independently of HMG-CoA reductase activity (6). Functionally, this mechanism suppresses critical T cell–costimulatory events. Such cell contact–dependent interactions between T cells and macrophages are now recognized to be of fundamental importance in promoting chronic inflammation in several autoimmune diseases (7). Third, atorvastatin was shown to promote a Th2 bias and to reverse chronic and relapsing experimental autoimmune encephalomyelitis, a CD4+, Th1-mediated central nervous system demyelinating disease model of multiple sclerosis (8).
These important data add to the growing evidence in support of the notion that statins may possess immunomodulatory properties that are clinically useful in treating a variety of autoimmune diseases, particularly Th1-mediated diseases such as rheumatoid arthritis (RA) (9). Indeed, a pilot clinical study examining the efficacy of simvastatin in RA demonstrated that the statin can suppress the clinical symptoms of RA to some extent (10, 11).
We have previously shown that apoptosis-inducing agents such as FTY720 may be beneficial in the treatment of autoimmune diseases (12). Statins are also potent inducers of apoptosis in a wide range of normal and tumor-derived cell lines (13–16). These apoptotic effects can be primarily related to the inhibition of HMG-CoA reductase, since intermediates of mevalonate pathways can reverse such an effect (13–16).
RA is characterized by a proliferative disorder of synovial tissue associated with Th1-predominant immune dysregulation (17). At present, little is known about whether statins have direct effects on synoviocytes. The aim of the present study was to investigate whether statins are able to induce apoptosis in RA synoviocytes and to determine the mechanism underlying the statin-induced apoptosis. We address these issues herein and show that fluvastatin, a lipophilic statin, induces a significant apoptotic response in RA synoviocytes through a mitochondrial- and caspase 3–dependent pathway and by the blockage of mevalonate pathways, particularly through the inhibition of protein geranylgeranylation and RhoA/RhoA kinase pathways. These findings provide an additional scientific rationale for using statins as a novel therapeutic approach in the treatment of RA.
Human synoviocytes derived from patients with RA or osteoarthropathy were purchased from Toyobo (Tokyo, Japan). Synoviocytes were placed in a humidified chamber containing an atmosphere of 95% air and 5% CO2 and cultured in Dulbecco's modified Eagle's medium containing 10% (volume/volume) heat-inactivated fetal bovine serum and supplemented with antibiotics. Each passage was cultured for 1 week. Cells from passages 2–9 were used for the studies. The cells were cultured in 24-well flat-bottomed plates or in 10-cm dishes. The cells were counted and confirmed to be >95% viable using the erythrosin B exclusion test (18).
Fluvastatin (a lipophilic statin) and pravastatin (a hydrophilic statin) were kindly provided by Tanabe Pharmaceutical (Osaka, Japan) and Sankyo Pharmaceutical (Tokyo, Japan), respectively, and pitavastatin (a lipophilic statin) was kindly provided by Kowa Pharmaceutical (Tokyo, Japan). Mevalonate, GGPP, and FPP were obtained from Sigma (St. Louis, MO). For in vitro experiments, these reagents were dissolved in H2O and adjusted to appropriate concentrations for each experiment. The geranylgeranyl transferase inhibitor GGTI-298, the farnesyl transferase inhibitor FTI-277, and the RhoA kinase inhibitors HA-1077 and Y-27632 were purchased from Calbiochem (La Jolla, CA). These compounds were dissolved in DMSO, as suggested by the supplier.
Apoptosis was measured in synovial cells cultured in vitro for 0, 1, 2, 3, and 4 days. After stimulation, as indicated below, the cells were carefully detached with trypsin–EDTA. Apoptotic cell death was analyzed by 4 different flow cytometry methods (19).
First, DePsipher (JC-1, a carbocyanine dye; R&D Systems, Minneapolis, MN) staining was used to evaluate the mitochondrial transmembrane potential and was performed according to the manufacturer's protocol. DePsipher has the property of aggregating upon membrane polarization to form an orange-red fluorescent compound. If the mitochondrial potential is disturbed, the dye cannot access the transmembrane space and remains or reverts to its green monomeric form. The disruption of the mitochondrial membrane potential is characteristic of the earliest stage of apoptosis.
Second, active caspase 3 staining with a fluorescein isothiocyanate (FITC)–conjugated monoclonal antibody apoptosis kit I (BD PharMingen, San Diego, CA) was performed according to the manufacturer's protocol. Caspase 3 has been implicated as a key protease that is activated during the process of apoptosis.
Third, annexin V–FITC staining with an FITC-conjugated annexin V antibody (BD PharMingen) was performed. Annexin V identifies cells at an early stage of apoptosis by binding to the membrane phospholipid phosphatidylserine.
Fourth, staining with propidium iodide (PI; Sigma), a DNA-intercalating dye, was performed after permeabilization with ethanol. Briefly, cells were washed twice in phosphate buffered saline (PBS) containing 0.1% (weight/volume) NaN3, and 1 ml of cold 70% (v/v) ethanol was added to the cell pellet. The cells were then vortexed and incubated at 4°C for 1 hour, washed twice, and resuspended in 0.1 ml of PBS containing 0.1% NaN3. For DNA staining, 0.1 ml of 1 mg/ml RNase A (Sigma) was added to cells while mixing, followed by the addition of 0.2 ml of 100 μg/ml of PI (Sigma). Cells were incubated at room temperature for 20 minutes in the dark and then analyzed by flow cytometry. Apoptotic cells appeared in the <2N DNA peak. Apoptotic cells were distinguished from necrotic cells by the light scatter profile. Necrotic cells showed a large decline in forward-angle light and side scatter, while apoptotic cells showed a smaller decline. Necrotic cell fragments and cell debris were excluded by an appropriate gate, and data were recorded on a logarithmic scale. Stained cells were analyzed by an LSR instrument (BD Biosciences, San Jose, CA) using the CellQuest software. At least 10,000 cells were evaluated per sample in most experiments, and each experiment was repeated at least 3 times.
Separation of cytosolic and membrane fractions was performed as follows. The cells were suspended in hypotonic buffer (20 mM HEPES–NaOH, pH 7.5, 2 mM MgCl2, 3 mM EDTA, 2 mM EGTA, 2 μg/ml aprotinin, and 2 μg/ml leupeptin) and disrupted 3 times by sonication for 5 seconds on ice. Cell debris was removed by centrifugation at 1,400g for 5 minutes, and the supernatant was subjected to ultracentrifugation at 100,000g for 30 minutes. The resulting supernatant was collected as the cytosolic fraction.
The pellet was dissolved in a hypotonic buffer, sonicated briefly, and centrifuged at 100,000g for 30 minutes at 4°C. Then, the pellet was suspended in lysis buffer (50 mM Tris HCl, pH 8.0, 150 mM NaCl, 1% Nonidet P40 [NP40], 1 mM phenylmethylsulfonyl fluoride [PMSF; Sigma]) and solubilized by three 5-second cycles of sonication on ice. The final supernatant was collected as the membrane fraction. The cytosolic fraction and the membrane fraction were subjected to immunoblotting.
Anti-human RhoA (26C4) and RhoA kinase (Rock-2) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Cells from different experimental conditions were lysed on ice for 30 minutes in 50 mM Tris HCl, pH 8.0, containing 150 mM NaCl, 1% NP40, 1 mM PMSF, 2 μg/ml aprotinin, and 2 μg/ml leupeptin. The cell lysates were centrifuged at 10,000g for 10 minutes, and the supernatant was collected as the whole cell lysate.
The protein concentration was measured with a BCA bicinchoninic acid protein assay reagent kit (Pierce, Rockford, IL). Equal amounts of proteins (20 μg) were loaded onto 12% acrylamide gels and electrophoresed under reducing conditions in a MiniGel system (Bio-Rad, Richmond, CA). The resolved proteins were transferred onto nitrocellulose membranes using a semidry transfer system (both from Bio-Rad). The membranes were stained with ponceau S to confirm equal loading and transfer of samples.
Transferred proteins were detected by using a Mouse ExtrAvidin alkaline phosphatase staining kit (Sigma) according to the manufacturer's instructions, with some modifications. Briefly, nonspecific sites of the membrane were blocked by overnight incubation at room temperature in 5% bovine serum albumin (w/v) in PBS. The membrane was incubated for 2 hours at room temperature with anti-RhoA antibody diluted 1:500 in PBS containing 0.05% Tween 20 (PBST), washed with PBST, and incubated for 1 hour at room temperature with diluted biotinylated goat anti-mouse Ig in Tris buffered saline (TBS), pH 7.4 (1:1,000 dilution). The membrane was washed with TBS, incubated for 1 hour at room temperature with diluted ExtrAvidin alkaline phosphatase in TBS (1:1,000 dilution), washed with TBS, and then incubated in BCIP and 4-nitro-blue tetrazolium chloride liquid substrates (Sigma).
Data are expressed as mean ± SD. Statistical analysis was performed with the StatView V system (Abacus Concepts, Berkeley, CA) for the Apple computer. One-way analysis of variance was applied for comparison among multiple groups. Differences among group means were considered significant at a P value of less than 0.05.
Previous studies showed that fluvastatin as well as other lipophilic statins induced apoptosis in various cell types (13–16). To determine whether fluvastatin had a similar effect on RA synoviocytes, apoptosis was measured in synoviocytes that had been cultured for 96 hours. In the absence of statins, apoptosis was seen in <10% of cultured synoviocytes, as determined by PI (Figure 1A) and annexin V–FITC (Figure 1D) staining. Since the concentration of statins that induced maximal apoptosis of various cell lines was reported to be 100 μM, the first experiment was performed at this concentration. Fluvastatin (Figures 1B and E), but not pravastatin (Figures 1C and F), induced a significant increase in the apoptosis of synoviocytes. In sharp contrast, fluvastatin showed little increase in apoptosis of synoviocytes from patients with osteoarthropathy (Figures 1G and H).
We then used annexin V staining to examine dose- and time-dependency of apoptosis. Synoviocytes were subjected to increasing concentrations of fluvastatin for 120 hours and then analyzed for apoptosis by flow cytometry (Figure 2A). The cells were sensitive to fluvastatin, exhibiting an apoptotic frequency of 11.8 ± 2.8% (mean ± SD) at 1 μM and reaching a plateau of 65.7 ± 15.4% at 10 μM. Since the lowest concentration of fluvastatin to induce maximal apoptosis was 10 μM, subsequent experiments were performed in this concentration. Kinetic studies verified that fluvastatin-induced apoptosis was time-dependent, with a linearly increasing apoptotic response ranging from 9.6 ± 1.0% after 24 hours up to 47.4 ± 19.6% after 96 hours (Figure 2B). These data indicated that fluvastatin induced apoptosis in RA synoviocytes in a dose- and time-dependent manner. Another lipophilic statin, pitavastatin, induced apoptosis in RA synoviocytes to the same extent as fluvastatin (data not shown).
Recently, the existence of at least 2 distinguishable cell death signaling pathways, the mitochondrial (intrinsic) pathway and the death receptor (extrinsic) pathway, was well established (20). We further examined the cell death signaling pathway for fluvastatin in synoviocytes. Fluvastatin disrupted the mitochondrial transmembrane potential (Figure 3A) and increased the number of active caspase 3–positive cells (Figure 3B), thereby suggesting that fluvastatin-induced cell death may be mediated, at least in part, through a mitochondrial- and caspase 3–dependent pathway.
To evaluate whether mevalonate metabolites can reverse the apoptotic effect of fluvastatin on synoviocytes, the cells were cultured with fluvastatin in the presence of mevalonate (100 μM), FPP (5 μM), or GGPP (5 μM) (add-back experiments). Apoptosis was determined by annexin V–FITC staining. Neither mevalonate metabolite induced apoptosis in synoviocytes.
FPP and GGPP are involved in the farnesylation and geranylgeranylation of proteins, respectively (21). Addition of mevalonate or GGPP almost completely reversed, and FPP partly reversed, the apoptotic effect of fluvastatin (Figure 4a). These results suggest that the apoptotic effect of fluvastatin may be independent of its lipid-lowering properties and may be mainly related to changes in protein geranylgeranylation.
To confirm that fluvastatin induced apoptosis by inhibiting protein geranylgeranylation, we next examined whether the geranylgeranyl transferase inhibitor GGTI-298 mimicked the effect of fluvastatin. Synoviocytes were treated with increasing concentrations of GGTI-298 or the farnesyl transferase inhibitor FTI-277, and apoptosis was measured by annexin V–FITC staining after 96 hours. GGTI-298 caused a significant increase in apoptosis at 30 μM and 100 μM concentrations (Figure 4b, parts D and E). In contrast, FTI-277 at the same concentrations showed very little increase in apoptosis compared with the control cells (Figure 4b, part C). These results also indicate that inhibition of protein geranylgeranylation predominantly induces apoptosis and may be involved in fluvastatin-induced apoptosis in RA synoviocytes.
Since RhoA is one of the geranylgeranylated proteins, the inhibition of protein geranylgeranylation should affect the RhoA pathway. We further examined whether inhibition of RhoA kinase, the immediate downstream effector of RhoA, caused apoptosis. To confirm this, we used 2 structurally different RhoA kinase inhibitors, Y-27632 and HA-1077. Synoviocytes were incubated in the presence or absence of inhibitors at a concentration of 100 μM for 96 hours. Treatment with Y-27632, but not HA-1077, significantly increased the apoptosis of synoviocytes to a much lower extent than did treatment with fluvastatin at a concentration of 10 μM (Figure 5). These data show that inhibition of RhoA kinase can be related to the process of apoptotic cell death in RA synoviocytes.
Since the small G protein Rho must be targeted to the plasma membrane for its activation, we next examined the effect of fluvastatin on the translocation of RhoA protein from the cytosolic fraction to the membrane fraction in synoviocytes. Treatment with fluvastatin decreased RhoA protein in the membrane fraction and increased RhoA protein in the cytosolic fraction in a time-dependent manner (24–72 hours) (Figure 6a). In addition, we analyzed the expression of RhoA kinase. Fluvastatin did not decrease the expression of RhoA kinase in whole cell lysates of RA synoviocytes (Figure 6b).
The major findings of this study were as follows. First, fluvastatin, a lipophilic statin, but not pravastatin, a hydrophilic statin, induced apoptosis in vitro in RA synoviocytes, but not in synoviocytes from patients with osteoarthropathy. Second, fluvastatin-induced apoptosis was mitochondrial- and caspase 3–dependent. Third, the apoptosis was reversed by mevalonate or GGPP. Fourth, the geranylgeranyl transferase inhibitor GGTI-298 and the RhoA kinase inhibitor Y-27632 increased apoptosis. Fifth, fluvastatin decreased the membrane fraction of RhoA in RA synoviocytes. Taken together, these findings suggest that fluvastatin induces apoptosis in RA synoviocytes through the inhibition of protein geranylgeranylation and subsequent suppression of small G proteins, particularly RhoA, and may play a crucial role in the pathogenesis of synovial cell proliferation in patients with RA. To our knowledge, this is the first study to demonstrate that statins can induce apoptosis in synovial cells from patients with RA, but not those from patients with osteoarthropathy.
Recent studies have shown that apoptosis of synovial cells plays an important role in the pathogenesis of RA (22, 23). The activation and proliferation of synovial cells is thought to be a key step in the destruction of cartilaginous and bony tissues in RA joints (24). RA synovitis may be exacerbated when apoptosis of synovial cells is insufficient or is resistant to treatment. Therefore, if apoptosis can be induced selectively in the RA synovium by an exogenous agent, it may prove to be a novel approach to the treatment of RA. Indeed, the efficacy of apoptosis-inducing therapy in experimental animal model of RA has been reported (25–27). More recently, Leung et al (28) demonstrated that simvastatin, a lipophilic statin, can effectively suppress murine collagen-induced arthritis via the specific suppression of pathogenic Th1 and proinflammatory responses. Moreover, Kanda et al (10) reported that simvastatin can suppress clinical symptoms in RA patients, although the mechanism of antirheumatic effects was not clarified. In the present study, we demonstrated that fluvastatin induced apoptosis in synoviocytes from patients with RA, but not in those from patients with osteoarthropathy, suggesting that the apoptotic effect of fluvastatin is a mechanism for suppression of inflammatory arthritis such as RA by statins.
In contrast to lipophilic statins, the hydrophilic statin pravastatin did not increase apoptosis in synovial cells from RA patients. Consistent with our findings, several groups of investigators (13, 29) have also reported that apoptosis of various cell types is induced by lipophilic statins, but not by pravastatin. Those studies revealed that pravastatin inhibited sterol synthesis in hepatocytes, with a potency equivalent to that of other statins. Thus, the reason for this apparent difference in the apoptotic potential between fluvastatin and pravastatin may be related to the hydrophilic nature of pravastatin and the lack of a pravastatin-specific carrier in nonhepatic cells. Consequently, it may be more difficult for pravastatin to penetrate the cell membrane of nonhepatic cells such as synovial cells.
We demonstrated that inhibition of protein geranylgeranylation is required for fluvastatin-induced apoptosis in synovial cells from patients with RA. Protein geranylgeranylation has been shown to be involved in the survival of vascular smooth muscle cells (13), osteoclasts (30), and cardiac myocytes (29). Our results are consistent with the findings of these studies. Furthermore, recent studies have also suggested that the pleiotropic effects of statins are linked closely to the inhibition of geranylgeranylation of small G proteins (1). Among the geranylgeranylated proteins, Rho may be a candidate molecule because Rho/Rho kinase has been shown to be involved in the process of apoptosis (15). In addition, inhibition of Rho activation has been reported to trigger apoptosis in fibroblasts (31) and T lymphoblasts (32). RhoA kinase is an immediate effector of RhoA, and it has been implicated in most of the identified functions of RhoA (33).
A new synthetic compound, Y-27632, has been widely used as a specific inhibitor of RhoA kinase for identifying the roles of the RhoA pathway in a variety of systems (34–36). In this study, we used Y-27632 to determine the role of RhoA kinase in apoptotic cell death. Treatment with a high concentration of Y-27632 (100 μM) significantly induced apoptosis, but to a lesser extent than fluvastatin, whereas the same concentration of another RhoA kinase inhibitor, HA-1077, which is structurally unrelated to Y-27632, showed little increase in apoptosis in synovial cells from patients with RA, as determined by PI staining. However, synoviocytes treated with HA-1077 also showed morphologic changes, including cell detachment, pseudopodia, and rounding up, suggesting that the RhoA/RhoA kinase pathway is a survival pathway of RA synovial cells. We further demonstrated that treatment with fluvastatin decreased the amount of RhoA protein in the membrane fraction. Consistent with these findings, Ogata et al (29) reported that the translocation of RhoA protein from the membrane fraction to the cytosolic fraction in cardiac myocytes was also seen after fluvastatin treatment.
Our results suggest that RhoA is involved in the regulation of apoptosis in RA synovial cells. We cannot rule out the possibility that, in addition to RhoA, other geranylgeranylated proteins are involved, because it is estimated that 0.5–1% of all cellular proteins are geranylgeranylated (37). Furthermore, we showed that the expression of Rock-2 was not reduced in fluvastatin-treated RA synovial cells. It is also possible that RhoA-interacting kinases other than Rock-2 act as effectors in RA synovial cells. Further investigations are therefore required for an understanding of the precise signaling pathways of fluvastatin-induced apoptosis and for application of our data to the pathogenesis and treatment of RA. Nevertheless, our results indicate that RhoA/RhoA kinase may be one of the critical signaling pathways in fluvastatin-triggered apoptosis in RA synovial cells.
The value and limitations in this study must be clearly pointed out. The concentrations used to demonstrate the apoptotic effects of statins in the in vitro cell cultures, especially with regard to inhibition of Rho geranylgeranylation, were ∼10 times higher than clinically prescribed concentrations (3). However, it is possible that the concentrations of statins needed to affect proliferating cells are lower than those needed to affect normal resting cells. It is also possible that in vivo, relatively low, but sustained, blood levels of statins would exert an effect similar to that seen in vitro with higher concentrations and short incubation times. Although fluvastatin induced apoptosis in RA synovial cells in vitro, we found no direct evidence of fluvastatin-induced apoptosis in RA synovium in vivo. The contribution of fluvastatin-induced apoptosis on the therapeutic effect in vivo remains to be determined.
In conclusion, we have shown that fluvastatin induces apoptosis in RA synovial cells via the inhibition of protein geranylgeranylation and the subsequent inhibition of the RhoA/RhoA kinase pathway. The induction of apoptosis in RA synovial cells by fluvastatin and the biologic antiatherosclerotic properties of the statins suggest that they may turn out to be ideal therapeutic agents in RA. Based on these results, we propose that the statins warrant clinical trials as potential modifiers of RA.
The authors thank Tanabe Pharmaceutical Company (Osaka, Japan), Sankyo Pharmaceutical Company (Tokyo, Japan), and Kowa Pharmaceutical Company (Tokyo, Japan) for donating the fluvastatin, pravastatin, and pitavastatin, respectively.