- Email: [email protected]
Apoptosis induced by nitric oxide is associated with nuclear p53 protein expression in cultured osteoarthritic synoviocytes
Osteoarthritis and Cartilage (1999) 7, 203–213 1999 OsteoArthritis Research Society International Article No. joca.1998.0209, available online at http://www.idealibrary.com on
1063–4584/99/020203+11 $12.00/0
Apoptosis induced by nitric oxide is associated with nuclear p53 protein expression in cultured osteoarthritic synoviocytes
BY DIDIER BORDERIE*, PASCAL HILLIQUIN†, ALAIN HERNVANN*‡, HERVEu LEMARECHAL*, CHARLES J. MENKES† AND OHVANESSE G. EKINDJIAN*‡ *Laboratoire de Biochimie A, †Institut de Rhumatologie, Hoˆpital Cochin, Universite´ Rene´ Descartes, 27 rue du fg St Jacques, 75014 Paris, France ‡Biochimie cellulaire de l’inflammation, Universite´ Paris XI, rue J.B. Clement, Chatenay-Malabry, France Summary Objective. Nitric oxide (NO) is a free radical molecule endogenously produced by NO synthases that may play a critical role in inflammation. It inhibits cell proliferation and may be involved in the induction of apoptosis in various cellular models. Recently, the presence of apoptotic cells was reported in the synovium of osteoarthritic (OA) patients. The aim of this study was to determine whether synovial fibroblasts are target cells for NO-induced apoptosis. The expression of p53 protein was also studied to evaluate the ability of synovial cells to repair DNA fragmentation. Methods. Synoviocytes from OA patients were treated with two NO donors: sodium nitroprusside (SNP) and S-nitroso-N-acetyl-penicillamine (SNAP). Apoptosis was analysed by transmission electron microscopy. DNA content was evaluated by flow cytometric analysis after propidium iodide staining and recognition of DNA strand break determined by the TUNEL (TdT-mediated dUTP nick end labeling) method. P53 protein expression was studied by immunofluorescence using a monoclonal antibody. Results. After 6 hours, cells treated with NO donors (1.25 mM) showed a cytoplasmic condensation and vacuolization. DNA strand break analysis by the TUNEL method confirmed the presence of a DNA fragmentation after 24 hours of NO treatment. There was also a progressive decrease in the DNA diploid peak in response to NO donors. In parallel, p53 protein, constitutively expressed in cytoplasmic synovial cells, showed markedly increased expression after a 6-hour NO exposure and displayed prominent nuclear staining after 12 hours. Conclusions. This study demonstrates the potential role of NO for the induction of synoviocyte apoptosis in OA. The increased expression of p53 in the nucleus may play a protective role in the control of apoptosis. Key words: Nitric oxide, Apoptosis, Synovial cells, Osteoarthritis, p53 protein.
Introduction
(IL-1â), tumor necrosis factor á (TNF-á) and interleukin-6 (IL-6) [5–6], which exert a negative e#ect on cartilage turnover. Synovial membrane is exposed very early to the aggressive factors contained in the synovial fluid. Apoptosis is characterized by nuclear chromatin condensation and DNA digestion into nucleosomesized fragments. Several proteins can regulate the apoptosis such as the p53 tumor suppressor that increases in response to DNA damage. P53 serves as a regulator of cell survival and proliferation and regulates a variety of DNA repairs [7]. Nitric oxide (NO) is considered as an extracellular messenger. It can di#use very rapidly and acts on other neighboring cells. Also, it has been shown that NO inhibits cell proliferation and can induce apoptosis in other cell models such as murine macrophages [8] or mouse thymocytes [9]. Recently, human chondrocytes and rabbit synoviocytes were found to synthesize substantial amounts of NO in response to IL-1â [10–12].
OSTEOARTHRITIS (OA) is characterized by morphological and quantitative alterations of articular cartilage leading to its destruction. The early changes in osteoarthritic joints seem partly to involve the synovial membrane as observed in experimental models [1]. Synovial membrane in OA exhibits morphological changes and early increase in enzymatic activity [2]. Moreover, synovial inflammation is often observed during the flares of the disease and probably plays an important role in the pathogenesis of OA [3]. Previous studies have shown that synovitis may be prominent next to areas of severe cartilage lesions [4]. Synoviocytes in OA produce metalloproteases and proinflammatory cytokines such as interleukin-1â Received 7 May 1998; accepted 25 August 1998. Address correspondence to: Didier Borderie, Laboratoire de Biochimie A, Hopital Cochin, 27 rue du faubourg Saint Jacques 75014 Paris, France.
203
204
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis
Implication of NO in joint diseases is also suggested by the report of substantial concentrations of metabolites of NO in human OA synovial fluids, such as nitrite, nitrate, and nitrosoproteins [13–15]. Chondrocytes in OA are able to produce high amounts of NO. Moreover, it was shown that larger amounts of NO may be produced by superficial chondrocytes than by their deep counterparts [16]. In addition, Blanco et al. [17] have shown NO involvement in the induction of cell death in human OA chondrocytes. The presence of apoptotic cells has been reported in the synovial membrane in RA as well as in OA. Apoptosis could thus serve as a regulatory process to control synovial cell metabolism in OA [18–19]. This study aimed to investigate the potential role of NO in the modulation of apoptosis of synoviocytes in OA. We show that apoptosis, assessed by several independent methods, can be regulated by high-dose NO donors despite p53 overexpression. Materials and methods REAGENTS
Reagents were: Dulbecco’s Minimal Essential Medium (DMEM), Phosphate bu#ered saline (PBS) (Gibco, Grand Island, NY); fetal calf serum (FCS) (Biowittaker Walkerville, MD), sodium nitroprussiate (SNP), S-nitroso-N-acetyl-penicillamine (SNAP), collagenase type I, trypsine, RNAse A, bovine serum albumin (BSA), oxyhemoglobin, sodium nitrite, sulfanilamide, N-1-naphthylethylenediamide dihydrochloride, Nicotinamide Adenine dinucleotide reduced form (NADH), pyruvate, paraformaldehyde, glutaraldehyde, Tris base, osmium tetroxide, lead citrate, N-MethylArginine (NMA), (Sigma Chemical Co. St Louis, MO), in-situ fluorescein cell death detection kit (Boehringer Mannheim Biochemicals, Indianapolis, IN) propidium iodide (Becton Dickinson Mountain View CA), p53 polyclonal antibody (Biosys SA, France), fluorescein conjugated anti-rabbit IgG antibody (Amersham Corp., Arlington Heights, IL), monoclonal biotinyled anti-rabbit IgG antibody, streptavidine-rhodamine system (Vector Laboratories, Inc, Burlingame, CA). All recombinant cytokines were purchased from Genzyme Corp., Cambridge,MA. SYNOVIOCYTE ISOLATION AND CULTURE
Human synovial cells from eight patients with OA of the hip (seven females and one male, mean
age 6212 years) were isolated from synovium obtained during joint surgery. After dissection, the superficial layer of synovium was digested with collagenase and trypsin as previously described [20], Cells were suspended in DMEM containing 10% FCS, amphotericin B, netromycin, ceftazidim, vancomycin and cultured at 5106 cells in 75 cm2 culture flasks at 37C in a humidified incubator with an atmosphere containing 5% CO2. After overnight cultured, nonadherent cells were removed and adherent cells were cultured in DMEM plus 10% FCS. When confluence was attained, cells were trypsinized and cultured on glass slides placed in 9.6 cm2 plates (2105 cells per well) or in 75 cm2 culture flasks (1106 cells per flask) at 37C (5% CO2). Experiments were performed on confluent cultures at first passage.
EVALUATION OF MORPHOLOGICAL CELLULAR CHANGES
To identify morphological changes induced by NO donors, synoviocytes were examined using a phase contrast microscope. Cells displaying apoptotic changes were identified using previously defined morphological criteria including chromatin condensation, nuclear fragmentation, and blebbing of cytoplasmic membrane [21–22].
RECOGNITION OF DNA STRAND BREAKS BY THE TERMINAL DEOXYNUCLEOTIDYL TRANSFERASE MEDIATED dUTP NICK-END LABELING (TUNEL) METHOD
To estimate the percentage of synovial cells undergoing apoptosis, cells containing DNA strand breaks were visualized by the TUNEL method as described by Gavrieli et al. [23] with an in-situ fluorescein cell death detection kit according to the recommendations of the manufacturer. Briefly, cells on slides were washed in PBS and fixed with 4% paraformaldehyde for 10 minutes and then incubated for 1 h with fluorescein labeled dUTP and deoxytransferase. After washing, the slides were examined by fluorescence microscopy (100). Controls consisted of omission of the deoxytransferase from the labeling reaction. Quantification of data was obtained by examination of a defined area of each plate. The number of cells with apoptotic nuclei and total cell numbers per field were determined by two independent observers and used for statistical analysis as previously described [24]. To assess the specific e#ect of NO aggression on DNA, cells were co-treated with
Osteoarthritis and Cartilage Vol. 7 No. 2
205
FIG. 1. Analysis of synovial cells by phase contrast microscopy. A: untreated cells at confluence, B: SNP (1.25 mM) for 12 hours: morphological changes consistent with apoptosis. Arrows show plasma membrane blebbing. Magnification 400.
NO donors and oxyhemoglobin (100 ìmol/l), a scavanger of NO.
were stained with 2% uranyl acetate in ethanol and lead citrate.
TRANSMISSION ELECTRON MICROSCOPY
FLOW CYTOMETRY OF DNA CONTENT
For electron microscopy, first passage synoviocytes were grown as monolayers in 75 cm2 flasks. At confluence, cells received 1.25 mM SNP or SNAP for the times indicated below. They were fixated in 2.5% glutaraldehyde bu#ered in PBS then scraped, pelleted, rinsed with PBS, fixed for 1 h at room temperature with PBS, post-fixed for 1 h in 2% osmium tetroxide bu#ered with PBS, dehydrated in a graded ethanol series and embedded in Epon resin. Thin sections of cell pellets
Nuclear DNA content was analysed by flow cytometry after iodide propidium staining using the method described by Vindelov et al. [25]. Briefly, cells were centrifuged at 400g for 5 minutes at room temperature and trypsinized (0.3 mg/ml) in a citrate bu#er pH 7.6, 0.4 mM. They were then incubated with RNAse A (0.4 mg/ml) for 15 minutes at +4C in the dark. After filtering the samples through 50 ìm nylon mesh, propidium iodide fluorescence of nuclei was measured on a
206
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis
FIG. 2. Electron micrographs of cultured synovial cells. A: untreated cells, B: after treatment by SNP (1.25 mM) for 3 hours: vacuolization of the cytoplasm; C: after treatment by SNP (1.25 mM) for 6 hours: important vacuolization and condensation of the cytoplasm; D: detail of the cytoplasmic membrane: blebbing of the cytoplasmic membrane, disorganization of the Golgi apparatus. Bar equals 2 ìm A–C, and 0.5 ìm D.
FACScan instrument (Becton Dickinson Mountain View CA) with a 560 nm dichromatic mirror and a 600 nm band pan filter. Results were analyzed by CellFIT software on the FACScan flow cytometer. IMMUNOFLUORESCENCE STUDY OF P53 EXPRESSION
After treatment by NO donors, cells on glass slides were fixed and permeabilized with an acetone/methanol mixture (V/V) at 20C for 10 minutes and then rehydrated in PBS plus 1% BSA. The slides were preincubated with 10% goat serum for 30 minutes and then incubated 1 hour at +37C with the p53 rabbit polyclonal antibody diluted in PBS plus 1% BSA to 1/500 or normal rabbit Ig as a negative control. The slides were then incubated with a fluorescein conjugated anti-rabbit IgG antibody diluted in PBS to 1/100. For co-localization studies of DNA strand break and p53 expression, slides were first processed for the TUNEL method as described, followed by p53 detection using a biotinylated second antibody diluted to 1/100. Visualization was performed
using a streptavidin/rhodamine system. Quantification of data was obtained by examination of defined area on each slide. Numbers of TUNEL and/or p53-positive cells and total cell numbers per field were determined by two independent observers and were used for statistical analysis as previously described [26]. QUANTIFICATION OF NITRITE
NO formation was detected by nitrite accumulation in the cultured cell supernatants by the diazotization reaction of Griess with sodium nitrite as standard [27]. Briefly, 100 ìl of culture supernatant were incubated with 100 ìl of 1% sulfanilamide, 0.1% N-1-naphthylethylenediamide dihydrochloride in 25% H3PO4 at room temperature for 10 minutes. Optical densities were measured at 570 nanometers using a Dynatech MR 5000 microplate reader (Dynex Technology S.A., France). Background levels of nitrite were determined in the cell-free DMEM with or without additives and were subtracted from the total amount of nitrite formed. The nitrite concentration was
Osteoarthritis and Cartilage Vol. 7 No. 2
207
FIG. 3. DNA fragmentation analysed by the TUNEL method. A: untreated cells, B: cells treated for 24 hours by SNP or SNAP (1.25 mM). Magnification 1000.
calculated from a NaNO2 standard curve. The detection limit for nitrite was 1 ìmol/ml. DATA PRESENTATION AND STATISTICAL ANALYSIS
All experiments were repeated at least three times. The microphotographs are representative of the results obtained in these experiments. The quantitative data are the mean values1 standard deviation (S.D.) from representative experiments performed in triplicate. Statistical significance was evaluated by Wilcoxon’s signed rank test. Results NO DONORS INDUCE MORPHOLOGICAL APOPTOTIC CHANGES
When NO was applied through SNP (1.25 mM), apoptosis was observed at the first passage by
phase contrast microscopy. After 6 hours, cells started to change morphology, became rounded and started to detach from dishes. At 12 hours, cells showed cytoplasmic condensation and blebbing of the cytoplasmic membrane (Fig. 1). To confirm these data, cells treated with NO donors were examined by electron microscopy after 0, 3 and 6 hours. The micrograph in Fig. 2 shows the changes at the 3rd hour, including cytoplasmic condensation; at this time we observed the persistence of a continuous plasma membrane and the protuberances on the cell surface that precede the shedding of apoptotic bodies. Later at the 6th hour, treated cells showed a paucity of cytoplasmic organelles and a digitization of the plasmalemma, ribosomes were not well individuated and the reticules showed a large
208
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis Table I Dose-effects of two NO donors for the induction of synoviocyte apoptosis % of apoptotic cells T=6 hours
Control SNP 0.01 mmol/l SNP 0.1 mmol/l SNP 0.5 mmol/l SNP 1 mmol/l SNP 1.25 mmol/l SNAP 0.01 mmol/l SNAP 0.1 mmol/l SNAP 0.5 mmol/l SNAP 1 mmol/l SNAP 1.25 mmol/l
% of apoptotic cells T=24 hours
Without oxyhemoglobin
With oxyhemoglobin (100 ìmol/l)
Without oxyhemoglobin
With oxyhemoglobin (100 ìmol/l)
<1 <1 <1 202 457 485 <1 <1 202 387 459
<1 <1 <1 <1 51 52 <1 <1 <1 42 32
<1 1 152 407 858 9010 <1 101 186 508 7010
<1 <1 <1 52 102 123 <1 <1 42 73 75
Cultured synovial cells from OA patients were treated with di#erent concentrations of two NO donors, SNP (sodium nitroprussiate) and SNAP (S-nitroso-N-acetyl-penicillamine). Experiments were performed without or with oxyhemoglobin to inhibit NO e#ects. Apoptosis was determined by the TUNEL method as described in the Materials and methods section. Results from duplicate experiments are expressed as the mean percentage of apoptotic cellsSD (N=5).
number of small dilated saccules. At the same time, a vacuolization in the cytoplam appeared. All these events seemed to precede the chromatin condensation. The overall e#ects of SNAP were similar at the same concentration (1.25 mM) but the time course was di#erent: the e#ect was faster with SNP than with SNAP. After 12 hours, a higher proportion of apoptotic cells was observed with SNP than with SNAP (SNP 693% versus SNAP 522%). DNA ALTERATION ANALYSIS
To confirm that NO can induce apoptosis in synoviocytes, cells were treated with 0.01 to 1.25 mM SNP or SNAP. The DNA fragmentation was analysed by the TUNEL method. This technique allows detection of DNA strand breaks by means of their 3 -OH extremities. Under normal conditions of culture, cells showed no fluorescence staining. After incubation with NO donors, there was a dose and timedependent increase in nuclear binding which was maximal after 24 hours (Fig. 3). This NO e#ect was inhibited when cells were co-incubated with NO donors and oxyhemoglobin (Table I). To follow the kinetics of these changes, cells were analysed by flow cytometry after staining with propidium iodide. In response to treatment with 1.25 mM SNP or SNAP, there was a progressive decrease in the diploid peak and an increase in the number of hypoploid cells as shown in Fig. 4. The e#ects of the NO donors were observed from
the 6th hour. At 12 hours, the flow analysis showed a complete DNA breakdown. APOPTOSIS AFTER NO PRODUCTION
To assess if endogenous NO production by synovial cells was able to induce apoptosis, synoviocytes were treated with rhIL-1â (1 ng/ml) or a combination of IL-1â (1 ng/ml), TNF-á (500 pg/ml) and IFN-ã (104 IU/ml) for 24 hours). Conditioned media were collected to measure nitrite concentrations. Only the combination of IL-1â, TNF-á and
Table II Effects of endogenous NO on synoviocyte apoptosis Treatment
Nitrite (ìmol/l)
% of apoptotic cells
Control IL-1â TNF-á IFN-ã IL-1â+TNF-á+IFN-ã IL-1â+TNF-á+IFN-ã+NMA
1.20.8 5.31.1* 2.80.3 1.90.4 112.6† 1.70.5
<1 21 42* <1 63* 21
Cultured synovial cells from OA patients were treated with the following cytokines, alone or in combination: IL-1â (1 ng/ ml), TNF-á (500 pg/ml), IFN-ã (104 IU/ml). NMA (N-methylarginine) was used to inhibit the inducible NO synthase. Nitrite in synoviocyte supernatants was measured by the Griess reaction and apoptosis was determined by the TUNEL method as described in the Materials and methods section. Results from duplicate experiments are expressed as the mean nitrite concentrationSD and as the mean percentage of apoptotic cellsSD (N=5). *P<0.01, †p<0.001 versus control cells.
Osteoarthritis and Cartilage Vol. 7 No. 2
209
IFN-ã was able to increase nitrite in cell supernatants but induced fewer than 10% apoptotic cells when analysed by both light microscopy and the TUNEL method (Table II). The synthesis of NO was inhibited when the cytokine combination was co-incubated with NMA (P<0.01) (Table II).
220 A
T0
EXPRESSION OF P53 PROTEIN IN SYNOVIOCYTES
Immunofluorescence experiments were performed to determine the subcellular localization of p53. Before NO donor treatment, we observed a granular cytoplasmic labeling in the majority of synovial cells (828% of cells); some synoviocytes also showed a weak nuclear labeling (<10%). After NO donor exposure, cells demonstrated a time and dose-dependent nuclear staining with a parallel decrease in the cytoplasmic labeling (Fig. 5). Using double-labeling for p53 and the TUNEL method, the majority of the strongly TUNELpositive cells also showed subnuclear positive immunostaining for p53 protein (Fig. 6).
0 200 B
T4
0 40 C
Discussion Numerous studies have shown that NO is able to induce apoptosis in a great number of cultured cell types, such as murine peritoneal macrophages [8], mouse thymocytes [9], or human chondrocytes [17]. Our results are the first to provide morphological and biochemical characterization of NO-induced cell apoptosis in human osteoarthritis synoviocytes. These data show evidence of cytoplasmic events such as vacuolization and protuberances of cell membranes that appear before DNA condensation observed by the TUNEL method, propidium iodide staining and flow cytometry analysis. Only the high NO donor concentrations induced synoviocyte apoptosis. However, apoptotic cells are certainly underestimated under these conditions, since some of the cells can detach from the dishes. Moreover, the LDH activity measured in the supernatants of treated cells was not significantly di#erent from that of control cells, proving the cytoplasmic membrane integrity of cells remaining adherent (data not shown). To assess the specific e#ect of the NO radical, we treated cells with two chemically di#erent NO donors. SNP requires electron transfer to release
T6
0 30 D
T12
0 20 E
T24
0
200
400
600
800
1000
FIG. 4. Analysis of synovial cell DNA content by flow cytometry after staining by propidium iodide. A: Control, B: 4 hours after SNP treatment (1.25 mM); C: 6 hours after SNP treatment; D: 12 hours after SNP treatment; E: 24 hours after SNP treatment.
210
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis
FIG. 5. Determination of p53 protein expression in cultured synovial cells. p53 was determined by immunofluorescence using a polyclonal antibody as described in the Materials and methods section. A: untreated cells containing cytoplasmic p53; B: synovial cells treated with SNP (1.25 mM) for 6 hours: staining of the synovial cells nucleus. Magnification 1000.
NO radicals whereas SNAP releases NO spontaneously, but the observed results were similar. Several mechanisms have been suggested to explain NO induced apoptosis. In rat insulinoma cells [28], it has been reported that high NO levels induced mitochondrial damage associated with DNA fragmentation and cell death. A more direct mechanism for NO-mediated apoptosis is suggested by the work of Nguen et al. [29] who demonstrated that NO could deaminate purine and pyrimidine bases in DNA and increase DNA strand breaks.
Several cell sources can produce NO in the OA articular sites. Cultures of human chondrocytes synthesize large amounts of NO in response to IL-1â [30]. The chondrocyte NO synthase presents close analogies with the brain isoform [31]. More recently, Hayashi et al. [16] have shown that larger amounts of NO may be produced by superficial chondrocytes than by their deep counterparts, showing the important role of NO at the cartilagesynovial membrane interface. Even though NO is limited by its short half life (4 sec) [32], this highly di#usible radical can theoretically exert relatively
Positive cells per 100 basal nuclei
Osteoarthritis and Cartilage Vol. 7 No. 2 ** **
100
**
** 80 * 60
*
40 20 0
0.0
6.0 12.0 Time (hours)
24.0
FIG. 6. Co-expression of p53 and DNA strand breaks in synovial cells. p53 expression was determined by immunofluorescence using a polyclonal antibody and DNA breaks were analyzed by the TUNEL method as described in the Materials and methods section. Cells were treated by SNP (1.25 mM). *P< 0.01 versus cells at t0, ** P<0.001 versus cells at t0.
long intercellular action [33], and act on synovial cells. NO could act as a synoviocyte stimulating agent in OA, leading to the production of cartilage-degrading molecules such as TNFá [34]. OA synoviocytes can also express inducible NO synthase, though to a lesser extent than in rheumatoid arthritis [34–35]. Stefanovic et al. [12] have reported an inducible form in cultured rabbit synovial cells but it seems that NO production is three or four times weaker than in chondrocytes [36]. The NO production induced by rhIL-1â (1 ng/ml) or by combination of cytokines in cultured human synovial fibroblasts during 24 hours led only few apoptotic cells (<10%). The apoptosis induced by either TNF-á or the combination of IL-1â, TNF-á, and IFN-ã seems to be related to NO production, as it is inhibited by NMA, a specific NO synthase inhibitor. Our results suggest that only high NO levels derived from exogenous cellular sources induce apoptosis in a high proportion of cells. Thus, initially, NO could play the role of an intercellular messenger. When cell capacities to neutralize the NO side e#ects by reactive oxygen species (such as superoxides) are exhausted, NO has a toxic e#ect and would cause cell death [37]. Paradoxically, we have shown that OA cultured synovial fibroblasts are able to express prominent p53 protein in the cytoplasm before addition of NO donors, and an increased nuclear staining after NO donor incubation. P53 protein has a short half-life and is rarely detected in non-neoplasic tissue, except after irradiation or exposure to DNA damaging agents [38–39]. It is known to mediate cellular response to DNA damage and maintain
211
genomic stability [40–41]. In severe cases p53 can lead to apoptosis [42]. The presence of p53 protein is consistent with the finding of Firestein et al. [43], who have reported that RA and OA cultured synovial cells express abundant immunoreactive p53, although the amount was substantially less in OA than in RA and was very small in dermal fibroblasts used as controls. Our data suggest that p53 may migrate to the DNA damage site, thus representing a mechanism of DNA repair as suggested by Bakalkin et al. [44]. However, this mechanism would be ine$cient when the DNA damage induced by NO is too great. We did not observe a reversible e#ect after eliminating NO donors from the conditioned medium and cultured cells for 24 hours. This reflects inability of the cells to repair the DNA damage and the irreversible induction of cell death. However it remains unclear why p53 was initially expressed in the cytoplasm of OA synovial cells. P53 seems to play a key role in the cellular cycle in OA synoviocytes, leading to cell growth arrest and permitting DNA repair or inducing apoptosis if DNA damage is irreversible. This conclusion is consistent with previous observations reporting a low frequency of apoptosis in OA in regard to RA synovium [18–19] whereas NO metabolites were found at increased levels in OA synovial fluids [13]. NO-induced apoptosis could play a role in the elimination of the cells involved in the inflammatory process [45]. Thus, p53 overexpression in synovial cells could limit radical aggression and permit DNA repair, high NO levels being able to induce apoptosis only when cell capacities to repair DNA damage are exceeded. So this conclusion is consistent with our observation; in the first step only cytoplasmic alterations are observed without DNA condensation. Messner et al. [40] proposed that p53 activation was the consequence of NO-induced DNA damage. In conclusion, this study shows that high NO donor concentrations induce synoviocyte apoptosis in OA. The results highlight the potential regulating role of NO produced in the OA joints. Moreover, we show that OA synoviocytes accumulate a cytoplasmic form of p53 protein probably due to the local production of inflammatory cytokines and oxide radicals in the joint [6, 13] and have a role in the regulation of apoptosis.
References 1. Lukoschek M, Scha%er MB, Burr DB, Boyd RD, Radin EL. Synovial membrane and cartilage changes in experimental osteoarthritis. J Orthop Res 1988;6:475–92.
212
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis
2. Cohen AS, Goldenberg D. Synovial fluid. In: Cohen AS, Ed. Laboratory Diagnostic Procedures in the Rheumatic Diseases. Grune and Stratton: Orlando 1985:1–54. 3. Pelletier JP, Di Batista JA, Roughley P, McCollum R, Martel-Pelletier J. Cytokines and inflammation in cartilage degradation. In: Moskowitz RW, Ed. Osteoarthritis, Rheum Dis North Am. 1993:545–68. 4. Lindblad S, Hedfors E. Arthroscopic and immunohistologic characterization of knee joint synovitis in osteoarthritis. Arthritis Rheum 1987;30:1081–8. 5. Keyszer GM, Heer AH, Kriegsmann J, et al. Comparative analysis of cathepsin L, cathepsin D, and collagenase messenger RNA expression in synovial tissues of patients with rheumatoid arthritis and osteoarthritis by in situ hybridization. Arthritis Rheum 1995;38:976–84. 6. Dalton BJ, Connor JR, Johnson WJ. Interleukin-1 induces interleukin-1 alpha and interleukin-1 beta gene expression in synovial fibroblasts and peripheral blood monocytes. Arthritis Rheum 1989;32: 279–87. 7. Lane DP. Cancer: p53, guardian of the genome. Nature 1992;358:15–6. 8. Albina JE, Cui S, Mateo RB, Reichner JS. Nitric oxide mediated apoptosis in murine peritoneal macrophages. J Immunol 1993;150:5080–5. 9. Fehsel K, Kroncke KD, Meyer ML, Huber H, Wahn V, Kolb-Backrofen V. Nitric oxide induces apoptosis in mouse thymocytes. J Immunol 1995;155: 2858–65. 10. Stadler J, Stefanovic-Racic M, Biliar TR. Articular chondrocytes synthesize nitric oxide in response to cytokines and lipopolysaccharides. J Immunol 1991;147:3915–20. 11. Palmer RM, Hickery MS, Charles IG, Moncada S, Bayliss MT. Induction of nitric oxide synthase in chondrocytes. Biochem Biophys Res Commun 1993;193:398–405. 12. Stefanovic-Racic M, Stadler J, Georgescu HI, Evans CH. Nitric oxide: synthesis and its regulation by rabbit synoviocytes. J Rheumatol 1994;21:1892–8. 13. Farrell AJ, Blake DR, Palmer RMJ, Moncada J. Concentrations of nitrite in synovial fluid and serum samples suggest increased nitric oxide synthesis in rheumatic diseases. Ann Rheum Dis 1992;51:1219–22. 14. McNicholas MJ, Aldaberth T, Drysdale MJ, et al. Synovial fluid from knees with secondary osteoarthrosis has much higher nitrite concentrations than serum suggesting local production of nitric oxide. Br J Rheum 1996;35:105 (Abstr.). 15. Hilliquin P, Borderie D, Hernvann A, Menkes CJ, Ekindjian OG. Presence of nitric oxide complexed to S-nitrosoproteins in rheumatoid arthritis. Arthritis Rheum 1997;40:1512–7. 16. Hayashi T, Abe E, Yamate T, Taguchi Y, Jasin HE. Nitric oxide production by superficial and deep articular chondrocytes. Arthritis Rheum 1997;40: 261–9. 17. Blanco FJ, Ochs RL, Schawz H, Lotz M. Chondrocyte apoptosis induced by nitric oxide. Am J Pathol 1995;146:75–85. 18. Firestein GS, Yeo M, Zvaifler NJ. Apoptosis in rheumatoid arthritis synovium. J Clin Invest 1995;96:1631–8.
19. Sugiyama M, Tsukasaki T, Yonekura A, Matsuzaki S, Yamashita S, Iwasaki K. Localisation of apoptosis and expression of apoptosis related proteins in the synovium of patients with rheumatoid arthritis. Ann Rheum Dis 1996;55:442–9. 20. Hernvann A, Aussel C, Cynober L, Moatti N, Ekindjian OG. IL-1â, a strong mediator for glucose uptake by rheumatoid and non rheumatoid cultured human synoviocytes. FEBS Lett 1992;303: 77–80. 21. Willie AH. Cell death. Int Rev Cytol 1987;17 (suppl):755–85. 22. Cohen JJ. Programmed cell death in immune system. Adv Immunol 1991;50:55–85. 23. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 1992;199:493–501. 24. Nitsch DD, Ghilardi N, Mu¨hl H, Nitsch C, Bru¨ne B, Pfeilschifter J. Apoptosis and expression of inductible nitric oxide synthase are mutually exclusive in renal mesengial cells. Am J Pathol 1997;150: 889–900. 25. Vindelov L, Christensen L, Nissen N. A detergenttrypsin method for the preparation of nuclei for flow cytometric DNA analysis. Cytometry 1993;3: 317–22. 26. Coates PJ, Save V, Ansari B, Hall PA. Demonstration of DNA damage/repair in individual cells using in situ end labelling: association of p53 with sites of DNA domage. J Pathol 1995;176:19–26. 27. Ding AH, Nathan CF, Stuehr DJ. Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages. Comparison of activating cytokines and evidence for independent production. J Immunol 1988;141: 2407–13. 28. Suarez-Pinzon WC, Stryaadka K, Schulz R, Rabinovitch A. Mechanisms of cytokine-induced destruction of rat insulinoma cells: The role of nitric oxide. Endocrinology 1994;134:1006–10. 29. Nguen J, Brunson D, Cresp CC, Penman BW, Wisanok JJ, Tannenbaum SR. DNA damage and mutation in human cells exposed to nitric oxide in vitro. Proc Natl Acad Sci USA 1992;89:3030–4. 30. Charles IG, Palmer RMJ, Hickery MS, et al. Cloning, characterization and expression of cDNA encoding an inducible nitric oxide synthase from the human chondrocyte. Proc Natl Acad Sci USA 1993;90:11 419–23. 31. Amin AR, Di Cesare PE, Vyas P, et al. The expression and regulation of nitric oxide synthase in human osteoarthritis-a#ected chondrocytes: evidence for up-regulated neuronal nitric oxide synthase. J Exp Med 1995;182:2097–102. 32. Lancaster JR Jr. Simulation of the di#usion and reaction of endogenously produced nitric oxide. Proc Natl Acad Sci USA 1994;91:8137–41. 33. Evans CH, Stefanovic-Racic M, Lancaster J. Nitric oxide and its role in orthopaedic disease. Clin Orthop Related Res 1995;312:275–94. 34. McInnes IB, Leung BP, Field M, et al. Production of nitric oxide in the synovial membrane of rheumatoid and osteoarthritis patients. J Exp Med 1996;184:1519–24. 35. Sakurai H, Kohsaka H, Liu MF, et al. Nitric oxide production and inducible nitric oxide synthase
Osteoarthritis and Cartilage Vol. 7 No. 2
36. 37.
38.
39.
40.
expression in inflammatory arthritides. J Clin Invest 1995;96:2357–63. Stefanovic-Racic M, Stadler J, Evans CH. Nitric oxide and arthritis. Arthritis Rheum 1993;36: 1036–44. Sandau K, Pfeilschifter J, Bru¨ne B. The balance between nitric oxide and superoxide determines apoptotic and necrotic death of rat mesangial cells. J Immunol 1997;158:4938–46. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B, Craig RW. Participation of p53 protein in the cellular response to DNA damage. Cancer Res 1991;51:6304–11. Merritt AJ, Potten CS, Kemp CJ, et al. The role of p53 in spontaneous and radiation-induced apoptosis in the gastrointestinal tract of normal and p53-deficient mice. Cancer Res 1994;54:451–6. Messner UK, Bru¨ne B. Nitric oxide-induced apoptosis: p53-dependent and p53-independent signaling pathways. Biochem J 1995;319:299–305.
213
41. Kuerbitz SJ, Plunkett BS, Walsh WV, Kastan MB. Wild-type p53 is a cell checkpoint determinant following irradiation. Proc Natl Acad Sci USA 1992;89:7491–5. 42. Messner UK, Ankarcrona M, Nicotera P, Bru¨ne B. P53 expression in nitric oxide induced apoptosis. FEBS Lett 1994;355:23–6. 43. Firestein GS, Nguen K, Aupperle KR, Yeo M, Boyle DL, Zvaifler NJ. Apoptosis in rheumatoid arthritis: P53 overexpression in rheumatoid arthritis synovium. Am J Pathol 1996;149:2143–51. 44. Bakalkin G, Yakovleva T, Selivanova G, et al. P53 binds single-stranded DNA ends and catalyses DNA renaturation and strand transfer. Proc Natl Acad Sc USA 1994;91:413–7. 45. Okuda Y, Sakoda S, Fujimura H, Yanagihara T. Nitric oxide via an inductible isoform of nitric oxide synthase is a possible factor to eliminate inflammatory cells from the central nervous system of mice with experimental allergic encephalomyelitis. J Neuroimmunol 1997;73:107–16.
1063–4584/99/020203+11 $12.00/0
Apoptosis induced by nitric oxide is associated with nuclear p53 protein expression in cultured osteoarthritic synoviocytes
BY DIDIER BORDERIE*, PASCAL HILLIQUIN†, ALAIN HERNVANN*‡, HERVEu LEMARECHAL*, CHARLES J. MENKES† AND OHVANESSE G. EKINDJIAN*‡ *Laboratoire de Biochimie A, †Institut de Rhumatologie, Hoˆpital Cochin, Universite´ Rene´ Descartes, 27 rue du fg St Jacques, 75014 Paris, France ‡Biochimie cellulaire de l’inflammation, Universite´ Paris XI, rue J.B. Clement, Chatenay-Malabry, France Summary Objective. Nitric oxide (NO) is a free radical molecule endogenously produced by NO synthases that may play a critical role in inflammation. It inhibits cell proliferation and may be involved in the induction of apoptosis in various cellular models. Recently, the presence of apoptotic cells was reported in the synovium of osteoarthritic (OA) patients. The aim of this study was to determine whether synovial fibroblasts are target cells for NO-induced apoptosis. The expression of p53 protein was also studied to evaluate the ability of synovial cells to repair DNA fragmentation. Methods. Synoviocytes from OA patients were treated with two NO donors: sodium nitroprusside (SNP) and S-nitroso-N-acetyl-penicillamine (SNAP). Apoptosis was analysed by transmission electron microscopy. DNA content was evaluated by flow cytometric analysis after propidium iodide staining and recognition of DNA strand break determined by the TUNEL (TdT-mediated dUTP nick end labeling) method. P53 protein expression was studied by immunofluorescence using a monoclonal antibody. Results. After 6 hours, cells treated with NO donors (1.25 mM) showed a cytoplasmic condensation and vacuolization. DNA strand break analysis by the TUNEL method confirmed the presence of a DNA fragmentation after 24 hours of NO treatment. There was also a progressive decrease in the DNA diploid peak in response to NO donors. In parallel, p53 protein, constitutively expressed in cytoplasmic synovial cells, showed markedly increased expression after a 6-hour NO exposure and displayed prominent nuclear staining after 12 hours. Conclusions. This study demonstrates the potential role of NO for the induction of synoviocyte apoptosis in OA. The increased expression of p53 in the nucleus may play a protective role in the control of apoptosis. Key words: Nitric oxide, Apoptosis, Synovial cells, Osteoarthritis, p53 protein.
Introduction
(IL-1â), tumor necrosis factor á (TNF-á) and interleukin-6 (IL-6) [5–6], which exert a negative e#ect on cartilage turnover. Synovial membrane is exposed very early to the aggressive factors contained in the synovial fluid. Apoptosis is characterized by nuclear chromatin condensation and DNA digestion into nucleosomesized fragments. Several proteins can regulate the apoptosis such as the p53 tumor suppressor that increases in response to DNA damage. P53 serves as a regulator of cell survival and proliferation and regulates a variety of DNA repairs [7]. Nitric oxide (NO) is considered as an extracellular messenger. It can di#use very rapidly and acts on other neighboring cells. Also, it has been shown that NO inhibits cell proliferation and can induce apoptosis in other cell models such as murine macrophages [8] or mouse thymocytes [9]. Recently, human chondrocytes and rabbit synoviocytes were found to synthesize substantial amounts of NO in response to IL-1â [10–12].
OSTEOARTHRITIS (OA) is characterized by morphological and quantitative alterations of articular cartilage leading to its destruction. The early changes in osteoarthritic joints seem partly to involve the synovial membrane as observed in experimental models [1]. Synovial membrane in OA exhibits morphological changes and early increase in enzymatic activity [2]. Moreover, synovial inflammation is often observed during the flares of the disease and probably plays an important role in the pathogenesis of OA [3]. Previous studies have shown that synovitis may be prominent next to areas of severe cartilage lesions [4]. Synoviocytes in OA produce metalloproteases and proinflammatory cytokines such as interleukin-1â Received 7 May 1998; accepted 25 August 1998. Address correspondence to: Didier Borderie, Laboratoire de Biochimie A, Hopital Cochin, 27 rue du faubourg Saint Jacques 75014 Paris, France.
203
204
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis
Implication of NO in joint diseases is also suggested by the report of substantial concentrations of metabolites of NO in human OA synovial fluids, such as nitrite, nitrate, and nitrosoproteins [13–15]. Chondrocytes in OA are able to produce high amounts of NO. Moreover, it was shown that larger amounts of NO may be produced by superficial chondrocytes than by their deep counterparts [16]. In addition, Blanco et al. [17] have shown NO involvement in the induction of cell death in human OA chondrocytes. The presence of apoptotic cells has been reported in the synovial membrane in RA as well as in OA. Apoptosis could thus serve as a regulatory process to control synovial cell metabolism in OA [18–19]. This study aimed to investigate the potential role of NO in the modulation of apoptosis of synoviocytes in OA. We show that apoptosis, assessed by several independent methods, can be regulated by high-dose NO donors despite p53 overexpression. Materials and methods REAGENTS
Reagents were: Dulbecco’s Minimal Essential Medium (DMEM), Phosphate bu#ered saline (PBS) (Gibco, Grand Island, NY); fetal calf serum (FCS) (Biowittaker Walkerville, MD), sodium nitroprussiate (SNP), S-nitroso-N-acetyl-penicillamine (SNAP), collagenase type I, trypsine, RNAse A, bovine serum albumin (BSA), oxyhemoglobin, sodium nitrite, sulfanilamide, N-1-naphthylethylenediamide dihydrochloride, Nicotinamide Adenine dinucleotide reduced form (NADH), pyruvate, paraformaldehyde, glutaraldehyde, Tris base, osmium tetroxide, lead citrate, N-MethylArginine (NMA), (Sigma Chemical Co. St Louis, MO), in-situ fluorescein cell death detection kit (Boehringer Mannheim Biochemicals, Indianapolis, IN) propidium iodide (Becton Dickinson Mountain View CA), p53 polyclonal antibody (Biosys SA, France), fluorescein conjugated anti-rabbit IgG antibody (Amersham Corp., Arlington Heights, IL), monoclonal biotinyled anti-rabbit IgG antibody, streptavidine-rhodamine system (Vector Laboratories, Inc, Burlingame, CA). All recombinant cytokines were purchased from Genzyme Corp., Cambridge,MA. SYNOVIOCYTE ISOLATION AND CULTURE
Human synovial cells from eight patients with OA of the hip (seven females and one male, mean
age 6212 years) were isolated from synovium obtained during joint surgery. After dissection, the superficial layer of synovium was digested with collagenase and trypsin as previously described [20], Cells were suspended in DMEM containing 10% FCS, amphotericin B, netromycin, ceftazidim, vancomycin and cultured at 5106 cells in 75 cm2 culture flasks at 37C in a humidified incubator with an atmosphere containing 5% CO2. After overnight cultured, nonadherent cells were removed and adherent cells were cultured in DMEM plus 10% FCS. When confluence was attained, cells were trypsinized and cultured on glass slides placed in 9.6 cm2 plates (2105 cells per well) or in 75 cm2 culture flasks (1106 cells per flask) at 37C (5% CO2). Experiments were performed on confluent cultures at first passage.
EVALUATION OF MORPHOLOGICAL CELLULAR CHANGES
To identify morphological changes induced by NO donors, synoviocytes were examined using a phase contrast microscope. Cells displaying apoptotic changes were identified using previously defined morphological criteria including chromatin condensation, nuclear fragmentation, and blebbing of cytoplasmic membrane [21–22].
RECOGNITION OF DNA STRAND BREAKS BY THE TERMINAL DEOXYNUCLEOTIDYL TRANSFERASE MEDIATED dUTP NICK-END LABELING (TUNEL) METHOD
To estimate the percentage of synovial cells undergoing apoptosis, cells containing DNA strand breaks were visualized by the TUNEL method as described by Gavrieli et al. [23] with an in-situ fluorescein cell death detection kit according to the recommendations of the manufacturer. Briefly, cells on slides were washed in PBS and fixed with 4% paraformaldehyde for 10 minutes and then incubated for 1 h with fluorescein labeled dUTP and deoxytransferase. After washing, the slides were examined by fluorescence microscopy (100). Controls consisted of omission of the deoxytransferase from the labeling reaction. Quantification of data was obtained by examination of a defined area of each plate. The number of cells with apoptotic nuclei and total cell numbers per field were determined by two independent observers and used for statistical analysis as previously described [24]. To assess the specific e#ect of NO aggression on DNA, cells were co-treated with
Osteoarthritis and Cartilage Vol. 7 No. 2
205
FIG. 1. Analysis of synovial cells by phase contrast microscopy. A: untreated cells at confluence, B: SNP (1.25 mM) for 12 hours: morphological changes consistent with apoptosis. Arrows show plasma membrane blebbing. Magnification 400.
NO donors and oxyhemoglobin (100 ìmol/l), a scavanger of NO.
were stained with 2% uranyl acetate in ethanol and lead citrate.
TRANSMISSION ELECTRON MICROSCOPY
FLOW CYTOMETRY OF DNA CONTENT
For electron microscopy, first passage synoviocytes were grown as monolayers in 75 cm2 flasks. At confluence, cells received 1.25 mM SNP or SNAP for the times indicated below. They were fixated in 2.5% glutaraldehyde bu#ered in PBS then scraped, pelleted, rinsed with PBS, fixed for 1 h at room temperature with PBS, post-fixed for 1 h in 2% osmium tetroxide bu#ered with PBS, dehydrated in a graded ethanol series and embedded in Epon resin. Thin sections of cell pellets
Nuclear DNA content was analysed by flow cytometry after iodide propidium staining using the method described by Vindelov et al. [25]. Briefly, cells were centrifuged at 400g for 5 minutes at room temperature and trypsinized (0.3 mg/ml) in a citrate bu#er pH 7.6, 0.4 mM. They were then incubated with RNAse A (0.4 mg/ml) for 15 minutes at +4C in the dark. After filtering the samples through 50 ìm nylon mesh, propidium iodide fluorescence of nuclei was measured on a
206
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis
FIG. 2. Electron micrographs of cultured synovial cells. A: untreated cells, B: after treatment by SNP (1.25 mM) for 3 hours: vacuolization of the cytoplasm; C: after treatment by SNP (1.25 mM) for 6 hours: important vacuolization and condensation of the cytoplasm; D: detail of the cytoplasmic membrane: blebbing of the cytoplasmic membrane, disorganization of the Golgi apparatus. Bar equals 2 ìm A–C, and 0.5 ìm D.
FACScan instrument (Becton Dickinson Mountain View CA) with a 560 nm dichromatic mirror and a 600 nm band pan filter. Results were analyzed by CellFIT software on the FACScan flow cytometer. IMMUNOFLUORESCENCE STUDY OF P53 EXPRESSION
After treatment by NO donors, cells on glass slides were fixed and permeabilized with an acetone/methanol mixture (V/V) at 20C for 10 minutes and then rehydrated in PBS plus 1% BSA. The slides were preincubated with 10% goat serum for 30 minutes and then incubated 1 hour at +37C with the p53 rabbit polyclonal antibody diluted in PBS plus 1% BSA to 1/500 or normal rabbit Ig as a negative control. The slides were then incubated with a fluorescein conjugated anti-rabbit IgG antibody diluted in PBS to 1/100. For co-localization studies of DNA strand break and p53 expression, slides were first processed for the TUNEL method as described, followed by p53 detection using a biotinylated second antibody diluted to 1/100. Visualization was performed
using a streptavidin/rhodamine system. Quantification of data was obtained by examination of defined area on each slide. Numbers of TUNEL and/or p53-positive cells and total cell numbers per field were determined by two independent observers and were used for statistical analysis as previously described [26]. QUANTIFICATION OF NITRITE
NO formation was detected by nitrite accumulation in the cultured cell supernatants by the diazotization reaction of Griess with sodium nitrite as standard [27]. Briefly, 100 ìl of culture supernatant were incubated with 100 ìl of 1% sulfanilamide, 0.1% N-1-naphthylethylenediamide dihydrochloride in 25% H3PO4 at room temperature for 10 minutes. Optical densities were measured at 570 nanometers using a Dynatech MR 5000 microplate reader (Dynex Technology S.A., France). Background levels of nitrite were determined in the cell-free DMEM with or without additives and were subtracted from the total amount of nitrite formed. The nitrite concentration was
Osteoarthritis and Cartilage Vol. 7 No. 2
207
FIG. 3. DNA fragmentation analysed by the TUNEL method. A: untreated cells, B: cells treated for 24 hours by SNP or SNAP (1.25 mM). Magnification 1000.
calculated from a NaNO2 standard curve. The detection limit for nitrite was 1 ìmol/ml. DATA PRESENTATION AND STATISTICAL ANALYSIS
All experiments were repeated at least three times. The microphotographs are representative of the results obtained in these experiments. The quantitative data are the mean values1 standard deviation (S.D.) from representative experiments performed in triplicate. Statistical significance was evaluated by Wilcoxon’s signed rank test. Results NO DONORS INDUCE MORPHOLOGICAL APOPTOTIC CHANGES
When NO was applied through SNP (1.25 mM), apoptosis was observed at the first passage by
phase contrast microscopy. After 6 hours, cells started to change morphology, became rounded and started to detach from dishes. At 12 hours, cells showed cytoplasmic condensation and blebbing of the cytoplasmic membrane (Fig. 1). To confirm these data, cells treated with NO donors were examined by electron microscopy after 0, 3 and 6 hours. The micrograph in Fig. 2 shows the changes at the 3rd hour, including cytoplasmic condensation; at this time we observed the persistence of a continuous plasma membrane and the protuberances on the cell surface that precede the shedding of apoptotic bodies. Later at the 6th hour, treated cells showed a paucity of cytoplasmic organelles and a digitization of the plasmalemma, ribosomes were not well individuated and the reticules showed a large
208
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis Table I Dose-effects of two NO donors for the induction of synoviocyte apoptosis % of apoptotic cells T=6 hours
Control SNP 0.01 mmol/l SNP 0.1 mmol/l SNP 0.5 mmol/l SNP 1 mmol/l SNP 1.25 mmol/l SNAP 0.01 mmol/l SNAP 0.1 mmol/l SNAP 0.5 mmol/l SNAP 1 mmol/l SNAP 1.25 mmol/l
% of apoptotic cells T=24 hours
Without oxyhemoglobin
With oxyhemoglobin (100 ìmol/l)
Without oxyhemoglobin
With oxyhemoglobin (100 ìmol/l)
<1 <1 <1 202 457 485 <1 <1 202 387 459
<1 <1 <1 <1 51 52 <1 <1 <1 42 32
<1 1 152 407 858 9010 <1 101 186 508 7010
<1 <1 <1 52 102 123 <1 <1 42 73 75
Cultured synovial cells from OA patients were treated with di#erent concentrations of two NO donors, SNP (sodium nitroprussiate) and SNAP (S-nitroso-N-acetyl-penicillamine). Experiments were performed without or with oxyhemoglobin to inhibit NO e#ects. Apoptosis was determined by the TUNEL method as described in the Materials and methods section. Results from duplicate experiments are expressed as the mean percentage of apoptotic cellsSD (N=5).
number of small dilated saccules. At the same time, a vacuolization in the cytoplam appeared. All these events seemed to precede the chromatin condensation. The overall e#ects of SNAP were similar at the same concentration (1.25 mM) but the time course was di#erent: the e#ect was faster with SNP than with SNAP. After 12 hours, a higher proportion of apoptotic cells was observed with SNP than with SNAP (SNP 693% versus SNAP 522%). DNA ALTERATION ANALYSIS
To confirm that NO can induce apoptosis in synoviocytes, cells were treated with 0.01 to 1.25 mM SNP or SNAP. The DNA fragmentation was analysed by the TUNEL method. This technique allows detection of DNA strand breaks by means of their 3 -OH extremities. Under normal conditions of culture, cells showed no fluorescence staining. After incubation with NO donors, there was a dose and timedependent increase in nuclear binding which was maximal after 24 hours (Fig. 3). This NO e#ect was inhibited when cells were co-incubated with NO donors and oxyhemoglobin (Table I). To follow the kinetics of these changes, cells were analysed by flow cytometry after staining with propidium iodide. In response to treatment with 1.25 mM SNP or SNAP, there was a progressive decrease in the diploid peak and an increase in the number of hypoploid cells as shown in Fig. 4. The e#ects of the NO donors were observed from
the 6th hour. At 12 hours, the flow analysis showed a complete DNA breakdown. APOPTOSIS AFTER NO PRODUCTION
To assess if endogenous NO production by synovial cells was able to induce apoptosis, synoviocytes were treated with rhIL-1â (1 ng/ml) or a combination of IL-1â (1 ng/ml), TNF-á (500 pg/ml) and IFN-ã (104 IU/ml) for 24 hours). Conditioned media were collected to measure nitrite concentrations. Only the combination of IL-1â, TNF-á and
Table II Effects of endogenous NO on synoviocyte apoptosis Treatment
Nitrite (ìmol/l)
% of apoptotic cells
Control IL-1â TNF-á IFN-ã IL-1â+TNF-á+IFN-ã IL-1â+TNF-á+IFN-ã+NMA
1.20.8 5.31.1* 2.80.3 1.90.4 112.6† 1.70.5
<1 21 42* <1 63* 21
Cultured synovial cells from OA patients were treated with the following cytokines, alone or in combination: IL-1â (1 ng/ ml), TNF-á (500 pg/ml), IFN-ã (104 IU/ml). NMA (N-methylarginine) was used to inhibit the inducible NO synthase. Nitrite in synoviocyte supernatants was measured by the Griess reaction and apoptosis was determined by the TUNEL method as described in the Materials and methods section. Results from duplicate experiments are expressed as the mean nitrite concentrationSD and as the mean percentage of apoptotic cellsSD (N=5). *P<0.01, †p<0.001 versus control cells.
Osteoarthritis and Cartilage Vol. 7 No. 2
209
IFN-ã was able to increase nitrite in cell supernatants but induced fewer than 10% apoptotic cells when analysed by both light microscopy and the TUNEL method (Table II). The synthesis of NO was inhibited when the cytokine combination was co-incubated with NMA (P<0.01) (Table II).
220 A
T0
EXPRESSION OF P53 PROTEIN IN SYNOVIOCYTES
Immunofluorescence experiments were performed to determine the subcellular localization of p53. Before NO donor treatment, we observed a granular cytoplasmic labeling in the majority of synovial cells (828% of cells); some synoviocytes also showed a weak nuclear labeling (<10%). After NO donor exposure, cells demonstrated a time and dose-dependent nuclear staining with a parallel decrease in the cytoplasmic labeling (Fig. 5). Using double-labeling for p53 and the TUNEL method, the majority of the strongly TUNELpositive cells also showed subnuclear positive immunostaining for p53 protein (Fig. 6).
0 200 B
T4
0 40 C
Discussion Numerous studies have shown that NO is able to induce apoptosis in a great number of cultured cell types, such as murine peritoneal macrophages [8], mouse thymocytes [9], or human chondrocytes [17]. Our results are the first to provide morphological and biochemical characterization of NO-induced cell apoptosis in human osteoarthritis synoviocytes. These data show evidence of cytoplasmic events such as vacuolization and protuberances of cell membranes that appear before DNA condensation observed by the TUNEL method, propidium iodide staining and flow cytometry analysis. Only the high NO donor concentrations induced synoviocyte apoptosis. However, apoptotic cells are certainly underestimated under these conditions, since some of the cells can detach from the dishes. Moreover, the LDH activity measured in the supernatants of treated cells was not significantly di#erent from that of control cells, proving the cytoplasmic membrane integrity of cells remaining adherent (data not shown). To assess the specific e#ect of the NO radical, we treated cells with two chemically di#erent NO donors. SNP requires electron transfer to release
T6
0 30 D
T12
0 20 E
T24
0
200
400
600
800
1000
FIG. 4. Analysis of synovial cell DNA content by flow cytometry after staining by propidium iodide. A: Control, B: 4 hours after SNP treatment (1.25 mM); C: 6 hours after SNP treatment; D: 12 hours after SNP treatment; E: 24 hours after SNP treatment.
210
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis
FIG. 5. Determination of p53 protein expression in cultured synovial cells. p53 was determined by immunofluorescence using a polyclonal antibody as described in the Materials and methods section. A: untreated cells containing cytoplasmic p53; B: synovial cells treated with SNP (1.25 mM) for 6 hours: staining of the synovial cells nucleus. Magnification 1000.
NO radicals whereas SNAP releases NO spontaneously, but the observed results were similar. Several mechanisms have been suggested to explain NO induced apoptosis. In rat insulinoma cells [28], it has been reported that high NO levels induced mitochondrial damage associated with DNA fragmentation and cell death. A more direct mechanism for NO-mediated apoptosis is suggested by the work of Nguen et al. [29] who demonstrated that NO could deaminate purine and pyrimidine bases in DNA and increase DNA strand breaks.
Several cell sources can produce NO in the OA articular sites. Cultures of human chondrocytes synthesize large amounts of NO in response to IL-1â [30]. The chondrocyte NO synthase presents close analogies with the brain isoform [31]. More recently, Hayashi et al. [16] have shown that larger amounts of NO may be produced by superficial chondrocytes than by their deep counterparts, showing the important role of NO at the cartilagesynovial membrane interface. Even though NO is limited by its short half life (4 sec) [32], this highly di#usible radical can theoretically exert relatively
Positive cells per 100 basal nuclei
Osteoarthritis and Cartilage Vol. 7 No. 2 ** **
100
**
** 80 * 60
*
40 20 0
0.0
6.0 12.0 Time (hours)
24.0
FIG. 6. Co-expression of p53 and DNA strand breaks in synovial cells. p53 expression was determined by immunofluorescence using a polyclonal antibody and DNA breaks were analyzed by the TUNEL method as described in the Materials and methods section. Cells were treated by SNP (1.25 mM). *P< 0.01 versus cells at t0, ** P<0.001 versus cells at t0.
long intercellular action [33], and act on synovial cells. NO could act as a synoviocyte stimulating agent in OA, leading to the production of cartilage-degrading molecules such as TNFá [34]. OA synoviocytes can also express inducible NO synthase, though to a lesser extent than in rheumatoid arthritis [34–35]. Stefanovic et al. [12] have reported an inducible form in cultured rabbit synovial cells but it seems that NO production is three or four times weaker than in chondrocytes [36]. The NO production induced by rhIL-1â (1 ng/ml) or by combination of cytokines in cultured human synovial fibroblasts during 24 hours led only few apoptotic cells (<10%). The apoptosis induced by either TNF-á or the combination of IL-1â, TNF-á, and IFN-ã seems to be related to NO production, as it is inhibited by NMA, a specific NO synthase inhibitor. Our results suggest that only high NO levels derived from exogenous cellular sources induce apoptosis in a high proportion of cells. Thus, initially, NO could play the role of an intercellular messenger. When cell capacities to neutralize the NO side e#ects by reactive oxygen species (such as superoxides) are exhausted, NO has a toxic e#ect and would cause cell death [37]. Paradoxically, we have shown that OA cultured synovial fibroblasts are able to express prominent p53 protein in the cytoplasm before addition of NO donors, and an increased nuclear staining after NO donor incubation. P53 protein has a short half-life and is rarely detected in non-neoplasic tissue, except after irradiation or exposure to DNA damaging agents [38–39]. It is known to mediate cellular response to DNA damage and maintain
211
genomic stability [40–41]. In severe cases p53 can lead to apoptosis [42]. The presence of p53 protein is consistent with the finding of Firestein et al. [43], who have reported that RA and OA cultured synovial cells express abundant immunoreactive p53, although the amount was substantially less in OA than in RA and was very small in dermal fibroblasts used as controls. Our data suggest that p53 may migrate to the DNA damage site, thus representing a mechanism of DNA repair as suggested by Bakalkin et al. [44]. However, this mechanism would be ine$cient when the DNA damage induced by NO is too great. We did not observe a reversible e#ect after eliminating NO donors from the conditioned medium and cultured cells for 24 hours. This reflects inability of the cells to repair the DNA damage and the irreversible induction of cell death. However it remains unclear why p53 was initially expressed in the cytoplasm of OA synovial cells. P53 seems to play a key role in the cellular cycle in OA synoviocytes, leading to cell growth arrest and permitting DNA repair or inducing apoptosis if DNA damage is irreversible. This conclusion is consistent with previous observations reporting a low frequency of apoptosis in OA in regard to RA synovium [18–19] whereas NO metabolites were found at increased levels in OA synovial fluids [13]. NO-induced apoptosis could play a role in the elimination of the cells involved in the inflammatory process [45]. Thus, p53 overexpression in synovial cells could limit radical aggression and permit DNA repair, high NO levels being able to induce apoptosis only when cell capacities to repair DNA damage are exceeded. So this conclusion is consistent with our observation; in the first step only cytoplasmic alterations are observed without DNA condensation. Messner et al. [40] proposed that p53 activation was the consequence of NO-induced DNA damage. In conclusion, this study shows that high NO donor concentrations induce synoviocyte apoptosis in OA. The results highlight the potential regulating role of NO produced in the OA joints. Moreover, we show that OA synoviocytes accumulate a cytoplasmic form of p53 protein probably due to the local production of inflammatory cytokines and oxide radicals in the joint [6, 13] and have a role in the regulation of apoptosis.
References 1. Lukoschek M, Scha%er MB, Burr DB, Boyd RD, Radin EL. Synovial membrane and cartilage changes in experimental osteoarthritis. J Orthop Res 1988;6:475–92.
212
D. Borderie et al.: Nitric oxide induces synoviocyte apoptosis
2. Cohen AS, Goldenberg D. Synovial fluid. In: Cohen AS, Ed. Laboratory Diagnostic Procedures in the Rheumatic Diseases. Grune and Stratton: Orlando 1985:1–54. 3. Pelletier JP, Di Batista JA, Roughley P, McCollum R, Martel-Pelletier J. Cytokines and inflammation in cartilage degradation. In: Moskowitz RW, Ed. Osteoarthritis, Rheum Dis North Am. 1993:545–68. 4. Lindblad S, Hedfors E. Arthroscopic and immunohistologic characterization of knee joint synovitis in osteoarthritis. Arthritis Rheum 1987;30:1081–8. 5. Keyszer GM, Heer AH, Kriegsmann J, et al. Comparative analysis of cathepsin L, cathepsin D, and collagenase messenger RNA expression in synovial tissues of patients with rheumatoid arthritis and osteoarthritis by in situ hybridization. Arthritis Rheum 1995;38:976–84. 6. Dalton BJ, Connor JR, Johnson WJ. Interleukin-1 induces interleukin-1 alpha and interleukin-1 beta gene expression in synovial fibroblasts and peripheral blood monocytes. Arthritis Rheum 1989;32: 279–87. 7. Lane DP. Cancer: p53, guardian of the genome. Nature 1992;358:15–6. 8. Albina JE, Cui S, Mateo RB, Reichner JS. Nitric oxide mediated apoptosis in murine peritoneal macrophages. J Immunol 1993;150:5080–5. 9. Fehsel K, Kroncke KD, Meyer ML, Huber H, Wahn V, Kolb-Backrofen V. Nitric oxide induces apoptosis in mouse thymocytes. J Immunol 1995;155: 2858–65. 10. Stadler J, Stefanovic-Racic M, Biliar TR. Articular chondrocytes synthesize nitric oxide in response to cytokines and lipopolysaccharides. J Immunol 1991;147:3915–20. 11. Palmer RM, Hickery MS, Charles IG, Moncada S, Bayliss MT. Induction of nitric oxide synthase in chondrocytes. Biochem Biophys Res Commun 1993;193:398–405. 12. Stefanovic-Racic M, Stadler J, Georgescu HI, Evans CH. Nitric oxide: synthesis and its regulation by rabbit synoviocytes. J Rheumatol 1994;21:1892–8. 13. Farrell AJ, Blake DR, Palmer RMJ, Moncada J. Concentrations of nitrite in synovial fluid and serum samples suggest increased nitric oxide synthesis in rheumatic diseases. Ann Rheum Dis 1992;51:1219–22. 14. McNicholas MJ, Aldaberth T, Drysdale MJ, et al. Synovial fluid from knees with secondary osteoarthrosis has much higher nitrite concentrations than serum suggesting local production of nitric oxide. Br J Rheum 1996;35:105 (Abstr.). 15. Hilliquin P, Borderie D, Hernvann A, Menkes CJ, Ekindjian OG. Presence of nitric oxide complexed to S-nitrosoproteins in rheumatoid arthritis. Arthritis Rheum 1997;40:1512–7. 16. Hayashi T, Abe E, Yamate T, Taguchi Y, Jasin HE. Nitric oxide production by superficial and deep articular chondrocytes. Arthritis Rheum 1997;40: 261–9. 17. Blanco FJ, Ochs RL, Schawz H, Lotz M. Chondrocyte apoptosis induced by nitric oxide. Am J Pathol 1995;146:75–85. 18. Firestein GS, Yeo M, Zvaifler NJ. Apoptosis in rheumatoid arthritis synovium. J Clin Invest 1995;96:1631–8.
19. Sugiyama M, Tsukasaki T, Yonekura A, Matsuzaki S, Yamashita S, Iwasaki K. Localisation of apoptosis and expression of apoptosis related proteins in the synovium of patients with rheumatoid arthritis. Ann Rheum Dis 1996;55:442–9. 20. Hernvann A, Aussel C, Cynober L, Moatti N, Ekindjian OG. IL-1â, a strong mediator for glucose uptake by rheumatoid and non rheumatoid cultured human synoviocytes. FEBS Lett 1992;303: 77–80. 21. Willie AH. Cell death. Int Rev Cytol 1987;17 (suppl):755–85. 22. Cohen JJ. Programmed cell death in immune system. Adv Immunol 1991;50:55–85. 23. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 1992;199:493–501. 24. Nitsch DD, Ghilardi N, Mu¨hl H, Nitsch C, Bru¨ne B, Pfeilschifter J. Apoptosis and expression of inductible nitric oxide synthase are mutually exclusive in renal mesengial cells. Am J Pathol 1997;150: 889–900. 25. Vindelov L, Christensen L, Nissen N. A detergenttrypsin method for the preparation of nuclei for flow cytometric DNA analysis. Cytometry 1993;3: 317–22. 26. Coates PJ, Save V, Ansari B, Hall PA. Demonstration of DNA damage/repair in individual cells using in situ end labelling: association of p53 with sites of DNA domage. J Pathol 1995;176:19–26. 27. Ding AH, Nathan CF, Stuehr DJ. Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages. Comparison of activating cytokines and evidence for independent production. J Immunol 1988;141: 2407–13. 28. Suarez-Pinzon WC, Stryaadka K, Schulz R, Rabinovitch A. Mechanisms of cytokine-induced destruction of rat insulinoma cells: The role of nitric oxide. Endocrinology 1994;134:1006–10. 29. Nguen J, Brunson D, Cresp CC, Penman BW, Wisanok JJ, Tannenbaum SR. DNA damage and mutation in human cells exposed to nitric oxide in vitro. Proc Natl Acad Sci USA 1992;89:3030–4. 30. Charles IG, Palmer RMJ, Hickery MS, et al. Cloning, characterization and expression of cDNA encoding an inducible nitric oxide synthase from the human chondrocyte. Proc Natl Acad Sci USA 1993;90:11 419–23. 31. Amin AR, Di Cesare PE, Vyas P, et al. The expression and regulation of nitric oxide synthase in human osteoarthritis-a#ected chondrocytes: evidence for up-regulated neuronal nitric oxide synthase. J Exp Med 1995;182:2097–102. 32. Lancaster JR Jr. Simulation of the di#usion and reaction of endogenously produced nitric oxide. Proc Natl Acad Sci USA 1994;91:8137–41. 33. Evans CH, Stefanovic-Racic M, Lancaster J. Nitric oxide and its role in orthopaedic disease. Clin Orthop Related Res 1995;312:275–94. 34. McInnes IB, Leung BP, Field M, et al. Production of nitric oxide in the synovial membrane of rheumatoid and osteoarthritis patients. J Exp Med 1996;184:1519–24. 35. Sakurai H, Kohsaka H, Liu MF, et al. Nitric oxide production and inducible nitric oxide synthase
Osteoarthritis and Cartilage Vol. 7 No. 2
36. 37.
38.
39.
40.
expression in inflammatory arthritides. J Clin Invest 1995;96:2357–63. Stefanovic-Racic M, Stadler J, Evans CH. Nitric oxide and arthritis. Arthritis Rheum 1993;36: 1036–44. Sandau K, Pfeilschifter J, Bru¨ne B. The balance between nitric oxide and superoxide determines apoptotic and necrotic death of rat mesangial cells. J Immunol 1997;158:4938–46. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B, Craig RW. Participation of p53 protein in the cellular response to DNA damage. Cancer Res 1991;51:6304–11. Merritt AJ, Potten CS, Kemp CJ, et al. The role of p53 in spontaneous and radiation-induced apoptosis in the gastrointestinal tract of normal and p53-deficient mice. Cancer Res 1994;54:451–6. Messner UK, Bru¨ne B. Nitric oxide-induced apoptosis: p53-dependent and p53-independent signaling pathways. Biochem J 1995;319:299–305.
213
41. Kuerbitz SJ, Plunkett BS, Walsh WV, Kastan MB. Wild-type p53 is a cell checkpoint determinant following irradiation. Proc Natl Acad Sci USA 1992;89:7491–5. 42. Messner UK, Ankarcrona M, Nicotera P, Bru¨ne B. P53 expression in nitric oxide induced apoptosis. FEBS Lett 1994;355:23–6. 43. Firestein GS, Nguen K, Aupperle KR, Yeo M, Boyle DL, Zvaifler NJ. Apoptosis in rheumatoid arthritis: P53 overexpression in rheumatoid arthritis synovium. Am J Pathol 1996;149:2143–51. 44. Bakalkin G, Yakovleva T, Selivanova G, et al. P53 binds single-stranded DNA ends and catalyses DNA renaturation and strand transfer. Proc Natl Acad Sc USA 1994;91:413–7. 45. Okuda Y, Sakoda S, Fujimura H, Yanagihara T. Nitric oxide via an inductible isoform of nitric oxide synthase is a possible factor to eliminate inflammatory cells from the central nervous system of mice with experimental allergic encephalomyelitis. J Neuroimmunol 1997;73:107–16.