Interleukin-15 inhibits sodium nitroprussideinduced apoptosis of synovial fibroblasts and vascular endothelial cells.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 46, No. 11, November 2002, pp 3010–3014 DOI 10.1002/art.10610 © 2002, American College of Rheumatology Interleukin-15 Inhibits Sodium Nitroprusside–Induced Apoptosis of Synovial Fibroblasts and Vascular Endothelial Cells Lin Yang, Sherry Thornton, and Alexei A. Grom Objective. One of the pathologic hallmarks of juvenile rheumatoid arthritis (JRA) is a tumor-like expansion of inflamed synovial tissue, or pannus, which causes much of the joint damage in this disease. The expansion of pannus is supported by extensive formation of new blood vessels. We have previously shown that revascularization of minced JRA synovial tissues engrafted into SCID mice correlated with the intensity of inflammatory activity in the tissues and with interleukin-15 (IL-15) expression. Since synovial vascular endothelial cells (VECs) expressed IL-15 receptors, the present study was undertaken to investigate the hypothesis that IL-15 might play a role in neovascularization of the pannus. Methods. To evaluate IL-15 for possible angiogenic activity, we assessed the ability of recombinant human IL-15 (rHuIL-15) to induce VEC growth directly and to stimulate synovial cells to produce endothelial growth factors. Since IL-15 had been shown to inhibit apoptosis of certain immune cells, we were also interested in whether it might have similar effects on VECs. Apoptosis was induced by addition of sodium nitroprusside (SNP) at 1–2 mM to >80% confluent primary VECs, and numbers of apoptotic cells were determined by annexin V assay. Results. Addition of rHuIL-15 at 10–100 ng/ml to primary synovial fibroblast cultures failed to upregulate expression of vascular endothelial growth fac- tor and angiopoietin 1 by these cells. Although rHuIL-15 failed to induce a mitogenic response of VECs, it promoted survival of these cells on Matrigel. Preincubation of VECs with rHuIL-15 at 50 ng/ml significantly reduced the proportion of VECs undergoing apoptosis. Conclusion. IL-15 promotes survival of VECs on Matrigel and inhibits SNP-induced apoptosis of endothelial cells. We hypothesize that this mechanism may be relevant to the stabilization of newly formed vascular structures in JRA synovium. Expansion of inflamed synovial tissue, or pannus, is a characteristic pathologic feature of both rheumatoid arthritis (RA) and juvenile rheumatoid arthritis (JRA). With the progression of the diseases, the pannus invades and destroys intraarticular cartilage. The expansion of the pannus is supported by extensive formation of new blood vessels (1). Suppression of such neovascularization in animal models of arthritis leads to a significant reduction in the severity of synovial inflammation (2). Although in healthy individuals the existing vasculature is extremely stable, under inflammatory conditions the volume of blood vessels may rapidly change (3). Such changes are related to the ability of many inflammatory cytokines to modulate the expression of the direct-acting vascular-specific growth factors, leading to the changes in the overall angiogenic activity. For instance, in adult RA synovium, interleukin-1 (IL-1), tumor necrosis factor ␣ (TNF␣), and transforming growth factor ␤ have emerged as major mediators of vascular endothelial growth factor (VEGF) release (4). More recently, IL-15, another proinflammatory cytokine, has been associated with vascular endothelial cell (VEC) functions in rheumatoid synovium. (5). The expression of IL-15 in rheumatoid synovium and the presence of IL-15 receptors on VECs have been reported by several groups (5–7). Furthermore, Angiolillo Supported in part by NIH grants P30-AR-4736, P30-HD28827, and MAMDC P60-AR-44059-01, by Children’s Hospital Research Foundation of Cincinnati, and by the Schmidlapp Foundation. Lin Yang, MD, Sherry Thornton, PhD, Alexei A. Grom, MD: Children’s Hospital Medical Center, Cincinnati, Ohio. Address correspondence and reprint requests to Alexei A. Grom, MD, Division of Rheumatology, Pavilion Building 2-129, Children’s Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229. E-mail: firstname.lastname@example.org. Submitted for publication November 8, 2001; accepted in revised form July 26, 2002. 3010 IL-15 INHIBITION OF FIBROBLAST AND ENDOTHELIAL CELL APOPTOSIS et al showed that IL-15, injected subcutaneously into nude mice, consistently induced neovascularization of Matrigel plugs (8), suggesting direct angiogenic activity of the cytokine. Our group’s previous studies of JRA synovial tissue demonstrated increased vascularization and VEC activation. These features remained prominent in JRA synovial tissue fragments engrafted into SCID mice, and correlated with the intensity of inflammatory activity and IL-15 expression (9). Based on these observations, we hypothesized that IL-15 may play a major role in neovascularization of pannus tissue in JRA. To evaluate this cytokine for possible angiogenic activity, we assessed the ability of recombinant human IL-15 (rHuIL-15) to promote VEC growth directly and to stimulate other synovial cells to produce endothelial growth factors. PATIENTS AND METHODS Synovial tissue samples. Synovial tissue specimens were obtained from JRA patients undergoing joint replacement surgery as a normal part of their clinical care (the tissues would otherwise have been discarded). All patients satisfied the American College of Rheumatology criteria for the diagnosis of JRA (10). Recombinant proteins and cell lines. Recombinant human IL-15 and recombinant human basic fibroblast growth factor (bFGF) were purchased from R&D Systems (Minneapolis, MN). To establish primary synovial fibroblast cultures, synovial tissue samples were minced, resuspended in serumfree RPMI 1640 with type I collagenase (Worthington, Lakewood, NJ) at a final concentration of 4 mg/ml, and incubated at 37oC with 10% CO2 with gentle stirring for 12 hours. Cell suspensions were then filtered through sterile mesh, washed, and transferred to fresh RPMI 1640 with 10% fetal calf serum (FCS). Primary human umbilical vein endothelial cells (HUVECs) were purchased from BioWhittaker (San Diego, CA). Cells were expanded in basal media supplemented with 2% FCS and endothelial growth factors (EGM-2 BulletKit System; BioWhittaker). Western blotting. Cell pellets (106 cells/sample) were lysed in 1% Triton X-100 lysis buffer, subjected to sodium dodecyl sulfate–12.5% polyacrylamide gel electrophoresis, and transferred to polyvinylidene difluoride membranes. The membranes were immunoblotted with goat IgG polyclonal anti-human IL-15 receptor ␣ chain (IL-15R␣) antibodies (R&D Systems) at 0.1 g/ml and washed. Biotinylated antigoat IgG (Jackson ImmunoResearch, West Grove, PA) was used as secondary antibody. The membranes were then incubated with streptavidin–alkaline phosphatase, washed, and developed in BCIP/nitroblue tetrazolium. Human peripheral blood lymphocytes were used as a positive control. Cell proliferation assay. HUVECs or primary fibroblasts from the second and third passages were plated in triplicate at 5 ⫻ 103 in 0.2 ml of complete media with or without additives in 96-well flat-bottomed plates. The plates were incubated for 72 hours. DNA synthesis was determined 3011 by 3H-thymidine deoxyribose incorporation (1 Ci/well; ICN Biomedicals, Costa Mesa, CA) during the last 18 hours of culture. Endothelial cell tube formation on Matrigel. HUVECs plated on Matrigel-coated wells rearrange themselves to form tube-like structures. The advantage of using growth factor– reduced Matrigel (GFR-Matrigel) as opposed to standard Matrigel is that differences in tube density in the presence of exogenous stimulants are more readily recognized (11). GFRMatrigel was purchased from Collaborative Biomedical Products (Bedford, MA). Tube formation was evaluated as described previously (11). Briefly, 300 l of GFR-Matrigel used at 4oC to coat a 24-well plate was allowed to polymerize at 37oC for 2 hours. HUVECs (2 ⫻ 105/well) in a final volume of 1.0 ml of EGM-2–2% FCS (with no endothelial cell growth factor supplementation) were added to the wells and incubated overnight at 37oC with or without rHuIL-15 (100 ng/ml). Basic FGF at 25 ng/ml was used as a positive control. Three independent experiments were performed for each culture condition. Tube formation was examined visually at 10⫻ magnification after 24 hours of incubation. Photographs of the central part of each well were taken, and the overall length of tube-like structures was measured. Induction of apoptosis by sodium nitroprusside (SNP). The nitric oxide (NO) donor SNP (Sigma-Aldrich, St. Louis, MO) was chosen as a proapoptotic agent because the overproduction of NO has been demonstrated in RA synovium (12,13). Primary synovial fibroblasts or HUVECs were cultured in appropriate media until ⬎80% confluence was achieved. HUVECs were grown in EGM-2 supplemented with endothelial growth factors and 2% FCS. Primary synovial fibroblasts were grown in RPMI supplemented with 10% FCS. After preincubation of cells with or without rHuIL-15 for 48 hours, they were treated for an additional 24 hours with SNP at various concentrations. Cells were then harvested and analyzed for numbers of apoptotic cells. All experiments were performed in triplicate. Detection of apoptosis. The annexin V assay was used to quantify numbers of apoptotic cells (Annexin Apoptosis Detection Kit; BD PharMingen, San Diego, CA). After exposure to rHuIL-15 for 48 hours and incubation with SNP for an additional 24 hours, cells were analyzed by flow cytometry on a FACScan (Becton Dickinson, Mountain View, CA). Results were integrated with CellQuest software (Becton Dickinson) and expressed as the percentage of annexin-positive, propidium iodide–negative cells. The concentration of SNP was titrated to determine conditions in which considerable numbers of VECs were in the early apoptotic stage but the numbers of dead cells were still low. Ribonuclease protection assay (RPA). Total cellular RNA was isolated from either fresh HUVECs or primary fibroblasts, utilizing RNA Stat-60 according to the instructions of the manufacturer (Tel-Test, Friendswood, TX). The expression of angiogenic factors and their receptors was measured by the Riboquant Multi-Probe RPA (BD PharMingen), as previously described (9). The template set hAngio1, which includes the key angiogenic factors VEGF, angiopoietin 1, and their receptors, was selected for this study. Statistical analysis. Normal distribution of samples was confirmed using the Wilk-Shapiro test, and the data were 3012 YANG ET AL Figure 1. Human interleukin-15 receptor ␣ chain (IL-15R␣) gene expression in primary synovial fibroblasts and endothelial cells. Cell lysates obtained from 106 cells were subjected to sodium dodecyl sulfate–12.5% polyacrylamide gel electrophoresis, transferred to polyvinylidene difluoride membranes, and immunoblotted with anti–IL15R␣ antibodies. Lane 1, Negative control. Lane 2, Human peripheral blood lymphocytes (positive control). Lane 3, Primary synovial fibroblasts. Lane 4, Human umbilical vein endothelial cells. Molecular mass standards are shown on the right. analyzed using Student’s t-test. P values less than or equal to 0.05 were considered significant. RESULTS IL-15R␣ expression by primary synovial fibroblasts and VECs. In previous studies, investigators at our institution demonstrated high levels of IL-15 in the majority of JRA synovial tissue samples (7). In this study, the expression of IL-15R␣ in primary synovial fibroblast and VEC cultures was initially demonstrated by reverse transcriptase–polymerase chain reaction analysis (results not shown) and then confirmed by Western blot analysis (Figure 1). IL-15R␣ is detectable by Western blotting as a protein of 60–65 kd (14). IL-15 promotion of human VEC survival on GFR-Matrigel. Most biologic effects of IL-15 have been demonstrated with concentrations in the range of 10– 100 ng/ml. This range is also comparable with the levels of IL-15 protein found in synovial fluid of patients with RA (6). In our initial experiments, addition of IL-15 to the culture media at final concentrations of 10 ng/ml and 100 ng/ml did not induce proliferation of the cells (results not shown). Since proliferation of VECs also depends on their adhesive interactions with extracellular matrix proteins, we assessed the ability of rHuIL-15 to stimulate VEC growth on GFR-Matrigel. Previous studies have shown that VECs grown on Matrigel in the presence of appropriate growth factors form tube-like structures (11). HUVECs from the second or third passage were plated on GFR-Matrigel (2 ⫻ 105/well) with or without IL-15 at 100 ng/ml. Basic FGF (at 25 ng/ml), a known angiogenic agent, was used as a positive control. Although no proliferative response to IL-15 was noted (results not shown), we observed a higher density of surviving tube-like structures when cells were grown in the presence of rHuIL-15. As shown in Figure 2, no tube formation was seen in the wells in which HUVECs were grown on GFR-Matrigel alone, and the vast majority of the cells in these wells were dead. In contrast, addition of rHuIL-15 was associated with formation of tube-like structures similar to those observed in the wells containing bFGF. This observation, combined with the known antiapoptotic effects of IL-15 on human immune cells and fibroblasts, led us to hypothesize that IL-15 might protect VECs against apoptotic death. IL-15 inhibition of SNP-induced apoptosis of VECs in vitro. To test the above hypothesis, we examined the effects of IL-15 on apoptosis of VECs induced Figure 2. Effects of recombinant human interleukin-15 (hrIL-15) on endothelial cell survival on growth factor–reduced Matrigel (GFR-Matrigel). Human umbilical vein endothelial cells (2 ⫻ 105/well) in a final volume of 1.0 ml of growth factor–free EGM-2–2% fetal calf serum with (A) or without (B) hrIL-15 (at 100 ng/ml) were placed on GFR-Matrigel in a 24-well plate and incubated overnight at 37oC. Basic fibroblast growth factor (bFGF) at 25 ng/ml was used as a positive control (C). Three independent experiments were performed for each culture condition. Tube formation was examined visually at 10⫻ magnification after 24 hours of incubation. Photographs of the central portion of each well were taken, and the overall length of tube-like structures was measured. IL-15 INHIBITION OF FIBROBLAST AND ENDOTHELIAL CELL APOPTOSIS Figure 3. Effects of recombinant human interleukin-15 (rHuIL-15) on sodium nitroprusside (SNP)–induced apoptosis of endothelial cells. Human umbilical vein endothelial cells from the second or third passage were grown in EGM-2–2% fetal calf serum with or without rHuIL-15 at 50 ng/ml for 48 hours prior to addition of SNP at various concentrations. Cells were then incubated for 24 hours with or without SNP and harvested. The proportion of apoptotic cells was determined using the annexin V assay as described in Patients and Methods. Apoptotic cells were distinguished from dead cells based on propidium iodide (PI) uptake. A, Mean and SEM proportion of apoptotic cells as determined in 3 independent experiments for each culture condition. Light grey–shaded bars represent cells preincubated with rHuIL-15; dark grey–shaded bars represent cells not preincubated with rHuIL-15. With SNP concentrations of 1.5 mM and 2 mM, respectively, P ⫽ 0.01 and P ⫽ 0.004 for the proportion of apoptotic cells with rHuIL-15 preincubation versus without rHuIL-15 preincubation. B, Typical flow cytometry results in a representative experiment (1.5 mM SNP). 3013 by SNP, a donor of NO. The overproduction of NO in rheumatoid synovium and the ability of NO to induce apoptosis of human synoviocytes have been demonstrated by several groups (12,13). The effects of NO on VECs, however, are more complex. Physiologic levels of NO regulate vascular tone and protect microvasculature from injury, whereas higher NO concentrations are pathologic and promote VEC apoptosis both in vivo and in vitro (13). Therefore, in initial experiments, we determined the length of the exposure and the concentration of SNP at which a considerable proportion of cells become apoptotic but the numbers of dead cells are still low. Twenty-four–hour exposure to SNP at concentrations in the 1.5–2.0 mM range was found to be optimal for analysis of the antiapoptotic activities of IL-15. At these concentrations, after 24 hours of incubation, up to 25% of VECs showed staining characteristic of early apoptosis. We then tested rHuIL-15 for its ability to inhibit VEC apoptosis induced by 1–2 mM SNP. As shown in Figure 3, preincubation of HUVECs with rHuIL-15 at 50 ng/ml for 48 hours prior to addition of SNP resulted in a statistically significant reduction in the proportion of apoptotic cells. IL-15 inhibition of SNP-induced apoptosis of primary synovial fibroblasts in vitro. Similar results were obtained with primary fibroblast cultures derived from fresh JRA synovial tissue samples (data not shown). Preincubation of primary fibroblasts with rHuIL-15 at 50 ng/ml for 48 hours prior to addition of SNP resulted in a statistically significant reduction in the proportion of apoptotic primary synovial fibroblasts. Failure of IL-15 to up-regulate expression of VEGF in primary fibroblasts in vitro. Since proinflammatory cytokines such as TNF␣ and IL-1 exert their angiogenic effects through stimulation of VEGF production by primary fibroblasts (4), we were interested in whether IL-15 might have similar properties. Primary fibroblast cultures were established from a fresh synovial tissue sample, and the synovial cells were stimulated with rHuIL-15 at a final concentration 50 ng/ml for 48 hours. Three independent experiments were performed for each culture condition, and levels of expression of VEGF were assessed by RPA. No differences were noted between stimulated and unstimulated fibroblasts (data not shown). DISCUSSION In our evaluation of the angiogenic activity of IL-15, we showed that addition of IL-15 to the media promoted VEC survival and tube formation on Matrigel. This observation combined with the known antiapo- 3014 YANG ET AL ptotic effects of IL-15 on immune cells and fibroblasts led us to hypothesize that IL-15 might protect VECs from apoptotic death. Indeed, we found that preincubation of VECs with rHuIL-15 made them more resistant to SNP-induced apoptosis. We also demonstrated that rHuIL-15 had similar effects on synovial fibroblasts. We used the NO donor SNP as a proapoptotic agent for two reasons: first, NO has been shown to induce apoptosis of synovial fibroblasts in vitro, and second, overproduction of NO in RA synovium has been documented in many studies (12,13). The levels of NO in our experimental system are likely to be higher than the physiologic levels of NO in inflamed synovium. This certainly makes it difficult to extrapolate the findings to JRA pathogenesis. However, combined with the ability of IL-15 at physiologic concentrations to promote survival of VECs on Matrigel, our data suggest that the observed effects may be relevant to the in vivo situation. Fraser et al have recently demonstrated high levels of VEC apoptosis in RA synovium, suggesting increased turnover of these cells in vivo (15). Based on the present findings, we hypothesize that IL-15 may promote survival of endothelial cells in JRA synovium and thus be involved in the stabilization of newly formed blood vessels. REFERENCES 1. Firestein GS. Starving the synovium: angiogenesis and inflammation in rheumatoid arthritis. J Clin Invest 1999;103:3–4. 2. Peacock DJ, Banquerigo ML, Brahn E. Angiogenesis inhibition suppresses collagen arthritis. J Exp Med 1992;175:1135–8. 3. Yancopoulos GD, Davis S, Gale NW, Rudge JS, Wiegand SJ, Holash J. Vascular-specific growth factors and blood vessel formation. Nature 2000;407:242–8. 4. Paleolog EM, Young S, Stark AC, McCloskey RV, Feldmann M, Maini RN. Modulation of angiogenic vascular endothelial growth factor by tumor necrosis factor ␣ and interleukin-1 in rheumatoid arthritis. Arthritis Rheum 1998;41:1258–65. 5. Oppenheimer-Marks N, Brezinschek RI, Mohamadzadeh M, Vita R, Lipsky PE. IL-15 is produced by endothelial cells and increases the transendothelial migration of T cells in vitro and in the SCID mouse-human rheumatoid arthritis model. J Clin Invest 1998;101: 1261–72. 6. McInnes IB, al-Mughales J, Field M, Leung BP, Huang FP, Dixon R, et al. The role of IL-15 in T-cell migration and activation in rheumatoid arthritis. Nat Med 1996;2:175–82. 7. Scola MP, Thompson S, Brunner HI, Tsoras M, Witte D, van Dijk MA, et al. IFN gamma:IL-4 ratios and associated type 1 cytokine expression in JRA synovial tissue. J Rheumatol 2002;29:369–78. 8. Angiolillo AL, Kanegane H, Sgadari C, Reaman GH, Tosato G. Interleukin-15 promotes angiogenesis in vivo. Biochem Biophys Res Commun 1997;233:231–7. 9. Scola PM, Imagawa T, Boivin GP, Giannini EH, Glass DN, Hirsch R, et al. Expression of angiogenic factors in juvenile rheumatoid arthritis: correlation with revascularization of human synovium engrafted into SCID mice. Arthritis Rheum 2001;44:794–801. 10. Cassidy JT, Levinson JE, Bass JC, Baum J, Brewer EJ Jr, Fink CW, et al. A study of classification criteria for a diagnosis of juvenile rheumatoid arthritis. Arthritis Rheum 1986;29:274–81. 11. Blair RJ, Meng H, Marchese MJ, Ren S, Schwartz LB, Tonnesen MG, et al. Human mast cells stimulate vascular tube formation: tryptase is a novel, potent angiogenic factor. J Clin Invest 1997; 99:2691–700. 12. Aupperle KR, Boyle DL, Hendrix M, Seftor EA, Zvaifler NJ, Barbosa M, et al. Regulation of synoviocyte proliferation, apoptosis, and invasion by the p53 tumor suppressor gene. Am J Pathol 1998;152:1091–8. 13. Shen YH, Wang XL, Wilcken DE. Nitric oxide induces and inhibits apoptosis through different pathways. FEBS Lett 1998; 433:125–31. 14. Bulanova E, Budagian V, Pohl T, Krause H, Dürkop H, Paus R, et al. The IL-15R␣ chain signals through association with Syk in human B cells. J Immunol 2001;167:6292–302. 15. Fraser A, Fearon U, Reece R, Emery P, Veale DJ. Matrix metalloproteinase 9, apoptosis, and vascular morphology in early arthritis. Arthritis Rheum 2001;44:2024–8.