FLIP and MAPK play crucial roles in the MLN51-mediated hyperproliferation of fibroblast-like synoviocytes in the pathogenesis of rheumatoid arthritis Ju-Eun Ha 1 , Young-Eun Choi 1 , Jinah Jang 2 , Cheol-Hee Yoon 1 , Ho-Youn Kim 3 and Yong-Soo Bae 1,2 1 Department of Biological Science, Sungkyunkwan University, Suwon, Gyeonggi-do, South Korea 2 Division of DC Immunotherapy, CreaGene Research Institute, Seongnam-shi, Gyeonggi-do, South Korea 3 Department of Medicine, Division of Rheumatology, Center for Rheumatoid Diseases and Rheumatism Research Center (RhRC), Catholic Research Institutes of Medical Sciences, Catholic University of Korea, Seoul, South Korea Rheumatoid arthritis (RA) is a chronic inflammatory arthritis characterized by synovial hyperplasia with local invasion of bone and cartilage. Accumulating evi- dence suggests that RA fibroblast-like synoviocytes (FLSs) possess unique characteristics in RA patho- genesis [1]. FLSs play a key role in the development of sustained inflammation and angiogenesis in arthritic joints [2–4]. Several cytokines and RA factors existing in the RA environment stimulate the overgrowth of FLSs, leading to the aggravation of disease. Amongst these factors, granulocyte–macrophage colony-stimu- lating factor (GM-CSF) also plays an important role in the pathogenesis of RA [5–7]. GM-CSF blockade results in less severe disease and reduces cytokine levels in tissue in vivo [8]. In RA synovium, FLSs express both tumor necrosis factor-a (TNF-a) and Fas receptors, and their ligands are detected in nearby macrophages or T cells [9,10]. However, previous studies have demonstrated that Fas activation induces apoptosis in only a small proportion of FLSs, because of their constitutive expression of the FLICE-inhibitory protein (FLIP) [11]. FLIP expression mediates the recruitment and activation of nuclear fac- tor-jB kinase and several mitogen-activated protein Keywords FLICE-inhibitory protein; granulocyte– macrophage colony-stimulating factor; metastatic lymph node 51; mitogen- activated protein kinase; rheumatoid arthritis fibroblast-like synoviocyte Correspondence Y S. Bae, Department of Biological Science, Sungkyunkwan University, 300 Chunchun- dong, Jangan-gu, Suwon, Gyeonggi-do 440-746, South Korea Fax: +82 31 290 7087 Tel: +82 31 290 5911 E-mail: ysbae04@skku.edu (Received 23 January 2008, revised 5 April 2008, accepted 9 May 2008) doi:10.1111/j.1742-4658.2008.06500.x One of the characteristic features of the pathogenesis of rheumatoid arth- ritis is synovial hyperplasia. We have reported previously that metastatic lymph node 51 (MLN51) and granulocyte–macrophage colony-stimulating factor (GM-CSF) are involved in the proliferation of fibroblast-like synovi- ocytes in the pathogenesis of rheumatoid arthritis. In this study, we have found that: (1) GM-CSF-mediated MLN51 upregulation is attributable to both transcriptional and post-translational control in rheumatoid arthritis fibroblast-like synoviocytes; (2) p38 mitogen-activated protein kinase plays a key role in the upregulation of MLN51; and (3) FLICE-inhibitory pro- tein is upregulated downstream of MLN51 in response to GM-CSF, result- ing in the proliferation of fibroblast-like synoviocytes. These results imply that GM-CSF signaling activates mitogen-activated protein kinase, followed by the upregulation of MLN51 and FLICE-inhibitory protein, resulting in fibroblast-like synoviocyte hyperplasia in rheumatoid arthritis. Abbreviations DC, dendritic cell; ERK, extracellular signal-regulated kinase; FLIP, FLICE-inhibitory protein; FLS, fibroblast-like synoviocyte; JNK, c-Jun N-terminal kinase; MAPK, mitogen-activated protein kinase; MLN51, metastatic lymph node 51; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyl-tetrazolium bromide; OA, osteoarthritis; RA, rheumatoid arthritis; TNF-a, tumor necrosis factor-a. 3546 FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS kinases (MAPKs), leading to cell survival, proliferation or proinflammatory gene expression [12–14]. FLIP expression is also upregulated during the differentia- tion of monocytes into dendritic cells (DCs) [15]. In addition, GM-CSF is a key component in the differen- tiation of monocytes into DCs [16], and is generally detected in RA synovial fluid [7,17]. Taken together, these reports suggest that GM-CSF is probably associ- ated with FLIP expression. Our previous studies have demonstrated that active RA FLSs express substantial amounts of metastatic lymph node 51 (MLN51) in the presence of GM- CSF, and the upregulation of MLN51 is associated with the hyperproliferation of FLSs [7]. MLN51 was first identified in breast cancer cells. Later, it was reported that MLN51 is associated with the exon junction complex in the nucleus and remains stably associated with mRNA in the cytoplasm [18,19]. A recent study has found that MLN51 is essential for the formation of stress granules occurring in malig- nant tumors [20]. In the present study, we have investigated the mech- anism underlying the GM-CSF-mediated and MLN51- associated hyperproliferation of FLSs using an RA FLS cell line, MH7A [21]. We found that GM-CSF upregulates MLN51 expression through the activation of MAPK, followed by the induction of FLIP expres- sion. The present data strongly suggest that MLN51 and MLN51-induced FLIP play critical roles in FLS hyperplasia under GM-CSF conditions by facilitating cell proliferation and blocking apoptosis in the patho- genesis of RA. Results GM-CSF induces the proliferation of MH7A cells and the expression of MLN51 Our recent study has shown that GM-CSF is involved in the proliferation of primary RA FLSs [7]. In the present study, we investigated cell proli- feration and MLN51 expression using an RA FLS cell line, MH7A, in the presence of GM-CSF. As shown in Fig. 1A, MH7A cell proliferation was enhanced by GM-CSF treatment in a dose-dependent manner at concentrations up to 100 ngÆmL )1 , but 0 10 50 100 300 500 (ng·mL –1 ) MLN51 0 10 50 100 300 500 (ng·mL –1 ) MLN51 GM-CSF GM-CSF 0 0.25 0.5 1 1.5 2 6 12 24 (h) 0 0.25 0.5 1 1.5 2 6 12 24 (h) MLN51 MLN51 GM-CSF (ng·mL –1 ) 0.0 0.5 1.0 1.5 2.0 2.5 A B C 0 10 100 1000 Number of cells (×10 4 ) * WB RT-PCR hMLN51 β -actin β -actin β -actin β -actin β -actin 0 0.5 1 3 6 12 0 0.5 1 3 6 12 (h) RA FLS (#1) RA FLS (#2) GM-CSF Fig. 1. GM-CSF induces FLS proliferation and MLN51 expression. (A) MH7A cells were incubated with GM-CSF at various concentrations for 24 h. Cell proliferation was assessed by MTT assay. The results are expressed as the mean ± standard devi- ation in triplicate. *P < 0.01. (B) Dose kinet- ics (left) and time kinetics (right) of GM-CSF effects on MLN51 expression in MH7A cells. MH7A cells were treated with various concentrations of GM-CSF for 1 h for dose kinetics, and with 100 ngÆmL )1 of GM-CSF for different periods of time for time kinet- ics. MLN51 expression was then assessed by western blot (WB) analysis and RT-PCR at the protein (top) and mRNA (bottom) levels. (C) Time kinetics of GM-CSF effects on MLN51 expression in primary RA FLSs. Primary RA FLS cells were treated with 100 ngÆmL )1 of GM-CSF for the periods of time shown in the figure. MLN51 expres- sion was then assessed by RT-PCR. J E. Ha et al. FLIP and MAPK in MLN51-mediated FLS hyperproliferation FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS 3547 not at extreme concentrations such as 500 ngÆmL )1 (Fig. 1A). We have reported previously that MLN51 is upregu- lated in the FLSs of RA patients, and is enhanced 6 days after GM-CSF treatment [7]. In the present experiments, we assessed the dose and time kinetics of the effects of GM-CSF on MLN51 expression in MH7A cells. MLN51 was fully induced at the protein level fol- lowing treatment with GM-CSF at a concentration of 50–100 ngÆmL )1 (left panel in Fig. 1B) for 1–2 h (right panel in Fig. 1B). However, GM-CSF treatment did not affect the mRNA level of MLN51 over the entire range of kinetics in MH7A cells (Fig. 1B). These data suggest that GM-CSF-mediated MLN51 upregulation in MH7A cells is likely to depend on translational or post- translational control. However, in the case of primary RA FLSs, MLN51 expression was induced at both mRNA and protein levels within 12 h following GM- CSF treatment (Fig. 1C), suggesting that GM-CSF- mediated MLN51 upregulation in primary RA FLSs is, at least in part, dependent on transcriptional control. GM-CSF-mediated MLN51 upregulation is attributable to both transcriptional and post-translational control in RA FLSs As a next step, we investigated the control mecha- nism underlying the expression of MLN51 following GM-CSF treatment. When MH7A cells were pretreat- ed with a-amanitin or cycloheximide, the mRNA and protein levels, respectively, of MLN51 were signifi- cantly decreased compared with those of untreated cells (Fig. 2A,B). However, pretreatment with a-amani- tin (Fig. 2A) or cycloheximide (Fig. 2B) did not affect the GM-CSF-mediated upregulation of MLN51 in MH7A cells, suggesting that the GM-CSF-mediated upregulation of MLN51 in MH7A cells is not depen- dent on transcriptional or translational control. In contrast, GM-CSF-mediated MLN51 upregulation was completely obliterated and significant amounts of MLN51 were detected even in the absence of GM-CSF in MH7A cells following pretreatment with MG-132, a proteasome inhibitor (Fig. 2C). In order to confirm this result, we treated the MH7A cells with MG-132 in the presence of GM-CSF, and measured MLN51 pro- tein expression at different time points. In accordance with the results shown in Fig. 2C, once cells had been pretreated with MG-132, significant amounts of MLN51 were detected from the beginning, and were maintained for longer than 4 h without any additional effects of GM-CSF (Fig. 2D). However, in the case of RA FLSs, MG-132 pretreatment showed additive effects on the GM-CSF-mediated upregulation of MLN51 expression at the protein level (Fig. 2E). These findings suggest that GM-CSF-mediated MLN51 upregulation in primary RA FLSs is, to some extent, MG-132 0 0.5 1 4 0 0.5 1 4 (h) MLN51 GM-CSF WB MLN51 GM-CSF A B C E D MLN51 β -actin β -actin β -actin β -actin β -actin β -actin β -actin β -actin – + – + α -amanitin WB RT-PCR CHX GM-CSF – + – + MLN51 MLN51 WB RT - PCR MG-132 MLN51 MLN51 GM-CSF – + – + WB RT-PCR GM-CSF – + – + MG132 Relative band intensity GM-CSF – + – + MLN51 MG132 WB RA-FLS 0 5 10 15 20 25 30 Fig. 2. GM-CSF-induced MLN51 expression acts as a post-transcriptional regulator rather than a transcriptional or translational control. MLN51 expressed in MH7A cells was assessed in the presence or absence of 10 lgÆmL )1 of a-amanitin for 20 h (A), 25 lgÆmL )1 of cycloheximide (CHX) for 20 h (B) and 20 l M MG-132 for 6 h (C). The expression of MLN51 was assessed by RT-PCR and western blot (WB) analysis. (D) The expression of MLN51 was assessed via WB analysis after treatment with GM-CSF for the indicated periods of time in the pres- ence or absence of MG-132. (E) MLN51 expressed in primary RA FLSs was assessed in the presence or absence of 20 l M MG-132 for 6 h. The expression of MLN51 was assessed by WB analysis (left). The histogram (right) represents the relative band intensity of the WB data, assessed by IMAGE J software (http://rsb.info.nih.gov/ij). FLIP and MAPK in MLN51-mediated FLS hyperproliferation J E. Ha et al. 3548 FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS dependent on both post-translational and transcrip- tional control, probably by blocking of the proteasome degradation pathway. GM-CSF-induced MLN51 is involved in the hyperproliferation and anti-apoptosis of MH7A cells via the upregulation of FLIP Next, we investigated the contribution of MLN51 to MH7A cell proliferation. When MLN51 was knocked down by transfection of siRNA (si-MLN51), cell pro- liferation was completely abrogated, whereas the over- expression of MLN51 enhanced MH7A cell proliferation (Fig. 3A). These results strongly suggest that MLN51 plays a critical role in the proliferation of MH7A cells. In order to determine whether MLN51 is involved in the anti-apoptosis as well as cell proliferation of MH7A in response to GM-CSF, we examined the apoptosis of normal and MLN51-knockdown MH7A cells in the presence or absence of GM-CSF. As shown in Fig. 3B, cell apoptosis was substantially increased by MLN51-knockdown, and the increased apoptosis was not attenuated by additional GM-CSF treatment. These data suggest that MLN51 is also involved in anti-apoptosis. Amongst the several anti-apoptotic molecules, FLIP mRNA was markedly enhanced by MLN51 over- expression in MH7A cells (Fig. 3C). Once treated with GM-CSF, FLIP mRNA was also increased, together with MLN51 mRNA, in primary RA FLSs, but was undetectable in osteoarthritis (OA) FLSs even in the presence of GM-CSF (Fig. 3D). Transient expression of MLN51 induced the expression of FLIP (Fig. 3E), whereas MLN51-knockdown attenuated FLIP expres- sion in MH7A cells regardless of GM-CSF treatment (Fig. 3F). These results indicate that MLN51 causes the upregulation of FLIP expression, followed by the blocking of FLS apoptosis. FLIP upregulated by MLN51 plays a crucial role in the anti-apoptosis of MH7A cells We examined whether FLIP was involved in the anti- apoptosis of MH7A cells. As shown in other cells [22], FLIP-knockdown (si-FLIP) completely abrogated MH7A cell proliferation, even in the presence of GM- CSF, when compared with control cells (si-con) (Fig. 4A). In addition, FLIP-knockdown markedly increased the cell apoptosis of MH7A, and the apopto- tic ratio was not attenuated by GM-CSF treatment (Fig. 4B). These data suggest that FLIP plays an important role in GM-CSF-mediated cell proliferation and anti-apoptosis. In contrast, FLIP-knockdown did not show any discernible effects on the expression of MLN51 (Fig. 4C), implying that FLIP works down- stream of MLN51 in the GM-CSF-mediated signaling pathway to FLS proliferation. MAPK functions in the upregulation of MLN51 under GM-CSF conditions Activation of the GM-CSF receptor leads to the acti- vation of multiple cytoplasmic signaling molecules, including MAPK. The MAPKs are key regulators of cytokine and metalloproteinase production, and thus may be targeted in RA. It has been reported previ- ously that all three MAPK families, extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK) and p38, are expressed in rheumatoid synovial tissue, and also play a key role in RA FLS activation [23]. We investigated whether or not GM- CSF activates MAPK in MH7A cells. Following the addition of GM-CSF to cultures of MH7A cells, JNK and p38 were dramatically phosphorylated within approximately 1 h, whereas ERK phosphoryla- tion was slightly enhanced during the same period of time (Fig. 5A). In good accordance with the data shown in Fig. 5A, pretreatment of MH7A cells with SB203580 (p38 inhibitor) completely abrogated the effects of GM-CSF on the upregulation of MLN51, whereas pretreatment with SP600125 (JNK inhibitor) or PD98059 (ERK inhibitor) partially or barely atten- uated the effects of GM-CSF on the expression of MLN51 and FLIP, respectively (Fig. 5B). These data indicate that MAPKs, particularly p38 and partly JNK, but not ERK, play an important role in the upregulation of MLN51 and FLS overgrowth upstream of MLN51 under the GM-CSF signaling pathway, as summarized in Fig. 6. Discussion Inflammatory cell infiltration and the expansion of an aggressive FLS population in the synovial membrane are the pathological hallmarks of RA [1,24]. A number of growth factors and cytokines have been described in association with the proliferative response of FLSs, including transforming growth factor-b, platelet- derived growth factor, fibroblast growth factor, inter- leukin-1b, TNF-a and interleukin-6. However, these factors are not sufficient to cover the active prolifera- tion capacity of RA FLSs, thus indicating that other factors must be involved in this proliferation. In our previous studies, we have determined that GM-CSF in the synovial fluid plays an important role in the hyper- J E. Ha et al. FLIP and MAPK in MLN51-mediated FLS hyperproliferation FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS 3549 proliferation of RA FLSs through the upregulation of MLN51 [7]. In the present study, we have investigated the mechanism underlying the GM-CSF-mediated and MLN51-associated hyperproliferation of FLSs using an RA FLS cell line, MH7A. As shown previously with primary RA FLSs, GM-CSF treatment increased the number of MH7A cells in culture, as well as the expression of MLN51 in these cells (Fig. 1). MLN51 Bcl2 c-IAP x-IAP NF K B (p65) NF K B (p50) FLIP Control pcDNA3.1 pcDNA-MLN51 GM-CSF – + – + si-con si-MLN51 MLN51 FLIP MLN51 FLIP si- con pcDNA3.1 si - MLN51 pcDNA-MLN51 * * * 2.5 AB CD EF 2.0 1.5 1.0 0.5 0.0 Cell number (×10 4 ) pcDNA3.1 -hMLN51 Untreated GM-CSF MLN51 β-actin MLN51 -siRNA con -siRNA β -actin β -actin β -actin β -actin 11.05 si-con 25.81 si-MLN51 + + GM-CSF Annexin V 26.52 si-MLN51 hMLN51 FLIP GM-CSF – + – + – + RA FLS (#1) RA FLS (#2) OA FLS (#1) Fig. 3. MLN51 induces FLIP expression in the GM-CSF-mediated proliferation of FLSs. (A) MH7A cells were transfected with 200 pmol siR- NA against MLN51 (si-MLN51) or with 3 lg of pcDNA3.1-MLN51 plasmids using Lipofectamine 2000. One day later, the cells were stimu- lated with GM-CSF for 24 h or were left unstimulated. Cell numbers were assessed by MTT assay, and expressed as the mean ± standard deviation in triplicate. *P < 0.01. (B) MH7A cells transfected with 200 pmol of MLN51 siRNA (si-MLN51) or non-targeting siRNA (si-con) were stimulated with 100 ngÆmL )1 of GM-CSF for 24 h, or were left unstimulated. Apoptosis of each sample was assessed by flow cytome- try after Annexin V–FITC staining. (C) MH7A cells were transfected with 1 lg of mock vector or pcDNA3.1-MLN51 plasmids. The levels of several anti-apoptotic gene mRNAs were assessed by semi-quantitative RT-PCR with specific PCR primer sets. (D) Primary RA and OA FLS samples were cultured for 6 h in the presence or absence of GM-CSF. The mRNAs of MLN51 and FLIP were assessed by semi-quantitative RT-PCR with specific PCR primer sets. (E) MH7A cells transfected with 1 lg of pcDNA3.1-MLN51 or 200 pmol MLN51 siRNA were har- vested at 24 h post-transfection, and subjected to western blot analysis for the expression of MLN51and FLIP. (F) MH7A cells transfected with control (si-con) or MLN51 (si-MLN51) siRNAs were treated with 100 ngÆmL )1 of GM-CSF, or were left untreated. The cells were har- vested and assessed by western blot analysis for the expression of MLN51and FLIP. FLIP and MAPK in MLN51-mediated FLS hyperproliferation J E. Ha et al. 3550 FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS In our previous paper [7], we examined the mRNA and protein levels of MLN51 6 days after GM-CSF treatment of a culture of RA FLSs. In the present study, however, MLN51 expression was examined within 24 h after GM-CSF treatment in MH7A cells at both mRNA and protein levels. In the case of MH7A cells, MLN51 was constitutively expressed at the mRNA level under normal conditions, and was not changed by GM-CSF treatment over 24 h. In con- trast, the protein level of MLN51 was low in the untreated control, but rapidly increased over 1–2 h following GM-CSF treatment, and the enhanced level lasted longer than 24 h. When the cells were pretreated with MG-132, the protein level of MLN51 was as high as that of GM-CSF-treated cells, even in the absence of GM-CSF, suggesting that GM-CSF-mediated MLN51 upregulation in MH7A cells is probably post- translational. However, when examined in primary RA FLSs over 12 h, both mRNA and protein levels of MLN51 were enhanced at 3–12 h following GM-CSF treatment (Fig. 1C). The pretreatment of RA FLSs with MG-132 showed an additional increment in the protein level of GM-CSF-induced MLN51 (Fig. 2E). These data indi- cate that GM-CSF not only induces the expression of MLN51, but also blocks the proteasome-mediated degradation of MLN51 in RA FLSs by an un- known mechanism. In other words, GM-CSF-mediated MLN51 upregulation is attributable to both transcrip- tional and post-translational control in RA FLSs. The overgrowth of RA FLSs may result from unbal- anced proliferation and apoptosis, and both processes have been detected on tissue sections of rheumatoid synovium [25,26]. In the present study, the MLN51- knockdown and ectopic expression of MLN51 (Fig. 3A,B) experiments have shown that MLN51 plays an important role in GM-CSF-mediated MH7A cell proliferation. In order to determine whether MLN51 is also involved in anti-apoptosis, we investigated the anti-apoptotic molecules, and found that FLIP expres- sion was upregulated by MLN51 (Fig. 3C,D). MLN51 is a subunit of the exon junction complex, which is involved in post-splicing events, such as mRNA export, nonsense-mediated mRNA decay and translation [27–29]. Taken together, these findings allow us to assume that MLN51 may facilitate the export of FLIP mRNA from the nucleus, or stabilize FLIP mRNA in the cytoplasm, followed by the blocking of cell apopto- sis, and therefore involvement in FLS overgrowth. Apoptosis stimulators, such as TNF-a and FasL, nor- mally induce cell apoptosis. In RA FLSs, however, activation of the TNF-a receptor or Fas receptor 0.0 0.5 1.0 1.5 2.0 si-con A B C si-FLIP Untreated GM-CSF Cell number (×10 4 ) * si-con si-FLIP + GM-CSF si-FLIP Annexin V 46.92 14.06 44.57 si-con si-FLIP GM-CSF –+ –+ MLN51 β -actin FLIP Fig. 4. MAPK and FLIP are involved in the GM-CSF-mediated FLS proliferation mechanism upstream and downstream of MLN51, respectively. (A) MH7A cells transfected with 100 pmol of FLIP (si-FLIP) or control (si-con) siRNAs were stimulated with 100 ngÆmL )1 of GM-CSF 24 h post-transfection, or were left unstimulated. At 24 h after treatment, cell numbers were assessed by MTT assay. The results are expressed as the mean ± standard deviation in triplicate. *P < 0.01. (B) Control (si-con) or FLIP si-RNA- transfected (si-FLIP) cells were treated with 100 ngÆmL )1 of GM-CSF for 24 h, or were left untreated. They were assessed for cell apoptosis by flow cytometry after Annexin V–FITC staining. (C) Control (si-con) or FLIP si-RNA-transfected (si-FLIP) cells were trea- ted with 100 ngÆmL )1 of GM-CSF for 24 h, or were left untreated. FLIP and MLN51 expression was assessed in the cells by western blot analysis. J E. Ha et al. FLIP and MAPK in MLN51-mediated FLS hyperproliferation FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS 3551 induces NF-jB translocation, which leads to increased FLIP expression [30]. This NF-jB loop may protect RA FLSs from TNF-a ⁄ FasL-mediated cell death, resulting in FLS overgrowth. However, we found that MLN51 induced FLIP expression in the absence of TNF-a or FasL stimulation. These data suggest that RA FLSs are resistant to cell apoptosis via, at least in part, MLN51-mediated FLIP upregulation under GM-CSF conditions. The activation of MAPK is almost exclusively found in synovial tissue from RA patients. This activation is induced by inflammatory cytokines [23]. Amongst these proinflammatory cytokines, GM-CSF induces phosphorylation of Ser345 in the MAPK consensus sequence [31]. We have found that GM-CSF induces the phos- phorylation of p38 and JNK predominantly (Fig. 5A), and that a p38 inhibitor (SB203580) com- pletely abrogates the GM-CSF-mediated upregulation of MLN51 and FLIP in MH7A cells (Fig. 5B). Although ERK (p42 ⁄ 44) is constitutively activated in MH7A cells (Fig. 5A), as reported previously [21], ERK inhibitor (PD98059) does not affect MLN51 and FLIP induction in MH7A cells (Fig. 5B), indi- cating that ERK is unlikely to be involved in the GM-CSF-mediated induction of MLN51 and FLIP in RA FLSs. These data suggest that MAPK, partic- ularly p38, is activated by GM-CSF, and plays an important role in the post-translational modification of MLN51, resulting in the protection of MLN51 from ubiquitin-mediated degradation. In summary, in RA FLSs: (1) GM-CSF signaling activates p38 MAPK; (2) this is followed by MLN51 upregulation via both transcriptional and post-transla- tional control; (3) FLIP expression is induced; and (4) this results in the anti-apoptotic proliferation of FLS, contributing to the pathogenesis of RA (Fig. 6). MLN51 and MLN51-induced FLIP are believed to play important roles in FLS hyperplasia by participat- ing in FLS proliferation and anti-apoptosis in RA pathogenesis. Thus, MLN51 and FLIP are attractive targets for the development of new RA therapeutics. Experimental procedures Isolation and establishment of RA FLSs from RA patients Fibroblast-like synoviocyte samples were obtained from synovectomized tissue of RA and OA patients undergoing Fig. 6. Summary diagram showing the role of MLN51 in the GM-CSF-mediated proliferation of RA FLSs via MAPK activation and induction of FLIP expression. A B Fig. 5. MLN51 expression via MAPK activation by GM-CSF. (A) MH7A cells were treated with 100 ngÆmL )1 of GM-CSF for the indi- cated periods of time. Cells were assessed by western blotting for the expression of three different MAPKs and their phospho-forms. (B) MH7A cells were pre-incubated with 20 l M SP600125, 50 lM PD98059 and 20 lM SB203580 for 1 h, and were then cultured in the presence or absence of GM-CSF (100 ngÆmL )1 ) for an additional 1 h. MLN51 and FLIP expression in each sample was assayed by western blot analysis. DMSO, dimethylsulfoxide. FLIP and MAPK in MLN51-mediated FLS hyperproliferation J E. Ha et al. 3552 FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS joint replacement surgery at Kangnam St Mary Hospital, Catholic University of Korea, Seoul, South Korea. Institu- tional Review Board (IRB) approval and patient informed consent from each enrolled participant were obtained. RA and OA FLS cells were prepared as described previously [7]. Cell line, chemicals and antibodies A human synovial cell line (MH7A), which was prepared from FLSs isolated from the knee joint of an RA patient, was obtained from Riken Cell Bank, Tsukuba, Ibaraki, Japan. MH7A cells were maintained in RPMI1640 (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY, USA) and 100 lgÆmL )1 each of penicillin and streptomycin. GM-CSF (LG Life Science, Seoul, Korea), cycloheximide (Calbiochem, San Diego, CA, USA), a-amanitin (Sigma, St Louis, MO, USA), MAPK inhibitors SP600125, PD98059, SB203580 (Calbiochem) and MG-132 (AG Scien- tific Inc., San Diego, CA, USA) were used in the present experiments. FLIP antibodies were purchased from Santa Cruz Co. (Santa Cruz, CA, USA). SAPT ⁄ JNK, phospho- SAPT ⁄ JNK, ERK, phospho-ERK, p38 and phospho-p38 antibodies were purchased from Cell Signaling Inc. (Dan- vers, MA, USA). Anti-b-actin (Sigma), anti-rabbit and anti-mouse IgG-HRP (Sigma) and Annexin V–fluorescein isothiocyanate (Becton Dickinson, Mountain View, CA, USA) IgG were also used in this study. Recombinant plasmids MLN51-expressing plasmids (pcDNA3.1-MLN51 and pET28-MLN51) were prepared by cloning full-length hMLN51 cDNA into pcDNA3.1 (Invitrogen, San Diego, CA, USA) at the EcoRI ⁄ XhoI site and partial hMLN51 cDNA into the pET28a(+) vector (Novagen, Madison, WI, USA) at the HindIII ⁄ XhoI site, respectively. Full-length and partial cDNAs of hMLN51 [18] were prepared by RT-PCR amplification of MH7A mRNA using appropriate primer pairs for cDNAs of hMLN51: full-length, 5¢-TATG AATTCGTTCTCCGTAAGATGGCGGAC-3¢ and 5¢-TA TCTCGAGTTAACTGGAACCCCTGCTTACAA-3¢; par- tial length, 5¢-ATCAAGCTTTGGTGCGTAAGGAGCT GAC-3¢ and 5¢-ATACTCGAGCTTAGCAGCTGGAGTC GTTT-3¢. Preparation of recombinant MLN51 protein and antiserum Recombinant BL21(DE3) cells that had been transformed by pET28a(+)-MLN51 were cultured in 2· yeast extract and tryptophan medium. Recombinant proteins were purified using a nickel nitrilotriacetic acid-conjugated bead column. MLN51 antiserum was prepared by immu- nizing New Zealand white rabbits three times at 3-week intervals with recombinant hMLN51 proteins emulsified with Freund’s adjuvant (Sigma). New Zealand white rab- bits were obtained from Orient Bio (Gyeonggi-do, South Korea) and were maintained in the Animal Care Facility of Sungkyunkwan University according to the Korean Experimental Animal Care Guidelines. Cell proliferation assay MH7A cells were seeded in 96-well plates overnight at a density of (1–5) · 10 3 cells per well in 100 lL of RPMI1640 containing 10% fetal bovine serum. Cells that had been pretreated with appropriate reagents or transfected with siRNA were cultured in the presence or absence of various concentrations of GM-CSF. After 24 h, 3-(4,5-dim- ethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) dye solution (20 lL per well, Promega, Madison, WI, USA) was added to each well, and incubated for another 4 h at 37 °C. The reaction was stopped by the addition of stop solution (150 lL per well), and the absorbance of each sample was subsequently measured by a spectrophotometer (Molecular Device, Union City, CA, USA) at 570 nm. Western blot analysis Each experiment was conducted as described previously [7]. In brief, cell lysates were normalized with Bradford reagent (Bio-Rad, Hercules, CA, USA), and 40–70 lg of lysate was subjected to 8–12% SDS-PAGE and transferred to poly(vinylidene difluoride) membranes (Millipore, Eschborn, Germany). The membranes were blocked and probed with appropriate antibody as described previously [7], and were then analyzed by an enhanced chemiluminescence western blotting system (Millipore ⁄ Amersham Biosciences, Freiburg, Germany). Quantitative RT-PCR Quantitative RT-PCR was conducted as described previously [7,32]. In brief, total RNAs were extracted using Tri-zol reagent (Invitrogen) and were then normalized. RT-PCR was conducted using the pre-Mix kit (Intron Biotech, Seoul, Korea) and the following primer pairs: MLN51: sense, 5¢-AAGACACCGAGGACGAGGAATC-3¢; anti-sense, 5¢-CCTTCCATAGCTTTCGCTGACG-3¢; FLIP: sense, 5¢-GAATGTGGAATTCAAGGCTCA-3¢; anti-sense, 5¢-AT ACAGGTACCCACACCCACA-3¢; Bcl-2: sense, 5¢-TTC CTCTGGGAAGGATGGCG-3¢; anti-sense, 5¢-CGTCCC TGAAGAAGCTCCTCC-3¢; IAP: sense, 5¢-TGTTGTGGC CTGATGCTGGA-3¢; anti-sense, 5¢-CAGGCAAAGCAAG CCACTCTG-3¢; XIAP: sense, 5¢-TGGTGACCAAGTGC AGTGCT-3¢; anti-sense, 5¢-AGGGTCTTCACTGGGCTT J E. Ha et al. FLIP and MAPK in MLN51-mediated FLS hyperproliferation FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS 3553 CC-3¢; NF-kB(p50): sense, 5¢-AGTTTCGGCGGTGGT AGTGG-3¢; anti-sense, 5¢-GCCAGCAGCATCTTCACG TC-3¢; NF-kB(p65): sense, 5¢-GACAATCGTGCCCCCAA CAC-3¢; anti-sense, 5¢-TGGGTCCGCTGAAAGGACT-3¢; human b-actin: sense, 5¢-TGACGGGGTCACCCACACT GTGCCCATCTA-3¢; anti-sense, 5¢-AGTCATAGTCCGC CTAGAAGCATTTFCGGT-3¢. siRNA transfection Human MLN51 and c-FLIP siRNAs were designed and synthesized by Invitrogen (StealthÔ) with sequences of 5¢-GGGCCCUAAGCAUUUGGAUGAUGAU-3¢ and 5 ¢-CC CUGGGCUAUGAAGUCCAGAAAUU-3¢, respectively. Cell transfection with siRNA was conducted using Lipo- fectamine 2000 (Invitrogen) according to the protocol of the manufacturer. After 5 h of incubation, the media were completely replaced and incubated further. Apoptotic analysis by flow cytometry MH7A cells were cultured in six-well plates at 5 · 10 5 cells per well. After 24 h, cells were transfected with MLN51 or FLIP siRNA, and cultured in the presence or absence of 100 ngÆmL )1 of GM-CSF at 37 °C for an additional 24 h. The cells were harvested and incubated for 15 min with fluorescein isothiocyanate-conjugated Annexin V antibody (Becton Dickinson) at room temperature in the dark. The cells were then analyzed using a FACS Calibur system (Becton Dickinson) with cell quest software. Acknowledgement This work was supported by Bio New Drug grants (A060115 and A040010) from the Korean Ministry of Health and Welfare. References 1 Firestein GS (2003) Evolving concepts of rheumatoid arthritis. Nature 423, 356–361. 2 Smith RS, Smith TJ, Blieden TM & Phipps RP (1997) Fibroblasts as sentinel cells. Synthesis of chemokines and regulation of inflammation. Am J Pathol 151, 317– 322. 3 Pap T, Muller-Ladner U, Gay RE & Gay S (2000) Fibroblast biology. Role of synovial fibroblasts in the pathogenesis of rheumatoid arthritis. Arthritis Res 2, 361–367. 4 Paleolog EM & Miotla JM (1998) Angiogenesis in arthritis: role in disease pathogenesis and as a potential therapeutic target. Angiogenesis 2, 295–307. 5 Alvaro-Gracia JM, Zvaifler NJ & Firestein GS (1989) Cytokines in chronic inflammatory arthritis. IV. Granu- locyte ⁄ macrophage colony-stimulating factor-mediated induction of class II MHC antigen on human mono- cytes: a possible role in rheumatoid arthritis. J Exp Med 170, 865–875. 6 Metcalf D (1989) The molecular control of cell division, differentiation commitment and maturation in haemo- poietic cells. Nature 339, 27–30. 7 Jang J, Lim DS, Choi YE, Jeong Y, Yoo SA, Kim WU & Bae YS (2006) MLN51 and GM-CSF involvement in the proliferation of fibroblast-like synoviocytes in the pathogenesis of rheumatoid arthritis. Arthritis Res Ther 8, R170. 8 Cook AD, Braine EL, Campbell IK, Rich MJ & Hamil- ton JA (2001) Blockade of collagen-induced arthritis post-onset by antibody to granulocyte–macrophage col- ony-stimulating factor (GM-CSF): requirement for GM-CSF in the effector phase of disease. Arthritis Res 3, 293–298. 9 Deleuran BW, Chu CQ, Field M, Brennan FM, Mitch- ell T, Feldmann M & Maini RN (1992) Localization of tumor necrosis factor receptors in the synovial tissue and cartilage–pannus junction in patients with rheuma- toid arthritis. Implications for local actions of tumor necrosis factor alpha. Arthritis Rheum 35, 1170–1178. 10 Asahara H, Hasumuna T, Kobata T, Yagita H, Okum- ura K, Inoue H, Gay S, Sumida T & Nishioka K (1996) Expression of Fas antigen and Fas ligand in the rheumatoid synovial tissue. Clin Immunol Immunopathol 81, 27–34. 11 Palao G, Santiago B, Galindo M, Paya M, Ramirez JC & Pablos JL (2004) Down-regulation of FLIP sensitizes rheumatoid synovial fibroblasts to Fas-mediated apop- tosis. Arthritis Rheum 50, 2803–2810. 12 Kataoka T, Budd RC, Holler N, Thome M, Martinon F, Irmler M, Burns K, Hahne M, Kennedy N, Kovac- sovics M et al. (2000) The caspase-8 inhibitor FLIP promotes activation of NF-kappaB and Erk signaling pathways. Curr Biol 10, 640–648. 13 Chaudhary PM, Eby MT, Jasmin A, Kumar A, Liu L & Hood L (2000) Activation of the NF-kappaB pathway by caspase 8 and its homologs. Oncogene 19, 4451–4460. 14 Imamura R, Konaka K, Matsumoto N, Hasegawa M, Fukui M, Mukaida N, Kinoshita T & Suda T (2004) Fas ligand induces cell-autonomous NF-kappaB activa- tion and interleukin-8 production by a mechanism dis- tinct from that of tumor necrosis factor-alpha. J Biol Chem 279, 46415–46423. 15 Willems F, Amraoui Z, Vanderheyde N, Verhasselt V, Aksoy E, Scaffidi C, Peter ME, Krammer PH & Gold- man M (2000) Expression of c-FLIP(L) and resistance to CD95-mediated apoptosis of monocyte-derived den- dritic cells: inhibition by bisindolylmaleimide. Blood 95, 3478–3482. FLIP and MAPK in MLN51-mediated FLS hyperproliferation J E. Ha et al. 3554 FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS 16 Sallusto F & Lanzavecchia A (1994) Efficient presenta- tion of soluble antigen by cultured human dendritic cells is maintained by granulocyte ⁄ macrophage colony-stimu- lating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J Exp Med 179, 1109–1118. 17 Alvaro-Gracia JM, Zvaifler NJ, Brown CB, Kaushan- sky K & Firestein GS (1991) Cytokines in chronic inflammatory arthritis. VI. Analysis of the synovial cells involved in granulocyte–macrophage colony-stimulating factor production and gene expression in rheumatoid arthritis and its regulation by IL-1 and tumor necrosis factor-alpha. J Immunol 146 , 3365–3371. 18 Degot S, Regnier CH, Wendling C, Chenard MP, Rio MC & Tomasetto C (2002) Metastatic lymph node 51, a novel nucleo-cytoplasmic protein overexpressed in breast cancer. Oncogene 21, 4422–4434. 19 Degot S, Le Hir H, Alpy F, Kedinger V, Stoll I, Wen- dling C, Seraphin B, Rio MC & Tomasetto C (2004) Association of the breast cancer protein MLN51 with the exon junction complex via its speckle localizer and RNA binding module. J Biol Chem 279, 33702–33715. 20 Baguet A, Degot S, Cougot N, Bertrand E, Chenard MP, Wendling C, Kessler P, Le Hir H, Rio MC & Tomasetto C (2007) The exon-junction-complex- component metastatic lymph node 51 functions in stress-granule assembly. J Cell Sci 120, 2774–2784. 21 Miyazawa K, Mori A & Okudaira H (1998) Establish- ment and characterization of a novel human rheuma- toid fibroblast-like synoviocyte line, MH7A, immortalized with SV40 T antigen. J Biochem 124, 1153–1162. 22 Wilson TR, McLaughlin KM, McEwan M, Sakai H, Rogers KM, Redmond KM, Johnston PG & Longley DB (2007) c-FLIP: a key regulator of colorectal cancer cell death. Cancer Res 67, 5754–5762. 23 Schett G, Tohidast-Akrad M, Smolen JS, Schmid BJ, Steiner CW, Bitzan P, Zenz P, Redlich K, Xu Q & Stei- ner G (2000) Activation, differential localization, and regulation of the stress-activated protein kinases, extra- cellular signal-regulated kinase, c-JUN N-terminal kinase, and p38 mitogen-activated protein kinase, in synovial tissue and cells in rheumatoid arthritis. Arthri- tis Rheum 43, 2501–2512. 24 Harris ED Jr (1990) Rheumatoid arthritis. Pathophysi- ology and implications for therapy. N Engl J Med 322, 1277–1289. 25 Aupperle KR, Boyle DL, Hendrix M, Seftor EA, Zvai- fler NJ, Barbosa M & Firestein GS (1998) Regulation of synoviocyte proliferation, apoptosis, and invasion by the p53 tumor suppressor gene. Am J Pathol 152, 1091– 1098. 26 Lories RJ, Derese I, Ceuppens JL & Luyten FP (2003) Bone morphogenetic proteins 2 and 6, expressed in arthritic synovium, are regulated by pro- inflammatory cytokines and differentially modulate fibroblast-like synoviocyte apoptosis. Arthritis Rheum 48, 2807–2818. 27 Ballut L, Marchadier B, Baguet A, Tomasetto C, Seraphin B & Le Hir H (2005) The exon junction core complex is locked onto RNA by inhibition of eIF4AIII ATPase activity. Nat Struct Mol Biol 12, 861–869. 28 Shibuya T, Tange TO, Stroupe ME & Moore MJ (2006) Mutational analysis of human eIF4AIII identifies regions necessary for exon junction complex formation and nonsense-mediated mRNA decay. RNA 12, 360– 374. 29 Tange TO, Shibuya T, Jurica MS & Moore MJ (2005) Biochemical analysis of the EJC reveals two new factors and a stable tetrameric protein core. RNA 11, 1869– 1883. 30 Ahn JH, Park SM, Cho HS, Lee MS, Yoon JB, Vilcek J & Lee TH (2001) Non-apoptotic signaling pathways activated by soluble Fas ligand in serum-starved human fibroblasts. Mitogen-activated protein kinases and NF- kappaB-dependent gene expression. J Biol Chem 276, 47100–47106. 31 Dang PM, Stensballe A, Boussetta T, Raad H, Dewas C, Kroviarski Y, Hayem G, Jensen ON, Gougerot- Pocidalo MA & El-Benna J (2006) A specific p47phox- serine phosphorylated by convergent MAPKs mediates neutrophil NADPH oxidase priming at inflammatory sites. J Clin Invest 116, 2033–2043. 32 Lee ES, Yoon CH, Kim YS & Bae YS (2007) The dou- ble-strand RNA-dependent protein kinase PKR plays a significant role in a sustained ER stress-induced apopto- sis. FEBS Lett 581, 4325–4332. J E. Ha et al. FLIP and MAPK in MLN51-mediated FLS hyperproliferation FEBS Journal 275 (2008) 3546–3555 ª 2008 The Authors Journal compilation ª 2008 FEBS 3555 . FLIP and MAPK play crucial roles in the MLN51-mediated hyperproliferation of fibroblast-like synoviocytes in the pathogenesis of rheumatoid arthritis Ju-Eun. (2006) MLN51 and GM-CSF involvement in the proliferation of fibroblast-like synoviocytes in the pathogenesis of rheumatoid arthritis. Arthritis Res Ther 8, R170. 8