Eur. J. Immunol. 2014. 44: 2721–2736
The xanthine oxidase–NFAT5 pathway regulates macrophage activation and TLR-induced inflammatory arthritis Nam-Hoon Kim ∗1 , Susanna Choi ∗1 , Eun-Jin Han1 , Bong-Ki Hong1 , Soo Youn Choi2 , H. Moo Kwon2 , Sue-Yun Hwang3 , Chul-Soo Cho1,4 and Wan-Uk Kim1,4 1
POSTECH-CATHOLIC BioMedical Engineering Institute, The Catholic University of Korea, Seoul, Korea 2 School of Nano-Bioscience and Chemical Engineering, Ulsan National Institute of Science and Technology, Ulsan, Korea 3 Department of Animal Biotechnology, Graduate School of Biology and Information Technology, Hankyoung National University, Ansung, Kyunggi-Do, Korea 4 Division of Rheumatology, Department of Internal Medicine, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea NFAT5 (nuclear factor of activated T cells), a well-known osmoprotective factor, can be activated by isotonic stimuli such as Toll-like receptor (TLR) triggering. However, it is unclear how NFAT5 discriminates between isotonic and hypertonic stimuli to produce different functional and molecular outcomes. Here, we identified a novel XO–ROS–p38 MAPK–NFAT5 pathway (XO is xanthine oxidase, ROS is reactive oxygen species) that is activated in RAW 264.7 macrophages upon isotonic TLR stimulation. Unlike what is seen under hypertonic conditions, XO-derived ROS were selectively required for the TLR-induced NFAT5 activation and NFAT5 binding to the IL-6 promoter in RAW 264.7 macrophages under isotonic conditions. In mouse peritoneal macrophages and human macrophages, TLR ligation also induced NFAT5 activation, which was dependent on XO and p38 kinase. The involvement of XO in NFAT5 activation by TLR was confirmed in RAW 264.7 macrophages implanted in BALB/c mice. Moreover, allopurinol, an XO inhibitor, suppressed arthritis severity and decreased the expression of NFAT5 and IL-6 in splenic macrophages in C57BL/6 mice. Collectively, these data support a novel function of the XO–NFAT5 axis in macrophage activation and TLR-induced arthritis, and suggest that XO inhibitor(s) could serve as a therapeutic agent for chronic inflammatory arthritis.
Keywords: Inflammatory arthritis Xanthine oxidase
Reactive oxygen species
Additional supporting information may be found in the online version of this article at the publisher’s web-site
Introduction The nuclear factor of activated T cells (NFAT) family is a group of five versatile transcription factors that are crucial to the devel-
Correspondence: Dr. Wan-Uk Kim e-mail: [email protected]
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
opment and function of the immune system. Among the five members of the NFAT family (NFAT1–5), NFAT5 is unique in that it is the only protein that responds to hypertonicity . NFAT5 is very similar to a Drosophila transcription factor involved
These authors contributed equally to this work.
Nam-Hoon Kim et al.
in the defense against hypertonicity , indicating that it may represent an ancient prototype among members of the NFAT family. Indeed, the ability to guard cells against osmotic damage is commonly required in invertebrates as well as in vertebrates including humans. Accordingly, the role of NFAT5 is most pronounced in tissues constantly exposed to osmotic stress, such as kidney medulla epithelia, intestine epithelia, skin epidermis, and cornea . Interestingly, NFAT5 is also activated in nonhypertonic conditions. Biologic processes that accompany NFAT5 activation include integrin-mediated carcinoma invasion , migration and differentiation of myoblasts , and T-cell proliferation . Additionally, we have demonstrated for the first time that NFAT5 activity is increased by tumor necrosis factor alpha (TNF-α) and interleukin (IL) 1β stimulation under isotonic conditions . NFAT5-deficient mice show a marked reduction of Ab-induced arthritis suggesting that NFAT5 is a critical regulator of rheumatoid arthritis (RA) . Moreover, NFAT5 has been recently shown to participate in innate immunity by activating gene expressions in the macrophage upon Toll-like receptor (TLR) ligation, independent of hypertonicity . Thus, NFAT5 may be operative in a regulatory pathway distinct from that instigated by hypertonicity. Yet, little is known about the nature of upstream signaling pathway(s) operating in isotonic conditions. RA is a chronic autoimmune disease that primarily attacks synovial joints . In the RA joints, various inflammatory cells interact with each other via an array of cytokines and/or cell-to-cell contacts, leading to prolonged inflammation and the destruction of cartilage and bone . For example, synovial macrophages, the major effectors of synovitis, secrete a number of pro-inflammatory mediators, including TNF-α, IL-6, prostaglandins, and nitric oxide (NO), involving inflammatory cascades and joint destruction . In turn, TLR ligation, pro-inflammatory cytokines, and cell contact (e.g. with activated T cells) drive macrophage activation. In particular, TLR2 and TLR4 are highly expressed in RA synovial fluid macrophages, and are thought to be responsible for the persistent activation of macrophages . Since NFAT5 is crucial to synoviocyte proliferation [7, 11], Ab-induced arthritis in mice , and TLR-dependent innate immunity , it might be required for the development of TLR-dependent inflammatory arthritis. In this study, the NFAT5 signaling pathway was investigated in macrophages triggered by TLR ligation. We identified a novel ROS–p38 MAPK–NFAT5 pathway (ROS is reactive oxygen species, MAPK is mitogen-activated protein kinase) activated in RAW 264.7 macrophages upon TLR stimulation. We also found first that XO-derived (xanthine oxidase) ROS acted as the upstream regulator of NFAT5 operating under inflammatory conditions. Moreover, the XO–NFAT5 pathway was required for TLR-mediated inflammatory arthritis in mice, inducing pro-inflammatory cytokine production. Taken together, these data provide evidence for a novel function of the XO–NFAT5 axis in macrophage activation and TLRinduced arthritis, and support a potential role for XO inhibitors, such as allopurinol and febuxostat, as a new therapeutic approach in patients with RA. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Immunol. 2014. 44: 2721–2736
Results Induction of NFAT5 activity by TLR in RAW 264.7 macrophages To analyze the transcriptional activity of NFAT5, the NFAT5 consensus sequence fused to the green fluorescence protein (GFP) reporter construct was stably transfected into RAW 264.7 macrophages, which showed basal NFAT5-dependent reporter activity at high levels (Supporting Information Fig. 1A). We found that TLR ligation by LPS or heat-inactivated Escherichia Coli was capable of inducing NFAT5-dependent reporter activity in the cells (Supporting Information Fig. 1A and B), consistent with a previous report . This effect was time- and dose-dependent (data not shown). In addition, triggering of TLR2 with peptidoglycan (PG), but not TLR3 (by poly I:C) nor TLR9 ligation (by CpG), substantially induced NFAT5-dependent reporter activity and increased its protein expression to a degree comparable to TLR4 ligation (Supporting Information Fig. 1C and D). Moreover, LPS, PG, and heat-inactivated E. coli all increased the trafficking of NFAT5 from the cytoplasm to the nucleus of macrophages (data not shown). Taken together, these results indicate that TLR ligation induces the reporter activity of NFAT5. Next, the effect of modulating NFAT5 activity on the gene expression in RAW 264.7 macrophages was examined by stably transfecting NFAT5 short hairpin RNA (shRNA) into the cells and monitoring the expression of NFAT5-governed genes (Fig. 1A). NFAT5 divergently regulated the expression of pro- and antiinflammatory genes in RAW 264.7 macrophages. Particularly, LPSor PG-induced nitrite, IL-6, and cyclooxygenase 2 (COX-2) expressions were significantly decreased by NFAT5 shRNA (Fig. 1B, C, and F), whereas IL-10 production was markedly increased (Fig. 1E). Of note, in RAW 264.7 macrophages, NFAT5 shRNA almost completely abolished the LPS- or PG-induced production of IL-6 (Fig. 1C), while the expressions of COX-2 and TNF-α were partially decreased (Fig. 1F and data not shown). These results indicate that NFAT5 is one of the major transcription factors responsible for IL-6 production by RAW 264.7 macrophages upon TLR ligation. Both the N-terminal DNA-binding domain (NDBD) of NFAT5 and NFAT5 decoy, which were introduced into the cells to exclude the off-target effect of shRNA, similarly decreased the productions of nitrite and IL-6 (Fig. 1D and data not shown). Collectively, these data support the view that NFAT5 differentially regulates nitrite, IL-6, and IL-10 production from RAW 264.7 macrophages activated by TLR-2 and TLR-4 ligation.
Differential effect of NFAT5 on gene expression in response to varying stimuli The concentration of LPS employed in the present was high (5 μg/mL), which may influence NFAT5 induction via cell damage and hypertonicity. To eliminate this concern, we examined viability of RAW 264.7 macrophages treated with high-dose LPS (5 μg/mL) over 24 h using Trypan blue exclusion and www.eji-journal.eu
Eur. J. Immunol. 2014. 44: 2721–2736
Figure 1. NFAT5 differentially regulates NO, IL-6, and IL-10 production by macrophages. (A) RAW 264.7 cells were stably transfected with lentiviral particles harboring NFAT5 shRNA or scrambled shRNA. The NFAT5 reporter gene was then transiently co-transfected into the cells, which were then stimulated with heat-inactivated E. coli (3 × 106 CFU/mL). GFP expression was assessed by fluorescence microscopy (left). Reduction in NFAT5 expression was confirmed by Western blot analysis (right), using α-tubulin as a loading control. (B) After stable knockdown with NFAT5 shRNA, RAW 264.7 macrophages were stimulated with LPS (1 μg/mL; left panel and 5 μg/mL; right panel), E. coli or peptidoglycan (PG; 10 μg/mL) for 24 h, and nitrite levels in the culture supernatants were measured by the Griess reaction. The degree of cell viability was assessed using an MTT assay. Data are presented as nitrite or IL-6 concentration relative to the percentage of the viability of NFAT5 shRNA-treated cells to the viability of control cells. p < 0.01 versus control (scrambled) shRNA. (C) RAW 264.7 macrophages were transfected stably with NFAT5 shRNA, and then stimulated with LPS (0.1, 1, 5, and 10 μg/mL) or PG (0.1, 1, 5, and 10 μg/mL) for 24 h. IL-6 production was determined by ELISA. *p < 0.01 versus control shRNA-transfected cells. (D) Macrophages were stably transfected with an NFAT5 decoy or NDBD, and then stimulated with the indicated concentrations of LPS for 24 h. IL-6 production in response to an NFAT5 decoy (left panel) encompassing the NFAT5 consensus sequence or NDBD of NFAT5 (right panel), was assessed by ELISA. NDBD fused to the nuclear localization sequence (NLS) sized 27 kDa was identified by Western blot analysis using anti-NFAT5 antibodies (right upper corner of the graph). β-actin was used as a loading control (data not shown). *p < 0.01 versus control NFAT5 decoy (left) or mock DNA (right). (E) IL-10 production by RAW 264.7 cells was assessed 24 h after stimulation with LPS or PG. p < 0.01 versus control shRNA. (F) RAW 264.7 macrophages were stimulated with LPS (5 μg/mL) or PG (10 μg/mL) for 24 h, and then COX-2 and iNOS expression in the cells was determined by Western blot analysis. α-tubulin was used as a loading control. (B–D) Data are shown as mean + SD and are representative of one of at least three independent experiments, each performed in triplicate. (F) Data are representative of one of three independent experiments with similar results. Comparison of the data between groups was performed by the Mann–Whitney U-test.
PI staining. We found that cell viability was minimally (