- Email: [email protected]
Pathological activation of canonical nuclear-factor κB by synergy of tumor necrosis factor α and TNF-like weak inducer of apoptosis in mouse acute colitis
Cytokine 69 (2014) 14–21
Contents lists available at ScienceDirect
Cytokine journal homepage: www.journals.elsevier.com/cytokine
Pathological activation of canonical nuclear-factor jB by synergy of tumor necrosis factor a and TNF-like weak inducer of apoptosis in mouse acute colitis Taeko Dohi a,⇑,2, Rei Kawashima a,c, Yuki I. Kawamura a, Takeshi Otsubo a, Teruki Hagiwara a, Aldo Amatucci b, Jennifer Michaelson b,1, Linda C. Burkly b,2 a Department of Gastroenterology, Research Center for Hepatitis and Immunology, Research Institute, National Center for Global Health and Medicine, 1-7-1 Kohnodai, Ichikawa, Chiba 272-8516, Japan b Biogen Idec, Cambridge, MA 02142, USA c Kitasato University Graduate School of Medical Sciences, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa 252-0373, Japan
a r t i c l e
i n f o
Article history: Received 7 January 2014 Received in revised form 15 April 2014 Accepted 1 May 2014
Keywords: Epithelial cell Acute colitis Cytokines
a b s t r a c t Tumor necrosis factor (TNF)-a is a major effector in various inflammatory conditions. TNF-like weak inducer of apoptosis (TWEAK) is a member of the TNF superfamily that promotes inflammatory tissue damage through its receptor, FGF-inducible molecule 14 (Fn14). Since both TWEAK and TNF-a have been shown to mediate pathological responses through inter-dependent or independent pathways by in vitro, the potential interplay of these pathways was investigated in a mouse colitis model. Acute colitis was induced by rectal injection of trinitrobenzene sulfonic acid (TNBS), with administration of control IgG, TNF receptor (TNFR)-Ig chimeric protein, anti-TWEAK monoclonal antibody, or the combination of TNFR-Ig and anti-TWEAK antibody. On day 4, disease severity was evaluated and gene expression profiling was analyzed using whole colon tissue. NF-jB activation was investigated with Western blot. Levels of transcript of TWEAK, Fn14 and NF-jB-related molecules were measured in purified colon epithelial cells (ECs). As a result, activation of the canonical (p50/RelA), but not noncanonical (p100/RelB)-mediated pathway was the hallmark of inflammatory responses in this model. Inflammation induced upregulation of Fn14 only in ECs but not in other cell types. Combination treatment of TNFR-Ig and anti-TWEAK antibody synergistically reduced disease severity in comparison with the control antibody or single agent treatment. Gene expression profile of the colon indicated downregulation of canonical NF-jB pathway with combination treatment. In conclusion, synergistic activation of canonical NF-jB by TWEAK and TNF-a is critical for the induction of inflammatory tissue damage in acute inflammation. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction TNF-like weak inducer of apoptosis (TWEAK) is a cytokine of the TNF ligand superfamily that is constitutively expressed by many
Abbreviations: ECs, epithelial cells; Fn14, fibroblast growth factor-inducible molecule 14; LPCs, lamina propria cells; LPS, lipopolysaccharide; mAb, monoclonal antibodies; MLN, mesenteric lymph node; NF, nuclear factor; PBS, phosphatebuffered saline; TLR, toll-like receptor; TNBS, trinitrobenzene sulfonic acid; TNF-a, tumor necrosis factor; TNFR2-Ig, TNF Receptor 2 fused with the murine IgG2a Fc region; TWEAK, TNF-like weak inducer of apoptosis. ⇑ Corresponding author at: 1-7-1 Kohnodai, Ichikawa, Chiba 272-8516, Japan. Tel.: +81 4 7375 4754; fax: +81 4 7375 4766. E-mail address: [email protected] (T. Dohi). 1 Current Address: Formerly Biogen Idec, Jounce Therapeutics, Cambridge, MA, 02142, USA. 2 Share senior authorship. http://dx.doi.org/10.1016/j.cyto.2014.05.001 1043-4666/Ó 2014 Elsevier Ltd. All rights reserved.
innate and adaptive immune cell types [1,2]. In various animal models of inflammation, infiltrating leukocytes including macrophages and activated T cells produce TWEAK. On the other hand, the TWEAK receptor, fibroblast growth factor-inducible molecule 14 (Fn14), is a highly inducible molecule in mesenchymal cells, epithelial cells and endothelial cells in response to basic fibroblast growth factor [3], platelet-derived growth factor and vascular endothelial growth factor [4], indicating the activation of the TWEAK/Fn14 pathway in the context of tissue injury and inflammation occurs through induction of Fn14. Indeed, Fn14 expression is upregulated in many different organs systems in contexts of injury or disease, including the liver, heart, vasculature, skeletal muscle, kidney, intestinal epithelium, and spinal cord; and blocking or deficiency in the TWEAK/Fn14 pathway ameliorates disease in multiple models [5], including several models of colitis [6–8]. In a mouse hapten-induced colitis model [6], we found that Fn14 was
T. Dohi et al. / Cytokine 69 (2014) 14–21
upregulated in the inflamed colonic epithelial cells (ECs), and TWEAK deficiency or TWEAK blocking mAbs significantly ameliorated this colitis, apparently by reducing inflammation, limiting chemokine and matrix metalloproteinase expression in EC and EC death [7]. Thus, TWEAK is involved in intestinal inflammation and tissue damage through inflammatory responses in EC. Tumor necrosis factor (TNF)-a is a major cytokine that mediates inflammatory responses and tissue damage in various inflammatory conditions (reviewed in [9]). The source of TNFa is mainly inflammatory cells type including macrophages and T cells. There are 2 cell surface receptors for TNF-a, TNFR1 and TNFR2, which are differentially expressed depending on cell types and normal/pathological conditions. TNFR1 and TNFR2 are both interact with TNF-a but generally TNFR1 is known to mediate major pathways for tissue damage by activating proinflammatory and the programmed cell-death related signals. Compared to TNFR1, TNFR2 signaling has not been well characterized, but shown to share some downstream pathways with TNFR1 and also promote tissue repair and angiogenesis. Activation of nuclear factor (NF)-jB is the key component of TNF-a signaling, leading to phosphorylation of IjB proteins to promote its ubiquitination and degradation, resulting in the nuclear translocation of NF-jB molecules, where NF-jB initiates gene transcription. Although both TNF-a and TWEAK are thus profoundly involved in inflammatory tissue damage, interactions or independence between TNF-a and TWEAK signal pathway has not been fully elucidated. For example, TWEAK and TNF-a appear to act through independent pathways in mediating some pathological responses. The response of cultured synoviocytes, which produce inflammatory chemokines in response to TWEAK, is not inhibited by blocking TNF-a, and conversely, synoviocyte responses to TNF-a are not inhibited by anti-TWEAK mAb [10]. In addition, in the same study it was shown that combined TNF-a and TWEAK stimulation was synergistic for production of multiple proinflammatory mediators. It is also known that TNF-a induces rapid activation of the canonical (p50/RelA) pathway of nuclear factor (NF)-jB signaling. TWEAK signaling through Fn14 can activate canonical NF-jB, however in contrast to TNF-a TWEAK/Fn14 engagement reportedly induce the delayed and prolonged induction of the noncanonical (p100/RelB) NF-jB pathway [11–14]. These results suggest that TWEAK and TNF-a stimulations both yield inflammatory output but involve differential activation of NF-jB signaling pathways. In the case of intestinal epithelial cells, we recently found that in IL-13-induced apoptosis, presence of both TWEAK/Fn14 and TNF-a is indispensable, suggesting largely shared pathway of TWEAK and TNF-a in induction of programmed cell death [7]. In these intestinal explants, we also found that the process of TNF-a activation in EC, shedding from membrane of epithelial cells, was TWEAK/Fn14 dependent. Based on these in vitro studies, we wished to explore the interdependency of TNF-a and TWEAK in vivo in an inflammatory setting. For this purpose, we chose hapten-induced colitis model in C57BL/6 strain, where acute inflammation can be stably induced. Our preliminary experiment indicated that full dose of TNF-a inhibitor was effective in prevention of colitis, while anti-TWEAK antibody administration was less effective, suggesting that other pathogenic pathways than TWEAK, especially TNF-a, are in play in this model. Thus to explore interplay between TNF-a and TWEAK and unveil the potential impact of a TWEAK-dependent pathway, we examined the effect of a suboptimal dose of TNF-a inhibitor in combination with TWEAK/Fn14 pathway blockade. This experimental system disclosed an unexpected synergy of TWEAK and TNF-a in acute inflammation.
15
2. Methods and materials 2.1. Mice Wild type, 6-week C57BL/6 male mice for colitis studies were purchased from CLEA Japan and kept under specific pathogen-free conditions during experiments performed at 7–8 weeks old. All protocols were approved by the institutional animal care and use committees in the Research Institute, National Center for Global Health and Medicine. 2.2. Reagents Trinitrobenzene sulfonic acid (TNBS) was purchased from Research Organics, (Cleveland, OH). For inhibition of endogenous TNF-a or TWEAK in mice, murine TNF Receptor 2 fused with the murine IgG2a Fc region (TNFR2-Ig) or murine anti-TWEAK IgG2a mAb (mP2D10) [15] were used, or mice were administered an isotype-matched control anti-human CD20 mAb that does not cross react with murine CD20, all prepared by Biogen Idec, Inc. (Cambridge, MA). 2.3. Induction and evaluation of colitis TNBS colitis was induced by intrarectal administration of a 2% solution of TNBS in phosphate-buffered saline (PBS): ethanol (1:1) [16,17]. For acute inflammatory responses, 70 lg/g body weight of TNBS was given on day 0 and animals sacrificed on day 4. One hour prior to administration of TNBS, groups of mice were injected i.p. with the control IgG2a mAb (anti-human CD20) (10 mg/kg), TNFR-Ig (0.3 mg/kg), anti-TWEAK (mP2D10, 10 mg/kg), the combination of TNFR-Fc (0.3 mg/kg) and anti-TWEAK mP2D10 (10 mg/kg), or were untreated. For single agent and combination treatments groups, 0.3 mg/kg TNFR-Ig was employed, since 1 mg/kg of TNFR-Ig markedly ameliorated TNBS colitis but 0.3 mg/kg was much less effective as monitored by the effect on colon length and body weight. Open ulcer area was measured in captured macroscopic images using image J (NIH) software. Colon tissue was rolled and snap-frozen in liquid nitrogen. Specimens for RNA extraction or Western blotting were cut from the frozen rolled colon. Frozen sections, showing the whole length of the colon, were stained with hematoxylin and eosin. Histological scores were blindly assigned to each proximal, middle and distal segment as follows: 0 – normal, 1 – ulcer or cell infiltration limited to the mucosa, 2 – ulcer or limited cell infiltration in the submucosa, 3 – focal ulcer involving all layers of the colon, 4 – multiple lesions involving all layers of the colon or necrotizing ulcer larger than 3 mm in length. Thus the total possible histological score is 12. A set of experiments using 5–8 mice for each experimental group was performed 3 times. Total number of mice for each experimental condition was as follows; untreated TNBS colitis group, 12; control antibody, 20; TNFR-Ig, 24; anti-TWEAK mAb, 24; combination of TNFR-Ig and anti-TWEAK mAb, 24. 2.4. Western blotting Snap-frozen colon tissue was crushed and total protein was extracted using standard methods. Extracts from individuals in a given treatment group pooled, resolved by SDS–PAGE and transferred to PVDF membranes (Bio Rad, Hercules, CA, USA) which were probed using the primary antibodies against IjBa (4812; Cell Signaling Technology), phospho-IjBa (9246; Cell Signaling Technology), phosho-p100 (4810; Cell Signaling Technology), p100/ p52 (4882; Cell Signaling Technology), and GAPDH (AM4300; Applied Biosystems). Secondary antibodies were anti-mouse
16
T. Dohi et al. / Cytokine 69 (2014) 14–21
IgG-HRP (NA931VS; Amersham Biosciences) and anti-rabbit IgGHRP (NA934VS; Amersham Biosciences). Samples were visualized using SuperSignal West Pico Chemiluminescent Substrate (34079; Thermo Scientific) and BioMax Light Film (8761520; Kodak). For semi-quantification of bands, obtained images were analyzed using NIH image J and relative value to that of GAPDH of the identical sample was calculated.
(lods) greater than zero and fold change greater than 2.0 were considered significantly different. Ingenuity Pathway Analysis software (Ingenuity Systems, Inc. CA 94063) was used for analysis of upstream regulators. Upstream regulators with activation Z-scores greater than 2 or smaller than 2 were considered significant.
2.5. Cell separation and quantitative RT-PCR
Data were statistically analyzed using a software Prism 4 (GraphPad Sofware, Inc) using the method indicated in the legend.
Isolation of purified ECs, lamina propria cells (LPCs) and mesenteric lymph node cells (MLN) was performed as described previously [8]. In brief, released ECs were stained with 7-Aminoactinomycin D (7AAD), fluorescein isothiocyanate-anti-CD45 and Rphycoerythrin-anti-EpCAM antibodies (BD Biosciences, Franklin Lakes, NJ) and viable EpCAM+CD45 cells sorted as purified ECs (Moflo, Beckman Coulter, Tokyo, Japan). To separate LPCs, the ECdenuded residue was treated with collagenase A (Roche Diagnostics, Tokyo, Japan). Total RNA was prepared from EC, LPMC or whole MLN tissue describe as above. Complementary DNA was synthesized from RNA by reverse transcription, and gene expression levels were determined with Taqman gene expression assay kit for TWEAK (Mm02583406_s1), Fn14 (Mm00489103_ m1), Rela (Mm00501346_m1), Relb (Mm00485664_m1), Tnfrsf1a (Mm00441875_m1), Tnfrsf1b (Mm00441889_m1), Nfkb2 (Mm004 79810_g1) and 18s rRNA (Hs99999901_s1) and ABI PRISM 7900 Sequence Detector (Applied Biosystems). The results were transformed by the DCt/DDCt method as described by the manufacturer, with 18s rRNA used as the calibrator gene. 2.6. Gene expression profiling Total RNA was prepared individually from the whole colon mucosa of 4–8 mice for each experimental group by using RNABee RNA Isolation solvent (TEL-TEST, Inc. TX) purified with DNAse (RNAse-free DNA set) and RNeasy Mini kit (Qiagen, Tokyo, Japan). Affymetrix scans were subject to quality control (QC) measures. These tests included a visual inspection of the distribution of raw signal intensities and an assessment of RNA degradation, relative log expression (RLE), and normalized unscaled standard error (NUSE). All sample scans passed these QC metrics. CEL files were subjected to GC-content-based Robust Multi-array Average normalization [18,19]. Expression levels were log (base 2) transformed. All calculations and analyses were carried out using R and Bioconductor computational tools [20]. To identify differentially expressed genes (DEGs) in colon between groups of samples, we applied the linear modeling approach (MANOVA) to fit gene expression levels (log2 transformed) according to the defined groups of samples and Bayesian posterior error analysis as implemented by Smyth [21]. Genes that exhibited a log-odds score
2.7. Statistical analysis
3. Results 3.1. Activation of NF-jB canonical pathway in the acute colitis In order to explore the potential cooperative involvement of both TNF-a and TWEAK pathways, we induced colitis in the C57BL/6 strain, where anti-TWEAK mAb treatment alone was only modestly effective, suggesting that other pathogenic pathways, especially involving TNF-a, are in play [22,23]. First, we investigated the activation status of NF-jB canonical and non-canonical pathway activation in this model. To determine activation of NFjB in colitis at biochemical level, Western blot analysis was performed on whole colon tissue (Fig. 1A and B). Activation of IjBa was clearly observed in the mice with colitis compared with naïve colon. In contrast, there was constitutive activation of NF-jB noncanonical pathway as shown by the presence of phosph-p100 and p52 in naïve colon; however, increase of phosphorylated p100 or p52 in colitis was not apparent. Overall these results indicate that NF-jB canonical pathway was mainly activated in the whole colon of acute TNBS-induced colitis. 3.2. Fn14 and Rel A gene expression was upregulated in ECs but not other cell types Since protein extract from whole colon tissue might mask celltype specific inflammatory responses, we investigated inflammatory responses at the cellular level. For this purpose, we obtained highly purified live colonic ECs with flow cytometry, LPC and MLN from mice with colitis and performed quantitative qRT-PCR for TWEAK, Fn14, TNFR1 and 2. We also performed qRT-PCR for canonical and noncanonical NF-jB components, since their transcript levels can be modulated by NF-jB signaling [24] (Fig. 2). When compared with the level of naïve mice, statistically significant induction of TWEAK was not seen in any of the cell types, although LPC showed a trend of upregulation. Significant upregulation of Fn14 was seen only in ECs with colitis, whereas there were no changes in LPMC or MLN. Thus, in this colitis model, TWEAK/Fn14 signaling was induced mainly in ECs, rather than
Fig. 1. NF-jB canonical but not noncanonical pathway was activated. (A) Western blot analysis. Whole colon tissue was obtained from naïve mice and day 4 colitic mice treated with control IgG and blots probed with antibodies for IjBa, phospho (p)- IjBa, p-p100, p100, p52 and GAPDH. Each lane represents individual mouse. (B) Result of panel A was semiquantified normalized with levels of GAPDH. Short bars indicate mean value.
T. Dohi et al. / Cytokine 69 (2014) 14–21
17
Fig. 2. NF-jB canonical pathway was activated in epithelial cells colitis was induced and EC, LPS, MLN cells isolated as described, with the purified EC being EpCAM+CD45 . Total RNA extract was subjected to quantitative RT-PCR. Fold changes from naïve cells for each cell type are shown, as average + SD, n = 3. Two tailed P values with unpaired ttest are indicated.
hematopoietic cell types. In addition, the expression of TNFRSF1A (coding TNFR1) and TNFRSF1B (TNFR2) both showed a trend of upregulation in ECs with colitis, and induction of TNFRSF1B reached statistically significant level. The canonical NF-jB component RelA (p65) was also upregulated in EC with colitis whereas the noncanonical NF-jB components RelB (p52) and NFKB2 (p100) gene expression were not upregulated in any of the cell types. These results suggest that the EC is a key cell type in which the canonical NF-jB pathway is activated in the colitis. 3.3. Synergistic protective effect of combination treatment in colitis As a tool for TNF or TWEAK inhibition in the mouse system, we used suboptimal dose of TNFR-Ig fusion protein and anti-TWEAK monoclonal antibody, respectively. We found that combination treatment resulted in a more than additive protective effect from acute colitis as compared to single agent treatment (Fig. 3).
As shown in Fig. 3, administration of either single agent, 0.3 mg/kg of TNFR-Ig or 10 mg/kg anti-TWEAK mAb, showed limited or no efficacy with respect to the body weight (Fig. 3A), colon length (Fig. 3B) and macroscopic/histological analysis (Fig. 3C–F). The TNFR-Ig-, anti-TWEAK mAb- and control Ig-treated groups each showed a trend of better recovery of body weight as compared to the untreated group, with no significant difference between the three treated groups on day 4. In marked contrast, treatment with the combination of TNFR-Ig and anti-TWEAK mAb resulted in significantly less weight loss and more rapid recovery of body weight than any of the other groups. In addition, the colon was significantly longer in the combination-treatment group than any of other experimental groups (Fig. 3B), and closer in value to that of naïve mice (mean ± SE, 8.76 ± 0.13 cm), whereas the monotherapyor control Ig-treated groups were not significantly different from untreated in colon length. Macroscopically, large necrotizing ulcers in middle to distal colon are typically seen in acute TNBS-induced
18
T. Dohi et al. / Cytokine 69 (2014) 14–21
Fig. 3. Combined treatment with TNF-a and TWEAK inhibitors protects against TNBS-induced colitis. (A) Body weight changes (% of day 0) in each treatment group. Data are shown as a mean ± SEM. Statistically significance was found in ‘‘untreated vs. combo’’ on day 3 and 4, ‘‘Control IgG vs. combo’’ on day 3 and 4, ‘‘TNFR-Ig vs. combo’’ on day 3 and 4, and ‘‘anti-TWEAK vs. combo’’ on day 4, using two-way anova with post Bonferroni multiple comparisons. (B) Colon length of day 4. Horizontal dotted line indicated mean value for naïve colon. (C) Macroscopic appearance of colon. (D) Area of open ulcer. (E) Histological score. (F) Typical histological finding in each treatment. Frozen sections were stained with hematoxylin and eosin. Scale bar = 200 lm. In panels b–e, each dot represents an individual mouse and solid horizontal bars indicate mean value. * Difference from combo-treated group was significant with Mann Whitney analysis (P < 0.05).
colitis. Many mice treated with the combination exhibited minimal open ulcers with only redness of the mucosal surface (Fig. 3C) and the area of open ulcer was significantly less in the combinationtreated as compared to all other groups (Fig. 3D). Minimal open ulcers were also seen in many of the mice treated with
anti-TWEAK mAb alone, and this was significant as compared to untreated (P = 0.039), but did not reach significance as compared to control Ig- or TNFR-Ig treatment. Matching these macroscopic observations, combination treatment resulted in a significantly lower histological score as compared to all other groups (Fig. 3E
T. Dohi et al. / Cytokine 69 (2014) 14–21
19
and F). TNFR-Ig treatment showed some benefit, with scores significantly lower than untreated or control IgG (P = 0.0038 and 0.004, respectively), as did anti-TWEAK mAb treatment, which was significantly lower than that of the untreated group (P = 0.0197) though not as compared to the control IgG treated group. Thus, administration of each of the single agents, TNF-Ig or anti-TWEAK mAb, had subtle protective effects on the histological scores and open ulcer area, respectively. Treatment with the combination of these two agents resulted in a much greater than additive effect of the monotherapies. 3.4. Gene expression profiling indicates synergistically suppressed NF-
jB pathway with combination treatment in colitis To investigate the mechanism for the synergistic protective effect of combined TNF-a and TWEAK inhibitor treatment in mouse colitis, gene expression profiles were analyzed using RNA extracted from whole colon. The whole data is available at GSE53835. A total of 781 genes changed more than 2-fold in both the control Ig-treated and untreated mice as compared to naïve mice. We investigated whether any of these genes were specifically changed as a result of monotherapy. However, in a comparison of the TNFR-Ig treated vs. untreated or control Ig-treated group, no such genes were found. Likewise, there was only 1 gene (2810416G20Rik) detected in comparison of the anti-TWEAK mAb-treated and untreated group or control Ig-treated group. These results indicated that neither the TNFR-Ig nor anti-TWEAK mAb treatment significantly affected colitis-induced changes in gene expression, consistent with our observation that single agent treatment had minimal effect on the measures of colitis. In contrast, from this list of 781 genes, there were 142 genes that were changed significantly by combination treatment as compared to both of the untreated and control Ig-treated groups, including 131 downregulated and 11 upregulated genes. Among these downregulated genes, we identified the 62 genes whose levels of expression were the most strongly affected such that they were not significantly different (less than 2-fold) between the group with combination treatment and naïve mice. This list of 62 genes (Fig. 4) indicates those upregulated in colitis induction but maintained by combination treatment at nearly normal levels, and thus the major molecular changes that combo treatment altered to protect against colitis. Using this list of 62 downregulated genes, we searched the possible upstream pathways that were affected by combination treatment using the Upstream Regulator Analysis Software, leading to a list of possible upstream pathways. In addition to TNF-a, TNFRSF1A (TNF receptor 1), TWEAK, lipopolysaccharide (LPS) and innate sensors like Toll-like receptor (TLR), were identified as potential as upstream regulators. Of note, IKBKG (IKK-c/NEMO) and NFKBIA (IjBa were also included as upstream regulators, suggesting that the canonical pathway of NF-jB was significantly downregulated. Fig. 4 shows the target genes of these upstream pathways that were found in our data set. Since differential target genes for the NF-jB canonical and noncanonical pathways have not been well identified, we could not discern canonical vs. noncanonical signaling from this analysis. Thus, the result of gene expression profiling was generally interpreted that downregulation of the NF-jB pathway and gene expression downstream of TNF-a, TWEAK, and TLR stimuli were the major events downregulated by combo treatment.
Fig. 4. A gene set strongly downregulated by combined TNF-a inhibitor and antiTWEAK mAb treatment in mouse TNBS-induced colitis. Shown are 62 genes, which were upregulated in whole colon in both the untreated and control Ig-treated groups, with levels in the combination treatment group comparable to those in naïve mice. Fold changes are as indicated with the color key. Black squares indicate the target molecules in the data set that were the basis for determination of the upstream regulators, LPS, NFKBIA, TNF-a and TWEAK, using Ingenuity Pathway Analysis software.
4. Discussion In this study, we unveiled a striking synergistic contribution of the TNF-a and TWEAK pathways in a model of acute colitis using combination treatment with pathway inhibitors, which cannot be
simply explained as an additive effect of both agents. This effect is apparently mediated through the canonical NF-jB signaling in EC. Based on these novel findings, we discuss possible mechanisms for
20
T. Dohi et al. / Cytokine 69 (2014) 14–21
the TWEAK and TNF-a pathways and their contributions to colitis as illustrated in Fig. 5. Interplay between TWEAK and TNF-related pathways has been suggested in in vitro assays [25–27]. In this study, TWEAK signaling is highly inducible in colon EC of C57BL/6 mice as indicated by Fn14 upregulation in EC from colitis mice. We also found that TNFR are broadly expressed in the EC, and mRNA for TNFR2 was notably increased with acute colitis. These results indicated that the site of synergy of TNF-a and TWEAK was in EC but not other cell types, involving Fn14 and TNFR2 signaling pathways in addition to TNFR1. Previous studies demonstrated activation of non-canonical NF-jB pathway induced by TWEAK/Fn14 [11–14], and that TNFR2 was also able to activate non-canonical NF-jB pathway interfering canonical pathway induced with TNFR1 [28]. Therefore, we examined the involvement of NF-jB noncanonical pathway in this model. However, in this acute colitis model, noncanonical NF-jB activation was constitutively high, and neither further increase nor effect of any treatment apparent, even in transcript measures dissected at the cellular level. In contrast, NF-jB canonical pathway activation was prominent in this model and also the target of combination therapy with anti-TWEAK antibody and TNF-inhibitor therapy. Thus the interplay between TWEAK and TNF-a in this acute colitis model is presumably occurring through their signaling of the canonical NF-jB pathway. Clues as the mechanisms underlying the synergy between the TNF-a and TWEAK pathways derive from our further analysis of the colitis tissues, namely the mild effects of TNF-a inhibitor and anti-TWEAK as single agents. In the anti-TWEAK treated group there was trend of more limited ulcer-area than TNFR-Ig treatment, which may reflect a direct preventive effect from EC damage by blocking TWEAK. The TNFR-Ig-treated group showed a lower histological score than untreated and control-Ig treated group, whereas the anti-TWEAK treated group did not. This seems to due to less cell infiltration reflecting the anti-inflammatory effect achieved by partially neutralizing TNF-a. Perhaps each of these effects alone is not sufficient to protect against colitis, but synergistic when combined. Further supporting this notion, we note that in
contrast to spontaneous inflammatory polyarthritis and inflammatory bowel disease in TNF-overexpressing mice [29], TWEAK overexpression in naïve mice induces tissue changes in the liver and heart without overt inflammation [30,31] though TWEAK may potentiate and sustain the inflammatory response through a feed forward loop, inducing the production by EC of cytokines, chemokines and matrix metalloproteinases [6] which then further recruit and activate macrophages and lymphocytes to produce TNF-a as well as TWEAK. While TWEAK also directly supports the activation of latent, membrane bound TNF-a from intestinal ECs in the presence of IL-13 [7] and thereby enhances EC death, our observation that anti-TWEAK but not TNFR-Ig monotherapy reduced ulcer area in acute colitis suggests that TWEAK promotes EC death through other TNF-a-independent mechanisms, consistent with the TWEAK-induced cell death that is TNF-a-independent in other systems [32]. In addition to TWEAK and TNF-a signaling, LPS appears to be a major upstream regulator based on the colitis gene expression analysis. There may also be an unknown mechanism whereby TWEAK cross talks with some of downstream pathways of TLRs in a TNF-a-independent or dependent manner. In addition, we recently demonstrated that the TWEAK/Fn14 pathway promotes chronic TNBS-induced colitis with fibrosis through the production of TSLP by ECs [33]. It is unknown if this is mediated by canonical or noncanonical NF-jB. However, we suggest that delayed and prolonged noncanonical NF-jB activation by TWEAK [14] may be important in chronic colitis. In conclusion, our study indicates that TNF-a and TWEAK/Fn14 pathways mutually support activation of NF-jB canonical pathway in a significant synergistic manner, which is critical in the tissue damage of acute inflammation.
Database linking Database: GSE53835. Acknowledgements This work was supported partly by grants and contracts from the program Grants-in-Aid for Scientific Research (C) (T. D. and Y.I. K.) and Grants-in Aid for Young Scientists (B) (T. O.) from the Ministry of Education, Cultures, Sports, Science, and Technology; a grant from the National Center for Global Health and Medicine 21-110, 22-205, 21-129, 25-104, (T. D. and Y.I. K.), Ministry of Health, Labor, and Welfare; Grants for Research on Measures for Intractable Diseases and Research on Health Sciences focusing on Drug Innovation from Ministry of Health, Labor and Welfare of Japan (T. D. and Y.I. K.); MEXT-Supported Program for the Strategic Research Foundation at Private Universities (T. D.). We thank Dr. M. Tamura-Nakano in NCGM EM Support Unit for her contribution to pathological analysis and Dr. Normand E. Allaire in Biogen Idec Inc for his valuable discussion on gene profiling analysis. References
Fig. 5. Model of TWEAK and TNF-a actions in colitis. TNF-a is produced by activated white blood cells (WBC, including macrophages and T cells). ECs also contribute active TNF-a by TWEAK/Fn14-mediated release from an endogenous pool. TNF-a activates NF-jB canonical pathway in white blood cell (WBC) types, inducing tissue damage and release of other proinflammatory mediators. TWEAK and TNF-a support and sustain canonical NF-jB pathway in EC. TWEAK may also activate the noncanonical NF-jB pathway. Signals from TLRs (not shown) also contribute to promote infiltration and activation of WBC and EC and may interplay with signaling associated with TNF-a and TWEAK.
[1] Kaplan MJ, Lewis EE, Shelden EA, Somers E, Pavlic R, McCune WJ, et al. The apoptotic ligands TRAIL, TWEAK, and Fas ligand mediate monocyte death induced by autologous lupus T cells. J Immunol 2002;169:6020–9 (Baltimore, Md: 1950). [2] Maecker H, Varfolomeev E, Kischkel F, Lawrence D, LeBlanc H, Lee W, et al. TWEAK attenuates the transition from innate to adaptive immunity. Cell 2005;123:931–44. [3] Meighan-Mantha RL, Hsu DK, Guo Y, Brown SA, Feng SL, Peifley KA, et al. The mitogen-inducible Fn14 gene encodes a type I transmembrane protein that modulates fibroblast adhesion and migration. J Biol Chem 1999;274: 33166–76.
T. Dohi et al. / Cytokine 69 (2014) 14–21 [4] Burkly LC, Michaelson JS, Hahm K, Jakubowski A, Zheng TS. TWEAKing tissue remodeling by a multifunctional cytokine: role of TWEAK/Fn14 pathway in health and disease. Cytokine 2007;40:1–16. [5] Burkly LC, Dohi T. The TWEAK/Fn14 pathway in tissue remodeling: For better or for worse. Adv Exp Med Biol 2011;691:305–22. [6] Dohi T, Borodovsky A, Wu P, Shearstone JR, Kawashima R, Runkel L, et al. TWEAK/Fn14 pathway: a nonredundant role in intestinal damage in mice through a TWEAK/intestinal epithelial cell axis. Gastroenterology 2009;136: 912–23. [7] Kawashima R, Kawamura YI, Oshio T, Son A, Yamazaki M, Hagiwara T, et al. Interleukin-13 damages intestinal mucosa via TWEAK and Fn14 in Mice-A pathway associated with ulcerative colitis. Gastroenterology 2011;141: 2119–29. [8] Son A, Oshio T, Kawamura YI, Hagiwara T, Yamazaki M, Inagaki-Ohara K, et al. TWEAK/Fn14 pathway promotes a T helper 2-type chronic colitis with fibrosis in mice. Mucosal Immunol 2013. [9] Bradley JR. TNF-mediated inflammatory disease. J Pathol 2008;214: 149–60. [10] Chicheportiche Y, Chicheportiche R, Sizing I, Thompson J, Benjamin CB, Ambrose C, et al. Proinflammatory activity of TWEAK on human dermal fibroblasts and synoviocytes: blocking and enhancing effects of anti-TWEAK monoclonal antibodies. Arthritis Res 2002;4:126–33. [11] Brown SA, Ghosh A, Winkles JA. Full-length, membrane-anchored TWEAK can function as a juxtacrine signaling molecule and activate the NF-kappaB pathway. J Biol Chem 2010;285:17432–41. [12] Brown SA, Richards CM, Hanscom HN, Feng SL, Winkles JA. The Fn14 cytoplasmic tail binds tumour-necrosis-factor-receptor-associated factors 1, 2, 3 and 5 and mediates nuclear factor-kappaB activation. Biochem J 2003;371:395–403. [13] Roos C, Wicovsky A, Muller N, Salzmann S, Rosenthal T, Kalthoff H, et al. Soluble and transmembrane TNF-like weak inducer of apoptosis differentially activate the classical and noncanonical NF-kappa B pathway. J Immunol 2010;185:1593–605 (Baltimore, Md: 1950). [14] Saitoh T, Nakayama M, Nakano H, Yagita H, Yamamoto N, Yamaoka S. TWEAK induces NF-kappaB2 p100 processing and long lasting NF-kappaB activation. J Biol Chem 2003;278:36005–12. [15] Campbell S, Burkly LC, Gao HX, Berman JW, Su L, Browning B, et al. Proinflammatory effects of TWEAK/Fn14 interactions in glomerular mesangial cells. J Immunol 2006;176:1889–98 (Baltimore, Md: 1950). [16] Dohi T, Fujihashi K, Kiyono H, Elson CO, McGhee JR. Mice deficient in Th1- and Th2-type cytokines develop distinct forms of hapten-induced colitis. Gastroenterology 2000;119:724–33. [17] Dohi T, Fujihashi K, Rennert PD, Iwatani K, Kiyono H, McGhee JR. Hapteninduced colitis is associated with colonic patch hypertrophy and T helper 2type responses. J Exp Med 1999;189:1169–80.
21
[18] Irizarry RA, Bolstad BM, Collin F, Cope LM, Hobbs B, Speed TP. Summaries of Affymetrix GeneChip probe level data. Nucl Acids Res 2003;31:e15. [19] Li C, Hung Wong W. Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application. Genome Biol 2001:2. RESEARCH0032. [20] Gentleman R. Bioinformatics and computational biology solutions using R and bioconductor. Springer Science+Business Media; 2005. [21] Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genetics Mol Biol 2004:3. Article3. [22] Shen C, de Hertogh G, Bullens DM, Van Assche G, Geboes K, Rutgeerts P, et al. Remission-inducing effect of anti-TNF monoclonal antibody in TNBS colitis: mechanisms beyond neutralization? Inflamm Bowel Dis 2007;13:308–16. [23] Neurath MF, Fuss I, Pasparakis M, Alexopoulou L, Haralambous S, Meyer, et al. Predominant pathogenic role of tumor necrosis factor in experimental colitis in mice. Eur J Immunol 1997;27:1743–50. [24] Allen IC, Wilson JE, Schneider M, Lich JD, Roberts RA, Arthur JC, et al. NLRP12 suppresses colon inflammation and tumorigenesis through the negative regulation of noncanonical NF-kappaB signaling. Immunity 2012;36:742–54. [25] Ikner A, Ashkenazi A. TWEAK induces apoptosis through a death-signaling complex comprising Receptor-interacting Protein 1 (RIP1), Fas-associated Death Domain (FADD), and Caspase-8. J Biol Chem 2011;286:21546–54. [26] Vince JE, Chau D, Callus B, Wong WW, Hawkins CJ, Schneider P, et al. TWEAKFN14 signaling induces lysosomal degradation of a cIAP1-TRAF2 complex to sensitize tumor cells to TNFalpha. J Cell Biol 2008;182:171–84. [27] Wicovsky A, Salzmann S, Roos C, Ehrenschwender M, Rosenthal T, Siegmund D, et al. TNF-like weak inducer of apoptosis inhibits proinflammatory TNF receptor-1 signaling. Cell Death Differ 2009;16:1445–59. [28] Cabal-Hierro L, Lazo PS. Signal transduction by tumor necrosis factor receptors. Cell Signal 2012;24:1297–305. [29] Kontoyiannis D, Pasparakis M, Pizarro TT, Cominelli F, Kollias G. Impaired on/ off regulation of TNF biosynthesis in mice lacking TNF AU-rich elements: implications for joint and gut-associated immunopathologies. Immunity 1999;10:387–98. [30] Jain M, Jakubowski A, Cui L, Shi J, Su L, Bauer M, et al. A novel role for tumor necrosis factor-like weak inducer of apoptosis (TWEAK) in the development of cardiac dysfunction and failure. Circulation 2009;119:2058–68. [31] Jakubowski A, Ambrose C, Parr M, Lincecum JM, Wang MZ, Zheng TS, et al. TWEAK induces liver progenitor cell proliferation. J Clin Invest 2005;115: 2330–40. [32] Burkly LC, Michaelson JS, Zheng TS. TWEAK/Fn14 pathway: an immunological switch for shaping tissue responses. Immunol Rev 2011;244:99–114. [33] Cultrone A, de Wouters T, Lakhdari O, Kelly D, Mulder I, Logan E, et al. The NFkappaB binding site located in the proximal region of the TSLP promoter is critical for TSLP modulation in human intestinal epithelial cells. Eur J Immunol 2013;43:1053–62.
Contents lists available at ScienceDirect
Cytokine journal homepage: www.journals.elsevier.com/cytokine
Pathological activation of canonical nuclear-factor jB by synergy of tumor necrosis factor a and TNF-like weak inducer of apoptosis in mouse acute colitis Taeko Dohi a,⇑,2, Rei Kawashima a,c, Yuki I. Kawamura a, Takeshi Otsubo a, Teruki Hagiwara a, Aldo Amatucci b, Jennifer Michaelson b,1, Linda C. Burkly b,2 a Department of Gastroenterology, Research Center for Hepatitis and Immunology, Research Institute, National Center for Global Health and Medicine, 1-7-1 Kohnodai, Ichikawa, Chiba 272-8516, Japan b Biogen Idec, Cambridge, MA 02142, USA c Kitasato University Graduate School of Medical Sciences, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa 252-0373, Japan
a r t i c l e
i n f o
Article history: Received 7 January 2014 Received in revised form 15 April 2014 Accepted 1 May 2014
Keywords: Epithelial cell Acute colitis Cytokines
a b s t r a c t Tumor necrosis factor (TNF)-a is a major effector in various inflammatory conditions. TNF-like weak inducer of apoptosis (TWEAK) is a member of the TNF superfamily that promotes inflammatory tissue damage through its receptor, FGF-inducible molecule 14 (Fn14). Since both TWEAK and TNF-a have been shown to mediate pathological responses through inter-dependent or independent pathways by in vitro, the potential interplay of these pathways was investigated in a mouse colitis model. Acute colitis was induced by rectal injection of trinitrobenzene sulfonic acid (TNBS), with administration of control IgG, TNF receptor (TNFR)-Ig chimeric protein, anti-TWEAK monoclonal antibody, or the combination of TNFR-Ig and anti-TWEAK antibody. On day 4, disease severity was evaluated and gene expression profiling was analyzed using whole colon tissue. NF-jB activation was investigated with Western blot. Levels of transcript of TWEAK, Fn14 and NF-jB-related molecules were measured in purified colon epithelial cells (ECs). As a result, activation of the canonical (p50/RelA), but not noncanonical (p100/RelB)-mediated pathway was the hallmark of inflammatory responses in this model. Inflammation induced upregulation of Fn14 only in ECs but not in other cell types. Combination treatment of TNFR-Ig and anti-TWEAK antibody synergistically reduced disease severity in comparison with the control antibody or single agent treatment. Gene expression profile of the colon indicated downregulation of canonical NF-jB pathway with combination treatment. In conclusion, synergistic activation of canonical NF-jB by TWEAK and TNF-a is critical for the induction of inflammatory tissue damage in acute inflammation. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction TNF-like weak inducer of apoptosis (TWEAK) is a cytokine of the TNF ligand superfamily that is constitutively expressed by many
Abbreviations: ECs, epithelial cells; Fn14, fibroblast growth factor-inducible molecule 14; LPCs, lamina propria cells; LPS, lipopolysaccharide; mAb, monoclonal antibodies; MLN, mesenteric lymph node; NF, nuclear factor; PBS, phosphatebuffered saline; TLR, toll-like receptor; TNBS, trinitrobenzene sulfonic acid; TNF-a, tumor necrosis factor; TNFR2-Ig, TNF Receptor 2 fused with the murine IgG2a Fc region; TWEAK, TNF-like weak inducer of apoptosis. ⇑ Corresponding author at: 1-7-1 Kohnodai, Ichikawa, Chiba 272-8516, Japan. Tel.: +81 4 7375 4754; fax: +81 4 7375 4766. E-mail address: [email protected] (T. Dohi). 1 Current Address: Formerly Biogen Idec, Jounce Therapeutics, Cambridge, MA, 02142, USA. 2 Share senior authorship. http://dx.doi.org/10.1016/j.cyto.2014.05.001 1043-4666/Ó 2014 Elsevier Ltd. All rights reserved.
innate and adaptive immune cell types [1,2]. In various animal models of inflammation, infiltrating leukocytes including macrophages and activated T cells produce TWEAK. On the other hand, the TWEAK receptor, fibroblast growth factor-inducible molecule 14 (Fn14), is a highly inducible molecule in mesenchymal cells, epithelial cells and endothelial cells in response to basic fibroblast growth factor [3], platelet-derived growth factor and vascular endothelial growth factor [4], indicating the activation of the TWEAK/Fn14 pathway in the context of tissue injury and inflammation occurs through induction of Fn14. Indeed, Fn14 expression is upregulated in many different organs systems in contexts of injury or disease, including the liver, heart, vasculature, skeletal muscle, kidney, intestinal epithelium, and spinal cord; and blocking or deficiency in the TWEAK/Fn14 pathway ameliorates disease in multiple models [5], including several models of colitis [6–8]. In a mouse hapten-induced colitis model [6], we found that Fn14 was
T. Dohi et al. / Cytokine 69 (2014) 14–21
upregulated in the inflamed colonic epithelial cells (ECs), and TWEAK deficiency or TWEAK blocking mAbs significantly ameliorated this colitis, apparently by reducing inflammation, limiting chemokine and matrix metalloproteinase expression in EC and EC death [7]. Thus, TWEAK is involved in intestinal inflammation and tissue damage through inflammatory responses in EC. Tumor necrosis factor (TNF)-a is a major cytokine that mediates inflammatory responses and tissue damage in various inflammatory conditions (reviewed in [9]). The source of TNFa is mainly inflammatory cells type including macrophages and T cells. There are 2 cell surface receptors for TNF-a, TNFR1 and TNFR2, which are differentially expressed depending on cell types and normal/pathological conditions. TNFR1 and TNFR2 are both interact with TNF-a but generally TNFR1 is known to mediate major pathways for tissue damage by activating proinflammatory and the programmed cell-death related signals. Compared to TNFR1, TNFR2 signaling has not been well characterized, but shown to share some downstream pathways with TNFR1 and also promote tissue repair and angiogenesis. Activation of nuclear factor (NF)-jB is the key component of TNF-a signaling, leading to phosphorylation of IjB proteins to promote its ubiquitination and degradation, resulting in the nuclear translocation of NF-jB molecules, where NF-jB initiates gene transcription. Although both TNF-a and TWEAK are thus profoundly involved in inflammatory tissue damage, interactions or independence between TNF-a and TWEAK signal pathway has not been fully elucidated. For example, TWEAK and TNF-a appear to act through independent pathways in mediating some pathological responses. The response of cultured synoviocytes, which produce inflammatory chemokines in response to TWEAK, is not inhibited by blocking TNF-a, and conversely, synoviocyte responses to TNF-a are not inhibited by anti-TWEAK mAb [10]. In addition, in the same study it was shown that combined TNF-a and TWEAK stimulation was synergistic for production of multiple proinflammatory mediators. It is also known that TNF-a induces rapid activation of the canonical (p50/RelA) pathway of nuclear factor (NF)-jB signaling. TWEAK signaling through Fn14 can activate canonical NF-jB, however in contrast to TNF-a TWEAK/Fn14 engagement reportedly induce the delayed and prolonged induction of the noncanonical (p100/RelB) NF-jB pathway [11–14]. These results suggest that TWEAK and TNF-a stimulations both yield inflammatory output but involve differential activation of NF-jB signaling pathways. In the case of intestinal epithelial cells, we recently found that in IL-13-induced apoptosis, presence of both TWEAK/Fn14 and TNF-a is indispensable, suggesting largely shared pathway of TWEAK and TNF-a in induction of programmed cell death [7]. In these intestinal explants, we also found that the process of TNF-a activation in EC, shedding from membrane of epithelial cells, was TWEAK/Fn14 dependent. Based on these in vitro studies, we wished to explore the interdependency of TNF-a and TWEAK in vivo in an inflammatory setting. For this purpose, we chose hapten-induced colitis model in C57BL/6 strain, where acute inflammation can be stably induced. Our preliminary experiment indicated that full dose of TNF-a inhibitor was effective in prevention of colitis, while anti-TWEAK antibody administration was less effective, suggesting that other pathogenic pathways than TWEAK, especially TNF-a, are in play in this model. Thus to explore interplay between TNF-a and TWEAK and unveil the potential impact of a TWEAK-dependent pathway, we examined the effect of a suboptimal dose of TNF-a inhibitor in combination with TWEAK/Fn14 pathway blockade. This experimental system disclosed an unexpected synergy of TWEAK and TNF-a in acute inflammation.
15
2. Methods and materials 2.1. Mice Wild type, 6-week C57BL/6 male mice for colitis studies were purchased from CLEA Japan and kept under specific pathogen-free conditions during experiments performed at 7–8 weeks old. All protocols were approved by the institutional animal care and use committees in the Research Institute, National Center for Global Health and Medicine. 2.2. Reagents Trinitrobenzene sulfonic acid (TNBS) was purchased from Research Organics, (Cleveland, OH). For inhibition of endogenous TNF-a or TWEAK in mice, murine TNF Receptor 2 fused with the murine IgG2a Fc region (TNFR2-Ig) or murine anti-TWEAK IgG2a mAb (mP2D10) [15] were used, or mice were administered an isotype-matched control anti-human CD20 mAb that does not cross react with murine CD20, all prepared by Biogen Idec, Inc. (Cambridge, MA). 2.3. Induction and evaluation of colitis TNBS colitis was induced by intrarectal administration of a 2% solution of TNBS in phosphate-buffered saline (PBS): ethanol (1:1) [16,17]. For acute inflammatory responses, 70 lg/g body weight of TNBS was given on day 0 and animals sacrificed on day 4. One hour prior to administration of TNBS, groups of mice were injected i.p. with the control IgG2a mAb (anti-human CD20) (10 mg/kg), TNFR-Ig (0.3 mg/kg), anti-TWEAK (mP2D10, 10 mg/kg), the combination of TNFR-Fc (0.3 mg/kg) and anti-TWEAK mP2D10 (10 mg/kg), or were untreated. For single agent and combination treatments groups, 0.3 mg/kg TNFR-Ig was employed, since 1 mg/kg of TNFR-Ig markedly ameliorated TNBS colitis but 0.3 mg/kg was much less effective as monitored by the effect on colon length and body weight. Open ulcer area was measured in captured macroscopic images using image J (NIH) software. Colon tissue was rolled and snap-frozen in liquid nitrogen. Specimens for RNA extraction or Western blotting were cut from the frozen rolled colon. Frozen sections, showing the whole length of the colon, were stained with hematoxylin and eosin. Histological scores were blindly assigned to each proximal, middle and distal segment as follows: 0 – normal, 1 – ulcer or cell infiltration limited to the mucosa, 2 – ulcer or limited cell infiltration in the submucosa, 3 – focal ulcer involving all layers of the colon, 4 – multiple lesions involving all layers of the colon or necrotizing ulcer larger than 3 mm in length. Thus the total possible histological score is 12. A set of experiments using 5–8 mice for each experimental group was performed 3 times. Total number of mice for each experimental condition was as follows; untreated TNBS colitis group, 12; control antibody, 20; TNFR-Ig, 24; anti-TWEAK mAb, 24; combination of TNFR-Ig and anti-TWEAK mAb, 24. 2.4. Western blotting Snap-frozen colon tissue was crushed and total protein was extracted using standard methods. Extracts from individuals in a given treatment group pooled, resolved by SDS–PAGE and transferred to PVDF membranes (Bio Rad, Hercules, CA, USA) which were probed using the primary antibodies against IjBa (4812; Cell Signaling Technology), phospho-IjBa (9246; Cell Signaling Technology), phosho-p100 (4810; Cell Signaling Technology), p100/ p52 (4882; Cell Signaling Technology), and GAPDH (AM4300; Applied Biosystems). Secondary antibodies were anti-mouse
16
T. Dohi et al. / Cytokine 69 (2014) 14–21
IgG-HRP (NA931VS; Amersham Biosciences) and anti-rabbit IgGHRP (NA934VS; Amersham Biosciences). Samples were visualized using SuperSignal West Pico Chemiluminescent Substrate (34079; Thermo Scientific) and BioMax Light Film (8761520; Kodak). For semi-quantification of bands, obtained images were analyzed using NIH image J and relative value to that of GAPDH of the identical sample was calculated.
(lods) greater than zero and fold change greater than 2.0 were considered significantly different. Ingenuity Pathway Analysis software (Ingenuity Systems, Inc. CA 94063) was used for analysis of upstream regulators. Upstream regulators with activation Z-scores greater than 2 or smaller than 2 were considered significant.
2.5. Cell separation and quantitative RT-PCR
Data were statistically analyzed using a software Prism 4 (GraphPad Sofware, Inc) using the method indicated in the legend.
Isolation of purified ECs, lamina propria cells (LPCs) and mesenteric lymph node cells (MLN) was performed as described previously [8]. In brief, released ECs were stained with 7-Aminoactinomycin D (7AAD), fluorescein isothiocyanate-anti-CD45 and Rphycoerythrin-anti-EpCAM antibodies (BD Biosciences, Franklin Lakes, NJ) and viable EpCAM+CD45 cells sorted as purified ECs (Moflo, Beckman Coulter, Tokyo, Japan). To separate LPCs, the ECdenuded residue was treated with collagenase A (Roche Diagnostics, Tokyo, Japan). Total RNA was prepared from EC, LPMC or whole MLN tissue describe as above. Complementary DNA was synthesized from RNA by reverse transcription, and gene expression levels were determined with Taqman gene expression assay kit for TWEAK (Mm02583406_s1), Fn14 (Mm00489103_ m1), Rela (Mm00501346_m1), Relb (Mm00485664_m1), Tnfrsf1a (Mm00441875_m1), Tnfrsf1b (Mm00441889_m1), Nfkb2 (Mm004 79810_g1) and 18s rRNA (Hs99999901_s1) and ABI PRISM 7900 Sequence Detector (Applied Biosystems). The results were transformed by the DCt/DDCt method as described by the manufacturer, with 18s rRNA used as the calibrator gene. 2.6. Gene expression profiling Total RNA was prepared individually from the whole colon mucosa of 4–8 mice for each experimental group by using RNABee RNA Isolation solvent (TEL-TEST, Inc. TX) purified with DNAse (RNAse-free DNA set) and RNeasy Mini kit (Qiagen, Tokyo, Japan). Affymetrix scans were subject to quality control (QC) measures. These tests included a visual inspection of the distribution of raw signal intensities and an assessment of RNA degradation, relative log expression (RLE), and normalized unscaled standard error (NUSE). All sample scans passed these QC metrics. CEL files were subjected to GC-content-based Robust Multi-array Average normalization [18,19]. Expression levels were log (base 2) transformed. All calculations and analyses were carried out using R and Bioconductor computational tools [20]. To identify differentially expressed genes (DEGs) in colon between groups of samples, we applied the linear modeling approach (MANOVA) to fit gene expression levels (log2 transformed) according to the defined groups of samples and Bayesian posterior error analysis as implemented by Smyth [21]. Genes that exhibited a log-odds score
2.7. Statistical analysis
3. Results 3.1. Activation of NF-jB canonical pathway in the acute colitis In order to explore the potential cooperative involvement of both TNF-a and TWEAK pathways, we induced colitis in the C57BL/6 strain, where anti-TWEAK mAb treatment alone was only modestly effective, suggesting that other pathogenic pathways, especially involving TNF-a, are in play [22,23]. First, we investigated the activation status of NF-jB canonical and non-canonical pathway activation in this model. To determine activation of NFjB in colitis at biochemical level, Western blot analysis was performed on whole colon tissue (Fig. 1A and B). Activation of IjBa was clearly observed in the mice with colitis compared with naïve colon. In contrast, there was constitutive activation of NF-jB noncanonical pathway as shown by the presence of phosph-p100 and p52 in naïve colon; however, increase of phosphorylated p100 or p52 in colitis was not apparent. Overall these results indicate that NF-jB canonical pathway was mainly activated in the whole colon of acute TNBS-induced colitis. 3.2. Fn14 and Rel A gene expression was upregulated in ECs but not other cell types Since protein extract from whole colon tissue might mask celltype specific inflammatory responses, we investigated inflammatory responses at the cellular level. For this purpose, we obtained highly purified live colonic ECs with flow cytometry, LPC and MLN from mice with colitis and performed quantitative qRT-PCR for TWEAK, Fn14, TNFR1 and 2. We also performed qRT-PCR for canonical and noncanonical NF-jB components, since their transcript levels can be modulated by NF-jB signaling [24] (Fig. 2). When compared with the level of naïve mice, statistically significant induction of TWEAK was not seen in any of the cell types, although LPC showed a trend of upregulation. Significant upregulation of Fn14 was seen only in ECs with colitis, whereas there were no changes in LPMC or MLN. Thus, in this colitis model, TWEAK/Fn14 signaling was induced mainly in ECs, rather than
Fig. 1. NF-jB canonical but not noncanonical pathway was activated. (A) Western blot analysis. Whole colon tissue was obtained from naïve mice and day 4 colitic mice treated with control IgG and blots probed with antibodies for IjBa, phospho (p)- IjBa, p-p100, p100, p52 and GAPDH. Each lane represents individual mouse. (B) Result of panel A was semiquantified normalized with levels of GAPDH. Short bars indicate mean value.
T. Dohi et al. / Cytokine 69 (2014) 14–21
17
Fig. 2. NF-jB canonical pathway was activated in epithelial cells colitis was induced and EC, LPS, MLN cells isolated as described, with the purified EC being EpCAM+CD45 . Total RNA extract was subjected to quantitative RT-PCR. Fold changes from naïve cells for each cell type are shown, as average + SD, n = 3. Two tailed P values with unpaired ttest are indicated.
hematopoietic cell types. In addition, the expression of TNFRSF1A (coding TNFR1) and TNFRSF1B (TNFR2) both showed a trend of upregulation in ECs with colitis, and induction of TNFRSF1B reached statistically significant level. The canonical NF-jB component RelA (p65) was also upregulated in EC with colitis whereas the noncanonical NF-jB components RelB (p52) and NFKB2 (p100) gene expression were not upregulated in any of the cell types. These results suggest that the EC is a key cell type in which the canonical NF-jB pathway is activated in the colitis. 3.3. Synergistic protective effect of combination treatment in colitis As a tool for TNF or TWEAK inhibition in the mouse system, we used suboptimal dose of TNFR-Ig fusion protein and anti-TWEAK monoclonal antibody, respectively. We found that combination treatment resulted in a more than additive protective effect from acute colitis as compared to single agent treatment (Fig. 3).
As shown in Fig. 3, administration of either single agent, 0.3 mg/kg of TNFR-Ig or 10 mg/kg anti-TWEAK mAb, showed limited or no efficacy with respect to the body weight (Fig. 3A), colon length (Fig. 3B) and macroscopic/histological analysis (Fig. 3C–F). The TNFR-Ig-, anti-TWEAK mAb- and control Ig-treated groups each showed a trend of better recovery of body weight as compared to the untreated group, with no significant difference between the three treated groups on day 4. In marked contrast, treatment with the combination of TNFR-Ig and anti-TWEAK mAb resulted in significantly less weight loss and more rapid recovery of body weight than any of the other groups. In addition, the colon was significantly longer in the combination-treatment group than any of other experimental groups (Fig. 3B), and closer in value to that of naïve mice (mean ± SE, 8.76 ± 0.13 cm), whereas the monotherapyor control Ig-treated groups were not significantly different from untreated in colon length. Macroscopically, large necrotizing ulcers in middle to distal colon are typically seen in acute TNBS-induced
18
T. Dohi et al. / Cytokine 69 (2014) 14–21
Fig. 3. Combined treatment with TNF-a and TWEAK inhibitors protects against TNBS-induced colitis. (A) Body weight changes (% of day 0) in each treatment group. Data are shown as a mean ± SEM. Statistically significance was found in ‘‘untreated vs. combo’’ on day 3 and 4, ‘‘Control IgG vs. combo’’ on day 3 and 4, ‘‘TNFR-Ig vs. combo’’ on day 3 and 4, and ‘‘anti-TWEAK vs. combo’’ on day 4, using two-way anova with post Bonferroni multiple comparisons. (B) Colon length of day 4. Horizontal dotted line indicated mean value for naïve colon. (C) Macroscopic appearance of colon. (D) Area of open ulcer. (E) Histological score. (F) Typical histological finding in each treatment. Frozen sections were stained with hematoxylin and eosin. Scale bar = 200 lm. In panels b–e, each dot represents an individual mouse and solid horizontal bars indicate mean value. * Difference from combo-treated group was significant with Mann Whitney analysis (P < 0.05).
colitis. Many mice treated with the combination exhibited minimal open ulcers with only redness of the mucosal surface (Fig. 3C) and the area of open ulcer was significantly less in the combinationtreated as compared to all other groups (Fig. 3D). Minimal open ulcers were also seen in many of the mice treated with
anti-TWEAK mAb alone, and this was significant as compared to untreated (P = 0.039), but did not reach significance as compared to control Ig- or TNFR-Ig treatment. Matching these macroscopic observations, combination treatment resulted in a significantly lower histological score as compared to all other groups (Fig. 3E
T. Dohi et al. / Cytokine 69 (2014) 14–21
19
and F). TNFR-Ig treatment showed some benefit, with scores significantly lower than untreated or control IgG (P = 0.0038 and 0.004, respectively), as did anti-TWEAK mAb treatment, which was significantly lower than that of the untreated group (P = 0.0197) though not as compared to the control IgG treated group. Thus, administration of each of the single agents, TNF-Ig or anti-TWEAK mAb, had subtle protective effects on the histological scores and open ulcer area, respectively. Treatment with the combination of these two agents resulted in a much greater than additive effect of the monotherapies. 3.4. Gene expression profiling indicates synergistically suppressed NF-
jB pathway with combination treatment in colitis To investigate the mechanism for the synergistic protective effect of combined TNF-a and TWEAK inhibitor treatment in mouse colitis, gene expression profiles were analyzed using RNA extracted from whole colon. The whole data is available at GSE53835. A total of 781 genes changed more than 2-fold in both the control Ig-treated and untreated mice as compared to naïve mice. We investigated whether any of these genes were specifically changed as a result of monotherapy. However, in a comparison of the TNFR-Ig treated vs. untreated or control Ig-treated group, no such genes were found. Likewise, there was only 1 gene (2810416G20Rik) detected in comparison of the anti-TWEAK mAb-treated and untreated group or control Ig-treated group. These results indicated that neither the TNFR-Ig nor anti-TWEAK mAb treatment significantly affected colitis-induced changes in gene expression, consistent with our observation that single agent treatment had minimal effect on the measures of colitis. In contrast, from this list of 781 genes, there were 142 genes that were changed significantly by combination treatment as compared to both of the untreated and control Ig-treated groups, including 131 downregulated and 11 upregulated genes. Among these downregulated genes, we identified the 62 genes whose levels of expression were the most strongly affected such that they were not significantly different (less than 2-fold) between the group with combination treatment and naïve mice. This list of 62 genes (Fig. 4) indicates those upregulated in colitis induction but maintained by combination treatment at nearly normal levels, and thus the major molecular changes that combo treatment altered to protect against colitis. Using this list of 62 downregulated genes, we searched the possible upstream pathways that were affected by combination treatment using the Upstream Regulator Analysis Software, leading to a list of possible upstream pathways. In addition to TNF-a, TNFRSF1A (TNF receptor 1), TWEAK, lipopolysaccharide (LPS) and innate sensors like Toll-like receptor (TLR), were identified as potential as upstream regulators. Of note, IKBKG (IKK-c/NEMO) and NFKBIA (IjBa were also included as upstream regulators, suggesting that the canonical pathway of NF-jB was significantly downregulated. Fig. 4 shows the target genes of these upstream pathways that were found in our data set. Since differential target genes for the NF-jB canonical and noncanonical pathways have not been well identified, we could not discern canonical vs. noncanonical signaling from this analysis. Thus, the result of gene expression profiling was generally interpreted that downregulation of the NF-jB pathway and gene expression downstream of TNF-a, TWEAK, and TLR stimuli were the major events downregulated by combo treatment.
Fig. 4. A gene set strongly downregulated by combined TNF-a inhibitor and antiTWEAK mAb treatment in mouse TNBS-induced colitis. Shown are 62 genes, which were upregulated in whole colon in both the untreated and control Ig-treated groups, with levels in the combination treatment group comparable to those in naïve mice. Fold changes are as indicated with the color key. Black squares indicate the target molecules in the data set that were the basis for determination of the upstream regulators, LPS, NFKBIA, TNF-a and TWEAK, using Ingenuity Pathway Analysis software.
4. Discussion In this study, we unveiled a striking synergistic contribution of the TNF-a and TWEAK pathways in a model of acute colitis using combination treatment with pathway inhibitors, which cannot be
simply explained as an additive effect of both agents. This effect is apparently mediated through the canonical NF-jB signaling in EC. Based on these novel findings, we discuss possible mechanisms for
20
T. Dohi et al. / Cytokine 69 (2014) 14–21
the TWEAK and TNF-a pathways and their contributions to colitis as illustrated in Fig. 5. Interplay between TWEAK and TNF-related pathways has been suggested in in vitro assays [25–27]. In this study, TWEAK signaling is highly inducible in colon EC of C57BL/6 mice as indicated by Fn14 upregulation in EC from colitis mice. We also found that TNFR are broadly expressed in the EC, and mRNA for TNFR2 was notably increased with acute colitis. These results indicated that the site of synergy of TNF-a and TWEAK was in EC but not other cell types, involving Fn14 and TNFR2 signaling pathways in addition to TNFR1. Previous studies demonstrated activation of non-canonical NF-jB pathway induced by TWEAK/Fn14 [11–14], and that TNFR2 was also able to activate non-canonical NF-jB pathway interfering canonical pathway induced with TNFR1 [28]. Therefore, we examined the involvement of NF-jB noncanonical pathway in this model. However, in this acute colitis model, noncanonical NF-jB activation was constitutively high, and neither further increase nor effect of any treatment apparent, even in transcript measures dissected at the cellular level. In contrast, NF-jB canonical pathway activation was prominent in this model and also the target of combination therapy with anti-TWEAK antibody and TNF-inhibitor therapy. Thus the interplay between TWEAK and TNF-a in this acute colitis model is presumably occurring through their signaling of the canonical NF-jB pathway. Clues as the mechanisms underlying the synergy between the TNF-a and TWEAK pathways derive from our further analysis of the colitis tissues, namely the mild effects of TNF-a inhibitor and anti-TWEAK as single agents. In the anti-TWEAK treated group there was trend of more limited ulcer-area than TNFR-Ig treatment, which may reflect a direct preventive effect from EC damage by blocking TWEAK. The TNFR-Ig-treated group showed a lower histological score than untreated and control-Ig treated group, whereas the anti-TWEAK treated group did not. This seems to due to less cell infiltration reflecting the anti-inflammatory effect achieved by partially neutralizing TNF-a. Perhaps each of these effects alone is not sufficient to protect against colitis, but synergistic when combined. Further supporting this notion, we note that in
contrast to spontaneous inflammatory polyarthritis and inflammatory bowel disease in TNF-overexpressing mice [29], TWEAK overexpression in naïve mice induces tissue changes in the liver and heart without overt inflammation [30,31] though TWEAK may potentiate and sustain the inflammatory response through a feed forward loop, inducing the production by EC of cytokines, chemokines and matrix metalloproteinases [6] which then further recruit and activate macrophages and lymphocytes to produce TNF-a as well as TWEAK. While TWEAK also directly supports the activation of latent, membrane bound TNF-a from intestinal ECs in the presence of IL-13 [7] and thereby enhances EC death, our observation that anti-TWEAK but not TNFR-Ig monotherapy reduced ulcer area in acute colitis suggests that TWEAK promotes EC death through other TNF-a-independent mechanisms, consistent with the TWEAK-induced cell death that is TNF-a-independent in other systems [32]. In addition to TWEAK and TNF-a signaling, LPS appears to be a major upstream regulator based on the colitis gene expression analysis. There may also be an unknown mechanism whereby TWEAK cross talks with some of downstream pathways of TLRs in a TNF-a-independent or dependent manner. In addition, we recently demonstrated that the TWEAK/Fn14 pathway promotes chronic TNBS-induced colitis with fibrosis through the production of TSLP by ECs [33]. It is unknown if this is mediated by canonical or noncanonical NF-jB. However, we suggest that delayed and prolonged noncanonical NF-jB activation by TWEAK [14] may be important in chronic colitis. In conclusion, our study indicates that TNF-a and TWEAK/Fn14 pathways mutually support activation of NF-jB canonical pathway in a significant synergistic manner, which is critical in the tissue damage of acute inflammation.
Database linking Database: GSE53835. Acknowledgements This work was supported partly by grants and contracts from the program Grants-in-Aid for Scientific Research (C) (T. D. and Y.I. K.) and Grants-in Aid for Young Scientists (B) (T. O.) from the Ministry of Education, Cultures, Sports, Science, and Technology; a grant from the National Center for Global Health and Medicine 21-110, 22-205, 21-129, 25-104, (T. D. and Y.I. K.), Ministry of Health, Labor, and Welfare; Grants for Research on Measures for Intractable Diseases and Research on Health Sciences focusing on Drug Innovation from Ministry of Health, Labor and Welfare of Japan (T. D. and Y.I. K.); MEXT-Supported Program for the Strategic Research Foundation at Private Universities (T. D.). We thank Dr. M. Tamura-Nakano in NCGM EM Support Unit for her contribution to pathological analysis and Dr. Normand E. Allaire in Biogen Idec Inc for his valuable discussion on gene profiling analysis. References
Fig. 5. Model of TWEAK and TNF-a actions in colitis. TNF-a is produced by activated white blood cells (WBC, including macrophages and T cells). ECs also contribute active TNF-a by TWEAK/Fn14-mediated release from an endogenous pool. TNF-a activates NF-jB canonical pathway in white blood cell (WBC) types, inducing tissue damage and release of other proinflammatory mediators. TWEAK and TNF-a support and sustain canonical NF-jB pathway in EC. TWEAK may also activate the noncanonical NF-jB pathway. Signals from TLRs (not shown) also contribute to promote infiltration and activation of WBC and EC and may interplay with signaling associated with TNF-a and TWEAK.
[1] Kaplan MJ, Lewis EE, Shelden EA, Somers E, Pavlic R, McCune WJ, et al. The apoptotic ligands TRAIL, TWEAK, and Fas ligand mediate monocyte death induced by autologous lupus T cells. J Immunol 2002;169:6020–9 (Baltimore, Md: 1950). [2] Maecker H, Varfolomeev E, Kischkel F, Lawrence D, LeBlanc H, Lee W, et al. TWEAK attenuates the transition from innate to adaptive immunity. Cell 2005;123:931–44. [3] Meighan-Mantha RL, Hsu DK, Guo Y, Brown SA, Feng SL, Peifley KA, et al. The mitogen-inducible Fn14 gene encodes a type I transmembrane protein that modulates fibroblast adhesion and migration. J Biol Chem 1999;274: 33166–76.
T. Dohi et al. / Cytokine 69 (2014) 14–21 [4] Burkly LC, Michaelson JS, Hahm K, Jakubowski A, Zheng TS. TWEAKing tissue remodeling by a multifunctional cytokine: role of TWEAK/Fn14 pathway in health and disease. Cytokine 2007;40:1–16. [5] Burkly LC, Dohi T. The TWEAK/Fn14 pathway in tissue remodeling: For better or for worse. Adv Exp Med Biol 2011;691:305–22. [6] Dohi T, Borodovsky A, Wu P, Shearstone JR, Kawashima R, Runkel L, et al. TWEAK/Fn14 pathway: a nonredundant role in intestinal damage in mice through a TWEAK/intestinal epithelial cell axis. Gastroenterology 2009;136: 912–23. [7] Kawashima R, Kawamura YI, Oshio T, Son A, Yamazaki M, Hagiwara T, et al. Interleukin-13 damages intestinal mucosa via TWEAK and Fn14 in Mice-A pathway associated with ulcerative colitis. Gastroenterology 2011;141: 2119–29. [8] Son A, Oshio T, Kawamura YI, Hagiwara T, Yamazaki M, Inagaki-Ohara K, et al. TWEAK/Fn14 pathway promotes a T helper 2-type chronic colitis with fibrosis in mice. Mucosal Immunol 2013. [9] Bradley JR. TNF-mediated inflammatory disease. J Pathol 2008;214: 149–60. [10] Chicheportiche Y, Chicheportiche R, Sizing I, Thompson J, Benjamin CB, Ambrose C, et al. Proinflammatory activity of TWEAK on human dermal fibroblasts and synoviocytes: blocking and enhancing effects of anti-TWEAK monoclonal antibodies. Arthritis Res 2002;4:126–33. [11] Brown SA, Ghosh A, Winkles JA. Full-length, membrane-anchored TWEAK can function as a juxtacrine signaling molecule and activate the NF-kappaB pathway. J Biol Chem 2010;285:17432–41. [12] Brown SA, Richards CM, Hanscom HN, Feng SL, Winkles JA. The Fn14 cytoplasmic tail binds tumour-necrosis-factor-receptor-associated factors 1, 2, 3 and 5 and mediates nuclear factor-kappaB activation. Biochem J 2003;371:395–403. [13] Roos C, Wicovsky A, Muller N, Salzmann S, Rosenthal T, Kalthoff H, et al. Soluble and transmembrane TNF-like weak inducer of apoptosis differentially activate the classical and noncanonical NF-kappa B pathway. J Immunol 2010;185:1593–605 (Baltimore, Md: 1950). [14] Saitoh T, Nakayama M, Nakano H, Yagita H, Yamamoto N, Yamaoka S. TWEAK induces NF-kappaB2 p100 processing and long lasting NF-kappaB activation. J Biol Chem 2003;278:36005–12. [15] Campbell S, Burkly LC, Gao HX, Berman JW, Su L, Browning B, et al. Proinflammatory effects of TWEAK/Fn14 interactions in glomerular mesangial cells. J Immunol 2006;176:1889–98 (Baltimore, Md: 1950). [16] Dohi T, Fujihashi K, Kiyono H, Elson CO, McGhee JR. Mice deficient in Th1- and Th2-type cytokines develop distinct forms of hapten-induced colitis. Gastroenterology 2000;119:724–33. [17] Dohi T, Fujihashi K, Rennert PD, Iwatani K, Kiyono H, McGhee JR. Hapteninduced colitis is associated with colonic patch hypertrophy and T helper 2type responses. J Exp Med 1999;189:1169–80.
21
[18] Irizarry RA, Bolstad BM, Collin F, Cope LM, Hobbs B, Speed TP. Summaries of Affymetrix GeneChip probe level data. Nucl Acids Res 2003;31:e15. [19] Li C, Hung Wong W. Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application. Genome Biol 2001:2. RESEARCH0032. [20] Gentleman R. Bioinformatics and computational biology solutions using R and bioconductor. Springer Science+Business Media; 2005. [21] Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genetics Mol Biol 2004:3. Article3. [22] Shen C, de Hertogh G, Bullens DM, Van Assche G, Geboes K, Rutgeerts P, et al. Remission-inducing effect of anti-TNF monoclonal antibody in TNBS colitis: mechanisms beyond neutralization? Inflamm Bowel Dis 2007;13:308–16. [23] Neurath MF, Fuss I, Pasparakis M, Alexopoulou L, Haralambous S, Meyer, et al. Predominant pathogenic role of tumor necrosis factor in experimental colitis in mice. Eur J Immunol 1997;27:1743–50. [24] Allen IC, Wilson JE, Schneider M, Lich JD, Roberts RA, Arthur JC, et al. NLRP12 suppresses colon inflammation and tumorigenesis through the negative regulation of noncanonical NF-kappaB signaling. Immunity 2012;36:742–54. [25] Ikner A, Ashkenazi A. TWEAK induces apoptosis through a death-signaling complex comprising Receptor-interacting Protein 1 (RIP1), Fas-associated Death Domain (FADD), and Caspase-8. J Biol Chem 2011;286:21546–54. [26] Vince JE, Chau D, Callus B, Wong WW, Hawkins CJ, Schneider P, et al. TWEAKFN14 signaling induces lysosomal degradation of a cIAP1-TRAF2 complex to sensitize tumor cells to TNFalpha. J Cell Biol 2008;182:171–84. [27] Wicovsky A, Salzmann S, Roos C, Ehrenschwender M, Rosenthal T, Siegmund D, et al. TNF-like weak inducer of apoptosis inhibits proinflammatory TNF receptor-1 signaling. Cell Death Differ 2009;16:1445–59. [28] Cabal-Hierro L, Lazo PS. Signal transduction by tumor necrosis factor receptors. Cell Signal 2012;24:1297–305. [29] Kontoyiannis D, Pasparakis M, Pizarro TT, Cominelli F, Kollias G. Impaired on/ off regulation of TNF biosynthesis in mice lacking TNF AU-rich elements: implications for joint and gut-associated immunopathologies. Immunity 1999;10:387–98. [30] Jain M, Jakubowski A, Cui L, Shi J, Su L, Bauer M, et al. A novel role for tumor necrosis factor-like weak inducer of apoptosis (TWEAK) in the development of cardiac dysfunction and failure. Circulation 2009;119:2058–68. [31] Jakubowski A, Ambrose C, Parr M, Lincecum JM, Wang MZ, Zheng TS, et al. TWEAK induces liver progenitor cell proliferation. J Clin Invest 2005;115: 2330–40. [32] Burkly LC, Michaelson JS, Zheng TS. TWEAK/Fn14 pathway: an immunological switch for shaping tissue responses. Immunol Rev 2011;244:99–114. [33] Cultrone A, de Wouters T, Lakhdari O, Kelly D, Mulder I, Logan E, et al. The NFkappaB binding site located in the proximal region of the TSLP promoter is critical for TSLP modulation in human intestinal epithelial cells. Eur J Immunol 2013;43:1053–62.