Fig 2: Mesenchymal IKK2-dependent signals promote arthritic manifestations in TNF-mediated arthritis. a Representative histological images of H&E, Toluidine Blue (TB) and Tartrate-Resistance Acid Phosphatase (TRAP) stained ankle joint sections and histological score of synovial hyperplasia, cartilage destruction and bone erosion of hTNFtg Ikk2f/f and hTNFtg Ikk2Ms-KO mice (n = 10; age: 11 weeks). b Representative histological images of H&E-, TB- and TRAP-stained ankle joint sections and histological score of synovial hyperplasia, cartilage destruction and bone erosion of Ikk2f/f (n = 7) and Ikk2Ms-KO mice (n = 8) subjected to CAIA at the peak of disease (Day 8). (3 experiments, n = 20–22 mice per genotype). Scale bar: 100 and 50 µm. Data are presented as the mean ± SEM. **P < 0.01 and ***P < 0.001, by two-tailed Student’s t-test

Fig 3: Increased death sensitization of Ikk2-deficient SFs in TNF-mediated signals. a Representative images depicting TUNEL+ or CC3+ cells in sections of Ikk2f/f and Ikk2Ms-KO mice (naive, hTNFtg (week8) or CAIA-treated (D8)) (n = 6 per genotype). Note the CC3+ cartilage cells (blue arrow), lining (black) and sublining (red) SFs, and other cells (open arrow) within necroptic areas. b Quantitation of TUNEL+ cells found in synovial membrane area of Ikk2f/f and Ikk2Ms-KO in naive (n = 5–7) and arthritic conditions (hTNFtg (n = 4–7); CAIA (n = 4)). c Survival rates of Ikk2f/f, hTNFtg Ikk2f/f SFs, and WT MEFs treated with different TNF concentrations, in the presence or absence of ML120b inhibitor (representative experiment, n = 5). d Representative immunoblot of IKK2 expression after the application of TAT-Cre in SF cultures derived from Ikk2f/f and hTNFtg Ikk2f/f mice, and IkB degradation and JNK1/2 phosphorylation patterns after TNF or IL-1ß treatment of Ikk2-deficient SFs. Relative quantitations are depicted for each blot. e Representative survival rates of Ikk2?/? and hTNFtg Ikk2?/? SFs and Ikk2-/- MEFs and control Ikk2f/f and hTNFtg Ikk2f/f SFs and WT MEFs treated with different TNF concentrations (n = 15, 4 experiments). Scale bar: 100 µm. Data are presented as the mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001, by two-tailed Student’s t-test (b) or one-way ANOVA (c, d). Unprocessed original scans of blots are shown in Supplementary Fig. 8

Fig 4: Ripk3-mediated signals are indispensable for persisting synovitis in hTNFtg Ikk2Ms-KO mice. a Representative histological images of H&E-, TB- and TRAP-stained ankle joint sections of 8-week-old hTNFtg Ikk2f/f Ripk3-/- (n = 11) and hTNFtg Ikk2Ms-KO Ripk3-/- mice (n = 13) in comparison to hTNFtg Ikk2Ms-KO and Ikk2f/f mice. Note the differences in synovium and adjacent tissues in respective areas indicated by the arrows. b Survival rates of Ikk2?/? Ripk3-/- and hTNFtg Ikk2?/? Ripk3-/- SF cultures treated with TNF (T), zVAD (Z), and nec-1 (N) as indicated. c Representative immunodetection of pMlkl, Mlkl and Ripk3 levels in the cell extracts of Ikk2?/? Ripk3-/- and hTNFtg Ikk2?/? Ripk3-/- SFs treated as indicated compared to control Ikk2?/? Ripk3+/+ and hTNFtg Ikk2?/? Ripk3+/+ samples, respectively (n = 4 per genotype, 2 experiments). d Representative immunodetection of Casp8 (C8), Cleaved Casp3 (CC3), Xiap, cFLIP(s) and IKK2 levels in cell extracts of Ikk2?/? Ripk3-/- and hTNFtg Ikk2?/? Ripk3-/- SFs treated as indicated compared to control Ikk2f/f Ripk3-/- and hTNFtg Ikk2f/f Ripk3-/- samples, respectively (the non-hTNFtg control cultures are indicated as WT) (n = 4 per genotype, 2 experiments). Scale bar: 500 µm. Data are presented as the mean ± SEM. ***P < 0.001 one-way ANOVA test (b). Unprocessed original scans of blots are shown in Supplementary Fig. 8

Fig 5: Activation of NF-?B by over-expression of IKK2-EE induces cFlip and rescues TAK1-/- MEFs from TRAIL induced cell death.Inducible over-expression of dominant active IKK2-EE in TAK1-/- MEFs was accompanied by an increase in phosphorylation of p65 and I?B-a degradation (A), and blocked sensitivity to TRAIL (1 µg/ml, 48 h) (B). Expression of both cIAP2 and cFlipL genes was elevated as determined by real time RT-PCR, using the housekeeping gene 18S rRNA as internal control (C). Western blot for cFlip and cIAP1 in TAK1-/- MEFs stably infected with IKK2EE (TAK1-/-uasIKK2EE) after addition of 10 nM 4HT for different times (* indicates non-specific band).