A pathogenic role for JNK1 in passive K/BxN serum transfer arthritis. (A) Clinical arthritis scores and ankle swelling in WT (gray circles, n = 15), Jnk1^−/− (black squares, n = 15), and Jnk2^−/− mice (black circles, n = 8) injected with 150 μL of K/BxN serum on day 0. Values are means ± SEM, *P < 0.05 vs. WT or Jnk2^−/− mice by two-way ANOVA with Bonferroni post hoc comparisons. (B) Histological scores for joint inflammation, erosion, and cartilage damage in WT, Jnk1^−/−, and Jnk2^−/− mice (eight mice per group) on day 10 after serum transfer. Values are means ± SEM, *P < 0.05 vs. WT or Jnk2^−/−. (C) Whole-joint RNA from control (gray bars) and arthritic (black bars) mice was extracted, and cytokines and MMP mRNAs were analyzed by Q-PCR. Jnk1^−/− vs. WT, *P < 0.05. (D) Serum IL-6 was measured by ELISA on day 10 after serum transfer. (E) Whole-joint protein was extracted, protein content was quantified, and IL-1β levels were analyzed by ELISA. Results are expressed as means of six mice per group ± SEM, *P < 0.05 vs. WT or Jnk2^−/− mice.

JNK1 in mast cells is critical for development of serum transfer arthritis. (A) WT neutrophils do not restore arthritis. WT, Jnk1^−/−, and Jnk1^−/− mice that were i.v. injected each day for 4 d with CD45.1⁺ congenic BM received 150 μL K/BXN sera. Arthritis clinical scores are expressed as means ± SEM (five mice per group). WT (gray diamonds), Jnk1^−/− + CD45.1⁺ BM (black triangles), and Jnk1^−/− (black squares). *P < 0.05 vs. WT using two-way ANOVA with post hoc Bonferroni comparisons. (B–D) WT, but not Jnk1^−/−, BMMC reconstitution restores arthritis. BM was harvested from femurs and tibias of WT and Jnk1^−/− donors and cultured for 4 wk, serially monitoring the cultures until they were >99% pure for c-Kit⁺/FcεRI⁺ cells. c-Kit mutant (Pretty2) recipient mice were injected with 10⁷ BMMCs and after 6 wk were injected with pooled K/BXN sera. (B) Toluidine blue staining of joints from MCs of reconstituted mice. Synovial MC were detected in Pretty2 mice after either WT or Jnk1^−/− BMMC engraftment. (C) MCs of reconstituted mice and controls (five mice per group) received 150 μL K/BXN sera. Arthritis clinical scores are expressed as means ± SEM. WT (black squares), Pretty2 (gray squares), Pretty2 + WT BMMC (black triangles), Pretty2 + Jnk1^−/− BMMC (gray diamonds). Clinical scores for Pretty2 mice and Pretty2 + Jnk1^−/− BMMCs were significantly lower than the scores for WT and Pretty2 + WT BMMCs by two-way ANOVA with Bonferroni post hoc test (*P < 0.05). (D) Histological scores for joint inflammation and erosion in control and reconstituted mice (five mice per group) on day 10 after serum transfer. Values are means ± SEM, *P < 0.05 vs. WT and Pretty2 + WT BMMCs by two-way ANOVA and Bonferroni post hoc test.

IL-1β secretion through FcγRs is impaired in Jnk1^−/− mast cells. BM from WT and Jnk1^−/− mice was harvested and cultured with IL-3–rich medium for 4 wk. (A) IL-1β in lysates of WT, Jnk1^−/−, and Jnk2^−/− BMMCs after treatment with the indicated stimuli was measured by ELISA in triplicate (representative of four experiments). (B) Whole RNA was extracted from WT and Jnk1^−/− BMMCs, and IL-1β mRNA was analyzed by Q-PCR in triplicate. Results are expressed as means ± SEM, *P < 0.05 vs. WT or Jnk2^−/−. Representative of three independent experiments.

JNK1 in bone marrow-derived cells is critical for serum transfer arthritis. (A and B) Adult WT and Jnk1^−/− mice were irradiated and reconstituted with WT or Jnk1^−/− bone marrow (eight mice per group). After 8 wk, the chimeras were injected with 150 μL of pooled K/BxN serum i.p. on day 0. Ankles from each mouse were prepared for histological scoring as above. Shown are clinical scores expressed as means ± SEM, *P < 0.05 vs. Jnk1^−/− BM > Jnk1^−/−. **P < 0.05 vs. WT BM> WT using two-way ANOVA with post hoc Bonferroni comparisons. (C and D) Histological scores of joints from the mice described above. Results are expressed as means ± SEM, *P < 0.05 vs. WT BM > WT. **P < 0.05 vs. WT BM > WT and WT BM > Jnk1^−/−. (E and F) Whole-joint protein from the mice above was extracted, and IL-1β content was analyzed by ELISA. Gray bars are from uninjected mice and black bars are from day 8 injected mice. Results are expressed as means ± SEM, *P < 0.05 vs. WT BM > WT. **P < 0.05 vs. WT BM > WT and WT BM > Jnk1^−/−.

Fewer degranulated MCs in Jnk1^−/− mice. Ankles from WT and Jnk1^−/− mice were prepared for histology staining with toluidine blue. (A) Representative toluidine blue-stained sections on day 10 after serum transfer. Arrows indicate degranulated MCs. (B) Number of intact MCs located in the joints of WT (n = 15), Jnk1^−/− (n = 15), and Jnk2^−/− (n = 8) mice on day 10 after serum transfer. MCs were visualized by toluidine blue staining and quantified as described. Results are expressed as means ± SEM, *P < 0.05 vs. WT or Jnk2^−/−. (C) Cytospin preparations of cells obtained by peritoneal lavage 30 min after PBS (a and c) or by K/BxN serum (b) injection were stained with anti-phospho-c-Jun antibody (a and b) or toluidine blue (c). Arrows indicate peritoneal MCs. (D) Number of intact MCs located around ankle joints from control and Jnk1^−/− mice (five mice per group) 16 h after K/BxN sera injection, quantified by a laser-scanning cytometer. Results are expressed as means ± SEM, *P < 0.05 vs. WT.

Injection of a JNK inhibitor prevents and attenuates inflammatory arthritis. (A) WT mice (six per group) were i.p. injected with 150 μL of pooled K/BxN sera on day 0 and treated with control peptide or D-JNKi daily (20 μg/g) starting at day 0 (A) or day 4 (B). D-JNKi treatment significantly reduced the onset of swelling (A) and reduced swelling in established disease (B). Results are expressed as means ± SEM, *P < 0.05 vs. control-treated WT. Representative of two independent experiments. (C) Whole-joint RNA from WT control (gray bars) and WT arthritic (black bars) mice was extracted on day 7, and IL-6 and IL-1β mRNAs were analyzed by Q-PCR. IL-1 and IL-6 mRNA expression was lower in D-JNKi peptide-treated mice (*P < 0.05). (D) WT mice (five per group) were injected with 150 μL of pooled K/BxN sera and D-JNKi or peptide control i.p. and killed after 16 h. The number of intact MCs located in their joints was determined by toluidine blue staining and quantified by a laser-scanning cytometer. Results are expressed as means ± SEM, *P < 0.05 vs. WT. (E) MC-deficient Pretty2 mice (four per group) were injected with 200 μL of pooled K/BxN sera i.p. on day 0 and treated with control peptide or D-JNKi daily (20 μg/g) starting at d 0. Shown are the clinical scores. Results are expressed as means ± SEM, *P < 0.05 vs. control-treated WT.