Background/Purpose: Disease flares, or episodic escalating inflammation, is a hallmark of autoimmune diseases, like Rheumatoid Arthritis (RA). They are particularly hard to predict and treat and often signify an overall worsening of disease. The biologic mechanism underpinning disease flares is not well understood, potentially due to limitations of mouse models. Type 1 interferon (IFN) is thought to be a pathogenic cytokine in autoimmune diseases by pre-conditioning cells for hyper-reactive responses during episodic inflammatory challenges. However, type 1 IFN has not been investigated in an in vivo episodic inflammatory arthritis model. Therefore, we tested the hypothesis that loss of type 1 IFN will reduce the flaring inflammation in an episodic zymosan induced arthritis (ZIA) challenge.

Methods: We performed intra-articular knee injections of 180 μg of zymosan (Zym) or PBS into female C57BL6 and IFN receptor 1 knock out (Ifnar1-/-) mice (n=4-10 per group). The same knees were then injected either 42 days or 68 days later with 180 μg of Zym (Fig1 A). Serial cheek bleeds for serum amyloid A (SAA) ELISA and joint widths were collected. After 21 days the mice were euthanized for joint μCT and histology. Patella surface was quantified and automated histomorphometry was performed to measure synovial area and cell counts within the synovium. ANOVA with Tukey’s post-hoc, t-test or paired t-test were used to assess differences

Results: In clinically resolved joints, residual elements of synovial pathology were still modestly increased compared to PBS injected controls with some infiltrates and thickening of the synovial lining (Fig. 1B). Strikingly, flaring joints demonstrated a large increase (~2x) in synovial area and cell counts compared to acute inflammation controls demonstrating escalating inflammation after episodic zymosan challenges (Fig 1B-C). Patella surface area is also increased in flared joints suggesting increased bone metabolism (Fig 1D). When Zym was instead re-injected at D68, the trained phenotype persisted as demonstrated by the increased AUC of joint widths (Fig1 E). However, when Ifnar1 null mice are re-injected at D42 there was no difference in SAA, joint widths or D21 synovial area compared to WT controls (Fig 2A-C).

Conclusion: Our results establish ZIA as a useful model for investigation of training and arthritis disease flare, and demonstrate that type 1 IFN does not play a role in changing joint inflammatory response to episodic zymosan challenges. This is in contrast with the known role of type 1 IFN in controlling many aspects of inflammation in acute and distinct episodic contexts. Further work to dissect mechanisms of training and the temporal and cell specific IFN responses to episodic inflammation is on-going.

Figure 1. Episodic zymosan induces a joint flare. A) The study design. B) Average synovial area and cell counts via automated histomorphometry, each point is one knee average across 3 anatomic levels in the medial compartment, One-Way ANOVA with Tukey’s Post-hoc. C) Representative images of Control, Resolved, Acute and Flare knees. D) uCT analysis of patella bone surface area, t-test. E) Joint widths percent of starting size over time with the 2nd Zym injection at D68 and AUC calculated from D0-D68 vs D68-D140 (right panel, paired t-test). Un-injected contralateral joints used as controls.

Figure 2. IFNAR1 KO does not modify the flare. A) Serum amyloid A concentration after 2nd ZIA induction at D42. B) Joint widths difference compared to D0 after 2nd ZIA induction at D42. # indicates p < 0.05 IFNAR KO Acute vs IFNAR KO Flare; $ indicates p < 0.05 WT KO Acute vs WT KO Flare in A and B, repeated one-way ANOVA with Tukey’s post-hoc. C) Average synovial area in acute and flaring inflammation between WT and Ifnar1-/-. each point is one knee average across 3 anatomic levels in the medial compartment.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/training-for-increased-inflammatory-arthritis-in-mice-is-not-modulated-by-type-1-interferon/