Background/Purpose: Pain persisting beyond the resolution or control of clinical signs of rheumatoid arthritis (RA) decreases quality of life for millions of people. Unfortunately, this pain does not respond well to typical analgesics, and, while tumor necrosis factor (TNF) is a pivotal cytokine in RA, the persistence of pain in the face of anti-TNF treatment indicates that control of this cytokine alone is insufficient to treat this pain. We have shown that in the K/BxN serum transfer model male C57Bl/6 (WT) mice develop transient inflammation and a corresponding tactile allodynia (TA), which persists beyond the resolution of inflammation. Here, we aimed to understand the role of TLR4 associated spinal cytokines, specifically TNF and interferon (IFN) β in the development of this persistent, post-inflammatory pain state, which is mediated through Toll-like receptor (TLR) 4 signaling.

Methods: K/BxN sera (100μl) was injected into male WT, Tnf^-/–, Tlr4^-/–, and Ifnar1^-/- mice on Days 0 and 2. Ankle width and withdrawal thresholds were examined over 28 days. Separate groups of mice were injected intrathecally (IT) with anti-TNF antibody or with IFNβ.

Results: In male mice, there is a pivotal time (~2 weeks) beyond which administration of a TNF inhibitor or a TLR4 antagonist can no longer abrogate the neuropathic pain state in this animal model. Hence, we examined spinal cords from arthritic mice on day 10 for differences in gene expression of key inflammatory cytokines associated with allodynia in male Tlr4^-/- and WT mice, as the WT male mice develop persistent pain, but the Tlr4^-/- do not. We found IFNβ transcripts decreased in WT mice (average fold change (AFC): 0.41) and were increased in Tlr4^-/- mice (AFC: 18.84). In contrast, TNF transcripts increased in WT mice (AFC 1.33), and remained unchanged in Tlr4^-/- mice (AFC 0.96). The levels of mRNA expression for IL-1β and IL-6 were equivalent between the two strains.

To further understand the roles of TNF and IFNβ in K/BxN-induced arthritis and pain, we assessed male Ifnar1^-/- and Tnf^-/- mice. The inflammatory phase TA was attenuated in Ifnar1^-/- mice (1.18g relative to 0.5g in WT males, p <0.05); however, these mice developed persistent TA while the late phase TA was reduced in Tnf^-/- mice (1.29g relative to 0.74g in WT males, p < 0.05), suggesting neither TNF nor IFNβ is completely responsible for the persistent pain state. Next, male WT mice were injected IT with anti-TNF antibody or with IFNβ. Neither pharmacologic treatment alone affected TA (p >0.05). However, when Tnf^-/- male mice were given IT IFNβ, we saw a persistent reversal in TA. Finally, WT male mice were treated with the combination of both anti-TNF and IFNβ and showed a reversal in their persistent pain state, with a day 28 threshold of 1.4g relative to 0.27g in vehicle treated controls (p < 0.05).

Conclusion: These results suggest that the pharmacological co-modulation of TNF and IFNβ is necessary for the treatment of persistent arthritis-related pain.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/effective-treatment-of-persistent-arthritis-pain-requires-co-modulation-of-tnf-and-type-i-interferon/