Interleukin-6 (IL-6) is an important mediator of inflammation. In addition to cells involved in inflammation, sensory nociceptive neurons express the IL-6 signal-transducer glycoprotein 130 (gp130). These neurons are not only involved in pain generation but also produce neurogenic inflammation by release of neuropeptides such as calcitonin gene-related peptide (CGRP). Whether IL-6 activation of sensory neurons contributes to the induction of inflammation is unknown. This study explored whether the action of IL-6 on sensory neurons plays a role in the generation of neurogenic inflammation and arthritis induction. In SNS-gp130−/− mice lacking gp130 selectively in sensory neurons and appropriate control littermates (SNS-gp130flox/flox), we induced antigen-induced arthritis (AIA), and assessed swelling, histopathological arthritis scores, pain scores, expression of CGRP in sensory neurons, serum concentrations of CGRP and cytokines, and the cytokine release from single cell suspensions from lymph nodes and spleens. In wild-type mice CGRP release was determined during development of AIA and, in cultured sensory neurons, upon IL-6 stimulation. Compared to SNS-gp130flox/flox mice SNS-gp130−/− mice showed significantly weaker initial swelling, reduced serum concentrations of CGRP, IL-6, and IL-2, no inflammation-evoked upregulation of CGRP in sensory neurons, but similar histopathological arthritis scores during AIA. During the initial swelling phase of AIA, CGRP was significantly increased in the serum, knee and spleen. In vitro, IL-6 augmented the release of CGRP from cultured sensory neurons. Upon antigen-specific restimulation lymphocytes from SNS-gp130−/− mice released more interleukin-17 and interferon-γ than lymphocytes from SNS-gp130flox/flox mice. In naive lymphocytes from SNS-gp130flox/flox and SNS-gp130−/− mice CGRP reduced the release of IL-2 (a cytokine which inhibits the release of interleukin-17 and interferon-γ). IL-6 signaling in sensory neurons plays a role in the expression of arthritis. Selective deletion of gp130 signaling in sensory neurons reduces the swelling of the joint (most likely by reducing neurogenic inflammation) but increases some proinflammatory systemic cellular responses such as the release of interleukin-17 and interferon-γ from lymphocytes upon antigen-specific restimulation. Thus IL-6 signaling in sensory neurons is not only involved in pain generation but also in the coordination of the inflammatory response.

In tumor cells, interleukin-6 (IL-6) signaling can lead to activation of the epidermal growth factor receptor (EGFR), which prolongs Stat3 activation. In the present experiments, we tested the hypothesis that IL-6 signaling activates EGFR signaling in peripheral and spinal nociception and examined whether EGFR localization and activation coincide with pain-related behaviors in arthritis. In vivo in anesthetized rats, spinal application of the EGFR receptor blocker gefitinib reduced the responses of spinal cord neurons to noxious joint stimulation, but only after spinal pretreatment with IL-6 and soluble IL-6 receptor. Using Western blots, we found that IL-6-induced Stat3 activation was reduced by gefitinib in microglial cells of the BV2 cell line, but not in cultured DRG neurons. Immunohistochemistry showed EGFR localization in most DRG neurons from normal rats, but significant downregulation in the acute and most painful arthritis phase. In the spinal cord of mice, EGFR was highly activated mainly in the chronic phase of inflammation, with localization in neurons. These data suggest that spinal IL-6 signaling may activate spinal EGFR signaling. Downregulation of EGFR in DRG neurons in acute arthritis may limit nociception, but pronounced delayed activation of EGFR in the spinal cord may be involved in chronic inflammatory pain.