A key role for gp130 expressed on peripheral sensory nerves in pathological pain

Abstract

Interleukin-6 (IL-6) is a key mediator of inflammation. Inhibitors of IL-6 or of its signal transducing receptor gp130 constitute a novel class of anti-inflammatory drugs, which raise great hopes for improved treatments of painful inflammatory diseases such as rheumatoid arthritis. IL-6 and gp130 may enhance pain not only indirectly through their proinflammatory actions but also through a direct action on nociceptors (i.e., on neurons activated by painful stimuli). We found indeed that the IL-6/gp130 ligand-receptor complex induced heat hypersensitivity both in vitro and in vivo. This process was mediated by activation of PKC-delta via Gab1/2/PI(3)K and subsequent regulation of TRPV1, a member of the transient receptor potential (TRP) family of ion channels. To assess the relevance of this direct pain promoting effect of IL-6, we generated conditional knock-out mice, which lack gp130 specifically in nociceptors, and tested them in models of inflammatory and tumor-induced pain. These mice showed significantly reduced levels of inflammatory and tumor-induced pain but no changes in immune reactions or tumor growth. Our results uncover the significance of gp130 expressed in peripheral pain sensing neurons in the pathophysiology of major clinical pain disorders and suggest their use as novel pain relieving agents in inflammatory and tumor pain.

Figure 1.

Small-size neurons from SNS-gp130^−/− lacked gp130 immunoreactivity but were otherwise normal. A, Double labeling of SNS-gp130^−/− and gp130^fl/fl DRG neurons with anti-gp130 and anti-Na_v1.8. Scale bar, 15 μm. B, Live imaging of DRG neurons showed colabeling of gp130 immunoreactivity and Na_v1.8 immunoreactivity in 90% of neurons from gp130^fl/fl mice, but lack of gp130 membrane immunoreactivity in 90% of Na_v1.8-immunoreactive neurons in SNS-gp130^−/− mice.

Figure 2.

Normal morphology of sensory ganglia and peripheral nerve in SNS-gp130^−/−. A, B, Distributions of IB4 and CGRP immunoreactivity were similar in DRG sections obtained from gp130^fl/fl and SNS-gp130^−/− mice. Percentages of IB4- or CGRP-immunoreactive neurons in DRG sections were also comparable for the two mouse strains. This indicates that peptidergic and nonpeptidergic small-size neuron populations were in general normal in the SNS-gp130^−/− mice. All experiments were repeated in at least 3 animals. C, D, Electron microscopy on nonmyelinated nerve fibers of saphenous nerves of gp130^fl/fl control and SNS-gp130^−/− mice at the age of 4–6 months. Asterisks demarcate nonmyelinated axons; arrows point to cytoplasmic processes of the nonmyelinating Schwann cells. Dark structures surrounding larger caliber axons are myelin sheaths. Note that there were no pathological alterations in the mutant nerves.

Figure 3.

Heat hyperalgesia in pain models required neuronal gp130. A, Changes in heat sensitivity following unilateral hindpaw injection of CFA in gp130^fl/fl and SNS-gp130^−/− mice. Paw withdrawal latency in response to ramp heat stimuli applied to the inflamed area was significantly attenuated at all time points in SNS-gp130^−/− mice (n = 18) compared with gp130^fl/fl (n = 17). Asterisks indicate significant differences (p < 0.001, ANOVA) between the two groups. B, In contrast, no difference in the paw diameters was observed between strains, suggesting that an equal degree of CFA inflammation occurred despite the deletion of neuronal gp130. C, D, PWL in response to ramp-shaped heat stimuli applied to the plantar side of the hindpaw ipsilateral (□) and contralateral (■) to tumor in gp130^fl/fl (C, n = 23) and SNS-gp130^−/− (D, n = 17) mice. After tumor induction in gp130^fl/fl mice, paw withdrawal latency decreased significantly (p < 0.05, ANOVA), starting from day 1 after injection and persisted over the 10 d of investigation. Mice with sensory neuron-specific deletion of gp130 developed significant heat hypersensitivity only on day 8 after tumor induction. Asterisks indicate significant differences (p < 0.001) between ipsilateral and contralateral paw. E, F, Discharge profiles of heat-sensitive C-fibers innervating the dorsal site of the hindpaw recorded in an in vitro skin-nerve preparation from gp130^fl/fl (E) and SNS-gp130^−/− (F) mice. Mean discharge rates in heat-sensitive C-fibers from gp130^fl/fl tumor mice (n = 8) were significantly higher compared with those from control (naive) mice (E, n = 12, p < 0.05, ANOVA). Tumor induction in SNS-gp130^−/− mice did not affect the discharge rates of the heat-sensitive fibers (F, p > 0.05, ANOVA). G, Weights of plantar and dorsal tumors were similar in mutant and floxed mice (n.s., not significant).

Figure 4.

IL6-induced heat hyperalgesia and significance of neuronal gp130. A, IL-6 mRNA was expressed in tumors of both mouse strains but absent in LL2 cell cultures; and B, IL-6 levels (in pg cytokine/mg protein) were significantly higher in tumor homogenates (plantar and dorsal) compared with control tissue (muscle and spinal cord) isolated from 11 mice. C, D, Intraplantar injection of HIL-6 (1 μg in 10 μl) in gp130^fl/fl (n = 8) and SNS-gp130^−/− mice (n = 10) evoked a drop of PWL in response to heat stimulation. Asterisks indicate the data points at which a significant difference (p < 0.001, ANOVA) between the ipsilateral (□) and contralateral (■) side was observed. PWL decreased to significantly lower values in gp130^fl/fl (n = 8) compared with SNS-gp130^−/− mice (p < 0.01, ANOVA, n = 10). E, F. Discharge profiles of heat-sensitive C-fibers before and after HIL-6 (1 ng/ml) application on the receptive fields in an in vitro skin-nerve preparation from gp130^fl/fl (E) and SNS-gp130^−/− (F) mice. HIL-6 application for 5 min induced a significant increase in the mean rate of discharge of the heat C-nociceptors from gp130^fl/fl (E, p < 0.01, paired t test, n = 7). Heat-sensitive C-fibers from SNS-gp130^−/− mice were not affected by HIL-6 stimulation (F, p > 0.05, Wilcoxon rank test, n = 7). G, H, I_heat was significantly facilitated after HIL-6 (H), and the increase in peak current amplitudes was accompanied by a shift in activation threshold temperature (G).

Figure 5.

Downstream signaling of IL-6/gp130 via Gab1/2, PI₃K and PKC-δ. A, B, Capsaicin-induced ionic currents were transiently and significantly facilitated by HIL-6. C, HIL-6 induced phosphorylation of the adapter proteins Gab1 and Gab2 and PI₃K, respectively. D, The facilitation of I_caps was dose-dependently inhibited by the PI₃K inhibitor wortmannin (n = 6). E, Stimulation of neurons with HIL-6 resulted in a translocation of PKC-δ toward the plasma membrane in neurons from gp130^fl/fl but not from SNS-gp130^−/− mice. F, To quantify PKC-δ translocation, fluorescence intensities were measured at the periphery and the center of the cell by performing a confocal line scan. The ratio k of the two was then calculated as detailed in Materials and Methods: k = 1, no translocation; k > 1, translocation to the PM; k < 1, cytoplasmic localization of PKC-δ. G, The facilitation of I_caps was inhibited by the PKC-δ inhibitor rottlerin (n = 6). H, The drop in PWL after HIL-6 injection was significantly attenuated in TRPV1^−/− mice compared with wt littermates (n = 6).