Abstract
G-CSF or CSF-3, originally defined as a regulator of granulocyte lineage development via its cell surface receptor (G-CSFR), can play a role in inflammation, and hence in many pathologies, due to its effects on mature lineage populations. Given this, and because pain is an extremely important arthritis symptom, the efficacy of an anti-G-CSFR mAb for arthritic pain and disease was compared with that of a neutrophil-depleting mAb, anti-Ly6G, in both adaptive and innate immune-mediated murine models. Pain and disease were ameliorated in Ag-induced arthritis, zymosan-induced arthritis, and methylated BSA/IL-1 arthritis by both prophylactic and therapeutic anti-G-CSFR mAb treatment, whereas only prophylactic anti-Ly6G mAb treatment was effective. Efficacy for pain and disease correlated with reduced joint neutrophil numbers and, importantly, benefits were noted without necessarily the concomitant reduction in circulating neutrophils. Anti-G-CSFR mAb also suppressed zymosan-induced inflammatory pain. A new G-CSF-driven (methylated BSA/G-CSF) arthritis model was established enabling us to demonstrate that pain was blocked by a cyclooxygenase-2 inhibitor, suggesting an indirect effect on neurons. Correspondingly, dorsal root ganglion neurons cultured in G-CSF failed to respond to G-CSF in vitro, and Csf3r gene expression could not be detected in dorsal root ganglion neurons by single-cell RT-PCR. These data suggest that G-CSFR/G-CSF targeting may be a safe therapeutic strategy for arthritis and other inflammatory conditions, particularly those in which pain is important, as well as for inflammatory pain per se.