Sensory nerve-derived CGRP controls osteoclastogenesis by limiting macrophage bioenergetics in bone repair

Bone healing is a tightly orchestrated, multiphase process that requires coordinated interactions between immune cells and skeletal cells. Sensory nerves act as intrinsic effectors of the inflammatory response, whose role in osteoimmunology during healing remains poorly defined. Using a bone healing model with sensory denervation, it's shown that sensory nerves protect bone repair by suppressing excessive osteoclastogenesis. During the acute inflammatory phase, sensory nerves are upstream regulators of macrophage activation. At the molecular level, calcitonin gene-related peptide (CGRP), a sensory neuron–derived neuropeptide, is identified to modulate macrophage activation by restricting key functions such as migration, phagocytosis, and pro-inflammatory cytokine production. Importantly, CGRP rapidly constrains adenosine triphosphate (ATP) synthesis with limited mitochondrial respiration in activating macrophages, accompanied by downregulating genes associated with oxidative phosphorylation and mitochondrial complex components. Following the metabolic checkpoint, macrophages exposed to CGRP show attenuated osteoclastogenic capacity, with decreased secretion of multiple key factors that support osteoclast differentiation and survival. Together, these findings indicate the neuro–immune–metabolic axis in bone healing, where sensory nerve–derived CGRP influences macrophage bioenergetics and thereby contributes to osteoimmunological regulation. It emphasizes the potential of incorporating sensory signals into therapeutic strategies, particularly those targeting immunometabolism in bone regeneration. Overall design: For in vivo samples, healing tissue within the bone defect in different groups at different time points was collected. Samples were tested using the DNBSEQ platform. Reads of low quality, adapter contamination, and high content of unknown bases (N) were removed. After obtaining clean reads, we used HISAT to align the clean reads to the reference genome sequence (Rattus_norvegicus_10116.NCBI.GCF_015227675.2_mRatBN7.2.v2201). Bioinformatic analysis was performed using Dr. Tom platform and RStudio (GSVA). For in vitro samples, RAW 264.7 were then treated with LPS alone or in combination with CGRP (10 nm) for 6 h. RNA was extracted by PureLink RNA Mini Kit (Thermo Fisher Scientific). Samples were tested using the Illumina NextSeq 2000 platform. Reads of low quality, adapter contamination, and high content of unknown bases (N) were removed using the Cutadapt tool. After obtaining clean reads, we used STAR to align the clean reads to the reference genome sequence (Mus_musculus_mm10/GCF_000001635.20). Bioinformatic analysis was conducted with RStudio (key package info e.g. FeatureCounts, DESeq2, FactoMineR, etc.). Samples that failed quality control were not included in the analysis.