Identification of a novel small-molecule modulator targeting SNX10 to inhibit osteoclastic bone resorption

Current antiresorptive therapies reduce bone loss by eliminating osteoclasts or inhibiting their formation. However, these approaches could disrupt osteoclast-osteoblast communication and cause serious complications with long-term usage. There is a need for developing new therapies that selectively inhibit resorptive function, while preserving osteoclast-mediated coupling effects to osteoblasts. Sorting nexin 10 (SNX10), an autosomal recessive osteopetrosis (ARO)-associated gene, plays a role mainly in osteoclast bone resorptive function. However, its potential as a target for developing therapeutic agents for bone disorders remains unexplored. In this study, we employed a multi-step approach combining artificial intelligence (AI)-driven virtual screening with high-throughput screening methods and functional assays to identify small molecules targeting SNX10 that inhibit bone resorption without impairing osteoclast formation. Our lead compound AW-006 emerged as the top candidate across all validation methods, selectively inhibiting osteoclast resorptive function while maintaining normal osteoclastogenesis in vitro. Mechanistic analyses indicated that AW-006 interacts with SNX10 and reduces its thermal stability, while molecular docking predicted binding within the PI(3)P-binding pocket and associated conformational changes affecting residues involved in PI(3)P binding and structural integrity. We discovered that SNX10 interacts with the key vesicular trafficking regulator Rab7 in living cells, and AW-006 abnormally enhances this interaction, dysregulating normal podosome belt formation in osteoclasts. Furthermore, the compound's anti-resorptive efficacy was validated in ovariectomized mice, demonstrating its therapeutic potential in estrogen deficiency-induced bone loss. Our study identifies AW-006 as a novel anti-resorptive candidate and highlights SNX10 as a promising therapeutic target for bone disorders.