1506 Multimodal transcriptomics highlight fibroblast heterogeneity and pathological signaling networks in keloid

Keloid, a tumor like scarring disorder, is caused by wounding in genetically predisposed individuals with poorly defined pathogenesis. To gain pathomechanistic details on keloid development, we performed single-cell RNA-sequencing and spatial transcriptomics on keloid edge and adjacent non-lesional skin. Upon single-cell analysis, we characterized ten major cell types based on expression of known markers. Among those, fibroblasts significantly enriched at keloid edge compared to the adjacent skin. Further sub-clustering of fibroblasts revealed three fibroblast subtypes, marked by ASPN, CCL19, and THSD4, respectively. Notably, ASPN+ and CCL19+ fibroblasts consistently increased in keloid edge in all patients and were mutually marked by known markers of human reticular dermal fibroblasts. Spatial mapping of fibroblast subtypes via spatial transcriptomics and RNAscope confirmed the deep dermal location of ASPN+ and CCL19+ fibroblasts and revealed their spatial organization at the keloid edge. Differential gene expression and regulon analysis indicated that keloid edge fibroblasts possessed many extracellular matrix organization and inflammation related genes, driven by their uniquely active transcription factors. To probe potential upstream signaling to fibroblasts, we performed cell-cell interaction analysis on fibroblasts and their neighboring cells, which revealed abundant signaling interactions specifically activated in keloid edge, including TGFb, CXCL, and CCL signaling pathways. Overall, our research suggested that hyperproliferation of reticular fibroblasts in keloid edge may drive keloid pathogenesis, partially by producing excessive amounts of extracellular matrix and promoting inflammation. These findings highlight the role of reticular fibroblasts in keloid pathogenesis and provide proof for targeting reticular fibroblasts at the keloid edge as an effective approach for keloid treatments.