Full text 2026

Integrating bulk and single-cell RNA sequencing identifies and validates lactylation-related signatures in diabetic foot ulcers

Peng X, Wang W, Nie X, et al.

Full text

Loading PDF… Expand reader Download

Abstract

Diabetic foot ulcers (DFU) is severe complications of diabetes mellitus, affecting 15–25% of diabetic patients and imposing substantial healthcare burdens. Lactylation has emerged as a critical regulator in inflammation and wound healing processes. However, the role of lactylation-related mechanisms in DFU pathogenesis remains poorly understood. We integrated the lactylation-related genes (LRGs) and systematically analysed the expression levels of these genes to characterize a gene signature related to DFUs through bulk RNA-seq and single-cell RNA-seq (scRNA-seq) datasets. Single-cell RNA sequencing identified ten major cell populations in DFU tissues. Augur prioritization and AUCell analysis revealed significantly elevated lactylation-related gene (LRG) activity in fibroblasts and multiple immune-related cell types in DFU, with fibroblasts exhibiting the strongest transcriptional perturbation. Subclustering analysis defined a DFU-enriched fibroblast subset (DFU_Fibroblasts), in which hdWGCNA identified red and brown gene modules tightly associated with this population. Bulk RNA-seq integration identified six differentially expressed LRGs enriched in mitochondrial respiratory chain organization and pyruvate metabolism, suggesting metabolic remodeling. Machine learning approaches further identified four hub LRGs (USB1, COX5A, LDHA, and NFU1) with robust diagnostic performance (AUC = 0.904 in the training cohort). Immune infiltration analysis revealed altered macrophage polarization and NK cell imbalance in DFU. At the single-cell level, hub LRG–high fibroblasts displayed glycolysis-oriented metabolic features, early pseudotime positioning, and enhanced outgoing signaling toward monocytes via the PPIA–BSG and ANGPTL2–ITGA5 + ITGB1 axes. Molecular docking indicated favorable binding between capsaicin and COX5A and LDHA, and qPCR validation confirmed significant upregulation of all four hub genes in DFU samples. Our study identifies a lactate-associated transcriptional program in DFU that is linked to fibroblast heterogeneity, metabolic reprogramming, and alterations in the immune microenvironment. The identified hub LRGs (COX5A, LDHA, USB1 and NFU1) exhibit diagnostic potential and may represent candidate molecular targets within this metabolic–inflammatory network. These findings provide a mechanistic framework for understanding the involvement of lactylation-related pathways in DFU progression and support further investigation into targeted therapeutic strategies.

Keywords

Machine Learning Diabetic Foot Ulcer Immune Microenvironment Metabolic Reprogramming Fibroblast Heterogeneity Single-cell Rna Sequencing Lactylation