Back to 2025 Abstracts
SINGLE-CELL MULTI-OMICS OF MOUSE DIABETIC WOUND HEALING REVEALS STROMAL CELL CROSSTALK DYSFUNCTION
Mateusz S. Wietecha
*1, Jingbo Pang
2, Avin Hafedi
1, Shalyn Keiser
1, Timothy J. Koh
21Oral Biology, College of Dentistry, University of Illinois Chicago, Chicago, IL; 2Kinesiology and Nutrition, College of Applied Health Sciences, University of Illinois Chicago, Chicago, IL
Type II diabetes mellitus is one of the most prevalent diseases associated with diabetic foot ulcers and impaired wound healing. The diabetic mouse model with dysfunctional leptin receptor (db/db) has been used in basic and translational studies due to its phenotypes (hyperphagia, hypometabolism, obesity, diabetes), including for delayed diabetic wound healing research, although a characterization of its temporal cellular dynamics has not been performed.
Excisional wounds were harvested from wild-type (WT) and diabetic (db/db) mice at 3, 6 and 10 days post-injury. Epithelium was discarded and stromal cells from digested dermal tissue were isolated, purified for live cells, and tagged using a panel of 18 antibodies against surface markers of major stromal cell types, including immune cells (neutrophils (CD45,CD11b,Ly6G), monocytes/macrophages (CD45,CD11b,F4/80), T lymphocytes (CD45,CD3,CD4) and non-immune cells (endothelial cells (CD31), fibroblasts (CD26,CD140a)). The cells were processed for 10X Genomics single-cell RNA-sequencing, followed by stringent quality control measures, resulting in a final dataset of 49963 high-quality cells. Downstream bioinformatics data analysis was conducted in R (4.4.1) using RStudio (2024.09.0) interface and the single-cell analysis package Seurat (5.1.0) and the cell-cell communication analysis package CellChat (2.1.2). Cell clustering was performed at multiple resolutions followed by cell typing analyses using both protein- and RNA-level data. Temporal cell abundance and pseudobulk analyses were performed to evaluate healing progression. Sub-clustering analyses were performed to analyze cell subtypes. Ligand-receptor analyses were performed to analyze putative cell-cell communication pathways. FACS analyses were performed on additional samples to validate major bioinformatics findings.
We identified 8 major cell types (in decreasing abundance): Fibroblasts, Macrophages, Neutrophils, Endothelial cells, Smooth muscle cells, T cells, Muscle progenitor cells, Skeletal muscle cells. Pseudobulk and gene co-expression analyses showed a marked delay in wound healing progression in db/db vs WT mice. Correlation of protein- and RNA-level data revealed high concordance between cell markers and their gene expressions in major cell types. Protein-level data was useful for cell typing and subtyping, identifying 4 fibroblast, 4 macrophage, and 2 neutrophil subtypes. These stromal cell subtypes showed dysregulated dynamics across the timecourse of healing in db/db vs WT mice. CellChat analyses revealed striking differences in cell-cell communication dynamics between fibroblasts, macrophages and neutrophils across healing, with the SPP1 pathway being especially dysregulated between these cells in db/db vs WT mice.
We present a comprehensive multi-omic characterization of mouse skin diabetic wound healing at the single-cell level, revealing temporal dysfunctions in major cell type abundances and stromal cell crosstalk dynamics.
Back to 2025 Abstracts
