Objective: To analyze the types and functions of CD34+ cells in full-thickness skin defect wounds of normal mice and diabetic mice by single-cell RNA sequencing. Methods: This study was an experimental study. The CD34+ cell lineage tracing mouse was produced, and the visualization of CD34+ cells under the fluorescent condition was realized. Six male CD34+ cell lineage tracing mice aged 7-8 weeks (designated as diabetic group) were intraperitoneally injected with streptozotocin to establish a diabetic model, and full-thickness skin defect wounds were prepared on their backs when they reached 13 weeks old. Another 6 male CD34+ cell lineage tracing mice aged 13 weeks (designated as control group) were also subjected to full-thickness skin defect wounds on their backs. On post-injury day (PID) 4, wound tissue was collected from 3 mice in control group and 2 mice in diabetic group, and digested to prepare single-cell suspensions. CD34+ cells were screened using fluorescence-activated cell sorting, followed by single-cell RNA sequencing. The Seurat 4.0.2 program in the R programming language was utilized for dimensionality reduction, visualization, and cell clustering analysis of CD34+ cell types, and to screen and annotate the marker genes for each CD34+ cell subpopulation. Kyoto encyclopedia of genes and genomes (KEGG) and gene ontology (GO) enrichment analysis was performed to analyze the differentially expressed genes (DEGs) of CD34+ fibroblasts (Fbs), smooth muscle cells (SMCs), keratinocytes (KCs), and chondrocyte-like cells (CLCs) in the wound tissue of two groups of mice for exploring cellular functions. Results: On PID 4, CD34+ cells in the wound tissue of both groups of mice were consisted of 7 cell types, specifically endothelial cells, Fbs, KCs, macrophages, T cells, SMCs, and CLCs. Among these, Fbs were further classified into 5 subpopulations. Compared with those in control group, the proportions of CD34+ endothelial cells, Fbs subpopulation 1, Fbs subpopulation 4, KCs, and CLCs in the wound tissue of mice were increased in diabetic group, while the proportions of CD34+ Fbs subpopulation 2, Fbs subpopulation 3, and SMCs were decreased. The marker genes for annotating CD34+ CLCs, endothelial cells, Fbs subpopulation 1, Fbs subpopulation 2, Fbs subpopulation 3, Fbs subpopulation 4, Fbs subpopulation 5, KCs, macrophages, SMCs, and T cells were respectively metastasis-associated lung adenocarcinoma transcript 1, fatty acid binding protein 4, Gremlin 1, complement component 4B, H19 imprinted maternally expressed transcript, Dickkopf Wnt signaling pathway inhibitor 2, fibromodulin, keratin 5, CD74 molecule, regulator of G protein signaling 5, and inducible T-cell co-stimulator molecule. KEGG and GO enrichment analysis revealed that, compared with those in control group, DEGs with significant differential expression (SDE) in CD34+ Fbs from the wound tissue of mice in diabetic group on PID 4 were significantly enriched in terms related to inflammatory response, extracellular matrix (ECM) organization, regulation of cell proliferation, and aging (with Pvalues all <0.05), DEGs with SDE in CD34+ SMCs were significantly enriched in terms related to cell migration, apoptotic process, positive regulation of transcription, and phagosome (with P values all <0.05), DEGs with SDE in CD34+ KCs were significantly enriched in terms related to mitochondrial function, transcription, and neurodegenerative diseases (with P values all <0.05), and DEGs with SDE in CD34+ CLCs were significantly enriched in terms related to rhythm regulation, ECM, and viral infection (with P values all <0.05). Conclusions: CD34+ cells display high heterogeneity in the healing process of full-thickness skin defect wounds in both normal mice and diabetic mice. The significantly enriched functions of DEGs with SDE in CD34+ cell subpopulations in the wound tissue of the two mouse groups are closely related to the wound healing process.
Read full abstract