Published online May 26, 2026. doi: 10.4252/wjsc.v18.i5.116280
Revised: December 1, 2025
Accepted: February 24, 2026
Published online: May 26, 2026
Processing time: 199 Days and 21.5 Hours
Chronic non-healing diabetic wounds are driven by persistent inflammation, impaired fibroblast function, and defective neovascularization. Mesenchymal stem cell (MSC)-based therapies show promise, but their efficacy is limited by suboptimal paracrine and immunomodulatory activity.
To determine whether erythropoietin-overexpressing MSCs (EPO-MSCs) enhance diabetic wound healing and to delineate the underlying immune mechanisms, with particular focus on serum amyloid A3-positive (Saa3+) macrophages and C-C motif chemokine ligand (CCL)-centered cell-cell communication.
Human umbilical cord-derived MSCs were transduced with an EPO-expressing lentiviral vector and compared with negative control MSCs (NC-MSCs). Fibrobl
EPO-MSCs retained typical MSCs surface markers and tri-lineage differentiation while secreting higher levels of EPO. Compared with NC-MSCs, EPO-MSCs sig
EPO engineering augments the pro-regenerative and immunomodulatory functions of MSCs, promoting expansion of Saa3+ macrophages and CCL-centric crosstalk with neutrophils to accelerate diabetic wound repair. EPO-MSCs there
Core Tip: This study shows that erythropoietin-overexpressing mesenchymal stem cells (EPO-MSCs) accelerate diabetic wound repair by enhancing fibroblast migration, an
- Citation: Ma BD, Zhang SJ, Shao YM, Jin RR, Sun L, Lv PJ, Yue H, Hu SK, Ma XW. Erythropoietin-overexpressing mesenchymal stem cells accelerate diabetic wound healing via steroid signaling pathway modulation. World J Stem Cells 2026; 18(5): 116280
- URL: https://www.wjgnet.com/1948-0210/full/v18/i5/116280.htm
- DOI: https://dx.doi.org/10.4252/wjsc.v18.i5.116280
Chronic non-healing wounds are a serious and devastating complication of diabetes mellitus, affecting a substantial proportion of people with diabetes and accounting for the majority of non-traumatic lower limb amputations worldwide[1]. The lifetime risk of developing a diabetes-related foot ulcer has been estimated at approximately 19%-34%, and ulcer recurrence and 5-year mortality rates remain unacceptably high, com
Researchers have focused on mesenchymal stem cells (MSCs) as promising therapeutic options in recent years owing to their remarkable regenerative capabilities and immunomodulatory functions[4]. MSCs promote wound repair by releasing cytokines and growth factors that enhance cell migration, stimulate angiogenesis, and regulate innate and adaptive immune responses[5,6]. Preclinical studies and early-phase clinical trials suggest that MSC-based interventions are generally safe and can accelerate closure of diabetic and other chronic cutaneous ulcers, yet response rates are heterogeneous and many wounds still fail to heal completely under MSC therapy. To overcome limitations related to cell survival, engraftment, and paracrine potency, next-generation strategies have been developed in which MSCs are genetically engineered or preconditioned to express angiogenic, anti-inflammatory, or cytoprotective molecules. Several of these engineered MSC products have entered early clinical testing in indications such as cardiovascular, neurological, and inflammatory diseases, underscoring the translational momentum toward more potent, mechanism-tailored MSC therapies[7,8].
Erythropoietin (EPO) is a hormone produced in the kidneys that promotes red blood cell generation[9]. Beyond its hematopoietic role, EPO exerts tissue-protective effects by binding to specific receptors and blocking pro-inflammatory cytokines, thereby reducing apoptosis and promoting cell survival. Preclinical evidence indicates that EPO treatment increases local microvessel density, hydroxyproline content, and vascular endothelial growth factor levels, resulting in considerably faster wound closure[10]. In a high-glucose microenvironment, EPO stimulates MSCs to secrete vascular endothelial growth factor, and the interaction between EPO and MSC-derived factors has been shown to enhance wound healing while attenuating monocyte infiltration[11].
The integration of the molecular machinery of EPO with MSC-based delivery systems represents a novel therapeutic paradigm for diabetic wound repair. Genetic engineering of MSCs to constitutively express EPO (EPO-MSCs) may help overcome the pharmacokinetic limitations of recombinant protein therapy while sustaining localized biological activity in the wound bed. In this diabetic mouse study, we investigated whether EPO-MSCs exhibit superior efficacy to unmodified MSCs in facilitating cutaneous wound healing and sought to elucidate the underlying mechanisms. By integrating transcriptomics, proteomics, and single-cell RNA sequencing, we systematically characterized changes in gene expre
The umbilical cords of full-term human neonates were aseptically extracted following approval from the Institutional Ethics Committee (Approval No. 2024-004), with written informed consent obtained from the donors. The tissue fra
The surface marker expression of the MSCs was evaluated by conducting flow cytometry assays. After the MSCs were resuspended in phosphate buffered saline (PBS), they were incubated at room temperature for 15 minutes with 4 μL of the designated antibody in the dark. The following antibodies were used for the phenotypic evaluation of the MSCs: Fluorescein isothiocyanate-conjugated CD14, CD19, CD34, CD45, allophycocyanin-conjugated CD73, CD105, phy
Adipogenic differentiation: About 2 × 104 MSCs were inoculated into each well of 24-well microplates for 10 days of induction with an adipogenic differentiation medium containing 10% foetal bovine serum, 0.5 mmol/L 3-isobutyl-1-methylxanthine, 0.5 μM dexamethasone, and 0.2 mmol/L insulin. After induction, the MSCs were immobilized with 4% paraformaldehyde, and oil red O staining (Solarbio, China) was performed to visualize intracellular lipid droplets.
Osteogenic differentiation: The MSCs were maintained for 14 days in osteogenic differentiation medium containing 10% foetal bovine serum, 10 mmol/L β-glycerophosphate, and 0.1 μM dexamethasone. After they were fixed with 4% para
Chondrogenic differentiation: Pellet cultures were established by seeding 2.5 × 105 MSCs per pellet and maintaining them for 21 days in a chondrogenic medium containing 50 μg/mL chondroitin sulfate and 10 ng/mL transforming growth factor-β3. Glycosaminoglycans were identified via Alcian blue staining (Solarbio, China), and images were captured using a polarized light microscope.
The coding sequence for EPO was amplified and subsequently cloned and inserted into the pLVX-Puro lentiviral vector. Using Lipofectamine 3000 (Invitrogen, CA, United States), HEK293T cells were cotransfected with the EPO transfer vector along with packaging plasmids (pMD2.G and psPAX2) to produce the lentiviral particles. The viral supernatants were harvested 48 hours after transfection, filtered using a 0.45-μm filter, and ultracentrifuged for 2 hours at 50000 × g and
Total protein was extracted by lysing different subgroups of MSCs using RIPA buffer. Following concentration quantification via the BCA assay, the extracted proteins were separated through sodium-dodecyl sulfate gel electrophoresis and then transferred onto polyvinylidene fluoride membranes. The blots were incubated for 1 hour at room temperature with 5% milk in 0.5% Tween-20-supplemented Tris-buffered saline. The membranes were probed with the designated primary antibodies overnight at 4 °C and then with horseradish peroxidase-labeled secondary antibodies for 2 hours at 37 °C. An Omni-ECL™ Chemiluminescence Kit (Epizyme, China) was used to visualize the immunoreactive bands, while a ChemiDoc XRS Plus system (Bio-Rad, CA, United States) was used to image the bands. The following antibodies were used in this analysis: Anti-EPO (1:1000, ThermoFisher, MA, United States), anti-glyceraldehyde-3-phosphate dehydrogenase (1:5000, Bioworld Technology, MN, United States), and horseradish peroxidase-labeled anti-rabbit IgG (1:5000, Abcam, United Kingdom).
A scratch assay was performed using human foreskin fibroblast (HFF) to assess cell migration. HFF fibroblasts were inoculated in six-well microplates, and upon reaching 90% confluence, uniform “scratches” were generated using a sterile pipette tip across the cell monolayer. EPO-MSCs and negative control MSCs (NC-MSCs) were then added to the upper Transwell compartment (0.4 μm pore size; Corning, NY, United States), and cell migration was observed after 24 hours. The scratched area was photographed using a microscope, and the extent of wound healing was quantified using the ImageJ software (National Institutes of Health, United States). The healing rate was calculated using the following formula: (Post-migration scratch area/initial scratch area) × 100%.
Cell viability was evaluated using a Cell Counting Kit-8 (Epizyme, China). Briefly, HFF cells were plated in 24-well microplates, and when they reached the logarithmic growth phase, the MSCs were placed in the upper Transwell (0.4 μm pore size) compartment for coculture with the HFF cells. After 24 hours, Cell Counting Kit-8 solution (50 μL) was added to each well, followed by 2 hours of incubation. The optical density (OD) was measured spectrophotometrically (Mo
This study was approved by the Ethics Committee of Zhengzhou Central Hospital, affiliated with Zhengzhou University (Ethics No. 2024004). All animal procedures were performed according to the NIH Guide for the Care and Use of Laboratory Animals and the ARRIVE guidelines. Male db/db mice (eight weeks old) were obtained from China Saiyuan Biotechnology and housed in specific pathogen-free settings. We maintained the ambient temperature at 22 ± 1 °C with a relative humidity of 50% ± 5% and a 12-hour/12-hour light/dark photoperiod. The mice were watered and fed ad libitum throughout the study. Diabetic wound models were established in db/db mice by creating two circular wounds, each 6 mm in full thickness, at the dorsal region. These wounds were made about 1 cm from the midline to prevent injury to the spine and important muscle groups. The control group was administered local PBS injections, whereas the experimental group was administered local injections of NC-MSCs and EPO-MSCs. The wound was divided into four qu
Hematoxylin and eosin staining, along with Masson’s trichrome staining, was performed to examine the wound healing of the tissue sections. Tissue samples were collected from the wound area and subjected to routine fixation, dehydration, embedding, and sectioning. Hematoxylin and eosin staining was performed to assess the epidermal thickness of the wound area, whereas Masson’s trichrome staining was performed to assess collagen fiber deposition. All stained sections were viewed under a light microscope, and quantitative analysis was performed using ImageJ.
Total RNA was isolated from mouse skin tissues using TRIzol reagent (Invitrogen, CA, United States). The quality and quantity of RNA were determined using a NanoDrop™ One/OneC spectrophotometer (Thermo Fisher, MA, United States). RNA integrity was assessed using an Agilent 4200 TapeStation system (Agilent Technologies, CA, United States). High-throughput transcriptome sequencing was subsequently conducted on the Illumina platform. Poly(A)-enriched mRNA was isolated from total RNA, followed by reverse transcription into cDNA and polymerase chain reaction amplification to construct the sequencing library. Before sequencing on the Illumina platform in the PE150 paired-end mode, the quality of the constructed library was evaluated using an Agilent 4200 TapeStation system.
The raw sequencing data were subjected to quality control procedures, which included the elimination of low-quality reads and adapter contamination. Clean reads were aligned to the reference genome, and gene-level count matrices were generated. Differentially expressed genes (DEGs) between groups were identified using the DESeq2 package. Raw P values from the Wald test were adjusted for multiple comparisons using the Benjamini-Hochberg method to control the false discovery rate (FDR), and genes with an adjusted P value (FDR) < 0.05 were considered significantly differentially expressed. Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) enrichment analyses were subsequently performed on the set of DEGs, and enrichment P values were likewise corrected for multiple testing using the Benjamini-Hochberg procedure.
Proteins were isolated from the skin tissues using RIPA buffer. The concentrations of the extracted proteins were mea
The obtained mass spectrometry data were analyzed using the iProteome software. Protein intensities were normalized across samples before statistical testing. Differentially expressed proteins between the NC-MSC and EPO-MSC groups were identified using DESeq2 based on normalized intensity values. Multiple-testing correction was performed using the Benjamini-Hochberg procedure, and proteins with an adjusted P value (FDR) < 0.05 were defined as differentially expressed. GO and KEGG enrichment analyses were then carried out on these differentially expressed proteins, and enrichment P values were similarly adjusted for multiple comparisons using the Benjamini-Hochberg method.
Skin tissue samples from the mice were dissected and underwent mechanical and enzymatic dissociation. Collagenase IV and DNase I digestion of the tissue yielded a single-cell suspension, which was filtered via a 70-μm strainer to eliminate undigested tissue fragments and then washed with PBS. The cell concentration was determined using an automated cell counter, and the cell density was adjusted to about 1 × 106 cells/mL. Single-cell RNA sequencing libraries were sub
Raw sequencing data were processed using the CellRanger pipeline (10 × Genomics, CA, United States) to generate gene-cell count matrices. Quality control was performed at the cell and gene levels. Cells with fewer than 200 detected genes, more than 5000 detected genes, or with > 20% of total counts derived from mitochondrial genes were excluded to remove low-quality cells, potential doublets, and stressed/dying cells. Genes expressed in fewer than three cells were also filtered out. The remaining high-quality cells were normalized and log-transformed, and principal-component analysis was carried out in Seurat. Graph-based clustering and uniform manifold approximation and projection were used for dimensionality reduction and visualization.
DEGs between clusters or conditions were identified using the Wilcoxon rank-sum test implemented in Seurat. P values were adjusted for multiple comparisons with the Benjamini-Hochberg method, and genes with an adjusted P value (FDR) < 0.05 were considered significant. Cell types were annotated based on canonical marker genes and reference datasets. Cell-cell communication analyses were performed using dedicated R packages, and the resulting ligand-receptor interaction P values or scores were corrected for multiple testing where applicable.
All statistical analyses of non-omics data were performed using Prism 8 (GraphPad). Data are presented as the mean ± SD, unless otherwise stated. Comparisons among more than two groups at a single time point were conducted using one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test. When only two groups were com
For high-throughput datasets, differential expression analyses of bulk RNA sequencing, proteomic, and single-cell RNA sequencing data were performed as described in sections from “Transcriptome sequencing of skin-tissue” to “Single-cell sequencing of skin tissue”. To account for multiple hypothesis testing in these analyses, raw P values from gene- or protein-level tests and from functional enrichment analyses were adjusted using the Benjamini-Hochberg method to control the FDR, and an adjusted P value (FDR) < 0.05 was regarded as statistically significant unless otherwise specified.
For the in vivo wound-healing experiments, the primary endpoint was the difference in wound closure between groups at predefined time points. Group sizes were determined a priori based on our pilot experiments, which indicated a large treatment effect of EPO-MSCs compared with NC-MSCs and PBS, and on previously published db/db mouse wound models using MSC-based therapies that employed comparable numbers of animals per group. These considerations suggested that the chosen sample size would be adequate to detect biologically meaningful differences at a two-sided α of 0.05 with a statistical power of approximately 80% for large effect sizes. We acknowledge that a formal prospective power calculation was not performed, and this is recognized as a methodological limitation of the present study.
The flow cytometry results indicated that the MSCs were negative for CD14, CD19, CD34, CD45, and HLA-DR but positive for CD73, CD90, and CD105 (Supplementary Figure 1). The formation of lipid droplets was detected through Oil Red O staining. Matrix mineralization was observed via Alizarin Red S staining, and acid proteoglycans were identified via Alcian blue staining (Supplementary Figure 1). These findings match the established characteristics of MSCs. In this study, two types of MSCs were used: EPO-MSCs and NC-MSCs. The results of western blotting assays revealed sig
Both EPO-MSCs and NC-MSCs facilitated the healing of cutaneous wounds in diabetic mice (Figure 2A and B). EPO-MSCs exhibited better wound healing properties than NC-MSCs (Figure 2A and B). EPO-MSCs more effectively increased epidermal thickness, whereas both groups showed similar effects on collagen deposition (Figure 2C and D).
Transcriptomic sequencing of the cutaneous wound tissue from diabetic mice in both the NC-MSC and the EPO-MSC groups was performed to analyze changes in gene expression. Compared to the NC-MSC mice, the EPO-MSC mice presented 114 upregulated DEGs and 78 downregulated DEGs (Figure 3A, Supplementary Table 1). The KEGG pathway enrichment results revealed the involvement of these genes in key pathways associated with diabetic complications, including the tumor necrosis factor (TNF), mitogen-activated protein kinases, interleukin (IL)-17, phosphatidylinositol 3-kinase/protein kinase B, and advanced glycation end product-receptor for advanced glycation end product signaling pathways (Figure 3B). The GO enrichment results further suggested that the DEGs were primarily associated with biological processes, including angiogenesis, positive regulation of cytokine production, and the proliferation of epithelial cells (Figure 3C).
Proteomic analysis on the cutaneous wound tissue from diabetic mice in the NC-MSC and EPO-MSCs groups was performed to evaluate the changes in protein expression. Compared to the NC-MSC group, the EPO-MSCs group presented 66 upregulated and seven downregulated proteins that were differentially expressed (Supplementary Table 2). The biological processes associated with the upregulated differentially expressed proteins were linked mainly to intermediate cytoskeleton organization, intermediate filament-based processes, and axodendritic transport (Figure 4A). The cellular component categories were predominantly associated with the mitochondrial envelope, mitochondrial inner membrane, and organelle envelope (Figure 4A). The main molecular function categories included microtubule minus-end binding, transcription coactivator activity, and electron transfer activity (Figure 4A). The KEGG enrichment results revealed the primary implications of the upregulated differentially expressed proteins in pathways related to amino sugar and nucleotide sugar metabolism, oxidative phosphorylation, apoptosis, and sphingolipid metabolism (Figure 4B).
In contrast, the downregulated differentially expressed proteins were associated with biological processes such as heterocycle catabolic processes, aromatic compound catabolic processes, and cellular responses to IL-6 (Figure 4C). Their cellular component categories were primarily associated with the U2-type prespliceosome, U12-type spliceosomal complex, and prespliceosome (Figure 4C). The molecular function categories were predominantly associated with GDP phosphatase activity, eukaryotic initiation factor 4E binding, and UDP phosphatase activity (Figure 4C). The KEGG enrichment results suggested a predominant correlation of downregulated differentially expressed proteins with longevity-regulating pathways across multiple species, nucleotide metabolism, RNA degradation, and taste transduction (Figure 4D).
Single-cell RNA sequencing was conducted on cutaneous wound tissues obtained from the NC-MSC-treated and EPO-MSCs-treated diabetic mice (Figure 5A). In total, 12 distinct cell populations were identified in the cutaneous wound tissues of both groups, with neutrophils and the mononuclear phagocytes (MPs) representing the most abundant cell types (Figure 5B). Compared to the NC-MSC-treated mice, the EPO-MSC-treated mice presented significant upregulation of both neutrophils and MPs. Within the macrophage lineage, macrophages played a predominant role in both groups (Figure 5C). Differential expression analysis of macrophages revealed that the expression of Arg1, a marker of M2 macrophages, was significantly higher in the EPO-MSCs group (Figure 5D). These findings suggested that compared to the NC-MSC group, the EPO-MSC group had a greater proportion of anti-inflammatory M2 macrophages.
The macrophages were classified into nine subpopulations, with cluster 5 showing a minimal presence in the NC-MSC group and a notable increase in the EPO-MSC group (Figure 5E). Differential expression analysis of macrophages in cluster 5 revealed high expression of serum amyloid A3 (Saa3), which indicated that Saa3+ macrophages played an important role in the enhanced wound healing observed among EPO-MSCs-treated diabetic mice (Figure 5F).
To elucidate the function of Saa3+ macrophages, differential gene expression analysis of this cell population was performed. The results of the GO enrichment analysis revealed that the biological processes predominantly associated with Saa3+ macrophages included acute inflammatory response, regulation of angiogenesis, and the chemotaxis and migration of neutrophils (Figure 6A). Regarding cellular components, the enriched categories were related primarily to the myelin sheath, membrane raft, and membrane microdomain (Figure 6A). The results of the molecular functions analysis revealed significant associations with cytokine activity, chemokine activity, and receptor-ligand interactions (Figure 6A). The KEGG enrichment results revealed that the DEGs in Saa3+ macrophages were associated mainly with the IL-17, hypoxia-inducible factor-1, and TNF signaling pathways (Figure 6B).
Cell-cell communication analysis was conducted to evaluate the strength of interaction among different cell types. Our results indicated that the weight of interaction between MPs and neutrophils was the highest (Figure 6C). The C-C motif chemokine ligand (CCL) signaling pathway was identified as the primary mediator of this interaction (Figure 6D). Finally, we quantified and visualized the contribution of each ligand-receptor pair in the CCL signaling pathway to the overall communication between MPs and neutrophils (Figure 6E).
In this study, we performed integrated multiomics analysis to elucidate the molecular mechanisms underlying EPO-MSC-mediated diabetic wound healing. Our findings indicated that EPO-MSCs facilitate wound healing by remodeling the paracrine profile, modulating the immune microenvironment, and activating key signaling pathways, thereby forming a multidimensional repair network. These effects can significantly alleviate the “healing paralysis” state in the diabetic wound microenvironment. These findings provide a strong theoretical foundation for the formulation of gene-engineered stem cell-based therapeutic strategies for chronic diabetic wound treatment. These effects can significantly alleviate the “healing paralysis” state in the diabetic wound microenvironment and form a multidimensional repair network (Figure 7).
The results of the scratch assay confirmed that, compared to NC-MSCs, EPO-MSCs significantly enhanced the migration of HFF cells. In vivo results revealed that EPO-MSCs and NC-MSCs promoted the healing of cutaneous wounds in diabetic mice. This finding aligns with the findings of previous studies, indicating the potential of MSCs in diabetic wound healing[12]. EPO-MSCs showed better results in promoting wound healing, especially in enhancing epidermal thickness. Next, we performed transcriptomic and proteomic analyses on cutaneous wound tissues from diabetic mice to investigate the molecular mechanisms underlying the effects of EPO-MSCs on wound healing. The results of the transcriptomic analysis revealed the involvement of DEGs in several key signaling pathways associated with diabetic wound healing. The roles of pathways such as the TNF, mitogen-activated protein kinases, and IL-17 pathways in diabetic wound healing have been well established in previous studies[13-15]. The GO enrichment results suggested that the DEGs were primarily associated with angiogenesis and epithelial cell proliferation, both of which are closely related to the wound healing process in diabetes.
The results of the proteomic analysis showed that the upregulated differentially expressed proteins in terms of cellular components were predominantly enriched in the mitochondrial envelope, inner membrane, and organelle envelope. These findings suggest that EPO-MSCs may increase mitochondrial function and energy metabolism, thereby supporting metabolic demands during wound healing. The KEGG enrichment results suggested that the upregulated differentially expressed proteins play an important role in pathways such as amino sugar and nucleotide sugar metabolism, oxidative phosphorylation, apoptosis, and sphingolipid metabolism. These pathways play crucial roles in the cellular energy supply, tissue repair, and anti-inflammatory responses. In contrast, the downregulated differentially expressed proteins exhibited predominant correlations with the metabolism of heterocyclic compounds, aromatic compounds, and cellular responses to IL-6. These proteins may negatively regulate inflammatory responses and remodeling of the extracellular matrix.
Single-cell sequencing revealed dynamic changes in macrophage subpopulations, providing crucial evidence for understanding the immune regulatory mechanisms of EPO-MSCs. The significant upregulation of Arg1 suggested that EPO-MSCs may increase the proportion of M2 macrophages during diabetic wound repair and perform unique functions through the Saa3+ macrophage subpopulation. Macrophages coordinate the wound healing process by transitioning from predominantly pro-inflammatory (M1-like) phenotypes, which appear early after injury, to anti-inflammatory (M2-like) phenotypes, which emerge later to facilitate skin regeneration and wound closure[16]. Researchers have extensively studied the acute-phase lipoprotein SAA as a reliable marker for various inflammatory and autoimmune conditions[17]. In murine models, SAA3 is the isoform that is commonly expressed in hematopoietic and epithelial cells[18]. SAA3-deficient mice exhibit metabolic disorders and obesity, along with altered secretion of cytokines in bone marrow-derived dendritic cells[19]. These findings indicate that SAA3 not only serves as a biomarker for infection and inflammation but is also crucial for developing immunity and maintaining metabolic homeostasis. In this study, Saa3 played an important regulatory role in facilitating the polarization of macrophages to the M2 phenotype, further enhancing wound repair. The results of the intercellular communication analysis revealed that the interaction between macrophages and neutrophils was the most intense, with the CCL signaling pathway being the primary mediator of this interaction. Lingering hy
From a translational standpoint, EPO-MSCs should be viewed within the broader framework of gene-modified cell therapies, which face distinct regulatory and manufacturing hurdles[22]. Because EPO-MSCs are generated using integrating viral vectors, regulatory agencies are likely to classify them as advanced therapy medicinal products and require rigorous characterization of vector copy number, insertional mutagenesis risk, and long-term transgene ex
Another key challenge for clinical translation is the limited retention and survival of transplanted MSCs in the harsh microenvironment of chronic diabetic wounds. In our model, local injection of EPO-MSCs was sufficient to demonstrate proof-of-concept efficacy; however, in human lesions with larger size and more complex architecture, cell loss due to mechanical washout, hypoxia, and proteolytic stress may substantially blunt therapeutic benefit[26]. The use of bio
To clarify the mechanistic scope of the present work, we note that although EPO-MSCs demonstrated superior the
This study also had several other limitations. First, while it demonstrated the therapeutic potential of EPO-MSCs both in vitro and in vivo models, the clinical applicability of these findings remains uncertain. Translating these results into human clinical trials is challenging owing to the complexity of human diabetic wounds, which may have molecular and immune characteristics different from those observed in animal models. Second, although single-cell RNA sequencing and proteomic analysis offer a high-resolution view of the wound microenvironment, they primarily provide descriptive and correlative insights. Third, we did not assess the long-term efficacy and safety of EPO-MSCs in the context of wound healing, including potential side effects or immune responses in diabetic individuals. Finally, the role of other cell types and extracellular matrix components in the EPO-MSC-mediated healing process was not fully addressed, preventing a comprehensive understanding of the overall repair mechanism.
To summarize, this study provides compelling evidence that EPO-MSCs have therapeutic potential to improve the efficacy of wound healing in diabetic mice, primarily via immune modulation and the facilitation of macrophage polarization. Using a multiomics approach, we identified several key signaling pathways and cellular processes involved in wound repair, including angiogenesis, immune cell recruitment, and macrophage polarization. These findings improved our knowledge of the complex molecular landscape of diabetic wound healing and highlighted the potential of EPO-MSCs as a promising cell-based therapeutic agent for treating diabetic wounds. However, further studies are needed to fully investigate the clinical relevance, underlying mechanisms of action, and long-term safety of EPO-MSCs in human diabetic wound healing, as well as to determine the contributions of other cell types and extracellular matrix factors to the repair process.
| 1. | Armstrong DG, Tan TW, Boulton AJM, Bus SA. Diabetic Foot Ulcers: A Review. JAMA. 2023;330:62-75. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 1177] [Cited by in RCA: 998] [Article Influence: 332.7] [Reference Citation Analysis (2)] |
| 2. | Falanga V, Isseroff RR, Soulika AM, Romanelli M, Margolis D, Kapp S, Granick M, Harding K. Chronic wounds. Nat Rev Dis Primers. 2022;8:50. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 14] [Cited by in RCA: 593] [Article Influence: 148.3] [Reference Citation Analysis (0)] |
| 3. | McDermott K, Fang M, Boulton AJM, Selvin E, Hicks CW. Etiology, Epidemiology, and Disparities in the Burden of Diabetic Foot Ulcers. Diabetes Care. 2023;46:209-221. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 901] [Cited by in RCA: 701] [Article Influence: 233.7] [Reference Citation Analysis (7)] |
| 4. | Otero-Viñas M, Falanga V. Mesenchymal Stem Cells in Chronic Wounds: The Spectrum from Basic to Advanced Therapy. Adv Wound Care (New Rochelle). 2016;5:149-163. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 94] [Cited by in RCA: 111] [Article Influence: 11.1] [Reference Citation Analysis (1)] |
| 5. | Farabi B, Roster K, Hirani R, Tepper K, Atak MF, Safai B. The Efficacy of Stem Cells in Wound Healing: A Systematic Review. Int J Mol Sci. 2024;25:3006. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 16] [Cited by in RCA: 52] [Article Influence: 26.0] [Reference Citation Analysis (0)] |
| 6. | Guillamat-Prats R. The Role of MSC in Wound Healing, Scarring and Regeneration. Cells. 2021;10:1729. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 339] [Cited by in RCA: 281] [Article Influence: 56.2] [Reference Citation Analysis (0)] |
| 7. | Wang Y, Jing L, Lei X, Ma Z, Li B, Shi Y, Zhang W, Li Y, Zhou H, Hu K, Xue Y, Jin Y. Umbilical cord mesenchymal stem cell-derived apoptotic extracellular vesicles ameliorate cutaneous wound healing in type 2 diabetic mice via macrophage pyroptosis inhibition. Stem Cell Res Ther. 2023;14:257. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 37] [Reference Citation Analysis (0)] |
| 8. | Huerta CT, Voza FA, Ortiz YY, Liu ZJ, Velazquez OC. Mesenchymal stem cell-based therapy for non-healing wounds due to chronic limb-threatening ischemia: A review of preclinical and clinical studies. Front Cardiovasc Med. 2023;10:1113982. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 39] [Cited by in RCA: 34] [Article Influence: 11.3] [Reference Citation Analysis (0)] |
| 9. | Peng B, Kong G, Yang C, Ming Y. Erythropoietin and its derivatives: from tissue protection to immune regulation. Cell Death Dis. 2020;11:79. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 147] [Cited by in RCA: 126] [Article Influence: 21.0] [Reference Citation Analysis (0)] |
| 10. | Hamed S, Ullmann Y, Masoud M, Hellou E, Khamaysi Z, Teot L. Topical erythropoietin promotes wound repair in diabetic rats. J Invest Dermatol. 2010;130:287-294. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 67] [Cited by in RCA: 78] [Article Influence: 4.9] [Reference Citation Analysis (0)] |
| 11. | Lu H, Wu X, Wang Z, Li L, Chen W, Yang M, Huo D, Zeng W, Zhu C. Erythropoietin-activated mesenchymal stem cells promote healing ulcers by improving microenvironment. J Surg Res. 2016;205:464-473. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 22] [Cited by in RCA: 28] [Article Influence: 2.8] [Reference Citation Analysis (0)] |
| 12. | Huang YZ, Gou M, Da LC, Zhang WQ, Xie HQ. Mesenchymal Stem Cells for Chronic Wound Healing: Current Status of Preclinical and Clinical Studies. Tissue Eng Part B Rev. 2020;26:555-570. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 185] [Cited by in RCA: 145] [Article Influence: 24.2] [Reference Citation Analysis (3)] |
| 13. | Elhabal SF, Abdelaal N, Saeed Al-Zuhairy SAK, Elrefai MFM, Elsaid Hamdan AM, Khalifa MM, Hababeh S, Khasawneh MA, Khamis GM, Nelson J, Mohie PM, Gad RA, Rizk A, Kabil SL, El-Ashery MK, Jasti BR, Elzohairy NA, Elnawawy T, Hassan FE, El-Nabarawi MA. Green Synthesis of Zinc Oxide Nanoparticles from Althaea officinalis Flower Extract Coated with Chitosan for Potential Healing Effects on Diabetic Wounds by Inhibiting TNF-α and IL-6/IL-1β Signaling Pathways. Int J Nanomedicine. 2024;19:3045-3070. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 33] [Reference Citation Analysis (0)] |
| 14. | Luo B, Ding X, Hu Y, Tian M, Wu J, Shi H, Lu X, Xia X, Guan W, Jiang W. Shikonin hastens diabetic wound healing by inhibiting M1 macrophage polarisation through the MAPK signaling pathway. Mol Immunol. 2025;177:73-84. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 1] [Cited by in RCA: 7] [Article Influence: 7.0] [Reference Citation Analysis (0)] |
| 15. | Mu X, Gu R, Tang M, Wu X, He W, Nie X. IL-17 in wound repair: bridging acute and chronic responses. Cell Commun Signal. 2024;22:288. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 23] [Cited by in RCA: 47] [Article Influence: 23.5] [Reference Citation Analysis (0)] |
| 16. | Hassanshahi A, Moradzad M, Ghalamkari S, Fadaei M, Cowin AJ, Hassanshahi M. Macrophage-Mediated Inflammation in Skin Wound Healing. Cells. 2022;11:2953. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 164] [Cited by in RCA: 231] [Article Influence: 57.8] [Reference Citation Analysis (1)] |
| 17. | Uhlar CM, Whitehead AS. Serum amyloid A, the major vertebrate acute-phase reactant. Eur J Biochem. 1999;265:501-523. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 783] [Cited by in RCA: 820] [Article Influence: 30.4] [Reference Citation Analysis (0)] |
| 18. | Ramadori G, Sipe JD, Colten HR. Expression and regulation of the murine serum amyloid A (SAA) gene in extrahepatic sites. J Immunol. 1985;135:3645-3647. [PubMed] |
| 19. | Ather JL, Poynter ME. Serum amyloid A3 is required for normal weight and immunometabolic function in mice. PLoS One. 2018;13:e0192352. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 26] [Cited by in RCA: 32] [Article Influence: 4.0] [Reference Citation Analysis (0)] |
| 20. | R AS, Nambi N, Radhakrishnan L, Prasad MK, Ramkumar KM. Neutrophil Migration is a Crucial Factor in Wound Healing and the Pathogenesis of Diabetic Foot Ulcers: Insights into Pharmacological Interventions. Curr Vasc Pharmacol. 2024. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 12] [Cited by in RCA: 14] [Article Influence: 7.0] [Reference Citation Analysis (0)] |
| 21. | Lin S, Wang Q, Huang X, Feng J, Wang Y, Shao T, Deng X, Cao Y, Chen X, Zhou M, Zhao C. Wounds under diabetic milieu: The role of immune cellar components and signaling pathways. Biomed Pharmacother. 2023;157:114052. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 40] [Reference Citation Analysis (0)] |
| 22. | Wang X, Wang H, Lu J, Feng Z, Liu Z, Song H, Wang H, Zhou Y, Xu J. Erythropoietin-Modified Mesenchymal Stem Cells Enhance Anti-fibrosis Efficacy in Mouse Liver Fibrosis Model. Tissue Eng Regen Med. 2020;17:683-693. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 8] [Cited by in RCA: 18] [Article Influence: 3.0] [Reference Citation Analysis (0)] |
| 23. | Iansante V, Brooks A, Coney L. Considerations in the Design of Non-Clinical Development Programmes to Support Non-Viral Genetically Modified Mesenchymal Stromal Cell Therapies. Pharmaceutics. 2021;13:823. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 3] [Reference Citation Analysis (0)] |
| 24. | Shirley Ding SL, Kumar S, Ali Khan MS, Ling Mok P. Human Mesenchymal Stem Cells Expressing Erythropoietin Enhance Survivability of Retinal Neurons Against Oxidative Stress: An In Vitro Study. Front Cell Neurosci. 2018;12:190. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 11] [Cited by in RCA: 14] [Article Influence: 1.8] [Reference Citation Analysis (0)] |
| 25. | Suttorp MM, Hoekstra T, Ocak G, van Diepen AT, Ott I, Mittelman M, Rabelink TJ, Krediet RT, Dekker FW. Erythropoiesis-stimulating agents and thrombotic events in dialysis patients. Thromb Res. 2014;134:1081-1086. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 6] [Cited by in RCA: 7] [Article Influence: 0.6] [Reference Citation Analysis (0)] |
| 26. | Huang JN, Cao H, Liang KY, Cui LP, Li Y. Combination therapy of hydrogel and stem cells for diabetic wound healing. World J Diabetes. 2022;13:949-961. [PubMed] [DOI] [Full Text] |
| 27. | Song Y, You Y, Xu X, Lu J, Huang X, Zhang J, Zhu L, Hu J, Wu X, Xu X, Tan W, Du Y. Adipose-Derived Mesenchymal Stem Cell-Derived Exosomes Biopotentiated Extracellular Matrix Hydrogels Accelerate Diabetic Wound Healing and Skin Regeneration. Adv Sci (Weinh). 2023;10:e2304023. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 1] [Cited by in RCA: 174] [Article Influence: 58.0] [Reference Citation Analysis (0)] |
| 28. | Huang J, Deng Q, Tsang LL, Chang G, Guo J, Ruan YC, Wang CC, Li G, Chan HF, Zhang X, Jiang X. Mesenchymal stem cells from perinatal tissues promote diabetic wound healing via PI3K/AKT activation. Stem Cell Res Ther. 2025;16:59. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 1] [Cited by in RCA: 17] [Article Influence: 17.0] [Reference Citation Analysis (0)] |