Human umbilical cord mesenchymal stem cells repair islet structure and function in diabetic rats through metabolic reprogramming

Figure 1 Effects of human umbilical cord mesenchymal stem cells on body weight, blood glucose levels, and the performance of pancreatic islets in rat diabetic models. A: Diagram illustrating the therapeutic procedure using human umbilical cord mesenchymal stem cells in diabetic rats; B: Changes in rat body weight over time across all groups; C: Temporal fluctuations in the fasting blood glucose levels among the different groups; D: Fasting insulin concentrations in the serum of rats per group; E: Measurements of serum C-peptide levels in rats from each group; F: Glycated hemoglobin levels in the serum of rats across groups; G: Glycated serum protein levels in rats evaluated by group; H: Homeostasis model assessment of insulin resistance indices assessed for each group of rats; I: Representative images of pancreatic tissues stained with hematoxylin and eosin from rats in all groups. Scale bars = 400 μm (left) or 200 μm (right); J: Representative TUNEL staining images of rat pancreatic tissues from each group. Scale bar = 100 μm; K: Bar graph shows quantitative analysis of apoptosis rate; L: Immunohistochemical staining of insulin in the rat pancreatic tissues. Scale indicators = 400 μm (left) or 200 μm (right); the bar chart depicts the quantified percentage of insulin-positive regions. Control, n = 8; diabetes mellitus (DM), n = 6; DM + human umbilical cord mesenchymal stem cells, n = 5. ^aP < 0.001. STZ: Streptozotocin; SD: Sprague-Dawley; hUC-MSCs: Human umbilical cord mesenchymal stem cells; DM: Diabetes mellitus; FBG: Fasting blood glucose; FINS: Fasting insulin; GHb: Glycated hemoglobin; GSP: Glycated serum protein; HOMA-IR: Homeostasis model assessment of insulin resistance; TUNEL: Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling; DAPI: 4’,6-Diamidino-2-phenylindole dihydrochloride.

Figure 2 Pancreatic metabolomics data analysis of differential metabolites in rats from different groups. A: Schematic diagram of pancreatic metabolomic analysis in rats; B: Two-dimensional principal component analysis score scatterplot showing rats across all groups; C: Orthogonal partial least squares-discriminant analysis (OPLS-DA) score scatterplot comparing the control and diabetes mellitus (DM) groups; D: OPLS-DA score scatterplot comparing the DM group with the DM + human umbilical cord mesenchymal stem cells (hUC-MSCs) group; E: Validation plots for OPLS-DA models comparing the control and DM groups; F: Validation plots of OPLS-DA models between the DM group and the DM + hUC-MSCs group; G and H: OPLS-DA S-plots comparing the control and DM groups (G), S-plots from OPLS-DA for the contrast between the DM group and the DM + hUC-MSCs group (H), the orange dots denote metabolites exhibiting variable importance in projection (VIP) > 1, while the green dots correspond to metabolites with VIP ≤ 1; I: Volcano plot depicting differential metabolites in the DM group compared with the control group; J: Clustered heatmap of metabolites with significant differences between the control group and the DM group; K: Volcano plot of differential metabolites between the DM + hUC-MSCs and DM groups; L: Heatmap with hierarchical clustering illustrating differential metabolites between the DM + hUC-MSCs and DM groups. Control, n = 8; DM, n = 6; DM + hUC-MSCs, n = 5. hUC-MSCs: Human umbilical cord mesenchymal stem cells; DM: Diabetes mellitus; OPLS-DA: Orthogonal partial least squares-discriminant analysis; VIP: Variable importance projection.

Figure 3 Correlation analysis of differential pancreatic metabolites between rats from different groups. A and B: Correlation heatmaps of differential metabolites for the indicated comparisons [A: Diabetes mellitus (DM) vs control; B: DM + human umbilical cord mesenchymal stem cells (hUC-MSCs) vs DM], with red and green representing positive and negative correlations, respectively, and color intensity scaled to the absolute value of the sample correlation coefficient; C and D: Correlation chord diagrams for differential metabolites in the indicated comparisons (C: DM vs control; D: DM + hUC-MSCs vs DM), with dot size proportional to log₂FC, dot color indicating metabolite class, and red and blue lines representing positive and negative correlations, respectively; E and F: Correlation network diagrams of differential metabolites for the indicated comparisons (E: DM vs control; F: DM + hUC-MSCs vs DM). Nodes represent metabolites, with node size reflecting the degree value. Red and blue lines denote positive and negative correlations, respectively, and line thickness is proportional to correlation strength. Only the top 50 metabolites (ranked by variable importance projection) are displayed in all correlation heatmaps, chord diagrams, and network diagrams. Control, n = 8; DM, n = 6; DM + hUC-MSCs, n = 5. DM: Diabetes mellitus; hUC-MSCs: Human umbilical cord mesenchymal stem cells; VIP: Variable importance projection.

Figure 4 Alterations in pancreatic metabolites and metabolic pathways induced by human umbilical cord mesenchymal stem cells in a rat model of diabetes. A: Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment bubble plot depicting differential metabolites between diabetes mellitus (DM) and control rats; B: KEGG enrichment analysis of metabolites differentially expressed after human umbilical cord mesenchymal stem cell intervention relative to the diabetic control group, presented as a bubble plot; C: Venn diagram illustrating key differential metabolites (KDMs) across the indicated group comparisons; D: Heatmap depicting 30 KDMs; E: KEGG enrichment bubble plot for the 30 KDMs. Control, n = 8; DM, n = 6; DM + human umbilical cord mesenchymal stem cells, n = 5. KEGG: Kyoto Encyclopedia of Genes and Genomes; DM: Diabetes mellitus; hUC-MSCs: Human umbilical cord mesenchymal stem cells; PPAR: Peroxisome proliferator-activated receptor.

Figure 5 Metabolic pathway diagrams significantly enriched with key differential metabolites. A: Pathway diagrams of the peroxisome proliferator-activated receptor signaling pathway; B: Choline metabolism in cancer; C: Efferocytosis; D: Linoleic acid metabolism; E: Biosynthetic pathway of phenylalanine, tyrosine, and tryptophan; F: Neuroactive ligand-receptor signaling pathway. PPAR: Peroxisome proliferator-activated receptor.

Figure 6 Receiver operating characteristic curves of 30 key differential metabolites comparing the diabetes mellitus and control groups. Control, n = 8; diabetes mellitus, n = 6. ROC: Receiver operating characteristic; AUC: Area under the curve.

Figure 7 Receiver operating characteristic curves for 30 key differential metabolites comparing the diabetes mellitus + human umbilical cord mesenchymal stem cells group with the diabetes mellitus group. Diabetes mellitus (DM), n = 6; DM + human umbilical cord mesenchymal stem cells, n = 5. ROC: Receiver operating characteristic; AUC: Area under the curve.

Figure 8 Importance dot plots and receiver operating characteristic curves of random forest models for key differential metabolites in each group. A and B: Dot plots of random forest importance ranking for key differential metabolites (KDMs) between different groups with the X-axis representing importance indices; C and D: Comprehensive receiver operating characteristic curves of KDMs between different groups generated based on the random forest model. Control, n = 8; diabetes mellitus (DM), n = 6; DM + human umbilical cord mesenchymal stem cells, n = 5. ROC: Receiver operating characteristic; AUC: Area under the curve.

Figure 9 Correlation analysis between key differential metabolites and biochemical indicators in rats from each group. A: Spearman clustering heatmap illustrates key differential metabolites (KDMs) - biochemical correlations in diabetes mellitus (DM) vs controls, with orange-red indicating positive associations, blue indicating negative associations, and darker shades denoting stronger correlations. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001 for the correlations; B: Spearman correlation network diagram of KDMs and biochemical indicators in the DM vs control groups, where red lines denote positive, green lines denote negative correlations, and line thickness reflects correlation magnitude (P < 0.05); C: Spearman correlation cluster heatmap of KDMs and biochemical indicators in the DM + human umbilical cord mesenchymal stem cells (hUC-MSCs) group vs the DM group; D: Spearman correlation network diagram of KDMs and biochemical indicators in the DM + hUC-MSCs group vs DM group. Control, n = 8; DM, n = 6; DM + hUC-MSCs, n = 5. FINS: Fasting insulin; FBG: Fasting blood glucose; HOMA-IR: Homeostasis model assessment of insulin resistance; GHb: Glycosylated hemoglobin; GSP: Glycosylated serum protein.