Published online Oct 26, 2026. doi: 10.4252/wjsc.122760
Revised: July 27, 2026
Accepted: October 8, 2026
Published online: October 26, 2026
Processing time: 182 Days and 4 Hours
Diabetes mellitus (DM) is a widespread metabolic disorder primarily characterized by persistent hyperglycemia and impaired pancreatic islet function. Human umbilical cord mesenchymal stem cells (hUC-MSCs) represent a promi
To investigate the therapeutic effect of hUC-MSCs in DM model rats and the underlying mechanism.
Streptozotocin was used to induce DM in Sprague-Dawley rats, and hUC-MSCs were administered interventional therapy. We then combined classic physiological and biochemical indices [such as body weight, fasting blood glucose (FBG), glycated hemoglobin (GHb), glycated serum protein (GSP), fasting insulin (FINS), and C-peptide, along with the homeostasis model assessment of insulin resistance (HOMA-IR)], histopathological evaluations [including hematoxylin and eosin (HE), TUNEL staining, and immunohistochemistry], and untargeted metabolomics to screen for key differential metabolites (KDMs) in the pancreas.
Treatment with hUC-MSCs significantly reduced the levels of FBG, GHb, GSP, and HOMA-IR and increased the levels of FINS and C-peptide in DM rats. HE staining showed that hUC-MSCs effectively improved islet structure by restoring the area and number of islets and reducing the cellular apoptosis rate in the pancreatic tissue. Immunohistochemical staining showed that hUC-MSCs significantly increased the number of insulin-positive β cells. Metabolomic analysis showed that hUC-MSCs improved pancreatic metabolism of DM rats by modulating various metabolic processes, such as those involved in lipid and amino acid metabolism. Receiver operating characteristic curve analysis showed that all 30 KDMs had excellent diagnostic performance, suggesting that KDMs are potential DM diagnostic biomarkers and predictors of the treatment efficacy of the hUC-MSCs. Spearman analysis showed significant correlation of KDMs with FBG, insulin, and HOMA-IR.
hUC-MSC therapy lowers blood glucose levels and enhances both the structure and functionality of islets in diabetic rats, while also supporting improvements in impaired pancreatic metabolism. Treatment with hUC-MSCs improves pancreatic metabolic disorders in the DM rats by modulating 30 KDMs.
Core Tip: Diabetes mellitus (DM) is a prevalent metabolic disease. This study investigated the treatment efficacy of the human umbilical cord mesenchymal stem cells (hUC-MSCs) in repairing diabetic rat islets and the potential underlying mechanisms. Results showed that hUC-MSCs lowered blood glucose, improved insulin secretion and islet structure, and regulated pancreatic lipid and amino acid metabolism. Thirty key differential metabolites were identified as potential diagnostic and treatment efficacy prediction biomarkers. The results indicate that hUC-MSCs protects islets and improves glucose metabolism in DM by regulating pancreatic metabolism.
- Citation: Xu J, Yang X, Zhang HY, Wang LL, Luo YD, Yang SL, Zhou YJ, Liu Y, Yang XF. Human umbilical cord mesenchymal stem cells repair islet structure and function in diabetic rats through metabolic reprogramming. World J Stem Cells 2026; 18(10): 122760
- URL: https://www.wjgnet.com/1948-0210/full/v18/i10/122760.htm
- DOI: https://dx.doi.org/10.4252/wjsc.122760
Diabetes mellitus (DM) is a prevalent chronic metabolic disorder characterized by sustained hyperglycemia, which arises from insufficient insulin production, impaired insulin sensitivity, or a combination of these factors[1]. Type 1 diabetes arises from an autoimmune response that destroys pancreatic β-cells, resulting in complete insulin deficiency[2,3]. Type 2 diabetes is primarily defined by reduced insulin sensitivity accompanied by impaired β-cell function[4,5]. Moreover, special types of diabetes, including maturity-onset diabetes of the young, are caused by genetic mutations that compromise crucial β-cell function[5]. Although existing drugs can control blood glucose levels, they cannot reverse the structural damage and functional decline of the islets. Therefore, there is an urgent need to investigate new therapies that can protect or repair islet structure and function. In recent years, stem cell-based regenerative medicine has provided new avenues for DM treatment. Human umbilical cord mesenchymal stem cells (hUC-MSCs) have shown clinical potential in improving glucose metabolism and islet structure in DM animal models because of their abundant source, low immunogenicity, multidirectional differentiation potential, and strong paracrine and immunomodulatory functions[6]. However, the specific molecular mechanisms by which hUC-MSCs exert their therapeutic effects, especially their systemwide regulatory effects on the metabolic network of the damaged pancreatic microenvironment, have not been fully elucidated. Therefore, this study investigated the systemwide regulatory effect of treatment with hUC-MSCs on the pancreatic metabolic microenvironment in DM. Previous studies have confirmed that hUC-MSCs alleviate islet inflammation, suppress β-cell apoptosis, and foster β-cell regeneration via multiple pathways, including cell homing[7], differentiation[8], cytokine secretion[9], and exosomes release[10]. These effects are closely related to the improvement of systemic glucose and lipid metabolism. However, as the core organ of metabolic regulation, the pancreas exhibits the dynamic metabolite changes directly reflect cellular function and signaling pathways activity. Accordingly, correlating the therapeutic effect of hUC-MSCs with pancreatic metabolomic profiles enables more precisely elucidation of its mechanism of action at the metabolite level. This correlational study also aims to link the observed phenotypic improvements (such as reduced hyperglycemia and restored islet structure) with potential metabolic reprogramming, thereby more comprehensively evaluating the therapeutic potential of hUC-MSCs and discovering novel biomarkers.
This study employed a DM rat model induced by streptozotocin (STZ), and hUC-MSCs were administered via tail-vein injection for intervention. The research methods combined measurements of classic physiological and biochemical indices [such as blood glucose and serum glycated hemoglobin (GHb), glycated serum protein (GSP), insulin, and C-peptide levels), homeostasis model assessment of insulin resistance (HOMA-IR)], histopathological analysis [hematoxylin and eosin (HE) staining, TUNEL staining, and immunohistochemistry], and cutting-edge systems biology technology-untargeted metabolomics. The strength of this multi-dimensional research strategy was that it could verify not only the therapeutic effect of hUC-MSCs in DM rats at the whole-animal level but also hUC-MSC-mediated the repair of the islet structure and function at the morphological level, as well as permit unbiased assessment of the global alterations in endogenous small-molecule metabolite levels within pancreatic tissue. Overall, we aimed to systematically elucidate the potential mechanism by which hUC-MSCs improved glucose metabolism in DM rats.
Primary hUC-MSCs were purchased from Chuyuan Cloud Medical Technology Co., Ltd. (Located in Foshan, Guangdong Province, China). These cells were derived from the Wharton’s jelly of human umbilical cords. The cells were positive positivity for CD44, CD73, and CD105 markers, while negative for HLA-DR, CD45, and CD34. Additionally, no microbial contamination was detected. hUC-MSCs were maintained in complete α-MEM medium (PM150421, Procell, Wuhan, Hubei Province, China) enriched with 10% fetal bovine serum (IC-1908, InCellGene, Germany) and 1% penicillin-streptomycin (PB180120, Procell, Wuhan, Hubei Province, China). The cells were incubated at 37 °C in a humidified atmosphere with 5% CO2. Upon reaching 80%-90% confluence, they were passaged at a 1:2 ratio, and the culture medium was replaced every three days. Cells at passages 3-5 were subjected to tail-vein injection into rats.
The cells were rinsed with phosphate buffered saline (PBS) and enzymatically dissociated with trypsin for approximately 2 minutes. After trypsin treatment, the cells were centrifuged at 3000 rpm for 5 minutes. The supernatant was discarded, and the cell pellet was resuspended in PBS to achieve a final density of 1 × 106 cells/mL. To preserve maximal cell viability, the hUC-MSC suspension was freshly prepared and used immediately.
Preparation of STZ solution: A citric acid buffer solution was prepared by dissolving 2.1 g of citric acid (60347ES25, YEASEN, Shanghai, China) with 2.94 g of trisodium citrate (60348ES25, YEASEN, Shanghai, China) in 100 mL of ultrapure water. STZ (S8050, Solarbio, Beijing, China) was dissolved in the citric acid buffer (pH = 4.5) to obtain a 2% (w/v) working solution.
Establishment of the rat diabetes model: Specific pathogen-free male Sprague-Dawley (SD) rats aged 6-8 weeks and weighing 200 ± 20 g were purchased from Beijing Weitong Lihua Experimental Animal Technology Co., Ltd. (Beijing, China). The rats animals were housed under standard laboratory conditions at 23 ± 2 °C with a relative humidity of 55%-60% and a 12-hour light-dark cycle, with free access to food and water. All experimental procedures were performed in accordance with the ARRIVE guidelines and approved by the Institutional Animal Care and Use Committee of the Hunan Evidence-Based Biotechnology Co. Ltd. (Research Collaborating Institution of the Affiliated Nanhua Hospital, Hengyang Medical School, University of South China) (Approval No. 2024-XZSW-240301).
After 1-week of acclimatization, 24 SD rats were randomly assigned to the control, DM, and DM + hUC-MSCs groups (n = 8 per group) using a random number table. The sample size was determined based on common practice in comparable intervention studies[11,12]. As previously described[13], diabetes was induced in the DM and DM + hUC-MSCs groups via a single intraperitoneal injection of STZ (60 mg/kg), whereas rats in the control group received an equal volume of citric acid buffer. Food was restored immediately after STZ administration, and blood glucose levels were measured 72-hour post-injection. Tail-vein blood glucose was monitored for three consecutive days, and diabetes modeling was deemed successful when all three readings reached ≥ 16.7 mmol/L[1]. The model establishment success rate was 100% in the present study.
Following confirmation of diabetes, rats in the DM and DM + hUC-MSCs groups were fed a high-fat, high-sugar diet[11,14] (HFSD, 12451M, Beijing Boaigang Biological Technology Co., Ltd., Beijing, China), whereas rats in the control group continued to receive a standard laboratory diet for the duration of the experiment.
By week 10 of the experiment, two rats in the DM group had died of ketosis and infection, leaving six surviving animals. In the DM + hUC-MSCs group, three rats died during the same period, leaving five survivors, whereas all eight rats in the control group survived. Based on a previous study[15], rats in the DM + hUC-MSCs group received tail-vein injection of 1 mL of freshly prepared hUC-MSC suspension (1 × 106 cells/mL per rat). Rats in the control and DM groups received an equal volume of PBS via the tail-vein. Four weeks post-treatment, all rats were euthanized, and tissues were collected for subsequent analyses. To minimize mitigation of observer bias, blinded procedures were implemented to ensure that researchers performing histological assessments and quantitative image analysis were unaware of the group allocations. Throughout the experiment, body weight and fasting blood glucose (FBG) were measured every Wednesday following a 12-hour overnight fast.
At the end of the experiment (week 14), all rats were euthanized with an overdose of sodium pentobarbital. Serum samples were prepared from collected cardiac blood and stored at -80 °C. In addition, pancreatic tissues were harvested; one aliquot was rapidly transferred to a microcentrifuge tube and stored frozen at -80 °C for subsequent metabolomic analysis, while the remaining portion was fixed in 4% paraformaldehyde solution (P0099, Beyotime, Shanghai, China) and for histological studies.
Commercial ELISA kits for rat insulin (E-EL-R2466, Elabscience, Wuhan, Hubei Province, China), C-peptide (MBS704133, MyBioSource, CA, United States), GHb (JL21291-96T, JONLNBIO, Shanghai, China) and GSP (BY-ER330799, BYabscience, Nanjing, Jiangsu Province, China) were used to measure serum fasting insulin (FINS), C-peptide, GHb, and GSP levels, respectively. Subsequently, HOMA-IR was calculated to assess IR according to the following formula: HOMA-IR = [FBG (mmol/L) × FINS (mU/L)]/22.5.
Pancreatic tissues were fixed with 4% paraformaldehyde for 24 hours, embedded in paraffin and cut into 4-μm-thick sections. After dewaxing and rehydration of the sections, HE staining was performed according to the manufacturer’s instructions (C0105S, Beyotime, Shanghai, China). After staining, morphological analysis of the pancreatic tissue was performed using an optical microscope (Olympus, Tokyo, Japan) and the number of islets was counted.
A one-step TUNEL cell apoptosis detection kit (CX107, Epizyme Biotech, Shanghai, China) was used to evaluate cellular apoptosis in the pancreatic tissues. Briefly, paraffin sections were dewaxed and washed with PBS thrice, 5 minutes each. Then, sections were covered with a working solution of proteinase K and incubated at 37 °C for 10 minutes. The sections were then washed thrice with PBS, 5 minutes each. The sections were incubated with the TUNEL labeling working solution in a humidified chamber at 37 °C in the dark for 60 minutes. Then, they were washed thrice with PBS, 5 minutes each. The sections were counterstained with DAPI staining solution (C1006, Beyotime, Shanghai, China) at room temperature in the dark for 5 minutes. The sections were quickly washed twice with PBS, 3 minutes each. The sections were mounted with anti-fade glycerol mounting medium. The TUNEL and DAPI single-channel images were acquired under a Leica SP8 laser confocal microscope in 3-5 randomly selected fields per sample and merged with the accompanying software. Quantitative analysis was performed using ImageJ software to calculate the percentage of apoptotic (TUNEL-positive) cells (averaged across multiple fields).
After dewaxing and rehydration, paraffin sections were placed in a 98 °C water bath and heated for 15 minutes using sodium citrate antigen retrieval solution (C1032, Solarbio, Beijing, China). After cooling, they were incubated with 3% hydrogen peroxide aqueous solution (H6520, Sigma-Aldrich, MO, United States) at room temperature for 10 minutes. Subsequently, the sections were permeabilized with 0.3% Triton X-100 (P0096, Beyotime, Shanghai, China) for 10 minutes and blocked with 5% BSA (SW3015, Solarbio, Beijing, China) for 1 hour. The sections were incubated with primary anti-insulin antibody (1:100, #4590, CST, MA, United States) at 4 °C overnight. After washing, they were incubated with HRP-conjugated goat anti-rabbit IgG secondary antibody (1:1000, ab6721, Abcam, United Kingdom) at room temperature for 1 hour. After DAB-mediated color development (P0203, Beyotime, Shanghai, China), the sections were observed under an optical microscope (Olympus, Tokyo, Japan) and imaged. Finally, the percentage of positive cells was analyzed using ImageJ software.
After thawing the rat pancreatic tissue samples on ice, mechanical disruption was performed in a grinder (at 30 Hz, 20 seconds). Subsequently, we precisely weighed 20 mg of tissue homogenate, added 400 μL of precooled extract (methanol:water = 7:3, containing an isotope internal standard), and vortexed for 5 minutes (at 1500 rpm). After the mixture was allowed to stand on ice for 15 minutes, it was centrifuged at 12000 rpm for 10 minutes at 4 °C, and 300 μL of the supernatant was stored at -20 °C for 30 minutes. Subsequently, 200 μL of supernatant was collected following high-speed centrifugation (12000 rpm, 3 minutes) at 4 °C and subjected to LC-MS analysis using a Sciex TripleTOF 6600 system, with both positive- and negative-ion modes operated simultaneously. Gradient elution was performed using a Waters ACQUITY Premier HSS T3 column (2.1 mm × 100 mm, 1.8 μm). The mobile phase consisted of 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (solvent B). Mass spectrometric detection was performed using the information-dependent acquisition mode, and raw data were acquired using Analyst TF 1.7.1 software. Gradient elution procedures and data acquisition parameters were based on previously published methods[16]. Quality control (QC) samples were prepared by mixing equal volumes of extracts from all study samples, and used to monitor instrument performance and data quality. During the instrumental analysis, one QC sample was injected after every 10 experimental samples.
First, the raw mass spectrometry data were converted to mzXML format using ProteoWizard, and peak detection, alignment, and retention time correction were performed with the XCMS program. Peaks with a missing rate exceeding 50% across all groups were filtered out, and missing values were imputed using the k-nearest-neighbor algorithm. Subsequently, the support-vector regression method was applied to correct peak areas to reduce systematic errors. The corrected peak data were searched against an in-house database, public databases, and predicted databases, with the MetDNA method employed for metabolite identification. Metabolites with a comprehensive identification score > 0.5 and a coefficient of variation (CV) < 30% in QC samples were ultimately selected. For metabolites identified in both positive and negative ion modes, the result with the highest identification level and the lowest CV was retained, and these selected entries were merged to generate the final metabolite data matrix for subsequent analysis. Afterwards, the data were scaled using unit variance scaling, and principal component analysis (PCA) was performed using the “prcomp” function in R software to evaluate the trend of inter-group separation. Next, to enhance the discriminative efficiency of the model, the data were subjected to log2 transformation and mean centering, and orthogonal partial least squares-discriminant analysis (OPLS-DA) was performed using the MetaboAnalyst R package. Subsequently, 200 permutation tests were conducted to evaluate the robustness of the model and mitigate overfitting. Variable importance in projection (VIP) > 1 and adjusted P-value < 0.05 were used as the threshold parameters for screening differential metabolites. R software packages such as ComplexHeatmap, circlize, and igraph were employed to generate correlation heatmaps, chord diagrams and correlation networks for the top 50 differential metabolites (ranked by VIP), respectively. Functional annotation of differential metabolites was performed based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) Compound database (http://www.kegg.jp/kegg/compound/), while pathway enrichment analysis was conducted using the KEGG pathway database (http://www.kegg.jp/kegg/pathway.html), with a threshold of P < 0.05 for the statistical significance of enriched pathways. In addition, Venn diagram analysis was carried out using the Xiantao Tool platform (https://www.xiantaozi.com/) to screen for key differential metabolites (KDMs) that were commonly differentially abundant across different groups. Receiver operating characteristic (ROC) curve analysis and random forest modeling of the KDMs were then performed using the Oebiotech cloud platform (https://cloud.oebiotech.cn/task/).
Statistical analysis of experimental data was performed using GraphPad Prism 10.1.2 (San Diego, CA, United States). The measurement data are expressed as mean ± SEM. One-way analysis of variance with Tukey’s HSD post-hoc test was applied for comparisons among multiple groups when the data were normally distributed and exhibited homogeneous variances. When the data deviated from normality or exhibited heterogeneous variances, the Kruskal-Wallis test was employed. The correlation between variables was evaluated using the Spearman correlation coefficient. A P value < 0.05 was considered to indicate statistically significant.
To evaluate the potential therapeutic benefits of hUC-MSCs for diabetes management, a diabetic rat model was established via STZ induction and subsequently treated with hUC-MSCs (Figure 1A). Compared with the control group, rats in the DM group exhibited prominent diabetic phenotypes, including marked reduction in body weight, decreased FINS and C-peptide concentrations, and elevated levels of FBG, GHb, and GSP. hUC-MSC therapy markedly reduced FBG, GHb, and GSP concentrations, while increasing FINS and C-peptide levels. No marked changes in body weight were observed (Figure 1B-G). Consistent improved glycemic control and insulin secretion, HOMA-IR values decreased significantly following hUC-MSC administration (Figure 1H). It should be noted, however, that STZ-induced diabetes mainly represents an insulin-deficient model caused by pancreatic β-cell destruction. Because HOMA-IR was originally developed and validated for insulin-resistant states, its interpretation is limited under conditions of markedly reduced circulating insulin concentrations. Therefore, the observed reduction in HOMA-IR is likely to reflect a combined effect of improved endogenous insulin secretion, as evidenced by increased FINS and C-peptide levels, together with potential improvements in insulin sensitivity, rather than indicating solely the reversal of peripheral insulin resistance.
HE staining demonstrated marked pancreatic islet damage in the DM group, characterized by decreased islet area, fewer islets, and indistinct boundaries between the islets and surrounding exocrine pancreatic tissue. In contrast, hUC-MSC treatment markedly improved islet morphology, increasing both islet area and number, and restored clear islet boundaries (Figure 1I). In addition, the rate of apoptosis in pancreatic tissue was significantly higher in the DM group but was substantially reduced upon hUC-MSC treatment (Figure 1J and K). Immunohistochemical staining further showed that the number of insulin-positive β-cells was markedly decreased in the DM group and was significantly restored after hUC-MSC treatment (Figure 1L). In summary, these findings suggest that hUC-MSC treatment ameliorates the diabetic phenotype primarily through the protection and functional restoration of pancreatic β-cells, resulting in secondary improvements to overall metabolic homeostasis.
We collected pancreatic samples from rats in different groups and performed untargeted metabolomic analysis to further evaluate the effect of hUC-MSC intervention on pancreatic metabolic processes in diabetic rats (Figure 2A). PCA revealed that samples from different groups were clearly separated, whereas samples within the same group were tightly clustered, confirming good stability and reproducibility of the metabolic dataset (Figure 2B). OPLS-DA showed clear distinctions between the DM and control groups, as well as between the DM + hUC-MSCs group and the DM group (Figure 2C and D). Furthermore, permutation tests verified the reliability and validity of the OPLS-DA models (Figure 2E and F). Subsequently, we screened for differential metabolites between different groups (VIP > 1) (Figure 2G and H) and further filtered these metabolites based on statistical significance (P < 0.05). As a result, 669 distinct metabolites distinguished the DM group from the control group (Figure 2I and J), whereas 78 unique metabolites discriminated the DM + hUC-MSCs group from the DM group (Figure 2K and L).
We then selected the top 50 differential metabolites on the basis of VIP ranking and performed Spearman correlation analysis (Figure 3A-F). Spearman correlation analysis revealed significant correlations among the differential metabolites between the DM and control groups. These metabolites were primarily alcohols and amines, bile acids, and carbohydrates and their derivatives; the correlation network of differential metabolites between the DM + hUC-MSCs and DM groups mainly included metabolites such as alcohols and amines, flavonoids, and hormones and hormone-related compounds (Figure 3A-F). These findings suggest that hUC-MSCs modulate the pancreatic metabolic patterns in diabetic rats, highlighting their promising role in ameliorating pancreatic metabolic disorders associated with diabetes.
To further identify the potential pathways involved in the pathogenesis of DM and the therapeutic effects of hUC-MSCs, we performed KEGG enrichment analysis of the altered metabolites. Differential metabolites between the DM and control groups, as well as between the DM + hUC-MSCs and DM groups, were mainly enriched in four metabolic pathways: Retrograde endocannabinoid signaling, α-linolenic acid metabolism, vitamin B6 metabolism, and choline metabolism in cancer (Figure 4A and B).
Next, we performed a comprehensive examination of the distinct regulatory effects of hUC-MSCs on metabolites and identified 30 metabolites with differential abundance across groups, as visualized by a Venn diagram (Figure 4C). Treatment with the hUC-MSCs reversed the pancreatic levels of 15 metabolites that were abnormally increased or decreased in the DM rats (Figure 4D, Table 1). These findings suggest that hUC-MSCs ameliorate pancreatic metabolic disorders in DM rats via modulation of these 15 KDMs.
| Metabolite ID | Metabolite name | DM vs control | DM + hUC-MSCs vs DM | ||
| Log2FC | Type | Log2FC | Type | ||
| MEDP1339 | LPC(18:1/0:0) | -0.327 | Down | -0.240 | Down |
| MW0015937 | Asclepin | -0.748 | Down | -0.621 | Down |
| MW0141637 | 17-phenyl trinor Prostaglandin F2 cyclopropyl amide | -1.688 | Down | -0.950 | Down |
| MW0055257 | N-Cyclopropyl-trans-2-cis-6-nonadienamide | -1.118 | Down | 1.541 | Up |
| MW0006200 | MS-PPOH | -0.870 | Down | -0.697 | Down |
| MEDTN02450 | N-(Cyclopropylmethyl)-7-[3,5-dihydroxy-2-(3-hydroxy-5-phenylpent-1-EN-1-YL)cyclopentyl]hept-5-enamide | -0.916 | Down | -2.084 | Down |
| MW0152407 | Leu-Leu-Trp | -1.244 | Down | 1.942 | Up |
| MW0144473 | Ala-Arg-His | -0.681 | Down | 1.264 | Up |
| MW0123754 | Doramapimod | -2.032 | Down | 2.050 | Up |
| MW0054549 | LysoPC(22:2(13Z,16Z)) | -0.553 | Down | 0.715 | Up |
| MW0053755 | 2-[[4-(3,12-dihydroxy-10,13-dimethyl 2,3,4,5,6,7,8,9,11,12,14,15,16,17-tetradecahydro-1H-cyclopenta[a]phenanthren-17-yl)-1-oxopentyl]amino]acetic acid | -1.625 | Down | 1.519 | Up |
| MW0014969 | 6-{[1-(2-{[2-(3,4-dimethoxyphenyl)ethyl]-C-hydroxycarbonimidoyl}eth-1-en-1-yl)-3-methoxy-4-oxocyclohexa-2,5-dien-1-yl]oxy}-3,4,5-trihydroxyoxane-2-carboxylic acid | -0.962 | Down | -0.553 | Down |
| MW0128176 | (4-{2-[(1R,16Z,24E,26E,28Z)-1,18-dihydroxy-23-(hydroxymethyl)-19,30-dimethoxy-15,17,21,29,35-pentamethyl-2,3,10,14,20-pentaoxo-11,36-dioxa-4-azatricyclo[30.3.1.0,]hexatriaconta-16,24,26,28-tetraen-12-yl]propyl}-2-methoxycyclohexyl)oxidanesulfonic acid | -4.293 | Down | -0.637 | Down |
| MEDN0785 | LTB4 | -1.503 | Down | -0.862 | Down |
| MW0113931 | Amygdalin | -1.227 | Down | 1.407 | Up |
| MW0106148 | Chorismic acid | -3.216 | Down | 2.256 | Up |
| MW0055456 | O-Decanoyl-L-carnitine | -1.405 | Down | -0.717 | Down |
| MW0012191 | 12,13-DiHOME | -1.340 | Down | -0.627 | Down |
| MW0154243 | N-acetyl-N6,N6,O-tridemethylpuromycin-5’-phosphate | -0.731 | Down | -0.548 | Down |
| MW0148399 | Diacetyl-10-gingerdiol | -1.228 | Down | -1.269 | Down |
| MW0157815 | Thr-Thr-Gly-Leu-Ile | -0.496 | Down | -0.726 | Down |
| MW0149707 | Gln-His-Ser-Ser-Leu | 0.989 | Up | -1.118 | Down |
| MW0123548 | Deoxykhivorin | 1.904 | Up | -0.992 | Down |
| MEDP1343 | LPC(0:0/18:3) | 0.483 | Up | -1.125 | Down |
| MW0152097 | L-a-lysophosphatidylserine | 0.919 | Up | -1.303 | Down |
| MW0151632 | Ile-Tyr-Phe-Glu | 0.800 | Up | -1.079 | Down |
| MW0155794 | Pro-Asn-Gln-Met-Ser | 0.931 | Up | -0.736 | Down |
| MW0114597 | HMBOA-Glc | 0.684 | Up | 0.467 | Up |
| ZINC1629864 | Gyrophoric acid | 0.562 | Up | -0.764 | Down |
| FDATN00900 | D-Arabinose | 1.707 | Up | 0.582 | Up |
To further identify the potential metabolic pathways associated with the KDMs, we performed KEGG pathway enrichment analysis. The results revealed that these differential metabolites were mainly associated with six major metabolic pathways, including the peroxisome proliferator-activated receptor (PPAR) signaling pathway, choline metabolism in cancer, efferocytosis, linoleic acid metabolism, and phenylalanine, tyrosine, and tryptophan biosynthesis, as well as neuroactive ligand-receptor interaction (Figures 4E and 5). These pathways are primarily associated with important biological processes, including lipid and amino acid metabolism (Table 2).
| Pathway | Classification level 1 | Classification level 2 | P value | Enrichment score |
| PPAR signaling pathway | Organismal systems | Endocrine system | 0.0037 | 268.6667 |
| Choline metabolism in cancer | Human diseases | Cancer: Overview | 0.0136 | 73.2727 |
| Efferocytosis | Cellular processes | Transport and catabolism | 0.0258 | 38.3810 |
| Linoleic acid metabolism | Metabolism | Lipid metabolism | 0.0343 | 28.7857 |
| Phenylalanine, tyrosine and tryptophan biosynthesis | Metabolism | Amino acid metabolism | 0.0427 | 23.0286 |
| Neuroactive ligand-receptor interaction | Environmental information processing | Signaling molecules and interaction | 0.0499 | 19.6585 |
| Glycerophospholipid metabolism | Metabolism | Lipid metabolism | 0.0677 | 14.3929 |
| Folate biosynthesis | Metabolism | Metabolism of cofactors and vitamins | 0.0701 | 13.8966 |
| Ubiquinone and other terpenoid-quinone biosynthesis | Metabolism | Metabolism of cofactors and vitamins | 0.0853 | 11.3521 |
| Arachidonic acid metabolism | Metabolism | Lipid metabolism | 0.0945 | 10.2025 |
In summary, hUC-MSC therapy significantly reshapes the pancreatic metabolomic profile, modulates multiple key metabolic pathways, and ameliorates glycemic parameters in diabetic rats. However, these changes reflect combined signals from both endocrine and exocrine tissues rather than islet-specific effects; nevertheless, our findings provide preliminary support for the therapeutic potential of hUC-MSCs in diabetes, pending confirmation via cell-type-specific analyses.
ROC curve analysis revealed that all 30 KDMs demonstrated potential biomarker value in DM development [Figure 6, area under the curve (AUC) > 0.75] as well as in anti-diabetic efficacy (Figure 7, AUC > 0.8). Subsequently, we performed random-forest-based importance ranking of the KDMs in different comparison groups (Figure 8A and B). Using the random forest algorithm, we constructed a comprehensive ROC model containing 30 KDMs. The model achieved an AUC of 0.99 for both diagnosis (Figure 8C) and efficacy prediction (Figure 8D), with favorable sensitivity and specificity; these results are based on the current dataset and warrant further validation.
Spearman correlation analysis was conducted as an exploratory step to assess the associations between KDMs and DM-related biochemical indicators. The results showed that, when comparing the DM group compared to the control group, C-peptide, FINS, FBG, HOMA-IR, GHb, and GSP were significantly correlated with 16, 18, 22, 18, 22, and 21 metabolites, respectively (P < 0.05) (Figure 9A and B, Table 3). When comparing the DM + hUC-MSCs group to the DM group, C-peptide, FINS, FBG, HOMA-IR, GHb, and GSP were significantly correlated with 24, 16, 17, 17, 19, and 20 metabolites, respectively (P < 0.05) (Figure 9C and D, Table 4). These correlational findings indicate that aberrant KDMs are associated with DM-related hyperglycemia and impaired islet function; however, such associations do not imply causation. Rather, they provide hypothesis-generating evidence for the potential involvement of KDMs in the metabolic improvements observed following hUC-MSC treatment, and warrant further validation through targeted functional experiments.
| Biochemical index | Differential metabolite ID | Correlation coefficient | P value |
| C-peptide | MW0015937 | 0.788548168 | < 0.001 |
| C-peptide | MW0128176 | 0.876654109 | < 0.001 |
| C-peptide | MW0053755 | 0.689428985 | < 0.01 |
| C-peptide | MW0106148 | 0.766521683 | < 0.01 |
| C-peptide | MW0141637 | 0.766521683 | < 0.01 |
| C-peptide | MEDTN02450 | 0.674010446 | < 0.01 |
| C-peptide | MW0123548 | -0.621146881 | < 0.05 |
| C-peptide | MW0014969 | 0.570485965 | < 0.05 |
| C-peptide | MW0006200 | 0.546256832 | < 0.05 |
| C-peptide | MW0123754 | 0.557270074 | < 0.05 |
| C-peptide | MW0054549 | 0.605728342 | < 0.05 |
| C-peptide | MW0148399 | 0.546256832 | < 0.05 |
| C-peptide | MW0113931 | 0.610133639 | < 0.05 |
| C-peptide | FDATN00900 | -0.640970718 | < 0.05 |
| C-peptide | ZINC1629864 | -0.557270074 | < 0.05 |
| C-peptide | MW0055257 | 0.539648886 | < 0.05 |
| FINS | MW0106148 | 0.815384615 | < 0.001 |
| FINS | MW0006200 | 0.824175824 | < 0.001 |
| FINS | MW0054549 | 0.920879121 | < 0.001 |
| FINS | MW0053755 | 0.736263736 | < 0.01 |
| FINS | MW0123548 | -0.736263736 | < 0.01 |
| FINS | MW0015937 | 0.679120879 | < 0.01 |
| FINS | MW0014969 | 0.78021978 | < 0.01 |
| FINS | MW0055456 | 0.696703297 | < 0.01 |
| FINS | MW0141637 | 0.753846154 | < 0.01 |
| FINS | MEDP1339 | 0.608791209 | < 0.05 |
| FINS | MW0157815 | 0.661538462 | < 0.05 |
| FINS | MW0154243 | 0.635164835 | < 0.05 |
| FINS | MW0012191 | 0.657142857 | < 0.05 |
| FINS | MEDN0785 | 0.643956044 | < 0.05 |
| FINS | MW0123754 | 0.657142857 | < 0.05 |
| FINS | MW0128176 | 0.652747253 | < 0.05 |
| FINS | FDATN00900 | -0.595604396 | < 0.05 |
| FINS | ZINC1629864 | -0.538461538 | < 0.05 |
| FBG | MW0053755 | -0.879828407 | < 0.001 |
| FBG | FDATN00900 | 0.793830142 | < 0.001 |
| FBG | MW0106148 | -0.749728467 | < 0.01 |
| FBG | MW0014969 | -0.670345453 | < 0.01 |
| FBG | MW0006200 | -0.685781039 | < 0.01 |
| FBG | MW0123754 | -0.705626793 | < 0.01 |
| FBG | MW0141637 | -0.756343718 | < 0.01 |
| FBG | MW0054549 | -0.725472546 | < 0.01 |
| FBG | MW0148399 | -0.72767763 | < 0.01 |
| FBG | MW0128176 | -0.776189472 | < 0.01 |
| FBG | MW0113931 | -0.679165788 | < 0.01 |
| FBG | MW0157815 | -0.641679365 | < 0.05 |
| FBG | MW0123548 | 0.606398025 | < 0.05 |
| FBG | MW0154243 | -0.557886183 | < 0.05 |
| FBG | MW0012191 | -0.551270932 | < 0.05 |
| FBG | MW0015937 | -0.624038695 | < 0.05 |
| FBG | MEDN0785 | -0.546860764 | < 0.05 |
| FBG | MW0055456 | -0.544655681 | < 0.05 |
| FBG | MEDTN02450 | -0.599782774 | < 0.05 |
| FBG | MW0152407 | -0.538040429 | < 0.05 |
| FBG | ZINC1629864 | 0.56229635 | < 0.05 |
| FBG | MW0055257 | -0.617423444 | < 0.05 |
| HOMA-IR | MW0053755 | -0.872248812 | < 0.001 |
| HOMA-IR | MW0106148 | -0.775332277 | < 0.01 |
| HOMA-IR | MW0014969 | -0.676213094 | < 0.01 |
| HOMA-IR | MW0123754 | -0.735684604 | < 0.01 |
| HOMA-IR | MW0141637 | -0.696036931 | < 0.01 |
| HOMA-IR | MW0148399 | -0.707050173 | < 0.01 |
| HOMA-IR | MW0128176 | -0.777534926 | < 0.01 |
| HOMA-IR | FDATN00900 | 0.773129629 | < 0.01 |
| HOMA-IR | MW0123548 | 0.535243589 | < 0.05 |
| HOMA-IR | MW0154243 | -0.660794555 | < 0.05 |
| HOMA-IR | MW0114597 | 0.570485965 | < 0.05 |
| HOMA-IR | MW0015937 | -0.618944233 | < 0.05 |
| HOMA-IR | MW0144473 | -0.610133639 | < 0.05 |
| HOMA-IR | MW0006200 | -0.550662129 | < 0.05 |
| HOMA-IR | MW0054549 | -0.570485965 | < 0.05 |
| HOMA-IR | MW0113931 | -0.632160124 | < 0.05 |
| HOMA-IR | MW0152407 | -0.568283317 | < 0.05 |
| HOMA-IR | MW0055257 | -0.625552178 | < 0.05 |
| GHb | MW0141637 | -0.815384615 | < 0.001 |
| GHb | MW0054549 | -0.806593407 | < 0.001 |
| GHb | MW0053755 | -0.683516484 | < 0.01 |
| GHb | MW0123548 | 0.793406593 | < 0.01 |
| GHb | MW0106148 | -0.793406593 | < 0.01 |
| GHb | MW0015937 | -0.679120879 | < 0.01 |
| GHb | MW0014969 | -0.775824176 | < 0.01 |
| GHb | MW0006200 | -0.797802198 | < 0.01 |
| GHb | MW0123754 | -0.740659341 | < 0.01 |
| GHb | MW0128176 | -0.758241758 | < 0.01 |
| GHb | FDATN00900 | 0.714285714 | < 0.01 |
| GHb | MEDP1339 | -0.551648352 | < 0.05 |
| GHb | MW0157815 | -0.538461538 | < 0.05 |
| GHb | MW0154243 | -0.665934066 | < 0.05 |
| GHb | MW0114597 | 0.657142857 | < 0.05 |
| GHb | MW0012191 | -0.604395604 | < 0.05 |
| GHb | MEDN0785 | -0.648351648 | < 0.05 |
| GHb | MW0144473 | -0.591208791 | < 0.05 |
| GHb | MW0055456 | -0.661538462 | < 0.05 |
| GHb | MEDTN02450 | -0.621978022 | < 0.05 |
| GHb | MW0113931 | -0.648351648 | < 0.05 |
| GHb | MW0055257 | -0.657142857 | < 0.05 |
| GSP | MW0053755 | -0.705494505 | < 0.01 |
| GSP | MW0123548 | 0.714285714 | < 0.01 |
| GSP | MW0012191 | -0.687912088 | < 0.01 |
| GSP | MW0106148 | -0.692307692 | < 0.01 |
| GSP | MW0015937 | -0.789010989 | < 0.01 |
| GSP | MW0014969 | -0.701098901 | < 0.01 |
| GSP | MEDN0785 | -0.714285714 | < 0.01 |
| GSP | MW0006200 | -0.723076923 | < 0.01 |
| GSP | MW0055456 | -0.723076923 | < 0.01 |
| GSP | MW0141637 | -0.731868132 | < 0.01 |
| GSP | MW0054549 | -0.753846154 | < 0.01 |
| GSP | MEDTN02450 | -0.692307692 | < 0.01 |
| GSP | MW0128176 | -0.723076923 | < 0.01 |
| GSP | MW0152407 | -0.701098901 | < 0.01 |
| GSP | MEDP1339 | -0.556043956 | < 0.05 |
| GSP | MW0154243 | -0.595604396 | < 0.05 |
| GSP | MW0114597 | 0.556043956 | < 0.05 |
| GSP | MW0144473 | -0.547252747 | < 0.05 |
| GSP | MW0123754 | -0.542857143 | < 0.05 |
| GSP | FDATN00900 | 0.591208791 | < 0.05 |
| GSP | ZINC1629864 | 0.569230769 | < 0.05 |
| Biochemical index | Differential metabolite ID | Correlation coefficient | P value |
| C-peptide | MW0152097 | -0.874717531 | < 0.001 |
| C-peptide | MW0006200 | -0.943054838 | < 0.001 |
| C-peptide | MW0148399 | -0.86560589 | < 0.001 |
| C-peptide | MW0149707 | -0.797268583 | < 0.01 |
| C-peptide | MW0114597 | 0.76993366 | < 0.01 |
| C-peptide | MW0151632 | -0.815491865 | < 0.01 |
| C-peptide | MW0154243 | -0.829159327 | < 0.01 |
| C-peptide | MW0155794 | -0.829159327 | < 0.01 |
| C-peptide | MW0012191 | -0.788156942 | < 0.01 |
| C-peptide | MW0144473 | 0.829159327 | < 0.01 |
| C-peptide | MEDP1343 | -0.742598738 | < 0.01 |
| C-peptide | MW0054549 | 0.76993366 | < 0.01 |
| C-peptide | FDATN00900 | 0.774489481 | < 0.01 |
| C-peptide | MW0113931 | 0.760822019 | < 0.01 |
| C-peptide | MW0152407 | 0.760822019 | < 0.01 |
| C-peptide | MW0055257 | 0.824603506 | < 0.01 |
| C-peptide | MEDP1339 | -0.619591585 | < 0.05 |
| C-peptide | MW0123548 | -0.697040533 | < 0.05 |
| C-peptide | MW0015937 | -0.66970561 | < 0.05 |
| C-peptide | MW0014969 | -0.66970561 | < 0.05 |
| C-peptide | MEDN0785 | -0.678817251 | < 0.05 |
| C-peptide | MW0055456 | -0.683373071 | < 0.05 |
| C-peptide | MEDTN02450 | -0.706152174 | < 0.05 |
| C-peptide | ZINC1629864 | -0.605924123 | < 0.05 |
| FINS | MW0152097 | -0.872727273 | < 0.001 |
| FINS | MW0149707 | -0.918181818 | < 0.001 |
| FINS | MW0154243 | -0.945454545 | < 0.001 |
| FINS | MW0113931 | 0.927272727 | < 0.001 |
| FINS | MW0151632 | -0.781818182 | < 0.01 |
| FINS | MW0155794 | -0.827272727 | < 0.01 |
| FINS | MW0014969 | -0.790909091 | < 0.01 |
| FINS | MW0054549 | 0.8 | < 0.01 |
| FINS | MW0148399 | -0.8 | < 0.01 |
| FINS | MW0055257 | 0.827272727 | < 0.01 |
| FINS | MW0144473 | 0.618181818 | < 0.05 |
| FINS | MW0123754 | 0.681818182 | < 0.05 |
| FINS | MW0006200 | -0.7 | < 0.05 |
| FINS | MW0152407 | 0.7 | < 0.05 |
| FINS | MEDTN02450 | -0.681818182 | < 0.05 |
| FINS | MW0128176 | -0.672727273 | < 0.05 |
| FBG | MW0054549 | -0.851938429 | < 0.001 |
| FBG | MW0152097 | 0.742598738 | < 0.01 |
| FBG | MW0149707 | 0.76993366 | < 0.01 |
| FBG | MW0151632 | 0.756266199 | < 0.01 |
| FBG | MW0154243 | 0.756266199 | < 0.01 |
| FBG | MW0155794 | 0.738042917 | < 0.01 |
| FBG | MW0123754 | -0.820047686 | < 0.01 |
| FBG | MW0113931 | -0.774489481 | < 0.01 |
| FBG | MW0128176 | 0.738042917 | < 0.01 |
| FBG | MW0053755 | -0.706152174 | < 0.05 |
| FBG | MW0123548 | 0.728931276 | < 0.05 |
| FBG | MW0144473 | -0.615035764 | < 0.05 |
| FBG | MW0014969 | 0.683373071 | < 0.05 |
| FBG | MW0006200 | 0.610479944 | < 0.05 |
| FBG | MW0152407 | -0.615035764 | < 0.05 |
| FBG | MW0148399 | 0.642370687 | < 0.05 |
| FBG | MW0055257 | -0.733487097 | < 0.05 |
| HOMA-IR | MW0123754 | -0.890909091 | < 0.001 |
| HOMA-IR | MW0144473 | -0.836363636 | < 0.01 |
| HOMA-IR | MW0006200 | 0.772727273 | < 0.01 |
| HOMA-IR | MW0055456 | 0.772727273 | < 0.01 |
| HOMA-IR | MW0152407 | -0.8 | < 0.01 |
| HOMA-IR | MW0128176 | 0.772727273 | < 0.01 |
| HOMA-IR | MW0123548 | 0.727272727 | < 0.05 |
| HOMA-IR | MW0157815 | 0.690909091 | < 0.05 |
| HOMA-IR | MW0151632 | 0.681818182 | < 0.05 |
| HOMA-IR | MW0155794 | 0.618181818 | < 0.05 |
| HOMA-IR | MW0012191 | 0.745454545 | < 0.05 |
| HOMA-IR | MEDN0785 | 0.7 | < 0.05 |
| HOMA-IR | MW0054549 | -0.618181818 | < 0.05 |
| HOMA-IR | FDATN00900 | -0.681818182 | < 0.05 |
| HOMA-IR | MW0148399 | 0.727272727 | < 0.05 |
| HOMA-IR | MEDTN02450 | 0.654545455 | < 0.05 |
| HOMA-IR | MW0055257 | -0.663636364 | < 0.05 |
| GHb | MW0054549 | -0.872727273 | < 0.001 |
| GHb | MW0055257 | -0.936363636 | < 0.001 |
| GHb | MW0152097 | 0.854545455 | < 0.01 |
| GHb | MW0149707 | 0.827272727 | < 0.01 |
| GHb | MW0151632 | 0.781818182 | < 0.01 |
| GHb | MW0155794 | 0.790909091 | < 0.01 |
| GHb | MW0144473 | -0.809090909 | < 0.01 |
| GHb | MW0006200 | 0.8 | < 0.01 |
| GHb | MEDP1339 | 0.645454545 | < 0.05 |
| GHb | MW0053755 | -0.663636364 | < 0.05 |
| GHb | MW0123548 | 0.690909091 | < 0.05 |
| GHb | MW0154243 | 0.745454545 | < 0.05 |
| GHb | MW0012191 | 0.663636364 | < 0.05 |
| GHb | MW0123754 | -0.618181818 | < 0.05 |
| GHb | MW0113931 | -0.7 | < 0.05 |
| GHb | MW0152407 | -0.645454545 | < 0.05 |
| GHb | MW0148399 | 0.636363636 | < 0.05 |
| GHb | MEDTN02450 | 0.745454545 | < 0.05 |
| GHb | ZINC1629864 | 0.690909091 | < 0.05 |
| GSP | MW0154243 | 0.872727273 | < 0.001 |
| GSP | MW0113931 | -0.881818182 | < 0.001 |
| GSP | MW0148399 | 0.936363636 | < 0.001 |
| GSP | MW0015937 | 0.818181818 | < 0.01 |
| GSP | MW0006200 | 0.827272727 | < 0.01 |
| GSP | MW0054549 | -0.772727273 | < 0.01 |
| GSP | MW0152407 | -0.818181818 | < 0.01 |
| GSP | MW0152097 | 0.754545455 | < 0.05 |
| GSP | MW0149707 | 0.754545455 | < 0.05 |
| GSP | MW0123548 | 0.627272727 | < 0.05 |
| GSP | MW0114597 | -0.645454545 | < 0.05 |
| GSP | MW0151632 | 0.727272727 | < 0.05 |
| GSP | MW0155794 | 0.754545455 | < 0.05 |
| GSP | MW0012191 | 0.636363636 | < 0.05 |
| GSP | MW0144473 | -0.7 | < 0.05 |
| GSP | MW0106148 | -0.618181818 | < 0.05 |
| GSP | MW0014969 | 0.690909091 | < 0.05 |
| GSP | MW0123754 | -0.690909091 | < 0.05 |
| GSP | MEDTN02450 | 0.654545455 | < 0.05 |
| GSP | MW0055257 | -0.754545455 | < 0.05 |
DM is a chronic metabolic disorder primarily characterized by persistent hyperglycemia, which gradually damages multiple organ systems. Hyperglycemia associated with DM triggers both acute and chronic complications via diverse pathological processes. For instance, sustained hyperglycemia can damage vital organs including the kidneys, eyes, heart, nerves, and blood vessels. It constitutes a leading cause of complications including vision loss, myocardial infarction, kidney failure, strokes, and lower-limb amputation in individuals with DM[17,18]. Therefore, effective glycemic control and interventions that mitigate downstream pathological processes represent key strategies for preventing and managing DM-related complications. The present study found that hUC-MSC intervention reduced the levels of FBG, GHb, and GSP in STZ-induced DM rats, indicating a prominent therapeutic effect of the hUC-MSCs against hyperglycemia in these DM rats. This finding is consistent with previously published reports[19].
Pancreatic islet β-cells are crucial for regulating blood glucose levels through the release of insulin. The integrity of their structure and function is directly related to the efficacy of insulin secretion and the stability of blood glucose control. This research revealed through HE staining that rats in the DM group exhibited a decrease in both the size and number of islets, alongside unclear boundaries between the islets and the surrounding exocrine pancreatic tissue. However, treatment with hUC-MSCs restored the morphology, area, and number of islets in DM rats and rendered the boundary between the islets and exocrine pancreatic tissue clearer. TUNEL staining analysis revealed a marked increase in the apoptosis levels within pancreatic tissues of the DM rats, which was notably reduced following hUC-MSC treatment. Immunohistochemistry demonstrated a marked decrease in the number of insulin-expressing β-cells in the diabetic group, which was significantly recovered following hUC-MSC therapy. Biochemical analysis revealed that treatment with the hUC-MSCs significantly reduced FBG, GHb, GSP, and HOMA-IR index in diabetic rats, increased FINS and C-peptide levels, and improved overall islet function. These results indicate that hUC-MSCs have a significant repairing effect on pancreatic tissue and β-cell function in DM rats. These results are consistent with the outcomes reported in previous studies[20]. Research indicates that hUC-MSCs contribute to the restoration of islets, promote regeneration of β-cells, and inhibit programmed cell death in DM rats[21]. Furthermore, treatment with hUC-MSCs effectively reduces HOMA-IR in the DM rats, thereby suggesting that they can ameliorate diabetes-related insulin resistance. This therapeutic effect may be related to the immunomodulatory and metabolic regulatory functions of mesenchymal stem cells[22].
Previous studies have shown that DM induces extensive pancreatic metabolic disorders, involving alterations in glucose metabolism, lipid metabolism, bile acid composition, and amine levels[23]. In the present study, metabolomic analysis revealed that administering hUC-MSCs significantly ameliorated in pancreatic metabolic imbalances in diabetic rats. A total of 669 significantly differential metabolites were identified between the DM group and the control group, whereas 78 significant differential metabolites were identified in the comparison between the DM group and the hUC‑MSC‑treated DM group. These findings suggest that hUC-MSCs contribute to a broad restoration of pancreatic metabolism to a normal state. Furthermore, hUC-MSCs modulate multiple key classes of metabolites, including alcohols, amines, flavonoids, and hormone-related compounds[24]. KEGG pathway enrichment analysis of differential metabolites further revealed that hUC-MSCs modulated multiple key metabolic pathways associated with the pathogenesis and restoration of DM, including the PPAR signaling pathway, linoleic acid metabolism, vitamin B6 metabolism, and choline metabolism in cancer. These pathways are tightly associated with energy metabolism and insulin sensitivity. For example, activation of the PPAR signaling pathway promotes lipid utilization and improves insulin responsiveness[25], whereas dysregulation of vitamin B6 metabolism impairs γ-aminobutyric acid and neurotransmitter synthesis, thereby compromising pancreatic islet β-cell function[26]. Collectively, these findings are consistent with the systematic regulatory capacity of hUC-MSCs in correcting pancreatic metabolic imbalances.
This study also demonstrated via ROC curve analysis that the 30 DM-related differential metabolites identified in pancreatic tissue possess robust predictive value for DM onset and the therapeutic effect of hUC-MSCs (AUC > 0.75 for DM diagnosis; AUC > 0.80 for treatment prediction). These findings are consistent with previous metabolomic studies, which have found unique metabolic signatures associated with DM progression, especially those involving lipid, amino acid, and nucleotide metabolism[23]. Furthermore, the integrated ROC model constructed using all 30 KDMs yielded an AUC value of 0.99, indicating the potential of metabolite combinations as highly sensitive and specific biomarkers for diagnosis and treatment monitoring. Previous studies have confirmed changes in the levels of specific bile acids, amino acid derivatives, and lipid mediators in DM patients and animal models, some of which can be reversed by anti-diabetic drugs or cell therapy[27]. This study expands the understanding within this field and indicates that hUC-MSCs are associated with both the reversal of pathological metabolic perturbations in the pancreas and alterations in KDMs that correlate with functional improvement; however, these findings do not establish a direct causal relationship.
In summary, the KDMs screened from the present dataset serve as both predictive biomarkers for hUC-MSC-mediated therapeutic effects in DM and potential effectors. They may provide molecular insights into the mechanisms by which hUC-MSCs improve blood glucose control and maintain islet integrity. Future studies should focus on functional validation of the roles of these individual key metabolic biomarkers via targeted metabolomics, gene knockout/overexpression assays, or pharmacological regulation.
hUC-MSC therapy reduces blood glucose levels and improves both the structure and function of islets in diabetic rats, while ameliorating impaired pancreatic metabolism. Treatment with hUC-MSCs improves pancreatic metabolic disorders in the DM rats by regulating 30 KDMs.
| 1. | Ghanem SS, Al-Barazenji T, Badi R, Abdelalim EM. Bridging pancreatic and hepatic development: overlapping genes and their role in diabetes. Cell Mol Biol Lett. 2025;30:113. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 3] [Reference Citation Analysis (0)] |
| 2. | Abel ED, Gloyn AL, Evans-Molina C, Joseph JJ, Misra S, Pajvani UB, Simcox J, Susztak K, Drucker DJ. Diabetes mellitus-Progress and opportunities in the evolving epidemic. Cell. 2024;187:3789-3820. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 204] [Cited by in RCA: 181] [Article Influence: 90.5] [Reference Citation Analysis (1)] |
| 3. | Siehler J, Blöchinger AK, Meier M, Lickert H. Engineering islets from stem cells for advanced therapies of diabetes. Nat Rev Drug Discov. 2021;20:920-940. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 106] [Cited by in RCA: 87] [Article Influence: 17.4] [Reference Citation Analysis (0)] |
| 4. | Niu F, Liu W, Ren Y, Tian Y, Shi W, Li M, Li Y, Xiong Y, Qian L. β-cell neogenesis: A rising star to rescue diabetes mellitus. J Adv Res. 2024;62:71-89. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 54] [Cited by in RCA: 52] [Article Influence: 26.0] [Reference Citation Analysis (1)] |
| 5. | Yong J, Johnson JD, Arvan P, Han J, Kaufman RJ. Therapeutic opportunities for pancreatic β-cell ER stress in diabetes mellitus. Nat Rev Endocrinol. 2021;17:455-467. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 52] [Cited by in RCA: 189] [Article Influence: 37.8] [Reference Citation Analysis (4)] |
| 6. | Li L, Li J, Guan H, Oishi H, Takahashi S, Zhang C. Human umbilical cord mesenchymal stem cells in diabetes mellitus and its complications: applications and research advances. Int J Med Sci. 2023;20:1492-1507. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 19] [Reference Citation Analysis (3)] |
| 7. | Liu P, Cao B, Zhou Y, Zhang H, Wang C. Human umbilical cord-derived mesenchymal stem cells alleviate oxidative stress-induced islet impairment via the Nrf2/HO-1 axis. J Mol Cell Biol. 2023;15:mjad035. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 2] [Cited by in RCA: 17] [Article Influence: 5.7] [Reference Citation Analysis (0)] |
| 8. | Xie T, Huang Q, Huang Q, Huang Y, Liu S, Zeng H, Liu J. Dysregulated lncRNAs regulate human umbilical cord mesenchymal stem cell differentiation into insulin-producing cells by forming a regulatory network with mRNAs. Stem Cell Res Ther. 2024;15:22. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 1] [Cited by in RCA: 5] [Article Influence: 2.5] [Reference Citation Analysis (0)] |
| 9. | Tang S, Feng K, Yang R, Cheng Y, Shi N, Zhang H, Wei Z, Ma Y. A dual-action strategy: Wound microenvironment responsive hydrogel and exosome-mediated glucose regulation enhance inside-out diabetic wound repair. J Control Release. 2025;382:113716. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 43] [Cited by in RCA: 46] [Article Influence: 46.0] [Reference Citation Analysis (0)] |
| 10. | Wang L, Qi C, Cao H, Zhang Y, Liu X, Qiu L, Wang H, Xu L, Wu Z, Liu J, Wang S, Kong D, Wang Y. Engineered Cytokine-Primed Extracellular Vesicles with High PD-L1 Expression Ameliorate Type 1 Diabetes. Small. 2023;19:e2301019. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 1] [Cited by in RCA: 25] [Article Influence: 8.3] [Reference Citation Analysis (0)] |
| 11. | Wen W, Cao Y, Chen P, Li J, Li W, Huang G, Zheng H, Zhu X, Zhang H, Chen Y, Huang X, Hu Y, Huang Y. A reliable strategy for establishment of an animal model of diabetic cardiomyopathy: Induction by a high-fat diet combined with single or multiple injections of low-dose streptozotocin. Life Sci. 2024;358:123161. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 3] [Cited by in RCA: 13] [Article Influence: 6.5] [Reference Citation Analysis (0)] |
| 12. | Li X, Guo L, Chen J, Liang H, Liu Y, Chen W, Zhou L, Shan L, Wang H. Intravenous injection of human umbilical cord-derived mesenchymal stem cells ameliorates not only blood glucose but also nephrotic complication of diabetic rats through autophagy-mediated anti-senescent mechanism. Stem Cell Res Ther. 2023;14:146. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 15] [Cited by in RCA: 24] [Article Influence: 8.0] [Reference Citation Analysis (4)] |
| 13. | Sedik AA, El-Nasr NMEA, Khalil WKB, El-Mosallamy AEMK. Unravelling the mechanism by which vildagliptin and linagliptin inhibit pyroptosis in lung injury through the NLRP3 inflammatory pathway in type 1 diabetic rats. Sci Rep. 2025;15:20292. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 1] [Cited by in RCA: 5] [Article Influence: 5.0] [Reference Citation Analysis (0)] |
| 14. | Selim SM, El Fayoumi HM, El-Sayed NM, Mehanna ET, Hazem RM. Alogliptin attenuates STZ-induced diabetic nephropathy in rats through the modulation of autophagy, apoptosis, and inflammation pathways: Targeting NF-κB and AMPK/mTOR pathway. Life Sci. 2025;361:123307. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 13] [Reference Citation Analysis (0)] |
| 15. | Liu Y, Chen J, Liang H, Cai Y, Li X, Yan L, Zhou L, Shan L, Wang H. Human umbilical cord-derived mesenchymal stem cells not only ameliorate blood glucose but also protect vascular endothelium from diabetic damage through a paracrine mechanism mediated by MAPK/ERK signaling. Stem Cell Res Ther. 2022;13:258. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 97] [Cited by in RCA: 85] [Article Influence: 21.3] [Reference Citation Analysis (3)] |
| 16. | Lindahl A, Heuchel R, Forshed J, Lehtiö J, Löhr M, Nordström A. Discrimination of pancreatic cancer and pancreatitis by LC-MS metabolomics. Metabolomics. 2017;13:61. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 34] [Cited by in RCA: 41] [Article Influence: 4.6] [Reference Citation Analysis (2)] |
| 17. | Jiang H, Xia C, Lin J, Garalleh HA, Alalawi A, Pugazhendhi A. Carbon nanomaterials: A growing tool for the diagnosis and treatment of diabetes mellitus. Environ Res. 2023;221:115250. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 49] [Cited by in RCA: 26] [Article Influence: 8.7] [Reference Citation Analysis (0)] |
| 18. | Lam BQ, Srivastava R, Morvant J, Shankar S, Srivastava RK. Association of Diabetes Mellitus and Alcohol Abuse with Cancer: Molecular Mechanisms and Clinical Significance. Cells. 2021;10:3077. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 2] [Cited by in RCA: 17] [Article Influence: 3.4] [Reference Citation Analysis (0)] |
| 19. | Chin SP, Abd Rahim ENA, Nor Arfuzir NN, Then KY, Cheong SK. Repeated Intravenous Dosing of Human Umbilical Cord Mesenchymal Stem Cells Improves Glycemic Control and Organ Protection in a Type 2 Diabetes Rat Model. Diabetes Metab Syndr Obes. 2025;18:4319-4334. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 3] [Reference Citation Analysis (0)] |
| 20. | Khatri R, Petry SF, Linn T. Intrapancreatic MSC transplantation facilitates pancreatic islet regeneration. Stem Cell Res Ther. 2021;12:121. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 33] [Cited by in RCA: 34] [Article Influence: 6.8] [Reference Citation Analysis (0)] |
| 21. | Bao Y, Zhao Z, Gao H. Effect of hTIMP-1 overexpression in human umbilical cord mesenchymal stem cells on the repair of pancreatic islets in type-1 diabetic mice. Cell Biol Int. 2021;45:1038-1049. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 2] [Cited by in RCA: 12] [Article Influence: 2.4] [Reference Citation Analysis (0)] |
| 22. | Shi Y, Wang Y, Li Q, Liu K, Hou J, Shao C, Wang Y. Immunoregulatory mechanisms of mesenchymal stem and stromal cells in inflammatory diseases. Nat Rev Nephrol. 2018;14:493-507. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 994] [Cited by in RCA: 918] [Article Influence: 114.8] [Reference Citation Analysis (8)] |
| 23. | Yedjou CG, Grigsby J, Mbemi A, Nelson D, Mildort B, Latinwo L, Tchounwou PB. The Management of Diabetes Mellitus Using Medicinal Plants and Vitamins. Int J Mol Sci. 2023;24:9085. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 80] [Reference Citation Analysis (0)] |
| 24. | Yang H, Xie Y, Li T, Liu S, Zeng S, Wang B. A novel minimally invasive OFM technique with orthotopic transplantation of hUC-MSCs and in vivo monitoring of liver metabolic microenvironment in liver fibrosis treatment. Stem Cell Res Ther. 2021;12:534. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 2] [Cited by in RCA: 6] [Article Influence: 1.2] [Reference Citation Analysis (1)] |
| 25. | Wang B, Fu W, Ueda A, Shah H, Wu CF, Chi W, Zhuang X. Genetic vitamin B6 deficiency and alcohol interaction in behavior and metabolism. bioRxiv. 2025;2025.03.06.641947. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 1] [Reference Citation Analysis (2)] |
| 26. | Noso S, Babaya N, Hiromine Y, Taketomo Y, Niwano F, Yoshida S, Ikegami H. Metabolic signatures of β-cell destruction in type 1 diabetes. J Diabetes Investig. 2023;14:48-57. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 7] [Reference Citation Analysis (0)] |
| 27. | Meneses MJ, Silva BM, Sousa M, Sá R, Oliveira PF, Alves MG. Antidiabetic Drugs: Mechanisms of Action and Potential Outcomes on Cellular Metabolism. Curr Pharm Des. 2015;21:3606-3620. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 43] [Cited by in RCA: 63] [Article Influence: 5.7] [Reference Citation Analysis (1)] |