From in vitro to in vivo: hypoxia attenuates replicative senescence and preserves therapeutic activity in long-term passaged human umbilical cord-derived mesenchymal stem cells

Abstract

Background

Mesenchymal stem cells (MSCs) hold great potential for regenerative medicine and tissue engineering. However, their clinical use is limited by low in vivo availability and rapid senescence during ex vivo expansion, posing a major challenge for scaled production. Growing evidence indicates that microenvironmental oxygen tension critically regulates MSCs fate. This study aims to systematically evaluate how different oxygen tensions modulate MSCs aging and function, providing evidence to optimize MSCs expansion protocols for clinical applications.

Methods

Human umbilical cord derived-mesenchymal stem cells (hUC-MSCs) were maintained under normoxic (20%O₂) and hypoxic (5%O₂) conditions through continuous passaging until senescence occurred. In vitro analysis systematically compared proliferation, cell cycle distribution, apoptosis, and senescence biomarkers across serially expanded populations from both oxygen concentrations. For in vivo evaluations, a type 2 diabetic nephropathy mouse model was used to investigate therapeutic efficacy. Mice received systemic transplantation of hUC-MSCs from each group, and renal functional recovery along with histopathological changes were assessed.

Results

Cells cultured under both 20%O₂ (normoxia) and 5%O₂ (hypoxia) stably expressed MSC-specific surface markers throughout serial passages. Hypoxia-cultured cells retained enhanced adipogenic and osteogenic differentiation potential, along with superior proliferative capacity and growth factor secretion, through at least 16 passages, consistently outperforming normoxic counterparts. Hypoxic conditions also attenuated replicative senescence, suppressed DNA damage, and reduced secretion of SASP-associated pro-inflammatory factors (including IL-6, IL-8, and TGF-β), while decreasing apoptosis. In a diabetic nephropathy mouse model, P16 MSCs cultured under hypoxia still significantly improved renal function and attenuated fibrosis post-transplantation, as demonstrated by reduced 24-h urinary protein, serum creatinine, and blood urea nitrogen levels compared to the normoxic MSCs group.

Conclusion

We provide the first in vivo evidence that hypoxic culture conditions enables high-passage hUC-MSCs(P16) to retain therapeutic potency in diabetic nephropathy. Our findings demonstrate that hypoxia effectively mitigates replicative senescence through suppression of oxidative damage and SASP secretion, thereby preserving the proliferative capacity and paracrine repair functions of long-term cultured cells.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s40659-026-00703-z.

Background

Human mesenchymal stem cells, known for their self-renewal and multipotent capacities, have been successfully isolated and expanded from a variety of tissues including bone marrow, adipose tissue, umbilical cord tissue, dental pulp, and placental tissues [1–3]. Moreover, Mesenchymal stem cells (MSCs) also exhibit potent immunomodulatory functions [4], autocrine/paracrine activities [5], and low immunogenicity, thereby attracted significant interest in the fields of regenerative medicine and tissue engineering. Mounting evidence has made MSCs the most promising alternative to conventional treatments for autoimmune diseases, degenerative diseases, and ageing related diseases. Achieving optimal therapeutic outcomes requires the administration of sufficient quantities of high-quality MSCs. Since their natural abundance in donor tissues is low, in vitro expansion becomes a necessary step before clinical application. However, during large-scale and long-term serial passaging, MSCs reach the Hayflick limit, entering a state of replicative senescence, accompanied by markedly reduced proliferation eventually leading to growth arrest [6]. Meanwhile, MSCs undergo morphological and functional alterations such as cell enlargement and flattening, loss of differentiation potential, diminished secretory function, and reduced therapeutic efficacy [7–11]. Therefore, it is of utmost importance to develop and optimize manufacturing protocols that not only supports large-scale expansion of MSCs but also effectively delay cellular senescence while preserving their functional biological properties.

While the importance of nutrients, growth factors, temperature, and pH in cell culture is well recognized and routinely controlled in practice, the role of oxygen tension has received far less attention [12]. Currently, ambient air conditions—with oxygen concentrations around 21% or slightly lower in standard CO₂ incubators (− 19%)—remain the standard for in vitro MSCs expansion and are typically referred to as “normoxia”. However, in vivo oxygen levels in most organs or tissues range from 2 to 10%, and can even fall below 1% in certain microenvironments such as the kidney, thymus, or solid tumor [13–15]. Thus, while often labeled “normoxic”, conventional cell culture conditions actually represent a non-physiological hyperoxic state relative to the physiologically hypoxic conditions in native tissue niches. Despite this significant discrepancy, the difference between in vitro and in vivo oxygen environments is frequently overlooked.

Given that elevated oxygen tension can induce oxidative stress, leading to DNA damage and cell senescence [16], it has been postulated that maintaining MSCs under hypoxic conditions may improve their overall quality. Consequently, a growing number of studies have explored the influence of low O₂ tension on the biological properties of MSCs. Although outcomes vary across studies, largely due to differences in MSCs tissue sources, oxygen concentrations (ranging from 1 to 10%), and culture conditions, most reports have consistently observed beneficial effects of hypoxic culture on MSCs survival, proliferation, migration, and differentiation potential. In particular, multiple lines of data suggested that hypoxic condition can delay cellular senescence and reduce the expression of senescence-associated markers [17–22].

However, most existing studies have focused on short-term or transient hypoxic exposure, while comparative studies examining the effects of prolonged hypoxia versus normoxia on MSCs senescence and biological functionality in vitro remain limited. More importantly, robust in vivo evidence validating the functional relevance of these hypoxic adaptations, particularly whether the therapeutic efficacy of MSCs is maintained after long-term serial passaging under hypoxia is still absent [23, 24].

This study systematically evaluates the effects of sustained hypoxia on Human umbilical cord derived-mesenchymal stem cells (hUC-MSCs) both in vitro and in vivo. Low-passage hUC-MSCs were maintained under normoxic (20%O₂) or hypoxic (5%O₂) conditions and subjected to serial passaging. The anti-senescence benefits and functional alterations resulting from hypoxic conditioning were assessed at multiple passage time points. Most significantly, to directly assess the physiological relevance, we established a type 2 diabetic nephropathy (DN) mouse model to evaluate and compare the therapeutic efficacy of cells expanded under normoxic versus hypoxic conditions across successive passages.

Methods

Expansion and characterization of hUC-MSCs

Primary hUC-MSCs (P0 cells) were obtained from Fengzekang Biopharmaceutical Co., Ltd. (Shenzhen,China) and cultured at 37 °C with 95% humidity, 5%CO₂ under either 5%O₂ (hypoxic) or 20%O₂ (normoxic) conditions. Cells were harvested at 80–90% confluency, with experiments conducted on passages4-16. Multilineage differentiation potential was confirmed through Oil Red O staining for adipogenesis and Alizarin Red staining for osteogenesis using Thermo Fisher staining kits. Specific surface marker expression was analyzed by flow cytometry (BD Accuri C6), assessing MSCs-positive markers (CD29, CD44, CD73, CD90, CD105) and exclusion markers (HLA-DR, CD31, CD34) with FlowJo software (BD Biosciences).

Cell proliferation assay

Passage4 and passage16 cells maintained under hypoxic or normoxic conditions were plated at a density of 500 cells/cm² in a 96-well plate. Cell proliferation assay were conducted everyday, day 0 through day 6 culture. After diluting the CCK-8 reagent with hUC-MSCs culture medium at a ratio of 1:10, 100 μL of the diluted CCK-8 solution was added to each well in the experimental groups and blank group. The plate was then incubated at 37 °C for 1 h in dark, then 96 well plate was measured the absorbance at 450 nm using SYNERGY H1 full-wave length microplate reader (BioTek, USA) and OD values were obtained, and the data were analyzed with the CHS3.8 software.

Starting from Passage4 and continuing every four passages until Passage16, cells maintained under hypoxic or normoxic conditions were trypsinized and harvested, followed by washing with PBS. The cells were then resuspended in 300 μL of PBS, and 900 μL of 95% ethanol was added for fixation for 12 h, followed by PBS washing. The cells were then resuspended in 100 μL of PBS and incubated at room temperature with 2 μL of Ki67 (BioLegend, USA) solution for 15 min. Cell proliferation was assessed using a flow cytometer (BD Accuri C6), and the data were analyzed using FlowJo software (BD, USA). The fluorescence intensity of the cells was evaluated (BD FACSVerse™) and analyzed using BD FACSuite™ software.

Cell viability analysis

Cells were trypsinized and harvested, followed by PBS washing. Then, the cells were resuspended in 500 μL of binding buffer and then incubated at room temperature with 3 μL of Annexin V-Alexa Fluor 647 and 5 μL of 7-aminoactinomycin D (7-AAD) solution for 15 min (provided by 4A Biotech, Beijing, China). Cell apoptosis was evaluated by flow cytometry (BD Accuri C6), and the data were analyzed using FlowJo software (BD, USA). The fluorescence intensity of the cells was evaluated (BD FACSVerse™), and were analyzed using the BD FACSuite™ software.

Cell cycle analysis by flow cytometry

Cells were trypsinized and harvested, followed by PBS washing. Then, the cells were resuspended in 300 μL of PBS, and 900 μL of 95% ethanol was added dropwise to fix them for 12 h, followed by PBS washing. Then, 400 μL of staining buffer, 15 μL of propidium iodide staining solution (25X), and 4 μL of RNase A (2.5 mg/mL) were added to each sample and incubated at 37 °C in the dark for 30 min (provided by 4A Biotech, Beijing, China). Cell cycle analysis was performed by flow cytometry (BD Accuri C6), and Modfit LT™ was used to assess the distribution of hUC-MSCs across different cell cycle phases.

Reactive oxygen species detection by DCFH-DA fluorescent probe

DCFH-DA (Sigma, USA) was diluted to a concentration of 5 μM using DMEM/F12 basal medium, and the solution was then mixed evenly and added to cells maintained under hypoxic or normoxic conditions. The cells were incubated at 37 °C for 20 min, avoiding light throughout the process. Reactive oxygen species (ROS) were detected using a flow cytometer (BD Accuri C6), and the data were analyzed using FlowJo software (BD, USA). The fluorescence intensity of the cells was evaluated (BD FACSVerse™) and analyzed using BD FACSuite™ software.

RNA extraction and real-time PCR

Cells maintained under different oxygen concentrations and passage numbers were collected. RNA purification was conducted using the EasyPure® RNA Kit (TransGen, Beijing, China) in accordance with the manufacturer's instructions. Following this, genomic DNA was removed, and cDNA was synthesized using the Evo M-MLV Reverse Transcription Premix Kit (AGbio, China). Subsequently, quantitative real-time polymerase chain reaction (ABI StepOne Plus) was performed on the synthesized cDNA using SYBR Green Master Mix (AGbio, China) and the primers detailed in Table 1. Following the manufacturer's instructions, the probe-based qPCR kit (tjs, CAT#17–001-htert) was used to detect the core telomerase gene TERT. All analyses and measurements were conducted using the LightCycler® 480 II Real-Time PCR Detection System (Roche).

Table 1.

Primer pairs for each gene

Gene	Primer sequence	Product size (bp)
LAMNB1	F:5'-GAGAGCAACATGAGCCAAGTG-3'	129
	R:5'-GTTCTTCCCTGGCACTGTTGAC-3'
P16	F:5'-GAGCAGCATGGAGCCTTC-3'	116
	R:5'-CGTAACTATTCCGTGCGTTC-3'
P21	F:5'-GCCATTACCCCATCACACT-3'	140
	R:5'-ACCGAGGCACTCAGAGGAG-3'

Telomere length analysis

Genomic DNA was extracted from cells, and telomere length was assessed using the Telomere Length Quantification qPCR Kit (tjs, CAT#14-001rTL) according to the manufacturer's protocol. Relative telomere length (T/S ratio) was calculated by comparing the copy number of telomeric repeat sequences to that of a single-copy reference gene, with normalization performed using primers targeting a 78 bp region on human chromosome 11. The T/S ratio is a widely accepted approach for relative telomere length quantification, in which T denotes the telomeric repeat copy number and S refers to the copy number of a stably expressed single-copy reference gene. This ratio normalizes variations in DNA input and amplification efficiency across samples, thereby allowing reliable and consistent comparison of relative telomere length among different cell passage groups [25]. All qPCR reactions were run on the CFX 96 Touch™ Real-Time PCR Detection System (Bio-Rad).

Enzyme-linked immunosorbent assays

P4, P10, and P16 hUC-MSCs maintained under hypoxic or normoxic conditions were seeded in a 6-well plate at 20,000 cells/cm². Following cell attachment, the cells were washed with PBS, and incubated with 1 mL phenol red-free DMEM/F12 basal medium (Procell, China) for 24 h. Supernatants were then collected for analysis. Secreted levels of FGF, VEGF, IL-6, IL-8, and TGFβ were quantified using ELISA kit (4A Biotech, Beijing, China). DNA damage biomarker 8-OHdG was measured with a separate ELISA kit (spbio, Wuhan, China).

Senescence associated β-galactosidase (SA-β-gal) staining

Passage16 hUC-MSCs maintained under hypoxic or normoxic conditions were seeded in a 6-well plate at 20,000 cells/cm². Upon reaching approximately 80% confluency, cells were fixed for 15 min, washed thrice with PBS, and incubated overnight in a 37 °C with 1 mL of β-galactosidase staining solution (Beyotime Biotechnology, Shanghai, China). Stained cells were Imaged using an inverted phase contrast microscope (Leica, Germany).

Mouse model of diabetic nephropathy and hUC-MSCs treatment

All procedures were approved by the Medical Ethics Committee of the First Affiliated Hospital of Shantou University (ethical approval number: SUMC2021-267), Male ICR mice (4-week-old, n = 60; Beijing Vital River Laboratory Animal Technology) were housed under specific pathogen-free (SPF) conditions (24–26 °C, 12-h light/dark cycle) at the Experimental Animal Center of Shantou Medical College, each mice considered an experimental unit. The mice were randomly assigned to the following groups: control group, DN group, HypoxiaP4 treatment group, NormoxiaP4 treatment group, HypoxiaP16 treatment group, and NormoxiaP16 treatment group. We sample size (n = 6/group) was determined based on animal welfare considerations and in line with a previous study [26]. Mice that died unexpected during the extended modeling period for diabetic nephropathy, which prevented data collection, were excluded from the analysis. A final sample size of n = 4 per group was ensured to maintain data accuracy and consistency. After 1 week of normal diet, the model mice received high-fat diet (HFD, D12492 60% kcal% fat) for 3 weeks. Following overnight fasting, streptozotocin (STZ; 40 mg/kg) was administered intraperitoneally for 4 consecutive days. Diabetic mice (blood glucose ≥ 16.7 mmol/L at 1 week post-STZ) continued HFD for 6 more weeks. Sustained hyperglycemia with urinary protein levels ≥ 20 mg/kg significantly exceeding control group levels, along with accompanying renal pathological changes, is defined as DN models. The treatment group mice received weekly tail vein injections of hUC-MSCs (2 × 10⁵ cells/200 μL PBS) for a duration of 4 weeks. To avoid potential confounding effects from the cell injection, an equivalent volume of PBS was administered to the control group. The study was performed by two investigators: the first investigator performed the random allocation of mice to treatment groups, while all treatment were administered exclusively by the second investigator to minimize variability and bias. Anesthesia was administered using isoflurane (RWD, R510-22–10), with the trachea fully exposed in the anesthesia machine (Matrx, VIP 3000). The oxygen flow rate was maintained at 0.5–1 L/min, with an isoflurane concentration of 3% during induction and 1.5% during maintenance. Mice were euthanized by cervical dislocation 24 h after the final injection for sample collection. All procedures performed on animals were conducted in accordance with ARRIVE guidelines.

In vivo tracking of CM-DiL-labeled hUC-MSCs

Cell tracking was performed using an IVIS Kinetic small animal live imaging system (PerkinElmer) at the Animal Center of Shantou University Medical College. Mice were anesthesia with isoflurane inhalation according to animal ethics protocols. Control and DN groups (n = 3/group) received injections of either PBS or CM-DiL-labeled hUC-MSCs (2 × 10⁵ cells/200μL PBS). Prior to injection, hUC-MSCs were incubated with 0.2 µM CM-DiL working solution (Bestbio, Shanghai, China) for 20 min at 37 °C followed by three PBS washes prior to transplantation. Then imaging was conducted at 8 h, 24 h, 48 h, 4 days, and 7 days post-injection.

Histological analysis

Kidneys were sagittally sectioned, fixed in 10% neutral buffered formalin, paraffin-embedded, and cut into 4 μm sections. Tissue morphology was assessed by hematoxylin and eosin (HE; Wexis, China) and Masson trichrome (Kaiqi Biology, China) staining. Slides were imaged using an CX31 Olympus upright microscope (Olympus, Japan).

Immunohistochemistry staining

Kidney sections were processed using the UltraSensitive SP Mouse/Rabbit Immunohistochemistry Kit (MXB, China). After blocking, sections were incubated overnight at 4 °C with primary antibodies: mouse anti-α-SMA (1:2000) and rabbit anti-TGF-β (1:1000), followed by species-matched secondary antibodies (goat anti-mouse/rabbit IgG). DAB development and hematoxylin counterstaining were performed after PBS washes. Sections were dehydrated through graded ethanol/xylene and imaged on an CX31 Olympus upright microscope (Olympus, Japan).

Scanning electron microscope analysis

After the mice were euthanized via cervical dislocation, their kidney tissues were promptly excised within 1 min. The samples were then fixed in 2.5% glutaraldehyde solution and trimmed to around 0.5 mm³ before their skin was removed. Subsequently, the kidney tissue samples were washed thrice in PBS 15 min/wash. Following this, the tissue samples were infiltrated with osmium tetroxide for 1 h until a darkened appearance was achieved. After additional PBS washes (3 × 15 min), the tissue samples tissues were dehydrated through graded acetone series (30%, 50%, 70%, 90%, 100%; 30 min each) To prepare for the final viewing of the slices using the scanning electron microscope, a dehydration solution was meticulously prepared with specific ratios of anhydrous acetone and Embedding Agents A (1:3), B (1:5), and C (pure embedding agent) for a 12-h sequential tissue infiltration process. Ultrathin Sects. (70 nm) were collected on copper grids for SEM imaging (JEM-F200, Japan).

Western blotting

Kidney tissues (around 20 mg each) were homogenized in 200 μL ice-cold strong RIPA lysis buffer (Solarbio, China) supplemented with 2μL PMSF, protease and phosphatase inhibitors. Protein concentrations were measured by BCA assay. Protein samples (around 10 μg) were loaded and separated on 10% SDS-PAGE gel, and transferred to polyvinylidene difluoride (PVDF) membrane (Millipore, US). After 1-h blocking with 5% non-fat milk, membranes were incubated overnight at 4 °C with primary antibodies: anti-Col3 (1:1000, ABclonal, China), anti-β-tublin (1:2000, CST, USA), and anti-α-SMA (1:1000, ABclonal, China). Following thrice washes with TBST, membranes were incubated with species-matched HRP-conjugated secondary antibodies (1:5000) for 1 h at room temperature. Specific protein bands were visualized using ECL reagent (NCM, China).

RNA sequencing

Total RNA was extracted from hUC-MSCs using Trizol reagent (invitrogen USA). Libraries were sequenced on an Illumina HiSeq 6000 platform by Novogene. Differential gene expression analysis was performed with DEGSeq R package (version 1.16.1). Unsupervised hierarchical clustering of significantly dysregulated genes was visualized by heatmap. Pathway enrichment analysis utilized the KEGG database (p < 0.05).

Statistical analysis

All data are presented as mean ± SEM and analyzed using GraphPad Prism 10. Statistical significance level was defined as p < 0.05. Group comparisons were performed by one-way or two-way ANOVA followed by LSD post-hoc tests where appropriate. Flow cytometry data were processed with modfit LT and FlowJo (BD Biosciences).

Result

The effects of hypoxia on biological characteristics of hUC-MSCs across serial passaging

To assess the impacts of oxygen concentrations on biological properties of hUC-MSCs during continuous passaging, in vitro differentiation assay revealed that hUC-MSCs cultured under both hypoxia and normoxia maintained adipocyte and osteoblast differentiation potential after serial passaging, although the capacity progressively diminished with increasing passage numbers (Supplemental Fig. 1A). Then flow cytometry analysis of surface marker expression showed that regardless of the oxygen concentration used during culture, high-passage hUC-MSCs consistently expressed CD29, CD44, CD73, CD90, and CD105, while remaining negative for HLA-DR, CD31 and CD34 (Supplemental Fig. 1B).

Hypoxia increases and maintains cell proliferative capacity of hUC-MSCs during serial passaging

To assess the impact of oxygen levels on cell proliferation, the CCK-8 assay revealed significantly enhanced proliferation cultured under 5%O₂ hypoxia versus 20%O₂ normoxia across serial passages(P4-P16), evidenced by elevated OD450 values over 7-day culture. Strikingly, high-passage(P16) normoxic cultured cells exhibited near-complete proliferative arrest, while hypoxic cultured Passage16 hUC-MSCs retained relatively high proliferative capacity (Fig. 1A).

Fig. 1.

Effects of hypoxia on high-passage hUC-MSCs. (A) Proliferative capacity measured by CCK-8 assay (OD450) under normoxia (20%O₂) and hypoxia (5%O₂) at P4, P10, and P16 (days 0–6). PI staining. N = 3 (per group), data are mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. corresponding Normoxia group; ^#p < 0.05, ^###p < 0.001, ^####p < 0.0001 vs. corresponding Normoxia group; ^△△P < 0.01 vs. corresponding Normoxia group. (B) Cell cycle distribution analyzed by (C) Ki67 staining mean fluorescence intensity (MFI) quantification measured by Flow cytometry. N = 3 per group; ns = no significance, *p < 0.05 vs. corresponding Normoxia group

To further confirm this phenomenon, the expression of cell proliferation marker, Ki67 and cell cycle were also determined by flow cytometry after Immunofluorescence staining and PI staining. In consistent with CCK-8 assay, cell cycle profiling revealed that normoxia induced significant G0/G1 phase accumulation and decline in the S + G2/M phase populations, while hypoxia suppressed G0/G1 arrest and high-passage(P16) cells maintained significantly lower G0/G1 proportions than normoxic counterparts (Fig. 1B). The Ki67 staining results showed that hypoxic cultures maintained markedly higher fluorescence intensity versus normoxia, indicating sustained proliferative potential. However, progressively diminished mean fluorescence intensity in normoxia was observed with continuous passages (Fig. 1C). Collectively, although in vitro serial passaging progressively impairs hUC-MSCs proliferation regardless of oxygen levels, physiologically-relevant hypoxia (5%O₂) dramatically mitigates this decline, preserving > 30% cycling cells even at late passages(P16) through reduced G0/G1 blockade and sustained proliferative marker expression.

Hypoxic culture conditions protects hUC-MSCs from oxidative damage and effectively delays the cellular senescence during extended serial passaging

To ascertain the impact of continuous passaging and oxygen tension on cellular integrity of hUC-MSCs, Annexin V/PI flow cytometry analysis revealed a significant, oxygen level-dependent elevation in apoptosis rates during serial passaging (P4-P16). Normoxic cultured high-passage(P16) cells exhibited profound apoptosis rate, whereas hypoxic cultures markedly attenuated both early and late apoptosis (Fig. 2A). Since 20%O₂ is sharply higher than the physiological concentration, it may cause oxidative stress and DNA damage [16]. We measured the levels of ROS and 8-OH guanine, a DNA oxidative damage marker. Our results showed that the ROS levels were higher in hUC-MSCs cultured under 20%O₂ normoxia versus 5%O₂ hypoxia and kept increasing during serial passaging (Fig. 2B). Accordingly, hypoxia substantially reduced 8-OH guanine level, particularly in high-passage(P16) cells compared with normoxia (hypoxia: 80.99 ± 2.83 ng/mL vs. normoxia 197.20 ± 17.02), thereby indicating the protective role of hypoxia against oxidative damage (Fig. 2C).

Fig. 2.

Hypoxia attenuates apoptosis and oxidative stress in high-passage hUC-MSCs. (A) Flow cytometry plots of Annexin V-APC/PI staining under normoxia (20%O₂) and hypoxia (5%O₂) at P4, P8, P12, and P16. Quadrants: Q1 (necrotic), Q2 (late apoptotic), Q3 (viable), Q4 (early apoptotic). Quantification of total apoptotic cells (Q2 + Q3). (B) The ROS levels measured by DCFH-DA mean fluorescence intensity (MFI). (C) 8-OHDG concentrations in the supernatant (DNA oxidative damage marker). N = 3 (per group); data are mean ± SEM. Significance vs. Normoxia group: ns = no significance, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Given that oxidative stress leads to damage of biomolecules such as DNA, proteins, and lipids, thereby contributing to cellular senescence, we next explored the impact of oxygen levels on cell senescence during serial passaging by detecting different hallmarks. The morphological alteration like enlarged and fattened cell body is the key feature of in vitro senescence, flow cytometric analysis of cell size showed that hUC-MSC volumes increased during passaging (P4-P18), as evidenced by a rightward shift on the forward scatter (FSC, indicating the cell size) axis [27]. Notably, 5%O₂ hypoxic conditions markedly suppressed cell size increase in high-passage(P12-P18) hUC-MSCs compared with 20%O₂ normoxia (Supplemental Fig. 2A).

Then, we detected senescence-associated β-galactosidase activity, a widely used marker for cellular aging. β-Galactosidase staining revealed that normoxic (20%O₂) cultures (P16) accumulated blue-green senescent cells markedly earlier and more than their hypoxic (5%O₂) counterparts, confirming the anti-aging advantage of low-oxygen expansion (Fig. 3A). Mouse embryonic fibroblasts (MEF) were used as a positive control for SA-β-gal staining.

Fig. 3.

Hypoxia mitigates senescence phenotypes in hUC-MSCs. (A) SA-β-gal staining (green-yellow: senescent cells). quantitative analysis of SA-β-gal positive area (normalized to total cell area, area%). Groups: Negative control (Neg), MEF, mouse embryonic fibroblasts (positive control). (B, C) Secretome analysis by ELISA: TGF-β, IL-6, IL-8, and VEGF, FGF concentrations in the supernatant. (D, E) mRNA expression of senescence markers (LMNB1, CDKN2A, CDKN1A) and telomerase (hTERT) by qRT-PCR. (F) Relative telomere length quantification. Data = mean ± SEM (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. Normoxia P16 group. ns = no significance. Scale bar indicates 50 μm

Senescence-Associated Secretory Phenotype (SASP) refers to the complicated and dynamic secretory profile of senescent cells, including a wide range of chemokines, cytokines, growth factors, and proteases. ELISA analysis revealed that the levels of SASP factorssuch as IL-6, IL-8, and TGF-β kept increasing in supernatants of hUC-MSCs cultured under 20%O₂ during serial passaging particularly in passage16, whereas hypoxia dramatically inhibited the secretion of these factors in high-passage hUC-MSCs (Fig. 3B). In contrast, hypoxia maintained growth factor secretion, including VEGF and FGF in high-passage hUC-MSCs compared to normoxic culture condition (Fig. 3C).

We next determined the expression of biomarkers of senescence including CDKN2A, CDKN1A and LMNB1. RT-qPCR analysis showed that hypoxia was associated with reduced expression of CDKN2A and CDKN1A, and preservation of LMNB1 expression across serial passaging compared to normoxic conditions (Fig. 3D).

Given the established link between replicative senescence and telomere attrition, we further measured telomerase expression and telomere length. Quantitative PCR analysis revealed significant upregulation of telomerase catalytic subunit hTERT (182.44 copy versus 0 copy) in hypoxic P10 hUC-MSCs compared to normoxic controls (Fig. 3E). Consistent with enhanced telomerase activity, telomere length was maintained at 3.19-fold higher levels under hypoxia in P10 cells and substantially preserved telomere length till P16 (1.17 folds) (Fig. 3F). These results demonstrate that hypoxia mitigates replicative senescence by preserving telomere integrity partially through hTERT-mediated mechanisms. These findings collectively demonstrate that 5%O₂ culture conditions are more effective in inhibiting cell size enlargement, suppressing the production of SASP factors, and delaying the processes of cellular growth and aging.

RNA sequencing (RNA-Seq) reveals hypoxia-delayed senescence through suppression of pro-inflammatory pathways

To elucidate the molecular mechanisms underlying the anti-senescence effects of hypoxic culture conditions on hUC-MSCs, RNA-Seq analysis was performed on low-passage and high-passage hUC-MSCs cultured under both 5%O₂ hypoxic and 20%O₂ normoxic conditions.

Comparative transcriptomic profiling of four experimental groups, low-passage(P4) and high-passage(P16) hUC-MSCs under hypoxia (5%O₂) versus normoxia (20%O₂) identified 11,825 differentially expressed genes (DEGs) as illustrated in Fig. 4A. Hierarchical clustering of DEGs revealed distinct transcriptional profiles between the groups. Notably, compared to high-passage(P16) hUC-MSCs cultured under 20%O₂ normoxic conditions, the expression of genes such as pro-aging mediators GPNMB and SQSTM1 were significantly downregulated, while rejuvenation-associated regulators like DUSP2 and HGF were significantly upregulated in hypoxic P16 cells (Fig. 4B, C). To validate the hypothesis that hypoxic culturing mitigates replicative senescence by modulating cell-cycle progression and suppressing the SASP, KEGG, Reactome, and GO enrichment analyses of the DEGs identified significant enrichment in pathways governing cell-cycle regulation, cellular senescence, and growth factor secretion (Fig. 4D–F). The protein–protein interaction (PPI) network further confirmed significant downregulation of proteins related to senescence-driving pathways, including downregulation of the cell cycle inhibitor CDKN2A, the NF-κB pathway activator SQSTM1, and the antioxidant-related protein SOD2. In contrast, cultivation under normal oxygen conditions promotes cellular senescence by inhibiting the cell cycle and upregulating oxidative stress pathways (Fig. 4G). FPKM quantification revealed significant suppression of pro-inflammatory cytokines such as IL-6, IFN, and TGF-β coupled with enhanced expression of VEGF and FGF, growth factors in hypoxic high-passage hUC-MSCs (Fig. 4H). Hypoxia delays senescence through transcriptional inhibition of pro-inflammatory and pro-aging pathways while promoting cell-cycle progression and regenerative factor secretion.

Fig. 4.

Transcriptomic profiling reveals hypoxia-regulated pathways in hUC-MSCs. (A) Venn diagram of overlapping gene expression between hypoxia (5%O₂) and normoxia (20%O₂) groups. (B) Heatmap of differentially expressed genes (DEGs) gene expression patterns. (C) Volcano plot of DEGs. (D) KEGG pathway analysis. (E) Reactome pathway analysis. (F) Gene Ontology (GO) enrichment analysis. (G) Protein–protein interaction (PPI) analysis. (H) Validation of key DEGs by FPKM value of genes. N = 3 (per group, biological replicates), data = mean ± SEM

Establishment of diabetic nephropathy mouse model and homing capacity of hUC-MSCs to damaged renal tissue

To obtain the in vivo evidence of the therapeutic potential of high-passage hUC-MSCs cultured under hypoxia, a type 2 diabetic nephropathy mouse model was established via HFD and STZ injection. Following STZ treatment, the DN group exhibited random blood glucose levels exceeding 16.7 mmol/L from the 11th week indicating successful establishment of the diabetic model. Then, followed by a 4-week period of tail vein injection of hUC-MSCs (1 × 10⁵ cells/week), the animals were euthanized, and samples were collected for further analysis (Fig. 5A). Notably, progressive body weight loss and was observed in DN mice post-disease induction, with a reduction from 35.18 g to 29.05 g at week 15, lower than control group(34.05 g), along with elevated blood glucose levels (Supplemental Fig. 3A, B). Furthermore, levels of Cr, BUN, and 24-h urine protein in DN mice significantly increased at week 11 and week 15 compared to the control group, with urine protein exceeding 20 mg/24 h at the 11th week (Supplemental Fig. 3C–E). In addition, urine protein levels exceeded 20 mg/24 h at week 11, significantly higher than threefold that of the control group. Strikingly, administration of hUC-MSCs(P6, cultured under normoxia) at week 11 effectively attenuated these elevations (Supplemental Fig. 3C–E).

Fig. 5.

Hypoxic high-passage hUC-MSCs ameliorate renal injury in DN mice. (A) Experimental design and timeline of DN model establishment by Figdraw. (B-D) Level changes of renal function markers after transplantation of hUC-MSCs: (B)Serum creatinine (Cr), (C) Blood urea nitrogen (BUN), (D) 24-h urinary protein. The transplantation of high-passage hUC-MSCs cultured under 5%O₂ hypoxia resulted in better preservation of kidney injury markers. (E) H&E staining of kidney tissue. (F) Electron microscopy examination of kidney tissue. (G) Masson trichrome staining of kidney tissue quantifying renal fibrosis (blue: collagen deposition). Groups: Control (healthy), DN (untreated), Hypoxia P4/P16 (5%O₂-cultured hUC-MSCs), Normoxia P4/P16 (20%O₂-cultured hUC-MSCs). N = 4 (per group), data = mean ± SEM. *p < 0.05, ***p < 0.001 vs. Normoxia P16 group. G:glomerulus. Scale bars: 200 μm (E), 1 μm (F), 100 μm (G)

To track in vivo biodistribution, CM-Dil-labeled hUC-MSCs were intravenously infused into DN mice and homing dynamics were monitored via in vivo small animal imaging at different time points within one week (Supplemental Fig. 4A). Systemic biodistribution analysis revealed peak fluorescence intensity at 24 h post-infusion and then gradually decreased (Supplemental Fig. 4B). Intriguingly, DN kidneys exhibited higher CM-Dil signal intensity versus non-diabetic controls (11.73 ± 0.37 vs. 9.81 ± 0.80 radiance units), confirming renal-targeted homing of hUC-MSCs to injured kidney (Supplemental Fig. 4C).

Hypoxic culture preserves therapeutic efficacy of high-passage hUC-MSCs in diabetic nephropathy mice: Improved renal function and pathological injury

As our data found the protective roles of hypoxia against cellular senescence in vitro, we further investigated whether hypoxia was able to maintain the therapeutic potential of hUC-MSCs across serial passaging using diabetic nephropathy mouse model. Four group of cells, low-passage(P6) and high-passage(P16) hUC-MSCs cultured under hypoxia (5%O₂) and normoxia (20%O₂), were administered via tail vein injection from week 11 as shown in Fig. 5A, once per week for four weeks. The blood tests results showed that transplantation of both hypoxic and normoxic low-passage(P6) hUC-MSCs significantly ameliorated renal dysfunction in DN mice compared to control group, no difference of therapeutic efficacy was found between these two types of cell. In contrast, the therapeutic efficacy of high-passage(P16) hUC-MSCs varied depending on O₂ tension. Administration of hypoxic P16 hUC-MSCs markedly decreased the levels of Cr, BUN, and 24-h urine protein in DN mice. MSC-treated DN mice exhibited 34.63% lower Cr (28.80 ± 2.60 vs. 44.06 ± 7.37 μmol/L), 45.12% reduced BUN (7.23 ± 0.60 vs. 13.18 ± 1.11 mmol/L), and 81.57% decreased 24-h proteinuria (4.61 ± 1.79 vs. 25.02 ± 1.19 mg/24 h) compared to untreated DN controls, while normoxic P16 hUC-MSCs treatment showed no therapeutic effect (Fig. 5B–D).

To compare and evaluate how the oxygen milieu during passaging influences the reparative potency of hUC-MSCs, we subjected murine kidneys to histological and ultrastructural interrogation. HE and Masson’s trichrome staining revealed extensive inflammatory-cell infiltration, glomerulor enlargement, hyaline arteriolar change, exuberant collagen deposition within glomeruli and the tubulointerstitium of DN kidneys (Fig. 5E, G). These lesions were markedly reduced after treatment with low-passage(P4) and hypoxic high-passage(P16) hUC-MSCs, whereas infusion of normoxic high-passage(P16) hUC-MSCs did not show much improvement.

We next investigate and compare the tissue repair capacity of different hUC-MSCs at the ultrastructural level, TEM disclosed pronounced thickening of the glomerular basement membrane (GBM) and widespread effacement of podocyte foot processes in DN kidneys. Mice receiving hypoxic high-passage hUC-MSCs exhibited near-normal GBM thickness and restoration of a well-defined interdigitating foot-process architecture. Conversely, DN kidneys treated with 20%O₂-expanded high-passage hUC-MSCs still displayed segmental GBM redundancy and incomplete podocyte repair or protection (Fig. 5F). These data establish that hypoxic culture conditions preserve superior anti-inflammatory, anti-fibrotic, and glomerular-protective properties of high-passage hUC-MSCs, enhancing their therapeutic potential in DN. In summary, the therapeutic efficacy of these cells showed significant oxygen-dependence and passage-dependency during expansion.

Hypoxia sustains potency of high-passage hUC-MSCs potentiating anti-fibrotic mechanisms in DN kidneys

Since our Masson’s staining found hUC-MSCs exuberant collagen deposition within glomeruli and the tubulointerstitium of DN kidneys, to dissect how the culture oxygen tension influences the anti-fibrotic capacity of serial passaged hUC-MSCs, we further profiled key fibrogenic markers. Immunohistochemistry revealed robust up-regulation of α-SMA and TGF-β in DN kidneys (Fig. 6A–D). High-passage hUC-MSCs expanded at 5%O₂ markedly attenuated this expression, whereas cells cultured at 20%O₂ conferred only modest suppression. Western blot confirmed these findings: treatment with 5%O₂-expanded MSCs significantly reduced renal Collagen III and α-SMA protein levels relative to their 20%O₂ counterparts (Fig. 6E–G). Thus, hypoxia preserves the intrinsic anti-fibrotic efficacy of high-passage hUC-MSCs, limiting myofibroblast activation and extracellular-matrix deposition in diabetic kidneys.

Fig. 6.

Hypoxic high-passage hUC-MSCs attenuate renal fibrosis in DN. (A) Immunohistochemical staining (IHC) of α-SMA in kidney tissues (groups: Control, DN, HypoxiaP4, NormoxiaP4, HypoxiaP16, NormoxiaP16). (B) Quantification of α-SMA positive area (%Area). (C) IHC of TGF-β in kidney tissues (same groups as A). (D) Quantification of TGF-β positive area (%Area). (E) Western blot analysis of fibrosis markers: Collagen III and α-SMA protein. (F) Quantification of Collagen III protein expression (normalized to β-tublin). (G) Quantification of α-SMA protein expression (normalized to β-tublin). N = 4 (per group), data = mean ± SEM. *p < 0.05, **p < 0.01 vs. NormoxiaP16 group. Groups: Control (healthy), DN (untreated diabetic nephropathy), Hypoxia P4/P16 (DN + 5%O₂-cultured hUC-MSCs), Normoxia P4/P16 (DN + 20%O₂-cultured hUC-MSCs). Scale bars: 100 μm (A)

RNA-Seq reveals hypoxia-preserved renoprotective capacity of high-passage hUC-MSCs via transcriptional repression of inflammatory and fibrosis pathways

To dissect how oxygen tension during expansion influences the in vivo tissue repair capacity and possible mechanisms of high-passage hUC-MSCs, we profiled whole-kidney transcriptomes after four-week transplantation of high-passage(P16) hUC-MSCs cultured under 5% or 20%O₂. RNA-Seq of whole-kidney tissue identified 28,753 differentially expressed genes (DEGs) between hypoxic and normoxic hUC-MSCs recipients (fpkm > 1) (Fig. 7A). Hierarchical clustering highlighted distinct transcriptomic signatures in hypoxic P16 hUC-MSCs recipients diverged sharply from their normoxic counterparts: significant downregulation of collagen proliferation and inflammatory fibrosis genes like Col1a1 and Tnfsf10, whereas selective upregulation of cell- homeostasis related genes Adm2 and Lsm5, indicating a metabolic shift from proliferative to homeostatic programs (Fig. 7B, C). KEGG, Reactome, and GO functional enrichment analyses converged on NF-κB signaling pathway, ECM-receptor interaction, and Extracellular matrix organization as the most prominently enriched pathways (Fig. 7D–F). The PPI network further indicated that P16 kidneys cultured under hypoxic conditions exhibited significant suppression of the inflammatory response pathway NF-κB, along with marked inhibition of classical pro-fibrotic responses, including downregulation of cell adhesion proteins Cx3cl1 and Cd44, excessive extracellular matrix deposition proteins Col1a1 and Col6a3, pro-fibrotic TGF-β signaling pathway-associated proteins Itgav and Fbln1, as well as NF-κB pathway-related Bcl2 proteins. Although residual transcripts for Tgfb and Col3a1 remained detectable, their abundance was significantly lower than in kidneys treated with normoxia-expanded hUC-MSCs (Fig. 7G, H). Our data indicate that hypoxia preserves the renoprotective potency of high-passage hUC-MSCs by transcriptionally repressing fibrosis- and inflammation- associated gene networks while reinforcing metabolic and structural homeostasis within the injured diabetic kidney.

Fig. 7.

Transcriptomic profiling of renal tissues reveals potential molecular mechanisms underlying MSC therapy efficacy. (A) Venn diagram of overlapping gene expression between different groups: NC group (healthy control), DN group (untreated), Hypoxia group (DN + 5%O₂-cultured hUC-MSCs), Normoxia group (DN + 20%O₂-cultured hUC-MSCs). (B) Heatmap of differentially expressed genes (DEGs) gene expression patterns. (C) Volcano plot of DEGs. (D) KEGG pathway analysis. (E) Reactome pathway analysis. (F) Gene Ontology (GO) enrichment analysis. (G) Protein–protein interaction (PPI) analysis. (H) Validation of key DEGs by FPKM value of genes. N = 3 (per group, biological replicates), data = mean ± SEM

Discussion

Mesenchymal stem cells derived from various sources hold transformative potential in the field of regenerative medicine, tissue repair, and basic therapeutic strategies due to their unique biological properties [28]. As we mentioned above, In vitro expansion of MSCs is the critically essential prerequisite for their clinical application, and the quality of cells is the key determinant of therapeutic effectiveness. It is therefore of great significance optimizing expansion protocols for MSCs to ensure the cell quality and quantity. Oxygen is an essential nutrient and plays a crucial role in cellular aerobic metabolism, particularly the tricarboxylic acid cycle [29]. Thus, oxygen tension is also a vital environmental parameter in cell culture and an important consideration for optimizing cell culture protocols, directly influencing cell viability and function.

Previous studies have largely highlighted the beneficial effects of physiological hypoxic culture conditions (1–5%O₂) over conventional atmospheric oxygen levels (20%O₂) on MSCs [30–33]. Yet, rigorous data elucidating the impact of hypoxia on MSC senescence and biological activity during long-term serial passaging remain limited, particularly in vivo evidence is still absent. Here, we systemically evaluated and compared the impacts of different oxygen tensions on hUC-MSC cellular senescence and functions during serial passaging, both in vitro and in vivo. Our data confirm that culturing under hypoxic condition (5%O₂) effectively delays replicative senescence and maintains the bioactivity of hUC-MSCs during serial passaging in vitro. Importantly, for the first time, we provided the in vivo data supporting that hypoxic culture conditions preserve the therapeutic efficacy of hUC-MSCs after continuous passaging till high passage numbers making use of DN mouse model.

In 1961, Hayflick and Moorhead reported limited proliferative capacity of primary human fibroblasts in vitro which was the first formal description of cellular senescence [34]. Following studies demonstrate that even stem cells including MSCs also undergoes cellular senescence during in vitro expansion in spite of their self-renewal capacity. Exogenous and endogenous factors that induce cell damage or stress—including DNA damage, dysfunctional telomeres, and genome instability—contribute to cellular senescence. Oxidative stress from environmental oxygen levels is one of the key contributing factors. Our results showed that ROS levels were higher under 5%O₂ hypoxia than 20%O₂ normoxia and increased with serial passaging. However, hypoxia significantly reduced oxidative DNA damage, especially in high-passage(P16) cells. Consequently, the detailed detection of cell senescence hallmarks such as cell enlargement [35], β-galactosidase activity [36], secretion of SASP [37, 38], and biomarkers of senescence (CDKN2A, CDKN1A and LMNB1) [39] confirmed that hypoxia effectively delays replicative senescence of hUC-MSCs during in vitro serial passaging.

Given that cellular senescence entails permanent growth arrest, we further examined cell cycle progression and proliferation in our model. During continuous passaging of hUC-MSCs under either 5%O₂ or 20%O₂ concentrations, a key finding was that hypoxic culture significantly enhanced cell proliferation across both low- and high-passage cells compared to normoxia, while a marked decline in proliferative capacity was observed predominantly in normoxic hUC-MSCs during passaging. Accordingly, hypoxic high-passage hUC-MSCs also exhibited a marked reduced G0/G1-phase cell cycle arrest and significantly lower apoptotic rate relative to their normoxic counterparts. Most previous studies reported similar findings, but only a few documented inhibition of cell growth in hypoxia [40].

Telomeres are highly conserved short tandem repeats located at the end of chromosomes in eukaryotic cells and play key roles maintaining genomic stability and integrity. Any factor causing telomere attrition contributes to senescence and telomere shortening also become one of the features of cell senescence [41].

Telomeric DNA is synthesized by telomerase, a specialized reverse transcriptase, but not replicative DNA polymerases [42]. Accordingly, most somatic cells and stem cells with too low level or without telomerase expression will undergo gradual telomere shortening after several rounds of cell division, which induces genomic instability and further leads to cell death or replicative senescence [43].

Consistent with our observation that 5%O₂ hypoxic culture effectively delays replicative senescence in serially passaged hUC-MSCs, we also found that hypoxia significantly attenuated telomere shortening compared to normoxia, even at high passage numbers. Similar results were reported by Estrada et al. using 3%O₂[44]. This effect may be attributed to the elevated hTERT expression level (Fig. 3E) under hypoxic condition, which could partially compensate for telomere attrition during continuous passaging—despite the fact that hTERT mRNA levels remain relatively low and even have previously been reported as undetectable in MSCs [45]. Mechanistically, the hypoxia-induced upregulation of hTERT may be mediated by HIF-1α, which has been reported to directly transactivate the hTERT promoter via binding to two hypoxia response elements located in its proximal promoter region [46]. Such HIF-1α-mediated transcriptional regulation provides a plausible mechanistic explanation for our observation that hypoxic culture increases hTERT mRNA expression in hUC-MSCs. Of note, the regulation of hTERT by hypoxia appears to be multilayered. In addition to such transcriptional activation, hypoxia has also been reported to promote alternative splicing of hTERT pre-mRNA toward the active variant [47]. Such post-transcriptional regulation may further contribute to the increased hTERT mRNA expression observed under hypoxic conditions.

Additionally, RNA sequencing analysis of low-passage(P4) and high-passage(P16) hUC-MSCs cultured under 5%O₂ and 20%O₂ respectively, further supported our findings. In high-passage hUC-MSCs maintained under 5%O₂, mRNA expression of pro-inflammatory factors such as IL-6, IFN, and TGF-β was reduced, while the mRNA expression of growth factors associated with anti-inflammatory and anti-fibrotic effects (VEGF, FGF, HGF, PDGF) was significantly increased. These findings align with previous reports confirming the roles of these factors in anti-fibrotic and anti-inflammatory processes [31, 48, 49]. Moreover, hypoxic high-passage(P16) hUC-MSCs exhibited downregulation of genes associated with oxidative stress (TP53,ICAM1), followed by a decrease in the stimulation of antioxidant genes (SESN2,GSR,SOD2), underscoring a synergistic interaction between oxidative stress and inflammatory responses [50–54]. Mechanistically, it is well established that ROS, the key mediators of oxidative stress, can stimulate the production of pro-inflammatory cytokines including TNF-α and IL-6, which further amplify ROS production, thereby forming a self-reinforcing vicious cycle that sustains chronic inflammation [55, 56]. Together, our results demonstrate that hypoxia effectively interrupts this positive feedback loop between oxidative stress and inflammatory responses.

Then we hypothesized that the 5%O₂ hypoxic environment may suppress the production of aging-associated secretory factors via a paracrine mechanism, thereby reducing oxidative stress-induced damage and delaying the emergence of cellular senescence phenotypes, while preserving cellular repair activity. Gene Set Enrichment Analysis (KEGG, Reactome) on differentially expressed genes from RNA sequencing revealed that hypoxic high-passage hUC-MSCs downregulated multiple pathways promoting inflammatory responses, including NF-κB, interleukin, SASP signaling and TGF-β signaling. Concurrently, pathways related to oxidative phosphorylation, the cell cycle, and cellular aging were also suppressed. These pathways interact and exhibit positive regulatory effects on each other. Previous studies indicate that they regulate the expression of pro-inflammatory factors and play key roles in mediating inflammatory responses [57, 58]. Among these interconnected signaling pathways, the crosstalk between TGF-β and NF-κB has been well documented. Accumulating evidence indicates that TGF-β interacts with NF-κB to synergistically drive renal inflammation and fibrotic progression [59]. The concurrent downregulation of both pathways under hypoxia therefore suggests that low oxygen tension attenuates this synergistic crosstalk, which is consistent with our observation of reduced inflammatory and fibrotic gene expression in hypoxic hUC-MSCs.

Thus far, we have systemically evaluated and compared the effects of hypoxic vs normoxic conditions on hUC-MSCs during prolonged in vitro passaging. In agreement with most previous reports, our results indicate that hypoxic culture effectively delays cellular senescence and preserves biological functions such as growth factor secretion of MSCs. However, all these findings are obtained from in vitro models, and the robust in vivo evidence supporting the maintenance of therapeutic efficacy in hypoxic-cultured MSCs after extended passaging remains lacking. To address this gap, we performed in vivo transplantation experiments using a mouse model of type 2 diabetic nephropathy. DN is a severe and progressive complication of diabetes with limited therapeutic options, representing one of the leading causes of end-stage renal disease and mortality [60]. The STZ combined with HFD model was selected based on the following rationale: STZ induces pancreatic β-cell apoptosis and consequent hyperglycemia, whereas HFD promotes insulin resistance and metabolic disorders. As summarized in recent reviews, this combined strategy reliably recapitulates the metabolic features of human type 2 diabetes mellitus [61], and has been widely validated for establishing stable diabetic kidney disease models in mice with the capacity to mimic key pathological features including albuminuria, glomerulosclerosis, and tubulointerstitial fibrosis [62]. Regarding MSC-based therapy for DN, most previous studies have focused on bone marrow-derived MSCs (BM-MSCs) or adipose-derived MSCs (AD-MSCs), which have been shown to improve glycemic control, ameliorate proteinuria, and suppress renal inflammation [63]. However, the therapeutic potential of hUC-MSCs, particularly those cultured under different oxygen tensions during in vitro expansion, remains less explored in the context of DN. Based on our in vitro findings that hypoxic culture preserves the anti-senescent and paracrine functions of hUC-MSCs, we further investigated whether these beneficial properties could translate into enhanced reno-protective efficacy in vivo using the STZ/HFD-induced DN model.

We first detected the biodistribution of CM-Dil-labeled hUC-MSCs by in vivo small animal imaging after intravenous infusion, and our results (Fig. S4B) revealed fluorescent signals primarily in major organs, peaking at 24 h post-injection and gradually declining thereafter, with faint fluorescence still detectable in the kidneys after one week. Importantly, the DN group exhibited higher fluorescence intensity in the kidneys compared to normal control, indicating enhanced homing and retention of hUC-MSCs to injured renal tissue. This is consistent with previous reports that MSCs can migrate to sites of damage and participate in repair processes via targeted differentiation or immunomodulation [64, 65].

DN involves damage to key components of the glomerular filtration barrier-podocytes, glomerular basement membrane, and endothelial cells-resulting in functional impairment characterized by elevated blood creatinine, urea nitrogen, and proteinuria [66, 67].

We found that transplantation of both hypoxic and normoxic low-passage(P4) hUC-MSCs led to significant and similar improvement of kidney function. Strikingly, hypoxic high-passage(P16) hUC-MSCs provided comparable improvement to low-passage hUC-MSCs, attenuating podocyte injury, interstitial fibrosis and other pathological changes, improving serum renal damage indicators and reducing proteinuria levels. By contrast, normoxic high-passage (P16) hUC-MSCs showed almost no therapeutic effect.

In DN, elevated cytokine levels such as IL-6, IL-8, TNF-α, and TGF-β drive inflammatory cell infiltration and interstitial fibrosis. TGF-β, in particular, induces early renal hypertrophy by disrupting the cell cycle and promotes interstitial fibrosis via epithelial-to-myofibroblast transition [68, 69]. Our data demonstrate that hypoxic but not normoxic high-passage hUC-MSCs significantly reduced TGF-β levels and downregulated the expression of myofibroblast markers α-SMA and type III collagen in DN mice renal tissues.

In renal RNA sequencing analysis, compared to the 20%O₂ group, treatment with high-passage hUC-MSCs cultured under 5%O₂ condition led to significant downregulation of inflammatory factors (e.g., Tnfsf10) and fibrosis-related markers (e.g., Tgf-β, Col3a1). Concurrently, key pathways mediating renal inflammation and fibrosis, TGF-β and Nf-κB were suppressed, along with reduced expression of cytoskeletal proteins and fibronectin.

Collectively, our in vivo results demonstrate that hypoxic culture effectively preserves therapeutic efficacy of high-passage hUC-MSCs. Their potent anti-inflammatory, anti-fibrotic, and glomerular-protective functions appear to be primarily attributed to paracrine mechanisms. Further studies are warranted to elucidate the specific roles and mechanisms of endocrine and paracrine actions mediated by these factors in the context of DN under hypoxic conditions.

Conclusion

Our study demonstrates that hypoxic culture conditions during in vitro expansion significantly delay cellular senescence and maintain the biological and reparative functions of hUC-MSCs compared to normoxic culture. This work is the first provision of in vivo evidence confirming that high-passage hUC-MSCs cultured under hypoxia retain considerable therapeutic efficacy, supporting their potential for clinical application even after extended serial passaging. These findings offer novel insights for improving the large-scale production of functional hUC-MSCs as off-the-shelf regenerative therapeutics. Further studies are needed to elucidate the precise anti-inflammatory and anti-apoptotic mechanisms through which hypoxia mitigates cellular aging.

Abbreviation

MSCs

Mesenchymal stem cells

hUC-MSs

Human umbilical cord derived-mesenchymal stem cells

MEF

mouse embryonic fibroblasts

DN

Diabetic nephropathy

ROS

Reactive oxygen species

SA-β-gal

Senescence associated β-galactosidase

SPF

Specific pathogen-free

HFD

High-fat diet

STZ

Streptozotocin

HE

Hematoxylin and eosin

PVDF

Polyvinylidene difluoride

SASP

Senescence-associated secretory phenotype

RNA-Seq

RNA sequencing

PPI

Protein–protein interaction

Cr

Serum creatinine

BUN

Blood urea nitrogen

GBM

Glomerular basement membrane

Author contributions

CJ: designed the study, conducted experiments, interpreted the results and wrote the draft of the manuscript. ZL, TZ, LH, YL, JL, XL, YZ, BC, MZ, LW: contributed to resources provision, animal experimentation, data curation, and manuscript review. XT Z, CR and XZ: interpreted the results, revised the manuscript and approved the final version. All authors critically reviewed the manuscript and approved the final version.

Funding

Funding was provided by the Natural Science Foundation of Guangdong Province (Grant No. 2025A1515012353) and the 2022 Science and Technology Innovation Strategy Special Project of Guangdong Province (Grant No. 220927127645167).

Availability of data and material

All data generated or analyzed in this study are included in this published article. Any other data will be available on reasonable request from corresponding author.

Declarations

Ethics approval and consent to participate

Animal experiments were approved by the Institutional Animal Care and Use Committee of Shantou University Medical College (No. SUMC2021-267). The approval date for the project titled “The efficacy of human umbilical cord mesenchymal stem cells prepared under different oxygen concentrations in treating diabetic nephropathy in mice” was 29 Jun. 2021. This study involving human umbilical cord tissues was conducted in accordance with the Declaration of Helsinki. The study protocol, including the collection and use of umbilical cord samples, was reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Shantou University Medical College (No. B-2022–249). The approval date for the project titled “Mechanism of Action and Translational Study on Injectable Thermosensitive Hydrogels Loaded with Walton's Jelly-derived MSCs Extract for Regulating the Microenvironment of Hard-to-Heal Wounds” was 04 Oct. 2022. Informed consent was obtained from all participating mothers prior to the collection of umbilical cord tissues.

Consent for publication

Not applicable.

Competing interests

The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.