Background

Tendons are the essential connective tissues that tether skeletal muscles to bones and transmit force. The tendon is easily torn because it is subjected to a high tensile strain. Tendon injuries are among the most common musculoskeletal disorders, affecting more than 32 million people in the United States [1]. Fibrotic scarring is inevitable during healing, resulting in the disruption of tendon matrix continuity and disability [2]. The existing effective treatment options for tendon injuries are limited.

One reason for the inferiority of healed tendons is the limited number of resident cells in the tendon tissues [3,4,5,6,7]. Therefore, cell-based therapies for tendon injury have attracted increasing attention. Because of their self-renewal and differentiation abilities, stem cells are used in multiple tissue regeneration and disease treatments [8,9,10]. The human umbilical cord is an abundant source of multipotent stem cells, and human umbilical cord mesenchymal stem cells (hUCMSCs) exhibit multiple lineage differentiation potential and immune modulation capacity [11]. hUCMSCs possess many advantages, such as fewer ethical issues, a painless collection process, and a lack of immunity compared to other mesenchymal stem cells (MSCs) [12]. hUCMSCs have general benefits for tissue regeneration in the bone, cartilage, uterus, and brain [13,14,15]. Jo et al. have reported that hUCMSCs induce rotator cuff tendon regeneration in rat models [16]. Lee et al. observed partial healing in a rabbit model of tendon rupture treated with hUCMSCs [17]. Therefore, hUCMSCs therapy is a potential treatment modality for tendon injuries.

Excessive reactive oxygen species (ROS) are crucial mediators of multiple pathological processes that induce cellular senescence, apoptosis, and dysfunction [18,19,20]. The therapeutic activity of MSCs is permanently impaired in injured tissues owing to abnormally high oxidative stress [21]. In tendons, both acute and chronic injuries can promote ROS generation, resulting in only a slight improvement in some MSC-related tendon healing studies [1, 22, 23]. Thus, effective antioxidant therapies must be developed in conjunction with stem cell therapies to enhance the efficacy of hUCMSCs.

Cerium is a lanthanide metal element that exhibits repeatable reducibility because of the conversion of cerium ions between trivalent and tetravalent ions. Recently, cerium oxide nanoparticles (CeONPs) have emerged as antioxidants owing to their superoxide dismutase, catalase, and peroxidase activities [24]. Consequently, CeONPs have been widely used in ROS-associated diseases. Du et al. suggested that injection with atorvastatin-loaded CeONPs mitigates acute kidney injury [25]. Lee et al. confirmed that CeONPs can protect against ischemic stroke by scavenging ROS [26]. Similar effects have been demonstrated in liver diseases and glaucoma [27,28,29]. However, a CeONPs-based defense strategy against oxidative injury during cellular transplantation and tendon injury needs to be developed.

In the present study, we synthesized methyl polyethylene glycol 2000 distearylphosphatidylethanolamine (mPEG2k-DSPE)-packed CeNOPs (PEG-CeONPs) and constructed PEG-CeONPs-carrying hUCMSCs. The protective effects of the PEG-CeONPs against oxidative stress were evaluated in vitro. We also assessed the therapeutic effects of the PEG-CeONPs-carrying hUCMSCs in a rat model of tendon injury. This study aimed to develop a novel stem cell modification strategy for tendon injury.

Methods

Cell culture

Clinical-grade hUCMSCs were obtained from Wingor Biotechnology Co., Ltd. (Shenzhen, China) and authenticated using cell surface markers and a trilineage differentiation assay. Briefly, 107 hUCMSCs were subjected to analysis for surface markers CD34 (E-AB-F1143D, Elabscience), CD44 (E-AB-F1100D, Elabscience), CD45 (E-AB-F1137D, Elabscience), CD29 (E-AB-F1049D, Elabscience), CD90 (E-AB-F1167D, Elabscience), and CD105 (E-AB-F1310D, Elabscience) by flow cytometry following the manufacturer’s protocol. The lipogenic differentiation ability of hUCMSCs was evaluated using an hUCMSC lipid-induced differentiation medium (PD-019, Procell, Wuhan, China). After lipid droplets appeared in hUCMSCs cultured in lipogenic differentiation medium, they were stained using Oil Red O solution. For the osteogenic differentiation assay, hUCMSCs were cultured in an osteogenic induction differentiation medium (PD-017, Procell, Wuhan, China). After 4 weeks, hUCMSCs were stained with Alizarin Red. The chondrogenic differentiation ability of hUCMSCs was measured using a chondrogenic induction differentiation medium (PD-018, Procell, Wuhan, China) following the manufacturer’s protocol. After 4 weeks, the hUCMSCs cytospheres were fixed, embedded, sliced, and stained with Alcian Blue staining solution and Nuclear Fast Red. Images were captured using an inverted microscope.

Synthesis and characterization of CeONPs

CeONPs were synthesized using previously reported methods with minor modifications [30]. Briefly, Ce(NO3)3·6H2O (1.736 g, 10294-41-4, Aladdin, Shanghai, China) and oleylamine (3.208 g, 112-90-3, Aladdin) were dispersed in 20 g 1-octadecene (112-88-9, Aladdin) and stirred for 2 h. After heating under a vacuum at 80 °C for 1 h, the mixed solution was heated and maintained at 260 °C for 2 h in an argon atmosphere. CeONPs were washed with cyclohexane and anhydrous ethanol at least three times to collect CeONPs. mPEG2k-DSPE (S28722, Yuanye Bio, Shanghai, China) was used to transfer the CeONPs into the aqueous phase through evaporation in an ultrasonic water bath. Finally, the PEG-CeONPs were purified by dialysis (MW cutoff = 8–14 kDa). An inductively coupled plasma optical emission spectrometer was used to calculate the molality of PEG-CeONPs. For cyanine3 (Cy3)-labeled PEG-CeONPs, mPEG2k-DSPE was replaced with Cy3-PEG2k-DSPE and the above procedure was repeated. The morphology and energy dispersive spectrometer of PEG-CeONPs were detected using an JEM-2100 TEM (JEOL, Tokyo, Japan). FTIR spectroscopy using a FTIR5700 spectrometer (Thermo Electron Corporation, MA, USA). The X-ray photoelectron spectra (XPS) were obtained via an ESCALAB 250 Xi (Thermo Electron Corporation, MA, USA) XPS system.

Antioxidant capacity of CeONPs

The antioxidant capacity of the CeONPs was assessed using multiple assays according to the manufacturer’s protocol. The free radical scavenging effects of CeONPs were tested using a hydroxyl free radical assay kit (A018-1-1, Nanjingjiancheng, Nanjing, China) and a 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical scavenging capacity assay kit (A153-1-1, Nanjingjiancheng). Superoxide dismutase (SOD) assay kits (A001-3-2, Nanjingjiancheng) and catalase (CAT) assay kits (A007-1-1, Nanjingjiancheng) were used to evaluate the SOD and CAT enzyme activities of CeONPs, respectively.

Intervention for hUCMSCs

For constructing the H2O2-elicited oxidative injury model, hUCMSCs were treated with 500 µM H2O2 (7722-84-1, Aladdin) for 24 h. hUCMSCs were cultured with 50 µg/mL PEG-CeONPs for 24 h. For pathway activator treatment, hUCMSCs were cultured with 10 ng/mL human recombinant TNFα and 50 µM dehydrocrenatine (DE) for 24 h [31].

Concentration of cellular ROS

The concentration of cellular ROS was assessed using dichlorodihydrofluorescein diacetate (DCFH-DA, S0033S, Beyotime, Shanghai, China) and dihydroethidium (DHE, 50102ES02, Yeason, Shanghai, China) fluorescent probes following the manufacturer’s protocol. hUCMSCs were washed with PBS and cultured in a serum-free medium supplemented with 10 µM DCFH-DA or 10 µM DHE for 30 min. After washing out the free probe with PBS, images were captured using an inverted fluorescence microscope. The mean fluorescence intensity was measured using ImageJ software (version: 1.8.0).

Mitochondrial membrane potential

The membrane potential of the isolated mitochondria was measured using the JC-1 Mitochondrial Membrane Potential Assay Kit (G1515, Servicebio) following to the manufacturer’s instructions. After washing with PBS, hUCMSCs were incubated with the JC-1 staining solution at 37 °C for 20 min. The hUCMSCs were then washed three times with JC-1 staining buffer. The images were visualized using an inverted fluorescence microscope and analyzed using ImageJ software.

Western blotting

Total proteins from hUCMSCs were extracted using RIPA buffer (Servicebio, G2008) supplemented with PMSF (G2008, Servicebio), phosphatase inhibitors (G2007, Servicebio), and 50×cocktail (G2006, Servicebio). Protein concentrations were measured using the BCA protein assay kit (G2026, Servicebio). Protein samples were reduced in SDS sample buffer and separated by 12% SDS-PAGE. After transferring onto PVDF membranes and blocking with 5% skimmed milk, the membranes were incubated with primary antibodies against-β-actin (1:2000, GB11001, Servicebio), BAX (1:2000, 50599-2, Proteintech, Wuhan, China), BCL2 (1:1000, CPA3144, Cohesion, London, UK), P16 (1:1000, A0262, Abclonal, Wuhan, China), P21 (1:1000, A1483, Abclonal), P65 (1:5000, T55034, Abmart, Shanghai, China), P-P65 (1:1000, TP56372, Abmart), IkBα (1:500, TA5002, Abmart), P-IkBα (1:500, TA2002, Abmart), IL-1β (1:1000, A16288, Abclonal), IL-6 (1:1000, CPP1813, Cohesion), TNFα (1:2000, A0277, Abclonal), JNK (1:1000, A0288, Abclonal), P-JNK (1:1000, AP0631, Abclonal), P38 (1:1000, A14401, Abclonal), and P-P38 (1:500, AP0526, Abclonal) overnight at 4 °C. The membranes were washed three times with PBST and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (1:5000, GB23303, Servicebio) for 1 h. The ECL substrate (SQ101, Epizyme Biotech, Shanghai, China) was used to visualize the protein bands on a ChemiDoc Touch (Bio-Rad, Hercules, CA, America), and the semi-quantitative analysis of images was conducted through ImageJ software.

Annexin V/propidium iodide (PI) staining

The percentage of apoptotic hUCMSCs was assessed by Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) staining (40302ES20, Yeason). Briefly, the hUCMSCs were collected and incubated with 5 µL Annexin V-FITC and 10 µL PI staining solution for 15 min, following the manufacturer’s protocol. Apoptosis rates were evaluated by flow cytometry (Beckman Coulter, USA).

Colony formation assay

hUCMSCs or PEG-CeONPs-carrying hUCMSCs were exposed to H2O2 and trypsinized. Five hundred hUCMSCs were replanted in six-well plates and cultured for 3 weeks. After fixing with 4% paraformaldehyde, the plates were stained with a crystal violet solution. ImageJ software was used to test the colony expansion ability.

Immunofluorescence staining

hUCMSCs were fixed using a 2% paraformaldehyde solution for 15 min, followed by permeabilization using 0.5% Triton X-100 for 20 min at room temperature. The hUCMSCs were washed with PBS and blocked with 5% goat serum. After incubating with primary antibody Ki67 (1:50, A11390, Abclonal) at 4 °C overnight and fluorescent secondary antibody (1:250 GB21303, Servicebio) at 37 °C for 1 h, cells were stained with DAPI, and the images were analyzed under a fluorescent microscope.

Real-time quantitative PCR (RT-qPCR)

Briefly, 1000 ng of RNA was isolated using RNA isolation kits (R0032, Beyotime) and reverse transcribed using a double-stranded cDNA synthesis kit (G3331, Servicebio). The mixture incubation conditions were: 5 min at 25 °C, 55 °C for 15 min, and 85 °C for 5 s. RT-qPCR was performed in triplicate using the Universal SYBR green fast qPCR mix kit (G3320, Servicebio) on LightCycler® 480 Software (Roche, Swiss Confederation). The primer sequences used in this research were as follows: β-actin, forward primer CACCCAGCACAATGAAGATCAAGAT, reverse primer CCAGTTTTTAAATCCTGAGTCAAGC; P16, forward primer CTGCCCAACGCACCGAATAG, reverse primer AGCTCCTCAGCCAGGTCCAC; P21, forward primer ACCACTGGAGGGTGACTTC, reverse primer CGGCGTTTGGAGTGGTAG; IL-1β, forward primer GTGCACGATGCACCTGTACG, reverse primer ACGGGCATGTTTTCTGCTTG; IL-6, forward primer AAGCAGCAAAGAGGCACTGG, reverse primer TGGGTCAGGGGTGGTTATTG; TNFα, forward primer GAACCCCGAGTGACAAGCCT, reverse primer CCCTTGAAGAGGACCTGGGA.

In vivo study

This study design was approved by the Laboratory Animal Welfare and Ethics Committee of the Renmin Hospital of Wuhan University (Approval No: 20230101 A) and conducted in compliance with the National Research Council’s Guide for the Care and Use of Laboratory Animals. Eight-week-old male Wistar rats were obtained from SiPeiFu Biotechnology Co, Ltd. (Beijing, China) and divided into the four following groups: sham group (n = 6), tendon injury group (n = 6), tendon injury + hUCMSCs group (n = 6), and tendon injury + PEG-CeONP-carrying hUCMSCs group (n = 6). The rats were anesthetized using 5% isoflurane inhalation, and a single defect (1 mm) was created in the patellar tendon. The latter two groups received 106 hUCMSCs or 106 PEG-CeONP-carrying hUCMSCs through intra-tendon injections once a week [32]. After 4 weeks, the rats were sacrificed and the patellar tendon was collected and fixed for further experiments.

Histological experiments

Embedded tendons were cut into 5 μm sections and stained with hematoxylin-eosin (HE) and Masson’s trichrome reagent. For immunofluorescence experiments, sections were subjected to antigen retrieval and incubated overnight at 4 °C with anti-SCX (1:200, ab58655, Abcam, Cambridge, UK), anti-TNMD (1:200, ABIN872740, Sizhengbai Biotechnology, Suzhou, China), anti-COL1A1 (1:200, A1352, Abclonal), and anti-COL3A1 (1:200, A3795, Abclonal) antibodies, followed by the corresponding secondary antibodies conjugated with FITC or Cy3. Part of the tendon was freshly frozen in optimal cutting temperature compound for DHE staining. Visualization was performed using a fluorescent microscope and quantification using ImageJ software.

Statistical analysis

All data in this research are shown as mean ± standard deviation. Shapiro–Wilk normality test was used to perform the normality test. For the data with normal distribution, one-way analysis of variance (ANOVA) followed by Bonferroni’s test (multiple groups) was administered. For the data with non-normal distribution, we performed the Kruskal–Wallis H-test, followed by Dunn’s test (multiple groups). For repeated measurements, ANOVA for repeated measurements was administered. p < 0.05 was considered a statistically significant difference.

Result

Characterization of hUCMSCs

Primary hUCMSCs were identified as described previously [33]. hUCMSCs showed fibroblast-like morphology under a light microscopy (Fig. 1A). Flow cytometry results indicated that hUCMSCs were positive for CD29, CD44, CD90, CD105 and negative for CD45, CD34 (Fig. 1B). Next, hUCMSCs were treated with differentiation media to assess their differentiation potential. Oil red O, Alizarin Red, and Alcian Blue staining showed that the hUCMSCs successfully differentiated into adipocytes, osteocytes and chondrocytes, respectively (Fig. 1C). Thus, hUCMSCs used in this study possessed high purity and excellent differentiation potential.

Fig. 1
figure 1

Characterization of hUCMSCs. A Primary hUCMSCs were spindle-shaped under light microscopy. B Flow cytometry analysis showed that 98.5% of cells expressed CD29, 97.4% expressed CD44, 98.6% of cells expressed CD90, and 98.1% of cells expressed CD105. Meanwhile, only 0.48% and 0.28% expressed CD45 and CD34. C Lipid droplets were stained by oil red O after adipogenic differentiation induction of hUCMSCs. Mineralized nodules were stained by Alizarin Red after osteogenic differentiation induction of hUCMSCs. Acid mucopolysaccharide was stained by Alcian Blue after chondrogenic differentiation induction of hUCMSCs

Full size image

Synthesis and antioxidant analysis of PEG-CeONPs

The hydrophobic CeONPs were synthesized by a previously reported thermal decomposition method [30]. After coating with mPEG2k-DSPE, CeONPs were transferred to the aqueous phase. As shown in Fig. 2A, the obtained PEG-CeONPs exhibited uniform morphologies with sizes of 5.45 ± 1.08 nm. Compared with hydrophobic CeONPs, there are two additional absorption peaks at 1737 and 1104 cm−1, which could be attributed to C=O and C–O–C of DSPE-mPEG respectively [34, 35], indicating the successful surface modified (Fig. 2B). XPS analysis indicates that Ce3+ and Ce4+ co-exist on the surface of hydrophobic CeONPs and PEG-CeONPs, which provide the chemical basis for the catalytic activities (Fig. 2C) [36]. EDS results revealed a Ce:O atomic ratio of 0.50 (Fig. 2D). Moreover, CeONPs were stable in an aqueous solution for at least 2 weeks, as evidenced by their appearance and the results of the UV-visible spectra analysis (Fig. 2E, F).

The conversion of cerium ions into their trivalent and tetravalent forms typically endows CeONPs with multiple antioxidant enzyme activities [37]. To substantiate this function, we performed a series of in vitro antioxidant assays. The results of SOD and CAT enzyme mimetic activity analysis revealed that PEG-CeONPs inhibited the generation of superoxide anions and hydrogen peroxide in a dose-dependent manner (Fig. 2G, H). DPPH free radical and hydroxyl radicals scavenging assays are widely used to evaluate antioxidant properties [38, 39]. PEG-CeONPs decreased the concentration of free radicals in a dose-dependent manner (Fig. 2I, J). The data suggested that the preparation of PEG-CeONPs with excellent antioxidant properties was successful.

Fig. 2
figure 2

Synthesis and antioxidant analysis of CeONPs. A Representative TEM images of CeONPs and PEG-CeONPs. FITR (B) and XPS (C) analysis of CeONPs and PEG-CeONPs. D EDS spectrum of PEG-CeONPs. Appearance (E) and UV–visible spectra (F) of PEG-CeONPs in aqueous solution for 2 weeks. Inhibition rate of superoxide anion (G), hydrogen peroxide (H), hydroxyl radicals (I), and DPPH (J) with different PEG-CeONP concentrations treatment (n = 3). **p < 0.01 versus 0 mg/mL PEG-CeONP

Full size image

PEG-CeONPs decreased the concentration of ROS in hUCMSCs

We loaded PEG-CeONPs into hUCMSCs to enhance their tolerance to high ROS levels in the injured area (Fig. 3A). CCK8 assay showed that PEG-CeONPs were safe for hUCMSCs with a concentration of less than 50 µg/mL both at 24- and 48-h exposure (Fig. 3B). Thus, we administrated 50 µg/mL PEG-CeONPs to hUCMSCs for 24 h to harvest PEG-CeONPs-carrying hUCMSCs. As shown in Fig. 3C, Cy3-PEG-CeONPs were intaken by hUCMSCs and dispersed around the nucleus. High levels of ROS are characteristic of the injured lesions and are responsible for the limited efficacy of stem cell transplants. Accordingly, hydrogen peroxide was used to mimic the oxidative stress in the injured area of tendon. DCFH-DA and DHE are cell-permeable probes that are widely used to detect the intracellular concentration of H2O2 or superoxide [40, 41]. DCFH probe imaging suggested that PEG-CeONPs reduced the concentration of ROS after H2O2 treatment; there was no significant difference between the NC and PEG-CeONPs groups (Fig. 3D). Subsequently, a DHE experiment was conducted and similar results were obtained (Fig. 3E). The results showed that PEG-CeONPs were successfully carried by hUCMSCs and reduced the intracellular ROS levels under H2O2 intervention.

Fig. 3
figure 3

PEG-CeONPs decreased the concentration of ROS in hUCMSCs. A A schematic drawing of the hUCMSC carried PEG-CeONPs. B hUCMSC viability assessed by the CCK-8 assay after treatment with PEG-CeONPs (n = 3). C hUCMSC uptake Cy3-labeled PEG-CeONPs and fluorescence imaging (n = 3). D, E Representative fluorescence imaging of intracellular ROS evaluated by DCFH and DHE probes and mean fluorescence intensity (n = 3). NS means not significant versus NC or 0 mg/mL PEG-CeONPs, **p < 0.01 versus NC or 0 mg/mL PEG-CeONPs, #p < 0.05 versus H2O2

Full size image

PEG-CeONPs-carrying hUCMSCs resisted ROS-mediated apoptosis

High ROS levels promote apoptosis through diverse signaling pathways [42]. Therefore, enhancing the regenerative and anti-apoptotic capacities of stem cells before treatment is considered an interventional strategy for improving therapeutic efficacy [43]. Here, we investigated the effect of PEG-CeONPs on hUCMSCs apoptosis under oxidative stress. Intrinsic cellular apoptosis is always accompanied by loss of mitochondrial membrane potential [43]. We performed a mitochondrial membrane potential assay, and the results showed that healthy polarized mitochondria (JC-1 aggregates, red) decreased and unhealthy depolarized mitochondria (JC-1 monomers, green) increased in the H2O2 group (Fig. 4A). However, PEG-CeONPs treatment partially reversed this effect. Additionally, PEG-CeONPs treatment reduced the expression of BAX (a pro-apoptotic protein) and increased the expression of BCL2 (an anti-apoptotic protein) under oxidative stress (Fig. 4B). PEG-CeONPs also reduced the percentage of apoptotic hUCMSCs under oxidative stress as shown by the results of flow cytometry (Fig. 4C). These results suggest that PEG-CeONPs-carrying hUCMSCs could resist ROS-mediated apoptosis.

Fig. 4
figure 4

PEG-CeONP-carried hUCMSCs resisted ROS-mediated apoptosis. A Representative fluorescence images of mitochondrial potential measured by JC-1 staining and mean fluorescence intensity (n = 3). B WB assays and Semi-quantitative analysis of BAX and BCL2 in NC, PEG-CeONPs, H2O2 and PEG-CeONPs + H2O2 groups (n = 3). C Annexin V/PI staining and cytometry indicated analysis to assess the percentage of apoptotic cells (n = 3). NS means not significant versus NC, **p < 0.01 versus NC, #p < 0.05 versus H2O2, ##p < 0.05 versus H2O2

Full size image

PEG-CeONPs-carrying hUCMSCs resisted ROS-mediated senescence

Excessive ROS levels disrupt mitochondria and trigger cellular senescence, impairing stem cell function and tissue regeneration [44]. Therefore, we investigated whether PEG-CeONPs exerted their anti-senescent function by scavenging ROS. Irreversible growth arrest is a characteristic of cellular senescence [45]. We performed a clonal formation assay and a CCK-8 assay to evaluate the proliferative capacity. The results showed that PEG-CeONPs counteracted the inhibitory effects of H2O2 on clonal formation and proliferation (Fig. 5A, B). Moreover, PEG-CeONPs enhanced the expression of Ki-67, a proliferation maker, under oxidative stress (Fig. 5C). Compared with the H2O2 group, the PEG-CeONPs + H2O2 group showed decreased senescence, as revealed by the reduced level of β-galactosidase, and decreased expression of the senescence-related factors, including P16 and P21, at the mRNA and protein levels (Fig. 5D−F). Senescent cells secrete pro-inflammatory factors and proteases to alter the tissue microenvironment, which is collectively termed the senescence-associated secretory phenotype (SASP) [46]. RT-qPCR and Western blot results indicated that treatment with PEG-CeONPs decreased the expression of SASP makers, including IL-1β, IL-6, and TNFα, which were enhanced by H2O2 (Fig. 5E and F). In summary, PEG-CeONPs-carrying hUCMSCs counteracted senescence induced by high ROS levels.

Fig. 5
figure 5

PEG-CeONPs-carried hUCMSCs resisted ROS-mediated senescence. A Representative colony formation morphology with result quantification (n = 3). B Cell viability was assessed by the CCK-8 assay after treatment with PEG-CeONPs and H2O2(n = 3).  C Ki-67 immunofluorescence staining and percentage of positive cells (n = 3). D Representative images of β-galactosidase staining and percentage of senescent cells (n = 3). E WB analysis of P21, P16, IL-1β, IL-6 and TNFα in hUCMSC treated with PEG-CeONPs and H2O2 (n = 3). F RT-qPCR analysis of P21, P16, IL-1β, IL-6 and TNFα in hUCMSC treated with PEG-CeONPs and H2O2 (n = 3). NS means not significant versus NC, **p < 0.01 versus NC, #p < 0.05 versus H2O2, ##p < 0.05 versus H2O2

Full size image

PEG-CeONPs-carrying hUCMSCs resisted apoptosis and senescence through NFκB and MAPK signaling pathways

Given that nuclear factor kappa-B (NFκB) and mitogen-activated protein kinase (MAPK) signaling pathways are involved in the transduction of oxidative signal and cellular apoptosis, we performed further experiments to elucidate the mechanism underlying PEG-CeONPs in hUCMSCs exposed to excessive oxidative stress [47, 48]. As shown in Fig. 6A and B, H2O2 elevated the expression of P-IκBA, P-P65, P-JNK, and P-P38, and reduced the level of IκBA, indicating the activation of NFκB and MAPK signaling pathways. To investigate the roles of NFκB and MAPK signaling pathways in the anti-apoptosis and anti-senescence effect of PEG-CeONPs, we treated PEG-CeONPs-carrying hUCMSCs with NFκB activator human recombinant TNFα and JNK/P38 activator DE [49, 50]. The application of human recombinant TNFα and DE counteracted the downregulatory effect of PEG-CeONPs on the BAX, P16 and P21 at the protein level and decreased the expression of BCL2. Thus, our data verified that PEG-CeONPs resisted apoptosis and senescence by inactivating the NFκB and MAPK signaling pathways.

Fig. 6
figure 6

PEG-CeONPs-carried hUCMSCs resisted to apoptosis and senescence through NFκB and MAPK signaling pathway. A WB analysis of IκBα, P-IκBα, P65, P-P65, BAX, BCL2, P16 and P21 in hUCMSC treated with PEG-CeONPs, H2O2 and TNFα (n = 3). B WB analysis of JNK, P-JNK, P38, P-P38, BAX, BCL2, P16 and P21 in hUCMSC treated with PEG-CeONPs, H2O2 and DE (n = 3). **p < 0.01 versus NC, #p < 0.05 versus H2O2, ##p < 0.05 versus H2O2, $p < 0.05 versus PEG-CeONPs, $$p < 0.01 versus PEG-CeONPs. NS means not significant

Full size image

Local administration of PEG-CeONP-hUCMSCs effectively promotes the recovery of patellar tendon defect

High levels of oxidative stress contribute to the limited efficacy of stem cell transplantation in injured areas. Thus, to investigate the advantages of the local injection of PEG-CeONP-hUCMSCs in tendon healing, we established a patellar tendon defect (PTD) model induced by surgery [2]. hUCMSCs and PEG-CeONP-hUCMSCs were injected into the tendon once per week (Fig. 7A). After 4 weeks of treatment, tendon tissues collected from the PTD group presented a significant defect. In contrast, tendons from the hUCMSCs and the PEG-CeONP-hUCMSCs presented smaller defects, especially in the latter (Fig. 7B). HE and Masson staining showed that treatment with PEG-CeONP-hUCMSCs reversed the disorderly arrangement of collagen fibers and attenuated the formation of vacuole-like structures (Fig. 7C). To evaluate the oxidative stress levels, the ROS fluorescent probe DHE was loaded into the tissues. Fluorescent images and semiquantitative analysis showed that PEG-CeONP-hUCMSCs remarkably inhibited the increase in ROS levels evoked by the injury compared to the hUCMSCs intervention (Fig. 7D). Moreover, PEG-CeONP-hUCMSCs enhanced tendon repair as evidenced by the higher expression of tenogenic markers SCX, TNMD, COL1A1, and COL3A1 (Fig. 7E–H). Taken together, the local administration of PEG-CeONP-hUCMSCs effectively promoted the repair of patellar tendon defects in rat models.

Fig. 7
figure 7

Local administration of PEG-CeONP-hUCMSCs effectively promotes recovery of patellar tendon defection. A A schematic drawing of animal experiments. B The appearance of tendon-injured areas. C Masson and HE staining of the tendon with PTD, PTD + hUCMSCs and PTD + PEG-CeONP-hUCMSCs treatment. D Representative fluorescence imaging of intracellular ROS evaluated by DHE probes and mean fluorescence intensity (n = 6). E, F Representative fluorescence imaging of immunofluorescence of TNMD, SCX, COL3A1 and COL1A1 (n = 6). G, H mean fluorescence of immunofluorescence image. NS means not significant versus PTD (n = 6), **p < 0.01 versus Sham, ##p < 0.01 versus PTD, &p < 0.01 versus PTD + hUCMSCs, &&p < 0.01 versus PTD+ hUCMSCs

Full size image

Discussion

Tendon injuries cause prolonged disability with high incidence [51]. The repair of tendon injuries is incomplete because of fibrotic scarring. Plenty of treatment strategies have been developed to enhance tendon healing; however, their efficacy requires improvement. Stem cell transplantation holds great promise in multiple diseases, including tendon injuries, as stem cells secrete a diverse repertoire of growth factors such as vascular endothelial growth factors, fibroblast-like growth factors, and insulin-like growth factors [52,53,54]. However, the poor survival of the stem cells in injured tissues impedes their therapeutic development, which is attributed to the high ROS concentrations in the injured areas.

Cerium oxide is a stable lanthanide metal element and the pale-yellow oxide form of the most abundant rare-earth metal [55]. Driven by a ground-state Ce 4f electron, the powerful Ce3+ to Ce4+ redox couple contribute to considerable reducibility [56]. Possessing superoxide dismutase, catalase, and peroxidase activities, CeONPs have been considered antioxidant agents [57,58,59]. CeONPs, with a higher surface-area-to-volume ratio, have better reducibility than cerium oxide with larger particles because of oxygen vacancies and Ce3+ mostly existing on the surface [60]. In our study, ultrasmall PEG-CeONPs with a size of 5.45 nm were synthesized and exhibited significant reducibility. Previous studies have been conducted to enhance the therapeutic efficacy of cell transplantation by pre-treating stem cells with natural drugs such as curcumin, Exendin-4, and resveratrol or by genetically modulating [61,62,63]. The above options often face high cost, non-specificity, limited activity and consequent uncontrolled side effects and thus fail in clinical trials [28]. In comparison, CeONPs, the artificial enzyme, has the advantage of low cost, high efficiency and stability, massive production, and easy handling, which attracted us to explore its role in cell transplantation [64]. Nanoparticles can be actively taken up by cells via endocytosis, indicating PEG-CeONPs can be carried by hUCMSCs and transplanted to the injured areas [65]. Although generally safe, concentration-dependent cytotoxicity of CeONPs has been observed in various cell types [66, 67]. Therefore, we explored and verified the appropriate loading conditions under which PEG-CeONPs exhibited both biocompatibility and antioxidant capacity in hUCMSCs.

Apoptosis, a form of programmed cell death, markedly affects stem cell transplantation [21]. Oxidative stress can induce apoptosis in many ways, including activation of the mitochondrial pathway (intrinsic) and activation of death receptors at the cell surface (extrinsic) [68]. Apoptotic stem cells possess severely compromised regenerative potential and secret fewer growth factors; therefore, improving the outcome of stem cell therapy is a novel interventional strategy, as exemplified by endowing stem cells with pro-survival and anti-apoptotic genes [43]. In this work, we confirmed that PEG-CeONP-carrying hUCMSCs could resist oxidative damage-induced cellular apoptosis in vivo and in vitro. In addition to causing apoptosis in transplanted cells, high oxidative stress can damage in site tendon-derived stem cells and deteriorate tendon healing [69]. Our study suggests that injection of PEG-CeONP-carrying hUCMSCs remodeled the microenvironment in the injured area by scavenging and inhibiting apoptosis. Senescence is a cellular state characterized by irreversible growth arrest and an altered epigenetic mechanism [70]. Recently, the transplantation of senescent cells into young animals was shown to result in persistent physical dysfunction [71, 72]. Senescence can be triggered by various stressors, such as multiple generations, activation of oncogenes, and ROS-induced DNA damage [73]. The accumulation of senescent cells in tendon tissue is a possible pathogenesis mechanism underlying tendinopathy [74]. Our data revealed that PEG-CeONPs protected hUCMSCs from H2O2-evoked cellular senescence and decreased the expression of senescence-associated molecules P16 and P21 in rats.

NFκB is a pleiotropic, redox-sensitive, nuclear transcription factor regulating the expression of many genes and associating with multiple biological processes, including apoptosis and senescence [75, 76]. The NFκB signaling pathway can be activated by H2O2 at different sites [77]. The activation of the NFκB signaling pathway is involved in musculoskeletal diseases including tendon diseases [78]. The MAPK pathway is a representative stress-responsive signaling pathway that induces cellular responses to divergent environmental stimuli [79]. Increased ROS generation leads to the activation of MAPK cascades, including c-Jun NH2-terminal kinase (JNK), and p38 MAPK [80]. Apoptosis and senescence are mediated by the MAPK signaling pathway [81, 82]. Therefore, we sought to explain the function of PEG-CeONPs by assessing the activation of NFκB and MAPK signaling pathways. Our data showed that PEG-CeONPs inhibited the activation of NFκB and MAPK signaling pathway in H2O2-treated hUCMSCs, and human recombinant TNFα and DE reversed the effect of PEG-CeONPs loading, indicating that PEG-CeONPs acted through the NFκB and MAPK signaling pathways (Fig. 8).

Fig. 8
figure 8

Schematic illustration of PEG-CeONP-carrying hUCMSCs counteract oxidative damage and facilitates tendon regeneration

Full size image

Various biomaterials strategies have been developed for tendon regeneration because of the poor repairs induced by low cellularity sources [83], poor blood supply [84], inflammatory microenvironment, and excessive oxidative stress [1]. Currently, biomaterials strategies for tendon regeneration mainly include scaffolds to excel at mechanical stabilization and hydrogels to deliver biochemical cues or cells [85]. For instance, A polymeric three-dimensional scaffold was reported to retain tendon-like mechanical properties and accelerate the healing progression [86]. Ren et al. designed a high-tenacity shape-adaptive hydrogel to deliver fibroblast growth factor [87]. Ji et al. proposed a cocktail-like hydrogel to transmit bone marrow mesenchymal stem cells [88]. The main issue of our research would be concerns about is the excessive oxidative stress in injured areas. In previous studies, people used antioxidative drugs such as melatonin and hormone agonists against oxidative injury in injured areas [89, 90]. We took advantage of the excellent and repeatable reversibility of cerium oxide nanoparticles and proposed a novel strategy that constructed cerium oxide nanoparticle-carried hUCMSCs. Our results indicate that the carriage of cerium oxide nanoparticles could against oxidative stress and enhance the effect of hUCMSCs. In future studies, novel bioactive hydrogel-based delivery solutions are expected to join forces with stem cell modification programs to provide a more effective intervention strategy for tendon injury. Our study has some limitations. First, given the potential ethical issue, we must explore the effects of PEG-CeONP-carrying hUCMSCs in rats, and the impact of xenogeneic transplantation should be considered. Additionally, the biological behavior of PEG-CeONP-carrying hUCMSCs in vivo should be tracked using advanced experimental methods. Finally, a synthetic scheme for producing clinical-grade PEG-CeONPs should be developed for further clinical research.

Conclusion

This study indicates that PEG-CeONPs with repeatable reducibility were taken up by hUCMSCs by endocytosis. When exposed to excessive concentrations of H2O2, most cells carried with the PEG-CeONPs remained active to avoid senescence and apoptosis by inhibiting of ROS-induced NFκB and MAPK activation. In addition, to intracellular oxidative stress, high levels of ROS in the injured microenvironment were reduced by PEG-CeONP-carrying hUCMSCs. Intra-tendon injection of PEG-CeONP-carrying hUCMSCs facilitated tendon regeneration by preventing senescence and apoptosis. The results in our study provide a novel nongenetic strategy to improve the therapeutic potential of stem cells, which could be widely used in the transplantation of stem cells target various diseases.