New strategy of bone marrow mesenchymal stem cells against oxidative stress injury via Nrf2 pathway: oxidative stress preconditioning.

Clinically, bone marrow mesenchymal stem cells (BMSCs) have been used in treatment of many diseases, but the local oxidative stress (OS) of lesion severely limits the survival of BMSCs, which reduces the efficacy of BMSCs transplantation. Therefore, enhancing the anti-OS stress ability of BMSCs is a key breakthrough point. Preconditioning is a common protective mechanism for cells or body. Here, the aim of this study was to investigate the effects of OS preconditioning on the anti-OS ability of BMSCs and its mechanism. Fortunately, OS preconditioning can increase the expression of superoxide dismutase, catalase, NQO1, and heme oxygenase 1 through the nuclear factor erythroid 2-related factor 2 pathway, thereby decreased the intracellular reactive oxygen species (ROS) levels, relieved the damage of ROS to mitochondria, DNA and cell membrane, enhanced the anti-OS ability of BMSCs, and promoted the survival of BMSCs under OS.

INTRODUCTION

Bone marrow mesenchymal stem cells (BMSCs) have strong regenerative ability and multi‐directional differentiation potential.1 Clinically, BMSCs have been used in the treatment of various diseases, especially in tissue engineering.2, 3, 4 However, we have found that the clinical application of BMSCs still has problems that need to be solved: oxidative stress (OS) severely limits the survival of BMSCs after transplantation, and seriously affects the transplantation efficacy.5, 6 OS can produce excessive reactive oxygen species (ROS) in BMSCs, which can peroxidize DNA, protein, lipid, mitochondria, endoplasmic reticulum, and others, and eventually accelerate cell senescence and apoptosis.7, 8, 9 Therefore, enhancing the anti‐OS ability of BMSCs will be expected to further improve the effect of BMSCs transplantation.

At present, the methods of enhancing the anti‐OS ability of BMSCs mainly include the application of antioxidants or drugs, which mainly play a role through intervention of the signaling pathways of aging and apoptosis, but the effect is limited.10, 11 Clinically, we often find that preconditioning has protective effects on organisms or body, such as myocardial ischemic preconditioning.12 That being so, whether OS preconditioning can also enhance anti‐OS ability of BMSCs, protect the survival of BMSCs under OS conditions? For the possible mechanism of antioxidant stress in cells, nuclear factor erythroid 2‐related factor 2 (Nrf2) is the most important and classical one among many signaling pathways.13, 14, 15 Nrf2 nuclear translocation, which promotes the expression of antioxidant enzymes genes, is essential for antioxidant stress.16, 17

Therefore, this study will use low concentrations of H2O2 to treat BMSCs for OS preconditioning, and then use high concentration of H2O2 for oxidative damage. To observe whether OS preconditioning can enhance the anti‐OS ability of BMSCs via Nrf2 pathway and promote the survival of BMSCs under OS.

MATERIALS AND METHODS

Animals

A total of 10 young male New Zealand white rabbits (4‐6 weeks old, 1.0‐2.0 kg) were provided by the Laboratory Animal Center of Guizhou Medical University (Guiyang, China). All procedures were performed in accordance with our Institutional Guidelines for Animal Research, and the investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (publication no. 85‐23, revised in 1996).

Reagents and instruments

Low glucose Dulbecco's modified Eagle's medium (L‐DMEM; Gibco), fetal bovine serum (FBS; Gibco), trypsin (Gibco), double antibody (Hyclon), phosphate‐buffered saline (PBS; Hyclon), percoll separation solution (Pharmacia), dimethyl sulfoxide (Sigma, Germany), cell counting kit‐8 (CCK‐8) solution (Solarbio, Beijing, China), 30% hydrogen peroxide solution (Chengdu Jinshan Chemical Reagent Co, Ltd, Sichuan, China), ROC detection kit (sigma, Germany), cell apoptosis mitochondrial membrane potential detection kit (KeyGEN BioTECH, Jiangsu, China), Annexin V‐fluorescein isothiocyanate (FITC)/propidium iodide (PI) (BD), terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) (Beyotime, Shanghai, China), 4′,6‐diamidino‐2‐phenylindole (DAPI) (Solarbio), malondialdehyde (MDA) test kit (Beyotime), catalase (CAT), and superoxide dismutase (SOD) viability test kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Nuclear and Cytoplasmic Extraction Reagents (Invitrogen). Protein A/G Plus‐Agarose and anti‐NQO1 antibody (Santa Cruz), anti‐Nrf2 antibody and anti‐Keap1 antibody (Sigma). Anti‐SOD2 antibody, anti‐heme oxygenase 1 (HO‐1) antibody, anti‐CAT antibody, and anti‐β‐actin antibody (Abcam, England). Micro‐adjustable pipette (Eppendorf, Germany), biosafety cabinet (ESCO, Singapore), benchtop high speed refrigerated centrifuge (Beckman), nucleic acid and protein measuring instrument (Nanodrop), quantitative polymerase chain reaction instrument (BIO‐RAD), multi‐function microplate reader (Biotech), inverted fluorescence microscope (Zeiss, Germany), and laser confocal microscope (Zeiss).

Isolation and culture of BMSCs

New Zealand white rabbits, 4 to 6 weeks old, weight 1.0 to 2.0 kg, bone marrow fluid were extracted from the distal femur and proximal tibia under aseptic conditions. The bone marrow fluid was centrifuged to remove the suspended fat (600 g/min, 5 minutes), and the bone marrow fluid was dripped on the percoll separation solution (1.073g/mL), and then the nucleated cell was separated by density gradient centrifugation (900g/min, 30 minutes). Finally, nucleated cells were cultured in L‐DMEM medium (10% FBS and 1% double antibody) under the condition of 37℃ and 5% CO2. When primary cells converge reached 80% to 90%, they were passaged at 1:3.

Determination of optimal H2O2 preconditioning concentration

The H2O2 concentration of preconditioning is the key part of the experiment, to determine the optimal preconditioning concentration, according to different H2O2 concentration, the third‐generation BMSCs were divided into five groups: group A (control group/0 μM), group B (50 μM), group C (100 μM), group D (200 μM), and group E (300 μM). When the cell confluence was about 80%, L‐DMEM medium containing different concentrations of H2O2 was added according to the grouping conditions, and the treatment was continued for 8 hours. Cell viability was detected by CCK‐8 method: cells were washed three times with PBS, fresh medium was replaced, 10 μL of CCK‐8 solution was added to each well, and after incubation for 3 hours, the absorbance value was measured by a microplate reader (450 nm). Apoptosis was detected by Annexin V/PI double staining: cells were harvested by trypsinization without EDTA, and cells were resuspended in buffer, adjusted the cell density to 1 × 107/mL, and 100 μL of cell suspension was transferred into the flow tube, then added 5 μL of Annexin V‐FITC and 5 μL of PI, the mixture was incubated at room temperature for 15 minutes in the dark. Finally, apoptosis was detected by flow cytometry within 1 hour. The concentration of H2O2 which is nontoxic to cells is the optimal pre‐adapted concentration.

Determination of optimal H2O2 oxidative injury concentration

To determine the optimal H2O2 concentration of oxidative damage, the third‐generation BMSCs were divided into 5 groups: group A (control group/0 μM), group B (500 μM), group C (1000 μM), and group D (1500 μM). When the cell confluence was about 80%, L‐DMEM medium containing different concentrations of H2O2 was added according to the grouping conditions, and the treatment was continued for 24 hours, and cell viability and apoptosis were detected as described above. The H2O2 concentration can make the apoptosis reach about 20%, and the cell viability decreases by about 50%, which is the optimal oxidative damage concentration.

OS preconditioning of BMSCs

The third‐generation BMSCs were inoculated into plates or culture‐flask. The experiment was divided into three groups: control group (BMSCs + 0 μM H2O2), oxidative damage group (BMSCs + 1000 μM H2O2) and preconditioning group (BMSCs + 50 μM H2O2 + 1000 μM H2O2). When the cell confluence was about 80%, according to the experimental group, the preconditioning group was first treated with 50 μM H2O2 for 8 hours, then recovered for 12 hours, and finally treated with 1000 μM H2O2 for 24 hours, the oxidative damage group was directly treated with 1000 μM H2O2 for 24 hours, and the control group was routinely cultured.

Evaluation of the protective effect of OS preconditioning on BMSCs

Detection of intracellular ROS

The center of OS is the accumulation of excessive ROS, and 2',7'‐Dichlorofluorescin diacetate (DCF‐DA) fluorescence probe is used to detect intracellular ROS: the cells were washed twice with PBS, 400 μL of the main reaction mixture was added to each dish, and incubated in an incubator (37℃) for 30 minutes, PBS washing to remove the residual DCF‐DA, then using laser confocal microscopy (FITC green fluorescent) detection.

Observation of mitochondrial membrane potential

Mitochondria is the core position of OS, and the changes of mitochondrial membrane potential are detected by JC‐1 method: the reaction mixture was prepared according to the kit instructions. After washing the cells with PBS, the reaction mixture was added, incubated in an incubator (37℃) for 30 minutes, and then the cells were washed three times with PBS, ultimately, using laser confocal microscopy (red and green fluorescent) detection.

Observation of DNA damage

Excessive ROS not only damage mitochondria, but also injury DNA by peroxidation. DNA damage was detected by TUNEL/DAPI staining: the cells were washed with PBS, 4% paraformaldehyde was fixed at room temperature for 30 minutes, 0.3% Triton X‐100 was permeated for 6 minutes, added proper amount of TUNEL solution, and cells were incubated in an incubator (37℃) for 60 minutes in the dark, and DAPI stained for 4 minutes. After each step, the cells were thoroughly washed with PBS, finally, the antifluorescence quencher was added and the laser confocal microscopy was used to observe fluorescence (Cy3 red fluorescent).

Detection of MDA content, SOD, and CAT viability

MDA is a lipid peroxidation product, CAT and SOD is an important antioxidant enzyme in cells. MDA content were detected by thiobarbituric acid method: the cells were disrupted by sonication, 100 μL of lysate was aspirated, 200 μL MDA working solution was added to the lysate, after mixing, the mixture was heated (100℃) for 15 minutes, cooled by ice water, centrifuged at 1000g for 10 minutes, 200 μL of the supernatant was added to 96‐well plate, and then the absorbance was measured by a microplate reader (532 nm). CAT and SOD viability were detected by WST‐1 method: BCA method was firstly used to detect the protein concentration, and then the samples, enzyme diluents and substrate application solution were added in the 96‐well plate, incubated in an incubator (37℃) for 20 minutes, the absorbance value was measured by a microplate reader (450 nm). The CAT and SOD viability was calculated according to the instructions of the CAT and SOD viability test kit, respectively.

Cell viability and apoptosis detection

To determine whether OS preconditioning has a protective effect on cell survival under OS conditions, the third‐generation BMSCs were treated with H2O2 in 96‐well plates or 75 cm2 flasks, the cell viability and apoptosis were measured as described above methods.

Mechanism of OS preconditioning enhancing the anti‐OS of BMSCs

Interaction of Nrf2 and Keap1 in cells

After treatment of BMSCs with H2O2 according to experimental grouping conditions, the cell total protein was extracted and immunoprecipitation was used to detect the interaction between Nrf2 and Keap1. The cytosolic fractions were incubated with anti‐Nrf2 or anti‐Keap1 antibodies for 12 hours at 4℃. Then, Protein A/G Plus‐Agarose was added for 2 hours at 4°C on a rotating device. Immunoprecipitates were collected by centrifugation at 6000g at 4°C and washed with lysis buffer. The pellets were eluted by heating at 95°C for 5 minutes in electrophoresis sample buffer and subjected to immunoblotting.

Nrf2 nuclear translocation

After treatment of BMSCs with H2O2, nuclear and cytosolic fractions were isolated using the nuclear and cytoplasmic extraction reagents, these nuclei and cytoplasmic lysates were used for Western blot analysis. Equal amounts of proteins were added to the gel for electrophoresis, after electrophoresis, the protein was transferred to a nitrocellulose membrane. Anti‐Nrf2 antibody was incubated overnight, and the second antibody was incubated for 2 hours. Images were taken with the Gel Imaging System and quantified using ImageJ software.

Expression of antioxidant enzymes

After treatment of BMSCs with H2O2, cells were collected in radioimmunoprecipitation assay buffer and centrifuged at 13 000g for 5 minutes to obtain the total cellular protein. Then, the SOD, CAT, NQO1, and HO‐1 were detected by Western blot, the specific method is the same as above.

Statistics analysis

All data in study were shown in mean ± SD. The Kolmogorov‐Smirnov statistical testing was first utilized to assess the normality of data, and then the analysis of variance (ANOVA) was utilized to analyze the homogeneity of data. If the test met the demand, the one‐way ANOVA was used for comparison. If not, the Wilcoxon test was then utilized for comparison.

RESULTS

The primary BMSCs adhered and stretched after 2 days (Figure 1A), about 5 days, the cells were colony‐shaped growth (Figure 1B), after approximately 10 days, and the cells gradually covered about 90% of bottom, showing a long spindle (Figure 1C). After 24 hours of BMSCs resuscitation, more than 70% of the cells were adhered to the bottom, and the cells were spindle or polygonal (Figure 1D); after 2 days of resuscitation, the cells began to grow rapidly (Figure 1E), about 4 days, the cells covered about 90% of bottom, which showed a long spindle (Figure 1F).

Figure 1

Isolation and culture of BMSCs. A, Primary BMSCs cultured for 2 days (100×). B, Primary BMSCs cultured for 5 days (100×). C, Primary BMSCs cultured for 10 days (100×); (D) BMSCs recovered for 24 hours (100×). E, BMSCs recovered for 2 days (100×). F, BMSCs recovered for 4 days (100×). BMSC, bone marrow mesenchymal stem cell

Optimal H2O2 preconditioning concentration

BMSCs was treated with different low concentration H2O2. The cell viability decreased with the increase of concentration, and the apoptosis increased gradually. When concentrations of H2O2 were 100, 200, or 300 μM, the cell viability was lower than the control group, and the apoptosis was higher than the control group, the difference was statistically significant (P < .05). When the concentration of H2O2 was 50 μM, the cell viability was about (99.02 ± 1.01)%, and the apoptosis was about (1.08 ± 0.05)%; compared with the control group, the difference was not statistically significant, not only that, but there was no significant change in cell morphology and proliferation, so the optimal preconditioning concentration was 50 μM (Figure 2).

Figure 2

Viability and apoptosis of BMSCs in different low concentrations of H2O2. A‐E, Apoptosis was detected by flow cytometry, concentrations of H2O2 were 0, 50, 100, 200, and 300 μM, respectively. F, Flow cytometry analysis of apoptosis (n = 3). G, Analysis of cell viability by CCK‐8 method (n = 3). All experiments were repeated three times. All data were presented as means ± SD. Statistical significances were calculated by the Student t test. Data were all compared with control group. BMSC, bone marrow mesenchymal stem cell; CCK‐8, cell counting kit‐8. *P < .05, # P > .05

Optimal H2O2 oxidative damage concentration

BMSCs were treated with different high concentrations of H2O2. The cell vitality of each group decreased with the increase of concentration of H2O2, and the cell apoptosis increased obviously. The cell vitality and apoptosis of each group were higher than the control group. However, when concentration of H2O2 was 1000 μM, the cell viability was about (48.74 ± 1.65)% and the apoptosis was about (19.92 ± 0.37)%, and compared with the control group, these difference was significant (P < .05), which resulted in a typical oxidative damage cell model, so the optimal H2O2 oxidative damage concentration was 1000 μM (Figure 3).

Figure 3

Viability and apoptosis of BMSCs in different high concentrations of H2O2. A‐D, Apoptosis was detected by flow cytometry, concentrations of H2O2 were 0, 500, 1000, and 1500 μM, respectively. E, Flow cytometry analysis of apoptosis (n = 3). F, Analysis of cell viability by CCK‐8 method (n = 3). All experiments were repeated three times. All data were presented as means ± SD. Statistical significances were calculated by the Student t test. Data were all compared with control group. BMSC, bone marrow mesenchymal stem cell; CCK‐8, cell counting kit‐8; FITC, fluorescein isothiocyanate. * P < .05

Protective effect of OS preconditioning

Intracellular ROS

OS is excessive ROS damage DNA, lipids, proteins, and other biological macromolecules. ROS accumulation is the central of OS. We use DCF‐DA fluorescent probe to detect intracellular ROS. Under laser confocal microscopy, only a small amount of ROS was formed in the blank group (Figure 4A and 4D). A large amount of ROS accumulation was observed in the oxidative damage group (Figure 4B and 4E). However, the ROS in the cells was significantly reduced after pretreatment (Figure 4C and 4F), and the difference was statistically significant (P < .05) (Figure 4G).

Figure 4

DCF‐DA fluorescent probe detected intracellular ROS. A, D, Control group. B, E, Oxidative damage group. C, F, Preconditioning group. G, Analysis of DCF‐DA fluorescence intensity (n = 5). All experiments were repeated three times. All data were presented as means ± SD. Statistical significances were calculated by the Student t test. Data were compared with control and oxidative damage groups respectively. Vs control group (*P < .05). Vs oxidative damage group, (# P < .05). DCF‐DA, 2′,7′‐Dichlorofluorescin diacetate; ROS, reactive oxygen species

Mitochondrial membrane potential

Mitochondria is not only the main source of intracellular ROS, but also main target for ROS damage. The decline of mitochondrial membrane potential is an early change of mitochondrial injury and an early signal of apoptosis. We used JC‐1 to detect changes in mitochondrial membrane potential. Under the confocal microscope, the cells in the blank group showed deepred and deepgreen (Figure 5A, 5D, and 5G), in the oxidative damage group showed lightred and deepgreen (Figure 5B, E and H), and in the preconditioning group also showed deepred and deepgreen (Figure 5C, 5F, and 5I). The results showed that the mitochondrial membrane potential in the oxidative damage group decreased significantly compared with the control group (P < .05), but there was no significant change in the preconditioning group, and the mitochondrial membrane potential in the preconditioning group was higher than the oxidative damage group (P < .05) (Figure 5J).

Figure 5

Detection of mitochondrial membrane potential by JC‐1. A, D, and G, Control group. B, E, and H, Oxidative damage group. C, F, and I, Preconditioning group. J, Analysis of red and green fluorescence intensity (n = 3). All experiments were repeated three times. All data were presented as means ± SD. Statistical significances were calculated by the Student t test. Data were compared with control and oxidative damage groups respectively. Vs control group (# P < .05). Vs oxidative damage group (*P < .05)

DNA damage and apoptosis

ROS can oxidize DNA, resulting in DNA breakage and exposing 3'‐OH, so TUNEL/DAPI staining can detect DNA damage. The results showed that there was no red fluorescence in the blank group, and the morphology of the nucleus was normal (Figure 6A, 6D, and 6G). A large amount of high‐intensity red fluorescence was observed in the oxidative injury group, and the nuclear membrane was shrunk, the chromatin was highly coagulated and marginalized, and apoptotic bodies were observed (Figure 6B, 6E, and 6H). While the red fluorescence intensity in the preconditioning group was obviously weakened, the morphology of most nuclear was normal, no apoptotic bodies were found, and DNA damage was lighter (Figure 6C, 6F, and 6I). The results showed that the cell apoptosis of the oxidative damage group was higher than the control group (P < .05), however, the preconditioning group was lower than the oxidative damage group (P < .05) (Figure 6J).

Figure 6

TUNEL/DAPI staining for DNA damage. A, D, and G, Control group. B, E, and H, Oxidative damage group. C, F, and I, Preconditioning group. J, Analysis of cell apoptosis (n = 3). All experiments were repeated three times. All data were presented as means ± SD. Statistical significances were calculated by the Student t test. Data were compared with control and oxidative damage groups respectively. Vs control group (# P < 0.05). Vs oxidative damage group (*P < .05). DAPI, 4′,6‐diamidino‐2‐phenylindole; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling

MDA content, SOD, and CAT viability

MDA is a product of lipid peroxidation, which reflects the degree of lipid peroxidation, while SOD and CAT are the main antioxidant enzymes in cells, which can reflect the antioxidant capacity of cells. The results of MDA assay showed that the highest MDA in oxidative injury group was about (2.76 ± 0.03) nmol/mg prot, which was significantly higher than the preconditioning group (P < .05) (Figure 7A). The results of SOD and CAT viability showed that the SOD and CAT viability were the highest in blank group and the lowest in oxidative damage group, while the SOD and CAT viability in the preconditioning group were significantly higher than the oxidative injury group (P < .05) (Figure 7B and 7C).

Figure 7

Detection of MDA, viability of antioxidant enzymes and cells, apoptosis. A, Analysis of MDA content (n = 4). B, Analysis of SOD vitality (n = 4). C, Analysis of CAT vitality (n = 4). D, Analysis of cell viability (n = 4). E‐G, Detection of apoptosis by flow cytometry. H, Analysis of cell apoptosis (n = 3). All experiments were repeated three times. All data were presented as means ± SD. Statistical significances were calculated by the Student t test. Data were compared with control and oxidative damage groups respectively. Vs control group (# P < .05). Vs oxidative damage group (*P < .05). CAT, catalase; FITC, fluorescein isothiocyanate; MDA, malondialdehyde; SOD, superoxide dismutase

Cell viability and apoptosis

OS preconditioning can decrease intracellular ROS and MDA levels, relieve DNA and mitochondrial damage, but whether it can promote BMSCs survival under OS conditions? Therefore, we also tested cell viability and apoptosis, and the results showed that the cell viability of preconditioning group was about (73.92 ± 9.82)%, which was about 1.5 times higher than the oxidative damage group, the difference was statistically significant (P < .05) (Figure 7D), and the apoptosis of preconditioning group was about (8.75 ± 0.12)%, which was also significantly lower than the oxidative damage group, the difference was statistically significant (P < .05) (Figure 7E‐H).

Mechanism of OS preconditioning

Interaction of Nrf2 and Keap1

Nrf2 is the most important and classical signal pathway of antioxidant stress in cells. Uncoupling of Nrf2 from Keap1, and Nrf2 nuclear translocation, which promotes the expression of antioxidant enzymes genes, is essential for antioxidant stress. The results showed that when BMSCs were exposed to OS, the total Nrf2 in each group increased to some extent, compared with control and oxidative damage group, the Nrf2 in BMSCs pretreated with 50 μm H2O2 increased significantly. However, there was no significant change in the Keap1 levels. In addition, the interaction between Nrf2 and Keap1 was significantly reduced in BMSCs pretreated with 50 μm H2O2 compared with control and oxidative damage group (Figure 8B).

Figure 8

Western blot analysis of the interaction between Nrf2 and Keap1, and Nrf2 nuclear translocation. A, Western blot analysis of the purity of cytosolic proteins and nuclear proteins. B, Western blot analysis of the interaction between Nrf2 and Keap1. C, Western blot analysis of nuclear and cytoplasmic Nrf2. D, Analysis of the Nrf2 ratio in nucleus and cytoplasm. All data were presented as means ± SD. Statistical significances were calculated by the Student t test. Data were compared with control and preconditioning groups respectively. Vs control group (*P < .05). Vs preconditioning group (# P < .05). GAPDH, glyceraldehyde 3‐phosphate dehydrogenase; IP, immunoprecipitation; Nrf2, nuclear factor erythroid 2‐related factor 2

Nrf2 nuclear translocation

Subsequently, We used Western blot to analyze the purity of cytosolic proteins and nuclear proteins, and we detected the nuclear translocation of Nrf2. Nuclear extracts are free from cytoplasmic contamination and cytoplasmic extracts are free from nuclear contamination (Figure 8A). Compared with the control group, when BMSCs were exposed to OS, the content of Nrf2 in nucleus increased and the content of Nrf2 in cytoplasm decreased in each group8, however, compared with the oxidative damage group, the increase of Nrf2 in nucleus and decrease of Nrf2 in cytoplasm were most obvious in BMSCs pretreated with 50 μm H2O2, the difference was statistically significant (P < .05) (Figure 8C and 8D).

Expression of antioxidant enzymes

Antioxidant enzymes are antioxidative effector molecules of cells, and their levels represent the ability of cells to resist OS. It can be seen from our results that the levels of antioxidant enzymes such as SOD, CAT, NQO1, and HO‐1 in BMSCs pretreated with H2O2 were significantly higher than the control group and the oxidative damage group, the difference was statistically significant (P < .05) (Figure 9).

Figure 9

Western blot analysis of SOD, CAT, NQO1, and HO‐1. A, Western blot detection of SOD, CAT, NQO1, and HO‐1. B, Western blot analysis of SOD. C, Western blot analysis of CAT. D, Western blot analysis of NQO1. E, Western blot analysis of HO‐1. All data were presented as means ± SD. Statistical significances were calculated by the Student t test. Data were compared with control and oxidative damage groups respectively. Vs control group (*P < .05). Vs oxidative damage group (# P < .05). CAT, catalase; HO‐1, heme oxygenase 1; SOD, superoxide dismutase

DISCUSSION

This study provides new and powerful evidence for enhancing the anti‐OS ability of BMSCs. We used OS preconditioning for BMSCs with low concentration of H2O2 (50 μM), and then treated with high concentration of H2O2 (1000 μM). Through analysis of ROS, mitochondrial membrane potential, DNA damage, cell viability, apoptosis, interaction between Nrf2 and Keap1, Nrf2 nuclear translocation, and expression of antioxidant enzymes. It was proved that OS preconditioning can enhance the anti‐OS ability of BMSCs via the Nrf2 pathway and promote the survival of BMSCs under OS.

BMSCs have strong self renewal ability and multidirectional differentiation potential, and there is no immune rejection in BMSCs transplantation.18, 19, 20, 21 Therefore, it has become the most commonly used cell for cell therapy. However, local OS microenvironment in lesions, causing BMSCs to suffer from OS. OS can produce a large amount of ROS in cells, excessive ROS will damage biological macromolecules, such as DNA and lipids. Furthermore, ROS also impair mitochondria, and release of cytochrome c and apoptosis‐inducing factors, thereby, triggering a variety of signal transduction pathways, such as Akt, P53, and P38MAPK pathways, eventually inducing cell apoptosis.22, 23, 24, 25, 26 Some studies have shown that more than 80% of transplanted BMSCs will die from the OS induced apoptosis, which significantly reduced the effect of BMSCs transplantation.27, 28, 29 Therefore, enhancing the anti‐OS ability of BMSCs is the breakthrough point to improve the effect of transplantation. Although a large number of studies have shown that the application of antioxidants and upregulation of the expression of related antioxidant stress proteins can enhance the anti‐OS ability of BMSCs to some extent, but the effect is limited.30, 31, 32 However, we used low concentration of H2O2 (50 μM) for OS preconditioning, and then used high concentration of H2O2 (1000 μM) for oxidative damage, our results showed that OS preconditioning can reduce intracellular ROS levels and reduce the damage of ROS to DNA, lipids and mitochondria, increase the viability of SOD and CAT, enhance the antioxidant stress ability of BMSCs, and promote the survival of BMSCs under OS.

For this study, the stimulation intensity of preconditioning is the key to mobilize the protective mechanisms of cells and not to reduce cell viability. So, before the preconditioning of OS, the effect different H2O2 concentrations on the viability and apoptosis of BMSCs were studied. The results showed that the concentration of H2O2 was 50μm, which did not increase the apoptosis and had no effect on cell viability, moreover, there was no obvious change in cell morphology. So this concentration has no obvious toxicity to cells, and is the optimum H2O2 preconditioning concentration. When the concentration of H2O2 was 1000 μM, the BMSCs could be subjected to obvious OS, is the optimum oxidative damage concentration, which was consistent with the study of Román in 2017.31

There may be many mechanisms to enhance the anti‐OS of BMSCs, such as Nrf2, extracellular‐signal‐regulated kinase 1/2, Akt, and so on.33, 34 Nrf2 is one of the most important and classical antioxidant stress signaling pathways. In this signaling pathway, Nrf2 nuclear translocation, which promotes the expression of antioxidant enzymes genes, is essential for antioxidant stress.35 Therefore, we further explored whether OS preconditioning enhances the antioxidant stress ability of BMSCs through Nrf2 pathway. Our results showed that OS preconditioning can reduce the interaction between Nrf2 and Keap1, increase Nrf2 nuclear translocation, and promote the expression of antioxidant enzymes, such as SOD, CAT, NQO1, and HO‐1.

In conclusion, OS preconditioning can increase the expression of SOD, CAT, NQO1, and HO‐1 through the Nrf2 pathway, thereby decreased the intracellular ROS levels, relieved the damage of ROS to mitochondria, DNA and cell membrane, enhanced the anti‐OS ability of BMSCs, and promoted the survival of BMSCs under OS.

CONFLICT OF INTERESTS

The authors declare that there are no conflict of interests.