Sesn2 attenuates the damage of endothelial progenitor cells induced by angiotensin II through regulating the Keap1/Nrf2 signal pathway

EPCs identification and detection of Sesn2 and Nrf2 protein expression in EPCs treated with Ang-II

Firstly, we observed that the morphology of isolated cells exhibited a cobblestone pattern, which is considered to be typical morphology of EPCs (Figure 1A). Secondly, results indicated that the isolated cells expressed CD31, CD34, CD133, and VEGFR2, all of which are thought to be typical markers of EPCs (Figure 1B). Flow cytometry results showed that the positive rate of CD34, CD133 and VEGFR2 was 75.9%±1.6, 48%±2.1 and 81.9%±1.9% respectively (Supplementary Figure 1). Further, it is accepted that VEGFR2 is a classical endothelial cell marker and accepted that D34 and CD133 are classical progenitor cell markers. Therefore, the results suggested that EPCs expressing CD31, VEGFR2, CD34, and CD133 were late-EPCs.

Figure 1. Identification of EPCs and Ang-II increased Sesn2 and Nrf2 protein expression in EPCs. Identification and exposures of EPCs to Ang-II for prolonged periods of time. (A) EPCs (passage (P) 1) from human umbilical cord. Scale bars = 5 μm and Scale bars = 40 μm. (B) We used immunofluorescence to detect the measures of CD31, CD34, CD134, and VEGFR2. (C) The treatments with Ang-II caused an increased levels of Sesn2 mRNA in EPCs. (D) The levels of Nrf2 mRNA were unchanged in EPCs exposed to Ang-II. (E) The treatments with Ang-II increased the levels of Sesn2 and Nrf2 proteins in EPCs. (F) We used Hoechst 33258 to detect the apoptosis of EPCs and found that the treatments with Ang-II caused an increase in the apoptosis of EPCs. Scale bars = 5 μm. All experiments were performed in triplicate. *p < 0.05, **p < 0.01, ***p < 0.001, versus the control. Data are represented as mean +/- SEM. (G) We used Hoechst 33258 staining to detect the apoptosis of EPCs. Scale bars = 5 μm.

To investigate whether or not the expression of Sesn2 and Nrf2 in EPCs treated with Ang-II compared to untreated samples, we treated EPCs with 1 μM Ang-II over predetermined times (i.e., 0, 6, 12, 24 h). We used a dose of Ang-II according to information available in previous research [36–38]. The mRNA levels of Sesn2 and Nrf2 were detected by using qRT-PCR. The results indicated that Sesn2 mRNA expression was enhanced for a prolonged time (Figure 1C). In contrast the results suggested that Nrf2 mRNA expression had not changed in EPCs over the prolonged time (Figure 1D). Moreover, our results from Western blotting assessments revealed that the treatment of Ang-II induced increases in the protein levels of Sesn2 and Nrf2 (Figure 1E). The levels of Nrf2 proteins in nuclear of EPCs were also increased (Supplementary Figure 2D). These results showed that the entry of Nrf2 to nuclear of EPCs was enhanced significantly. Ang-II also caused the increased levels of LC-3 II and caused the decreased levels of P62 (Supplementary Figure 2A). The levels of autophagy were found to have been promoted by the treatment with Ang-II. In order to examine the role of Ang-II application in EPCs, we used Hoechst33258 to determine the apoptosis of EPCs, and we used CCK8 to detect the cell viability of EPCs. These results indicated that the apoptosis was increased and the cell viability was reduced in EPCs exposed to Ang-II for prolonged time (Figure 1F, 1G). These results indicated that autophagy, Sesn2, and Nrf2 might be involved in the dynamics of apoptosis of EPCs induced by the treatment of Ang-II. Further, there may be complex mechanisms and interactions between autophagy, Sesn2, and Nrf2 in EPCs exposed to Ang-II.

Sesn2 attenuates the apoptosis of EPCs induced by Ang-II through promoting the protein level of Nrf2

To investigate the role of Sesn2 in the EPCs treated with Ang-II, the LV3-pGLV-H1 + Puro plasmids with pcDNA-Sesn2 or control oligonucleotides (Lenti-Sesn2 and Lenti-NC) were designed such as to overexpress Sesn2 protein. The effect of inducing overexpression of Lenti-Sesn2 was confirmed by using qRT-PCR and Western blotting. The levels of Sesn2 mRNA were found to have been remarkably upregulated, and the levels of Sesn2 proteins were also significantly overexpressed (Figure 2A, 2B). We established that we achieved stable transfection of Lenti-Sesn2 and Lenti-NC in EPCs and these EPCs were used in following experiments. Recent research has reported that Keap1 induced downregulation of the levels and activity of Nrf2 [29, 30]. Therefore, the levels of Nrf2 and Keap1 proteins were measured by using Western blotting. As could be seen in Figure 2C, the treatments with Ang-II induced the decreased levels of Keap1 proteins and contrastingly increased the levels of Nrf2 proteins and the entry of Nrf2 to nuclear in EPCs (Supplementary Figure 2E). Interestingly, upregulation of Sesn2 was found to have further reduced the levels of Keap1 proteins and the entry of Nrf2 to nuclear in EPCs (Supplementary Figure 2E). We used immunofluorescence in order to further detect the levels of Keap1 proteins. The results indicated that levels of Keap1 proteins were reduced by the treatments of Ang-II. The immunofluorescence results indicated that there was downregulation of Keap1 proteins after the treatments of Ang-II, which Sesn2 upregulation consequently was able to further enhance this effect (Figure 2D). These results suggested that the levels of Nrf2 proteins were negatively regulated by Keap1 in EPCs. Thus, we postulated that upregulation of Sesn2 induced repression of Keap1 and thereby enhanced the levels of Nrf2 proteins.

Figure 2. Overexpression of Sesn2 promoted the survival of EPCs after Ang-II treatment. EPCs were transfected with Lenti-Sesn2 or NC before the treatments with Ang-II. (A) The effects of Lenti-Sesn2 were confirmed by using qRT-PCR. (B) The levels of Sesn2 proteins were upregulated significantly by the applications of Lenti-Sesn2. (C) The levels of Keap1, Nrf2, Bax, Bcl-2, Mit cyt-c, Cyto cyt-c, Cleaved-caspase 3 and GAPDH related proteins in EPCs were detected by using Western blotting. (D) The levels of Keap1 proteins in EPCs were detected by using Immunofluorescence. Scale bars = 2 μm. All experiments were performed in triplicate. *p < 0.05, **p < 0.01, ***p < 0.001, versus the control. Mit cyt-c: Mitochondria cytochrome C; Cyto cyt-c: Cytoplasm cytochrome C. Data are represented as mean +/- SEM.

The role of Sesn2 in EPCs treated with Ang-II needed to be confirmed as part of our study and research goals. The treatments with Ang-II were found to have induced apoptosis of EPCs. The treatments with Ang-II promoted the flow of cyt-c from mitochondria to cytoplasm, changed the protein levels of Cleaved-caspase3, and reduced the relative ratio of Bcl-2/Bax (Figure 2C). The promotion of these proteins meant that apoptosis was induced by Ang-II treatments in EPCs. An important finding was that overexpression of Sesn2 could alleviate apoptosis of EPCs induced by the treatments of Ang-II (Figure 2C). Besides, Sesn2 upregulation could enhanced the tube formation of EPCs. Importantly, Sesn2 upregulation also could repair the impaired tube formation of EPCs induced by Ang-II treatments (Supplementary Figure 3A). In sum, these results indicated that Sesn2 could be used to help protect EPCs from apoptosis induced by Ang-II.

Knockdown of Nrf2 can reverse the protective effect of Sesn2 on EPCs

To further investigate whether or not Sesn2 protected EPCs from Ang-II treatment through promotion of Nrf2, we designed Small interfering RNA (si-RNA) to silence the expression of Nrf2 mRNA and proteins. EPCs containing Lenti-Sesn2 or Lenti-NC were transfected with si-Nrf2. As could be seen in Figure 3A, 3B, the levels of Nrf2 mRNA were reduced significantly, and the levels of proteins also were decreased remarkably in EPCs. We used Si-Nrf2-1 in the following experiments described hereafter. Firstly, we used Western blotting to detect the apoptotic markers including such as Cleaved-caspase3, Bcl-2, and Bax. Our results indicated that upregulation of Sesn2 induced decreased levels of Cleaved-caspase3 but silencing of Nrf2 was able to induce reversal of this role in EPCs (Figure 3C and Supplementary Figure 1D). There were opposite effects with regards to the ratio of Bcl-2/Bax in EPCs (Figure 3C and Supplementary Figure 2F, 2D). Importantly, we observed that silencing of Nrf2 did not result in a reduction of the levels of Sesn2 in Lent-Sesn2 + Ang-II + si-Nrf2 treatment group. Our results also indicated that upregulation of Sesn2 could enhance the levels of Nrf2 and the entry of Nrf2 to nuclear in the Lent-Sesn2 + Ang-II + si-NC treatment group (Figure 3C and Supplementary Figure 2F). These results suggested that Sesn2 promoted the levels of Nrf2 proteins. The results from immunofluorescence assessments confirmed that although the levels of Cleaved-caspase 3were downregulated by Lenti-Sesn2, in contrast, si-Nrf2 enhanced the levels of expression of Cleaved-caspase3 in EPCs (Figure 3D). In order to verify whether or not si-Nrf2 had an influence upon the migration of EPCs, we used transwell assays. The results from these assessments indicated that the migration of EPCs were repressed by the treatments of Ang-II and indicated that Sesn2 could reverse this effect. However, silencing of Nrf2 resulted in the inhibition of the positive effect of Sesn2 on EPCs exposed to Ang-II (Figure 3E). These results revealed that Sesn2 protected EPCs from Ang-II through promoting the expression of Nrf2.

Figure 3. Knockdown of Nrf2 could reverse the protective effect of Sesn2 in EPCs treated with Ang-II. EPCs were divided into treatment groups including for a control, Ang-II, Ang-II + Lenti-Sesn2, and Ang-II+ Lenti-Sesn2 + si-Nrf2 treatment groups. (A) The levels of Nrf2 mRNA were found to have decreased in EPCs transfected with siNrf2. (B) The levels of Nrf2 proteins were found to have been reduced in EPCs transfected with si-Nrf2. (C) Measures for Sesn2, Nrf2, Cleaved-caspase3, Bcl-2, Bax, and GAPDH protein expression in EPCs were determined by using Western blotting. (D) We used immunofluorescence to detect levels of caspase3 protein in ECPs. Scale bars = 2 μm. (E) We determined the migration of EPCs by using Transwell assays. Scale bars = 5 μm. All experiments were performed in triplicate. *p < 0.05, **p < 0.01, ***p < 0.001, versus the control. Data are represented as mean +/- SEM.

Autophagy acts as an important mediator in Sesn2/Nrf2 sighing pathway

An important aspect of our research was to determine how Sesn2 promotes Nrf2 in EPCs as previously this was unknown. Previously, autophagy has been reported to serve as a mediator between Sesn2 and Nrf2. In the process of autophagy, Sesn2 was found to have induced the degradation of Keap1 and thereby promoted Nrf2 signaling through P62-dependent autophagy [33, 39]. To confirm the role of autophagy between Sesn2 and Nrf2, we used chloroquine (CQ, 10 μM) to treat EPCs transfected with either Lenti-Sesn2 or Lenti-NC. Our results indicated that CQ treatments alone were able to repress the levels of autophagy flux and influenced the dynamics of the Keap1/Nrf2 pathway. Restraint of autophagy was capable of inducing the impaired viability of EPCs (Figure 4A) [40, 41]. Accordingly, our results indicated that upregulation of Sesn2 was able to induce autophagy. Treatment with CQ further inhibited autophagy that was induced by overexpression of Sesn2 (Figure 4A). Chloroquine acted as an inhibitor of lysosomes and inhibited the fusion of autophagosomes and lysosomes. In this manner, the levels of LC-3II were increased as the autophagic flux was inhibited. The levels of expression of P62 confirmed such findings (Figure 4A). Interestingly, our results indicated that Keap1 was negatively expressed in EPCs when Sesn2 was correspondingly overexpressed. And, then consequently, Nrf2 protein expression and the entry to nuclear was increased in EPCs (Figure 4A and Supplementary Figure 3B). However, our results indicated that CQ inhibited autophagic flux induced by Sesn2, and then inhibited the degradation of Keap1. Further, Keap1 repressed the levels of Nrf2. Meanwhile, Keap1 combined with Nrf2and thereby resulting in the retention of Nrf2 in cytoplasm (Figure 4A and Supplementary Figure 3B). We used immunofluorescence to determine the distribution of Nrf2 in EPCs. The results clearly indicated that Sesn2 promoted Nrf2 to enter the nucleus of EPCs while CQ reversed this phenomenon (Figure 4B). Finally, we evaluated the effect of autophagy upon the migration of EPCs and found that CQ could reduce the promotion of migration of EPCs by Sesn2. These results demonstrated that autophagy might have acted as an important mediator in the Sesn2/Nrf2 signaling pathway.

Figure 4. Repression of autophagy inhibited the protective effect of Sesn2 in EPCs. EPCs were divided into treatment groups representing the control, CQ, Lenti-Sesn2, and Lenti-Sesn2 + CQ cohorts. (A) The levels of expression of proteins of Keap1, Nrf2, Cleaved-caspase3, Cleaved-caspase9, P62, LC-3, and GAPDH in EPCs were detected by use of Western blotting. (B) The levels of expression of proteins of Nrf2 in EPCs were determined by using Immunofluorescence. Scale bars = 2 μm. (C) Measures of the migration of EPCs were determined by using Transwell assays. All experiments were performed in triplicate. Scale bars = 5 μm. *p < 0.05, **p < 0.01, ***p < 0.001, versus the control. Data are represented as mean +/- SEM.

To further determine whether or not autophagy mediated the Sesn2/Nrf2 signaling pathway, we designed si-RNA to silence the expression of Sesn2. As shown in Supplementary Figure 2B, 2C, the levels of Sesn2 mRNA and proteins were reduced dramatically. We used si-Sesn2-1 in the following experiments described hereafter. Silencing of Sesn2 was found to have inhibited the upstream dynamics of autophagic flux, whereby we observed a decrease in LC-3 and an increase in P62 (Figure 5A). Meanwhile, the levels of Keap1 proteins were enhanced whereas the levels of Nrf2 and the entry of Nrf2 to nuclear were reduced significantly (Figure 5A and Supplementary Figure 3C). The rates of EPCs apoptosis also were found to have been increased when Sesn2 was silenced by si-Sesn2 (Figure 5A). Interestingly, CQ was found to have inhibited the fusion of autophagosomes and lysosomes, which ultimately resulted in the inhibition of the downstream dynamics of autophagy. Thus, the levels of P62 and LC-3 were increased when EPCs were treated with CQ (Figure 5A). The CQ treatment was found to have further suppressed Nrf2 levels and the entry to nuclear because the CQ treatment reduced the degradation of Keap1 through P62-dependent autophagic flux (Figure 5A and Supplementary Figure 3C). Many studies have already reported that as an antioxidant transcription factor, Nrf2 can activate many antioxidant enzymes such as to repress ROS levels in cells [42–44]. In our experiments, we found that ROS levels were increased remarkably in EPCs transfected with si-Sesn2 and found that CQ enhanced this effect (Figure 5B). Moreover, silencing of Sesn2 was found to have also inhibited the migration of EPCs, and CQ further repressed the migration of EPCs (Figure 5C). In summary, these results indicated that Sesn2 upregulated the Keap1/Nrf2 signaling pathway through inducing autophagy.

Figure 5. Sesn2 protected the EPCs through regulating the autophagic flux. EPCs were divided into treatment groups representing the control, si-Sesn2, and si-Sesn2 + CQ. (A) The levels of P62, LC-3, Cleaved-caspase3, Keap1, Nrf-2, and of GAPDH proteins in EPCs. (B) Measures of ROS production in EPCs were assessed by the use of fluorescent dye H2DCFDA. Scale bars = 5 μm. (C) Measures of the migration of EPCs were determined by using Transwell assays. Scale bars = 20 μm. All experiments were performed in triplicate. Scale bars = 5 μm. *p < 0.05, **p < 0.01, ***p < 0.001, versus the control. Data are represented as mean +/- SEM.

Upregulation of Nrf2 promotes the viability and tube formation ability of EPCs

To confirm the role of Nrf2 in Ang-II-treated EPCs, the LV3-pGLV-H1 + Puro plasmids with pcDNA-Nrf2 or control oligonucleotides (Lenti-Nrf2 and Lenti-NC) were designed to overexpress levels of Nrf2 proteins. As shown in Figure 6A, the levels of Nrf2 mRNA were significantly increased. Nrf2 protein expression also was enhanced significantly by the treatment of Lenti-Nrf2 (Figure 6B). Our previous results indicated that Ang-II treatments was able to facilitate apoptosis of EPCs. In this section, our approach was to have transfected EPCs with Lenti-Nrf2 or Lenti-NC before the treatments with Ang-II. Overexpression of Nrf2 was able to induce the alleviation of apoptosis of EPCs that had been caused by the treatments of Ang-II. Nrf2 was found to have reduced the leaking of cyt-c from mitochondria to the cytoplasm. Nrf2 also caused a decrease of Cleaved-caspsae3 and an increase of Bcl-2/Bax (Figure 6C). Nrf2 usually serves as an antioxidant transcription factor which helps to inhibit ROS production in cells. Therefore, we detected the levels of ROS by use of the fluorescent dye H2DCFDA. Our results indicated that Nrf2 could reduce the levels of ROS caused by the Ang-II treatments. We found that green staining positive cells were reduced significantly in the Ang-II + Lenti-Nrf2 treatment groups (Figure 6D). It is understood that tube formation and migration abilities reflect the viability of EPCs. Thus, in order to investigate whether or not Nrf2 promoted tube formation and the migration ability of EPCs treated with Ang-II, we performed Matrigel angiogenesis assays and transwell assays. Our results indicated that Ang-II repressed the tube formation of EPCs and indicated that upregulation of Nrf2 could reverse this effect (Figure 6E). The results of transwell assays indicated that Ang-II inhibited the migration of EPCs and indicated that overexpression of Nrf2 could reverse this effect (Supplementary Figure 3D). These results suggested that upregulation of Nrf2 attenuated Ang-II-induced inhibition of the viability and the tube formation ability in EPCs.

Figure 6. Upregulation of Nrf2 inhibited Ang-II-induced apoptosis and dysfunction of EPCs. EPCs were divided into treatment groups representative of the control, Ang-II, and Ang-II + Lenti-Nrf2. (A) The levels of Nrf2 mRNA were found to have increased significantly in EPCs transfected with Lenti-Nrf2. (B) The levels of Nrf2 proteins were found to have increased significantly in the Nrf2 samples transfected with Lenti-Nrf2. (C) The levels of for Bax, Bcl-2, Mit cyt-c, Cyto cyt-c, Cleaved-caspase 3, and GAPDH proteins in EPCs were detected by using Western blotting. (D) ROS production in EPCs was detected by the use of fluorescent dye H2DCFDA. Scale bars = 5 μm. (E) Tube formation abilities were assessed on Matrigel. All experiments were performed in triplicate. Scale bars = 5 μm. *p < 0.05, **p < 0.01, ***p < 0.001, versus the control. Data are represented as mean +/- SEM.

Nrf2 protects EPCs from Ang-II through inhibiting ROS production

It is known that Nrf2 plays an important role in oxidative stress and that Nrf2 protects tissues and cells from oxidative stress. The results of our analyses using Western blotting indicated that silencing of Nrf2 was able to increase the autophagy and apoptosis. Further, knocking-down of Nrf2 caused an increase of Cleaved-caspse3 and caused a decrease of the Bcl-2/Bax ratios (Figure 7A). Similarly, knocking-down of Nrf2 induced an increase in the levels of expression of Beclin-1 and LC-3 (Figure 7A). Thus, we postulated that promotion of autophagy might have played a protective role in EPCs when Nrf2 was silenced.

Figure 7. NAC facilitated the survival and migration of EPCs after silencing of Nrf2. EPCs were divided into treatment groups representative of the control, si-Nrf2 and, si-Nrf2 + NAC. (A) The levels of Cleaved-caspase3, Bax, Bcl-2, Beclin-1, LC-3, and GAPDH proteins in EPCs were determined by using Western blotting. (B) We used Hoechst 33258 staining to detect the apoptosis of EPCs. Scale bars = 5 μm. (C) Migration of EPCs was determined by the application of wound healing assays (details given in methods). All experiments were performed in triplicate. Scale bars = 20 μm. *p < 0.05, **p < 0.01, ***p < 0.001, versus the control. Data are represented as mean +/- SEM.

NAC (N-acetyl-L-cysteine, 10mM) [45] was used to investigate whether or not Nrf2 protected EPCs by suppressing ROS production. EPCs transfected with si-Nrf2 were treated with NAC. As can be seen in Figure 7A, Cleaved-caspase3 was found to have decreased and Bcl-2/Bax was found to have increased, indicating that apoptosis of EPCs induced by silencing of Nrf2 was reduced by NAC. Our results also indicated that autophagy was decreased in EPCs transfected with si-Nrf2. The results of Hoechst33258 confirmed that NAC attenuated the apoptosis of EPCs (Figure 7B). Importantly, our results indicated that NAC was capable of facilitating the migration and the tube formation ability of EPCs transfected with si-Nrf2 (Figure 7C and Supplementary Figure 3E). These results demonstrated that Nrf2 protected EPCs through inhibiting ROS production.

EPC isolation and culture from human umbilical cord blood

We isolated EPCs from samples of human umbilical cord derived blood by using density gradient centrifugation with Histopaque-1077 (Sigma). We gently added the samples of human blood to tubes pre-filled with separation solution and then centrifuged samples at 2000 rpm for 20 min at 4° C. After centrifugation, there were three resultant layers based upon differences in densities of constituents, including the uppermost layer which contained serum, the middle layer which contained white blood cells, and the lowermost layer which was contained red blood cells. During this process, the middle layer becomes suspended within the separation liquid. We removed white blood cells and constituents from the lowermost layer and individually resuspended them in complete endothelial cell growth medium (EGM)-2 (Lonza), followed by a final re-seeding of these samples into wells of six-well plates. We changed the medium every 72 hours, and we subcultured ECPs were according to 1:3 ratios cells to cells. We detected known biomarkers for EPCs by using immunofluorescence. For Immunofluorescence steps, we used primary antibodies against CD34, CD31, endothelial growth factor receptor 2 (VEGFR2), and CD133 which we purchased from Cell Signaling Technology. And then we detected the marker including CD34, CD133 and VEGFR2. The positive rate of CD34, CD133 and VEGFR2 was 75.9%±1.6, 48%±2.1 and 81.9%±1.9% respectively (Supplementary Figure 1). All aspects of our experiments were reviewed and approved by the Ethics Committee of Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine. Further, all aspects of our experiments were performed according to and following all guidelines in the Helsinki Declaration. The IRB number for our study was XHEC-D-2018-368. Samples of blood derived from umbilical cord material was obtained from a single donor, and corresponding information attributed to the donor is shown in Table 1.

Flow cytometry

Took newly isolated human umbilical cord blood mononuclear cells, added FITC-labeled anti-human CD34, PE-labeled anti- human VEGFR2 and APC-labeled anti-human 2 μL each, and sent the flow cytometer to the machine Detection.

Western blotting

We extracted total proteins from ECPs by using RIPA (Radioimmunoprecipitation lysis buffer, Beyotime, China) containing PMSF (Phenylmethanesulfonyl fluoride, Beyotime, China). Concentrations of protein were measured by using the bicinchoninic acid (BCA) method (Beyotime, China) following all manufacturer protocols. Proteins were separated by using 10-15% SDS-PAGE. We electrophoretically transferred proteins to polyvinylidene difluoride (PVDF) membranes (Millipore, USA). Membranes were blocked with 5 % skim milk and were then incubated with primary antibodies for: Sesn2 (Proteintech, USA), Nrf2 (Proteintech, USA), Keap1 (Abcam, UK), Bax (Proteintech, USA), Bcl-2 (Proteintech, USA), Cytochrome c (Proteintech, USA), Cleaved-caspase 3 (Proteintech, USA), Cleaved-caspase9 (Abcam, UK), P62 (Proteintech, USA), LC-3 (Abcam, UK), Beclin-1 (Proteintech, USA), CD31 (Abcam, UK), CD34 (Abcam, UK), CD133 (Abcam, UK), VEGFR2 (Abcam, UK), Lamin B (Proteintech, USA) and GAPDH (Beyotime, China). We next incubated the membranes with secondary antibodies. Results for the membranes post-incubation periods were visualized by using chemiluminescence (Millipore).

Real-time quantitative PCR

We extracted total RNA from EPCs by using TRIzol (TakaRa, China) according to all manufacturer protocols. For mRNA, we carried out reverse transcription and quantitative Real-Time polymerase chain reaction (qRT-PCR) as has been previously described. Briefly, we used extracted RNA for reverse transcription to produce cDNA by using a Prime-Script RT reagent kit with gDNA Eraser (Takara Biotechnology Co., Ltd., Japan) according to all manufacturer protocols. We carried out qRT-PCR out on a Roche LightCycler by using SYBR® Premix Dimer Eraser™ (Perfect Real Time; Takara, Japan) with coupled cDNA synthesis for the desired genes. The sequences for primer set we used are listed as follows: Sesn2 Forward: 5’-CTCATCACCAAGGAACACATC-3’, Reverse: 5’-CTCTGTTCACTAGGGGGTGTAG-3’; and Nrf2 Forward: 5’- ACGGTATGCAACAGGACATTGAGC-3’, Reverse: 5’ TTGGCTTCTGGACTT|GGAACCATG-3’.

Cell viability assays

We used CCK-8 (Beyotime, China) to determine cell viability after treatments of Ang-II. EPCs (1 × 10⁴ cells/well) were suspended in 100 μL of fresh medium containing Ang-II and were then seeded. After applying the Ang-II treatments, we added CCK-8 (10 μL/well) the mix with freshened medium and incubated samples for 3 h. Next, we determined measures of the absorbance at 450 nm.

Transwell assays

We used transwell assays in order to determine measures of the migration of EPCs. Complete fresh EGM-2 containing treatments of either Ang-II or CQ (each 600 μL) was added to the bottom chambers. We then added200 μL of serum-free medium containing 5 × 10⁴ EPCs to the top chambers. After incubating at 37° C for 12 h, EPCs transmigrating to membranes were fixed by using 4 % paraformaldehyde and were then stained with crystal violet. We selected three random microscopic fields to count stained cells.

Wound healing assays

We used approximately 1 × 10⁶ EPCs per sample seeded into six-well plates and then incubated at 37° C. When EPCs reached a level of confluence of at least 90 %, we manually generated wounds by scratching cell monolayers with a sterile 200 μL plastic pipette tip and then incubated the samples in fresh medium containing 1 % FBS for 24 h. We took photographs to assess the mean distance of migration per field. The measured area of the remaining unhealed scratch wounds was calculated as follows: migration area (%) = (A0 – An)/A0 × 100, where A0 represents the area of the initial wound and an represents the remaining area of wound at the time step the corresponding measurement was made.

Matrigel angiogenesis assays

Matrigel^TM (100 μL; BD Biosciences) was paved in the wells of 96-well plates and incubated at 37° C for 30 min. We then seeded 2 × 10⁴ EPCs on the Matrigel and incubated at 37° C for 8 h. Next, we photographed the results and produced images under magnification using microscopy (Leica, Germany), and analyzed the results by using Images J software.

Immunofluorescence analysis

We seeded EPCs upon glass coverslips in a six-well plate and then fixed the samples by using 4 % paraformaldehyde diluted in PBS for 30 min. EPCs were permeabilized with 0.1% Triton-X-100 in PBS for 15 min, and subsequently blocked in 5 % goat serum for 1 h. We incubated EPCs in primary antibodies overnight at 4° C including at all of the following dilutions: CD31 (1:100, CST, USA), CD34 (1:200, CST, USA), CD133 (1:200, Abcam, UK), VEGFR2 (1:100, Abcam, UK), Keap1 (1:400, Proteintech, USA), Nrf2 (1:200, Proteintech, USA) and Caspase3 (1:200, CST, USA). We washed EPCs with PBST three times and then incubated the samples in Cy3-conjugated AffiniPure goat Goat anti-rabbit Anti-Rabbit IgG (H+L) and Alexa Fluor 488-conjugated AffiniPure goat anti-mouse Goat Anti-Mouse IgG (H+L). Measures of immunofluorescence were determined by using fluorescence microscopy (Olympus BX51).

ROS measurement

We measured the levels of ROS production by using a Reactive Oxygen Species Assay Kit (Beyotime, China) following all manufacturer protocols. Briefly, we washed EPCs with PBS and incubated the samples with serum-free medium containing with DCFH-DA at 37° C for 20 min. Then the DCFH-DA in plates was carefully removed and discarded and we then washed EPCs with serum-free medium three times. We produced images for analysis using photography under microscopy (Olympus Fluoview, Japan) whereby positive cells were found to have emitted green color at various intensities.

Hoechst 33258 staining

Apoptotic EPCs were identified by using Hoechst 33258 staining (Beyotime, China). We seeded EPCs at densities of 6 × 10⁵ cells/well in wells of a 60 mm plate. After applying the variously composed treatments, we fixed EPCs with 4 % paraformaldehyde for 10 min, then washed samples with PBS three times, followed by staining with 2 μg/mL Hoechst 33258 (Beyotime, China) in Hank’s balance salt solution for 5 min. Changes in the morphology of apoptotic nuclei were evaluated under fluorescence using microscopy (Olympus BX51) at an absorbance wavelength of 460 nm.

Small interfering RNA transfection

Small interfering RNAs specifically targeting Sesn2 and Nrf2 were designed by and purchased from Gene Pharma, China. Si-Sesn2 and si-Nrf2 sequences are shown in Table 2. We mixed Si-Sesn2 and si-Nrf2 with Lipofectamine 2000 (Invitrogen, Carlsbad, California, USA) for 20 min in the serum-free medium. We then next added the mixtures to the plates.

Lentivirus transfection

The LV3-pGLV-H1+Puro plasmids with pcDNA-Sesn2 (Lenti-Sesn2), The LV3-pGLV-H1+Puro plasmids with pcDNA-Nrf2 (Lenti-Nrf2), and control Lentivirus (Lenti-NC) were all designed and synthesized by Gene Pharma, China. Briefly, we transfected ECPs with Lentivirus particle following all manufacturer protocols. The mRNA and protein levels of expression Sesn2 and Nrf2 were determined by using qRT-PCR and Western blotting.

Statistical analyses

All data were presented as the mean ± standard deviation (SD). We used GraphPad Prism Software to analyze all data and statistical analyses. We used Student’s t-tests to analyze comparisons of the measures between two treatment groups. For multiple comparisons among multiple treatment groups, we used two-way analysis of variance (ANOVA) followed by Tukey’s post-hoc tests. The levels of statistical significance was at a P value of *p < 0.05, and values of **p < 0.01 and ***p < 0.001 were considered as increasingly statistically significant.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author upon request.