Reactive oxygen species regulate the quiescence of CD34-positive cells derived from human embryonic stem cells.

Cardiovascular Research Advance Access published April 18, 2014

1

Reactive Oxygen Species Regulate the Quiescence of CD34-positive Cells Derived from Human Embryonic Stem Cells

Sun-Hwa Song1, Kyoungjong Kim1, Jeong Joo Park1, Kyung Hoon Min2, and Wonhee Suh1 1

College of Pharmacy, Ajou University, Suwon, Gyeonggi-do 443-749, Korea

2

College of Pharmacy, Chung-Ang University, Seoul, 156-756, Korea

Woncheon-Dong, Yeongtong-Gu, Suwon, Gyeonggi-do 443-749, Korea, Tel: 82-31-219-3445;

Accepted Manuscript

Fax: 82-31-219-3675; E-mail: [email protected]

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2014. For permissions please email: [email protected].

Downloaded from http://cardiovascres.oxfordjournals.org/ at University of New Hampshire on June 24, 2014

Corresponding author: Wonhee Suh, PhD, College of Pharmacy, Ajou University, San 5,

2

Abstract Aims Reactive oxygen species (ROS) are involved in a wide range of cellular processes. However, few studies have examined the generation and function of ROS in human embryonic vascular development. In this study, the sources of ROS and their roles in the vascular differentiation of human embryonic stem cells (hESCs) were investigated. Methods and Results During vascular differentiation of hESCs, CD34+ cells had quiescence-related

ROS, which were primarily generated through NOX4, were substantially higher in hESC-derived CD34+ cells than in hESC-derived CD34- cells. To determine whether excess levels of ROS induce quiescence of hESC-derived CD34+ cells, ROS levels were moderately reduced using selenium to enhance antioxidant activities of thioredoxin reductase and glutathione peroxidase. In comparison to

Accepted Manuscript

untreated CD34+ cells, selenium-treated CD34+ cells exhibited changes in gene expression that favored cell cycle progression, and had a greater proliferation and a smaller fraction of cells in G0 phase. Thus, selenium treatment increased the number of hESC-derived CD34+ cells, thereby enhancing the efficiency with which hESCs differentiated into vascular endothelial and smooth muscle cells. Conclusion This study reveals NOX4 produces ROS in CD34+ cells during vascular differentiation of hESCs, and shows that modulation of ROS levels using antioxidants such as selenium may be a novel approach to increase the vascular differentiation efficiency of hESCs.

Keywords: CD34, human embryonic stem cells, reactive oxygen species, selenium, vascular differentiation


gene expression profiles and a large fraction of these cells were in G0 phase. In addition, levels of

3

Introduction Human embryonic stem cells (hESCs), which are derived from the inner cell mass of a blastocyst, have a high self-renewal capacity and the potential to differentiate into all three germ layers. HESCs are a powerful tool for investigating the molecular mechanisms underlying human embryonic development1. In the vascular biology field, many researchers have used hESCs to study human embryonic vascular development in vitro. In terms of regenerative medicine, understanding and

sufficient quantities of vascular cells for potential therapeutic applications. During hESC differentiation, vascular cells, such as vascular endothelial cells (ECs) and smooth muscle cells (SMCs), are differentiated from common precursors called vascular progenitors or angioblasts. The emergence of vascular progenitors is critical during vascular differentiation of hESCs.

Accepted Manuscript

Therefore, many studies have isolated hESC-derived vascular progenitors and characterized their molecular and cellular properties to provide insight into the poorly defined mechanism underlying human embryonic vascular development2-4. In this regard, we previously identified genes that are differentially expressed in hESC-derived vascular progenitors by comparing the gene expression profiles of hESC-derived vascular progenitors and other differentiated cells5. Our previous results showed that biological processes involving mitosis and biogenesis of the ribonucleoprotein complex are substantially repressed in hESC-derived vascular progenitors. In particular, genes involved in the cell cycle, chromosome segregation, and spindle organization are considerably downregulated in hESC-derived vascular progenitors, suggesting that these cells do not proliferate and remain quiescent during vascular differentiation. Cell division is highly regulated by the cellular reduction and oxidation (redox) environment. The cellular redox status is influenced by the production and removal of reactive oxygen species (ROS); therefore, ROS play an important role in cell division. ROS are commonly thought to induce cell cycle arrest due to the oxidation of macromolecules. By contrast, low levels of ROS can serve as second messengers in mitogenic signaling cascades that promote cell cycle progression6. This dual action of ROS as negative and positive regulators of cell division, varies according to the ROS


optimizing the differentiation of hESCs into vascular lineages should facilitate the generation of

4

concentration, how quickly ROS are scavenged, and the cell types. Given the critical roles of ROS in cell proliferation and cell cycle arrest, we hypothesized that the quiescence of hESC-derived vascular progenitors is associated with ROS that are generated during hESC differentiation. In this study, we examined the sources and generation of ROS during vascular differentiation of hESCs and the effect of antioxidant selenium on the cell cycle of CD34+ cells containing vascular progenitors 5, 7. Elucidation of these processes might contribute to the optimization of the conditions required for

Materials and methods Cell culture H9 hESCs (Wicell Research Institute, Madison, WI) were grown on mitomycin-treated mouse

Accepted Manuscript

embryonic fibroblasts in hESC culture medium containing DMEM/F12 (Gibco, Grand Island, NY) supplemented with knockout serum replacement (20%; Gibco), penicillin-streptomycin (1%; Gibco), β-mercaptoethanol (0.1 mM; Gibco), non-essential amino acid solution (1%; Gibco), and basic fibroblast growth factor (bFGF, 5 ng/mL; ProSpec, Rehovot, Israel). Vascular differentiation of hESCs H9 cells were cultured on Matrigel (BD Biosciences, Bedford, MA) for 2 days before human embryoid bodies (hEBs) were formed. HEBs were differentiated in medium I [DMEM/F12 containing knockout serum replacement (10%), vascular endothelial growth factor (VEGF, 10 ng/mL; R&D System, Minneapolis, MN), bone morphogenetic protein 4 (BMP4, 10 ng/mL; ProSpec), bFGF (5 ng/mL; ProSpec) and activin-A (3 ng/mL; ProSpec)] under hypoxic conditions (3% O2) for 8 days. HEBs were then attached to a gelatin-coated dish and further differentiated for an additional 7 days in medium II [endothelial basal medium (Lonza, Walkersville, MD) containing VEGF (50 ng/mL), BMP-4 (20 ng/mL), and bFGF (5 ng/mL)] under normoxic conditions5, 7. For vascular differentiation of hESCs with selenium supplementation, sodium selenite (20 or 50 ng/mL; Sigma, St. Louis, MO) was added to medium I and II.


vascular differentiation of hESCs, which would enhance the efficiency of vascular differentiation.

5

Reverse transcription-polymerase chain reaction (RT-PCR) Total RNA was isolated from cells using TRIzol (Invitrogen, Carlsbad, CA) and cDNA was synthesized using the Superscript First-Strand Synthesis System (Invitrogen) and amplified through 25-35 PCR cycles with gene-specific primers. Densitometric analysis of PCR bands was performed with Image-Lab (MCM design, Birkerød, Denmark). For quantitative real-time PCR analysis, cDNA was analyzed using SYBR Green PCR master mix (Applied Biosystems, Forster City, CA). Data were

encoding glyceraldehyde-3-phosphate dehydrogenase (GAPDH). All reactions were performed in triplicate. Primer sequences are listed in Supplementary Table I. Flow cytometry analysis To separate CD34+ and CD34- cells differentiated from hESCs, hEBs were dissociated into single cells using dissociation buffer (ReproCell, Yokohama, Japan) and dispase (Roche, Mannheim,

Accepted Manuscript

Germany), and labeled with phycoerythrin (PE)-conjugated anti-CD34 IgG (Dako Inc., Carpinteria, CA). CD34+ and CD34- cells were purified using a fluorescence-activated cell sorting (FACS) Aria II cell sorter and CellQuest acquisition software (BD Biosciences). Controls were stained with appropriate isotype-matched nonspecific IgGs. Cell surface expression of CD34, CD31, and SM22α was analyzed using an Accuri flow cytometer (BD Biosciences). Cells pretreated with human FcR blocking reagent (Miltenyi Biotec, Bergisch Gladbach, Germany) were labeled with fluorescenceconjugated primary IgGs against CD34 (Dako Inc.), CD31 (BD Biosciences), or SM22α (Abcam, Cambridge, MA) for 1 h at 4°C and washed with ice-cold buffer. Cell cycle analysis Sorted cells were suspended in nucleic acid staining solution (Gibco) and stained with 10 µg/mL 7aminoactinomycin D (AAD; eBioscience, San Diego, CA) and 1 µg/mL pyronin Y (Sigma). After washing in cold PBS, cells were kept on ice during flow cytometry analysis. Quiescent cells (G0) were identified as the population with a 2N DNA content and a lower RNA content than cells in S phase.


analyzed using the ∆∆ Ct method and normalized against values obtained for the housekeeping gene

6

Cell proliferation analysis Cells were fixed and labeled with primary IgGs against Ki-67 (BD Biosciences) and phospho Histone 3 (pH3; Millipore, Billerica, MA). Nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Burlingame, CA). For quantitative analysis, five representative color images were randomly acquired per sample. Measurement of ROS

diacetate (H2DCFDA; Invitrogen). Briefly, cells were incubated in serum-free medium containing 20 µM H2DCFDA at 37°C for 20 min. After washing several times, images of the cells were acquired using a fluorescence microscope (Nikon, Tokyo, Japan). The relative H2DCFDA fluorescence intensity (corrected for background fluorescence) was measured using NIS-Elements BR 3.1 software.

Accepted Manuscript

Fluorescence emitted from H2DCFDA was also analyzed using an Accuri flow cytometer. For experiments with ROS inhibitors, cells were pretreated with N-acetylcysteine (NAC, 100 µM; Sigma), diphenyleneiodonium (DPI,10 µM; Sigma), Nw-nitro- L-arginine methyl ester (L-NAME, 100 µM; Calbiochem, Billerica, MA), apocynin (100 µM; Sigma), allopurinol (100 µM; Sigma), FCCP (60 µM; Sigma), auranofin (AF, 5 µM; Sigma), aurothioglucose (ATG, 20 µM; Sigma), PX12 (50 µM; Sigma), mercaptosuccinic acid (MSA, 5 µM; Sigma), or catalase (3,000 units/mL; Sigma) at 37°C for 5–30 min. Cells pretreated with hydrogen peroxide (100 µM; Sigma) were included as positive controls. For negative controls, cells were not incubated with H2DCFDA. Immunofluorescence staining Cells were fixed in 4% paraformaldehyde (Sigma), permeabilized with 0.5% Triton X-100 (Sigma), and blocked in 10% normal goat or rabbit serum (Vector Laboratories Inc.). Cells were incubated with primary IgGs against phospho-p53 (Ser15, Cell Signaling Technology, Danvers, MA), CD34 (Abcam), CD31 (R&D Systems), vascular endothelial (VE)-cadherin (R&D Systems), α-fetoprotein (AFP; R&D System), β-III tubulin (Chemicon International Inc., Temecular, CA), or α-smooth muscle actin (α-SMA; R&D Systems), or with irrelevant non-specific IgGs. The cells were then


Intracellular ROS levels were measured using 5,6-chloromethyl-2ʹ,7ʹ-dichlorodihydrofluorescein

7

incubated with Alexa Fluor 488- or 594-conjugated secondary IgGs (Invitrogen) and imaged using a fluorescence microscope (Nikon). Nuclei were stained with DAPI. Small interfering RNA (siRNA) transfection To silence the gene expression, hEBs were differentiated for 15 days and transfected with 50-200 nM of siRNA (Bioneer, Daejeon, Korea) using LipofectamineTM 2000 (Invitrogen). RT-PCR was performed to analyze gene expression 24 h after transfection. The sequences of the siRNAs are listed

Glutathione Peroxidase (GPx) assay Cells were homogenized in cold buffer containing Tris-HCl (50 mM, pH7.5), EDTA (5 mM), and dithiothreitol (1 mM). The cell homogenate was centrifuged and the supernatant was mixed with assay buffer (GPx Assay Kit; Cayman Chemical, Ann Arbor, MI) containing NADPH, glutathione, and glutathione reductase. Cumene hydroperoxide was added to initiate the reaction and the

Accepted Manuscript

absorbance at 340 nm was measured every minute, with measurements obtained for at least five time points. Samples pretreated with MSA (100 µM) were included to measure non-specific NADPH oxidation. GPx activity was expressed as nanomoles of NADPH oxidized per minute and milliliter of sample. Thioredoxin reductase (TRxR) assay Cells were homogenized in cold buffer containing potassium phosphate (50 mM; pH7.4) and EDTA (1 mM). The cell homogenate was centrifuged and the supernatant was mixed with assay buffer (TRxR Assay Kit; Cayman Chemical). NADPH and 5,5ʹ-dithiobis-(2-ninitrobenzoic acid) were added to initiate the reaction, and the absorbance at 405 nm was measured every minute, with measurements obtained for at least five time points. Samples pretreated with sodium aurothiomalate (20 µM) were included to measure non-specific TrxR activity. TrxR activity was expressed as micromoles of NADPH oxidized per minute and milliliter of sample. Statistical analysis All data are presented as mean values ± standard errors of the means (SEM). One-way analysis of variance, followed by Bonferroni’s post hoc multiple comparison test, was used to determine the


in Supplementary Table II.

8

differences between multiple groups. Unpaired Student’s t-test was used for comparison between two groups. P < 0.05 was considered as statistically significant. The n number indicates the number of samples in each experiment.

Results Quiescent hESC-derived CD34+ cells

progenitors, real-time RT-PCR analysis of hESC-derived CD34- cells and CD34+ cells containing vascular progenitors was performed. These cells were isolated using FACS from hESCs that had undergone vascular differentiation for 15 days (Figure 1A)7. A G1 checkpoint inhibitor (p21) that induces cell cycle arrest was markedly upregulated, while several genes associated with cell cycle

Accepted Manuscript

progression (CCNB1, CDC20, CDK1, and PCNA) were downregulated in hESC-derived CD34+ cells compared to hESC-derived CD34- cells (Figure 1B). In addition, the number of phospho-p53+ cells was much greater in CD34+ cell than in CD34- cells. Consistent with this, cell cycle analysis showed that 41.0 % of hESC-derived CD34+ cells were in G0 phase compared to 9.1% of hESC-derived CD34- cells (Figure 1D). Figure 1E and Supplementary Figure I also showed that the percentage of Ki-67+ or pH3+ cells was significantly lower in the hESC-derived CD34+ cell population than in the hESC-derived CD34- cell population. Although Ki-67 and pH3 are not perfect markers of cell division or proliferation, overall results indicate that a large proportion of hESC-derived CD34+ cells are not actively dividing, remaining in a quiescent state during vascular differentiation of hESCs. Generation of ROS in hESC-derived CD34+ cells Cell cycle progression is controlled by cellular redox status. In particular, p21 has been demonstrated to be highly induced by ROS and controls cell cycle progression in a redox-sensitive manner8. Thus, we hypothesized that ROS are associated with the quiescent state of hESC-derived CD34+ cells. First, we examined ROS generation during vascular differentiation of hESCs. The fluorescence intensity of H2DCFDA, an intracellular ROS indicator that becomes fluorescent upon oxidation by ROS, significantly increased during vascular differentiation of hESCs (Figure 2A). The


To examine the expression levels of cell cycle-related genes in hESC-derived vascular

9 cells with fluorescent H2DCFDA were mainly at the centers of hEBs, the region in which CD34+ cells are primarily located. Immunofluorescence analysis of hEBs revealed that H2DCFDA fluorescence colocalized with PE-conjugated anti-CD34 IgG fluorescence (Figure 2B). In cells isolated from differentiated hEBs, the fluorescence intensity of H2DCFDA was significantly higher in CD34+ cells than in CD34- cells (Figure 2C, Supplementary Figure II). This suggests that ROS are mainly produced in CD34+ cells during vascular differentiation of hESCs.

ROS are generated from a number of sources, including the mitochondrial electron transport system, xanthine oxidase, NOX, and uncoupled nitric oxide synthase (NOS)9. We examined the source of ROS produced during vascular differentiation of hESCs using pharmacological inhibitors that selectively block pathways of ROS generation. The dose of each inhibitor was optimized such

Accepted Manuscript

that ROS generation was transiently inhibited without markedly affecting cell viability. Treatment with NAC, a flavoprotein inhibitor (DPI), or a NOX inhibitor (apocynin) significantly reduced the level of ROS in hESC-derived CD34+ cells (Figure 3A). However, treatment with a xanthine oxidase inhibitor (allopurinol), a mitochondrial oxidative phosphorylation inhibitor (FCCP), or an endothelial NOS inhibitor (L-NAME) did not affect ROS generation in hESC-derived CD34+ cells. These data suggest that NOX is a major source of ROS in hESC-derived CD34+ cells. To further determine the isoforms of NOX involved in ROS generation, the expression profiles of five human NOX isoforms were analyzed using RT-PCR (Figure 3B). The mRNA level of NOX4 initially decreased and then gradually increased from day 12 of vascular differentiation, the point at which CD34+ cells began to appear in differentiated hEBs. The mRNA levels of NOX4 and p22phox were markedly higher in CD34+ cells than in CD34- cells (Figure 3C). Expression of the other NOX isoforms was not detected in CD34+ or CD34- cells. To investigate the involvement of NOX4 in ROS generation during vascular differentiation of hESCs, NOX4 expression was reduced in differentiated hEBs using siRNA (Figure 3D). In hEBs treated with NOX4-specific siRNA (Si-NOX4), few CD34+ cells displayed fluorescent H2DCFDA (Figure 3E). By contrast, in hEBs treated with control siRNA (Si-Cont), H2DCFDA fluorescence intensity was high in CD34+ cells. Quantitative analysis of


NADPH oxidase (NOX) 4 generates ROS in hESC-derived CD34+ cells.

10

H2DCFDA fluorescence revealed that ROS levels were significantly lower in Si-NOX4-treated hEBs than in Si-Cont-treated hEBs. Selenium reduces the levels of ROS in hESC-derived CD34+ cells by enhancing the antioxidant activities of TrxR and GPx. Next, we investigated whether the cell cycle status of hESC-derived CD34+ cells is altered when the level of intracellular ROS is reduced. Hydrogen peroxide, a major ROS produced by NOX4, is

intracellular ROS levels, NAC, which completely scavenges ROS similar to catalase, was added to the differentiation culture medium of cells. NAC treatment almost completely inhibited vascular differentiation of hESCs (Supplementary Figure III), suggesting that a certain level of ROS is required for vascular differentiation. To moderately reduce the level of ROS, the antioxidant activities of the

Accepted Manuscript

glutathione and thioredoxin systems were modulated using selenium. TrxR and GPx are key enzymes in these systems and both are selenoproteins that require selenium for their activities11. The mRNA levels of TrxR and GPx in hESC-derived CD34+ cells were examined (Figure 4A). All TrxR genes (TrxR1, TrxR2, and TrxR3) and four GPx genes (GPx1, GPx2, GPx3, and GPx4) were expressed in hESC-derived CD34+ cells. When hEBs were differentiated in the presence of non-cytotoxic concentrations of selenium (20 and 50 ng/mL), the antioxidant activities of TrxR and GPx were enhanced and the levels of ROS in hEBS was substantially reduced by day 15 of differentiation (Figure 4B-D). The fluorescence intensity of H2DCFDA was significantly lower in selenium-treated hESC-derived CD34+ cells than in untreated hESC-derived CD34+ cells (Figure 4E). However, the fluorescence intensity of H2DCFDA in selenium-treated hESC-derived CD34+ cells was not as low as that in hESC-derived CD34- cells. To confirm that changes in ROS levels in hESC-derived CD34+ cells in response to selenium treatment were due to the enhanced activities of TrxR and GPx, levels of ROS were measured after treatment with a TrxR inhibitor (ATG, AF, or PX12) or a GPx inhibitor (MSA). Treatment with each of these inhibitors substantially increased the fluorescence intensity of H2DCFDA in selenium-treated CD34+ cells, suggesting that selenium reduces ROS levels in hESCderived CD34+ cells by enhancing the antioxidant activities of TrxR and GPx (Figure 4F).


neutralized into water by catalase and the glutathione and thioredoxin redox systems10. To reduce

11 Selenium enhances the proliferation of hESC-derived CD34+ cells Next, we determined whether the decrease in ROS levels following selenium supplementation changes the cell cycle status of hESC-derived CD34+ cells. There were significantly more Ki-67+ cells in the selenium-treated hESC-derived CD34+ cell population than in the untreated hESC-derived CD34+ cell population (Figure 5A). Cell cycle analysis showed that the percentage of cells in G0 phase was lower in the selenium-treated hESC-derived CD34+ cell population than in the untreated

related genes (CDK1, CCNA2, CCND1, CCNB1, CDC20, and PCNA) were increased in seleniumtreated CD34+ cells, whereas expression of p21 was substantially reduced (Figure 5C). Flow cytometric analysis showed that selenium treatment increased the number of CD34+ cells during vascular differentiation of hESCs (Figure5D). This result was confirmed by immunocytochemical

Accepted Manuscript

data showing that the number of CD34+ cells was substantially increased in selenium-treated hEBs (Figure 5E). Selenium promotes vascular differentiation of hESCs Since CD34+ cells differentiate into ECs and SMCs during the differentiation of hESCs, we examined whether the selenium-induced increase in the number of CD34+ cells enhances vascular differentiation of hESCs. Real-time RT-PCR showed that mRNA levels of EC-specific genes (CD31, and VE-cadherin) and SMC-specific genes (Myocardin, and α-SMA) were higher in selenium-treated hEBs than in untreated hEBs (Figure 6A). Flow cytometric analysis showed that there were more CD31+ cells and smooth muscle22α (SM22α)+ cells in selenium-treated hEBs than in untreated hEBs (Figure 6B). Consistent with the RT-PCR results, immunocytochemical data revealed that protein levels of CD31, VE-cadherin, and α-SMA were substantially higher in selenium-treated hEBs than in untreated hEBs (Figure 6C). The selenium-induced increase in vascular differentiation of hESCs was abrogated when hEBs were co-treated with selenium and a TrxR inhibitor (ATG) or a GPx inhibitor (MSA) (Supplementary Figure IV). This suggests that selenium promotes vascular differentiation of hESCs by enhancing the activities of TrxR and GPx. By contrast, selenium supplementation did not affect the differentiation of hESCs into endoderm or ectoderm (Supplementary Figure V).


hESC-derived CD34+ cell population (Figure 5B). In RT-PCR analysis, the mRNA levels of mitosis-

12

Discussion In vascular biology, ROS are critical regulators of diverse physiological and pathophysiological processes. However, the specific roles of ROS during human embryonic vascular development have not been fully investigated. In this study, hESCs were used to investigate the generation, source, and role of ROS during vascular differentiation. NOX4 expression was substantially higher in CD34+ cells than in CD34- cells. NOX4 constitutively generates ROS (mainly hydrogen peroxide) without the

NOX4 expression level. Given this, it was hypothesized that the excess level of ROS in CD34+ cells is due to upregulation of NOX410. Indeed, knockdown of NOX4 using siRNA significantly reduced the level of ROS in CD34+ cells, indicating that NOX4 is the predominant source of ROS in hESCderived CD34+ cells. NOX4 expression is induced in response to a wide variety of agonists and

Accepted Manuscript

cellular stressors12. In particular, several NOX4 inducers, such as BMP4 and angiotensin II, as well as hypoxia, are crucial mediators of the vascular differentiation of hESCs. BMP4 promotes the development of vascular progenitors during the differentiation of hESCs13. Zambidis et al. showed that inhibitors of angiotensin II signaling significantly reduce the expansion of hESC-derived hemangioblasts that give rise to vascular progenitors and hematopoietic stem cells14. Hypoxia has recently been shown to accelerate the differentiation of murine ESCs into vascular lineage cells15, 16. Our previous study revealed that CD34+ cells appear mainly at the center of hEBs, which might be subjected to severe hypoxia relative to the periphery of hEBs 7. Given the importance and relevance of NOX4 inducers in the vascular differentiation of hESCs, BMP4, angiotensin II, and/or hypoxia may be involved in the upregulation of NOX4 expression in hESC-derived CD34+ cells. The present study implicated NOX4-induced ROS generation in the quiescence of hESC-derived CD34+ cells. Unlike CD34- cells, CD34+ cells exhibited a gene expression pattern that is characteristic of non-dividing cells. For example, genes involved in mitosis (CCNB1, CDC20, and CDK1) and DNA replication (PCNA) were significantly downregulated in CD34+ cells17. In particular, the number of phospho-p53+ cells in the hESC-derived CD34+ cell population is substantially greater than in the hESC-derived CD34- cell population. This suggests that p53 pathway may be highly activated in


need for a stimulus and the level of NOX4-induced ROS production is mostly determined by the

13 hESC-derived CD34+ cells. Given that p53 directly increases the transcription of p21, a significant increase in the mRNA level of p21 in hESC-derived CD34+ cells may be due to p53 activation. p21 is a molecular switch governing transition out of the cell cycle and maintenance in G0 phase. The expression of p21 is regulated in a redox-sensitive manner. In various cell types, an increase in endogenous ROS or treatment with sub-lethal doses of hydrogen peroxide induces cell cycle arrest that is associated with increased expression of p218, 18, 19. In the present study, the mRNA level of p21

reduced ROS levels by enhancing the antioxidant activities of TrxR and GPx. In addition, the expression levels of other genes involved in cell proliferation (CDK1, CCNA2, CCNB1, CCND1, CDC20, and PCNA) were increased in selenium-treated CD34+ cells, which led to a decrease in the percentage of cells in G0 phase and an enhanced proliferation rate. These data suggest that NOX4-

Accepted Manuscript

induced ROS generation may be closely associated with the quiescent state of CD34+ cells during vascular differentiation of hESCs. It was further supported by the siRNA-mediated knockdown results that a moderate knockdown of NOX4 or p22phox decreased ROS levels and subsequently enhanced the proliferation and vascular differentiation of hESC-derived CD34+ cells (Supplementary Figure VI and VII). Although a moderate reduction in the level of ROS using selenium caused quiescent CD34+ cells to enter the cell cycle and proliferate, complete scavenging of ROS using NAC inhibited vascular differentiation of hESCs. During angiogenesis, NOX-dependent ROS formation is required to enhance EC proliferation and migration20. In ischemic vascular disease models, administration of NAC and NOX inhibitors abrogates collateral vessel growth and neovascularization in injured tissues21. NOX-derived ROS inactivate protein tyrosine phosphatase and therefore enhance signaling cascades for angiogenic factors such as VEGF, bFGF, stromal derived factor-1, and erythropoietin22. Furthermore, NOX4-derived hydrogen peroxide increases the expression and phosphorylation of endothelial NOS23. A moderate level of NOX4-derived ROS may exert similar effects in the regulation of signaling pathways and gene expression required for vascular differentiation of hESCs.


in hESC-derived CD34+ cells was substantially reduced by selenium treatment that moderately

14

However, more research is needed to elucidate the molecular mechanisms through which ROS regulate the vascular differentiation of hESCs. Taken together, this study demonstrates that NOX4-derived ROS are produced in CD34+ cells during the vascular differentiation of hESCs and are implicated in the quiescent state of hESC-derived CD34+ cells. A moderate decrease in ROS levels using selenium significantly promotes the proliferation of CD34+ cells, thereby enhancing the vascular differentiation of hESCs. However, it

derived CD34+ cells. These findings may contribute to a better understanding of redox states in human embryo during vascular differentiation and help to optimize the conditions required for efficient vascular differentiation of hESCs.

Accepted Manuscript

Funding This work was supported by the Bio & Medical Technology Development Program [20100020275] [2012M3A9C6050368] through National Research Foundation grant funded by the Korean government.

Acknowledgements None.

Conflicts of Interest None declared.


cannot be ruled out that other mechanisms might be involved in the cell cycle control of hESC-

15

AAD

aminoactinomycin D

AF

auranofin

AFP

α-fetoprotein

ATG

aurothioglucose

bFGF

basic fibroblast growth factor

BMP

bone morphogenetic protein

DAPI

4,6-diamidino-2-phenylindole

DPI

diphenyleneiodonium

EB

embryoid body

ECs

endothelial cells

FACS

fluorescence-activated cell sorting

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

GPx

glutathione Peroxidase

H2DCFDA

5,6-chloromethyl-2ʹ,7ʹ-dichlorodihydrofluorescein diacetate

hEBs

human embryoid bodies

hESCs

human embryonic stem cells

L-NAME

Nw-nitro- L-arginine methyl ester

MSA

mercaptosuccinic acid

NAC

N-acetylcysteine

NOS

nitric oxide synthase

NOX

NADPH oxidase

PE

phycoerythrin

pH3

phospho Histone 3

redox

reduction and oxidation

ROS

Reactive oxygen species

RT-PCR

Reverse transcription-polymerase chain reaction


Accepted Manuscript

Abbreviations

Si-Cont

control siRNA

Si-NOX4

NOX4-specific siRNA

siRNA

small interfering RNA

SMA

smooth muscle actin

SMCs

smooth muscle cells

TRxR

thioredoxin reductase

VEGF

vascular endothelial growth factor


Accepted Manuscript

16

17

References 1.

Murry CE, Keller G. Differentiation of embryonic stem cells to clinically relevant

populations: lessons from embryonic development. Cell 2008;132:661-680. 2.

Ferreira LS, Gerecht S, Shieh HF, Watson N, Rupnick MA, Dallabrida SM, Vunjak-

Novakovic G, Langer R. Vascular progenitor cells isolated from human embryonic stem cells give rise to endothelial and smooth muscle like cells and form vascular networks in vivo. Circ Res

3.

Hill KL, Obrtlikova P, Alvarez DF, King JA, Keirstead SA, Allred JR, Kaufman DS. Human

embryonic stem cell-derived vascular progenitor cells capable of endothelial and smooth muscle cell function. Exp Hematol 2010;38:246-257 e241. 4.

Park SW, Jun Koh Y, Jeon J, Cho YH, Jang MJ, Kang Y, Kim MJ, Choi C, Sook Cho Y,

Accepted Manuscript

Chung HM, Young Koh G, Han YM. Efficient differentiation of human pluripotent stem cells into functional CD34+ progenitor cells by combined modulation of the MEK/ERK and BMP4 signaling pathways. Blood 2010;116:5762-5772. 5.

Song SH, Jung W, Kim KL, Hong W, Kim HO, Lee KA, Lee KY, Suh W. Distinct

transcriptional profiles of angioblasts derived from human embryonic stem cells. Exp Cell Res 2013. 6.

Boonstra J, Post JA. Molecular events associated with reactive oxygen species and cell cycle

progression in mammalian cells. Gene 2004;337:1-13. 7.

Kim KL, Song SH, Choi KS, Suh W. Cooperation of Endothelial and Smooth Muscle Cells

Derived from Human Induced Pluripotent Stem Cells Enhances Neovascularization in Dermal Wounds. Tissue Eng Part A 2013. 8.

Li JM, Fan LM, George VT, Brooks G. Nox2 regulates endothelial cell cycle arrest and

apoptosis via p21cip1 and p53. Free Radic Biol Med 2007;43:976-986. 9.

Kang DH, Kang SW. Targeting Cellular Antioxidant Enzymes for Treating Atherosclerotic

Vascular Disease. Biomol Ther (Seoul) 2013;21:89-96. 10.

Lassegue B, San Martin A, Griendling KK. Biochemistry, physiology, and pathophysiology

of NADPH oxidases in the cardiovascular system. Circ Res 2012;110:1364-1390.


2007;101:286-294.

18

11.

Steinbrenner H, Sies H. Protection against reactive oxygen species by selenoproteins.

Biochim Biophys Acta 2009;1790:1478-1485. 12.

Bedard K, Krause KH. The NOX family of ROS-generating NADPH oxidases: physiology

and pathophysiology. Physiological reviews 2007;87:245-313. 13.

Bai H, Gao Y, Arzigian M, Wojchowski DM, Wu WS, Wang ZZ. BMP4 regulates vascular

progenitor development in human embryonic stem cells through a Smad-dependent pathway. J Cell

14.

Zambidis ET, Park TS, Yu W, Tam A, Levine M, Yuan X, Pryzhkova M, Peault B.

Expression of angiotensin-converting enzyme (CD143) identifies and regulates primitive hemangioblasts derived from human pluripotent stem cells. Blood 2008;112:3601-3614. 15.

Han Y, Kuang SZ, Gomer A, Ramirez-Bergeron DL. Hypoxia influences the vascular

Accepted Manuscript

expansion and differentiation of embryonic stem cell cultures through the temporal expression of vascular endothelial growth factor receptors in an ARNT-dependent manner. Stem cells (Dayton, Ohio) 2010;28:799-809. 16.

Lee SW, Jeong HK, Lee JY, Yang J, Lee EJ, Kim SY, Youn SW, Lee J, Kim WJ, Kim KW,

Lim JM, Park JW, Park YB, Kim HS. Hypoxic priming of mESCs accelerates vascular-lineage differentiation through HIF1-mediated inverse regulation of Oct4 and VEGF. EMBO Mol Med 2012;4:924-938. 17.

Coller HA, Sang L, Roberts JM. A new description of cellular quiescence. PLoS Biol

2006;4:e83. 18.

Russo T, Zambrano N, Esposito F, Ammendola R, Cimino F, Fiscella M, Jackman J,

O'Connor PM, Anderson CW, Appella E. A p53-independent pathway for activation of WAF1/CIP1 expression following oxidative stress. J Biol Chem 1995;270:29386-29391. 19.

Barnouin K, Dubuisson ML, Child ES, Fernandez de Mattos S, Glassford J, Medema RH,

Mann DJ, Lam EW. H2O2 induces a transient multi-phase cell cycle arrest in mouse fibroblasts through modulating cyclin D and p21Cip1 expression. J Biol Chem 2002;277:13761-13770.


Biochem 2010;109:363-374.

19

20.

Ushio-Fukai M, Tang Y, Fukai T, Dikalov SI, Ma Y, Fujimoto M, Quinn MT, Pagano PJ,

Johnson C, Alexander RW. Novel role of gp91(phox)-containing NAD(P)H oxidase in vascular endothelial growth factor-induced signaling and angiogenesis. Circ Res 2002;91:1160-1167. 21.

Tojo T, Ushio-Fukai M, Yamaoka-Tojo M, Ikeda S, Patrushev N, Alexander RW. Role of

gp91phox (Nox2)-containing NAD(P)H oxidase in angiogenesis in response to hindlimb ischemia. Circulation 2005;111:2347-2355. Frey RS, Ushio-Fukai M, Malik AB. NADPH oxidase-dependent signaling in endothelial

cells: role in physiology and pathophysiology. Antioxid Redox Signal 2009;11:791-810. 23.

Craige SM, Chen K, Pei Y, Li C, Huang X, Chen C, Shibata R, Sato K, Walsh K, Keaney JF,

Jr. NADPH oxidase 4 promotes endothelial angiogenesis through endothelial nitric oxide synthase

Accepted Manuscript

activation. Circulation 2011;124:731-740.


22.

20

Figure legends Figure 1. hESC-derived CD34+ cells are quiescent (A) Flow cytometric analysis of CD34+ cells in differentiated hEBs. (B) Real-time RT-PCR analysis of cell cycle-related genes in hESC-derived CD34+ and CD34- cells. CD34+ and CD34- cells were sorted from hEBs on day 15 of vascular differentiation. mRNA levels in CD34+ cells are expressed relative to the levels in CD34- cells (set as 1) (mean ± SEM, *p < 0.05, n = 3). (C) Quantitative

vascular differentiation, sorted CD34+ and CD34- cells were plated at the same density and labeled with anti-phospho-p53 IgG. The mean number of phospho-p53+ cells per field is shown (mean ± SEM, *p < 0.05, n = 4). (D) Cell cycle analysis of hESC-derived CD34+ and CD34- cells. On day 15 of vascular differentiation, CD34+ and CD34- cells were sorted with the same gating strategy used in

Accepted Manuscript

Figure 1A. Sorted cells were then stained with pyronin Y and 7-AAD. Numbers indicate the percentage of cells in each cell cycle phase. Quiescent cells (G0 phase) were identified as the population with low 7-AAD and pyronin Y signals and quantified as the percentage of the total number of cells (mean ± SEM, *p < 0.05, n = 4). (E) Representative images of Ki-67-stained samples and the percentages of Ki-67 + cells in hESC-derived CD34+ and CD34- cell populations are shown (mean ± SEM, *p < 0.05, n = 4). On day 15 of vascular differentiation, sorted cells were labeled with anti-Ki-67 IgG (green) and DAPI (blue). The percentage of Ki-67+ cells was expressed as the mean number of Ki-67+ cells per total number of cells in a given image. Negative control (NC) samples were stained with isotype-matched control IgG. Scale bars indicate 50 µm. Figure 2. ROS generation in hESC-derived CD34+ cells (A) Representative images of H2DCFDA fluorescence in undifferentiated hESCs (D0) and hEBs cultured in vascular differentiation medium for 10 (D10) or 15 (D15) day, and quantitative evaluation of H2DCFDA fluorescence intensity (mean ± SEM, *p < 0.05 vs. D0, n = 4). Cellular ROS levels were measured using the redox-sensitive indicator H2DCFDA. Cells pretreated with 100 µM hydrogen peroxide were included as positive controls. As the NCs, cells were not incubated with H2DCFDA. (B) On day 15 of vascular differentiation, hEBs were stained with PE-conjugated anti-


analysis of phospho-p53+ cells in hESC-derived CD34+ and CD34- cell populations. On day 15 of

21 CD34 IgG (red) and H2DCFDA (green). (C) Cellular ROS levels in hESC-derived CD34+ and CD34cells. On day 15 of vascular differentiation, sorted cells were stained with H2DCFDA and the fluorescence intensities of 30-50 cells in each group were quantified (mean ± SEM, *p < 0.05, n = 3). All scale bars indicate 100 µm. Figure 3. NOX4 induces ROS generation in hESC-derived CD34+ cells (A) Representative images of H2DCFDA fluorescence after pretreatment of hESC-derived CD34+ and

On day 15 of vascular differentiation, CD34+ and CD34- cells were incubated with various ROS inhibitors before the addition of H2DCFDA. H2DCFDA fluorescence intensities of 30-50 cells in each group were quantified (mean ± SEM, *p < 0.05 vs. PBS-treated CD34+ cells, n = 3). (B) RTPCR analysis of the expression profiles of NOX genes and p22phox during vascular differentiation of

Accepted Manuscript

hESCs (D0, undifferentiated hESCs; D8, D12, and D15 hEBs differentiated for 8, 12, and 15 days, respectively). (C) mRNA levels of NOX genes and p22phox in hESC-derived CD34+ and CD34- cells on day 15 of vascular differentiation. (D) siRNA-mediated knockdown of NOX4 expression in hEBs. On day 15 of vascular differentiation, hEBs were transfected with Si-NOX4 or Si-Cont. mRNA levels were analyzed using RT-PCR. GAPDH was used as a loading control in (B-D). (E) Effect of NOX4 knockdown on ROS generation in hESC-derived CD34+ cells. HEBs were transfected with Si-NOX4 or Si-Cont and stained with anti-CD34 IgG (red) and H2DCFDA (green) the following day. H2DCFDA fluorescence intensities in each group were quantified (mean ± SEM, *p < 0.05, n = 3). All scale bars indicate 50 µm. Figure 4. Selenium moderately reduces the ROS level in hESC-derived CD34+ cells by enhancing the antioxidant activities of TrxR and GPx (A) RT-PCR analysis of mRNA levels of various TrxR and GPx genes in hESC-derived CD34+ cells on day 15 of differentiation. (B)(C) Effect of selenium (Se) on the antioxidant activities of TrxR and GPx in hEBs (mean ± SEM, *p < 0.05 vs. cells not treated with selenium (Se 0), n =3). Antioxidant activities of TrxR and GPx were examined in hEBs differentiated with or without selenium supplementation (20 or 50 ng/mL). (D) Effect of selenium on ROS generation during vascular


CD34- cells with ROS inhibitors, and quantification of relative H2DCFDA fluorescence intensities.

22

differentiation of hESCs. H2DCFDA fluorescence was imaged and quantified in hEBs differentiated with or without selenium supplementation for 10 (D10) or 15 (D15) days (mean ± SEM, *p < 0.05, n = 3). Scale bars indicate 100 µm. (E) Effect of selenium on ROS generation in hESC-derived CD34+ and CD34- cells. After hEBs were differentiated with or without selenium supplementation (20 or 50 ng/mL) for 15 days, CD34+ and CD34- cells were sorted and stained with H2DCFDA (mean ± SEM, *p < 0.05, n = 3). Scale bar indicates 50 µm. (F) Effect of TrxR inhibitors (ATG, AF, and PX12) and

cells. After hEBs were differentiated with or without selenium supplementation (50 ng/mL) for 15 days, CD34+ cells were treated with various ROS inhibitors before the addition of H2DCFDA. H2DCFDA fluorescence was imaged and quantified as previously described (mean ± SEM, *p < 0.05, #

p > 0.05 vs. cell treated with PBS without selenium supplementation (Se 0), n = 3). Scale bar

Accepted Manuscript

indicates 50 µm. Figure 5. Selenium promotes the proliferation of hESC-derived angioblasts HEBs were differentiated with or without selenium supplementation (0, 50 ng/mL) for 15 days. (A) Representative images of Ki-67-stained CD34+ cells and quantitative analysis of the percentage of Ki67+ cells (mean ± SEM, *p < 0.05, n = 3). CD34+ cells were sorted from differentiated hEBs and labeled with anti- Ki-67 IgG (green) and DAPI (blue). (B) Cell cycle analysis of hESC-derived CD34+ cells. CD34+ cells were sorted and stained with pyronin Y and 7-AAD. Numbers indicate the percentage of cells in each cell cycle phase. (C) RT-PCR analysis of the mRNA levels of cell cyclerelated genes in hESC-derived CD34+ cells. Intensities of PCR bands were densitometrically measured and divided by the corresponding value of the loading control (GAPDH). Relative mRNA levels were calculated by setting the mRNA level of CDK1 in untreated CD34+ cells to 1.0 (mean ± SEM, *p < 0.05 vs. Se 0 of each gene, n = 3). (D-E) Flow cytometric analysis and immunofluorescence assay of CD34+ cells in differentiated hEBs. In (D), numbers indicate the percentage of CD34+ cells. In (E), red fluorescence indicates anti-CD34 IgG staining. All scale bars indicate 50 µm.


a GPX inhibitor (MSA) on the selenium-mediated decrease in ROS levels in hESC-derived CD34+

23

Figure 6. Selenium enhances the vascular differentiation of hESCs HEBs were differentiated with or without selenium supplementation (20 or 50 ng/mL) for 15 days. (A) Real-time RT-PCR analysis of EC (CD31 and VE-cadherin) and SMC (α-SMA and Myocardin) marker genes in hEBs. The mRNA level of each gene is expressed relative to the level in cells differentiated without selenium supplementation (Se 0) (mean ± SEM, *p < 0.05 vs. Se 0, n =3). (B) Flow cytometric analysis of CD31+ and SM22α+ cells in hEBs. Numbers indicate the percentage of

Accepted Manuscript

or anti-α-SMA IgGs. Scale bars indicate 50 µm.


positively stained cells. (C) Representative images of hEBs stained with anti-CD31, anti-VE-cadherin,

24


25


26


27


28


29


PGC-1α and reactive oxygen species regulate human embryonic stem cell-derived cardiomyocyte function.

Reactive Oxygen Species in Stem Cells.

Reactive Oxygen Species Are Required for Human Mesenchymal Stem Cells to Initiate Proliferation after the Quiescence Exit.

CDK6 levels regulate quiescence exit in human hematopoietic stem cells.

CD38 Is Required for Neural Differentiation of Mouse Embryonic Stem Cells by Modulating Reactive Oxygen Species.

Electrical maturation of neurons derived from human embryonic stem cells.

Germ cells from human embryonic stem cells?

Reactive oxygen species in eradicating acute myeloid leukemic stem cells.

Reactive oxygen species in normal and tumor stem cells.

bFGF promotes adipocyte differentiation in human mesenchymal stem cells derived from embryonic stem cells.

Prostate tumor-induced angiogenesis is blocked by exosomes derived from menstrual stem cells through the inhibition of reactive oxygen species.

Islet Endothelial Cells Derived From Mouse Embryonic Stem Cells.

Molecular characterisation of stromal populations derived from human embryonic stem cells: Similarities to immortalised bone marrow derived stromal stem cells.

Neural Progenitor Cells Derived from Human Embryonic Stem Cells as an Origin of Dopaminergic Neurons.

Characterization of polyhormonal insulin-producing cells derived in vitro from human embryonic stem cells.

The miR-29b-Sirt1 axis regulates self-renewal of mouse embryonic stem cells in response to reactive oxygen species.

Reactive oxygen species regulate the differentiation of acute promyelocytic leukemia cells through HMGB1-mediated autophagy.

Excess reactive oxygen species production mediates monoclonal antibody-induced human embryonic stem cell death via oncosis.

Lysophosphatidic acid increases the proliferation and migration of adipose‑derived stem cells via the generation of reactive oxygen species.

Intermittent high oxygen influences the formation of neural retinal tissue from human embryonic stem cells.

Embryonic stem cell-derived haematopoietic progenitor cells down-regulate the CD3 ξ chain on T cells, abrogating alloreactive T cells.

Adipocytes derived from PA6 cells reliably promote the differentiation of dopaminergic neurons from human embryonic stem cells.

Comparison of immunological characteristics of mesenchymal stem cells derived from human embryonic stem cells and bone marrow.

Triploid human embryonic stem cells derived from tripronuclear zygotes displayed pluripotency and trophoblast differentiation ability similar to the diploid human embryonic stem cells.