|Home | About | Journals | Submit | Contact Us | Français|
Human pluripotent stem cell (hPSC)-derived endothelial cells and their progenitors are important for vascular research and therapeutic revascularization. Here, we report a completely defined endothelial progenitor differentiation platform that uses a minimalistic medium consisting of Dulbecco's Modified Eagle Medium and ascorbic acid, lacking of albumin and growth factors. Following hPSC treatment with a GSK-3β inhibitor and culture in this medium, this protocol generates more than 30% multipotent CD34+CD31+ endothelial progenitors that can be purified to>95% CD34+ cells via magnetic activated cell sorting (MACS). These CD34+ progenitors are capable of differentiating into endothelial cells in serum-free inductive media. These hPSC-derived endothelial cells express key endothelial markers including CD31, VE-cadherin, and von Willebrand factor (vWF), exhibit endothelial-specific phenotypes and functions including tube formation and acetylated low-density lipoprotein (Ac-LDL) uptake. This fully defined platform should facilitate production of proliferative, xeno-free endothelial progenitor cells for both research and clinical applications.
Human pluripotent stem cells (hPSCs) are increasingly used in vascular research, including disease modeling, drug screening, and development of regenerative therapies (Ashton et al., 2011; Bautch, 2011; Kinney et al., 2014; Kusuma et al., 2014; Murry and Keller, 2008; Segers and Lee, 2008). Recently, dramatic improvements in the efficiency of directed differentiation protocols to produce endothelial cells have been reported by stage-specific modulation of pathways including TGFβ superfamily (James et al., 2010; Rufaihah et al., 2011; Wang et al., 2007), VEGF (vascular endothelial growth factor) (Goldman et al., 2009; James et al., 2010; Rufaihah et al., 2011; Wang et al., 2007), and Notch signaling (Marcelo et al., 2013; Sahara et al., 2014). However, most of these approaches require animal cells, fetal bovine serum, or cytokines and growth factors, limiting their applications for large-scale endothelial cell production for research or therapeutic applications.
Recently, we reported a rapid and robust endothelial progenitor differentiation protocol under serum-free conditions, which only employs a Gsk-3β inhibitor in LaSR basal medium (Advanced DMEM/F12, 2.5 mM GlutaMAX and 60 μg/mL ascorbic acid) (Lian et al., 2014). The presence of bovine serum albumin (BSA) in this medium increases the cost, adds xenogenic components, and heightens lot-to-lot variability.
Toward developing a defined, xeno-free endothelial progenitor differentiation platform, we screened several commercially available basal media, supplemented with insulin and ascorbic acid, for the ability to generate CD34+CD31+ cells after treatment with 6 μM CHIR99021. We found that DMEM supplemented with 100 μg/mL ascorbic acid generated 20-30% CD34+CD31+ endothelial progenitors that were enriched to >95% CD34+ progenitors via MACS. This minimal, defined differentiation platform should facilitate generation of proliferative endothelial progenitor cells from hPSCs for both research and clinical applications.
hPSCs maintained on a Synthemax-coated surface in E8 were dissociated into single cells with Accutase (Life Technologies) at 37°C for 5 min and then seeded onto a Synthemax-coated cell culture dish at 50,000 cell/cm2 in E8 supplemented with 5 μM ROCK inhibitor Y-27632 (Selleckchem) (day -3) for 24 hr. Cells were then cultured in E8, changed daily. At day 0, cells were treated with 3-9 μM CHIR99021 (Selleckchem) for 2 days in DMEM (Life Technologies, 11965) supplemented with 100 μg/ml ascorbic acid (Sigma, A8960) (DMEM/Vc). After 2 days, CHIR99021-containing medium was aspirated and cells were maintained in DMEM/Vc without CHIR99021 for 3 to 4 additional days.
Day 5 differentiated populations were dissociated with Accutase for 10 min and purified with an EasySep Magnet kit (STEMCELL Technologies) using a CD34 antibody (Miltenyi Biotec) according to the manufacturer's instructions. After purification, the total number of enriched CD34+ cells were counted and the yields were calculated as number of CD34+ endothelial progenitors generated per hPSC seeded at day −3. The endothelial progenitor cells were resuspended at a density of 2×106 cells per mL of endothelial freezing medium, which consists of 30% FBS (Life Technologies), 10% DMSO (Sigma), 60% EGM-2 (Lonza) and 5 μM Y-27632. 1 mL of the cell suspension was aliquoted into each cryovial and frozen in a Mr. Frosty™ freezing container at −80°C. 24 hr later, the cryovials were transferred to liquid nitrogen for long-term storage. For recovery, frozen cells were partially thawed in a 37°C water bath, and were then transferred into a 15-mL conical tube containing 5 mL 10% FBS DMEM medium (Life Technologies). After centrifuging, cells were resuspended in EGM-2 medium (Lonza) or human endothelial-SFM supplemented with 20 ng/ml bFGF and 10 ng/ml EGF (Life Technologies) containing 5 μM Y27632 and plated into collagen IV-coated dishes (BD BioCoat) at a density of 0.05 million cells per cm2. The next day, medium was replaced with fresh room temperature EGM-2 medium (Lonza) or human endothelial-SFM supplemented with 20 ng/ml bFGF and 10 ng/ml EGF (Life Technologies).
Day 5 differentiated populations were dissociated with Accutase for 10 min and purified with an EasySep Magnet kit (STEMCELL Technologies) using an anti-CD34 antibody according to the manufacturer's instructions. The purified CD34+ cells were plated on collagen IV-coated dishes (BD BioCoat) in EGM-2 medium (Lonza) or human endothelial-SFM supplemented with 20 ng/ml bFGF and 10 ng/ml EGF (Life Technologies) and split every 3-4 days with Accutase.
Day 5 differentiated populations were dissociated with Accutase for 10 min and purified with an EasySep Magnet kit (STEMCELL Technologies) using an anti-CD34 antibody according to the manufacturer's instructions. The purified CD34+ cells were plated on collagen IV-coated dishes (BD BioCoat) in SmGM-2 medium (Lonza) and split every 3-4 days with Accutase.
To assess the formation of capillary structures, 1 ×105 day 15 endothelial cells in 0.4 ml EGM-2 medium (Lonza) supplemented with 50 ng/ml VEGF (R&D Systems) were plated into one well of 24-well tissue culture plate pre-coated with 250 μl Matrigel (BD Biosciences). Tube formation was observed by light microscopy after 24 hr of incubation.
Total RNA was prepared with the RNeasy mini kit (QIAGEN) and treated with DNase (QIAGEN). 1 μg RNA was reverse transcribed into cDNA via Oligo (dT) with Superscript III Reverse Transcriptase (Invitrogen). Real-time quantitative PCR was done in triplicate with iQSYBR Green SuperMix (Bio-Rad). RT-PCR was performed with Gotaq Master Mix (Promega) and then subjected to 2% agarose gel electrophoresis. ACTB was used as an endogenous housekeeping control. PCR primer sequences are provided in Supplementary Table 4.
Cells were singularized with Accutase for 10 min and then fixed with 1% paraformaldehyde for 20 min at room temperature and stained with primary and secondary antibodies (Supplemental Table 3) in PBS plus 0.1% Triton X-100 and 0.5% BSA. Data were collected on a FACSCaliber flow cytometer (Beckton Dickinson) and analyzed using FlowJo. For ICAM-1 expression, day 15 post-purified endothelial cells were treated with or without 10 ng/ml TNFα for 16 hr prior to flow cytometry analysis.
Cells were fixed with 4% paraformaldehyde for 15 min at room temperature and then stained with primary and secondary antibodies (Supplemental Table 3) in PBS plus 0.4% Triton X-100 and 5% non-fat dry milk (Bio-Rad). Nuclei were stained with Gold Anti-fade Reagent with DAPI (Invitrogen). An epifluorescence microscope (Leica DM IRB) with a QImaging® Retiga 4000R camera was used for imaging analysis.
We previously demonstrated that activation of canonical Wnt signaling in hPSCs in LaSR basal medium generates functional CD34+/CD31+ endothelial progenitors in numerous hPSC lines (Lian et al., 2014). Figures 1A and S1 show schematics of the endothelial differentiation and purification protocols. LaSR basal medium consists of advanced DMEM/F12 medium, which contains proteins including transferrin and BSA (AlbuMAX II) (Supplementary Table 1). To develop a defined, xeno-free medium for endothelial progenitor differentiation, we assessed the efficiency of endothelial progenitor differentiation induced in H13 human embryonic stem cells (hESCs) by 6 μM CHIR99021 treatment in 4 commercially available basal media supplemented with 10 μg/mL insulin and 60 μg/mL ascorbic acid, as these two factors were shown to enhance endothelial cell proliferation and differentiation (May and Harrison, 2013; Montecinos et al., 2007; Piecewicz et al., 2012; Zhao et al., 2011). Only DMEM generated more than 10% CD34+CD31+ endothelial progenitors. Supplementing DMEM with ascorbic acid significantly increased the percentage of endothelial progenitors at day 5, while insulin diminished endothelial progenitor purity. Other basal media yielded few, if any, CD34+CD31+ cells (Fig. 1B).
We optimized the concentrations of CHIR99021 (CH) and ascorbic acid in DMEM and found that 5 μM CH and 100 μg/mL ascorbic acid provided the greatest purity of endothelial progenitors (Fig. 1C, D). Next, we tested DMEM supplemented with ascorbic acid as an endothelial progenitor differentiation medium in multiple additional hESC (H1, H14) and iPSC (19-9-11, 6-9-9, 19-9-7) lines at passages between 20 and 100, and they all generated 20-30% CD34+CD31+ cells (Fig. S2, Supplementary Table 2), comparable to the differentiation efficiencies reported in LaSR basal medium (Lian et al., 2014).
Molecular analysis during endothelial progenitor differentiation showed dynamic changes in gene expression, with downregulation of the pluripotency markers NANOG, SOX2, and OCT4, and induction of mesoderm genes T, MIXL1 and EOMES in the first 24 hours after CHIR99021 addition (Fig. 2A). Expression of the endothelial progenitor markers KDR, CD34, CDH5 and CD31 was detected at day 4 and increased at day 5 (Fig. 2A). Immunofluorescent analysis revealed robust surface expression of both CD34 and CD31 on day 5 (Fig. 2B). In addition, flow cytometry profiling during endothelial progenitor differentiation showed a population of cells expressing CD144, but not ICAM-1, appeared at day 5 (Fig. S3A), consistent with our previous report of hPSC differentiation to endothelial progenitors in albumin-containing medium (Lian et al., 2014). To further investigate the multipotent nature of these CD34+/CD31+ cells, single step MACS using an anti-CD34 antibody was performed on day 5 of differentiation, yielding 99% pure CD34+ cells (Fig. 2C). Additional cell lines were also enriched to >95% CD34+ populations with a yield of 4-5 CD34+ endothelial progenitors for every input hPSC (Fig. S2, Supplementary Table 2). The purified CD34+ cells were plated on Collagen IV-coated 96-well plates at a density of one cell per well in either endothelial or smooth muscle medium. After 10 days of culture, they generated relatively pure populations of cells expressing smooth muscle myosin heavy chain (SMMHC), smooth muscle actin (SMA) and calponin, or VE-cadherin, vWF and CD31, respectively (Fig. 2D and E), indicating their multipotency. In addition, we tested whether day 5 CD34+ cells exhibit hematopoietic potential in IMDM medium supplemented with growth factor cocktails (300 ng/ml stem cell factor ((SCF), 300 ng/ml Flt-3, 50 ng/ml colony-stimulating factor 3 (CSF3), 10 ng/ml IL-3, and 10 ng/ml IL-6) shown to sustain human hematopoietic stem cells (Wang et al., 2004), but did not detect CD45+ cells after 7 days (Fig S3B).
To further assess the intrinsic properties of endothelial cells differentiated from CD34+ cells generated in this defined platform, MACS-sorted CD34+ cells were cultured in commercial endothelial media (EGM2 and human endothelial SFM) on collagen IV-coated plates. The resulting cells exhibited morphological characteristics typical of primary endothelial cells (Fig. S1). These hPSC-derived endothelial cells proliferated actively and were capable of 20 population doublings over 2 months in serum-containing EGM2 (Fig. 3A). Flow cytometry and immunostaining analysis of cells differentiated in serum-free human endothelial SFM revealed robust expression of CD31, VE-cadherin and vWF, comparable to primary human umbilical vein endothelial cells (HUVECs) (Fig. 3B, C).
Next, we assessed the endothelial nature of these hPSC-derived CD31+ cells differentiated in serum-free media by testing for tube formation and acetylated low-density lipoprotein (Ac-LDL) uptake. Upon treatment with VEGF, the cells organized into tube-like structures in Matrigel (Fig. 3D), and were able to take up Ac-LDL (Fig. 3E), demonstrating their endothelial function. In addition, these hPSC-derived endothelial cells upregulated expression of the adhesion molecule ICAM-1 upon TNF-α treatment (Fig. S3C), indicating their ability to respond to inflammatory mediators. Furthermore, they also maintained viability (Fig. S4A) and endothelial marker expression after storage in liquid nitrogen for a month (Fig. S4B, C), indicative of cryopreservation ability.
Existing methods for hPSC differentiation to endothelial progenitors require the addition of growth factors and/or xenogenic components, limiting their application for large-scale production and therapeutic applications (Bautch, 2011; Wilson et al., 2014). Here, we report a defined, albumin-free, non-xenogenic differentiation system for directing hPSCs to endothelial progenitors. We showed that a completely defined medium, DMEM supplemented with 100 μg/mL ascorbic acid, is sufficient to efficiently generate CD34+CD31+ endothelial progenitors from hPSCs following Gsk-3β inhibition. These hPSC-derived endothelial progenitors are multipotent and can be further directed into smooth muscle cells or endothelial cells upon subsequent culture in appropriate inductive media. CD31+/VE-cadherin+ endothelial cells differentiated under serum-free conditions exhibited uptake of acetylated low-density lipoprotein (Ac-LDL) and formed tube-like structures when cultured on Matrigel in the presence of VEGF. However, long-term expansion of these cells required serum-containing medium.
Albumin has been reported to increase growth rate and overall cell health (Ashman et al., 2005; Zoellner et al., 1996). Here, however, we demonstrate that albumin is dispensable in endothelial progenitor differentiation. In spite of the greater simplicity of this new albumin free-medium, it supported endothelial progenitor induction of hPSCs comparably to LaSR basal medium. This simplified medium offers several advantages in both research and clinical applications of hPSC-derived endothelial progenitors. First, it eliminates batch-to-batch variability of albumin, likely increasing reproducibility of differentiation processes. Second, it provides a simpler chemical background for examining and screening factors regulating gene expression, differentiation, and proliferation. For example, albumin can bind and sequester lipids, proteins and small molecules (Garcia-Gonzalo and Izpisúa Belmonte, 2008). Third, it can reduce the risk of potential pathogen contamination and cell immunogenicity, facilitating therapeutic applications of hPSC-derived endothelial progenitor cells. Finally, this new system can significantly reduce reagent cost and simplify quality control for endothelial progenitor cell differentiation.
This study demonstrates that a completely defined, xeno-free medium can be used to efficiently derive functional endothelial progenitors from hPSCs in the absence of exogenous proteins. This is an important step toward the ultimate clinical application of hPSC-derived endothelial progenitors.
This work was supported by NIH grant R01 EB007534.
Funding: This work was supported by NIH grant R01 EB007534.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Xiaoping Bao: Conception and design, Collection and/or assembly of data, Data analysis and interpretation, Manuscript writing, Final approval of manuscript
Xiaojun Lian: Conception and design, Collection and/or assembly of data, Data analysis and interpretation, Manuscript writing
Kaitlin K. Dunn: Collection and/or assembly of data, Data analysis and interpretation
Mengxuan Shi: Collection and/or assembly of data, Data analysis and interpretation
Tianxiao Han: Collection and/or assembly of data, Data analysis and interpretation
Tongcheng Qian: Collection and/or assembly of data, Data analysis and interpretation
Vijesh J. Bhute: Collection and/or assembly of data, Data analysis and interpretation
Scott G. Canfield: Collection and/or assembly of data, Data analysis and interpretation
Sean P. Palecek: Conception and design, Financial support, Data analysis and interpretation, Manuscript writing, Final approval of manuscript
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
The authors disclose no conflicts of interest.