Derivation of Endothelial Cells From Human Embryonic Stem Cells by Directed Differentiation
Analysis of MicroRNA and Angiogenesis In Vitro and In Vivo
Objective— To develop an embryoid body-free directed differentiation protocol for the rapid generation of functional vascular endothelial cells derived from human embryonic stem cells (hESCs) and to assess the system for microRNA regulation and angiogenesis.
Methods and Results— The production of defined cell lineages from hESCs is a critical requirement for evaluating their potential in regenerative medicine. We developed a feeder- and serum-free protocol. Directed endothelial differentiation of hESCs revealed rapid loss of pluripotency markers and progressive induction of mRNA and protein expression of vascular markers (including CD31 and vascular endothelial [VE]-cadherin) and angiogenic growth factors (including vascular endothelial growth factor), increased expression of angiogenesis-associated microRNAs (including miR-126 and miR-210), and induction of endothelial cell morphological features. In vitro, differentiated cells produced nitric oxide, migrated across a wound, and formed tubular structures in both the absence and the presence of 3D matrices (Matrigel). In vivo, we showed that cells that differentiated for 10 days before implantation were efficient at the induction of therapeutic neovascularization and that hESC-derived cells were incorporated into the blood-perfused vasculature of recipient mice.
Conclusion— The directed differentiation of hESCs is efficient and effective for the differentiation of functional endothelial cells from hESCs.
Tissue regeneration and stimulation of angiogenesis are important therapeutic considerations in the treatment of myocardial infarction, peripheral vascular disease, and stroke. Human embryonic stem cells (hESCs) can be maintained in a pluripotent state indefinitely, with the capacity for exponential scale up and differentiation into any cell type of the adult body,1 including endothelial cells (ECs) expressing functional EC-specific characteristics, genes, and proteins.2–6 hESC-ECs may be used to generate a large supply of transplantable, healthy, functional cells for the repair of ischemic tissues.
Presently, 2 approaches exist for differentiating hESCs to ECs: either 3D embryoid body (EB) differentiation or 2D monolayer-directed differentiation, with serum- and growth factor–supplemented differentiation systems. EB differentiation, in which cells are permitted to spontaneously differentiate into ECs7–12 concurrently with a multitude of other cell types from all 3 germ layers, yields low EC differentiation efficiency (range, 1% to 3%) and usually requires an entire colony of hESCs.7–9 EBs are notoriously difficult to dissociate and subsequently challenging to isolate and culture the preferred cell type to homogeneity, even by fluorescence-activated cell sorting (FACS) or magnetic-activated cell sorting.7,9 To circumvent this, modified approaches to increase endothelial differentiation efficiency via serial differentiation have been developed.4,10 These modifications have included coculture with mouse fibroblast feeder layers (mouse embryo fibroblasts,13 OP9,5,14 S17,14,15 MS-5,14 and mouse ECs15) before isolation of subculture of progenitor or mature ECs. Murine models of hind limb ischemia have been used in such systems to demonstrate that implantation of hESC-ECs can improve blood perfusion and limb salvage.4–6,8,10 Thus, hESC-ECs can potentially be used as a source of ECs for the treatment of ischemia. However, there are significant scientific obstacles that must be resolved with the current differentiation protocols before human clinical trials (eg, animal product contamination, such as animal-derived feeder cells and serum inclusion in differentiation media, and efficient strategies developed to enable robust production, isolation, and scalability of desired ECs to ensure sufficient cell numbers for transplantation). Herein, to the best of our knowledge, we describe the first feeder- and serum-free monolayer hESC-EC–directed differentiation protocol, with detailed analysis of microRNA (miRNA) and mRNA/protein markers and assessment of the hESC-EC therapeutic potential in a mouse model of limb ischemia.
Detailed methods can be found in the supplemental material (available online at http://atvb.ahajournals.org).
Cell Lines and Culture Conditions
The hESC lines SA461 and SA121 (Cellartis, Dundee, Scotland) were cultured in a feeder-free system. To maintain hESCs in prolonged pluripotent status, hESCs were cultured at 2×105 cells/cm2 on fibronectin (Merck Chemicals Ltd, Nottingham, England). Cells were cultured in chemically defined media16 (denoted “pluripotent maintenance media”), supplemented with 50% human umbilical Wharton inner layer (Cellartis) fibroblast-conditioned media (VitrohES; VitroLife, Gothenburg, Sweden) and 10 ng/mL basic fibroblast growth factor (Invitrogen, Paisley, Scotland). Endothelial differentiation was induced by incubation in an “endothelial differentiation media,” consisting of large-vessel endothelial growth media, 500 mL (TCS CellWorks, Buckingham, England), supplemented with hydrocortisone, 1 μg/mL; human epidermal growth factor, 10 ng/mL; basic fibroblast growth factor, 3 ng/mL; and heparin, 10 μg/mL (TCS CellWorks).
Taqman Low-Density Array mRNA Analysis
Total cellular RNA was isolated from hESC-ECs or time-matched pluripotent hESCs at each differentiation point with a kit (miRNAeasy Mini Kit; Qiagen, Crawley, England). mRNA expression was analyzed using the human stem cell pluripotency Taqman low-density array fluidic card. Mature miRNA expression was analyzed using the human microRNA array v1.0 Taqman low-density array card.
FACS, Immunocytofluorescence and Angiogensis Array
Cells were harvested by enzymatic dissociation using Tryple Select (Invitrogen) and washed in PBS containing 2% FCS (FACS buffer). Subsequently, cells were incubated for 1 hour at 4°C with mouse IgG1 monoclonal anti–human CD31–fluorescein isothiocyanate (BD Pharmingen, Oxford, England) and mouse IgG2b monoclonal antihuman VE-cadherin–PE (R&D Systems, Abingdon, England).
Cells were incubated at 4°C overnight with the following antibodies: mouse IgG monoclonal anti–human OCT4 (1:200; Santa Cruz Biotechnology, Inc, Heidelberg, Germany); mouse IgG monoclonal anti–human Nanog (1:200; Abcam plc, Cambridge, England); mouse IgG monoclonal anti–human CD31 (1:100; Dako UK Ltd, Ely, England); and goat IgG polyclonal VE-cadherin (1:100; R&D Systems). Detection was by a 1:500 dilution of fluorescein isothiocyanate (488) or tetramethylrhodamine B isothiocyanate (543)–conjugated goat antimouse/goat antirabbit/donkey antigoat antibody (Molecular Probes and Invitrogen).
Human angiogenesis proteome antibody arrays (Profiler; R&D Systems) were conducted according to the manufacturer’s instructions.
Assessment of Nitric Oxide Production
hESC-ECs at varying stages of differentiation were incubated with 5-μmol/L 4-amino-5-methylamino-2′, 7′-difluoro fluorescein (DAF-FM) diacetate (Invitrogen) for 30 minutes at 37°C. Thereafter, cells were washed with PBS and fluorescence visualized using a 200-mol/L microscope (Zeiss Axiovert 2OOM; Carl Zeiss Ltd, Welwyn Garden City, England). Endogenous nitric oxide (NO) production in filtered cell supernatant, harvested over a 72-hour collection period, was quantified using a kit (BIOMOL Nitric Oxide Colorimetric [Greiss reaction] Assay Kit; Enzo Life Sciences [UK] Ltd, Exeter, England) and conducted according to the manufacturer’s instructions. Stimulation of NO production was induced by the addition of 10-μmol/L carbachol (Sigma-Aldrich Company Ltd, Dorset, England) 72 hours before collection of supernatant.
Wound Closing Assay
A wound-closing assay was performed and quantified as previously described.17
Cell Transplantation Study
The experiments involving mice were performed in accordance with the Guide for the Care and Use of Laboratory Animals, prepared by the Institute of Laboratory Animal Resources and with prior approval of the UK Home Office and the University of Bristol. Unilateral hind limb ischemia was induced in immunocompromised NU/TO/LAC/BK mice (B&K, Hull, England), as previously described.18 hESC-ECs, 106 cells in 15 μL of culture medium (n=17 mice), were injected in 3 equidistant sites of the ischemic adductor muscle along the projection of the femoral artery. Control mice were injected with fresh endothelial differentiation medium (n=18 mice) or a cellular control, consisting of human microvascular ECs (hMVECs) (Cambrex), 106 cells (n=12 mice). Foot blood flow was measured at 7 and 14 days after ischemia by using a perfusion imaging system (Lisca color laser Doppler; Perimed, Sweden). At 14 days after surgery, the limbs of terminally anesthetized mice were perfusion/fixed and the adductor muscles were harvested and paraffin embedded to perform histological analyses of capillary densities.19,20 To determine the presence of transplanted cells, additional sections were stained for the antihuman nucleus antigen (hNAg) and the endothelial marker CD31. To determine the incorporation of transplanted cells into blood vessels of ischemic adductor muscles, in a separate experiment, mice were injected with hESC-ECs or cell medium in their ischemic adductor; after 14 days, they were IV injected with both a human-specific fluorescein isothiocyanate–conjugated lectin and a mouse-specific biotinylated lectin (each lectin at 100 μg/100 μL) (n=5 mice per group). After 10 minutes, mice were perfusion/fixed and adductor muscles were harvested and frozen in cryostat optimum cutting temperature (Cryo-OCT) compound (Fisher Scientific UK Ltd, Loughborough, England). Cryosections were then stained for the anti-hNAg in combination with human-specific lectin or with mouse-specific and human-specific lectins together.
Before any statistical analysis, data were tested for and shown to exhibit gaussian distribution. Gaussian distribution was determined by applying the Shapiro-Wilk normality test to the data. Where appropriate, values were presented as mean±SEM. Comparison of the different parameters for the various genes by Taqman low-density array analysis, NO production, and cell migration was determined by repeated-measures ANOVA. Significant differences were assigned using the Dunnett post hoc test. Comparison of capillary density was determined with an unpaired t test. Comparison of blood flow by Doppler flowmetry was determined by the Holm-Sidak t test for multiple comparisons and the 1-way ANOVA on rank test.
Assessment of a Novel Directed Differentiation Protocol
Morphologically, hESCs differentiated toward ECs assumed the characteristic cuboidal “cobblestone” morphological features (7 to 14 days) (supplemental Figure I). By 14 to 21 days of differentiation, the appearance of some “sprouting” cells, forming tubelike structures (in the absence of 3D matrices), was apparent.
Rapid Loss of Pluripotency and Induction of Vascular Endothelial and Angiogenesis-Associated Genes
We assessed both loss of pluripotency and induction of mesoderm/endothelial markers at the mRNA and protein level. A significant reduction in mRNA expression of the pluripotency markers, Nanog, Oct 4, and Sox 2, was observed after 4 days of differentiation (Figure 1A). Downregulation of the pluripotency genes continued concurrently with progression of endothelial differentiation for the duration of the experiment (Figure 1A). As shown in supplemental Figure II, this reduction in pluripotency-associated mRNAs was mirrored in assessments performed using a lentiviral-driven promoter-reporter system. Nanog promoter-driven destablished green fluorescent protein (dsGFP) expression was expressed highly in pluripotency and did not significantly change in samples maintained in pluripotent conditions. In contrast, hESCs transduced with plasmid green zeo (pGZ)-Nanog-dsGFP lentivirus demonstrate a substantial reduction in transgene expression in concordance with the progression of differentiation. No Nanog-dsGFP expression was found localized within cells expressing CD31 protein after 21 days of directed differentiation.
As shown in Figure 1B, induction of mesoderm-specific mRNAs was confirmed with maximal expression of brachyury, NKX2.5, and Mesp1 and Mixl1, observed at 4, 7, and 10 days of differentiation, respectively (Figure 1B). An early endothelial marker, kinase insert domain reception (KDR), was significantly induced after only 4 days of differentiation, with peak mRNA expression observed on day 7. Mature endothelial markers (ie, FMS-related tyrosine kinase-1 (FMT1), CD31, and VE-cadherin) were all significantly induced, concomitant with progression of endothelial differentiation. The induction of other germ layers or trophoblasts was not detected (supplemental Table I).
Protein expression of pluripotency (stage-specific embryonic antigen-4 [SSEA4] and Tra-1–60), and endothelial (KDR, CD31, C105, and VE-cadherin) and hematopoietic (CD45) markers was analyzed at progressing stages of differentiation. We observed a time-dependent increase in CD31 and VE-cadherin (Figure 2A and supplemental Figure III), consistent with the progression of differentiation, with a concomitant decrease in the percentage of cells expressing 1 or both of the pluripotency markers (SSEA4 and Tra-1–60), from 98.34±3.22% on 0 days to 8.15±3.48% on 10 days and to 6.54±5.26% by 21 days (supplemental Figure III). Within the pluripotent population, 27.74±1.79% of cells expressed KDR, with a peak at 7 days (84.78±1.39%) (supplemental Figure III), in agreement with mRNA data (Figure 1). We observed only 9.52±3.41% CD45-positive cells within the 21-day hESC-EC cell population, with 94.8±5.62% of cells expressing CD31 or CD105 (supplemental Figure III) and 81.59±2.11% of cells coexpressing CD31 and VE-cadherin. The expression of the previously mentioned endothelial markers increased in a spatial and temporal manner in accordance with the progression of endothelial differentiation (Figure 2A). Furthermore, cells retained the ability to proliferate, as determined by dual Ki67 and CD31 or VE-cadherin staining after subpassages (supplemental Figure IV).
An analysis of 55 angiogenesis-related proteins was performed at 0-, 10-, 14-, and 21-day points by angiogenesis arrays. The induction of all angiogenesis-related proteins, with the exception of basic fibroblast growth factor, was observed in a time-dependent manner (Figure 2B), with marked induction of insulinlike growth factor binding proteins 2 and 3, matrix metalloproteinase 9, tissue inhibitor of metalloproteinases-1 (TIMP1), and urokinase-type plasminogen activator proteins, observed in 10-day differentiated cells (Figure 2B and supplemental Figure V). At 14 days, hESC-ECs also expressed elevated dipeptidyl peptidase IV/CD26 and serpine1/plasminogen activator inhibitor type 1 proteins (Figure 2B and supplemental Figure V). At 21 days, hESC-ECs expressed a more complete repertoire of proteins examined, including vascular endothelial growth factor (VEGF) and VEGF-C (Figure 2B and supplemental Figure V).
hESC-ECs Produce NO and Respond to Pharmacological Stimulation
The capability of differentiated hESC-ECs to produce NO was visually assessed by DAF-FM diacetate fluorescence detection, with an increase in NO evidenced in a time-dependent manner from 0 to 21 days of differentiation (Figure 3A). This increase in NO was corroborated by quantification of NO using the Greiss reaction of cell supernatants. hESC-ECs that differentiated for 4 days or longer produced significantly greater NO than pluripotent cells (Figure 3A). Furthermore, 21-day hESC-ECs produce similar levels of NO (Figure 3A) to HUVECs and ECs prepared from the human vena saphena (hSVECs) (supplemental Figure VI). In addition, we assessed the response of differentiated hESC-ECs to an NO stimulator, carbachol,21 and demonstrated that hESC-ECs that differentiated for 10 days or longer produce significantly more NO in response (Figure 3B).
Dynamic Changes in MicroRNA Expression Profiles on Differentiation
The expression of miRNAs associated with angiogenesis or suppressed angiogenesis was analyzed in cells differentiated for 10 days. The expression of miRNAs associated with angiogenesis (let-7b, 7f, miR-126, 130a, 133a, 133b, 210, and 296)22 was induced (Figure 4), and the expression of miRNAs associated with impaired angiogenesis (miR-20a, 20b, 221, and 222)22 was suppressed in 10-day differentiated cells (Figure 4). Individual RT-PCR assays for the validation of a selection of the previously mentioned miRNAs were performed (supplemental Figure VII).
Evaluation of Wound-Closing Capability In Vitro
Pluripotent hESCs and hESC-ECs differentiated at varying stages, and hSVECs were subject to scratch wounding to evaluate cell migration.17 hESC-ECs that differentiated for 7, 10, 14, and 21 days demonstrated migration across a wound evaluated over 12 hours (Figure 5). hESC-ECs that differentiated for 21 days migrated significantly further than hSVECs after 12 hours, with complete wound closure evident (Figure 5). No significant differences were observed in 10-, 14-, and 21-day hESC-ECs and hSVECs over 3 to 9 hours (Figure 5).
Evaluation of Angiogenesis
In vitro, 10-, 14-, and 21-day differentiated hESC-ECs were successfully propagated into growth factor–reduced Matrigel and imaged 72, 96, and 120 hours after seeding (supplemental Figure VIII). Network tubelike structures were evident after only 72 hours, with no difference observed in the rate of tube formation between hESC-ECs differentiated for 10, 14, or 21 days (supplemental Figure VIII).
To determine the therapeutic potential of hESC-ECs for vascular regeneration in vivo, 10-day differentiated hESC-ECs were intramuscularly injected into the ischemic adductor muscle and hind limb blood flow recovery was monitored over 14 days. Figure 6A shows representative Doppler images at 14 days after ischemia. As shown in Figure 6B, at both 7 and 14 days after the induction of ischemia, blood flow to the ischemic foot was improved by hESC-ECs only. As shown in Figure 6C and supplemental Figure IX, hESC-EC transplantation enhanced capillary density of the ischemic adductor at 14 days after ischemia. As shown by Figure 6D, confocal microscopy of hESC-EC–injected muscles demonstrated the presence of transplantation-derived hNAg-positive cells within CD31-positive microvessels and within microvessels that had bound the IV-infused human-specific lectin. hNAg-positive staining was observed in mice injected with hMVECs; however, no coexpression of hNAg and CD31 was observed. The absence of hNAg-positive staining and human-specific lectin detection in medium-injected muscles confirmed specificity of lectin for human cells (Figure 6D and supplemental Figure XB). Dual detection of mouse- and human-specific lectin demonstrated that hESC-EC cells engrafted into the host vasculature (supplemental Figure XB).
We have reported an efficient and directed differentiation protocol for the derivation of ECs from hESCs using a serum- and feeder-free system. Our data show rapid downregulation of pluripotency factors, concomitant with induction of vascular endothelial markers at the mRNA, miRNA, and protein levels. Moreover, we report the ability of hESC-ECs to respond to an NO stimulator, migrate, and spontaneously produce tube-like structures in monolayer cultures and when cells are cultured in Matrigel. Finally, we show that the transplantation of cells into the muscle of immunodeficient mice subjected to hind limb ischemia promoted therapeutic neovascularization and blood flow recovery via engraftment into the vasculature, whereas mature hMVECs were not able to promote such beneficial effects.
There are several previously published protocols describing hESC-EC differentiation to ECs; however, even using a 2D cross-species coculture system using immunomagnetic selection to isolate CD34-positive progenitors, low EC differentiation efficiency in the order of 10% was reported.13 Nonetheless, these ECs did form vessels in vivo.13 More recently, a 2D protocol was described with a reported EC differentiation efficiency of approximately 50% at the end of the differentiation period.23 In addition, although the protocol could be successfully adapted to serum free by supplementation with a serum replacement, a considerable population of unknown lineage and pluripotent cells remained, requiring subsequent immunomagnetic separation to isolate CD31-positive cells.23 Recently, 3 improved protocols have been reported using VEGF in addition to media components,24 hypoxia-induced differentiation,25 and suppression of the transforming growth factor β pathway,26 with considerable improvement to the proportion of cells successfully differentiating to an endothelial lineage. However, all protocols still used an initial EB formation step. Our monolayer differentiation protocol produced a highly pure EC population without awkward isolation steps. We observed a discrepancy in expected CD34 and KDR expression across our differentiation protocol; however, the basal expression of KDR is reportedly high in pluripotent hESC cells and remains so during endothelial differentiation,7,15 as confirmed by FACS analysis. We cannot explain the late (21 days), modest, and yet significant induction in CD34 mRNA expression within our hESC-ECs; however, we report that all other mesoderm and progenitor markers examined displayed a transient induction, with only approximately 10% of cells differentiating toward a hematopoietic lineage after 21 days.
The performance of our hESC-derived ECs was evaluated in vivo in a suitable model through which to evaluate the angiogenic potential of injected cells. More important, we compared the hESC-derived ECs with hMVECs, a mature EC type. hESC-ECs induced angiogenesis rapidly, as suggested by improved blood flow at 7 and 14 days after transplantation. These hESC-EC cells were significantly more angiogenic than an adult counterpart hMVEC. The effect was rapid and sustained over this time course. The injected hESC-ECs were taken at 10 days of differentiation because cells at this stage had not demonstrated spontaneous sprouting of tubelike structures in vitro; however, as evidenced herein, miRNAs associated with hESC differentiation and loss of stemness (Let-7 family) and mesoderm and mature endothelial markers were efficiently expressed. Other studies7–12 have assessed the expression of EC markers at various stages of hESC differentiation, injecting cells differentiated for longer than 14 days in in vivo assays; however, none have assessed the induction of angiogenesis-associated miRNAs.22 We report that miRNAs associated with impaired angiogenesis are reduced in agreement with the increase in angiogenesis-associated proteins. The expression of angiogenesis-associated mature miRNAs (miR-126, 130a, 133a and b, and 210)22 in our 10-day hESC-ECs provides the rationale for injection in vivo. Furthermore, we demonstrate that 10-day hESC-ECs produce NO and respond appropriately to a stimulator, indicating that the cells would be capable of angiogenesis.27 Conversely, angiogenesis arrays show that maximal levels of angiogenic factors were present in cells assessed at later points (21 days in particular). Therefore, it may be possible to further improve angiogenesis in vivo by assessment of cells that have been under directed differentiation for 21 days or by incorporation of some of the recent discoveries by others (eg, VEGF inclusion24 or transforming growth factor β manipulation26). However, the induction and isolation of hESC-ECs at the correct differentiation stage most appropriate for transplantation seems to be important; too mature (hMVECs),28 highly differentiated,4,6 or very immature (VEGF) R2-positive Tra-1–60–positive differentiating hESC cells4 and no improvement in local blood flow have been reported. Therefore, our cells, isolated at 10 days of differentiation, still present high levels of mesoderm markers (eg, brachyury, Mesp1, and Nkx2.5), although levels of these genes are still elevated at 21 days, concurrent with miRNAs associated with angiogenesis. Furthermore, 10-day hESC-ECs formed network tubelike structures at the same rate as 14- and 21-day hESC-ECs. This may suggest that transplantation of cells at this stage may be considered a “progenitor” cell type, capable of further differentiation in vivo. Although there may remain potential for improving the optimal time for harvesting of cells, it is apparent that harvesting at 10 days appears effective. This is evident from our observations of efficient angiogenesis in vivo, evoked potentially as a response to paracrine mechanisms, including miRNA-mediated proangiogenic effects, with direct hESC-EC engraftment into vessels. Indeed, our method for differentiating ECs from hESCs may also provide an efficient and rapid in vitro system of screening for novel pharmaceutical therapeutic interventions, such as angiogenic or antiangiogenic small molecules, with mature endothelial markers and tube formation observed within 21 days.
In conclusion, this is the first study to demonstrate a fully scalable feeder- and serum-free method for the derivation of functional ECs without EB requirement. This differentiation system may provide a renewable source of ECs for potential applications, such as cellular therapy for the amelioration of regional ischemic tissue or for in vitro assays, to assist in the search for novel gene products to develop new therapeutic approaches for vascular regeneration.
Sources of Funding
This study was supported by the British Heart Foundation and ITI Life Sciences.
Received on: September 11, 2009; final version accepted on: April 15, 2010.
Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM. Embryonic stem cell lines derived from human blastocysts. Science. 1998; 282: 1145–1147.
Sone M, Itoh H, Yamahara K, Yamashita JK, Yurugi-Kobayashi T, Nonoguchi A, Suzuki Y, Chao T-H, Sawada N, Fukunaga Y, Miyashita K, Park K, Oyamada N, Sawada N, Taura D, Tamura N, Kondo Y, Nito S, Suemori H, Nakatsuji N, Nishikawa S-I, Nakao K. Pathway for differentiation of human embryonic stem cells to vascular cell components and their potential for vascular regeneration. Arterioscler Thromb Vasc Biol. 2007; 27: 2127–2134.
Yamahara K, Sone M, Itoh H, Yamashita JK, Yurugi-Kobayashi T, Homma K, Chao TH, Miyashita K, Park K, Oyamada N, Sawada N, Taura D, Fukunaga Y, Tamura N, Nakao K. Augmentation of neovascularizaiton in hindlimb ischemia by combined transplantation of human embryonic stem cells-derived endothelial and mural cells. PLoS One. 2008; 3: e1666.
Levenberg S, Golub JS, Amit M, Itskovitz-Eldor J, Langer R. Endothelial cells derived from human embryonic stem cells. Proc Natl Acad Sci U S A. 2002; 99: 4391–4396.
Cho SW, Moon SH, Lee SH, Kang SW, Kim J, Lim JM, Kim HS, Kim BS, Chung HM. Improvement of postnatal neovascularization by human embryonic stem cell derived endothelial-like cell transplantation in a mouse model of hindlimb ischemia. Circulation. 2007; 116: 2409–2419.
Vodyanik MA, Bork JA, Thomson JA, Slukvin II. Human embryonic stem cell-derived CD34+ cells: efficient production in the coculture with OP9 stromal cells and analysis of lymphohematopoietic potential. Blood. 2005; 105: 617–626.
Kaufman DS, Hanson ET, Lewis RL, Auerbach R, Thomson JA. Hematopoietic colony-forming cells derived from human embryonic stem cells. Proc Natl Acad Sci U S A. 2001; 98: 10716–10721.
Emanueli C, Minasi A, Zacheo A, Chao J, Chao L, Salis MB, Straino S, Tozzi MG, Smith R, Gaspa L, Bianchini G, Stillo F, Capogrossi MC, Madeddu P. Local delivery of human tissue kallikrein gene accelerates spontaneous angiogenesis in mouse model of hindlimb ischemia. Circulation. 2001; 103: 125–132.
Madeddu P, Emanueli C, Pelosi E, Salis MB, Cerio AM, Bonanno G, Patti M, Stassi G, Condorelli G, Peschle C. Transplantation of low dose CD34+KDR+ cells promotes vascular and muscular regeneration in ischemic limbs. FASEB J. 2004; 18: 1737–1739.
Invernici G, Emanueli C, Madeddu P, Cristini S, Gadau S, Benetti A, Ciusani E, Stassi G, Siragusa M, Nicosia R, Peschle C, Fascio U, Colombo A, Rizzuti T, Parati E, Alessandri G. Human fetal aorta contains vascular progenitor cells capable of inducing vasculogenesis, angiogenesis, and myogenesis in vitro and in a murine model of peripheral ischemia. Am J Pathol. 2007; 170: 1879–1892.
Nourse MB, Halpin DE, Scatena M, Mortisen DJ, Tulloch NL, Hauch KD, Torok-Storb B, Ratner BD, Pabon L, Murry CE. VEGF induces differentiation of functional endothelium from human embryonic stem cells: implications for tissue engineering. Arterioscler Thromb Vasc Biol. 2010; 30: 80–89.
Sonia P-L, Ana C, Ana A, Magdalena M-L, Carmen E-L, Carolina G, Sonia T, Dario M, David B, David M, Silvia S-G, Pilar S, Stefan G, José Enrique OC, Rubén M, Joaquín D, Deborah JB, Miodrag S. Hypoxia promotes efficient differentiation of human embryonic stem cells to functional endothelium. Stem Cells. 2010; 28: 407–418.
James D, Nam HS, Seandel M, Nolan D, Janovitz T, Tomishima M, Studer L, Lee G, Lyden D, Benezra R, Zaninovic N, Rosenwaks Z, Rabbany SY, Rafii S. Expansion and maintenance of human embryonic stem cell-derived endothelial cells by TGFbeta inhibition is Id1 dependent. Nat Biotechnol. 2010; 28: 161–166.
Kalka C, Masuda H, Takahashi T, Kalka-Moll WM, Silver M, Kearney M, Li T, Isner JM, Asahara T. Transplantation of ex vivo expanded endothelial progenitor cells for therapeutic neovascularization. Proc Natl Acad Sci U S A. 2000; 97: 3422–3427.