Therapeutic angiogenesis/vasculogenesis is promising for ischemia diseases. Due to the limited source of endothelial progenitor cells, hESCs represent a new and exciting avenue of angiogenic therapy. However, traditional spontaneous 3D method by EB differentiation of hESCs into endothelial cells is inefficient and the 2D method by co-culture with animal materials such as MEFs will limit its future clinic usage 
. Therefore, we sought to develop an extracellular matrix (ECM) culture system for increasing endothelial differentiation of hESCs and ensuring animal product free cells. In the past decade, there has been an increasing consensus that the ECM contains critical signals that regulate endothelial differentiation and blood vessel formation during development and disease progression 
Here we describe a two-step differentiation procedure that takes advantage of collagen I to efficiently promote differentiation of hESCs into endothelial lineage cells. Compared to previous 3D EB methods 
, this new protocol is more efficient for deriving hESC-ECs, and may be utilized to generate large numbers of endothelial cells for therapeutic application. Our qRT-PCR data confirmed that angiogenesis from EB sprouting in collagen triggered release of gelatinase, matrix metalloproteinases (MMPs), tissue inhibitors of MMP (TIMP), and upregulated endothelial progenitor markers. We believe changes in endothelial gene expression and signaling lead to events that not only stimulate vascular morphogenesis but also promote endothelial differentiation. Global gene expression profiling of hESCs during directed endothelial differentiation revealed significant enrichment of a number of endothelial genes and appropriate biological processes that correspond to each differentiation stage. Results were confirmed using qPCR analysis. Though hESC-ECs and HUVECs revealed a high degree of similarity between their respective expression profiles, hESC-ECs do not resemble HUVEC completely. Previous reports have shown that hemodynamic forces such as fluid shear stress can affect gene transcription in vivo
and in vitro 
. This suggests that shear stress on HUVEC may contribute to the discrepancy of expression profiles between HUVEC and hESC-ECs (which have not been exposed to hemodynamic forces during the 3-week differentiation process). Thus, to generate more functional and mature endothelial cells, shear stress conditioning of hESC-ECs may be needed in the future. Importantly, our microarray analysis provides a greater understanding of the patterns of activation of specific genes during endothelial differentiation, and identifies novel gene targets that may be used to enhance endothelial differentiation in the future.
To understand whether hESC-ECs also possess vasculogenic ability in vivo
, we next evaluated the in vivo
vascular formation potential of hESC-ECs after transplantation in mouse dorsal window chamber model. Previous data showed that supporting fibroblast cells were needed to stabilize engineered blood vessels 
. However, there are obvious drawbacks in using mouse support cells in the study of vessel maturation of human cell. Here we introduced collagen type I and Matrigel, without support cells, and showed hESC-ECs can form functional vasculature and are stable up to 2 months post transplantation. At present, we speculate that Matrigel, which is abundant in growth factor and extracellular matrix other than collagen I, may help stabilize the newly formed blood vessels.
An important observation in the present study is that the engraftment of hESC-ECs can improve cardiac function short-term. In recent years, several animal studies have shown that transplantation of EPCs isolated from the peripheral blood or bone marrow can improve cardiac function 
. In this study, we have shown that transplantation of hESC-ECs can also result in significant improvements in cardiac function at week 2. However, this was not sustained at a significant level beyond 4 weeks. We believe this is likely due to massive cell death within the first few weeks of transplantation, as indicated by our serial bioluminescence imaging studies to assess longitudinal cell fate. These findings indicate that other mechanisms such as activation of paracrine pathways may play an important role on cardiac performance amelioration at week 2 
, but additional studies will be needed in the future to test this hypothesis. Indeed, one can speculate that the acute donor cell death phenomenon may also explain the lack of long-term functional benefits in several of the clinical trials completed so far 
. The reasons for acute donor cell death are likely multi-factorial, and may involve immunogenicity due to xenotransplantation of human cells (note SCID mice still have natural killer cells), as well as the lack of regional nutrients and signaling factors for maintenance of cell viability. Following direct intramyocardial injection, the transplanted cells formed clumps as confirmed by histology. This is in contrast to the Matrigel plug and dorsal skin fold chamber data shown in , which however is a static and artificial environment loaded with collagen and Matrigel, and does not reflect the stress imposed upon hESC-ECs within a fast-beating ischemic heart. Without adequate host blood supply, nutrient transportation to the hESC-ECs is only by diffusion and is likely inadequate. The survival of transplanted cells could further be exacerbated in the context of myocardial infarction. Altogether, these factors contributed to the loss of bioluminescence imaging signals over time and the lack of long-term benefit in cardiac function, which is also consistent with recent findings based on transplantation of hESC-derived cardiomyocytes 
. Because of these obstacles, the development of a strategy to alleviate apoptotic cell death may be of primary importance for hESC-ECs therapy. Recent work on hESC-derived cardiomyocyte therapy revealed that Matrigel combined with pro-survival growth factor cocktail resulted in reliable formation of substantial myocardial grafts and significant functional improvement 
. Thus, tissue engineering techniques, rather than direct stem cell transplantation, may prove to be a more viable approach in the future 
In summary, our results show that (i) endothelial differentiation efficiency was increased by a two-step differentiation of hESCs and with animal product free procedure by using serum free culture medium, (ii) hESC-ECs possess functional vasculogenic ability in vivo, and (iii) transplantation of hESC-ECs can significantly improve short-term cardiac function at 2 weeks. However, the numbers of surviving hESC-ECs in infarcted tissue decreased significantly after the first few weeks, demonstrating the requirement for close long-term follow-up in all cardiac cell transplantation studies. These data suggest that prolonged heart function recovery may require more permanent graft survival of transplanted cells. Alternative transplantation protocols with larger numbers of hESC-ECs, multiple injection time points, or addition of matrix or prosurvival factors to prevent donor cell death after transplantation could eventually lead to the realization of sustained enhancement of heart function post myocardial infarction.