WO2014006228A1

WO2014006228A1 - Vascular bed-specific endothelial cells

Info

Publication number: WO2014006228A1
Application number: PCT/EP2013/064410
Authority: WO
Inventors: Aernout Luttun; Xabier Lopez ARANGUREN; Giulia COPPIELLO; Manu BEERENS; Felipe Prosper; Xabier AGUIRRE
Original assignee: Katholieke Universiteit Leuven; FOUNDATION FOR APPLIED MEDICAL RESEARCH
Current assignee: Katholieke Universiteit Leuven; FOUNDATION FOR APPLIED MEDICAL RESEARCH
Priority date: 2012-07-06
Filing date: 2013-07-08
Publication date: 2014-01-09
Anticipated expiration: 2015-01-06
Also published as: WO2014006228A9; GB201212111D0

Description

VASCULAR BED-SPECIFIC ENDOTHELIAL CELLS

Background and Summary BACKGROUND OF THE INVENTION

A. Field of the Invention

This invention relates to methods for the generation and subsequent validation of vascular bed-specific endothelial cells (ECs) with specific genetic and functional signatures starting from cellular precursors and use thereof.

These vascular bed-specific ECs with specific genetic and functional signatures from cellular precursors of the endothelial lineage of the present invention are in a particular embodiment of present invention used for the generation of in vitro bioengineered tissue culture equivalents. These in vitro bioengineered tissue culture equivalents are in an additional embodiment used for drug toxicity testing or for the generation of bioengineered vessel conduits for in vivo transplantation. In yet another embodiment these cells generated by present invention are used for treatment of vascular bed-specific diseases affecting the brain {e.g., stroke), the liver {e.g., sinusoidal obstruction syndrome), the heart (myocardial ischemia), the extremities {e.g., peripheral vascular disease), amongst others. Several documents are cited throughout the text of this specification. Each of the documents herein (including any manufacturer's specifications, instructions, etc.) are hereby incorporated by reference; however, there is no admission that any document cited is indeed prior art of the present invention. B. Description of the Related Art ECs line the inside ('endo') of blood and lymphatic vessels. Morphological, functional and molecular analyses revealed significant heterogeneity among ECs at different levels within the (lymph) vascular tree (Aird, 2007a, b). First, arterial, venous or lymphatic ECs clearly have different characteristics. A second level of heterogeneity is seen when comparing vascular beds of different organs (Aird, 2007b). The latter heterogeneity is mostly found in the microvasculature (i.e., the capillary bed), which adapts to the unique needs of the underlying tissue (Auerbach et al., 1985). Finally, ECs have distinct characteristics in different species (Fajardo, 1989). This remarkable diversity finds its origin in a combination of factors. On the one hand, each EC has properties that are intrinsic (i.e., independent of external cues) and preserved upon isolation from their in vivo microenvironment. On the other hand, another part of the unique EC signature is rapidly lost upon in vitro culture (a phenomenon called 'phenotypic drift') since it is determined by communication with surrounding tissues and by exposure to certain external biomechanical or biochemical stimuli (Amatschek et al., 2007; Wick et al., 2007). Interestingly, such heterogeneity has also been described for vascular smooth muscle cells (SMCs) and pericytes, which are the cells that surround the endothelium in a large part of the vascular bed (Majesky, 2007). Little is currently known about the mechanisms that determine endothelial heterogeneity. Morphological diversity can be demonstrated by electron microscopy and vascular casting, thereby focusing on the presence or absence of specialised features. In general, three different EC types are distinguished morphologically: those lacking fenestrations ('continuous ECs'; e.g., in cardiac muscle, brain), those featuring fenestrations sealed by a diaphragm ('fenestrated ECs'; e.g., endocrine glands) and those featuring fenestrations without diaphragm ('discontinuous or sinusoidal ECs'; e.g., liver) (Pries and Kuebler, 2006). Molecular heterogeneity has been studied on the DNA/RNA level by microarray, substractive hybridisation or serial analysis of gene expression (SAGE), or at the protein level using two-dimensional electrophoresis, mass spectrometry or substractive antibody expression cloning. In vivo approaches, most of which are not possible in humans, include specific antigen staining, injection of lectin variants, injection of antibodies, the generation of knock-in reporter mice, in vivo promoter activity analysis, phage display (in which bacteriophages expressing peptides or antibody fragments are injected and specifically retained by certain vascular beds), and most recently laser capture microdissection. In the Examples of present application we show an integrated approach to study EC heterogeneity involving stem cells.

Several stem cell types have been considered as potential EC precursors, including unfractionated bone marrow cells, embryonic stem cells (ESCs), induced pluripotent stem cells (iPSCs), mesenchymal stem cells (MSCs), umbilical cord or peripheral blood mononuclear cells, adipose tissue-derived cells, endothelial progenitor cells (EPCs), blood outgrowth endothelial cells (BOECs), tissue resident progenitor cells, multipotent adult progenitor cells (MAPCs) and mesoangioblasts (MABs). There is significant heterogeneity among these progenitors in terms of their phenotypic/functional characteristics, their ability to incorporate into nascent blood vessels and their potential to be specialised into different subtypes of endothelium (Aranguren et al., 2007; Gulati et al., 2003; Yoder et al., 2007). This may in part relate to epigenetic determinants, such as differential promoter methylation or it may be speculated that organ- or vascular bed-specific endothelial precursors exist in the adult. In this context, it is interesting to mention that endothelial precursors have been found in the vascular wall of adults and embryos. Another reason for the observed heterogeneity may be that there are clonal variations within the same precursor populations.

The existence of EC heterogeneity has a tremendous clinical impact. It forms the basis of vascular bed-specific diseases (e.g., atherosclerosis, varicosis or lymphedema, restricted to arteries, veins and lymphatics, respectively) and contributes to the disappointing results and side-effects obtained with 'broad spectrum' (anti-)angiogenic treatments in patients. In addition, it determines the vascular tropism of metastasising tumour cells and is the culprit for vascular bed- specific manifestations of acquired immunodeficiency syndrome (Conway and Carmeliet, 2004; Deng et al., 2006; Goerdt and Sorg, 1992; Ribatti et al., 2002). The existence of vascular bed-specific factors {e.g., endocrine gland vascular endothelial growth factor or EG-VEGF and gonadotropins) and inhibitors {e.g., chondromodulin-l) indeed suggests that (anti)angiogenic therapy with ubiquitous growth factors (e.g., VEGF) or inhibitors is not the ideal treatment. Moreover, EC progenitor-based revascularisation approaches have not asked whether the transplanted cells acquire the desired EC phenotype once engrafted in a diseased tissue where environmental cues are absent.

There is a need in the art for protocols to generate (stem cell-derived) pre- specialised ECs. Present invention fulfils this need by unravelling mechanisms of EC heterogeneity and offers new methods to design tailor-made therapies to cure vessel-related diseases. In addition, by revealing the 'vascular address' of each vessel, it allows targeting of systemically applied therapeutic agents to the region of interest, thereby minimising the risk for side-effects in other vascular territories.

This invention goes beyond the state-of-the-art with an unprecedented and innovative integrated in vitro/in vivo multi-disciplinary approach based on stem/progenitor cells and small animal models to: (/^') expand our knowledge of EC diversity by obtaining EC type and vascular bed-specific gene-profiles ('signatures'); (/^'/^') exploit that knowledge to design protocols to generate specialised ECs by differentiation from (stem/progenitor) cells in order to design specialised vascular therapies for (lymph)vascular disorders, to assay these signatures in pathological conditions or to use them in drug screening systems.

SUMMARY OF THE INVENTION

There is a need in the art for vascular bed-specific ECs to be used for therapeutic and diagnostic purposes, since these cells cannot be isolated in sufficient numbers from donors. According to the invention, this problem is solved by means of a method to generate these cells in sufficient numbers, which represents a significant advancement.

Such embodiment of present invention is schematically summarised in Figure 1 below.

A comparative transcriptomic screen on two categories of ECs: microvascular ECs (those from capillaries or 'microvessels' in three different murine/human organs) and macrovascular ECs (those from large vessels such as arteries or veins; we used arteries and veins from human umbilical cord and from adult mouse vena cava and thoracic aorta) delivered relevant information to develop such a method. Importantly, we used freshly isolated cells for the screen and not cultured cells (like the majority of existing screens), as the former are representative for vascular bed-specific ECs in their in vivo tissue context.

Further scope of applicability of the present invention will become apparent from the detailed description given hereinafter. However, it should be understood that the detailed description and specific examples, while indicating preferred embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

Some embodiments of the invention are set forth in claim format directly below:

An engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell, the cell being genetically altered by transduction or transfection of a source ceil with a transcription factor of the group consisting of Meox2, Tcf 15, Ppary, Wt1 , Ebf3, Zic3, Left , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthoiogue thereof, having the same biological function, or a combination thereof.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the ceil being genetically altered by transduction or transfection of a source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; and/or with a transcription factor of the group consisting of Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof. As will be evident from the examples hereinafter, the first group is associated with differentiation into arterial endothelial cells, and the second group is associated with differentiation into microvascular endothelial cells.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source ceil with a transcription factor of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof; in particular said source cell is transduced or transfected with the group of transcription factors consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3, eventually including human homologues or variants, member of the same family, or orthologues thereof. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with the transcription factor Prdm16 and one or more transcription factors selected of the group consisting of Emx2, Msx1 , Nkx2-3, Tox2, and Aff3; or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial ceil (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source eel! with a transcription factor Meox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Meox2 in combination with one ore more transcription factors selected from the group consisting of Tcf15, Pparv, Wt1 , Ebf3, Zic3, Lett , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twist 1 , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16: more in particular Meox2 in combination with one ore more transcription factors selected from the group consisting of Tcf15, Pparv, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2; even more in particular Meox2 in combination with one ore more transcription factors selected from the group consisting of Tctt S, Pparv, Wt1 , and Ebf3.

The engineered vascular bed-specific endothelial cell (the 'target ceil') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell a transcription factor with Tcf15 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular TcF15 in combination with one ore more transcription factors selected from the group consisting of Meox2, Pparv, Wt1 , Ebf3, Zic3, Lett , Foxf2, Foxfl a, Foxd , Tcfec, Hoxb5, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16: more in particular TcF15 in combination with one ore more transcription factors selected from the group consisting of Meox2, Pparv, Wt1 , Ebf3, Zic3, Lett , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2; even more in particular TcF15 in combination with one ore more transcription factors selected from the group consisting of Meox2, Pparv, Wt1 , and Ebf3.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the ceil being genetically altered by transduction or transfection of a source ceil with a transcription factor with Ppary or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Ppary in combination with one ore more transcription factors selected from the group consisting of eox2, TcF15, Wt1 , Ebf3, Zic3, Left , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16: more in particular Ppary in combination with one ore more transcription factors selected from the group consisting of Meox2, TcF15, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2; even more in particular Ppary in combination with one ore more transcription factors selected from the group consisting of Meox2, Meox2, TcF15, Wt1 , and Ebf3.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Wt1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Wt1 in combination with one ore more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16: more in particular Wt1 in combination with one ore more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2; even more in particular Wt1 in combination with one ore more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, and Ebf3.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Zic3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Zic3 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Leff , Foxf2, Foxfl a, Foxd , Tcfec, Hoxb5, af, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Zic3 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary,

TcF15, Ebf3, Wt1 , Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; even more in particular Zic3 in combination with one or more transcription factors selected from the group consisting of Lef1 , Foxf2, Foxfl a, and Foxd .

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Lef1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Lef1 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Lef1 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Foxf2, Foxfl a, Foxd , Tcfec, Hoxb5, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; even more in particular Lef1 in combination with one or more transcription factors selected from the group consisting of Zic3, Foxf2, Foxfl a, and Foxd .

The engineered vascular bed-specific endothelial ceil (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Foxf2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Foxf2 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Foxf2 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxfl a, Foxd , Tcfec, HoxbS, af, Cux2,

Gata4, Meis2, Twistl , and Zeb2; even more in particular Foxf2 in combination with one or more transcription factors selected from the group consisting of Zic3, Lef1 , Foxfl a, and Foxd .

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the ceil being genetically altered by transduction or transfection of a source cell with a transcription factor Foxfl a or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Foxfl a in combination with one or more transcription factors selected from the group consisting of Meox2, Ppar , TcF15, Ebf3, Wt1 , Zic3, Left , Foxf2, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Foxfl a in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; even more in particular Foxfl a in combination with one or more transcription factors selected from the group consisting of Zic3, Lef1 , Foxf2, and Foxd .

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Foxd or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Foxd in combination with one or more transcription factors selected from the group consisting of Meox2, Pparv, TcF1 5, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm 16; more in particular Foxd in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; even more in particular Foxd in combination with one or more transcription factors selected from the group consisting of Zic3, Lef1 , Foxf2, and Fox l a.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Tcfec or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Tcfec in combination with one or more transcription factors selected from the group consisting of Meox2, Pparv, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, Hoxb5,

Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Tcfec in combination with one or more transcription factors selected from Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; even more in particular Tcfec in combination with one or more transcription factors selected from HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor HoxbS or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular HoxbS in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and

Prdm16; more in particular HoxbS in combination with one or more transcription factors selected from Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; even more in particular HoxbS in combination with one or more transcription factors selected from Tcfec, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Maf or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Maf in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Maf in combination with one or more transcription factors selected from Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Cux2, Gata4, Meis2, Twistl , and Zeb2; even more in particular Maf in combination with one or more transcription factors selected from Tcfec, HoxbS, Cux2, Gata4, Meis2, Twistl , and Zeb2.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil of embodiment 1 , the ceil being genetically altered by transduction or transfection of a source cell with a transcription factor Cux2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Cux2 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lefl , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Cux2 in combination with one or more transcription factors selected from Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Gata4, Meis2, Twistl , and Zeb2; even more in particular Cux2 in combination with one or more transcription factors selected from Tcfec, HoxbS, Maf, Gata4, Meis2, Twistl , and Zeb2. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Gata4 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Gata4 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lefl , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Gata4 in combination with one or more transcription factors selected from Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Meis2, Twistl , and Zeb2; even more in particular Gata4 in combination with one or more transcription factors selected from Tcfec, HoxbS, Maf, Cux2, Meis2, Twistl , and Zeb2. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Meis2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Meis2 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Leff , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Gata4, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Meis2 in combination with one or more transcription factors selected from Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, Hoxb5, Maf, Cux2, Gata4, Twistl , and Zeb2; even more in particular Meis2 in combination with one or more transcription factors selected from Tcfec, HoxbS, Maf, Cux2, Gata4, Twistl , and Zeb2.The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the ceil being genetically altered by transduction or transfection of a source cell with a transcription factor Twistl or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Twistl in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Le l , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Twistl in combination with one or more transcription factors selected from Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, and Zeb2; even more in particular Twistl in combination with one or more transcription factors selected from Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, and Zeb2. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function; in particular Zeb2 in combination with one or more transcription factors selected from the group consisting of Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16; more in particular Zeb2 in combination with one or more transcription factors selected from Meox2, Ppary, TcF15, Ebf3, Wt1 , Zic3, Lef1 , Foxf2, Foxfl a, Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, and Twistl ; even more in particular Zeb2 in combination with one or more transcription factors selected from Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, and Twistl . The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Emx2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Msx1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Tox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , whereby the cell is being genetically altered by transduction or transfection of a source cell with a transcription factor of the group consisting of eox2, Tcf15, Ppary, WT1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, Hoxb5, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an organ-specific microvascular endothelial cell. The engineered vascular bed-specific endothelial cell (the 'target ceil') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell; in particular said source cell is transduced or transfected with the group of transcription factors consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3, eventually including human homologues or variants, member of the same family, or orthologues thereof. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , whereby the cell is being engineered by transduction or transfection of a source cell with the transcription factor Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into an arterial endothelial cell.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , whereby the is the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 and Ebf3 or a human homoiog or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source ceil into a cardiac microvascular endothelial cell; in particular said source cell is transduced or transfected with the group of transcription factors consisting of eox2, Tcf15, Ppary, Wt1 and Ebf3, eventually including human homologues or variants, member of the same family, or orthologues thereof.The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , whereby the is the cell being engineered by transduction or transfection of a source ceil with a transcription factor of the group consisting of Zic3, Lef1 , Foxf2, Foxf 1 a and Foxd or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source ceil into a brain microvascular endothelial cell; in particular said source cell is transduced or transfected with the group of transcription factors consisting of Zic3, Lef1 , Foxf2, Foxf 1 a and Foxd , eventually including human homologues or variants, member of the same family, or orthologues thereof. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target ceil of embodiment 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a liver microvascular endothelial ceil; in particular said source cell is transduced or transfected with the group of transcription factors consisting of Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2, eventually including human homologues or variants, member of the same family, or orthologues thereof.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell; in particular said source cell being engineered by transduction or transfection of a source cell with the transcription factor Prdm16 in combination with one or more transcription factors selected from the group consisting of Emx2, Nkx2-3, Msx1 , Tox2, and Aff3; more in particular said source cell is transduced or transfected with the group of transcription factors consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16, eventually including human homologues or variants, member of the same family, or orthologues thereof.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous embodiments, whereby the transcription factors are overexpressed.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous embodiments, whereby the source cell is a blood outgrowth endothelial cell (BOEC).

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous embodiments, whereby the source cell is a dedifferentiated cultured human umbilical vein endothelial cell (HUVEC) or a dedifferentiated HUAEC.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous embodiments, whereby the source cell is a mammalian eel! of embryonic or non-embryonic origin.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous embodiments, whereby the source ceil is a source cell of the group consisting of endothelial progenitor cells (EPCs), mesenchymal stem cells (MSCs), embryonic stem cells (ESCs), induced pluripotency stem ceils (iPSCs), mesoangioblasts (MABs), multipotent adult progenitor cells (MAPCs), blood outgrowth endothelial cells (BOECs), induced endothelial cells (iECs), unfractionated bone marrow cells, umbilical cord or peripheral blood-derived mononuclear cells, adipose tissue-derived cells, tissue resident progenitor ceils, human umbilical vein endothelial cells (HUVECs), human umbilical artery endothelial cells (HUAECs), human dermal microvascular endothelial cells (HDMECs), cultured endothelial cell lines from heart, cultured endothelial cell lines from liver, cultured endothelial cell lines from brain.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous embodiments, whereby the transduction or transfection is lentiviral.

An engineered vascular bed-specific endothelial cell, characterized in that the ceil is engineered by induced differentiation of a source cell by transferring directly to source cells the expression product of a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof the expression product being linked to a protein transduction domain (PTD), such as poly-arginine and HIV-derived Tat, or to small cationic peptide domains to enhance its cell membrane crossing capacity.

An engineered vascular bed-specific endothelial cell, characterized in that the cell is engineered by induced differentiation of a source cell by transferring directly to source cells the expression product of a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lefl , Foxf2, Foxfl a, Foxd , Tcfec, Hoxb5, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; and/or by induced differentiation of a source cell by transferring directly to source cells the expression product of a transcription factor of the group consisting of Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof the expression product being linked to a protein transduction domain (PTD), such as poly-arginine and HIV-derived Tat, or to small cationic peptide domains to enhance its cell membrane crossing capacity.

The engineered vascular bed-specific endothelial cell of embodiment 40, whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Left , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate said source ceil into a vascular endothelial target cell. The engineered vascular bed-specific endothelial cell of embodiment 41 , whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Left , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; and/or of the group consisting of Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate said source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial cell of embodiments 40 or 41 , whereby the expression product is of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a vascular endothelial target cell. 45. The engineered vascular bed-specific endothelial cell of embodiment 40 or 41 , whereby the expression product is of Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

As will be evident to the skilled artisan, in this and each of the following embodiments to the engineered vascular bed-specific endothelial cells, the particular embodiments for each of said transcription factors and the combination of each of said transcription factors with the further transcription factors as provided herein above, also apply. Said specific embodiments to each of said transcription factors and the combination of each of said transcription factors with the further transcription factors, also apply to the therapeutic and diagnostic applications (infra) as well as in the methods to obtain said cells as provided hereinafter.

46. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Meox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

47. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Tcf15 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

48. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Ppary or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

49. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Wt1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial ceil of any one of the previous embodiments, whereby the expression product is of Zic3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Lef1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target ceil.

The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Foxf2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial ceil of any one of the previous embodiments, whereby the expression product is of Foxf 1 a or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Foxd or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial ceil of any one of the previous embodiments, whereby the expression product is of Tcfec or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source ceil into a vascular endothelial target cell. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of HoxbS or a human homologue or a variant, member of the same family, orthoiogue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial ceil of any one of the previous embodiments, whereby the expression product is of Maf or a human homologue or a variant, member of the same family, orthoiogue thereof, having the same biological function, or a human homolog to differentiate source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial ceil of any one of the previous embodiments, whereby the expression product is of Cux2 or a human homologue or a variant, member of the same family, orthoiogue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target ceil.

The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Gata4 or a human homologue or a variant, member of the same family, orthoiogue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Meis2 or a human homologue or a variant, member of the same family, orthoiogue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target ceil.

The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Twistl or a human homologue or a variant, member of the same family, orthoiogue thereof, having the same biological function, to differentiate source ceil into a vascular endothelial target cell.

The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Zeb2 or a human homologue or a variant, member of the same family, orthoiogue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

63. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Emx2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

64. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

65. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Msx1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

66. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Tox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

67. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a,

Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an organ-specific microvascular endothelial cell. 68. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell.

69. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into an arterial endothelial cell.

70. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 and Ebf3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a cardiac microvascular endothelial cell.

71 . The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of the group consisting of Zic3, Lef1 , Foxf2, Foxf 1 a and Foxd or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a brain microvascular endothelial cell.

72. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of the group consisting of Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a liver microvascular endothelial cell.

73. The engineered vascular bed-specific endothelial cell of any one of the previous embodiments, whereby the expression product is of the group consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell. 74. The engineered vascular bed-specific endothelial cell according to any one of the previous embodiments, whereby the transcription factors are overexpressed.

75. The engineered vascular bed-specific endothelial cell according to any one of the previous embodiments, whereby the source cell is a blood outgrowth endothelial cell (BOEC).

76. The engineered vascular bed-specific endothelial cell according to any one of the previous embodiments, whereby the source cell is a dedifferentiated cultured human umbilical vein endothelial cell (HUVEC) or a dedifferentiated HUAEC.

77. The engineered vascular bed-specific endothelial cell according to any one of the previous embodiments, whereby the source ceil is a mammalian cell of embryonic or non-embryonic origin.

78. The engineered vascular bed-specific endothelial cell according to any one of the previous embodiments, whereby the source cell is a source cell of the group consisting of endothelial progenitor cells (EPCs), mesenchymal stem cells (MSCs), embryonic stem cells (ESCs), induced pluripotency stem cells (iPSCs), mesoangiobiasts (MABs), multipotent adult progenitor cells (MAPCs), blood outgrowth endothelial cells (BOECs), induced endothelial cells (iECs), unfractionated bone marrow cells, umbilical cord or peripheral blood-derived mononuclear cells, adipose tissue-derived cells, tissue resident progenitor cells, human umbilical vein endothelial cells (HUVECs), human umbilical artery endothelial cells (HUAECs), human dermal microvascular endothelial ceils (HDMECs), cultured endothelial cell lines from heart, cultured endothelial cell lines from liver, cultured endothelial cell lines from brain.

79. The engineered vascular bed-specific endothelial cell according to any one of the previous embodiments, whereby the cell is terminally differentiated or in partially differentiated form.

80. The engineered ceil according to any one of the previous embodiments, for use in a treatment of vascular bed-specific disorder. 81 . The engineered cell according to any one of the previous embodiments, for use in a treatment of a disorder of vascular bed-specific haemostasis or for a disorder of generating and/or maintaining vascular bed-specific phenotypes.

82. The engineered cell according to any one of the previous embodiments 5, for use in a treatment of macrovascular complications evoked by a vascular bed-specific (haemostasis) disorder.

83. The engineered cell according to any one of the previous embodiments, for use in a treatment of microvascular complications evoked by a vascular bed- specific (haemostasis) disorder.

84. The engineered cell according to any one of the previous embodiments, for use in a treatment of for decreasing severity of vascular bed-specific disorder in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment with the engineered cell, thereby decreasing severity of disorder in the subject.

85. The engineered ceil according to any one of the previous embodiments, for use in a treatment of for decreasing severity of vascular bed-specific disorder or diseases or circulatory or hypoxic conditions that can be treated with the compositions and the methods of the invention comprise but are not limited to: atherosclerosis, preeclampsia, erectile dysfunction, renal failure, transplant accelerated arteriosclerosis, deep vein thrombosis, sleep apnea, hypoxia during sleep, fetal hypoxia, smoking, anemia, endothelial dysfunction, sinusoidal obstruction syndrome, regional perfusion deficits (e.g., limb, gut, renal ischemia), congestic heart failure, peripheral vascular disease, frost bite, decubitus ulcers, asphyxiation, poisoning (e.g., carbon monoxide, heavy metal), altitude sickness, pulmonary hypertension, sudden infant death syndrome, asthma, chronic obstructive pulmonary disease, congenital circulatory anomalities (e.g., Tetralogy of Fallot), erythroblastosis, myocardial infarction, aortic stenosis, or any other cardiac disease which can lead to heart failure, plaque rupture, both primary and secondary (in-stent) restenosis in coronary or peripheral arteries, coronary vascular disease, hypertension, diabetes, obesity, stroke, aneurysm, thrombosis, arrythmia, tachycardia, surgical or physical trauma in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment the engineered cell, thereby decreasing severity of disorder in the subject.

86. The engineered cell according to any one of the previous embodiments, for use in a treatment for decreasing seventy of vascular bed-specific diseases affecting the brain (e.g., stroke), the liver (e.g., sinusoidal obstruction syndrome), the heart (myocardial ischemia) or the extremities (e.g., peripheral vascular disease) in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment the engineered cell, thereby decreasing severity of disorder in the subject.

87. The engineered ceil for use in a treatment according to any one of the previous embodiments, whereby the cell is genetically altered or unaltered, by direct introduction to a site of interest, e.g., on or around the surface of an acceptable matrix, or systemicaily, in combination with a pharmaceutically acceptable carrier so as to repair, replace or promote the growth of existing and/or new blood vessels.

88. The engineered cell to any one of the previous embodiments,, whereby said cell is administered in the treatment by any one of the following methods: localised injection, catheter administration, systemic injection, intraperitoneal injection, parenteral administration, oral administration, intracranial injection, intra-arterial injection, intra-venous injection, intra-ventricular infusion, intra- placental injection, intra-uterine injection, surgical intra-myocardial injection, transendocardial injection, transvascular injection, intra-coronary injection, intra-muscuiar injection, surgical injection into a tissue of interest or via direct application to tissue surfaces (e.g., during surgery or on a wound).

89. A pharmaceutical composition, comprising an engineered ceil according to any one of the previous embodiments.

90. An in vitro bioengineered tissue, comprising an engineered cell according to any one of the previous embodiments.

91 . The use of an engineered cell according to any one of the previous embodiments, in a method of screening for non-toxic compounds.

92. The use of an engineered cell according to any one of the previous embodiments, in a method of in vitro bioengineering a tissue. 93. The use of an engineered cell according to any one of the previous embodiments, in in vitro drug toxicity testing, either using the cell product alone, or as a (or one of the) cellular component(s) in the context of engineered 2D or 3D tissue equivalents;

94. The use of an engineered cell according to any one of the previous embodiments as a (or one of the) cellular component(s) of tissue engineered constructs for implantation in patients (for example to coat the inside of artificial arterial conduits or as the intimal cellular component of a fully biological tissue-engineered vascular graft or to coat cardiac valves).

95. An in vitro method of diagnosing a vascular bed-specific disorder phenotype in a subject, said method comprising (a) analysing the level of expression or activity of expression product of at least 5 genes of the '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs) in a sample isolated from said subject, and (b) compare said level of expression or activity with the level of expression or activity in '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs); whereby a deviated level of expression or activity relative to such '(differential) reference signature' is an indication of such disorder phenotype or a propensity thereto.

96. An in vitro method of diagnosing a vascular bed-specific disorder phenotype in a subject, said method comprising: (a) genotyping one or more genes in the '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs) in a sample isolated from said subject, and (b) analyse the DNA sequence of said gene(s); whereby polymorphisms (e.g., SNPs) in said genes of the 'reference signatures' is an indication of such disorder phenotype or a propensity thereto.

97. Use of the '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs), to design tailored therapy, e.g., vascular bed-specific drug delivery or activation of reference signature genes to increase vascularisation.

98. Use of '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs) to read-out the (side) effects on the ECs of the targeted organ/tissue by current or novel (anti-)angiogenic or other therapies.

99. Use of '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs) to read-out the efficiency of existing or to be developed alternative (combinatorial) approaches not described herein to specify source ceils towards a vascular bed-specific target cell.

100. The engineered ceil according to any one of the previous embodiments, for use in a treatment of for decreasing severity of vascular bed-specific disorder or diseases or circulatory or hypoxic conditions that can be treated with the compositions and the methods of the invention comprise but are not limited to: atherosclerosis, preeclampsia, erectile dysfunction, renal failure, transplant accelerated arteriosclerosis, deep vein thrombosis, sleep apnea, hypoxia during sleep, fetal hypoxia, smoking, anemia, endothelial dysfunction, sinusoidal obstruction syndrome, regional perfusion deficits (e.g., limb, gut, renal ischemia), congestic heart failure, peripheral vascular disease, frost bite, decubitus ulcers, asphyxiation, poisoning (e.g., carbon monoxide, heavy metal), altitude sickness, pulmonary hypertension, sudden infant death syndrome, asthma, chronic obstructive pulmonary disease, congenital circulatory anomalities (e.g., Tetralogy of Fallot), erythroblastosis, myocardial infarction, aortic stenosis, or any other cardiac disease which can lead to heart failure, plaque rupture, both primary and secondary (in-stent) restenosis in coronary or peripheral arteries, coronary vascular disease, hypertension, stroke, aneurysm, thrombosis, arrythmia, tachycardia, surgical or physical trauma in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment the engineered cell, thereby decreasing seventy of disorder in the subject. 101 . A method to differentiate a source cell into a vascular endothelial cell (the target cell), said method comprising genetically altering the source cell by transduction or transfection of said source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppar , Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Fox l a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof; in particular said method comprising genetically altering the source cell by transduction or transfection of said source cell with a transcription factor of the group consisting of Meox2, Tcf15, Pparv, Wt1 , Ebf3, Zic3, Leff , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , and Zeb2; and/or with a transcription factor of the group consisting of Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof.

102. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with a transcription factor of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2- 3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof; in particular said source cell is transduced or transfected with the group of transcription factors consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3, eventually including human homoiogues or variants, member of the same family, or orthologues thereof.

103. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with a transcription factor Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

104. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Meox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function. 105. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source ceil with transcription factor Tcf15 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

106. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Pparv or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

107. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Wt1 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

108. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Zic3 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

109. The method of embodiment 101 , the source ceil being genetically altered by transduction or transfection of said source cell with transcription factor Lef1 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

1 10. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Foxf2 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

1 1 1 . The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Foxf 1 a or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

1 12. The method of embodiment 101 , the source ceil being genetically altered by transduction or transfection of said source cell with transcription factor Foxd or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function. . The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Tcfec or a human homologue or a variant, member of the same family, ortho!ogue thereof, having the same biological function.

. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor HoxbS or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Maf or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Cux2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Gata4 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of embodiment 1 , the ceil being genetically altered by transduction or transfection of a source cell with a transcription factor Meis2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Twistl or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Zeb2 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

121 . The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Emx2 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

122. The method of embodiment 101 , the source ceil being genetically altered by transduction or transfection of said source ceil with transcription factor Nkx2-3 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

123. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Msx1 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

124. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Tox2 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function.

125. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, WT1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, Hoxb5, af, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an organ-specific microvascular endothelial cell.

126. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with a transcription factor of the group of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate said source ceil into an arterial endothelial cell; in particular said source ceil is transduced or transfected with the group of transcription factors consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3, eventually including human homologues, variants, members of the same family, or orthoiogues thereof, to differentiate said source cell into an arterial endothelial cel.

127. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with transcription factor Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into an arterial endothelial cell.

128. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 and Ebf3 or a human homolog or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source ceil into a cardiac microvascular endothelial cell; in particular said source cell is transduced or transfected with the group of transcription factors consisting of Meox2, Tcf15, Pparv, Wt1 and Ebf3, eventually including human homologues, variants, members of the same family, or orthoiogues thereof, to differentiate said source cell into a cardiac microvascular endothelial cell

129. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source ceil with a transcription factor of the group consisting of Zic3, Lef1 , Foxf2, Foxfl a and Foxd or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a brain microvascular endothelial cell; in particular said source ceil is transduced or transfected with the group of transcription factors consisting of Zic3, Lef1 , Foxf2, Foxfl a and Foxd , eventually including human homologues, variants, members of the same family, or orthoiogues thereof, to differentiate said source cell into a brain microvascular endothelial ceil.

130. The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source ceil with a transcription factor of the group consisting of Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2,

Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a liver microvascular endothelial cell; in particular said source cell is transduced or transfected with the group of transcription factors consisting of Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2, eventually including human homologues, variants, members of the same family, or orthologues thereof, to differentiate said source cell into a liver microvascular endothelial cell.

131 . The method of embodiment 101 , the source cell being genetically altered by transduction or transfection of said source cell with a transcription factor of the group consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell; in particular said source cell is transduced or transfected with the group of transcription factors consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16, eventually including human homologues, variants, members of the same family, or orthologues thereof, to differentiate said source cell into an arterial endothelial cell .

132. The method according to any one of the previous embodiments 97 to

127, whereby the transcription factors are overexpressed.

133. The method according to any one of the previous embodiments 97 to

128, whereby the source cell is a blood outgrowth endothelial cell (BOEC).

134. The method according to any one of the previous embodiments 97 to 128, whereby the source cell is a dedifferentiated cultured human umbilical vein endothelial cell (HUVEC) or a dedifferentiated HUAEC.

135. The method according to any one of the previous embodiments 97 to 128, whereby the source cell is a mammalian cell of embryonic or non- embryonic origin.

136. The method according to any one of the previous embodiments 97 to 128, whereby the source cell is a source cell of the group consisting of endothelial progenitor cells (EPCs), mesenchymal stem cells (MSCs), embryonic stem cells (ESCs), induced pluripotency stem cells (iPSCs), mesoangioblasts (MABs), multipotent adult progenitor cells (MAPCs), blood outgrowth endothelial cells (BOECs), induced endothelial cells (iECs), unfractionated bone marrow cells, umbilical cord or peripheral blood-derived mononuclear cells, adipose tissue-derived ceils, tissue resident progenitor cells, human umbilical vein endothelial cells (HUVECs), human umbilical artery endothelial cells (HUAECs), human dermal microvascular endothelial cells (HDMECs), cultured endothelial cell lines from heart, cultured endothelial cell lines from liver, cultured endothelial cell lines from brain. 137. The method according to any one of the previous embodiments, whereby the transduction or transfection is lentiviral.

Further embodiments of the present invention in claim format are;

1 . An engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell, the cell being genetically altered by transduction or transfection of a source cell with a transcription factor of the group consisting of eox2, Tcf15, Pparv, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof.

2. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil of claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof.

3. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the ceil being genetically altered by transduction or transfection of a source ceil with a transcription factor Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target ceil') or the engineered source cell to differentiate into a vascular endothelial target ceil according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Meox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial ceil (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source ceil a transcription factor with Tcf15 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor with Pparv or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Wt1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil according to claim 1 , the cell being genetically altered by transduction or transfection of a source ceil with a transcription factor Zic3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function. 9. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Lef1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

10. The engineered vascular bed-specific endothelial cell (the 'target ceil') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell a transcription factor with Foxf2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

1 1 . The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Foxf 1 a or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

12. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Foxd or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

13. The engineered vascular bed-specific endothelial ceil (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source ceil with a transcription factor Tcfec or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

14. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor HoxbS or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target ceil') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source ceil with a transcription factor Maf or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Cux2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the ceil being genetically altered by transduction or transfection of a source cell with a transcription factor Gata4 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Meis2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Twistl or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function. 20. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

21 .The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Emx2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

22. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

23. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Msx1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

24. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor Tox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function.

25. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , whereby the cell is being genetically altered by transduction or transfection of a source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, Hoxb5, af, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an organ-specific microvascular endothelial cell. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target ceil according to claim 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell.

The engineered vascular bed-specific endothelial ceil (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , whereby the cell is being engineered by transduction or transfection of a source cell with the transcription factor Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into an arterial endothelial cell .

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , whereby the is the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of eox2, Tcf15, Ppary, Wt1 and Ebf3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a cardiac microvascular endothelial cell.

The engineered vascular bed-specific endothelial ceil (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell according to claim 1 , whereby the is the cell being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Zic3, Lef1 , Foxf2, Foxf 1 a and Foxd or a human homo!ogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a brain microvascular endothelial cell.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source ceil into a liver microvascular endothelial cell.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil according to claim 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil according to any one of the previous claims 1 to 31 , whereby the transcription factors are overexpressed.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims 1 to 32, whereby the source cell is a blood outgrowth endothelial cell (BOEC).

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims 1 to 32, whereby the source cell is a dedifferentiated cultured human umbilical vein endothelial cell (HUVEC) or a dedifferentiated HUAEC. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims 1 to 32, whereby the source cell is a mammalian ceil of embryonic or non-embryonic origin.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims 1 to 32, whereby the source cell is a source cell of the group consisting of endothelial progenitor cells (EPCs), mesenchymal stem cells (MSCs), embryonic stem cells (ESCs), induced pluripotency stem cells (iPSCs), mesoangioblasts (MABs), muitipotent adult progenitor cells ( APCs), blood outgrowth endothelial cells (BOECs), induced endothelial ceils (iECs), unfractionated bone marrow cells, umbilical cord or peripheral blood-derived mononuclear cells, adipose tissue-derived cells, tissue resident progenitor cells, human umbilical vein endothelial cells (HUVECs), human umbilical artery endothelial cells (HUAECs), human dermal microvascular endothelial ceils (HDMECs), cultured endothelial cell lines from heart, cultured endothelial cell lines from liver, cultured endothelial cell lines from brain.

The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims, whereby the transduction or transfection is lentiviral.

An engineered vascular bed-specific endothelial cell, characterized in that the cell is engineered by induced differentiation of a source cell by transferring directly to source cells the expression product of a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof the expression product being linked to a protein transduction domain (PTD), such as poly-arginine and HlV-derived Tat, or to small cationic peptide domains to enhance its ceil membrane crossing capacity. 39. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxff a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a vascular endothelial target cell.

40. The engineered vascular bed-specific endothelial ceil of claim 38, whereby the expression product is of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a vascular endothelial target ceil.

41 . The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

42. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Meox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

43. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Tcf15 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

44. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Ppary or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell. 45. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Wt1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell. 46. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Zic3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source ceil into a vascular endothelial target cell. 47. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Lef1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

48. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Foxf2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

49. The engineered vascular bed-specific endothelial ceil of claim 38, whereby the expression product is of Foxf 1 a or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

50. The engineered vascular bed-specific endothelial ceil of claim 38, whereby the expression product is of Foxd or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source ceil into a vascular endothelial target cell.

51 . The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Tcfec or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source ceil into a vascular endothelial target cell. 52. The engineered vascular bed-specific endothelial ceil of claim 38, whereby the expression product is of HoxbS or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell. 53. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Maf or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a human homolog to differentiate source cell into a vascular endothelial target ceil. 54. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Cux2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell. 55. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Gata4 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

56. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Meis2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

57. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Twistl or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell.

58. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target ceil.

59. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Emx2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell. 60. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell. 61 . The engineered vascular bed-specific endothelial ceil of claim 38, whereby the expression product is of Msx1 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a vascular endothelial target cell. 62. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Tox2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source eel! into a vascular endothelial target cell. 63. The engineered vascular bed-specific endothelial ceil of claim 38, whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an organ-specific microvascular endothelial ceil.

64. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial ceil.

65. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into an arterial endothelial cell.

66. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of the group consisting of Meox2, Tcf15, Pparv, Wt1 and Ebf3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a cardiac microvascular endothelial cell.

67. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of the group consisting of Zic3, Lef1 , Foxf2, Foxfl a and Foxd or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a brain microvascular endothelial cell. The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of the group consisting of Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a liver microvascular endothelial cell.

The engineered vascular bed-specific endothelial cell of claim 38, whereby the expression product is of the group consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell.

The engineered vascular bed-specific endothelial ceil according to any one of the previous claims 38 to 69, whereby the transcription factors are overexpressed.

The engineered vascular bed-specific endothelial cell according to any one of the previous claims 38 to 70, whereby the source cell is a blood outgrowth endothelial cell (BOEC).

The engineered vascular bed-specific endothelial cell according to any one of the previous claims 38 to 70, whereby the source ceil is a dedifferentiated cultured human umbilical vein endothelial cell (HUVEC) or a dedifferentiated HUAEC.

The engineered vascular bed-specific endothelial cell according to any one of the previous claims 38 to 70, whereby the source cell is a mammalian cell of embryonic or non-embryonic origin.

The engineered vascular bed-specific endothelial cell according to any one of the previous claims 38 to 70, whereby the source cell is a source cell of the group consisting of endothelial progenitor cells (EPCs), mesenchymal stem cells (MSCs), embryonic stem cells (ESCs), induced pluripotency stem cells (iPSCs), mesoangiobiasts (MABs), multipotent adult progenitor cells

(MAPCs), blood outgrowth endothelial cells (BOECs), induced endothelial cells (iECs), unfractionated bone marrow cells, umbilical cord or peripheral blood-derived mononuclear cells, adipose tissue-derived cells, tissue resident progenitor cells, human umbilical vein endothelial cells (HUVECs), human umbilical artery endothelial ceils (HUAECs), human dermal microvascular endothelial cells (HDMECs), cultured endothelial cell lines from heart, cultured endothelial cell lines from liver, cultured endothelial cell lines from brain.

75. The engineered vascular bed-specific endothelial ceil according to any one of the previous claims 38 to 70, whereby the cell is terminally differentiated or in partially differentiated form.

76. The engineered cell according to any one of the previous claims 1 to 75, for use in a treatment of vascular bed-specific disorder.

77. The engineered cell according to any one of the previous claims 1 to 75, for use in a treatment of a disorder of vascular bed-specific haemostasis or for a disorder of generating and/or maintaining vascular bed-specific phenotypes.

78. The engineered cell according to any one of the previous claims 1 to 75, for use in a treatment of macrovascular complications evoked by a vascular bed-specific (haemostasis) disorder.

79. The engineered cell according to any one of the previous claims 1 to 75, for use in a treatment of microvascular complications evoked by a vascular bed- specific (haemostasis) disorder.

80. The engineered cell according to any one of the previous claims 1 to 75, for use in a treatment of for decreasing severity of vascular bed-specific disorder in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment with the engineered cell, thereby decreasing severity of disorder in the subject.

81 .The engineered cell according to any one of the previous claims 1 to 75, for use in a treatment of for decreasing severity of vascular bed-specific disorder or diseases or circulatory or hypoxic conditions that can be treated with the compositions and the methods of the invention comprise but are not limited to: atherosclerosis, preeclampsia, erectile dysfunction, renal failure, transplant accelerated arteriosclerosis, deep vein thrombosis, sleep apnea, hypoxia during sleep, fetal hypoxia, smoking, anemia, endothelial dysfunction, sinusoidal obstruction syndrome, regional perfusion deficits (e.g., limb, gut, renal ischemia), congestic heart failure, peripheral vascular disease, frost bite, decubitus ulcers, asphyxiation, poisoning (e.g., carbon monoxide, heavy metal), altitude sickness, pulmonary hypertension, sudden infant death syndrome, asthma, chronic obstructive pulmonary disease, congenital circulatory anomalities (e.g., Tetralogy of Fallot), erythroblastosis, myocardial infarction, aortic stenosis ro any other cardiac disease which can lead to heart failure, plaque rupture, both primary and secondary (in-stent) restenosis in coronary or peripheral arteries, coronary vascular disease, hypertension, diabetes, obesity, stroke, aneurysm, thrombosis, arrythmia, tachycardia, surgical or physical trauma in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment the engineered cell, thereby decreasing severity of disorder in the subject.

82. The engineered ceil according to any one of the previous claims 1 to 75, for use in a treatment for decreasing severity of vascular bed-specific diseases affecting the brain (e.g., stroke), the liver (e.g., sinusoidal obstruction syndrome), the heart (myocardial ischemia) or the extremities (e.g., peripheral vascular disease) in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment the engineered cell, thereby decreasing severity of disorder in the subject.

83. The engineered cell for use in a treatment according to any one of the claims 76 to 82, whereby the cell is genetically altered or unaltered, by direct introduction to a site of interest, e.g., on or around the surface of an acceptable matrix, or systemically, in combination with a pharmaceutically acceptable carrier so as to repair, replace or promote the growth of existing and/or new blood vessels.

84. The engineered cell according to any one ofthe previous claims 76 to 83, whereby said cell is administered in the treatment by any one of the following treatments: localised injection, catheter administration, systemic injection, intraperitoneal injection, parenteral administration, oral administration, intra-cranial injection, intra-arterial injection, intra-venous injection, intra-ventricuiar infusion, intra-placental injection, intra-uterine injection, surgical intra-myocardial injection, transendocardial injection, transvascular injection, intra-coronary injection, intra-muscular injection, surgical injection into a tissue of interest or via direct application to tissue surfaces (e.g., during surgery or on a wound).

85. A pharmaceutical composition, comprising an engineered cell according to any one of the previous claims 1 to 75.

86. An in vitro bioengineered tissue, comprising an engineered cell according to any one of the previous claims 1 to 75.

87. The use of an engineered cell according to any one of the previous claims 1 to 75, in a method of screening for non-toxic compounds.

88. The use of an engineered cell according to any one of the previous claims 1 to 75, in a method of in vitro bioengineering a tissue.

89. The use of an engineered cell according to any one of the previous claims 1 to 75, in in vitro drug toxicity testing, either using the cell product alone, or as a (or one of the) cellular component(s) in the context of engineered 2D or 3D tissue equivalents;

90. The use of an engineered cell according to any one of the previous claims 1 to 75 as a (or one of the) cellular component(s) of tissue engineered constructs for implantation in patients (for example to coat the inside of artificial arterial conduits or as the intimai cellular component of a fully biological tissue-engineered vascular graft or to coat cardiac valves).

91 . An in vitro method of diagnosing a vascular bed-specific disorder phenotype in a subject, said method comprising (a) analysing the level of expression or activity of expression product of at least 5 genes of the '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs) in a sample isolated from said subject, and (b) compare said level of expression or activity with the level of expression or activity in '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs); whereby a deviated level of expression or activity relative to such 'reference signature' is an indication of such disorder phenotype or a propensity thereto. 92. An in vitro method of diagnosing a vascular bed-specific disorder phenotype in a subject, said method comprising: (a) genotyping one or more genes in the '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs) in a sample isolated from said subject, and (b) analyse the DNA sequence of said gene(s); whereby polymorphisms (e.g., SNPs) in said genes of the 'reference signatures' is an indication of such disorder phenotype or a propensity thereto.

93. Use of the '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs), to design tailored therapy, e.g., vascular bed-specific drug delivery or activation of reference signature genes to increase vascularisation.

94. Use of '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous

ECs) to read-out the (side) effects on the ECs of the targeted organ/tissue by current or novel (anti-)angiogenic or other therapies.

95. Use of '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs) to read-out the efficiency of existing or to be developed alternative

(combinatorial) approaches not described herein to specify source cells towards a vascular bed-specific target cell.

Detailed Description

DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION

The following detailed description of the invention refers to the accompanying drawings. The same reference numbers in different drawings identify the same or similar elements. Also, the following detailed description does not limit the invention. Instead, the scope of the invention is defined by the appended claims and equivalents thereof. Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It is intended that the specification and examples be considered as exemplary only.

Each and every claim is incorporated into the specification as an embodiment of the present invention. Thus, the claims are part of the description and are a further description and are in addition to the preferred embodiments of the present invention.

Each of the claims set out a particular embodiment of the invention.

DEFINITIONS

As used herein, the terms below are defined by the following meanings:

The terms "comprises", "comprising", and the like can have the meaning ascribed to them in U.S. Patent Law and can mean "includes", "including" and the like. As used herein, "includes", "including" or the like means including, without limitation.

The term "source cell" is defined as the cell that is converted, by methods described in the invention, to a vascular bed-specific endothelial cell. Any mammalian cell of embryonic or non-embryonic origin with endothelial (differentiation) potential can be used as the source cell. Examples of such cells include, but are not limited to: endothelial progenitor cells (EPCs), mesenchymal stem cells (MSCs), embryonic stem cells (ECS), induced pluripotency stem cells (iPSCs), mesoangioblasts (MABs), multipotent adult progenitor cells (MAPCs), blood outgrowth endothelial cells (BOECs), induced endothelial cells (iECs), unfractionated bone marrow cells, umbilical cord or peripheral blood-derived mononuclear cells, adipose tissue-derived cells, tissue resident progenitor cells, human umbilical vein endothelial cells (HUVECs), human umbilical artery endothelial cells (HUAECs), human dermal microvascular endothelial cells (HDMECs) or cultured endothelial cell lines from heart, liver, brain.

The term "target cell" or the "cell product" is defined as the vascular bed- specific endothelial cell that results from the conversion imposed by the methods described in the invention, wherein the vascular-bed identifies the desired endothelial cell type; for example, the pulmonary vascular bed, refers to vascular endothelial cells of the lungs. The target cell or cell product thus refers to vascular endothelial cells in general, including subtypes based on the vascular bed, such as for example an organ-specific microvascular endothelial cell, an arterial endothelial cell, a cardiac microvascular endothelial cell, a brain microvascular endothelial cell, or a liver microvascular endothelial cell

The term "(differential) reference signature" designates a list of genes that can be used as a reference to define the endothelial subtype identity of the cell product. The latter needs to have certain expression levels of the said list of genes in order to be defined as a certain endothelial subtype, e.g., an arterial endothelial cell, a capillary endothelial cell from the heart, the brain or the liver.

"Embryonic stem cells (ESCs)" are well known in the art and have been prepared from many different mammalian species for many years. ESCs are stem cells derived from the inner cell mass of an early stage embryo known as blastocyst. They are able to differentiate into all derivatives of the three germ layers (ectoderm, mesoderm and endoderm). These include each more than 220 cell types in the adult body. ESCs can become any tissue in the body, excluding placenta. Only the morula's cells are totipotent, able to become all tissues and placenta.

"Induced pluripotency stem cells (iPSCs)" are somatic cells that have been reprogrammed, for example, by introducing exogenous genes that confer on the somatic cell a less differentiated phenotype. The cells can then be induced to differentiate into less differentiated progeny. IPSCs have been derived using modifications of an approach originally discovered in 2006 (Takahashi et al., 2007). For example, in one instance, to produce iPSCs, scientists started with skin cells that were then modified by standard laboratory technique using retroviruses to insert genes into the cellular DNA. In one instance, the inserted genes were Oct4, Sox2, Klf4 and c-myc, known to act together as natural regulators to keep cells in an embryonic stem cell-like state. These cells have been described in the literature (Brambrink et al., 2008; Hanna et al., 2008; Jaenisch and Young, 2008; Wernig et al., 2008). These references are incorporated by reference for teaching iPSCs and methods for producing them. It is also possible that such cells can be created by specific culture conditions (exposure to specific agents).

The term "MAPC" is an acronym for "multipotent adult progenitor cell". The term "adult", with respect to MAPCs is non-restrictive. It refers to a non-embryonic somatic cell. "Multipotent", with respect to MAPCs, refers to the ability to give rise to cell types of more than one embryonic lineage. MAPCs can form cell lineages of all three primitive germ layers (i.e., endoderm, mesoderm and ectoderm). Human MAPCs and methods for their isolation and growth are described in U.S. Patent 7,015,037 and U .S. Patent Application Serial No. 10/467,963 (PCT/US02/04652, published as WO 02/064748). MAPCs have also been derived from other mammals: mice (U.S. Patent 7,015,037 and U.S. Patent Application No. 10/467,963); rats (U .S. Patent Application No. 10/467,963); pigs (U.S. Patent Application PCT/US2005/038979).

"Progenitor cells" are cells produced during differentiation of a stem cell that have some, but not all, of the characteristics of their terminally-differentiated progeny. Defined progenitor cells, such as "endothelial progenitor cells", are committed to a lineage, but not to a specific or terminally-differentiated cell type. The phrase "endothelial cells" encompasses not only terminally-differentiated cell types, but also cells that are committed to a specific endothelial lineage (e.g., venous and/or arterial lineage), but are not terminally-differentiated. The term "progenitor" as used in the acronym "MAPC" does not limit these cells to a particular lineage.

"Stem cell" means a cell that can undergo self-renewal (i.e., progeny with the same differentiation potential) and also produce progeny cells that are more restricted in differentiation potential .

"Effective amount" generally means an amount which provides the desired local or systemic effect. For example, an effective amount is an amount sufficient to effectuate a beneficial or desired clinical result. The effective amounts can be provided all at once in a single administration or in fractional amounts that provide the effective amount in several administrations. The precise determination of what would be considered an effective amount may be based on factors individual to each subject, including their size, age, injury, and/or disease or injury being treated, and amount of time since the injury occurred or the disease began. One skilled in the art will be able to determine the effective amount for a given subject based on these considerations which are routine in the art.

"Subject" means a vertebrate, such as a mammal, such as a human. Mammals include, but are not limited to, humans, dogs, cats, horses, cows and pigs.

"Therapeutically effective amount" refers to the amount determined to produce any therapeutic response in a mammal. For example, effective amounts of therapeutic cells or cell-associated agents may prolong the survivability of the patient, an/or inhibit overt clinical symptoms. Treatments are therapeutically effective within the meaning of the term as used herein, include treatments that improve a subject's quality of life even if they do not improve the disease outcome per se. Such therapeutically effective amounts are ascertained by one of ordinary skill in the art through routine application to subject populations such as in clinical and pre-clinical trials. Thus, to "treat" means to deliver such an amount.

"Treat", "treating" or "treatment" are used broadly in relation to the invention and each such term encompasses, among others, preventing, ameliorating, inhibiting, or curing a deficiency, dysfunction, disease, or other deleterious process, including those that interfere with and/or result from therapy.

METHODS AND COMPOSITION OF THE INVENTION

The invention is based on genetically altering the source cells by overexpression of transcription factor genes. It is accordingly an object of the present invention to provide methods of altering source cells into vascular endothelial cells by overexpression of transcription factor genes as provided throughout this application. As described in detail in the Examples provided herein, this has been accomplished by lentiviral transduction. Nevertheless, genetic modification by introducing DNA or RNA into the source cell can be accomplished by a variety of methods available to those skilled in the art, and are within the admit of the present invention. These methods comprise means for either permanent or transient expression of the gene introduced and are generally grouped into four major categories: (i) viral transfer, including the use of DNA or RNA viral vectors, such as retroviruses (including lentiviruses), Simian virus 40 (SV40), adenovirus, adeno-associated viruses, alpha virus, including Sindbis virus (U.S. Patent No. 5,843,723), herpes virus and bovine papillomavirus; (ii) chemical transfer, including calcium phosphate transfection, DEAE dextran transfection methods; (iii) membrane fusion transfer, using DNA-loaded membranous vesicles such as liposomes, red blood cell ghosts and protoplasts; and (iv) physical transfer techniques, such as microinjection, microprojectile, electroporation, nucleofection or direct "naked" DNA transfer.

The genetic material can be introduced using promoters that will allow for the gene of interest to be positively or negatively induced using certain chemicals/drugs, to be eliminated following administration of a given drug/chemical/temperature, or can be tagged to allow induction by chemicals (including, but not limited to, the tamoxifen responsive mutated estrogen receptor) in specific cell compartments (including, but not limited to, the cell membrane).

Since lentiviral transduction will lead to random integration in the genome, and since random integration may cause untoward effects {e.g., silencing of the transgene, or insertional activation of tumour suppressor genes), the genetic material can be introduced by site-directed homologous recombination, using Zinc finger nucleases or TALE nucleases, into gene loci that are known to have favourable chromatin configurations and are thus transcriptionally active, such as the AASV1 locus on human chromosome 19 or the murine ROSA26 locus.

Successful transduction or transfection of the source cells can be demonstrated using genetic markers, a technique that is known to those of skill in the art. Fluorescent proteins, e.g., green fluorescent protein of Aequorea Victoria, Cherry protein, for example, have been routinely used. Alternative selectable markers include the β-Gal gene, the truncated nerve growth factor receptor, drug selectable markers (including, but not limited to: NEO, MTX, hygromycin, blasticidin).

Not only overexpression of the TFs mentioned above and below can increase their expression level or activity can also be increased in alternative ways. Certain chemical compounds have been described to mimic or act as ligands for several of these TFs or to stabilise the protein {e.g., rosiglitazone for PPARy and PRDM16). Other chemical compounds can act by repressing the effect of specific TFs. In addition, TFs {e.g., PRDM16) often act through complexing with co-factors which modulates the effect of the TFs. Thus, methods in which such chemical components or co-factors are used as a substitute for their 'target' TF represent another embodiment of the current invention.

Assessment of successful conversion of the source cell to the desired cell product can be done by quantitative real-time polymerase chain reaction (qRT- PCR) for genes from the (differential) reference signature or for genes previously described in the literature to be specific for the desired vascular bed-specific EC. In addition, whole genome gene expression of the cell product can be assayed by microarray or equivalent methods {e.g., RNA-seq). Also contemplated are methods to identify successfully converted cells by their expression of proteins encoded by genes from the (differential) reference signature or genes previously published in the literature, by a variety of methods including but not limited to: fluorescence activated cell sorting (FACS), immunofluorescence staining or Western blotting. Assessment of the purity of the cell product can be performed by qRT-PCR for potentially contaminating cell types or, at the protein level, by FACS, immunofluorescence or Western blotting. In addition, successful conversion can also be monitored on life cells using reporter constructs, stably incorporated in the source cells, based on vascular bed-specific EC promoters that then become activated upon successful conversion. These promoters can drive expression of a fluorescence gene, such that successfully converted cells can be identified by FACS.

In case the cell product is not homogenous, i.e., comprises a mixture of successfully converted and not successfully converted cells, then, if desirable, the successfully converted cells can be separated from the not successfully converted cells by FACS or magnetic beads, using antibodies against cell surface markers uniquely expressed on the successfully converted cells. The reporter technology based on fluorescence genes described above to monitor successful conversion of the source cells can also be used to separate them from the not successfully converted cells. Alternatively, the promoters of the reporter constructs can drive the expression of an antibiotic resistance gene, allowing for positive selection of the successfully converted cells by exposure to antibiotics (including, but not limited to G418, hygromycin, blasticidin).

As an alternative to genetic modification by introducing DNA or RNA into the source cell, the corresponding proteins can be transferred directly to cells when they are linked to a protein transduction domain (PTD), small cationic peptide domains that can freely and rapidly cross cell membranes. Several PTDs such as poly-arginine and HIV-derived Tat have been identified that allow a fused protein to efficiently cross membranes.

In addition to the TF genes described herein, variants, members of the same family {e.g., the Prdm family), homologues or orthologues of the factors/genes, which have the same biological function/activity, can be used or assayed for in methods of the invention. For example, variants, homologues or orthologues of use in the present invention may be homologous or have sequence identity (nucleotide or amino acid sequence) with the transcription factors provided herein. "Homology" refers to the percent identity between two polynucleotide or two polypeptide sequences. Examples of assays and programs to determine if a factor/gene is homologous are known in the art. Determination of the percent identity between any two sequences can be accomplished using a mathematical algorithm. Computer implementations of the mathematical algorithms can be utilised for comparison of sequences to determine sequence identity. Such implementations include, but are not limited to: CLUSTAL in the PC/Gene program, the ALIGN program and GAP, BESTFIT, BLAST, FASTA, and TFASTA.

In addition to the use of TF (combinations) described herein, to obtain vascular bed-specific ECs (i.e., the cell product), the invention is also directed to use the cell product for distinct therapeutic and diagnostic purposes. Thus in a further embodiment the present invention provides the target cells obtained using the methods of the present invention, as well as the use thereof, in particular in any pharmaceutical compositions containing the cell product. Such compositions are suitable for administration to subjects in need of such cells. The cells would be administered in therapeutically effective amounts. The choice of the pharmaceutical composition for administering cells for a given application to patients will depend on a variety of factors. Prominent among these will be the species of subject, the nature of the disorder, dysfunction, or disease being treated and its state and distribution in the subject, the nature of other therapies and agents that are being administered, the optimum route of administration, survivability via the route, the dose regimen, and other factors that will be apparent to those skilled in the art. In particular, for instance, the choice of suitable carriers and other additives will depend on the exact route of administration and the nature of the particular dosage form. A pharmaceutically acceptable preservative or stabiliser can be employed to increase the life of cell/medium compositions. If such preservatives are included, it is well within the purview of the skilled artisan to select compositions that will not affect the viability or efficacy of the cells. Those skilled in the art will recognise that the components of the compositions should be chemically inert. This will present no problem to those skilled in chemical and pharmaceutical principles.

The cell products can be administered by a variety of methods available to the art, including but not limited to: localised injection, catheter administration, systemic injection, intraperitoneal injection, parenteral administration, oral administration, intracranial injection, intra-arterial injection, intra-venous injection, intra-ventricular infusion, intra-placental injection, intra-uterine injection, surgical intra-myocardial injection, transendocardial injection, transvascular injection, intra- coronary injection, intra-muscular injection, surgical injection into a tissue of interest or via direct application to tissue surfaces {e.g., during surgery or on a wound). Methods to administer the cells may be combined with methods to increase cell survival, including, but not limited to: incorporation into a biopolymer {e.g., fibronectin, fibrin, fibrinogen, thrombin, collagen, proteoglycans) or synthetic polymer, encapsulation achieved by polymers (such polymeric encapsulation systems include, but are not limited to: alginate, polysaccharide hydrogels, chitosan, calcium or barium alginate, a layered matrix of alginate and polylysine, a photopolymerisable poly(ethylene glycol) (PEG) polymer, a polyanionic material termed Biodritin (U.S. Patent 6,281 ,341 ), polyacrylates, and polymers such as hydroxyethyl methacrylate methyl methacrylate), encapsulation in silicon capsules with pores applied by photolithographic techniques, encapsulation using immune- compatible polycations, including but not limited to, poly-l-lysine polycation or poly- l-ornithine or poly(methylene-co-guanidine) hydrochloride, encapsulation in biocompatible semipermeable membranes ("macroencapsulation"). USE OF THE INVENTION

For the gene signatures or the '(differential) reference signatures', the following uses can be defined:

• since the gene or reference signature changes upon pathological conditions, it can be used as a diagnostic in patients with vascular bed-specific disorders (for instance by genotyping for the signature); furthermore, the gene or reference signature can be used to estimate the risk for acquiring certain cardiovascular disorders by looking for polymorphisms {e.g., SNPs) in one or more genes of the signature;

• the gene or reference signature can be used to design tailored therapy, e.g., vascular bed-specific drug delivery or activation of reference signature genes to increase vascularisation;

• the gene or reference signature can be used as a read-out for the (side) effect on the ECs of the targeted organ/tissue by current or novel (anti-)angiogenic or other therapies;

• the gene or reference signature can be used as a read-out for existing or to be developed alternative (combinatorial) approaches not described herein to specify source cells towards a vascular bed-specific EC phenotype.

For the vascular bed-specific ECs ('the target cells' or desired 'cell products'), the following applications can be defined:

• cell transplantation in patients with vascular bed-specific disorders. For the purposes described herein, either autologous, allogeneic or xenogeneic cells can be administered to a patient, either in terminally differentiated or in partially differentiated form, genetically altered or unaltered, by direct introduction to a site of interest, e.g., on or around the surface of an acceptable matrix, or systemically, in combination with a pharmaceutically acceptable carrier so as to repair, replace or promote the growth of existing and/or new blood vessels;

• in vitro drug toxicity testing, either using the cell product alone, or as a (or one of the) cellular component(s) in the context of engineered 2D or 3D tissue equivalents; as a (or one of the) cellular component(s) of tissue-engineered constructs for implantation in patients (for example to coat the inside of artificial arterial conduits or as the intimal cellular component of a fully biological tissue-engineered vascular graft or to coat cardiac valves).

Examples of (vascular bed-specific) conditions or diseases or circulatory or hypoxic conditions that can be treated with the compositions and the methods of the invention comprise but are not limited to: atherosclerosis, preeclampsia, erectile dysfunction, renal failure, transplant accelerated arteriosclerosis, deep vein thrombosis, sleep apnea, hypoxia during sleep, fetal hypoxia, smoking, anemia, endothelial dysfunction, sinusoidal obstruction syndrome, regional perfusion deficits {e.g., limb, gut, renal ischemia), congestic heart failure, peripheral vascular disease, frost bite, decubitus ulcers, asphyxiation, poisoning (e.g., carbon monoxide, heavy metal), altitude sickness, pulmonary hypertension, sudden infant death syndrome, asthma, chronic obstructive pulmonary disease, congenital circulatory anomalities {e.g., Tetralogy of Fallot), erythroblastosis, myocardial infarction, aortic stenosis, or any other cardiac disease which can lead to heart failure, plaque rupture, both primary and secondary (in-stent) restenosis in coronary or peripheral arteries, coronary vascular disease, hypertension, diabetes, obesity, stroke, aneurysm, thrombosis, arrythmia, tachycardia, surgical or physical trauma.

EXAMPLES

The following examples are provided in order to demonstrate and further illustrate aspects of the present invention and its many advantages and are not to be construed as limiting the scope thereof.

Example 1 : Identification of and induction in dedifferentiated cultured heart endothelial cells or blood outgrowth endothelial cells (BOECs) of a gene (and functional) signature specific for heart microvascular endothelial cells using individual transcription factors or combinations thereof and studies on the in vivo role of these transcription factors

1.1. ABSTRACT Endothelial cells (ECs) lining the inside of blood vessels in different organs show significant heterogeneity caused by cell-intrinsic and -extrinsic factors. They are morphologically and functionally adapted to meet the unique demands of the tissue in which they reside. In this study, by performing a comparative microarray analysis on microvascular ECs freshly isolated from liver, heart and brain, we established 'ex vivo' gene-signatures and partly validated them at the protein level and in samples of human origin. When zooming in on heart microvascular ECs, we found that the lack of tissue microenvironment in cultured ECs wiped out the expression of the majority of heart EC-specific genes. Functional annotation analysis suggested a prominent active function of heart ECs in fatty acid transport, a process the molecular regulation of which is very elusive at present. Interestingly, when evaluating the expression of the heart EC fingerprint in a larger array of tissues, we found that ECs from metabolically highly active tissues (i.e., fat and skeletal muscle) were very similar to heart ECs, unlike ECs from other tissues (i.e., lung, kidney and pancreas). Furthermore, we identified a pool of 5 transcription factors (TFs; mesenchyme homeobox 2 or MEOX2, transcription factor 15 or TCF15, Early B-cell factor 3 EBF3, peroxisome proliferation-activated receptor gamma or PPARy, and Wilms tumour 1 or WT1) which were able to induce the heart EC gene and functional signature when overexpressed alone or in combination in vitro, in cultured heart ECs or in blood outgrowth endothelial cells (BOECs). Accordingly, compound heterozygosity for two of these TFs (Meox2 and Tcf15) in mice resulted in a significant drop in expression of heart EC signature genes - but not liver EC genes - in vivo. Furthermore, fatty acid transporter gene expression was decreased in ECs - but not cardiomyocytes - of Meox2^+/~:Tcf15^+/~ hearts compared to their wild-type (WT) littermates, while glucose transporter Glutl was upregulated. Accordingly, Meox2^+/~:Tcf15^+/~ hearts, but not livers, had a reduced uptake of fatty acids, while glucose uptake was increased. Finally, aged - but not young - Meox2^+/~:Tcf15^+/~ m\ce developed cardiac fibrosis and had impaired heart function. The availability of ECs with a heart-specific molecular gene and functional signature will be useful for in vitro drug toxicity testing and for designing in vivo revascularisation strategies tailored to the heart for patients with ischemic heart disease. In addition, this strategy based on vascular bed-specific TF overexpression to induce a heart EC-specific gene signature in BOECs, can also be tested for other EC types, e.g., liver and brain ECs. Furthermore, our findings highlight a regulatory role for ECs in fatty acid transfer to the heart parenchyma and describe two of its intrinsic regulators. Our insights could be used to develop new strategies based on endothelial targeting of Meox2/Tcf 15 to remedy cardiac dysfunction as a result of changes in the relative contribution of fatty acids and glucose to energy production. 1.2. BACKGROUND AND AIMS

ECs lining the inside of blood vessels in different organs show significant heterogeneity caused by cell-intrinsic and -extrinsic factors and are morphologically and functionally adapted to meet the unique demands of the tissue in which they reside. For instance, the ECs of the liver - a filter and storage organ - are organised in sieve plates and lack a basal lamina to facilitate the passive exchange of big molecules between the blood and the hepatocytes (Aird, 2007a). Brain ECs, on the other hand, rest on a basal lamina, are in close contact with astrocyte end feet and pericytes, have tight junctions between them and are equipped with specific carrier molecules to transport only the necessary substances and to avoid entrance of pathogens or toxic compounds into the central nervous system. Heart capillary ECs influence cardiac function by secreting signalling molecules {e.g., nitric oxide or NO), possess a basal lamina and have a high vesicular transport activity (Brutsaert, 2003). The existence of endothelial diversity has tremendous implications for the development and treatment of diseases, because often the vasculature of the targeted organ is affected or has a direct role in the pathogenesis. From this perspective, adequate and customised EC repair is indispensable for successful restoration of organ function (Aird, 2007a). Moreover, since little is known about specific functions of heart microvascular ECs, we focussed our efforts on the study of these cells in relation to the organ where they reside. As a continuously contracting organ, the heart needs to produce -20 times its own weight in adenosine triphosphate (ATP) each day. Mainly glucose and fatty acids (FAs) serve as substrates for ATP production. Under physiological conditions FAs account for 60-90% of ATP generation, whereas glucose contributes for 10- 40%. A broad range of cardiomyopathies that can lead to cardiac hypertrophy and failure are associated with chronic alterations in cardiac energy metabolism in general and FA utilisation in particular (Neubauer, 2007). Besides the demonstrated role of molecules involved in FA metabolism in cardiomyocytes, recent findings suggest that proteins involved in FA uptake and transport in both ECs and cardiomyocytes may also play a role in these pathologies (van der Vusse, 2000). Nevertheless, it has not been established so far whether cardiac ECs are specialised in order to fulfil the high FA delivery demanded by the heart. Recent studies identified vascular endothelial growth factor (VEGF)-B as a paracrine mediator of FA uptake through ECs in heart (and skeletal muscle) and showed that VEGF-B inhibition protected against type 2 diabetes by decreasing FA uptake and reverting muscle metabolism towards glucose consumption (Hagberg et al., 2010 and 2012). However, the EC-intrinsic (transcriptional) regulators that determine active uptake and transfer of FAs are unknown, except for peroxisome proliferator- activated receptor γ (PPARy) (Kanda et al., 2009). We now identified two novel transcriptional regulators of the gene signature and metabolic substrate transport in heart ECs, i.e., Meox2 (also known as Gax or Mox2) and Tcf15 (also known as paraxis). Meox2 and Tcf15 play a role in early specification of paraxial mesoderm to somitic dermomyotome and a defect in heart development was not reported in Meox2- or Tcf15-deficient mice (Burgess et al., 1996; Mankoo et al., 1999). Nevertheless, expression of both factors was shown in the adult mouse heart, without indication of specific cellular localisation. Meox2 participates in regulating EC homeostasis, being either pro- or anti-angiogenic depending on the expression level and EC type. Tcf15, on the other hand, has never been studied in ECs, except that it was initially cloned from an endothelial library. Here we demonstrate that within the adult heart, Meox2 and Tcf15 are exclusively expressed in ECs and, using gain- and loss-of-function in vitro and in vivo studies, that they are critical regulators of the balance between FA and glucose transport across heart ECs. Our first aim was to learn more about molecular and functional differences between capillary (microvascular) ECs from three clinically relevant vascular beds (i.e., those in brain, heart and liver). Certain studies have offered some insight into specific expression of brain and liver capillary ECs (Calabria and Shusta, 2008; Geraud et al., 2010), however, for heart capillaries, almost nothing is known since most research efforts have focused on endocardial ECs or the coronary macrovasculature (Brutsaert, 2003). This transcriptional profiling effort was then complemented with functional annotation studies and a cross-species comparison in human samples. Since it is known for macrovascular (Aranguren et al., in revision; see Example 2) and some {e.g., brain and liver) microvascular ECs (Calabria and Shusta, 2008; Geraud et al., 2010) that vascular bed-specific genes are significantly downregulated once the cells are deprived of their microenvironment during cell culture, we here investigated the 'ex vivo' vascular bed-specific gene expression profile of freshly isolated cells and compared it to their cultured 'in vitro' equivalents.

A second objective was to restore or induce the 'ex vivo' signature in cultured heart ECs or EC progenitors, respectively. From studies in macrovascular ECs, it is known that TFs play a central role in determining their specific gene signature. Hence, within our 'ex vivo' microvascular EC gene profiles, we expected to find - and indeed found - TFs that co-determine their unique signature. To demonstrate the latter, we overexpressed heart EC-specific TFs - alone or in combination - in cultured heart ECs as well as in unspecified EC progenitors from peripheral or cord blood (blood outgrowth endothelial cells or 'BOECs') which we previously showed to be functionally integrating in newly formed vessels in wounds (Hendrickx et al., 2010).

A final aim was to explore the in vivo role of heart EC-specific TFs. To analyse in more detail the in vivo role of some TFs, corresponding knock-out mice were obtained with the specific intention to (i) explore whether lack of these TFs in vivo had repercussions for the specific EC signature and to (ii) study potential effects of TF deficiency on vascular and heart function. 1.3. MATERIALS AND METHODS

Procedures to achieve Aim 1

Animals and human biopsies. Tie2-GFP mice (Motoike et al., 2000) were used as EC donors for expression profiling. After obtaining informed consent, human ECs were isolated from heart biopsies (right atrial appendage from patients undergoing left-sided valve surgery without pulmonary hypertension or right heart failure), brain biopsies (cortical tissue from epileptic patients undergoing amygdalo- hippocampectomy) or liver biopsies (patients undergoing elective cholecystectomy). Experimental procedures with animals and human-derived samples were approved by the Ethics Committee on Animal Use of KU Leuven and of University Hospitals Leuven, respectively. Human samples were handled according to the Declaration of Helsinki. EC isolation and culture. For murine microvascular EC isolation, tissues from 8- 12 weeks-old mice were dissected out, surrounding connective tissue and visible large vessels removed and tissues enzymatically digested using optimised procedures for each organ (i.e., 1 .2 U/ml dispase, followed by Percoll gradient centrifugation for liver; 0.7 mg/ml crude collagenase + 39 U/ml DNAse I followed by BSA density gradient centrifugation for brain; 0.7 mg/ml crude collagenase for kidney, lung and pancreas; 1 .5 mg/ml collagenase I for heart, skeletal muscle, BAT and WAT). After a final wash in PBS, cells were resuspended in FACS buffer (PBS/EDTA I mM/HEPES 25 mM l\ % BSA, pH7), filtered with a 40 μηη mesh and sorted directly in RLT (Qiagen) or TRIzol® (+ 1 % β-mercapto-ethanol [BME]) for RNA extraction or in Gey's buffer for protein extraction. For profiling, sorting was done based on the GFP⁺ fraction from Tie2-GFP donors and for each organ, samples were a mix of male and female in a 1 :1 ratio. For assaying gene expression in Meox2^+/~:Tcf15^+/~ donors (Aim 3) and their single-heterozygous or Wild-type littermates, sorting was based on the CD31 ⁺CD34⁺CD45^" fraction and either males or females were used. For human heart ECs, epicardial tissue was removed and biopsies were digested with 1 .5 mg/ml collagenase I; for human liver ECs, biopsies were digested with 0.08 Wunsch U/ml liberase and 39 U/ml DNAse I; for human brain ECs, meninges and the most external layer of white matter were removed and biopsies were digested with 0.7 mg/ml crude collagenase and 39 U/ml DNAse I followed by BSA density gradient centrifugation. After a final wash in PBS, cells were resuspended in FACS buffer, filtered with a 40 μηη mesh and sorted directly in RLT or TRIzol® (+ 1 % BME) for RNA extraction. Sorting was done either based on the Tie2⁺podoplanin^"CD45^" or the CD31 ⁺CD34⁺CD45^" fraction. An antibody list for FACS, immunofluorescence and Western blot is provided in Table 1. To obtain cultured human heart ECs, cells obtained from freshly digested biopsies were grown on plates coated with 0.1 % gelatine in EBM2 medium supplemented with EGM2-MV bullet kit (Lonza) in standard conditions (95% O₂/5% CO₂/37° C). When confluent, the Tie2⁺podoplanin^"CD45^" or CD31 ⁺Tie2⁺CD45^" EC fraction was sorted and replated, or RNA from ~10⁵ sorted ECs was collected for qRT-PCR analysis. Before every experiment cell purity was assessed by CD31 FACS staining. Cardiomyocyte isolation and culture. Single ventricular myocytes were enzymatically dissociated from 3- to 4-months-old mice. Mice were injected i.p. with heparin, anaesthetised with pentobarbital, and the heart was quickly excised. After cannulation of the aorta, hearts were mounted on a Langendorff perfusion set. The heart was briefly rinsed with normal Tyrode solution, containing 137 mM NaCI, 5.4 mM KCI, 0.5 mM MgCI₂, 1 mM CaCI₂, 1 1 .8 mM Hepes, 10 mM glucose and 10 mM 2,3-butanedione monoxime (BDM; Sigma Aldrich), pH adjusted to 7.4 with NaOH. Subsequently it was perfused with a Ca²⁺-free Tyrode solution for 10 min. The Ca²⁺-free Tyrode solution contained 130 mM NaCI, 5.4 mM KCI, 1 .2 mM KH₂PO₄, 1 .2 mM MgSO₄, 6 mM Hepes, 20 mM glucose and 10 mM BDM, pH adjusted to 7.2 with NaOH. Collagenase type II 672 U/ml (Worthington), and 30 μΜ CaCI₂ added to the Ca²⁺-free Tyrode solution, were recirculated for 8-10 min. The enzymes were washed out with low Ca²⁺Tyrode solution, i.e., the Ca²⁺-free solution to which 0.18 mM CaCI₂ was added, supplemented with 0.5% Bovine Serum Albumin (BSA; Sigma) for 3 min. and then again with low Ca²⁺Tyrode solution without BSA for 3 min. All solutions used were continuously gassed with 95% O₂/5%CO₂. The heart was then removed from the perfusion apparatus, the ventricles dissociated into single cells by pipetting and afterwards with 5 min. gentle shaking. Cells were sieved through a 250 μηη mesh opening, washed twice with Ca²⁺-free Tyrode solution supplemented with increasing doses of Ca²⁺ (0.5 and 1 mM) and pelleted by centrifugation at 50g for 1 min. after each wash. Finally, cardiomyocytes were collected in TRIzol® or RIPA buffer for ex-vivo qPCR or Western blot analysis, or resuspended in culture medium (M-199 medium, Gibco-Life Technologies), supplemented with 2 mM carnitine, 5 mM taurine, 5 mM creatine (all Sigma Aldrich), and 10 mM BDM, plated onto dishes previously coated with laminin (Sigma Aldrich), and allowed to attach for four hours under standard conditions (95% O₂/5% CO₂, 37° C) before using them for the in vitro fatty acid (FA) uptake experiments.

Microarray analysis and data filtering. Microarrays on pooled EC samples were performed by the VIB Nucleomics Core. Briefly, RNA quality control was done using a Bioanalyser 2100 (Agilent Technologies) and 500 pg of RNA from the EC fraction of 5 selected samples per tissue was amplified, biotin-labelled and run on a mouse genome-wide microarray (Affymetrics Mo Gene1 -0ST). Microarray data were filtered to obtain a validated tissue-specific EC signature: first, only genes that were statistically significantly and differentially overexpressed at least 4-fold (Log₂> 2) versus the other two tissues-ECs and that had a mean Log₂ probe intensity greater than 6 were retained. Next, for the heart, only genes that were significantly enriched in the EC fraction (by comparing versus the non-EC fraction and/or freshly isolated cardiomyocytes) were retained. For brain and liver, the validaton against the non-EC fraction was performed for only 30 genes (representing all TFs and the most differentially expressed genes). Functional annotation analysis was performed with DAVID (http://david.abcc.ncifcrf.gov/) software on the filtered list of the heart and on the full lists of brain and liver.

RNA/protein isolation, cDNA preparation, qRT-PCR and Western blot. RNA isolation, qRT-PCR and Western blot were performed as described extensively in Example 3. Briefly, total RNA from cell lysates was extracted using TRIzol® reagent or RLT lysis buffer. mRNA was reverse transcribed using Superscript III Reverse Transcriptase (Invitrogen) and cDNA underwent 40 rounds of amplification on an ABI PRISM 7700 cycler (Perkin Elmer/Applied Biosystems) for standard SYBER GREEN qRT-PCR. Primer sequences are listed in Table 2. mRNA levels were normalised using GAPDH, ACTB or TUBB as housekeeping gene. Proteins were extracted with RIPA buffer (Sigma) supplemented with protease inhibitors. For Western blotting, 10 to 40 g of proteins was used. Blot pictures were recorded with a Bio-Rad Chemidoc XRS+ molecular imager, equipped with Image Lab software (Bio-Rad laboratories). An antibody list for Western blotting is provided in

Table 1.

Procedures to achieve Aim 2

Lentivirus production and overexpression. The lentiviral construct for overexpressing human MEOX2 (EX-Z3242-Lv1 14) was purchased from Genecopoeia; PPARG1 was cloned from human heart EC cDNA and other TFs were amplified starting from cDNA plasmids (Thermo Scientific Molecular Biology; (maps for the used lentiviruses can be found in Figure 2.A-D and cloning information in Table 3). Lentiviruses were produced in HEK293 cells using two helper plasmids (psPax2 and PMD2G) and Fugene® transfection reagent (Roche Applied Science) and virus productions were titered to use the minimum amount of virus giving 100% transduction. For transduction, cultured human heart ECs and BOECs were plated on 0.1 % gelatin-coated and collagen type l-coated 24-well plates, respectively, and transduced the next day. The medium was changed 12 hours post-transduction. When multiple transductions were performed, the different viruses were added with an interval of 24 hours. Cells were collected 3 days after the last overexpression for RNA extraction, cDNA synthesis and qRT-PCR analysis or used for in vitro FA uptake or Western blot experiments.

BOEC isolation and culture. BOECs were isolated and grown from peripheral and umbilical cord blood (after obtaining informed consent), as previously published (Hendrickx et al., 2010). Briefly, two times 25 ml peripheral (venous) blood was taken using a 20 ml syringe filled with 2.5 ml heparin (5,000 U/ml). Cord blood (50 ml) was collected in dedicated plastic collector bags (Fenwal) containing sodium citrate as anticoagulant. After dividing the blood equally in two 50 ml falcon tubes, the blood was diluted with an equal volume of PBS containing 1 % PSA (penicillin/streptomycin/amfotericinB; Invitrogen). Ten falcon tubes (10 ml) were prepared by filling them with 2.5 ml histopaque (Ficoll Paque™ Plus, GE Healthcare). The layer of 10 ml diluted blood was carefully put onto each histopaque layer taking care not to mix the two layers and the 10 ml falcon tubes were centrifuged for 30 min. at 750 g in a centrifuge with a swing-out rotor of which the brake was switched off. Buffy coats were collected into a new 50 ml falcon tube. 45 ml of PBS/1 % PSA/2% FBS (FBS: fetal bovine serum; Hyclone) was added to the falcon tube and the pellet was washed thoroughly by resuspension. After centrifugation for 5 min. at 750 g (with the brakes on), the supernatant was taken off and the cell pellet was resuspended in 50 ml PBS/1 % PSA/2% FBS. After repeating the last two steps one more time, the supernatant was taken off and the remaining cell pellet was resuspended in EBM-2 (EGM-2 basal medium + EGM-2 singlequots; Lonza). The cells were plated on one or more collagen type-l coated six-well plate(s) (3 ml/well; Greiner Bio-One) and incubated at 37°C and 5% CO2. The media of the cells was changed 100% every day during the first week and thereafter every other day until colonies appeared (usually between day 1 1 -28 days after plating). Once colonies had reached a sufficient size (usually 2-4 weeks after plating), the colonies were harvested together by trypsinisation. Cells were resuspended in EBM-2 growth medium and 3 ml/well was plated in collagen-coated six-wells. Once the wells were confluent, cells were split (1 :3). From then on cells were plated in 10 cm collagen type l-coated plates (Greiner Bio-One).

In vitro FA uptake. FA uptake assays were performed in human heart ECs 72 hours post-transduction or in freshly isolated adult murine cardiomyocytes, after 4 hours of culture. Briefly, cells were medium-deprived and washed with Gey's buffer. For FFA uptake measurement, cells were incubated with 500 nM BODIPY-PA (Molecular Probes) in Gey's buffer for 30 min. (for cardiomyocytes 1 mM 2,3- butanedione monoxime was also added), or 5 μΜ BODIPY-PA in PBS containing 0.1 % FA free-BSA. For VLDL-associated FA uptake, cells were incubated for 2 hours with 10 g/ml Dil-Human VLDL (Kalen Biomedical) in Gey's buffer. Cells were washed twice and maintained in Gey's buffer for 1 hour, then fixed with 4% paraformaldehyde and stained with DAPI. Pictures were taken with a Zeiss Axiovert 200M microscope at 40X magnification. The number of fluorescent vesicles in the cytoplasm was determined by a blinded investigator in 20-40 cells per condition, in 3-5 independent experiments.

Procedures to achieve Aim 3

Animals. Meox2^+/~:Tcf15^+/~ mice were obtained by intercrossing Meox2^Cre/+ (C57BI/6 background; Jackson Laboratories; stockN°003755) and Tcf15^+/~ mice (129S7/SvEvBrd^*C57BI/6 background; provided by E.N. Olson and J.A. Rawls, Dallas, TX and Tempe, AZ, USA; Burgess et al., 1996). Mice were analysed at 3-4 months of age or at 1 1 months of age. Heart ECs or liver ECs (CD31 ⁺CD34⁺CD45^~ cells) were isolated as described above from 7 to 17 weeks-old littermates, either males or females, and gene expression was evaluated by qRT-PCR.

Histology. Oil Red-O, Hematoxylin-Eosin, Sirius red and immunofluorescence stainings were performed as described (Hendrickx et al., 2010). Briefly, for Oil Red- O staining mice were anesthetised with pentobarbital, hearts were rapidly excised, washed in KCI 1 M and PBS, directly snap-frozen in liquid N₂ and conserved at - 80°C until used. Frozen hearts were embedded in Tissue-tek freezing medium (Leica Biosystems), cryo-sectioned at 10 μητι, air-dried for 10 min. and immediately stored at -20°C to be used the same day. Tissue slides were rinced with milliQ water and fixed 1 h with 3.7% formaldehyde in milliQ water. After one rinse in milliQ water, slides were immersed in a freshly prepared, Wattman-filtered Oil Red-O solution (0.3% Oil Red-O in 60% Triethyl-phosphate-H₂O) for 30 min. Then rinsed with milliQ water and deionised water and mounted with Aquadrop (Merck). Pictures were taken with a Zeiss Axiovert 200M microscope at 40X magnification and quantification was carried out with the use of ImageJ software evaluating the percentage of stained tissue surface. The procedure for immunofluorescence/immunohistochemical staining was performed on 3-6 μιτι thick sections of paraffin embedded tissues. An antibody list for IF is provided in Table 1. Images were recorded on a Zeiss Axiovert 200M microscope equipped with a Zeiss MRc5 camera and Axiovision 4.8 software. Oil red-O staining, capillary/cardiomyocyte ratio, cardiomyocyte cross-sectional area and Sirius red staining were quantified using Image J software by a blinded investigator.

Echocardiography. Mice (n=8-9) were sedated with 1 .5% isoflurane and standard views were obtained in 2 dimensions by transthoracic echocardiography. Echocardiography was performed using an MS400 transducer (18-38 mHz probe frequency) on a Vevo 2100 scanner (VisualSonics). Body temperature was monitored and maintained at 37°C. Image analysis was performed by a blinded investigator using the manufacturer's software package. Mice were analysed at 3-4 or 1 1 months of age. In vivo FA uptake. For the assessment of FA uptake in vivo, 7 Wild-type and 8 Meox2^+/~:Tcf15^+/~ mice were injected intravenously with a 2 Ci dose of [1 -¹⁴C]- Oleic Acid (¹⁴C-OA; Perkin Elmer) dissolved in saline (1 :4). For the analysis of ¹⁴C- OA in tissues, and after 30 min. of intravenous injection, mice were anaesthetised with an i.p. injection of ketamine (75 mg/kg) and xylazine (10 mg/kg) mixture and perfused with 0.9% saline solution. Organs (heart and liver) were dissected, weighed and solubilised by the addition of Solvable (Perkin Elmer, 1 ml/100 mg tissue) in a glass scintillation vial and incubated at 60°C overnight. After cooling to room temperature, 30% (w/w) hydrogen peroxide (100 μΙ per 1 ml of sample) was added followed by heating at 60°C for another hour to minimise colour quenching of samples. Finally, scintillation fluid (4 ml per 200 μΙ; Normascint, Scharlab) was added to each sample followed by vigorous shaking. The vials were then allowed to equilibrate in the dark for at least 60 min. before scintillation counting using an LKB Wallac Rackbeta 1214 Counter (Perkin Elmer). Final data are expressed as % injected dose per gram of tissue.

[¹⁸F]FDG PET imaging. Glucose metabolism in the heart was measured by positron emission tomography (PET) with the radiotracer 2-deoxy-2-[¹⁸F]fluoro-D- glucose ([¹⁸F]FDG), synthetised by standard nucleophilic substitution methods at the Clinica Universidad de Navarra PET-GMP laboratory. PET imaging was performed in a dedicated small animal scanner (Philips Mosaic, Cleveland, OH), with 2 mm resolution at full width half maximum (FWHM), 1 1 .9 cm axial field of view (FOV) and 12.8 cm transaxial FOV. Mice (n=12 Wild-type; n=1 1 Meox2⁺ :Tcf15^+/') were anaesthetised with 2% isoflurane in 100% O2 gas and placed horizontally on the PET scanner bed. The [¹⁸F]FDG (7.9 ± 1 .5 MBq in 100 μΙ saline) was injected through the tail vein simultaneously at the beginning of a list mode study of 60 min. For each study, a summed sinogram of the whole emission study and an 18 frame dynamic sinogram (2x 15"; 7x30"; 1 x60"; 1 x120"; 1 x180"; 2x300"; 4x300") were created. From these sinograms, a summed and a dynamic image were generated containing information about the corresponding time intervals. All images were reconstructed using the 3D Ramla algorithm with 2 iterations and a relaxation parameter of 0.024 into a 128x 128 matrix with a 1 mm voxel size applying dead time, decay, random and scattering corrections. For the assessment of [¹⁸F]FDG uptake in the heart, all studies were exported and analysed using the PMOD software (PMOD Technologies Ltd., Adliswil, Switzerland). For each mouse, a volume of interest (VOI) of the heart was drawn on coronal 1 -mm-thick slices of summed PET images. Then, VOI was applied to the dynamic image to obtain the [¹⁸F]FDG uptake values for each time point. The obtained data were normalised by the specific dose at the time of injection.

Statistics. All data are expressed as mean ± s.e.m. from pooled data of 3-12 independent experiments. For single comparisons P-values were calculated with two-tailed unpaired Student's i-test, for multiple comparisons P-values were calculated with one-way ANOVA, followed by contrast analysis. For a time course of a single variable comparison, repeated measures ANOVA was applied, followed by Student's i-test for each time point, P<0.05 was considered significant. Statistical analysis of the microarray data was performed by the VIB Nucleomics Core. Briefly, differential expression was assessed via the moderated f-statistic and to control the false discovery rate, multiple testing correction was performed according to Benjamini and Hochberg. Differences > 4-fold with P<0.05 and Log₂ probe intensities > 6 were considered significant.

1.4. RESULTS

Results for Aim 1: To gain insight into the molecular heterogeneity of microvascular ECs

1 ° Generate and validate tissue-specific microvascular EC 'ex vivo' gene signatures

Step 1: Defining a differential gene sic/nature.

To obtain a specific signature of heart microvascular ECs, we performed a whole genome microarray comparison of GFP⁺ ECs freshly sorted from hearts, livers and brains of adult Tie2-GFP reporter mice (in which the Tie2 promoter- enhancer is active in blood vascular but not lymphatic ECs; Motoike et al., 2000). FACS analysis confirmed that the sorted GFP⁺ fraction constituted a pure population of cells positive for the general EC marker CD31 and for the microvascular EC marker CD36 (see Supplementary Note 1 at the end of Example 1) and negative for the hematopoietic marker CD45 {Figure 3A). Cluster and principal component analysis of microarray data revealed that the 5 samples for each tissue nicely clustered together and away from the two other tissues, suggesting that there was a high degree of intra-organ homogeneity and inter- organ heterogeneity (Figure 3B). Despite the large number of genes commonly expressed by all three microvascular EC types studied, a high degree of heterogeneity was present: we identified a pool of 49, 168 and 1 17 genes (corresponding to 92, 191 and 159 probes) specifically enriched in heart, liver or brain ECs, respectively {Figure 3C and Table 4).

The microarray output was representative for the three microvascular EC types, as evidenced by the high expression of general (i.e., Pecaml and Tek) EC markers, the low expression of genes corresponding to contaminating cells and from the expected enrichment of previously known tissue-specific microvascular EC markers (Figure 3D-E).

Step 2: Filtering against non-EC penes.

Since the design of the microarray experiment did not allow excluding detection of transcripts shared between ECs and the surrounding stromal cells, we performed a qRT-PCR comparison between the sorted EC and non-EC fractions. For liver and brain we performed this comparison for only 30 genes (i.e., the most differentially expressed genes including all TFs. For the liver -77%, and for brain -93% of the tested genes were enriched in the EC fraction (Figure 4A-B, and in Table 5 a list of the TFs belonging to the fingerprints of the three vascular beds studied). For the heart, we conducted this analysis for all 49 genes, and we determined the enrichment in the endothelial fraction in comparison to both the non-EC fraction and a population enriched in cardiomyocytes (Figure 4C). 31 genes (-63%) were retained after this analysis, corresponding to the final list of validated murine heart ECs (Figure 5 and Table 6). Next, we performed a functional annotation analysis on the gene lists. Since only a limited number of genes was not enriched in the EC fraction, for brain and liver, we considered the whole list, since the amount of genes that was not enriched in the EC fraction was more substantial for the heart, and a functional role specific for heart ECs has not been determined so far, we limited the functional analysis for this organ to the validated gene list. The main functions represented by this gene set were 'lipid binding' and 'lipid/FA transport' (Table 7 A). For brain ECs '(selective) transport' and 'metabolism' were highly represented functional terms, whereas for liver 'coagulation' and 'scavenger receptor activity' were listed as highly represented functions {Table 7B,C), in accordance with previously reported data sets.

Step 3: Test the heart EC sic/nature in additional tissue ECs.

To consolidate the heart EC-specific gene signature, we measured its expression by quantitative RT-PCR in ECs freshly isolated from additional tissues, i.e., pancreas, lung and kidney. We found that most genes (-81 % compared to at least 2 of these tissues and -48% compared to all three tissues) were at least 4- fold enriched in heart ECs {Figure 6A). On the other hand, ECs isolated from tissues with continuous endothelium that, like the heart, are highly active in FA uptake for energy production or storage {i.e., skeletal muscle or SkM, brown and white adipose tissue or BAT and WAT, respectively) shared the expression of most genes with heart ECs {Figure 6B). This observation strengthened our hypothesis that the molecular signature of heart ECs reflects their specialisation in active FA transport.

Step 4: Cross-over analysis with human ECs.

A cross-species validation in human ECs confirmed a partially retained enrichment of the signature in human heart ECs when compared with brain and liver ECs {Figure 7). Similar results were obtained for brain and liver ECs (data not shown).

2° Compare the 'ex vivo' with the in vitro heart EC signature To determine the effect of the lack of microenvironment on the gene expression profile, we cultured the bulk cells from the digested human heart biopsies, sorted the blood EC fraction (Tie2⁺:podoplanin^":CD45^"; or Tie2⁺:CD31 ^+" :CD45^"; Figure 8 A) and analysed their gene expression profile by q RT-PCR directly or either after further passages in vitro. As previously shown for other EC types (Geraud et al., 2010), when human heart ECs were cultured in vitro, they lost their specific gene expression profile {Figure 8B). Similar results were obtained for brain and liver ECs (data not shown). Results for Aim 2. To induce the heart EC gene expression and specific function by overexpressing heart EC-specific TFs

1 ° Restoring the 'ex vivo' heart EC gene and functional signature in cultured heart ECs Step 1: Transciptional regulation of the heart EC fingerprint

In addition to genes encoding FA transport-related proteins, the heart EC gene signature also contained five genes encoding a TF, i.e., Meox2, Tcf15, early B-cell factor 3 (Ebf3), Pparg and Wilms tumour 1 (Wt1; Table 5C and Table 6). Meox2, Tcf15 and Ebf3 have never been associated with the heart vasculature, whereas Ppary has been described for its involvement in FA uptake in ECs in the heart (among other tissues; Kanda et al., 2009) and Wt1 for its detection in heart vessels under ischemic conditions (Wagner et al., 2002). Moreover, Wt1 is involved in cardiac vascular development (Wagner et al., 2005). We confirmed at the protein level that Meox2 and Tcf15 were exclusively expressed in the EC compartment of the murine heart and not detectable in liver or brain tissue under identical staining conditions {Figure 9A,B). We next tested the hypothesis that these TFs intrinsically regulate the heart EC signature by assessing in a stepwise manner whether their lentiviral overexpression would restore this signature in cultured human heart ECs. When overexpressing MEOX2 and Tcf15 alone or in combination in cultured heart ECs, we observed a synergistic effect between them, together significantly inducing -31 % of the genes in the signature (Figure 10A). Interestingly, MEOX2/Tcf15 did neither affect each other's expression nor that of WT1, but had a small yet significant inductive effect on PPARG. Expression of EBF3, known to be involved in differentiation into multiple non-endothelial cell types, was strongly induced by MEOX2/Tcf15. Yet, overexpression of EBF3 alone only induced a limited number of signature genes and thus did not recapitulate the combined effect of MEOX2 and Tcf15. Furthermore, combining EBF3 overexpression with MEOX2/Tcf15 only additionally induced aquaporin 7 (AQP7) and FABP9, and only boosted the upregulation of EEPD1 {Figure 10B). When we overexpressed PPARG alone, we measured a strong upregulation of three genes, two of which were known targets - RBP7 and AQP7 - and one novel target (FABP9). When PPARG was combined with MEOX2/Tcf15, for multiple genes, a superior inductive effect was obtained compared to MEOX2/Tcf15, resulting in the significant induction of -48% of the gene signature {Figure 10C). When we finally overexpressed the most abundant isoform (Kamarzova et al., 2012 and Supplementary Note 2 at the end of Example 1) of WT1, we detected a significant upregulation of multiple genes (-38%) of the signature, however combining overexpression of WT1 with MEOX2ITcf151 PPARG did not have an additive effect {Figure WD). Induction of mRNA expression for a random gene subset was confirmed at the protein level {Figure WE).

Step 2: Transcriptional regulation of FA uptake in heart ECs

Since lipid binding and FA transport are major functions represented by the heart EC gene signature, we next wondered whether the uptake of FAs in cultured human heart ECs can be restored by overexpressing TF (combination)s with a significant inductive effect on the heart EC signature. In vivo, FAs reach the heart EC barrier in mainly two forms, either bound to albumin (free-fatty acids or FFA) or contained in circulating lipoproteins {e.g., chylomicrons or very low density lipoproteins [VLDL]) from which they can be released by lipoprotein lipase (Lpl), present at the luminal EC surface. While it was previously assumed that, within the heart, cardiomyocytes are the prime - if not the only - source of Lpl (Davies et al., 2012), we document for the first time Lpl mRNA expression in heart ECs {Table 5). Furthermore, LPL was the gene most upregulated by MEOX2/Tcf15, MEOX2/Tcf15/PPARG or MEOX2/Tcf15/PPA RG/WT1 overexpression {Figure 10A-D), thereby restoring its expression to levels approximating those detected in freshly isolated heart ECs (data not shown). As a test model for FFA uptake we incubated cultured human heart ECs overexpressing the TFs or a reporter gene with fluorescently-labelled palmitic acid (BODIPY-PA) and counted the number of lipid-containing vesicles {Figure 11A). Furthermore, as a model for Lpl-dependent FA uptake we incubated ECs overexpressing the TFs or a reporter gene with VLDL containing Dil-labelled FA (Dil-VLDL; Figure 11B). In both models, we measured that all TF conditions improved FA uptake compared to control and that combined overexpression of MEOX2/Tcf15 outperformed single overexpression of PPARG or WT1 in VLDL-derived FA uptake. Moreover, in the FFA model, the uptake induced by MEOX2/Tcf15 was further enhanced when PPARG was added, while this synergistic effect was not observed for Lpl-dependent FA uptake. Finally, in accordance with the observed effect on the heart EC signature {Figure 10D), the four-TF combination did not significantly increase the effect produced by MEOX2/Tcf15/PPARG. In both tests, the uptake was saturable in a competition assay with non-labelled FA. Thus, we here demonstrate for the first time that FA uptake from VLDL can be heart EC-autonomous, i.e., independent of Lpl produced by cardiomyocytes and that in addition to PPARG (Kanda et al., 2009), also MEOX2/Tcf15 and WT1 regulate FA uptake in heart ECs.

2° Inducing the 'ex vivo' heart EC signature in endothelial progenitors We next attempted to induce a specific heart EC signature by TF overexpression in totally unrelated and unspecialised endothelial progenitors (BOECs). The latter TF-induced differentiation approach is in fact opposite to the TF-induced reprogramming of differentiated cells to stem cells ('induced pluripotent stem cells'), first described by Yamanaka et al. (Takahashi et al., 2007). Like we observed with cultured heart ECs, our TF cocktail (MEOX2 + TCF15 + PPARy + WT1 + rosiglitazone 10 μΜ) induced strong upregulation of the heart EC fingerprint in BOECs (Figure 12).

Results for Aim 3: To explore the in vivo role of heart EC-specific TFs

1 ° Meox2/Tcf 15 determine heart EC identity in vivo

To address whether Meox2/Tcf 15 synergised to affect the adult heart EC expression profile in vivo, transgenic mice lacking one allele of each TF were generated by intercrossing Tcf15^+/~ with Meox2^Cre/+ (further referred to as Meox2^+/~) mice. Mice homozygously deficient for Tcf15 (Burgess et al., 1996) or Meox2 (Supplementary Note 3 at the end of Example 1) die perinatally, while heterozygously deficient mice are viable. ECs were isolated from the heart and, as a control, from the liver of adult Meox2^+/~:Tcf15^+/~ mice and their single- heterozygous or Wild-type littermates. While haplodeficiency for Meox2 or Tcf15 alone only slightly affected the heart EC signature, we observed a significant downregulation of -45% of the signature genes in Meox2^+/~:Tcf15^+/~ hearts, supporting the synergistic genetic interaction of both TFs also in vivo {Figure 13). Complementary to our in vitro findings, Pparg and Ebf3 were significantly downregulated, whereas Wt1 was not affected, confirming the transcription factor hierarchy we unravelled in vitro {Figure 10A-D) General EC markers were not affected in heart ECs, and specific liver sinusoidal EC genes were not altered in liver ECs {Figure 14A,B), suggesting that Meox2/Tcf15 specialise ECs and do not affect EC identity per se in the heart, and that these factors do not affect EC specialisation in unrelated tissues.

2° Meox2/Tcf 15 regulate energy substrate uptake in heart ECs in vivo

To further investigate whether Meox2 and Tcf15 regulate energy substrate uptake by heart ECs in vivo, we tested additional genes encoding glucose or FA transport-related molecules, previously described in heart ECs but not included in our signature, either because they were not enriched in heart ECs, or because they were less than 4-fold enriched in comparison to both liver and brain ECs {Figure 15). CD36, Fabp4, Fabp5, and Gpihbpl, the latter encoding the anchoring protein for Lpl on the luminal EC surface (Davies et al., 2012), were significantly downregulated in Meox2^+/~: Tcf15^+/~ heart ECs (but not in liver sinusoidal ECs), while Fatp3, a FA transporter regulated by VEGF-B interaction with heart ECs (Hagberg et al., 2010) as well as the glucose transporter Glutl were upregulated in Meox2^+/~ :Tcf15^+/~ heart ECs, possibly in a compensatory fashion {Figure 16A,B). Notably, expression of Fatp4, another FA transporter gene induced by VEGF-B in heart ECs, VEGF-B receptors {Flt1 and Nrp1) or Glut4, were not affected by Meox2/Tcf15 (combined) haplodeficiency {Figure 16B and data not shown). Reduced CD36 expression in Meox2^+/':Tcf15^+/'heart ECs was confirmed at the protein level {Figure 16C). Upon Meox2/Tcf15 combined heterozygous deficiency the expression of none of the lipid or glucose transport-related genes tested was affected in cardiomyocytes {Figure 17 A). Accordingly, isolated adult Meox2^+/~ :Tcf15^+/~ cardiomyocytes did not have an impaired FA uptake in vitro {FigureUB). Together, these observations demonstrate an EC-autonomous role within the heart, as also supported by the unique expression of Meox2 and Tcf15 in the endothelial fraction {Figure 9A).

To monitor FA uptake by the heart in vivo we administered ¹⁴C-labelled oleic acid (¹⁴C-OA) via tail vein injection to 12-16 weeks-old Meox2^+/~ :Tcf15^+/~ and Wild- type males, and measured the radioactive signal in the heart after 30 min. We detected a -30% reduction in FA uptake in Meox2^+/~:Tcf15^+/~ compared to Wild-type hearts {Figure 18A ), while FA uptake in the liver was not compromised (injected dose/g tissue: 12 ± 1 in Wild-type versus 10 ± 3 in Meox2^+/~:Tcf15^+/~ livers; n=7-8; P=NS). To investigate whether reduced FA handling is compensated by increased glucose uptake, we measured 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG) uptake in the heart, and detected a -38% increase in Meox2^+/~:Tcf15^+/~ compared to Wild- type mice at -60 min. post-injection {Figure 18B. C).

3° Aged Meox2^+/~:Tcf15^+/~ mice develop cardiac dysfunction Since a shift from FA to glucose uptake due to a defect in FA transport- related proteins can be associated with cardiac dysfunction in animal models upon aging, we performed histology and functional testing on young and aged Meox2^+/~ :Tcf15^+/~ and Wild-type male mice. In accordance with the in v/Vo¹⁴C-OA uptake assay, in young (8-16 weeks-old) mice, Oil red-O staining revealed that lipid accumulation decreased by -61 % in the heart muscle of Meox2^+/~:Tcf15^+/~ compared to Wild-type hearts. However, this did not cause fibrosis or cardiac contractile dysfunction, nor was it accompanied by changes in cardiomyocyte size or vessel number (n° of capillahes/n° of cardiomyocytes: 0.99 ± 0.07 in Wild-type versus 0.97 ± 0.08 in Meox2^+/~:Tcf15^+/~ hearts. n=4; P=NS; Figure 19). However, aged (32-44 weeks-old) Meox2^+/~:Tcf15^+/~ mice, in which we measured a more pronounced (-79%) decrease in heart lipid accumulation, displayed fibrosis and heart systolic dysfunction (as evidenced by a reduced ejection fraction and fractional shortening), in the absence of significant differences in cardiomyocyte size or vessel counts {Figure 20 and data not shown). 1.5. CONCLUSIONS AND PERSPECTIVES

We determined for the first time a unique 'ex vivo' molecular fingerprint for heart capillary ECs (as well as brain and liver microvascular ECs) and demonstrated that this signature can be restored or induced in cultured heart ECs or BOECs, respectively, using a combination of heart EC-specific TFs. For two of these TFs, Meox2 and Tcf15, we showed an equivalent important role in determining the heart EC signature in vivo. These findings will be a platform for studies in which we will exploit the use of the generated heart ECs, including their use for in vitro drug toxicity tests and as in vivo, for treatment in animal models of ischemic heart disease. We hypothesise that BOECs pre-specified to heart ECs by overexpression of our heart-specific TF cocktail will have a more profound and durable effect on revascularisation (and refunctionalisation) upon transplantation in the ischemic heart. We will test this hypothesis first in mice and then later in a clinically more relevant porcine model.

In addition, this strategy based on TF overexpression to induce a heart EC- specific gene signature in BOECs, can also be tested for other EC types, e.g., liver and brain ECs. Indeed, in the gene list of murine liver ECs we found 8 genes functionally annotated as having TF activity (i.e., Tcfec, Hoxb5, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2; Table 5A) and in the murine brain ECs, we indentified 5 TFs {i.e., Zic3, Lef1, Foxf2, Foxfla and Foxd; Table 5B) Resulting pre-specified ECs can than be tested for their therapeutic efficacy in models of liver {e.g., sinusoidal obstruction syndrome) or brain {e.g., stroke) vascular disease.

Moreover, here we unravel for the first time that heart, skeletal muscle, BAT and WAT, the major tissues involved in whole body metabolism, share specialised endothelial cells, active in the transport of fatty acids to the parenchyma, and we provide a fingerprint of genes specifically expressed by these ECs. We show that Meox2 and Tcf15 EC-autonomously determine the expression of these fingerprint genes both in vitro and in vivo, and that they can modulate energy substrate delivery to contracting cardiomyocytes. Meox2/Tcf15 acted upstream of another endothelial regulator of FA uptake, i.e., PPARy (Kanda et al., 2009), as suggested by the increased PPARG expression upon MEOX2/Tcf15 overexpression and by the significant downregulation of Pparg and some of its known target genes {CD36, Fabp4 and Aqp7) in Meox2^+/~:Tcf15^+/~ heart ECs. On the other hand, MEOX2/Tcf15 and PPARG also had a synergistic effect in vitro on the regulation of most genes of the signature, which were not induced by the overexpression of PPARG alone. These findings thus establish a previously unrecognised TF network in charge of endothelial FA uptake regulation

Understanding the intrinsic regulation of FA uptake by heart ECs offers new therapeutic perspectives for the use of ECs as a target to modify the metabolic status of the heart, which has been extensively described to be affected in a large list of pathologies leading to heart failure. For example, pathological situations where there is an increase in circulating FAs, like obesity or type 2 diabetes, lead to an excess of fat uptake in tissues, resulting in lipotoxicity and the so-called "diabetic heart". The latter results from metabolic reprogramming of the cardiomyocytes which occurs in parallel with the overload of FA, leading to the development of insulin resistance, oxidative stress, mitochondrial damage and ultimately heart failure (Abel et al., 2012). It is also known that a variety of pathological insults in the heart, which initially are not related to cardiac metabolism, including hypertension, myocardial infarction, and aortic stenosis, if protracted or severe, can lead to the reprogramming of the cardiac energy production machinery and lead to activation of lipid anabolic pathways, inefficient ATP production, and therefore impaired contractility, cytotoxicity and heart failure (Neubauer, 2007). Further studies will be necessary to assess whether our insights could be used to develop new strategies based on endothelial targeting of Meox2/Tcf 15 to remedy cardiac dysfunction as a result of changes in the relative contribution of FAs and glucose to energy production. Such therapeutic strategies could be complementary to other approaches aimed at shifting the energy substrate source of the heart (Jaswal et al., 201 1 ; Stanley et al., 2012).

Another EC-targeted approach, i.e., inhibition of VEGF-B as a paracrine factor affecting the ECs' FA transporter expression, was shown to decrease FA uptake and increase glucose uptake in the heart (Hagberg et al., 2010). Meox2/Tcf15 in vivo regulated a set of membrane FA and glucose transporters (Cd36 and Glutl) that was not overlapping with those regulated by VEGF-B. Furthermore, Meox2^+/~ :Tcf15^+/~ heart ECs had lower expression the glycerol transporter Aqp7 and of intracellular FA transporters (Fabp4 and FabpS). Finally, we revealed for the first time that besides cardiomyocytes, cardiac ECs - in large part driven by Meox2 and Tcf15 - produced their own Lpl and that combined deficiency of the two transcription factors reduced expression of both Lpl and its prime endothelial transporter, Gpihbpl (Davies et al., 2012) in vivo. Collectively, this suggests that Meox2 and Tcf15 co-regulate several molecules involved at different levels of endothelial FA uptake, complementary to VEGF-B. Hence, combined targeting of these intrinsic and paracrine pathways could synergistically reinforce the endothelial barrier of the heart against pathological fat uptake.

Of note, within the heart EC fingerprint, there were also genes encoding secreted molecules that could be involved in local EC-to-cardiomyocyte communication and influence the cardiomyocyte's FA usage. For instance, C1qtnf9 - the expression of which was decreased in Meox2^+/~ :Tcf15^+/~ heart ECs and increased in vitro upon MEOX2-Tcf15 combined overexpression - is a paralogue of the adipokine Adipoq. Although adipose tissue is the prime organ for C1 qtnf9 secretion, significant local production has also been detected in the heart. In mice, type 2 diabetes and obesity, as well as cardiac ischaemic damage, induced a significant decrease in cardiac C1 qtnf9, and its exogenous supplementation reduced cardiac damage after ischaemia-reperfusion (Kambara et al., 2012). Mechanistic studies revealed that C1 qtnf9 exerts its effect in cardiomyocytes through activation of the AMPK pathway, which is known to increase FA oxidation in myocytes. Finally, the heart EC fingerprint also harboured genes encoding proteins not linked to FA handling but relevant for cardiac remodelling, such as metalloproteinase Adamts9 and metalloproteinase inhibitor Timp4. Given their known involvement in collagen homeostasis, their lowered expression in Meox2^+/~ :Tcf15^+/~ heart ECs may have contributed to the cardiac fibrosis we observed, which is a phenotype common to other animal models of impaired cardiac FA uptake (Augustus et al., 2006).

In summary, here we provide for the first time a fingerprint of genes enriched in ECs lining the capillaries of tissues with high metabolic capacity for FA, suggesting a common specialisation of these vascular beds. We identified Meox2 and Tcf15 as transcriptional determinants of the heart EC fingerprint and demonstrated that they are critical regulators of the balance between FA and glucose transport across heart ECs (Figure 21). These findings have broadened our understanding of the genetic identity and physiological role of heart microvascular endothelium and provide a platform for further mechanistic studies for therapeutic exploitation of specific EC targeting aimed at modulation of the energy substrate uptake in the heart. SUPPLEMENTARY NOTES TO EXAMPLE 1

Supplementary note 1. Microvascular EC purity.

As previously described, CD36 is highly expressed in microvascular ECs and absent in ECs from large vessels (Chi et al., 2003). Nevertheless, between capillary ECs from different organs there is a different degree of expression of CD36, being very low in brain ECs and high in liver and heart ECs (as evident from our microarray analysis; Figure 15). Indeed, for heart and liver, microvascular EC purity could be determined by FACS for CD36 (as shown for the heart in Figure 3A), whereas under identical FACS staining conditions we could not detect CD36 positive cells in brain ECs. Therefore, to confirm microvascular EC purity of brain EC preparations, we relied upon the significant enrichment for known brain-specific microvascular EC markers such as Glutl, Lat1, Ocln and Tfrc (Figure 3E).

Supplementary note 2. Relative abundance of WT1 isoforms in cardiac ECs.

Several WT1 isoforms have been described with different tissue specificities and functions (Kamarzova et al., 2012). We quantified the relative abundance of the long versus short isoforms (the latter generated from an alternative start codon in intron 1 ) and of the isoforms carrying or not the KTS domain in freshly isolated human and mouse heart ECs and found that the most abundant WT1 isoform in heart ECs is the isoform expressing exonl and the KTS domain (ratio isoform - exonl /isoform + exonl = 1 :1 ,500, and 71 ±8% carrying the KTS element). We therefore overexpressed the isoform D (expressing exonl , exon5 and carrying KTS), the most prevalent isoform of WT1 in normal tissues, to test its effect on the heart EC gene signature.

Supplementary note 3. Mendelian ratios of Meox2^+/~ x Meox2^+/~ intercrosses.

Meox2^Cre/+ mice were obtained from the Jackson Laboratories

(stockN "003755) and crossed for one additional generation in C57BI/6 to establish the colony. When setting up Meox2^Cre/+ x Meox2^Cre/+ crosses to obtain Meox2^cre/cre (or Meox2^~/~) mice, we observed that Meox2^~/~ pups died within a few hours after birth, while the ratios just before birth (at embryonic day 19.5) were Mendelian. This observation is at variance with what is reported on the Jackson Laboratory website where it is mentioned that Meox2 mice are viable at birth and die just before weaning (http://jaxmice.jax.org/strain/003755.html).

Example 2: Identification of and induction in dedifferentiated cultured human umbilical vein endothelial cells of a gene signature specific for human arterial endothelial cells using individual transcription factors or combinations thereof 2.1. ABSTRACT

Endothelial cells (ECs) lining arteries and veins have distinct molecular and functional features. The underlying regulatory mechanisms in human ECs are incompletely understood. Here, we established a specific fingerprint of freshly isolated arterial (HUAECs) and venous EC (HUVECs) from human umbilical cord comprising 64 arterial and 12 venous genes, representing distinct functions and pathways. Among the arterial genes were 8 transcription factors (TFs), including Notch target HEY2, the current 'golden standard' denominator for arterial (A)EC specification. Culture of HUAECs or HUVECs abrogated differential gene expression at least in part due to loss of canonical Notch activity and HEY2 expression. Notably, restoring HEY2 expression or Delta-like4-induced Notch signalling in cultured ECs only partially reinstated the aEC gene signature while combined overexpression of the 8 TFs restored this fingerprint more robustly. While some TFs only stimulated few genes, others boosted a large proportion of arterial genes. Interestingly, although there was some overlap and cross-regulation, the TFs largely complemented each other in regulating the aEC gene profile. Additionally, overexpression of the 8 arterial TFs improved the arteriogenic capacity of HUVECs in a Matrigel plug assay in vivo. Thus, our study showed that Notch signalling determines only part of the AEC signature and identified additional novel and complementary transcriptional players in the complex regulation of human arteriovenous EC identity.

2.2. BACKGROUND AND AIMS Endothelial cells (ECs) of different vessels across the body differ in morphology, function and gene expression profile, a phenomenon known as 'endothelial heterogeneity' (Aird, 2007a, b). Endothelial diversity is due to exposure to different microenvironments (extracellular matrix, surrounding cells, blood flow) and different intrinsic genetic programs which are present at very early stages of development even before the vascular system is functional (Aird, 2007a). Genetic programs determining arterial or venous EC identity have been mainly studied in zebrafish and mice and much less attention has been given to human cells, mainly because of the difficulty to obtain ECs of human origin. It is well documented that canonical Notch signalling including the ligand delta-like 4 (DII4), the receptors Notch 1 and Notch4 and the downstream TFs Hey1 and Hey2 (gridlock in zebrafish) determine the arterial phenotype across species (Swift and Weinstein, 2009), in part by blocking the TF chicken ovalbumin upstream promoter-TF (COUP-TF)II which is expressed in venous ECs (VECs) (You et al ., 2005). The Notch-Hey pathway induces ephrinB2 expression and blocks the expression of the corresponding receptor EphB4 in arterial ECs (AECs) whereas in vECs COUP-TFII mediates the opposite effect (Swift and Weinstein, 2009; You et al., 2005). This ephrinB2-EphB4 differential expression establishes a polarity that assists in segregating arteries from veins down to the capillary level (Wang et al., 1998). Many other molecules and pathways like Shh, VEGF, FoxC1 /C2, adrenomedullin or cAMP, spreds, Wnt, Sry-related HMG box (Sox), TGF-β and phosphatidylinositol-3 kinase play a role in arteriovenous specification (Corada et al., 2010; Swift and Weinstein, 2009). Nevertheless, most studies have been focused on one or only a few regulatory elements in this specification process. An exception to this is the study by Chi et al. (Chi et al., 2003) who elegantly showed endothelial heterogeneity between arterial and venous ECs as well as microvascular ECs on a genome-wide scale. Importantly, for this analysis they used human ECs in culture. However, studies in microvascular ECs (Calabria and Shusta, 2008; Geraud et al., 2010) and lymphatic ECs (Amatschek et al ., 2007; Wick et al., 2007) have shown that the culture process, even short-term, can erase a significant part of the specific gene expression profile. This loss of identity upon culture has not been systematically studied for arterial and venous ECs. Therefore, our aim was to perform a stringent genome-wide profiling study on ECs isolated from arteries and veins from different human vascular beds, including cultured and freshly isolated cells, and to analyse in more detail intrinsic regulators of arteriovenous endothelial heterogeneity.

2.3. MA TERIALS AND METHODS

EC isolation and culture. Commercial EC lines used: HAECs (Lonza, Barcelona, Spain; CatN°CC-2535), HCAECs (Lonza, CatN°CC-2585), HIAECs (ATCC, Barcelona, Spain; CatN° CRL-2475), HPAECs (ATCC, CatN° CRL-2598), HIVECs (ATCC, CatN° CRL-2606), and HPVECs (ATCC, CatN° CRL-2607). EC lines were cultured according to the provider's instructions. HHAECs, HHVECs, HUVECs and HUAECs were isolated at the Clinica Universidad de Navarra (after obtaining informed consent) by perfusing the corresponding vessel with collagenase type I (Invitrogen, Carlsbad, CA). Harvested cells were cultured for 24 hours, washed to discard non-attached cells, grown until 100% confluence and split 1 :3 every 3-4 days. For the analysis of fresh cells or short time culture for time course analysis, ECs from umbilical arteries or veins were magnetically selected using anti-human CD34 magnetic beads (Miltenyi-Biotec, Madrid, Spain) and an AutoMACS magnetic selector (Miltenyi-Biotec) according to the manufacturer's instructions. Purity of magnetically sorted HUVEC-F or HUAEC-F was assayed by FACS. Only samples with more than 95% purity for CD31 ⁺CD34⁺CD45^" cells were eligible for microarray hybridisation or short time culture (Figure 22). All TFs and a random subset of the genes from the fingerprint were validated using FACS sorting for the Tie2⁺CD31 ⁺CD45^" population (data not shown).

RNA isolation, quality control and qRT-PCR. Total RNA from cell lysates was extracted using TRIzol® reagent or RLT lysis buffer (Qiagen). The RNA integrity/quality of the samples used for microarray hybridisation were determined with a Bioanalyser 2100 (Agilent Technologies, Santa Clara, CA). When necessary, mRNA was reverse transcribed using Superscript III Reverse Transcriptase (Invitrogen, Carlsbad, CA) and cDNA underwent 40 rounds of amplification on an ABI PRISM 7700 cycler (Perkin Elmer/Applied Biosystems, Foster City, CA) for standard quantitative Real-Time PCR as described in Example 3. Primer sequences are listed in Table 8. mRNA levels were normalised using GAPDH as housekeeping gene. Data, expressed as mean ± s.e.m. comparing two groups were analysed by Student's i-test. SPSS software was used for statistical analyses and differences were considered significant when P<0.05.

Microarray hybridisation and statistical analysis. RNA hybridisation of the 38 human EC samples was done in collaboration with the Department of Hematology, Hospital Universitario de Salamanca, using the Affymetrix HG-U133 Plus 2.0 GeneChip Oligonucleotide Microarray (Affymetrix, Santa Clara, CA, USA). All steps were carried out according to the manufacturer's protocol. 100 ng of RNA was amplified and 15 g of amplified and labeled cRNA was hybridised on the array. Arrays were scanned using a GeneChip Scanner 7G. Background correction and normalisation were done using the RMA (Robust Multichip Average) algorithm. The method for differential gene expression analysis was the one contained in the LIMMA Bioconductor package. To control the false discovery rate, multiple testing correction was performed and probes with a corrected P-value below 0.05 were selected. For the classification analysis, a filtering process was applied first to eliminate probe sets with low expression values. Applying the criterion of an expression value greater than 32 in 5 samples for each experimental condition, 32,939 probe sets were selected for statistical analysis using the LIMMA Bioconductor package. Prediction analysis for microarrays was applied to classify arterial and venous samples and identify genes that were associated with each specific class. This algorithm ranked genes using a penalised i-test and identified a gene set for classification with soft thresholding. Gene number was controlled by a thresholding parameter, which was determined with a 10-fold cross-validation. This parameter was manually selected to minimise the overall error rate (t=7). The obtained classifier required 78 probes. Functional and pathway enrichment analysis was done using Ingenuity Pathway Analysis software (Ingenuity Systems, Redwood City, CA, http://www.ingenuity.com). Time course analysis with TLDA. Taqman® Low Density Array (TLDA) plates with Taqman® primers for our described HUAECfresh/HUVECfresh fingerprint, previously described arteriovenous markers and some general endothelial markers were obtained from Applied Biosystems. HUAECfresh/HUVECfresh samples and samples from HUAECs and HUVECs cultured for 24 hours, 48 hours or 6 days were run on a 7900 HT fast real time PCR system (Applied Biosystems) and analysed according to the manufacturer's instructions. GAPDH was used as housekeeping gene. Each condition was run in quadruplicate. nCounter analysis. DLL4-Fc activated or BSA-treated HUAECs (A/=3) and HUVECs transduced with cherry, each of the 8 TFs or a combination of them (A/=4- 6) were used for nCounter analysis. Briefly, RNA, extracted using TRIzol® reagent was quantified and quality controlled with a Bioanalyser 2100 (Agilent Technologies) and samples were processed in collaboration with the VIB Nucleomics Core Facility. 100-500 ng of total RNA was hybridised according to the manufacturer's instructions. Some of the results were confirmed by qRT-PCR and genes for which the probe intensity value was low were analysed by qRT-PCR. Data were normalised by scaling to the GAPDH gene. The scaled counts were base log 2-transformed. Testing whether a contrast was significantly different from 0 was done by using a moderated i-test, as implemented in LIMMA. The resulting P-values were corrected for multiple testing with Benjamini-Hochberg to control the false discovery rate. P<0.05 and at least 50% change differences in comparison with control (BSA-treated HUAECs or cherry-transduced HUVECs) was considered as significant. The Nanostring probe list is provided in Table 9. The global expression profiles of the samples were represented in the form of a heat map, using the gplots package of R. Hierarchical clustering was used to cluster the individual gene expression profiles based on Pearson correlation and complete linkage. Immunofluorescence staining and Western blot. The procedure for IF staining was done on cord blood samples as described previously (Aranguren et al., 2007). Antibodies used were: Rabbit anti-human Rasgrf2 (Sigma, CatN⁰ HPA018679), Alexa488-labelled mouse anti-human smooth muscle a-actin (Sigma, CatN° F3777), Rabbit anti-human Nr3c2 (Santa-Cruz, CatN° SC-1 1412) and goat anti- human Msx1 (R&D Systems, CatN° AF5045). Western blot was performed as described (Hendrickx et al., 2010). Samples were collected in RIPA buffer (Sigma, Bornem, Belgium) and protein concentration was measured by the BCA assay. 40 μg of protein was used for blotting. Antibodies used were: Rabbit anti-human Rasgrf2 (Sigma, CatN⁰ HPA018679), rabbit anti-human A2M (Sigma, CatN° HPA002265), Rabbit anti-human MAP9 (Sigma, CatN" HPA037864) and mouse anti-human a-tubulin (Sigma, CatN°T6199; used as housekeeping reference). siRNA treatment and Notch activity assays. siRNA knockdown was performed using Silencer® Select pre-designed siRNA from Applied Biosystems for RBPJ (siRNA ID#: s7251 and s7253) or Negative Control 1 (siRNA ID#: am4636). Briefly, 2,500 HUAECs/cm² were cultured overnight. The next day, cells were transfected with 5 pmol siRNA mixed with 0.5 μΙ of lipofectamine 2000 (Invitrogen) in 100 μΙ of OPTI-MEM (Invitrogen). The day after transfection, media was replaced and cells were maintained for 6 days, with an additional siRNA transfection at day 3. The canonical Notch pathway was induced by immobilised DLL4 ligand activation. Briefly, DLL4-Fc (R&D Systems, CatN° 1389-D4) was incubated overnight at 4°C at 1 μg ml in 0.1 % gelatin 1 % BSA in PBS with gently shaking to allow its adsorption to the cell culture dish. The next day, DLL4-Fc coated plates were incubated at 37°C for 1 hour. Non-attached DLL4-Fc was removed by washing and 2,500 HUAECs/cm² were seeded and cultured for 72 hours. The canonical Notch pathway was blocked by γ-secretase inhibitor DAPT (Calbiochem, San Diego, CA, USA; CatN° 565784; alone or in combination with immobilised DLL4-Fc) at 3 μΜ concentration. The corresponding DMSO volume was used as a control.

Lentivirus production and overexpression. The lentiviral construct for overexpression of HEY2 was obtained from Genecopoeia (Rockville, USA; Table 10). Open reading frames (ORF) for MSX1 , EMX2, Prdm16, NKX2-3, Aff3 and TOX2 were cloned from cDNA (Open Biosystems) or total cDNA (BD) after the cytomegalovirus (CMV) promoter in pRRL2-CMV-PGK-Cherry {Figure 23 and Table 10). The lentiviral construct for overexpression of SOX17 was kindly provided by C. Verfaillie (Stem Cell Institute, KU Leuven; Figure 23). For lentiviral particle production HEK293 cells were plated (5x10⁶ cells/10 cm dish) and the next day transfected with the plasmid of interest together with two helper plasmids (psPax2 and PMD2G) using Fugene transfection reagent (Roche). In brief, 400 μΙ of OPTIMEM (Gibco) was mixed with 1 μg PMD2G, 3 μg psPax2 and 4 μg lentiviral construction plasmid {Table 10). 24 Ι of Fugene was added and the mixture was incubated for 20 min. at RT and gently applied to the cells. The next day, medium was replaced and lentiviral particle-containing supernatant was collected 36 hours later. Viruses were concentrated by centrifugation using 50.000 MWCO Vivaspin(R) 20ml centrifugal concentrators (Sartorius Stedim). Transduced cells were kept for 6 days and collected into TRIzol® buffer. For long-term overexpression of HEY2 and Prdm16, transduced cells were cultured up to 28 days.

In vivo Matrigel implantation assay. 0.5x10⁶ HUVECs transduced with control Cherry lentivirus or with the combination of the 8 TFs were mixed with pre-cooled 0.5 ml of Growth Factor Reduce Matrigel containing 300 ng/ml VEGFi₆5 (R&D Systems) and 700 ng/ml bFGF (R&D Systems) and subcutaneously injected in the back of 8-weeks-old athymic nu/nu mice (N=5 per group). 14 days later, mice were sacrificed by cervical dislocation and the Matrigel plug was dissected out. The Matrigel plug was divided in two equal pieces, one piece was processed for cryosectioning and the other for paraffin sectioning. Human cells were detected by the Cherry fluorescent signal or by human-specific CD31 staining (Dako). Smooth muscle coverage was analysed on smooth muscle a-actin (Sigma) stained sections and collagen deposition was quantified on Sirius red stained sections. 2.4. RESULTS

Freshly isolated but not cultured AECs and VECs differ in gene expression profile

To identify an arteriovenous fingerprint in human ECs across different vascular beds, we used microarrays on RNA from 38 human (H) EC samples {Figure 24) corresponding to 6 cultured AEC types (hepatic artery ECs or HHAECs, A/=3; aorta ECs or HAECs, N=2; coronary artery ECs or HCAECs, N=2; iliac artery ECs or HIAECs, N=2; pulmonary artery ECs or HPAECs, A/=3; and umbilical artery ECs or HUAEC-C, N=5), 4 cultured VEC types (hepatic vein ECs or HHVECs, A/=3; iliac vein ECs or HIVECs, A/=3; pulmonary vein ECs or HPVECs, N=2; and umbilical vein ECs or HUVEC-C, N=5), freshly isolated HUAECs (HUAEC-F, N=4) and freshly isolated HUVECs (HUVEC-F, N=4). Due to the difficulty to obtain biopsies from healthy donors, we did not have access to freshly isolated AECs or VECs matched for all cultured EC types. Despite the homogeneity within EC subtypes, when considering all 30 cultured EC samples together, we did not find statistically significant differences in gene expression of AECs versus vECs.

We next performed profiling analysis on the freshly isolated HUVEC/HUAEC subset using two statistical approximations: differential gene expression or classification analysis {Figure 25 A). Both analyses detected significant differences between HUAEC-F and HUVEC-F yielding a combined panel of 76 genes (-102 probes; Table 12) with 64 HUAEC-F genes and 12 HUVEC-F genes, -55% of which were commonly picked up by both statistical methods {Figure 25 A). In the remainder of the text, we refer to this 76 genes set as the 'arteriovenous fresh profile'. This profile included 16 genes with documented (differential) expression in AECsA/ECs and 13 genes associated with a vascular phenotype in mice, 4 of which were described to be involved in AEC specification/development {NOTCH4, HEY2, KDR and SOX17), revealing that our extracted gene list was reliable {Figure 25 A, Table 11). Functional and pathway enrichment analysis produced several relevant terms, including (cardio)vascular disease, cardiovascular system development and Notch signalling {Figure 25B-C). Thorough study of the literature revealed that 10 genes (or a related family member or orthologue) had a previously documented interaction with the Notch pathway {Table 11). Finally, the arteriovenous fresh profile harboured 9 genes encoding a TF (8 arterial and 1 venous; Figure 25D). We validated some of the new AEC {i.e., RASGRF2; Figure 25E-F) and VEC {i.e., NR3C2; Figure 25G-H) markers at the protein level. All TFs and a random subset of the genes from the fingerprint were validated using FACS sorting for Tie2⁺CD31 ⁺CD45^" population (data not shown).

The culturing process rapidly erases differential arteriovenous gene expression

Hierarchical clustering analysis for the arteriovenous fresh profile showed a perfect separation of the 4 HUAEC-F and HUVEC-F samples, confirming a high degree of difference between the two EC subtypes {Figure 26 A). The influence of the culture process on EC gene expression has been previously described for brain microvascular ECs, lymphatic ECs and venular ECs but not for AECs or large VECs. Hence, we next compared freshly isolated HUVECs/HUAECs with their cultured counterparts. Global differential expression analysis showed dramatic changes in gene expression upon culturing, with 410 probe sets (-327 annotated genes) upregulated and 442 probe sets (-332 annotated genes) downregulated in HUAEC-C versus HUAEC-F and 66 probe sets (-56 annotated genes) upregulated and 105 probe sets (-69 annotated genes) downregulated in HUVEC-C versus HUVEC-F. Strikingly, when considering the arteriovenous fresh profile genes alone, there were no obvious differences between HUVEC-C and HUAEC-C {Figure 26B, Table 12). Expression of the 8 arterial TFs was completely lost upon culturing HUAECs (Figure 26C). This culture-induced assimilation was also underscored by incorrect clusters generated to distinguish HUVEC-C and HUAEC-C, unlike their fresh equivalents, which clustered according to their arterial or venous origin {Figure 26 A). These incorrect clusters were also apparent when considering all cultured AEC and VEC types from our screen, suggesting that AECs and VECs from these different vascular beds had lost their specific expression profiles upon culturing {Figure 24).

Since during the observed assimilation process arterial and venous-specific genes of the arteriovenous fresh profile as well as the two cell types may behave differently, we next analysed the arterial and venous part of the signature separately in HUAECs and HUVECs. When HUAECs were cultured for several passages, there was strong silencing of most of the 64 arterial genes of the signature, while there was no obvious change in the venous arm {Figure 26D-F, Table 12). Conversely, when culturing HUVECs, venous genes were significantly - albeit less dramatically than arterial genes in HUAECs - downregulated, while arterial genes did not change {Figure 26E-F, Table 12). Thus, HUAECs were more susceptible than HUVECs to assimilation. To determine the kinetics of the downregulation, we cultured freshly isolated cells for different periods of time before determining the arteriovenous fresh profile. Twenty-four hours of culture was sufficient to induce changes for a majority of the genes (62% of arterial genes in HUAECs and 73% of venous genes in HUVECs; Figure 27A-B). The changes became more obvious after 48 hours and 6 days (Figure 27C-F), revealing that the majority of gene expression changes occurred very rapidly independent of passaging.

Reactivation of Notch signaiiing only partially restores arterial gene expression in HUAEC-C

It is well established that canonical Notch signalling induces arterial specification during vascular development, by boosting the expression of AEC markers like ephrinB2, and blunting the expression of VEC markers like EphB4. Due to the silencing of the Notch target HEY2 in HUAEC-C compared to HUAEC-F {Figure 26, Table 12), we hypothesised that the canonical Notch pathway might be inactive in vitro, which could explain the loss of the arterial phenotype upon culture. During canonical Notch signalling, ligand binding to the Notch receptor induces γ- secretase-mediated cleavage of the receptor, thereby releasing the Notch intracellular domain (NICD). The latter subsequently migrates to the nucleus and binds to recombination signal binding protein for immunoglobulin kappa J region (RBPJ). This association releases RBPJ-associated co-inhibitors and makes RBPJ accessible to co-activators that induce expression of Notch-responsive genes like HEY1/2 {Figure 28 A). If the canonical Notch pathway were active in HUAEC-C, then blocking it by using DAPT (a γ-secretase inhibitor) should induce silencing of Notch targets HEY1/2 and downstream gene EFNB2, while compensatorily upregulating VEC markers {i.e., NR2F2, NRP2 and EPHB4; Figure 28Bleft). None of this happened, suggesting that the canonical Notch pathway is inactive in HUAEC-C {Figure 28C). In this inactive state, there is no NICD that binds to RBPJ, so the latter would be associated to co-inhibitors and repress HEY1/2 expression {Figure 28Bleft). Accordingly, upon siRNA-mediated knockdown of RBPJ expression this repressive mark was lost and HEY1/2 expression was induced, suggesting that RBPJ in HUAEC-C is in a repressive mode {Figure 28Bmiddle, Figure 28D). In accordance, HEY1, HEY2 and EFNB2 expression could be reactivated in HUAEC-C by stimulation with the Notch-ligand DLL4, and this could be blocked again by DAPT treatment (Figure 28Bright, Figure 28 E). To evaluate the overall importance of canonical Notch signalling in the arteriovenous fresh profile regulation in HUAEC-C, we exposed them to anchored DLL4-Fc or bovine serum albumin (BSA). Only a subset (20/64: -31 %) of arterial markers were significantly upregulated upon DLL4-mediated Notch activation and 4 were even downregulated, suggesting additional pathways are involved to fully specify AECs (Figure 28F, Table 13).

Combined overexpression of 8 TFs robustly induces the arteriovenous fresh profile in HUVEC-C The HUAEC-F-specific fingerprint contained 8 TFs (Figure 25D), only 2 of which were previously associated with arterial specification, i.e., HEY2 and SOX17 ( Table 11). Since canonical Notch pathway induction in HUAEC-C was not able to completely restore the arteriovenous fresh profile, some of the newly identified TFs might be important for arterial specification. To determine the role of the 8 TFs we attempted to convert HUVEC-C into cells with a HUAEC-Fexpression profile by overexpression of these factors. Six days after transduction, we analysed expression of the genes within the arteriovenous fresh profile (Figure 29A-B), of markers previously reported to be enriched in AEC or VEC (Figure 29C) and of general EC markers (Figure 29D). Figure 29A shows the heat map analysis of the arteriovenous fresh profile for individual TFs or the combination of all 8 of them. More distant localisation from the cherry control sample indicates a stronger capacity of the TF (combination) to induce arterial specification. As can be observed from the red gene blocks in the hierarchical clustering analysis, the TF largely complemented each other for arteriovenous fingerprint regulation, although some of these blocks also overlapped (Figure 29B). Overall, both analyses revealed that the 8 TF combination delivered the most complete restoration of the arteriovenous fresh profile, while the 'golden standard' for arterialisation HEY2, did not induce strong changes in the arteriovenous fresh profile (Figure 29A-B). Many established AEC markers were also significantly upregulated, while classical venous markers were downregulated by overexpression of the TF combination (Figure 29C). Importantly, except for TEK, we did not observe significant changes in general EC marker expression, suggesting that the TF combination specifically drives arterial specification (Figure 29D). When comparing the converted HUVEC- C to HUAEC-F, for the majority (73%) of arterial genes, the TF combination induced expression levels that were above 2%, 60% of which were in the mid- range (2-25% of HUAEC-F levels) and 40% in the high-range (> 25% of HUAEC-F levels) of expression {Figure 29E). Some of the gene expression changes were validated at protein level by Western blot (Figure 29F). Additionally, analysis of gene expression profile at 28 days post-transduction showed that long-term induction of the arterial phenotype is possible {Figure 30)

Individual TFs interact in a complex network to regulate the arteriovenous fresh profile

We next zoomed in on the relative contribution of individual TFs to the regulation of the arteriovenous fresh profile by evaluating their effect on a gene-by- gene basis. When looking at the arterial arm of the profile, overall, a large proportion of genes (-73%) were regulated by at least 1 TF, with -51 % co- regulated by at least 2 TFs (Figure 31 A). In contrast, only 1 third of the genes was regulated at the venous side (Figure 31B). While some TFs, like Aff3, almost did not exhibit any regulatory effect, others regulated a high percentage of arterial genes from the arteriovenous fresh profile (-41 % for Prdm16, -36% for NKX2-3 and -30% for EMX2; Table 14). Consistent with the global heat map and hierarchical clustering analyses, HEY2 contributed less robustly to arterial gene regulation (-22%; Table 14). Although the hierarchical cluster analysis revealed a significant degree of complementarity between TFs, we also observed overlap, suggesting complex interactions. Therefore, we mapped these interactions in a network (Figure 31C). While some genes (e.g., APO, GRB14, ARL 15) were exclusively regulated by one TF, most genes were co-regulated by more than one TF (e.g. FAT1 was co-regulated by 6 TFs). Furthermore, we also observed cross- regulation between the different TFs, for instance MSX1 being regulated by EMX2, NKX2-3, Prdm16 and TOX2 and HEY2 being regulated by Prdm16 and MSX1 . Finally, some TFs (i.e., TOX2 and SOX17) modulated their own expression. Thus, induction of the arterial gene signature requires a complex interaction of multiple TFs. Arterialised HUVECs form more arterial-like vascular structures in a Matrigel plug implantation assay in vivo.

To determine whether HUVECs transduced with the combination of the 8 arterial TFs were able to form vascular structures which possess more arterial characteristics in vivo, we performed a subcutaneous Matrigel plug implantation assay {Figure 32). In comparison to Cherry-infected HUVECs, cells infected with the 8 TFs formed more elaborate vascular networks {Figure 32A,B), which were significantly more covered by smooth muscle cells {Figure 32C,D). Moreover, the smooth muscle-covered fractional area {Figure 32E) as well as organised collagen deposition was greater in the Matrigel containing the 8 TF-treated HUVECs {Figure 32F).

2.5. CONCLUSIONS AND PERSPECTIVES

Several studies have addressed the question how the identity of arterial endothelium is established. While each of these studies has highlighted a specific pathway active during the arterial specification process, none of them has taken a genome-wide approach. Only one study reported a differential gene screen between several arterial and venous EC types (Chi et al., 2003). However, this study only considered cultured cells. Here, we have shown that the culturing process rapidly 'erases' the specific gene expression profile of arterial ECs. Therefore, to get insight into the in vivo expression pattern of arterial endothelium, we performed a genome-wide differential screen on freshly isolated arterial and venous ECs from human umbilical cord. In this way we established a specific arteriovenous 'fresh' signature of 76 genes. This signature was not only determined by canonical Notch signalling - currently considered as the 'golden standard' for arterial specification - but was co-regulated by additional TFs many of which have not been previously associated with arterial endothelial specification. Overexpression of these TFs in HUVECs revealed that they regulated arterial identity in a complementary but also interactive fashion. Some of the newly identified TFs, such as Prdm16, Emx2 and Nkx2-3, performed better than the current golden standard. Simultaneous overexpression of all eight TFs resulted in the induction of the majority of arterial genes within the newly established arteriovenous 'fresh' signature. To have insight in arterial specification can lead to important applications. Many diseases with a vascular origin are restricted to a certain vascular bed. For instance, atherosclerosis, leading to ischemia, only affects arteries. Restoration of the perfusional defect mostly requires the expansion of arterial blood (i.e., oxygen) supply to the tissues affected by ischemia. Current 'general' revascularisation strategies based on growth factors or stem cells have not taken into account this specific need for arterial supply and this may in part explain their limited clinical success (Conway and Carmeliet, 2004; Deng et al., 2006; Pearson, 2009). Although the Matrigel plug implantation assay we used here is not a pathological model, the fact that we found a more elaborate and arterial-like vascular network (characterised by smooth muscle coating and collagen deposition) in the presence of cells overexpressing the 8 arterial TFs suggests that this overexpression strategy could improve the outcome of cell transplantation in clinically relevant animal models of arterial growth, such as limb ischemia. In addition, increased insight in the arterial specification may provide new targets for specific growth factor therapy. The arterialising TF combination that we identified, along with the arteriovenous signature, can be used in different approaches. One approach would be to pre- specify endothelial progenitors (e.g., blood outgrowth endothelial cells or BOECs) by TF overexpression to an arterial fate before transplantation into ischemic tissues. Another approach would be to use these arterialised BOECs to coat the inside of artificial vessel conduits to reduce their thrombogenicity when implanted in ischemic patients. These cells could also be used as the endothelial component of a 'biological' arterial conduit consisting of an endothelial and smooth muscle cell layer.

Example 3: Identification and validation of Prdm16 as an important arterialising transcription factor acting in part upstream of Notch which induces an arterial endothelial-specific gene (and functional) signature in dedifferentiated cultured human umbilical vein endothelial cells (HUVECs) or blood outgrowth endothelial cells (BOECs)

3.1. ABSTRACT By using a genome-wide transcriptomic screen on freshly isolated murine arterial and venous endothelial cells (ECs), we identified a murine arteriovenous fingerprint that was strikingly different from our human arteriovenous signature. However, meta-analysis on our human and murine arterial and venous endothelial samples revealed a conserved set of genes, defining the arterial and venous identity of ECs, independent of species, developmental stage, anatomical location and haemodynamic parameters. Intriguingly, we identified an arterial-exclusive expression pattern for the transcription factor Prdm16 in both humans and mice. In addition, Prdm16 retained this arterial expression pattern throughout evolution, as demonstrated by whole-mount in situ hybridisation in zebrafish. Prdm16 deficiency in both zebrafish and mice resulted in arterial vascular defects. Furthermore, the lack of a single Prdm16 allele in adult mice impaired their recovery from an ischemic insult, a process that requires expansion of pre-existing arterial collateral vessels. Micro-array analysis of Prdm16-, Hey2-, or control (Cherry) lentivirus- treated endothelial progenitors (blood outgrowth endothelial cells or BOECs) revealed that Prdm16 also instructs ECs to adopt the murine arterial gene signature in a superior fashion than Hey2. These in vitro studies further demonstrated that Prdm16 induces many conserved arterial signature genes in BOECs. Intriguingly, using our conserved arteriovenous fingerprint as a read-out, this arterialising capacity again exceeded that of the Notch target Hey2, the current 'golden standard'. Moreover, the arterialising effect of Prdm16 on the conserved arteriovenous fingerprint was more profound than its effect on our species- restricted fingerprints. Hence, Prdm16 orchestrates the arterial programme of ECs via regulating a set of conserved arterial- and venous-specific genes, in addition to its effect on the species-restricted arteriovenous signatures. Notably, the arterialising effect of Prdm16 on all three arteriovenous fingerprints described outcompeted the arterialising effect of Hey2. Of interest, ectopic Prdm16 overexpression resulted in a robust induction of many Notch components in both HUVECs and BOECs, raising the hypothesis that Prdm16 lies directly upstream of Notch during the arterial specification of ECs. Indeed, concomitant blocking of the canonical Notch pathway upon Prdm16 overexpression severely hampered or completely abolished the inductive effect of Prdm16 on key Notch pathway genes. Likewise, Prdm16 was able to induce several conserved arterial-specific genes, including Gja5 and Dkk2 in the presence of DMSO, but not DAPT. Intriguingly, treatment with DAPT only partially blunted the arterialising effect of Prdm16. Hence, Prdm16 arterialises ECs via canonical Notch-independent and -dependent manners, thereby outperforming the current 'golden standard' Hey2. Indeed, in vivo studies clearly indicated that prdm16 and notch genetically interact during arterial development in zebrafish. In addition, using mutational studies, we found that both the DNA binding- and the CtBP binding-domain of Prdm16 are required for fullblown Prdm16-mediated arterialisation. Finally, Prdm16 strongly induced the expression of semaphorin 3C and - to a lesser extent - semaphorin 3G, known to be attractants for smooth muscle cells (SMCs). Accordingly, SMC coating of arteries of Prdm 76-deficient embryos was strongly reduced. Thus, Prdm16 induced a specific function of arterial ECs, i.e. SMC attraction. Therefore, our data indicate for the first time a role for Prdm16 in the establishment of an arterial gene and functional signature in ECs, in part by acting directly upstream of Notch.

3.2. BACKGROUND AND AIMS

Multiple studies have been devoted to understanding the molecular pathways that determine the development of arteries and veins. Several factors and pathways have been identified, the canonical Notch pathway being considered as the master regulator of arterial fate (Chi et al., 2003). Nevertheless, due to the lack of genome-wide approaches in studying arteriovenous specification, many of the regulating factors remain unknown. Our genome-wide transcriptional profiling studies on freshly isolated human arterial and venous ECs revealed a specific arteriovenous signature (see Example 2). Within this signature we identified eight transcription factors (TFs) that co-regulate arterial endothelial identity of which Prdm (PRD1 -BF1 -RIZ1 homologous domain containing)16 emerged as one of the most powerful TFs.

Prdm 16 is a member of a large family of factors characterised by a PR domain at their N-terminal side. In humans, 17 members of the Prdm family have been identified, and 15 members in mice. Given the fact that members of the Prdm family control processes such as cell commitment, differentiation, growth and apoptosis, it is not surprising that Prdm proteins play important roles during bidirectional cell fate decisions. Perhaps this is reflected by their highly specific expression profiles. Indeed, Prdm members are generally expressed in a cell type- and tissue-specific manner. Prdm16, for instance, is highly enriched in brown adipose tissue (BAT), while its expression is absent in white adipose tissue (WAT) and induces browning of pre-adipocytes (Seale et al., 2007). Although Prdm family members have been extensively described as key molecules in cellular differentiation processes, the underlying molecular mechanism by which they do so remains to be further elucidated. Prdm family members can regulate gene expression according to different mechanisms. First, they can cause epigenetic changes in gene expression profiles, depending or not on their intrinsic histone methyltransferase capacity. Secondly, Prdm molecules also have DNA-binding capacities, through their Zn-finger domains. Thirdly, Prdm molecules may regulate gene expression independent of their DNA binding capacity by complexing with other DNA-binding molecules. Prdm16 for instance can form transcriptional complexes with C-terminal binding proteins (CtBPs) and peroxisome proliferator-activated receptor gamma co- activator 1 (PGC1 ) in adipose tissue. Both are in direct competition with each other to bind Prdm16 and define whether Prdm16 acts as a transcriptional repressor or activator (Kajimura et al., 2008; Seale et al., 2007).

Interestingly, while several Prdm family members have been described in bidirectional cell fate decisions, none, with the exception of Prdm6, has been described to play a role in the vasculature in general, nor in the establishment of endothelial specification in particular. Interestingly, interaction with CtBP molecules is a common feature of multiple Prdm family members and CtBPs have been implicated in vascular development since CtBP2^~/~ mice display extraembryonic vascularisation defects. In addition, CtBP1^+/~:CtBP2^~/~ mice suffered from severe haemorrhaging, a hallmark of vascular defects (Hildebrand and Soriano, 2002). Moreover, a complex formed by Tell (Etv6) and CtBPs regulates the endothelial response to angiogenic stimuli by linking VEGF receptor signalling with intracellular Notch signalling (Roukens et al., 2010). Together with the fact that Prdm16 binds CtBPs in multiple tissues (Endo et al ., 2012; Kajimura et al., 2008), this raises the intriguing question whether a Prdm16-CtBP complex could act at the level of arteriovenous cell fate decisions.

As mentioned above, (canonical) Notch signalling has been hailed as the holy grail for arterialising ECs. Indeed, DII4^+I~ mice, Hey1:Hey2 combined knockout and Su(7-/)-deficient mice all display vascular defects characterised by the reduced expression of arterial markers (most notably EfnB2) and ectopic expression of venous markers (mainly identified by EphB4) on the arterial branch of the vasculature (Swift and Weinstein, 2009). Zebrafish studies confirmed the absolute requirement for Notch signalling for proper arterial development (Swift and Weinstein, 2009). However, although its role in arterial differentiation seems to be evolutionary conserved, Notch-deficient animals retain expression of some arterial- specific genes indicating that other factors, potentially upstream of or in parallel to Notch, must act during the arteriovenous cell fate decisions. Intriguingly, Hamlet (the Drosophila orthologue of Prdm16/Prdm3) modifies the accessibility of Su(H) to Notch target gene promoters (Endo et al., 2012). Moreover, mice deficient for Notch downstream Hes effectors display reduced levels of Prdm16 in the telencephalon during murine neurogenesis (Kinameri et al., 2008). Given the arteriovenous expression pattern of Prdm16 in humans, an interesting question one therefore might raise is whether Prdm16 is tightly linked to Notch in the context of endothelial differentiation.

This study was aimed at further elucidating the potential involvement of Prdm16 in (arterial) vascular development in both physiological and pathological conditions, investigating its potential interaction with Notch signalling and testing its potency to arterialise endothelial (progenitor) cells.

3.3. MATERIALS AND METHODS

Isolation of macrovascular ECs from different species

Mice. The thoracic aorta and vena cava from Tie2-GFP mice (expressing GFP in the blood-vascular ECs or but not in lymphatic ECs (Motoike et al., 2000)) were dissected out and placed in ice-cold MCDB131 medium (Gibco). Excess fat and contaminating tissue around the vessels were removed and vessels were cut into smaller pieces before incubation in 1 X PBS (Gibco, pH 7.4) containing 2 mg/ml crude extract collagenase (Roche) at 37°C until fully dissociated. To obtain a single cell solution, tubes were repeatedly shaken, filtered through a 100 m nylon mesh (BD Biosciences), spun down (600 g, 7 min.), resuspended in PBS supplemented with 1 % BSA and run through a FACS Aria™ device (Beckton Dickinson) to separate the endothelial (GFP⁺) from the non-endothelial (GFP^") fraction.

Humans. After dissecting the umbilical cord after birth, the arterial and venous vessels were flushed with PBS to remove remaining blood cells prior to incubation with collagenase type I (Gibco) at 37°C. EC-containing collagenase solution was collected and vessels were flushed with PBS once more to obtain maximal cell yield. Arterial or venous ECs (HUAECs and HUVECs, respectively) were washed with PBS and resuspended in 300 μΙ MACS buffer (Miltenyi-biotec), with addition of 100 μΙ FcyR blocking reagent and 100 μΙ anti-CD34 magnetic beads (Miltenyi-biotec). After incubation at 4°C, cells were washed with PBS to remove non-attached anti-CD34 magnetic beads prior to resuspension in 500 μΙ MACS buffer and magnetic isolation by the AutoMACS system (Miltenyi-biotec). Afterwards, purity was assessed by FACS using EC markers (CD31 and CD34) and a hematopoietic marker (CD45). Purity typically exceeded 95%.

Zebrafish. FACS sorting of zebrafish ECs was performed on single cell suspensions of 24 hours post-fertilisation (hpf), 48 hpf, 72 hpf and 6 days post- fertilisation (dpf) Tg(Kdr:eGFP)^s843 embryos (Jin et al., 2005) obtained by trypsinisation. Vascular (eGFP⁺) and non-vascular (eGFP^") cell fractions were sorted with high purity in TRIzol® using a FACS Aria™ device.

RNA extraction and cDNA preparation

RNA extraction. To extract total RNA, 1 ml TRIzol® (Invitrogen) was added to the cells and stored at -80°C. After thawing, samples were vortexed and 200 μΙ chloroform was added (Merck) prior to incubation for 10 min. at RT. The samples were centrifuged at full speed for 15 min., after which the upper transparent layer was taken off and transferred into a new Eppendorf tube. 500 μΙ isopropanol (Merck) was added, 200 μΙ 4 M LiCI (Merck) and 1 μΙ glycogen (Invitrogen) was added and the samples were stored for at least 10 min. at RT. After centrifugation at full speed for 10 min., the isopropanol layer was carefully discarded and 1 ml of cold 70% EtOH was added. Afterwards, the samples were centrifuged at top speed for 5 min. and EtOH was carefully discarded. Again, 1 ml of cold 70% EtOH was added and samples were centrifuged at full speed for 10 min.. After discarding the entire EtOH layer, samples were air-dried for at least 10 min. and the remaining pellet was resuspended in 33 μΙ RNase-free-water prior to storage at -80°C. cDNA preparation. For cDNA preparation from RNA samples, 32 μΙ of an RNA sample was mixed with 4 μΙ random hexamer primers and 4 μΙ 10 mM dNTP mix (Superscript III First Strand Kit). The sample was denatured for 5 min. at 65°C and cooled down for 2 min. at 4°C. Simultaneously, a solution of 8 μΙ 10X RT buffer, 16 μΙ MgCI₂ (25 mM), 8 μΙ DTT (0.1 M), 6 μΙ H₂O, 1 μΙ RNase OUT and 1 μΙ SSIII reverse transcriptase (Superscript III First Strand Kit) was prepared. Subsequently, 40 μΙ of the solution was added to each sample and the first cDNA synthesis cycle was started: 10 min. at 25°C, 50' min. at 50°C and 5 min. at 85°C. The samples were cooled down to 4°C and 1 μΙ RNaseH was added prior to incubation for 20 min. at 37°C.

Quantitative reverse-transcription PCR

cDNA was subjected to PCR-based amplification and detected with a nonspecific fluorescent dye (SYBR green) which intercalates in the de novo formed double stranded DNA. Each PCR reaction was performed by adding 1 μΙ cDNA to 6 μΙ SYBR green (Applied Biosystems), 4 μΙ MQ water and 1 μΙ primer mix (2.5 μΜ forward and reverse primer; a primer list is provided in Table 15). The polymerase reaction was performed on a Real-Time PCR system (Step One Plus, Applied Biosystems): 2 min. at 50°C, 10 min. at 95°C and subsequently 40 rounds of amplification at 95°C for 50 sec, each time followed by 45 sec. at 60°C. GAPDH or β-actin were used as housekeeping genes to standardise for the total amount of cDNA in the samples.

Microarray on HUVECs and HUAECs

Microarray analysis was also performed on RNA from freshly isolated HUAECs and HUVECs (N=5 per group), and hybridised on a human genome-wide microarray (Affymetrix Human U133 array), as more elaborately described in Example 2. Statistical analysis and data interpretation was performed in close collaboration with the University of Navarra, (Pamplona, Spain) and Integromics (Madrid, Spain).

Microarray on mAECs and mVECs

Pooled RNA of ECs from 8-12 week-old mice was extracted (RNeasy minikit, Qiagen). The quality and purity of the RNA samples was analysed by the VIB Nucleomics Core facility with a Bioanalyser 2100 and validated by PCR analysis prior to further proceedings. 500 pg of RNA from 5 selected isolations per tissue was amplified, labeled with biotin and hybridised on a mouse genome-wide microarray (Affymetrics Mo Gene1 -0 ST array). The qualitative and statistical analysis of the microarray output was performed by the Nucleomics Core. The analysis was based on the RMA expression levels of the probe sets that had at least once a present MAS 5.0 detection call. Differential expression was assessed via the moderated f-statistic. To control the false discovery rate, multiple testing correction was performed according to Benjamini and Hochberg. We considered genes to be differentially expressed between mAECs and mVECs if they displayed at least 2-fold enrichment in either of these two fractions ([log-value]> 1 ) with a corrected P-value < 0.001 . Such stringent approach was chosen to obtain 'workable' amounts of genes.

To further define a conserved arteriovenous signature, we conducted a meta-analysis on all microarray samples discussed above, using a Fischer i-test on the combined data sets. To map the probe sets of both microarray platforms (i.e., Human Genome U133 Plus 2.0 Array and the GeneChip Mouse Gene 1 .0 ST Array), we mapped the probe sets of the Human U133 Plus 2.0 array to the Mouse Gene 1 .0 ST array. This was done by mapping of homologous/orthologous probe sets with the package annotation Tools of BioConductor (annotation files HG- U133_Plus_2.na32.annot.csv and Mouse Gene_1.na32.annot.csv as downloaded from the Affymetrix website and HomoloGene database homologene.data build 66 as downloaded from NCBI). In case a probe set of one array maps to several probe sets of the other array, they are mentioned multiple times in the data table.

Nanostring on HUVECs overexpressing arterialising TFs RNA from Prdm16-overexpressing, Hey2-overexpressing or Cherry control- overexpressing HUVECs (A/=4 per group) was used for a comparative Nanostring experiment (which also included 6 additional TFs, as decribed in detail in Example 2). RNA quality/concentration were determined with a Bioanalyser 2100 (Agilent Technologies, Santa Clara, CA) and 100-500 ng of total RNA was hybridised, according to the manufacturers' instructions, in collaboration with the VIB Nucleomics Core Facility at KU Leuven. Some of the results were confirmed by qRT-PCR and for those genes for which the Nanostring probe intensity value was low, results were analysed by qRT-PCR. Whole-mount in situ hybrydisation (WISH) on zebrafish embryos

Probe synthesis. Primers were designed using Primer 3.0 software (http://frodo.wi.mit.edu/primer3/) to amplify a 601 bp long fragment of zebrafish prdm16. The following primers were used: 5 TGACCAGTGCCCCAAAG3' and 5'TTCTTTCCCTCGCAAAAGC3'. A cDNA sample containing a mix of cDNA derived from whole zebrafish (uniZF) was used as a scaffold for the PCR based amplification of prdm16. Half of the PCR product was run on a gel to confirm proper fragment length and 2 μΙ of the remaining product was ligated according to the manufacturers' protocol into a pCRII-TOPO® TA vector (Invitrogen). Linearised DNA (1 g) underwent transcription to produce either labelled sense or antisense probe by incubation for 2 h at 37°C with labelled NTPs, 10X buffer (Roche), DEPC- treated water, RNA polymerase (either for Sp6 or for T7; Roche) and Protector RNAse inhibitor (Roche). At last, the RNA probe was purified added to hybridisation buffer (Hyb+) and stored at -20°C. From this stock concentration, a 1 :100 working concentration was used for performing WISH. WISH. AB zebrafish embryos were collected at 24 and 48 hpf and fixed using Memfa fixative (1 M MOPS, 10 mM MgSO₄ and 20 mM EGTA) for 1 h on a shaker. Afterwards, the fixative was replaced by 100% EtOH. After a few minutes, fresh EtOH was added to the embryos for storage at -20°C. Alternatively, embryos were directly submitted to WISH. First, embryos were rehydrated and rinsed three times with PBS-T (PBS containing 0.1 % Tween20). Afterwards, embryos were permeabilised with 1 X proteinase K in PBS-T for 15 min. and subsequently transferred into 0.1 M triethanolamine, into 0.1 M triethanolamine with 1 X acetic anhydride solution, into 0.1 M t ethanolamine with 2X acetic anhydride solution and washed two times with PBS-T before fixation with formaldehyde-PBS-T for 20 min.. After four washes with PBS-T, the embryos were shortly incubated in the settling buffer (50:50 mix of hybridisation buffer and PBS-T). After a brief pre- incubation in hybridisation buffer (10 min. at 60°C), embryos were incubated overnight at 60°C in the hybridisation solution containing the digoxygenin (DIG)- labelled prdm16 probe. The next day, embryos were washed several times at room temperature (RT) with 2X saline sodium citrate (SSC), 2 times with 0.2X SSC and ultimately two times with MAB (maleic acid buffer). Subsequently, 1 ml of MAB-BR (blocking reagent) was added for 30 min., prior to incubation for 4 hours with 1 ml of MAB-BR-serum with anti-DIG-AP antibody (1 :200, Roche). Finally, embryos were four times incubated with MAB for 20 min. and one time overnight at 4°C. The following day, the embryos were rinsed two times with alkaline phosphatase (AP) buffer before putting them overnight on 1 ml BM purple AP substrate in the dark (i.e. covered with aluminum foil) at RT. The next day (or after several days depending on the staining intensity), the chromogenic reaction was stopped by adding PBS-T. Subsequently embryos were fixed with multiple aliquots of Memfa fixative and finally, embryos were bleached, rehydrated and stored in glycerol until further processing. Whole mount pictures were taken using a Zeiss Axio Imager Z1 microscope. Alternatively, embryos were paraffin-embedded, cross-sectioned (7 μιτι) on a microtome (Leica) and counterstained with nuclear fast red (NFR). Briefly, slides were deparaffinised and placed into the NFR solution for 5-10 min.. Afterwards, slides were rinsed with MQ water, dehydrated and mounted with DPX. Cross-sections were examined on a Zeiss Axio Imager Z1 microscope. Production of wild-type and mutant Prdm16 expressing ientivirus

Vector cloning. A pYX-Asc vector containing the full cDNA for mPrdm16 was purchased from Open Biosystems (clone ID 6409778). mPrdm16 was subsequently amplified by means of PCR (primer list: see Table 10) using Phusion® Hot Start II DNA polymerase according to the manufacturers' protocol and cloned into a pRRL2-PGK-Cherry vector using Xbal and Xhol restriction enzymes. Plasmids were transformed and purified by miniprep. Correct insertion was confirmed by sequencing and maxipreps were made prior to lentiviral production.

Mutagenesis. To insert mutations in the sequence coding for the DNA- binding region of Prdm16 (Prdm164D/V/A), we used the QuikChange® Site- Directed Mutagenesis Kit (Stratagene) according to the manufacturers' protocol. In brief, the wild-type pRRL2-Prdm16 PGK-Cherry plasmid served as a template for PCR amplification, using primers carrying the desired mutation (primers used for mutation were: 5'AACCTCCAGCGGCACGTGCAG3' and

5TTCTCTTTGTTGTGGATGTTC3'). After amplification of the plasmid DNA with the designed mutagenic primers, the non-mutated template string was eliminated by restriction digest with Dpnl for 1 hour at 37°C, specifically breaking down parental adenine-methylated GATC sites. Mutated plasmid was then transformed into TOP10 bacteria and plated on ampicillin containing agar plates. Single bacterial colonies were picked from which plasmid DNA was isolated as miniprep. The vector samples were sequenced to verify if the desired mutation was present. The Prdm16 mutant inable to interact with CtBP {Prdm16ACtBP) was kindly provided by S. Kajimura (UCSF, CA, USA).

Virus production. HEK293 cells were plated (5x10⁶ cells/10 cm dish) and the next day transfected with our plasmid of interest together with two helper plasmids (psPax2 and PMD2G) using Fugene transfection reagent (Roche). In brief, 400 μΙ of OPTIMEM (Gibco) was mixed with 1 g PMD2G, 3 g psPax2 and 4 g pRRL2 plasmid. 24 μΙ of Fugene was added and the mixture was incubated for 20 min. at RT and gently applied to the cells. The next day, medium was replaced and lentiviral particle-containing supernatant was collected 36 h later. Viruses were concentrated by centrifugation and used directly to transduce HUVECs/BOECs (or stored at -80°C). The lentiviral construct for overexpression of HEY2 was obtained from Genecopoeia (Rockville, USA; Table 10).

<xSMA/Prdm16 double staining on frozen sections

E14.5 and E17.5 FvB embryos were dissected out, rinsed in RNA later and put in a 15 ml falcon tube to snap-freeze in liquid nitrogen. Unfixed cryo-preserved embryos were sectioned (7 μιτι) on a cryostat (Leica, CM3000) and sections were stained for Prdm16. Therefore, air-dried sections were incubated with 4% paraformaldehyde (PFA) for 10 min. at 4°C, sections were washed with MQ water and three times with Tris-HCI-NaCI-Tween (TNT) before they were immersed in PBS-Tr (PBS with 0.1 % Triton-X) for 30 min.. The sections were blocked for 1 h with tris-NaCI-BMP buffer (TNB) containing 20% pre-immune donkey serum (PID; Sigma-Aldrich) and incubated overnight on a shaker with Sheep-anti-Prdm16 primary antibody (1 :20 in 20% PID/TNB; R&D Systems) at 4°C. After washing three times with TNT buffer, the sections were incubated with secondary Donkey-anti- Sheep-Texas Red antibody (Jackson Laboratories; 1 :100) in TNB and FITC- conjugated anti-aSMA (Sigma; 1 :500) for 2 h at RT and once again washed three times with TNT before mounting with a glue (prolong gold) containing DAPI.

Haematoxylin and eosin staining on paraffin sections

Slides were briefly submerged in MQ water prior to deparaffinisation. Afterwards, they were incubated in hemaluin dye for 3 min., washed with flowing water and put in eosin dye for 10 sec. Finally, slides were washed with MQ water until its red colour disappeared and subsequently dehydrated again before being mounted with DPX. aSMA/Coup-TFII double staining on paraffin sections

Sections were deparaffinised and subjected to antigen retrieval (based on a pH 6-citrate buffer) and microwave heating. Slides were washed three consecutive times with tris-buffered saline (TBS) prior to blocking with 10% pre-immune goat serum (PIG) in TNB for 1 h. In a next step, slides were incubated overnight with primary Ab against Coup-TFII (Perseus Proteomics; 1 :200) at 4°C. The next day, slides were washed multiple times with TNT, incubated with Goat-anti-Mouse Alexa 568 secondary Ab to visualise Coup-TFII and with FITC-conjugated anti-aSMA antibody (Sigma; 1 :500) to visualise the SMC layer. Finally, slides were washed several times with TNT and mounted with prolong gold containing DAPI. SMC density of the jugular aorta was measured as the difference between the outer and inner (luminal) surface area of the artery using NIH Image J software. 5 to 10 images were analysed for each mouse.

Whoie-mount β-galactosidase staining on murine tissues Adult vessels (aorta, vena cava) from Prdm16^+/~ embryos, containing the β- galactosidase gene (http[//www.informatics.jax.org/javawi2/serylet/WlFetch ?page=alleleDetail&key=37049), or E10.5 embryos derived from Prdm16^+/~ intercrosses, were processed as follows: tissues were briefly washed with PBS prior to fixation at 4°C with fixing solution for 20 min.. Tissues were subsequently washed repeatedly with PBS and incubated with staining solution at 30°C overnight. Next day, tissues were washed three times with PBS and post-fixed with 4% PFA for 2 h at RT. Tissues were washed multiple times and further processed for paraffin sectioning and nuclear fast red (NFR) staining. Adult tissues from Prdm16^+/+ mice were used as an internal negative control.

Morpholino knockdown and live screening in zebrafish embryos

Knockdown of zPrdm16 was achieved by injecting Tg(kdr:eGFP)^s843 zebrafish with a morpholino (Mo) at the one cell stage. The prdm16 Mo is a 25 bp oligomer (purchased from GeneTools LLC; sequence 5'-3': ATATGCTGCCCAAGACTAGAAATAC) which complementary binds to the region containing the ATG start codon and thereby blocks translation. The Mo was diluted in phenol red (resulting in a better contrast during and after injection) to obtain the correct Mo concentration. First, the injection needle was calibrated at 5X magnification of a Sterna 2000-C microscope (Zeiss). After injection at 0.8X of the same microscope, at 200-300 hPa for 0.2 sec, the embryos were incubated at 28°C (in a Binder incubator). We used up to 14.4 ng prdm16 Mo per embryo. Our findings were confirmed with a second splice-site morpholino (5'- ACTCACACTATCACCCACCTTATCA-3' ) . To address the genetic interaction between prdm16 and notch in vivo, we injected Tg(kdr-eGFP)^s843 zebrafish with a suboptimal dose (10.8 ng) of prdm16 Mo at the single cell stage and manually dechorionated the embryos 6 to 7 h later. Dechorionated embryos were incubated in 0.3% Danieau water to which we added a final concentration of 12.5 μΜ DAPT or an equal volume of DMSO. Zebrafish embryos were screened for vascular defects at 48 hpf. Alternatively, we injected Tg(kdrl-eGFP)³⁸⁴³ zebrafish with morpholinos targeting either prdm16 or grl or the combination of the latter two. Briefly, 10.8 ng prdm16 Mo or 3.6 ng grl Mo were injected alone or together. Ns Mo was used to compensate for the differential amount of Mo between conditions. A bolus injection of 14.4 ng ns Mo was used as a control.

Live screening or confocal imaging of Tg(kdr:eGFP)^s843 zebrafish embryos was performed using a Zeiss Lumar V.12 fluorescence stereomicroscope equipped with a Zeiss AxioCam MrC5 digital camera or a laser-scanning microscope LSM510, respectively. The development of the intersomitic vessels (ISVs) was analysed in 10 consecutive somite segments. Scores ranging from 0 to 10 were given to each embryo, depending on the number of normally developed ISV in those 10 segments. Typically, blood vascular defects were analysed at 48 hpf. Only embryos that were alive and had a heart beat were included in the analyses. Hind limb ischemia in mice

Prior to surgery, mice were anaesthetised with 200 μΙ of a (26:8:66) mixture of ketamin: xylazin: NaCI. During the surgery, mice were placed on a heating pad and the temperature was monitored to remain at 37°C. The fur of both hind limbs was removed by Veet® treatment. After making an incision to open the skin, the nerve was carefully dissociated from the femoral artery in the right upper limb. Subsequently, to induce femoral artery occlusion, two surgical clamps were placed onto the right femoral artery, one above and one below the branching point with the arteria caudalis femoralis. Additionally, to ensure a complete lack of blood flow, an incision was made in the femoral artery between the two clamps. The skin was stitched and the closed wound was disinfected with iodine. While the mouse was still under anesthesia, successful interruption of blood flow was monitored by Laser Doppler scanning (Lisca; see below). Mice were kept on the heating pad until they awoke.

The follow-up of the ischemic mice implied non-invasive monitoring of blood flow during recovery of the ischemic insult by Laser Doppler scanning. This system scans a preselected area of the hind limb with a laser beam. When the laser beam hits red blood cells in motion, a sound wave is sent back to a detector in the scanning head. The intensity of this feedback signal is proportional to the amount of moving red blood cells (i.e., the blood flow) and is transformed into a colour code by the Lisca software, red representing high blood flow, blue/black representing low/no blood flow. The left non-ligated limb was also scanned and used as an internal reference to calculate the relative perfusion in the ligated right limb. In addition to the scanning session immediately following surgery, Laser Doppler measurements were performed at day 3, 7, 10, 14 and 21 after surgery, under isoflurane anaesthesia and temperature monitored (37°C) conditions.

At day 21 , mice were sacrificed. Therefore, mice were sedated with a (26:8:66) mixture of ketamine: xylazine: NaCI. The vessels were perfused (by inserting a needle into the apex (left ventricle) of the heart and making an incision in the right atrium) with adenosine to induce vasodilatation. Next, vessels were perfused with Zinc Fix (zinc formalin fixative) for 10 min. The right adductor and gastrocnemius muscles were dissected out and placed in Zinc Fix overnight. The next day, after washing the tissues three times with MQ water, they were stored in 70% EtOH at 4°C until further histological processing.

Sirius red staining on paraffin sections and analysis

To measure fibrosis, cross-sections of the gastrocnemius muscle were stained with the sulphonated azo-dye Sirius red. Briefly, after deparaffinisation, sections were incubated for 90 min. in freshly prepared Sirius red dye solution, followed by differentiation in HCI . Slides were dehydrated and mounted with DPX. Fibrosis was determined by measuring the area taken up by Sirius red staining using ImageJ software on 4 images randomly chosen from three different sections per mouse. Lentiviral transduction of HUVECs or BOECs

HUVECs were obtained as described in Example 2. BOECs from peripheral blood or umbilical cord blood were obtained as described under Example 1. HUVECs or hBOECs were plated one day prior to transduction into a 24-well plate (25,000 cells/well). The next day, 0.5 to 10 μΙ of virus (depending on the virus and the virus production) was added to the medium. The morning following transduction, medium was changed and cells were directly harvested 5 days later (6 days after transduction) on TRIzol® lysis buffer. RNA cDNA was made and analysed by qRT-PCR. Alternatively, we added 50 μΜ DAPT or its solvent DMSO to the medium. For luciferase assays, cells were transduced as described above. However, cells were co-transduced with a RBPJK-luciferase reporter virus (Cignal Lenti RBP- Jk Reporter (luc) Kit: CLS-014L; SABiosciences; Qiagen) and a lentiviral Renilla control to normalise for transduction efficiency (Cignal™ Lenti Renilla Control (luc): CLS-RCL; SABiosciences; Qiagen). Medium was changed the next day and again at day 3 and cells were collected in 50 μΙ 1 x luciferase lysis buffer 6 days after transduction. Firefly and Renilla luciferase activity were measured from 20 μΙ of lysate according to the manufacturers' protocol using the Dual-Luciferase® Reporter Assay System (E1910) from Promega on a Microplate Luminometer LB 96V (EG&G Berthold).

Statistics

Data, expressed as mean ± s.e.m comparing two groups were analysed by

Student's i-test: for single comparisons P-values were calculated with two-tailed unpaired Student's i-test; for multiple comparisons P-values were calculated with one-way ANOVA, followed by contrast analysis. For the limb ischemia study, repeated measures ANOVA was applied, followed by Student's i-test for each time point. For phenotypic scoring of zebrafish, the penetrance of the phenotype, the number of embryos exhibiting the different severities of morphant phenotype was counted and Chi-square analysis was used to determine whether this fraction differed between control or dose groups. SPSS software was used for statistical analysis and differences were considered significant when P<0.05. 3.4. RESULTS

Prdm16 has an arterial-exclusive expression pattern throughout evolution

Our microarray screen on human ECs from human umbilical cord arteries and veins revealed that PRDM16 was highly enriched in arterial ECs {Example 2). In a next step, we were eager to establish an arteriovenous fingerprint and identify their transcriptional regulators in another species, since evolutionary conserved mechanisms most likely play the most pivotal role in the establishment of the arterial/venous identity of ECs. Since it was technically not feasible to derive sufficient ECs from murine umbilical arteries and veins, we sorted GFP⁺ ECs from the thoracic aorta and vena cava of adult Tie2-GFP mice and performed gene profiling on these samples. As for the human profiling studies, we used a stringent statistical comparison to define our murine arteriovenous fingerprint. Remarkably, there was only a very limited overlap of our human arteriovenous fingerprint with the murine arteriovenous gene signature. More specifically, almost none of the genes from the human fingerprint were also significantly differentially expressed between murine (m)AECs and mVECs. Thus, the murine arteriovenous signature represented a completely novel set of 68 and 85 probe sets differentially expressed between mAECs and mVECs, corresponding to 65 arterial- and 62 venous-specific genes, respectively {Table 16).

COUP-TFII, a TF known as the prime VEC denominator (You et a/., 2005) was significantly enriched in mVECs compared to mAECs. However, due to our stringent statistical standards, COUP-TFII was not part of our human arteriovenous fingerprint, despite its differential expression pattern. This prompted us to evaluate more broadly the expression profile of all TFs differentially expressed between AECs and VECs of either human or murine origin, to avoid missing out on potentially strong candidates which could play pivotal roles in arteriovenous specification. A summary of the average probe intensities of these TFs in both AECs and VECs from human and murine origin is listed in Table 17. Intriguingly, this analysis revealed that only few TFs retained their differential expression pattern across species. Indeed, many of the arterial TFs of the human (including AFF3, EMX2 and TOX2) and the murine (Dlx5 and Isl1) fingerprint only displayed a differential expression in the original species in which they were identified. Intriguingly however, Coup-TFII remained a valid VEC determinant in both species, confirming the validity of our data. Additionally, some arterial TFs were significantly enriched in both human and murine AECs, compared to their venous counterparts. Indeed, both Hey2 and Sox17, as well as Prdm16 were significantly differentially expressed between both HUAECs versus HUVECs and mAECs versus mVECs, although only Hey2 made the stringent cut-off values set by us in both species. As Hey2 and Sox17 have been previously associated with the arterial vasculature and given the strong arterialising capacity of Prdm16 in HUVECs (see Example 2), we decided to further elaborate on the role of Prdm16 during arterial development. Our profiling studies suggested that, even though the arteriovenous gene signature may be different depending on species (or other variables), these signatures may be determined by common TFs. Hence, we next wondered whether Prdm16 also regulated the murine arteriovenous signature, and whether it outperformed Hey2, parallel to our findings concerning the human signature in HUVECs. Therefore, we overexpressed Prdm16, HEY2 or a negative control (Cherry) in human endothelial progenitor cells (i.e. blood outgrowth endothelial cells or BOECs) and performed microarray analysis on these samples (A/=3 for each condition). Prdm16 overexpression resulted in the significantly increased expression of a subset of genes (14 out of 48 detectable genes; -29%) preferentially expressed on mAECs. Conversely, unlike Prdm16 overexpression in HUVECs, Prdm16 significantly suppressed the expression of -31 % (15 out of 48 detectable genes) of the mVEC-specific genes in BOECs, while only enhancing the levels of -10% of these genes. Thus, Prdm16 clearly induced an arterial shift in BOECs when considering the murine arteriovenous fingerprint. Surprisingly, unlike previously shown in HUVECs (Chi et ai, 2003), under our experimental conditions in BOECs, HEY2 significantly upregulated or downregulated only a limited number of genes upon overexpression and none of these were part of the murine arteriovenous fingerprint. This further suggests that also for the murine arteriovenous fingerprint, Prdm16 was a stronger 'arterialiser' than the golden standard Hey2.

Determination of the arteriovenous fingerprint across species: towards a conserved arteriovenous signature

When performing stringent statistical analysis on the human and murine data sets separately, we obtained two distinct, almost non-overlapping arteriovenous gene signatures. Yet, some of the transcriptional regulators were common to both species. We next wondered whether a meta-analysis considering both data sets together and thereby using an alternative biostatistics approach would deliver a common - thus conserved - gene signature, the regulation of which would potentially be better than for the two separate signatures by the common TFs, i.e., Prdm16, Hey2 and Coup-TFII. When only considering genes that were at least twofold enriched in AECs or VECs in both species, the meta-analysis on our human and murine microarray data gave us a third list of genes differentially expressed between arteries and veins (21 arterial and 30 venous genes; Table 18). The arterial versus venous log2 ratios of these genes are plotted in Figure 33 for both human and murine samples.

Notably, Prdm16, Hey2 and Coup-TFII were retained in this list. In addition to these master regulators of arteriovenous identity, among the arterial-specific genes, several were previously associated with arteriovenous differentiation or arterial defects in mice or zebrafish (e.g. HEY2, GJA5 and PTPRJ) upon knockdown/knockout {Table 18). Thus, this third fingerprint exhibits hallmarks of a conserved signature that defines the arterial or venous identity of ECs across species. Therefore, we took this third list as our read-out for mechanistic studies discussed below.

Like our 2 species-restricted fingerprints, this conserved arteriovenous signature responded in the expected fashion to lentiviral Prdm16 overexpression, as evidenced by our microarray results in BOECs. More importantly, the arterialising effect of Prdm16 on this final list exceeded that on our species- restricted fingerprints, as Prdm16 induced -63% of the arterial genes, while suppressing -64% of the venous markers, thereby clearly shifting the balance towards a more arterialised phenotype (A/=3-3; Figure 34 A). Furthermore, like for the species-specific arteriovenous gene signatures, comparative microarray profiling of HEY2 overexpressing or Prdm16 overexpressing BOECs, revealed that the arterialising effect of HEY2 was inferior to that of Prdm16. In fact, when considering the conserved arteriovenous gene list, HEY2 overexpression only induced the expression of -47% of the arterial genes, while suppressing -43% of the venous markers {Figure 34B).

Finally, to further validate this conserved arteriovenous signature, we conducted microarray analysis on HUVECs treated with an siRNA against NR2F2 (siNR2F2) or a negative control (siNS). We hypothesised that silencing of NR2F2 - a prime determinant of venous specification by inhibiting Notch (You et ai, 2005) - would augment the expression of our newly identified arterial-specific genes, while suppressing the venous-specific genes. As expected, knockdown of COUP-TFII enhanced the levels of several arterial-specific genes, including HEY2. However, the arterialising effect of siCOUP-TFII was less dramatic than that observed in our Prdm16 overexpression studies, mainly because of a modest inhibitory effect on the venous genes (Figure 34C). In conclusion, the evolutionary conserved arterial- specific TF Prdm16 is a firm driver of EC arterialisation by regulating a set of well- established arterial and venous enriched genes, thereby outperforming the Notch pathway. To confirm our microarray results, we performed PCR analysis for PRDM16 and two endogenous controls, HEY2 (for arterial ECs) and NR2F2 (also known as COUP-TFII; for venous ECs), on independently derived human umbilical cord samples and ECs from murine aorta and vena cava. PCR data confirmed the arterial enrichment of PRDM16 and HEY2, while NR2F2 was clearly enriched in venous ECs ([fold upregulation HUAECs versus HUVECs]: 15 ± 2, P<0.001 for HEY2; 45 ± 6, P<0.001 for PRDM16; 0.1 1 ± 0.01 , P<0.01 for NR2F2; N=4; Figure 35A). Similar results were obtained in ECs isolated from murine aorta (mAECs) and vena cava (mVECs; [fold upregulation mAECs versus mVECs]: 4.7 ± 0.2, P<0.05 for Hey2; 47 ± 18, P<0.05 for Prdm16; 0.16 ± 0. 12, P<0.05 for Nr2f2; N=4; Figure 35B).

PRDM16 staining confirmed the presence of human PRDM16 protein on HUAECs, while absent on HUVECs {Figure 35C,D). To further validate the strictly arterial expression pattern of Prdm16 in mice, we studied Prdm16 protein expression taking advantage of Prdm16^+/~ mice carrying one allele in which the β- galactosidase gene was knocked-in into the Prdm16 locus. 5-bromo-4-chloro-3- indolyl- -D-galactosidase (X-gal) stainings revealed clear Prdm16 expression in the arterial endothelium of E10.5 embryos, including in the DA and the vitelline artery {Figure 35E.F). Conversely, we could not detect Prdm16 on ECs from the (common) cardinal or vitelline vein {Figure 35E,F). Prdm16 immunofluorescence staining using a commercially available antibody confirmed our findings on later developmental stages: at E14.5, Prdm16 was readily detected in the endothelium of intercostal arteries and the jugular artery, while it was absent in their venous counterparts {Figure 35G,H). Similarly, at E17.5, ECs from coronary arteries, but not coronary veins stained for Prdm16 {Figure 35I,J). Likewise, X-gal staining on adult thoracic aorta and vena cana of Prdm16^+/~ mice confirmed the persistent exclusive arterial expression profile of Prdm16 {Figure 35K-M). Thus, the expression of Prdm16 is restricted to the arterial branch of the vasculature in both humans and mice from early development through adulthood. Notably, Prdm16 was not only detected on AECs, but was also frequently found on arterial SMCs (Figure 35G,H).

Early vascular expression of rdm16 in zebrafish

Next, we investigated whether Prdm16 is also expressed in the vasculature of the zebrafish, a vertebrate organism that is taxonomically quite distant from humans and mice, but has been well documented for vascular research. Therefore, we separated endothelial from non-endothelial fractions using Tg(kdr:eGFP)^s843 zebrafish embryos, at different developmental stages after which we analysed zebrafish prdm16 expression by means of qRT-PCR. With this approach, a temporal expression profile during embryogenesis was obtained revealing an endothelial expression peak at 24 hours post-fertilisation (hpf; Figure 36 A). This is a very relevant timeframe since the two major trunk vessels (the dorsal aorta or DA and posterior cardinal vein or PCV) are properly formed and obtain their arteriovenous identity around this time in zebrafish. After 24 hpf, prdm16 expression - although still detectable - gradually declined on the vasculature.

These timeline experiments did however not allow us to make a subdivision between arterial and venous ECs. Therefore, to gain insight into the spatial distribution of prdm16 expression, we performed whole-mount in situ hybridisation (WISH; Figure 36B,C). Whole mount pictures of 24 hpf AB zebrafish, demonstrated prdm16 expression in the DA but not the PCV. Additionally, some embryos were cross-sectioned and counterstained with nuclear fast red (NFR). A representative picture {Figure 36C) shows signal in the DA, while prdm 16 expression remained absent in the PCV. Since zebrafish do not have SMCs or pericytes in the trunk vessels at this developmental stage, vascular expression of prdm16 at this stage can be regarded as arterial endothelial-exdusive. At 48 hpf the expression profile of prdm16 remained exclusively arterial, however less prominent as its expression in the notochord became predominant (data not shown). Therefore, we conclude that Prdm16 displays an arterial exclusive expression pattern in the zebrafish, murine and human vasculature. Prdm16 deficiency causes severe (arterial) vascular defects in zebrafish and mice

Prdm16 deficiency in zebrafish results in (arterial) vascular defects

To investigate its functional importance in AECs, we knocked down prdm16 in Tg(kdr-eGFP)^s843 zebrafish embryos by injecting a morpholino specifically targeting its ATG region (prdm16 Mo). Prdm16 morphants displayed a clear vascular phenotype {Figure 36D-F) at 48 hpf, characterised by defective development of the (arterial) ISVs and an improperly formed DLAV. Indeed, 48% of zebrafish injected with prdm16 Mo suffered impaired ISV formation, while nearly all (92%) control (ns Mo) embryos had a normal ISV phenotype (P<0.0001 versus ns Mo; A/=99-99). Moreover, in more than half (52%) of the prdm16 morphants, the DLAV did not or only partially form, while 97% ns Mo-treated embryos showed normal DLAV development (P<0.0001 versus ns Mo, A/=99-99). In addition, prdm16 morphants often displayed aortic hypoplasia, making it hard to distinguish between the DA and PCV, although there was no evident fusion between the two axial vessels. All these defects could be recapitulated via injection of a splice-site morpholino (data not shown).

As Prdm16 is a likely candidate to act during arteriovenous cell fate decisions, we analysed the expression of several arterial and venous markers on sorted eGFP⁺ ECs from prdm16 and ns Mo-treated embryos at both 24 and 48 hpf. However, we failed to detect any sigificant downregulation of either dll4, efnb2a or gri Vice versa, eGFP⁺ ECs from prdm16 Mo-treated embryos did not have elevated levels of the venous marker coup-tfll (data not shown). As these results were unexpected, we turned to WISH to see whether dll4, efnb2a and gri displayed ectopic expression on the PCV or other vascular structures. WISH for dll4, efnb2a and gri did however not reveal a mispatterned expression profile as none of these genes showed ectopic expression on the PCV or a reduced arterial expression. WISH for the venous marker coup-tfll also did not reveal an aberrant expression pattern (data not shown).

Prdm16 deficiency in mice results in (arterial) vascular defects In order to address the role of Prdm16 in murine vascular development, we performed timed-matings between Prdm16^+/' x Prdm16^+/~ intercrosses and initially collected embryos at E14.5 as we had done for previous studies. Collecting embryos at this specific stage allowed us to examine all three macrovascular vessels within one region {Figure 37A). Indeed, at this developmental stage, embryos possess a major artery, vein and lymphatic within their jugular region, and are therefore particularly useful to study macrovascular endothelial heterogeneity. Close examination of these embryos revealed subtle, but consistent vascular defects. Indeed, H&E stainings demonstrated that Prdml ^'6^{'Α '}, but not Prdm16^+/+ or Prdm l 6^+/~ E14.5 embryos displayed haemorrhages throughout their body {Figure 37B,C). Moreover, in some of these Prdm l 6^'^ embryos, red blood cells could be seen leaking out of the artery {Figure 37D-G). This vascular leakage is most likely secondary to the fragile structure of arteries of Prdm 76-deficient embryos. Indeed, aSMA staining on sections of the jugular region of these embryos revealed a reduced coverage by vascular SMCs in the artery compared to their Prdm16^+/~ or Prdml 6^+/+ littermates ([aSMA⁺area in μηη²]: 3777 ± 102 versus 2662 ± 88 for Prdml 6^+/+ versus Prdml 6^~ mice; P<0.0001 ; N=7-7 and 3625 ± 213 versus 2662 ± 88 for Prdm l 6^+/~ versus Prdm 16^'/' mice; P<0.01 ; A/=1 1 -7; Figure 37D-H). Moreover, the residual SMCs that were present failed to organise properly and revealed altered morphology. Analogous to our experiments in zebrafish, we failed to detect ectopic expression of the venous TF Coup-TFII on AECs (not shown).

Heterozygous deficiency of Prdml 6 impairs collateral arterial reperfusion

The subtle but consistent vascular defects seen in our knockout embryos, prompted us to hypothesise that Prdm 76-deficient mice might have impaired vascular recovery upon an ischemic insult. As Prdm16^' mice do not survive beyond birth, we therefore submitted Prdm16^+/+ and Prdm16^+/~ mice to a model of moderate limb ischemia - also known as intermittent claudication, as the recovery of such an ischemic insult is highly dependent on the expansion of native arterial collaterals, in a process termed adaptive arteriogenesis. The recovery of Prdm16^+/~ mice was indeed significantly impaired compared to their Prdm16^+/+ littermates as determined by laser Doppler perfusion measurements (P<0.001 ; Λ/=12-12; Figure 38A-C). One week after surgery, for example, Prdm16^+/+ mice had already 37 ± 3% of normal blood perfusion, while Prdm16^+/~ mice only returned to 25 ± 2% of their normal perfusion profile (P<0.01 ; A/=1 1 -12), suggesting impaired collateral growth specifically in adductor muscles of Prdm16^+/~ mice, as Pecaml staining revealed no difference in vascular density of the gastrocnemius muscle between the two genotypes (Figure 38D,E).

In accordance with their reduced blood flow during the recovery phase, a higher percentage of Prdm16^+/~ mice displayed macroscopic signs of necrosis, compared to their control littermates throughout the experiment ([% of mice with macroscopic signs]: 92% of Prdm16^+/~ mice versus 50% of Prdm16^+/+ mice at day 3; 83% of Prdm16^+/- mice versus 17% Prdm16^+/+ mice at day 21 ; P<0.05; A/=12-12; Figure 39 A). Representative images of the lack of macroscopically observable necrosis in Prdm16^+/+ or its presence in Prdm 16^+/~ mice are given in Figure 39B and C, respectively.

Finally, histological examination of the gastrocnemius muscle of these mice revealed a two-fold higher degree of fibrosis in Prdm16^+/~ mice, compared to their Prdm16^+/+ littermates ([% fibrotic area]: 4 ± 1 in Prdm16^+/+ mice versus 8 ± 1 in Prdm 16^+/- mice (P<0.05; A/=1 1 -12; Figure 40).

Prdm16 induces an arterial phenotype in ECs in part through canonical Notch

Prdm16 and Hey 2 co-regulate multiple arterial-specific genes

Since the ultimate goal of this project was to generate new tailor-made vascular therapies, using pre-differentiated stem/progenitor cells, we were keen on finding out whether Prdm16 could direct ECs towards an arterial phenotype. Therefore, we first overexpressed Prdm16 in established venous ECs (HUVECs) using a lentiviral construct. Nanostring analysis revealed that Prdm16 significantly induced many (23/64; -36%) of the arterial genes we found in our human microarrays (see also Example 2). Moreover, plotting the arterialising effect of Prdm16 against that of the 'golden standard' Hey2 in these experiments, clearly indicated that Prdm16 induced a more robust arterial phenotype in HUVECs than HEY2. Indeed, not only was the overall number of genes significantly induced by Prdm16 higher than for HEY2 (see Example 2), the increase in expression of many genes was far more enhanced upon Prdm16 treatment. Of note, while several arterial specific genes were indeed only induced by either Prdm16 or HEY2, we detected several arterial genes regulated by both TFs. Thus, Prdm16 and Hey2 not only act in a parallel fashion to direct ECs towards an arterial phenotype, they also co-regulate part of the arterial signature. A striking overlap between the genes regulated by HEY2 and Prdm16 was indeed observed: 9 out of the 13 (69%) arterial genes (MFAP5, NPR3, RASGRF2, SEMA3C, SYTL2, FAT1, MAP9, ODAM, CUBN) significantly induced by HEY2 also revealed significantly elevated expression levels upon Prdm16 overexpression.

Prdm16 activates Hey1/2 and their downstream target ephrinB2

The arterial-specific TFs Prdm16 and Hey2 were significantly differentially expressed in our different comparisons of arterial versus venous ECs and therefore are most likely key determinants of arterial identity. Secondly, the capacity of Prdm16 exceeded Hey2's capability of arterialising both HUVECs (see Example 2) and BOECs. Thirdly, several members of the Prdm family - including Prdm16 - genetically interact with the Notch pathway. Fourthly, Prdm16 and HEY2 co- regulate many human arterial-specific genes upon overexpression in HUVECs (see above). Finally, even though these genes were not part of the conserved arterial fingerprint, half of the - few - genes significantly regulated by HEY2 in BOECs are driven in the same direction by Prdm16. Indeed, microarray analysis of control (Cherry), Prdm16 or HEY2-overexpressing BOECs (A/=3 for all) indicated that 13 out of the 27 genes significantly regulated by HEY2, were driven in the same direction by Prdm16 (data not shown). Therefore, we wondered whether Prdm16 and Notch acted through a single cascade to drive endothelial arterialisation. To address this question, we analysed in more depth the effect of Prdm16 overexpression on the Notch pathway via qRT-PCR. Intriguingly, overexpressing Prdm16 in HUVECs resulted in a robust induction of key canonical Notch pathway related members, including DLL4, HEY1 and HEY2 ([fold upregulation Prdm 16 versus Cherry]: 10.6 ± 3.4; P<0.05 for DLL4, 6.8 ± 1 .3; P<0.01 for HEY1; 4.5 ±1 .1 ; PO.05 for HEY2). Conversely, the established venous markers EPHB4, NRP2 and COUP-TFII were not or only slighty suppressed by Prdm16 ([fold upregulation Prdm16 versus Cherry]: 1 .0 ± 0.4; P=NS for EPHB4, 0.4 ± 0.1 ; P<0.05 for NRP2; 0.7 ±0.1 ; P=NS for COUP-TFII). Moreover, the expression of the downstream Notch target gene EFNB2 was also elevated by ectopic Prdm16 expression ([fold upregulation Prdm16 versus Cherry]: 4.0 ±0.6; PO.01 for EFNB2; Figure 41). Similar results were obtained in BOECs ([fold upregulation Prdm16 versus Cherry]: 5.5 ± 3.6; P=NS for DLL4, 5.6 ± 1 .2; P<0.01 for HEY1; 8.5 ±4.0; P<0.05for HEY2, 3.7 ± 1 .0; P<0.05 for EFNB2, 1 .1 ± 0.2; P=NS for EPHB4, 0.5 ± 0.1 ; P<0.05 for NRP2 and1 .1 ± 0.2; P=NS for COUP-TFII; Figure 41). Hence, Prdm16 directly activates Notch signalling in vitro. Since no clear upregulation of PRDM16 was observed upon ectopic Hey2 expression, we hypothesised that Prdm16 lays directly upstream Hey2 during arterial endothelial differentiation.

Prdm16 lies directly upstream of canonical Notch during arterial differentiation

Prdm16 enhances RBPJK 'S transcriptional activity To confirm these findings, we conducted co-transduction experiments with

Prdm16 lentiviral particles and a luciferase-based canonical Notch (RBPJK) reporter virus. Co-transduction with Prdm16 lentivirus increased the luciferase signal compared to co-transduction with a control virus (Cherry). Data were normalised to Renilla luciferase for transduction efficiency ([ratio luciferase/Renilla signal x 10^"3]: 3.0 ± 0.5 and 1 .2 ± 0.2 for Prdm16 versus Cherry, respectively; *P<0.05; N=3 for all; Figure 42).

The arteriaiising effect of Prdm16 in vitro is partially abolished by DAPT

If Prdm16 would act at least in part through canonical Notch signalling, its ability to induce the Notch ligand DLL4, the Notch downstream effectors HEY1/2 and the Notch downstream target EFNB2 should be attenuated by addition of a γ- secretase inhibitor, such as DAPT. Indeed, qRT-PCR demonstrated that the induction of all four of these genes was severely compromised in the presence of DAPT (N-[N-(3,5-Difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester), compared to its control DMSO ([fold upregulation Prdm16 DMSO/Cherry DMSO]: 3.4 ± 0.7 for DLL4; P<0.05, 13.5 ± 5.8; P<0.05 for HEY1, 21 A ± 5.7; P<0.01 for HEY2 and 2.9 ± 0.4; P=NS for EFNB2; [fold upregulation Prdm16 DAPT/Cherry DAPT]: 1 .4 ± 0.3; P=NS for DLL4, 2.7 ± 0.8; P=NS for HEY1, 6.3 ± 2.3; P<0.05 for HEY2, 1 .7 ± 0.5; P=NS for EFNB2). Data are depicted in Figure 43 as expression levels versus Cherry-treated BOECs that were not treated with a compound. Furthermore, we found that the Prdm16-mediated induction of several arterial genes of our conserved arteriovenous signature, including GJA5 and DKK2, was also severely hampered or completely abrogated in the presence of DAPT {Figure 44). Prdm16 knockdown in zebrafish does not lead to aberrant canonical Notch activity

To assess whether Prdm16 also instructs Notch signalling in vivo, we purchased rbpja reporter zebrafish, expressing eGFP under the control of an element containing 12 RBPJK-binding sites (Tg(Tp1-Mmu.Hbb:eGFP)). However, we failed to detect differences between ns Mo and prdm16 Mo-treated Tg(Tp1- Mmu.Hbb:eGFP) embryos, suggesting that canonical notch signalling was not abolished by prdm16 silencing {Figure 45). Thus, similar to what we observed in mice and consistent with our WISH results in zebrafish, prdm16 deficiency does not lead to reduced Notch signalling or diminished expression of key Notch effectors, such as grl, despite the strong induction of Notch pathway genes upon ectopic Prdm16 expression in vitro. Most likely, other Prdm family members compensate for the loss of Prdm16.

Prdm 16 and Notch genetically interact in vivo

DAPT induces aortic abnormalities in prdm16 Mo, but not ns Mo-treated zebrafish

Collectively, the results mentioned in the previous section indicate that ectopic Prdm16 expression drives arterialisation, at least in part by directly acting through canonical Notch signalling. However, despite the obvious (arterial) vascular defects, Prdm16 deficiency in mice nor in zebrafish resulted in impaired Notch signalling or dramatic shifts in the arteriovenous identity of ECs. In order to confirm our in vitro data, we incubated Tg(kdr-eGFP)^s843 zebrafish treated with suboptimal doses of ns or prdm16 Mo with the canonical Notch inhibitor DAPT or its control DMSO and scored these embryos for aortic hypoplasia or other morphological anomalies of the DA at 48 hpf {Figure 46). Intriguingly, while DAPT treatment in ns Mo-injected zebrafish did not cause increased levels of aortic defects, compared to DMSO-treated zebrafish (P=0.64; A/=99-100), it had a clear deleterious effect on the appearance of the DA in prdm16 morphants (P=0.01 ; A/=99-109). Consequently, while injection of a suboptimal dose of prdm16 Mo in zebrafish incubated with DMSO only resulted in marginal arterial defects (ns Mo/DMSO versus prdm16 Mo/DMSO; P<0.05; A/=99-99), the phenotype of zebrafish injected with the same dose of morpholino was severely aggravated in the presence of DAPT (ns Mo/DAPT versus prdm16 Mo/DAPT; P<0.0001 ; ΛΜ 00-109). Representative images of the zebrafish embryos subjected to the different treatments are shown in Figure 46B-E. Grl and prdm16 co-operate during arterial development in zebrafish

As grl - the zebrafish orthologue of Hey2 - is the main determinant of the arterial identity in zebrafish, we next performed co-injection experiments in which we simultaneously knocked-down grl and prdm16. Indeed, co-injecting grl and prdm16 morpholino's significantly worsened the vascular defects seen with suboptimal dose injections of either of these morpholino's alone (P<0.0001 for grl Mo versus grl Mo + prdm16 Mo; A/=100-100; P<0.0001 for prdm16 Mo versus grl Mo + prdm16 Mo; A/=100-100; Figure 47A). Hence, prdm16 and notch/grl signalling genetically interact during the establishment of the arterial vasculature in zebrafish. Representative images of the zebrafish embryos subjected to the different treatments are shown in Figure 47B-E.

Prdm 16-m edia ted arterialisation is partially dependent on its interaction with CtBPs and DNA

The Prdm16-mediated arterial shift requires CtBP and DNA binding

To unravel the precise mechanisms behind the driving forces of Prdm16- mediated arterialization, we conducted parallel overexpression experiments in BOECs with wild-type (WT) Prdm16 and two well-described Prdm16 mutants, one that fails to bind to DNA (Prdm164D/V/A) and another that is incapable of forming a complex with CtBP proteins (Prdm164CfBP). Intriguingly, both mutant Prdm16 isoforms exhibited only a partial inductive effect on HEY1, HEY2 and EFNB2 expression levels compared to WT Prdm16, underscoring the importance of both regions for Prdm16-mediated Notch signaling. Likewise, Prdm164CfBP, but not Prdm164D/V/A, displayed lower inductive activity on DII4 expression levels compared to WT Prdm16 ([fold upregulation versus Cherry]: 5.1 ± 2.0 for WT Prdm16, 1 .7 ± 0.4; for Prdm164CfBP, 4.4 ± 2.0 for Prdm164D/V/¾). Moreover, like we observed with our compound experiments (see 5.3.2.2), both GJA5 ([fold upregulation versus Cherry]: 93 ± 39 for WT Prdm16, 3.6 ± 1 .8 for Prdm164CfBP, 45 ± 20 for Prdm164D/V/¾) and DKK2 ([fold upregulation versus Cherry]: 268 ± 91 for WT Prdm16, 85 ± 29 for Prdm164CfBP, 34 ± 19 for Prdm164D/V/A) elicited only a limited response upon ectopic overexpression of the mutated Prdm16 variants (Figure 48), despite the fact that all Prdm16 isoforms were equally overexpressed (data not shown). Prdm16-mediated Notch signalling requires CtBP and DNA binding

From the above, it was clear that Prdm16 must interact with CtBP and DNA to establish part of its arterial ising effect. We next wondered whether these interactions were also involved in Prdm16-mediated canonical Notch signalling. Analogous to the previous section, we assayed the Prdm16 mutants and their WT equivalent for their capacity to induce canonical Notch signalling, using the RBPJK- luc reporter virus as a read-out. We found that, while Prdm16 was able to strongly activate canonical Notch signalling, the mutants did so to a far lesser extent (Figure 49). Thus, Prdm16-mediated canonical Notch signalling is dependent on its interaction with CtBP and DNA. Prdm16 leads the way to proper arterial SMC coating

Our initial microarray analysis on freshly isolated HUAECs and HUVECs revealed arterial enrichment of two semaphorins, SEMA3C and SEMA3G. Intriguingly, these genes exhibited highly enhanced expression levels, upon Prdm16 overexpression in HUVECs, especially SEMA3C (fold-upregulation versus control virus: 155 ± 25.3; P<0.0001 and 1 .8 ± 0.24; P<0.05 for SEMA3C and SEMA3G, respectively; Figure 50 A). Since semaphorins act as guidance molecules for migrating SMCs, as was already described for Sema3G, Prdm16 might specifically instruct arterial ECs to secrete these molecules to attract multiple layers of SMCs, explaining our in vivo phenotype of deficient arterial SMC coating in Prdm16 knockout embryos (Figure SOB). Hence, Prdm16 defines both the molecular and functional identity of arterial endothelial cells.

3.5. SUMMARY AND CONCLUSIONS

Together, our data indicate for the first time a role for Prdm16 during arterial endothelial differentiation. Indeed, via whole-genome transcriptomic profiling, we identified an arterial-exclusive expression pattern of Prdm16 in both human and murine ECs. Moreover, via meta-analysis of our human and murine arrays, we report here a conserved arteriovenous fingerprint, defining the arterial and venous state of ECs independent of species, anatomical location and haemodynamic parameters. Interestingly, this list contained Hey2 and Coup-TFII - two well- described key determinants of arterial and venous specification - and Prdm16. Ectopic Prdm16 induced many of the arterial-specific genes of both our human and murine aretriovenous fingerprint, clearly indicating that Prdm16 orchestrates the arterial program of both human and murine ECs. More importantly, the arterialising capacity of Prdm16 on the conserved arteriovenous signature exceeded that observed on the species-restricted fingerprints. Hence, Prdm16 is a key determinant of the arterial identity of ECs, by regulating a conserved set of genes in addition to the regulation of a species-specific set of genes. One intriguing finding was that the arterialising capacity of Prdm16 was superior to that of the current 'golden standard' Hey2. Using a variety of assays, we were able to place Prdm16 directly upstream of canonical Notch signalling and thus Hey2, both in vitro and in vivo. Indeed, Prdm16 and Hey2 co-regulate several genes and concommittant inhibition of canonical Notch attenuated in great part the arterialising power of Prdm16 in vitro. Vice versa, reduced Notch signalling in vivo aggravated the vascular defects observed prdm16 Mo-treated zebrafish. Finally, we identified 2 domains of Prdm16 that are indispensable for its full arterialising capacity. Several, non-mutually exclusive mechanisms might underlie the observed effects {Figure SOB). First, Prdm16 might directly activate the expression of the Notch ligand DII4, thereby triggering the release of NICD, which will ultimately lead to the induction of Notch target genes HEY1/2. Notably however, the DNA binding deficient mutant of Prdm16 fails to induce the expression levels of HEY1/2 to the same extent as its WT variant, despite comparable DLL4 expression levels between these two variants. Secondly, Prdm16 might bind CtBP proteins, thereby converting the RBPJK-complex from an inhibitory to an activating state, similar to the function of NICD. Moreover, both Prdm16-CtBP and NICD could be necessary for full activation of the RBPJK complex. Finally, Prdm16 could merely sequester CtBP, thereby releasing it from the RBPJK-complex and hence augmenting endogenous Hey1/2 levels. However, if this would be the case, inhibiting CtBP1/2 via small interfering RNAs would result in elevated levels of Hey1/2 in BOECs, a finding we did not observe. Moreover, adding DAPT to Prdm16-treated BOECs severely attenuated or nearly abolished the induction of Hey1/2, indicating that Prdm16 does not just act as a permissive factor by sequestering CtBPs, but actively drives Hey1/2 expression. Finally, Prdm16 is one of the few Prdm family members with intrinsic methyltransferase activity. Hence, once directed to the promoter of its target genes by transcriptional co-regulators, Prdm16 might induce methylation and thus alter the transcriptional activity of the loci poised with binding sites for the Prdm16 complex. Moreover, Endo et al. reported that the binary switches instructed by hamlet occur in a methylation-dependent fashion: hamlet binds to RBPJK binding sites and trimethylates H3K27, while preventing trimethylation of H3K4. Hamlet also increases histone H3 levels, all hallmarks of a dense chromatin structure, not compatible with active transcription. Alternatively, as described for multiple Prdm proteins, histone acetyltransferases {e.g., HDACs) might be at play in the Prdm16-mediated arterial isation of ECs. On a more speculative basis, Prdm16 might have non-histone post-transcriptional targets. Nevertheless, further studies are warranted to pinpoint the exact mechanism via which Prdm16 induces canonical Notch activity. Altogether, Prdm16 is a main orchestrator of the arterial molecular and functional fingerprint. Thus, Prdm16 could be used as a novel therapeutic target in vascular therapies. Furthermore, the generation of an endothelial-specific Prdm16 knockout mouse could lead to a better understanding of its role and importance during early arteriovenous specification. Finally, it would be of great interest to assess the role of Prdm16 during the development of atherosclerotic lesions or potentially other arterial defects. Drawing Description

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will become more fully understood from the detailed description given herein below and the accompanying drawings which are given by way of illustration only, and thus are not limitative of the present invention, and wherein:

FIGURE 1 provides a chart which presents a schematic summary of an embodiment of the invention' and of the examples. Example 1 relates to studies on the establishment of microvascular EC gene and reference signatures and on the use of TF (combinations) to induce a heart EC-specific target cell. Example 2 relates to studies on the establishment of macrovascular EC gene and reference signatures and on the use of TF (combination)s to induce an arterial EC-specific target cell. Example 3 relates to studies validating one of the 8 arterial TFs, i.e., Prdm16, discovered in the studies related to Example 2. FIGURE 2 displays the lentiviral maps for overexpression of Tcf15. EBF3. PPARy. and WT1. A-E. Schematic circular plasmid maps of the Cherry control vector (panel A; without insertion of a gene in the multiple cloning site or MCS) or constructs containing the open reading frame of murine Tcf15 (mTcf15; panel B), human EBF3 (hEBF3; panel C), human PPARy {hPPARy panel D) and human Wilms tumour 1 (hWT1; panel E) inserted in the MCS. Expression of the Cherry reporter gene is driven by the phosphoglycerate kinase (PGK) promoter while expression of the inserted transcription factor gene is driven by the cytomegalovirus (CMV) promoter.

FIGURE 3 displays the microarray results and validation. A. Representative FACS plots of heart homogenates from Tie2-GFP mice showing gate (G1 ) setting for sorting of GFP⁺ endothelial cells (ECs) and G1 analysis revealing -99% purity for EC marker CD31 and -96% for microvascular EC marker CD36, and negligible contamination (<1 %) with hematopoietic cells (expressing CD45). B. Microarray heat map and principal component analysis of heart EC, liver EC and brain EC samples. For the heat map dark grey represents a high degree of similarity in gene expression profile between the matching samples on the two axes of the plot; light grey represents a low degree of similarity. C. Venn diagramme representing the genes differentially expressed in the three vascular beds studied. D. Representative plot of microarray probe intensities for EC markers and markers of potential contaminant cell types for a single EC preparation from brain, heart and liver. E. The diagram shows the proportions of microarray probe intensities for known vascular bed-specific markers for each of the three vascular beds. Data represent mean ± s.e.m.; ^*P<0.05 versus corresponding specific organ and a minimum 4-fold difference and Log₂ probe intensity >6.

FIGURE 4. provides a graphic display on the sic/nature cross-validation versus the non-EC fraction of murine organs. A. Validation of a selected subset of the murine liver EC signature: transcription factors and genes most differentially expressed according to the microarray vs. heart ECs and liver ECs for a total number of 30 genes. qRT-PCR data represent fold enrichment in the EC fraction versus the non- EC fraction sorted from mouse liver and expressed as mean ± s.e.m.; A/=3-5; ^*P<0.05. B. Validation of a selected subset of the murine liver EC signature: transcription factors and genes most differentially expressed according to the microarray versus heart ECs and brain ECs for a total number of 30 genes. qRT- PCR data represent fold enrichment in the EC fraction versus the non-EC fraction sorted from mouse brain and expressed as mean ± s.e.m.; A/=3-5; ^*P<0.05; #P<0.1 . C. Validation of the murine heart EC signature. qRT-PCR data represent fold enrichment in the EC fraction versus the non-EC fraction (depicted with white bars) and in the EC fraction versus the freshly isolated cardiomyocytes (depicted as black bars). Data are expressed as mean ± s.e.m.; A/=3-5; ^*P<0.05.

FIGURE 5. provides a graphic display of Log₂ probe-set intensity of validated genes from the heart EC fingerprint in murine heart, liver and brain ECs. N=5; ^*P<0.05 versus heart ECs.

FIGURE 6 provides graphics related to the expanded expression analysis of the heart EC signature in additional tissues revealing that the signature was very similar to that of other metabically active tissues (i.e., brown adipose tissue, white adipose tisse and skeletal muscle; panel A), but very different from other tissues (i.e., lung, pancreas and kidney; panel B). Data are presented relative to expression in heart ECs (=1 ). (A/=3-4). *P<0.05 versus heart ECs and a minimum 4-fold difference.

FIGURE 7 provides a graphic display of the mRNA expression determined by qRT- PCR of genes of the heart EC fingerprint in human (h) heart, brain or liver ECs relative to human heart ECs, revealing that part of the signature is also enriched in human heart ECs versus brain and liver ECs. ND: not detectable. Data represent mean ± s.e.m. (n=3-10; *P<0.05; ^#P<0.1 versus heart ECs)

FIGURE 8 displays the heart EC signature analysis in cultured ECs from human biopsies. A. Schematic representation of isolation and culture of ECs from human biopsies with subsequent sorting set-up to have a pure Tie2⁺:Podoplanin^" (non- lymphatic) population. Right: heart ECs in culture uniformly stained for the EC marker VE-Cadherin, DAPI was used as nuclear counterstain. B. qRT-PCR analysis of the heart EC signature in cultured ECs relative to freshly isolated heart ECs. Data represent mean ± s.e.m.; A/=3-10; *P<0.05; #P<0.1 . FIGURE 9 provides pictures that demonstrate the validation of the heart EC- specific TFs at protein level. A. Western blots show Meox2 and Tcf15 in sorted cardiac EC and their absence in non-EC from the heart and freshly isolated cardiomyocytes. B, Murine heart, brain and liver tissue cross-sections stained for Meox2, and Tcf15 and co-stained for an EC marker (BS-I lectin for brain and heart, CD105 for liver; DAPI was used as nuclear counterstain). Arrowheads indicate nuclei positively stained for the corresponding TF. Scale bars correspond to 20 μηη.

FIGURE 10 provides graphics on the effect of TF overexpression on the heart EC signature in cultured human heart ECs. All diagrams in (A-D) represent mRNA expression determined by qRT-PCR for 29 genes of the heart EC fingerprint in cultured human heart ECs overexpressing certain transcription factor (combination)s relative to a 'Cherry' reporter control. (n=3-9 independent experiments; in each experiment all the conditions were tested and analyzed in parallel. *P<0.05 or ^#P<0.1 versus Cherry shown under the bars; or versus another condition as indicated above the bars). Primers in the 3'UTR region were used to distinguish endogenous expression from that caused by lentiviral overexpression. Data represent means ± s.e.m. Two of the 31 signature genes were not included, i.e., Klra9 and Klra10, as there is no human equivalent. A, Overexpression of MEOX2, Tcf15 or MEOX2/Tcf15versus Cherry. B, Overexpression of MEOX2/Tcf15, EBF3 or MEOX2/Tcf15/EBF3 versus Cherry. C, Overexpression of MEOX2/Tcf15,PPARG or MEOX2/Tcf15/PPARGversus Cherry (in all experiments where PPARG was overexpressed, an agonist, rosiglitazone, was added at 10 μΜ). D, Overexpression of MEOX2/Tcf15/PPARG , WT1 or MEOX2/Tcf15/PPARG/WT1 versus Cherry. E, Western blots for ZDHHC2, RBP7 and TIMP4 in cultured human heart ECs transduced with Cherry or MEOX2/Tcf15/PPARG. a-TUBULIN was used as loading control.

FIGURE 11 provides photographs on the fatty acid uptake in cultured heart ECs overexpressing TFs. A, Fluorescence micrographs of cultured human heart ECs transduced with Cherry, WT1 (W), PPARG (P), MEOX2/Tcf15 (MT), MEOX2/Tcf15/PPARG (MTP) or MEOX2/Tcf15/PPARG/WT1 (MTPW) exposed to BODIPY-palmitic acid (PA; green), counterstained with DAPI for nuclei (blue). An excess (10X) of non-labeled PA was added in a competition assay. Corresponding quantification of the number of lipid-laden vesicles per cell is shown as mean ± s.e.m. (n=4-5; ^#P=0.059, ^*P<0.05, ^**P<0.01 versus Cherry shown inside the bars; or versus another condition as indicated above the bars). B, Fluorescence micrographs of cultured human heart ECs transduced with GFP, W, P, ΜΤ,ΜΤΡ or MTPW exposed to Dil-VLDL (red), counterstained with DAPI for nuclei (blue). An excess (10X) of non-labeled VLDL was added in a competition assay. Corresponding quantification of the number of lipid-laden vesicles per cell is shown as mean ± s.e.m. (n=3; ^*P<0.05, ^**P<0.01 versus GFP shown inside the bars; or versus another condition as indicated above the bars). Scale bars: 20μηη. NS: not significant. FIGURE 12 provides a graphic display on results of the programming of BOECs into heart ECs by TF overexpression. Bar diagram reporting expression levels of heart EC-specific genes following overexpression of MEOX2+TCF15+PPARy+WT1 with addition of rosiglitazone (10 μΜ; Ros.) in blood outgrowth endothelial cells (BOECs). Data represent fold change ± s.e.m. relative to BOECs transduced with a negative control (Cherry virus); A/=4; ^*P<0.05; #P<0.1 .

FIGURE 13 provides a graphic display on the cardiac EC expression pattern of Meox2/Tcf15 deficient mice. Bar diagrams reporting mRNA expression determined by qRT-PCR of genes of the heart EC fingerprint in heart ECs from of adult Meox2^+/~, (white bars), Tcf15^+/~ (grey bars) or Meox2^+/~:Tcf15^+/~ (black bars) mice, relative to Wild-type (=1 , indicated by dashed line) littermates. Data represent mean ± s.e.m.; n=3-8; ^*P<0.05 versus WT. FIGURE 14 provides a graphic display on the analysis of genes outside the heart EC fingerprint in Meox2^+/~:Tcf15^+/~ mice. A, mRNA expression determined by qRT- PCR of general EC markersin heart ECs sorted from Meox2^+/~, Tcf15^+/~, or Meox2^+/~ :Tcf15^+/~ littermates relative to Wild-type heart ECs (n=6-8) revealing no significant differences in expression. B, mRNA expression determined by qRT-PCR of liver sinusoidal (LS)EC markers in liver ECs sorted from Meox2^+/~, Tcf15^+/~, or Meox2^+/~ :Tcf15^+/~ littermates relative to Wild-type heart ECs (n=4-8) revealing no significant differences in expression. Data represent mean ± s.e.m.

FIGURE 15 provides a graphic display of the expression levels of fatty acid and glucose transporter genes in heart, brain and liver ECs. The graph shows Log₂ probe set intensity in heart, brain and liver ECs. Fabp4 and Cd36, two well-known fatty acid transporter molecules, are significantly higher in heart ECs compared to brain and liver. (n=5). ^*P<0.05.

FIGURE 16 provides a graphic display of the expression levels of fatty acid and glucose transporters in Meox2^+/~, Tcf15^+/~ mice_^ . qRT-PCR analysis of genes involved in lipid uptake commonly expressed between heart and liver ECs, show heart-specific downregulation in Meox2^+/~:Tcf15^+/~ ECs. (A/=5-8). ^*P<0.05. B. qRT- PCR analysis of glucose transporters on freshly isolated ECs from

Meox2^+/~:Tcf15^+/~ mice show increased expression of Glutl and the FA transporter Fatp3, a downstream target of VEGFB in Meox2^+/~:Tcf15^+/~ mice. (A/=3) ^*P<0.05. C. Representative FACS plot showing downregulation of Cd36 protein expression in heart ECs from Meox2⁺ Tcf15^{+ "} mice (darg grey line) compared to Meox2^+/+ :Tcf15^+/+ (light grey line). The experiment was repeated three times.

FIGURE 17 provides graphic displays showing the lack of effect of Meox2/Tcf15 heterozygous deficiency on FA uptake in cardiomyocytes. A, mRNA expression determined by qRT-PCR of genes encoding FA or glucose transporters in cardiomyocytes isolated from Meox2^+/~:Tcf15^+/~ littermates relative to Wild-type cardiomyocytes (n=4) revealing no significant differences in expression. B, Representative fluorescence micrographs of cultured cardiomyocytes from Wild- type or Meox2^+/~ :Tcf15^+/~ littermates exposed to BODIPY-palmitic acid (PA; in green), counterstained with DAPI for nuclei (in blue), revealing no differences in PA uptake between genotypes. An excess (10X) of non-labelled PA was used as competition assay. Quantitative data represent mean ± s.e.m.

FIGURE 18 provides a graphic display of the in vivo uptake of fatty acids and glucose in Meox2^+/~:Tcf15^+/~ mice. A. Graph showing the uptake of radio-labelled ¹⁴C-oleic acid in Meox2^+/~:Tcf15^+/~ and Meox2^+/+ :Tcf15^+/+ littermates showing a reduction in uptake upon Meox2/Tcf15 heterozygous deficiency. (A/=7-8) ^*P<0.05. B. Time-lapse quantification of of [¹⁸F]FDG uptake into Meox2^+/+ :Tcf15^+/+ and Meox2^+/\ cf/5^+Ahearts showing a significant increase in glucose uptake upon Meox2/Tcf15 heterozygous deficiency. C. Representative μΡΕΤ image of Meox2^+/+:Tcf15^+/+ an6 Meox2^+/-;Tcf15 ^/- hearts (arrow) is shown. (A/=1 1 -12) ^*P<0.05 FIGURE 19 provides graphic displays related to the heart phenotype and function of young adult Meox2^+/~:Tcf15^+/~ mice. All histological data are relative to Meox2^+/+:Tcf15^+/+ or Meox2^+/~:Tcf15^+/~ hearts from 8-16 weeks-old male mice. A. Representative Oil Red-O staining and relative quantification (A/=4; ^*P<0.05), scale bar 20 μιτι. B. Representative Sirius red staining showing the absence of fibrosis. (A/=4), scale bar 200 μιτι. C. Representative merged pictures of laminin staining of extracellular matrix (in green) and lectin staining of blood vessels (in red) and relative quantification of the cardiomyocyte cross-sectional area revealing no significant differences (A/=4), scale bar 20μηη. D. Echocardiographic phenotype of Meox2^+/+ :Tcf15^+/+ or Meox2^+/~:Tcf15^+/~ male mice of 12-16 weeks old expressed as % ejection fraction (%EF) and % fractional shortening (%FS) (A/=9) revealing no functional impairment.

FIGURE 20 provides graphic displays related to the heart phenotype and funtion of aged adult Meox2^+/~:Tcf15^+/~ mice .All histological data are relative to Meox2^+/+:Tcf15^+/+ or Meox2^+/~: Tcf 15^+/~ hearts from 1 1 months-old male mice. A. Oil Red-O staining and relative quantification revealing a significant reduction in fat accumulation in cardiomyocytes upon Meox2/Tcf15 heterozygous deficiency (A/=3; ^*P<0.05), scale bar 20 μιτι. B. Sirius red staining and relative quantification revealing a significant increase in fibrosis upon Meox2/Tcf 15 heterozygous deficiency (A/=3; *P<0.05), scale bar 200 μηη. C. Merged pictures of laminin staining of extracellular matrix (in green) and lectin staining of blood vessels (in red) and relative quantification the cardiomyocyte cross-sectional area revealing no significant differences (A/=3;), scale bar 20 μιτι. D. Echocardiographic phenotype of Meox2^+/+ :Tcf15^+/+ or Meox2^+/~:Tcf15^+/~ male mice (1 1 months old) expressed as % ejection fraction (%EF) and % fractional shortening (%FS) revealing functional impairment upon Meox2/Tcf15 heterozygous deficiency. (A/=8-9; *P<0.05).

FIGURE 21 displays a schematic summary of the findings in Example 1. In Wild- type heart endothelial cells (ECs), Meox2 and Tcf15 together regulate the balance between fatty acid (FA) and glucose uptake by orchestrating the expression of multiple signature genes, encoding membranous or intracellular regulators of FA transport, supporting the preferential use of FAs as a source of energy production in cardiomyocytes. Combined Meox2/Tcf15 heterozygous deficiency results in downregulation of these genes and an increase in Glutl, together causing a shift to higher glucose and lower FA delivery to cardiomyocytes. This combined deficiency in the long run causes fibrosis and systolic dysfunction. Lpl: lipoprotein lipase; FAPB: fatty acid binding protein; HSPG: heparan sulphate proteoglycans; Glutl : glucose transporter 1 ; Alb: albumin; VLDL: very low density lipoprotein; CM: chylomicron.

FIGURE 22 displays a graphical display of the purity of EC preparations from human umbilical cord. FACS analysis of freshly MACS column-sorted HUAEC (A) or HUVEC (B) revealing only minimal (< 1 % in both cell populations) contamination with CD45⁺ blood or inflammatory cells and high purity (> 97% in both cell populations) for CD31 ⁺ and CD34⁺ endothelial cells.

FIGURE 23 displays the lentiviral maps for overexpression of the arterial TFs. A-G. Schematic circular plasmid maps of the Cherry control vector (panel A; without insertion of a gene in the multiple cloning site or MCS) or constructs containing the open reading frame of murine Aff3 (mAff3; panel B), human MSX1 (hMSX1; panel C), human EMX2 (hEMX2; panel D), human NKX2-3 (hNKX2-3; panel E), human 70X2 (hTOX2; panel F), or murine Prdm16 (mPrdm16; panel G) inserted in the MCS of the pRRL2 backbone. Expression of the Cherry reporter gene is driven by the phosphoglycerate kinase (PGK) promoter while expression of the inserted transcription factor gene is driven by the cytomegalovirus (CMV) promoter. H. Schematic circular plasmid map of the construct containing the open reading frame of human SOX17 (hSOXU) in the pLVX backbone. Expression of the transcription factor is driven by the CMV promoter.

FIGURE 24 provides a graphic display of how the cell culture process assimilates arterial and venous endothelial cells. Hierarchical clustering analysis of all 38 endothelial cell (EC) samples for all 102 probes (corresponding to 76 annotated genes of the arteriovenous fresh profile) reveals that freshly isolated cells (on the left) cluster according to their venous or arterial origin, while for cultured cell types (on the right) the clustering does not classify the sample groups correctly, suggesting that the differences in expression profile have been largely erased. Arterial cell types are represented by a red colour while venous cell types are represented by a blue colour in the colour bar below. The colour code for expression levels is displayed on top. HUAEC: human umbilical artery EC; HUVEC: human umbilical vein EC; HPVEC: human pulmonary vein EC; HHAEC: human hepatic artery EC; HHVEC: human hepatic vein EC; HIVEC: human iliac vein EC; HCAEC: human coronary artery EC; HAEC: human aortic EC; HPAEC: human pulmonary artery EC; H IAEC: human iliac artery EC; NA: not assigned. FIGURE 25 provides photographs and graphics on the genome-wide analysis of freshly isolated HUVECs and HUAECs. A. Upper. Pie diagrams showing the number of annotated arterial (red) and venous (blue) genes emerging as differentially expressed between freshly isolated HUAEC and HUVEC from the differential gene statistical analysis (left) or classification analysis (right). Lower. Diagram showing the total amount of annotated genes and corresponding percentages (black lettering; arterial-enriched genes in red, venous-enriched genes in blue lettering) uniquely identified by the differential gene analysis (pink), from the classification analysis (yellow) or emerging from both analyses (orange; which included 4 genes, i.e., HEY2, NOTCH4, SOX17 and KDR, previously associated with arterial specification). Other displayed names correspond to the most differentially expressed genes in each part of the diagram. B,C. Bar diagrams showing functional terms (B) and pathways (C) significantly enriched in the 76 differential gene set. Numbers of associated genes per function/pathway are mentioned on top of the bars. Yellow bars represent subcategories of the main functional term 'cardiovascular diseases'. D. Bar diagram showing microarray probe set intensities (± s.e.m.; N=4) for the 8 arterial (A) and 1 venous (V) transcription factors identified among the arteriovenous fresh profile, revealing their differential expression in freshly isolated HUAEC (red) and HUVEC (blue). *: P<0.05 versus arterial probe set intensity. E-H. Immunofluorescence stainings on paraffin cross-sections of human umbilical cord (umbilical vein is shown in E,G, umbilical artery in F-H) for a-smooth muscle cell actin (aSMA; in green) and RASGRF2 (E-F; in red) or NR3C2 (G-H; in red). DAPI (blue) was used as nuclear counterstain. White arrowheads indicate positive staining in the endothelial lining. Scale bars: 20 μηη.

FIGURE 26 displays how the culturing process rapidly erases differential arteriovenous gene expression. A. Hierarchical clustering analysis for freshly isolated or cultured HUAEC (red) or HUVEC (blue) revealing that for fresh samples replicates of each cell type tightly cluster together and both clusters are nicely separated. For cultured samples, incorrect clustering occurred. B. Diagram representing the probe set intensity difference between HUVEC and HUAEC for each gene of the arteriovenous fresh profile (showing arterial genes (A) on the left and venous genes (V) on the right) for cultured (filled diamond) or freshly isolated (open triangles) cells, revealing only minor differences in cultured cells. C. Diagram representing the expression (± s.e.m.; N=4) determined by qRT-PCR for arterial transcription factors in freshly isolated cells (open triangles) or cultured cells (filled diamonds) relative to their expression in freshly isolated cells. *: P<0.05 versus freshly isolated cells. D-E. Diagrams showing probe set intensity (± s.e.m.; N=4) for arterial (A) and venous (V) genes of the arteriovenous fresh profile for cultured (filled diamonds) or freshly isolated (open triangles) HUAEC (D; symbols in red) or HUVEC (E; symbols in blue). F. Diagram showing that the average probe set intensity for all arterial genes (dashed lines) or venous genes (full lines) in freshly isolated HUAEC (left; red) or HUVEC (right; blue) 'bleach' to a default average expression level (grey dashed line) upon culturing (middle; pink). *: P<0.05 versus fresh; NS: not significant versus fresh (A/=4-5). FIGURE 27 is a graphic display of the kinetics of the culture-induced assimilation process. Diagrams on the left represent expression levels for all arterial genes (A/=64) of the arteriovenous fresh profile in HUAEC. Panel A shows expression levels after 24 hours of culture (filled red diamonds) relative to those in freshly isolated HUAEC (open black triangles). Panel C represents expression levels after 48 hours of culture (open red circles) relative to those after 24 hours of culture (filled black diamonds). Panel E represents expression levels after 6 days of culture (filled red squares) relative to those after 48 hours of culture (open black circles). Diagrams on the right show expression levels for all venous (A/=12) genes of the arteriovenous fresh profile in HUVEC. Panel B represents expression levels after 24 hours of culture (filled blue diamonds) relative to those in freshly isolated HUVEC (open black triangles). Panel D represents expression levels after 48 hours of culture (open blue circles) relative to those after 24 hours of culture (filled black diamonds). Panel F represents expression levels after 6 days of culture (filled blue squares) relative to those after 48 hours of culture (open black circles). All diagrams together reveal that loss of expression occurs within 24 hours after culturing for the majority of arterial and venous genes in HUAEC and HUVEC, respectively, an effect that further increases rapidly independent of cell passaging. The percentage of genes with expression lower than the reference (corresponding to the expression levels in freshly isolated cells indicated by a dashed line) are mentioned on the right side of each diagram.

FIGURE 28 provides a display of how reactivation of Notch signalling only partially restores arterial gene expression in HUAEC-C. A-B. Schematic diagrams showing 4 numbered consecutive steps during normal canonical Notch signalling/^'n vivo (A) or under several experimental conditions in vitro (B), i.e. in the presence of DAPT (left), siRNA against RBPJ (middle) or Delta-like (DLL)4-Fc anchored to the cell culture dish (right). C. Diagram representing expression (± s.e.m.; N=4) of several Notch pathway members and downstream genes in cultured HUAEC treated with DAPT (black) or DMSO (white) relative to DMSO. ^*: P<0.05 versus DMSO. D. Diagram representing expression (± s.e.m.; N=4) of the same gene panel in cultured HUAEC treated with siRBPJ (black) or non-silencing siRNA (siNS; white) relative to siNS. ^*: P<0.05 versus siNS. E. Diagram representing expression (± s.e.m.; N=4) of the same gene panel in cultured HUAEC treated with bovine serum albumin (BSA)/DAPT (light grey), BSA/DMSO (white), DLL4-Fc/DMSO (black) or DLL4-Fc/DAPT (dark grey) relative to BSA/DMSO. ^*: P<0.05 versus BSA DMSO or DLL-4-Fc/DMSO. F. Diagram representing expression of all arterial (left) and venous (right) genes of the arteriovenous fresh profile in cultured HUAEC exposed to DLL4-Fc relative to BSA (red dotted line indicates expression in BSA-treated cells). Genes with no change in expression are in white, those that are upregulated in black and those that are downregulated in grey. ^*: P<0.05 versus BSA.

FIGURE 29 provides a graphic display of how combined overexpression of eight transcription factors robustly induces the arteriovenous fresh profile in HUVEC-C.

A. Heat map analysis of the arteriovenous fresh profile upon transduction of cultured HUVEC with cherry control virus, lentivirus expressing one transcription factor, or a combination of all 8 transcription factor-expressing lentiviruses ('ALL'). The farther the condition is removed from the cherry control condition (or the greener the colour in the cherry column for a certain condition), the better the induction of the arteriovenous fresh profile. Note that the 'golden standard' HEY2 certainly is not the best performing individual transcription factor. B. Hierarchical clustering analysis of the arteriovenous fresh profile upon transduction of cultured HUVEC with cherry control virus, lentivirus expressing one transcription factor, or a combination of all 8 transcription factor-expressing lentiviruses ('ALL'). Red colour in the bar on the left represents an arterial gene, while the blue colour represents a venous gene. While almost all the genes are expressed at low levels (in green) in the cherry control condition, each of the individual transcription factors, except for Aff3, upregulated a subset of (mostly arterial) genes in a largely complementary but in some cases overlapping fashion. As expected from the complementarity, the 8 transcription factors together induce the majority of the (arterial) genes. C,D. Diagrams representing expression for classical arterial (C, left) or venous (C, right) or general endothelial (D) markers in cultured HUVEC transduced with cherry control virus (white) or all 8 transcription factor-expressing lentiviruses ('ALL TF') relative to cherry control. ^*: P<0.05 versus cherry control. E. Upper. Pie diagram representing the proportion of arterial genes with expression levels < 2% (light pink) or > 2% (light red) in cultured HUVEC transduced with all 8 transcription factor- expressing lentiviruses ('ALL TF') relative to those in freshly isolated HUAEC. Lower. Pie diagram representing, within the subset of arterial genes with > 2 % values, the proportion of genes with mid-range (< 25%; dark pink) or high-range (> 25%; dark red) expression. F. Western blot showing the validation of some genes being upregulated by trancription factor overexpression. a-TUBULIN was used as loading control. 'D' represents proportional values for the density of the protein band, kDA represents expected protein size.

FIGURE 30 provides a graphic representation of how TFs can induce long-term expression of the arterial gene profile. Diagram representing the expression of selected genes of the fingerprint upon long-term overexpression of transcription factors Prdm16 and Hey2, respectively (data are presented as mean ± s.e.m.; N=3). Although some variation across time is present, transduced cells acquire and maintain TF-dependent gene expression for up to 1 month.

FIGURE 31 provides a graphic representation of how individual transcription factors interact in a complex network to regulate the arteriovenous fresh profile. A,B- Pie diagrams representing the proportion of arterial (A) or venous (B) genes of the arteriovenous fresh profile regulated by 0 (blue), 1 (purple), 2 (orange), 3 (green) or more than 3 (beige) transcription factors (TF). Corresponding absolute numbers are listed in the table on the right of the panel. C. To summarise all information on the effect of transcription factor overexpression on the arteriovenous fresh profile, we composed an interaction network of the 8 transcription factors and the arteriovenous fresh profile. Since Aff3 did not regulate any gene of the signature, nor was it regulated by the other transcription factors, it is not included in the network. Each transcription factor hub (represented by rounded boxes) is drawn in a different colour and interactions originating from each hub are shown by arrows (induction) or vertical lines (inhibition) in the corresponding colour. Genes are assigned to the hubs according to the transcription factor by which they were most strongly regulated. For those genes exclusively regulated by 1 transcription factor the box of that gene is drawn in dark colour, whereas for genes that are regulated by more than one transcription factor the box is drawn in light colour. Transcription factors are in oval boxes, arterial genes are in rectangular boxes with full line and venous genes in rectangular boxes with dashed lines.

FIGURE 32 displays the effect of the 8 TFs on HUVECs in an in vivo Matrigel plug implantation assay system. A,B- Fluorescent micrographs of Matrigel implants containing HUVECs transduced with Cherry reporter gene or HUVECs transduced with the combination of the 8 arterial TFs and Cherry. (A) Ex vivo explant and (B) cryosections showing more elaborate Cherry⁺ vascular structures (white arrows) upon overexpression of the 8 arterial TFs. C. Cryosections stained with smooth muscle a-actin, and the corresponding quantification (right panel), showing a higher % of smooth muscle-coated Cherry⁺ vessels in Matrigels with the 8 TF-transduced HUVECs. D. Human-specific CD31 (red) and smooth muscle a-actin (green) staining, demonstrating the presence of vessels of human origin in the implant containing the 8 TF-transduced HUVECs. The white arrow indicates an autofluorescent erythrocyte inside a functional human vessel. E. Smooth muscle a- actin (green) staining and the corresponding quantification (right panel) showing a larger smooth muscle-covered area in the Matrigels containing HUVECs transduced with the 8 arterial TFs. F. Sirius red staining, analysed under polarised light, and its corresponding quantification, show more deposition of organised collagen in Matrigels with the 8 TFs transduced HUVECs (right panel). Dashed lines delineate the borders of the Matrigel. All quantitative data are shown as mean ± s.e.m. (N=5 per group); ^*P<0.05. Scale bars: 100 μηη in Α,Β^ Ο μηη in C,D, and 200 μηη in E,F.

FIGURE 33 displays the conserved arteriovenous fingerprint between human and murine EC samples. AECA/EC log-ratio's of probe set intensities for the conserved arteriovenous signature genes. Black bars represent log-ratio's of our human microarray, white bars indicate log-ratio's of the murine array.

FIGURE 34 displays microarray results for the conserved arteriovenous sic/nature in Prdm16 overexpressing BOECs, HEY2 overexpressing BOECs or siNR2F2- treated HUVECs. A. Prdm16 overexpression in BOECs resulted in augmented levels of several arterial-specific genes (dark grey), while the expression of the bulk of venous genes (light grey) was suppressed. A Cherry containing vector was used as a control. B. Overexpression of HEY2 in BOECs only resulted in a modest modulation of the conserved arteriovenous fingerprint in BOECs. C. siNR2F2 treatment of HUVECs resulted in a strong induction of multiple arterial genes (dark grey) compared to a non-silencing siRNA (siNS). However, only half of the venous- specific genes (light grey) were inhibited by siNR2F2 treatment. Figure 35 displays the validation of the arterial-exclusive expression pattern of Prdm16 across species at different developmental stapes. A. qRT-PCR demonstrating the arterial enrichment for HEY2 and PRDM16 in HUAECs versus HUVECs. Vice versa, COUP-TFII was predominantly expressed on HUVECs (A/=4- 4). B. qRT-PCR indicating the arterial enrichment for Hey2 and Prdm16 in mAECs versus mVECs. Vice versa, Coup-TFII was predominantly expressed on mVECs (A/=4-4). C,D. Immunofluorescence staining for PRDM16 (in grey) on cross- sections of human umbilical cords confirming its presence on HUAECs (white asterisks in C), while absent on HUVECs (white arrows in D). Smooth muscle cells (SMCs) are stained with a-SMC-actin in white. E, E', E". X-gal staining (dark grey) in E10.5 Prdm16^+/~ 'knock-in' embryos demonstrating its expression on the dorsal aorta (DA; E), while the common cardinal vein (CCV) is devoid of Prdm16 expression (E"). F, F', F". X-gal staining in E10.5 Prdm16^+/~ 'knock-in' mice demonstrating its expression in the DA, but not the cardinal vein (F). Likewise, Prdm16 is expressed on the vitelline artery, but not its venous counterpart (F"). G,H. Prdm16 staining (grey) on E14.5 embryos indicating its presence on intercostal arteries and the carotid artery (white asterisks in G,H). Conversely, Prdm16 was not detected on ECs from intercostal veins or the carotid vein (white arrows in G,H). Slides were counterstained with a-SMC-actin in white. I,J. Prdm16 immunofluorescence staining on an E17.5 embryo visualising its presence on coronary AECs (white asterisks in J), but not on coronary VECs (white arrows in /). K,L. X-gal staining (dark grey) on adult tissues: Prdm16 is expressed on ECs from the aorta (black asterisks in K), but not the vena cava (black arrows in L). M. X-gal staining on Prdm16^+/+ aorta demonstrating the specificity of the staining as X-gal could not be detected on aortic ECs of a Prdm16^+/+ mouse (black asterisks). Scale bars: 50 μΜ in E, E', E", F, F', F",G, I and J and 100 μΜ in C, D, H, K, L and M. ^*P<0.05 versus VECs.

FIGURE 36 displays the Prdm16 expression profile and knockdown phenotype in Tg(kdr-eGFP)⁵⁸⁴³ zebrafish. A. Expression levels of prdm16 in Tg(kdr-eGFP)⁸⁸⁴³ zebrafish at different developmental stages for vascular (black bars) and nonvascular (white bars) fractions. Vascular expression of prdm16 peaks at 24 hpf. Data are expressed as fold-expression versus universal zebrafish RNA (uniZ). B. Prdm16 whole-mount in situ hybridisation (WISH) shows prdm16 expression (dark grey) in the dorsal aorta (DA; arrowheads), but not in the posterior cardinal vein (PCV). C. Panel C shows a cross-section of an embryo stained for prdm16 RNA by WISH. Inset in C zooms in on the region of the DA and PCV, revealing prmd16 expression in the DA, but not in the PCV. D,E. Confocal images of Tg(kdr- eGFPf⁸⁴³ zebrafish at 48 hpf, injected either with non-silencing (D) morpholino (ns Mo) or prdm16 Mo (E). Prdm16 morphants show a clear vascular phenotype, characterised by impaired formation of the dorsal longitudinal anastomotic vessel (DLAV) and improperly formed or absent intersomitic vessels (ISVs, indicated by arrowheads). D',E'. Higher magnification of the images in D,E. F. Quantification of the ISV developmental defects in ns Mo and prdm16 Mo-treated embryos, expressed as the percentage of embryos showing either normal (white), abnormal (grey) or abolished (black) ISV formation. Numbers of embryos analysed are mentioned in the bar graphs. Scale bars represent: 80 μιτι in B, 40 μιτι in C, 100 μηη in D,E and 200 μηη in D',E'.^****P<0.0001 versus ns Mo.

FIGURE 37 displays Prdm16^{- "} mouse embryos with haemorrhages and signs of reduced SMC coverage. A. Schematic drawing of an E14.5 embryo, highlighting the region of interest in these embryos. Inset corresponds to region shown in {B,C). B,C. Haematoxylin&Eosin (H&E) staining of Prdm16^+/+ {A) or Prdm16^' (B) embryonic day (E)14.5 embryos showing severe bleedings (dotted area in (C) in the Prdm16^' mice. D,E. a-smooth muscle cell-actin (aSMA) staining (white) for Prdm16^+/+ (D) and Prdm16^' (E) E14.5 embryos showing reduced SMC coverage, with abnormal morphology. F,G. Areas corresponding to the insets in panels D,E revealing reduced/discontinuous SMC coating and blood cells leaking out (arrows) of the artery in Prdm 16^' mice (G) but not in WT littermates (F). (H) Quantification of SMC coating in Prdm16^+/+ (black bar), Prdm16^+/~ (grey bar), and Prdm16^' (white bar) E14.5 embryos (expressed as aSMA⁺ area in μηη ± s.e.m. clearly revealing reduced SMC investment in Prdm16^' mice versus

Prdm16^+/~ mice. Scale bars correspond to 40 μηη.^**Ρ<0.01 versus Prdm16^+/~, ^****P<0.0001 versus Prdm16^+/+.

FIGURE 38 displays Prdm16^{+ "} mice with impaired perfusional recovery upon limb ischemia. A. Graph displaying the time-dependent gradual recovery of blood perfusion in the paw of the operated limb, relative to its perfusion in the control limb for Prdm16^+/+ (full line) and Prdm16^+/~ (dotted line) mice. Prdm16^+/~ mice display hampered recovery as evidenced by their reduced perfusion levels (Λ =12-12; P<0.001 ). B,C. Representative laser Doppler images of a Prdm16^+/+ (B) and Prdm16^+/~ (C) mouse 1 week after surgery, demonstrating reduced blood flow in Prdm16^+/~ mice compared to their Prdml 6^+/+ siblings in the paw of the ischemic limb (compare left squares in B and C), while perfusion in the non-ischemic control limb is similar in both genotypes (compare right squares in B and C). D,E. Pecaml immunofluorescence staining (in grey) on gastrocnemius muscle of Prdm16^+/+ (D) and Prdm16^+/~ (E) mice indicating equal capillary density between both genotypes (A =12-12). Scale bars represent 100 μηη.

FIGURE 39 displays Prdml 6^{+ "} mice with macroscopic signs of necrosis upon limb ischemia. A. Graph representing the fraction of mice (%) displaying macroscopic signs of necrosis during the recovery of an ischemic insult. Prdm16^+/~ mice (black bars) clearly exhibit more necrosis compared to their WT littermates (white bars) throughout the experiment (A =12-12). B,C. Representative images of the ischemic paw of a Prdml 6^+/+ (B) or Prdm16^+/~ (C) mouse 7 days after ligation, demonstrating necrotic digits (arrows in C) and signs of inflammation in Prdm16^+/~, but not Prdm16^+/+ mice (compare brackets in B and C). *P<0.05 versus Prdm16^+/+. FIGURE 40 displays Prdml 6^{+ "} mice with increased fibrosis in ischemic hind limbs compared to their Prdml 6^+l+ littermates. A,B Sirius red images of the gastrocnemius muscle of Prdm16^+/+ (A) and Prdm16^+/~ (B) mice 21 days after induction of hind limb ischemia, demonstrating elevated levels of fibrosis (grey) in Prdm16^+/~ mice compared to their Prdm16^+/+ littermates. Also note the increased replacement of muscle by adipose tissue in Prdml 6^+/~ mice. C. Quantification of the amount of fibrosis between Prdm16^+/+ (white bar) and Prdm16^+/~ (black bar) mice showing a 2-fold increase in the amount of fibrotic tissue in Prdm16^+/~ versus Prdml 6^+/+ mice. Scale bars represent 100 μΜ. *P<0.05 versus Prdml 6^+/+.

FIGURE 41 shows how ectopic Prdml 6 drives the onset of multiple Notch pathway penes. A. qRT-PCR showing the relative expression levels of several Notch-related genes in HUVECs treated with Prdml 6 lentivirus (black bars) or a control (Cherry; white bars) virus, demonstrating clear upregulation of DLL4, HEY1, HEY2 and EFNB2 upon ectopic Prdm16 expression. Prdm16 had no effect on the expression of the venous markers EPHB4 and COUP-TFII, but was able to suppress NRP2. B. qRT-PCR showing the relative expression levels of several Notch-related genes in BOECs treated with Prdm16 lentivirus (black bars) or a control (Cherry; white bars) virus, revealing enhanced expression levels of Notch pathway genes DLL4, HEY1, HEY2 and EFNB2 upon excess Prdm16. Vice versa, Prdm16 was able to suppress NRP2, nut not EPHB4 or COUP-TFII. ^*P<0.05 versus Cherry; ^**P<0.01 versus Cherry.

FIGURE 42 shows how Prdm16 induces canonical Notch signalling in vitro. BOECs were transduced with a Renilla control virus and an RBPJK-luciferase (RBPJK-IUC) reporter virus alone (white bar) or in combination with either a control (Cherry; grey bar) or Prdm 76-encoding lentivirus (black bar). Co-transduction with Prdm16 resulted in a 2.4 fold induction of canonical Notch signalling, as measured by the ratio of luciferase versus Renilla signal. ^*P<0.05 versus Cherry; N=3. FIGURE 43 shows how Prdm16 activates the canonical Notch pathway. BOECs were treated with a control (Cherry, white bars) or a Prdm16-containing (black bars) lentivirus in the presence of the canonical Notch inhibitor DAPT, or its control DMSO. qRT-PCR analysis revealed that Prdm16-mediated induction of the Notch ligand DLL4 (A) and the Notch downstream target EFNB2 (D) was completely abolished in the presence of DAPT, but not DMSO. Likewise, Prdm16 induces HEY1 (B) and HEY2 (C) in the presence of DMSO, while its effect on the latter two Notch effector genes was severely hampered in the presence of DAPT. ^*P<0.05 versus Cherry; ^**P<0.01 versus Cherry; N=3.

FIGURE 44 shows how Prdm 16 arterialises BOECs through activation of canonical Notch. qRT-PCR analysis of BOECs treated with a control (Cherry, white bars) or a Prc/m76-containing (black bars) virus demonstrating that Prdm16 augments the expression levels of both the arterial-specific genes GJA5 (A) and DKK2 (B) in the presence of DMSO, but not upon incubation with the canonical Notch inhibitor DAPT. ^*P<0.01 versus Cherry; N=3. FIGURE 45 shows how Prdm 16 deficiency in zebrafish does not result in reduced notch signalling. A. Representative image of ns Mo-treated canonical notch reporter zebrafish (Tg(Tp1-Mmu.Hbb:eGFP)) embryo at 72 hpf. Note the clear signal in the dorsal aorta (DA), while the posterior cardinal vein (PCV) is devoid of canonical notch activity. Canonical Notch activity is also high in the neural tube (NT). B. Representative image of a prdm16 Mo-treated Tg(Tp1-Mmu.Hbb:eGFP) embryo at 72 hpf, indicating no observable difference between notch activity in the DA of prdm16 Mo versus ns Mo (>A)-treated zebrafish. Scale bars represent 100 μΠΓΙ .

FIGURE 46 shows how DAPT induces aortic abnormalities in prdm16 Mo, but not ns Mo-treated zebrafish. A. Quantification of the aortic defects seen in ns Mo or prdm16 Mo-treated Tg(kdr-eGFP)⁵⁸⁴³ zebrafish simultaneously treated with DMSO or DAPT, demonstrating that, while DAPT had no effect on ns Mo-treated, it clearly resulted in increased aortic defects in prdm16 Mo-treated embryos. Numbers of embryos analysed per condition are mentioned in the graph bars. B-E. Representative images of ns Mo DSMO (S), ns Mo DAPT (C), prdm16 Mo DMSO

(D) and prdm16 Mo DAPT (E) treated embryos, illustrating the aortic hypoplasia readily observed in prdm16 Mo DAPT treated zebrafish embryos (see also inset in

(E) ). Dimensions of the DA are demarcated by dotted lines. Scale bars represent 200 μΓη . *P<0.05; **P<0.01 ; ****P<0.0001 .

FIGURE 47 displays how co-injections of grl Mo with prdm16 Mo severely aggravates the vascular defects observed in zebrafish treated with only prl Mo or prdm16 Mo. A. Quantification of the aortic defects seen in ns Mo, grl Mo, prdm16 Mo or grl + prdm16 Mo-treated Tg(kdr-eGFP)^s843 zebrafish, demonstrating increased percentage of embryos presenting aortic defects (hypoplasia) in grl Mo + prdm16 Mo-treated zebrafish compared to single Mo-treated zebrafish. Numbers of embryos analysed per condition are mentioned in the graph bars. B-E. Representative images of ns Mo (B), grl Mo (C), prdm16 Mo (D) and grl Mo + prdm16 Mo DAPT (E) treated embryos, illustrating the aggravated aortic hypoplasia observed in grl Mo + prdm16 Mo-treated zebrafish embryos. Dimensions of the DA are demarcated by dotted lines. Scale bars represent 200 μηη. *P<0.05; **P<0.01 ; ****P<0.0001 .

FIGURE 48 shows that Prdm16, but not Prdm16ACtBP or PrdmWADNA exerts full arterialising capacity in BOECs. qRT-PCR analysis revealing only partial or completely abolished upregulation of Notch pathway genes {DLL4, HEY1, HEY2 and EFNB2) in BOECs upon overexpression of Prdml 64CfSP (dark grey) or Prdml 64D/V/4 (light grey) mutants, compared to overexpression of WT Prdm16 (black). Expression values are relative to control (Cherry; indicated by dotted line) treated BOECs. Similarly, WT Prdm16, but not Prdml 64CfSP or Prdml resulted in enhanced levels of the arterial-specific genes GJA5 and DKK2. *P<0.05 versus WT Prdml 6; A/=8-1 1 .

FIGURE 49 shows that Prdml 6, but not Prdml 6ACtBP or Prdml 6ΔΡΝΑ induces full canonical Notch activation. BOECs were transduced with a Renilla control virus and an RBPJK-luciferase (RBPJK-IUC) reporter virus in combination with a control (Cherry; white), WT Prdml 6 (black), Prdml GACtBP (dark grey) or Prdml 6ADNA (light grey) mutants. While WT Prdml 6 clearly activated canonical Notch activity as measured by luciferase/Renilla signal, this effect was drastically reduced when either of the mutant Prdml 6 variants was applied. Results are given as fold- induction versus Cherry treated BOECs, the latter indicated by a dotted line. *P<0.05 versus WT Prdml 6; N=7.

FIGURE 50 shows how Prdml 6 leads the way to proper arterial differentiation and development .A. Diagram representing qRT-PCR results demonstrating that both class 3 semaphorins, SEMA3C (left diagram) and SEMA3G (right diagram) are upregulated by ectopic mPrdm16 expression compared to transduction with a Cherry-control virus. *P<0.05; ***P<0.001 versus Cherry-control virus; N=4. B. (Left panel) Since Prdml 6 is able to induce SEMA3G and more notably SEMA3C, Prdml 6 might instruct AECs, but not VECs to secrete a gradient of Sema3c/Sema3g to attract multiple layers of SMCs during arterial development, thereby functionally defining arterial identity. (Right panel) Alternatively, but not mutually exclusive, Prdml 6 interacts with the Notch pathway to induce arterial differentiation via multiple potential mechanisms. Indeed, Prdml 6 might directly bind to CtBP to convert the RBPJK-containing repressor complex to an activating state. Alternatively, Prdml 6 exerts epigenetic functions via its PR domain or via its interaction with histone modifying enzymes such as HMTs and HDACs. Finally, Prdml 6 might activate both Hey1 and Hey2 in a DII4-independent or DII4- dependent manner by directly triggering the release of NICD. DLL4: delta-like ligand 4; SEMA: semaphorin; SMC: smooth muscle cell : (A/V)EC: arterial/venous endothelial cell; NICD: Notch intracellular domain; NECD: Notch extracellular domain; HDAC: histone deacetylase; CtBP: C-terminal binding protein; PRDM16: PR domain containing protein 16; RBPJK: recombination signal binding protein for immunoglobulin kappa J region. TABLE 1 provides an Antibodies list for work described in Example 1

TABLE 2 provides a Primer list for work described in Example 1

TABLE 3 provides heart TF cloning information

TABLE 4 on on General differential signatures of brain, liver and heart murine endothelial cells (ECs) TABLE 5 provides the transcription factors in the tissue EC-related fingerprints TABLE 6 provides the Reference signature of (murine) heart ECs TABLE 7 provides afunctional annotation list for the (validated) signatures TABLE 8 provides a Primer list for work described in Example 2 TABLE 9 provides a Nanostring probe list TABLE 10 provides the arterial TF cloning information

TABLE 1 1 provides characteristics of the arteriovenous human differential reference signature

TABLE 12 provides the probe set intensities in HUAEC/HUVEC of genes contained within the arteriovenous fresh profile. The table represents the average microarray probe set intensities for all 102 differentially expressed probe sets (corresponding to 64 annotated arterial genes and 12 annotated venous genes) in freshly isolated HUAEC (column A; N=4) and HUVEC (column B; N=4) or cultured HUAEC (column C; N=5) or HUVEC (column D; N=5). Rows for both arterial and venous markers are sorted according to the degree of differential expression between freshly isolated HUVEC and HUAEC (column E). Calculated differences between freshly isolated HUAEC or HUVEC and their cultured counterparts reveal dramatic loss of expression of arterial markers in cultured HUAEC (column F) and venous markers in cultured HUVEC (column G). As a result, expression differences between cultured HUAEC and HUVEC for the majority (-73%) of the probe sets in the arteriovenous fresh profile were small (> -1 or < 1 ; column H). NA: not annotated.

TABLE 13 demonstrates the expression of genes from the arteriovenous fresh profile in DLL4-Fc treated HUAEC. The table represents relative expression levels (± s.e.m.; N=4) for the 76 genes in the arteriovenous fresh profile upon exposure of cultured HUAEC to DLL4-Fc compared to bovine serum albumin (BSA)-treated HUAEC. Genes are sorted according to the same order as in Table 12. ^*: P<0.05 versus BSA. is TABLE 14 provides expression of genes from the arteriovenous fresh profile in cultured HUVEC overexpressing transcription factors. The table represents relative expression levels (± s.e.m.; A/=4-6) for the 76 genes in the arteriovenous fresh profile upon lentiviral transduction of HUVEC with a single transcription factor or a combination of all 8 transcription factors, compared to those in HUVEC transduced with a cherry control lentivirus. Genes are sorted according to the same order as in Table 12. ^*: P<0.05 versus cherry control. ^#: x10³; aEC: arterial endothelial cell; vEC: venous endothelial cell; ND: not detectable.

TABLE 15 provides the primer list for work described in Example 3

TABLE 16 provides the gene list of arterial- and venous-specific genes differentially expressed ([logvalue]>1 and P<0.001 ) between mAECs and mVECs TABLE 17 provides the average arterial and venous probe intensities and corresponding log2 ratios for all TFs identified in our human and murine microarrays

TABLE 18 provides the list of conserved arterial/venous-specific genes

References to this application Abbott NJ, Patabendige AA, Dolman DE, Yusof SR, Begley DJ.(2010). Structure and function of the blood-brain barrier. Neurobiol D/s.37: 13-25

Abel ED, O'Shea KM, Ramasamy R. (2012). Insulin resistance: Metabolic mechanisms and consequences in the heart. Arterioscler Thromb Vase S/o/.32:2068-2076 Aird, W.C. (2007a). Phenotypic heterogeneity of the endothelium: I. Structure, function, and mechanisms. Circ Res 100, 158-173.

Aird, W.C. (2007b). Phenotypic heterogeneity of the endothelium: II. Representative vascular beds. Circ Res 100, 174-190.

Amatschek, S., Kriehuber, E., Bauer, W., Reininger, B., Meraner, P., Wolpl, A., Schweifer, N., Haslinger, C, Stingl, G., and Maurer, D. (2007). Blood and lymphatic endothelial cell-specific differentiation programs are stringently controlled by the tissue environment. Blood 109, 4777-4785.

Aranguren, X.L., Luttun, A., Clavel, C, Moreno, C, Abizanda, G., Barajas, M.A., Pelacho, B., Uriz, M., Arana, M., Echavarri, A., et al. (2007). In vitro and in vivo arterial differentiation of human multipotent adult progenitor cells. Blood 109, 2634- 2642.

Auerbach, R., Alby, L, Morrissey, L.W., Tu, M., and Joseph, J. (1985). Expression of organ-specific antigens on capillary endothelial cells. Microvasc Res 29, 401 - 41 1 .

Augustus AS, Buchanan J, Park TS, Hirata K, Noh HL, Sun J, Homma S, D'Armiento J, Abel ED, Goldberg IJ. (2006). Loss of lipoprotein lipase-derived fatty acids leads to increased cardiac glucose metabolism and heart dysfunction. J Biol Cftem.281 :8716-8723

Brambrink, T., Foreman, R., Welstead, G.G., Lengner, C.J., Wernig, M., Suh, H., and Jaenisch, R. (2008). Sequential expression of pluripotency markers during direct reprogramming of mouse somatic cells. Cell Stem Cell 2, 151 -159.

Brutsaert, D.L. (2003). Cardiac endothelial-myocardial signalling: its role in cardiac growth, contractile performance, and rhythmicity. Physiol Rev 83, 59-1 15.

Burgess, R., Rawls, A., Brown, D., Bradley, A., and Olson, E.N. (1996). Requirement of the paraxis gene for somite formation and musculoskeletal patterning. Nature 384, 570-573.

Conway, E.M., and Carmeliet, P. (2004). The diversity of endothelial cells: a challenge for therapeutic angiogenesis. Genome Biol 5, 207.

Corada, M., Nyqvist, D., Orsenigo, F., Caprini, A., Giampietro, C, Taketo, M.M., Iruela-Arispe, M.L., Adams, R.H., and Dejana, E. (2010). The Wnt/beta-catenin pathway modulates vascular remodeling and specification by upregulating DII4/Notch signalling. Dev Cell 18, 938-949. Davies BS, Beigneux AP, Fong LG, Young SG. (2012). New wrinkles in lipoprotein lipase biology. Curr Opin Lipidol.23:35-42

Deng, D.X., Tsalenko, A., Vailaya, A., Ben-Dor, A., Kundu, R., Estay, I., Tabibiazar, R., Kincaid, R., Yakhini, Z., Bruhn, L, et al. (2006). Differences in vascular bed disease susceptibility reflect differences in gene expression response to atherogenic stimuli. Circ Res 98, 200-208.

Endo, K., Karim, M.R., Taniguchi, H., Krejci, A., Kinameri, E., Siebert, M., Ito, K., Bray, S.J., and Moore, A.W. (2012). Chromatin modification of Notch targets in olfactory receptor neuron diversification. Nat Neurosci 15, 224-233.

Fajardo, L.F. (1989). The complexity of endothelial cells. A review. Am J Clin Pathol 92, 241 -250.

Fischer, A., Schumacher, N., Maier, M., Sendtner, M., and Gessler, M. (2004). The Notch target genes Hey1 and Hey2 are required for embryonic vascular development. Genes Dev 18, 901 -91 1 .

Geraud, C, Schledzewski, K., Demory, A., Klein, D., Kaus, M., Peyre, F., Sticht, C, Evdokimov, K., Lu, S., Schmieder, A., et a/.(2010). Liver sinusoidal endothelium: a microenvironment-dependent differentiation program in rat including the novel junctional protein liver endothelial differentiation-associated protein-1 . Hepatology 52, 313-326.

Goerdt, S., and Sorg, C. (1992). Endothelial heterogeneity and the acquired immunodeficiency syndrome: a paradigm for the pathogenesis of vascular disorders. Clin Investig 70, 89-98.

Goupille, O., Saint Cloment, C, Lopes, M., Montarras, D., and Robert, B. (2008). Msx1 and Msx2 are expressed in sub-populations of vascular smooth muscle cells. Dev Dyn 237, 2187-2194.

Gulati, R., Jevremovic, D., Peterson, T.E., Chatterjee, S., Shah, V., Vile, R.G., and Simari, R.D. (2003). Diverse origin and function of cells with endothelial phenotype obtained from adult human blood. Circ Res 93, 1023-1025.

Hagberg, E., Falkevall, A., Wang, X., Larsson, E., Huusko, J., Nilsson, I., van Meeteren, L., Samen, E., Lu, L., Vanwildemeersch, M. et al. (2010). Vascular endothelial growth factor B controls endothelial fatty acid uptake. Nature 464, 917- 921 . Hagberg CE, Mehlem A, Falkevall A, Muhl L, Farm BC, Ortsater H, Scotney P, Nyqvist D, Samen E, Lu L, Stone-Elander Set a/.(2012). Targeting VEGF-B as a novel treatment for insulin resistance and type 2 diabetes. A/afure.490:426-430 Hanna, J., Markoulaki, S., Schorderet, P., Carey, B.W., Beard, C, Wernig, M., Creyghton, M.P., Steine, E.J., Cassady, J. P., Foreman, R., et al. (2008). Direct reprogramming of terminally differentiated mature B lymphocytes to pluripotency. Cell 133, 250-264.

Hendrickx, B., Verdonck, K., Van den Berge, S., Dickens, S., Eriksson, E., Vranckx, J.J., and Luttun, A. (2010). Integration of blood outgrowth endothelial cells in dermal fibroblast sheets promotes full thickness wound healing. Stem Cells 28, 1 165-1 177.

Hildebrand, J.D., and Soriano, P. (2002). Overlapping and unique roles for C- terminal binding protein 1 (CtBP1 ) and CtBP2 during mouse development. Mol Cell Biol 22, 5296-5307.

Jaenisch, R., and Young, R. (2008). Stem cells, the molecular circuitry of pluripotency and nuclear reprogramming. Cell 132, 567-582.

Jin, S.W., Beis, D., Mitchell, T., Chen, J.N., and Stainier, D.Y. (2005). Cellular and molecular analyses of vascular tube and lumen formation in zebrafish. Development 132, 5199-5209.

Jaswal JS, Keung W, Wang W, Ussher JR, Lopaschuk GD. (201 1 ). Targeting fatty acid and carbohydrate oxidation~a novel therapeutic intervention in the ischemic and failing heart. Biochimica et biophysica acta 813:1333-1350

Kajimura, S., Seale, P., Tomaru, T., Erdjument-Bromage, H., Cooper, M.P., Ruas, J.L., Chin, S., Tempst, P., Lazar, M.A., and Spiegelman, B.M. (2008). Regulation of the brown and white fat gene programs through a PRDM16/CtBP transcriptional complex. Genes Dev 22, 1397-1409.

Kanda, T., Brown, J.D., Orasanu, G., Vogel, S., Gonzalez, F.J., Sartoretto, J., Michel, T., and Plutzky, J. (2009). PPARgamma in the endothelium regulates metabolic responses to high-fat diet in mice. J Clin Invest 119, 1 10-124.

Kambara T, Ohashi K, Shibata R, Ogura Y, Maruyama S, Enomoto T, Uemura Y, Shimizu Y, Yuasa D, Matsuo K, Miyabe M, Kataoka Yet al. (2012). CTRP9 protein protects against myocardial injury following ischemia-reperfusion through AMP- activated protein kinase (AMPK)-dependent mechanism. J Biol Cftem.287: 18965- 18973

Kinameri, E., Inoue, T., Aruga, J., Imayoshi, I., Kageyama, R., Shimogori, T., and Moore, A.W. (2008). Prdm proto-oncogene transcription factor family expression and interaction with the Notch-Hes pathway in mouse neurogenesis. PLoS One 3, e3859.

Kishida, K., Shimomura, I., Nishizawa, H., Maeda, N., Kuriyama, H., Kondo, H ., Matsuda, M., Nagaretani, H., Ouchi, N., Hotta, K., ef a/.(2001 ). Enhancement of the aquaporin adipose gene expression by a peroxisome proliferator-activated receptor gamma. J Biol Chem 276, 48572-48579.

Kramarzova K, Stuchly J, Willasch A, Gruhn B, Schwarz J, Cermak J, Machova- Polakova K, Fuchs O, Stary J, Trka J, Boublikova L. (2012). Real-time PCR quantification of major Wilms' tumor gene 1 (WT1 ) isoforms in acute myeloid leukemia, their characteristic expression patterns and possible functional consequences. Leukemia.26:2086-2095

Liao, W.P., Uetzmann, L, Burtscher, I., and Lickert, H. (2009). Generation of a mouse line expressing Sox17-driven Cre recombinase with specific activity in arteries. Genesis 47, 476-483.

Majesky, M.W. (2007). Developmental basis of vascular smooth muscle diversity. Arterioscler Thromb Vase Biol 27, 1248-1258.

Mankoo, B.S., Collins, N .S., Ashby, P., Grigorieva, E., Pevny, L.H ., Candia, A.,

Wright, C.V., Rigby, P.W., and Pachnis, V. (1999). Mox2 is a component of the genetic hierarchy controlling limb muscle development. Nature 400, 69-73.

Motoike, T., Loughna, S., Perens, E., Roman, B.L., Liao, W., Chau, T.C., Richardson, CD., Kawate, T., Kuno, J., Weinstein, B.M., et al. (2000). Universal

GFP reporter for the study of vascular development. Genesis 28, 75-81 .

Neubauer S. (2007). The failing heart-an engine out of fuel. N Engl J

Mec/.356:1 140-1 151

Pearson, J.D. (2009). Endothelial progenitor cells - hype or hope? J Thromb Haemost 7, 255-262.

Pries, A.R., and Kuebler, W.M. (2006). Normal endothelium. Handb Exp Pharmacol, 1 -40. Ribatti, D., Nico, B., Vacca, A., Roncali, L, and Dammacco, F. (2002). Endothelial cell heterogeneity and organ specificity. J Hematother Stem Cell Res 11, 81 -90. Roukens, M.G., Alloul-Ramdhani, M., Baan, B., Kobayashi, K., Peterson-Maduro, J., van Dam, H., Schulte-Merker, S., and Baker, D.A. (2010). Control of endothelial sprouting by a Tel-CtBP complex. Nat Cell Biol 12, 933-942.

Seale, P., Kajimura, S., Yang, W., Chin, S., Rohas, L.M., Uldry, M., Tavernier, G., Langin, D., and Spiegelman, B.M. (2007). Transcriptional control of brown fat determination by PRDM16. Cell Metab 6, 38-54.

Stanley WC, Dabkowski ER, Ribeiro RF, Jr., O'Connell KA. (2012). Dietary fat and heart failure: Moving from lipotoxicity to lipoprotection. Circ Res.1 10:764-776

Swift, M.R., and Weinstein, B.M. (2009). Arterial-venous specification during development. Circ Res 104, 576-588.

Takahashi, K., Tanabe, K., Ohnuki, M., Narita, M., Ichisaka, T., Tomoda, K., and Yamanaka, S. (2007). Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131, 861 -872.

Wagner KD, Wagner N, Bondke A, Nafz B, Flemming B, Theres H, Scholz H.(2002). The Wilms' tumor suppressor Wt1 is expressed in the coronary vasculature after myocardial infarction. FASEB J.16:1 1 17-1 1 19

Wagner, N., Wagner, K.D., Theres, H., Englert, C, Schedl, A., and Scholz, H. (2005). Coronary vessel development requires activation of the TrkB neurotrophin receptor by the Wilms' tumor transcription factor Wt1 . Genes Dev 19, 2631 -2642. Wang, H.U., Chen, Z.F., and Anderson, D.J. (1998). Molecular distinction and angiogenic interaction between embryonic arteries and veins revealed by ephrin-B2 and its receptor Eph-B4. Cell 93, 741 -753.

Wernig, M., Zhao, J. P., Pruszak, J., Hedlund, E., Fu, D., Soldner, F., Broccoli, V., Constantine-Paton, M., Isacson, O., and Jaenisch, R. (2008). Neurons derived from reprogrammed fibroblasts functionally integrate into the fetal brain and improve symptoms of rats with Parkinson's disease. Proc Natl Acad Sci U S A 105, 5856- 5861 .

Wick, N., Saharinen, P., Saharinen, J., Gurnhofer, E., Steiner, C.W., Raab, I., Stokic, D., Giovanoli, P., Buchsbaum, S., Burchard, A., ef a/.(2007). Transcriptomal comparison of human dermal lymphatic endothelial cells ex vivo and in vitro. Physiol Genomics 28, 179-192. Yoder, M.C., Mead, L.E., Prater, D., Krier, T.R., Mroueh, K.N., Li, F., Krasich, R., Temm, C.J., Prchal, J.T., and Ingram, D.A. (2007). Redefining endothelial progenitor cells via clonal analysis and hematopoietic stem/progenitor cell principals. Blood 109, 1801 -1809.

You, L.R., Lin, F.J., Lee, C.T., DeMayo, F.J., Tsai, M.J., and Tsai, S.Y. (2005). Suppression of Notch signalling by the COUP-TFII transcription factor regulates vein identity. Nature 435, 98-104.

TABLE 1 : Antibodies list for work described in Exampie 1

Antigen Species Supplier, catalog N° Label

CD31 Mouse BD 558738 FITC

CD31 Mouse BD 551262 APC

CD34 Mouse BD 55138 PE

CD45 Mouse BD 559864 APC

CD36 Mouse Biolegend 102612 APC

CD31 Human BD 555446 PE

CD31 Human BD 555445 FITC

CD34 Human BD 345804 APC

CD45 Human BD 345808 FITC

PODOPLANIN Human Angiobio 1 1-009PE PE

TIE2 Human Reliatech 101 -NBi54 biotin

Endoglin Mouse R&D AF1320 none

Laminin Mouse Sigma L9393 none

Meox2 Mouse Eurogentec none

Tcf15 Mouse Santa Cruz sc-46438 none

Gapdh Human/Mouse Cell Signaling 21 18 none a-Tubulin Human/Mouse Sigma T6199 none

TIMP4 Human Abeam ab58425 none

ZDHCC2 Human Santa Cruz sc-98219 none

RBP7 Human Sigma HPA034749 none

BSI Lectin Mouse Sigma L3759 biotin

TABLE 2: Primerlist for work described in Exampie 1

Gene Species Forward primer (5'3') Reverse primer (5'3')

GAPDH H TGGTATCGTGGAAGGACTCATGAC ATGCCAGTGAGCTTCCCGTTCAGC

ACTB H TGGCACCACACCTTCTACAATG TAGCAACGTACATGGCTGGG

ABLIM3 H CTTCATCACAGGCGAAGTCA TTGGTCCACGAATCTTGATG

ADAMTS9 H TACACCGCAAACGACTGTGT TCACGATCGGGAGGTTTATC

ADH1 H ATGGAGGTGTGGA I I I I TCG GTGCGTCCAGTCAGTAGCAG

AKAP5 H GACGCCCTACGTTGATCTTT CCAGGCATATTCATACCTTTCA

AQP7 H GACTGGG GACACAG G G ATAG GGATGCTTGAACCATG I I I I G

C1QTNF9 H AATGGTCTGCCTGGAAGAGA CCTTTGATGCC I I I I GCTTC

EBF3 H TGCCACCGTCATCATAATTG TGCAGAACTGCTTGGATTTG

EEPD1 H GTCACGACCTGACCCTTGTT GGGTTCTTGGTGCTGATGTT

FAPB9 H TGATGGGAAAATGATGACCA CTCTTTGCCAAGCCATTTTT

H19 H GAGCTCTCAGGAGGGAGGAT CCAGCCTAAGGTGTTCAGGA

HN1I H TCCTTCCAGCAGGCCTAATA CCAAAAATGTCGCTGGTCTT

KCNA5 H CAGTTCCCCAACACACTCCT CAGGGGCTTCTCCTCTTCTT

LPL H GTCCGTGGCTACCTGTCATT TGTCCCACCAGTTTGGTGTA

MAP3K5 H GCCCCTGCATGAAGAAATAG TTAATGGACCCCATTTGGAA

MCTP1 H TAATGCAGAAGTCCCCTTGG CCACACAGGGTTGAGGTTCT

MEOX2 H CCAG G AAAG AAAG GACAG CA TCCACCCTTTACCCTCTTCC

MEOX2 UTR H TCTGGGACCACCTTCTTTTG CCACCACCCTCTGTCACTTT

NAV3 H GGACCGAGTGGTACTGGAAA CAGAGAGCCCACATGATGAA

NTS H AAGCACATGTTCCCTCTTGG AAGCCCTGCTGTGACAGATT

PA PS S 2 H GAGGTGGCTAAGCTGTTTGC AAG CACAC G CTCAG GAG TTT

PDPN H TGC TCT TCG TTT TGG GAA GC TCGCTGGTTCCTGGAGTCAC

PMP2 H GGGGTTAGCCAC C AG AAAAC TCTCTTTGCCATCCCATCTC

PPARG H GACCACTCCCACTCCTTTGA CGACATTCAATTGCCATGAG

PPARG UTR H CTCGAGGACACCGGAGAG GGCTGACTCTCGTTTGAGAAA

PRND H ACCAAGGAGGCATTTGTCAC GCTGCTGCACTCTGTACTGC

RBP7 H TGCTGAAGCCACAGAAAGTG AGGTGAGCCTGTCATTGTCC

RTN1 H GAGTTTCCTGCTGCTGCTCT AGGGTGATCTCAAGCTCCAA

SLC28A2 H CCTTGGGGACATGACACTCT GCTTTACCCCCTCCTCACTC

TCF15 H GCAGCTGCTTGAAGGTGAG CGGTCCCTACACAAAGAAGG

TCF15 UTR H TCCTGGAGAGCTGTGAGGAT CCCAGAACATCTGTGCCTCT

TIMP4 H CAGACCCTGCTGACACTGAA AGACTTTCCCTCTGCACCAA

WT1 H CAAATGACATCCCAGCTTGA GACACCGTGCGTGTGTATTC

WT1 UTR H CAGGCTGCTAACCTGGAAAG CTCCATTTGTGCAAGGAGGT

ZDHHC2 H TTG CG G CAACAG ATTTACAG TG CTG ACTAG CCAACAATG A

ACTA2 H TGGTGTGTGACAATGGCTCT C I I I I CCATGTCGTCCCAGT

PROX1 H CAGTACTGAAGAGCTGTCTATAACCAGAG TCTG AG CAACTTC CAG G AATCTC

PTPRC H G ACAACAGTG G AG AAAG G AC G GGGAAGGTGTTGGGCTTTG

TIE2 H ACACCTGCCTCATGCTCAGC AG CAGTACAG AG ATG GTTG CATTC

PECAM1 H TCTGCACTGCAGGTATTGACAA CTGATCGATTCGCAACGGA

Tubb M GGGAGGTGATAAGCGATGAA CCCAGGTTCTAGATCCACCA Gapdh M CCGCATCTTCTTGTGCAGT GAATTTGCCGTGAGTGGAGT

Actb M GACGGCCAGGTCATCACTAT CTTCTGCATCCTGTCAGCAA

Ablim3 M CAGTCACTTCTGGCACTGGA CCCCTCAGTGAACATCTGGT

Add M CAGCTGCATCTTCCTCCTCT ATGGTGGTGCCTCCAGATAG

Adamst9 M ACCCGGATGAGATACGTCAG GGTCAGTCTCGGGGATGTAA

Adh1 M ACAAACCCTTCACCATCGAG CCTTCTCCAACGCTCTCAAC

Aim M CGGCAGTTTGGTTTACAGGT CTGGTGTCCCTGTTGGTTCT

Akap5 M AGCAGAGCTGGCCAGAGAC TGTGACTTCTTCAGGACACGA

Akr1d9 M AGCTCTAGAGGCTGGGTTCC CAGCTCTGGCCTATGGAAAG

Aqp7 M CTCGGTGTCAACTTGGGTTT CCTTGGAACCTGTCACCAAC

C1qtnf9 M AGGTCACAATGGCCTACCTG AGCCTGGATCACCTTTGATG

Cox8b M TGCGAAGTTCACAGTGGTTC TGCTGCGGAGCTCTTTTTAT

Dmtp M CAATGGAACTACGCCTGCAT AATATGGGCACCTCTTGCTG

Ebf3 M CTCGACACATCCCTGGAGTT GTAACCTTTCGGGGTCACCT

Eepdl M TTTGGAGGTATGAGGGATGG GCCTCTTTGTCAAGCAGCTC

Enpp3 M CAGCTACGGGAACAATGGAT CCTCTGTGCGATGAGTCAAA

Fabp3 M GACGAGGTGACAGCAGATGA TGCCATGAGTGAGAGTCAGG

Fabp9 M ATG G ACATCCAAG CAG GAAG CTGTTTGCCAAGCCATTTTT

Gda M GTGGTGCTTCAAACCATGTG TTCTTCAGCCACATCAGTGC

H19 M AATGGTGCTACCCAGCTCAT GCAGAGTTGGCCATGAAGAT

Hn1l M AGGCAAAGGAAGTGGGATCT TGGAGGATTTGTCCTCTTGG

Kcna5 M CGCTACTTCGATCCCTTGAG CAGCTGGTAAAAGCGGATCT

KlralO M GGCCAAGCAATGAACTTCTG ATGCTGGCAGTTCGCTTTAC

Klra9 M AGATTCCTCACGGGACACAG ATTGGCCATTGTCAATCCAT

Lambl M GGAACCGGAGTTCAGCTATG GGGTTGAGGGTCTCGTGATA

Lpl M TTTGGCTCCAGAGTTTGACC GTCTTGCTGCTGTGGTTGAA

Map3k5 M GGCCAACAACATCATCCTCT GCCACCTTCAAAAGCTGAAC

Mb M CACCCTGAGACCCTGGATAA GAGCATCTGCTCCAAAGTCC

Mdp1 M CAGTAATGCAGATGTCCCTTTG CAAGAAAGGCTGAGCCCATA

Me ox2 M GGCAGAATTTGCCCATCATA TGCTCAGAGCTGTGGTCACT

Myh6 M CCACCCAAGTTCGACAAGAT AGGCGTTGTCAGAGATGGAG

Myl2 M GACCCAGATCCAGGAGTTCA TCAGCCTTCAGTGACCCTTT

Myl3 M AAGGGCGAGATGAAGATCAC GCAGGAACGTCTCAAAATCC

Myoz2 M CCATGCAGAATGGGAGAGTT TTTGGGTACAAAGCCTCCAG

Nav3 M AAGCAAGCAGTGACCTGGAT AGGGGACAGTGATGTTGGAG

Nts M TTCACAATGAGTTGCCAGGA GTGTGGACCTGCTTGTCAGA

Papss2 M GTCTACCAGGCCCATCATGT GACATTGTCCCCATCCAGAG

Pmp2 M GTGGGGTTAGCCAACAGAAA TCTCCAGTGTCACGATGCTC

Pparg M CAGGCCTCATGAAGAACCTT GGATCCGGCAGTTAAGATCA

Prg4 M GCCACCTGCAACTGTGATTA CTG CACAG CACTTG CCATAC

Prnd M CTGGGTTTCGTTTGGTTCAT CCCTGGCAGATACTCCGTTT

Rbp7 M TCAGCGGTACCTGGAATCTT AGGCTCGTGCATTTTCTGTT

Rtn1 M CATCAGCTTCCGCATCTACA AAAGCCTCCGTAGCTCCTTC

Slc28a2 M GAGCAGCTGATCTCCTTTGC ATCTGATCTCCCAGCCATTG

Td15 M GCTCCATCTGCACCTTCTGT TTGTCCTCCGGTCCTTACAC Timp4 M ACACGCCATTTGACTCTTCC CAGCCACAGTTCTGGTGGTA

Tnnd M CAGCAAAGGGAAGTCTGAGG CGTGCAAGACCAGCATCTAC

Tnni3 M G AAG CAG G AG ATG G AACG AG TGACTTTTGCTTCCACGTCA

Tnnt2 M CTGAGACAGAGGAGGCCAAC TTCTCGAAGTGAGCCTCGAT

Wt1 M AGGTTTTCTCGCTCAGACCA G CTG AAG G G CTTTTCACTTG

Zdhhc2 M GCTGTACTGCCTTTTCATTGC CCAAACAG AG AG G ACAAG CTG

Acta2 M CGCTGTCAGGAACCCTGAGA CGAAGCCGGCCTTACAGA

Ptprc M GCCCCGGGATGAGACAGT TTTGAAAGCCCGAGTGCCT

Cdh5 M ATTGAGACAGACCCCAAACG TTCTGGTTTTCTGGCAGCTT

Ng2 M CCTTCCTACAAGTGACCATTGC CCTGAATTCACTGCCTCCTG

Pdpn M GCCAGTGTTGTTCTGGGTTT AGAGGTGCCTTGCCAGTAGA

Pecaml M GTCATGGCCATGGTCGAGTA CTCCTCGGCGATCTTGCTGAA

Prox1 M CGCGTGGGTTTCTTCTCTGC GGGCTGTGCTGTCATGGTCA

Tek M GAAACATCCCTCACCTGCAT TGGCC I I I I CTCTCTTCCAA

Fabp5 M CAAAACCGAGAGCACAGTGA CCCTCATTGCACCTTCTCAT

Slc2a4 M ATTCTGGTTGCC CAGGTGCT ATTGGACGCTCTCTCTCCAA

Slc2a1 M CCGCCTCATGTTGGCTGT TGTGGTGGATGGGATGGG

Gpihbpl M GGGCACAAGAAGATGGTGAT CTGGAGCAGCTCTGTGTCTG

Slc27a1 M ACCACTCTGCAGGGAACATC AGCGGCAGATTTCACCTATG

Slc27a4 M CTACCCAAAGCTGCCATTGT GAACTTCTTCCGGATCACCA

CD36 M TGCCAGTCGGAGACATGCT GCCACGTCATCTGGGTTTTG

Fabp4 M TGGAAGCTTGTCTCCAGTGA TCGACTTTCCATCCCACTTC

Flt1 M TGGCCAGAGGCATGGAGT TCGCAAATCTTCACCACATGG

Nrp1 M ACACCTGAGCTTCGGACGTT CCACTGTGTGTGGCTCTCTCA

H: human; M: mouse

TABLE3. Heart TF cloning information

Cloned

mRNA Restriction

Gene sequence enzymes

Symbol Forward Primer (5'->3') Reverse Primer (5'->3') (bases) used Backbone

Genecopoeia (ref. control GFP EX-EGFP-Lvl 14)

Genecopoeia (ref. hMeox2 EX-Z3242-Lv1 14)

PRRL2-GPK- hWT1 CCCTCTAGACGATCCTGGACTTCCTCTTG TTTGTCGACTGGAGAGTCAGACTTGAAAGCA 171 -1817 Xbal-Sall Cherry

PRRL2-GPK- mTCF15 CCCTCTAGAATGGCGTTCGCGCTGCTGC TTTGCTAGCTTGTCCTCCGGTCCTTACAC 18-733 Xbal-Nhel Cherry

PRRL2-GPK- hPPRGamma CCCTCTAGAATGACCATGGTTGACACAGA TTTCTCGAGGTACAAGTCCTTGTAGATCT 248-1678 Xbal-Xhol Cherry

PRRL2-GPK- hEBF3 CCCTCTAGAGCAGTTTTCATGTTTGGGAT TTTGTCGACTCCCTCACATTGGCGGGACT 51-1719 Xbal-Sall Cherry

TABLE4. General differential signature of brain, liver and heart murine endothelial cells (ECs)

A. ENRICHED IN BRAIN 159 PROBES, 117 DESCRIBED GENES, SORT ORDER: HIGHEST DIFFERENCE BRAIN-HEART

Brain ECs Heart ECs Liver ECs Heart - Brain Brain - Liver Heart - Liv

ANNOTATION 1 SYMBOL Probe intensity (Log2 Probe intensity (Log2

NM_021471 Slcolcl 12.1 5.2 4.7 -6.9 7.4 0.5

NM_030687 Slcola4 10.9 4.0 3.7 -6.9 7.2 0.3

NM_030556 Slcl9a3 11.2 4.5 4.5 -6.6 6.7 0.0

NM_013657 Sema3c 11.1 5.0 4.7 -6.0 6.4 0.3

ENSMUST00000049544 2610301F02Rik 10.8 4.9 4.1 -6.0 6.7 0.7

NM_029662 Mfsd2 10.9 5.1 5.0 -5.8 5.9 0.1

NM_008973 Ptn 11.6 5.9 5.0 -5.7 6.7 1.0

NM_001025576 2610301F02Rik 10.8 5.3 4.5 -5.5 6.3 0.7

NM_011400 Slc2al 12.7 7.2 6.4 -5.5 6.2 0.8

ENSMUST00000049544 2610301F02Rik 10.9 5.4 4.9 -5.5 6.0 0.5

NM_031194 Slc22a8 10.9 5.6 5.3 -5.4 5.7 0.3

ENSMUST00000049544 2610301F02Rik 10.0 4.7 3.4 -5.3 6.6 1.3

NM_009738 Bche 9.4 4.1 4.5 -5.3 4.9 -0.4

BC116955 1190003M12Rik 10.3 5.1 4.6 -5.3 5.7 0.4

NM_008756 Ocln 10.0 4.9 4.7 -5.1 5.3 0.2

NM_008409 Itm2a 12.0 6.9 4.5 -5.1 7.6 2.5

NM_021436 Tmeffl 9.8 4.8 4.9 -5.0 4.9 -0.1

NM_009402 Pglyrpl 11.7 6.7 6.0 -5.0 5.6 0.7

NM_016672 Ddc 10.2 5.2 5.2 -5.0 5.0 0.0

NM_010703 Lefl 10.9 6.0 5.9 -5.0 5.0 0.1

NM_023805 Slc38a3 10.6 5.7 5.7 -4.9 4.9 0.0

NM_028060 Slc35f2 9.7 4.9 4.9 -4.9 4.9 0.0

NM_133237 Apcddl 10.6 5.9 5.7 -4.7 5.0 0.2

NM_052994 Spock2 11.1 6.4 6.0 -4.7 5.1 0.4

BC047154 A930038C07Rik 9.0 4.3 4.1 -4.7 4.9 0.2

NM_172471 Itih5 10.3 5.6 5.0 -4.7 5.3 0.6

NM_172479 Slc38a5 10.1 5.4 5.3 -4.6 4.8 0.2

NM_009728 AtplOa 10.1 5.5 5.1 -4.6 5.1 0.5

NM_011638 Tfrc 11.2 6.7 5.3 -4.4 5.9 1.5

NM_009575 Zic3 9.0 4.6 4.7 -4.4 4.4 0.0

NM_027299 Degs2 10.9 6.6 5.5 -4.3 5.4 1.0

NM_009196 Slcl6al 9.8 5.5 5.4 -4.3 4.4 0.1

ENSMUST00000098351 Foxfla 10.7 6.4 5.9 -4.3 4.8 0.5

NM_011198 Ptgs2 10.5 6.3 5.5 -4.3 5.1 0.8

NM_178797 Far2 9.7 5.5 5.4 -4.2 4.3 0.1

NM_172476 Tmc7 9.2 5.1 5.0 -4.1 4.2 0.2

ENSMUST00000049544 2610301F02Rik 7.4 3.4 3.3 -4.0 4.1 0.1

NM_013825 Ly75 8.9 4.9 4.8 -4.0 4.1 0.2

NM_028943 Sgms2 9.4 5.5 5.1 -3.9 4.3 0.4

NM_008935 Proml 11.3 7.4 5.5 -3.9 5.9 2.0

NM_175105 Aqpll 9.7 6.0 5.8 -3.7 3.9 0.2

NM_029001 Elovl7 10.7 7.1 7.0 -3.6 3.7 0.2

NM_028802 Prei4 10.6 7.0 7.4 -3.6 3.2 -0.4

NM_007515 Slc7a3 8.6 5.0 4.8 -3.5 3.8 0.2

NM_001038602 Marveld2 8.8 5.2 5.3 -3.5 3.4 -0.1

NM 010225 Foxf2 8.2 4.7 4.5 -3.5 3.7 0.2 Brain ECs Heart ECs Liver ECs Heart - flratn Brain - Liver Heart - Liver

ANNOTATION 1 SYMBOL Probe intensity (Log2 Probe intensity (Log2 )

NM_025807 Slcl6a9 9.8 6.3 6.6 -3.5 3.1 -0.3

NM_016719 Grbl4 9.3 5.8 5.6 -3.4 3.7 0.2

NM_013467 Aldhlal 9.5 6.1 6.2 -3.4 3.4 -0.1

NM_016678 Reck 10.2 6.8 5.9 -3.4 4.3 0.9

NM_029116 Kbtbdll 8.8 5.4 5.7 -3.4 3.2 -0.2

NM_026385 Pllp 9.1 5.7 5.3 -3.4 3.8 0.5

NM_009255 Serpine2 9.3 6.0 5.2 -3.4 4.2 0.8

NM_146136 Slcl6a4 10.8 7.5 6.3 -3.4 4.6 1.2

NM_008256 Hmgcs2 8.6 5.3 5.7 -3.4 2.9 -0.4

NM_001080813 Rabllfipl 8.6 5.2 5.1 -3.3 3.5 0.2

NM_015744 Enpp2 10.7 7.4 5.6 -3.3 5.1 1.8

NM_008982 Ptprj 9.3 6.1 6.2 -3.2 3.2 -0.1

NM_001099634 Myof 9.0 5.9 6.0 -3.1 3.0 -0.2

NM_008982 Ptprj 9.1 6.0 6.0 -3.1 3.0 -0.1

NM_009199 Slclal 10.2 7.1 5.1 -3.1 5.0 2.0

NM_028916 Efhc2 7.1 4.0 3.9 -3.1 3.1 0.1

NM_001004173 Sgpp2 8.7 5.7 5.4 -3.0 3.3 0.2

NM_145447 Mfsd7c 8.7 5.6 5.2 -3.0 3.5 0.4

NM_133857 Usp53 10.6 7.6 6.9 -3.0 3.8 0.7

NM_026778 Cthrcl 9.6 6.6 6.2 -3.0 3.4 0.4

NM_172778 Maob 7.9 4.9 4.7 -3.0 3.3 0.3

NM_053073 Lrp8 8.6 5.6 5.4 -3.0 3.2 0.2

NM_173740 Maoa 10.9 8.0 5.6 -3.0 5.3 2.4

NM_018811 Abhd2 11.4 8.5 7.8 -2.9 3.6 0.6

NM_009199 Slclal 9.6 6.7 5.7 -2.9 3.9 1.0

NM_008869 Pla2g4a 8.2 5.3 5.4 -2.9 2.8 -0.1

NR_000004 Snord35b 10.5 7.6 5.3 -2.9 5.2 2.3

NM_153574 Faml3a 7.7 4.8 5.1 -2.9 2.6 -0.3

NM_009367 Tgfb2 9.2 6.3 5.6 -2.9 3.6 0.7

NM_021451 Pmaipl 8.9 6.1 6.3 -2.8 2.6 -0.2

NM_028757 Nebl 8.9 6.1 5.0 -2.8 3.9 1.1

NM_007470 Apod 8.7 5.9 5.5 -2.8 3.3 0.5

NM_013869 Tnfrsfl9 8.3 5.5 5.4 -2.8 2.9 0.1

NM_007513 Slc7al 10.3 7.5 6.5 -2.8 3.7 1.0

NM_027153 Pir 8.8 6.1 5.7 -2.7 3.1 0.3

NM_198702 Lphn3 7.7 4.9 4.6 -2.7 3.1 0.3

NM_010513 Igflr 10.5 7.8 7.5 -2.7 3.0 0.3

NM_001126490 Isml 8.4 5.6 5.2 -2.7 3.2 0.5

NM_009890 Ch25h 8.2 5.5 5.3 -2.7 2.9 0.2

NM_028748 Paqr5 9.4 6.7 6.4 -2.7 2.9 0.2

NM_010104 Ednl 10.0 7.3 6.1 -2.7 3.9 1.2

NM_001042611 Cp 10.0 7.3 5.8 -2.7 4.2 1.5

NM_029614 Prss23 10.8 8.2 6.9 -2.7 3.9 1.3

NM_153546 Mboatl 7.5 4.9 4.5 -2.6 3.0 0.4

NM_001033336 Abcc4 10.1 7.4 7.5 -2.6 2.5 -0.1

NM_015732 Axin2 9.0 6.4 6.6 -2.6 2.5 -0.1

NM_146102 Afapll2 8.8 6.2 5.5 -2.6 3.3 0.7

NM_001039556 Rad54b 7.4 4.8 4.7 -2.6 2.7 0.1

NM_001024720 Hmcnl 8.3 5.8 6.1 -2.5 2.2 -0.3

NM 019671 Netl 9.7 7.2 6.3 -2.5 3.4 0.9 Brain ECs Heart ECs Liver ECs Heart - Brain Brain - Liver Heart - Live

ANNOTATION 1 SYMBOL Probe intensity (Log2 Probe intensity (Log2

NM_026672 Gstm7 8.2 5.8 5.5 -2.5 2.7 0.3

NM_009768 Bsg 11.4 9.0 9.1 -2.5 2.4 -0.1

NM_023395 Wfdcl 8.3 5.9 5.9 -2.4 2.4 0.0

NM_001024720 Hmcnl 8.1 5.7 5.7 -2.4 2.4 0.0

NM_009759 Bmx 8.8 6.4 5.3 -2.4 3.5 1.1

NM_009846 Cd24a 10.3 7.9 7.3 -2.4 3.0 0.6

NM_172752 Sorbs2 10.0 7.6 5.1 -2.4 4.9 2.5

NM_008846 Pip5klb 7.8 5.4 4.7 -2.4 3.1 0.7

NM_017405 Lsr 8.6 6.2 5.8 -2.3 2.7 0.4

NM_001024720 Hmcnl 10.1 7.8 7.5 -2.3 2.6 0.2

NM_010357 Gsta4 9.5 7.2 6.1 -2.3 3.4 1.1

NM_001024720 Hmcnl 10.4 8.1 8.1 -2.3 2.3 -0.1

NM_054041 Antxrl 8.7 6.3 5.7 -2.3 3.0 0.7

NM_001024720 Hmcnl 9.9 7.6 7.9 -2.3 2.0 -0.3

NM_172653 Slc39al0 11.9 9.6 8.3 -2.3 3.6 1.3

NM_001024720 Hmcnl 8.5 6.2 6.0 -2.3 2.5 0.3

NM_172872 Kank4 7.4 5.1 4.9 -2.3 2.5 0.2

NM_020559 Alasl 9.2 6.9 7.1 -2.3 2.1 -0.2

NM_001033289 Slc9a2 7.2 5.0 5.0 -2.3 2.3 0.0

— NA 9.4 7.2 4.7 -2.3 4.8 2.5

NM_175193 Golim4 11.3 9.1 8.8 -2.2 2.5 0.2

NM_001024720 Hmcnl 9.5 7.3 7.0 -2.2 2.6 0.3

NM_177025 Coblll 10.9 8.6 6.1 -2.2 4.8 2.5

NM_001024720 Hmcnl 10.4 8.1 8.3 -2.2 2.1 -0.1

BC048939 A130022J15Rik 11.3 9.1 7.9 -2.2 3.4 1.2

NM_001024720 Hmcnl 10.0 7.8 7.5 -2.2 2.5 0.3

NM_001024720 Hmcnl 9.8 7.6 7.5 -2.2 2.4 0.2

NM_001024720 Hmcnl 9.5 7.3 7.1 -2.2 2.3 0.1

NM_001024720 Hmcnl 10.4 8.2 8.2 -2.2 2.2 0.0

NM_001024720 Hmcnl 9.2 7.0 6.8 -2.2 2.4 0.2

NM_001024720 Hmcnl 9.6 7.4 7.3 -2.2 2.3 0.1

NM_001024720 Hmcnl 11.2 9.0 9.4 -2.2 1.8 -0.3

NM_008592 Foxcl 7.9 5.7 5.0 -2.2 2.9 0.7

NM_001024720 Hmcnl 8.9 6.7 6.7 -2.2 2.2 0.0

NM_175263 Notum 7.7 5.6 5.6 -2.2 2.2 0.0

NM_001024720 Hmcnl 10.4 8.2 8.4 -2.1 2.0 -0.1

NM_001024720 Hmcnl 11.3 9.1 9.1 -2.1 2.2 0.1

NM_001024720 Hmcnl 11.1 9.0 8.6 -2.1 2.5 0.3

NM_001024720 Hmcnl 8.7 6.6 6.6 -2.1 2.1 0.0

NM_001024720 Hmcnl 10.4 8.3 8.0 -2.1 2.4 0.3

NM_178728 Napepld 7.4 5.3 4.8 -2.1 2.6 0.5

NM_001082974 Neurl2 9.1 7.0 6.5 -2.1 2.6 0.5

NM_011158 Prkar2b 6.9 4.8 4.9 -2.1 2.0 -0.2

NM_010274 Gpd2 10.3 8.2 8.1 -2.1 2.2 0.1

NM_198161 Bhlhb9 7.7 5.6 4.5 -2.1 3.1 1.0

NM_010216 Figf 8.6 6.5 5.7 -2.1 2.9 0.8

NM_023627 Isynal 8.8 6.7 5.9 -2.1 2.9 0.8

NM_010233 Fnl 11.6 9.5 6.3 -2.1 5.3 3.2

NM_028803 Gbel 10.3 8.2 8.3 -2.1 1.9 -0.1

NM 001024720 Hmcnl 10.3 8.2 7.7 -2.1 2.6 0.5 Brain ECs Heart ECs Liver ECs Heart - Brain Brain - Liver Heart - Live

ANNOTATION 1 SYMBOL Probe intensity (Log2 Probe intensity (Log2

NM_001024720 Hmcnl 9.5 7.4 7.0 -2.1 2.4 0.4

ENSMUST00000095134 Nebl 7.6 5.5 4.8 -2.1 2.8 0.8

NM_001024720 Hmcnl 10.5 8.5 8.5 -2.1 2.0 0.0

NM_011125 Pltp 11.8 9.8 7.6 -2.0 4.2 2.2

NM_001024720 Hmcnl 10.2 8.2 8.3 -2.0 2.0 -0.1

NM_147776 Vwal 9.0 7.0 5.5 -2.0 3.5 1.5

NM_001024720 Hmcnl 10.2 8.1 8.2 -2.0 2.0 -0.1

NM_029614 Prss23 8.9 6.8 6.0 -2.0 2.9 0.8

NM_001024720 Hmcnl 10.0 7.9 8.0 -2.0 2.0 0.0

NM_001024720 Hmcnl 10.8 8.8 8.4 -2.0 2.3 0.3

NM_001024720 Hmcnl 8.3 6.3 6.3 -2.0 2.0 0.0

NM_001024720 Hmcnl 11.1 9.1 8.8 -2.0 2.2 0.2

NM 001024720 Hmcnl 11.7 9.7 9.8 -2.0 1.9 -0.1

B. ENRICHED IN LIVER 191 PROBES, 168 DESCRIBED GENES, SORT ORDER HIGEST DIFFERENCE LIVER-HEART

Liver ECs Heart ECs Brain ECs Heart - Liver Brain - Liver Heart - Brain

ANNOTATION 1 SYMBOL Probe intensity (Log2 ) Probe intensity (Log2 )

NM_001134675 Mup7 10.5 3.0 2.9 -7.6 -7.6 0.0

NM_029465 Clec4g 12.5 5.1 5.1 -7.4 -7.4 0.0

NM_008647 Mup2 10.3 2.9 2.9 -7.4 -7.4 0.0

NM_007870 Dnasell3 12.0 4.7 4.8 -7.3 -7.2 -0.1

NM_008647 Mup2 10.1 2.9 2.9 -7.2 -7.2 0.0

NM_010554 Ilia 11.2 4.1 4.2 -7.1 -7.1 -0.1

NM_031195 Msrl 11.2 4.0 3.6 -7.1 -7.6 0.5

NM_008625 Mrcl 12.2 5.1 4.8 -7.1 -7.4 0.3

NM_001045550 Mup2 9.9 2.9 3.0 -6.9 -6.9 0.0

NM_008647 Mup2 9.8 3.0 3.0 -6.8 -6.8 0.0

NM_001045550 Mup2 9.9 3.1 3.1 -6.8 -6.8 0.0

NM_008278 Hpgd 11.9 5.3 4.9 -6.6 -7.0 0.4

NM_001045550 Mup2 9.5 2.9 2.9 -6.6 -6.6 0.0

NM_001045550 Mup2 9.7 3.0 3.1 -6.6 -6.6 -0.1

NM_007553 Bmp2 11.7 5.2 6.1 -6.5 -5.6 -0.9

NM_018797 Plxncl 11.7 5.2 4.7 -6.4 -7.0 0.5

NM_138673 Stab2 12.2 5.9 5.9 -6.3 -6.3 0.0

NM_031198 Tcfec 10.2 4.2 4.1 -6.0 -6.1 0.1

NM_001077189 Fcgr2b 11.2 5.2 5.0 -5.9 -6.2 0.2

NM_008012 Akrlb8 10.6 4.7 3.8 -5.8 -6.8 0.9

NM_001163011 Mupl 8.5 2.7 2.8 -5.8 -5.8 0.0

NM_007977 F8 11.9 6.3 5.7 -5.7 -6.2 0.6

NM_009654 Alb 9.9 4.3 4.2 -5.6 -5.7 0.1

NM_010959 Oit3 11.0 5.4 5.2 -5.6 -5.8 0.2

NM_182783 Faml67b 11.7 6.2 6.5 -5.5 -5.2 -0.3

NM_008356 Ill3ra2 9.0 3.6 3.4 -5.5 -5.6 0.2

NM_020578 Ehd3 11.3 5.9 6.3 -5.4 -5.0 -0.4

AF285583 Texl8 8.4 3.0 3.2 -5.4 -5.2 -0.2

NM_177909 Slc9a9 9.6 4.3 3.9 -5.3 -5.7 0.3

NM_017399 Fabpl 8.8 3.5 3.4 -5.3 -5.4 0.1

NM_019985 Cleclb 10.2 5.3 5.0 -4.9 -5.2 0.3

NM 001136072 Meis2 10.4 5.6 5.1 -4.9 -5.3 0.5 Liver ECs Heart ECs Brain ECs Heart - Liver Brain - Liver Heart - Brain

ANNOTATION 1 SYMBOL Probe intensity (Log2 ] Probe intensity (Log2

NM_007934 Enpep 11.2 6.3 5.5 -4.8 -5.7 0.8

NM_007412 Gprl82 11.7 7.1 6.6 -4.6 -5.1 0.5

NM_010016 Cd55 11.1 7.0 5.9 -4.2 -5.2 1.0

NM_010074 Dpp4 11.9 6.3 8.0 -5.6 -3.8 -1.8

NM_177789 Vsig4 8.9 4.1 4.1 -4.8 -4.8 0.0

NM_138304 Calml4 10.7 6.1 7.0 -4.5 -3.7 -0.8

NM_008555 Maspl 9.8 5.3 5.0 -4.5 -4.9 0.4

NM_009924 Cnr2 9.3 5.0 4.7 -4.4 -4.7 0.3

NM_033314 Slco2al 10.8 6.5 6.5 -4.3 -4.4 0.1

NM_010427 Hgf 8.6 4.3 4.3 -4.3 -4.3 0.0

NM_027711 Iqgap2 9.7 5.5 5.6 -4.3 -4.2 -0.1

NM_009245 Serpinalc 8.1 3.8 4.0 -4.3 -4.1 -0.1

NM_011175 Lgmn 11.5 7.3 8.0 -4.2 -3.6 -0.7

NM_013489 Cd84 8.4 4.3 4.2 -4.2 -4.2 0.0

NM_172508 Dse 9.1 5.1 5.0 -4.0 -4.2 0.1

NM_028749 Npl 10.0 6.0 5.9 -4.0 -4.0 0.0

NM_152803 Hpse 8.5 4.5 4.3 -4.0 -4.2 0.2

NM_173036 Gpr97 10.2 6.3 6.1 -3.9 -4.1 0.2

NM_019413 Robol 9.5 5.6 5.6 -3.9 -3.9 0.0

BC027185 2210023G05Rik 9.5 5.7 6.0 -3.8 -3.5 -0.3

NM_133720 Cysltr2 7.1 3.3 3.8 -3.8 -3.3 -0.5

NA 9.8 6.1 6.0 -3.7 -3.8 0.1

NM_013474 Apoa2 9.9 6.2 6.0 -3.7 -3.9 0.2

NM_001001309 Itga8 9.3 5.7 5.4 -3.6 -3.9 0.3

NM_001159965 Ralgps2 9.4 5.8 6.4 -3.6 -2.9 -0.7

NM_011026 P2rx4 9.5 5.9 6.6 -3.6 -2.9 -0.7

NM_001033767 Gm4951 9.9 6.3 6.2 -3.6 -3.7 0.1

NM_139198 Plac8 7.5 3.9 3.7 -3.6 -3.7 0.2

NM_144830 Tmeml06a 10.1 6.6 6.2 -3.5 -4.0 0.4

NM_013737 Pla2g7 10.9 7.4 6.5 -3.5 -4.4 0.9

NA 9.9 6.4 6.6 -3.5 -3.3 -0.2

NM_007469 Apocl 8.5 5.1 5.0 -3.5 -3.5 0.0

NM_001081957 Gmll428 7.9 4.4 4.3 -3.5 -3.6 0.1

NM_145429 Arrb2 10.4 7.0 6.9 -3.4 -3.5 0.1

NM_008772 P2ryl 9.2 5.8 7.0 -3.4 -2.2 -1.2

NM_028351 Rspo3 7.6 4.3 4.4 -3.4 -3.3 -0.1

NM_030750 Sgppl 10.2 6.9 7.2 -3.3 -3.0 -0.3

NM_028810 Rnd3 11.1 7.8 8.4 -3.3 -2.6 -0.6

ENSMUST00000055842 Armcx4 9.4 6.2 7.3 -3.3 -2.1 -1.2

NM_007547 Sirpa 9.5 6.2 6.3 -3.3 -3.2 0.0

NM_008495 Lgalsl 11.9 8.8 7.5 -3.2 -4.4 1.3

BC080730 3200002M19Rik 10.5 7.4 7.1 -3.2 -3.4 0.3

NM_008867 Pla2rl 8.8 5.7 5.1 -3.1 -3.7 0.5

BC006736 3200002M19Rik 10.5 7.4 7.2 -3.1 -3.3 0.2

NM_009696 Apoe 12.4 9.3 10.2 -3.1 -2.1 -0.9

NM_138672 Stabl 10.7 7.6 7.5 -3.0 -3.1 0.1

NM_022024 Gmfg 12.0 9.0 8.2 -3.0 -3.8 0.8

NM_016923 Ly96 9.5 6.4 6.2 -3.0 -3.2 0.2

NM_133721 Itga9 10.7 7.7 6.9 -3.0 -3.8 0.8

NM_011594 Timp2 10.4 7.4 7.7 -3.0 -2.6 -0.4 Liver Ecs Heart ECs Brain Ecs Heart - Liver Brain - Liver Heart - Brain

ANNOTATION 1 SYMBOL Probe intensity (Log2 ) Probe intensity (Log2 )

NM_008096 Gc 7.7 4.7 4.8 -3.0 -2.9 -0.1

NM_009899 Clcal 8.3 5.3 4.7 -3.0 -3.6 0.6

NM_010696 Lcp2 8.8 5.8 6.0 -3.0 -2.8 -0.2

NM_008608 Mmpl4 9.5 6.6 6.0 -2.9 -3.5 0.6

NM_001012323 Mup20 5.8 2.9 2.8 -2.9 -3.0 0.0

NM_008029 Flt4 10.0 7.1 7.6 -2.9 -2.4 -0.6

NM_053247 Lyvel 9.9 7.0 6.2 -2.9 -3.7 0.8

NM_019517 Bace2 10.3 7.4 6.5 -2.9 -3.8 0.9

BC116748 9030625A04Rik 7.7 4.8 4.8 -2.9 -3.0 0.1

NM_001039544 Mup3 6.7 3.8 3.9 -2.9 -2.8 -0.1

NM_010774 Mbd4 9.8 7.0 7.1 -2.9 -2.7 -0.1

NM_177373 Ppfia2 7.8 5.0 5.0 -2.8 -2.8 0.0

NM_020282 Nqo2 9.7 6.9 7.0 -2.8 -2.7 -0.1

NM_025408 Acer3 8.7 5.9 5.8 -2.8 -2.9 0.1

NM_011662 Tyrobp 9.9 7.1 7.0 -2.8 -2.9 0.1

NM_021272 Fabp7 9.5 6.7 5.8 -2.8 -3.6 0.8

NM_026439 Ccdc80 9.8 7.0 5.4 -2.8 -4.5 1.7

NM_008599 Cxcl9 9.1 6.3 5.2 -2.8 -4.0 1.2

— NA 8.9 6.1 6.2 -2.8 -2.8 0.0

NM_133862 Fgg 6.1 3.3 3.3 -2.8 -2.8 0.1

NM_023635 Rab27a 9.9 7.2 7.5 -2.8 -2.4 -0.4

NM_031997 Tmem2 10.8 8.0 7.3 -2.8 -3.5 0.7

NM_011882 Rnasel 7.6 4.8 4.2 -2.8 -3.4 0.6

NM_001101479 Pabpc4l 7.7 5.0 4.7 -2.7 -3.0 0.3

NM_026731 Ppplrl4a 9.7 7.0 6.9 -2.7 -2.9 0.2

NM_007398 Ada 9.6 6.9 6.5 -2.7 -3.1 0.4

NM_010158 Khdrbs3 9.1 6.4 6.5 -2.7 -2.6 -0.1

— NA 5.3 2.6 2.6 -2.7 -2.7 0.1

NM_139232 Fgd4 7.4 4.7 4.6 -2.7 -2.8 0.1

NM_016667 Sntbl 9.4 6.8 5.9 -2.6 -3.5 0.8

NM_032541 Hamp 7.3 4.7 4.5 -2.6 -2.8 0.2

NM_007454 Aplbl 9.3 6.7 7.3 -2.6 -2.0 -0.6

NM_010517 Igfbp4 9.7 7.1 7.2 -2.6 -2.5 -0.1

BC150711 AI607873 7.5 4.9 4.2 -2.6 -3.3 0.7

NM_010045 Dare 7.5 5.0 5.4 -2.6 -2.1 -0.5

NM_013584 Lifr 11.8 9.2 9.6 -2.6 -2.2 -0.4

NM_009178 St3gal4 9.8 7.2 6.7 -2.6 -3.1 0.5

NM_023132 Renbp 8.5 5.9 6.0 -2.6 -2.5 -0.1

NM_024495 Carl3 7.2 4.6 4.4 -2.6 -2.8 0.2

XR_033405 Gm5970 6.9 4.3 4.1 -2.5 -2.7 0.2

NM_023537 Rab3b 7.1 4.6 4.6 -2.5 -2.5 0.0

NM_010230 Fmnl 8.0 5.5 5.4 -2.5 -2.7 0.1

NM_001014423 Abi3bp 8.4 5.9 5.5 -2.5 -2.9 0.3

NM_016751 Clec4f 8.2 5.7 5.4 -2.5 -2.8 0.3

ENSMUST00000111752 Cux2 8.7 6.2 5.9 -2.5 -2.8 0.3

BC053070 Rnfl57 8.3 5.8 5.5 -2.5 -2.8 0.4

NM_177157 Gchfr 9.0 6.5 6.1 -2.5 -2.9 0.4

NM_007853 Degsl 11.1 8.7 8.4 -2.5 -2.7 0.3

NM_018761 Ctnnall 8.5 6.0 6.3 -2.5 -2.2 -0.3

NM 172961 Abat 8.8 6.3 6.3 -2.5 -2.4 0.0 Liver Ecs Heart ECs Brain Ecs Heart - Liver Brain - Liver Heart - Brain

ANNOTATION 1 SYMBOL Probe intensity (Log2 ) Probe intensity (Log2 )

NMJ.73388 Slc43a2 7.6 5.2 5.5 -2.5 -2.1 -0.4

NMJD11815 Fyb 7.0 4.6 4.3 -2.4 -2.8 0.3

NMJD08969 Ptgsl 9.7 7.2 6.6 -2.4 -3.1 0.7

NMJD54077 Prelp 9.8 7.3 6.9 -2.4 -2.8 0.4

NMJD11243 Rarb 8.9 6.4 5.7 -2.4 -3.1 0.7

ENSMUST00000099728 Gm 10808 7.5 5.1 4.5 -2.4 -3.0 0.6

ENSMUST00000090166 Adh6b 7.7 5.3 4.4 -2.4 -3.3 0.9

NM_028275 1700112E06Rik 7.2 4.8 4.6 -2.4 -2.6 0.3

NMJ.46126 Sord 9.1 6.8 6.7 -2.4 -2.4 0.0

NMJD08092 Gata4 9.0 6.7 6.3 -2.4 -2.7 0.4

NM_026792 Agpat5 9.8 7.4 7.2 -2.4 -2.6 0.3

NM_011405 Slc7a7 9.6 7.3 6.9 -2.4 -2.8 0.4

NMJD09690 Cd5l 6.6 4.3 4.1 -2.4 -2.5 0.2

NMJ.46126 Sord 9.2 6.8 6.7 -2.3 -2.4 0.1

NMJD08268 Hoxb5 8.8 6.4 5.5 -2.3 -3.2 0.9

NMJD07574 Clqc 7.6 5.3 5.5 -2.3 -2.2 -0.2

NMJD01025577 Maf 8.3 6.0 6.2 -2.3 -2.2 -0.2

NMJD23653 Wnt2 7.8 5.4 5.4 -2.3 -2.4 0.1

NMJ.45839 Rasgeflb 7.9 5.6 4.9 -2.3 -3.0 0.6

NMJD19467 Aifl 8.0 5.7 5.4 -2.3 -2.6 0.3

NMJ.81988 Rerg 8.0 5.7 5.5 -2.3 -2.5 0.2

NM_020283 B3galtl 7.0 4.7 4.6 -2.3 -2.4 0.1

NMJL53528 Gramdlc 7.2 4.9 5.0 -2.3 -2.2 -0.1

NM_010807 Marcksll 9.9 7.7 6.3 -2.3 -3.6 1.4

NM_010353 Gsg2 7.5 5.3 5.1 -2.3 -2.4 0.2

NMJD08327 Ifi202b 7.2 5.0 4.0 -2.2 -3.2 0.9

NMJL72523 Slcl8a2 7.6 5.4 5.3 -2.2 -2.3 0.1

NMJD09692 Apoal 8.5 6.3 6.2 -2.2 -2.3 0.1

NMJD11576 Tfpi 10.7 8.5 8.7 -2.2 -2.1 -0.2

NMJD24413 Plekhfl 7.7 5.5 5.4 -2.2 -2.2 0.0

NM_024225 Snx5 11.1 8.9 9.0 -2.2 -2.1 -0.2

NM_017372 Lyz2 7.9 5.7 4.7 -2.2 -3.3 1.1

NM_022880 Slc29al 9.2 7.0 6.6 -2.2 -2.6 0.4

NM_010378 H2-Aa 8.2 6.0 5.2 -2.2 -3.0 0.8

NM_026820 Ifitml 8.4 6.3 6.4 -2.2 -2.0 -0.1

AK148800 2810055G20Rik 5.5 3.3 3.3 -2.2 -2.2 0.0

NM_029023 Scpepl 9.2 7.0 7.1 -2.2 -2.1 -0.1

NMJ.45594 Fgll 6.8 4.7 4.7 -2.2 -2.1 0.0

NMJD09658 Akrlb3 10.1 8.0 7.9 -2.2 -2.2 0.0

NMJD08663 Myo7a 8.2 6.1 5.7 -2.2 -2.5 0.4

NMJD23158 Cxcll6 8.5 6.3 5.5 -2.2 -3.0 0.8

NMJD01142701 Hmhal 7.7 5.5 5.6 -2.1 -2.1 0.0

NMJ508511 Lrmp 7.9 5.8 5.4 -2.1 -2.5 0.3

NM_008492 Ldhb 9.9 7.8 7.2 -2.1 -2.7 0.5

NM_011658 Twistl 7.7 5.6 5.5 -2.1 -2.2 0.1

NM_001162532 Faml74b 10.2 8.1 6.9 -2.1 -3.3 1.2

NM_007490 Art2a 7.1 5.0 5.0 -2.1 -2.1 0.0

NM_016673 Cntfr 8.7 6.6 6.4 -2.1 -2.3 0.2

NM_026820 Ifitml 9.7 7.6 7.6 -2.1 -2.1 0.0

NM 001037717 Slc38a6 8.6 6.6 6.6 -2.1 -2.0 -0.1 Liver Ecs Heart ECs Brain Ecs Heart - Liver Brain - Liver Heart - Brain

ANNOTATION 1 SYMBOL Probe intensity (Log2 ) Probe intensity (Log2 )

NM_001033335 Serpina3f 6.5 4.5 4.5 -2.1 -2.1 0.0

NM_175034 Slc24a5 9.0 6.9 6.7 -2.1 -2.3 0.2

NM_015753 Zeb2 9.1 7.0 6.6 -2.0 -2.4 0.4

NM_007651 Cd53 7.7 5.7 5.5 -2.0 -2.2 0.2

NM_026972 Cd209b 6.6 4.6 4.6 -2.0 -2.0 0.0

NM_008859 Prkcq 8.9 6.8 5.2 -2.0 -3.7 1.6

NM_172546 Cnksr3 9.4 7.4 7.3 -2.0 -2.1 0.1

NM_008152 Gpr65 6.0 4.0 3.9 -2.0 -2.1 0.1

NM_007646 Cd38 11.2 9.2 7.5 -2.0 -3.8 1.8

C. ENRICHED IN HEART 88 PROBES, 48 DESCRIBED GENES, SORT ORDER: HIGEST DIFFERENCE HEART-LIVER

Liver EC :s Heart ECs Br ain ECs Heart - Brain Heart - Liver Brain - Liver

ANNOTATION 1 SYMBOL |~~ Probe intensity (Log2 ) Probe intensity (Log2 )

NM_010861 Myl2 10.5 4.4 4.4 6.0 6.1 0.0

NM_022020 Rbp7 12.0 7.7 6.6 4.2 5.4 1.2

NM_007473 Aqp7 10.5 5.0 5.2 5.5 5.4 -0.1

NM_001130174 Tnnt2 11.4 6.3 6.3 5.1 5.1 0.0

NM_183175 Clqtnf9 10.2 6.3 5.1 3.9 5.1 1.1

ENSMUST00000090647 Gm 10021 10.7 6.1 5.7 4.6 5.0 0.4

NM_008509 Lpl 11.0 6.5 6.1 4.6 5.0 0.4

NM_008584 Meox2 9.2 4.1 4.3 5.1 4.9 -0.2

NM_001024712 544988 9.7 5.1 5.0 4.5 4.7 0.1

ENSMUST00000090652 544988 9.7 5.3 5.1 4.5 4.6 0.2

NM_080639 Timp4 9.6 5.2 5.1 4.4 4.6 0.1

EF651836 ENSMUSG00000C 9.6 5.1 5.0 4.5 4.5 0.0

NM_009393 Tnncl 9.9 5.9 5.4 4.0 4.5 0.6

NM_009608 Actcl 10.3 5.7 5.9 4.6 4.5 -0.2

EF651808 Hnll 10.0 5.7 5.5 4.3 4.5 0.1

NM_145983 Kcna5 9.8 6.5 5.4 3.3 4.4 1.1

NM_001024712 544988 10.0 5.7 5.6 4.3 4.4 0.1

EF651808 Hnll 9.7 5.4 5.3 4.3 4.4 0.1

EF651810 B930046C15Rik 9.7 5.4 5.4 4.2 4.3 0.1

BC172015 LOC100038847 9.6 5.4 5.3 4.1 4.3 0.1

EF651820 Gm3264 9.2 4.9 4.9 4.3 4.3 -0.1

NM_001024706 Gm5458 9.2 5.0 5.0 4.2 4.2 0.0

NM_177135 D830030K20Rik 9.8 5.6 5.6 4.1 4.2 0.1

NM_198649 Ablim3 8.6 4.5 4.4 4.1 4.2 0.1

AK049619 ENSMUSG00000C 7.3 3.4 3.1 4.0 4.2 0.2

NM_177135 D830030K20Rik 9.7 5.6 5.6 4.1 4.1 0.0

— NA 8.8 4.9 4.7 3.9 4.1 0.3

EF651835 ENSMUSG00000C 9.7 5.6 5.5 4.0 4.1 0.1

NR_002853 Aim 10.1 6.9 6.1 3.3 4.0 0.7

NM_001024706 Gm5458 8.2 4.2 4.2 4.1 4.0 -0.1

NM_134005 Enpp3 8.8 5.5 4.8 3.3 4.0 0.6

EF651825 D830030K20Rik 9.5 5.6 5.6 3.9 3.9 0.0

NM_177135 D830030K20Rik 9.5 5.6 5.6 3.9 3.9 0.0

NM_177135 D830030K20Rik 7.4 3.5 3.5 4.0 3.9 0.0

NM_008459 KlralO 7.1 3.3 3.3 3.8 3.8 0.0 Liver ECs Heart ECs Brain ECs Heart - Brain Heart - Liver Brain - Liver

ANNOTATION 1 SYMBOL Probe intensity (Log2 Probe intensity (Log2

NM_175314 Adamts9 9.7 7.0 6.0 2.7 3.8 1.1

BC151018 A430107O13Rik 10.0 7.0 6.3 3.0 3.7 0.7

NM_001024706 Gm5458 9.6 5.9 5.9 3.7 3.7 0.0

NM_175314 Adamts9 10.5 7.7 6.8 2.8 3.7 0.9

NM_011864 Papss2 8.9 5.5 5.3 3.4 3.6 0.3

NM_010651 Klra9 7.0 3.3 3.4 3.7 3.6 -0.1

NM_011598 Fabp9 7.7 3.8 4.1 3.9 3.6 -0.3

NM_177135 D830030K20Rik 6.6 2.9 3.0 3.6 3.6 -0.1

NM_001081035 Nav3 8.7 5.3 5.1 3.3 3.5 0.2

BC100412 1700001E04Rik 8.8 5.3 5.3 3.5 3.5 -0.1

NM_008482 Lambl-1 9.5 7.0 6.0 2.5 3.5 1.0

NM_177135 D830030K20Rik 7.0 3.5 3.6 3.6 3.5 -0.1

NM_001127330 Pparg 10.1 5.5 6.7 4.6 3.4 -1.2

NM_008580 Map3k5 9.4 7.1 6.1 2.4 3.4 1.0

ENSMUST00000067972 Gm9956 8.8 6.0 5.5 2.9 3.4 0.5

NM_019759 Dpt 7.8 4.5 4.5 3.3 3.3 0.0

NM_175314 Adamts9 9.8 7.5 6.6 2.4 3.2 0.9

NM_010651 Klra9 7.0 3.6 3.8 3.4 3.2 -0.2

NM_021503 Myoz2 7.2 4.0 4.0 3.2 3.2 0.0

NM_172980 Slc28a2 6.9 3.9 3.8 3.0 3.1 0.1

NM_010859 Myl3 8.8 5.7 5.7 3.1 3.1 0.0

NM_023043 Prnd 8.2 6.0 5.1 2.1 3.0 0.9

NM_175314 Adamts9 10.2 8.0 7.3 2.2 2.9 0.7

— NA 6.9 3.9 4.0 3.0 2.9 -0.1

NM_024435 Nts 6.9 3.8 4.0 3.0 2.8 -0.2

NM_010174 Fabp3 7.5 5.3 4.6 2.2 2.8 0.6

NM_026189 Eepdl 8.7 6.2 5.9 2.5 2.8 0.3

NM_178395 Zdhhc2 7.2 4.9 4.4 2.3 2.8 0.5

NM_001113415 Ebf3 8.1 5.5 5.3 2.6 2.8 0.2

NM_001101471 Akap5 8.9 6.0 6.1 2.9 2.8 -0.1

BC100412 1700001E04Rik 7.7 4.9 4.9 2.8 2.8 0.0

NM_001024706 Gm5458 7.7 4.9 4.9 2.8 2.8 0.0

ENSMUST00000104964 Gml6503 7.3 5.0 4.5 2.3 2.8 0.5

NM_007751 Cox8b 8.0 5.9 5.3 2.1 2.7 0.6

NM_009328 Tcfl5 9.1 6.9 6.4 2.1 2.7 0.6

NM_010266 Gda 8.0 5.7 5.3 2.3 2.7 0.4

ENSMUST00000049189 Adamts9 7.8 5.7 5.2 2.1 2.6 0.5

NM_013593 Mb 8.8 6.2 6.2 2.6 2.6 0.0

NM_153457 Rtnl 8.1 5.8 5.6 2.3 2.5 0.3

NM_010856 Myh6 7.7 5.2 5.2 2.5 2.5 0.0

NM_010174 Fabp3 7.5 5.5 5.0 2.0 2.5 0.5

NM_009406 Tnni3 10.2 7.5 7.7 2.8 2.5 -0.3

NR_001592 H19 7.5 5.4 5.0 2.2 2.5 0.3

NM_144783 Wtl 8.8 6.2 6.3 2.5 2.5 -0.1

NM_001024706 Gm5458 7.2 4.8 4.8 2.4 2.4 0.0

NM_001024712 544988 8.3 6.1 6.0 2.2 2.3 0.1

NM_001013785 Akrlcl9 6.9 3.4 4.6 3.5 2.3 -1.3

NM_007409 Adhl 9.8 7.8 7.6 2.0 2.2 0.2

NM_030174 Mctpl 8.3 5.8 6.1 2.5 2.2 -0.3

NM_001030305 Pmp2 6.2 3.8 4.0 2.4 2.2 -0.2 Liver ECs Heart ECs Brain ECs Heart - Brain Heart - Liver Brain - Liver

ANNOTATION 1 SYMBOL Probe intensity (Log2 Probe intensity (Log2

NM_133187 1110032E23Rik 9.3 7.3 7.2 2.1 2.1 0.0 NM_021400 Prg4 7.0 4.9 4.9 2.1 2.0 0.0

TABLE 5. Transcription factors in the tissue-related ECs fingerprints

A. TRANSCRIPTION FACTORS IN THE LIVER ECs FINGERPRINT

Liver EC Heart EC Brain EC Heart -Liver Brain -Liver Heart - Brain

ANNOTATION 1 SYMBOL llill Probe intesity (Log2 ) Probe intesity (Log2 )

NM_031198 Tcfec 10.2 4.2 4.1 -6.0 -6.1 0.1

NM_001136072 Meis2 10.4 5.6 5.1 -4.9 -5.3 0.5

ENSMUST000001117 Cux2 8.7 6.2 5.9 -2.5 -2.8 0.3

NM_008092 Gata4 9.0 6.7 6.3 -2.4 -2.7 0.4

NM_008268 Hoxb5 8.8 6.4 5.5 -2.3 -3.2 0.9

NM_001025577 Maf 8.3 6.0 6.2 -2.3 -2.2 -0.2

NM_011658 Twistl 7.7 5.6 5.5 -2.1 -2.2 0.1

NM_015753 Zeb2 9.1 7.0 6.6 -2.0 -2.4 0.4

B. TRANSCRIPTION FACTORS IN THE BRAIN ECs FINGERPRINT

Brain EC Heart EC Liver EC Heart - Brain Brain -Liver Heart -Liver

Probe intesity (Log2 ] Probe intesity (Log2

NM_010703 Lefl 10.9 6.0 5.9 -5.0 5.0 0.1

NM_009575 Zic3 9.0 4.6 4.7 -4.4 4.4 0.0

ENSMUST00000098351 Foxfla 10.7 6.4 5.9 -4.3 4.8 0.5

NM_010225 Foxf2 8.2 4.7 4.5 -3.5 3.7 0.2

NM 008592 Foxcl 7.9 5.7 5.0 -2.2 2.9 0.7

C. TRANSCRIPTION FACTORS IN THE HEART ECs FINGERPRINT

Heart ECs Brain ECs Liver ECs Heart - Brain Heart -Liver Brain -Liver

ANNOTATION 1 SYMBOL [ Probe intesity (Log2 ] Probe intesity (Log2

NM_008584 Meox2 9.2 4.1 4.3 5.1 4.9 -0.2

NM_001127330 Pparg 10.1 5.5 6.7 4.6 3.4 -1.2

NM_001113415 Ebf3 8.1 5.5 5.3 2.6 2.8 0.2

NM_144783 Wtl 8.8 6.2 6.3 2.5 2.5 -0.1

NM 009328 Tcfl5 9.1 6.9 6.4 2.1 2.7 0.6

TABLE6. Validated signature of (murine) heart endothelial cells (ECs)

Heart ECs Brain ECs Liver ECs Heart - Brain Heart -Liver Brain -Liver

AN NOTATION 1 SYMBOL Probe intesity (Log2 ) Probe intesity (Log2 )

N M_007473 Aqp7 10.5 5.0 5.2 5.5 5.4 -0.1

NM_008584 Meox2 9.2 4.1 4.3 5.1 4.9 -0.2

NM_001127330 Pparg 10.1 5.5 6.7 4.6 3.4 -1.2

N M_008509 Lpl 11.0 6.5 6.1 4.6 5.0 0.4

N M_080639 Timp4 9.6 5.2 5.1 4.4 4.6 0.1

EF651808 Hnll 10.0 5.7 5.5 4.3 4.5 0.1

N M_022020 Rbp7 12.0 7.7 6.6 4.2 5.4 1.2

N M_198649 Ablim3 8.6 4.5 4.4 4.1 4.2 0.1

N M_183175 Clqtnf9 10.2 6.3 5.1 3.9 5.1 1.1

N M_011598 Fabp9 7.7 3.8 4.1 3.9 3.6 -0.3

N M_008459 KlralO 7.1 3.3 3.3 3.8 3.8 0.0

N M_010651 Klra9 7.0 3.3 3.4 3.7 3.6 -0.1

N M_011864 Papss2 8.9 5.5 5.3 3.4 3.6 0.3

N M_001081035 Nav3 8.7 5.3 5.1 3.3 3.5 0.2

N M_145983 Kcna5 9.8 6.5 5.4 3.3 4.4 1.1

N M_172980 Slc28a2 6.9 3.9 3.8 3.0 3.1 0.1

N M_024435 Nts 6.9 3.8 4.0 3.0 2.8 -0.2

N M_001101471 Akap5 8.9 6.0 6.1 2.9 2.8 -0.1

N M_175314 Adamts9 9.7 7.0 6.0 2.7 3.8 1.1

NM_001113415 Ebf3 8.1 5.5 5.3 2.6 2.8 0.2

NM_144783 Wtl 8.8 6.2 6.3 2.5 2.5 -0.1

N M_026189 Eepdl 8.7 6.2 5.9 2.5 2.8 0.3

N M_030174 Mctpl 8.3 5.8 6.1 2.5 2.2 -0.3

N M_001030305 Pmp2 6.2 3.8 4.0 2.4 2.2 -0.2

N M_008580 Map3k5 9.4 7.1 6.1 2.4 3.4 1.0

N M_178395 Zdhhc2 7.2 4.9 4.4 2.3 2.8 0.5

N M_153457 Rtnl 8.1 5.8 5.6 2.3 2.5 0.3

N _001592 H19 7.5 5.4 5.0 2.2 2.5 0.3

NM_009328 Tcfl5 9.1 6.9 6.4 2.1 2.7 0.6

N M_023043 Prnd 8.2 6.0 5.1 2.1 3.0 0.9

N M 007409 Adhl 9.8 7.8 7.6 2.0 2.2 0.2

TABLE7. Functional annotation list for the validated heart EC signature and the brain and liver EC signatures

A. HEART EC VALIDATED SIGNATURE FUNCTIONAL ANNOTATIONS

Term Genes PVal

GO:0008289~lipid binding LPL, RBP7, FABP9, PPARG, PMP2, MCTP1 0,00 BP00028:Lipid and fatty acid transport LPL, RBP7, FABP9, PMP2 0,00 mmu03320:PPAR signaling pathway LPL, PPARG, AQP7 0,00 GO:0005215~transporter activity RBP7, FABP9, AQP7, KCNA5, PMP2, SLC28A2 0,03 BP00274:Cell communication RBP7, FABP9, NTS, PPARG, PMP2 0,05

B. BRAIN EC SIGNATURE FUNCTIONAL ANNOTATIONS Term Genes PValu

PTGS2, SGMS2, PIP5K1B, ALDH1A1, FAR2, PRKAR2B, PLA2G4A, ISYNA1,

GO:0044255~cellular lipid metabolic process NAPEPLD, HMGCS2, CH25H, MBOAT1, ELOVL7, DEGS2 0,00

LY75, LPHN3, A930038C07RIK, BSG, SERPINE2, ENPP2, SPOCK2, PGLYRPl, PTN, GO :0030246~carbo ydrate binding CD24A, FN1 0,00

CTHRC1, PTGS2, SLC39A10, ENPP2, SPOCK2, EDN1, PGLYRPl, NOTUIvl,

APCDD1, LSR, CD24A, ISM1,

signal peptide TGFB2, LPHN3, A930038C07RIK, IGF1R, APOD, SERPINE2, BCHE, TNFRSF19,

PTN, ITIH5, SEMA3C, VWA1, FIGF,

PLTP, FN1, PTPRJ, RECK, LY75, BSG, TMEFF1, PROM1, A130022J15RIK, WFDC1, LRP8, ANTXR1, CP, PRSS23

ALDH1A1, FAR2, ISYNA1, SGMS2, PTGS2, HMGCS2, CH25H, MBOAT1, ELOVL7,

GO:0008610~lipid biosynthetic process DEGS2 0,00

SLC38A3, SLC39A10, SLC9A2, SLC22A8, ATP10A, SLC19A3, SLC01A4, SLC16A1,

GO:0022857~transmembrane transporter activity TFRC, SLC7A3, SLC7A1, SLC16A9, SLC2A1, ABCC4, PLLP, SLC1A1 0,00 lipid synthesis FAR2, PTGS2, HMGCS2, CH25H, ELOVL7, DEGS2 0,00

GO:0001568~blood vessel development PTPRJ, RECK, FOXF1A, EDN 1, SEMA3C, FOXC1, FIGF, TGFB2 0,00 IPR011497:Protease inhibitor, Kazal-type RECK, SLC01A4, SPOCK2, SLC01C1 0,00 BP00016:Amino acid transport SLC7A3, SLC7A1, SLC1A1, NET1 0,00

GO:0009308~amine metabolic process A930038C07RIK, DDC, SPOCK2, MAOA, PGLYRPl, ITIH5, FOXCl, TGFB2 0,01 GO:0006066~alcohol metabolic process GPD2, DDC, ISYNAl, GBEl, HMGCS2, CH25H, MAOA, TGFB2 0,01

OCLN, SGMS2, SGPP2, SLC9A2, ATPIOA, APCDDl, LSR, LPHN3, FAR2, SLC16A1, AQP11,

CH25H, SLC2A1, SLCOICI, ELOVL7, SLClAl, PTPRJ, TMEFFl, BSG, SLC22A8, PLLP, DEGS2, SLC38A5,

GO:0016021~integral to membrane

SLC38A3, ENPP2, SLC39A10, SLC19A3, ABHD2, PAQR5, ITM2A, SLC01A4,

IGFIR, MFSD7C,

MARVELD2, SLC35F2, TNFRSF19, MYOF, LY75, TMC7, MAOA, MAOB, GOLIM4, PROMl, SLC7A3, TFRC, SLC7A1, SLC16A9, MBOATl, ABCC4, LRP8, ANTXRl

GO:0006082~organic acid metabolic process ALDHlAl, GPD2, FAR2, PRKAR2B, DDC, PTGS2, CH25H, ELOVL7, DEGS2 0,01 BP00081:Coenzyme and prosthetic group metabolism GPD2, ALAS1, APOD, HMGCS2, SLC19A3 0,01 GO:0016053~organic acid biosynthetic process FAR2, PTGS2, CH25H, ELOVL7, DEGS2 0,01 GO:0042180~cellular ketone metabolic process ALDHlAl, GPD2, FAR2, PRKAR2B, DDC, PTGS2, CH25H, ELOVL7, DEGS2 0,01 GO:0006584~catecholamine metabolic process DDC, MAOA, TGFB2 0,02 GO:0006885~regulation of pH AQP11, EDN1, SLC9A2 0,02 GO:0031012~extracellular matrix CTHRCl, A930038C07RIK, HMCNl, SPOCK2, PTN, VWAl, FNl 0,02 mmu00260:Glycine, serine and threonine metabolism ALAS1, MAOA, MAOB 0,02 MF00102:Protease inhibitor RECK, SERPINE2, SPOCK2, ITIH5, WFDC1 0,02 GO:0005975~carbohydrate metabolic process GPD2, A930038C07RIK, ISYNAl, GBEl, SPOCK2, PGLYRPl, ITIH5, FOXCl 0,02 oxidoreductase ALDHlAl, GPD2, FAR2, PTGS2, CH25H, MAOA, MAOB, CP, DEGS2 0,03 IPR003961:Fibronectin, type III PTPRJ, A930038C07RIK, IGFIR, VWAl, FNl 0,03

GO:0055080~cation homeostasis TFRC, AQP11, EDN1, SLC9A2, CD24A 0,03 BPOOllO ther receptor mediated signaling pathway FOXFIA, FOXF2, SEMA3C, FOXCl, TGFB2 0,03 BP00013:Amino acid metabolism DDC, SLC7A3, SLC7A1, SLClAl, NET1 0,04 BP00001:Carbohydrate metabolism GPD2, IGFIR, FOXFIA, GBEl, FOXF2, SLC2A1, PGLYRPl, FOXCl 0,04

C. LIVER EC SIGNATURE FUNCTIONAL ANNOTATIONS

Term Genes PValue

SCPEPl, ART2A, MASPl, LGMN, PTGSl, DNASE1L3, DSE, CIQC, WNT2, APOA2, LYZ2, BMP2, MUP1, MUP3, FLT4,

APOAl, HPSE, APOE, HAMP, ROBOl, RSP03, OIT3, IL13RA2, GPR97, LY96, LIFR, F8,

signal peptide SLC24A5, MMP14, SIRPA, PRELP, CD84, BACE2, PLA2G7, TFPI, H2-AA, GC, ₀₎₀₀

PLXNC1, CXCL9,

APOC1, CNTFR, TIMP2, FGG, ALB, SERPINA1C, FGL1, PLA2R1, TYROBP, MRC1, FAM174B, MUP7,

MUP2, CCDC80, CD5L, STAB2, HGF, LYVE1, CD55, FCGR2B, STAB1, ITGA8,

CXCL16, IGFBP4

GO:0008034~lipoprotein binding AP0A2, APOAl, MSR1, STAB1, APOE, CXCL16 0,00

BMP2, MASPl, LY96, F8, CXCL9, CIQC, CD55, STAB1, DARC, CNR2, PLA2G7,

GO:0006954~inflammatory response IGFBP4, ILIA 0,00

MRC1, LGALS1, CCDC80, STAB2, CD209B, ABI3BP, LYVE1, CLEC4F, APOE,

GO:0030246~carbohydrate binding STAB1, RSP03, CLEC4G, PLA2R1, CLEC1B 0,00 IPR016186:C-type lectin-like MRC1, LYVE1, CLEC4F, STAB1, CLEC4G, STAB2, PLA2R1, CD209B, CLEC1B 0,00 GO:0030247~polysaccharide binding LYVE1, STAB1, APOE, RSP03, CCDC80, STAB2, CD209B, ABI3BP 0,00 GO:0016044~membrane organization PLEKHF1, M RC1, MSR1, AP1B1, FCGR2B, CLEC4F, MY07A, STAB2, PLA2R1,

CD209B, SIRPA 0,00

IPR000566:Lipocalin-related protein and Bos/Can/Equ allergen MUP7, MUP20, MUP1, MUP3, MUP2, FABP1, FABP7 0,00 mmu04610:Complement and coagulation cascades FGG, CD55, MASPl, F8, TFPI, CIQC 0,00

GO:0016616~oxidoreductase activity, acting on the CH-OH group of donors, or

NADP as acceptor AKR1B8, LDHB, BMP2, SORD, AKR1B3, HPGD 0,01 GO:0006869~lipid transport APOA2, APOAl, MSR1, APOE, APOC1, SLC02A1 0,01 GO:0005044~scavenger receptor activity MSR1, STAB1, CXCL16, CD5L 0,01 GO:0030169~low-density lipoprotein binding MSR1, STAB1, CXCL16 0,01 GO:0050670~regulation of lymphocyte proliferation CD38, PRKCQ, FCGR2B, VSIG4, ADA 0,01 GO:0031012~extracellular matrix WNT2, ALB, HPSE, LGALS1, CCDC80, TIMP2, MMP14, ABI3BP, PRELP 0,01 GO:0045576~mast cell activation FYB, FCGR2B, LCP2 0,01

SLC29A1, ITGA9, P2RX4, GPR97, CLCA1, FCGR2B, STAB1, FLT4, ITGA8, SIRPA,

GO:0005887~integral to plasma membrane SLC02A1, CLEC1B 0,01 GO:0002682~regulation of immune system process CD38, PRKCQ, CD55, MASPl, FCGR2B, H2-AA, VSIG4, CIQC, ADA 0,01

MRCl, MSRl, AP1B1, FCGR2B, CLEC4F, MY07A, STAB2, PLA2R1, CD209B,

GO:0016192~vesicle-mediated transport SIRPA, RAB27A

GO:0030301~cholesterol transport APOA1, MSRl, APOE

GO:0005764~lysosome PLEKHFl, HPSE, MY07A, LGMN, H2-AA, RAB27A

GO:0042311~vasodilation WNT2, P2RX4, APOE

GO:0042157~lipoprotein metabolic process APOA2, APOA1, APOE, APOC1

GO:0002274~myeloid leukocyte activation FYB, FCGR2B, LCP2

Table 8. Primer list for qRT-PCR.

Gene

Forward primer Reverse primer Forward primer Reverse primer symbol symbol

A2M CGCCTCAGTCTCTGGAAAAC GCCTCCAGGTCAGTGAAGAG MFAP5 ATACCCCTGGGGGCTAAATAG CGTCGTAAACTGGTGAAGCA

ACE2 ATACTGTGACCCCGCATCTC ATGCTAGGGTCCAGGGTTCT MGSTl GACCTCACCCAGGTAATGGA TACAGGAGGCCAATTCCAAG

ADRB1 CGAGACCCTGTGTGTCATTG AGCACTTGGGGTCGTTGTAG MPP7 AAAAAGCCTGCATTCATTGG AGGCATAGGAGGCAACACTG

AFF3 TTTCAGTCATCAGCCAGCAGi AAGTGTTCTGGATCCGGTTG MPPED2 GGTTTAATGGATGGGGCTTT CATGGATTCCACCAAACACA

AFF3-UTR GCCGCCTGTGTATGTGTGTA ACGGTTTAGCACTGGAATGG MSX1 ACTCCTCAAGCTGCCAGAAGA AGCTCTGCCTCTTGTAGTCTCTTT

APM2 TGCTCTTGACGACTCCACAG TCAGAGGCCTGGTTAGCAGT NAV1 CCCAAAGAACTTCGGATCAA GGAGGGTGAAGCAGTCTCTG

ARL 15 AAGAACTTGGAGGGGCTGAT CTGCTGGCTTGTCTTGATGA NKX2-3 GAGCCCAAGGAACATGGAGA CTGGGCTTGCGAGAAGAG

ACTB TGGCACCACACCTTCTACAATG TAGCAACGTACATGGCTGGG NKX2-3-UTR ATTTCGTTTCGCAGCTGTCT AAAGGAAACCCGGTAACACC

CCDC3 TTGCCTCACGGAGTCAATTT CGAGGAGCACATGAGCCTA NOS1 CCCTTCAGTGGCTGGTACAT CGGAGTGATGGTCAACAATG

CDH5 GTTCACGCATCGGTTGTTC TCTGCATCCACTGCTGTCA NOTCH 1 CCACGGGCGACGTCACCC TCCACTCTGGCGGGCACG

CMBL GCAGTTGCCCAATACCAGAT TCTGGGCATGACACTGTTGT N0TCH4 ATGTCTCAATGGCGGCTCC GGAGAAGGTGCCAGGCCT

CNTN3 CCAGCAGCTAAAGGTTCGAC GTAAGTGAGACGCCCTCTGG NPR3 GGAGACCGATATGGGGATTT CACTGCCGATTCTTCTAGGC

COL4A 1 CTGGCCAGAAAGGAGAGATG TCATTGCCTTGCACGTAGAG NR2F2 CGCCTCAAAAAGTGCCTCA GCATCCTGCCCCTCTGC

COL4A2 CACCTTCCACCCAGATCAGT CTCTGGCACCTTTTGCTAGG NR3C2 AACAGGTAGACGGCGAGAGA TTTGGAATAGCACCGGAAAC

CUBN ACTGTGAAGGGGGTTCTGTG GACAGGCCCCACAGTAGAAA NRCAM TCACCATTGTGGACCAAAGA CTCTGGGAGGACATTGGAAA

CYSL TR1 TGAGAACAAACGCAAAAGGA CTTGATTGCGGAAGTCATCA NRP2 GCGAGTGGATTGTTTACGCC CAGTCTTTGCCCAGGAGGTC

DLL4 ATGACCACTTCGGCCACTATG GCCCGAAAGACAGATAGGCTG ODAM CAGGCCAAGTTGATCCCTTA TGAGGTTGTTCCCAGGGTAG

EFNB1 GTTCTCGACCCCAACGTGTT CAGGCTTCCATTGGATGTTGA OLFML3 AACGCCGACTAGCTGCTTTA GCTCCAGACGATCCACTCTC

EFNB2 CTGCTGGATCAACCAGGAAT TCCGGTACTTCAGCAAGAGG PAPLN GGCAAGAGGGATGTCTGTGT AGCTCGGCTGAGGTCATTAG

EMX2 ACCTTCTACCCCTGGCTCAT AGCCTTCTTCCTCCAGCTTC PDE4D CCAGTCTGCGAACTGTACGA TGCCTGGTCTGTAGGGTCTC

ENG TGCCACTGGACACAGGATAA CCTTCGAGACCTGGCTAGTG PEC AMI TCTGCACTGCAGGTATTGACAA CTGATCGATTCGCAACGGA

EPHB4 GAGCTGTGTGGCAATCAAGA GAACTGTCCGTCGTTTAGCC P RDM 16 CAAACGCTTCGAATGTGAAA CGTGTAGGACTTGTGGCAGA

FAM19A5 CGCAGTTCCTCAAAGAAGGT GACACGGAAGCATGTCACAC PRDM16-UTR CTGCAAAACACTTGCCTGAA GACCAGGAGCAGCTATGTCC

FAP CCAGGAGATCCACCTTTTCA ACGCAGGGTAAGTGGTATCG PSMAL TTGGAATCTTCCTGGAGGTG CTGCTATCTGGTGGTGCTGA

FAT1 GTGGAAAAGGGGACAGTGAA CCTGATCGGTTGCAAAGACT PTPRR TTGTCTGCCAGCTTCGTATG GGGCTTCAGAGCTTCTCCTT

FGL2 TTGGATGGCAAATGTTCAAA CCATGGTCTCCATGTCACAG RASGRF2 ACCTTGCCATCGAAAGAATG ACAGAACCCCACCTGTCTTG

FLRT3 ACAGTGTATCCTGCCCAAGG CTGCGTTCCCCTGTTACAAT RBPJ GGCAGTGGATGGAAGAAAA CTTTTATCCGCTTGCTGAGG

FOLH1 TTGGAATCTTCCTGGAGGTG CTGCTATCTGGTGGTGCTGA RUNDC3B GTATCTGCAGCATTGAAAATATGG CAATAGCATTGAGTCCTAGAAGCA

FREM1 ATGGATGTAGTGGGGCAGAC GTCAGCTGAAGGAGGTCAGG RYR3 AAATTGCTGGCTGCTCTCAT TTAAGGCTTCTGGGCTTTCA

GALNTL4 CGACAAGAAGCTGGAGGAAG GGTAGGCGTTGTAGCCGTAG SCARA3 CGCTGCCAGAAGAACCTATC CAGCTCCTCCTGCAGTTTTC

GAPHD TGGTATCGTGGAAGGACTCATGAC ATGCCAGTGAGCTTCCCGTTCAGC SEMA3C TCATTCCATGATTGCTCGAA CCTCGGGTTATCAGTTTCCA

GLIPR2 GCTCTGCAAGAACCTCAACC ATACCATGGCCGTGAAGTGT SEMA3G ACGGAGCACAATAGCACCTT GACCACAGTCTGGGAGAAGC

GRB14 AAATCCCACTGAAGCCCTTT CCCGTACCAAGAAAACTCCA SLC2A 1 CCTGCAGGAGATGAAGGAAG ACAGCGACACGACAGTGAAG

H†9 GAGCTCTCAGGAGGGAGGAT CCAGCCTAAGGTGTTCAGGA SLC2A3 TGGTTATTGGCCTCTTCTGC AAGGGCTGCACTTTGTAGGA

HECW2 TCCAGCATTCCCTATGAAGG GCTTCTGAAACTGCCTGTCC SNTB2 GCAAAGATACAGCCACAGCA TGTCCACGGCATACAGTCAT

HEY1 GAGAAGGCTGGTACCCAGTG GCTCAGATAACGCGCAACTT S0X17 CAGCAGAATCCAGACCTGCA CAGCGCCTTCCACGACTT

HEY2 CCCCTGCGAGGAGACGA ATCTAATCACAGAGCTAGTACTTTGCCC S0X17-UTR TGACTCCGGTGTGAATCTCC GCAACAACAAAAACCCAGGA

HEY2-UTR TTCAAGGCAGCTCGGTAACT CAGGCACTTACGAAACACGA S0X18 AGAACCCGGACCTGCACA CAGCTCCTTCCACGCTTTG

HSF2BP TCGTTAGACGGTGATGTCCA CCCAGAAGCTGCAATATGGT S0X7 GGCGGCCCATGAACG TCCACGTACGGCCTCTTCTG

JAG 1 CCAATGACTGCAGCCCTCAT GCTCCAAAGGCACAAGGTGA SYTL2 TGTTTTTGTGGCCCAGTGTA AGTTCCACCTCCCCTAGGAA

JAG2 TCATCCCCTTCCAGTTCGC AGGCTCTTCCAGCGGTCCT TFPI2 GTCGATTCTGCTGCTTTTCC CACTGGTCGTCCACACTCAC

KDR TGGCATCGCGAAAGTGTATC AAAGGGAGGCGAGCATCTC TEK ACACCTGCCTCATGCTCAGC AGCAGTACAGAGATGGTTGCATTC

KLHL6 TTTGAGACCGTGATGAGCTG CTCATTGCCAGAAAGGTGGT TMEM200A GAGCAGCATTTGCATTCTGA ACCAGTGTACATGCCGTTCA

LAMA2 AGGTGAATGTGGAAGGCATC GGGGTAGAATGGTCTGCTCA TMEM200C CAGTAGCAGTGGCAGCAAAA CCGAAGACCTTGAGCTTGTC

LMCD1 CCAAGAGGACCACTGCCTAA GTAAAAGGCACCCTCTGTGC T0X2 TCAGGAAGAGGAGTCGGAAG CACGATTTTGGACACGTCAC

LOC40 W22 TCAGGTCCCTGGATAAGGTG GTGAGAGCTTTCCCTGCAAC T0X2-UTR ATCTCTGAGTTCCCCAGCAG TTTTCTGCTGTCAGCCCTTC

LRCHi GAGCATACAAGCATGCCAGA TGGCACATTGAGGCTAACAG VWF TGCTGGTATGGAGTATAGGCAGTG CCGGAATGCACGCAGG

MAP9 ACATGGAGGAGAAGGATGGA CAGATGCGTTTCCCTCAGAT XG AGCTGGGAGACCAGAAGTCA GGCATCTGCCAAATCAAAGT

Table 9. Nanostring probelist

Probe set ID Gene J^{arge e} Target sequence

d region

Arterial genes fron the arteriovenous fresh profile

N _003480.2 MFAP5 55-155 CTCATCTCATTGTTTCAGCGGAGGCCAAATCTGAAGTCCTTTCCAGGGAGTGGCTCTGTTCATCTTATTCGCCAGCCAAAGTAGGAACAGCGTAAGAGGA

NM_01 5381 .3 FAM19A5 320-420 GCCGGCACCACGAGAGCCCGGCCCGCCTGTGTGGACGCAAGAATCATCAAGACCAAGCAGTGGTGTGACATGCTTCCGTGTCTGGAGGGGGAAGGCTGCG

NM_1 30846.1 PTPRR 670-770 GTCCAACGTATCTCTTACATTGGACATGAGTAGCTTGGGGAACATTGAACCCTTTGTGTCTATACCAACACCACGGGAGAAGGTAGCAATGGAGTATCTG

NR_0021 96.1 H19 1592-1692 GGCCTTTGAATCCGGACACAAAACCCTCTAGCTTGGAAATGAATATGCTGCACTTTACAACCACTGCACTACCTGACTCAGGAATCGGCTCTGGAAGGTG

N _006682.2 FGL2 250-350 CAATTCAGCAGGATCGAGGAGGTGTTCAAAGAAGTCCAAAACCTCAAGGAAATCGTAAATAGTCTAAAGAAATCTTGCCAAGACTGCAAGCTGCAGGCTG

NM_000908.2 NPR3 35-135 TTGCAGAGAAGGACGCTTCCTCTCTATCTTTTGGCGCATTAGTGAAGGGGGTATTCTATTTTGTTAAAGCGCCCAAGGGGGCGCAGGGACCTTGGAGAGA

NM_020872.1 CNTN3 3030-3130 CCAGTAGTGAACAGATCAGGATTCCACGAATAACCAGTATGGATGCAAGAGGATCCACTTCAGCCATCTCGAATGTCCACCCTATGTCAAGTTATATGCC

NM_006909.1 RASGRF2 2675-2775 GAGTCCACCAGGCTTTAACAACACCGAGAGAACATGTGATAAAGAGTTTATTATACGGAGAACGGCTACCAATCGAGTTCTGAACGTCCTCCGTCACTGG a NM_O201 63.1 SEMA3G 2591 -2691 CCCAAGGGATCTGCCATTTGTTCTCAGAGATGGCCTGGCTTCCGCAACACATTTCCGGGTGTGCCCAGAGGCAAGAGGGTTGGGTGGTTCTTTCCCAGCC

4455-4555 AACCCGCACTGTGCCTGGGTGTATTCTTTATACTGTAGATAATGGAGAAATTTTCTATCTCTGTCCCTATTTGTATAAGCCAAGGTGATGCTGGGTGCCC

550-650 TGAAACCAGAGGACTTTGACAATGTTCAGTCCCTCCTGACAAACAGTATTTACTTACAAGATTCGGAGGTAACAGTGAAGGGATTCAGGATATACGGTGC

1685-1 785 TCCATCTCAATCCCTGTGAAGTCAGACATTGCTCCTGTCGCTCGGTTGCTCATCTATGCTGTTTTACCTACCGGGGACGTGATTGGGGATTCTGCAAAAT

765-865 TCCGAGGACCTGCTAAAATCAGCTACTAGAATCTGCTGCCAGAGGGGACAAAGACGTGCACTCAACCTTCTACCAGGCCACTCTCAGGCTCACCTTAAAA

3855-3955 GGCAAACTCAAGTATGCAATCTATTTCGAGGCTCGGGAAGAAACAGGTTTCTCTACATATAATCCTCAAGTGATCATTCGAGGTGGGACACCTACTCATG

3355-3455 CCTTGTGGTTCCTTTGCCCAACACCAGTGAATCCTTCCAGCCCGTCAGCACAGTGCTACCAAGGAATAATTCCATTGGGGAGTCGTTGTCGAGTCAGTAC 795-895 TGTGCATCATGGCCTTCGTGTACCTGCGGGTGTTCCGCGAGGCCCAGAAGCAGGTGAAGAAGATCGACAGCTGCGAGCGCCGTTTCCTCGGCGGCCCAGC

600-700 TTTAATCCAAGATACAGAACCTGTGATGCTTTCACCTATACTGGCTGTGGAGGGAATGACAATAACTTTGTTAGCAGGGAGGATTGCAAACGTGCATGTG

1374-1474 AGCCACGGGGCCATTTCCTCGGTGGTGTCCGACGCCAGCTCCGCGGTATATTACTGCAACTATCCTGACGTGTGACAGGTCCCTGATCCGCCCCAGCCTG

35-1 35 GGCTGGGGGCTTGTCGCCCTTTCAGGCTCCACCCTTTGCGGAGATTATAAATAGTCATGATCCCAGCGAGACCCAGAGATGCTGTAATGGTAAGACTTTG

1760-1860 CATCATGTGTTTCACACCACAATTCTGTGCCACAGTTAAGAGGGTCTGGTACCCTTGCAGGACCTTTGTAGGTTGTGGGAAAAAGTCGCAGAAAGATACT

3365-3465 CCGGAACTGTCATTGCCAGACAGAATTGTACCTTTAGCACATTCTGGTGCCCTGGCCTCTTAGTGTTCACCAGTTGTACTGCAATATAGACAAGATGTCT

1646-1746 CTCCGCGAAGCGTTGGCGGGGAGCCCAAGGACATAACAAATTAAAAGCATGAAGGAGAGAAAAATGGGGGTCGTGGCTTGAGAAATTCCAGGCCCTACCG

3895-3995 GAAAAGACCTGCAGAGAAATAAGCTATATGTCACCTTCGTTGGGGAGGAAGGGCTGGATTACAGTGGGCCTTCTAGAGAGTTTTTCTTCCTGGTATCCAG

1740-1840 GAGGTCAGACCTTGGACAGTAGTCTTGACTTCCTGCTATAGAACACATTGTTAACACTGAAAAAGATGATCTGTTCTAGGGGAATGGTGAAAGCTGACTC

1942-2042 GCCATTCTGAGAGGAGACATGATGAGGTAGAATTGGAGGTATACGACTTCTTCTTTGAAAGGAGTGCACCTATGACAGTCCACATCTCCATCAGAACAGC

1 150-1250 GGCATTTCCTTGAAGGGAGAAGAAGGAATCATGGGCTTTCCTGGACTGAGGGGTTACCCTGGCTTGAGTGGTGAAAAAGGATCACCAGGACAGAAGGGAA

NM_012259.2 HEY2 1575-1675 GAGAGAGAGTGAGAGAGTCGTGTTTCGTAAGTGCCTGAGCTTAGGAAGTTTTCTTCTGGATATATAACATTGCACAAGGGAAGACGAGTGTGGAGGATAG

NM_00501 0.4 NRCAM 875-975 CATTGTTGTCCGCCCATCCAGATCACCATTGTGGACCAAAGAAAAACTTGAACCAATCACACTTCAAAGTGGTCAGTCTTTAGTACTTCCCTGCAGACCC

N _001 845.4 C0L4A 1 780-880 TGGGCTTAAGTTTTCAAGGACCAAAAGGTGACAAGGGTGACCAAGGGGTCAGTGGGCCTCCAGGAGTACCAGGACAAGCTCAAGTTCAAGAAAAAGGAGA

NM_006379.2 SEMA3C 945-1045 TGTCTGTGGGAGTGGCGCTTTCAGTCCTGTCTGTACTTACTTGAACAGAGGGAGGAGATCAGAGGACCAAGTTTTCATGATTGACTCCAAGTGTGAATCT

NM_004460.2 FAP 1490-1590 GCATTGGAAGCTATCCTCCAAGCAAGAAGTGTGTTACTTGCCATCTAAGGAAAGAAAGGTGCCAATATTACACAGCAAGTTTCAGCGACTACGCCAAGTA

NM_001 098798.1 T0X2 784-884 GAAGCCTGTGTCGGCCTACGCACTCTTCTTCAGAGACACTCAGGCCGCCATCAAGGGTCAGAACCCCAGTGCCACTTTCGGTGACGTGTCCAAAATCGTG

NM_002253.2 KDR 1420-1520 CAATCACACAATTAAAGCGGGGCATGTACTGACGATTATGGAAGTGAGTGAAAGAGACACAGGAAATTACACTGTCATCCTTACCAATCCCATTTCAAAG

NM_1 75569.2 XG 27-127 AGCTGTGGAGTTTGGGATCTGAGCTTGGAGCCCATTTGTTTCTGGCAGTTCCGCTCATATTTTCCACTTGAAGACATCGCCTCCCTTCCTTCCAAGCTGG

NM_005245.3 FAT1 780-880 ACCCAACCAGTGGTGTGATAGTGTTAACTGGTAGACTTGATTACCTAGAGACCAAGCTCTATGAGATGGAAATCCTCGCTGCGGACCGTGGCATGAAGTT

NM_001 01 4986.1 F0LH1 1530-1630 GAAGAATTTGGTCTTCTTGGTTCTACTGAGTGGGCAGAGGAGAATTCAAGACTCCTTCAAGAGCGTGGCGTGGCTTATATTAATGCTGACTCATCTATAG

N _1 9851 6.1 GALNTL4 995-1095 ACAGCAGTAACGAGGAACTGAAGGAGAAGCTGACCGAATATGTGGACAAGGTGAACAGCCAGAAGCCAGGCTTCATCAAAGTCGTGCGTCACAGCAAGCA

NM_001 036.3 RYR3 3655-3755 TCTCAGATCGGCCGCATGAATCTCGGGACAGATGCCAGTACCTTCAAGTTTTATACCATGTGCGGTCTCCAAGAGGGCTTTGAGCCTTTTGCTGTCAACA

N _001 0251 08.1 AFF3 1 174-1274 AAGCACCTGGCAAAGTGGAACCAACCAAATTTCCATTTCCAAATAAGGACTCTCAGCTTGTATCCTCTGGACACAATAATCCAAAGAAAGGTGATGCAGA

NM..004557.3 NOTCH4 6420-6520 ACCCCTAAGTTGGAACCAAGAATTGCAGGCATATGGGATGTAAGATGTTCTTTCCTATATATGGTTTCCAAAGGGTGCCCCTATGATCCATTGTCCCCAC

NM. .004098.3 EMX2 2505-2605 ACATTCCCTTTCCTAACATCCTGAGGCTTAAAACCCTGATGCAAACTTCTCCTTTCAGTGGTTGGAGAAATTGGCCGAGTTCAACCATTCACTGCAATGC

NM. .001 1 67738.1 NAV1 294-394 CTCAGTTCAGAAGAATTCAATGCCAGCTCCTCACTCAACTCCCTCCCAAGTACTCCCACTGCTTCTCGCAGGAACTCAACAATAGTGCTACGCACAGACT

NM. .01 9087.2 ARL 15 75-175 GGATGCTTTAGGCTGCCGGATGTCTGATCTCCGAATAACTGAGGCGTTTCTGTACATGGATTATCTGTGTTTTAGAGCACTTTGCTGCAAGGGACCACCA

NM. .001 039580.1 MAP9 547-647 ATAAATCAAACGGTAACATAACCAAAGATGAGCCAGTGTGTGCCATCAAAAATGAAGAGGAAATGGCACCTGATGGGTGTGAAGACATTGTTGTAAAATC

NM. .022343.2 GLIPR2 21 6-316 CCAGTGTGGGGAGAACCTTGCATGGGCATCCTATGATCAGACAGGAAAGGAGGTGGCTGATAGATGGTACAGTGAAATCAAGAACTATAACTTCCAGCAG

NM. .001 1 93329.1 APO 1375-1475 AGACCCTTCTGCGGCTGATCCCTCCTTGCCTCTCAGCAGCACATTCTGTTCTGGGAGCACACCCGTTCTCTCGGCTGGATGTTCTCATCGTCCCTGCCAA

NM. .01 51 1 6.2 LRCH1 71 5-81 5 CTGCCTGTGTGGTCTGCCTCTCAAAGTCTTAATCGCAAGTAACAACAAACTTGGATCATTACCAGAAGAGATAGGTCAGCTCAAACAGTTAATGGAGCTG

NM. .002448.3 MSX1 1225-1325 GGCACCGCCAGCCGCCTTCCCTTTAACCCTCACACTGCTCCAGTTTCACCTCTTTGCTCCCTGAGTTCACTCTCCGAAGTCTGATCCCTGCCAAAAAGTG

NM. .00651 6.2 SLC2A 1 2500-2600 AGGCTCCATTAGGATTTGCCCCTTCCCATCTCTTCCTACCCAACCACTCAAATTAATCTTTCTTTACCTGAGACCAGTTGGGAGCACTGGAGTGCAGGGA

NM¹. .004490.2 GRB14 980-1080 ACAGCTGGACCCTTTTTGAGCACCTGCCTCACATAGGTGTAGAAAGAACAATAGAAGACCACGAACTGGTGATTGAAGTGCTATCCAACTGGGGGATAGA

NM. .01 7855.3 ODAM 445-545 TCAATACTATCCAGTTTACATGGTCCTACCCTGGGAACAACCTCAGCAAACAGTTCCAAGGTCACCTCAACAAACAAGACAGCAACAGTATGAGGAGCAG

NM. .031 455.3 CCDC3 470-570 ACCGTGGTCCAGGACTACTCCTATTTCTTCTTCCTCAGGATGGATGAAAATTATAACCTCTTGCCTCACGGAGTCAATTTCCAAGATGCCATCTTCCCAG

NM. .001 081 .3 CUBN 8930-9030 TGACTTTTGTGTCCTTCCACTTAGAAGCTCGTTCCGCTGTGACGGGAAGCTGTGTCAACGATGGCGTGCACATTATCAGAGGTTACAGCGTCATGTCCAC

NM. .001 1 97221 .1 PDE4D 21 20-2220 ATTGTTTCCAAAGTGCATGTCACATGCCACAACCACGGTCACACCTCACTGTCATCTGCCAGGACGTTTGTTGAACAAAACTGACCTTGACTACTCAGTC

CO NM. .001 1 34406.1 RUNDC3B 929-1029 GCTTCTAGGACTCAATGCTATTGATTTCAGTTTCTGCCTAAAGGGAGAGGGGCTGGATGGCAGTTTTCCTGCTGTAATAGACTATACACCATATTTGAAG

NM. .006639.2 CYSLTR 1 90-190 AAAGAAGCTTGCCTATAGAGCAGGCACTCTGTGAATGGACTGTGCTTTTACGACCCTACAGGGTATCAAGATACTGTGCAGCTCGCCAACAAGGATTAAT

NM. .021 804.2 ACE2 1825-1 925 GGCCCTCTGCACAAATGTGACATCTCAAACTCTACAGAAGCTGGACAGAAACTGTTCAATATGCTGAGGCTTGGAAAATCAGAACCCTGGACCCTAGCAT

NM. .001 080392.1 LCHN 4440-4540 CCAGGTGGGAGTGGAATTCAGGTTTGGGGCTCGTTGGTATCCATGCAAAATATGACAAAGGCCTGTTCAAGAGGGCATTTTCAATTCTGTAGGCTCAGCA

NM. .01 4583.2 LMCDl 1470-1 570 TTGTTTCCCAGCTACAACCAACTAAAGACACAAATGGCGTTCTGCAAGGGGACTCTGGGAGGAGTTTTCCAGAATGCAATTCCGAGTGAGCAAATCGCAT

NM. .007031 .1 HSF2BP 800-900 ACTGGTCAAACAATGGAGAGTTTTGTGAAGTCGTTAGACGGTGATGTCCAGGAGCTGGATTCGGATGAAAGTCAGTTTGTTTTCGCTCTGGCTGGAATTG

NM. .001 1 35032.1 F AM 176 A 510-610 CGTGGAGATGGCTTTGCTCAGCAACATCCTAGCGGCCTATTCCTTTGTCTCAGAAAATCCTGAGCGAGCAGCTCTGTACTTTGTTTCTGGCGTGTGCATC

Venous genes from the arteriovenous fresh profile

NM_000901 .3 NR3C2 430-530 CCTACAGAGAGGACCGATGAGAATAACTACATGGAGATTGTCAACGTAAGCTGTGTTTCCGGTGCTATTCCAAACAACAGTACTCAAGGAAGCAGCAAAG

NM_01 6240.2 SCARA3 280-380 AGAACCTATCTTTGCACACATCGGTGCGGATTCTTTACCTCTTCCTGGCCCTGCTCCTGGTGGCCGTGGCTGTGTTGGCCTCTCTGGTTTTCAGAAAAGT

CD NM_1 73462.3 PAPLN 424-524 CCTACTACAGCGCCCCAAACAAGTGTGAACTGAACTGCATTCCCAAGGGGGAGAACTTCTACTACAAGCACAGGGAGGCTGTGGTTGATGGGACGCCCTG

NM_006750.3 SNTB2 2645-2745 GTTTGAACAAGATAAATCCTAAAGAGCCAGCCTGCTTGAGGAGTAGACTTGGTGGGTGAAGCCAGCAATTCCGCACAAACGTCATGTTGAATTGTTTTGG

NM_001 042521 .1 MGC13057 550-650 ATCTTGGAATATGCACACCGCCTGTCTCAGGATATCTTGTGTGATGCCTTGCAGCAATGGGCATGCAATAACATCAAGTACCATGACATTCCATACATTG

NM_01 3281 .2 FLRT3 1 145-1245 GACAATATAACACAACTGATTCTTCGCAACAATCCCTGGTATTGCGGGTGCAAGATGAAATGGGTACGTGACTGGTTACAATCACTACCTGTGAAGGTCA

NM_j006829.2 APM2 8-108 CCCTGGGGCGCCTTAAAAACCGGAGCTGGCGCTTGGCATCGCCACTCTGGGCAGGATCCAACGTCGCTCCAGCTGCTCTTGACGACTCCACAGATACCCC

NM_1 38809.3 CMBL 2425-2525 TTGTCTCCTAAGCAGTGGATTTCCAAGCACTGGCATTTTTGGGTCCCAGGGCCTGAGGCTACCTCAGAAATATCTCCTGTGTTTCTGGTGGTGGCGTAAA

NM_0201 90.2 OLFML3 1 525-1625 GCCCGAAGAGTCAAAACCCTCAATGTTCCCTCCTGCTCTCCTGCCCCATGTCAACAAATTTCAGGCTAAGGATGCCCCAGACCCAGGGCTCTAACCTTGT

NM_1 45764.1 MGST1 330-430 TCTTGGAATTGGCCTCCTGTATTCCTTGAGTGGTCCCGACCCCTCTACAGCCATCCTGCACTTCAGACTATTTGTCGGAGCACGGATCTACCACACCATT

NM_1 73496.3 MPP7 573-673 TACACTACTATGAGAAGCAGAGTCCGGTGCCCATTCTCCATGGTGCGGCGGCCTTGGCCGATGATCTGGCCGAAGAGCTTCAGAACAAGCCATTAAACAG

NM_001 20421 3.1 N0S1 1201 -1 301 GGCGCTGGTGGAGATCAATATCGCGGTTCTCTATAGCTTCCAGAGTGACAAAGTGACCATTGTTGACCATCACTCCGCCACCGAGTCCTTCATTAAGCAC

Established arterial EC genes

NM_01 841 9.2 S0X18 1475-1 575 CCTGCGAGGGTGCCTGGAGTTCCCACGTGTCCCGGGGCTTTTCCAGGAAGCCCGAGCCCAGGACCTGTTGGCAGAGTTGCCAGGGTTACATTTTTGAAGC

NM_004093.2 EFNB2 620-720 TTGTAAAACCAAATCCAGGTTCTAGCACAGACGGCAACAGCGCCGGACATTCGGGGAACAACATCCTCGGTTCCGAAGTGGCCTTATTTGCAGGGATTGC

NM_01 761 7.3 NOTCH! 735-835 CTGCCAGGCTTCACCGGCCAGAACTGTGAGGAAAATATCGACGATTGTCCAGGAAACAACTGCAAGAACGGGGGTGCCTGTGTGGACGGCGTGAACACCT

N _000214.2 JAG1 915-101 5 TTGCTTGTGGAGGCGTGGGATTCCAGTAATGACACCGTTCAACCTGACAGTATTATTGAAAAGGCTTCTCACTCGGGCATGATCAACCCCAGCCGGCAGT

NM_004429.4 EFNB 1 137-237 CGAGGCTTCGGGGGCGCAAACTAATGGGACTGGCTCGCTCGGCAGCATCTCCCCGCTCTTCTAAGTACACTGAGCAGGGCCCGCGCTGAAGTAGAAGCTG

NM_01 9074.2 DLL4 893-993 AATGACCACTTCGGCCACTATGTGTGCCAGCCAGATGGCAACTTGTCCTGCCTGCCCGGTTGGACTGGGGAATATTGCCAACAGCCTATCTGTCTTTCGG

Targete

Probe set ID Gene Target sequence

d region

Established arterial EC genes continued

NM_1451 59.1 JAG2 4225-4325 ATTTTTGTAAAGTTTCCGTGCGTGGCACTCGCTGTATGAAAGGAGAGAGCAAAGGGTGTCTGCGTCGTCACCAAATCGTAGCGTTTGTTACCAGAGGTTG NM_001 001 392.1 CD44 429-529 ACACCATGGACAAGTTTTGGTGGCACGCAGCCTGGGGACTCTGCCTCGTGCCGCTGAGCCTGGCGCAGATCGATTTGAATATAACCTGCCGCTTTGCAGG NM_031 439.2 S0X7 2635-2735 CTGTGAGAATTTGTCTTCCTCACCAGCCAGGTCCTCAGGCAAAGTCCTCAGCCAGTGCTTTAGAGCAACTTCCCGCAAATCAGAAACTCACTGTGATTCC NM_01 2258.3 HEYI 398-498 AACAGTTTGTCTGAGCTGAGAAGGCTGGTACCCAGTGCTTTTGAGAAGCAGGGATCTGCTAAGCTAGAAAAAGCCGAGATCCTGCAGATGACCGTGGATC

Established venous EC genes

NM_201 264.1 NRP2 805-905 TCTCACCTGGGTTTTCTTAGCCCTCTACTTTTCAAGACACCAAGTGAGAGGCCAACCAGACCCACCGTGCGGAGGTCGTTTGAATTCCAAAGATGCTGGC NM_021 005.2 NR2F2 1 530-1 630 CCATAGTCCTGTTCACCTCAGATGCCTGTGGTCTCTCTGATGTAGCCCATGTGGAAAGCTTGCAGGAAAAGTCTCAGTGTGCTTTGGAAGAATACGTTAG NM_004444.4 EPHB4 1680-1 780 GTCCTGACTTCACCTATACCTTTGAGGTCACTGCATTGAACGGGGTATCCTCCTTAGCCACGGGGCCCGTCCCATTTGAGCCTGTCAATGTCACCACTGA

General EC genes

NM_000442.3 PEC AMI 1365-1465 ATCTGCACTGCAGGTATTGACAAAGTGGTCAAGAAAAGCAACACAGTCCAGATAGTCGTATGTGAAATGCTCTCCCAGCCCAGGATTTCTTATGATGCCC NM_000552.3 VWF 81 1 5-8215 CACCTGCAACCCCTGCCCCCTGGGTTACAAGGAAGAAAATAACACAGGTGAATGTTGTGGGAGATGTTTGCCTACGGCTTGCACCATTCAGCTAAGAGGA NM_000459.2 TEK 615-71 5 CGAGTTCGAGGAGAGGCAATCAGGATACGAACCATGAAGATGCGTCAACAAGCTTCCTTCCTACCAGCTACTTTAACTATGACTGTGGACAAGGGAGATA NM_001 795.3 CDH5 3405-3505 TCTCCCCTTCTCTGCCTCACCTGGTCGCCAATCCATGCTCTCTTTCTTTTCTCTGTCTACTCCTTATCCCTTGGTTTAGAGGAACCCAAGATGTGGCCTT NM_001 1 1 4753.1 ENG 1480-1 580 GTCCTTGATCCAGACAAAGTGTGCCGACGACGCCATGACCCTGGTACTAAAGAAAGAGCTTGTTGCGCATTTGAAGTGCACCATCACGGGCCTGACCTTC

Housekeeping genes

NM_0001 94.1 HPRT1 240-340 TGTGATGAAGGAGATGGGAGGCCATCACATTGTAGCCCTCTGTGTGCTCAAGGGGGGCTATAAATTCTTTGCTGACCTGCTGGATTACATCAAAGCACTG NM_001 1 01 .2 ACTB 101 0-1 1 10 TGCAGAAGGAGATCACTGCCCTGGCACCCAGCACAATGAAGATCAAGATCATTGCTCCTCCTGAGCGCAAGTACTCCGTGTGGATCGGCGGCTCCATCCT N _022551 .2 RPS18 256-356 TGCAGAATCCACGCCAGTACAAGATCCCAGACTGGTTCTTGAACAGACAGAAGGATGTAAAGGATGGAAAATACAGCCAGGTCCTAGCCAATGGTCTGGA NM_002046.3 GAPDH 35-135 TCCTCCTGTTCGACAGTCAGCCGCATCTTCTTTTGCGTCGCCAGCCGAGCCACATCGCTCAGACACCATGGGGAAGGTGAAGGTCGGAGTCAACGGATTT

TABLE10. Arterial TF cloning information

Gene Forward Primer (5'->3') Reverse Primer (5'->3') Cloned Restriction Backbone Symbol mRNA enzymes

sequenc used

e (bases)

hHey2 Genecopoeia (ref.

EX-U0515-L.V1 14) mPRDMI CCATCTAGAATGCGATCCAAGGCGAGGG TCACTCGAGTCAGAGGTGGTTGATGGGG 1 15-3770 Xbal-Xhol pRRL2-GPK-Cher 6 mAFF3 AG G G CTAG C ATGG AC AGCTTCG ACTTAG C TGAGCTAGCCTAGGACAGGTGTACGCTGC 202-3888 Nhel pRRL2-GPK-Cher hTox2 AAGTCTAGACCTACGTGGGGATGAGTGAC GTCCTCGAGCTCTGAGCAGTGCCTTCCA 209-1663 Xbal-Xhol pRRL2-GPK-Cher hMsxl TTTTGCTAGCATGGCCCCGGCTGCTGACA GGGCTCGAGCTATGTCAGGTGGTACATGC 236-1 147 Nhel-Xhol pRRL2-GPK-Cher hEmx2 CCCCGCTAGCATGTTCCAGCCGGCGCCCA AAACTCGAGTTAATCATCTGAGGTCACGT 824-1582 Nhel-Xhol pRRL2-GPK-Cher

A

hNkx2-3 CCCCAGCTAGCATGATGTTACCAAGCCCG TAGCTCGAGCTACCAGGCCCGGATGCCCT 200-1294 Nhel-Xhol pRRL2-GPK-Cher

G

hSox17 ATCGTCCTCGAGCACCATGAGCAGCCCGG ATGACTCTAGATCACACGTCAGGATAGTTG 203-1449 Xhol-Xbal pLVX-IRES-Hyg

ATGCGGG CAGT (Clontech ref.

632185)

Table 11. Characteristics of the arteriovenous fresh profile gene list

Number of

Characteristic Gene names

genes

COL4A1, COL4A2, CUBN, CYSLTR1,

Knock-out mouse has vascular phenotype* 13 EMX2, FGL2, HEY2, KDR, MSX1,

NKX2-3, NOTCH4, SEMA3C, SOX17

Previously associated with cardiovascular 7 ARL15, CNTN3, COIAA1, GRB14,

/

disease/risk factors in humans^† MFAP5, NPR3, TOX2

A2M, ACE2, ADRB1, CNTN3, CYSLTR1, FAP, FAT I, FOLH1, KDR,

Coding for cell surface protein 22 NOTCH4, NPR3, NRCAM, PTPRR,

SLC2A1, SLC2A3, SYTL2, XG, FLRT3, MPP7, NOS1, TMEM200A, TMEM200C

CCDC3, COL4A1, COL4A2, FAM19A5,

FGL2, FREM1, LAMA2, MFAP5,

Coding for secreted protein 14

ODAM, SEMA3C, SEMA3G, TFPI2, OLFML3, PAPLN

AFF3, APO, CUBN, EMX2, FAM176A, GALNTL4, GLIPR2, GRB14, HECW2,

HEY2, KLHL6, LMCD1, MAP9, MSX1,

Coding for intracellular protein 27

NAV1, NKX2-3, PDE4D, PRDM16, PSMAL, RGC-32, RYR3, SOX17, TOX2, CMBL, MGST1, SCARA3, SNTB2

AFF3, EMX2, HEY2, MSX1, NKX2-3,

Coding for transcription factor 9

PRDM16, SOX17, TOX2, NR3C2

APO, CCDC3, COL4A1, COL4A2,

CYSLTR1, GLIPR2, HEY2, KDR,

Documented expression in aEC and/or vEC 16

MFAP5, MSX1, NKX2-3, NOTCH4, RGC-32, SEMA3G, SOX17, NR3C2

Previously associated with arteriovenous

4 HEY2, KDR, NOTCH4, S0X17 specification

CNTN3, EMX2, HECW2, HEY2, KDR,

Previously linked with the Notch pathway^† 10 MFAP5, MSX1, NKX2-3, NOTCH4,

PRDM16

gene from the arteriovenous profile list alone and/or in combination with another gene;^† gene from the arteriovenous profile list or an orthologue/family member Table 12. Probe set intensities in HUAEC/HUVEC of genes contained within the arteriovenous fresh profile.

A B C D E F G H

Probe set ID Gene HUAEC-F HUVEC-F HUAEC-C HUVEC-C B-A C-A D-B D-C

Arterial probe sets

213764_s_at MFAP5 12,0 3,9 6,0 4,1 -8,1 -6,0 0,2 -1,9

229459_at FAM19A5 11,6 3,7 4,1 3,6 -7,9 -7,5 -0,1 -0,5

213765_at MFAP5 11,4 3,9 5,7 4,2 -7,6 -5,7 0,4 -1,5

206084_at PTPRR 11,4 4,0 6,7 4,6 -7,4 -4,6 0,6 -2,2

224646_x_at H19 13,5 6,3 6,6 6,8 -7,1 -6,9 0,4 0,2

209758_s_at MFAP5 11,3 4,6 6,9 5,6 -6,7 -4,4 0,9 -1,3

237094_at FAM19A5 10,1 3,5 3,8 3,7 -6,6 -6,4 0,1 -0,1

224997_x_at ΗΊ9 11,6 5,1 5,3 5,1 -6,6 -6,4 0,0 -0,2

227265_at FGL2 10,4 4,1 4,0 3,8 -6,3 -6,5 -0,4 -0,2

219789_at NPR3 10,0 3,9 4,8 4,1 -6,1 -5,3 0,1 0,7

229831 _at CNTN3 9,6 3,5 7,3 5,0 -6,0 -2,3 1,4 -2,3

210675_s_at PTPRR 10,2 4,2 5,4 4,2 -6,0 -4,7 0,1 -1,2

228109_at RASGRF2 11,0 5,0 7,9 7,4 -6,0 -3,1 2,4 -0,5

219689_at SEMA3G 10,5 4,6 5,1 6,0 -6,0 -5,4 1,4 0,8

232424_at PROMT 6 9,3 3,4 4,8 3,6 -5,9 -4,6 0,2 -1,2

204834_at FGL2 10,0 4,1 4,3 4,0 -5,9 -5,7 0,0 -0,2

205 13_at MPPED2 9,1 3,2 3,3 3,4 -5,8 -5,7 0,2 0,1

230943_at NA 10,3 4,5 7,8 8,9 -5,8 -2,6 4,4 1,1

217757_at A2M 10,7 4,9 4,6 4,7 -5,8 -6,1 -0,2 0,1

237727_at FREM1 10,1 4,4 4,0 4,3 -5,7 -6,1 -0,2 0,3

,211966_at COL4A2 10,8 5,3 10,0 8,6 -5,5 -0,9 3,3 -1,4

219743_at HEY2 9,8 4,3 6,2 5,8 -5,5 -3,6 1,5 -0,5

204105_s_at NRCAM 9,3 3,9 10,3 8,6 -5,5 1,0 4,7 -1,7

211981_at COL4A 7 12,2 6,8 10,8 9,9 -5,4 -1,4 3,1 -0,9

203789_s_at SEMA3C 9,6 4,3 10,1 7,3 -5,3 0,5 3,1 -2,8

209955_s_at FAP 11,2 5,9 9,9 6,8 -5,3 -1,3 0,9 -3,1

222921 _s_at HEY 2 10,9 5,6 8,5 8,0 -5,3 -2,3 2,5 -0,5

228737_at T0X2 9,8 4,5 6,2 7,6 -5,2 -3,6 3,1 1,5

219054_at NPR3 12,0 6,8 6,9 6,3 -5,2 -5,1 -0,5 -0,6

203934_at KDR 10,6 5,5 7,4 8,7 -5,1 -3,2 3,2 1,3

1554062_at XG 8,8 3,7 7,2 4,5 -5,1 -1,5 0,8 -2,7

229523_at TMEM200C 10,3 5,4 7,4 7,1 -5,0 -2,9 1,7 -0,3

218723_s_at RGC_32 9,1 4,2 6,0 7,9 -4,9 -3,1 3,6 1,9

216840_s_at LAMA2 8,7 4,0 4,5 4,5 -4,8 -4,2 0,5 0,0

234994_at TMEM200A 8,6 3,9 9,1 8,8 -4,7 0,5 4,9 -0,4

A B C D E F G H

Probe set ID Gene HUAEC-F HUVEC-F HUAEC-C HUVEC-C B-A C-A D-B D-C

236031 _x_at FREMl 8,4 3,7 3,6 3,7 -4,7 -4,8 0,0 0,1

229309_at ADRB1 8,3 3,6 4,3 4,1 -4,7 -4,0 0,5 -0,2

209278_s_at TFPI2 12,0 7,4 8,8 11,3 -4,6 -3,2 3,9 2,5

219993_at SOX 17 8,8 4,3 6,3 7,8 -4,6 -2,5 3,6 1,5

202498_s_at SLC2A3 10,6 6,2 5,4 5,8 -4,4 -5,1 -0,5 0,3

209277_at TFPI2 10,8 6,4 7,7 10,0 -4,3 -3,0 3,5 2,2

232914_s_at SYTL2 9,4 5,1 5,5 4,7 -4,3 -3,9 -0,5 -0,9

228233_at FREM1 8,8 4,5 3,9 4,1 -4,2 -4,8 -0,4 0,2

228167_at KLHL6 8,1 3,9 5,2 5,3 -4,2 -2,9 1,5 0,1

1553808_a_at NKX2-3 8,1 4,0 4,4 4,1 -4,1 -3,7 0,1 -0,3

232080_at HECW2 8,6 4,4 9,7 9,0 -4,1 1,1 4,5 -0,7

226809_at CYP4F30P 8,5 4,5 4,4 4,4 -4,0 -4,1 -0,1 0,0 in 201579_at FAT1 9,7 5,8 11,8 9,0 -3,9 2,0 3,1 -2,8

CO 215363_x_at FOLH1 9,9 6,0 4,6 4,6 -3,9 -5,4 -1,4 0,0

213519_s_at LAMA2 8,8 4,9 5,2 5,3 -3,9 -3,6 0,3 0,1

1554079_at GALNTL4 9,7 5,8 5,4 6,1 -3,9 -4,3 0,2 0,7

206306_at RYR3 7,2 3,3 3,4 3,3 -3,9 -3,8 0,0 0,0

227198_at AFF3 7,3 3,6 3,8 3,8 -3,7 -3,5 0,2 -0,1

222088_s_at SLC2A3 11,6 8,0 6,2 6,8 -3,7 -5,5 -1,2 0,6

205247_at N0TCH4 9,6 5,9 5,3 6,5 -3,7 -4,3 0,5

-3- 1,1

205116_at LAMA2 8,6 4,9 5,3 5,2 -3,7 -3,3 0,3 -0,1

221950_at EMX2 7,0 3,4 4,8 4,0 -3,6 -2,2 0,6 -0,8

224772_at NAVl 10,4 6,8 10,4 9,6 -3,5 0,0 2,8 -0,8

219842_at ARL15 9,2 5,6 6,6 5,6 -3,5 -2,5 0,0 -1,0

225496_s_at SYTL2 8,8 5,4 5,3 5,3 -3,4 -3,5 -0,1 0,0

211303_x_at FOLH1 7,6 4,3 3,8 3,7 -3,3 -3,8 -0,5 0,0 l\J

CD 228423_at MAP9 6,9 3,6 8,1 5,7 -3,3 1,2 2,1 -2,4

1556128_a_at RASGRF2 9,3 6,1 7,2 7,9 -3,2 -2,0 1,8 0,6

225602_at GLIPR2 8,1 4,9 9,7 9,4 -3,2 1,6 4,5 -0,2

237716_at APO 8,0 4,8 4,4 4,2 -3,2 -3,6 -0,6 -0,2

226795_at LRCH1 8,5 5,3 8,1 7,7 -3,2 -0,3 2,4 -0,4

205932_s_at MSX1 8,4 5,3 7,3 7,9 -3,1 -1,1 2,6 0,6

201250_s_at SLC2A1 11,3 8,2 9,5 8,6 -3,1 -1,9 0,4 -0,8

229655_at FAM19A5 8,0 4,9 5,0 5,2 -3,0 -2,9 0,3 0,2

206204_at GRB14 10,5 7,5 6,6 6,8 -3,0 -4,0 -0,7 0,2

220133_at ODAM 6,8 3,8 5,0 4,2 -3,0 -1,9 0,3 -0,8

223316_at CCDC3 13,8 10,9 8,0 7,9 -2,9 -5,7 -3,0 -0,1

201249_at SLC2A 7 7,1 4,3 5,5 4,7 -2,8 -1,7 0,4 -0,8

A B C D E F G H

Probe set ID Gene HUAEC-F HUVEC-F HUAEC-C HUVEC-C B-A C-A D-B D-C

238169_at NA 6,4 3,7 4,3 5,4 -2,7 -2,1 1,6 1,0

202497_x_at SLC2A3 10,3 7,7 6,2 7,0 -2,7 -4,1 -0,6 0,8

206775_at CUBN 8,6 5,9 6,3 6,0 -2,7 -2,3 0,1 -0,2

228151_at NA 12,7 10,0 9,9 8,8 -2,6 -2,8 -1,3 -1,1

228962_^at PDE4D 7,6 5,0 6,1 5,6 -2,6. -1,5 0,6 -0,4

241703_at RUNDC3B 6,2 3,6 6,7 4,5 -2,6 0,5 0,9 -2,2

225604_s_at GLIPR2 6,3 3,9 8,3 8,3 -2,4 2,0 4,4 0,0

213484_at NA 7,2 4,9 4,8 5,1 -2,3 -2,4 0,2 0,3

231747_at CYSLTR1 6,5 4,2 3,9 4,1 -2,3 -2,6 -0,1 0,2

227828_s_at FAM176A 7,0 4,7 8,8 8,8 -2,3 1,7 4,1 0,0

209227_at NA 6,0 3,8 6,4 5,6 -2,2 0,4 1,8 -0,8

222257_s_at ACE 2 6,8 4,6 4,5 4,6 -2,2 -2,3 0,0 0,1

228601 _at LOC401022 8,8 6,7 7,4 7,8 -2,2 -1,4 1,1 0,4

227502_at LCHN 9,0 6,9 6,3 6,3 -2,1 -2,7 -0,6 0,0

218574_s_at LMCD1 11,7 10,1 8,3 7,7 -1,6 -3,4 -2,4 -0,7 3 207020_at HSF2BP 6,5 4,9 6,5 6,0 -1,5 0,0 1,1 -0,4 on Venous probe sets

205259_at NR3C2 5,6 7,3 5,6 5,3 1,7 0,0 -2,0 -0,3

219416_at SCARA3 7,6 9,4 6,9 8,1 1,8 -0,7 -1,3 1,2

219250_s_at FLRT3 4,8 6,7 7,4 5,9 1,9 2,6 -0,8 -1,5

226435_at PAPLN 4,2 6,2 4,5 4,4 1,9 0,3 -1,8 -0,1 in

≡ 3 229004_at NA 5,3 7,3 6,7 5,5 2,0 1,4 -1,9 -1,3

227312_at SNTB2 6,2 8,4 6,1 5,5 2,2 -0,2 -3,0 -0,6

228195_at MGC13057 5,0 7,2 3,8 3,9 2,2 -1,2 -3,3 0,1

203571 _s_at APM2 5,8 8,1 5,8 6,2 2,3 -0,1 -1,9 0,4

227522_at CMBL 6,6 9,0 8,8 9,6 2,4 2,2 0,6 0,8

218162_at 0LFML3 5,1 7,6 5,2 7,0 2,4 0,1 -0,6 1,8

224918_x_at MGST1 4,0 6,6 8,5 9,6 2,6 4,5 2,9 1,0 l\J

238778_at MPP7 5,0 8,4 5,1 3,3 3,3 0,0 -5,1 -1,8

239132_at NO SI 6,6 10,0 5,0 4,9 3,4 -1,6 -5,0 -0,1

Table 13. Expression of genes from the arteriovenous fresh profile in DLL4-Fc treated HUAEC.

Transcription factor code for arterial EC identi Table 14. Expression of genes from the arteriovenous fresh profile in cultured HUVEC overexpressing transcription factors. aEC

genes cherry HEY2 MSX1 E X2 Prdml6

MFAP5 100 ± 69 919 + 229* 91 ± 71 696 + 209 919 ± 0*

FAM19A5 100 + 15 158 ± 11 91 + 22 151 + 29 148 + 15

PTPRR 100 ± 57 207 + 133 165 + 21 643 + 92 441 + 107

H19 100 + 4 129 + 44 196 + 68 310 + 85* 39540 + 10556*

FGL2 100 + 71 706 ± 539* 212 + 130 121 + 60 339 + 160

NPR3 100 + 77 798 + 403* 182 + 168 490 + 193* 419 + 64*

CNTN3 100 + 96 180 + 111 51 + 28 27 + 22 546 + 63*

RASGRF2 100 + 37 919 + 378* 31 + 16 319 + 101 393 + 78*

SEMA3G 100 ± 52 ^• 81 ± 31 288 ± 212 174 ± 45 179 - + 25*

PRDMT6 100 + 20 208 + 38 184 + 44 177 + 42 125 + 27

MPPED2 100 ± 79 371 ± 605 197 ± 237 222 + 275 217 + 370

A2M 100 + 41 70 + 22 50 ± 7 141 + 31 2061 + 394*

FREM1 100 + 12 125 + 35 106 + 6 167 + 22 96 + 56

C0L4A2 100 + 21 179 + 49 146 + 42 384 + 54* 77 ± 8

HEY2 100 + 116 152 + 144 746 + 272* 255 + 208 767 + 263*

NRCAM 100 + 39 221 ± 62 55 + 42 43 + 7* 110 + 10

C0L4A 1 100 + 22 193 ± 56 141 + 55 340 + 49* 86 + 10

SEMA3C 100 ± 41 985 ± 820* 356 + 75 43 + 19 15578 + 2529*

FAP 100 ± 9 125 + 4 117 ± 12 115 + 10 129 + 8

T0X2 100 + 16 91 + 11 116 + 12 104 + 9 77 + 5

KDR 100 + 28 61 + 16 120 + 33 65 + 10 29 + 3*

TMEM200

C 100 + 91 56 + 22 83 + 108 85 + 19 14 ± 9*

XG 100 + 25 201 ± 54 297 + 37* 222 + 39 373 + 71*

RGC-32 100 ± 21 41 + 14* 82 + 22 113 ± 20 169 + 8

LAMA 2 100 ± 12 107 + 10 122 + 12 159 ± 8* 120 + 7

TMEM200

A 100 + 31 130 + 15 64 ± 20 52 ± 4 53 + 2

ADRB1 100 + 10 102 ± 24 113 + 14 120 ± 7 60 + 12

TFPI2 100 + 27 342 ± 268* 154 + 45 1706 ± 197* 94 + 9

SOX 17 100 + 12 105 + 2 112 + 7 110 ± 11 86 ± 3

o o

R

^ in N

co o

1 I s*:

a Co

TABLE15. Primerlist for work described in Example 3

Gene working concentration

Symbol Species Forward Primer (5'->3') Reverse Primer (5'->3') (nM)

PRDM 16 H CAAACGCTTCGAATGTGAAA CGTGTAGGACTTGTGGCAGA 200

DLL4 H ATGACCACTTCGGCCACTATG GCCCGAAAGACAGATAGGCTG 200

HEY1 H GAGAAGGCTGGTACCCAGTG G CTC AG ATAAC GCG C AACTT 200

HEY2 H CCCCTGCGAGGAGACGA ATCTAATCACAGAGCTAGTACTTTGCCC 200

NOTCH 1 H CCACGGGCGACGTCACCC TCCACTCTGGCGGGCACG 200

NOTCH4 H ATGTCTCAATGGCGGCTCC GGAGAAGGTGCCAGGCCT 200

EFIMB2 H CTGCTGGATCAACCAGGAAT TCCGGTACTTCAGCAAGAGG 200

GJA5 H TACAACACAAGGCAGCAAGC TCGTATCACACCGGAAATCA 200

NR2F2 H CGCCTCAAAAAGTGCCTCA GCATCCTGCCCCTCTGC 200

NRP2 H GCGAGTGGATTGTTTACGCC CAGTCTTTGCCCAGGAGGTC 200

EPHB4 H GAGCTGTGTGGCAATCAAGA GAACTGTCCGTCGTTTAGCC 200

B-ACTIN H TGGCACCACACCTTCTACAATG TAGCAACGTACATGGCTGGG 200

ODAM H CAGGCCAAGTTGATCCCTTA TGAGGTTGTTCCCAGGGTAG 200

SEMA3C H TCATTCCATGATTGCTCGAA CCTCGGGTTATCAGTTTCCA 200

SEMA3G H ACGGAGCACAATAGCACCTT GACCACAGTCTGGGAGAAGC 200

FGL2 H TTGGATGGCAAATGTTCAAA CCATGGTCTCCATGTCACAG 200

MFAP5 H ATACCCCTGGGGGCTAAATAG C GTCGTAAACTG GTG AAG C A 200

RASGRF2 H ACCTTGCCATCGAAAGAATG ACAGAACCCCACCTGTCTTG 200

NPR3 H GGAGACCGATATGGGGATTT CACTGCCGATTCTTCTAGGC 200

ACE2 H ATACTGTGACCCCGCATCTC ATGCTAGGGTCCAGGGTTCT 200

FAT H GTGGAAAAGGGGACAGTGAA CCTGATCGGTTGCAAAGACT 200

SYTL2 H TG I 1 1 1 1 GTGGCCCAGTGTA AGTTCCACCTCCCCTAGGAA 200

MAP9 H AC ATG G AGG AG AAG G ATG G A CAGATGCGTTTCCCTCAGAT 200

PTPRR H TTGTCTGCCAGCTTCGTATG GG G CTTCAG AG CTTCTC CTT 200

CUBN H ACTGTGAAGGGGGTTCTGTG GACAGGCCCCACAGTAGAAA 200

XG H AGCTGGGAGACCAGAAGTCA GGCATCTGCCAAATCAAAGT 200

Prdm l6 M CACGGAAGAGCGTGAGTACA GGCAGACCTCGCATATGAAT 200

Nr2f2 M GACTCCGCCGAGTATAGCTG CCTACCAAACGGACGAAAAA

Hey2 M GCGTCGGGATCGAATAAATA AGCATGGGCATCAAAGTAGC

B-actin M GACGGCCAGGTCATCACTAT CTTCTGGATCCTGTCAGCAA prdm l6 ZF CATTCGCCATCAGATGTCAC TGTGACGGCACAGGTTAGAG efla ZF TCACTGGTACTTCTCAGGCTGACT TTCTTGGAGATACCAGCCTCAAA

Table 16. Gene list of arterial- and venous-specific genes differentially expressed ([log- value]:.1 and P<0.001 ) between mAECs and mVECs

ARTERIAL GENES

Ace Edn1 Gkn3 Krt80 Pcolce2 Sod3

Atp2b2 Ehd4 Glee Lox Pde3b Sponl

Bgn Enpp6 Glt8d4 Mall Plcg2 Ssfa2

Bmp6 Ephx2 Gria3 Mettl7a1 Plxdc2 Sulfl

Cbr2 Fam3c Hey2 Mkl2 Prss23 Tagln

Ccdc3 Fbliml Htral Naaladl2 Ptprj Thbsl

Clu Fbln2 isn Nmt2 Serpingl Timp2

Cmklrl Fbln5 Kazaldl Nox4 Serpinil Tnfrsf26

Cp Fgfrl KlkW Nuprl Sfrpl Wwp1

CytH Frzb Klk11 Pam Slc4a4

Dlx5 Gja5 Krt18 Pcdh9 Srnod

VENOUS GENES

Abcbla Emcn Hamp Ltbpl Ms Serpina3n

Aqp1 Eng Hmcnl Lyvel Pitpnd Stra6

Arhgap18 Enpp2 ΙΠ203 Mctpl Pla2g7 TmemlOO

Bmp2k Ephal Igfbp5 Mest Plxntf Tppp

Casp6 Flt1 Igfbp6 Myo10 Postn Vav3

Ccdc141 Fmnl2 Inhba Myolb Proxl Zfp462

Cdh11 Fmo1 Itga6 Nckap5 Rasgeflb Zfp521

Ctla2a Gimap6 Kdr Npr3 Rasgrp3

Ctla2b Gja1 Lama4 Nr2†2 Rgs16

Eb†1 Gpm6a Lbh Nrp2 Rgs5

Eltdl Gpr116 Lsamp Nsg1 Rspo3

Transcription factors are indicated in boldface.

Table 17. Average arterial and venous probe intensities and corresponding log₂ ratios for all TFs identified in our human and murine microarrays.

Log₂ ratio

(HUAECs

GENE HUAECs HUVECs mAECs mVECs (mAECs/

/

mVECs)

HUVECs)

Arterial

Aff3 7.23 3.59 3.64 5.91 6.12 -0.21

Dlx5 4.03 4.06 -0.03 8.49 6.19 2.30

Emx2 6.91 3.42 3.49 5.02 4.83 0.19

Hey2 10.9 5.59 5.26 9.96 8.03 1.93

Isl1 4.49 4.89 -0.40 7.80 5.86 1.94

Msx1 8.28 5.30 2.98 7.54 7.16 0.38

Nkx2-3 7.83 3.95 3.88 6.05 5.89 0.16

Prdm16* 9.40 3.38 6.02 8.27 7.12 1.15

Sox17 8.84 4.30 4.54 8.03 7.22 0.81

Tox2 9.66 4.53 5.13 7.01 7.04 -0.03

Venous

Eb†1 5.01 5.06 -0.05 7.59 9.41 -1.82

Coup-TFII** 6.67 10.0 -3.33 6.69 8.07 -1.38

Nr3c2 5.62 7.30 -1.68 6.53 6.32 0.21

Proxl 4.63 5.97 -1.34 5.63 8.32 -2.65

Zn†462 6.97 9.15 -2.18 7.09 8.39 -1.30

Zn†521 9.26 10.3 -1.04 7.03 8.78 -1.75

^*While Prdm16 was not present in our murine arteriovenous fingerprint due to the stringent tresholds we utilized, this TF was significantly differentially expressed between mAECs and mVECs (P<0.01 ) and was more than 2-fold enriched in AECs in both human and mouse ECs. "Likewise, Nr2f2 did not make our cut-off in the human signature although it was significantly enriched in HUVECs versus HUAECs (ΡΟ.0 ).

Table 18. List of conserved arterial/venous-specific genes

ARTERIAL GENES VENOUS GENES

Antxr1^$ Hev2* Abcg2* Gpm6a Proxl

Atp2a3* Igf2 Bex1 Lrg1 * Pthlh

Bcap29 Lox* Cdh11 Ltbp1 * Rgs5^$

Camk4* Nox4* Chn2 Lyvel Samd5

Ccdc3 Prdm16 Ctsk* Meox2* Sntb2

Cp Ptpri* Dcn^$ Mest* Tbx5*

Dkk2^$ Rps6ka5 Emcn Nostrin* Zfp462

Edn1* Slc4a4 Enpp2* Nr2f2* Zfp521

Efna5* Thbs Entpdl* Nrp2*

Ehd4** Tox Flrt2* Piptnd

GjaS* Fmnl2 Pkhd1l1

^*knockouts mice has cardiovascular phenotype; ^**Ehd4-/-Ehd3-/- has vascular phenotype; ^$described to have a role during pathological angiogenesis; genes already described to be expressed in an arterial/venous-specific manner are in red, TFs are in bold.

Claims

VASCULAR BED-SPECIFIC ENDOTHELIAL CELLS

Claims

What is claimed is:

96. An engineered vascular endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell, the cell being genetically altered by transduction or transfection of a source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lett , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof.

97. A method to differentiate a source cell into a vascular endothelial cell (the 'target cell'), said method comprising genetically altering the source cell by transduction or transfection of said source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Let , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof.

98. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell of claim 1 , the cell being genetically altered by transduction or transfection of a source cell with a transcription factor of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof.

99. The engineered vascular bed-specific endothelial cell (the 'target ceil') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , whereby the cell is being genetically altered by transduction or transfection of a source ceil with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an organ-specific microvascular endothelial cell.

100. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil according to claim 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell.

101 . The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , whereby the ceil is being engineered by transduction or transfection of a source cell with the transcription factor Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into an arterial endothelial cell.

102. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source ceil to differentiate into a vascular endothelial target cell according to claim 1 , whereby the is the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Meox2, Tcf15, Ppary, Wt1 and Ebf3 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into a cardiac microvascular endothelial ceil.

103. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , whereby the is the cell being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Zic3, Lef1 , Foxf2, Foxfl a and Foxd or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a brain microvascular endothelial cell.

104. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to claim 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Tcfec, Hoxb5, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source ceil into a liver microvascular endothelial cell.

105. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target ceil according to claim 1 , whereby the cell is being engineered by transduction or transfection of a source cell with a transcription factor of the group consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell.

106. The engineered vascular bed-specific endothelial cell (the 'target ceil') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims 1 to 10, whereby the transcription factors are overexpressed.

107. The engineered vascular bed-specific endothelial cell (the 'target ceil') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims 1 to 1 1 , whereby the source cell is a source cell of the group consisting of endothelial progenitor cells (EPCs), mesenchymal stem cells (MSCs), embryonic stem cells (ESCs), induced pluripotency stem cells (iPSCs), mesoangiobiasts (MABs), multipotent adult progenitor cells (MAPCs), blood outgrowth endothelial cells (BOECs), induced endothelial cells (iECs), unfractionated bone marrow cells, umbilical cord or peripheral blood-derived mononuclear cells, adipose tissue-derived cells, tissue resident progenitor cells, human umbilical vein endothelial cells (HUVECs), human umbilical artery endothelial cells (HUAECs), human dermal microvascular endothelial cells (HDMECs), cultured endothelial cell lines from heart, cultured endothelial cell lines from liver, cultured endothelial ceil lines from brain.

108. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims 1 to 1 1 , whereby the source cell is a dedifferentiated cultured human umbilical vein endothelial ceil (HUVEC) or a dedifferentiated HUAEC.

109. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims 1 to 1 1 , whereby the source cell is a mammalian cell of embryonic or non-embryonic origin.

1 10. The engineered vascular bed-specific endothelial cell (the 'target cell') or the engineered source cell to differentiate into a vascular endothelial target cell according to any one of the previous claims, whereby the transduction or transfection is lentiviral.

1 1 1 . An engineered vascular bed-specific endothelial cell, characterized in that the cell is engineered by induced differentiation of a source cell by transferring directly to source cells the expression product of a transcription factor of the group consisting of Meox2, Tcf15, Pparv, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof the expression product being linked to a protein transduction domain (PTD), such as poly-arginine and HlV-derived Tat, or to small cationic peptide domains to enhance its cell membrane crossing capacity.

1 12. The engineered vascular bed-specific endothelial cell of claim 16, whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl , Zeb2, Emx2, Msx1 , Nkx2-3, Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a vascular endothelial target cell.

13. The engineered vascular bed-specific endothelial ceil of claim 16, whereby the expression product is of the group consisting of Prdm16, Emx2, sx1 , Tox2, Aff3 and Nkx2-3 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a vascular endothelial target cell.

14. The engineered vascular bed-specific endothelial ceil of claim 16, whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 , Ebf3, Zic3, Lef1 , Foxf2, Foxfl a, Foxd , Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an organ- specific microvascular endothelial cell.

15. The engineered vascular bed-specific endothelial cell of claim 16, whereby the expression product is of the group consisting of Prdm16, Emx2, Msx1 , Tox2, Aff3 and Nkx2-3 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell.

16. The engineered vascular bed-specific endothelial cell of claim 16, whereby the expression product is of Prdm16 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, to differentiate source cell into an arterial endothelial cell.17. The engineered vascular bed-specific endothelial cell of claim 16, whereby the expression product is of the group consisting of Meox2, Tcf15, Ppary, Wt1 and Ebf3 or a human homoiogue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a cardiac microvascular endothelial cell.

18. The engineered vascular bed-specific endothelial cell of claim 16, whereby the expression product is of the group consisting of Zic3, Lef1 , Foxf2, Foxf 1 a and Foxd or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into a brain microvascular endothelial ceil.

1 19. The engineered vascular bed-specific endothelial cell of claim 16, whereby the expression product is of the group consisting of Tcfec, HoxbS, Maf, Cux2, Gata4, Meis2, Twistl and Zeb2 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source ceil into a liver microvascular endothelial cell.

120. The engineered vascular bed-specific endothelial cell of claim 16, whereby the expression product is of the group consisting of Emx2, Nkx2-3, Msx1 , Tox2, Aff3 and Prdm16 or a human homologue or a variant, member of the same family, orthologue thereof, having the same biological function, or a combination thereof to differentiate source cell into an arterial endothelial cell.

121 . The engineered vascular bed-specific endothelial cell according to any one of the previous claims 16 to 25, whereby the transcription factors are overexpressed.

122. The engineered vascular bed-specific endothelial cell according to any one of the previous claims 16 to 26, whereby the source cell is a source cell of the group consisting of endothelial progenitor cells (EPCs), mesenchymal stem cells (MSCs), embryonic stem cells (ESCs), induced pluri potency stem cells (iPSCs), mesoangioblasts (MABs), multipotent adult progenitor cells (MAPCs), blood outgrowth endothelial ceils (BOECs), induced endothelial ceils (iECs), unfractionated bone marrow cells, umbilical cord or peripheral blood-derived mononuclear cells, adipose tissue-derived ceils, tissue resident progenitor cells, human umbilical vein endothelial cells (HUVECs), human umbilical artery endothelial cells (HUAECs), human dermal microvascular endothelial cells (HDMECs), cultured endothelial cell lines from heart, cultured endothelial cell lines from liver, cultured endothelial cell lines from brain.

123. The engineered vascular bed-specific endothelial cell according to any one of the previous claims 16 to 26, whereby the source cell is a dedifferentiated cultured human umbilical vein endothelial cell (HUVEC) or a dedifferentiated HUAEC.

124. The engineered vascular bed-specific endothelial cell according to any one of the previous claims 16 to 26, whereby the source cell is a mammalian cell of embryonic or non-embryonic origin.

125. The engineered cell according to any one of the previous claims 16 to 29, for use in a treatment of vascular bed-specific disorder.

126. The engineered cell according to any one of the previous claims 1 to 29, for use in a treatment of a disorder of vascular bed-specific haemostasis or for a disorder of generating and/or maintaining vascular bed-specific phenotypes.

127. The engineered cell according to any one of the previous claims 1 to 29, for use in a treatment of for decreasing severity of vascular bed-specific disorder or diseases or circulatory or hypoxic conditions that can be treated with the compositions and the methods of the invention comprise but are not limited to: atherosclerosis, preeclampsia, erectile dysfunction, renal failure, transplant accelerated arteriosclerosis, deep vein thrombosis, sleep apnea, hypoxia during sleep, fetal hypoxia, smoking, anemia, endothelial dysfunction, sinusoidal obstruction syndrome, regional perfusion deficits (e.g., limb, gut, renal ischemia), congestic heart failure, peripheral vascular disease, frost bite, decubitus ulcers, asphyxiation, poisoning (e.g., carbon monoxide, heavy metal), altitude sickness, pulmonary hypertension, sudden infant death syndrome, asthma, chronic obstructive pulmonary disease, congenital circulatory anomalities (e.g., Tetralogy of Fallot), erythroblastosis, myocardial infarction, aortic stenosis ro any other cardiac disease which can lead to heart failure, plaque rupture, both primary and secondary (in-stent) restenosis in coronary or peripheral arteries, coronary vascular disease, hypertension, diabetes, obesity, stroke, aneurysm, thrombosis, arrythmia, tachycardia, surgical or physical trauma in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment the engineered cell, thereby decreasing severity of disorder in the subject.

128. The engineered cell according to any one of the previous claims 1 to 29, for use in a treatment for decreasing severity of vascular bed-specific diseases affecting the brain (e.g., stroke), the liver (e.g., sinusoidal obstruction syndrome), the heart (myocardial ischemia) or the extremities (e.g., peripheral vascular disease) in a subject, the method comprising autologous, allogeneic or xenogeneic cell transplantation; in a subject in need of treatment the engineered ceil, thereby decreasing severity of disorder in the subject.

129. A pharmaceutical composition, comprising an engineered cell according to any one of the previous claims 1 to 29.

130. An in vitro bioengineered tissue, comprising an engineered cell according to any one of the previous claims 1 to 29.

131 . The use of an engineered cell according to any one of the previous claims 1 to 29, in a method of screening for non-toxic compounds.

132. The use of an engineered cell according to any one of the previous claims 1 to 29, in a method of in vitro bioengineering a tissue.

133. The use of an engineered ceil according to any one of the previous claims 1 to 29 as a (or one of the) cellular component(s) of tissue engineered constructs for implantation in patients (for example to coat the inside of artificial arterial conduits or as the intimal cellular component of a fully biological tissue-engineered vascular graft or to coat cardiac valves).

134. An in vitro method of diagnosing a vascular bed-specific disorder phenotype in a subject, said method comprising (a) analysing the level of expression or activity of expression product of at least 5 genes of the '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs) in a sample isolated from said subject, and (b) compare said level of expression or activity with the level of expression or activity in '(differential) reference signatures' of table 4 (brain ECs), of table 5 (liver ECs), of table 6 (heart ECs) and/or tables 12/16/18 (arterial or venous ECs); whereby a deviated level of expression or activity relative to such 'reference signature' is an indication of such disorder phenotype or a propensity thereto.