CA2429317A1

CA2429317A1 - Specific promoter of the human vascular endothelial cadherin-2 (hve-cad-2) gene and the therapeutic uses thereof

Info

Publication number: CA2429317A1
Application number: CA002429317A
Authority: CA
Inventors: Robert Zweigerdt; Manfred Rudiger; Thomas G. Moll
Original assignee: CARDION AG
Current assignee: Individual
Priority date: 2000-11-16
Filing date: 2001-11-16
Publication date: 2002-05-23
Also published as: IL155907A0; AU2002220711A1; EP1392836A2; WO2002040692A2; WO2002040692A3

Abstract

The invention relates to a nucleic acid fragment which contains the regulatory sequence of the VE-Cad-2- gene, a complete or a functional active variation thereof. The invention also relates to a nucleic acid construct containing the nucleic acid fragment and a functional nucleic acid sequence or a heterological gene, a vector or a vector system, a cell or a cell line, a transgenic non-human animal, a medicament, a diagnostic, an array, a method for identifying pharmaceutically active substances, a method for identifying functionally active variants, a method for isolating endothelium cells from stem cells, a method for treatment, a method for administering a medicament, the use of the nucleic acid fragment, the nucleic acid construct, the transformed cell or cell lines of the tests and the arrays.

Description

Vascular endothelium-specific promoter of the human vascular endothelial cadherin-Z (hVE--Cad-2) gene and its therapeutic uses The invention relates to a nucleic acid fragment comprising the regulatory sequence of the VE-Cad-2 gene, completely or a functionally active variant thereof, to a nucleic acid construct comprising the nucleic acid fragment and a 1o functional nucleic acid sequence or a heterologous gene, to a vector or a vector system, to a cell or cell line, to a transgenic non-human animal, to a pharmaceutical, to a diagnostic, to an array, to a method for the identification of pharmacologically active substances, to a method of identifying functionally active variants, to a method of isolating endothelial cells from stem cells, to a method of treatment, to a method of administering a pharmaceutical, to the use of the nucleic acid fragment, of the nucleic acid construct, of the transformed cell or cell line, of the test, and of the array.
The vascular endothelium is a monolayer assemblage consisting of a large 2o number of partially overlapping individual cells and lines all arterial and venous blood vessels of higher vertebrates. The cells of the vascular endothelium form a barrier preventing interaction of blood cells with the vessel wall. The vascular endothelium plays an important part in the regulation of coagulation, the adhesion of leukocytes, the growth of smooth muscle cells of the vessels and the control of the transvascular diffusion of metabolites.
The wide distribution of the vascular endothelium in the body therefore provides the opportunity, on specific expression of genes in vascular-endothelial cells, of expressing a therapeutically active gene either systemically in the whole body or, 3o depending on the transfer system, locally in particular organs or specifically in segments of the vascular system.

For example, donor organs taken for transplantation can be perfused via the vascular system outside the body. The use of suitable vectors such as adenoviruses or liposomes makes it possible for the vascular endothelium to be transfected efficiently and be used for expressing genes which protect the transplanted organ and prevent rejection in the recipient.
Numerous experimental and clinical studies have shown that supply of a tumor through blood vessels and capillaries is crucial for its growth. Once again, the endothelium, which lines even the smallest capillaries of highly vascularized to tumors, provides an opportunity to introduce growth-inhibiting or cytotoxic substances efficiently into the interior of tumors. The aim is to retard the growth of the degenerate cells or bring about definitive death of the cancer cells.
Said gene therapy approaches require promoters which control efficient and specific expression of therapeutically relevant genes in the vascular endothelium.
A number of DNA sequences which bring about expression in vascular-endothelial cells have been published. However, with some of these sequences the gene activation is not confined to the vascular endothelium but leads to expression 2o in other cells and tissues too. These include regulatory DNA elements of the following genes: von Willebrand factor (Guar J. et al. (1999). Blood.
94(10):3405-12.), platelet/endothelial cell adhesion molecule-1 (PECAM-1;
Botella LM. et al. (2000) J Imrnunol. 164(3):1372-1378), preproendothelin-1 (Aversa CR. et al. (1997). Am J Physiol 273(4 Pt 1):L848-855), Tie-1 (Hewett PW. et al. (1998) Biochem Biophys Res Commun. 252(3):546-551) and P-selectin (Pan J. et al. (1998). J Biol Chem. 273(16):10058-10067). Because of said limitation, these regulatory sequences are substantially unsuitable for use in gene therapy.

Although other genes such as the receptor tyrosine kinase TIE-2 (Dube A. et al.
(1999). Circ Res. 84(10):1177-1185), E-selectin (Smith GM. et al. (1993).
Biochem Biophys Res Commun. 194(1):215-221) and the "vascular endothelial growth factor receptor-1" (Flt-1; Wakiya K. et al. (1996) J Biol Chem.
271(48):30823-30828) and -2 (Flk-l; Kappel A. et al. (1999) Blood. 93(12):4284-4292) are specifically expressed in the vascular endothelium, this occurs mainly after cell activation by inflammatory cytokines or only during vascular proliferation.
to By contrast, other well-characterized promoters show a substantially uniform vascular-endothelial cell-specific gene expression in transgenic mice. These include the promoter of the vascular endothelial cadherin-1 gene (Gory S. et al.
(1999). Blood. 93(1):184-192) and regulatory elements of the ICAM-2 gene (Cowan PJ. et al. (1998). J Biol Chem. 273(19):11737-11744).
Uniform gene expression in the vascular endothelium of all organs of the body may certainly be an advantage or even the precondition for some applications.
On the other hand, this may give rise to crucial disadvantages for other gene therapies, as will be explained in detail below.
Numerous gene therapy approaches for controlling cancer have recently been developed. In principle, two strategies are distinguished in this connection.
One possibility is the systemic administration of nontoxic precursor substances which are converted into cytotoxic products only in the local milieu, i.e.
environment, of the tumor (Connors TA. (1995). Gene Ther. 2: 702-709; Chrystal RG. (1999).
Cancer Chemother Pharmacol 43 (Supply: 90-99).
This is done by introducing into the tumor a gene whose protein product brings about the conversion of the systemically administered chemotherapeutic agent 3o from the precursor form into the active form. Expression of this transgene in the vascular endothelium of highly vascularized cancerous growths leads to a high local concentration of the active chemotherapeutic agent in the interior of the tumor. At the same time, the systemic exposure of the patient to the histotoxic agent is low.
Examples of this prodrug therapy are the transfection of tumors with the gene for thymidine kinase (TK) from the herpes simplex virus, which converts ganciclovir into the active, cytocidal form (Culver KW. et al. (1992). Science 256: 1550-Link CJ. et al. (1995). Hybridoma 14: 143-147; Tanaka T. et al. (1996). Cancer 1o Res. 56: 1341-1345), and the use of the gene for cytosine deaminase (CD) from fungi or bacteria, which catalyzes the conversion of the base S-fluorocytosine (5-FC) into toxic 5-fluorouracil (5-F~ and is employed for treating colorectal carcinomas (Topf N. et al. (1998). Gene Ther. 5: 507-513).
The second strategy is based on expression of genes whose products cause initiation of programmed cell death, called apoptosis, and thus are able to cause the death of cancer cells.
Examples of the "pro-apoptosis" therapy are the successful expression of tumor 2o necrosis factor-alpha (TNF-a) in vascular endothelial cells of tumors (Jaggar RT.
et al. (1997) Hum Gene Ther. 8 (18): 2239-2247) or the expression of TRAIL
("TNF-related apoptosis-inducing ligand"), a member of the TNF-a superfamily.
It was possible to show for both cytokines that their expression may successfully lead to the death of numerous tumors (Griffith TS. et al. (2000). J Immunol.

(5): 2886-2894; Jaggar RT. et al. (1997). Hum Gene Ther. 8 (18): 2239-2247).
Another paradigm in the gene therapy control of cancer is the use of the Fas ligand (Fast; CD95 ligand). According to this, expression of Fast, a transmembrane protein which likewise belongs to the TNF-a family, is suitable for initiating programmed cell death in degenerate cells in vivo (Shimizu M.
et al.
(1996). Biochem Biophys Res Commun. 228(2):375-379; Seino K. et al. (1997).
Nat Med. 1997 3(2):165-170).
The induction of apoptosis is the result of the specific interaction between the ligand and its receptor, Apol/Fas (CD95).
Besides induction of cell death in cancer cells carrying the Apol/Fas receptor, it 1o was possible by expressing Fast locally in the vessel wall also to prevent successfully neointima formation after balloon dilation (Sata M. et al.
(1998).
Proc Natl Acad Sci USA. 95(3):1213-1217). Another property of Fast is maintenance of an immunoprivileged status in some organs such as the brain, the anterior chamber of the eye and the testis. In this connection, expression of Fast in vascular-endothelial cells of transplanted organs was able to prevent rejection of the transplants (Swenson KM. et al. (1998). Transplantation. 65(2):155-160;
Tran TH. et al. (1998). Transplantation. 66(9):1126-1131).
All said therapeutic approaches - irrespective of the gene used - require efficient 2o gene transfer into vascular endothelial cells.
Besides liposomal formulations and retroviruses, probably the most efficient and therefore most frequently used vehicles at present for introducing foreign genes into mammalian cells in vitro and in vivo are adenoviral vectors (AdV). In this case, use is made of the broad host spectrum, the episomal presence of the vector and the infectability of mitotically inactive cells. Because of their tropism, the vectors are particularly suitable for transduction of vascular-endothelial cells, in which case high-level expression of the transgene is achieved (Crystal RG.
(1999). Cancer Chemother Pharmacol. 43 Supp1:S90-99; O'Brien T, Simari RD.
(2000). Mayo Clin Proc. Aug;75(8):831-834).

In spite of the numerous advantages of AdV, there are also limitations.
Studies in mice in which adenoviral vectors were administered systemically using reporter genes have shown that the liver is the organ with by far the greatest transduction on intravenous administration of AdV (Kurata H, et al. (1999). J Allergy Clin Immunol. 103(5 Pt 2):5471-84; Ye X. et al. (2000). Hum Gene Ther. Mar 1;11 (4):621-7).
Accordingly, in animal experiments both using the precursor substance therapy and employing proapoptotic genes there was found to be systemic toxicity which 1o was mainly attributable to hepatotoxicity (Brand K. et al. (1997). Cancer Gene Ther. 4(1):9-16; Brand K. et al. (1998). Gene Ther. 5(10):1363-1371; Okuyama T.
et al. (1998). Gene Ther. 5(8):1047-1053).
However, this finding was made not only following systemic virus administration by intravenous injection. Perfusion of organs for transplantation with VasL-expressing AdV led to liver damage in the recipient organism just as did local injection of vectors directly into tumors, because the viruses used eventually reached the bloodstream and accumulate in the liver (Morelli AE. et al. (1999) J
Gen Virol. 80 ( Pt 3):571-83; Putzer BM. et al. {2000). Gene Ther. 7(15):1317-25;
2o Aoki K. et al. (2000). Mol Ther. (6):555-565).
In summary, there are gene therapy strategies suitable for killing cancer cells in vivo. It has also been possible to show that transplant rejection is prevented by expression of protective genes in the vascular endothelium of transplanted organs.
However, applications to date are limited by the fact that the vectors used accumulate and are expressed in the liver of the treated animals. This results in extensive damage to the organ.
There is thus a pressing need for promoters which make it possible for genes to be 3o expressed efficiently in the endothelium of blood vessels. It is in particular the object of the invention to provide promoters which make it possible for genes to be exposed efficiently in the endothelium of blood vessels but, at the same time, preferably preclude any activation in the liver, including the vascular endothelium of the liver.
The invention was thus based on the object of providing nucleic acid fragments for targeted in vivo gene transfer which make it possible in particular in vivo for genes to be switched on as specifically as possible in venous and arterial endothelial cells of all organs and tissues preferably with the exception of 1 o expression in the liver.
It is intended, for example, to achieve a high level of safety on expression of therapeutic genes for controlling tumors, in the induction of tolerance for transplantation and other gene therapy methods, which is of very great importance because of the cytotoxic nature of the genes used. The vascular endothelium-specific promoters previously available do not comply with this requirement.
The object of the invention has been achieved surprisingly by providing a nucleic acid fragment, wherein the nucleic acid fragment contains a regulatory sequence of the human VE-Cad-2 gene extending 5' upstream from the translation start or a functionally active variant thereof, such that the nucleic acid fragment allows vascular endothelium-specific expression. In a preferred embodiment of the invention the nucleic acid fragment according to the invention allows vascular endothelium-specific expression excluding expression in the liver.
As a result of the experiments conducted for this study a regulatory sequence of the human VE-Cad-2 gene contained in the BAC clone SOg21 was identified (Accession number AC005740, W. Kimmerly et al., submitted 1998 at the Human Genome Center, DOE Joint Genome Institute, Lawrence Berkeley National 3o Laboratory, USA). The nucleic acid fragment according to SEQ ID No. 1 extends _g_ 5' upstream from the translation start located in position 110948-110954 (atg atg) in the BAC clone SOg2l. It should be pointed out that the genomic database entry AC005740 of the National Center for Biotechnology Information accessible through the PubMedline does not contain any reference or information to the human VE-Cad-2 gene nor to the regulatory sequence of the human VE-Cad-2 gene.
The invention also relates to a nucleic acid fragment containing the sequence according to SEQ ID No. 1 or a functionally active variant thereof. In a preferred 1o embodiment of the invention the nucleic acid fragment according to the invention allows vascular endothelium-specific expression excluding expression in the liver.
Recently, Ludwig et al. (Mammalian Genome 2000, 11, pp. 1030-1033) have cloned the human VE-Cad-2 gene and a sequence extending 5' upstream from the translation start ( Figure 3B) that represents a 1717 by long part of the 5021bp long nucleic acid fragment according to the invention according to SEQ ID No.

sharing 100 % sequence homology. Even though figure 3B of Ludwig et al.
(2000, supra) depicts putative transcription factor binding sites located in the 500 by long sequence 5' upstream from the transcription start (Figure 3B) determined 2o by means of database search, the study does not provide experimental evidence for the functioning or activity of this potential regulatory sequence to enable a person skilled in the art to design or obtain a regulatory sequence of the human VE-Cad-2 gene that would allow vascular-endothelium-specific expression.
Furthermore the data on the expression of the VE-Cad-2 gene also demonstrate a solid expression of the gene in the liver. A person skilled in the art would therefore have no indication and no motivation to assume that the putative regulatory sequences of the VE-Cad-2 gene of Ludwig et al. (2000, supra) allow vascular endothelium-specific expression excluding expression in the liver.
3o In a preferred embodiment this object has surprisingly been achieved by providing a nucleic acid fragment which comprises a functional nucleotide sequence under the control of a 5'regulatory sequence of the human VE cadherin-2 gene (hVE-CAD-2); and by providing vector systems which comprise such a fragment, in particular as part of an expression cassette, and transgenic animals, and cell lines which are obtainable by introducing such a nucleic acid fragment.
The person skilled in the art understands a nucleic acid fragment to be a nucleic acid, particularly a DNA or RNA sequence, preferably a single- or double-stranded, in particular a double-stranded DNA sequence.
1o A regulatory sequence means for the purpose of the present invention in general a nucleic acid sequence which is located upstream of the translation start (+1) of the human VE-Cad-2 gene and which controls the transcription or modulates the control of transcription of a nucleic acid sequence which is located downstream and which is connected to said sequence in the direction of the 3' end, in particular in relation to the correct start of transcription, the transcription rate, kinetics and/or the tissue specificity for vascular endothelium. The regulatory sequence has, in particular, promoter or enhancer activity.
The person skilled in the art understands a heterologous gene in relation to the 2o regulatory sequence of the VE-Cad-2 gene to be all genes which are not naturally connected to this sequence in the direction of the 3' end. The genes comprise naturally occurring genes, mutated genes or genes encoding fusion proteins.
The genes may be derived from humans, animals, plants, algae or bacteria. A
heterologous gene may also comprise two or more genes that are serially arranged and the genes may for example be separated by an internal ribosome entry site (IRES) (Vagner et al. 2001, EMBO Rep 2001 Oct;2(10):893-8) allowing all genes of such bi- or mufti-cistronic construct to be transcribed under control of the nucleic acid fragment according to the invention, the latter being preferentially located 5' upstream of the genes to be expression regulated.

A functionally active variant means for the purpose of the present invention a nucleic acid sequence which has been obtained from the sequence according to SEQ B~ No. 1 by addition, insertion, substitution or deletion of one or more nucleotides and has a sequence homology of at least 25% to the sequence of SEQ
ID No. 1, and makes vascular endothelium-specific expression possible, preferably with the exception of expression in the liver. Furthermore functionally active variants are understood as meaning all DNA sequences which are complementary to a DNA sequence, which hybridize with the reference sequence under stringent conditions and have a similar activity to the nucleic acid fragment of SEQ ID No. 1 according to the invention. A functionally active variant may also contain sequences located upstream of 105932 of the BAC clone SOg21 (Accession number AC005740) which represents the first nucleotide of the nucleic acid fragment according to SEQ ID No. 1.
1 s "Stringent hybridization conditions" are understood as meaning those conditions in which hybridization takes place at 60°C in 2.5 x SSC buffer, followed by a number of washing steps at 37°C in a lower buffer concentration, and remains stable.
2o Functionally active variants within the meaning of the present invention are also nucleic acid fragments which preferably have a sequence homology, of at least about 30%, 50%, 65% or 80%, preferentially at least about 90 %, particularly preferred of at least about 95% to the sequence according to SEQ ID No. 1.
Examples of such functionally active variants are accordingly the nucleic acid 25 fragments homologous to the nucleic acid fragment according to the invention, which originate from organisms other than the human or the mouse, preferably from non-human mammals such as, for example monkeys, pigs and rats.
In order to decide, whether a candidate nucleic acid fragment is a functionally 3o active variant, the activity of the candidate nucleic acid fragment may for example be compared with the activity of a nucleic acid fragment according to SEQ ID
No.
1. Assuming that the candidate nucleic acid fragment fulfills the criteria of a functionally active variant on the level of % sequence homology the candidate nucleic acid fragment represents a functionally active variant if the activity in the functional assay is similar to or identical with the activity exhibited by the nucleic acid fragment according to SEQ ID No. 1.
Such a functional assay is comprised for example by the assays described in examples 2 and 3 wherein the activity of the nucleic acid fragment according to the invention on the expression of a reporter gene is assessed for different cells transfected with the luciferase reporter gene construct, allowing evaluation of the strength of controlled expression and cell-specificity of the regulatory activity of the nucleic acid fragment according to the invention. By replacing the nucleic acid fragment according to the invention with the candidate nucleic acid fragment and by assessing the level of expression of the reporter gene under control of the candidate nucleic acid fragment in different cells, it is possible to identify those nucleic acid fragments among the candidate nucleic acid fragment tested which show an activity similar or identical to the activity of the nucleic acid fragment according to the invention according to SEQ ID No. 1 and thus represent a 2o functionally active variant.
In a preferred embodiment of the invention, a nucleic acid fragment according to the invention comprises a functional part of the sequence of SEQ ID No. 1.
Particular preference is given to a functional part of 1-3804 by and special preference is given to a functional part of 2724-3804 by of the sequence of SEQ
ID No. 1.
A functional part means for the purpose of the present invention a nucleic acid sequence which has been obtained from the sequence according to SEQ ID No. 1 3o by nucleotide deletion of the 5' or 3' end and which has the same function as the unmodified nucleic acid fragment according to the invention.

Sequence homology means for the purposes of the present invention the degree of similarity (% identity) of two sequences, that in the case of polynucleotides is determined by means of for example BLASTN 2.014, wherein the Filter is set off and BLOSUM is 62 (Altschul et al., 1997, Nucleic Acids Res., 25:3389-3402).
This can be checked with current sequence homology programs, for example in the Internet under http://www.h~sc.bcm.tmc.edu/SearchLauncher/.
In another embodiment the invention relates to nucleic acid fragments according to the invention having a sequence homology of at least about 80%, preferentially 1o at least about 90 %, particularly preferred of at least about 95% to the sequence according to SEQ ID No. 1 or to a functional part thereof.
Within the meaning of the invention "vascular endothelium-specific expression"
is understood to mean that the expression of a gene or a functional part thereof, or a gene coding for a fusion protein is confined to cells of the vascular endothelium.
The vascular endothelium comprises for example vascular cells which line the vessels (arteries and veins) and capillaries of the body of an adult or of an embryo, including precursors of these cells, and endothelial cells lining the cardiovascular system, including the endocardium, the heart valves and venous valves. Also included are cell lines that are derived from vascular endothelial cells such as for 2o example bovine aortic endothelial cells (BAEC) and human umbilical vein endothelial cells (HUVEC).
Within the meaning of the invention "vascular endothelium-specific expression excluding expression in the liver" is understood to mean that the expression of a gene or a functional part thereof, or a gene coding for a fusion protein is confined to cells of the vascular endothelium or cell lines just mentioned. Excluded from the expression, i.e. the expression is absent or cannot be detected, are cells of the liver comprising endothelial cells in the liver. The endothelial cells in the liver preferably include vascular endothelial cells, mesenchymal cells and smooth 3o muscle cells of the vessels and capillaries in the liver or cell lines derived from liver cells. It is further preferred for expression to be substantially absent in all non-endothelial cells of the liver, including for example hepatocytes, Kupffer cells and epithelial cells.
A nucleic acid construct means for the purposes of the present invention a nucleic s acid fragment which comprises a nucleic acid fragment according to the invention and, functionally linked thereto, one or more heterologous genes. This gene may be, for example, a marker gene or a gene which codes for a therapeutically active gene product. A marker gene might be, for example, a fluorescent protein gene such as for example green fluorescent protein (GFP), beta-galactosidase, luciferase, red fluorescent protein, yellow fluorescent protein or His, Myc or Flag tag bound to a heterologous gene. The heterologous gene will be a heterologous gene as defined hereinbefore.
A therapeutically active gene product might be for example the isoforms of heme oxygenase, the isoforms of MCP (monocyte chemoattractant protein), e.g.
15 MCP-l, GM-CSF, the isoforms of nitric oxide synthase (e.g. iNOS: inducible nitric oxide synthase; eNOS: endothelial nitric oxide synthase; nNOS: neuronal nitric oxide synthase) or the Fas ligand. Particular preference is given to heme oxygenase l, MCP-1, iNOS.
2o Further classes of genes that are of interest for use as heterologous genes applicable in genetherapy include tumor suppresser genes such as p53 (Takahashi et al., 1992, Cancer Res. 52, 2340-2343,) and reticuloblastoma or RB; cell cycle blockers such as GATA-6 (Suzuki et al, Genomics, 1996, 38, 283-290); anti-angiogenesis genes such as endostatin and angistatin (Folkman K., Nature Med.
1, 2s 27-31, 1995), anti-sense gene sequences (Wang & Becker, Nature Med. 3, 887-893, 1997), and genes encoding viral subunit vaccines (Viral Subunit Vaccines, Donnelly et al. Nature Med. 1, 583-587, 1995).
It has been found, surprisingly, that gene expression under the control of a 30 5'-hVE-CAD-2 promoter is vascular-endothelium-specific in particular in vivo in mammals. It has thus been shown for the first time that, for example, ~i-galactosidase expression under the control of the human VE-CAD-2 promoter sequence which comprises 5015 by (figure 1, -5015 by upstream of the translation start of the hVE-CAD-2 gene) is tissue-specific in transgenic mice.
Surprising observations have shown that, for example, the (3-galactosidase reporter gene was expressed under the control of the 5'-hVE-CAD-2 promoter in many, if not all, vascular endothelial cells of transgenic animals. A
significant exception to this comprised the vessels of the liver, in which switching on of the 1o reporter gene, i.e. the expression of the reporter gene was essentially undetectable.
Thus, in contrast to the prior art, the regulatory sequence, i.e. the nucleic acid fragment according to the invention makes it possible to switch on therapeutic genes in vascular endothelium but, at the same time, precludes expression of toxic gene products in the liver is provided. This makes it possible for the first time to use for example in gene therapy many genes with cytotoxic properties which are used, for example, for controlling tumors.
The invention relates to an isolated nucleic acid fragment which is composed of 2o the promoter of the human vascular endothelial cadherin-2 (hVE-CAD2) gene or functional parts and is used, for example, for expression of a transgene or heterologous gene in vascular endothelial cells. The hVE-CAD2 promoter or regulatory sequence preferably comprises the nucleic acid sequence according to the invention indicated in figure 1 according to SEQ ID No. 1 of 5015 bp, which is located directly 5' above the translation start of the hVE-CAD2 gene, or a functional equivalent of this sequence.
The invention further relates to the provision of recombinant nucleic acid molecules which are suitable for gene expression in vascular-endothelial cells and comprise the following functionally linked components:

- promoter or regulatory sequence of the hVE-CAD2 gene as described above - one or more genes, i.e. heterologous genes, or active parts thereof - a polyadenylation signal The gene or genes, i.e. heterologous genes, used may comprise any gene for which vascular-endothelium-specific expression is required.
The polyadenylation signal can be of any suitable type. A preferred embodiment comprises preferably an SV40 polyadenylation signal in the 3' position relative to to the functional sequence.
The recombinant nucleic acid molecules may comprise other conventional regulatory nucleotide sequences such as, for example, leader sequences, IRES
sequences, enhancer sequences, polyadenylation signals and sequences which ~5 control the rate of expression in terms of quantity or in its time course.
The invention further relates to the provision of vector systems which comprise the aforementioned recombinant nucleic acid molecules. Vector systems within the meaning of the invention are understood to mean vectors, comprising a 2o nucleic acid fragment according to the invention or a nucleic acid construct according to the invention. A particularly preferred vector is selected from the group consisting of plasmids, shuttle vectors, phagemids, cosmids, first or third generation of adenoviral vectors, expression vectors, and gene therapeutically active vectors.
The nucleic acid construct or the vector according to the invention may also comprise other regulatory sequences that are operatively linked to the regulatory sequences according to the invention. The regulatory sequences are "operably linked" when they are linked, preferably covalently in such a way as to place the genes under the influence of transcription regulation of all the regulatory sequences operably linked.
The additional regulatory sequences for example comprise tetracycline inducible sequences or other regulatory sequences that can be controlled by addition or removal of a transcription inducing agent. In addition cell-cycle dependant regulatory sequences such as for example control of expression during S-phase by the promoter of the E2F-1 gene allowing to drive the expression of transgenes with a specificity of action in dividing cells such as tumor cells (Parr et al. Nature 1o Medicine, 1997,3: 1145-1149) may be employed. By operatively linking different regulatory sequences one may control expression of the heterologous gene based on a combination of parameters such as tissue- or cell-specificity (i.e.
vascular endothelium-specific expression excluding expression in the liver), cell-cycle status-specificity and expression based on the presence or absence of a transcription inducing agent such as for example tetracycline, and thus confine the expression of a genetherapeutically active gene to a defined subset of cells.
Such systems are versatile and advantageous as they are adaptable to different diseases and different therapeutic needs.
2o An expression vector within the meaning of the present invention, comprises at least one nucleic acid fragment according to the invention, at least one translation initiation signal, at least one heterologous gene, one translation termination signal and/or one polyadenylation signal for the expression in prokaryotes and/or eukaryotes.
Suitable expression vectors can be prokaryotic or eukaryotic expression vectors.
Examples of prokaryotic expression vectors are, for expression in E. coli, e.g. the vectors pGEM or pUC derivatives, examples of eukaryotic expression vectors are for expression in Saccharomyces cerevisiae, e.g. the vectors p426Met25 or 3o p426GAL1 (Mumberg et al. (1994) Nucl. Acids Res., 22, 5767-5768), for expression in insect cells, e.g. baculovirus vectors such as disclosed in EP-127 839 or EP-B1-0 549 721, and for expression in mammalian cells, e.g. the vectors Rc/CMV and Rc/RSV or SV40 vectors, which are all generally obtainable.
The invention also relates to the use of a nucleic acid fragment according to the invention or a vector according to the invention, for the expression of at least one heterologous gene.
The invention furthermore relates to a knock-out gene construct, wherein the 1o knock-out gene construct comprises a nucleic acid fragment according to the invention. Knock-out gene constructs are known to the person skilled in the art, for example, from the US patents 5,625,122; US 5,698,765; US 5,583,278 and US 5,750,825.
Additionally the invention relates to a cell which comprises the nucleic acid construct according to the invention, or a vector according to the invention or a knock-out gene construct according to the invention. The cell according to the invention can be used for the expression of a heterologous gene. Another preferred cell according to the invention is a cell selected from the group 2o consisting of embryonic stem cells, embryonic germ cells, and stem cells derived from adult tissue. Particularly preferred stem cells derived from adult tissue include, but are not restricted thereto, neuronal stem cells, bone marrow stem cells, mesenchymal stem cells, hematopoietic stem cells, epithelial stem cells, digestive tract stem cells and duct stem cells.
Duct means for the purposes of the invention all ducts including the ductus arteriosus.
The cells according to the invention can be produced for example by transfecting at least one cell with a nucleic acid fragment according to the invention, with a 3o nucleic acid construct according to the invention, with a vector according to the invention or with a knock-out gene construct according to the invention. In order to make possible the introduction of nucleic acid fragment according to the invention into a cell and thus the expression of the heterologous gene in a eu-or prokaryotic cell by transfection, transformation or infection, the nucleic acid fragment can be present as a plasmid, as part of a viral or non-viral vector.
Suitable viral vectors here are particularly: baculoviruses, vaccinia viruses, adenoviruses, adeno-associated viruses and herpesviruses. Suitable non-viral vectors here are particularly: virosomes, liposomes, cationic lipids, or polylysine-conjugated DNA.
Examples of vectors applicable in genetherapy are virus vectors, for example adenovirus vectors, retroviral vectors or vectors based on replicons of RNA
viruses (Lindemann et al., 1997, Mol. Med. 3: 466-76; Springer et al., 1998, Mol.
Cell. 2: 549-58, Khromykh, 2000, Curr. Opin. Mol Ther.;2:555-569). Eukaryotic expression vectors are suitable in isolated form for gene therapy use, as naked DNA can penetrate, for example, into skin cells on topical application (Hengge et al., 1996, J. Clin. Invest. 97: 2911-6; Yu et al., 1999, J. Invest. Dermatol.
112:
370-5).
Gene therapeutically active vectors can also be obtained by complexing the 2o nucleic acid fragments according to the invention with liposomes. In the case of lipofection, small unilamellar vesicles are prepared from cationic lipids by ultrasonic treatment of the liposome suspension. The DNA is bound ionically to the surface of the liposomes, namely in such a ratio that a positive net charge remains and the plasmid DNA is complexed to 100% of the liposomes. In addition to the lipid mixtures DOTMA (1,2-dioleyloxypropyl-3-trimethylammonium bromide) and DPOE (dioleoxylphosphatidylethanolamine), meanwhile numerous novel lipid formulations were synthesized and tested for their efficiency in the transfection of various cell lines (Behr et al. 1989, Proc. Natl. Acad. Sci.
USA 86:
6982-6986; Felgner et al., 1994, J. Biol. Chem. 269:2550-2561; Gao, X. and 3o Huang, 1991, Biochim. Biophys. Acta 1189:195-203). Examples of the novel lipid formulations are DOTAP N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium ethyl-sulphate or DOGS (TRANSFECTAM; diocta-decylamidoglycylspermine). Auxiliaries which increase the transfer of nucleic acids into the cell can be, for example, proteins or peptides which are bound to DNA or synthetic peptide-DNA molecules which make possible the transport of the nucleic acid into the nucleus of the cell (Schwartz et al., 1999, Gene Therapy 6:282; Branden et al., 1999, Nature Biotech. 17:784). Auxiliaries also include molecules which make possible the release of nucleic acids into the cytoplasm of the cell (Planck et al., 1994, J. Biol. Chem. 269:12918; Kichler et al. (1997) Bioconj. Chem. 8:213) or, for example, liposomes (Uhlmann and Peymann, 1990, l0 Chem. Rev. 90, 544). The cells according to the invention can be used for the expression of a heterologous gene.
The invention further relates to a cell wherein the cell is in particular a mammalian cell, including human cell.
The invention also relates to cells according to the invention, wherein the cell is from a cell line. Preferably the cell line is selected from the group consisting of embryonic stem cells, embryonic germ cells, and stem cells, and stem cells derived from adult tissue. Particularly preferred stem cells derived from adult 2o tissue include, but are not restricted thereto, neuronal stem cells, bone marrow stem cells, mesenchymal stem cells, hematopoietic stem cells, epithelial stem cells, digestive tract stem cells and duct stem cells.
A cell line according to the invention can be produced by transfection, transformation or infection of a cell line with a nucleic acid fragment according to the invention or a vector according to the invention by methods described in detail above.
In a further embodiment, the invention also relates to a cell, wherein the cell is a 3o transgenic non-human stem cell. The stem cell comprises a nucleic acid construct according to the invention, a vector according to the invention, or a knock-out gene construct according to the invention. Transgenic non-human stem cells can be produced by transfecting the stem cell with a nucleic acid fragment according to the invention or with a vector according to the invention or with a knock-out gene construct according to the invention. Processes for the transformation of stem cells are well known to a person skilled in the art and include, for example, electroporation or microinjection.
The invention further relates to the provision of a method for the expression of to one or more genes in the vascular endothelium of animals. The method comprises the transfer of one or more genes, i.e. heterologous genes, into said animal, in which case each nucleic acid construct comprises the nucleic acid fragment according to the invention and, functionally linked thereto, one or more heterologous genes.
Said nucleic acid construct can be transferred, for example, by means of viral vectors or liposomal administration systems as described, for example, by Evans R. et al. (1994) Ann N Y Acad Sci., 716:257-264.
2o Alternatively, said construct can be transferred into an animal in such a manner that it is injected by microinjection into the fertilized egg cell from which said animal originates. This comprises a standard technique which can be carned out by a trained person skilled in the art.
The invention thus further relates to transgenic non-human animals, preferably mammals which comprise at least one nucleic acid construct according to the invention or at least one cell according to the invention, i.e. at least one non-human stem cell. Transgenic non-human animals according to the invention can for example be produced by regenerating a cell according to the invention, i.e. a 3o transgenic non-human stem cell into a transgenic non-human animal. The transgenic non-human animal according to the invention may be used, for example, for the expression of a heterologous gene according to the invention or for analyzing genetherapeutically active nucleic acids or vectors according to the invention.
The invention further relates to a method of producing a transgenic non-human animal according to the invention, wherein a cell according to the invention, i.e. a transgenic non-human stem cell is regenerated to a transgenic non-human animal.
Methods for producing transgenic animals, in particular transgenic mice, are known to the person skilled in the art from DE 196 25 049 and US 4,736,866;
US 5,625,122; US 5,698,765; US 5,583,278 and US 5,750,825 and include transgenic animals which can be produced, for example, by means of direct injection of expression vectors according to the invention into embryos or spermatocytes or by means of the transfection of expression vectors into embryonic stem cells (Polites and Pinkert: DNA Microinjection and Transgenic Animal Production, p. 15 to 68 in Pinkert, 1994: Transgenic Animal Technology:
A Laboratory Handbook, Academia Press, London, UK; Houdebine, 1997, Harwood Academic Publishers, Amsterdam, The Netherlands; Doetschman: Gene Transfer in Embryonic Stem Cells, p. 115 to 146 in Pinkert, 1994, supra; Wood:
2o Retrovirus-Mediated Gene Transfer, p. 147 to 176 in Pinkert, 1994, supra;
Monastersky: Gene Transfer Technology; Alternative Techniques and Applications, p. 177 to 220 in Pinkert, 1994, supra).
If the above described nucleic acids are integrated into so-called "targeting"
vectors or "knock-out" gene constructs (Pinkert, 1994, supra), it is possible after transfection of embryonic stem cells and homologous recombination, for example, to generate knock-out mice which, in general, as heterozygous mice, show decreased expression of the nucleic acid, while homozygous mice no longer exhibit expression of the nucleic acid. Animals produced in this way can also be 3o used for analysis, for example for the screening and for the identification of pharmacologically active substances interacting with nucleic acid fragments according to the invention.
The invention also relates to a test for the identification of pharmacologically active substances which modulate the functioning of a nucleic acid fragment according to the invention, wherein the test comprises at least one nucleic acid fragment according to the invention, at least one vector according to the invention and/or at least one cell according to the invention, if appropriate together or combined with suitable additives or auxiliaries. Such test can be used for the 1 o identification of pharmacologically active substances which modulate the functioning of the nucleic acid fragment according to the invention.
The term " pharmacologically active substance" in the sense of the present invention is understood as meaning all those molecules, compounds and/or ~5 compositions and substance mixtures which can interact under suitable conditions with the nucleic acid fragments according to the invention, if appropriate combined or together with suitable additives and/or auxiliaries. Possible pharmacologically active substances are simple chemical (organic or inorganic) molecules or compounds, but can also include peptides, proteins or complexes 2o thereof. Examples of pharmacologically active substances are organic molecules that are derived from libraries of compounds that have been analyzed for their pharmacological activity. On account of their interaction, the pharmacologically active substances can influence the functions) of the nucleic acid fragment according to the invention, in vivo or in vitro or alternatively only bind to the 25 nucleic acid fragments according to the invention or enter into other interactions of covalent or non-covalent manner with them.
Within the meaning of the invention "functioning" of the nucleic acid fragments according to the invention is understood to mean the regulatory activity of the nucleic acid fragments according to the invention onto the transcription of genes, 3o heterologues or nucleic acid sequences that can be transcribed into mRNA
and that are operatively linked to the nucleic acid fragments according to the invention. Preferably the nucleic acid sequences whose transcription is controlled or modulated by the nucleic acid fragments according to the invention are located 3' downstream from the nucleic acid fragments according to the invention. The activity of the nucleic acid fragments according to the invention comprises s initiation of transcription, modulation of the transcription, i.e.
activating or inhibiting the transcription of the transcription-controlled nucleic acid sequences.
In the case of activating the transcription, the nucleic acid fragments according to the invention may stimulate the rate of transcription resulting in a faster production of transcripts, or prolong the time of transcription leading to a larger 1o number of transcripts compared to control values. In the case of inhibiting transcription, the nucleic acid fragments according to the invention may reduce the rate of transcription resulting in a faster production of transcripts, or shorten the time of transcription leading to a reduced number of transcripts compared to control values.
In a preferred embodiment of the test, the pharmacologically active substance effects the functioning of the nucleic acid fragment according to the invention in an activating way or in an inhibitory way. In the case of "activating"
pharmacologically active substances, the transcription controlled by or modulated 2o by the nucleic acids according to the invention should be enhanced, thus leading to an increase in the amount of the gene expression product compared to the expression determined in the absence of the pharmacologically active substance.
In the case of "inhibitory" pharmacologically active substances, the transcription controlled by or modulated by the nucleic acid fragments according to the invention should be inhibited, thus leading to a decrease in the amount of the gene expression product compared to the expression level determined in the absence of the pharmacologically active substance.
A particularly preferred test is provided by modifying examples 2 and 3. For 3o example it is possible to use the hVE2-5.0 vector. Alternatively during the cloning of the luciferase reporter gene constructs the cloned PCR-fragment hVE-PCR1 employed in example 2 can be replaced for any nucleic acid fragment according to the invention to be used in the test. Then upon transfection of the luciferase reporter constructs to the different cells, for example HUVEC cells, BAEC
cells, HeLa cells, NIH-3T3 cells, the candidate pharmacologically active substances can be added to one group of transfected cells and the cells are cultured for, for example 48 hours. A control group of transfected cells does not receive candidate pharmacologically active substances and the cells are cultured in the same way as the other group of cells. Then the expression of the reporter gene is determined in the different cells and the effect of the candidate pharmacologically active 1o substance onto the transcription control of the nucleic acid fragments according to the invention is assessed. Pharmacologically active substances within the meaning of the invention are those substances that exert a modulatory, i.e. inhibitory or activating effect onto the transcription controlled and/or modulated by nucleic acid fragments according to the invention, determined by the rate and kinetics of expression of the reporter gene. The test is not limited to the use of the vectors and reporter gene employed in examples 2 and 3. Other expression vectors and other reporter genes mentioned in detail further above can also be utilized.
Such tests according to the invention may be produced by combining at least one 2o nucleic acid fragment according to the invention, at least one vector according to the invention andlor at least one cell according to the invention, with suitable additives or auxiliaries.
The identified pharmacologically active substances can, if appropriate, be combined or together with suitable additives and/or auxiliaries for the production of a diagnostic or a pharmaceutical for the prevention, treatment and/or diagnosis of disorders selected from the group consisting of vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory 3o disorders and/or immunogenic disorders. The pharmacologically active substances can be, for example inorganic or organic molecules, for example nucleic acids or analogues of nucleic acids, antisense-sequences of nucleic acid fragments according to the invention, peptides, proteins or antibodies. Examples of pharmacologically active substances are furthermore organic molecules, contained in substance libraries that have been tested for their pharmacological activity.
The invention also relates to an array immobilized on a support material comprising at least one nucleic acid fragment according to the invention and/or at least one cell according to the invention. Such array may for example be used for identifying substances that bind to nucleic acid fragments according to the 1o invention which may be used as candidate pharmacologically active substances in the test described above. Alternatively the array can be used for the analysis in connection with disorders selected from the group consisting of vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory disorders and/or immunogenic disorders.. An array according to the invention may be produced by immobilizing at least one nucleic acid fragment according to the invention, andlor at least one cell according to the invention, on a support material.
2o Carner material means for the purposes of the invention, as described for example in WO 98/18961, for example porous materials such as, for example, nitrocellulose or else nonporous materials such as, for example, glass, chemically sensitized glass.
Methods of producing and preparing such arrays by means of spotting, printing or solid-phase chemistry in connection with photolabile protective groups are for example known from US 5,744,305. Such arrays can also brought into contact with substances or a substance libraries and tested for interaction, for example for binding or change of conformation. Thus it is possible that a substance to be 3o tested contains a detectable marker, for example the substance can be labeled radioactively, fluorescence labeled or luminescence labeled or a label allowing indirect detection such as for example biotin.
In another embodiment of the invention a diagnostic is provided, wherein the diagnostic comprises at least one nucleic acid fragment according to the invention, at least one nucleic acid construct according to the invention, at least one vector according to the invention, and/or at least one cell according to the invention, if appropriate together or combined with suitable additives or auxiliaries. Such diagnostic may for example be used for the diagnosis of disorders selected from the group consisting of vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory disorders and/or immunogenic disorders.
A diagnostic according to the invention may be produced by combining at least one nucleic acid fragment according to the invention, at least one nucleic acid construct according to the invention, at least one vector according to the invention, and/or at least one cell according to the invention, with suitable additives or auxiliaries.
2o A diagnostic according to the invention preferably comprises a nucleic acid fragment according to the invention; where the nucleic acid fragment is a DNA
probe.
Suitable probes are, for example, DNA or RNA fragments having a length of about 100-1000 nucleotides, preferably having a length of about 200-500 nucleotides, particularly preferably having a length of about 300-400 nucleotides, whose sequence can be derived from the nucleic acid fragments according to SEQ
JD No. 1. or from functionally active variants thereof.

Alternatively, it is possible with the aid of the nucleic acid fragments according to the invention to synthesize oligonucleotides which are suitable as primers for a polymerase chain reaction. Using this, the nucleic acid fragments described above or parts of this can be amplified or isolated from genomic DNA. Suitable primers are, for example, DNA fragments having a length of about 10 to 100 nucleotides, preferably having a length of about 15 to 50 nucleotides, particularly preferably having a length of 20 to 30 nucleotides, whose sequence can be derived from the nucleic acid fragment according to SEQ m No. 1. This opens up a further possibility of identifying mutations of the regulatory sequence of the VE-Cad-1o gene which may cause disorders, especially genetic disorders of the vascular endothelium. Such methods are generally known to the person skilled in the art.
The invention furthermore relates to a pharmaceutical, wherein the pharmaceutical comprises at least one nucleic acid fragment according to the invention, at least one nucleic acid construct according to the invention, at least one vector according to the invention, and/or at least one cell according to the invention, if appropriate together or combined with suitable additives or auxiliaries. Particularly preferred is the use of the pharmaceutical for somatic gene therapy.
A pharmaceutical according to the invention can be produced by combining at least one nucleic acid fragment according to the invention, at least one nucleic acid construct according to the invention, at least one vector according to the invention and/or at least one cell according to the invention with suitable additives and/or excipients.
The invention also relates to a method of treating a mammal or a human by administering to the mammal or human a pharmaceutically effective amount of a pharmaceutical according to the invention. Preferably the pharmaceutical is 3o administered by means of a method selected from the group consisting of systemic injection, local injection, perfusion, or catheter-based administration.

The pharmaceutical can be introduced into the organism by either the ex vivo approach, in which cells are removed from a patient, genetically modified by DNA transfection, and subsequently re-introduced into the patient, or by in vivo approaches whereby the gene therapeutic vectors are introduced into the patient's body as naked DNA or through the use of viral and non-viral vectors or cells according to the invention.
A suitable pharmaceutical is for example one which contains the nucleic acid fragment according to the invention in naked form or in the form of one of the 1o genetherapeutically active vectors described above or in a form complexed with liposomes or gold particles. Suitable additives or auxiliaries comprise, for example, a physiological buffer solution, preferably having a pH of about 6.0-8.0, preferably of about 6.8-7.8, particularly preferably of about 7.4, and/or an osmolarity of about 200-400 milliosmol/liter, preferably of about 290-310 milliosmol/liter. In addition, additives or auxiliaries can contain suitable stabilizers, such as nuclease inhibitors, preferably complexing agents such as EDTA andlor other auxiliaries known to the person skilled in the art. The nucleic acid fragment described can be optionally administered in the form of the virus vectors described above or as liposome complexes or a gold particle complex, by 2o means of perfusion, systemic injection, or local injection or catheter-based administration.
The pharmaceutical according to the invention may also be administered through oral dosage forms, such as, for example, tablets or capsules, via the mucous membranes, for example, the nose or the oral cavity, in the form of sprays via the lung, or in the form of dispositories implanted under the skin. Transdermal therapeutic systems (TTS) are known for example, from EP 0 944 398 A1, EP 0 916 336 A1, EP 0 889 723 A1 or EP 0 852 493 A1.
3o A therapy based on the use of cells according to the invention, which express at least one heterologous gene as described above can be achieved by using cells selected from the group consisting of epithelial cells, vascular cells, liver cells, embryonic stem cells, embryonic germ cells, and stem cells derived from adult tissue. Particularly preferred stem cells derived from adult tissue include, but are not restricted thereto, neuronal stem cells, bone marrow stem cells, mesenchymal stem cells, hematopoietic stem cells, epithelial stem cells, digestive tract stem cells and duct stem cells.
In a preferred embodiment of the invention the pharmaceutical according to the invention can be used for the prevention and/or treatment of disorders selected 1o from the group consisting of vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory disorders and/or immunogenic disorders.
Furthermore, the invention relates to a method for identification of a nucleic acid fragment according to the invention comprising the steps of (1) Combining at least a nucleic acid fragment comprising a regulatory sequence of the human VE-Cad-2 gene extending 5' upstream from the translation start or a variant thereof, with a reporter gene to form a reporter 2o gene expression vector;
(2) Introducing the reporter gene expression vector into at least two different cells;
(3) Measuring the level of expression of the reporter gene;
(4) Comparing the levels of expression of the reporter gene of the different cells employed; and (5) Identifying nucleic acid fragments which allow vascular endothelium-specific expression but preferably no expression in the liver.

A variant means for the purpose of the present invention a nucleic acid sequence which has been obtained from the sequence according to SEQ ID No. 1 by addition, insertion, substitution or deletion of one or more nucleotides.
Methods for combining nucleic acid fragments with a reporter gene to form a reporter gene expression vector are for example introduced in example 2 and are generally known to the person skilled in the art.
In a preferred embodiment of the invention, the reporter gene used in the method 1o for identifying nucleic acid fragments according to the invention can be selected from the group consisting of beta-galactosidase, luciferase, green fluorescent protein (GFP), red fluorescent protein, yellow fluorescent protein or His, Myc or Flag tag bound to a heterologous gene. The heterologous gene will be a heterologous gene as defined hereinbefore.
The introduction of a reporter gene expression vector into the cells or cell lines is not limited to transfection but may also be achieved by transformation, by infection, by gene-gun bombardment or by other methods of introducing nucleic acids into cells generally known to the person skilled in the art. For the purpose of 2o the method for identification of nucleic acid fragments, the cells and cell lines according to the invention described above may be employed. In the case the reporter gene is luciferase the measurement of the level of expression of the reporter gene can for example be achieved by the Luciferase assay system or the BCA protein assay system described in example 3. Other reporter genes may require different assays for quantification of reporter gene expression products that are generally known to the person skilled in the art. By using different cells or cell lines containing the reporter gene constructs for these experiments the cell-specificity of the transcription control properties of candidate nucleic acid fragments can be compared with the transcription control properties of nucleic 3o acid fragments according to the invention, preferably nucleic acid fragments of the sequence according to SEQ ID No. 1, resulting in the identification of nucleic acid fragments that display similar or different transcription control properties compared with the nucleic acid fragments according to the invention allowing vascular endothelium-specific expression, preferably excluding expression in the liver. The identified nucleic acid fragments rnay for example induce even more persistent and/or even stronger levels of expression of the reporter gene as compared with the nucleic acid fragments according to the invention, preferably the sequence according to SEQ m No.l. Such identified nucleic acid fragments could be even better suited for use in gene therapy as the genetherapeutically active gene product could be produced over a longer period of time in larger 1o amounts.
The invention further relates to a method for the production of a gene therapeutically active vector, a pharmaceutical, or a diagnostic, wherein the nucleic acid fragment identified with the aid of the method just described is inserted into a vector containing at least one heterologous gene, and vascular endothelium-specific expression is made possible, preferably excluding expression in the liver. The methods used for the production of gene therapeutically active vectors, the pharmaceutical, or diagnostics have been described above.
The invention further relates to the use of a gene therapeutically active vector, pharmaceutical or diagnostic produced by the method just described for the diagnosis, prevention and/or therapy of disorders selected from the group consisting of vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory disorders and/or immunogenic disorders.
The invention also relates to a method for selecting and/or immortalizing vascular-endothelial cells from stem cells into which a construct of the invention 3o has been transferred by standard methods or which can be isolated by a person skilled in the art using well-characterized methods from transgenic animals or their embryos which comprise a construct of the invention.
The invention further relates to a method for selecting endothelial cells from stem cells comprising the following steps:
(1) combining at least one nucleic acid fragment according to the invention with a reporter gene to form a reporter gene expression vector;
(2) introducing the reporter gene expression vector into at least one 1o stem cell;
(3) cultivating the stem cell(s);
(4) initiating the differentiation of the cultivated stem cell(s); and (5) isolating the endothelial cells) from the cultivated cell(s).
~5 The stem cells) used in the selection method comprises) embryonic stem cells, embryonic germ cells, and stem cells derived from adult tissue. The stem cells derived from adult tissue preferably include, but are not restricted thereto, neuronal stem cells, bone marrow stem cells, mesenchymal stem cells, hematopoietic stem cells, epithelial stem cells, digestive tract stem cells and duct 20 stem cells.
Differentiation of the cultivated stem cells used in the selection method according to the invention can be initiated for example by embryoid body formation, preferably by cultivating the stem cells in solutions, by cultivating the stem cells 25 in high density, by adding cytokines, growth factors, retinoic acid or DMSO
to the cultivated cells or by adding other substances known to initiate differentiation.
Methods for selecting cells from differentiated embryonic stem cells are described in Klug et al. (J. Clin. Invest. 1996 Jul l; 98 (1):216-24) and Soria et al.
(Diabetes.
2000 Feb.; 49 (2):157-62).

A preffered embodiment of the invention relates to a selection method where the reporter gene is an antibiotic resistance gene, and the endothelial cells) is (are) isolated by collecting the differentiated endothelial cells) after addition of a suitable antibiotic in step (3) or (4).
An antibiotic according to the invention means an antibiotic against which the antibiotic resistance gene used in the reporter gene expression vector generates resistance. After addition of the antibiotic to the cultivated stem cells, the only stem cells which survive and differentiate are essentially those containing the 1o reporter gene expression vector.
A further preferred embodiment of the invention provides a selection method where the antibiotic resistance gene is selected from the group consisting of hygromycin resistance gene (hph), zeocin resistance gene (Sh ble), puromycin resistance gene (pacA) and gentamycin or 6418 resistance gene (aph).
In a further particularly preferred embodiment of the invention, the selection method according to the invention relates to a selection method where the reporter gene is selected from the group consisting of luciferase, green fluorescent protein, 2o red fluorescent protein, and yellow fluorescent protein, and the endothelial cells) is (are) isolated from the cultivated cells) by means of fluorescence-activated cell sorting (FACS).
Also provided is a selection method according to the invention where a reporter gene is selected from the group consisting of beta-galactosidase, luciferase, green fluorescent protein, red fluorescent protein, yellow fluorescent protein, or a His, Myc or Flag tag bound to a heterologous gene, and the endothelial cells) is (are) isolated from the cultivated cells by means of affinity purification. A
heterologous gene means a heterologous gene as defined hereinbefore.

The selected and/or immortalized vascular-endothelial cells can be employed for cell-mediated transplantation, for generating artificial vessels in vitro, for producing artificial heart valves or venous valves and for somatic gene transfer in vivo.
The invention further relates to a method for producing an artificial tissue or organ comprising at least one endothelial cell, where at least one endothelial cell obtained by the selection method according to the invention which is described hereinbefore is combined and cultivated with at least one suitable cell andlor a 1o support in order to generate the artificial tissue or organ. The artificial tissue is preferably selected from the group consisting of vessels, heart valves and venous valves.
A suitable cell means for the purposes of the invention a feeder cell or another cell apart from the endothelial cell obtained by the selection method, which is necessary to form the artificial organ or tissue or which is part of the artificial organ or tissue. Such suitable cells are known to the person skilled in the art.
A support means for the purposes of the invention a substance, molecule or 2o material or matrix which serves as chemical, physiological or mechanical support for the tissue or organ to be produced. Such supports are known to the person skilled in the art.
The invention further relates to a method for testing the pharmacological activity of a pharmacological substance, where at least one endothelial cell obtained by the selection method according to the invention is exposed to the pharmacological substances, and the pharmacological activity of the pharmacological substance is determined. The pharmacological substance means for the purposes of the invention any substances, compounds, mixtures or compositions. Possible 3o pharmacological substances are simple chemical (organic or inorganic) molecules or compounds, but they may also include peptides, proteins or complexes thereof.
Examples of pharmacological substances are toxic substances, compounds, mixtures or compositions; or organic molecules which are derived from compound libraries and which have been investigated for their pharmacological activity.
Pharmacological activity means for the purposes of the invention every response of a tested cell exposed to the pharmacological substance, at the level of morphology, metabolism, physiology or genetic activity.
The pharmacological activity of the tested cell can be investigated for example at the level of vitality or apoptosis, i.e. the selected endothelial cell is exposed to the toxic substance, compound, mixture or composition, and the vitality of the cell is determined. This makes it possible for example to identify endothelial cells which are resistant to a tested toxic substance or make it possible to identify toxic substances, compounds, mixtures or compositions which are useful for inducing apoptosis. Vitality and apoptosis tests are generally known to a person skilled in the art.
2o Methods according to the invention for testing the pharmacological activity can also be used for high throughput screening investigations on pharmacological substances which show interesting diagnostic or therapeutic properties in a selected cell type. It is possible in the same way to find the susceptibility or sensitivity to the pharmacological substance of a selected endothelial cell exposed to the pharmacological substance, making it possible to identify pharmacological substances having interesting susceptibility or sensitivity properties.
These cells can also be used for establishing an in vitro vascular-endothelial cell model. Such an in vitro model is suitable according to the invention for investigating substances potentially with therapeutic activity, in particular for pharmacological investigations.
The invention is described in detail in the following exemplary embodiments.
Reference is made therein to the appended examples and figures. It must be taken into account in this connection that the following descriptions comprise only an illustration. The general validity of the invention as described above is not restricted by the examples.
1o Description of the fi ug-re_s Fig. 1 shows the nucleic acid sequence of a promoter, which is 5015 base-pairs long and is located 5'-upstream of the translation start (ATGATG;
position 5016-5021), of the human VE-CAD2 gene. The start of transcription is located at position 3805 (+1). The recognition sequence for the restriction endonuclease EcoRV is located at position 513-518, for SacI at position 2725-2730, for HindIII at position 3704-3709 and for SmaI at position 4385-4390.
Fig.2 shows the structure of luciferase reporter gene constructs for 2o investigating the promoter activity in cell culture. The top diagram (A) shows the genomic structure of the human VE-CAD2 promoter. The first exon after the translation start has been indicated by a box. The arrow inside the box indicates the direction of reading of the VE-CAD2 gene.
Cleavage sites for restriction endonucleases used for the cloning have been indicated (EcoRV, SacI, HindIII, SmaI).
The constructs were cloned using a PCR fragment which comprises 5015 by of the human VE-CAD2 promoter and which is called hVE-PCR1 (B) (cf. example 1). hVE-PCR1 comprises at the 5' end a cleavage site for the restriction enzyme KpnI and at the 3' end a cleavage 3o site for the restriction enzyme ScaI.

The other diagrams in this figure (C-G) show the structure of reporter gene constructs obtained by cloning promoter fragments into the plasmid pGL3-basic (Promega). pGL3-basic comprises the coding sequence of the luciferase marker. The luciferase (Luc) gene has been indicated by a box (Luc). The bacterial portions of the circular plasmids have not been depicted. The length of the cloned promoter fragments has been indicated in base pairs (bp) and is 5015 by for hVE2-S.0 (C), 4500 by for hVE2 4.5 (D), 2289 by for hVE2-2.3 (E), 1310 by for hVE2-1.3 (F) and 627 by for hVE2-0.6 (G). Restriction cleavage sites used for cloning the 1o constructs and destroyed during this are depicted in parentheses.
(bp = base pairs, kb = kilobases) Fig. 3 shows the luciferase activity in various cells after transfection with the various reporter gene constructs from fig. 2 in cell culture. The means and the standard deviation (SEM) for the measured light units relative to the promoterless plasmid pGL3-basic, which was used as control plasmid, have been indicated. The following cells were transfected:
primary endothelial cells from the human umbilical vein (human umbilical vein endothelial cells, HUVEC), primary endothelial cells from 2o the bovine aorta (bovine aortic endothelial cells, BAEC) human tumor cell line (HeLa), fibroblast cell line from rodents (NIH-3T3). Each transfection was repeated 6 times.
Fig. 4 shows the diagrammatic structure of constructs for generating transgenic mice. (A) shows the genomic structure of the hVE-CAD2 promoter which has already been described for figure 1 (A). The construct pRZ-hVE-Z1 (B) comprises 5015 by of the VE-CAD2 promoter coupled to the ~i-galactosidase reporter gene (LacZ gene; depicted as a white box). The graph shows the proportion of the construct obtained by 3o cutting with the enzymes SaII and NotI and isolated for injection into fertilized mouse egg cells. Transgenic founder animals were identified by carrying out a Southern blot analysis with the sample Z2. This comprises a 1938 by EcoRI-EcoRV fragment from the LacZ gene. Indicated cleavage sites for the restriction enzymes EcoRI, EcoRV, SacI, HindIII, SmaI, SaII and NotI were used for cloning the constructs and for analyzing transgenic animals.
Fig. 5 shows the X-Gal staining in transgenic pRZ-hVE-Z1 mice of the two established transgenic strains (the strains were called pRZ-hVE-Z1-#10 and pRZ-hVE-Z1-#54). The pattern and intensity of the staining were the same in all investigated developmental stages in both strains. (A) shows 1o an embryo on day 10.5 of development, while the embryo in (B) was 12.5 days old (embryos of the transgenic strain pRZ-hVE-Z1-#10 are shown). In these embryos the lacZ reporter gene is expressed in developing vessel structures such as, for example, capillaries in the head region, vessels located between the somites, the head primordia and the developing primordia of the limbs (limb buds) (embryonic day 12.5; fig.
5B). It was confirmed in sections of these embryos that the beta-galactosidase protein was restricted to the vascular endothelium (data on sections not shown). The embryo in (C) was stained on embryonic day 15.5. This was done by first removing the skin and then dividing the 2o embryo lengthwise in order to allow the dye to penetrate in. In this stage, the reporter gene was intensively expressed in all blood vessels of the developing skeletal muscle of the entire embryo. The intense staining leads to all blood vessels appearing black or dark gray in the black and white image in fig. 5 (C). Organs in this stage were also analyzed by beta-galactosidase staining (data not shown). Staining was restricted to the vascular endothelium of the kidney, in the heart to every size of vessels and in the lung. In contrast thereto, no staining was observable in blood vessels of the liver, including the central vein. Individual organs from embryos in day 18.5 of development were investigated for 3o expression of (3-galactosidase. Reporter gene expression was restricted in all the organs investigated, including skeletal muscle, lung, spleen, kidney, heart and pancreas, to the monocellular vascular endothelium of the blood vessels and capillaries. For example, (D) shows the ventral view and (E) shows the dorsal view of the heart. (F) shows an enlarged partial view of figure (E), with the X-Gal staining in all the veins and arteries opening into the heart having been made clear. In the original color figures, only the vascular cells lining the developing arteries and veins appear bright blue (positive X-Gal stain). (G)-(I) depict cross sections of the heart shown in (D)-(F). (G) shows the staining in the vascular endothelium of the aorta, while (3-galactosidase-positive vascular-endothelial cells in a coronary vessel have been depicted in (H).
to (I) shows a cross section in which a coronary vessel is cut lengthwise. It is evident in these images (G, H) that the X-Gal-positive cells (which appear black or dark gray in the black and white images) are entirely restricted to the vascular endothelial cells which line the blood vessels, as is also confirmed by analysis of the color images (data not shown).
Fig. 6 Northern Blot analysis of VE-Cad-2 expression in adult mouse and human tissue. (A) mRNA of the marine VE-Cad-2 is located at the approximately 7 kb position. Highly vascularized organs such as the lung, heart, spleen, liver and kidney show pronounced expression of 2o VE-Cad-2. (B) mRNA of the human VE-Cad-2 is located approximately at the 4.4 kb position. Human VE-Cad-2 expression is detectable in all the organs investigated, except the brain, with markedly high intensities of expression in the heart, placenta, lung and liver.
Fig. 7 RT-PCR analysis of VE-Cad-2 and beta-galactosidase expression in tissue from adult transgenic pRZ-hVE-Z1#10 (left-hand column), pRZ-hVE-Z1 (middle column) and wild-type CD1 mice (right-hand column).
The PCR products in the middle and right-hand columns (separated by a broken line) are located on the same gel, and thus there is only one 3o molecular weight marker lane (M) for these two columns, whereas the PCR products in the left-hand column were run on a separate gel.

Ubiquitously expressed genes - the housekeeping gene GAPDH and the ribosomal gene L7 - were detected in comparable intensities in the cDNA
from all investigated tissues in both mice, the transgenic and the wild-type. VE-Cad-2 expression was also detectable in slightly different intensities in all mice and all investigated tissues including the lung, liver, heart, spleen and kidney. Beta-galactosidase expression was observed only in transgenic animals of the established strain pRZ-hVE-Z 1 # 10 and pRZ-hVE-Z 1 #54, but not in the wild-type CD 1 mice. In the transgenic animals, a beta-galactosidase-specific PCR product of 400 by 1o was detected in the cDNA derived from the lung, heart, spleen and kidney but, surprisingly, the PCR product was completely absent from the liver.
Examples Example 1' Cloning of the human VE-CAD2 promoter The published cDNA for the mouse VE-CAD2 gene (P. Telo et al., JBC 1998;
accession number Y08715) was used for a Genbank search (NCBI, basic BLAST). In this, sequence sections showing a sequence homology of 81-96% to 2o the mouse cDNA were found on a human BAC (bacterial artificial chromosome).
The BAC is described as follows: Homo sapiens chromosome 5p, BAC clone 50g21 (LBNL H154), complete sequence. The accession number is AC005740.
The published sequence comprises 18670 base pairs. The BAC was obtained as E. coli stab culture from the Humane Genome Center, DOE Joint Genome Institute, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA.
The bacterial clone was cultivated and the BAC DNA was isolated using a DNA
purification kit (Qiagen, Hilden, Germany). It was possible on the basis of the homology to the cDNA of the marine VE-CAD2 gene to identify the translation start of the human gene, which is located at position 110948 of the published 3o BAC sequence. The promoter was isolated by the PCR technique (polymerase chain reaction). The primers were defined on the basis of the BAC sequence.
The 5' primer hVE2-chrF2 (forward primer) is located between position 105933 and 105959 of the BAC sequence. At the 5' end, the primer comprises a recognition site for the restriction enzyme KpnI and 6 random bases (n). This makes it possible to cut the resulting PCR product. The primer sequence is: 5'- nnn nnn ggt acc cag aag tag tgc cct tcc tct cga - 3' (SEQ ID No 2). The orientation of the 3' primer hVE2-UP-ATG (reverse primer) is defined contrary to the direction of reading of the VE-CAD2 gene. The primer is located between position 110947 and 110924 of the BAC sequence and is positioned so that the translation start of the human VE-CAD2 gene (adenosine at position 110948) is not present.
1o hVE2-UP-ATG comprises at the 5' end a recognition site for the restriction enzyme ScaI and 6 randomly chosen nucleotides and has the following sequence:
5'- nnn nnn agt act get tac cgc aac gtg ggc tag att - 3' (SEQ ID No 3). 50 ng of purified BAC DNA are used in a SO ~1 PCR mixture for amplifying the promoter.
The resulting fragment, which is called hVE-PCR1, comprises 5015 by of the hVE-CAD2 promoter and is cut with KpnI and ScaI for cloning reporter gene constructs (cf. example 2 and example 4). The fragment was checked by sequencing.
Example 2: Cloning of luciferase reporter gene constructs 2o The constructs were cloned by using the PCR fragment hVE-PCRI (cf.
example 1). hVE-PCRl contains at the 5' end a cleavage site for the restriction enzyme KpnI and at the 3' end a cleavage site for the restriction enzyme ScaI.
The construct hVE2-5.0 kb (cf. fig. 1 C) is prepared by cutting hVE-PCR1 with KpnI
and ScaI and cloning into the KpnI and SmaI cleavage sites of the plasmid pGL3-basic (ScaI and SmaI cleavage sites are compatible). All the other constructs are truncations of the construct hVE-2 5.0 kb at the 5' end of the promoter. The 3' end of the promoter at the junction with the luciferase gene is identical in all the constructs. In order to obtain hVE2-4.5 kb (cf. fig. 1 D), hVE-PCR1 is cut with EcoRV and ScaI, and the 4.5 kb fragment is cloned into the SmaI cleavage site of 3o pGL3-basic. All the cleavage sites are compatible. hVE2-2.3 kb (cf. fig. 1 E) is obtained by cutting hVE-PCR1 with SacI and ScaI. The 2.3 kb fragment is cloned into the plasmid pGL3 basic cut with SacI and SmaI. For cloning hVE2-I.3 kb (cf. fig. 1 F), hVE-PCRI is cut with HindIII and ScaI and cloned into the vector pGL3-basic which has been cut with KpnI and SmaI. This is preceded by blunt-ending by treatment with T4-DNA polymerase. The construct hVE2-0.6 (cf.
fig. 1 g) is produced in the same way, cutting hVE-PCRl with SmaI and ScaI and cloning the resulting 0.6 kb fragment into the pGL3-basic which has been cut with SmaI.
Example 3: Transfection experiments in cell culture 1o Luciferase reporter gene constructs (cf. example 2) were used to transfect the following cells: primary endothelial cells from the human umbilical vein (human umbilical vein endothelial cells, HUVEC), primary endothelial cells from the bovine aorta (bovine aortic endothelial cells, BAEC), human tumor cell line (HeLa), fibroblast cell line from rodents (NIH-3T3). 2-3x10$ cells are seeded in DMEM with 10% FCS in a 6-well cell culture plate. The transfection is carried out 12 hours after seeding the cells. 1.5 ~g of the respective plasmid DNA are mixed with 3.25 p1 of ExGen solution (MBI Fermentas) and a I50 mM sodium chloride solution in a I00 ~1 mixture. The transfection solution is mixed with 1 ml serum-free medium and placed on the cells for one hour. The cells are then 2o cultivated in serum-containing medium for 48 hours. To determine luciferase (Luciferase Assay System, Promega, Cat. No. E4030) and measure protein (BCA-Protein Assay, Pierce, Cat. No. 23223), the cells are washed with PBS buffer and dissolved in 200 p1 of lysis buffer. 20 p1 portions of the lysate are used for each assay. The procedure is in accordance with the manufacturers' instructions.
The luciferase measurement took place in a Packard luminometer. Each sample is measured twice. The luciferase values were adjusted via the protein determination. Besides the luciferase constructs mentioned in example 2, the promoterless plasmid pGL3-basic was transfected. The results of 6 independent transfection experiments were evaluated and averaged. These values were divided 3o by the mean values after transfection with pGL3-basic and thus related to this control plasmid. The results are depicted in figure 3.

1. Expression of luciferase reporter gene constructs in cell culture The results depicted in figure 3 show that the hVE-CAD-2 promoter is specifically switched on in cultivated endothelial cells. Reporter gene constructs containing the promoter fragments comprising 5.0 kb, 4.5 kb and 2.3 kb respectively control gene expression which, in venous (HUVEC) and arterial (BAEC) endothelial cells, is 35-37 times that of the promoterless control plasmid.
At the same time, only a background activity is observed in control cells. It is not possible on the basis of these data to detect a difference in specificity and level of expression between the three constructs hVE2-5.0, hVE2-4.5 and hVE2-2.3.
1o By comparison with these, the construct hVE2-1.3 causes clear switching on of the luciferase gene only in arterial endothelial cells, while only a background activity not differing from expression in control cell lines is to be observed in venous cells.
The cDNA for the hVE-CAD-2 gene is published in the NCBI database under accession number AF240635. According to this, the gene transcription start is located 1211 by above the translation start (compare figure 1 ). The construct hVE2-1.3 contains 1310 by above the translation starting point and thus evidently only 99 by of the promoter 5' above the transcription start. This short promoter fragment nevertheless leads to limited endothelial cell-specific expression.
In contrast thereto, the construct hVE2-0.6, which contains exclusively transcribed sequence, as expected shows no promoter activity at all, irrespective of the cell line used.
In summary, the data show that the hVE-CAD-2 promoter mediates specific gene expression in vascular-endothelial cells in vitro. In this connection, no unambiguous difference between constructs which contain promoter fragments 3o S.0 kb, 4.5 kb and 2.3 kb in size is evident. Accordingly, even a sequence comprising 2.3 kb is sufficient to cause high-level and vascular-endothelial cell-specific gene expression in cell culture. However, this observation does not permit any reliable statement to be made about possible differences between said promoter fragments on gene expression in vivo.
Example 4: Production of transgenic mice by microinjection of ~i_galactosidase constructs.
For cloning pRZ-hVE-Z1, the PCR product hVE-PCRl (cf. example 1 and figure 1) which comprises the human VE-CAD2 promoter is cut with KpnI and 1o ScaI and cloned into the plasmid pPD 46.21 (A. Fire et al., Gene 93, 189-98, 1990) which has been cut with SaII and SmaI. For this purpose, the DNA
fragments are blunt-ended by treatment with T4-DNA polymerase. The plasmid pPD 46.21 comprises the coding sequence for the ~3-galactosidase reporter gene (lacZ) from E. coli coupled to a nuclear localization signal which causes the protein to be transported into the cell nucleus. The overall size of the construct pRZ-hVE-Z1 is 11.1 kb. For microinjection into mouse egg cells, the construct is cut with SaII and NotI to result in the fragment 8.5 kb in size which is depicted in fig. 4 (B) and which comprises the 5015 by VE-CAD2 promoter and the LacZ
gene (with an SV-40 polyadenylation signal; not depicted).
After purification of the fragment by electrophoresis, transgenic mice are produced as described by Hogan B. (Hogan B. et al. Manipulating the Mouse Embryo; a laboratory manual; Second Edition; Cold Spring Harbor Laboratory Press). The DNA solution is injected into the pronucleus of fertilized oocytes of the mouse strain CD 1. For the analysis, DNA is isolated from pieces cut from the tails of founder animals. For this purpose, the pieces are incubated in 500 u1 of lysis buffer (50 ~M tris/HCL, pH 8.0, 100 mM EDTA 100 mM NaCI, 1% SDS
addition of 35 ~M proteinase K (10 mg/ml)) at 55°C overnight, and the DNA is obtained by isopropanol precipitation. To detect a 6.4 kb transgene band, the mouse DNA is digested with EcoRI. The probe used for the Southern blot hybridization is the 1.9 kb EcoRV-EcoRI fragment (Z2) (cf. fig. 4). For further investigation, positive animals are crossed with female or male wild-type CD 1 partners. The progeny show mendelian distribution of the transgene. It was thus possible to confirm integration of the construct into the germ line.
Example 5: X-Gal staim'ng of trans~enic mice Embryos were dissected out on the desired embryonic day and incubated in fixing solution (2% paraformaldehyde; 0.1 M PIPES pH 6.9; 2 mM MgClZ; 2 mM
EGTA) at room temperature for 10-20 min. Embryos with an age greater than 14.5 exibryonic days were halved medianly and fixed for a further 20 min.
After 1o washing in PBS/0.01% sodium deoxycholate/0.02% NP-40 three times, the embryos were incubated in staining solution (6 mM K3Fe(CN)6; 6 mM
K4Fe(CN)6; 0.02% NP-40; 0.25 mM sodium deoxycholate; 1 x PBS; 0.01%
X-GaI/DMSO) at 30°C for 12-16 h. They were then washed three times with PBS.
For further analyses, an incubation was carned out in 15% sucroselPBS
overnight.
Whole embryos or organs were then transferred into OCT freezing medium (Miles), were frozen on dry ice and stored at ~80°C. For the analysis, cryosections were prepared with a thickness of 10-40 ~,m.
2o 2. Production of transgenic mouse lines having the construct pRZ h VE-ZI
It was found by Southern hybridization that, of 62 progeny obtained after microinjection of the construct, a total of 5 animals contained the transgene in their genome. Two of these seven animals transmitted the construct to the progeny. Comparable (3-galactosidase expression was observed in both lines.
3. Vascular endothelium-specific activity of the S.0 kb hYE-CAD-2 promoter in transgenic mice Analysis of the [3-galactosidase staining in transgenic animals shows promoter activity even during early embryonic development on day I 1.5 in most if not all vessels of transgenic animals. Detailed analysis of whole embryos and of the corresponding sections showed reporter gene activation in the vascular endothelium of the umbilical cord, of the brain, of the muscles and of the developing organs. Expression in fine capillaries was also observed. The promoter s activity was absolutely specific in this case, so that there was no detectable ectopic reporter gene expression in non-endothelial cells. The two lines obtained show a substantially identical expression pattern. It is evident that the promoter fragment used contains all the regulatory elements for bringing about insertion site-independent gene expression in transgenic mice.
Further analysis took place on older stages (on embryonic day 15.5 and at the time of delivery on embryonic day 18.5). There was also detection in these stages of (3-galactosidase staining in most if not all vessels of the organs investigated, including the muscles. Whereas the vessels in organs such as heart, lung, kidney, 15 stomach and intestine showed intense (3-galactosidase activity, in contrast thereto no staining at all was detectable in the liver. This observation is absolutely surprising because it was possible to show a Northern blot analysis expression both of the human and of the marine VE-CAD-2 gene in all vascularized organs including the liver (data not shown). It is evident that the hVE-CAD-2 promoter 2o consists of a plurality of regulatory elements, of which some specifically bring about switching on of the gene in the vascular endothelium of liver vessels.
Accordingly, the human promoter sequence comprising 501 S by used in this case comprises all the regulatory elements for gene expression throughout the vascular endothelium with the exception of the vascular-endothelial cells in the liver.
An 25 alternative possibility is that these liver-specific elements are present but, because of specific inhibiting elements present in the promoter fragment, they do not lead to gene expression in this organ.
In summary, it can be stated that the 5015 kb promoter, which is preferably used 30 according to the invention, of the human VE-CAD-2 gene surprisingly comprises all the regulatory elements for vascular-endothelial cell-specific expression in the blood vessels but, at the same time, precludes switching on of the gene in the blood vessels of the liver. The promoter is thus suitable for numerous gene therapy approaches which require gene expression in the vascular endothelium but with which, owing to the therapeutic gene used, there is expected to be a risk of systemic side effects due to liver damage.
Example 6: Northern blot analysis of adult tissue 1. Production of the probes for the Northern blot analysis 1o Mouse VE-Cad-2 probes: primers for amplifying a 437 by fragment were produced using the published sequence of the mouse VE-Cad-2 cDNA (Telo et al.
1998, J. Biol. Chem. 273, 17565-17572). The fragment was called mVE-Northern. The primer sequences are: mVE-2F2 (forward primer): TAC TTC TGC
CAT TCC TGC TAG G (SEQ >D No. 4) and mVE-2R2 (reverse primer): GAC
CTT TAG AGT CTC TCA CGG A (SEQ m No. 5). A cDNA generated from mouse cardiac tissue was used as template for amplification of the probe. The following PCR conditions were used: 2.0 ~1 (50 ng) of cDNA, 0.5 w1 (5 pM) of primer mVE-2F2, 0.5 w1 (5 pM) of primer mVE-2R2, 1.5 mM MgCl2 and 9.5 ~.1 of H20 were mixed to form a 12.5 ~1 master mix. The master mix was made up 2o with Taq polymerase and H20 to a total volume of 25 p1. The reaction vessel was transferred into a Thermocycler, and the following PCR program was used:
15 min at 95°C, followed by 30 cycles of: 30 sec at 95°C, 30 sec at 55°C, 60 sec at 72°C, followed by 7 min at 72°C and 5 min at 95°C.
Human VE-Cad-2 probes: the presumed open reading frame of the human VE-Cad-2 gene was identified using the murine VE-Cad-2 gene sequence (Telo et al., 1998, supra) in a Genbank database analysis. On the basis of the aligned mouse VE-Cad-2 gene sequence and the BAC clone 50621 sequence (accession number AC005740), primers which recognized the human VE-Cad-2 gene 3o translation start were designed, and another primer which recognizes a sequence located 3' downstream from the translation start was also designed, the two primers covering a human VE-Cad-2 fragment with a length of 432 bp. The following primer sequences were used to amplify a 432 by human VE-Cad-2 fragment by PCR: hVE-2S2 (forward primer): GTA AGC ATG ATG CAA CTT
CTG (SEQ ID No. 6) and an hVE-2A1 (reverse primer): AAG GGT TTC TGC
TCG TCC TTG (SEQ ID No. 7). A cDNA derived from HUVEC cells was used as template for the PCR. The PCR conditions used were the same as were used to generate the mouse VE-Cad-2 probes. The isolated human fragment was called hVE-Northern.
2. Northern blot hybridization:
Multiple tissue Northern blots (MTN), Clonetech, were used. Mouse: (MTN) blot cat. No.: 7762-1 lot. No. 9030523 and human: (MTN) blot cat. No.: 7760-1 lot No. 9010567. All buffers and solutions were provided by the manufacturer and used according to the manufacturer's instructions. 50 ng of the Northern probes described above were used with the alpha-3zP-dCTP (ICN) using the primer IT
kit (Stratagene) in accordance with the manufacturer's instructions. After a final washing step, a Kodak x-ray film was exposed using the hybridized blot for 96 h.
3. Results Results are shown in figure 6. The mRNA of the marine VE-Cad-2 is located in an approximately 7 kb position. Highly vascularized organs such as the lung, heart, spleen, liver and kidney show pronounced expression of VE-Cad-2 in the mouse (fig. 6 (A)). The mRNA of the human VE-Cad-2 is located in an approximately 4.4 kb position. Expression of human VE-Cad-2 is detectable in all the organs investigated except the brain, of markedly high intensity in the heart, placenta, lung and liver.

Example 7' RT-PCR analysis of adult trans~enic mice RNA was isolated from adult transgenic mice produced as in example l and 2.
Organs from a one-year old pRZ-hVE-Z1#10 female, from a one-year old pRZ-hVE-Z1#54 female and from a one-year old CD1 wild-type female (control) were cautiously isolated and used to prepare RNA by the Trizol method (Life Technologies, cat. No. 15596-026 lot No.: 1011120) in accordance with the manufacturer's instructions. Tissue was homogenized with an Ultra-Turrax. The following organs were tested: lung, liver, heart, spleen and kidney. 500 ng of RNA from each organ were used as template for the reverse transcriptions (RT) to using the expand reverse transcriptase kit which comprises a DNase digestion step (Roche cat. No: 1785-834 lot No: 92035623). The resulting cDNA was diluted in a total volume of 20 ~,1. 2 ~,l of the cDNA preparation were used for the PCR, using the hot star tag TM master mix kit (Qiagen 1000 units cat. No: 203445 lot No: 10921919). The following primer pairs were used:
1. GAPDH R3 (forward primer): CAC CAC CTT CTT GAT GTC ATC A
(SEQ ID No. 8) and F2 (reverse primer): GCC ATC AAT GAC CCC TTC
ATT G (SEQ m No. 9). The amplified PCR fragment is 693 by long.
2. L7 up (forward primer): GGT AGT GGT CAA ATG GCG ATT (SEQ m No. 10) and L7 down (reverse primer): GCC ACC AAT CCC CAT ATG
2o GAA (SEQ ID No. 11). The amplified PCR fragment is 206 by long.
3. mVE F1 (forward primer): GTC CGG TCC TCA TCA GAT TCT (SEQ m No. 12) and R2 (reverse primer): GTG TGC TGC CCC AAC AAC ATT
(SEQ ID No. 13). The amplified PCR fragment is 612 by long.
4. LacZ F1 (forward primer): GGT AGT GGT CAA ATG GCG ATT (SEQ m No. 14) and R3 (reverse primer): GCC ACC AAT CCC CAT ATG GAA
(SEQ ID No. 15). The amplified PCR fragment is 400 by long.
The following PCR conditions were used: the following were mixed in separate reaction vessels: 2.0 w1 of cDNA, 0.5 w1 (5 pM) of primer (reverse primer), 0.5 w1 (5 pM) of primer (forward primer), 1.5 mM of MgCl2 and 9.5 u1 of H20 to form a 12.5 ~1 master mix. Then each reaction vessel was made up to a total volume of 25 ~1 with Taq polymerase and H20. The reaction vessels were transferred into a Thermocycler, and the following PCR program was used: 15 min at 95°C, followed by 30 cycles of 30 sec at 95°C, 30 sec at 55°C and 60 sec at 72°C
followed by 7 min at 72°C and S min at 95°C. Then 15 ~1 of each PCR were fractionated by electrophoresis on a 1.5% agarose gel with EtBr and then analyzed.
As shown in figure 7 it was possible for the isolated cDNAs to be amplified successfully and for the ubiquitously expressed genes, the GAPDH gene 1o (housekeeping gene) and L7 gene (ribosomal gene) to be observed in all tissues tested. All the cDNAs used were therefore of suitable quality. Expression of the mouse VE-Cad-2 gene was likewise observed in all organs tested. This observation is consistent with the Northern blot analysis of the adult mice and of the human tissue as described in example 6 and shown in figure 6. As expected, RT-PCR analysis of beta-galactosidase gene expression showed no detectable PCR product in the transgenic CD 1 wild-type mouse (negative control, fig. 7, right-hand column, foot of image). In contrast thereto, it was possible in both transgenic strains pRZ-hVE-Z1#10 (fig. 7, left-hand column, foot of image) and pRZ-hVE-Z1#54 (fig. 7, middle column, foot of image) to obtain a beta-2o galactosidase product with a length of 400 by in tissues derived from adult lung, heart, spleen and kidney (fig.7, left-hand and middle column, foot of image, lanes 2, 4, 5 and 6). The hVE promoter with a length of 5015 by can therefore be used to control reporter gene expression in all these organs irrespective of the site of integration of the transgene in the genome. Surprisingly, essentially no beta-galactosidase gene expression was detectable in adult liver from the two independently generated transgenic strains (fig. 7, left-hand and middle column, foot of image, lane 3). In view of the Northern blot data in figure 6 and the RT-PCR expression patterns of the endogenous mouse VE-Cad-2 gene as discussed hereinbefore, this is absolutely unexpected, and both methods show that 3o the VE-Cad-2 gene is unambiguously expressed in adult mouse and human liver.
The essentially absent transgene expression in the adult liver of both transgenic strains is evidently a characteristic feature of the 5015 by hVE promoter. The RT-PCR data are consistent with the beta-galactosidase expression patterns observed in the transgenic embryos, as shown in fig. 5. In summary, these results clearly show that the SO15 by hVE-Cad promoter can be used for the expression of any transgene of interest in vascular endothelial cells of any organ in embryos s or adult animals, with expression being substantially absent from the liver.
These features are also useful and advantageous for many applications in medicine, gene therapy and related areas.
It will be apparent to those skilled in the art that various modifications can be 1o made to the compositions and processes of the invention. Thus, it is intended that the present invention should also cover such modifications and variations, provided they come within the scope of the claims and their equivalents.
Priority application US 60/249,390 was filed on November 16, 2001. All 1 s publications cited herein are incorporated in their entireties by reference.

_1_ SEQUENCE LISTING
SEQUENZ PROTOKOLL
<1i0> Cardioa AG
Zweigerdt, Robert Rildiger, Manfred Thomas, Moll <120> Vascular endothelium-spec'rfic promoter of the human vascular endothelial Cadherin-2 (hVE-Cad-2) gene and its therapeutic use <230> C34480aS
<1S0> US 60/249,390 <151> cool-11-is <160> 15 <170> PatentTn version 3.1 <210> 1 <2i1> 5021 <212> DNA
<2i3> Homo Sapiens <400> 1 cagaagtagt gcccttcctc tcgaaggttg gggagaaaga catgtgtgtg taactcttat 60 caacccagaa ctcaagtggt ccagttttgg gtccatgtta tggttactaa cgcacaaaat 120 aaatttggaa aagaacatac aaatcttttc ttgtattttt cttgaaaatc atgaaggagg 180 tctataattt gcaataacga attcsaaagt ccttttcatc ttcattctac ctccttcaca 240 gcctctgctg ttgtctgctg tattgctctc cagaattttc tcagattgca gaaatctccc 300 tcctcccgtc tatttatttt tatcacgtaa atcascttag cctctctttc cagcaaaaaq 360 gtcaaatcaa attccccaac cctgtcccaa atgasgggtg cggaactgag ctggcaatta 920 cagactattc tttggcagta ttctgggaaa gggaaaagta tgaattttct tgccctgtcc 480 ctggtatttt gtaaagcttt ccaagtgact tcgatatctg gacccttttg agaactgccg 540 ctgttaacca agacataaag ctttttcatt tcttgaaagg tcctatttct tggcaaattt 600 ctccctgcaa agtgcattct tttttctgca ctgaccccag aascaqgtgc cattgtgtgg 660 gagacagcac ctgtcatctg tctcaacaaa gtccacaaag tgccctgcta ggagcagata 720 acccagaaca tctacctgag tctcccagtc aagattcctg gtctgactag gaatctaact 780 tcaacaagat cttcctgaaa aaagacgctg agcttggacg gtccaaccac ctccctacct 840 gggataagcc ggaggatgct tcttggcttc cccaaagctg tcttggtgqt gatgctgtgg 900 caaccacagg ggagattcac gaggaqaaag cctggaagac cagagccctg gaagtggggc 960 agccagccca gcgggacatt cgtaggggcg aggtgtgatc tcagccttca ttcagctggc 1020 ctgggagtcc tccaagggtc agagtcacac cttactagtc tgacccctca aggttcctta 1080 ggactccagg tgttttasga agattctctg aacccccacc cttcccactt cacccctcag 1140 ccctccctgt ggttaagaac ccaagtgtga acaaccccaa gtttggtggt tagtgaggcc 1200 ttcaagcaca aaccctaatc ctqccctttt aaaaaaaata ttttggactt aagggaggct 1260 gtcctcaatt gtttcaaacc tatttcagag tacctccact ccttcaaacc tattttcaaa 1320 actcaggaaa acgtcgcctt tccagacqaa ctctagtgac aqctccattc accccttggc 1380 atttgactgc taaagatcag ctgacecctg cagccagaag gaggagagaa aattgctaaa 1940 tggggacctc atttccagag gtgatcacct taaaaaagtc tggggccttg tttgccagct 1500 ttgaaatctc aaagggtaat cttagcttca tgtccccagt gtgaataaaa caaacaataa 1560 aaaccaccta tgaaatctaa attcaaactt tcttqcgacc tatgcatcat ttttttgcat 1620 tcacattccc agtcaaaaca cagacacaca cacacacaca cccaacagat atataactgc 1680 atgtaaaata tatatagata aaacaatqtc ctctgaagaa tataaatgtt aacacaagta 1740 aaattatcat gaattccata actccttctg tggccttggt ccacgttagg ttttccatta 1800 agagatacta ttgtcatcag tgaqtagctt ctcctgttgt attcttctgt gtagcacggg 1860 gctggagctg tggatattta tgaaccccag gcacccactt cctgttttcg tggctgtccc 1920 cggtccctgc ctctgtgccg ttttctctgc ctttggccac cagatggcgc tctggaaacg 1980 agttttctgc atcaaagctc ccctgcatta ctctcaacta ggctccctcc ctcttctctg 2040 ttcccttcca agccattttc cagggtgcag ccaaaaggcc tttctggagg gcacatctgt 2100 gtgatcttgg gcaagttacc taacgctgtg tgctcagcat cttcaatgtg atggtatggc 2160 tgtgagtgcg tcatgagagg atgtaggtaa agcacttagg acaaggcctg gcatgtggtc 2220 gctgttctca ttatcctgcg tagagctttt caaaagcact ccacgacctt caggacaagg 2280 tctaaattct tagccatggg aacccaggcc actgcatcat ctgaggaccc ctgcttatgt 2390 cccctgccca accttcctcc ctacactact gcactgaccc tcttgcagca cagccaggag 2400 ccagcctgcc cacttcgggg cctttgccca ttttgcccct gccaccagct ttgcctgtct 2460 cctgcctatt caacttcctc gqgctcactc ctccaggtcc gtcaggattc agcgtgtgcc 2520 ctccatgaqg agcctcccgg acccccaggc ctgattccag agccattcta ctggcctctc 2580 gcctttcatt ttccctgccc ccacctgcga ctgtcagccc cttgccttgt tcactgtcag 2690 gcacacaggg ggacgattgc ggggcggata gtgagactgt cacgtgggct gccagcaggg 2700 ggcaatgggc acccttgtgt ccctgagctc ccaggggctc ctagtagctt aggcagcacc 2760 ttgcccacac caagctgagg gagacctgag aaatggqatt cctgccaaag aggcaaagaa 2820 aaaagtgaag ggacaaatgc aaggccatct ttgccatttg cgtatgaggc ataaaacttg 2880 gaagstasta aacagcaaaa tcatagtcct tatgtgctga gtgctttctc tgagccaggc 2940 cctgtgctga gtccttaccc caccttgtga gatcgatgac agccctcaga gaaggaaaca 3000 gaggctagga gaggttaagt cccttgtgca gtgtctcaga gcttgtgagt ggcctgtctg 3060 tctccagggc catgctgtag gaggtggccc tggattcagg ccccttgcta tggtttaatg 3120 ggtttcagtg aaatagttgg gaaacctaag gaqtgggata tggagttaag ggaaagaaaa 3180 gccagagaga aagggaagga gtgcaggtga ctgtccctta gccaccccag tcctgatgac 3240 cacaggcctc catgccaata agccctgact agtgccactt gggtccaaac atggcatctc 3300 tggccccaag ggcttagtag caaacaccca tctagggaag ctggcgttca ttctcatcac 3360 ctcaaatgct tcatgagcct cagggatcaa ccttgaagtg ggtataaggc tgggagaatg 3920 ttgggggcag caaactgaag ggcacaataa gaaqcaataa ggccacctca aagcccaccc 3480 aagcaaactg ctcattcacc tccttccttc ctgaatttca ccttatgagg aggtgagtgg 3590 aagatagggt atcccttaaa acatctaaaa ggagagttgg gggcagcaaa ggagatgtgc 3600 ttcacgggac tcttataaac aaaactggqg agagangaat tggaggaagg gggaaagaca 3660 tagatqaaag ggaggggaag gtgtgggaga gggaaacgta tasaagcttc caagaggagt 3720 gggaggctgg 9ggttcccca gacagagact cagtctggac cagatgcaga gaacaatgga 3780 cttcaaggct ggaggggggc agaagggaag cgggaggaga gccacacggt caagttgcac 3840 aggttcttgc agcttctgga atcaagacca tgggcaccct cataagtcag tgtgggcagg 3900 gactqcccca gggccaatcc aagatccaga ggtagccata gggtgtgaca agttgtgcag 3960 attacaacac tcaccccttq caataacgtc actgcctgtg actcggggcc aggcccaggc 4020 caaagccctt cctacatcat ttcgtttaat cctcacagtt tcctgctgaa agggctacta 9080 ttcttactcc catccccact ctacagatga ggtaatgqag gcccaggaaa gttaagtgac 4140 ttgtcccaga tgacaccgct ggtaagttgc aaagtcagaa tttgaactca gqcagtttac 9200 ctctgatggc tgctctgtta atcacagctg ctttccagtq agacaaaaac gggtgatcag 9260 ggcagagtca agacagagag gtaaacaaqa ttggqaaaaa gacaggaatg agaggggaac 9320 aatgggggaa aagataggaa caaagagagt tggggaaggg gagagaaaca qqaaacatga 9380 cttgcccgqq aggggcatca gtccacgtgc aagcaggtgg aggctcaagt tttctgctca 9990 cttggtgatg cagaggctcc ctttccctca gcagccgcct tgctgcgtgg acagcagctt 9500 cccatctggc ctgtccccgg agccccggcc tcatcctcct cagcggcagg ccacttagct 9560 tcacaggaaa tgctctttct ctaattqgca ttgasactca cagccctccc ttttcctgta 4620 ggtggggttt ccataggaaa aagctgcttc tctgtttccc cagcctagca actqtttggc 4680 agtcagagtc ccacatcctg ctcaactggg tcaggtccct cttagaccag ctcttqtcca 4790 tcatttgctg aagtggacca actagttccc cagtaggggg tctcccctgg caattcttga 4800 tcggcgtttg gacatctoag atcgcttcca atgaagatgg ccttgccttg gggtcctgct 9860 tgtttcataa tcatctaact atgggacaag gttgtgccgg cagctctqgg ggaaggagca 4920 cqgggctgat caagccatcc aggaaacact ggaggacttg tccagccttg aaagaactct 4980 aqtggtttct gaatctagcc cacttggcgg taagcatgat g 5021 <210> 2 <211> 36 <212> DNA
<213> Artificial sequence <220>
<223> Primersequence <220>
<221> mist feature <222> (17..(6) <223> n~g, c, a or t <Qac> 2 nnnnnnggta cccagaagta gtgcccttcc tctcga 36 <210> 3 <211> 36 <212> DNA
<213> Artificial sequence _7_ <220>
<223> Primer sequence <220>
<221> misc feature <222> (1)..(61 <223> n~g, c, a or t <900> 3 nnnnnnagta ctgcttaccg caacgtgggc tagatt 36 <210>4 <211>22 <212>DNA

<213>Mus musculus <900> 9 tacttctgcc attcctgcta gg 22 <210>5 <211>22 <212>DNA

<213>Mus musculus <400> 5 gacctttaga gtctctcacg ga 22 <210> 6 <211> 21 <212> DNA
<213> Homo sapiens _g_ <900> 6 gtaagcatga tgcaacttct g 21 <210> 7 <211> 21 <212> ONA
<213> Homo Sapiens <400> 7 aagggtttct gctcgtcctt g 21 <210> 8 <211> 22 <212> DNA
<213> Mus musculus C900> B
caccaccttc ttgatgtcat ca 22 <210>9 <z11>2z <212>DNA

<213>Mus musculus <900> 9 gccatcaatg accccttcat tg 22 <210>10 <z11>21 <212>DNA

<213>Mus musculus <900> 10 ggtagtggtc aaatggcgat t 21 <210> 11 <2I1> 21 <212> DNA
<213> Mus musculus <400> 11 gccaccaatc cccatatgga a 21 <210>12 <211>21 <212>DNA

<213>Mus musculus <400> 12 gtccggtcct catcagattc t 21 <210>13 <211>21 <212>DNA

<213>Mus musculus <900> 13 gtgtgctgcc ccaacaacat t 21 <210> 14 <211> a1 <212> DNA
<213> Escherichia coli - 1~ -<900> 14 ggtagtggtc aaatggcgat t 21 <210> 15 <2I1> 21 <212> DNA
<213> Escherichia coli <400> 15 gccaccaatc cccatatgga a 21

Claims

We claim:

1. A nucleic acid fragment, wherein the nucleic acid fragment contains a regulatory sequence of the human VE-Cad-2 gene extending 5' upstream from the translation start or a functionally active variant thereof, such that the nucleic acid fragment allows vascular endothelium-specific expression.

2. A nucleic acid fragment according to claim 1, wherein the nucleic acid fragment contains the sequence according to SEQ ID No. 1 or a functionally active variant thereof.

3. A nucleic acid fragment according to either of claims 1 and 2, wherein the nucleic acid fragment allows vascular-endothelium-specific expression excluding expression in the liver.

4. A nucleic acid fragment according to at least one of claims 1 to 3, wherein the nucleic acid fragment has a sequence homology of at least 80% to the sequence according to SEQ ID No. 1 or to a functional part thereof.

5. A nucleic acid fragment according to at least one of claims 1 to 3, wherein the nucleic acid fragment has a sequence homology of at least 90% to the sequence according to SEQ ID No. 1 or to a functional part thereof.

6. A nucleic acid fragment according to at least one of claims 1 to 3; wherein the nucleic acid fragment has a sequence homology of at least 95% to the sequence according to SEQ ID No. 1 or to a functional part thereof.

7. A nucleic acid fragment according to at least one of claims 1 to 6, wherein the nucleic acid fragment contains a functional part of the sequence according to SEQ ID No. 1.

8. A nucleic acid fragment according to claim 7, wherein a preferred functional part is 1-3804 by of the sequence according to SEQ ID No. 1.

9. A nucleic acid fragment according to claim 7, wherein a preferred functional part is 2724-3804 by of the sequence according to SEQ ID No. 1.

10. A nucleic acid construct containing a nucleic acid fragment according to at least one of claims 1-9 and a heterologous gene.

11. A nucleic acid construct according to claim 10, wherein the heterologous gene codes for a therapeutically active gene product.

12. A nucleic acid construct according to either of claims 10 and 11, wherein the therapeutically active gene product is heme oxygenase, MCP-1, GM-CSF, iNOS, eNOS or a Fas ligand.

13. A vector, comprising a nucleic acid fragment according to at least one of claims 1 to 9 or a nucleic acid construct according to at least one of claims 10 to 12.

14. A vector according to claim 13, wherein the vector is selected from the group consisting of plasmids, shuttle vectors, phagemids, cosmids, first and third generation of adenoviral vectors, expression vectors, and gene therapeutically active vectors.

15. A knock-out gene construct, wherein the knock-out gene construct comprises a nucleic acid fragment according to at least one of claims 1 to 9.

16. A cell which comprises the nucleic acid fragment according to at least one of claims 1 to 9, a nucleic acid construct according to at least one of claims 10 to 12, a vector according to either of claims 13 and 14 or a knock-out gene construct according to claim 15.

17. A cell according to claim 16, wherein the cell is selected from the group consisting of embryonic stem cells, embryonic germ cells and stem cells derived from adult tissue.

18. A cell according to claim 17, wherein the cell derived from adult tissue is selected from the group consisting of neuronal stem cells, bone marrow stem cells, mesenchymal stem cells, hematopoietic stem cells, epithelial stem cells, digestive tract stem cells and duct stem cells.

19. A cell according to at least one of claims 16 to 18, wherein the cell is a mammalian cell including a human cell.

20. A cell according to at least either of claims 17 and 18, wherein the cell is a transgenic non-human stem cell.

21. A transgenic non-human animal, wherein the animal comprises at least one cell according to claim 20.

22. A test for the identification of pharmacologically active substances which modulate the functioning of a nucleic acid fragment according to at least one of claims 1 to 9, wherein the test comprises at least one nucleic acid fragment according to at least one of claims 1 to 9, at least one vector according to either of claims 13 and 14, and/or at least one cell according to at least one of claims 16 to 20, if appropriate together or combined with suitable additives and/or auxiliaries.

23. The test according to claim 22, wherein the pharmacologically active substance effects the functioning of the nucleic acid fragment in an activating way.

24. The test according to claim 22, wherein the pharmacologically active substance effects the functioning of the nucleic acid fragment in an inhibitory way.

25. An array immobilized on a support material, wherein the array comprises at least one nucleic acid fragment according to at least one of claims 1 to 9 and/or at least one cell according to at least one of claims 16 to 20.

26. A diagnostic wherein the diagnostic comprises at least one nucleic acid fragment according to at least one of claims 1 to 9, at least one nucleic acid construct according to at least one of claims 10 to 12, at least one vector according to either of claims 13 and 14, and/or at least one cell according to at least one of claims 16 to 20, if appropriate together or combined with suitable additives or auxiliaries.

27. A diagnostic according to claim 26, wherein said nucleic acid fragment is a DNA probe.

28. A pharmaceutical, wherein the pharmaceutical comprises at least one nucleic acid fragment according to at least one of claims 1 to 9, at least one nucleic acid construct according to at least one of claims 10 to 12, at least one vector according to either of claims 13 and 14, and/or at least one cell according to at least one of claims 16 to 20, if appropriate together or combined with suitable additives or auxiliaries.

29. A method for the identification of a nucleic acid fragment according to at least one of claims 1 to 9 comprising the steps of:
(1) Combining at least a nucleic acid fragment comprising a regulatory sequence of the human VE-Cad-2 gene extending 5' upstream from the translation start or a variant thereof, with a reporter gene to form a reporter gene expression vector;
(2) Introducing the reporter gene expression vector into at least two different cells;
(3) Measuring the level of expression of the reporter gene;
(4) Comparing the levels of expression of the reporter gene of the different cells employed; and (5) Identifying nucleic acid fragments which allow vascular endothelium-specific expression, excluding expression in the liver.

30. The method for identification according to claim 29, wherein said reporter gene is selected from the group consisting of beta-galactosidase, luciferase, green fluorescent protein (GFP), red fluorescent protein, yellow fluorescent protein, or a His, Myc or Flag tag bound to a heterologous gene.

31. A method of producing a gene therapeutically active vector, pharmaceutical and/or diagnostic, wherein a nucleic acid fragment identified according to either of claims 29 and 30 is inserted into a vector which comprises at least one heterologous gene and which makes vascular endothelium-specific expression possible, excluding expression in the liver.

32. A method for selecting endothelial cells from stem cells comprising the following steps:
(1) combining at least one nucleic acid fragment according to at least one of claims 1 to 9 with a reporter gene to form a reporter gene expression vector;
(2) introducing the reporter gene expression vector into at least one stem cell;
(3) cultivating the stem cell(s);
(4) initiating the differentiation of the cultivated stem cell(s); and (5) isolating the endothelial cells) from the cultivated cell(s).

33. The selection method according to claim 32, wherein the reporter gene is an antibiotic resistance gene, and the endothelial cells) are isolated by collecting the differentiated endothelial cells) after addition of a suitable antibiotic in step (3) or (4).

34. The selection method according to claim 33, wherein the antibody resistance gene is selected from the group consisting of hygromycin resistance gene (hph), zeocin resistance gene (Sh ble), puromycin resistance gene (pacA), and gentamycin or G418 resistance gene (aph).

35. The selection method according to claim 32, wherein the reporter gene is selected from the group consisting of luciferase, green fluorescent protein (GFP), red fluorescent protein and yellow fluorescent protein, and the endothelial cell(s) is (are) isolated from the cultivated cell(s) by fluorescence-activated cell sorting (FACS).

36. The selection method according to claim 32, wherein the reporter gene is selected from the group consisting of luciferase, beta-galactosidase, green fluorescent protein (GFP), red fluorescent protein, yellow fluorescent protein and a His, Myc or Flag tag bound to a heterologous gene, and the endothelial cells) is (are) isolated from the cultivated cells) by affinity purification.

37. A method for producing a cell according to at least one claims 16 to 20, wherein the cell is transfected with a nucleic acid fragment according to at least one of claims 1 to 9, a nucleic acid construct according to at least one of claims 10 to 12, a vector according to either of claims 13 and 14 or a knock-out gene construct according to claim 15.

38. A method of producing a transgenic non-human animal according to claim 20, wherein a cell according to claim 20 is regenerated into a transgenic non-human animal.

39. A method of producing a test according to at least one of claims 22 to 24, wherein at least one nucleic acid fragment according to at least one of claims 1 to 19, at least one vector according to either of claims 13 and 14 and/or at least one cell according to at least one of claims 16 to 20, is combined with suitable additives or auxiliaries.

40. A method of producing an array according to claim 25, wherein at least one nucleic acid fragment according to at least one of claims 1 to 9, and/or at least one cell according to at least one of claims 16 to 20, is immobilized on a support material.

41. A method of producing a diagnostic according to either of claims 26 or 27, wherein at least one nucleic acid fragment according to at least one of claims 1 to 9, at least one nucleic acid construct according to at least one of claims 10 to 12, at least one vector according to either of claims 13 or 14, and/or at least one cell according to at least one of claims 16 to 20, is combined with suitable additives or auxiliaries.

42. A method of producing a pharmaceutical according to claim 28, wherein at least one nucleic acid fragment according to at least one of claims 1 to 9, at least one nucleic acid construct according to at least one of claims 10 to 12, at least one vector according to either of claims 13 or 14, and/or at least one cell according to at least one of claims 16 to 20, is combined with suitable additives or auxiliaries.

43. A method of treating a mammal or a human by administering to the mammal or human a pharmaceutically effective amount of a pharmaceutical according to claim 28.

44. The method according to claim 43, wherein said pharmaceutical is administered by means of a method selected from the group consisting of systemic injection, local injection, perfusion, and catheter-based administration.

45. The use of a nucleic acid fragment according to at least one of claims 1 to 9, a nucleic acid construct according to at least one of claims 10 to 12, a vector according to either of claims 13 or 14, a cell according to at least one of claims 16 to 20, or a transgenic non-human animal according to claim 21, for the expression of a heterologous gene.

46. The use of a test according to at least one of claims 22 to 24 for the identification of pharmacologically active substances which modulate the functioning of the nucleic acid fragment according to at least one of claims 1 to 9.

47. The use of an array immobilized on a support material according to claim for analysis in connection with disorders selected from vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory disorders and/or immunogenic disorders.

48. The use of a diagnostic according to either of claims 26 or 27 for the diagnosis of disorders selected from the group consisting of vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory disorders and/or immunogenic disorders.

49. The use of a pharmaceutical according to claim 28 for the prevention and/or treatment of disorders selected from the group consisting of vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory disorders and/or immunogenic disorders.

50. The use according to claim 49 wherein the pharmaceutical is used for somatic gene therapy.

51. The use of the gene therapeutically active vector, a pharmaceutical or a diagnostic produced according to claim 31 for the diagnosis, prevention and/or treatment of disorders selected from the group consisting of vascular disorders, genetic disorders, disorders associated with pathological vasodilatation or vasoconstriction, atherosclerosis, diabetes, cancerous disorders, inflammatory disorders and/or immunogenic disorders.

52. A method of producing an artificial tissue or organ comprising at least one endothelial cell, wherein an endothelial cell obtained by the method according to at least one of claims 32 to 36 is combined or cultivated with at least one suitable cell and/or support in order to generate the artificial tissue or organ.

53. The method according to claim 52, wherein the artificial tissue is selected from the group consisting of vessels, heart valves and venous valves.

54. A method for testing the pharmacological activity of a pharmacological substance, wherein an endothelial cell obtained by the method according to at least one of claims 32 to 36 is exposed to the pharmacological substance and the pharmacological activity of the pharmacological substance is determined.