DE29916160U1 - Modulation of gene transcription in vascular cells - Google Patents

Description

Cardiogene Gene Therapy Systems AG

C29861 BQ/ZW/cp/sa

Modulation the Transcription of genes in vascular cells

A double-stranded nucleic acid capable of binding sequence-specifically to the transcription factor AP-1, C/EBP or a related transcription factor.

Coronary heart disease (CHD) is the leading cause of death in industrialized nations. In recent decades, aortocoronary venous bypass graft surgery and percutaneous transluminal coronary angioplasty (PTCA) have been performed on a larger scale for the treatment of coronary heart disease, which have significantly increased the survival rate as well as the quality of life of cardiac patients.

Although these methods have been established for several decades, they still have a high probability of relapse or total occlusion of the treated vessels (30-50%) and this complication often appears just a few months after the operation. This is a significant medical problem and an economic burden for the health systems of the industrialized world.

One of the main reasons for complications observed in PTCA and coronary artery bypass surgery appears to be the induction of migration and growth of cells of the vessel wall. This inappropriate and undesirable growth of vascular tissue leads to a remodeling process that

·

• a · · · t I

&bull; <

ultimately causing relapse or closure of surgically treated vessels.

To prevent postoperative relapse or occlusion of the treated vessel segment, the strategy of implanting stents, tubular structures designed to maintain the postoperative shape and diameter of a vessel after PTCA, has recently been adopted. This method is also complicated by the development of restenosis within the stent, which is a result of excessive growth and migration of smooth muscle cells due to mechanical stress on the vessel wall. In addition, several clinical trials using a variety of adjuvant drug therapies have failed to successfully suppress the side effects of the surgical procedure.

The result of therapeutic failure is the (re-)occlusion of myocardial blood vessels, leading to ischemic episodes and, in severe cases, myocardial infarction (MI). This is accompanied by massive death of cardiomyocytes in areas of the heart muscle with insufficient oxygen supply, resulting in non-contractile scar tissue. Therefore, triggering vascular growth in ischemic areas of the heart muscle and/or inducing cardiomyocyte proliferation could represent a viable therapeutic approach for the treatment of myocardial infarction.

Most cells in the body are in the Go phase of the cell cycle, also known as the resting phase. Almost all resting cells in the body still have the capacity to grow and can be stimulated to re-enter the cell cycle by a number of stimuli, the most important being growth factors and injury. Growth as well as remodeling processes are primarily regulated at the transcriptional level.

-3-

The physical stress of, for example, coronary angioplasty or stent implantation will therefore lead to the induction of a number of genes, the most important being cyclins, cell cycle-specific phosphatases, cell cycle-specific transcription factors such as cyclin E, cyclin A, cyclin B, cdc25C, cdc25A, E2F family members as well as a number of metabolically important genes or genes involved in DNA duplication such as PCNA, histones, dhfr . While these factors are newly synthesized after entry into the cell cycle, many transcription factors involved in the first steps of growth, the so-called "immediate-early" genes, are already present in the cell and are activated by a specific signal; a well-known member of this class is AP-I.

AP-I is a heterodimeric transcription factor consisting of c-Jun and c-Fos proteins (Curran and Franza (1988) Cell 55, 395), which interact with each other via leucine zipper motifs. Unlike c-Jun, which is able to homodimerize and bind DNA alone, c-Fos relies on binding to c-Jun to bind DNA in a sequence-specific manner. Both proteins are members of a larger family of proteins that includes JunB and JunD (Jun-related) and Fral and FosB (Fos-related) (Curran and Vogt

(1992) in Transcriptional Regulation, 797 (McKnight and Yamamoto) Cold Spring Harbor Laboratory Press). AP-I is able to bind to the consensus sequence, TGACTCA motif, but many variations of this sequence can be strongly bound by the various homo- and heterodimers of the family members (Franza et al. (1988) Science 239, 1150; Rauscher et al.

(1988) Genes Dev. 2, 1687; Risse et al. (1989) EMBO J. 8, 3825; Yang-Yen et al. (1990) New Biol. 2,351).

Although the simultaneous expression of Fos and Jun can lead to dramatic synergistic activation of AP-I-dependent transcription (Chiu et al. (1988) Cell 54, 541), a significant degree of experimental variability was observed depending on the cell type used. Therefore, it is likely that

the activities of Fos and Jun are influenced by other proteins that may be expressed in a cell type-specific manner (Baichwall and Tjian (1990) Cell 63, 815) and that in each cell type the presence of pre-existing or inducible transcription factors influences both the selection of gene targets and the transcriptional effect of Fos-Jun family hetero- or homodimers. Consistent with the promoter and cell type specificity of AP-I, it has been shown that Fos does not activate the c-fos promoter but represses transcription (Sassone-Corsi (1988) Cell 54, 553; Lucibello et al. (1989) Cell 59, 999). Consequently, it is not possible to predict what effect AP-I will have on a specific promoter in a given cell type.

A method for specifically influencing the transcriptional activity of a particular transcription factor is disclosed in WO 95/11687. This teaches that it is possible to interfere with the activation function of transcription factors by treating a cell with a double-stranded DNA molecule, called a cis-element decoy, containing a binding site for the specific transcription factor. The exogenous delivery of a large number of transcription factor binding sites to a cell, in particular in much higher numbers than those present in the endogenous promoter in the genome, creates a situation in which the majority of a particular transcription factor binds specifically to the respective cis-element decoy and not to its endogenous target genes. This approach to inhibiting the binding of transcription factors to their endogenous binding site is also referred to as squelching. Squelching of transcription using DNA decoys has been successfully used to inhibit cell growth using DNA fragments specifically targeted to the transcription factor E2F (Morishita et al. (1995) 92, 5855).

Based on the extrapolation of the current results of the human genome project, it is speculated that about 50% of all human genes are transcription

· · i
·
· ·· ·
· ·
·
&bull;
&bull; ·
·
·· · ·
·
·· · &bull; ·
&bull;
·· ·
· &bull; · · ·
» · · ·
» ··· · ·
I · · · ·
&igr; · · · * · ·
· ··
· ·
· I
· I ·
·· · &igr; · · · &bull;
·· · · &bull;
· · »· ·· &bull; ■ · ·· ·· · · ·

factors. It is known that the complexity and diversity of organisms is due in large part to the restriction of the expression of certain transcription factors to specific cell types and/or to specific developmental stages. A transcription factor that has a positive effect on growth in one cell type may have an opposite effect in another cell type or in another organism. Therefore, the success of treatment with a given transcription factor decoy may vary not only between species but also between different tissues, for example veins and arteries, and between cell types in a given tissue, for example endothelial cells and smooth muscle cells of a cell wall within an organism.

Therefore, one of the objects of the present invention is to target transcription factors that modulate transcription in vascular cells, including endothelial and smooth muscle cells, or cardiac cells and which are therefore suitable targets for squelching with double-stranded nucleic acids having a sequence specific for binding a transcription factor. Preferred transcription factor targets are transcription factors that are directly or indirectly involved in inducing growth and/or remodeling of vascular and/or cardiac muscle tissue.

Surprisingly, it was found that the activation of the endothelin-1 (ET-I) gene in venous endothelial cells is mediated by AP-I. This activation was triggered by excessive mechanical stress. The ET-I peptide produced in response to this stimulus is highly effective in vasoconstriction and growth stimulation. Release of ET-I has been correlated with excessive smooth muscle cell growth and remodeling. Subsequently, it was possible to show that the activation of the ET-I gene in endothelial cells could be blocked by the introduction of double-stranded DNA carrying the AP-I binding site.

&bull; a * ·

i*

Accordingly, a method of the present invention is the modulation of the transcription of at least one gene in endothelial or cardiac cells, the method comprising the step of contacting said cell with a mixture containing one or more double-stranded nucleic acids capable of binding sequence-specifically to the transcription factor AP-I or a related transcription factor. This method is particularly applied to endothelial cells which are part of a vessel or a vascular graft. Typically, the method is applied to a vessel in vivo or to a vascular graft in vivo before or after implantation or ex vivo before implantation. Preferably, the cardiac cell is a cardiac muscle cell or a fibroblast.

Another unexpected result of the present invention was the discovery that CCAAT/Enhancer-Binding Protein (C/EBP) is involved in the activation of ET-I in carotid aortic smooth muscle cells following the application of mechanical stress. In agreement with the above-described differential effects of transcription factors in different tissues, AP-I was not involved in the activation of ET-I genes in this cell type. When C/EBP-specific double-stranded nucleic acids were introduced into smooth muscle cells, the activation of the ET-I gene could be blocked.

C/EBPs comprise a family of transcription factors that are critical for normal cellular differentiation and function in a variety of tissues (reviewed in Lekstrom-Himes and Xanthopoulos (1996) J. Biol. Chem. 44, 28645). There are at least six members of the C/EBP family (Cao et al. (1991) Genes Dev. 5, 1538) that bind to DNA only after dimerization. C/EBP family members have been studied specifically for their role in the liver and are known to be involved in the acute phase response following liver injury (reviewed in Fey et al. (1990) in Progress in Liver Diseases

* i

·

-7-

(Popper and Schaffner) Vol. 9, 89, W.B. Saunders Co., Philadelphia). C/EBP family members have been shown to play a major role in inflammatory processes, and C/EBP binding motifs have been recognized in promoters of inflammatory cytokines such as interleukin-6 (IL-6), IL-Iß and tumor necrosis factor α (TNFa), as well as other cytokines such as IL-8 and IL-12 (Matsusaka et al. 1993) Proc. Natl. Acad. Sci. 90, 10193; Plevy et al. (1997) Mol. Cell. Biol. 17, 4572). The involvement of C/EBP in the transcriptional activation of ET-I, a factor known to promote growth and remodeling of vascular tissue, suggests another suitable transcription factor target in vascular cells.

Accordingly, another method of the present invention is influencing the transcription of one or more genes in vascular or cardiac cells, the method comprising the step of contacting said cell with a composition containing one or more double-stranded nucleic acids capable of binding sequence-specifically to the transcription factor C/EBP or a related factor.

It is known that individual transcription factors can often activate a promoter to some extent alone, but that the complete activation of this promoter depends on synergistic activation by two or more transcription factors (Sauer et al. (1995) Science 270, 1783). Consequently, in order to achieve a stronger modulation of a promoter than with a double-stranded nucleic acid directed against a single transcription factor, it is also intended that the method of the present invention be carried out with a mixture containing one or more sequences specifically binding C/EBP and, in addition, one or more double-stranded nucleic acids capable of sequence-specifically binding the transcription factor AP-1. This combination will, for example, result in a stronger modulation of the transcription of promoters containing both AP-I and C/EBP-

contain binding sites, allow, or make it possible to influence two or more promoters independently at the same time.

When the method of the present invention is applied to vascular cells, the preferred targets are smooth muscle cells (SMC) or endothelial cells, '· while for cardiac cells, the preferred targets are cardiac muscle cells or fibroblasts. These cells are treated, for example, directly in the animal or human, in the vessel in situ , or they may be part of an autologous or heterologous vascular graft. These grafts, for example arterial or venous bypass grafts, may be treated prior to implantation by ex vivo application of the method of the present invention or after implantation by in vivo application of the method. In a preferred embodiment, the vessel treated is a coronary or peripheral artery or an arterio-venous fistula (e.g. Brescia-Cimino fistula) and the vascular graft is an arterial or venous bypass graft or a biological or prosthetic arterio-venous shunt.

For example, in order to prevent vascular occlusion induced by a surgical procedure or by a pathological process, it is intended to specifically modulate the expression of genes involved in the growth or transformation of vascular cells. This modulation may be a direct effect on the promoter of said gene by squelching the binding of AP-I, C/EBP or a related protein, or it may be an indirect effect by interfering with the regulation of a transcription factor which itself modulates a gene involved in proliferation or migration of vascular cells. Genes involved in proliferation or migration include genes encoding proteins that are active inside a cell, such as cyclins, cyclin-dependent kinases, inhibitors of apoptosis and genes encoding proteins that are secreted and therefore active outside the cell, such as metalloprotinases, collagenases or growth factors. In particular, growth factors that are secreted by

-9-

secreted by a vascular cell can stimulate the growth and migration of a number of other vascular cells. Therefore, the modulation of genes that exert effects outside the target cell is of particular importance for the present invention.

For example, the sequence of a nucleic acid used to prevent the binding of the transcription factor AP-I, C/EBP or a related transcription factor is a consensus sequence derived from the sequence of AP-I or C/EBP binding sites in many promoters, such as 5'CGCTTGATGACTCAGCCGGAA-S' (for AP-I) or 5'-TGCAGATT GCGCAATTG- 3' (for C/EBP), or it may be identical to the AP-I or C/EBP binding site in the particular promoter or enhancer whose transcriptional regulation is targeted. As already mentioned above, AP-I and C/EBP are members of a family of closely related transcription factors that are able to hetero- or homodimerize (Roman et al. (1990) Genes Dev. 4, 1404; Cao et al. (1991) Genes Dev. 5, 1538). There are a large number of different heterodimers that probably recognize different bona fide AP-I or C/EBP binding sites with different affinities. Therefore, the slightly different AP-I or C/EBP binding times present in different promoters may be preferentially bound only by certain hetero- or homodimers of AP-I or C/EBP or related transcription factors (Akira et al. (1990) EMBO J. 9, 1897, Risse et al. (1989) supra). Therefore, the use of promoter-specifically designed nucleic acids may allow targeted squelching of transcription factors involved in the regulation of that particular promoter. If a promoter contains more than one version of an AP-I and/or C/EBP binding site, as is often the case, nucleic acids representing the sequences of all of these binding sites can be used simultaneously. Similarly, one or more double-stranded nucleic acids containing slightly different versions of the consensus transcription factor binding site can be used simultaneously.

·

·

-10-

The affinity of binding of a nucleic acid sequence to AP-I, C/EBP or a related transcription factor can be determined by using the Electrophoretic Mobility Shift Assay (EMSA) (Sambrook et al. (1989) Molecular Cloning. Cold Spring Harbor Laboratory Press; Krzesz et al. (1999) FEBS Lett. 453, 191). This test system is suitable for quality control of nucleic acids intended for use in the method of the present invention or determinations of the optimal length of a binding site. It is also suitable for the identification of other sequences bound by AP-I, C/EBP or related transcription factors. For an EMSA designed to isolate new binding sites, purified or recombinantly expressed versions of AP-I, C/EBP or related transcription factors are best used in several alternating rounds of PCR amplification and selection by EMSA (Thiesen and Bach (1990) Nucleic Acids Res. 18, 3203).

As already described above, genes containing binding sites for AP-I and/or C/EBP or related transcription factors are suitable targets for the method of the present invention, as are genes that are indirectly affected. Genes encoding proteins involved in the growth and/or remodeling of vascular tissue as well as in other pathological processes are the endothelin gene family, the macrophage chemotactic protein (MCP) gene family and the nitric oxide synthase (NOS) gene family. Examples of genes with particular relevance to vascular cells are the prepro-endothelin-1 gene, the MCP-1 gene and the inducible NOS gene.

Genes known to contain AP-I binding sites in their promoter or enhancer regions and which are therefore putative targets for specific squelching by the method of the present invention include the prepro-endothelin-1 gene, the endothelin receptor B gene, the

- 11 -

inducible NOS gene (iNOS), the E-selectin gene, the monocyte chemotactic protein-1 gene (MCP-I), the intercellular adhesion molecule-1 (ICAM-I) and the interleukin-8 (IL-8) gene.

Genes known to contain C/EBP binding sites in their promoter or enhancer region and are therefore putative targets for specific squelching using the method of the present invention include the prepro-endothelin-1 gene, the endothelin receptor B gene, the iNOS gene, the E-selectin gene, the intercellular adhesion molecule-1 gene (ICAM-1), the IL-8 gene and the IL-6 gene.

The method of the present invention modulates the transcription of a gene or genes in such a way that the gene or genes are activated. Activation in the context of the present invention means that the transcription rate is increased compared to cells which have not been treated with a composition containing a double-stranded nucleic acid. Such an increase can be determined, for example, by Northern blot (Sambrook et al. (1989) supra) or RT-PCR (Sambrook et al. (1989) supra). Typically, such an increase is at least a 2-fold, more preferably at least a 5-fold, at least a 20-fold and most preferably a 100-fold increase. Activation can be achieved, for example, when AP-I and/or C/EBP or a related transcription factor acts as a transcriptional repressor at a particular gene and therefore squelching of the repressor results in a loss of repression. Loss of repression does not necessarily equate with activation, but it is known for several promoters that loss of repressor binding can be sufficient for activation (Zwicker et al. (1995) EMBO J. 14, 4514).

The method of the present invention also modulates the transcription of a gene or genes in such a way that the gene or genes activate the

·

- 12-

activator or are repressed. Repression in the context of the present invention means that the transcription rate is reduced compared to cells which have not been treated with a mixture containing a double-stranded nucleic acid. Such an increase can be determined, for example, by Northern blot (Sambrook et al. (1989) supra) or RT-PCR (Sambrook et al. (1989) supra). Typically, such a reduction is at least a 2-fold, more preferably at least a 5-fold, at least a 20-fold and most preferably at least a 100-fold reduction. The loss of activation or repression can be achieved, for example, when AP-I and/or C/EBP or a related transcription factor act as a transcription activator at a particular gene and therefore squelching of the activator leads to the loss of activation or repression.

In a further embodiment of the present invention, the modulation results in activation of one gene or genes at the same time as another gene or genes lose activation or are repressed. The different effects on individual genes can be easily followed, for example by Northern blot (Sambrook et al. (1989) supra) or RT-PCR (Sambrook et al. (1989) supra) or also with DNA chip array technology (US 5,837,466).

The double-stranded nucleic acid used in the present invention contains, in a preferred embodiment, one or more copies of a sequence that specifically binds AP-I and/or C/EBP or a related transcription factor. Synthetic nucleic acids are typically at most 100 bp long and can therefore recognize one or up to 10 complete transcription factor binding sites, depending on the length of the specific transcription factor recognition site selected. Nucleic acids amplified by enzymatic methods such as PCR or propagated in a suitable prokaryotic or eukaryotic host organism contain at least about 10 copies, preferably at least about 30

copies, at least about 100 copies or at least about 300 copies of the respective transcription factor binding site.

The double-stranded nucleic acid used in the method of the present invention is, for example, a vector or an oligonucleotide. In a preferred embodiment, the vector is a plasmid vector and in particular a plasmid vector capable of autosomally replicating (Wohlgemuth et al. (1996) Gene Ther. 6, 503-12), thereby increasing the stability of the introduced double-stranded nucleic acid. The length of the double-stranded oligonucleotide is typically chosen between about 10-100 bp, preferably between about 20-60 bp and most preferably between 30-40 bp.

Oligonucleotides are usually rapidly degraded by endo- and exonucleases, in particular DNases and RNases in the cell. Therefore, the nucleic acid is modified to stabilize it against degradation so that a high concentration of the nucleic acid is maintained in the cell over a long period of time. Typically, such stabilization can be obtained by the introduction of one or more modified internucleotide phosphorus groups or by the introduction of one or more non-phosphorus internucleotides.

A successfully stabilized nucleic acid does not necessarily contain such an internucleotide at every internucleotide position. It is intended that various modifications will be introduced into the nucleic acid and that the resulting double-stranded nucleic acids will be tested for sequence-specific binding to AP-I, C/EBP or related proteins using the routine EMSA test system. This test system allows the determination of the binding constant of the nucleic acid and thus the determination of whether the affinity has been changed by the modification. Modified nucleic acids that still

exhibit sufficient binding, wherein sufficient binding means at least about 50% or at least about 75%, and most preferably about 100% of the binding of the unmodified nucleic acid.

A nucleic acid which still shows sufficient binding will be tested to see if it is more stable in the cell than the unmodified nucleic acid. The in vitro test system employed uses the method of the present invention to introduce the modified nucleic acid into a cell. Subsequently, transfected cells are removed at various times and can be assayed for the amount of nucleic acid remaining by Northern blot (Sambrook et al. (1989) supra), Southern blot (Sambrook et al. (1989) supra), PCR (Sambrook et al. (1989) supra), RT-PCR (Sambrook et al. (1989) supra) or DNA chip array (US 5,837,466) techniques. A successfully modified nucleic acid has a half-life in the cell of at least about 48 hours, more preferably at least about 4 days, most preferably at least about 14 days.

Suitable modified internucleotides are summarized in Uhlmann and Peyman ((1990) Chem. Rev. 90, 544). Modified internucleotide phosphate residues and/or non-phosphorus bridges in a nucleic acid that can be used in a method of the present invention include, for example, methylphosphonate, phosphorothioate, phosphorodithioate, phosphoramidate, phosphate ester, while non-phosphorus internucleotide analogs include, for example, siloxane bridges, carbonate bridges, carboxymethyl ester bridges, acetamidate bridges and/or thioether bridges. It is also intended that this modification improves the stability of a mixture intended for use in a method of the present invention.

A further embodiment of the invention is the stabilization of nucleic acids by introducing structural features into the nucleic acid that

Increase the half-life of the nucleic acid. Such structures containing hairpin and bell DNA are disclosed in US 5,683,985. The simultaneous introduction of modified internucleotide phosphate residues and/or non-phosphorus bridges, together with the mentioned structures, is also intended. The resulting nucleic acids can be tested for binding and stability in the test system described above.

The mixture containing double-stranded nucleic acid is contacted with vascular cells or cardiac cells. The aim of this contacting is the transfer of the double-stranded nucleic acid which binds AP-I and/or C/EBP or a related transcription factor into the cell and in particular into the nucleus of the cell. Therefore, it is intended to use nucleic acid modification and/or additives or excipients known to increase membrane permeability within the scope of the present invention (Uhlmann and Peyman (1990) Chem. Rev. 90, 544).

A mixture according to the invention contains in a preferred embodiment essentially only nucleic acid and buffer. This method uses high concentrations of the double-stranded nucleic acid to enable efficient uptake of the nucleic acid into the cell and ultimately high concentration of nucleic acids in the nucleus. A suitable concentration of the nucleic acids is at least 1 μμ, more preferably 10 μμ and most preferably 50 μμ, with one or more suitable buffers added. An example of such a buffer is Tyrode solution containing 144.3 mmol/l Na ⁺ , 4.0 mmol/l K ⁺ , 138.6 mmol/l Cl", 1.7 mmol/l Ca ²⁺ , 1.0 mmol/l Mg ²⁺ , 0.4 mmol/l HPO ₄ ² ', 19.9 mmol/l HCO ₃ ', 10.0 mmol/l D-glucose.

In a further embodiment of the invention, the mixture additionally contains at least one additive and/or excipient. Additives and/or excipients such as lipid, cationic lipids, polymers, nucleic acid aptamers, peptides and

Proteins are intended, for example, to increase the delivery of nucleic acids into the cell, to target the mixture to only a subset of cells, to prevent degradation of the nucleic acid in the cell, to facilitate storage of the nucleic acid mixture before use, and/or to enhance introduction into the nucleus of the cell.

Excipients that increase the transfer of nucleic acids into the cell can be, for example, proteins or peptides bound to DNA or synthetic peptide-DNA molecules that enable 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 Biotechnology 17, 784). Excipients also include molecules that enable the release of nucleic acids into the cytoplasm of the cell (Planck et al. (1994) J. Biol. Chem. 269, 12918; Kichler et al. (1997) Bioconjug. Chem. 8, 213) or, for example, liposomes (Uhlmann and Peyman (1990) supra).

In order to specifically target the mixture to a subset of cells, the auxiliary substance can be selected so that it recognizes a protein of the target cell, preferably a protein or protein domain displayed on the surface of the cell. Specific targeting can also be described as specific recognition of cells. There are at least two classes of suitable auxiliary substances to achieve this recognition. One class of auxiliary substances contains antibodies or antibody fragments that preferentially recognize proteins or protein domains located on the extracellular side of the cell membrane. Antibodies directed against such membrane structures of endothelial cells are described, for example, in Burrows et al. ((1994) Pharmac. Ther. 64, 155), Hughes et al. ((1989) Cancer Res. 49, 6214) and Maruyama et al. ((1990) Proc. Nat. Aca. Sci. 87, 5744). Antibodies intended for use are directed against membrane structures of smooth muscle cells, such as:

·

- the antibody 10F3 (Printseva et al. (1987) Exp. Cell Res. 169, 85)

- Antibodies against actin (Desmoliere et al. (1988) Comptes Reudus des Seances de la Soc. de Biol et de ses Branches 182, 391)

- Antibodies against angiotensin II receptors (Butcher et al. (1993) BBRA 196, 1280)

- Antibodies against receptors for growth factors (review in Mendelsohn (1988) Prog. All. 45, 147; Sato et al. (1989) J. Nat. Cane. Inst. 81, 1600; Hynesetal. (1994) BBA 1198, 165)

or antibodies directed against, for example,

- EGF receptor (Fan et al. (1993) Cancer Res. 53, 4322; Bender et al. (1992) Cancer Res. 52, 121; Aboud-Pirak et al. (1988) J. Nat. Cancer Inst. 80 , 1605)

- PDGF receptor (Yu et al. (1994) J. Biol. Chem. 269, 10668; Kelly et al. (1991)266,8987)

- FGF receptor (Vanhalteswaran et al. (1991) J. Cell Biol. 115, 418; Zhan et al. (1994) J. Biol. Chem. 269, 20221)

The other class of auxiliary substances are ligands that bind to cell surface receptors with high affinity. Such ligands can be the natural ligands of the mentioned receptors or modified ligands with an even higher affinity. Typically such a ligand is a peptide known to bind to a given membrane receptor. In addition, it is possible to isolate new peptide ligands by using the receptor of interest to screen a combinatorial peptide library. (Lu, Z. et al. (1995) Biotechnol. 13, 366; U.S. 5,635,182; Koivunen, E. et al. (1999) J. Nucl. Med. 40, 883).

Membrane receptors on endothelial cells to which the ligands mentioned can be directed include, for example, PDGF, bFGF, VEGF and TGFp (Pusztain et

-18-

al. (1993) J. Pathol. 169, 191). In addition, the ligand also comprises adhesion molecules that bind proliferating/migrating endothelial cells. Adhesion molecules of this type such as SLex, LFA-I, MAC-I, LECAM-I or VLA-4 have already been described (reviews in Augstein-Voss et al. (1992) J. Cell Biol. 119, 483; Pauli et al. (1990) Cancer Metast. Rev. 9, 175; Honn et al. (1992) . Cancer Metast. Rev. 11, 353). But the ligand is not limited to peptide ligands, it can also be a nucleic acid aptamer that specifically recognizes the cell surface of a target cell. (Hicke, BJ. et al. (1996) J. Clin. Invest. 98, 2688).

Membrane receptors expressed in smooth muscle cells that are targeted by the mentioned ligands include, for example, PDGF, EGF, TGFa, FGF, endothelin A and TGFβ (reviewed in Pusztain et al. (1993) J. Pathol. 169, 191; Harris (1991) Current Opin. Biotechnol. 2, 260).

Additives that stabilize the nucleic acid in the cell include nucleic acid condensing substances such as cationic polymers, poly-L-lysine or polyethyleneimine.

The mixture used in the method of the present invention is preferably applied locally by injection, catheters, trocars, projectiles, pluronic gels, sustained drug release polymers, coated stents, or any other device that allows local access. Ex vivo application of the mixture used in the method of the present invention also allows local access.

Another embodiment of the present invention is a method of treating vascular disease, preferably vascular proliferative disease, the method comprising the step of contacting vascular cells with a mixture as described above in an amount

sufficient to modulate transcription of one or more genes. Diseases intended to be treated by this method include, for example, restenosis, intimal hyperplasia of arterial or venous bypass grafts, graft vasculopathy and hypertrophy. It is also intended that this method be used as an adjuvant therapy for treatments commonly used in vascular diseases such as PTCA, PTA, bypass surgery or use of arteriovenous shunts. The scope of the present invention also includes diseases known to be the result of vascular disease such as coronary heart disease and myocardial infarction.

It is also intended to use the method of the present invention as an in vitro method. For example, an in vitro screening method allowing the identification of genes modulated in vascular cells by AP-I and/or C/EBP or related proteins. Such a method would comprise in vitro contacting said cells with a mixture containing one or more double-stranded nucleic acids capable of sequence-specifically binding transcription factor AP-I and/or C/EBP or related proteins. The transcription level of thousands of genes could then be monitored simultaneously using a DNA chip array, for example as described in US 5,837,466, and compared with and without the mixture added.

Another object of the present invention is a double-stranded nucleic acid which is capable of binding sequence-specifically to the transcription factor AP-I or a related transcription factor which comprises one of the sequences of SEQ ID 5 or 9 to 18 with the sequences: 5' CGCTTG ATGAC TCAGC CGGAA 3' (consensus AP-I); 5' GTGCT GACTC AGCAC 3' (constructed AP-I); 5' CGCTT AGTGA CTAAG CG 3' (constructed AP-I); 5' TGTGC TGACT CAGCA CA 3' (constructed AP-I); 5' TTGTG CTGAC TCAGC ACAA 3' (constructed AP-I); 5' TCGCT TAGTG ACTAA GCGA 3'

-20-

(engineered AP-I); 5' TGCTG ACTCA TGAGT CAGCA 3' (engineered AP-1); 5' TGCTG ACTAA TTAGT CAGCA 3'(engineered AP-I); 5' GTCGC TTA GT GACTA AGCGAC 3' (engineered AP-I); 5' CTTGT GCTGA CTCAG CACAA G 3' (engineered AP-I); 5' TTGCT GACTC ATGAG TCAGC AA 3' (engineered AP-I).

Another aspect of the present invention is a double-stranded nucleic acid capable of binding sequence specifically to the transcription factor C/EBP or a related transcription factor comprising one of the sequences of SEQ ID 6 or 19 to 26 with the sequences: 5' TGCAG ATTGC GCAAT CTGCA 3' (consensus C/EBP); 5' GACAT TGCGC AATGT C 3' (engineered C/EBP); 5' AGCAT TGGCC AATGC T 3' (engineered C/EBP); 5' CGACA TTGCG CAATG TCG 3' (engineered C/EBP); 5' AGGCA TTGGC CAATG CCT 3' (engineered C/EBP); 5' ACGAC ATTGC GCAAT GTCGT 3' (engineered C/EBP); 5' TAGGC ATTGG CCAAT GCCTA 3' (constructed C/EBP); 5' CTGTT GCGCA ATTGC GCAAC AG 3' (constructed C/EBP); 5' GACTT GCGCA ATTGC GCAAG TC 3' (constructed C/EBP).

Double-stranded nucleic acids of the present invention are between 12 and 22 nucleotides long. The optimal length of a given nucleic acid is selected to optimize binding to the transcription factor and uptake into the cell. Typically, a double-stranded nucleic acid shorter than 12 bp binds only weakly to its target protein, while one longer than 22 bp, although binding strongly, is taken into the cell only with a low efficiency. Binding strength can be determined by EMSA, while uptake of the double-stranded nucleic acid can be analyzed by Northern blot (Sambrook et al. (1989) supra), Southern blot (Sambrook et al. (1989) supra), PCR (Sambrook et al. (1989) supra), RT-PCR (Sambrook et al. (1989) supra) or DNA chip array technologies (US 5,837, 466). A nucleic acid of the present invention can be stabilized as described above.

&bull; *

i * i si

-21 -

A preferred embodiment of the present invention are nucleic acids containing a palindromic binding site and thus comprising at least two transcription factor binding sites in a short double-stranded nucleic acid. A double-stranded nucleic acid constructed in this manner has a higher statistical probability of binding to AP-I and/or CfEBP or a related factor. To enable the construction of a palindromic binding sequence, a mismatch in one side of the palindromic core sequence (AP-I: TGAC and C/EBP: GCAA) can be tolerated. Preferably, the core sequence is located in the middle of the nucleic acid to allow optimal binding to AP-I and/or C/EBP or a related transcription factor.

A nucleic acid of the present invention is rapidly taken up into the cell. Sufficient uptake is characterized by the modulation of one or more genes that can be modulated by AP-I and/or C/EBP or a related transcription factor. The double-stranded nucleic acid of the present invention preferably modulates the transcription of a gene or genes after about 2 hours of contact with the cell, more preferably after about 1 hour, after about 30 min, after about 10 min, and most preferably after about 2 min. A typical mixture used in such an assay contains 10 μg/ml of double-stranded nucleic acid.

The following figures and examples are for illustration purposes only and in no way limit the scope of the invention.

Fig. I: Effect of (A) endothelial removal, (B,E) Ro 31-8220 (0.1 μg/L), (C ₅ F) herbimycin A (0.1 μg/L) and (D) actinomycin D (1 μg/L) on the abundance of ppET-1 mRNA in isolated segments of rabbit carotid arteries (RbCA) or rabbit jugular vein (RbW),

-22-

perfused for 6 hours with different levels of intraluminal pressure. RT-PCR analysis, comparable results were obtained in at least four other experiments with segments from other animals.

' Fig. 2: Time-dependent increase in nuclear shift of (A) AP-I and (B) C/EBP in endothelium-intact segments of RbJV perfused at 20 mmHg. Representative EMSA, comparable results were obtained in at least two other experiments.

Fig. 3: Effect of specific decoy oligodeoxynucleotides (dODN) against (B ₁ D) C/EBPmut, (C) GATA-2, and (D) C/EBP on the abundance of ppET-1 mRNA in endothelium-intact segments of RbCA or RbJV perfused for 6 hours at different intraluminal pressures. (E,F) Effect of (E) Ro 31-8220 (0.1 μαλ/Γ) and

(F) Herbimycin A (0.9 μg/L) on the abundance of ppET-1 mRNA in PAEC (primary culture) incubated under static conditions or stretched for 6 hours. Representative RT-PCR analysis, comparable results were obtained for at least two additional experiments using segments or cells from other animals.

Fig. 4: Effects of specific decoy ODN against AP-I and C/EBP alone or in combination on (A,C) the abundance of ppET-1 mRNA (n=3-5) and (B) intravascular ET-I peptide (n=3) in isolated segments of RbCA (C) or RbJV (A,B) perfused for 6 hours with different levels of intraluminal pressure. *P<0.05 vs. 0 or 2 mmHg, ^# P<0.05 vs. 20 or 160 mmHg, ^f P<0.05 vs. 20 mmHg with C/EBP dODN.

Fig. 5: (A,B) Expression of different rat ppET-1 promoter-luciferase constructs (the asterisk indicates the -168 bp construct without the AP-I response element) in cultured PAEC incubated under static conditions or stretched for 6 hours in (A) the absence or (B) presence of different dODN (indicated by the plus symbol; n=3-7, duplicate determinations). (C) ß-galactosidase activity in cultured PAEC cotransfected with an SV40/ß-galactosidase expression vector and different rat ppET-1 promoter-luciferase constructs as a measure of similar transfection efficiency (n=3 - 7, duplicate determinations).

Fig. 6: Scheme of the possible transcription factor binding sites in the promoter of the rat ppET-1 gene, which also indicates the length of the different promoter constructs.

EXAMPLES

1. In situ model

Male New Zealand White rabbits (2.1 ±0.1 kg body weight, η = 47) were anesthetized with pentobarbitone (60 mg/kg iV; Sigma-Aldrich, Deisenhofen, Germany) and exsanguinated. The left and right common carotid aorta (RbCA) as well as the external jugular vein (RbJV) were dissected, cleaned of attached adipose and connective tissue, and cut in half. The segments were mounted in a specifically designed four-position perfusion chamber, where they were re-stretched to their in situ length (20.6 ± 0.3 mm) using a flexible cannula. Their diameter was continuously monitored by video microscopy (Visitron Instruments, Munich, Germany). The lumen of the segments and the surrounding tissue bath were individually perfused (lumen: 1 ml/min.; bath 0.5 ml/min.) with warmed (37°C) oxygen-enriched (lumen: 75% N ₂ , 20% O ₂ , 5% CO ₂ , PO; = 140 mmHg, PCO ₂ =15-20 mmHg, pH 7.4; bath: 95% CO ₂ , 5% O ₂ , PO ₂ > 300 mmHg, PCO ₂ =13-38 mmHg, pH 7.4) Tyrode solution of the following composition: 144.3 mmol/1 Na ⁺ , 4.0 mmol/1 K ⁺ , 138.6 mmol/1 Cl", 1.7 mmol/1 Ca ² \ 1.0 mmol/1 Mg ²⁺ , 0.4 mmol/1 HPO ₄ ² ", 19.9 mmol/1 HCO ₃ ', 10.0 mmol/1 D-glucose. An IPC roller pump (Ismatec, Wertheim, Germany) pumping at a frequency of 1.33 Hz was used for perfusion. After a 30 min equilibration time, the segments were perfused at 2, 90 or 160 mmHg for 3 -12 hours with the aid of an adjustable reloading system (Hugo Sachs Elektronik, March, Germany).

2. Cell culture

Endothelial cells were treated with 1 U/ml dispase in Hepes-

modified Tyrode solution for 7 min. at 37 ⁰ C, isolated and mounted on gel tin-coated 60 mm culture dishes (2 mg/ml gelatin in 0.1 M HCl for 30

Min. at ambient temperature) in DMEM/Ham F12 (1:1, v/v) containing 10

U/ml nystatin, 50 U/ml penicillin, 50 μ§/π&igr;1 streptomycin, 5 mmol/l Hepes, 5 mmol/l TES and 20% fetal calf serum. They were passaged once using 0.5% trypsin/0.2% EDTA (w/v) and seeded in BioFlex™ Collagen Type I six-well plates (Flexcell, Hillsborough, NC, USA) additionally coated with gelatin. They were identified by their typical cobblestone morphology, positive immunostaining for von Willebrand factor (vWF) and negative immunostaining for smooth muscle α-actin (Krzesz et al. (1999) 453, 191).

3. RT-PCR analysis

The frozen segments were ground under liquid nitrogen using a mortar and pestle. Total RNA was isolated using the Quiagen RNeasy Kit (Quiagen, Hilden, Germany), followed by cDNA synthesis using a maximum of 3 μg RNA and 200 U Superscript™ II reverse transcriptase (Gibco Life Technologies, Karlsruhe, Germany) in a total volume of 20 μl according to the manufacturer's instructions. For cDNA loading adjustment, 5 μl (approximately 75 ng cDNA) of the resulting cDNA solution and 20 pmol of each primer (Gibco) were used for elongation factor 1 (EF-I) PCR with 1 U Taq DNA polymerase (Gibco) in a total volume of 50 μl. Elongation factor-1 (EF-I) served as a standard for PCR. PCR products were separated on 1.5% agarose gels containing 0.1% ethidium bromide and band intensity was determined densiometrically using a CCD camera system and One-Dscan gel analysis software from Scanaltytics (Billerica, MA, USA) to adjust the volume of cDNA in subsequent PCR analyses.

All PCR reactions were performed individually for each primer pair in a Hybaid OmnE Thermocycler (AWG; Heidelberg, Germany). The individual PCR conditions for the cDNA of PAEC were as follows: ppET-1 (product size 432 bp, 30 cycles, annealing temperature 53 ⁰ C, (forward primer) 5'

GGAGC TCCAG AAACA GCTGT C 3', (reverse primer) 5' CTGCT GATAA ATACA CTTCT TTCC 3' (corresponds to nucleotide sequence 233-664 of the rat ppET-1 gene, GenBank accession number M64711); EF-2 (218 bp, 22 cycles, 58°C, 5' GACAT CACCA AGGGT GTGCA G 3', 5' GCGGT CAGCA CAATG GCATA 3' (1990-2207, human EF-2, Z11692).

4. Intravascular ET-I concentration

ET-I was extracted from weighed segments as described (Hisaki et al. (1994) Am. J. Physiol. 266, 422; Moreau et al. (1997) Circulation 96, 1593), and its concentration was determined using an ELISA kit (Amersham, Freiburg, Germany).

5. Reporter gene analysis

Partial sequences of the rat ppET-1 promoter, amplified by PCR, were cloned blunt-end into the Sma-1 restriction site of the multiple cloning region of the luciferase expression vector pCMV TK luc+ after excision of the CMV promoter (Paul et al. (1995) 25, 683). The pCMV TK Luc+ construct with and without the CMV promoter served as a control. Sequence-directed mutagenesis of the AP-I and GATA binding site in -168 bp construct was performed using the Transformer™ Site Directed Mutagenesis Kit (Clontech, Heidelberg, Germany). Co-transfection to estimate the transfection efficiency was performed with the SV40/ß-galactosidase expression vector pUC19 (Paul et al. (1995) supra).

For transfection, 40% confluent porcine aortic endothelial cells were incubated with 1.5 μ§ plasmid DNA and 15 μδ Effecten ™ (Qiagen, Hilden, Germany) for 6 hours , then the medium was replaced and the cells were cultured until they reached 80% confluence (usually after 18-24 h). They were then either incubated statically or were subjected to a 20% cyclic stress at

Ill ···

-27-

0.5 Hz in a Flexercell FX-3000 computerized stretch apparatus for 6 h. Luciferase and ß-galactosidase activity in the cell lysates was determined using the corresponding chemiluminescence and photometric assay kits (Promega, Mannheim, Germany) and normalized on the basis of protein content. Transfection efficiency was estimated by light microscopy by staining the cells with ß-galactosidase substrate, o-nitrophenyl-ß-D-galactopyranoside (Lim and Cha (1989) Biotechniques 7, 576).

6. Electrophoretic mobility shift analysis (EMSA)

Nuclear extracts and [32P]-labeled double-stranded consensus oligonucleotides (Santa Cruz Biotechnology, Heidelberg, Germany), nondenaturing polyacrylamide gel electrophoresis, autoradiography and supershift analysis were performed as described (Krzesz et al. (1999) supra). Oligonucleotides with the following single-stranded sequence were used:

(core sequences are underlined): AP-I, 5'

CGCTTGA TGACTCA GCCGGAA 3'; C/EBP, 5 ^; TGCAGA TTGCGCAA TCTG CA 3'.

7. Decoy oligodeoxynucleotide (dODN) technique

Double-stranded dODN were prepared from the complementary single-stranded phosphorothioate-linked oligodeoxynucleotides (Eurogentec, Cologne, Germany) as described (Krzesz et al. (1999) supra). The lumens of the aorta carotid or jugular vein segments were either filled with medium containing the double-stranded dODN or the dODN were pre-incubated with the cultured PAEC for 4 hours at a concentration of 10 μμ, conditions that were previously optimized based on EMSA and RT-PCR analysis. Thereafter, the dODN-containing medium was washed out by perfusion (segments) or replaced with fresh medium (PAEC). The single-stranded sequences of the dODN were as follows (underlined letters indicate phosphorothioate-linked bases): AP-I, 5'

-28-

CGCT TGATGACTCAGCC GGAA 3'; C/EBP, 5 ¹ TGCA GATTGCGCAATC TGCA 3'; C/EBPmut, 5' TGCA GAGACTAGTCTC TGCA 3'; GATA-2, 5' CACTTGATAACAGAAA GTGATAACTCTS'.

8. Statistical analysis

Unless otherwise indicated, all data in figures and text are expressed as mean ± SEM of η experiments with segments or cells from different animals. Statistical analysis was performed using the Students t-test for unpaired data with a P-value <0.05 considered statistically significant.

9. Pre-Pro ET-I mRNA expression

Pressurization of RbCA to 160 mmHg and RbJV to 20 mraHg resulted in maximal expansion with an average outer diameter expansion of 208 and 274%, respectively. The mRNA level of the housekeeping control gene EF-I was not altered by the change in perfusion pressure and no significant loss of EC from the perfused segments was observed. Increasing the perfusion pressure from 2 or 90 to 160 mmHg (RbCA) or from 0 or 5 to 20 mmHg (RbJV) resulted in a significant increase in both ppET-1 mRNA and intravascular ET-I (Figs. 1, 3 and 4). This presumably deformation-induced expression of ppET-1 was restricted to the endothelium, as it was strongly reduced after denudation of the segments (Fig. la). Pretreatment of RbJV with protein kinase C (PKC) inhibitor (Wilkinson et al. (1993) Biochem J. 295, 335), Ro 31-8220 (0.1 μg/L, Fig. Ib), but not with the Src family-specific tyrosine kinase inhibitor (Banes et al. (1995) 73, 349; Birukov et al. (1997) 81, 895), herbimycin A (0.1 μg/L, Fig. Ic) strongly inhibited the basal and pressure-dependent expression of ppET-1. In contrast, Ro 31-8220 had no effect in RbCA (Fig. Ie), whereas the deformation-induced increase of ppET-1 Ex-

t in* »ti*

-29-

pressure was almost abolished after exposure of the segments to herbimycin A (Fig. If). Blockade of actinomycin D RNA synthesis (I μπα/δ) abolished the pressure-induced increase in ppET-1 mRNA (Fig. Id) and intravascular ET-I in both vessel types.

Result: The ET-I gene is expressed de novo in response to pressure-dependent deformation of the EC.

10. Transcription factor activation

The use of AP-I and C/EBP oligos in EMSA revealed a transient (i.e., 30-60 min) and significant (two- to three-fold) increase in activity after exposure of RbJV to an intraluminal pressure of 20 mmHg (n=3-9 for each transcription factor). This nuclear translocation was most pronounced and reproducible in the case of AP-I (Fig. 2a) and C/EBP (Fig. 2b), which consisted primarily of the β- and δ-isoform according to supershift analysis. Furthermore, the abundance of AP-I in nuclear extracts was reduced both under basal conditions and in the presence of elevated perfusion pressures after pretreatment of RbJV with Ro 31-8220 (from 247±39 to 146±13% of control after 1 h perfusion at 20 mmHg, n=3). Almost identical results were obtained for AP-I and C/EBP in RbCA, with additional pressure-induced activation of C/EBPß and δ being blocked by herbimycin A.

Results: AP-I and C/EBP show increased binding activity in extracts of RbJV and RbCA after an increase in intraluminal pressure.

11. Role of AP-I and C/EBP

Both AP-I and C/EBP dODN showed significant but different effects.

Thus, in RbJV, the AP-I dODN abolished pressure-dependent, but not basal, ppET-1 mRNA (Fig. 4a) and ET-I peptide abundance (Fig. 4b). The C/EBP dODN, on the other hand, significantly increased both basal and pressure-dependent ppET-1 expression to the same extent (Fig. 4a and b). In contrast, no effect was observed with the mutated dODN, C/EBPmut, thus underlining the specificity of the dODN technique (Fig. 3b). The combination of AP-I with the C/EBP dODN significantly reversed the enhancing effect of the latter on ppET-1 expression (Fig. 4a and

Unlike in RbW, C/EBP, but not C/EBPmut dODN, abolishes the pressure-induced increase in ppET-1 expression in RbCA without affecting the basal level (Fig. 4c). In these segments, the AP-I dODN modestly reduces ppET-1 expression in a manner similar to that of the GATA-2 dODN in the RbJV (Fig. 3c).

Results: ppET-1 induction in response to pressure can be inhibited in RbJV by AP-I dODNs and in RbCA by C/EBP ODNs.

12. Stretch response of the rat ppET-1 promoter

PAEC were transiently transfected with various rat ppET-1 promoter-luciferase constructs. These cells were chosen because they express much higher levels of both ppET-1 mRNA and ET-I peptide in response to cyclic stress, which effect is reduced by both Ro 31-8220 (Fig. 3e) and herbimycin A (Fig. 3f). According to ß-galactosidase staining and measurement of ß-galactosidase activity, the transfection efficiency was estimated to be about 15%. All transfected promoter constructs (-98, -168, -350, -

-31 -

590, -729, -1070, -1250 and -1329 bp) were expressed by PAEC under static conditions, but at different levels, and, except for the -98 bp construct, showed a significant increase in luciferase activity when subjected to cyclic stress for 6 h (Fig. 5a). The differences in luciferase activity determined between the individual promoter constructs were not dependent on differences in ß-galactosidase transfection efficiency, since the activity of the enzyme appeared relatively constant in doubly transfected cells (Fig. 5c).

The full-length promoter construct showed the highest activity in stretched PAEC (approximately 40% of the luciferase activity of a CMV-driven gene under static conditions). The smallest construct with stretch response was the 168 bp fragment, whose expression level was approximately half of the full-length promoter construct (Fig. 5a and b). Deletion of the only AP-I binding site between -110 to -100 bp abolished the stretch response (Fig. 5a). A similar effect was obtained when PAEC transfected with this construct were treated with AP-I dODN (63 and 93% inhibition, respectively, Fig. 5b) or with 0.1 μg/L AP-I dODN (Fig. 5c). Ro 31-8220 to inhibit PKC activity (abrogating basal and reducing stretch-induced luciferase activity by 84%). In contrast, a GATA-2 consensus dODN reduced basal and stretch-induced luciferase activity in PAEC transfected with the -168 bp construct by only 28 and 31%, respectively. Furthermore, both basal and stimulated expression of the full-length promoter construct were significantly inhibited after exposure to AP-I dODN (by 58 and 74%, respectively) (Figs. 5a and b).

The C/EBP dODN had no significant effect on luciferase activity in PAEC transfected with the -590, -729, or -1250 bp construct but significantly increased stretch-dependent luciferase activity in PAEC transfected with the -1070 bp construct (Fig. 5b).

&bull; *

-32-

Result: The stretch response of the ppET-1 promoter in PAECs depends on AP-I. The ppET-1 promoter activity can be inhibited by AP-I dODN, but not by C/EBP dODN.

13. Transfection of interposition vein grafts in rabbits

To characterize the effect of AP-I decoys on neointimal lesion formation after arterial venous grafting using the rabbit external jugular vein and common carotid aorta, the establishment of an in vivo model is planned. In this experimental model, a surgical success rate and graft opening rate of about 75% is expected. Therefore, a group of 8 rabbits will be used to obtain sufficient data for morphometric analyses. Another group of 6 animals will be used to demonstrate efficient delivery of the decoy into the venous segment. General anesthesia is induced intravenously in New Zealand white rabbits (3-4 kg), followed by standard inhalation anesthesia using halothane. A vertical neck incision to expose both the external jugular vein and the ispsilateral common carotid aorta is performed. A four-centimeter segment of the vein is harvested using a traumatic technique, is then treated with topical papaverine and preserved in Ringer's solution containing heparin. Transfection of the vein segment is performed as follows: A narrow catheter is advanced into the vein segment to flush the segment free of Ringer's solution. After proximal ligation of the vessel, the decoy solution is infused into the vein, avoiding distension of the vein. After distal control is obtained by a second ligation, the decoy solution is incubated with a dwell time of 30 min to achieve sufficient transfection rates. During the incubation period, the ipsilateral carotid aorta is dissected and mobilized. This segment of the artery is divided by clamps, the transfected vein segment is reversed and sutured in an end-to-end manner using standard sutures between the carotid aorta and the venous incision.

After removal of the staples, homeostasis is established and the wound is closed. The animals in the control group receive a scrambled decoy that has no effect on AP-I activity.

Experiments subjected to efficiency study are terminated after 2 days and electrophoretic mobility shift assays (EMSA) are performed to demonstrate efficient squelching of the target transcription factor (AP-I). The biological effect of decoy treatment is demonstrated 3 weeks after transfection using standard computerized morphometry.

14. Transfection of the injured carotid aorta to prevent restenosis

To characterize the effect of C/EBP decoys on nointimal lesion formation in the rabbit carotid aorta after PTA, the establishment of an in vivo model of restenosis and atherosclerosis is intended.

Adult New Zealand white rabbits (3-4 kg) are used. Some of the animals are fed a diet containing 1% cholesterol for 2 weeks before balloon injury and they are maintained on this diet until removal of the transfected vessel. Balloon injury is performed after induction of anesthesia as described in 13. After a vertical nuchal incision, the right common carotid aorta is dissected free from surrounding tissues and collateral branches are ligated. The rabbits are systemically heparinized by intravenous injection of heparin (100 U/kg) before clamping of the carotid aorta. A 2-French Fogarty embolectomy catheter is introduced through a branch of the external carotid aorta and inflated in the mid-portion of the common carotid aorta and then withdrawn to the bifurcation. This procedure is repeated three times, achieving complete endothelial denudation. Then an infusion catheter filled with normal saline is advanced into the artery to flush the carotid aorta. After proximal control is achieved

·

&bull; · ·» &bull; ·

-34-

decoy solution is infused and incubated for 30 min. Blood circulation through the common carotid aorta is restored after litigation of the side branch used for catheter insertion, homeostasis is established and the wound is closed.

*' At the time intervals described above, rabbits are re-anesthetized and vessel segments are harvested. Experiments that are part of the efficiency study of dODN treatment are re-anesthetized and vessel segments are harvested. Experiments that are part of the efficiency study are stopped after 2 days, EMSA is performed to demonstrate efficient decoying of the target transcription factor (C/EBP). The biological effect of decoy treatment is demonstrated 4 weeks after transfection by standard computerized morphometry.

SEQUENCE LIST

<110> Cardiogene Gentherap. Systeme AG <120> Modulation of transcription of genes in vascular cells <160> <170> Word 6.0, PC-DOS/MS-DOS

<212>DNA

<213> Oryctolagus cuniculus <400> 1 ggagctccag aaacagctgt c

<210>2

<211>24

<212>DNA

<213> Oryctolagus cuniculus <400> 2 ctgctgataa atacacttct ttcc

<210>3

« &bull; &bull; &bull; &bull; · &bull; &bull; · &bull; · · &bull; ·· · · &bull; · · &bull; &bull; &bull; * *· &bull; · · &bull; · · * ·

<212>DNA <213> Oryctolagus cuniculus

<400> 3

J.
gacatcacca agggtgtgca g

<210>4 <211>20 <212>DNA <213> Oryctolagus cuniculus

<400> 4

gcggtcagca caatggcata 20

<210>5

<212>DNA <213> artificial sequence

<220>

<223> consensus binding site for AP-I

<400> 5

cgcttgatga ctcagccgga a 21

<210>6 <211>20

<212>DNA <213> artificial sequence

<223> consensus binding site for C/EBP

tgcagattgc gcaatctgca

<211>20 <212>DNA <213> artificial sequence

<223> mutated consensus binding site for C/EBP

tgcagagact agtctctgca

<211>27 <212>DNA <213> artificial sequence

<223> consensus binding site for GATA-2

<400> 8 cacttgataa cagaaagtga taactct 27

<2&Idigr;0>9 <212>DNA

<213> artificial sequence <220>

<223> constructed AP-I <400> 9 gtgctgactc agcac

<210> 10

<211> 17

<212>DNA

<213> artificial sequence <220>

<223> constructed AP-I

<400> 10 cgcttagtga ctaagcg

<211> 17

<212>DNA

<213> artificial sequence

<223> constructed AP-I

<400> 11 tgtgctgact cagcaca

<210> 12 <212>DNA

<213> artificial sequence

<223> constructed AP-I <400> 12 ttgtgctgac tcagcacaa

<2lO> 13 <212>DNA

<213> artificial sequence

&bull; · · e * · · · ■ «&phgr; * %&phgr; φ ^ _% φ _{&Lgr; φ} φ

<223> constructed AP-I

<400> 13

tcgcttagtg actaagcga

<210> 14 <211>20 <212>DNA <213> artificial sequence

<223> constructed AP-I

<400> 14

tgctgactca tgagtcagca

<210> 15 <211> 20 <212>DNA <213> artificial sequence

<223> constructed AP-I

<400> 15

tgctgactaa ttagtcagca

<210> 16 <212>DNA

<213> artificial sequence

<223> constructed AP-I <400> 16 gtcgcttagt gactaagcga C

<210> 17 <212>DNA

<213> artificial sequence

<223> constructed AP-1 <400> 17 cttgtgctga ctcagcacaa g

<210> 18

<211>22

<212>DNA

<213> artificial sequence

<220>

<223> constructed AP-I

<400> 18 i

ttgctgactc atgagtcagc aa 22

<210>19 <212>DNA

<213> artificial sequence <220>

<223> constructed C/EBP <400> 19 gacattgcgc aatgtc 16

<210>20

<211> 16

<212>DNA

<213> artificial sequence <220>

<223> constructed C/EBP <400> 20

<210>21

<2&Iacgr;2> DNA <213> artificial sequence

<220> <223> constructed C/EBP

<400>21

<210>22 , <211> 18 <212>DNA <213> artificial sequence

<220> <223> constructed C/EBP

<400> 22

<210>23 <2U> <212>DNA

agcattggcc aatgct 16

cgacattgcg caatgtcg

aggcattggc caatgcct 18

<213> artificial sequence

<220> <223> constructed C/EBP

<400> 23

acgacattgc gcaatgtcgt 20

<210>24 <211>20 <212>DNA <213> artificial sequence

<220> <223> constructed C/EBP

<400> 24

taggcattgg ccaatgccta 20

<210> 25 <211>22 <212>DNA <2 1 3> artificial sequence

<220><223> constructed C/EBP

<400> 25

ctgttgcgca attgcgcaac ag

<210>26

<k1> 22

<212>DNA

<213> artificial sequence

<223> constructed C/EBP <400> 26 gacttgcgca attgcgcaag te

Claims (2)

1. A double-stranded nucleic acid capable of binding sequence-specifically to the transcription factor AP-1 or a related transcription factor having any of the sequences of SEQ ID 5 or 9 to 18 . 2. A double-stranded nucleic acid capable of binding sequence specifically to the transcription factor C/EBP or a related transcription factor having any of the sequences of SEQ ID 6 or 19 to 26 .