DE19845434C1 - Tissue binding peptides and related nucleotide sequences useful for tissue specific gene transfer - Google Patents

Abstract

Tissue binding peptides chosen from 8 peptides (given in the specification) are new. Independent claims are also included for: (1) a nucleic acid coding for a tissue binding peptide as above; (2) a vector containing a nucleic acid as above; (3) a method to produce a tissue binding peptide; and (4) a method to identify a tissue binding peptide comprises contacting a tissue with one or more peptides and isolating a tissue binding peptide.

Description

The present invention relates to tissue-binding peptides, their identification with Aid to tissues, their manufacture and use as a medicine or diagnostic.

The pathogenesis of arteriosclerosis is increasingly due to damage to the Endothelium, which results in endothelial dysfunction. An endothelial Dysfunction is primarily reflected in a change in the expression pattern of Surface proteins (selectins, adhesion molecules) but also in modified ones Synthesis rates of growth hormones again. So z. B. in the early stages of Arteriosclerosis an increased expression of the E- and P-selectins and the cellular Adhesion molecules VCAM and ICAM for the immigration of monocytes into the vessel wall. The conversion of the monocytes into macrophages leads to an altered synthesis of Growth factors, in particular smooth muscle cells for proliferation and Stimulate the release of mediators that have an inflammatory effect.

Drug therapy for atherosclerosis tries the causes early, if possible decrease before the manifestation. A main goal is to increase the Elimination of cholesterol metabolites (bile acids) or through a reduction in Cholesterol biosynthesis to reduce the elevated plasma cholesterol level. Due to the Counter-regulation initiated by the organism can be treated with drug therapy lower the total cholesterol level by only about 30%. Because this procedure also doesn't is suitable for inducing regression of an existing arteriosclerosis percutaneous transluminal angioplasty (PTA), d. H. the expansion of the stenosed Vascular section using a balloon catheter, the therapy of choice today. disadvantage However, this form of therapy is that in 30-50% of the treated patients there is a Restenosis, d. H. comes to a reclosure. Alternatively, therefore, a surgical one Treatment by bypassing. But also with this therapy occurs in approx. 50% of patients restenosed within 3-6 months after treatment Bypass vessel.  

A number of gene therapy processes have therefore been developed in which a second catheter is used, via which a gene therapeutic vector locally the affected section of the vessel is applied (see, for example, WO 95/27070).

Various strategies have been developed for the local application of the therapeutic gene Based on modified balloon catheters developed, the direct application of a Allow substance or the gene into the vessel wall. After a local application with a double balloon catheter z. B. Nabel, E.R. et al. (1990) Science 249, 1285 a transient expression of the β-galactosidase gene in transfected cells of the femoral Detect pig artery by liposomal and retroviral transfection. A Long-term expression (up to 5 months) was only possible after retroviral Achieve transfection while expression in liposomally transfected cells only up to was detectable for 4 weeks. In addition to the duration of the transfection, the short duration is particularly important Transfection efficiency is a limiting factor of liposomal transfer. The Transfection efficiency could be improved, for example, by using HVJ liposomes can be increased (Morishita, R. et al. (1994) Gene 149, 13). In this system inactivated Sendai viruses (hemagglutinating virus of Japan, HVJ) with DNA-containing Liposomes fused. Due to the viral fusion and binding proteins (F and HN Protein) the genetic material can go directly into the bypassing the lysosomes Cytoplasm of the cell can be efficiently introduced. Finally there is the possibility cell-specific antibodies to liposomal (Vingerhoeds, H. et al. (1994) Immunmeth. 4, 259) or viral (Wickham, T.J. et al. (1996) J. Virol. 70, 6831), the therapeutic gene under the control of a strong viral promoter to be able to target only the desired cell type.

However, the disadvantage of this method is that an additional intervention is necessary and secondly, the efficiency of local gene transfer in the organism in general is not is very high. After all, not all vessels are reached with a catheter, especially not the small-caliber arteries and capillaries. This is important because Arteriosclerotic wall changes not only affect the large vessels, but also occur in smaller arteries.  

The optimization of a tissue-specific expression is particularly useful for the therapy of vascular diseases of great medical importance. It doesn't just open up Possibilities of direct vascular diseases, such as B. arteriosclerosis and their Treat secondary diseases (stenosis, restenosis, heart attack), but also all over organs to reach the bloodstream as well as tumors. Ultimately, one can Tissue-directed gene transfer into vascular cells also expresses therapeutic Proteins occur that are due to pathological or genetic changes in the Target organism is not or no longer available to a sufficient extent, e.g. B. factor VIII deficiency, insulin deficiency etc.

An essential goal of somatic gene therapy is therefore to seek out a therapeutic gene systemic or local administration as specific as possible in the target cells of the Introducing the organism and the therapeutic gene essentially only in these cells to express. For this purpose, the therapeutic gene should be packaged on the one hand so that it Transport through the blood vessel system is essentially not degraded by nucleases, and on the other hand it should be in a form that expresses cell-specific gene expression the target cells. The specific recognition of cells is also called "Targeting". There are basically two options for this targeting: on the one hand, one can use antibodies against receptors on the cell surface, which in viral or liposomal vector systems are integrated (Vingerhoeds, H. et al (1994) supra; Wickham, T. J. (1996) supra), and on the other hand peptides with high Use binding affinity for receptors on the cell surface.

With certain antibodies, for example, certain tumor cells could be recognized and can be detected even in intact tissue by means of a suitable marking. A major disadvantage of antibodies, however, is that they are mostly for technical reasons are produced in the mouse and often have serious side effects on patients due to an immune response. Even so-called humanized antibodies are not free from this problem. In individual cases, such pronounced immune complexes can occur come that drastically restrict the blood flow to the kidney and thus become a lethal lead to renal failure.

With peptide-presenting banks, for example, interactions of proteins were possible  other macromolecules such as sugars, antibodies, lipids or proteins studied in vitro become. Peptide presenting libraries that are chemically produced have been developed e.g. B. from Geysen, H.M. et al. (1996) Mol. Immunol. 23, 709 or de Koster, H. S. (1995) J. Immunol. Methods 187, 179-188. In peptide synthesis, the Solid phase amino acid mixtures are used so that the amino acid sequence of the resulting peptides statistically, i. H. was randomly distributed. The peptide banks could can also be used in resin-bound form (de Koster, H.S. et al. (1995) supra).

With the use of peptide-presenting banks, which are based, for example, on the presentation of the peptides on the surface of phages, it became possible to present a large number of different peptides (up to approximately 10 ¹¹ peptides) at the same time. The peptides are z. B. at the N-terminus of the pIII coat protein of filamentous phages (see e.g. Kay, BK et al. (1993) Gene 128, 59 or Jellis, CL et al. (1993) Gene 137, 63-68). The length of the peptides presented varied from 6 to 38 amino acids. The phage banks were used, among other things, for the epitope screening of antibodies (Cortese, R. et al. (1994) Trends Biotechnol. 12, 262; Grihalde, ND et al. (1995) Gene 166, 187), for the characterization of protein- or peptide-binding Domains (Adey, NB & Kay, BK (1996) Gene 169, 133; Blond-Elguindi, S. et al. (1993) Cell 75, 717-728; Balass, M. et al. (1993) Proc. Natl. Acad. Sci. USA 90, 10638-10642), for the discovery of DNA-binding motifs (Wang, B. et al. (1995) J. Biol. Chem. 270, 23239), for the identification of unknown receptors which e.g. B. to viral binding proteins (Hong, SS & Boulanger, P. (1995) EMBO J. 14, 4714), for the identification of protein-binding peptides (Daniels, DA & Lane, DP (1994) J. Mol. Biol. 243, 639-652; Hong, SS & Boulanger, P. (1995) EMBO J. 14, 4714-4727, No. 19) or single chain Fv (scFv) antigen binding fragments (Vaughan, TJ et al. (1996) Nature Biotechnology 14, 309-314) and proposed for targeting cells (Barry, MA et al. (1996) Nature Medicine 2, 299-305, No. 3).

In addition to peptide-presenting phage banks, peptide-presenting bacterial banks are known. An example of this are fusions of a combinatorial peptide library with thioredoxin, which led to the identification of peptide aptamers which recognize the cyclin-dependent kinase 2 (Colas, P. et al. (1996) Nature, 380, 548-550). Browsing a combinatorial peptide library was e.g. B. also with fusions to the C-terminus of the alpha subunit of G _t (340-350) to identify rhodopsin-binding sequences (Martin, EL et al. (1996) J. Biol. Chem. 271, 361-366 , No. 1). Another example is peptide insertions in a thioredoxin-flagellin fusion protein, which is presented on the E. coli cell surface, with which protein-protein interactions can be investigated (Lu, Z. et al. (1995) Bio / Technology 13, 366- 372; U.S. Pat. No. 5,635,182).

The applications of the peptide presenting banks described above were limited however on in vitro investigations with isolated and cleaned components (sugar, Proteins, antibodies).

The object of the present invention was therefore to find a means which is targeted and treatment and diagnosis of diseased organs as gentle as possible.

One object of the present invention is therefore a tissue-binding peptide selected from a peptide with the amino acid sequence

GEGRTVVLSF, AWCRGGILGDAM, GNLVDLVVGFDD, RVSPPKKSGGGV, GSSKWGLTXKCG, RGGVRQRSRGRR, GEGRTVVCRS or SQRWTALWQWIG

as well as variants of it.

According to the present invention, variants are understood to mean deletions, additions or substitutions of one or more amino acids, the tissue-specific Binding of the peptide is essentially unchanged, i.e. H. the peptide still has retain its ability to bind better to one tissue than another.

In case of deletion, preferably no more than 4 amino acids, especially not deleted more than 2 amino acids. The deletion is preferably carried out on the N- or on the C- Terminus of the peptide, especially at the N-terminus.

According to the present invention, the term addition does not only mean that Addition of one or more amino acids, for example 1-25 amino acids,  but also fusion proteins from the peptide according to the invention with another C- and / or N-terminal peptide or protein. For example, a merger leads one tissue-specific peptide with a peptide that contains several histidines, for example the peptide GLFHAIAHFIHGGWHGLIHGWYG not only for tissue-specific Binding, but also to permeabilize the cell membrane at slightly acidic pH (See, e.g., Midoux, P. et al. (1998) Bioconjugate Chem. 9, 260-267). Exists Fusion protein, for example from the peptide according to the invention and several lysines, for example 16 sequential lysines or the nucleocapsid protein (NCp) 7 of HIV-1 or NCp7-derived peptides, the peptide according to the invention can be particularly simple complexes a nucleic acid, for example a DNA containing a therapeutic gene (see e.g. Harbottle, R.P. et al. (1998) Human Gene Therapy 9, 1037-1047 or Bachmann, A.S. et al. (1998) J. Mol. Med. 76, 126-132). According to the present The invention can thus also be used, for example, to produce a fusion protein that contains a tissue-specific peptide, a peptide which effects permeabilization of the cell membrane and contains a nucleic acid binding peptide, which is particularly advantageous.

In the case of substitutions, conservative substitutions are preferably understood, i. H. Substitutions where, for example, one or more amino acids with one hydrophobic side chain by one or more other amino acids with another hydrophobic side chain can be replaced. Examples of amino acids with hydrophobic Side chains are glycine, valine, leucine, isoleucine, methionine, phenylalanine, tryptophan or proline. Likewise, amino acids with a polar neutral residue can be mutually be replaced. Examples include serine, threonine, cysteine, tyrosine and asparagine or glutamine. Furthermore, amino acids with a polar acidic residue can be mutually be replaced. For example, this includes aspartic acid or glutamic acid. Amino acids with a polar basic residue can also be exchanged with one another be such. B. lysine, arginine or histidine. Preferably no more than up to 4 amino acids, especially not more than up to 2 amino acids, especially not more than substituted an amino acid.

According to the present invention, the term amino acid means not only that in proteins naturally occurring amino acids, but also others in a peptide chain build-in amino acids, such as. B. hydroxylysine, hydroxyproline, ornithine, citrulline or  Homocysteine.

The individual amino acids of the peptide according to the invention can also be modified. For example, the peptide can be marked and / or protected by a modification. A suitable label is, for example, the exchange of the hydroxyl group of threonine or serine with the isotope ¹⁸ F, which has a half-life of 2 hours and can therefore be used well in clinical practice. Alternatively, the peptide can also be labeled with ¹⁸ F at the carboxy terminus via a fluoroethylamide bond. The C-terminal fluoramidation is carried out, for example, with fluoroethylamine and a urea derivative (for example TBU) for activation.

Another suitable label is the labeling of cysteine or methionine with the sulfur isotope ⁷³ S. The isotope ⁷³ S has the advantage over the isotope ¹⁸ F that it has a longer half-life.

The peptide according to the invention can also be labeled with other compounds, for example with β-galactosidase or biotin or avidin / streptavidin. A radioactive one However, labeling with in particular short-lived isotopes has the advantage that the structure of the peptide remains essentially unchanged and the distribution of the peptide in the organism for example with the help of positron emission tomography (PET) in an advantageous manner can be analyzed.

For labeling, it is generally advantageous if the functional side chains of the amino acids are protected beforehand. For example, the peptide on the solid phase is converted from the C- to the N-terminus with the aid of an acid-labile resin (e.g. SASRIN, a 2-methoxy-4-alkoxybenzyl alcohol resin, Fréchet, JMJ et al. (1979) Polymer 20, 675) using the FastMoc® method (Merrifield, RB (1963) J. Am. Chem. Soc. 95, 1328; Wang, SS (1973) J. Am. Chem. Soc. 95, 1328; Perkin-Elmer Applied Biosystems, Foster City, CA). In this way, the peptide can be cleaved from the resin under conditions in which both the side chain protecting groups and the N-terminal protecting group (Fmoc, N-fluorenylmethoxycarbonyl) remain stable. The marking can then be done, for example, with ¹⁸ F. This is followed by the elimination of the protective groups and the purification via HPLC.

Another object of the present invention is a nucleic acid coding for a tissue-binding peptide according to the invention. A nucleic acid is particularly preferably selected from a nucleic acid with the nucleotide sequence

as well as variants of it.

According to the present invention, the term variants is understood in particular Variants due to the degeneracy of the genetic code for the same peptide encode. The term variants is also understood to mean those listed Deoxyribonucleic acids (DNA) corresponding ribonucleic acids (RNA). Can too the DNA or RNA may be modified to e.g. B. better protected against nuclease degradation to be. Suitable modifications can be found e.g. B. from Uhlmann, E. & Peyman A. (1990) Chemical Reviews 90, 543-584 No. 4th

Another embodiment of the present invention relates to a vector containing one or more nucleic acids according to the invention. For example, a vector can be a Expression vector for the production of the peptides according to the invention in prokaryotic or be eukaryotic cells. Examples of prokaryotic expression vectors are for the Expression in E. coli e.g. B. the vectors pGEM or pUC derivatives and for eukaryotic Expression vectors for expression in Saccharomyces cerevisiae e.g. B. the vectors p426Met25 or p426GAL1, for expression in insect cells e.g. B. Baculovirus Vectors as disclosed in EP-B1-0 127 839 or EP-B1-0 549 721, and for expression in Mammalian cells e.g. B. the vectors Rc / CMV and Rc / RSV or SV40 vectors, all of which are generally available.  

In general, the expression vectors also contain suitable ones for the respective host cell Promoters such as B. the trp promoter for expression in E. coli (see, for example, EP-B1-0 154 133), the ADH2 promoter for expression in yeasts (Russel et al. (1983), J. Biol. Chem. 258, 2674-2682), the baculovirus polyhedrin promoter for expression in Insect cells (see e.g. EP-B1-0 127 839) or the early SV40 promoter or LTR Promoters e.g. B. from MMTV (mouse mammary tumor virus; Lee et al. (1981) Nature 214, 228-232).

Examples of vectors which are active in gene therapy are virus vectors, preferably Adenovirus vectors, especially replication-deficient adenovirus vectors, or adeno associated virus vectors, e.g. B. an adeno-associated virus vector consisting exclusively of two inverted terminal repeat sequences (ITR).

For example, in adenoviral vectors, in particular of type 5 (see sequence for Chroboczek, J. et al. (1992) Virol. 186, 280-285) and especially from subgroup C, in general the E1 gene region by a foreign gene with its own promoter or by nucleic acid according to the invention replaced. By exchanging the E1 gene region, which for the expression of the downstream adenoviral genes is a prerequisite, a does not arise replication-capable adenovirus. These viruses can then only be found in one cell line multiply, which replaces the missing E1 genes.

Replication-deficient adenoviruses are therefore generally formed by homologous recombination in the so-called 293 cell line (human embryonic kidney cell line), which has a copy of the E1 region stably integrated in the genome. For this purpose, the nucleic acid according to the invention is cloned into recombinant adenoviral plasmids under the control of a suitable promoter. Then the homologous recombination with an E1-deficient ten adenoviral genome, such as. B. d1327 or de11324 (adenovirus 5), in the helper cell line 293. When recombination is successful, viral plaques are harvested. The replication-deficient viruses produced in this way are used in high titers (for example 10 ⁹ to 10 ¹¹ "plaque forming units" or plaque-forming units) for infection of the cell culture or for somatic gene therapy.

In general, the exact insertion point of the nucleic acid according to the invention is in the  adenoviral genome not critical. It is Z. B. also possible, the invention Cloning nucleic acid in place of the deleted E3 gene (Karlsson, S. et al. EMBO J. 5, 1986, 2377-2385). Preferably, however, the E1 region or parts thereof, e.g. B. the E1A or E1B region (see, for example, WO 95/00655) by the nucleic acid according to the invention replaced, especially if the E3 region is also deleted. Furthermore are suitable Third generation adenoviral vectors, i. H. Vectors other than the ITR and the Packaging signal contain no further sequences coding for viral proteins (Kochanek, S. (1996) Proc. Natl. Acad. Sci. USA 93, 5731). The recombination in the Helper cell line takes place e.g. B. with the help of a helper plasmid.

However, the present invention is not limited to the adenoviral vector system, but also adeno-associated virus vectors are suitable in combination with the Nucleic acid according to the invention in a special way, since the transfer of nucleic acid according to the invention are carried out in resting, differentiated cells by AAV can, which is particularly advantageous for the treatment of vascular diseases. For the Vector functions generally suffice for the two inverted approx. 145 bp long terminal repetition sequences (ITR: inverted terminal repeats; see e.g. WO 95/23867). They carry the signals necessary for replication, packaging and in "cis" Integration into the host cell genome. The ability to integrate can take a long time sustained gene expression in vivo can be ensured, which in turn is special is advantageous. Another advantage of AAV is that the virus is not pathogenic to the Humans and is relatively stable in vivo. The cloning of the invention Nucleic acid in the AAV vector or parts thereof is carried out according to the person skilled in the art Methods such as B. in WO 95/23867, at Chiorini, J.A. et al. (1995) Human Genes Therapy 6, 1531-1541 or Kotin, R.M. (1994) Human Gene Therapy 5, 793-801 are described.

Vectors with gene therapy effects can also be obtained by complexing the nucleic acid according to the invention with liposomes (neutral or cationic), i. H. the Nucleic acid is essentially included in the liposome because it is very high Transfection efficiency, especially of cardiac muscle cells, can be achieved (see e.g. WO 95/27070) and the nucleic acid is essentially protected against DNases. Especially Transfection with nucleic acid-liposome complexes using Sendai is advantageous  Viruses in the form of so-called HVJ liposomes (virosomes), because of this the transfection rate can be further increased.

In lipofection, small unilamellar vesicles are made from cationic lipids Ultrasound treatment of the liposome suspension produced. The nucleic acid becomes ionic bound on the surface of the liposomes in such a ratio that a positive net charge remains and the nucleic acid 100% from the liposomes is complexed. In addition to those of Felgner et al. (Flegner, P.L. et al. (1987) Proc. Natl. Acad. Sci USA 84, 7413-7414) used lipid mixtures DOTMA (1,2- Dioleyloxypropyl-3-trimethyl-ammonium bromide) and DOPE (Dioleoylphosphatidy lethanolamine), numerous new lipid formulations have now been synthesized and based on tested their efficiency of transfection of different cell lines (Behr, J.P. et al. (1989), Proc. Natl. Acad. Sci. USA 86, 6982-6986; Felgner, J.H. et al. (1994) J. Biol. Chem. 269, 2550-2561; Gao, X. & Huang, L. (1991), Biochim. Biophys. Acta 1189, 195-203). Examples of the new lipid formulations are DOTAP N- [1- (2,3-dioleoyloxy) propyl] - N, N, N-trimethylammonium ethyl sulfate or DOGS (TRANSFECTAM; Dioc tadecylamidoglycylspermin). An example of the production of DNA liposome com complexes of phosphatidylcholine, phosphatidylserine and cholesterol and their successful Application in the transfection of vessel walls with the help of Sendai viruses is in the WO 95/27070.

It is particularly advantageous if the nucleic acid-liposome complex comprises nucleic acid Binding proteins, for example chromosomal proteins, preferably HMG proteins (high Mobility Group Proteins), especially HMG-1 or HMG-2, or nucleosomal histones such as H2A, H2B, H3 or H4, because this expresses the desired Nucleic acid can be increased at least 3-10 times. The chromosomal proteins can, for example, from calf thymus or rat liver by generally known methods isolated or genetically engineered. Human HMB-1 can, for example, after methods known to the person skilled in the art are particularly easy to use by genetic engineering human cDNA sequence from Wen, L. et al. (1989) Nucleic Acids Res., 17 (3), 1197-1214, getting produced.

Another object of the present invention is therefore a method for  Preparation of a peptide according to the invention, in which the peptide is either chemical synthesized or, as already explained above, is produced genetically. The chemical Synthesis is carried out, for example, using the generally known Merrifield technique.

However, tissue-binding peptides can also be used according to the present invention obtained that one or more peptides are brought into contact with a tissue and then the bound peptide (s) are isolated.

Another object of the present invention is therefore a method for finding a tissue-binding peptide comprising the following steps:

• a) contacting a tissue with one or more peptides, and
• b) isolating one or more tissue-binding peptides.

Preferably, after contacting a tissue with one or more Peptides according to step (a) unbound peptides, for example by washing the Remove tissue with a suitable buffer solution.

According to the present invention, peptide is preferably understood to mean peptides with a Length of approx. 5 to approx. 40 amino acids, in particular with a length of approx. 5 to approx. 20 Amino acids, especially with a length of about 10 to about 20, particularly preferably with a length of about 10 to about 15 amino acids.

According to the present invention, tissue is understood to mean cell groups, in particular Cellular groups including their intercellular substance, for example epithelial tissue, Connective tissue, supporting tissue, muscle tissue or nerve tissue. The term tissue according to the present invention also includes whole organs or parts thereof as well Vessels. The organs or vessels are preferably still "living" Organs or vessels.

The intercellular substance is a substance embedded in the intercellular spaces, the one contains structurally appearing basic substance and fibrous connective tissue fibers. It was therefore completely surprising that by bringing tissue into contact with one or  several peptides tissue-binding peptides were found, which are essentially for the Tissues used were specific and not predominantly related to non-specific structures such as e.g. B. the intercellular substance were bound.

The specificity of the tissue-binding peptides can preferably be increased by the fact that the Peptides isolated according to step (b) are repeated one or more times with the tissue in Be brought in contact. The peptides are preferably after their respective isolation brought into contact with the corresponding tissue up to 10 times. That way you can cross-specific peptides that are specific for several tissue types were also found when the peptides are sequentially contacted with different tissues become. For example, if the peptides are taken in portions with both healthy and brought into contact with diseased tissue, the Peptides are found that are specific for healthy or diseased tissue. Peptides, which are essentially only specific for diseased tissue are suitable, for example particularly advantageous for diagnosis, for example in the context of a PET examination.

It is also known that by external circumstances such. B. stress or inflammation, Tissues can be changed pathologically. So change z. B. smooth muscle cells in restenotic vascular areas their entire phenotype and go from resting, secretory into the proliferative phenotype. With the help of the present invention can therefore also be found tissue-binding peptides that are in vivo Being able to recognize conditions specific to pathologically modified tissue. It is too Advantageously possible pathological cell types in the tissue that previously existed could not yet be sufficiently characterized, specifically using the to characterize selected peptides according to the method of the invention identify and, for example, treat with gene therapy.

It is to find tissue-binding peptides according to the method according to the invention advantageous if a peptide bank, for example a combinatorial peptide bank (see e.g. B. Colas, P. et al. (1996) supra) is used. Under peptide library according to the present Invention is understood to be a collection of several different peptides that are free or are available in bound form.  

For example, a peptide library of free peptides contains a collection of chemically synthesized peptides. (see e.g. Geysen, H.M. et al. (1996) supra). In the Peptide synthesis was used on the solid phase amino acid mixtures, so that the Amino acid sequence of the resulting peptides was statistically distributed. The peptide banks could also be used in resin-bound form (de Koster, H. S. et al. (1995) supra).

Peptides in bound form are, for example, so-called peptide-presenting banks, e.g. B. a peptide-presenting phage bank or preferably a peptide-presenting bacterial bank, as already described in more detail above. At the peptide-presenting banks, the peptides are presented in the form of a fusion protein on the surface of, for example, phages or bacteria. Preferred fusions are insertions, preferably fusions with thioredoxin, with thioredoxin (TrxA) -flagellin (FliC), with the alpha subunit of G _t , as already described in more detail above, or z. B. with LamB (Charbit, A. et al. (1988) Gene 70, 181-189) or Lpp-OmpA (Francisco, JA et al. (1993) Proc. Natl. Acad. Sci. USA 90, 10444-10448 ).

For the present invention, peptide-presenting bacterial banks and in particular those with a thioredoxin fusion protein are particularly preferred especially for selection in an in vivo approximate perfusion model of isolated vessels and organs for the following reasons:

• 1. The peptides are presented as fusion proteins together with a protein, for example thioredoxin, the three-dimensional structure of which is generally known. The randomized peptides are generally presented in a domain which is very easily accessible and is located on the surface of the total protein, for example the so-called "Cys-Cys active site loop" of thioredoxin. The fact that the peptide is generally presented sufficiently far from the bacterium and is thus easily accessible generally enables a good interaction with the desired binding partner.
  Consequently, it can generally be ensured that, for example, in the case of viral gene transfer into vascular cells, the peptides can also develop their effect in the form of fusion proteins with a protein having a therapeutic effect without essentially losing their binding capacity. Peptides which are fused to the terminus of a protein and are thus presented in phage libraries generally lose their binding capacity as soon as they are expressed as fusion proteins together with a therapeutically active protein, which is disadvantageous for use in gene therapy. An important reason for this is that the peptides themselves have a free N or C terminus during the presentation which is available for binding. For this reason, it is particularly advantageous if the peptides are presented in the form of inserts into selected surface proteins by a peptide-presenting bank.
• 2. For example, the E. coli thioredoxin is stable against proteases. Hence is a protease resistant protein in the form of a fusion protein with the peptide, for example for selection in the ex-vivo perfused organ model is particularly preferred, especially since Proteases can be found in the vessels on the cells. Especially organs, e.g. B. isolated perfused liver, but also tumors, release a high level of proteases, that could cleave the fusion protein in the event of protease sensitivity.
• 3. Because of their size, bacteria cannot move as quickly as phages in the reach extravascular space and thus, for example, in the perfused organ on cells reach that are not accessible for later therapy. In contrast to phages appropriate bacterial banks can also be incubated at body temperature without that the results are essentially falsified by endocytosis. Hereby the in vivo conditions can also be essentially met ex vivo, which for example, is particularly advantageous in the case of "living" organs or vessels.

According to the present invention, the term “living” means organs or vessels that are still functional. To maintain functionality over a longer period of time, it is therefore advantageous if the organ or the vessel are adequately supplied with oxygen and nutrient substrates. A nutrient solution which contains Ca ²⁺ , Mg ²⁺ and a protein, for example an albumin such as BSA (bovine serum albumin), is preferably suitable for this. The concentration of calcium ions is preferably 2.5 mM and of magnesium ions 1.1 mM. Albumin is preferably present in the nutrient solution at a concentration of 1%. The so-called Krebs-Henseleit solution, which is composed as follows (data in mM), is particularly suitable for the perfusion of isolated vessels and isolated organs: NaCl 116; KCl 4.6; MgSO ₄ 1.1; NaHCO ₃ 24.9; CaCl ₂ 2.5; KH ₂ P0 ₄ 1.2; Glucose 8.3; Pyruvate 2.0 and 1% BSA equilibrated with 95% O ₂ and 6% CO ₂ (pH 7.4; 37 °) (see e.g. Decking, UKM et al. (1997) Circulation Research 81, 154-164 , No. 2).

In a preferred embodiment, for example, a peptide-presenting Bacterial bank used in an in vitro vascular perfusion model after previously in vivo in the whole animal (rat) on one vessel a restenosis by denudation of the endothelium had been induced by means of a balloon catheter. This procedure has the decisive advantages that the occurring as a result of pathological vascular processes Changes in expression patterns of membrane-associated or membrane-bound proteins are retained, the cell polarity does not change and all cells remain in their natural tissue structure.

The peptides found with the aid of the method according to the invention can, as above already described, modified. A suitable modification is for example a radioactive or non-radioactive label for use in a test system or diagnostic.

For the tissue-specific peptides that can be found according to the method according to the invention there are numerous applications, for example tissue-specific transfer of Substances, especially of pharmaceutically active compounds, especially for the tissue-specific gene transfer. Tissue-specific gene transfer serves, for example for the diagnosis and / or therapy of vascular and / or organ diseases with the help for example viral and / or non-viral vectors, preferably with the aid of Liposomes, as already described in more detail above.

For tissue-specific gene transfer, the peptide can be physical, chemical or be genetically linked to the desired nucleic acid (see above). For example the peptide either together with a therapeutic protein, e.g. B. nitric oxide Synthase (see e.g. WO 95/27020), insulin (see e.g. EP-B1-0 001 929), erythropoietin  (see e.g. EP-B1-0 148 605), or blood coagulation factors such as e.g. B. factor VIII, Interferons, cytokines, hormones, growth factors, etc. are expressed, or to one Nucleic acid which is a therapeutic gene, for example a gene coding for the exemplified therapeutic proteins or for an antisense nucleic acid for targeted switching off of expression products. Suitable Coupling methods are e.g. B. on lysine residues or nucleocapsid proteins, as above explained in more detail (see, for example, Harbottle, R.P. et al. (1998) supra or Bachmann, A.S. et al. (1998) supra). It is particularly advantageous if one or more tissue-binding peptides via a positively charged domain, e.g. B. polylysine, to one or more nucleic acids are bound.

To increase the membrane permeation during gene transfer, a peptide is also suitable, which Membrane permeation increased, e.g. B. a hisitidine-containing polypeptide (see above). Preferably a so-called polyfection solution containing a nucleic acid with the desired therapeutic gene, a fusion protein of tissue-specific peptide and a DNA binding portion, e.g. B. a positively charged domain, and a peptide that the Membrane permeation increased, used. A suitable polyfection solution is, for example in Midoux, P. (1998) supra. Furthermore, a coupling of the peptide to the Liposomes via an inserted C-terminal cysteine to an activated one Lipid component, e.g. B. over MPB-PE, maleinimidophenylbutyroylphosphatidylethanolamine, (Zuidam, N.J. et al. (1995) Biochim. Biophys. Acta 1240, 101).

The peptides according to the invention can also be used to change the tropism of viruses, preferably adenoviruses, for example for tissue-specific gene transfer. This is done, for example, by attaching the peptide via a suitable linker to the C-terminus of the Knob domain (see, for example, Douglas, JT & Curiel, DT ( 1997 ) Neuromuscular Disorders 7, 284-298) or by exchanging the for the binding region responsible in the knob domain of the fiber protein. Alternatively, the RGD motif in the hypervariable region of the penton base (e.g. Ad12 19AS, Ad2.5 82 AS) can also be replaced by a cell-specific peptide analogous to the experiments carried out by Wickham with the FLAG epitope, with preference natural tropism can also be switched off (see Wickham et al. (1996) J. Virol. 70, 6831). The tropism can be changed, for example, via a shuttle plasmid which codes for the corresponding construct.

Another possible use of the tissue-binding peptides is the use for the diagnosis of pathologically altered tissue, especially in a pathological one altered vessel, inflammation, arteriosclerosis, vessels, the tumor tissue supply, and tissue with proliferating smooth muscle cells of the vascular system, as well the representation of different sections of the vascular system, such as. B. small / large vessels, Veins, organ-specific endothelium.

The diagnosis can either indirectly via z. B. Peptide-Recognizing Antibodies for example in a generally known ELISA test (see e.g. Voller, A. et al. (1976) Bull. World Health Organ., 53, 55-63) or directly by the tissue-binding Peptide is additionally marked. The marking can be a non-radioactive one Marking, e.g. B. about biotin or avidin / streptavidin, or a radioactive Marking (see above).

Radiolabelling is particularly important when examining the whole body Imaging patients, such as B. PET (positron emission tomography) or SPECT (single photon emission computer tomography) for the diagnosis of, for example Inflammation, vascular changes or tumors, advantageous. For this, the patient the labeled peptide is administered intravenously and the distribution of the peptide in the organism analyzed for example by means of "whole-body PET".

Another object of the present invention is therefore a drug and / or Test system for example in the form of a diagnostic agent containing one or more Peptides according to the invention or one or more peptides according to the invention Have been found, or one or more nucleic acids according to the invention, and if appropriate, suitable auxiliaries and / or additives.

Another object of the present invention also relates to a Composition, for example in the form of a transfection or polyfection system containing one or more peptides according to the invention or one or more peptides which were found by the method according to the invention, or one or more  nucleic acids according to the invention and a further substance, preferably a pharmaceutically active compound, as already described in more detail above, and optionally suitable auxiliaries and / or additives.

Suitable auxiliaries and / or additives are, for example, generally known protease or nuclease inhibitors.

The following figures, tables and examples are intended to describe the invention in more detail, without restricting them:

FIGURES AND TABLES

Fig. 1 shows the carotid model from Clowes. After ligating the internal and external carotid arteries, a balloon catheter is inserted through the external carotid artery into the common carotid artery, dilated and the vessel denuded over a distance of 1-1.5 cm. The catheter is then removed again, the external carotid artery ligated and the internal carotid artery reopened.

Fig. 2 shows the selection of the peptides in the ex vivo perfused rat carotid.

Fig. 3 shows the relative binding strength of the peptide P36 to a dilated Rattencarotis, which was taken at the eighth postoperative day, compared to the untreated control carotid.

Figure 4 shows the binding constants of peptide P36 to primary aortic porcine endothelial cells (PAEC), primary porcine smooth muscle cells (VSMC) and monkey kidney cells (COS). The endothelial cells were stimulated by LPS and TNFα (PAEC stimulated).

Tab. 1 shows the DNA sequences and peptide sequences derived from them of the isolated ones Clones.  

ABBREVIATIONS

BSA: bovine serum albumin
DCM: dichloromethane, synonymous: methylene chloride
DIEA: diisoethylamine
EDT: ethanedithiol
HBTU / HOBt: hydroxybenzotriazole
IMC: Medium, contains (1 x M9 salts, 0.2% (casamino acids, 0.5% glucose, 1 mM MgCl ₂ ) 10 x M9 salts: 0.42 M Na ₂ HPO ₄ , 0.22 M KH ₂ PO _4, 85 mM NaCl, 187 mM NH ₄ Cl, pH 7.4)
NMP: N-methyl-2-pyrrolidone
PBS: phosphate buffer
SASRIN: "super acid sensitive resin"
TFA: trifluoroacetic acid

EXAMPLES 1. Prepare a peptide-presenting bacterial bank

A peptide-presenting bacterial library was used, which is commercially available from Invitrogen BV, the Netherlands (FliTrx ^™ Random Peptide Display Library, catalog No. K1125-01; Z. Lu, et al. (1995) Biotech. 13, 366). In this bacterial bank, the bacteria contain an expression cassette which codes for a fusion protein of flagellin and thioredoxin. The flagellin is used to anchor the fusion protein in the bacterial membrane, while the thioredoxin presents a foreign 12 amino acid long peptide in an exposed domain, whose coding 36 base pair long oligonucleotide has been cloned at the appropriate position in the coding sequence of the thioredoxin. The oligonucleotide used for cloning was synthesized at random, so that the bacterial bank represents a total of 10 ⁹ different peptides. The expression cassette is under the control of the tryptophan promoter, so that the expression of the fusion protein can be switched on by the administration of tryptophan at the time of an exponentially growing bacterial culture. Since each bacterium only takes up one plasmid, only one type of peptide with a defined amino acid sequence is expressed from each bacterium and presented on the surface.

2. Growing the bacterial culture

50 ml of IMC medium were mixed with 10 ⁹ bacteria and incubated overnight for 15 hours at +25 ^o C. The cell number was then determined by measuring the OD ₆₀₀ , 10 ¹⁰ cells were incubated in 50 ml IMC medium with 100 μg / ml ampicillin and the expression of the bacterial bank was induced by adding 100 μg / μl tryptophan. The cells were incubated for another 6 hours at 25 ° C before being used for selection.

3. In vitro vascular perfusion model

Restenosis was induced in an established restenosis model in the rat (See, e.g., Clowes, A.W. et al. (1983) Lab. Invest. 49, 327).

As shown in FIG. 1, the common carotid artery in an anesthetized rat was freed from the endothelial cells by introducing a balloon catheter (denudation). This procedure stimulates the smooth vascular muscle cells to proliferate and changes their phenotype from the resting, secretory smooth muscle cell to the proliferating smooth muscle cell.

On the day of the strongest proliferation of smooth muscle cells (10th postoperative day), the vascular segment marked in FIG. 1 was dissected out and perfused in a special ex vivo perfusion system with a saline medium with the addition of glucose and gassing with oxygen. After an equilibration time of 10 min, which was used to wash out any blood cells and blood components, this vascular segment with a total of 10 ¹⁰ peptide-presenting bacteria in oxygen-gassed saline medium with a flow of 0.2 ml / min. Recirculated perfused for 60 min. The non-binding bacteria were detached by treatment with saline medium and the specifically binding bacteria were eluted by perfusion for ten minutes under stringent conditions with 0.1 M glycine pH 2.0. The eluted bacteria were cultivated again and subjected to a further five selection cycles (see FIG. 2) in order to increase the specificity.

4. Test protocol for the perfused rat carotid

Rats weighing approximately 500 g were anesthetized by administering 100 μl / 100 g body weight Dormicum and 100 μl / 100 g body weight Hypnorm. The carotid artery was then dissected cranially to the bifurcation, the internal carotid artery and all branches branching from it were ligated and the carotid communis was denuded starting from the external carotid about 2 cm along the common carotid with a 2F balloon catheter at 1 bar. The internal carotid artery ligature was reopened postoperatively. On the 10th postoperative day, the denuded common carotid artery was dissected and perfused ex vivo with 0.1 ml / min PBS with 0.1 M glucose. Meanwhile, 10 ml of the tryptophan-induced bacterial suspension was centrifuged at 1000 rpm for 3 min, washed in PBS, centrifuged again and the sediment in 6 ml total volume of PBS, 0.1 M glucose, 0.1% BSA, 1 mM CaCl ₂ , 10 mM MgCl ₂ suspended. The carotid was then perfused recirculating with this suspension with gassing of oxygen at 0.2 ml / min for 1 h. The suspension was then washed out by perfusion with 2-3 ml of PBS at 0.2 ml / min before the bound bacteria were eluted by elution with 3 ml of 0.1 M HCl (pH 2.2 adjusted with glycine) and in 50 ml of IMC Medium, 100 µg / ml ampicillin were grown overnight at + 25 ° C.

5. Result of the selection attempts

According to the experiments described above, bacteria were obtained which were on their upper express surface peptides that bind specifically to proliferating smooth muscle cells. The bacteria obtained after six selections were then placed on nutrient medium cultures were plated and starting from 60 individual colonies. Of these cultured bacteria, the plasmid DNA was isolated and sequenced. With this Surprisingly, models were able to bind specific peptides with one of the in vivo The closest method to the situation (organ culture model) can be detected.

After sequencing the DNA of the isolated bacteria, the DNA sequences and peptide sequences derived therefrom were obtained (see Table 1). The sequence of peptide 36 shows homologies to the α ₄ integrin, which is expressed on leukocytes and is responsible for the attachment of the leukocytes to activated vascular cells. The recognition is mediated by the vascular cellular adhesion molecule VCAM-1. The homology is not in the area that is characterized as the binding area of the integrin (RGD-Motif).

6. Attachment studies

The peptide from clone 36 (peptide 36) used for the binding studies was Moc solid phase synthesis on a fully automated peptide synthesizer from Applied Biosystems manufactured on a 0.1 mmol scale. To determine the binding affinities the peptide was radioactively labeled with the desired cell type.

7. Radioactive labeling of a tissue-specific peptide

Since the bacteria used for the selection presented the peptides on their surface in the context of a fusion protein (thioredoxin), the N-terminal amino acid (glycine) was used for the radioactive labeling. For this purpose, ¹⁴ C-gycin was protected with activated N-fluorenyl-methoxycarbonyl-succinimide, purified and coupled covalently N-terminally to the peptide located on the resin and protected in the side chains by means of dicyclohexyl-carbodiimide. The peptide was then cleaved from the resin, purified by HPLC and examined for binding to the target cells in the perfused vessel model and in the cell culture.

8. Test protocol for the synthesis of the radiolabelled ¹⁴ C-Fmoc-glycine

20 µmol of the radioactively labeled ¹⁴ C-glycine (corresponding to 1 mCi; DuPont) in 5 ml of water were combined in one with 38 mg of glycine (0.5 mmol) and 53 mg of Na ₂ CO ₃ (0.5 mmol) 25 ml of the round-bottom flask were added, and 165 mg (0.49 mmol) of N-fluorenylmethoxycarbonyl carbonate (Fmoc carbonate) dissolved in 5 ml of acetone were added dropwise with stirring. During the dropping (60 min), the pH was kept constant by adding 1 M Na ₂ CO ₃ . The mixture was then left to stand at room temperature overnight, the aqueous phase was acidified by adding 2 M HCl and extracted twice with 10 ml of ethyl acetate each time. The combined ethyl acetate phases were washed neutral with water several times, dried with MgSO ₄ and the product was precipitated by adding petroleum ether and filtered off. For purification, the product was dissolved in ethyl acetate and slowly precipitated by adding small amounts of petroleum ether.

Yield: 95 mg (corresponding to 68% of theory), 0.32 mmol
specific activity: 0.0064 MBq / µmol

9. Test protocol for the synthesis of radioactively labeled peptides

The labeled Fmoc-Glycine was used for the synthesis of a total of 0.4 mmol peptide used so that each peptide was 75% labeled.

The appropriate amount of the resin / peptide mixture (0.1 mmol) was weighed out and placed in a glass frit with a stopper and a tap. 25 mg of the ¹⁴ C-Fmoc-glycine (0.075 mmol) were dissolved in 1 ml of NMP, 155 μl of 0.45 M HBTU / HOBt and 750 μl of 2 M DIEA in NMP were added to the resin / peptide mixture and added for one hour incubated at room temperature with gentle shaking. To complete the coupling, 150 mg of Fmoc-glycine (0.5 mmol) were then dissolved in 2 ml of NMP, 2 ml of 0.45 M HBTU / HOBt and 1 ml of 2 M DIEA were added and the mixture was added to the reaction mixture. The mixture was 30 min. incubated at room temperature, the unreacted Fmoc-glycine filtered off and the mixture washed four times with 6-10 ml of NMP each. The protective group was split off by incubation for 30 minutes with 22% piperidine in NMP with stirring. The piperidine was filtered off, and the resin / peptide mixture was then washed four times with NMP and DCM.

10. Test protocol for splitting off the protective groups

To remove the protective groups, the mixture was ice-cooled and with a solution from 0.75 g phenol, 0.25 ml EDT, 0.5 ml thioanisole, 0.5 ml water and 10 ml TFA under Stir added. It was then stirred for two hours at room temperature incubated, the peptide released is filtered off and by adding five times the volume ice-cold diethyl ether. After centrifugation, it was in the sediment located peptide washed several times with dieethyl ether and centrifuged again.

The peptides were chromatographed using an Äkta Purifier 100 over an RPC resource column (3 ml, Pharmacia) with a 20 minute gradient of water, 0.1% TFA to 95% acetonitrile, 0.07% TFA at a flow rate of 2 ml / min cleaned and then lyophilized.
Yields: 65.6 and 70.8% (depending on the peptide)
specific activity: 0.0165 to 0.0182 MBq / µmol

11. Attachment studies

For the binding studies in the perfused vessel model and in cell culture for binding the vessels or the cells were attached to the target cells for different time intervals different concentrations of the peptides to be examined and the amount determined on bound peptide per cm vessel or per cell number.

11.1 Binding to the perfused carotid

It can be seen from FIG. 3 that the peptide 36 binds approximately four times more strongly to the dilated carotid than to the untreated control in the ex vivo perfusion.

11.2 Binding to pig endothelial cells in cell culture

Since the peptide 36 also has homologies to the porcine α ₄ integrin sequence, cell culture studies were also carried out on porcine endothelial cells which were stimulated by treatment with lipopolysaccharides (LPS) and the tumor necrosis factor (TNFα). This stimulation corresponds to an in vivo activation of cells which are located in vascular areas exposed to stress or inflammatory stimuli and thus also in restenotic vascular areas.

The binding constants of the peptide compared to activated and non-activated endothelial cells differ by almost 3 powers of ten ( FIG. 4). The peptide practically does not bind to control cells (COS).

SEQUENCE LOG

Claims (25)

1. Tissue-binding peptide selected from a peptide with the amino acid sequence
GEGRTVVLSF, AWCRGGILGDAM, GNLVDLVVGFDD, RVSPPKKSGGGV, GSSKWGLTXKCG, RGGVRQRSRGRR, GEGRTVVCRS or SQRWTALWQWIG
as well as variants of it. 2. Nucleic acid coding for a tissue-binding peptide according to claim 1. 3. Nucleic acid according to claim 2 selected from a nucleic acid with the nucleotide sequence
as well as variants of it 4. Vector containing a nucleic acid according to claim 2 or 3.   5. A method for producing a tissue-binding peptide according to claim 1, characterized characterized in that the peptide is either chemically synthesized or genetically engineered will be produced. 6. A method for finding a tissue-binding peptide comprising the following steps:

• a) contacting a tissue with one or more peptides, and
• b) isolating one or more tissue-binding peptides.

7. The method according to claim 6, characterized in that the isolated according to step (b) Peptides are repeated one or more times with the same or with a different tissue Be brought in contact. 8. The method according to claim 6 or 7, characterized in that the tissue is diseased tissue. 9. The method according to any one of claims 6-8, characterized in that the or Peptides are contained in a peptide bank. 10. The method according to claim 9, characterized in that the peptide bank peptide-presenting bank, preferably a peptide-presenting bacterial bank. 11. Use of a tissue-binding peptide according to claim 1 or obtained after Method according to one of claims 6-10 for the tissue-specific transfer of Substances. 12. Use of a tissue-binding peptide according to claim 11 for the tissue-specific transfer of pharmaceutically active compounds and / or for the tissue-specific gene transfer.   13. Use according to claim 11 or 12, characterized in that the gene transfer viral and / or non-viral, preferably with the help of liposomes. 14. Use according to any one of claims 11-13, characterized in that one or several tissue-binding peptides via a positively charged domain to one or several nucleic acids are bound. 15. Use of a tissue-binding peptide according to claim 1 or obtained after Method according to one of claims 6-10 for changing the tropism of viruses. 16. Use of a tissue-binding peptide according to claim 15 to change the Tropism of adenoviruses. 17. Use of a tissue-binding peptide according to claim 1 or obtained after Method according to one of claims 6-10 for the diagnosis of pathologically altered Tissue and / or to represent various sections of the vascular system, veins or organ-specific endothelium. 18. Use of a tissue-binding peptide according to claim 17 for illustration different sections of the small and / or large vessels. 19. Use according to claim 17 or 18, characterized in that the pathological altered tissue is selected from a pathologically altered vessel; at Inflammation; Arteriosclerosis; Vessels that supply tumor tissue; Fabric with proliferating smooth muscle cells of the vascular system. 20. Use according to one of claims 17-19, characterized in that the tissue-binding peptide is labeled, preferably radioactively labeled.   21. Medicament and / or diagnostic agent containing one or more tissue-binding Peptides according to claim 1, one or more nucleic acids according to one of the Claims 2-4 or one or more tissue-binding peptides obtained according to the Method according to one of claims 6-10. 22. Medicament and / or diagnostic agent according to claim 21 containing suitable auxiliary and / or additives. 23. A composition containing one or more tissue-binding peptides according to Claim 1, one or more nucleic acids according to any one of claims 2-4 or one or more tissue-binding peptides are obtained by the method according to one of claims 6-10, and another substance. 24. The composition according to claim 23, characterized in that the further Substance is a pharmaceutically active compound. 25. Composition according to one of claims 23 or 24 containing suitable Auxiliaries and / or additives.