EP0774006A2

EP0774006A2 - Chimerical peptide-nucleic acid fragment, process for producing the same and its use for appropriately introducing nucleic acids into cell organelles and cells

Info

Publication number: EP0774006A2
Application number: EP95921691A
Authority: EP
Inventors: Peter Seibel; Andrea Seibel
Original assignee: Individual
Current assignee: Individual
Priority date: 1994-06-16
Filing date: 1995-06-11
Publication date: 1997-05-21
Also published as: US20010008771A1; AU2667995A; DE19520815A1; WO1995034665A3; DE19520815C2; DE4421079C1; WO1995034665A2; US6921816B2

Abstract

In order to appropriately introduce genetic material into cell organelles, the corresponding nucleic acid is bound to a cell-specific, compartment-specific or membrane-specific peptide. The nucleic acid is directed by the peptide to the aimed compartment through the natural protein transport paths and is transported through the membrane (the membrane system). For nucleic acids to unfold their replication and transcription activity in cells and cell organelles and to be projected into the cell through the protein import route, they must on the one hand have all required genetic elements and on the other hand must not exceed the pore import capacity. These two requirements caused the development of a linear-cyclic plasmid system which contains all required genetic elements, may be absorbed through the protein import route thanks to its pseudo-linear structure (cyclic DNA ends) and may finally build cyclic replication intermediates. By combining in any desired manner nucleic acids and compartment-specific protein sequences, this process allows nucleic acids to be appropriately introduced into cells and cell organelles. This system is thus suitable both for mutagenesis and in the molecular therapy of genetic diseases.

Description

Chimeric peptide-nucleic acid fragment, process for its preparation, and its use for the targeted introduction of nucleic acid into cell organelles and cells

This invention relates to a chimeric peptide nucleic acid fragment, the process for its preparation and its use for the targeted introduction of nucleic acids into cell organelles and cells.

It is known that cellular membrane systems are largely impermeable to nucleic acids. However, cell membranes can be overcome very efficiently through physical processes (transformation) and biological processes (infection). The transformation, that is, the immediate absorption of the naked nucleic acid by the cell is preceded by a treatment of the cells. Different methods for generating these 'competent cells' are available. Most methods are based on the observations by Mandel and Higa (M. Mandel et al. (1970), "Calcium-dependent bacteriophage DNA infection", J. Mol. Biol. 53: 159-162), who were able to show for the first time that the yields in the absorption of lambda DNA by bacteria in the presence of calcium chloride can be increased significantly. For the first time, this method was also developed by Cohen et al. (SN Cohen et al. (1972), "Nonchromosomal antibiotic resistance in bacteria: genetic transformation of Escherichia coli by R-factor DNA", Proc. Natl. Acad. Sci. USA 69: 2110-2114) were successfully used for plasmid DNA and have been improved by many modifications (M. Dagert et al. (1979), "Prolonged ineubation in calcium chloride improves the competence of Escherichia coli cells", Gene 6: 23-28). Another transformation method is based on the observation that high-frequency alternating current fields can break open cell membranes (electroporation). This technique can be used to introduce bare DNA not only into prokaryotic cells but also into eukaryotic cell systems (K. Shigekawa et al. (1988), "Electroporation of eukaryotes and prokaryotes: a general approach to the introduction of macromolecules into cells" , Biotechniques 6: 742-751). Two very gentle methods of introducing DNA into eukaryotic cells have been described by Capecchi (MR Capecchi, (1980), "High effϊciency transformation by direct microinjection of DNA into cultured mammalian cells ", Cell 22: 479-488) and Klein et al. (TM Klein et al. (1987)," High velocity microprojectiles for delivering nucleic acids into living cells ", Nature 327: 70-73): They are based on the direct injection of the DNA into the individual cell (microinjection), as well as on the bombardment of a cell population with microprojectiles made of tungsten, to the surface of which the nucleic acid in question was bound ('shotgun') The biological infection methods have proven their worth in physical transformation of cells, in particular the highly efficient viral introduction of nucleic acids into cells (KL Berkner, (1988), "Development of adenovirus vectors for the expression of heterologous genes", Biotechniques 6: 616-629; LK Miller, (1989) "Insect baculoviruses: powerful gene expression vectors", Bioessays JJ: 91-95; B. Moss et al. (1990), "Product review. New mammalian expression vectors ", Nature 348: 91-92) and lipofection-mediated lipofection (RJ Mannino et al. (1988) Liposome Mediated Gene Transfer, Biotechniques 6: 682-690; PL Feigner et al. (1987), "Lipofection: a highly efficient, lipid-mediated DNA transfection procedure", Proc. Natl. Acad. Be. USA 84: 7413-7417). All methods described so far deal with the overcoming of the prokaryotic or eukaryotic plasma membrane by naked or packaged nucleic acids. While the site of action has already been reached by introducing the nucleic acid into the prokaryotic cell, further biochemical processes take place in the compartmented eukaryotic cell that support the penetration of the nucleic acid into the cell nucleus under certain conditions (e.g. viral infection pathway in HTV). Analogous infection processes in which exogenous nucleic acids are actively introduced into other cell organelles (e.g. in mitochondria) have not been described so far.

In addition to the introduction of the nucleic acid into the cell or the cell organelle, the transcription and above all the replication of the introduced nucleic acid plays a decisive role. In this context, it is known that DNA molecules can have a special property that allows doubling in a cell under certain conditions. A special structural element contributes to this Point of origin of DNA replication at (ori, origin), the presence of which gives the ability to replicate DNA (KJ Marians (1992), "Prokaryotic DNA replication", Annu. Rev. Biochem. 6] _: 673-719; ML DePamphilis (1993), "Eukaryotic DNA replication: anatomy of an origin", Annu. Rev. Biochem. 62: 29-63; H. Echols and MF Goodman (1991), "Fidelity mechanisms in DNA replication", Annu. Rev. Biochem 60: 477-511). The strictly controlled process of DNA replication in E. coli begins, for example, with the binding of a protein (K. Geider and H. Hoffmann Berling (1981), "Proteins Controlling the helical structure of DNA", Annu. Rev. Biochem. 50: 233-260) to the highly specific initiation site and thus defines the starting point for a specific RNA polymerase (primase). It synthesizes a short strand of RNA (-10 nucleotides, 'primer') that is complementary to one of the DNA template strands. The 3 'hydroxyl group of the terminal ribonucleotide of this RNA chain serves as a' primer 'for the synthesis of new DNA by a DNA polymerase. DNA-unwinding proteins de-spiral the DNA double helix (JC Wang (1985), "DNA topoisomerases", Annu. Rev. Biochem. 54: 665-697). The separated single strands are stabilized in their conformation by DNA-binding proteins (JW Chase and KR Williams (1986), "Single-stranded DNA binding proteins required for DNA replication", Annu. Rev. Biochem. 55: 103-136) to enable a trouble-free function of the DNA polymerases (TS Wang (1991), "Eukaryotic DNA polymerases", Annu. Rev. Biochem. 60: 513-552). A multienzyme complex, the holoenzyme of DNA polymerase III, synthesizes most of the new DNA. The RNA part of the chimeric RNA-DNA molecule is then cleaved from the DNA polymerase III. Removing the RNA from the newly created DNA chains creates gaps between the DNA fragments. These gaps are filled by DNA polymerase I, which can rebuild DNA from a single-stranded template. While one of the two newly synthesized DNA strands is continuously built up during replication (5'-3 'direction, lead strand), Ogawa and Okazaki observed that a large part of the newly synthesized counter strand (3'-5' direction, delay strand) short DNA fragments (T. Ogawa and T. Okazaki (1980), "Discontinuous DNA replication", Annu. Rev. Biochem. 49: 421-457). So-called primases initiate here by the synthesis of several RNA primers the beginning of the DNA synthesis of the opposite strand. As replication progresses, these fragments are freed from their RNA primers, the gaps are closed and covalently linked to long daughter strands by the DNA ligase. After the end of the replication cycle, two chromosomes are formed.

In contrast, the DNA replication of many plasmids is controlled via an origin of replication, which does not synthesize the delay strand (3'-5 'direction) and can initiate the synthesis of two continuous DNA strands bidirectionally (each in 5'- 3 'direction along the two matrices). The prerequisite for complete DNA replication is the cyclic form of the nucleic acid. It ensures that the DNA polymerases return to the starting point at the end of the new synthesis of the complementary DNA strands, where ligases now ensure the covalent linkage of the ends of the two newly synthesized daughter strands.

An interesting form of linear-cyclic nucleic acids is known from smallpox viruses: so-called 'hairpin loops' at the ends of their linear genomes mean that they have a cyclic molecular structure while maintaining a predominantly linear conformation (DN Black et al. (1986), "Genomic relationship between capripoxviruses ", Virus Res. 5: 277-292; JJ Esposito and JC Knight (1985)," Orthopoxvirus DNA: a comparison of restriction profiles and maps ", Virology 143: 230-251). Covalently closed 'hairpin' nucleic acids have not only been found in smallpox viruses, but also for the ribosomal RNA from Tetrahymena (EH Blackburn and JG Gall (1978), "A tandemly repeated sequence at the termini of the extrachromosomal ribosomal RNA genes in Tetrahymena", J. Mol. Biol. 120: 33-53), and the genomes of the parvoviruses (SE Straus et al. (1976), "Concatemers of altemating plus and minus Strands are intermediates in adenovirus-associated virus DNA synthesis", Proc. Natl Acad. Sci. USA 73: 742-746; P. Tattersall and DC Ward (1976), "Rolling hairpin model for replication of parvovirus and linear chromosomal DNA ", Nature 263: 106-109).

With the plasmids or nucleic acid constructs known to date, however, it is not possible to introduce nucleic acids into cells or cell organelles in a targeted manner via the protein import route. However, this is, for example, a prerequisite for treating changes in the mitochondrial genome of patients with neuromuscular and neurodegenerative diseases at the genetic level, or for being able to carry out targeted mutagenesis in mitochondria or other cell organelles.

The object of the present invention was therefore to develop a Konstnikt based on nucleic acid, which allows the targeted introduction of nucleic acids into cells and compartments of eukaryotic cells. In addition, a method is to be provided for how this constituent can get into cell compartments or cells. Furthermore, the nucleic acid introduced should be such that it can also be taken up as a replicable nucleic acid via cellular protein import routes. In addition, properties should be present which lead to controlled transcription and / or replication in cells or in defined target compartments of a cell. The method is to be used for the therapy of genetic diseases (changes in the mitochondrial genome) and for targeted mutagenesis in eukaryotic and prokaryotic cells. The invention is intended to meet the following requirements:

- universal applicability

- cell-, compartment- and membrane-specific infiltration behavior

- high effectiveness - low immunogenicity

- minimizing the risk of infection

- The introduced nucleic acid (plasmid molecule) should be replicable - The introduced nucleic acid (plasmid molecule) should be transcribable

- The introduced nucleic acid (plasmid molecule) should be stable against nucleases

- The structure of the introduced nucleic acid (plasmid molecule) should be universally usable

This object is achieved by the features of claims 1), 25), 54), 56), 58), 60) and 61). Advantageous refinements result from the subclaims.

In order to be able to transport a protein within a cell from the place of origin to another compartment or another cell organelle (e.g. the site of action), the protein is usually synthesized as a preprotein (R. Zimmermann et al. (1983), " Biosynthesis and assembly of nuclear-coded mitochondrial membrane proteins in Neurospora crassa ", Methods Enzymol. 97: 275-286). In addition to the matured amino acid sequence, the preprotein has a so-called signal sequence. This signal sequence is specific for the target compartment and ensures that the preprotein can be recognized by surface receptors. The natural barrier 'membrane' is then overcome by translocating the preprotein through an active (multiple transport proteins' are involved in this process) or a passive process (direct passage without the involvement of other proteins) through the membrane. The signal sequence is then usually separated off at the site of action by a specific peptidase, provided that it is not itself a component of the matured protein. The matured protein can now develop its enzymatic activity.

It was recognized by the inventors that this mechanism can be used to target nucleic acids through membranes. The nucleic acid is not subject to any restriction, ie any desired or known nucleic acid can be used. For this purpose, a cell-, compartment- or membrane-specific signal sequence is coupled to the desired nucleic acid, whereby a chimeric peptide-nucleic acid fragment arises. In this context, it is known that the coupling between a nucleic acid and a peptide via the ε-Maleimidocaproic acid-N-hydroxysuccinimdester modified α-amino group of a synthetic KDEL peptide can take place (K. Arar et al, (1993), "Synthesis of oligonucleotide-peptide conjugates containing a KDEL signal sequence", Tetrahedron Lett. 34: 8087- 8090). However, this linking strategy is completely useless for the introduction of nucleic acids in cell organelles and cells, since translocation is to take place analogously to natural protein transport. Such a transport is not to be expected when the α-amino group of a synthetic peptide is blocked by a nucleic acid. Coupling via a carboxy-terminal amino acid was therefore chosen by the inventors. This ensures a 'linear' link on the one hand, and on the other hand the free amino-terminal end of the signal peptide is thus available for the essential steps of the import reaction.

In order to be able to use the described transport mechanism for the introduction of replicative and transcription-active nucleic acids, the nucleic acid is preferably integrated into an existing genome via homologous recombination, or is itself the carrier of the genetic elements, which ensures autonomous initiation of replication and transcription. Only the last variant fulfills the criterion of universal applicability, since recombination into an existing cellular genome is only possible under certain conditions and in selected cells.

One possibility here is the use of cyclic DNA, since the DNA polymerases return to the starting point at the end of the new synthesis of the daughter strands and thus ensure complete DNA replication. While the use of a double-stranded cyclic plasmid fulfills all physical criteria for a successful replication in every target compartment of the cell, this physical DNA form contrasts with the size capacity of the import pore, which is crucial for the targeted translocation: the compact diameter of a superhelical plasmid is comparable to globular proteins and makes translocation through a membrane system via the protein import route seem hopeless. The solution is to use it linear-cyclic DNA molecules that have modified (cyclized) ends, but still only have the diameter of linear DNA molecules. On the one hand, these do not represent an obstacle to the size of the import pore, on the other hand, these linear-cyclic DNA molecules have all the physical requirements to be able to form replicative and transcription-active plasmids in the mitochondria.

For the construction of the chimeric peptide-nucleic acid fragment according to the invention, as well as for the construction of a replicative and transcription-active nucleic acid portion (plasmid), one preferably needs:

- Signal peptide or signal sequence (cell, compartment or membrane specific) - Coupling agent

Nucleic acid (oligonucleotide), which can preferably include the following further information:

- Information on the initiation and regulation of transcription and replication

- Information for the termination of transcription and replication - Multiple cloning site for an additional insert (to be expressed)

nucleic acid

- Modification options so that 'hairpin loops' can be added (cyclization of the ends), which allow a link to the signal peptide

The selection of the signal sequence depends on which membrane or membrane system is to be overcome and which target compartment of the cell (cell nucleus, mitochondrion, chloroplast) or cell organelle is to be reached. Proteins that are to be inserted into one of the four mitochondrial compartments (outer mitochondrial membrane, intermembrane space, inner mitochondrial membrane, matrix space), for example, have compartment-specific signal sequences. Signal sequences which contain a cell-, compartment- or membrane-specific recognition signal are generally selected for the introduction of nucleic acids and thereby directing the attached nucleic acid to its place of action (eg inner side of the inner mitochondrial membrane or matrix space). There is a choice of signal sequences that can transport proteins in the presence or absence of a membrane potential. Signal sequences that work independently of membrane potentials are preferred for the nucleic acid introduction, for example the signal sequence of ornithine transcarbamylase (OTC) for transport into the matrix space of the mitochondria (AL Horwich et al. (1983), "Molecular cloning of the cDNA coding for rat ornithine transcarbamoylase ", Proc. Natl. Acad. Sci. USA 80: 4258-4262; JP Kraus et al. (1985)," A cDNA clone for the precursor of rat mitochondrial ornithine transcarbamylase: comparison of rat and human leader sequences and conservation of catalytic sites ", Nucleic. Acids. Res. L3: 943-952). Basically, the pure signal sequence is sufficient for the transport into the target compartment. However, signal sequences are preferably selected which additionally have a cell- or compartment-specific peptide cleavage site. In the most favorable case, this 'cleavage point' lies within the signal sequence, but can also be added to it by additional amino acids in order to ensure that the signal sequence is split off after reaching the target compartment (e.g. the signal sequence of human OTC can be extended by another ten amino acids of the matured OTC become). This ensures that the nucleic acid in the target compartment can be separated from the signal peptide and that the action of the nucleic acid is fully developed. The selected signal sequence is produced biologically (purification of natural signal sequences or cloning and expression of the signal sequence in a eukaryotic or prokaryotic expression system), but preferably by a chemical-synthetic route.

In order to ensure a linear chemical link between nucleic acid and signal peptide, the signal peptide is coupled via a coupling agent, which is generally via amino acids, preferably via amino acids with reactive side groups, preferably via a single cysteine or lysine at the carboxy-terminal end of the signal peptide connected is. A bifunctional crosslinker, preferably a heterobifunctional crosslinker, is used as the coupling reagent Cysteins as a coupling point on the signal peptide has a second reactive group, preferably an amino-reactive group, in addition to a thiol-reactive group (e.g. m-maleimidobenzoyl-N-hydroxy-succinimide ester, MBS and its derivatives).

The nucleic acid also has a coupling site that should be compatible with the selected crosslinker. When using MBS, the oligonucleotide should have an amino or thiol function. The coupling group of the nucleic acid can be introduced via the chemical synthesis of the oligonucleotide and is generally located at the 5 'end, at the 3' end, but preferably directly on a modified base, e.g. B. as 5'-amino linker (TFA-amino linker Amidite ^R , 1, 6- (N-trifluoroacetylamino) -hexyl-ß-cyanoethyl-N, N-diisopropyl phosphoramidite, Pharmacia) or as 5'-thiol linker (THIOL-C6 phosphoramidite *, MWG Biotech) on a free 5'-hydroxy / phosphate group, as a 3 'amino linker (3' amino modifier C7-CPG synthesis columns ^R , MWG Biotech) on a free 3 'hydroxy / phosphate group, but preferably as an arnino-modified one Base analog, preferably amino-modified deoxyuridine (Amino-Modifier-dT ^R , 5 '-Dimemoxy-trityl-5- [N- (trifluoroacetylaminohexyl) -3-acrylimido] -2'-deoxy uridine, 3' - [(2-cyanoethyl ) - (N, N-diisopropyl)] - phosphoramidite, Glen Research) within the sequence. The reactive group compatible with the crosslinker used is at least distanced from the 5 'or 3' end of the oligonucleotide or the modified base by a C2 spacer unit, but preferably by a C6 spacer unit. The nucleic acid (oligonucleotide) with a reactive coupling group comprises at least two nucleotides.

In order to increase the stability of the nucleic acid (oligonucleotide) against cellular and extracellular nucleases, the chemically synthesized nucleic acids can be protected by a sulfurizing reagent (Beaucage reagent *, MWG-Biotech). In chemical synthesis, the phosphorus diester bonds of the nucleic acid are converted into phosphorothioate bonds. This oligonucleotide can then be used for the enzymatic amplification of nucleic acids, extended by further linking reactions with other nucleic acids or used directly. In order to use the chimeric peptide-nucleic acid fragment directly, the nucleic acid (oligonucleotide) should have a hybridization-capable secondary structure, preferably without internal homologies, in order to be able to form a linear single-strand structure. This ensures that the nucleic acid (oligonucleotide) of the chimeric peptide-nucleic acid fragment can develop a biochemical / therapeutic effect without further nucleic acid coupling.

For coupling with the signal sequence, however, nucleic acids (oligonucleotides) are preferably used which have two further properties:

1. The sequence is preferably partially palindromic, with a smooth 5 '-3' end ('blunt end'), overhanging 3 'end (' sticky end '), but preferably with overhanging, phosphorylated 5' end (' sticky end '), very preferably with an overhanging 5' end comprising 4 nucleotides and which has no self-homology (palindromic sequence). This allows a stable, monomeric secondary structure ('hairpin loop') to be formed. The overhanging 5 'end serves to couple defined nucleic acids, antisense oligonucleotides, but preferably transcribable and replicable genes.

2. The oligonucleotide carries at the apex of the 'loop' a modified base which carries a group reactive towards the crosslinker, preferably an amino-modified 2'-deoxythymidine. The amino function of this modified base enables the coupling reaction between MBS and oligonucleotide.

The chimeric peptide nucleic acid fragment is suitable for the targeted introduction of nucleic acids into cells and cell organelles (e.g. cell nucleus, chloroplast), in particular for the introduction of ribonucleic acids (mRNAs, 'antisense' oligonucleotides) and deoxyribonucleic acids (complete genes, 'antisense 'Oligonucleotides). It is particularly suitable for introducing transcribable and processable genes into Mitochondria, but especially for introducing replicative, transcription-active and processable linear-cyclic nucleic acids (plasmids).

In a preferred embodiment, a transcribable gene is coupled to the nucleic acid containing the reactive coupling site or to the chimeric peptide-nucleic acid fragment. This is preferably done by amplifying a gene, preferably a cloned gene, which consists of a mitochondrial promoter, preferably the promoter of the light DNA strand (O _L , nt 490 - nt 369) and the gene to be expressed in a processable form, preferably a mitochondrial Gen, preferably a mitochondrial transfer RNA, preferably the mitochondrial tRNA ^{Leu (UUR)} (nt 3204 - nt 3345), exists (S. Anderson et al. (1981), "Sequence and organization of the human mitochondrial genome", Nature 290: 457-465). After the enzymatic amplification of the gene, the coupling to the nucleic acid containing the reactive coupling site or to the chimeric peptide-nucleic acid fragment can be coupled via a 'blunt-end' ligation, but preferably a 'sticky-end' ligation . For this purpose, the nucleic acid to be coupled has at least one end which can be coupled, which preferably consists of a 5 'overhang comprising 4 nucleotides and has no self-homology (palindromic sequence). If both ends are to be coupled with 'hairpin loops', a nucleic acid is preferably selected which has different 5' overhangs, which preferably comprise 4 nucleotides and have no self-homology. Nucleic acids whose 5 'ends also have no homology with one another are very preferably used. Two different 'hairpin loops' are then preferably used for the modification of the ends (cyclization), one specific (complementary) for the 'left' plasmid end, the other specific for the 'right' plasmid end of the nucleic acid. In order to increase the stability of the nucleic acid against cellular and extracellular nucleases, the phosphorodiester bonds of the nucleic acid can be substituted by phosphorothioate bonds and thus protected if modified phosphorothioate nucleotides are already used in the enzymatic amplification. A method which has the following steps is suitable for producing the chimeric peptide-nucleic acid fragment:

(a) Reaction of a nucleic acid (oligonucleotide) containing a functional coupling group with a coupling agent. (b) Implementation of the construct from (a) with amino acids on

Carboxy-terminal end of a peptide containing a signal sequence, except a KDEL signal sequence, and

(c) optionally extending the chimeric peptide nucleic acid fragment from (b) by further DNA or RNA fragments.

In another preferred embodiment, the chimeric peptide nucleic acid fragment can be produced by the following steps:

(a) Optional extension of the nucleic acid, containing a functional coupling group, by further DNA or RNA fragments.

(b) Implementation of the nucleic acid with a functional coupling group or the extended nucleic acid from (a) with a coupling agent.

(c) Reaction of the construct from (b) with amino acids at the carboxy-terminal end of a peptide, comprising a signal sequence, with the exception of a KDEL signal sequence.

In another embodiment, which is a linear cyclic nucleic acid in the form of a plasmid, the selection of the nucleic acid depends on which genetic information is to be expressed in which cell and in which target compartment of the cell. In this context, nucleic acids that are to be transcribed must have a suitable promoter. If, for example, a gene is to be expressed in the mitochondrial matrix, a mitochondrial promoter can be selected, preferably the promoter of the light mtDNA strand. The control the transcription takes place in other cell compartments (eg cell nucleus, chloroplast) by compartment-specific promoters.

The regulation of the transcription is usually carried out by so-called transcription regulation sequences, preferably mitochondrial transcription regulation sequences. In general, these sequences comprise at least binding sites for factors that initiate transcription (transcription initiation factor) and the binding site for the RNA synthesizer. If a transcription is to be initiated in the mitochondria, binding sequences of the mitochondrial transcription factors and the RNA polymerase, in particular the mitochondrial transcription factor 1 and the mitochondrial RNA polymerase, are suitable here. In other cell compartments (e.g. nucleus, chloroplast) the transcription can be controlled by compartment-specific transcription regulation sequences.

In order to be able to regulate the transcription, the plasmid has transcription regulation sequences, which are preferably added in the 3 'direction to the transcription initiation site (promoter). If, for example, the transcription of a mitochondrial transformation plasmid is to be regulated, the control elements are suitable for the H and L strand transcription of the mitochondrial genome, but preferably the so-called 'conserved sequence blocks ¹ , which terminate the transcription of the L strand and at the same time the Enable transition to DNA replication. In order to induce the exclusive transcription of the desired gene (possibly the desired genes in a polycistronic transcription), the transcription is interrupted at a suitable point behind the 3 'end of the expression-capable gene / genes. This is achieved by inserting a suitable transcription termination site, preferably arranged in the 3 'direction to the promoter. The binding sequence for a bidirectionally acting transcription termination factor is particularly suitable for regulated expression. For the transcription termination in the mitochondria, a binding motif is preferred here mitochondrial transcription termination factor selected. At the same time, the use of a transcription termination factor binding sequence suppresses the formation of 'antisense RNA' of the head-to-head linked dimeric plasmids.

The selection of transformed cells can be checked by expressing a reporter gene. Particularly suitable reporter or selection genes are expression-capable genes, the expression of which leads to a macroscopic change in the phenotype. There is a choice of genes that trigger resistance to antibiotics, for example. The resistance genes for oligomycin (OLI) or chloramphenicol (CAP) are particularly suitable for use in a mitochondrial transformation system. The mitochondrial chloramphenicol resistance gene appears to be a particularly suitable selection gene in this context, since CAP-sensitive cell lines already change their phenotype with a proportion of approximately 10% of the 16 S rRNA ^{CAP +} gene.

Replication of the nucleic acid can be achieved by an initiation site for DNA replication (origin of replication). The chimeric

A peptide-nucleic acid fragment in the form of a plasmid must therefore have at least one origin of replication. The orientation of the origin of replication can be arranged independently of the expression-capable gene (genes), but the origin of replication is preferably arranged in the 3 'direction to the promoter. A suitable origin of replication for a mitochondrial transformation plasmid would be a mitochondrial origin of replication. The origin of the replication of the heavy mtDNA strand, which preferably has at least one conserved sequence block, is particularly suitable here. Replication can be controlled via so-called regulatory sequences for replication. For this purpose, the plasmid must have at least one such sequence motif, which is preferably arranged in the 3 'direction to the promoter and to the origin of replication. If the replication in the mitochondria is to be regulated, a mitochondrial replication regulation sequence is particularly suitable here. A motif is preferably used that at least contains one of the 'termination associated sequences'. In other cell compartments (eg cell nucleus, chloroplast), replication is preferably initiated via at least one compartment-specific origin of replication and controlled via compartment-specific replication regulation sequences.

In order to allow different genes to be cloned into the plasmid molecule, the plasmid nucleic acid must also have a suitable cloning module (multiple cloning site) which has as different as possible recognition sequences for restriction endonucleases. Rare recognition sequences that do not occur elsewhere on the plasmid are particularly suitable here. The cloning module can be installed at any point on the transformation plasmid. If the area of the cloning site is to be integrated into the transcription of the selection gene, the insertion of the multiple cloning site in the 3 'direction to the promoter and in the 5' direction to the transcription termination site is suitable. The integration of the multiple cloning site in the 5 'direction to the selection gene is particularly suitable since, when using the selection system, the region of the multiple cloning site is transcribed at the same time.

In order to allow autonomous replication in each target compartment of a cell when using a nucleic acid, it must be ensured that the DNA replication enzymes return to the starting point of synthesis after the synthesis of the daughter strand in order to covalently link the 3'- to the 5'- Ensure end of the newly synthesized daughter strand by appropriate enzymes. A linear nucleic acid plasmid that can be converted into a cyclic nucleic acid is suitable for this. The plasmid ends can be cyclized using so-called ligation-capable (phosphorylated) ends of the nucleic acid. The use of a 'blunt-end' nucleic acid or a nucleic acid with 3 'overhanging ends, but preferably a nucleic acid with 5' overhanging ends, is particularly suitable for this purpose. The overhanging ends should each comprise at least one nucleotide. However, 5 'overhanging ends are preferably used are formed from four nucleotides. These preferably have no self-homology (palindromic sequence) and are also preferably not complementary to one another in order to suppress dimer formation in a later nucleic acid linkage.

The cyclization of the prepared plasmid ends is mediated by synthetic oligonucleotides. These have a partial self-homology (partial palindromic sequence) and are therefore capable of forming so-called 'hai in-loop' structures. The partially palindromic sequence leads to the formation of a stable, preferably monomeric, secondary structure ('hairpin loop'), with a smooth 5'-3 'end (blunt end), an overhanging 3' end ('sticky end') '), but preferably with an overhanging 5' end. These oligonucleotides are particularly suitable if they have a 5 'phosphorylated end. The linear plasmid DNA can now be converted into a linear-cyclic system using synthetic oligonucleotides with a hairpin-loop structure. The ends of the two oligonucleotides are preferably complementary to one end of the prepared plasmid nucleic acid. For this purpose, two different 'hairpin loops' are preferably used, one specific (complementary) for the 'left' plasmid end, one specific (complementary) for the 'right' plasmid end in order to suppress dimer formation. At least one of the two hairpin-loop oligonucleotides can have at least one modified nucleotide. This ensures the coupling site to a signal peptide so that the nucleic acid transport can be mediated via the protein import route. Ideally, this coupling site (modified nucleotide) is located at one of the unpaired positions of the 'loop'. A chemically reactive group, in particular an amino or thiol diffusion, is suitable as the coupling point.

In order to prepare the ends of the transformation plasmid for the modification (cyclization), it must be ensured that the ends of the plasmid are complementary to the ends of the oligonucleotides ('hairpin loops'). This is achieved on the one hand by amplifying the plasmid DNA with suitable oligonucleotides that have at least have a restriction endonuclease recognition sequence. Detection sequences for restriction endonucleases which do not occur repeatedly in the plasmid sequence are suitable here. The use of recognition sequences for restriction endonucleases which generate sticky ends, in particular those which generate 5 'overhanging ends, preferably outside the own recognition sequence, is particularly suitable. In this context, the recognition sequence for the restriction endonuclease Bsa I (GGTCTCN, N ₅ ) is particularly suitable. On the other hand, the use of a cloned nucleic acid which already has the recognition sequences for a restriction endonuclease, preferably Bsa I, is suitable here. This means that the enzymatic amplification can be dispensed with and the nucleic acid obtained via a plasmid preparation / restriction enzyme treatment can be used directly. The cloned nucleic acid preferably already contains the recognition sequence for the restriction endonuclease Bsa I at both ends.

Different methods are available for cleaning the transformation plasmid. The primary goal here is to separate the cyclized plasmid molecules from unreacted starting materials. The use of DNA-degrading enzymes has proven to be suitable in this connection, and the use of enzymes which have a 5'-3 'or 3'-5'-exonuclease activity is particularly recommended here. In particular, the use of exonuclease III leads to the complete hydrolysis of unreacted starting materials, while the cyclized plasmid DNA remains intact (no free 5 'or 3' ends). The reaction products can be purified either by electrophoretic or chromatographic processes, but also by precipitation. There are different cleaning methods to choose from. On the one hand, the cyclized nucleic acid conjugated with the coupling agent and the signal peptide can be treated with an exonuclease, preferably exonuclease III, and then purified by chromatographic, electrophoretic purification or precipitation. On the other hand, the cyclized plasmid DNA can be treated with an exonuclease, preferably exonuclease III, purified, and then with the coupling agent and the Conjugate signal peptide and be purified by chromatographic, electrophoretic purification or precipitation.

By using modified oligonucleotides, the linkage with a signal peptide can be realized, which directs the transformation plasmid into the desired cell compartment in vivo. For this purpose, either the transformation plasmid can first be reacted with the modified oligonucleotide (ligation) and then conjugation with the coupling agent and the signal peptide, or the modified oligonucleotide is first conjugated with the coupling agent and the signal peptide and then for the cyclization of the transformation plasmid end used (ligation).

The cell membrane can be overcome by the transformation system (cellular transformation) using various methods. The 'particle gun' system or microinjection are suitable, but electroporation and lipotransfection are preferred. All methods ensure the introduction of the linear cyclic peptide nucleic acid plasmid into the cytosol of the cell, from where the plasmid is directed to the target site (target compartment) by the conjugated signal peptide.

Compared to the transformation and infection methods of the prior art mentioned in the introduction to the description, this method offers for the first time the possibility of introducing nucleic acids into cells and cell organelles in a targeted manner. The target compartment to be reached (cytosol, nucleus, mitochondrium, chloroplast, etc.) can be determined by the choice of the signal sequence. In addition to the compartment- and cell-specific infiltration behavior, this process is characterized by its universal applicability. Both prokaryotic and eukaryotic cells and cell systems can be treated with the translocation vector. Since a natural membrane transport system is used for targeted infiltration, there is no need to treat the cells or cell organelles with membrane-permeabilizing agents (eg calcium chloride method, see above). When using a replicative and transcription-active nucleic acid, the plasmid only unfolds its full size after the completion of the first replication cycle: as a true cyclic plasmid (artificial chromosome) it now has double genetic information (head-to-head linked plasmid dimers). In particular with regard to the use of this system for somatic gene therapy, this behavior is deliberately induced and is of crucial importance since the genes to be expressed have to compete with the defective genes in the cells. In addition to this highest possible effectiveness, the system impresses in that it does not have to be integrated into a genome via a recombination process, such as retroviral systems, in order to become replicative. As a result, uncontrollable side effects (undesired recombination) are suppressed as far as possible in advance. The use of this plasmid system can therefore be promised without any major security risk.

The present invention is explained in particular by the figures, which show:

1: signal peptide of the ornithine transcarbamylase of the rat, and a DNA sequence suitable for the introduction. Above: signal peptide of rat ornithine transcarbamylase (32 amino acids), extended by ten N-terminal amino acids of the matured protein and an additional cysteine as a coupling site. The peptide sequence is shown in the international single-letter code; Middle: a partially palindromic DNA sequence of 39 nucleotides suitable for the introduction with an amino-modified T at nucleotide position 22; Bottom: Pronounced secondary structure of the oligonucleotide with an overhanging 5 'end and a modified nucleotide at the apex of the' loop '.

Fig. 2: Structure of the amino-modified 2'-deoxythymidine. R: Nucleic acid residues. 3: Schematic representation of the chimeric peptide-nucleic acid fragment, consisting of amino-modified oligonucleotide (39 nucleotides) with a pronounced hairpin loop, crosslinker and signal peptide. CL: Crosslinker.

4: Electrophoretic separation of the coupling product from amino-modified oligonucleotide (39 nucleotides), m-maleimidobenzoyl-N-hydroxy-succinimide ester (MBS) and signal peptide of rat ornithine transcarbamylase (42 amino acids, extended by one cysteine at the C-terminus).

5a: Flow diagram of the peptide-DNA fusion, cloning, amplification and coupling of the transcribable and processable mitochondrial tRNA gene to be introduced (S. Anderson et al. (1981), "Sequence and organization of the human mitochondrial genome", Nature 290 : 457-465). CL:

Crosslinker (MBS); MCS: Multiple cloning site of pBluescript *

(Stratagene); mtTF: binding site of the mitochondrial transcription factor; RNA-Pol: binding site of the mitochondrial RNA polymerase; tRNA

Leucine: Mitochondrial transfer RNA gene for leucine (UUR); Sac II, Apa

I, Eco RI: interfaces for restriction endonucleases; The cloned mitochondrial sequences were numbered in accordance with the published sequence of the human mitochondrial genome (S. Anderson et al. (1981), "Sequence and organization of the human mitochondrial genome", Nature 290: 457-465).

5b: Sequence of the cloned tRNA ^{Leu (uυR)} gene.

6a / b: Representation of the ³² P radiation of the DNA and of the enzyme activities for adenylate kinase, cytochrome c oxidase and malate dehydrogenase (y-axes) in 11 fractions (x-axes) of a mitochondrial sucrose gradient density centrifugation . The proportion of the respective radiation / Enzyme activity expressed as a percentage of the total radiation / enzyme activity applied to the gradient. ADK: adenylate kinase; COX: cytochrome c oxidase; MDH: malate dehydrogenase.

7a / b: Representation of the ³² P radiation of the DNA and the enzyme activities for adenylate kinase, cytochrome c oxidase and malate dehydrogenase (y-axes) in 11 fractions (x-axes) of a mitoplast-sucrose gradient density centrifugation . The proportion of the respective radiation / enzyme activity is shown, expressed as a percentage of the total radiation / enzyme activity that was applied to the gradient. ADK: adenylate kinase; COX: cytochrome c oxidase; MDH: malate dehydrogenase.

8: Cloning of the nucleic acid portion of the peptide-nucleic acid plasmid in pBluescript (plasmid 1). Using the two oligonucleotides (primers 1 and 2), the gene section from nucleotide 15903 to nucleotide 677 was enzymatically amplified from mitochondrial HeLa DNA (includes: promoter, characterized by the binding sites for the mitochondrial

Transcription factors and RNA polymerase; Origin of replication, characterized by the so-called 'conserved sequence blocks'; Regulation of DNA replication, characterized by the TAS 'motifs). Since the oligonucleotides contain recognition sequences for the restriction endonucleases Xho I and Pst I, the ends of the amplified nucleic acid can be modified so that on the one hand they are compatible with a vector arm of pBluescript and on the other hand they are compatible with the hybrid of the oligonucleotides MCS / TTS 1 and 2 . In addition to a multiple cloning site, these also contain a transcription termination sequence which is responsible for the regulated transcription. The ligation product is then transformed into E. coli XL1 Blue. Following plasmid isolation of insert-bearing E. coli colonies, the nucleic acids were subjected to an RFLP and sequence analysis. 9: Sequence of the oligonucleotides MCS / TTS 1 and 2. The oligonucleotides MCS 1 and 2 were produced synthetically and contain recognition sequences for nine different restriction endonucleases, as well as a sequence motif which is able to bi-directionally prevent transcription. The oligonucleotides are complementary and can therefore form a hybrid.

The overhanging ends are part of the recognition sequences for the restriction endonucleases Pst I and Barn HI.

10: Nucleotide sequence of the nucleic acid portion of the peptide nucleic acid plasmid (plasmid 1).

11: Cloning of the reporter gene in the nucleic acid portion of the peptide-nucleic acid plasmid in pBluescript (plasmid 2). Using the two oligonucleotides (primers 3 and 4), the gene section from nucleotide 1562 to nucleotide 3359 was amplified enzymatically from a DNA extract of a human CAP-resistant cell line (includes: part of the 12 S rRNA gene, tRNA ^Val gene, 16 S rRNA ^{CAP +} gene, tRNA ^ ^u gene, part of the ND 1 gene).

Since the oligonucleotides contain recognition sequences for the restriction endonucleases Hind HI and Bei I, the ends of the amplified nucleic acid can be modified so that they are compatible with the multiple cloning site (MCS) of the peptide-nucleic acid plasmid (plasmid 1). The ligation product is then transformed into E. coli XL1 Blue. Following the plasmid isolation of insert-bearing E. coli colonies, the nucleic acids were subjected to RFLP and sequence analysis and are available for the experiments described.

12: Nucleotide sequence of the nucleic acid portion of the peptide-nucleic acid plasmid including the reporter gene (plasmid 2). 13a: reaction sequence of the cyclization of the nucleic acid portion and the conjugation of the nucleic acid portion with a signal peptide. The nucleic acid portion of the peptide nucleic acid plasmid can be obtained via a plasmid preparation or an enzymatic amplification. In both cases, treatment with the restriction endonuclease Bsa I leads to an intermediate which can be ligated. This can be implemented directly with the monomerized hairpin loops. The reaction product is separated from non-specific (non-cyclic) reaction products and starting materials by an exonuclease III treatment, purified and conjugated to the signal peptide via a crosslinker. Alternatively, one of the two

'hairpin-loops' are first conjugated to the signal peptide via a crosslinker before the cyclizing ligation reaction is carried out. Here, too, the reaction product is cleaned by an exonuclease III treatment.

13b: Structure and sequence of the 'hairpin-loop' oligonucleotides HP 1 and 2.

Fig. 14: Monomerization of a hairpin loop oligonucleotide. The synthetic hairpin loops HP 1 and 2 can be monomerized by thermal or alkaline denaturation. A standard agarose gel is shown in this figure: Lane 1, molecular weight standard (ΦX 174 RF DNA treated with the restriction endonuclease Hae III); Lane 2: HP 1,

Synthesis product; Lane 3: HP 1, thermally monomerized.

15a: Ligation reaction between the nucleic acid portion of the peptide nucleic acid plasmid (plasmid 2) and the hairpin loops HP 1 and 2. This figure shows a standard agarose gel: lane 1, cloned nucleic acid portion of the peptide nucleic acid portion treated in pBluescript with the restriction endonuclease Bsa I; Lane 2: ligation of the reaction products from lane 1 with the hairpin loops HP 1 and 2; Lane 3, Treatment of the reaction products from lane 2 with exonuclease III; Lane 4, molecular weight standard (λ DNA treated with the restriction endonucleases Hind III and Eco RI).

Fig. 15b: Review of the purified ligation product by a Mae III-RFLP analysis. This figure shows a standard agarose gel: lane 1, enzymatically amplified nucleic acid fraction after a Mae /// treatment; Lane 2: purified ligation product of the enzymatically amplified nucleic acid fraction after a Mae III treatment; Lane 3: purified product of plasmid DNA ligation after Mae III treatment; Lane 4, molecular weight standard (ΦX 174 RF DNA treated with the restriction endonuclease Hae III). Fig. 16: Transcription and replication of the peptide nucleic acid plasmid. This figure shows a standard agarose gel: lane 1, molecular weight standard (λ DNA treated with the restriction endonucleases Hind III and Eco RI); Lane 2, untreated peptide nucleic acid plasmid; Lane 3: transcription products of the peptide nucleic acid plasmid obtained in vitro; lane 4: replication and transcription products of the peptide nucleic acid plasmid obtained in vitro; Lane 5, replication and transcription products of the peptide nucleic acid plasmid obtained in vivo; Lane 6, untreated peptide nucleic acid plasmid.

The present invention will now be explained with reference to the following examples, which, however, are in no way intended to limit the invention.

Example 1:

Introduction of a chimeric peptide nucleic acid fragment into the mitochondria

In order to be able to provide evidence that nucleic acids can be transported through membranes in a targeted manner using the method described above, the Overcome the mitochondrial double membrane system with a DNA translocation vector. For this purpose, the mitochondrial signal sequence of ornithine transcarbamylase (AL Horwich et al. (1983), "Molecular cloning of the cDNA coding for rat ornithine transcarbamoylase", Proc. Natl. Acad. Sci. USA 80: 4258-4262) (enzyme of the urea cycle, naturally located in the matrix of the mitochondria) chemically manufactured and cleaned. As a reactive group for later connection to the DNA, the original sequence was expanded to include a cysteine at the C terminus (see FIG. 1). This ensured that the heterobifunctional crosslinker (MBS) could only be coupled to the thiol group of the single cysteine. A DNA oligonucleotide (39 nucleotides) was selected as the linking partner, which is distinguished by two special features:

1. The sequence is partially palindromic with an overhanging, phosphorylated 5 'end (see Figure 1). This allows a so-called 'hairpin loop' to be formed. The overhanging 5 'end serves to ligate defined nucleic acids to this oligonucleotide, which can then be imported into the mitochondria.

2. The oligonucleotide carries a modified base at the apex of the loop (see FIG. 1). It is an amino-modified 2'-deoxythymidine (see Figure 2). The amino function of the modified base enables the coupling reaction between MBS and oligonucleotide.

The three reaction partners (oligonucleotide, MBS and peptide) are linked in individual reaction steps. First the oligonucleotide (50 pmol) is reacted in a buffer (100 μl; 50 mM potassium phosphate, pH 7.6) together with MBS (10 nmol dissolved in DMSO) (reaction time: 60 min .; reaction temperature: 20 ° C). Unreacted MBS is separated off via a 'Nick Spin Column ^R ' (Sephadex G 50, Pharmacia), which was equilibrated with 50 mM potassium phosphate (pH 6.0). The eluate contains the desired reaction product and is reacted with the peptide (2.5 nmol) in a further reaction step (reaction time: 60 min .; reaction temperature: 20 ° C.). The was stopped Reaction by adding dithiothreitol (2 mM). The coupling product (chimera, see FIG. 3) was separated from unreacted starting materials by preparative gel electrophoresis and isolated from the gel by electroelution (see FIG. 4). Different nucleic acids can now be coupled to the overhanging 5 'end of the oligonucleotide by simple ligation.

For the experiment described below, a 283 bp long double-stranded DNA (dsDNA) was amplified via an enzymatic reaction (PCR). A DNA fragment cloned in pBluescript ^R (Stratagene) was used as template DNA, which, in addition to the human mitochondrial promoter of the light strand (P _L , nt 902 - nt 369), contains the gene for the mitochondrial transfer RNA leucine (tRNA ^{Leu (UUR)} , nt 3204 - nt 4126) (see Figure 5). Two oligonucleotides were used as amplification primers, primer 1 having a non-complementary 5 'end (see FIG. 5). The dsDNA was modified by the 3'-5 'exonuclease activity of the T4-DNA polymerase (incubation in the presence of 1 mM dGTP), which can produce overhanging 5' ends under conditions known to the person skilled in the art (C. Aslanidis et al. (1990) "Ligation-independent cloning of PCR products (LIC-PCR)", Nucleic. Acids. Res. 18: 6069-6074).

Together with the already conjugated peptide-MBS oligonucleotide, the PCR-amplified DNA could be assembled using the T4 DNA ligase. In order to be able to easily detect the linkage partners after introduction into the mitochondria, the free 5'-OH group of the ligated DNA was radioactively phosphorylated by an enzymatic reaction (A. Novogrodsky et al. (1966), "The enzymatic phosphorylation of ribonucleic acid and deoxyribonucleic acid. I. Phosphorylation at 5 "-hydroxyl termini", J. Biol. Chem. 241 .: 2923-2932; A. Novogrodsky et al. (1966), "The enzymatic phosphorylation of ribonucleic acid and deoxyribonucleic acid. π. Further properties of the 5'-hydroxyl polynucleotide kinase ", J. Biol. Chem. 241: 2933-2943). For the isolation of mitochondria, a fresh rat liver was crushed, suspended in 25 mM HEPES, 250 mM sucrose, 2 mM EDTA, 52 μM BSA and homogenized in a glass homogenizer (50 ml). Cell membranes, cell debris and cell nuclei were centrifuged off at 3000 g and the supernatant prepared for a further centrifugation. For this purpose, the supernatant was transferred to cooled centrifuge beakers and centrifuged at 8000 g. The isolated mitochondria were resuspended in 200 ml of the same buffer and centrifuged again at 8000 g. The clean mitochondrial pellet was resuspended in an equal volume of the same buffer and energized by adding 25 mM succinate, 25 mM pyruvate and 15 mM malate. The protein content of the suspension was determined using a Bradford test kit * (Pierce). 200 μg mitochondrial protein (energized mitochondria) were together with 10 pmol of the chimeric at 37 ° C. for 60 min. incubated (0.6 M sorbitol, 10 mM potassium phosphate pH 7.4, 1 mM ATP, 2 mM MgCl ₂ , 1% BSA). The mitochondria were reisolated by centrifugation at 8000 g, resuspended in 0.6 M sorbitol, 10 mM potassium phosphate pH 7.4, 2 mM MgCl ₂ , 1% BSA, 10 U / ml DNAse I and 30 min. incubated at 37 ° C. This washing step was repeated twice to remove non-specifically adhering molecules. To demonstrate that the chimera is associated with the mitochondria, the re-isolated mitochondria were purified using sucrose gradient density centrifugation. The individual fractions of the gradient were analyzed to localize the chimeras and mitochondria. The adenylate kinase, the cytochrome c oxidase and the malate dehydrogenase activity were determined as markers of the mitochondria, while the chimera could be identified by measuring the ³² P radiation (see FIG. 6). An analogous experiment to determine the unspecific DNA incorporation was carried out with the same DNA that was not linked to the signal peptide (see FIG. 6). It was derived from the measurements that 65% of the chimeras used specifically segregated with the mitochondria, while the non-specific DNA incorporation was less than 5% of the DNA used. In order to show that the chimera is not only associated with the surface of the mitochondria (membrane, import receptor), the re-isolated mitochondria were divided into the three compartments outer mitochondrial membrane / intermembrane space, inner mitochondrial membrane and Fractionated matrix space. For this purpose, the mitochondria were incubated with digitonin (final concentration: 1.2% w / v digitonin) and the resulting mitoplasts were separated using sucrose gradient density centrifugation, collected in fractions and the activities of marker enzymes (adenylate kinase: intermembrane space; cytochrome c oxidase: inner mitochondrial membrane; Dehydrogenase: matrix space) according to Schnaitmann and Greenawalt (C. Schnaitman et al. (1968), "Enzymatic properties of the inner and outer membranes of rat liver mitochondria", J. Cell Biol. 3_8: 158-175; C. Schnaitman et al (1967), "The submitochondrial localization of monoamine oxidase. An enzymatic marker for the outer membrane of rat liver mitochondria", J. Cell Biol. 32: 719-735) (see FIG. 7). An analogous experiment to determine the unspecific DNA incorporation was carried out with the same DNA that was not linked to the signal peptide (see FIG. 7). It was deduced from the measurements that 45% of the chimeras was associated with the mitoplasts, while the non-specifically adhering DNA could be estimated at less than 3%. The isolated mitoplasts (loss of the outer membrane and the intermembrane space) were lysed by Lubrol ^R (0.16 mg / mg protein; ICN) and separated into the compartments inner mitochondrial membrane (pellet) and matrix space (supernatant) by ultracentrifugation at 144000 g. The compartments were assigned by measuring the activities of the cytochrome c oxidase (inner mitochondrial membrane) and the malate dehydrogenase (matrix space). The chimeras were measured by detecting the ² P radiation in the scintillation counter and showed 75% segregation with the matrix of the mitochondria, while 25% of the chimeras remained associated with the inner membrane of the mitochondria (incomplete translocation).

Example 2: Introduction of a regenerative and transcription-active chimeric peptide nucleic acid fragment (plasmid) into the mitochondria of living cells

In order to be able to prove that a linear peptide-nucleic acid plasmid with cyclized ends ('hairpin-loops') membranes in vivo via the protein import route overcome and can be transcribed and replicated despite the chemical coupling with a signal peptide, the transcription and replication behavior after the transfection of cells and the import into the matrix of the mitochondria was studied. For this purpose, the signal peptide of mitochondrial ornithine transcarbamylase was synthesized, purified and linked to a nucleic acid plasmid.

A prerequisite for checking the correct transcription and replication behavior is the physical structure of the plasmid: for the experiment described below, a 3232 bp long double-stranded vector DNA (dsDNA) was cloned into pBluescript ¹¹ (Stratagene). For this purpose, the region of the mitochondrial genome was amplified via two modified oligonucleotides (primer 1, hybridized with nucleotides 15903-15924 of the human mtDNA, contains an extension at the 5 'end by the sequence TGTAGctgcag to introduce a Pst I site; primer 2, hybridizes with nucleotides 677-657 of the human mtDNA, includes at the 5 'end an extension by the sequence TTGCATGctcgagGGTCTCAGGG for the introduction of an Xho I interface), which is the promoter of the light DNA strand, the origin of the mtDNA replication of the heavy strand , the regulatory motifs for transcription (CODs, 'conserved sequence blocks ¹ ), and the regulatory body for DNA replication (' TAS ', termination associated sequences, (DC Wallace (1989), "Report of the committee on human mitochondrial DNA", Cytogenet Cell Genet. _5:... 612-621) contained (see Figure 8) Behind this fragment (3 'direction) has been inserted a multiple cloning site, which should allow simple linking to a gene to be expressed. The multiple cloning site (MCS / TTS) was generated by chemical synthesis of two complementary oligonucleotides (MCS / TTS 1 and 2), which contain recognition sequences for different restriction endonucleases (see FIG. 9). Under conditions known to those skilled in the art, the two oligonucleotides form hybrids which, after phosphorylation with T4 DNA polynucleotide kinase, can be used for the ligation. The hybrids are characterized in this context by 5'- and 3'-single-strand overhanging ends, which are complementary to a Pst I, on the one hand others are complementary to a Bam HI interface (see FIG. 9). Together with the multiple cloning site, the synthetic oligonucleotides MCS / TTS 1 and 2 also contain a bidirectional mitochondrial transcription termination sequence (see FIG. 9). It is arranged in the 3 'direction to the MCS and ensures that the transcription is interrupted at this point and that correctly terminated transcripts are created. This sequence motif also ensures that no 'antisense RNA' is expressed in the cyclic plasmid system. The ligation reaction between pBluescript, PCR-amplified fragment and the MCS / TTS hybrid was carried out in a stoichiometry of 1: 2: 2 under conditions known to the person skilled in the art. After the transformation, several E. coli colonies (clones) could be isolated and characterized. For this purpose, the corresponding plasmid DNA was subjected to dideoxy sequencing (FIG. 10) under conditions known to the person skilled in the art.

A so-called reporter gene was inserted into the multiple cloning site for experimental verification of replication and transcription. The chloramphenicol-resistant human mitochondrial 16 S ribosomal RNA was selected as the reporter gene, which differs from the naturally occurring ribosomal RNA only by a modified nucleotide (polymorphism). The polymerase chain reaction was used to convert a DNA extract of chloramphenicol-resistant HeLa cells into a fragment with two modified oligonucleotides (primer 3, hybridized with nucleotides 1562-1581 of mitochondrial DNA, at the 5 'end around the CCTCTaagctt sequence to introduce a Hind πi -Interface extended; primer 4, hybridized with nucleotides 3359-3340, amplified at the 5 'end by the sequence GCATTactagt to introduce an At I-interface) amplified under conditions known to the person skilled in the art. In order to ensure correct processing of the later transcript, the amplification ^product comprised the two flanking tRNA genes (tRNA ^VaI and tRNA ^Leu ). The amplified DNA was treated with the restriction endonucleases Hind III and Bei I, purified by precipitation and with the pBluescript plasmid 1 treated with Hind III and Bei I (see FIGS. 8, 9 and 10) in a stoichiometry of 1: 1 in a ligation reaction under conditions known to those skilled in the art. The cloning strategy is shown in Figure 11.

Several E. coli colonies (clones) could be isolated and characterized. For this purpose, the corresponding plasmid DNA was subjected to dideoxy sequencing under conditions known to the person skilled in the art (see FIG. 12). In order to prepare the cloned DNA for use on cell cultures and mitochondria, the cloning insert (mitochondrial transformation plasmid) was separated from the pBluescript vector under the conditions known to the person skilled in the art by using the restriction endonuclease Bsa I. Alternatively, the insert DNA could be amplified by means of the polymerase chain reaction via two oligonucleotides (primers 2 and 5; nucleotide sequence of primer 5: GATCCGGTCTCATTTTATGCG). When using S-dNTPs, this allowed the production of 'thionated' DNA that is stabilized against cellular nucleases. In both cases, the subsequent use of the restriction endonuclease Bsa I leads to two different 5 'overhangs. These are complementary to the Ηairpin loops' used in order to achieve a cyclization of the linear nucleic acid (see FIGS. 13a and b). The oligonucleotides are produced by chemical synthesis. As a result, they have no phosphorylated 5 'ends and have to be phosphorylated by a kinase reaction under conditions known to the person skilled in the art (in order to be able to check the cellular transformation later, [γ- ³² P] -ATP was partly used as substrate in this reaction to radioactively label the plasmid). The 'haiφin-loop' structure of the oligonucleotides is formed spontaneously to a large extent because the palindromic sequence is able to hybridize with itself. Dimers of the hairpin loops can also be converted into monomers by using them in the largest possible volume (<0.1 μM) for at least 5 min. denatured at 93 ° C and immediately fixed in a solid matrix by freezing. The oligonucleotides are then slowly thawed at 4 ° C. and are then 99% present in the desired monomeric 'shark-in-loop' structure (see FIG. 14). The plasmid DNA was cyclized together with the two monomerized 'haiφin-loops' (HP 1 and 2) in a single reaction. The molar ratio of plasmid DNA to the two 'haiφin loops' was 1: 100: 100 (plasmid: HPl: HP2). Using the T4 DNA ligase, the individual reactants could be assembled under conditions known to the person skilled in the art (see FIG. 15). The ligation products were purified by treatment with exonuclease HI (reaction conditions: 37 ° C., 60 min.). While nucleic acids with free 3 'ends are degraded by the nuclease, the plasmid DNA associated with the two' haiφin loops' remains stable towards the 3 '-5' exonuclease activity of the enzyme. The only reaction product (see FIG. 15a) was separated using preparative agarose gel electrophoresis and purified by electroelution or using QIAquick (Qiagen) according to the manufacturer's recommendation.

The ligation product was checked using an RFLP analysis (restriction fragment length polymorphism). For this purpose, the ligated and purified plasmid DNA was treated with the restriction endonuclease Mae III under conditions known to the person skilled in the art. The DNA has five cleavage sites, so that fragments of different sizes are formed, which can be analyzed using an agarose gel (4%). FIG. 15b shows an example of the Mae III cleavage pattern that is obtained after the ligation of the plasmid DNA with the two haiφin loops ¹ . The DNA bands marked with arrowheads represent the left and right ends of the amplified (lane 1) and the linear-cyclic (lane 2 and 3) mitochondrial plasmid.

To conjugate the circularized plasmid with the synthetic signal peptide of rat ornithine transcarbamylase (H ₂ N-MLSNLRILLNKAALRKAHTSMVRNFRYGK-PVQSQVQLKPRDLC-COOH), the nucleic acid with a 20-fold molar excess of m-maleimidobenzoyl-in-hydroxysaminophenocyclin. incubated at 20 ° C (incubation medium: 50 mM potassium phosphate pH 7.8). The excess coupling agent was separated by a "nick spin column" (Pharmacia-LKB) under conditions known to the person skilled in the art. The conjugation of the "activated" Nucleic acid was obtained by reacting the nucleic acid with a 50-fold molar excess of the signal peptide at 20 ° C. (incubation medium: 50 mM potassium phosphate pH 6.8). After 45 min. the reaction was terminated by the addition of 1 mM dithiothreitol and the conjugate was available for the following experiments.

In order to demonstrate the usability of the peptide-nucleic acid plasmid in vivo, the plasmid had to be introduced into eukaryotic cells. For this purpose, a chloramphenicol-sensitive B-lymphocyte or fibroblast cell culture was transfected via lipotransfection with the peptide-nucleic acid plasmid: 1 μg of the radioactively labeled peptide-nucleic acid plasmid (the label was identified as ³² P-label during the kinase reaction of the 'haiφin loops' (HPl) was pre-incubated together with 2-6 μl LipofectAmine ^R (Gibco-BRL) in 200 μl serum-free Optimem ^R (Gibco-BRL) (15 min., 20 ° C). During the incubation, the polycationic lipid of the LipofectAmine ^R reagent forms DOSPA (2,3-dioleyloxy-N- [2- (sperminecarboxamido) -ethyl] - N, N-dimethyl-l-propanaminium trifluoroacetate) with the support of the neutral lipid DOPE (dioleoylphosphatidylethanolamine ) unilamellar liposomes, which are able to complex the DNA. The reaction mixture was then added to the prepared cells, which had been adjusted to a density of approx. 2.5 * 10 ⁶ cells per 0.8 ml (35 mm culture dishes, 4 h, 37 ° C., CO ₂ incubator). The transfection medium was then replaced by 5 ml of DMEM medium (Gibco-BRL), which had previously been supplemented with 10% fetal calf serum and 100 μg / ml chloramphenicol. The transformation efficiency was determined by measuring the ³² P radiation of the construct. As a rule, a cellular incorporation rate of 80-85% was measured. This means that 80-85% of the chimeric construct was associated with the transformed cells and 15-20% of the chimeric peptide-DNA plasmid remained in the supernatant of the transfection reaction.

After about 21-28 days, chloramphenicol-resistant colonies formed in the transformed cells. The resistant cells were separated and expanded under conditions known to the person skilled in the art. From about 1 * 10 ⁵ cells could sufficient DNA can be obtained from conditions known to the person skilled in the art to carry out genotyping. For this purpose, the isolated DNA was separated by agarose gel electrophoresis and transferred to a nylon membrane (Southern blot). The nucleic acids were detected by hybridization with a specific, radioactively labeled probe (see FIG. 16). In addition to the introduced circularized 'linear' vector (lanes 2 and 6), this figure shows an 'in vitro' transcription (lane 3), an 'in vitro' replication (lane 4) and the intermediates obtained in vivo (isolated Nucleic acids of a transformed clone) is shown. While the three smaller bands can be generated in vitro by incubating the circularized vector with the four nucleoside triphosphates (RNA) and a mitochondrial enzyme extract (lane 3), the addition of the deoxynucleoside triphosphates to the reaction mixture is observed to produce a dimer, circular plasmid ( largest band in lane 4): an identical picture emerges when analyzing the nucleic acids that can be obtained from transformed cell colonies (lane 5). Sequence analysis confirmed that the largest DNA band in lanes 4 and 5 is actually the dimeric and therefore replicated mitochondrial plasmid.

A lipotransfection approach was used as the control experiment, in which the unconjugated plasmid which was not linked to the signal peptide was used. As expected, this plasmid was not taken up into the mitochondria of the transfected cells and consequently did not lead to the formation of chloramphenicol resistant cells. These cells stopped growing after 10 days and died completely within the next 8 to 10 days.

Claims

1) Chimeric peptide nucleic acid fragment comprising:

(a) cell-, compartment- or membrane-specific signal peptide with the exception of a KDEL signal sequence, (b) coupling agent,

(c) Nucleic acid (oligonucleotide), where the coupling of the signal peptide takes place via the coupling agent which is connected to the signal peptide via amino acids at the carboxy-terminal end, thereby ensuring the targeted introduction of nucleic acids into cell organelles and cells.

2) Chimeric peptide-nucleic acid fragment according to claim 1, characterized in that the nucleic acid consists of at least two bases.

3) Chimeric peptide-nucleic acid fragment according to claim 1 or 2, characterized in that the nucleic acid has a hybridizable secondary structure.

4) Chimeric peptide-nucleic acid fragment according to one of claims 1-3, characterized in that the nucleic acid has a palindromic sequence.

5) Chimeric peptide-nucleic acid fragment according to claim 4, characterized in that the nucleic acid can form a 'haiφin-loop'.

6) Chimeric peptide nucleic acid fragment according to claim 5, characterized in that the nucleic acid can hybridize with itself and form an overhanging 3 'or 5' end ('sticky end'). 7) Chimeric peptide nucleic acid fragment according to any one of claims 1-6, characterized in that the nucleic acid is a ribonucleic acid, preferably a deoxyribonucleic acid.

8) Chimeric peptide-nucleic acid fragment according to claim 7, characterized in that the nucleic acid has chemically modified 'phosphorothioate' bonds.

9) Chimeric peptide-nucleic acid fragment according to one of claims 1-8, characterized in that the nucleic acid carries a reactive coupling group.

10) Chimeric peptide nucleic acid fragment according to claim 9, characterized in that the reactive coupling group contains an amino function if the coupling agent contains an amino-reactive group.

11) Chimeric peptide-nucleic acid fragment according to claim 9, characterized in that the reactive coupling group contains a thiol function if the coupling agent contains a thiol-reactive group.

12) Chimeric peptide nucleic acid fragment according to claim 10 or 11, characterized in that the coupling group at least one

C2 spacer, but preferably a C6 spacer is present bound to the nucleic acid.

13) Chimeric peptide-nucleic acid fragment according to claim 12, characterized in that the coupling group is located at the 3 'hydroxy / phosphate terminus or at the 5' hydroxy / phosphate terminus of the nucleic acid, but preferably at the base.

14) Chimeric peptide nucleic acid fragment according to any one of claims 10-13, characterized in that defined nucleic acids, antisense oligonucleotides, messenger RNAs or transcribable and / or replicable genes are coupled to the 5 'end and / or 3' end become. 15) Chimeric peptide-nucleic acid fragment according to claim 14, characterized in that the nucleic acid to be coupled contains chemically modified 'phosphorothioate' bonds.

16) Chimeric peptide nucleic acid fragment according to claim 14 or 15, characterized in that the gene to be coupled contains a promoter, preferably a mitochondrial promoter.

17) Chimeric peptide-nucleic acid fragment according to one of claims 1-16, characterized in that the signal peptide has a reactive amino acid at the carboxy-terminal end, preferably a lysine or cysteine, if the coupling agent contains an amino- or thiol-reactive group.

18) Chimeric peptide-nucleic acid fragment according to one of claims 1-17, characterized in that the signal peptide carries a cell-, compartment- or membrane-specific recognition signal.

19) Chimeric peptide-nucleic acid fragment according to one of claims 1-18, characterized in that the signal peptide has a cell-, compartment- or membrane-specific peptide cleavage site.

20) Chimeric peptide-nucleic acid fragment according to one of claims 1-19, characterized in that the peptide consists of the compartment-specific cleavable signal peptide of human mitochondrial ornithine transcarbamylase, extended by an artificial cysteine at the C-terminus.

21) Chimeric peptide-nucleic acid fragment according to one of claims 1-20, characterized in that the coupling agent is a bifunctional, preferably a heterobifunctional, crosslinker. 22) Chimeric peptide-nucleic acid fragment according to one of claims 1-21, characterized in that the coupling agent contains thiol-reactive and / or amino-reactive groups if the signal peptide and the nucleic acid carry thiol and / or amino groups as coupling sites.

23) Chimeric peptide nucleic acid fragment according to one of claims 1-22, characterized in that the coupling agent is m-maleimido-benzoyl-N-hydroxy-succinimide ester or a derivative thereof.

24) Chimeric peptide nucleic acid fragment according to any one of claims 1-23, characterized in that the molecule can overcome membranes with and without membrane potential using natural transport mechanisms.

25) Chimeric peptide-nucleic acid fragment in the form of a linear cyclic plasmid, characterized in that the plasmid comprises at least one origin of replication and that both ends of the nucleic acid portion are cyclized, with at least one cyclized end having a modified nucleotide which has a coupling agent a cell-, compartment- or membrane-specific signal peptide can be coupled.

26) Chimeric peptide nucleic acid fragment according to claim 25, characterized in that the nucleic acid portion further comprises at least one promoter, preferably a mitochondrial promoter, most preferably the mitochondrial promoter of the light strand.

27) Chimeric peptide nucleic acid fragment according to one of claims 25 and 26, characterized in that the neucleic acid portion further comprises transcriptional regulatory sequences, preferably mitochondrial transcriptional regulatory sequences. 28) Chimeric peptide nucleic acid fragment according to one of claims 25-27, characterized in that the transcription regulatory sequences have at least one binding site of a transcription initiation factor.

29) Chimeric peptide-nucleic acid fragment according to one of claims 25-28, characterized in that the transcription regulatory sequences via at least one

Binding site for the RNA synthesizer, preferably have the binding site for the mitochondrial transcription factor 1 and the mitochondrial RNA polymerase.

30) Chimeric peptide nucleic acid fragment according to one of claims 25-29, characterized in that the transcription regulatory sequences in the 3 'direction to

Promoter are arranged.

31) Chimeric peptide nucleic acid fragment according to one of claims 25-30, characterized in that the regulation of the transcription is carried out by elements of the mitochondrial H and L strand transcription control.

32) Chimeric peptide nucleic acid fragment according to claim 31, characterized in that the so-called 'conserved-sequence blocks' of the L-strand transcription function as transcription control elements.

33) Chimeric peptide nucleic acid fragment according to one of claims 25-32, characterized in that the plasmid further comprises at least one transcription

Termination point includes.

34) Chimeric peptide nucleic acid fragment according to one of claims 25-33, characterized in that the transcription termination site has a binding sequence of a mitochondrial transcription termination factor. 35) Chimeric peptide nucleic acid fragment according to claim 34, characterized in that the transcription termination site has the binding sequence of a preferably bi-directional transcription termination factor.

36) Chimeric peptide-nucleic acid fragment according to one of claims 25-35, characterized in that the origin of replication is a mitochondrial origin of replication, preferably the origin of replication of the heavy mtDNA strand, which has at least one conserved sequence block ''.

37) Chimeric peptide nucleic acid fragment according to one of claims 25-36, characterized in that the plasmid further comprises at least one regulatory sequence for replication.

38) Chimeric peptide nucleic acid fragment according to one of claims 25-37, characterized in that the regulatory sequence for replication is a mitochondrial sequence motif.

39) Chimeric peptide nucleic acid fragment according to one of claims 25-38, characterized in that the plasmid further comprises a selection gene, preferably an antibiotic resistance gene, preferably the oligomycin or chloramphenicol resistance gene.

40) Chimeric peptide nucleic acid fragment according to one of claims 25-39, characterized in that the plasmid further contains a multiple cloning site which allows the expression of 'foreign genes'. 41) Chimeric peptide-nucleic acid fragment according to one of claims 25-40, characterized in that the multiple cloning site contains recognition sequences for restriction endonucleases, which preferably do not occur at any other site on the plasmid.

42) Chimeric peptide nucleic acid fragment according to one of claims 25-41, characterized in that the multiple cloning site is arranged in the 3 'direction to the promoter and in the 5' direction to the transcription termination site.

43) Chimeric peptide nucleic acid fragment according to one of claims 25-42, characterized in that the multiple cloning site is arranged in the 5 'direction to the selection gene.

44) Chimeric peptide-nucleic acid fragment according to one of claims 25-43, characterized in that the nucleic acid fragment has ligation-capable (phosphorylated) ends.

45) Chimeric peptide-nucleic acid fragment according to one of claims 25-44, characterized in that the nucleic acid fragment via 'blunt ends' or

3 'overhanging ends, preferably 5' overhanging ends.

46) Chimeric peptide-nucleic acid fragment according to one of claims 25-45, characterized in that the nucleic acid fragment has 5 'overhangs comprising 4 nucleotides, which have no self-homology (palindromic sequence) and are also not complementary to one another.

47) Chimeric peptide nucleic acid fragment according to one of claims 25-46, characterized in that the ends of the nucleic acid fragment are cyclized via synthetic oligonucleotides. 48) Chimeric peptide nucleic acid fragment according to one of claims 25-47, characterized in that the overhanging 5 'ends of the two oligonucleotides are each complementary to a different end of the nucleic acid.

49) Chimeric peptide nucleic acid fragment according to one of claims 25-48, characterized in that two different 'haiφin-loops' are used for the cyclization, one specific (complementary) for the 'left' plasmid end, the other specific for the 'right' plasmid end of the nucleic acid.

50) Chimeric peptide-nucleic acid fragment according to one of claims 25-49, characterized in that the modified nucleotide is preferably located within the 'loop'.

51) Chimeric peptide-nucleic acid fragment according to one of claims 25-50, characterized in that the plasmid DNA is enzymatically amplified with suitable oligonucleotides which have at least one recognition sequence for a restriction endonuclease, which is preferably not repeated in the

Plasmid sequence occurs.

52) Chimeric peptide nucleic acid fragment according to claim 51, characterized in that the restriction endonuclease to be used generates overhanging ends, preferably 5 'overhanging ends, the cleavage site preferably being located outside the recognition sequence.

53) Chimeric peptide nucleic acid fragment according to claim 51 or 52, characterized in that the restriction endonuclease is Bsa I. 54) Method for producing a chimeric peptide-nucleic acid fragment according to one of claims 1-53, characterized by the following steps: (a) reaction of a nucleic acid (oligonucleotide) containing a functional coupling group with a coupling agent. (b) Implementation of the construct from (a) with amino acids on

Carboxy-terminal end of a peptide containing a signal sequence, with the exception of a KDEL signal sequence, and (c) optionally extending the chimeric peptide nucleic acid fragment from (b) by further DNA or RNA fragments.

55) Method according to claim 54, characterized in that the DNA in step (c) is a PCR amplified DNA fragment containing the human mitochondrial promoter of the light strand (P _L ) and the gene for the mitochondrial transfer RNA Leucine (tRNALeu ^(UUR) ).

56) Method for producing a chimeric peptide-nucleic acid fragment according to one of claims 1-53, characterized by the following steps:

(b) Implementation of the nucleic acid with a functional coupling group or the extended nucleic acid from (a) with a coupling agent. (c) Implementation of the construct from (b) with amino acids on

Carboxy-terminal end of a peptide containing a signal sequence, except for a KDEL signal sequence.

57) Method according to claim 56, characterized in that the DNA in step (a) is a PCR-amplified DNA fragment containing the human mitochondrial promoter of the light strand (P _L ) and the gene for the mitochondrial transfer RNA Leucine (tRNALeu ^^). 58) Use of the chimeric peptide nucleic acid fragments according to one of the

Claims 1-53 for targeted nucleic acid introduction in cell organelles and cells, characterized by the implementation of the fragment with cells or pretreated cell compartments.

59) Use according to claim 58, characterized in that the pretreated cell compartments are energized mitochondria.

60) Use of the chimeric peptide nucleic acid fragment according to any one of claims 1-59 for introduction into eukaryotic cells.

61) Use of the chimeric peptide-nucleic acid fragment according to claim 60, characterized in that the 'Particle Gun' system, electroporation,

Microinjection or lipotransfection is used for introduction into eukaryotic cells.