EP1352071A1

EP1352071A1 - Vector for introducing molecules in eukaryotic cells and molecules vectorized by same

Info

Publication number: EP1352071A1
Application number: EP02712014A
Authority: EP
Inventors: Denis Michel
Original assignee: Universite de Rennes 1
Current assignee: Universite de Rennes 1
Priority date: 2001-01-19
Filing date: 2002-01-18
Publication date: 2003-10-15
Also published as: FR2819811B1; WO2002057462A1; US20040152195A1; FR2819811A1; IL157005A0; CN1491283A; CA2435094A1; JP2004522439A

Abstract

The invention concerns a vector for introducing a molecule (X) having at least an amino-terminal end into a eukaryotic cell, characterised in that it consists of an association of a cell incorporating module enabling membrane translocation, and an uniquitin module or other related peptide capable of conjugation, said vector having a carboxy-terminal end designed to be fused by peptide bond to the amino-terminal end of said molecule (X). The invention is characterised in that said vector is rapidly destroyed after separation from said molecule (X). The invention also concerns any molecule vectorized by such a vector.

Description

VECTOR FOR INTRODUCING MOLECULES IN EUKARYOTIC CELLS AND MOLECULES VECTORIZED BY SUCH A VECTOR

The invention relates to the field of biochemistry, cell biology and medicine. More specifically, the invention relates to a molecular tool, hereinafter called

"vector", intended to be used for the introduction of all types of molecules into eukaryotic cells.

The invention particularly finds its application for the intracellular incorporation of amino acids, or peptides of any size and sequence (dipeptides, tripeptides, polypeptides, proteins, etc.), which can be chemically modified.

The invention also relates to any molecule vectorized by such a vector. It will be noted that in the present description the terms "vectorized molecule" are understood to refer to the assembly constituted by the vector fused with the molecule to be transported in the cell. According to the prior art, the supply of molecules into eukaryotic cells can be carried out by fusing the molecule to be transported with cellular incorporation modules ("cell-permeable modules" CPM), such as membrane translocation polypeptides, capable of cross cell membranes and gain direct access to the cytoplasmic compartment of cells.

Several CPMs are already known and used to introduce peptides and proteins, which do not spontaneously cross the membranes of eukaryotic cells. CPMs can correspond to natural protein domains (Schwarze et al. 2000. Trends Cell Biol. 10, 290-295, Fujihara and Nadler 1999. EMBO J. 18, 411-419) or to peptides of artificial sequences ( Futaki et al. 2000. J. Biol. Chem. 276, 5836-5840; Ho et al. 2001. Cancer Res. 61, 474-477). Several properties of translocation peptides described according to the prior art, make them promising tools in cellular and medical biology, such as for example:

- an efficient passage through the blood-brain barrier (tight junctions), while preserving its integrity.

- the ability to bypass the drug resistance system (MDR), by which tumor cells become resistant to many anti-cancer agents.

- direct passage into the cytoplasmic compartment, independent of endocytosis phenomena. This peculiarity, not very widespread among the other modes of non-viral vectorization, allows efficient cellular incorporation at low temperature, and avoids the problem of the exit of endosomes towards the cytoplasm. - rapid vectorization in all regions of the body shortly after intravenous injection, preventing a prolonged presence in the blood. - the possible coupling of these peptides to non-peptide compounds and to nucleic acids, which are hydrophilic and do not spontaneously cross cell membranes.

The main application of this prior art is the introduction of proteins or peptides into cells, by means of CPM consisting of translocation peptides. According to this prior art, the proteins or peptides (X) to be introduced into the cells are prepared in fusion with the CPMs via a peptide bond, in the form of recombinant CPM-X proteins. These proteins are produced in appropriate expression systems, for example in bacteria, or else synthesized using peptide synthesizers. When brought into contact with cells, the recombinant proteins CPM-X are fully incorporated by these cells, and part X can exercise the biological activity for which it is designed (Schwarze et al. 2000. Trends Cell Biol. 10 , 290-295).

Simple CPM-X merges have several drawbacks:

- the fact that the CPM remains associated with the molecule X after cellular incorporation, can prevent the good intracellular functioning of the transported molecule (X), by decreasing the specificity and the affinity of interaction of X with its intracellular targets, and by unfavorable to its authentic three-dimensional folding. These negative interferences are particularly probable in the case where X is small, such as a peptide.

- the fact that the CPM remains associated with the molecule X after cellular incorporation can also prevent the appropriate subcellular locahsation of X. For example, many translocation peptides currently known have tendencies to accumulate in certain subcellular compartments (linked nuclear localization to an activity "NLS", re secretion etc.), bringing with them the molecule X. Conversely, fusion with a vector can prevent the functioning of certain localization sequences which would have been provided for in the transported molecule X, such as for example a N-terminal mitochondrial addressing signal.

- CPMs can present, beyond a certain concentration, a certain degree of toxicity for cells and the organism, so as to limit their use in cell and medical biology. The possible toxicity can come from the destabilizing properties of the membranes of certain translocation peptides, such as for example fragments of viral envelope proteins or bacterial toxins, or else from the presence of electrically charged domains such as for example the translocation peptides rich in basic amino acids, or polyhistidine sequences, labeling and affinity purification for metals. Finally, the various modules of the vectorized molecule other than the transported molecule X, may present risks of immunogenicity. Such limitations of the state of the art constitute the drawbacks thereof. The objective of the present invention is to propose a technique for introducing molecules into eukaryotic cells, using CPMs but presenting new advantages compared to the techniques of the prior art. In particular, an objective of the present invention is to propose a vector making it possible to transport into eukaryotic cells any type of molecule and in particular any type of peptide, including very short peptides such as simple amino acids, dipeptides or tripeptides, and peptides whose first amino acid is not methionine. These objectives are achieved thanks to the invention which relates to a vector for the introduction into a eukaryotic cell of a molecule (X) having at least one amino-terminal end, characterized in that it is constituted by the association of a cellular incorporation module allowing the membrane translocation, and an ubiquitin or other related conjugable peptide module, said vector having a carboxy-terminal end intended to be fused by peptide reason at the amino-terminal end of said molecule ( X).

It will be noted that in the present description is meant the terms "ubiquitin or other related conjugable peptides", hereinafter called "UBL", as designating ubiquitin or related peptides such as NEDD-8, SUMO (Small Ubiquitin- related MOdifier), UCRP (Ubiquitin Cross-Reacting Protein), RUB (Related to UBiquitin) and others (Jentsch and Pyrowolakis 2000. Trends Cell Biol. 10, 335-342). The use of modified or shortened versions of UBL in the context of the present invention is possible if their capacity to induce cleavage at their C-terminal end is preserved. Preferably, UBL is ubiquitin. UBLs are naturally synthesized in fusion with carboxyl terminal extensions, consisting of simple amino acids, polypeptides or proteins. There are cellular enzymatic activities capable of cleaving precisely between the UBL domains and their C-terminal extensions.

In the case of ubiquitin, the existence of such activities has long been proven (Lund et al. 1985. J. Biol. Chem. 260, 7609-7613; Pickart and Rose 1985. J. Biol. Chem. 260, 7903-7910).

Responsible enzymes have been identified (Mayer et al. 1989. Biochemistry 28, 166-172). However, knowledge and possession of the enzymes responsible for these cleavage activities are in no way necessary for the implementation of the present invention. Several uses of these enzymatic activities have already been described in the field of biotechnology. The present invention takes advantage of the natural existence of these enzymes in cells for a new and original objective: the intracellular cleavage of molecules pre-synthesized outside the cells, after incorporation by the cells. In the context of the present invention, the presence of a UBL module at the N-terminal part of the transported molecule X, in fact allows the cleavage between UBL and X, and therefore the release of the molecule X from the rest of the vector.

The use of the C-terminal cleavage of UBL to achieve this objective has several advantages: - The enzymatic activities responsible for these cleavages are widespread in most eukaryotic cells and constitutively active, which makes it possible to generalize the use of this system vector.

- The C-terminal cleavage of UBL is only conditioned by the recognition of the amino acid sequence of the UBL module, located exclusively on the NH2 side of the cleavage site, which makes it possible to introduce any sequence on the COOH side. This property makes it possible to impose no condition or restriction as to the nature of the X peptides. In particular, very short peptides, such as dipeptides, can be as efficiently transported according to this invention.

- Since the cleavage is precisely at the C-terminal end of the UBLs, no constraint exists as to the nature of the N-terminal amino acid of molecule X. In particular, this amino acid may be different from a methionine, which is not the case for peptides synthesized from transgenes, after transcription and intracellular translation.

The use of modified or shortened versions of UBL is possible if their ability to induce cleavage at their C-terminal end is preserved.

Very many types of molecules (X) can be vectorized by the vector according to the invention and in particular simple amino acids and peptides of any size, and independently of any constraint of composition and sequence. The vector according to the invention also makes it possible to vectorize chemically modified peptides.

The vector according to the invention can also be used for the cellular incorporation of non-peptide molecules, such as for example oligonucleotides or hydrophilic chemical compounds which will have been grafted beforehand at the carboxy-terminal end of an oligopeptide. It will also be noted that it will be possible to chemically modify the vector molecules, provided that the peptide bond between the vector molecule and the vector remains accessible for C-terminal hydrolases from PUBL. For example, a C-terminal acylation of the molecule (X) will allow its intracellular membrane anchoring.

The vector according to the present invention also has advantages compared to the peptides simply associated with membrane translocation peptides as stated previously.

More specifically, the present invention allows the intracellular release of the molecules (X) transported, in a form completely free of any fusion after cleavage between UBL and X.

The physical dissociation between molecule X and the rest of the vector according to the present invention has several advantages: a- The absence of functional interference between the vector and molecule X. If molecule X must ensure a biological function in cells, the efficiency of this function can be considerably reduced if the other modules of the vectorized molecule remain associated with it (CPM, UBL and others). By steric hindrance, these modules can hinder:

-> the specificity and the affinity of interaction of molecule X with endogenous molecules of the cell, as therapeutic targets. This last point is particularly likely in the case where X is a peptide.

-> the authentic three-dimensional folding of the molecule X. -> the appropriate subcellular localization of the molecule X. Many translocation peptides currently known have tendencies to move towards certain localizations (nuclear locahsation by NLS activity ("Nuclear Localization Signal") ), re-secretion etc.), carrying with them the molecules transported X. Conversely, fusion with a vector can prevent the functioning of certain locahsation sequences which would have been provided for in molecule X, such as for example an N-terminal signal mitochondrial addressing. b- In the case where molecule X is a peptide, this system allows total freedom of choice over the nature of this peptide. The intracellular release mechanism induced by the UBL module makes it possible to deliver peptides of any size, composition and sequence into cells, in particular having the desired amino- and carboxy-terminal regions, including having an amino acid amino-terminal other than methionine. By way of comparison, it should be recalled that peptides synthesized in situ from transgenes are necessarily synthesized with an initiating methionine in the amino-terminal position, and that the transgenes do not lend themselves well to the synthesis of small proteins and peptides. c- Finally, the intracellular dissociation between the molecule X and the vector modules makes it possible to introduce an additional device into this vectorization system: the subsequent destruction of all the modules of the vectorized molecule other than the transported molecule X ( CPM, UBL and others). The scheduled destruction of these modules is the second objective of the present invention. In the simplest case where all of the vector components are peptide in nature, this objective is achieved by the introduction of a module into the vectorized molecule.

Several categories of destabilization modules can be used for this system, in particular destruction boxes ("destruction boxes") present in certain naturally unstable proteins such as, for example, swans (Glotzer et al.

1991. Nature 349, 132-138), or else an amino-terminal end exposing a destabilizing amino acid followed by a region comprising one or more lysines.

Preferably, it is the latter type of destabilization module that is used. The N-terminal destabilization module (DN, "Destabilizing N-end") is located at the amino-terminal end of the vectorized molecule.

The nature of this type of N-terminal destabilization module is based on the observation that the half-life of intracellular proteins is very precisely determined by the nature of the N-terminal sequences and mainly by the nature of the first N-terminal amino acid (Bachmair et al. 1986. Science 234, 179-186). We propose to exploit this natural activity of eukaryotic cells in a new and original way, to program the intracellular destruction of a protein pre-synthesized outside the cells, then incorporated by the cells.

Depending on the DN chosen, it is possible to program the destruction, after the release of the transported molecule X, of the rest of the recombinant protein (comprising at least DN-CPM-UBL).

It is important to note that DN destabilization signals only take effect in the intracellular context, and therefore in no way affect the stability of the recombinant protein in extracellular fluids. This point is demonstrated by the existence of many extracellular proteins which are naturally very stable although having a destabilizing amino acid at their amino-terminal end, exposed after the chvage of the signal peptide secretion.

We propose several methods to expose destabilizing amino acids at the N-terminal end of the vectorized protein:

A- By enzymatic cleavage of the recombinant protein by a protease: - Or a protease, the recognition site of which is located exclusively on the NH ₂ side of the storage site (such as factor Xa for example), having provided a destabilizing residue just downstream from its cleavage site (FIG. 3 A).

- or a protease cutting inside its recognition pattern, but chosen because the amino acid forming part of its recognition site and located just downstream of its cleavage site, happens to be a destabilizing residue.

B - By chemical cleavage, for example with cyanogen bromide (CNBr), at the level of a methionine of the purified recombinant protein, having provided for a destabilizing amino acid just downstream in the vectorized molecule. A prerequisite for this method is that the recombinant protein is designed so that it does not contain other methionines (Figure 3B).

C - By enzymatic elimination of the N-terminal initiating methionine from the purified recombinant protein, by an appropriate memionine aminopeptidase (MAP)

(Figure 3C). D - Alternatively, the exposure of an N-terminal destabilizing amino acid can be carried out not during the preparation of the recombinant protein as above, but by the receptor cells themselves. For this, an additional UBL module is inserted into the recombinant protein just upstream of the DN (Figure

3D). This method is complex but offers the advantage of guaranteeing that the destruction of the vector module will only occur after the C-terminal chvage of UBL, and therefore after the release of the transported molecule X. In this case, it is possible to choose a destabilizing amino acid dictating a very short protein half-life (such as arginine for example).

Preferably, methods A, B and D are used. In the case of method B, the presence of internal methionines in the recombinant protein must first be avoided. In particular, the methionine corresponding to the first amino acid of UBL is replaced by a leucine.

We can ensure that the degradation of the vector module will be well delayed compared to the release of the transported molecule, in two ways: a- by the choice of a DN and in particular of an N-terminal amino acid, inducing the destabilization in a time significantly greater than the time of action of the C-terminal hydrolases from PUBL. b- by the DN exposure strategy by cleavage of a UBL placed upstream according to point D of the methods for inserting a DN, illustrated in Figure 3D. In this method, the additional UBL module is not destroyed in the cells, but is completely harmless insofar as it is identical to normal endogenous cellular UBLs.

Several molecules are known that can be used in the context of the present invention to constitute the UBL module, such as ubiquitin, SUMO, NEDD-8, UCRP, RUB or others (see Jentsch and Pyrowolakis, 2000. Trends Coll Biol. 10, 335-342). As already indicated above, the use of modified or shortened versions of UBL in the context of the present invention is also possible if their capacity to induce cleavage at their C-terminal end is preserved.

Preferably, this module is ubiquitin. Several methods are known that can be used to constitute the intracellular destabilization module of the vector.

Preferably, this module consists of an amino-terminal destabilizing acid followed by a region comprising one or more lysine residues. Preferably, the amino-terminal amino acid destabilizer is glutamine. It will also be noted that in the context of the present invention, said incorporation module is preferably a membrane translocation peptide such as a peptide derived from the TAT protein of the AIDS virus (HIV), of natural or modified sequence.

The invention also relates to any vectorized molecule, characterized in that it is constituted by the fusion of a molecule (X) having at least one amino-terminal end of which said amino-terminal end is fused by peptide bond at the end carboxy-terminal of the vector as described above.

According to a variant, said molecule (X) is an amino acid or a peptide, optionally chemically modified. According to another variant, said molecule (X) is a non-peptide molecule grafted to the carboxy-terminal end of an oligopeptide.

According to another variant, an N-terminal protein destabilization module can be added to the amino-terminal end of the vectorized protein.

In the simplest case where the modules of the system are of a polypeptide nature, it is possible to produce the assembly as a recombinant protein in appropriate expression systems, prokaryotes (bacteria) or eukaryotes (for example bacculovirus). However, the use of a eukaryotic production system requires ensuring that it does not include the enzymatic activities leading to the C-terminal chvage of UBL. Eventually, these activities, if they exist, must be deactivated beforehand. These problems do not arise for bacterial systems, predominantly lacking these enzymatic activities. Alternatively, it is possible to chemically synthesize the vectorized molecule by means of peptide synthesizers.

According to a variant of the invention, said vectorized molecule also comprises at least one labeling module. Preferably, this labeling module is a polypeptide such as poly-histidine.

Such a labeling module could in particular be used to facilitate the purification of the vectorized polypeptide, for example by affinity for metals in the case of polyhistidine.

It will be noted that the integration of such a labeling module into the vector is absolutely not mandatory. It can in particular be removed after the preparation of the vectorized molecule, before its use.

According to another variant of the invention, said vectorized molecule additionally comprises at least one cell addressing module. It is indeed often desirable to vectorize molecules only in certain specific cell types. Translocation peptides are already known to be preferentially incorporated by certain cell types (Fujihara and Nadler, 1999), but most of the translocation peptides appear to be effective in all tissues (Schwarze et al. 2000. Trends Cell Biol. 10 , 290-295). However, even in this case, addressing to certain defined cells can be specified for this vectorization system, by adding additional modules to the vectorized molecule. For example, it has been shown that the presence of a tripeptide RGD (Arginine - Glycine - Aspartate), allows preferential addressing of metastatic cancer cells, overexpressing αv / β3 integrins.

Together, the specific devices of this system (release and programmed destruction of the vector) cooperate to achieve the following net result: natural peptides of arbitrary composition and sequence, are transported within the cells themselves, rid of all additional molecules, and without alteration of the organism.

The invention will be better understood thanks to the appended drawings in which:

- Figure 1 shows a vector molecule according to the invention;

- Figure 2 schematically shows the operation of the invention; - Figure 3 schematically represents methods A B, C and D to expose an amino acid particuher at the N-terminal end of the vectorized molecule (see above);

- Figure 4 shows a spectrum reflecting the incorporation of TAT-Ubi proteins coupled to FITC, into thymocytes; - Figure 5 shows a polyacrylamide gel transferred to the membrane observed on a fluorescence reader With reference to FIG. 1, a vectorized molecule is constituted by the fusion, of NH2 into COOH, of an intracellular destabilization module 2, of a cellular incorporation module 3 which can for example be constituted by peptide derived from the protein TAT of the AIDS virus (HIV), of an ubiquitin module or other related conjugable peptide (UBL) 4 and of a molecule to be transported 5. Once incorporated into the cells, the vectorized molecule is cleaved between 4 and 5 by a specific hydrolase 6. The molecule to be transported 5 can then exercise the biological activity for which it was designed. After release of the molecule to be transported 5, the rest of the vectorized molecule 2-3-4 is destroyed by an intracellular enzymatic system 7 recognizing the destabilization module 2. With reference to FIG. 2, the vectorized molecule 1 is introduced into the extracellular medium A. Thanks to the module of cellular incorporation module 2 which it contains, this molecule crosses the cell membrane C and is found in cytoplasm B. Thanks to a specific hydrolase 6, the peptide bond between molecule 5 and UBL 4 is broken and molecule 5 is released with its free amino-terminal end. This molecule can exert there the biological activity for which it was designed 8. The future of the vector l ', constituted by the fusion of the modules 2,3,4, does not have any more influence on the function the transported molecule 5. According to a variant, the presence of a destabilization module 2 induces the destruction of the vector l 'by intracellular machinery.

The invention offers a wide range of applications, from basic research to therapy. Since it does not impose any conditions on the nature of molecule X, it allows the introduction into polypeptides of both natural and artificial sequences. Natural sequence polypeptides can for example correspond to domains of existing proteins. Peptides of artificial sequences can be selected as specific ligands of known regulatory proteins, through screening strategies for random peptide banks, direct or reverse double hybrid, or even computer-aided modeling.

In the particular case where the peptides which it is desired to transport in the cells are precisely normal or mutated UBLs, such as for example the ubiquitins mutated at the level of lysine residues, the UBL module provided in the vector can be used, in which case its C-terminal extension can be any.

The invention is also directly adaptable to the screening of random peptide banks, which offer billions of possible configurations. For this, a library can be constructed by directly using the vector described above, by inserting random nucleotide sequences 3 'to the region encoding the UBL. The molecular tools defined in the present invention: intracellular release of the transported molecule X and programmed destruction of the other modules of the molecule vectorized, are adaptable to already identified translocation modules or in future developments. They are also adaptable to the administration modalities which will be defined for the cellular incorporation modules.

In the field of medicine, the present invention is adaptable to the treatment of any type of pathology, from the moment when effective transported X molecules can be defined for these pathologies.

The present invention takes advantage of duly confirmed cellular activities in various fields of biology, so that its functionality is largely assured. The functionalities of the present invention are set out below. Exposure of any amino acid after cleavage by factor Xa:

The possibility of using factor Xa to expose the desired amino acid at the N-terminal end of a recombinant protein, is guaranteed by its specificity of chvage of this protease. The cleavage of FXa is directed by the recognition of a sequence of four amino acids (preferably IEGR) entirely contained on the NH2 side of its cleavage site, which does not impose any condition on the nature of the next amino acid. Factor Xa is commercially available and widely used. Release of the molecule transported X inside the eukaryotic cells:

This release is ensured by CHVage C-terminal d'UBL. The efficiency and specificity of the enzymatic activities responsible for these cleavages are totally guaranteed by their biological importance. Indeed, the C-terminal extensions of UBLs must imperatively be eliminated to allow the later conjugation of UBLs on their substrates. Planned destruction of the vector after cell incorporation:

No exception has been reported with regard to the efficiency of intracellular degradation of proteins having a destabilizing amino acid at their N-terminal end. Like the C-terminal chvage of UBL, this mechanism is a natural process of any eukaryotic cell, ensuring normal physiological functions. For example, this mechanism would be involved in:

- the degradation of secreted proteins which are badly localized in the cell cytoplasm, or the degradation of protein waste residues.

- large-scale protein degradation in damaged neurons. In this case, intracellular ligases graft a destabilizing supernumerary amino acid (often arginine) to the N-terminal end of the proteins to be destroyed.

Stability of molecules containing a UBL and a DN in the extracellular medium: The enzymatic activities of chvage and protein destruction, requested respectively by the UBL and DN modules, exist only inside the cells eukaryotes. It is therefore excluded that the presence of these modules leads to any alteration of the vector molecules as long as they are in the extracellular medium. Membrane permeability:

The transmembrane transport properties are also duly confirmed for numerous translocation peptides according to the prior art.

Examples of manufacturing vector production systems according to the present invention are detailed below:

Several plasmids have been manufactured for the production in bacteria of recombinant proteins according to the present invention. These plasmids are only nonlimiting examples of application of the present invention.

- First plasmid.

The plasmid described below allows the production in bacteria of a recombinant protein called "base", comprising all the modules of the vector, but not of molecules transported X. The production of recombinant proteins for the transport of proteins or peptides X , requires the insertion into this plasmid of the nucleotide sequence encoding these proteins or peptides, according to conventional techniques of nucleotide cloning.

The plasmid is made from the construction pQE-30 sold by the company QIAGEN for bacterial protein production.

In this construct, the coding sequence for ubiquitin was imported from

Xenopus laevis and not mammals, for better use of codons in bacteria.

The species of origin of the ubiquitin coding sequence can be freely chosen since it does not affect the derived amino acid sequence (ubiquitin is 100% conserved between vertebrates).

This plasmid makes it possible to produce in bacteria according to standard protocols, a basic protein, purified by affinity on Nickel. The amino acid sequence of this protein is as follows, in letter code: l-MRGSHHHHHHGSKLIEGRQLGYGRKKRRQR-30 31-RRGGSASSHMQIFVKTLTGKTITLEVEPSD-60 61 -TIENVKAKIQDKEGIPPDQQRLIFAGKQLETLDHL 91-DHL 91-DHL

The "poly-histidine tag", used to selectively purify the recombinant protein by affinity for metals, is the <RGSHHHHH> region. In addition, there are specific commercial antibodies directed against this epitope (QUIAGEN). The recognition motif for factor Xa protease is <IEGR>. The chvage site being located on the carboxy side of the residue R, the cleavage by factor Xa leads to exposing the amino acid glutamine <Q> at the end of the residual recombinant protein. The destabilizing function of this amino acid can take effect due to the presence of lysine K residues nearby.

The translocation peptide is the TAT peptide derived from the HIN virus <YGRKKRRQRRR>

The UBL module is ubiquitin <MQIFVKTLTG KTITLEVEPS DTIEΝVKAKI QDKEGIPPDQ QRLIFAGKQL EDGRTLSDYΝ IQKESTLHLV LRLRGG> The transported molecule X is the dipeptide <AC>, without biological interest and intended to be replaced by additional peptides or proteins in peptides or proteins the plasmid. The carboxy-terminal cysteine residue of this dipeptide is provided in the event that it is envisaged to graft on its thiol group, a non-peptide molecule according to claim 14. No other cysteine exists in the basic recombinant protein, so as to facilitate such a coupling reaction. Note that alanine <A> is a stabilizing amino acid.

In the example chosen, the presence of a glutamine at the amino-terminal end will fix a half-life for the recombinant protein at approximately 10 minutes after its introduction into the cells, while the release of X is rehosed between 1 and 2 minutes after cellular incorporation.

A method for making a recombinant protein where the dipeptide AC is replaced by any peptide or protein is set out below.

The sequence encoding this peptide or protein is prepared in the form of double-stranded DNA having at the 5 ′ end of the coding region a cut with a blunt end, and on the 3 ′ side a cohesive end corresponding to one of the enzymes restriction present in the plasmid (in this case Kpn I, Sal I, or Pst I). The plasmid is prepared by double cutting on the one hand with the enzyme Sfo I, and on the other hand with the enzyme whose cohesive end corresponds to that chosen for the insert. The Sfo I double-stranded cleavage site is located precisely 3 'from the last codon of ubiquitin (marked v on the diagram below). The following diagram represents the sequences encoding the COOH end of the basic protein:

Sfol

5 'GGT GGC GCC TGT TGA GCT C 3'

3 'CCA CCG CGG ACA ACT CGA G- -5'

NH2 ... GGAC * The restriction enzyme cleavage sites used for cloning are marked (v). In the example chosen, the first site (Sfo I) releases ends with blunt ends, and the second site releases ends with cohesive ends. The amino acids GG co-esponde at the COOH limit of ubiquitin. Dipeptide AC is the C-terminal extension of ubiquitin in the base protein. * represents a "stop" codon (end of translation).

The following diagram shows the insertion of the sequence encoding any protein in C-terminal fusion with ubiquitin.

Sfol

5 'GGT GGC NNN NNN NNN NNN N-

3 'CCA CCG NNN NNN NNN NNN N- • 5'

NH2 ... G G X X X X -

After cutting Sfo I, the inserted nucleotide sequence, represented with "N" s, is linked to the coding sequence of ubiquitin. This ligation allows the production of proteins X in C-terminal fusion of ubiquitin, provided that care has been taken to preserve the translational reading phase between the coding sequence of ubiquitin and that of the C-terminal peptide.

The insert encoding molecule X may come from another plasmid, from a PCR reaction, or else from a nucleotide synthesis, in the case of peptides of reasonable size.

Second plasmid.

A variant construct was fabricated, where the wild-type TAT peptide is replaced by a modified version according to Ho et al. (2001. Cancer Res. 61, 474-477), the cell permeability properties of which are improved, and the amino acid sequence of which is YARAAARQARA.

This plasmid codes for the following protein: l-MRGSHHHHHHGSIEGRQKYARAAARQARAG-30 3 l-SASSHMQIFVKTLTGKTITLEVEPSDTIEN-60 6 l-VEAKIQDKEGIPPDQQRLIFAGKQLEDGRT-90 91 -LSD YNIQKESTLHLV

Finally a third version was built, or the methionine residue corresponding to the first amino acid of ubiquitin, is replaced by a leucine (the ATG codon has been replaced by CTG). This construction is intended to test the possibility of exposing an amino-terminal amino acid destabilizer by removal of the initiating methionine by the reaction to cyanogen bromide, according to a variant of the method presented in FIG. 3 B. By way of illustration of the present invention, the experiments detailed below were carried out.

Recombinant proteins "poly-Histidine-TAT-Ubiquitin-C-terminal extension" were efficiently incorporated by cells reputed to be refractory to conventional transfection techniques: normal thymocytes in primary culture, extracted from 4-week-old mice.

The correct incorporation has been visualized in two ways:

- By immunological detection of the recombinant protein in cell extracts previously incubated with the recombinant proteins and then washed. This detection was carried out by the western blot technique using primary antibodies directed against the RGSHHHHHH peptide (sold by the company QUIAGEN).

- By incorporating fluorescence. The recombinant "poly-Histidine-TAT-Ubiquitin-C-terminal extension" proteins were covalently coupled to fluorescein (FITC), then freed from free fluorescein. After incubation of mouse thymocytes with these recombinant fluorescent proteins, the cellular incorporation of the fluorescence was measured by fluorimetry (FACS Cahbur Becton Dickinson). This measurement made it possible to verify that all of the thymocytes incorporated the protein, in a homogeneous and rapid manner.

The intraceUular release of the C-terminal extension of ubiquitin was verified as follows: Proteins extracted from thymocytes previously incubated with the protein "poly-Histidine-TAT-Ubiquitin-C-terminal extension" labeled with FICT, were subjected to electrophoresis. Examination of this fluorescent electrophoresis gel revealed that the fluorescent ubiquitin has been covalently conjugated on cellular protein substrates.

- This result confirms the cellular incorporation of the recombinant protein, since peptide conjugation calls for intracellular machinery.

- This result confirms the C-terminal ubiquitin chvage and therefore the release of its C-terminal extension. Indeed, the conjugation of ubiquitin is only achievable after exposure of its authentic C-terminal end, terminated by a GG dipeptide (Gly-Gly). These experiences will be better understood thanks to the following figures Figure 4: Thymocytes extracted from 4-week-old mice were incubated or not (control) with FITC alone, or TAT-Ubi protein coupled to FTTC, for 15 minutes. After washing, the cells were analyzed by flow cytometry. The spectrum represented in FIG. 4 shows that the protein is rapidly and homogeneously incorporated by the cells of the population.

Figure 5: To demonstrate intraceUular chvage of the recombinant protein at the C-terminal end of the ubiquitin module, thymocytes in primary culture were incubated or not with recombinant proteins TAT-Ubi- C-terminal extension coupled at FITC. Protein extracts were prepared from these cells, then subjected to a denaturing electrophoresis in polyaciylamide gel.

(SDS-PAGE, now only covalent bonds). The content of the gel was transferred to the membrane and observed on a fluorescence reader (Storm scanner, Molecular

Dynamics) The result obtained is shown in Figure 5 on which we can distinguish different channels. Channel 1. Recombinant protein TAT-Ubi-Extension C-terminal produced in bacteria and purified by affinity for Nickel, then coupled to fluorescein

(FITC). The recombinant protein is clearly visible (RP). Some contaminating bacterial proteins (C) of high molecular weight are visible. Channel 2. Proteins extracted from control thymocytes.

Channel 3. Proteins extracted from thymocytes incubated with the recombinant protein preparation shown in channel 1. A size band equivalent to RP (arrow) is clearly detected in the cells, but the majority of the fluorescence is found in protein proteins. highest molecular weight (CS). These bands must correspond to conjugation substrates of fluorescent ubiquitin derived from the recombinant protein. The very intense band corresponds to an unidentified protein, predominant in thymocytes. This result shows that:

- the recombinant protein is effectively incorporated by thymocytes - the recombinant protein is effectively cleaved at the C-terminal end of the ubiquitin domain. Indeed, the conjugation of ubiquitin on intracellular protein substrates requires that the authentic end of ubiquitin be exposed and stripped of its C-terminal extension. REFERENCES

Bachmair A, Finley D, Varshavsky A. 1986. In vivo half-life of a protein is a function of its amino-terminal residue. Science 234, 179-186 Fujihara, S.M., and Nadler, S.G. (1999) Intranuclear targeted delivery of functional

NF-kappaB by 70 kDa heat shock protein. EMBO J. 18, 411-419.

Futaki, S., Suzuki, T., Ohashi, W., Yagami, T., Tanaka, S., Ueda, K., and Sugiura, Y. (2000). Arginine-rich peptides: An abundant source of membrane-permé peptides having potential as carriers for intraceUular protein delivery. J. Biol. Chem. 276, 5836-5840.

Glotzer M, Murray AW, and Kirschner MW. (1991). Cyclin is degraded by the ubiquitin pathway. Nature 349, 132-138.

Ho A, Schwarze SR, Mermelstein SJ, Waksman G, Dowdy SF (2001). Synthetic protein transduction domains: enhanced transduction potential in vitro and in vivo. Cancer Res. 61, 474-477.

Lund et al. 1985. Nucleotide sequence analysis of a cDNA encoding human ubiquitin reveals that ubiquitin is synthesized as a precursor. J. Biol. Chem. 260, 7609- 7613

Mayer and coU. 1989. Detection, resolution, and nomenclature of multiple ubiquitin carboxyl-terminal esterases from bovine calf thymus. Biochemistry 28, 166-172

Pickart and Rose 1985. Ubiquitin carboxyl-terminal hydrolase acts on ubiquitin carboxyl-terminal amides. J. Biol. Chem. 260, 7903-7910.

Schwarze, S. R., Hruska, K. A., and Dowdy, S. F. (2000). Protein transduction: unrestricted delivery into ail cells? Trends CeU Biol. 10, 290-295

Claims

1. Vector for the introduction of a molecule (X) having at least one ajtτώ o-terminal end in a eukaryotic cell characterized in that it is constituted by the association of a ceUular incorporation module allowing the translocation membrane, and a UBL module, that is to say an ubiquitin module or other related conjugable peptide or a modified or shortened version of UBL, said vector having a carboxy-terminal end intended to be fused by peptide bond at the amino terminus of said molecule (X)

2. Vector according to claim 1 characterized in that it contains an intraceUular destabilization module.

3. Vector according to claim 2 characterized in that said intracellular destabilization module is constituted by a "destruction box" or by an amino-terminal end exposing an destabilizing amino acid followed by a region comprising one or more lysines.

4 Vector according to claim 3 characterized in that said amino-terminal amino acid is glutamine.

5. Vector according to claim 3 characterized in that a destabilizing amino acid is exposed at the N-terminal end after cleavage by a protease.

6. Vector according to claim 5 characterized in that said protease is the factor EXa.

7. Vector according to any one of claims 1 to 6 characterized in that it contains a labeling module.

8. Vector according to claim 7 characterized in that said labeling module is a polypeptide.

9. Vector according to claim 8 characterized in that said labeling module is poly-histidine.

10. Vector according to any one of claims 1 to 9 characterized in that said incorporation module is a membrane translocation peptide.

11. Vector according to claim 10 characterized in that said incorporation module is a peptide derived from the TAT protein of the AIDS virus (HIV).

12. Vector according to any one of claims 1 to 11 characterized in that the UBL module is an ubiquitin module.

13. Vectorized molecule characterized in that eUe is constituted by the fusion of a molecule (X) having at least one amino-terminal end of which said end amino-terminal is fused by peptide bond to the carboxy-terminal end of the vector according to any one of claims 1 to 12.

14. Vectorized molecule according to claim 13 characterized in that said molecule (X) is an amino acid or a peptide.

15. Vectorized molecule according to claim 14 characterized in that said molecule (X) is an amino acid or a chemically modified peptide.

16. Vectorized molecule according to claim 15 characterized in said molecule (X) is a non-peptide molecule grafted to the carboxy-terminal end of an oligopeptide.