EP1668037A1

EP1668037A1 - Interacting polypeptide comprising a heptapeptide pattern and a cellular penetration domain

Info

Publication number: EP1668037A1
Application number: EP04787492A
Authority: EP
Inventors: Pierre Jalinot; Armelle Roisin; Jean-Philippe Robin
Original assignee: Centre National de la Recherche Scientifique CNRS; Ecole Normale Superieure de Lyon; Ecole Normale Superierure de Lyon
Current assignee: Centre National de la Recherche Scientifique CNRS; Ecole Normale Superieure de Lyon
Priority date: 2003-09-30
Filing date: 2004-09-30
Publication date: 2006-06-14
Also published as: FR2860237B1; WO2005033147A1; US7709606B2; JP2008500267A; CA2540520A1; JP4927546B2; FR2860237A1; US20060240516A1

Abstract

The invention relates to an interacting polypeptide consisting of or comprising a heptapeptide pattern of sequence X1X2X3X4X5X6X7 and a transduction domain, characterized in that it is a chimera polypeptide, the amino acid X7 is located between 5 and 35 amino acids of the C-terminal end of said polypeptide, and that the domain (b) is situated in C-terminal relative to pattern (a). The invention also relates to screening methods for identifying interacting polypeptides capable of modifying the phenotype of a cell and to uses of interacting polypeptides as mentioned in phenotypic screens or for therapeutic purposes. Lastly, the invention concerns interacting polypeptides capable of modifying the function of the HIV-1 Rev viral protein.

Description

INTERACTION POLYPEPTIDE COMPRISING A HEPTAPEPTIDE PATTERN AND A CELL PENETRATION DOMAIN

The present invention relates to interaction polypeptides comprising a heptapeptide motif and a cell penetration domain, and to screening methods for the identification of interaction polypeptides capable of modifying the phenotype of a cell and for the identification of other molecules capable of interacting with an intracellular target. The present invention also relates to uses of the interaction polypeptides as mentioned in phenotypic screens or for therapeutic purposes. Finally, the present invention is directed to particular interaction polypeptides capable of modifying the function of the viral protein Rev and their uses in screening methods to identify other molecules capable of interacting with Rev.

Peptide interactions are responsible for an important part of intra and intercellular communication; it is through a complex set of interactions between a protein and its ligand that the cell functions and adapts to its environment. When new proteins are identified, this knowledge may seem, at first, devoid of interest as long as the partners of this new protein are not, in turn, identified, which makes it possible to determine the cellular function of the protein. .

In addition, a protein is, in most cases, capable of interacting with different partners. From these interactions can arise either the same function vis-à-vis each of the partners (for example a protein phosphorylating different peptide ligands), or else perfectly distinct functions can thus arise. This is particularly the case for membrane proteins, which independently of their natural function within the cell, also serve as a gateway for viruses. The function of any given protein is therefore largely determined by the peptides or proteins with which this protein interacts.

Protein interactions are generally the combination of local interactions between very specific sites of the protein (some amino acids) thanks to different bonds, such as electrostatic bonds, hydrogen, and repulsive interactions due to spatial congestion. A given protein can interact specifically with a protein domain so that this protein can interact with a whole class of other proteins that share that same domain.

Although for a given protein we generally know only a limited number of natural peptide partners, its capacity for interaction is much greater. Indeed, a given protein can have many sites capable of generating interactions and which are accessible to other peptides. The determination of new interaction partners can therefore allow either to modify (disturb, prevent, improve, change the kinetics, the specificity, ...) of interactions, or to mimic existing interactions, or to bring to light new interactions.

The identification of new interaction partners is in general! performed in two different situations. In the first case, we are looking for a partner for a specific target, intracellular or extracellular. In such a case, it is possible to develop a particular polypeptide capable of interacting with the target. The sequence of this polypeptide is determined by taking into account in particular the structure of the target protein with which it must interact, the distribution of charges, the attractive and repulsive forces. This step is carried out most of the time by molecular modeling. This approach requires a very thorough knowledge of the target protein and in particular of its three-dimensional structure. It is also possible to perform a screen to test a very large number of potential polypeptides against the target. The screen must make it possible to demonstrate an interaction between one of the polypeptides tested and the determined target. A screen generally used in this case is the two-hybrid screen ("two-hybrid system") in yeast (US Pat. No. 5,580,736), where the determined target and the potential polypeptide are expressed simultaneously.

In a second case, the target protein is not determined, that is to say new interaction partners can also be screened for their ability to modify a given phenotype of the cell, without the target protein with which they interact with is known. In this case, phenotypic screens are used which consist in making the cell express different polypeptides and in identifying those capable of modifying the phenotype in the expected direction (US Pat. No. 6,153,380). A variant of this situation consists in screening new interaction partners capable of preventing, destroying, modifying or destabilizing an interaction between two other proteins. Also in this case, phenotypic screens are generally used. The mode by which the partner acts on the interaction between the two other proteins is not necessarily known.

For the construction of polypeptides to be used in screens as mentioned, numerous constructions have been proposed, characterized in particular by the simultaneous presence of a fixed protein sequence (sometimes called platform) into which a motif is inserted. whose sequence is random (see patent application WO96 / 02561). By varying this sequence, it is therefore possible to generate a whole family of polypeptides and to test them successively for their ability to interact within the cell. When this approach is used to determine new interaction partners of intracellular proteins, it is generally resorted to cloning and transformations which make it possible to express the construct to be tested in the cell where the protein of which a new one is sought is expressed. interaction partner. In the therapeutic field, this means that, if a potential partner is detected, it must then be able to make a drug which will act inside the cell, that is to say find a way to make it penetrate into the cell. Gene therapy does not normally allow this step to be carried out easily. The other solution consisting in grafting onto the interaction partner a motif promoting cell penetration has the drawback of modifying the interaction partner in an unpredictable manner. After this modification, it is not guaranteed that it is still capable of interacting with the protein of which it has been chosen to be a ligand.

Another approach is to develop protein mimes by modeling, but this method is not yet developed. In addition, we often end up with molecules which are not or hardly able to cross the cell membrane, and which therefore do not generate the interaction for which they were developed. If a motif ensuring cell penetration has to be grafted onto these molecules, then, as in the previous case, there is no guarantee that they are still capable of interacting with the protein of which they have been chosen to be a ligand. . The object of the present invention is to propose new interaction peptides having a particular construction such that a cell penetration domain is associated with the interaction partner. If such a partner proves to be, after a first screen, capable of interacting with the partner of interest, the passage from the screen to the generation of a drug does not add any new reason which would risk disturbing the interaction brought into play. evidence during the screening step. The interaction peptides according to the invention have the dual property of cell penetration, for example in lymphocytes and / or in macrophages, and of interaction with a partner.

The inventors have developed a process making it possible to produce proteins of small size having a heptapeptide motif and capable of binding to a protein target, of viral or cellular origin, and thereby inhibiting its activity. These small proteins have been designed to be able to enter cells. The interaction polypeptide of the invention may also have a stabilization domain, linkers or other components.

The present invention also provides various screening methods making it possible to obtain new polypeptides capable of modifying a given phenotype or of interacting with a given target. The present invention also provides a method of screening for chemical molecules capable of interacting with a given protein target.

One area where determining such partners is important is virus replication. Indeed, the replication of certain viruses in human or animal cells is allowed by the expression of a limited number of viral proteins, some of which have an activity essential for the realization of this process. By blocking these proteins with an interaction polypeptide defined according to the screens of the present invention, therefore capable of penetrating into cells and preventing the necessary contacts with other viral or cellular proteins, it is possible to prevent replication. of the virus. The implementation of the invention made it possible to identify partners capable of interacting with the Rev protein and in particular the Rev protein of HIV-1 and thus preventing replication of the virus. These polypeptides are also part of the present invention. The invention can also be implemented in methods aimed at identifying partners capable of interacting with different proteins implicated in certain cancers.

In the context of the present invention, the terms below are defined as follows:

Heptapeptide or heptapeptide motif: linear sequence of 7 consecutive amino acids covalently linked. Domain (peptide): peptide sequence comprising at least 5 amino acids, and which, within a sequence comprising it, can be delimited by a deletion-mutation analysis. Such a domain is generally responsible for a function or a role and is characterized by the presence of essential amino acids whose mutation results in the loss of function. As examples, mention may be made of the NLS (Nuclear Localization Signal) domain present in many nucleus proteins, the cell penetration domain, the NES (Nuclear Export Signal) domain, the protein-protein interaction domains, the catalytic domains. Domain or motif of cell penetration or transduction: peptide sequence, comprising from 5 to 35 amino acids, capable of ensuring, in vivo, ex vivo or in vitro, the penetration inside the cell of a protein containing this field.

In order to ensure cell penetration, the domain may need to be placed at one end of the protein, for example at the C-terminal end; the cell penetration function can also be indifferent to the positioning of the motif within the protein. The cell penetration function is possibly limited to a certain size of the protein, a size beyond which penetration is no longer ensured.

The pattern of cell penetration is either general to all cell types, or specific to certain membranes, for example to membranes of prokaryotic cells or to membranes of Gram + bacteria or - or to the membranes of certain human cell types, such as lymphocytes, for example primary lymphocytes, and / or macrophages.

The pattern of penetration can also ensure penetration inside the nucleus for eukaryotic cells. Stabilization domain or motif: amino acid sequence, comprising at least 30 amino acids, the secondary structure of which is stable over time and under certain stress conditions, and which has the capacity to stabilize any chimeric protein which comprises it. In particular, the structure must be insensitive to denaturation and to degradation by proteases and to stress conditions in general; this structure must also be only slightly disturbed when inserted within or at the ends of this domain. The stabilization domain is also characterized by a weak immunogenic character. Preferably, a stabilization domain is of relatively moderate size, therefore has less than 300 amino acids, preferably less than 200 amino acids. In general, a stabilization domain is in most cases a fragment of a protein naturally present within the cell, a protein which must be chosen from proteins abundant in the cell, ubiquitous, and not involved in processes of degradation.

Random sequence: sequence which is defined or constructed by a random process. In the case of a DNA sequence, the random process consists in choosing, one by one, deoxyribonucleotides, among the four possible, each having (equiprobability) or not (biased choice), the same probability of being chosen. Due to the degeneracy of the genetic code, a random DNA sequence, with equiprobability of the 4 bases, will generate a random peptide sequence, but with a bias, because the amino acids coded by several codons will be overrepresented compared to the others.

The probability that a 21 base random DNA sequence (ie 7 amino acids) does not contain a stop codon in phase is 71.5%. It can also be a randomly defined peptide sequence, that is to say that the amino acids are determined successively, by choice from among all the possible amino acids, with equiprobability or not. Bait: in a screen having the objective of determining new molecules capable of interacting with each other, the bait is the defined, determined molecule, for which a ligand is sought by means of the screen. The bait may or may not be fused to a molecule acting as a reporter.

Prey: in this same screen, the prey is the molecule which is tested for its ability to interact with a determined bait.

The prey can be fused or not with a molecule playing the role of reporter. Two-hybrid screen (or double hybrid) (“two-hybrid system”): this is a special screen, which was first developed in yeast but whose principle can be adapted to other cell types.

In a two-hybrid screen, the DNA sequence coding for the bait is fused in phase to a sequence "d1" coding for a first domain "D1". The sequence coding for the prey is in turn fused in phase to a sequence "d2" coding for a second domain "D2". The domains “D1” and “D2” are characterized by the fact that their union “D1 + D2” has a particular property or function which the elements “D1” and “D2” do not have taken separately. The meeting of "D1" and "D2" requires an external intervention to put them and keep them in contact. After translation, the possible interaction of the bait (merged with "D1"), and the prey (merged with "D2"), will bring together "D1" and "D2". The particular property or function which results from this “D1 + D2” meeting makes it possible to highlight the interaction between the bait and the prey. Chimeric polypeptide: polypeptide comprising a covalent fusion of at least two amino acid sequences which are not, naturally, contiguous within the same protein. The covalent fusion can be carried out by a direct or indirect covalent bond (via a linker). It may be two sequences originating from two proteins of the same cell, or else two proteins originating from cells of a close or distant genus, for example from different animal species, or else a sequence of a protein of eukaryotic cell and the other prokaryotic cell. It may also be the covalent fusion of a sequence of a protein with a synthetic sequence, which is not naturally present within said protein. Linker or "spacer": amino acid sequence, very short, generally comprising 1 to 10 amino acids, preferably 1 to 5, present between two domains of a polypeptide which it separates. The linkers are chosen so as not to interfere functionally with the two domains that they separate. The use of linkers is also particularly recommended to allow each domain to adopt a three-dimensional folding independently of each other.

Linkers are generally rich in amino acids glycines because they have a very small footprint at the side chain and are not very reactive. Pralines are also very often inserted in linkers for their properties promoting the formation of elbows and therefore greater independence of the two domains that the linker separates. Percentage of identity between two protein sequences: This percentage indicates the degree of identity between two amino acid sequences along the entire sequence. If the sequences considered are of different size, the% identity is expressed as a function of the total length of the longest sequence. To calculate the% identity, the two sequences are superimposed so as to maximize the number of identical amino acids by allowing discontinuities of finite length, then the number of identical amino acids is divided by the total number of acids longest sequence amines. This definition is that adopted in the context of the present invention.

According to a first aspect, the invention relates to an interaction polypeptide comprising a heptapeptide motif of sequence ^N X ₁ X ₂ X ₃ X ₄ X ₅ X6X ₇ ^c and a cell penetration domain, where Xi, X ₂ , X ₃ , X ₄ , X ₅ , X ₆ and X ₇ are amino acids. The interaction polypeptide according to the invention is a chimeric polypeptide; its “chimeric” character comes from the fusion of the heptapeptide motif and the cell penetration domain, that is to say that this chain is not naturally present in any protein.

In addition, a polypeptide according to the invention is characterized by a particular arrangement of the heptapeptide motif with respect to the cell penetration domain. In fact, the amino acid X ₇ is located between 5 and 35 amino acids from the C-terminal end of said polypeptide, preferably between 7 and 25, preferably between 9 and 20, for example 12 to 15, and the domain of cell penetration is located in C-terminal with respect to the heptapeptide motif. The transduction domain can be placed at the C-terminus of the interaction polypeptide; it can also be arranged in a C-terminal relative to the heptapeptide motif, without however being at the C-terminal end, for example it is followed by a His-Tag or another motif allowing the purification of the interaction polypeptide or well its detection. The heptapeptide motif present in the interaction polypeptide according to the invention is a sequence of 7 amino acids. Preferably, the amino acids are chosen from the 20 naturally occurring amino acids. However, it can also, in the context of the invention, be incorporated modified amino acids. The heptapeptide motif is, for example, generated by a random process. It can also be encoded by a DNA sequence which has itself been generated. by a random process. Conversely, the heptapeptide motif can be perfectly chosen and determined, possibly after a molecular modeling step when the interaction polypeptide is screened for its ability to interact with a determined protein. The length of 7 amino acids for the motif is particularly advantageous, in particular when the motif is coded by a DNA sequence generated by a random process. Indeed, when the DNA sequence is generated randomly, this can lead to obtaining "stop codons" (TAA, TAG and TGA). A length of 7 amino acids, or 21 nucleotides, makes it possible to obtain a relatively small occurrence of “stop” codons and of unnecessary sequences while being long enough to obtain a domain capable of interaction. The number 7 for the length of the heptapeptide motif therefore appears to be an advantageous compromise. Another advantage of number 7 is that a sequence of this length is unlikely to be immunogenic. Indeed, it is known that the antigens presented by the major histocompatibility class I complex (MHC-class I) have a minimum length of 8 to 9 amino acids, therefore greater than the length of the heptapeptide. An interaction polypeptide according to the present invention therefore comprises at least 12 to 42 amino acids. It is generally accepted that polypeptides of this size can sometimes be unstable inside the cell and that they are preferred targets for proteases, with the exception, however, of a few peptides such as antimicrobial peptides. Therefore, an interaction polypeptide according to the present invention is preferably coupled to a molecule which will stabilize the whole. Preferably, such stabilization is carried out by integrating a stabilization domain into the sequence of the interaction polypeptide. The stabilization domain is characterized by a stability over time of the polypeptide comprising it greater than that of the same polypeptide without stabilization domain. It is also characterized by its stability with respect to stress conditions, for example denaturation and cleavage conditions which can occur in vitro or in vivo. A given stabilization domain can also be chosen for its stability in certain particular biological media such as the intestinal medium or the serum medium, for example so that the interaction partners containing such a domain are stable in the event of ingestion, passage in the stream blood. Regardless of its stabilizing role, a stabilization domain can also have characteristics allowing its detection. Another advantage of the stabilization domain is its capacity to be produced in bacteria, or in another organism allowing the production of recombinant proteins, in large quantity and in stable form.

Among the preferred stabilization domains are natural protein fragments. Indeed, certain proteins are particularly abundant in the cell, ubiquitous, non-immunogenic and stable. Fragments of such proteins are then particularly preferred for use as a stabilization domain. This is particularly the case for proteins of the ubiquitin family (ubiquitin-like).

When the interaction polypeptide is intended to be introduced into a given cell, it is particularly advantageous to choose as a stabilization domain a fragment of a protein present in said cell, or else a protein belonging to the species of which leaves the cell. In the context of the present invention, it is therefore advantageous to choose the stabilization domain from the protein fragments present in human or animal cells. The inventors have in particular shown the very interesting properties, as a stabilization domain, of a member of the ubiquitin family, truncated in its C-terminal part in order to suppress the digiycin motif as well as all the sequences which are found downstream of this pattern.

A particularly preferred protein in the context of the present invention is the protein homologous to Ubiquitin, called SUMO-1. A fragment of the SUMO-1 protein which is particularly suitable for implementing the present invention is the fragment illustrated in FIG. 6 (SEQ ID No. 1), where the digiycin motif and all the sequences downstream have also been truncated. For use as a stabilization domain in the context of the present invention, any sequence having at least 80% identity, preferably at least 90%, or at least 95% identity with the previously mentioned sequence is particularly advantageous .

Other proteins or protein fragments are also adapted to this “stabilization domain” function, in particular chaperone proteins, such as shockeat shock proteins' (HSP). An alternative to using a stabilization domain is to cyclize the interaction partners, to obtain the same stabilization effect. By cyclization, is meant the peptide fusion of the N-terminal part, upstream of the heptapeptide domain, with the C-terminal end of the interaction polypeptide.

According to the present invention, it is envisaged that very short amino acid sequences, called linker or “spacer”, are present either between the stabilization domain if there is one and the heptapeptide motif, or between the heptapeptide motif and the transduction domain. The main role of these linkers is to link the different domains of the polypeptide.

This role is therefore for example to isolate the domains in such a way that there is no interaction between the domains, for example simply by separating them by a sufficient number of amino acids not interacting either with a domain , nor with the other. The two domains can then adopt a folding which is only little or not even influenced by the presence of the other domain.

The role of the linker can also be to impose a particular positioning of a domain with respect to the other, in particular by forming a bend, which also generally has the consequence of isolating one domain from the other.

Finally, the field can also include amino acids which are known to be sites of cleavage by certain proteases, preferably intracellular.

This approach allows, after cellular penetration, to separate the two domains by clicking at the location of the linker. Conversely, a linker can be chosen such that it is particularly resistant to proteases, in order to avoid for example the potential cleavage of the cell penetration domain before the penetration step.

Preferably, a linker according to the invention mainly comprises slightly reactive and space-saving amino acids, that is to say amino acids whose side chain contains only a few atoms, this is particularly the case for the amino acids glycine and praline. In addition, proline generally forms an elbow within the chain in which it is inserted, this property can be exploited to isolate one area from the other.

Preferably, a linker according to the present invention comprises 5 amino acids or less than 5 amino acids.

A linker according to the invention makes it possible to introduce a degree of flexibility between the different functional parts of the protein.

A linker particularly suitable for the implementation of the present invention comprises less than 5 amino acids and at least about 20% of amino acids Glycine or Proline, preferably at least 50%. Depending on a particular situation, a linker according to the present invention contains only amino acids chosen from glycine and proline. For example, a linker according to the invention can have the sequence GGGG or PG, where G signifies Glycine, P signifies Proline, and the sequences are written according to the conventional convention from the N-terminal end to the C-terminal end.

An interaction polypeptide according to the invention comprises a cell penetration domain of 5 to 35 amino acids which is in its C-terminal part; this domain of cell penetration allows the penetration of the polypeptide inside the cell. Different areas identified in the literature can be used for this purpose. Among the areas particularly preferred in the context of the present invention is the area of penetration of the HIV-Tat protein, or else areas of other viral strains having the same activity. The cell penetration domain of the HIV-Tat protein is characterized by the following sequence: ^N RKKRRQRRR ^c (SEQ ID No. 9). where the one letter code to represent amino acids was used. It is also envisaged, in the context of the present invention, to use a cell penetration domain corresponding essentially to the preceding sequence where a few amino acids have been mutated, in particular in order to improve the penetration properties or else to increase the resistance to proteases or decrease a possible immune response.

The intracellular penetration domain of the interaction polypeptide according to the invention can also comprise a polyarginine motif of sequence RRRRRRR, RRRRRRRR or RRRRRRRRRR (7 to 9 amino acids arginine) whose penetration role has been shown in particular for lymphocytes (Wender and al, 2000).

Another transduction domain also envisaged in the present invention has a sequence partially corresponding to that of the Tat transduction domain but incorporating certain modifications: RRKARRQRRR (SEQ ID No. 21). A particular construction for an interaction polypeptide according to the invention consists in placing the amino acid X ₇ of the heptapeptide motif between 10 and 30 amino acids of the C-terminal end of the polypeptide, preferably between 12 and 28, preferably between 15 and 25 amino acids from the C-terminus. For the choice of the stabilization domain of an interaction polypeptide according to the present invention, a particularly suitable domain is the protein SUMO-1. Of preferably, the SUMO-1 protein fragment, the sequence of which is illustrated in FIG. 6 (SEQ ID No. 1), or any domain sharing at least 80% identity, preferably 90% d, is used as the stabilization domain. identity with this sequence. Another field which can be used in the context of the invention is the fragment of ubiquitin defined by the sequence illustrated in FIG. 6.

Other domains or elements advantageously form part of an interaction polypeptide or are grafted onto such a polypeptide. In particular, it is particularly advantageous to incorporate into the sequence of the polypeptide an addressing sequence, or else a sequence facilitating its detection and monitoring, for example of its localization or of its degradation. As a sequence facilitating detection, it is possible to envisage sequences having an easily detectable enzymatic property, or Dien a reactivity with a determined antibody, or else fluorescent properties. It is also possible to graft or incorporate into the sequence of a polypeptide a sequence facilitating its purification, for example by providing a poly-histidine or His-Tag tail. Addressing signals making it possible to deliver the polypeptide of the invention preferentially to certain types of cells are also particularly advantageous in the context of the present invention. Particularly attractive target cells are lymphocytes, macrophages, Langerhans cells, dendritic cells, stem cells, muscle cells, etc. Such addressing signals can be part of the stabilization domain, they can also be part of a polypeptide of the invention in addition to the stabilization domain, linkers, heptapeptide and the penetration domain. Alternatively, cell penetration domains which are specific or preferential to certain types of cells can be used.

Among the sequences which are particularly preferred for the heptapeptide motif, mention may be made of the sequences having at least one of the following characteristics: X ₂ is an amino acid Tryptophan, X ₄ or / and X ₅ is an amino acid Cysteine. Other sequences according to the invention also have a leucine for the amino acid X ₆ . Preferably, the sequence of the heptapeptide motif comprises a tryptophan at X ₂ and a cysteine at X ₄ or / and X ₅ . A heptapeptide motif preferably has the 3 characteristics listed. Sequences of heptapeptide motifs which are very particularly preferred in the context of the present invention as interacting with the Rev protein of HIV-1 are the following sequences:

^N FWFCGLK ^G (SEQ ID N ° 2), ^N NWLCCLN ^c (SEQ ID N ° 3), ^N KLGCFWF ^c (SEQ ID N ° 10), ^N NLCCLWN ^c (SEQ ID N ° 11), ^N FWFCGLA ^c (SEQ ID N ° 27), ^N AWLCCLN ^c (SEQ ID N ° 25), ^N NWLCCLA ^c (SEQ ID N ° 26), ^" FWFCGAK (SEQ ID N ^Q 45), FWFCGAA (SEQ-ID N ° 46), NWACCLN (SEQ ID N ° 47), NWLACLN (SEQ ID N ° 48), AWACCLN (SEQ ID N ° 49), AWLACLN (SEQ ID N ° 50), AWLCCLA (SEQ ID N ° 51), NWAACLN (SEQ ID N ° 52) , NWACCLA (SEQ ID N ° 53), NWLACLA (SEQ ID N ° 54), AWAACLN (SEQ ID N ° 55), AWACCLA (SEQ ID N ° 56), AWLACLA (SEQ ID N ° 57), NWAACLA (SEQ ID N ° 58) and AWAACLA (SEQ ID N ° 59).

Also preferred are the 7 amino acid sequences obtained from the preceding sequences by substitution of a single amino acid. Among the sequences which are particularly preferred for an interaction polypeptide according to the invention, mention may be made of the following sequences:

FWFCGLKPGRKKRRQRRRG (SEQ ID N ° 4), NWLCCLNPGRKKRRQRRRG (SEQ ID N ° 5), FWFCGAKPGRKKRRQRRRG (SEQ ID N ° 60),

FWFCGLAPGRKKRRQRRRG (SEQ ID N ° 61), FWFCGAAPGRKKRRQRRRG (SEQ ID N ° 62), AWLCCLNPGRKKRRQRRRG (SEQ ID N ° 63), NWACCLNPGRKKRRQRRRG (SEQ ID NR 64) ° 66), AWACCLNPGRKKRRQRRRG (SEQ ID N ° 67), AWLACLNPGRKKRRQRRRG (SEQ ID N ° 68), AWLCCLAPGRKKRRQRRRG (SEQ ID N ° 69),

NWAACLNPGRKKRRQRRRG (SEQ ID N ° 70), NWACCLAPΘRKKRRQRRRG (SEQ ID N ° 71), NWLACLAPGRKKRRQRRRG (SEQ ID N ° 72), AWAACLNPGRKKRRQRRRRG (SEQ ID NAPRRKRRRRRRRRRRRRRRRRRRG) 75),

NWAACLAPGRKKRRQRRRG (SEQ ID No. 76) and AWAACLAPGRKKRRQRRRG (SEQ ID No. 77). A polypeptide of the invention therefore has the capacity to penetrate inside a cell without external intervention when it is brought into contact with the cell, it also has the capacity to interact with a protein or a protein domain which can be intracellular or extracellular. In addition, the very structure of a polypeptide of the invention provides it with great flexibility, in particular at the level of the variable part which is the motif. heptapeptide. Indeed, this motif is placed within 35 amino acids of the C-terminal part of the polypeptide, preferably less than 25 amino acids. This privileged position, in a part of the polypeptide undergoing less conformational constraints than in the more central part of the polypeptide, provides the heptapeptide with greater freedom of folding. This increased freedom is likely to translate into an increased ability to adapt to one's target. Indeed, when the stresses are low at each end of the heptapeptide, it can adopt a greater number of conformations because the energy barrier to pass from one conformation to another is weaker. This flexibility generates, for the same heptapeptide, different possible conformations, and therefore a greater probability of interaction with the target.

According to another aspect, the present invention relates more particularly to peptides capable of interacting with the Rev protein, in order to inhibit replication thereof, characterized in that they consist of or include the heptapeptide sequence ^N X ^ XaX ^ XeX? ^c , where X- ,, X ₃ , X ₄ , X ₅ , Xe, X, are amino acids independently chosen from natural or modified amino acids and X ₄ and / or X is a cysteine, W is tryptophan. Preferably X ₅ is a cysteine. The Rev protein is that of a viral strain of HIV type, for example HIV-1 or HIV-2, or else of SIV (simian immunodeficiency virus), SHIV (hybrid between human immunodeficiency virus and simienne), FIV (feline immunodeficiency virus) or any other viral strain having a protein homologous to the HIV-1 Rev protein; it can also be the equivalent Rex protein of HTLV-1, HTLV-2 or BLV. Such peptides comprise or consist of a sequence capable of interacting with the Rev protein in a double-hybrid yeast test; in addition, these peptides are also capable of inhibiting viral replication in vivo. Preferred heptapeptide sequences are the following sequences: ^N FWFCGLK ^c (SEQ ID No 2), ^N NWLCCLN ^c (SEQ ID No 3), ^N FWFCGLA ^c (SEQ ID No 27), ^N AWLCCLN ^c (SEQ ID No 25) and ^N NWLCCL ^c (SEQ ID N ° 26), FWFCGAK (SEQ ID N ° 45), FWFCGAA (SEQ ID N ° 46), NWACCLN (SEQ ID N ° 47), NWLACLN (SEQ ID N ° 48), AWACCLN (SEQ ID N ° 49), AWLACLN (SEQ ID N ° 50), AWLCCLA (SEQ ID N ° 51), NWAACLN (SEQ ID N ° 52), NWACCLA (SEQ ID N ° 53), NWLACLA (SEQ ID N ° 54), AWAACLN (SEQ ID N ° 55), AWACCLA (SEQ ID N ° 56), AWLACLA (SEQ ID N ° 57), NWAACLA (SEQ ID N ° 58) and AWAACLA (SEQ ID N ° 59). In order to test the ability of a peptide to interact with the Rev protein of HIV-1 in a double-hybrid yeast test, the following test, implemented in the experimental part, can be carried out: Bait: plasmid pLexRev (see Example 1) Prey: plasmid containing the activation domain of GAL4 in fusion with the sequence coding for the peptide to be tested, for example a plasmid derived from pGAD424.

Yeast: HF7c strain of S. cerevisiae. Example 1 provides more information on possible operating conditions for the implementation of this test.

In order to test the ability of the peptide to inhibit viral replication in vivo, the following test, implemented in the experimental part, can be carried out:

Cells: human cells, peripheral blood mononuclear

Virus: reference lymphotropic strain HIV-1-LAI (Barré-Sinόussi et al; 1983). Protocol: the cells are pretreated for 30 minutes with 5 concentrations of the peptide to be tested and then infected with the HIV-1-LAI strain. The peptide is maintained in the medium throughout the time of the culture; the cell supernatant is collected 7 days after infection and the reverse transcription activity is measured. The experimental part provides more information on the operating conditions, concentrations and buffers which can be used to carry out this test.

Polypeptides comprising the sequences ^N FWFCGLKPGRKKRRQRRRG ^c (SEQ ID No. 4), ^N NWLCCLNPGRKKRRQRRRG ^c (SEQ ID No. 5), FWFCGAKPGRKKRRQRRRG (SEQ ID No. 60), FWFCGLAPGRKKRRRRRRRRRR # 62)

AWLCCLNPGRKKRRQRRRG (SEQ ID N ° 63), NWACCLNPGRKKRRQRRRG (SEQ ID N ° 64), NWLACLNPGRKKRRQRRRG (SEQ ID N ° 65), NWLCCLAPRKKRRQRRRG (SEQ ID NR 66) 68),

AWLCCLAPGRKKRRQRRRG (SEQ ID N ° 69), NWAACLNPGRKKRRQRRRG (SEQ ID N ° 70), NWACCLAPGRKKRRQRRRG (SEQ ID N ° 71),

NWLACLAPGRKKRRQRRRG (SEQ ID N ° 72), AWAACLNPGRKKRRQRRRG (SEQ ID N ° 73), AWACCLAPGRKKRRQRRRG (SEQ ID N ° 74), AWLACLAPGRKKRRQRRRG (SEQ ID NG 75RRQR) ID No. 76) and AWAACLAPGRKKRRQRRRG (SEQ ID No. 77) are particularly preferred for use as inhibitors or modulators of the Rev protein and very particularly of the Rev protein of HIV-1.

Such peptides are capable of interacting either globally with the Rev protein, or possibly with one of its domains. ^' In fact, proteins are generally characterized by the presence of different domains with sometimes different functions within their sequence. Depending on the three-dimensional structure and the folding adopted by the protein, these domains sometimes have the particularity of being accessible to partners independently of each other.

This similar situation is found for the different epitopes of the same protein making it possible to generate distinct monoclonal antibodies, the different epitopes being domains of the protein distinct from each other, and being each accessible.

Within the Rev protein, a signal called NES (Nuclear Export Signal) has been identified, which addresses the protein outside the nucleus. This signal meets the domain definition as given in the present application. Therefore, a peptide interacts either with the Rev protein globally, or specifically with the NES domain of the Rev protein. The NES domain is defined by the sequence: ^N LQLPPLERLTLD ^c (SEQ ID No. 8) where the one-letter code to represent the amino acids is used.

According to yet another aspect, the present invention also relates to a family of interaction polypeptides according to the invention, or population of interaction molecules according to the invention, the members of the family / population being interaction polypeptides do not differentiating from each other only by the sequence of the heptapeptide motif.

Different members of the same family / population are therefore identical for their area of penetration, their area of stabilization if they have one, their possible linkers. A family / population is defined by the presence of at least two members of which only the sequence of the heptapeptide motif differs. However, a family generally comprises more than two members, in general at least 10 and preferably at least 50. Families which are particularly preferred in the context of the present invention are families comprising at least 100 members distinct, preferably 1000. It is not excluded that a family or population has a certain number of members which are identical. In the case where the heptapeptide motif consists of 7 amino acids chosen from the 20 natural ones, a family or population as defined according to the invention comprises up to 20 ⁷ distinct members.

The advantage of such a family is that only the heptapeptide motif is responsible for the different properties possessed by each member of the family, since it is the only variable between the members. Preferably, a family of interaction polypeptides according to the present invention comprises the sequence ^N ^c (SEQ ID No. 6), where the sequence ^N XιX ₂ X ₃ X ₄ X ₅ X ₆ X ₇ ^c corresponds to the sequence of the heptapeptide motif as defined in the invention and the sequence PGKKRRQRRRG (SEQ ID No. 1 2) corresponds to the sequence of a linker and a cell penetration domain, where the one letter code to represent the amino acids is used. X _{1 τ} X ₂ , X ₃ , X, X ₅ , Xe and X ₇ represent amino acidos. The heptapeptide domain is the only variable region of a family member polypeptide (or population) compared to another family member polypeptide.

Even more preferably, a family of interaction polypeptides comprises members whose sequence comprises or consists of the following sequence: (SEQ ID NO: 7), or the sequence MSDQEAKPSTEDLGDKKEGEYIKLKVIGQDSSEIHFKVKMTTHLKKLKESYCQRQG VPMNSLRFLFEGQRIADNHTPKELGMEEEDVIEVYQEQTARGGGGSX ^ ^ xsx sXe PGKKRRQRRRG X ₇ (SEQ ID NO: 78). The heptapeptide motif, corresponding to the sequence ^N X., X ₂ X ₃ X X5X ₆ X ₇ ^c is the only region different from one member of the family to another. X- ,, X ₂ , X ₃ , X, X ₅ , X ₆ and X ₇ represent amino acids. Interaction polypeptides of this family are called "SUMO-1 Heptapeptide Protein Transduction Domain" or SHP. The PGKKRRQRRRG sequence (SEQ ID No. 12) comprises the cell penetration domain required for the polypeptides of the invention. A sequence exhibiting less than 3 modifications with respect to this sequence is also a preferred sequence within the framework of the invention. By modification, one understands addition or deletion of an amino acid, or substitution of an amino acid of this sequence by another amino acid. The sequence :

MSDQEAKPSTEDLGDKKEGEYIKLKVIGQDSSEIHFKVKMTTHLKKLKESYCQRQG VPMNSLRFLFEGQRIADNHTPKELGMEEEDVIEVYQEQTARPPNPKKEIELGGGGS (SEQ ID NO 16) includes a stabilization domain for IDQ # 16). Another sequence comprising a stabilization domain is the sequence:

MSDQEAKPSTEDLGDKKEGEYIKLKVIGQDSSEIHFKVKMTTHLKKLKESYCQRQG VPMNSLRFLFEGQRIADNHTPKELGMEEEDVIEVYQEQTARGGGGS (SEQ ID NO 79). Sequences sharing at least 80%, preferably at least 90%, of identity with this sequence are also sequences which can be used in order to constitute the stabilization domain of a family of polypeptides according to the invention.

The invention also relates to all the DNA sequences coding for a polypeptide of the invention, unique or belonging to a family or population, or coding for any other fragment or domain mentioned in the present invention. It can be double stranded DNA or single stranded DNA. It is considered in the context of the present invention that a single-stranded DNA codes for a polypeptide when this sequence or indeed the complementary sequence indeed contains the coding part. The DNA sequence is in linear or circular form, for example in a plasmid.

The various RNAs which could be derived from the transcription of a DNA sequence coding for a polypeptide of the invention are also included in the invention.

An interaction polypeptide as defined in the present invention has the potential ability to interact with a protein partner, determined or not, in order to modify the behavior of the cell. Inside the cell, the presence of such a polypeptide can therefore make it possible to block certain cellular functions, in particular by playing the role of competitive inhibitors. This action has therapeutic effects in many conditions if the cellular modification consists in attenuating or preventing a harmful cellular function to the cell. It is also envisaged that the presence of the polypeptide outside the cell modifies the property of the cell by interacting with a messenger or else with a protein on the surface of the cell. Therefore, the polypeptides of the invention as defined have a privileged application in the field of therapy where they can be used as active ingredient in medicaments. They are then very often accompanied by various pharmaceutically acceptable excipients. By therapy is meant both curative and prophylactic aims.

The interaction polypeptides according to the invention specifically tested for their interaction with the Rev protein are very particularly preferred for uses in therapy against HIV infections, preferably in human therapy. The polypeptides of the invention can also be used in veterinary applications, to treat animals infected with HIV homologous viruses having a Rev protein or a strong homolog. They can also be used as a preventive measure.

Since they have been specifically designed to have a way to enter cells, namely their area of penetration, ^they can be administered in a variety of forms, without limitation. They can in particular be ingested in the form of a tablet, capsule, syrup, or else be injected muscularly or intravenously, by penetration of a lotion, a gel or an ointment. Any other form of administration can also be considered. They can also be packaged in the form of liposomes and then administered in this form.

It is not excluded that the medicament also consists of the DNA molecule coding for an interaction polypeptide of the invention. This is particularly the case when gene therapy is used. The invention also relates to cells containing a polypeptide of the invention, unique or belonging to a family, or any other fragment or domain mentioned in the present invention, and also cells containing DNA sequences encoding such polypeptides. Preferred cells are lymphocytes and macrophages, dendritic cells, Langerhans cells, stem cells, preferably such cells are mammalian and in particular human cells. Other cells are yeast cells and bacterial cells.

Cells as presented can be obtained by transformation, gene therapy, or by contacting the cell and an interaction polypeptide of the invention. Such cells have particularly interesting applications in therapy. Transformed cells, especially bacteria with a DNA coding for an interaction polypeptide according to the invention, are particularly advantageous for producing said polypeptide. The polypeptide produced in bacteria is then optionally purified and can be used as an active ingredient in a medicament.

According to another aspect, the present invention also relates to screening methods involving interaction polypeptides as described. Indeed, these polypeptides are specially designed to act against protein targets against which they have been tested. A first screening method envisaged by the invention aims to identify polypeptides which are capable of modifying the phenotype of a cell. The method comprises a step of bringing a polypeptide of the invention into contact with the cell whose phenotype is sought to be modified. This step is followed by the detection of the change in cell phenotype. These steps are then optionally supplemented by a step of determining the sequence of the heptapeptide motif included in the sequence of the polypeptide which has been tested. The term "phenotype" should be understood in the broad sense as encompassing all the morphological or functional characteristics of the cell. A modification of the phenotype can in particular be characterized by a change in the shape of the cell, a modification of the composition of the membrane, secretion of a given protein, a change in color or a change in reactivity under given conditions.

The method aims to identify polypeptides which are capable of generally modifying the phenotype of a cell, it may be an expected or sought-after modification, it may also be a surprising modification. or that was not wanted.

Indeed, according to one use of the method, it can be used in order to identify polypeptides capable of interacting with a given protein. In such a situation, the interaction of the polypeptide with its target will lead to a modification of the phenotype which was expected, for example the death of the cell, or resistance or sensitivity to an antibiotic, or to a virus, or to heat. . Optionally, the interaction can be demonstrated via a reporter, in particular a reporter gene. This reporter element is responsible for the modification of the phenotype, this modification being expected. The method can also be used to identify a polypeptide capable of interacting with a given protein group or for example with a given metabolic pathway or immune cascade. In such a situation, the modification of the expected phenotype is the modification of the function of the group or else the modification of the immune pathway or cascade. The effect of such a change is not always determined in advance.

It is also possible to use the method of the invention to screen for polypeptides for, for example, a beneficial modification to the cell, without the exact nature of the modification being fixed at the start of the screening process.

The first step of the method is characterized by bringing a polypeptide of the invention into contact with the cell. The expression “bringing into contact” is understood to mean both the action consisting in bringing the polypeptide close to the cell, the bringing into contact as well as the action consisting in introducing the polypeptide into the cell. By the expression "bringing into contact", it is also included the case where the DNA coding for the polypeptide is introduced into the cell and this DNA is then translated to give rise to the polypeptide. In this situation, it is the cell itself that helps generate the polypeptide. The first step of the method is therefore for example carried out by the extracellular addition of the polypeptide in a medium containing the cell whose phenotype is to be modified, for example the addition of the polypeptide in a fluid containing the cell. As the polypeptide according to the invention has a penetration domain, it can optionally cross the cell membrane. The method described is either an extracellular screening method or an intracellular screening method. Screening is considered intracellular if the interaction between the polypeptide and its target takes place in the cell, regardless of whether the polypeptide has been added extracellularly or introduced intracellularly. Screening is considered extracellular if, on the contrary, the interaction between the polypeptide and its target takes place outside the cell, for example on its surface.

The screening method of the invention can be implemented in vivo, or in vitro. When it is carried out in vivo, the bringing together of the polypeptide and the cell can consist of bringing together the polypeptide with a fluid comprising the cell or else a tissue containing such a cell. A particularly preferred application of the process of the invention consists in making use of a two-hybrid screen. This application is recommended when the target protein, for which a partner is sought in the form of an interaction polypeptide, is known and when its sequence is available. It is then possible to detect an interaction between the target protein and an interaction polypeptide according to the invention. Such an interaction is generally manifested by the induction of a reporter gene - the expression of which modifies the phenotype in an easily detectable manner, for example by a change in cell color. A particularly preferred application of the method is in the field of screening for new interaction partners against viral proteins, including the Rev protein. Such a protein is found in various viruses and in particular HIV. In this application, the gene coding for the Rev protein is for example cloned in a two-hybrid system to serve as bait. Such a screen makes it possible to identify polypeptides capable of rendering a cell resistant to HIV infection, by preventing or limiting viral replication in the host cell. Another application of the method consists in making use of a triple-hybrid screen. In such a screen, two peptide partners, the interaction of which is known, are cloned in a two-hybrid system and a third plasmid comprises the interaction partner according to the invention. The latter is tested for its ability to modify the interaction between the first two partners. Thanks to the reporter system of the two-hybrid screen, it is possible to observe and measure the action of the interaction partner on the interaction between the two peptide partners. A second method covered by the present invention is a method of screening molecules for the identification of one of them which interacts with a determined intracellular target.

Such a method comprises a first step of generating polypeptides of the invention. The polypeptides thus generated are placed in the presence of the intracellular target, a detection of the interactions between the polypeptide and the target is carried out, possibly followed by the determination of the sequence of the heptapeptide motif of the polypeptide.

The polypeptide can be brought into the presence of the target directly by introducing the polypeptide into the cell, or else indirectly by bringing the polypeptide and the cell into contact, the polypeptide then entering the cell. The detection step can be carried out by any technique known to a person skilled in the art. A possible detection is in particular the observation of a change phenotypic; the screen can be adapted so that the modification of the phenotype is the result of the expression of a reporter gene. Detection may also involve the assay of an entity, for example the assay of a protein, or a metabolite. The various preferred characteristics or uses mentioned in the context of the first process according to the invention are also applicable for this second process. In particular, the method can be implemented in vivo or in vitro. It is preferably implemented by making use of a two-hybrid system. Finally, a preferred application of such a screening method aims to identify interaction molecules capable of interacting with viral proteins essential for the replication of viruses, and most particularly the HIV Rev protein. A third method covered by the invention is a method for modulating the properties of an intracellular target molecule. Such a method comprises a step of bringing into contact a cell which contains the target molecule and an interaction polypeptide of the invention.

The target molecule is preferably a protein or a protein fragment comprising at least 5 amino acids, preferably at least 10. The polypeptide of the invention is specifically designed with a heptapeptide motif which is capable of interacting with the molecule target. Such a heptapeptide motif is advantageously determined by the implementation of the first screening method of the invention.

By “modulating properties” is meant all the modifications of the target which are likely to affect its function or its properties, in particular the modifications which will modify its ability to interact with its partners; but also the modifications which will lead to greater or lower stability under certain conditions, the modifications of enzymatic kinetics, of specificity or of selectivity.

As already mentioned, the various preferred characteristics or uses mentioned in the context of the first process according to the invention are also applicable for this third process. In particular, the method can be implemented in vivo or in vitro. It is preferably implemented by making use of a two-hybrid system. Finally, a preferred application of such a method aims to modify the properties of viral proteins essential for the replication of viruses, and more particularly the HIV Rev protein. According to another aspect, the invention relates to different uses. In particular, the invention comprises the use of a DNA sequence coding for a polypeptide of the invention so that the polypeptide then translated serves as prey in a phenotypic screen; preferably the DNA sequence is cloned in a two-hybrid system.

In fact, a two-hybrid system is characterized by the search for an interaction between a given molecule (protein), generally called "bait", and a potential partner to be tested, generally called "prey". A two-hybrid system generally comprises the introduction into a cell of the sequences coding for the bait, the prey and a reporter gene, linked in a specific way to other elements. It is in this context, as prey, that the DNA sequence coding for a polypeptide of the invention can be used. A particularly preferred situation in the context of the invention consists in using the DNA sequence coding for a polypeptide of the invention as prey in a two-hybrid system. In this situation, a potential bait is the Rev protein of the human immunodeficiency virus (HIV). Indeed, this viral protein is a particularly interesting target for the identification of new inhibitors. The present invention therefore envisages very particularly the use of the DNA sequence coding for a polypeptide of the invention in a two-hybrid system with the HIV Rev protein.

As indicated above, the interaction sought between the polypeptide of the invention and a given protein, for example HIV-Rev, can occur globally, or else with a particular domain of the protein. In the context of the present invention, it is therefore envisaged to clone, to serve as bait, only a sequence coding for a domain of the target protein. This strategy makes it possible to identify interaction polypeptides specific to this domain. Indeed, it has been discovered by the inventors that the potential for interaction between two given polypeptides is increased when the two polypeptides have comparable sizes. The interaction polypeptide of the invention has a fairly short sequence responsible for the interaction, these are the 7 amino acids of the heptapeptide, with possibly the combination of the interaction generated by the penetration motif. Therefore, interaction polypeptides are particularly recommended for their capacity to interact with polypeptides of comparable size, in a preferred manner, with domains of comparable size. Particularly advantageous uses of interaction polypeptides in a two-hybrid screen therefore make use, for the bait, of sequences coding for domains of 5 to 30 amino acids, preferably from 6 to 20, more preferably from 7 to 15 amino acids. Such domains are for example binding domains, addressing signals, catalytic domains.

Another example of a field particularly preferred in the context of these uses is le-NLS (Nuclear Localization Signal). Another area also preferred is the NES (Nuclear Export Signal). One such NES domain is that of the HIV Rev protein. Other “NES” signals are known having more or less similar sequences and sizes in the same order of magnitude. The two domains NLS and NES have sizes which are perfectly compatible with the optimal sizes for interaction with a polypeptide of the invention. A sequence for an NES domain is in particular that of the HIV-1 Rev protein, LQLPPLERLTLD (SEQ ID No. 8). When the HIV Rev protein is the target protein for which an interaction partner is sought, then it is particularly advantageous to clone in a two-hybrid screen a domain of the Rev protein, and preferably the NES domain of REV, to play the role of bait.

The present invention also relates to a screening method for identifying, among candidate molecules, molecules which are capable of interacting with a given intracellular target. According to this aspect of the invention, use is made of the interaction properties of the intracellular target with an interaction polypeptide according to the invention. In fact, according to this method, it is sought to detect a variation in this interaction between target and interaction polypeptide, when the candidate molecule to be tested is added. The modification of the interaction between target and interaction polypeptide reflects the potential capacity of the candidate molecule to interact with the intracellular target. In this application, the interaction polypeptide according to the present invention, capable of interacting with a given intracellular target, serves as a tool for identifying new drugs, proteinaceous or not, which are also capable of interacting with the target intracellular given. The method therefore comprises the following steps: i. bringing the target molecule into contact with an interaction polypeptide according to the invention, ii. detecting the interaction between the target molecule and the interaction polypeptide, iii. adding a candidate molecule, iv. detecting the modification of the interaction between the target molecule and the interaction polypeptide.

According to this method, it is possible to test different candidate molecules, in particular by high throughput screens. The interaction between the target molecule and the interaction polypeptide according to the invention can be detected by any suitable means, and in particular by the use of a reporter gene. It is conventionally made use of a "two hybrid" system. In such a situation, the interaction between the target molecule and the interaction polypeptide is detected using a reporter gene.

To implement the method, once the target molecule has been fixed, it may be advantageous, in a preliminary step, to identify an interaction polypeptide according to the invention capable of interacting with said target molecule.

The present invention also covers the candidate molecules identified by this method, such as protein molecules, polynucleotides, small organic molecules, for example having a molecular weight of less than 1000, lipids, carbohydrates, etc. According to a particularly preferred implementation in the context of the present invention, the intracellular target molecule is the Rev protein and the interaction polypeptide used is as defined in the present invention, preferably with one of the following heptapeptide sequences: ^N FWFCGLK ^c (SEQ ID N ° 2), ^N NWLCCLN ^c (SEQ ID N ° 3), ^N FWFCGLA ^c (SEQ ID N ° 27), ^N AWLCCLN ^c (SEQ ID N ° 25) and ^N NWLCCLA ^c (SEQ ID N ° 26), FWFCGAK (SEQ ID N ° 45), FWFCGAA (SEQ ID N ° 46), NWACCLN (SEQ ID N ° 47), NWLACLN (SEQ ID N ° 48), AWACCLN (SEQ ID N ° 49), AWLACLN ( SEQ ID N ° 50), AWLCCLA (SEQ ID N ° 51), NWAACLN (SEQ ID N ° 52), NWACCLA (SEQ ID N ° 53), NWLACLA (SEQ ID N ° 54), AWAACLN (SEQ ID N ° 55) ), AWACCLA (SEQ ID N ° 56), AWLACLA (SEQ ID N ° 57), NWAACLA (SEQ ID N ° 58) and AWAACLA (SEQ ID N ° 59). For example, the interaction polypeptide used is one of those described in the experimental part under the name SHPR. The present invention also relates to a method for the identification of molecules capable of modulating the properties of an intracellular target molecule. This method is characterized in particular by the following different stages: Firstly, nucleic acid molecules are generated comprising sequences coding for different members of a family of interaction polypeptides according to the invention. These different members differ from each other only in the sequence of the heptapeptide motif. The method also includes screening the molecules thus generated using a two-hybrid yeast system. In this screening, use is made of the target molecule, for which interaction partners are sought, as bait. The molecules generated play the role of prey.

The method further includes a step of detecting an interaction. This interaction can be manifested in particular by the expression of a reporter gene included in the two-hybrid system. A phenotypic change in the cell or the assay of an entity can also allow the detection of an interaction.

Optionally, the method also includes verifying this interaction in human cells. This step can be done in different ways. One way is to create the equivalent of the two-hybrid system but adapted to human cells. The system is advantageously implemented with a cell line. It is also possible to verify the interaction by co-immunoprecipitation or by verifying the effects generated by this interaction. In this way, the screened polypeptide is brought into contact with the cell containing the intracellular target. It is observed whether this polypeptide induces an effect on the cell due to its interaction with the target. A subsequent step also envisaged consists in producing in quantity the interaction polypeptide identified in the context of the present invention. Such production is advantageously carried out in bacteria, for example under the action of an inducible promoter capable of over-expressing the polypeptide. The polypeptide can be produced in soluble form or as an inclusion body. The production in quantity is also possible in mammalian cells, for example in Sf9 cells.

This method is particularly suitable for detecting interaction polypeptides capable of modifying the function of viral proteins, such as those of HIV: Tat, Rev. In such a case, the method for obtaining and characterizing interaction polypeptides according to the invention is carried out in several stages and can be carried out according to the following methodology, employed by the inventors:

Screening of a bank of random heptapeptides in yeast against the targeted protein:

A library of random heptapeptides in fusion with the activating domain of a transcription factor, in this case GAL4, at its N-terminal end, and the nine amino acids of the basic Tat domain at its C-terminal end has been constructed. in a yeast expression vector. The sequence coding for the protein of interest, viral or cellular is cloned downstream from that coding for a DNA binding domain, in this case LexA, in a vector allowing the expression of the protein in yeast. The basic Tat domain associated with the heptapeptide motif is used as the motif ensuring intracellular penetration. The design of the final protein sequence is such that the different domains are spaced apart by glycine or proline amino acids which ensure their separation. The yeast strain L40 which has the reporter genes His and LacZ dependent on LexA binding motifs is transformed by the vector expressing the fusion LexA target protein, which was done with the proteins LexA-Tat, LexA-Rev and LexA- NES Rev, then by the vector bank having the random heptapetides. The Tat and Rev proteins of HIV-1 are regulatory proteins essential for viral replication. A first selection of the clones is made on medium without histidine. The positive clones are then subjected to a β-galactosidase test on a filter. The clearly positive colonies are recovered and the vector containing the heptapeptide motif isolated by transformation in E. coli. Vector sequencing allows the identification of the heptapeptide motif.

Verification of the existence of the interaction in the context of mammalian cells: To ensure that the motifs selected in the yeast environment also work in the biochemical context of human cells, a test similar to the previous one is carried out in cells. HeLa. The heptapeptide motif is re-cloned from the yeast vector into a plasmid allowing expression in human cells downstream of sequences coding for the Flag epitope, a nuclear localization signal and the activating domain of the Vp16 protein. This construct is co-transfected into HeLa cells with a vector expressing the LexA-target protein fusion and a vector possessing the reporter gene SEAP (Secreted Alkaline Phosphatase) under the control of LexA binding sequences. The measurement of SEAP activity makes it possible to evaluate the capacity of the heptapeptide motif to interact with the target protein in the nucleus of human cells. The vector expressing the heptapeptide motif also makes it possible to evaluate, by immunoblot experiments, made with extracts of cells transfected by this construct, the level of expression of the fusion protein with this motif, and therefore to verify its stability in the cells. These experiments made it possible to confirm the ability of several heptapeptide motifs to interact with Tat and Rev.

Association of the heptapeptide / basic amino acid sequence of Tat with a stabilization protein: Short linear peptides sometimes exhibit poor stability in extra- and intra-cellular media. To overcome this problem, the inventors decided to associate the heptapeptide / basic amino acid sequence of Tat with a small, stable and abundant protein in human cells. They initially chose members of the Ubiquitin family. The constructions were made with Ubiquitin (Ub) itself and a homologous protein, SUMO-1. The sequences coding for these proteins were introduced into an expression vector for mammalian cells. The heptapeptide / basic amino acid sequence of Tat is introduced downstream of SUMO-1 or Ubiquitin at the digiycin motif which is lost. This digiycin motif is normally linked to the side chains of the lysines of proteins modified by these polypeptides. The correct expression of the SUMO-1-heptapetide fusion proteins specific for Tat (designated SHPT) and Rev (designated SHPR) could be verified. For Ub constructions these tests have not yet been carried out in the absence of a good antibody directed against Ubiquitin. Using these constructions, the inventors carried out functional tests to evaluate whether the SHPRs and SHPTs directed respectively against Rev and Tat were capable of preventing the function of these proteins. For Tat, the expression vectors SHPTs were co-transfected with a vector expressing Tat and an indicator construct having the HIV-1 promoter in front of the CAT sequence. Two out of six anti-Rev SHPRs show complete inhibition of the effect of Rev on a construct having the CAT sequence in association with Rev Response Element (RRE) in an intron. The RRE RNA sequence allows the binding of the Rev protein. This type of indicator construction makes it possible to evaluate the effect of Rev, which by causing the transport to the cytoplasm of the RNA not spliced allows the expression of the enzyme CAT. Production in quantity of SHPTs and SHPRs interaction peptides:

As a result of these results, the inventors have developed a system for producing large quantities of interacting peptides in order to be able to assess their capacity to penetrate cells and to inhibit the function of their target protein from the extracellular medium. The sequence SUMO-1 / anti-Rev heptapeptide / basic Tat domain was cloned into a vector allowing production in bacteria. A protocol for purification on a heparin column, then by gel filtration, was defined. The first results show that SHPRs are well produced and can be purified by this process. The production of a large batch of these two molecules makes it possible to evaluate the possible pharmacological interest of these proteins.

The whole process described here made it possible to establish the feasibility of the method of identifying ligands antagonistic to the function of target proteins. The interaction peptides obtained can be used directly as therapeutic molecules, or serve as models for developing small organic molecules mimicking their structure.

The diagram illustrated in figure 1 recapitulates the various stages and the constructions employed. The steps and constructions are given by way of example of implementation and can without major difficulty be adapted or replaced by equivalents by a person skilled in the art.

Legend of figures

Figure 1: It illustrates the different steps and constructs used for the generation of interaction polypeptides capable of interacting with a given target protein. The double arrows symbolize an interaction between two molecules. The different cell types are indicated for each step. Step 3 (functional tests) was carried out with the REV viral protein as target.

Figure 2: Figure 2 illustrates the maps of the different plasmids used for the "two-hybrid" step in yeast.

Figure 2 A: plasmid pGAD-CR which is derived from pGAD424. plasmid pGAD-CR-P which derives from pGAD-CR and includes the sequence coding for the heptapeptide motif inserted between the BamHI and Xmal restriction sites.

Figure 2B: plasmid pLex-Tat which contains the Tat cDNA inserted between the EcoRI and BamHI restriction sites of the plasmid pLex9. plasmid pLexNES which contains the sequence coding for the NES motif of Rev inserted between the BamHI and Xhol restriction sites of the plasmid pLex9.

Figure 2C: plasmid pLexRev which contains the Rev cDNA inserted between the BamHI and SalI restriction sites of the plasmid pLex9. Figure 3: Figure 3 illustrates the maps of the different plasmids used for the “two-hybrid” step in mammalian cells.

Figure 3A: plamside pSG-FNV-P. The X7 PTD motif was amplified from pGAD-CR-P, digested with the restriction enzymes SalI and Bglll and then inserted between the restriction sites Xhol and Bglll of the plasmid pSG-FNV.

PSG5LexA-Tat plasmid, derived from pSG5LexA and includes Tat cDNA.

Figure 3B:

PSG5LexA-Rev plasmid, derived from pSG5LexA and includes Rev cDNA.

Figure 4: Figure 4 illustrates the SUMO-1-P expression vector maps for functional tests. pTL1-SUMO-1-CP: the coding sequence of SUMO-1 including the restriction sites Xhol and Bglll has been inserted in pTL1. pTL1-SUMO-1-P: the sequence of the X7 PTD motif was inserted between the Xhol and BglII restriction sites of pTL1-SUMO-1-CP. Figure 5: Figure 5 illustrates the map of the plasmid pFLAG-SUMO-1-P used as expression vectors in bacteria. This plasmid was constructed by inserting the sequence coding for the SUMO-1-P motif between the Hindlll and

Bglll of the plasmid pFLAG®-MAC.

FIG. 6: FIG. 6 illustrates the wild-type sequences of 'SUMO-1' (numbered 1) and of Ubiquitin (numbered 3) as well as the sequences of the interaction peptides according to the invention comprising, as stabilization domain, a fragment of

SUMO-1 (SUMO interaction peptides numbered 2) or ubiquitin (UB interaction peptides numbered 4). The sequences of wild SUMO-1 and of an interaction peptide comprising a fragment of SUMO-1 were aligned. The star symbolizes the presence of an identical amino acid in the two sequences.

A similar alignment was made between the sequence of Ubiquitin and the interaction peptide comprising a fragment of ubiquitin.

The motif XXXXXXX in the interaction peptides represents the sequence of 7 amino acids which can be defined randomly.

Figure 7: Figure 7 is a schematic representation of proteins used as bait (A) and prey (B, C) in two-hybrid tests in yeast (A) or in mammalian cells (C). A. The baits consist of the LexA protein in fusion with Tat (amino acids aa 1 to 86), Rev (aa 1 to 116) or the nuclear export domain of Rev (aa 70 to 96). B. Yeast prey includes the GAL4 activator domain, a series of 7 amino acids of random sequence and the Protein Transduction Domain PTD. C. Prey in mammalian cells include the FLAG epitope, the NLS (nuclear localization signal) of SV40 (large T), the Vp16 activation domain of HSV and the heptapeptide + PTD module of Tat. Figure 8: Figure 8 illustrates the analysis of the heptapeptide + PTD modules of Tat, selected against Tat and Rev, by two-hybrid mammalian screening. A. HeLa cells are transfected with the reporter construct pSEAP LEX5X and pSG5LexA-Tat, alone or with pSG-FNV-P-T8, -T9, -T10, or -T24. The quantity of SEAP is measured and the average of the values obtained for two independent transfection points is represented. The error bar is half the difference between the two values. B. The same test was carried out with pSEAP LEX5X and pSG5LexA-Rev together with pSG-FNV-P-T24, R15, -R115, -R190, -N142, -N31 and -N7. The results are presented as for FIG. 8A. Figure 9: Figure 9 illustrates the inhibition of Tat and Rev activities for the SHPT and SHPR constructs. A. schematic representation of the SHP construct which includes the entire sequence of SUMO-1, in which the digiycin (GG) motif has been mutated into AR, associated at its C-terminal end with the Heptapeptide - PTD module from Tat. B. HeLa cells are transfected with the LTR HIV-CAT construct with pSG-Tat and either pTL1-SUMO-1, pTL1-SHPT-8, -9, -10, or -24. The quantity of these latter plasmids was either 0.5 μg (light gray bars) or 2 μg (dark gray bars). CAT activity is measured and the average of the values obtained for two independent transfection points is represented. The error bar is half the difference between the two values. C. In order to evaluate the activity of Rev, HeLa cells are transfected with the plasmids pDM128 and pSG-Rev together with either pTL1-SUMO-1, pTL1- SHPR-15, -115, -190, -R7, -R31 and -R142. CAT activities are shown as in Figure 9B.

Figure 10: Figure 10 shows the direct protein-protein interaction between SHPR and Rev. GST-Rev was produced in bacteria and loaded onto an agarose-glutathione column. SHPR-142, -190 as well as SHPT-8 are loaded in the column. After washing, the proteins are eluted with glutathione. An aliquot of the loading (line 1), washing (line 2) and elution (lines 3, 4, 5) fractions was analyzed by immunoblot for GST-Rev (top graph) and for SHP using antibodies against GST or FLAG, respectively. This shows that there is co-elution of SHPR-142 and SHPR-190 with GST-Rev, except that SHPT-8 does not bind to the protein on the column.

Figure 11: Figure 11 illustrates the intracellular entry of SHP. Mononuclear cells are prepared from peripheral blood, either directly (A) or after activation with PHA (phytohemagglutinin) and IL-2 (interleukin 2) (B), then incubated with 2 μM of SHPR-190. At the times indicated after the addition of the interaction peptide, an aliquot of the supernatant is analyzed by immunoblot using antibodies against FLAG (A, lines 1 to 4; B, lines 1 to 3). The cells are collected and lysed in RIPA buffer. The extract thus obtained is analyzed by immunoblotting as described for the supernatant (A, lines 5 and 6; B, line 4). In part A, the exposure times are different between lines 5 and 6 (30s) and lines 1 to 4 (5s).

These analyzes show that the protein is stable in the supernatant and also that it is present inside the cells.

In part C, Jurkat cells were incubated with 2 μM of SHPT-8, SHPR-15, SHPR-142 or SHPR-190 for one hour and 24 hours later the cells were collected and analyzed by immunofluorescence using an antibody against FLAG. Light transmission and fluorescence images for representative cells are shown. The cells exhibit a diffuse fluorescence which is not observed for the controls.

Figure 12: Figure 12 shows the inhibition of HIV-1 replication in lymphocytes and macrophages. A. mononuclear cells are prepared from peripheral blood and are activated. The cells are treated with different quantities of AZT, Indinavir, SHPR-15, -142, -190 or SHPT-8 and infected with HIV-1-LAI. Viral replication is measured by assaying the reverse transcription activity in the supernatants and the inhibition percentages are represented as a function of the amount of the compound. B. Macrophages are prepared from monocytes, treated with different concentrations of drugs and SHPs and infected with HIV-1 / Ba-L. Viral replication is measured by assaying the reverse transcription activity. The percentages of inhibition at 7 days post-infection are represented as in FIG. 12A.

EXAMPLES

Example 1: Description of the constructs and operating conditions used to identify peptides interacting with the Tat and Rev proteins of HIV-1.

HIV-1 expresses several regulatory proteins whose action on essential cellular factors ensures rapid and efficient production of viral particles. After two decades of intense research, significant progress has been made in understanding the molecular mechanisms inducing the diversion actions of these regulatory proteins. Particular interest was paid to two of them, namely the Tat and Rev proteins. Tat activates transcription of the integrated provirus by establishing contacts with the transcription factors and the TAR motif located at the 5 ′ end of the viral RNA. Rev also has the dual capacity to interact with RNA, in this case the RRE motif present in the intronic position in the 3 'part of viral RNA, and with nuclear factors of the cell. This viral protein includes both a nuclear localization signal (NLS) and a nuclear export sequence (NES) signal, and by shuttling between the nucleus and the cytoplasm, this protein allows the export and therefore the translation of non-spliced and partially spliced viral RNAs. Since the proteins Tat and Rev are small proteins which act via protein-protein interaction, the inventors therefore considered that it should be possible to inhibit their function, which is absolutely necessary for viral replication , thanks to peptide ligands (interaction peptides) capable of competitively interfering with the associations in which these proteins of interest are involved.

As it makes it possible to demonstrate interactions in an intracellular medium, and very particularly in the nucleus, the double hybrid (or two-hybrid) test represents a method of choice for a first step in the selection of such interaction peptides.

An important problem linked to the use of peptides is their intracellular penetration. In order to avoid the use of complex procedures in the context of gene therapy, the inventors added the property of cell penetration at the very screening stage, thanks to the addition of a protein transduction domain, more particularly in this case the basic domain of Tat. This method allowed the inventors to identify short peptide sequences which bind to Tat or Rev. The association of these peptide sequences with SUMO-1 in order to stabilize them, has led to small proteins, called SUMO-1 Heptapeptide Protein Transduction Domain or SHP, which are capable of effectively penetrating into the lymphoytes, and of inhibiting, for some of them ; the functions of Rev. These proteins, which have also been observed to inhibit viral replication, both in lymphocytes and in macrophages, represent potential new therapeutic agents useful in combating the generation of new viral particles. The SHPs developed against the Tat protein are called SHPT and those developed against the Rev protein are called SHPR.

I. Two yeast hybrids:

A. Random peptide bank: 1- Construction pGAD-CR:

This plasmid is a derivative of pGAD424 (Clontech). This vector was digested by

EcoRI and Bglll.

The fragment containing the linker regions, the cloning sites of the random sequence fragment and the basic Tat domain was obtained by hybridization of the following oligonucleotides: 5 'pAATTGσGTGGTGGCGGATCCGGTTTGCCCGGGAGAAAGAAGCGTAGACAAAGAAGACGTGGTTA CCCACCACCGCCTAGGCCAAACGGGCCCTCTTTCTTCGCATCTGTTTCTTCTGCACCAATTCTAGp5' (p means phosphate). This fragment was inserted between the Ecorl and Bglll sites of pGAD424

2- Construction of the library of interaction peptides (pGAD-CR-P): The fragment of random sequence was generated by hybridization of the following two oligonucleotides: 5 '-bACTCGGATCCNBT Nl ^ ^{^} ^ GGCCCCAGCGTCACb - 5',

(b means biotin) repair by DNA polymerase and digestion by the enzymes BamHI and Xmal.

The digestion product was mixed with paramagnetic streptavidin beads to remove the biotinylated ends.

The BamHI-Xmal fragment contained in the supernatant was inserted between the sites

BamHI and Xmal of the vector pGAD-CR, to give the plasmids pGAD-CR-Ps.

The ligation products were transformed into the E.coli XL1-blue strain and striated on LB agar plates containing Pampicillin. 2.10 ⁶ independent colonies were recovered and cultured for 1 hour in LB medium containing ampicillin. The plasmids are then extracted and purified according to the standard PEG procedure.

B. Bait:

1. Vector pLex9: This vector is a derivative of pGBT9 (Clontech). The DNA binding domain of

GAL4 has been replaced by that of LexA (Farjot et al., 1999).

2- pLex-Tat construction:

The repaired Ncol / BamHI fragment obtained from the vector pSG-Tat (Veschambre et al.,

1995) was cloned between the Smal and BamHI sites of pLexθ. 3- Construction of pLexrev and pLex-NES:

These constructions contain the entire Rev sequence (pLexRev) or only that of the NES (pLex-NES) in fusion with that of LexA (Farjot et al., 1999).

The double hybrid screens with either pLexTat, pLexRev or pLex-NES as bait and pGAD-CR-P as prey were carried out in the HF7c strain of S. cerevisiae as already described (Rousset et al, Oncogene, 1998, 16, 643 -654). The colonies were cultivated on a minimal medium, devoid of histidine, and were analyzed for the expression of β-galactosidase, by the assay on filter. previously described (Rousset et al, Oncogene, 1998, 16, 643-654). Plasmids pGAD-CR-P from positive colonies are recovered and the heptapeptide motif sequenced.

II. Two hybrids in mammalian cells:

A. pSG-FNV-P constructions:

These constructs are derivatives of the plasmid pSG-FNV (Desbois et al., 1996). The region containing the heptapeptide was amplified from the pGAD-CR-P constructs using the following oligonucleotides: 5'- TGAAGGTCGACCACCAAACCCAAAAAAAGAG -3 '(OligO 5')

5'- CCTGAGAAAGCAACCTGACC -3 '(OligO 3')

The PCR fragment was digested with SalI and Bglll and inserted between the Xhol and Bglll sites of the vector pSG-FNV, thus giving the plasmid pSG-FNV-P.

B. Baits: 1. Construction of the vector pSG-LexA-Rev:

This vector is a derivative of pSG5 (Green et al., 1988) expressing in mammalian cells the LexA-Rev fusion (Farjot et al., 1999).

2. Construction of the vector pSG-LexA-Tat:

The sequence coding for Tat was introduced downstream from that of LexA in pSG-LexA.

C. Indicator construction (built rapporteur):

This plasmid has the SEAP sequence under the control of five binding sites for LexA (Farjot et al., 1999).

III. SUMO-1 and SUMO-1-peptide expression vectors for mammalian cells:

A. Wild pTL1-SUMO-1 construction:

The SUMO-1 sequence was amplified from a cDNA (IMAGE clone obtained from

HGMP - UK) by the following oligos: 5'- GGGTCGACGTCCATATGTCTGACCAGGAGG -3 '5'- AAAAGATCTCTAAACTGTTGAATGACC -3'

The amplified fragment was digested with SalI and Bglll and inserted between the Xhol and Bglll sites of the expression vector pTL1 which is a derivative of pSG5. B. Construction of the vector PTL1-SUMO-CP:

This construction allows the insertion of heptapeptides in fusion with SUMO-1 by suppressing the digiycin motif. The SUMO-1 sequence was amplified with the following oligonucleotides:

5'- GGGTCGACGTCCATATGTCTGACCAGGAGG -3 '

5'- ATCAGATCTGAATCTCGAGCCGTTTGTTCCTGATAAAC -3 '

The amplified fragment was digested with SalI and Bglll and inserted between the Xhol and Bglll sites of the vector pTL1. C. Construction of the vectors pTL1-SUMO-P

The sequence containing the heptapeptide and the basic domain of Tat was amplified as described in II. A., digested with Sali and Bglll and inserted between the Xhol and Bglll sites of the vector pTL1-SUMO-CP.

IV. Ubiquitin and Ubiquitin peptide expression vectors for mammalian cells:

A. Construction pTL1-Ub-CP:

The Ubiquitin sequence was amplified from a cDNA (IMAGE clone obtained from HGMP - UK) with the following oligonucleotides: 5'- CTAAGAATTCAAAATGCAAATCTTCGTGAAAACC -3 '

5 '- CACGAAGATCTACACTCGAGCTCTCAGACGCAGGACC - 3'

The amplified fragment was digested with EcoRI and Bglll and inserted between the EcoRI and Bglll sites of pTL.1.

B. Construction of the vectors pTL1-Ub-P

The sequence containing the heptapeptide and the basic domain of Tat was amplified as described in II. A., digested with Sali and Bglll and inserted between the Xhol and Bglll sites of pTL1-Ub-CP

V. Construction of expression vectors in pFLAG-SUMO-P bacteria:

The plasmids pTL1-SUMO-P were digested with Hindlll and Bglll. The fragment containing the SUMO-1 sequence in fusion with the heptapeptide / basic Tat domain part was inserted between the Hindlll and Bglll sites of the vector pFLAGMac (IBI FLAG® Biosystems). VI. Construction of expression vectors in pFLAG-Ub-P bacteria:

The plasmids pTL1-Ub-P were digested with EcoRI and Bgl11. The fragment containing the Ub sequence in fusion with the peptide / basic domain part of Tat was inserted between the EcoRI and Bglll sites of the vector pFLAG-2 (IBI FLAG® Biosystems).

VII. Cell culture and transfection:

The HeLa and COS7 cells are cultured in DMEM medium supplemented with 5% fetal calf serum in an atmosphere humidified with 5% CO ₂ . The transfections were carried out by the calcium / phosphate co-precipitation method.

For double-hybrid tests in mammalian cells, the transfection is carried out in 6-well plates seeded with 80,000 HeLa cells. The DNA mixture contains 125 ng of the SEAP reporter construct, 25 ng of the expression vectors Tat or Rev in fusion with LexA, with a quantity defined as optimal of plasmid expressing the prey, namely 250 ng for the motifs selected against Tat and 25 ng for the patterns selected against Rev, with the exception of clones 142 and 190 (50ng). The SEAP activity is measured using the “SEAP Reporter Gene Assay Chemiluminescent” kit (Roche) according to the manufacturer's instructions. The Tat functional test is carried out by transfection of 300,000 HeLa cells into 60 mm petri dishes with the LTR HIV-CAT reporter construct (50ng), pSG-Tat (2ng) and the expression vector of SUMO-1 or an interaction peptide (0.5 and 2μg). The Rev functional test is carried out in a similar manner by transfecting the reporter plasmid pDM128 (50ng), pSG-Rev (5ng) and the expression vector of SUMO-1 or of an interaction peptide (0.5 and 2 μg) . CAT activity is measured using the CAT ELISA kit (Roche). Jurkat cells are cultured in RPMI 1640 supplemented with 10% fetal calf serum, at 37 ° C., in a humidified atmosphere, at 5% CO ₂ . Human mononuclear blood cells (PBMC) and monocytes are isolated from the blood of healthy HIV negative donors by density gradient centrifugation using Ficoll-Hypaque (Eurobio). Human mononuclear blood cells (PBMC) are activated for 3 days with 1 μg of phytohemagglutinin-P (PHA-P, Difco Laboratories) and 5 lU / ml of human recombinant interleukin-2 (rhlL-2; Roche products). Then, the human mononuclear blood cells (PBMC) are arranged on a 96-well microplate (100,000 cells per well) in 200 μL of medium A (RPMI 1640 cell culture medium; Invitrogen, 10% fetal calf serum (FCS, Bio West) inactivated by heat (+ 56 ° C. for 45 minutes), and 1% tri-antibiotic mixture (penicillin, streptomycin and neomycin; PSN, Invitrogen)) supplemented with 20 lU / mL rhlL-2. The cells are maintained at 37 ° C. in a humidified atmosphere, with 5% of CO ₂ .

Macrophages derived from monocytes (“monocyte-derived macrophages” MDM) are differentiated from monocytes at 7 days by adhesion. On day 3, 300,000 cells are dispersed per well in 48-well plates in 1 ml of culture medium. The differentiation of monocytes and the culture of MDMs are carried out in cell culture medium A ': DMEM glutamax ™ supplemented with 10% of heat-inactivated FCS and the tri-antibiotic mixture PSN 1X at 37 ° C. in an atmosphere of 5 % of C0 ₂ , humidified.

VIII. Production of bacteria and purification of SHP

The vectors described in section VI are used to transform E.coli BL21-CodonPlus ™ -RP (Stratagene) cells which have been cultured in 2L of LB medium until reaching an optical density of 0.9. After 30 minutes at 18 ° C, 0.1 mM IPTG is added to the cultures which are continued overnight at 18 ° C. The bacteria are sonicated in a lysis buffer: 200 mM NaCI, 0.1 M Tris-HCl, pH 7.4, 10 mM MgCI2 which is supplemented with Roche's "complete" antiprotease (Roche complete antiprotease), 0 , 5mg / mL of lysosyme and 20u / mL of Benzonase (Sigma). After centrifugation at 4,000 revolutions per minute for 30 minutes, the supernatant is loaded onto a 5mL column of Heparin HyperD (Biosepra). Elution is carried out with the DE 600 buffer (600mM KCI, 20mM Hepes, pH7.9, 10μM ZnCI ₂ , 1.5mM MgCI ₂ , 1mM EDTA and 1mM DTT). The elution product is fractionated by a column of Hiload 16/60 Superdex 200 filtration gel (Amersham Pharmacia Biotech). The fractions containing the SHP are combined and, after dialysis against a DE 50 buffer (same composition as the DE 600 buffer, except for the concentration of KCI which is 20 mM), are loaded onto a 5mL mono Q HyperD column ( Biosepra). The flow-through is dialyzed against PBS 1X tamppon, concentrated 10 times by lyophilization and sterile filtered. IX- Immunoblot and immunofluorescence:

The techniques used are those conventionally used in the field (polyacrylamide gels, polyvinylidene difluorinated membranes, PVDF, monoclonal antibody directed against FLAG of Sigma, diluted to 1: 1000 and revelation with the ECL kit from AMersham Biosciences).

The immunofluorescence tests are carried out with a primary anti-FLAG antibody at a dilution of 1: 500 incubated for 2 hours and a secondary antibody coupled to the fluorophore Alexa Fluor 488 (Molecular Probes) diluted to 1: 1000, incubated for one hour.

X. Viruses and antiviral tests:

PBMC cells are infected with the reference lymphotropic strain HIV-1-LAI (Barre-Sinoussi et al, Science, 1983, 220, 868-871), and MDM cells with the reference strain having a tropism for HIV macrophages -1 / Ba-L (Gartner et al, Science, 1086, 233, 215-219). These viruses are amplified in vitro with umbilical blood mononuclear cells UBMC activated with PHA-P. The cell-free supernatant is centrifuged at 360,000 xg for 10 minutes to remove soluble factors, and the pellet is re-suspended in cell culture medium. Virus stocks are titrated using PHMC-P activated PBMC cells and the 50% tissue infection doses in culture (50% Tissue Culture Infectious Doses TCID50) are calculated using the formula of Kârber. PBMC cells are pretreated for 30 minutes with 5 concentrations of each molecule and infected with 75 TCID50 of the HIV-1-LAI strain. AZT (1.6, 8, 40, 200 and 1 OOOnM) and indinavir (1.6, 8, 40, 200 and 1 OOOnM) are used as controls. The interaction peptides SHPR-8, -15, -190 and SHPT-142 are tested at concentrations of 31, 62.5, 125, 250, 500, 1000 and 2000 nM. The molecules of which maintained throughout the culture, and the cell supernatant is harvested 7 days after infection and stored at -20 ° C. in order to measure the viral replication by assaying the reverse transcription activity (“reverse transcriptase »RT). PBMC cells are observed microscopically on the seventh day in order to detect possible drug-induced cytotoxicity. The MDM cells are pretreated for 30 minutes in 5 concentrations of the different molecules and infected with 30,000 TCID50 of the HIV-1 / Ba-L strain. AZT (0.8, 4, 20, 100 and 500 nM) and indinavir (0.8, 4, 20, 100 and 500 nM) are used as controls. The interaction peptides SHPR-8, -15, -190 and SHPT-142 are tested at concentrations of 62.5, 125, 250, 500, 1000 and 2000 nM. The molecules are maintained for 7 days after infection, and the cell supernatant is harvested 7, 14 and 21 days after infection and stored at -20 ° C in order to measure viral replication by assaying the transcription activity reverse (RT reverse transcriptase). MDM cells are observed at the microscopic level on days 7, 14 and 21 in order to detect a possible drug-induced cytotoxicity. HIV replication is analyzed by assaying RT activity in cell culture supernatants using the RetroSys RT kit (Innovagen). The experiments are carried out 3 times and the results are expressed as the average of the RT activity +/- the standard deviation (SD). 50, 70 and 90% of the effective doses (ED50, ED70 and ED90) are calculated using the percentages of untreated controls and computer software (J. and TC Chou, Biosoft, Cambridge).

Example 2: Results

Identification of peptide sequences binding to Tat and Rev

The search for interaction peptides against the regulatory proteins Tat and Rev was carried out initially by a two-hybrid screening in yeast. The baits contain the LexA sequence in fusion with the entire Tat, Rev or NES sequence of rev (Figure 7A). The prey comprises 3 different parts: the GAL4 activation domain, a random sequence of 7 amino acids and the Tat transduction domain (Figure 7B). The prey expression vector library was constructed by cloning DNA fragments prepared from synthetic oligonucleotides including a series of 21 degenerate positions. This bank was used in 3 different screens with either LexTat, LexRev or LexNES as bait (See Table 1a). All the vectors of the clones growing on a medium devoid of histidine and positive for the β-galactosidase test were isolated and re-tested. None of these vectors triggers the expression of β-galactosidase when LexA alone is used as bait (See Table 1b, c, d). Four vectors selected against Tat, 3 against Rev and 19 against NES were thus obtained (See Table 1b, c and d). These positive clones were then analyzed using a double-hybrid mammalian cell test. The constructions produced have been described previously and are illustrated in FIG. 7C. When LexA, alone or with the Prey proteins, is transiently expressed, no expression of secreted alkaline phosphatase (“Secreted alkaline phosphatase SEAP) is observed from the reporter construct. Both LexTat and LexRev are capable, by themselves, of triggering the expression of SEAP, but this production is increased by cotransfection with the vectors of prey. The results are presented in FIG. 8. They show that the selected sequences in yeast are also capable of binding to the Tat and Rev proteins in the nucleus of human cells. Table 1a: number of clones obtained by double hybrid screening of the bank of random peptides:

Table 1b: β-galactosidase test for the clones obtained from the screening carried out with Lex-Tat as bait:

Table 1c: β-galactosidase test for the clones obtained from the screening carried out with Lex-Rev as bait:

Table 1d: β-galactosidase test for the clones obtained from the screening carried out with Lex-NES as bait:

Inhibition of Tat and Rev Functions Given the capacity of the interaction peptides isolated in the preceding steps, the next step is to verify their capacity to act as an antagonist of the activity of Tat and Rev, by disturbing their association with cellular effectors. Functional tests were therefore carried out. In order to stabilize the identified interaction peptides, constructions were carried out with the SUMO-1 protein. The entire coding sequence, including a mutation in the C-terminal di-glycine motif, was inserted upstream of the coding sequence for the heptapeptide-transduction domain module. The artificial or chimeric proteins thus produced were called Sumo-1-heptapetide - grotein Transuction domain Tat or Rev, either SHPT or SHPR according to their ability to bind to Tat or Rev respectively. The effect of the 4 SHPTs on the transactivation of the HIV-1 promoter by Tat is evaluated by transient expression experiments in HeLA cells. At low concentrations, the various SHPTs do not reduce the transactivation of Tat; at higher concentrations, a slight reduction in trans-activation is observed, in particular with SHPT-8 (see FIG. 9A). These data show that the selected SHPTs are not very good inhibitors of Tat function. The activity of SHPR was evaluated using the reporter construct pDM128 which includes the CAT coding sequence as well as an RRE motif in an intron. The expression of Rev stimulates the expression of CAT by allowing the export of the non-spliced messenger to the cytoplasm. Using this test, it was observed that the different SHPRs show different activities, the results are illustrated in FIG. 9B. The data clearly show that 2 SHPR, SHPR-142 and SHPR-190 are a priori effective inhibitors of Rev function. It has been verified whether this inhibition is in fact directly correlated with a protein-protein interaction as expected. To this end, both Rev and SHPR have been produced in bacteria. A protocol was developed in order to purify SHPs (see example 1). Rev is expressed in fusion with GST and coupled to glutathione-agarose beads. SHPR-142 and SHPR-190 are loaded into the column, as well as SHPT-8 as a control. When the GST-Rev fusion is discharged with glutathione, a co-elution of SHPR-142 and SHPR-190 is observed while no SHPT is present in the fractions containing GST-Rev (see FIG. 10). This shows that the interaction between SHPR-142 and SHPR-190 on the one hand and Rev on the other hand is direct and does not involve any intermediary.

Penetration of SHPs inside cells from the extracellular medium The activity of SHPs identified from the extracellular medium was then tested. Initially, cytotoxicity was examined. At concentrations up to 1 μM, these molecules do not seem to modify the percentage of dying cells in a population of peripheral blood mononuclear cells (“peripheral blood mononuclear cell” PBMC) cultured in vitro, activated with PHA and IL-2.

In order to evaluate the capacity of SHP to penetrate into cells, PBMCs are cultured in a medium supplemented with 2 μM of SHPR-190. At different times, an aliquot of the supernatant is removed and the cells are collected. After several washes, the cells are lysed in RIPA buffer and the supernatant as well as what is extracted are analyzed by immunoblot using an antibody directed against the FLAG epitope, an epitope present at the N-terminal end of the SHPs produced in bacteria. Both for the resting lymphocytes (FIG. 11A) and those activated (FIG. 11B), it is observed that SHPR-190 is stable in the culture medium and that a fraction of the protein is present inside the cell (Figure 11). Considering that a lymphocyte is a sphere of 12μ in diameter, the quantification experiments which have been carried out with antibodies labeled with a fluorophore, have shown that an external concentration of 2 μM leads to an intracellular concentration of 15 μM. This indicates that SHPs are likely to be concentrated inside the cell. In order to confirm that the SHPs enter the interior of the cell, immunofluorescence analyzes were also carried out. Jurkat cells were incubated with 1 μM SHP for 4 hours and after two washes, the cells were cultured for 24 hours. The immunofluorescence analysis using an antibody against FLAG shows a clear fluorescence present diffusely throughout the cell interior for the 4 SHPs tested (FIG. 11C), whereas this is not the case for the control cells.

All of these experiments therefore provide proof that lymphocytes can be cultured with purified SHPs and that these latter molecules can actually enter cells.

Inhibition of HIV-1 replication by SHPs

In a final step, it was examined whether the SHPs were indeed capable of inhibiting the replication of HIV-1. This was done in PBMC activated by PHA-P and in MDM. In the model of PBMC activated by PHA-P, replication of HIV-1-LAI is optimal in the 7 ^th day. The effects of the different SHPs were therefore tested on this date. AZT and indinavir were used as controls. As expected, replication of HIV-1-LAI is effectively inhibited by AZT and indinavir (Figure 12A and Table 2a). SHPT-8 has no anti-HIV activity and SHPR-15 has weak activity. In contrast, SHPR-142 and SHPR-190 decrease replication in PBMCs cells activated with PHA-P. HIV-1 replication is reduced by SHPR-142 to 2μM (73 ± 6%, see table 2a), and by SHPR-190 to 1 and 2μM (80 ± 7% and 100% respectively, see table 2a), La comparison of the effective doses at 50% confirm that SHPR-190 is more active than SHPR-142 (see Table 2b). Table 2a: anti-HIV effects of AZT, indinavir IDV and of SHPR-15, SHPT-8, SHPR-142 and SHPR-190 in PBMC infected with HIV-1 LAI. Percentage inhibition.

49 Table 2b: anti-HIV effects of AZT, indinavir IDV and SHPR-142 and SHPR-190 in PBMC infected with HIV-1 LAI: 50, 70 and 90% of the effective doses. The results are expressed in nM.

The reference strain HIV-1 / Ba-L replicates effectively in MDM (Monocytes derived macrophages). The reverse transcription activity in the culture supernatant is detected from the 7 ^th day post-infection (pi) and is maximum between 14 and 21 days pi. As expected, the replication of HIV-1 / Ba-L is reduced in a dose-dependent manner by AZT and indinavir (FIG. 12B and Table 3a). In these cells, as in PBMCs, SHPR-15 shows very low activity. Unlike the results obtained for activated PBMCs, SHPT-142 also has a weak inhibition of HIV replication (44 ± 10% for 2 μM on day 14, and no effect at 1 μM, table 3a). On the contrary, viral replication is inhibited by both SHPT-8 and SHPR-190 at 2 μM (69 ± 6% and 92 ± 2% on day 14, respectively, table 3a). Unlike SHPT-8, the effects of SHPR-190 on HIV replication are dose dependent (Figure 12B). In addition, as illustrated by the ED90 values (Table 3b), the interaction peptide SHPR-190 shows greater anti-HIV activity. These effects decline after 14 days, probably due to the degradation of the interaction peptide.

Table 3a: anti-HIV effects of AZT, indinavir IDV and of SHPT-8 and SHPR-190 in MDM infected with HIV-1 / Ba-L. Percentage inhibition.

The results therefore show that the interaction peptide SHPR-190 exhibits real anti-HIV effects in the two main cellular targets of HIV, namely CD4 + T lymphocytes and macrophages.

In addition, since the SHPR-142 peptide was selected against the Rev NES and the SHPR-190 peptide also appears to interact with the Rev NES signal, these two molecules potentially interfere with the export properties of protein Rev. Table 3b: anti-HIV effects of AZT, indinavir IDV and of SHPT-8 and SHPR-190 in MDM infected with HIV-1 / Ba-L: 50, 70 and 90% of the effective doses. The results are expressed in nM. NC means not calculated.

Example 3: Sequences of SHP identified

SHPT: SHPT-8 P e Thr Met Arg Gly Val Asp SHPT-9 # Thr Arg Arg Island Glu Met Pro Gly Arg Asp Ile Pro Gly Val Asp Gly Ser Ile Leu Arg Gly Cys Trp Asp SHPT-10- Gly Ala Val Asp Lys Ser His SHPT-24 Ser Arg Val Asp Arg Lys Asp # This clone results from the insertion of several heptapeptide sequences. SHPR: SHPR-15 Met Cys Val Asp Leu Leu Leu SHPR-31 Arg Gin Val Gly Met Leu Tyr SHPR-115 Leu Ala Pro Arg Asn Leu Leu SHPR-142 * Phe Trp Phe Cys Gly Leu Lys SHPR-190 * Asn Trp Leu Cys Cys Leu Asn *: these two heptapeptides are those having an inhibitory power on the function of the Rev protein. Example 4: Mutations in SHPR-142 and SHPR-190

Mutational analyzes were carried out to better determine the mechanism of action of SHPs directed against Rev and the importance of the residues at each position in SHPR-142 and -190.

1. Effect of the binding of SHPR-Rev on the binding of the viral protein to its target RNA sequence, the Rev Response Element (RRE): The Rev protein causes the export to the cytoplasm of mRNA HIV-1 by binding to a particular RNA sequence: the RRE. The inventors have therefore clarified the question of whether the binding of Rev inhibitory SHPRs interferes with this Rev / RRE association. For this, delay gel experiments were carried out with a RNA molecule corresponding to the RRE element as a probe. The Rev protein produced in bacteria binds to this RNA and induces a migration delay in non-denaturing polyacrylamide gel. The addition of the proteins SHPR190 and 142 does not inhibit this association, but induces a greater delay indicating that the SHPR-Rev complex is still capable of binding to the RRE. In these experiments the addition of a control protein like SHPT8 has no particular effect.

It can therefore be concluded from these experiments that the inhibitory SHPRs 190 and 142 do not modify the interaction of Rev with the RRE element. Their inhibitory action must therefore rather relate to the export of the Rev-RRE complex to the cytoplasm. This is consistent with the observation that SHPR 190 and 142 interact with the Rev NES.

2. Sequences required for the inhibitory activity of SHPR-142 and 190: The experiments carried out by the inventors are as follows:

- During the cloning of the heptapeptide motif downstream of the SUMO-1 sequence, ten amino acids coding for the GAL4 protein (PPNPKKEIEL) had been preserved in order not to modify the immediate environment of the selected binding motif.

Mutants (called ΔGAL4) of SHPR-142 and 190 have been produced where this sequence coding for a short motif of GAL4 has been eliminated. - To evaluate the contribution of each of the amino acids of the heptapeptide motif to the inhibitory activity, each of the seven positions of SHPR-142 and 190 was mutated into alanine (Ala) and these mutants tested for mutants having anti activity -viral increased. These different mutants were tested using the transient expression system consisting of transfection into HeLa cells of the reporter plasmid having the CAT sequence associated with the RRE in the intronic position with expression vectors for the Rev protein and the SHPR protein. . The inhibitory activity of these SHPR derivatives was measured and is expressed in the following tables relative to that of the parental SHPR. The values indicated correspond to the average of those obtained for two independent measurements. The difference between the two values is less than 20%.

Table 4

Table 4: The expression vectors in mammalian cells of SHPR142 (column 2) and SHPR190 (column 3) were mutated at each of the positions of the heptapeptide (changed to alanine) and by deletion of the amino acids coding for GAL4. These vectors were transfected with the reporter plasmid pDM128 and the expression vector Rev as described in the preceding sections (and in Roisin et al. JBC (2004) 279: 9208.) The amount of CAT protein is expressed in% of that corresponding to the transfection of the parental SHPR. A value greater than 100 therefore indicates a negative effect on the inhibitory power of SHPR and a lower value an increase in this inhibitory property. Table 5

Table 5: The SHPR expression vector was modified by deletion of the amino acids coding for GAL4 and at positions 1 or 4 of the heptapeptide. These expression vectors were tested as described in the legend in Table 4. These results show that the sequence coding for the amino acids of GAL4 plays a negative role. For sequence 142, the modification of the first five residues has a negative effect, that of amino acids 6 and 7 increases the activity, especially for position 7 (Table 4). For heptapeptide 190 it appears that the substitution of amino acid 1 fairly significantly increases the activity, likewise to a lesser degree than that of positions 3, 4 and 7 (Table 4). In addition, the mutation of position 1 or position 4 further increases the effect of eliminating the GAL4 sequences (Table 5). It can be concluded from these results that the residual GAL4 part can be eliminated from the final hybrid proteins. For heptapeptide 190, it also appears advantageous to modify position 1. A modification of position 4 could also be advantageous in order to limit the amount of cysteine residues. These are indeed likely to cause modifications by oxidation. Considering the results obtained for the two motifs 142 and 190, it appears that the tryptophan residue in position 2 is important and that a close cysteine residue is required which can be in position 4 or preferably in position 5. These results lead to new potentially more active SHPR190 derivatives. Studies have also been carried out to better understand the role of the transduction domain. To this end, the basic tat domain used for SHPR 142 and 190 has either been truncated in order to delete most of it, or else replaced by a polyarginine motif. In both cases, it has been shown that the intracellular capacity of the polypeptide to interact with its Rev target is conserved. These results confirm the essential nature of the heptapeptide for the ability to interact with the target, interaction in which the transduction domain does not participate.

Therefore, if a polypeptide according to the invention is identified for its ability to interact with a given target, it is then possible to vary the transduction domain to optimize its properties (transduction efficiency, safety, non-immunogenicity,. ..) without modifying the heptapeptide, while being assured that the polypeptide thus obtained is always capable of interacting with the given target.

References: - Desbois, C, Rousset, R., Bantignies, F., and Jalinot, P. (1996). Exclusion of lnt-6 from PML nuclear bodies by binding to the HTLV-I Tax oncoprotein. Science 273, 951-3.

- Farjot, G., Sergeant, A., and Mikaelian, I. (1999). A new nucleoporin-like protein interacts with both HIV-1 Rev nuclear export signal and CRM-1. J Biol Chem 274, 17309-17.

- Green, S., Issemann, I., and Sheer, E. (1988). A versatile in vivo and in vitro eukaryotic expression vector for protein engineering. Nucleic Acids Res 16, 369.

- Veschambre, P., Simard, P., and Jalinot, P. (1995). Evidence for functional interaction between the HIV-1 Tat transactivator and the TATA box binding protein in vivo. J Mol Biol 250, 169-80.

- Wender PA, Mitchell DJ, Pattabiraman K, Pelkey ET, Steinman L, Rothbard JB. (2000). The design, synthesis, and evaluation of molecules that enable or enhance cellular uptake: peptoid molecular transporters. Proc NatI Acad Sci U S A; 97 (24): 13003-8. - Krâber, G. Beitag Zur Kollektiven Behandlung Pharmakologisher Reihenversuche. (1931) Arch. Exp. Path. Pharmak. 162, 956-959.

Claims

claims

1. Interaction polypeptide comprising the following elements: (a) A heptapeptide motif of sequence X ₁ X ₂ X ₃ XX ₅ X ₆ X ₇ , (b) A transduction domain, characterized in that it is '' a chimeric polypeptide, that amino acid X ₇ is located between 5 and 35 amino acids from the C-terminal end of said polypeptide, and that domain (b) is located at C-terminal with respect to motif (a) .

2. Interaction polypeptide according to claim 1 characterized in that said motif has a sequence defined by a random process.

3. Interaction polypeptide according to claim 1 further comprising: (c) A stabilization domain, characterized in that the stabilization domain is located in N-terminal with respect to the heptapeptide motif of said polypeptide.

4. Interaction polypeptide according to one of claims 1 to 3, characterized in that "spacers" (linkers) are inserted between said elements a and b, and / or a and c.

5. Polypeptide according to claim 4, characterized in that said "spacers" are sequences of less than 5 amino acids and comprising at least 20% of amino acids Glycine or Proline.

6. Polypeptide according to one of claims 1 to 5, characterized in that the transduction domain is that of the HIV-Tat protein.

7. Polypeptide according to claim 6, characterized in that said transduction domain comprises the sequence RKKRRQRRR.

8. Polypeptide according to one of claims 3 to 7, characterized in that said stabilization domain comprises the sequence: MSDQEAKPSTEDLGDKKEGEYIKLKVIGQDSSEIHFKVKMTTHLKKLKESYCQ RQGVPMNSLRFLFEGQRIADNHTPKELGMEEEDVIEVYQEQT (SEQ ID No.1), or includes a sequence having at least 90% identity with SEQ ID No.

9. Polypeptide according to one of claims 1 to 8, characterized in that the amino acid X ₇ is between 10 and 30 amino acids from the C-terminal end of said polypeptide.

10. The polypeptide according to claim 9, characterized in that the amino acid X ₇ is located between 15 and 25 amino acids from the C-terminal end of said polypeptide.

11. Polypeptide according to one of claims 1 to 10, comprising a motif (a) having the following sequence: where Xi, X ₃ , X ₄ , X ₅ , X ₆ , X ₇ , are amino acids chosen independently of one another from natural or modified amino acids, where X ₄ or / and X ₅ is a cysteine and where W is tryptophan.

12. The polypeptide according to claim 11, wherein the cysteine is preferably in position X ₅ .

13. The polypeptide according to claim 11, comprising as motif (a) a sequence chosen from the following sequences: FWFCGLK (SEQ ID No. 2), NWLCCLN (SEQ ID No. 3), FWFCGLA (SEQ ID No. 27), AWLCCLN (SEQ ID N ° 25), NWLCCLA (SEQ ID N ° 26), FWFCGAK (SEQ ID N ° 45), FWFCGAA (SEQ ID N ° 46), NWACCLN (SEQ ID N ° 47), NWLACLN (SEQ ID N ° 48), AWACCLN (SEQ ID N ° 49), AWLACLN (SEQ ID N ° 50), AWLCCLA (SEQ ID N ° 51), NWAACLN (SEQ ID N ° 52), NWACCLA (SEQ ID N ° 53), NWLACLA ( SEQ ID N ° 54), AWAACLN (SEQ ID N ° 55), AWACCLA (SEQ ID N ° 56), AWLACLA (SEQ ID N ° 57), NWAACLA (SEQ ID N ° 58) and AWAACLA (SEQ ID N ° 59 ).

14. The polypeptide according to claim 13, comprising a sequence chosen from the following sequences: FWFCGLKPGRKKRRQRRRG (SEQ ID No. 4), NWLCCLNPGRKKRRQRRRG (SEQ ID No. 5), FWFCGAKPGRKKRRQRRRG (SEQ ID N ° 60), FWFCGLAPGRKKRRQRRRG (SEQ ID N ° 61), FWFCGAAPGRKKRRQRRRG (SEQ ID N ° 62), AWLCCLNPGRKKRRQRRRG (SEQ ID N ° 63), NWACCLNPGRKKRRQRRRNRRNRRNRRNRRNRRNRRRNRRNRRNRRNRRRNRRRNRNRRNRRRNRRNRRNRRNRNRRNRRNRRNRNRRNRRNRRNRRNRRNRRR) 65), NWLCCLAPGRKKRRQRRRG (SEQ ID N ° 66), AWACCLNPGRKKRRQRRRG (SEQ ID N ° 67), AWLACLNPGRKKRRQRRRG (SEQ ID N ° 68), AWLCCLAPRKKRRQRKRKRRNRK (SEQ ID N ° 71), NWLACLAPGRKKRRQRRRG (SEQ ID N ° 72), AWAACLNPGRKKRRQRRRG (SEQ ID N ° 73), AWACCLAPGRKKRRQRRRG (SEQ ID N ° 74), AWLACLAPGRKKRRQRRRRRRRRRRQ 76) and AWAACLAPGRKKRRQRRRG (SEQ ID No. 77).

15. Polypeptide comprising a heptapeptide of sequence: ^N X ₁ WX ₃ X ₄ X ₅ X ₆ X ₇ ^c , where Xi, X ₃ , X, X ₅ , X ₆ , X ₇ , are amino acids chosen independently of each others among natural or modified amino acids, where X ₄ or / and X ₅ is a cysteine and where W is tryptophan, and interacting with the Rev protein or one of its domains.

16. The polypeptide according to claim 15, wherein said Rev protein is the Rev protein of HIV-1, HIV-2, SIV, SHIV or FIV.

17. The polypeptide according to claim 15, wherein said heptapeptide is capable of interacting with the HIV-1 Rev protein in a double-hybrid yeast test, and wherein said polypeptide is capable of inhibiting replication of the HIV-1 virus in vivo. .

18. The polypeptide according to claim 15 or 16, wherein the sequence of the heptapeptide is chosen from the following sequences: FWFCGLK (SEQ ID No. 2), NWLCCLN (SEQ ID No. 3), FWFCGLA (SEQ ID No. 27) , AWLCCLN (SEQ ID N ° 25), NWLCCLA (SEQ ID N ° 26), FWFCGAK (SEQ ID N ° 45), FWFCGAA (SEQ ID N ° 46), NWACCLN (SEQ ID N ° 47), NWLACLN (SEQ ID N ° 48), AWACCLN (SEQ ID N ° 49), AWLACLN (SEQ ID N ° 50), AWLCCLA (SEQ ID N ° 51), NWAACLN (SEQ ID N ° 52), NWACCLA (SEQ ID N ° 53), NWLACLA (SEQ ID N ° 54), AWAACLN (SEQ ID N ° 55), AWACCLA (SEQ ID N ° 56), AWLACLA (SEQ ID N ° 57), NWAACLA (SEQ ID N ° 58) and AWAACLA (SEQ ID N ° 59).

19. Population of interaction molecules, each member of the population consisting of or comprising a polypeptide according to claim 1, 2 or 3 and distinguished from the other members only by the sequence of the heptapeptide.

20. Population of interaction molecules according to claim 19, characterized in that this population comprises at least 100 distinct molecules, preferably at least 1000.

21. Population of interaction molecules according to claim 19, characterized in that each member comprises the following sequence: X ₁ X ₂ X ₃ X ₄ XsXβX ₇ PGKKRRQRRRG (SEQ ID No. 6).

22. Population of interacting molecules according to one of claims 19 to 21 characterized in that each member comprises the following sequence MSDQEAKPSTEDLGDKKEGEYIKLKVIGQDSSEIHFKVKMTTHLKKLKESYCQ RQGVPMNSLRFLFEGQRIADNHTPKELGMEEEDVIEVYQEQTARPPNPKKEI ELGGGGSX ₁ X ₂ X ₃ X ₄ X ₅ X ₆ X ₇ PGKKRRQRRRG (SEQ ID NO: 7 ), or else each member includes the sequence: MSDQEAKPSTEDLGDKKEGEYIKLKVIGQDSSEIHFKVKMTTHLKKLKESYCQ RQGVPMNSLRFLFEGQRIADNHTPKELGMEEEDVIEVYQEQTARGGGGSX _! X ₂ X3X X5X ₆ X7PGKKRRQRRRG (SEQ ID No. 78).

23. A nucleic acid molecule comprising a sequence coding for any one of the polypeptides according to claims 1 to 22.

24. A polypeptide according to any one of claims 1 to 18 for use as a medicament.

25. Pharmaceutical or veterinary preparation comprising as active ingredient a polypeptide according to one of claims 1 to 18.

26. Pharmaceutical preparation according to claim 25, packaged in the form of an injectable solution.

27. Pharmaceutical preparation according to claim 25 for intravenous or oral administration.

28. Use of a polypeptide according to one of claims 1 to 18, for the manufacture of a medicament intended to treat infections by HIV or by a virus homologous in other animal species.

29. A screening method for the identification of polypeptides capable of modifying the phenotype of a cell comprising: i. bringing an interaction polypeptide according to any one of claims 1 to 14 and 19 to 22 into contact with the cell; ii. detecting a change in the cell phenotype; iii. optionally determining the sequence of the heptapeptide motif of said polypeptide.

30. Screening method according to claim 29 characterized in that the contacting consists in the synthesis of the interaction polypeptide by the cell.

31. A screening method according to claim 29 characterized in that the contacting consists in the extracellular addition of the polypeptide.

32. A screening method according to claim 29 characterized in that it is an intracellular screening.

33. A screening method according to claim 29 characterized in that it is an extracellular screening.

34. A screening method according to claim 29 characterized in that it is an in vivo screening process.

35. A screening method according to claim 29 characterized in that it is a process implemented in vitro.

36. A screening method according to claim 29 characterized in that the phenotypic change observed is induced by a reporter gene.

37. Screening method according to claim 36 characterized in that said reporter gene is that of a "two-hybrid" system.

38. A method of screening molecules for the identification of a molecule interacting with a determined intracellular target comprising: i. generating polypeptides according to one of claims 1 to 14 or 19 to 22, bringing the target into contact with the generated polypeptides; detecting the interaction of a polypeptide with the target and optionally determining the sequence of the heptapeptide motif.

39. The screening method according to claim 38, wherein the detection of the interaction is carried out by the observation of a phenotypic change, for example by means of a two-hybrid system, or by the assay of an entity.

40. A method for modulating the properties of an intracellular target molecule comprising bringing in vitro a cell containing the target molecule and an interaction polypeptide according to any one of claims 1 to 14, said polypeptide comprising a heptapeptide motif interacting with the target molecule.

41. Use of a molecule according to claim 23 as prey in a phenotypic screen.

42. Use according to claim 41, characterized in that said phenotypic screen is a two-hybrid screen.

43. Use according to claim 42 characterized in that the bait in said screen comprises a sequence coding for the HIV-Rev protein or one of its domains.

44. Use according to claim 42 characterized in that the bait in said screen comprises a sequence coding for a domain of 7 to 15 amino acids.

45. Use according to claim 44 characterized in that said domain is an NLS (Nuclear Localization Signal).

46. Use according to claim 44 characterized in that said domain is an NES (Nuclear Export Signal).

47. Use according to claim 46 characterized in that said NES is that of the HIV-REV protein.

48. Use according to claim 46 characterized in that said NES has the following peptide sequence: LQLPPLERLTL (SEQ ID No. 8).

49. A method of screening for candidate molecules capable of interacting with an intracellular target molecule comprising: i. bringing the target molecule into contact with an interaction polypeptide according to one of claims 1 to 14 to 19 to 22; ii. detecting the interaction between the target molecule and the interaction polypeptide, iii. the addition of a candidate molecule, iv. detecting the modification of the interaction between the target molecule and the interaction polypeptide.

50. The method of claim 49, wherein said interaction between the target molecule and the interaction polypeptide is observed through the use of a reporter gene, for example in a two-hybrid test.

51. Method according to one of claims 49 and 50, wherein said target protein is the Rev protein and wherein said interaction polypeptide is a polypeptide according to one of claims 15 to 18.

2. Method for identifying molecules capable of modulating the properties of an intracellular target molecule comprising: i. The generation of nucleic acid molecules comprising sequences coding for different members of a population of polypeptides according to one of claims 19 to 22, ii. Screening of these molecules using a two-hybrid yeast system, comprising the target molecule as bait, iii. Detection of an interaction; iv. Possibly, the verification of this interaction in human cells.