EP1570089A1 - Molecules inhibiting hepatitis c virus protein synthesis and method for screening same - Google Patents

Abstract

The invention concerns a method for screening molecules which consists, in vitro: a) in incubating together the subunit p116 (SEQ ID4) of the elF3 protein, the nucleotide sequence of region II (SEQ ID2) of the HCV IRES or any sequence containing at least 10 successive nucleotides of the region II (SEQ ID2) of the HCV IRES and the molecule to be tested; b) in detecting the possible formation of p116/IRES region II complex, the absence of complex indicating the inhibiting capacity of the tested molecule, to inhibit the formation of said complexes; c) in selecting the molecules inhibiting the formation of complexes.

Description

HEPATITIS C VIRUS PROTEIN SYNTHESIS MOLECULES AND METHOD FOR SCREENING SUCH INHIBITORY MOLECULES

The invention relates to the treatment of viral or non-viral pathologies in which proteins are involved, the synthesis of which is initiated through an internal ribosome entry site (IRES), of which at least part of the sequence is similar from one IRES to another. Among these pathologies are notably but not limited to, among viral pathologies, viruses belonging to the Flaviridae family such as hepatitis C virus (HCV), swine fever (CSFV), bovine diarrhea ( BVDV), and among non-viral pathologies, cancers in which certain proteins are involved, such as for example fibroblast growth factors responsible for the neovascularization of developing tumors, the proto-oncogene c-myc etc ... More specifically, the treatment proposed in the invention consists in preventing the binding of the translation initiation factor, eIF3, to RNA constituting the 5 ′ non-coding part of the IRES (Internai Ribosome Entry Site) sequence of viral genomes or of certain genes involved in the abovementioned pathologies, so as to inhibit protein synthesis. Consequently, the subject of the invention is also a method for screening for molecules capable of inhibiting the formation of the complex: sequence of TIRES / eIF3, in particular nucleotides 56 to 92 of domain II of TIRES and the recombinant polypeptide corresponding to the part central (amino acids 185 to 279) of the protein subunit pi 16 (also called pi 10, eIF3b, eIF-3eta (BLAST P55884)) of eIF3.

The process of research and development of new therapeutic drug discovery molecules requires above all the identification of new targets associated with diseases (protein, RNA or DNA, or their complex) and their validation. The identified and validated target is then used in molecular screening tests, which allow the selection of active molecules. It is this approach which is proposed by the Applicant, the target being constituted by a specific sequence of HCV TIRES (domain II) and a recombinant polypeptide derived from the sequence of the subunit pl l6 of eIF3. In the following description, the invention is more particularly described in connection with the treatment of the HCV virus although this also applies to the swine fever virus (CSFV) or that of bovine diarrhea (BVDV). , and this, taking into account the strong homology existing between these viruses belonging to the same family.

The hepatitis C virus has been identified as being responsible for non-A non-B hepatitis developed frequently during chronic malignant pathologies, such as cirrhosis of the liver or hepatocellular carcinoma. HCV is transmitted by blood transfusion or blood derivatives. The HCV genome is in the form of single-stranded RNA with a size of around 9.4 kb and coding for a single polyprotein consisting of 3,010 amino acids (CHOO et al., 1989).

Contrary to the classic scheme, the initiation of the translation of HCV messenger TARN is not done by recognition of the cap (or CAP), since the latter is absent (translation called "cap-dependent"), but through an internal ribosome entry site (IRES), positioned at the 5 'untranslated region (5'-UTR) of HCV, between nucleotides 40 and 372 of the HCV sequence (translation known as "cap- independent ") (equivalent to SEQ ID1 according to the invention). As the mechanism of synthesis of viral proteins is very different from that of the host cell, a possible strategy for the development of new therapeutic molecules consists in inhibiting viral protein synthesis without any influence on the protein synthesis of the host cell. In addition, the sequence of TIRES being a very conserved region in this virus reputed to be very variable (92% homology), it can be expected that the use of this sequence as target, will be particularly interesting.

Different structural studies have shown that TIRES of HCV is folded in on itself to form three domains or regions, in a loop, respectively regions II (Ha, Ilb), III (ma, IHb, IIIc, Illd, Ille, Illf) and IV as shown in FIG. 1 (ZHAO et al, 2001), TIRES further comprising a start AUG codon. The single-stranded RNA of CSFV and BVDV also contains an IRES sequence containing a start AUG codon, the structure of TIRES being similar to that of HCV (FIG. 1). In addition, the alignment of the sequences consisting of the genome of these three viruses shows a strong homology of the region II of TIRES, in particular of the MRR site (RNA recognition pattern), which tends to suggest that the molecules acting on TIRES of the HCV could also act on that of the CSFV or the BVDV. In practice, the initiation of translation of mRNA begins with the recognition and fixation by TIRES of the 40S ribosomal subunit and of initiation factors, in particular, the initiation factor called "eIF3".

The initiation factor eIF3 is a multiprotein complex consisting of 10 different subunits such as for example p36, p44, p47, p66, pi 10, pi 16 and pl70. Secondary structure prediction studies have shown that the pi 16 subunit has in its central part, located between amino acids 185 and 279, a recognition pattern for TARN β-α-β-β-α- β. The location of the recognition motif of the pi 16 subunit of eIF3 is represented in FIG. 2. Similarly, the C-terminal part of the p44 subunit also has a similar structure β-α- β-β-α-β, corresponding to a hypothetical MRR. This type of motif is found in a large number of proteins binding to RNA (RNA binding protein), such as for example the proteins lnRNP or even snRNP, but also in some proteins binding to single-stranded DNA. According to the secondary structure prediction algorithms, the central part of the pi 16 subunit is folded in a conformation similar to that of the known MRRs for the conserved amino acids IVVD and TK / RGF / YVE located in leaves 1 and 3 corresponding to the RNP-2 and RNP-1 recognition patterns (see Figure 2). Although by its secondary structure and its homology, the MRR of pi 16 of eIF3 meets the criteria of "putative RNA-binding proteins", its real capacity to fix TARN has never been demonstrated before.

In fact, document FR-A-2 815 358 describes a method of treating hepatitis C which consists in preventing the protein synthesis of HCV by supposed inhibition of the binding of the pi 16 subunit of eIF3 to region III of TIRES. The candidate molecules for this inhibition correspond to polypeptides having an affinity with region III of TIRES greater than that of the pi16 subunit of eIF3. In practice, the polypeptide inhibitors are obtained by screening for pi16 proteins mutated with the IRES sequence of HCV. More precisely, only the central part corresponding to the recognition motif (MRR) is mutated, the polypeptide being capable of binding to the IIIb loop of HCV TIRES with an affinity greater than or equal to that of the non-mutated MRR of pi 16. In in practice, the mutations are introduced into the MRR by random mutagenesis or by targeted mutagenesis according to the phage display technique. Here again, no indication is given concerning the nucleotide sequence of region III of TIRES capable of interacting with the mutated MRR. In addition, no results of a possible inhibition are given in the examples. Sizova et al, 1998, showed that eIF3 protected the Illb apical region of TIRES of HCV and CSFV, in particular nucleotides 204, 212, 214, 215 and 220 (see Figure 1, labeled nucleotides ©), from enzymatic cleavage or chemical changes. More recently, Kieft et al, 2001, using the same methods as those implemented by SIZOVA supra, identified the nucleotides of the Illa stem, Illb loop and Illb stem as the main elements of the interaction (see Figure 1 , labeled nucleotides ⁽ D). In addition, using the technique known as “filter-binding assay”, these various authors have shown that the deletion of the apical region Illb leads to a reduction in the eIF3-IRES interaction of at least 10 Thus, the Illb apical loop is currently considered to be the most likely site of fixation of eIF3, however, none of these documents precisely shows the existence of an interaction between the isolated Illb domain and eIF3. Likewise, none of them identify a specific RNA sequence binding to eIF3.

Buratti et al, 1998, showed that the proteins pi 70 and 116 / pl 10 of eIF3 bind to region III of TIRES of HCV without however, again, identifying the RNA sequence of TIRES envisaged.

The document WO 01/44266 also reports the interaction between the pi 16 subunit of the initiation factor eIF3 and the region III of HCV TIRES, more precisely at the level of a domain capable of pairing and defining two sequences. 7 base and 9 base nucleotides, respectively. The definition of this minimal motif makes it possible to implement a test to identify compounds capable of competing in the formation of the eIF3-HCV complex.

Furthermore, document US-A-6,001,990 describes a series of oligonucleotides, selected for their capacity to inhibit the translation of TNA of HCV. Among them, Toligonucleotide of 28 nucleotides of sequence TAGACGCTTTCTGCGTG AAGACAGTAGT, corresponding to the sequence SEQ ID 3 of this document, effectively hybridizes with region II of TIRES of HCV.

It is known that the RNA-binding proteins of the MRR family recognize short single-stranded sequences (<10 nt) belonging to a loop in RNA structures of the loop-stem type or to a stem extension. These short fragments integrated into an appropriate structural context are essential for the specific binding of MRRs to messenger RNAs. or premessagers comprising 1000 nucleotides or more. Therefore, it is important to identify the minimum sequence of TIRES of TIRES interacting with the MRR. Indeed, the identification of this minimum sequence makes it possible first of all to understand the mechanism of the interaction, but also to design complementary antisense oligonucleotides (of size generally between 30-35nt) capable of inhibiting the formation of the complex RNA / protein or in the case of RNAi (interference RNA or silencing of size between 21-23 nt) to target the interaction region. The identification of the minimum sequence is also essential for carrying out the structural studies necessary for in silico screening as well as for the optimization of active molecules. According to a technique known to those skilled in the art, the atomic structure of the RNA / protein complex or RNA alone in 3 dimensions is sought by NMR. It is known that this technique can only be used for small or complexed RNA fragments, of small size (less than 25 nt). A second technique corresponds to X-ray crystallography, a technique which can be applied to larger RNA fragments, however limited to 70 nt. On the contrary, the crystallography of proteins is not limited by size but can however only be applied to isolated proteins and not to multiprotein complexes, such as eIF3 (deletion).

In other words, one of the problems which the invention proposes to solve is to precisely identify the smallest RNA sequence of TIRES which binds to the MRR of pi 16, so that this sequence can be used in screening methods for molecules of interest (deletion).

During its research, the Applicant not only discovered that the pi 16 subunit of eIF3 did not bind to region III but to region II of HCV TIRES (SEQ 2 according to the invention), but also succeeded to precisely identify the nucleotide sequence of TIRES, hereinafter called the consensus sequence (SEQ 3 according to the invention), interacting with the MRR of pi 16.

Furthermore, the functionality of the MRR (SEQ 5 according to the invention) predicted in the pi 16 subunit of eIF3 (SEQ 4 according to the invention) has been demonstrated by the Applicant. Thus, the expression of this central part of the pi 16 subunit of eIF3 in recombinant form and the demonstration of its specific binding to domain II of HCV TIRES makes it possible to use this polypeptide in place of the multiprotein complex eIF3 , requiring purification from a lysate of cultured cells or reticulocytes. This makes it possible, on the one hand, to make screening much less expensive, but above all to apply methods of structural biology, such as NMR or crystallography, to resolve the atomic structure of this RNA-protein complex and to design inhibitors.

Given the Thomology existing between the IRES sequence of HCV and those of CSFV and BVDV, the discovery of the consensus sequence makes it possible to envisage treating the different pathologies in which these viruses are involved, thanks to molecules capable of blocking the synthesis protein by inhibiting the binding of the pi 16 protein subunit of eIF3, in particular its MRR, to region II of these viruses.

The candidate molecules can be existing or future molecules whose inhibitory properties are tested by screening.

Consequently, the invention firstly relates to a method for screening molecules according to which, in vitro: a the pi 16 subunit (SEQ ID4) of the protein eIF3, the nucleotide sequence of region II (SEQ) is incubated together ID2) of TIRES of HCV or any sequence containing at least 10 successive nucleotides of region II (SEQ ID 2) of TIRES of HCV and the molecule to be tested, b / we then detect the possible formation of complex pi 16 / region II IRES , the absence of a complex testifying to the inhibitory capacity of the molecule tested, to inhibit the formation of said complexes, c / the molecules inhibiting the formation of complexes are selected.

By molecule, we mean any chemical molecule of synthetic or natural origin, known or future. This term also designates multiprotein complexes such as antibodies, proteins, peptides, ribonucleotides or natural or modified deoxyribonucleotides, and PNA molecules (peptides-nucleic acids).

As already said, the pi16 subunit of eIF3 contains a TARN recognition motif (MRR) located in the central part, more specifically between amino acids 175 and 279 of the sequence SEQ ID4. The amino acid sequence of the pi 16 MRR corresponds to the sequence SEQ ID5. In other words, and in an advantageous embodiment of the screening method of the invention, only the sequence of the recognition motif of the pi 16 protein (SEQ ID5) is incubated. The MRR polypeptide from pi 16 (SEQ ID5) is preferably produced in recombinant form, in association with a tag facilitating its purification. It can be labeled, during preparation, with radioactive, biotinylated or fluorescent amino acids, making it possible to detect the formation of the protein / RNA complex.

Furthermore and as will be demonstrated in the examples, the Applicant has identified precisely the sequence of the region II of TIRES of the HCV binding to the recognition motif of the pi 16 protein. This sequence, called in the following description "sequence consensus "(SEQ ID3), contains 37 nucleotides located between nucleotides 56 and 92 of the IRES sequence of HCV. This sequence can be produced by chemical synthesis or by in vitro transcription, and labeled by radioactivity, biotinylation or fluorescence.

Consequently and in a preferred embodiment, only part of the region II is incubated and corresponds to the consensus nucleotide sequence SEQ ID3 or a sequence comprising at least 8 successive of the sequence SEQ ID 3.

Since RNA is an unstable molecule, the RNA molecule to be incubated according to the invention can be modified in order to increase its stability. For this purpose, it can contain phosphorothioate, methylphosphonate, phosphoramidate, acetamidate, carbamate, etc. skeletons. It can also contain modified bases, such as 2'-deoxynucleosides, 2 '-O-alkylnucleosides, 2' -fluoro-2 '-deoxynucleosides.

In practice, the incubation is carried out in a buffer solution at room temperature. Advantageously, increasing concentrations of molecules to be tested are incubated in order to detect a possibly dose-dependent efficacy.

The second step of the process consists in detecting the formation of protein / RNA complex. Any detection method known to a person skilled in the art can be implemented. If radiolabelled RNA is used, the RNA / protein / molecule mixture is advantageously filtered through a nitrocellulose membrane, then the detection is done by measuring the radioactivity bound to the membrane, corresponding to the amount of RNA fixed on protein. Alternatively, TARN can be labeled non-radioactively (by example with biotin), incubated with the protein, filtered through a nitrocellulose membrane and revealed using streptavidin or specific antibodies.

Other techniques can be used to detect RNA / protein interaction, such as SPA (Scintillation Proximity Assay), FRET (Fluorescence Resonance Energy Transfer), HTRF (Homogeneous Time-Resolved Fluorescence), LANCE (Lanthanide Chelation Excitation), FP (Fluorescence Polarization), FCS (Fluorescence Correlation Spectroscopy), FL (Fluorescence Lifetime Measurements).

In the SPA approach, the purified pi 16 MRR polypeptide is immobilized using antibodies specific for the myc epitope, present in the C-terminal part of the polypeptide, or Ni ²⁺ chelators, on a 96-well plate impregnated with glitter. The radiolabelled consensus RNA is added. A signal is only detected if TARN is attached to the immobilized polypeptide.

As will be described in the examples (FIG. 8), this screening technique was used by the Applicant to study and compare the capacity of 15 different aminoglycosides to dissociate the MRR complex from pi 16 / consensus RNA.

In an advantageous embodiment, the screening results according to the invention are correlated with those of additional tests consisting in testing, ex vivo, the influence of the selected molecule on the cap-independent translation (dependent on TIRES) and the translation cap-dependent using a bi-cistronic construction.

This step can be implemented by any method known to those skilled in the art, in particular by the construction of bicistronic vectors consisting of two luciferases framing the sequence of region II (SEQ ID 2) or any sequence containing at least 10 nucleotides successive of region II (SEQ ID 2), or the consensus sequence (SEQ ID 3) or a flanking sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3; the first luciferase being translated in a cap-dependent manner and the second in a cap-independent manner or vice versa. Cells are then transfected with the bicistronic vectors and the rate of translation by Dual Luciferase is measured. The cells capable of being transfected are chosen in a conventional manner by a person skilled in the art, such as, for example, HeLa cells or even Huh 7 cells. The comparison of the results obtained with the screening test proposed by the Applicant and those of the bicistronic tests makes it possible to retain only the molecules capable both of preventing the fixation of pi 16 on the domain II of TIRES in vitro and of inhibiting specifically IRES-dependent translation in a cellular model (ex vivo). In addition, as shown in Example 4, the fact that tobramycin is both capable of dissociating the pi 16/11 complex in vitro at all the concentrations tested and is the most specific inhibitor of the translation controlled by TIRES demonstrates the validity and relevance of the choice of the pi 16 / IRES complex as a screening target.

Consequently, the invention also relates to the use of the molecules identified at the end of the screening process described above for the preparation of a medicament intended for the treatment of hepatitis C (HCV), of swine fever (CSFV) , bovine diarrhea (BVDV).

More generally, any molecule capable of inhibiting in vitro the binding of the pi 16 protein, in particular its recognition motif (MRR) to region II or a sequence containing at least 10 successive nucleotides of region II (SEQ ID 2), in particular a part of region II corresponding to the sequence SEQ ID3 or a sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3, can be used for the manufacture of a medicament intended for the treatment of hepatitis C (HCV) , swine fever (CSFV), bovine diarrhea (BVDV).

In the context of a first trial implementing the screening method of the invention, the Applicant has found that the aminoglycosides, in particular tobramycin, were capable of inhibiting the binding of the MRR of pi 16 to the consensus sequence of the region II of TIRES and that, in addition, this inhibition did not affect cap-dependent translation.

Consequently, the invention also relates to the use of aminoglycosides, in particular tobramycin, for the manufacture of a composition intended for the treatment of hepatitis C (HCV), swine fever (CSFV), bovine diarrhea (BVDV). They may also be derivatives of aminoglycosides, in particular derivatives of tobramycin, having improved properties in the pharmaceutical context. These aminoglycosides, preferably tobramycin, or their derivatives can be administered in combination with liposomes for better absorption.

In particular, the Taminogroup in position 6 ′ in tobramycin is particularly exposed and can be selectively acetylated and then used to graft other groups. The invention therefore also relates to the use of tobramycin analogs, particularly those modified in the 6 'amino position, for the treatment of hepatitis C.

Furthermore, the discovery of the consensus sequence makes it possible to use an antisense oligonucleotide complementary to the sequence SEQ ID 3 or any sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3, with the exception of the sequence TAGACGCTTTCTGCGTGAAGACAGTAGT, as a medicament, in particular for the treatment of hepatitis C (HCV), swine fever (CSFV), bovine diarrhea (BVDV). Similarly, RNAi containing 19 nucleotides of the sequence SEQ ID 3 (consensus sequence) flanked by UU can be used as a medicament for the treatment of the same pathologies as above.

In a first embodiment, the invention therefore also relates to a pharmaceutical composition comprising an antisense oligonucleotide complementary to the sequence SEQ ID 3 or any sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3, exception of the sequence TAGACGCTTTCT GCGTGAAGACAGTAGT.

As already said, the molecules tested in the screening process can be known molecules such as for example aminoglycosides but also molecules still to be developed.

In the latter case, it appears possible to identify in silico, from a library of molecules, molecules capable of inhibiting the protein synthesis of viruses belonging to the family of Flaviridae.

Consequently, the invention also relates to a method for screening a library of molecules in silico consisting of: - to determine the atomic coordinates either of region II of TIRES (SEQ ID 2) of HCV or of any sequence containing at least 10 successive nucleotides of region II (SEQ ID 2) of HCV TIRES, or of the binding sequence specifically to the MRR of the protein PI 16 of eIF3 (SEQ ID 3) or a sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3, either of the region II complex (SEQ ID 2) or of the specific sequence (SEQ ID 3) with the recognition motif of the protein pi 16 of eIF3 (SEQ ID 5)

- then to screen the library of chemical molecules with the atomic coordinates thus determined.

Any software known to those skilled in the art may be used for determining the atomic coordinates.

The molecules thus identified can then be tested in the method described above, consisting in detecting RNA / protein complexes in vitro.

The invention and the advantages which ensue therefrom will emerge more clearly from the following embodiment in support of the appended figures.

Figure 1 is a representation of the structure of HCV TIRES. This consists of 3 loop domains, II (lia, Ilb), III (Illa, Illb, IIIc, Illd, Ille, Illf) and IV. References © and ® indicate the nucleotides involved in the binding of eIF3 according to references Sizova et al. (1998) and Kieft et al. (2001), respectively. Figure 2 shows the location of the TARN Recognition Pattern in the pi 16 subunit of eIF3 and the prediction of its secondary structure.

FIG. 3 compares the affinity of the MRR of pi 16 for regions II, Illabc, IlIeflV and the whole of HCV TIRES, measured after retention on nitrocellulose. On the left, a graphic representation of TIRES of HCV makes it possible to locate the different fragments tested. FIG. 4 is a diagram showing the principle of the method for producing random sub-fragments of HCV TIRES.

FIG. 5 is a diagram showing the principle of the selection process of the random sub-fragments specific to the MRR of pi 16 of eIF3 obtained according to the diagram of FIG. 4 (5 A), and the sequences of the transcription matrices and of the primers used (5B). FIG. 6 represents the results of alignment of the RNA sequences, selected at the end of the 4 ^th and 5 ^th selection / amplification cycles (6A), and the location of the “consensus” sequence (direction orientation) in TIRES of HCV (6B).

FIG. 7A shows the capacity of the consensus sequence (DOR4-35 and DOR5-4) to inhibit the interaction between IRES and MRR of pi 16, compared with that of TIRES, II, Ilabc, IlIeflV and transfer RNA.

FIG. 7B shows that the consensus sequence (nt 56-92) has an affinity for the MRR-pl 16 of eIF3 greater than that of the Illa fragment (nt 153-173) and that of the apical part of fragment II (nt 73 -92). FIG. 8 represents the capacity of the aminoglycosides to inhibit the binding of the MRR of eIF3 to the consensus sequence of the region II of HCV TIRES. FIG. 9 represents the effect of aminoglycosides on cap-dependent and cap-independent translation in cell culture. The bicistronic construction, the cloning scheme of which is shown in FIG. 9A, is used for the transient transfection of HeLa cells with 1 μg of plasmid DNA (Figure 9B) and 2.5 μg of pDNA (Figure 9C).

Example 1: Demonstration of the capacity of the recognition motif (MRR) of pl16 to bind to the region II of PIRES of HCV

1 / Cloning and expression of the recognition motif of pi 16 of eIF3

The amino acid sequence of the recognition motif (MRR) of the pi 16 protein corresponds to the sequence SEQ ID5 located between amino acids 175 and 279 of the sequence SEQ ID4 (corresponding to the sequence of the pi 16 protein). The cDNA coding for the MRR is amplified by RT-PCR from DNA extracted from HeLa cells in the presence of the following primers:

- SEQ ID6: CATATGGATCGGCCCCAGGAAGCAGATGGAATC

- SEQ ID7: GTGCTCGAGCCACTCGTCACTGATCGTCATATA The amplified fragment is cloned in a plasmid pET-30b (Novagen) in fusion with His ₆ -Tag C-terminal between the Nde and Xho sites. The protein is then produced in E. Coli (strain BL211ysS) then purified on Ni2 + - NTA agarose under native conditions. 2 / Synthesis of total TIRES and its Illabc fragments. IlIeflV and Iiab

a / Principle Four different nucleotide sequences are synthesized and cloned, respectively: - a nucleotide sequence corresponding to the totality of TIRES located between nucleotides 40 and 372 of the HCV DNA (b),

a nucleotide sequence corresponding to the Illabc region, located between nucleotides 141 and 252 of the DNA of HCV ©,

a nucleotide sequence corresponding to the IlIeflV region located between nucleotides 250 and 372 of the HCV DNA (d),

- A nucleotide sequence corresponding to the Iiab region located between nucleotides 40 and 119 of the HCV DNA (e).

b / Cloning of the entire nucleotide sequence of TIRES (SEQ IDl The cDNA of TIRES (SEQ IDl) is amplified by RT-PCR from total RNA isolated from patients suffering from HCV (genotype lb) in the presence of the primers following nucleotides:

SEQ ID8: ACCGCTAGCCTCCCCTGTGAGGAACTACT SEQ ID9: GAAAGCTTTTTTCTTTGAGGTTTAGGATTTGTGCTCATGATG CACG

The amplified fragment is first cloned into a plasmid pGEM-T then then into pSP-luc + (Promega) between Nhel and Hind III sites. The plasmid pSP-IRES-lucH- thus obtained contains TIRES of HCV clone in fusion with the luciferase under control of the promoter SP6.

Once sequenced (GenomeExpress, Grenoble), the TIRES sequence was aligned and compared to the other IRES sequences deposited in the banks (such as D49374 or AF139594). The identity observed was 96.6%, which corresponds to the average rate of genomic variability of IRES between different strains of HCV.

c / Summary of the Illabc region

The synthesis of cDNA from the Illabc region is carried out as follows. Two overlapping oligonucleotides, the first of which, SEQ ID10, consists of the promoter for T7 polymerase and the nucleotide sequence of the Illa and Illb region (Nt 139-215 of HCV TARN) and the second, SEQ ID11, of the nucleotide sequence of the Illb region and IIIc (Nt 193-252 of HCV TARN) are hybridized in the presence of a Klenow fragment. The oligonucleotides have the following sequences:

- SEQ ÏÏDIO: TAATACGACTCACTATAGGGTAGTGGTCTGCGGAACCGGT GAGTACACCGGAATTGCCAGGACGACCGGGTCCTTTCTTGGATAAACCCGCT CAA

- SEQ IDl1: TAGCAGTCTCGCGGGGGCACGCCCAAATCTCCAGGCATTG AGCGGGTTGATCCAAGAAAG

The double stranded cDNA fragment obtained is then amplified by PCR in the presence of T7 corresponding to SEQ ID 12: TAATACGACTCACTATAGGG. And of a flanking oligonucleotide whose sequence is as follows: SEQ IDl 3: TAGCAGTCTCGCGGGGGCACG.

άl Synthesis of the IlIeflV region The cDNA corresponding to the IlIeflV region (Nt 250-372) was obtained by PCR amplification of the plasmid pSP-ΔIRES-luc + using primers whose nucleotide sequences correspond to those of SP6 (SEQ IDl 4: TATTTAGGTGACACTATAGAAT) and SEQ IDl 3. The plasmid pSP-ΔIRES-luc + results from the digestion of the plasmid pSP-IRES-luc + by Nhel, the cleavage sites being located between nucleotides 39/40 and 248/249 of TIRES. The amplification product SP6 -> SEQ ID13 is then used as a matrix in the in vitro transcription reaction using SP6-polymerase (SP6 MEGAscript, Ambion).

e / Synthesis of the Iiab Region The cDNA corresponding to the Iiab region was obtained by PCR amplification of the plasmid pSP-IRES-luc + using the primers SP6 (SEQ ID 14) and SEQ ID 15 GTCCTGGTGGCTGCAGGACACTCATAC. The amplification product SP6 -> SEQ IDl 5 is then used as a template in the in vitro transcription reaction using SP6-polymerase.

3 / Fixation of the MRR of pi 16 on TIRES and its Illabc domains. IlIeflV. IIAB

Radiolabeled RNA fragments are obtained by in vitro transcription of the above-mentioned matrices in the presence of [α-32P] UTP. The RNA fragments are purified in a 6% acrylamide-urea gel and precipitated. The RNA pellets are taken up in 25mM Tris-HCl, pH 7.4. In order to allow renaturation, TARN was incubated at 65 ° C in the aforementioned buffer for 5-7 min and then slowly cooled to room temperature. The renatured RNAs were incubated with increasing concentrations of protein in the same 25mM Tris-HCl buffer, pH 7.4, at room temperature for 5 min. The mixture of proteins and RNA is then deposited on a nitrocellulose membrane previously washed with the same buffer. The radioactivity of the filter containing the RNA-protein complexes was measured using the MicroBeta Trilux radioactivity counter (PerkinElmer).

In the hypothesis of a competitive inhibition, the MRR of pi 16 was previously incubated with non-radiolabelled RNA (concentration: protein 0.7 μM, RNA: 0.1 to 1 μM) for 30 min at room temperature followed by the addition of radiolabelled TIRES TIRES. The analysis of binding of TARN to the protein was carried out exactly as described above.

4 / Results

The affinity of the RNA recognition motifs (MRR) of the pi 16 subunit of eIF3 for whole TIRES and its fragments II, Illabc and IlIeflV was studied by retention on nitrocellulose. As shown in Figure 3, the TIRES fixed protein with an apparent Kd of 0.8μM. However, the affinity of MRRpl 1ό for the Illabc fragment (putative eIF3 binding site) is significantly lower than that for TIRES and comparable to that for IlIeflV used as a negative control. This was unexpected, especially since the previously published results assumed that the apical part of the loop forming the Illb region was the probable site of binding of eIF3 (Sizova D, 1998; Buratti, 1998; Kieft et al, 2001; FA- 2,815,358). In reality and as shown in this figure, the recognition reason for eIF3 is not found in the Illabc region but in region I Example 2: Identification of the consensus sequence binding to the MRR p! 16

1 / Production of random sub-fragments of HCV TIRES and method of selection of specific fragments binding to the MRRpl 16 of eIF3

The method called SERF (Selection of Random Fragments) described by STELZ (2000) is used to synthesize the random sequences of TIRES. Its principle is shown in Figure 4.

A / Production of the sub-fragments

2 ug of TIRES cDNA are digested with 5U of Dnase I (Rnase-free, Amersham), at room temperature, for 15 minutes, making it possible to obtain cDNA fragments, the size of which varies between 30 and 100 nucleotides. Blunt ends are generated at the end of the cDNA fragments obtained, by Taq-polymerase at 72 ° C, for 10 minutes in a PCR buffer based on dNTP 1 mMol. Taq-polymerase simultaneously adds additional “dA” residues to the T '3' end of fragments (Figure 4). This makes it possible to increase the efficiency of ligation of the fragments obtained in the vector pGEM-T-Easy (Promega), which in turn is provided with complementary "dT" at the 5 'end (FIG. 4). The DNA fragments are then cloned in the presence of T4 DNA ligase (BioLabs) in a pGEM-T Easy vector (Promega) between the T7 and SP6 promoters. The DNA fragments are then amplified in the presence of the oligonucleotides T7 and SP6 and then the amplification product is used as a template for transcription by SP6 (MEGAscript, Ambion). The transcripts larger than 200 nt corresponding to the transcripts with Tinsert> 60 nt were purified on 10% 8M urea acrylamide gel (FIG. 4, M corresponding to RNA markers “Century markers”, Ambion).

B / Selection of the subfragments The recombinant protein MRRpl 16 of eIF3 is purified on a Ni-NTA-agarose column under native conditions (FIG. 5). The purified protein is then incubated with the library consisting of the purified RNA fragments obtained above in a 25 mM Tris-HCl buffer, pH 7.4 for 15 min at room temperature. The concentration of TARN is, initially equal to 0.2 μM and that of the protein, equal to 0.8 μM. The protein / RNA mixture is then deposited on a nitrocellulose membrane previously washed with the same buffer. The filter containing the RNA-protein complexes is then cut into pieces and TARN is extracted with an SDS solution, 0.1%), sodium acetate 0.3M pH 5.0 for one hour at room temperature. The RNA is then recovered by precipitation in tethanol in the presence of tRNA used to facilitate precipitation. The RNA pellet is then taken up in 10 μl of water and subjected to reverse transcription in the presence of the reverse transcriptase "Stratascript" of T oligonucleotide T7 (Stratagene). The single-stranded DNA fragments are then amplified by PCR using Toligonucleotide T7 (SEQ ID 14), T oligonucleotide SP6 (SEQ ID 14) and the sequence SEQ ID 16 corresponding to the linker region adjacent to SP6.

- SEQ ID 16: TATTTAGGTGACACTATAGAATACTCAAGCTATGCAT CCAACGCGTTG A control PCR is carried out in parallel with the oligonucleotides SP6 and T7 in order to confirm the absence of contaminating template DNA, among the selected RNAs. The fragments amplified by PCR are then purified and then used as a transcription matrix in the following cycle. The selection / amplification cycle is repeated 5 times. The RT-PCR products are analyzed on 2% agarose gel (FIG. 5: Φx are DNA markers (stratagene), 1 and 2 are amplification products obtained with the primers SP6 or SEQ IDl 6. The concentration d RNA in subsequent cycles is equal to 0.058 μM and that of the protein is regularly reduced from a value of 1.2 μM in the second cycle to a value of 0.2 μM in the fifth cycle. obtained after the fourth and fifth cycles are cloned into a plasmid pTrcHis2-TOPO (Invitrogen) chosen to facilitate the cloning process in the absence of the T7 promoter. The plasmids were purified and sequenced. The sequences obtained were aligned with the software assistant Clustal W DNA (Thompson, JDet al CLUSTAL W: improving the sensitivity of progressive multiple sequence aligned through sequence weighting, positions-specific gap penalties and weight matrix choice. (1994) Nucleic Acids Research, 22, 4673-4680) available during e on the Pôle Bio-Informatique Lyonnais website.

2 / Results

As illustrated in FIG. 6A, among 16 selected RNA sequences, cloned after 4 and 5 cycles and sequences using the T7 primer, 11 clones contain the sequence UACUGUCUUCACGCAGAAAGCGUCUAGCAUGGCGUU corresponding to nucleotides 56 to 92 of the sequence SEQ ID1, 2 clones contain the sequence CGCCTCATGCCTGGAGAT (nt 61-72 of SEQ IDl) and one clone shows homology with the part 84-90 of SEQ IDl. Thus, these results identify the region of TIRES 56-92 of sequence TACTGTCTTCACGCAGAAAGCGTCTAGCCATGGCGTT (SEQ ID3) as corresponding to the binding site of MRRpl 16 (FIG. 6B).

The hypothesis of competitive inhibition offers an additional means of studying the specificity of the interaction in question. As shown in Figure 7A, the consensus sequences of clones 4-35 (DOR 4-35) and 5-4 (DOR 5-4) are the most effective inhibitors (after TIRES itself) of the interaction IRES-MRR of pi 16. These results confirm that the consensus sequence identified is a determinant of the binding of MRR ρl 16 to whole TIRES.

In addition, the results of the affinity studies of MRRpl 16 for the different fragments of TIRES (FIG. 7B) by “filter-binding assay” show that the consensus fragment nt 56-92 is sufficient to allow the binding of MRRpl 16. Au on the contrary, the region 73-92 corresponding to the apical loop of region II (Ilb) is not sufficient for the binding of this polypeptide.

EXAMPLE 3 In Vitro Screening Test

The advantage of the present discovery is to seek to inhibit the binding of the MRR of pi 16 to the consensus sequence SEQ ID 3 of the region II of TIRES to prevent the initiation of translation and consequently protein synthesis by HCV.

Among the potential molecules, the Applicant has selected aminoglycosides. Aminoglycosides represent a class of chemical molecules that interact specifically with certain folded RNA molecules, such as 16S ribosomal RNA, ribozymes, and the TAR region of HIV. However, the specificity of these molecules with regard to HCV TARN as well as their capacity to inhibit IRES-dependent translation of HCV TIRES has not been previously demonstrated.

The screening test is carried out as follows. The pi 16 MRR and the region II consensus sequence are incubated in the presence of different aminoglycosides. The RNA mixture is then deposited on a nitrocellulose membrane under the same conditions as in Example 2. The results are shown in FIG. 8. Among the 15 aminoglycosides tested at 4 different concentrations, the compounds tobramycin and streptomycin inhibit the formation of the RNA-protein complex at all the concentrations tested. Tobramycin inhibits 43% of the MRRpl 16/11 complex at a concentration of 20 μM and 54% at 40 μM. Streptomycin, on the other hand, inhibits 25% of the MRRpl 16/11 complex at a concentration of 20 μM and 36% at 40 μM In contrast, neomycin and sisomycin only inhibit the complex at concentrations greater than 40 μM .

The aminoglycosides kanamycin A, kanamycin B and tobramycin are molecules of very similar structure. However, tobramycin (43% inhibition at 20 μM) is more active than kanamycin B (22% inhibition at 20 μM) which is in turn more active than kanamycin A (0% inhibition at 20 , 40 and 80 μM). This indicates that the presence of an aminogroup in position R2 (tobramycin and kanamycin B) promotes the dissociation of the complex, while the presence of the hydroxyl group in position RI is unfavorable there (kanamycin A) On the other hand, the amino group in position 6 is not directly involved in the interaction and can therefore be used to introduce modifications allowing the necessary active concentrations to be reduced.

Example 4; Inhibition of cap-independent translation ex vivo

In this example, a correlation is established between the results obtained in the in vitro screening system (Example 2) and those of a bicistronic cell system. This makes it possible to verify that the inhibition of the formation of the protein pi 16 eIF3 / RNA complex of TIRES of the HCV by a chemical molecule also results in an inhibition of the translation dependent on TIRES in cells ex vivo. Furthermore, this test makes it possible to detect a possible toxicity of the molecule in question for the cell itself, while measuring its effect on the cap-dependent translation.

a / Preparation of bicistronic vectors Bicistronic constructs consisting of a first cistron corresponding to the Renilla luciferase gene, followed by the IRES sequence, followed by a second cistron corresponding to the Firefly luciferase gene (pRluc-IRES-Fluc) are prepared as follows. A plasmid pRL-SV40 (Promega) is linearized with Xba I and dephosphorylated. At the same time, TIRES is amplified with the Firefly luciferase gene by PCR, in the presence of complementary oligonucleotides containing the Xbal sites. The PCR products are then subcloned into the plasmid pTrcHis2-TOPO (Invitrogen) in order to control digestion. The Tinsert ligation containing TIRES with the luciferase gene Firefly and the linearized vector pRL-SV40 is carried out using T4 DNA ligase (Biolabs).

b / HeLa cell transfection 10 ⁷ HeLa cells suspended in serum-free DMEM are transfected with 1 to 2.5 μg of plasmid pRluc-IRES-Fluc by electroporation at 0.5 V for 30 milliseconds using a Puiser gene (BioRad). The cells are then cultured in 24 or 96-well plates in the presence of different aminoglycosides, at concentrations between 2 and 5 mM for 24-36 hrs. The activity of Renilla luciferase (cap-dependent translation) and that of Firefly luciferase (cap-independent translation = viral) in cell lysates is measured and compared using the Dual-luciferase test (Promega) and the Lumat LB9507 luminometer ( Berthold).

c / Results The effect of 10 different aminoglycosides on the TIRES-dependent translation and on the cap-dependent translation is studied using HeLa cells transfected with the bicistronic construct (FIG. 9A).

According to the results appearing in FIG. 9B, among the 9 aminoglycosides tested, at ImM concentration, tobramycin inhibits the synthesis of Firefly luciferase controlled by TIRES of the hepatitis C virus by 90.4%, while the synthesis of Renilla luciferase, controlled by "cap", is not inhibited (168% of the control). A similar effect is observed at a concentration of tobramycin of 2 mM (the synthesis of IRES-luciferase is inhibited at 83% and that of cap-luciferase is inhibited only at 27%).

Hygromycin and G418 inhibit both cap-dependent and IRES-dependent translation in an IRES-nonspecific manner.

The effect of paromomycin at concentrations 1, 2 and 5 mM is more pronounced on IRES-dependent translation (inhibited at 37%) than on cap-dependent translation (inhibited at 6.3%) and is therefore moderately IRES- specific. When the quantity of RNA produced in the cells is increased, using a higher concentration of plasmid DNA (FIG. 9C), the effect of tobramycin is less pronounced (inhibition of IRES-dependent translation by 36.5% to 2mM and from 69% to 5mM), with a cap-dependent synthesis not inhibited (268 and 134% of control).

Under the same conditions, streptomycin inhibits the translation of the two cistrons in an IRES-non-specific manner (from 5mM in concentration). Thus, certain aminoglycosides are capable of inhibiting TIRES-dependent translation in an IRES-specific manner, without inhibiting the translation of the host cell. These non-toxic molecules for the cell at the indicated concentrations can be used to treat hepatitis C.

The results obtained with the bicistronic system are consistent with those of the screening test developed by the Applicant (MRRpl 16 / domain II): the same molecule, tobramycin, has been identified as the most active in the two systems. This shows the relevance of the claimed screening test, which can be used to identify new inhibitors of eIF3 binding on TIRES and of IRES-dependent translation.

BIBLIOGRAPHY

1. Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton M. (1989) Science 244: 359-62 "Isolation of a cDNA clone derived from a blood-borne non-A, non-B viral hepatitis genome. "

2. Sizova DV, Kolupaeva VG, Pestova TV, Shatsky IN, Hellen CU (1998) J Virol 72: 4775-82, Specifies interaction of eukaryotic translation initiation factor 3 with the 5 'nontranslated regions of hepatitis C virus and classical swine fever virus RNAs. 3. Kieft J., Zhou K., Jubin R., Doudna J., RNA (2001), 7: 194-206, Mechanism of ribosome recruitment by hepatitis C IRES RNA 4. Buratti E, Tisminetzky S, Zotti M, Baralle FE (1998) Nucleic Acids Res 26: 3179-

87 Functional analysis of the interaction between HCV 5'UTR and putative subunits of eukaryotic translation initiation factor eIF3. 5. Zhao WD, Wimmer E. (2001) J Virol 75: 3719-30 Genetic analysis of a poliovirus / hepatitis C virus chimera: new structure for domain II of the internai ribosomal entry site of hepatitis C virus.

6. Block KL, Vornlocher HP, Hershey JW. Characterization of cDNAs encoding the p44 and p35 subunits of human translation initiation factor eIF3. (1998) J Biol Chem. 273: 31901-8.

7. Asano K, Vornlocher HP, Richter-Cook NJ, Merrick WC, Hinnebusch AG, Hershey JW. (1997) Structure of cDNAs encoding human eukaryotic initiation factor 3 subunits. Possible roles in RNA binding and macromolecular assembly. J Biol Chem. 272: 27042-52. 8. Stelz U, Spahn C, Nierhaus KH, (2000) Proc Natl Acad Scie USA 97, 4597-4602 "Selecting rRNA binding sites for the ribosomal proteins L4 and L6 from randomly fragmented rRNA application of method called SERF"

Claims

1 / Method for screening molecules according to which, in vitro: a / we incubate together the pi 16 subunit (SEQ ID4) of the protein eIF3, the nucleotide sequence of region II (SEQ ID2) of HCV TIRES or any sequence containing at least 10 successive nucleotides of region II (SEQ ID 2) of TIRES of HCV and the molecule to be tested, b / then detecting the possible formation of complex pi 16 / region II IRES, the absence of a complex evidencing the inhibitory capacity of the molecule tested, to inhibit the formation of said complexes, c / the molecules inhibiting the formation of complexes are selected. 2 / A method according to claim 1, characterized in that only the sequence of the recognition motif of the pi 16 protein (SEQ ID5) is incubated. 3 / A method according to claim 1, characterized in that only part of the region II is incubated and corresponds to the consensus nucleotide sequence SEQ ID3 or a sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3. 4 / A method according to claim 1, characterized in that the molecule to be tested is incubated in increasing doses. 5 / A method according to claim 1, characterized in that the detection is carried out by filtration of the mixture through a nitrocellulose membrane, then by measurement of the radioactivity bound to the membrane corresponding to the amount of RNA fixed on the membrane . 6 / A method according to claim 1, characterized in that Ton then tests, ex vivo, the influence of the molecule selected in c) on the cap-independent translation and the cap-dependent translation to retain only the molecules inhibiting the translation cap-independent without influencing cap-dependent translation. 7 / A method according to claim 6, characterized in that constructs bicistronic vectors consisting of two luciferase framing the sequence of region II (SEQ ID 2) or any sequence containing at least 10 successive nucleotides of region II (SEQ ID 2), or the consensus sequence (SEQ ID 3) or a sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3; the first luciferase being translated in a cap-dependent manner and the second in a cap-independent manner or vice versa. 8 / Use of the molecules selected at the end of the screening process which is the subject of one of claims 1 to 7 for the manufacture of a medicament intended for the treatment of hepatitis C (HCV), swine fever (CSFV), bovine diarrhea (BVDV). 9 / Use of an aminoglycoside for the manufacture of a medicament intended for the treatment of hepatitis C (NHC), swine fever (CSFV), bovine diarrhea (BVDV). 10 / Use according to claim 9, characterized in that Taminoglycoside is tobramycin. 11 / Pharmaceutical composition comprising an antisense oligonucleotide complementary to the sequence SEQ ID 3 or of any sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3, with the exception of T oligonucleotide of sequence TAGACGCTTTCTGCGTGAAGACAGTAGT. 12 / Use of an antisense oligonucleotide complementary to the sequence SEQ ID 3 or of any sequence comprising at least 8 successive nucleotides of the sequence SEQ ID 3, with the exception of the oligonucleotide of sequence TAGACGCTTTCTGCG TGAAGACAGTAGT, as a medicament, for the treatment of hepatitis C (HCV) (deletion).