WO1996032473A1 - Dumbbell sense oligonucleotide for inhibiting herpes simplex virus (hsv) - Google Patents

Abstract

A dumbbell sense oligonucleotide for inhibiting herpes simplex virus (HSV), comprising a paired double-stranded oligodeoxyribonucleotide fragment closed at each end by a loop consisting of a non-self-paired covalent bond, and being characterised in that said paired double-stranded fragment comprises an ICP4 protein binding site sequence.

Description

 OLIGONUCLEOTIDE SENSE INHIBITOR OF HERPES SIMPLEX VIRUS (HSV) WITH HALTERED STRUCTURE

The present invention relates to closed double-stranded oligodeoxynucleotides of the sense-type herpes simplex virus inhibitors and anti-viral pharmaceutical compositions comprising them.

Herpes simplex viruses type 1 and type 2 (HSV-1 and HSV-2) are human viruses which belong to the family of herpesviridae, an important group of viruses, six of which are strictly human: HSV-1 and HSV-2 , chickenpox and shingles virus (VZV), cytomegalovirus (CMV), Epstein-Barr virus (EBV) and herpesvirus type 6 (HHV-6) (Roizman, 1990).

Herpes simplex viruses, both dermotropic and neurotropic, give infections of remarkable diversity in terms of their clinical manifestations and their degree of severity. Besides certain more or less well tolerated infections (oral herpes, genital herpes), the two types of herpes (HSV-1 and HSV-2) can be at the origin of infections of extreme gravity: herpetic encephalitis postnatal, neonatal herpes, herpetic keratitis, immunocompromised herpes (Whitley, 1990).

In the past twenty years, the frequency of infections with herpes simplex viruses, in particular genital herpes, and their prevalence is increasing in western and industrialized countries. In the USA, between 5 and 25 million people are estimated to be affected and more than 300,000 new cases are diagnosed each year. In France, an estimate of 310,000 annual outbreaks is given, including 82% of recurrent forms and 18% of initial forms.

Most of the antivirals available today are analogues of the bases or nucleosides that make up nucleic acids.

This class of drugs works by blocking viral replication, by acting on replication enzymes. For example, aciclovir, a purine analog, is used in the treatment of virus infections herpes simplex type 1 (HSV-1) and type 2 (HSV-2). After phosphorylation by viral thymidine kinase in infected cells, aciclovir becomes active and constitutes a decoy for viral DNA polymerase. Herpes virus 1 and 2 strains resistant to treatment with aciclovir appear, especially in immunocompromised patients. Viral strains resistant to aciclovir usually carry a mutation in the gene encoding thymidine kinase, more rarely in that of DNA polymerase. We are looking for an antiviral therapy which acts selectively on the multiplication processes of the virus without affecting the host cell, therefore devoid of toxicity.

In order to specifically combat herpes simplex virus infections, new strategies based on the use of synthetic oligonucleotides have been developed. A major research effort has been devoted to improving the stability of therapeutic oligonucleotides. It is indeed known that the oligonucleotides are degraded, mainly by 3 'exonucleases in cells and in serum (Eder et al., 1991). In the context of antisense approaches, numerous chemical modifications of nucleotides and internucleotide bonds have been developed in order to obtain oligonucleotides resistant to nuclear attacks but retaining properties of pairing to their targets (Stein and Cheng, 1993; Milligan and al., 1993; Hélène and Saison-Behmoaras, 1 994). More recently, phosphodiester antisense oligonucleotides capable of adopting a so-called "staple" structure have been used to inhibit the proliferation of the herpes simplex virus type 1 or 2 (Poddevin et al., 1994).

Sense oligonucleotides are pharmacological agents making it possible to inhibit the binding of a transcriptional regulator to a promoter whose activity it controls (Clusel et al., 1993; Blumenfeld and Vasseur, 1994). The inhibition of the biological activity of a transcriptional regulator by competition with oligonucleotide decoys comprising its specific binding sequence makes it possible to modulate the expression of the genes placed under the control of the targeted transcription factor. The therapeutic activity of an oligonucleotide depends on the target chosen. According to the present invention, an oligonucleotide is used comprising a binding sequence for the transcription factor ICP4 encoded by the Herpes Simplex viruses.

In the case of HSV, the first genes to be expressed after infection and before any synthesis of viral protein, are the immediate early genes where α: α4, αO, α27, α22 and α47 (Roizman and Sears, 1990) . Among the α gene products, at least three (α4, α 0 and α27) are involved in regulating the transcription of the early (β) and late (γ) genes of HSV-1. ICP4, the product of the α4 gene from HSV-1, is involved in activating the transcription of the early and late genes of the virus (Preston, 1979; Dixon and Schaffer, 1980, DeLuca and Schaffer, 1985). Among the many possible targets for a sense oligonucleotide, it has been observed that the choice of the ICP4 binding site constitutes a very effective target for blocking the progression of the viral cycle.

The ICP4 protein is a very early viral protein, which has an affinity for certain DNA sequences, and whose activity of transactivation of viral genes is required for the progression of the viral cycle.

The biological activity of sense oligonucleotides depends both on their stability under physiological conditions and on their binding affinity to the targeted proteins. The stability of sense oligonucleotides can be increased by introducing covalent bonds between the two strands to give a circular or closed double-stranded structure of the dumbbell type (WO92 / 19732; Clusel et al., 1993; Chu and Orgel, 1992). The oligonucleotides with a dumbbell-type structure consist of a paired double-stranded oligonucleotide structure framed and closed at each end by a loop consisting of a covalent link joining the two strands of said double-stranded fragment. This loop is generally a non-self-paired oligonucleotic sequence, but it may be a non-nucleotide chemical covalent link. However, the binding affinity of circular or closed double-stranded oligonucleotides of their targets may nevertheless decrease due to steric constraints linked to the closed structure. In this case, the closure or circularization of a sense double-stranded oligonucleotide does not represent an advantage, but a loss of the biological potential of the sense compounds.

It has been discovered according to the present invention that with respect to a closed double-stranded dumbbell-type oligonucleotide capable of binding the transcription factor ICP4 encoded by the gene α4 of the HSV virus and having properties to inhibit the multiplication of the virus, the production of a therapeutically effective oligonucleotide does not depend on the simple addition of covalent bonds to the ends of a double-stranded sequence comprising a sequence corresponding to said binding site.

The design of a closed double-stranded sense oligonucleotide with a dumbbell-type structure according to the present invention indeed requires the establishment of certain proper rules, the protein ICP4 binding to DNA, in order to establish the minimum length which must separate the site for fixing the ends of the double-stranded portion of the molecule.

According to the present invention, the interactions between the ICP4 protein and different binding sequences have been studied on different structures of open, hairpin or closed double-stranded oligonucleotide. Closed double-stranded oligonucleotides are circular molecules that exhibit greater stability than open or hairpin molecules of the same sequence. It has been discovered that dumbbell oligonucleotides may have a specific binding affinity for the ICP4 protein equal to or less than that of double-stranded oligonucleotides, depending on the position of the fixing sequence relative to the ends of the double-stranded portion of the oligonucleotide closed. Closed double-stranded oligonucleotides which exhibit an ICP4 binding site sequence identical to that of double-stranded or hairpin oligonucleotides have higher antiviral activity than that obtained with double-stranded or hairpin structures. The present invention therefore relates to oligonucleotides meaning inhibitor of herpes simplex virus (HSV) with dumbbell structure comprising a paired oligodeoxyribonucleotide fragment closed at each end by a loop consisting of a covalent link, characterized in that said paired double fragment comprises a binding site sequence for the ICP ₄ protein.

ICP4 is a protein which specifically binds to DNA sequences containing the ATCGT C consensus motif. It also binds to a series of other sequences diverging from this consensus sequence (Faber and Wilcox, 1988; DiDonato et al. , 1991; Everett et al., 1991).

In a particularly suitable embodiment, the oligonucleotide according to the present invention comprises the sequence ATCGTC.

The dumbbell type oligonucleotides according to the invention comprise from 20 to 100, preferably 30 to 70 nucleotides. Preferably said sequence of the ICP4 binding site itself comprises from 8 to 20 nucleotides.

In one embodiment, the dumbbell oligonucleotide therefore comprises a double-stranded structure of 13 base pairs containing a sequence for recognition of the viral transcription factor ICP4 of the herpes simplex virus type 1 consisting of one of the sequences: ATCGTCCATACCG ATCGTCCACACGG ATCGTCTCTCCGG Preferably the recognition sequence for the transcription factor ICP4 is the sequence ATCGTCTCTCCGG.

The sequences mentioned above correspond to the direct strand shown from left to right in the 5 ′ to 3 ′ direction. It has been discovered that preferably in the paired double-stranded fragment, the sequence of the ICP4 protein binding site must be flanked by at least 3 nucleotides at each end (which therefore corresponds to at least 3 base pairs in the double-stranded fragment) .

Advantageously, the sequence of the ICP4 binding site is framed by 3 to 6 nucleotides at each end.

In one embodiment, said nucleotides which frame the sequence of the binding site correspond to the nucleotides naturally present in this position.

Advantageously, in the oligonucleotides the internucleotide links are natural phosphodiester links.

The loop at each end connecting the two strands of the double-stranded fragment can be formed by various molecular structures and not necessarily nucleotide. However, advantageously, the oligonucleotides according to the invention comprise a double-stranded structure the ends of which are connected by a short loop formed by 3, 4 or 5 non-self-paired nucleotides.

The following sequences constituting loops having the best properties are cited in particular: G A A A, G A A, T T T T,

The preparation of closed oligonucleotides from linear oligonucleotides by ligation techniques is known. In particular, the closed double-stranded sense oligonucleotides which are the subject of the invention can be obtained chemically, biologically, or by approaches calling upon combinations of the techniques of synthetic chemistry and molecular biology. The dumbbell oligonucleotides used here can therefore be prepared from linear double-stranded oligonucleotides, then closed by chemical or enzymatic techniques. Various methods of chemical synthesis of natural oligonucleotides have been developed and are well known. For example, and without limitation, the oligonucleotides can be synthesized by the phosphoramidite method (Caruthers 1985) or by phosphonate chemistry (Froehler et al., 1986).

By using one or the other of these techniques or any other sequential procedure allowing the chemical synthesis of polynucleotide chains of sequence determined in advance, linear oligonucleotides are obtained which it is possible to close by biological means, with the using ligation enzymes, or chemically.

In the case of enzymatic ligation, the oligonucleotides must contain a 5 * phosphate terminal group, the phosphorylation of the 5 ′ terminal having been carried out chemically, or else biologically using a kinase (polynucleotide kinase) and ATP or any other phosphate donor.

In the case of closure by chemical ligation, the oligonucleotides can comprise a terminal phosphate group either in 3 ′ or in 5 ′, and the chemical ligation can be carried out in the presence of reagents such as carbodiimide or cyanogen bromide or any other reagent capable of catalyzing the formation of an internucleotide link (Dolinnaya et al., 1993).

Other chemical routes for the preparation of closed double-stranded oligonucleotides can be used. Linear oligonucleotides can be synthesized by the usual routes, then closed by chemical ligation or by a link involving terminal nucleotides.

Different nucleotides can be used in the formulation of oligonucleotides. The oligonucleotides forming the subject of the present invention are composed of a nucleotide base sequence comprising in particular adenine (A), guanine (G), cytosine (C), thymine (T) and uracil (U), linked together by internucleotide bonds, in particular natural, that is to say phosphodiesters. The oligonucleotides according to the invention can also contain rare nucleotides (Inosine, I, or ri for example) or modified nucleotides, either in deoxyribo- series or in ribo- series.

The oligonucleotides according to the invention can comprise reactive nucleotides, capable of establishing links with the sequence of the target molecule complementary to the oligonucleotide, or, in another application, intra-molecular links within the oligonucleotide .

Thus, the oligonucleotides according to the invention can carry reactive groups grafted onto the nucleotides, such as for example psoralen groups, or other bridging agents or intercalating agents which can react with the sequence of the target molecule complementary to the oligonucleotide In a particularly interesting case of the reactive groups grafted onto some of the nucleotides of the oligonucleotide could induce the formation of an intramolecular bridging within the molecule itself.

The oligonucleotide may have internal bonds produced by reactive agents belonging to or not belonging to the structure of the molecule itself.

Also part of the invention of so-called chimeric oligonucleotides, constituted by the covalent assembly of nucleotide and non-nucleotide fragments, in particular as regards the loops at the ends.

The oligonucleotides according to the invention can be associated with a molecular structure promoting penetration, targeting or cellular or intra-cellular addressing.

Among molecules making it possible to increase their intracellular penetration, mention may in particular be made of lipophilic groups, polypeptides or proteins. The present invention also relates to a pharmaceutical composition comprising, as active principle, at least one oligonucleotide according to the invention and an excipient in particular suitable for topical or systematic application.

These compositions are particularly useful for combating viral infections due to the HSV-1 and HSV-2 viruses. They help control viral proliferation.

They can comprise another antiviral active principle such as an anti-HSV anti-sense oligonucleotide or an anti-HSV anti-viral nucleoside analogous compound, such as acyclovir.

In one embodiment, the antisense oligonucleotide is directed against a sequence of pre-mRNA IE ₄ and IE ₅ of the HSV virus.

Preferably it comprises an oligonucleotide comprising the sequence A T C G T C T C T C C G G as an ICP4 binding site. It is in particular the dumbbell oligonucleotide GT5210 represented in FIG. 1.

Adequate dosage formulations can be established to optimize the delivery of these molecules to their target cells. Thus, for example, the compounds according to the invention could be encapsulated in liposomes, nanoparticles, LDL particles, or in any other type of microspheres allowing adequate preservation and promoting targeting. The oligonucleotides can also be combined with cationic surfactants.

Other characteristics and advantages of the present invention will appear in the light of the examples which will follow with reference to FIGS. 1 to 5. Figure 1 shows the oligonucleotides used in the examples. The oligonucleotides were synthesized and used in the experiments described in the examples, either in open double-stranded form, or in the form of a hairpin, or in closed double-stranded form. Four families of oligonucleotides were synthesized, the sequences of which are presented in Figure 1.

The consensus binding sequence of ICP4 (DiDonato et al., 1991) is underlined in each type of oligonucleotide. The oligonucleotide GT7202, double-stranded, 24 base pairs (bpd), contains a binding site for ICP4 present in the promoter of the gene coding for the glycoprotein gD (Faber and Wilcox, 1986); in this oligonucleotide the binding sequence of ICP4 (13 bpd, underlined in the figure) is surrounded by 5 and 6 bpd upstream and downstream of the site, respectively.

The oligonucleotide GT7201 has 17 bpd and contains the same ICP4 binding site as GT7202 but having only two bpb upstream and downstream of the ICP4 binding sequence (13 bpb, underlined in the figure).

The double-stranded GT 7209 oligonucleotide of 24 base pairs contains an ICP4 binding site present in the promoter of the α4 gene coding for ICP4 (Faber and Wilcox, 1988); in this oligonucleotide the binding sequence of ICP4 comprises 13 bpd (underlined in the figure), it is surrounded by 5 and 6 bpd upstream and downstream of the site, respectively.

The 24-base pair double-stranded oligonucleotide GT 7210 contains an ICP4 binding site located 740 bp upstream of the α0 gene, and forming part of a functional promoter of the α / β gene classes (Bohenzky et al., 1993 ); in this oligonucleotide the binding sequence of ICP4 (13 bpd, underlined in the figure) is surrounded by 5 and 6 bpd upstream and downstream of the site, respectively. The double-stranded oligonucleotide GT7212 of 24 base pairs corresponds to the negative control for binding of ICP4. In this oligonucleotide, the ICP4 binding sequence corresponding to the site present in GT7210 was mutated at seven positions (nucleotides indicated in small letters in the figure). This mutated sequence is no longer linked by the ICP4 protein (see Figure 3).

The oligonucleotides GT6201, GT6202, GT6209, GT6210 and GT6212, have a hairpin structure, and correspond to the double-stranded sequences GT7201, GT7202, GT7209, GT7210 and GT7212, respectively, linked by a 5T loop at one end.

The oligonucleotides GT5201, GT5202, GT5209, GT5210 and GT52 12, have a double-stranded dumbbell structure, and correspond to the double-stranded sequences GT7201, GT7202, GT7209, GT7210 and GT7212, respectively, linked by a 5T loop at each end.

FIG. 2 represents the results of binding of ICP4 to a closed double-stranded oligonucleotide and shows that the binding depends on the position of the binding site in the double-stranded part of the closed molecule.

The relative affinities of ICP4 with respect to the oligonucleotides GT7201, GT7202, GT6201, GT6202, GT5201 and GT5202 were determined by delay gel competition experiments according to the technique described in Mat. and Meth., using GT7202 as a radioactive probe. The radioactivity bound to ICP4 in the presence of a 100-fold molar excess of each oligonucleotide was quantified, and the relative affinities were determined by considering as 100% the competition observed for the cold oligonucleotide GT7202.

The sequences of the oligonucleotides are those described in Figure 1.

Figure 3 shows the autoradiography of a gel delay showing that the ICP4 protein binds to different binding sites with different relative affinities. The sequences of the oligonucleotides are described in Figure 1.

The gel retardation technique is described in Mat. and Meth.

The specific band corresponding to ICP4 is indicated by an arrow; nsMm and nsMi indicate, respectively, non-specific cellular or viral binding activities.

ICP4 binding reactions are carried out with the radioactive probe GT7210 and 10 μg of nuclear proteins from MRC5 cells infected with HSV-1, in the absence (tracks 2, 3 and 10) or in the presence of the indicated quantities of GT7210 competitors (track 4-6), GT6210 (tracks 7-9),

GT7209 (tracks 11-13) or GT6209 (tracks 14-16).

The lower panel shows a longer exposure of the same autoradiography.

FIG. 4 represents the autoradiography of a gel delay showing that the closed double-stranded oligonucleotides containing an ICP4 site of high relative affinity are recognized specifically by ICP4 and with an affinity similar to that of hairpin molecules or double-stranded molecules.

In this Figure, the competition experiments with the non-radioactive oligonucleotides GT7210, GT6210, GT52 10 and controls GT7212, GT6212 or GT5212 are represented.

The sequences of the oligonucleotides are those described in Figure 1. The gel delay technique is described in Mat. and Meth.

The specific band corresponding to ICP4 is indicated by an arrow; nsMm indicates non-specific cell binding activity. The ICP4 binding reactions are carried out with the radioactive probe GT7210 and 10 μg of nuclear proteins from MRC5 cells infected with HSV-1, in the absence (tracks 2 and 12) or in the presence of the indicated quantities of competitors GT7210 (tracks 3- 5), GT6210 (tracks 6-8), GT5210 (9-11). The control oligonucleotides GT7212 (lane 13), GT6212 (lane 14) and GT5212 (lane 15) were used in molar excess of 500 times.

FIG. 5 represents the comparative study of the anti IISV-1 activities of the double-stranded sense oligonucleotides GT7210, the hairpin GT6210 and the closed double-stranded GTS 210.

Duplicate MRC5 cells are treated with the indicated concentrations of oligonucleotides GT72 10, GT62 10 or GT52 10, or control oligonucleotides GT7210, GT6212 or GT5212 (see FIG. 1), in the presence of 10 μg / ml of lipofectin, as described in Mat. and Meth.

The 0% inhibition activity (virus control) corresponds to the titer of the virus treated with lipofectin in the absence of oligonucleotides. Under the experimental conditions described, the titer of the virus control treated with lipofectin corresponds to 90% of the titer of the untreated virus.

The results presented correspond to an average of 2 independent experiments; the values of the standard deviations do not exceed 10% GT5210 (circles); GT6210 (squares); GT7210 (rhombus); GT52 12 (cross); GT6212 (dash); GT7212 (triangle).

EXAMPLE 1 EVIDENCE OF THE FIXING OF THE ICP4 TRANSCRIPTION FACTOR OF THE HSV-1 VIRUS BY SENSE-TYPE OGONUCLEOTIDES

1. Preparation of closed sense oligonucleotides from linear oligonucleotides by the chemical ligation technique. The experimental conditions allowing efficient chemical ligation of the oligonucleotides synthesized in 3'P form are as follows: 100 DO (1 DO ₂ 60 = 33 μg / ml) of linear oligonucleotide 3'P are dissolved in 450 μl of MES pH buffer 7.5, and are incubated 1-2 min at 95 ° C, then cooled slowly for 2-3 h, to finally be transferred to 4 ° C. 20 mM MgCl ₂ and 0.5 M CNBr are added (final volume = 500 μl) and the mixture is incubated at 4 ° C for 1 min. After incubation, the reaction is stopped by adding 100 μl of 3M potassium acetate and 1 ml of absolute ethanol. The oligonucleotides are precipitated, and the closed double-stranded oligonucleotides are purified by reverse-phase high pressure liquid chromatography (RP-HPLC), using a PRP-3 Cl 8 column (300 A, 10 μm).

2. Preparation of nuclear extracts from MRC5 cells infected with HSV-1.

MRC5 cells (2 x 10 ⁷ cells) are infected with HSV-1 at an ego (multiplicity of infection) of 10 pfu / cell. After 1 hour of adsorption at 37 ^β C, the cells are rinsed 2 times with PBS buffer (Phosphate Buffered Saline), then covered with complete MEM medium supplemented with 2% fetal calf serum. Infected cells are harvested 16 hours after infection.

Uninfected control cells are prepared in parallel according to the same protocol. Nuclear extracts from infected and uninfected cells were prepared according to the method described by Cereghini et al.

(1987), with as modifications the introduction of 5 μg / ml of aprotinin and leupeptin.

3. Gel delay technique.

In order to envisage a possible therapeutic application of the closed double-stranded oligonucleotides described in the present invention, it is essential to examine the capacity of these oligonucleotides to bind specifically to their target, in an in vitro study system. The The gel retardation technique (Fried and Crothers, 1981; Garner and Revzin, 1981) is a very sensitive method which allows the specificity and the relative affinities of DNA-binding proteins to be analyzed quickly and simply. for their binding sites present in oligonucleotides.

The oligonucleotides are labeled at the 5 ′ end by phosphorylation with ATP (γ-32p = 3000 Ci / mmol) and the polynucleotide kinase of bacteriophage T4, according to standard protocols. The binding reactions of the ICP4 protein with the different oligonucleotides are carried out in a reaction mixture of 20 μl containing: 10 mM Hepes pH 7.9, 10% (vol / vol) of glycerol, 100 mM NaCl, 0.1 mM EDTA, 0.5 mM DTT , 0.25 mM PMSF, 2.5 μg / ml aprotinin, 2.5 μg / ml leupeptin, 1 ng (25,000 cpm) radioactive probe, 1 μg DNA of sonicated salmon sperm, 1 μg poly (dI-dC) .poly (dIdC) , and 2-10 μg of nuclear extract from HSV-1 infected MRC-5 cells. After an incubation of 10 minutes in ice, the samples are subjected to electrophoretic migration at 12 V / cm in a non-gel. denaturant at 6% polyacrylamide containing 0.25X TBE.

The gel is fixed and dried to then be subjected to autoradiography. The bands corresponding to the ICP4-radiaoctive probe complex and to the free probe are identified using autoradiography, and quantified in a liquid scintillation radioactivity counter.

The comparison of the relative affinities of ICP4 for the different oligonucleotides is carried out by competition experiments with non-radiolabelled oligonucleotides. In this case, the radioactive probe containing an ICP4 binding sequence is added to the reaction mixture at the same time as the cold competitor oligonucleotide.

4. Results

4.1. The affinity of ICP4 for its specific binding site contained in a closed double-stranded oligonucleotide depends on the position of the binding site relative to the ends of the double-stranded portion of the closed molecule. As shown in FIG. 2, the relative affinity of the ICP4 protein with respect to a particular binding sequence contained in an open double-stranded or hairpin structure, does not vary according to the position of this site relative to at the ends of the molecule. Thus, in the case of the site corresponding to the binding sequence present in the promoter of glycoprotein D, ICP4 binds with the same affinity to the consensus site contained in the double-stranded molecules GT7201 (FIG. 1, two bpds upstream and downstream of the consensus site) and GT7202 (Figure 1, five and six bps upstream and downstream of the consensus site, respectively), or hairpin GT6201 (Figure 1, two bps upstream and downstream of the consensus site) and GT6202 (Figure 1, five and six bps upstream and downstream of the consensus site, respectively).

On the other hand, with closed oligonucleotides, ICP4 binds with an affinity reduced by 5-10 times (Figure 2) when the consensus sequence is surrounded by only two bpds upstream and downstream (GT5201; Figure 1), but has an affinity comparable to other structures when the consensus sequence is surrounded by five and six bps upstream and downstream, respectively (GT5202; Figure 1).

These results show that these double-stranded molecules have constraints linked to their structure which induce conformations not suitable for the specific binding of DNA-binding proteins. In certain cases as for GT5201, the circularization of the double-stranded oligonucleotide GT7201 does not produce the expected advantage concerning the stabilization of the molecule, but rather induces the loss of the biological activity. The production of a closed double-stranded sense oligonucleotide effective for binding ICP4 therefore requires the establishment of certain rules specific to this type of DNA-binding protein as regards the minimum length which must separate the binding site from the ends of the double-stranded portion of the molecule. The production of a closed double-stranded sense oligonucleotide therefore does not involve the simple addition of covalent "linkers" at the ends of a known double-stranded sequence. 4.2. The affinity of ICP4 varies depending on the sequence of the binding site.

As shown in Figure 3, the competition profiles of the double-stranded oligonucleotides GT7209 and GT7210 are different, the affinity of ICP4 for the oligonucleotide GT7210 being 10 times that of GT7209 (Figure 3, compare lanes 5 and 6 with tracks 12 and 13). The same type of results are obtained with the hairpin oligonucleotides (GT6209 and GT6210; Figure 3: compare tracks 8 and 9 to tracks 15 and 16). Similar results were observed for the oligonucleotides GT6202 and GT7202 (results not shown).

In all oligonucleotides, the length of the double-stranded portion (24 bpd), as well as the position of the site (5 and 6 bpd upstream and downstream of the site, respectively) are comparable, the difference being in the sequences of the sites themselves. fixation. The oligonucleotides GT6209 and GT7209 contain a binding site for ICP4 present in the promoter of the α4 gene coding for ICP4 (Faber and Wilcox, 1988); the oligonucleotides GT6202 and GT7202 contain a binding site for ICP4 present in the promoter of the gene coding for the gD glycoprotein (Faber and Wilcox, 1986); while the oligonucleotides GT62 10 and GT7210 contain a potential binding site located 740 bp upstream of the α0 gene, and forming part of a functional promoter of the α / f gene classes. (Bohenzky et al., 1993). This potential binding site therefore represents a binding site of high relative affinity.

4.3. The binding of ICP4 to a closed double-stranded "dumbbell" oligonucleotide containing a site of high relative affinity is specific and comparable to that relating to an open or hairpin double-stranded oligonucleotide.

Figure 4 shows a gel delay illustrating the comparison of the affinities of ICP4 with respect to open double-stranded, hairpin or double-stranded oligucleotide of the dumbbell type (GT7210, GT6210 and GT5210, respectively; Figure 1), containing a high binding site affinity (see Figure 3). In all these cases, the ICP4 binding sequence is surrounded by 5 and 6 bpd upstream and downstream of the site, respectively. It is clear in this figure that the binding of ICP4 on the radioactive probe GT7210 is displaced in a comparative manner by the different competing oligonucleotides (Figure 4, compare tracks 3-5, 6-8 and 9-11 to track 2) .

Furthermore, the interaction of ICP4 with the closed double-stranded oligonucleotide GT5210 is specific, as indicated by the result of the competition with a large excess of mutated closed double-stranded oligonucleotide (GT5212 -see Figure 1-; Figure 4, compare tracks 9-1 1 to track 15).

EXAMPLE 2 INHIBITION OF THE MULTIPLICATION OF HSV-1 BY SENSE-LIKE OLIGONUCLEOTIDES.

1. Study of the activity of sense oligonucleotides.

To reach their targets, the oligonucleotides, polyanionic molecules, must cross the plasma membrane. By forming a stable complex with oligonucleotides, cationic lipids such as lipofectin (Gibco) facilitate penetration and significantly increase the activity of antisense oligonucleotides (Chiang et al., 1991; Bennett et al., 1992; Capaccioli et al ., 1993). We therefore studied the antiviral effect of sense oligonucleotides in the presence of lipofectin, at concentrations compatible with cell viability greater than 90%, and without effect on virus-cell interactions.

2. Infection of cells in the presence or absence of sense oligonucleotides.

MRC-5 cells (ATCC) are cultured in MEM medium (Gibco) supplemented with 5% fetal calf serum, L-glutamine, non-essential amino acids and penicillin-streptomycin. After seeding at a density of 60,000 cells per well of 2 cm ² , the cells are incubated for 16 to 24 h at 37 ° C in a humid atmosphere enriched in CO ₂ (7%). The cells are rinsed with PBS and then exposed for 4 h to the oligonucleotide-lipofectin complexes distributed in a volume of 500 μl per well. The concentrations of oligonucleotides vary from 0.05 μM to 1 μM while lipofectin is used at a rate of 10 μg / ml.

After a 4 h incubation at 37 ° C, the medium is aspirated from each well and the cells are again rinsed with PBS and then covered with OPTI-MEM medium (Gibco). After a further one hour incubation at 37 ° C, the cells are infected with HSV-1 at a multiplicity of infection (m.o.i.) of 3 p. f.u. / cell in the presence or absence of oligonucleotides without lipofectin.

A volume of 250 μl of oligonucleotides in solution (2.x concentrates) in OPTI-MEM medium is deposited on the cell monolayer and, 5 minutes after, a viral solution of 50 μl is deposited in each well of the microplate. The incubation is continued for 1 hour with gentle shaking. After adding 200 μl per well of opti-MEM medium, the plates are maintained for 24 h at 37 ° C. Under the conditions described above - above, the oligonucleotide-lipofectin co mp proves to be slightly toxic (<10%) or non-toxic to cells. After the onset of the cytopathic effect, the reactions are stopped by freezing the plates in liquid nitrogen.

3. Virus titration.

The titrations were carried out by "yield reduction" according to the protocol described by Poddevin et al. (1994).

All measurements are performed in duplicate. The intracellular viruses are released into the culture medium after 3 rapid cycles of freezing in liquid nitrogen and thawing at 37 ° C. They are then diluted in serum-free medium to carry out the actual titration. The cells are seeded the day before in complete medium (2.5 x IV wells of 4 cm). The next day, the medium is aspirated and 200 μl of different viral dilutions are deposited per well. After an incubation of one hour at 37 ° C with gentle but constant shaking, the medium is aspirated and the cells are covered with complete medium (2.5% in serum) containing 1.2% methyl cellulose for a 3-day incubation at 37 ° C.

After 3 days, the medium is eliminated, the cells are fixed with a PBS / 10% formalin solution for 20 minutes and then stained with crystal violet. The plates are then rinsed and the areas counted by transillumination with the negatoscope. The titrations are carried out relative to the viral titers observed in the absence of oligonucleotides.

4. Results

Figure 5 shows a comparative study of the inhibition of the proliferation of HSV-1 induced by oligonucleotides containing a site of high relative affinity for ICP4 in the form of open double-stranded structure (GT7210), in hairpin (GT6210) or double-stranded closed (GTS 210), for concentrations ranging from 50 nM to 1 μM. The curves have a biphasic appearance, the inhibitory effect systematically decreasing at higher oligonucleotide concentrations. This phenomenon seems to be due to saturation of cationic lipids at increasing concentrations of oligonucleotides.

The closed double-stranded oligonucleotide GT5210 is immediately very active at very low concentration, since it induces nearly 80% inhibition at 50 nM, while the other double-stranded oligonucleotides, GT7210 and hairpin GT6210, are inactive at that same concentration. The antiviral activity of GT5210 increases with the concentration of the oligonucleotide, reaching 90% inhibition at 100 nM.

The hairpin oligonucleotide GT6210 produces a reduction in viral titer of 5% at 50 nM and a maximum of 60% at 250 nM, while the double-stranded oligonucleotide GT7210 produces a 20% inhibition at 50 nM and reaches 55% at 500 nM (Figure 5). Whatever their structure, the mutated control oligonucleotides GT7212, GT6212 or GT5212 are inactive at the concentrations tested.

The results illustrated in FIG. 5 show that the 50% inhibitory concentration (IC ₅₀ ) of the closed double-stranded oligonucleotide GT 5210 is less than 50 nM, and that the inhibitory concentration 90% (IC90) is 100 nM. No cytoto.xicity being observed at concentrations up to 100 μM, the therapeutic index for the closed double-stranded oligonucleotide GT 5210 is greater than 1000.

BIBLIOGRAPHY

1. Bennett, CF., M-Y. Chiang, H. Chan, J.E.E. Shoemaker and C.K. Mirabelli (1992), Mol. Pharmacol. 41.1023-1033.

2. Blumenfeld, M. and M. Vasseur (1994), Medicine / Science 10, 274-281.

3. Bohenzky, R.A., A.G. Papavassiliou, I.H. Gelman and S. Silverstein (1993), J. Virol. 67, 632-642.

4. Capaccioli, S., G. DiPasquale, E Mini, T. Mazzei and A. Quattrone (1993), Biochem. Biophys. Res. Comm. 197, 818-825.

5. Caruther, M.H. (1985), Science 230, 281-285.

6. Cereghini, S., M. Raymondjean, A. Garcia Carranca, P. Herbomel and M. Yaniv (1987), Cell 50, 627-638.

7-Chiang, M-Y., H. Chan, M.A. Zounes, S.M. Freier, W.F. Lima and CF. Bennett (1991), J. Biol. Chem. 266, 18162-18171.

8. Chu, B.C.F. and LE. Orgel (1992), Nucleic Acids Res. 20, 5857-5858.

9. Clusel, C, E. Ugarte, N. Enjolras, M. Vasseur and M. Blumenfeld (1993), Nucleic Acids Res. 21, 3405-341 1.

10. DeLuca, N.A. and P.A. Schaffer (1985), Mol. Cell. Biol. 5, 1997-2008. 1 1. DiDonato, J.A., J.R. Spitzner and M.T. Muller (1991), J. Mol. Biol. 219, 451-470.

12. Dixon, R.A.F. and P.A. Schaffer (1980), J. Virol. 36, 189-203.

13. Dolinnaya, N.G., Blumenfeld, M., Mernkova, I.N., Oretskaya, T.S., Krynetskaya, N.F., Ivanovskaya, M. G., Vasseur, M. and Shabarova, Z.A. (1993), Nucleic Acids Res. 21, 5403-5407.

14. Everett, R.D., M. Hliot, G. Hope and A. Orr (1991), Nucleic Acids Res. 19, 4901-4908.

15. Faber, S.W. and K.W. Wilcox (1986), Nucleic Acids Res. 14, 6067-6083.

16. Faber, S.W. and K.W. Wilcox (1988), Nucleic Acids Res. 16, S55-570. 17. Fried, M. and Crothers, DM. (1981), Nucleic Acids Res. 9, 6505-6525.

18. Froehler, B.C., Ng, P. and Mateucci, M.D. (1986), Nucleic Acids Res. 14, 5399-5407.

19. Garner, M.M. and Revzin, A. (1981), Nucleic Acids Res. 9, 3047-3060.

20. Hélène, C. and E. Saison-Behmoaras (1994), Médecine / sciences 10, 253-273. 21. Milligan, JF, MD Mateucci and JC Martin (1993), J. Med. Chem. 36, 1923-1937.

22. Poddevin, B., S. Meguenni, I. Elias, M. Vasseur and M. Blumenfeld (1994), Antisense Res. Dev. 4, 147-154. 23. Preston, CM. (1979), J. Virol. 29, 275-284.

24. Roizman, B. (1990). In: Fields, B., Knipe, D.M. (eds) Virology, Raven Press, New York, ppl787-1793.

25. Roizman, B and Sears, A.E. (1990). In: Fields, B., Knipe, D.M. (eds) Virology, Raven Press, New York, pp 1794-1841. 26. Shepard, A.A. and N.A. DeLuca (1991), J. Virol. 65, 787-795.

27. Stein, C.A. and Y.-C Cheng (1993), Science 261, 1004-1012.

28. Whitley, R.J. (1990). In: Fields, B., Knipe, D.M. (eds) Virology, Raven Press, New York, ppl 843- 1887.

Claims

CLAIM

1. Herpes simplex virus (HSV) inhibitor sense oligonucleotide with dumbbell structure comprising a paired double-stranded oligodeoxyribonucleotide fragment closed at each end by a loop consisting of a covalent link, characterized in that said double-stranded paired fragment comprises a sequence of a site of fixation for the protein ICP ₄ .

2. Oligonucleotide according to claim 1, characterized in that said sequence of an ICP4 site comprises the sequence A T C G T C.

3. Oligonucleotide according to claim 1 or 2, characterized in that said sequence of the ICP4 binding site comprises from 10 to 20 nucleotides.

4. Oligonucleotide according to one of claims 2 or 3 characterized in that the sequence of the binding site of the protein ICP4 consists of one of the following sequences: ATCGTCCATACCG ATCGTCCACACGG ATCGTCTCTCCGG

5. Oligonucleotide according to claim 4 characterized in that the sequence of the binding site consists of the sequence ATCGTCTCTCCGG.

6. Oligonucleotide according to one of claims 1 to 5 characterized in that in the paired double-stranded fragment, the sequence of the binding site of the ICP4 protein is framed on the same strand by at least 3 nucleotides at each end before the loop.

7. Oligonucleotide according to claim 6 characterized in that the sequence of the ICP4 binding site is framed by 3 to 6 nucleotides at each end before the loop.

8. Oligonucleotide according to claim 6 or 7 characterized in that said nucleotides which frame the sequence of the binding site to the ICP4 protein correspond to the nucleotides naturally present in this position.

9. Oligonucleotide according to one of claims 1 to 8 characterized in that the internucleotide links are natural phosphodiester links.

10. Oligonucleotide according to one of claims 1 to 9 characterized in that the loop consists of 3 to 5 nucleotides.

1 1. Oligonucleotide according to claim 10 characterized in that the loop consists of one of the following sequences: G A A A, G A A, TTTT. TTTTT.

12. Oligonucleotide according to one of claims 1 to 1 1, characterized in that the oligonucleotide consists in the dumbbell oligonucleotide GT 5210, represented in FIG. 1.

13. Anti viral pharmaceutical composition against a Herpes Simplex virus, in particular HSV1 and HSV2 comprising an oligonucleotide according to one of claims 1 to 12 and a pharmaceutically acceptable vehicle.

14. Pharmaceutical composition according to claim 13 characterized in that it comprises a vehicle suitable for topical or systematic application.

15. Pharmaceutical composition according to one of claims 13 to

14 characterized in that it further comprises a vehicle facilitating targeting and cell penetration.

16. Pharmaceutical composition according to one of claims 13 to 15 characterized in that it comprises another antiviral active ingredient such as an anti-HSV antisense oligonucleotide or an anti-HSV antiviral nucleoside analogous compound, such as acyclovir.

17. Composition according to claim 16, characterized in that the antisense oligonucleotide is directed against a sequence of pre-mRNA IE ₄ and IE ₅ of the HSV virus.