CA2452663A1

CA2452663A1 - Method for screening molecules which can bind to the gp120 protein of the immunodeficiency virus

Info

Publication number: CA2452663A1
Application number: CA002452663A
Authority: CA
Inventors: Claudio Vita; Catherine Royer; Vivian Pogenberg
Original assignee: Individual
Current assignee: Institut National de la Sante et de la Recherche Medicale INSERM; Commissariat a lEnergie Atomique et aux Energies Alternatives CEA
Priority date: 2001-07-06
Filing date: 2002-07-04
Publication date: 2003-01-16
Also published as: WO2003005032A2; FR2827046A1; FR2827046B1; WO2003005032A3

Abstract

The invention relates to a method for screening molecules which can bind to the gp 120 protein of the immunodeficiency virus. The inventive method uses the fluorescene anisotropy technique and the peptide of the following formula (I): TPA- Xaaa- Xaab- Ala or Gln or His - Arg ou Phe - Cys - Xaac- Xaad- Arg -Cys - Lys - Xaae- Xaaf- Xaag- Xaah- Leu ou Lys - Xaai - Lys - Cys - Ala or Gln - Gly or (D)Asp or Ser - Ser or His or Asn - Xaaj- Cys - Thr or Ala - Cys -Xaak-NH2, (I) wherein TPA represents thiopropionic acid, Xaaa, Xaab, Xaac, Xaad, Xaae, Xaaf, Xaag, Xaah, and Xaai are naturals or non natural amino acids which are identical or different, Xaaj represents b-naphtylalanine or phenylalanine or bi-phenylalanine, and Xaak represents Gly or Val or Ile.

Description

METHOD FOR SCREENING MOLECULES CAPABLE OF ATTACHING

VIRUS
DESCRIPTION
Technical field The present invention concerns a method for screening molecules capable of attaching themselves to l0 the gp120 protein of the immunodeficiency virus or its analogues. It uses a fluorescence anisotropy technique and a specific family of peptides with a high affinity and specificity for the gp120 viral protein.
Numerous immunodeficiency viruses exist, comprising a gp120 viral protein envelope or a protein analogous to this.. Among said viruses one may cite the human immunodeficiency virus (HIV), the simian immunodeficiency virus (SIV), the ovine immunodeficiency virus (VISNA), the bovine immunodeficiency virus (BIV) and the feline immunodeficiency virus (FIV).
Some of the peptides of the present invention are capable of attaching themselves to the gp120 viral protein or its analogues, inhibiting the attachment of the gp120 viral protein to the CD4 receptor, with IC5o values between 0.1 and 400 nM as a function of their sequence. Some of these molecules are capable of inhibiting the infection of T lymphocytes by the AIDS
virus with, for example, a EDso (effective dose) of 100 to 900 nM. In addition, the attachment of said molecules to B 13837.3 EE

the recombinant gp120 viral protein induces a conformational variation in said protein which unblocks new epitopes, with the same efficiency as the soluble CD4 molecule.
The screening method of the present invention is a powerful research tool that makes it possible to select new molecules of interest for the manufacture of new anti-AIDS medicines.
It consequently finds an obvious application in the development of new therapies and new means of diagnosis enabling AIDS to be combated.
State of the prior art A lot of research work has been carried out to find efficient treatments and means of preventing infection by the immunodeficiency virus and, in particular, the human immunodeficiency virus (HIV). This is due, in particular, to the variability of the virus, the difficulty in understanding the mechanism of its entry into the target cells and the immune response necessary for the protection of the infection by the virus.
As the analysis of the three dimensional structure of the CD4 - gp120 complex shows [1] (see appended references), the structural elements of the CD4 that are critical for its attachment to the gp120 viral glycoprotein include the amino acids G1y38, G1n40, G1y41, Ser42, Phe43, Thr45, which are included in the region 36-47, and the arginine-59 of the CD4. The region 36-47 of the CD4 has been assimilated to the region CDR2 of B 13837.3 EE

immunoglobulins and has a hairpin ordered structured.
Said structure is formed of a ~ strand (C'), corresponding to the sequence 36-40, followed by a bend corresponding to the sequence 40-43, which allows a second ~ strand (C " ), corresponding to the sequence 43-47, to form with the first an antiparallel ~ structure.
Said structure is stabilised by hydrogen bonds between the amide nitrogen of the G1y38, G1n40, Phe43, Thr45 and G1y47 residues of the first R strand and the carbonylic l0 oxygen of the residues, respectively, Thr45, Phe43, G1n40, G1y38, I1e36 of the second ~ strand. Said hairpin shaped structure plays a key role in the interaction of the CD4 with the gp120 viral protein and fulfils a double function: firstly, it makes it possible for the amino acids G1y38, G1n40, G1y41, Ser42, Phe43 and Thr45 to form a molecular surface complementary to the interaction surface of the gp120 viral protein and, in particular, makes it possible for the lateral chain of the Phe43 to insert itself in the entrance of a hydrophobic cavity in the molecular surface of the gp120 viral protein:
secondly, it enables a ~ sheet type interaction between the polypeptide backbones of the (3 strand C" of the CD4 and the (315 strand (residues 365-368) of the gp120 viral protein. Arginine 59 (Arg59) also plays a key role in the CD4 -gp120 interaction, since the guanidinium group of its lateral chain forms a double hydrogen bridge with the lateral chain of the Asp369 residue of the gp120 viral protein, allowing the stabilisation of the (3 structure type interaction between the C " strands of the CD4 and B 13837.3 EE

X15 of the gp120 viral protein. Mutagenesis experiments that have been carried out have shown that all of said residues of the CD4 play a critical role in the interaction with the gp120 viral protein [2], [3].
The reproduction of this hairpin ordered structure and the conformation of the lateral chains of its residues is critical in a molecule which has to attach itself to the gp120 viral protein at the interaction site of the CD4. The fact that simple peptidic constructions, such as a linear peptide corresponding to the sequence 37-53 and a cyclic peptide corresponding to the sequence 37-46 of the CD4, do not have measurable affinity for the gp120 viral protein is a demonstration of this [4].
The infection of CD4 cells by the AIDS virus is mediated by a high affinity interaction between the viral envelope and the CD4. Mutagenesis experiments and competition with antibodies have made it possible to localise, in the D1 domain of the CD4, the contact surface with the gp120 protein of the viral envelope. The three dimensional structure of a functional recombinant form of the gp120 glycoprotein, but deleted from the loops Vl-V2-V3 and the N- regions and C-terminals, in complex with the D1D2 domains of the CD4 and with the Fab fragment of a monoclonal antibody, has been recently resolved by crystallography !1]. In this structure, the CD4 interacts with a large hollow in the molecular structure (800 A°2) of the gp120 viral protein, by using a large molecular surface (742 A°2), which is centred around the region that has been assimilated to the CDR2 region B 13837.3 EE

of immunoglobulins. At the heart of the molecular hollow of the gp120 viral protein, a narrow and deep cavity opens up, the "Phe43 cavity", in which the lateral chain of the Phe43 , at the peak of the CDR2 region of the CD4 , 5 occupies the entrance. The resolution of said structure confirms the previous mutagenesis data, which indicate the Phe43 residue and the whole CDR2 loop as being functionally very important, and proposes said CDR2 loop as a possible target for the inhibition of the attachment of HIV-1 virions to the cells. The gp120 - CD4 interaction allows the virus to attach to the membrane of the target cells and represents the first protein -protein interaction in the infection mechanism.
Nevertheless, other co-receptors are necessary for an IS efficient entry of the AIDS virus into the cells. CXCR-4 [5] , [6] , is .involved in the entrance of the lymphotropic viral strains (X4) and CCR-5 [7] , [8] , which is involved in the entry of the M-tropic strains (R5) and most of the primary isolates. Several pieces of evidence indicate that the attachment of the gp120 viral protein on the CD4 could produce conformational variations in the glycoprotein of the envelope, which could expose new sites, detectable for specific antibodies, called CD4i, and could increase the affinity of the envelope for the receptors of the chemokines [9], [10], the entry co-receptors. After attachment of the CD4 and the co-receptors by the gp120 viral protein, other conformational variations could lead to a structural reorganisation of the gp4l, which could lead to its B 13837.3 EE

fusogene peptide, then the fusion of the viral and cellular membranes and finally to the entry of the viral burden in the cell.
The crystallographic structure of the gp120 viral protein complexed to the CD4 has made it possible to elucidate one of the mechanisms that the AIDS virus uses to escape the immune response. It was already known that the protein of the gp120 envelope has hypervariable and highly glycosylated regions. The crystallographic structure has in fact demonstrated that the hypervariable regions and the highly glycosylated regions form the exposed molecular surface of the gp120 viral protein, which also corresponds to the surface accessible to the virions. Only several regions of the molecular surface of the gp120 viral protein are conserved and may be the target of antibodies. Said regions are not normally accessible and are probably protected by the hypervariable loops (V1-V2-V3) which, in the crystallised protein, have been deleted, and are therefore invisible.
One of said regions is a surface near to the interaction site for the chemokine receptors, and accessible to the CD4i antibodies, after attachment of the envelope to the CD4. The bonding site to the CD4 is another exposed and conserved surface: however, said surface is present in a cavity of the gp120 viral protein and is therefore not accessible to the antibodies, but may accommodate the CD4 which has a single immunoglobulin domain, unlike antibodies that use two domains for molecular recognition.
B 13837.3 EE

The inhibition of the interaction of the gp120 viral protein with the principal receptor of the entrance, the CD4, by a soluble recombinant CD4 and/or by immunoglobulin chimera containing domains of the CD4, has been one of the first therapeutic approaches suggested and tested for the inhibition of the infection by HIV-1.
These molecules are effective, in vitro, in the inhibition of the viral infection only in the case of strains adapted to the laboratory, but are ineffective in the case of numerous primary isolates. The rapid degradation, in vivo, of these constructions derived from the CD4 and their lower affinity for the viral envelope of the primary isolates have been proposed as principal causes for their ineffectiveness. However, studies have shown that the affinity, for the CD4, of the monomeric gp120 recombinants of certain primary isolates is not significantly different to that of gp120 from laboratory strains. The oligomeric nature of the gp120 viral protein at the surface of the virus, the density and the structure of the CD4 at the surface of the permissive cells and perhaps the presence of other molecular interactions that have not yet been clearly elucidated or other entry accessory molecules could play an important role in the entry mechanism and represent other reasons for the ineffectiveness of the soluble CD4 proteins as entry antagonists. The development of antiviral agents that target the gp120 viral protein presents a major challenge, due to the high genetic variability of the protein of the envelope, a factor which may explain the B 13837.3 EE

ineffectiveness of many inhibitors of the gp120 viral protein. Nevertheless, the fact that the residues that contribute to forming the heart of the bonding site of the gp120 viral protein for the CD4 are formed by very conserved residues suggests that the inhibitors of the gp120 viral protein may be effective against a wide spectrum of HIV-1 isolates.
Peptidic structures derived from the CD4 have been proposed as inhibitors of the viral infection: this is a structural mimic peptide of the CDR2 loop of the CD4 which incorporates a phenyl residue mimicking the Phe43 [14] and a cyclic peptide corresponding to the CDR3 loop of the CD4 and incorporating additional aromatic residues [15]. However, these constructions have a weak antiviral IS activity, limited to a laboratory isolate and, even if they have been designed from the outset as structural mimics of the CD4, they are not active in the entry step [16] .
At present, a certain number of inhibitors of the gp120 - CD4 interaction have been proposed and have a clinical application. These are, for example, a large recombinant protein, formed by the genetic fusion of the DID2 domains of the CD4 with the constant domains of an immunoglobulin IgG2 [17]. This PR0542 construction is in I/II clinical phase. The large recombinant protein seems to be well tolerated by the human organism, but its effectiveness depends on a high circulating dose which is difficult to attain. Smaller molecules have been proposed as inhibitors of the CD4 - gp120 interaction and are in B 13837.3 EE

the clinical testing phase. These are: a bis-azo colorant FP21399 [18], identified by a combinatory approach, the phosphorothioate oligonucleotide zintevir (AR177) [19], a polymer based on naphthalene sulphonate PR02000 [20] and a starch derivative dextrin 2-sulphate (D2S} [21], which inhibit the HIV-1 chemical isolates in cellular cultures, but require high concentrations. Among the entry inhibitors, there is also SPC3 [22], a multi-branched synthetic peptidic construction which, originally proposed as an inhibitor of the CD4 - gp120 interaction, then turned out to be active in a step following the attachment of the virions to the cells. Other small molecules have been proposed as entry inhibitors: the bicyclam AMD3100 [23], the NSC651016 [24), the peptide T22 [25] and its more recent version T134 or T140 [26]
and the TAK779 [27). These molecules are viral entry inhibitors, active on the co-receptors of the CXCR4 and CCR5 entry, but have no activity in the CD4 - gp120 interaction. Peptides derived from the sequence of the C-terminal sequence of the glycoprotein gp4l, the peptide T20 (or DP178} [28) and its more recent improved version DT1249 [29], are other inhibitors of the infections that target a transitory phase of the entry, after the attachment of the viral envelope to the CD4 and before the fusion of the virus and cell membranes: said peptides are inactive in the gp120 - CD4 interaction. Most of these products are presently in the I/II phase of clinical testing, but their therapeutic effectiveness still needs to be verified.
B 13837.3 EE

The anti-AIDS therapeutic regime presently used in clinics, tritherapy, involves a combination of three inhibitors that target two viral enzymes, inverse transcriptase and protease. Although tritherapy has 5 succeeded in significantly reducing the viral burden of many patients, it is not capable, even under the most favourable conditions, to eradicate the disease, because dormant infection reservoirs always remain [30]. The partial success of tritherapy on the one hand and the 10 impossibility of stopping the AIDS infection with present therapeutic protocols on the other hand, have increased the demand for new medicines that could be active at the level of other viral targets and increase the repertory and effectiveness of present clinical medicines. .
The banks of molecules presently available and potentially active in the inhibition of the entry and the infection by the AIDS virus are enormous. However, no tool for rapidly and efficiently searching for molecules that are indeed active in inhibiting the gp120 - CD4 interaction necessary for an effective infection by the AIDS virus is available. The present screening methods are slow, not very accurate and often require significant quantities of reagents. Moreover, the costs incurred in these screenings are high.
Description of the invention The precise aim of the present invention is to provide a method for screening molecules capable of attaching themselves with a defined affinity to the gp120 B 13837.3 EE

protein of the AIDS virus, in particular the human immunodeficiency virus (HIV).
Said screening method of the present invention includes at least one screening cycle comprising the following steps:
a) covalent labelling with a fluorescent probe of a peptide with said defined affinity for the gp120 protein, said peptide comprising the sequence (A) defined below:
b) bringing the labelled peptide into the presence of the gp120 protein, at a concentration of gp120 which allows between 20 and 500 of attachment of said labelled peptide, in such a way as to form a reversible complex [labelled peptide - gp120], c) measuring the standard fluorescence polarisation of the complex (labelled peptide - gp120), d) screening testing by bringing into competition the complex (labelled peptide - gp120) formed with at least one candidate molecule, capable of attaching itself to the gp120 protein, at a defined concentration of said molecules) and under physiochemical conditions that enable the competition between said molecule and the labelled peptide to form, if appropriate, a complex with the gp120 protein, e) measuring the fluorescence polarisation emitted by the screening test, and f) comparing the fluorescence polarisation measurement of step e) with the standard measurement of step c) , B 13837.3 EE

the candidate molecules) being considered as a molecule capable of attaching itself to the gp120 protein when the fluorescence polarisation measurement of step e) is less than the standard measurement of step c).
The molecules screened by means of the screening method of the present invention are therefore those that enter into competition with the peptide of the present invention.
Indeed, the method of the present invention is, in l0 particular, characterised in that it uses a specific family of peptides that mimic the CD4 and make up for the above-mentioned disadvantages of the prior art. Indeed, it uses a stable peptide, with a very high affinity for the envelope of the AIDS virus, in particular for the gp120 protein.
The peptide of the present invention is characterised in that is comprises the following sequence (A) TPA - P1 - Cys - PZ - Cys - P3 -Cys - Ala or Gln Gly or (D)Asp or Ser - Ser or His or Asn - Xaa~ - Cys Thr or Ala - Cys - Xaak - NH2, in which TPA represent thiopropionic acid, Xaa~ represents ~i-naphtylalanine, phenylalanine or bi-phenylalanine, Xaak represents Gly, Val or Ileu, P1 represents 3 to 6 amino acids, Pz represents 2 to 4 amino acids and P3 represents 6 to 10 amino acids, the amino acids in P1, PZ and P3 being natural or not natural, identical or different, and P1, PZ and P3 having or not a shared sequence, said peptide having (3 hairpin conformation in which the (3 B 13837.3 EE

bend is formed by the amino acid residues Ala or Gln -Gly or Dasp or Ser - Ser or His or Asn - Xaa~ of the sequence (A).
According to one embodiment of the present invention, the peptide of the present invention is characterised in that it comprises the following sequence (I) TPA - Xaaa - Xaab - Ala or Gln or His - Arg or Phe -Cys - Xaa° - Xaad - Arg - Cys - Lys - Xaae - Xaaf Xaag - Xaah - Leu or Lys - Xaal - Lys - Cys - Ala or Gln - Cly or (D)Asp or Ser - Ser or His or Asn -Xaa~ - Cys - Thr or Ala - Cys - Xaak - NH2, in which TPA representx thiopropionic acid, Xaaa, Xaab, Xaa°, Xaad, Xaae, Xaaf, Xaag, Xaah and Xaal are amino acids, 20 natural or non-natural, identical or different, Xaa~
represents (3-naphtylalanine or phenylalanine or bi-phenylalanine and Xaak represents Gly or Val or Ile.
The D(Asp) residue represents the D optical isomer of aspartic acid, Nal represents ~3-naphtylalanine, and 25 Bip represents bi-phenylalanine.
The inventors have shown that this peptide, defined by the sequence (A), in particular the sequence (I), comprising a thiopropionic acid in position 1 of the sequence, a Phe, Nal or Bip residue in position 23 of the B 13837.3 EE

sequence (Xaa~) and a Gly, Val or Ile residue in position 27 of the sequence (Xaak) , has a high affinity and a high bonding specificity for the gp120 protein of the viral envelope of the immunodeficiency virus.
Said peptide comprises only 27 or 28 residues, and has a hairpin structure comprising two antiparallel (3 strands linked by a bend ~i of four residues. The two antiparallel strands, being at a distance that could give rise to interactions at intramolecular hydrogen bridges, stabilise the structure. In particular, the distance between the amide nitrogen atoms and the carbonylic oxygen atom of the peptide bond is 2.5 to 3.6 A°. Said well defined and stable structure reproduces the structural elements of the CD4 that are critical for its bonding to the gp120 glycoprotein.
The three-dimensional structure of said peptide has been resolved experimentally by nuclear magnetic resonance (NMR). This analysis shows that its three-dimensional organisation reproduces the peptidic backbone of the hairpin structure of the CD4, and the region 37-46 of said peptide may be superimposed on the region 36-37 of the CD4 with a rms deviation of only 1.05 A°.
Moreover, the lateral chains of the Ala or Gln 20, Ser 22, Xaa~ 23 (Nal, Phe or Bip), and Thr or Ala 25 residues have an orientation comparable to that of the corresponding residues of the CD4, namely Gln 40, Ser 42, Phe 43 and Thr 45. In particular, the lateral chain of Xaa~ 23 rises from the hairpin pattern in a conformation similar to that of Phe of the CD4 to insert itself at the B 13837.3 EE

entry of a hydrophobic cavity of the gp120 viral protein, thus strengthening the bond with gp120. Arg and Lys at the positions 9 and 18 are topologically equivalent to the ARg59 and Lys35 residues of the protein CD4.
5 TPA and the 5 Cys of the peptide form disulphide bridges that have a type 1-3, 2-4 and 3-6 pairing and stabilise the hairpin structure.
The hairpin structural unit has a surface accessible to the solvent and/or a larger molecule such as a 10 protein, which may favour its interaction with the protein of the viral envelope.
The lateral chains of the amino acids of the surface accessible to the solvent of this structural unit are close in spatial terms and form a continuous molecular 15 structure which reproduces the structure of several residues that are important for the CD4 - gp120 bond.
The structural platform containing the hairpin structural unit also contains other structural regions capable of receiving additional chemical functionalities 2o in a well defined spatial position with respect to the hairpin structural unit which can thus reinforce its biological bonding function and/or have labelling groups.
For example, a biotin group or a fluorescein group has been incorporated at the level of the lateral chain of the lysine 11, without leading to the loss of the biological activity of the derivative. The incorporation of said groups has made it possible to use these derivatives in interaction tests with the protein of the envelope, as described above.
B 13837.3 EE

According to the invention, the sequence (A), for example the sequence (I), may further comprise a proline residue linked to the Xaak residue. A proline residue added to position 28 stabilises the C-terminal end and makes the C-terminal ~i strand of the hairpin structural unit less flexible.
According to the invention, in sequence (I), Xaal may be Gly.
For example, the peptide of the present invention l0 may comprise a sequence chosen from among the sequences ID n° 4, ID n° 5, ID n° 6, ID n° 7, ID
n° 7, ID n° 9, ID
n° 10, ID n° 11, ID n° 12, ID n° 13, ID n°
14, ID n° 15, ID n° 16, ID n° 17, ID n° 18, ID n° 19 and ID
n° 20 of the appended sequence list.
The peptide of the present invention may be obtained by solid phase chemical synthesis or by genetic recombination. The chemical synthesis may be carried out for example with an Applied Biosystems mod.433A
(registered trade mark) type automatic peptide 2o synthesiser. It may be carried out for example in Fmoc chemistry which uses the fluoromethyloxycarbonyl group for the temporary protection of the a-amine function of the amino acid.
The peptide of the invention may also be produced by means of a method comprising the following steps:
a) integrating a nucleic acid sequence in an expression vector, said nucleic acid sequence coding for the peptide of the present invention, B 13837.3 EE

b) introducing the expression vector comprising said nucleic acid sequence into a host cell, c) culturing the host cell comprising said nucleic acid under culture conditions that allow the synthesis of said peptide, d) recovering said synthesised peptide, and e) grafting the TPA in N-terminal position.
The technical aspects involved in carrying out this l0 peptide synthesis method are known to those skilled in the art. For example, they are described in the work by Sombrook, Fritsch and Maniatis, MOLECULAR CLONING, A
LABORATORY MANUAL, 2nd edition.
The grafting of a TPA in C-terminal position of the peptide may be achieved by means of a conventional organic method applicable to a peptide.
The inventors have synthesised a family of peptides reproducing a part of the structure of the region of the CD4 resembling the CDR2 region of the immunoglobulins which, as mutagenesis results and crystallographic analysis have shown, are among the critical regions of the CD4 bond to the gp120 viral protein. The peptides of the present invention are capable of attaching themselves to the region of the glycoprotein gp120 which interacts with the CD4, the principal receptor for the entry of the AIDS virus, particularly the HIV-1 virus, into the CD4 target cells.
These peptides constitute a family of inhibitors in which the affinity for the monomeric recombinant gp120 B 13837.3 EE

viral glycoprotein varies between ICso values from 0.1 nM
to 400 nM. They constitute specific and powerful inhibitors of the CD4 - gp120 interaction based on the structure of the CD4. The bond with the recombinant gp120 viral protein forms in competition with the soluble CD4 and is specific to the well structured shape.
Moreover, they may be labelled, for example with a fluorescent probe, without the bonding affinity to the protein of the gp120 viral envelope being altered.
Consequently, said peptides may be used as tracers in the screening method using a fluorescence anisotropy technique according to the present invention.
This method uses the fluorescence polarisation of said peptide labelled in a covalent manner by a fluorescent probe.
The peptide used is that which has an affinity in the range of that of the molecules screened. For example, if one aims to screen molecules with an affinity greater than or equal to 106 M-1, one uses a labelled peptide with an affinity of 106 M-1. Thus, for example, if one aims to screen molecules with a higher affinity, for example 109 M~1, one uses a peptide with a similar affinity.
Consequently, the screening method of the present invention may be fine tuned, as a function of the sought after affinity, by choosing the peptide used.
Moreover, the affinity for the gp120 viral protein of the molecules detected by the screening method of the present invention may be determined from that of the B 13837.3 EE

peptide according to the invention chosen for the screening.
The methods and devices that may be used for implementing the method of the present invention are those that are accessible to those skilled in the art in the literature concerning the study of molecular interactions, for example in documents [37] , [38] , [39]
and [40] and in the patent US 5 756 292. The analyses and simulations may be carried out by using the "BIOEQS"
programme described in documents [41], [42] and [43] or other interaction analysis programmes.
The fluorescent probe should preferably have a high quantum ef f iciency and a f luorescence in the visible for a high detection sensitivity. We have tested rhodamine green, fluorescein and Alexa488 (trademark). Obviously, other fluorescent probes with the above specific properties may be used.
The fluorescence polarisation is calculated from two measurements, the emission intensity of the parallel and perpendicular polarised fluorescence, with an excitation in parallel. Here, the parallel polarisation is defined along the direction z of the laboratory axis. The polarisation is calculated as follows:
P = (III - II) I (III + Iz) (1) One may also express the result in fluorescence anisotropy result, calculated as follows:
B 13837.3 EE

A = (III - Iz) / (III + 2I=) (2) The relation between polarisation and anisotropy being as follows:
A = 2/3 (1/P - 1/3)-1 (3) .
5 The fluorescence anisotropy or polarisation value is linked to the rate of rotational diffusion of the molecule in solution through the following relationship (4) Ao/A-1 = t/t~ (4), 10 Where Ao is the limit anisotropy of the fluorophor, a physical constant determined by the angle between the absorption and emission dipoles, t is the fluorescence lifetime, and t~ is the rotational correlation time. This depends on the hydrated volume (Vh) of the molecule or the 15 molecular complex according to the relationship (5):
t~ - ~lVh/R'r ( 5 ) .
where r~ is the viscosity of the solution, R is the gas constant and T is the temperature in Kelvin.
Consequently, when the labelled peptide is alone in 20 solution, its rotational diffusion rate is rapid and therefore the fluorescence anisotropy of the fluorophor linked in a covalent manner remains relatively low. On the other hand, when the labelled peptide interacts with the gp120 the fluorescence anisotropy of the probe increases considerably since the gp120 viral protein has a molecular weight of 120000 daltons, and thus the size of the peptide of the present invention - gp120 complex is significantly higher than that of the peptide alone.
B 13837.3 EE

Consequently, the rate of rotational diffusion drops, and the anisotropy increases.
In the screening method of the present invention, any molecule capable of inhibiting the interaction between a peptide of the present invention and a gp120 viral protein may represent an inhibitor of the CD4 -gp120 interaction, and thus a viral infection inhibitor.
It is therefore possible, thanks to the peptides of the present invention, to screen in a bank of molecules any molecule capable of interacting with the gp120 protein or a protein analogous to the 9p120, in particular molecules with antiviral activity or molecules useful f or the manufacture of medicines.
The screening method of the present invention IS functions by competition. The peptide of the present invention, chosen and labelled, is brought into the presence of gp120 at a concentration that allows between and 50 % of attachment. The quantity of gp120 must not be saturating, if not, a large amount of molecules to be 20 screened must be used to allow the competition. Moreover, the complex (labelled peptide - gp120) formed must be reversible in order to allow competition.
The concentration conditions of candidate molecules in the screening tests of the method of the present invention are determined in the same manner as in conventional competition tests for molecules for a receptor used in molecular biology. The concentration of the candidate molecules) is, preferably, 10 to 100 times B 13837.3 EE

higher than the concentration of the labelled peptide of the present invention.
The method of the present invention is a method that is both rapid, particularly due'to the fact that it uses fluorescence anisotropy techniques, and precise, because it uses the peptides of the present invention. Moreover, it requires very small quantities of reagents as shown by the examples below, which reduces the costs incurred in screening molecules capable of being used for the to manufacture of medicines intended for the treatment of AIDS compared to the methods of the prior art.
Moreover, the screening method of the present invention is completely suited to the screening of the banks of molecules presently available and potentially active in the inhibition of the entry and the infection of the AIDS virus. Consequently, it constitutes a rapid and efficient screening tool for molecules that are indeed active in inhibiting the gp120 - CD4 interaction required for an efficient infection by the AIDS virus.
For example, according to a first embodiment of the present invention, a single candidate molecule is tested in a screening cycle. In this case, if the fluorescence anisotropy measurement of step e) is less than the standard measurement. of step c), then the candidate molecule tested is considered as a molecule capable of attaching itself to the gp120 viral protein. If the fluorescence anisotropy measurement of step e) is unchanged with respect to the standard measurement of step c), then the candidate molecule tested is not B 13837.3 EE

considered as being capable of attaching itself to the gp120 viral protein. Thus, according to this embodiment, the cycles must follow on from each other as many times as there are different candidate molecules to test, at a rate of one molecule per cycle.
For example, according to a second embodiment of the present invention, a sample comprising several different candidate molecules may be tested in a single screening cycle. In this case, if the fluorescence anisotropy l0 measurement of step e) is less than the standard measurement of step c), then the sample is considered as comprising at least one molecule capable of attaching itself to the gp120 viral protein. If the fluorescence anisotropy measurement of step e) is unchanged with IS respect to the standard measurement of step c}, then the sample is considered as not comprising at least one molecule capable of attaching itself to the gp120 viral protein.
Consequently, according to this second embodiment of 20 the present invention, a single cycle allows a sample comprising several different candidate molecules to be tested. If, for one of the samples, the fluorescence anisotropy measurement of step e) is less than the standard measurement of step c), then at least one of the 25 candidate molecules of said sample is considered as capable of attaching itself to the gp120 viral protein.
The screening may then be f fined down to the molecules of said sample according to the first embodiment of the present invention in order to identify the molecules) of B 13837.3 EE

the sample capable of attaching itself (themselves) to the gp120 viral protein. This second embodiment of the present invention allows very rapid screening of candidate molecules.
These first and second embodiments of the present invention may be carried out on known candidate molecules presently available and potentially active in the inhibition of the entry and the infection by the AIDS
virus.
According to a third embodiment, the method of the present invention may be applied to non-purified biological liquids such as urine, cell extracts, bacterial extracts, etc. When the method of the invention enables such an extract to be considered as comprising at least one molecule capable of attaching itself to the gp120 viral protein, said extract may be analysed by conventional chemical analysis methods in order to identify the molecules that it contains. A more precise screening, according to the first embodiment of the present invention, can then be carried out in order to reveal, among all of the identified molecules in the extract, the molecules) capable of attaching itself (themselves) to the gp120 viral protein.
Consequently, the method of the present invention constitutes a screening tool that is powerful, rapid, reproducible, sensitive and which can be fine tuned to a wide range of affinities. It is carried out in solution and does not require the complexes of free molecules to be separated.
B 13837.3 EE

The screening method of the present invention has numerous advantages that those skilled in the art may easily identify.
Among these advantages, one may cite the capability 5 of the method to be able to select molecules with a pre defined affinity. Indeed, if the conditions of the competition reaction make it possible to maintain constant the Kd between the labelled peptide of the present invention and the gp120 viral protein, the 10 affinity of the molecule selected after the competition step will be at least equal to that of the labelled peptide.
Another advantage is that the labelling of the peptide of the present invention may be carried out with 15 different compounds such as rhodamine green, fluorescein, etc. Moreover, the use of compounds such as fluorophors, in which the absorption and emission properties in the visible may be particularly advantageous for the detection of molecules in non-purified extracts such as 20 urine, cell extracts, bacterial extracts, etc.
Furthermore, this process makes it possible to use small amounts of gp120 viral protein, since it can be carried out in the presence of a non-saturating concentration of gp120, or even preferably at a 25 concentration that allows 20 to 50 % attachment.
Unlike ELISA tests widely carried out in laboratories in order to highlight competition between given molecules, the method of the present invention uses peptides and the anisotropy measurements defined above B 13837.3 EE

make it possible to carry out measurements entirely in solution. Moreover, the method of the present invention may be adapted to very small volumes, of several microlitres, and thus to microwell measurements. It also makes it possible to carry out very rapid measurements, in one or two minutes for the competitions.
Another advantage of the present invention resides in the reproducibility of the results. Indeed, the anisotropy value for a labelled peptide of the present l0 invention, for example under the conditions defined in the examples given below, is always the same.
A further advantage of the present invention resides in the fact that the temperature and solution conditions may easily be modified and controlled.
A yet further advantage resides in the fact that, since the signal/noise ratio is very high, the method of the present invention enables very low attachment levels to be detected.
Yet other advantages will become clearer to those skilled in the art on reading the following examples and by referring to the list of appended sequences and figures.
Brief description of the list of sequences The appended list of sequences provides the following peptide sequences:
- sequence ID n° 1: sequence sCD4.
- sequence ID n° 2: sequence of scyllatoxin (ScTx).
B 13837.3 EE

27 .
- sequence ID n° 3: sequence CD4M9: peptide derived from scyllatoxin, with no TPA in the 1 position of the sequence.
sequence ID n° 4 -16: sequences designated CD4M9T, CD4M26, CD4M27A, CD4M27B, CD4M27C, CD4M31, CD4M32, CD4M33, CD4M35, K15, K16 and CD4M9BIP and CD4M9V
respectively of the present invention.
- sequence n° 17: sequence designated CD4M3 derived from scyllatoxin with no TPA in position 1 of the sequence.
- sequence n° 18: sequence designated CD4M0 derived from charbydotoxin.
Brief description of the drawings - Figure 1: Inhibition curves for the bond of the CD4 to the gp120 viral protein, by peptides of the present invention, obtained by ELISA in competition. The tested peptides were: CD4M9, CD4M9T, CD4M27, CD4M32, CD4M33, and CD4M35. The experimental data is expressed in 2o percentage bonding (% F) as a function of the concentration of peptides.
- Figure 2: Inhibition curves for the bond of the CD4 to the gp120 viral protein, by the peptide labelled with CD4M33-fluroescein, CD4M33-F, obtained by ELISA in competition. The curves for the peptides CD4M9 (sequence ID n° 3), CD4M33 (sequence ID n° 13) and CD4M35 (sequence ID n° 14) are also represented for comparison purposes.
The experimental data is expressed in percentage bonding (a F) as a function of the concentration of peptides.
B 13837.3 EE

- Figure 3: Interaction curves for the recombinant viral protein gp120 - HXB2 with the peptide CD4M33 (sequence ID n° 13) attached to the surface of a bio-chip of a surface plasmon resonance instrument. The association (0-300 s) and dissociation (300-800 s) curves have been recorded after injecting the gp120 viral protein at concentrations 13.2 (1), 19.8 (2), 29.6 (3), 44.4 (4), 66.6 (5) and 100 (6) nM. The data is expressed in resonance units (RU) as a function of time (t) in l0 seconds.
- Figures 4(a) and (b): interaction curve of the HIV-I, inactivated by AT-2 [34], with the antibody 48d [11] (a) and CG10 ~35] (b), attached to the surface of a bio-chip of a surface plasmon resonance instrument in the presence of the peptide CD4M33 (10 ~.zg/ml) , of an inactive peptide (A1a23) CD4M9 (Phe replaced by Ala in the peptide sequence of CD4M9) (10 ug/ml) and in the absence of peptide. The association (0-180s) and dissociation (180-600 s) curves have been recorded after injecting the HIV-1, which was pre-incubated with the peptides at 20 °C for 1 h. The data is expressed in resonance units (RU) as a function of time (t) in seconds.
Figures 5 (A) to (C): effects of peptides CD4M30 (sequence ID n° 10) (Figure 4A) , and CD4M33 (sequence ID
n° 13) (Figure 4B), CD4M35 (sequence ID n° 14) (Figure 4C) on the replication of the HIV-lr,p,I strain in the CMSP
activated by the PHA-P. The data is expressed in percentage inhibition (o I) of the Inverse Transcriptase B 13837.3 EE

(IT) activity in the culture supernatant as a function of the concentration in nM of peptides.
- Figures 6 (A) and (B): effect of the presence of soluble CD4 and the peptides of the present invention on the interaction of the recombinant gp120Hxsz viral protein with the CG10 (A) and 48d (B) antibodies, attached to the surface of a bio-chip of a surface plasmon resonance instrument. The association (0 -180s) and dissociation (180 - 400, 180 - 450 s) curves are shown for the gp120 viral protein alone and the gp120 viral protein in the presence of the CD4M25 (1.5 equivalent, sequence ID n°
6), of the CD4M9 (1.5 equivalent), of the inactive mutant (A1a23) CD4M9 (1.5 equivalent) and the soluble CD4 (1.5 equivalent). The data is expressed in resonance units (RU) as a function of time (t) in seconds.
- Figure 7: examples of peptide sequences of the present inventions represented with the one letter codes of amino acid residues. TPA and B represent thiopropionic acid and bi-phenylalanine respectively and "d" the D
optical form of aspartic acid.
- Figures 8 a), b) and c): curves illustrating fluorescence anisotropy measurements (An), expressed in millianisotropy (mA), of the peptide K16 of the ,present invention labelled with rhodamine green, in solution at a concentration of 2 nM, and brought into the presence of the HXB2 strain of the gp120 viral protein at different nanomolar (nM) concentrations [GP120] .
- Figures 9 and 10: curves illustrating fluorescence anisotropy measurements (An), expressed in B 13837.3 EE

millianisotropy (mA), respectively of peptides CD4M33 and K15 of the present invention labelled with rhodamine green, in solution at a concentration of 2 nM, and brought into the presence of the HXB2 strain of the gp120 5 viral protein at different nanomolar (nM) concentrations [GP120] .
- Figure 11: curves illustrating fluorescence anisotropy measurements (An), expressed in millianisotropy (mA), of the peptide CD4M9 of the present 10 invention labelled with fluoroscein, in solution at a concentration of 6 nM, and brought into the presence of the HXB2 strain of the gp120 viral protein at different nanomolar (nM) concentrations [GP120] .
- Figures 12a) and 12b): curves illustrating 15 fluorescence anisotropy measurements (An), expressed in millianisotropy (mA), of the peptide CD4M33 of the present invention labelled with rhodamine green, in solution at a concentration of 6 nM, and brought into the presence of the HXB2 and SF2 strains respectively of the 20 gp120 viral prctein at different nanomolar (nM) concentrations [GP120] and [SF2] .
- Figure 13: theoretical curves illustrating theoretical fluorescence anisotropy values (An), expressed in millianisotropy (mA), of a competition 25 between a labelled peptide CD4M33 and a non-labelled peptide CD4M33, as a function of the nanomolar (nM) concentration of the non-labelled peptide CD4M33, at different nanomolar (nM) concentrations of a gp120 (HXB2) B 13837.3 EE

viral protein. These curves have been calculated from the Kd=l4nM of the CD4M33 determined by direct assaying.
- Figures 14 a) to i): curves illustrating fluorescence anisotropy measurements (An), expressed in millianisotropy (mA), of a competition between a fluorescein labelled peptide CD4M33 and different non labelled peptides (respectively CD4M33, CD4M35, CD4M32, sCD4, CD4M27, CD4M9, CD4M9TPA and CD4M9BIP), as a function of the nanomolar (nM) concentration of the non labelled peptide.
Examples EXAMPLE l: SYNTHESIS OF THE PEPTIDES OF THE PRESENT
INVENTION
The peptides of the present invention have been produced in this example by solid phase chemical synthesis with an Applied Biosystems mod.433A automatic peptide synthesiser, and using Fmoc chemistry, which employs the fluroenylmethyloxycarbonyl (Fmoc) group for the temporary protection of the a-amine function of amino acids. The protector groups used to prevent secondary reactions of the lateral chains of amino acids, in this Fmoc strategy, were t-butyl ether (tBu) for the Ser, Thr and Tyr residues; t-butyl ester (OtBu) for Asp, Glu;
trityle (Trt) for GIn, Asn, Cys, His; t-butyloxycarbonyl (Boc) for Lys and 2,2,5,7,8 - pentamethylchromane - 6 -sulfonyl (Pmc) for Arg.
The coupling reaction was carried out with an excess of 10 equivalents of amino acids (Immol) with respect to B 13837.3 EE

the resin (0.1 mmol). The protected amino acid was dissolved in 1 ml of N-methylpyrollidone (NMP) and 1 ml of a 1M solution of 1- N- hydroxy- 7- azabenzotriazole (HOAt) in the solvent NMP. 1m1 of a 1M solution of N,N' -dicyclohexylcarbodiimide (DCC) was then added. After 40 to 50 minutes of activation, the active ester formed was transferred to the reactor which contains the resin.
Before this step of transfer and then coupling, the resin was deprotected of its Fmoc group by a solution of 20 °s l0 piperidine in NMP. The excess piperidine was removed by washing with NMP after around 5 to 10 minutes.
During the deprotection, the detection of dibenzofulvene - piperidine adducts at 305 nm made it possible to monitor the correct course of the synthesis.
Indeed, the quantification of the adduct made it possible to estimate the efficiency of the deprotection of the Fmoc group and following the coupling of the last amino acid incorporated.
CHEMICAL MODIFICATIONS MADE TO THE PEPTIDES OF THE
PRESENT INVENTION
A fluorescent probe and a biotin group were coupled on two of the studied peptides of the present invention, the CD4M9T (sequence ID n° 4) and the CD4M33 (sequence ID
n° 13). The incorporation on these two compounds was carried out on the lysine 11 residue, at the level of the face opposite the bonding site of the gp120 viral protein. This choice of lysine was determined by its possibilities of protection of its lateral chain with a B 13837.3 EE

viral protein. These curves protective group, called orthogonal, such as the group of (1- (4,4- dimethyl- 2,6- dioxo- cyclohexylidene) ethyl) (or Dde group). Said group is in fact stable to the treatment at 20 % of piperidine used for the deprotection of the Fmoc group, but was separated in a specific manner by treatment with a 2 % solution of hydrazine. Once the Dde group had been removed, it was possible to couple the fluorescein and biotin.
INTRODUCTION OF FLUORESCEIN
The fluorescent group was introduced on the lateral chain of the lysine 11 of the peptide of the present invention CD4M33 and the lysine 15 or 16 of peptides K15 (sequence ID n° 15) or K16 (sequence ID n° 16), respectively, of the present invention. The Dde group was therefore used for the protection of the lateral chains of the lysine during the synthesis of the peptide and then liberated by two treatments of 5 minutes with 2 % of hydrazine in dimethylformamide (DMF). The fluorescein molecule was coupled directly to the synthesised peptide.
The coupling of the probe was carried out by means of the fluorescein- (5- (6)- carboxylate ester of N-hydroxysuccinimidyl. To achieve this, to one equivalent of resin were added 4 equivalents of activated ester of fluorescein in the presence of 14 equivalents of diisopropylethylamine (DIEA). The reaction was left to take place in NMP solvent for one night. The final deprotection was finally realised.
B 13837.3 EE

INTRODUCTION OF BIOTIN
An 8- amino- 3,6- dioxaoctanoic spacer was firstly added to the lysine 11, previously deprotected of the Dde group. The reaction is similar to that described in the previous paragraph. Biotin was then incorporated in the form of a biotinamidocaproate ester of N-hdyrosuccinimidyl: to one equivalent of resin were thus added 4 equivalents of activated ester of biotin in the presence of 14 equivalents of diisopropylethylamine l0 (DIEA). The reaction was left to take place at ambient temperature for one night. The resin was then washed before being treated by the final deprotection solution.
INTRODUCTION OF A THIOL GROUP
The peptide CD4M9 was synthesised according to the procedure described above with the lysine in position 11 having a Dde group as the protector group of the lateral chain. After synthesis of the peptide, the peptide -resin was treated 5 times with a 2 o solution of hydrazine in DMF. The coupling of a link arm was carried out for one hour at ambient temperature in DMF with 10 equivalents of Fmoc- 8- amino- 3,6- dioxaoctanoic acid using the HBTU reagent in the presence of diisopropylethylamine. The Fmoc group was then deprotected with 20 % piperidine in DMF. The peptide -resin was then treated with 10 equivalents of Traut reagent (2- iminothiolane hydrochloride from Sigma) in the presence of DIEA. The peptide was finally liberated and deprotected as described above.
B 13837.3 EE

FINAL DEPROTECTION OF THE RESIN
The separation of the resin and the protector groups present on the lateral chains was carried out 5 simultaneously by treating the peptide linked to the resin by trifluoroacetic acid (TFA). Before carrying out the separation, the resin was washed several times with dichloromethane (DCM) and finally dried. The reagent used during the separation was an acid mixture containing 81.5 l0 % of TGA and phenol traps ( 5 °s ) , thioanisol ( 5 % ) , water (5 0), ethanedithiol (2.5 0) and tri-isopropylsilane (1 %). The resin was treated with this mixture for three hours under agitation and at ambient temperature, at a level of 100 ml of solution per gram of resin. The free 15 peptide in solution was recovered by filtration. The peptide was then precipitated and cold washed in diisopropyl ether and then dissolved in 20 % acetic acid and lyophilised.

FORMATION OF DISULPHIDE BRIDGES
The peptide recovered after lyophilisation, the synthetic crude, was in the reduced form, in other words the intrachain disulphide bridges were not formed. The 25 formation of these covalent bonds was carried out by using the cystamine / cysteamine redox couple. The synthetic crude was taken up in water with 0.1 % TFA
added (v/v) and 6M guanidium chloride to facilitate its dissolution, at a level of 2.0 mg.ml-1. This solution was B 13837.3 EE

then added, drop by drop, diluted to 0.2 mg.ml-l, to the reducing buffer, composed of 100 mM Tris/HCl, pH 7.8, and mM cysteamine. The final 0.5 mM cystamine (oxidant) was added after 45 minutes reaction time at ambient 5 temperature. The medium was brought to pH 3.0 after 30 minutes.
The cysteamine make it possible to reduce the thiol groups present on the peptide. In free air, it oxidises and allows the oxidation of cysteines and thus the folding of the peptide through formation of intrachain disulphide bridges. The cystamine added at the end of the treatment makes it possible to perfect the folding. The proper course of the oxidation was verified by analytical chromatography by comparing the retention times of the crude and oxidised products, higher for the former.
The oxidation of the peptide CD4M9, with a thiol group in position 11, differs somewhat from that described in the above paragraph. The folding reaction medium, composed of a 20 mM phosphate, 200 mM NaCl buffer adjusted to pH 7.8 was degassed for a long period by argon and then 5 mM of cysteamine and S mM of cytamine added. The peptide comprising seven free SH functions was then added and the oxidation reaction was stopped by adding acid after only 10 minutes, sufficient time for the folding and the formation of three natural disulphide bridges and sufficiently short to limit the formation of secondary product corresponding to higher oxidation states.
B 13837.3 EE

PURIFICATION OF THE PEPTIDES OF THE PRESENT INVENTTON
The peptides of the present invention were purified by high performance liquid chromatography in inverse phase on a Vydac C18 (trademark) preparatory column (1.0 x 25.0 cm). A 0-60 °s linear gradient of acetonitrile in an aqueous solution of 0.1 o trifluoroacetic acid was used over 90 minutes. The fractions of the major peak were analysed by analytic HPLC; the fractions with only a single peak were collected and lyophilised. The resulting l0 product was analysed by mass spectrometry.
EXAMPLE 2: COMPETITION TESTS BY INDIRECT ELISA
The inhibition of the rgp120 - CD4 interaction was measured by indirect ELISA (Enzyme Linked ImmunoSorbent Assay). 50 ng per well of anti-gp120 antibody D7324 were immobilised on a plate with 96 wells (Maxisorb, Nunc) for one night at 4 °C. After passivation by Bovine Serumalbumin and three washings with the washing buffer (Tris 10 mM, pH 7.8, Tween 20 0.05%) , 15 ng of rgp120~B2 per well were added. Different concentrations of competitors were then added, after three washings, as well as 0.4 ng of soluble CD4 per well. 100 o controls only containing soluble CD4 were carried out. After one night at 4 °C and three washings, 5 ng of mouse antibody anti-CD4 L120.3 were added to each well, then a goat antibody anti-IgG combined with peroxydase (GAMPO). The development was achieved by adding 3,3',5,5'-tetramethylbenzidine, a fluorescent substrate of the peroxydase. The reaction was stopped after 30 minutes by B 13837.3 EE

adding 2M sulphuric acid. The inhibition of the rgp120 -CD4 interaction was calculated by reading the absorbance A at 450 nm. Controls without competitor enabled the absorbance Aloo% (absorbance in the absence of competitor) to be determined. The percentage inhibition for each concentration of peptide of the present invention was calculated from the formula:
Percentage inhibition = 100 x (Aloo% - A) / Aloo~.
Duplicate tests were carried out and the results expressed as the average of the experimental doubles.
The biological activities of the peptide CD4M0 (sequence ID n° 18) derived from the charybdotoxin and the peptide CD4M3 (sequence ID n° 17), derived from the scyllatoxin (sequence ID n° 2), were tested by indirect ELISA. These peptides have a capability of inhibiting the rgp120~i - soluble CD4 interaction, with an inhibition concentration at 50 0 (or ICso) of 4.0 x 10-5 and 2.0 x 10-M, respectively (Table I) [4]. These values are 10 000 times higher than the ICso presented by the soluble CD4 (ICso = 4.0 x 10-9 M) [4] .
The appended Figure 1 groups together the results obtained in this example with the peptides of the present invention.
It shows the percentage attachment (°s F) of the viral protein gp120~B2 - CD4 as a function of the molar concentration (C(M)) of competitor, in other words of peptide of the present invention. This Figure shows the inhibition of the CD4 - gp120~B2 interaction.
B 13837.3 EE

The CD4M9 (sequence ID n° 3) was tested in ELISA by competition. It shows the capacity to inhibit the rgp120LAI - soluble CD4 interaction with an inhibition concentration at 50 0 (or ICSO) of 4.0 x 10-' M (Table I) , i.e. an increase in the inhibition capacity by~.a factor of 50 compared to the peptide of the present invention CD4M3 (sequence ID n° 17). The peptide of the present invention CD4M9, in ELISA test by competition is also capable of inhibiting the interaction of the CD4 with l0 other recombinant gp120 proteins, from the T- and M-tropic viruses : HIV-lIII$, HIV-Ice, HIV-lBa_L, HIV-l,~_FL, HIV-1y~61D, with an inhibition concentration at 50 % (or ICso) of between 0.1 and 1.0 x IO-6 M [4] .
Said peptide inhibits the gp120~B2 - soluble CD4 interaction at a Cso of 9 x 10-e M (Figure l, Table I) .
A mutant CD4M9V, comprising the substitution of the Gly - Pro C-terminal sequence by Val, has an ICso of 3.0 x 10-' M in the inhibition of the gp120~I - CD4 interaction (Table 1), which represents a slight increase in the inhibiting capacity. The mutant CD4M9Bip comprises the substitution Phe23Bip compared to the CD4M9; this peptide has an IC5o of 1.0 x 10-' M in the inhibition of the gp120~i - CD4 interaction (Table 1), which represents an additional increase in the inhibiting capacity. The mutant CD4M9T (sequence ID n° 4), comprising a thin -propionic acid substituting the Cys C-terminal amino acid, has an ICso of 3.0 x 10-$ M in the inhibition of the gp120~I - CD4 interaction (Figure 1); the ICso in the gp120~$2 interaction 1.1 x 10-a M, Table 1), which B 13837.3 EE

represents an increase in the inhibiting capacity of a factor close to 10 compared to the CD4M9.
The following mutants all have a thio - propionic (TPA) and valine (Val or Ile) residue in the N- and C
5 terminal position, respectively, and were all tested with the gp120, derived from the T tropic strains of HIV - l~B
and HIV - 1~I.
The peptide of the present invention CD4M27 (sequence ID n° 6) comprises ARgSPhe mutations (to remove l0 a non essential arginine and protect the disulphide bridge 6 - 24), Gly27Va1 and the deletion of the C-terminal protein residue (to better stabilise the ~i structure). It provided an increase in the inhibiting capacity of CD4 - gp120~,I bond, ICS° of 7.0 x 10-9 M
15 (Figure 1, Table I). The peptides CD4M27 A, B and C are obtained respectively by a Gly2lSer (a), Ser22His (b) and Ser22Asn (c) mutation from the peptide CD4M27. They have an inhibiting capacity comparable to that of the peptide CD4M27.
20 The peptide of the present invention CD4M32 (sequence ID n° 12), compared to the CD4M9T, comprises the deletion of the C-terminal praline residue, the G1y27Val mutation and the mutation of the Phe23 in bi-phenylalanine (Bip): the Phe23Bip mutation adds a 25 hydrophobic extension to the lateral chain of the Phe23, thus increasing the interaction with the hydrophobic cavity of the gp120 viral protein. CD4M32 provides an increase in the inhibition of the CD4 - gp120~$z bond, (ICS° of 8.10 x 10-1° M, Figure 1) . The mutant CD4M30 B 13837.3 EE

41 . _ (sequence ID n° 10), compared to the CD4M32, comprises the mutation ARg4Gln, thus increasing the solubility of the molecule. This peptide of the present invention has an ICS° of 3.0 nM, in the inhibition tests of the CD4 -gp120I,p,I interaction. The mutant CD4M33 (sequence ID n°
13) groups together the mutations present in the two previous peptides, particularly CysITPA, ArgSPhe, Phe23Bip, Gly27Val, the deletion of a residue in C-terminal position and, in addition, the mutation Ala4His (in order to increase the solubility of the molecule).
Said mutations have an additive effect and allow the peptide of the present invention CD4M33 to inhibit the CD4 - gp120~BZ and CD4 - gp120~i interaction with an ICso of 1.2 x 10-1° M and 2.5 x 10-1° M, ( Figure 1 and Table 1), i.e. an increase in the inhibiting capacity of the CD4 - gp120 bond by a factor of 1000 compared to the CD4M9. A supplementary peptide Cd4M35 (sequence ID n° 14) was synthesised with, compared to the peptide Cd4M33 of the present invention, the mutations Va127A1e and G1y21 (D)Asp. Said mutations allow the peptide of the present invention to inhibit the CD4 - gp120~$2 interaction with an ICS° of 7.0 x 10-11 M (Figure l, Table 1) .
Two other peptides K15 (sequence ID n° 15) and K15 (sequence ID n° 15) K16 (sequence ID n° 16) were synthesised: they comprise the substitutions Gln7Val, LeuBGln, LysllHis, compared to the peptide CD4M33 of the present invention, and a Lys residue in position 15 or 16, respectively. A fluorescent group of fluorescein was then attached, in a covalent manner, at the level of the B 13837.3 EE

lateral chain of said lysine residue, according to the protocol described in example 1. The fluorescent peptides K15 and K16 have an ICso of 5.0 x 10-8 M and 6.0 x 10-9 M
in the inhibition of the CD4 - gp120~i interaction, in which the affinity for the gp120 viral protein is not modified by the labelling.
In conclusion, the present invention provides a family of peptides that represent powerful inhibitors of the CD4 - gp120 interaction.
Table 1 50 % inhibition concentration (ICso) of the mimicking peptides of the CD4 in the interaction of the soluble recombinant CD4 with the protein of the gp120LAI and gp120~$2 viral envelope. The standard deviation for each value is less than 30 %.
Name Sequence ID ICso ICso n (gp120~I) (gp120~B2) CD4M0 18 4 0 l,.iM -CD4M3 17 2 0 ltM -CD4M9 3 400 nM 90 nM

CD4M9V - 300 nM -CD4M9T 4 30 nM 11 nM

CD4M9Bip - 100 nM -CD4M27 6 10 nM 7 nM

CD4M30 10 3.0 nM -CD4M32 12 2.0 nM 800 pM

B 13837.3 EE

CD4M33 13 250 pM 120 pM

CD4M35 14 - 70 pM

K15FR 15 50 nM -K16FR 16 6 nM -Exam 1e 3: Surface plasmon resonance experiments Surface plasmon resonance experiments were carried out with a Biocore 2000 (Biacore, Uppsala, Sweden) system. The peptides to be tested were coupled with biotin (as described above) then immobilised on a bio-chip on which streptavidin had been previously attached.
The streptavidin was immobilised on the bio-chip as follows: the surface of the bio-chip was first activated by injection of 50 ~zl of coupling agent provided by the manufacturer and capable of forming amide bonds: N
ethyl- N' (dimethylaminpropyl) carbodiimide (EDC} / N
hydroxysuccinimide (NHS), 50/50; then 20 ~Zl of streptavidin, 0.2 mg. ml-i in 10 mM sodium acetate, pH 4.5, were then injected at 5 ul.min-1, followed by a neutralisation of the carboxylic groups activated by 2 x 20u1 of 1.0 M ethanolamine, pH 8.5. The biotinised molecules were. then injected (10 ul.min-1) in a 10 mM
Hepes buffer, 0.3 M NaCl, pH 7.4) on three of the four tracks of the bio-chip. In the first track, the CD4M9 (10 M} was immobilised up to an RU (resonance unit) value of 100; the second track was left untouched (control); the third track was covered with the soluble CD4 ( 10-' M) up to an RU value of 450; on the fourth track, the CD4M33 B 13837.3 EE

(10-' M) was immobilised to an RU value of 100. For each analysis, several concentrations of different gp120 (HXB2, BAL, JRFL strains) were injected, at a rate of 50 ul.min-1 at 25 °C, for an association time of 4 minutes.
The bio-chip was then rinsed with a buffer, 10 mM Hepes, 150 mM NaCl, 3.4 mM EDTA and 0.05 % P20, pH 7.4, in order to analyse the dissociation phase. After each experiment, the bio-chip was regenerated with 25 u1 of 1.0 M formic acid. The dissociation constants were calculated from the to kinetic constants determined by the Bia-evaluation 3.0 software.
Example 4: Antiviral activity of the CD4 mimickers Handling of the infectious material was carried out in an L3 type high security laboratory. In order to be as close as possible to physiopathological conditions, the study was carried out with the aid of primary cultures of mononucleated cells of human peripheral blood (CMSP). In all of the experiments, the effects of the new molecules were compared to those of AZT.
ISOLATION, CULTURE AND ACTIVATION OF THE CELLS
Culture medium The medium A was composed of RPMI cellular culture medium (Life Technologies) supplemented by 10 % of foetal calf serum (FCS, Roche Product) decomplemented by heat at +56 °C for 30 minutes, 2 mM of L-Glutamine (Roche Product) and a 100 ~.zg/ml solution of three antibiotics (penicillin, streptomycin and neomycin; PSN, Life B 13837.3 EE

Technologies). The medium B was composed of the medium A
supplemented by 20 UI/ml of IL-2 human recombinant (Roche Product).
Isolation and activation of the CMSP
5 The CMSP were separated from the other elements appearing from the blood by ficoll gradient centrifugation (MSL 2000, Eurobio): 30 ml of blood, from a healthy donor, diluted to one third were deposited on a 20 ml cushion of ficoll. After 20 minutes of 10 centrifugation at 850 g, the ring of CMSP was removed and washed twice with RPMI 1640, after 10 minutes of centrifugation at 750 g and 5 minutes at 400 g. The CMSP
were then activated for 48 h by 1 lzg/ml of phytoheagglutinine-P (PHA-P; Difco Laboratories). The 15 CMSP were cultured at +37 °C, in an atmosphere saturated in humidity, under 5 o CO2. At the end of 48 hours of mitogenic activation, they were cultured in medium B.
Throughout the culture, the culture supernatants were removed, and the culture media were renewed every three 20 or four days. At each renewal of the culture media, the cellular viability was evaluated by microscopic observation.
EVALUATION OF THE ANTIRETROVIRAL ACTIVITY OF THE

The compounds of the present invention CD4M30, CD4M33 and CD4M35 were solubilised in ppi sterile water (Aguettant), aliquoted then conserved at -20 °C. The compound SAH-CD4 (soluble CD4 coupled to the human B 13837.3 EE

albumin serum supplied by Aventis (Vitry-sur-Seine, France) was conserved at -80 °C up to its use. The solutions and the dilutions were then formed extemporaneously in the medium A. The CMSP were pretreated with the compounds for 1 hour then infected by the lymphocytic HIV - 1~,I tropism reference isolate or by the clinical isolate HIV - lLfiN~ The biological characteristics of this isolate are: rapid/high, scyncitia inducing (SI), X4: it is therefore l0 preferentially suited to infecting the lymphocytes. The viral stock was constituted by amplifying, in vitro, this strain with the aid of mononucleated cells of umbilical blood (CMSO) activated beforehand by lug/ml of PHA-P and cultured in the medium B. In order to eliminate the soluble factors such as the cytokines, the culture supernatants were ultracentrifuged at 360 000 g for 5 minutes, and the pellets resuspended in RPMI 1640. The viral stock thus constituted was then titrated with CMSP
activated by the PHA-P. The TCIDso (50 o Tissue Culture Infectious Dose) was calculated using the Karber formula.
The CMSP were infected with different 10-100 TCIDSO viral doses of the HIV - 1~I strain and by 50 TCIDso of the HIV
- lLSN strain (multiplicity of infection m.o.i. - 0.001).
DOSAGE OF THE VIRAL REPLICATION IN THE CULTURE
SUPERNATANTS
The viral replication was measured on day 7 of the culture, by dosing the Inverse Transcriptase activity in the culture supernatants with the aid of the RetroSys B 13837.3 EE

dosage kit (registered trademark) according to the recommendations of the Innovagen Company.
Analysis of the results and determination of the 50 effective doses.
The 50 % effective doses (EDSO) were calculated using the "Dose-effects analysis with microcomputers"
software developed by J. Chou & T. C. Chou.
EXAMPLE 5: PROTOCOL FOR PRODUCING 'T'HE RECOMBINANT GP120 VIRAL PROTEIN
The fragment coding the gp120 viral protein (amino acid V12 at 8481) was amplified by PCR in the plasmid HIVIIIB / HXB2R, with the aid of two seeds making it possible to generate a fragment containing the BamHI
sites (before the Kpnl site of the gp120 viral protein) and PstI (after the stop codon added to the end of the sequence of gp120 viral protein). The BamHI / PstI
fragment thus obtained was cloned in the Bluescript vector (pBs, Stratagene), leading to the plasmid pBSml.
The sequence coding the N-terminal end of the gp120 viral protein (T1 to G11), as well as the one coding the signal peptide of the ecdysteroid glycosltransferase of the baculovirus of Autographa californica, were inserted between the BamHI and KpnI sites of pBSml, leading to pBSm2.
The BamHI - Pstl fragment of pBSm2 was then inserted between the BglII - PstI sites of the transfer vector p119P of the baculovirus P10. Sf9 insect cells were cotransfected with purified viral DNA of the modified B 13837.3 EE

baculovirus AcSLPIO and the DNA of the recombinant vector p119P gp120. The recombinant viruses were purified by a standard method.
The Sf9 scla cells (line adapted to growth without serum, deposited in the collection of the Institut Pasteur) were maintained in a spinner in a medium without serum. Said cells (5.105 cells / ml) were then infected by the recombinant viruses at a multiplicity of infection of 1 PFU (plaque forming unit) per cell and incubated at 28 °C. After six days of infection, the cells were centrifuged (500 g), and the crude gp120 viral protein was concentrated and directly purified from the culture supernatant by affinity chromatography on a Sepharose -bromacetylate coupled to the anti-gp120 D7324 antibody.
EXAMPLE 6: ATTACHMENT OF THE VIRAL ENVELOPE
In order to better characterise its biological activity, the peptide of the present invention CD4M33 (sequence ID n° 13) was labelled with biotin and with the fluorescent probe fluorescein as described in example 1, at the level of the Lys-11 positioned on the surface opposite the active surface of the peptide of the present invention CD4M33. Its activity remained unchanged in the ELISA tests (Figure 2).
Figure 2 groups together the results obtained in this example. It illustrates curves showing the change in the percentage attachment (o F) of peptides of the present invention to the gp120HxBa viral protein as a B 13837.3 EE

function of the molar concentration of said peptides (C (M) ) .
These results show that the peptide CD4M33 may incorporate a labelling group without its biological activity being altered.
In order to obtain a first evaluation of the affinity of the peptide of the present invention CD4M33 for the gp120 viral protein, an analysis of biospecific interactions (described in example 3), based on the l0 detection by surface plasmon resonance. In this technique, the biotinised peptide CD4M33 of the present invention (prepared as in example 1) was attached in a specific manner to the carboxylated dextran matrix of a bio-chip that had been pregrafted with streptavidin.
Solutions of different recombinant gp120 proteins (gp120rix$a, gp120B~, and gp120JRSL) were then injected onto said matrix and a signal indicating a specific and high affinity association was then detected. Figure 3 shows the change in the plasmonic resonance signal as a function to time, after interaction of the gp120HxBa protein (at different concentrations) with the peptide CD4M33 of the present invention.
After non-linear regression of the association and dissociation curves obtained, one obtains a dissociation constant KD of 2.4.10-9 M for the viral protein gp120xxaz.
of 7.4.10-9 M for the viral protein gp120B~, and of 2.4.10-9 M for the viral protein gpl2OJRFL. These values are comparable to the value of KD = 1.9 x 10-9 M, reported in the literature [34~ for the CD4 - gp120 interaction. The B 13837.3 EE

same technique based on the detection by surface plasmon resonance was also used to check if the peptide of the present invention CD4M33 could attach the viral envelope in its native form. To do this, a suspension of viral 5 particles inactivated by adrithiol-2 [35] was injected on the same bio-chip, grafted with the antibodies 48d and CG10. A specific and high affinity interaction signal was then clearly detected, in the case where the suspension was incubated with the peptide of the present invention 10 CD4M33, but not in the case where the viral suspension was incubated with the peptide CD4M9, much less active, or in the absence of peptide (Figure 4).
Figure 4 shows that the presence of the peptide CD4M33 is essential for the interaction of the viral 15 envelope with the antibodies 48d and CG10, specific to the viral envelope, and that the HIV-1 does not attach itself to the antibodies in its absence: this is an indirect demonstration of the attachment of the peptide CD4M33 to the HIV-1 viral envelope.
20 These experiments show that the peptide of the present invention CD4M33, labelled or not, has a capacity of attaching to the HIV viral envelope both in its isolated and purified recombinant form and in its native form present at the surface of the virus.
EXAMPLE 7: INHIBITION OF INFECTION BY THE AIDS VIRUS
In order to evaluate the antiviral capacity of the peptides of the present invention, the inventors carried out infection experiments by the virus HIV-lLAZ and by the B 13837.3 EE

chemical isolate HIV-1LEN, of primary cultures of mononucleosis cells of human peripheral blood (CMSP). In all of the experiments, the effects of new molecules were compared to those of AZT. The HIV-lei virus was added to the different viral doses (10 - 100 TCIDSO, Table 2) in the presence of variable concentrations of CD4M30, CD4M33, CD4M35 and SAH-CD4. The HIV-lLErr virus was added at TCIDso (Table 4) in the presence of CD4M33. Mimicking toxicity tests of the CD4 were then carried out on the l0 same cells.
A. TOXICITY OF THE CD4 MIMICS
At the tested doses, none of the compounds, AZT, SAH-CD4 or anti-CD4 mimics reduced the viability of the CMSP activated by the PHA-P.
B. ANTI-HIV-h,p,= ACTIVITY OF THE CD4 MIMICS
b.1. Replication of the HIV-1~,I strain in the CMSP.
The HIV-1~I strain replicates to a large extent in the CMSP activated by the PHA-P. The viral replication peak is on day 7 of the culture, and the effects of the peptides CD4M30, CD4M33 and CD4M35 were quantified at this post-infection time.
b.2. Effects of AZT on the re lication of the HIV-1~I
strain in the CMSP.
AZT strongly inhibited the replication of the HIV-1~I
strain in the CMSP activated by the PHA-P (Table 2) . The EDso was equal to 3 - 14 nM.
B 13837.3 EE

b.3. Effects of the SAH-CD4 compound on the replication of the HIV-1~I strain in the CMSP.
The antiretroviral activity of the SAH-CD4 compound towards the CMSP activated by the PHA-P and infected by the HIV-1LAI strain was demonstrated by an inhibition of 85~15 °s of the viral replication at the concentration of 2 . 5 uI~I .
b.4. Effects of CD4 mimics on the replication of the HIV-l~z strain in the CMSP.
The peptides of the present invention CD4M30, CD4M33 and CD4M35 showed an anti-retroviral activity (Table 3) in the cultures of CMSP activated by the PHA-P and infected by the HIV-1~I strain. The compound CD4M33 was the most antiviral of the three at the different viral doses tested. Its activity was lower than that of AZT: EDso -100 - 500 nM vs AZT: EDso = 3 nM (Table 3 below) , but it was higher by at least one base 10 logarithm than that of the derivative of the CD4 receptor, SAH-CD4.
Figures 5A-C group together the results obtained:
24 percentage inhibition (% I) as a function of the peptide concentration in nM.
C. ANTIVIRAL ACTIVITY OF THE COMPOUND CD4M33 TOWARDS THE
CLINICAL ISOLATE HIV-1L~
c1. Effects of AZT on the replication of the HIV-lLfiN
isolate in the CMSP.
AZT strongly inhibited the replication of the HIV-1L~
strain in the CMSP activated by the PHA-P and infected with the HIV-lLSrr isolate, with an EDso equal to 2.2 nM
B 13837.3 EE

Table 4). The degree of inhibition was identical to that observed towards the HTV-1~I lymphocytic tropism reference strain.
c1. Effects of the CD4M33 mimic on the replication of the HIV-1LEN clinical isolate in the CMSP.
The retroviral activity of the CD4M33 mimic towards the CMSP activated by the PHA-P and infected with the HIV-1L~
isolate, was demonstrated by a dose dependent reduction in the viral replication and an EDso equal to 367 nM
(Table 4). The CD4M33 mimic thus conserves a significant anti-HIV-1 activity, even though it is slightly less compared to that observed with the HIV-1~I strain.
These experiments demonstrate that these peptides of the present invention are capable of inhibiting the infection of the cells, even at high viral doses, with EDso values of 900-35 nM (Table 3 below) . The peptide of the present invention CD4M33 is the most antiviral and its EDso is around 100 nM for standard viral infection doses (Figure 4 a-c, Table 3 below).
Furthermore, the peptide CD4M33 has an antiviral activity higher by 1 logarithm of base 10 to that observed with another derivative of the receptor CD4 receptor, SAH-CD4, and above all has shown a significant anti-HIV activity towards a clinical isolate (Table 4).
In conclusion, the peptide of the present invention CD4M33, which is capable of attaching itself to the recombinant gp120 viral protein with a Kd of 2.4 - 8.0 nM
(surface plasmon resonance experiments) and inhibiting the interaction between the recombinant. proteins CD4 -B 13837.3 EE

gp120 in ELISA, with an ICso of 120 - 250 pM, also has the capacity of inhibiting this interaction in primary cultures of human cells, of inhibiting the initial step of the entry and thus blocking the infection even of a HIV-1 clinical isolate.
Table 2 Effect of AZT on the replication of the HIV-l~,i strain in the CMSP activated by the PHA-P. The results are expressed in percentage inhibition t standard deviation:
AZT (nM) % Inhibition 1 39 + 10 10 59 + 28 %

100 87 + 18 %

100 100 t 0 %

10 000 100 + 0 Table 3 50 % effective doses (EDso) of the peptides of the present invention and AZT in cultures of mononucleosis cells of human peripheral blood (CMSP), activated by phytohemagglutinin-P (PHA-P) and infected by the HIV-1~,I
virus at different infectious doses (TCIDso) : 50 % Tissue Culture Infectious Dose):
B 13837.3 EE

5$
_ _ EDso (nM) .

Sequences 100 TCID5o 50 TCIDso 10 TCIDso Table 4 Effect of AZT and the CD4M33 mimic on the replication of the HIV-lL~ clinical isolate in cultures of mononucleosis cells of human peripheral blood (CMSP), activated by phytohemagglutinin-P (PHA-P). The cells were infected by 50 TCIDso of virus HIV- 11,EN AZT CD4M3 3 EDso (nM) 2.2 367 ED~o (nM) 4.8 542 ED9o (nM) 16.5 1006 EXAMPLE 8: EXPOSURE OF THE ANTIGENIC SITES OF THE
ENVELOPE
In order to evaluate the capacity of the peptide of the present invention CD4M33 to induce conformational variations in the gp120 protein, characterised by the exposure of epitopes sensitive to neutralising antibodies, the interaction of such human antibodies as 48d and CG10 with the HIV envelope in the presence of the B 13837.3 EE

CD4M33 and, by comparison, of the recombinant CD4, was undertaken, using the plasmon surface resonance technique. The antibodies 48d and CG10 were covalently attached to the surface of bio-chips, then solutions of gp120, in the absence of CD4 and in the presence of an excess of CD4 and CD4M33 were injected. When the CD4M33 was added to the gp120~$2 viral protein, with a molar excess of 1.5 and 20 times, the viral protein showed a strong interaction with the antibodies (Figures 6a-b); on the other hand, in its absence (and in the absence of soluble CD4), the protein of the envelope interacted with the antibodies very weakly (Figures 6a-b). Moreover, the effect of adding CD4M33 on the increase of the interaction affinity of the gp120 viral protein for the antibodies was comparable to that which is obtained by adding equivalent quantities of soluble CD4.
The inventors also carried out similar plasmon resonance experiments using the HXB2 strain of HIV-1 directly. Figure 4 shows the results obtained in these experiments and shows the change of the plasmon resonance signal as a function of the time after interaction of a suspension of HIV-182 viral particles with a bio-chip with immobilised 48d and CG10 antibodies: the HIV-1 interacted with the antibodies only after incubation of the peptide CD4M33 of the present invention and, on the other hand, in its absence or in the presence of the inactive peptide [A1a23]CD4M9, the interaction was practically zero.
B 13837.3 EE

The plasmon surface resonance technique therefore made it possible to demonstrate that the bond of the CD4M33 to the protein of the gp120 envelope is capable of inducing a modification of the viral protein which, subsequently, exposes in a preferential manner certain molecular surfaces to the neutralising antibodies (17b, CG10), isolated from persons affected by the HIV-1 [12], [13] .
The exposure of these cryptic epitopes, induced by the peptide CD4M33 of the present invention, is effective both in the protein of the envelope in recombinant form and in the envelope of the virions. Moreover, given recent experiments that have shown that the gp120 protein complexed to the CD4 may be a molecular form capable of inducing the formation of neutralising antibodies [31,32, 33], the complex CD4M33 - gp120 therefore represents a very interesting candidate as an immunogen for the induction of neutralising antibodies, particularly in a vaccinal application.
The CD4M33 has the property of unmasking the epitopes of the gp120 viral protein, as does the soluble CD4, with the advantage that it cannot induce an immune response against the CD4 and that, in addition, due to its small size, it allows an even more favourable access to the unmasked epitopes of the gp120 viral protein.
EXAMPLE 9: SCREENING
A) EQUIPMENT AND METHODS
EQUIPMENT
B 13837.3 EE

The equipment used for the measurements was that described in the referenced documents [41], [42] and [43]
and in patent US 5 756 292.
THE PEPTIDES
The peptides of the present invention, labelled or not labelled, in the form of 50 ug lyophilised (powder) were solubilised by adding to an Eppendorf (trademark) containing the powder, 50 u1 of a potassium phosphate buffer, 10 nM, 100 mM KC1, pH 7.0, in order to obtain a solution concentrated to between 300 and 750 ~ZM.
The labelling was carried out with fluorescein (F) or with rhodamine green (R).
Two other solutions, 1 . 10 and 1 . 100 dilutions of the above-mentioned solutions, were prepared with the same buffer.
These labelled peptides were used firstly in bonding experiments to the gp120 in order to determine their affinity and then were used as tracers, in screening experiments by competition of the interaction of said peptides with the gp120 viral protein.
The sCD4 (soluble CD4) used is a recombinant protein produced in the Baculovirus expression system by Intracell (trademark) sold in France by Neosystem Laboratoire (Strasbourg, France).

The gp120 viral proteins were prepared in the form of solutions at a concentration of around 1 mg/ml.
B 13837.3 EE

DIRECT ASSAYS
Assays were carried out in the "reverse" sense.
Firstly, 4 ml of a solution of peptide of the present invention, labelled, and at a concentration of 6 nM in the phosphate buffer mentioned above was prepared. Then, 200 um of a solution containing 10 u1 of gp120 at a concentration of 0.421 uM, 4 u1 of peptide of the present invention labelled with fluorescein and at a l0 concentration of 6 nM, and 186 u1 of buffer were prepared.
All of the anisotropy measurements were carried out using a Beacon 2000 polarimeter (trademark) (Panvera Corp., Madison, WI) with the set of filters corresponding to fluorescein.
The anisotropy of 200 u1 of the solution of labelled peptide alone was firstly tested. Depending on the probe used on the peptide and the adjustment of the apparatus, in particular the gain, the anisotropy of the labelled peptide of the present invention alone varied between 80 and 100 millianisotropy units.
Then, the anisotropy of the tube containing the labelled peptide and the gp120 at 410 nM was measured.
Generally, the value of the labelled peptide in the presence of this high concentration of gp120 was around 200 millianisotropy units.
The assay was carried out by taking 40 u1 samples of the solution and then adding 40 u1 of the solution of labelled peptide alone at 6 nM. In this manner, the gp120 B 13837.3 EE

was diluted, by a value of 0.1 unit log by dilution, while maintaining constant the concentration of labelled peptide. This reverse assaying protocol uses the least possible amount of gp120.
5 The anisotropy measurements were recorded up to the stabilisation of the signal. In the case of this interaction, the stabilisation was relatively long, since it lasted more than 20 minutes for the first point, but then the dilution points balanced out more rapidly, in 2 10 to 4 minutes.
All of the assays were carried out at 21 °C, in a Beacon 2000 (trademark) temperature controlled sample holder.
The labelled peptides showed a tendency to stick to 15 the walls of the recipients. Several dilution solutions were then performed in order to avoid a progressive reduction in the concentration of labelled peptide.
COMPETITIONS
20 For the competition tests, simulations of the competition of the peptide CD4M33 labelled with fluorescein (CD4M33 - F) with the non-labelled peptide CD4M33 were carried out. The results of these simulations are given in appended Figure 13.
25 These simulations showed that the ideal conditions for the competition are 20 mM of gp120 and 6 nM of labelled peptide.
The competition was then carried out between 10 and 100 nM of non-labelled peptide CD4M33.
B 13837.3 EE

The competitions were experimentally carried out by adding to 200 ~.zl a 6 nM solution of peptide CD4M33 - F
and 20 nM gp120 of aliquots of 1 u1 of non-labelled competitor peptide CD4M33 in the phosphate buffer above in order to cover concentrations between 10 and 100 nM.
The anisotropy values were then measured after each addition.
The resulting curves of these competitions were analysed in order to determine the Kd of interaction l0 between the competitor and the gp120 while maintaining constant the Kd between the peptide CD4M33 - F and the gp120.
The analyses and the simulations were carried out using the "BIOESQ" programme described in documents [41], [42] and [43] .
B) RESULTS
1) AFFINITY MEASUREMENTS Y~fITH DIFFERENT PEPTIDES OF THE

The principle of the screening method of the present invention is shown in appended Figures 8a), 8b) and 8c).
In order to obtain the results shown in these figures, a peptide of the present invention labelled with rhodamine green, called K16 - R, in solution at a concentration of 6 nM was assayed by a solution of HXB2 strain gp120.
In Figure 8a), the preparation of gp120 was a little less active than that used in Figure 8 b). Indeed, in the first case (I K16 / gp120), the affinity (Kd} extracted B 13837.3 EE

from the analysis of the data according to a simple attachment model, in other words one molecule of gp120 attaching to one molecule of labelled peptide of the present invention, was 176 nM, whereas in the second case (II K16 / gp120), shown in Figure 8b), the affinity was 46 nM.
A third preparation was therefore made (III K16 /
gp120). The measurements carried out on this preparation are shown in Figure 8c). In this case, the measured affinity (Kd) was 97 nM.
Three other labelled peptides, CD4M33, CD4K15 and CD4M9, were tested for their interaction with the HXB2 strain of gp120. The results of these tests are shown in Figures 9, 10 and 11. Their affinity for the gp120 viral protein was respectively 14 nM, 195 nM and 323 nM.
2) AFFINITY MEASUREMENTS CARRIED OUT WITH A PEPTIDE OF
THE PRESENT INVENTION FOR DIFFE1~ENT GP120 VIRAL PROTEINS
The affinities of the peptide CD4M33 for gp120 proteins derived from different viral strains: HXB2 and SF2, were also determined.
They are shown in Figures 12a) and 12b).
The affinity of 8 nM for the HXB2 gp120 is very close to the value determined from the results shown in Figure 9, in other words 14 nM.
On the other hand, the preparation of SF2 gp120 showed a stronger affinity for the peptide CD4M33, of around 1 nM.
B 13837.3 EE

3) DISPLACEMENT OF A LABELLED PEPTIDE OF THE PRESENT

LABELLED PEPTIDE
A displacement of a labelled peptide with a determined affinity, represented by the peptide CD4M33, by a non-labelled peptide CD4M33 was carried out in order to simulate a screening according to the method of the present invention. The results of this example are shown in Figure 13.
l0 These results show that a concentration of 20 nM of gp120 enables an attachment of around 50 0, and therefore a relatively low initial anisotropy of around 180 millianisotropy units.
It should be noted that the large change of IS anisotropy observed during the attachment of the labelled peptides of the present invention by the gp120, DA ~ 180 millianisotropy units, confers a wide dynamic range to these measurements, and enables a very high signal to noise ratio. Consequently, with a standard deviation of 2 20 millianisotropy units in the measurements, it is possible to detect changes of 2 % in the quantity attached.
Consequently, a competition with a molecule with the same affinity as the CD4M33 results in 100 a displacement with a concentration of the competitor close to 100 nM.
25 The inventors therefore used the competition of the labelled CD4M33 with other non-labelled peptides of the present invention in order to determine the affinity of said peptides for the HXB2 gp120.
B 13837.3 EE

The results of these competitions are grouped together in Figures 14a) to i).
A competition of the labelled CD4M33 with a non labelled CD4M33 was first carried out in order to determine if the labelling perturbs the affinity for the gp120.
In direct attachment, represented in Figures 10 and 11, the affinity of the labelled CD4M33 for the HXB2 gp120 was 11~ 3 nM.
The analysis of the competition data was carried out by setting the affinity value of the labelled CD4M33 to its value of 11 nM and allowing the affinity value of the competitor to evolve, in this case the non-labelled CD4M33 peptide. The affinity value corresponding to the line crossing through the dots in Figure 14a) was 14 nM.
It was therefore concluded that the labelling of the CD4M33 does not affect its affinity for the gp120.
Then, several other non-labelled peptides of the present invention as well as the soluble domain of the human CD4 (sCD4) were tested. The results of these tests are represented in Figures 14b) to 14i). Their affinity differs by a f actor of 40. The results obtained are given in Table 5 below.
It should be noted that the affinity of the peptide CD4M9 for the gp120 viral protein obtained by competition, Kd = 359 nM, was the same, taking account of the measurement errors estimated at ~%, as that found by direct assaying of the labelled CD4M9 (Kd = 323 nM).
B 13837.3 EE

Given that the screened molecules are not observed directly, but in competition with the labelled peptide of the present invention, this screening test may highlight any molecule, whatever its chemical structure, which 5 attaches to the gp120 in a competitive manner with this peptide.
Table I
Peptide of Method Gp120 Affinity Rd the present (nM) invention K16 Direct HXB2 176 K16 Direct HXB2 46 K16 Direct HXB2 97 K15 Direct HXB2 195 CD4M9 Direct HXB2 323 CD4M9 Competition HXB2 359 CD4M33 Direct HXB2 14 CD4M33 Direct HXB2 8 CD4M33 Direct SF2 1 CD4M33 Competition HXB2 14 CD4M35 Competition HXB2 9 CD4M32 Competition HXB2. 23 sCD4 Competition HXB2 9 CD4M27 Competition HXB2 148 CD4M9TPA Competition HXB2 26 CD4M9BIP Competition HXB2 9?

(seq. ID n 19) Competition HXB2 320 B 13837.3 EE

(seq. ID n 20) The peptides of the present invention thus represent in vitro inhibitors of the infection by the AIDS virus, particularly the HIV-1 virus.
Consequently, the screening method of the present invention, which uses the fluorescence anisotropy of a labelled peptide of the present invention, constitutes a new screening tool for candidate molecules for the manufacture of medicines for treating AIDS.
B 13837.3 EE

REFERENCES
[1] Kwong, P. D. , Wyatt, R. , Robinson, J. , Sweet, R. W. , Sodroski, J., Hendrickson, W. A. (1998) Nature 393, 648 - 659.
[2] Sweet, R. W., Truneh, A., Hendrickson, W. A.
(1991) Curr. Opin. Biotechnol. 2, 622 - 633.
[3] Ryu, S. E., Trueh, A., J., Sweet, R. W., Hendrickson, W. A. (1994) Structure 2, 59 -74.
[4] Vita, C., Drakopoulou, E., Vizzavona, J., Rochette, 5., Martin, L., Menez, A., Roumestand, C., yang, Y.
S., Ylistagui, L., Benjouad, A., Gluckman, J. C. 1999, Proc. Natl. Acad. Sci. USA, 96, 13091 - 13096.
[5] Feng, Y., Broder, C. C, Kennedy, P. E., Berger, E.
IS A. (1996) Science 272, 872 - 877.
[6] Choe, H., Farzan, M., Sun, Y., Sullivan, N., Rollins, B., Ponath, P. D., Wu, L., Mackay, C.R., LaRosa, G., Newman, W., Gerard, N., Gerard, C., Sodroski, J. (1996) CeII 85, 1135 - 1148.
[7] Dragic, T., Litwin, V., Allaway, G. P., Martin, S.
R., Huang, Y., Nagashima, K. A., Cayanan, C., Maddon, P. J., Koup, R. A., Moore, J. P., Paxton, W. A. (1996) Nature 381, 667 - 673.
[8] Alkhatib, G., Combadiere, C., Broder, C. C., Feng, Y., Kennedy, P. E., Murphy, P. M., Berger, E. A.
(1996) Science 272, 1955 - 1958.
[9] Wu, L., Gerard, N. P., Wyatt, R., Chloe, H., Parolin, C., Buffing, A., Cardoso, A. A., Desjardin, B 13837.3 EE

E., Newman, W., Gerard, C., Sodroski, J. (1996) Nature 384, 179 - 183.
[10] Trkola, A., Dragic, T., Arthos, J., Binley, J. M., Olson, W.C., Allaway, G. P., Cheng-Mayer, C., Robinson, J., Maddon, P. J., Moore, J. P. (1996) Nature 384, 184 - 187.
[11] Thali, M., Moore, J. P., Furman, C., Charles, M., Ho, D. D., Robinson, J., Sodroski, J. (1993) J.
Virol. 67, 3978 - 3988.
14 [12] Wyatt, R., Moore, J., Accola, M., Desjardin, E., Robinson, J., Sodroski, J. (1995) J. ViroZ. 69, 5723 -5733.
[13] Moore, J., Sodroski, J. (1996) J. Virol. 70, 1863 - 1872.
[14] Chen, S., Chrusciel, R. A., Nakanishi, Ho, Raktabutr, A., Jonhson, M. E., Sato, A., Weiner, D., Howie, J, Sagarovi, H. U., Greene, M. I., Kahn, M.
(1992), Proc. Natl. Acad. Sci., USA, 89, 5872 -5876.
[15] Zhang, X., Gaubin, M., Briant, L., Srikantan, V., Murali, R., Saragoui, H. U., Weiner, D., Devaux, C., Autiero, M., Piatier-Tonneau, D., Greene, M. I. (1997) Nature Biotechnol., 15, 150 - 154.
[16] Moore, J. P., Sweet, R. W. (1993), Perspect. Drug Disc. Design I, 235 - 250.
[17] Allaway, G. P., Davis-Bruno, K. L., Beaudry, G. A., Garcia, E. B., Wong, E. L., Ryder, A. M., Hasel, K. W., Gauduin, M. C., Koup, R. A., McDougal, J. S., et al., (1995) AIDS ReS. Hum. Retroviruses ZL, 533 - 539.
B 13837.3 EE

[18] Ono, M., Wada, Y., Wu, Y., Nemori, R., Jinbo, Y., Wang, H., Lo, K. M., Yamaguchi, N., Brunkhorst, B., Otomo, H., Wesolowski, J., Way, J. C., Itoh, I., Gillies, S., Chen, L. B. (1997) Nat. Biotechnol. 15, 343 - 348.
[19] Ojwang, J. 0., Buckheit, R. W., Pommier, Y., Mazumder, A., De Vreese, K., Este, J. A., Reymen, D., Pallansch, L. A., Lackman-Smith, C., Wallace, T. L., et al (1995) Antimicrob. Agents Chemother. 39, 2426 - 2435.
[20] Rusconi, S., Moonis, M., Merrill, D. P., Pallai, P.
V., Neidhardt, E. A., Singh, S. K., Willis, K. J., Osburne, M. S., Profy, A. T., Jenson, J. C., Hirsch, M.
S. (1996) Antimicrob. Agents Chemother. 40, 234 - 236.
[21] De Clercq, E. (1998) Pure Apps. Chem. 70, 567 - 577.
[22] Yahi, N., Fantini, J., Baghdiguian, S., Mabrouk, K., Tamalet, C., Rochat, H., Van Rietschoten, J., Sabatier, J. M. (1995) Proc. Natl. Acad. Sci. USA 92, 4867 - 4871.
[23] De Clercq, E., Yamamoto, N., Pauwels, R., Balzarini, J., Witvrouw, M., De Vreese, K., Debyser, Z., Rosenwirth, B., Peichl, P., Datema, R., Thornton, D., Skerlj, R, Gaul, F., Padmanabhan, S., Bridger, G., Henson, G., Abrams, M. (1994) Antimicrob. Agents Chemother. 38, 668 - 674.
[24] Howard, O. M. , Oppenheim, J. J. , Hollignshead, M. G, Covey, J. M., Bigelow, J., McCormack, J. J., Buckheit, R. W, Jr., Clanton, D. J., Turpin, J. A., Rice, W. G.
(1998) J. Med. Chem. 41, 2184 - 2193.
B 13837.3 EE

[25] Tamamura, H., Muratami, T., Masuda, M., Otaka, A., Takeda, W., Ibuka, T., Nakashira, H., Waki, M., Matsumoto, A., Yamamoto, N. et al. (1994), Biochem.
Biophys. Res. Commun, 205, 1729 - 1735.
5 [26] Tamamura, H., Omagari, A., Oishi, S., Kanamoto, T., Yamamoto, N., Peiper, S. C., Nakasima, H., Otaka, A., Pujiri, N. (2000) Bioorg. Med. Chem. Lett. 10, 2633 -2637.
[27] Baba, M., Nishimura, O., Kanzaki, N., Okamoto, M., 10 Sawada, H., Iizawa, Y., Shiraishi, M., Aramaki, Y., Okonogi, K., Ogawa, Y., Meguro, K., Fujino, M. (1999) Proc. Natl. Acad. Sci. USA 96, 5698 - 5703.
[28] Wild, C., Dubay, J. W., Greenwell, T., Baird, T.
Jr., Oas, T. G., McDanal, C., Hunter, E., Matthews, 15 T. (1994) Proc. Natl. Acad. Sci. USA 91, 9770 - 9774.
[29] Rimsky, L. T., Shugars, D. C., Matthews, T. J.
(1998) J. Virol. 72, 986 - 993.

[30] Chun, T. W., Fauci, A. S. (1999) Proc. Natl. Acad.
Sci. USA 96, 10958 - 10961.
20 [31] LaCasse, R. A., Follis, K. E., Txahey, M., Scarborough, J. D., Littman, D. R., Nunberg, J. H.
(1999) Science 283, 357 - 362.
[32] Devico, A., Silver, A., Thronton, A. M., Sarngadharan, M. G., Pal, R. (1996) Virology 218, 258 25 - 263.
[33] Fouts, T. R., Tuskan, R., Godfrey, K., Reitz, M., Hone, D., Lewis, G. K., DeVico, A. L. (2000) J. Virol.
74, 11427 - 11436.
B 13837.3 EE

[34] Rossio, J. L., Esser, M. T., Suryanarayana, K., Schnieder, D. K., Bess, J. W. Jr., Vasquez, G. M., Wiltrout, T. A., Chertova, E., Grimes, M: K., Sattentau, Q., Arthur, L. O., Henderson, L. E., Lifson, J. D. (1998) J. Virol. 72, 7992 - 8001.
[35] Lee, S., Peden, K., Dimitrov, D. S., Horder, C. C., Manischewitz, J., Denisova, G., Gershoni, J. M., Golding, H. (1997) J. Virol. 71, 6037 - 6043.
[36] Wu,M., Myszka, D. G., Tendian, S. W., Brouillette, C. G., Sweet, R. W., Chaiken, I. M., Heudrickson, W.
A. (1996} Proc. Natl. Acad. Sci. USA 93, 120530 15035.
[37] McMahon, J. B., Beutler, J. A., O'Keefe, B. R., Goodrum, C. H., Myers, M. A., Boyd, M. R. (2000) J.
Biomol. Screen. 5, 169 - 176.
[38] Sportman, J. R., Leytes, L. J. (2000} Drug Discov.
Today 1, 27 - 32.
[39] Mathis, G., Biochemistry, 1993, 32(43), 11682 - 7.
[40] Pernelle, C. et al. , Clin. Chem. , 1993, 39 (9) , 1953 - 9.
[41] Royer, C. A. , Smith, W. R. , Beechem, J. M. , Analysis of Binding in Macromolecular Complexes: A Generalized Numerical Approach, Analytical Biochem. 210, 91 - 97 (1993).
[42] Royer, C. A. , Beecham, Numerical Analysis of Binding Data: Advantages, Practical Aspects and Implications, Methods in Enzymology, Vol. 210, 481 (1992).
B 13837.3 EE

[43~ Royer, C. A., Improvement in the Numerical Analysis of Thermodynamic Data from Biomolecular Complexes, Analytical Biochem., 210, 91 - 97 (1993).
B 13837.3 EE

LIST OF SEQUENCES
<110> COMMISSARIAT A L'ENERGIE ATOMIQUE
INSERM
<120> METHOD FOR SCREENING MOLECULES CAPABLE OF ATTACHING THEMSELVES TO THE

<130> B13837EE
<140>
<141>
<150> FR N° 0109015 <151> 2001 - 07 - 06 <160> 20 <170> PatentIn Ver. 2.1 <210> 1 <211> 16 <212> PRT
<213> Homo Sapiens <220>
<223> Sequence G1n33 at Pro48 of the human CD4 <400> 1 Gln Ile Lys Ile Leu Gly Asn Gln Gly Ser Phe Leu Thr Lys Gly Pro <210> 2 <211> 31 <212> PRT
<400> scorpion <400> 2 Ala Phe Cys Asn Leu Arg Met Cys Gln Leu Ser Cys Arg Ser Leu Gly Leu Leu Gly Lys Cys Ile Gly Asp Lys Cys Glu Cys Val Lys His <210> 3 <211> 28 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <400> 3 Cys Asn Leu Ala Arg Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Phe Cys Ala Cys Gly Pro <210> 4 <211> 28 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <400> 4 Xaa Asn Leu Ala Arg Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Phe Cys Ala Cys Gly Pro <210> 5 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <220>
<221> MOD_RES
<222> (23) <223> bi-phenylalanine or naphtylalanine <400> 5 Xaa Asn Leu His Phe Cys Val Gln Arg Cys His Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Xaa Cys Ala Cys Val <210> 6 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <400> 6 Xaa Asn Leu Ala Phe Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Phe Cys Ala Cys Val <210> 7 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <400> 7 Xaa Asn Leu Ala Phe Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Ser Ser Phe Cys Ala Cys Val <210> 8 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <400> 8 Xaa Asn Leu Ala Phe Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly His Phe Cys Ala Cys Val <210> 9 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD RES

<222> (1) <223> thiopropionic acid <400> 9 Xaa Asn Leu Ala Phe Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Asn Phe Cys Ala Cys Val <210> 10 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <220>
<221> MOD_RES
<222> (23) <223> bi-phenylalanine or naphtylalanine <400> 10 Xaa Asn Leu Gln Phe Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Xaa Cys Ala Cys Val <210> 11 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <220>
<221> MOD_RES
<222> (23) <223> bi-phenylalanine or naphtylalanine <400> 11 Xaa Asn Leu His Phe Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Gln Gly Ser Xaa Cys Tyr Cys Val <210> 12 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <220>
<221> MOD_RES
<222> (23) <223> bi-phenylalanine or naphtylalanine <400> 12 Xaa Asn Leu Ala Arg Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Xaa Cys Ala Cys Val <210> 13 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <220>
<221> MOD_RES
<222> (23) <223> bi-phenylalanine or naphtylalanine <400> 13 Xaa Asn Leu His Phe Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Xaa Cys Ala Cys Val <210> 14 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <220>
<221> MOD_RES
<222> (23) <223> bi-phenylalanine or naphtylalanine <220>
<221> MOD_RES
<222> (21) <223> D isomer of Asp <400> 14 Xaa Asn Leu His Phe Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Xaa Ser Xaa Cys Ala Cys Ile <210> 15 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <400> 15 Xaa Asn Leu His Phe Cys Val Gln Arg Cys His Ser Leu Gly Lys Leu Gly Lys Cys Ala Gly Ser Phe Cys Ala Cys Val <210> 16 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (1) <223> thiopropionic acid <400> 16 Xaa Asn Leu His Phe Cys Val Gln Arg Cys His Ser Leu Gly Leu Lys Gly Lys Cys Ala Gly Ser Phe Cys Ala Cys Val <210> 17 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <400> 17 Cys Asn Leu Ala Arg Cys Gln Leu Ser Cys Lys Ser Leu Gly Leu Lys Gly Gly Cys Gln Gly Ser Phe Cys Thr Cys Gly <210> 18 <211> 33 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <400> 18 Val Ser Cys Thr Thr Ser Lys Glu Cys Trp Ser Val Cys Gln Arg Leu His Asn Thr Ser Lys Gly Gly Cys Gln Gly Ser Phe Cys Thr Cys Gly Pro <210> 19 <211> 28 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <220>
<221> MOD_RES
<222> (23) <223> bi-phenylalanine <400> 19 Cys Asn Leu Ala Arg Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Xaa Cys Ala Cys Gly Pro <210> 20 <211> 27 <212> PRT
<213> Artificial sequence <220>
<223> Description of the artificial sequence: sequence derived from scyllatoxin <400> 20 Cys Asn Leu Ala Arg Cys Gln Leu Arg Cys Lys Ser Leu Gly Leu Leu Gly Lys Cys Ala Gly Ser Phe Cys Ala Cys Val

Claims

73

1. Method for screening molecules capable of attaching themselves with a defined affinity to the gp120 protein of the immunodeficiency virus, said method comprising at least one screening cycle comprising the following steps:
a) covalent labelling with a fluorescent probe of a peptide with said defined affinity for the gp120 protein, said peptide comprising the following sequence (A):
TPA - P1 - Cys - P2 - Cys - P3 -Cys - Ala or Gln -Gly or (D)Asp or Ser - Ser or His or Asn - Xaa j - Cys -Thr or Ala - Cys - Xaa k - NH2, in which TPA represent thiopropionic acid, Xaa j represents .beta.-naphtylalanine, phenylalanine or bi-phenylalanine, Xaa k represents Gly, Val or Ileu, P1 represents 3 to 6 amino acids, P2 represents 2 to 4 amino acids and P3 represents 6 to 10 amino acids, the amino acids in P1, P2 and P3 being natural or not natural, identical or different, and P1, P2 and P3 having or not a shared sequence, said peptide having a .beta. hairpin conformation in which the .beta.
bend is formed by the amino acid residues Ala or Gln -Gly or Dasp or Ser - Ser or His or Asn - Xaa j of the sequence (A), b) bringing the labelled peptide into the presence of the gp120 protein, at a concentration of gp120 which allows between 20 and 50 0 of attachment of said labelled peptide, in such a way as to form a reversible complex [labelled peptide - gp120], c) measuring the standard fluorescence polarisation of the complex (labelled peptide - gp120), d) screening testing by bringing into competition the complex (labelled peptide - gp120) formed with at least one candidate molecule, capable of attaching itself to the gp120 protein, at a defined concentration of said molecule(s) and under physiochemical conditions that enable the competition between said molecule and the labelled peptide to form, if appropriate, a complex with the gp120 protein, e) measuring the fluorescence polarisation emitted by the screening test, and f) comparing the fluorescence polarisation measurement of step e) with the standard measurement of step c), the candidate molecule(s) being considered as a molecule capable of attaching itself to the gp120 protein when the fluorescence polarisation measurement of step e) is less than the standard measurement of step c).

2. Method according to claim 1, in which the sequence (A) may further comprise a proline residue linked to the Xaa k residue.

3. Method according to claim 1, in which the sequence is chosen from among the sequences ID n o 4, ID
n o 5, ID n o 6, ID n o 7, ID n o 7, ID n o 9, ID n o 10, ID n o 11, ID n o 12, ID n o 13, ID n o 14, ID n o 15, ID n o 16, ID

n o 17, ID n o 18, ID n o 19 and ID n o 20 of the appended sequence list.

4. Method according to claim 1, in which the fluorescent probe is a fluorophor with absorption properties in the visible.