FR3018919A1 - METHOD FOR SCREENING ERBB2 INHIBITORS AND APPLICATIONS THEREOF - Google Patents

Abstract

The present invention relates to the field of modulators of the ErbB2 receptor, and proposes a novel screening method for identifying inhibitors of said receptor, as well as a method for preparing a pharmaceutical composition comprising the said (s) inhibitor (s) for the prevention and / or treatment of pathologies associated with overactivation and / or overexpression of ErbB2. The present invention also provides a kit that can be used in said methods.

Description

INTRODUCTION The present invention relates to the field of modulators of the ErbB2 receptor, and proposes a new screening method for identifying inhibitors of said receptor, as well as a method for preparing a pharmaceutical composition comprising the said (s) ) inhibitor (s) for the prevention and / or treatment of pathologies associated with overactivation and / or overexpression of ErbB2. The present invention also provides a kit that can be used in said methods. ErbB2 (receptor for human epidermal growth factors, also known as HER-2) is a 185 kDa cellular receptor belonging to the family of tyrosine kinase receptors, which structurally consists of an ectodomain, a transmembrane segment, and a cytoplasmic region comprising a tyrosine kinase domain. Activation of this receptor results in autophosphorylation of tyrosines present in the intracellular domain of the latter, which in turn exerts, via various intracellular signaling pathways (PI-3 kinase, ERK, p38, etc.), an effect on many cellular processes such as transformation, proliferation, survival, apoptosis and cell development. It has long been known that deregulation of this receptor is responsible for the development of many types of human cancers, including breast, ovarian, stomach and salivary cancers. To date, it is estimated that about 20 to 30% of breast cancers are caused by abnormal gene amplification leading to overexpression of ErbB2 (so-called ErbB2 positive tumors, or ErbB2 + tumors) (Yarden et al. , 2001). This overexpression promotes ligand-independent activation of ErbB2 and induces an increase in cell proliferation and metastatic potential. Fast-growing tumors related to this overexpression are notably more aggressive and much less sensitive to chemotherapy or hormone therapy, and it has been shown that patients with ErbB2 + breast tumors have a higher proportion of visceral and cerebral metastases. These tumors are therefore associated with a poor clinical prognosis.

Constitutive activation of the ErbB2 receptor may also result from the presence of somatic mutations in the transmembrane domain or the kinase domain, as has been observed in mammary, gastric and colorectal carcinomas (Lee et al., 2006). In addition, it has been shown that in ErbB2 receptors, the ErbB2 receptor undergoes proteolytic cleavage generating a 95 kDa fragment (p95) of the latter (truncated ErbB2 receptor p95ErbB2) which exhibits greater kinase activity than the wild type receptor. The expression of this truncated form, devoid of extracellular domain, very oncogenic, is correlated with an increased appearance of ganglionic metastases, and is associated with a poor clinical prognosis (Saez et al., 2006). As a result, the ErbB2 receptor is a major therapeutic target in the treatment of cancers, particularly ErbB2 + breast cancers, which have a relatively high prevalence in the population (20 to 30% of primary breast cancers, or approximately 8,000 patients). per year in France).

However, despite all the interest in this receptor and the efforts made to date to develop an effective therapy, current strategies to target the extracellular domain of this receptor (anti-EBb2 monoclonal antibody therapies such as Trastuzumab or Pertuzumab, used alone or in combination with chemotherapy) or receptor kinase activity (therapies using small kinase inhibitory molecules, such as Lapatinib, also known as Tykerb) ( Spektor et al., 2007) have been shown to have limited action, and are highly costly for patients. These molecules are also not effective on the mutated and truncated ErbB2 receptor forms. With respect to Trastuzumab, approximately 66 to 88% of treated patients do not respond to treatment (ie, patients with so-called primary or intrinsic resistance), and nearly one-third of patients initially responding to this antibody develop following year, a so-called acquired resistance, with metastasis occurring in 15% of cases (Bublil et al., 2007). The appearance of this type of resistance has been shown to be mainly related to the expression of the p95ErbB2 fragment, since this highly active truncated form lacks the Trastuzumab recognition site (Bublil et al., 2007). In addition, this antibody has no effect on cancer cells expressing low levels or mutated forms of the receptor. Another problem with this antibody is the risk of cancer progression in the central nervous system, because of its high molecular weight which limits its access to the brain: in fact, about 20 to 30% of women on Trastuzumab-based therapy develop metastases in the brain. Treatments with small molecules that inhibit kinase activity, such as Lapatinib, are generally associated with the development of significant toxic effects due to their lack of specificity, which greatly limits their use. for security reasons. In particular, it has been shown that the median duration of Lapatinib response is less than one year, and that a majority of patients pre-treated with Trastuzumab (-80%) do not respond.

There is therefore an urgent need to identify new ErbB2 receptor inhibitors that do not have these disadvantages. It has been previously discovered that a new family of ErbB2 receptor partners controls the physiological activity of the ErbB2 receptor: it is the family of ERM proteins (Ezrin, Radixin, Moesin) which are naturally present in cells and which are, moreover, known for their ability to establish a link between the proteins expressed on the surface of the cells and the cellular skeleton. These ERM molecules have the particular feature of binding directly to the ErbB2 receptor via a subdomain that they have in common within their amino-terminal end, referred to as the FERM domain (F for protein 4.1, E for Ezrin, R for Radixin, M for Moesin). To do this, ErbB2 has, unlike the other members of the ErbB receptor family, a specific binding motif to this FERM domain, called EBM or EBD (Ezrin binding domain) in its juxtamembrane cytosolic region (that is, the acids Amines 674 to 689). It has thus been demonstrated that the ErbB2 receptor interacts specifically with the Ezrin, Moesin and related Merlin FERM domain, and that this interaction stabilizes ErbB2 in a catalytically inactive state by exerting molecular stress on the domain. juxtamembrane of ErbB2, thus limiting the access of the kinase domain to tyrosine substrates. Interestingly, the expression of Moesin is absent in 97% of ErbB2 + tumors. The introduction of the Ezrin or Moesin FERM domain into ErbB2 overexpressing breast cancer cells is sufficient to inhibit the functional activity of ErbB2 and the growth of these cells. Importantly, this domain also has an effective action on mutated forms of the receptor, including the highly metastatic p95ErbB2 form that is insensitive to current therapies. The discovery of this new ErbB2 receptor-specific regulatory mechanism, which was the subject of patent application VVO / 2011/036211 A1, thus paves the way for the development of new, more targeted therapeutic approaches, based on the use of of the FERM domain to inhibit the activation of overexpressed ErbB2 receptors, as well as that of mutated and truncated forms of the latter. It is an object of the present invention to provide a novel strategy for identifying compounds mimicking the action of the FERM domain and thereby blocking the activity of the ErbB2 receptor. To do this, the inventors have developed a new, robust, reproducible and inexpensive two-step screening method based on the Alphascreen (homogeneous amplified luminescence proximity test) technology, which can, if necessary, be made in a format suitable for high throughput screening. The first step of this method relies more particularly on the study of the destabilization of the interaction between the juxtamembrane region of ErbB2 (Ezrin binding domain or EBM) and the Ezrin FERM domain by one or more candidate compounds. . In order to discriminate the compounds destabilizing the interaction between ErbB2 and FERM by interacting with ErbB2 of those interacting with the FERM domain, this method comprises a second so-called counter-screening step, which measures the interaction between the FERM domain and the motif. ERM binding in the juxtamembrane region of a protein other than ErbB2, such as CD44. This method makes it possible to reliably identify compounds which bind the juxtamembrane region of ErbB2 and which can thus inhibit the activity of this receptor.

This process has been successfully validated by screening 1,200 FDA-approved bioactive compounds and 320 phytochemicals, all selected on the basis of their high chemical and pharmacological diversity, bioavailability and availability. their possible toxicity in humans. Indeed, this screening has made it possible to identify efficiently and selectively several compounds of interest, whose ErbB2 inhibitory activity has been validated in cellulo. In addition to their ability to mimic the action of the FERM domain, these compounds are small molecules and are likely to circumvent the difficulties encountered with current therapies against ErbB2 + cancers.

The present invention thus proposes a new screening method, easy to implement, making it possible to select ErbB2 inhibitors efficiently and reliably and at a lower cost, based on the study of the interaction between the FERM domain and the EBM domain of ErbB2. The present invention also provides a process for the preparation of a pharmaceutical composition comprising an inhibitor identified by this screening, as well as a kit for carrying out the methods described herein. DETAILED DESCRIPTION OF THE INVENTION Unless otherwise indicated, the scientific and technical terms used in the context of the present invention will have the meanings that are commonly accepted by those skilled in the art. By "ErbB" is meant herein a tyrosine kinase receptor belonging to the family of epidermal growth factor receptors, which includes in particular ErbB1, ErbB2, ErbB3 and ErbB4 receptors. ErbB receptors generally comprise an extracellular domain, capable of binding to a ligand specific thereto, a lipophilic transmembrane domain, a highly conserved intracellular tyrosine kinase domain, as well as a C-terminal intracellular signaling domain having tyrosine residues. whose phosphorylation controls growth, survival, adhesion, migration as well as cell differentiation. The terms "ErbB2" and "HER2" can be used interchangeably herein, and refer to the human HER2 protein described, for example, by Semba et al. (1985) and Yamamoto et al. (1986). As discussed above, this term also covers truncated and mutated forms of this protein (Lee et al., 2006, Saez et al., 2006). Moreover, in addition to the domains common to ErbB receptors, ErbB2 has in its cytosolic juxtamembrane region a motif of interaction with ERM domains of ERM proteins, called EBM (Ezrin-binding domain), which has been described. in the patent application VVO 2011/036211. By "FERM" (F for protein 4.1, E for Ezrin, R for Radixin, M for Moesin), "FERM domain", "4.1N30" or "membrane attachment domain" is here referred to as the N-terminal portion. highly conserved ERM proteins (Ezrin, Radixin, Moesin), which has been described in particular by Chishti et al. (1998), and whose ability to bind to the EBM motif of the ErbB2 receptor and thereby prevent ligand-independent activation of the latter and the signaling events leading to cell proliferation has been demonstrated in the VVO patent application. 2011/036211. The FERM domain generally consists of three subdomains, called subdomains A, B and C (or lobes F1, F2 and F3), the subdomain C (or lobe F3) being sufficient for binding to the EBM motif. . By "polypeptide", "protein", "peptide" is meant here an amino acid sequence connected by peptide bonds (-NHCO-) - which (s) are their length or modification (s) post-translational (s) - which may be of natural or synthetic origin, and which may play a structural and / or functional role in a cell, in vitro and / or in vivo. This term further encompasses functional variants and functional fragments of a given protein. By "functional variant" or "functional fragment" of a given protein is meant a variant (for example a mutated protein) or a fragment thereof which retains the same biological function as that of said protein. When the polypeptide according to the invention comprises or consists of a particular domain of a protein, a fragment or variant of said polypeptide is said to be functional if it has the same biological function as that of said domain. For example, in the context of the present invention, a variant or fragment of a polypeptide comprising or consisting of the EBM domain of ErbB2 is said to be functional if it is capable of binding to the ERM domain of the ERM proteins. By "ErbB2 receptor inhibitor" or "ErbB2 inhibitor" is meant herein an active agent capable of inhibiting the ErbB2 receptor tyrosine kinase activity. In the context of the present invention, said inhibitor is preferably specific for the ErbB2 receptor, that is to say that it does not inhibit the tyrosine kinase activity of other receptors of the same family, such as ErbB1 receptors, ErbB3 and / or ErbB4. In addition to the screening method according to the invention, the inhibition of the tyrosine kinase activity can be measured using tests that are well known to those skilled in the art, such as the measurement of the tyrosine phosphorylation (s) of the ErbB2 intracellular domain (Graus-Porta et al., 1997; Dankort et al., 2011), for example using commercial kits such as the Mesoscale "PhosphoErbB2 (Tyr1248) Assay VVhole Eye Lysate" kit. By "homogeneous amplified luminescence proximity test" (more commonly referred to as "Amplified Luminescent Proximity Homogeneous Assay Screen") is meant here a non-radioactive test to study the interactions between biomolecules at a distance generally equal to or less than 200 nm. The term "homogeneous" means here that this test does not require a separation or washing step, unlike tests qualified as heterogeneous. This test is better known as AlphaScreen or AlphaLISA technology which is marketed by PerkinElmer (http://www.perkinelmer.com). The principle of this technology is based on the conjugation of two biomolecules to donor and acceptor particles, which, when these biomolecules bind to each other, lead to a transfer of energy, producing a luminescent signal (Arkin et al. al., 2012). More particularly, the photo-excitation of the donor particles leads to the formation of singlet oxygen which diffuses towards the acceptor particles, the latter emitting a luminescent signal when they are close to the donor particles. In the absence of binding between the biomolecules tested, and thus of bringing together the donor and acceptor particles, the singlet oxygen falls into a basal state and no luminescent signal can be emitted by the acceptor particles. Unlike FRET (Rirster type resonance energy transfer), the emission of the fluorescent signal by the acceptor particles takes place at a wavelength shorter than that allowing the excitation of the donor particles. By "donor particles" is meant here energy-giving particles, while the term "acceptor particles" qualifies here energy-accepting particles. Generally, the donor particles contain a "photosensitizer", i.e. a light-excitable agent that has the ability to convert ambient oxygen to singlet oxygen after illumination, while the acceptor particles contain an agent. "Chemiluminescent", that is to say an agent capable of emitting a luminescent signal in the presence of singlet oxygen.

Other definitions are given below in the description.

The inventors have developed a new and innovative screening method for identifying molecules useful for treating and / or preventing the pathologies associated with overactivation and / or overexpression of the ErbB2 receptor. This process is fast, sensitive, reliable and inexpensive, and can be performed in high throughput screening.

Thus, in a first aspect, the present invention provides an in vitro screening method for selecting an ErbB2 inhibitor, comprising the following steps: a) contact in a first medium of at least one candidate compound with: - a polypeptide comprising the Ezrin binding domain (EBM) of ErbB2, said polypeptide being conjugated to one or more donor particles, and - a polypeptide comprising the FERM domain, said polypeptide being conjugated to one or more acceptor particles, b) excitation of the medium of the step a) using an illumination stimulating the activation of the donor particles, and C) measuring the luminescent signal emitted by the acceptor particles following excitation b), a significant variation Al of the luminescent signal in the presence of said candidate compound with respect to the luminescent signal in the absence of said candidate compound being indicative of the inhibition of the binding of the FERM domain to the EBM domain of ErbB2 by the compound candidate ; d) selecting at least one candidate compound capable of inhibiting the binding of the FERM domain to the EBM domain of ErbB2 (said compound being identified after carrying out steps a) to c) described above), and contact in a second medium of said compound with: - a polypeptide comprising the EBM domain of a protein other than ErbB2, said polypeptide being conjugated to one or more donor particles, and - a polypeptide comprising the FERM domain, said polypeptide being conjugated to one or more acceptor particles e) exciting the medium of step b) using an illumination stimulating the activation of the donor particles, and f) measuring the luminescent signal emitted by the accepting particles following the excitation e), a variation A non-significant signal of the luminescent signal in the presence of said candidate compound being indicative of the ErbB2 inhibition capacity of said compound selected in step d). Preferably, said inhibition is a specific inhibition of the ErbB2 receptor, that is to say that the candidate compound inhibits the ErbB2 receptor tyrosine kinase activity, and not that of other receptors of the same family, such as ErbB1 receptors. , ErbB3 and / or ErbB4. At the end of step f), a luminescent signal is thus observed which may or may not reveal the binding of the FERM domain to the EBM domain of a protein other than ErbB2. With this information, it will be possible to conclude as to the inhibition capacity, preferably specific, of ErbB2 by the tested candidate compound. Preferably, said method further comprises a step g) of selecting the candidate compound as an ErbB2 inhibitor. This method thus has the particularity of discriminating the compounds inhibiting the interaction between ErbB2 and FERM by interacting with ErbB2, of those interacting with the FERM domain, thanks to the use of a counterscreening in the second selection step ( steps d to f, preferably d to g) which measures the interaction between the FERM domain and the ERM binding motif contained in the juxtamembrane region of proteins other than the ErbB2 receptor. Only those compounds capable of inhibiting the binding of the FERM domain to the EBM domain of ErbB2 are in fact subjected to a counter-screening according to steps d) to f), in order to identify among these those binding to the EBM domain of ErbB2 (and thus inhibiting ErbB2). The examples described below demonstrate that this new approach makes it possible to identify compounds capable of effectively and specifically inhibiting ErbB2. The inhibitory activity of the latter has indeed been validated in vitro in a breast cancer cell line overexpressing ErbB2. More precisely, by "variation 3.1" of the luminescent signal is meant here the variation of the luminescent signal evaluated at the end of step c), while the term "variation 32" of the luminescent signal designates the variation of the evaluated luminescent signal. at the end of step f).

In the context of the present invention, the variation 3.1 should be significant at the end of step c). By "significant variation" of the luminescent signal is meant here a total absence of luminescent signal or a significant decrease of the signal relative to the luminescent signal in the absence of the candidate compound. By "significant decrease" is meant here a decrease in the luminescence signal of at least 30%, at least 40%, more preferably at least 50%, at least 60%, and still more more preferably at least 70% or at least 80% with respect to the luminescent signal in the absence of said candidate compound. Outside of these conditions, the variation 3.1 is not significant, and it is therefore not carried out the counter-screening steps d) to f) (or d) to g)) of the process of the invention.

Furthermore, it is appropriate in the context of the present invention that the variation 32 is insignificant at the end of step f). By "non-significant variation" is meant here that the variation 32 is a fraction of the variation M, which does not represent more than one third of said variation M, preferably not more than one quarter of said variation M, and so more preferably not more than one fifth of said variation M. Those skilled in the art will understand that in order to objectively evaluate the variation 32 with respect to the variation M, an identical concentration of the test compound must be used in the screening step a) to c) and in the counter-screening step d) to f) or d) to g). Preferably, a non-significant variation of the luminescent signal may correspond to a variation of less than 10%, preferably less than 5% in the presence of the candidate compound relative to the luminescent signal in the absence of said candidate compound. Said variation may be zero with respect to the luminescent signal in the absence of said candidate compound (i.e. luminescent signal emitted in the presence of the candidate compound identical to that emitted in the absence of said compound). Outside of these conditions, the variation 32 is significant, and the tested candidate compound is not selected.

The measurement of the luminescent signal can be carried out using any suitable measuring device. These devices are well known to those skilled in the art, and include, without limitation, luminescence detectors such as the "New infinite M1000 PRO" apparatus marketed by TECAN. In the context of the present invention, ErbB2 is preferably of human origin. Thus, according to a preferred embodiment of the invention, the EBM domain of the ErbB2 receptor is of SEQ ID NO: 1 sequence. One skilled in the art will nevertheless be able to identify the amino acid sequence of the EBM domain of the ErbB2 receptor of other animal species, for example by sequence homology. In the context of the present invention, the FERM domain is preferably of human origin, and / or is preferably selected from the group consisting of the FERM domain of Ezrin protein, Moesin, Radixin and Merlin. More preferably, the FERM domain is the FERM domain of the Ezrin protein. Thus, according to a preferred embodiment of the invention, the FERM domain is of sequence SEQ ID NO: 2. One skilled in the art will readily understand that the FERM domain can be used here in whole or in part provided that either retained at least the EBM domain binding subdomain C. Those skilled in the art will also be able to identify the amino acid sequence of the FERM domain of other animal species, for example by sequence homology. The proteins other than ErbB2 used in the counter-screening step of the method of the invention are proteins containing an EBM domain in their juxtamembrane region. These proteins are well known to those skilled in the art and include adhesion molecules such as the CD44 (differentiation cluster 44) and sialomucin CD43 (differentiation cluster), or the the superfamily of immunoglobulins such as ICAM-1, -2 and -3. Thus, according to a preferred embodiment of the invention, the EBM domain of a protein other than ErbB2 that is included in the polypeptide used in step d) of the method is selected from the group consisting of CD44 EBM domains. , CD43, ICAM-1, ICAM-2 and ICAM-3. Even more preferably, said EBM domain is the EBM domain of CD44. In the context of the present invention, this EBM domain is preferably of human origin. Thus, according to a particularly preferred embodiment of the invention, the EBM domain of human CD44 is of SEQ ID NO: 3 sequence. The person skilled in the art will nevertheless be able to identify the amino acid sequence of the domain FERM of other animal species, for example by sequence homology. The screening method of the invention is based on energy transfer between donor and acceptor particles that are conjugated to biomolecules (i.e., to a polypeptide comprising an EBM domain and to a polypeptide comprising an FERM domain, respectively). In a preferred manner, the screening method according to the invention is a homogeneous amplified luminescence proximity test. Thus, according to a preferred embodiment of the invention, said donor particles comprise at least one photosensitizer and said acceptor particles comprise at least one chemiluminescent agent, said agent generating the luminescent signal when the photosensitizer converts the ambient oxygen to singlet oxygen following excitation with said illumination and that the donor and acceptor particles are close to each other. According to a preferred embodiment of the invention, said photosensitizer is selected from the group consisting of acetone, endoperoxide, benzophenone, chlorophylls, eosin, erythrosine, fullerenes, methylene blue porphyrins such as hematoporphyrin, phthalocyanines such as phthalocyanines Al3 +, phthalocyanines Cu2 + or phthalocyanines Zn2 +, rose Bengal, 9-thioxanthone, and any derivative thereof. These compounds as well as other examples of photosensitizers which can be used in the screening method according to the invention are described in particular in US patent documents 5340716, US 5516636, US 5536834, US 5709994, US 5763602, US 6251581, US 6406667, US 6797481, US 7033775, and US 7101682, as well as Ullman et al. (1994) and DeRosa et al. (2002). The particularly preferred photosensitizer according to the invention is phthalocyanine, preferably complexed to a metal such as Al3 + Cu2 + or Zn2 +.

According to a also preferred embodiment of the invention, said chemiluminescent agent is selected from the group consisting of thioxene, anthracene (eg 9,10-bis (phenylethynyl) anthracene, 9,10-diphenylanthracene), rubrene, europium, terbium, and any derivative thereof. These compounds as well as other examples of chemiluminescent agents that can be used in the screening method according to the invention are described in particular in the documents of US Pat. Nos. 5,340,716, 5,156,616, US 5,356,834, US Pat. US 6406667, US 6797481, US 7033775, and US 7101682, as well as Ullman et al. (1994).

The chemiluminescent agents which are particularly preferred according to the invention are thioxene, anthracene and rubrene. As indicated above, a signal luminescent signal is generated when the donor particles and the accepting particles are in close proximity to each other. This approximation of said particles takes place when the biomolecules with which they are conjugated interact with each other. More particularly, said donor and acceptor particles are said to be close if the distance between them is less than or equal to that traveled by the singlet oxygen in said medium. Preferably, said donor and acceptor particles are said to be close if the distance between them is less than or equal to about 200 nm. In the context of the present invention, the term "about" means that a variation of plus or minus 5% of a given value can be tolerated. Illumination for exciting the donor particles can be performed using any suitable light source, such as a laser. According to a preferred embodiment of the invention, the illumination is performed at a wavelength greater than that of the measurement of the luminescent signal. It is understood that the wavelength of this illumination, and that at which the luminescent signal is emitted and therefore must be measured, vary according to the nature of the photosensitizer and the chemiluminescent agent. These wavelengths can be easily determined by those skilled in the art.

Thus, according to a preferred embodiment of the invention, when the photosensitizer is phthalocyanine, illumination is performed at a wavelength of about 680 nm. According to a also preferred embodiment of the invention, when the final chemiluminescent agent is rubrene, the measurement of the luminescent signal is carried out at a wavelength of between about 520 and 620 nm. The donor and acceptor particles used herein are preferably polymer beads, such as polystyrene, polyacrilamide, polyvinyl chloride, polyvinylnaphthalene and / or polymethacrylate beads. In a particularly preferred manner, said particles are latex beads. In addition, in order to prevent these particles from settling in the medium, the diameter of said particles is preferably less than or equal to about 250 nm. Particles can also be covered with a hydrogel to minimize nonspecific interactions and self-aggregation. As indicated above, the polypeptides according to the invention are "conjugated" to said particles, i.e. on the surface of said particles. By "conjugate" or "conjugation" is meant here a covalent or non-covalent bond. The methods making it possible to carry out these types of connection are well known to those skilled in the art and therefore do not need to be detailed further. Non-covalent bonds are particularly preferred according to the invention, and include, without limitation, streptavidin or avidin-biotin binding, histidine-metal ion binding (nickel, cobalt), and hydrogen bonding between glutathione S-transferase and GSH.

It is understood that the screening method according to the invention is carried out under experimental conditions allowing interaction between the EBM and FERM domains, in order to be able to test the ability of a candidate compound to inhibit ErbB2. The medium used for this purpose is generally an aqueous medium. The pH of said medium is in turn between pH5 and pH10, and preferably between pH6 and pH8. A buffer may also be used in said medium, such as a borate buffer, cacodylate, carbonate, citrate, HEPES, MES, MOPS, phosphate, PIPES, TAPS, TES or Tris. In another aspect, the present invention relates to an ErbB2 inhibitor capable of being selected by the screening method according to the invention as described above. Preferably, said inhibitor ErbB2 is isoliquiritigenin. Another aspect of the present application relates to a process for the preparation of a pharmaceutical composition for the prevention and / or treatment of pathologies associated with overactivation and / or overexpression of ErbB2, comprising the steps of: a) selection of at least one ErbB2 inhibitor according to the screening method as described above; and b) adding a pharmaceutically acceptable carrier to said inhibitor.

Preferably, said inhibitor ErbB2 is isoliquiritigenin. By "pathologies associated with overactivation and / or overexpression of ErbB2" is meant here any pathology resulting from an increase in the activation and / or overexpression of ErbB2. We talk about overactivation of the ErbB2 receptor when the activity of the latter is significantly higher in cells of a patient compared to healthy cells from the same type of tissue. This overactivation may result from overexpression of the ErbB2 receptor, a mutation of the ErbB2 receptor, its cleavage or heterodimerization with other members of the ErbB receptor family, and can be determined with the aid of tests well known to those skilled in the art, such as the measurement of the tyrosine phosphorylation (s) of the ErbB2 intracellular domain (Graus-Porta et al., 1997, Dankort et al., 2001), for example using o commercial kits such as Mesoscale's "Phospho-ErbB2 (Tyr1248) Assay VVhole Cell Lysate" kit. We talk about overexpression of the ErbB2 receptor when the latter is expressed at a significantly higher level on the surface of cells of a patient compared to healthy cells from the same type of tissue. This overexpression may result from an abnormal increase in the amplification of the gene coding for this receptor and / or an increase in the transcription and / or translation of ErbB2 or even a mutation, and may be determined by tests well known to those skilled in the art for measuring the expression of proteins, in particular with the aid of specific antibodies directed against the ErbB2 receptor, such as immunohistochemistry, immunoprecipitation, Western blotting, immunoblot, bound enzyme immunoadsorption assay (ELISA) and its variants (ELISPOT), radioimmunoassay, flow cytometry, but also using assays evaluating the expression of the gene encoding ErbB2. such as fluorescence in situ hybridization (FISH), or polymerase chain reaction (PCR, RT-PCR). For example, WO91 / 05264 discloses measuring the expression of the ErbB2 receptor from a biological fluid of a patient. The pathologies associated with overactivation and / or overexpression of ErbB2 are well known to those skilled in the art and include, among others, benign or malignant tumors such as cancer, including in particular breast cancer, cancer. cellular squamous cell cancer, small cell lung cancer, non-small cell lung cancer, gastrointestinal cancer, pancreatic cancer, glioblastoma, cervical cancer, ovarian cancer, liver cancer, bladder cancer, hepatoma, colon cancer, colorectal cancer, carcinoma of the endometrium, salivary gland carcinoma, kidney cancer, liver cancer, cancer of the liver prostate, vulvar cancer, thyroid cancer, hepatic carcinoma and various types of head and neck cancer. By the term "prevention" it is necessary to understand here the means of preventing the pathology from declaring itself or of slowing down its appearance, while by the term "treatment" is meant here the means of treating a declared pathology or of to reduce its symptoms and / or progression. By "pharmaceutically acceptable carrier" is meant here an inactive or inert, non-toxic and, therefore, devoid of pharmacological action, which can be used to improve the properties of a pharmaceutical composition, such as its shelf life or his administration. Examples of such carriers include, inter alia, surfactants (cationic, anionic or neutral), surface stabilizers, preservatives, wetting or emulsifying agents, solvents, buffers, saline solutions, dispersion media, isotonic and absorption retarding agents, and their analogs, which are physiologically compatible. Such carriers are described in the "Remington's Pharmaceutical Sciences" manual (22nd edition, edited by Allen, Loyd V., Jr), to which a person skilled in the art may refer. The present invention also relates, in another aspect, to a pharmaceutical composition comprising at least: a) an ErbB2 inhibitor capable of being selected according to the screening method as described above; and b) a pharmaceutically acceptable carrier as described above. Said composition is particularly useful for preventing and / or treating pathologies associated with overactivation and / or overexpression of ErbB2 as set forth above.

Preferably, said inhibitor ErbB2 is isoliquiritigenin. Thus, in another aspect, the invention relates to a method for preventing and / or treating a condition associated with overactivation and / or overexpression of ErbB2, comprising administering to a subject a pharmaceutically effective amount. effective of the pharmaceutical composition according to the invention. Preferably, said composition comprises isoliquiritigenin as an ErbB2 inhibitor capable of being selected according to the screening method as described above. In other words, the invention relates to a pharmaceutical composition as described above, for its use in the prevention and / or treatment of a pathology associated with overactivation and / or overexpression of ErbB2. More specifically, the invention relates to the use of the pharmaceutical composition according to the invention for the manufacture of a medicament for preventing and / or treating a pathology associated with overactivation and / or overexpression of ErbB2. The invention also relates to a method for preventing and / or treating a condition associated with overactivation and / or overexpression of ErbB2, comprising administering to a subject a pharmaceutically effective amount of isoliquiritigenin. In other words, the invention relates to isoliquiritigenin for its use in the prevention and / or treatment of a condition associated with overactivation and / or overexpression of ErbB2. More specifically, the invention relates to the use of isoliquiritigenin for the manufacture of a medicament for preventing and / or treating a pathology associated with overactivation and / or overexpression of ErbB2. By "subject" is meant to describe here any member of the animal kingdom, and preferably the man. The term "administration" or "administering" herein means that a compound of interest is delivered or dispensed to a subject by any suitable mode of administration, which may be readily determined by one skilled in the art depending on the nature of the subject. said compound. For example, said compound may be delivered or dispensed to said subject as appropriate orally, transdermally, or parenterally such as by subcutaneous, intravenous, intramuscular, or intraperitoneal injection. In the context of the present invention, the term "pharmaceutically effective amount" is intended here to mean a quantity or a prophylactic or therapeutic concentration of a compound of interest, that is to say a quantity or concentration of said compound sufficient to prevent the pathology related to overactivation and / or overexpression to declare or delay its onset, or to treat said pathology once reported or to mitigate its symptoms and / or progression. Those skilled in the art are able to determine this so-called pharmaceutically effective amount. In order to perform the screening according to the invention, and thus to select an ErbB2 inhibitor that can be used to prevent and / or treat pathologies associated with the overactivation and / or overexpression of said receptor, it is particularly useful to have a kit with the necessary components. Thus, the invention relates in another aspect to a kit for use in any of the methods described above, comprising: a) a polypeptide comprising the EBM domain of ErbB2; b) a polypeptide comprising the FERM domain; and optionally, a polypeptide comprising the EBM domain of a protein other than ErbB2. The preferred embodiments are as described above. In particular, the EBM domain of a protein other than ErbB2 that is included in polypeptide c) is preferably selected from the group consisting of EBM domains of CD44, CD43, ICAM-1, ICAM-2 and ICAM-3, and even more preferably is the EBM domain of CD44. It is understood that said polypeptides may be labeled with a marker allowing its conjugation to a solid support, such as donor or accepting particles. Preferably, said kit further comprises: d) donor particles capable of being conjugated to said polypeptide a); e) acceptor particles capable of being conjugated to said polypeptide b); and f) optionally, acceptor particles capable of being conjugated to said polypeptide c). Optionally, said kit may further include instructions for performing said method.

The present invention will be better understood in light of the examples below. DESCRIPTION OF THE DRAWINGS Figure 1. Development of the Alpha-screen test: robustness and validation of its specificity. (A) Principle of the Alphascreen measurement of the interaction between the HER2 juxtamembrane region (Biot-EBM) and the Ezrin FERM domain (GST-FERM). (B) Sequence of the biotinylated polypeptide corresponding to the wild-type HER2 juxtamembrane region (HER2 VVT), comprising the EBM domain of ErbB2 and the biotinylated polypeptide corresponding to the juxtamembrane region of HER2 carrying mutations on the critical residues involved in the interaction with the FERM domain of Ezrin (HER2 MUT). (C) Quantification of the signals measured in Alphascreen in absence (Bic) or in the presence of the biotinylated polypeptide corresponding to the juxtamembrane region of HER2 and the Ezrin FERM domain (Pept / FERM) making it possible to calculate a robustness score (Z ' = 0.8). (D) Quantification of the competition efficiency of the HER2 / FERM interaction by the non-biotinylated polypeptide of the wild-type HER2 juxtamembrane region (HER2 VVT) or mutated on the critical residues involved in the interaction with the domain FERM of Ezrin (HER2 MUT). Figure 2. Development of counterscreening: validation of its specificity. (A) Principle of the Alphascreen measurement of the interaction between the Ezrin FERM domain (GST-FERM) and the known ERM binding motif contained in the juxtamembrane region of CD44 (Biot-CD44). (B) Sequence of the biotinylated polypeptide corresponding to the juxtamembrane region of CD44 containing the known ERM binding motif. (C) Quantification of the competition efficiency of the CD44 / FERM interaction by the non-biotinylated polypeptide of the CD44 juxtamembrane region containing the known ERM binding motif. Figure 3. Procedures of the high-throughput screening according to the invention.

Figure 4. Alphascreen characterization of the hit 8 identified in the phytochemical library: confirmation and against screening. (A) Alphascreen confirmation of inhibition of HER2 / FERM interaction in the presence of increasing concentrations of compound 8 for 1h (FERM_EBM_1h). (B) Alphascreen confirmation of inhibition of HER2 / FERM interaction in the presence of increasing concentrations of compound 8 for 20 h (FERM_EBM_20h). (C) Result of the cross-screen showing quantification of inhibition of CD44 / FERM interaction in the presence of increasing concentrations of compound 8 for 1h (FERM-CD44 1h). (D) Result of the cross-screen showing quantification of the inhibition of the CD44 / FERM interaction in the presence of increasing concentrations of Compound 8 for 20h (FERM-CD44 20h). Figure 5. Confirmation of hit 8: interaction with ErbB2 and inhibition of its phosphorylation. (A) Surface Plasmonic Resonance Measurement of the Interactions Between Compound 8 and the Wild-type ErbB2 Juvam Region Polypeptide ('/ VT ErbB2) or Mutated on Critical Residues Involved in Interaction with the FERM Domain of Ezrin (Mut ErbB2) or a non-relevant peptide (CD147). (B) Surface Plasmon Resonance Measurement of Interactions Between Ezrin Compound 8 and Ezrin Domain (GST-FERM) or GST Only (GST). (C) Dot blot analysis of tyrosine phosphorylation inhibition of ErbB2 in SKBR3 cells treated with increasing concentrations of Compound 8 (0.625-80 μM) for 24 h with anti-phospho antibody total tyrosine (4G10). (D) Quantification of the signals represented in C. (E) Dot blot analysis of ErbB2 tyrosine phosphorylation inhibition in SKBR3 cells treated with increasing concentrations of Compound 8 (0.625-80 μM) for 48 h using a total anti-phospho-tyrosine antibody (4G10). (F) Quantification of the signals represented in E. Figure 6. Confirmation of hit 8: inhibition of ErbB2-dependent cell proliferation. (A) Quantification of the proliferation of untreated SKBR3 cells (SKBR3) or treated for three days with compound 3.2 at 3.2 μM (SKBR3_8). (B) Quantification of the proliferation of untreated HBMEC cells (H BM EC) or treated for three days with Compound 8 at 3.2 μM (HBMEC_8). (C) Quantification of the proliferation of MCF7 cells stably transfected with a vector allowing expression of untreated HER2 (HER2) or treated for three days with Compound 8 at 5 μM (HER2_8_5pM). (D) Quantification of the proliferation of MCF7 cells stably transfected with an empty vector untreated (pIRES) or treated for three days with Compound 8 at 5 μM (pl RES_8_5pM). (E) Quantification of the proliferation of untreated SKBR3 cells (SKBR3) or treated for three days with Compound 8 at 20 μM (SKBR3_8). (F) Quantification of the proliferation of untreated HBMEC cells (HBMEC) or treated for three days with Compound 8 at 20 μM (HBMEC_8). The significance (* P <0.05 and ** P <0.01) was evaluated by statistical tests "student T test". Figure 7. Alphascreen characterization of the hit T identified in the chemical compound bank: confirmation and against sieving. (A) Alphascreen confirmation of inhibition of HER2 / FERM interaction in the presence of increasing concentrations of compound T for 1h (FERM EBM_1h). (B) Alphascreen confirmation of the inhibition of the HER2 / FERM interaction in the presence of increasing concentrations of the compound T during 20h (FERM_EBM_20h). (C) Result of the cross-screen showing the quantification of the inhibition of the CD44 / FERM interaction in the presence of increasing concentrations of the compound T for 1h (FERM-CD44 1h). (D) Result of the cross-screen showing the quantification of the inhibition of the CD44 / FERM interaction in the presence of increasing concentrations of compound T for 20 h (FERM-CD44 20h). Figure 8. Complementary experiments on hit T: interaction with ErbB2 and inhibition of its phosphorylation. (A) Surface Plasmon Resonance Measurement of Interactions Between the T-Compound and the Wild-type ErbB2 Juxtamembrane Region (VVT ErbB2) Polypeptide or Mutated on Critical Residues Involved in Interaction with the Ezrin FERM Domain (Mut ErbB2) or a non-relevant peptide (CD147). (B) Surface Plasmon Resonance Measurement of Interactions Between Ezrin Compound T and Ezrin Domain (GST-FERM) or GST Only (GST). (C) Dot Blot Analysis of Inhibition of ErbB2 Tyrosine Phosphorylation in SKBR3 Cells Treated with Increasing Concentrations of T Compound (0.625-80 μM) for 24 Hours Using a Total Anti-Phosphor Tyrosine Antibody (4G10) ). (D) Quantification of the signals represented in C. (E) Dot blot analysis of ErbB2 tyrosine phosphorylation inhibition in SKBR3 cells treated with increasing concentrations of compound T (0.625-80 μM) for 48 hours with using total anti-phospho-tyrosine antibody (4G10). (F) Quantification of the signals represented in E.

Figure 9. Complementary experiments on hit T: inhibition of ErbB2-dependent cell proliferation. (A) Quantification of the proliferation of untreated SKBR3 cells (SKBR3) or treated for three days with the 3.2 mM T compound (SKBR3_T). (B) Quantification of the proliferation of untreated HBMEC cells (HBMEC) or treated for three days with the 3.2 mM T compound (HBMEC_T). (C) Quantification of the proliferation of HBMEC cells transfected with a vector allowing the expression of HER2 (HBMEC + HER2) untreated (CTRL) or treated for three days with the compound T at 2.5 μM (T). (D) Quantification of the proliferation of HBMEC cells transfected with an empty vector (H BM EC) untreated (CTRL) or treated for three days with the compound T at 2.5 μM (T). (E) Quantification of the proliferation of untreated SKBR3 cells (SKBR3) or treated for three days with compound 8pM T (SKBR3_T). (F) Quantification of the proliferation of untreated HBMEC cells (HBMEC) or treated for three days with compound T at 8 μM (HBMEC_T). The significance (* P <0.05, ** P <0.01, *** P <0.001) was evaluated by statistical tests "student T test". EXAMPLES 1. MATERIAL AND METHODS 1.1. AlphaScreene ("Amplified Luminescent Proximity Homogenous Assay" or Homogeneous Amplified Luminescence Proximity Test) 1.1.1. Reagents AlphaScreen® technology was used to measure the interaction between the Ezrin FERM domain and the Ezrin binding motif (EBM) contained in the juxtamembrane regions of ErbB2 or CD44. The FERM domain of Ezrin in fusion with glutathione-S-transferase (GST) was coupled to GSH coated acceptor beads. GST alone was used as a negative control. A biotinylated ErbB2 peptide corresponding to the binding motif of Ezrin (amino acids 674 to 689: biotin-ILIKRRQQKIRKYTMRRL (SEQ ID NO: 1), 26 amino acids) was synthesized, as well as a control peptide where the residues The keys involved in the interaction with the FERM domain were mutated (biotin-ILIGGGQQKIGKATMRRL (SEQ ID NO: 4)). These peptides were coupled to donor beads coated with streptavidin. For counterscreening, the biotinylated peptide corresponding to the juxtamembrane region of CD44 (biotin-NSRRRCGQKKKLVINSG (SEQ ID NO: 3) was synthesized and coupled to donor-streptavidin beads.The reagents for AlphaScreen® (acceptor glutathione and donor-streptavidin beads) from PerkinElmer 1.1.2 Measurement of interaction The interaction was carried out in 384 well Optiplate plates (PerkinElmer, VVhalham, MA, USA) in a total reaction volume of 20 μl. GST-FERM (or GST) and biotinylated EBM peptide were prepared in Alphascreen® reaction buffer, PBS, pH 7.4, containing 5 mM MgCl2 and 0.02% CHAPS. FERM (0 to 1 μM) was incubated with biotinylated EBM peptide (0 to 1 μM) for 30 minutes at room temperature The donor (20 μg / ml) and acceptor (20 μg / ml) beads were then added 40 minutes or overnight in the dark and at room temperature 1 .1.3 Screening Banks Banks of phytochemical and chemical molecules are marketed by Prestwick Chemical. For high throughput screening, dilution of the compounds at 100 μM in 1% DMSO was performed by an automated workstation (Automation VVorkstation, PerkinElmer). 2.5 μl of the diluted compounds were automatically transferred to the 384-well Optiplate plates containing 2.5 μl of buffer. 5 μl of a solution containing 0.625 μM GST-FERM and 20 nM biotin-EBM was added to the compounds for 30 minutes at room temperature. 10 μl of a solution containing 20 μg / ml acceptor-glutathione beads and 20 μg / ml of streptavidin-donated beads were then added to the wells for 40 minutes or overnight in the dark and at room temperature. To eliminate possible non-specific interference of the compounds with the fluorescent signal, they were incubated with a mixture of 20 μg / ml of donor-streptavidin beads and acceptor-biotin beads. Light signals were read with the EnVision® Plate Reader (PerkinElmer). All experiments involving AlphaScreen® beads were performed in the dark. 1.2. Surface Plasmon Resonance (SPR) interaction test The interaction of the active compounds with ErbB2 selected by the AlphaScreen® screening was measured on a Biacore T200 (IECB, Bordeaux). Briefly, the biotinylated peptide corresponding to the juxtamembrane region of ErbB2 (biotin-ILIKRRQQKIRKYTMRRL (SEQ ID NO: 1)) was immobilized on streptavidin-coated chips (Series S sensor chip SA, GE Healthcare). As negative controls, the ErbB2 peptide mutated on residues essential for the binding of ERMs (biotinILIGGGQQKIGKATMRRL (SEQ ID NO: 4)) and the non-relevant biotinylated peptide corresponding to the cytosolic sequence of CD147 (biotin-AYDKRRKPEDVLDDDDAGSAPLKSSGQHQNDKGKNVRQRNSS (SEQ ID NO : 5), 42 amino acids) were used. The active compounds, diluted in PBS buffer, 0.02% Tween 20 were used as analytes, with the Ezrin FERM domain as a positive control. In addition, the compounds were tested for their ability to interact with the FERM domain. For this purpose, the GST-FERM (or GST) was captured by an anti-GST antibody immobilized on the chips (Series S sensor chip CM5, GE Healthcare). Active compounds in PBS buffer, 0.02% Tween 20 were used as analytes, with ErbB2 peptide as positive control. This approach allowed to characterize in detail (kinetics and specificity) the molecular interaction between these active compounds of low molecular weight and ErbB2. 1.3. Dot blots The total phosphotyrosine signal in SKBR3, an ErbB2 overexpressing breast cancer cell line, was analyzed by dot blots with antiphosphotyrosine antibody (clone 4G10). In these cells, ErbB2 is the tyrosine-phosphorylated major protein, so the decrease in the total phosphotyrosine signal reflects the inhibition of ErbB2 activity. The quantifications shown correspond to the results of two independent experiments. 1.4. Proliferation assays ErbB2 overexpressing SKBR3 cells, MCF-7 cells, an ErbB2-independent breast cancer cell line stably transfected with an empty vector (MCF-7 / pIRES) or with an ErbB2 coding vector (MCF-7 / ERBB2), or HBMEC cells transfected or not with a vector encoding ErbB2 were treated with the compounds before determining cell proliferation every day for 3 days using an MTT test (3- (4,5-dimethylthiazole) bromide). 2-yl) -2,5-diphenyl tetrazolium).

AG1478, a non-specific inhibitor of ErbB2 kinase activity, was used as a positive control and untreated cells serve as a negative control. 2. RESULTS AND DISCUSSION Breast cancer is the leading cause of cancer death among women worldwide. In 2008, this cancer accounted for 23% (1.38 million) of all new cancer cases and 14% (458,400) of cancer-related deaths (J Emal et al., 2011). It is estimated that almost half of all breast cancer cases and 60% of the associated deaths are from developing countries. In Europe, France belongs to countries where the incidence of this type of cancer is the highest and continues to increase. Twenty to 30% of primary breast cancers in women are due to the dysregulated expression of ErbB2, which represents approximately 8,000 patients a year in France and approximately 450,000 patients a year worldwide. Only 25% of patients respond to current therapies, not effective on mutated or truncated forms of this receptor or with high toxicity because of their lack of selectivity. There is therefore an urgent need to develop alternative approaches to specifically target ErbB2, reduce the risk of toxicity and effectively act on mutated or truncated forms of ErbB2. The approach developed here makes it possible to identify attractive molecules that meet these requirements and that have a clinical benefit for these patients. It has indeed been discovered that the interaction between the ERM domain of the proteins of the ERM family stabilizes ErbB2 in a catalytically inactive state by exerting a molecular stress on its juxtamembrane region, thus preventing access of the tyrosine substrates to the kinase domain. The purpose of this approach is to identify small molecules mimicking the FERM domain to inhibit ErbB2.

To do this, an Alphascreen test, robust and reproducible and which is also suitable for high throughput screens (HTS) has been developed. This test is based on the destabilization of the interaction between the juxtamembrane region of ErbB2 (EBM) and the FERM domain of Ezrin (Figure 1). Produced in 384-well microplates, this test is suitable for screen automation. After optimization, the test is robust, having a Z 'of 0.8 (this dimensionless score is used to evaluate the reproducibility and the quality of a test), indicating that it makes it possible to discriminate well the active compounds of the inactive compounds. This interaction is effectively inhibited by the ErbB2 soluble peptide corresponding to the ERM binding motif (SEQ ID NO: 1) and this inhibition is much less effective in the presence of an ErbB2 peptide mutated on the key residues involved in the interaction with the FERM domain (SEQ ID NO: 4).

To discriminate between compounds that inhibit ErbB2 / FERM interaction by interacting with ErbB2 from those interacting with the FERM domain, a counter screen (or counterscreen) in Alpha screen to measure the interaction between the FERM domain and the motif. known ERM binding contained in the juxtamembrane region of CD44 (SEQ ID NO: 3) (Figure 2). This interaction is efficiently inhibited by the soluble peptide of CD44. After the development of these tests, a pilot screen was conducted with pharmacologically active chemical compounds from banks marketed by Prestwick Chemical on a platform dedicated to Alphascreen (BMIScreen, Bordeaux, France) (Figure 3). These banks contain 1,200 FDA-approved bioactive compounds and 320 phytochemicals, all selected on the basis of their high chemical and pharmacological diversity and knowledge of their bioavailability and toxicity in humans. The use of molecules whose use in humans has already been approved makes it possible to reposition them rapidly and at a lower cost by reducing the time necessary for their development, as well as the risks since clinical trials have already been carried out and their toxicological and pharmacokinetic profiles are known. These compounds were first tested for their ability to inhibit the FERM / ErbB2 interaction in AlphaScreen®. In parallel, the compounds were tested for possible interference with the fluorescence signals of the assay in a false positive screen (donor and acceptor beads alone). To validate the first hits, their effectiveness in inhibiting the FERM / ErbB2 interaction was analyzed by performing dose effects in AlphaScreen® and their IC50 was determined. The compounds thus validated were then tested in a cross-screen based on the interaction between the FERM domain and the CD44 peptide to determine the selectivity of their inhibitory effect. This method allowed good discrimination between active compounds and inactive compounds, between compounds that specifically compete with the ErbB2 / FERM interaction and those that inhibit both ErbB2 / FERM and CD44 / FERM interactions. Of the 1500 molecules tested, two compounds were identified: a phytochemical compound (8) and a chemical compound (T) that statistically significantly inhibited the FERM / ERbB2 interaction (> 40%) and showed little or no interference with the technique in the false positive screen (Table 1). Table 1. Compounds identified by the pilot screen Hit NOM CRIBLE Non-specific interference with the signal Alpha screen Inhibition max,% Inhibition max,% 1h 20h PHYTOHEMICAL BANK 8 lsoliquiritigenin 37.66 56.83 4.04 CHEMICAL BANK T Alexidine dihydrochloride 49.36 18.55 14.61 Validation and characterization of the compound To validate compound 8, its effectiveness in inhibiting the FERM / ErbB2 interaction was analyzed by performing dose effects in AlphaScreen® to determine its 1050.

Table 2. Characteristics of Compound 8 Inhibition max,% IC50 pM Inhibition max, IC50 pM, Hit NOM% FERM-CD44 1h 20h 1h 20h 1h 20h 1h 20h 43, 8 lsoliquiritigenin 33.3 9 20.3 4 14.8 0 N / DN These results demonstrate that compound 8 inhibits the FERM / ERbB2 interaction (43.9%) at 24 hours in a dose-dependent manner (IC50 = 4 μM). Moreover, this compound does not interfere with the FERM / CD44 interaction and therefore represents a very attractive hit (Figure 4). Figure 5 shows the validation of compound 8: the direct interaction between compound 8 and ErbB2 or the FERM domain was tested by Plasmonic Surface Resonance experiments using a Biacore T200.

Compound 8 interacts specifically with ErbB2 without any interaction with the mutant ErbB2 peptide or the non-relevant peptide (CD147). Moreover, only a weak interaction of compound 8 with the FERM domain is detected (FIGS. 5A and 5B). These results confirm the direct interaction between compound 8 and ErbB2; the ability of compound 8 to inhibit the activation of ERbB2 in cellulo was then assayed on SKBR3, an ErbB2 overexpressing breast cancer cell line frequently used as a model for studying these cancers. The SKBR3 were treated 24h or 48h with increasing concentrations of compound 8 (0.625-80 μM) and the phosphorylation of ErbB2 was analyzed in "dot blot". Compound 8 decreases ErbB2 phosphorylation at 24h and 48h in a dose-dependent manner (Figures 50 to 5F). To further characterize the effects of compound 8, its ability to inhibit cell proliferation due to overexpression of ERbB2 was analyzed. SKBR3 cells were used for this purpose, as well as normal human endothelial cells (HBMEC) as a control to eliminate the risk of non-specific toxic effects. Treatment with Compound 8 at 3.2 μM weakly decreases proliferation of SKBR3 cells at days 1 and 2, and at 20 μM, this inhibition becomes significant with no effect on the proliferation of HBMECs. The selective effects of compound 8 were then confirmed using MCF7 cells, ErbB2-negative breast cancer cells stably transfected with an ErbB2 expression vector (MCF-HER2) or an empty vector (MCF7 pIRES). Treatment with Compound 8 at 5 μM significantly inhibited the proliferation of MCF7-ErbB2 cells at days 2 and 3, without affecting that of the control MCF7 cells (Figure 6).

Compound 8 from the phytochemical library has therefore been validated for its direct interaction with the juxtamembrane domain of ErbB2 as well as for its ability to inhibit ErbB2 phosphorylation (and therefore activation) and ErbB2-cell proliferation. dependent. This compound specifically inhibits the ErbB2 / FERM interaction (negative screen), which results in a specificity of ErbB2 inhibition and an absence of toxicity at high concentrations in the control cells. Validation and characterization of the compound T To validate the compound T, its effectiveness in inhibiting the FERM / ERbB2 interaction was evaluated by performing dose effects in AlphaScreen®. Compound T inhibits the FERM / ERbB2 interaction by 64.7% at 1 hour in a dose-dependent manner (1050 = 6.2 μM) without effect at 24 hours. Table 3. Characteristics of the compound T Hit NOM Inhibition max, IC50 pM, FERM-EBM Inhibition max,% IC50 pM, FERMCD44% lh 20h lh 20h lh 20h lh 20h T Alexidine 64.7 41.2 6.2 157 66.8 28, However, this compound also interferes drastically with the FERM / CD44 interaction in a dose-dependent manner (Inhibition> 60%) suggesting a low specificity of this compound for ErbB2 (Figure 7). Additional experiments on the compound, shown in Figure 8: the direct interaction of the compound T with ErbB2 or the field FERM was tested in SPR. The compound T interacts very strongly with the peptide corresponding to the juxtamembrane region of ErbB2 whereas little or no interaction is detected with the ErbB2 mutant peptide or the non-relevant peptide. Moreover, no interaction of the compound T with the FERM domain is detected. These results confirm the direct interaction between the compound T and ErbB 2 (FIGS. 8A and 8B); the ability of compound T to inhibit ErbB2 activation was then analyzed in SKBR3 cells. Compound T strongly decreases the phosphorylation of ErbB2 at 24 hours and 48 hours in a dose-dependent manner (FIGS. 8C to 8F). To further characterize the effects of the T compound, its ability to inhibit cell proliferation due to overexpression of ErbB2 was analyzed. Treatment with the 3.2 mM T compound significantly decreased the proliferation of SKBR3 cells from day 1 to day 3 without any effect on the proliferation of HBMECs. Furthermore, treatment with the 2.5 μM T compound significantly inhibited the proliferation of ErbB2 overexpressing HBMEC cells on days 2 and 3 without affecting that of HBMEC cells that do not overexpress ErbB2. On the other hand, at doses greater than or equal to 5 μM, nonspecific toxic effects are observed on HBMECs (FIG. 9). The compound T from the first screen was therefore validated for its direct interaction with ErbB2 as well as for its ability to inhibit ErbB2 activation and ErbB2-dependent cell proliferation. This compound, however, exhibits toxicity at moderate concentrations in the control cells. 3. CONCLUSION It has previously been demonstrated that targeting the juxtamembrane domain of ErbB2 by the Ezrin FERM domain selectively inhibits activation of ErbB2 by a mechanism different from that of Trastuzumab and Lapatinib and effectively blocks activation. various oncogenic forms of ErbB2, including the highly metastatic p95ErbB2 form that is insensitive to current targeted therapies. In addition, the FERM domain interacts with ErbB2, but not with other members of the ErbB family, because the consensus pattern allowing this interaction is not present in other members of the ErbB family. Thus, compounds targeting the juxtamembrane domain of ErbB2 should not exhibit toxicity due to non-specific inhibition of ErbB and non-ErbB kinases. A compound mimicking the effects of the FERM domain should therefore make it possible to overcome the main difficulties encountered to date with existing therapies. The screening method according to the invention has made it possible to identify those types of compounds which: interact with the juxtamembrane region of ErbB2; inhibit the phosphorylation of ErbB2; and - inhibit ErbB2-dependent cell proliferation. It should be noted that these compounds have already been approved for use in humans, offering the possibility of therapeutic repositioning, quickly and cost-effectively. Thanks to the counter-screening of the method of the invention, it is also possible to distinguish the compounds having a selective effect on the inhibition of ErbB2. Thus, this innovative screening strategy is adapted to the identification of new molecules that inhibit ErbB2 activity in an efficient and selective manner, which can be used in new targeted therapies for ErbB2-overexpressing breast cancers.

REFERENCES Arkin M.R., Glicksman MA, Fu H., Havel J.J., and Du Y. (2012). Assay guidance manual editors Sittampalam G.S., Gal-Edd N., Arkin M. et al., Bethesda (MD): Eh Lilly & Company and the National Center for Advancing Translational Sciences. Bublil E.M. and Yarden Y. (2007). Curr Opin This !! Biol; 19: 124-34. Chishti AH, Kim AC, Marfatia SM, Lutchman M., Hanspal M., Jindal H., Liu SC, Low PS, GA Roll, Mohandas N., Chasis JA, Conboy JG, Gascard P., Takakuwa Y., Huang SC , Benz EJ, Bretscher A, Fehon RG, Gusella JF, Ramesh V, Solomon F., Marchesi VT, Tsukita S., Tsukita S., Hoover KB (1998). Trends Biochem. Sci. ; 23 (8): 281-2. Dankort D., Jeyabalan N., Jones N., Dumont D.J. and Muller W. J. (2001). J. Biol. Chem. ; 276: 38921-38928 (2001). DeRosa M.C., and Crutchley R.J. (2002). Coordination Chemistry Reviews: 233-234,351-371. Graus-Porta D., Beerli R.R., Daly J.M. and Hynes N.E. (1997). Embo J; 16: 1647-55. Jemal A., Bray F., MM Center, Ferlay J., VVard E., and Forman D. (2011). CA Cancer J Clin; 61: 69-90. Lee J.VV., Soung YH, Seo SH, Kim SY, Park OH, Wang YP, Park K., Nam SW, Park W.5., Kim SH, Lee JY, Yoo NJ, and Lee SH (2006) . Clin Cancer Res; 12: 57-61. Sàez R., Molina M.A., Ramsey E.E., Rojo F., Keenan E.J., Albanell J., Lluch A., GarciaConde J., Baselga J., and Clinton G.M. (2006). Clin Cancer Res; 12: 424-31. Semba K., Kamata N., Toyoshima K., and Yamamoto T. (1985). Proc Nati Acad Sci U SA; 82 (19): 6497-6501.

Spector N., Xia W., El-Hariry I., Yarden Y. and Bacus S. (2007). Breast Cancer Res; 9: 205. Remington's Pharmaceutical Sciences, 22nd Edition, edited by Allen, Loyd V., Jr. Ullman EF, Kirakossian H., Singh S., Wu ZP, Irvin BR, Pease JS, Switchenko AC, Irvine JD, Dafforn A Skold ON, et al. (1994). Proc Nat! Acad Sci U S A. 1994; 30 91 (12): 5426-30.

Yamamoto T., Ikawa S., Akiyama T., Semba K., Nomura N., Miyajima N., Saito T., and Toyoshima K. (1986). Nature, 319 (6050): 230-4. Yarden Y. and Sliwkowski M.X. (2001). Nat Rev Mol This !! Biol; 2: 127-37.

Claims (10)

REVENDICATIONS1. An in vitro screening method for selecting an ErbB2 inhibitor, said method comprising the steps of: a) contacting in a first medium of at least one candidate compound with: - a polypeptide comprising the Ezrin binding domain (EBM) of ErbB2, said polypeptide being conjugated to one or more donor particles, and - a polypeptide comprising the FERM domain, said polypeptide being conjugated to one or more acceptor particles, b) exciting the medium of step a) with the aid of an illumination stimulating the activation of the donor particles, and c) measuring the luminescent signal emitted by the accepting particles following the excitation b), a significant variation M of the luminescent signal in the presence of said candidate compound with respect to the luminescent signal in the absence of said candidate compound being indicative of the inhibition of the FERM domain binding to the ErbB2 EBM domain by the candidate compound; d) selecting at least one candidate compound capable of inhibiting the binding of the FERM domain to the EBM domain of ErbB2, and contact in a second medium of said compound with: a polypeptide comprising the EBM domain of a protein other than ErbB2 , said polypeptide being conjugated to one or more donor particles, and - a polypeptide comprising the FERM domain, said polypeptide being conjugated to one or more acceptor particles, e) exciting the medium of step b) using a illumination stimulating the activation of the donor particles, and f) measuring the luminescent signal emitted by the accepting particles following the excitation e), a non-significant A2 variation of the luminescent signal in the presence of said candidate compound being indicative of the inhibition capacity ErbB2 of said compound selected in step d). 2. Method according to claim 1, in which: said EBM domain of ErbB2 is of sequence SEQ ID NO: 1; and / or said FERM domain is of sequence SEQ ID NO: 2; and / or - said EBM domain of a protein other than ErbB2 is selected from the group consisting of EBM domains of CD44, CD43, ICAM-1, ICAM-2 and ICAM-3. The method of claim 1 or 2, wherein said donor particles comprise at least one photosensitizer and said acceptor particles comprise at least one chemiluminescent agent, said agent generating the luminescent signal when the photosensitizer converts ambient oxygen to singlet oxygen following the excitation with said illumination and the donor and acceptor particles are close to each other. The method of claim 3, wherein said photosensitizer is selected from the group consisting of acetone, endoperoxide, benzophenone, chlorophylls, eosin, erythrosine, fullerenes, methylene blue, porphyrins, phthalocyanines, rose bengal, 9-thioxanthone, and any derivative thereof. The method of claim 3, wherein said chemiluminescent agent is selected from the group consisting of thioxene, anthracene, rubrene, europium, terbium, and any derivative thereof. The method of any one of claims 3 to 5, wherein said donor and acceptor particles are in proximity to one another if the distance therebetween is less than or equal to about 200 nm. The method of any one of claims 1 to 6, wherein the illumination is at a wavelength greater than that of the measurement of the luminescent signal. A process for preparing a pharmaceutical composition for the prevention and / or treatment of pathologies associated with overactivation and / or overexpression of ErbB2, comprising the steps of: a) selecting at least one ErbB2 inhibitor according to the method as defined in any one of claims 1 to 7; and b) adding a pharmaceutically acceptable carrier to said inhibitor, wherein said ErbB2 inhibitor is isoliquiritigenin. A kit for use in a method as defined in any one of claims 1 to 8, comprising: a) a polypeptide comprising the EBM domain of ErbB2; b) a polypeptide comprising the FERM domain; and c) optionally, a polypeptide comprising the EBM domain of a protein other than ErbB2. The kit of claim 9, further comprising: d) donor particles capable of being conjugated to said polypeptide a); and e) acceptor particles capable of being conjugated to said polypeptide b).