The present invention concerns molecules, in particular active as BRAG2 inhibitors and applications thereof.

In particular, the invention concerns BRAG2 inhibitors in the treatment of a cancer or angiogenesis.

Prior Art

Cells response to variations in their environment by assembling dynamic signalling complexes at the surface of membranes, which collect signals and transmit information. Such signalling platforms are often dysfunctional in disease, either because of mutations that affect their regulation or because they are appropriated by pathological pathways. Drugs modulating their activity are thus highly sought-after, but signalling nodes have remained challenging targets by conventional competitive inhibitors because of their structural flexibility, which commonly involves large conformational changes, their widespread protein-protein interfaces and their multiple protein-lipid interactions. Consequently, a current drug discovery challenge is to develop novel strategies that use the structural features of membrane-associated signalling complexes.

Small GTPases and their regulators belong to the category of peripheral membrane proteins involved in pathologies. Proteins of the large family of small GTPases are chief organizers of signalling platforms at the surface of membranes with crucial functions in signal transduction, cell motility, membrane traffic and the coordination between these pathways. Because of their importance in normal cell homeostasis, small GTPase functions are twisted or hijacked in diverse pathologies, such as cancer, cardiovascular diseases and bacterial or viral infections. Inhibiting their activities in pathological contexts is therefore a compelling, unmet need in drug discovery. Small GTPases regulation is highly complex as it combined a GDP/GTP switch and a cytosol/membrane cycle where the GTP-bound GTPase is attached to the membrane. Added to this, the output of GTPase signalling involves multiple activators (guanine nucleotide exchange factors or GEFs, which stimulate the GDP/GTP exchange), inhibitors (GTPase-activating proteins or GAPs, which accelerate GTP hydrolysis and GDIs, which wrap their lipidic anchor to solubilize them), and effectors, which collectively assemble signalling platforms. GEFs, GAPs and effectors are themselves highly regulated through structural rearrangements and protein-membrane interactions. To date, some strategies inhibit the membrane/cytosol cycle by targeting either enzymes involved in the anchoring of their lipid post-translational modification or GDIs. Yet, inhibitors that directly target the regulatory protein-membrane interactions have never been described.

Peripheral membrane proteins coordinate cell responses to signals coming from their environment and are hence involved in numerous diseases. Despite their significance, they have remained elusive targets for conventional competitive inhibitors and alternate approaches are highly needed.

Aims of the invention

The present invention aims to solve the technical problem of providing a BRAG2 inhibitor and applications thereof.

The present invention aims to solve the technical problem of providing new route for the treatment of a cancer or angiogenesis.

More particularly, the present invention aims to solve the technical problem of providing new route for the treatment of a cancer or angiogenesis by a BRAG2 inhibitor.

The present invention also aims to solve the technical problem of providing a molecule binding a protein-membrane interface.

Detailed description

The present invention solves at least one and preferably all technical problem set forth in the present invention.

In particular, the present invention relates to molecules having the following chemical structure (I) or a pharmaceutically acceptable salt thereof or a prodrug thereof, for use in a method of therapeutic treatment:

wherein:

R1 is a fluorinated alkyl, preferably CF3;

R3 is a chemical group comprising at least one oxygen and/or a nitrogen;

R2, R4, R5 and R6 are independently atoms or groups of atoms.

The invention also relates to molecules having the following chemical structure (I):

wherein:

R1 is a fluorinated alkyl, preferably CF3;

R3 is a chemical group comprising at least one oxygen and/or a nitrogen;

R6 is an atoms or group of atoms different than hydrogen;

R2, R4, R5 and are independently atoms or groups of atoms.

In particular, the present invention relates to a BRAG2 inhibitor having a structure of a molecule as defined in the present invention.

BRAG2 designates the protein Brefeldin-resistant Arf-GEF 2 protein (SEQ ID NO:1 ) (IQ motif and SEC7 domain-containing protein 1 , i.e. see UniProtKB - Q6DN90 (IQEC1_HUMAN) - SEQ ID NO:1 ). This protein is also designated as ADP-ribosylation factors guanine nucleotide-exchange protein 100 or ADP-ribosylation factors guanine nucleotide-exchange protein 2. Another isoform of BRAG2 exists, referenced as UniProtKB - A0A087WWK8 (SEQ ID NO:10). This isoform shares a high sequence identity (same PH domain), but differs mainly by a longer N-terminus. Inhibitors of BRAG2 of the present invention inhibiting BRAG2 (Q6DN90) are considered to inhibit also this isoform (A0A087WWK8) as they bind to PH domain.

In particular the invention relates to non-competitive inhibition of its protein- membrane interactions having potent and selective inhibition of a membrane-associated regulator of small GTPases. Molecules according to the invention designated as Bragsin, inhibit the activation of Arf GTPases. Such inhibition is performed for example by their guanine nucleotide exchange factor BRAG2 in vitro, and this effect is specific and manifests only in the presence of membranes. Advantageously, in cells, molecules according to the invention affect the trans-Go\g\ network, and this effect is rescued by ectopic expression of BRAG2 or constitutively active Arf and is phenocopied by BRAG2 gene silencing.

In one embodiment, said molecule is selected from the group consisting of:

wherein R’ is a chemical group of atoms, for example an alkyl optionally substituted or COOR’ form an ester salt, for example a sodium ester.

In one embodiment, R4 is selected from the group consisting of an hydrogen, an hydroxy, an alkyl, preferably a methyl (Me) or ethyl (Et), an O-alkyl (or alkoxy), preferably OMe or OEt, an alkene, an O-alkylene, an alkyne, preferably -CCH, or an O-alkyne, preferably -OCH2-CCH.

In one embodiment, R6 is selected from the group consisting of an hydrogen, an hydroxy, an alkyl, preferably a methyl (Me) or ethyl (Et), an O-alkyl, preferably OMe or OEt, an alkene, an O-alkylene, an alkyne, preferably -CCH or an O-alkyne, preferably - OCH2-CCH.

In one embodiment, R5 is selected from the group consisting from the group consisting of an hydrogen, an hydroxy, an alkyl, preferably a methyl (Me) or ethyl (Et), an O-alkyl, preferably OMe or OEt, an alkene, an O-alkylene, an alkyne, preferably -CCH or an O-alkyne, preferably -OCH2-CCH.

In one embodiment, R2 is H.

In one embodiment, R4 is H.

In one embodiment, R5 is H.

In one embodiment, R6 is H.

In one embodiment, R6is selected from the group consisting of an hydroxy, an alkyl, preferably a methyl (Me) or ethyl (Et), an O-alkyl (or alkoxy), preferably OMe or OEt, an alkene, an O-alkylene, an alkyne, preferably -CCH, or an O-alkyne, preferably -

OCH2-CCH.

In one embodiment, R6 is OMe. In one embodiment, R2, R5 and R6 are hydrogen atoms.

In one embodiment, said molecule is selected from the group consisting of:

In one embodiment, said molecule is selected from the group consisting of:

In one embodiment, wherein R1 is CF ₃ . In one embodiment, wherein R3 is NO2.

In one embodiment, wherein R3 is NO2 and R1 is CF ₃ .

As used herein, the term "alkyl group" means a saturated straight chain or branched non-cyclic hydrocarbon having from 1 to 10 carbon atoms. Representative saturated straight chain alkyls include methyl, ethyl, n-propyl, n-butyl, n-pentyl, n-hexyl, n-heptyl, n- octyl, n-nonyl and n-decyl; while saturated branched alkyls include isopropyl, sec-butyl, isobutyl, tert-butyl, isopentyl, 2-methylbutyl, 3-methylbutyl, 2-methylpentyl, 3-methylpentyl, 4-methylpentyl, 2-methylhexyl, 3-methylhexyl, 4-methylhexyl, 5-methylhexyl, 2,3- dimethylbutyl, 2,3-dimethylpentyl, 2,4-dimethylpentyl, 2,3-dimethylhexyl, 2,4- dimethylhexyl, 2,5-dimethylhexyl, 2,2-dimethylpentyl, 2,2-dimethylhexyl, 3,3- dimtheylpentyl, 3,3-dimethylhexyl, 4,4-dimethylhexyl, 2-ethylpentyl, 3-ethylpentyl, 2- ethylhexyl, 3-ethylhexyl, 4-ethylhexyl, 2-methyl-2-ethylpentyl, 2-methyl-3-ethylpentyl, 2- methyl-4-ethylpentyl, 2-methyl-2-ethylhexyl, 2-methyl-3-ethylhexyl, 2-methyl-4-ethylhexyl, 2,2-diethylpentyl, 3,3-diethylhexyl, 2,2-diethylhexyl, 3,3-diethylhexyl and the like. Alkyl groups included in compounds of this invention may be optionally substituted with one or more substituents.

As used herein, the term "alkoxy" or“O-alkyl” refers to an alkyl group which is attached to another moiety by an oxygen atom. Examples of alkoxy groups include methoxy, isopropoxy, ethoxy, tert-butoxy, and the like. Alkoxy groups may be optionally substituted with one or more substituents.

As used herein, the term "heterocycloalkyl" means a monocyclic or polycyclic group having at least one heteroatom selected from O, N or S, and which has 2-1 1 carbon atoms, which may be saturated or unsaturated, but is not aromatic. Examples of heterocycloalkyl groups include (but are not limited to): piperidinyl, piperazinyl, 2- oxopiperazinyl, 2-oxopiperidinyl, 2-oxopyrrolidinyl, 4-piperidonyl, pyrrolidinyl, hydantoinyl, valerolactamyl, oxiranyl, oxetanyl, tetrahydropyranyl, tetrahydrothiopyranyl, tetrahydropyrindinyl, tetrahydropyrimidinyl, tetrahydrothiopyranyl sulfone, tetrahydrothiopyranyl sulfoxide, morpholinyl, thiomorpholinyl, thiomorpholinyl sulfoxide, thiomorpholinyl sulfone, 1 ,3-dioxolane, tetrahydrofuranyl, dihydrofuranyl-2-one, tetrahydrothienyl, and tetrahydro-1 ,1 -dioxothienyl. Typically, monocyclic heterocycloalkyl groups have 3 to 7 members. Preferred 3 to 7 membered monocyclic heterocycloalkyl groups are those having 5 or 6 ring atoms. A heteroatom may be substituted with a protecting group known to those of ordinary skill in the art, for example, the hydrogen on a nitrogen may be substituted with a tert-butoxycarbonyl group. Furthermore, heterocycloalkyl groups may be optionally substituted with one or more substituents. In addition, the point of attachment of a heterocyclic ring to another group may be at either a carbon atom or a heteroatom of a heterocyclic ring. Only stable isomers of such substituted heterocyclic groups are contemplated in this definition.

As used herein the term "substituent" or "substituted" means that a hydrogen radical on a compound or group is replaced with any desired atom or group of atoms. Examples of substituents are those found in the exemplary compounds and embodiments disclosed herein when he molecule still present a BRAG2 inhibitor activity. Examples of substituents are haloge; alkyl; alkenyl; alkynyl; hydroxy; alkoxy; nitro; thiol; thioether; imine; cyano; amido; phosphonato; phosphine; carboxyl; thiocarbonyl; sulfonyl; sulfonamide; ketone; aldehyde; ester; oxygen (-0); haloalkyl (e.g., trifluoromethyl); cycloalkyl, which may be monocyclic or fused or non-fused polycyclic, or a heterocycloalkyl, which may be monocyclic or fused or non-fused polycyclic, monocyclic or fused or non-fused polycyclic aryl or heteroaryl; amino (primary, secondary, or tertiary); C02CH3; CONH2; OCH2CONH2; NH2; S02NH2; OCHF2; CF3; OCF3; and such moieties may also be optionally substituted by a fused-ring structure or bridge, for example -0CH20-. These substituents may optionally be further substituted with a substituent selected from such groups. In one embodiment, the term "substituent" or the adjective "substituted" refers to a substituent selected from the group consisting of an alkyl, an alkenyl, an alkynyl, an cycloalkyl, an cycloalkenyl, a heterocycloalkyl, an aryl, a heteroaryl, an aralkyl, a heteraralkyl, a haloalkyl, -C(0)NR1 1 R12, -NR13C(0)R14, a halo, -OR13, cyano, nitro, a haloalkoxy, -C(0)R13, -NR1 1 R12, -SR13, -C(0)0R13, -0C(0)R13, -NR13C(0)NR1 1 R12, -0C(0)NR1 1 R12, -NR13C(0)0R14, -S(0)rR13, -NR13S(0)rR14, -0S(0)rR14,

S(0)rNR1 1 R12, -O, -S, and -N-R13, wherein r is 1 or 2; R1 1 and R12, for each occurrence are, independently, H, an optionally substituted alkyl, an optionally substituted alkenyl, an optionally substituted alkynyl, an optionally substituted cycloalkyl, an optionally substituted cycloalkenyl, an optionally substituted heterocycloalkyl, an optionally substituted aryl, an optionally substituted heteroaryl, an optionally substituted aralkyl, or an optionally substituted heteraralkyl; or R1 1 and R12 taken together with the nitrogen to which they are attached is optionally substituted heterocycloalkyl or optionally substituted heteroaryl; and R13 and R14 for each occurrence are, independently, H, an optionally substituted alkyl, an optionally substituted alkenyl, an optionally substituted alkynyl, an optionally substituted cycloalkyl, an optionally substituted cycloalkenyl, an optionally substituted heterocycloalkyl, an optionally substituted aryl, an optionally substituted heteroaryl, an optionally substituted aralkyl, or an optionally substituted heteraralkyl.

The term "halogen" means -F, -Cl, -Br or -I. In one embodiment, Halogen is -Br. In one embodiment, Halogen is -Cl. The crystal structure of the complex between BRAG2 and molecules according to the invention and structure-activity analysis using BRAG2 mutants and analogs of molecules according to the invention revealed that

Preferably, a molecule according to the invention binds at the interface between the PH domain of BRAG2 and the lipid bilayer in a manner that renders BRAG2 unable to activate lipidated Arf.

Preferably, a molecule according to the invention depletes a cancer stem cell population in mammary tumor cell lines.

Preferably, a molecule according to the invention is for use in a treatment of a breast cancer, especially a breast cancer involving a BRAG2 overexpression.

A molecule according to the invention thus pioneers a novel class of cell-active inhibitors that impair protein-membrane interactions without disruption. The present invention is a new class of drugs targeting peripheral membrane proteins.

The invention also relates to an inhibitor having one or more protein-membrane interactions and inhibiting a mammal BRAG2, and preferably a human BRAG2, for use in a treatment of a cancer.

The invention also relates to an inhibitor of a mammal BRAG2, and preferably a human BRAG2, said inhibitor having a structure according to any one of claims 1 to 6.

The invention also relates to a method, notably in vitro or in cellulo, for inhibiting a mammal BRAG2, and preferably a human BRAG2, said method comprising putting BRAG2 in contact with a BRAG2 inhibitor as defined in the present invention.

The invention also relates to an inhibitor of a mammal BRAG2, and preferably a human BRAG2, said inhibitor having a structure as defined in the present invention.

The invention also relates to a method, notably in vitro or in cellulo, for inhibiting a mammal BRAG2, and preferably a human BRAG2, said method comprising putting BRAG2 in contact with a BRAG2 inhibitor as defined in the present invention.

The present invention focused notably on Arf GTPases, which orchestrate a variety of regulatory functions in lipid and membrane trafficking, and their GEF BRAG2. BRAG2 belongs to the BRAG family, whose members activate Arf GTPases to control signaling and/or endocytosis of integrins and other receptors and cell adhesion (D'Souza, R. S., & Casanova, J. E. (2016). - see below).

In one embodiment, molecules according to the present are inhibitors of Arf pathways.

In one embodiment, molecules according to the present invention are inhibitors of the activation of myristoylated Arf 1 and myristoylated Arf6 by BRAG2. In one embodiment, molecules according to the present invention do not inhibit the nucleotide exchange activity of BIG1 , a Golgi ArfGEF, ARNO and EFA6a.

In one embodiment, molecules according to the present invention bind to PH domain of BRAG2.

In one embodiment, molecules according to the present invention are inhibitors of Sec7 domain of BRAG2.

Surprisingly, in one preferred embodiment, molecules according to the present invention are specific to BRAG2 over BIG1 , a Golgi ArfGEF, and of ARNO, EFA6a.

Surprisingly, in one preferred embodiment, molecules according to the present invention are BRAG2 inhibitors in the presence of an artificial membrane, and even more surprisingly of a membrane containing PIP ₂ . In one embodiment, molecules according to the present invention bind at the interface between the PH domain and the membrane.

In one embodiment, molecules according to the present invention disperse TGN46 immuno-staining in cells.

The present invention also relates to a complex of BRAG2, especially human BRAG2, and more particularly BRAG2 ^{Sec7 PH} , with a molecule according to the present invention.

BRAG2 is implicated in severe pathologies, including breast cancer (Morishige, M. et al. (2008)), uveal melanoma (Yoo, J.H. et al. (2016)), diabetic retinopathy Zhu, W. et al. (2017)) and mental retardation Shoubridge, C. et al. (2010)). BRAG family members contain a Sec7 domain, which is responsible for stimulating GDP/GTP exchange, followed by a pleckstrin homology (PH) domain, which binds PIP2-containing membranes with high affinity. Previous work has shown that PIP2-containing membranes potentiate the GEF activity of BRAG2 by more than three orders of magnitude (Aizel, K. et al. Integrated conformational and lipid-sensing regulation of endosomal ArfGEF BRAG2. PLoS Biol 11 , e1001652 (2013) and Jian, X., Gruschus, J.M., Sztul, E. & Randazzo, P.A. The pleckstrin homology (PH) domain of the Arf exchange factor Brag2 is an allosteric binding site. J Biol Chem 287, 24273-83 (2012). This large increase in activity is determined by the interaction of BRAG2 with multiple lipids, resulting in its precisely oriented apposition to the membrane (Karandur, D., Nawrotek, A., Kuriyan, J. & Cherfils, J. Multiple interactions between an Arf/GEF complex and charged lipids determine activation kinetics on the membrane. Proc Natl Acad Sci U S A 114, 1 1416-1 1421 (2017).

The present invention also relates to a pharmaceutical composition comprising at least one molecule as defined in the present invention, said composition comprising one or more excipients and optionally one other pharmaceutically active ingredient. As is known to the person skilled in the art, various forms of excipients can be used adapted to the mode of administration and some of them can promote the effectiveness of the active molecule, e.g. by promoting a release profile rendering this active molecule overall more effective for the treatment desired.

The pharmaceutical compositions of the invention are thus able to be administered in various forms, more specially for example in an injectable, pulverizable or ingestible form, for example via the intramuscular, intravenous, subcutaneous, intradermal, oral, topical, rectal, vaginal, ophthalmic, nasal, transdermal or parenteral route. A preferred route is oral administration. The present invention notably covers the use of a compound according to the present invention for the manufacture of pharmaceutical composition.

Such medicament can take the form of a pharmaceutical composition adapted for oral administration, which can be formulated using pharmaceutically acceptable carriers well known in the art in suitable dosages. Such carriers enable the pharmaceutical compositions to be formulated as tablets, pills, dragees, capsules, liquids, gels, syrups, slurries, suspensions, and the like, for ingestion by the patient. In addition to the active ingredients, these pharmaceutical compositions may contain suitable pharmaceutically- acceptable carriers comprising excipients and auxiliaries which facilitate processing of the active compounds into preparations which can be used pharmaceutically. Further details on techniques for formulation and administration may be found in the latest edition of Remington's Pharmaceutical Sciences (Maack Publishing Co., Easton, Pa.).

The present invention also relates to a molecule or a pharmaceutical composition as defined in the present invention for use in a method of a therapeutic treatment of disease presenting a deregulated expression of BRAG2, preferably an overexpression of BRAG2.

The present invention also relates to a molecule or a pharmaceutical composition as defined in the present invention as a drug having a protein-membrane interaction for use in a method of a therapeutic treatment of disease in need of a protein-membrane interaction.

In one embodiment, said disease is selected from the group consisting of a cancer, in particular an invasive cancer, a cancer with metastasis, a cancer resistant to an EGFR and/or ErbB2 modulator, angiogenesis, diabetic retinopathy, nonsyndromic intellectual disability, etc.

The present invention also relates to a molecule or a pharmaceutical composition as defined in the present invention for use in a method of a therapeutic treatment of a disease is selected from the group consisting of a cancer, in particular an invasive cancer, a cancer with metastasis, a cancer resistant to an EGFR and/or ErbB2 modulator, angiogenesis, diabetic retinopathy, non-syndromic intellectual disability, etc. The present invention also relates to a method of therapeutic treatment, said method comprising administering to a mammal in need thereof an effective amount of at least one molecule as defined in the present invention.

Preferably, said mammal is a human patient.

In one embodiment, said method of treatment is for treating breast cancer, and in particular breast cancer in need of targeting bCSC population.

A method of treatment of a cancer or angiogenesis according to the invention is for example supported by:

Braq2 and cancers

D'Souza, R. S., & Casanova, J. E. (2016). The BRAG/IQSec family of Arf GEFs. Small GTPases, 7(4), 257-264. http://doi.org/10.1080/21541248.2016.1219442

Matsumoto, Y., Sakurai, H., Kogashiwa, Y., Kimura, T., Matsumoto, Y., Shionome, T., et al. (2017). Inhibition of epithelial-mesenchymal transition by cetuximab via the EGFR-GEP100-Arf6-AMAP1 pathway in head and neck cancer. Head & Neck, 39(3), 476—485. http://doi.Org/10.1002/hed.24626

Xie, C.-G., Wei, S.-M., Chen, J.-M., Xu, X.-F., Cai, J.-T., Chen, Q.-Y., & Jia, L.-T. (2012). Down-Regulation of GEP100 Causes Increase in E-Cadherin Levels and Inhibits Pancreatic Cancer Cell Invasion. PLoS ONE, 7(5), e37854. http://doi.org/10.1371/journal.pone.0037854

Hu, Z., Du, J., Yang, L., Zhu, Y., Yang, Y., Zheng, D., et al. (2012). GEP100/Arf6 is required for epidermal growth factor-induced ERK/Rad signaling and cell migration in human hepatoma HepG2 cells. PLoS ONE, 7(6), e38777. http://doi.org/10.1371/journal.pone.0038777

Menju, T., Hashimoto, S., Hashimoto, A., Otsuka, Y., Handa, H., Ogawa, E., et al. (201 1 ). Engagement of overexpressed Her2 with GEP100 induces autonomous invasive activities and provides a biomarker for metastases of lung adenocarcinoma. PLoS ONE, 6(9), e25301. http://doi.org/10.1371/journal.pone.0025301

Yoo, J.H. et al. ARF6 Is an Actionable Node that Orchestrates Oncogenic GNAQ Signaling in Uveal Melanoma. Cancer Cell 29, 889-904 (2016).

Breast cancer

Hu, Z., Xu, R., Liu, J., Zhang, Y., Du, J., Li, W., et al. (2013). GEP100 regulates epidermal growth factor-induced MDA-MB-231 breast cancer cell invasion through the activation of Arf6/ERK/uPAR signaling pathway. Experimental Cell Research, 319( 13), 1932-1941 . http://doi.Org/10.1016/j.yexcr.2013.05.028 Sabe, H., Hashimoto, S., Morishige, M., Ogawa, E., Hashimoto, A., Nam, J.-M., et al. (2009). The EGFR-GEP100-Arf6-AMAP1 signaling pathway specific to breast cancer invasion and metastasis. Traffic (Copenhagen, Denmark), 10(8), 982-993. http://doi.Org/10.1 1 1 1/j.1600-0854.2009.00917.x

Sabe, H., Hashimoto, S., Morishige, M., Ogawa, E., Hashimoto, A., Nam, J.-M., et al. (2009). The EGFR-GEP100-Arf6-AMAP1 signaling pathway specific to breast cancer invasion and metastasis. Traffic (Copenhagen, Denmark), 10(8), 982-993. http://doi.Org/10.1 1 1 1/j.1600-0854.2009.00917.x

Morishige, M. et al. GEP100 links epidermal growth factor receptor signalling to Arf6 activation to induce breast cancer invasion. Nat Cell Biol 10, 85-92 (2008).

Angiogenesis

Manavski, Y., Carmona, G., Bennewitz, K., Tang, Z., Zhang, F., Sakurai, A., et al. (2014). Brag2 differentially regulates b1 - and 33-integrin-dependent adhesion in endothelial cells and is involved in developmental and pathological angiogenesis. Basic Research in Cardiology, 109(2), 404. http://doi.org/10.1007/s00395-014-0404-2

Hashimoto, A., Hashimoto, S., Ando, R., Noda, K., Ogawa, E., Kotani, H., et al. (201 1 ). GEP100-Arf6-AMAP1 -cortactin pathway frequently used in cancer invasion is activated by VEGFR2 to promote angiogenesis. PLoS ONE, 6(8), e23359. http://doi.org/10.1371/journal.pone.0023359

Zhu, W., Shi, D. S., Winter, J. M., Rich, B. E., Tong, Z., Sorensen, L. K., et al. (2017). Small GTPase ARF6 controls VEGFR2 trafficking and signaling in diabetic retinopathy. The Journal of Clinical Investigation, 127(12). http://doi.Org/10.1 172/JCI91770

Sakurai, A., Jian, X., Lee, C. J., Manavski, Y., Chavakis, E., Donaldson, J., et al. (201 1 ). Phosphatidylinositol-4-phosphate 5-kinase and GEP100/Brag2 protein mediate antiangiogenic signaling by semaphorin 3E-plexin-D1 through Arf6 protein. Journal of Biological Chemistry, 286(39), 34335-34345. http://doi.org/10.1074/jbc.M1 1 1.259499

And also:

Zhu, W. et al. Small GTPase ARF6 controls VEGFR2 trafficking and signaling in diabetic retinopathy. J Clin Invest (2017).

Shoubridge, C. et al. Mutations in the guanine nucleotide exchange factor gene IQSEC2 cause nonsyndromic intellectual disability. Nat Genet 42, 486-8 (2010).

In the figures:

Figure 1. Bragsin2 affects Arf pathways in cells a. Chemical structures of Bragsinl and Bragsin2 and derivatives used in this study. Chemical synthesis and structural characterization of the compounds are described in example 6.

b. Bragsin2 disperses the TGN46 and GM130 markers. HeLa cells were treated with either DMSO (0.25%) or Bragsin2 (50 mM) for 30 min, immunostained for TGN46, GM130 or EEA1 (green channel) and analysed by confocal microscopy. The effect of Bragsinl treatments are shown in Figure 11.

c. Dispersion of GM130 and TGN46 by Bragsin2 is reversible. HeLa cells were treated as in Figure 1a and were incubated for an additional 30 min in fresh medium (wash panel) prior to immunostaining and confocal microscopy analysis. Reversibility of the effect of Bragsinl is shown in Figure 12.

d. Expression of Arf-mCherry constructs carrying an activating mutation rescues the effect of Bragsin2. HeLa cells were transfected with constitutively active Q/L mutants of Arf-mCherry. Note the difference between non-transfected cells (white asterisk) and transfected cells. Rescue of Bragsinl phenotype by constitutively active Arf mutants is shown in Figure 13. Scale bars, 10 mhi.

Figure 2. Bragsin is a specific inhibitor of the ArfGEF BRAG2

a. Bragsinl has no effect on the Sec7 domain of human ArfGEFs in solution. Nucleotide exchange kinetics were measured by fluorescence kinetics in the presence of Bragsinl (50 mM) or DMSO using purified Sec7 domains and a truncated version of Arf 1 (D17Arf1 ), which can be activated in solution. Representative kinetic profiles are given in

Figure 14.

b. Bragsinl specifically inhibits BRAG2 in the presence of liposomes. Nucleotide exchange kinetics were determined in the presence of Bragsinl (50 mM) or DMSO with ArfGEF constructs carrying membrane-binding domains and myristoylated Arf 1 . Rad was artificially tethered to liposomes by a C-terminal hexahistidine tag as described in (Peurois, F. et al. Biochem J 474, 1259-1272 (2017). Representative kinetic profiles are given in Figure 15.

c. Bragsinl inhibits the activation of myristoylated Arf6 by BRAG2 on liposomes. Experiments were carried out as in Figure 2b. Representative kinetic profiles are given in

Figure 16.

d. Dose-response of Bragsinl and Bragsin2 towards myristoylated Arf 1 and BRAG2 on liposomes. Experiments were carried out as in Figure 2b.

e. Silencing of BRAG2 phenocopies the effect of Bragsin2 on the TGN46 compartment. HeLa cells were treated with DMSO or Bragsin2 (50 mM) or transfected with siRNAs targeting BRAG2, ARNO or GBF1 or with a control siRNA (siCTRL). Immunofluorescence staining of TGN46 is in green (white-grey on the black-white figures). Cell boundaries were highlighted by actin staining (magenta (grey on the black- white figures). SiRNA silencing efficiencies are shown in Figure 17. Scale bars, 10 pm. Statistical analysis of the dispersal of TGN46 staining is shown in Figure 19.

f. Overexpression of BRAG2 rescues the dispersion of the TGN46 compartment induced by Bragsin. HeLa cells were transfected with BRAG2-mCherry (magenta channel) and treated with Bragsin2 (50 mM). Immunofluorescence staining of TGN46 is in green (white-grey on the black-white figures).

Figure 3. Bragsin binds to the PH domain of BRAG2

a. Crystallographic structure of Bragsin 1 bound to the PH domain of BRAG2. The inset shows an electron density omit map of the inhibitor. The Sec7 domain is in pink, the linker in yellow, the PH domain in blue.

b. Interactions of Bragsinl with residues of the PH domain. Hydrogen bonds are shown in dotted lines.

c. Analysis of the inhibition of BRAG2 mutants by Bragsin2. GEF efficiencies were measured by fluorescence kinetics using myristoylated Arf 1 as in Figure 2b.

d. Structure-activity relationship analysis of Bragsin analogs. All compounds were used at 20 mM. GEF activities were measured by fluorescence kinetics using myristoylated Arf 1 as in Figure 2b.

e. Effect of Bragsin analogs on the TGN46 compartment. HeLa cells treated with the compounds (50 mM) were immunostained and analysed by confocal microscopy as in

Figure 1 b.

Figure 4. Bragsin is a non-competitive inhibitor of protein-membrane interactions

a. Bragsinl (in violet) overlaps with the canonical phosphoinositide-binding site of the PH domain of BRAG2. IP3 (in red) is from the GRP1 -IP3 complex (DiNitto, J.P. et al. Structural basis and mechanism of autoregulation in 3-phosphoinositide-dependent Grp1 family Arf GTPase exchange factors. Mol Cell 28, 569-83 (2007)). The position of the membrane is indicated by an arrow.

b. Bragsin2 does not impair binding of BRAG2 to liposomes. Binding was measured by liposome flotation. U: bottom fraction, containing unbound proteins ; B: top fraction, containing liposome-bound proteins. Proteins are revealed by Instant Blue staining after SDS-PAGE. Quantification is shown below. c. Inhibition of BRAG2 by Bragsin2 is stronger in the presence of liposomes that contain PIP ₂ lipids. Kinetics were measured as in Figure 2b.

Figure 5. Bragsin affects breast cancer stem cells

a. Evolution of the proportion of ALDH ^br cells after treatment with Bragsin2 using the Aldefluor assay. Results are expressed as mean ± SD.

b. Tumorsphere-forming efficiency (SFE) was calculated using an extreme limiting dilution analysis (ELDA) algorithm. Results are expressed as the estimated number of tumorspheres obtained for 100 cells plated.

Figure 6. A model for protein-membrane interfacial inhibition by Bragsin

Figures 7-9. Analysis of the stability of Bragsinl and Bragsin2

Fig. 7. Bragsinl that has been kept in a stock solution in DMSO for several months lost activity towards BRAG2 (not shown). 1 H-NMR analysis shows that it completely decomposed to hydrated compound (13)

Fig. 8-9. Mass spectroscopy analysis shows Bragsinl is less stable than Bragsin2. Bragsinl and Bragsin2 were dissolved in PBS. Bragsinl slowly decomposed to hydrolyzed compound within 72 hours (Fig. 8) while Bragsin2 remained stable (Fig. c).

Figures 10-13 : Bragsinl has the same cellular effects as Bragsin2

Fig. 10 : Bragsin2 has no effect on the tubulin and actin networks in Hela cells. HeLa cells were treated with either DMSO (0.25%) or Bragsin2 (50 mM) for 30 min, immunostained for oc-tubulin (green channel) or F-actin (magenta channel) and analysed by confocal microscopy.

Fig. 1 1 : Bragsinl disperses the TGN46 and GM130 markers. HeLa cells were treated as in Figure 10 with 50 mM Bragsinl and immunostained for TGN46 or GM130 (green channel) prior to confocal microscopy analysis.

Fig. 12 : Dispersion of GM130 and TGN46 by Bragsinl is reversible. After treatment with either DMSO (0.25%) or Bragsinl (50 mM) for 30 min, HeLa cells were washed and incubated for an additional 30 min in fresh medium (wash) prior to immunostaining for TGN46 or GM130 (green channel) and confocal microscopy analysis.

Fig. 13 : Expression of constitutively active Arf1 , Arf5 and Arf6 constructs rescues the effect of Bragsinl . HeLa cells were transiently transfected with a plasmid encoding the Q/L mutant of the indicated Arf-mCherry isoform; non transfected cells are stamped with a white asterisk. Scale bars, 10 mhi. Figures 14-19: Bragsin is a specific inhibitor of BRAG2

14 : Representative fluorescence kinetic traces of the experiments shown in Figure 2a. The GDP/GTP nucleotide exchange of N-terminally truncated Arf-GDP was followed in solution by tryptophan fluorescence in the presence of 50 mM of Bragsinl (light grey) or 0.25% of DMSO (dark grey) after addition of catalytic amount of the indicated ArfGEF Sec7 domain and an excess of GTP (see material and methods for details). The spontaneous curve (black) corresponds to Arf activation in the absence of GEF.

15 : Representative fluorescence kinetic traces of the experiments shown in Figure 2b. ^myr Arf1 or Rad activation by catalytic amount of the indicated GEF was performed as in Figure 14 except that the experiments contained 100 mM liposomes (see material and methods for details; example 7).

16: Representative fluorescence kinetic traces of the experiments shown in Figure 2c. ^myr Arf6 activation was performed in the presence of liposomes as described in Figure 15.

17-18 : RT-qPCR analysis of ARNO, BRAG2 and GBF1 mRNA levels. HeLa cells were either transfected with siRNAs targeting BRAG2 (siBRAG2, light grey bars), ARNO (siARNO, medium grey bars), GBF1 (siGBFI , dark grey bars) or with a nontargeting siRNA (siCTRL, white bars) as a control (Fig. 17), or treated with Bragsin2 (dark grey bars) or with DMSO (white bars) as a control (Fig. 18). Total RNAs were extracted, reverse-transcribed and submitted to real-time PCR using primer pairs specific for mRNAs encoding ARNO, BRAG2 or GBF1 (as indicated below graphs). The graphs show the amounts of mRNAs relative to control values, which are set to 100%.

19 : Statistical analysis of the dispersal of TGN46 staining of the experiments shown in Figure 2e. Area of cells (polygon) and TGN (minimal ellipse) were determined manually after sum Z projection of stacks and measured using F lmageJ. Their ratio expressed the relative area (%) occupied by the TGN46 staining in the cell. Statistical analyses were done using ANOVA with Tukeys post-test. Counted cells > 200, ^* p<0.05, ^**** p<0.0001 .

Figures 20-21 :

20: Shows that Bragsin3 (compound 14) strongly inhibits BRAG2 in the presence of liposomes. Nucleotide exchange kinetics were determined in the presence of Bragsin3 (50 microM) or DMSO with BRAG2 ^Sec7PH construct and myristoylated Arf 1 (see material and methods for details). 21 : Bragsin3 disperses the TGN46 marker. HeLa cells were treated as in Figure 1 b with 50 mM Bragsin3 and immunostained for TGN46 prior confocal microscopy analysis.

Figures 22-29:

22: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line MDA-MB-231 .

23: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line MCF7.

24: Shows the viability of Bragsine 2, Compound 14 and Compound 3 on Cell line SUM149.

25: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line SUM159.

26: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line A549.

27: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line U87-MG.

28: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line PANC-1.

29: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line.

Viability tests (measure of the quantity of ATP after cell lysis, CelltiterGlo, Promega) and cytotocixity (measure of LDH after cell lysis, CytoTox-One, Promega.

Results in figures 22-29 were analysed by one-way ANOVA followed by Tukey’s multiple comparisons (vs DMSO), ^* P<0.05, ^** P<0.01 , ^*** P<0.001 , ^**** P<0.0001.

Bragsin2: 6-methoxy-5-nitro-2-(trifluoromethyl)-4H-chromen-4-one

Compound 14 (2MetO): 6,8-dimethoxy-5-nitro-2-(trifluoromethyl)-4H-chromen-4-one

Compound 3 (-N02): 6-methoxy-2-(trifluoromethyl)-4H-chromen-4-one (negative control).

The present invention is further illustrated by means of the following examples.

The data presented in these examples, and also in parts of the patent description, are in part taken from preliminary analysis and as such represent a close approximation to the final, validated dataset. However, this fully supports the present invention.

Other aims, characteristics and advantages of the invention will appear clearly to the person skilled in the art upon reading the explanatory description which makes reference to the Examples which are given simply as an illustration and which in no way limit the scope of the invention.

The Examples make up an integral part of the present invention, and any characteristic which appears novel with respect to any prior state of the art from the description taken in its entirety, including the Examples, makes up an integral part of the invention in its function and in its generality.

Thus, every example has a general scope.

Furthermore, in the Examples, all percentages are given by mass, unless indicated otherwise, temperature is expressed in degrees Celsius unless indicated otherwise, and the pressure is atmospheric pressure, unless indicated otherwise.

EXAMPLES

A small molecule Bragsinl was originally discovered in a yeast chemogenomic screen; Bragsin perturbs the function of the yeast ArfGEF Sec7p, and a related analog, such as Bragsin2 or molecules according to the present invention, especially as defined by any of the claimed structures, referred to collectively as Bragsin (Figure 1 a). Bragsin is a specific inhibitor of the ArfGEF BRAG2 in vitro and in cells. Cell-based assays and in vitro reconstitution of GEF activities in artificial membranes using pure proteins were performed. Bragsin is a non-competitive inhibitor, which binds at the GEF/membrane interface and acts to orient the GEF on the membrane in a manner that makes BRAG2 incompetent for nucleotide exchange activity according to crystallographic analysis coupled to structure-activity relationship (SAR) analysis. Consistent with a role of BRAG2 in breast cancer, Bragsin depletes the cancer stem cell population in mammary tumor cell lines. Bragsin is as a novel type of inhibitor that targets protein-membrane interactions with potential in breast cancer treatment.

Example 1 - Bragsin inhibits the activation of Art GTPases in cells.

Bragsinl may affect Arf pathways in cells because Bragsin 1 perturbs the function of the yeast ArfGEF Sec7p. Bragsinl was found to be chemically unstable after a few days in aqueous solution leading to a biologically inactive hydrated derivative. A related analog bearing a methoxy group instead of a methyl group (Bragsin2) was resistant to hydration and stable (Figures 7-9). In initial experiments Bragsin2 was used to assess possible effects on the structure of the c/s- Golgi and trans-Go\g\ network (TGN) compartments, which are regulated by Arf-dependent pathways. In HeLa cells, the c/s- Golgi marker GM130 and the TGN marker TGN46 was dispersed by Bragsin2 into punctate structures of heterogeneous sizes (Figure 1 b), and this effect was reversible (Figure 1c). In contrast, Bragsin2 had no effect on the early endosome marker EEA1 (Figure 1 b) nor did it affect tubulin and actin networks (Figure 10). Cells were transfected with Arf-mCherry constructs carrying a QL mutation that makes Arf constitutively active to assess if this effect was mediated via interference with Arf GTPase pathways,. Arf1 , Arf5 and Arf6 expression rescued Bragsin2-induced dispersion of the TGN46 compartment, consistent with an effect of the compound on Arf pathways (Figure 1d). The phenotypes of cells treated with freshly prepared Bragsinl were identical to those treated with Bragsin2 (Figure 11 ), were reversible (Figure 12) and rescued by expression of constitutively active Arf GTPases (Figure 13). Together, these data suggest that Bragsinl and Bragsin2 inhibit Arf pathways in cells, in agreement with the original activity observed in a chemogenomic screen of Bragsinl towards the yeast ArfGEF Sec7p.

Example 2 - Bragsin is a specific inhibitor of BRAG2 that necessitates the presence of membranes for inhibition

Arf GTPases are activated in human cells, by several ArfGEF subfamilies, all of which contain a conserved Sec7 domain decorated with variable appended domains (Nastou, K.C., Tsaousis, G.N., Kremizas, K.E., Litou, Z.l. & Hamodrakas, S.J. The human plasma membrane peripherome: visualization and analysis of interactions. Biomed Res Int 2014, 397145 (2014); DiNitto, J.P. et al. Mol Cell 28, 569-83 (2007)).

The effect of Bragsinl on the GEF efficiency of representative human ArfGEFs was evaluated by fluorescence kinetics using highly purified recombinant Arf GTPases and ArfGEFs. In a first series of assays, the Sec7 domains of BIG1 , a Golgi ArfGEF, and of ARNO, EFA6a and BRAG2, which function at the plasma membrane, were used. When tested in solution, Bragsinl had no effect on the activation of Arf 1 by any of these Sec7 domains (Figures 2a and 14). Separately, we reconstituted the activation of myristoylated Arf1 by the same ArfGEFs on liposomes using ArfGEF constructs that contain a membrane-binding domain (DCB-HUS domain for BIG1 , PH domains for the other GEFs). Remarkably, Bragsinl strongly inhibited BRAG2 on liposomes, while it had no effect on the other ArfGEFs (Figures 2b and 15) or on Rad /Trio, an unrelated small GTPase/GEF system (Figures 2b and 15). BRAG2 activates several Arf isoforms, including the plasma membrane-located Arf6 isoform. Bragsinl also inhibited the BRAG2-mediated activation of myristoylated Arf6, indicating that inhibition is independent of the Arf isoform used (Figures 2c and 16). Bragsinl and Bragsin2 have the same IC ₅ o towards BRAG2 (3 mM) according to dose-response analysis, confirming that Bragsinl and Bragsin2 can be used interchangeably (Figure 2d). The specific inhibition of BRAG2 by Bragsin in vitro predicts that BRAG2 silencing by siRNA should yield the same phenotype as treatment by the small molecule. Silencing of BRAG2 induced the redistribution of TGN46 into tubular or punctate structures, which phenocopied the effects induced by Bragsin (Figures 2e, 17 and 18). As a control, silencing of the Golgi ArfGEF GBF1 resulted in a markedly different phenotype, in which TGN46 was either concentrated into punctate structures or became diffuse throughout the cytosol. Silencing of ARNO, which is located at the plasma membrane, had no effect on TGN46 distribution. Overexpression of BRAG2-mCherry rescued the dispersion of the TGN46 structure induced by Bragsin, which is consistent with the silencing experiments (Figure 2f). Bragsin is a specific inhibitor of the ArfGEF BRAG2 in vitro and in cells according to these experiments. These oexperiments identify a mechanism of inhibition that requires the membrane to manifests itself.

Example 3 - Bragsin binds to the PH domain of BRAG2

In vitro and cellular assays support that the BRAG2 ^{Sec7 PH} construct used in vitro recapitulates the inhibitory effects seen with full-length BRAG2 in cells. This construct was used to obtain the crystal structure of the BRAG2 ^{Sec7 PH} Bragsinl complex (Table 1 ). Unambiguous electron density was observed near the PH domain in which the inhibitor could be modeled (Figure 3a), and its position was not involved in crystal contacts. Bragsinl interacts with the PH domain by a combination of polar and hyrdophobic contacts involving Leu 651 , His 652 and Arg 654 in strand b1 , Lys 667 in strand b3 and Arg 681 in strand b4 (Figure 3b). CF ₃ and carbonyl groups of Bragsin2 interact with the PH domain, but not the methyl group, which explains why, where this latter group is replaced by a methoxy group, is as potent as Bragsinl . Conversely, hydration of Bragsinl leading to a sp3 geometry and a different orientation of the carbon bearing the CF ₃ substituent, was in agreement with the loss of activity of Bragsinl over time (see above). R654E or R681 E mutations were introduced in the PH domain of BRAG2 to confirm the location of the binding site, the. Inhibition was affected for both mutants (Figure 3c). The R681 E mutation rendered BRAG2 insensitive to Bragsin and, interestingly, the R654E mutant was more inhibited by Bragsin than the wild type BRAG2. Next, Bragsin analogs were synthesized wherein NO2 or CF ₃ functional groups, which establish interactions with BRAG2 in the crystal, were removed or modified (compounds (3) to (6), Figure 1a and Example 6). All analogs displayed reduced inhibition of BRAG2 in vitro (Figure 3d) and were not capable to disperse TGN46 (compounds 3, 4, and 6) or had less effect (compound 5) in cells (Figure 3e). Together, the crystallographic, mutagenesis and structure-activity relationship analyses consistently pinpointed a specific binding site for Bragsin within the PH domain of BRAG2.

Example 4 - Bragsin is a non-competitive inhibitor of BRAG2/membrane interactions.

Bragsin may partially overlap with a lipid bound at this position was predicted by comparison of the PH domain of a cytohesin ArfGEF with a phosphoinositide headgroup bound to the canonical lipid-binding site (DiNitto, J.P. et al. Mol Cell 28, 569-83 (2007))(Figure 4a). Bragsin inhibition could act by blocking the interaction of BRAG2 with membranes. Surprisingly, Bragsin2 did not displace BRAG2 from PIP ₂ -containing liposomes, showing that it does not inhibit BRAG2 by disrupting its association with the membrane (Figure 4b). Alternatively, Bragsin may bind at the interface between the PH domain and the membrane if inhibition is sensitive to the lipid composition of the membrane. Using liposomes that support strong BRAG2 activity, inventors found that Bragsin2 inhibits BRAG2 more efficiently on liposomes that contain PIP ₂ compared to liposomes that contain PS as the sole negatively charged lipids (Figure 4c).These experiments support that Bragsin inhibits BRAG2 by binding at the interface between BRAG2 and the lipid bilayer in a non-competitive manner, causing a mode of binding of BRAG2 that is not competent for promoting nucleotide exchange on membrane-attached myristoylated Arf.

Example 5 - Bragsin affects breast cancer stem cells

Cancer stem cells represent the cell population that sustains tumor growth, metastasis, resistance to chemo- and radio-therapies and recurrence after treatment. Targeting stem cell populations has become a key step in the design of efficient anticancer strategies because their relative abundance in tumors correlates with poor prognosis in patients,. It was tested whether Bragsin affects breast cancer stem cells (bCSCs) as BRAG2 has been reported to be involved in breast cancer cell invasion (Morishige, M. et al. GEP100 links epidermal growth factor receptor signalling to Arf6 activation to induce breast cancer invasion. Nat Cell Biol 10, 85-92 (2008)). It was first evaluated the effect of Bragsin on the bCSC population in three different breast cancer cell lines (SUM159, SUM149, S68), using aldehyde dehydrogenase (ALDH ^br ) activity as a marker of the malignant bCSC population (Charafe-Jauffret, E. et al. ALDH1 -positive cancer stem cells predict engraftment of primary breast tumors and are governed by a common stem cell program. Cancer Res 73, 7290-300 (2013); Ginestier, C. et al. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 1 , 555-67 (2007).. Bragsin treatment (50 mM) depleted the bCSC population in two of the three cell lines tested (SUM149 and S68) (Figure 5a). A single-cell tumorsphere formation assay was performed to functionally validate these observations. A marked decrease was observed in tumorsphere-forming efficiency (SFE) in SUM149 and S68 cell lines following treatment by Bragsin, consistent with the depletion of the bCSC ALDH ^br population (Figure 5b). In contrast, the SUM159 cell line, whose bCSC ALDH ^br activity was not decreased by Bragsin, presented a similar SFE, whether they were treated by Bragsin or not. Together, these results indicate that Bragsin may target the bCSC population of some breast cancers.

Example 6- Synthesis of Bragsinl , Bragsin 2 and analogs used in this study (Figure 1a)

6.1. General information

Unless otherwise stated, all glassware was flame-dried or oven dried before use and all reactions were performed under an atmosphere of argon. Dichloromethane, acetonitrile, toluene, methanol and DMSO were purchased from Sigma-Aldrich anhydrous grade and used as received; all other solvents are distilled before use. All reagents were used as received from commercial suppliers unless stated otherwise. Reaction progress was monitored by thin layer chromatography (TLC) performed on aluminium plates coated with silica gel F254. Visualization was achieved by fluorescence quenching with UV light at 254 nm or by staining using potassium permanganate, phosphomolibdic acid solution, p- anisaldehyde solution or vanillin solution and heating. Flash column chromatography was performed using silica gel 60 (230-400 mesh, Merck and co.). ¹ H NMR and ¹³ C NMR spectra were recorded using a Bruker AV-300, AV-400 and AV-500 spectrometer at 300K. Chemical shifts were given in parts per million (ppm, d), referenced to the solvent peak of CDCI3, defined at d = 7.26 ppm ( ¹ H NMR) and d = 77.16 ( ¹³ C NMR). Coupling constants are quoted in Hz ( J ). ¹ H NMR splitting patterns are designated as singlet (s), doublet (d), doublet of doublet (dd), triplet (t), quartet (q), pentet (p). Splitting patterns that could not be interpreted or easily visualized are designated as multiplet (m) or broad (br).

6.2. Experimental Procedures

Synthesis of Bragsinl, Bragsin2 and compound (3) relflux

R = Me R = Me (Bragsin1 )....(1)

= OMe OMe (Bragsin2)....(2)

6.2.1. General procedure

Step 1 and 2

Dry THF (1.5 ml) and finely dispersed LiH (3.4 equivalent) were placed in a round- bottom two-necked flask under argon and the mixture was brought to reflux. A solution of the desired ketone (1.0 equivalent) and ethyl 2,2,2-trifluoroacetate (1.5 equivalent) in dry THF was added over 10-15 min, under stirring. The mixture was refluxed for 2 h and then quenched with an aqueous solution of aqueous 1 N HCI. The reaction mixture was extracted with EtOAc three times and the organic layers were combined, washed with brine, dried over MgS0 ₄ , filtered, and concentrated under reduced pressure vacuo. The product was used in the next step without further purifications.

A drop of a 37% aq. HCI solution was added to a solution containing the reaction product obtained after step 1 (1.0 equivalent) in acetic acid (5 ml), under stirring and the solution was refluxed for 1 h. After this time, the reaction was diluted with water and the solvent was removed under vacuum. The crude mixture was obtained and purified by column chromatography (cyclohexane: ethyl acetate, gradient from 0% to 60% ethyl acetate) to deliver designed chromenone derivatives.

Step 3

A mixture of concentrated H2SO4 (0.6 ml) and concentrated HN0 ₃ (0.6 ml) was added to a solution of purified product from step 2 (1 equiv) in cone. H2SO4 (2 ml). The reaction mixture was stirred at 75°C for 35 min and then diluted with ice-cold water. The reaction mixture was extracted by using ethyl acetate 3 times, washed with brine, dried over MgSC>4, filtered and concentrated under reduced pressure. The crude mixture was purified by chromatography (reverse/normal phase) to deliver designed nitro-chromenone products.

6-methyl-5-nitro-2-(trifluoromethyl)-4H-chromen-4-one (Bragsinl , Figure 1 a):

The title compound was prepared from commercially available 2'-hydroxy-5'-methylacetophenone, according to general procedure. Bragsinl was obtained in 1 1 % yield (in 3 steps). Data is in accordance with the literature. ¹ H-NMR (500 MHz, CDCI3): d 7.71 (d, J = 8.9 Hz, 1 H), 7.63 (d, J = 8.9 Hz, 1 H), 6.74 (s, 1 H), 2.39 (s, 3H).

6-methoxy-5-nitro-2-(trifluoromethyl)-4H-chromen-4-one (Bragsin2, Figure 1 a):

The title compound was prepared from commercial available 2'-hydroxy-5'-methoxyacetophenone, according to general procedure. Bragsin2 was obtained in 78% yield (in 3 steps). Data is in accordance with the literature. ¹ H-NMR (500 MHz, CDCI ₃ ): d 7.71 (d, J = 9.5 Hz, 1 H), 7.54 (d, J = 9.5 Hz, 1 H), 6.70 (s, 1 H), 3.98 (s, 3H).

6-methoxy-2-(trifluoromethyl)-4H-chromen-4-one (3) (compound 3, Figure 1 a):

The title compound was prepared from commercial available 2'-hydroxy-5'-methoxyacetophenone , according to general procedure (step 1 and 2). Compound 3 was obtained in 90% yield (in 2 steps). Data is in accordance with the literature. ¹ H-NMR (300 MHz, CDCI ₃ ): d 7.54 (d, J = 3.1 Hz, 1 H), 7.48 (d, J = 9.2 Hz, 1 H), 7.34 (dd, J = 9.2, 3.1 Hz, 1 H), 6.70 (s, 1 H), 3.91 (s, 3H).

Quality analysis of compound (4) (compound 4, Figure 1 a)

6-methoxy-5-nitro-2-(trifluoromethyl)-4H-chromen-4-one (4):

The title compound is commercial available compound (CAS: 354128-13-5). ¹ H-NMR (500 MHz, CDCI ₃ ): d 7.88 (d, J = 6.0 Hz, 1 H), 7.00 (s, 1 H), 6.37 (d, J= 6.0 Hz, 1 H), 4.03 (s, 3H), 2.37 (s, 3H).

Synthesis of compound (5) (compound 5, Figure 1 a)

6-hydroxy-2-(trifluoromethyl)-4H-chromen-4-one (7):

Chromenone 3 (0.1 g, 0.41 mmol, 1 .0 equiv.) was dissolved in DCM. Over the reaction mixture was added dropwise 1 ml_ of BBr ₃ (1 M, 0.82 mmol, 2.0 equiv.) at 0 °C. The reaction was stirred at room temperature for 2 hours. Then, it was quenched with water, extracted with DCM, washed with NaHC0 ₃ and dried over MgS0 ₄ . The volatiles were removed under reduced pressure. The collected residues were purified via column chromatography (hexane: EtOAc = 1 :2). Compound 7 was obtained in 69 % yield (65 mg, 0.28 mmol). ¹ H-NMR (300 MHz, CDCI ₃ ): d 7.84 (d, J = 3.0 Hz, 1 H),

7.50 (d, J = 9.2 Hz, 1 H), 7.38 (dd, J = 9.2, 3.0 Hz, 1 H), 6.75 (s, 1 H). ¹³ C-NMR (75 MHz, CDCI ₃ ): d 178.2, 155.3, 152.9 (q, J = 39.2 Hz), 150.4, 125.2, 124.7, 120.1 , 1 18.7 (q, J = 274.2 Hz), 109. 4 (q, J = 2.8 Hz), 109.2. HRMS (ESI): calculated for CI ₀ H ₆ F ₃ O ₃ [M+H] ⁺ : 231 .0259, found: 231 .0264.

6-hydroxy-4-oxo-2-(trifluoromethyl)-4H-chromene-5-carbald ehyde (8):

A mixture of chromenone 7 (100 mg, 0.44 mmol, 1 .0 equiv.) and hexamethylenetetramine (121 .9 mg, 0.87 mmol, 2.0 equiv.) was dissolved in TFA (2 ml). The mixture was heated up at 120 °C in microwave equipment for 30 minutes. Then, the reaction was cooled to room temperature and quench with ice water, extracted with Et ₂ 0 and dried over MgS0 ₄ . After the solvent was removed, the collected residues were purified via column chromatography (hexane: EtOAc = 3:2). Compound 8 was obtained in 48 % yield (55 mg, 0.21 mmol). ¹ H-NMR (300 MHz, CDCI ₃ ): d 13.09 (s, 1 H), 1 1 .43 (s, 1 H), 7.75 (d, J = 9.4 Hz, 1 H), 7.43 (d, J = 9.4 Hz, 1 H), 6.78 (s, 1 H). ¹³ C- NMR (75 MHz, CDCI ₃ ): d 198.6, 178.1 , 162.5, 151 .3 (q, J = 39.7 Hz), 150.3, 128.0, 127.2, 122.4, 1 18.5 (q, J = 273.9 Hz), 1 15.3, 1 1 1 .8 (q, J = 2.7 Hz). HRMS (ESI): calculated for C _H H ₆ F ₃ 0 ₄ [M+H] ⁺ : 259.0218, found: 259.0212.

6-methoxy-4-oxo-2-(trifluoromethyl)-4H-chromene-5-carbald ehyde (5)

(compound 5, Figure 1 a)

Under argon, compound 8 (50 mg, 0.19 mmol, 1 .0 equiv.) was dissolved in dry DMF followed by the addition of K ₂ C0 ₃ (80 mg, 0.582 mmol, 3.0 equiv.) at 0 °C. Then, methyl iodide (15 mI_, 0.23 mmol, 1 .2 equiv.) was added dropwise to the mixture at 0 °C. The reaction was stirred for 12 hours at room temperature. The mixture was quenched with NH4CI, extracted with ethyl acetate, washed with brine and dried over MgS0 ₄ . The organic layers were combined and the volatiles were removed under reduced pressure. The crude product was purified via column chromatography (hexane: EtOAc, 3:2) to afford the titled product in 86% yield (55 mg, 0.21 mmol). ¹ H-NMR (300 MHz, CDCI ₃ ): d 10.57 (s, 1 H), 7.67 (d, J = 9.3 Hz, 1 H), 7.46 (d, J = 9.3 Hz, 1 H), 6.67 (s, 1 H), 3.91 (s, 3H). ¹³ C-NMR (75 MHz, CDCI3): d 191.9, 177.5, 154.7, 152.5 (q, J = 39.4 Hz), 149.6, 126.0, 123.6, 121 .9, 120.0, 1 18. 6 (q, J= 274.2 Hz), 1 10.0 (q, J= 2.8 Hz), 57.1 . HRMS (ESI): calculated for C12H8F3O4 [M+H] ⁺ : 273.0374, found: 273.0369.

Synthesis of compound (6) (compound 6, Figure 1 a)

methyl 2-acetyl-3-hydroxy-6-methoxybenzoate (11):

To a stirred solution of furan 9 (389 mg 3.96 mmol, 1.0 equiv.) in toluene (1.5 ml_) at 0 °C was added keto ester 10 (500 mg, 3.96 mmol, 1 .0 equiv.) in one portion. Upon complete addition the amber solution was allowed to warm to 90 °C. After 1 hour the reaction mixture was concentrated in vacuo to give a bicyclic mixture as 1 :3 regioselectivity

(observed by NMR) as viscous burgundy oil. To a stirred solution of this bicyclic mixture (875 mg, 3.97 mmol, 1.0 equiv.) in THF (4 ml_) was slowly added a solution of dry hydrochloric acid in ether (1.0 M, 0.8 ml_, 0.78 mmol, 0.2 equiv.) at 0 °C over 5 min. Upon complete addition the amber solution was allowed to warm to room temperature. After 2 hours the reaction mixture was concentrated in vacuo to give an amber oil. The crude material (412 mg, 43% yield over 2 steps) was purified via silica gel column chromatography (hexane:EtOAc, 10:1 ) to give product 11 (89 mg) as a clear light yellow oil . ¹ H-NMR (300 MHz, CDCI ₃ ): d 1 1.91 (s, 1 H), 7.17 (d, J = 9.2 Hz, 1 H), 7.04 (d, J = 9.2 Hz, 1 H), 3.95 (s, 3H), 3.81 (s, 3H), 2.52 (s, 3H). ¹³ C-NMR (75 MHz, CDCI ₃ ): d 203.8, 168.6, 156.5, 149.1 , 124.0, 121.3, 121.0, 1 17.5, 57. 7 53.2, 29.3. HRMS (ESI): calculated for CiiHi ₂ 0 ₅ Na [M+Na] ⁺ : 247.0582, found: 247.0580. methyl 6-methoxy-4-oxo-2-(trifluoromethyl)-4H-chromene-5-carboxylat e (12):

The title compound was prepared from benzoquinone 11 , according to general procedure. Compound 12 was obtained in 85% yield. ¹ H-NMR (300 MHz, CDCI ₃ ): d 7.63 (d, J = 9.4 Hz, 1 H), 7.42 (d, J = 9.4 Hz, 1 H), 6.65 (s, 1 H), 4.01 (s, 3H), 3.93 (s, 3H). 1 ³ C-NMR (75 MHz, CDCI ₃ ): d 176.0, 166.9, 154.2, 152.0 (q, J =

39.4 Hz), 149.6, 121.9, 120.7, 120.4, 1 19.0, 1 18. 6 (q, J = 274.5 Hz), 109.9 (q, J = 2.7 Hz), 57.0, 53.3. HRMS (ESI): calculated for C _{I 3} H _{I 0} F ₃ O ₅ [M+H] ⁺ : 303.0430, found: 303.0475. 6-methoxy-4-oxo-2-(trifluoromethyl)-4H-chromene-5-carboxylic acid (6)

(compound 6, Figure 1a)

Chromenone 12 (50 mg, 0.16 mmol, 1 .0 equiv.) was dissolved in DCM. Over the reaction mixture was added dropwise 0,3 ml_ of BBr ₃ (1 M, 0.33 mmol, 2.0 equiv.) at 0 °C. The reaction was stirred at room temperature for 1 hour. Then, it was quenched with water, extracted with DCM, washed with NaHC0 ₃ and dried over MgS0 ₄ . The volatiles removed under reduced pressure. The collected residues were purified via column chromatography (hexane: EtOAc = 2:3). After chromatography separation, compound 6 was obtained in 22 mg (46% yield) as white solid. ¹ H-NMR (300 MHz, CDCI ₃ ): d 7.56 (d, J = 9.3 Hz, 1 H), 7.40 (d, J = 9.3 Hz, 1 H), 6.69 (s, 1 H), 3.96 (s, 3H). ¹³ C-NMR (75 MHz, CDCI ₃ ): d 175.6, 168.8, 154.8, 151 .7 (q, J = 39.3 Hz), 150.2, 125.3, 122.9, 122.8, 1 18. 6 (q, J = 274.3 Hz), 1 13.3, 1 10.7 (q, J = 2.7 Hz), 53.3. HRMS (ESI): calculated for Ci ₂ H ₈ F ₃ 0 ₅ [M+H] ⁺ : 289.0324, found: 289.0317.

2-hydroxy-6-methyl-5-nitro-2-(trifluoromethyl)chroman-4-o ne (13):

Data is in accordance with the literature ¹ H-NMR (300 MHz, CDCI ₃ ): d 7.49 (d, J = 8.6 Hz, 1 H), 7.15 (d, J = 8.6 Hz, 1 H), 5.71 (s (br), 1 H, OH), 3.10 (q, J = 9.6 Hz, 2H), 2.24 (s, 3H).

6,8-dimethoxy-5-nitro-2-(trifluoromethyl)-4H-chromen-4-on e (14)

Step 1

Charge Pd(OAc) ₂ (4.4 mg, 0.02 mmol), K ₂ S ₂ 0 ₈ (216 mg, 0.8 mmol) in a 4 mL intillation vial, followed by 2 mL TFA and ketone substrate (0.4 mmol). Seal the reaction with a teflon-lined cap. Heat the reaction at 50 °C for 1 .5 hours on a pie block.Monitor the reaction by TLC. Remove the solvent in vacuo. Subject the residue to flash chromatography (silica gel) using Hex/DCM to obtain hydroxy-3E ¹ ,5E ¹ - dimethoxyacetophenone.

¹ H-NMR (400 MHz, CDCI ₃ ): d 12.21 (s, 1 H), 6.69 (d, J = 0.7 Hz, 2H), 3.86 (s, 3H), 3.79 (s, 3H), 2.61 (s, 3H)

General procedure

Step 2

Dry THF (1 .5 ml) and finely dispersed LiH (3.4 equivalent) were placed in a round- bottom two-necked flask under argon and the mixture was brought to reflux. A solution of containing, corresponding ketone, commercial starting material (1 .0 equivalent) and ethyl 2,2,2-trifluoroacetate (1 .5 equivalent) in dry THF was added over 10-15 min, under stirring. The mixture was refluxed for 2 h and then quenched with an aqueous solution of aqueous 1 N HCI. The reaction mixture was extracted with EtOAc three times and the organic layers were combined, washed with brine, dried over MgS0 ₄ , filtered, and concentrated in vacuo. The product was used in the next step without further purifications.

Step 3

One drop of cone. HCI was added to a solution of crude from step 1 (1 .0 equivalent) in acetic acid, under stirring and the solution was refluxed for 1 h. After this time, the reaction was diluted with water and the solvent was removed under vacuum. The crude mixture was obtained and purified by chromatography to deliver designed chromenone derivatives.

Step 4 (Nitration)

A mixture of cone. H2S04 (0.6 ml) and cone. HN03 (0.6 ml) was added to a solution of purified product from step 2 (1 equiv) in cone. H2S04 (2.3 ml). The reaction mixture was stirred at 75°C for 35 min and then diluted with ice-cold water. The precipitate was filtered off, washed with water, dried and recrystallized from BuOH. Compound 14 (Bragsin3) was obtained as a white crystal solid.

Example 7 - Material and methods.

7.1. Chemicals.

Nucleotides were purchased from Jena Bioscience. BFA was from Sigma. Bragsinl (6-methyl-5-nitro-2-(trifluoromethyl)-4H-chromen-4-one) and Bragsin2 (6-methoxy-5-nitro- 2-(trifluoromethyl)-4H-chromen-4-one) were purchased from Vitas-M laboratory, further purified and analyzed as described in example 6. Analysis of the stability of Bragsinl and Bragsin2 is shown in Figures 7-9. Bragsinl became commercially unavailable in the course of this study, and some experiments were therefore carried out only with Bragsin2 (effect on BRAG2 mutants-Fig 3c, effect on BRAG2 membrane binding-Fig 4b, and effect of membrane composition on Bragsin efficiency- Fig 4c). Bragsinl and Bragsin2 can be used indifferently. Synthesis of Bragsin analogs shown in Figure 1a is described in example 6.

7.2. Antibodies and cDNAs.

Mouse monoclonal anti-GM130 (cis- Golgi matrix protein of 130 kDa) and anti-EEA1 (early endosome antigen 1 ) were from Transduction Laboratories. Mouse monoclonal anti-oc-tubulin was from Sigma. Sheep antibody against TGN46 (trans-Go\g\ network protein of 46 kDa) was purchased from AbD Serotec. Alexa 647-labelled phalloidin was from Invitrogen. For secondary antibodies, Alexa 488-conjugated with goat anti-mouse or donkey anti-sheep IgGs (Invitrogen) was used for immunofluorescence and horseradish peroxidase-conjugated chicken anti-mouse IgG (Santa Cruz Biotechnology) was used for Western blotting. Plasmids encoding full-length Arf 1 Q71 L, Arf5 Q71 L and Arf6 Q67L mutants were kindly provided by Julie Menetrey (LEBS, CNRS, Gif-sur-Yvette, France) and used as templates for sub-cloning in the pmCherry-N1 vector (Clontech) for expression of mCherry C-terminal fusion mutant proteins in mammalian cells. Full-length sequence coding for human BRAG2b (1 -963) was synthetized by ProteoGenix and sub cloned into pmCherry-N1 (Clontech) for transient expression of BRAG2-mCherry in mammalian cells. BRAG2 ^Sec7PH mutants were generated by site-directed mutagenesis using Quickchange kit (Stratagene) as per manufacturer’s protocol.

7.3. Cell culture, transfection and treatment with inhibitors.

HeLa cells were grown in Dulbecco’s modified Eagle’s medium supplemented with Glutamax™ and 10% fetal bovine serum (Invitrogen). For immunofluorescence studies, cells were grown on Labtek glass slides (Nunc) and transfected for 18 to 24 h using Lipofectamine 2000 (Invitrogen), according to the supplier’s instructions. When specified, cells were treated with the indicated concentration of small molecule or corresponding volume of vehicle (DMSO) in growth medium for 30 min at 37 ^° C.

7.4. Immunofluorescence and confocal microscopy.

Immunostaining procedure was as described in (Viaud, J. et al. Structure-based discovery of an inhibitor of Arf activation by Sec7 domains through targeting of protein- protein complexes. Proc Natl Acad Sci U S A 104, 10370-5 (2007)) except that secondary antibody incubation was performed for 1 h using Alexa 488-conjugated goat anti-mouse (1 :600) or donkey anti-sheep (1 :500) IgG and Alexa 647-conjugated phalloidin (1 :100). Anti-GM130 was used at dilution 1 :200. Images were bidirectionally recorded using an inverted Leica TCS SP8 laser-scanning confocal microscope with a 100c (N. A. 1.40) oil objective (HCX APO, Leica). Fluorochromes were detected sequentially using excitation laser lines at 488 nm (Alexa 488), 594 nm (mCherry) and 633 nm (Alexa 647). Stacks were generated using a z-step of 0.5 pm and processed using F///71 m age J (Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat Methods 9, 676-82 (2012)). Images are representative of at least two independent experiments.

7.5. siRNA knockdown.

Gene silencing was achieved using siRNAs targeting all the known transcript variants of each GEF gene: siBRAG2, Hs_IQSEC1_5 (Qiagen, SI03019408); siARNO, Hs_PSCD2_3 (Qiagen, SI00061299); siGBFI , Hs_GBF1_3 (Qiagen, SI00425418). Nontargeting control siRNA: siCTRL, ANStars Negative Control siRNA (Qiagen, 1027280). HeLa cells were submitted to a double reverse transfection using Lipofectamine® RNAiMAX (Invitrogen, 12323563) following the manufacturer's instructions. Briefly, trypsinized cells were seeded at 15,000 cells/cm ² into the wells of a 12-well plate, each containing the transfection mix: 10 pmol siRNA and 3 mI Lipofectamine® RNAiMAX in 200 mI Opti-MEM® for a final volume of 1 ml. 48 hours later, cells were replated and transfected again following the same protocol, in 6-well plates for RNA extraction (final volume 2.5 ml/well) and in 8-well Lab-Tek® glass slides for immunofluorescence staining (final volume 200 mI/well). Analyses were performed 48 hours after the second transfection. As determined by RT-qPCR, BRAG2, ARNO and GBF1 mRNA levels were specifically decreased by about 50%, 83% and 88% in the presence of their respective siRNAs (Figure 17) and these levels were not modified by treatment with Bragsin (Figure 18). Total RNA was extracted with the RNeasy® Mini Kit (Qiagen, 74104) following the manufacturer's instructions. Each sample RNA (1 pg) was reverse-transcribed using the iScript™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad, 170-8840). cDNAs (100 ng) were quantified by real-time PCR in a CFX Connect instrument (Bio-Rad) using the SsoAdvanced™ Universal SYBR® Green Supermix (Bio-Rad, 172-5270). Amplification was performed according to the Bio-Rad standard Prime-PCR protocol. Primers were chosen with the NCBI Primer-BLAST algorithm as follows: BRAG2, 5'- CGTGGCATTTCTTTGGTGTC-3' (SEQ ID NO:2) and 5'- ACCCG AG AATT G ATG AGTCG- 3' (SEQ ID NO:3); ARNO, 5'-TTGTGTCAAGGATAGGGCTG-3' (SEQ ID NO:4) and 5'- ACTT CACCTT CAT AG AGGCG-3' (SEQ ID NO:5); GBF1 , 5'-

TGT CACT CT CT ACCTTTGCG-3' (SEQ ID NO:6) and 5'-AAAT CT CCGCT GT GT CCAT C-3' (SEQ ID NO:7) ; GAPDH, 5'-ACAAGAGGAAGAGAGAGACCC-3' (SEQ ID NO:8) and 5'- TACATGACAAGGTGCGGCTC-3' (SEQ ID NO:9). 7.6. Proteins.

Bovine A17Arf1 and human A13Arf6, full-length myristoylated Arf 1 and Arf6, human BRAG2 ^Sec7 (390-594) and BRAG2 ^Sec7PH (390-81 1 ), human EFA6 ^Sec7 (527-727) and EFA6 ^Sec7PHCt (527-1024), ARNO ^Sec7 (50-256), BIG1 ^Sec7 (691 -889), BIG1 ^DcbHusSec7 (2-888) and ARNO ^Sec7PH (50-399) were expressed and purified as in (Benabdi, S. et al. Family wide Analysis of the Inhibition of Arf Guanine Nucleotide Exchange Factors with Small Molecules: Evidence of Unique Inhibitory Profiles. Biochemistry 56, 5125-5133 (2017)) and references therein. BRAG2 ^Sec7PH mutants were purified as the wild type protein. Purification of human full-length Rad carrying a 6xHis tag in C-terminus and human TRIO ^{DH1 PH1} (1232-1550) was described in (Peurois, F. et al. Characterization of the activation of small GTPases by their GEFs on membranes using artificial membrane tethering. Biochem J 474, 1259-1272 (2017)).

7.7. Liposomes and flotation assay.

Lipids (of natural origin) were from Avanti Polar Lipids, except NBD-PE from Sigma. Liposomes were prepared as described in (Aizel, K. et al. Integrated conformational and lipid-sensing regulation of endosomal ArfGEF BRAG2. PLoS Biol 11 , e1001652 (2013)) and extruded at 0.2 pm. For the Arf GEF specificity studies, liposomes contained 48% phosphatidylcholine (PC), 20% phosphatidylethanolamine (PE), 30% phosphatidylserine (PS) and 2% phosphatidylinositol-4,5-bisphosphate (PIP2). For Rac activation assay, liposomes contained 43% PC, 20% PE, 10% PS, 20% cholesterol, 2% PIP ₂ , 5% NiNTA lipids and 0.2% NBD-PE. To exclude aggregation or disruption of the liposomes, their size distribution was controlled before and after experiments by dynamic light scattering as described in (Benabdi, S. et al. Biochemistry 56, 5125-5133 (2017)). Dose-response and flotation assays were carried out with liposomes containing 37.9% PC, 20% PE, 20% PS, 2% PIP2, 20% cholesterol and 0.1 % NBD-PE. Flotation assays were carried out as in ¹⁸ .

7.8. Nucleotide exchange assays.

Nucleotide exchange kinetics were monitored by tryptophan fluorescence with excitation/emission wavelengths of 292/340 nm using a Cary Eclipse fluorimeter (Varian) at 37 ^° C and under continuous stirring as described in ^{50 51} . For specificity assays, 50 mM of Bragsinl or 0.25% DMSO were incubated for 2 min at 37 ^° C in HKM buffer (50 mM HEPES pH 7.4, 120 mM potassium acetate, 1 mM MgCL and 1 mM DTT) with either 1 mM of N-terminally truncated Arf-GDP and 100 nM of indicated Sec7 domain (in solution) or 0.4 mM of ^myr Arf1 in the presence of 100 mM of liposomes and 2 to 100 nM of ArfGEFs as described in(Benabdi, S. et al. Biochemistry 56, 5125-5133 (2017) and Peurois, F. et al. Biochem J 474, 1259-1272 (2017)). Nucleotide exchange rates (I ) were determined from monoexponential fits and means are given as the percentage of control activity ± SD. All experiments were done in triplicate.

7.9. Crystallization and structure determination.

BRAG2 was concentrated to 5 mg/ml for crystallization and crystals were obtained at 293 K by vapor diffusion in 18% PEG 20000, 0.1 M Tris-HCI pH 8.5. Crystals were transferred to the reservoir solution supplemented with 10% glycerol and soaked with 20 mM Bragsinl in a final volume of 100 mI and incubated for 24 hours at room temperature. Diffraction data were collected at PROXIMA2 beamline (SOLEIL Synchrotron, Gif-sur- Yvette, France) from a single crystal and processed with autoPROC (Vonrhein, C. et al. Data processing and analysis with the autoPROC toolbox. Acta Crystallogr D Biol Crystallogr 67 , 293-302 (201 1 )). The structure was solved by molecular replacement with Phaser (McCoy, A.J. et al. Phaser crystallographic software. J Appl Crystallogr 40, 658- 674 (2007)) using unbound BRAG2 (PDB 5NLY) (Karandur, D. et al., J. Proc Natl Acad Sci U S A 114, 1 1416-1 1421 (2017)) as a model. The ligand fit was done using RHOFIT (Global Phasing Ltd.) or the ligand fit option in Phenix, which both gave the same ligand orientation. The structure was refined with Phenix (Adams, P.D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D Biol Crystallogr 66, 213-21 (2010)) and Buster (Blanc, E. et al. Refinement of severely incomplete structures with maximum likelihood in BUSTER-TNT. Acta Crystallogr D Biol Crystallogr 60, 2210-21 (2004)), in alternation with model building in Coot(Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta crystallographica. Section D, Biological crystallography 60, 2126-32 (2004)). Statistics for data processing and refinement are reported in Table 1. Coordinates and structure factors have been deposited to the Protein Data Bank with entry code 6FNE.

7.10. Cancer stem cell assays

Three breast cancer cell lines (BCL) from three distinct molecular subtypes (SUM149/basal, SUM159/mesenchymal and S68/luminal) were used in this study. All BCLs were grown in standard medium as previously described (Charafe-Jauffret, E. et al. Cancer Res 73, 7290-300 (2013)). The ALDEFLUOR Kit (Stem Cell Technologies) was used to isolate the population with high aldehyde dehydrogenase enzymatic activity using an LSR2 cytometer (Becton-Dickinson Biosciences) as previously described (Ginestier, C. et al. Cell Stem Cell 1 , 555-67 (2007)). For the tumorsphere assay, BCLs were grown in adherent condition under Bragsin2 treatment (50 mM) or vehicle for 72 hours, then seeded as single cells in ultra-low attachment plates (Corning) following a limiting dilution. Tumorspheres were grown in a serum-free mammary epithelium basal medium. The capacity of cells to form tumorspheres was quantified under microscope. Statistical analysis of tumorsphere-forming efficiency was done with Extreme LDA software (http://bioinf.wehi.edu.au/software/elda/).

Conclusions of examples

In the present invention, Bragsin has been identified as a potent and selective inhibitor of BRAG2 that affects breast cancer stem cells in view of reconstitution of lipidated Arf GTPases and GEFs on artificial membranes together with cell-based assays. The mechanism of inhibition was supported by crystallography, mutagenesis, SAR and membrane-binding assays. Inventors supporte by these experiments implications for the cell biology of BRAG2 and its role in breast cancer. Interfacial inhibition of protein- membrane interactions as a new concept in drug discovery was also established.

The characterization of Bragsin reveals a previously overlooked aspect of BRAG2 functions in controlling the integrity of the TGN compartment. Previous studies reported a general role of BRAG2 in regulating plasma membrane receptor signaling, such as AMPA, EGF, VEGF and GNAQ receptors and trafficking of adhesion proteins including b1 - integrins and N-cadherin, but the underlying functional pathways have remained unclear. Interestingly, recent studies showed that b1 -integrins use a retrograde route to the TGN to be secreted in a polarized manner leading to cell adhesion or persistent migration, and that recycling b1 -integrins transiently localize to TGN46-positive post-Golgi carriers. The conspicuous effect of Bragsin on the structure of the TGN suggest a role of BRAG2 in regulating main aspects of the traffic of b1 -integrins and other receptors to and from the TGN.

Bragsin involves a unique mechanism of action at the protein-membrane interface. Bragsin binds at the edge of the canonical lipid-binding site of the PH domain without disrupting the interaction of BRAG2 with the liposomes. It is thus likely that Bragsin is able to contact BRAG2 and the membrane simultaneously.

PIP ₂ -containing membranes potentiate the GEF activity of BRAG2 as highlighted by previous studies (Aizel, K. et al. Integrated conformational and lipid-sensing regulation of endosomal ArfGEF BRAG2. PLoS Biol 11 , e1001652 (2013) ; Jian, X., Gruschus, J.M., Sztul, E. & Randazzo, P.A. The pleckstrin homology (PH) domain of the Arf exchange factor Brag2 is an allosteric binding site. J Biol Chem 287, 24273-83 (2012). This involves contacts with multiple lipids leading to a well-defined apposition of the Arf-BRAG2 complex on the membrane (Karandur, D., Nawrotek, A., Kuriyan, J. & Cherfils, J. Multiple interactions between an Arf/GEF complex and charged lipids determine activation kinetics on the membrane. Proc Natl Acad Sci U S A 114, 1 1416-1 1421 (2017)). It can be predicted that mispositioning of BRAG2 on the membrane should affect its efficiency. The biophysical and structural data reported in the present invention robustly suggest that this is the mechanism whereby Bragsin inhibits BRAG2, as summarized in the model shown in Figure 6. Given that Bragsin targets a PH domain, which is widespread in signaling peripheral membrane proteins (Lemmon, M.A. Membrane recognition by phospholipid binding domains. Nat Rev Mol Cell Biol 9, 99-1 1 1 (2008)), it is remarkable and surprising that it did not inhibit the exchange activity of the ArfGEFs ARNO and EFA6 and of the RacGEF Trio, all of which also carry PH domains that bind to the plasma membrane. Such selectivity is supported by the mounting evidence that lipid cooperativity and/or recognition of several types of phosphoinositides is a common principle for the specific membrane recruitment of many PH domains and other peripheral membrane-binding domains. As a consequence, the collective physico-chemical properties of a lipid bilayer encode specificity determinants that are recognized by peripheral membrane proteins in addition to, or in place of, a particular interaction with an individual lipid, leading to unique protein-membrane interfaces. Accordingly, the specificity of Bragsin likely results from its interfacial mechanism, whereby it recognizes BRAG2 in the context of its unique interface with the lipid bilayer.

Membrane peripheral proteins constitute a large class of signaling proteins that control vital cellular processes, yet the“peripherome” is still considered poorly druggable according to the prior art. Targeting the interactions of peripheral membrane proteins with membranes exemplifies an emerging paradigm shift in drug development, where focus has traditionally rather been on the inhibition of catalytic activities and protein-protein interactions, notably in cancer. Bragsin represents an entirely novel class of inhibitors, which exploit the protein-membrane interface for efficient inhibition. This mode of inhibition is reminiscent of interfacial inhibitors of protein-protein interactions to some extent, such as the natural compound Brefeldin A, which traps another subfamily of ArfGEFs in a non productive complex with the small GTPase Arf1 , and a growing list of protein-protein stabilizers that work according to the same principle. The present invention demonstrates that the protein-membrane interface is a novel Achille’s heel in peripheral membrane proteins owing to specificity determinants and energetic features that can be exploited by small molecules for inhibition. As illustrated by inventor’s observation, Bragsin affects the sternness of breast cancer cells. The present invention has characterized novel inhibitors that exploit protein- membrane interactions. The present invention represents an exquisite tool to dissect pathways reliant on BRAG2 involved in receptor signaling and cancer. The present invention also defines a novel concept that can apply to a broad range of peripheral membrane signaling proteins and opens avenues for drug development based on the mechanisms described in this study.

Table 1. Crystal structure of Brag2/Bragsin1 complex. Crystallographic statistics.

X-rav source Proxima 2, synchrotron SOLEIL

Data collection:

Space group P2i2i2i

Cell dimensions a,b,c (A) 65.92, 66.16, 218.79

a,b,g (°) 90, 90, 90

Wavelength (A) 0.980105

Resolution (A) 65.9-2.50 (2.78-2.50)

Rsym (%) 18 (142)

Rpim(%) 5 (46)

Completeness (%) 93.9 (70)

Redundancy 13.0 (10.5)

I/s 9.9 (1 .7)

CC(1/2) (%) 99.7 (54)

Total number of reflection 286171 (1 1487)

Refinement:

Rwork / Rfree 0.22/ 0.27

Nb. Atoms

Protein 5641

Water 18

B-factors

Protein 66

Bragsinl 74

Water 43

R.M.S. deviations

Bond lengths (A) 0.009

Bond angles (°) 1 .136

PDB ID 6FNE Human BRAG2 (IQ motif and SEC7 domain-containing protein 1 ) has the following sequence: (https://www.uniprot.org/uniprot/Q6DN90.fasta)

>sp|Q6DN90|IQEC1_HUMAN IQ motif and SEC7 domain-containing protein 1 OS=Homo sapiens QX=9606 GN=IQSEC1 PE=1 SV=1

MWCLHCNSERTQSLLELELDSGVEGEAPSSETGTSLDSPSAYPQGPLVPGSSLSPDH Y

EHTSVGAYGLYSGPPGQQQRTRRPKLQHSTSILRKQAEEEAIKRSRSLSESYELSSD LQ

DKQVEMLERKYGGRLVTRHAARTIQTAFRQYQMNKNFERLRSSMSENRMSRRIVLSN M

RMQFSFEGPEKVHSSYFEGKQVSVTNDGSQLGALVSPECGDLSEPTTLKSPAPSSDF A

DAITELEDAFSRQVKSLAESIDDALNCRSLHTEEAPALDAARARDTEPQTALHGMDH RKL

DEMTASYSDVTLYIDEEELSPPLPLSQAGDRPSSTESDLRLRAGGAAPDYWALAHKE DK

ADTDTSCRSTPSLERQEQRLRVEHLPLLTIEPPSDSSVDLSDRSERGSLKRQSAYER SLG

GQQGSPKHGPHSGAPKSLPREEPELRPRPPRPLDSHLAINGSANRQSKSESDYSDGD N

DSINSTSNSNDTINCSSESSSRDSLREQTLSKQTYHKEARNSWDSPAFSNDVIRKRH YRI

GLNLFNKKPEKGVQYLIERGFVPDTPVGVAHFLLQRKGLSRQMIGEFLGNRQKQFNR DV

LDCVVDEMDFSTMELDEALRKFQAHIRVQGEAQKVERLIEAFSQRYCICNPGVVRQF RN

PDTIFILAFAIILLNTDMYSPNVKPERKMKLEDFIKNLRGVDDGEDIPREMLMGIYE RIRKRE

LKTNEDHVSQVQKVEKLIVGKKPIGSLHPGLGCVLSLPHRRLVCYCRLFEVPDPNKP QKL

GLHQREIFLFNDLLVVTKIFQKKKNSVTYSFRQSFSLYGMQVLLFENQYYPNGIRLT SSVP

GADIKVLINFNAPNPQDRKKFTDDLRESIAEVQEMEKHRIESELEKQKGVVRPSMSQ CSS

LKKESGNGTLSRACLDDSYASGEGLKRSALSSSLRDLSEAGKRGRRSSAGSLESNVE FQ

PFEPLQPSVLCS

Human IQ motif and SEC7 domain-containing protein 1 is also known with the following sequence:

(https://www.uniprot.org/uniprot/A0A087WWK8.fasta):

>tr|A0A087WWK8|A0A087WWK8_HUMAN IQ motif and SEC7 domain-containing protein 1 OS=Homo sapiens QX=9606 GN=IQSEC1 PE=1 SV=1

MACRRRYFVEGEAPSSETGTSLDSPSAYPQGPLVPGSSLSPDHYEHTSVGAYGLYSG P

PGQQQRTRRPKLQHSTSILRKQAEEEAIKRSRSLSESYELSSDLQDKQVEMLERKYG GR

LVTRHAARTIQTAFRQYQMNKNFERLRSSMSENRMSRRIVLSNMRMQFSFEGPEKVH S

SYFEGKQVSVTNDGSQLGALVSPECGDLSEPTTLKSPAPSSDFADAITELEDAFSRQ VKS

LAESIDDALNCRSLHTEEAPALDAARARDTEPQTALHGMDHRKLDEMTASYSDVTLY IDE

EELSPPLPLSQAGDRPSSTESDLRLRAGGAAPDYWALAHKEDKADTDTSCRSTPSLE RQ

EQRLRVEHLPLLTIEPPSDSSVDLSDRSERGSLKRQSAYERSLGGQQGSPKHGPHSG AP KSLPREEPELRPRPPRPLDSHLAINGSANRQSKSESDYSDGDNDSINSTSNSNDTINCSS ESSSRDSLREQTLSKQTYHKEARNSWDSPAFSNDVIRKRHYRIGLNLFNKKPEKGVQYLI ERGFVPDTPVGVAHFLLQRKGLSRQMIGEFLGNRQKQFNRDVLDCVVDEMDFSTMELD EALRKFQAHIRVQGEAQKVERLIEAFSQRYCICNPGVVRQFRNPDTIFILAFAIILLNTD MY SPNVKPERKMKLEDFIKNLRGVDDGEDIPREMLMGIYERIRKRELKTNEDHVSQVQKVEK LIVGKKPIGSLHPGLGCVLSLPHRRLVCYCRLFEVPDPNKPQKLGLHQREIFLFNDLLVV T KIFQKKKNSVTYSFRQSFSLYGMQVLLFENQYYPNGIRLTSSVPGADIKVLINFNAPNPQ D RKKFTDDLRESIAEVQEMEKHRIESELEKQKGVVRPSMSQCSSLKKESGNGTLSRACLD DSYASGEGLKRSALSSSLRDLSEAGKRGRRSSAGSLESNVEGSIISSPHMRRRATSTRE CPSRPHQTMPNSSSLLGSLFGSKRGKPPPQAHLPSAPALPPPHPPVVLPHLQHSVAGH HLGPPEGLPQAAMHGHHTQYCHMQNPPPYHHHHHHHPPQHIQHAHQYHHGPHGGHP AYGAHAHGHPPLPSAHVGHTVHHHGQPPAPPPPTSSKAKPSGISTIVSAHVGHTVHHHG QPPAPPPPTSSKAKPSGISTIV