BRAG2 INHIBITORS AND APPLICATIONS THEREOF

Abstract

The present invention relates to molecules having the following chemical structure (I). 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.

##STR00001##

Claims

1. A molecule having the following chemical structure (1) or a pharmaceutically acceptable salt thereof or a prodrug thereof, for use in a method of therapeutic treatment: ##STR00024## wherein: R1 is a fluorinated alkyl; R3 is a chemical group comprising at least one oxygen and/or a nitrogen; and R2, R4, R5 and R6 are independently atoms or groups of atoms.

2. The molecule for use according to claim 1, wherein said molecule is selected from the group consisting of: ##STR00025## wherein R′ is a chemical group of atoms and R2, R4, R5 and R6 are as defined in claim 1.

3. The molecule for use according to claim 1, wherein R4 is selected from the group consisting of an hydrogen, an hydroxy, an alkyl, an O-alkyl (or alkoxy), an alkene, an O-alkylene, an alkyne, and an O-alkyne.

4. The molecule for use according to claim 1, wherein R6 is selected from the group consisting of hydrogen, an hydroxy, an alkyl, an O-alkyl, an alkene, an O-alkylene, an alkyne, and an O-alkyne.

5. The molecule for use according to claim 1, wherein R2, R5 and R6 are hydrogen atoms.

6. The molecule for use according to claim 1, wherein said molecule is selected from the group consisting of: ##STR00026##

7. (canceled)

8. A molecule having the following chemical structure (1): ##STR00027## wherein: R1 is a fluorinated alkyl; R3 is a chemical group comprising at least one oxygen and/or a nitrogen; R6 is an atom or group of atoms different than hydrogen; R2, R4, R5 and are independently atoms or groups of atoms.

9. The molecule according to claim 8, wherein said molecule is selected from the group consisting of: ##STR00028## wherein R′ is a chemical group of atoms and R2, R4, R5 and R6 are as defined in claim 8.

10. The molecule according to claim 8, wherein R4 is selected from the group consisting of an hydrogen, an hydroxy, an alkyl, an O-alkyl (or alkoxy), an alkene, an O-alkylene, an alkyne, and an O-alkyne.

11. The molecule according to claim 8, wherein R6 is selected from the group consisting of an hydroxy, an alkyl, an O-alkyl, an alkene, an O-alkylene, an alkyne, and an O-alkyne.

12. The molecule according claim 8, wherein said molecule is selected from the group consisting of: ##STR00029##

13. The molecule of claim 1 wherein said molecule is an inhibitor of a mammalian BRAG2.

14. An in vitro or in cellulo method for inhibiting a mammalian BRAG2, said method comprising contacting the mammalian BRAG2 with a BRAG2 inhibitor according to claim 7.

15. A pharmaceutical composition comprising at least one molecule according to claim 1, said composition comprising one or more excipients and optionally one other pharmaceutically active ingredient.

16-19. (canceled)

20. A method of therapeutic treatment, said method comprising administering to a mammal in need thereof an effective amount of at least one molecule according to claim 1.

21. The method of claim 20, wherein said method is for treating a disease presenting a deregulated expression of BRAG2.

22. The method according to claim 20, wherein said method is for treating a disease is selected from the group consisting of a cancer, angiogenesis, diabetic retinopathy, or non-syndromic intellectual disability.

23. The method of claim 20, wherein said method is for treating a disease 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.

24. The method of claim 20, wherein said molecule is an inhibitor having one or more protein-membrane interactions and inhibiting a mammal BRAG2, and wherein said method is for the treatment of a cancer.

25. The molecule according to claim 3, wherein the alkyl is methyl (Me) or ethyl (Et); and/or the O-alkyl is OMe or OEt; and/or the alkyne, is —CCH; and/or the O-alkyne is —OCH2-CCH.

26. The molecule according to claim 4, wherein the alkyl is methyl (Me) or ethyl (Et); and/or the O-alkyl, is OMe or OEt; and/or the alkyne is —CCH; and/or the O-alkyne is —OCH2-CCH.

Description

[0090] In the figures:

[0091] FIG. 1. Bragsin2 affects Arf pathways in cells

[0092] a. Chemical structures of Bragsin1 and Bragsin2 and derivatives used in this study. Chemical synthesis and structural characterization of the compounds are described in example 6.

[0093] b. Bragsin2 disperses the TGN46 and GM130 markers. HeLa cells were treated with either DMSO (0.25%) or Bragsin2 (50 μM) for 30 min, immunostained for TGN46, GM130 or EEA1 (green channel) and analysed by confocal microscopy. The effect of Bragsin1 treatments are shown in FIG. 11.

[0094] c. Dispersion of GM130 and TGN46 by Bragsin2 is reversible. HeLa cells were treated as in FIG. 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 Bragsin1 is shown in FIG. 12.

[0095] 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 Bragsin1 phenotype by constitutively active Arf mutants is shown in FIG. 13. Scale bars, 10 m.

[0096] FIG. 2. Bragsin is a specific inhibitor of the ArfGEF BRAG2

[0097] a. Bragsin1 has no effect on the Sec7 domain of human ArfGEFs in solution. Nucleotide exchange kinetics were measured by fluorescence kinetics in the presence of Bragsin1 (50 μM) or DMSO using purified Sec7 domains and a truncated version of Arf1 (D17Arf1), which can be activated in solution. Representative kinetic profiles are given in FIG. 14.

[0098] b. Bragsin1 specifically inhibits BRAG2 in the presence of liposomes. Nucleotide exchange kinetics were determined in the presence of Bragsin1 (50 μM) or DMSO with ArfGEF constructs carrying membrane-binding domains and myristoylated Arf1. Rac1 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 FIG. 15.

[0099] c. Bragsin1 inhibits the activation of myristoylated Arf6 by BRAG2 on liposomes. Experiments were carried out as in FIG. 2b. Representative kinetic profiles are given in FIG. 16.

[0100] d. Dose-response of Bragsin1 and Bragsin2 towards myristoylated Arf1 and BRAG2 on liposomes. Experiments were carried out as in FIG. 2b.

[0101] e. Silencing of BRAG2 phenocopies the effect of Bragsin2 on the TGN46 compartment. HeLa cells were treated with DMSO or Bragsin2 (50 μM) 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 FIG. 17. Scale bars, 10 μm. Statistical analysis of the dispersal of TGN46 staining is shown in FIG. 19.

[0102] 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 μM). Immunofluorescence staining of TGN46 is in green (white-grey on the black-white figures).

[0103] FIG. 3. Bragsin binds to the PH domain of BRAG2

[0104] a. Crystallographic structure of Bragsin1 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.

[0105] b. Interactions of Bragsin1 with residues of the PH domain. Hydrogen bonds are shown in dotted lines.

[0106] c. Analysis of the inhibition of BRAG2 mutants by Bragsin2. GEF efficiencies were measured by fluorescence kinetics using myristoylated Arf1 as in FIG. 2b.

[0107] d. Structure-activity relationship analysis of Bragsin analogs. All compounds were used at 20 mM. GEF activities were measured by fluorescence kinetics using myristoylated Arf1 as in FIG. 2b.

[0108] 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 FIG. 1b.

[0109] FIG. 4. Bragsin is a non-competitive inhibitor of protein-membrane interactions

[0110] a. Bragsin1 (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.

[0111] 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.

[0112] c. Inhibition of BRAG2 by Bragsin2 is stronger in the presence of liposomes that contain PIP.sub.2 lipids. Kinetics were measured as in FIG. 2b.

[0113] FIG. 5. Bragsin affects breast cancer stem cells

[0114] a. Evolution of the proportion of ALDH.sup.br cells after treatment with Bragsin2 using the Aldefluor assay. Results are expressed as mean±SD.

[0115] 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.

[0116] FIG. 6. A model for protein-membrane interfacial inhibition by Bragsin

[0117] FIGS. 7-9. Analysis of the stability of Bragsin1 and Bragsin2

[0118] FIG. 7. Bragsin1 that has been kept in a stock solution in DMSO for several months lost activity towards BRAG2 (not shown). 1H-NMR analysis shows that it completely decomposed to hydrated compound (13)

[0119] FIG. 8-9. Mass spectroscopy analysis shows Bragsin1 is less stable than Bragsin2. Bragsin1 and Bragsin2 were dissolved in PBS. Bragsin1 slowly decomposed to hydrolyzed compound within 72 hours (FIG. 8) while Bragsin2 remained stable (Fig. c).

[0120] FIGS. 10-13: Bragsin1 has the same cellular effects as Bragsin2

[0121] 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 μM) for 30 min, immunostained for α-tubulin (green channel) or F-actin (magenta channel) and analysed by confocal microscopy.

[0122] FIG. 11: Bragsin1 disperses the TGN46 and GM130 markers. HeLa cells were treated as in FIG. 10 with 50 μM Bragsin1 and immunostained for TGN46 or GM130 (green channel) prior to confocal microscopy analysis.

[0123] FIG. 12: Dispersion of GM130 and TGN46 by Bragsin1 is reversible. After treatment with either DMSO (0.25%) or Bragsin1 (50 μM) 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.

[0124] FIG. 13: Expression of constitutively active Arf1, Arf5 and Arf6 constructs rescues the effect of Bragsin1. 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 μm.

[0125] FIGS. 14-19: Bragsin is a specific inhibitor of BRAG2

[0126] 14: Representative fluorescence kinetic traces of the experiments shown in FIG. 2a. The GDP/GTP nucleotide exchange of N-terminally truncated Arf-GDP was followed in solution by tryptophan fluorescence in the presence of 50 μM of Bragsin1 (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.

[0127] 15: Representative fluorescence kinetic traces of the experiments shown in FIG. 2b. .sup.myrArf1 or Rac1 activation by catalytic amount of the indicated GEF was performed as in FIG. 14 except that the experiments contained 100 μM liposomes (see material and methods for details; example 7).

[0128] 16: Representative fluorescence kinetic traces of the experiments shown in FIG. 2c. .sup.myrArf6 activation was performed in the presence of liposomes as described in FIG. 15.

[0129] 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 (siGBF1, 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%.

[0130] 19: Statistical analysis of the dispersal of TGN46 staining of the experiments shown in FIG. 2e. Area of cells (polygon) and TGN (minimal ellipse) were determined manually after sum Z projection of stacks and measured using Fiji/ImageJ. 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.

[0131] FIGS. 20-21:

[0132] 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.sup.Sec7PH construct and myristoylated Arf1 (see material and methods for details).

[0133] 21: Bragsin3 disperses the TGN46 marker. HeLa cells were treated as in FIG. 1b with 50 mM Bragsin3 and immunostained for TGN46 prior confocal microscopy analysis.

[0134] FIGS. 22-29:

[0135] 22: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line MDA-MB-231.

[0136] 23: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line MCF7.

[0137] 24: Shows the viability of Bragsine 2, Compound 14 and Compound 3 on Cell line SUM149.

[0138] 25: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line SUM159.

[0139] 26: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line A549.

[0140] 27: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line U87-MG.

[0141] 28: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line PANC-1.

[0142] 29: Shows the viability and cytotoxicity of Bragsine 2, Compound 14 and Compound 3 on Cell line.

[0143] Viability tests (measure of the quantity of ATP after cell lysis, CelltiterGlo, Promega) and cytotocixity (measure of LDH after cell lysis, CytoTox-One, Promega.

[0144] Results in FIGS. 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.

[0145] Bragsin2: 6-methoxy-5-nitro-2-(trifluoromethyl)-4H-chromen-4-one

[0146] Compound 14 (2MetO): 6,8-dimethoxy-5-nitro-2-(trifluoromethyl)-4H-chromen-4-one

[0147] Compound 3 (—NO.sub.2): 6-methoxy-2-(trifluoromethyl)-4H-chromen-4-one (negative control).

[0148] The present invention is further illustrated by means of the following examples.

[0149] 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.

[0150] 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.

[0151] 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.

[0152] Thus, every example has a general scope.

[0153] 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

[0154] A small molecule Bragsin1 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 (FIG. 1a). 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 Arf GTPases in Cells

[0155] Bragsin1 may affect Arf pathways in cells because Bragsin 1 perturbs the function of the yeast ArfGEF Sec7p. Bragsin1 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 (FIGS. 7-9). In initial experiments Bragsin2 was used to assess possible effects on the structure of the cis-Golgi and trans-Golgi network (TGN) compartments, which are regulated by Arf-dependent pathways. In HeLa cells, the cis-Golgi marker GM130 and the TGN marker TGN46 was dispersed by Bragsin2 into punctate structures of heterogeneous sizes (FIG. 1b), and this effect was reversible (FIG. 1c). In contrast, Bragsin2 had no effect on the early endosome marker EEA1 (FIG. 1b) nor did it affect tubulin and actin networks (FIG. 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 (FIG. 1d). The phenotypes of cells treated with freshly prepared Bragsin1 were identical to those treated with Bragsin2 (FIG. 11), were reversible (FIG. 12) and rescued by expression of constitutively active Arf GTPases (FIG. 13). Together, these data suggest that Bragsin1 and Bragsin2 inhibit Arf pathways in cells, in agreement with the original activity observed in a chemogenomic screen of Bragsin1 towards the yeast ArfGEF Sec7p.

Example 2—Bragsin is a Specific Inhibitor of BRAG2 that Necessitates the Presence of Membranes for Inhibition

[0156] 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. I. & 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)).

[0157] The effect of Bragsin1 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, Bragsin1 had no effect on the activation of Arf1 by any of these Sec7 domains (FIGS. 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, Bragsin1 strongly inhibited BRAG2 on liposomes, while it had no effect on the other ArfGEFs (FIGS. 2b and 15) or on Rac1/Trio, an unrelated small GTPase/GEF system (FIGS. 2b and 15). BRAG2 activates several Arf isoforms, including the plasma membrane-located Arf6 isoform. Bragsin1 also inhibited the BRAG2-mediated activation of myristoylated Arf6, indicating that inhibition is independent of the Arf isoform used (FIGS. 2c and 16). Bragsin1 and Bragsin2 have the same IC.sub.50 towards BRAG2 (3 mM) according to dose-response analysis, confirming that Bragsin1 and Bragsin2 can be used interchangeably (FIG. 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 (FIGS. 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 (FIG. 2f). Bragsin is a specific inhibitor of the ArfGEF BRAG2 in vitro and in cells according to these experiments. These experiments identify a mechanism of inhibition that requires the membrane to manifests itself.

Example 3—Bragsin Binds to the PH Domain of BRAG2

[0158] In vitro and cellular assays support that the BRAG2.sup.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.sup.Sec7-PH Bragsin1 complex (Table 1). Unambiguous electron density was observed near the PH domain in which the inhibitor could be modeled (FIG. 3a), and its position was not involved in crystal contacts. Bragsin1 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 (FIG. 3b). CF.sub.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 Bragsin1. Conversely, hydration of Bragsin1 leading to a sp3 geometry and a different orientation of the carbon bearing the CF.sub.3 substituent, was in agreement with the loss of activity of Bragsin1 over time (see above). R654E or R681E mutations were introduced in the PH domain of BRAG2 to confirm the location of the binding site, the. Inhibition was affected for both mutants (FIG. 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 NO.sub.2 or CF.sub.3 functional groups, which establish interactions with BRAG2 in the crystal, were removed or modified (compounds (3) to (6), FIG. 1a and Example 6). All analogs displayed reduced inhibition of BRAG2 in vitro (FIG. 3d) and were not capable to disperse TGN46 (compounds 3, 4, and 6) or had less effect (compound 5) in cells (FIG. 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

[0159] 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))(FIG. 4a). Bragsin inhibition could act by blocking the interaction of BRAG2 with membranes. Surprisingly, Bragsin2 did not displace BRAG2 from PIP.sub.2-containing liposomes, showing that it does not inhibit BRAG2 by disrupting its association with the membrane (FIG. 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.sub.2 compared to liposomes that contain PS as the sole negatively charged lipids (FIG. 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

[0160] 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.sup.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 μM) depleted the bCSC population in two of the three cell lines tested (SUM149 and S68) (FIG. 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.sup.br population (FIG. 5b). In contrast, the SUM159 cell line, whose bCSC ALDH.sup.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 Bragsin1, Bragsin 2 and Analogs Used in this Study (FIG. 1a)

[0161] 6.1. General Information

[0162] 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 F.sub.254. 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.). .sup.1H NMR and .sup.13C 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, δ), referenced to the solvent peak of CDCl.sub.3, defined at δ=7.26 ppm (.sup.1H NMR) and δ=77.16 (.sup.13C NMR). Coupling constants are quoted in Hz (J). .sup.1H 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).

[0163] 6.2. Experimental Procedures

[0164] Synthesis of Bragsin1, Bragsin2 and Compound (3)

##STR00007##

[0165] 6.2.1. General Procedure

[0166] Step 1 and 2

[0167] Dry THE (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 THE 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 HCl. The reaction mixture was extracted with EtOAc three times and the organic layers were combined, washed with brine, dried over MgSO.sub.4, filtered, and concentrated under reduced pressure vacuo. The product was used in the next step without further purifications.

[0168] A drop of a 37% aq. HCl 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.

[0169] Step 3

[0170] A mixture of concentrated H.sub.2SO.sub.4 (0.6 ml) and concentrated HNO.sub.3 (0.6 ml) was added to a solution of purified product from step 2 (1 equiv) in conc. H.sub.2SO.sub.4 (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 MgSO.sub.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 (Bragsini, FIG. 1a)

[0171] ##STR00008##

[0172] The title compound was prepared from commercially available 2′-hydroxy-5′-methylacetophenone, according to general procedure. Bragsin1 was obtained in 11% yield (in 3 steps). Data is in accordance with the literature. .sup.1H-NMR (500 MHz, CDCl.sub.3): δ 7.71 (d, J=8.9 Hz, 1H), 7.63 (d, J=8.9 Hz, 1H), 6.74 (s, 1H), 2.39 (s, 3H).

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

[0173] ##STR00009##

[0174] 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. .sup.1H-NMR (500 MHz, CDCl.sub.3): δ 7.71 (d, J=9.5 Hz, 1H), 7.54 (d, J=9.5 Hz, 1H), 6.70 (s, 1H), 3.98 (s, 3H).

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

[0175] ##STR00010##

[0176] 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. .sup.1H-NMR (300 MHz, CDCl.sub.3): δ 7.54 (d, J=3.1 Hz, 1H), 7.48 (d, J=9.2 Hz, 1H), 7.34 (dd, J=9.2, 3.1 Hz, 1H), 6.70 (s, 1H), 3.91 (s, 3H).

Quality Analysis of Compound (4) (Compound 4, FIG. 1a)

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

[0177] ##STR00011##

[0178] The title compound is commercial available compound (CAS: 354128-13-5). .sup.1H-NMR (500 MHz, CDCl.sub.3): δ 7.88 (d, J=6.0 Hz, 1H), 7.00 (s, 1H), 6.37 (d, J=6.0 Hz, 1H), 4.03 (s, 3H), 2.37 (s, 3H).

[0179] Synthesis of Compound (5) (Compound 5, FIG. 1a)

##STR00012##

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

[0180] ##STR00013##

[0181] 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.sub.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 7 water, extracted with DCM, washed with NaHCO.sub.3 and dried over MgSO.sub.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). .sup.1H-NMR (300 MHz, CDCl.sub.3): δ 7.84 (d, J=3.0 Hz, 1H), 7.50 (d, J=9.2 Hz, 1H), 7.38 (dd, J=9.2, 3.0 Hz, 1H), 6.75 (s, 1H). .sup.13C-NMR (75 MHz, CDCl.sub.3): δ 178.2, 155.3, 152.9 (q, J=39.2 Hz), 150.4, 125.2, 124.7, 120.1, 118.7 (q, J=274.2 Hz), 109.4 (q, J=2.8 Hz), 109.2. HRMS (ESI): calculated for C.sub.10H.sub.6F.sub.3O.sub.3 [M+H].sup.+: 231.0259, found: 231.0264.

6-hydroxy-4-oxo-2-(trifluoromethyl)-4H-chromene-5-carbaldehyde (8)

[0182] ##STR00014##

[0183] 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.sub.2O and dried over MgSO.sub.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). .sup.1H-NMR (300 MHz, CDCl.sub.3): δ 13.09 (s, 1H), 11.43 (s, 1H), 7.75 (d, J=9.4 Hz, 1H), 7.43 (d, J=9.4 Hz, 1H), 6.78 (s, 1H). .sup.13C-NMR (75 MHz, CDCl.sub.3): δ 198.6, 178.1, 162.5, 151.3 (q, J=39.7 Hz), 150.3, 128.0, 127.2, 122.4, 118.5 (q, J=273.9 Hz), 115.3, 111.8 (q, J=2.7 Hz). HRMS (ESI): calculated for C.sub.11H.sub.6F.sub.3O.sub.4 [M+H].sup.+: 259.0218, found: 259.0212.

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

(Compound 5, FIG. 1a)

[0184] ##STR00015##

[0185] Under argon, compound 8 (50 mg, 0.19 mmol, 1.0 equiv.) was dissolved in dry DMF followed by the addition of K.sub.2CO.sub.3 (80 mg, 0.582 mmol, 3.0 equiv.) at 0° C. Then, methyl iodide (15 μL, 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 NH.sub.4Cl, extracted with ethyl acetate, washed with brine and dried over MgSO.sub.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). .sup.1H-NMR (300 MHz, CDCl.sub.3): δ 10.57 (s, 1H), 7.67 (d, J=9.3 Hz, 1H), 7.46 (d, J=9.3 Hz, 1H), 6.67 (s, 1H), 3.91 (s, 3H). .sup.13C-NMR (75 MHz, CDCl.sub.3): δ 191.9, 177.5, 154.7, 152.5 (q, J=39.4 Hz), 149.6, 126.0, 123.6, 121.9, 120.0, 118.6 (q, J=274.2 Hz), 110.0 (q, J=2.8 Hz), 57.1. HRMS (ESI): calculated for C.sub.12H.sub.8F.sub.3O.sub.4 [M+H].sup.+: 273.0374, found: 273.0369.

[0186] Synthesis of Compound (6) (Compound 6, FIG. 1a)

##STR00016##

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

[0187] ##STR00017##

[0188] 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 THE (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. .sup.1H-NMR (300 MHz, CDCl.sub.3): δ 11.91 (s, 1H), 7.17 (d, J=9.2 Hz, 1H), 7.04 (d, J=9.2 Hz, 1H), 3.95 (s, 3H), 3.81 (s, 3H), 2.52 (s, 3H). .sup.13C-NMR (75 MHz, CDCl.sub.3): δ 203.8, 168.6, 156.5, 149.1, 124.0, 121.3, 121.0, 117.5, 57.7 53.2, 29.3. HRMS (ESI): calculated for C.sub.11H.sub.12O.sub.5Na [M+Na].sup.+: 247.0582, found: 247.0580.

methyl 6-methoxy-4-oxo-2-(trifluoromethyl)-4H-chromene-5-carboxylate (12)

[0189] ##STR00018##

[0190] The title compound was prepared from benzoquinone 11, according to general procedure. Compound 12 was obtained in 85% yield. .sup.1H-NMR (300 MHz, CDCl.sub.3): δ 7.63 (d, J=9.4 Hz, 1H), 7.42 (d, J=9.4 Hz, 1H), 6.65 (s, 1H), 4.01 (s, 3H), 3.93 (s, 3H). 1.sup.3C-NMR (75 MHz, CDCl.sub.3): δ 176.0, 166.9, 154.2, 152.0 (q, J=39.4 Hz), 149.6, 121.9, 120.7, 120.4, 119.0, 118.6 (q, J=274.5 Hz), 109.9 (q, J=2.7 Hz), 57.0, 53.3. HRMS (ESI): calculated for C.sub.13H.sub.10F.sub.3O.sub.5[M+H].sup.+: 303.0430, found: 303.0475.

6-methoxy-4-oxo-2-(trifluoromethyl)-4H-chromene-5-carboxylic acid (6)

(Compound 6, FIG. 1a)

[0191] ##STR00019##

[0192] 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.sub.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 NaHCO.sub.3 and dried over MgSO.sub.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. .sup.1H-NMR (300 MHz, CDCl.sub.3): δ 7.56 (d, J=9.3 Hz, 1H), 7.40 (d, J=9.3 Hz, 1H), 6.69 (s, 1H), 3.96 (s, 3H). .sup.13C-NMR (75 MHz, CDCl.sub.3): δ 175.6, 168.8, 154.8, 151.7 (q, J=39.3 Hz), 150.2, 125.3, 122.9, 122.8, 118.6 (q, J=274.3 Hz), 113.3, 110.7 (q, J=2.7 Hz), 53.3. HRMS (ESI): calculated for C.sub.12H.sub.8F.sub.3O.sub.5[M+H].sup.+: 289.0324, found: 289.0317.

2-hydroxy-6-methyl-5-nitro-2-(trifluoromethyl)chroman-4-one (13)

[0193] ##STR00020##

[0194] Data is in accordance with the literature .sup.1H-NMR (300 MHz, CDCl.sub.3): δ 7.49 (d, J=8.6 Hz, 1H), 7.15 (d, J=8.6 Hz, 1H), 5.71 (s (br), 1H, OH), 3.10 (q, J=9.6 Hz, 2H), 2.24 (s, 3H).

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

[0195] Step 1

[0196] Charge Pd(OAc).sub.2 (4.4 mg, 0.02 mmol), K.sub.2S.sub.2O.sub.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-3Ê.sup.1,5Ê.sup.1-dimethoxyacetophenone.

##STR00021##

[0197] .sup.1H-NMR (400 MHz, CDCl.sub.3): δ 12.21 (s, 1H), 6.69 (d, J=0.7 Hz, 2H), 3.86 (s, 3H), 3.79 (s, 3H), 2.61 (s, 3H)

[0198] General Procedure

[0199] Step 2

[0200] 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 THE 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 HCl. The reaction mixture was extracted with EtOAc three times and the organic layers were combined, washed with brine, dried over MgSO.sub.4, filtered, and concentrated in vacuo. The product was used in the next step without further purifications.

[0201] Step 3

[0202] One drop of conc. HCl 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.

##STR00022##

[0203] Step 4 (Nitration)

[0204] A mixture of conc. H2SO4 (0.6 ml) and conc. HNO3 (0.6 ml) was added to a solution of purified product from step 2 (1 equiv) in conc. H2SO4 (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.

##STR00023##

Example 7—Material and Methods

[0205] 7.1. Chemicals.

[0206] Nucleotides were purchased from Jena Bioscience. BFA was from Sigma. Bragsin1 (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 Bragsin1 and Bragsin2 is shown in FIGS. 7-9. Bragsin1 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). Bragsin1 and Bragsin2 can be used indifferently. Synthesis of Bragsin analogs shown in FIG. 1a is described in example 6.

[0207] 7.2. Antibodies and cDNAs.

[0208] 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-α-tubulin was from Sigma. Sheep antibody against TGN46 (trans-Golgi 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 Arf1 Q71 L, Arf5 Q71 L and Arf6 Q67L mutants were kindly provided by Julie Ménétrey (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.sup.Sec7PH mutants were generated by site-directed mutagenesis using Quickchange kit (Stratagene) as per manufacturer's protocol.

[0209] 7.3. Cell Culture, Transfection and Treatment with Inhibitors.

[0210] 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 3TC.

[0211] 7.4. Immunofluorescence and Confocal Microscopy.

[0212] 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 USA 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 100×(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 μm and processed using Fiji/ImageJ (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.

[0213] 7.5. siRNA Knockdown.

[0214] Gene silencing was achieved using siRNAs targeting all the known transcript variants of each GEF gene: siBRAG2, Hs_IQSEC1_5 (Qiagen, S103019408); siARNO, Hs_PSCD2_3 (Qiagen, S100061299); siGBF1, Hs_GBF1_3 (Qiagen, S100425418). Nontargeting control siRNA: siCTRL, AllStars 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.sup.2 into the wells of a 12-well plate, each containing the transfection mix: 10 pmol siRNA and 3 μl Lipofectamine® RNAiMAX in 200 μl 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 μl/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 (FIG. 17) and these levels were not modified by treatment with Bragsin (FIG. 18). Total RNA was extracted with the RNeasy® Mini Kit (Qiagen, 74104) following the manufacturer's instructions. Each sample RNA (1 μg) 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′-ACCCGAGAATTGATGAGTCG-3′ (SEQ ID NO:3); ARNO, 5′-TTGTGTCAAGGATAGGGCTG-3′ (SEQ ID NO:4) and 5′-ACTTCACCTTCATAGAGGCG-3′ (SEQ ID NO:5); GBF1, 5′-TGTCACTCTCTACCTTTGCG-3′ (SEQ ID NO:6) and 5′-AAATCTCCGCTGTGTCCATC-3′ (SEQ ID NO:7); GAPDH, 5′-ACAAGAGGAAGAGAGAGACCC-3′ (SEQ ID NO:8) and 5′-TACATGACAAGGTGCGGCTC-3′ (SEQ ID NO:9).

[0215] 7.6. Proteins.

[0216] Bovine Δ17Arf1 and human A13Arf6, full-length myristoylated Arf1 and Arf6, human BRAG2.sup.Sec7(390-594) and BRAG2.sup.Sec7PH (390-811), human EFA6.sup.Sec7(527-727) and EFA6.sup.Sec7PHCt (527-1024), ARNO.sup.Sec7 (50-256), BIG1.sup.Sec7 (691-889), BIG1.sup.DcbHusSec7 (2-888) and ARNO.sup.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.sup.Sec7PH mutants were purified as the wild type protein. Purification of human full-length Rac1 carrying a 6×His tag in C-terminus and human TRIO.sup.DH1PH1 (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)).

[0217] 7.7. Liposomes and Flotation Assay.

[0218] 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 μm. For the Arf GEF specificity studies, liposomes contained 48% phosphatidylcholine (PC), 20% phosphatidylethanolamine (PE), 30% phosphatidylserine (PS) and 2% phosphatidylinositol-4,5-bisphosphate (PIP.sub.2). For Rac activation assay, liposomes contained 43% PC, 20% PE, 10% PS, 20% cholesterol, 2% PIP.sub.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% PIP.sub.2, 20% cholesterol and 0.1% NBD-PE. Flotation assays were carried out as in.sup.18.

[0219] 7.8. Nucleotide Exchange Assays.

[0220] 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.sup.50,51 For specificity assays, 50 μM of Bragsin1 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.sub.2 and 1 mM DTT) with either 1 μM of N-terminally truncated Arf-GDP and 100 nM of indicated Sec7 domain (in solution) or 0.4 μM of .sup.myrArf1 in the presence of 100 μM 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 (k.sub.obs) were determined from monoexponential fits and means are given as the percentage of control activity ±SD. All experiments were done in triplicate.

[0221] 7.9. Crystallization and Structure Determination.

[0222] 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-HCl pH 8.5. Crystals were transferred to the reservoir solution supplemented with 10% glycerol and soaked with 20 μM Bragsin1 in a final volume of 100 μl 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 (2011)). 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 USA 114, 11416-11421 (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.

[0223] 7.10. Cancer Stem Cell Assays

[0224] 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

[0225] 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.

[0226] 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.

[0227] 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.

[0228] PIP.sub.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 USA 114, 11416-11421 (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 FIG. 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, 99-111 (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.

[0229] 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 stemness of breast cancer cells.

[0230] 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-US-00001 TABLE 1 Crystal structure of Brag2/Bragsin1 complex. Crystallographic statistics. X-ray source Proxima 2, synchrotron SOLEIL Data collection: Space group P212121 Cell dimensions a,b,c  (Å) 65.92, 66.16, 218.79 a, b, g (°) 90, 90, 90 Wavelength (Å) 0.980105 Resolution (Å) 65.9-2.50 (2.78-2.50) R.sub.sym (%) 18 (142) R.sub.pim(%) 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 (11487) Refinement: R.sub.work/R.sub.free 0.22/0.27 Nb. Atoms Protein 5641 Water 18 B-factors Protein 66 Bragsin1 74 Water 43 R.M.S. deviations Bond lengths (Å) 0.009 Bond angles (°) 1.136 PDB ID 6FNE

TABLE-US-00002 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 OX = 9606 GN = IQSEC1 PE = 1 SV = 1 MWCLHCNSERTQSLLELELDSGVEGEAPSSETGTSLDSPSAYPQGPLVPGSSLSPDHY EHTSVGAYGLYSGPPGQQQRTRRPKLQHSTSILRKQAEEEAIKRSRSLSESYELSSDLQ DKQVEMLERKYGGRLVTRHAARTIQTAFRQYQMNKNFERLRSSMSENRMSRRIVLSNM RMQFSFEGPEKVHSSYFEGKQVSVTNDGSQLGALVSPECGDLSEPTTLKSPAPSSDFA DAITELEDAFSRQVKSLAESIDDALNCRSLHTEEAPALDAARARDTEPQTALHGMDHRKL DEMTASYSDVTLYIDEEELSPPLPLSQAGDRPSSTESDLRLRAGGAAPDYWALAHKEDK ADTDTSCRSTPSLERQEQRLRVEHLPLLTIEPPSDSSVDLSDRSERGSLKRQSAYERSLG GQQGSPKHGPHSGAPKSLPREEPELRPRPPRPLDSHLAINGSANRQSKSESDYSDGDN DSINSTSNSNDTINCSSESSSRDSLREQTLSKQTYHKEARNSWDSPAFSNDVIRKRHYRI GLNLFNKKPEKGVQYLIERGFVPDTPVGVAHFLLQRKGLSRQMIGEFLGNRQKQFNRDV LDCVVDEMDFSTMELDEALRKFQAHIRVQGEAQKVERLIEAFSQRYCICNPGVVRQFRN PDTIFILAFAIILLNTDMYSPNVKPERKMKLEDFIKNLRGVDDGEDIPREMLMGIYERIRKRE LKTNEDHVSQVQKVEKLIVGKKPIGSLHPGLGCVLSLPHRRLVCYCRLFEVPDPNKPQKL GLHQREIFLFNDLLVVTKIFQKKKNSVTYSFRQSFSLYGMQVLLFENQYYPNGIRLTSSVP GADIKVLINFNAPNPQDRKKFTDDLRESIAEVQEMEKHRIESELEKQKGVVRPSMSQCSS LKKESGNGTLSRACLDDSYASGEGLKRSALSSSLRDLSEAGKRGRRSSAGSLESNVEFQ 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 OX = 9606 GN = IQSEC1 PE = 1 SV = 1 MACRRRYFVEGEAPSSETGTSLDSPSAYPQGPLVPGSSLSPDHYEHTSVGAYGLYSGP PGQQQRTRRPKLQHSTSILRKQAEEEAIKRSRSLSESYELSSDLQDKQVEMLERKYGGR LVTRHAARTIQTAFRQYQMNKNFERLRSSMSENRMSRRIVLSNMRMQFSFEGPEKVHS SYFEGKQVSVTNDGSQLGALVSPECGDLSEPTTLKSPAPSSDFADAITELEDAFSRQVKS LAESIDDALNCRSLHTEEAPALDAARARDTEPQTALHGMDHRKLDEMTASYSDVTLYIDE EELSPPLPLSQAGDRPSSTESDLRLRAGGAAPDYWALAHKEDKADTDTSCRSTPSLERQ EQRLRVEHLPLLTIEPPSDSSVDLSDRSERGSLKRQSAYERSLGGQQGSPKHGPHSGAP KSLPREEPELRPRPPRPLDSHLAINGSANRQSKSESDYSDGDNDSINSTSNSNDTINCSS ESSSRDSLREQTLSKQTYHKEARNSWDSPAFSNDVIRKRHYRIGLNLFNKKPEKGVQYLI ERGFVPDTPVGVAHFLLQRKGLSRQMIGEFLGNRQKQFNRDVLDCVVDEMDFSTMELD EALRKFQAHIRVQGEAQKVERLIEAFSQRYCICNPGVVRQFRNPDTIFILAFAIILLNTDMY SPNVKPERKMKLEDFIKNLRGVDDGEDIPREMLMGIYERIRKRELKTNEDHVSQVQKVEK LIVGKKPIGSLHPGLGCVLSLPHRRLVCYCRLFEVPDPNKPQKLGLHQREIFLFNDLLVVT KIFQKKKNSVTYSFRQSFSLYGMQVLLFENQYYPNGIRLTSSVPGADIKVLINFNAPNPQD RKKFTDDLRESIAEVQEMEKHRIESELEKQKGVVRPSMSQCSSLKKESGNGTLSRACLD DSYASGEGLKRSALSSSLRDLSEAGKRGRRSSAGSLESNVEGSIISSPHMRRRATSTRE CPSRPHQTMPNSSSLLGSLFGSKRGKPPPQAHLPSAPALPPPHPPVVLPHLQHSVAGH HLGPPEGLPQAAMHGHHTQYCHMQNPPPYHHHHHHHPPQHIQHAHQYHHGPHGGHP AYGAHAHGHPPLPSAHVGHTVHHHGQPPAPPPPTSSKAKPSGISTIVSAHVGHTVHHHG QPPAPPPPTSSKAKPSGISTIV