Inhibitors of RAC1 and uses thereof for inducing bronchodilatation

Abstract

The present invention concerns a compound having the following formula (I): wherein: —A is in particular —N(R′.sub.a)—C(═O)—R, R′.sub.a being H or a (C.sub.1-C.sub.6)alkyl group, and R being preferably a group having the following formula (II): —X is in particular chosen from the group consisting of: —SO.sub.2—N(R′.sub.b)—, R′.sub.b being H or a (C.sub.1-C.sub.6)alkyl group, —N(R″.sub.b)—SO.sub.2—, R″b being H or a (C.sub.1-C.sub.6)alkyl group, —CO—NH—, and —NH—CO—, for use for the treatment of pathologies characterized by bronchoconstriction, such as asthma. ##STR00001##

Claims

1. A method for treating a pathology characterized by bronchoconstriction, comprising administering to a mammal in need thereof a therapeutically effective amount of a compound having the following formula (VI): ##STR00064## wherein: p is an integer from 1 to 3 inclusive, X′ is chosen from the group consisting of: —S—, —NH—, —NR.sub.d—, —CH.sub.2—, —SO.sub.2—, and —SO—, R.sub.d being H or a (C.sub.1-C.sub.6)alkyl group; q is 0 or is an integer from 1 to 5 inclusive, the R.sub.3 groups, identical or different, are chosen from the group consisting of: (C.sub.1-C.sub.6)alkyl groups, halogen atoms, (C.sub.1-C.sub.6)alxoxy groups, (C.sub.1-C.sub.6)thioalkyl groups, and —NR.sub.aR.sub.b groups, R.sub.a and R.sub.b, identical or different, being H or a (C.sub.1-C.sub.6)alkyl group; m is 0 or is an integer from 1 to 5 inclusive, the R.sub.2 groups, identical or different, are chosen from the group consisting of: halogen atoms, (C.sub.1-C.sub.6)alkyl groups, (C.sub.1-C.sub.6)alkoxy groups, (C.sub.1-C.sub.6)thioalkyl groups, —SCF.sub.3, —SF.sub.5, and —NR.sub.aR.sub.b groups, R.sub.a and R.sub.b, identical or different, being H or a (C.sub.1-C.sub.6)alkyl group.

2. The method of claim 1, wherein the pathology characterized by bronchoconstriction is asthma.

3. The method of claim 1, wherein the compound has the following formula (VII): ##STR00065## wherein: R.sub.5 is a (C.sub.1-C.sub.6)alkyl group; and the R.sub.4 groups, identical or different, are chosen from the (C.sub.1-C.sub.6)alkyl groups.

4. The method of claim 1, wherein the compound has the formula (VI) wherein X′ is —S— or —CH.sub.2—.

5. The method of claim 1, wherein the compound has the formula (VI) wherein X′ is —S—.

6. The method of claim 1, wherein the compound has the formula (VI) wherein q is 0 or 1, and the R.sub.3 groups, identical or different, are chosen from the group consisting of: (C.sub.1-C.sub.6)alkyl groups, and (C.sub.1-C.sub.6)alxoxy groups.

Description

FIGURES

(1) FIG. 1 relates to the inhibition of Rac activation by A4.1 (or compound (2)). It represents immunoblot analysis of Rac-GTP and Rac total expression in fibroblasts stimulated by a Rac activator and pre-incubated or not with NSC23766 or (2) for 1 h.

(2) FIG. 2 relates to the analysis of A4.1 selectivity. FIG. 2A: Representative surface plasmon resonance (SPR) sensograms. (2) was injected at 0.78, 1.56, 3.1, 6.25, 12.5 and 25 μM into sensor chip coated by indicated purified small GTPase. When indicated, EDTA was added in the running buffer. n>3. FIG. 2B: Representative real-time kinetics of nucleotide exchange assay of indicated small GTases. The increase of mant-GTP or the decrease of man-GDP was recorded in presence or absence of (2) (10 μM). n=3.

(3) FIGS. 3 to 5 relate to the inhibition of the Rac-induced cell functions by (2). FIG. 3: (2) blocks Rac activator-induced actin reorganization. 3T3 cells were incubated in serum-free growth medium, either alone, or supplemented with (2) or NSC23766 at indicated concentration 1 h before Rac activation. Ruffles are indicated by arrows (left panel). Percentages of cells with ruffles were quantified (right panel). Results shown are representative of 3 independent experiments. FIG. 4: (2) decreases cell migration. 3T3 cells were incubated or not with (2) or NSC23766. Left panel, representative records with arrows indicating the cell location at different times. Right panel, quantification of cell speed in each experimental conditions. Results shown are representative of 2 independent experiments. FIG. 5: (2) decreases cell adhesion. Upper panel, representative kinetics of fibroblast adhesion pre-treated or not with 10 μm (2) or NSC23766. Lower panel, quantification of cell adhesion. Results shown are representative of 3 independent experiments.

(4) FIG. 6 relates to the inhibition of aSMC contraction by (2). FIG. 6A: Contractile responses to methacholine in bronchi from control mice. When indicated, murine bronchial rings were pre-treated with (2) before methacholine stimulation (n=5-7). FIG. 6B: Analysis of airway reactivity to methacholine challenges by non-invasive (plethysmography) in naive (AP) and ovalbumin-challenged mice (00) treated with (2) or vehicle (PBS) nebulization (n=3).

BIOLOGY RESULTS

(5) Materials and Methods

(6) In silico screening. The structure of Rac1 in complex with GDP was first extracted from the crystal structure of Rac1-GDP complexed with arfaptin (PDB code 114D; Tarricone et al, Nature 2001). Pharmacophore models were created from the binding site of GDP with Rac1 using the Receptor-Ligand Pharmacophore Generation tools within Accelrys Discovery Studio 4.0 (DS4.0) software package.

(7) The pharmacophore model was used as a search query against three dimensional multi-conformational molecular databases. The HitFinder™ collection (14,400 compounds) from Maybridge (www.maybridge.com) and the DIVERSet™-EXP (50,000 compounds) and the DIVERSet™-CL (50,000 compounds) from Chembridge (www.chembridge.com) were used in the virtual screening. For the preparation of ligands, duplicate structures were removed and 3D coordinates were generated. A multi-conformational ligand database was then created using Catalyst within the Build 3D Database tool under DS4.0. The query was performed using the Search 3D Database tool with the FAST search method under DS4.0, retrieving as hits only compounds matching all features of the query.

(8) The docking studies were performed using LigandFit option of receptor-ligand interactions protocol section available in DS4.0. Initially, Rac1 protein was prepared, by adding the hydrogen atoms and removing the water molecules, and then minimized using CHARMm force field. The protein molecule thus prepared was then defined as the total receptor, after removing GDP. The ligand molecules retained by the pharmacophore model were docked into the binding site of the Rac1 and the interaction energies in the form of dock score (Venkatachalam et al, J Mol Graph Model. 2003) between each ligand and the protein were calculated. Docking was performed using CFF as the energy grid. Penality of 200 kcal/mol/atom was set up to reduce the dock score of poses that occurred outside of the binding site. The conformational search of the ligand poses was performed by the Monte Carlo trial method. Maximum internal energy was set at 10000 kcal/mol. A short rigid body minimization was then performed (steepest descent and Broyden Fletcher Goldfarb Shanno (BFGS) minimizations). Ten poses were saved for each ligand after docking and 100 steps of BFGS rigid body minimization were then carried out. Scoring was performed with six scoring functions: LigScore1, Ligscore2 (Krammer et al, J Mol Graph Model. 2005), PLP1, PLP2 (Gehlhaar et al, Chem Biot 1995), PMF (Venkatachalam et al, J Mol Graph Model. 2003; Muegge and Martin, J Med Chem 1999) and Jain (Jain, j Comput-Aided Mol Design 1996). CFF force field was used for LigScore calculations. Best scored compounds were retained based on the calculation of a consensus score and binding energies under DS4.0.

(9) Cell culture, transfection and actin staining. NIH3T3 cells grew up in DMEM (Gibco; Invitrogen) containing 10% foetal bovine serum, 100 units/mL penicillin and 100 μg/mL streptomycin at 37° C. and 5% CO.sub.2. The culture medium was changed every 72 hours.

(10) After treatments, fibroblasts were fixed with 4% paraformaldehyde in PBS, permeabilized in PBS 0.5% Triton X-100, and incubated with 130 μg/mL of FITC-conjugated phalloidin (Sigma) to visualize F-actin. After staining, images were captured by a fluorescence microscope (Nikon). The actin cytoskeleton organization was analyzed to observe Rac1-dependent ruffle formation.

(11) Analysis of Rac1 activity. In NIH3T3 cells lysates, Rac1 activity was evaluated by active Rac immunoprecipitation using anti-Rac-GTP antibody (26903, NewEast Biosciences). The precipitated active Rac was subjected to SDS-PAGE and detected by immunoblot with anti-Rac1 antibody (BD biosciences).

(12) Surface plasmon resonance studies. SPR immobilization was performed at 25° C. Rac1, RhoA and Cdc42 purified proteins were diluted to 5 μg/mL in Na.sup.+ acetate buffer (pH 5.0) and injected into sensor chip CM5 (GE Healthcare) in a Biacore T200 (GE Healthcare) that was activated with NHS/EDC buffer. Approximately 5,000 response units of the purified protein were captured on the biosensor chip. Biosensor chips were blocked by an injection of 1 mM ethanolamine (pH 8.5). The Rac1 biosensor chip was validated by the injection of a dose-response curve of NSC23766 at the start of each experiment. SPR analysis was performed at 25° C. in HBSEP running buffer (5% DMSO). When indicated, EDTA (20 mM) was added in running buffer.

(13) Unidirectional cell migration. Cells (1000/well) were seeded in a 96 well plate with 10 mm fibronectin stripes (CytooPlates Motility, CYTOO) in medium with 1% SVF and allowed to spread for 4 hours before capturing time-lapse images for 24 hours (image/10 minutes) on a Widefield Leica DMI 6000B drove with Metamorph software. Cells speed was measured with ImageJ software.

(14) Cell adhesion assay using impedance technology. Cells (10000/well) were seeded in a 96 well plate microtiter xCELLigence assay plate (E-Plate) (ACEA Biosciences Inc.) and placed on the Real-time xCELLigence Cell Analyzer (Roche Applied Science) platform at 37° C. to measure the “cell index” every 5 min for a period of 6 hours. The cell index unit is defined as (R.sub.n−R.sub.b)/15. R.sub.n is the cell electrode impedance of the well when it contains cells. R.sub.b is the background impedance of the well with the media alone.

(15) Nucleotide exchange assays. Full-length human small GTPases carrying a 6-histidine tag fused to their C-terminus were expressed in E. coli and purified to homogeneity. Small GTPases were loaded with GDP or N-methylanthraniloyl-GDP (GDP/mant-GDP, JenaBiosciences) before nucleotide exchange kinetics experiments.

(16) Nucleotide exchange kinetics were monitored by fluorescence of the mant fluorophore (λexc=360 nm, λem=440 nm) or tryptophan fluorescence (for Arf6; λexc=280 nm, λem=292 nm) using a Cary Eclipse fluorimeter (Varian, Toulouse, France) at 30° C. under stirring. All kinetics assays were carried out in a buffer containing 50 mM Tris at pH 8, 300 mM NaCl, 2 mM MgCl.sub.2, 1 mM DTT and were started by addition of 100 μM N-methylanthraniloy-GTP or GTP (mant-GTP/GTP, JenaBiosciences). Nucleotide exchange kinetics were carried out at a concentration of small GTPases of 1 μM, either without GEF for spontaneous exchange, in the presence of 50 nM GEF for single kobs (s-1) determination. The kobs was determined from single-exponential fit of the fluorescence change. All experiments were carried out at least in triplicate.

(17) Airways reactivity ex vivo. Murine primary bronchi were cleaned, cut in rings and mounted on a multichannel isometric myograph in Krebs-Henseleit physiological solution (118.4 mM NaCl, 4.7 mM KCl, 2 mM CaCl.sub.2, 1.2 mM MgSO.sub.4, 1.2 mM KH.sub.2PO.sub.4, 25 mM NaHCO.sub.3 and 11 mM glucose) at 37° C. under oxygen. A pre-tension of 0.5 mN was applied. We constructed dose-response curves to methacholine (Sigma). When indicated, rings were pre-incubated 1 h before contraction with (2) (or compound A4.1). The wire myograph was connected to a digital data recorder (MacLab/4e, AD Instruments) and recordings were analyzed using LabChart v7 software (AD Instruments).

(18) Animals use and airways responsiveness measurement in vivo. All experimental procedures and animal care were performed in accordance with the European Community Standards on the Care and Use of Laboratory Animals and approved by the local ethics committee (Comité d'Ethique en Expérimentation Animale des Pays de Loire).

(19) Airway responsiveness was assessed in conscious, unrestrained mice using a barometric, whole-body plethysmography (EMKA Technologies) by recording respiratory pressure curves in response to inhaled methacholine (Sigma) at concentrations of 0-40 mg/ml for 1 min. Airway responsiveness was expressed in enhanced pause (Penh) units. The Penh values measured after stimulation were averaged and expressed as the fold-increase over baseline values. When required, the Rac inhibitor (2) was nebulised (300 μl at 5 mM) 10 min before methacholine challenge.

(20) Statistics. All data are expressed as the mean±SEM of sample size n. For multiple comparisons, the non-parametric Kruskal-Wallis test was used followed by Dunns' post-test. For individual comparisons, statistical analysis was performed using non-parametric t-test (Mann-Whitney). Data analysis was performed using the GraphPad Prism software. The threshold for statistical significance was set at P<0.05.

(21) Results

(22) Pharmacophore Modeling and Virtual Screening

(23) The pharmacophore model was built using HBA (hydrogen bond acceptor) and Ring_A (ring aromatic) features. These features were created based on the observation of Rac1/GDP interactions. One HBA was centered on the oxygen atom of the guanine group of GDP and was oriented toward the N atom of residue Ala159 of Rac1. The other HBA feature was centered on the saturated oxygen of the ended phosphate group of the GDP and was oriented toward the centroid of the N atom of residues Gly12, Ala13, Val14, Gly15 and Lys16, and the NZ atom of Lys16 of Rac1. A Ring_A feature was added, centered on the imidazole group of GDP and oriented toward the aromatic group of residue Phe28 of Rac1. Location constraints were defined by spheres with radius of 1.6 and 2.2 Å on the head and tail of the latest features, respectively. Sixteen exclusion spheres were generated automatically, using the Receptor-Ligand Pharmacophore Generation tool of DS4.0. Finally, the pharmacophore model containing all the features described above, was used to search a database of 116,000 chemical compounds using the Search 3D Database tool within DS4.0. The Fit Value threshold was fixed to 1.6 and allowed to extract 9362 compounds for the docking process.

(24) To further reduce the number of compounds to be evaluated in vitro, molecular docking studies were conducted using LigandFit module of Receptor-Ligand Interactions section available under DS4.0. Ligands molecules retained by the pharmacophore-based approach were docked into a binding site defined as the volume filled by GDP in the Rac1/GDP complex. The volume of the binding site was 606 Å3 and contained 4851 points. Among the 9362 compounds retained after the Pharmacophore-based search, 9189 were actually docked to the target. To improve the screening accuracy, a consensus strategy was adopted. The top 20% of the docked database, ranked by at least five of the six scoring functions used, were retained and for compounds with a dock score above 70, binding free energies were calculated after in situ ligand minimization. The ligands were then ranked based on the lowest binding free energy after the withdrawal of the compound poses having a high ligand free energy (threshold 20 kcal/mol). The top 100 were retained to be purchased and evaluated in vitro on Rac-dependent cellular functions: cell adhesion and migration. Thus, the hit (2) was identified as the best potential Rac inhibitor.

(25) Compound (2) is a Potent and Selective Inhibitor of Rac Proteins

(26) The potential of (2) to inhibit Rac activity was first evaluated by pull-down assay. As expected, the level of Rac-GTP is increased in culture cells stimulated by a Rac activator. This activation is prevented by the Rac inhibitor NSC23766 and also by (2) (FIG. 1). These results suggest that the compound (2) is a new Rac inhibitor. To further delineate the selectivity of (2), its interaction with Rho family proteins was examined by surface plasmon resonance (SPR). The SPR sensograms reveal that (2) binds Rac1 but not RhoA or Cdc42 (FIG. 2A), suggesting a selective interaction of (2) for Rac proteins. EDTA chelates Mg.sup.2+ ions and promotes nucleotide release from Rho small GTPases. The interaction Rac1: (2) is significantly increased in Rac1 nucleotide free (FIG. 2A), strengthening the hypothesis that (2) binds Rac1 and docks into the GDP/GTP pocket To proceed with the analysis of (2) selectivity, a screening was conducted with purified proteins representing members of the Rho, Rab and Arf small GTPases subfamilies. The effect of (2) on the small GTPase ability to exchange nucleotide was analyzed by a real-time assay of mant-GTP or mant-GDP binding kinetics. The presence of the small molecule inhibitor (2) decreased the B.sub.max GTP-binding on Rac1 and Rac2. In contrast, (2) did not alter nucleotide exchange kinetics of RhoA, RhoG, Rab35 and Arf6 (FIG. 2B and data not shown). These results suggest that the compound (2) is a selective inhibitor of the small GTPase Rac but without specificity for Rac isoforms.

(27) Compound (2) Inhibits Rac-Dependent Cell Functions

(28) The small GTPase Rac is extensively described to play a crucial role in actin cytoskeleton organization, cell adhesion and migration. To evaluate the ability of (2) to inhibit Rac-mediated cell functions, the actin structures of the cells stimulated by Rac activator was examined in the presence or absence of (2). Rac activator stimulated membrane ruffling in fibroblastes (FIG. 3A). However, in the presence of (2) or NSC23766, the efficiency of Rac activator to induce ruffle is strongly decreased. Interestingly, the dose-dependent inhibition observed in fibroblastes suggest that the small molecule (2) (IC.sub.50=0.67 nM) is a powerful Rac inhibitor compared to NSC23766 (IC.sub.50=2.6 μM). This hypothesis is reinforced by the cell migration (FIG. 3B) and adhesion (FIG. 3C) assays. Indeed, NSC23766 and (2) slow down the migration and adhesion rate of the cells but a higher inhibition is always recorded with cells treated with the compound (2).

(29) These in vitro assays demonstrate that (2) inhibits Rac-dependent cell functions with a higher efficiency than NSC23766.

(30) Compound (2) Prevents Bronchoconstriction and Airway Hyperresponsiveness

(31) Excessive contraction of airways smooth muscle cells (aSMC) is one of the main characteristics of asthma. The degree of this airway hyperresponsiveness (AHR) correlates with asthma severity and the need for therapy. However, the molecular mechanisms leading to AHR are not completely understood. Recently, we unveiled an unexpected and essential role of Rac1 in the regulation of intracellular Ca.sup.2+ and contraction of aSMC, and the development of AHR. Rac1 thus appears as an attractive therapeutic target in asthma, with a combined beneficial action on both bronchoconstriction and pulmonary inflammation. First, the functional impact of (2) in aSMC was studied by measuring the contractile response of bronchial rings from control mice. The maximal contraction induced by the muscarinic receptor agonist methacholine was dose-dependent reduced by (2), suggesting that this small molecule could be used to induce bronchodilation (FIG. 4A). To confirm in vivo the potential therapeutic of (2), the pulmonary resistance was measured in a mouse model of human allergic asthma, induced by percutaneous sensitization and intranasal challenge with house dust mite extract-Dermatophagoides farinae (Der f). Der f sensitization induces AHR that is prevented by acute (2) nebulization (FIG. 4B), suggesting that (2) inhibits in vivo Rac to induce bronchodilation.

(32) In conclusion, the lead molecule (2) is a new selective and potent Rac inhibitor that could open up a new avenue for the treatment of pulmonary pathologies characterized by AHR.