Étude De L'action Du PBRM, Un Inhibiteur De La 17B

Étude de l’action du PBRM, un inhibiteur de la 17β-hydroxystéroïde déshydrogénase (17β-HSD) type 1 …Qui mena à la découverte fortuite d’un 1er activateur de la 17β-HSD type 12

Mémoire

Alexandre Trottier

Maîtrise en médecine moléculaire Maître ès sciences (M. Sc.)

Québec, Canada

Résumé

Les 17β-hydroxystéroïdes déshydrogénases (17β-HSD) sont un groupe de 15 enzymes connues avant tout pour leur rôle dans le métabolisme des hormones sexuelles. La 17β-HSD1 est responsable de la toute dernière étape dans la fabrication des estrogènes actifs. Cela en fait une cible intéressante pour traiter l’endométriose et le cancer du sein qui sont stimulées par ces hormones.

Le dérivé stéroïdien PBRM, conçu dans notre laboratoire, est l’une des rares molécules ayant démontré une inhibition forte et spécifique de la 17β-HSD1. Lors des présents travaux, l’effet de l’inhibiteur s’est avéré irréversible, sélectif et durable tout en présentant un profil intéressant chez la souris.

Durant ce processus, plusieurs composés n’ayant pas les qualités requises ont été mis de côté. Parmi eux, l’un s’est avéré être un activateur de la 17β-HSD12, une enzyme essentielle dans l’élongation des acides gras. Il s’agit là du premier activateur rapporté pour la famille des 17β-HSD.

III

Abstract

17β-Hydroxysteroid dehydrogenases (17β-HSD) are a group of 15 enzymes known firstly for their involvement in sexual hornomes metabolism. 17β-HSD1 is responsible of the last step in the biosynthesis of potent estrogens. It is thus an interesting target to treat diseases stimulated by those hormones such as endometriosis and breast cancer.

PBRM, a steroidal inhibitor developed in our laboratory, is one of the few molecules that shown a strong and specific inhibition of 17β-HSD1. The present works showed that the inhibitory effect is irreversible, selective and long-lasting while showing an interesting profil in mice.

During that process, many other compounds were tested but didn’t have the required qualities. Among them, one seemed to stimulate the activity of 17β-HSD12, an essential enzyme for fatty acids elongation also involved in estrogen metabolism. It is the first reported activator for a member of 17β-HSD family.

Table des matières

Résumé ...... III Abstract ...... V Table des matières ...... VII Liste des tableaux ...... IX Liste des schémas ...... XI Liste des figures ...... XIII Liste des abréviations et symboles ...... XVII Épigraphe ...... XXIII Remerciements ...... XXV Avant-propos ...... XXVII Introduction ...... 1 1. Les estrogènes ...... 1 1.1 Définition ...... 1 1.2 Récepteurs estrogéniques ...... 2 1.2.1 Récepteur des estrogènes α (ERα) ...... 3 1.2.2 Récepteur des estrogènes β (ERβ) ...... 4 1.2.3 Récepteur estrogénique couplé aux protéines G (GPER) ...... 4 1.2.4 Autres ...... 5 1.3 Note ...... 5 1.4 Rôles physiologiques des estrogènes ...... 5 1.4.1 Reproduction ...... 6 1.4.2 Squelette et muscles ...... 8 1.4.3 Système digestif et métabolisme ...... 8 1.4.4 Neurologique ...... 9 1.4.5 Autres ...... 10 2. Maladies sensibles aux estrogènes ...... 10 2.1 Cancers ...... 10 2.1.1 Cancer du sein ...... 11 2.1.2 Cancers du système reproducteur ...... 12 2.1.3 Autres cancers ...... 13 2.2 Endométriose ...... 13

VII 2.3 Autres ...... 14 3. Modulation de la stéroïdogenèse comme approche thérapeutique ...... 14 3.1 Stéroïdogenèse ...... 15 3.2 Inhiber la biosynthèse des hormones sexuelles ...... 16 3.2.1 Agoniste/Antagoniste GnRH ...... 16 3.2.2 17α-Hydroxylase/17,20 lyase (CYP17A1) ...... 16 3.2.3 Aromatase ...... 18 3.2.4 Stéroïde sulfatase ...... 19 3.2.5 17β-Hydroxystéroïde déshydrogénase ...... 21 3.3 Inhibiteur de la 17β-HSD1 ...... 23 3.3.1 Intérêt thérapeutique...... 23 3.3.2 Avancées actuelles ...... 24 4. Activateurs enzymatiques ...... 25 4.1 Définition et mécanismes ...... 25 4.2 La 17β-HSD12 ...... 26 5. Objectifs et aperçu des travaux ...... 27 Chapitre 1 ...... 29 Résumé ...... 30 Discovery of a non-estrogenic irreversible inhibitor of 17β-hydroxysteroid dehydrogenase type 1 from 3- substituted-16β-(m-carbamoylbenzyl)-estradiol derivatives...... 31 Chapitre 2 ...... 81 Résumé ...... 82 Mode of action of PBRM as 17β-HSD1 inhibitor ...... 83 Chapitre 3 ...... 105 Résumé ...... 106 Identification of a first enzymatic activator of a hydroxysteroid dehydrogenase ...... 107 Conclusion ...... 123 Bibliographie ...... 127

VIII

Liste des tableaux

INTRODUCTION

Tableau 1: Portrait des nouveaux cas de cancer au Canada en 2013, produit par la Société canadienne du cancer en collaboration avec Statistique Canada et l’Agence de la Santé Publique du Canada80...... 11

CHAPITRE 1

Table 1. Inhibition of 17β-HSD1 (Serie I) ...... 42 Table 2. Inhibition of 17β-HSD1 (Serie II) ...... 44 Table 3. Inhibition of 17β-HSD1 (Serie III) ...... 46 Table 4. Selectivity of Compounds 1 and 23b on Four Enzymes (17β-HSD2, 17β-HSD7, 17β-HSD12, and CYP3A4) ...... 48

CHAPITRE 2

Table 1. Amino acids in key position for inactivation by PBRM ...... 92

Liste des schémas

CHAPITRE 1

Scheme 1. Reagents and conditions: (a) 3-carboxamide-benzaldehyde, KOH, EtOH, rx; (b) NaBH4, MeOH; (c)

H2, Pd/C, MeOH, rt...... 36

Scheme 2. Reagents and conditions: (a) R1R2NH, BOP, DIPEA, DMF, rt; (b) (i) BOP, DIPEA, THF, rt, (ii)

NaBH4, rt; (c) PPh3, CBr4, DCM, rt; (d) NHR1R2, Et3N, DCM, rt; (e) (i) NaN3, DMF, 60 °C, (ii) H2, Pd/C (10%), MeOH, rt...... 37 Scheme 3. Reagents and conditions: (a) styrene, Grubb (II) catalyst, dichloroethane, reflux; (b) HCl 10% in

MeOH, rt; (c) 3-carboxamide-benzaldehyde, KOH, EtOH, reflux; (d) NaBH4, MeOH, rt; (e) H2, Pd/C (10%),

MeOH, rt; (f) (i) BH3-DMS, THF, −78 °C, (ii) H2O2, NaHCO3; (g) NaH, benzylbromide; (h) CPMA, DCM, rt; (i)

PPh3, CBr4, DCM, rt; (j) NaI, acetone, rt; (k) (i) Dess–Martin reagent, DCM, rt, (ii) NaClO2, t-BuOH, 2-methyl- butene, KH2PO4, rt; (l) CH3NH2 in THF, BOP, DIPEA, DMF, rt...... 38 Scheme 4. Reagents and conditions: (a) Grubb II catalyst, allyloxymethyl-benzene; (b) HCl 10% in MeOH, rt;

PPh3, CBr4, DCM, rt; (g) allylbromide, NaOH, acetone, reflux; (h) (i) NaIO4, RuCl3–H2O, EtOAc/ACN, 0 °C, (ii)

NaBH4, THF:H2O (1:1), rt...... 39

Scheme 5. Reagents and conditions: (a) oxone, NaHCO3, acetone/ACN (1:2), rt; (b) Pd/C (10%), ammonium acetate, MeOH, 70°C; (c) 3-carboxamide-benzaldehyde, KOH, EtOH, reflux; (d) NaBH4, MeOH, rt; (e) H2,

Pd/C (10%), MeOH, rt; (f) PPh3, CBr4, DCM, rt...... 40

CHAPITRE 3

Scheme 1. Chemical synthesis of compounds A1 and A3. Reagents and conditions: (a) methoxymethylchloride, diisopropylethylamine (DIPEA), dichloromethane (DCM), rt; (b) 3-CN-benzaldehyde, KOH, EtOH, 100

°C; (c) NaBH4, MeOH, rt; (d) m-chloroperbenzoic acid, DCM, rt; (e) appropiate alkylamine, EtOH, microwave, 180 °C; (f) triphosgene, DIPEA, DCM, rt; (g) HCl 10% (v/v) in MeOH, 50 °C...... 110

Liste des figures

INTRODUCTION

Figure 1. Structures d’estrogènes communs de divers types et leur affinité relative pour leurs récepteurs nucléaires (taille des bulles proportionnelle à l’affinité pour ERα)...... 2 Figure 2: Voies de signalisation principales des estrogènes, tiré de Nilsson et Gustafsson en 201110. Dans un premier temps, l’estradiol lie GPER, un récepteur transmembranaire qui active la voie de signalisation de la protéine kinase A. Par la suite, il lie également certains ER situés à la membrane cellulaire, mais surtout au noyau ce qui activement respectivement d’autres voies de signalisation et la transcription de certains gènes. . 3 Figure 3: Effets principaux des estrogènes à travers le corps, tiré du site internet http://www.women-health- info.com/...... 6 Figure 4: Contrôle hormonal du cycle reproductif féminin, tiré du site http://www.santecheznous.com/. On y voie l’interrelation entre les hormones hypothalamo-hypophysaires, l’estradiol, la progestérone et la maturation des ovocytes ce qui affecte également l’endomètre au cours cycle ovarien...... 7 Figure 5: Représentation de la biosynthèse des principaux stéroïdes dont font partie les estrogènes dont la plupart apparaissent au-delà de la ligne pointillée rouge, qui est tirée du site http://ceri.com/q_v7n2q3.html. Le cholestérol est transformé à la fois en 3 types d’hormones stéroïdiennes : les minéralocorticoïdes (aldostérone), les corticostéroïdes et les hormones sexuelles par une pléthore d’enzymes...... 15 Figure 6: Principaux inhibiteurs à ce jour de la CYP17A1 et leur action sur les 2 réactions séparées catalysées par l’enzyme. L’ortéronel est dit spécifique à la seconde et n’affecte ainsi pas la production de corticostéroïdes qui se fait à partir de 17-Hydroxyprégnénolone contrairement à l’abiratérone qui inhibe les 2...... 17 Figure 7: Inhibiteurs de l’aromatase qui sont aujourd’hui utilisé pour empêcher la biosynthèse des estrogènes. Les substrats, qui ont un effet plus ou moins androgénique, et produits estrogéniques de l’enzyme sont aussi illustrés...... 18 Figure 8: Principal inhibiteur de la STS et principales activités de cette enzyme en bleu. Les flèches jaunes représentent l’action de sulfotransférases responsables de la réaction inverse de la STS...... 20 Figure 9: Principales activités dans la synthèse des hormones sexuelles rapportées pour les 17β-HSD, tiré de Samson, Labrie et Luu-The en 2009148. Pour la plupart, les formes actives se trouvent à droite et les inactivent à gauche de la figure. Les 17β-HSD sont responsables de l’interconvertion entre ces deux formes...... 21 Figure 10: Structure des inhibiteurs de la 17β-HSD1 les plus prometteurs de différentes séries rapportées... 24 Figure 11: Types de mécanismes d’activation enzymatique par des petites molécules identifiés jusqu’à maintenant et les enzymes pour lesquelles un tel mode d’action est connu, tiré de Zorn et Wells en 2010177. Les mécanismes de type A impliquent la liaison de sites allostériques et ceux de types B celle de sous-unités par l’activateur. Cela a pour effet de rendre le site catalytique plus disponible et/ou actif en causant le clivage d’une partie d’une proenzyme dans le cas de A2 et la dimérisation dans celui de B2...... 26 Figure 12: Structure du premier activateur de la 17β-HSD12 à être décrit...... 27

XIII CHAPITRE 1

Figure 1. Two pathways involved in the formation of strong estrogen E2 (pathway II) and weak estrogen 5-diol (pathway I) from key steroid DHEA...... 33 Figure 2. Key interactions observed in a ternary complex of 17β-HSD1/inhibitor 1/cofactor NADP and representation of new E2 derivatives modified at position 3 (series I, II, and III). The scope of this side-chain (R) is dual: (1) reaching a third interaction with an amino acid and (2) removing the undesirable estrogenic activity of the first generation inhibitor 1...... 34 Figure 3. 17β-HSD1 inhibitory potency of compounds 1, 14, 23b, 28, and 30 in T-47D intact cells. Breast cancer cells expressing 17β-HSD1 were incubated with various concentrations of inhibitor for 24 h in presence 14 of labeled [ C]-E1 (60 nM). IC50 represents the concentration that inhibited the transformation of E1 into E2 by 50%. Results are representative of two experiments performed in triplicate except for compound 14 (tested one time)...... 47 Figure 4. Time-dependent inactivation of 17β-HSD1 by compound 23b. The transformation of [14C]-E1 to [14C]-E2 by purified enzyme was assessed after preincubation with compound 23b (0, 100, or 500 nM), with or without the natural substrate E1 (500 nM), and expressed as the percentage of initial enzyme activity. See the Experimental Section for the conditions of the enzymatic assay...... 49 Figure 5. Representation of compound 1 in the active site of 17β-HSD1. Coordinates are from PDB 3HB5.(33) Compound 1 is represented by the thick purple sticks, NADP+ by the small cyan sticks, and the amino acids are represented by the small green sticks. H-bonds between compound 1 hydroxyl at position C3 and Glu-282/His-221 are highlighted...... 50 Figure 6. Results from the docking of compound 23b at the binding site of 17β-HSD1: (A) using the crystallographic binding site conformation, (B) with Glu-282 side chain solvent-oriented, (C) with Glu-282 side chain solvent-oriented and His-221 mutated into Ala, and (D) superposition of receptor from (B) with the docked compound 23b from (C). The structure of compound 1 is shown for purposes of comparison with the docked compound 23b and is not included in the docking calculations. Atoms from PDB 3HB5 are represented in green (protein in cartoon and compound 1 in sticks), the modified residues are represented by the purple sticks, and docked compound 23b by the orange (A), white (B), and cyan (C,D) sticks...... 51 Figure 7. Proposed irreversible mechanism of compound 23b (X = Br) and 23c (X = I) on 17β-HSD1...... 52 Figure 8. Carbon numbering used for the assignment of representative 1H NMR signals...... 54

CHAPITRE 2

Figure 1. Structure of 17β-HSD1 inhibitors PBRM, PIRM and [3H]-PBRM (radiolabeled PBRM)...... 85 Figure 2. Kitz-Wilson analysis of 17β-HSD1 inhibition by PBRM (A) and PIRM (B). Purified enzyme activity was evaluated (1 h inhibition time) after a defined pre-incubation time with different inhibitor concentrations and after the inhibitor was washed out. Slope (Kobs) from each experiment (left panel) was calculated and then plotted in a second graph as 1/ Kobs on 1/[inhibitor] (right panel). These graphes show the results of one assay representative of two independent assays conducted in triplicate. Values of Ki and kinact are expressed as the mean ± standard deviation of these two assays...... 87 Figure 3. 17β-HSD1 activity recovery of T47D (A) and JEG-3 (B) cells after a 24 h treatment with 1 µM PBRM (black spot) or with the vehicle only (white square). Cells were washed and enzyme activity measured after different times. The results of one assay representative of two conducted in triplicate is shown here...... 88

XIV

Figure 4. Inhibition of mice, pig, monkey and human 17β-HSD1 by 1 µM (red) and 10 µM (blue) of PBRM. Homogenized ovaries were used as source of enzyme for the three first species (mouse, pig and monkey), while highly purified enzyme was used for human. The values are expressed as the mean ± standard deviation of triplicate from one assay that is representative of two independent experiments for which control transformation of E1 to E2 has always ranged between 11% and 19%...... 90 Figure 5. Structural comparison of human 17β-HSD1 in green with monkey (purple), pig (cyan) and mouse (pink) (left to right respectively)...... 91 Figure 6. Tissue distribution of [3H]-PBRM in the digestive track (A) and other tissues (B) of mice. Three animals were sacrificed for each time after subcutaneous injection then organs were collected, dissolved and radioactivity quantified individually. The results are the mean value for three mice ± standard error...... 93 Figure 7. Mass balance of [3H]-PBRM in mice after subcutaneous administration. Feces (red) and urine (blue) were recuperated while cages were washed at determined times. The results are the mean ± standard deviation for two mice of the cumulative radioactivity measured in excrements in times in terms of injected dose...... 95

CHAPITRE 3

Figure 1. Chemical structure of compounds tested for 17β-HSD12 modulation (activation and inhibition). ... 110 Figure 2. Modulation (activation or inhibition) of the transformation of [14C]-E1 to [14C]-E2 by 17β-HSD12 in transfected HEK-293 intact cells. A) Screening of compounds 55, A1, A2, B1, B2, B3 and B4 tested at 10 µM. Ctrl: vehicle only. B) Stimulation of 17β-HSD12 activity by compounds A3 (grey) and A1 (black). Results of one experiment performed in triplicate (mean ± standard deviation)...... 111 Figure 3. Stimulation of [14C]-E1 to [14C]-E2 conversion by compound A1. A) Experiment performed in intact HEK-293 cells stably expressing 17β-HSD12. B) Experiment performed in microsomal extract obtained from homogenized HEK-293 cells stably expressing 17β-HSD12. Each figure is the result of one experiment performed in triplicate (mean ± standard deviation) and is representative of two assays...... 112 Figure 4. A) Evaluation of compound A1’s selectivity at 1 µM (grey) and 10 µM (black) for the transformation of [14C]-E1 by 17β-HSD1, 17β-HSD7 and 17β-HSD12, and for the transformation of [14C]-E2 by 17β-HSD2. T47D cells were used for 17β-HSD1 assays and transfected HEK-293 cells for the other enzymes. B) Stimulation by compound A1 of [14C]-E1 to [14C]-E2 transformation in T47D cells cultured at confluence and treated with a potent 17β-HSD1 inhibitor (PBRM). These are the results of one experiment in triplicate (mean ± standard deviation) representative of two assays...... 114

CONCLUSION

Figure 1. Structure du PBRM ...... 123 Figure 2. Structure des activateurs efficaces identifiés jusqu’à maintenant pour la 17β-HSD12 où X est une chaîne alkyle...... 125

Liste des abréviations et symboles

Chimie et biologie

Δ15-16 Double liaison entre les carbones 15 et 16 Δ5-diol 5-Androstène-3β,17β-diol 17β-HSD 17β-hydroxystéroïde déshydrogénase 3β-HSD 3β-hydroxystéroïde déshydrogénase 4-dione 4-Androstène-3,17-dione, plus communément appelé Androstènedione 5-diol 5-Androstène-3β,17β-diol 5α-diol 5α-Androstane-3β,17β-diol ACN Acétonitrile AKR Aldo-céto réductase Ala Alanine Arg Arginine Asn Asparagine Bh3-dms Complexe de diméthylsulfure de borane BOP (Benzotriazol-1-yloxy)tris(diméthylamino) phosphonium hexafluorophosphate BPA Bisphénol phosphate A BSA Albumine de sérum bovin CA Californie ou Canada CHU Centre hospitalier universitaire CHUL Centre hospitalier de l'Université Laval CIHR Instituts de recherche en santé du Canada (IRSC) CPMA (Chloro-phénylthio-méthylène)diméthylammonium chloride CREMOGH Centre de recherche en endocrinologie moléculaire et oncologique et génomique humaine CYP Cytochrome P450 CYP17A1 17α-Hydroxylase/17,20 lyase CYP3A4 Cytochrome P450 3A4 DBF Dibenzylfluorescéine DCM Dichlorométhane DES Diéthylstilbestrol DHEA Déhydroépiandrostérone

XVII DHT Dihydrotestostérone DIPEA Diisopropyléthylamine DMEM F-12 Milieu de culture de Dulbecco modifié de type F-12 DMF Diméthylformamide DMSO Diméthylsulfoxide E1 Estrone E2 Estradiol E3 Estriol EDTA Acide éthylène diamine tétraacétique ER Récepteur des estrogènes ERRγ Récepteur orphelin de molécules associées aux estrogènes γ ER-X Récepteur des estrogènes X ERα Récepteur des estrogènes α ERβ Récepteur des estrogènes β ERγ Récepteur des estrogènes γ EtOAc Acétate d'éthyle EtOH Éthanol FBS Sérum de veau fétal Fig Figure FRSQ Fond de recherche du Québec - Santé FSH Hormone folliculo-stimulante Glu Glutamine GLUT4 Transporteur de glucose de type 4 Gly Glycine GnRH Hormone de libération des gonadotrophines hypophysaires GPER Récepteur couplé aux protéines G des estrogènes GPER-1 Récepteur couplé aux protéines G des estrogènes de type 1 GPR30 Récepteur couplé aux protéines G 30 HDL Lipoprotéines de haute densité HEK-293 Cellules humaines de rein embryonaire 293 His Histidine HMG-CoA 3-Hydroxy-3-méthylglutaryl-coenzyme A HPLC Chromatographie en phase liquide à haute performance HSP Protéines de choc thermique IBIS Institut de Biologie Intégrative et des Systèmes

XVIII

IR Infrarouge KAR 3-Cétoacyl-CoA réductase LDL Lipoprotéines de faible densité LH Hormone lutéinisante LPL Lipoprotéine lipase LRMS Spectrométrie de masse à basse résolution MCF-7 Cellules "Michigan Cancer Foundation-7" MEM Milieu de culture minimal essentiel MeOH Méthanol MO Missouri NAD+ Nicotinamide adénine dinucleotide NADP Nicotinamide adénine dinucleotide phosphate NADPH Nicotinamide adénine dinucleotide phosphate NIS Symport Na/I NJ New Jersey NMR Résonance magnétique nucléaire ON Ontario P450scc Cholestérol desmolase PBRM Produit bromé René Maltais ou 3-{[(16β)-3-(2-bromoethyl)-17-hydroxyestra-1(10),2,4-trien- 16-yl]methyl}benzamide PDB Banque de données de protéines PIRM Produit iodé René Maltais ou 3-{[(16β)-3-(2-iodoethyl)-17-hydroxyestra-1(10),2,4-trien-16- yl]methyl}benzamide PMID Identifiant unique de Pubmed PROTEO Centre de Recherche sur la Fonction, la Structure et l’Ingénierie des Protéines p-TSA para-toluène sulfonique acide QC Québec RCPG Récepteur couplé aux protéines G ou GPCR de l’abbréviation anglaise Ref Référence rt Température ambiante SAR Relation structure à activité SDR Déshydrogénase à chaînes courtes SERM Modulateurs sélectifs des récepteurs des estrogènes SRY Région du chromosome Y determinant le sexe STS Stéroïde sulfatase

XIX t-BuOH 2-Méthylpropan-2-ol ou tert-butanol TEA Triéthylamine THF Tetrahydrofurane TLC Chromatographie sur couche mince Tris Trishydroxyméthylaminométhane, ou 2-amino-2-hydroxyméthyl-1,3-propanediol USA États-Unis d'Amérique UV Ultraviolet vs Versus

Unités de mesure et mathématiques

% Pourcent SD Écart type µL Micro litre µM Micro molaire Å Ångström (équivalent de 10-10m) Cm3/mol Centimètre cube par mole °C Degré Celsius DPM/g Désintégration par minute par gramme

EC50 Concentration efficace médiane g Gramme h Heure Hz Hertz

IC50 Concentration inhibitrice médiane IU Unité internationale J Joule

Ki Constante d'inhibtion kinact Constante d'inactivation

Km Constante de Michaelis

Kobs Constante observée M Molaire mCi Millicurie (mesure radioactivité) mg Milligramme MHz Mégahertz

min Minute min-1 Réciproque de la minute (mesure de vitesse) mL Millilitre mM MilliMolaire mm Millimètre mmol Millimole ng Nanogramme nM Nanomolaire pH Potentiel hydrogène RMSD Racine carrée de l'erreur quadratique moyenne

T½ Temps de demi-vie v/v Volume/volume

Vmax Vitesse initiale maximale

XXI

À la mémoire d’Uranie.

À une véritable science Qui devrait être bien plus qu’une pute Qui s'agenouille pour la jouissance de quelques-uns

Parce qu’elle est plus qu’une technique, un métier ou une activité Mais bien une part de notre société et un art digne de sa muse

XXIII

Remerciements

J’ai quand même beaucoup appris au cours des deux (ou plutôt trois) dernières années. Peut-être pas autant que ce à quoi je m’attendais, mais à vrai dire pas le type de connaissances que j’avais envisagé non plus. Pour cela, je dois remercier l'équipe du laboratoire, mais particulièrement Donald et René. Pour l'accueil bien sûr, mais au-delà des courtoisies qui sont de mise, c'est pour la possibilité de réellement tester mes idées que je suis réellement reconnaissant. Sans oublier Charles, Guy, Lucie, Jenny, Diana, Marie-Claude de même que tous les autres que j’oublie certainement et qui ont avantageusement fait partie de mon quotidien au cours des dernières années… en recherche et autour.

Un merci un peu plus spécial à Amélie, Yann, Coraline et tous les quelques rares autres (« chanceux ») dont la journée est souvent loin d'être terminée à 17h. Les soirées au laboratoire auraient clairement été beaucoup plus pénibles sans vous... Et à coup sûr moins intéressantes.

J’aimerais également remercier toutes les personnes qui m’ont aidé à tracer ma voie au cours des dernières années et qui ont fait que je suis rendu là où je suis aujourd’hui et maintenant. Qu’ils aient été des professeurs, des directeurs de recherche, des amis, des collègues, des coéquipiers ou autres connaissances, je leur en suis tous reconnaissant autant qu’ils puissent être.

Parmi eux se trouve évidemment ma famille qui m’a toujours soutenu sur bien des aspects (et généralement encouragé) à continuer dans la voie que j’ai choisie. Pour cette maîtrise, je me sais gré à Audrey. Elle qui m’a enduré tout au long, que les choses roulent bien ou non… des activités perturbées par les imprévus jusqu’au sommeil qui l’était par diverses pensées, idées et muses (malheureusement c’est relativement fréquent).

Je souhaite finalement exprimer ma gratitude à toutes les sources de mon inspiration. Qu’il s’agisse de plantes dont j’ai pris soin ou non, des couchers de soleil, de la magnifique vue que l’on a du centre de recherche ou encore d’autres scènes qui m’ont inspiré par leur beauté et/ou leur complexité, je ne peux qu’en être reconnaissant.

XXV

Avant-propos

À la base, le but des travaux de notre laboratoire est de développer des traitements pour divers cancers. Ces maladies seraient la cause de mortalité la plus importante au Canada à l’instar des maladies cardiovasculaires. Une bonne partie de ces recherches visent les cancers du sein et du système reproducteur. Le PBRM, qui est le principal sujet de ce travail, cible une enzyme nécessaire à la formation des estrogènes actifs.

En tant que tel, il pourrait devenir une option thérapeutique intéressante dans le cancer du sein entre autres, particulièrement dans une optique de médecine personnalisée où les traitements seraient formulés sur mesure pour une personne donnée. Toutefois, Il s’agit peut-être d’un traitement encore plus intéressant pour l’endométriose, une maladie pour laquelle les choix de traitements acceptables sont très peu nombreux, voire inexistants pour bien des femmes atteintes. Un inhibiteur de la 17β-HSD1 pourrait être un grand avantage pour la qualité de vie de nombreuses femmes atteintes et pour qui aucun traitement n’est réellement efficient.

Après l’introduction de ce mémoire, les deux premiers chapitres sont dédiés à des expériences menées dans le but de développer cette molécule prometteuse. Il s’agit d’articles rédigés en anglais pour des revues scientifiques, le premier ayant été publié alors que le second est destiné à l’être dans les prochains mois.

Au premier chapitre, l’article écrit par Dr René Maltais fait suite à deux manuscrits qui traitaient du PBRM. Le premier a rapporté que le groupement 2-bromoéthyl, caractéristique de cette molécule et positionné sur le carbone 3, avait permis d’éviter les effets estrogéniques observés pour les précurseurs du PBRM. Le second a montré que l’inhibiteur résultant avait un effet optimal in vitro et in vivo. Dans ce troisième article paru dans le Journal of Medicinal Chemistry, c’est la découverte du PBRM, parmi trois séries de molécules synthétisées dans le but d’obtenir des inhibiteurs dépourvus d’activité estrogénique, qui est rapportée. À cela est venue se greffer une analyse de sa sélectivité et de son mécanisme d’action, ce qui constitue ma contribution à cet article. J’ai aussi fait les reprises d’autres tests biologiques présentés pour les IC50 notamment. Les essais de modélisations explorant le processus d’inhibition ont été menés par Xavier Barbeau, étudiant du Pr Patrick Lagüe.

Dans le sillage de ces travaux, le chapitre deux aborde plus en profondeur le mode d’action du PBRM. Il s’agit d’un manuscrit à un stade préliminaire de son écriture avec Pr Donald Poirier, mon directeur de recherche, et quelques résultats devraient s’y rattacher dans les prochains mois. N’en demeure pas moins que les travaux qui s’y trouvent représentent une part importante du projet principal de ma maîtrise. J’ai effectué et conçu en grande partie les tests biologiques de l’article tant in vitro qu’in vivo en collaboration notamment avec Dre Jenny

XXVII Roy dans les derniers temps. La synthèse chimique et le radiomarquage ont été effectués par Dr René Maltais alors que la modélisation moléculaire a été effectuée par Xavier Barbeau, étudiant du Pr Patrick Lagüe.

Le troisième chapitre traite d’une découverte qui est indirectement issue de travaux pour développer des anticancéreux. Il est constitué d’un article, récemment publié dans la revue ACS Chemical Biology, qui a pour objet la découverte inattendue d’un activateur de la 17β-HSD12. Vu l’omniprésence des inhibiteurs, autant dans notre laboratoire qu’en dehors, on en vient à oublier qu’il existe aussi de rares cas d’enzymes dont l’activité est stimulée par certaines molécules, des activateurs. À l’occasion d’un essai d’inhibition de la 17β-HSD12, quelques molécules du laboratoire ont été sélectionnées au hasard, pour remplir l’espace disponible surtout. Si aucune n’inhibait l’enzyme, un inhibiteur inefficace de la 17β-HSD1 en augmentait au contraire l’activité jusqu’à 300%.

Après avoir répété ce résultat, il a été convenu de caractériser sommairement cet activateur même si l’intérêt que cela avait n’était pas très clair. À noter qu’un activateur de la 17β-HSD12 est assurément destiné à un usage en recherche fondamentale plutôt éloigné des orientations pratiques du laboratoire. Ainsi, ces travaux découlent d’une observation que j’ai faite, puis que j’ai investiguée. J’ai donc développé et effectué l’ensemble des expériences pour rédiger l’article sous la supervision de Dr René Maltais et Pr Donald Poirier.

XXVIII

Introduction

1. Les estrogènes

1.1 Définition

Traditionnellement, les estrogènes sont considérés comme « les hormones femelles » qui régulent la différenciation et la croissance des cellules de divers tissus en régulant la transcription de divers gènes, et ce, pour tous les vertébrés et même certains insectes1. Il est généralement accepté aujourd’hui que cette vision des estrogènes est bien simpliste. Ces hormones régulent maints systèmes physiologiques à la fois par des effets génomiques et non génomiques via plusieurs récepteurs distincts2,3.

Chez l’humain, les estrogènes sont des hormones stéroïdiennes majoritairement produites dans les gonades et le placenta, dont le plus puissant et le plus important représentant est l’estradiol (E2). Quatre autres stéroïdes exercent un effet estrogénique, soit : l’estrone (E1), un précurseur de l’estradiol, l’estriol (E3), produit en grande quantité par le placenta lors de la grossesse, le 5-androstène-3β,17β-diol (Δ5-diol), un estrogène faible qui deviendrait le principal estrogène après la ménopause, et le 5α-androstane-3β,17β-diol, un métabolite de la dihydrotestostérone (DHT) (voir Figure 1)4.

1 Représentation graphique produite à partir des données publiées par : - Kuiper et al.11 - Harris et al.185

Figure 1. Structures d’estrogènes communs de divers types et leur affinité relative pour leurs récepteurs nucléaires (taille des bulles proportionnelle à l’affinité pour ERα).

Par ailleurs, nous sommes au quotidien exposés à une variété de composés exogènes dans notre environnement qui ont un effet estrogénique. Ces xénoestrogènes peuvent être différenciés en trois catégories, soit : les phytoestrogènes, produits par les plantes5 (β-zearalanol, génistéine, coumestrol, etc.), les produits de synthèse ou contaminants comme certains pesticides (methoxychlore entre autres) et additifs commerciaux (comme le bisphénol A (BPA))6,7, et les médicaments (éthinylestradiol, tamoxifène, diéthylstilbestrol (DES), etc.) (voir Figure 1)6.

1.2 Récepteurs estrogéniques

À l’instar des autres hormones stéroïdiennes, les estrogènes agissent principalement en liant des récepteurs nucléaires. Les récepteurs des estrogènes (ER) qui se retrouvent dans le cytoplasme (environ 5%) et dans le noyau (95%) des cellules sous forme d’un complexe stable avec des protéines de choc thermique (HSP)7. Ces protéines chaperonnes sont relâchées suite à la liaison d’un ligand, ce qui entraîne l’homo- ou l’hétérodimérisation du récepteur et permet son action comme facteur de transcription en modulant l’expression d’une variété de gènes.

En plus de ce rôle dit classique des estrogènes, il est maintenant bien reconnu que ces hormones exercent également certains effets non génomiques via l’activation d’un récepteur membranaire (GPER) et/ou d’une population de récepteur classiques (ER) liés à la membrane cellulaire8. Pour ajouter à la complexité, il est également à noter que les ER exercent certains effets, même en l’absence de ligand via certaines voies de signalisation9.

Figure 2: Voies de signalisation principales des estrogènes, tiré de Nilsson et Gustafsson en 201110. Dans un premier temps, l’estradiol lie GPER, un récepteur transmembranaire qui active la voie de signalisation de la protéine kinase A. Par la suite, il lie également certains ER situés à la membrane cellulaire, mais surtout au noyau ce qui activement respectivement d’autres voies de signalisation et la transcription de certains gènes.

1.2.1 Récepteur des estrogènes α (ERα)

Le principal récepteur responsable de l’effet des estrogènes serait ERα. Il est également celui qui serait le plus impliqué dans la stimulation de cancers par les estrogènes2. Son expression est surtout observée au niveau du système reproducteur, des reins et de la glande surrénale11. De nombreux autres tissus présentent également une expression plus faible.

3 Trois sous-types ont été rapportés jusqu’à maintenant dont une forme dite « Wild-type » (ERα1) et deux isoformes tronquées (ERα2 et ERα3)12. Les implications de chacune de ces isoformes demeurent toutefois largement méconnues et peu de distinction entre les 3 types est faite dans la littérature de même que dans le présent ouvrage. ERα a longtemps été tout simplement nommé ER avant d’être renommé quelques décennies plus tard, suite à la découverte d’un second type de ER alors nommé ERβ.

1.2.2 Récepteur des estrogènes β (ERβ)

Découvert plus récemment, en 1996, la fonction de ERβ demeure méconnue et sujette à débat13. Ce récepteur a d’abord été considéré comme étant la contrepartie de ERα qui exercerait des effets opposés à ce dernier14. Des données plus récentes tendent plutôt à montrer que ERβ aurait un rôle bien différent, voire à deux faces15. Alors qu’en coexpression avec ERα, l’activation d’ERβ semble effectivement avoir un effet souvent opposé (antiprolifératif, proapoptotique, etc.), lorsqu’il est exprimé seul il peut au contraire montrer un effet semblable à ERα.

Par ailleurs, les 2 récepteurs présentent une distribution étendue bien que des profils d’expression passablement différents. ERβ est plus fortement exprimé au niveau de la prostate, des ovaires, de la vessie et des poumons11. De plus, cinq isoformes de ERβ, une « Wild-type » (ERβ1) et 4 formes tronquées (ERβ2 à 5) ont été rapportées16. Ces isoformes auraient des profils d’expression et des actions différents alors que l’une d’entre elles ne lie pas les estrogènes.

1.2.3 Récepteur estrogénique couplé aux protéines G (GPER)

Les estrogènes exercent des effets non génomiques qui seraient majoritairement dus à GPR30, un récepteur couplé aux protéines G (RCPG) récemment renommé GPER ou GPER117,18. Jusqu’à maintenant, E2 est le seul ligand endogène connu de ce récepteur ce qui en fait un récepteur des estrogènes19. Étrangement, alors que GPER a une affinité semblable à celle des ER pour E2, il ne lie ni E1 ni E320,21. La caractérisation de ce RCGP étant relativement nouvelle, son rôle physiologique demeure sujet à controverse, mais il aurait un rôle important dans la prolifération de certains cancers de même qu’au niveau cardiovasculaire.

L’expression de ce récepteur membranaire a été rapportée au niveau des vaisseaux sanguins, du cerveau, du cœur et du système reproducteur22,23. Le rôle de GPER demeure sujet à discussion, mais il semblerait que plusieurs des effets néfastes autant que bénéfiques soient médiés par ce récepteur24,25. Le développement

d’agonistes et d’antagonistes sélectifs devrait permettre d’en apprendre davantage sur celui-ci dans les prochaines années26,27.

1.2.4 Autres

L’existence de récepteurs membranaires des estrogènes a longtemps été un mystère même après avoir été confirmée dès 197728. Mais, suite à la caractérisation de GPER au début du nouveau millénaire, une variété de nouveaux récepteurs ont fait leur apparition dans la littérature3. Bien que peu caractérisés et largement ignorés au profit de leurs vis-à-vis intracellulaires, il n’en demeure pas moins qu’ils exerceraient des rôles essentiels de signalisation rapide des estrogènes particulièrement au niveau du cerveau8. Faute d’être bien identifiés à ce jour, ces récepteurs sont souvent rapportés en tant que ER-X3,29,30.

Un autre récepteur liant des molécules associées aux estrogènes (ERRγ), impliqué dans la glucogénèse au niveau du foie, a été caractérisé31. Certains estrogènes exogènes comme le tamoxifène sont connus pour lier ce récepteur intracellulaire, mais aucun composé endogène. Il existe également un 3e récepteur intracellulaire des estrogènes confirmé, soit le récepteur des estrogènes γ (ERγ). Cependant, celui-ci n’est pas exprimé chez l’humain32.

1.3 Note

Aux fins du présent mémoire, le terme estrogène réfère aux estrogènes endogènes. Par ailleurs, le terme ER signifie autant ERα que ERβ à l’image de l’ambiguïté relativement fréquente dans la littérature autour de l’utilisation de ce terme. GPER sera également considéré pour la suite. Les autres récepteurs seront ignorés faute d’intérêt et/ou d’informations suffisantes pour en discuter. Il en va de même pour la distinction entre les différentes isoformes d’ERα et d’ERβ.

1.4 Rôles physiologiques des estrogènes

Tel que l’expression quasi ubiquitaire de leurs récepteurs le suggère, les estrogènes sont impliqués dans une grande variété de processus physiologiques, et ce, dans divers tissus et organes qui ne se limitent pas au système sexuel (Figure 3).

Figure 3: Effets principaux des estrogènes à travers le corps, tiré du site internet http://www.women-health- info.com/.

1.4.1 Reproduction

En tant « qu’hormone féminine », les estrogènes sont populairement associés à la fonction reproductive de la femme. Le cycle menstruel est probablement le système le plus notoirement régi par les hormones stéroïdiennes que sont les progestatifs et les estrogènes33. Ces dernières augmentent puis diminuent brutalement juste avant l’ovulation ce qui déclenche une montée rapide des taux d’hormone lutéinisante (LH) et de l’hormone folliculo- stimulante (FSH), les hormones responsables du contrôle de la biosynthèse d’estrogènes34. Celles-ci à leur tour déclenchent une série de processus qui causent la rupture folliculaire et la relâche d’un ovocyte vers les trompes de Fallope et l’utérus pour la fécondation (Figure 4)

Les estrogènes favorisent notamment le processus de reproduction en augmentant le désir sexuel35. Advenant la fécondation, les concentrations sériques d’estrogènes augmentent constamment au cours de la grossesse grâce au placenta jusqu’à la naissance. Cela joue un rôle crucial dans le développement utérin et mammaire. D’un côté, la stimulation estrogénique va causer le développement et la vascularisation de l’endomètre, la muqueuse utérine, afin de fournir un environnement favorable au fœtus36. De l’autre, elle va entrainer la

différenciation et la prolifération du système galactophorique dans le but de permettre l’allaitement du nourrisson à terme37. Ces effets marqués sur les organes reproductifs de la femme expliquent à la fois l’utilisation des estrogènes comme anovulants et leur appellation d’hormone femelle38.

Figure 4: Contrôle hormonal du cycle reproductif féminin, tiré du site http://www.santecheznous.com/. On y voie l’interrelation entre les hormones hypothalamo-hypophysaires, l’estradiol, la progestérone et la maturation des ovocytes ce qui affecte également l’endomètre au cours cycle ovarien.

Chez l’homme, au niveau sexuel, un ratio estrogène/androgène élevé favorise également la gynécomastie (développement mammaire)39. De plus, l’exposition foetale à des composés estrogéniques est associée à de l’hypospadias (ouverture de l’urètre sous le pénis)40, de l’hypogonadisme41, de la cryptorchidie (positionnement d’un testicule ou des deux à l’extérieur du scrotum)42 et une diminution de la fertilité43. Paradoxalement, ces mêmes hormones sont impliquées dans la spermatogénèse44 et dans l’éjaculation,45 des processus ayant bien évidemment un effet favorable pour la reproduction.

Par ailleurs, les hormones sexuelles jouent un rôle important dans la différenciation sexuelle. On sait depuis longtemps que les androgènes amorcent le développement de la prostate et des gonades (testicules) chez l’homme conséquemment à la présence du gène SRY46, faute de quoi s’amorce la différenciation ovarienne41. Les estrogènes joueraient également divers rôles dans la formation des appareils génitaux en causant une certaine féminisation des organes génitaux femelles47 et mâles43 48.

1.4.2 Squelette et muscles

Un autre rôle bien familier des estrogènes est celui de protection de l’intégrité des os49. Un effet bien visible avec la diminution de la densité osseuse et l’ostéoporose qui apparaissent typiquement après la ménopause, au moment où les taux sanguins d’estrogènes sont à leur plus bas50. Les hormones sexuelles ont en fait un impact considérable sur les ostéoblastes et ostéoclastes, les cellules responsables respectivement de la formation et de la résorption osseuses51.

Si la régénération du tissu osseux par les estrogènes est un processus bien connu, celle du tissu musculaire l’est bien moins. La baisse de la production d’estrogènes à la ménopause est également synonyme d’une augmentation du temps de récupération après un dommage musculaire52 et des symptômes de sarcopénie (perte de masse musculaire)53. Si le mécanisme reste à clarifier, les estrogènes diminuent clairement l’infiltration de cellules inflammatoires dans les muscles et stimulent la réparation de ces derniers 53,54.

1.4.3 Système digestif et métabolisme

Les récepteurs des estrogènes sont également exprimés au niveau du tractus gastro-intestinal et des organes qui y sont associés55. Leur activation va stimuler la relâche d’insuline glucodépendante par le pancréas56 et augmenter l’expression de GLUT4, un important transporteur du glucose57.

En plus de ce rôle dans l’homéostasie du glucose, les estrogènes régulent la lipolyse par le tissu adipeux en diminuant l’expression de la lipoprotéine lipase (LPL)58 et la lipogenèse en réduisant l’expression de gènes nécessaires à la synthèse de lipides au niveau du foie59. Toujours au niveau hépatique, ces hormones vont faciliter l’excrétion du cholestérol en diminuant les taux sanguins de lipoprotéines de faible densité (LDL) et en augmentant ceux de lipoprotéine de haute densité (HDL), tout en stimulant la production de cholestérol par la stimulation de la HMG-CoA réductase60.

Par ailleurs, les hormones sexuelles seraient en partie responsables des différences entre hommes et femmes dans la manière dont les gras sont disposés (forme pomme et poire respectivement)61. Une différence majeure vient du fait que les femmes accumulent moins de gras viscéraux durant la période qui s’étend de la puberté à la ménopause. Après cette dernière, considérant l’atténuation de ce trait distinctif et la diminution de la production d’estrogènes conséquente, il est généralement considéré que les estrogènes seraient les principales responsables des différences observées62.

Il a également été montré que la motilité gastro-intestinale serait globalement diminuée et l’absorption des lipides augmentée par les composés estrogéniques. Ces derniers exerceraient au niveau du tractus gastro-intestinal plusieurs rôles physiologiques qui demeurent peu étayés à ce jour par rapport à leur vis à vis pathologiques, lesquels seront abordés plus loin63.

1.4.4 Neurologique

Parallèlement à l’impact sur la digestion et la reproduction, les estrogènes agissent également au niveau du cerveau pour augmenter l’apport énergétique en augmentant l’appétit et en modifiant le goût64. Ces deux effets communément associés à la grossesse, lors de laquelle les taux d’estrogènes sanguins sont à leur plus haut, relèvent du rôle neuroendocrinien de ces hormones65. Ces dernières réguleraient également la disposition des gras en agissant au niveau de l’hypothalamus, en agissant plus particulièrement sur les astrocytes61.

La différence homme/femme dans la résistance à la douleur est un autre effet relié aux estrogènes que tout un chacun a pu observer66. Même si les mécanismes impliqués sont sujets à étude, elle serait en partie due aux estrogènes qui auraient, grosso modo, un effet antinociceptif67. Chez la femme, une certaine analgésie est observable à la grossesse tandis qu’au contraire, les périodes lors desquelles les taux d’estrogènes sont plus bas sont généralement associées avec une plus grande sensibilité à la douleur68. Les hormones estrogéniques modulent le système opioïdergique autant au niveau de la production endogène d’opioïdes que de l’activité et l’expression des récepteurs des opiacés67, ce qui expliquerait une bonne partie de l’effet antinociceptif.

L’estradiol tend à augmenter la neurogénèse, induire la plasticité neuronale et la connectivité neuronale69.Ces actions exerceraient un rôle fondamental dans des processus comme la mémoire, l’apprentissage et la cognition en général70. Globalement, les estrogènes sont reconnus pour avoir un effet neuroproteur. Cela est en partie dû à l’altération de l’humeur et des processus cognitifs observés aux environs de la ménopause71, mais aussi à l’effet protecteur contre les dommages au cerveau associé à ces hormones72. En effet, une augmentation du risque de dépression et de déclin cognitif accompagne la chute de la production d’hormones sexuelles caractéristique de la ménopause71.

Il est intéressant de constater que les diverses actions des estrogènes au niveau du cerveau sont médiées par une variété de récepteurs3. Dans l’ensemble, l’effet neuroendocrinien des composés estrogéniques demeure nébuleux alors qu’autant le contrôle de leur production, l’effet, les voies de signalisation que les récepteurs eux- mêmes sont peu connus à l’heure actuelle.

9 1.4.5 Autres

Les récepteurs des estrogènes sont également exprimés par certaines cellules immunitaires, ce qui fait que l’estradiol a des effets directs en plus de certains effets indirects autant sur l’immunité innée qu’acquise73–75. Les bénéfices sur le système immunitaire font d’ailleurs partie des avantages de la thérapie hormonale après la ménopause76. L’effet sur ce système aurait également des effets cardiovasculaires positifs77.

2. Maladies sensibles aux estrogènes

Les estrogènes sont impliqués dans un certain nombre d’états pathologiques, que leur effet soit bénéfique ou néfaste. Ce chapitre portera sur les maladies qui sont stimulées par ces hormones.

2.1 Cancers

Un cancer est une maladie consistant en une prolifération excessive de cellules anormales et malignes. Ensemble, les cancers seraient aujourd’hui la plus grande cause de mortalité au Canada (Tableau 1)78. Plusieurs d’entre eux sont affectés positivement et/ou négativement par les estrogènes via leurs effets sur la prolifération et l’inflammation. Pour certains, les traitements hormonaux sont même une option thérapeutique intéressante, voire efficace et reconnue dans certains cas79.

2.1.1 Cancer du sein

Que ce soit chez l’homme ou chez la femme, les estrogènes jouent un rôle important dans le cancer du sein81,82. Cette maladie a été diagnostiquée chez 200 hommes et 23 800 femmes en 201378. Or, la majorité des cas du cancer du sein sont dépendants des estrogènes à un moment ou à un autre de leur développement. En fait, les tumeurs exprimant ER seraient de plus en plus fréquentes83. Ce lien entre les estrogènes et le cancer du sein a été rapporté il y a bien longtemps. Dès la fin du 19e siècle, l’ovariectomie a été utilisée comme traitement du cancer du sein84. Depuis, le traitement et le diagnostic de ces tumeurs ont, fort heureusement, considérablement évolué grâce à des efforts importants faits en recherche depuis des décennies. En effet, il est beaucoup plus opportun aujourd’hui d’empêcher l’effet des estrogènes à l’aide de médicaments. Les antiestrogènes sont par ailleurs largement utilisés comme traitement du cancer du sein, mais aussi pour le prévenir85.

Si les ER sont bien connus pour leur implication dans le développement mammaire normal autant que pathologique, c’est également le cas de GPER86. Les hormones estrogéniques (ou les estrogènes) sont d’autant plus importantes qu’elles tendent à se concentrer dans les tumeurs par rapport au sang87. Un certain nombre d’effets des estrogènes de même que certains facteurs génétiques qui y sont liés ont d’ailleurs été associés à l’apparition du cancer du sein88.

11 2.1.2 Cancers du système reproducteur

En plus de celui du sein, plusieurs cancers du système reproducteur de la femme sont connus pour avoir une composante hormonale. Par exemple, plus de 90% des tumeurs endométriales de type 1 (qui comptent pour environ 80% des cancers de l’endomètre) expriment ER contre au plus 30% pour le type 289. En pratique, c’est environ 20% des femmes atteintes d’un cancer de l’endomètre pour qui les traitements avec un anti estrogènes seraient efficaces90. Le col de l’utérus réagit également aux estrogènes qui stimulent la prolifération de ce tissu. L’implication des estrogènes dans le cancer de cette portion de l’utérus demeure sujette à discussion, mais certains laboratoires ont montré un effet synergique avec le virus du papillome humain, voire une stimulation nécessaire de ERα91, pour le développement de ce type de cancer92. Tout récemment, une étude chez la souris a montré qu’une thérapie anti-estrogénique à base de raloxifène pourrait être efficace dans le traitement du cancer cervical93.

Une forme particulière de cancer du vagin causée par des estrogènes est également assez célèbre. De 1938 à 1971, le DES (Diéthylstilbestrol, voir figure 1) était prescrit aux femmes pour éviter les fausses couches et certains problèmes au cours de la grossesse. En plus de ne pas être particulièrement efficace, l’exposition in utero au DES est la principale cause de l’adénome vaginal qui ne se développe que plusieurs années plus tard, à l’âge adulte94,95

Les estrogènes jouent également un rôle dans un autre cancer gynécologique; celui des ovaires. Environ 30% à 40% des tumeurs ovariennes exprimeraient ER ce qui a été associé à un meilleur pronostic pour les patientes96. D’un autre côté, il semble que bloquer pharmacologiquement les voies estrogéniques ait montré un certain intérêt thérapeutique pour contrer le cancer de l’ovaire97. Dans tous les cas, il est clair que les ovaires expriment fortement ERβ dans la majorité des tissus ovariens, mais que cette expression est perdue au cours de l’oncogenèse. Au contraire, l’expression d’ERα et de GPER, souvent associés à l’effet des estrogènes favorisant la prolifération cellulaire, tend à augmenter durant le développement tumoral98,99.

La prostate également exprime les trois principaux ER, mais moins uniformément que les ovaires. ERβ est exprimé au niveau épithélial alors que ERα et GPER le sont dans les cellules stromales100,101. Toutefois, l’expression d’ERβ s’estompe au cours de l’évolution de la tumeur pour être recouvrée dans d’éventuelles métastases tandis que l’expression d’ERα progresse inversement100. L’augmentation du taux E2/DHT au cours du vieillissement qui s’amplifie considérablement en cas de prolifération anormale de la prostate pointe également vers un rôle des estrogènes dans le cancer de la prostate102. Par ailleurs, diverses thérapies ciblant l’effet de ces hormones ont été testées in vitro, in vivo et/ou lors d’essais cliniques. Les résultats et conclusions de ces études sont plutôt variables et l’intérêt thérapeutique de l’approche antiestrogénique pour le cancer de la prostate est d’autant plus incertain103.

2.1.3 Autres cancers

Après les cancers du sein et de la prostate, le cancer colorectal est l’un des plus fréquents dans la population canadienne78. Les estrogènes y provoquent également certains effets, mais leur implication demeure pour le moins confuse104. C’est principalement ERβ qui est exprimé au niveau du côlon, mais cette expression serait perdue au cours du développement d’un cancer ce qui suggère un rôle protecteur de ce récepteur105. Une situation semblable est aussi observée pour les autres cancers gastro-intestinaux en général pour lesquels des effets autant anti- que pro- oncogéniques sont rapportés.106,107

ERβ aurait également un rôle important dans la formation des lymphocytes et certains liens ont été établis avec la leucémie108. Dans le gliome, le type de tumeur du cerveau le plus fréquent, les estrogènes auraient aussi un rôle important via ERβ109. En général, les estrogènes semblent jouer un rôle au niveau des tumeurs cérébrales110,111 tout comme pour les cancers du foie112, des poumons113, de la peau114 et de l’os (ostéosarcome)115.

2.2 Endométriose

Cette maladie est caractérisée par la présence de tissu endométrial en dehors de la cavité utérine, ce dernier est alors caractérisé d’ectopique. Environ 10% des femmes en seraient atteintes bien que le diagnostic soit difficile à établir116. Il s’agit d’une pathologie qui est rarement mortelle, mais qui a des conséquences considérables sur la qualité de vie incluant des douleurs durant les règles et les relations sexuelles, des douleurs pelviennes, des règles anormales et même des hémorragies internes117. Il appert également que l’endométriose serait responsable d’environ 50% des cas d’infertilité chez la femme118.

L’endométriose est largement considérée comme une maladie dépendante des estrogènes. À ce titre, des antagonistes ERα, des agonistes de l’hormone de libération des gonadotrophines hypophysaires (GnRH) et des inhibiteurs de l’aromatase ont été testés comme traitement avec un certain succès119,120. Par ailleurs, la production d’estrogènes serait augmentée dans le tissu ectopique121. Ces hormones activeraient à la fois ERα, ERβ et GPER qui auraient tous un effet sur le développement de la maladie de la maladie122,123. Qui plus est, le risque d’endométriose semble augmenté par certaines mutations de gènes impliqués dans la production et l’effet des estrogènes124.

13 2.3 Autres

L’angioœdème héréditaire est une maladie rare qui prend la forme d’un gonflement épisodique important de la peau et/ou des muqueuses125. Le type 3 de cette pathologie est considéré comme dépendant des estrogènes. Définie il y a une dizaine d’années, cette forme autosomale dominante aussi particulière que rarissime touche presque seulement des femmes. Elle est aggravée par les hauts taux d’estrogènes lors d’une grossesse et par la prise de contraceptifs oraux126.

Les estrogènes ont également une influence sur certaines maladies auto-immunes, dont le lupus érythémateux qu’elles semblent promouvoir127. Le raloxifène a d’ailleurs été testé pour le traitement de cette pathologie128. Au contraire, une stimulation estrogénique peut avoir des effets bénéfiques dans le traitement d’autres maladies du même type comme l’encéphalite auto-immune, les scléroses multiples et l’arthrite129.

Les estrogènes ont également des effets positifs dans un certain nombre de pathologies, dont les maladies cardiovasculaires et neurodégénératrices de même que l’ostéoporose. Le contrôle des voies métaboliques par les estrogènes leur confère également un rôle bénéfique dans l’obésité qui est mis en évidence par l’augmentation des graisses à la ménopause avec la baisse des taux d’estrogènes130.

3. Modulation de la stéroïdogenèse comme approche thérapeutique

Deux stratégies visant à empêcher les estrogènes d’exercer leurs effets délétères s’imposent: empêcher les estrogènes de lier leur récepteur ou bloquer leur fabrication (Figure 5). La première option a été largement appliquée au cours des dernières décennies, à la fois avec des antagonistes totaux (Fulvestrant) que par des modulateurs sélectifs des récepteurs des estrogènes ou SERM (tamoxifène, raloxifène, lasoxifène, etc.). Ces derniers agissent comme agonistes ou comme antagonistes selon les tissus tandis que les antiestrogènes purs bloquent indistinctement les récepteurs estrogéniques. Il est à noter que ces molécules ont été développées bien avant que ne soit identifié GPER et il appert qu’à la fois le tamoxifène et le fulvestrant stimulent ce récepteur20. Cette stimulation délétère pourrait expliquer pourquoi de récents résultats ont montré que, par rapport aux antagonistes des estrogènes, des inhibiteurs de la stéroïdogenèse seraient plus avantageux pour le traitement des maladies sensibles aux estrogènes. S’il n’y a que les inhibiteurs de l’aromatase et les agonistes/antagonistes GnRH qui sont utilisés actuellement, plusieurs autres cibles potentielles pourraient aider à réguler la biosynthèse des hormones sexuelles.

Figure 5: Représentation de la biosynthèse des principaux stéroïdes dont font partie les estrogènes dont la plupart apparaissent au-delà de la ligne pointillée rouge, qui est tirée du site http://ceri.com/q_v7n2q3.html. Le cholestérol est transformé à la fois en 3 types d’hormones stéroïdiennes : les minéralocorticoïdes (aldostérone), les corticostéroïdes et les hormones sexuelles par une pléthore d’enzymes.

3.1 Stéroïdogenèse

La synthèse des estrogènes est d’abord sujette à un contrôle de l’axe hypothalamo-hypophysaire (Schéma 2). En réaction à certains stimuli, dont la baisse des niveaux d’estrogènes dans le sang, il y a sécrétion de GnRH par l’hypothalamus dans le cerveau. Cette hormone va par la suite stimuler le relâchement de FSH et de LH par l’adénohypophyse, lesquelles vont, à leur tour, réguler des enzymes responsables de la fabrication des hormones sexuelles131. Une série d’enzymes sont nécessaires pour passer de la matière première, le cholestérol, à des stéroïdes sexuels qui vont activer leurs récepteurs (Figure 5). Certaines d’entre elles peuvent être ciblées afin de diminuer la production des estrogènes.

15 3.2 Inhiber la biosynthèse des hormones sexuelles

3.2.1 Agoniste/Antagoniste GnRH

Comme leur nom l’indique, ces molécules ne sont pas des inhibiteurs à proprement parler, mais bien des ligands des récepteurs de la GnRH. Il n’en demeure pas moins qu’ils ont comme effet d’inhiber la stéroïdogenèse. La GnRH est un peptide produit au niveau de l’hypothalamus dont le relâchement est pulsatile132. L’amplitude et le rythme de ces pulsations dans les concentrations de l’hormone régulent l’expression de son récepteur même. Utiliser un agoniste de la GnRH résulte en une stimulation constante qui sature les récepteurs de l’hormone, ce qui a pour effet, à terme, de diminuer la quantité de récepteurs disponibles au niveau de l’hypophyse133. Au final, il en résulte une forme de castration via inhibition de la relâche de LH et de FSH qui sont nécessaires à l’expression des enzymes de la stéroïdogenèse.

Ainsi, il existe deux manières d’empêcher l’effet de la GnRH: bloquer directement le récepteur à l’aide d’un antagoniste ou en entraver l’expression avec un agoniste. Paradoxalement, autant les agonistes que les antagonistes sont utilisés pour le traitement de cancers hormonaux, de l’endométriose ou du syndrome des ovaires polykystiques,134–136 entre autres. Les agonistes (leuprolide, histrelin, goserelin, etc.) sont utilisés depuis déjà plusieurs années tandis que l’utilisation des antagonistes est plus récente (cetrorelix, abarelix, degarelix, etc.)136. Pour ajouter au paradoxe, il est à noter que ces molécules sont également d’intérêt pour la promotion de la fertilité135,137, alors qu’elles sont utilisées pour leur effet inverse dans le cancer.

3.2.2 17α-Hydroxylase/17,20 lyase (CYP17A1)

Cette enzyme de la glande surrénale exerce deux activités distinctes et successives dans la stéroïdogenèse : la 17α-hydroxylation transforme les progestatifs en précurseurs des glucocorticoïdes qui peuvent ensuite être transformés par l’activité 17,20 lyase en déhydroépiandrostérone (DHEA) ou en 4-androstène-3,17-dione (4- dione). Ces deux derniers stéroïdes sont des précurseurs peu et/ou pas actifs des hormones sexuelles. Considérant qu’il n’existe pas actuellement d’inhibiteur sélectif à l’une ou l’autre de ces activités, l’inhibition de cette enzyme cause l’arrêt de la production autant des glucocorticoïdes que des androgènes et des estrogènes (Figure 5).

Figure 6: Principaux inhibiteurs à ce jour de la CYP17A1 et leur action sur les 2 réactions séparées catalysées par l’enzyme. L’ortéronel est dit spécifique à la seconde et n’affecte ainsi pas la production de corticostéroïdes qui se fait à partir de 17-Hydroxyprégnénolone contrairement à l’abiratérone qui inhibe les 2.

Le kétoconazole a déjà été utilisé comme inhibiteur de la CYP17A1 pour traiter le cancer de la prostate,138 et ce, malgré son manque de sélectivité avéré, voir son effet ubiquitaire. L’abiratérone est actuellement le seul inhibiteur (sélectif celui-là) de cette catégorie à être utilisé en clinique139 (Figure 2). Ce médicament doit être pris en combinaison avec de la prednisone pour compenser le blocage de la synthèse des corticostéroïdes qu’il cause, sans compter qu’il tend à augmenter la synthèse des minéralocorticoïdes ce qui explique en partie le suivi serré des patients qui prennent cet inhibiteur140. Un autre inhibiteur de la CYP17A1 est actuellement à l’étude (phase III), l’ortéronel138 (Figure 6). Celui-ci inhibe préférentiellement l’activité 17,20 lyase de l’enzyme ce qui devrait permettre une diminution des effets secondaires observés avec l’abiratérone (Figures 5 et 6).

17 3.2.3 Aromatase

Elle a déjà été appelée estrogène synthase (peu utilisé de nos jours) de par son rôle essentiel dans la formation des principaux estrogènes endogènes. En effet, l’aromatisation du cycle A stéroïdien par cette enzyme transforme irréversiblement des stéroïdes à caractère androgénique (4-dione, testostérone et 16α- hydroxyandrostènedione) en d’autres ayant un effet estrogénique (E1, E2 et E3).

Figure 7: Inhibiteurs de l’aromatase qui sont aujourd’hui utilisé pour empêcher la biosynthèse des estrogènes. Les substrats, qui ont un effet plus ou moins androgénique, et produits estrogéniques de l’enzyme sont aussi illustrés.

L’inhibition de l’aromatase apparait donc assez tôt comme une stratégie intéressante pour traiter des maladies stimulées par les estrogènes (Figure 5). Dès le début des années 70, l’aminoglutethimide a été utilisé pour le traitement du cancer du sein en tant qu’inhibiteur de la cholestérol desmolase (P450scc), mais elle a rapidement montré une qualité intéressante d’inhibition de l’aromatase141. Après une première génération d’inhibiteurs, une

deuxième fut développée avec cette fois une sélectivité appréciable, mais une efficience thérapeutique discutable. Les inhibiteurs de troisième génération se sont finalement montrés supérieurs au traitement de référence des maladies sensibles aux estrogènes, utilisant le tamoxifène, lors des essais cliniques142. Cette dernière génération (Figure 7), dont fait partie l’exémestane (stéroïdien), de même que l’anastrozole, le létrozole et le vorozole (non stéroïdiens), est un véritable succès de l’industrie pharmaceutique. À titre d’exemple, les ventes de l’anastrozole seulement s’élèvent à près de 2 milliards pour 2009143, bien que ces chiffres sont aujourd’hui moindres après l’expiration de plusieurs brevets.

Dans tous les cas, les inhibiteurs de l’aromatase sont d’un grand intérêt, autant pour l’aspect commercial que pour le traitement de la puberté précoce, de l’endométriose, du cancer du sein, léiomyome utérin, etc. Ce succès a encouragé le développement d’autres inhibiteurs de la synthèse des estrogènes pouvant avoir un intérêt thérapeutique dont il est discuté dans les prochaines sections.

3.2.4 Stéroïde sulfatase

La stéroïde sulfatase ou estrone sulfatase (STS) est indirectement impliquée dans la synthèse des estrogènes, mais joue un rôle prépondérant dans la disponibilité des précurseurs de l’estradiol. La DHEA et l’estrone sont majoritairement retrouvées sous une forme sulfatée, ce qui leur confère une demi-vie beaucoup plus longue que leur vis-à-vis non sulfaté, tout en leur faisant perdre leur affinité pour les récepteurs et les enzymes de biosynthèse des hormones. Ces formes inactives sont celles majoritairement retrouvées dans le sang, ce qui en fait une sorte de forme de transport des stéroïdes entre les glandes principales et les organes cibles. Pour que le stéroïde soit utilisable dans l’organe ciblé, il faut que le groupement sulfate soit retiré par hydrolyse, ce que fait la STS144.

Figure 8: Principal inhibiteur de la STS et principales activités de cette enzyme en bleu. Les flèches jaunes représentent l’action de sulfotransférases responsables de la réaction inverse de la STS.

Inhiber la STS permet donc de garder la DHEA et l’estrone sous une forme sulfatée inactive, ce qui les empêche d’être transformées en androgènes et/ou en estrogènes actifs dans les organes périphériques, où ces hormones exercent leur action (Figures 5 et 8). Des molécules capables d’exercer cette action ont fait l’objet de bien des travaux au cours des deux dernières décennies. Récemment, l’un de ces inhibiteurs, nommé Irosustat ou STX64 (Figure 8), a montré des résultats fort encourageants, malgré un manque d’efficacité lors d’essais cliniques de phase II145.

3.2.5 17β-Hydroxystéroïde déshydrogénase

Les 17β-hydroxystéroïdes déshydrogénases (17β-HSD) forment une superfamille responsable de l’interconversion des hormones sexuelles. D’un côté, les enzymes réductrices (types 1, 3, 5, 7, 12 et 15) vont activer les hormones sexuelles avec le NADPH comme cofacteur, de l’autre, leurs vis à vis oxydatives (types 2, 4, 6, 8, 9, 10, 11, 13 et 14) vont désactiver les hormones en utilisant le NAD+ (Figure 9). Si les deux premières isozymes sont connues depuis les années 1950, seulement trois avaient été découvertes il y a 20 ans146. Des 15 17β-HSD connues à ce jour, 12 sont exprimées chez l’humain dont trois seulement semblent exclusivement impliquées dans la stéroïdogenèse. Les autres exercent d’autres rôles dans la synthèse du cholestérol ou dans le métabolisme des acides gras par exemple. En fait, pour certains, leur implication dans le métabolisme des hormones sexuelles est fortement mise en doute147.

Figure 9: Principales activités dans la synthèse des hormones sexuelles rapportées pour les 17β-HSD, tiré de Samson, Labrie et Luu-The en 2009148. Pour la plupart, les formes actives se trouvent à droite et les inactivent à gauche de la figure. Les 17β-HSD sont responsables de l’interconvertion entre ces deux formes.

Bref, avec environ 20% d’homologie entre ces membres qui n’ont parfois que quelques éléments caractéristiques en commun, il s’agit d’une famille d’enzymes pour le moins diversifiées. Le développement d’inhibiteurs de 17β-HSD a commencé il y a plus de 30 ans et a principalement porté sur l’inhibition du type 1. Toutefois, aucun inhibiteur n’a été testé lors d’essais cliniques jusqu’à maintenant. N’en demeure pas moins qu’il y a un réel intérêt thérapeutique pour des inhibiteurs de certaines isozymes.

21 La 17β-HSD1 est responsable de l’activation d’E1 en E2, l’estrogène le plus puissant. Elle est également capable de transformer la DHEA, un stéroïde inactif, en Δ5-diol, un estrogène faible qui aurait un rôle crucial après la ménopause149. Il est également possible que la 17β-HSD1 soit capable de désactiver la DHT, l’androgène le plus puissant150. Cette enzyme spécifique aux estrogènes a été associée à des maladies comme l’endométriose, le syndrome métabolique de même que dans les cancers du sein, de l’endomètre, du côlon, de l’ovaire et de la prostate147. De plus, il semblerait y avoir un lien entre la pré-éclampsie et une expression augmentée de la 17β-HSD1151.

Au contraire, la 17β-HSD2 est la principale enzyme à catalyser la réaction inverse, soit de désactiver les estrogènes (E2 en E1 et Δ5-diol en DHEA), mais également les androgènes. Son implication importante dans la diminution des taux d’hormones sexuelles actives qui ont un effet bénéfique au niveau des os en fait une cible attrayante pour le traitement de l’ostéoporose152. Cette enzyme est exprimée entre autres dans le foie, l’intestin, l’endomètre, le pancréas, la prostate et le sein147.

Tout comme pour la 17β-HSD1, l’expression de la 17β-HSD3 est beaucoup plus confinée à des tissus spécifiques. En fait, on ne la retrouve que dans les testicules. Globalement, son action est spécifique à la transformation des androgènes peu actifs en d’autres plus actifs comme la testostérone et la DHT153. Cette enzyme exerce un rôle important dans la production des androgènes mis en évidence par le fait qu’une déficience en 17β-HSD3 est la cause du pseudohermaphrodisme, un trouble du développement dans lequel le mâle développe des caractéristiques sexuelles « féminisées »154. Inhiber cette enzyme s’apparenterait à une castration chimique qui pourrait être intéressante pour traiter le cancer de la prostate entre autres155.

L’autre enzyme de cette famille impliquée dans l’activation des androgènes est la 17β-HSD5. Contrairement à l’isozyme de type 3, son expression est beaucoup plus étendue jusqu’à être considérée pratiquement ubiquitaire et son action n’est pas restreinte qu’aux androgènes147. Un inhibiteur de cette isozyme pourrait être un traitement intéressant du cancer de la prostate considérant qu’environ la moitié des androgènes actifs qui s’y trouvent sont produits localement et ne sont donc pas affectés par la production des testicules153. Il s’agit de la seule 17β- HSD qui appartient à la famille des aldo-céto réductases (AKR), alors que les autres appartiennent à la famille des déshydrogénases de courtes chaînes (SDR). C’est pourquoi elle porte également le nom de AKR1C3156.

La 17β-HSD10 est une autre des isozymes ayant une expression quasi ubiquitaire. Elle oxyde des substrats aussi différents que des acides gras, des acides biliaires, des acides aminés, des hormones sexuelles et d’autres stéroïdes. Inhiber cette enzyme a été proposé comme un traitement potentiel de l’Alzheimer147. Il a d’ailleurs été montré que la 17β-HSD10 interagissait avec la bêta-amyloïde, un peptide souvent évoqué pour son lien avec certaines maladies neurodégénératives157.

3.3 Inhibiteur de la 17β-HSD1

3.3.1 Intérêt thérapeutique

Les efforts pour développer un inhibiteur de la 17β-HSD1 visent principalement à ajouter une option thérapeutique pour le traitement du cancer du sein et de l’endométriose qui atteindraient respectivement environ 1/9 et 1/10 des femmes au cours de leur vie. Dans ces deux maladies, une surexpression de l’enzyme par rapport au tissu sain a été rapportée. Ces résultats restent à confirmer, mais il n’en demeure pas moins que le taux E2/E1 est augmenté dans ces deux pathologies ce qui cause une stimulation estrogénique délétère accrue121,158.

Bien qu’encore une fois les chiffres divergent, environ 50% des tumeurs mammaires et pratiquement tous les tissus endométriotiques exprimeraient la 17β-HSD1 et plus de la moitié de chacun expriment ER121,159,160. À noter que l’importance de GPER dans le développement de ces maladies est de mieux en mieux connue et ils ne devraient pas être négligés122,161,162. Actuellement, les SERM et les inhibiteurs de l’aromatase utilisés respectivement avant et après la ménopause seraient les meilleurs traitements pharmacologiques de première ligne pour les cancers du sein exprimant ER163. Toutefois, ces traitements ont souvent une efficience limitée pour l’endométriose. Pour les inhibiteurs de l’aromatase, la faible expression de cette enzyme dans l’endomètre est possiblement en cause121. Dans tous les cas, les effets secondaires sont souvent un frein à l’utilisation des médicaments existants pour les femmes atteintes120.

Ainsi, un inhibiteur de la 17β-HSD1 pourrait permettre de bloquer la production de Δ5-diol, ce que les inhibiteurs de l’aromatase ne peuvent faire. Sachant que ces derniers seraient les meilleurs traitements du cancer du sein hormono-dépendant chez les femmes post-ménopausées164, une combinaison avec un inhibiteur de la 17β- HSD1 pourrait être une nouvelle option thérapeutique avantageuse. Sans compter qu’il semblerait qu’un traitement à partir d’un inhibiteur d’aromatase augmente l’expression de la 17β-HSD1165.

Inhiber plus spécifiquement la dernière étape de la production d’E2, l’estrogène physiologiquement actif et le seul ligand endogène connu de GPER pourrait signifier un traitement comportant moins d’effets secondaires. Il s’agit d’un net avantage par rapport au traitement en usage dont l’action est moins spécifique, particulièrement pour le traitement de l’endométriose. Une molécule bloquant cette stimulation du récepteur membranaire des estrogènes pourrait être un avantage par rapport aux antagonistes utilisés qui sont des agonistes de ce récepteur20. Cet effet stimulant pourrait expliquer, en partie à tout le moins, l’avantage que peuvent avoir les inhibiteurs de l’aromatase. En général, considérant les effets diversifiés des estrogènes, cibler spécifiquement

23 les effets indésirables des estrogènes dans une situation donnée sans affecter des processus essentiels est un défi important qui peut nécessiter davantage de molécules aux effets plus spécifiques comme un inhibiteur de la 17β-HSD1.

3.3.2 Avancées actuelles

Figure 10: Structure des inhibiteurs de la 17β-HSD1 les plus prometteurs de différentes séries rapportées.

Jusqu’à présent, après des années à développer des inhibiteurs de la 17β-HSD1, quelques laboratoires ont rapporté des molécules efficaces, sélectives et dépourvues du moindre effet estrogénique même résiduel146. Un groupe de Solvay Pharmaceutical notamment a synthétisé un inhibiteur constitué d’un noyau estrone substitué en position 15. Ils ont rapporté une forte inhibition et une absence de liaison à ERα in vitro par ce dernier, en plus de démontrer son efficacité et son manque d’effet estrogénique in vivo166,167. Il y a également le groupe de Michael J. Reed et Barry V.L. Potter qui a publié un inhibiteur substitué en position 16 d’un noyau estrone (STX1040) ayant des caractéristiques semblables (Figure 10)168.

Du côté des inhibiteurs non-stéroïdiens, des dérivés de l’acide coumarique et des hydroxybenzothiazoles rapportés par des groupes slovènes et allemands ont également montré des effets intéressants169–171. Parmi les inhibiteurs de la 17β-HSD1, certains des plus prometteurs, ayant démontré à la fois leur puissance, leur sélectivité et leur efficacité in vivo font partie des (hydroxyphenyl)naphtols publiés par le groupe de Rolf W. Hartmann (Figure 10)172.

Au sein de notre laboratoire, la majorité de nos efforts ont porté sur des inhibiteurs stéroïdiens depuis une vingtaine d’années173. Jusqu’à tout récemment, l’estrogénicité des inhibiteurs a été un problème important, malgré le développement de composés relativement puissants substitués en position 16 de l’estradiol174. L’ajout d’un groupement éthyle halogéné au niveau du carbone trois a permis d’obtenir des inhibiteurs efficaces et dépourvus d’effet estrogénique autant in vitro qu’in vivo175,176. Notamment, le PBRM, le fruit principal de cette série, a totalement annulé la stimulation induite par E1 de tumeurs xénogreffées chez la souris (Figure 6).

4. Activateurs enzymatiques

Alors que les inhibiteurs d’enzymes inondent les tablettes de nos pharmacies… et accessoirement de nos laboratoires, il est facile d’oublier qu’il existe également des activateurs d’enzymes. De telles molécules sont loin d’être courantes, mais elles sont généralement de grand intérêt pour la recherche, voire thérapeutique dans certains cas177.

4.1 Définition et mécanismes

Un activateur d’enzyme est une molécule qui lie directement une enzyme pour en stimuler l’activité177. Un rapide coup d’œil dans la littérature permet de voir que ces molécules sont souvent confondues avec les inducteurs d’enzyme, qui en augmentent l’expression, et/ou les autres molécules qui stimulent indirectement l’activité d’une enzyme. À titre d’exemple, l’effet d’activateur direct du resveratrol, pourtant l’un des activateurs ayant été les plus étudiés, est aujourd’hui mis en doute178.

Très peu d’enzymes sont connues pour être activées par des ligands endogènes, mais on peut supposer que de nombreuses autres le sont également (Figure 11). Le fait est que de tels mécanismes étant difficiles à identifier, leur nombre est certainement plus élevé que ce qui est actuellement connu179. L’existence même d’une molécule capable d’activer directement une enzyme sous-tend qu’il existe un mécanisme de régulation endogène de l’activité de cette même enzyme ce qui constitue une information cruciale sur une enzyme.

25 Il n’y aurait que 12 enzymes pour lesquelles le mode d’activation a été identifié et caractérisé177. Ceux-ci peuvent être séparés entre 4 types de mécanismes (Figure 11). Dans tous les cas, la liaison du ligand à un site allostérique ou sur une sous-unité régulatrice favorise une forme active de l’enzyme. L’un de ces mécanismes implique une activation irréversible via une modification post-traductionnelle de l’enzyme tandis que les autres sont réversibles.

Figure 11: Types de mécanismes d’activation enzymatique par des petites molécules identifiés jusqu’à maintenant et les enzymes pour lesquelles un tel mode d’action est connu, tiré de Zorn et Wells en 2010177. Les mécanismes de type A impliquent la liaison de sites allostériques et ceux de types B celle de sous-unités par l’activateur. Cela a pour effet de rendre le site catalytique plus disponible et/ou actif en causant le clivage d’une partie d’une proenzyme dans le cas de A2 et la dimérisation dans celui de B2.

4.2 La 17β-HSD12

Aucun activateur de cette enzyme n’avait été rapporté dans la littérature avant le début de nos travaux sur le sujet. En fait, la connaissance de son existence même est relativement récente. Elle a d’abord été rapportée sous le nom de 3-cétoacyl-CoA réductase (KAR) en 2003 pour son rôle dans l’élongation des acides gras180, avant d’être « découverte » à nouveau en 2006 en tant que 17β-HSD12 pour une implication dans la synthèse des estrogènes181. À l’heure actuelle, les travaux sur cette enzyme demeurent rares avec à peine une vingtaine d’articles sur le sujet dans la dernière décennie.

L’ombre d’une controverse demeure à savoir si la 17β-HSD12 est réellement capable de transformer E1 en E2 dans un contexte physiologique, mais la grande majorité des auteurs tendent à dire que non. Dans tous les cas, le rôle principal, voire essentiel, de l’enzyme est de catalyser l’une des 4 étapes de l’élongation des acides gras. Elle est conséquemment exprimée ubiquitairement, mais en plus grande quantité dans le foie.

Figure 12: Structure du premier activateur de la 17β-HSD12 à être décrit.

Dans ce contexte, la découverte d’une activation endogène ayant un effet sur l’une et/ou l’autre des activités de l’enzyme pourrait avoir de grandes répercussions sur la manière dont on la considère. D’autant plus que, considérant le peu d’outils disponibles pour l’étudier, un activateur enzymatique pourrait grandement contribuer à la recherche. Jusqu’à maintenant, peu d’inhibiteurs de cette enzyme ont été rapportés et ceux qui l’ont été manquent d’efficacité et autant que de sélectivité182. Du côté des activateurs, on peut affirmer qu’un stéroïde synthétisé dans notre laboratoire (Figure 12) stimule la transformation de l’estrone en estradiol par ce qui semble être une activation directe de la 17β-HSD12. Beaucoup de travail reste à faire cependant.

2EMHFWLIVHWDSHUoXGHVWUDYDX[

La majorité des efforts ont été centrés sur l’étude de l’action du PBRM dans le but de promouvoir le développement de cette molécule183. Cela passait par 3 objectifs principaux :

1. Caractériser le mode d’action du PBRM

2. Poursuivre l’évaluation de l’efficacité de l’inhibiteur in vitro et in vivo

3. Vérifier la sélectivité du PBRM pour la 17β-HSD1 par rapport à d’autres enzymes clefs

À cette fin, son mode d’action de même que sa sélectivité ont été abordés dans le chapitre 1. Il est vite apparu qu’il s’agissait d’un inhibiteur irréversible de la 17β-HSD1 qui n’affecte pas quatre des autres membres

27 importants de cette famille, soit les types 2, 3, 7 et 12. L’article en question porte plus largement sur la découverte de la molécule parmi des séries de composés stéroïdiens substitués en position 3 synthétisés pour annuler l’effet estrogénique résiduel d’un précurseur. Des essais de modélisation indiquent également que le PBRM interagirait avec l’enzyme via la formation d’un lien covalent avec une histidine potentialisée par un glutamate. Ces deux acides aminés du site catalytique sont normalement impliqués dans la reconnaissance du substrat naturel159.

Cette hypothèse est ensuite appuyée par des essais d’inhibitions in vitro et in silico de 17β-HSD1 orthologues dans le chapitre 2. Ce dernier évalue également le potentiel d’inhibition irréversible du PBRM et la durée de son effet en cellules intactes. Ces essais ont également permis d’estimer une valeur pour le taux de resynthèse de la 17β-HSD1 jusqu'alors inconnu. La caractérisation du profil pharmacocinétique est aussi rapportée dans ce chapitre, mais il s’agit d’un travail en cours.

En parallèle de la réalisation de ces objectifs, d’autres projets ont été entamés au cours des dernières années, mais n’ont pas été aussi fructueux. C’est notamment le cas de l’évaluation du potentiel d’un inhibiteur de la 17β- HSD1 dans le cancer de la prostate. Les caractéristiques d’inhibition et d’estrogénicité d’autres séries de composés dont certains inhibiteurs non stéroïdiens ont aussi été évaluées184. Ceux-ci ont fait l’objet de publications et ce sera vraisemblablement le cas éventuellement pour des séries d’analogues du PBRM (dérivés de E1 et E2 substitués en position 3) ou d’autres sans substitution en position 3 et ayant un cycle [1,3] oxazinan- 2-one en position 17.

Toutefois, cette dernière série a permis de faire une découverte étonnante rapportée dans le chapitre 3. Afin d’évaluer la sélectivité du PBRM, un essai d’inhibition de la 17β-HSD12 a été fait en cellules intactes. Du même coup, quelques autres composés du laboratoire ont été testés. L’un d’entre eux, une molécule de prime abord peu intéressante de la série des carbamates, a triplé l’activité de l’enzyme contre toute attente. Comme il s’agissait d’une première pour une 17β-HSD, un autre volet avec de nouveaux objectifs s’est ajouté :

A. Vérifier l’hypothèse que le composé identifié soit un activateur enzymatique

B. Évaluer la relation structure-activité de l’effet observé

L’effet d’activation a été caractérisé dans diverses cellules et dans une préparation microsomale ce qui est l’objet du chapitre 3. En travaillant sur l’objectif B, de nouveaux activateurs dont certains plus puissants ont été découverts. Malgré le petit nombre de molécules effectives, certains groupements importants pour l’effet d’activation ont été identifiés. Ces résultats n’ont pas été publiés jusqu’à maintenant.

Chapitre 1

29 Résumé

La 17β-HSD1 est vue comme jouant un rôle clef dans la progression de maladies sensibles aux estrogènes comme le cancer du sein et l’endométriose. Trois séries successives de dérivés substitués en C3 du 16β-(m- carbamoylbenzyl)-E2, un inhibiteur puissant, ont été synthétisées afin d’éliminer l’effet estrogénique indésirable de ce composé phare. Ainsi, la synthèse et la caractérisation de même que l’évaluation de la capacité à inhiber la 17β-HSD1 et à induire une prolifération estrogénique de 20 nouveaux dérivés d’E2 sont rapportées.

Une étude de relation structure-activité a permis d’obtenir un nouvel inhibiteur stéroïdien puissant et dépourvu d’effet estrogénique; le 3-{[(16β,17β)-3-(2-bromoethyl)-17hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide, plus simplement appelé PBRM. Ce dernier a inhibé la transformation de E1 en E2 en cellules T47D (IC50 = 83 nM) par la 17β-HSD1 sans inhiber significativement les types 2, 7 et 12 ou la CYP3A4. La cinétique d’inhibition et la modélisation suggèrent finalement un mode d’action irréversible et compétitif pour le PBRM.

Journal of Medicinal Chemistry, 2014, 57(1), 204-222

Discovery of a non-estrogenic irreversible inhibitor of 17β- hydroxysteroid dehydrogenase type 1 from 3-substituted-16β-(m- carbamoylbenzyl)-estradiol derivatives.

René Maltais,1 Diana Ayan,1 Alexandre Trottier,1 Xavier Barbeau,2 Patrick Lagüe,3 Jean-Emmanuel Bouchard,1 and Donald Poirier1*

1Laboratory of Medicinal Chemistry, Oncology and Nephrology Unit, CHU de Québec - Research Center (CHUL, T4-42) and Faculty of Medicine, Laval University, Québec City, QC, Canada.

2Département de chimie, Institut de biologie intégrative et des systèmes (IBIS), and Centre de recherche sur la fonction, la structure et l'ingénierie des protéines (PROTEO),Université Laval, Québec City, QC, Canada.

3Département de biochimie microbiologie et bio-informatique, Institut de biologie intégrative et des systèmes (IBIS), and Centre de recherche sur la fonction, la structure et l'ingénierie des protéines (PROTEO),Université Laval, Québec City, QC, Canada.

* Corresponding author:

Dr. Donald Poirier

Laboratory of Medicinal Chemistry

CHU de Québec - Research Center (CHUL, T4-42)

2705 Laurier Boulevard, Québec City, QC, G1V 4G2, Canada

Phone: 1(418) 654-2296; Fax : 1(418) 654-2761

E-mail: [email protected]

Abstract

17β-Hydroxysteroid dehydrogenase type 1 (17β-HSD1) is thought to play a pivotal role in the progression of estrogen-sensitive breast cancer by transforming estrone (E1) into estradiol (E2). We designed three successive series of E2-derivatives at position C3 of the potent inhibitor 16β-(m-carbamoylbenzyl)-E2 to remove its unwanted estrogenic activity. We report the chemical synthesis and characterization of 20 new E2-derivatives, their evaluation as 17β-HSD1 inhibitors, and their proliferative (estrogenic) activity on estrogen-sensitive cells. The structure–activity relationship study provided a new potent and steroidal nonestrogenic inhibitor of 17β- HSD1 named 3-{[(16β,17β)-3-(2-bromoethyl)-17-hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide (23b). In fact, this compound inhibited the transformation of E1 into E2 by 17β-HSD1 in T-47D cells (IC50 = 83 nM), did not inhibit 17β-HSD2, 17β-HSD7, 17β-HSD12, and CYP3A4, and did not stimulate the proliferation of estrogen- sensitive MCF-7 cells. We also discussed the results of kinetic and molecular modeling (docking) experiments, suggesting that compound 23b is a competitive and irreversible inhibitor of 17β-HSD1.

Introduction

Breast cancer is the most frequent cancer among American women, with an estimate of 288130 new cases diagnosed and 39520 related deaths in 2011.(1, 2) This disease is more frequent in postmenopausal women, representing about three-quarters of total breast cancer cases. Importantly, almost 65% of postmenopausal breast cancer tumors are known to be estrogen-dependent in their progression.(3) 17β-Hydroxysteroid

dehydrogenase type 1 (17β-HSD1) is suspected to play a pivotal role in the progression of these breast cancers, transforming estrone (E1) into estradiol (E2), the most potent endogenous ligand for estrogen receptors (ERα and ERβ).(4-6) 17β-HSD1 also catalyzes the reduction of dehydroepiandrosterone (DHEA) into 5-androstene- 3β,17β-diol (5-diol), a weaker estrogen that becomes more important after menopause and could also stimulate breast cancer cell proliferation.(7) Importantly, several immunohistochemical studies have reported the presence of 17β-HSD1 in breast carcinoma tissue and the enzyme was detected in approximately 50–60% of cancer cells.(8-14) Furthermore, a significant increment of 17β-HSD1 expression following antiaromatase therapy in breast cancer patients has also been observed, suggesting a compensatory response of the enzyme to estrogen depletion.(15) These observations suggest that a 17β-HSD1 inhibitor could be useful in blocking estrogen biosynthesis in a large number of breast cancers and could be advantageous toward a maximal estrogen biosynthesis blockade, considering that aromatase inhibitors are unable to block 5-diol production (Figure 1).

17-HSD 1 Pathway I 5-diol (complementary)

CHOL DHEA Estrogenic effect (proliferation of breast 3-HSD cancer cells)

Aromatase 17-HSD 1 Pathway II 4-dione E1 E2 (classical)

Figure 1. Two pathways involved in the formation of strong estrogen E2 (pathway II) and weak estrogen 5-diol (pathway I) from key steroid DHEA.

With the emergence of personalized medicine and diagnostic tests, the arrival of a potent 17β-HSD1 inhibitor in a clinical setting is highly expected to give a new option for the treatment of women detected with a high expression of 17β-HSD1 and a low expression of aromatase in breast cancer tumor biopsies. Finally, the use of 17β-HSD1 inhibitors is also a promising approach for the treatment of other estrogen-dependent diseases, such as endometrial cancer(16) and endometriosis,(17) where the enzyme has been shown to be overexpressed.(18- 20)

Over the past 30 years, many efforts were dedicated to designing potent inhibitors of the key steroidogenic enzyme 17β-HSD1, but it is only recently that lead candidates have been reported with very high inhibitory activity.(21-28) The presence of residual estrogenic activity associated with steroidal inhibitors, which are often built around an estrane nucleus, is a major drawback to their development and their use as therapeutic agents. Thus, despite the strong potential of 17β-HSD1 inhibitors for the treatment of estrogen-dependent diseases, validation of this pharmaceutical target remains to be confirmed. New potent and specific inhibitors with a

33 nonestrogenic profile as well as selectivity toward other 17β-HSD isoforms, especially type 2, are thus strongly needed to validate the therapeutic in vivo approach and to engage the first clinical trial with human subjects.

16β-(m-Carbamoylbenzyl)-E2 (1) has already been reported as a potent inhibitor of 17β-HSD1.(29, 30) Despite its good inhibitory potency, this compound was found to stimulate the MCF-7 and T-47D estrogen-sensitive breast cancer cell lines in vitro, thus compromising its therapeutic potential.(29) To remove the undesirable residual estrogenic activity of E2 derivative 1, we explored the impact on both 17β-HSD1 inhibition and estrogenicity of a series of differently functionalized small chains in the replacement of the hydroxyl (OH) group at position 3. In fact, this OH is well-known to be very important for ER binding affinity and, consequently, to produce an estrogenic effect.(31) Because replacing the 3-OH group by a hydrogen atom did not allow full blockade of the estrogenic activity, as assessed by the proliferation of estrogen-sensitive cells,(29, 32) additional modifications should be tested. To reach a new anchoring point with an amino acid in proximity to position 3 of compound 1 (Figure 2),(33) and to potentially remove the undesirable estrogenic activity by disturbing the binding on ER, we selected different functional groups and side chain lengths (0, 1, or 2 carbon spacer) in order to promote hydrogen bonding (B(OH)2, NH2, F, CONH2, COOH, CH2OH, CONR1R2, CH2NH2) or hydrophobic interactions (CH3, CH2Br, CH2Cl, CH2I, CH2Ph).

Figure 2. Key interactions observed in a ternary complex of 17β-HSD1/inhibitor 1/cofactor NADP and representation of new E2 derivatives modified at position 3 (series I, II, and III). The scope of this side-chain (R) is dual: (1) reaching a third interaction with an amino acid and (2) removing the undesirable estrogenic activity of the first generation inhibitor 1.

The successive synthesis of three series of E2 derivatives, combined with structure–activity relationships (SAR), provided a new potent and nonestrogenic steroidal inhibitor of 17β-HSD1 named 3-{[(16β,17β)-3-(2-bromoethyl)-

17-hydroxyestra-1(10),2,4-trien-16-yl]methyl} benzamide (23b). After publication of preliminary results,(34) we now report the full details of chemical synthesis and characterization of 20 new E2 derivatives, their evaluation as 17β-HSD1 inhibitors, and their proliferative (estrogenic) activity on estrogen-sensitive cells. We also determine the selectivity of the best inhibitor 23b for other enzymes and used molecular modeling to obtain insight into its binding mechanism.

Results and Discussion

Chemistry

Compounds reported in this SAR study were all synthesized from E1 as a common synthetic precursor (Schemes 1–4). The 3-substituted-16β-(m-carbamoylbenzyl)-E2 derivatives were regrouped in three different series (I–III) relative to the spacers present between the functional group and the A-ring of steroid core (Figure 2).

Series I Compounds (Schemes 1 and 2)

The functional groups (B(OH)2, NH2, F, CONH2, COOH, CH2OH, CONR1R2, CH2Br, and CH2NH2) that constituted the first series of derivatives were selected with the intention of exploring the tolerance of the enzyme for substituents of different natures (H-donor, H-acceptor, hydrophilic, and hydrophobic) and to provide useful SAR data. These functional groups were directly attached to position 3 of the A-ring as illustrated with compounds 7, 8, 9, 10, 11, 12a–c, 13, 14, and 15a–d (Table 1). Our strategy toward obtaining these compounds consisted in functionalizing E1 with the appropriate substituent at C3 and then introducing the m- carbamoylbenzyl moiety at C16β. The inverse strategy consisted in introducing the different chains at C3 after the installation of carbamoylbenzyl moiety and was much more uncertain considering the reactivity of the carboxamide functionality. Thus, we first prepared the intermediates 2 (boronate-ester),(35) 3 (amino),(36) 4 (fluoro),(37) 5 (carboxamide),(38) and 6 (carboxy)(39) following previously reported synthetic procedures (Scheme 1). These intermediates were then submitted to a sequence of three chemical steps to give their corresponding 16β-(m-carbamoylbenzyl) derivatives 7, 8, 9, 10, and 11. These three steps, already reported for the synthesis of 1,(29, 40, 41) consisted in an aldolization reaction with the 3-carboxamide-benzaldehyde followed by a reduction with NaBH4 and a catalytic (Pd/C) hydrogenation of the allylic alcohol. This sequence of

35 reactions gave low to modest yields (6–49% for 3 steps) depending of the 3-substituted E1 derivative used as starting material, with the lower yields observed for the boronic acid 7, aniline 8, and carboxamide 10.

Scheme 1. Reagents and conditions: (a) 3-carboxamide-benzaldehyde, KOH, EtOH, rx; (b) NaBH4, MeOH; (c) H2, Pd/C, MeOH, rt.

The other derivatives of series I, compounds 12a–c, 13, 14, and 15a–d, were synthesized from 3-carboxy derivative 11 (Scheme 2). An amidation of the mixed anhydride generated from the 3-carboxy group of 11 with appropriate amines using BOP as a coupling agent gave the carboxamide derivatives 12a–c in low yields (13– 30%). The 3-methylalcohol derivative 13 was obtained by activating the carboxy group with BOP and then reducing the anhydride with NaBH4. The bromide derivative 14 was obtained in good yield (60%) by bromination of 13 using triphenylphosphine and carbon tetrabromide in DCM. From 14, we generated the corresponding amines 15a–c by a nucleophilic displacement reaction using the appropriate amines in DCM and TEA as base. The amine 15d was however synthesized in two steps by using first the sodium azide in DMF for a substitution of the bromide and next by reducing the intermediate azide.

Scheme 2. Reagents and conditions: (a) R1R2NH, BOP, DIPEA, DMF, rt; (b) (i) BOP, DIPEA, THF, rt, (ii) NaBH4, rt; (c) PPh3, CBr4, DCM, rt; (d) NHR1R2, Et3N, DCM, rt; (e) (i) NaN3, DMF, 60 °C, (ii) H2, Pd/C (10%), MeOH, rt.

Series II Compounds (Scheme 3)

In this second series of compounds represented by 18, 20, 22, 23a–c, 24, and 25, one additional carbon spacer

(CH2) was introduced between the functional group (CH2Ph, CH3, CH2OH, CH2Cl, CH2Br, CH2I, COOH, and

CONHCH3) and the A-ring of the steroid core (Scheme 3). The vinyl group was identified as the common functionality for the synthesis of each compound of this series. Thus, the key intermediate 16 was obtained from E1 by a dioxolane protection of the C17-ketone,(42) an activation of the phenol by a triflate formation, and a carbonylative vinylation. The 2-phenylethyl derivative 18 was obtained from a metathesis reaction between the vinyl intermediate 16 and styrene using the Grubb (II) catalyst, followed by deprotection of dioxolane to give 17, and the addition of the 16β-(m-carbamoylbenzyl) moiety using the previously described three-step sequence of reactions. The synthesis of the ethyl derivative 20 was generated by deprotection of 16 and installation of the 16β-side chain. The oxidative hydroboration of 16 using a dimethylsulfide borane complex and hydrogen peroxide gave the corresponding primary alcohol, which was protected as the benzyl ether 21. This compound was next transformed to 22 using the three-step sequence of reactions. The last step, the catalytic hydrogenation, also allows regeneration of the free alcohol. The hydroxyl group of 22 was substituted using

(chloro-phenylthio-methylene)dimethylammonium chloride (CPMA),(43) CBr4 and PPh3, or NaI in acetone to provide the chloride 23a, bromide 23b, or iodide 23c. The alcohol 22 was also transformed into the corresponding carboxylic acid 24 in three steps: a) a Dess−Martin oxidation, b) an in situ oxidation with NaClO2, and c) a NaBH4 reduction of the C17-ketone. Finally, the N-methylamide 25 was obtained from 22 by activating the carboxylic acid group the intermediate keto acid with BOP reagent to permit the methylamine nucleophilic displacement.

Scheme 3. Reagents and conditions: (a) styrene, Grubb (II) catalyst, dichloroethane, reflux; (b) HCl 10% in MeOH, rt; (c) 3-carboxamide-benzaldehyde, KOH, EtOH, reflux; (d) NaBH4, MeOH, rt; (e) H2, Pd/C (10%), MeOH, rt; (f) (i) BH3-DMS, THF, −78 °C, (ii) H2O2, NaHCO3; (g) NaH, benzylbromide; (h) CPMA, DCM, rt; (i) PPh3, CBr4, DCM, rt; (j) NaI, acetone, rt; (k) (i) Dess–Martin reagent, DCM, rt, (ii) NaClO2, t-BuOH, 2-methyl- butene, KH2PO4, rt; (l) CH3NH2 in THF, BOP, DIPEA, DMF, rt.

Series III Compounds (Scheme 4)

Compounds 27, 28, and 30 contain a long spacer ((CH2)3 or OCH2CH2) at position 3 of the steroid nucleus. This series was designed to see the impact of spacer length on 17β-HSD1 inhibitory activity of compound 23b, the best inhibitor identified in the series II. The 3-vinyl-17-dioxolane-estra-1(10),2,4-triene (16) was first submitted to a metathesis reaction with the allyloxymethyl-benzene to give 26 in a low yield. The usual sequence of reactions was next used to install the 16β-side chain to give the diol 27. The primary alcohol of 27 was then transformed into bromide 28 using CBr4 and PPh3. Finally, compounds 29 and 30, two analogues of 27 and 28 both bearing an oxygen atom rather than a CH2 group as point of attachment on the ring A (OCH2CH2 instead of CH2CH2CH2), were synthesized in three steps. Starting from 1, an O-allylation with allylbromide followed by the ruthenium-catalyzed oxidative cleavage of the terminal olefin gave the corresponding aldehyde,(44) which was then reduced with NaBH4 to give the alcohol 29. This compound was next transformed to bromide 30. We were particularly interested by derivative 30 considering the presence of an oxygen atom directly attached to the

steroid A-ring. In fact, the CH2CH2O side chain of 30 could allow supplemental interaction with the enzyme compared to the (CH2)3 spacer of 28.

Scheme 4. Reagents and conditions: (a) Grubb II catalyst, allyloxymethyl-benzene; (b) HCl 10% in MeOH, rt; (c) 3-carboxamide-benzaldehyde, KOH, EtOH, reflux; (d) NaBH4, MeOH, rt; (e) H2, Pd/C (10%), MeOH, rt; (f) PPh3, CBr4, DCM, rt; (g) allylbromide, NaOH, acetone, reflux; (h) (i) NaIO4, RuCl3–H2O, EtOAc/ACN, 0 °C, (ii) NaBH4, THF:H2O (1:1), rt.

Optimization of the Chemical Synthesis of 23b (Scheme 5)

Small quantities of compounds 23b and 23c were initially obtained for the purpose of in vitro assays, but larger quantities were thereafter necessary for in vivo assays. We thus designed a shorter and more efficient chemical route that enabled the preparation of multigram quantities of 23b or 23c. Briefly, 3-vinyl-estra-1(10),2,4-triene- 17-one (19)(45) was oxidized with oxone(46) to give the oxirane 31 in excellent yield. The epoxide group of 31 was then transformed to primary alcohol 32 by performing a regiospecific palladium catalyzed transfer hydrogenation(47) using ammonium formate and palladium on charcoal in refluxing methanol. The usual sequence of three reactions for the introduction of the 16β-carbamoyl-m-benzamide side chain(34) was used to provide the diol 22 in a very good yield of 84%. Finally, the bromination of 22 using triphenylphosphine and carbon tetrabromide in DCM gave the bromide 23b. With only eight steps compared to 10 steps (from estrone), this new strategy to generate 23b was found advantageous over the first one. Importantly, the second chemical synthesis of 23b was achieved with a global yield of 17% compared to 7% for the first synthesis. Interestingly, only four chromatographic purifications were required along the chemical synthesis, thus reducing the time and cost needed to obtain multigram of 23b in very good HPLC purity of 98.5%.

Structure–Activity Relationship Study

Series I Compounds (Inhibition of 17β-HSD1 and Estrogenicity)

The first series of 3-substituted-16β-(m-carbamoylbenzyl)-E2 derivatives contains 15 compounds with substituents of different natures that were selected to study the impact on enzyme inhibition of various factors like hydrophobicity, H-bond donor/acceptor capacity, electrostatic charge, and steric hindrance. Compounds were tested for their ability to inhibit the transformation of E1 into E2 by 17β-HSD1 in intact T-47D cells, a cell line that expresses endogenous 17β-HSD1.(30, 48) As an initial observation (Table 1), five compounds of this series showed a significant inhibition of 17β-HSD1 (>30% at 0.1 μM), including boronic acid 7, amines 8 and 15d, methylalcohol 13, and bromomethyl 14, whereas nine compounds gave weak inhibition values (<30% at 0.1 μM), including fluoride 9, amides 10 , 12a, 12b, and 12c, carboxylic acid 11, and tertiary amines 15a, 15b, and 15c. When we consider these results more closely, we observe a good inhibition for derivatives with a potential H-bond donor group like primary amines, boronic acid, and alcohol. This could be due to the capacity of these substituents to form H-bond with Glu-282 or His-221 as previously observed for the phenolic derivative 1 (Figure 2). For example, the aromatic amine 8 (31% of inhibition at 0.1 μM) showed a lower inhibition value than the aminomethyl 15d (44% at 0.1 μM). This could be explained by the fact that the NH2 of aniline 8 is a poor H-bond acceptor group contrary to the CH2NH2 group of 15d, which has an acceptor/donor capacity close to that of the OH of 1, a phenol that can form two H-bond interactions with Glu-282 and His-221. A protonation of the CH2NH2 group of 15d could also lead to a stronger interaction with Glu-282 and His-221. This limited H-

bond acceptor capacity of 8 could thus explain its lower activity compared to 1 and 15d. Having a H-bond acceptor/donor capacity, the primary alcohol 13 gave an inhibition close to that of the corresponding primary amine 15d (37 and 44% at 0.1 μM) and lower than the inhibition of phenol 1 (70%). Also, the interaction between the substituent at position C3 and the enzyme seems to be influenced by the distance between the functionality and the amino acid to favor the formation of H-bond as seen by the difference of activity between the alcohol 13 and phenol 1.

The very low inhibition at 0.1 μM of the carboxamide 10 (2%) compared to aminomethyl 15d (44%), H-donor and H-acceptor, and of dimethylcarboxamide 12a (1%) compared to corresponding amine 15a (19%), only H- acceptor, reveals that the flexibility and orientation of the substituent to deliver the H-donor group could be an important factor for enzyme inhibition. The presence of a C═O, responsible for the difference of basicity between the amide and amine functionalities or producing a repulsive interaction with Glu-282, could also explain the lower inhibitory activity of amides 10 and 12a. Furthermore, in the case of a tertiary amine like dimethylamine 15a, where the amine acted only as H-bond acceptor group via the lone electron pair on nitrogen, the interaction with His-221 seems to be less favored than in the case of the corresponding primary amine 15d (H-donor and H-acceptor groups). However, the lower inhibition of tertiary amines could also be explained by steric hindrance of the dimethyl group of 15a (182.5 cm3/mol), and a loss of inhibitory activity becomes obvious with more bulky chains like diethylamine 15b (294.5 cm3/mol) and pyrolidine derivative 15c (238.5 cm3/mol). Despite these first interesting SAR observations, the most surprising result of the series was the good inhibition (30% at 0.1 μM) obtained with the hydrophobic bromomethyl substituent (compound 14), which we had not anticipated.

41 Table 1. Inhibition of 17β-HSD1 (Serie I)

a b compd no. R1 inhibition % estrogenicity at 0.1 μM at 0.1 μM at 1.0 μM 1 –OH 72 ± 6 +++ +++ 7 –B(OH)2 36 ± 4 +++ +++ 8 –NH2 31 ± 6 – + 9 –F 10 ± 5 – + 10 –CONH2 2 ± 2 – ++ 11 –COOH 3 ± 3 + +++ 12a –CON(CH3)2 1 ± 4 + ++ 12b –CON(Et)Pr 10 ± 12 – – 12c –CON(CH2)4 3 ± 3 – – 13 –CH2OH 37 ± 14 – + 14 –CH2Br 30 ± 12 – – 15a –CH2N(CH3)2 19 ± 9 – – 15b –CH2N(Et)Pr 14 ± 12 – – 15c –CH2N(CH2)4 10 ± 12 – + 15d –CH2NH2 44 ± 8 +++ +++ a Inhibition of the transformation of [14C]-E1 (60 nM) into [14C]-E2 by 17β-HSD1 in T-47D intact cells. Inhibition values are represented as means (±SD) of at least two independent experiments performed in triplicate. bEffect of inhibitors on the growth of estrogen-starved estrogen-sensitive MCF-7 cells after 7 days of treatment. Legend for estrogenicity: “–” = no (0–5%), “+” = weak (5–15%), “++” = medium (15–30%), “+++” = strong (>30%) vs control (basal cell proliferation fixed at 0%).

After investigating the inhibitory potency on 17β-HSD1, we evaluated the proliferative (estrogenic) activity of synthesized compounds on MCF-7 estrogen-sensitive cell line. In fact, for a potential use in breast cancer, an enzyme inhibitor should be devoid of estrogenic activity. Despite the promising 17β-HSD1 inhibitory results obtained with boronic acid 7, aniline 8, methylalcohol 13, and methylamine 15d, these compounds were all estrogenic at 1 μM and, consequently, they were disqualified for their use in the context of breast cancer treatment. However, we were pleased to see that bromomethyl derivative 14 did not stimulate the cell proliferation at 0.1 and 1 μM. Other compounds, such as 12b, 12c, 15a, and 15b, were nonestrogenic on MCF- 7 cells, but they inhibit weakly 17β-HSD1. Finally, compounds 9, 10, 11, and 12a were found to be estrogenic and weak inhibitors. Thus, on the basis of the 17β-HSD1 inhibition and estrogenicity results of series I compounds, we planned the synthesis of new compounds (series II) in order to investigate the effect of an

additional carbon spacer to move away the functionality from A-ring on both enzyme inhibition and estrogenic activity.

Series II Compounds (Inhibition of 17β-HSD1 and Estrogenicity)

To extend our SAR studies, we added a methylene (CH2) spacer between some of the most active functionalities of the first series of compounds and the steroid A-ring. The tolerance of this enzyme to different types of hydrophobic substituents (Ph, H, Cl, Br, and I) was also investigated. The results obtained in this second series were very interesting (Table 2) because the bromoethyl derivative 23b and the iodoethyl derivative 23c gave good 17β-HSD1 inhibition values (51 and 54% at 0.1 μM, respectively). An improvement in inhibitory activity was thus observed with 23b (2 × CH2) over 14 (1 × CH2) when tested at 0.1 μM (51 and 30%, respectively). However, replacement of the bromide of 23b by a lipophilic and nonelectronegative substituent like phenyl or methyl produced very weak inhibitions at 0.1 μM (17 and 16% for 18 and 20, respectively). As seen with tertiary amide 25, the size of the phenyl ring substituent is also detrimental to inhibition. In the same manner, the chain extension was ineffective for carboxylic acid 11 and alcohol 13 because the homologue compounds 24 and 22 had a weak inhibitory effect on 17β-HSD1.

43 Table 2. Inhibition of 17β-HSD1 (Serie II)

compd no. R inhibitiona % estrogenicityb at 0.1 μM at 0.1 μM at 1.0 μM 11 (series I) –COOH 3 ± 3 + +++ 13 (series I) –CH2OH 37 ± 14 – + 14 (series I) –CH2Br 30 ± 12 – – c c 18 –CH2CH2Ph 17 ± 5 – – 20 –CH2CH3 6 ± 8 – – 22 –CH2CH2OH 18 ± 1 + + c c 23a –CH2CH2Cl 12 ± 7 – – 23b –CH2CH2Br 51 ± 2 – – c c 23c –CH2CH2I 54 ± 2 – – 24 –CH2COOH 7 ± 7 – + 25 –CH2CONHCH3 11 ± 8 – – aInhibition of the transformation of [14C]-E1 (60 nM) into [14C]-E2 by 17β-HSD1 in T-47D intact cells. Inhibition values are represented as means (±SD) of at least two independent experiments performed in triplicate. bEffect of inhibitors on the growth of estrogen-starved estrogen-sensitive MCF-7 cells after 7 days of treatment. cEffect of inhibitors on the growth of estrogen-starved estrogen-sensitive T-47D cells after 7 days of treatment. Legend for estrogenicity: “–” = no (0–5%), “+” = weak (5–15%), “++” = medium (15–30%), “+++” = strong (>30%) vs control (basal cell proliferation fixed at 0%).

Despite the results pointing toward the importance of electronegativity of the substituent, this factor does not seem to be the most important as observed with the poor inhibition value (12% at 0.1 μM) obtained with chloroethyl derivative 23a, the more electronegative substituent of the series. The leaving group capability could represent a more coherent explanation for the stronger inhibition observed with bromide and iodide atoms compared to lower leaving capability of the chloride atom. This hypothesis suggests an irreversible inhibition of the enzyme by a nucleophilic attack of the bromide or iodide atom by an amino acid residue. In fact, it was previously reported that a bromopropyl chain added at position 16 of E2 provided an irreversible inhibition of 17β-HSD1.(49, 50)

The estrogenicity of series II compounds was assessed by evaluating the proliferation of estrogen-sensitive cells. The presence of OH group (compound 22) is sufficient to induce a cell proliferation that is higher than the level observed with alcohol 13. A weak estrogenicity was also obtained for the carboxylic acid derivative 24. None of the other compounds stimulated the proliferation of cells at 0.1 and 1 μM, including the two most active inhibitors of the series II (compounds 23b and 23c). These two compounds have a hydrophobic group at position

3 of steroid core that, contrary to 22 and 24, cannot generate a hydrogen bonding with ER. In fact, the C3-OH of E2 is well-known to be involved in binding with ER.(51, 52)

Series III Compounds (Inhibition of 17β-HSD1 and Estrogenicity)

A short series of compounds was designed to study the tolerance of the enzyme for a longer side chain at position 3 (Table 3). We first added a CH2 group to the bromoethyl side chain of compound 23b, the most potent inhibitor and nonestrogenic compound of the first two series, to obtain the bromopropyl derivative 28. We also tested the intermediate alcohol 27 to validate the tendency observed in series I and II with alcohol derivatives 13 and 22. Finally, we tested compound 30, a derivative that integrated an oxygen atom in the bromopropyl side chain to reach additional interactions with the enzyme. The inhibition assay revealed that the longer (CH2)3Br side chain did not increase the inhibition compared to the shorter (CH2)2Br side chain (41 and 51% at 0.1 μM for 28 and 23b, respectively). For the different alcohol derivatives synthesized in the three series, the proximity of the hydroxyl group to the steroid A-ring was an important factor and inhibition values decreased with the spacer length (72, 37, 18, and 20% of inhibition at 0.1 μM for n = 0–3 or compounds 1, 13, 22, and 27, respectively). As a second relevant result of series III, the presence of an oxygen atom in the side chain of compound 28 (compound 30) did not increase the inhibition (41 and 37% for 28 and 30, respectively). Regarding the ability of series III compounds to stimulate estrogen-sensitive cell proliferation, the alcohol derivative 27 was found to be estrogenic but interestingly not the bromide derivatives 28 and 30. This result once again illustrates the importance of a hydrogen bonding between the C-3 OH and an amino acid known to be involved in the binding of E2 to ER.(51, 52)

45 Table 3. Inhibition of 17β-HSD1 (Serie III)

compd no. R X inhibitiona estrogenicityb % at 0.1 μM at 0.1 μM at 1.0 μM

23b (series II) –Br CH2 51 ± 2 – – 27 –CH2OH CH2 20 ± 5 – + 28 –CH2Br CH2 41 ± 5 – – 30 –CH2Br O 37 ± 3 – – a Inhibition of the transformation of [14C]-E1 (60 nM) into [14C]-E2 by 17β-HSD1 in T-47D intact cells. Inhibition values are represented as means (±SD) of at least two independent experiments performed in triplicate. bEffect of inhibitors on the growth of estrogen-starved estrogen-sensitive T-47D cells after 7 days of treatment at different concentrations. Legend for estrogenicity: “–” = no (0–5%), “+” = weak (5–15%), “++” = medium (15– 30%), “+++” = strong (>30%) vs control (basal cell proliferation fixed at 0%).

Determination of IC50 Values (17β-HSD1 Inhibition)

The IC50 values of bromide derivatives 14, 23b, 28, and 30, the most promising compounds for inhibitory as well as nonestrogenic activity, were determined to compare their potency to inhibit 17β-HSD1 in T-47D cells (Figure

3). The bromomethyl derivative 14 with a shorter spacer (1 × CH2) was found to be six times less potent than the bromoethyl derivative 23b (2 × CH2) (IC50 = 430 nM and 68 nM, respectively), and the bromopropyl derivative

28 with a longer spacer (3 × CH2) was found to be half as potent (IC50 = 153 nM) than 23b. The presence of an oxygen atom instead of a CH2 group in the side chain of 30 did not increase the inhibition value of 28 (IC50 = 172 and 153 nM, respectively). Thus, the bromoethyl derivative 23b was the most potent 17β-HSD1 inhibitor of the bromide series with IC50 value of 68 nM in this assay (Figure 3) and 97 nM in another assay both performed in T-47D cells (mean value = 83 nM). On the basis of these screening results, 23b was selected for subsequent biological in vivo assays.(53)

Figure 3. 17β-HSD1 inhibitory potency of compounds 1, 14, 23b, 28, and 30 in T-47D intact cells. Breast cancer cells expressing 17β-HSD1 were incubated with various concentrations of inhibitor for 24 h in presence of labeled 14 [ C]-E1 (60 nM). IC50 represents the concentration that inhibited the transformation of E1 into E2 by 50%. Results are representative of two experiments performed in triplicate except for compound 14 (tested one time).

Selectivity of 23b over Other Enzymes

Inhibition of 17β-HSD2, 17β-HSD7, and 17β-HSD12

The selectivity of the17β-HSD1 inhibitor 23b over 17β-HSD2 was evaluated to ensure that it does not deactivate the oxidation of E2 into E1 by inhibiting 17β-HSD2 (Table 4). The assay was performed with stably transfected 17β-HSD2 in intact HEK-293 cells using [14C]-E2 as substrate.(54) As a result, compound 23b did show a very low inhibition of 17β-HSD2 (only 7% at the higher concentration of 10 μM) for the conversion of E2 to E1. We were also interested in the selectivity of compound 23b for other 17β-HSDs, such as type 7 and type 12,(55, 56) which have been reported to convert E1 to E2 in breast cancer cells.(30) Similarly to 17β-HSD2, the assays

47 were done with stably transfected 17β-HSD7 or 17β-HSD12 in HEK-293 cells but using [14C]-E1 as substrate instead of [14C]-E2. The compound 23b was found to be highly selective over 17β-HSD7 (9% of inhibition at 10 μM) and less selective over 17β-HSD12 (34% of inhibition at 10 μM).

Table 4. Selectivity of Compounds 1 and 23b on Four Enzymes (17β-HSD2, 17β-HSD7, 17β-HSD12, and CYP3A4)

compd no. inhibition of inhibition of inhibition of inhibition of inhibition of 17β-HSD1b 17β-HSD2% 17β-HSD7% 17β-HSD12% CYP3A4 IC50 (μM) at 10 μM at 10 μM at 10 μM IC50 (μM) 1 0.036 ± 0.012 0 ± 12 0 ± 1 6 ± 7 1.5 ± 0.4 23b 0.083 ± 0.021 7 ± 16 9 ± 5 34 ± 10 3.9 ± 0.5 INH-2c 48 ± 13 INH-7d 81 ± 3 INH-12e 39 ± 3 ketoconazole 0.024 ± 0.004

aInhibition values are represented as means (±SD) of at least two experiments performed in triplicate. b17β-HSD1 in T-47D cells. cPotent inhibitor of 17β-HSD2 (see compound 1 in ref 54). dPotent inhibitor of 17β-HSD7 (see compound 81 in ref 55). eWeak inhibitor of 17β-HSD12 (see compound 55 in ref 56).

Inhibition of CYP3A4

We were concerned about the selectivity of action of compound 23b toward CYP3A4 (Table 4), one of the most important liver enzymes involved in drug metabolism. We used the P450 Inhibition Kit CYP3A4/DBF as suggested by the manufacturer with the exception that 23b was dissolved in a mixture of DMSO/ACN (5:95) instead of only acetonitrile. Ketoconazole was used as a positive reference inhibitor giving an IC50 of 0.024 μM, which is in agreement with reported values.(57) When tested for CYP3A4 inhibition, the bromoethyl derivative

23b and the phenolic derivative 1, we obtained IC50 values of 3.9 ± 0.5 μM for 23b and 1.5 ± 0.4 μM for 1, thus indicating a very low risk of drug–drug interactions for both compounds but especially for 23b.(57)

Competitive Nature and Irreversibility of Inhibitor 23b

The leaving group ability of halogen atoms (Cl, Br, and I) and the potency of corresponding compounds to inhibit the transformation of [14C]-E1 to [14C]-E2 by 17β-HSD1 suggests the formation of a covalent bond. To verify this hypothesis and to provide some information about the mechanism of action, we performed an inactivation assay with a representative inhibitor and purified 17β-HSD1. Figure 4 shows the inhibition curves for the bromide derivative 23b (0, 0.1, and 0.5 μM) according to preincubation time. As expected, a slight decrease of the enzyme activity was observed in the absence of inhibitor, but a progressive inhibition of 17β-HSD1 was observed with the time at both concentrations for 23b. The rate of this time-dependent inhibition was higher at 0.5 μM than at 0.1 μM of inhibitor. The very fact that the preincubation time affects the enzyme activity indicates an irreversible effect and strongly suggests the inactivation of the enzyme caused by a covalent binding of the inhibitor considering its nature. As previously reported,(58) a reversible inhibitor would generate an inhibition curve closely similar to that of the control because the same enzymatic activity would be recovered after washing out the inhibitor independently of the preincubation time. In the inactivation assay discussed above, we also addressed the competitiveness of inhibitor 23b. Moreover, the rate of 17β-HSD1 inactivation ([14C]-E1 to [14C]- E2) is slowed by the presence of unlabeled E1 (0.5 μM), which thus appear to compete with 23b (0.5 μM) for the substrate binding site of the enzyme.

Figure 4. Time-dependent inactivation of 17β-HSD1 by compound 23b. The transformation of [14C]-E1 to [14C]- E2 by purified enzyme was assessed after preincubation with compound 23b (0, 100, or 500 nM), with or without the natural substrate E1 (500 nM), and expressed as the percentage of initial enzyme activity. See the Experimental Section for the conditions of the enzymatic assay.

49 Molecular Modeling

Since we demonstrated the competitive nature of the inhibitor 23b for the substrate binding site (Figure 4), we were interested in studying the potential 17β-HSD1/23b interactions by performing some docking experiments. For discussion purposes, we also studied the competitive reversible inhibitor 1, an E2 derivative that was crystallized with 17β-HSD1.(33) The hydroxyl group at position C3 of compound 1 produces H-bonds with side chains of residues Glu-282 and His-221 (Figure 5). Compound 23b differs from compound 1 as it does not have an H-bond donor group at position C3. Its inhibition potency was however shown to be very good, suggesting a different mode of action.

Figure 5. Representation of compound 1 in the active site of 17β-HSD1. Coordinates are from PDB 3HB5.(33) Compound 1 is represented by the thick purple sticks, NADP+ by the small cyan sticks, and the amino acids are represented by the small green sticks. H-bonds between compound 1 hydroxyl at position C3 and Glu-282/His- 221 are highlighted.

The first docking attempt for compound 23b led to a root-mean-square deviation (RMSD) value of 1.49 Å when compared to the compound 1 position obtained from the crystallographic structure. As shown in Figure 6A, the Glu-282 side chain is oriented toward the binding site to make an H-bond with compound 1, leaving no space for the bromoethyl side chain of compound 23b. However, Glu-282 is a solvent exposed flexible residue as indicated by its high B-factor value of 45. To account for this flexibility in the docking calculations, we generated a binding site conformation for which the Glu-282 side chain is exposed to the solvent (see Experimental

Section). As shown in Figure 6B, this latter conformation accommodates the docking of compound 23b with a much better RMSD of 0.76 Å. Nonetheless, the distance between His-221 side chain and the bromide of compound 23b is 3.8 Å, too long to account for an irreversible binding as hypothesized earlier. To get around the force field limitations that do not allow for covalent reactions between the bromoethyl moiety and His-221 side chain, this latter residue was mutated into an Ala, which has a smaller side chain. The docking results using this binding site conformation are presented in Figure 6C,D. The RMSD of the resulting compound 23b structure is 0.59 Å, and the distance between the CH2 of the bromoethyl side chain and the NH of reconstituted His-221 side chain is now 1.7 Å, indicating the possibility of a covalent reaction (Figure 6D). This doubly mutated docking was made to demonstrate that without the conformational limitations of His-221 and Glu-282, the core with the bromoethyl moiety is very well placed in the enzyme pocket to generate a covalent bond.

Figure 6. Results from the docking of compound 23b at the binding site of 17β-HSD1: (A) using the crystallographic binding site conformation, (B) with Glu-282 side chain solvent-oriented, (C) with Glu-282 side chain solvent-oriented and His-221 mutated into Ala, and (D) superposition of receptor from (B) with the docked compound 23b from (C). The structure of compound 1 is shown for purposes of comparison with the docked compound 23b and is not included in the docking calculations. Atoms from PDB 3HB5 are represented in green (protein in cartoon and compound 1 in sticks), the modified residues are represented by the purple sticks, and docked compound 23b by the orange (A), white (B), and cyan (C,D) sticks.

On the basis of the results of kinetics and molecular modeling with bromide derivative 23b, and the similar reactivity of both bromide and iodide, we are confident that the iodide of 23c will be released in the enzyme active site after the formation of a covalent bond with an amino acid. Therefore, it could be caught by this amino acid to form the corresponding iodoimidium salt(59) and to be retained within the active site of 17β-HSD1 (Figure 7). In the event that the iodide would be expelled from the enzyme active site and then to the cell, it could be caught by the NIS symporter,(60) a specialized protein known to fix iodide and to be selectively expressed in thyroid and breast cancer cells.(61) Interestingly, the use of an irreversible inhibitor of 17β-HSD1 could open the door to molecular imaging and radiotherapeutics by adding the appropriate iodo-radioisotope (123I and 131I,

51 respectively) on parent inhibitor 23c. Advantageously, the iodo-radioisotope derivatives could be readily accessible from a simple substitution reaction between compound 23b and Na*I.

Figure 7. Proposed irreversible mechanism of compound 23b (X = Br) and 23c (X = I) on 17β-HSD1.

Conclusion

Three successive series (I–III) of 3-substituted-16β-(m-carbamoylbenzyl)-E2 derivatives has provided important SAR data regarding enzyme tolerance for substituents of different natures (hydrophilic, hydrophobic, H-bond donor, H-bond acceptor, basic, acidic, etc.) and different chain lengths (0–3 atom spacer). In the first series of synthesized compounds, different types of substituents like alcohol and amine gave acceptable inhibition levels but their significant estrogenic activity disqualified them for treatment of estrogen-related diseases. The most promising substituent of the first series was undoubtedly the bromomethyl derivative 14, which showed an acceptable level of inhibition and a nonestrogenic profile. The second series of compounds thus focused on the carbon chain extension of bromomethyl derivative 14 as well as other promising compounds of the first series. These new series of compounds converged toward the identification of 3-(2-bromoethyl)-16β-(m- carbamoylbenzyl)-estra-1(10),2,4-trien-17β-ol (23b) as potent inhibitor of 17β-HSD1 in T-47D cells (IC50 = 68 and 97 nM) without any estrogenic activity detected on estrogen-sensitive cells. This bromo derivative was found to be a competitive and irreversible inhibitor of 17β-HSD1. Molecular modeling with 23b docked in 17β-HSD1 showed the potential key interactions with His-221, highlighting the possibility of a nucleophilic attack of the His- 221 on the bromoethyl group. Otherwise, this novel inhibitor represents an important evolution relative to lead compound 1, which was a potent inhibitor of 17β-HSD1 but with an unwanted estrogenic activity. Also, compound 23b had a selectivity of action over 17β-HSD2, 17β-HSD7, 17β-HSD12, and liver enzyme CYP3A4, thus showing a promising profile toward in vivo assays. Importantly, the evaluation of bromo derivative 23b in vivo reveals the efficiency of this inhibitor to completely block the tumor growth of estrone stimulated cancer cells (T-47D) expressing the 17β-HSD1 enzyme.(53) Finally, as interesting perspective, the iodide derivative 23c could provide an opportunity for molecular imaging of tissues expressing 17β-HSD1 as well as selective

radiotherapeutic treatment. These new 17β-HSD1 inhibitors 23b and 23c, developed through a SAR study, thus represent promising candidates toward clinical studies for the treatment and diagnosis of estrogen-dependent diseases like breast cancer and endometriosis.

Experimental Section

Chemical reagents were purchased from Sigma-Aldrich Canada Ltd. (Oakville, ON, Canada). The usual solvents were obtained from Fisher Scientific (Montréal, QC, Canada) and were used as received. Anhydrous dichloromethane (DCM), dimethylformamide (DMF), and tetrahydrofuran (THF) were obtained from Sigma- Aldrich. Thin-layer chromatography (TLC) and flash-column chromatography were performed on 0.20- mm silica gel 60 F254 plates (E. Merck; Darmstadt, Germany) and with 230–400 mesh ASTM silica gel 60 (Silicyle, Québec, QC, Canada), respectively. Infrared spectra (IR) were recorded on a Horizon MB 3000 ABB FTIR spectrometer (Québec, QC, Canada), and only the significant bands reported in cm–1. Samples were prepared as KBr pellet. Nuclear magnetic resonance (NMR) spectra were recorded at 400 MHz for 1H and 100.6 MHz for 13C on a Bruker Avance 400 digital spectrometer (Billerica, MA, USA). The chemical shifts (δ) were expressed in ppm and referenced to chloroform (7.26 and 77.0 ppm), acetone (2.06 and 29.1 ppm), or methanol (3.33 and 49.0 ppm) for 1H and 13C NMR, respectively. The numbering reported in Figure 8 was used for the assignment of 1H and 13C NMR signals. Low-resolution mass spectra (LRMS) were recorded on a PE Sciex API-150ex apparatus (Foster City, CA, USA) equipped with a turbo ion-spray source. The HPLC purity of the final compounds to be tested was determined with a Shimadzu apparatus using a Shimadzu SPD-M20A photodiode array detector, an Altima HPC18 reversed-phase column (250 mm × 4.6 mm, 5 μM), and a solvent gradient of

MeOH:H2O. The wavelength of the UV detector was selected between 190 and 205 nm. All final compounds shown a purity ≥95% (95.0–99.9%) except for compounds 13, 15b, 15c, 18, 20, 24, 25, and 27 (90.2–93.8%). The IUPAC nomenclature was used in the experimental part and the names of steroid derivatives were generated using ACD/Laboratories (Chemist’version) software (Toronto, ON, Canada).

Figure 8. Carbon numbering used for the assignment of representative 1H NMR signals.

Synthesis of 7, 8, 9, 10, and 11

General Procedure for the Introduction of 16β-Carbamoyl-m-benzamide Side Chain

(a) Aldolization reaction: To a solution of the appropriate steroidal ketone 2, 3, 4, 5, or 6 (0.3 mmol) in EtOH (0.04 M) was added 3-formyl-benzamide (0.6 mmol) and an aqueous KOH (10%) solution (15% v/v). The solution was heated at reflux for 30 min. The resulting solution was diluted with water, neutralized with aqueous HCl

(10%), and extracted with EtOAc. The organic layers were combined, washed with brine, dried with MgSO4, and evaporated under reduced pressure. The crude compound was purified by flash chromatography to give the corresponding enone. (b) Reduction of enone: To a solution of enone (0.25 mmol) in MeOH (0.03 M) was added

NaBH4 (0.38 mmol), and the mixture was stirred at room temperature. After 1 h, the resulting solution was concentrated under vacuo, diluted with EtOAc, washed with water and brine, and dried with MgSO4. After the organic phase was evaporated under reduced pressure, the crude compound was purified by flash chromatography to give the corresponding allylic alcohol. (c) Catalytic hydrogenation of allylic alcohol: To a solution of allylic alcohol (0.2 mmol) in EtOH (0.03 M) under argon atmosphere at room temperature was added palladium on charcoal (10% w/w). The reaction vessel was flushed three times with hydrogen and stirred for 24 h at room temperature. The resulting solution was filtered on Celite and then evaporated under reduced pressure. The crude compound was purified by flash chromatography using DCM/MeOH (95:5) as eluent to give the desired 16β-carbamoyl-benzamide derivative 7, 8, 9, 10, or 11.

[(16β,17β)-16-(3-Carbamoylbenzyl)-17-hydroxyestra-1(10),2,4-trien-3-yl]boronic Acid (7)

1 Yield: 13 mg, 8% (for 3 steps). IR (KBr): 1659 (C═O, amide), 3380 (OH and NH2). H NMR (CD3OD): 0.91 (s,

18-CH3), 1.13 (m, 14α-CH and 15β-CH), 1.20–1.55 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.69 (t, J = 6.0 Hz, 15α-CH), 1.84 (broad d, J = 11.8 Hz, 7β-CH), 2.05 (d, J = 12.2 Hz, 12β-CH), 2.26 (m, 9α-CH), 2.35 (m, 11α-

CH), 2.47 (q, J = 12.5 Hz, 1H of 1′-CH2), 2.52 (m, 16α-CH), 2.83 (m, 6-CH2), 3.17 (d, J = 12.1 Hz, 1H of 1′-CH2),

3.84 (d, J = 9.4 Hz, 17α-CH), 7.22–7.50 (m, 4-CH, 2-CH, 1-CH, 5″-CH, and 6″-CH), 7.70 (dd, J1 = 1.2 Hz, J2 =

13 7.4 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.2 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15), 38.8 (C1′), 39.0 (C12), 39.6 (C8), 43.4 (C16), 45.4 (C13), 46.0 (C9), 50.1 (C14), 83.0 (C17), 125.4 (C1), 126.0 (C4″), 129.1 (C2″), 129.4 (C5″), 131.9 (C2), 133.5 (C6″), 134.8 (C3 and C3″), 135.4 (C4), 136.5 (C5), 136.7

+ (C10), 144.4 (C1″), 172.7 (CONH2). LRMS for C27H37BNO5 [M + CH3OH + H] = 466.3. HPLC purity of 97.9% (retention time = 10.6 min).

3-[(16β,17β)-3-Amino-17-hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide (8)

1 Yield: 5 mg, 8% yield (for 3 steps). IR (KBr): 1663 (C═O, amide), 3364 (OH and NH2). H NMR (CD3OD): 0.91

(s, 18-CH3), 1.14 (m, 14α-CH and 15β-CH), 1.20–1.50 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.66 (t, J = 7.0 Hz, 15α-CH), 1.84 (m, J = 11.8 Hz, 7β-CH), 2.02 (d, J = 12.4 Hz, 12β-CH), 2.15 (m, 9α-CH), 2.30 (m, 11α-

CH), 2.47 (q, J = 12.3, 1H of 1′-CH2), 2.50 (m, 16α-CH), 2.72 (m, 6-CH2), 3.17 (dd, J1 = 2.2 Hz, J2 = 12.0 Hz, 1H of 1′-CH2), 3.83 (d, J = 9.4 Hz, 17α-CH), 6.46 (d, J = 2.1 Hz, 4-CH), 6.54 (dd, J1 = 2.3 Hz, J2 = 8.2 Hz, 2-CH), 7.04 (d, J = 8.3 Hz, 1-CH), 7.40 (m, 5″-CH and 6″-CH), 7.70 (d, J = 7.4 Hz, 4″-CH), 7.75 (s, 2″-CH). 13C NMR

(CD3OD): 13.3 (C18), 27.5 (C11), 28.7 (C7), 30.7 (C6), 33.0 (C15), 38.9 (C12), 39.0 (C1′), 40.0 (C8), 43.4 (C16), 45.1 (C13), 45.5 (C9), 49.9 (C14), 83.1 (C17), 114.9 (C2), 117.2 (C4), 126.0 (C4″), 126.8 (C1), 129.2 (C2″),

129.5 (C5″), 132.0 (C10), 133.5 (C6″), 134.8 (C3”), 138.2 (C5), 144.4 (C1″), 145.5 (C3), 172.7 (CONH2). LRMS

+ for C27H37N2O3 [M + CH3OH + H] = 437.3. HPLC purity of 96.9% (retention time = 5.3 min).

3-{[(16β,17β)-3-Fluoro-17-hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide (9)

1 Yield: 59 mg, 26% (3 steps). IR (KBr): 1639 (C═O, amide), 3356 (OH and NH2). H NMR (CD3OD): 0.91 (s, 18-

CH3), 1.15 (m, 14α-CH and 15β-CH), 1.22–1.56 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.67 (t, J = 7.0 Hz, 15α-CH), 1.82 (m, 7β-CH), 2.04 (d, J = 12.2 Hz, 12β-CH), 2.21 (m, 9α-CH), 2.35 (m, 11α-CH), 2.47 (q, J = 12.5,

1H of 1′-CH2), 2.52 (m, broad, 16α-CH), 2.81 (m, 6-CH2), 3.17 (dd, J1 = 3.2 Hz, J2 = 12.6 Hz, 1H of 1′-CH2), 3.84

(d, J = 9.4 Hz, 17α-CH), 6.75 (dd, J1 = 2.6. Hz, J2 = 9.8 Hz, 4-CH), 6.81 (dt, J1 = 2.8 Hz, J2 = 8.6 Hz, 2-CH), 7.27

(dd, J1 = 6.0 Hz, J2 = 8.4 Hz, 1-CH), 7.40 (5″-CH and 6″-CH), 7.71 (td, J1 = 1.5 Hz, J2 = 7.5 Hz, 4″-CH), 7.76 (s,

13 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.4 (C11), 28.3 (C7), 30.5 (C6), 33.0 (C15), 38.8 (C1′ and C12), 39.4

(C8), 43.3 (C16), 45.2 (C9), 45.3 (C13), 49.8 (C14), 82.9 (C17), 113.1 (d, JCC-F = 21.0 Hz, C2), 115.7 (d, JCC-F =

20.2 Hz, C4), 125.9 (C4″), 127.9 (d, JCC-F = 8.0 Hz, C1), 129.1 (C2″), 129.4 (C5″), 133.5 (C6″), 134.7 (C3″), 137.3

55 (C10), 140.1 (d, JC–F = 6.8 Hz, C5), 144.3 (C1″), 162.2 (d, JC–F = 242.0 Hz, C3), 172.6. LRMS for C27H35FNO3 [M

+ + CH3OH + H] = 440.3. HPLC purity of 98.1% (retention time = 17.5 min).

(16β,17β)-16-(3-Carbamoylbenzyl)-17-hydroxyestra-1(10),2,4-triene-3-carboxamide (10)

1 Yield: 21 mg, 6% (for 3 steps). IR (KBr): 1647 (C═O, amide), 3213, 3329, 3483, and 3533 (OH and NH2). H

NMR (CD3OD): 0.93 (s, 18-CH3), 1.18 (m, 14α-CH and 15β-CH), 1.27–1.60 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.70 (m, 15α-CH), 1.88 (m, 7β-CH), 2.07 (d, J = 12.6 Hz, 12β-CH), 2.32 (m, 9α-CH), 2.40 (m, 11α-

CH), 2.48 (q, J = 12.4, 1H of 1′-CH2), 2.52 (m, broad, 16α-CH), 2.89 (m, 6-CH2), 3.17 (m, 1H of 1′-CH2), 3.86 (d,

J = 9.4 Hz, 17α-CH), 7.35–7.45 (m, 1-CH, 5″-CH and 6″-CH), 7.58 (s, 4-CH), 7.62 (dd, J1 = 1.8 Hz, J2 = 8.2 Hz,

13 2-CH), 7.70 (dd, J1 = 1.4 Hz, J2 = 7.0 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.2 (C11), 28.4 (C7), 30.5 (C6), 33.1 (C15), 38.9 (C12 and C1″), 39.3 (C8), 43.4 (C16), 45.4 (C13), 45.9 (C9), 50.0 (C14), 82.9 (C17), 126.0 (C1 and C2), 126.5 (C4″), 129.2 (C2″), 129.3 (C5″), 129.5 (C4), 132.0 (C3), 133.6 (C6″), 134.8

+ (C3″), 138.2 (C5), 144.4 (C1″), 146.0 (C10), 172.6 (2 × CONH2). LRMS for C28H37N2O4 [M + CH3OH + H] = 465.3. HPLC purity of 96.4% (retention time = 6.4 min).

(16β,17β)-16-(3-Carbamoylbenzyl)-17-hydroxyestra-1(10),2,4-triene-3-carboxylic Acid (11)

Yield: 320 mg, 49% (for 3 steps). IR (KBr): 1666 (C═O, amide, and acid), 3198, 3283, 3383, and 3553 (OH and

1 NH2). H NMR (CD3OD): 0.91 (s, 18-CH3), 1.17 (m, 14α-CH and 15β-CH), 1.26–1.58 (m, 7α-CH, 12α-CH, 8β- CH, and 11β-CH), 1.70 (t, J = 7.0 Hz, 15α-CH), 1.87 (m, 7β-CH), 2.06 (d, J = 12.4 Hz, 12β-CH), 2.31 (m, 9α-

CH), 2.42 (m, 11α-CH), 2.48 (q, J = 12.6, 1H of 1′-CH2), 2.53 (m, 16α-CH), 2.88 (m, 6-CH2), 3.18 (dd, J1 = 3.2

Hz, J2 = 12.6 Hz, 1H of 1′-CH2), 3.86 (d, J = 9.4 Hz, 17α-CH), 7.36–7.45 (m, 1-CH, 5″-CH and 6″-CH), 7.69 (d,

13 J = 7.3 Hz, 2-CH), 7.71(s, 4-CH), 7.75 (d, J = 7.4 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.2 (C11), 28.3 (C7), 30.4 (C6), 33.0 (C15), 38.8 (C12), 38.9 (C1′), 39.3 (C8), 43.3 (C16), 45.4 (C13), 46.1 (C9), 50.0 (C14), 82.9 (C17), 126.0 (C1), 126.4 (C4”), 128.0 (C2), 129.1 (C2″), 129.4 (C5″), 131.3 (C4), 133.5

(C6″), 134.8 (C3″), 138.0 (C3 and C5), 144.3 (C1″), 146.8 (C10), 171.0 (COOH), 172.7 (CONH2). LRMS for

+ C28H36NO5 [M + CH3OH + H] = 466.3. HPLC purity of 95.1% (retention time = 13.8 min).

Synthesis of 12a–c

General Procedure for N-Acylation of 11

To a solution of acid 11 (0.12 mmol) in DMF (3 mL) was added BOP (0.14 mmol), the appropriate amine (0.36 mmol), and DIPEA (28 μL, 0.17 mmol). The solution was stirred at room temperature for 2 h. The mixture was poured into water, extracted with EtOAc, washed with water and brine, dried over MgSO4, and evaporated under reduced pressure. The crude N-acylated derivative 12a, 12b, or 12c was purified by flash chromatography (typically DCM/MeOH, 95:5 to 9:1).

(16β,17β)-16-(3-Carbamoylbenzyl)-17-hydroxy-N,N-dimethylestra-1(10),2,4-triene-3-carboxamide (12a)

1 Yield: 7 mg, 13%. IR (KBr): 1666 (C═O, amide), 3379 and 3456 (OH and NH2). H NMR (CD3OD): 0.92 (s, 18-

CH3), 1.17 (m, 14α-CH and 15β-CH), 1.22–1.60 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.70 (t, J = 7.0 Hz,

15α-CH), 1.87 (m, 7β-CH), 2.08 (dt, J1 = 2.9 Hz, J2 = 12.3 Hz, 12β-CH), 2.30 (m, 9α-CH), 2.35–2.70 (m, 11α-

CH, 1H of 1′-CH2 and 16α-CH), 2.86 (m, 6-CH2), 3.02 (s, NCH3), 3.10 (s, NCH3), 3.17 (dd, J1 = 3.2 Hz, J2 = 12.5

Hz, 1H of 1′-CH2), 3.87 (d, J = 9.4 Hz, 17α-CH), 7.11 (s, 4-CH), 7.17 (d, J = 8.0 Hz), 7.38–7.44 (m, 1-CH, 5″-CH

13 and 6″-CH), 7.70 (d, J = 7.5 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.2 (C11), 28.3 (C7),

30.4 (C6), 33.0 (C15), 38.8 (C12), 38.9 (C1′), 39.4 (C8), 39.7 and 40.1 (N(CH3)2), 43.3 (C16), 45.4 (C13), 45.8 (C9), 50.0 (C14), 83.0 (C17), 125.3 (C1), 126.0 (C2), 126.5 (C4″), 128.5 (C4), 129.2 (C2″), 129.4 (C5″), 133.5

(C6″), 134.3 (C3), 134.8 (C3″), 138.3 (C5), 143.6 (C10), 144.3 (C1″), 172.6 (CONH2), 174.2 (CON(CH3)2). LRMS

+ for C29H37N2O3 [M + H] = 461.3. HPLC purity of 97.6% (retention time = 10.2 min).

(16β,17β)-16-(3-Carbamoylbenzyl)-N-ethyl-17-hydroxy-N-propylestra-1(10),2,4-triene-3-carboxamide (12b)

1 Yield: 7 mg, 30%. IR (KBr): 1666 (C═O, amide), 3209 and 3406 (OH and NH2). H NMR (CD3OD): 0.78 and

1.00 (2t, J = 7.1 Hz, CH3CH2CH2N), 0.92 (s, 18-CH3), 1.12 and 1.26 (2t, J = 6.4 Hz, CH3CH2N), 1.17 (m, 14α-

CH and 15β-CH), 1.20–1.75 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH, 15α-CH and CH3CH2CH2N), 1.85 (m, 7β-

CH), 2.05 (d, J = 12.2 Hz, 12β-CH), 2.30 (m, 9α-CH), 2.48 (m, 11α-CH, 1H of 1′-CH2 and 16α-CH), 2.85 (m, 6-

CH2), 3.10–3.60 (m, 1H of 1′-CH2, CH3CH2CH2N and CH3CH2N), 3.85 (d, J = 9.4 Hz, 17α-CH), 7.11 (s, 4-CH), 7.04 (d, J = 8.0 Hz, 2-CH), 7.35–7.44 (m, 1-CH, 5″-CH, and 6″-CH), 7.70 (d, J = 7.5 Hz, 4″-CH), 7.76 (s, 2″-CH).

13 C NMR (CD3OD): 11.7 (CH3CH2N), 13.3 (C18), 14.3 (CH3CH2CH2N), 23.0 (CH3CH2CH2N), 27.2 (C11), 28.3

57 (C7), 30.4 (C6), 33.0 (C15), 38.8 (C12), 38.9 (C1′), 39.4 (C8), 43.3 (C16), 45.4 (C13), 45.8 (C9), 47.5 (CH3CH2N),

50.0 (C14), 51.9 (CH3CH2CH2N), 83.0 (C17), 124.6 (C2), 126.1 (C1), 126.5 (C4”), 127.8 (C4), 129.2 (C2″), 129.4

(C5″), 133.5 (C6″), 134.3 (C3), 134.8 (C3″), 138.3 (C5), 143.5 (C10), 144.3 (C1″), 172.6 (CONH2), 174.2 (CON).

+ LRMS for C32H43N2O3 [M + H] = 503.3. HPLC purity of 98.8% (retention time = 15.4 min).

3-{[(16β,17β)-17-Hydroxy-3-(pyrrolidin-1-ylcarbonyl)estra-1(10),2,4-trien-16-yl]methyl}benzamide (12c)

1 Yield: 17 mg, 30%. IR (KBr): 1666 (C═O, amide), 3402 (OH and NH2). H NMR (CD3OD): 0.91 (s, 18-CH3), 1.15 (m, 14α-CH and 15β-CH), 1.25–1.60 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.69 (t, J = 7.0 Hz, 15α-CH),

1.89–1.99 (2t, J = 6.5 Hz, CH2CH2 of pyrolidine and 7β-CH), 2.03 (m, 12β-CH), 2.29 (m, 9α-CH), 2.39 (m, 11α-

CH), 2.47 (q, J = 12.5, 1H of 1′-CH2), 2.52 (m, 16α-CH), 2.85 (m, 6-CH2), 3.17 (dd, J1 = 3.0 Hz, J2 = 12.6 Hz, 1H of 1′-CH2), 3.47 and 3.58 (2t, J = 7.0 Hz, 2 × CH2N of pyrolidine), 3.85 (d, J = 9.4 Hz, 17α-CH), 7.21 (s, 4-CH), 7.26 (d, J = 8.2 Hz, 2-CH), 7.34–7.44 (m, 1-CH, 5″-CH and 6″-CH), 7.70 (d, J = 7.5 Hz, 4″-CH), 7.76 (s, 2″-CH).

13 C NMR (CD3OD): 13.3 (C18), 25.3 (CH2 of pyrolidine), 27.2 (C11 and CH2 of pyrolidine), 28.3 (C7), 30.4 (C6),

33.0 (C15), 38.8 (C12), 38.9 (C1′), 39.3 (C8), 43.3 (C16), 45.4 (C13), 45.8 (C9), 47.4 and 50.0 (2 × CH2N of pyrolidine), 50.0 (C14), 82.9 (C17), 125.3 (C2), 126.0 (C1), 126.4 (C4”), 128.6 (C4), 129.1 (C2″), 129.2 (C5″),

133.5 (C6″), 135.1 (C3″), 134.9 (C3), 138.2 (C5), 144.1 (C10), 144.3 (C1″), 172.1 (CON), 172.6 (CONH2). LRMS

+ for C31H39N2O3 [M + H] = 487.3. HPLC purity of 95.1% (retention time = 13.3 min).

Synthesis of 13

To a solution of acid 11 (300 mg, 0.70 mmol) in anhydrous THF (20 mL) was successively added BOP (338 mg, 0.76 mmol) and DIPEA (145 μL, 0.84 mmol) under an argon atmosphere at room temperature. The solution was stirred for 10 min and NaBH4 (30 mg, 0.79 mmol) was added in one portion and stirred again for 1 h. The resulting solution was poured into water, extracted with EtOAc, washed with brine, dried with MgSO4, and evaporated under reduced pressure. The crude compound was purified by two successive flash chromatography procedures, first using DCM/MeOH (95:5) and second acetone/hexanes (1:1) to give 88 mg (30%) of 13 (3- {[(16β,17β)-17-Hydroxy-3-(hydroxymethyl)estra-1(10),2,4-trien-16-yl]methyl}benzamide). IR (KBr): 1663 (C═O,

1 amide), 3356 (OH and NH2). H NMR (CD3OD): 0.92 (s, 18-CH3), 1.16 (m, 14α-CH and 15β-CH), 1.24–1.54 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.69 (t, J = 7.0 Hz, 15α-CH), 1.85 (m, 7β-CH), 2.05 (m, J = 12.2 Hz, 12β-

CH), 2.26 (m, 9α-CH), 2.37 (m, 11α-CH), 2.47 (q, J = 12.4, 1H of 1′-CH2), 2.53 (m, broad, 16α-CH), 2.83 (m, 6-

CH2), 3.17 (dd, J1 = 3.6 Hz, J2 = 13.0 Hz, 1H of 1′-CH2), 3.85 (d, J = 9.5 Hz, 17α-CH), 7.03 (s, 4-CH), 7.09 (d, J

= 8.0 Hz, 2-CH), 7.27 (d, J = 8.0 Hz, 1-CH), 7.38–7.45 (m, 5″-CH and 6″-CH), 7.70 (d, J = 6.0 Hz, 4″-CH), 7.76

13 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.4 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15), 38.8 (C1′), 39.0 (C12),

39.7 (C8), 43.4 (C16), 45.4 (C13), 45.8 (C9), 50.0 (C14), 65.1 (CH2OH), 83.0 (C17), 125.5 (C1), 126.0 (C2), 126.3 (C4”), 128.6 (C4), 129.1 (C2″), 129.4 (C5″), 133.5 (C6″), 134.8 (C3″), 137.6 (C3), 139.6 (C5), 140.6 (C10),

+ 144.4 (C1″), 172.7 (CONH2). LRMS for C28H38NO4 [M + CH3OH + H] = 452.3. HPLC purity of 91.8% (retention time = 9.7 min).

Synthesis of 14

To a solution of alcohol 13 (65 mg, 0.15 mmol) in DCM (7 mL) was added at 0 °C triphenylphosphine (61 mg, 0.23 mmol), DIPEA (58 mg, 80 μL, 0.45 mmol), and carbon tetrabromide (77 mg, 0.23 mmol). The solution was stirred at room temperature for 15 h. The resulting solution was poured into water, extracted with DCM, dried with MgSO4, and evaporated under reduced pressure. The crude compound was purified by flash chromatography (DCM/MeOH, 97:3) to give 45 mg (60%) of 14 (3-{[(16β,17β)-3-(bromomethyl)-17-

1 hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide). H NMR (CD3COCD3): 0.92 (s, 18-CH3), 1.16 (m, 14α- CH and 15β-CH), 1.20–1.56 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.69 (t, J = 6.2 Hz, 15α-CH), 1.84 (m, 7β-CH), 2.02 (m, under solvent peak, 12β-CH), 2.24 (m, 9α-CH), 2.35 (m, 11α-CH), 2.47 (q, J = 12.2 Hz, 1H of

1′-CH2), 2.52 (m, broad, 16α-CH), 2.81 (m, 6-CH2), 3.22 (dd, J1 = 2.9 Hz, J2 = 12.4 Hz, 1H of 1′-CH2), 3.87 (m,

17α-CH and 17β–OH), 4.58 (s, CH2Br), 6.5 (broad s, 1H of CONH2), 7.12 (s, 4-CH), 7.19 (d, J = 8.1 Hz, 2-CH),

7.29 (d, J = 8.0 Hz, 1-CH), 7.35 (t, J = 7.6 Hz, 5″-CH). 7.41 (d, J = 7.6 Hz, 6″-CH), 7.42 (broad s, 1H of CONH2),

13 7.75(d, J = 7.5 Hz, 4″-CH), 7.83 (s, 2″-CH). C NMR (CD3COCD3): 12.4 (C18), 26.1 (C11), 27.3 (C7), 29.2 (C6),

32.1 (C15), 34.0 (CH2Br), 37.8 (C1′ and C12), 38.2 (C8), 42.3 (C16), 44.4 (C13), 44.6 (C9), 48.9 (C14), 81.3 (C17), 124.8 (C4″), 125.8 (C1), 126.6 (C2), 128.0 (C2″), 128.2 (C5″), 129.7 (C4) 131.9 (C6″), 134.5 (C3″), 135.4

+ (C3), 137.1 (C5), 141.0 (C10), 143.3 (C1″), 168.4 (CONH2). LRMS for C28H37BrNO3 [M + CH3OH + H] = 514.3 and 516.3. HPLC purity of 97.8% (retention time = 18.3 min).

Synthesis of 15a–c

General Procedure for N-Akylation of Bromide 14

To a solution of 14 (25 mg, 0.06 mmol) in DCM (3 mL) was added triethylamine (43 μL, 3.0 mmol) and the appropriate amine (3.0 mmol). The solution was stirred at room temperature for 3 h. The resulting solution was poured into water, extracted with DCM, dried over a phase separator device (Biotage), and evaporated under

59 reduced pressure to give the desired N-alkylated derivative after purification by flash chromatography (typically DCM/MeOH, 95:5 to 9:1).

3-{[(16β,17β)-3-[(Dimethylamino)methyl]-17-hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide (15a)

1 Yield: 10 mg, 43%. IR (KBr): 1663 (C═O, amide), 3364 and 3429 (OH and NH2). H NMR (CDCl3): 0.91 (s, 18-

CH3), 1.16 (m, 14α-CH and 15β-CH), 1.25–1.57 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.69 (t, J = 7.0 Hz,

15α-CH), 1.84 (m, 7β-CH), 2.05 (d, J = 12.2 Hz, 12β-CH), 2.24 (s, 2 × NCH3), 2.26 (m, 9α-CH), 2.37 (m, 11α-

CH), 2.47 (q, J = 12.3 Hz, 1H of 1′-CH2), 2.52 (m, 16α-CH), 2.83 (m, 6-CH2), 3.17 (dd, J1 = 2.7 Hz, J2 = 12.5 Hz,

1H of 1′-CH2), 3.41 (s, CH2N(CH3)2), 3.85 (d, J = 9.4 Hz, 17α-CH), 7.00 (s, 4-CH), 7.05 (d, J = 8.0 Hz, 2-CH), 7.27 (d, J = 8.0 Hz, 1-CH), 7.35–7.45 (m, 5″-CH and 6″-CH), 7.70 (d, J = 6.0 Hz, 4″-CH), 7.76 (s, 2″-CH). 13C

NMR (CD3OD): 13.3 (C18), 27.3 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15), 38.8 (C12), 39.0 (C1′), 39.6 (C8), 43.4

(C16), 45.1 (N(CH3)2), 45.4 (C13), 45.8 (C9), 50.0 (C14), 64.6 (CH2N), 83.0 (C17), 126.0 (C4″), 126.3 (C1), 128.0 (C2), 129.2 (C2″), 129.4 (C5″), 131.3 (C4), 133.5 (C6″), 134.8 (C3″), 135.4 (C3), 137.7 (C5), 140.9 (C10),

+ 144.4 (C1″), 172.7 (CONH2). LRMS for C29H39N2O2 [M + H] = 447.2. HPLC purity of 97.4% (retention time = 4.1 min).

3-{[(16β,17β)-3-[Ethyl(propyl)amino)methyl]-17-hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide (15b)

1 Yield: 5 mg, 21%. IR (KBr): 1663 (C═O, amide), 3398 (OH and NH2). H NMR (CD3OD): 0.90 (t, J = 7.2 Hz,

CH3CH2CH2N), 0.92 (s, 18-CH3), 1.10 (t, J = 7.2 Hz, CH3CH2N), 1.16 (m, 14α-CH and 15β-CH), 1.25–1.60 (m,

7α-CH, 12α-CH, 8β-CH, 11β-CH, and CH2CH2CH3), 1.69 (t, J = 7.1 Hz, 15α-CH), 1.85 (m, 7β-CH), 2.05 (d, J =

12.8 Hz, 12β-CH), 2.26 (m, 9α-CH), 2.33–2.60 (11α-CH, 1H of 1′-CH2, 16α-CH, NCH2CH3, and NCH2CH2CH3),

2.83 (m, 6-CH2), 3.17 (dd, J1 = 3.1 Hz, J2 = 12.6 Hz, 1H of 1′-CH2), 3.58 (s, ArCH2N), 3.85 (d, J = 9.4 Hz, 17α- CH), 7.01 (s, 4-CH), 7.07 (d, J = 8.1 Hz, 2-CH), 7.26 (d, J = 8.0 Hz, 1-CH), 7.36–7.45 (m, 5″-CH and 6″-CH),

13 7.70 (d, J = 7.5 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 11.1 (CH3), 12.1 (CH3), 13.3 (C18), 20.1

(CH3CH2CH2N), 27.3 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15), 38.8 (C12), 39.0 (C1′), 39.6 (C8), 43.4 (C16), 45.4

(C13), 45.8 (C9), 48.1 (CH3CH2CH2N), 50.0 (C14), 55.8 (CH3CH2N), 58.4 (ArCH2N), 83.0 (C17), 126.0 (C4″), 126.3 (C1), 128.1 (C2), 129.1 (C2″), 129.4 (C5″), 131.3 (C4), 133.5 (C6″), 134.8 (C3 and C3″), 137.7 (C5), 140.8

(C10), 144.4 (C1″), 172.7 (CONH2). LRMS for C32H45N2O2 [M + H] = 489.4. HPLC purity of 91.5% (retention time = 2.1 min).

3-{[(16β,17β)-17-Hydroxy-3-(pyrrolidin-1-ylmethyl)estra-1(10),2,4-trien-16-yl]methyl}benzamide (15c)

1 Yield: 7 mg, 28%. IR (KBr): 1663 (C═O, amide), 3205 and 3383 (OH and NH2). H NMR (CD3OD): 0.91 (s, 18-

CH3), 1.16 (m, 14α-CH and 15β-CH), 1.25–1.57 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.68 (t, J = 7.0 Hz,

15α-CH), 1.82 (broad s, 7β-CH and CH2CH2 of pyrolidine), 2.04 (d, J = 12.3 Hz, 12β-CH), 2.25 (m, 9α-CH), 2.37

(m, 11α-CH), 2.47 (q, J = 12.3 Hz, 1H of 1′-CH2), 2.55 (broad s, 2 × CH2N of pyrolidine and 16α-CH), 2.82 (m,

6-CH2), 3.17 (dd, J1 = 3.0 Hz, J2 = 12.6 Hz, 1H of 1′-CH2), 3.57 (s, CH2NAr), 3.85 (d, J = 9.4 Hz, 17α-CH), 7.02 (s, 4-CH), 7.08 (d, J = 8.1 Hz, 2-CH), 7.26 (d, J = 8.0 Hz, 1-CH), 7.36–7.44 (m, 5″-CH and 6″-CH), 7.70 (d, J =

13 7.5 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 24.0 (CH2CH2 of pyrolidine), 27.3 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15), 38.8 (C12), 39.0 (C1′), 39.6 (C8), 43.4 (C16), 45.4 (C13), 45.8 (C9), 50.0 (C14),

54.8 (2(CH2N), 61.1 (ArCH2N), 83.0 (C17), 126.0 (C4″), 126.3 (C1), 128.2 (C2), 129.1 (C2″), 129.4 (C5″), 131.0

(C4), 133.5 (C6″), 134.9 (C3″), 136.0 (C3), 137.7 (C5), 140.8 (C10), 144.4 (C1″), 172.7 (CONH2). LRMS for

+ C31H41N2O2 [M + H] = 473.3. HPLC purity of 93.8% (retention time = 2.0 min).

Synthesis of 15d

To a solution of bromide 14 (30 mg, 0.06 mmol) in anhydrous DMF (3 mL) was added sodium azide (12 mg, 0.18 mmol). The solution was stirred at 60 °C for 3 h under an argon atmosphere. The resulting solution was poured into water, extracted with EtOAc, washed with brine, dried with MgSO4, and evaporated under reduced pressure. The crude compound (25 mg) was dissolved in ethanol (3 mL), and palladium on charcoal (10%) was added (10 mg). The reaction vessel was then flushed three times with hydrogen and the solution stirred for 24 h under an argon atmosphere at room temperature. The resulting solution was filtered on Celite and evaporated under reduced pressure. The crude compound was purified by flash chromatography using DCM/MeOH (95:5) as eluent to give 15d (3-{[(16β,17β)-3-(aminomethyl)-17-hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide).

1 Yield: 15 mg, 58%. IR (KBr): 1663 (C═O, amide), 3418 (OH and NH2). H NMR (CD3OD): 0.91 (s, 18-CH3), 1.16 (m, 14α-CH and 15β-CH), 1.22–1.56 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.69 (t, J = 7.0 Hz, 15α-CH),

1.84 (m, 7β-CH), 2.04 (m, 12β-CH), 2.24 (m, 9α-CH), 2.37 (m, 11α-CH), 2.47 (q, J = 12.4 Hz, 1H of 1′-CH2),

2.52 (m, broad, 16α-CH), 2.82 (m, 6-CH2), 3.17 (dd, J1 = 2.9 Hz, J2 = 12.4 Hz, 1H of 1′-CH2), 3.84 (d, J = 9.4 Hz, 17α-CH), 6.94 (s, 4-CH), 7.00 (d, J = 9.2 Hz, 2-CH), 7.25 (d, J = 8.0 Hz, 1-CH), 7.35–7.46 (m, 5″-CH and 6″-

13 CH), 7.70 (d, J = 6.0 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.4 (C11), 28.5 (C7), 30.5

(C6), 33.0 (C15), 38.8 (C12), 39.0 (C1′), 39.7 (C8), 43.4 (C16), 45.4 (CH2NH2), 45.7 (C9), 45.9 (C13), 50.0 (C14), 83.0 (C17), 126.0 (C4″), 126.2 (C1), 126.7 (C2), 129.1 (C4), 129.2 (C2″), 129.4 (C5″), 133.5 (C6″), 134.8 (C3″),

61 + 138.0 (C5), 138.8 (C3), 140.7 (C10), 144.4 (C1″), 172.7 (CONH2). LRMS for C27H35N2O2 [M + H] = 419.3. HPLC purity of 95.5% (retention time = 2.0 min).

Synthesis of 17

To a solution of compound 16(34) (350 mg, 1.07 mmol) in DCM (75 mL) under argon atmosphere was added styrene (257 μL, 233 mg, 2.24 mmol). The solution was purged by argon bubbling for 5 min, and Grubb (II) catalyst (1,3-bis(2,4,6-trimethylphenyl)-2- imidazolidinylidene)dichloro(phenylmethylene)(tricyclohexylphosphine)ruthenium) (48 mg, 0.056 mmol) was added. The solution was heated at reflux for 24 h under argon atmosphere. The solution was poured into water, extracted two times with DCM, filtered under separator phase device (Biotage), and evaporated under reduced pressure. The crude compound was purified by flash chromatography (EtOAc/hexanes: 95:5) to give 50 mg (11%) of metathesis product. This later compound (42 mg, 0.105 mmol) was then dissolved in methanol (3 mL), and an aqueous solution of HCl 10% (1 mL) was added. The reaction mixture was stirred at room temperature for 2 h. The solution was poured into a saturated NaHCO3 solution (50 mL), extracted with EtOAc, washed with brine, dried with MgSO4, and evaporated under reduced pressure to give compound 17 (3-[(E)-2-

1 phenylethenyl]estra-1(10),2,4-trien-17-one). Yield: 40 mg, 78%. IR (KBr): 1736 (C═O). H NMR (CDCl3): 0.92

(s, 18-CH3), 1.40–2.58 (residual CH and CH2), 2.95 (m, 6-CH2), 7.08 (s, CH═CH), 7.23–7.38 (m, 1-CH, 2-CH,

13 4-CH, and 3-CH of styrene), 7.51 (d, J = 7.2 Hz, 2CH of styrene). C NMR (CDCl3): 13.8 (C18), 21.6 (C15), 25.7 (C11), 26.5 (C7), 29.4 (C6), 31.6 (C12), 35.9 (C16), 38.2 (C8), 44.5 (C14), 48.0 (C13), 50.5 (C9), 124.0 (C4), 125.7 (C2), 126.4 (CH of Ph), 127.1 (C1), 127.4 (CH of Ph), 128.0 (CH of Ph), 128.4 (CH═CH), 128.6 (CH of Ph), 134.9 (C10), 136.7 (C5), 137.5 (C of Ph), 139.4 (C3), 225 (too weak not recorded). LRMS for C26H29O [M + H]+ = 357.3.

Synthesis of 18

Compound 18 (6 mg, 11% yield) was prepared from 17 using the general three-step procedure for introducing the 16β-carbamoyl-m-benzamide side chain. 3-{[(16β,17β)-17-hydroxy-3-(2-phenylethyl)estra-1(10),2,4-trien-

1 16-yl]methyl}benzamide (18). IR (KBr): 1643 (C═O, amide), 3379 (OH and NH2). H NMR (CDCl3): 0.88 (s, 18-

CH3), 1.15 (m, 14α-CH and 15β-CH), 1.26–1.60 (m, 7α-CH, 12α-CH, 8β-CH and 11β-CH), 1.75 (t, J = 7.2 Hz, 15α-CH), 1.83 (m, 7β-CH), 2.00 (m, 12β-CH), 2.25 (m, 9α-CH), 2.35 (m, 11α-CH), 2.48 (q, J = 12.8, 1H of 1′-

CH2), 2.55 (m, 16α-CH), 2.80 (m, 6-CH2 and PhCH2CH2Ph), 3.17 (dd, J1 = 3.8 Hz, J2 = 12.9 Hz, 1H of 1′-CH2),

3.88 (m, 17α-CH), 5.65 and 6.10 (2 broad s, CONH2), 6.94 (s, 4-CH), 7.01 (d, J = 8.0 Hz, 2-CH), 7.18–7.42 (m,

13 1-CH, 5″-CH, 6″-CH and PhCH2), 7.60 (d, J = 7.3 Hz, 4″-CH), 7.72 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18),

27.4 (C11), 28.7 (C7), 30.5 (C6), 33.0 (C15), 38.7 (PhCH2), 38.8 (C12), 39.0 (C1′), 39.2 (PhCH2), 39.7 (C8), 43.4 (C16), 45.4 (C13), 45.7 (C9), 50.0 (C14), 83.0 (C17), 126.0 (C4″), 126.1 (C2), 126.8 (C1), 129.1

(PhCH2CH2), 129.2 (2 × PhCH2CH2), 129.4 (PhCH2CH2), 129.5 (2 × PhCH2CH2), 129.9 (C4), 133.5 (C6″), 134.8

(C3″), 137.4 (C3), 139.0 (C5), 140.1 (C10), 143.3 (PhCH2CH2), 144.4 (C1″), 167.6 (CONH2). LRMS for

+ C34H40NO2 [M + H] = 494.3. HPLC purity of 93.5% (retention time = 21.6 min).

Synthesis of 19

To a solution of compound 16(34) (180 mg, 0.64 mmol) in acetone (18 mL) was added an aqueous solution of HCl 10% (2 mL), and the mixture was stirred at room temperature for 2 h. The solution was then poured into a saturated solution of NaHCO3 (50 mL), extracted with EtOAc, washed with brine, dried with MgSO4, and evaporated under reduced pressure to give 140 mg of deprotected compound 19 (3-ethenylestra-1(10),2,4-trien-

1 17-one).(42) H NMR (CD3COCD3): 0.90 (s, 18-CH3), 1.37–2.48 (residual CH and CH2), 2.88 (m, 6-CH2), 5.16

(d, J = 10.9 Hz, 1H of CH═CH2), 5.74 (d, J = 17.6 Hz, 1H of PhCH═CH2), 6.69 (dd, J1 = 10.9 Hz, J2 = 17.6 Hz,

13 CH═CH2), 7.16 (s, 4-CH), 7.26 (m, 1-CH and 2-CH). C NMR (CD3COCD3): 13.4 (C18), 21.4 (C15), 25.8 (C11), 26.5 (C7), 29.3 (C6, under solvent peaks), 31.9 (C12), 35.3 (C16), 38.3 (C8), 44.6 (C14), 47.7 (C13), 50.5 (C9),

112.4 (CH═CH2), 123.6 (C4), 125.7 (C2), 126.9 (C1), 135.2 (C10), 136.7 (C5), 137.1 (CH═CH2), 139.9 (C3),

+ 218.6 (C17). LRMS for C20H25O [M + H] = 281.2.

Synthesis of 20

Compound 20 (40 mg, 54% yield) was prepared from 19 using the general procedure for introduction of the 16β- carbamoyl-m-benzamide side chain. 3-{[(16β,17β)-3-ethyl-17-hydroxyestra-1(10),2,4-trien-16-

1 yl]methyl}benzamide (20). IR (KBr): 1666 (C═O, amide), 3186 and 3367 (OH and NH2). H NMR (CDCl3): 0.87

(s, 18-CH3), 1.12 (m, 14α-CH and 15β-CH), 1.22 (t, J = 7.6 Hz, CH3CH2), 1.20.-1.65 (m, 7α-CH, 12α-CH, 8β- CH, and 11β-CH), 1.74 (t, J = 7.0 Hz, 15α-CH), 1.83 (m, 7β-CH), 2.00 (m, 12β-CH), 2.24 (m, 9α-CH), 2.34 (m,

11α-CH), 2.48 (q, J = 12.3, 1H of 1′-CH2), 2.53 (m, 16α-CH), 2.58 (q, J = 7.6 Hz, CH3CH2Ph), 2.82 (m, 6-CH2),

3.16 (dd, J1 = 4.3 Hz, J2 = 12.8 Hz, 1H of 1′-CH2), 3.87 (m, 17α-CH), 5.75 and 6.12 (broad s, CONH2), 6.92 (s, 4-CH), 6.99 (d, J = 8.1 Hz, 2-CH), 7.22 (d, J = 8.0 Hz, 1-CH), 7.34–7.42 (m, 5″-CH and 6″-CH), 7.60 (d, J = 8.8

13 Hz, 4″-CH), 7.72 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 16.3 (CH3CH2), 27.4 (C11), 28.7 (C7), 29.4

(CH3CH2), 30.5 (C6), 33.0 (C15), 38.9 (C1′), 39.0 (C12), 39.7 (C8), 43.3 (C16), 45.4 (C13), 45.6 (C9), 50.0 (C14), 83.0 (C17), 126.0 (C4″), 126.1 (C2), 126.2 (C1), 129.1 (C4), 129.2 (C2″), 129.4 (C5″), 133.5 (C6″), 134.8 (C3″),

63 + 137.4 (C3), 138.6 (C5), 142.4 (C10), 144.3 (C1″), 172.7 (CONH2). LRMS for C29H40NO3 [M + CH3OH + H] = 450.3. HPLC purity of 93.6% (retention time = 19.8 min).

Synthesis of 21 and 22

These compounds were prepared from 16 as previously published in our preliminary report.(34)

Synthesis of 23a

To a solution of alcohol 22(34) (20 mg, 0.05 mmol) in DCM (1.0 mL) was added chlorodimethyl(phenylthio)- chloride methanaminium (CPMA) (45 mg, 0.19 mmol) at 0 °C under an argon atmosphere. The solution was then allowed to return at room temperature and stirred for 3 h. The crude compound was purified by flash chromatography (DCM/MeOH, 97:3) to give 12 mg (57%) of chloride 23a (3-{[(16β,17β)-3-(2-chloroethyl)-17- hydroxyestra-1(10),2,4-trien-16-yl]methyl}benzamide). IR (KBr): 1639 (C═O, amide) and 3364 (OH and NH2).

1 H NMR (CD3OD) 0.91 (s, 18-CH3), 1.15 (m, 14α-CH and 15β-CH), 1.25–1.56 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.68 (t, J = 7.0 Hz, 15α-CH), 1.83 (m, 7β-CH), 2.04 (m, 12β-CH), 2.23 (m, 9α-CH), 2.36 (m, 11α-CH),

2.47 (q, J = 12.3, 1H of 1′-CH), 2.53 (m, 16α-CH), 2.81 (m, 6-CH2), 2.96 (t, J = 7.3 Hz, CH2CH2Cl), 3.17 (dd, J1

= 2.9 Hz, J2 = 12.4 Hz, 1H of 1′-CH2), 3.70 (t, J = 7.4 Hz, CH2CH2Cl), 3.84 (d, J = 9.4 Hz, 17α-CH), 6.92 (s, 4- CH), 6.98 (d, J = 8.0 Hz, 2-CH), 7.22 (d, J = 8.0 Hz, 1-CH), 7.36–7.44 (m, 5″-CH and 6″-CH), 7.70 (d, J = 7.5

13 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.3 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15), 38.8

(C1′), 39.0 (C12), 39.6 (C8), 39.8 (CH2CH2Cl), 43.3 (C16), 45.4 (C13), 45.7 (C9), 46.0 (CH2CH2Cl), 50.0 (C14), 83.0 (C17), 126.0 (C4″), 126.4 (C1), 127.1 (C2), 129.1 (C2″), 129.4 (C5″), 130.3 (C4), 133.5 (C6″), 134.8 (C3″),

+ 136.7 (C3), 137.8 (C5), 140.0 (C10), 144.4 (C1″), 172.7 (CONH2). LRMS for C29H39ClNO3 [M + CH3OH + H] = 484.2. HPLC purity of 98.7% (retention time = 19.2 min).

Synthesis of 23b

This compound was prepared from 22 as previously published in our preliminary report.(34)

Synthesis of 23c

To a solution of bromide 23b (35 mg, 0.07 mmol) in acetone (5 mL) was added sodium iodide (15 mg, 0.1 mmol), and the solution was stirred at room temperature under argon atmosphere for 24 h. Another portion of sodium iodide (52 mg, mmol) was added and the solution stirred for an additional 24 h. The resulting solution was poured into water (100 mL) and extracted with EtOAc. The combined organic layer was washed with brine, dried with

MgSO4, and evaporated under reduced pressure. The crude compound was purified by flash chromatography (DCM/MeOH, 95:5) to give 18 mg (47%) of iodure 23c (3-{[(16β,17β)-17-hydroxy-3-(2-iodoethyl)estra-1(10),2,4-

1 trien-16-yl]methyl} benzamide). IR (KBr): 1636 (C═O, amide), 3379 (OH and NH2). H NMR (CD3OD): 0.91 (s,

18-CH3), 1.14 (m, 14α-CH and 15β-CH), 1.22–1.55 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.69 (t, J = 7.1 Hz, 15α-CH), 1.82 (m, 7β-CH), 2.04 (m, 12β-CH), 2.22 (m, 9α-CH), 2.36 (m, 11α-CH), 2.47 (q, J = 12.3 Hz, 1H of 1′-CH), 2.52 (m, 16α-CH), 2.82 (m, 6-CH2), 3.07 (t, J = 7.6 Hz, CH2CH2I), 3.17 (dd, J1 = 3.0 Hz, J2 = 12.5 Hz,

1H of 1′-CH2), 3.36 (t, J = 7.6 Hz, CH2CH2I), 3.84 (d, J = 9.4 Hz, 17α-CH), 6.89 (s, 4-CH), 6.95 (d, J = 8.0 Hz, 2-

CH), 7.22 (d, J = 8.0 Hz, 1-CH), 7.36–7.44 (m, 5″-CH and 6″-CH), 7.69 (dd, J1 = 1.4 Hz, J2 = 6.0 Hz, 4″-CH),

13 7.76 (s, 2″-CH). C NMR (CD3OD): 6.3 (CH2I), 13.3 (C18), 27.3 (C11), 28.4 (C7), 30.5 (C6), 33.0 (C15), 38.8

(C1′), 39.0 (C12), 39.6 (C8), 41.1 (CH2CH2I), 43.4 (C16), 45.4 (C13), 45.7 (C9), 50.0 (C14), 83.0 (C17), 126.0 (C4″), 126.5 (C1), 126.6 (C2), 129.1 (C4), 129.4 (C2″), 129.8 (C5″), 133.5 (C6″), 134.8 (C3″), 137.9 (C5), 139.2

+ (C3), 140.0 (C10), 144.4 (C1″), 172.7 (CONH2). LRMS for C29H39INO3 [M + CH3OH + H] = 576.2. HPLC purity of 95.0% (retention time = 16.6 min).

Synthesis of 24

Dess–Martin periodinane (67 mg, 0.16 mmol) was added in one portion to a solution of alcohol 22 (50 mg, 0.12 mmol) in DCM (4 mL) at room temperature. After 1 h, the reaction mixture was treated with saturated NaHSO3

(0.25 mL), followed by a saturated solution of NaHCO3 (5 mL) and the aqueous layer was extracted with EtOAc.

The combined organic phase was dried over MgSO4, filtered, and evaporated to give a crude compound (49 mg) that was taken up in t-BuOH (2.2 mL) and water (0.2 mL). 2-Methyl-2-butene (64 μL, 0.76 mmol), NaClO2

(13 mg, 0.14 mmol), and KH2PO4 (19 mg, 0.14 mmol) were then sequentially added, and the mixture was stirred for 12 h at room temperature. The organic solvent was evaporated under vacuum, and the resulting aqueous solution was acidified using 1 N HCl (1 mL) and extracted with EtOAc. The organic phase was washed with brine, dried over MgSO4, and evaporated. The crude compound was purified by trituration from MeOH to give 30 mg of an aldehyde. To a solution of the later compound (30 mg, 0.067 mmol) in MeOH (5 mL) was added

NaBH4 (7 mg, 0.18 mmol). The solution was stirred at room temperature for 2 h, and two other portions of NaBH4 were sequentially added (7 mg, 0.18 mg) over a period of 2 h. The resulting solution was concentrated under vacuo, diluted with DCM (25 mL), washed with water, dried with MgSO4, and evaporated under reduced

65 pressure. The crude compound was purified by flash chromatography (DCM/MeOH, 9:1) to give 15 mg (50%) of compound 24 ([(16β,17β)-16-(3-carbamoylbenzyl)-17-hydroxyestra-1(10),2,4-trien-3-yl]acetic acid). IR (KBr):

1 1659 (C═O, amide), 1705 (C═O, acid), 2300–3600 (OH, acid), 3388 (OH and NH2). H NMR (CD3OD): 0.91 (s,

18-CH3), 1.15 (m, 14α-CH and 15β-CH), 1.24 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.68 (t, J = 7.1 Hz, 15α-CH), 1.82 (m, 7β-CH), 2.04 (m, 12β-CH), 2.23 (m, 9α-CH), 2.36 (m, 11α-CH), 2.47 (q, J = 12.3 Hz, 1H of

1′-CH2), 2.53 (m, 16α-CH), 2.81 (m, 6-CH2), 3.17 (dd, J1 = 3.1 Hz, J2 = 12.7 Hz, 1H of 1′-CH2), 3.51 (s,

CH2COOH), 3.84 (d, J = 9.4 Hz, 17α-CH), 6.96 (s, 4-CH), 7.01 (d, J = 8.0 Hz, 2-CH), 7.23 (d, J = 8.0 Hz, 1-CH),

13 7.36–7.46 (m, 5″-CH and 6″-CH), 7.70 (td, J1 = 1.5 Hz, J2 = 7.5 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.3 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15), 38.8 (C1′), 39.0 (C12), 39.6 (C8), 43.4 (C16), 45.4

(C13), 45.7 (C9), 49.2 (under solvent peaks (CH2COOH)), 50.0 (C14), 83.0 (C17), 126.0 (C4”), 126.4 (C1), 127.5 (C2), 129.1 (C2″), 129.4 (C5″), 130.8 (C4), 133.3 (C3), 133.5 (C6″), 134.8 (C3″), 137.8 (C5), 140.1 (C10), 144.4

+ (C1″), 163.0 (COOH), 172.7 (CONH2). LRMS for C29H38NO5 [M + CH3OH + H] = 480.3. HPLC purity of 91.2% (retention time = 5.7 min).

Synthesis of 25

To the C17-ketonic form of 24 (37 mg, 0.08 mmol), obtained from Dess−Martin oxidation of 22, dissolved in anhydrous DMF (3 mL) and under argon atmosphere was added BOP (40 mg, 0.09 mmol), methyl amine (115 μL, 0.03 mmol, 2.0 M in THF), and DIPEA (18 μL, 0.11 mmol). The solution was stirred at room temperature.

After 3 h, the mixture was poured into water, extracted with EtOAc, washed with brine, dried with MgSO4, and evaporated under reduced pressure to give 41 mg of a crude amide. This compound was taken up into a mixture of MeOH/DCM (9:1), then treated with NaBH4 (15 mg, 0.40 mmol) and stirred for 30 min at room temperature.

The resulting solution was poured into water, extracted with EtOAc, washed with brine, dried with MgSO4, and evaporated under reduced pressure. The crude amide compound was purified by flash chromatography (DCM/MeOH, 95:5) to give 6 mg (15%) of 25 (3-{[(16β,17β)-17-hydroxy-3-[2-(methylamino)-2-oxoethyl]estra-

1 1(10),2,4-trien-16-yl]methyl}benzamide). IR (KBr): 1655 (C═O, amide), 3360 (OH and NH2). H NMR (CD3OD):

0.91 (s, 18-CH3), 1.15 (m, 14α-CH and 15β-CH), 1.25–1.50 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.68 (t, J = 7.0 Hz, 15α-CH), 1.83 (m, 7β-CH), 2.04 (m, 12β-CH), 2.24 (m, 9α-CH), 2.36 (m, 11α-CH), 2.47 (q, J = 12.3,

1H of 1′-CH2), 2.52 (m, 16α-CH), 2.71 (s, CH3NHCO), 2.86 (m, 6-CH2), 3.17 (dd, J1 = 2.5 Hz, J2 = 12.3 Hz, 1H of 1′-CH2), 3.41 (s, ArCH2CO), 3.84 (d, J = 9.4 Hz, 17α-CH), 6.96 (s, 4-CH), 7.01 (d, J = 8.0 Hz, 2-CH), 7.23 (d, J = 8.0 Hz, 1-CH), 7.36–7.44 (m, 5″-CH and 6″-CH), 7.70 (d, J = 7.4 Hz, 4″-CH), 7.76 (s, 2″-CH). 13C NMR

(CD3OD): 13.3 (C18), 26.5 (CH3NHCO), 27.3 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15), 38.8 (C1′), 39.0 (C12),

39.7 (C8), 43.3 (C16), 43.4 (CH2CONH), 45.4 (C13), 45.7 (C9), 50.0 (C14), 83.0 (C17), 126.0 (C4″), 126.5 (C1),

127.3 (C2), 129.1 (C2″), 129.4 (C5″), 130.5 (C4), 133.5 (C6″), 133.8 (C3), 134.8 (C3″), 138.0 (C5), 140.2 (C10),

+ 144.4 (C1″), 167.7 (CONHCH3), 175.1 (CONH2). LRMS for C29H37N2O3 [M + H] = 461.2. HPLC purity of 90.2% (retention time = 4.1 min).

Synthesis of 26

To a solution of compound 16(34) (1.5 g, 4.82 mmol) in DCM (400 mL) was added allyloxymethyl-benzene (1.3 g, 9.55 mmol), and the mixture was stirred under an argon atmosphere for 5 min. Grubb’s catalyst (204 mg, 0.24 mmol) was then added, and the solution was heated at 60 °C under argon atmosphere for 48 h. The solution was evaporated and the residue purified by flash chromatography using EtOAc/hexanes (5:95) to give 130 mg (6%) of metathesis product. To a solution of this compound (120 mg, 0.28 mmol) in acetone (3 mL) was added aqueous 10% HCl (3 mL), and the solution was stirred at room temperature for 6 h. The resulting solution was diluted with water (60 mL), neutralized with a saturated NaHCO3 solution, and extracted with EtOAc. The organic layers were combined and washed with brine, dried with MgSO4, and evaporated under reduced pressure. The crude compound was purified by flash chromatography (EtOAc/hexanes, 5:95) to give 80 mg (69%) of compound

1 26 (3-[(1E)-3-(benzyloxy)prop-1-en-1-yl]estra-1(10),2,4-trien-17-one). H NMR (CDCl3): 0.91 (s, 18-CH3), 1.40–

2.45 (residual CH and CH2), 2.52 (dd, J1= 8.6 Hz, J2 = 18.7 Hz, 16β-CH), 2.92 (m, 2H, 6-CH2), 4.19 (dd, J1 = 1.1

Hz, J2 = 6.1 Hz, OCH2CH═CH), 4.57 (s, OCH2Ph), 6.30 (m, CH═CHCH2O), 6.58 (d, J = 16.0 Hz, CH═CHCH2O),

13 7.14 (s, 4-CH), 7.18–7.39 (m, OCH2Ph, 1-CH, and 2-CH). C NMR (CDCl3): 13.8 (C18), 21.6 (C15), 25.7 (C11),

26.5 (C7), 29.4 (C6), 31.6 (C12), 35.8 (C16), 38.1 C8), 44.4 (C14), 48.0 (C13), 50.5 (C9), 70.8 (OCH2Ph), 72.0

(PhOCH2═CH), 123.9 (C4), 125.4 (C2), 125.6 (CH of Ph), 127.1 (C1), 127.6 (CH of Ph), 127.8 (CH of Ph), 128.4

(CH of Ph), 132.4 (Ph-CH═CH2), 134.3 (C10), 136.6 (C5), 138.3 (C of Ph), 139.4 (C3), 217.0 (C17). LRMS for

+ C28H33O2 [M + H] = 401.3.

Synthesis of 27

Compound 27 (38 mg, 55%) was obtained from 26 using the general three-step procedure used for introducing the 16β-carbamoyl-m-benzamide side chain. 3-{[(16β,17β)-17-hydroxy-3-(3-hydroxypropyl)estra-1(10),2,4-

1 trien-16-yl]methyl}benzamide (27). IR (KBr): 1663 (C═O, amide), 3383 (OH and NH2). H NMR (CD3OD): 0.91

(s, 18-CH3), 1.15 (m, 14α-CH and 15β-CH), 1.20–1.58 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.68 (t, J =

7.0 Hz, 15α-CH), 1.80 (m, 7β-CH and CH2CH2CH2OH), 2.04 (m, 12β-CH), 2.21 (m, 9α-CH), 2.35 (m, 11α-CH),

2.47 (q, J = 12.2, 1H of 1′-CH), 2.52 (m, 16α-CH), 2.59 (t, J = 7.2 Hz, ArCH2CH2), 2.79 (m, 6-CH2), 3.17 (d, J =

67 12.5 Hz, 1H of 1′-CH2), 3.56 (t, J = 6.5 Hz, CH2CH2OH), 3.84 (d, J = 9.3 Hz, 17α-CH), 6.87 (s, 4-CH), 6.93 (d, J = 8.0 Hz, 2-CH), 7.18 (d, J = 8.0 Hz, 1-CH), 7.36–7.44 (m, 5″-CH and 6″-CH), 7.70 (d, J = 6.0 Hz, 4″-CH), 7.76

13 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.4 (C11), 28.7 (C7), 30.5 (C6), 32.6 (CH2CH2CH2), 33.0 (C15),

35.5 (ArCH2CH2), 38.9 (C1′), 39.0 (C12), 39.7 (C8), 43.4 (C16), 45.4 (C13), 45.7 (C9), 50.0 (C14), 62.3 (CH2OH), 83.0 (C17), 126.0 (C4″), 126.2 (C1), 127.2 (C2), 129.1 (C2″), 129.4 (C5″), 129.9 (C4), 133.5 (C6″), 134.8 (C3″),

137.5 (C5), 138.9 (C3), 140.3 (C10), 144.3 (C1″), 172.7 (CONH2). LRMS for C30H42NO4 [M + CH3OH + H] = 480.3. HPLC purity of 90.6% (retention time = 13.5 min).

Synthesis of 28

To a solution of compound 27 (28 mg, 0.063 mmol) in DCM (3 mL) was added at 0 °C triphenylphosphine (33 mg, 0.13 mmol) and carbon tetrabromide (42 mg, 0.13 mmol). The solution was stirred at 0 °C for 40 min, and second portions of triphenylphosphine (13 mg, 0.05 mmol) and carbon tetrabromide (17 mg, 0.05 mmol) were added. After 1 h at 0 °C, the resulting solution was poured into water (50 mL), extracted with DCM, dried with

MgSO4, and evaporated under reduced pressure. The crude compound was purified by flash chromatography (DCM/MeOH: 97:3) to give 8 mg (25%) of 287 (3β-{[(16β,1 )-3-(3-bromopropyl)-17-hydroxyestra-1(10),2,4-trien-

1 16-yl]methyl} benzamide). IR (KBr): 1663 (C═O, amide), 3383 (OH and NH2). H NMR (CDCl3): 0.88 (s, 18-

CH3), 1.12 (m, 14α-CH and 15β-CH), 1.20–1.58 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.75 (t, J = 7.1 Hz,

15α-CH), 1.83 (m, 7β-CH), 2.00 (m, 12β-CH), 2.15 (m, CH2CH2Br), 2.23 (m, 9α-CH), 2.33 (m, 11α-CH), 2.48 (q,

J = 12.3, 1H of 1′-CH), 2.53 (m, 16α-CH), 2.70 (t, J = 7.2 Hz, ArCH2CH2), 2.82 (m, 6-CH2), 3.17 (dd, J1 = 4.3 Hz,

J2 = 12.9 Hz, 1H of 1′-CH2), 3.41 (t, J = 6.6 Hz, CH2CH2Br), 3.86 (m, 17α-CH), 5.60 and 6.10 (broad s of CONH2), 6.91 (s, 4-CH), 6.98 (d, J = 7.9 Hz, 2-CH), 7.22 (d, J = 8.0 Hz, 1-CH), 7.34–7.42 (m, 5″-CH and 6″-CH), 7.60 (d,

13 J = 7.3 Hz, 4″-CH), 7.72 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.4 (C11), 28.6 (C7), 30.5 (C6), 33.0 (C15),

33.7 (CH2CH2CH2Br), 34.4 (ArCH2CH2), 35.7 (CH2Br), 38.9 (C1′), 39.0 (C12), 39.7 (C8), 43.3 (C16), 45.4 (C13), 45.7 (C9), 50.0 (C14), 83.0 (C17), 126.0 (C4″), 126.4 (C1), 126.8 (C2), 129.1 (C2″), 129.4 (C5″), 130.0 (C4),

133.5 (C6″), 134.8 (C3″), 137.7 (C5), 139.0 (C3), 139.3 (C10), 144.3 (C1″), 172.7 (CONH2). LRMS for

+ C30H41BrNO3 [M + CH3OH + H] = 542.2 and 544.3. HPLC purity of 97.5% (retention time = 16.6 min).

Synthesis of 29

To a solution of compound 1(29) (150 mg, 0.37 mmol) in acetone (3 mL) were added NaOH (50 mg, 1.25 mmol) and allyl bromide (40 μL, 0.46 mmol). After the mixture was stirred at 60 °C for 5 h, the resulting solution was

diluted with EtOAc, and the solution was washed with a saturated aqueous solution of ammonium chloride, brine, dried with MgSO4, and evaporated under reduced pressure to give 165 mg of the corresponding 3-O-allyl compound. The crude compound was found sufficiently pure to pursue to the next step without further purification. Sodium periodate (108 mg, 0.50 mmol) was added to water (0.5 mL) and the solution stirred at 0

°C for 5 min followed by the addition of RuCl3–H2O (4 mg, 0.02 mmol), EtOAc (1 mL), and acetonitrile (1 mL). The 3-O-allyl compound (150 mg, 0.33 mmol) was added to the previous solution and stirred for 90 s. The reaction was quenched by the addition of a saturated aqueous solution of Na2S2O3 (2 mL). The aqueous layer was extracted with EtOAc, and the combined organic phase was dried with Na2SO4 and evaporated under reduced pressure. The residue was dissolved in a mixture of THF (1 mL) and water (1 mL), and NaBH4 (13 mg, 0.34 mmol) was added. The solution was stirred at room temperature for 20 min, water was then added (10 mL), and the mixture was extracted with DCM. The organic phase was washed with a saturated NaHCO3 solution, dried with Na2SO4, and evaporated under reduced pressure. The residue was dissolved in THF (1 mL) and water (1 mL) at 0 °C, sodium periodate (144 mg, 0.67 mmol) was added in small portions, and the solution was stirred for 20 min at room temperature. Ethylene glycol (50 μL) was added, the reaction mixture was diluted with water

(3 mL), and the mixture was extracted with EtOAc. The combined organic phase was dried with Na2SO4 and evaporated under reduced pressure. The crude compound was purified by flash chromatography (EtOAc/hexanes, 9:1) to give 25 mg (15%) of compound 29 (3-{[(16β,17β)-17-hydroxy-3-(2-

1 hydroxyethoxy)estra-1(10),2,4-trien-16-yl]methyl} benzamide). H NMR (CD3OD): 0.91 (s, 18-CH3), 1.14 (m, 14α-CH and 15β-CH), 1.20–1.50 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.67 (m, 15α-CH), 1.81 (m, 7β-CH),

2.02 (m 12β-CH), 2.17 (m, 9α-CH), 2.33 (m, 11α-CH), 2.47 (q, J = 12.3, 1H of 1′-CH2), 2.52 (m, 16α-CH), 2.78

(m, 6-CH2), 3.17 (dd, J1 = 3.0 Hz, J2 = 12.5 Hz, 1H of 1′-CH2), 3.85 (m, 17α-CH and CH2OH), 3.99 (m,

OCH2CH2OH), 6.62 (d, J = 2.7 Hz, 4-CH), 6.70 (dd, J1 = 2.7 Hz, J2 = 8.6 Hz, 2-CH), 7.17 (d, J = 8.6 Hz, 1-CH),

13 7.36–7.44 (m, 5″-CH and 6″-CH), 7.70 (dd, J1 = 1.5 Hz, J2 = 6.0 Hz, 4″-CH), 7.76 (s, 2″-CH). C NMR (CD3OD): 13.3 (C18), 27.5 (C11), 28.6 (C7), 30.8 (C6), 33.0 (C15), 38.8 (C1′), 39.0 (C12), 39.9 (C8), 43.4 (C16), 45.4

(C9), 45.5 (C13), 49.9 (C14), 61.8 (CH2OH), 70.4 (OCH2CH2), 83.0 (C17), 112.2 (C2), 115.4 (C4), 126.0 (C4″), 127.2 (C1), 129.1 (C2″), 129.4 (C5″), 133.5 (C6″), 134.0 (C5), 134.8 (C3″), 138.9 (C10), 144.4 (C1″), 158.2 (C3),

+ 172.7 (CONH2). LRMS for C29H40NO5 [M + CH3OH + H] = 450.3.

Synthesis of 30

To a solution of compound 29 (20 mg, 0.46 mmol) in DCM (2 mL) and anhydrous THF (1 mL) was added at 0 °C triphenylphosphine (23 mg, 0.87 mmol) and carbon tetrabromide (29 mg, 0.87 mmol). The solution was stirred at 0 °C for 40 min, and second portions of triphenylphosphine (20 mg, 0.46 mmol) and carbon

69 tetrabromide (23 mg, 0.87 mmol) were added. The solution was stirred at 0 °C for 40 min, and then third portions of triphenylphosphine (20 mg, 0.46 mmol) and carbon tetrabromide (23 mg, 0.87 mmol) were added. After 1 h at 0 °C, the resulting solution was poured into water (100 mL), extracted with DCM, dried with MgSO4, and evaporated under reduced pressure. The crude compound was purified by flash chromatography (DCM/diethyl ether/MeOH, 75:20:5) to give 12 mg (52%) of 30 (3-{[(16β,17β)-3-(2-bromoethoxy) -17-hydroxyestra-1(10),2,4-

1 trien-16-yl]methyl}benzamide). IR (KBr): 1663 (C═O, amide), 3367 (OH and NH2). H NMR (CD3OD): 0.91 (s,

18-CH3), 1.14 (m, 14α-CH and 15β-CH), 1.24–1.52 (m, 7α-CH, 12α-CH, 8β-CH, and 11β-CH), 1.67 (t, J = 6.8 Hz, 15α-CH), 1.82 (m, 7β-CH), 2.03 (m, 12β-CH), 2.18 (m, 9α-CH), 2.33 (m, 11α-CH), 2.47 (q, J = 12.3 Hz, 1H of 1′-CH2), 2.52 (m, 16α-CH), 2.80 (m, 6-CH2), 3.17 (d, J = 10.4 Hz, 1H of 1′-CH2), 3.67 (t, J = 5.6 Hz,

OCH2CH2Br), 3.83 (d, J = 9.2 Hz, 17α-CH), 4.25 (t, J = 5.6 Hz, OCH2CH2Br), 6.62 (s, 4-CH), 6.70 (d, J1 = 8.7 Hz, 2-CH), 7.19 (d, J = 8.6 Hz, 1-CH), 7.36–7.44 (m, 5″-CH and 6″-CH), 7.70 (d, J = 7.4 Hz, 4″-CH), 7.76 (s, 2″-

13 CH). C NMR (CD3OD): 13.3 (C18), 27.5 (C11), 28.6 (C7), 30.7 (C6 and CH2Br), 33.0 (C15), 38.8 (C1′), 39.0

(C12), 39.8 (C8), 43.4 (C16), 45.4 (C9), 45.4 (C13), 49.9 (C14), 69.2 (OCH2CH2Br), 83.0 (C17), 113.3 (C2), 115.6 (C4), 126.0 (C4″), 127.4 (C1), 129.1 (C2″), 129.4 (C5″), 133.5 (C6″), 134.5 (C5), 134.8 (C3″), 139.1 (C10),

+ 144.4 (C1″), 157.5 (C3), 172.7 (CONH2). LRMS for C29H39BrNO4 [M + CH3OH + H] = 544.2 and 546.3. HPLC purity of 98.0% (retention time = 18.6 min).

Multigram Synthesis of 23b

Synthesis of 19

3-Vinyl-estra-1(10),2,4-trien-17-one (19) was synthesized in two steps (6.4 g, 72%) from estrone (8.6 g, 31.7 mmol) using a published procedure.(45)

Synthesis of 31 from 19

To a solution of alkene 19 (6.1 g, 21.7 mmol) in a mixture of acetone and acetonitrile (1:2) 450 mL was added a saturated aqueous solution of NaHCO3 (300 mL) and oxone (20.0 g, 65.1 mmol). The solution was stirred at room temperature for 4 h, then poured into water and extracted with EtOAc. The organic phase was washed with brine, dried over Na2SO4, and evaporated under reduce pressure. Purification by flash chromatography with hexanes/EtOAc (8:2) yielded 4.9 g (76%) of 3-(oxiran-2-yl)estra-1(10),2,4-trien-17-one (31). IR (film): 1736

1 (C═O, ketone). H NMR (CD3COCD3): 0.91 (s, 18-CH3), 1.40–2.50 (residual CH and CH2), 2.76 and 3.06 (2m,

CH2OCH), 2.89 (m, 6-CH2), 3.79 (dd, J1 = 4.0 Hz, J2 = 2.6 Hz, CHOCH2), 7.01 (s, 4-CH), 7.06 (d, J = 8.1 Hz, 2-

13 CH), 7.29 (d, J = 8.1 Hz, 1-CH). C NMR (CD3COCD3): 13.4 (C18), 21.4 (C15), 25.8 (C11), 26.5 (C7), 29.3 (C6)

(under solvent peaks), 31.9 (C12), 35.3 (C16), 38.3 (C8), 44.6 (C14), 47.7 (C13), 50.2 (CH2O of epoxide), 50.5 (C9), 51.6 (CHO of epoxide), 123.0 (C4), 125.6 (C2), 126.2 (C1), 135.6 (C10), 136.8 (C5), 140.0 (C3), 227.0

+ (C17). LRMS for C20H25O2 [M + H] = 297.2. HPLC purity of 99.9% (retention time = 13.2 min).

Synthesis of 32 from 31

To a solution of oxirane 31 (4.9 g, 16.5 mmol) in MeOH in a Schlenck reactor were added ammonium formate (10.4 g, 165 mmol) and 10% palladium on charcoal (2.50 g) under an argon atmosphere at room temperature. The solution was heated at 70 °C for 2 h. The suspension was filtered over Celite and the filtrate evaporated under reduced pressure to dryness. Purification by flash chromatography with hexanes/EtOAc (7:3) yielded 3.6 g (73%) of 3-(2-hydroxyethyl)estra-1(10),2,4-trien-17-one (32). IR (film): 1728 (C═O, ketone), 3464 (OH). 1H

NMR (CDCl3): 0.91 (s, 18-CH3), 1.38–2.44 (residual CH and CH2), 2.51 (dd, J1 = 8.6 Hz, J2 = 18.8 Hz, 16β-CH),

2.82 (t, J = 6.5 Hz, CH2CH2OH), 2.91 (m, 6-CH2), 3.86 (t, J = 6.5 Hz, CH2CH2OH), 6.98 (s, 4-CH), 7.03 (d, J =

13 8.0 Hz, 2-CH), 7.25 (d, J = 9.0 Hz, 1-CH). C NMR (CDCl3): 13.8 (C18), 21.6 (C15), 25.7 (C11), 26.5 (C7), 29.4

(C6), 31.6 (C12), 35.9 (C16), 38.2 (C8), 38.7 (CH2CH2OH), 44.3 (C14), 48.0 (C13), 50.5 (C9), 63.6 (CH2OH),

125.6 (C1), 126.4 (C2), 129.7 (C4), 135.8 (C3), 136.7 (C5), 137.9 (C10), 221.5 (C17). LRMS for C20H27O2 [M + H]+ = 299.2. HPLC purity of 99.3% (retention time = 9.7 min).

Transformation of 32 to 22 and Then to 23b

The compound 22(34) was synthesized in a yield of 84% (4.4 g) from compound 32 (3.6 g, 12.1 mmol) using the three-step general procedure for the introduction of 16β-carbamoyl-m-benzamide side chain described above. The bromination of 22 (4.4 g, 10.1 mmol) was performed using the same procedure described above and provided 23b (2.5 g, 50%) in excellent HPLC purity (98.5%).

17β-HSD1 Inhibition Assay

Breast cancer T-47D cells were seeded in a 24-well plate (3000 cells/well) in RPMI medium supplemented with insulin (50 ng/mL) and 5% dextran-coated charcoal-treated fetal bovine serum (FBS), which was used rather than untreated 10% FBS, to remove the remaining steroid hormones. Stock solution of each compound to be

71 tested was previously prepared in EtOH and diluted with culture medium to achieve appropriate concentrations prior to use. After 24 h of incubation, a diluted solution was added to the cells to obtain the appropriate final concentration (0.1 or 1 μM for screening and ranging from 1 nM to 10 μM for IC50 value determination). The final concentration of EtOH in the well was adjusted to 0.1%. Additionally, a solution of [14C]-E1 (American Radiolabeled Chemicals, Inc., St. Louis, MO, USA) was added to obtain a final concentration of 60 nM. Cells were incubated for 24 h, and each inhibitor was assessed in triplicate. After incubation, the culture medium was removed and labeled steroids (E1 and E2) were extracted with diethyl ether. The organic phase was evaporated to dryness with nitrogen. Residues were dissolved in DCM, dropped on silica gel thin layer chromatography plates (EMD Chemicals Inc., Gibbstown, NJ, USA), and eluted with toluene/acetone (4:1) as solvent system. Substrate [14C]-E1 and metabolite [14C]-E2 were identified by comparison with reference steroids (E1 and E2) and quantified using the Storm 860 system (Molecular Dynamics, Sunnyvale, CA, USA). The percentage of transformation and the percentage of inhibition were calculated as follow: % transformation = 100[14C]-E2/([14C]- E1 + [14C]-E2) and % of inhibition = 100(% transformation without inhibitor – % transformation with inhibitor)/% transformation without inhibitor.

17β-HSD1 Inactivation Assay

Purified 17β-HSD1 kindly provided by Dr. Sheng-Xiang Lin (CHU de Québec—Research Center)(33) was used for inactivation/competition assays. An enzyme solution was diluted in physiological buffer (100 mM Tris, 20% glycerol, 0.2 mg/mL BSA, 2 mM EDTA) containing 1 mM of NADPH as cofactor and was treated in triplicate with an ethanolic solution of compound 23b to reach the accurate concentration, with or without unlabeled E1 (500 nM). The mixture was then preincubated at 37 °C with shaking before simultaneous dilution 1:20 in physiological buffer and addition of radiolabeled substrate to a final concentration of 60 nM [14C]-E1. Transformation of substrate was stopped after 45 min of incubation at 37 °C with shaking by cooling down on ice the enzyme medium and adding equal volume of diethyl ether for further extractions, separation by TLC, and quantification of radiolabeled steroids as described above.

Estrogen-Sensitive Cell Proliferation Assays (Estrogenic Activity)

Quantification of cell growth was determined by using CellTiter 96 aqueous solution cell proliferation assay (Promega, Nepean, ON, Canada) following the manufacturer′s instructions. MCF-7 or T-47D cells were resuspended in their medium (DMEM-F12 or RPMI, respectively) supplemented with insulin (50 ng/mL) and 5%

dextran-coated charcoal treated FBS to remove remaining estrogenic hormones present in the serum and medium. Aliquots (100 μL) of the cell suspension were seeded in 96-well plates (3000 cells/well). After 48 h, the medium was changed for a new one containing an appropriate concentration of the steroid derivative to be tested. The medium was replaced every two days. Cells were left to grow for seven days, either in presence or absence of the compound to be tested.

Inhibition Assays for 17β-HSD2, 17β-HSD7, and 17β-HSD12

Selectivity of compounds 1 and 23b were assessed as previously described(30) in stably transfected HEK-293 cells kindly provided by Dr. Van Luu-The (CHU de Québec—Research Center). Cells were seeded in 24-well plates in protocol medium (MEM medium supplemented with 5% dextran-coated charcoal stripped, G418 (700 μg/mL), penicillin (100 IU/mL), streptomycin (100 μg/mL), insulin (50 ng/mL), glutamine (2 mM), nonessential amino acids (0.1 mM), and pyruvate (1 mM)). After 48 h incubation, treatment with DMSO solution (<0.5% final) of compound 1 or 23b in protocol medium was conducted. Radiolabeled substrates ([14C]-E2 for 17β-HSD2 and [14C]-E1 for 17β-HSD7 and 17β-HSD12), obtained from American Radiolabeled Chemicals Inc. (St. Louis, MO, USA), were added to obtain a final concentration of 60 nM. Once a substrate transformation of 12–25% was reached, radiolabeled steroids (E1 and E2) were extracted from the supernatant with diethyl ether. Separation and quantification of substrates and metabolites were conducted as previously described in the section on 17β- HSD1 inhibition assays.

Inhibition of CYP3A4

We used the commercially available P450 Inhibition Kit CYP3A4/DBF (BD Gentest) of BD Biosciences (Mississauga, ON, Canada) according to manufacturer’s instructions, with the exception that 23b was dissolved in a mixture of dimethylsulfoxide/acetonitrile (5:95) instead of only acetonitrile. The enzyme activity was measured by the fluorescence caused by the enzymatic transformation of dibenzylfluorescein (DBF). Inhibitory potencies were determined as IC50 values, which were calculated with GraphPad Prism 5 software to express inhibition potency.

73 Molecular Modeling

Docking simulations were performed using MOE 2012.10.(62) The crystal structure coordinates of 17β-HSD1, including inhibitor 1 and cofactor NADP, were taken from PDB ID 3HB5 (PMID: 19929851). Hydrogen atoms were added using the Protonate 3D tool included in MOE. The protein complex was prepared using the LigX tool, included in MOE, to adjust H, rotamers, and to minimize the system’s energy as previously described (PMID: 22566074). Docking simulations were performed using the rigid receptor protocol and default parameters. Validation of the docking protocol was carried out by a self-docking of compound 1, leading to an RMSD of 0.37 Å between the docked and the crystallographic structures. Because compound 23b shares its core structure with compound 1, no further optimization of the docking protocol was considered.

Compound 23b was built in MOE based on compound 1. Hydrogens were readjusted and molecules were energy-minimized prior to docking using the same protocol as for compound 1. Three docking calculations were done for compound 23b, each using a different binding site conformation: (A) using the crystallographic conformation, (B) with Glu-282 side chain removed from the binding site and exposed to the solvent, and (C) in addition to point 2, the mutation of His-221 to Ala. The Glu-282 conformation was modified using the Rotamer Explorer tool in MOE, and the lowest energy conformer not pointing toward the binding site was selected. The mutation of His-221 to Ala was done using the Sequence tool in MOE, and no further energy minimization was required.

The authors declare the following competing financial interest(s): R. Maltais and D. Poirier have ownership interest in a patent application. No potential conflicts of interest were disclosed by the other authors.

Acknowledgement

We thank the Canadian Institutes of Health Research (CIHR) for an operating grant. We also thank Charles Ouellet for performing an experiment for the estrogenicity, Dr. Van Luu-The for providing HEK-293 transfected cells, and Dr. Sheng-Xiang Lin for providing purified 17β-HSD1. Careful reading of the manuscript by Micheline Harvey is also greatly appreciated.

Abbreviations Used

ACN acetonitrile BOP (benzotriazol-1-yloxy)tris(dimethylamino) phosphonium hexafluorophosphate DCM dichloromethane DHEA dehydroepiandrosteone 5-diol 5-androsten-3β,17β-diol 4-dione 4-androstene-3,17-dione DIPEA diisopropylethylamine DMF dimethylformamide DMSO dimethyl sulfoxide E1 estrone E2 estradiol ERα estrogen receptor alpha ERβ estrogen receptor beta EtOAc ethyl acetate EtOH ethanol FBS fetal bovine serum 3β-HSD 3β-hydroxysteroid dehydrogenase 17β-HSD 17β-hydroxysteroid dehydrogenase MeOH methanol p-TSA para-toluene sulfonic acid SAR structure–activity relationship TEA triethylamine THF tetrahydrofuran

75 References

1. U.S. Breast Cancer Statistics; Breastcancer.org: Ardmore, PA, 2013; http://www.breastcancer.org/symptoms/understand_bc/statistics.

2. Cancer Trends Progress Report—2011/2012 Update; National Cancer Institute, NIH, DHHS: Bethesda, MD, August 2012; http://progressreport.cancer.gov.

3. Brueggemeier, R. W.; Hackett, J. C.; Diaz-Cruz, E. S.Aromatase inhibitors in the treatment of breast cancer Endocr. Rev. 2005, 26, 331– 345

4. Moeller, G.; Adamski, J.Integrated view on 17beta-hydroxysteroid dehydrogenase Mol. Cell. Endocrinol. 2009, 301, 7– 19

5. Penning, T. M.17beta-Hydroxysteroid dehydrogenase: inhibitors and inhibitor design Endocr.-Relat. Cancer 1996, 3, 41– 56

6. von Angerer, E. The estrogen receptor as a target for rational drug design. In Molecular Biology Intelligence Unit; R.G. Landes Company: Austin, TX, 1995.

7. Simard, J.; Vincent, A.; Duchesne, R.; Labrie, F.Full oestrogenic activity of C19-delta 5 adrenal steroids in rat pituitary lactotrophs and somatotrophs Mol. Cell. Endocrinol. 1988, 55, 233– 242

8. Nagasaki, S.; Miki, Y.; Akahira, J. I.; Suzuki, T.; Sasano, H.17β-Hydroxysteroid dehydrogenases in human breast cancer Ann. N. Y. Acad. Sci. 2009, 1155, 25– 32

9. Sasano, H.; Frost, A. R.; Saitoh, R.; Harada, N.; Poutanen, M.; Vihko, R.; Bulun, S. E.; Silverberg, S. G.; Nagura, H.Aromatase and 17 beta-hydroxysteroid dehydrogenase type 1 in human breast carcinoma J. Clin. Endocrinol. Metab. 1996, 81, 4042– 4046

10. Poutanen, M.; Isomaa, V.; Lehto, V. P.; Vihko, R.Immunological analysis of 17beta-hydroxysteroid dehydrogenase in benign and malignant human breast tissue Int. J. Cancer. 1992, 50, 386– 390

11. Ariga, N.; Moriya, T.; Suzuki, T.; Kimura, M.; Ohuchi, N.; Satomi, S.; Sasano, H.17beta- Hydroxysteroid dehydrogenase type 1 and type 2 in ductal carcinoma in situ and intraductal proliferative lesions of the human breast Anticancer Res. 2000, 20, 1101– 1108

12. Suzuki, T.; Moriya, T.; Ariga, N.; Kaneko, C.; Kanazawa, M.; Sasano, H.17beta-Hydroxysteroid dehydrogenase type 1 and type 2 in human breast carcinoma: a correlation to clinicopathological parameters Br. J. Cancer 2000, 82, 518– 523

13. Gunnarsson, C.; Hellqvist, E.; Stal, O.17beta-Hydroxysteroid dehydrogenases involved in local oestrogen synthesis have prognostic significance in breast cancer Br. J. Cancer 2005, 92, 547– 552

14. Miyoshi, Y.; Ando, A.; Shiba, E.; Taguchi, T.; Tamaki, Y.; Noguchi, S.Involvement of up-regulation of 17beta-hydroxysteroid dehydrogenase type 1 in maintenance of intratumoral high estradiol levels in postmenopausal breast cancers Int. J. Cancer 2001, 94, 685– 689

15. Chanplakorn, N.; Chanplankorn, P.; Suzuki, T.; Ono, K.; Chan, M. S. M.; Miki, Y.; Saji, S.; Ueno, T.; Toi, M.; Sasano, H.Increased estrogen sulfatase (STS) and 17β-hydroxysteroid dehydrogenase type

1 (17β-HSD1) following neoadjuvant aromatase inhibitor therapy in breast cancer patients Breast Cancer Res. Treat. 2010, 120, 639– 648

16. Cremoux, P.Hormone therapy and breast cancer Bull. Cancer 2011, 98, 1311– 1319

17. Saloniemi, T.; Järvensivu, P.; Koskimies, P.; Jokela, H.; Lamminen, T.; Ghaem-Maghami, S.; Dina, R.; Damdimopoulou, P.; Mäkelä, S.; Perheentupa, A.; Kujari, H.; Brosens, J.; Poutanen, M.Novel hydroxysteroid (17β) dehydrogenase 1 inhibitors reverse estrogen-induced endometrial hyperplasia in trangenic mice Am. J. Pathol. 2010, 176, 11443– 145

18. Theobald, A. J.Management of advanced breast cancer with endocrine therapy: the role of the primary healthcare team Int. J. Clin. Pract. 2000, 54, 665– 669

19. Dizerega, G. S.; Barber, D. L.; Hodgen, G. D.Endometriosis: role of ovarian steroids in initiation, maintenance, and suppression Fertil. Steril. 1980, 33, 649– 653

20. Lanisnik-Rizner, T.Estrogen metabolism and action in endometriosis Mol. Cell. Endocrinol. 2009, 307, 8– 18

21. Poirier, D.Inhibitors of 17β-hydroxysteroid dehydrogenase Curr. Med. Chem. 2003, 10, 453– 477

22. Day, J. M.; Tutill, H. J.; Purohit, A.; Reed, M. J.Design and validation of specific inhibitors of 17β- hydroxysteroid dehydrogenases for therapeutic application in breast and prostate cancer, and in endometriosis Endocr.-Relat. Cancer 2008, 15, 665– 692

23. Brozic, P.; Lanisnik-Risner, T.; Gobec, S.Inhibitors of 17beta-hydroxysteroid dehydrogenase type 1 Curr. Med. Chem. 2008, 15, 137– 150

24. Poirier, D.Advances in development of inhibitors of 17β-hydroxysteroid dehydrogenases Anticancer Agents Med. Chem. 2009, 9, 642– 660

25. Poirier, D.17β-Hydroxysteroid dehydrogenase inhibitors: a patent review Expert. Opin. Ther. Pat. 2010, 20, 1123– 1145

26. Marchais-Oberwinkler, S.; Henn, C.; Moller, G.; Klein, T.; Negri, M.; Oster, A.; Spadaro, A.; Werth, R.; Wetzel, M.; Xu, K.; Frotscher, M.; Hartmann, R. W.; Adamski, J.17β-Hydroxysteroid dehydrogenases (17β-HSDs) as therapeutic target: protein, structures, functions, and recent progress in inhibitor development J. Steroid Biochem. Mol. Biol. 2011, 125, 66– 82

27. Poirier, D.Contribution to the development of inhibitors of 17β-hydroxysteroid dehydrogenase type 1 and 7: key tools for studying and treating estrogen-dependent diseases J. Steroid Biochem. Mol. Biol. 2011, 125, 83– 94

28. Lin, S. X.; Poirier, D.; Adamski, J.A challenge for medicinal chemistry by the 17β-hydroxysteroid dehydrogenase superfamily: an integrated biological function and inhibition study Curr. Top. Med. Chem. 2013, 13, 1164– 1171

29. Laplante, Y.; Cadot, C.; Fournier, M. C.; Poirier, D.Estradiol and estrone C-16 derivatives as inhibitors of type 1 17β-hydroxysteroid dehydrogenase: blocking of ER+ breast cancer cell proliferation induced by estrone Bioorg. Med. Chem. 2008, 16, 1849– 1860

77 30. Laplante, Y.; Rancourt, C.; Poirier, D.Relative involvement of three 17β-hydroxysteroid dehydrogenases (types 1, 7 and 12) in the formation of estradiol in various breast cancer cell lines using selective inhibitors Mol. Cell. Endocrinol. 2009, 301, 146– 153

31. Fang, H.; Tong, W.; Shi, L. M.; Blair, R.; Perkins, R.; Branham, W.; Hass, B. S.; Xie, Q.; Dial, S. L.; Moland, C. L.; Sheehan, D. M.Structure–activity relationships for a large diverse set of natural, synthetic and environmental estrogens Chem. Res. Toxicol. 2001, 14, 280– 294

32. Cadot, C.; Laplante, Y.; Kamal, F.; Luu-The, V.; Poirier, D.C6-(N,N-Butyl-methyl- heptanamide) derivatives of estrone and estradiol as inhibitors of type 1 17β-hydroxysteroid dehydrogenase: chemical synthesis and biological evaluation Bioorg. Med. Chem. 2007, 15, 714– 726

33. Mazumdar, M.; Fournier, D.; Zhu, D. W.; Cadot, C.; Poirier, D.; Lin, S. X.Binary and ternary crystal structure analyses of a novel inhibitor with 17beta-HSD type 1: a lead compound for breast cancer therapy Biochem. J. 2009, 10, 357– 366

34. Maltais, R.; Ayan, D.; Poirier, D.Crucial role of 3-bromoethyl side-chain in removing the undesirable estrogenic activity of potent 17β-hydroxysteroid dehydrogenase type 1 inhibitor 16β-(m- carbamoylbenzyl) estradiol ACS Med. Chem. Lett. 2011, 2, 678– 681

35. Ahmed, V.; Liu, Y.; Silvestro, C.; Taylor, S. D.Boronic acids as inhibitors of steroid sulfatase Bioorg. Med. Chem. 2006, 14, 8564– 8573

36. Radu, I. I.; Poirier, D.; Provencher, L.New efficient pathway for the synthesis of 3-amino-estrone Tetrahedron Lett. 2002, 43, 7616– 7619

37. Furuya, T.; Strom, A. E.; Ritter, T.Silver-mediated fluorination of functionalized aryl stannanes J. Am. Chem. Soc. 2009, 131, 1662– 1663

38. Morera, E.; Ortar, G.A palladium-catalyzed carbonylative route to primary amides Tetrahedron Lett. 1998, 39, 2835– 2838

39. Cacchi, S.; Lupi, A.Palladium-catalyzed hydroxycarbonylation of vinyl and aryl triflates: synthesis of α- β-unsaturated and aromatic carboxylic acids Tetrahedron Lett. 1992, 33, 3939– 3942

40. Poirier, D.; Chang, H. J.; Azzi, A.; Boivin, R. P.; Lin, S. X.Estrone and estradiol C-16 derivatives as inhibitors of type 1 17β-hydroxysteroid dehydrogenase Mol. Cell. Endocrinol. 2006, 27, 236– 238

41. Allan, G. M.; Lawrence, H. R.; Cornet, J.; Bubert, C.; Fischer, D. S.; Vicker, N.; Smith, A.; Tutill, H. J.; Purohit, A.; Day, J. M.; Mahon, M. F.; Reed, M. J.; Potter, B. V. L.Modification of estrone at the 6, 16, and 17 positions: novel potent inhibitors of 17β-hydroxysteroid dehydrogenase type 1 J. Med. Chem. 2006, 49, 1325– 1345

42. Skoda-Földes, R.; Kollàr, L.; Marinelli, F.; Arcadi, A.Direct and carbonylative vinylation of steroid triflates in the presence of homogeneous palladium catalysts Steroids 1994, 59, 691– 695

43. Gomez, L.; Gellibert, F.; Wagner, A.; Mioskowski, C.(Chloro-phenylthio- methylene)dimethylammonium chloride (CPMA) an efficient reagent for selective chlorination and bromation of primary alcohols Tetrahedron Lett. 2000, 41, 6049– 6052

44. Yang, D.; Zhang, C.Ruthenium-catalyzed oxidative cleavage of olefins to aldehydes J. Org. Chem. 2001, 66, 4814– 4818

45. Yasu, Y.; Koike, T.; Akita, M.Intermolecular aminotrifluoromethylation of alkenes by visible-light-driven photoredox catalysis Org. Lett. 2013, 15, 2136– 2139

46. Rousseau, C.; Christensen, B.; Bols, M.Artificial epoxidase II. Synthesis of cyclodextrin ketoesters and epoxidation of alkenes Eur. J. Org. Chem. 2005, 413–, 273 2739

47. Varghese, J. P.; Sudalai, A.; Iyer, S.Pd-catalysed regiospecific reductive ring opening of epoxides and glycidic esters Synth. Commun. 1995, 25, 2267– 2273

48. Ducan, L. J.; Reed, M. J.The role and proposed mechanism by which oestradiol 17β-hydroxysteroid dehydrogenase regulates breast tumour estrogen concentrations J. Steroid Biochem. Mol. Biol. 1995, 55, 565– 572

49. Pelletier, J. D.; Poirier, D.Synthesis and evaluation of estradiol derivatives with 16α- (bromoalkylamide), 16α-(bromoalkyl) or 16α-(bromoalkynyl) side chain as inhibitors of 17β- hydroxysteroid dehydrogenase type 1 without estrogenic activity Bioorg. Med. Chem. 1996, 4, 1617– 1628

50. Sam, K. M.; Boivin, R. P.; Tremblay, M. R.; Auger, S.; Poirier, D.C16 and C17 derivatives of estradiol as inhibitors of 17β-hydroxysteroid dehydrogenase type 1: chemical synthesis and structure–activity relationships Drug Des. Discovery 1998, 15, 157– 180

51. Matthews, J.; Celius, T.; Halgren, R.; Zacharewski, T.Differential estrogen receptor binding of estrogenic substances: a species comparison J. Steroid Biochem. Mol. Biol. 2000, 74, 223– 234

52. Kuiper, G. G.; Carlsson, B.; Grandien, K.; Enmark, E.; Haggblad, J.; Nilsson, S.; Gustafsson, J. A.Comparison of the ligand binding specificity and transcript tissues distribution of estrogen receptors α and β Endocrinology 1997, 138, 863– 870

53. Ayan, D.; Maltais, R.; Roy, J.; Poirier, D.A new nonestrogenic steroidal inhibitor of 17β-hydroxysteroid dehydrogenase type 1 blocks the estrogen-dependent breast cancer tumor growth induced by estrone Mol. Cancer Ther. 2012, 11, 2096– 2104

54. Bydal, P.; Auger, S.; Poirier, D.Inhibition of type 2 17β-hydroxysteroid dehydrogenase by estradiol derivatives bearing a lactone on the D-rings: structure–activity relationship Steroids 2004, 69, 325– 342

55. Bellavance, E.; Luu-The, V.; Poirier, D.Potent and selective steroidal inhibitors of 17β-hydroxysteroid dehydrogenase type 7, an enzyme that catalyzes the reduction of the key hormones estrone and dihydrotestosterone J. Med. Chem. 2009, 52, 7488– 7502

56. Farhane, S.; Fournier, M. A.; Poirier, D.Synthesis and evaluation of amido-deoxyestradiol derivatives as inhibitors of 17β-hydroxysteroid dehydrogenase type 12 Curr. Enzyme Inhib. 2011, 7, 134– 146

57. Marchais-Oberwinkler, S.; Wetzel, M.; Ziegler, E.; Kruchten, P.; Werth, R.; Henn, C.; Hartmann, R. W.; Frotscher, M.New drug-like hydroxyphenylnaphthol steroidomimetics as potent and selective 17β- hydroxysteroid dehydrogenase type 1 inhibitors for the treatment of estrogen-dependent diseases J. Med. Chem. 2011, 54, 534– 547

58. Poirier, D.; Dionne, P.; Auger, S.A 6β-(thiaheptanamide) derivative of estradiol as inhibitor of 17β- hydroxysteroid dehydrogenase type 1 J. Steroid Biochem. Mol. Biol. 1998, 64, 83– 90

59. Yoh, S. D.; Lee, O. S.; Park, J. H.; Kim, S. H.; Lee, Y. D.Solvent effects on the benzylation with substituted imidazoles in acetonitrile and methanol Bull. Korean Chem. Soc. 2000, 21, 653– 655

60. Ahn, B. C.Sodium iodide symporter for nuclear molecular imaging and gene therapy: from bedside to bench and back Theranostics 2012, 2, 392– 402

61. Upadhyay, G.; Singh, R.; Agarwal, G.; Mishra, S. K.; Pal, L.; Pradham, P. K.; Das, B. K.; Godbole, M. M.Functional expression of sodium iodide symporter (NIS) in human breast cancer tissue Breast Cancer Res. Treat. 2003, 77, 157– 165

62. Molecular Operating Environment (MOE) 2012.10; Chemical Computing Group Inc., 1010 Sherbrooke St. West, Suite #910, Montréal, QC, Canada H3A 2R7, 2012.

Chapitre 2

81 Résumé

En tant que dernière enzyme impliquée dans la production du principal estrogène biologiquement actif (E2), il appert que la 17β-HSD1 est une cible attrayante pour le traitement sélectif de maladies stimulées par les estrogènes. Le PBRM est un inhibiteur des plus prometteurs de cette enzyme qu’il semble affecter de manière irréversible. En fonction de cette nouvelle information, son potentiel d’inhibition a été réévalué par la mesure de ses constantes d’inhibition et par la durée de son action en cellules intactes. Cela a permis d’estimer que la resynthèse de l’enzyme pouvait prendre quelques jours.

Une variation importante dans la puissance du PBRM d’une espèce à l’autre a également été observée. L’étude de celle-ci soutient l’hypothèse que l’inhibition surviendrait par la formation d’un lien covalent avec l’His221 potentialisé par la Glu282. Le PBRM s’est également avéré être excrété par voie fécale et se retrouver principalement au niveau du tractus gastro-intestinal chez la souris.

Manuscrit en cours de rédaction

Mode of action of PBRM as 17β-HSD1 inhibitor

Alexandre Trottier1, Jenny Roy1, René Maltais1, Xavier Barbeau2, Patrick Lagüe3 and Donald Poirier1

1 - Laboratory of Medicinal Chemistry, Endocrinology and Nephrology Unit, CHU de Québec - Research Center (CHUL, T4-42) and Faculty of Medicine, Laval University, Québec City, Québec G1V 4G2, Canada

2 – Département de Chimie, Institut de Biologie Intégrative et Des Systèmes (IBIS), and Centre de Recherche sur la Fonction, la Structure et l’Ingénierie des Protéines (PROTEO), Université Laval, Québec City, Québec G1V 4G2, Canada

3 - Département de Biochimie Microbiologie et Bio-informatique, Institut de Biologie Intégrative et des Systèmes (IBIS), and Centre de Recherche sur la Fonction, la Structure et l’Ingénierie des Protéines (PROTEO), Université Laval, Québec City, Québec G1V 4G2, Canada

83 Abstract

17β-Hydroxysteroid dehydrogenase type 1 (17β-HSD1) appears to be needed for estradiol biosynthesis, the main bioactive endogenous estrogen, and is thus an interesting target to treat estrogen-dependent diseases. A lead compound from our laboratory named PBRM has recently been reported to be a non-estrogenic, highly selective and irreversible 17β-HSD1 inhibitor. We here reported a thorough evaluation of its potency by measuring inhibition constants and turnover rate in vitro. The mechanism of enzyme inactivation has then been studied by both interspecies comparison and in sillico analysis. Pharmacokinetic of PBRM in mice has also been evaluated.

Introduction

Estrogens mediate both important physiological processes and initiation of illness affecting millions of men and women around the world1–5. Targeting specifically the estrogenic stimulation of important diseases is a real challenge as well as an interesting option of treatment. Two different strategies have been used successfully in clinic so far to interfere with the effect of these hormones: blocking their receptors (with full or tissue selective antagonists) or blocking their biosynthesis (with GnRH agonists or aromatase inhibitors).

Selective estrogen-receptor modulators (SERM) and aromatase inhibitors are considered as the best first line treatments for estrogen-dependent breast cancers, but the latter (enzyme inhibitor) seems to be a slightly suitable option for post-menopausal women according to recent studies6,7. Even if both are efficient drugs, they have some secondary effects and are not optimal treatment for diseases like endometriosis, a pathology in which endometrial tissue is found outside of the uterine cavity 8,9. This condition affect 5 to 10% of women and causes them infertility, abdominal pain and even internal bleedings in some cases10,11.

17β-Hydroxysteroid dehydrogenase type 1 (17β-HSD1) inhibitors are promising compounds to help fighting such gynecological condition and are possible complementary option for breast cancer treatment12–14. This enzyme is first responsible of the activation of estrone (E1) into estradiol (E2), a potent estrogen. It is also involved in the transformation of dehydroepiandrosterone (DHEA) into 5-androstene-3β,17β-diol (5-diol), which is suspected to act as the principal estrogen after the menopause15. The biosynthesis of 5-diol is not affected by aromatase inhibitor so combination with a 17β-HSD1 inhibitor might have complementary effect for treatment of postmenopausal women cancer.

Despite decades of research, no 17β-HSD1 inhibitor has yet reached the clinical trials. Among the few showing an adequate inhibition potential16 the lack of specificity notably over estrogen receptor (ER) activation has been

an important drawback especially for steroidal compounds12,17–19. Reaching an acceptable selectivity over other types of 17β-HSD has also been a major problem of some inhibitors. At least 15 enzymes belong to the 17β- HSD superfamily, they are reductive (types 1, 3, 5, 7, 12 and 15) or oxidative (types 2, 4, 6, 8, 9, 10, and 14) and cause an activation or a deactivation of steroidal hormones, while most of them are also involved in nonsteroidal synthesis as cholesterol, retinoic acid or long chain fatty acid 20,21. Some of these enzymes, such as 17β-HSD2 that is responsible of the opposite action of 17β-HSD1 in estrogen metabolism, must not be inhibited by a 17β-HSD1 inhibitor.

The poor interspecies homology of 17β-HSD1 even in the catalytic site is another challenge to the development of efficient inhibitors. Consequently, it have been clearly demonstrated that the effect of inhibitors is highly dependent of the species used for the test22. However, a handful of 17β-HSD1 inhibitors were recently reported to show interesting features and as such are interesting candidates for further preclinical studies23–28. One of them, an estradiol derivative named PBRM (Fig. 1) was developed in our laboratory and showed interesting efficiency in both cancer cells and mice29,30. Moreover further investigation revealed a highly selective inhibition and covalent binding to 17β-HSD128. Irreversible inhibition could be advantageous to the potency aspect but also raise fear about possible damageable reaction with random proteins31–33.

The measure of concentration needed to inhibit half of an enzyme activity (IC50), as used previously, is not the most insightful value to evaluate the potency of an irreversible inhibitor. The efficacy of PBRM is then measured in vitro by determining affinity and inactivation values as well as estimating turnover of the enzyme. Its capacity to interact with 17β-HSD1 of other species is also assessed both in vitro and in sillico. The development of PBRM of course depends on its pharmacokinetic profile, which was the object of distribution and excretion experiments conducted in mice and that are reported in the present article.

PBRM (X=Br; Y=H) PIRM (X=I; Y=H) [3H]-PBRM (X=Br; Y=3H)

Figure 1. Structure of 17β-HSD1 inhibitors PBRM, PIRM and [3H]-PBRM (radiolabeled PBRM).

85 Results

Evaluation of PBRM inhibition potential

Measure of potency

28,30 The PBRM potency as inhibitor of 17β-HSD1 has been only evaluated by IC50 values until now . Such measure is not appropriate to evaluate an irreversible inhibitor for which the effect is mostly time dependent contrary to the exclusively concentration dependent effect of reversible ones. As the first non-estrogenic and irreversible 17β-HSD1 inhibitor of interest to date, it is primordial to measure its potency in order to provide a point of comparison with eventual inhibitors that would be discovered in the future. In order to do that, the affinity constant (Ki) and the maximal enzyme inactivation rate (kinact) values have been determined through time dependent inhibition assay of purified 17β-HSD1 by PBRM and PIRM at many concentrations. PBRM, and its iodo analogue PIRM, were the best we obtained until now and both were tested to confirm the proposed mechanism of inhibition. As an iodide is more reactive (better leaving group) toward a nucleophilic amino acid than bromide, PIRM is expected to inactivate the enzyme faster and to have a value of kinact consequently higher than PBRM while affinity should be comparable for both.

A Q0 A’ Q0 Q0 Q0 Q0 Q0 Q0 Q0 Q0

(% of control) of (% Q0 Enzyme Activity Enzyme Ki= 381 nM -1 kinact= 0.084 min

7LPH PLQ .

B B’ Q0 Q0 Q0 Q0 Q0 Q0 Q0 Q0 (% of control) of (% Enzyme Activity Enzyme Ki= 394 nM -1 kinact= 0.104 min

7LPH PLQ

Figure 2. Kitz-Wilson analysis of 17β-HSD1 inhibition by PBRM (A) and PIRM (B). Purified enzyme activity was evaluated (1 h inhibition time) after a defined pre-incubation time with different inhibitor concentrations and after the inhibitor was washed out. Slope (Kobs) from each experiment (left panel) was calculated and then plotted in a second graph as 1/ Kobs on 1/[inhibitor] (right panel). These graphes show the results of one assay representative of two independent assays conducted in triplicate. Values of Ki and kinact are expressed as the mean ± standard deviation of these two assays.

The inhibition of irreversible inhibitors is traditionally evaluated by a Kitz-Wilson analysis for which the enzyme is pre-incubated with various concentrations of the compound and, at time intervals, sample of this mixture is highly diluted in a reaction medium containing substrate to measure the remaining enzyme activity. If an irreversible inhibition occurs, time-dependent inhibition of substrate transformation is observed as it is the case in Fig. 2A and 2B (left panels), for PBRM and PIRM, respectively. The slope for each concentration is then plotted again to determine Ki and kinact values.

From these data, similar Ki values of 381 nM and 394 nM were calculated for PBRM and PIRM respectively (Fig.

-1 1C and 1D, right panels). However, they differ from their inactivation potency (kinact) with 0.084 min and 0.104 min-1 as values for PBRM and PIRM, respectively. These results are coherent with the hypothesis that inhibition is at least partly due to a reaction of the C3 halogeno-ethyl group with the enzyme, which substitution reaction is known to be easier with iodide than bromide. Thus, the nature of the halogen only affects the reactivity (kinact) and not the affinity (Ki) component of the inhibition. This result can also be used to explain the lack of inhibition we observed with chloride and hydroxide analogue of PBRM, which are worst leaving groups 28.

Recovery of 17β-HSD1 activity

As far as we know, the turnover of 17β-HSD1, the time needed to synthetize the enzyme, has never been reported in any way. This rate is susceptible to be highly different depending of the kind of cells, the organ and/or the species where the enzyme is expressed. To have an estimation of the time needed for this process we choose the well-known T47D and JEG-3 cells. These two cell lines, immortalized from placenta and breast cancers, are representative of high and low expression profiles, respectively. Cells were cultured to confluence in plates and were treated with 1 µM of PBRM for 24 h in culture medium deprived in steroids before being thoroughly washed. Complete medium were then added and the enzyme activity measured at different times after inhibitor removal.

A 100 B 100

75 75

50 50

Vehicle Vehicle 25 PBRM pre-treated 25

Enzyme (%Activity of control) Enzyme PBRM pre-treated Enzyme Activity (%Activity of control) Enzyme

0 0 0 2 4 6 8 10 12 0 2 4 6 8 10 12 Time (d) Time (d) Figure 3. 17β-HSD1 activity recovery of T47D (A) and JEG-3 (B) cells after a 24 h treatment with 1 µM PBRM (black spot) or with the vehicle only (white square). Cells were washed and enzyme activity measured after different times. The results of one assay representative of two conducted in triplicate is shown here.

Just after the inhibitor was removed, at time 0, the transformation of E1 to E2 is negligible and not significantly different from 0 with 7.7% and 9.7% (Fig. 3A and 3B respectively). Then the enzyme activity is gradually

recovered reaching half of the maximal enzyme activity in 3.5 days for T47D cells and 2.1 days for JEG-3 cells. The recovery appears to reach a plateau of 82% for T47D and 93% for JEG-3 after 6 days. Thus the capacity of cells to produce E2 is recovered at the levels of untreated cells for JEG-3 cells, but not for T47D cells. This could be explained by the fact that cells were cultured at a confluent state for an unusually long period in these experiments. There is also noticeably faster recovery of enzyme activity for JEG-3 than T47D cells, which is coherent with the expression level of 17β-HSD1 in each cell line.

The same experiment has also been conducted in hormones deprived medium and with non-confluent cells (data not shown). The recovery of enzyme activity appeared to be slightly slower in these last conditions than with confluent cells using complete medium. Considering that the regain of enzyme activity after the treatment with an irreversible inhibitor such as PBRM needed the synthesis of new enzyme molecules, it is therefore possible to evaluate the turnover rate of 17β-HSD1. Thus, 6 days were needed for a full recovery of 17β-HSD1 activity in both cell lines while half recovery is reached in 2.1 and 3.5 days for JEG-3 and T47D cells, respectively.

Inhibition of orthologous enzymes

In order to evaluate the interspecies inhibition of the inhibitor PBRM, ovaries of mice, pig and monkeys were homogenized and the 17β-HSD1 partially purified by centrifugation (Fig. 4). Homogenates were then treated with PBRM to measure the inhibition of E2 production. This inhibition was then compared with the one measured with highly purified 17β-HSD1 in the same conditions.

As expected, the inhibition of 17β-HSD1 by PBRM differs between species. With only 13 % and 37 % inhibition of mouse 17β-HSD1 at 1 and 10 µM, respectively. Inhibition in pig homogenized ovaries is also weaker than for the human enzyme with 32% and 76% at the same concentrations. Monkeys enzyme though were inhibited in a comparable order than its human counterpart with respectively 71% and 90% compared to 90% and 100% at 1 µM and 10 µM.

Molecular modelling

If differences between species inhibition were expected, the explanation of these divergences could bring important insight about PBRM inhibition mechanism. Comparison of the respective sequences and modeling studies were conducted to address this point (Fig. 5).

Figure 5. Structural comparison of human 17β-HSD1 in green with monkey (purple), pig (cyan) and mouse (pink) (left to right respectively).

Important differences between species were observed in amino acids forming the catalytic site of 17β-HSD1. From the docking studies, His221 and Glu282 in human and their orthologous counterpart were identified as the more significant (Table 1). The histidine is preserved in monkey and pig but is mutated to glycine in mouse. As this amino acid cannot form a covalent link with PBRM bromide moiety, this dissimilarity may explain the poor inhibition observed with PBRM in rodent as well as the lack of time-dependent inhibition (data not shown). Even tough, the other three species were not equally prone to PBRM inhibition what could be related to variance in the second gating residues. Glutamate in position 282 is conserved in mice and human but is replaced by asparagine and arginine in monkey and pig, respectively.

Docking and inhibition assays are thus coherent with the hypothesis of histidine and glutamic acid involvement in the mechanism of PBRM. These results tend to confirm the hypothesis formulated after our first docking studies 28.

91 Human Monkey Pig Mouse Position His His His Gly 221 (AA H residue)

Position Glu Asn Arg Glu 282 (AA residue)

Table 1. Amino acids in key position for inactivation by PBRM

Pharmacokinetic of PBRM

Distribution of PBRM

A long plasmatic half-life (T1/2) of PBRM had already been measured, but little more was known about its exact kinetic30. Distribution was first evaluated after subcutaneous administration of [3H]-PBRM to mice that were sacrificed by group of three for each time courses. The 3H atom was introduced at position C17α of PBRM, a position that should be not reactive considering the bulky environment provided by the C18-methyl and C16- benzylamide chain which protect the 17-alcohol from a potential oxidation, glucuronidation or sulfatation reaction. Their organ were collected and dissolved in a solution of KOH. Samples were then bleached with H2O2 and the radioactivity quantified.

Figure 6. Tissue distribution of [3H]-PBRM in the digestive track (A) and other tissues (B) of mice. Three animals were sacrificed for each time after subcutaneous injection then organs were collected, dissolved and radioactivity quantified individually. The results are the mean value for three mice ± standard error.

93 Most tissues appeared to reach maximal level of [3H]-PBRM 6 h after injection. The radioactivity was mostly observed in the digestive tract for the first 6 h after injection. Mice sacrificed 12 h after injection shower lower level in intestine and colon than those sacrificed after 6 or 24 h. The fact that later were administrated in the morning and the former in the evening could explain this deviation. Moreover, both tissue and content of each stomach, intestine and colon appeared to have high radioactivity level (Fig. 6A). In fact, intestine was the only organ where the presence of inhibitor was still detected after 3 days. This observation point out to a strong fecal excretion of PBRM.

Thus PBRM did not appear to accumulate in any organ. As a matter of fact, the inhibitor was nearly detectable or not at all in most of tissues outside of the digestive system (Fig. 6B). Overall, the lack of inhibitor accumulation in any organ is reassuring in regard of safety concerns raised by a covalent mechanism of inhibition31.

Excretion of PBRM

In order to have a better idea of the excretion way of PBRM, a mass bilan assay was conducted in mice. Inhibitor was administrated by intravenous injection to mice kept in metabolic cages. Feces and urine were collected regularly and cages washed each time. Radioactivity was then quantified in feces, urine and wash liquid.

The PBRM were mostly recuperated in feces with 93% against 7% in urine (Fig. 7). The radioactivity recovered in washing liquid account for less than 0.5% of the injected radioactivity. 68% of the elimination occurred in the first 24 h after what it continued at a lower rate until 5 days after the injection when it reached a plateau. The excretion has also been measured in a third mouse at higher dose (data not shown) showing a similar profile.

Figure 7. Mass balance of [3H]-PBRM in mice after subcutaneous administration. Feces (red) and urine (blue) were recuperated while cages were washed at determined times. The results are the mean ± standard deviation for two mice of the cumulative radioactivity measured in excrements in times in terms of injected dose.

Discussion

The present work had permitted to deepen our knowledge of both potential and mechanism of 17β-HSD1 inhibition by PBRM. The comparable Ki of PBRM (381 nM) and PIRM (394 nM) combined with the slightly better inactivation value for the later (0.084 min-1 against 0.104 min-1 respectively) measured are coherent with the suspected formation of a covalent link by substitution reaction of the 2-halogenoethyl part of both inhibitors. Iodine is generally known to be more reactive in this condition. As the first of its kind, the inhibition values reported for PBRM will serve as reference for any eventual 17β-HSD1 inhibitors.

In a competitive point of view, the PBRM appeared to have an affinity slightly lower than natural substrate and

35, 18 previous inhibitors . However, for irreversible inhibitors kinact is more meaningfull value than Ki to evaluate inhibition potential. For instance, an inhibitor with comparable Ki for two enzymes could be selective for the one

33 irreversibly inhibited (kinact ≠ 0) if the other is strictly reversibly inhibited (kinact = 0) . This highlights the fact that irreversible inhibition is mostly a matter of time. In fact, the potential of inhibition appeared to be as much important as duration of its effect. This one is mostly depending on time required to resynthesize the enzyme in each tissues.

95 To assess that point, recovery of the 17β-HSD1 was measured after the cells being washed following PBRM treatment. T47D and JEG-3 cell lines were used as representatives of breast and placenta respectively, two organs known for their expression of 17β-HSD1. The former required as much as 3.5 days to recover half of its capacity to transform E1 to E2 while the later needed 2.1 days. These values are comparable to those reported for irreversible inhibition of aromatase, another enzyme involved in estrogen production34.

The recovery time observed could be considererd as a descent estimate of 17β-HSD1 turnover rate. This assumption supposes that PBRM binds the enzyme both irreversibly and covalently in such a way that synthesis of new enzyme molecules is the only way enzymatic activity can be recoverered by cells. Also, the cell proliferation, the possible remaining free inhibitor at time 0 and the effect of other enzymes are considered to have neglectible effect. All those supposition are fairly probable considering the conditions used. However a rigourous determination of 17β-HSD1 turnover rate would have to address all those conscerns. Without such data, we find the estimate provided here fully satisfying.

Anyway, the recovery took days to happen in both cell lines cultured in complete or in steroid and growth factor deprived medium. A long lasting effect could then reasonably be expected which could be a great and unique advantage over any other 17β-HSD1 inhibitors. Low dose and/or low frequency administration of PBRM could therefore be efficient and caused less undesirable effect than comparable reversible 17-HSD1 inhibitor32.

Inhibitions of 17β-HSD1 present in other species also provided insightful new information about PBRM mechanism. The differences between species were attended considering works of Adamski group22 who shown similar profile for an E2 derivative substituted in position 15 instead of 16 for PBRM. However, our experiences of docking pointed out amino acids around position 3 to explain the interspecies divergences. The apparent necessity of His221 for irreversible inhibition is coherent with its reaction with PBRM to inactivate 17β-HSD1 highlighted earlier. Results with orthologous enzymes also tend to confirm the hypothesis that Glu282 potentiate the reaction28 and can explain the variation of inhibition between human, monkey and pig. This also highly support the idea that 2-bromoethyl group of PBRM needs a particular environment for the formation of a covalent bond which is unlikely to be common through human proteins. Thus, the reaction is probably highly selective to 17β-HSD1 and the risk of adverse reaction by random covalent binding should be consequently very low.

Until now, no other enzyme seemed to be significantly inhibited by PBRM, a good omen considering the traditional fears and concerns about covalent inhibitors specificity and safety. Additionally, no adverse effect related to the inhibitor was noticed in experiments conducted in rodent. Among those reported here, distribution of tritiated PBRM showed no particular accumulation in any organ. However, the levels found in gastrointestinal tract were fairly high and for couple of days as some radioactivity were still detected three days after injection. These results are coherent with those of excretion, which appeared to be mainly fecal (93% found in feces)

against 7% in urine. Elimination appeared to be fairly slow as about one third of the dose was still in the animals 24 h after injection and took about 4 days before everything get out.

Then the slow fecal excretion may explain the high PBRM level found in gastrointestinal tract for a long time. This pointing out a very slow metabolism of the inhibitor that may also explain the long plasmatic half-life of cold PBRM as previously reported. However, the metabolism has not yet been directly addressed. Appropriate experiments should be conducted soon to enlighten this point as well as the oral availability of PBRM that have been observed in a preliminary test (data not shown).

Further distribution study will also have to consider the effect of the excretion. Mice of 12 h-time point were injected in the evening while others were injected in the morning. The goal was to avoid nightly sacrifice but it seemed to affect distribution as less PBRM were founded in intestine and colon of these mice than in those sacrificed 6 h and 24 h after injection. Difference between night and days were also observed in feces production but concentrations in excrements were not seemed much affected by the time of the day chosen for injection.

Conclusion

Overall, the pharmacokinetic profile of PBRM is favorable. Distribution showed no accumulation and excretion took days, causing a long staying time which is favorable to the efficiency of an irreversible inhibitor. Besides of the long time needed for breast and placenta cells to recover 17β-HSD1 activity after treatment with PBRM, our results highlight the possibility of efficient treatment at low dose and low frequency. The strong potency of enzyme inactivation measured here is also favorable for such effect. Moreover, modeling and inhibition assays of orthologous 17β-HSD1 showed encouraging insights about PBRM mechanism suggesting a very specific action.

Acknowledgment

We are thankful to Canadian Institutes of Health Research for supporting this research. We would like to thank Dr. Sheng-Xiang Lin (CHU de Québec – Research Center) for providing purified human 17β-HSD1 and Abattoir Bolduc of Buckland (Québec, Canada) for providing the pig ovaries.

97 Method

Measure of enzymes activity

Enzymatic assays were performed by quantifying the transformation of radiolabeled substrate, namely [14C]-E1 in ethanolic solution (American Radiolabeled Chemicals, Inc., St. Louis, MO, USA). Steroids (E1 and E2) were first extracted from aqueous medium with diethyl ether. The organic phase was evaporated and the residues dissolved in dichloromethane to be dropped on silica gel thin layer chromatography (TLC) plates (EMD Chemicals Inc., Gibbstown, NJ, USA). Radiolabeled substrate and product were then separated by TLC with toluene/acetone (4:1) solvent system and finally quantified using Storm 860 system (Molecular Dynamics, Sunnyvale, CA, USA).

Measure of potency

Purified 17β-HSD1 kindly provided by Dr. Sheng-Xiang Lin (CHU de Québec - Research Center)35 was used for inactivation/competition assays. An enzyme solution was diluted in physiological buffer (100 mM Tris, 20% glycerol, 0.2 mg/mL BSA, 2 mM EDTA) containing 1 mM of NADPH as cofactor and was treated with PBRM (compound 23b in Maltais et al. 2014)) solubilized in DMSO to reach the accurate concentration. The mixture was then pre-incubated at 37°C with shaking before a simultaneous dilution 1:100 with physiological buffer in duplicate and addition of radiolabeled substrate to a final concentration of 10 nM [14C]-E1. Transformation of substrate was stopped after 45 min of incubation at 37°C with shaking by cooling down on ice the enzyme medium and adding equal volume of diethyl ether for further extractions, separation by TLC and quantification of radiolabeled steroids (E1 and E2) as described above.

36 Ki and kinact were measured according to Kitz-Wilson analysis . The percentage of transformation and remaining activity were calculated as % transformation = 100 x [14C]-E2 ⁄ ([14C]-E1 + [14C]-E2) and remaining enzyme activity = 100 x (% transformation / % transformation (time = 0)). Remaining enzyme activities were plotted against time for each inhibitor concentration in a semi-log graph. The slope of each curve was calculated using

GraphPad Prism 5 as the Kobs. 1/Kobs were re-plotted against 1/inhibitor concentration. The slope and the y- intercept of this second plot were calculated to obtain kinact and Ki as 1/y-intercept and slope x kinact.

Turnover rate in cells

T47D and JEG-3 cells were cultured in complete RPMI medium (supplemented with 10% fetal bovine serum (FBS), 2 nM glutamax, 100 IU/mL peniciline, 100 mg/mL streptomycin and 1 nM E2). They were seeded in 24- wells culture dishes at 1 mL/well for two days to reach confluence. 17β-HSD1 was inactivated by gently washing the cells with RPMI to remove complete medium and then were treated overnight with PBRM (1 µM) or equivalent DMSO control (<0.1% DMSO concentration) in 1 mL of steroid-deprived medium which contain 50 ng/mL insulin and 5% charcoal-stripped FBS instead of 10% FBS and no added insulin for complete medium.

Inhibitor was ultimately removed by thorough washes of the cells with RPMI and replaced by complete RPMI medium renewed every two days. The recovery of 17β-HSD1 was evaluated at various times by washing again the cells and incubating them with 60 nM [14C]-E1 in 1 mL protocol medium. Substrate transformation was stopped 3 and 1 h later for T47D and JEG-3 cells, respectively, by removing medium for further separation and quantification of radiolabeled steroids (E1 and E2). Both PBRM- and only DMSO- treated cells were tested in triplicate for each testing time.

Inhibition of orthologous 17β-HSD1

Monkey, pig and mice ovaries were homogenized with Polytron and enzyme was separated by two successive centrifugations of supernatant (12500 and 100000 x g) in phosphate buffer (20 mM KH2PO4, 0.25 M sucrose, 1 mM EDTA, pH 7.4, 20% glycerine) containing protease inhibitors cocktail (Roche Diagnostics, Laval, QC, Canada).

Homogenates were diluted in 950 µL of phosphate buffer containing 10 µM NADPH and were treated with a DMSO solution of PBRM (<0.1% final DMSO concentration) for 30 minutes at 37°C. Then inhibition assay was started by adding [14C]-E1 to final concentration of 10 nM [14C]-E1 in 1 mL of reactive medium. Substrate transformation was stopped 1 h later by removing medium. Radiolabeled steroids (E1 and E2) were immediately extracted on ice with diethyl ether for further separation and quantification of radiolabeled steroids.

In silico comparison of orthologous inhibition

Protein sequences of 17-HSD1 human (NP_000404), marmoset (AAG01115), pig (NP_001121944),

99 mouse (NP_034605) and rat (NP_036983) were taken from the protein database of NCBI. Multiple alignment was performed using t-coffee 10.00.r1613 and was used with MOE (www.chemcomp.com) to build the homology modeling for each species with default parameters using PDB 3HB4 as template. The best models based on packing energy were evaluated with MolProbity web interface. The one with the best score for each specie was chosen for comparison with the others. Presentation of the sequence alignment was generated using TeXShade 37 and structural representation were generated with Pymol 1.5.0.1. Residues numbering is on the right of each sequence.

Chemical synthesis of PBRM, PIRM and [3H]-PBRM

Reagents, solvents and apparatus: Non radiochemical reagents were purchased from Sigma-Aldrich Canada

Ltd. (Oakville, ON, Canada). NaBT4 (100 mCi in 2.5 mL of MeOH, 10.9 Ci/mmol) was purchased from Perkin Elmer (Boston, MA, USA). Anhydrous dichloromethane (DCM) was obtained from Sigma-Aldrich. Ethyl acetate (EtOAc), hexanes and methanol (MeOH) were purchased from Fisher Scientific (Montréal, QC, Canada). TLC and flash-column chromatography were performed on 0.20-mm silica gel 60 F254 plates (E. Merck; Darmstadt, Germany) and with 230-400 mesh ASTM silica gel 60 (Silicyle, Québec, QC, Canada), respectively. Nuclear magnetic resonance (NMR) spectra were recorded at 400 MHz for 1H and 100.6 MHz for 13C on a Bruker Avance 400 digital spectrometer (Billerica, MA, USA). The chemical shifts (δ) were expressed in ppm and referenced to acetone (2.06 and 29.24 ppm, for 1H and 13C NMR, respectively). Low-resolution mass spectra (LRMS) were recorded on a Shimadzu apparatus (Kyoto, Japan) equipped with a turbo ion-spray source.

17β-HSD1 inhibitors PBRM and PIRM were synthesized as previously published 28.

[3H]-PBRM was synthesized from PBRM following the sequence of two reaction steps (oxidation and reduction) reported below.

Oxidation:To a solution of PBRM (70 mg, 0.14 mmol) in DCM (3 mL) was added N-methylmorpholine-N-oxide (25 mg, 0.21 mmol), molecular sieves (70 mg) and tetrapropylammonium-perruthenate (7 mg, 0.02 mmol). The solution was stirred at 0°C for 1 h and then 3 h at room temperature. The resulting solution was directly purified by flash chromatography using DCM/MeOH as eluent to give 36 mg (51 %) of oxidated PBRM (3-{[(16β)-3-(2-

1 bromoethyl)-17-oxoestra-1(10),2,4-trien-16-yl]methyl}benzamide). H NMR: δH (acetone-d6) 0.77 (3 H, s), 1.26 – 1.61 (6 H, m), 1.78 – 2.09 (3 H, m), 2.22 – 2.31 (1 H, m), 2.36 – 2.45 (1 H, m), 2.52 (1 H, ddt, J 9.8, 7.8, 4.9), 2.75 (1 H, dd, J 13.6, 10.0), 2.83 (2 H, t, J 7.6), 3.08 (2 H, t, J 7.5), 3.19 (1 H, m), 3.63 (2 H, t, J 7.5), 6.68 (1 H, s), 6.97 (1 H, s), 7.03 (1 H, d, J 8.0), 7.23 (1 H, d, J 8.0), 7.35 – 7.46 (2 H, m), 7.48 (1 H, s), 7.81 (1 H, dt,

100

13 J 7.2, 1.5), 7.86 (1 H, s). C NMR: δC (acetone-d6) 14.1, 26.4, 27.4, 28.5, 30.0, 32.9, 34.2, 38.0, 38.5, 39.5, 45.3, 49.0, 49.6, 51.6, 126.2, 126.2, 126.9, 128.9, 129.1, 130.0, 132.8, 135.4, 137.2, 137.3, 139.1, 141.4, 168.9,

+ 221.1. LRMS (APCI) for C28H33O2NBr [M + H] : 494.2 and 496.2 m/z. HPLC purity of 95.6 % (retention time = 18.7 min). The HPLC purity was determined using a RP-C18 column (Altima, HP, C18-AQ, 4.6 mm x 250 mm,

5 µm), a gradient of solvents, from MeOH/H2O (70:30) to 100% MeOH over 30 min, and a PDA detector at a maximum wavelength absorption of 190 nm.Reduction: To a solution of oxidated PBRM (10 mg, 0.02 mmol) in

MeOH (3 mL) was added NaBT4 (50 µL, 2 mCi) at room temperature under argon atmosphere. The solution was stirred for 1 h and then NaBH4 (4 mg, 0.1 mmol) was added to complete the reaction. The reaction was monitored by TLC until completion. The resulting solution was poured into water (75 mL) and extracted two times with EtOAc (15 mL). The organic phase was dried over sodium sulfate, filtered and evaporated under reduce pressure. The residu was characterized by TLC (DCM/MeOH; 95:5) as a single spot corresponding to PBRM (same retention time for both tritiated and untritiated PBRM). The mixture of PBRM and [3H]-PBRM was conserved in ethanol (150 µL) at -20°C and used as such for biodistribution experiments. A specific activity of 19.2 mCi/mmol was measured giving a radiochemical yield of 76%.

Measurement of radioactivity

Measurement of [3H]-PBRM was conducted by liquid scintillation. Samples were disposed in appropriate polypropylene vials and 10 mL of scintillation cocktail (BCS Biodegradable Couting Scintillant, Amersham Biosciences) was added. Radioactivity was quantified using Wellac 1411 scintillation counter.

Animals

All animals were acclimatized to the environmental conditions (temperature, 22 ± 3°C; humidity, 50 ± 20%; 12- h light/dark cycles, lights on at 07:15 h) for at least 3 days before starting the experiment. The animals were allowed free access to water and a certified commercial rodent food (Rodent Diet #T.2018.15, Harlan Teklad). The experiments with animals were conducted in an animal facility approved by the Canadian Council on Animal Care (CCAC) and the Association for Assessment and Accreditation of Laboratory Animal Care. The study was conducted in accordance with the CCAC Guide for Care and Use of Experimental Animals. Institutional approval was obtained.

101

Biodistribution in mice

Female BALB/c mice weighing approximately 20 g were obtained from Charles River, Inc. and were housed 3 to 5 per cage. 14 µCi/100 µg of [3H]-PBRM in 100 µL ethanol/propylene glycol (8:92) was administrated by subcutaneous injection to mice. Animals were sacrificed 3, 6, 12, 24, 48 or 72 h after administration. Blood was preserved into Microvette potassium-EDTA-coated tube (Sarstedt, Aktiegesellchaft & Co., Montreal, Canada) and thereafter centrifuged at 3200 rpm for 10 minutes at 4°C. Other tissues were then collected and dissolved with KOH 10 N as well as the remaining carcass.

100 µL samples of dissolved tissues were bleached with 50 µL H2O2 30% for 1 h at 37°C in scintillation vials. Each was then aerated and radioactivity was quantified 48 h later when chemiluminescence caused by KOH was reduced to background level.

Mass bilan in mice

Mice were home individually in metabolic cages. After at least 48 h of acclimation, a 10 µCi/ 72 µg dose of [3H]- PBRM in ethanol/propylene glycol/saline (8:46:46) vehicle was injected intravenously (caudal vein). The animals stayed in the metabolic cages for one week during which urine and feces were collected at different times (0.5, 1, 2, 3, 4, 5, 6 and 7 days). After each collection, the cage were thoroughly washed with ethanol and water successively. Samples of urine and wash liquid were counted directly by liquid scintillation counting as describe above. Feces were beforehand diluted and bleach in sodium hypochlorite 3% solution. Mice were also sacrifice and their tissue collected, dissolved and remaining radioactivity quantified as done for biodistribution experiment.

References

(1) Deroo, B. J. Estrogen Receptors and Human Disease. J. Clin. Invest. 2006, 116, 561–570. (2) Henderson, B. E.; Feigelson, H. S. Hormonal Carcinogenesis. Carcinogenesis 2000, 21, 427–433. (3) Hernández-Bello, R.; Nava-Castro, K.; Muñiz-Hernández, S.; Nava-Luna, P.; Trejo-Sánchez, I.; Tiempos-Guzmán, N.; Mendoza-Rodríguez, Y.; Morales-Montor, J. Beyond the Reproductive Effect of Sex Steroids: Their Role During Immunity to Helminth Parasite Infections. Mini Rev. Med. Chem. 2012, 12, 1071–1080. (4) Hogan, A. M.; Collins, D.; Baird, A. W.; Winter, D. C. Estrogen and Its Role in Gastrointestinal Health and Disease. Int. J. Colorectal Dis. 2009, 24, 1367–1375.

102

(5) Morris, J. A.; Jordan, C. L.; Breedlove, S. M. Sexual Differentiation of the Vertebrate Nervous System. Nat. Neurosci. 2004, 7, 1034–1039. (6) Burstein, H. J.; Prestrud, A. A.; Seidenfeld, J.; Anderson, H.; Buchholz, T. A.; Davidson, N. E.; Gelmon, K. E.; Giordano, S. H.; Hudis, C. A.; Malin, J.; Mamounas, E. P.; Rowden, D.; Solky, A. J.; Sowers, M. R.; Stearns, V.; Winer, E. P.; Somerfield, M. R.; Griggs, J. J.; American Society of Clinical Oncology. American Society of Clinical Oncology Clinical Practice Guideline: Update on Adjuvant Endocrine Therapy for Women with Hormone Receptor-positive Breast Cancer. J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 2010, 28, 3784–3796. (7) Chlebowski, R. T. Current Concepts in Breast Cancer Chemoprevention. Pol. Arch. Med. Wewnętrznej 2014, 124, 191–199. (8) Soares, S. R.; Martínez-Varea, A.; Hidalgo-Mora, J. J.; Pellicer, A. Pharmacologic Therapies in Endometriosis: a Systematic Review. Fertil. Steril. 2012, 98, 529–555. (9) Dunn, J.; Gotay, C. Adherence Rates and Correlates in Long-term Hormonal Therapy. In Vitamins & Hormones; Elsevier, 2013; Vol. 93, pp. 353–375. (10) De Graaff, A. A.; D’Hooghe, T. M.; Dunselman, G. A. J.; Dirksen, C. D.; Hummelshoj, L.; WERF EndoCost Consortium; Simoens, S. The Significant Effect of Endometriosis on Physical, Mental and Social Wellbeing: Results from an International Cross-sectional Survey. Hum. Reprod. Oxf. Engl. 2013, 28, 2677–2685. (11) Eskenazi, B.; Warner, M. L. Epidemiology of Endometriosis. Obstet. Gynecol. Clin. North Am. 1997, 24, 235–258. (12) Marchais-Oberwinkler, S.; Henn, C.; Möller, G.; Klein, T.; Negri, M.; Oster, A.; Spadaro, A.; Werth, R.; Wetzel, M.; Xu, K.; Frotscher, M.; Hartmann, R. W.; Adamski, J. 17β-Hydroxysteroid Dehydrogenases (17β-HSDs) as Therapeutic Targets: Protein Structures, Functions, and Recent Progress in Inhibitor Development. J. Steroid Biochem. Mol. Biol. 2011, 125, 66–82. (13) Poirier, D. 17beta-Hydroxysteroid Dehydrogenase Inhibitors: a Patent Review. Expert Opin. Ther. Patents 2010, 20, 1123–1145. (14) Purohit, A.; Tutill, H. J.; Day, J. M.; Chander, S. K.; Lawrence, H. R.; Allan, G. M.; Fischer, D. S.; Vicker, N.; Newman, S. P.; Potter, B. V. L.; Reed, M. J. The Regulation and Inhibition of 17beta- hydroxysteroid Dehydrogenase in Breast Cancer. Mol. Cell. Endocrinol. 2006, 248, 199–203. (15) Simard, J.; Vincent, A.; Duchesne, R.; Labrie, F. Full Oestrogenic Activity of C19-delta 5 Adrenal Steroids in Rat Pituitary Lactotrophs and Somatotrophs. Mol. Cell. Endocrinol. 1988, 55, 233–242. (16) Lin, S.-X.; Poirier, D.; Adamski, J. A Challenge for Medicinal Chemistry by the 17β-hydroxysteroid Dehydrogenase Superfamily: An Integrated Biological Function and Inhibition Study. Curr. Top. Med. Chem. 2013, 13, 1164–1171. (17) Brozic, P.; Lanisnik Risner, T.; Gobec, S. Inhibitors of 17beta-hydroxysteroid Dehydrogenase Type 1. Curr. Med. Chem. 2008, 15, 137–150. (18) Poirier, D. Contribution to the Development of Inhibitors of 17β-hydroxysteroid Dehydrogenase Types 1 and 7: Key Tools for Studying and Treating Estrogen-dependent Diseases. J. Steroid Biochem. Mol. Biol. 2011, 125, 83–94. (19) Day, J. M.; Tutill, H. J.; Purohit, A.; Reed, M. J. Design and Validation of Specific Inhibitors of 17beta- hydroxysteroid Dehydrogenases for Therapeutic Application in Breast and Prostate Cancer, and in Endometriosis. Endocr. Relat. Cancer 2008, 15 , 665–692. (20) Moeller, G.; Adamski, J. Integrated View on 17beta-hydroxysteroid Dehydrogenases. Mol. Cell. Endocrinol. 2009, 301, 7–19. (21) Vihko, P.; Härkönen, P.; Soronen, P.; Törn, S.; Herrala, A.; Kurkela, R.; Pulkka, A.; Oduwole, O.; Isomaa, V. 17 Beta-hydroxysteroid Dehydrogenases--their Role in Pathophysiology. Mol. Cell. Endocrinol. 2004, 215, 83–88. (22) Keller, B.; Ohnesorg, T.; Mindnich, R.; Gloeckner, C. J.; Breitling, R.; Scharfe, M.; Moeller, G.; Blöcker, H.; Adamski, J. Interspecies Comparison of Gene Structure and Computational Analysis of Gene Regulation of 17beta-hydroxysteroid Dehydrogenase Type 1. Mol. Cell. Endocrinol. 2006, 248, 168–171.

103 (23) Messinger, J.; Husen, B.; Koskimies, P.; Hirvelä, L.; Kallio, L.; Saarenketo, P.; Thole, H. Estrone C15 derivatives—A New Class of 17β-hydroxysteroid Dehydrogenase Type 1 Inhibitors. Mol. Cell. Endocrinol. 2009, 301, 216–224. (24) Brožič, P.; Kocbek, P.; Sova, M.; Kristl, J.; Martens, S.; Adamski, J.; Gobec, S.; Lanišnik Rižner, T. Flavonoids and Cinnamic Acid Derivatives as Inhibitors of 17β-hydroxysteroid Dehydrogenase Type 1. Mol. Cell. Endocrinol. 2009, 301, 229–234. (25) Starčević, š.; Kocbek, P.; Hribar, G.; Lanišnik Rižner, T.; Gobec, S. Biochemical and Biological Evaluation of Novel Potent Coumarin Inhibitor of 17β-HSD Type 1. Chem. Biol. Interact. 2011, 191, 60–65. (26) Spadaro, A.; Negri, M.; Marchais-Oberwinkler, S.; Bey, E.; Frotscher, M. Hydroxybenzothiazoles as New Nonsteroidal Inhibitors of 17β-Hydroxysteroid Dehydrogenase Type 1 (17β-HSD1). PLoS ONE 2012, 7, e29252. (27) Henn, C.; Einspanier, A.; Marchais-Oberwinkler, S.; Frotscher, M.; Hartmann, R. W. Lead Optimization of 17β-HSD1 Inhibitors of the (hydroxyphenyl)naphthol Sulfonamide Type for the Treatment of Endometriosis. J. Med. Chem. 2012, 55, 3307–3318. (28) Maltais, R.; Ayan, D.; Trottier, A.; Barbeau, X.; Lagüe, P.; Bouchard, J.-E.; Poirier, D. Discovery of a Non-Estrogenic Irreversible Inhibitor of 17β-Hydroxysteroid Dehydrogenase Type 1 from 3- Substituted-16β-(m-carbamoylbenzyl)-estradiol Derivatives. J. Med. Chem. 2014, 57, 204–222. (29) Maltais, R.; Ayan, D.; Poirier, D. Crucial Role of 3-Bromoethyl in Removing the Estrogenic Activity of 17β-HSD1 Inhibitor 16β-(m-Carbamoylbenzyl)estradiol. ACS Med. Chem. Lett. 2011, 2, 678–681. (30) Ayan, D.; Maltais, R.; Roy, J.; Poirier, D. A New Nonestrogenic Steroidal Inhibitor of 17β- hydroxysteroid Dehydrogenase Type I Blocks the Estrogen-dependent Breast Cancer Tumor Growth Induced by Estrone. Mol. Cancer Ther. 2012, 11, 2096–2104. (31) Singh, J.; Petter, R. C.; Baillie, T. A.; Whitty, A. The Resurgence of Covalent Drugs. Nat. Rev. Drug Discov. 2011, 10, 307–317. (32) Smith, A. J. T.; Zhang, X.; Leach, A. G.; Houk, K. N. Beyond Picomolar Affinities: Quantitative Aspects of Noncovalent and Covalent Binding of Drugs to Proteins. J. Med. Chem. 2009, 52, 225– 233. (33) Gierse, J. K.; Koboldt, C. M.; Walker, M. C.; Seibert, K.; Isakson, P. C. Kinetic Basis for Selective Inhibition of Cyclo-oxygenases. Biochem. J. 1999, 339 ( Pt 3), 607–614. (34) Brueggemeier, R. W.; Katlic, N. E. Aromatase Inhibition by an Enzyme-activated Irreversible Inhibitor in Human Carcinoma Cell Cultures. Cancer Res. 1990, 50, 3652–3656. (35) Mazumdar, M.; Fournier, D.; Zhu, D.-W.; Cadot, C.; Poirier, D.; Lin, S.-X. Binary and Ternary Crystal Structure Analyses of a Novel Inhibitor with 17beta-HSD Type 1: a Lead Compound for Breast Cancer Therapy. Biochem. J. 2009, 424, 357–366. (36) Kitz, R.; Wilson, I. B. Esters of Methanesulfonic Acid as Irreversible Inhibitors of Acetylcholinesterase. J. Biol. Chem. 1962, 237, 3245–3249. (37) Beitz, E. TEXshade: Shading and Labeling of Multiple Sequence Alignments Using LATEX2 Epsilon. Bioinforma. Oxf. Engl. 2000, 16, 135–139.

104

Chapitre 3

105 Résumé

Les petites molécules qui modulent directement une enzyme pour en stimuler l’activité sont plutôt rares. À notre connaissance, un tel activateur n’avait jamais été rapporté pour aucun membre de la famille des 17β-HSD. Nous aurions ainsi fait la découverte fortuite d’un premier activateur, un stéroïde causant une augmentation allant jusqu’à 300% de l’activité de la 17β-HSD12. La stimulation de la transformation de l’estrone (E1) en estradiol (E2) a été observée dans des cellules HEK 293 intactes surexprimant l’enzyme et en homogénats partiellement purifiés de même que dans des cellules du cancer du sein (T47D). Une étude de relation structure-activité montre que cette activité est liée à la nature et la position de certains substituants sur le cycle [1,3]oxazinan-2- one de E2 ce qui pourrait aider à l’identification d’une voie d’activation endogène de la 17β-HSD12 insoupçonnée. Ainsi, l’activateur découvert pourrait être un outil précieux pour étudier cette enzyme peu connue.

106

ACS Chemical Biology, 2014, 9, 1668-1673

Identification of a first enzymatic activator of a hydroxysteroid dehydrogenase

Alexandre Trottier, René Maltais and Donald Poirier*

Laboratory of Medicinal Chemistry, Endocrinology and Nephrology Unit, CHU de Québec (CHUL, T4), and Faculty of Medicine, Laval University, Québec (Québec), G1V 4G2, Canada

Keywords: enzyme; activator; hydroxysteroid dehydrogenase; 3-ketoacyl-CoA reductase, steroidogenesis; estradiol derivative.

107 Abstract

Small molecule activators that directly modulate the activity of an enzyme are uncommon entities and such activators had never yet been identified for any 17-hydroxysteroid dehydrogenase (17-HSD). We hereby report the fortuitous discovery of a steroid derivative that caused an up to 3-fold increase in the activity of 17- HSD12. The stimulation of estrone to estradiol conversion has been characterized in intact and homogenized stably transfected HEK-293 cells and has also been observed in T47D breast cancer cells. Structure-activity relationships closely linked to the nature of the substituent on the [1,3]oxazinan-2-one ring of an estradiol derivative emerged from this study and may help in the identification of a previously unsuspected endogenous activation of 17-HSD12. This activator will therefore be a useful tool to study this relatively unknown enzyme as well as the possible activation of other 17-HSD family members.

Introduction

A decade after its discovery, 17β-hydroxysteroid dehydrogenase type 12 (17-HSD12) still remains relatively unknown. It is one of the most recently identified isoenzymes of the 17-HSD family, a group of key enzymes that are well-known to be involved in the last step of sex hormone biosynthesis.1,2 This enzyme was however initially reported as 3-ketoacyl-CoA reductase (KAR) for its essential role in fatty acid elongation cascade.3 It was later found to be able to reduce estrone (E1) in estradiol (E2) and was consequently renamed 17-HSD12.4 Since then, its physiological role remains a matter of debate. 17-HSD12 appeared to be responsible for the increase of potent estrogen biosynthesis in differentiated adipocytes.5 On the other hand, some recent works with 17-HSD12 knockout mice 6 and cancer cells 7,8 tend to show a greater involvement in fatty acid elongation. Both products, namely E2 and long chain fatty acids, are known to be involved in a variety of cancers; the former by the activation of estrogen receptors (ER) 2,9 and the later via further transformation into eicosanoids.10 The expression of 17-HSD12 has been linked to poor prognoses of breast,7 ovary 11 and prostate 12 cancers and is known to be stimulated by sterol regulatory element-binding protein-1, a transcription factor related to fatty acid metabolism.13 Until now, no post-translational modulation of the enzyme had been brought to our attention.

We recently evaluated the selectivity of certain enzyme inhibitors developed in our laboratory by testing their 17-HSD12 inhibition abilities in stably transfected HEK-293 cells. Surprisingly, a compound first designed as a 17-HSD1 inhibitor was found to more than double the E1 to E2 conversion basal activity. This was an

108

unexpected result, considering that no molecule had been reported to turn on a 17-HSD nor any other dehydrogenases or steroidogenesis enzymes, as far as we know. The discovery of a small molecule activator that directly interacts with the enzyme to enhance its activity is a rather rare event as only twelve enzymes have well-known synthetic activators to date.14 The selective activation of an enzyme activity with a small molecule may potentially offer crucial knowledge about its endogenous regulation and could also have a great interest in therapeutic ends as well as in various aspects of fundamental research.15

The existence of a 17-HSD12 activator is interesting on its own, considering how little is known about this essential ubiquitous enzyme.6,16 Such an activator could therefore become a useful tool, which is currently lacking, to study 17-HSD12. Moreover, it sheds light on an unsuspected regulation pathway that could change the way we believe this crucial enzyme works. Thus, following the discovery of an activation effect caused by a first small molecule, both the activator and its effect were primarily characterized.

Results and discussion

Screening for stimulation of 17-HSD12 activity (intact cells).

After an initial fortuitous observation of enzymatic activity stimulation, available compounds synthesized in our laboratory with a closely related structure (Figure 1) were screened for 17-HSD12 inhibition/activation of [14C]- E1 to [14C]-E2 conversion in stably transfected HEK-293 cells. These compounds (A1, A2, A3, B1, B2, B3 and B4) were obtained by adapting a diversity oriented synthesis (DOS) methodology recently reported by our research group for the synthesis of various fused steroidal azacycles.17 For the synthesis of representative compounds A1 and A3, 3-methoxymethyl-O-estrone was submitted to an aldol condensation with 3-formyl- benzonitrile to give the intermediate enone (Scheme 1). The carbonyl of this conjugated ketone was then stereoselectively reduced with sodium borohydride to give the 17-OH derivative, which was treated with a solution of m-chloroperbenzoic acid in chloroform. After a separation of both epoxide derivatives by chromatography, the major epoxide with an alpha stereochemistry was heated at a high temperature in a microwave apparatus with an amine (ethylamine or butylamine) to afford the corresponding aminodiol intermediate. Finally, the cyclisation between the secondary amine and the OH group using triphosgene followed by the hydrolysis of the 3-methoxymethylether protecting group provided the final compounds A1 and A3. Each compound was purified by chromatography and characterized to confirm their chemical structure and purity (Supporting Information).

109

Scheme 1. Chemical synthesis of compounds A1 and A3. Reagents and conditions: (a) methoxymethylchloride, diisopropylethylamine (DIPEA), dichloromethane (DCM), rt; (b) 3-CN-benzaldehyde, KOH, EtOH, 100 °C; (c) NaBH4, MeOH, rt; (d) m-chloroperbenzoic acid, DCM, rt; (e) appropiate alkylamine, EtOH, microwave, 180 °C; (f) triphosgene, DIPEA, DCM, rt; (g) HCl 10% (v/v) in MeOH, 50 °C.

Figure 1. Chemical structure of compounds tested for 17β-HSD12 modulation (activation and inhibition).

As initially observed, cells treated with compound A1 (10 M) showed an increase of 291% in E2 production when compared to the level of untreated cells (Figure 2A). In this screening assay, compound 55,18 a known 17-HSD12 inhibitor, was used as control. None of the other compounds stimulated nor inhibited the enzyme activity at 10 M. As a preliminary structure-activity relationship (SAR) statement, the R2 hydrophobic side chain

110

seems crucial to the activity since A2 only differs from A1 by a 2-hydroxy-ethyl chain in R2 instead of a butyl group. Also, the activating effect of this chain is highly affected by R1 substitution on the estratriene core. This could explain the lack of stimulation observed by compounds from the B series, particularly B1, which only differs from A1 by an unsaturation at position 15-16 (Δ15-16) instead of a 16-hydroxy group on a saturated D-ring. From our first results, no stimulation of enzyme activity could be imputed to R3 substitution, but B3 and B4 were the two only tested compounds with other substitutions than carboxamide and they showed no activation just like their carboxamide counterparts B2 and B1. Thus, a 16-OH-estratriene core as well as an adequately positioned hydrophobic chain on the E-ring seems favorable for the stimulation of E2 production by 17-HSD12 stably overexpressed in HEK-293 cells.

Figure 2. Modulation (activation or inhibition) of the transformation of [14C]-E1 to [14C]-E2 by 17β-HSD12 in transfected HEK-293 intact cells. A) Screening of compounds 55, A1, A2, B1, B2, B3 and B4 tested at 10 µM. Ctrl: vehicle only. B) Stimulation of 17β-HSD12 activity by compounds A3 (grey) and A1 (black). Results of one experiment performed in triplicate (mean ± standard deviation).

In order to confirm the activation effect of a hydrophobic R2 group, compound A3 was synthesized and tested for the transformation of [14C]-E1 by 17-HSD12 (Figure 2B). This compound only differs from A1 by the replacement of a butyl by an ethyl group in R2. The production of [14C]-E2 was stimulated by a treatment with A3 at concentrations of 10 and 20 M, but not at 1 M. This stimulation by A3 was however considerably lower than by A1 (36 and 57% stimulation against 89 and 202% at 10 and 20 M, respectively). This result confirms that the activation effect of A1 is not an artifact and also importantly validates the significant role of the hydrophobic R2 side chain for the observed stimulation. It would also be interesting to test compounds with

111 various R2 hydrophobic substitutions and different R3 moieties for an accurate SAR determination which may lead to synthesis of better synthetic activators or to the identification of endogenous compounds susceptible to exert similar or better activation of 17-HSD12.

Characterization of the activation effect in HEK-293 intact cells and cell-free assays.

Since compound A1, as a steroid derivative, might interact with various targets other than 17-HSD12, further tests were needed to determine the possible mode of action responsible for the observed increase of E2 production. To verify the hypothesis of a receptor-mediated enzymatic stimulation of activity and/or expression, A1 was tested at many concentrations in both intact cells and cell-free assay.

Figure 3. Stimulation of [14C]-E1 to [14C]-E2 conversion by compound A1. A) Experiment performed in intact HEK-293 cells stably expressing 17β-HSD12. B) Experiment performed in microsomal extract obtained from homogenized HEK-293 cells stably expressing 17β-HSD12. Each figure is the result of one experiment performed in triplicate (mean ± standard deviation) and is representative of two assays.

Activation assays with compound A1 in intact cells showed a clear dose-dependent stimulation of E1 to E2 conversion (Figure 3A) in the range of concentrations tested. The maximal activation was however not reached considering that higher doses have cytotoxic effect, the EC50 (effective concentration necessary to stimulate 50% of the maximal activity) and the maximal activation of basal activity have consequently not been determined for A1. For cell-free assays, a microsomal extract was isolated from a homogenate of HEK-293 cells stably expressing 17-HSD12 as previously described.4,19 A dose-dependent stimulation of E1 to E2 conversion by microsomal extract has been shown from 2.5 to 10 M of A1 with the maximal activity observed at around 20

112

M (Figure 3B). An EC50 of 7.5 M and a maximal activation of 120% were evaluated for A1. To determine whether compound A1 interacts with 17-HSD12 by an allosteric site, we calculated the ratio between the compound A1 (S) concentrations at 90% and 10% of Vmax (S0.9/S0.1). If an enzyme follows Michaelis-Menten

20 kinetic in which the substrate concentration is equal to Km at Vmax/2, the S0.9/S0.1 value will be equal to 81. From data obtained in cell-free assay (Figure 3B), we determined a S0.9/S0.1 ratio of 8.9, which value indicates that compound A1 activates the 17-HSD12 by an allosteric site. This ratio value also allowed us to estimate the number of interacting sites as 2.

The hypothesis of any genomic, transcriptional or proliferative effect of compound A1 is overturned by the observed activation in a cell-free assay, without other enzymes involved in E1 to E2 metabolism than 17- HSD12,4 and with the lack of transduction pathways.21 In fact, the doubled enzyme activity observed in as little as one hour is fairly unlikely to be mediated by a receptor or any other enzyme than the 17-HSD12 in microsomal extract of stably transfected HEK-293 cells. A direct post-translational activation of the 17-HSD12 is thus the only plausible explanation for the results in the cell-free assay. Nonetheless, the consistently higher maximal stimulation in intact cells than in microsomal faction might point out that compound A1 has additional effect in intact cells. Microsomal extraction process, change of enzymatic environment, assay conditions (different substrate concentration and time) or other factor affecting enzyme activity are most likely to be responsible of this difference as the effective concentrations remain unchanged in intact cells and in cell-free assay. These similar EC50 values are consistent with the effect of an enzyme activator, because neither the absence of intact cells nor the lack of cytoplasmic element seems to affect the potency of compound A1.

Activation of 17-HSD12 in intact T47D cells.

To confirm the activation effect in more physiological-like conditions than the enzyme overexpressed in HEK- 293 cells, activation assays were conducted in T47D cells (breast adenocarcinoma). These cells are well-known to show a high endogenous expression of 17-HSD12 as well as a weaker expression of 17-HSD1, 17-HSD7 and ER, which are prone to interact with the measurement of E1 to E2 conversion.22 Prior to the assay, modulation of 17-HSD activities that may interfere with the results has first been assessed (Figure 4A) using T47D cells for 17-HSD1 and transfected HEK-293 cells for the three other 17-HSDs (types 2, 7 and 12). Compound A1 showed a good selectivity with 17-HSD12 stimulation (235%) and a weak 17-HSD1 inhibition (50%) at 10 M only, while no modulation of 17-HSD2 and 17-HSD7 was observed at both 1 and 10 M. It should be noted that the 17-HSD1 is by far the most efficient enzyme and is considered as being mainly

113 responsible for E1 to E2 conversion.8,22 In fact, T47D cells are often used to assess 17-HSD1 inhibition as the activities of 17-HSD7 and 17-HSD12 are not detectable in normal conditions.8,22 In order to detect any modulation of 17-HSD12 activity, the 17-HSD1 was totally inactivated with 1 M of PBRM, a selective and irreversible inhibitor reported to have a long-acting effect.23,24 No inhibitor of 17-HSD2 and 17-HSD7 was used considering the lack of enzyme expression and activity, respectively, in the T47D cells.22 However, the cells had to be cultured at their maximal density to avoid any effect through estrogenic stimulation caused by ER activation.

Figure 4. A) Evaluation of compound A1’s selectivity at 1 µM (grey) and 10 µM (black) for the transformation of [14C]-E1 by 17β-HSD1, 17β-HSD7 and 17β-HSD12, and for the transformation of [14C]-E2 by 17β-HSD2. T47D cells were used for 17β-HSD1 assays and transfected HEK-293 cells for the other enzymes. B) Stimulation by compound A1 of [14C]-E1 to [14C]-E2 transformation in T47D cells cultured at confluence and treated with a potent 17β-HSD1 inhibitor (PBRM). These are the results of one experiment in triplicate (mean ± standard deviation) representative of two assays. The transformation of E1 to E2 by the T47D cells increased considerably from 0 to 72 h under all the conditions reported in Figure 4B. The rate of transformation was enhanced in a concentration-dependent manner for cells treated with 5, 10 and 20 M of compound A1, which showed higher transformation values than control after 24, 72 and 120 h. At 1 M, a small difference was observed only after 120 h. E1 to E2 conversion appeared to reach a plateau after 72 h at all concentrations except at 20 M. In fact, no more E2 appeared to be produced from 72 to 120 h at 0 M (27 vs. 26 %), 1 M (28 vs. 29 %), 5 M (32 vs. 34 %) and 10 M (38 vs. 41 %) of compound A1, whereas a small increase was observed at 20 M (47 vs. 55%). Such lowered and/or leveled off transformation after 72 h is not surprising considering that at this point cells are confluent in culture dishes for 5 days without a medium change. It should be noted that the observed production of E2 by the 17-HSD12 is very weak and the activity would not have been observed if the 17-HSD1 was effective. Without the use of a 17-

114

HSD1 inhibitor, all the substrate would have been transformed 22 in only a few hours while the effect of 17- HSD12 needs days to be adequately observed.

The previous experiments thus confirmed the activation potential of compound A1 in a breast cancer cell lines. Such activation of 17-HSD12 confirms the existence of an endogenous activation of the enzyme. Indeed, the activator we synthesized is most likely to mimic an endogenous activation factor as it is the case for already characterized enzyme activators.15 This possible modulation of 17-HSD12, as observed for the transformation of E1 to E2, could also stimulate, inhibit or have no effect on fatty acid elongation. It was however impossible to test this potential activity of 17-HSD12 because of the actual lack of suitable substrates and assays. We do not even know if both enzyme activities are catalyzed by the same catalytic site and how compound A1 could stimulate these two activities. These two issues should be addressed for a better understanding of the observed activation and its implications.

For now, compound A1 may become a valuable tool to investigate the probable endogenous activation of 17- HSD12. At this time, we have found no pharmacological interest in activating this enzyme considering its relatively unknown role in cancers 7,12,22 and the fact that no physiological issue caused by 17-HSD12 deficiency and/or lack of activity has yet been reported in humans, as far as we know. Moreover, the transformation catalyzed by this enzyme is not a rate-limiting step in the production of essential long chain fatty acids, so its activation might have little or no effect 25,26 and a significant physiological involvement in E2 production remains uncertain. However, knowledge about the stimulation of 17-HSD12 activity might lead to insight in the development of potent inhibitors which could be interesting for the treatment of some cancers and inflammatory diseases.27 Additionally, this first example of a 17-HSD activator gives rise to the possible identification of other 17-HSD activators which could be of great interest to treat, prevent and/or study many diseases such as cancer, 1,2,27 endometriosis,28 osteoporosis,29 or disorders in sexual development (formerly named pseudohermaphrodism) 30 for instance.

In summary, we report the identification of the first known activator of a 17-HSD family member. Thus, a treatment with compound A1 has produced stimulation of 17-HSD12 activity (transformation of E1 to E2) in intact and homogenized stably transfected HEK-293 cells as well as in intact T47D breast cancer cells. The exact mechanism of action is not yet known, but our results using a microsomal preparation suggest an allosteric

115 binding of compound A1 to 17-HSD12. This activator could be an insightful tool to further study this little known enzyme and its modulation.

METHODS Cells and methods.

T47D cells were obtained from the American Type Culture Collection (ATCC) and HEK-293 cells stably transfected with 17-HSD12, 17-HSD7 or 17-HSD2 were kindly provided by Dr. Van Luu-The (CHUQ-CHUL Research Center, Québec, Canada). Cell culture was conducted as previously described.31 The final concentration of dimethylsulfoxide (DMSO) and ethanol in the culture medium was adjusted for each assay and did not exceed 0.5% (v/v) together. [14C]-radiolabeled substrate in ethanol solution, namely [14C]-E1 and [14C]- E2 (both from American Radiolabeled Chemicals, Inc.), were used to assess enzyme activities. Steroids were extracted from medium with diethyl ether, separated by thin layer chromatography (plates from EMD Chemicals Inc.) with toluene/acetone (4:1) and finally quantified using Storm 860 system (Molecular Dynamics) as described.32

Modulation of 17-HSD12 in HEK-293 cells.

HEK-293 cells were plated at 3 x 105 cells/well in 24-well culture dishes in MEM medium as reported,31 but complete steroid-deprived medium was obtained by using 10% (w/v) charcoal-stripped medium FBS (Wisent Inc.) instead of 10% FBS. After 48 h of incubation, cells were treated by adding a DMSO solution of desired

14 compound (0, 1, 10 and/or 20 M) and 60 nM of [ C]-E1. For EC50 determination, final compound concentrations of 10 nM to 20 M were tested. Culture medium was removed 24 h later for further quantification of radiolabeled steroids (E1 and E2).

Modulation of 17-HSD7 in HEK-293 cells.

HEK-293 cells were plated in MEM complete steroid-deprived medium as above at a density of 1.5 x 103 cells/well in 24-well culture dishes. After 48 h of incubation, cells were treated with a DMSO solution of compound A1 (0, 1 and 10 M) and [14C]-E1 (60 nM). Culture medium was removed 18 h later for further quantification of radiolabeled steroids (E1 and E2).

116

Modulation of 17-HSD2 in HEK-293 cells.

HEK-293 cells were plated in MEM complete steroid-deprived medium as above at a density of 5 x 103 cells/well in 24-well culture dishes. After 48 h of incubation, cells were treated by adding a DMSO solution of compound A1 and 60 nM of [14C]-E2. Culture medium was removed 1 h later for further quantification of radiolabeled steroids (E1 and E2).

Cell-free assay.

Homogenates of HEK-293 cells expressing 17-HSD124 were obtained as previously reported19 except that cells were lysed by five freeze/thaw (–80 °C/4 °C) cycles and pipetting (up and down) between each cycle instead of sonifying. Cell lysates were centrifuged at 10000g and the obtained supernatant centrifuged again at 100000g to isolate the microsomal fraction. The microsomal fraction was suspended in Tris buffer containing 20% (v/v) glycerol and BSA (0.2 mg/mL), and a sample of 100 µL was added to a test tube. The resulting suspension was treated with [14C]-E1 (10 nM) and a DMSO solution of compound A1 (final concentration of 0, 2.5, 5, 10, 20 or 40 M) in a total volume of 1 mL. After 1 h of incubation at 37 °C with shaking, the tubes were immediately cooled down in ice and the steroids (E1 and E2) extracted for quantification.

Inhibition of 17-HSD1 in T47D cells.

T47D cells were plated in RPMI complete steroid-deprived medium at a density of 3 x 103 cells/well in 24-well culture dishes as already described.24 After 48 h of incubation, cells were treated by adding a DMSO solution of desired compound (0, 1 or 10 M) and [14C]-E1 (60 nM). Culture medium was removed 24 h later for further quantification of radiolabeled steroids (E1 and E2).

Activity of 17-HSD12 in T47D cells.

T47D cells were plated in RPMI complete steroid-deprived medium at a density of 2 x 105 cells/well in 24-well culture dishes. After 24 h of incubation, DMSO solution of 17-HSD1 inhibitor PBRM (compound 23b in ref. 23) was added to the wells to reach a final concentration of 1 M. After a second 24 h incubation, cell medium was removed and replaced by 1 mL of RPMI complete steroid-deprived medium containing 1 M of PBRM and 0, 1,

117 5, 10 or 20 M of compound A1 both in DMSO solution. [14C]-E1 (60 nM final) was then added in the wells. Culture medium was removed after 0, 24, 72 or 120 h for further quantification of radiolabeled steroids (E1 and E2).

Data analysis.

Both “Allosteric sigmoidal” and “log (agonist) vs. response -- Variable slope” fits were performed using GraphPad Prism version 5.00 for Windows (GraphPad Software, www.graphpad.com) to analyze dose-response assays in intact cells and microsomal preparation. Figures 3A and 3B were drawn with the first fit, while EC50 was evaluated using the latter.

Associated Content

Supporting Information

Details of the chemical synthesis and characterisation of compounds A1 and A3 and intermediate compounds. This material is available free of charge via the Internet at http://pubs.acs.org.

Author Information

Corresponding Author

*E-mail: [email protected]

Notes

The authors declare no competing financial interest.

118

Ackowledgement

This work was supported by a grant from the Canadian Institutes of Health Research (grant number MOP43994) to DP, an FRQS fellowship to AT and a CREMOGH fellowship to AT. We would like to thank V. Luu-The (CHUQ- CHUL Research Center) for providing transfected HEK-293 cells and M. Harvey for careful reading of the manuscript.

References

(1) Moeller, G., and Adamski, J. (2009) Integrated view on 17beta-hydroxysteroid dehydrogenases, Mol. Cell. Endocrinol. 301, 7–19.

(2) Day, J. M., Tutill, H. J., Purohit, A., and Reed, M. J. (2008) Design and validation of specific inhibitors of 17beta-hydroxysteroid dehydrogenases for therapeutic application in breast and prostate cancer, and in endometriosis, Endocr. Relat. Cancer 15, 665–692.

(3) Moon, Y.-A., and Horton, J. D. (2003) Identification of two mammalian reductases involved in the two-carbon fatty acyl elongation cascade, J. Biol. Chem. 278, 7335–7343.

(4) Luu-The, V., Tremblay, P., and Labrie, F. (2006) Characterization of type 12 17β-hydroxysteroid dehydrogenase, an isoform of type 3 17β-hydroxysteroid dehydrogenase responsible for estradiol formation in women, Mol. Endocrinol. 20, 437–443.

(5) Bellemare, V., Laberge, P., Noël, S., Tchernof, A., and Luu-The, V. (2009) Differential estrogenic 17β- hydroxysteroid dehydrogenase activity and type 12 17β-hydroxysteroid dehydrogenase expression levels in preadipocytes and differentiated adipocytes, J. Steroid Biochem. Mol. Biol. 114, 129–134.

(6) Rantakari, P., Lagerbohm, H., Kaimainen, M., Suomela, J.-P., Strauss, L., Sainio, K., Pakarinen, P., and Poutanen, M. (2010) Hydroxysteroid (17β) dehydrogenase 12 is essential for mouse organogenesis and embryonic survival, Endocrinology 151, 1893–1901.

(7) Nagasaki, S., Suzuki, T., Miki, Y., Akahira, J., Kitada, K., Ishida, T., Handa, H., Ohuchi, N., and Sasano, H. (2009) 17β-Hydroxysteroid dehydrogenase type 12 in human breast carcinoma: A prognostic factor via potential regulation of fatty acid synthesis, Cancer Res. 69, 1392–1399.

119 (8) Day, J. M., Foster, P. A., Tutill, H. J., Parsons, M. F. C., Newman, S. P., Chander, S. K., Allan, G. M., Lawrence, H. R., Vicker, N., Potter, B. V. L., Reed, M. J., and Purohit, A. (2008) 17beta-hydroxysteroid dehydrogenase type 1, and not type 12, is a target for endocrine therapy of hormone-dependent breast cancer, Int. J. Cancer 122, 1931–1940.

(9) Thomas, C., and Gustafsson, J.-Å. (2011) The different roles of ER subtypes in cancer biology and therapy, Nat. Rev. Cancer 11, 597–608.

(10) Tucker, S. C., and Honn, K. V. (2013) Emerging targets in lipid-based therapy, Biochem. Pharmacol. 85, 673–688.

(11) Szajnik, M., Szczepanski, M. J., Elishaev, E., Visus, C., Lenzner, D., Zabel, M., Glura, M., DeLeo, A. B., and Whiteside, T. L. (2012) 17β Hydroxysteroid dehydrogenase type 12 (HSD17B12) is a marker of poor prognosis in ovarian carcinoma, Gynecol. Oncol. 127, 587–594.

(12) Audet-Walsh, É., Bellemare, J., Lacombe, L., Fradet, Y., Fradet, V., Douville, P., Guillemette, C., and Lévesque, É. (2012) The impact of germline genetic variations in hydroxysteroid (17-beta) dehydrogenases on prostate cancer outcomes after prostatectomy, European Urology 62, 88–96.

(13) Nagasaki, S., Miki, Y., Akahira, J., Suzuki, T., and Sasano, H. (2009) Transcriptional regulation of 17beta- hydroxysteroid dehydrogenase type 12 by SREBP-1, Mol. Cell. Endocrinol. 307, 163–168.

(14) Zorn, J. A., and Wells, J. A. (2010) Turning enzymes ON with small molecules, Nat. Chem. Biol. 6, 179– 188.

(15) Bishop, A. C., and Chen, V. L. (2009) Brought to life: targeted activation of enzyme function with small molecules, J. Chem. Biol. 2, 1–9.

(16) Sakurai, N., Miki, Y., Suzuki, T., Watanabe, K., Narita, T., Ando, K., Yung, T. M. C., Aoki, D., Sasano, H., and Handa, H. (2006) Systemic distribution and tissue localizations of human 17beta-hydroxysteroid dehydrogenase type 12, J. Steroid Biochem. Mol. Biol. 99, 174–181.

(17) Maltais, R., and Poirier, D. (2012) Diversity-oriented synthesis of spiro- and fused azacycles from ketone molecular templates, Eur. J. Org. Chem. 5435–5439.

(18) Farhane, S., Fournier, M.-A., and Poirier, D. (2011) Synthesis and evaluation of amido-deoxyestradiol derivatives as inhibitors of 17β-hydroxysteroid dehydrogenase type 12, Curr. Enzyme Inhib. 7, 134–146.

120

(19) Luu-The, V., Zhang, Y., Poirier, D., and Labrie, F. (1995) Characteristics of human types 1, 2 and 3 17 beta- hydroxysteroid dehydrogenase activities: oxidation/reduction and inhibition, J. Steroid Biochem. Mol. Biol. 55, 581–587.

(20) Segel, I. H. (1976) Biochemical calculations: How to solve mathematical problems in general Biochemistry, 2nd ed., pp 303-317, John Wiley and Sons, New York.

(21) Pertoft, H. (2000) Fractionation of cells and subcellular particles with Percoll, J. Biochem. Biophys. Methods 44, 1–30.

(22) Laplante, Y., Rancourt, C., and Poirier, D. (2009) Relative involvement of three 17beta-hydroxysteroid dehydrogenases (types 1, 7 and 12) in the formation of estradiol in various breast cancer cell lines using selective inhibitors, Mol. Cell. Endocrinol. 301, 146–153.

(23) Maltais, R., Ayan, D., Trottier, A., Barbeau, X., Lagüe, P., Bouchard, J.E, and Poirier, D. (2014) Discovery of a non-estrogenic irreversible inhibitor of 17b-hydroxysteroid dehydrogenase type 1 from 3-substituted 16β- (m-carbamoylbenzyl)estradiol derivatives, J. Med. Chem. 57, 204-222.

(24) Ayan, D., Maltais, R., Roy, J., and Poirier, D. (2012) A new nonestrogenic steroidal inhibitor of 17β- hydroxysteroid dehydrogenase type 1 blocks the estrogen-dependent breast cancer tumor growth induced by estrone, Mol. Cancer Ther. 11, 2096-2104.

(25) Plourde, M., and Cunnane, S. C. (2007) Extremely limited synthesis of long chain polyunsaturates in adults: implications for their dietary essentiality and use as supplements, Appl. Physiol. Nutr. Metab. 32, 619–634.

(26) Jump, D. B. (2009) Mammalian fatty acid elongases, Methods Mol. Biol. 579, 375–389.

(27) Bozza, P. T., Bakker-Abreu, I., Navarro-Xavier, R. A., and Bandeira-Melo, C. (2011) Lipid body function in eicosanoid synthesis: an update, Prostaglandins Leukot. Essent. Fatty Acids 85, 205–213.

(28) Dassen, H., Punyadeera, C., Kamps, R., Delvoux, B., Van Langendonckt, A., Donnez, J., Husen, B., Thole, H., Dunselman, G., and Groothuis, P. (2007) Estrogen metabolizing enzymes in endometrium and endometriosis, Human Reproduction 22, 3148–3158.

(29) Marchais-Oberwinkler, S., Henn, C., Möller, G., Klein, T., Negri, M., Oster, A., Spadaro, A., Werth, R., Wetzel, M., Xu, K., Frotscher, M., Hartmann, R. W., and Adamski, J. (2011) 17β-Hydroxysteroid dehydrogenases (17β-HSDs) as therapeutic targets: Protein structures, functions, and recent progress in inhibitor development, J. Steroid Biochem. Mol. Biol. 125, 66–82.

121 (30) George, M. M., New, M. I., Ten, S., Sultan, C., and Bhangoo, A. (2010) The clinical and molecular heterogeneity of 17βHSD-3 enzyme deficiency, Horm. Res. Paediatr. 74, 229–240.

(31) Cadot, C., Laplante, Y., Kamal, F., Luu-The, V., and Poirier, D. (2007) C6-(N,N-butyl-methyl-heptanamide) derivatives of estrone and estradiol as inhibitors of type 1 17β-hydroxysteroid dehydrogenase: Chemical synthesis and biological evaluation, Bioorg. Med. Chem. 15, 714–726.

(32) Ngatcha, B. T., Laplante, Y., Labrie, F., Luu-The, V., and Poirier, D. (2006) 3β-Alkyl-androsterones as inhibitors of type 3 17β-hydroxysteroid dehydrogenase: inhibitory potency in intact cells, selectivity towards isoforms 1, 2, 5 and 7, binding affinity for steroid receptors, and proliferative/antiproliferative activities on AR+ and ER+ cell lines, Mol. Cell. Endocrinol. 248, 225–232

122

&RQFOXVLRQ

Les 17β-HSD sont définitivement des enzymes difficiles d’approche. En plusieurs décennies de recherche, aucun inhibiteur pour l’une d’entre elles n’est parvenu au stade des études cliniques, bien souvent faute d’atteindre la puissance et/ou la sélectivité minimale dès le début du développement. Le PBRM est l’un des quelques inhibiteurs susceptibles d’être le premier à parvenir au stade des études cliniques. Il a précédemment été démontré qu’il était dépourvu de toute activité estrogénique et qu’il était très efficace, cela autant en laboratoire que chez l’animal.

Figure 1. Structure du PBRM

Dans le cadre des travaux de ce mémoire, le PBRM s’est également avéré sélectif vis-à-vis des 17β-HSD de types 2, 3, 7 et 12 tout en ne présentant qu’une faible inhibition de la CYP3A4, une enzyme importante du métabolisme des médicaments. S’il a bien été établi par diverses méthodes qu’il n’avait aucun effet sur ERα, l’affinité pour ERβ et GPER reste cependant à établir tout comme sa capacité à inhiber les autres 17β-HSD et CYP450.

Le mode d’action irréversible et compétitif de cet inhibiteur a également été dévoilé ce qui a nécessité le calcul de son Ki, et de son kinact. Un tel effet pourrait être un avantage pour le traitement de maladies chroniques comme le cancer du sein et l’endométriose ce qui permet généralement une administration de doses plus faibles et/ou moins fréquentes surtout considérant la longue demi-vie plasmatique déjà rapportée. En fait, les essais visant à mesurer le taux de resynthèse de la 17β-HSD1 ont fourni une première preuve qu’un tel effet à long terme est probable pour le PBRM. L’inhibition de l’enzyme perdurait plusieurs jours après que l’inhibiteur soit retiré des deux lignées de cellules cancéreuses. De plus, les travaux effectués dernièrement au niveau pharmacocinétique tendent à montrer un profil favorable de l’inhibiteur dont une bonne absorption par voie orale et une distribution acceptable.

123 Un mécanisme d’action a d’ailleurs été proposé sur la base d’une étude de modélisation moléculaire des séries d’inhibiteurs ayant mené au PBRM, ce qui a été appuyé par l’inhibition et la modélisation des formes orthologues de la 17β-HSD1. La porte est ainsi ouverte pour le développement d’une nouvelle génération d’inhibiteurs qui viseraient spécifiquement la formation d’un lien covalent avec l’His221 de l’enzyme. Pour l’instant, après 20 ans de travaux au sein de notre laboratoire pour y arriver, le PBRM présente toutes les caractéristiques désirées d’un inhibiteur de la 17β-HSD1.

On peut également déduire que l’effet de l’inhibiteur sera hautement sélectif en faisant une synthèse des informations obtenues jusqu’à maintenant. En effet, il est peu probable qu’une autre protéine possède le domaine précis nécessaire pour que le PBRM la lie irréversiblement ou y réside suffisamment longtemps. Dans le cas d’un tel inhibiteur irréversible, c’est la capacité à inhiber à long terme qui est importante sur le plan de la spécificité et non l’affinité pour une cible ou une autre. S’il faut bel et bien plusieurs jours avant que l’enzyme soit pleinement renouvelée, l’effet d’inhibition en est d’autant plus long.

Comme le PBRM est le seul inhibiteur ayant montré de telles capacités jusqu’à maintenant, il pourrait bien devenir un outil avantageux pour le traitement de l’endométriose, une maladie pour laquelle aucune option efficiente n’a été trouvée. Celles atteintes du cancer du sein pourraient aussi bénéficier d’un blocage plus complet et spécifique de la production des estrogènes par la combinaison d’inhibiteurs de l’aromatase et de la 17β-HSD1. L’inhibition que de cette dernière pourrait également être une option intéressante en prophylaxie si ses effets secondaires sont moindres que les autres traitements hormonaux. Il s’agit là d’un effet attendu puisque l’enzyme intervient à la toute dernière étape de la production endogène des estrogènes.

Toutefois, certains de ces résultats reposent sur des suppositions qui gagneraient à être vérifiées. À titre d’exemple, le taux de dissociation du PBRM une fois lié à la 17β-HSD1 n’a jamais été vérifié, même si tous les résultats pointent vers une valeur nulle pour ce taux. La valeur calculée pour le taux de resynthèse tiendrait moins de l’estimation si la démonstration formelle d’un taux de dissociation nul était faite et que l’expression de l’enzyme était mesurée au cours de l’expérience.

Dans un même ordre d’idée, les essais de sélectivité, autant par rapport aux autres espèces au chapitre 2 qu’aux différentes enzymes au chapitre 1, ont été menés dans des conditions passablement différentes. Nous assumons, probablement à juste titre, que ces différences n’ont aucun effet sur le type de valeur mesurée. En effet, le taux d’inhibition de l’activité initiale d’une enzyme est difficilement altérable et il n’y a aucune raison pour que des facteurs comme l’abondance ou l’expression de l’enzyme affectent les résultats. Toutefois, on ne peut exclure que d’autres facteurs inconnus viennent moduler l’inhibition ou l’activité de l’enzyme de manière différentielle entre différentes sources d’enzymes (type de cellules, homogénat vs enzyme purifiée, etc.).

124

Figure 2. Structure des activateurs efficaces identifiés jusqu’à maintenant pour la 17β-HSD12 où X est une chaîne alkyle.

Une nouvelle facette des 17β-HSD a aussi été révélée avec la caractérisation d’un premier activateur pour cette famille d’enzymes. Une poignée d’autres molécules parvenant à activer la 17β-HSD12 autant en microsomes qu’en cellules intactes l’exprimant de manière endogène ont par la suite été découvertes. L’existence même de ces activateurs synthétiques suggère fortement que l’enzyme est sujette à une activation endogène. Une telle information est fort importante vu l’expression forte et quasi ubiquitaire de la 17β-HSD12 de même que son implication à la fois dans l’élongation des acides gras, la production d’estrogènes et certains cancers.

Si les bénéfices de cette découverte sont purement fondamentaux, il sera intéressant de voir si d’autres 17β- HSD sont sujettes à une telle activation. La stimulation de certaines isozymes pourrait avoir un intérêt thérapeutique dont, par exemple, la 17β-HSD3 qui présente une bonne homologie avec la 17β-HSD12. Dans tous les cas, cette trouvaille inattendue d’un activateur ouvre bien des portes et soulève un certain nombre de questions. À la lumière des réponses qui y seront fournies, il faudra vraisemblablement réévaluer le rôle physiologique de cette enzyme.

À noter que ces travaux ne sont qu’un début qui visait à dévoiler l’existence d’une activation de la 17β-HSD12 plutôt qu’à l’étudier à proprement parler. Ainsi, puisqu’à aucun moment nous n’avons montré une activation de l’enzyme pure, il demeure une parcelle d’incertitude quant au fait que les activateurs identifiés agissent bel et bien directement sur l’enzyme. Comme des éléments pouvant potentiellement moduler l’activité de l’enzyme étaient présents dans chacun des tests, le doute est de mise, même si la perspective d’un activateur direct demeure la plus probable. Tester l’élongation des acides gras suite au traitement par nos molécules activatrices serait très éclairant quant à leur effet réel sur la 17β-HSD12

Pour les deux sujets abordés au cours de ma maîtrise, beaucoup de travail reste à faire pour que, dans le cas du premier, un inhibiteur de la 17β-HSD1 soit testé chez l’humain et, pour le second, que l’activation de la 17β- HSD12 et l’implication de cette découverte soit bien comprise. Dans tous les cas, les 17β-HSD demeurent une famille d’enzymes intéressantes à étudier qui réserve sans doute bien des surprises autant sur le plan thérapeutique que fondamental.

125

Bibliographie

(1) Thornton, J. W. Evolution of Vertebrate Steroid Receptors from an Ancestral Estrogen Receptor by Ligand Exploitation and Serial Genome Expansions. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 5671– 5676. (2) Thomas, C.; Gustafsson, J.-Å. The Different Roles of ER Subtypes in Cancer Biology and Therapy. Nat. Rev. Cancer 2011, 11, 597–608. (3) Toran-Allerand, C. D. Minireview: A Plethora of Estrogen Receptors in the Brain: Where Will It End? Endocrinology 2004, 145, 1069–1074. (4) Baker, M. E. What Are the Physiological Estrogens? Steroids 2013, 78, 337–340. (5) Wocławek-Potocka, I.; Mannelli, C.; Boruszewska, D.; Kowalczyk-Zieba, I.; Waśniewski, T.; Skarżyński, D. J. Diverse Effects of Phytoestrogens on the Reproductive Performance: Cow as a Model. Int. J. Endocrinol. 2013, 2013, 1–15. (6) Benotti, M. J.; Trenholm, R. A.; Vanderford, B. J.; Holady, J. C.; Stanford, B. D.; Snyder, S. A. Pharmaceuticals and Endocrine Disrupting Compounds in U.S. Drinking Water. Environ. Sci. Technol. 2009, 43, 597–603. (7) Welshons, W. V.; Krummel, B. M.; Gorski, J. Nuclear Localization of Unoccupied Receptors for Glucocorticoids, Estrogens, and Progesterone in GH3 Cells. Endocrinology 1985, 117, 2140–2147. (8) Levin, E. R. Minireview: Extranuclear Steroid Receptors: Roles in Modulation of Cell Functions. Mol. Endocrinol. 2011, 25, 377–384. (9) Coleman, K. M.; Smith, C. L. Intracellular Signaling Pathways: Nongenomic Actions of Estrogens and Ligand-independent Activation of Estrogen Receptors. Front. Biosci. 2001, 6, D1379–1391. (10) Nilsson, S.; Gustafsson, J.-Å. Estrogen Receptors: Therapies Targeted to Receptor Subtypes. Clin. Pharmacol. Ther. 2011, 89, 44–55. (11) Kuiper, G. G.; Carlsson, B.; Grandien, K.; Enmark, E.; Häggblad, J.; Nilsson, S.; Gustafsson, J. A. Comparison of the Ligand Binding Specificity and Transcript Tissue Distribution of Estrogen Receptors Alpha and Beta. Endocrinology 1997, 138, 863–870. (12) Gibson, D. A.; Saunders, P. T. K. Estrogen Dependent Signaling in Reproductive Tissues - a Role for Estrogen Receptors and Estrogen Related Receptors. Mol. Cell. Endocrinol. 2012, 348, 361–372. (13) Mosselman, S.; Polman, J.; Dijkema, R. ER Beta: Identification and Characterization of a Novel Human Estrogen Receptor. FEBS Lett. 1996, 392, 49–53. (14) Böttner, M.; Thelen, P.; Jarry, H. Estrogen Receptor Beta: Tissue Distribution and the Still Largely Enigmatic Physiological Function. J. Steroid Biochem. Mol. Biol. 2013, 139, 245-251. (15) Leygue, E.; Murphy, L. C. A Bi-faceted Role of Estrogen Receptor β in Breast Cancer. Endocr. Relat. Cancer 2013, 20, R127–139. (16) Leung, Y.-K.; Mak, P.; Hassan, S.; Ho, S.-M. Estrogen Receptor (ER)-beta Isoforms: a Key to Understanding ER-beta Signaling. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 13162–13167. (17) Filardo, E. J.; Quinn, J. A.; Bland, K. I.; Frackelton, A. R., Jr. Estrogen-induced Activation of Erk-1 and Erk-2 Requires the G Protein-coupled Receptor Homolog, GPR30, and Occurs via Trans- activation of the Epidermal Growth Factor Receptor through Release of HB-EGF. Mol. Endocrinol. 2000, 14, 1649–1660. (18) Srivastava, D. P.; Evans, P. D. G-protein Oestrogen Receptor 1: Trials and Tribulations of a Membrane Oestrogen Receptor. J. Neuroendocrinol. 2013, 25, 1219–1230. (19) Prossnitz, E. R.; Oprea, T. I.; Sklar, L. A.; Arterburn, J. B. The Ins and Outs of GPR30: a Transmembrane Estrogen Receptor. J. Steroid Biochem. Mol. Biol. 2008, 109, 350–353. (20) Thomas, P.; Pang, Y.; Filardo, E. J.; Dong, J. Identity of an Estrogen Membrane Receptor Coupled to a G Protein in Human Breast Cancer Cells. Endocrinology 2005, 146, 624–632. (21) Revankar, C. M.; Cimino, D. F.; Sklar, L. A.; Arterburn, J. B.; Prossnitz, E. R. A Transmembrane Intracellular Estrogen Receptor Mediates Rapid Cell Signaling. Science 2005, 307, 1625–1630.

127 (22) Isensee, J.; Meoli, L.; Zazzu, V.; Nabzdyk, C.; Witt, H.; Soewarto, D.; Effertz, K.; Fuchs, H.; Gailus- Durner, V.; Busch, D.; Adler, T.; de Angelis, M. H.; Irgang, M.; Otto, C.; Noppinger, P. R. Expression Pattern of G Protein-coupled Receptor 30 in LacZ Reporter Mice. Endocrinology 2009, 150, 1722– 1730. (23) Chimento, A.; Sirianni, R.; Casaburi, I.; Pezzi, V. GPER Signaling in Spermatogenesis and Testicular Tumors. Front Endocrinol (Lausanne) 2014, 5, 30. (24) Yuguchi, H.; Tanabe, A.; Hayashi, A.; Tanaka, Y.; Okuda, K.; Yamashita, Y.; Terai, Y.; Ohmichi, M. The Expression Status of G Protein-coupled Receptor GPR30 Is Associated with the Clinical Characteristics of Endometriosis. Endocr. Res. 2013, 38, 223–231. (25) Prossnitz, E. R.; Barton, M. The G-protein-coupled Estrogen Receptor GPER in Health and Disease. Nat Rev Endocrinol 2011, 7, 715–726. (26) Prossnitz, E. R.; Barton, M. Estrogen Biology: New Insights into GPER Function and Clinical Opportunities. Mol. Cell. Endocrinol. 2014. (27) Dennis, M. K.; Burai, R.; Ramesh, C.; Petrie, W. K.; Alcon, S. N.; Nayak, T. K.; Bologa, C. G.; Leitao, A.; Brailoiu, E.; Deliu, E.; Dun, N. J.; Sklar, L. A.; Hathaway, H. J.; Arterburn, J. B.; Oprea, T. I.; Prossnitz, E. R. In Vivo Effects of a GPR30 Antagonist. Nat. Chem. Biol. 2009, 5, 421–427. (28) Pietras, R. J.; Szego, C. M. Specific Binding Sites for Oestrogen at the Outer Surfaces of Isolated Endometrial Cells. Nature 1977, 265, 69–72. (29) Qiu, J.; Bosch, M. A.; Tobias, S. C.; Grandy, D. K.; Scanlan, T. S.; Ronnekleiv, O. K.; Kelly, M. J. Rapid Signaling of Estrogen in Hypothalamic Neurons Involves a Novel G-protein-coupled Estrogen Receptor That Activates Protein Kinase C. J. Neurosci. 2003, 23, 9529–9540. (30) Kampa, M.; Notas, G.; Pelekanou, V.; Troullinaki, M.; Andrianaki, M.; Azariadis, K.; Kampouri, E.; Lavrentaki, K.; Castanas, E. Early Membrane Initiated Transcriptional Effects of Estrogens in Breast Cancer Cells: First Pharmacological Evidence for a Novel Membrane Estrogen Receptor Element (ERx). Steroids 2012, 77, 959–967. (31) Kim, D.-K.; Ryu, D.; Koh, M.; Lee, M.-W.; Lim, D.; Kim, M.-J.; Kim, Y.-H.; Cho, W.-J.; Lee, C.-H.; Park, S. B.; Koo, S.-H.; Choi, H.-S. Orphan Nuclear Receptor Estrogen-related Receptor γ (ERRγ) Is Key Regulator of Hepatic Gluconeogenesis. J. Biol. Chem. 2012, 287, 21628–21639. (32) Sabo-Attwood, T.; Kroll, K. J.; Denslow, N. D. Differential Expression of Largemouth Bass (Micropterus Salmoides) Estrogen Receptor Isotypes Alpha, Beta, and Gamma by Estradiol. Mol. Cell. Endocrinol. 2004, 218, 107–118. (33) Evans, H. M. Endocrine Glands: Gonads, Pituitary, and Adrenals. Annual Review of Physiology 1939, 1, 577–652. (34) Gay, V. L.; Midgley, A. R., Jr; Niswender, G. D. Patterns of Gonadotrophin Secretion Associated with Ovulation. Fed. Proc. 1970, 29, 1880–1887. (35) Warnock, J. K.; Swanson, S. G.; Borel, R. W.; Zipfel, L. M.; Brennan, J. J.; ESTRATEST Clinical Study Group. Combined Esterified Estrogens and Methyltestosterone Versus Esterified Estrogens Alone in the Treatment of Loss of Sexual Interest in Surgically Menopausal Women. Menopause 2005, 12, 374–384. (36) Christensen, A.; Dewing, P.; Micevych, P. Membrane-initiated Estradiol Signaling Induces Spinogenesis Required for Female Sexual Receptivity. J. Neurosci. 2011, 31, 17583–17589. (37) Cowie, A. T.; Folley, S. J.; Malpress, F. H.; Richardson, K. C. Studies on the Induction of Mammary Growth and Lactation in Virgin Goats. J. Endocrinol. 1951, 7, xxi–xxii. (38) Pincus, G.; Garcia, C. R.; Rock, J.; Paniagua, M.; Pendleton, A.; Laraque, F.; Nicolas, R.; Borno, R.; Pean, V. Effectiveness of an Oral Contraceptive; Effects of a Progestin-estrogen Combination Upon Fertility, Menstrual Phenomena, and Health. Science 1959, 130, 81–83. (39) Charlot, M.; Béatrix, O.; Chateau, F.; Dubuisson, J.; Golfier, F.; Valette, P. J.; Réty, F. Pathologies of the Male Breast. Diagn. Interv. Imaging 2013, 94, 26–37. (40) Boillot, B.; Teklali, Y.; Moog, R.; Droupy, S. [Penile congenital abnormalities]. Prog. Urol. 2013, 23, 664–673.

128

(41) Blaschko, S. D.; Cunha, G. R.; Baskin, L. S. Molecular Mechanisms of External Genitalia Development. Differentiation 2012, 84, 261–268. (42) Habert, R.; Delbes, G.; Duquenne, C.; Livera, G.; Levacher, C. [Effects of estrogens on the development of the testis during fetal and neonatal life]. Gynecol Obstet Fertil 2006, 34, 970–977. (43) Eisenberg, M. L.; Hsieh, M. H.; Walters, R. C.; Krasnow, R.; Lipshultz, L. I. The Relationship Between Anogenital Distance, Fatherhood, and Fertility in Adult Men. PLoS ONE 2011, 6, e18973. (44) Carreau, S.; Bouraima-Lelong, H.; Delalande, C. Estrogen, a Female Hormone Involved in Spermatogenesis. Adv. Med. Sci. 2012, 57, 31–36. (45) Corona, G.; Jannini, E. A.; Vignozzi, L.; Rastrelli, G.; Maggi, M. The Hormonal Control of Ejaculation. Nat. Rev. Urol. 2012, 9, 508–519. (46) Sinclair, A. H.; Berta, P.; Palmer, M. S.; Hawkins, J. R.; Griffiths, B. L.; Smith, M. J.; Foster, J. W.; Frischauf, A. M.; Lovell-Badge, R.; Goodfellow, P. N. A Gene from the Human Sex-determining Region Encodes a Protein with Homology to a Conserved DNA-binding Motif. Nature 1990, 346, 240–244. (47) Yang, J. H.; Menshenina, J.; Cunha, G. R.; Place, N.; Baskin, L. S. Morphology of Mouse External Genitalia: Implications for a Role of Estrogen in Sexual Dimorphism of the Mouse Genital Tubercle. J. Urol. 2010, 184, 1604–1609. (48) Hsieh, M. H.; Grantham, E. C.; Liu, B.; Macapagal, R.; Willingham, E.; Baskin, L. S. In Utero Exposure to Benzophenone-2 Causes Hypospadias through an Estrogen Receptor Dependent Mechanism. J. Urol. 2007, 178, 1637–1642. (49) Trémollières, F. [What patients need to know about the risk of bone fracture and its prevention]. J Gynecol Obstet Biol Reprod (Paris) 2012, 41, F20–27. (50) Rodan, G. A.; Martin, T. J. Therapeutic Approaches to Bone Diseases. Science 2000, 289, 1508– 1514. (51) Svoboda, M.; Hamilton, G.; Thalhammer, T. Steroid Hormone Metabolizing Enzymes in Benign and Malignant Human Bone Tumors. Expert Opin. Drug Metab. Toxicol. 2010, 6, 427–437. (52) Kendall, B.; Eston, R. Exercise-induced Muscle Damage and the Potential Protective Role of Estrogen. Sports Med. 2002, 32, 103–123. (53) Enns, D. L.; Tiidus, P. M. The Influence of Estrogen on Skeletal Muscle: Sex Matters. Sports Med. 2010, 40, 41–58. (54) Velders, M.; Diel, P. How Sex Hormones Promote Skeletal Muscle Regeneration. Sports Med. 2013, 43, 1089–1100. (55) Barros, R. P. A.; Gustafsson, J.-Å. Estrogen Receptors and the Metabolic Network. Cell Metab. 2011, 14, 289–299. (56) Ropero, A. B.; Pang, Y.; Alonso-Magdalena, P.; Thomas, P.; Nadal, A. Role of ERβ and GPR30 in the Endocrine Pancreas: A Matter of Estrogen Dose. Steroids 2012, 77, 951–958. (57) Barros, R. P. A.; Gabbi, C.; Morani, A.; Warner, M.; Gustafsson, J.-A. Participation of ERalpha and ERbeta in Glucose Homeostasis in Skeletal Muscle and White Adipose Tissue. Am. J. Physiol. Endocrinol. Metab. 2009, 297, E124–E133. (58) Homma, H. Estrogen Suppresses Transcription of Lipoprotein Lipase Gene. Existence of a Unique Estrogen Response Element on the Lipoprotein Lipase Promoter. J. Biol. Chem. 2000, 275, 11404– 11411. (59) Gao, H. Long-Term Administration of Estradiol Decreases Expression of Hepatic Lipogenic Genes and Improves Insulin Sensitivity in Ob/ob Mice: A Possible Mechanism Is through Direct Regulation of Signal Transducer and Activator of Transcription 3. Mol. Endocrinol. 2005, 20, 1287–1299. (60) Faulds, M. H.; Zhao, C.; Dahlman-Wright, K.; Gustafsson, J.-Å. The Diversity of Sex Steroid Action: Regulation of Metabolism by Estrogen Signaling. J. Endocrinol. 2012, 212, 3–12. (61) Fuente-Martin, E.; Garcia-Caceres, C.; Morselli, E.; Clegg, D. J.; Chowen, J. A.; Finan, B.; Brinton, R. D.; Tschöp, M. H. Estrogen, Astrocytes and the Neuroendocrine Control of Metabolism. Rev. Endocr. Metab. Disord. 2013, 14, 331–338.

129 (62) Goodman-Gruen, D.; Barrett-Connor, E. Sex Differences in Measures of Body Fat and Body Fat Distribution in the Elderly. Am. J. Epidemiol. 1996, 143, 898–906. (63) Hogan, A. M.; Collins, D.; Baird, A. W.; Winter, D. C. Estrogen and Its Role in Gastrointestinal Health and Disease. Int. J. Colorectal Dis. 2009, 24, 1367–1375. (64) Faas, M. M.; Melgert, B. N.; de Vos, P. A Brief Review on How Pregnancy and Sex Hormones Interfere with Taste and Food Intake. Chemosens. Percept. 2010, 3, 51–56. (65) Sinchak, K.; Wagner, E. J. Estradiol Signaling in the Regulation of Reproduction and Energy Balance. Front. Neuroendocrinol. 2012, 33, 342–363. (66) Bartley, E. J.; Fillingim, R. B. Sex Differences in Pain: a Brief Review of Clinical and Experimental Findings. Br. J. Anaesth. 2013, 111, 52–58. (67) Amandusson, A.; Blomqvist, A. Estrogenic Influences in Pain Processing. Front. Neuroendocrinol. 2013, 34, 329-349. (68) Stening, K.; Eriksson, O.; Wahren, L.; Berg, G.; Hammar, M.; Blomqvist, A. Pain Sensations to the Cold Pressor Test in Normally Menstruating Women: Comparison with Men and Relation to Menstrual Phase and Serum Sex Steroid Levels. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2007, 293, R1711–1716. (69) Brinton, R. D. Estrogen-induced Plasticity from Cells to Circuits: Predictions for Cognitive Function. Trends Pharmacol. Sci. 2009, 30, 212–222. (70) Srivastava, D. P.; Waters, E. M.; Mermelstein, P. G.; Kramár, E. A.; Shors, T. J.; Liu, F. Rapid Estrogen Signaling in the Brain: Implications for the Fine-tuning of Neuronal Circuitry. J. Neurosci. 2011, 31, 16056–16063. (71) Weber, M. T.; Maki, P. M.; McDermott, M. P. Cognition and Mood in Perimenopause: A Systematic Review and Meta-analysis. J. Steroid Biochem. Mol. Biol. 2013, 142, 90-98. (72) Simpkins, J. W.; Singh, M.; Brock, C.; Etgen, A. M. Neuroprotection and Estrogen Receptors. Neuroendocrinology 2012, 96, 119–130. (73) Nadkarni, S.; McArthur, S. Oestrogen and Immunomodulation: New Mechanisms That Impact on Peripheral and Central Immunity. Curr. Opin. Pharmacol. 2013, 13, 576–581. (74) Hernández-Bello, R.; Nava-Castro, K.; Muñiz-Hernández, S.; Nava-Luna, P.; Trejo-Sánchez, I.; Tiempos-Guzmán, N.; Mendoza-Rodríguez, Y.; Morales-Montor, J. Beyond the Reproductive Effect of Sex Steroids: Their Role During Immunity to Helminth Parasite Infections. Mini Rev. Med. Chem. 2012, 12, 1071–1080. (75) Kovats, S. Estrogen Receptors Regulate an Inflammatory Pathway of Dendritic Cell Differentiation: Mechanisms and Implications for Immunity. Horm. Behav. 2012, 62, 254–262. (76) Ghosh, M.; Rodriguez-Garcia, M.; Wira, C. R. The Immune System in Menopause: Pros and Cons of Hormone Therapy. J. Steroid Biochem. Mol. Biol. 2013, 142, 171-175. (77) Nadkarni, S.; Cooper, D.; Brancaleone, V.; Bena, S.; Perretti, M. Activation of the Annexin A1 Pathway Underlies the Protective Effects Exerted by Estrogen in Polymorphonuclear Leukocytes. Arterioscler. Thromb. Vasc. Biol. 2011, 31, 2749–2759. (78) Société canadienne du cancer, http://www.cancer.ca/fr-ca/cancer-information/cancer- type/breast/statistics/?region=qc, consulté le 30 novembre 2014. (79) Gustafsson, J.-A. What Pharmacologists Can Learn from Recent Advances in Estrogen Signalling. Trends Pharmacol. Sci. 2003, 24, 479–485. (80) Statistiques Canadiennes Sur Le Cancer 2013, Sujet Particulier: Le Cancer Du Foie; Société canadienne du cancer, 2013. (81) Takagi, K.; Moriya, T.; Kurosumi, M.; Oka, K.; Miki, Y.; Ebata, A.; Toshima, T.; Tsunekawa, S.; Takei, H.; Hirakawa, H.; Ishida, T.; Hayashi, S.-I.; Kurebayashi, J.; Sasano, H.; Suzuki, T. Intratumoral Estrogen Concentration and Expression of Estrogen-induced Genes in Male Breast Carcinoma: Comparison with Female Breast Carcinoma. Horm. Cancer 2013, 4, 1–11. (82) Yang, G.; Nowsheen, S.; Aziz, K.; Georgakilas, A. G. Toxicity and Adverse Effects of Tamoxifen and Other Anti-estrogen Drugs. Pharmacol. Ther. 2013, 139, 392–404.

130

(83) Glass, A. G.; Lacey, J. V.; Carreon, J. D.; Hoover, R. N. Breast Cancer Incidence, 1980 2006: Combined Roles of Menopausal Hormone Therapy, Screening Mammography, and Estrogen Receptor Status. JNCI Journal of the National Cancer Institute 2007, 99, 1152–1161. (84) Beatson, G. On the Treatment of Inoperable Cases of Carcinoma of the Mamma: Suggestions for a New Method of Treatment, with Illustrative Cases. The Lancet 1896, 148, 162–165. (85) Sestak, I. Preventative therapies for healthy women at high risk of breast cancer. Cancer Manag. Res. 2014, 17, 423-430. (86) Scaling, A. L.; Prossnitz, E. R.; Hathaway, H. J. GPER Mediates Estrogen-Induced Signaling and Proliferation in Human Breast Epithelial Cells and Normal and Malignant Breast. Horm Cancer 2014. (87) Honma, N.; Saji, S.; Hirose, M.; Horiguchi, S.; Kuroi, K.; Hayashi, S.; Utsumi, T.; Harada, N. Sex Steroid Hormones in Pairs of Tumor and Serum from Breast Cancer Patients and Pathobiological Role of Androstene-3β, 17β-diol. Cancer Sci. 2011, 102, 1848–1854. (88) Henderson, B. E.; Feigelson, H. S. Hormonal Carcinogenesis. Carcinogenesis 2000, 21, 427–433. (89) Zagouri, F.; Bozas, G.; Kafantari, E.; Tsiatas, M.; Nikitas, N.; Dimopoulos, M.-A.; Papadimitriou, C. A. Endometrial Cancer: What Is New in Adjuvant and Molecularly Targeted Therapy? Obstet. Gynecol. Int. 2010, 2010, 749579. (90) Lee, W.-L.; Yen, M.-S.; Chao, K.-C.; Yuan, C.-C.; Ng, H.-T.; Chao, H.-T.; Lee, F.-K.; Wang, P.-H. Hormone Therapy for Patients with Advanced or Recurrent Endometrial Cancer. J. Chin. Med. Assoc. 2014, 77, 221-226. (91) Chung, S.-H.; Shin, M. K.; Korach, K. S.; Lambert, P. F. Requirement for Stromal Estrogen Receptor Alpha in Cervical Neoplasia. Horm. Cancer 2013, 4, 50–59. (92) Arbeit, J. M.; Howley, P. M.; Hanahan, D. Chronic Estrogen-induced Cervical and Vaginal Squamous Carcinogenesis in Human Papillomavirus Type 16 Transgenic Mice. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 2930–2935. (93) Spurgeon, M. E.; Chung, S.-H.; Lambert, P. F. Recurrence of Cervical Cancer in Mice after Selective Estrogen Receptor Modulator Therapy. Am. J. Pathol. 2014, 184, 530–540. (94) Kinch, R. A. Diethylstilbestrol in Pregnancy: An Update. Can Med Assoc J 1982, 127, 812–813. (95) Merino, M. J. Vaginal Cancer: The Role of Infectious and Environmental Factors. Am. J. Obstet. Gynecol. 1991, 165, 1255–1262. (96) Høgdall, E. V. S.; Christensen, L.; Høgdall, C. K.; Blaakaer, J.; Gayther, S.; Jacobs, I. J.; Christensen, I. J.; Kjaer, S. K. Prognostic Value of Estrogen Receptor and Progesterone Receptor Tumor Expression in Danish Ovarian Cancer Patients: From the “MALOVA” Ovarian Cancer Study. Oncol. Rep. 2007, 18, 1051–1059. (97) Yokoyama, Y. Recurrent Epithelial Ovarian Cancer and Hormone Therapy. World Journal of Clinical Cases 2013, 1, 187. (98) Modugno, F.; Laskey, R.; Smith, A. L.; Andersen, C. L.; Haluska, P.; Oesterreich, S. Hormone Response in Ovarian Cancer: Time to Reconsider as a Clinical Target? Endocr. Rel. Cancer 2012, 19, R255–R279. (99) Liu, H.; Yan, Y.; Wen, H.; Jiang, X.; Cao, X.; Zhang, G.; Liu, G. A Novel Estrogen Receptor GPER Mediates Proliferation Induced by 17β-estradiol and Selective GPER Agonist G-1 in Estrogen Receptor α (ERα)-negative Ovarian Cancer Cells. Cell Biol. Int. 2014, 38, 631-638. (100) Prins, G. S.; Korach, K. S. The Role of Estrogens and Estrogen Receptors in Normal Prostate Growth and Disease. Steroids 2008, 73, 233–244. (101) Park, I. I.; Zhang, Q.; Liu, V.; Kozlowski, J. M.; Zhang, J.; Lee, C. 17Beta-estradiol at Low Concentrations Acts through Distinct Pathways in Normal Versus Benign Prostatic Hyperplasia- derived Prostate Stromal Cells. Endocrinology 2009, 150, 4594–4605. (102) Krieg, M.; Klötzl, G.; Kaufmann, J.; Voigt, K. D. Stroma of Human Benign Prostatic Hyperplasia: Preferential Tissue for Androgen Metabolism and Oestrogen Binding. Acta Endocrinol. 1981, 96, 422–432. (103) Bonkhoff, H.; Berges, R. The Evolving Role of Oestrogens and Their Receptors in the Development and Progression of Prostate Cancer. Eur. Urol. 2009, 55, 533–542.

131 (104) D’Errico, I.; Moschetta, A. Nuclear Receptors, Intestinal Architecture and Colon Cancer: An Intriguing Link. Cell. Mol. Life Sci. 2008, 65, 1523–1543. (105) Konstantinopoulos, P. A.; Kominea, A.; Vandoros, G.; Sykiotis, G. P.; Andricopoulos, P.; Varakis, I.; Sotiropoulou-Bonikou, G.; Papavassiliou, A. G. Oestrogen Receptor Beta (ERbeta) Is Abundantly Expressed in Normal Colonic Mucosa, but Declines in Colon Adenocarcinoma Paralleling the Tumour’s Dedifferentiation. Eur. J. Cancer 2003, 39, 1251–1258. (106) Kennelly, R.; Kavanagh, D. O.; Hogan, A. M.; Winter, D. C. Oestrogen and the Colon: Potential Mechanisms for Cancer Prevention. Lancet Oncol. 2008, 9, 385–391. (107) Hogan, A. M.; Collins, D.; Baird, A. W.; Winter, D. C. Estrogen and Gastrointestinal Malignancy. Mol. Cell. Endocrinol. 2009, 307, 19–24. (108) Chute, J. P.; Ross, J. R.; McDonnell, D. P. Minireview: Nuclear Receptors, Hematopoiesis, and Stem Cells. Mol. Endocrinol. 2010, 24, 1–10. (109) Kabat, G. C.; Etgen, A. M.; Rohan, T. E. Do Steroid Hormones Play a Role in the Etiology of Glioma? Cancer Epidemiol. Biomarkers Prev. 2010, 19, 2421–2427. (110) Liu, C.; Zhang, Y.; Zhang, K.; Bian, C.; Zhao, Y.; Zhang, J. Expression of Estrogen Receptors, Androgen Receptor and Steroid Receptor Coactivator-3 Is Negatively Correlated to the Differentiation of Astrocytic Tumors. Cancer Epidemiol. 2014. (111) Michaud, D. S.; Gallo, V.; Schlehofer, B.; Tjønneland, A.; Olsen, A.; Overvad, K.; Dahm, C. C.; Kaaks, R.; Lukanova, A.; Boeing, H.; Schütze, M.; Trichopoulou, A.; Bamia, C.; Kyrozis, A.; Sacerdote, C.; Agnoli, C.; Palli, D.; Tumino, R.; Mattiello, A.; Bueno-de-Mesquita, H. B.; Ros, M. M.; Peeters, P. H. M.; van Gils, C. H.; Lund, E.; Bakken, K.; Gram, I. T.; Barricarte, A.; Navarro, C.; Dorronsoro, M.; Sánchez, M. J.; Rodríguez, L.; Duell, E. J.; Hallmans, G.; Melin, B. S.; Manjer, J.; Borgquist, S.; Khaw, K.-T.; Wareham, N.; Allen, N. E.; Tsilidis, K. K.; Romieu, I.; Rinaldi, S.; Vineis, P.; Riboli, E. Reproductive Factors and Exogenous Hormone Use in Relation to Risk of Glioma and Meningioma in a Large European Cohort Study. Cancer Epidemiol. Biomarkers Prev. 2010, 19, 2562–2569. (112) Shi, L.; Feng, Y.; Lin, H.; Ma, R.; Cai, X. Role of Estrogen in Hepatocellular Carcinoma: Is Inflammation the Key? J. Transl. Med. 2014, 12, 93. (113) Siegfried, J. M.; Stabile, L. P. Estrogenic Steroid Hormones in Lung Cancer. Seminars in Oncology 2014, 41, 5–16. (114) Yao, P.-L.; Gonzalez, F. J.; Peters, J. M. Targeting Estrogen Receptor-β for the Prevention of Nonmelanoma Skin Cancer. Cancer Prev. Res. (Phila) 2014, 7, 182–185. (115) Dohi, O.; Hatori, M.; Suzuki, T.; Ono, K.; Hosaka, M.; Akahira, J.; Miki, Y.; Nagasaki, S.; Itoi, E.; Sasano, H. Sex Steroid Receptors Expression and Hormone-induced Cell Proliferation in Human Osteosarcoma. Cancer Sci. 2008, 99, 518–523. (116) Eskenazi, B.; Warner, M. L. Epidemiology of Endometriosis. Obstet. Gynecol. Clin. North Am. 1997, 24, 235–258. (117) De Graaff, A. A.; D’Hooghe, T. M.; Dunselman, G. A. J.; Dirksen, C. D.; Hummelshoj, L.; WERF EndoCost Consortium; Simoens, S. The Significant Effect of Endometriosis on Physical, Mental and Social Wellbeing: Results from an International Cross-sectional Survey. Hum. Reprod. 2013, 28, 2677–2685. (118) D’Hooghe, T. M.; Debrock, S.; Hill, J. A.; Meuleman, C. Endometriosis and Subfertility: Is the Relationship Resolved? Semin. Reprod. Med. 2003, 21, 243–254. (119) Pavone, M. E.; Bulun, S. E. Aromatase Inhibitors for the Treatment of Endometriosis. Fertil. Steril. 2012, 98, 1370–1379. (120) Soares, S. R.; Martínez-Varea, A.; Hidalgo-Mora, J. J.; Pellicer, A. Pharmacologic Therapies in Endometriosis: a Systematic Review. Fertil. Steril. 2012, 98, 529–555. (121) Delvoux, B.; Groothuis, P.; D’Hooghe, T.; Kyama, C.; Dunselman, G.; Romano, A. Increased Production of 17beta-estradiol in Endometriosis Lesions Is the Result of Impaired Metabolism. J. Clin. Endocrinol. Metab. 2009, 94, 876–883.

132

(122) Plante, B. J.; Lessey, B. A.; Taylor, R. N.; Wang, W.; Bagchi, M. K.; Yuan, L.; Scotchie, J.; Fritz, M. A.; Young, S. L. G Protein-coupled Estrogen Receptor (GPER) Expression in Normal and Abnormal Endometrium. Reprod. Sci. 2012, 19, 684–693. (123) Bulun, S. E.; Monsavais, D.; Pavone, M. E.; Dyson, M.; Xue, Q.; Attar, E.; Tokunaga, H.; Su, E. J. Role of Estrogen Receptor-β in Endometriosis. Semin. Reprod. Med. 2012, 30, 39–45. (124) Hu, X.; Zhou, Y.; Feng, Q.; Wang, R.; Su, L.; Long, J.; Wei, B. Association of Endometriosis Risk and Genetic Polymorphisms Involving Biosynthesis of Sex Steroids and Their Receptors: An Updating Meta-analysis. Eur. J. Obstet. Gynecol. Reprod. Biol. 2012, 164, 1–9. (125) Miranda, A. R.; Ue, A. P. F. de; Sabbag, D. V.; Furlani, W. de J.; Souza, P. K. de; Rotta, O. Hereditary Angioedema Type III (estrogen-dependent) Report of Three Cases and Literature Review*. Anais Brasileiros de Dermatologia 2013, 88, 578–584. (126) Riedl, M. A. Hereditary Angioedema with Normal C1-INH (HAE Type III). J. Allergy Clin. Immunol. Pract. 2013, 1, 427–432. (127) Chan, K. L.; Mok, C. C. Development of Systemic Lupus Erythematosus in a Male-to-female Transsexual: The Role of Sex Hormones Revisited. Lupus 2013, 22, 1399–1402. (128) Mok, C. C.; Ying, S. K. Y.; Ma, K. M.; Wong, C. K. Effect of Raloxifene on Disease Activity and Vascular Biomarkers in Patients with Systemic Lupus Erythematosus: Subgroup Analysis of a Double-blind Randomized Controlled Trial. Lupus 2013, 22, 1470–1478. (129) Guéry, J.-C. Estrogens and Inflammatory Autoimmune Diseases. Joint Bone Spine 2012, 79, 560– 562. (130) Yager, J. D.; Davidson, N. E. Estrogen Carcinogenesis in Breast Cancer. N. Engl. J. Med. 2006, 354, 270–282. (131) Hillier, S. [The respective roles of gonadotrophins on follicular growth and oocyte maturation]. J. Gynecol. Obstet. Biol. Reprod. (Paris) 2004, 33, 3S11–14. (132) Thompson, I. R.; Kaiser, U. B. GnRH Pulse Frequency-dependent Differential Regulation of LH and FSH Gene Expression. Mol. Cell. Endocrinol. 2014, 385, 28–35. (133) Tammela, T. L. J. Endocrine Prevention and Treatment of Prostate Cancer. Mol. Cell. Endocrinol. 2012, 360, 59–67. (134) DiVasta, A. D.; Laufer, M. R. The Use of Gonadotropin Releasing Hormone Analogues in Adolescent and Young Patients with Endometriosis. Curr. Opin. Obstet. Gynecol. 2013, 25, 287–292. (135) Xiao, J.; Chen, S.; Zhang, C.; Chang, S. Effectiveness of GnRH Antagonist in the Treatment of Patients with Polycystic Ovary Syndrome Undergoing IVF: a Systematic Review and Meta Analysis. Gynecol. Endocrinol. 2013, 29, 187–191. (136) Limonta, P.; Montagnani Marelli, M.; Mai, S.; Motta, M.; Martini, L.; Moretti, R. M. GnRH Receptors in Cancer: From Cell Biology to Novel Targeted Therapeutic Strategies. Endocr. Rev. 2012, 33, 784– 811. (137) Kol, S.; Humaidan, P. GnRH Agonist Triggering: Recent Developments. Reprod. Biomed. Online 2013, 26, 226–230. (138) Zhu, H.; Garcia, J. A. Targeting the Adrenal Gland in Castration-resistant Prostate Cancer: a Case for Orteronel, a Selective CYP-17 17,20-lyase Inhibitor. Curr Oncol Rep 2013, 15, 105–112. (139) Hoy, S. M. Abiraterone Acetate: a Review of Its Use in Patients with Metastatic Castration-resistant Prostate Cancer. Drugs 2013, 73, 2077–2091. (140) Basset, V.; Flamand, V.; Crouzet, S.; Ploussard, G. [Choice of new drugs in castration-resistant prostate cancer: predictive factors and effectiveness assessment]. Prog. Urol. 2013, 23 Suppl 1, S44–48. (141) Samojlik, E.; Santen, R. J.; Wells, S. A. Adrenal Suppression with Aminoglutethimide. II. Differential Effects of Aminoglutethimide on Plasma Androstenedione and Estrogen Levels. J. Clin. Endocrinol. Metab. 1977, 45, 480–487. (142) Berry, J. Are All Aromatase Inhibitors the Same? A Review of Controlled Clinical Trials in Breast Cancer. Clin. Ther. 2005, 27, 1671–1684.

133 (143) Astra Zeneca. http://www.astrazeneca- annualreports.com/documents/2010/therapy_review_area_factsheets/oncology.pdf; 2014. (144) Reed, M. J.; Purohit, A.; Woo, L. W. L.; Newman, S. P.; Potter, B. V. L. Steroid Sulfatase: Molecular Biology, Regulation, and Inhibition. Endocr. Rev. 2005, 26, 171–202. (145) Williams, S. J. Sulfatase Inhibitors: a Patent Review. Expert Opin Ther Pat 2013, 23, 79–98. (146) Poirier, D. 17beta-Hydroxysteroid Dehydrogenase Inhibitors: a Patent Review. Expert Opin. Ther. Patent 2010, 20, 1123–1145. (147) Moeller, G.; Adamski, J. Integrated View on 17beta-hydroxysteroid Dehydrogenases. Mol. Cell. Endocrinol. 2009, 301, 7–19. (148) Samson, M.; Labrie, F.; Luu-The, V. Specific Estradiol Biosynthetic Pathway in Choriocarcinoma (JEG-3) Cell Line. J. Steroid Biochem. Mol. Biol. 2009, 116, 154–159. (149) Simard, J.; Vincent, A.; Duchesne, R.; Labrie, F. Full Oestrogenic Activity of C19-delta 5 Adrenal Steroids in Rat Pituitary Lactotrophs and Somatotrophs. Mol. Cell. Endocrinol. 1988, 55, 233–242. (150) Aka, J. A.; Mazumdar, M.; Chen, C.-Q.; Poirier, D.; Lin, S.-X. 17beta-hydroxysteroid Dehydrogenase Type 1 Stimulates Breast Cancer by Dihydrotestosterone Inactivation in Addition to Estradiol Production. Mol. Endocrinol. 2010, 24, 832–845. (151) Ishibashi, O.; Ohkuchi, A.; Ali, M. M.; Kurashina, R.; Luo, S.-S.; Ishikawa, T.; Takizawa, T.; Hirashima, C.; Takahashi, K.; Migita, M.; Ishikawa, G.; Yoneyama, K.; Asakura, H.; Izumi, A.; Matsubara, S.; Takeshita, T.; Takizawa, T. Hydroxysteroid (17-β) Dehydrogenase 1 Is Dysregulated by miR-210 and miR-518c That Are Aberrantly Expressed in Preeclamptic Placentas: a Novel Marker for Predicting Preeclampsia. Hypertension 2012, 59, 265–273. (152) Bagi, C. M.; Wood, J.; Wilkie, D.; Dixon, B. Effect of 17beta-hydroxysteroid Dehydrogenase Type 2 Inhibitor on Bone Strength in Ovariectomized Cynomolgus Monkeys. J. Musculoskelet. Neuronal Interact. 2008, 8, 267–280. (153) Luu-The, V.; Labrie, F. The Intracrine Sex Steroid Biosynthesis Pathways. Prog. Brain Res. 2010, 181, 177–192. (154) George, M. M.; New, M. I.; Ten, S.; Sultan, C.; Bhangoo, A. The Clinical and Molecular Heterogeneity of 17βHSD-3 Enzyme Deficiency. Horm. Res. Paediatr. 2010, 74, 229–240. (155) Marchais-Oberwinkler, S.; Henn, C.; Möller, G.; Klein, T.; Negri, M.; Oster, A.; Spadaro, A.; Werth, R.; Wetzel, M.; Xu, K.; Frotscher, M.; Hartmann, R. W.; Adamski, J. 17β-Hydroxysteroid Dehydrogenases (17β-HSDs) as Therapeutic Targets: Protein Structures, Functions, and Recent Progress in Inhibitor Development. J. Steroid Biochem. Mol. Biol. 2011, 125, 66–82. (156) Byrns, M. C.; Jin, Y.; Penning, T. M. Inhibitors of Type 5 17β-hydroxysteroid Dehydrogenase (AKR1C3): Overview and Structural Insights. J. Steroid Biochem. Mol. Biol. 2011, 125, 95–104. (157) Morley, J. E.; Farr, S. A. The Role of Amyloid-beta in the Regulation of Memory. Biochem. Pharmacol. 2014, 88, 479–485. (158) Purohit, A.; Tutill, H. J.; Day, J. M.; Chander, S. K.; Lawrence, H. R.; Allan, G. M.; Fischer, D. S.; Vicker, N.; Newman, S. P.; Potter, B. V. L.; Reed, M. J. The Regulation and Inhibition of 17beta- hydroxysteroid Dehydrogenase in Breast Cancer. Mol. Cell. Endocrinol. 2006, 248, 199–203. (159) Vihko, P.; Härkönen, P.; Soronen, P.; Törn, S.; Herrala, A.; Kurkela, R.; Pulkka, A.; Oduwole, O.; Isomaa, V. 17 Beta-hydroxysteroid Dehydrogenases--their Role in Pathophysiology. Mol. Cell. Endocrinol. 2004, 215, 83–88. (160) Shao, R.; Cao, S.; Wang, X.; Feng, Y.; Billig, H. The Elusive and Controversial Roles of Estrogen and Progesterone Receptors in Human Endometriosis. Am. J. Transl. Res. 2014, 6, 104–113. (161) Sjöström, M.; Hartman, L.; Grabau, D.; Fornander, T.; Malmström, P.; Nordenskjöld, B.; Sgroi, D. C.; Skoog, L.; Stål, O.; Leeb-Lundberg, L. M. F.; Fernö, M. Lack of G Protein-coupled Estrogen Receptor (GPER) in the Plasma Membrane Is Associated with Excellent Long-term Prognosis in Breast Cancer. Breast Cancer Res. Treat. 2014, 145, 61–71. (162) Lin, B. C.; Suzawa, M.; Blind, R. D.; Tobias, S. C.; Bulun, S. E.; Scanlan, T. S.; Ingraham, H. A. Stimulating the GPR30 Estrogen Receptor with a Novel Tamoxifen Analogue Activates SF-1 and Promotes Endometrial Cell Proliferation. Cancer Res. 2009, 69, 5415–5423.

134

(163) Burstein, H. J.; Prestrud, A. A.; Seidenfeld, J.; Anderson, H.; Buchholz, T. A.; Davidson, N. E.; Gelmon, K. E.; Giordano, S. H.; Hudis, C. A.; Malin, J.; Mamounas, E. P.; Rowden, D.; Solky, A. J.; Sowers, M. R.; Stearns, V.; Winer, E. P.; Somerfield, M. R.; Griggs, J. J.; American Society of Clinical Oncology. American Society of Clinical Oncology Clinical Practice Guideline: Update on Adjuvant Endocrine Therapy for Women with Hormone Receptor-positive Breast Cancer. J. Clin. Oncol. 2010, 28, 3784–3796. (164) Chlebowski, R. T. Current Concepts in Breast Cancer Chemoprevention. Pol. Arch. Med. Wewn. 2014, 124, 191–199. (165) Chanplakorn, N.; Chanplakorn, P.; Suzuki, T.; Ono, K.; Chan, M. S. M.; Miki, Y.; Saji, S.; Ueno, T.; Toi, M.; Sasano, H. Increased Estrogen Sulfatase (STS) and 17beta-hydroxysteroid Dehydrogenase Type 1(17beta-HSD1) Following Neoadjuvant Aromatase Inhibitor Therapy in Breast Cancer Patients. Breast Cancer Res. Treat. 2010, 120, 639–648. (166) Messinger, J.; Husen, B.; Koskimies, P.; Hirvelä, L.; Kallio, L.; Saarenketo, P.; Thole, H. Estrone C15 derivatives—A New Class of 17β-hydroxysteroid Dehydrogenase Type 1 Inhibitors. Mol. Cel. Endocrinol. 2009, 301, 216–224. (167) Saloniemi, T.; Järvensivu, P.; Koskimies, P.; Jokela, H.; Lamminen, T.; Ghaem-Maghami, S.; Dina, R.; Damdimopoulou, P.; Mäkelä, S.; Perheentupa, A.; Kujari, H.; Brosens, J.; Poutanen, M. Novel Hydroxysteroid (17β) Dehydrogenase 1 Inhibitors Reverse Estrogen-Induced Endometrial Hyperplasia in Transgenic Mice. Am. J. Pathol. 2010, 176, 1443–1451. (168) Day, J. M.; Foster, P. A.; Tutill, H. J.; Parsons, M. F. C.; Newman, S. P.; Chander, S. K.; Allan, G. M.; Lawrence, H. R.; Vicker, N.; Potter, B. V. L.; Reed, M. J.; Purohit, A. 17beta-hydroxysteroid Dehydrogenase Type 1, and Not Type 12, Is a Target for Endocrine Therapy of Hormone-dependent Breast Cancer. Int. J. Cancer 2008, 122, 1931–1940. (169) Spadaro, A.; Negri, M.; Marchais-Oberwinkler, S.; Bey, E.; Frotscher, M. Hydroxybenzothiazoles as New Nonsteroidal Inhibitors of 17β-Hydroxysteroid Dehydrogenase Type 1 (17β-HSD1). PLoS ONE 2012, 7, e29252. (170) Brožič, P.; Kocbek, P.; Sova, M.; Kristl, J.; Martens, S.; Adamski, J.; Gobec, S.; Lanišnik Rižner, T. Flavonoids and Cinnamic Acid Derivatives as Inhibitors of 17β-hydroxysteroid Dehydrogenase Type 1. Mol. Cel. Endocrinol. 2009, 301, 229–234. (171) Starčević, š.; Kocbek, P.; Hribar, G.; Lanišnik Rižner, T.; Gobec, S. Biochemical and Biological Evaluation of Novel Potent Coumarin Inhibitor of 17β-HSD Type 1. Chemico-Biological Interactions 2011, 191, 60–65. (172) Henn, C.; Einspanier, A.; Marchais-Oberwinkler, S.; Frotscher, M.; Hartmann, R. W. Lead Optimization of 17β-HSD1 Inhibitors of the (hydroxyphenyl)naphthol Sulfonamide Type for the Treatment of Endometriosis. J. Med. Chem. 2012, 55, 3307–3318. (173) Poirier, D. Contribution to the Development of Inhibitors of 17β-hydroxysteroid Dehydrogenase Types 1 and 7: Key Tools for Studying and Treating Estrogen-dependent Diseases. J. Steroid Biochem. Mol. Biol. 2011, 125, 83–94. (174) Laplante, Y.; Cadot, C.; Fournier, M.-A.; Poirier, D. Estradiol and Estrone C-16 Derivatives as Inhibitors of Type 1 17beta-hydroxysteroid Dehydrogenase: Blocking of ER+ Breast Cancer Cell Proliferation Induced by Estrone. Bioorg. Med. Chem. 2008, 16, 1849–1860. (175) Ayan, D.; Maltais, R.; Roy, J.; Poirier, D. A New Nonestrogenic Steroidal Inhibitor of 17β- hydroxysteroid Dehydrogenase Type I Blocks the Estrogen-dependent Breast Cancer Tumor Growth Induced by Estrone. Mol. Cancer Ther. 2012, 11, 2096–2104. (176) Maltais, R.; Ayan, D.; Poirier, D. Crucial Role of 3-Bromoethyl in Removing the Estrogenic Activity of 17β-HSD1 Inhibitor 16β-(m-Carbamoylbenzyl)estradiol. ACS Med. Chem. Lett. 2011, 2, 678–681. (177) Zorn, J. A.; Wells, J. A. Turning Enzymes ON with Small Molecules. Nat. Chem. Biol. 2010, 6, 179– 188. (178) Dittenhafer-Reed, K. E.; Feldman, J. L.; Denu, J. M. Catalysis and Mechanistic Insights into Sirtuin Activation. Chembiochem 2011, 12, 281–289.

135 (179) Bishop, A. C.; Chen, V. L. Brought to Life: Targeted Activation of Enzyme Function with Small Molecules. J. Chem. Biol. 2009, 2, 1–9. (180) Moon, Y.-A.; Horton, J. D. Identification of Two Mammalian Reductases Involved in the Two-carbon Fatty Acyl Elongation Cascade. J. Biol. Chem. 2003, 278, 7335–7343. (181) Luu-The, V.; Tremblay, P.; Labrie, F. Characterization of Type 12 17beta-hydroxysteroid Dehydrogenase, an Isoform of Type 3 17beta-hydroxysteroid Dehydrogenase Responsible for Estradiol Formation in Women. Mol. Endocrinol. 2006, 20, 437–443. (182) Farhane, S.; Fournier, M.-A.; Poirier, D. Synthesis and Evaluation of Amido-Deoxyestradiol Derivatives as Inhibitors of 17β-Hydroxysteroid Dehydrogenase Type 12. Cur. Enzyme Inhib. 2011, 7, 134–146. (183) Pelchat, P. Université Laval: des recherches sur un traitement prometteur du cancer du sein. Le Soleil 2014. (184) Djigoué, G. B.; Maltais, R.; Ouellet, C.; Trottier, A.; Poirier, D.; Chemical Synthesis and NMR Characterization of Non Steroidal Mimics of an Estradiol Derivative Used as Inhibitor of 17β- Hydroxysteroid Dehydrogenase Type 1. Int. J. Chem. 2012, 4. (185) Harris, H. A.; Bapat, A. R.; Gonder, D. S.; Frail, D. E.; The ligand binding profiles of estrogen receptors alpha and beta are species dependent. Steroids. 2002, 67, 379-384.

136