Projet De Nouvelle Page Titre

L’IMPLICATION DE LA 17ΒETA- HYDROXYSTEROIDE DESHYDROGENASE TYPE 1 ET DE LA TROPOMYOSINE-1 ALPHA DANS LA PROGRESSION ET L’INVASION DES CELLULES CANCÉREUSES DU SEIN

Thèse

MOUNA ZERRADI

Programme de doctorat en biologie cellulaire et moléculaire Philosophiae Doctor (Ph.D.)

Québec, Canada

L’IMPLICATION DE LA 17ΒETA- HYDROXYSTEROIDE DESHYDROGENASE TYPE 1 ET DE LA TROPOMYOSINE-1 ALPHA DANS LA PROGRESSION ET L’INVASION DES CELLULES CANCÉREUSES DU SEIN

Thèse

MOUNA ZERRADI

Sous la direction de :

Sheng Xiang Lin, directeur de recherche Jacques Huot, codirecteur de recherche

Résumé

Au Canada et partout dans le monde, le cancer du sein est un enjeu de taille en santé publique. L’exposition élevée à l'estradiol (E2) est considérée comme un facteur de risque majeur du cancer du sein. La synthèse de cette hormone est catalysée par la 17β- hydroxystéroïde déshydrogénase (17β-HSD), en particulier la 17β-HSD de type 1 (17β- HSD1) qui utilise le NADPH comme cofacteur et catalyse la conversion de l’estrone (E1) en E2. Le signal d’E2 est médié par le récepteur d’estrogène, une fois ce dernier activé par E2, on constate une stimulation de la croissance et de la prolifération cellulaire. Différents inhibiteurs sont utilisés en clinique pour bloquer la synthèse finale d’E2. Cependant aucun inhibiteur de la 17β-HSD1 n’est encore utilisé en clinique malgré l’importance de cette enzyme dans la conversion de l’E1 en E2. Ceci est certainement dû au fait que les inhibiteurs de la 17β-HSD1 sont des dérivés d’E2 d’où la difficulté d’éliminer complètement l’activité estrogénique non desirée. Dans nos travaux, nous avons tenté d’inhiber l’activité ou l’expression de la 17β-HSD1 pour étudier son effet sur la régulation du profil protéique et génomique, sur le cycle cellulaire et sur l’invasion cellulaire des cellules cancéreuses MCF7 et T47D. Nos résultats montrent que l’inhibition de la 17β-HSD1 module l’expression de diverses protéines impliquées dans la prolifération cellulaire tels que la tumor protein D54, dans le cycle cellulaire tels que la 14-3-3 epsilon et la tumor protein D53, et dans l’invasion cellulaire tels que la nm23. Aussi, l'inhibition de la 17β-HSD1 dans les cellules T47D régule l'expression des gènes impliqués dans le transport (RANBP3L, APOD), la liaison d'ADN (HIST1H2BM), le traitement de l'antigène et la présentation de l'antigène de peptide ou de polysaccharide par l'intermédiaire du CMH de classe II (HLA- DQA2), la régulation transcriptionnelle (TP63) et dans l'adhérence cellulaire (CD36). Au niveau des deux lignées cellulaires utilisées T47D et MCF7, l’inhibition de la 17β-HSD1 diminue la prolifération cellulaire, la formation d’E2, l’invasion et la migration cellulaire et mène aussi à l’arrêt du cycle cellulaire en phase G0/G1. Nos résultats montrent l'importance majeure de l’inhibition de la 17β-HSD1 dans un contexte de cancer du sein estrogéno-dépendant. Le cancer du sein est aussi une pathologie génétique associée à des modifications quantitatives et /ou

iii qualitatives des gènes tels que le gène codant pour la tropomyosine (Tm). Il a été démontré que l’expression de la Tm1 est perdue dans les cellules cancéreuses métastatiques du sein. Mon deuxième projet de recherche consistait à étudier l’effet de la phosphorylation de la tropomyosine 1 alpha sur la prolifération, l’adhérence, la formation de colonies et la migration des cellules cancéreuses du sein MDA-MB231. Pour cet effet des transfectants MDA-MB231 stables exprimant soit la Tm-1α sauvage, soit la Tm-1α mutante non- phosphorylable (Ser283Ala) ou encore la Tm-1α mutante pseudophosphorylée (Ser283Glu) ont été générés. Nos résultats montrent que les cellules exprimant la forme phosphomimétique (pseudo-phosphorylée) de la Tm1 S283E/Tm1 sont caractérisées par une adhérence accrue au substrat. Aussi, la migration transendothéliale et la migration dans un essai de cicatrisation de ces cellules est réduite par rapport aux cellules parentales ou ceux exprimant la forme non-phosphorylable de la Tm1 (S283A). En outre, nous avons constaté que les cellules exprimant les mutants S283A forment plus de colonies en agar mou que ceux exprimant les mutants S283E, montrant ainsi que la phosphorylation de la Tm1 sur la Ser283 contribue à ses propriétés anti-tumorales.

Table des matières

Résumé ...... iii Table des matières...... v Liste des tableaux ...... xi Liste des figures ...... xiii Abréviations ...... xvii Remerciements ...... xxiii Avant-Propos ...... xxvii Chapitre 1 : Introduction générale ...... 1 1.1. Le cancer du sein ...... 3 1.2. Cancer du sein hormono-dépendant...... 3 1.3. Les enzymes impliquées dans la stéroïdogénèse ...... 6 1.3.1. L’aromatase ...... 6 1.3.2. La stéroïde sulfatase ...... 7 1.3.3. La 17β-HSD1 Type 1 ...... 8 1.4. Mode d’action d’E2 ...... 11 1.4.1. La voie ER-dépendante ...... 11 1.4.1.1. La voie génomique classique ...... 11 1.4.1.2. La voie génomique non classique ...... 12 1.4.1.3. La voie non génomique ...... 12 1.4.2. La voie ER-indépendante ...... 13 1.5. Récepteur d’E2 : ER ...... 14 1.6. Les anti-estrogènes...... 15 1.6.1. Le tamoxifène et ses analogues ...... 16 1.6.2. Les anti-estrogènes purs ...... 19 1.7. Cancers non hormonaux dépendants : les oncogènes et les suppresseurs de tumeurs ...... 20 1.8. La tropomyosine ...... 21 1.8.1. Les isoformes de la tropomyosine ...... 24 1.9. La relation entre la tropomyosine et le cytosquelette d’actine ...... 27 1.10. Le niveau d’expression de la tropomyosine dans les tumeurs humaines ...... 27

1.11. Les modifications post-traductionnelles de la tropomyosine ...... 28 1.12. Les mécanismes de régulation de la tropomyosine...... 29 1.13. Les hypothèses de recherche ...... 30 1.14. Les objectifs de travail ...... 32 Chapitre 2: Differential proteome of estrogen-dependent breast cancer cells by 17β- HSD type 1 inhibition and knockdown: modulation of nm23-H1 expression, cell cycle and cell invasion ...... 33 Summary ...... 37 Résumé en français ...... 39 1. Introduction ...... 41 2. Materials and methods ...... 44 2.1. Cell culture ...... 44 2.2. Plasmid construction and stable transfection ...... 44 2.3. siRNA synthesis and transfection ...... 44 2.4. 17β-HSD1 steroidal inhibitor ...... 45 2.5. Western blot...... 45 2.6. Activity assay ...... 45 2.7. Cell proliferation assay ...... 46 2.8. Colony formation ...... 46 2.9. Cell apoptosis assay ...... 47 2.10. Cell cycle analysis ...... 47 2.11. Cell invasion ...... 48 2.12. Determination of E2 level ...... 48 2.13. Protein extraction ...... 49 2.14. Two-dimensional gel electrophoresis and image analysis ...... 49 2.15. Mass spectrometry analysis and protein identification ...... 50 2.16. Ingenuity Pathway analysis (IPA) analysis ...... 51 2.17. Quantitative real-time RT-PCR ...... 51 2.18. Statistics ...... 52 3. Results ...... 52 3.1. 17β-HSD1 knockdown and inhibition in T47D, MCF7 and MCF7-17β-HSD1 cells .... 52 3.2. Effect of 17β-HSD1 modulation on T47D, MCF7 and MCF7-17β-HSD1 cell proliferation ...... 53

3.3. Effect of 17β-HSD1 modulation on estradiol formation and colony formation in T47D, MCF7 and MCF7-17β-HSD1 cells ...... 54 3.4. Effect of 17β-HSD1 modulation on T47D, MCF7 and MCF7-17β-HSD1 cell invasion ...... 55 3.5. Effect of 17β-HSD1 inhibition and down regulation on the cell cycle and apoptosis in T47D, MCF7 and MCF7-17β-HSD1 cells ...... 55 3.6. The knockdown of 17β-HSD1 modulates the protein profile of breast cancer cell line T47D...... 56 3.7. The inhibition of 17β-HSD1 modulates the protein profile of breast cancer cell line T47D...... 57 3.8. The knock-down of 17β-HSD1 modulates the protein profile of MCF7 breast cancer cells ...... 58 3.9. The role of nm23-H1 expression in T47D cell proliferation, invasion, estradiol formation and its correlation with 17β-HSD1 expression ...... 59 3.10. Common proteins to both treatments: 17β-HSD1 siRNA transfection and 17β- HSD1 inhibition ...... 59 3.11. Network, canonical pathways, biological functions and upstream regulators identified by IPA analysis ...... 60 4. Discussion ...... 62 5. Conclusions ...... 65 References ...... 67 Chapitre 3: The effect of 17β-HSD1 inhibition on the transcriptome of T47D breast cancer cells ...... 117 Abstract ...... 120 Résumé en français ...... 121 1. Introduction ...... 122 2. Materials and methods ...... 123 2.1. Cell culture ...... 123 2.2. 17β-HSD1 steroidal inhibitor ...... 123 2.3. Activity assay ...... 123 2.4. DNA microarray processing ...... 124 2.5. Microarray analysis ...... 124 2.6. Quantitative real-time RT-PCR ...... 125 2.7. Statistics ...... 125 3. Results and discussion ...... 125 3.1. 17β-HSD1 expression and inhibition in T47D cells ...... 126

vii

3.2. The inhibition of 17β-HSD1 modulates the genomic profile of breast cancer cell line T47D 126 3.3. Networks and canonical pathways identified following 17β-HSD1 inhibition ...... 128 4. Conclusion ...... 128 References ...... 131 Chapitre 4: Regulation of breast cancer progression by phosphorylation of the tumor suppressor tropomyosin-1 alpha ...... 145 Abstract ...... 148 Résumé en français ...... 149 1. Introduction ...... 150 2. Methods...... 153 2.1. Reagents and antibodies ...... 153 2.2. Cells ...... 153 2.3. Plasmids ...... 153 2.4. Western blot...... 154 2.5. Cell Proliferation and Viability Assay ...... 154 2.6. Cell adhesion assays ...... 155 2.7. Anchorage-independent growth assays ...... 155 2.8. Cell migration assays ...... 156 2.8.1. Wound healing assay ...... 156 2.8.2. Transendothelial cell migration assays ...... 156 2.9. Statistics ...... 157 3. Results ...... 158 3.1. Adhesion of MDA MB231 cells depends on phosphorylation of Tm1 at Ser 283 ...... 158 3.2. Phosphorylation of Tm1 at S283 reduced cell migration ...... 159 3.3. Phosphorylation of Tm1 at S283 reduces the formation of MDA MB231 colonies on soft agar ...... 160 4. Discussion ...... 161 5. Conclusion ...... 163 References ...... 165 Chapitre 5: Conclusion générale ...... 175 Bibliographie (Chapitres 1 et 5) ...... 185 Annexe ...... 201

viii

Annexe 1: Functional proteomic analyses of breast cancer cells: effects of the modulation of 17beta-hydroxysteroid dehydrogenase type 1 expression on protein and transcript profiles ...... 203 Annexe 2: Androgens, body fat distribution and adipogenesis ...... 239

Liste des tableaux

Chapitre 2

Table 1. Common proteins to both treatments: 17β-HSD1 siRNA transfection and 17β- HSD1 inhibition ...... 85

Table 2. Upstream regulator analysis of proteins identified after 17β-HSD1 inhibition .. 98

Table 3. Upstream regulator analysis of proteins identified after 17β-HSD1 siRNA transfection ...... 99

Table 4. Upstream regulator analysis of proteins identified after 17β-HSD1 siRNA transfection in MCF7 cells ...... 100

Additional file 1, Table S1. Mass spectrometry identification of proteins differentially expressed between T47D control cells and T47D cells transfected with siRNA against 17β-HSD1 ...... 103

Additional file 2, Table S2. Mass spectrometry identification of proteins differentially expressed between T47D control cells and T47D cells treated with inhibitor against 17β- HSD1...... 106

Additional file 3, Table S3. Mass spectrometry identification of proteins differentially expressed between MCF7 control cells and MCF7 cells transfected with 17β-HSD1 siRNA ...... 110

Additional file 4, Table S4. Primers used for quantitative real-time RT-PCR ...... 115

Chapitre 3

Table 1. Differentially expressed genes (≥1.8-fold) in T47D cells after 17β-HSD1 inhibition by a specific inhibitor ...... 135

Table 2. The comparison between differentially expressed genes obtained by microarray and q-RT-PCR analysis in T47D cells after 17β-HSD1 by specific inhibitor ...... 138

Additional file 1, Table S1. Primers used for quantitative real-time RT-PCR ...... 143

xi xii

Liste des figures

Chapitre 1

Figure 1. Taux d’incidence annuelle des différents types de cancers chez les femmes aux États-Unis de 1975 à 2011 ...... 5

Figure 2. Les voies intracrines principales de biosynthèse et d'inactivation des androgènes et des estrogènes dans les tissus périphériques chez l’humain ...... 10

Figure 3. L’action de l’estradiol (E2) passe par l’intermédiaire de ER situé dans le noyau. . 13

Figure 4. La structure et l’homologie entre ERα et ERβ humains...... 15

Figure 5. Modèle moléculaire de l'action anti-estrogénique ...... 18

Figure 6. Les principaux analogues du tamoxifène : le torémifène, l’idoxifène, le droloxifène et le TAT-59 ...... 19

Figure 7. Les anti-estrogènes purs, ICI 164,384 et ICI 182,780, sont des dérivés de l'estradiol ...... 20

Figure 8 . Les changements d’expression des isoformes de Tm dans le cancer ...... 23

Figure 9. Les isoformes de tropomyosine (Tm) sont produites suite à l’épissage alternatif de différents gènes...... 26

Chapitre 2

Figure 1. Protein and mRNA level of 17β-HSD1 after knockdown by specific siRNA in MCF7-17β-HSD1 and T47D cells...... 73

Figure 2. Inhibition and the knock down of 17β-HSD1 reduce T47D, MCF7 and MCF7- 17β-HSD1 cell proliferation...... 75

Figure 3. Inhibition and the knock down of 17β-HSD1 reduce T47D, MCF7 cells and MCF7-17β-HSD1 cell estradiol level and colony formation...... 76

xiii

Figure 4. The Effect of 17β-HSD1 expression on cell invasion in T47D, MCF7 and MCF7-17β-HSD1 cells...... 78

Figure 5. The effect of 17β-HSD1 knockdown and inhibition on cell cycle and apoptosis in T47D cells...... 81

Figure 6. Knockdown and the inhibition of 17β-HSD1 modulate the protein profile of T47D breast cancer cell lines...... 82

Figure 7. The knock down of 17β-HSD1 modulates the MCF7 protein profile ...... 84

Figure 8. Western blot and RT-qPCR confirmation in T47D cells ...... 86

Figure 9. Western blot and RT-qPCR confirmation in MCF7 cells...... 87

Figure 10. Role of nm23-H1 expression in T47D cell proliferation, invasion, estradiol formation and correlation with 17β-HSD1 expression...... 88

Figure 11. The common networks to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D cells. Two common networks were identified...... 92

Figure 12. The common canonical pathways to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D cells...... 95

Figure 13. The common biological functions to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D cells...... 97

Figure 14. TP53 as a common upstream regulator to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D...... 101

Chapitre 3

Figure 1. 17β-HSD1 inhibition in T47D cells by a specific inhibitor. Activity assays were carried out in T47D using [14C] E1 as substrate...... 139

Figure 2 . Overpresentation of functional categories. The genes are grouped into cellular components, biological processes and molecular function...... 140

Figure 3. Network identified following 17β-HSD1 inhibition in T47D cells...... 141

Figure 4. Canonical pathways identified following 17β-HSD1 inhibition in T47D cells...... 142

xiv

Chapitre 4

Figure 1. Generation of MDA MB-231 cells stably expressing wild type forms of Tm1 or phosphomimetic or nonphosphorylatable forms of Tm1...... 169

Figure 2. Tropomyosin-1 expression in MDA MB231 cell lines reduces their proliferation...... 170

Figure 3. Phosphorylation of Tm1 at S283 increases their adhesion to substratum...... 171

Figure 4. Phosphorylation of Tm1 at S283 reduced cell migration in MDA MB231. ... 172

Figure 5. Phosphorylation of Tm1 at S283 reduced transendothelial cell migration of MDA MB231...... 173

Figure 6. Phosphorylation of Tm1 at S283 reduces the formation of MDA-MB231 colonies on soft agar by cells...... 174

xvi

Abréviations

14C carbone 14 17β-HSD 17 β-hydroxystéroïde déshydrogénase 17 β-HSD1 17 β-hydroxystéroïde déshydrogénase type 1 2-D bidimensionnel 2-MeOE2 2-methoxyestradiol 2-OHE2 2-hydroxyestradiol 4-dione 4-androstenedione 4-OHE2 4-hydroxyestradiol ADN acide désoxyribonucléique ADNc ADN complémentaire AF1 fonction d'activation ARN acide ribonucléique ARNm ARN messager DBD domaine de liaison d'ADN DHEA déhydroépiandrostérone DHEA-S déhydroépiandrostérone sulfate DMEM Dulbecco 's Modified Eagle Medium El estrone El-S estrone sulfate E2 estradiol ER récepteur des estrogènes (estrogen receptor) ER+ ER-positif ERE élément de réponse aux estrogènes FBS fetal bovine serum FDA Food and Drug Administration G-418 Geneticin IA inhibiteurs de l'aromatase kDa kilodalton 1 litre

xvii

LBD domaine de liaison du ligand M molaire mA milliampère mg milligramme min minute mL millilitre mM millimolaire MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide NADPH nicotinamide adénine dinucléotide phosphate nm nanomètre nM nanomolaire pb paires de bases PBS phosphate buffer saline PCR réaction de polymérisation en chaîne (polymerase chain reaction) pH potentiel hydrogen pI point isoélectrique Q-RT-PCR quantitative real-time RT-PCR Rpm rotations par minute RT-PCR reverse transcription PCR SDR déshydrogénases/réductases à chaîne courte SDS-PAGE sodium dodecyl sulphate - polyacrylamide gel electrophoresis sec seconde siRNA petits ARN interférents (small interfering RNA) Ser61 Sérine 61 Ser118 Sérine 118 Ser167 Sérine 167 Ser283 Sérine 283 SERM modulateur sélectif des récepteurs des estrogènes STS stéroïde sulfatase T testostérone TAM tamoxifène

xviii

TLC chromatographie sur couche mince Tm tropomyosin Tm1α tropomyosin 1 alpha Tris-HCl tris (hydroxymethyl) aminomethane hydrochloride Wt wild type µg microgramme µl microlitre µM micromolaire °C degré celcius

xix

Je dédie cette thèse

À mes très chers parents

À mon très cher mari

xxi xxii

Remerciements

Mes vifs remerciements s'adressent à mon directeur de recherche Dr Sheng-Xiang Lin pour mon accueil dans son laboratoire et pour son encadrement tant apprécié ainsi que ses conseils judicieux.

Je tiens aussi à remercier mon co-directeur Dr Jacques Huot qui m'a accueilli dans son laboratoire et m'a permis de venir à Québec afin de réaliser ma thèse. Je vous remercie d'avoir participé à mon développement scientifique malgré toutes les difficultés rencontrées. Vous trouverez ici l'expression de mon admiration et de ma reconnaissance.

Un très grand merci au Dr André Tchernof pour la prélecture et la correction de ma thèse. Les mots ne seront jamais suffisants pour exprimer ma grande reconnaissance pour tous ce que vous avez faits pour m’aider afin de finaliser ma thèse et commencer une nouvelle carrière. Merci du plus profond de mon cœur André.

J'adresse également mes remerciements aux membres du jury, pour avoir accepté d’évaluer ma thèse, notamment Dr Donald Poirier, Dr André Tchernof et Dr. Yi-Wei Huang.

Un grand merci à l'équipe du laboratoire qui par leur compétence et leur dévouement, m'ont aidé au cours des différentes étapes de ma thèse. Merci Dan Xu, ma collègue de bureau, Xiaoqiang Wang, Leyi Lin, Bo Zhan, Chenyan Zhang, Yan wang, Jean François Thériault, Ming Zhou, Dao-Wei Zhu et François Houle. Merci à tous ceux qui ont croisé mon chemin.

Je tiens à remercier le Dr Donald Poirier qui nous a fourni certains inhibiteurs stéroïdiens utilisés lors de cette étude.

Je remercie Gina Racine pour son aide dans l'analyse des gels d'électrophorèses bidimensionnelles. Le travail avec toi était un vrai plaisir, ta gentillesse et ta façon de faire rendent les tâches tellement faciles. Je salue ton professionnalisme et ta disponibilité.

xxiii

Je remercie le Dr Ezékiel Calvo et Annick Ouellet pour leur aide très précieuse dans l'apprentissage des analyses génomiques des biopuces à ADN.

Je tiens à remercier Carl St-Pierre pour son support technique en microscopie confocale, Ronald Maheux pour l'analyse d'images, Nathalie Paquet pour la quantification en temps réel des ARNm, Alexandre Brunet pour l’analyse en cytométrie de flux. Je salue votre professionnalisme ainsi que l'efficacité avec laquelle vous réalisez vos tâches.

Je tiens à remercier le Gouvernement du Royaume du Maroc et spécifiquement le Ministère de l’enseignement supérieur de m’avoir accordé la bourse d’exemption des frais majorés. Votre aide est bien appréciée.

Je remercie tous mes amis qui m’ont apporté leur soutien lors des différentes étapes de mon doctorat et plus spécifiquement mes deux amies doctorantes Wafae et Sanaa qui ont toujours été là pour me remonter le moral.

Je remercie mon frère jumeau Issam, ma petite nièce adorée Ines et mes beaux parents Mohamed et Halima, merci pour votre soutien moral.

Je remercie l’amour de ma vie, mon cher mari Adil de m’avoir soutenu et encouragé pour traverser les moments les plus difficiles durant ces dernières années. Tes encouragements et surtout ta confiance en moi m’ont toujours redonné espoir.

Un grand merci pour mes très chers parents, ma mère Naima et mon père Mohamed, les mots ne sont pas suffisants pour exprimer mon amour et ma gratitude envers vous. Je vous remercie d’avoir accepté et supporté mon absence au cours de ces dernières années et de m’avoir aidé financièrement tout au long de mes études et spécialement lors de la période de rédaction de ma thèse. Vous étiez toujours là au moment les plus difficiles de ma vie, les moments où je n’étais plus capable d’avancer. Merci maman de m’avoir appelé chaque jour de l’année sans omettre aucun jour, tes appels me font sentir que je suis avec vous et que je vis votre quotidien, grâce à cela je ne me suis jamais sentie loin de vous ni loin de mon pays,

xxiv

le Maroc. Je vous dédie cette thèse en reconnaissance de tous vos sacrifices tout au long de ma vie.

xxv xxvi

Avant-Propos

Le projet de recherche portant sur l’implication de la 17β-hydroxystéroide déshydrogénase type 1 dans la progression du cancer du sein a été effectué au Centre de recherche du Centre Hospitalier de l'Université Laval (CHUQ-CHUL, Québec) à l'unité de recherche en endocrinologie moléculaire et oncologique et génomique humaine. Le deuxième portant sur l’effet de la tropomyosine-1 et de sa phosphorylation dans la progression du cancer du sein a été effectué au Centre de recherche de cancérologie de l’hôtel Dieu de Québec (CHUQ- HDQ, Québec). Durant mes études doctorales, j’ai réalisé deux projets de recherche dont la majorité des résultats sont présentés dans cette thèse.

Cette thèse est composée de cinq chapitres. Le premier chapitre porte sur une revue bibliographique des deux projets de recherche. Les hypothèses et les objectifs de recherche sont présentés à la fin de ce premier chapitre.

Les chapitres 2, 3 et 4 présentent les résultats obtenus à l’issue de mes deux projets de recherche de doctorat. Ces chapitres sont présentés sous forme d’articles scientifiques rédigés en anglais avec un résumé en français au début de chaque article. Je suis la première auteure de ces 3 articles.

Le chapitre 2 est un article soumis à Genome Medicine. J'ai réalisé la totalité du travail de laboratoire et j’ai rédigé la totalité de cet article.

Le chapitre 3 est un article en voie de soumission. J’ai réalisé la majorité des expériences et j’ai rédigé la totalité de cet article.

Le chapitre 4 est un article publié dans Journal of Cancer Therapy (JCT, 2015, Volume 6, 783-792). J'ai effectué la totalité du travail de laboratoire et j’ai rédigé l’article en collaboration avec le Dr Jacques Huot.

xxvii

Le chapitre 5 présente une conclusion générale des résultats obtenus dans mes deux projets de recherche ainsi que les perspectives futures de chacun des projets.

L’annexe 1 présente un article publié dans Breast Cancer Research (Breast Cancer Research, 2012, Volume 14 : R92). J’ai participé à la réalisation des expériences notamment les expériences sur la migration cellulaire. Cet article montre l’effet de la 17β-HSD type 1 sur le profil protéique des cellules cancéreuses du sein ainsi que sur la migration cellulaire.

L’annexe 2 comprend également un article de revue sur les androgènes publié à titre de première auteure dans Current Obesity Reports (Curr Obes Rep (2014) 3:396–403).

xxviii

Chapitre 1 : Introduction générale

1.1. Le cancer du sein

Au Canada et partout dans le monde, le cancer du sein est un enjeu de taille en santé publique. Quelle que soit la façon de l’évaluer, il s’agit d’un phénomène répandu, qui représente la deuxième cause de décès par cancer chez les femmes à travers le monde [1, 2]. Le cancer du sein est le cancer le plus fréquemment diagnostiqué [2]. Au Canada et aux États-Unis, le cancer du poumon est la première cause de décès par cancer chez les femmes suivi par le cancer du sein [3]. Chez les femmes, le cancer du sein devrait représenter 29% de tous les nouveaux cas de cancers, ce qui représente 231 840 cas du total des estimations des nouveaux cas et 40 290 cas de décès par cancer en 2015 [4]. Le taux d’incidence annuelle du cancer du sein est supérieur aux autres types de cancers chez les femmes aux États-Unis de 1975 à 2011 [4] (Figure 1). Au Canada, le taux de cancer du sein figure parmi le plus élevé qui soit au monde. En général, les taux d'incidence sont élevés en Europe de l'Ouest et du Nord, l'Australie / Nouvelle-Zélande et l'Amérique du Nord; intermédiaire en Amérique du Sud, dans les Caraïbes, et en Afrique du Nord; et faible en Afrique sub-saharienne et en Asie [2]. Huit cas de cancer du sein sur dix surviennent chez les femmes âgées de 50 ans ou plus. En 2013, on a estimé que 6000 femmes au Québec (23 800 au Canada) seront diagnostiquées avec un cancer du sein. Environ 1.350 femmes au Québec (5000 au Canada) vont mourir de la maladie [5]. En dépit de sa forte incidence, le dépistage précoce et les traitements modernes ont considérablement augmenté la survie des patientes.

1.2. Cancer du sein hormono-dépendant

L’estradiol (E2) régule un large éventail de réactions physiologiques dans divers tissus cibles, et joue un rôle important dans le développement et la progression des cancers du sein [6]. La concentration d’E2 est 2.3 fois plus élevée dans le tissu du cancer du sein en comparaison avec le tissu normal [7].

Les données du registre mondial montrent que l'incidence du cancer du sein a augmenté dans la plupart des régions du monde depuis 1973 [8]. L'importance des estrogènes dans l'étiologie du cancer du sein est largement reconnue et le gouvernement des États-Unis a ajouté les estrogènes stéroïdiens à la liste des cancérogènes humains connus [9-15]. L’exposition élevée à E2 est considérée comme un facteur de risque majeur du cancer du sein. Il existe aussi d’autres facteurs de risque liés au cancer du sein tels que les menstruations précoces, la grossesse tardive, la ménopause tardive, l'utilisation prolongée de contraceptifs oraux et le traitement hormonal substitutif [16-18].

Les niveaux d'E2 intra-tumoraux ne sont pas significativement différents entre les patientes atteintes de cancer du sein avant et après la ménopause, mais le ratio estradiol/estrone intratumoral est significativement plus élevé chez les femmes ménopausées en comparaison aux femmes non ménopausées atteintes de cancer du sein [19, 20]. Le ratio de la concentration d'E2 dans le tissu tumoral/plasma est de 23 dans les carcinomes du sein chez les femmes ménopausées, mais est de 5 dans les carcinomes du sein avant la ménopause [21, 22].

Figure 1. Taux d’incidence annuelle des différents types de cancers chez les femmes aux États-Unis de 1975 à 2011. Les cancers présentés sont le cancer du sein, du colon, du poumon, de l’utérus, le mélanome de la peau, de la thyroïde et du foie. Source : Siegel, R. L. et al, 2015 [4].

1.3. Les enzymes impliquées dans la stéroïdogénèse

Il est bien connu que l’estradiol (E2) est impliqué dans la progression des cancers estrogéno- dépendants. La synthèse d’E2 de fait via 2 voies : la voie de l’aromatase et la voie de la stéroïde sulfatase (STS). L’aromatase catalyse la conversion de la testostérone (T) en E2 et la conversion du 4-androstènedione (4-dione) en estrone (El). Dans la voie de la STS, l’estrone- sulfate (E1-S) est transformé en El par la STS, suivie par une étape d'activation pour produire l’E2. La dernière étape de la synthèse d’E2 est catalysée par la 17β-hydroxystéroïde déshydrogénase (17β-HSD), en particulier la 17β-HSD de type 1 (17β-HSD1) [23] (Figure 2).

1.3.1. L’aromatase

La biosynthèse des estrogènes est catalysée par un complexe enzymatique nommé aromatase, dont l'activité se traduit par l’aromatisation du cycle A des androgènes, pour former le cycle A phénolique, qui est une caractéristique des estrogènes, avec une perte concomitante du groupe méthyle angulaire en position 19 [24]. L’aromatase catalyse la conversion des stéroïdes androgéniques C-19 en stéroïdes estrogéniques C-18, cette étape est critique dans la biosynthèse des estrogènes [25]. L’aromatase P-450, une enzyme de 55 kDa, est le produit du gène CYP19, un membre de la superfamille du gène cytochrome P-450 [26]. Le gène CYP19 de l’aromatase est situé sur le chromosome 15q21 [27]. L'enzyme catalyse la dernière étape de la biosynthèse des estrogènes par trois réactions d'hydroxylation successives du cycle A des androgènes [28, 29]. L'activité de l'aromatase P-450 est détectée dans différents types cellulaires et sites tissulaires humains, tels que le placenta, les cellules granulosa de l'ovaire, les cellules testiculaires de Sertoli et de Leydig, le tissu adipeux, le foie, la peau et le cerveau [24, 30]. L'enzyme est localisée dans le reticulum endoplasmique et forme un complexe fonctionnel avec le NADPH-cytochrome P-450 réductase qui est une flavoprotéine ubiquitaire [31].

Des inhibiteurs de l'aromatase (IA) ont été développés et permettent de réduire considérablement la production des estrogènes, offrant ainsi une approche thérapeutique pour le traitement des tumeurs du sein estrogèno-dépendantes [3]. Il y a seulement trois IA disponibles sur le marché et approuvés par la FDA pour le traitement du cancer du sein, notamment, l’inhibiteur stéroïdien exémestane, le létrozole et l'anastrozole, ces deux derniers étant des composés non stéroïdiens [32].

1.3.2. La stéroïde sulfatase

La stéroïde sulfatase (STS) est une enzyme qui permet la synthèse des précurseurs de l’E2. La STS permet la conversion de l'estrone sulfate (ElS) en estrone (E1) [33] et le déhydroépiandrostérone sulfate (DHEA-S) en déhydroépiandrostérone (DHEA) [34]. E1 est ensuite converti en E2 et le DHEA en Δ5-diol, lequel est un précurseur de la biosynthèse d’E2 via l’aromatase, menant ainsi à une stimulation de la croissance des tumeurs de sein estrogéno-dépendantes. La STS est exprimée dans les ovaires, l’endomètre, les testicules, la prostate, la peau, le cerveau, le foie, la glande surrénale, le placenta et l’os [35-38]. L’expression de l'ARNm du STS dans les tissus malins du sein est beaucoup plus élevée que dans les tissus normaux et également beaucoup plus élevée que l'expression de l'ARNm de l'aromatase [39]. Des études ont confirmé que l'expression de l'ARNm du STS dans les tumeurs du sein ER-positives est un indicateur de pronostique indépendant dans la prédiction de la survie sans rechute, avec des niveaux d'expression élevés associés à un mauvais pronostic [40-42].

L'inhibition de la STS représente une nouvelle approche pour bloquer la formation des stéroïdes possédant des propriétés estrogéniques, et pour cet effet un inhibiteur non stéroïdien de la STS (667COUMATE) est actuellement en phase clinique pour le traitement du cancer du sein hormono-dépendant [43].

1.3.3. La 17β-HSD1 Type 1

Les 17β-HSD catalysent l’inter-conversion des estrogènes et des androgènes [20]. La 17β- HSD1 a été cristallisée en 1993 [44] et sa structure a été rapportée en 1995 comme premier exemple d’une enzyme stéroïdienne humaine de conversion [45, 46]. Depuis, plusieurs études structure-fonction ont été réalisées, et une douzaine de structures 3D de l'enzyme avec divers stéroïdes, cofacteurs et inhibiteurs ont été déterminées [47, 48]. La 17-βHSD1 utilise le NADPH comme cofacteur et catalyse la conversion de E1 en E2 [49, 50]. Le gène codant pour la 17β-HSD1 (HSD17β1) est localisé sur le chromosome 17q12-21 [50]. L’enzyme est membre de la famille des déshydrogénases/réductases à chaîne courte (SDR) et est localisée dans le cytoplasme et principalement exprimée dans le placenta, les ovaires et les glandes mammaires [51, 52]. En plus, la 17β-HSD1mène à la synthèse d’E2 qui stimule la croissance des cellules du cancer du sein [53, 54]. Dans le cancer du sein, l’expression et l’activité de la 17β-HSD1 sont significativement plus élevées que dans le tissu normal du sein, menant ainsi à une stimulation et un développement tumorale [22, 55, 56]. Aussi, une forte expression de l’ARNm de la 17β-HSD1 est significativement associée à un mauvais pronostic dans le cancer du sein [55]. Le niveau d'ARNm de la 17β-HSD1 ainsi que le ratio E2/E1 sont supérieurs chez les femmes ménopausées en comparaison avec les femmes pré- ménopausées atteintes du cancer du sein [19]. Ces données ont clairement montré que les niveaux élevés d’E2 intratumoral chez les femmes ménopausées sont dus à une forte expression de 17β-HSD1 [19, 57] . La surexpression de la 17β-HSD1 permettra une conversion élevée de E1 en E2 chez les femmes ménopausées, ce qui explique le fait que les femmes ménopausées atteintes d'un cancer du sein ont un ratio E2/E1 intratumorale plus élevé que les patientes pré-ménopausées [19]. Ces données ont clairement montré l'implication de l'enzyme 17β-HSD1 dans le cancer du sein hormono-dépendant.

À cet effet, des inhibiteurs de la 17β-HSD1 ont été synthétisés afin d’inhiber l’activité de l’enzyme. EM-1745 est un inhibiteur stéroïdien avec un niveau d'estrogénicité élevé et un pouvoir élevé d’inhibition [58]. Le composé 18 est un inhibiteur moins estrogénique que l’EM-1745 avec une forte inhibition de la conversion d’E1 en E2 de 81% à 10 µM [59].

Récemment, un inhibiteur non estrogénique in vivo et in vitro (PBRM) a été synthétisé par l’équipe de Poirier et al [60]. Cet inhibiteur est un stéroïde dérivé d’E2 qui possède deux caractéristiques majeures: une chaîne carbamoylbenzyl en position C16 pour inhiber la 17β- HSD1 et une chaîne latérale 2-bromoéthyle en position C3 pour éliminer l'activité estrogénique résiduelle due au noyau E2 [61]. PBRM a un IC50 de 68 nmol/L et il est capable de réduire la taille de la tumeur de 100% in vivo [60]. Plusieurs inhibiteurs non stéroïdiens appartenant à la famille des flavones et des coumestanes ont aussi été utilisés pour l’inhibition de la 17β-HSD1 [62].

Figure 2. Les voies intracrines principales de biosynthèse et d'inactivation des androgènes et des estrogènes dans les tissus périphériques chez l’humain. TESTO, testostérone; El, estrone; E2, estradiol; E2-S, l'estradiol sulfate; El-S, estrone sulfate; DHEA, dehydroépiandrostérone; DHEA-S, dehydroépiandrostérone sulfate; 4-DIONE, 4-androstènedione; 5-DIOL, androstenediol; DHT, dihydrotestostérone [23].

1.4. Mode d’action d’E2

Le signal d’E2 est transmis par les récepteurs des estrogènes. Les estrogènes exercent leur effet carcinogène via deux voies de signalisation : la voie ER-dépendante et la voie ER- indépendante [11, 63-65].

1.4.1. La voie ER-dépendante

L'estrogène exerce ses effets biologiques en se liant au ER, qui est principalement exprimé dans le noyau en tant que membre de la superfamille des récepteurs nucléaires des facteurs de transcription. ER agit en formant des homos ou hétéros-dimères d’ERα et ERβ. ERα a été largement utilisé comme un marqueur prédictif pour l'hormonothérapie. Il existe trois voies génomiques et non génomiques distinctes par lesquelles l’estrogène régule l'expression d'une grande variété de gènes impliqués dans la croissance cellulaire [6].

1.4.1.1. La voie génomique classique

Dans la voie classique, les estrogènes diffusent à travers la membrane plasmique des cellules, où ils se lient au ER [66]. Une fois liées, les protéines de choc thermique (HSP90) se dissocient et un changement de conformation et une homodimérisation du ER se produisent. Ensuite, ER fonctionne comme un facteur de transcription qui se lie à une séquence d'ADN spécifique, l'élément de réponse aux estrogènes (ERE), qui est présente dans les régions promotrices des gènes cibles (Figure 3) [67]. Ces ERE sont des séquences palindromiques de 13 paires de bases situées en amont du site de début de transcription [68]. ER se lie à des ERE par son domaine de liaison à l'ADN (DBD) et recrute des co-activateurs tels que SRC-1,

AIB1 et p300/CBP pour former un complexe ER fonctionnel [69]. Cette activation du ER par E2, mène à la stimulation de la croissance et la prolifération cellulaire [10, 11, 63-65].

1.4.1.2. La voie génomique non classique

Dans une voie génomique non-classique, ER peut réguler la transcription sans se lier directement à l’ADN [70]. ER agit comme un co-activateur et interagit avec d'autres facteurs de transcription tels que AP-1, SP-1 et NF-kB par l'intermédiaire d'interactions protéine- protéine, et peut réguler la transcription de gènes qui ne possède pas d’ERE. Les gènes activés de cette manière sont l'ovalbumine, l'IGF-1, la collagénase, c-Myc, la cycline D1, c- fos et le récepteur de lipoprotéines à faible densité [69-73].

1.4.1.3. La voie non génomique

Dans la voie non génomique, la liaison d’E2 au ER membranaire permet l’activation non génomique de la MAPK et PI3K/Akt qui à leur tour activent par phosphorylation des facteurs de transcription cibles et/ou des co-activateurs notamment l’ERα nucléaire [69]. Ce dernier possède plusieurs sites de phosphorylation différents tels que la Ser118 et Ser167 qui sont phosphorylés par diverses kinases, notamment la MAPK et la PI3K/Akt [74]. Ces ER phosphorylés dimérisent et peuvent agir comme un facteur de transcription pour ERE [69].

Figure 3. L’action de l’estradiol (E2) passe par l’intermédiaire de ER situé dans le noyau. Une fois E2 lié au ER, ce dernier s’homodimérise et interagit avec les ERE situés dans la région promotrice des gènes activés par E2. Ces événements déclenchent une réponse estrogénique dans la cellule [67].

1.4.2. La voie ER-indépendante

La voie ER-indépendante du cancer du sein estrogèno-dépendant implique la production des métabolites d’estrogène génotoxiques, qui sont très réactifs et provoquent des dommages à l'ADN [64, 65, 75]. E2 est métabolisé en 2-hydroxyestradiol (2-OHE2) et 4-hydroxyestradiol (4-OHE2) par le cytochrome P450 1A1 (CYP1A1) et le cytochrome P450 1B1 (CYP1B1), respectivement [9, 76]. Le 2-OHE2 et 2- methoxyestradiol (2-MeOE2) ont des caractéristiques chimio-protectrice possibles, alors que 4-OHE2 est très génotoxique [77-81]. Les métabolites cancérigènes d'estrogènes tels que le 4-OHE2 subissent une oxydation pour former des quinones électrophiles qui réagissent facilement avec l'ADN [9, 63, 75, 82]. En outre, le cycle redox des quinones et semiquinones conduit à la formation de radicaux libres et des réactifs d'oxygène, créant ainsi plus de possibilités pour des dommages génétiques [12, 77, 83].

1.5. Récepteur d’E2 : ER

Les récepteurs des estrogènes (ERα et ERβ) sont des membres de la superfamille des récepteurs nucléaires des facteurs de transcription [84]. ERα a été le premier récepteur des estrogènes cloné et il a été isolé à partir des cellules humaines MCF-7 de cancer du sein dans les années 1980 [85-87]. Dix ans plus tard, ERβ a été cloné à partir de la prostate de rat [88]. Aussi des formes humaines [89-91] et de souris [92] d’ERβ ont été clonées. Le gène d’ERα humain est situé sur le chromosome 6 et le gène ERβ est sur le chromosome 14 [90]. ERα et ERβ sont exprimés dans le cerveau, le sein, le système cardiovasculaire, les os et le tractus urogénital [93-96].

La protéine ERα est composée de 595 acides aminés avec un poids moléculaire de 66 kDa [87]. Alors que la protéine ERβ est constituée de 530 acides aminés (28). ER comprend six domaines fonctionnels différents (Figure 4) [97, 98]. Chez l’humain, ERα et ERβ partagent des domaines structuraux communs, qui sont désignés AF. Le domaine A/B contient la fonction d'activation (AF1), une fonction d'activation constitutive qui contribue à l'activité transcriptionnelle du ER. Ce domaine est l'un des domaines les moins conservés entre ERα et ERβ, présentant seulement 30% d’homologie. Le domaine de liaison d'ADN (DBD), ou le domaine C, est la région la plus conservée entre ERα et ERβ, avec 96% d’homologie. Ceci permet aux deux récepteurs de se lier à des sites cibles similaires. Le DBD se compose de deux doigts de zinc qui sont des éléments essentiels du ER vu que des études ont démontré que ER ne peut pas se lier à l'ADN in vitro ou in vivo lors de la perte du DBD [98, 99]. Le domaine D contient le signal de localisation nucléaire avec une homologie de 30% entre les deux récepteurs. Finalement, la région E/F comprend le domaine de liaison du ligand (LBD), une surface de liaison de co-régulation, le domaine de dimérisation, un second signal de localisation nucléaire et la fonction d'activation 2 (AF2). Contrairement à AF1, AF2 est une fonction d'activation dépendante du ligand. Les domaines E/F du ERα et ERβ présentent 53% d’homologie [84].

E2 et son récepteur (ER) jouent un rôle important dans la genèse et la progression maligne du cancer du sein. ERα régule la transcription de divers gènes agissant comme un facteur de

transcription qui se lie à des éléments de réponse aux estrogènes (ERE) en amont des gènes cibles. L'expression du ERα est étroitement associée à la biologie du cancer du sein, en particulier le développement des tumeurs; par exemple, les cancers du sein qui sous- expriment l’ERα révèlent souvent des phénotypes plus agressifs. En outre, l'expression du ERα dans les tissus tumoraux est un indicateur de pronostic favorable dans le traitement endocrinien [6]. La surexpression du ERα est fréquemment observée dans le stade précoce du cancer du sein [6].

Figure 4. La structure et l’homologie entre ERα et ERβ humains. Les domaines A-F sont représentés, ainsi que le pourcentage d’homologie entre les domaines individuels au niveau de l'acide aminé. Les fonctions d'activation 1 (AF1) et 2 (AF2) sont aussi indiquées. Adapté à partir de [84, 91].

1.6. Les anti-estrogènes

Les anti-estrogènes peuvent être classés en deux groupes: les analogues du tamoxifène ou de ses métabolites et les anti-estrogènes purs [67].

1.6.1. Le tamoxifène et ses analogues

Le tamoxifène (Nolvadex ®, Astra Zeneca Pharmaceuticals Wilmington, DE) a été développé il y a plus de 30 ans et a été approuvé par la US Food and Drug Administration (FDA) en 1977 pour le traitement du cancer du sein avancé ER-positif [100]. Le tamoxifène est un anti-estrogène et un modulateur sélectif des récepteurs des estrogènes (SERM) qui se lie à l’ER et, dans les cellules du cancer du sein, antagonise l'effet des estrogènes sur une variété de gènes de régulation de la croissance cellulaire [101] (Figure 5). L'effet prédominant du tamoxifène est cytostatique avec l'induction de l’arrêt du cycle cellulaire en phase G1, diminuant ainsi la prolifération des cellules cancéreuses [100]. Le tamoxifène a aussi un effet estrogénique dans l'os, préservant ainsi la densité osseuse chez les femmes ménopausées [102, 103]. L'effet secondaire le plus grave du tamoxifène est une conséquence de son activité estrogénique dans l'endomètre, qui se traduit par une hyperplasie de l'endomètre et un cancer de l'endomètre [104, 105].

La découverte que le tamoxifène (TAM) est métabolisé en 4 hydroxy-tamoxifène (4-OHT), un anti-estrogène puissant [106], a eu un impact majeur sur le développement de ce type de médicaments. Aussi, la structure triphényléthylénique du TAM a fourni la base pour la synthèse de plusieurs nouveaux analogues présentés à la Figure 6 [67].

Le torémifène ou chlorotamoxifène (Fareston®; Schering Corp, Kenilworth, NJ) diffère du TAM seulement par la présence d'un atome de chlore en position 4, et ses activités précliniques et cliniques sont très similaires à celles du TAM [107]. C’est un anti-estrogène et un agent anti-tumoral [108, 109], utilisé pour le traitement du cancer du sein avancé et testé en tant que thérapie adjuvante (Figure 6). Le toremifène a montré moins d’effet génotoxique que le TAM mais possède les mêmes effets estrogéniques dans l'utérus que ce dernier [110, 111].

L’idoxifène est aussi un analogue stable du TAM du point de vue métabolique, il est synthétisé pour éviter la toxicité causée par le TAM dans le foie de rat [112, 113]. Il se lie au

ER avec une affinité supérieure à celle du TAM et c’est un inhibiteur plus puissant de la croissance des tumeurs mammaires chez le rat et de la croissance des cellules cancéreuses humaines du sein dans des modèles de xénogreffes [114, 115].

Le droloxifène, ou 3-hydroxytamoxifène, a été largement étudié comme un anti-estrogène et un agent anti-tumoral dans des tests de laboratoire [116] (Figure 6). Il possède un spectre d'activité similaire à celui du tamoxifène et du torémifène avec une plus grande affinité pour ER [107].

Le TAT-59 est un anti-estrogène développé pour le traitement du cancer du sein avancé. Ce médicament inhibe la croissance des cellules du cancer du sein ER+ stimulées par l’estrogène et transplantées dans des souris athymiques [117, 118] (Figure 6). Comparé au TAM, le TAT-59 possède une affinité de liaison plus élevée pour ER [107].

Le raloxifène (Evista ®), un dérivé benzothiophénique, se lie à ER avec une affinité égale à celle de E2 [119]. Le développement de ce médicament comme un anti-estrogène pour le cancer du sein a été abandonné dans les années 1980. Selon l'Institut national du cancer, la FDA a approuvé le raloxifène en 2007 pour traiter les femmes ménopausées souffrant d'ostéoporose et celles à risque élevé de développer un cancer du sein [120]. Il n’est pas utilisé pour le traitement du cancer du sein mais plutôt pour la prévention étant donné qu’un traitement de trois ans permet de réduire de 76% le risque du cancer du sein chez les femmes ménopausées souffrant d'ostéoporose [120, 121]. .

Figure 5. Modèle moléculaire de l'action anti-estrogénique [67].

Figure 6. Les principaux analogues du tamoxifène : le torémifène, l’idoxifène, le droloxifène et le TAT-59 [67].

1.6.2. Les anti-estrogènes purs

Les anti-estrogènes purs ont été découverts par Wakeling et ses collègues [122]. Le composé principal, ICI 164,384, est un dérivé de E2 en position 7α qui n'a pas de propriétés estrogéniques détectables in vivo ou in vitro [67, 123]. Il existe aussi un autre anti-estrogène pur plus puissant que le ICI 164,384 et c’est le composé ICI 182,780 (FASLODEX®) [124]. Contrairement aux SERM, ICI 182,780 bloque la transactivation du ER provenant à la fois des domaines AF-1 et AF-2 [125]. Ce médicament induit la dégradation du ER (Figure7), avec une diminution marquée de sa concentration cellulaire [126-128].

Figure 7. Les anti-estrogènes purs, ICI 164,384 et ICI 182,780, sont des dérivés de l'estradiol, mais n’ont pas de propriétés estrogéniques [67].

1.7. Cancers non hormonaux dépendants : les oncogènes et les suppresseurs de tumeurs

Les cancers sont des pathologies génétiques et épigénétique associées à des modifications quantitatives et /ou qualitatives des gènes. Le nombre et la nature des gènes associés au cancer sont impressionnants. L’initiation et la progression du cancer du sein sont caractérisées par de multiples altérations génétiques et épigénétiques qui activent des oncogènes et perturbent les fonctions de gènes suppresseurs de tumeurs spécifiques [129]. Les analyses de génomique indiquent qu'il y a seulement quelques gènes qui sont couramment mutés dans le cancer du sein. Ces gènes sont les oncogènes ErbB2, PI3KCA, MYC, et CCND1 ainsi que les suppresseurs de tumeur BRCA1 / 2 et p53 [129]. PTEN est un autre suppresseur de tumeur important qui contre le signal de survie transmis par PI3K des

oncogènes activés tels que ErbB2 [129]. D'autre part, de nombreux gènes sont moins fréquemment mutés ce qui fournit une explication à la grande hétérogénéité du cancer du sein.

Les gènes codant pour la tropomyosine (Tm) sont également fréquemment dérégulés dans le cancer du sein, et des changements significatifs dans l'expression des différentes isoformes de la tropomyosine accompagnent la transformation cellulaire. Les cellules perdent l’expression des isoformes de tropomyosine de haut poids moléculaire en général, et celle de la tropomyosine 1 en particulier (Figure 8). Plusieurs études ont montré que l’expression de la Tm1 est perdue dans les cellules cancéreuses métastatiques du sein [130, 131]. Les changements d’expression de la tropomyosine sont aussi associés à l’acquisition des propriétés métastatiques [132].

1.8. La tropomyosine

La tropomyosine (Tm) est une protéine qui se lie au filament d’actine et joue un rôle primordial dans la régulation des fonctions du cytosquelette au niveau des cellules musculaires et non-musculaires [133]. Elle est exprimée au niveau des opisthocontes y compris les animaux et les champignons, mais son expression n’est pas documentée dans les plantes, les protistes et les procaryotes [134, 135]. L'étude de la fonction de la Tm en dehors du sarcomère musculaire est également devenu importante [136], en particulier dans le contexte de la motilité cellulaire altérée de l'actomyosine dans les cellules cancéreuses [132, 137-139]. On distingue des isoformes musculaires et non musculaires de la Tm. Au niveau du muscle strié, la Tm musculaire régule l’interaction entre l’actine et la myosine par le biais du complexe troponine qui joue le rôle d’un médiateur de la contraction musculaire en réponse au calcium [140, 141]. Au niveau des cellules non musculaires, la Tm joue un rôle dans la formation et la stabilisation des fibres d’actines en facilitant les interactions actomyosine et en protégeant l’actine contre l’action de la cofilin et la gelsolin [140]. La Tm non-musculaire est la protéine la plus étudiée du cytosquelette d’actine. Une diminution

21 d’expression de cette forme de Tm est souvent associée au phénotype transformé ce qui conduit à une altération de l’organisation des filaments d’actine au niveau des cellules cancéreuses. Les filaments d’actine sont impliqués dans plusieurs fonctions cellulaires notamment la contraction musculaire, l’adhésion et la motilité cellulaire, le transport vésiculaire, l’endocytose, l’exocytose, la fonction de l’appareil de Golgi et la cytokinèse. La Tm contribue à la diversité des fonctions du cytosquelette d’actine, vu le grand nombre d’isoformes [133].

Figure 8 . Les changements d’expression des isoformes de Tm dans le cancer. Les astrocytes en division, les fibroblastes et des cellules épithéliales expriment aussi bien la Tm de haut poids moléculaire (HMW) que la Tm de bas poids moléculaire (LMW). Suite à la transformation cellulaire et à l’acquisition du potentiel de malignité, les cellules perdent l'expression de la Tm de haut poids moléculaire en général et de la Tm1 en particulier. L'expression ectopique de la Tm1 peut inverser la transformation des cellules NIH3T3 et des cellules cancéreuses du sein. Au niveau du cerveau, les astrocytes quiescentes n'expriment aucune isoforme de Tm HMW, l’expression de ces dernières induit l'entrée de ces cellules dans le cycle cellulaire. Source : Gunning P. et al, 2008 [133].

1.8.1. Les isoformes de la tropomyosine

Les mammifères possèdent 4 gènes (α, β, γ, δ) de tropomyosine (Tm) (Figure 9), lesquels génèrent plus de 40 isoformes d’épissage alternatif [142-145]. Les gènes ont une organisation similaire, mais il est intéressant de noter que chez les mammifères, les gènes montrent des variations entre les espèces au niveau des exons fonctionnels. Chez l’Homme, les gènes α, β, γ, δ sont connus comme étant TPM1 (localisé sur le chromosome 15q22) [146], TPM2 (localisé sur le chromosome 9p13) [147], TPM3 (localisé sur le chromosome 1q22) [148] et TPM4 (localisé sur le chromosome 19p13) [149], respectivement. La transcription est initiée par l’exon 1a ou 1b. Les cellules non-musculaires expriment des isoformes de Tm de haut poids moléculaire (HMW : 284 acides aminées ; présence de l’exon 1a et 2), notamment la Tm1, Tm2, Tm3 et Tm6, et des isoformes de bas poids moléculaire (LMW : 248 acides aminées ; présence de l’exon 1b) incluant la Tm4, Tm5 (NM1), Tm-5a et Tm-5b. Les isoformes de Tm de haut poids moléculaire sont impliqués dans la transformation oncogénique. L’altération de leur expression a été démontrée dans plusieurs lignées cellulaires transformées menant ainsi à une désorganisation des fibres d’actine [150, 151]. Plusieurs études ont montré que l’expression de la Tm1 est perdue dans les cellules cancéreuses métastatiques du sein [130, 131]. Les isoformes de bas poids moléculaire peuvent aussi être impliquées dans le développement tumoral. Lin et al ont identifié une nouvelle isoforme de bas poids moléculaire TC22 fortement associée au cancer du côlon [152].

Figure 9. Les isoformes de tropomyosine (Tm) sont produites suite à l’épissage alternatif de différents gènes. L'organisation des quatre gènes chez les mammifères est remarquablement semblable. Chez l’Homme, les gènes α-, β-, γ-, and δ-Tm, correspondent à TPM 1, 2, 3 et 4. Tous ces produits ont été vérifiés par Northern et Western blots, l’ancienne nomenclature des isoformes est incluse, où c'est nécessaire. Il y a un certain nombre d’ARNm qui sont détectés seulement par RT-PCR et qu’on ne montre pas sur cette figure. Source : Gunning P. et al, 2008 [133].

1.9. La relation entre la tropomyosine et le cytosquelette d’actine

Dans plusieurs cas, la diminution d’expression de Tm de haut poids moléculaire observée dans les cancers humains est associée à l’inhibition de la formation des fibres d’actine au niveau des cellules transformées. Aussi, l’expression ectopique de Tm de haut poids moléculaire dans les fibroblastes transformés Ras et Src restore la formation des fibres d’actine et diminue significativement la motilité cellulaire [153, 154]. Par contre, les études faites jusqu’à présent ne sont pas toutes capable de montrer que l’expression ectopique de la Tm renverse la transformation associée au changement du cytosquelette d’actine. Yager et al montrent que l’expression exogène de la Tm1 au niveau des cellules du neuroblastome ne mène pas à l’organisation des filaments d’actine [155]. De même, Shields et al sont aussi incapable de montrer l’effet de l’expression stable de Tm de haut poids moléculaire dans des cellules épithéliales transformées RIE-1 [131]. Ces études suggèrent donc que d’autres facteurs sont impliqués.

1.10. Le niveau d’expression de la tropomyosine dans les tumeurs humaines

Plusieurs études ont montré que l’expression de la Tm varie dans des échantillons de tumeurs humaines. Il a même été démontré que dans des stades avancés de cancers du sein, de prostate et du cerveau, la Tm est faiblement exprimée en comparaison avec un tissu normal [132, 156, 157]. Des études portant sur des tumeurs de sein montrent que l’expression de la Tm de haut poids moléculaire diminue dans les lésions malignes du sein comparée au tissu bénin ou normal [158]. Une autre étude montre que la Tm1 est surexprimée dans des cancers primaires de sein avec métastases [132]. Pawlak et al ont trouvé que l’expression de Tm1, Tm2 et Tm3 diminue dans les cellules cancéreuses de la vessie [159]. Des études portant sur des tumeurs du système nerveux central montrent que les tumeurs astrocytiques, présentant un stade avancé, n’expriment pas de Tm à haut poids moléculaire en comparaison avec des tumeurs de stade primaire, ce qui suggère que le stade de la tumeur peut être corrélé à

27 l’expression de Tm de haut poids moléculaire [156]. Une autre étude montre une augmentation d’expression de Tm de haut poids moléculaire dans des astrocytes néoplasiques en comparaison avec des astrocytes normaux [160]. Ces études suggèrent que l’altération de l’expression de la tropomyosine est un aspect important dans la biologie tumorale. Il est possible que ces différences d’expression de Tm de haut poids moléculaire dans les tumeurs soient dépendantes du type cellulaire, étant donné que chaque cellule exprime ses propres isoformes de Tm. D’autres études sont nécessaires pour caractériser différentes tumeurs afin de mieux comprendre les changements d’expression de Tm dans les cancers humains.

1.11. Les modifications post-traductionnelles de la tropomyosine

La Tm subit deux formes de modifications post-traductionnelles; l’acétylation NH2- terminale et la phosphorylation. L’acétylation est essentielle pour le fonctionnement normal de l’α-tropomyosine musculaire, permettant ainsi une forte interaction avec les filaments d’actine [161, 162]. L’acétylation NH2-terminale est une modification constitutive de la Tm et non pas un mécanisme physiologique qui régule la fonction de la Tm. À l’inverse, la phosphorylation de la tropomyosine est liée à la régulation de sa fonction. La phosphorylation de la tropomyosine de haut poids moléculaire par la PI3K joue un rôle dans la régulation de l’endocytose [163]. La phosphorylation de la Tm2 non-musclaire au niveau de la Ser61 par la phosphoinoside 3-kinase se produit à un stade tardif de l'endocytose, ce qui est cohérent avec un rôle dans la polymérisation de l'actine [163]. La tropomyosine est aussi phosphorylée au niveau du muscle lisse, squelettique et cardiaque, cette modification module son interaction avec d’autres protéines comme la caldesmon et HSP27 [164, 165]. La phosphorylation de la tropomyosine joue un rôle de modulateur au niveau du muscle cardiaque [166, 167]. Dans les cellules endothéliales, il a été démontré que la Tm1 alpha est phosphorylée sur la Ser283 par la DAP kinase en aval de la voie ERK, suite à un stress oxydatif, induisant ainsi la formation des fibres d’actine et conférant une résistance au stress oxydatif médié par le blebbing membranaire [168-171]. La phosphorylation de la Tm-1α est

donc nécessaire au maintien de l’intégrité de l’endothélium vasculaire.

1.12. Les mécanismes de régulation de la tropomyosine

Le mécanisme par lequel l’expression de la Tm est régulée dans les cellules transformées est encore mal compris. Des études montrent que la diminution d’expression de la Tm est due à une méthylation du promoteur ou à une régulation par un micro ARN [158, 172, 173]. Bharadwaj et al ont montré que dans des lignées cancéreuses de sein, la Tm1 est réprimée suite à une méthylation du promoteur [172]. D’autres études suggèrent que l’expression de la Tm1 est réprimée par le micro ARN-21. Le mir21 a récemment été identifié comme oncogène suite à sa surexpression dans plusieurs types de tumeurs [174, 175].

1.13. Les hypothèses de recherche

Le premier projet de recherche porte sur l’effet de la 17β-HSD1 dans la modulation du cancer du sein hormono-dépendant. La 17β-HSD1 joue un rôle majeur dans la progression du cancer du sein de type hormono-dépendant étant donné qu’elle est responsable de la conversion de l’estrone (E1) en estradiol (E2). Il est bien connu que l’E2 induit la prolifération des cellules tumorales menant ainsi à la progression du cancer du sein. Pour cette étude deux lignées de cellules cancéreuses hormono-dépendantes du sein ont été utilisées notamment les cellules T47D et MCF7. Ces cellules expriment à la fois les récepteurs des estrogènes et des androgènes. En se basant sur toutes ces données, nous avons émis les hypothèses suivantes :

- L’inhibition de la 17β-HSD1 module le niveau d’E2 dans les cellules cancéreuses du sein T47D et MCF7.

- La baisse du niveau d’E2 suite à l’inhibition de la 17β-HSD1 mène à la diminution de la prolifération des cellules cancéreuses T47D et MCF7.

- L’inhibition de la 17β-HSD1 a un effet sur le cycle cellulaire, la migration et l’invasion des cellules cancéreuses T47D et MCF7.

- Il y a un changement du profil protéique des cellules cancéreuses T47D et MCF7 suite à l’inhibition de la 17β-HSD1.

- L’inhibition de la 17β-HSD1 mène à la modulation du profil génomique des cellules cancéreuses T47D.

Le deuxième projet de recherche porte sur l’effet de la phosphorylation de la Tm1α sur la progression du cancer du sein. Des travaux antérieurs du laboratoire ont montré que l’activation de la voie ERK par le stress oxydant conduit à la phosphorylation rapide de la

TM1α des cellules endothéliales [169]. Les données de la littérature montrent clairement que la Tm joue un rôle crucial dans le cancer du sein. Néanmoins, les mécanismes intimes de la régulation de la progression du cancer du sein par la Tm sont compliqués par le fait que cette protéine existe sous différentes isoformes dont certaines subissent des modifications post- traductionnelles comme la phosphorylation et l’acétylation. Pour réaliser ce projet, la lignée cellulaire MDA MB231 a été utilisée. Nous avons donc généré des transfectants MDA MB231 stables exprimant soit la tropomyosine-1α sauvage, soit la tropomyosine-1α mutante non-phosphorylable (Ser283Ala) ou encore la tropomyosine-1α mutante pseudophosphorylée (Ser283Glu). Par la suite, ces cellules ont été utilisées pour vérifier différents paramètres de la transformation néoplasique. Pour ce deuxième projet, nous avons donc émis les hypothèses suivantes :

1) La phosphorylation de la Tm-1α renforce ses propriétés de suppresseur de tumeur.

2) La phosphorylation de la Tm-1α est impliquée dans la modulation et la progression du cancer du sein.

1.14. Les objectifs de travail

Suite aux hypothèses émises, plusieurs objectifs ont été vérifiés :

 Objectif 1 : Étudier les changements protéiques des cellules cancéreuses du sein en

réponse à l’inhibition de la 17β-HSDl.

 Objectif 2 : Établir le rôle direct de la 17β-HSD1 dans le développement du cancer du sein hormono-dépendant en précisant l'impact de son inhibition et de sa surexpression sur la prolifération, l’invasion, la migration cellulaire, la formation de colonies, la concentration d’estradiol, l’apoptose et le cycle cellulaire des cellules cancéreuses humaines du sein T47D et MCF7.

 Objectif 3 : Voir l’effet de l’inhibition de la 17β-HSDl sur le profil génomique des cellules cancéreuses T47D.

 Objectif 4 : Déterminer l’effet de la phosphorylation de la tropomyosine 1 alpha sur la prolifération, l’adhérence, la formation de colonies et la migration des cellules cancéreuses du sein MDA MB231.

Chapitre 2: Differential proteome of estrogen-dependent breast cancer cells by 17β-HSD type 1 inhibition and knockdown: modulation of nm23-H1 expression, cell cycle and cell invasion

Differential proteome of estrogen-dependent breast cancer cells by 17β-HSD type 1 inhibition and knockdown: modulation of nm23-H1 expression, cell cycle and cell invasion

Mouna Zerradi1 and Sheng-Xiang Lin1*

1 Laboratory of Molecular Endocrinology and oncology, CHU (CHUL) Research Centre and Laval University, Québec, Canada G1V 4G2;

*Corresponding author: Sheng-Xiang Lin, Endocrinology and Nephrology Axe, CHUL Research Center, 2705 Boulevard Laurier, Sainte-Foy, Québec, Canada G1V 4G2. Tel.: 418 654 2296; Fax: 418 654 2761; E-mail: [email protected]

Cet article est soumis dans le journal Genome Medicine.

Running title The inhibition of 17β-HSD1 in breast cancer cells

Abbreviations 17β-HSD1: 17beta-hydroxysteroid dehydrogenase type 1; 4-dione: 4-androstenedione; E1-S: estrone sulfate; E1: estrone; E2: estradiol; ER+: estrogen receptor positive; FBS: fetal bovine serum; BC: breast cancer; INH: inhibitor; IPA: Ingenuity Pathway analysis; MCF7-1: MCF7- 17β-HSD1; MS: mass spectrometry; NADPH: nicotinamide adenine dinucleotide phosphate; PBS: phosphate-buffered saline; PBS-T: PBS-tween 20; RT-qPCR: reverse transcription quantitative real-time polymerase chain reaction; SDR: dehydrogenases/reductases short chain; SDS-PAGE: sodium dodecyl sulphate-polyacrylamide gel electrophoresis; STS: steroid sulfatase; T: testosterone; TLC: thin layer chromatography.

Summary

Background. The enzyme 17β-hydroxysteroid steroid dehydrogenase type 1 catalyzes estradiol formation and plays a major role in the induction and progression of hormone- dependent breast cancer (ER+ BC). The aim of the study was to characterize the effect of 17β-HSD1 inhibition on the proteome, cell cycle, apoptosis and invasiveness of breast cancer cells. Methods. The T47D and MCF7 cell lines, representative for ER+ BC, were used to perform all experiments. A proteomic analysis was performed, by using 2D gel electrophoresis, to compare the proteome of T47D cells treated with 17β-HSD1 inhibitor or transfected with 17β-HSD1 siRNA and control T47D cells, as well as the proteome of MCF7 cells transfected with 17β-HSD1 siRNA and control MCF7 cells. Cell proliferation, apoptosis, cell cycle, estradiol formation and cell invasion were measured. Results. In T47D and MCF7 cells, the inhibition of 17β-HSD1 modulated several proteins such as tumor protein D54 (TPD54), involved in cell proliferation, 14-3-3 protein epsilon (YWHAE) and tumor protein D53 (TPD53), involved in cell cycle regulation, and the metastasis inhibition factor nm23 (nm23-H1) involved in cell invasion. In MCF7 cells, the software Ingenuity Pathway analysis (IPA) analysis associated those identified proteins to 8 networks, 20 significant pathways and four categories of biological functions such as apoptosis, necrosis, cell death and cell survival. In T47D, following inhibition of 17β-HSD1 by a specific inhibitor or siRNA, proliferation decreased by 42% and 33% respectively. Decreases in colony formation of 30% and 40%, in estradiol formation of 74% and 56% and in cell invasion of 35% and 31% respectively, were also observed. Our results show that MCF7-17β-HSD1 cells grow faster (25%), form more estradiol (70%) and are more invasive (31%) than wild-type MCF7 cells. Cell cycle results revealed a significant reduction of cells in G0/G1 phase at MCF7-17βHSD1 in comparison to MCF7 cells, and a significant increase of cells in S phase and G2/M. Apoptotic results show an increase in MCF7-17β-HSD1 living cells and a decrease in MCF7-17β-HSD1 apoptotic cells compared to MCF7 cells. Conclusion. This study shows that the inhibition of 17β-HSD1 affect several signaling pathways by regulating important networks and proteins with critical functions. 17β-HSD1

37 regulates cell cycle, proliferation and invasiveness of MCF7 and T47D cells. These results further support the nature of drug target of the enzyme in breast cancer.

Résumé en français

Introduction. L’enzyme stéroïdogénique 17β-hydroxystéroïde déshydrogénase type 1 catalyse la formation de l'estradiol, jouant ainsi un rôle majeur dans l’induction et la progression des cancers du sein de type hormono-dépendants. L’objectif de l’étude était de caractériser l’effet de l’inhibition de la 17β-HSD1 sur le protéome, le cycle cellulaire, l’apoptose et la capacité d’invasion des cellules du cancer du sein. Méthodologie. Les deux lignées cellulaires T47D et MCF7 ont été utilisées pour réaliser toutes les expériences. L’analyse protéomique a été réalisée en utilisant l’électrophorèse bidimensionnelle pour comparer le protéome des cellules T47D contrôles à celui des cellules traitées avec un inhibiteur stéroïdien ou transfectées avec un siRNA spécifique contre la 17β- HSD1 et le protéome des cellules MCF7 contrôles a été comparé à celui des cellules transfectées avec un siRNA spécifique contre la 17β-HSD1. La prolifération, l’apoptose et le cycle cellulaire ont été étudiés en plus de la formation d’estradiol et l’invasion cellulaire. Résultats. Dans les cellules T47D et MCF7, l'inhibition de la 17β-HSD1 module l’expression de plusieurs protéines telles que la tumor protein D54 (TPD54), impliquée dans la la prolifération cellulaire, la 14-3-3 protein epsilon (YWHAE) et la tumor protein D53 (TPD53), impliquées dans la régulation du cycle cellulaire, et le metastasis inhibition factor nm23 (nm23-H1) impliquée dans l’invasion cellulaire. Dans les cellules MCF7, le logiciel Ingenuity Pathway analysis (IPA) a associé ces protéines identifiées à 8 réseaux, 20 pathways et quatre catégories de fonctions biologiques tels que l'apoptose, la nécrose, la mort cellulaire et la survie cellulaire. Suite à l’inhibition de la 17β-HSD1 par un inhibiteur ou un siRNA spécifique, on a noté une diminution de la prolifération des cellules T47D de 42% et 33%, de la formation de colonies de 30% et 40%, de la formation d’estradiol de 74% et 56% et de l’invasion cellulaire de 35% et 31%, respectivement. Nos résultats montrent aussi que les cellules MCF7-17β-HSD1 prolifèrent plus rapidement (25%), forment plus d’estradiol (70%) et sont plus invasives (31%) que les cellules MCF7 sauvages. Les résultats du cycle cellulaire ont montré une diminution significative des cellules MCF7-17βHSD1 en phase GO/G1 en comparaison avec les cellules MCF7, ainsi qu’une augmentation significative des cellules en phase S et G2/M. Les résultats d’apoptose montrent une augmentation des cellules

39 vivantes MCF7-17βHSD1 et une diminution des cellules apoptotiques MCF7-17βHSD1 en comparaison avec les cellules MCF7. Conclusion. Cette étude montre que l’inhibition de la 17β-HSD1 par un inhibiteur ou un siRNA spécifique touche des voies de signalisation tout en régulant différentes protéines ayant d’importantes fonctions physiologiques. On a aussi montré pour la première fois que la 17β-HSD1 régule le cycle cellulaire et l’invasion cellulaire des cellules cancéreuses du sein T47D et MCF7.

1. Introduction

Breast cancer is the most frequently diagnosed cancer and the leading cause of cancer death in females worldwide (1). In women, breast cancer is expected to account for 29% of all new cancers, accounting for 231,840 of total estimated numbers of new cases and 40,290 of total cancer deaths in 2015(2). In Canada and United States, lung cancer is the first cause of cancer death in women followed by breast cancer (3). In general, incidence rates are high in Western and Northern Europe, Australia/New Zealand, and North America; intermediate in South America, the Caribbean, and Northern Africa; and low in sub-Saharan Africa and Asia (1). Eight breast cancer cases of ten occur in women aged 50 years or more. In 2013, an estimated 6,000 women in Quebec (23,800 in Canada) will be diagnosed with breast cancer. About 1,350 women (5,000 in Canada) will die of the disease (4). Despite its high incidence, early detection and modern treatments have significantly increased patient survival.

Several observations from epidemiological studies indicate that certain factors increase the risk of breast cancer in women. Hereditary factor plays a role in the development of breast cancer. For example, cancers are caused by the presence of BRCA1 and BRCA2 genes in some individuals and 3% to 8% of all women with breast cancer will be found to carry a mutation in BRCA1 or BRCA2 genes (5). Estrogen-dependent breast cancer is still overwhelming, representing about 60% and 75% of premenopausal and postmenopausal patients, respectively (6), in such cases steroid-converting enzymes in peripheral play critical roles.

The major determinant of breast cancer is suggested to be prolonged exposure to high levels of estrogens (7). Estradiol (E2) is a steroid hormone which can regulate the differentiation and proliferation of normal breast epithelial cells. Most breast cancers are hormone-dependent (8), in which E2 plays a major role in the induction and progression of hormone-dependent breast cancers (9). Also, there are other risk factors for breast cancer such as early menstruation, late pregnancy, late menopause, prolonged use of oral contraceptives and hormone replacement therapy (10-12). It has been demonstrated that E2

41 concentration is 2.3 times higher in the breast cancer tissue in comparison with normal tissue (13).

In premenopausal women, the ovary is the major source of estrogens, and cancers take these produced estrogens to stimulate tumor growth (14). Premenopausal women have a serum E2 levels ten times higher than postmenopausal women (15, 16). However, postmenopausal women with breast cancer have an equivalent intratumoral E2 levels to that of premenopausal women (16). In postmenopausal women, the intratumoral biosynthesis of estrogens is supposed to maintain the high intratumoral E2 levels (17-19). The intratumoral ratio estradiol/estrone is significantly higher in postmenopausal women compared to premenopausal women with ER+ breast cancer (14, 20).

The majority of E2 is produced by two biosynthetic pathways: aromatase and steroid sulfatase pathways. Aromatase catalyzes the conversion of testosterone (T) to E2 and the conversion of 4-androstenedione (4-dione) to estrone (El). In the sulfatase pathway, estrone sulfate (E1-S) is converted to El by steroid sulfatase (STS), followed by an activation step to produce E2. The 17β-hydroxysteroid dehydrogenases (17β-HSDs) are involved in both biosynthetic pathways (21, 22). Human 17β-HSD1 is an important enzyme in breast cancer development because it catalyzes the conversion of E1 to E2 with nicotinamide adenine dinucleotide phosphate (NADPH) as a cofactor (23-27). In addition, the enzyme inactivates DHT, and thus additionally stimulates breast cancer proliferation (28). The gene encoding for the 17βHSD1 (HSD17β1) is located on chromosome 17q12-21 (26). The enzyme is a member of dehydrogenases/reductases short chain (SDR) family and is localized in the cytoplasm and mainly expressed in the placenta, ovaries and mammary glands (29, 30). In addition to the synthesis of E2 that stimulates the growth of breast cancer cells (31, 32).

In breast cancer, the 17β-HSD1 expression and activity are significantly higher than in normal breast tissue, thus, leading to tumor stimulation and development (33-35). Also, a high 17β-HSD1 mRNA expression is significantly associated with poor prognosis in breast cancer (34). The mRNA level of 17β-HSD1 is higher in postmenopausal women compared to premenopausal women with breast cancer (14). These data have clearly shown that the high

intratumoral E2 levels in postmenopausal patients are due to high expression of 17β-HSD1 (14, 36). The upregulated 17β-HSD1 will allow a higher E1 conversion to E2 in postmenopausal patients, thus explaining the fact that postmenopausal women with breast cancer have a higher intratumoral E2/E1 ratio than premenopausal patients (14). All these data show the importance of possible application of 17β-HSD1 in ER+ breast cancer treatment.

For this study, we used T47D and MCF7 cells, hormone-dependent breast cancer cells that express both the estrogen and androgen receptors. These cell lines are widely used for the studies of hormone-dependent breast cancer when tested in vitro and in vivo. T47D strongly expressed 17β-HSD1, while MCF7 shows a lower level of 17β-HSD1 (37).

In the present study, we investigated the role of 17β-HSD1 inhibition and expression on the proteome profile, cell cycle, apoptosis and invasiveness of breast cancer cells.

2. Materials and methods

2.1. Cell culture

MCF7 and T47D cells were purchased from the American Type Culture Collection (ATCC, Manassas, USA). T47D cells were cultured in DMEM high-glucose phenol red-free, supplemented with 7.5 mg/L bovine insulin (Sigma, Oakville, Ontario, Canada). MCF7 cells were cultured in DMEM low glucose phenol red-free. The cells were cultured in the presence of 10% FBS at 37°C in a humidified atmosphere of 95% air and 5% CO2. For hormone-free medium, FBS was treated overnight with 2% dextran-coated charcoal at 4oC to remove the residual steroids present in the serum.

2.2. Plasmid construction and stable transfection

The recombinant plasmid containing 17β-HSD1 cDNA and the stably-transfected MCF7 6 cells (MCF-1 cells) were generated as previously described (28). 1xl0 MCF7 cells were then transfected in 6-well plates with the recombinant plasmid (1 pg) by using Lipofectamine 2000 (Invitrogen, Burlington, Ontario, Canada). Stable transfectants were selected for four weeks with 500 ug/mL of G-418 (Invitrogen, Burlington, Ontario, Canada) (28).

2.3. siRNA synthesis and transfection

The sense and antisense sequences of three 17β-HSD1 siRNAs were selected and synthesized as previously described (28). Two days before transfection, MCF7, MCF7-1 or T47D cells were cultured in a cell medium. 3x105 cells were transfected in 6-well plates with 100-200 nM mixed 17β-HSD1 specific siRNAs or control siRNA (scramble siRNA) (GenePharma, Shanghai, China) using Lipofectamine 2000 (Invitrogen, Burlington, Ontario, Canada). Transfected cells were cultured for two to four days before experiments.

2.4. 17β-HSD1 steroidal inhibitor

The 2-methoxy-E2 C-16 derivative inhibitor of 17β-HSD1 (compound 18) (38) was used in the present study as previously described (28). The 2-methoxy-E2 C-16 derivative inhibitor, with an IC50 value of 72 nm, shows 17β-HSD1 activity an inhibition of 81%, 69%, and 37% at 10, 1, and 0.1 µM, respectively, in T47D cells (38).

2.5. Western blot

Total proteins were extracted from cells with RIPA buffer supplemented with 1% protease inhibitors cocktail (EMD Chemicals, Gibbs-town, USA). The Bradford method was used to quantify proteins. The same amounts of proteins from each sample were separated on 12% SDS-PAGE, and the gels were transferred into nitrocellulose membranes for Western blotting. Membranes were blocked with 5% non-fat milk in PBS-Tween 20 for one hour at room temperature. Membranes were then incubated in blocking buffer containing 1:75000 dilution of the primary anti-17β-HSD1 monoclonal rabbit antibody (Abcam, Cambridge, MA, USA) for two hours at room temperature. For controls, membranes were incubated in blocking buffer containing 1:7500 dilution of anti-actin monoclonal mouse antibody (Abcam, Cambridge, MA, USA) for two hours at room temperature. Membranes were then washed and incubated in blocking buffer containing 1:10 000 dilution of the respective horseradish peroxidase-conjugated secondary antibody (anti-IgG antibody) (Santa Cruz Biotechnology, Santa Cruz, CA, USA) for one hour at room temperature. Protein signals were visualized with chemiluminescence Reagent (PerkinElmer, Waltham, MA, USA).

2.6. Activity assay

25,000 T47D cells were plated in 24-well plates and cultured in medium treated with dextran-coated charcoal. Enzyme activity experiments for T47D cells were carried out after treatment with various concentrations of 17β-HSD1 inhibitor. A final concentration of 0.2 µM [14C]-E1 (American Radiolabeled Chemicals, St Louis, USA) was added to the culture medium and the reaction was stopped at 0 min, 10 min, 30 min and 60 min. Then, steroids 45 were extracted by using 3 volumes of diethyl ether. The organic phase was evaporated, and the residue dissolved in 50 µL dichloromethane and transferred on Silica gel 60 thin layer chromatography plates (Merck, Darmstad, Germany). Steroid migration was performed in toluene-acetone (4:1, v/v). TLC plates were let to dry and quantified with storm imaging system (Molecular Dynamics, Sunnyvale, CA, USA).

2.7. Cell proliferation assay

Cell proliferation was evaluated with the Cyquant cell proliferation assay kit. Six thousand T47D, MCF7 and MCF7-1 cells were plated in 96-well plates and incubated at 37°C. At the desired time, the medium was removed from wells and washed once with phosphate buffered saline (PBS). Cells in the microplate were frozen and stored at 70°C until the day of experiment. For sample quantification, plates were thawed at room temperature, then 200 μL of the CyQUANT® GR dye/cell-lysis buffer were added to each sample well and incubated 5 minutes at room temperature. Absorbance of each well was measured at 480 nm excitation and 520 nm emission.

2.8. Colony formation

Three hundred T47D cells were transfected with 100 nM control and 17β-HSD1 siRNA as described above or treated with17β-HSD1 inhibitor and vehicle ethanol. Cells were cultured in the presence of 10% FBS and incubated for 14 days. Thereafter, cells were washed once with PBS, fixed and stained with trypan blue (0.25% trypan blue; 3% acetic acid) (Sigma- Aldrich, Oakville, Ontario, Canada) for 30 minute at room temperature. Then, they were washed twice with water and counted.

2.9. Cell apoptosis assay

The presence of apoptotic, viable or necrotic cells were assessed by adding annexin V- allophycocyanin (APC) and propidium iodide (PI) to the cells (Invitrogen, Burlington, Ontario, Canada). To do so, 500,000 T47D, MCF7 and MCF7-1 cells were plated in 6-well plates and incubated for 24 hours. Cells were then treated with vehicle ethanol or 17β-HSD1 specific inhibitor for another 4 days or transfected with control siRNA or 17β-HSD1 specific siRNA for another 4 days. For positive controls, cells were treated with 3% hydrogen peroxide (H2O2) for 30 minute to induce apoptosis or fixed with pre-cold ethanol for 1 hour at -20°C to induce necrosis. A negative control was prepared by incubating cells in the absence of inducing agent. Positive and negative controls as well as samples were then washed with PBS 1X and resuspended in 500 mL annexin-binding buffer (10 mM HEPES,

140 mM NaCl, and 2.5 mM CaCl2, pH 7.4). Then, 10 μL of annexin V-APC and 1 μL of PI (20 μg/mL) were added to the corresponding positive controls and to the samples and incubated for 30 minute on ice in the dark. Samples were then analyzed by flow cytometry by using BD FACSCanto II (BD Biosciences, Ontario, Canada). The analysis was done with BD FACSDiva 6.1.2 software (BD Biosciences, Ontario, Canada). This assay was done in duplicate.

2.10. Cell cycle analysis

500 000 T47D, MCF7 and MCF7-1 cells were plated in 6-well plates and incubated for 24 hours. Cells were then treated with vehicle ethanol or 17β-HSD1 specific inhibitor for another 4 days or transfected with control siRNA or 17β-HSD1 specific siRNA for another 4 days. Cells were washed with PBS 1X and resuspended in the corresponding medium. Then, cells were washed and resuspended with PBS 0.5% FBS. 70% chilled ethanol was added to the cells for 1 hour at -20°C. Cells were then washed with PBS 1X and resuspended with 500 µL PI staining buffer (50 µg/mL PI, 100 µg/mL RNAse A, 0.2% Triton X-100 and PBS 1X). Samples were incubated at 4°C for 30minute in the dark and then analyzed by flow cytometry using BD FACSCanto II (BD Biosciences, Ontario, Canada). Data analysis was

47 done using BD FACSDiva 6.1.2 software (BD Biosciences, Ontario, Canada). This assay was performed in duplicate.

2.11. Cell invasion

Cell invasion was assessed using a Cell Invasion Assay Kit (Cell biolabs, Inc, San Diego, CA, USA). T47D, MCF7 and MC7-1 cell suspension containing 0.5 x 106 cells/mL in serum free media (DMEM, 10 mM HEPES pH 7.4, 1 mM MgCl2, and 0.5 % bovine serum albumin) were prepared and 0.3 mL of the cell suspension solution were added to the insert.

Plates were incubated for 48h at 37°C in a 5% CO2 atmosphere. Cells in the upper part of the chamber were removed with a cotton swab. The insert was then transfered to a clean well containing 400 μL of Cell Stain Solution and incubated for 10 minutes at room temperature. The stained inserts were washed several times in a beaker of water and transfered to an empty well and 200 μL of extraction solution were added per well. Wells were then incubated 10 minutes on an orbital shaker. 100 μL from each sample were transferred to a 96-well microtiter plate and absorbance was measured at 560 nm.

2.12. Determination of E2 level

6000 T47D cells were transfected with 100 nM control and 17β-HSD1 siRNA as described above or treated with17β-HSD1 inhibitor. T47D, MCF7 and MCF7-1 cell supernatants were collected from wells after 96 hours and frozen at -80°C until the day of the experiment. A commercial estradiol enzyme immunoassay kit (Cayman Chemical, Michigan, USA) was used to determine the level of E2 in T47D, MCF7 and MCF7-1 cells supernatant according to the manufacturer’s protocol. Cell supernatant was thawed and 50 µL were mixed to 50 µL tracer and 50 µL antibody. Each sample was done in duplicate wells. Plates were read at 420 nm in a plate reader (Spectra Max 340PC, Molecular Devices). For the control, medium without cells was used to determine basal E2 level.

2.13. Protein extraction

T47D and MCF7 cells were plated in 10 cm diameter dishes. After 24 hours, cells were treated with ethanol vehicle or 17β-HSD1 specific inhibitor for another 8 days or transfected with control siRNA or 17β-HSD1 specific siRNA for another 4 days. After treatments, cells were washed twice with PBS 1x and proteins were extracted with 500 L lysis buffer T8 (Urea 7M, Thiourea 2M, 3-[(3-Cholamidopropyl) dimethylammonio]-1-propanesulfonate (CHAPS) 3%, 1,4-Dimercapto-2,3-butanediol (DTT) 20 mM, Tris (2-carboxyethyl) phosphine hydrochloride (TCEP) 5 mM) containing 1% protease inhibitors cocktail (EMD Chemicals, Gibbs-town, NJ, USA) and 50 mM Tris-HCl pH 8.8. Cells were then incubated for 2 hours at room temperature and precipitated with 2-D clean-up kit (GE Healthcare, Piscataway, NJ, USA). The 2-D Quant Kit was used to determine protein concentration (GE Healthcare, Piscataway, NJ, USA). The experience was done in four independent biological replicates for each condition (39).

2.14. Two-dimensional gel electrophoresis and image analysis

The two-dimensional gel electrophoresis was performed at the Proteomic platform of Infectious Disease Research Center (Québec, Canada). 200 µg of protein were loaded on to 24 cm Immobiline Dry Strip (GE Healthcare, Piscataway, NJ, USA) pH 4-7 on IPGPhor isoelectric focusing system (GE Healthcare, Piscataway, NJ, USA) for first gel dimension as recommended by the manufacturer. Strips were then incubated in equilibration buffer (50 mM Tris-Cl, pH 8.8, 6 M urea, 30% glycerol, 2%SDS, trace of bromphenol blue) containing 10 mg/mL dithiothreitol for 15 min and then in equilibration buffer containing 25 mg/mL iodoacetamide for 15 min. The second dimension was run on 2D gel 12% acrylamide gel using Ettan Dalt twelve System (GE Healthcare, Piscataway, NJ, USA). Gels were stained with Sypro Ruby (Invitrogen, Burlington, Ontario, Canada) and scanned by using ProXpress scanner (Perkin Elmer, Waltham, MA, USA). Comparative analysis of combination of 4 replicates of T47D control cells or T47D treated cells and 4 replicates of MCF7 control cells or MCF7 transfected cells with 17β-HSD1 siRNA are done by using Progenesis Same Spots software (Nonlinear Dynamics). Spots of interest (fold change 1.5 or higher and anova value

49 p ≤ 0.05) were cut from gels using a ProXcision_Spot cutter (Perkin Elmer, Waltham, MA, USA), conserved in 1% acetic acid and submited to trypsin digestion before mass spectrometry analysis.

2.15. Mass spectrometry analysis and protein identification

Mass spectrometry (MS) experiments were performed by the Proteomics platform of the Eastern Quebec Genomics Center (Québec, Canada). Spots of interest were extracted from gels and placed in 96-well plates and then washed with water. Tryptic digestion was performed on a MassPrep liquid handling robot (Waters, Milford, MA, USA) according to the manufacturer’s specifications and to the protocol of Shevchenko et al (40) with the modifications suggested by Havlis et al (41). Peptide samples were separated by online reversed-phase (RP) nanoscale capillary liquid chromatography (nanoLC) and analyzed by electrospray mass spectrometry (ES MS/MS). The experiments were performed with a Agilent 1200 nano pump connected to a 5600 mass spectrometer (AB Sciex, Framingham, MA, USA) equipped with a nanoelectrospray ion source. Peptide separation took place on a self-packed PicoFrit column (New Objective, Woburn, MA, USA) packed with Jupiter (Phenomenex) 5u, 300A C18, 15 cm x 0.075 mm internal diameter. Peptides were eluted with a linear gradient from 2-50% solvent B (acetonitrile, 0.1% formic acid) in 30 minutes, at 300 nL/min. Mass spectra were acquired using a data dependent acquisition mode using Analyst software version 1.6. Each full scan mass spectrum (400 to 2000 m/z) was followed by collision-induced dissociation of the seven most intense ions. The dynamic exclusion (30 seconds exclusion duration) function was enabled, and the relative collisional fragmentation energy was set to 35%. All MS/MS peak list were generated with Protein Pilot (AB Sciex, Framingham, MA, USA, Version 4,5) and samples MGF were analyzed using Mascot (Matrix Science, London, UK; version 2.3.02). Mascot was set up to search the Uniref100- Homo sapiens database (release 13-03) assuming the digestion enzyme trypsin. Mascot was searched with a fragment ion mass tolerance of 0.10 Da and a parent ion tolerance of 0.10 Da. Iodoacetamide derivative of cysteine was specified as a fixed modification and oxidation of methionine was specified as a variable modification. Two missed cleavages were allowed. Scaffold (version 4.0.1, Proteome Software Inc., Portland, OR, USA) was used to validate

MS/MS based peptide and protein identifications. Peptide identifications were accepted if they could be established at greater than 95.0% probability as specified by the Peptide Prophet algorithm (42). Protein identifications were accepted if they could be established at greater than 95.0% probability and contained at least 2 identified peptides. Protein probabilities were assigned by the Protein Prophet algorithm (43). Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony.

2.16. Ingenuity Pathway analysis (IPA) analysis

The proteins identified by MS analysis were analyzed with ingenuity pathway analysis (IPA) (44).

2.17. Quantitative real-time RT-PCR

Total RNA was extracted from T47D, MCF7 or MCF7-1 cells with the Qiazol Lysis Reagent (Qiagen, Hilden, DE) and assayed by spectrophotometry (Nano Drop) in duplicate. The quality of total RNA samples was verified with bioanalyzer (Agilent technologies, Mississauga, Ontario, Canada). The analysis showed high quality for all the RNA with RIN (RNA Integrity Number) greater than 9.1 on a scale of 10. The quantification by real-time PCR (qPCR) for each condition was done in duplicate as previously described (28, 45). Normalization was performed using the reference genes shown to have stable expression levels from embryonic life through adulthood in various tissues (46): ATP synthase O subunit (ATP5O), hypoxanthine guanine phosphoribosyl transferase 1 (HPRT1), glucose-6- phosphate dehydrogenase (G6PD) and 18S ribosomal RNA (18S). As additional file 1 shows the sequence of specific primers used for the amplification. Messenger RNA level was expressed as copies / µg total RNA. Quantitative Real-Time PCR measurements were performed by the CHU de Québec Research Center (CHUL) Gene Expression Platform Québec, Québec, Canada.

2.18. Statistics

Values are expressed as mean ± SD. Student t tests were used for comparison between two means. p value < 0.05 was considered as statistically significant.

3. Results

3.1. 17β-HSD1 knockdown and inhibition in T47D, MCF7 and MCF7-17β- HSD1 cells

MCF7 cells were transfected with the recombinant plasmid by using Lipofectamine 2000. Stable transfectant MCF7-17β-HSD1 were selected for four weeks with 500 ug/mL of G- 418. The transfection of 17β-HSD1 in MCF7 cells was worked because we obtain a band in stable transfectants MCF7-17β-HSD1 compared to MCF7 cells (Figure 1A). A mix of three siRNA against 17β-HSD1 and negative control siRNA were used to determine whether siRNA can knock down the expression of 17β-HSD1 in cells (28). Total protein was extracted from cells 96 h after transfection and was analyzed by Western blot using a 17β- HSD1 antibody. The specificity of 17β-HSD1 siRNA was confirmed, resulting a quasi- knock-out of the enzyme (Figure 1B).

The mRNA transcription of endogenous 17β-HSD1 in T47D, MCF7 and MCF7-17β-HSD1 was evaluated by quantitative RT-PCR. Results show that MCF7-17β-HSD1 over-expresses 17β-HSD1 in comparison to MCF7 (Figure 1C). Also, there is a decrease of 17β-HSD1 mRNA expression in MCF7-17β-HSD1 (Figure 1D) and T47D (Figure 1E) cells transfected with 17β-HSD1 siRNA in comparison with control MCF7-17β-HSD1 and T47D, respectively.

To determine the capacity of the 17β-HSD1 inhibitor, total protein extracts from T47D cells were used in activity assays to evaluate the conversion of 14C-E1 into 14C-E2 after 0, 10, 30 and 60 minutes, in the absence or presence of the inhibitor at different concentrations (3 and 52

10 µM). After 60 minutes of treatment, there was a decrease of E1 conversion to E2 of 2.1 times at an inhibitor concentration of 3 µM and 2.25 times at 10 µM. The inhibitory capacity of 17β-HSD1 inhibitor was confirmed, as there was a correlation between 17β-HSD1 expression and E1 conversion to E2 (Figure 1F).

3.2. Effect of 17β-HSD1 modulation on T47D, MCF7 and MCF7-17β-HSD1 cell proliferation

To determine the impact of 17β-HSD1 knock down on cell proliferation, T47D cells were transfected with 17β-HSD1 specific siRNA in complete growth medium. The growth of T47D cells was reduced by 33% after 96 h of transfection (Figure 2A). These observations show that knockdown of 17β-HSD1 expression in T47D decreased cell proliferation.

To investigate the effect of 17β-HSD1 activities on cell proliferation, T47D cells were treated for 8 days with various concentrations of 17β-HSD1 inhibitor in complete growth medium. Growth of T47D cells was reduced by 13, 15 and 42% with 1, 3 and 5 µM 17β-HSD1 inhibitor; respectively (Figure 2B). These results confirm that the inhibition of 17β-HSD1 activity decreased T47D cell proliferation.

MCF7 and MCF7-17β-HSD1 proliferation were evaluated by CyQuant test. MCF7-17β- HSD1 grows 25% faster than MCF7 cells (Figure 2G). MCF7 and MCF7-17β-HSD1 cells were transfected with 17β-HSD1 specific siRNA in complete growth medium. The proliferation of MCF7 cells decreased by 15% after 96 h of transfection. While, MCF7-17β- HSD1 growth decreased by 53% after 96 h of transfection (Figures 2C and 2D). These observations show that the knockdown of 17β-HSD1 expression in MCF7 and MCF7-17β- HSD1 decreased cell proliferation.

To determine the effect of 17β-HSD1 activities on breast cancer cell proliferation, MCF7 and MCF7-17β-HSD1 cells were treated for 8 days with 17β-HSD1 inhibitor in complete growth medium. The growth of MCF7 and MCF7-17β-HSD1 cells were reduced by 20% and 33%

53 with 5 µM 17β-HSD1 inhibitor, respectively (Figures 2E and 2F). These results confirm that the inhibition of 17β-HSD1 activities decreased MCF7 and MCF7-17β-HSD1 cell proliferation.

3.3. Effect of 17β-HSD1 modulation on estradiol formation and colony formation in T47D, MCF7 and MCF7-17β-HSD1 cells

The effect of 17β-HSD1 knockdown was determined by measuring E2 concentration in cell culture medium. T47D cells were transfected with 17β-HSD1 specific siRNA in complete growth medium for 4 days. After 17β-HSD1 knockdown, the level of E2 was reduced by 56 % in comparison with control T47D cell (Figure 3A). The impact of 17β-HSD1 activity inhibition was also determined by assessing E2 levels in cell culture medium. Cells were treated with 17β-HSD1 inhibitor (5 µM) in complete growth medium. After 8 days of treatment, the level of E2 was reduced by 74 % in comparison with control T47D cells (Figure 3B). These results confirm that modulation of 17β-HSD1 activity or expression in T47D affects E2 concentration.

The level of E2 in MCF7 cells was compared to that of MCF7-17β-HSD1 by performing ELISA assay using cell culture medium. E2 formation in MCF7-17β-HSD1 was 70% higher than MCF7 cells (Figure 3C). These observations show that the 17β-HSD1 expression in MCF7 cells modulates the E2 concentration.

We tested the effect of 17β-HSD1 inhibition on colony formation. As shown in figure 3D, down regulation of 17β-HSD1 by specific siRNA reduced by 40% the number of colonies formed compared to T47D control. Inhibition of 17β-HSD1 with the inhibitor reduced by 30% the number of colonies formed compared to the T47D control (Figure 3E).

3.4. Effect of 17β-HSD1 modulation on T47D, MCF7 and MCF7-17β-HSD1 cell invasion

The cell invasion assay was performed by using the CytoSelect™ Cell Invasion Assay Kit to evaluate the effect of 17β-HSD1 expression on T47D, MCF7 and MCF7-17β-HSD1 cell invasion. As shown in Figure 4A, cell invasion of T47D cells transfected with 17β-HSD1- specific siRNA was reduced by 31%. Inhibition of 17β-HSD1 by the specific inhibitor reduced by 35% the number of invasive cell (Figure 4B).

MCF7-17β-HSD1 cell invasion was 31% higher than MCF7 cells (Figure 4E). The cell invasion of MCF7-17β-HSD1 cells transfected with 17β-HSD1-specific siRNA was reduced by 29% (Figure 4C). The inhibition of 17β-HSD1 by specific inhibitor reduced MCF7-17β- HSD1 cell invasion by 31% (Figure 4D). These results indicate that there is a positive correlation between 17β-HSD1 expression and MCF7 cell invasion.

3.5. Effect of 17β-HSD1 inhibition and down regulation on the cell cycle and apoptosis in T47D, MCF7 and MCF7-17β-HSD1 cells

As shown in Figures 5A and 5C, the percentage of T47D cells transfected with siRNA in the G1/G0 phase was 69% compared with 60% of the control T47D, and in the G2 phase was 21% compared with 27%, while the percentage of the cells in S phase was 9% compared with 12% (p < 0.05). Following inhibition of 17β-HSD1 by the specific inhibitor in complete growth medium, the percentage of T47D cells treated with inhibitor in the G1/G0 phase was 68% compared with 63% of the control T47D, and in the G2 phase was 21% compared with 25%, and the percentage of the S phase was 11% compared with 10% (figures 5B and 5D). These differences did not reach significance.

The effect of 17β-HSD1 expression on the cell cycle was evaluated by flow cytometry. The percentage of MCF7-17β-HSD1 cells in the G1/G0 phase was 75% compared with 81% of

MCF7 cells, and in the G2 phase was 17% compared with 12% of MCF7 cells, and the percentage of the S phase was 7% compared with 6% of MCF7 cells (Figures 5E and 5F).

When we examined the effect of 17β-HSD1 on T47D cell apoptosis, there was no significant variation of the Annexin V expression in T47D transfected with 17β-HSD1 specific siRNA or treated with 17β-HSD1 specific inhibitor compared to the control cells (Figures 5G and H).

The apoptosis test was performed by flow cytometry. Results showed that the percentage of living cells is significantly higher in MCF7-17β-HSD1 (89%) than MCF7 cells (84%). While the percentage of apoptotic cells and necrosis cells is less in MCF7-17β-HSD1 (4% and 0.6% respectively) than MCF7 cells (7% and 1.2% respectively) (Figure 5J).

3.6. The knockdown of 17β-HSD1 modulates the protein profile of breast cancer cell line T47D

Eight two-dimensional electrophoresis gels were used to perform proteomic analysis. Gels were made from four independent biological repetitions of protein samples from T47D control and T47D transfected with 17β-HSD1 siRNA. Figure 6A shows that the two cell lysates had similar spot patterns. We identified 10 significant differential protein spots between T47D control and T47D transfected with 17β-HSD1. 8 spots were downregulated and 2 spots were upregulated after siRNA transfection (Figure 6C). The 10 spots were identified by MS, giving a total of 90 proteins. Some proteins were present in more than one spot. Thus, the total number of distinct proteins was 86 in which 56 and 30 proteins from spots downregulated and upregulated after 17β-HSD1 siRNA transfection respectively (Figure 6C). UniProt database [www.uniprot.org] was used to determine the function of each of the proteins identified by MS analysis. The knockdown of 17β-HSD1 in T47D cell decreased the expression of the metastasis inhibition factor nm23 (nm23-H1), a regulator of breast cancer metastasis (additional file 1 Table S1).

The expression of nm23-H1 in T47D was verified by Western blotting and the decrease in nm23-H1 expression after transfection with 17β-HSD1 specific siRNA was confirmed at the protein and mRNA levels (Figures 8A and 8B). We also found a decrease in 17β-HSD1 expression after nm23-H1 transfection with its specific siRNA (Figure 8A).

3.7. The inhibition of 17β-HSD1 modulates the protein profile of breast cancer cell line T47D

Figure 6B shows that the two cell lysates had similar spot patterns. We identified 11 significant differential protein spots between T47D control and T47D treated with the inhibitor. Four spots were downregulated and 7 spots were upregulated after inhibitor treatment (Figure 6D). The 11 spots were identified by MS, giving a total of 107 proteins. Some proteins were present in more than one spot. Thus, the total number of distinct proteins is 103 in which 65 and 38 proteins from spots down regulated and up-regulated after 17β- HSD1 inhibition respectively (Figure 6D). UniProt database [www.uniprot.org] was used to determine the function of each protein identified by MS analysis. The inhibition of 17β- HSD1 in T47D cell decreased the expression of important proteins implicated in various biological functions such as the tumor protein D54; regulating cell proliferation; and alpha- actinin-4 which is a regulator of apoptotic process (additional file 2 Table S2). While, the expression of proteins involved in cell cycle regulation like septin 9, 14-3-3 protein epsilon and tumor protein D53 were increased after 17β-HSD1 inhibition (additional file 2 Table S2).

The expression of TPD54 in T47D was verified by Western blotting and the decrease in TPD54 expression after 17β-HSD1 inhibition with the specific inhibitor was confirmed (Figure 8A). The mRNA level of several genes was verified by RT-qPCR and compared to protein level obtained in the 2D gel analysis. After 17β-HSD1 knockdown in T47D, we found that Prelamin-A/C (LMNA), TPD54 and GSTMu3 were downregulated at the mRNA and protein level. TPD53 and YWHAE were upregulated at both transcript and protein levels (Figure 8B). These results show a correlation between protein and mRNA level (Figure 8C).

However, septin 9 showed a differential regulation between the protein and transcriptional levels. Septin 9 was upregulated at protein level and downregulated at transcription level (Figures 8B and 8C).

3.8. The knock-down of 17β-HSD1 modulates the protein profile of MCF7 breast cancer cells

The effect of 17β-HSD1 knock down on MCF7 protein profile was analyzed by performing two-dimensional gel. Total protein lysates of MCF7 control and MCF7 transfected with 17β- HSD1 specific siRNA was used. The analysis identified 12 significant differential protein spots between MCF7 control and MCF7 transfected with 17β-HSD1 (Figures 7A and B). Nine spots are downregulated and 3 spots are upregulated after siRNA transfection (Figure 7C). The 12 spots were identified by MS, giving a total of 136 proteins in which 91 and 45 proteins from spots down regulated and up-regulated after 17β-HSD1 siRNA transfection respectively (Figure 7C). UniProt database [www.uniprot.org] was used to determine the function of each of protein identified by MS analysis. The 17β-HSD1 inhibition in MCF7 cells decrease the expression the metastasis inhibition factor nm23 (nm23-H1), a regulator of breast cancer metastasis, and the tumor protein D54, a regulator of cell proliferation (additional file 3 table S3). While, the expression of proteins involved in cell cycle regulation like septin 8, was increased after the 17β-HSD1 inhibition in MCF7 cells (additional file 3 table S3).

The expression of TPD54 in MCF7 was verified by Western blotting and the decrease in TPD54 expression after 17β-HSD1 knockdown was confirmed (Figure 9A). The mRNA level of nm23-H1 was verified by qRT-PCR and compared to protein level obtained by the 2D gel analysis. The MCF7-17β-HSD1 showed a higher mRNA level of nm23-H1 than MCF7 cells (Figure 9B). The knockdown of 17β-HSD1 decreased the mRNA level of nm23- H1 in MCF7 and MCF7-17β-HSD1 cells (Figures 9C and 9D). These results proved that there is a positive correlation between 17β-HSD1 and nm23-H1 expression.

3.9. The role of nm23-H1 expression in T47D cell proliferation, invasion, estradiol formation and its correlation with 17β-HSD1 expression

To determine the impact of nm23-H1 on breast cancer progression, T47D cells were transfected with nm23-H1 specific siRNA. A mixture of three siRNA against nm23-H1 and negative control siRNA were used to determine whether siRNA can knock down the expression of nm23-H1 in T47D cells. Proteins were extracted from cells 96 h after transfection and analyzed by Western blot using an nm23-H1 antibody. The specificity of nm23-H1 siRNA was confirmed (Figure 10A). A proliferation test was performed by Cyquant to evaluate the effect of the knock down of nm23-H1 on T47D cell proliferation. Results show that the growth of T47D cells was reduced by 23 % after 96 h of transfection (Figure 10B). These observations show that the knockdown of nm23-H1 expression in T47D, led to a decrease of cell proliferation.

We also tested the effect of nm23-H1 down regulation on T47D cell invasion. Results show that cell invasion of T47D cells transfected with nm23-H1 specific siRNA was reduced by 33 % (Figure 10C).

T47D cells were transfected with nm23-H1 specific siRNA for 4 days. After nm23-H1 knock down, the level of E2 was reduced by 43 % in comparison with control T47D cells (Figure 10D).

3.10. Common proteins to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition

After 17β-HSD1 inhibition in T47D cells with an inhibitor or specific siRNA, four common proteins to both treatments were down regulated, namely alpha actinin 4 (ACTN4), involved in protein transport; cathepsin D, involved in the pathogenesis of several diseases such as breast cancer; heat shock protein beta-1 (HSPB1), involved in stress resistance and actin

59 organization; and heterogeneous nuclear ribonucleoprotein C the-like1 (HNRCL) involved in nucleosome assembly (Table 1).

3.11. Network, canonical pathways, biological functions and upstream regulators identified by IPA analysis

In T47D cells, the knockdown and inhibition of 17β-HSD1 generated modification of 89 and 99 proteins respectively. Proteins identified by the two dimensional gel analysis were used to identify common networks to both treatments, 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition, in T47D cells by using software Ingenuity Pathway analysis (IPA, Ingenuity® Systems; http://www.ingenuity.com). Two common networks were identified. The first one is involved in infectious disease, cancer, cell death and cell survival (Figure 11A). The second network is involved in humoral immune response, protein synthesis, cell death and cell survival (Figure 11B).

In MCF7 cells, the knockdown of 17β-HSD1 generated modification of 132 proteins which have been associated to 8 networks. Network that shows the highest score and contains 32 proteins of the list and 3 partner proteins added by IPA is shown in Figure 11C. The first function associated is cell death and survival. Second network contains 17 proteins of the list and 12 partner proteins added by IPA. The three main functions associated are posttranslational modification, protein folding and cellular development (Figure 11D).

The software Ingenuity Pathway analysis (IPA) was also used to identify common canonical pathways to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D cells using proteins identified by the two dimensional gel analyses. Sixteen common canonical pathways were identified (Figure 12A). Figure 12B shows the percentage of genes involved in various pathways, the ratio gives an indication on which pathway has been affected, mainly based on the percentage of genes used for IPA analysis. The p-value was used to determine an association between a specific pathway and genes used for IPA analysis.

In MCF7 cells, proteins identified by the two dimensional gel analyses were used to identify canonical pathways. The list of 132 proteins may be associated to 275 pathways. In Figure 12C, the histogram shows the top 20 most significant pathways. The most significant pathway is 14-3-3-mediated signaling pathway (Figure 12D).

Fifty five common biological functions to the two treatments, 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D cells, are obtained through the IPA analysis. The most significant functions are cell death, apoptosis and necrosis (Figure 13A). The proteins identified following the knockdown of 17β-HSD1 in MCF7 cells are associated with four categories of biological functions, significantly represented, such as apoptosis, necrosis, cell death and cell survival (Figure 13B).

To determine the common upstream regulators to the two treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition by using proteins identified by the two dimensional gel analysis, the software Ingenuity Pathway analysis (IPA) was also used. Two common upstream regulators were identified TP53 and NFE2L2. The predictions in TP53 were highly consistent (Tables 2 and 3). Figure 14A shows the effect of the upstream regulator TP53 on the proteins identified by two dimensional gel analysis after 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition. TP53 activated proteins upregulated following 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition and inactivated proteins downregulated following 17β- HSD1 siRNA transfection and 17β-HSD1 inhibition.

Upstream regulator analysis of proteins identified after 17β-HSD1 knockdown in MCF7 cells is shown in Table 4. CD28 is one of the upstream regulators that have a very significant z- score (Figure 14B). CD28 is a glycoprotein localized in the cell surface. According to the IPA prediction, CD28 would have an activated state, taking account of the inhibition of 10 target protein and the overexpression of the two others.

4. Discussion

Estradiol (E2) is involved in the development and the progression of hormone-dependent breast cancers (9, 31, 32, 47, 48). Clinical and animal studies, demonstrated that cumulative exposure of the breast epithelium to this hormone is one of the risk factors associated with breast cancer (49). Studies have shown that in breast carcinomas, the tissue concentrations of E2 were more than 10 fold higher than in plasma (50, 51). Human breast cancer tissues contain all the enzymes involved in the final steps of E2 biosynthesis, including enzymes in the rate-limiting steps such as aromatase, steroid sulfatase, and 17β-hydroxysteroid dehydrogenase type 1 (13, 52). 17β-HSD1 has an estrogen activating roles (activation of E1 into the potent estrogen, E2) and androgen inactivating (inactivation of the most active androgen, DHT) in breast cancer cells (28, 53-55).

In the present study, we investigate the effect of 17β-HSD1 inhibition on T47D, MCF7 and MCF7-17β-HSD1 cell proliferation, invasion, cell cycle regulation, apoptosis and proteome profile. For the inhibitory capacity of 17β-HSD1 inhibitor, there was a correlation between 17β-HSD1 expression and E1 conversion to E2 (Figure 1F). Our results show that inhibition and knockdown of 17β-HSD1 decreased T47D and MCF7 cell proliferation and E2 concentration significantly (Figures 2A, B, E and F; Figures 3A and B). Also, 17β-HSD1 overexpression in MCF7 cell permits an increase in cell growth and E2 formation (Fig. 2G; Figure 3C). It has been shown that estrogen is the major factor in the proliferation of breast cancer cells (56). Also, Suzuki et al demonstrated that 17β-HSD1 is responsible for regulating the process leading to the accumulation of E2 in human breast cancer tissues (57). Another study showed that inhibition of 17β-HSD1 leads to a decrease in T47D cell proliferation (28, 38), whereas 17β-HSD1 overexpression leads to the stimulation of MCF7 growth (28). A high [E2]/[E1] ratio is positively correlated with 17β-HSD1 expression (58) and to the proliferation of BC cells (28). Our results confirm the positive correlation between 17β-HSD1 expression, the level of E2 and the proliferation of T47D and MCF7 breast cancer cells.

We also showed that inhibition and knockdown of 17β-HSD1 decreased the cell invasion and colonies formation of the T47D and MCF7-17β-HSD1 cells (Figures 4A, B, C, and D). In MCF7-17β-HSD1, cell invasion was 31% higher than MCF7 cells (Figure 4E). We have already demonstrated that 17β-HSD1 expression is positively correlated with the migration of MCF7 breast cancer cells (39). Estrogens are promoters of cell movement in several tissues, including the breast (59), with the finding that ER+ breast cancers are driven to invade and metastasize by endogenous or exogenous estrogens (60, 61). It has also been shown that in response to estrogen treatment, MCF-10F forms colonies in agar and shows a higher invasive capacity in Matrigel than MCF-10F in the absence of estrogens (62, 63). Our results suggest that 17β-HSD1 inhibition decreases T47D cells invasive capacity by decreasing the E2 level and that there is a positive correlation between 17β-HSD1 expression and MCF7 cell invasion.

Following knockdown of 17β-HSD1 with a specific siRNA in T47D cells, cell cycle analyses showed a significant increase of cells in G0/G1 phase and a significant decrease of cells in S phase and in G2 / M phase (Figure 5C). Our results also show that inhibition of 17β-HSD1 has no significant influence on T47D cells apoptosis. In MCF-7 cells, the overexpression of 17β-HSD1 leads to a significant decrease of cells in G1/G0 phase and a significant increase of cells in S phase and in G2 / M phase (Figure 5F). Also, in MCF7-17β- HSD1, the percentage of living cells is higher, while the percentage of apoptotic cells and necrosis cells is less than MCF7 cells (Figure 5J). It has been shown that treatment of cells with E2, significantly decreased the percentage of cells in G0/G1 phase of the cell cycle and stimulated S phase entry (64). Antiestrogen (tamoxifen) treatment arrests cell cycle progression in the G0/G1 phase (65). Our results show that knockdown of 17β-HSD1 expression, like for tamoxifen, leads to the inhibition of G1/S progression and cell cycle arrest. This is probably due to the lower level of E2.

Proteomic studies of T74D and MCF7 cells were analyzed by performing two- dimensional gel. Several proteins involved in cell proliferation, apoptotic process, cell invasion and cell cycle are modulated following 17β-HSD1 knockdown and inhibition in T47D cells. The expression of metastasis inhibition factor nm23 (nm23-H1), a regulator of

63 breast cancer metastasis (additional file 1 Table S1), tumor protein D54 (TPD54), Glutathione S-transferase Mu 3 (GSTMu3) and Prelamin-A/C (LMNA) (additional file 2 Table S2) was decreased. TPD54 is tumor protein D52 family member (66). TPD52 is the first gene of the tumor protein D52 family which was identified as being overexpressed in human breast cancer (67, 68). The expression of zinc finger protein (ZPR1) (Additional file 1 Table S1), 14-3-3 protein epsilon (YWHAE) and tumor protein D53 (TPD53) (Additional file 2 Table S2) was increased. In breast cancer cells, TPD53, member of the D52-like family, is identified as the binding partner of 14-3-3 (69). TPD53 is a regulator of cell cycle progression (70). TPD53 is a protein that has a strong expression at the G2/M phase and low expression when cells cycles back into G0/G1 and its expression is correlated with cyclin B1 expression (70).

In MCF7 cells, the knock down of 17β-HSD1 increase the expression of proteins involved in cell cycle regulation like septin 8 (Additional file 3 table S3). While, the expression of the tumor protein D54, a regulator of cell proliferation, and metastasis inhibition factor nm23 (nm23-H1), a regulator of breast cancer metastasis was increased (Additional file 3 table S3). We have already demonstrated that 17β-HSD1 expression positively regulates the nm23-H1 gene and increases the migration of MCF7 breast cancer cells (39). Our results show again that there is a positive correlation between 17β-HSD1 and nm23-H1 expression in T47D and MCF7 breast cancer cells. We also noted a decrease in 17β-HSD1 expression after nm23-H1 transfection with specific siRNA in T47D cells (Figure 10A), accompanied by a reduction of cell proliferation (23%) (Figure 10B), cell invasion (33 %) (Figure 10C) and E2 formation (43 %) (Figure 10D) in comparison with control T47D cells. Thus showing that the knockdown of nm23-H1 lead to a decrease of 17β-HSD1 expression, which causes a decrease in E1 conversion to E2. Accordingly, there is a decrease in cell growth and invasiveness of T47D cells. These results support the conclusion of positive correlation between nm23-H1 and 17β-HSD1 expression.

In this study, we used the software Ingenuity Pathway analysis to identify the common proteins and upstream regulators to the two treatments, 17β-HSD1 inhibitor and 17β-HSD1 siRNA in T47D cells. Four common proteins to the two treatments were down regulated

especially alpha actinin 4 (ACTN4), cathepsin D (CTSD), heat shock protein beta-1 (HSPB1) and heterogeneous nuclear ribonucleoprotein C the-like1 (HNRCL). Moreover, two common upstream regulators to both treatments were identified TP53 and NFE2L2. The predictions in TP53 are highly consistent (Tables 2 and 3). The TP53 activates proteins upregulated following 17β-HSD1 siRNA transfection and inhibition and inactivates proteins downregulated following 17β-HSD1 siRNA transfection and inhibition.

5. Conclusions

This study shows that the inhibition of 17β-HSD1 by a specific inhibitor or siRNA affect several signaling pathways in T47D and MCF7 cells by regulating various proteins involved in several physiological functions such as tumor protein D54 (TPD54) (involved in cell proliferation), 14-3-3 protein epsilon (YWHAE) and tumor protein D53 (TPD53) (involved in cell cycle regulation) and metastasis inhibition factor nm23 (nm23-H1) (regulator of breast cancer metastasis). We also proved the positive correlation between 17β-HSD1 and nm23-H1 expression in T47D and MCF7 cells. For the first time, we showed that 17β-HSD1 regulates the cell cycle and the cell invasion of T47D and MCF7 breast cancer cells through generation of E2 level.

Competing interests The authors declare that they have no competing interests.

Authors' contributions SXL and MZ designed the study. MZ carried out the experimental studies. MZ and SXL prepared the manuscript. All authors read and approved the final manuscript.

Authors’ information Laboratory of Molecular Endocrinology and Oncology, Centre Hospitalier Universitaire de Québec Research Center (CHUQ - CHUL) and Department of Molecular Medicine, Laval University, 2705 boulevard Laurier, Québec G1V 4G2, Canada.

Acknowledgements We thank Mrs Gina Racine for her assistance in 2-D gel image analyses; we acknowledge Dr. E-L. Calvo for his assistance in 2-D gel analyses. We thank Mrs Nathalie Paquet for her help in quantitative real-time RT-PCR analysis. We thank Dr Alexandre Brunet for his assistance in flow cytometry analyses. We thank Dr Donald Poirier for giving us the 17β- HSD type 1 inhibitor. This work was supported by the Canadian Institutes of Health Research, with grant MOP 97917 and 89851 to S.-X. Lin (PI).

References

1. Jemal, A., Bray, F., Center, M. M., Ferlay, J., Ward, E., and Forman, D. (2011) Global cancer statistics. CA: a cancer journal for clinicians 61, 69-90 2. Siegel, R. L., Miller, K. D., and Jemal, A. (2015) Cancer statistics, 2015. CA: a cancer journal for clinicians 65, 5-29 3. Jemal, A., Siegel, R., Ward, E., Hao, Y., Xu, J., Murray, T., and Thun, M. J. (2008) Cancer statistics, 2008. CA: a cancer journal for clinicians 58, 71-96 4. Society, C. C. (2013) Cancer facts from Canadian Cancer Statistics 2013: https://www.cancer.ca/~/media/cancer.ca/QC/about%20us/media/Facts%20sheets/13-05-29- Cancer-Stats-Facts-2013-VFIN-angl.pdf. Consulté le 24/03/2015 5. Brody, L. C., and Biesecker, B. B. (1998) Breast cancer susceptibility genes. BRCA1 and BRCA2. Medicine 77, 208-226 6. Russo, J., Hasan Lareef, M., Balogh, G., Guo, S., and Russo, I. H. (2003) Estrogen and its metabolites are carcinogenic agents in human breast epithelial cells. The Journal of steroid biochemistry and molecular biology 87, 1-25 7. Fernandez, S. V., and Russo, J. (2010) Estrogen and xenoestrogens in breast cancer. Toxicologic pathology 38, 110-122 8. Russo, I. H., and Russo, J. (1998) Role of hormones in mammary cancer initiation and progression. Journal of mammary gland biology and neoplasia 3, 49-61 9. Nilsson, S., Makela, S., Treuter, E., Tujague, M., Thomsen, J., Andersson, G., Enmark, E., Pettersson, K., Warner, M., and Gustafsson, J. A. (2001) Mechanisms of estrogen action. Physiological reviews 81, 1535-1565 10. Nelson, H. D., Humphrey, L. L., Nygren, P., Teutsch, S. M., and Allan, J. D. (2002) Postmenopausal hormone replacement therapy: scientific review. Jama 288, 872-881 11. Rossouw, J. E., Anderson, G. L., Prentice, R. L., LaCroix, A. Z., Kooperberg, C., Stefanick, M. L., Jackson, R. D., Beresford, S. A., Howard, B. V., Johnson, K. C., Kotchen, J. M., Ockene, J., and Writing Group for the Women's Health Initiative, I. (2002) Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial. Jama 288, 321-333 12. McPherson, K., Steel, C. M., and Dixon, J. M. (2000) ABC of breast diseases. Breast cancer-epidemiology, risk factors, and genetics. Bmj 321, 624-628 13. Chetrite, G. S., Cortes-Prieto, J., Philippe, J. C., Wright, F., and Pasqualini, J. R. (2000) Comparison of estrogen concentrations, estrone sulfatase and aromatase activities in normal, and in cancerous, human breast tissues. The Journal of steroid biochemistry and molecular biology 72, 23-27 14. Miyoshi, Y., Ando, A., Shiba, E., Taguchi, T., Tamaki, Y., and Noguchi, S. (2001) Involvement of up-regulation of 17beta-hydroxysteroid dehydrogenase type 1 in maintenance of intratumoral high estradiol levels in postmenopausal breast cancers. International journal of cancer. Journal international du cancer 94, 685-689 15. Thorsen, T., Tangen, M., and Stoa, K. F. (1982) Concentration of endogenous oestradiol as related to oestradiol receptor sites in breast tumor cytosol. Eur J Cancer Clin Oncol 18, 333-337 16. van Landeghem, A. A., Poortman, J., Nabuurs, M., and Thijssen, J. H. (1985) Endogenous concentration and subcellular distribution of estrogens in normal and malignant human breast tissue. Cancer research 45, 2900-2906 67

17. Miller, W. R., and O'Neill, J. (1987) The importance of local synthesis of estrogen within the breast. Steroids 50, 537-548 18. O'Neill, J. S., Elton, R. A., and Miller, W. R. (1988) Aromatase activity in adipose tissue from breast quadrants: a link with tumour site. Br Med J (Clin Res Ed) 296, 741-743 19. Reed, M. J., Aherne, G. W., Ghilchik, M. W., Patel, S., and Chakraborty, J. (1991) Concentrations of oestrone and 4-hydroxyandrostenedione in malignant and normal breast tissues. International journal of cancer. Journal international du cancer 49, 562-565 20. Suzuki, T., Miki, Y., Nakamura, Y., Moriya, T., Ito, K., Ohuchi, N., and Sasano, H. (2005) Sex steroid-producing enzymes in human breast cancer. Endocrine-related cancer 12, 701-720 21. Aka, J. A., Mazumdar, M., and Lin, S. X. (2009) Reductive 17beta-hydroxysteroid dehydrogenases in the sulfatase pathway: critical in the cell proliferation of breast cancer. Molecular and cellular endocrinology 301, 183-190 22. Lin, S. X., Chen, J., Mazumdar, M., Poirier, D., Wang, C., Azzi, A., and Zhou, M. (2010) Molecular therapy of breast cancer: progress and future directions. Nat Rev Endocrinol 6, 485-493 23. Peltoketo, H., Luu-The, V., Simard, J., and Adamski, J. (1999) 17beta-hydroxysteroid dehydrogenase (HSD)/17-ketosteroid reductase (KSR) family; nomenclature and main characteristics of the 17HSD/KSR enzymes. Journal of molecular endocrinology 23, 1-11 24. Nguyen, B. L., Chetrite, G., and Pasqualini, J. R. (1995) Transformation of estrone and estradiol in hormone-dependent and hormone-independent human breast cancer cells. Effects of the antiestrogen ICI 164,384, danazol, and promegestone (R-5020). Breast cancer research and treatment 34, 139-146 25. Peltoketo, H., Isomaa, V., Maentausta, O., and Vihko, R. (1988) Complete amino acid sequence of human placental 17 beta-hydroxysteroid dehydrogenase deduced from cDNA. FEBS letters 239, 73-77 26. Luu The, V., Labrie, C., Zhao, H. F., Couet, J., Lachance, Y., Simard, J., Leblanc, G., Cote, J., Berube, D., Gagne, R., and et al. (1989) Characterization of cDNAs for human estradiol 17 beta-dehydrogenase and assignment of the gene to chromosome 17: evidence of two mRNA species with distinct 5'-termini in human placenta. Molecular endocrinology 3, 1301-1309 27. Lin, S. X., Yang, F., Jin, J. Z., Breton, R., Zhu, D. W., Luu-The, V., and Labrie, F. (1992) Subunit identity of the dimeric 17 beta-hydroxysteroid dehydrogenase from human placenta. The Journal of biological chemistry 267, 16182-16187 28. Aka, J. A., Mazumdar, M., Chen, C. Q., Poirier, D., and Lin, S. X. (2010) 17beta- hydroxysteroid dehydrogenase type 1 stimulates breast cancer by dihydrotestosterone inactivation in addition to estradiol production. Molecular endocrinology 24, 832-845 29. Martel, C., Rheaume, E., Takahashi, M., Trudel, C., Couet, J., Luu-The, V., Simard, J., and Labrie, F. (1992) Distribution of 17 beta-hydroxysteroid dehydrogenase gene expression and activity in rat and human tissues. The Journal of steroid biochemistry and molecular biology 41, 597-603 30. Luu-The, V. (2001) Analysis and characteristics of multiple types of human 17beta- hydroxysteroid dehydrogenase. The Journal of steroid biochemistry and molecular biology 76, 143-151 31. Thomas, D. B. (1984) Do hormones cause breast cancer? Cancer 53, 595-604

32. Vihko, R., and Apter, D. (1989) Endogenous steroids in the pathophysiology of breast cancer. Critical reviews in oncology/hematology 9, 1-16 33. Pasqualini, J. R. (2004) The selective estrogen enzyme modulators in breast cancer: a review. Biochimica et biophysica acta 1654, 123-143 34. Gunnarsson, C., Olsson, B. M., Stal, O., and Southeast Sweden Breast Cancer, G. (2001) Abnormal expression of 17beta-hydroxysteroid dehydrogenases in breast cancer predicts late recurrence. Cancer research 61, 8448-8451 35. Oduwole, O. O., Li, Y., Isomaa, V. V., Mantyniemi, A., Pulkka, A. E., Soini, Y., and Vihko, P. T. (2004) 17beta-hydroxysteroid dehydrogenase type 1 is an independent prognostic marker in breast cancer. Cancer research 64, 7604-7609 36. Vermeulen, A., Deslypere, J. P., and Paridaens, R. (1986) Steroid dynamics in the normal and carcinomatous mammary gland. Journal of steroid biochemistry 25, 799-802 37. 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. Molecular and cellular endocrinology 301, 146-153 38. Laplante, Y., Cadot, C., Fournier, M. A., and Poirier, D. (2008) Estradiol and estrone C-16 derivatives as inhibitors of type 1 17beta-hydroxysteroid dehydrogenase: blocking of ER+ breast cancer cell proliferation induced by estrone. Bioorganic & medicinal chemistry 16, 1849-1860 39. Aka, J. A., Zerradi, M., Houle, F., Huot, J., and Lin, S. X. (2012) 17beta- hydroxysteroid dehydrogenase type 1 modulates breast cancer protein profile and impacts cell migration. Breast cancer research : BCR 14, R92 40. Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. (1996) Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Analytical chemistry 68, 850-858 41. Havlis, J., Thomas, H., Sebela, M., and Shevchenko, A. (2003) Fast-response proteomics by accelerated in-gel digestion of proteins. Analytical chemistry 75, 1300-1306 42. Keller, A., Nesvizhskii, A. I., Kolker, E., and Aebersold, R. (2002) Empirical statistical model to estimate the accuracy of peptide identifications made by MS/MS and database search. Analytical chemistry 74, 5383-5392 43. Nesvizhskii, A. I., Keller, A., Kolker, E., and Aebersold, R. (2003) A statistical model for identifying proteins by tandem mass spectrometry. Analytical chemistry 75, 4646-4658 44. Systems, I. (2011) www.ingenuity.com. August 2011 version 45. Luu-The, V., Paquet, N., Calvo, E., and Cumps, J. (2005) Improved real-time RT- PCR method for high-throughput measurements using second derivative calculation and double correction. BioTechniques 38, 287-293 46. Warrington, J. A., Nair, A., Mahadevappa, M., and Tsyganskaya, M. (2000) Comparison of human adult and fetal expression and identification of 535 housekeeping/maintenance genes. Physiological genomics 2, 143-147 47. Deroo, B. J., and Korach, K. S. (2006) Estrogen receptors and human disease. The Journal of clinical investigation 116, 561-570 48. Frasor, J., Chang, E. C., Komm, B., Lin, C. Y., Vega, V. B., Liu, E. T., Miller, L. D., Smeds, J., Bergh, J., and Katzenellenbogen, B. S. (2006) Gene expression preferentially regulated by tamoxifen in breast cancer cells and correlations with clinical outcome. Cancer research 66, 7334-7340

49. Henderson, B. E., and Feigelson, H. S. (2000) Hormonal carcinogenesis. Carcinogenesis 21, 427-433 50. Miller, W. R., Hawkins, R. A., and Forrest, A. P. (1982) Significance of aromatase activity in human breast cancer. Cancer research 42, 3365s-3368s 51. Miller, W. R. (1991) Aromatase activity in breast tissue. The Journal of steroid biochemistry and molecular biology 39, 783-790 52. Evans, T. R., Rowlands, M. G., Law, M., and Coombes, R. C. (1994) Intratumoral oestrone sulphatase activity as a prognostic marker in human breast carcinoma. British journal of cancer 69, 555-561 53. Gangloff, A., Shi, R., Nahoum, V., and Lin, S. X. (2003) Pseudo-symmetry of C19 steroids, alternative binding orientations, and multispecificity in human estrogenic 17beta- hydroxysteroid dehydrogenase. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 17, 274-276 54. Nahoum, V., Gangloff, A., Shi, R., and Lin, S. X. (2003) How estrogen-specific proteins discriminate estrogens from androgens: a common steroid binding site architecture. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 17, 1334-1336 55. Lin, S. X., Poirier, D., and Adamski, J. (2013) A challenge for medicinal chemistry by the 17beta-hydroxysteroid dehydrogenase superfamily: an integrated biological function and inhibition study. Curr Top Med Chem 13, 1164-1171 56. Castoria, G., Migliaccio, A., Giovannelli, P., and Auricchio, F. (2010) Cell proliferation regulated by estradiol receptor: Therapeutic implications. Steroids 75, 524-527 57. Suzuki, T., Moriya, T., Ariga, N., Kaneko, C., Kanazawa, M., and Sasano, H. (2000) 17Beta-hydroxysteroid dehydrogenase type 1 and type 2 in human breast carcinoma: a correlation to clinicopathological parameters. British journal of cancer 82, 518-523 58. Zhang, C. Y., Chen, J., Yin, D. C., and Lin, S. X. (2012) The contribution of 17beta- hydroxysteroid dehydrogenase type 1 to the estradiol-estrone ratio in estrogen-sensitive breast cancer cells. PLoS One 7, e29835 59. Ricketts, D., Turnbull, L., Ryall, G., Bakhshi, R., Rawson, N. S., Gazet, J. C., Nolan, C., and Coombes, R. C. (1991) Estrogen and progesterone receptors in the normal female breast. Cancer research 51, 1817-1822 60. Yager, J. D., and Davidson, N. E. (2006) Estrogen carcinogenesis in breast cancer. The New England journal of medicine 354, 270-282 61. Platet, N., Cathiard, A. M., Gleizes, M., and Garcia, M. (2004) Estrogens and their receptors in breast cancer progression: a dual role in cancer proliferation and invasion. Critical reviews in oncology/hematology 51, 55-67 62. Russo, J., and Russo, I. H. (2006) The role of estrogen in the initiation of breast cancer. The Journal of steroid biochemistry and molecular biology 102, 89-96 63. Russo, J., Fernandez, S. V., Russo, P. A., Fernbaugh, R., Sheriff, F. S., Lareef, H. M., Garber, J., and Russo, I. H. (2006) 17-Beta-estradiol induces transformation and tumorigenesis in human breast epithelial cells. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 20, 1622-1634 64. Doisneau-Sixou, S. F., Cestac, P., Chouini, S., Carroll, J. S., Hamilton, A. D., Sebti, S. M., Poirot, M., Balaguer, P., Faye, J. C., Sutherland, R. L., and Favre, G. (2003) Contrasting effects of prenyltransferase inhibitors on estrogen-dependent cell cycle

progression and estrogen receptor-mediated transcriptional activity in MCF-7 cells. Endocrinology 144, 989-998 65. Doisneau-Sixou, S. F., Cestac, P., Faye, J. C., Favre, G., and Sutherland, R. L. (2003) Additive effects of tamoxifen and the farnesyl transferase inhibitor FTI-277 on inhibition of MCF-7 breast cancer cell-cycle progression. International journal of cancer. Journal international du cancer 106, 789-798 66. Nourse, C. R., Mattei, M. G., Gunning, P., and Byrne, J. A. (1998) Cloning of a third member of the D52 gene family indicates alternative coding sequence usage in D52-like transcripts. Biochimica et biophysica acta 1443, 155-168 67. Ross, D. T., and Perou, C. M. (2001) A comparison of gene expression signatures from breast tumors and breast tissue derived cell lines. Disease markers 17, 99-109 68. Balleine, R. L., Fejzo, M. S., Sathasivam, P., Basset, P., Clarke, C. L., and Byrne, J. A. (2000) The hD52 (TPD52) gene is a candidate target gene for events resulting in increased 8q21 copy number in human breast carcinoma. Genes, chromosomes & cancer 29, 48-57 69. Boutros, R., Bailey, A. M., Wilson, S. H., and Byrne, J. A. (2003) Alternative splicing as a mechanism for regulating 14-3-3 binding: interactions between hD53 (TPD52L1) and 14-3-3 proteins. Journal of molecular biology 332, 675-687 70. Boutros, R., and Byrne, J. A. (2005) D53 (TPD52L1) is a cell cycle-regulated protein maximally expressed at the G2-M transition in breast cancer cells. Experimental cell research 310, 152-165

FIGURE

A B

C D

Figure 1. Protein and mRNA level of 17β-HSD1 after knockdown by specific siRNA in MCF7-17β-HSD1 and T47D cells. A. Western blot was carried out with protein extracted from MCF7 and MCF7-17β-HSD1 cells (B) and by using 17β-HSD1 antibody. C. Relative 17β-HSD1 mRNA expression after control siRNA or 17β-HSD1 siRNA transfection in MCF7, MCF7-17β-HSD1 and T47D cells. mRNA level was quantified by RT-qPCR. Values are averaged from an experiment performed in duplicate. Values are average from an experiment done in duplicates. p value was determined by a Student’s t-test (** P<0.01). C. Activity assays were carried out in T47D cells using 14C-E1 as substrate. The assay was carried out for four incubation times (0, 10, 30 and 60 minutes) and represents the conversion of 14C-E1 into 14C-E2: c1, T47D control cells incubated without inhibitor; c2, T47D cells incubated with 3 µM 17β-HSD1 inhibitor; c3, T47D cells incubated with 10 µM 17β-HSD1 inhibitor; c4, The graph shows the quantification of activity assays.

A B

C D

E F

Figure 2. Inhibition and the knock down of 17β-HSD1 reduce T47D, MCF7 and MCF7- 17β-HSD1 cell proliferation. (A) 6000 T47D cells were transfected with 100 nM 17β-HSD1 specific siRNA or control siRNA in complete growth medium. (B) 6000 T47D cells were treated for 6 days with different concentrations of 17β-HSD1 inhibitor (1, 3 and 5 µM) in complete growth medium. (C) 6000 MCF7 cells or MCF7-17β-HSD1 (D) were transfected with 100 nM 17β-HSD1 specific siRNA or control siRNA in complete growth medium. (E) 6000 MCF7 cells or MCF7-17β-HSD1 (F) cells were treated for 6 days with 17β-HSD1 inhibitor (5 µM) in complete growth medium. (G) Cell proliferation was evaluated by Cyquant cell proliferation assay kit. Experiments were performed at least three times in triplicates. p value was determined by a Student’s t-test (* p<0.05; ** p<0.01).

A B

D E

Figure 3. Inhibition and the knock down of 17β-HSD1 reduce T47D, MCF7 cells and MCF7-17β-HSD1 cell estradiol level and colony formation. (A) T47D cells were

transfected with 100 nM 17β-HSD1 specific siRNA or control siRNA. (B) T47D cells were treated with 5 µM of 17β-HSD1 inhibitor or vehicle ethanol. (C) The E2 level in MCF7 cell supernatants was compared to that of MCF7-17βHSD1 cells and was measured by ELISA analysis. After 96 h of transfection or treatment, the levels of E2 in T47D, MCF7 cells and MCF7-17β-HSD1 cells supernatants were measured by ELISA analysis. (D) 300 T47D cells were transfected with 100 nM control or 17β-HSD1 siRNA. (E) 300 T47D cells were treated with17β-HSD1 inhibitor or vehicle ethanol. Cells were cultured in the presence of 10% FBS and incubated for 14 days. After which colonies were counted.

A B

C D

Figure 4. The Effect of 17β-HSD1 expression on cell invasion in T47D, MCF7 and MCF7-17β-HSD1 cells. (A) 150,000 cells transfected with control siRNA or 17β-HSD1 specific siRNA were plated in each insert. (B) 150,000 cells treated with vehicle ethanol or 17β-HSD1 specific inhibitor were plated in each insert. (C) 150,000 cells transfected with

control siRNA or 17β-HSD1 specific siRNA or treated with vehicle ethanol or 17β-HSD1 specific inhibitor (D) were plated in each insert. (E) MCF7 cell invasion was compared to that of MCF7-17β-HSD1. Plates were incubated for 48 hours at 37°C in a 5% CO2 atmosphere. Cells in the upper part of the chamber were removed with a cotton swab and cell invasion was assessed by using a Cell Invasion Assay. Finally, 100 μL from each sample were transferred to a 96-well microtiter plate and the absorbance was measured at 560 nm in a plate reader. Values are average ± SD from an experiment done in triplicates. The experiments were performed at least three times. p value was determined by a Student’s t-test (* P<0.05; ** P<0.01).

A B

C D

E F

G H

Figure 5. The effect of 17β-HSD1 knockdown and inhibition on cell cycle and apoptosis in T47D cells. (A) 500,000 cells were plated in 6 well plates and incubated for 24 hours. Cells were then transfected with control siRNA or 17β-HSD1 specific siRNA for another 4 days (C) or treated with vehicle ethanol or 17β-HSD1 specific inhibitor for another 4 days (B, D). (E), (F) 500,000 MCF7 and MCF7-17β-HSD1 cells were plated in 6 well plates and incubated for 96 hours. Cells were stained with propidium iodide (PI) and analysed by flow cytometry. (G) 500,000 cells were plated in 6 well plates and incubated for 24 hours. Cells were then transfected with control siRNA or 17β-HSD1 specific siRNA for another 4 days or treated with vehicle ethanol (CTL) or 17β-HSD1 specific inhibitor (1, 5 and 10 µM) for another 4 days (H). (J) 500,000 MCF7 and MCF7-17β-HSD1 cells were plated in 6 well plates and incubated for 96 hours. The differentiation between apoptotic cells and viable or necrotic cells were determined by adding annexin V-allophycocyanin (APC) and propidium iodide (PI) to the cells. Cells were then stained with Annexin V-APC and PI for 30 min. Samples were then analyzed by flow cytometry. Values are average ± SD from an experiment done in duplicates. Experiments were performed at least three times. p value was determined by a Student’s t-test (* p<0.05; ** p<0.01).

A B

C Total variant Spots Number of spot picked identified for MS Protein T47D 8 down 8 56 control regulated siRNA after 17β- HSD1 inhibition T47D 2 up 2 30 17β- regulated HSD1 after 17β- siRNA HSD1 inhibition)

D Total Spots Number of variant picked identified spot for MS Protein T47D 4 down 4 65 control regulated after 17β- HSD1 inhibition T47D-17β- 7 up 7 38 HSD1 regulated inhibitor after 17β- HSD1 inhibition

Figure 6. Knockdown and the inhibition of 17β-HSD1 modulate the protein profile of T47D breast cancer cell lines. (A) Two-dimensional gel was performed. Total protein lysates of T47D control and T47D transfected with 17β-HSD1 specific siRNA or T47D control and T47D treated with 17β-HSD1 specific inhibitor (B) was used. Progenesis software and a t-test (with a p-value < 0.05) were used for the proteomic analyses. (C) The analysis identified 10 significant differential protein spots between T47D control and T47D transfected with 17β-HSD1. 8 spots are downregulated and 2 spots are upregulated after siRNA transfection. The 10 spots were identified by MS, giving a total of 86 proteins in

which 56 and 30 proteins from spots down regulated and up-regulated after 17β-HSD1 siRNA transfection respectively. (D) The analysis of T47D control and T47D treated with 17β-HSD1 inhibitor gives 11 significant differential protein spots. 4 spots are downregulated and 7 spots are upregulated after inhibition treatment and the total number of distinct proteins is 103 proteins in which 65 and 38 proteins from spots down regulated and up-regulated after 17β-HSD1 inhibition respectively. Progenesis software and a t-test (with a p-value < 0.05) were used for the proteomic analyses.

A B

Total Spots Number of variant spot picked for identified MS Protein MCF7 control 9 down 9 91 regulated after 17β- HSD1 knockdown MCF7-17βHSD1 3 up 3 45 siRNA regulated after 17β- HSD1 knockdown

Figure 7. The knock down of 17β-HSD1 modulates the MCF7 protein profile. Two- dimensional gel was performed. Total protein lysates of MCF7 control and MCF7 transfected with 17β-HSD1 siRNA was used. Progenesis software and a t-test (with a p-value < 0.05) were used for the proteomic analyses. The analyses identified 12 significant differential protein spots between MCF7 control (A) and MCF7 transfected with 17β-HSD1 (B). C. 9 spots are downregulated and 3 spots are upregulated after siRNA transfection. The 12 spots were identified by MS, giving a total of 136 proteins in which 91 and 45 proteins from spots down regulated and up-regulated after 17β-HSD1 siRNA transfection respectively.

Table 1. Common proteins to both treatments: 17β-HSD1 siRNA transfection and 17β- HSD1 inhibition

Proteins T47D transfected with T47D treated with 17β- 17β-HSD1 siRNA HSD1 inhibitor FC FC Alpha actinin 4 (ACTN4) 1.5 1.8

Cathepsin D (CTSD) 2 1.5

Heat shock protein beta-1 2 1.5 (HSPB1) Heterogeneous nuclear 1.5 1.6 ribonucleoprotein C the- like1 (HNRCL)

FC: Fold change

C Correlation gel 2D et RT- Description qPCR Metastasis inhibition factor nm23 (nm23- H1) Yes Prelamin-A/C (LMNA) Yes Tumor protein D54 (TPD54) Yes Glutathione S-transferase Mu 3 Yes Septin-9 (Sept9) No 14-3-3 protein epsilon (YWHAE) Yes TPD53 Tumor protein D53 Yes

Figure 8. Western blot and RT-qPCR confirmation in T47D cells. (A) Western blot was carried out with protein extracted from T47D cells by using nm23-H1, 17β-HSD1 and TPD54 antibodies. (B) 500,000 T47D cells were transfected with 100 nM 17β-HSD1 specific siRNA or control siRNA. Then, the mRNA level of several genes was verified by RT-qPCR and compared to protein level obtained by the 2D gel analysis (C).

A B

C D

Figure 9. Western blot and RT-qPCR confirmation in MCF7 cells. A. Western blot was carried out with protein extracted from T47D cells and by using nm23-H1 and TPD54 antibodies. B. the mRNA level of nm23-H1 gene in MCF7 and MCF7- 17β-HSD1 cells was verified by RT-QPCR. C. 500,000 MCF7 and MCF7-17β-HSD1 cells (D) were transfected with 100 nM 17β-HSD1 specific siRNA or control siRNA. Then, the mRNA level of nm23- H1 gene was verified by RT-qPCR and compared to protein level obtained by the 2D gel analysis. Values are average ± SD from an experiment done in duplicates. p value was determined by Student’s t-test (* p<0.05).

B C

Figure 10. Role of nm23-H1 expression in T47D cell proliferation, invasion, estradiol formation and correlation with 17β-HSD1 expression. (A) Western blot was carried out with protein extracted from T47D cells and by using nm23-H1 antibody. (B) 6000 T47D cells were transfected with 100 nM nm23-H1 specific siRNA or control siRNA. Cell proliferation was evaluated by Cyquant cell proliferation assay kit. (C) 150,000 cells

transfected with control siRNA or nm23-H1 specific siRNA were plated in each insert. The plate was incubated for 48 h at 37°C in a 5% CO2 atmosphere. Cells in the upper part of the chamber were removed with a cotton swab and the cell invasion was assayed by using the CytoSelect™ Cell Invasion Assay. Finally, 100μL from each sample were transferred to a 96- well microtiter plate and the absorbance was measured at 560nm in a plate reader. (D) T47D cells were transfected with 100 nM nm23-H1 specific siRNA or control siRNA. After 96 h of transfection, the level of E2 in T47D cell supernatants was measured by ELISA. Values are average ± SD from an experiment done in triplicates. Experiments were performed at least three times in triplicates. p value was determined by a Student’s t-test (* P<0.05; ** P<0.01).

Figure 11. The common networks to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D cells. Two common networks were identified. A. The first one is involved in infectious disease, cancer, cell death and survival. B. The second network is involved in humoral immune response, protein synthesis, cell death and survival. The proteins identified following the 17β-HSD1 knockdown in MCF7 cells are associated with 8 networks. C. Network 1 shows the highest score and contains 32 proteins of the list and 3 partner proteins added by IPA. The first function associated is cell death and survival. D. Network 2 contains 17 proteins of the list and 12 partner proteins added by IPA. The three main functions associated are post-translational modification, protein folding and cellular development.

Figure 12. The common canonical pathways to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D cells. The software Ingenuity Pathway analysis (IPA) is used to identify common canonical pathways to both treatment: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition by using proteins identified by the two dimensional gel analysis. Sixteen common canonical pathways were identified (A). (B) The percentage of genes involved in various pathways, the ratio gives an indication on which pathway has been affected the most, based on the percentage of genes used for IPA analysis. The p-value is used to determine an association between a specific pathway and genes used for IPA analysis. IPA is also used to identify canonical pathways obtained after 17β-HSD1 knockdown in MCF7 cells. C. The histogram shows the top 20 most significant pathways. D. The most significant pathway is 14-3-3-mediated signaling pathway.

Figure 13. The common biological functions to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D cells. A. Fifty five common biological functions to the two treatments are obtained through the IPA analysis. The most significant functions are cell death, apoptosis and necrosis. B. The proteins identified following the knockdown of 17β-HSD1 in MCF7 cells are associated with four categories, significantly represented, such as apoptosis, necrosis, cell death and cell survival.

Table 2. Upstream regulator analysis of proteins identified after 17β-HSD1 inhibition

p-value Upstream Activation Molecule Type of Target molecules in dataset Regulator z-score overlap transcription ACTG1,APCS,CTSD,HSP90AA1,HSP NFE2L2 -1.000 0.0000 regulator 90AB1,LMNA,PDIA3,PDIA4,VCP ACTB,ACTN1,ACTN4,AHCY,ANXA transcription 6,CARHSP1,CNN3,FASN,GPD1L,GS TP53 1.051 0.0000 regulator N,HSP90AA1,HSP90AB1,HSPB1,MX 1,PSMD2,TRIM28,UBA1 ¸

Table 3. Upstream regulator analysis of proteins identified after 17β-HSD1 siRNA transfection

p-value Upstream Activation Molecule Type of Target molecules in dataset Regulator z-score overlap transcription CTSD,EIF2S1,HSP90AA1,HSP90B1,PD NFE2L2 0.000 0.0041 regulator IA3,VCP ACTB,ACTN4,CLU,FKBP4,HSP90AA1, transcription TP53 0.209 0.0001 HSPB1,HSPD1,P4HB,PRDX2,TRIM28, regulator UBA1

Table 4. Upstream regulator analysis of proteins identified after 17β-HSD1 siRNA transfection in MCF7 cells

100

Figure 14. TP53 as a common upstream regulator to both treatments: 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition in T47D. The common upstream regulator was determined by using the software IPA. The effect of the upstream regulator TP53 on the proteins identified by two dimensional gel analysis. The TP53 activates proteins upregulated after 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition and inactivates proteins

101 downregulated after 17β-HSD1 siRNA transfection and 17β-HSD1 inhibition. IPA is also used to identify upstream regulators of proteins identified after 17β-HSD1 siRNA transfection in MCF7 cells. B. CD28 is one of the upstream regulator that has a very significant z-score.

102

Additional file 1, Table S1. Mass spectrometry identification of proteins differentially expressed between T47D control cells and T47D cells transfected with siRNA against 17β-HSD1

Spot Fold Identified proteins Accession Gene name MW pI Peptide Function Number exp/pred number

DOWN REGULATED PROTEIN FOLLOWING siRNA TRANSFECTION

1691 2,1 ATP synthase subunit beta, P06576 ATP5B 45/57 5,1 11 ATP synthesis mitochondrial 60 kDa heat shock protein, P10809 HSPD1 45/61 5,1 8 chaperone-mediated protein mitochondrial complex assembly Tubulin beta chain P07437 TUBB 45/50 5,1 6 cell division Hsc70-interacting protein P50502 ST13 45/41 5,1 2 protein homooligomerization Proteasomal ubiquitin receptor Q16186 ADRM1 45/42 5,1 2 RNA elongation from RNA ADRM1 polymerase II promoter 4201 2 Serine/threonine-protein kinase Q13177 PAK2 24/58 5,59 4 cell structure disassembly during PAK 2 apoptosis Peroxiredoxin-2 P32119 PRDX2 24/22 5,59 3 Involved in redox regulation of the cell, anti apoptosis Electron transfer flavoprotein P38117 ETFB 24/28 5,59 2 respiratory electron transport subunit beta chain Cathepsin D P07339 CTSD 24/45 5,59 2 proteolysis, Involved in the pathogenesis of breast cancer Heat shock protein beta-1 P04792 HSPB1 24/23 5,59 2 cell death, stress resistance and actin organization 3360 1,8 Metastasis inhibition factor nm23 P15531 NME1 14/17 5,1 3 regulation of apoptotic process (nm23) and cell proliferation ATP synthase subunit alpha, P25705 ATP5A1 14/60 5,1 2 mitochondrial ATP synthesis mitochondrial coupled proton transport 2906 1,9 Sorcin P30626 SRI 22/22 5,09 5 calcium ion transport Ras-related protein Rab-18 Q9NP72 RAB18 22/23 5,09 7 Protein transport Hornerin Q86YZ3 HRNR 22/282 5,09 4 Keratinization Gamma-glutamylcyclotransferase O75223 GGCT 22/21 5,09 4 glutathione biosynthetic process Protein disulfide-isomerase A3 H7BZJ3 PDIA3 22/14 5,09 3 cell redox homeostasis (Fragment) F-actin-capping protein subunit P47756 CAPZB 22/31 5,09 2 actin cytoskeleton organization beta Aminoacylase-1 Q03154 ACY1 22/46 5,09 3 Proteolysis PDZ domain-containing protein O14908 GIPC1 22/36 5,09 2 glutamate secretion GIPC1 Ganglioside GM2 activator P17900 GM2A 22/21 5,09 2 sphingolipid metabolic process 2004 1,9 Isoform C1 of Heterogeneous P07910-2 HNRNPC 40/32 5,22 20 mRNA processing nuclear ribonucleoproteins C1/C2 Tubulin beta-4B chain P68371 TUBB4B 40/50 5,22 16 microtubule-based process, G2/M transition of mitotic cell cycle Protein disulfide-isomerase P07237 P4HB 40/57 5,22 19 cell redox homeostasis NSFL1 cofactor p47 Q9UNZ2 NSFL1C 40/41 5,22 14 centrosome cycle Nucleophosmin P06748 NPM1 40/33 5,22 8 Involved in cell proliferation and regulation of tumor suppressors p53/TP53 and ARF NHL repeat-containing protein 2 Q8NBF2 NHLRC2 40/79 5,22 6 Unknown function Glycogenin-1 P46976 GYG1 40/39 5,22 4 glycogen biosynthetic process Heat shock 70 kDa protein 1A/1B P08107 HSPA1A 40/70 5,22 5 cell proliferation, apoptotic process Hepatoma-derived growth factor P51858 HDGF 40/27 5,22 3 Transcription regulation, cell proliferation Methylosome protein 50 Q9BQA1 WDR77 40/37 5,22 4 ncRNA metabolic process Tubulin alpha-1B chain P68363 TUBA1B 40/50 5,22 2 microtubule cytoskeleton organization

103

Plexin-B2 O15031 PLXNB2 40/205 5,22 2 neural tube closure Inosine-5'-monophosphate H0Y4R1 IMPDH2 40/51 5,22 3 purine nucleotide biosynthetic dehydrogenase 2 (Fragment) process Guanine nucleotide-binding protein P63092 GNAS 40/46 5,22 3 activation of adenylate cyclase G(s) subunit alpha isoforms short activity Ezrin P15311 EZR 40/69 5,22 3 cytoskeletal anchoring at plasma membrane T-complex protein 1 subunit P48643 CCT5 40/60 5,22 2 'de novo' posttranslational protein epsilon folding Glutamyl-tRNA(Gln) Q9H0R6 QRSL1 40/57 5,22 2 mitochondrial translation amidotransferase subunit A, mitochondrial Ubiquitin-like modifier-activating P22314 UBA1 40/118 5,22 2 cell death enzyme 1 Protein phosphatase 1 regulatory Q15435 PPP1R7 40/42 5,22 2 Regulatory subunit of protein subunit 7 phosphatase 1 3408 1,6 Phosphoglycerate kinase 1 P00558 PGK1 19/45 5,23 9 Gluconeogenesis Hippocalcin-like protein 1 P37235 HPCAL1 19/22 5,23 2 calcium ion binding ADP-ribosylation factor 3 P61204 ARF3 19/21 5,23 2 small GTPase mediated signal transduction 2818 1,8 Ras-related protein Rab-1B Q9H0U4 RAB1B 24/22 5,4 3 small GTPase mediated signal transduction, protein transport 4206 1,5 Succinyl-CoA ligase [GDP- Q96I99 SUCLG2 40/47 5,15 8 tricarboxylic acid cycle forming] subunit beta, mitochondrial Endophilin-A2 Q99961 SH3GL1 40/41 5,15 6 signal transduction Farnesyl pyrophosphate synthase P14324 FDPS 40/48 5,15 5 farnesyl diphosphate biosynthetic process Eukaryotic translation initiation P05198 EIF2S1 40/36 5,15 5 Translation regulation, Protein factor 2 subunit 1 biosynthesis Acid ceramidase Q13510 ASAH1 40/45 5,15 2 cell death Suppressor of G2 allele of SKP1 Q9Y2Z0 SUGT1 40/41 5,15 4 Mitosis homolog Heterogeneous nuclear O60812 HNRNPCL1 40/32 5,15 2 nucleosome assembly ribonucleoprotein C-like 1 Peptidyl-prolyl cis-trans isomerase Q02790 FKBP4 40/52 5,15 3 steroid hormone receptor FKBP4 complex assembly, androgen receptor signaling pathway Biliverdin reductase A P53004 BLVRA 40/33 5,15 2 heme catabolic process Clusterin P10909 CLU 40/52 5,15 2 regulation of apoptotic process Na(+)/H(+) exchange regulatory O14745 SLC9A3R1 40/39 5,15 3 Wnt signaling pathway cofactor NHE-RF1 Alpha-actinin-4 O43707 ACTN4 40/105 5,15 2 platelet degranulation, response to hypoxia, regulation of apoptotic process

UP REGULATED PROTEIN FOLLOWING siRNA TRANSFECTION

1498 1,7 WD repeat-containing protein 55 Q9H6Y2 WDR55 51/42 4,78 17 rRNA processing Nuclease-sensitive element-binding P67809 YBX1 51/36 4,78 10 nuclear mRNA splicing, via protein 1 spliceosome Ribonuclease inhibitor P13489 RNH1 51/50 4,78 12 mRNA catabolic process Protein TMED8 Q6PL24 TMED8 51/36 4,78 6 Transport 45 kDa calcium-binding protein Q9BRK5 SDF4 51/42 4,78 8 fat cell differentiation Heat shock protein HSP 90-alpha P07900 HSP90AA1 51/85 4,78 9 G2/M transition of mitotic cell cycle Ataxin-3 P54252 ATXN3 51/42 4,78 6 actin cytoskeleton organization 26S proteasome non-ATPase Q5VWC4 PSMD4 51/41 4,78 7 ubiquitin-dependent protein regulatory subunit 4 catabolic process Alpha-enolase P06733 ENO1 51/47 4,78 4 transcription, DNA-dependent Alpha-N-acetylgalactosaminidase P17050 NAGA 51/47 4,78 5 glycosylceramide catabolic process

104

Elongation factor 1-gamma P26641 EEF1G 51/50 4,78 6 Protein biosynthesis Adipocyte plasma membrane- Q9HDC9 APMAP 51/46 4,78 3 biosynthetic process associated protein 6-phosphogluconate P52209 PGD 51/53 4,78 4 pentose-phosphate shunt, dehydrogenase, decarboxylating oxidative branch Striatin-3 Q13033 STRN3 51/87 4,78 4 negative regulation of intracellular estrogen receptor signaling pathway Eukaryotic initiation factor 4A-II Q14240 EIF4A2 51/46 4,78 5 Host-virus interaction, Protein biosynthesis Actin, cytoplasmic 1 P60709 ACTB 51/42 4,78 4 cell motion Zinc finger protein ZPR1 O75312 ZNF259 51/51 4,78 3 cell proliferation Sialidase-1 Q99519 NEU1 51/45 4,78 3 glycosphingolipid metabolic process Protein TSSC4 Q9Y5U2 TSSC4 51/34 4,78 3 tumor-suppressing subtransferable fragments Protein FAM203A Q9BTY7 FAM203A 51/42 4,78 3 unknown function Endoplasmin P14625 HSP90B1 51/92 4,78 2 ER-associated protein catabolic process Elongation factor 1-alpha 2 Q05639 EEF1A2 51/50 4,78 2 Protein biosynthesis, negative regulation of apoptotic process Transcription intermediary factor Q13263 TRIM28 51/89 4,78 2 DNA repair 1-beta Transitional endoplasmic reticulum P55072 VCP 51/89 4,78 2 retrograde protein transport, ER ATPase to cytosol Cell division cycle protein 123 O75794 CDC123 51/39 4,78 2 cell cycle, cell division homolog Catenin alpha-1 P35221 CTNNA1 51/100 4,78 2 regulation of apoptotic process Receptor-type tyrosine-protein Q15262 PTPRK 51/162 4,78 2 Cell migration, negative phosphatase kappa regulation of cell cycle Dolichyl- P39656 DDOST 51/51 4,78 2 T cell activation diphosphooligosaccharide--protein glycosyltransferase 48 kDa subunit 3413 1,7 Hydroxymethylglutaryl-CoA Q01581 HMGCS1 58/57 5,44 15 cellular lipid metabolic process synthase, cytoplasmic Serine/arginine-rich splicing factor Q13247 SRSF6 58/40 5,44 2 mRNA processing 6

Fold, fold change; MW, molecular weight; pI, isoelectric point. The function or the biological process were obtained from the UniProt database [www.uniprot.org].

105

Additional file 2, Table S2. Mass spectrometry identification of proteins differentially expressed between T47D control cells and T47D cells treated with inhibitor against 17β- HSD1

Spot Fold Identified proteins Accession Gene name Number MW pI Function Number of exp/pred unique (kDa) pepides

SPOT DOWN-REGULATED IN T47D TREATED WITH INHIBITOR AGAINT 17B-HSD1 AS COMPARED TO WT T47D

430 2 Interferon-induced GTP- P20591 MX1 46 87/76 5,96 Induction of apoptosis, GTP binding protein Mx1 catabolic process Prelamin-A/C P02545 LMNA 10 87/74 5,96 Regulation of cell migration and

apoptotic process Programmed cell death 6- Q6NUS1 PDCD6IP 5 87/97 5,96 Apoptotic process, cell cycle interacting protein Acetyl-CoA carboxylase 1 Q13085 ACACA 3 87/266 5,96 Acetyl-CoA metabolic process, Fatty acid biosynthesis Fatty acid synthase P49327 FASN 4 87/273 5,96 Fatty acid biosynthesis Poly(A)-specific O95453 PARN 5 87/73 5,96 RNA modification ribonuclease PARN General transcription factor Q499G6 GTF2I 3 87/110 5,96 Unknown function II, i Synaptotagmin binding, B7Z645 SYNCRIP 4 87/52 5,96 Nucleic acid binding cytoplasmic RNA interacting protein, isoform CRA_b HLA-B associated transcript B0UX83 BAG6 4 87/119 5,96 Plays a key role in apoptosis 3 Epidermal growth factor Q8TE68 EPS8L1 4 87/80 5,96 Stimulates guanine exchange receptor kinase substrate 8- activity of SOS1 like protein 1 CTP synthase 1 P17812 CTPS1 4 87/67 5,96 Response to drug, glutamine metabolic process Protein S100-A9 P06702 S100A9 3 87/13 5,96 Actin cytoskeleton reorganization, induction of apoptosis Ubiquitin-like modifier- P22314 UBA1 3 87/13 5,96 Cell death, protein ubiquitination activating enzyme 1 Transcription intermediary Q13263 TRIM28 2 87/89 5,96 DNA repair factor 1-beta Protein RUFY3 Q7L099 RUFY3 2 87/53 5,96 Negative regulation of axonogenesis Coatomer subunit alpha P53621 COPA 2 87/138 5,96 Intracellular protein transport Clathrin interactor 1 Q14677 CLINT1 2 87/68 5,96 Cellular membrane organization, endocytosis 3201 1,8 Early endosome antigen 1 Q15075 EEA1 19 88/16 5,27 Early endosome to late endosome transport Heat shock protein HSP 90- P08238 HSP90AB1 20 88/83 5,27 Stress response beta Actinin 1 smooth muscle B3V8S3 ACTN1 9 88/103 5,27 Regulation of apoptotic process variant Heat shock protein HSP 90- P07900 HSP90AA1 9 88/85 5,27 Stress response, G2/M transition alpha of mitotic cell cycle Desmoplakin Ia D7RX09 DSP 7 88/279 5,27 Apoptotic process Putative uncharacterized Q658U4 DKFZp666D193 7 88/64 5,27 Protein folding protein DKFZp666D193 Synaptotagmin-like protein 1 Q8IYJ3 SYTL1 6 88/62 5,27 Exocytosis, intracellular protein transport Glutathione S-transferase P21266 GSTM3 5 88/27 5,27 Response to estrogen stimulus, Mu 3 cellular detoxification of nitrogen 106

compound 26S proteasome non-ATPase Q13200 PSMD2 4 88/100 5,27 DNA damage response, signal regulatory subunit 2 transduction by p53 class mediator resulting in cell cycle arrest 78 kDa glucose-regulated P11021 HSPA5 5 88/72 5,27 ER overload response; ER- protein associated protein catabolic process Alpha-actinin-4 O43707 ACTN4 4 88/105 5,27 Actin crosslink formation, response to hypoxia, regulation of apoptotic process Sorting nexin 1, isoform A6NKH4 SNX1 3 88/53 5,27 intracellular protein transport CRA_d Src substrate cortactin Q14247 CTTN 2 88/62 5,27 Regulation of cell migration, plays a role in the invasiveness of cancer cells and the formation of metastases Nucleolin, isoform CRA_c B3KM80 NCL 2 88/59 5,27 Nucleic acid binding Tyrosine-protein kinase C1PHA2 KIF5B-ALK 4 88/168 5,27 Transmembrane receptor protein receptor tyrosine kinase signaling pathway Ran GTPase-activating P46060 RANGAP1 4 88/64 5,27 Mitotic cell cycle protein 1 Plectin Q15149 PLEC 3 88/532 5,27 Cellular component disassembly involved in execution phase of apoptosis Na(+)/H(+) exchange Q5T2W1 PDZK1 2 88/57 5,27 Cell proliferation regulatory cofactor NHE- RF3 DNA damage-binding Q16531 DDB1 2 88/113 5,27 Wnt receptor signaling pathway protein 1 Galectin-3-binding protein Q08380 LGALS3BP 3 88/65 5,27 Cell adhesion Acylamino-acid-releasing P13798 APEH 3 88/82 5,27 Proteolysis enzyme Clathrin heavy chain 1 Q00610 CLTC 2 88/192 5,27 Cellular membrane organization Catenin alpha-1 P35221 CTNNA1 2 88/100 5,27 Cell adhesion, response to estrogen stimulus Transitional endoplasmic P55072 VCP 2 88/89 5,27 ER-associated protein catabolic reticulum ATPase process Exosome component 9 A5PLM5 EXOSC9 2 88/49 5,27 RNA processing Zinc finger protein 326 Q5BKZ1 ZNF326 2 88/66 5,27 Transcription, DNA-dependent NADH-ubiquinone P28331 NDUFS1 2 88/79 5,27 Apoptotic mitochondrial changes oxidoreductase 75 kDa subunit, mitochondrial 4182 1,6 Glutaredoxin-3 O76003 GLRX3 12 43/37 5,47 Cell redox homeostasis Anamorsin Q6FI81 CIAPIN1 5 43/34 5,47 Apoptotic process Adenosylhomocysteinase P23526 AHCY 11 43/48 5,47 S-adenosylhomocysteine catabolic process Gamma-soluble NSF Q99747 NAPG 6 43/35 5,47 Intracellular protein transport attachment protein Transcriptional activator Q96QR8 PURB 5 43/33 5,47 Transcription, DNA-dependent protein Pur-beta Histidyl-tRNA synthetase, B3KWE1 HARS 2 43/50 5,47 Histidyl-tRNA aminoacylation isoform CRA_a Calponin-3 Q15417 CNN3 2 43/36 5,47 Actomyosin structure organization, tropomyosin binding Actin, cytoplasmic 2 P63261 ACTG1 3 43/42 5,47 Cellular membrane organization Guanine nucleotide-binding A8MTJ3 GNAT3 2 43/40 5,47 Adenylate cyclase-modulating G- protein G(t) subunit alpha-3 protein coupled receptor signaling pathway Radixin isoform d A7YIK3 RDX 2 43/30 5,47 Actin filament capping

107

Heterogeneous nuclear O60812 HNRNPCL1 2 43/32 5,47 Nucleotide binding ribonucleoprotein C-like 1 2-oxoisovalerate P21953 BCKDHB 2 43/43 5,47 Response to glucocorticoid dehydrogenase subunit beta, stimulus mitochondrial 1882 1,5 Actin, cytoplasmic 1 P60709 ACTB 10 28/42 5,37 Cellular membrane organization Protein disulfide-isomerase P30101 PDIA3 9 28/57 5,37 Positive regulation of apoptotic A3 process Apolipoprotein A-I-binding Q8NCW5 APOA1BP 7 28/32 5,37 Protein homotetramerization protein Tumor protein D54 O43399 TPD52L2 7 28/22 5,37 Regulation of cell proliferation Heat shock protein beta-1 P04792 HSPB1 3 28/23 5,37 Cell death Cathepsin D P07339 CTSD 3 28/45 5,37 Proteolysis, Involved in the pathogenesis of breast cancer Proteasome activator Q06323 PSME1 2 28/29 5,37 DNA damage response, signal complex subunit 1 transduction by p53 class mediator resulting in cell cycle arrest Elongation factor 2 P13639 EEF2 2 28/95 5,37 Translational elongation Cytokine-like nuclear factor D3DUE6 N-PAC 2 28/61 5,37 Pentose-phosphate shunt n-pac, isoform CRA_c

SPOT UP-REGULATED IN T47D TREATED WITH INHIBITOR AGAINT 17B-HSD1 AS COMPARED TO WT T47D

445 2,7 Lysosomal alpha-glucosidase P10253 GAA 13 86/105 5,79 Essential for the degradation of glygogen to glucose in lysosomes Gelsolin P06396 GSN 5 86/86 5,79 Actin filament polymerization Filaggrin-2 Q5D862 FLG2 5 86/248 5,79 Calcium ion binding Protein disulfide-isomerase P13667 PDIA4 7 86/73 5,79 Cell redox homeostasis A4 Annexin A6, isoform CRA_c A6NN80 ANXA6 4 86/75 5,79 Calcium-dependent phospholipid binding Alanyl-tRNA synthetase, P49588 AARS 4 86/109 5,79 Protein biosynthesis cytoplasmic Keratinocyte proline-rich Q5T749 KPRP 6 86/64 5,79 Cellular protein modification protein process Stress-70 protein, P38646 HSPA9 4 86/72 5,79 Protein folding, negative mitochondrial regulation of apoptotic process Septin-9 Q9UHD8 SEPTIN9 2 86/65 5,79 Cell cycle, cell division Vesicle-fusing ATPase P46459 NSF 2 86/83 5,79 Protein transport Desmoglein-1 Q02413 DSG1 2 86/114 5,79 Cell adhesion, response to progesterone stimulus WD repeat-containing Q9H7D7 WDR26 3 86/72 5,79 Involved in MAPK pathways protein 26 Apolipoprotein D C9JF17 APOD 2 86/24 5,79 Lipid binding SEC16A protein A4QN18 SEC16A 2 86/211 5,79 COPII vesicle coating SEC31A protein B7ZL00 SEC31A 2 86/130 5,79 Cellular protein metabolic process Skin-specific protein 32 Q5T750 XP32 2 86/26 5,79 Epidermis development 2198 2,3 14-3-3 protein epsilon P62258 YWHAE 13 23/29 5,25 G2/M transition of mitotic cell cycle Lamin-B1 P20700 LMNB1 3 23/66 5,25 Cellular component disassembly involved in execution phase of apoptosis Prostaglandin reductase 2 Q8N8N7 PTGR2 2 23/38 5,25 Prostaglandin metabolic process 2-aminoethanethiol Q96SZ5 ADO 2 23/30 5,25 Metal ion binding dioxygenase 2779 2 Histone H4 P62805 HIST1H4A 2 15 11 5,86 CENP-A-containing nucleosome assembly at centromere 2230 1,9 14-3-3 protein zeta/delta P63104 YWHAZ 13 23/28 5,6 Negative regulation of apoptotic process 2177 1,8 Adenine P07741 APRT 7 23/20 5,93 Purine salvage

108

phosphoribosyltransferase Diphosphoinositol O95989 NUDT3 3 23/19 5,93 Cell-cell signaling polyphosphate phosphohydrolase 1 Malectin Q14165 MLEC 3 23/32 5,93 Post-translational protein modification Ras-related C3 botulinum A4D2P0 RAC1 2 23/23 5,93 Regulation of cell migration, toxin substrate 1 (Rho Wnt receptor signaling pathway family, small GTP binding protein Rac1) 9.5S alpha-1-glycoprotein P02743 APCS 2 23/25 5,93 Protein folding Calcium-regulated heat stable Q9Y2V2 CARHSP1 2 23/16 5,93 Intracellular signal transduction protein 1 2064 1,7 Splicing factor, B4E241 SRSF3 9 25/14 6,32 Nucleotide binding arginine/serine-rich 3, isoform CRA_a Transgelin-2 P37802 TAGLN2 3 25/22 6,32 Muscle organ development 1936 1,6 Putative RNA-binding Q9Y383 LUC7L2 5 27/46 5,68 May bind to RNA via its protein Luc7-like 2 Arg/Ser-rich domain Glycerol-3-phosphate Q8N335 GPD1L 3 27/38 5,68 glycerol-3-phosphate catabolic dehydrogenase 1-like protein process Tumor protein D53 Q16890 TPD52L1 2 27/22 5,68 G2/M transition of mitotic cell cycle Uncharacterized O43709 WBSCR22 2 27/32 5,68 Methylation methyltransferase WBSCR22 Exosome complex Q5RKV6 EXOSC6 2 27/28 5,68 DNA deamination component MTR3 RAB11B protein A5YM50 RAB11B 2 27/25 5,68 Protein transport, small GTPase mediated signal transduction Transmembrane emp24 Q9Y3B3 TMED7 2 27/25 5,68 Protein transport domain-containing protein 7 Cytosol aminopeptidase P28838 LAP3 2 27/56 5,68 Proteolysis

Fold, fold change; MW, molecular weight; pI, isoelectric point. The function or the biological process was obtained from the UniProt database [www.uniprot.org]

109

Additional file 3, Table S3. Mass spectrometry identification of proteins differentially expressed between MCF7 control cells and MCF7 cells transfected with 17β-HSD1 siRNA

Spot Fold Identified proteins Accession Gene name Number MW pI Function Number of unique pepides DOWN REGULATED PROTEIN FOLLOWING siRNA TRANSFECTION 2086 2,4 V-type proton ATPase subunit d 1 F5GYQ1 ATP6V0D1 11 40/40 4,79 ATP hydrolysis coupled proton transport Actin, cytoplasmic 1 P60709 ACTB 11 40/42 4,79 involved in various types of cell motility Elongation factor 1-delta P29692 EF1D 10 40/71 4,79 regulation of transcription Cytoskeleton-associated protein 4 Q07065 CKAP4 15 40/66 4,79 Mediates the anchoring of the endoplasmic reticulum to microtubules Mitochondrial heat shock 60kD B3GQS7 HSPD1 9 40/61 4,79 Stress response protein 1 variant 1 Hepatoma-derived growth factor P51858 HDGF 5 40/27 4,79 Transcription regulation Heterogeneous nuclear G3V4C1 HNRNPC 3 40/34 4,79 Nucleic acid binding ribonucleoproteins C1/C2 Guanine nucleotide-binding protein- Q9BVP2 GNL3 5 40/62 4,79 regulation of cell proliferation like 3 Actin-related protein 3 B4DXW1 ACTR3 3 40/42 4,79 positive regulation of actin filament polymerization Soluble calcium-activated Q8WVQ1 CANT1 3 40/39 4,79 positive regulation of I-kappaB nucleotidase 1 kinase/NF-kappaB cascade IQ motif containing GTPase A4QPB0 IQGAP1 3 40/189 4,79 positive regulation of Ras GTPase activating protein 1 activity Serine/arginine-rich splicing factor Q13247 SRSF6 2 40/39 4,79 mRNA splicing 6 14-3-3 protein zeta/delta (Fragment) E7EX29 YWHAZ 2 40/28 4,79 negative regulation of apoptotic process DNA damage-binding protein 1 B4DG00 DDB1 2 40/51 4,79 nucleic acid binding 3157 2,1 Phosphoglycerate kinase B4DHB3 PGK 2 23/30 4,92 Glycolysis Protein unc-45 homolog A Q9H3U1 UNC45A 4 23/102 4,92 cell differentiation Chromobox protein homolog 3 Q13185 CBX3 3 23/21 4,92 chromatin remodeling Proteasome subunit alpha type-6 B4DXJ9 PSMA6 2 23/25 4,92 proteolysis involved in cellular protein catabolic process GTPase Nras P01111 NRAS 2 23/21 4,92 Ras protein signal transduction BH3 interacting domain death B3KT21 BID 2 23/27 4,92 positive regulation of apoptotic agonist process 2431 2 Alpha-enolase P06733 ENO1 32 36/47 5,97 negative regulation of cell growth T-complex protein 1 subunit epsilon E7ENZ3 CCT5 10 36/60 5,97 protein folding Glycerol-3-phosphate Q8N335 GPD1L 10 36/38 5,97 NADH metabolic process dehydrogenase 1-like protein COP9 signalosome complex subunit Q13098 GPS1 5 36/55 5,97 cell cycle 1 Voltage-dependent anion-selective B4DKM5 VDAC2 4 36/27 5,97 voltage-gated anion channel channel protein 2 activity Procollagen galactosyltransferase 1 Q8NBJ5 COLGALT1 4 36/72 5,97 lipopolysaccharide biosynthetic process

110

Asparagine synthetase B4DXZ1 ASNS 4 36/62 5,97 L-asparagine biosynthetic process Palmitoyl-protein thioesterase 1 P50897 PPT1 4 36/37 5,97 apoptotic DNA fragmentation Voltage-dependent anion-selective P21796 VDAC1 3 36/31 5,97 apoptotic process channel protein 1 Calcineurin-like phosphoesterase Q9BRF8 CPPED1 3 36/36 5,97 hydrolase activity domain-containing protein 1 DNA replication licensing factor Q14566 MCM6 3 36/93 5,97 G1/S transition of mitotic cell cycle MCM6 Acyl-CoA synthetase family B4DFQ6 ACSF2 3 36/71 5,97 ligase activity member 2, mitochondrial F-actin-capping protein subunit beta P47756 CAPZB 3 36/31 5,97 actin cytoskeleton organization Elongation factor 1-gamma B4DTG2 EEF1G 2 36/56 5,97 Protein biosynthesis Retinoic acid receptor RXR-alpha P19793 RXRA 2 36/51 5,97 Transcription regulation 3843 1,9 Leukotriene A-4 hydrolase P09960 LTA4H 12 39/69 5,21 Leukotriene biosynthesis T-complex protein 1 subunit theta B4DEM7 CCT8 9 39/58 5,21 protein folding Truncated nucleolar phosphoprotein A4ZU86 NPM1 6 39/30 5,21 nucleic acid binding B23 Endoplasmin P14625 HSP90B1 7 39/92 5,21 ER-associated protein catabolic process SEC23-interacting protein Q9Y6Y8 SEC23IP 7 39/111 5,21 Plays a role in the organization of endoplasmic reticulum exit sites Serine/arginine-rich-splicing factor G5E9M3 SRSF7 5 39/26 5,21 mRNA processing 7 Ubiquitin carboxyl-terminal Q9Y5K5 UCHL5 5 39/38 5,21 DNA repair hydrolase isozyme L5 Transcription intermediary factor 1- Q13263 TRIM28 3 39/89 5,21 Transcription regulation beta HUMAN Hexokinase-1 P19367 HK1 4 39/102 5,21 cell death Heat shock protein HSP 90-alpha P07900 HSP90AA1 4 39/85 5,21 G2/M transition of mitotic cell cycle Glyceraldehyde-3-phosphate E7EUT5 GAPDH 3 39/28 5,21 glyceraldehyde-3-phosphate dehydrogenase dehydrogenase (NAD+) (phosphorylating) activity Eukaryotic translation initiation P05198 EIF2S1 3 39/36 5,21 activation of signaling protein factor 2 subunit 1 activity involved in unfolded protein response Calcium-binding mitochondrial Q6NUK1 SLC25A24 2 39/53 5,21 transmembrane transport carrier protein SCaMC-1 [ Histone-lysine N-methyltransferase Q86TU7 SETD3 2 39/67 5,21 Transcription regulation setd3 2856 1,8 Tubulin alpha-1A chain Q71U36 TUBA1A 2 29/46 5,28 G2/M transition of mitotic cell cycle 3397 1,8 Annexin H0YMU9 ANXA2 15 19/26 5,35 calcium ion binding E9PK25_HUMAN Cofilin-1 E9PK25 CFL1 3 19/23 5,35 actin filament depolymerization Metastasis inhibition factor nm23 E7ERL0 NME1 2 19/15 5,35 regulation of apoptotic process and cell proliferation 4666 1,7 Tubulin, beta 2C Q8N6N5 TUBB2C 6 19/50 4,77 microtubule-based process CNPY2_HUMAN Protein canopy Q9Y2B0 CNPY2 7 19/21 4,77 negative regulation of gene homolog 2 expression Lysosomal alpha-glucosidase P10253 GAA 3 19/105 4,77 Essential for the degradation of glygogen to glucose in lysosomes Trafficking protein particle complex A6NDN0 TRAPPC3 2 19/15 4,77 ER to Golgi vesicle-mediated 3, isoform CRA_a transport SEC23-interacting protein Q9Y6Y8 SEC23IP 2 19/111 4,77 Golgi organization Transforming protein RhoA P61586 RHOA 2 19/22 4,77 Rho protein signal transduction Diablo homolog, mitochondrial F5GXT8 DIABLO 2 19/15 4,77 Activation of cysteine-type

111

(Fragment) endopeptidase activity involved in apoptotic process DNA polymerase epsilon subunit 3 Q9NRF9 POLE3 2 19/17 4,77 DNA replication 4665 1,7 Prohibitin P35232 PHB 13 32/30 5,6 DNA biosynthetic process Latexin Q9BS40 LXN 5 32/26 5,6 Inflammatory response Cathepsin D P07339 CTSD 8 32/45 5,6 proteolysis, Involved in the pathogenesis of breast cancer Polyamine-modulated factor 1 Q6P1K2 PMF1 10 32/23 5,6 cell division Phosphomannomutase 1 Q92871 PMM1 10 32/30 5,6 post-translational protein modification Protein disulfide-isomerase A3 F5H119 PDIA3 9 32/57 5,6 cell redox homeostasis Heat shock protein beta-1 P04792 HSPB1 7 32/23 5,6 Involved in stress resistance and actin organization Heterogeneous nuclear G8JLB6 HNRNPH1 6 32/51 5,6 nucleic acid binding ribonucleoprotein H, N-terminally processed Glutathione S-transferase Mu 3 P21266 GSTM3 7 32/27 5,6 response to estrogen stimulus Synaptosomal-associated protein 29 O95721 SNAP29 8 32/29 5,6 protein transport Tumor protein D54 O43399 TPD52L2 5 32/22 5,6 regulation of cell proliferation E3 ubiquitin-protein ligase CHIP Q9UNE7 STUB1 5 32/35 5,6 DNA repair Eukaryotic initiation factor 4A-III P38919 EIF4A3 5 32/47 5,6 mRNA processing Alpha-soluble NSF attachment P54920 NAPA 4 32/33 5,6 Protein transport protein Chloride intracellular channel O00299 CLIC1 3 32/27 5,6 signal transduction protein 1 Pre-mRNA-splicing factor SPF27 O75934 BCAS2 3 32/26 5,6 mRNA processing Transitional endoplasmic reticulum P55072 VCP 3 32/89 5,6 ER-associated protein catabolic ATPase process Alpha N-terminal protein Q9BV86 NTMT1 3 32/25 5,6 N-terminal peptidyl-alanine methyltransferase 1A trimethylation Transmembrane emp24 domain- Q9Y3B3 TMED7 3 32/25 5,6 Protein transport containing protein 7 E3 ubiquitin-protein ligase UBR5 E7EMW7 UBR5 2 32/309 5,6 protein ubiquitination 40S ribosomal protein S4, X P62701 RPS4X 2 32/30 5,6 positive regulation of cell isoform proliferation Anamorsin H3BT65 CIAPIN1 2 32/29 5,6 apoptotic process Ras-related protein Rab-21 Q9UL25 RAB21 2 32/24 5,6 small GTPase mediated signal transduction Histone H3.1 P68431 HIST1H3A 2 32/16 5,6 regulation of gene silencing 3245 1,6 14-3-3 protein gamma P61981 YWHAG 9 22/28 5,73 G2/M transition of mitotic cell cycle Fatty acid synthase P49327 FASN 6 22/273 5,73 Fatty acid biosynthesis Replication factor C (Activator 1) 4, B4DM41 RFC4 3 22/34 5,73 ATP binding 37kDa, isoform CRA_b Sialic acid synthase Q9NR45 NANS 3 22/40 5,73 lipopolysaccharide biosynthetic process Nucleolin P19338 NCL 2 22/68 5,73 Angiogenesis Glucose-6-phosphate 1- P11413 G6PD 2 22/59 5,73 Carbohydrate metabolism dehydrogenase

UP REGULATED PROTEIN FOLLOWING siRNA TRANSFECTION

3746 2,1 Rho GTPase-activating protein 1 Q07960 ARHGAP1 25 54/50 6,14 Rho protein signal transduction Serine/threonine-protein P63151 PPP2R2A 15 54/52 6,14 signal transduction

112

phosphatase 2A 55 kDa regulatory subunit B alpha isoform Pyruvate kinase isozymes M1/M2 P14618 PKM 7 54/58 6,14 Glycolysis RuvB-like 1 (Fragment) B5BUB1 RUVBL1 7 54/50 6,14 DNA duplex unwinding Heat shock 70 kDa protein 1A/1B P08107 HSPA1A 6 54/70 6,14 Stress response Septin-8 A6NFQ9 SEPTIN 8 7 54/51 6,14 cell cycle E3 ubiquitin-protein ligase TRIM21 F5H012 TRIM21 6 54/54 6,14 zinc ion binding Heterogeneous nuclear G8JLB6 HNRNPH1 6 54/51 6,14 nucleic acid binding ribonucleoprotein H, N-terminally processed 4-trimethylaminobutyraldehyde P49189 ALDH9A1 6 54/54 6,14 hormone metabolic process dehydrogenase Serine hydroxymethyltransferase B4DJ63 SHMT1 4 54/42 6,14 One-carbon metabolism Pre-mRNA-processing factor 19 Q9UMS4 PRPF19 4 54/55 6,14 DNA damage Aldehyde dehydrogenase, dimeric E9PNN6 ALDH3A1 3 54/44 6,14 cellular aldehyde metabolic process NADP-preferring DnaJ homolog subfamily A member O60884 DNAJA2 3 54/46 6,14 positive regulation of cell 2 proliferation Mitochondrial intermediate Q99797 MIPEP 2 54/81 6,14 Proteolysis peptidase Elongation factor 1-alpha A8K9C4 EF1A 2 54/50 6,14 Protein biosynthesis Cold shock domain-containing E9PGZ0 CSDE1 2 54/91 6,14 regulation of transcription, DNA- protein E1 dependent UDP-glucose 6-dehydrogenase O60701 UGDH 2 54/55 6,14 UDP-glucose metabolic process Heterogeneous nuclear P55795 HNRNPH2 2 54/49 6,14 mRNA splicing, via spliceosome ribonucleoprotein H2 895 2,1 Nuclear receptor-binding protein Q9UHY1 NRBP1 8 71/58 4,92 ER to Golgi vesicle-mediated transport PACSIN2 protein Q6FIA3 PACSIN2 9 71/48 4,92 Eyes absent homolog 3 B4DIR7 EYA3 7 71/58 4,92 multicellular organismal development Signal transducing adapter molecule O75886 STAM2 6 71/58 4,92 Protein transport 2 DNA replication licensing factor P25205 MCM3 6 71/91 4,92 G1/S transition of mitotic cell cycle MCM3+ Protein disulfide-isomerase A4 P13667 PDIA4 6 71/73 4,92 cell redox homeostasis ITGAV protein A5YM53 ITGAV 5 71/116 4,92 Cell adhesion Ubiquitin-like modifier-activating P22314 UBA1 5 71/118 4,92 Ubl conjugation pathway enzyme 1 Yorkie homolog F5H202 YAP1 4 71/53 4,92 positive regulation of cell proliferation Heat shock 70 kDa protein 4 P34932 HSPA4 3 71/94 4,92 chaperone-mediated protein complex assembly Retinoblastoma-binding protein 5 Q15291 RBBP5 2 71/45 4,92 regulation of transcription, DNA- dependent Protein BANP B2RCF7 BANP 2 71/51 4,92 regulation of transcription, DNA- dependent Golgin subfamily A member 1 Q92805 GOLGA1 2 71/62 4,92 protein targeting to Golgi Pyridoxal-dependent decarboxylase H3BNZ1 PDXDC1 2 71/85 4,92 carboxylic acid metabolic process domain-containing protein 1 4672 1,7 Serine/threonine-protein kinase O95747 OXSR1 9 62/58 6,05 response to oxidative stress OSR1 Sorting nexin-27 Q96L92 SNX27 6 62/61 6,05 Protein transport Serine/threonine-protein kinase Q13177 PAK2 7 62/58 6,05 regulation of growth PAK 2

113

Dopamine receptor interacting Q4W4Y1 DRIP4 9 62/96 6,05 Unknown function protein 4 T-complex protein 1 subunit alpha P17987 TCP1 7 62/60 6,05 de novo' posttranslational protein folding Pyruvate kinase isozymes M1/M2 P14618 PKM 7 62/58 6,05 Glycolysis Torsin-1A-interacting protein 1 Q5JTV8 TOR1AIP1 6 62/66 6,05 nuclear membrane organization Copine-3 O75131 CPNE3 4 62/60 6,05 lipid metabolic process Prolyl 4-hydroxylase subunit alpha- C9JL12 P4HA1 4 62/59 6,05 collagen fibril organization 1 Coiled-coil domain-containing Q6PII3 C3orf19 3 62/54 6,05 protein 174 Glucosamine fructose-6-phosphate Q06210 GFPT1 2 62/79 6,05 Controls the flux of glucose into the aminotransferase [isomerizing] 1 hexosamine pathway U2 small nuclear RNA auxiliary B5BU25 U2AF2 2 62/53 6,05 mRNA processing factor 2 isoform b Pre-B-cell leukemia transcription H0YLB0 PBX1 2 62/192 6,05 positive regulation of cell factor 1 proliferation

Fold, fold change; MW, molecular weight; pI, isoelectric point. The function or the biological process was obtained from the UniProt database [www.uniprot.org]

114

Additional file 4, Table S4. Primers used for quantitative real-time RT-PCR

T Gene Taille Séquences des amorces 5'→ 3' Description GenBank Région annealing Symbol (pb) S/AS (°C) TTCTTTGTCCCCTGGGTCTG Homo sapiens hydroxysteroid (17-beta) 1986- 17β-HSD1 NM_000413 215 60 TGT/ATGGGGGTCTCACTGT dehydrogenase 1 2200 GTTGCT

Homo sapiens NME/NM23 nucleoside TGTGGAGAGTGCAGAGAAG NME1 diphosphate kinase 1 (NME1), région NM_198175 694-833 140 60 GAGA/GAAGGAGGGGAAAT commune aux 2 transcrits GGATGTGA TCGGCTACGTGGGGATTGAC Homo sapiens septin 9 (SEPT9), région 1585- SEPT9 NM_006640 234 60 T/CGCCTTTCTCCTCAATATC commune aux 7 transcrits 1818 GTG CGTCACCAAAAAGCGCAAA Homo sapiens lamin A/C (LMNA), 1491- LMNA NM_170707 204 60 CTG/CGGTAAGTCAGCAAGG région commune aux 3 transcrits 1694 GATCATCT Homo sapiens tumor protein D52-like 2 GCCGGCCAAGATATCAACC TPD52L2 (TPD52L2), région commune aux 6 NM_199360 148-332 185 60 TG/TGGCGCAGAGTGACAAT transcrits (TPD54) TTCCT Homo sapiens tumor protein D52-like 1 CTAATGGAGGCAGTTTTGAG NM_0010033 TPD52L1 (TPD52L1), région commune aux 4 749-848 100 60 GAG/TGCAGCTCCTCCTCCTT 95 transcrits (TPD53) GGTC Homo sapiens tyrosine 3- CAGGCTGAGCGATACGACG monooxygenase/tryptophan 5- YWHAE NM_006761 198-387 190 60 AA/GCTTGTCTTCTCCTCCCT monooxygenas activation protein, TGTTTTCT epsilon polypeptide (YWHAE) GGTTCTCGGGTACTGGGATA Homo sapiens glutathione S-transferase GSTM3 NM_000849 331-424 94 60 TTC/CGTGTACCGTTTCTCCT mu 3 (brain) (GSTM3) CATAAG Homo sapiens ATP synthase, H+ ATTGAAGGTCGCTATGCCAC Atp5o transporting, mitochondrial F1 NM_001697 200-466 267 60 AG/AACGACTCCTTGGGTAT complex, O subunit TGCTTAA AGTTCTGTGGCCATCTGCTT Homo sapiens hypoxanthine Hprt1 NM_000194 814-971 157 60 AGTAG/AAACAACAATCCGC phosphoribosyltransferase 1 CCAAAGG Homo sapiens glucose-6-phosphate GATGTCCCCTGTCCCACCAA 2490- G6PD dehydrogenase (G6PD), nuclear gene NM_000402 121 60 CTCTG/GCAGGGCATTGAGG 2610 encoding mitochondrial protein TTGGGAG ACGGACCAGAGCGAAAGCA 983- 18S Homo sapiens 18S ribosomal RNA NR_003286 226 60 TT/TCCGTCAATTCCTTTAAG 1208 TTTCAGCT

115

116

Chapitre 3: The effect of 17β-HSD1 inhibition on the transcriptome of T47D breast cancer cells

117

118

Effect of 17β-HSD1 inhibition on the transcriptome of T47D breast cancer cells

Mouna Zerradi1 and Sheng-Xiang Lin1*

1Molecular Endocrinology and Genomics Research Centre, CHUL (CHUQ) Research Centre and Laval University, Québec, Canada G1V 4G2

RUNNING TITLE: Inhibition of 17β-HSD1 in breast cancer cell

KEYWORDS: 17β-HSD1, Breast cancer cells, T47D, transcriptome.

Cet article est présentement en voie de soumission dans un journal scientifique.

119

Abstract

Background. Estrogens play important role in the development and progression of hormone- dependent breast carcinomas. The enzyme 17β-hydroxysteroid dehydrogenase type 1 (17β- HSD1) catalyzes significantly the conversion of inactive estrogen estrone to biologically active estradiol. The aim of this study was to investigate the impact of 17β-HSD1 inhibition on gene expression profiles in breast cancer cells.

Methods. The breast cancer epithelial cell line T47D was used to perform all experiments. A transcriptome analysis was performed to compare the genome of control T47D cells and T47D cells treated with 17β-HSD1 inhibitor. The genomic analysis was performed using affymetrix DNA microarrays followed by Q-RT-PCR validation.

Results. After the inhibition of 17β-HSD1 in T47D cells by a specific inhibitor, 33 genes were up-regulated and 12 genes were down-regulated. The expression of genes implicated in transport (RANBP3L, APOD), DNA binding (HIST1H2BM), antigen processing and presentation of peptide or polysaccharide antigen via MHC class II (HLA-DQA2) decreased. The expression of TP63 (transcriptional regulator) and CD36 (involved in cell adhesion) increased. The Q-RT-PCR validated the microarray results.

Conclusion. This study shows that inhibition of 17β-HSD1 modulates the transcriptome of T47D cells and affects the expression of genes involved in transport, DNA binding and cell adhesion.

120

Résumé en français

Introduction. Les estrogènes jouent un rôle important dans le développement et la progression du cancer du sein hormono-dépendant. L’enzyme 17β-hydroxystéroïde déshydrogénase de type 1 (17β-HSD1) catalyse significativement la conversion de l’estrone, un estrogène inactif, en estradiol, un estrogène biologiquement actif. L’objectif de cette étude était de caractériser l’impact de l’inhibition de la 17β-HSD1 sur le profil d’expression des cellules du cancer du sein.

Méthodologie. La lignée cellulaire du cancer du sein T47D a été utilisée pour effectuer toutes les expériences. Une analyse génomique a été réalisée pour comparer le transcriptome des cellules T47D contrôle et des cellules T47D traitées avec un inhibiteur spécifique à la 17β-HSD1. L'analyse a été réalisée à l'aide de puces à ADN suivie par une Q -RT- PCR pour confirmer les résultats.

Résultats. Suite à l'inhibition de la 17β-HSD1 dans les cellules T47D par un inhibiteur spécifique, 33 gènes ont été surexprimés et 12 gènes ont été sousexprimés. L'expression des gènes impliqués dans le transport (RANBP3L, APOD), la liaison d'ADN (HIST1H2BM), le traitement de l'antigène et la présentation de l'antigène de peptide ou de polysaccharide par l'intermédiaire du CMH de classe II (HLA- DQA2) était diminuée. Tandis que l'expression de TP63 (régulateur transcriptionnel), et CD36 (impliquée dans l'adhérence cellulaire) a augmenté. Les résultats de Q-RT-PCR ont validé les résultats des puces à ADN.

Conclusion. Cette étude montre que l'inhibition de la 17β-HSD1 module le profil génomique des cellules T47D et affecte l'expression de plusieurs gènes impliqués dans le transport, la liaison à l’ADN et l’adhésion cellulaire. .

121

1. Introduction

Breast cancer is one of the most common forms of cancer observed in women [1]. Estrogens have an important role in the development of hormone-dependent breast carcinomas [2, 3]. Estradiol (E2) is produced by two biosynthetic pathways: the aromatase and steroid sulfatase pathways. 17β-HSD type 1 catalyzes the conversion of inactive estrogen estrone (E1) to biologically active estrogen E2 using nicotinamide adenine dinucleotide phosphate (NADPH) as a cofactor [4-9]. In breast cancer specimens, a high 17β-HSD1 expression is associated with a poor prognosis [10].

Activation of the estrogen receptor (ER) through binding of E2 is crucial for the growth and development of many mammary tumors [11]. Estrogen effects are exerted through two types of specific receptor: estrogen receptors alpha (ERα) and beta (ERβ) [12, 13]. Estrogen receptors act mainly by regulating the expression of target genes whose promoters contain specific sequences called estrogen-responsive element (ERE). After ERE-binding of ligand- bound ER dimers, modulation of transcription occurs via interaction with coactivators or corepressors. Together, these complexes play an important role in the recruitment of transcriptional machinery, the modulation of chromatine structure, and then in the regulation of ER target-gene expression [1, 14].

Inhibition of 17β-HSD1 activity has been used to blunt the final step of E2 synthesis. A specific inhibitor (2-methoxy-E2 C-16 derivatives, Compound 18) bearing a methoxy group at position 2 generates strong inhibition of the transformation of E1 to E2 in T47D breast cancer cells (81%, 69%, and 37% at 10, 1, and 0.1 µM, respectively) [15]. The 17β-HSD1 can regulate estrogen-related genes such as pS2 which is an estrogen-responsive gene [9, 16]. Moreover, 17β-HSD1 expression decreases the mRNA levels of androgen receptor by 64%. While, increasing that of the estrogen receptors ER alpha and ER beta by 171 and 120%,

122

respectively [17]. The aim of this study was to investigate the role of 17β-HSD1 inhibition in regulating the transcriptome of T47D breast cancer cells.

2. Materials and methods

2.1. Cell culture

T47D cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA). T47D cells were cultured in high-glucose, phenol red-free DMEM, supplemented with 7.5 mg/1 bovine insulin (Sigma, Oakville, Ontario, Canada). The cells were cultured in the presence of 10% FBS at 37°C in a humidified atmosphere of 5% CO2.

2.2. 17β-HSD1 steroidal inhibitor

The 2-methoxy-E2 C-16 derivative inhibitor of 17β-HSD1 (compound 18) [15] was used in the present study as previously described [16]. The 2-methoxy-E2 C-16 derivative inhibitor showed an inhibition rates of 81%, 69%, and 37% at 10, 1, and 0.1 µM, respectively in T47D cells [15].

2.3. Activity assay

25,000 T47D cells were plated onto 24-well plates and cultured in medium treated with dextran-coated charcoal. Time course experiments were carried out after treatment with various concentrations of 17β-HSD1 inhibitor. A final concentration of 0.2 µM [14C]-E1 (American Radiolabeled Chemicals, St Louis, USA) was added to the culture medium and the reaction was stopped at 0, 10, 30 and 60 min. After the incubation, steroids were extracted with 3 volumes of diethyl ether on ethanol dry ice bath. The organic phase was evaporated, dissolved in 50 µl dichloromethane, applied to Silica gel 60 thin layer

123 chromatography plates (Merck, Darmstad, Germany), and separated by migration in a toluene-acetone (4:1, v/v) solvent system. TLC plates were exposed and quantified using a storm imaging system (Molecular Dynamics, Sunnyvale, CA).

2.4. DNA microarray processing

T47D cells treated with 17β-HSD1 inhibitor (5 µM) or ethanol vehicle were cultured for 10 days in complete growth medium. Total RNA was then isolated from cells using Qiazol Lysis Reagent (Qiagen, Hilden, DE). The experiment was performed in duplicate. DNA microarray analyses were carried out with Affymetrix Human Gene 1.0 ST arrays according to the Affymetrix standard protocol. Briefly, total RNA (200 ng per sample) was labeled using the Affymetrix GeneChip® WT cDNA Synthesis and Amplification Kit protocol, and hybridized to the arrays as described by the manufacturer (Affymetrix, Santa Clara, CA). The cRNA hybridization cocktail was incubated overnight at 45°C while rotating in a hybridization oven. After 16 h of hybridization, the cocktail was removed and the arrays were washed and stained in an Affymetrix GeneChip fluidics station 450, according to the Affymetrix-recommended protocol (http://media.affymetrix.com/support/downloads/manuals/wt_sensetarget_label_manual. pdf). The arrays were scanned using the Affymetrix GCS 3000 7G and the Gene-Chip Operating Software (Affymetrix, Santa Clara, CA), to produce the intensity files. Microarray analyses were performed by the CHU de Québec Gene Expression Platform, Québec, Québec, Canada.

2.5. Microarray analysis

The background subtraction and normalization of probe set intensities was performed using the method of Robust Multiarray Analysis (RMA) described by Irizarry et al. [18]. To identify differentially-expressed genes, gene expression intensity was compared using a moderated t test and a Bayes smoothing approach developed for a low number of replicates [19]. To correct for the effect of multiple testing, the false discovery rate was estimated from

124

p values derived from the moderated t test statistics [20]. The analysis was performed using the affylmGUI Graphical User Interface for the Limma microarray package [21] and gene ontology analysis by GO enrichment functionality of Partek Genomics Suite software.

2.6. Quantitative real-time RT-PCR

Total RNA was extracted from T47D cells by using the Qiazol Lysis Reagent (Qiagen, Hilden, DE) and assayed by spectrophotometry (Nano Drop) in duplicate. The quality of total RNA samples was verified with a bioanalyzer (Agilent technologies, Mississauga, Ontario, Canada). The analysis showed high quality for all RNA samples with RIN (RNA Integrity Number) above than 9.1 on a scale of 10. The quantification by real-time PCR (qPCR) for each condition was performed in duplicate as previously described [16, 22]. Normalization was performed using the reference genes shown to have stable expression levels from embryonic life through adulthood in various tissues [23]: ATP synthase O subunit (ATP5O), hypoxanthine guanine phosphoribosyl transferase 1 (HPRT1), glucose-6-phosphate dehydrogenase (G6PD) and 18S ribosomal RNA (18S). The additional file 1 shows the design of specific primers used for the amplification. Messenger RNA abundance was expressed as copies/µg total RNA. Quantitative real-time PCR measurements were performed by the CHU de Québec Gene Expression Platform, Québec, Québec, Canada.

2.7. Statistics

Values are expressed as mean ± SD. Student t tests were used for comparaison between two means. p value < 0.05 was considered as statistically significant.

3. Results and discussion

125

3.1. 17β-HSD1 expression and inhibition in T47D cells

To assess the specifity of the 17β-HSD1 inhibitor, total protein extracts from T47D cells were used in activity assays for the conversion of 14C-E1 into 14C-E2 after 0, 10, 30 and 60 min, in absence (vehicle ethanol) or presence of inhibitor (5 µM). The specificity of 17β- HSD1 inhibitor was confirmed, because the inhibition of 17β-HSD1 leads to inhibition of the conversion of E1 to E2 (Figure 1). It has also been demonstrated that the inhibitor lead inhibition rates of 81%, 69%, and 37% at 10, 1, and 0.1 µM respectively in T47D cells [15].

3.2. The inhibition of 17β-HSD1 modulates the genomic profile of breast cancer cell line T47D

To assess the effect 17β-HSD1 inhibition on T47D cell transcriptome, we performed microarray analyses. Out of over 28,000 transcripts on the array, 45 had a fold change ≥ 1.8 and considered for the present study. 33 genes were up-regulated and 12 genes were downregulated following the 17β-HSD1 inhibition in T47D cells. The UniProt database [www.uniprot.org] was used to determine the function of protein related to each transcripts coding (Table 1). 17β-HSD1 inhibition in T47D cells decrease the expression of transcript coding for proteins implicated in transport (RANBP3L, APOD), DNA binding (HIST1H2BM), antigen processing and presentation of peptide or polysaccharide antigen via MHC class II (HLA-DQA2). While, the expression of Tumor protein p63 (transcriptional regulator) and CD36 (involved in cell adhesion) was up-regulated.

CD36 was among the transcripts that were upregulated after 17β-HSD1 inhibition. CD36 is a transmembrane receptor that is involved in apoptosis, cell-ECM interactions, activation of TGF-β, adipocyte differentiation, angiogenesis, and immune signaling [24]. In all tumor subtypes obtained from women (ER+/ PR+, HER2+, or ER-/PR-/HER2), CD36 expression was absent in comparison with the normal adjacent tissue [25]. Also, in reduction mammoplasty tissues (RMFs), the repression of CD36 expression, decreased adipocyte differentiation and increased matrix accumulation [25]. In endothelial cells, CD36 has been

126

shown to block VEGF action [26]. In our study the high expression of CD36 following 17β- HSD1 inhibition could promote an anti-angiogenic profile.

Another upregulated transcript after 17β-HSD1 inhibition was p63, which is a member of the p53 family, involved in cancer pathophysiology [27]. In human cancers, p63 is often overexpressed or deregulated [28]. The p63 has two major isoforms TA or ΔN and each isoform have different biological effects [27]. In response to chemotherapeutic drugs and, independently of p53, TAp63 is induced in cancer cells inducing cell-cycle arrest, apoptosis, senescence and DNA repair [29-31]. While, ΔNp63 isoforms are often overexpressed in tumors and it is associated with a poor prognosis [32]. Guo et al showed that, when p53 is deficient, TAp63 deficiency improved Ras-mediated oncogenesis in vivo and increased proliferation [30]. They also proved that senescence is induced by the gain of TAp63 in mouse embryonal fibroblasts [30]. Other studies discovered that p63 acts like a suppressor of metastasis. In epithelial tumors, TAp63 inhibits invasiveness and metastasis by monitoring the expression of important metastasis-inhibitor genes [33-35]. In our case p63 was upregulated following 17β-HSD1 inhibition, which can improve its metastasis suppressor action.

APOD expression was downregulated following 17β-HSD1 inhibition. ApoD is a glycoprotein identified in plasma with a size of 29 to 30 kDa [36, 37]. In advanced stages of prostate cancer, an increase in expression of apoD has been noted [38]. Moreover, the expression of apoD was higher in prostate carcinomas than in normal prostate [38]. ApoD is accumulated in the cyst fluid from women with breast gross cystic disease [39, 40]. It is also high in neurological diseases such as Alzheimer's disease, dementia, meningoencephalitis, stroke and motor neuron disease [41], Parkinson disease [42], multiple sclerosis [43] and schizophrenia [44, 45]. In breast cancer cells, exposure to interleukin-1α (IL-1α) increased apoD expression and decreased ZR-75-1 and T-47D cell proliferation [46], while treatment of ZR-75-1 cells with IL-6 decreased apoD secretion and cell proliferation [47]. Thus, the correlation between cell proliferation and apoD secretion is not observed in all situations [48]. ApoD plays an important role in the migration of smooth muscle cells (SMCs). Leung WC et al, showed that apoD is highly expressed in motile smooth muscle cells (SMCs) [49].

127

The stimulation of migration by wounding with platelet-derived growth factor (PDGF)-BB increased the level of APOD in ovine aortic smooth muscle cells (Ao SMCs). Also, ApoD works with PDGF-BB to mediate migration of human pulmonary artery SMCs (hPASMCs) [49]. The exact function of ApoD is still not totally clear [50]. ApoD could be a marker for steroid signalling [51] and a very interesting putative predictive and prognostic marker in breast cancer [50].

Q-RT-PCR was performed to confirm the results obtained in microarray analyses. Two genes identified by microarray analysis APOD and TP63 (Table 2) were tested. Treatment effects were validated. Figure 2 shows functional categories, such as biological processes and molecular function.

3.3. Networks and canonical pathways identified following 17β-HSD1 inhibition

Genes identified by microarray analysis were used to identify networks and canonical pathways modulated following 17β-HSD1 inhibition. Using Ingenuity Pathway analysis software (IPA, Ingenuity® Systems), five networks were identified; the first one is involved in lipid metabolism, molecular transport and small molecule biochemistry (Figure 3). The second network is involved in cellular movement, nervous system development and function, cellular growth and proliferation. The third is involved in cellular and organismal development and gene expression. The fourth is involved in cellular movement, cell signaling, vitamin and mineral metabolism, and the fifth one is involved in cell signaling, small molecule biochemistry and cellular movement. Five canonical pathways were identified namely, Wnt/β-catenin signaling (Figure 4), role of IL-17A in Psoriasis, bladder cancer signaling, cell cycle: G1/S checkpoint regulation and polyamine regulation in colon cancer.

4. Conclusion

128

Our results show that 17β-HSD1 inhibition modulates the expression profile of T47D cells and affects the expression of genes involved in several physiological functions such as APOD (involved in protein transport), TP63 (involved in cancer pathophysiology) and CD36 (involved in angiogenesis). The inhibition of 17β-HSD1 may promote an anti-angiogenic profile by increasing CD36 expression and improve p63 metastasis suppressor action.

129

List of abbreviations

17β-HSD1: 17beta-hydroxysteroid dehydrogenase type 1; E1: estrone; E2: estradiol; ER: estrogen receptor; ERE: estrogen-responsive element; FBS: fetal bovine serum; NADPH: nicotinamide adenine dinucleotide phosphate; RT-qPCR: reverse transcription quantitative real-time polymerase chain reaction; TLC: thin layer chromatography.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

SXL and MZ designed the study. MZ carried out the experimental studies. MZ and SXL prepared the manuscript. All authors read and approved the final manuscript.

Authors’ information

Laboratory of Molecular Endocrinology and Oncology, Centre Hospitalier Universitaire de Québec Research Center (CHUQ - CHUL) and Department of Molecular Medicine, Laval University, 2705 boulevard Laurier, Québec G1V 4G2, Canada.

Acknowledgements

We thank Dr. E-L. Calvo and Annick Ouellet for their assistance with the microarray analysis. We thank Mrs Nathalie Paquet for her help in quantitative real-time RT-PCR analysis. We thank Dr Donald Poirier for giving us the 17β-HSD type 1 inhibitor. This work was supported by the Canadian Institutes of Health Research, with grant MOP 97917 and 89851 to S.-X. Lin (PI).

130

References

1. Platet N, Cathiard AM, Gleizes M, Garcia M: Estrogens and their receptors in breast cancer progression: a dual role in cancer proliferation and invasion. Crit Rev Oncol Hematol 2004, 51(1):55-67. 2. Thomas DB: Do hormones cause breast cancer? Cancer 1984, 53(3 Suppl):595- 604. 3. Vihko R, Apter D: Endogenous steroids in the pathophysiology of breast cancer. Critical reviews in oncology/hematology 1989, 9(1):1-16. 4. Peltoketo H, Isomaa V, Maentausta O, Vihko R: Complete amino acid sequence of human placental 17 beta-hydroxysteroid dehydrogenase deduced from cDNA. FEBS letters 1988, 239(1):73-77. 5. Luu The V, Labrie C, Zhao HF, Couet J, Lachance Y, Simard J, Leblanc G, Cote J, Berube D, Gagne R et al: Characterization of cDNAs for human estradiol 17 beta- dehydrogenase and assignment of the gene to chromosome 17: evidence of two mRNA species with distinct 5'-termini in human placenta. Molecular endocrinology 1989, 3(8):1301-1309. 6. Gast MJ, Sims HF, Murdock GL, Gast PM, Strauss AW: Isolation and sequencing of a complementary deoxyribonucleic acid clone encoding human placental 17 beta-estradiol dehydrogenase: identification of the putative cofactor binding site. American journal of obstetrics and gynecology 1989, 161(6 Pt 1):1726-1731. 7. Lin SX, Yang F, Jin JZ, Breton R, Zhu DW, Luu-The V, Labrie F: Subunit identity of the dimeric 17 beta-hydroxysteroid dehydrogenase from human placenta. The Journal of biological chemistry 1992, 267(23):16182-16187. 8. Ghosh D, Pletnev VZ, Zhu DW, Wawrzak Z, Duax WL, Pangborn W, Labrie F, Lin SX: Structure of human estrogenic 17 beta-hydroxysteroid dehydrogenase at 2.20 A resolution. Structure 1995, 3(5):503-513. 9. Lin SX, Poirier D, Adamski J: A challenge for medicinal chemistry by the 17beta- hydroxysteroid dehydrogenase superfamily: an integrated biological function and inhibition study. Curr Top Med Chem 2013, 13(10):1164-1171. 10. Oduwole OO, Li Y, Isomaa VV, Mantyniemi A, Pulkka AE, Soini Y, Vihko PT: 17beta-hydroxysteroid dehydrogenase type 1 is an independent prognostic marker in breast cancer. Cancer research 2004, 64(20):7604-7609. 11. Theobald AJ: Management of advanced breast cancer with endocrine therapy: the role of the primary healthcare team. International journal of clinical practice 2000, 54(10):665-669. 12. Korach KS: Insights from the study of animals lacking functional estrogen receptor. Science 1994, 266(5190):1524-1527.

131

13. Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M: Effect of single and compound knockouts of estrogen receptors alpha (ERalpha) and beta (ERbeta) on mouse reproductive phenotypes. Development 2000, 127(19):4277- 4291. 14. McKenna NJ, Lanz RB, O'Malley BW: Nuclear receptor coregulators: cellular and molecular biology. Endocrine reviews 1999, 20(3):321-344. 15. Laplante Y, Cadot C, Fournier MA, 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(4):1849-1860. 16. Aka JA, Mazumdar M, Chen CQ, Poirier D, Lin SX: 17beta-hydroxysteroid dehydrogenase type 1 stimulates breast cancer by dihydrotestosterone inactivation in addition to estradiol production. Molecular endocrinology 2010, 24(4):832-845. 17. Aka JA, Zerradi M, Houle F, Huot J, Lin SX: 17beta-hydroxysteroid dehydrogenase type 1 modulates breast cancer protein profile and impacts cell migration. Breast cancer research : BCR 2012, 14(3):R92. 18. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP: Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 2003, 4(2):249-264. 19. Smyth GK: Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Statistical applications in genetics and molecular biology 2004, 3:Article3. 20. Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I: Controlling the false discovery rate in behavior genetics research. Behavioural brain research 2001, 125(1-2):279- 284. 21. Wettenhall JM, Simpson KM, Satterley K, Smyth GK: affylmGUI: a graphical user interface for linear modeling of single channel microarray data. Bioinformatics 2006, 22(7):897-899. 22. Luu-The V, Paquet N, Calvo E, Cumps J: Improved real-time RT-PCR method for high-throughput measurements using second derivative calculation and double correction. Biotechniques 2005, 38(2):287-293. 23. Warrington JA, Nair A, Mahadevappa M, Tsyganskaya M: Comparison of human adult and fetal expression and identification of 535 housekeeping/maintenance genes. Physiol Genomics 2000, 2(3):143-147. 24. Silverstein RL, Febbraio M: CD36, a scavenger receptor involved in immunity, metabolism, angiogenesis, and behavior. Science signaling 2009, 2(72):re3. 25. DeFilippis RA, Chang H, Dumont N, Rabban JT, Chen YY, Fontenay GV, Berman HK, Gauthier ML, Zhao J, Hu D et al: CD36 repression activates a multicellular stromal program shared by high mammographic density and tumor tissues. Cancer discovery 2012, 2(9):826-839. 26. Primo L, Ferrandi C, Roca C, Marchio S, di Blasio L, Alessio M, Bussolino F: Identification of CD36 molecular features required for its in vitro angiostatic activity. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2005, 19(12):1713-1715.

132

27. Vanbokhoven H, Melino G, Candi E, Declercq W: p63, a story of mice and men. The Journal of investigative dermatology 2011, 131(6):1196-1207. 28. Crum CP, McKeon FD: p63 in epithelial survival, germ cell surveillance, and neoplasia. Annual review of pathology 2010, 5:349-371. 29. Gressner O, Schilling T, Lorenz K, Schulze Schleithoff E, Koch A, Schulze- Bergkamen H, Lena AM, Candi E, Terrinoni A, Catani MV et al: TAp63alpha induces apoptosis by activating signaling via death receptors and mitochondria. The EMBO journal 2005, 24(13):2458-2471. 30. Guo X, Keyes WM, Papazoglu C, Zuber J, Li W, Lowe SW, Vogel H, Mills AA: TAp63 induces senescence and suppresses tumorigenesis in vivo. Nature cell biology 2009, 11(12):1451-1457. 31. Lin YL, Sengupta S, Gurdziel K, Bell GW, Jacks T, Flores ER: p63 and p73 transcriptionally regulate genes involved in DNA repair. PLoS genetics 2009, 5(10):e1000680. 32. Deyoung MP, Ellisen LW: p63 and p73 in human cancer: defining the network. Oncogene 2007, 26(36):5169-5183. 33. Adorno M, Cordenonsi M, Montagner M, Dupont S, Wong C, Hann B, Solari A, Bobisse S, Rondina MB, Guzzardo V et al: A Mutant-p53/Smad complex opposes p63 to empower TGFbeta-induced metastasis. Cell 2009, 137(1):87-98. 34. Muller PA, Caswell PT, Doyle B, Iwanicki MP, Tan EH, Karim S, Lukashchuk N, Gillespie DA, Ludwig RL, Gosselin P et al: Mutant p53 drives invasion by promoting integrin recycling. Cell 2009, 139(7):1327-1341. 35. Su X, Chakravarti D, Cho MS, Liu L, Gi YJ, Lin YL, Leung ML, El-Naggar A, Creighton CJ, Suraokar MB et al: TAp63 suppresses metastasis through coordinate regulation of Dicer and miRNAs. Nature 2010, 467(7318):986-990. 36. McConathy WJ, Alaupovic P: Isolation and partial characterization of apolipoprotein D: a new protein moiety of the human plasma lipoprotein system. FEBS letters 1973, 37(2):178-182. 37. Provost PR, Weech PK, Tremblay NM, Marcel YL, Rassart E: Molecular characterization and differential mRNA tissue distribution of rabbit apolipoprotein D. Journal of lipid research 1990, 31(11):2057-2065. 38. Aspinall JO, Bentel JM, Horsfall DJ, Haagensen DE, Marshall VR, Tilley WD: Differential expression of apolipoprotein-D and prostate specific antigen in benign and malignant prostate tissues. The Journal of urology 1995, 154(2 Pt 1):622-628. 39. Balbin M, Freije JM, Fueyo A, Sanchez LM, Lopez-Otin C: Apolipoprotein D is the major protein component in cyst fluid from women with human breast gross cystic disease. The Biochemical journal 1990, 271(3):803-807. 40. Diez-Itza I, Vizoso F, Merino AM, Sanchez LM, Tolivia J, Fernandez J, Ruibal A, Lopez-Otin C: Expression and prognostic significance of apolipoprotein D in breast cancer. The American journal of pathology 1994, 144(2):310-320. 41. Terrisse L, Poirier J, Bertrand P, Merched A, Visvikis S, Siest G, Milne R, Rassart E: Increased levels of apolipoprotein D in cerebrospinal fluid and hippocampus of Alzheimer's patients. Journal of neurochemistry 1998, 71(4):1643-1650.

133

42. Ordonez C, Navarro A, Perez C, Astudillo A, Martinez E, Tolivia J: Apolipoprotein D expression in substantia nigra of Parkinson disease. Histology and histopathology 2006, 21(4):361-366. 43. Reindl M, Knipping G, Wicher I, Dilitz E, Egg R, Deisenhammer F, Berger T: Increased intrathecal production of apolipoprotein D in multiple sclerosis. Journal of neuroimmunology 2001, 119(2):327-332. 44. Thomas EA, Dean B, Pavey G, Sutcliffe JG: Increased CNS levels of apolipoprotein D in schizophrenic and bipolar subjects: implications for the pathophysiology of psychiatric disorders. Proceedings of the National Academy of Sciences of the United States of America 2001, 98(7):4066-4071. 45. Mahadik SP, Khan MM, Evans DR, Parikh VV: Elevated plasma level of apolipoprotein D in schizophrenia and its treatment and outcome. Schizophrenia research 2002, 58(1):55-62. 46. Blais Y, Sugimoto K, Carriere MC, Haagensen DE, Labrie F, Simard J: Potent stimulatory effect of interleukin-1 alpha on apolipoprotein D and gross cystic disease fluid protein-15 expression in human breast-cancer cells. International journal of cancer Journal international du cancer 1994, 59(3):400-407. 47. Blais Y, Sugimoto K, Carriere MC, Haagensen DE, Labrie F, Simard J: Interleukin- 6 inhibits the potent stimulatory action of androgens, glucocorticoids and interleukin-1 alpha on apolipoprotein D and GCDFP-15 expression in human breast cancer cells. International journal of cancer Journal international du cancer 1995, 62(6):732-737. 48. Rassart E, Bedirian A, Do Carmo S, Guinard O, Sirois J, Terrisse L, Milne R: Apolipoprotein D. Biochimica et biophysica acta 2000, 1482(1-2):185-198. 49. Leung WC, Lawrie A, Demaries S, Massaeli H, Burry A, Yablonsky S, Sarjeant JM, Fera E, Rassart E, Pickering JG et al: Apolipoprotein D and platelet-derived growth factor-BB synergism mediates vascular smooth muscle cell migration. Circulation research 2004, 95(2):179-186. 50. Soiland H, Skaland I, Janssen EA, Gudlaugsson E, Korner H, Varhaug JE, Soreide JA, Baak JP: Comparison of apolipoprotein D determination methods in breast cancer. Anticancer Res 2008, 28(2B):1151-1160. 51. Soiland H, Soreide K, Janssen EA, Korner H, Baak JP, Soreide JA: Emerging concepts of apolipoprotein D with possible implications for breast cancer. Cellular oncology : the official journal of the International Society for Cellular Oncology 2007, 29(3):195-209.

134

FIGURE

Table 1. Differentially expressed genes (≥1.8-fold) in T47D cells after 17β-HSD1 inhibition by a specific inhibitor

Accession number Gene name Gene Symbol Regulation Fold- Function Change NM_001161429 RAN binding protein 3- RANBP3L down 2,91 Intracellular transport like ENST00000361453 NADH-ubiquinone ND2 down 2,40 Respiratory chain complex I oxidoreductase chain 2 NM_003521 histone cluster 1, H2bm HIST1H2BM down 2,31 DNA binding; responsible for the nucleosome structure of the chromosomal fiber in eukaryotes NR_002969 small nucleolar RNA, SNORA36A down 2,13 Biosynthesis of stable cellular H/ACA box 36A, small RNAs such as tRNAs, rRNAs, nucleolar RNA snRNAs, and snoRNAs is aided by covalent nucleotide modification after transcription. NM_002543 oxidized low density OLR1 down 1,96 Cell adhesion; proteolysis lipoprotein (lectin-like) receptor 1 NM_001647 apolipoprotein D APOD down 1,90 Transporter activity NM_001185 alpha-2-glycoprotein 1, AZGP1 down 1,86 ribonuclease activity; fatty acid zinc-binding binding NM_182511 cerebellin 2 precursor CBLN2 down 1,84 integral to membrane BX648100 C21orf15 chromosome C21orf15 down 1,84 21 open reading frame 15 AK172782 glycerol-3-phosphate GPAM down 1,82 secretion; defense response to acyltransferase, virus; fatty acid homeostasis mitochondrial NR_002327 small nucleolar RNA, SNORA10 down 1,81 Unknown H/ACA box 10 , small nucleolar RNA NM_020056 major HLA-DQA2 down 1,80 antigen processing and histocompatibility presentation of peptide or complex, class II, DQ polysaccharide antigen via MHC alpha 2 class II NM_194247 heterogeneous nuclear HNRNPA3 up 1,82 nucleotide binding ribonucleoprotein A3 NM_017770 elongation of very long ELOVL2 up 1,83 very long-chain fatty acid chain fatty acids metabolic process NM_001920 decorin , transcript DCN up 1,85 kidney development

135

variant A1 NM_003106 SRY (sex determining SOX2 up 1,88 negative regulation of region Y)-box 2 transcription from RNA polymerase II promoter NM_003722 tumor protein p63, TP63 up 1,89 p53 binding; response to tumor transcript variant 1 cell NM_015169 RRS1 ribosome RRS1 up 1,90 condensed nuclear chromosome biogenesis regulator homolog (S. cerevisiae) NM_032279 ATPase type 13A4 ATP13A4 up 1,94 nucleotide binding NM_001160133 potassium voltage-gated KCNQ5 up 1,94 ion transport channel, KQT-like subfamily, member 5, transcript variant 4 NM_018485 G protein-coupled GPR77 up 1,98 G-protein coupled receptor receptor 77 activity NM_001193582 neuronal cell adhesion NRCAM up 1,99 neuron migration molecule, transcript variant 4 NM_002272 keratin 4 KRT4 up 2,01 cytoskeleton organization NM_006061 cysteine-rich secretory CRISP3 up 2,04 defense response protein 3, transcript variant 1 NM_012302 latrophilin 2 LPHN2 up 2,05 G-protein coupled receptor activity NM_002964 S100 calcium binding S100A8 up 2,08 protein binding protein A8 NM_004274 A kinase (PRKA) AKAP6 up 2,10 Protein binding anchor protein 6 NM_020808 signal-induced SIPA1L2 up 2,11 regulation of small GTPase proliferation-associated mediated signal transduction 1 like 2 NM_014279 olfactomedin 1, OLFM1 up 2,11 protein binding transcript variant 1 NM_001001548 CD36 molecule CD36 up 2,13 positive regulation of tumor (thrombospondin necrosis factor production receptor), transcript variant 1 NM_014689 dedicator of cytokinesis DOCK10 up 2,14 Rho GTPase binding 10 NM_172362 potassium voltage-gated KCNH1 up 2,14 protein binding channel, subfamily H (eag-related), member 1, transcript variant 1 NM_003246 thrombospondin 1 THBS1 up 2,19 negative regulation of fibrinolysis NR_033410 makorin ring finger MKRN9P up 2,26 chromatin binding protein 1 NM_000421 keratin 10 KRT10 up 2,29 structural molecule activity NM_031950 fibroblast growth factor FGFBP2 up 2,33 growth factor binding binding protein 2 NM_018837 sulfatase 2, transcript SULF2 up 2,39 hydrolase activity variant 1 NM_007231 solute carrier family 6 SLC6A14 up 2,62 amino acid transport (amino acid transporter),

136

member 14 NM_020796 sema domain, SEMA6A up 2,66 apoptosis; cytoskeleton transmembrane domain organization (TM), and cytoplasmic domain, (semaphorin) 6A NM_000165 gap junction protein, GJA1 up 2,71 protein binding alpha 1, 43kDa NM_000499 cytochrome P450, CYP1A1 up 2,87 steroid hydroxylase activity; family 1, subfamily A, response to hypoxia; cell polypeptide 1 proliferation NM_000609 chemokine (C-X-C CXCL12 up 3,73 cell adhesion motif) ligand 12, transcript variant 2 NM_005097 leucine-rich, glioma LGI1 up 3,83 protein binding inactivated 1 NM_004190 lipase, gastric, transcript LIPF up 4,93 lipid catabolic process variant 2 NM_001756 serpin peptidase SERPINA6 up 5,01 serine-type endopeptidase inhibitor, clade A (alpha- inhibitor activity 1 antiproteinase, antitrypsin), member 6

137

Table 2. The comparison between differentially expressed genes obtained by microarray and q-RT-PCR analysis in T47D cells after 17β-HSD1 by specific inhibitor

Gene symbol Microarray q-RT-PCR Fold change of up-regulated gene after 17β-HSD1 inhibition TP63 1,89 5 Fold change of down-regulated gene after 17β-HSD1 inhibition APOD 1,9 1,8

138

A B

Figure 1. 17β-HSD1 inhibition in T47D cells by a specific inhibitor. Activity assays were carried out in T47D cells using 14C-E1 as substrate. The assay was carried out at 0, 10, 30 and 60 minutes and represent the conversion of 14C-E1 into 14C-E2: A, T47D control cells incubated without inhibitor. B, T47D cells incubated with 5 µM 17β-HSD1 inhibitor. The experiment was done in duplicates.

139

Figure 2 . Overpresentation of functional categories. The genes are grouped into cellular components, biological processes and molecular function.

140

Figure 3. Network identified following 17β-HSD1 inhibition in T47D cells. The Ingenuity Pathway analysis (IPA) software was used to identify networks by using genes identified by microarray analysis. The green color indicates down regulated genes and the red color indicates up regulated genes. Lines indicate known relationships. Dashed lines indicate proposed relationships.

141

Figure 4. Canonical pathways identified following 17β-HSD1 inhibition in T47D cells. Genes identified by microarray analysis were used to identify the canonical pathways by using the Ingenuity Pathway analysis (IPA) software. The green color indicates down regulated genes and the red color indicates up regulated genes.

142

Additional file 1, Table S1. Primers used for quantitative real-time RT-PCR

Gene size Primer sequence 5'→3' Description GenBank Symbol (pb) S/AS

Homo sapiens tumor protein p63 CGTCAGAACACACATG TP63 (TP63), région commune aux 6 NM_003722 176 GTATCC/TGCCTGTACG transcrits TTTCAATTGTGTGC ATGGAACTGTGAATCA Homo sapiens apolipoprotein D AATCGAAGGTG/GTTG APOD NM_001647 168 (APOD) GATGATGCAGGTACAG GA Homo sapiens ATP synthase, H+ ATTGAAGGTCGCTATG Atp5o transporting, mitochondrial F1 NM_001697 267 CCACAG/AACGACTCCT complex, O subunit TGGGTATTGCTTAA AGTTCTGTGGCCATCT Homo sapiens hypoxanthine Hprt1 NM_000194 157 GCTTAGTAG/AAACAA phosphoribosyltransferase 1 CAATCCGCCCAAAGG GATGTCCCCTGTCCCA Homo sapiens glucose-6- G6PD NM_000402 121 CCAACTCTG/GCAGGG phosphate dehydrogenase CATTGAGGTTGGGAG ACGGACCAGAGCGAA 18S Homo sapiens 18S ribosomal RNA NR_003286 226 AGCATT/TCCGTCAATT CCTTTAAGTTTCAGCT Homo sapiens 3-beta- GAAGGGCAGAGGTGG hydroxysteroid AACTAGAA/AACAAAG ADNg dehydrogenase/delta-5-delta-4- M38180 260 ACCAAAGACCAGTGAG isomerase (3-beta-HSD) gene A (intron)

143

144

Chapitre 4: Regulation of breast cancer progression by phosphorylation of the tumor suppressor tropomyosin-1 alpha

145

146

Regulation of breast cancer progression by phosphorylation of the tumor suppressor tropomyosin-1 alpha

Mouna Zerradi, François Houle, Jacques Huot*

Le Centre de recherche du CHU de Québec and Le Centre de recherche en cancérologie de l’Université Laval, l’Hôtel-Dieu de Québec, 9 rue McMahon, Québec G1R 2J6 Canada [email protected] [email protected] [email protected]

Running title: Phosphorylation of tropomyosin 1 in breast cancer

To whom correspondence should be addressed at: [email protected]

Cet article est publié dans le Journal of Cancer Therapy, 6, 783-792.

147

Abstract

Background: Tropomyosin 1 alpha chain (Tm1) is an actin-binding protein that regulates the endothelial cell response to oxidative stress following its phosphorylation at Serine 283 (S283). Tm1 is also a major tumor suppressor in breast cancer. In the present study, we investigated the role of phosphorylation of Tm1 in regulating its tumor suppressor properties. Methods: MDA MB231 breast cancer cells stably overexpressing wild type form of Tm1 or Tm1 mutants (S283A and S283E) were generated. Proliferation and cell viability were assayed by means of the enzymatic cleavage of the tetrazolium salt WST-1 to formazan dye by cellular mitochondrial dehydrogenases. Adhesion assays were performed at various periods of time on cells grown on plastic. Cell migration was evaluated by using the wound- healing assay and by measuring transendothelial migration of cancer cells. Malignant transformation in vitro was determined by using the anchorage-independent growth assay on soft agar. Results: We found that cells expressing the phosphomimetic form of Tm1 S283E/Tm1 are characterized by an increased adhesion to the substratum. Moreover, the migration of MDA MB231/S283E/Tm1 cells in a wound closure assay is reduced compared to parental cells or those expressing the non-phosphorylatable form of Tm1 (S283A). Similarly, the transendothelial migration of MDA MB231/S283E/Tm1 cells is also reduced as compared to the other cell lines. Moreover, we found that the cells expressing the S283A mutants form more colonies in soft agar that those expressing the S283E mutants. Conclusion: Phosphorylation of Tm1 at Ser283 contributes to its anti-tumor properties, and this effect results mainly from an increase in cell adhesion associated with a decrease in their migratory and invasive potentials. Key words: Tropomyosin 1, phosphorylation, F-actin, cell migration, invasion.

148

Résumé en français

Introduction: La tropomyosine chaîne alpha 1 (Tm1) est une protéine qui se lie au filament d’actine tout en régulant la réponse des cellules endothéliales au stress oxydatif suite à sa phosphorylation sur la sérine 283 (S283). Tm1 est également un suppresseur de tumeur majeur dans le cancer du sein. Dans la présente étude, nous avons étudié le rôle de la phosphorylation de la Tm1 dans la régulation de ses propriétés de suppresseur de tumeur. Méthodes: Des cellules MDA MB231 du cancer du sein surexprimant de manière stable la forme sauvage ou mutée de la Tm1 (S283A et S283E) ont été générées. La prolifération et la viabilité des cellules ont été analysées au moyen du test MTT qui mesure la transformation des sels de tétrazolium du substrat en cristaux insolubles de formazan par la déshydrogénase mitochondriale. Les tests d'adhésion ont été effectués à différentes périodes de temps sur des cellules cultivées sur plastique. La migration cellulaire a été évaluée en utilisant le test de la cicatrisation et en mesurant la migration trans-endothéliale des cellules cancéreuses. La transformation maligne in vitro a été déterminée en utilisant le test de croissance indépendante de l'ancrage sur agar mou. Résultats: Nous avons trouvé que les cellules exprimant la forme phosphomimétique (pseudo-phosphorylée) de la Tm1 S283E/Tm1 sont caractérisées par une adhérence accrue au substrat. En outre, la migration des cellules MDA MB231/S283E/Tm1 dans un essai de cicatrisation est réduite par rapport aux cellules parentales ou ceux exprimant la forme non- phosphorylable de la Tm1 (S283A). De même, la migration transendothéliale des cellules MDA MB231/S283E/Tm1 est également réduite par rapport aux autres lignées cellulaires. En outre, nous avons constaté que les cellules exprimant les mutants S283A forment plus de colonies en agar mou que ceux exprimant les mutants S283E. Conclusion: La phosphorylation de la Tm1 sur la Ser283 contribue à ses propriétés anti- tumorales, et cet effet résulte principalement d'une augmentation de l'adhérence des cellules cancéreuses associée à une diminution de leurs potentiels migratoire et invasif. Mots clés: Tropomyosine 1, phosphorylation, F-actin, migration cellulaire, invasion.

149

1. Introduction

Breast cancer is the most common malignancy among women in Europe and North America. Despite its high incidence, early detection and modern treatments have significantly increased patient survival. Nevertheless, breast cancer remains the second leading cause of cancer death in women (Hooks 2010). Breast cancer initiation and progression are characterized by multiple genetic and epigenetics alterations that activate dominant-acting oncogenes and disrupt the functions of specific tumor suppressor genes (Lee and Muller 2010). Genome analyses indicate that there are only a few genes that are commonly mutated in breast cancer. These genes include the oncogens ErbB2, PI3KCA, MYC, and CCND1 as well as the tumor suppressors BRCA1/2 and p53 [2]. PTEN is another important tumor suppressor that counteracts PI3K-mediated survival signaling from activated oncogenes such as ErbB2 (Lee and Muller 2010). On the other hand, many genes are less frequently mutated, which provides an explanation for the high heterogeneity of breast cancer.

The genes encoding for tropomyosins (Tm) are also frequently deregulated in breast cancer, and significant changes in the expression of different tropomyosin isoforms accompany cell transformation. Mammals have four genes (α, β, γ, δ) that code for tropomyosins generating more than 40 alternatively spliced isoforms (Perry 2001). The Tm genes have a similar organization. In humans, the α, β, γ, δ genes are known as TPM1 (located on the chromosome 15q22) (Eyre et al. 1995), TPM2 (located on chromosome 9p13) (Hunt et al. 1995), TPM3 (located on chromosome 1q22) (Wilton et al. 1995) and TPM4 (located on chromosome 19p13) (Wilton et al. 1996), respectively. Intriguinly, a lot of confusion exists in the nomenclature of the various forms of tropomyosins. In particular, the gene product of TPM2 is Tpm2, TM2 or TM2β and has previously been ascribed as coding for Tm1 in fibroblasts (Geeves et al. 2014; Gunning et al. 2008). On the other hand, TPM1 encodes for tropomyosin 1 alpha chain that is also frequently referred as tropomyosin-1 or Tm1 (Houle et al. 2007; Houle et al. 2003; Simoneau et al. 2012) (www.uniprot.org/uniprot/P09493). For the sake of understanding and to keep in line with

150

previous literature, we used herein TPM2/Tm1 to refer as the gene product of TPM2 and Tm1 as the gene product of TPM1.

The tropomyosins are members of a family of actin-binding proteins that play an important role in regulating the function of actin filament in both muscle and non-muscle cells (Gunning 2008; Gunning et al. 2008). In striated muscle, tropomyosin regulates the interaction between myosin and actin by forming a troponin complex that mediates muscle contraction in response to calcium (Cooper 2002). In nonmuscle cells, tropomyosins play a role in the formation and stabilization of stress fibers by facilitating actomyosin interactions and protecting actin against the action of cofilin and gelsolin (Cooper 2002). The mechanims by which TPM2/Tm1 exerts its tumor suppressor functions in breast cancer seem to be associated with cytoskeletal remodeling. As mentioned above, the TPM2/Tm1 re-expression in MCF7 null cells suppress the malignant phenotype, which alters the interaction of the E- cadherin-catenin complex with the cytoskeleton, indicating that TPM2/Tm1-induced cytoskeleton dysregulation could play a significant role in suppression of the malignant phenotype (Mahadev et al. 2002). Moreover, the loss of TPM2/Tm1 in cancer cells is associated with disorganization of actin filaments and thereby impaired cell migration (Hendricks and Weintraub 1981). Tropomyosins are subject to two types of posttranslational modification namely,

NH2-terminal acetylation and phosphorylation. The acetylation of Tm is essential for normal functions of muscle α-Tm and it is required for the strong binding of most Tms to actin (Pittenger et al. 1995; Urbancikova and Hitchcock-DeGregori 1994 ). On the other hand, Tm1 phosphorylation modulates its interaction with other proteins required for actin polymerization and stabilization, including caldesmon and HSP27 (Mak et al. 1978; Somara et al. 2005 ). Along these lines, the phosphorylation of cytoskeletal non-muscle Tm2 at Ser- 61 by phosphoinositide 3-kinase occurs at a late stage of endocytosis, which is consistent with a role in actin polymerization (Naga Prasad et al. 2005). In endothelial cells, we found that Tm1 alpha chain (herein named Tm1) is phosphorylated at Ser283 in response to activation of the ERK/DAP kinase axis by oxidative stress. In turn, this triggers the formation of stress fibers and confers resistance to oxidative stress-mediated membrane blebbing (Houle and Huot 2006; Houle et al. 2007; Houle et al. 2003; Simoneau et al. 2012).

151

In the present study, we investigated the role of phosphorylation in regulating the tumor suppression properties of Tm1 in breast cancer.

152

2. Methods

2.1. Reagents and antibodies

Anti-tropomyosin (clone TM311) monoclonal mouse antibody and anti-α-actin antibody produced in rabbit were purchased from Sigma-Aldrich (Oakville, On, Canada). The anti- mouse and anti-rabbit IgG-horseradish peroxidase were purchased from The Jackson Laboratory (Bar Harbor, ME, USA).

2.2. Cells

MDA MB231 cells were obtained from the American Type Culture Collection (ATCC). MDA MB231 cells were maintained in Dulbecco's Modified Eagle Medium (DMED) containing 10% fetal bovine serum (FBS). Cultures were incubated at 37°C in a humidified atmosphere containing 95% air and 5% CO2. MDA MB-231 were transfected with vectors expressing wild-type form of Tm1 (Tm1wt), a non phosphorylatable (Tm1 S283A: MDA TM1 A) or a phosphomimetic (Tm1S283E: MDATM1 E) mutated form of Tm1 using the calcium chloride precipitation technique to generate cancer cell lines stably expressing tropomyosin-1. The cells were selected by incubation in medium containing G418 at 400 µg/ml. A pool of resistant cells was obtained and frozen or sub-cultured for the experiments of this project.

2.3. Plasmids

TPM1 (RecName :Full= Tropomyosin alpha-1 chain) cDNA was cloned as previously described (Houle et al. J Cell Sci 2007) by PCR amplification from IMAGE clone 562592 (ATCC) into pIRES-hrGFP2a (Stratagene, La Jolla, CA) vectors using the following primers: 5’ TAGAATTCTATGGACGCCATCAAGAAGAAGATGCAGATGC-3’ and 5’-CCTGCTCGAGTATATGGAAGTCATATCGTTGAGAGC- 3’. The tropomyosin-1

153

Ser283Ala was generated by PCR site-directed mutagenesis on pIRES-hrGFP2a- tropomyosin-1 construct using the primers 5’- ATGACTGCTATATAACTCGAGTACCCATATGACG- 3’ and 5’- TTATATAGCAGTCATATCGTTGAGAGCGTGG- 3’. The tropomyosin-1 S283E mutant was generated similarly using the primers 5’ CGATATGACTGAAATAATACTCGAGTACCCATATG-3’ and 5’- CGAGTTATATTTCAGTCATATCGTTGAGAGCG-3’ (NP_000357) (Houle et al. 2007).

2.4. Western blot

Total proteins from cells grown in 60 mm Petri dishes were extracted with 100 µl loading buffer. Proteins were quantified using the Lowry method and equal amounts of proteins from each sample were separated on 10% SDS-PAGE, and the gels were transferred onto nitrocellulose membranes for Western blotting. The membranes were blocked with 5% non- fat milk in PBS-Tween 20 for one hour at room temperature. Membranes were then incubated over night at 4°C in blocking buffer containing 1 :1000 dilution of the primary anti-tropomyosin (clone TM311) monoclonal mouse antibodies (Sigma-Aldrich, Oakville, On, Canada). For loading control, 1 :3000 dilution of anti-α-actin antibody produced in rabbit (Sigma-Aldrich, Oakville, On, Canada) was used. Afterward, membranes were washed and incubated for one hour with the respective horseradish peroxidase-conjugated secondary antibody (anti-IgG antibody) diluted 5000 times. Protein signals were visualised with chemiluminescence Reagent (PerkinElmer, Waltham, MA, USA).

2.5. Cell Proliferation and Viability Assay

Cell viability has been determined by using the Quick Cell Proliferation Assay Kit from BioVision. This cell proliferation and viability assay is based on the enzymatic cleavage of the tetrazolium salt WST-1 into formazan dye by cellular mitochondrial dehydrogenases that are active in viable cells only. The formazan dye produced by viable cells is quantified by

154

measuring the absorbance of the dye at 440 nm. The exact procedure that was followed was as described in the BioVision protocol sheet.

2.6. Cell adhesion assays

MDA MB231-Tm1 cells (3x104 cells/well) were seeded on plastic in 96 well plates containing DMEM 10% FBS for 1 hour at 37°C. Then, the cells were washed twice with fresh DMEM 10% FBS and fixed with 100 µl formaldehyde 3.7% for 30 minutes at room temperature. Thereafter, the cells were washed once with Phosphate Buffered Saline (PBS) and stained with 50 µl of crystal violet for 15 min at room temperature. Then, they were washed twice with PBS and lysed with 100 µl of SDS 1%. Absorbance of each well was measured at 550 nm.

2.7. Anchorage-independent growth assays

Liquefied 2% agarose was mixed with an equal volume of DMEM 2X growth medium lacking serum and supplemented with G418 (400 µg/ml). One milliliter of the mixture was layered on 35 mm Petri dishes to create a 1% agarose base. Liquefied 0.6% agarose was mixed with an equal volume of DMEM 2X medium, and 10 ml of this solution was mixed with 1 ml of growth medium containing 104 cells in 0.27% agarose. One ml of this cell suspension was layered on top of the 1% agarose base, and 1 ml of DMEM medium containing 10% FBS was added. The cells were incubated for 14 days, after which colonies present in representative fields were photographed using phase-contrast microscopy (20X and 40X on a Nikon-TE300 inverted microscope) (Desprez et al. 1998).

155

2.8. Cell migration assays

Cell migration was evaluated by using two methods:

2.8.1. Wound healing assay

MDA MB231, MDA MB231/Tm1wt, MDA MB231/Tm1 S283A and MDA MB231/Tm1 S283E were cultivated in complete growth medium. Cells were seeded at high density in 35 mm Petri dishes in complete growth medium. One day later, a straight scratch was done using a p200 micropipette tip. Thereafter, cells were washed three times with fresh DMEM and incubated in the same medium. The movements of cells in the scratched area were monitored by capturing images every 15 minutes for a total duration of 24 h using a 10X objective lens of a phase contrast microscope Nikon- TE2000. Then, migrated cells were counted manually.

2.8.2. Transendothelial cell migration assays

Cell migration was assayed using a modified Boyden chamber assay. HUVEC (150x103) were grown to confluence (48 h) on an 8.0m pore size gelatinized polycarbonate membranes (Corning, NY, USA) separating the two compartments of a 6.5mm Transwell. Tumor cells in suspension were stained with 500nM calcein-AM (Sigma-Aldrich, Oakville, On, Canada) for 30 min at 37°C, and then added in migration buffer (199 medium, 10 mM

HEPES pH 7.4, 1 mM MgCl2, and 0.5 % bovine serum albumin) on the endothelial monolayer. The chambers were incubated for 6h at 37°C in a 5% CO2 atmosphere. The cells in the upper part of the chamber were removed with a cotton swab. Then, fluorescent tumor cells that crossed the membrane were counted in five fields using a 20x lens on a Nikon- TE300 inverted microscope (Gout et al. 2006).

156

2.9. Statistics

Values are expressed as mean ± SD. Student t tests were used for comparaison between two means. A p value < 0.05 was considered as statistically significant.

157

3. Results

3.1. Adhesion of MDA MB231 cells depends on phosphorylation of Tm1 at Ser 283

Different isoforms of tropomyosins have been shown to exert tumor suppressor properties in breast cancer cells (Gunning et al. 2008). Along these lines, parental MDA MB231 cells are highly invasive and they do not express Tm1 (Figure 1) and (Yu et al. 1995). Accordingly, they proliferate faster than pooled clonal MDA MB231 cell lines expressing a wt form of Tm1 (Figure 2). However, the mechanisms underlying the tumor suppressors functions of Tm1 are still unclear. We reported previously that Tm1 is phosphorylated at S283 in endothelial cells exposed to oxidative stress, which contributes to maintain the integrity of the endothelium (Houle and Huot 2006; Houle et al. 2007; Houle et al. 2003). Here, we hypothesize that phosphorylation of Tm1 at S283 contributes to regulate its tumor suppressor functions. To this end, we generated clonal MDA MB231 breast cancer cells that stably express a non-phosphorylatable (S283A) or phosphomimetic (S283E) mutated form of the protein (Figure 1). Thereafter, we pooled the clonal cell lines and used them to investigate the role of Tm1 phosphorylation in modulating its function of tumor suppressor.

We first verified whether the phosphorylation of Tm1 regulates cellular adhesion by means of adhesion assays performed directly on plastic dishes. Results showed that the cells that express the S283E/Tm1 phosphomimetic mutant were those that adhere more strongly to the substratum either after 1 h and 2 h. Conversely, the parental cells and the cells that express the non-phosphorylatable mutant S283A/Tm1 were those that adhere the less (Figure 3). Overall, these findings suggest that phosphorylation of Tm1 at S283 increased the adhesion of MDA MB231 cells. Interestingly, consistent with the fact that MDA MB231/S283E/Tm1 have a higher adhesive ability than the two other cell lines, we previously reported that these cells express thick stress fibers whereas parental MDA MB231 cells as well as MDA MB231/S283A/Tm1 cells are devoid of stress fibers (Houle et al.

158

2007). Indeed, stress fibers are well recognized to be associated with increased cellular adhesion (Kirchenbuchler et al. ; O'Neill 2009).

3.2. Phosphorylation of Tm1 at S283 reduced cell migration

Increased cell migration is an inherent feature that characterized invasiveness during metastatic dissemination (Geho et al. 2005). We thus evaluated next whether phosphorylation of Tm1 at S283 regulates cell migration by means of a wound-healing assay applied to the different MDA MB231 cell lines. We found that the migration potential of MDA MB231/S283E/Tm1 cells was reduced as compared to parental cells that do not express Tm1 or to the MDA MB 231/S283A/Tm1 cells. This is evidenced by the larger number of cells from the latter two cell lines that invade the scratch (Figures 4A, 4B). These findings suggest that phosphorylation of Tm1 at S283 reduced the invasive properties of metastatic MDA MB231 cells. Incidentally, these results are in line with the fact that Tm1 phosphorylation at S283 has conferred increased adhesion of the cells to the substratum rendering them less motile.

Transendothelial migration and extravasation of circulating cancer cells are key events of metastatic dissemination (Gout and Huot 2009; Gout et al. 2008). Here we found that the transendothelial migration of MDA MB231/S283E/Tm1 cells was reduced in comparison to the other cell lines (Figure 5). This result indicates that the phosphorylation of Tm1 impairs the migration of MDA MB231 through endothelial layer.

Overall, this series of results indicate that phosphorylation of Tm1 at S283 reduces the migration of MDA MB231 cells and thereby their metastatic potential, which is consistent with the fact that phoshorylation of Tm1 regulates its functions of tumor suppressor.

159

3.3. Phosphorylation of Tm1 at S283 reduces the formation of MDA MB231 colonies on soft agar

The anchorage-independent growth assay for colony formation in soft agar, is a stringent assay for detecting malignant transformation of cells in vitro and is considered as a hallmark of malignant transformation (Mori et al. 2009). We thus used this assay as an additional end-point to evaluate the role of Tm1 phosphorylation as a regulatory mechanism underling the tumor suppressor properties of Tm1 in breast cancer. As shown in figure 6, the expression of the S283E mutant reduced by 70% the number of colonies formed by the parental MDA MB231 cells in soft agar. In contrast, the formation of colonies by cells expressing Ser283A was twice higher than those expressing the S283E mutant. The results indicate that phosphorylation of Tm1 at Ser283 reduces the anchorage-independent growth of MDA MB231 cells, which is consistent with the fact that phosphorylation of Tm1 contributes to its tumor suppressor properties.

160

4. Discussion

High-molecular weight tropomyosins such as Tm1 play major roles as tumor suppressor in breast cancer (Gunning 2008 ; Shah et al. 1998). However, the mechanisms underlying the tumor suppressor properties of Tm1 are ill defined. Here we provided the first evidence showing that phoshorylation of Tm1 is a major mechanism that regulates its tumor suppressor functions. This is supported first by the observation showing that MDA MB231 cells that express the phosphomimetic mutant S283E adhere more strongly to the substratum than the parental cells and those that express the non-phosphorylatable mutant S283A. Indeed, this finding is consistent with the fact that MDA MB231/S283E/Tm1 are less motile and invasive given than are sticked to the substratum, an effect explained by the fact that MDA MB231/S283E/Tm1 cells contains more stress fibers (Houle et al. 2007). Along these lines, a second argument supporting the role of phosphorylation of Tm1 at Ser283 in mediating its tumor suppressor functions is that MDA MB231/S283E/Tm1 are less motile than the parental cells and the cells that express the non-phosphorylatable mutant of Tm1 in the wound closure assay. Indeed, it is well known that increased invasive properties are associated with increased cell migration (Geho et al. 2005). Thirdly, MDA MB231/S283E/Tm1 cells have a lower capacity to transmigrate across an endothelial layer than the parental cells and the cells that express the non-phosphorylatble mutant of Tm1. This observation is a strong argument favoring the point that phosphorylation of Tm1 at S283 is involved in conferring tumor suppression and antimetastatic potential to Tm1 given that TEM (transendothelial migration) is a well-known prerequesite to metastasis (Gout et al. 2008). We previously reported that TEM of colon cancer cells require interactions between adhesion receptors present on endothelial cells and their counter-receptors on cancer cells. We showed that E-selectin is a typical endothelial adhesion receptor that is induced by inflammatory cytokines and that binds to Death Receptor 3 expressed by colon cancer cells to enable their transendothelial migration and extravasation (Gout et al. 2006; Gout et al. 2008; Tremblay et al. 2008). However, it does not seem that the TEM of MDA MB231 cells do require the expression of E-selectin given that they cross the endothelial layer even when endothelial cells do not express E-selectin. Hence, in the present study, our results suggest that Tm1-dependent TEM of MDA MB231 cells is importantly dependent on its

161 phosphorylation at Ser283. Finally, the fact that cells expressing the phosphomimetic mutant of Tm1 form less colonies in soft agar is a major argument supporting that Tm1 phosphoryation at Ser 283 confers tumor suppressor functions to Tm1, given that this test is well accepted as reflecting malignancy in vitro (Mori et al. 2009). Nevertheless, the finding that the cells expressing S283A mutant also form less colonies that the parental cells, but more than the S283E, cells indicate that other factors than just phosphorylation are involved in tumor suppression by Tm1.

Overall our results constitute the first evidence that phosphorylation of Tm1 regulates its tumor suppressor functions in breast cancer cells.

162

5. Conclusion

In this study, we have shown that MDA-MB231 cells stably expresssing a pseudophosphorylated form of Tm1 at Ser283 (S283E) adhere more firmly to the substratum and are less motile that the parental cells and cells that express a non-phosphorylatable form of the protein (S283A). Moreover, the Tm1 S283E expressing MDA-231 cells form less colonies in soft agar. We thus conclude that phosphoryation of Tm1 at Ser283 is a major mechanism that contributes to the tumor and metastatic suppressor properties of Tm1.

163

List of abbreviations

Tm1: Tropomyosin 1 alpha chain; Tm1 S283A: non phosphorylatable mutated form of Tm1; Tm1S283E: phosphomimetic mutated form of Tm1; DMED: Dulbecco's Modified Eagle Medium; FBS: fetal bovine serum; PBS: phosphate-buffered saline; PBS-T: PBS-tween 20; SDS-PAGE: sodium dodecyl sulphate-polyacrylamide gel electrophoresis; WT: wild type.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

JH and MZ designed the study. MZ and FH carried out the experimental studies. MZ, JH and FH prepared the manuscript. All authors read and approved the final manuscript.

Authors' information

Cancer Research Center of Laval University and Centre Hospitalier Universitaire de Québec Research Center (CHUQ - L’Hôtel-Dieu de Québec), 9 rue McMahon, Québec G1R 2J6, Canada.

Acknowledgments

This work was supported by the Canadian Institutes of Health Research.

164

References

Cooper, J. 2002. Actin dynamics: tropomyosin provides stability. Curr Biol 12(15): R523. Desprez, P.Y., Lin, C.Q., Thomasset, N., Sympson, C.J., Bissell, M.J., and Campisi, J. 1998. A novel pathway for mammary epithelial cell invasion induced by the helix-loop-helix protein Id-1. Mol Cell Biol 18(8): 4577-4588. Eyre, H., Akkari, P.A., Wilton, S.D., Callen, D.C., Baker, E., and Laing, N.G. 1995. Assignment of the human skeletal muscle alpha-tropomyosin gene (TPM1) to band 15q22 by fluorescence in situ hybridization. Cytogenet Cell Genet 69(1-2): 15-17. Geeves, M.A., Hitchcock-DeGregori, S.E., and Gunning, P.W. 2014. A systematic nomenclature for mammalian tropomyosin isoforms. Journal of muscle research and cell motility. doi: 10.1007/s10974-014-9389-6. Geho, D.H., Bandle, R.W., Clair, T., and Liotta, L.A. 2005. Physiological mechanisms of tumor-cell invasion and migration. Physiology (Bethesda) 20: 194-200. doi: 20/3/194 [pii] 10.1152/physiol.00009.2005. Gout, S., and Huot, J. 2009. Role of cancer microenvironmentin metastasis: Focus on colon cancer. Cancer Microenvironment 1(1): 69-83. doi: 10.10007/s12307-008-0007-2. Gout, S., Morin, C., Houle, F., and Huot, J. 2006. Death receptor-3, a new E-Selectin counter-receptor that confers migration and survival advantages to colon carcinoma cells by triggering p38 and ERK MAPK activation. Cancer Res 66(18): 9117-9124. Gout, S., Tremblay, P.L., and Huot, J. 2008. Selectins and selectin ligands in extravasation of cancer cells and organ selectivity of metastasis. Clin Exp Metastasis 25(4): 335-344. doi: 10.1007/s10585-007-9096-4. Gunning, P. 2008. Emerging issues for tropomyosin structure, regulation, function and pathology. Adv Exp Med Biol 644: 293-298. Gunning, P., O'Neill, G., and Hardeman, E. 2008. Tropomyosin-based regulation of the actin cytoskeleton in time and space. Physiol Rev 88(1): 1-35. doi: 88/1/1 [pii] 10.1152/physrev.00001.2007. Hendricks, M., and Weintraub, H. 1981. Tropomyosin is decreased in transformed cells. Proc Natl Acad Sci U S A 78(9): 5633-5637. Hooks, M.A. 2010. Breast cancer: risk assessment and prevention. Southern medical journal 103(4): 333-338. doi: 10.1097/SMJ.0b013e3181d39038. Houle, F., and Huot, J. 2006. Dysregulation of the endothelial cellular response to oxidative stress in cancer. Molecular carcinogenesis 45(6): 362-367. doi: 10.1002/mc.20218. Houle, F., Poirier, A., Dumaresq, J., and Huot, J. 2007. DAP kinase mediates the phosphorylation of tropomyosin-1 downstream of the ERK pathway, which regulates the formation of stress fibers in response to oxidative stress. J Cell Sci 120(Pt 20): 3666-3677. Houle, F., Rousseau, S., Morrice, N., Luc, M., Mongrain, S., Turner, C.E., Tanaka, S., Moreau, P., and Huot, J. 2003. Extracellular signal-regulated kinase mediates phosphorylation of tropomyosin-1 to promote cytoskeleton remodeling in response to oxidative stress: impact on membrane blebbing. Molecular biology of the cell 14(4): 1418- 1432. doi: 10.1091/mbc.E02-04-0235. Hunt, C.C., Eyre, H.J., Akkari, P.A., Meredith, C., Dorosz, S.M., Wilton, S.D., Callen, D.F., Laing, N.G., and Baker, E. 1995. Assignment of the human beta tropomyosin gene (TPM2) to band 9p13 by fluorescence in situ hybridisation. Cytogenet Cell Genet 71(1): 94-95.

165

Kirchenbuchler, D., Born, S., Kirchgessner, N., Houben, S., Hoffmann, B., and Merkel, R. Substrate, focal adhesions, and actin filaments: a mechanical unit with a weak spot for mechanosensitive proteins. J Phys Condens Matter 22(19): 194109. doi: S0953- 8984(10)31742-5 [pii] 10.1088/0953-8984/22/19/194109. Lee, E.Y., and Muller, W.J. 2010. Oncogenes and tumor suppressor genes. Cold Spring Harbor perspectives in biology 2(10): a003236. doi: 10.1101/cshperspect.a003236. Mahadev, K., Raval, G., Bharadwaj, S., Willingham, M.C., Lange, E.M., Vonderhaar, B., Salomon, D., and Prasad, G.L. 2002. Suppression of the transformed phenotype of breast cancer by tropomyosin-1. Exp Cell Res 279(1): 40-51. Mak, A., Smillie, L.B., and Barany, M. 1978. Specific phosphorylation at serine-283 of alpha tropomyosin from frog skeletal and rabbit skeletal and cardiac muscle. Proc Natl Acad Sci U S A 75(8): 3588-3592. Mori, S., Chang, J.T., Andrechek, E.R., Matsumura, N., Baba, T., Yao, G., Kim, J.W., Gatza, M., Murphy, S., and Nevins, J.R. 2009. Anchorage-independent cell growth signature identifies tumors with metastatic potential. Oncogene 28(31): 2796-2805. doi: onc2009139 [pii] 10.1038/onc.2009.139. Naga Prasad, S.V., Jayatilleke, A., Madamanchi, A., and Rockman, H.A. 2005. Protein kinase activity of phosphoinositide 3-kinase regulates beta-adrenergic receptor endocytosis. Nat Cell Biol 7(8): 785-796. O'Neill, G.M. 2009. The coordination between actin filaments and adhesion in mesenchymal migration. Cell Adh Migr 3(4): 355-357. doi: 9468 [pii]. Perry, S.V. 2001. Vertebrate tropomyosin: distribution, properties and function. J Muscle Res Cell Motil 22(1): 5-49. Pittenger, M.F., Kistler, A., and Helfman, D.M. 1995. Alternatively spliced exons of the beta tropomyosin gene exhibit different affinities for F-actin and effects with nonmuscle caldesmon. Journal of cell science 108 ( Pt 10): 3253-3265. Shah, V., Braverman, R., and Prasad, G.L. 1998. Suppression of neoplastic transformation and regulation of cytoskeleton by tropomyosins. Somat Cell Mol Genet 24(5): 273-280. Simoneau, B., Houle, F., and Huot, J. 2012. Regulation of endothelial permeability and transendothelial migration of cancer cells by tropomyosin-1 phosphorylation. Vascular cell 4(1): 18. doi: 10.1186/2045-824X-4-18. Somara, S., Pang, H., and Bitar, K.N. 2005. Agonist-induced association of tropomyosin with protein kinase Calpha in colonic smooth muscle. Am J Physiol Gastrointest Liver Physiol 288(2): G268-276. Tremblay, P.L., Huot, J., and Auger, F.A. 2008. Mechanisms by which E-selectin regulates diapedesis of colon cancer cells under flow conditions. Cancer Res 68(13): 5167-5176. doi: 68/13/5167 [pii] 10.1158/0008-5472.CAN-08-1229. Urbancikova, M., and Hitchcock-DeGregori, S.E. 1994. Requirement of amino-terminal modification for striated muscle alpha-tropomyosin function. The Journal of biological chemistry 269(39): 24310-24315. Wilton, S.D., Eyre, H., Akkari, P.A., Watkins, H.C., MacRae, C., Laing, N.G., and Callen, D.C. 1995. Assignment of the human a-tropomyosin gene TPM3 to 1q22-->q23 by fluorescence in situ hybridisation. Cytogenet Cell Genet 68(1-2): 122-124.

166

Wilton, S.D., Lim, L., Dorosz, S.D., Gunn, H.C., Eyre, H.J., Callen, D.F., and Laing, N.G. 1996. Assignment of the human alpha-tropomyosin gene TPM4 to band 19p13.1 by fluorescence in situ hybridization. Cytogenet Cell Genet 72(4): 294-296. Yu, M., Sato, H., Seiki, M., and Thompson, E.W. 1995. Complex regulation of membrane- type matrix metalloproteinase expression and matrix metalloproteinase-2 activation by concanavalin A in MDA-MB-231 human breast cancer cells. Cancer Res 55(15): 3272-3277.

167

FIGURE

168

Figure 1. Generation of MDA MB-231 cells stably expressing wild type forms of Tm1 or phosphomimetic or nonphosphorylatable forms of Tm1. MDA MB-231 were transfected with vectors expressing wild-type form of Tm1 (Tm1wt), a non phosphorylatable (Tm1 S283A: MDA TM1A) or a phosphomimetic (Tm1S283E: MDATM1 E) mutated form of Tm1 using the calcium chloride precipitation technique. The cells were selected by incubation in medium containing G418 at 400 µg/ml. A pool of resistant cells were obtained and used in experiments described in the paper. Western blot was carried out with protein extracted from the cell lines expressing the various forms of Tm1, using tropomyosin antibody.

169

Figure 2. Tropomyosin-1 expression in MDA MB231 cell lines reduces their proliferation. Parental MDA MB231 cells and MDA MB231 expressing wt Tm1 (TMwt) were incubated with medium supplemented with 10% FBS for 96 h. Cell proliferation was determined using a colorimetric method based on conversion of tetrazolium salt WST-1 to a colored compound. Values are average ± SD from an experiment done in triplicates. P value was determined by a Student’s t-test. The experiment was performed at least three times in triplicates.

170

Figure 3. Phosphorylation of Tm1 at S283 increases their adhesion to substratum. Parental MDA MB231-Tm1 or MDA MB-231 cells stably expressing non phosphorylatable (Tm1S283A: TMA) or a phosphomimetic (Tm1S283E: TME) cells were incubated with medium supplemented with 10% FBS for 1-2 hours at 37°C. Thereafter, the cells were fixed with formaldehyde 3.7%, washed with PBS, stained with crystal violet and lysed with SDS1%. Absorbance was used as the measure of adhesion. Values are average ± SD from an experiment done in triplicates. The experiments were performed at least three times in triplicates. p value was determined by a Student’s t-test.

171

A B

Figure 4. Phosphorylation of Tm1 at S283 reduced cell migration in MDA MB231. A, Cell migration was evaluated by means of a scratch test assay applied to parental MDA MB231 and MDA MB231 cells stably expressing Tm1S283A (TMA) or Tm1S283E (TME). The black lines represent the initial wound and were used to visualize the cells that invaded the scratch. B. The cells that crossed the lines were manually counted from four different fields. Their average number ± SD from two separate experiments are indicated in ordinate. p was determined by a Student’s t-test.

172

Figure 5. Phosphorylation of Tm1 at S283 reduced transendothelial cell migration of MDA MB231. Transendothelial cell migration of parental MDA MB231 and MDA MB231 cells stably expressing Tm1S283A (TMA) or Tm1S283E (TME). Briefly, HUVEC cells (150x103) were grown to confluence (48 h) on an 8.0 µm pore size gelatinized polycarbonate membranes separating the two compartments of a 6.5 mm Transwell. MDA MB231 cells in suspension were stained with 500 nM calcein-AM for 30 min at 37°C, and then added in migration buffer on the endothelial monolayer. The chambers were incubated for 6 h at 37°C in a 5% CO2 atmosphere. The cells in the upper part of the chamber were removed with a cotton swab. Then, fluorescent MDA MB231 cells that crossed the membrane were counted in five fields using a 20x lens on a Nikon-TE300 inverted microscope (Gout et al. 2006). Values are average ± SD from two independent experiments done in triplicates. p value was determined by a Student’s t-test.

173

Figure 6. Phosphorylation of Tm1 at S283 reduces the formation of MDA-MB231 colonies on soft agar by cells. Liquefied 0.6% agarose was mixed with an equal volume of DMEM 2X medium. Ten ml of this solution was mixed with 1 ml of growth medium containing 104 cells in 0.27% agarose. One ml of this cell suspension was layered on top of 1% agarose base, and 1 ml of DMEM medium containing 10% FBS was added. The cells (parental MDA MB231, MDA MB231 S283A (TMA) or MDA MB231S283E (TME) were incubated for 14 days, after which colonies present in 5 representative fields were counted. p value was determined by a Student’s t-test.

174

Chapitre 5: Conclusion générale

175

176

Au Canada et partout dans le monde, le cancer du sein est un enjeu de taille en santé publique. L’E2 est impliqué dans le développement et la progression du cancer du sein hormono-dépendant [53, 54, 93, 176, 177]. Des études ont montré que dans les carcinomes du sein, les concentrations tissulaires d’E2 sont 10 fois plus élevées que dans le plasma [178, 179]. Les tissus du cancer du sein humain contiennent toutes les enzymes impliquées dans les étapes finales de la biosynthèse de l'E2, y compris des enzymes tels que l'aromatase, la stéroide sulfatase et la 17β-hydroxystéroïde déshydrogénase type 1 (17β-HSD1) [7, 180]. L’enzyme stéroïdienne 17β-HSD1 joue un rôle majeur dans la progression du cancer du sein de type hormono-dépendant. Cette enzyme catalyse la conversion de E1 (estrogène inactif), en E2 (estrogène biologiquement actif). Aussi, elle permet l’inactivation de l'androgène le plus actif, la dihydrotestostérone (DHT) [181-184]. Différents inhibiteurs sont utilisés en clinique pour inhiber la synthèse finale de E2. Cependant aucun inhibiteur de la 17β-HSD1 n’est encore utilisé en clinique malgré l’importance de cette enzyme dans la conversion de E1 en E2.

Mon premier projet de recherche consistait à démontrer l’effet de l’inhibition de la 17β- HSD1 sur la régulation du profil protéique et génomique, sur le cycle cellulaire et sur l’invasion cellulaire des cellules cancéreuses MCF7 et T47D.

L’effet de l’inhibition de la 17β-HSD1 sur le profil protéique des cellules cancéreuses T47D et MCF7

Une analyse protéique a été réalisée pour comparer le protéome des cellules T47D contrôles à celui des cellules traitées avec un inhibiteur stéroïdien ou transfectées avec un siRNA spécifique contre la 17β-HSD1 (Chapitre 2). L’analyse protéique a été réalisée en utilisant l’électrophorèse bidimensionnelle des protéines suivie par une spectrométrie de masse afin d’analyser les spots modulés. Suite à l’inhibition de la 17β-HSD1 par un inhibiteur spécifique, l'expression de la tumor protéine D54 (TPD54), la glutathion S- transférase Mu 3 (GSTMu3) et prelamine-A/C (LMNA) a diminué. Tandis que l'expression des protéines impliquées dans la régulation du cycle cellulaire telles que la protéine 14-3-3

177 epsilon (YWHAE) et la tumor protein D53 (TPD53) a augmenté. La TPD53, membre de la famille de D52-like, est identifiée comme partenaire de liaison de la protéine 14-3-3 [185]. La TPD53 régule la progression du cycle cellulaire, elle est fortement exprimée à la phase G2/M et faiblement exprimée à la phase G0/G1 et son expression est corrélée avec l'expression de la cycline B1 [186].

Nous avons également comparé le protéome des cellules T47D contrôle à celui des cellules T47D transfectées avec un siRNA spécifique contre la 17β-HSD1. Les résultats protéiques ont montré que l’inhibition de l’expression de la 17β-HSD1 augmente l'expression de la protéine (ZPR1) et diminue l’expression de la nm23 (nm23-H1), un régulateur de métastase du cancer du sein. Nous avons également constaté une diminution de l'expression de la 17β-HSD1 suite à l’inhibition de l’expression de la nm23-H1. Nous avons déjà démontré que l'expression de la 17β-HSD1 régule positivement le gène nm23-H1 et augmente la migration des cellules MCF7 du cancer du sein [187]. Nos résultats montrent également qu'il existe une corrélation positive entre la 17β-HSD1 et la nm23-H1 dans les cellules cancéreuses du sein T47D.

Une troisième analyse protéomique a été réalisée pour comparer le protéome des cellules MCF7 sauvages à celui des cellules transfectées avec un siRNA spécifique contre la 17β- HSD1. Suite à l’inhibition de l’expression de la 17β-HSD1, différentes protéines ont été modulées notamment la septin 8 (régulateur du cycle cellulaire) qui a été surexprimée tandis que la tumor protein D54 (régulateur de la prolifération cellulaire) et la nm23-H1 (régulateur de la métastase du cancer du sein) ont été sous-exprimées. Nos résultats montrent encore une fois qu’il existe une corrélation positive entre la 17β-HSD1 et la nm23-H1 dans les cellules cancéreuses MCF7.

Une analyse IPA a montré que suite à l’inhibition de la 17β-HSD1 dans les cellules T47D par un inhibiteur ou un siRNA, quatre protéines, communes aux deux traitements, ont été sous-exprimées notamment l’actinin alpha 4 (ACTN4), la cathepsin D (CTSD), la heat shock protein beta-1 (HSPB1) et la heterogeneous nuclear ribonucleoprotein C-like 1 (HNRCL). Il

178

est intéressant de noter que les trois premiers gènes sont associés à la physiopathologie du cancer du sein.

L’effet de l’inhibition de la 17β-HSD1 sur le profil génomique des cellules cancéreuses T47D

Une analyse génomique a été réalisée pour comparer le génome des cellules T47D contrôle et des cellules T47D traitées avec un inhibiteur spécifique à la 17β-HSD1 (Chapitre 3). L'analyse génomique a été réalisée à l'aide de puces à ADN. Suite à l'inhibition de la 17β- HSD1 au niveau des cellules T47D, l'expression des gènes impliqués dans le transport (RANBP3L, APOD), la liaison d'ADN (HIST1H2BM), le traitement de l'antigène et la présentation de l'antigène de peptide ou de polysaccharide par l'intermédiaire du CMH de classe II (HLA- DQA2) a diminué. Tandis que l'expression de TP63 (régulateur transcriptionnel) et CD36 (impliquée dans l'adhérence cellulaire) a augmenté. CD36 est un récepteur transmembranaire impliqué dans l'apoptose, les interactions cellule-ECM, l'activation de TGF-β, la différenciation des adipocytes, l'angiogenèse et de la signalisation immunitaire [188]. Dans les cellules endothéliales, CD36 bloque l'action du VEGF [189]. Dans notre étude, la surexpression de CD36 suite à l'inhibition de la 17β-HSD1 pourrait promouvoir un profil anti-angiogénique. Dans les cancers humains, p63 est souvent surexprimée ou dérégulée [190]. D'autres études ont découvert que p63 agit comme un suppresseur de métastases. Dans les tumeurs épithéliales, TAp63 inhibe l’invasion et les métastases en surveillant l'expression des gènes inhibiteurs de métastases [191-193]. Dans notre cas, p63 est surexprimée suite à l'inhibition de la 17β-HSD1, ce qui peut améliorer son action de suppresseur de métastases. ApoD est une glycoprotéine identifiée dans le plasma [194, 195]. Dans les stades avancés de cancer de la prostate, une augmentation de l'expression de apoD a été notée [196]. La fonction exacte de ApoD n’est pas encore totalement clair [197]. ApoD pourrait être un marqueur pour la signalisation des stéroïdes [198] et un marqueur de pronostic et de prédiction putatif très intéressant dans le cancer du sein [197]. Cette étude montre que l'inhibition de la 17β-HSD1 module le profil génomique

179 des cellules T47D et affecte l'expression de gènes impliqués dans plusieurs fonctions physiologiques.

L’effet de l’inhibition de la 17β-HSD1 sur la prolifération, le cycle cellulaire et l’invasion des cellules cancéreuses T47D et MCF7

Les résultats présentés au chapitre 2 montrent que l’inhibition de la 17β-HSD1 dans les cellules T47D et MCF7 par un inhibiteur spécifique ou un siRNA mène à la diminution de la prolifération et de la formation d’estradiol (E2) dans les cellules cancéreuses. Aussi, la surexpression de la 17β-HSD1 dans les cellules MCF7 permet une augmentation de la croissance cellulaire et la formation d’E2. Suzuki et al ont aussi montré que la 17β-HSD1 est responsable du processus de régulation menant à l'accumulation d’E2 dans les tissus cancéreux humains du sein [199]. D’autres études ont prouvé que l'inhibition de la 17β- HSD1 mène à une diminution de la prolifération des cellules T47D [59, 183], tandis que la surexpression de la 17β-HSD1 stimule la croissance des cellules MCF7 [183]. Un ratio [E2]/[E1] élevé est positivement corrélé à l’expression de la 17β-HSD1 [200] et à la prolifération des cellules cancéreuses du sein [183]. Nos résultats confirment la corrélation positive entre l'expression de la 17β-HSD1, le niveau d’E2 et la prolifération des cellules cancéreuses du sein T47D et MCF7.

Nous avons aussi montré que l’inhibition et le knockdown de la 17β-HSD1 mène à la diminution de l’invasion cellulaire et de la formation de colonies des cellules T47D et MCF7-17β-HSD1. Dans les cellules MCF7-17β-HSD1, l’invasion cellulaire est plus élevée que dans les cellules MCF7 contrôles. Nous avons déjà démontré que l'expression de la 17β- HSD1 est positivement corrélée à la migration des cellules cancéreuses du sein MCF7 et à l’expression de la nm23-H1 [187]. Les estrogènes sont des promoteurs du mouvement cellulaire dans plusieurs tissus, y compris le sein [201]. Il a aussi été démontré qu’en présence d’estrogènes, les cellules MCF-10F forme des colonies dans l’agar et montre une capacité d'invasion plus élevée dans le Matrigel que les cellules MCF-10F en absence d'estrogènes [202, 203]. Nos résultats suggèrent que l'inhibition de 17β-HSD1 diminue la

180

capacité invasive des cellules T47D en diminuant le niveau d’expression de nm23-H1 ainsi que le niveau d’E2. Il existe aussi une corrélation positive entre l'expression de la 17β-HSD1, la nm23-H1 et l'invasion des cellules MCF7.

Suite au knockdown de la 17β-HSD1, les résultats du cycle cellulaire ont montré une augmentation significative des cellules en phase GO/G1 et une diminution significative des cellules en phase S et G2/M. Aussi, l’inhibition de la 17β-HSD1 dans les cellules T47D n’a pas d’effet sur l’apoptose. Au niveau des cellules MCF7, la surexpression de la 17β-HSD1, mène à une diminution significative des cellules en phase GO/G1, ainsi qu’une augmentation significative des cellules en phase S et G2/M en comparaison avec les cellules MCF7. Les résultats d’apoptose montrent une augmentation des cellules vivantes MCF7-17β-HSD1 et une diminution des cellules apoptotiques MCF7-17β-HSD1 en comparaison avec les cellules MCF7. Il a été démontré que le traitement des cellules avec l’E2, diminue de manière significative le pourcentage de cellules en phase G0/G1 du cycle cellulaire et stimule l’entrée des cellules en phase S [204]. Par contre, le traitement avec un anti-estrogène, le tamoxifène, mène à l’arrêt du cycle cellulaire en phase G0/G1 [205]. Nos résultats montrent que le knock-down de la 17β-HSD1, comme pour le tamoxifène, conduit à l'inhibition de la progression G1/S et l'arrêt du cycle cellulaire. Ceci est probablement dû au faible niveau d’E2.

L’inhibition de la synthèse d’E2 par la voie de la stéroïde sulfatase via la 17β-HSD1 en combinaison avec des anti-aromatases peut ouvrir de nouvelles applications cliniques dans le cas du cancer du sein estrogèno-dépendant. Il a été démontré que l’inhibition de la 17β- HSD1 inhibe la prolifération des cellules T47D in vitro et diminue significativement l’E2 plasmatique et le volume des tumeurs dans un modèle animal [206]. Récemment, un inhibiteur non estrogénique in vivo et in vitro (PBRM) a été synthétisé [60, 61]. PBRM a un IC50 de 68 nmol/L et il est capable de réduire de 100% in vivo la taille de la tumeur dont la croissance est stimulée par E1 [60]. De plus l’inhibition de la 17β-HSD1 peut aussi inhiber la synthèse de 5-androstène-3β, 17β-diol (5-diol), un autre estrogène actif [207, 208]. Cette inhibition ne peut pas être obtenue en utilisant comme cible l'aromatase. Montrant ainsi

181 l'importance majeure de l’inhibition de la 17β-HSD1 et l'avantage d'une thérapie combinée avec l’inhibition de l'aromatase dans le cas du cancer du sein hormono-dépendant.

L’impact de la phosphorylation de la tropomyosine 1 alpha sur ses propriétés de suppresseur de tumeur

Mon deuxième projet de recherche portait sur l’étude de la phosphorylation de la tropomyosine chaîne alpha 1 (Tm1) qui est une protéine qui se lie au filament d’actine. La tropomyosine de haut poids moléculaire tel que la Tm1 est un suppresseur de tumeur majeur dans le cancer du sein [209 , 210]. Cependant, les mécanismes expliquant ses propriétés de suppresseur de tumeur sont mal définis. Ici, nous avons fourni la première preuve montrant que la phosphorylation de la Tm1 est un mécanisme majeur qui régule ses fonctions de suppresseur de tumeur.

Pour réaliser cette étude, les cellules MDA MB231 du cancer du sein surexprimant de manière stable la forme sauvage ou mutée de la Tm1 (S283A et S283E) ont été générées (Chapitre 4). Nous avons trouvé que les cellules exprimant la forme phosphomimétique (pseudo-phosphorylée) de la Tm1 S283E/Tm1 sont caractérisées par une adhérence accrue au substrat. En effet, ce résultat est cohérent avec le fait que les cellules MDA MB231 / S283E / Tm1 sont moins motiles et moins invasives étant donné qu’elles sont attachées au substrat et ceci est dû au fait que les cellules MDA MB231 / S283E / Tm1 contient plus de fibres de stress [169]. En outre, la migration des cellules MDA-MB231/S283E/Tm1 dans un essai de cicatrisation est réduite par rapport aux cellules parentales ou ceux exprimant la forme non- phosphorylable de la Tm1 (S283A). De même, la migration transendothéliale des cellules MDA-MB231/S283E/Tm1 est également réduite par rapport aux autres lignées cellulaires. Cette observation est un argument prouvant que la phosphorylation de la Tm1 au niveau de la Ser283 lui confère ses propriétés de suppresseur de tumeur ainsi que le potentiel anti- métastatique étant donné que la migration transendothéliale est prérequise pour la formation de métastases [211].

182

Nous avons aussi constaté que les cellules exprimant les mutants S283A forment plus de colonies en agar mou que celles exprimant les mutants S283E. Ce test de croissance sur agar mou reflète la malignité in vitro [212]. Néanmoins, le fait que les cellules exprimant les mutants S283A forment également moins de colonies que les cellules parentales, mais plus que les cellules exprimant les mutants S283E, indiquent que des facteurs autres que la phosphorylation sont impliqués dans la suppression des tumeurs par la Tm1.

En conclusion à cette étude, la phosphorylation de la Tm1 au niveau de la Ser283 contribue à ses propriétés anti-tumorales, et cet effet résulte principalement d'une augmentation de l'adhérence des cellules cancéreuses associée à une diminution de leurs potentiels migratoire et invasif. La tropomyosine (Tm) semble être un acteur important de la transformation maligne. Cependant, la compréhension de son rôle est compliquée par les nombreuses isoformes de tropomyosine et par le fait que certaines de ces fonctions sont régulées de façon postraductionnelles. Or, bien comprendre le rôle de la tropomyosine aurait une énorme implication sur la compréhension des cancers humains. Comme perspectives à ce projet, il faudrait vérifier le profil d’expression des isoformes de la tropomyosine dans des tissus cancéreux humains du sein et confirmer le rôle de l’expression et de la phosphorylation de la Tm dans des modèles de souris. En obtenant plus d’information sur son rôle dans le cancer, il serait donc possible d’envisager l’utilisation de la tropomyosine comme biomarqueur de diagnostique et de pronostique.

183

184

Bibliographie (Chapitres 1 et 5)

1. Hooks MA: Breast cancer: risk assessment and prevention. Southern medical journal 2010, 103(4):333-338. 2. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D: Global cancer statistics. CA: a cancer journal for clinicians 2011, 61(2):69-90. 3. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, Thun MJ: Cancer statistics, 2008. CA: a cancer journal for clinicians 2008, 58(2):71-96. 4. Siegel RL, Miller KD, Jemal A: Cancer statistics, 2015. CA: a cancer journal for clinicians 2015, 65(1):5-29. 5. Cancer facts from Canadian Cancer Statistics 2013: https://www.cancer.ca/~/media/cancer.ca/QC/about%20us/media/Facts%20shee ts/13-05-29-Cancer-Stats-Facts-2013-VFIN-angl.pdf 6. Hayashi SI, Eguchi H, Tanimoto K, Yoshida T, Omoto Y, Inoue A, Yoshida N, Yamaguchi Y: The expression and function of estrogen receptor alpha and beta in human breast cancer and its clinical application. Endocrine-related cancer 2003, 10(2):193-202. 7. Chetrite GS, Cortes-Prieto J, Philippe JC, Wright F, Pasqualini JR: Comparison of estrogen concentrations, estrone sulfatase and aromatase activities in normal, and in cancerous, human breast tissues. The Journal of steroid biochemistry and molecular biology 2000, 72(1-2):23-27. 8. IARC: Cancer incidence in five continents. International Agency for Research on Cancer, Lyon 2007, Volume 9. 9. Cavalieri EL, Stack DE, Devanesan PD, Todorovic R, Dwivedy I, Higginbotham S, Johansson SL, Patil KD, Gross ML, Gooden JK et al: Molecular origin of cancer: catechol estrogen-3,4-quinones as endogenous tumor initiators. Proceedings of the National Academy of Sciences of the United States of America 1997, 94(20):10937-10942. 10. Cavalieri E, Chakravarti D, Guttenplan J, Hart E, Ingle J, Jankowiak R, Muti P, Rogan E, Russo J, Santen R et al: Catechol estrogen quinones as initiators of breast and other human cancers: implications for biomarkers of susceptibility and cancer prevention. Biochimica et biophysica acta 2006, 1766(1):63-78. 11. Bhat HK, Calaf G, Hei TK, Loya T, Vadgama JV: Critical role of oxidative stress in estrogen-induced carcinogenesis. Proceedings of the National Academy of Sciences of the United States of America 2003, 100(7):3913-3918. 12. Mense SM, Remotti F, Bhan A, Singh B, El-Tamer M, Hei TK, Bhat HK: Estrogen- induced breast cancer: alterations in breast morphology and oxidative stress as a function of estrogen exposure. Toxicology and applied pharmacology 2008, 232(1):78-85. 13. IARC: IARC monographs on the evaluation of carcinogenic risks to humans: overall evaluations of carcinogenicity, an updating of IARC monographs. International Agency for Research on Cancer, Lyon 1987, volumes 1 to 42. Supplement 7.

185

14. IARC: IARC monographs on the evaluation of carcinogenic risks to humans: hormonal contraception and postmenopausal hormone therapy, 72. International Agency for Research on Cancer, Lyon, 1999:pp. 474–530. 15. NTP: Report on Carcinogens. National Toxicology Program, Research Triangle Park. 2002. 16. Nelson HD, Humphrey LL, Nygren P, Teutsch SM, Allan JD: Postmenopausal hormone replacement therapy: scientific review. Jama 2002, 288(7):872-881. 17. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC et al: Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial. Jama 2002, 288(3):321-333. 18. McPherson K, Steel CM, Dixon JM: ABC of breast diseases. Breast cancer- epidemiology, risk factors, and genetics. Bmj 2000, 321(7261):624-628. 19. 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. International journal of cancer Journal international du cancer 2001, 94(5):685-689. 20. Suzuki T, Miki Y, Nakamura Y, Moriya T, Ito K, Ohuchi N, Sasano H: Sex steroid- producing enzymes in human breast cancer. Endocrine-related cancer 2005, 12(4):701-720. 21. Pasqualini JR, Chetrite G, Blacker C, Feinstein MC, Delalonde L, Talbi M, Maloche C: Concentrations of estrone, estradiol, and estrone sulfate and evaluation of sulfatase and aromatase activities in pre- and postmenopausal breast cancer patients. The Journal of clinical endocrinology and metabolism 1996, 81(4):1460- 1464. 22. Pasqualini JR: The selective estrogen enzyme modulators in breast cancer: a review. Biochimica et biophysica acta 2004, 1654(2):123-143. 23. Aka JA, Mazumdar M, Lin SX: Reductive 17beta-hydroxysteroid dehydrogenases in the sulfatase pathway: critical in the cell proliferation of breast cancer. Molecular and cellular endocrinology 2009, 301(1-2):183-190. 24. Simpson ER, Kilgore MW, Mahendroo MS, Means GD, Corbin CJ, Mendelson CR: Regulation of human aromatase cytochrome P450 gene expression. The Journal of steroid biochemistry and molecular biology 1992, 43(8):923-930. 25. Jordan VC, Brodie AM: Development and evolution of therapies targeted to the estrogen receptor for the treatment and prevention of breast cancer. Steroids 2007, 72(1):7-25. 26. Nebert DW, Nelson DR, Coon MJ, Estabrook RW, Feyereisen R, Fujii-Kuriyama Y, Gonzalez FJ, Guengerich FP, Gunsalus IC, Johnson EF et al: The P450 superfamily: update on new sequences, gene mapping, and recommended nomenclature. DNA and cell biology 1991, 10(1):1-14. 27. Toda K, Shizuta Y: Molecular cloning of a cDNA showing alternative splicing of the 5'-untranslated sequence of mRNA for human aromatase P-450. European journal of biochemistry / FEBS 1993, 213(1):383-389. 28. Thompson EA, Jr., Siiteri PK: Utilization of oxygen and reduced nicotinamide adenine dinucleotide phosphate by human placental microsomes during

186

aromatization of androstenedione. The Journal of biological chemistry 1974, 249(17):5364-5372. 29. Akhtar M, Calder MR, Corina DL, Wright JN: Mechanistic studies on C-19 demethylation in oestrogen biosynthesis. The Biochemical journal 1982, 201(3):569-580. 30. Santen RJ, Brodie H, Simpson ER, Siiteri PK, Brodie A: History of aromatase: saga of an important biological mediator and therapeutic target. Endocrine reviews 2009, 30(4):343-375. 31. Peterson JA, Ebel RE, O'Keeffe DH, Matsubara T, Estabrook RW: Temperature dependence of cytochrome P-450 reduction. A model for NADPH-cytochrome P- 450 reductase:cytochrome P-450 interaction. The Journal of biological chemistry 1976, 251(13):4010-4016. 32. Varela CL, Amaral C, Tavares da Silva E, Lopes A, Correia-da-Silva G, Carvalho RA, Costa SC, Roleira FM, Teixeira N: Exemestane metabolites: Synthesis, stereochemical elucidation, biochemical activity and anti-proliferative effects in a hormone-dependent breast cancer cell line. European journal of medicinal chemistry 2014, 87C:336-345. 33. MacIndoe JH: The hydrolysis of estrone sulfate and dehydroepiandrosterone sulfate by MCF-7 human breast cancer cells. Endocrinology 1988, 123(3):1281- 1287. 34. Reed MJ, Purohit A, Woo LW, Newman SP, Potter BV: Steroid sulfatase: molecular biology, regulation, and inhibition. Endocrine reviews 2005, 26(2):171- 202. 35. Partanen S: Histochemistry of estrogen sulfatases in human breast diseases. Virchows Archiv B, Cell pathology including molecular pathology 1985, 49(1):53-60. 36. Dodgson KS, Spencer B, Thomas J: Studies on sulphatases. 6. The localization of arylsulphatase in the rat-liver cell. The Biochemical journal 1954, 56(2):177-181. 37. Willemsen R, Kroos M, Hoogeveen AT, van Dongen JM, Parenti G, van der Loos CM, Reuser AJ: Ultrastructural localization of steroid sulphatase in cultured human fibroblasts by immunocytochemistry: a comparative study with lysosomal enzymes and the mannose 6-phosphate receptor. The Histochemical journal 1988, 20(1):41-51. 38. Kawano J, Kotani T, Umeki K, Oinuma T, Ohtaki S, Aikawa E: A monoclonal antibody to rat liver arylsulfatase C and its application in immunohistochemistry. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 1989, 37(5):683-690. 39. Utsumi T, Yoshimura N, Takeuchi S, Maruta M, Maeda K, Harada N: Elevated steroid sulfatase expression in breast cancers. The Journal of steroid biochemistry and molecular biology 2000, 73(3-4):141-145. 40. Utsumi T, Yoshimura N, Takeuchi S, Ando J, Maruta M, Maeda K, Harada N: Steroid sulfatase expression is an independent predictor of recurrence in human breast cancer. Cancer research 1999, 59(2):377-381. 41. Miyoshi Y, Ando A, Hasegawa S, Ishitobi M, Taguchi T, Tamaki Y, Noguchi S: High expression of steroid sulfatase mRNA predicts poor prognosis in patients with estrogen receptor-positive breast cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 2003, 9(6):2288- 2293. 187

42. Yoshimura N, Harada N, Bukholm I, Karesen R, Borresen-Dale AL, Kristensen VN: Intratumoural mRNA expression of genes from the oestradiol metabolic pathway and clinical and histopathological parameters of breast cancer. Breast cancer research : BCR 2004, 6(2):R46-55. 43. Stanway SJ, Purohit A, Woo LW, Sufi S, Vigushin D, Ward R, Wilson RH, Stanczyk FZ, Dobbs N, Kulinskaya E et al: Phase I study of STX 64 (667 Coumate) in breast cancer patients: the first study of a steroid sulfatase inhibitor. Clinical cancer research : an official journal of the American Association for Cancer Research 2006, 12(5):1585-1592. 44. Zhu DW, Lee X, Breton R, Ghosh D, Pangborn W, Daux WL, Lin SX: Crystallization and preliminary X-ray diffraction analysis of the complex of human placental 17 beta-hydroxysteroid dehydrogenase with NADP+. Journal of molecular biology 1993, 234(1):242-244. 45. Ghosh D, Pletnev VZ, Zhu DW, Wawrzak Z, Duax WL, Pangborn W, Labrie F, Lin SX: Structure of human estrogenic 17 beta-hydroxysteroid dehydrogenase at 2.20 A resolution. Structure 1995, 3(5):503-513. 46. Azzi A, Rehse PH, Zhu DW, Campbell RL, Labrie F, Lin SX: Crystal structure of human estrogenic 17 beta-hydroxysteroid dehydrogenase complexed with 17 beta-estradiol. Nature structural biology 1996, 3(8):665-668. 47. Breton R, Housset D, Mazza C, Fontecilla-Camps JC: The structure of a complex of human 17beta-hydroxysteroid dehydrogenase with estradiol and NADP+ identifies two principal targets for the design of inhibitors. Structure 1996, 4(8):905-915. 48. Puranen T, Poutanen M, Ghosh D, Vihko P, Vihko R: Characterization of structural and functional properties of human 17 beta-hydroxysteroid dehydrogenase type 1 using recombinant enzymes and site-directed mutagenesis. Molecular endocrinology 1997, 11(1):77-86. 49. Peltoketo H, Isomaa V, Maentausta O, Vihko R: Complete amino acid sequence of human placental 17 beta-hydroxysteroid dehydrogenase deduced from cDNA. FEBS letters 1988, 239(1):73-77. 50. Luu The V, Labrie C, Zhao HF, Couet J, Lachance Y, Simard J, Leblanc G, Cote J, Berube D, Gagne R et al: Characterization of cDNAs for human estradiol 17 beta- dehydrogenase and assignment of the gene to chromosome 17: evidence of two mRNA species with distinct 5'-termini in human placenta. Molecular endocrinology 1989, 3(8):1301-1309. 51. Martel C, Rheaume E, Takahashi M, Trudel C, Couet J, Luu-The V, Simard J, Labrie F: Distribution of 17 beta-hydroxysteroid dehydrogenase gene expression and activity in rat and human tissues. The Journal of steroid biochemistry and molecular biology 1992, 41(3-8):597-603. 52. Luu-The V: Analysis and characteristics of multiple types of human 17beta- hydroxysteroid dehydrogenase. The Journal of steroid biochemistry and molecular biology 2001, 76(1-5):143-151. 53. Thomas DB: Do hormones cause breast cancer? Cancer 1984, 53(3 Suppl):595- 604. 54. Vihko R, Apter D: Endogenous steroids in the pathophysiology of breast cancer. Critical reviews in oncology/hematology 1989, 9(1):1-16.

188

55. Gunnarsson C, Olsson BM, Stal O, Southeast Sweden Breast Cancer G: Abnormal expression of 17beta-hydroxysteroid dehydrogenases in breast cancer predicts late recurrence. Cancer research 2001, 61(23):8448-8451. 56. Oduwole OO, Li Y, Isomaa VV, Mantyniemi A, Pulkka AE, Soini Y, Vihko PT: 17beta-hydroxysteroid dehydrogenase type 1 is an independent prognostic marker in breast cancer. Cancer research 2004, 64(20):7604-7609. 57. Vermeulen A, Deslypere JP, Paridaens R: Steroid dynamics in the normal and carcinomatous mammary gland. Journal of steroid biochemistry 1986, 25(5B):799- 802. 58. Berube M, Poirier D: Improved synthesis of EM-1745, preparation of its C17- ketone analogue and comparison of their inhibitory potency on 17beta- hydroxysteroid dehydrogenase type 1. Journal of enzyme inhibition and medicinal chemistry 2009, 24(3):832-843. 59. Laplante Y, Cadot C, Fournier MA, 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. Bioorganic & medicinal chemistry 2008, 16(4):1849-1860. 60. Ayan D, Maltais R, Roy J, Poirier D: A new nonestrogenic steroidal inhibitor of 17beta-hydroxysteroid dehydrogenase type I blocks the estrogen-dependent breast cancer tumor growth induced by estrone. Molecular cancer therapeutics 2012, 11(10):2096-2104. 61. Maltais R, Ayan D, Trottier A, Barbeau X, Lague P, Bouchard JE, Poirier D: Discovery of a non-estrogenic irreversible inhibitor of 17beta-hydroxysteroid dehydrogenase type 1 from 3-substituted-16beta-(m-carbamoylbenzyl)-estradiol derivatives. Journal of medicinal chemistry 2014, 57(1):204-222. 62. Makela S, Poutanen M, Kostian ML, Lehtimaki N, Strauss L, Santti R, Vihko R: Inhibition of 17beta-hydroxysteroid oxidoreductase by flavonoids in breast and prostate cancer cells. Proceedings of the Society for Experimental Biology and Medicine Society for Experimental Biology and Medicine 1998, 217(3):310-316. 63. Liehr JG, Han X, Bhat HK: 32P-postlabelling in studies of hormonal carcinogenesis. IARC scientific publications 1993(124):149-155. 64. Clemons M, Goss P: Estrogen and the risk of breast cancer. The New England journal of medicine 2001, 344(4):276-285. 65. Yager JD, Davidson NE: Estrogen carcinogenesis in breast cancer. The New England journal of medicine 2006, 354(3):270-282. 66. Rao GS: MOde of entry of steroid and thyroid hormones into cells. Molecular and cellular endocrinology 1981, 21(2):97-108. 67. MacGregor JI, Jordan VC: Basic guide to the mechanisms of antiestrogen action. Pharmacological reviews 1998, 50(2):151-196. 68. Gronemeyer H: Transcription activation by estrogen and progesterone receptors. Annual review of genetics 1991, 25:89-123. 69. Yamaguchi Y: Microenvironmental regulation of estrogen signals in breast cancer. Breast cancer 2007, 14(2):175-181. 70. Bjornstrom L, Sjoberg M: Mechanisms of estrogen receptor signaling: convergence of genomic and nongenomic actions on target genes. Molecular endocrinology 2005, 19(4):833-842.

189

71. Liu MM, Albanese C, Anderson CM, Hilty K, Webb P, Uht RM, Price RH, Jr., Pestell RG, Kushner PJ: Opposing action of estrogen receptors alpha and beta on cyclin D1 gene expression. The Journal of biological chemistry 2002, 277(27):24353-24360. 72. Bruning JC, Lingohr P, Gillette J, Hanstein B, Avci H, Krone W, Muller-Wieland D, Kotzka J: Estrogen receptor-alpha and Sp1 interact in the induction of the low density lipoprotein-receptor. The Journal of steroid biochemistry and molecular biology 2003, 86(2):113-121. 73. Kalaitzidis D, Gilmore TD: Transcription factor cross-talk: the estrogen receptor and NF-kappaB. Trends in endocrinology and metabolism: TEM 2005, 16(2):46-52. 74. Le Goff P, Montano MM, Schodin DJ, Katzenellenbogen BS: Phosphorylation of the human estrogen receptor. Identification of hormone-regulated sites and examination of their influence on transcriptional activity. The Journal of biological chemistry 1994, 269(6):4458-4466. 75. Patel MM, Bhat HK: Differential oxidant potential of carcinogenic and weakly carcinogenic estrogens: Involvement of metabolic activation and cytochrome P450. Journal of biochemical and molecular toxicology 2004, 18(1):37-42. 76. Cavalieri EL, Rogan EG: A unifying mechanism in the initiation of cancer and other diseases by catechol quinones. Annals of the New York Academy of Sciences 2004, 1028:247-257. 77. Liehr JG, Fang WF, Sirbasku DA, Ari-Ulubelen A: Carcinogenicity of catechol estrogens in Syrian hamsters. Journal of steroid biochemistry 1986, 24(1):353-356. 78. Li JJ, Li SA: Estrogen carcinogenesis in Syrian hamster tissues: role of metabolism. Federation proceedings 1987, 46(5):1858-1863. 79. Badawi AF, Cavalieri EL, Rogan EG: Role of human cytochrome P450 1A1, 1A2, 1B1, and 3A4 in the 2-, 4-, and 16alpha-hydroxylation of 17beta-estradiol. Metabolism: clinical and experimental 2001, 50(9):1001-1003. 80. Brueggemeier RW, Bhat AS, Lovely CJ, Coughenour HD, Joomprabutra S, Weitzel DH, Vandre DD, Yusuf F, Burak WE, Jr.: 2-Methoxymethylestradiol: a new 2- methoxy estrogen analog that exhibits antiproliferative activity and alters tubulin dynamics. The Journal of steroid biochemistry and molecular biology 2001, 78(2):145-156. 81. LaVallee TM, Zhan XH, Herbstritt CJ, Kough EC, Green SJ, Pribluda VS: 2- Methoxyestradiol inhibits proliferation and induces apoptosis independently of estrogen receptors alpha and beta. Cancer research 2002, 62(13):3691-3697. 82. Cao K, Stack DE, Ramanathan R, Gross ML, Rogan EG, Cavalieri EL: Synthesis and structure elucidation of estrogen quinones conjugated with cysteine, N- acetylcysteine, and glutathione. Chemical research in toxicology 1998, 11(8):909- 916. 83. Yager JD: Endogenous estrogens as carcinogens through metabolic activation. Journal of the National Cancer Institute Monographs 2000(27):67-73. 84. Pearce ST, Jordan VC: The biological role of estrogen receptors alpha and beta in cancer. Critical reviews in oncology/hematology 2004, 50(1):3-22. 85. Walter P, Green S, Greene G, Krust A, Bornert JM, Jeltsch JM, Staub A, Jensen E, Scrace G, Waterfield M et al: Cloning of the human estrogen receptor cDNA. Proceedings of the National Academy of Sciences of the United States of America 1985, 82(23):7889-7893. 190

86. Greene GL, Gilna P, Waterfield M, Baker A, Hort Y, Shine J: Sequence and expression of human estrogen receptor complementary DNA. Science 1986, 231(4742):1150-1154. 87. Green S, Walter P, Kumar V, Krust A, Bornert JM, Argos P, Chambon P: Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A. Nature 1986, 320(6058):134-139. 88. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA: Cloning of a novel receptor expressed in rat prostate and ovary. Proceedings of the National Academy of Sciences of the United States of America 1996, 93(12):5925-5930. 89. Mosselman S, Polman J, Dijkema R: ER beta: identification and characterization of a novel human estrogen receptor. FEBS letters 1996, 392(1):49-53. 90. Enmark E, Pelto-Huikko M, Grandien K, Lagercrantz S, Lagercrantz J, Fried G, Nordenskjold M, Gustafsson JA: Human estrogen receptor beta-gene structure, chromosomal localization, and expression pattern. The Journal of clinical endocrinology and metabolism 1997, 82(12):4258-4265. 91. Ogawa S, Inoue S, Watanabe T, Hiroi H, Orimo A, Hosoi T, Ouchi Y, Muramatsu M: The complete primary structure of human estrogen receptor beta (hER beta) and its heterodimerization with ER alpha in vivo and in vitro. Biochemical and biophysical research communications 1998, 243(1):122-126. 92. Tremblay GB, Tremblay A, Copeland NG, Gilbert DJ, Jenkins NA, Labrie F, Giguere V: Cloning, chromosomal localization, and functional analysis of the murine estrogen receptor beta. Molecular endocrinology 1997, 11(3):353-365. 93. Nilsson S, Makela S, Treuter E, Tujague M, Thomsen J, Andersson G, Enmark E, Pettersson K, Warner M, Gustafsson JA: Mechanisms of estrogen action. Physiological reviews 2001, 81(4):1535-1565. 94. Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA: Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 1997, 138(3):863-870. 95. Gustafsson JA: Estrogen receptor beta--a new dimension in estrogen mechanism of action. The Journal of endocrinology 1999, 163(3):379-383. 96. Taylor AH, Al-Azzawi F: Immunolocalisation of oestrogen receptor beta in human tissues. Journal of molecular endocrinology 2000, 24(1):145-155. 97. Kumar V, Green S, Staub A, Chambon P: Localisation of the oestradiol-binding and putative DNA-binding domains of the human oestrogen receptor. The EMBO journal 1986, 5(9):2231-2236. 98. Kumar V, Green S, Stack G, Berry M, Jin JR, Chambon P: Functional domains of the human estrogen receptor. Cell 1987, 51(6):941-951. 99. Kumar V, Chambon P: The estrogen receptor binds tightly to its responsive element as a ligand-induced homodimer. Cell 1988, 55(1):145-156. 100. Osborne CK: Tamoxifen in the treatment of breast cancer. The New England journal of medicine 1998, 339(22):1609-1618. 101. Dhingra K: Antiestrogens--tamoxifen, SERMs and beyond. Investigational new drugs 1999, 17(3):285-311. 102. Love RR, Mazess RB, Barden HS, Epstein S, Newcomb PA, Jordan VC, Carbone PP, DeMets DL: Effects of tamoxifen on bone mineral density in postmenopausal

191

women with breast cancer. The New England journal of medicine 1992, 326(13):852-856. 103. Kristensen B, Ejlertsen B, Dalgaard P, Larsen L, Holmegaard SN, Transbol I, Mouridsen HT: Tamoxifen and bone metabolism in postmenopausal low-risk breast cancer patients: a randomized study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 1994, 12(5):992-997. 104. Fisher B, Costantino JP, Redmond CK, Fisher ER, Wickerham DL, Cronin WM: Endometrial cancer in tamoxifen-treated breast cancer patients: findings from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14. Journal of the National Cancer Institute 1994, 86(7):527-537. 105. Fornander T, Rutqvist LE, Cedermark B, Glas U, Mattsson A, Silfversward C, Skoog L, Somell A, Theve T, Wilking N et al: Adjuvant tamoxifen in early breast cancer: occurrence of new primary cancers. Lancet 1989, 1(8630):117-120. 106. Jordan VC, Collins MM, Rowsby L, Prestwich G: A monohydroxylated metabolite of tamoxifen with potent antioestrogenic activity. The Journal of endocrinology 1977, 75(2):305-316. 107. Osborne CK, Zhao H, Fuqua SA: Selective estrogen receptor modulators: structure, function, and clinical use. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 2000, 18(17):3172-3186. 108. Kangas L, Nieminen AL, Blanco G, Gronroos M, Kallio S, Karjalainen A, Perila M, Sodervall M, Toivola R: A new triphenylethylene compound, Fc-1157a. II. Antitumor effects. Cancer chemotherapy and pharmacology 1986, 17(2):109-113. 109. Kangas L: Review of the pharmacological properties of toremifene. Journal of steroid biochemistry 1990, 36(3):191-195. 110. O'Regan RM, Cisneros A, England GM, MacGregor JI, Muenzner HD, Assikis VJ, Bilimoria MM, Piette M, Dragan YP, Pitot HC et al: Effects of the antiestrogens tamoxifen, toremifene, and ICI 182,780 on endometrial cancer growth. Journal of the National Cancer Institute 1998, 90(20):1552-1558. 111. Tomas E, Kauppila A, Blanco G, Apaja-Sarkkinen M, Laatikainen T: Comparison between the effects of tamoxifen and toremifene on the uterus in postmenopausal breast cancer patients. Gynecologic oncology 1995, 59(2):261-266. 112. McCague R, Leclercq G, Legros N, Goodman J, Blackburn GM, Jarman M, Foster AB: Derivatives of tamoxifen. Dependence of antiestrogenicity on the 4- substituent. Journal of medicinal chemistry 1989, 32(12):2527-2533. 113. McCague R, Parr IB, Haynes BP: Metabolism of the 4-iodo derivative of tamoxifen by isolated rat hepatocytes. Demonstration that the iodine atom reduces metabolic conversion and identification of four metabolites. Biochemical pharmacology 1990, 40(10):2277-2283. 114. Chander SK, McCague R, Luqmani Y, Newton C, Dowsett M, Jarman M, Coombes RC: Pyrrolidino-4-iodotamoxifen and 4-iodotamoxifen, new analogues of the antiestrogen tamoxifen for the treatment of breast cancer. Cancer research 1991, 51(21):5851-5858. 115. Johnston SR, Riddler S, Haynes BP, A'Hern R, Smith IE, Jarman M, Dowsett M: The novel anti-oestrogen idoxifene inhibits the growth of human MCF-7 breast cancer xenografts and reduces the frequency of acquired anti-oestrogen resistance. British journal of cancer 1997, 75(6):804-809.

192

116. Hasmann M, Rattel B, Loser R: Preclinical data for Droloxifene. Cancer letters 1994, 84(2):101-116. 117. Koh J, Kubota T, Asanuma F, Yamada Y, Kawamura E, Hosoda Y, Hashimoto M, Yamamoto O, Sakai S, Maeda K et al: Antitumor effect of triphenylethylene derivative (TAT-59) against human breast carcinoma xenografts in nude mice. Journal of surgical oncology 1992, 51(4):254-258. 118. Iino Y, Takai Y, Ando T, Ohwada S, Yokoe T, Sugamata N, Takei H, Horiguchi J, Iijima K, Morishita Y: A new triphenylethylene derivative, TAT-59; hormone receptors; insulin-like growth factor 1; and growth suppression of hormone- dependent MCF-7 tumors in athymic mice. Cancer chemotherapy and pharmacology 1994, 34(5):372-376. 119. Black LJ, Jones CD, Falcone JF: Antagonism of estrogen action with a new benzothiophene derived antiestrogen. Life sciences 1983, 32(9):1031-1036. 120. Yang G, Nowsheen S, Aziz K, Georgakilas AG: Toxicity and adverse effects of Tamoxifen and other anti-estrogen drugs. Pharmacology & therapeutics 2013, 139(3):392-404. 121. Cummings SR, Eckert S, Krueger KA, Grady D, Powles TJ, Cauley JA, Norton L, Nickelsen T, Bjarnason NH, Morrow M et al: The effect of raloxifene on risk of breast cancer in postmenopausal women: results from the MORE randomized trial. Multiple Outcomes of Raloxifene Evaluation. Jama 1999, 281(23):2189- 2197. 122. Wakeling AE, Bowler J: Steroidal pure antioestrogens. The Journal of endocrinology 1987, 112(3):R7-10. 123. wakeling AE: A new approach to breast cancer therapy: Total estrogen ablation with pure antiestrogens, in Long-Term Tamoxifen Treatment for Breast Cancer (Jordan VC ed). University of Wisconsin Press, Madison 1994:pp 219–234. 124. Wakeling AE, Dukes M, Bowler J: A potent specific pure antiestrogen with clinical potential. Cancer research 1991, 51(15):3867-3873. 125. Wakeling AE: Use of pure antioestrogens to elucidate the mode of action of oestrogens. Biochemical pharmacology 1995, 49(11):1545-1549. 126. Fawell SE, White R, Hoare S, Sydenham M, Page M, Parker MG: Inhibition of estrogen receptor-DNA binding by the "pure" antiestrogen ICI 164,384 appears to be mediated by impaired receptor dimerization. Proceedings of the National Academy of Sciences of the United States of America 1990, 87(17):6883-6887. 127. Dauvois S, Danielian PS, White R, Parker MG: Antiestrogen ICI 164,384 reduces cellular estrogen receptor content by increasing its turnover. Proceedings of the National Academy of Sciences of the United States of America 1992, 89(9):4037- 4041. 128. Dauvois S, White R, Parker MG: The antiestrogen ICI 182780 disrupts estrogen receptor nucleocytoplasmic shuttling. Journal of cell science 1993, 106 ( Pt 4):1377-1388. 129. Lee EY, Muller WJ: Oncogenes and tumor suppressor genes. Cold Spring Harbor perspectives in biology 2010, 2(10):a003236. 130. Mahadev K, Raval G, Bharadwaj S, Willingham MC, Lange EM, Vonderhaar B, Salomon D, Prasad GL: Suppression of the transformed phenotype of breast cancer by tropomyosin-1. Experimental cell research 2002, 279(1):40-51.

193

131. Shields JM, Mehta H, Pruitt K, Der CJ: Opposing roles of the extracellular signal- regulated kinase and p38 mitogen-activated protein kinase cascades in Ras- mediated downregulation of tropomyosin. Molecular and cellular biology 2002, 22(7):2304-2317. 132. Franzen B, Linder S, Uryu K, Alaiya AA, Hirano T, Kato H, Auer G: Expression of tropomyosin isoforms in benign and malignant human breast lesions. British journal of cancer 1996, 73(7):909-913. 133. Gunning P, O'Neill G, Hardeman E: Tropomyosin-based regulation of the actin cytoskeleton in time and space. Physiological reviews 2008, 88(1):1-35. 134. Barua B, Pamula MC, Hitchcock-DeGregori SE: Evolutionarily conserved surface residues constitute actin binding sites of tropomyosin. Proceedings of the National Academy of Sciences of the United States of America 2011, 108(25):10150-10155. 135. Cranz-Mileva S, Pamula MC, Barua B, Desai B, Hong YH, Russell J, Trent R, Wang J, Walworth NC, Hitchcock-DeGregori SE: A molecular evolution approach to study the roles of tropomyosin in fission yeast. PLoS One 2013, 8(10):e76726. 136. Lin JJ, Helfman DM, Hughes SH, Chou CS: Tropomyosin isoforms in chicken embryo fibroblasts: purification, characterization, and changes in Rous sarcoma virus-transformed cells. J Cell Biol 1985, 100(3):692-703. 137. Choi C, Kim D, Kim S, Jeong S, Song E, Helfman DM: From skeletal muscle to cancer: insights learned elucidating the function of tropomyosin. J Struct Biol 2012, 177(1):63-69. 138. Helfman DM, Flynn P, Khan P, Saeed A: Tropomyosin as a regulator of cancer cell transformation. Advances in experimental medicine and biology 2008, 644:124- 131. 139. Stehn JR, Haass NK, Bonello T, Desouza M, Kottyan G, Treutlein H, Zeng J, Nascimento PR, Sequeira VB, Butler TL et al: A novel class of anticancer compounds targets the actin cytoskeleton in tumor cells. Cancer research 2013, 73(16):5169-5182. 140. Cooper JA: Actin dynamics: tropomyosin provides stability. Current biology : CB 2002, 12(15):R523-525. 141. Gordon AM, Homsher E, Regnier M: Regulation of contraction in striated muscle. Physiological reviews 2000, 80(2):853-924. 142. Perry SV: Vertebrate tropomyosin: distribution, properties and function. Journal of muscle research and cell motility 2001, 22(1):5-49. 143. Vindin H, Gunning P: Cytoskeletal tropomyosins: choreographers of actin filament functional diversity. Journal of muscle research and cell motility 2013, 34(3-4):261-274. 144. Schevzov G, Whittaker SP, Fath T, Lin JJ, Gunning PW: Tropomyosin isoforms and reagents. Bioarchitecture 2011, 1(4):135-164. 145. Geeves MA, Hitchcock-DeGregori SE, Gunning PW: A systematic nomenclature for mammalian tropomyosin isoforms. Journal of muscle research and cell motility 2015, 36(2):147-153. 146. Eyre H, Akkari PA, Wilton SD, Callen DC, Baker E, Laing NG: Assignment of the human skeletal muscle alpha-tropomyosin gene (TPM1) to band 15q22 by fluorescence in situ hybridization. Cytogenetics and cell genetics 1995, 69(1-2):15- 17.

194

147. Hunt CC, Eyre HJ, Akkari PA, Meredith C, Dorosz SM, Wilton SD, Callen DF, Laing NG, Baker E: Assignment of the human beta tropomyosin gene (TPM2) to band 9p13 by fluorescence in situ hybridisation. Cytogenetics and cell genetics 1995, 71(1):94-95. 148. Wilton SD, Eyre H, Akkari PA, Watkins HC, MacRae C, Laing NG, Callen DC: Assignment of the human a-tropomyosin gene TPM3 to 1q22-->q23 by fluorescence in situ hybridisation. Cytogenetics and cell genetics 1995, 68(1- 2):122-124. 149. Wilton SD, Lim L, Dorosz SD, Gunn HC, Eyre HJ, Callen DF, Laing NG: Assignment of the human alpha-tropomyosin gene TPM4 to band 19p13.1 by fluorescence in situ hybridization. Cytogenetics and cell genetics 1996, 72(4):294- 296. 150. Hendricks M, Weintraub H: Tropomyosin is decreased in transformed cells. Proceedings of the National Academy of Sciences of the United States of America 1981, 78(9):5633-5637. 151. Takenaga K, Nakamura Y, Sakiyama S: Suppression of synthesis of tropomyosin isoform 2 in metastatic v-Ha-ras-transformed NIH3T3 cells. Biochemical and biophysical research communications 1988, 157(3):1111-1116. 152. Lin JL, Geng X, Bhattacharya SD, Yu JR, Reiter RS, Sastri B, Glazier KD, Mirza ZK, Wang KK, Amenta PS et al: Isolation and sequencing of a novel tropomyosin isoform preferentially associated with colon cancer. Gastroenterology 2002, 123(1):152-162. 153. Prasad GL, Fuldner RA, Cooper HL: Expression of transduced tropomyosin 1 cDNA suppresses neoplastic growth of cells transformed by the ras oncogene. Proceedings of the National Academy of Sciences of the United States of America 1993, 90(15):7039-7043. 154. Janssen RA, Mier JW: Tropomyosin-2 cDNA lacking the 3' untranslated region riboregulator induces growth inhibition of v-Ki-ras-transformed fibroblasts. Molecular biology of the cell 1997, 8(5):897-908. 155. Yager ML, Hughes JA, Lovicu FJ, Gunning PW, Weinberger RP, O'Neill GM: Functional analysis of the actin-binding protein, tropomyosin 1, in neuroblastoma. British journal of cancer 2003, 89(5):860-863. 156. Hughes JA, Cooke-Yarborough CM, Chadwick NC, Schevzov G, Arbuckle SM, Gunning P, Weinberger RP: High-molecular-weight tropomyosins localize to the contractile rings of dividing CNS cells but are absent from malignant pediatric and adult CNS tumors. Glia 2003, 42(1):25-35. 157. Wang FL, Wang Y, Wong WK, Liu Y, Addivinola FJ, Liang P, Chen LB, Kantoff PW, Pardee AB: Two differentially expressed genes in normal human prostate tissue and in carcinoma. Cancer research 1996, 56(16):3634-3637. 158. Varga AE, Stourman NV, Zheng Q, Safina AF, Quan L, Li X, Sossey-Alaoui K, Bakin AV: Silencing of the Tropomyosin-1 gene by DNA methylation alters tumor suppressor function of TGF-beta. Oncogene 2005, 24(32):5043-5052. 159. Pawlak G, McGarvey TW, Nguyen TB, Tomaszewski JE, Puthiyaveettil R, Malkowicz SB, Helfman DM: Alterations in tropomyosin isoform expression in human transitional cell carcinoma of the urinary bladder. International journal of cancer Journal international du cancer 2004, 110(3):368-373.

195

160. Galloway PG, Likavec MJ, Perry G: Tropomyosin isoform expression in normal and neoplastic astrocytes. Laboratory investigation; a journal of technical methods and pathology 1990, 62(2):163-170. 161. Urbancikova M, Hitchcock-DeGregori SE: Requirement of amino-terminal modification for striated muscle alpha-tropomyosin function. The Journal of biological chemistry 1994, 269(39):24310-24315. 162. Pittenger MF, Kistler A, Helfman DM: Alternatively spliced exons of the beta tropomyosin gene exhibit different affinities for F-actin and effects with nonmuscle caldesmon. Journal of cell science 1995, 108 ( Pt 10):3253-3265. 163. Naga Prasad SV, Jayatilleke A, Madamanchi A, Rockman HA: Protein kinase activity of phosphoinositide 3-kinase regulates beta-adrenergic receptor endocytosis. Nature cell biology 2005, 7(8):785-796. 164. Mak A, Smillie LB, Barany M: Specific phosphorylation at serine-283 of alpha tropomyosin from frog skeletal and rabbit skeletal and cardiac muscle. Proceedings of the National Academy of Sciences of the United States of America 1978, 75(8):3588-3592. 165. Somara S, Pang H, Bitar KN: Agonist-induced association of tropomyosin with protein kinase Calpha in colonic smooth muscle. American journal of physiology Gastrointestinal and liver physiology 2005, 288(2):G268-276. 166. Schulz EM, Wieczorek DF: Tropomyosin de-phosphorylation in the heart: what are the consequences? Journal of muscle research and cell motility 2013, 34(3- 4):239-246. 167. Marston S, Gautel M: Introducing a special edition of the Journal of Muscle Research and Cell Motility on tropomyosin: form and function. Journal of muscle research and cell motility 2013, 34(3-4):151-153. 168. Simoneau B, Houle F, Huot J: Regulation of endothelial permeability and transendothelial migration of cancer cells by tropomyosin-1 phosphorylation. Vascular cell 2012, 4(1):18. 169. Houle F, Poirier A, Dumaresq J, Huot J: DAP kinase mediates the phosphorylation of tropomyosin-1 downstream of the ERK pathway, which regulates the formation of stress fibers in response to oxidative stress. J Cell Sci 2007, 120(Pt 20):3666-3677. 170. Houle F, Huot J: Dysregulation of the endothelial cellular response to oxidative stress in cancer. Molecular carcinogenesis 2006, 45(6):362-367. 171. Houle F, Rousseau S, Morrice N, Luc M, Mongrain S, Turner CE, Tanaka S, Moreau P, Huot J: Extracellular signal-regulated kinase mediates phosphorylation of tropomyosin-1 to promote cytoskeleton remodeling in response to oxidative stress: impact on membrane blebbing. Molecular biology of the cell 2003, 14(4):1418-1432. 172. Bharadwaj S, Prasad GL: Tropomyosin-1, a novel suppressor of cellular transformation is downregulated by promoter methylation in cancer cells. Cancer letters 2002, 183(2):205-213. 173. Zhu S, Si ML, Wu H, Mo YY: MicroRNA-21 targets the tumor suppressor gene tropomyosin 1 (TPM1). The Journal of biological chemistry 2007, 282(19):14328- 14336. 174. Chan JA, Krichevsky AM, Kosik KS: MicroRNA-21 is an antiapoptotic factor in human glioblastoma cells. Cancer research 2005, 65(14):6029-6033. 196

175. Si ML, Zhu S, Wu H, Lu Z, Wu F, Mo YY: miR-21-mediated tumor growth. Oncogene 2007, 26(19):2799-2803. 176. Deroo BJ, Korach KS: Estrogen receptors and human disease. The Journal of clinical investigation 2006, 116(3):561-570. 177. Frasor J, Chang EC, Komm B, Lin CY, Vega VB, Liu ET, Miller LD, Smeds J, Bergh J, Katzenellenbogen BS: Gene expression preferentially regulated by tamoxifen in breast cancer cells and correlations with clinical outcome. Cancer research 2006, 66(14):7334-7340. 178. Miller WR, Hawkins RA, Forrest AP: Significance of aromatase activity in human breast cancer. Cancer research 1982, 42(8 Suppl):3365s-3368s. 179. Miller WR: Aromatase activity in breast tissue. The Journal of steroid biochemistry and molecular biology 1991, 39(5B):783-790. 180. Evans TR, Rowlands MG, Law M, Coombes RC: Intratumoral oestrone sulphatase activity as a prognostic marker in human breast carcinoma. British journal of cancer 1994, 69(3):555-561. 181. Gangloff A, Shi R, Nahoum V, Lin SX: Pseudo-symmetry of C19 steroids, alternative binding orientations, and multispecificity in human estrogenic 17beta-hydroxysteroid dehydrogenase. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2003, 17(2):274-276. 182. Nahoum V, Gangloff A, Shi R, Lin SX: How estrogen-specific proteins discriminate estrogens from androgens: a common steroid binding site architecture. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2003, 17(10):1334-1336. 183. Aka JA, Mazumdar M, Chen CQ, Poirier D, Lin SX: 17beta-hydroxysteroid dehydrogenase type 1 stimulates breast cancer by dihydrotestosterone inactivation in addition to estradiol production. Molecular endocrinology 2010, 24(4):832-845. 184. Lin SX, Poirier D, Adamski J: A challenge for medicinal chemistry by the 17beta- hydroxysteroid dehydrogenase superfamily: an integrated biological function and inhibition study. Curr Top Med Chem 2013, 13(10):1164-1171. 185. Boutros R, Bailey AM, Wilson SH, Byrne JA: Alternative splicing as a mechanism for regulating 14-3-3 binding: interactions between hD53 (TPD52L1) and 14-3-3 proteins. Journal of molecular biology 2003, 332(3):675-687. 186. Boutros R, Byrne JA: D53 (TPD52L1) is a cell cycle-regulated protein maximally expressed at the G2-M transition in breast cancer cells. Experimental cell research 2005, 310(1):152-165. 187. Aka JA, Zerradi M, Houle F, Huot J, Lin SX: 17beta-hydroxysteroid dehydrogenase type 1 modulates breast cancer protein profile and impacts cell migration. Breast cancer research : BCR 2012, 14(3):R92. 188. Silverstein RL, Febbraio M: CD36, a scavenger receptor involved in immunity, metabolism, angiogenesis, and behavior. Science signaling 2009, 2(72):re3. 189. Primo L, Ferrandi C, Roca C, Marchio S, di Blasio L, Alessio M, Bussolino F: Identification of CD36 molecular features required for its in vitro angiostatic activity. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2005, 19(12):1713-1715. 190. Crum CP, McKeon FD: p63 in epithelial survival, germ cell surveillance, and neoplasia. Annual review of pathology 2010, 5:349-371. 197

191. Adorno M, Cordenonsi M, Montagner M, Dupont S, Wong C, Hann B, Solari A, Bobisse S, Rondina MB, Guzzardo V et al: A Mutant-p53/Smad complex opposes p63 to empower TGFbeta-induced metastasis. Cell 2009, 137(1):87-98. 192. Muller PA, Caswell PT, Doyle B, Iwanicki MP, Tan EH, Karim S, Lukashchuk N, Gillespie DA, Ludwig RL, Gosselin P et al: Mutant p53 drives invasion by promoting integrin recycling. Cell 2009, 139(7):1327-1341. 193. Su X, Chakravarti D, Cho MS, Liu L, Gi YJ, Lin YL, Leung ML, El-Naggar A, Creighton CJ, Suraokar MB et al: TAp63 suppresses metastasis through coordinate regulation of Dicer and miRNAs. Nature 2010, 467(7318):986-990. 194. McConathy WJ, Alaupovic P: Isolation and partial characterization of apolipoprotein D: a new protein moiety of the human plasma lipoprotein system. FEBS letters 1973, 37(2):178-182. 195. Provost PR, Weech PK, Tremblay NM, Marcel YL, Rassart E: Molecular characterization and differential mRNA tissue distribution of rabbit apolipoprotein D. Journal of lipid research 1990, 31(11):2057-2065. 196. Aspinall JO, Bentel JM, Horsfall DJ, Haagensen DE, Marshall VR, Tilley WD: Differential expression of apolipoprotein-D and prostate specific antigen in benign and malignant prostate tissues. The Journal of urology 1995, 154(2 Pt 1):622-628. 197. Soiland H, Skaland I, Janssen EA, Gudlaugsson E, Korner H, Varhaug JE, Soreide JA, Baak JP: Comparison of apolipoprotein D determination methods in breast cancer. Anticancer Res 2008, 28(2B):1151-1160. 198. Soiland H, Soreide K, Janssen EA, Korner H, Baak JP, Soreide JA: Emerging concepts of apolipoprotein D with possible implications for breast cancer. Cellular oncology : the official journal of the International Society for Cellular Oncology 2007, 29(3):195-209. 199. 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. British journal of cancer 2000, 82(3):518-523. 200. Zhang CY, Chen J, Yin DC, Lin SX: The contribution of 17beta-hydroxysteroid dehydrogenase type 1 to the estradiol-estrone ratio in estrogen-sensitive breast cancer cells. PLoS One 2012, 7(1):e29835. 201. Ricketts D, Turnbull L, Ryall G, Bakhshi R, Rawson NS, Gazet JC, Nolan C, Coombes RC: Estrogen and progesterone receptors in the normal female breast. Cancer research 1991, 51(7):1817-1822. 202. Russo J, Russo IH: The role of estrogen in the initiation of breast cancer. The Journal of steroid biochemistry and molecular biology 2006, 102(1-5):89-96. 203. Russo J, Fernandez SV, Russo PA, Fernbaugh R, Sheriff FS, Lareef HM, Garber J, Russo IH: 17-Beta-estradiol induces transformation and tumorigenesis in human breast epithelial cells. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 2006, 20(10):1622-1634. 204. Doisneau-Sixou SF, Cestac P, Chouini S, Carroll JS, Hamilton AD, Sebti SM, Poirot M, Balaguer P, Faye JC, Sutherland RL et al: Contrasting effects of prenyltransferase inhibitors on estrogen-dependent cell cycle progression and estrogen receptor-mediated transcriptional activity in MCF-7 cells. Endocrinology 2003, 144(3):989-998. 198

205. Doisneau-Sixou SF, Cestac P, Faye JC, Favre G, Sutherland RL: Additive effects of tamoxifen and the farnesyl transferase inhibitor FTI-277 on inhibition of MCF-7 breast cancer cell-cycle progression. International journal of cancer Journal international du cancer 2003, 106(5):789-798. 206. Day JM, Foster PA, Tutill HJ, Parsons MF, Newman SP, Chander SK, Allan GM, Lawrence HR, Vicker N, Potter BV et al: 17beta-hydroxysteroid dehydrogenase Type 1, and not Type 12, is a target for endocrine therapy of hormone- dependent breast cancer. International journal of cancer Journal international du cancer 2008, 122(9):1931-1940. 207. Poulin R, Labrie F: Stimulation of cell proliferation and estrogenic response by adrenal C19-delta 5-steroids in the ZR-75-1 human breast cancer cell line. Cancer research 1986, 46(10):4933-4937. 208. Aspinall SR, Stamp S, Davison A, Shenton BK, Lennard TW: The proliferative effects of 5-androstene-3 beta,17 beta-diol and 5 alpha-dihydrotestosterone on cell cycle analysis and cell proliferation in MCF7, T47D and MDAMB231 breast cancer cell lines. The Journal of steroid biochemistry and molecular biology 2004, 88(1):37-51. 209. Gunning P: Emerging issues for tropomyosin structure, regulation, function and pathology. Adv Exp Med Biol 2008, 644:293-298. 210. Shah V, Braverman R, Prasad GL: Suppression of neoplastic transformation and regulation of cytoskeleton by tropomyosins. Somat Cell Mol Genet 1998, 24(5):273-280. 211. Gout S, Tremblay PL, Huot J: Selectins and selectin ligands in extravasation of cancer cells and organ selectivity of metastasis. Clin Exp Metastasis 2008, 25(4):335-344. 212. Mori S, Chang JT, Andrechek ER, Matsumura N, Baba T, Yao G, Kim JW, Gatza M, Murphy S, Nevins JR: Anchorage-independent cell growth signature identifies tumors with metastatic potential. Oncogene 2009, 28(31):2796-2805.

199

200

Annexe

201

202

Annexe 1: Functional proteomic analyses of breast cancer cells: effects of the modulation of 17beta-hydroxysteroid dehydrogenase type 1 expression on protein and transcript profiles

203

204

Functional proteomic analyses of breast cancer cells: effects of the modulation of 17beta-hydroxysteroid dehydrogenase type 1 expression on

protein and transcript profiles

Juliette Adjo AKA1, Mouna Zerradi1, François Houle2, Jacques Huot2 and Sheng-Xiang

Lin1*

1Molecular Endocrinology and Genomics Research Centre, CHUL (CHUQ) Research Centre and Laval University, Québec, Canada G1V 4G2; and 2Cancer Research Centre of Laval University, Québec, Canada G1R 2J6

*Corresponding author: Sheng-Xiang Lin, Molecular Endocrinology and Genomics axe, CHUL Research Center, 2705 Boulevard Laurier, Sainte-Foy, Québec, Canada G1V 4G2. Tel.: 418 654 2296; Fax: 418 654 2761; E-mail: [email protected]

RUNNING TITLE: Effect of 17β-HSD1 expression on breast cancer proteome.

KEYWORDS: 17β-HSD1, Breast cancer, Cell migration, Gene expression modulation, Proteomic profile.

Cet article est publié dans le journal Breast Cancer Research; 14(3):R92

205

Abstract

Introduction. Human 17beta-hydroxysteroid dehydrogenase type 1 (17β-HSD1) is a steroid- converting enzyme that has long been known to play critical roles in estradiol synthesis and more recently in dihydrotestosterone (DHT) inactivation, showing a dual function that promotes breast cancer cell proliferation. Previously, we reported the first observation of the influence of the enzyme on endogenous estrogen-responsive gene expression. Here, we demonstrate the impact of 17β-HSD1 expression on the breast cancer cell proteome and investigate its role in cell migration.

Methods. 17β-HSD1 was stably transfected in MCF7 cells and the proteome of the generated cells overexpressing 17β-HSD1 (MCF7-17βHSD1 cells) was compared to that of the wild type MCF7 cells. Proteomics study was performed using two-dimensional gel electrophoresis followed by mass spectrometry analysis of differentially expressed protein spots. Reverse transcription quantitative real-time PCR (RT-qPCR) was used to investigate the transcription of individual gene. The effect of 17β-HSD1 on MCF7 cell migration was verified by a wound-healing assay.

Results. Proteomic data demonstrate that the expression of more than 59 proteins is modulated following 17β-HSD1 overexpression. 17β-HSD1 regulates the expression of important genes and proteins that are relevant to cell growth control, such as BRCA2 and CDKN1A interacting protein (BCCIP) and proliferating cell nuclear antigen (PCNA) which are down- and up-regulated in MCF7-17βHSD1 cells, respectively. RT-qPCR data reveal that 17β-HSD1 increases the mRNA levels of estrogen receptors (ER) alpha and beta by 171 and 120%, respectively, while decreasing that of the androgen receptor by 64%. Interestingly, 17β-HSD1 increases the mRNA transcript (by 3.6 times) and the protein expression of the metastasis suppressor gene nm23-H1 and the expression of the two enzymes are closely correlated. We have further shown that 17β-HSD1 expression is associated with an increase of MCF7 cell migration.

206

Conclusions. In addition to the regulation of important genes, we have demonstrated for the first time that 17β-HSD1 increases breast cancer cell migration, in spite of its positive regulation of the antimetastatic gene NM23. This is also correlated to its stimulation of breast cancer cell growth, further confirming its targeting in ER positive breast cancer. The novel findings in this study suggest several directions for future research on the contribution of 17β-HSD1 to breast cancer progression and related treatment.

207

1. Introduction

Breast cancer is the most frequent cancer affecting women. The malignancy accounts for about 1 in 10 cancers in the world and is diagnosed in one million women each year [1, 2]. In North America, breast cancer is the second most important cause of death by cancer in women, after lung cancer, and the leading cause of cancer death among those between 20 and 59 years of age [3, 4]. After increasing through the 80s and 90s, breast cancer incidence rates showed a welcome decrease of 3.5% per year from 2001 to 2004 and the mortality rate decreased by 1.9% per year in the United States between 1998 and 2006 [3, 5]. This reflects an improvement in the diagnosis and treatment of the disease, yet it remains of prime importance. Epidemiological evidence indicates that most breast cancer risk factors are associated with prolonged exposure of the mammary gland to high levels of estradiol (E2). This potent estrogen plays a crucial role in the development and evolution of hormone-dependent breast cancer [6]. About 60% of pre-menopausal and 75% of post-menopausal breast cancer patients show a hormone dependency [7]. The final steps of E2 biosynthesis implicate two principal pathways in breast cancer tissue: the aromatase pathway transforms androgens into estrogens, and the sulfatase pathway converts the inactive hormones estrone sulfate (E1S) and dehydroepiandrosterone sulfate (DHEA-S) into estrone (E1) and dehydroepiandrosterone (DHEA), respectively, via the action of steroid sulfatase (STS). Different enzymes further convert DHEA to 5α-androstane-3β, 17β-diol (A-diol) and the latter to testosterone which can in turn be converted to E2 by aromatase. The inactive E1, synthesized by both STS and aromatase, is converted to the potent E2 by the action of reductive 17beta-hydroxysteroid dehydrogenases (17β-HSDs) [8–10]. In breast cancer cells, E2 is principally synthesized by 17β-HSD type 1 (17β-HSD1), with NADPH as cofactor. Previously, we reported the dual function of 17β-HSD1 in estradiol synthesis and DHT inactivation stimulating cell proliferation [11]. Analyses of 17β-HSD1 mRNA expression in breast carcinoma specimens from patients revealed that high expression of the enzyme correlates with a weak prognosis for breast cancer [12–14]. Despite these observations, the relationship between 17β-HSD1 expression and that of genes and proteins involved in breast cancer cell growth has not been established. The aim of the present study was to investigate the impact of 17β-HSD1

208

overexpression on the protein profile of breast cancer cells. The MCF7 cell line is a human hormone-dependent breast cancer cell line widely used for breast cancer studies that expresses both estrogen and androgen receptors. Since the cell line barely expresses endogenous 17β-HSD1 [11, 15], we used it as cell model for the increase of 17β-HSD1 expression. The proteomic approach using two-dimensional (2-D) gel electrophoresis is the most popular tool to study global changes in protein profile following biological or chemical treatments. We thus used this technique to analyse the proteomic modification of MCF7 cells in response to 17β-HSD1 overexpression. Following the proteomics analysis, reverse transcription quantitative real-time PCR (RT-qPCR) was used to investigate the gene transcription of a number of differentially expressed proteins, such as proliferating cell nuclear antigen (PCNA) and the metastasis suppressor gene nm23-H1. The overexpression experiments, combined with further siRNA knockdown analysis, demonstrated a strong positive correlation between nm23-H1 regulation and 17β-HSD1 expression. We thus hypothesized that 17β-HSD1 could be implicated in breast cancer cell metastasis and evaluated its effect on MCF7 cell migration.

2. Materials and methods

2.1. Cell culture and generation of stably-transfected MCF7-17βHSD1 cells Wild type (wt) MCF7 and T47D cells were cultured as previously described, with MCF7 cell culture medium containing 1 nM β-E2 [11]. Recombinant plasmid containing 17β-HSD1 cDNA and the 17β-HSD1-stably-transfected MCF7 cells (MCF7-17βHSD1 cells) were generated as previously described [11].

2.2. Generation of protein extracts for proteomics analysis

Wt MCF7 and stably transfected (MCF7-17βHSD1) cells were defrosted at the same time and cultured in T75 CM flasks in complete medium containing β-E2. After three passages, cells were plated in 100 x 2 CM dishes and cultured until reaching the desired confluence.

209

For protein sample preparation, cells having reached 80-90% confluence were washed two times with cold PBS 1x and scraped with a rubber policeman in 1.2 mL PBS. Cells were collected in an eppendorf and centrifuged at 3000 rpm for 5 min. The cell pellets were resuspended in 500 L lysis buffer T8 (7 M urea, 2 M thiourea, 3% CHAPS, 20 mM DTT, 5 mM TCEP, 0.5% IPG buffer pH 4-7, 0.25% IPG buffer pH 3-10) containing 50 mM tris-HCl pH 8.8, 1 mM PMSF and 1% protease inhibitors cocktail (EMD Chemicals, Gibbs-town, NJ). Protein samples were precipitated using 2-D Clean-Up Kit (GE Healthcare, Piscataway, NJ) and resolubilized in T8 buffer. The protein samples included three independent biological replicates (coming from three independent cell culture experiments), representing total proteins from each cell line (MCF7 and MCF7-17βHSD1) for a total of six samples. The protein concentrations were determined using the 2-D Quant Kit (GE Healthcare).

2.3. 2-D gel electrophoresis

For the first dimension, 200 µg total protein samples from MCF7 and MCF7-17βHSD1 cells were loaded onto 24-cm pH 4-7 Immobilized pH gradient (IPG) strips (Immobiline DryStrips; GE Healthcare). Strips were rehydrated for 10 hours at 30 volts and isoelectric focusing was performed on an IPGphorII IEF system (GE Healthcare). For the second- dimension SDS-PAGE, focused Immobiline DryStrips were equilibrated twice for 15 min in an equilibration buffer (50 mM tris-HCl pH 8.8, 6 M urea, 30% glycerol, 2% SDS, trace of bromophenol blue) containing 10 mg/mL DTT for the first equilibration and 25 mg/mL iodoacetamide for the second one. Immobiline DryStrips were then transferred onto the surface of a 12% acrylamide gel (20 x 25 x 0.1 cm) and sealed using 0.5% agarose. Gels were run in Ettan DALTtwelse system (GE Healthcare) in a standard tris-glycine SDS-PAGE buffer at 40 mA/gel and 15oC until the tracking dye reached the end of the gel. Three independent protein samples coming from three independent cell culture experiments were run for each cell line. Gels were fixed overnight in 40% methanol, 7% acetic acid, stained with Sypro Ruby (Invitrogen, Burlington, Ontario, Canada) and scanned with the ProXpress CCD scanner (PerkinElmer, Waltham, MA). The 2-D gel electrophoresis was performed on the Proteomic platform of the Infectious Disease Research Center (Québec, Canada).

210

2.4. 2-D gel image analysis

Protein spot detection, spot matching and semi-quantitative statistical analysis were performed using the Progenesis software version PG240 (Nonlinear Dynamics, Durham, NC). For each cell line, three different gel images were analysed and a corresponding synthetic image reference was obtained. After computer matching, detected spots and spot matches were manually edited for more accuracy. A spot had to be present in at least two of the three replicate gels to be considered in the analysis. The detection of protein spots differentially expressed was performed using the t-test (p < 0.05) and INCA volume and proteins that were differentially expressed 2-fold or higher were considered significant. 18 protein spots were selected among the differentially expressed spots and were excised from Sypro Ruby-stained 2-D gels using a ProXcision robot (PerkinElmer) and sent for mass spectrometry (MS) analysis.

2.5. Mass spectrometry and protein identification

MS experiments were performed by the Proteomics platform of the Eastern Quebec Genomics Center (Québec, Canada). Protein spots were washed with water and tryptic digestion was performed on a MassPrep liquid handling robot (Waters, Milford, MA) according to the manufacturer’s specifications and to the protocol of Shevchenko et al. [16] with the modifications suggested by Havlis et al. [17]. Peptide samples (an aliquot of the digested protein sample) were separated by online reversed-phase (RP) nanoscale capillary liquid chromatography (nanoLC) and analyzed by electrospray mass spectrometry (ES MS/MS). The experiments were performed with a Thermo Surveyor MS pump connected to an LTQ linear ion trap mass spectrometer (ThermoFisher, San Jose, CA) equipped with a nanoelectrospray ion source (ThermoFisher). Peptide separation took place on a PicoFrit column BioBasic C18, 10 cm x 0.075 mm internal diameter (New Objective, Woburn, MA) with a linear gradient from 2-50% solvent B (acetonitrile, 0.1% formic acid) in 30 minutes, at 200 nL/min (obtained by flow-splitting). Mass spectra were acquired using a data dependent acquisition mode using Xcalibur software version 2.0. Each full scan mass spectrum (400 to 2000 m/z) was followed by collision-induced dissociation of the seven most intense ions. The

211 dynamic exclusion (30 seconds exclusion duration) function was enabled, and the relative collisional fragmentation energy was set to 35%. All MS/MS samples were analyzed using the Mascot algorithm (Matrix Science, London, UK; version Mascot) and the Uniref100_14_0_Homo_sapiens_9606 database (version with 89892 entries). Mascot was searched with a fragment ion mass tolerance of 0.50 Da and a parent ion tolerance of 2.0 Da. Iodoacetamide derivative of cysteine was specified as a fixed modification and oxidation of methionine was specified as a variable modification. Two missed cleavages were allowed. Scaffold (version Scaffold_2_01_02, Proteome Software Inc., Portland, OR) was used to validate MS/MS based peptide and protein identifications. The protein identification cut off was set at a confidence level of 95% (MASCOT score > 33) with at least two peptides matching to a protein. Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony.

2.6. Reverse transcription quantitative real-time PCR and semiquantitative RT-PCR

Total RNA was isolated from cells using Trizol Reagent (Invitrogen) in 6-well plates and treated with DNase 1. Analysis of the RNA integrities using the Bioanalyzer 2100 (Agilent Technologies, Mississauga, Ontario, Canada) and the RNA 6000 Nano Chips (Agilent) showed good qualities for all the RNA samples with RNA Integrity Numbers (RIN) higher than 8/10. RNA samples for RT-qPCR analyses comprised two biological repetitions for each condition and cell line. mRNA quantifications were performed as previously described [11, 18] with Atp5o, Hprt1 and G6PD genes used as internal controls. The primers used for the amplification and the corresponding cDNA fragments of each mRNA are shown in Supplementary Table 1. The mRNA levels were expressed as mRNA copies/µg total RNA and S.D were < 10% of duplicates. Semiquantitative RT-PCR was carried out and analysed as previously described [11] except for the number of cycles, which was 30.

212

2.7. siRNA synthesis and transfection

The sense and antisense sequences of three 17β-HSD1 siRNAs were selected and synthesized as previously described [11]. Transfection of T47D cells with siRNA was carried out in 6- well plates using Lipofectamine siRNAMax (Invitrogen), 3x105 cells/well and 200 nM mixed 17β-HSD1 specific siRNAs. For cell migration assays, MCF7-17βHSD1 cells were transfected with 100 nM mixed 17β-HSD1 specific siRNAs in 35 x 10 mm2 dishes. Control cells were transfected with scramble (control) siRNA [11].

2.8. Cell migration assay

Cell migration was evaluated by using a wound-healing assay. First, MCF7 and MCF7- 17βHSD1 cells were cultivated in T75 culture flasks in complete growth medium. Cells, at low passage number, were seeded at high density in 35 x 10 mm2 dishes in estradiol-free medium containing 5% fetal bovine serum (FBS). Two days later, a straight scratch was made in triplicate across a confluent monolayer cultures using a p200 micropipette tip. Thereafter, cells were washed five times with fresh estradiol-free medium and were incubated in the same medium. Second, MCF7-17βHSD1 cells were transfected with 17β- HSD1 specific siRNAs or scramble siRNA (control siRNA) in 35 x 10 mm2 dishes in complete growth medium, and were incubated. Forty-eight hours after transfection, a wound was created by manually scraping the cell monolayer as described above. Cells were then washed five times and incubated in estradiol-free medium containing 5% FBS. The movements of cells in the scratched area were monitored by capturing images every 30 minutes for a total duration of 48 hours using the x10 objective lens of a phase-contrast microscope. The scratch widths were measured at specific time points using the freeware ImageJ.

2.9. Western blot

Wt MCF7 and MCF7-17βHSD1 cells were cultured in complete medium containing β-E2 and total proteins were extracted from cells with complete T8 lysis buffer. Equal volumes of 213 proteins were separated by 12% SDS-PAGE and then electroblotted onto nitrocellulose membranes. The membranes were blocked with 5% non-fat milk in PBS-tween 20 (PBS-T) for one hour at room temperature. After blocking, the membranes were incubated two hours at room temperature in 5% non-fat milk in PBS-T containing the following primary antibodies against the indicated proteins: 17β-HSD1 (1:100,000 dilution of ab51045) from Abcam (Cambridge, MA), PCNA (1:500 dilution of sc-7907), nm23-H1 (1:500 dilution of sc-343), BCCIP (1:300 dilution of sc-130898) from Santa Cruz Biotechnology (Santa Cruz, CA), and β-actin as internal control (1:7,500 dilution of a monoclonal antibody, from Sigma). Next, membranes were incubated for one hour at room temperature with a horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology) diluted 10,000 times. Protein signals were visualised with Chemiluminescence Reagent (PerkinElmer) and bands were quantified by using ImageJ and Agfa Arcus II scanner. The ratios between the signals of the protein of interest and β-actin were calculated to determine the relative protein expression values.

3. Results

3.1. 17β-HSD1 overexpression modulates protein profile of MCF7 cells

To investigate the proteomic modifications of MCF7 cells in response to 17β-HSD1 overexpression, we performed 2-D gel analysis using total protein lysates of the wt breast cancer cell line MCF7 and MCF7 cells overexpressing 17β-HSD1 cultured in E2-containing medium. We then compared the proteomic profile of the two cell lines. To do so, we first stably transfected wt MCF7 cells with a 17β-HSD1 plasmid to generate the 17β-HSD1-stably transfected cells (MCF7-17βHSD1 cells). Western blot showed an increase in 17β-HSD1 expression in the stably transfected MCF7-17βHSD1 cells (Figure 1A). Proteomic analyses were carried out on six 2-D electrophoresis gels made from three independent biological repetitions of protein samples from MCF7-17βHSD1 cells and the parent cells. The two cell lines displayed similar spot patterns (Figure 1B) which allowed a good spot alignment for the proteome comparison. MCF7-17βHSD1 protein samples exhibited a lower number of protein

214

spots (3024) than MCF7 (3132 spots). The proteomic analyses using the Progenesis software and a t-test (with a p-value < 0.05) identified 30 significant differential protein spots between MCF7 and MCF7-17βHSD1 as follows: 5 spots down-regulated and 13 spots up-regulated in MCF7-17βHSD1 as compared to MCF7, for a total of 18 spots that varied 2-fold or more, whilst 7 and 5 spots were found unique to MCF7 and MCF7-17βHSD1 samples, respectively (Figure 1C). The analyses by MS of 18 protein spots (Figure 1B), selected among the differentially expressed spots, allowed the identification of proteins with a known UniProt accession number among all the spots for a total of 73 proteins. The numbers of proteins found in each cell line are listed in Figure 1C. Some spots contained more than one protein and some proteins were present in more than one spot. For example, 17β-HSD1 was identified in the spot numbers 2300 and 2305 (Figure 1B and D). This results in the identification of 59 distinct proteins distributed as follows: 15 and 9 proteins from spots unique to MCF7 and MCF7-17βHSD1 respectively, and 35 proteins from spots up-regulated in either cell line. These results showed that 17β-HSD1 modulates protein profile in MCF7 cells. Using the UniProt database at www.uniprot.org, we determined the subcellular locations and functions (or biological processes) of each of the 59 proteins identified by MS analysis. The original spot for each protein, the spot fold-increase or fold-decrease in one cell line versus another cell line, the protein name, the molecular mass, the isoelectric point, the number of unique peptides allowing the protein identification in the MS analysis, and the UniProt accession number of the protein are listed (Table 1; see Supplementary Table 2 for additional data). The information about the molecular function and/or biological process was found for most proteins. Important proteins involved in cell proliferation were differentially expressed following 17β-HSD1 overexpression. These proteins include PCNA, peroxiredoxin 2, BRCA2 and CDKN1A interacting protein (BCCIP) and ribonuclease/angiogenin inhibitor 1 (RNH1). Intriguingly, the metastasis inhibition factor nm23 (nm23-H1), an enzyme known to act as a down-regulator of breast cancer metastasis, was up-regulated by 17β-HSD1 and found in a spot unique to MCF7-17βHSD1 as compared to MCF7. The repartitions of the functions and subcellular locations of the 59 differentially expressed proteins are illustrated and the percentages of proteins involved in each molecular function and found in each cellular location are indicated (Figure 2). Overexpression of 17β-HSD1 in MCF7 cells causes

215 a differential expression of proteins that act mainly in metabolism (5.5% up-regulated protein and 5.5% repressed), mRNA processing (3.1% induced and 7.9% repressed), protein biosynthesis (9%) and transport (8%). Differentially expressed proteins are mainly located in cytoplasm and nucleus. In order to verify the differential expression of individual proteins in MCF7 and MCF7- 17βHSD1, we performed Western blot analysis on total protein extracts for three of the identified proteins, PCNA, nm23-H1 and BCCIP. The differential expression of PCNA and nm23-H1 in MCF7 and MCF7-17βHSD1 was confirmed. However, BCCIP barely exhibited a protein band and thus, the observed relative protein expression values in the two cell lines should be considered with caution (Figure 3A and B).

3.2. 17β-HSD1 regulates the mRNA levels of enzymes involved in cell proliferation

Next, we investigated if 17β-HSD1 influences the transcription of genes involved in cell proliferation. To do this, six proteins involved in cell proliferation, cancerogenesis or metastasis regulation were selected: PCNA, peroxiredoxin 2, nm23-H1, S-phase kinase- associated protein 1 (SKP1), BCCIP and RNH1. Their mRNA levels were measured by RT- qPCR analysis of total RNA extracts from MCF7 and MCF7-17βHSD1 cell lines (Table 2). The increases in mRNA levels of PCNA, peroxiredoxin 2, nm23-H1 and BCCIP following 17β-HSD1 overexpression were significant (more than 2-fold). The most regulated mRNA is that of nm23-H1 which was 3.6-fold higher in MCF7-17βHSD1 cells than in MCF7 cells (Table 2). We further evaluated the correlation between mRNA and protein levels by comparing data from the RT-qPCR and the proteomic analyses. Proteomics and RT-qPCR data were considered to correlate if the mRNA level and protein spots were regulated in the same direction. It must be noted that the observed correlations are semiquantitative, since 2-D gel data are considered semiquantitative, some spots contained more than one protein. In addition, some proteins were found in several spots, which can be the effect of posttranslational modifications [19]. When comparing MCF7-17βHSD1 to MCF7, we found that 216

RNH1, a regulator of angiogenesis, was down-regulated at both protein and transcript levels, whereas PCNA, SKP1 and nm23-H1 were up-regulated at both protein and transcript levels. With the exception of peroxiredoxin-2 (which is an anti-apoptosis protein) and BCCIP (a promoter of cell cycle arrest), all the other four proteins for which the mRNA expression was evaluated, exhibited a regulation in the same direction for protein and mRNA in MCF7- 17βHSD1 as compared to MCF7 (Table 2). These data can indicate the existence of a semiquantitative correlation between protein and mRNA expression. Thus, it may be possible to predict the presence of a protein based on its gene expression or inversely. However, as suggested in a previous study [20], the correlation between mRNA and protein levels may not be sufficient to predict protein expression levels from quantitative mRNA data.

3.3. Transcription of various genes involved in E2 production

Because 17β-HSD1 is a pivotal enzyme in the synthesis of E2, a hormonal steroid playing a major role in breast cancer induction and progression, we were interested to know if its overexpression in MCF7 cells would influence the expression of other genes involved in the hormone synthesis, inactivation and action. The mRNA levels of these proteins, which include 17β-HSDs type 2 (17β-HSD2), type 5 (17β-HSD5), type 7 (17β-HSD7), type 12 (17β-HSD12), aromatase (or P450arom), estrogen sulfotransferase (EST), STS, androgen receptor (AR), estrogen receptor alpha (ERα) and estrogen receptor beta (ERβ), were quantified by RT-qPCR (Figure 3C). RT-qPCR analyses revealed that the overexpression of 17β-HSD1 in MCF7 cells induces an increase in the mRNA expression of 17β-HSD5, STS, 17β-HSD12 and ERβ by 20, 33, 73 and 120%, respectively, while inhibiting AR expression by 64%. The highest mRNA level modulation was observed with ERα, which exhibited an increase of 171%. 17β-HSD7 showed a low increase whereas no modulation was observed with 17β-HSD2 and aromatase expression (Figure 3C and D). These results show that the expression of 17β-HSD1 can influence that of other genes implicated in estradiol metabolism and action, especially the estrogen receptors (ER).

217

3.4. Correlation between 17β-HSD1 and nm23-H1 expression

Since our proteomic and RT-qPCR data showed that 17β-HSD1 overexpression increases the metastasis inhibition factor nm23-H1 mRNA and protein levels, we were interested to know if 17β-HSD1 knockdown would decrease nm23-H1 expression. T47D cells were chosen for this investigation because the cell line expresses a high level of endogenous 17β-HSD1 [11, 15]. Cells were transfected with 17β-HSD1 specific siRNA and with scramble siRNA (control siRNA) and total RNA was extracted 48 hours after transfection. 17β-HSD1 and nm23-H1 mRNAs were quantified by RT-qPCR. The efficacy of 17β-HSD1 knockdown by its specific siRNA was demonstrated since 92% inhibition of 17β-HSD1 mRNA was observed (Figure 3E). Nm23-H1 mRNA levels were compared in control-siRNA- and 17β- HSD1-siRNA-transfected T47D cells. A decrease of 31% of nm23-H1 mRNA expression was observed after 17β-HSD1 gene knockdown (Figure 3F). These results, combined with proteomic and RT-qPCR analyses of MCF7 and MCF7-17βHSD1, indicate a positive correlation between 17β-HSD1 and nm23-H1 expression.

3.5. Regulation of cell migration by 17β-HSD1

To investigate if 17β-HSD1 could have any implication in cancer metastasis, we evaluated the effect of its expression on MCF7 cell migration by means of a wound-healing assay. The migrations of wt MCF7 and MCF7-17βHSD1 cells were first compared. We found that cell migration was higher in MCF7-17βHSD1 stably overexpressing 17β-HSD1 than in wt MCF7 cells, showing that MCF7-17βHSD1 cells have more ability to invade a scratch than wt MCF7 cells (Figure 4A and B). The effect of 17β-HSD1 knockdown on cell migration was then tested in MCF7-17βHSD1 cells transfected with 17β-HSD1 specific siRNAs or control siRNA. Semiquantitative RT-PCR analysis showed a 98% decrease of 17β-HSD1 mRNA after transfection of MCF7-17βHSD1 cells with specific siRNAs as compared to the control cells (Figure 4C). Importantly, the knockdown of 17β-HSD1 in MCF7-17βHSD1 cells was associated with a decreased cell migration (by 16.1%, p<0.05) compared to cells transfected with control siRNA (Figure 4D and E). Taken together, these results show that 17β-HSD1 expression is positively correlated with MCF7 cell migration.

218

4. Discussion

4.1. Proteomic modifications of MCF7 cells in response to 17β-HSD1 overexpression

In a previous study, we showed that modulating the expression of the steroid-converting enzyme 17β-HSD1 in MCF7 and T47D cells led to a differential cell growth compared to the parent cells, cultured in E2-containing medium [11]. The present study compared the proteomes of the stably transfected MCF7-17βHSD1 and wt MCF7 cells and established the first differential profile of a cell line overexpressing the enzyme and its parent cell. Our proteomics data revealed that increasing 17β-HSD1 expression significantly modulates the expression of proteins involved in various functional activities such as cell cycle, cell growth, apoptosis and carcinogenesis. Examples include PCNA, BCCIP and peroxiredoxin-2. Considering the functions of these enzymes in breast cancer, the directions in which their expression is regulated by 17β-HSD1 agree with its role in increasing breast cancer cell growth. This could reveal the factors that make 17β-HSD1-stably-transfected MCF7 cells grow faster than the wt MCF7 cells when cultured in E2-containing medium [11]. The four most represented functional activities for the modulated proteins are metabolism (examples include alpha-N-acetylgalactosaminidase and mitochondrial enoyl-CoA hydratase), mRNA processing (eukaryotic initiation factor 4A-III and arginine/serine-rich 2 splicing factor), protein biosynthesis (elongation factor 1-gamma) and transport (endoplasmic reticulum resident protein ERp29 and RAB11B protein) (Figure 2, Table 1 and Supplementary Table 2). The predominant locations of differentially expressed proteins in the nucleus and cytoplasm might reflect their functions in mRNA processing and protein biosynthesis. These four functions are known to be essential in steroid signalling which involves fast nongenomic activities that include the transport and metabolism of signalling molecules and genomic mechanisms mediated by their specific receptor, which comprise gene transcription (RNA formation and mRNA processing) and the biosynthesis of proteins [14] related to cell growth and regulation. The presence of E2 in MCF7 and MCF7-17βHSD cell culture medium and the change in expression of a large number of proteins involved in these four functions following 17β-HSD1 overexpression suggest a modulation of E2 effects by 17β-HSD1. In

219 fact, the 17β-HSD enzymes fit well into the concept of pre-receptor regulator of steroid action as they efficiently alter the binding of steroids to their genomic and nongenomic receptors and effectors, acting as a metabolic switch prior to the function of these receptors [14]. The concept of pre-receptor regulation of E2 action by 17β-HSD1 corroborates with its effect on the modulation of E2 responsiveness of pS2 genes in T47D cells [11], since E2 exerts its biological effect on breast cancer predominantly via the mediation of ERα and ERβ [21, 22]. The strong stimulation of ERα (171% increase) and ERβ (120% increase) gene expression and the protein regulation of a large number of non-estrogen responsive genes caused by 17β-HSD1 overexpression further suggest that the ligand-independent transcriptions of ER target genes are also modulated by the enzyme. Indeed, this stimulation can influence the regulation of gene transcription by ER. A recent study showed high level of 17β-HSD2 in wt MCF7 [23]. Our data, on the contrary, showed negligible expression of 17β-HSD2 in this cell line, in conformity with other studies [24, 25]. While 17β-HSD1 has no effect on the expression of the E2-inactivating enzyme 17β-HSD2, it increases the mRNA levels of the E2-producing enzymes 17β-HSDs type 5, 7 and 12 with the higher increase being type 12. This suggests a concerted action of reductive 17β-HSDs to accelerate the cellular E2 biosynthesis.

4.2. 17β-HSD1 increases breast cancer cell migration despite a positive correlation with the metastasis suppressor gene nm23-H1

Two estrogen-responsive genes involved in metastasis regulation, cathepsin D (Table 1) [26] and nm23-H1 [27-29], were found to be differentially expressed at the protein levels following 17β-HSD1 overexpression. Cathepsin D, which is an independent marker of poor prognosis in breast cancer correlated with the incidence of clinical metastasis [30], was down-regulated. Nm23-H1 was up-regulated at the protein level, its mRNA increased 3.6- times with 17β-HSD1 overexpression (in MCF7-17βHSD1), and its gene expression decreased by 31% following 17β-HSD1 knockdown in T47D. These results demonstrated that 17β-HSD1 expression is positively and closely correlated to nm23-H1 expression. The down-regulation of cathepsin D can be related to the increase of nm23-H1 as their negative correlation has already been demonstrated [26]. Patients with malignant melanoma who

220

develop metastases during the first 2 years after diagnosis have significantly lower levels of tumor nm23-H1 expression (56% of the mean value) as compared to patients with a less aggressive disease (164%) [31]. The nm23-H1 gene, NM23, is known to function as a tumor metastasis suppressor gene and its transcript level is reduced in highly metastatic cells [31, 32]. It has been reported that nm23-H1 inhibits cell migration and cancer metastasis by modulating the activity of members of the Rho small GTPase family, enzymes known to play a key role in the actin cytoskeleton dynamics required for cancer cell migration and invasion [33]. In agreement with the strong enhancement of nm23-H1 mRNA level by 17β-HSD1, we hypothesized the implication of 17β-HSD1 in tumor metastasis. Until now, quantitative analyses of the transcripts of estrogen-producing enzymes in breast cancer metastases have not demonstrated any significant association between 17β-HSD1 mRNA level and metastases, although the sulfatase and aromatase mRNA levels were significantly associated with the presence of metastases by some studies [34-36]. Using the wound-healing assay, we demonstrated for the first time that increasing 17β-HSD1 expression led to the increase of MCF7 cell migration while 17β-HSD1 knockdown decreased MCF7 cell migration. Our study thus shows for the first time that 17β-HSD1 expression is positively correlated with the migration of the breast cancer cell line MCF7, revealing its role as a positive regulator of cell migration, contrary to nm23-H1. E2 induced time-dependent increases in the abundance of nm23-H1 mRNA and protein, with the extent of these effects correlating with the expression level of its receptor ERα [28], which has been shown to interact with the non-metastasis gene nm23-H1 [26, 28]. Since our cell models were cultivated in the presence of E2 and our data showed that 17β-HSD1 positively regulates ERα mRNA level, one could postulate, for the time being, that the positive correlation between nm23-H1 and 17β-HSD1 expression is in coincidence with the activation of nm23-H1 expression by ERα and E2, which are increased with 17β-HSD1 expression. Thus, 17β-HSD1 may indirectly affect nm23-H1 expression via ERα action. On the other hand, the increase of MCF7 cell migration by 17β-HSD1, demonstrated in the present study, corroborates with its role in stimulating breast cancer cell growth [11] and the poor prognosis for patents in whom 17β-HSD1 is highly expressed in the breast [13]. This may open a new study on the role of this multi-functional steroid enzyme appeared early in evolution, revealing a complex mechanism in breast cancer with its expression. The later

221 may involve protein-protein and protein-DNA interactions among ERα, nm23, 17β-HSD1, Cathepsin D and other genes.

5. Conclusions

Our study demonstrates that 17β-HSD1 affects breast cancer cell proteome and modulates expression of several genes at both mRNA and protein levels. Among the individual mRNA and proteins for which the regulation was investigated, the most strongly modulated by 17β- HSD1 are ERα and nm23-H1. Intriguing observations are that although 17β-HSD1 strongly stimulates nm23-H1 expression, it is associated with an increased MCF7 cell migration. Here, we report the general study of proteomics with 17β-HSD1 expression modification, while the mechanism on cell migration modification opens a new study of interest for additional roles of the well-known steroid-converting enzyme. It may be of great interest to investigate 17β-HSD1 role in cancer metastasis formation.

222

List of abbreviations used 17β-HSD1, 17beta-hydroxysteroid dehydrogenase type 1; 2-D, two-dimensional; AR, androgen receptor; BCCIP, BRCA2 and CDKN1A interacting protein; CHAPS, 3-[(3- cholamidopropyl)dimethylammonio]-1-propanesulfonate; cm; centimeter; CM, square centimetre; E1, estrone; E1S, estrone sulphate; E2, estradiol; ER, estrogen receptors; ERα, estrogen receptor alpha; ERβ, estrogen receptor beta; EST, estrogen sulfotransferase; FBS, fetal bovine serum; INCA, Intelligent Noise Correction Algorithm; mm, millimeter; PCNA, proliferating cell nuclear antigen; RNH1, ribonuclease/angiogenin inhibitor 1; RT-PCR, reverse transcription PCR; RT-qPCR, reverse transcription quantitative real-time PCR; SDS- PAGE, sodium dodecyl sulphate - polyacrylamide gel electrophoresis; SKP1, S-phase kinase-associated protein 1; STS, steroid sulfatase; wt, wild type.

Competing interests The authors declare that they have no competing interests.

Authors’ contributions JAA, MZ and FH carried out experimental studies. JAA and SXL designed the study. JAA, SXL and JH prepared the manuscript. All authors read and approved the final manuscript.

Acknowledgments We thank Mrs G. Racine for her advices in 2-D gel image analysis; we acknowledge Dr. E-L. Calvo for his critical reading of the manuscript, from CHUQ Research Center (Quebec, Canada). We thank Ms. S. Méthot for her valuable editing of the manuscript. This work was supported by the Canadian Institutes of Health Research, with grant to S-X. Lin (Principal Investigator for FRN57892).

223

References

1. Sasco AJ: Breast Cancer and the Environment. Horm Res 2003, 60:Suppl 3:50. 2. ESHRE Capri Workshop Group: Hormones and breast cancer. Hum Reprod Update 2004, 10:281-293. 3. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, Thun MJ: Cancer statistics, 2008. CA Cancer J Clin 2008, 58:71-96. 4. Canadian Cancer Society's Steering Committee: Canadian Cancer Statistics 2009. 2009, Toronto: Canadian Cancer Society. [www.cancer.ca/statistiques]. 5. American Cancer Society: Breast cancer facts & figures 2009-2010. 2009, Atlanta: American Cancer Society, Inc. [www.cancer.org]. 6. Russo IH, Russo J: Role of hormones in mammary cancer initiation and progression. J Mammary Gland Biol Neoplasia 1998, 3:49-61. 7. Russo J, Hasan Lareef M, Balogh G, Guo S, Russo IH: Estrogen and its metabolites are carcinogenic agents in human breast epithelial cells. J Steroid Biochem Mol Biol 2003, 87:1-25. 8. Nguyen BL, Chetrite G, Pasqualini JR: Transformation of estrone and estradiol in hormone-dependent and hormone-independent human breast cancer cells. Effects of the antiestrogen ICI 164,384, danazol, and promegestone (R-5020). Breast Cancer Res Treat 1995, 34:139-146. 9. Pasqualini JR: The selective estrogen enzyme modulators in breast cancer: a review. Biochim Biophys Acta 2004, 1654:123-143. 10. Aka JA, Mazumdar M, Lin S-X: Reductive 17beta-hydroxysteroid dehydrogenases in the sulfatase pathway: critical in the cell proliferation of breast cancer. Mol Cell Endocrinol 2009, 301:183-190. 11. Aka JA, Mazumdar M, Chen C-Q, Poirier D, Lin S-X: 17β-hydroxysteroid dehydrogenase Type 1 stimulates breast cancer by dihydrotestosterone inactivation in addition to estradiol production. Mol Endocrinol 2010, 24:832-845. 12. Oduwole OO, Li Y, Isomaa VV, Mäntyniemi A, Pulkka AE, Soini Y, Vihko PT: 17beta-hydroxysteroid dehydrogenase type 1 is an independent prognostic marker in breast cancer. Cancer Res 2004, 64:7604-7609. 13. Gunnarsson C, Jerevall PL, Hammar K, Olsson B, Nordenskjöld B, Jansson A, Stål O: Amplification of HSD17B1 has prognostic significance in postmenopausal breast cancer. Breast Cancer Res Treat 2008, 108:35-41. 14. Meier M, Möller G, Adamski J: Perspectives in understanding the role of human 17beta-hydroxysteroid dehydrogenases in health and disease. Ann N Y Acad Sci 2009, 1155:15-24. 15. Laplante Y, Rancourt C, Poirier D: 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 2009, 301:146-153. 16. Shevchenko A, Wilm M, Vorm O, Mann M: Mass spectrometric sequencing of proteins from silver-stained polyacrylamide gels. Analytical Chemistry 1996, 68:850-858.

224

17. Havlis J, Thomas H, Sebela M, Shevchenko A: Fast-Response Proteomics by Accelerated In-Gel Digestion of Proteins. Anal Chem 2003, 75:1300-1306. 18. Luu-The V, Paquet N, Calvo E, Cumps J: Improved real-time RT-PCR method for high-throughput measurements using second derivative calculation and double correction. Biotechniques 2005, 38:287-293. 19. Huber M, Bahr I, Krätzschmar J R, Becker A, Müller E C, et al: Comparison of Proteomic and Genomic Analyses of the Human Breast Cancer Cell Line T47D and the Antiestrogen-resistant Derivative T47D-r. Mo Cell Proteomics 2004, 3:43- 55. 20. Gygi SP, Rochon Y, Franza BR, Aebersold R: Correlation between Protein and mRNA Abundance in Yeast. Mol Cell Biol 1999, 19:1720-1730. 21. Kim J, Petz LN, Ziegler YS, Wood JR, Potthoff SJ, Nardulli AM: Regulation of the estrogen-responsive pS2 gene in MCF-7 human breast cancer cells. J Steroid Biochem Mol Biol 2000, 74:157-168. 22. Marchiori AC, Casolari DA, Nagai MA: Transcriptional up-regulation of PHLDA1 by 17beta-estradiol in MCF-7 breast cancer cells. Braz J Med Biol Res 2008, 41:579-582. 23. Hevir N, Trošt N, Debeljak N, Rižner TL: Expression of estrogen and progesterone receptors and estrogen metabolizing enzymes in different breast cancer cell lines. Chem Biol Interact 2011, 191:206-216. 24. Laplante Y, Rancourt C, Poirier D: 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 2009, 301:146-153. 25. Smuc T, Rizner TL: Expression of 17beta-hydroxysteroid dehydrogenases and other estrogen-metabolizing enzymes in different cancer cell lines. Chem Biol Interact 2009, 178:228-33. 26. Curtis CD, Likhite VS, McLeod IX, Yates JR, Nardulli AM: Interaction of the tumor metastasis suppressor nonmetastatic protein 23 homologue H1 and estrogen receptor alpha alters estrogen-responsive gene expression. Cancer Res 2007, 67:10600-10607. 27. Charpentier AH, Bednarek AK, Daniel RL, Hawkins KA, Laflin KJ, Gaddis S, MacLeod MC, Aldaz CM: Effects of estrogen on global gene expression: identification of novel targets of estrogen action. Cancer Res 2000, 60:5977-5983. 28. Lin KH, Wang WJ, Wu YH, Cheng SY: Activation of Antimetastatic Nm23-H1 Gene Expression by Estrogen and Its alpha-Receptor. Endocrinology 2002, 143:467-475. 29. Wang DY, Fulthorpe R, Liss SN, Edwards EA: Identification of estrogen-responsive genes by complementary deoxyribonucleic acid microarray and characterization of a novel early estrogen-induced gene: EEIG1. Mol Endocrinol 2004, 18:402-411. 30. Liaudet-Coopman E, Beaujouin M, Derocq D, Garcia M, Glondu-Lassis M, Laurent- Matha V, Prébois C, Rochefort H, Vignon F: Cathepsin D: newly discovered functions of a long-standing aspartic protease in cancer and apoptosis. Cancer Lett 2006, 237:167-179.

225

31. Flørenes VA, Aamdal S, Myklebost O, Maelandsmo GM, Bruland OS, Fodstad O: Levels of nm23 messenger RNA in metastatic malignant melanomas: inverse correlation to disease progression. Cancer Res 1992, 52:6088-6091. 32. Youn B, Kim HD, Kim J: Nm23-H1/nucleoside diphosphate kinase as a key molecule in breast tumor angiogenesis. Expert Opin Ther Targets 2008, 12:1419-30. 33. Boissan M, De Wever O, Lizarraga F, Wendum D, Poincloux R, Chignard N, Desbois- Mouthon C, Dufour S, Nawrocki-Raby B, Birembaut P, Bracke M, Chavrier P, Gespach C, Lacombe ML: Implication of metastasis suppressor NM23-H1 in maintaining adherens junctions and limiting the invasive potential of human cancer cells. Cancer Res 2010, 70:7710-7722. 34. Miyoshi Y, Ando A, Hasegawa S, Ishitobi M, Taguch, T, Tamaki Y, Noguchi S: High expression of steroid sulfatase mRNA predicts poor prognosis in patients with estrogen receptor-positive breast cancer. Clin Cancer Res 2003, 9:2288-2293. 35. Irahara N, Miyoshi Y, Taguchi T, Tamaki Y, Noguchi S: Quantitative analysis of aromatase, sulfatase and 17beta-HSD(1) mRNA expression in soft tissue metastases of breast cancer. Cancer Lett 2006, 243:23-31. 36. Salhab M, Reed MJ, Al Sarakbi W, Jiang WG, Mokbel K: The role of aromatase and 17-beta-hydroxysteroid dehydrogenase type 1 mRNA expression in predicting the clinical outcome of human breast cancer. Breast Cancer Res Treat 2006, 99:155- 162.

226

Figure legends

Figure 1. Proteomic analysis of wt MCF7 cells and MCF7 stably transfected with 17β- HSD1 (MCF7-17βHSD1). (A) 17β-HSD1 and β-actin expression between wt MCF7 and MCF7-17βHSD1 revealed by Western blots. (B) Representative 2-D gel images for wt MCF7 and MCF7-17βHSD1 cells. Whole cell lysates (200 µg) from each cell were separated by 2-D electrophoresis and visualized by Sypro Ruby staining. The 2-D gels were scanned and the differentially expressed (2-fold or higher, p < 0.05) proteins were detected using Progenesis software. The 18 differentially expressed protein spots that were selected for MS analysis are marked with circles. Protein spots up-regulated in MCF7-17βHSD1 are depicted in the MCF7-17βHSD1 proteome image; protein spots down-regulated in MCF7-17βHSD1 are depicted in the MCF7 proteome image. The numbers refer to the spot number listed in Table 1 and Supplementary Table 2. The squares represent the indicated area shown in more detail in (D). (C) Summary of the numbers of spots and proteins obtained from the proteomics data. * up-regulated (up) and down-regulated (down) proteins in MCF7-17βHSD1 as compared to wt MCF7 cells. (D) Zoom showing some differentially expressed protein spots from wt MCF7 and MCF7-17βHSD1 comparison. Arrows indicate 17β-HSD1 protein which was revealed by MS analysis to be present in the spot numbers 2305 (unique to MCF7-17βHSD1) and 2300 (up-regulated in MCF7-17βHSD1 as compared to wt MCF7 cells).

Figure 2. Functions and cellular locations of the differentially expressed proteins in MCF7-17βHSD1 and MCF7 cells. The Uniprot database was used to generate the cellular location and the molecular function and/or biological process of each of the 59 nonredundant (distinct) proteins identified by mass spectrometry analysis as differentially regulated. Of the 59 distinct proteins, the percentages of proteins involved in each molecular function and found in each cellular location are indicated in bracket.

Figure 3. mRNA and protein level modulation by 17β-HSD1. (A) Expression of 17β- HSD1, PCNA, nm23-H1 and BCCIP between wt MCF7 and MCF7 cells stably transfected with 17β-HSD1 (MCF7-17βHSD1) revealed by Western blots. β-actin protein amount was

227 used as internal control. The arrows show the positions of the protein bands. (B) The Western blot bands in (A) were quantified, and the ratio between the protein signals of interest and the β-actin signal was calculated to determine the relative protein expression values between wt MCF7 and MCF7-17βHSD1. (C) RT-qPCR values (mRNA copies/µg total RNA) of mRNAs encoding proteins involved in estradiol production and/or action within wt MCF7 and MCF7- 17βHSD1. N, negligible (RT-qPCR values < 1,000); 0, mRNA not detected after many rounds of amplification. SDs were < 10% of duplicates. (D) Relative mRNA expression values of enzymes involved in estradiol production in MCF7-17βHSD1 as compared to wt MCF7. The mRNA levels in wt MCF7 cells were fixed at 100. (E) and (F) Relative 17β- HSD1 (E) and nm23-H1 (F) mRNA expression in siRNA-transfected T47D cells. T47D cells were transfected with 17β-HSD1 siRNA or control siRNA and 17β-HSD1 mRNA was quantified by RT-qPCR. mRNA quantity in control-siRNA transfected cells was fixed at 100.

Figure 4. 17β-HSD1 is a positive regulator of MCF7 cell migration. (A) and (B) Comparison of cell migrations between wt MCF7 and MCF7 cells stably transfected with 17β-HSD1 (MCF7-17βHSD1). A scratch was applied to wt MCF7 and MCF7-17βHSD1 cells confluent in 35 x 10 mm2 dishes and the ability to invade the scratch was measured. (A) The relative migration of wt MCF7 and MCF7-17βHSD1 cells at 24 and 36 hours post- scratch was quantified. The scratch widths, two near the border and three in the middle of the scratch (as shown in B) were measured at the indicated time points using the freeware ImageJ and data were used to calculate the percentage of migration. Averages ± SD from two separate experiments performed in triplicate are indicated. (B) Results showed that wt MCF7 cells have less ability to invade the scratch than the MCF7-17βHSD1 cells. Lines represent measurements made to assess modifications in scratch width. (C) 17β-HSD1 knockdown by siRNA in MCF7-17βHSD1 cells. Semiquantitative RT-PCR was performed using 17β-HSD1 and β-actin primers and total RNA extracted from MCF7-17βHSD1 cells transfected with 17β-HSD1 specific siRNAs or control siRNA. (D) and (E) Effect of 17β-HSD1 knockdown on MCF7-17βHSD1 cell migration. MCF7-17βHSD1 cells were transfected with 17β-HSD1 specific siRNAs or control siRNA for 48 hours before creating a wound by scraping the cell monolayer. Cells transfected with 17β-HSD1 siRNA have less ability to invade the scratch than cells transfected with control siRNA. All experiments were done in duplicate and

228

repeated twice, and representative images of cell progression in the scratch are shown. Error bars represent SD. *, P<0.05 by Student’s t test.

229

Tables Table 1. Mass spectrometry identification of proteins differentially expressed between wild type MCF7 and MCF7 stably transfected with 17β-HSD1 (MCF7-17βHSD1). The function and/or biological process were from the UniProt database (www.uniprot.org). Spot, spot number; FC, fold change; U, unique; MW, molecular weight (kDa); pI exp, isoelectric point as determined from the 2-D gel experiments; Pep, number of unique peptides. Proteins in bold were used for RT-qPCR validation. The number after the protein name indicated the additional spot in which the protein was found. MW UniProt pI Function and/or biological Spot FC Description exp/ Pep number exp process pred SPOT DOWN-REGULATED IN MCF7-17βHSD1 AS COMPARED TO WT MCF7 4183 2.7 Cathepsin D P07339 28/45 5.0 20 Proteolysis, pathogenesis of diseases (breast cancer) Ezrin-radixin-moesin-binding phosphoprotein 50 O14745 58/39 5.3 17 Wnt signaling pathway 3252 2.3 Neudesin Q9UMX5 19/19 4.7 2 Neuronal differentiation and Proliferation 1703 2.2 Ribonuclease/angiogenin inhibitor 1 P13489 58/50 4.6 24 Regulation of angiogenesis, (RNH1)(3177) mRNA catabolism BRCA2 and CDKN1A interacting protein Q9P287 58/36 4.6 4 Promote cell cycle arrest (BCCIP) Cell division cycle protein 123 homolog O75794 58/39 4.6 2 Required for S phase entry of the cell cycle SPOT UNIQUE TO WT MCF7 2617 U Poly(rC)-binding protein 2 Q15366 32/39 6.4 11 RNA binding Purine nucleoside phosphorylase P00491 32/32 6.4 6 DNA modification BTB/POZ domain-containing protein KCTD15 Q96SI1 32/32 6.4 4 Potassium ion transport RING finger protein 114 Q9Y508 32/26 6.4 2 Cell differentiation 4335 U Peptidyl-prolyl cis-trans isomerase E Q9UNP9 35/33 5.4 10 Protein folding, mRNA Splicing Transgelin-2 P37802 20/22 5.6 8 Muscle development Splicing factor, arginine/serine-rich 2 Q01130 35/25 5.4 2 mRNA processing 3039 U RAB11B protein A5YM50 24/25 5.7 11 Protein transport, signal Transduction Peroxiredoxin-2 P32119 24/22 5.7 6 Cell redox regulation, anti- Apoptosis Splicing factor, arginine/serine-rich 3 P84103 24/19 5.7 2 RNA processing in relation with cell proliferation SPOT UP-REGULATED IN MCF7-17βHSD1 AS COMPARED TO WT MCF7 2714 2.5 Exosome complex exonuclease RRP41 Q9NPD3 29/26 6.2 3 rRNA processing Enoyl-CoA hydratase, mitochondrial P30084 29/31 6.2 6 Fatty acid and lipid Metabolism Heat shock 70 kDa protein 1 P08107 29/70 6.2 5 Stress response Eukaryotic translation initiation factor 4H Q15056 29/27 6.2 4 Host-virus interaction, protein biosynthesis 2496 4.9 Proliferating cell nuclear antigen (PCNA) P12004 32/29 4.6 16 DNA replication 5474 4.0 Myosin regulatory light chain 2, nonsarcomeric P19105 20/20 4.7 5 Cytokinesis, receptor capping, cell locomotion 2300 3.3 60S acidic ribosomal protein P0(2305) A8K4Z4 40/34 5.6 6 Ribosome biogenesis, translation elongation SPOT UNIQUE TO MCF7-17βHSD1 2305 U 17β-hydroxysteroid dehydrogenase type 1(2300) P14061 41/35 5.4 22 Steroid biosynthesis 3177 U Metastasis inhibition factor nm23 (nm23-H1) P15531 21/17 5.9 6 Cell cycle and proliferation, nucleotide metabolism 4667 U 60S ribosomal protein L11 P62913 21/20 5.9 2 Binds to 5S ribosomal RNA 4678 U S-phase kinase-associated protein 1 (SKP1) P63208 20/19 4.6 2 Ubl conjugation pathway

230

Table 2. mRNA quantification by RT-qPCR of genes involved in breast cancer cell proliferation within wt MCF7 and MCF7 stably transfected with 17β-HSD1 (MCF7-17βHSD1) and comparison with 2-D gel data. SDs were < 10% of duplicates. Correlation a) MCF7- Fold Description MCF7 b) 2-D gel and 17βHSD1 regulation RT-qPCR RT-qPCR value (mRNA copies/µg total RNA) Proliferating cell nuclear antigen (PCNA) 1,599,813 4,483,982 + 2.8 Yes c) Peroxiredoxin 2 4,078,760 8,585,424 + 2.1 No c) Metastasis inhibition factor nm23 (nm23-H1) 5,366,763 19,356,416 + 3.6 Yes c) S-phase kinase-associated protein 1 (SKP1) 3,810,452 5,714,509 + 1.5 Yes BRCA2 and CDKN1A interacting protein (BCCIP) 345,839 1,074,647 + 3.1 No c) Ribonuclease/angiogenin inhibitor 1 (RNH1) 1,015,193 727,425 - 1.4 Yes c) a) Proteins were selected for RT-qPCR after their identification by mass spectrometry analysis of 2-D gel protein spots. b) Fold regulation of mRNA levels in MCF7-17βHSD1 compared to wt MCF7 (+, fold increase; -, fold decrease). c) Mass spectrometry analysis showed that the 2-D spot contained several proteins including the indicated protein.

231

Figure 1

232

Figure 2

233

Figure 3

234

Figure 4

235

Additional files

Additional file 1, Table S1. Primers used for reverse transcription quantitative real-time PCR. GenBank cDNA Gene symbol number (NCBI) Primer sequence (5' → 3') Forword/reverse fragmenta

PCNA NM_002592 gggcttcgacacctaccgctg/ctttctcctggtttggtgcttca 404-570 PRDX2 NM_005809 gcccacgcagctttcagtcat/agtccgacagcttcacctctttg 132-247 NME1 (NDKA) NM_198175 tgtggagagtgcagagaaggaga/gaaggaggggaaatggatgtga 694-833 SKP1 NM_006930 tgcacccaccacaaggatgac/catttttgatattgaaggtcttgcgaa 363-609 BCCIP NM_016567 agctggacaagtttttaaatgacacc/tcctgcttccacaaatgtcttactaat 514-713 RNH1 NM_002939 ctgtggatctgggagtgtggc/caggacttcacccacagcgactc 1254-1441 HSD17B2 NM_002153 gcgcctctcggtgctccaaatg/cggccatgcattgtttgtagtcagtca 557-738 AKR1C3 NM_003739 caaccaggtagaatgtcatccgtat/acccatcgtttgtctcgttga 633-752 HSD17B7 NM_016371 tccaccaaaagcctgaatctctc/gggctcactatgtttctcaggc 826-1118 HSD17B12 NM_016142 ggctggtcttgaaatcggcat/tgccactgccagatgaaatgtt 439-650 CYP19A1 M28420 cgacaggctggtaccgcatgctc/aagaggcaataataaaggaaatccagac 734-856 SULT1E1 NM_005420 aagcgttccaggcaagaccagatg/tttgcacttttccacatcaccctctt 205-323 STS (ARSC1) M16505 agccctaatcctgacccttttcttgg/ccgcctccaccgttagcctct 889-1025 ESR1 NM_000125 tgcaaaatctaacccctaaggaagtg/ctcccagtacccacagtccatctc 5258-5550 ESR2a NM_001437 cgccgtgaccgatgctttgg/gcccttgttactcgcatgcctgac 1711-1834 AR NM_000044 agccattgagccaggtgtagtgt/catcctggagttgacattggtgaa 3152-3401 aDownstream position from the ATG start codon.

236

Additional file 2, Table S2. Additional data of mass spectrometry identification of proteins differentially expressed between wild type MCF7 and MCF7 stably transfected with 17β-HSD1 (MCF7-17βHSD1). The function and/or biological process were from the UniProt database [19]. Spot, spot number; FC, fold change; U, unique; MW, molecular weight (kDa); pI exp, isoelectric point as determined from the 2-D gel experiments; Pep, number of unique peptides. The number after the protein name indicated the additional spot in which the protein was found. MW UniProt pI Function and/or biological Spot FC Description exp/ Pep number exp process pred SPOT DOWN-REGULATED IN MCF7-17βHSD1 AS COMPARED TO WT MCF7 4183 2.7 Keratin, type II cytoskeletal 8(2196; 3499) P05787 58/54 5.3 60 Host-virus interaction 3252 2.3 Calpain small subunit 1 P04632 19/28 4.7 6 Proteolyse catalysis

2214 3.0 Keratin, type I cytoskeletal 19(2196) P08727 44/44 4.6 45 Myofiber organization, host-virus interaction 2196 2.6 cDNA FLJ75154, highly similar to Homo A8K9A4 44/34 4.8 3 Nucleic acid binding sapiens heterogeneous nuclear ribonucleoprotein C (C1/C2), mRNA 1703 2.2 Alpha-N-acetylgalactosaminidase P17050 58/47 4.6 5 Carbohydrate metabolic process Elongation factor 1-gamma P26641 58/50 4.6 3 Protein biosynthesis 1957 5.4 Heterogeneous nuclear ribonucleoprotein H P31943 51/49 6.3 11 mRNA processing cDNA, FLJ92536, highly similar to Homo B2R5M8 51/47 6.3 3 Glutathione and isocitrate sapiens isocitrate dehydrogenase 1 (NADP+), metabolic process soluble (IDH1), mRNA COP9 signalosome complex subunit 3 Q9UNS2 51/48 6.3 2 Involved in various cellular and developmental processes Actin, cytoplasmic 1(2305) P60709 47/42 4.5 7 Cell motility Phosphoglycerate kinase 1 P00558 51/45 6.3 2 Glycolysis SPOT UNIQUE TO WT MCF7 4335 U Peptidyl-prolyl cis-trans isomerase E Q9UNP9 35/33 5.4 10 Protein folding, mRNA splicing Uncharacterized protein SFRS7 (Splicing factor, A6NNE8 35/16 5.4 2 Unknown arginine/serine-rich 7, 35kDa, isoform CRA_a) Palmitoyl-protein thioesterase 1 P50897 35/34 5.4 2 Sensory transduction SPOT UP-REGULATED IN MCF7-17βHSD1 AS COMPARED TO WT MCF7 2496 4.9 UPF0368 protein Cxorf26 Q9BVG4 32/26 4.6 2 Unknown 2714 2.5 BTB/POZ domain-containing protein KCTD21 Q4G0X4 29/30 6.2 5 Potassium ion transport Endoplasmic reticulum protein ERp29 P30040 29/29 6.2 4 Protein transport, folding and secretion Haloacid dehalogenase-like hydrolase domain- Q9BSH5 29/28 6.2 4 Metabolic process containing protein 3 High-mobility group box 1 A5D8W9 29/25 6.2 4 DNA bending SPOT UNIQUE TO MCF7-17βHSD1 2305 U UPF0553 protein C9orf64 Q5T6V5 41/39 5.4 2 Unknown

237

238

Annexe 2: Androgens, body fat distribution and adipogenesis

239

240

Androgens, body fat distribution and adipogenesis

Mouna Zerradi1,2, Julie Dereumetz1,2, Marie-Michèle Boulet1,2, André Tchernof1,2

1Endocrinology and Nephrology, CHU de Québec, 2 Department of Nutrition, Laval University, Québec City, Canada G1V 4G2

Keywords: adipose tissue, androgen, DHT, PCOS, testosterone.

Running title: Androgens, body fat distribution and adipogenesis

Corresponding author: Andre Tchernof, Ph.D. Endocrinology and Nephrology CHU de Quebec Research Center 2705 Laurier Blvd. R-4779 Quebec City, PQ CANADA G1V 4G2 Tel: 418-654-2296 Email: [email protected]

Cet article est publié dans le journal Current Obesity Reports. 3:396–403.

241

ABSTRACT

Androgens are regulators of important adipocyte functions such as adipogenesis, lipid storage and lipolysis. Through depot-specific impact on the cells of each fat compartment, androgens could modulate body fat distribution patterns in humans. Testosterone and dihydrotestosterone have been shown to inhibit the differentiation of preadipocytes to lipid- storing adipocytes in several models including primary cultures of human adipocytes from both men and women. Androgen effects have also been observed on some markers of lipid metabolism such as LPL activity, fatty acid uptake and lipolysis. Possible depot-specific and sex-specific effects have been observed in some but not all models. Transformation of androgen precursors to active androgens or their inactivation by enzymes that are expressed and functional in adipose tissue may contribute to modulate the local availability of active hormones. These phenomena, along with putative depot-specific interactions with glucocorticoids may contribute to human body fat distribution patterns.

242

1. INTRODUCTION

Obesity is an important public health problem because of its association with serious diseases such as type 2 diabetes and cardiovascular disease [1, 2]. It is a very heterogeneous condition in humans, and a major sex dimorphism is observed [3]. Generally, women have higher body fat percentages than men, who have higher lean body mass including muscle and bone [4-7]. Moreover, men present an android fat distribution pattern and preferentially accumulate fat in the abdomen, while women present a gynoid fat distribution pattern and more likely accumulate lipids in the gluteal and femoral regions of the body [8]. Men also have more visceral or intra-abdominal fat, that is, adipose tissue located inside the abdominal cavity, compared to women, regardless of total adiposity [8]. However, in each sex, wide interindividual variation is observed [9]. For example, visceral fat varies by approximately 10 fold in samples of lean to moderately obese Caucasian men and women [9]. Interestingly, the amount of visceral fat has been repeatedly shown to be among the most critical determinants of the presence of clustering cardiometabolic alterations that increase the risk of type 2 diabetes and cardiovascular disease [10-12].

Among the mechanisms that have been proposed to explain the detrimental effect of intra- abdominal fat accumulation, the portal vein theory suggested that increased free fatty acid release by highly lipolytic visceral adipocytes [13-17] directly in the portal vein to the liver could be a causal factor in visceral obesity-related metabolic alterations [18]. However, this theory has never been fully supported and has been challenged at least by some studies on free fatty acid homeostasis [19-21]. There has since been a paradigm shift in our understanding of adipose tissue physiology, and it is now increasingly clear that impaired adipose tissue expansion is likely a primary determinant of metabolic dysfunction [22]. Adipose tissue expansion occurs through adipocyte hypertrophy (enlargement of existing adipocytes through triglyceride synthesis), and/or adipocyte hyperplasia (generation of new adipocytes through differentiation of preadipocytes) [23, 24]. Failure to adequately store excess dietary lipids in adipose tissues through these mechanisms leads to post-prandial fatty acid overflow to ectopic sites such as the liver, muscle, pancreas and heart [25]. Parallel to 243 this demonstration, obesity has been associated with low-grade, chronic inflammation [26- 28]. Adipose tissue-derived cytokines and immune cells infiltrating adipose tissues have been considered as potentially mediating the link between visceral obesity and metabolic alterations increasing type 2 diabetes and cardiovascular disease risk [29-32]. Overall, in the metabolic syndrome, adipose tissue function impairments include adipocyte hypertrophy, impaired adipogenesis, low free fatty acid uptake, reduced triglyceride synthesis, resistance to the inhibitory effect of insulin on lipolysis, immune cell infiltration and inflammatory cytokine secretion. These alterations collectively reduce the storage capacity of fat tissue and contribute to systemic metabolic perturbations [8]. Hence, body fat distribution patterns and their hormonal determinants play critical roles in the development of metabolic alterations in the face of excess energy intake. The present article will briefly review the impact of androgens on body fat distribution patterns and selected aspects of adipose tissue function in humans.

2. ANDROGENS AND ADRENAL ANDROGEN PRECURSORS

Reduced testosterone levels in circulation have been associated with many pathological states [33, 34]. For example, in addition to anabolic functions, testosterone deficiency may also be involved in the development of type 2 diabetes or other metabolic alterations [33, 34]. Testosterone can also be converted into a more active androgen, dihydrotestosterone (DHT), or it can be transformed to estradiol (E2) by P450 aromatase [35]. DHT is the most potent natural androgen in humans and is involved in sexual maturation during puberty; it is known to amplify testosterone signaling through gene transcription modulation as it binds to the androgen receptor at lower concentrations than testosterone [35, 36]. DHT can be inactivated to 5α-androstane-3α,17β-diol through type 3 3α-hydrohysteroid dehydrogenase (HSD) action [37] or synthesized from this precursor [36]. On the other hand, active androgens can be generated from adrenal androgen precursors such as dehydroepiandrosterone (DHEA) and androstenedione [38]. In humans, the sulfated derivative of DHEA (DHEA-S) is the most abundant steroid hormone in circulation [38]. However, blood concentrations of this steroid decrease with aging and other conditions such as obesity and the metabolic syndrome [39]. Whether DHEA has a direct hormonal effect is uncertain since it has been considered as a

244

weak hormone and no nuclear receptor have been identified or characterized for this steroid [40]. Yet, it was suggested that DHEA could play a role in endothelial function, cellular immunity, inflammation, insulin sensitivity and sexual function [35, 40].

3. CIRCULATING ANDROGENS AND BODY FAT DISTRIBUTION PATTERNS

Androgens have been related to body fat distribution patterns in humans [41, 42]. In men, data are quite consistent in demonstrating that low testosterone levels are associated with abdominal, visceral obesity and the metabolic syndrome in cross-sectional and longitudinal studies [39, 43-46]. After the age of 20-30 years, men often present a decrease in serum testosterone while the concentration of SHBG increases with age [47] and this relative state of hypogonadism in some individuals is commonly associated with abdominal obesity and the metabolic syndrome [46, 48]. Consistent with these notions, testosterone treatment in hypogonadal men leads to a decrease in abdominal fat accumulation and metabolic improvements including improved glucose homeostasis [49-52]. These effects seem to depend on the dose used [53] and have rather neutral effects on HDL-cholesterol concentrations [51, 54]. They are apparently observed when androgen levels reached during treatment remain within the physiological range [43]. Supra-physiological androgen treatments have a completely different series of effects including increased visceral fat accumulation and insulin resistance in the case of female-to-male transsexuals [55, 56] or important alterations of blood lipids and cardiac function in the case of anabolic steroid users [57]. In sum, when examining physiologically relevant data in men, relative testosterone deficiency is related to abdominal, visceral obesity and features of the metabolic syndrome and replacement restores metabolic homeostasis and reduces abdominal fat accumulation.

In women, a positive association between abdominal obesity and plasma androgen levels has often been postulated or assumed. This is based on women with the polycystic ovary syndrome (PCOS), who often show hyperandrogenism associated with abdominal obesity and hyperinsulinemia [58]. However, abdominal obesity is not always found in women with PCOS when total adiposity is controlled for [59]. In fact, a wider examination of literature available on the association between circulating androgen levels and abdominal obesity in

245 women with or without PCOS leads to a much more equivocal picture [41, 42]. Several studies showed no significant association between abdominal body fat distribution and free testosterone, testosterone, DHT or androstenedione [60, 61]. Côté et al. reported a negative association between circulating levels of DHT and visceral fat accumulation [42]. Menopause status may explain some of the discrepancies observed. After menopause, there is a decrease in the level of E2 leading to a decrease in sex hormone-binding globulin (SHBG) level and an increase of free testosterone (FT) [62]. Cao et al. [63] found that abdominal fat accumulation correlated positively with FT and DHEA-S levels in early and late postmenopausal women, respectively. In another study of non-obese premenopausal women, DHEA levels seemed to be negatively associated with visceral fat accumulation [61].

The associations between abdominal obesity and circulating adrenal androgens DHEA, DHEA-S and androstenedione are controversial in both sexes. In men, significant negative associations were found between DHEA levels and measures of abdominal fat accumulation [64-66]. In women, two groups found a negative correlation between plasma DHEA-S and WHR [67, 68], while other groups reported a positive association between these two parameters [69, 70]. In our study, plasma androstenedione and DHEA levels were not related to either total or abdominal adiposity [42]. In men, a negative correlation between DHEA-S plasma and abdominal fat accumulation was found by some groups [64, 65] whereas others showed opposite results [71].

Given the very equivocal nature of the literature on circulating androgen associations with abdominal obesity in women, we suggest that the assumption that high androgen levels are associated with android body fat distribution in women should be critically re-examined. Among the factors that may explain discrepancies in the associations between body fat distribution measurements and adrenal androgen precursor levels, we propose that the depot- specific impact of androgens on adipose tissue function and variation in the ability to convert steroids to active hormones in peripheral sites such as adipose tissue may be of significance. The next sections provide a short summary of the studies which have investigated the effects of androgens on some aspects of adipose tissue function and a discussion of how local

246

adipose tissue steroid conversions may affect them, with a particular emphasis on preadipocyte differentiation.

4. ANDROGENIC IMPACT ON SELECTED ASPECTS OF ADIPOSE TISSUE FUNCTION

Androgens influence various biological processes that may impact on body fat mass and adipose tissue distribution in both sexes, including adipogenesis (preadipocyte differentiation), as well as lipolysis and lipid accumulation. Other review articles have also addressed this topic [72-74].

4.1 Adipogenesis

The process of differentiation of preadipocytes into mature adipocytes is a highly regulated pathway which necessitates multiple transcriptional factors. Adipocyte maturation starts with the recruitment of mesenchymal precursors which will be committed to become preadipocytes that will stop their growth, enter a phase of mitotic clonal expansion and eventually undergo terminal differentiation, the final step of this process generating mature, functional adipocytes [75]. An important driver of preadipocyte differentiation is peroxisome proliferator-activated receptor- (PPAR-γ), along with other factors such as enhancer binding proteins-α, β and δ (C/EBPα, C/EBPβ and C/EBPδ), insulin-like growth factor-1 (IGF-1) and glucocorticoid receptors, among others [75, 76].

As summarized in Table 1, androgens have been reported to alter preadipocyte differentiation in a number of studies. Testosterone and DHT prevented in vitro differentiation of 3T3-L1 and C3H10T1/2 murine preadipocyte cell lines through an AR- dependent pathway [77-79]. This effect was partially blocked by androgen receptor antagonists flutamide or bicatulamide [77, 79]. Another study tested the effect of testosterone and DHEA in the 3T3-L1, 3T3-F442A and 3T3-A31 murine preadipocyte cell lines and showed that both steroids decreased 3T3-L1 adipogenic proliferation and differentiation [80].

247

DHEA also decreased 3T3-F442A cell proliferation [80]. Interestingly, the effects of DHEA were abolished in the presence of 3β-HSD inhibitor Trilostane, suggesting that conversion to androgens or other steroids is required to observe an effect of DHEA in adipocytes [80]. Studies have also been conducted with wild-type (WT) and androgen receptor-null (ARKO) mice [81]. Loss of the androgen receptor in bone marrow stromal cells (BMSC) isolated from ARKO mice had higher self-renewal rates compared to cells from WT mice [81].

Regarding the effects of androgens on adipogenesis in humans, studies have demonstrated that DHT inhibits differentiation of preadipocytes from the subcutaneous, mesenteric and omental fat depots [82]. We have also shown that DHT and testosterone have an inhibitory effect on primary preadipocytes from the subcutaneous and visceral compartment in both men and women [83]. The commitment of subcutaneous abdominal adipose stem cells (hASCs) obtained from subcutaneous abdominal adipose tissue of non-obese women to preadipocytes was impaired when these cells were treated with testosterone or DHT, suggesting that androgens affect the early stage of preadipocyte differentiation [84]. DHEA was also found to inhibit differentiation and proliferation of human subcutaneous preadipocytes from the Chub-S7 cell line [85].

The impact of anatomic origin of the cells was difficult to evaluate in many of the above- cited studies, including those performed in human preadipocytes, because in vitro culture conditions often make detection of quantitative differences among fat compartments challenging [83]. In male rats, castration increased the proliferative capacity of preadipocytes in epididymal and perirenal fat. Differentiation capacity was, however, increased in epididymal fat and reduced in perirenal fat and no effect was observed in the subcutaneous depot [86]. Interestingly, data on the androgen receptor seem to show that visceral adipocytes may be more sensitive to androgenic action or express higher levels than subcutaneous adipocytes [87, 88]. More studies are needed to firmly establish depot-differences in the anti- adipogenic and anti-mitogenic effects of androgens.

4.2 Lipolysis

248

Lipolysis is the pathway leading to triglyceride hydrolysis in adipocytes, providing energy to tissues in the form of fatty acids; it is extremely well coordinated and regulated, and can be affected by many stimuli [89]. Two major enzymes in the lipolytic cascade contribute to hydrolysis of triglycerides in adipose tissue: adipose triglyceride lipase (ATGL) and hormone sensitive-lipase (HSL) [90]. HSL is phosphorylated through the cAMP-dependent protein kinase (PKA) pathway and translocated to fat droplets of the adipocyte [90]. Among the many agents that regulate lipolysis are catecholamines, which stimulate (β-adrenergic) and insulin, which inhibits lipolysis to a greater extent in visceral than in subcutaneous adipocytes [17, 91]. Studies on androgens and lipolysis are not unanimous. Testosterone treatment increased norepinephrine-stimulated lipolysis in subcutaneous fat from normal men [92], which was confirmed in human and rodent adipocytes treated with testosterone [93] or DHEA [94]. However, in female rhesus macaques, testosterone supplementation blunted the stimulatory effect of the luteal phase on lipolysis and hormone-sensitive lipase expression in visceral adipose tissue [95]. Other studies reported an inhibitory effect of testosterone on catecholamine-induced lipolysis in differentiated primary subcutaneous preadipocytes from women [96]. These effects were shown to be mediated through androgen receptor-dependent [97] regulation of β-adrenoreceptors, hormone-sensitive lipase and adenylate cyclase activity [93, 96, 98-100]. DHEA-S stimulated lipolysis in adipocytes from subcutaneous but not visceral adipocytes from women whereas in men no effect was observed in the visceral or subcutaneous compartment [94]. Finally, we have shown a positive correlation between androgen levels in plasma or omental adipose tissue and the responsiveness of adipocytes to positive lipolytic stimuli [101]. Although not all studies are consistent, androgens, particularly testosterone, seem to have sex-specific effects on adipocyte lipolysis.

4.3 Lipid accumulation

Regarding lipid uptake, most studies reported that androgens reduce lipid uptake and synthesis in adipose tissue. Testosterone supplementation in men decreased lipoprotein lipase (LPL) activity and triglyceride uptake in abdominal adipose tissue compartments [92, 102]. Different effects were reported in isolated mature adipocytes [97] or in fat from monkeys that were castrated and testosterone-replaced [103]. In the latter study, castration of Japanese

249 macaques led to the formation of very small, multilocular adipocytes, a phenotype that was reversed by testosterone replacement [103]. Ex vivo treatment of retroperitoneal adipose tissues from female rhesus macaques with DHT decreased fatty acid uptake in insulin-treated conditions but increased it in the basal sate [103]. In other experiments, an interaction with the menstrual cycle was observed [95]. For example, chronic testosterone treatment increased insulin signaling and fatty acid uptake in omental adipose tissue during the menses but this effect was absent during the luteal phase [95]. We studied the effects of DHT and testosterone on LPL activity measured in whole adipose tissue explants from the visceral and subcutaneous fat compartments of men and women [83]. Both steroids had a dose-dependent inhibitory effect on LPL activity that was especially apparent in explants from men. However, not all effects reached statistical significance and rather high androgen doses were required to observe inhibition in this model [83]. More studies are needed to establish the depot-specific impact of androgens on the pathways of lipid storage in adipose tissues from men and women.

5. STEROID CONVERSIONS AND INTERACTION WITH GLUCOCORTICOIDS

Providing a detailed review of the many steroid-converting enzymes that have been detected in adipose tissue so far is beyond the scope of this article. With respect to androgens, we have reported that the conversion of DHT to inactive androgen metabolite 5α-androstane-3α, 17β-diol was detected in fat tissues of both men and women [104-106]. Activity was higher in subcutaneous than in visceral adipose tissue and androgen inactivation rates in visceral fat were positively associated with adiposity indices such as BMI, fat cell size and visceral adipose tissue area measured by computed tomography [104-106]. We have demonstrated that the enzyme responsible for most of the conversion of DHT to the inactive androgen metabolite in human adipocytes is 3α-HSD type 3, or AKR1C2 [37].

When testing various factors that could modulate DHT inactivation rates in preadipocytes, we found that dexamethasone treatment alone led to a strong stimulation of DHT inactivation that was apparent in the first 24 hours of treatment and was reversed by glucocorticoid receptor antagonist RU486 [107]. Active glucocorticoids are known to stimulate

250

adipogenesis and can be synthesized locally by 11β-HSD type 1 as a function of the size and number of adipocytes in a given compartment [108-110]. Conversely, as described above, androgens inhibit adipogenesis and are inactivated locally by enzymes that are responsive to glucocorticoids such as dexamethasone. We have suggested that the stimulation of AKR1C2 expression and DHT inactivation by glucocorticoids in preadipocytes may remove some of the inhibitory effect of this steroid and create a permissive hormonal environment allowing adipogenesis. Other interactions have been reported in adipose tissue between the androgen- and glucocorticoid-signaling pathways [111]. Interaction of these hormonal signals in a depot-specific manner possibly represents a significant mechanism in the modulation of body fat distribution patterns.

6. CONCLUSION

Androgens likely play an important role in the modulation of several aspects of adipose tissue function. Androgenic action in adipose tissue appears to regulate lipid storage and adipose tissue expansion through modulation of adipogenesis, LPL activity, fatty acid uptake and lipolysis. Possible depot-specific and sex-specific effects have been observed in some but not all models. On the other hand, transformation of androgen precursors to active androgens or their inactivation by enzymes that are expressed and functional in adipose tissue may alter local availability of active hormones. These reactions and their putative depot-specific interaction with glucocorticoids may contribute to determine body fat distribution patterns in humans.

251

ACKNOWLEDGEMENTS

Studies from our group cited in this manuscript were funded by operating grants from the Canadian Institutes of Health Research to A Tchernof and Co-investigators (MOP-53195, MOP-102642 and MOP-130313).

REFERENCE ANNOTATIONS

Reference 37: Of Interest – This study describes the stimulatory effect of glucocorticoids on DHT inactivation in adipose tissue and demonstrates that 3α-HSD-3 (AKR1C2) is responsible for this phenomenon.

Reference 50: Of interest – This study demonstrates that long-term testosterone therapy ameliorates features of the metabolic syndrome and can be clinically useful through such reduction of cardiometabolic risk in hypogonadal men.

Reference 51: Of interest – This study shows that long-term testosterone therapy resulted in significant improvements of cardiometabolic risk factors and produced clinical benefits in obese, diabetic men.

252

REFERENCES

[1] Tran BX, Nair AV, Kuhle S, Ohinmaa A, Veugelers PJ. Cost analyses of obesity in Canada: scope, quality, and implications. Cost effectiveness and resource allocation C/E. 2013;11:3. [2] Yang Z, Zhang N. The Burden of Overweight and Obesity on Long-term Care and Medicaid Financing. Medical care. 2014;52:658-63. [3] Veilleux A, Tchernof A. Sex differrences in body fat distribution. In: Symonds ME, editor. Adipose Tissue Biology. First ed. Nottingham: Springer; 2012. p. 123-66. [4] Lemieux S, Prud'homme D, Bouchard C, Tremblay A, Després JP. Sex differences in the relation of visceral adipose tissue accumulation to total body fatness. Am J Clin Nutr. 1993;58:463-7. [5] Wells JC. Sexual dimorphism of body composition. Best Pract Res Clin Endocrinol Metab. 2007;21:415-30. [6] Siervogel RM, Demerath EW, Schubert C, Remsberg KE, Chumlea WC, Sun S, et al. Puberty and body composition. Horm Res. 2003;60:36-45. [7] Van Loan MD. Total body composition: birth to old age. In: Roche AF, Heymsfield SB, Lohman TG, editors. Human body composition. Champaign, IL: Human Kinetics; 1996. p. 205-15. [8] Tchernof A, Després JP. Pathophysiology of human visceral obesity: an update. Physiol Rev. 2013;93:359-404. [9] Tchernof A. Sex differences in energy balance, body composition and body fat distribution. In: Brown FM, Wyckoff J, Tsatsoulis A, Editors. Diabetes in women 2009. [10] Smith JD, Borel AL, Nazare JA, Haffner SM, Balkau B, Ross R, et al. Visceral adipose tissue indicates the severity of cardiometabolic risk in patients with and without type 2 diabetes: results from the INSPIRE ME IAA study. JCEM. 2012;97:1517-25. [11] Borel AL, Nazare JA, Smith J, Almeras N, Tremblay A, Bergeron J, et al. Visceral and not subcutaneous abdominal adiposity reduction drives the benefits of a 1-year lifestyle modification program. Obesity. 2012;20:1223-33. [12] Després JP, Lemieux I. Abdominal obesity and metabolic syndrome. Nature. 2006;444:881-7. [13] Bergman RN, Van Citters GW, Mittelman SD, Dea MK, Hamilton-Wessler M, Kim SP, et al. Central role of the adipocyte in the metabolic syndrome. J Investig Med. 2001;49:119-26. [14] Mauriège P, Galitzky J, Berlan M, Lafontan M. Heterogeneous distribution of beta and alpha-2 adrenoceptor binding sites in human fat cells from various fat deposits: functional consequences. Eur J Clin Invest. 1987;17:156-65. [15] Hellmér J, Marcus C, Sonnenfeld T, Arner P. Mechanisms for differences in lipolysis between human subcutaneous and omental fat cells. J Clin Endocrinol Metab. 1992;75:15-20.

253

[16] Arner P. Differences in lipolysis between human subcutaneous and omental adipose tissues. Ann Med. 1995;27:435-8. [17] Tchernof A, Bélanger C, Morisset AS, Richard C, Mailloux J, Laberge P, et al. Regional differences in adipose tissue metabolism in women: Minor effect of obesity and body fat distribution. Diabetes. 2006;55:1353-60. [18] Björntorp P. "Portal" adipose tissue as a generator of risk factors for cardiovascular disease and diabetes. Arteriosclerosis. 1990;10:493-6. [19] Guo Z, Jensen MD. Intramuscular fatty acid metabolism evaluated with stable isotopic tracers. J Appl Physiol. 1998;84:1674-9. [20] Basu A, Basu R, Shah P, Vella A, Rizza RA, Jensen MD. Systemic and regional free fatty acid metabolism in type 2 diabetes. Am J Physiol Endocrinol Metab. 2001;280:E1000-E6. [21] Jensen MD, Cardin S, Edgerton D, Cherrington A. Splanchnic free fatty acid kinetics. Am J Physiol Endocrinol Metab. 2003;284:E1140-E8. [22] Gray SL, Vidal-Puig AJ. Adipose tissue expandability in the maintenance of metabolic homeostasis. Nutr Rev. 2007;65:S7-S12. [23] Drolet R, Richard C, Sniderman AD, Mailloux J, Fortier M, Huot C, et al. Hypertrophy and hyperplasia of abdominal adipose tissues in women. Int J Obes. 2008;32:283-91. [24] Tchoukalova YD, Votruba SB, Tchkonia T, Giorgadze N, Kirkland JL, Jensen MD. Regional differences in cellular mechanisms of adipose tissue gain with overfeeding. Proc Natl Acad Sci USA. 2010;107:18226-31. [25] Carpentier AC, Labbé SM, Grenier-Larouche T, Noll C. Abnormal dietary fatty acid metabolic partitioning in insulin resistance and Type 2 diabetes. Clin Lipidol. 2013;6:703-16. [26] Trayhurn P, Wood IS. Signalling role of adipose tissue: adipokines and inflammation in obesity. Biochem Soc Trans. 2005;33:1078-81. [27] Wisse BE. The inflammatory syndrome: the role of adipose tissue cytokines in metabolic disorders linked to obesity. J Am Soc Nephrol. 2004;15:2792-800. [28] Yudkin JS, Kumari M, Humphries SE, Mohamed-Ali V. Inflammation, obesity, stress and coronary heart disease: is interleukin-6 the link? Atherosclerosis. 2000;148:209- 14. [29] Hauner H. Secretory factors from human adipose tissue and their functional role. Proc Nutr Soc. 2005;64:163-9. [30] Fischer-Posovszky P, Wabitsch M, Hochberg Z. Endocrinology of adipose tissue - an update. Horm Metab Res. 2007;39:314-21. [31] Patsouris D, Li PP, Thapar D, Chapman J, Olefsky JM, Neels JG. Ablation of CD11c- positive cells normalizes insulin sensitivity in obese insulin resistant animals. Cell Metab. 2008;8:301-9. [32] Olefsky JM, Glass CK. Macrophages, inflammation, and insulin resistance. Annu Rev Physiol. 2010;72:219-46. [33] Bain J. The many faces of testosterone. Clinical interventions in aging. 2007;2:567- 76.

254

[34] Corona G, Monami M, Rastrelli G, Aversa A, Tishova Y, Saad F, et al. Testosterone and metabolic syndrome: a meta-analysis study. J Sex Med. 2011;8:272-83. [35] Matsumoto T, Shiina H, Kawano H, Sato T, Kato S. Androgen receptor functions in male and female physiology. J Steroid Biochem Mol Biol. 2008;109:236-41. [36] Auchus RJ. The backdoor pathway to dihydrotestosterone. Trends Endocrinol Metab. 2004;15:432-8. [37] Veilleux A, Cote JA, Blouin K, Nadeau M, Pelletier M, Marceau P, et al. Glucocorticoid-induced androgen inactivation by aldo-keto reductase 1C2 promotes adipogenesis in human preadipocytes. Am J Physiol Endocrinol Metab. 2012;302:E941-E9. [38] Labrie F. DHEA, important source of sex steroids in men and even more in women. Prog Brain Res. 2010;182:97-148. [39] Blouin K, Després JP, Couillard C, Tremblay A, Prud'homme D, Bouchard C, et al. Contribution of age and declining androgen levels to features of the metabolic syndrome in men. Metabolism. 2005;54:1034-40. [40] Traish AM, Kang HP, Saad F, Guay AT. Dehydroepiandrosterone (DHEA)--a precursor steroid or an active hormone in human physiology. J Sex Med. 2011;8:2960-82. [41] Tchernof A, Labrie F. Dehydroepiandrosterone, obesity and cardiovascular disease risk: a review of human studies. Eur J Endocrinol. 2004;151:1-14. [42] Côté JA, Lessard J, Mailloux J, Laberge PY, Rhéaume C, Tchernof A. Circulating 5 α -dihydrotestosterone, abdominal obesity, and adipocyte characteristics in women. Horm Mol Biol Clin Invest. 2012;12:391-400. [43] Blouin K, Boivin A, Tchernof A. Androgens and body fat distribution. Journal of Steroid Biochemistry and Molecular Biology. 2008;108:272-80. [44] Pasquali R, Casimirri F, Cantobelli S, Melchionda N, Morselli Labate AM, Fabbri R, et al. Effect of obesity and body fat distribution on sex hormones and insulin in men. Metabolism. 1991;40:101-4. [45] Gapstur SM, Gann PH, Kopp P, Colangelo L, Longcope C, Liu K. Serum androgen concentrations in young men: a longitudinal analysis of associations with age, obesity, and race. The CARDIA male hormone study. Cancer Epidemiol Biomarkers Prev. 2002;11:1041-7. [46] Laaksonen DE, Niskanen L, Punnonen K, Nyyssonen K, Tuomainen TP, Valkonen VP, et al. Testosterone and sex hormone-binding globulin predict the metabolic syndrome and diabetes in middle-aged men. Diabetes Care. 2004;27:1036-41. [47] Kaufman JM, Vermeulen A. The decline of androgen levels in elderly men and its clinical and therapeutic implications. Endocr Rev. 2005;26:833-76. [48] Seftel A. Male hypogonadism. Part II: etiology, pathophysiology, and diagnosis. Int J Impot Res. 2006;18:223-8. [49] Boyanov MA, Boneva Z, Christov VG. Testosterone supplementation in men with type 2 diabetes, visceral obesity and partial androgen deficiency. Aging Male. 2003;6:1-7.

255

[50] Traish AM, Haider A, Doros G, Saad F. Long-term testosterone therapy in hypogonadal men ameliorates elements of the metabolic syndrome: an observational, long-term registry study. Int J Clin Pract. 2014;68:314-29. [51] Haider A, Yassin A, Doros G, Saad F. Effects of long-term testosterone therapy on patients with "diabesity": results of observational studies of pooled analyses in obese hypogonadal men with type 2 diabetes. Int J Endocrinol. 2014;2014:683515. [52] Mårin P, Holmäng S, Jönsson L, Sjöström L, Kvist H, Holm G, et al. The effects of testosterone treatment on body composition and metabolism in middle-aged and obese men. Int J Obesity. 1992;16:991-7. [53] Woodhouse LJ, Gupta N, Bhasin M, Singh AB, Ross R, Phillips J, et al. Dose- dependent effects of testosterone on regional adipose tissue distribution in healthy young men. J CLin Endocrinol Metab. 2004;89:718-26. [54] Gruenewald DA, Matsumoto AM. Testosterone supplementation therapy for older men: potential benefits and risks. J Am Geriatr Soc. 2003;51:101-15. [55] Elbers JMH, Asscheman H, Seidell JC, Megens JA, Gooren LJG. Long-term testosterone administration increases visceral fat in female to male transsexuals. J Clin Endocrinol Metab. 1997;82:2044-7. [56] Elbers JMH, Giltay EJ, Teerlink T, Scheffer PG, Asscheman H, Seidell JC, et al. Effects of sex steroids on components of the insulin resistance syndrome in transsexual subjects. Clin Endocrinol. 2003;58:562-71. [57] Glazer G. Atherogenic effects of anabolic steroids on serum lipid levels. A literature review. Arch Intern Med. 1991;151:1925-33. [58] Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev. 1997;18:774-800. [59] Barber TM, Golding SJ, Alvey C, Wass JA, Karpe F, Franks S, et al. Global adiposity rather than abnormal regional fat distribution characterizes women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2008;93:999-1004. [60] Casson PR, Toth MJ, Johnson JV, Stanczyk FZ, Casey CL, Dixon ME. Correlation of serum androgens with anthropometric and metabolic indices in healthy, nonobese postmenopausal women. J Clin Endocrinol Metab. 2010;95:4276-82. [61] Keller JL, Casson PR, Toth MJ. Relationship of androgens to body composition, energy and substrate metabolism and aerobic capacity in healthy, young women. Steroids. 2011;76:1247-51. [62] Burger HG, Dudley EC, Cui J, Dennerstein L, Hopper JL. A prospective longitudinal study of serum testosterone, dehydroepiandrosterone sulfate, and sex hormone- binding globulin levels through the menopause transition. J Clin Endocrinol Metab. 2000;85:2832-8. [63] Cao Y, Zhang S, Zou S, Xia X. The relationship between endogenous androgens and body fat distribution in early and late postmenopausal women. PLoS One. 2013;8:e58448. [64] Field AE, Colditz GA, Willett WC, Longcope C, McKinlay JB. The relation of smoking, age, relative weight, and dietary intake to serum adrenal steroids, sex

256

hormones, and sex hormone- binding globulin in middle-aged men. J Clin Endocrinol Metab. 1994;79:1310-6. [65] Couillard C, Gagnon J, Bergeron J, Leon AS, Rao DC, Skinner JS, et al. Contribution of body fatness and adipose tissue distribution to the age variation in plasma steroid hormone concentrations in men: the HERITAGE family study. J Clin Endocrinol Metab. 2000;85:1026-31. [66] Tchernof A, Després JP, Bélanger A, Dupont A, Prud'homme D, Moorjani S, et al. Reduced testosterone and adrenal C19 steroid levels in obese men. Metabolism. 1995;44:513-9. [67] de Pergola G, Triggiani V, Giorgino F, Cospite MR, Garruti G, Cignarelli M, et al. The free testosterone to dehydroepiandrosterone sulphate molar ratio as a marker of visceral fat accumulation in premenopausal obese women. Int J Obes. 1994;18:659- 64. [68] De Simone M, Verrotti A, Lughetti L, Palumbo M, Farello G, Di Cesare E, et al. Increased visceral adipose tissue is associated with increased circulating insulin and decreased sex hormone binding globulin levels in massively obese adolescent girls. J Endocrinol Invest. 2001;24:438-44. [69] Barrett-Connor E, Ferrara A. Dehydroepiandrosterone, dehydroepiandrosterone sulfate, obesity, waist-hip ratio, and noninsulin-dependent diabetes in postmenopausal women: the Rancho Bernardo Study. J Clin Endocrinol Metab. 1996;81:59-64. [70] Williams DP, Boyden TW, Pamenter RW, Lohman TG, Going SB. Relationship of body fat percentage and fat distribution with dehydroepiandrosterone sulfate in premenopausal females. J Clin Endocrinol Metab. 1993;77:80-5. [71] Ravaglia G, Forti P, Maioli F, Boschi F, Bernardi M, Pratelli L, et al. The relationship of dehydroepiandrosterone sulfate (DHEAS) to endocrine-metabolic parameters and functional status in the oldest-old. Results from an italian study on healthy free-living over-ninety-year-olds. J Clin Endocrinol Metab. 1996;81:1173-8. [72] O'Reilly MW, House PJ, Tomlinson JW. Understanding androgen action in adipose tissue. J Steroid Biochem Mol Biol. 2014;143c:277-84. [73] Blouin K, Veilleux A, Luu-The V, Tchernof A. Androgen metabolism in adipose tissue: Recent advances. Mol Cell Endocrinol. 2009;301:97-103. [74] Veilleux A, Blouin K, Tchernof A. Mechanisms of androgenic action in adipose tissue. Clin Lipidol. 2009;4:367-78. [75] Ali AT, Hochfeld WE, Myburgh R, Pepper MS. Adipocyte and adipogenesis. Eur J Cel Biol. 2013;92:229-36. [76] Rosen ED, Spiegelman BM. What we talk about when we talk about fat. Cell. 2014;156:20-44. [77] Singh R, Artaza JN, Taylor WE, Gonzalez-Cadavid NF, Bhasin S. Androgens stimulate myogenic differentiation and inhibit adipogenesis in C3H 10T1/2 pluripotent cells through an androgen receptor-mediated pathway. Endocrinology. 2003;144:5081-8. [78] Dieudonne MN, Pecquery R, Leneveu MC, Giudicelli Y. Opposite effects of androgens and estrogens on adipogenesis in rat preadipocytes: evidence for sex and

257

site-related specificities and possible involvement of insulin-like growth factor 1 receptor and peroxisome proliferator-activated receptor gamma2. Endocrinology. 2000;141:649-56. [79] Singh R, Artaza JN, Taylor WE, Braga M, Yuan X, Gonzalez-Cadavid NF, et al. Testosterone inhibits adipogenic differentiation in 3T3-L1 cells: nuclear translocation of androgen receptor complex with beta-catenin and T-cell factor 4 may bypass canonical Wnt signaling to down-regulate adipogenic transcription factors. Endocrinology. 2006;147:141-54. [80] Fujioka K, Kajita K, Wu Z, Hanamoto T, Ikeda T, Mori I, et al. Dehydroepiandrosterone reduces preadipocyte proliferation via androgen receptor. Am J Physiol Endocrinol Metab. 2012;302:E694-704. [81] Huang CK, Tsai MY, Luo J, Kang HY, Lee SO, Chang C. Suppression of androgen receptor enhances the self-renewal of mesenchymal stem cells through elevated expression of EGFR. Biochim Biophys Acta. 2013;1833:1222-34. [82] Gupta V, Bhasin S, Guo W, Singh R, Miki R, Choong K, et al. Effects of dihydrotestosterone on differentiation and proliferation of human mesenchymal stem cells and preadipocytes. Mol Cell Endocrinol. 2008;296:32-40. [83] Blouin K, Nadeau M, Perreault M, Drolet R, Marceau P, Mailloux J, et al. Effects of Androgens on Adipocyte Differentiation and Adipose Tissue Explant Metabolism in Men and Women. Clin Endocrinol. 2009;72:176-88. [84] Chazenbalk G, Singh P, Irge D, Shah A, Abbott DH, Dumesic DA. Androgens inhibit adipogenesis during human adipose stem cell commitment to preadipocyte formation. Steroids. 2013;78:920-6. [85] McNelis JC, Manolopoulos KN, Gathercole LL, Bujalska IJ, Stewart PM, Tomlinson JW, et al. Dehydroepiandrosterone exerts antiglucocorticoid action on human preadipocyte proliferation, differentiation, and glucose uptake. Am J Physiol Endocrinol Metab. 2013;305:E1134-44. [86] Garcia E, Lacasa M, Agli B, Giudicelli Y, Lacasa D. Modulation of rat preadipocyte adipose conversion by androgenic status: involvement of C/EBPs transcription factors. J Endocrinol. 1999;161:89-97. [87] Barbosa-Desongles A, Hernandez C, Simo R, Selva DM. Testosterone induces cell proliferation and cell cycle gene overexpression in human visceral preadipocytes. Am J Physiol Cell Physiol. 2013;305:C355-9. [88] Rodriguez-Cuenca S, Monjo M, Proenza AM, Roca P. Depot differences in steroid receptor expression in adipose tissue: possible role of the local steroid milieu. Am J Physiol Endocrinol Metab. 2005;288:E200-7. [89] Kolditz CI, Langin D. Adipose tissue lipolysis. Curr Opin Clin Nutr Metab Care. 2010;13:377-81. [90] Chaves VE, Frasson D, Kawashita NH. Several agents and pathways regulate lipolysis in adipocytes. Biochimie. 2011;93:1631-40. [91] Zierath JR, Livingston JN, Thorne A, Bolinder J, Reynisdottir S, Lonnqvist F, et al. Regional difference in insulin inhibition of non-esterified fatty acid release from human adipocytes: relation to insulin receptor phosphorylation and intracellular

258

signalling through the insulin receptor substrate-1 pathway. Diabetologia. 1998;41:1343-54. [92] Rebuffé-Scrive M, Mårin P, Björntorp P. Effect of testosterone on abdominal adipose tissue in men. Int J Obes. 1991;15:791-5. [93] Xu X, de Pergola G, Björntorp P. The effects of androgens on the regulation of lipolysis in adipose precursor cells. Endocrinology. 1990;126:1229-34. [94] Hernandez-Morante JJ, Perez-de-Heredia F, Lujan JA, Zamora S, Garaulet M. Role of DHEA-S on body fat distribution: gender- and depot-specific stimulation of adipose tissue lipolysis. Steroids. 2008;73:209-15. [95] Varlamov O, Chu MP, McGee WK, Cameron JL, O'Rourke RW, Meyer KA, et al. Ovarian cycle-specific regulation of adipose tissue lipid storage by testosterone in female nonhuman primates. Endocrinology. 2013;154:4126-35. [96] Dicker A, Ryden M, Näslund E, Muehlen IE, Wiren M, Lafontan M, et al. Effect of testosterone on lipolysis in human pre-adipocytes from different fat depots. Diabetologia. 2004;47:420-8. [97] Anderson LA, McTernan PG, Harte AL, Barnett AH, Kumar S. The regulation of HSL and LPL expression by DHT and flutamide in human subcutaneous adipose tissue. Diab Obes Metab. 2002;4:209-13. [98] Pecquery R, Dieudonne MN, Cloix JF, Leneveu MC, Dausse JP, Giudicelli Y. Enhancement of the expression of the 2-adrenoreceptor protein and mRNA by a direct effect of androgens in white adipocytes. Biochem Biophys Res Commun. 1995;206:112-8. [99] Pecquery R, Leneveu MC, Giudicelli Y. Influence of androgenic status on the alpha 2/beta-adrenergic control of lipolysis in white fat cells: predominant alpha 2- antilipolytic response in testosterone-treated-castrated hamsters. Endocrinology. 1988;122:2590-6. [100] Xu XF, de Pergola G, Björntorp P. Testosterone increases lipolysis and the number of beta- adrenoceptors in male rat adipocytes. Endocrinology. 1991;128:379-82. [101] Bélanger C, Hould FS, Lebel S, Biron S, Brochu G, Tchernof A. Omental and subcutaneous adipose tissue steroid levels in obese men. Steroids. 2006;71:674-82. [102] Li M, Björntorp P. Effects of testosterone on triglyceride uptake and mobilization in different adipose tissues in male rats in vivo. Obesity Res. 1995;3:113-9. [103] Varlamov O, White AE, Carroll JM, Bethea CL, Reddy A, Slayden O, et al. Androgen effects on adipose tissue architecture and function in nonhuman primates. Endocrinology. 2012;153:3100-10. [104] Blouin K, Richard C, Bélanger C, Dupont P, Daris M, Laberge P, et al. Local androgen inactivation in abdominal visceral adipose tissue. J Clin Endocrinol Metab. 2003;88:5944-50. [105] Blouin K, Richard C, Brochu G, Hould FS, Lebel S, Marceau S, et al. Androgen inactivation and steroid-converting enzyme expression in abdominal adipose tissue in men. J Endocrinol. 2006;191:637-49. [106] Blouin K, Blanchette S, Richard C, Dupont P, Luu-The V, Tchernof A. Expression and activity of steroid aldoketoreductases 1C in omental adipose tissue as positive

259

correlates of adiposity in women. Am J Physiol Endocrinol Metab. 2005;288:E398- E404. [107] Blouin K, Nadeau M, Mailloux J, Daris M, Lebel S, Luu-The V, et al. Pathways of Adipose Tissue Androgen Metabolism in Women: Depot Differences and Modulation by Adipogenesis. Am J Physiol Endocrinol Metab. 2009;296:E244-E55. [108] Bujalska IJ, Walker EA, Tomlinson JW, Hewison M, Stewart PM. 11Beta- hydroxysteroid dehydrogenase type 1 in differentiating omental human preadipocytes: from de-activation to generation of cortisol. Endocr Res. 2002;28:449-61. [109] Masuzaki H, Paterson J, Shinyama H, Morton NM, Mullins JJ, Seckl JR, et al. A transgenic model of visceral obesity and the metabolic syndrome. Science. 2001;294:2166-70. [110] Veilleux A, Rheaume C, Daris M, Luu-The V, Tchernof A. Omental adipose tissue type 1 11 beta-hydroxysteroid dehydrogenase oxoreductase activity, body fat distribution, and metabolic alterations in women. J CLin Endocrinol Metab. 2009;94:3550-7. [111] Hartig SM, He B, Newberg JY, Ochsner SA, Loose DS, Lanz RB, et al. Feed-forward inhibition of androgen receptor activity by glucocorticoid action in human adipocytes. Chem Biol. 2012;19:1126-41.

260

Table 1: Studies which have examined the effect of androgens on preadipocyte differentiation

Study Model Androgen Dose Adipogenesis measurement Effect

Huang et al (2013) BMSC from WT - - AR deletion Oil Red O staining Lack of the androgen receptor (AR) in BMSC [81] and ARKO mice enhances self-replication Fujioka et al (2012) 3T3-L1 / 3T3- Testosterone/DHEA 1mM BrdU incorporation DHEA and testosterone inhibit 3T3-L1 preadipocyte [80] F442A / 3T3-A31 Gene expression (PPARγ2) proliferation and differentiation through AR. DHEA decreases 3T3-F442A but not 3T3-A31 proliferation

Singh et al (2006) 3T3-L1 Testosterone/DHT 0-100nM / 0- Oil Red O staining Testosterone and DHT decrease differentiation of [79] 10nM Gene expression (PPARγ2) 3T3-L1 cells through AR-dependent pathway (partially reversed by AR antagonist)

Singh et al (2003) C3H 10T1/2 Testosterone/DHT 0-300 nM Oil Red O staining Testoterone and DHT inhibit differentiation of [77] Gene expression (PPARγ2) C3H10T1/2 cells through AR-dependent pathway (partially reversed by AR antagonist)

Dieudonne et al (2000) SC and visceral Testosterone/DHT 10µM G3PDH activity Testosterone and DHT inhibit differentiation of [78] fat primary cells primary preadipocytes in rats (rat) Chazenbalk et al (2013) SC primary cells Testosterone/DHT 50nM / 5nM Gene expression Testosterone and DHT inhibit differentiation of [84] (human) Immunofluorescence preadipocytes at an early stage (reversed by AR microscopy antagonist)

McNelis et al (2013) Chub-S7 cell DHEA 0-25µM Gene expression : DHEA inhibits preadipocyte proliferation and [85] line (SC/ human) 10nM G3PDH, FABP4, UCP1, LPL differentiation through anti-glucocorticoid effect in PPARG, FASN human preadipocyte cell line (Chub-S7)

Blouin et al (2013) OM and SC Testosterone/DHT 0-1µM G3PDH activity Testosterone and DHT inhibit differentiation of [83] primary cells Oil Red O staining primary SC and OM preadipocytes from both sexes (human) in a dose-dependent manner

Gupta et al (2008) SC, OM and DHT 0-30nM Oil Red O staining DHT inhibits differentiation of hMSC and primary [82] mesenchymal Gene expression (PPARγ, preadipocytes from three different depots cells (human) aP2, Leptin) DHT: Dihydrotestosterone; DHEA: Dehydroepiandrosterone; G3PDH: Glycerol-3-phosphate dehydrogenase; SC: Subcutaneous; OM: omental; BMSC: Bone marrow stem cells; hMSC: human mesenchymal stem cells.

261

262