Implication of lysosomes on glucose-dependent E2F1-driven cell growth control and their novel role in mitotic progression

Eugènia Almacellas i Canals

ADVERTIMENT. La consulta d’aquesta tesi queda condicionada a l’acceptació de les següents condicions d'ús: La difusió d’aquesta tesi per mitjà del servei TDX (www.tdx.cat) i a través del Dipòsit Digital de la UB (diposit.ub.edu) ha estat autoritzada pels titulars dels drets de propietat intel·lectual únicament per a usos privats emmarcats en activitats d’investigació i docència. No s’autoritza la seva reproducció amb finalitats de lucre ni la seva difusió i posada a disposició des d’un lloc aliè al servei TDX ni al Dipòsit Digital de la UB. No s’autoritza la presentació del seu contingut en una finestra o marc aliè a TDX o al Dipòsit Digital de la UB (framing). Aquesta reserva de drets afecta tant al resum de presentació de la tesi com als seus continguts. En la utilització o cita de parts de la tesi és obligat indicar el nom de la persona autora.

ADVERTENCIA. La consulta de esta tesis queda condicionada a la aceptación de las siguientes condiciones de uso: La difusión de esta tesis por medio del servicio TDR (www.tdx.cat) y a través del Repositorio Digital de la UB (diposit.ub.edu) ha sido autorizada por los titulares de los derechos de propiedad intelectual únicamente para usos privados enmarcados en actividades de investigación y docencia. No se autoriza su reproducción con finalidades de lucro ni su difusión y puesta a disposición desde un sitio ajeno al servicio TDR o al Repositorio Digital de la UB. No se autoriza la presentación de su contenido en una ventana o marco ajeno a TDR o al Repositorio Digital de la UB (framing). Esta reserva de derechos afecta tanto al resumen de presentación de la tesis como a sus contenidos. En la utilización o cita de partes de la tesis es obligado indicar el nombre de la persona autora.

WARNING. On having consulted this thesis you’re accepting the following use conditions: Spreading this thesis by the TDX (www.tdx.cat) service and by the UB Digital Repository (diposit.ub.edu) has been authorized by the titular of the intellectual property rights only for private uses placed in investigation and teaching activities. Reproduction with lucrative aims is not authorized nor its spreading and availability from a site foreign to the TDX service or to the UB Digital Repository. Introducing its content in a window or frame foreign to the TDX service or to the UB Digital Repository is not authorized (framing). Those rights affect to the presentation summary of the thesis as well as to its contents. In the using or citation of parts of the thesis it’s obliged to indicate the name of the author.

Universitat de Barcelona Facultat de Farmàcia i Ciències de l’Alimentació Departament de Bioquímica i Fisiologia

Institut d’Investigació Biomèdica de Bellvitge (IDIBELL) Laboratori de Metabolisme del Càncer (LMC)

Implication of lysosomes on glucose-dependent E2F1-driven cell growth control and their novel role in mitotic progression

Doctoral thesis submitted by Eugènia Almacellas i Canals

Doctoral Program in Biomedicine

Directed by:

Dr. Caroline Mauvezin, Institut d’Investigació de Bellvitge (IDIBELL)

Dr. Albert Tauler, Universitat de Barcelona (UB)

Dr. Albert Tauler Dr. Caroline Mauvezin Eugènia Almacellas Co-director Co-director PhD Student

Barcelona, 2018 Anacrusa. Immunostaining of U2OS cells actin cytoskelleton A la meva família, la que m’ha tocat i la que m’ha trobat

ABSTRACT

Lysosomes are the primary degradative organelles in mammalian cells. Indeed, lysosome enzymatic cocktail allows the degradation of a vast repertoire of cellular material through different converging pathways including macroautophagy. Lysosome function relies on both its acidification capacity, mediated by the vacuolar ATPase, and its cytosolic positioning, driven by the motor proteins kinesins and dyneins. Emerging evidences point out a relevant role of lysosomes in cancer progression denoting their important function in cellular homeostasis maintenance. Indeed, lysosomotropic drugs, such as chloroquine-derivatives, are already being used in clinical trials for cancer treatment. This PhD thesis aims to study lysosome functions in cancer cells to identify novel vulnerabilities related to this organelle. Two different questions are addressed herein: 1) The implication of lysosomes on glucose-mediated E2F1-driven mTORC1 activation and 2) The role of lysosomes in cell division.

E2F1 is overexpressed in numerous human cancers, including lung, breast and hepatocellular carcinomas. Traditionally, the major role reported for E2F1 in cancer is the activation of cell cycle. Previously, our group reported that E2F1 regulates cell growth, through the activation of mTORC1, a major regulator of protein synthesis and autophagy and demonstrated that E2F1 induces the anterograde movement of lysosomes, which is associated with translocation of mTOR to lysosomes and v-ATPase activation. Here we demonstrate that E2F1-dependent mTORC1 activation relies on glucose availability. E2F1 transcriptionally regulates several glycolytic enzymes, thus increasing glycolytic flux. More specifically, we described that E2F1 up-regulates the

PFKFB3 isoenzyme of the PFK-2, a potent activator of glycolysis, and that PFKFB3 activity determines mTORC1 activation. We hypothesize that E2F1 is able to activate glycolysis and therefore increase v-ATPase and mTORC1 activity. Besides, E2F1 induces lysosome-dependent exocytosis which correlates with a metastatic phenotype. These novel functions of E2F1 in v-ATPase regulation and lysosomal trafficking provide insight into regulatory mechanisms by which E2F1 drives malignancy and highlight the potential role of lysosomes as a metabolic hub in mammalian cells.

Studies on lysosome function are mainly focused in cells in interphase. However, the implication of these organelles during cell division remains unclear. Mitosis is a key event during cell cycle, in which cells finally divide into two daughter cells. Mitotic progression comprises five active phases involving a dramatic rearrangement of cellular components in a short period of time. Until now, degradation of mitotic factors has been attributed only to ubiquitination and proteasome-dependent degradation.

In the present study, we show that impairment of lysosomal trafficking and function delays mitotic progression and increases mitotic errors, phenomena accompanied by an increase in toroidal-shaped nuclei a reflection of impaired mitosis. Finally, we use a proteomic approach to discover novel lysosome protein substrates involved in mitotic progression. Interestingly, we identified regulatory proteins of the cohesin complex necessary for correct chromosomal segregation. By characterizing a novel function of lysosomes specifically in mitosis, our work establishes a novel model of regulation of cell division beyond the proteasome.

In summary, this work provides new insights into the function of lysosomes down- stream of E2F1 oncogenic signaling, which modulates G1/S transition and cell proliferation but also into lysosomal function in the completion of cell cycleby regulating mitotic progression. Our work highlights the importance of targeting lysosomes for novel cancer therapies. RESUM

Els lisosomes són els principals orgànuls degradatius de la cèl·lula eucariota, capaços de degradar un ampli repertori de material cel·lular. La funció lisosomal depèn tant de la seva acidificació, sota el control de la v-ATPasa, com de la seva distribució citosòlica, regulada per kinesines i dineïnes. La funció lisosomal és essencial per preservar la homeòstasi cel·lular i la seva alteració està lligada al desenvolupament de malalties de diferent índole, des de trastorns neurodegeneratius fins al càncer. De fet, fàrmacs lisosomotròpics, tals com els derivats de la cloroquina, s’estudien actualment en assajos clínics com a tractament per un ampli ventall de tumors. L’objectiu d’aquesta tesi doctoral és l’estudi de les funcions lisosomals en cèl·lules canceroses per tal de definir noves vulnerabilitats relacionades a aquest orgànul. S’adrecen així doncs, dues preguntes experimentals: 1) La implicació dels lisosomes en l’activació de mTORC1 degut a E2F1 a través de la glucosa i 2) El rol dels lisosomes en la divisió cel·lular.

La funció oncogènica d’E2F1 s’ha atribuït tradicionalment al seu paper com a regulador del cicle cel·lular. Actualment, però, es coneix la seva implicació en diverses funcions relacionades amb la progressió tumoral, entre elles, el creixement cel·lular a través de l’activació de mTORC1. En aquest projecte, hem demostrat que l’activació d’mTORC1

és amplificada en presència de glucosa. E2F1 activa la transcripció d’enzims glucolítics, específicament l’isoenzim de la PFK-2 PFKFB3 i d’aquesta manera incrementa el flux glucolític. Hem descrit que l’activitat de PFKFB3 està directament relacionada amb l’activació d’mTORC1 suggerint la regulació del complex pel flux glucolític. En paral·lel, a través de la regulació del tràfic lisosomal, E2F1 incrementa la exocitosi lisosomal

àmpliament correlacionada amb el fenotip metastàtic. Les funcions descrites aquí destaquen la importància del lisosoma en el mecanisme oncogènic d’E2F1 i recolzen la idea dels lisosomes com a plataformes d’integració del metabolisme cel·lular.

L’estudi de la funció lisosomal s’ha limitat majoritàriament a cèl·lules en interfase i pocs estudis es centren en la seva implicació en la divisió cel·lular (mitosi). La mitosi

és un procés crucial en el cicle cel·lular en què la cèl·lula finalment es divideix en dues cèl·lules filles genèticament idèntiques. La progressió mitòtica requereix un ampli canvi morfològic, estrictament coordinat en un període de temps extremadament curt. En aquest context, la degradació de factors mitòtics és un element limitant per garantir la correcta progressió mitòtica. Fins ara, la degradació de factors mitòtics s’ha atribuït exclusivament al proteasoma deixant la funció lisosomal fora de l’escenari. En aquest estudi, demostrem que la inhibició de la funció lisosomal tant a nivell de tràfic com d’acidificació retarda la progressió mitòtica i incrementa la freqüència d’errors mitòtics, fenòmens acompanyats de l’aparició del nucli toroïdal el qual reflexa una mitosi aberrant. Finalment, hem usat una aproximació proteòmica per descobrir noves dianes lisosomals implicades en la mitosi. Entre aquestes, hem identificat proteïnes reguladores del complex de coesines, essencial pel correcte desenvolupament de la segregació cromosòmica. La caracterització de la funció lisosomal específicament en mitosi desenvolupada en aquest estudi implica un nou model de regulació de la divisió cel·lular més enllà de la degradació proteasomal.

Aquest estudi dóna llum a la funció lisosomal sota el senyal oncogènic d’E2F1, un regulador clau de la transició G1/S del cicle cel·lular així com demostra la implicació dels lisosomes per la culminació del cicle a través de la regulació de la mitosi. En conclusió, el nostre treball recolza la importància del lisosoma com a possible diana terapèutica en el càncer.

TABLE OF CONTENTS

TABLE OF CONTENTS...... 1 LIST OF FIGURES...... 5 LIST OF TABLES...... 9 ABBREVIATIONS LIST...... 11 INTRODUCTION...... 17 1. THE LYSOSOME...... 19 Lysosome structure...... 20

Lysosomal membrane proteins...... 20

Lysosomal matrix proteins...... 22

Lysosome acidification: the v-ATPase...... 23

Lysosomal trafficking...... 27

Dynein family...... 27

Kinesin family...... 28

Lysosome functions...... 31

Degradative function...... 31

Exocytosis...... 35

Signaling platform...... 36

Lysosomes in the context of cancer...... 41 2. E2F1 TRANSCRIPTION FACTOR...... 43 E2F transcription factor family...... 44

E2F1 in cancer...... 45

E2F1 functions...... 46

E2F1 regulation of cell cycle...... 46

E2F1 regulation of cell metabolism...... 47

E2F1 regulation of cell growth and lysosome trafficking...... 48

TABLE OF CONTENTS 1 3. MITOTIC CELL DIVISION...... 51 Mitosis progression and regulation...... 52

Mitotic entry: from G2 to metaphase...... 53

Mitotic exit: from metaphase to cytokinesis...... 55

Chromosome instability and cancer progression...... 57 RATIONAL OF THE STUDY AND AIMS...... 59 1. RATIONAL OF THE STUDY...... 61 2. AIMS...... 63 RESULTS...... 65 1. E2F1 REGULATES mTORC1 AND LYSOSOMAL TRAFFICKING THROUGH GLYCOLYSIS INDUCTION...... 67 Glucose potentiates E2F1-induced mTORC1 activation...... 67

E2F1 induces glycolysis...... 74

E2F1 modulates cytosolic pH...... 76

PFKFB3 regulates E2F1-driven mTORC1 activity...... 79

PFK-1, but not aldolase or GAPDH, affects mTORC1 activity...... 84

Glucose availability determines E2F1-dependent lysosome trafficking regulation... 86

E2F1 and v-ATPase drive lysosome-dependent exocytosis and cell migration...... 86 2. LYSOSOMES PREVENT DEFECTIVE MITOSIS AND MEDIATE MITOTIC PROTEINS DEGRADATION...... 91 Lysosomes are acidic vesicles active during cell division...... 91

Lysosome function and trafficking impairment delays mitotic progression...... 94

Mitotic errors correlate with a novel nuclear phenotype: the toroidal nucleus...... 99

Lysosome disruption induces toroidal nucleus...... 105

Macroautophagy prevents defective mitosis...... 111

Novel lysosome protein substrates are implicated in cell division...... 113

2 TABLE OF CONTENTS DISCUSSION...... 119 E2F1 regulates mTORC1 and lysosomal trafficking through glycolysis induction..... 122

Lysosomes prevent defective mitosis and mediate mitotic proteins degradation... 127

Integrative map of lysosomal function in cancer...... 133 CONCLUSIONS...... 137 MATERIALS AND METHODS...... 141 Cell culture...... 143

Protein extraction and Western Blot analysis...... 144

Immunoprecipitation...... 145

Transfection procedures...... 146

Immunofluorescence...... 147

Image processing and analysis...... 149

Vesicle acidification assay: Magic Red...... 149

Cytosolic pH measurement...... 149

Live-cell time-lapse videos...... 150

Real time quantitative PCR (qPCR)...... 151

Transmission electron microscopy (TEM)...... 152

Cell cycle analysis...... 152

Lysosome-dependent exocytosis measurement...... 153

Ultra-Performance Liquid Chromatography (UPLC)...... 154

Glycolytic rate measurement...... 154

Mass spectrometry...... 155

Statistical analysis...... 156 REFERENCES...... 159 ACKNOWLEDGEMENTS / AGRAÏMENTS...... 181

TABLE OF CONTENTS 3 4 LIST OF FIGURES INTRODUCTION Figure 1. Lysosome structure...... 21

Figure 2. v-ATPase structure...... 24

Figure 3. Reverse pH gradient is a feature of cancer cells...... 26

Figure 4. Kinesin family members...... 29

Figure 5. Lysosome motor core complex...... 30

Figure 6. Different types of autophagy in mammalian cells...... 32

Figure 7. mTORC1 regulatory landscape...... 38

Figure 8. E2F family members’ structure...... 44

Figure 9. E2F1 and RB1 alterations in cancer...... 45 Figure 10. E2Fs activity oscillations during cell cycle...... 46 Figure 11. E2F1 regulates metabolism...... 48

Figure 12. E2F1-driven mTORC1 regulation through v-ATPase and lysosomal trafficking modulation...... 49

Figure 13. Mitotic subphases...... 52

Figure 14. Sister chromatids separation during metaphase-anaphase transition...... 56 RESULTS- E2F1 REGULATES mTORC1 AND LYSOSOMAL TRAFFICKING THROUGH GLYCOLYSIS INDUCTION Figure 15. Glucose potentiates E2F1-induced mTORC1 activation...... 68

Figure 16. E2F1 activation maintains energetic balance...... 69

Figure 17. Acute glycolysis blockade inhibits E2F1-driven mTORC1 activation...... 70

Figure 18. E2F1-induced mTORC1 activation relies on glycolysis rather than mitochondrial oxidation...... 71

Figure 19. mTORC1 translocation to the lysosome is sensitive to glucose levels...... 72

Figure 20. mTORC1 lysosomal recruitment is affected by glucose levels...... 73

LIST OF FIGURES 5 Figure 21. Schematic representation of GlycoStress Test...... 74

Figure 22. E2F1 induces glycolysis...... 75

Figure 23. E2F1-induced glycolysis potentiation is independent of mTORC1 activation...... 75

Figure 24. E2F1 activation increases cytosolic ...... pH 76

Figure 25. E2F1 modulates the expression of pH regulators...... 77

Figure 26. E2F1 up-regulates pH regulators role on mTORC1 activation...... 78

Figure 27. MCT1 depletion does not affect cytosolic pH...... 78

Figure 28. E2F1 modifies the expression of several glycolytic enzymes...... 80

Figure 29. E2F1 regulates the expression of PFKFB3...... 80

Figure 30. PFKFB3 inhibition reduces mTORC1 activity...... 82

Figure 31. TIGAR overexpression reduces mTORC1 activity...... 83

Figure 32. PFKFB3 overexpression induces mTORC1 activity...... 83

Figure 33. PFK-1 plays a role on E2F1-driven mTORC1 activation...... 85

Figure 34. GAPDH or aldolase A deletion increase mTORC1 activity...... 85

Figure 35. Glucose availability affects lysosomes peripheral positioning induced by E2F1...... 87

Figure 36. E2F1 induces lysosome-dependent exocytosis in a v-ATPase-dependent manner...... 88

Figure 37. Lysosome activity is preserved from perinuclear to peripheral lysosomes...... 88

Figure 38. E2F1 and v-ATPase affect cell motility...... 89

Figure 39. E2F1 and v-ATPase promote cell migration and invasion...... 90 RESULTS - LYSOSOMES PREVENT DEFECTIVE MITOSIS AND MEDIATE MITOTIC PROTEINS DEGRADATION Figure 40. Lysosomes shape and number are modified in dividing cells...... 91

Figure 41. Lysosomes are active vesicles in mitotic cells...... 92

Figure 42. Autophagy persists during mitosis...... 93

6 LIST OF FIGURES Figure 43. Autophagic vesicles and lysosomes are present in dividing cells...... 94

Figure 44. Lysosome impairment delays mitotic progression...... 95

Figure 45. Lysosome impairment delays mitotic entry and exit...... 96

Figure 46. Number of mitosis is reduced upon v-ATPase inhibition...... 97

Figure 47. Lysosome impairment increases mitotic errors...... 98

Figure 48. Nucleus with chromatin-free structure compartment is formed upon mitosis...... 99

Figure 49. Chromatin-free structure does not correspond to enlarged nucleolus.. 100

Figure 50. Toroidal nucleus characterization...... 101

Figure 51. Toroidal nucleus containing cells can undergo DNA replication and cell division...... 102

Figure 52. Nuclear envelope formation precedes toroidal nucleus appearance..... 103

Figure 53. Toroidal nucleus formation correlates with mitotic errors occurrence... 104

Figure 54. Mitosis impairment generates toroidal nucleus...... 104

Figure 55. Lysosome acidification inhibition increases toroidal nucleus formation.106

Figure 56. Toroidal nucleus population do not persist after a replication cycle...... 108

Figure 57. Cell death and mitosis correlate in cells treated with Concanamycin A. 109

Figure 58. Lysosome trafficking impairment increases toroidal nucleus formation.110

Figure 59. Combination of lysosome trafficking impairment and lysosome acidification blockade are additive on toroidal nucleus formation...... 111

Figure 60. Autophagic proteins depletion causes toroidal nucleus formation...... 112

Figure 61. Atg5 KO MEFs exhibit a higher percentage of toroidal nucleus...... 113

Figure 62. Cell cycle analysis of cellular functions for Mass Spectrometry analysis.114

Figure 63. Representation of pathways affected by lysosome impairment during cell division...... 115

Figure 64. Gene ontology analysis of all proteins up-regulated or only present in Concanamycin A treated cells...... 115

Figure 65. Functional annotation clustering of proteins related to cell cycle regulation and mitotic progression...... 116

LIST OF FIGURES 7 Figure 66. Pds5, WAPL and PLK1 accummulate in mitotic cells upon lysosome inhibition...... 117 DISCUSSION Figure 67. Model of E2F1-induced cellular processes...... 126

Figure 68. Impairment of lysosomes in mitotic cells leads to chromosomal instability...... 132

Figure 69. Lysosomes as gatekeepers of cellular homeostasis and cell cycle progression in the context of cancer...... 134 MATERIALS AND METHODS Figure 70. Validation of Cell Event dye for cell death detection...... 150

Figure 71. Schematic representation of cell synchronization at. G2 ...... 153

8 LIST OF FIGURES LIST OF TABLES INTRODUCTION Table 1. Lysosome matrix proteins...... 22 RESULTS Table 2. Toroidal nucleus can be detected in a variety of cell lines...... 107 MATERIALS AND METHODS Table 3. Chemicals and drugs used...... 143

Table 4. List of primary and secondary antibodies used for Western Blotting...... 144

Table 5. List of plasmids used...... 146

Table 6. List of siRNA used...... 147

Table 7. List of primary and secondary antibodies used for immunofluorescence. 148

Table 8. List of primers used in real time quantitative PCR...... 151

LIST OF TABLES 9 10 ABBREVIATIONS LIST

ABBREVIATION COMPLETE NAME 2DG 2-deoxyglucose 3PO 3-(3-pyridinyl)-1-(4-pyridinil)-2-propen-1-one 4EBP 4E-Binding Protein A, B, C ACC Acetyl-CoA carboxylase ALDO A Aldolase A AMP Adenosine monophosphate AMPK AMP-activated kinase APC/C Anaphase promoting complex Arl8 ADP-ribosylation factor-like protein 8 Atg Autophagy-related proteins ATP Adenosine triphosphate BLOS 1/2 Biogenesis of lysosome-related organelles 1/2 BORC BLOC-one-related Complex CAD Carbamoyl-phosphate synthetase CA Carbonic anhydrases CASTOR Cytosolic arginine sensor for mTORC1 Cdc20 Cell division cycle protein 20 homolog Cdh1 APC/C activator protein Cdk Cyclin dependent kinase CENPE Centromere-associated protein E CHMP Charged multivesicular body proteins CIN Chromosome instability CMA Chaperone-mediated autophagy c-Myc Myc proto-oncogene protein Conc A Concanamycin A D, E, F DAPI 4,6-diamidino-2-penylinole DCTN1 Dynactin subunit 1 DDR DNA damage response DEPTOR Domain-containing mTOR-interacting protein DFCP1 Zinc finger FYVE domain-containing protein 1 DP Dimerization Partner DSP Dithiobis(succinimidyl propionate) E2F 1-8 E2F transcription factors 1-8

ABBREVIATIONS LIST 11 ECAR Extracellular acidification rate EE Early endosome ER Endoplasmic reticulum ESCRT-III Endosomal sorting complex required for transport III

Fru 1,6-P2 Fructose 1,6-bisphosphate

Fru 2,6-P2 Fructose 2,6-bisphosphate Fru 6-P Fructose 6-phosphate FYCO1 FYVE and coiled-coil domain-containing protein 1 G, H, I GADPH Glyceraldehyde-3-phosphate dehydrogenase GAP GTPase-activating protein GATOR GTPase activating proteins toward Rags complex GEF Guanine nucleotide Exchange Factor GO Gene ontology H2A / H2AX /gH2AX Histone 2A / 2AX / gamma H2AX H2B Histone 2B HCQ Hydroxychloroquine HEX β-hexosaminidase HIF1α Hypoxia inducible factor 1α HOPS homotypic fusion and protein sorting complex Hsp70 Heat-shock protein 70 INM Inner nuclear membrane J, K, L KIF Kinesin-like proteins KLC Kinesin Light Chain LAMP1/2 Lysosome associated membrane protein 1/2 LAMTOR Ragulator complex protein LC3 Microtubule-associated protein 1A/1B-light chain 3 LE Late endosome LIMP2 Lysosomal integral membrane protein 2 LINC Linker of nucleoskeleton and cytoskeleton LIR LC3 interacting region LKB1 Liver kinase B1 LMP Lysosome membrane protein M, N, O MCAK Mitotic centromere-associated kinesin/KIF2C MCT 1/4/7 Monocarboxylate transporter 1/4/7 MEF Mouse embryonic fibroblast mLST8 Mammalian lethal with SEC13 protein 8

12 ABBREVIATIONS LIST MTOC Microtubule organizing center mTOR Mammalian target of rapamycin mTORC1/2 mTOR complex 1/2 MVB Multivesicular body NE Nuclear envelope NEBD Nuclear envelope breakdown NHE 1 Na+ /H+ exchanger 1 NPC Nuclear pore complex NUP Nucleoporin OHT 4-hydroxytamoxifen OLIGO Oligomycin A P, Q, R PDK1/3 Pyruvate dehydrogenase kinase 1/3 PDMS Poly(dimethylsiloxane) PDS5 Cohesin associated factor B PE Phosphatidylethanolamine PFK-1 Phosphofructokinase-1 PFK-2 / FBPase-2 Phosphofructokinase-2 / Fructose bisphosphatase-2 PFKFB3 6-Phosphofructo-2-Kinase/Fructose-2,6-Biphosphatase isoenzyme 3 PFKL / M / P PFK-1 isoenzyme L / M / P PI3KC3 / Vps34 Phosphatidylinositol 3-kinase class 3 PI3P Phosphatidylinositol 3-phosphate PKA cyclic-AMP-dependent kinase PLK1 Polo-like kinase 1 PRAS40 proline-rich AKT substrate of 40 kDa pRB Retinoblastoma protein Rab7 Ras-related protein RagA/B/C/D Ras-related GTP-binding proteinA/B/C/D Raptor Regulatory-associated protein of mTOR Rheb RAS-homolog enriched in brain RILP Rab-interacting lysosomal protein S, T, U S6K1 p70S6 kinase 1 SAC Spindle assembly checkpoint SKIP / PLEKHM2 Pleckstrin homology domain-containing family M member 2 SLC38A9 Solute carrier 38 A9 SNAP29 synaptosomal-associated protein 29 SNARE Soluble N-ethylmaleimide-sensitive factor activating protein receptor SQSTM1 / p62 Sequestosome-1

ABBREVIATIONS LIST 13 SREBP Sterol regulatory element-binding protein STX17 Syntaxin 17 TEM Transmission electron microscopy TFEB Transcription factor EB TIGAR TP53-induced glycolysis and apoptosis regulator TOS TOR signaling TSC 1/2 Tuberous sclerosis complex 1/2 UBA Ubiquitin-associated domain ULK1 Unc-51 like autophagy activating kinase 1 UPLC Ultra-performance liquid chromatography UPS Ubiquitin-proteasome system V, W, X, Y, Z VAMP8 Vesicle-associated membrane protein 8 v-ATPase Vacuolar-ATPase WAPL Wings apart-like protein WIPI2 WD repeat domain phosphoinositide-interacting protein 2

14 ABBREVIATIONS LIST

Multiply or divide? Immunostaining of U2OS cells actin cytoskelleton INTRODUCTION

1. THE LYSOSOME

The term “lysosome” was originally coined by Christian de Duve in 1955 to describe a newly discovered organelle full of soluble hydrolases capable to degrade biological material (de Duve, 2005). Later studies with electron microscopy showed that lysosomes constitute up to 5% of the intracellular volume and are heterogeneous in size and morphology (Novikoff et al., 1956). Over 50 different acid hydrolases are contained within the lysosome, which offer a degradation capacity of a vast repertoire of biological substrates such as proteins, nucleic acids, carbohydrates, lipids and cellular debris. The breakdown products are then transported back to the cytosol by specific transporters localized in the lysosomal membrane, thus making the lysosome a key organelle for cellular homeostasis maintenance.

Lysosomes have a remarkable role in various cellular processes, including microbial killing, antigen presentation, detoxification, cholesterol homeostasis, apoptosis, metabolic signaling, plasma membrane repair and cell migration and invasion (Fennelly and Amaravadi, 2017; Luzio et al., 2007). Notably, lysosome positioning has emerged as a key phenomenon controlling at least some of these functions.

I will summarize here the current knowledge of lysosome structure and the regulation of the vacuolar-ATPase proton pump (v-ATPase), the lysosome transport machinery, as well as the different functions currently attributed to lysosomes.

INTRODUCTION 19 Lysosome structure

Lysosomes are vesicular organelles found in all animal cells except erythrocytes (Figure 1). The lysosomal membrane corresponds to a single lipid bilayer that contributes to the acidification of the lysosomal lumen, sequestration of lysosomal enzymes, membrane fusion events and transport of degraded products to the cytoplasm. The execution of lysosomal catabolic events occurs in the highly acidic lumen (pH 4.5) that is maintained by the constant pumping of protons into the lysosome mainly by the v-ATPase (Figure 1). This proton gradient also provides the driving force for proton- coupled transport of metabolites, ions and soluble substrates across the lysosomal membrane (Ishida et al., 2013). Perturbation of this gradient leads to inefficient cargo sorting, altered membrane traffic and impaired cellular clearance; stresses that will potentially activate cell death programs (Mindell, 2012; Xu and Ren, 2015).

A layer of luminal glycoproteins protects the lysosomal membrane from degradation by the enzymatic content Figure( 1). Thus, one function of the lysosomal membrane is to separate this harmful acidic environment from the cytosol. Lysosomal membrane permeabilization and the consequent leakage of the lysosomal content into the cytosol leads to lysosomal-cell death. This form of cell death is executed by the cathepsins resident in the lysosomal lumen which will activate apoptosis (Boya and Kroemer, 2008).

Based on recent studies, over 100 bona fide lysosomal resident proteins have been identified: approximately 70 lysosomal matrix proteins and around 50 lysosomal membrane proteins (LMPs). However, these numbers are likely to increase in the near future (Lübke et al., 2009; Schröder et al., 2010). Deficiencies in LMPs or lysosomal hydrolases may lead to lysosomal storage disorders, a group of human pathologies driven by the accummulation of lysosome substrates that can not be properly transported or degraded within the lysosome (Ballabio and Gieselmann, 2009). Lysosomal membrane proteins

Among the known LMPs, the most abundant correspond to Lysosome associated membrane proteins (LAMP1 and 2) and Lysosome integral membrane protein (LIMP2) (Figure 1). These multifunction proteins are involved in the protection of the lysosomal membrane degradation, vesicle fusion and the recruitment of specific enzymes to the lysosome (Schwake et al., 2013).

20 INTRODUCTION Figure 1. Lysosome structure. Simplified view of a lysosome including proteins separated by function (two axis). Protein examples involved in trafficking/fusion processes, Ion transport, metabolite trans- port and lysosomal structure are shown. CLC7: Cl-/H+ antiporter; CFTR: Cystic fibrosis transmembrane conductance regulator; NPC1: Niemann-Pick C1 cholesterol transporter; LIMP: Lysosome Integral Mem- brane Protein; LAMP1/2: Lysosome associated membrane protein 1 / 2; SNAREs: Soluble N-ethylmalei- mide-sensitive factor activating protein receptor; Rab7: Ras-related protein; BORC: BLOC-one-related complex. SLC38A9: Solute carrier 38 A9. V-ATPase: vacuolar-ATPase.

The glycoproteins LAMP1 and LAMP2 are also implicated in the recognition of specific cytosolic substrates, lysosome biogenesis, autophagy, melanosome biogenesis or lysosome exocytosis (Couto et al., 2017; Eskelinen, 2006; Schwake et al., 2013). LAMP1 and 2 molecules are commonly used as lysosomal markers to identify the lysosomal compartment. Interestingly LAMP2 undergoes alternative splicing leading to the generation of LAMP2A, B and C. LAMP2A has specific functions in chaperone- mediated autophagy (CMA) as discussed later (Kaushik and Cuervo, 2012). Mutations in LAMP2 cause Danon disease, a lysosomal glycogen storage disease (Rowland et

INTRODUCTION 21 al., 2016). Although LAMP1 or LAMP2 knockout mice are able to survive even with some defects in autophagy (Andrejewski et al., 1999; Stypmann et al., 2006; Tanaka et al., 2000), the double knockout mice (KO) are lethal at embryonic days 14.5-16.5 (Eskelinen, 2006; Eskelinen et al., 2004), suggesting overlapping functions between the family members compensated in the single KOs.

On the other hand, LIMP2 is required for lysosome biogenesis and maintenance and has been involved in targeting some lysosomal proteins such as glucocerebrosidase acting as a receptor (Gonzalez et al., 2014). Mutations of LIMP2 gene may lead to Gaucher disease (Rosenbloom and Weinreb, 2013). Lysosomal matrix proteins Lysosomes contain several matrix proteins which essentially degrade biological material, among them glycosidases, proteases and sulfatases (Table 1). Activity of lysosome matrix enzymes is optimal at pH 4.5, which is specific of the lysosomal lumen, thus restricting this degradative function to lysosomes. The lysosomal acidic pH is facilitated by the v-ATPase proton pump.

GLYCOSIDASES alpha-Galactosidase A Chondroitinase ABC Hyaluronan Lyase alpha-N-acetylgalactosamini- Chondroitinase AC Hyaluronidase 1 dase alpha-Galactosidase Cytosolic beta-Glucosidase Hyaluronidase 4 alpha-L-Fucosidase Endo-beta-Galactosidase alpha-L-Iduronidase Endo-beta-N-acetylglucosamin- Tissue alpha-L-Fucosidase Klotho idase F1 Endo-beta-N-acetylglucosamin- AMCase Klotho beta idase F3 Endo-beta-N-acetylglucosamin- beta-Galactosidase-1 Lactase-like Protein idase H beta-Glucuronidase Galactosylceramidase Lysosomal alpha-Glucosidase beta (1-3)-Galactosidase Glucosylceramidase MBD4 beta (1-4)-Galactosidase Heparanase alpha-N-acetylglucosaminidase Chitinase 3-like 1 Heparinase I Sialidase-1 Chitinase 3-like 2 Heparinase II O-GlcNAcase Chitobiase Heparinase III O-Glycosidase Chitotriosidase Hexosaminidase A Chondroitinase B Hexosaminidase B PROTEASES AMSH/STAMBP Cathepsin D Cathepsin V

22 INTRODUCTION Cathepsin 3 Cathepsin F Cathepsin X / Z / P Cathepsin 6 Cathepsin H Galactosylceramidase Cathepsin 7/Cathepsin 1 Cathepsin K Asparaginyl Endopeptidase Cathepsin A Cathepsin L Napsin A Cathepsin B Cathepsin O Cathepsin C Cathepsin S SULFATASES Glucosamine (N-acetyl)-6-Sul- Arylsulfatase A Sulfamidase fatase Arylsulfatase B Iduronate 2-Sulfatase Sulfatase-2 N-Acetylgalactosamine-6-Sulfa- Arylsulfatase G tase

Table 1. Lysosome matrix proteins. Lysosome matrix enzymes are mainly divided into glycosidases for glycosidic bounds degradation, proteases for proteolysis and sulfatases for cleavage of O- and N-sulfate esters from a variety of substrates (From The Human Lysosome Gene Database (Brozzi et al., 2013)). Lysosomal acidification: the v-ATPase

The v-ATPase is a highly conserved enzyme which uses the energy of ATP hydrolysis to pump protons into the lysosomal lumen. v-ATPase participates in several cellular processes, such as pH regulation, endocytosis, intracellular trafficking, maturation of endosomes and acidification of recycling vesicles (Cotter et al., 2015a).

Indeed, the progressive acidification of endocytic vesicles and the maintenance of their specific pH is controlled by the v-ATPase, ranging from 6.8-6.1 in early endosomes (EEs) and culminating in 6.0 – 4.8 in late endosomes (LEs) and about 4.5 in lysosomes. The proton pumping activity of the v-ATPase generates a transmembrane voltage which has to be compensated by either cation efflux or anion influx to allow lysosome acidificationFigure ( 1) (Ishida et al., 2013; Mindell, 2012).

Structure of the v-ATPase

Mammalian v-ATPase (formally referred as ATP6) is composed by 13 subunits divided into two structural domains (Figure 2): the cytosolic V1 domain consisting of eight subunits (ATP6V1A-H) which is responsible for the ATP hydrolysis and the transmembrane V0 domain which consists of five subunits (ATP6V0a-e) and allows the proton pumping (Cotter et al., 2015a; Marshansky et al., 2014).

INTRODUCTION 23 Some v-ATPase subunits have different isoforms which can be either ubiquitous or selectively expressed in specific cell types and perform specialized functions. Once V1 subunit docks at the lysosomal surface and binds to V0 complex, the proton pump becomes functional. The modular structure of the v-ATPase and the diversity of subunit isoforms allow the complex to adapt to different cellular and tissue environments and to be independently regulated (Capecci and Forgac, 2013; Son et al., 2016).

Figure 2. v-ATPase structure. v-ATPases are characterized by a rotatory mechanism that starts in the V1 domain after ATP hydrolysis and involves the V0 proteolipid domain, which is target of the lysosomo- tropic drug Concanamycin A (Bowman et al., 2004; Huss et al., 2002). Once V0 rotates, two protons are transferred from the cytoplasm to the lysosomal lumen. Relevant information about the proton pump is shown.

Regulation of the v-ATPase

Several mechanisms have evolved to coordinate the v-ATPase function such as transcriptional regulation of the composing subunits or the assembly/disassembly of the V1/V0 structural domains (Lee, 2012). However, the molecular mechanisms comprising the v-ATPase regulation in mammalian cells are only partially understood. Most of research on this field has been performed in yeast dissecting the regulation of the multimeric proton pump and more recently this knowledge is being translated to human cells.

Glucose starvation was proposed to trigger the V1/V0 reversible disassembly and consequent inhibition (Parra and Kane, 1998). Part of this regulation is mediated by

24 INTRODUCTION glycolytic enzymes that interact with the v-ATPase and modulate its function. Both in mammalian and yeast models, ATP6V1E, ATP6V0a and ATP6V1B subunits interact with aldolase, which is sensitive to the presence of glucose in the medium (Lu et al., 2001, 2007) (Figure 2). Additionally, phosphofructokinase 1 (PFK-1), a rate-limiting glycolytic enzyme, interacts with the kidney-specific v-ATPase subunit a4 andthe ubiquitous a1 (Su et al., 2003) (Figure 2). v-ATPase function can also be regulated by direct phosphorylation of the A-subunit by cyclic-AMP-dependent kinase (PKA), which has been suggested to be triggered by glucose starvation (Alzamora et al., 2010; Dechant et al., 2010). Functions of the v-ATPase

Besides its role regulating lysosomal pH, v-ATPase is involved in the regulation of endosome maturation, autophagy, vesicle intracellular trafficking and metabolic signaling (Rabanal-Ruiz and Korolchuk, 2018; Saftig and Klumperman, 2009; Settembre et al., 2013; Zhang et al., 2014). Localization of v-ATPase is not exclusive to the lysosome but it has been also found in the plasma membrane where it exerts a pivotal function on renal acidification, bone resorption, sperm maturation and cytosolic pH homeostasis (Forgac, 2007; Toyomura et al., 2003).

Several studies showed that v-ATPase hyperactivation is, in fact, associated with malignant transformation, invasion and metastasis (Cotter et al., 2015b; Hinton et al., 2009; McGuire et al., 2016). Indeed, a reverse pH gradient is a hallmark of cancer metabolism characterized by extracellular acidosis and intracellular alkalinization (Figure 3). Extracellular acidosis is related to drug resistance, invasiveness and extracellular matrix degradation (Kato et al., 2013).

V-ATPase localization in the plasma membrane together with higher expression of proton transporters has been described in cancer cells supporting extracellular acidosis (Kato et al., 2013; Webb et al., 2011). Extracellular acidosis is promoted in cancer cells by the increased expression of proton transporters in the plasma membrane. Indeed, members of the monocarboxylate transporters (MCTs), especially MCT1 and MCT4, or Na+/H+ exchangers (NHEs), NHE1 member are commonly found altered in cancer cells. Consequences of intracellular alkalinization, however, are incompletely understood, but they account for signaling pathways regulation and metabolic adaptation which provide a better cancer cell survival and proliferation abilities (Persi et al., 2018).

INTRODUCTION 25 Figure 3. Reverse pH gradient is a feature of cancer cells. Members of the monocarboxylate transpor- ter family (MCT1 and MCT4) and Na+/H+ exchangers (NHE1) are over-expressed in cancer cells to sup- port proton extrusion. Carbonic anhydrases family is also highly represented to maintain pH equilibrium.

26 INTRODUCTION Lysosomal trafficking

Cytoplasm is a dynamic environment where organelles constantly move, change in size and shape and establish transient contacts with each other. Cellular cytoskeleton supports cell structure and trafficking being composed by three different tracks, namely actin filaments (formed by actin), intermediate filaments (among them keratins, vimentin and lamins) and microtubules (formed by tubulin) which support intracellular transport, cell morphology, specialized structures such as lamellipodia and cell division. Lysosomes are transported along microtubule tracks.

Microtubules are polarized filaments composed of alpha and beta heterodimers. Their minus-end corresponds to the microtubule organizing center (MTOC) where the microtubule network nucleates, while their plus-end points to the cell periphery and polymerize faster.

The fundaments of intracellular transport are highly conserved and motor proteins constitute a common molecular machinery among different cell types. Each molecular motor has acquired the specificity to bind to its partner on the cargo complex, enabling the specific execution of the transport process. Lysosomes are not equally dispersed through the whole cell, indeed two interchangeable lysosome pools exist: peripheral and perinuclear lysosomes (Pu et al., 2016).

Noteworthy, lysosome positioning is altered during malignant transformation where peripheral lysosomes are more abundant, thus contributing to cell motility and signaling (Kallunki et al., 2013). Both retrograde and anterograde transport are mediated by coupling to dynein and kinesin motor proteins respectively (Caviston and Holzbaur, 2006; Zhu et al., 2005). Dynein family

Dynein family is composed by two major classes, axonemal and cytoplasmic dyneins. Axonemal dyneins coordinate ciliary and flagellar beating while cytoplasmic dyneins are involved in intracellular transport, mitosis, cell polarization and cell motility (Höök and Vallee, 2006). Two forms of cytoplasmic dyneins have been described: cytoplasmic dynein 1, found in all microtubule-containing cells and cytoplasmic dynein 2, specific for flagella and cilia base (Höök and Vallee, 2006). Cytoplasmic dynein 1 is responsible

INTRODUCTION 27 for retrograde transport of the Golgi apparatus, lysosomes and endosomes and is also associated with kinetochores in mitosis allowing the transition from metaphase to anaphase (Corthésy-Theulaz et al., 1992; Pfarr et al., 1990).

Dynein is a multimeric complex formed by two heavy chains and several light chains which interacts with dynactin, another multimeric complex, comprising more than 20 subunits formed by 11 different proteins (Ishikawa, 2012). Cytoplasmic dynein 1 interacts with lysosomes in a dynactin-dependent manner (Burkhardt et al., 1997; Deacon et al., 2003).

Adaptor proteins of lysosome minus-end transport

The recruitment of dynein-dynactin to late endosomes and lysosomes is dependent on Ras-related protein (Rab7), whose GTP-bound status is regulated by GTPase- activating proteins (GAPs) TBC1D15 and TC1D2 and the Guanine nucleotide exchange factor (GEF) complex Mon1-Ccz1 (Nordmann et al., 2010). Rab7 couples lysosomes to dynein-dynactin through different effectors such as Rab7-interacting lysosomal protein (RILP) which acts towards dynactin subunit 1 (DCTN1) (Cantalupo et al., 2001; Johansson et al., 2007; Jordens et al., 2001).

Kinesin family

Kinesins have a significant role in intracellular transport and have critical functions for cell morphogenesis, functioning and survival. The human genome encodes for 45 different kinesin-like proteins (KIFs) that can be divided into 15 kinesin families (Figure 4). KIFs are responsible for many transport processes of mammalian cells (Miki et al., 2001).

KIFs possess a conserved motor domain which hydrolyzes ATP and allows the movement, while the other regions are quite variable (Figure 4). Generally, motor proteins dimerize with each other, but some of them act as monomers. The variable tail region binds cargoes through the interaction with light chains or associated proteins (Hirokawa and Noda, 2008). In addition to transporting cargos, kinesins bind chromosomes and spindles and are functional during mitosis (Goshima and Vale, 2003; Wordeman, 2010). Some motor proteins participate in both intracellular transport and mitosis.

28 INTRODUCTION Figure 4. Kinesin family members. Kinesin proteins are divided into 15 families with some specialized functions. Kinesin structure is defined by the presence of a globular motor domain and variable domains such as coiled-coil domains, neck linkers or PH domains for the interaction with lipids. The represented kinesins correspond to processive kinesins related to vesicle distribution (kinesin-1, kinesin-2, kinesin-3) and to chromosome segregation (kinesin-5, kinesin-7) and the depolymerizing M-type kinesin-13 family. Phylogenetic tree was adapted from (Hirokawa et al., 2009). Structural information of the kinesin mem- bers was obtained from UniProt database, subdomains length corresponds to the annotated amino aci- dic sequence. CENPE: Centromere-associated protein E; MCAK: Mitotic centromere-associated kinesin.

Adaptor proteins of lysosome plus-end transport

Centrifugal movement is mediated by the small GTPase ADP-ribosylation factor-like protein 8 (Arl8), which links lysosome to kinesin-1 (KIF5) and kinesin-3 (KIF1A and 1β). Arl8 is a member of the Arf family of small GTPases that can bind the lysosome through the interaction with the BLOC-one-related complex (BORC). BORC is a multi-subunit complex formed by eight proteins: BLOS1, BLOS2, Snapin, Myrlysin, Lyspersin, KXD1, MEF2BNB and Diaskedin (Figure 5). This complex shares some subunits with BLOC- 1, a complex implicated in lysosome-related organelles biogenesis (BLOS1, BLOS2 and Snapin). Arl8 binds to different adaptor proteins depending on the interacting kinesin. For example, Arl8 interacts with Pleckstrin homology domain-containing

INTRODUCTION 29 family M member 2 (SKIP) and kinesin light chain (KLC) in the case of KIF5B (Rosa- Ferreira and Munro, 2011) or with an intrinsic CC3 domain in the case of KIF1Bβ- and KIF1A-dependent lysosomal transport (Wu et al., 2013). Interestingly, the biochemical properties of microtubule tracks also determine the binding of lysosomes to KIF5B- or KIF1A-motor complex, allowing their movement along more acetylated perinuclear microtubules or long and rigid tyrosinated microtubules, respectively (Guardia et al., 2016).

Figure 5. Lysosome motor core complex. Kinesin-1 KIF5B member is anchored to the lysosome through the BORC multisubunit complex which contains Myrlysin protein bound to the lysosome surface though a myristoil group and recruits Arl8b GTPase.

30 INTRODUCTION Lysosome functions

Lysosomes are mainly recognized as degradative organelles, but in the recent years their implication in other processes of high relevance in human pathologies and aging has been uncovered. In this section, three functions of lysosomes will be introduced: (1) degradative function and converging pathways, (2) lysosome as a signaling platform and (3) lysosome-dependent exocytosis. It remains unclear whether all lysosomes can exert all functions or if specialized pools coexist. Lysosomes are heterogeneous in size, localization and morphology. In fact, some specialized lysosomes have been already described (i.e. CMA-lysosomes…), but no clear definition of lysosome subpopulations has been yet established and molecular markers to label lysosomal pools remain to be identified.

Degradative function

There are multiple pathways delivering substrates to the lysosome. In general, extracellular material destined for degradation is delivered to the lysosome via endocytosis, whereas intracellular waste is disposed of by the lysosome via self- catabolic process known as autophagy.

Endocytosis

Cellular internalization of fluids, solutes, proteins and membrane components occurs via endocytosis. Several endocytic routes have emerged depending on the nature of the cargo. Important examples account for phagocytosis, micropinocytosis, clathrin- mediated endocytosis, caveolin-mediated endocytosis and clathrin- and caveolin- independent endocytosis.

After internalization process, cargos embedded in endocytic vesicles end up in EEs, which can in turn be recycled to the plasma membrane or fuse with LEs and lysosomes for degradation. Membrane proteins destined for degradation in lysosomes are mono- ubiquitinated and consequently recognized by receptors which will sort proteins into microdomains of EEs and form the multivesicular bodies (MVB) delivering cargo to the degradative pathway (Jovic et al., 2010; Mellman, 1996). MVBs mature and move to the perinuclear region of the cell where they fuse with each other and with the LEs.

INTRODUCTION 31 Autophagy

Autophagy is a highly conserved catabolic process which consists in the degradation of cellular components. Different types of autophagy coexist in the mammalian system, to note, macroautophagy, microautophagy and CMA (Figure 6) (Dikic and Elazar, 2018; Mizushima, 2007). Macroautophagy and microautophagy involve the sequential function of Autophagy-related proteins (Atg), while CMA utilizes a different machinery..

Figure 6. Different types of autophagy in mammalian cells. Mammalian cells utilize three autophagic routes: microautophagy, CMA and macroautophagy to allow cellular material clearance. ER: Endoplas- mic reticulum. Hsp70: Heat-shock protein 70. LC3: Microtubule-associated protein 1A/1B-light chain 3.

Microautophagy

Microautophagy is the process by which the lysosomal membrane directly engulfs small portions of cytoplasm. This process starts with the invagination of the lysosomal membrane followed by its extension and the generation of the “autophagic tube”. This event is an active ATP-dependent reaction. Microautophagy molecular basis are still controversial, although it has been shown that Microtubule-associated protein 1A/1B-light chain 3 (Atg8/LC3) or Atg10/Atg12 and Atg5 autophagic proteins are

32 INTRODUCTION involved (Li et al., 2012; Mijaljica et al., 2011). Due to a lateral sorting mechanism, proteins are excluded from the newly generated invagination, which has a high lipid content. After scission from autophagic tubes, the released protein-free vesicles are degraded in the lysosomal lumen (Li et al., 2012). Microautophagy is the less understood type of autophagy and it is still controversial whether it has a specific role on organelle size maintenance and nutrient starvation resistance or if it is simply a compensatory and spontaneous method for macroautophagy to consume the excess of membrane.

Chaperone-mediated autophagy

CMA is a type of selective autophagy, identified so far only in mammalian cells, which allows the degradation of specific target proteins (Fred Dice, 1990). Proteins degraded by CMA are directly recognized by the cytosolic chaperone heat-shock protein 70 (Hsp70) that delivers them to the lysosomal surface by interacting with LAMP2A. Once docked, substrate proteins unfold and are transported across the lysosomal membrane for their degradation.

All the proteins targeted for CMA contain the KFERQ pentapeptide in their amino acid sequence which is sufficient for their targeted degradation. Noteworthy, about 30% of all cytosolic proteins contain the KFERQ sorting motif (Kaushik and Cuervo, 2012; Tang et al., 2017). Macroautophagy

Macroautophagy is by far the most well studied autophagic pathway, mainly characterized by the formation of the autophagosome, a double-membrane vesicle containing the cytoplasmic material that will be degraded after fusion with the lysosome. Macroautophagy induction results in ATGs recruitment to a specific subcellular location and nucleation of membrane to form the phagophore. Gradual elongation results in expansion of the phagophore to form the autophagosome.

Autophagy starts by the induction and nucleation of the phagophore. Phagophores are mainly nucleated from an endoplasmic reticulum (ER) extension called “omegasome” (Axe et al., 2008), although it has also been shown phagophore nucleation from mitochondria-ER and plasma membrane-ER contact sites (Hamasaki et al., 2013; Nascimbeni et al., 2017). Autophagy initiation and autophagosome formation is a complex event that is supported by three protein complexes, successively involved in

INTRODUCTION 33 key steps. Autophagy initiation is mediated by serine/threonine protein kinase Unc- 51 like autophagy activating kinase 1 (ULK1), which functions in complex with FIP200, Atg13 and Atg101. How ULK1 is recruited to the nucleation site remains poorly understood. Once activated and translocated to phagophore site, ULK1-containing complex induces the recruitment of a second protein complex, the PI3KC3 complex, containing PI3KC3 (Vps34) as well as Beclin-1, Vps15 and Atg14L. PI3KC3 protein complex is responsible for the production of phosphatidylinositol 3-phosphate (PI3P) at the autophagosome formation site. PI3P-binding proteins such as WD repeat domain phosphoinositide-interacting protein (WIPI2B) and Zing finger FYVE domain- containing protein 1 (DFCP1) are then recruited to phagophore. ULK1 and PI3KC3 proteins also help the recruitment of Atg9-containing vesicles to the nucleation site (Karanasios et al., 2013; Nishimura et al., 2017) which may deliver lipids and proteins promoting membrane expansion. How ULK1 is recruited to the nucleation site remains poorly understood.

After nucleation, phagophore expansion is mediated by two ubiquitin-like conjugation systems: the Atg16L1-Atg5-Atg12 complex and the Atg8 conjugation machinery. This complex dictates the anchoring of the Ub-like Atg8 family members to the growing autophagosome membrane. Newly synthesized Atg8s are processed by the cysteine protease Atg4, exposing a Glycine that is essential for their conjugation to phosphatidylethanolamine (PE) (Slobodkin and Elazar, 2013). Processed Atg8s are activated by the E1-like enzyme Atg7 and conjugated to the membrane-associated PE by Atg3, converting the free diffusing form into membrane-anchored lipidated form (in the case of LC3 those forms correspond to LC3-I and LC3-II respectively) (Slobodkin and Elazar, 2013). For the correct PE conjugation Atg3 requires the E3-like activity of Atg12-Atg5 conjugate (Ichimura et al., 2000). Conjugation of PE to Atg8 is key for phagophore expansion and probably sealing. The next step for macroautophagy is the maturation of the autophagosome which implies the gradual clearance of Atg proteins from the outer membrane and the recruitment of the motor and fusion machinery for the delivery to lysosomes. Atg8s drive the maturation by linking autophagosome to kinesins through the autophagy-specific kinesin adaptors such as FYVE and coiled-coil domain-containing protein 1 (FYCO1) (Olsvik et al., 2015). Fusion of autophagosome with lysosome involves Soluble N-ethylmaleimide-sensitive factor activating protein receptor (SNAREs), syntaxin 17 (STX17) and synaptosomal- associated protein 29 (SNAP29) on the autophagosome and vesicle-associated membrane protein 8 (VAMP8) on the lysosome (Koyama-Honda et al., 2013; Stolz

34 INTRODUCTION et al., 2014) and the homotypic fusion and protein sorting (HOPS) complex, which mediates membrane tethering to allow membrane fusion (McEwan et al., 2015; Stolz et al., 2014).

Macroautophagy was initially understood as a largely unspecific degradative process that occurred upon nutrient deprivation. Bulk autophagy, though, serves to recycle cellular material and compensate for the lack of nutrients. However, it has become clear that autophagy contributes to cellular homeostasis in non-starved cells by selective degradation of damaged organelles or protein aggregates. Thus, selective autophagy is crucial for cellular homeostasis. New selective forms of autophagy are progressively emerging describing more and more players. Unlike bulk autophagy, selective autophagy needs to reach three criteria for efficient degradation: the cargo has to be selectively recognized, then efficiently tether to the nascent autophagosome and finally not to be excluded from the autophagosome. For this, selective autophagy depends on cargo receptor proteins that act as a bridge between the cargo and the autophagosomal membrane (Zaffagnini and Martens, 2016). One of these adaptor proteins is Sequestrasome-1 (SQSTM1/p62) which binds to ubiquitinated cargos via its C-terminal ubiquitin-associated domain (UBA) domain and to Atg8 proteins (LC3B among others), through its LC3-interacting region (LIR) (Pankiv et al., 2007; Seibenhener et al., 2004). p62 preferentially binds to certain polyubiquitin chain types: while K48-linked chains are mainly recognized by the proteasomal system, the K63-linked chains as well as mono-ubiquitin are associated with autophagy (Tan et al., 2008). Exocytosis

Conventional secretion is mediated by vesicles coming from the ER-Golgi-secretory pathway. However, several evidences point out unconventional secretory forms that include the fusion of MVBs (exosome release) or lysosomes to the plasma membrane (Nickel and Rabouille, 2009), a process known as lysosome-dependent exocytosis. Upon stimulation, lysosomes are translocated along the microtubules from the perinuclear region to the plasma membrane where they can dock and fuse to release lysosomal content to the extracellular space (Jaiswal et al., 2002). Nowadays, it is well accepted that all cells can, at some point and under proper stimulation, elicit lysosome-dependent exocytosis (Rodríguez et al., 1997). Lysosome-dependent exocytosis relies on calcium release at specific sites. In fact, endosomes and lysosomes

INTRODUCTION 35 represent important sources of calcium (~0.5mM in lysosomes and between 0.003- 2mM in the endosome) (Shen et al., 2012). Importantly, relevant cellular functions such as plasma membrane repair (Reddy et al., 2001) or cell spreading and motility (Dou et al., 2012) require lysosome-dependent exocytosis.

The involvement of lysosome-dependent exocytosis on cell migration and invasion arises from the possibility to deliver adhesion molecules (Dozynkiewicz et al., 2012), signaling scaffolds (Schiefermeier et al., 2014) and proton pumps to the plasma membrane together with acid hydrolases and protons to the extracellular space. Morphological studies of cell protrusions show that clusters of lysosomes are present in focal adhesions and cell motility structures such as invadopodia where they are more prone to fuse (Hastie and Sherwood, 2016; Naegeli et al., 2017). In this context, v-ATPase proton pumps have been found at the plasma membrane of cancer cells where they are able to extrude protons from the cytosol which facilitates extracellular acidification and correlates with invasive and metastatic phenotypes (Cotter et al., 2015b; Toyomura et al., 2003). Signaling platform Lysosome degradative products can be used by cells as building blocks for biomass production what turns the lysosome into a perfect candidate to sense the metabolic status of the cell. In fact, master regulators of cell metabolism such as mammalian Target of Rapamycin Complex 1 (mTORC1) and AMP-activated kinase (AMPK) are located at the lysosomal surface where they can be activated by up-stream signals. This regulation works also on the other way, as mTORC1 modulates lysosome biogenesis through the phosphorylation of the transcription factor EB (TFEB). TFEB phosphorylation determines its subcellular localization thus modulating the transcription of genes related to lysosome biogenesis and autophagy. mTORC1- dependent TFEB phosphorylation represses its nuclear shuttling consequently inhibiting lysosome biogenesis.

Thus, the lysosomal surface is starting to be apprehended as a metabolic platform where key metabolic regulators are controlled according to nutritional status. mTORC1 signaling pathway mTOR kinase coordinates a major eukaryotic signaling network able to connect cell growth with environmental conditions. mTOR is a serine/threonine kinase part of two

36 INTRODUCTION distinct complexes, mTORC1 and mTORC2. Here, we focus on the study of mTORC1 regulation at the lysosomal surface. mTORC1 is composed by three core components: mTOR, Raptor (regulatory protein associated with mTOR) and mLST8 (mammalian lethal with Sec13 protein 8). Raptor facilitates substrate recognition by mTORC1 through binding to the TOR signaling motif (TOS) found on several mTORC1 substrates and is also needed for the lysosomal localization of mTORC1 (Sancak et al., 2008a). mLST8 binds to the mTOR catalytic domain but its requirement for mTORC1 activity is not direct. mTORC1 contains also two inhibitory subunits PRAS40 (proline-rich Akt substrate 40kDa) and DEPTOR (DEP domain containing mTOR interacting protein). mTORC1 is a master regulator of biomass production for cell growth thus increasing the synthesis of proteins, lipids and nucleotides and suppressing catabolic processes such as autophagy. mTORC1 up-regulates protein synthesis mainly by the phosphorylation of two key substrates, p70S6 Kinase 1 (S6K1) and eIF4E Binding Protein (4EBP) which trigger translation. Apart from protein synthesis, mTORC1 promotes de novo lipid synthesis through the regulation of sterol responsive element binding protein (SREBP) and nucleotide synthesis through carbamoyl-phosphate synthetase (CAD) required for DNA replication and ribosome biogenesis in proliferating cells (Ben-Sahra et al., 2013) (Figure 7A). All these changes affecting biomass production are accompanied by a metabolic shift from oxidative metabolism to glycolysis through the activation of Hypoxia Inducible Factor 1α (HIF1α) (Yecies and Manning, 2011). Glycolysis induction is a common feature of cancer cells and it is believed not to be only an adaptation for rapid ATP production but also to generate intermediate metabolites for anabolic reactions (Gatenby and Gillies, 2004; Gillies and Gatenby, 2007).

In addition to the mentioned anabolic processes, mTORC1 also suppresses protein catabolism, mainly through autophagy inhibition. mTORC1 phosphorylates and impedes activation of ULK1 thus inhibiting autophagy induction (Kim et al., 2011). Furthermore, mTORC1 regulates autophagy by phosphorylating and inhibiting the nuclear translocation of TFEB, which drives the transcription of lysosome biogenesis- and autophagy-related genes (Roczniak-Ferguson et al., 2012; Settembre et al., 2012) (Figure 7A).

Activity of mTORC1 is modulated by several upstream signals, such as growth factors, energy, oxygen, DNA damage or nutrients availability. All these signals converge at the lysosomal surface where mTORC1 gets activated.

INTRODUCTION 37 Growth factors and mitogen-dependent signaling converge on the inhibition of the key mTORC1 negative regulator Tuberous Sclerosis Complex (TSC). TSC is a heterotrimeric complex formed by TSC1, TSC2 and TBX1D7 which acts as a GAP for the small GTPase Rheb (RAS-homolog enriched in brain), which directly binds and activates mTORC1 (Figure 7B). TSC phosphorylation results in its dissociation from the lysosomal surface where Rheb is localized allowing its GTP binding and activation of mTORC1 (Inoki et al., 2003; Menon et al., 2014).

Figure 7. mTORC1 regulatory landscape. A) Some of the mTORC1 downstream targets are represented as well as the impact on subsequent cellular functions. mTORC1 activates protein synthesis through the phosphorylation of 4EBP and S6K thus increasing translation initiation. The activation of S6K also activates nucleotide synthesis by S6K-dependent phosphorylation of CAD, an enzyme that promotes pyrimidine synthesis. mTORC1-dependent activation of lipid synthesis through SREBP is not-completely understood. Glycolysis indirectly activated by mTORC1 at least through HIF1α induction. mTORC1 inhi- bits autophagy and lysosome biogenesis through the inhibition of TFEB transcription factor and ULK1. B) mTORC1 upstream regulation players at the lysosomal surface. Growth factors signal inhibit TSC1/2 complex by phosphorylation, which in turn inhibits Rheb GTPase. Thus, growth factors end up by acti- vating mTORC1 kinase activity. Amino acids act through the Rag A/B – Rag C/D GTPases and induce the recruitment of mTORC1 in the close proximity of Rheb at the lysosomal surface. Intraluminal arginine is sensed by the amino acid transporter SLC38A9 through the v-ATPase which integrates the signal to the Rags activation branch. Dotted lines in the scheme indicate indirect effects.

38 INTRODUCTION mTORC1 translocation to the lysosome is triggered by the presence of amino acids through the Rag small GTPases (Ras-related GTP-binding protein) (Sancak et al., 2008b). Rag A/B – Rag C/D heterodimers are tethered to the lysosomal surface via Ragulator heteropentameric complex composed by LAMTOR1/p18, LAMTOR2/p14, LAMTOR3/MP1, LAMTOR4/p10 and LAMTOR5/HBXIP. LAMTOR1 anchors Ragulator to the lysosomal membrane. Interaction of Rag GTPases with Ragulator is necessary for mTORC1 activation.

Amino acids promote the Ragulator GEF activity towards Rag A/B, whose GTP-bound form allows the recruitment of mTOR (Bar-Peled et al., 2012). This complex machinery interacts with the v-ATPase which is required for the sensing of amino acids from the lysosomal lumen (Figure 7B). Other proteins are specialized on the sensing of different amino acids to this machinery, such as CASTOR1 (Cytosolic arginine sensor for mTORC1) complex for cytosolic arginine, Sestrin 1/2 for cytosolic leucine, which signal through GATOR2 and GATOR1 (GTPase activating proteins toward Rags 2 and 1) directly to the Rags or SLC38A9 which senses intraluminal arginine to the v-ATPase.

Recently, the crystal structure of Ragulator-Rag GTPases has been uncovered. Structural data suggest that the module built from three heterodimers may serve as a platform for additional factors involved in the nutrient-dependent regulation of mTORC1 on lysosomes (Yonehara et al., 2017).

Cellular energy is sensed by the AMPK which inhibits mTORC1 indirectly through phosphorylation of TSC and directly by phosphorylating Raptor. Glucose deprivation triggers AMPK activation through the decrease in AMP/ATP levels, thus inhibiting mTORC1. However, glucose deprivation also inhibits mTORC1 in cells lacking TSC or AMPK, through the inhibition of the Rag GTPases, suggesting that mTORC1 senses glucose through diverse mechanisms (Kalender et al., 2010). In this direction, it has been shown that AXIN, a scaffold protein involved in AMPK regulation is able to modulate Ragulator GEF activity, thus affecting mTORC1 recruitment to the lysosome upon glucose starvation (Zhang et al., 2014). Moreover, other regulatory mechanism has been shown through the glycolytic enzyme glyceraldehyde-3- phosphate dehydrogenase (GAPDH), which would be bound to Rheb when glucose is absence thus keeping mTORC1 activity repressed. After glucose addition GAPDH would be redirected towards glycolysis thus allowing Rheb interaction with mTORC1 and activating its kinase activity (Lee et al., 2009).

INTRODUCTION 39 AMPK regulation at the lysosome

AMPK is the central regulator of cellular response to energy levels. AMPK is a heterotrimeric protein comprised of a catalytic α subunit, a β subunit and a ɣ subunit. This complex is sensitive to changes in the cytosolic AMP/ATP and ADP/ATP ratios. Binding of AMP to ɣ subunit of AMPK modulates its phosphorylation by Liver kinase B1 (LKB1) (Cheung et al., 2000), while ADP binding prevents its dephosphorylation and inactivation (Xiao et al., 2011). Once activated, AMPK enhances catabolic processes for energy production and increases glucose uptake while inhibiting biosynthetic processes. One important pathway regulated by AMPK is macroautophagy by phosphorylating and activating ULK1/2 kinase and inhibiting mTORC1 pathway (Egan et al., 2011; Kim et al., 2011).

Recently, AMPK has been localized at the lysosomal surface through its interaction with AXIN, which is a scaffold protein for AMPK and its upstream activator LKB1 (Zhang et al., 2014). AXIN recruitment to the lysosome is mediated by a Ragulator subunit, LAMTOR1. Interestingly, glucose starvation has been found to enhance AXIN/LKB1 – AMPK association to LAMTOR1, while loss of LAMTOR1 impaired AMPK activation. Moreover, phosphorylation of AMPK has been shown to occur exclusively atthe lysosomal surface under glucose starvation (Zhang et al., 2014, 2017). Furthermore, the underlying mechanism for AMPK activation at the lysosomal surface to sense glucose levels is dependent on the glycolytic enzyme aldolase, by its direct interaction with this regulatory machinery. In addition, aldolase was proposed to sense the levels of Fructose 1,6 bisphosphate (Fru-1,6-P2), which linearly correlate with the amount of glucose (Zhang et al., 2017).

This common regulatory machinery in both AMPK and mTORC1 regulation at the lysosomal surface suggest an intrinsic cross-talk between anabolism and catabolism directly at the lysosomal surface (Carroll and Dunlop, 2017).

40 INTRODUCTION Lysosomes in the context of cancer

Being described as the engulfing organelle of damaged cellular biomass, lysosome has emerged recently as a versatile organelle implicated in many functions important for cellular homeostasis maintenance. Cell malignant transformation promotes aberrant growth and induces a fast consume of nutrients, accumulation of protein aggregates and damaged organelles making some cancer cells dependent on lysosome- dependent recycling pathways to sustain uncontrolled cell proliferation and growth. Additionally, lysosomes are relevant as signal scaffolds for the activation of metabolic master regulators, such as AMPK and mTORC1 and their positioning has been related to cell migration and invasion (Kallunki et al., 2013). During cancer invasion, cells need to move and degrade the extracellular matrix components, which is achieved by the activation of secreted proteolytic enzymes such as metalloproteinases and plasminogen or cathepsins coming from the lysosomes (Sevenich and Joyce, 2014). Moreover, the continuous formation and dissolution of focal adhesions is required for efficient cell motility. Indeed, a recent study points out that a subpopulation of LAMTOR2/3-expressing lysosomes are targeted to FA where they help to their turn- over (Schiefermeier et al., 2014).

All these effects suggest that targeting lysosomes for cancer therapy might have pleiotropic effects involving cell metabolism (Settembre et al., 2013), DNA damage (Kroemer et al., 2010), cell death (Boya and Kroemer, 2008; Kroemer and Jäättelä, 2005) and protein secretion (Kraya et al., 2015). Lysosome-targeting drugs can be divided at least in five categories including lysosomal hydrolase inhibitors, Hsp70 inhibitors, cationic amphiphilic compounds, v-ATPase inhibitors such as Concanamycin A and Chloroquine derivatives. Interestingly, several clinical trials using Hydroxychloroquine (HCQ) alone or in combination are currently being conducted worldwide in humans (Chude and Amaravadi, 2017; Onorati et al., 2018).

INTRODUCTION 41 42 INTRODUCTION 2. E2F1 TRANSCRIPTION FACTOR

Lysosomes are implicated in several hallmarks of cancer being involved in cell metabolism regulation, cell growth control and cell invasion capabilities through exocytosis. Interestingly, lysosome positioning can define the malignant behavior of a cancer cell and is related to mTORC1 activity. Indeed, mTORC1 activity is higher when lysosomes are peripheral, next to the cell membrane where they integrate mitogenic signals and are prone to fuse with the plasma membrane. How oncogenic signals impact on lysosome positioning is an emerging field still understudied. Previous studies of our research group described the regulation of mTORC1 and lysosomal trafficking by the oncogene E2F1 which is dependent on the v-ATPase activation (Meo-Evoli et al., 2015; Real et al., 2011). One of the focus of this project is to uncover the link between E2F1 oncogenic functions and mTORC1 activation at the lysosomal surface. In this section, I will discuss the main characteristics of E2F1 protein, its connection to cancer progression and its cellular functions as transcription factor.

INTRODUCTION 43 E2F transcription factor family

E2F proteins comprise a large family of transcription factors which act downstream of the retinoblastoma protein. This family is essential for cell cycle regulation specifically at the G1/S transition. However, members of this family have been also related to other cellular processes such as apoptosis regulation, differentiation, development or cell growth.

In mammals, E2F transcription factor family is composed by 8 members (E2F1-8) that belong either to activator (E2F1-3) or repressor (E2F4-8) subclasses Figure( 8). E2F1 was the first discovered member of the E2F family and its role has been extensively studied in the context of cancer. The activity of E2F1 is dependent on its binding partners, which include dimerization partners (DP) and the retinoblastoma family proteins or pocket proteins composed by pRB, p107 and p130. E2F1 - pRB interaction blocks the transcriptional activation domain of the E2F1-DP complex and prevents the recruitment of transcriptional co-activators.

Figure 8. E2F family members’ structure. E2F transcription factors can be divided into activators and repressors. They contain different functional domains indicated in the legend. Grey boxes represent structural domains with not known specialized function (Chen et al., 2009; Tsantoulis and Gorgoulis, 2005).

44 INTRODUCTION E2F1 in cancer

E2F1 function in tumors has been controversial as it can be considered either an oncogene or a tumor suppressor due to its ability to induce both cell cycle progression and p53-dependent apoptosis (Pierce et al., 1999).

E2F1 overexpression increases colony formation in soft agar, induces tumors in nude mice and, in cooperation with Ras, transforms rat embryo fibroblasts (Johnson et al., 1994). E2F1 also increases the susceptibility to develop skin tumors in p53-null mice (Pierce et al., 1998). In liver, overexpression of E2F1 results in hepatocellular adenomas (Conner et al., 2000). In contrast, E2F1 knockout mice develop spontaneous tumors suggesting a protective role of E2F1 against oncogenesis (Yamasaki et al., 1996). This bimodal function of E2F1 is highly dependent on the genetic context, the status of pRB and the expression of the other E2F family members.

Of note, studies in human patient samples showed that in non-small cell lung carcinomas, E2F1 protein is highly expressed compared to normal adjacent tissue (Gorgoulis et al., 2002; Karakaidos et al., 2004). Similar scenario applies for other lung tumors such as small cell lung carcinoma or large cell neuroendocrine carcinoma (Eymin et al., 2001), pancreatic ductal carcinoma (Yamazaki et al., 2003) or breast carcinoma (Zhang et al., 2000), indicating that in patients, E2F1 levels are increased (Figure 9).

Figure 9. E2F1 and RB1 alterations in cancer. Different alterations of DNA-copy number data of The Cancer Genome Atlas (TCGA) of all types of cancer. The data and representation were obtained from cBioportal database (Cerami et al., 2012; Gao et al., 2013).

INTRODUCTION 45 E2F1 functions

The main function of E2F1, shared with the other members of this family, isto transcriptionally regulate the G1/S transition thus regulating cell cycle progression. However, decades after the discovery of E2F1, it is now evident that E2F1 functions go beyond its classical role as a cell cycle regulator, controlling apoptosis, senescence, DNA-damage response, cell growth, autophagy and metabolism. Indeed, genome- wide analyses have confirmed that more that 20% of the human genome promoters contain E2F1-binding sites regulating genes of hundreds of cellular pathways (Bieda et al., 2006).

E2F1 regulation of cell cycle

Cell cycle is composed by four different phases: G1, S, G2 and M. E2F1 participates on eliciting the G1/S transition of the cell cycleFigure ( 10), which commit cells to progress independently from environmental signals (Bertoli et al., 2013). Cell cycle regulation is based on the waved activation of several Cdk-cyclin complexes which allow the temporary control of cell cycle phases.

Figure 10. E2Fs activity oscillations during cell cycle. Quiescent cells express E2F4 and 5 inhibitory E2Fs. Once cells enter G1 the activator forms E2F1-3 are expressed to allow the transcription of genes involved in DNA replication and G1/S transition. At the end of S phase, E2F6 – 8 take control and repress E2F promoters allowing the progression through G2 and M phases of the cell cycle (adapted from (Chen et al., 2009)).

46 INTRODUCTION Quiescent cells (G0) express E2F4 and E2F5 that bound to pocket proteins and other co-repressors, act as repressors of transcription. Upon mitogenic stimulation, sequential phosphorylation of pRB occurs what results in the accumulation of newly synthetized E2F1, E2F2 and E2F3 Figure( 10). The released E2F activators commit cells to undergo DNA replication driving the transcriptional program of the S phase. At the end of S phase and G2, E2F activators activity is attenuated by the action of the repressors E2F6-8, which act independently of pocket proteins. At late G2, pRB is dephosphorylated and sequesters again the activator E2Fs (Chen et al., 2009).

E2F1 regulation of cell metabolism

Metabolic reprogramming has emerged as one of the hallmarks of tumor cells. Thus, metabolic pathways are orchestrated to satisfy the requirements for cell growth, proliferation and differentiation. Growing evidences establish a regulatory crosstalk between metabolic pathways and cell cycle regulators. E2F1, as a main regulator of G1/S transition, is also able to induce a metabolic shift from oxidative to glycolytic metabolism. It has been shown that, while E2F1 knockout mice undergo normal development, possibly due to the compensation by other E2F family members, they exhibit metabolic alterations in several tissues such as adipose tissue, pancreas or muscle (Blanchet et al., 2011; Denechaud et al., 2017).

Most cancer cells predominantly use glycolysis to generate energy even in presence of abundant oxygen. This metabolic adaptation is known as the Warburg effect and explains the high rate of glucose uptake and lactate production of cancer cells even when mitochondria are not defective. In this regard, it has been described that E2F1 enhances glycolysis and represses mitochondrial oxidation (Denechaud et al., 2017). In hepatocellular carcinoma cells, E2F1 regulates the expression of genes involved in glycolysis and lactate export by recruiting Pontin and Reptin DNA helicases (Tarangelo et al., 2015) (Figure 11). Moreover, in bladder and prostate cancer cell lines, E2F1 also enhances glycolysis through the suppression of Sirtuin 6 expression, which inhibits the transcription of several glycolytic genes (Wu et al., 2015). Accordingly, in proliferating cells, E2F1 promotes the expression of the fetal form of the phosphofructokinase-2/fructose bisphosphatase-2 enzyme (PFK-2/FBPase-2), a potent stimulator of glycolysis (Fernández de Mattos et al., 2002). Besides glycolytic enzymes transcriptional regulation, E2F1 also inhibits mitochondrial oxidation by enhancing the expression of Pyruvate Dehydrogenase Kinases (PDK1 and PDK3) in pancreatic cells, which results in increased aerobic glycolysis (Wang et al., 2016) (Figure 11).

INTRODUCTION 47 Figure 11. E2F1 regulates metabolism. E2F1 regulates metabolism at both systemic and cellular level. The scheme shows the implication of this transcription factor in tissue function as well as in cancer cells promoting the Warburg effect (Denechaud et al., 2017). PDK: Pyruvate dehydrogenase kinase; PDH: Pyruvate dehydrogenase; fPFK-2: fetal PFK-2 / FBPase-2.

E2F1 regulation of cell growth and lysosome trafficking

Recent results from our laboratory demonstrated that E2F1 regulates cell growth by inducing mTORC1 activity (Real et al., 2011). Interestingly, E2F1-induced mTORC1 activation is independent on the PI3K/Akt/TSC1-2 pathway (Real et al., 2011). We showed that E2F1 drives the translocation of mTORC1 to the lysosomal compartment where it can be activated by the Rag GTPases. The molecular mechanism responsible for this regulation is still under investigation. E2F1 activation, in fact, increases v-ATPase activity and induces lysosome peripheral localization independently of mTORC1 activity. In this regard, E2F1 is able to increase v-ATPase activity through both

48 INTRODUCTION transcriptional regulation of the V0b subunit and increased docking of the V1 domain (Meo-Evoli et al., 2015). However, V0b depletion is not able to completely abolish E2F1-induced mTORC1 activation, suggesting that other independent mechanisms could be implicated on E2F1-driven mTORC1 activation (Meo-Evoli et al., 2015). The fact that E2F1 regulates the peripheral positioning of lysosomes throughout the cytoplasm potentially facilitates lysosome-dependent exocytosis, a process described to increase cell motility and invasion Figure( 12).

Figure 12. E2F1-driven mTORC1 regulation through v-ATPase and lysosomal trafficking modulation. E2F1 regulates v-ATPase activity through the transcriptional activation of V0b subunit and increased V1 docking at the lysosome and also induces lysosomal trafficking towards cell periphery. Both events impact on mTORC1 activation, which drives cell growth and inhibits autophagy (Meo-Evoli et al., 2015). These phenotype could induce an increased lysosome-dependent exocytosis, thus increasing cell migra- tion and invasion.

INTRODUCTION 49 50 INTRODUCTION 3. MITOTIC CELL DIVISION

Lysosome-focused research has been mainly performed in whole cell population where the majority of cells are in interphase. During cell division the degradative function has been restricted to the ubiquitin-proteasome system (UPS) under control of the Anaphase-Promoting Complex (APC/C). UPS is protein specific, ubiquitin triggered and faster than lysosome-dependent degradation. In agreement with this, two elegant studies described the inhibition of autophagy during cell division (Eskelinen et al., 2002; Furuya et al., 2010). The authors claim that autophagy signaling is shut down when cells enter mitosis and is restored in telophase and show that autophagic structures can barely be detected in dividing cells. However, lysosome-dependent degradation is wider and gets rid of protein aggregates and other cellular material. Indeed, autophagy protects against genomic instability under starvation conditions (Matsui et al., 2013). Although autophagic structures seem to be reduced during cell division, lysosomal degradation has been implicated in Cyclin A2 degradation during mitosis (Loukil et al., 2014) and other recent studies claim that autophagy and mitophagy persist during mitosis (Li et al., 2016; Liu et al., 2009). To clarify this controversy, one of the focus of this project is to study the possible participation of lysosomal-dependent degradation of mitotic proteins. In this section, I will present an overview of the molecular pathways involved in mitotic progression and introduce the essential players and regulators of cell division in the mammalian system.

INTRODUCTION 51 Mitosis progression and regulation

Cell division is the process by which a cell gives rise to two genetically equal daughter cells. This process is known as mitosis and comprises the last step of the cell cycle. To maintain genome integrity, all cell cycle phases must be tightly coordinated to ensure the faithful transmission of hereditary information between generations. Mitosis has been studied by cytologist for the past 100 years, but it is not until the 80’ that research unveiled the basic molecular factors underlying eukaryotic cell division coordination. The ordering and sequencing of events depends on central regulatory systems that initiate every step at appropriate time.

Mitosis itself consists of five active phases: prophase, prometaphase, metaphase, anaphase and telophase (Figure 13). The major regulation checkpoint during mitosis comprises the metaphase to anaphase transition, controlling the transition between the mitotic entry and the mitotic exitFigure ( 13). Through these phases, cells must undergo nuclear envelope break-down (NEBD), chromosomes condensation which will bind the mitotic spindle and equally separate into daughter cells.

Figure 13. Mitotic subphases.Mitosis comprises five morphological subphases: prophase, prometaphase, metaphase, anaphase and telophase. Finally, cytokinesis gives rise to the two daughter cells.

52 INTRODUCTION The main coordinators of this regulatory network are the cyclin-dependent kinases (Cdk), whose activity oscillates during the different mitotic subphases in association with different cyclins. Cdk activity controls the mitotic entry as far as metaphase when all chromosomes are aligned. From there, another key regulator lights up, the APC/C, an ubiquitin-protein ligase which ubiquitinates several regulatory proteins and targets them for proteasomal degradation.

Mitotic entry: from G2 to metaphase

Cyclins oscillation during early mitosis

In animal cells, two or three cyclins govern mitotic entry and coordination (Sullivan and Morgan, 2007). The first mitotic cyclin expressed is Cyclin A, which appears at the onset of S phase contributing to DNA synthesis. The levels of cyclin A remain high until prophase and contribute to chromosome condensation and possibly NEBD. Degradation of cyclin A occurs at prometaphase.

The next player is cyclin B whose levels rise in G2 and will activate Cdk1 activity. Specialization of different cyclins restricts both substrate recognition by theCdks and its localization, which probably is the underlying mechanism of this sequential regulation (Bloom and Cross, 2007).

Cyclin B culminates chromosome condensation and drives mitotic spindle assembly. Cyclin B is active until metaphase onset once chromosomes are aligned and ready to be separated. Cyclin B is then degraded during metaphase by the UPS (Balachandran et al., 2016). Finally, cyclin B3 has been described in some animal species although its function remains unclear. Studies in fruit fly proposed that cyclin B3 is degraded in anaphase, later than cyclin B (Parry and O’Farrell, 2001).

Nuclear envelope break-down

The eukaryotic nucleus is separated from the rest of the cellular content bythe nuclear envelope (NE). The NE is composed by the outer nuclear membrane (ONM) which is continuous with the endoplasmic reticulum and the inner nuclear membrane (INM) in contact with chromatin through lamin proteins (Magistris and

INTRODUCTION 53 Antonin, 2018). Vertebrates have three lamin genes (lamin A, B1 and B2) (de Leeuw et al., 2018; Stuurman et al., 1998). Of note, nuclear lamina is in contact with the cytoplasmic actin cytoskeleton through LInker of Nucleoskeleton and Cytoskeleton (LINC). The two nuclear membranes are in contact in many sites by the presence of nuclear pore complexes (NPCs) which allow the selective transport of molecules in and out of the nucleus. Approximately 500 individual proteins constitute a single NPC that are collectively known as nucleoporins (NUPs) (Alber et al., 2007). NE undergoes extreme morphological changes during mitosis when it breaks down and reforms in late anaphase and telophase.

During mitosis, phosphorylation cascades trigger the NEBD, when NUPs are phosphorylated by Cdk1 and Polo-like kinase 1 (PLK1) promoting NPC destabilization (Linder et al., 2017). Thus, lamins and other INM proteins get phosphorylated at least by Cdk1 (Heald and McKeon, 1990). Lamins disassembly is essential for NEBD during mitosis. The NEBD is coordinated with the formation of the bipolar mitotic spindle. During NEBD, microtubules mechanically tear the nuclear membrane forcing NE rupture. In late anaphase, the NE starts to reassemble from the bulk ER around the decondensing chromatin (Lu et al., 2011). This process requires events of membrane fusion mediated by ESCRT-III complex (Olmos et al., 2015; Vietri et al., 2015).

Cohesin complex disruption for sister chromatids individualization

After DNA replication in S phase, sister chromatids remain associated through the action of cohesin complexes. Cohesion assists the equal segregation of the duplicated genetic material (Haarhuis et al., 2013a). Cohesin is formed by several proteins that form a ring around both DNA strings and maintain their association along G2 phase.

Once prophase is initiated and chromatids individualize and condense, cohesion disruption is triggered except in the centromeric region of chromosomes that will remain bound until the metaphase-anaphase transition when all the chromosomes are aligned at the metaphase plate (Villa-Hernández and Bermejo, 2018).

Mitotic spindle formation

Chromosome segregation during mitosis rely on mitotic spindle, a complex macromolecular machinery that coordinates the chromosome movement. Mitotic spindle is composed by centrosomes, chromosomes and microtubules. In prophase, NPC-attached dyneins contribute to centrosome separation by pulling on microtubules

54 INTRODUCTION as centrosomes are pushed apart by kinesin 5 (Eg5; also known as KIF11) (Ferenz et al., 2010). During prometaphase, duplicated centrosomes nucleate a radial array of microtubules which capture chromosomes through the kinetochore (Hayden et al., 1990; Rieder and Alexander, 1990). The machinery of microtubule nucleation has been challenged and several evidences have shown that other mechanisms control this process. Chromosomes themselves can in fact induce the nucleation of microtubules thus facilitating the capture by the mitotic spindle (Gadde and Heald, 2004; Khodjakov et al., 2003; Maiato et al., 2004).

Mitotic exit: from metaphase to cytokinesis

APC/C-substrate destruction

APC/C is formed by at least 12 subunits and needs three cofactors to exert its function: the ubiquitin-activation (E1) enzyme, an ubiquitin-conjugating (E2) enzyme and a co- activator protein. The multiubiquitination of target proteins by APC/C is processive, so multiple lysins can be modified in a single substrate-binding event (Meyer and Rape, 2011). APC/C targets contain a short amino acid sequence that is required for mitotic degradation, namely destruction box (D-box) and KEN box (He et al., 2013; Morgan, 2013; Pfleger and Kirschner, 2000). Cdc20 and Cdh1 are the best studied co-activator proteins for APC/C which recognize and bind the D-box and KEN motifs of protein targets.

The ability of APC/C to recognize its substrates as well as the correct time of degradation is key for anaphase initiation, exit from mitosis and the preparation for the next round of DNA replication.

Cdk1 inactivation by cyclin B proteolysis is essential for the initiation ofmitotic exit. After cyclin B proteasome-dependent degradation, the remaining Cdk1 gets inactivated allowing the transition to anaphase and telophase. This degradative function is accomplished by APC/C bound to Cdc20 (APC/CCdc20).

Sister chromatides separation

The best-known function of APC/C is its role on directly promoting anaphase. To progress into anaphase, the bounds between sister chromatids must be dissolved. In metaphase, APC/CCdc20 ubiquitinates securin, a protein that acts as inhibitor of the protease separase. Once securin is degraded, separase is active and cleaves RAD21

INTRODUCTION 55 subunit of cohesin, a protein complex that anchors both sister chromatids together. Cohesin disruption allows the dissociation of the sister chromatidsFigure ( 14). The exact mechanisms that ensure sister kinetochores attachment to microtubules with opposing polarities remain partially obscure. It is known that after successful bi-orientation, sister chromatids are pulled in opposite directions while theyare still bound by the cohesin complex (Nasmyth, 2002). This might increase chromatin tension near the centromeres and activate the action of separase, probably by shutting off Mad2- Aurora B- dependent checkpoint. In parallel, unattached chromosomes trigger the production of a complex containing Mad1-3 and Bub3 which binds Cdc20 thus inhibiting APC/C activity and preventing cyclin B degradation and progression to anaphase. This control mechanism is known as the spindle assembly checkpoint (SAC) (Figure 14). Mitotic motors are mainly members of the kinesin superfamily, which are uniquely regulated timely and spatially during mitosis, thus contributing to bipolar spindle assembly, cell division axis definition and chromosome alignment and segregation. The subdivision of chromokinesins are those kinesins that interact with DNA and spindle microtubules during mitosis. In metazoans, three kinesin families of chromokinesins have been described (Kinesin-4, -10 and -12).

Figure 14. Sister chromatids separation during metaphase-anaphase transition. Chromosome attachment defects are sensed by the Bub-Mad checkpoint, which arrest cell cycle through the APC/C inhibition, thus avoiding the inactivation of the Cdk1. Once APC/C is activated, it allows the activity of separase which in turn cleaves cohesins thus promoting the sister chromatids separation. SMC1/3: Structural maintenance of chromosomes protein 1/3; RAD21: Double-strand-break repair protein 21; PDS5: Cohesin-associated factor B; STAG 1/2: Cohesin subunit SA 1/2; WAPL: Wings apart-like protein.

56 INTRODUCTION Chromosome instability and cancer progression

Chromosome instability (CIN) is defined as the loss or rearrangement of chromosomes during cell division. CIN is currently understood as a source for genetic variation that can favor tumor cells adaptation to environmental stresses and cytotoxic anticancer drugs (Bakhoum and Cantley, 2018; McClelland, 2017). CIN is a common characteristic of solid tumors and can be classified as numerical CIN (aneuploidies) or structural CIN (gain or loss of fractions of chromosomes).

The mechanisms underlying CIN remain poorly characterized but reflect defective chromosome duplication or segregation during mitosis. Among the described mechanisms, CIN is caused by kinetochore-microtubule attachment errors, abnormal sister chromatids cohesion, aberrant centrosome duplication, telomere dysfunctions or SAC abnormalities (Thompson et al., 2010). However, the consequences of CIN in cancer development are still debated; some researchers favor the idea that CIN is an early event that leads to genomic alterations that will help cancer cell progression, while others claim that CIN is a side effect of uncontrolled cell growth during the neoplastic transformation.

INTRODUCTION 57 The thinker. Immunostaining of U2OS cells actin cytoskelleton RATIONAL OF THE STUDY AND AIMS

1. RATIONAL OF THE STUDY

Emerging evidences link lysosome biology to the hallmarks of cancer (Kallunki et al., 2013). Lysosome function relies on both lysosome acidification modulated by the v-ATPase and on their localization in the cytoplasm, which is coordinated by the motor proteins kinesins and dyneins. Cancer cells display an altered lysosome positioning and degradation capacity which impacts on basic cellular functions such as autophagy, metabolic signaling and cell migration.

Previous results demonstrated that E2F1 transcriptionally regulates v-ATPase V0b subunit and activates mTORC1 at the lysosome surface (Meo-Evoli et al., 2015). However, V0b depletion can not completely abolish E2F1-induced mTORC1 activation, suggesting that other independent mechanisms might be implicated on E2F1 regulation of mTORC1. As E2F1 is a metabolic regulator able to induce a metabolic switch towards glycolysis and to transcriptionally up-regulate some glycolytic enzymes, wehypothesize that glucose may influence E2F1-driven mTORC1 regulation and lysosomal localizationAIM [ 1a]. Lysosome surface offers a platform to integrate environmental inputs, such as nutrient availability, into cell growth and metabolism coordination (Pu et al., 2016). Cell growth is controlled by the master regulator mTORC1. mTORC1 activity is modulated by amino acids which induce its translocation to the lysosomal surface. In parallel, mitogenic signaling impacts on mTORC1 activity through the regulation of TSC1/2 complex (Rabanal-Ruiz and Korolchuk, 2018). Despite the molecular machinery controlling mTORC1 sensing of amino acids is well established, the sensing machinery for other nutrients such as glucose is not fully understood. Glucose starvation reduces energy production inducing energetic stress which is sensed by AMPK. Once activated, AMPK directly phosphorylates TSC2 and Raptor, thus inhibiting mTORC1 activity (Gwinn et al., 2008). However, cells lacking AMPK are still able transduce glucose insufficiency to reduced mTORC1 activity (Kalender et al., 2010), suggesting that alternative mechanisms coordinate glucose availability and mTORC1 activation. Interestingly, AMPK regulation by glucose has been described to occur at the lysosomal surface and to be mediated by the glycolytic enzyme aldolase, which interacts with Ragulator and activates AMPK when glucose is absent (Zhang et al., 2014, 2017). Thus,we hypothesize that a glycolytic metabolon at the lysosomal surface could sense glucose to regulate mTORC1 [AIM 1b]. Lysosomes peripheral localization influence cell ability to migrate as peripheral lysosomes are more prone to fuse with the plasma membrane releasing hydrolases and protons to the extracellular matrix. Accordingly, v-ATPase activation at the cell

RATIONAL OF THE STUDY AND AIMS 61 surface has been related to the metastatic potential of several cancer cell lines, due to the acidification of the extracellular space (Cotter et al., 2015b; Martinez-Zaguilan et al., 1993; Toyomura et al., 2003). Recent results from our research group have shown that E2F1 activation enhances v-ATPase activity and lysosome redistribution towards cell periphery (Meo-Evoli et al., 2015). We hypothesize that increased v-ATPase activity together with lysosome proximity to plasma membrane could potentially induce lysosome-dependent exocytosis, thus contributing to metastatic functions of E2F1 [AIM 1c]. Cancer cells rely on high metabolic demand to maintain their uncontrolled proliferation rate. In this sense, most known oncogenes such as c-Myc, Ras or E2F1 are cell cycle regulators that promote cellular adaptation to convenient metabolic state. Cell division is characterized by a sequence of events by which a cell gives rise to two daughter cells. To maintain genome integrity, all the cell cycle phases must be coordinated ensuring the faithful transmission of hereditary information between generations. Among the four cell cycle phases, G1, S and G2 account for more than 90% of the cell cycle time and largely correlate with cell cycle length (Araujo et al., 2016). In contrast, the duration of mitosis is extremely short, remarkably constant and uncoupled from the variability in other cell cycle phases. Mitosis itself consists of five active phases timely and sequentially controlledFigure ( 13 - Introduction). In this context, degradative pathways such as the proteasome or the lysosome may play important roles on controlling cell division. Curiously, the role of proteasome- dependent degradation has been extensively studied through the APC/C, which ubiquitinates and targets several proteasome substrates (Peters, 2006). In contrast, activation of lysosomal degradation and autophagy participation in mitosis isstill understudied and remains unclear. Thus, some works state that during cell division autophagy is shut down while others favors an active autophagic flux (Li and Zhang, 2016). We hypothesize that correct mitotic progression might also rely on lysosome activity AIMS[ 2a and b]. Proteasome degrades key target proteins during mitosis allowing chromosome segregation, thus correct transition from metaphase to anaphase. Ubiquitination is the signal used to target proteasome protein substrates. However, ubiquitination is not only restricted to signal for proteasomal degradation. Only polyubiquitinated proteins will be recognized by the proteasomal machinery, while ubiquitin can also be recognized by autophagic adaptor proteins such as p62. p62 protein binds to ubiquitinated autophagic cargo and to Atg8 proteins through it LIR motif at the autophagosome leading to lysosomal-dependent degradation of the selected cargo. We hypothesize that some mitotic proteins might be targeted and actively degraded by the lysosome in dividing cells to enable correct mitotic progression, thus preventing chromosome instability [AIM 2c].

62 RATIONAL OF THE STUDY AND AIMS 2. AIMS

The main goal of this PhD project is to elucidate lysosome functions of cancer cells to find novel vulnerabilities related to this organelle. In this context we have stablished the following aims and subaims:

1. Study the implication of glucose on E2F1-induced mTORC1 activation and lysosomal trafficking and E2F1 function on cell migration.

a. Determine the importance of glucose availability on E2F1-induced mTORC1 regulation and lysosome localization

b. Characterize the regulation of mTORC1 pathway by glycolytic enzymes

c. Investigate the function of E2F1 on lysosome-dependent exocytosis and cell migration

2. Uncover a novel role of lysosomes during cell division

a. Analyze the lysosome function in dividing cells

b. Study the implication of the lysosome acidification and trafficking in mitosis

c. Identify novel lysosomal protein substrates implicated in cell division

RATIONAL OF THE STUDY AND AIMS 63 Side by side. Immunostaining of U2OS cells actin cytoskelleton RESULTS

1. E2F1 REGULATES mTORC1 AND LYSOSOMAL TRAFFICKING THROUGH GLYCOLYSIS INDUCTION

Glucose potentiates E2F1-induced mTORC1 activation

To evaluate the effect of glucose on E2F1-induced mTORC1 activation, ER-E2F1 stable U2OS cells were glucose starved and treated or not with 4-hydroxytamoxifen (OHT) and mTORC1 activity was analyzed by quantifying the phosphorylation of its downstream target S6K. Results showed that E2F1 activates mTORC1 both in presence and absence of glucose. However, while E2F1-induced activation in glucose starved cells is limited and not significant, the response in glucose full medium is significantly amplifiedFigure ( 15A and B). As expected, glucose starved cells display an increased AMPK phosphorylation. AMPK is known to inhibit mTORC1 directly (through Raptor phosphorylation) and indirectly (through TSC1/2 phosphorylation) what could explain, at least partially, the low mTORC1 activity observed in glucose starvation conditions. Interestingly, AMPK phosphorylation as well as its downstream target Acetyl-CoA carboxylase (ACC) are not affected by E2F1 induction suggesting that energetic balance is not involved on E2F1-induced mTORC1 activationFigure ( 15A). These results support a possible utilization of glucose for E2F1-driven mTORC1 activation. To clarify whether E2F1 activation impacts on the energetic balance and consequently AMPK activation, we measured the adenosine nucleotide levels upon E2F1 induction. Targeted metabolomics was performed by Ultra-Performance Liquid Chromatography (UPLC) and AMP/ADP/ATP levels were measured. Results showed that upon E2F1 induction, AMP and ADP levels tend to increase Figure( 16A), while AMP/ATP ratio, which triggers AMPK response, is maintained after OHT treatment Figure( 16B). Cells treated with 2-deoxyglucose (2DG), an analogue of glucose that bocks glycolysis and Oligomycin A, a mitochondrial complex V inhibitor, were used as a positive control for energetic stress. As expected, under these conditions cells display an increased levels of AMP and ADP and a dramatic reduction in ATPFigure ( 16A). To finally discard any role of E2F1 on AMPK activation, we performed a time-course experiment of OHT treatment and analyzed the AMPK specific phosphorylation sites on its down-

RESULTS 67 stream targets: ACC phosphorylation on Serine 79, Raptor phosphorylation on Serine 792, ULK phosphorylation on Serine 555. Results showed that none of these phosphorylation sites are altered upon E2F1 induction, while mTORC1 is activated in a time dependent manner upon E2F1 induction as shown by the phosphorylation of S6K (Figure 16C). We also checked the autophosphorylation site of AMPK Threonine 172 which slightly increases upon E2F1 induction; however this change does not correlate with mTORC1 inactivation confirming that AMPK is not responsible for the effect of E2F1 on mTORC1 activation in presence of glucose (Figure 16C).

Figure 15. Glucose potentiates E2F1-induced mTORC1 activation. U2OS ER-E2F1 cells were serum- deprived overnight and glucose-starved for 1 hour prior to treatment with OHT for 6 hours. A) Indicated proteins were analyzed by Western Blot. B) Densitometry of protein gels was performed. Error bars correspond to S.E.M. for n=3 independent experiments. Statistical significance is represented as: * p < 0.05, ** p < 0.005, ns: p > 0.05. a.u.: arbitrary units (Fold change to CTRL).

As E2F1 effect on mTORC1 activation is independent on energetic balance but relies on glucose availability, we studied whether direct inhibition of glycolytic flux would affect E2F1-induced mTORC1 activation. To tackle this question, E2F1-induced and control cells were treated with 2DG. As shown here, acute treatment of 2DG completely abolishes E2F1-induced mTORC1 activationFigure ( 17). As expected, AMPK is activated upon 2DG treatment reflecting the energetic unbalanceFigure ( 17).

68 RESULTS Figure 16. E2F1 activation maintains energetic balance. A) U2OS ER-E2F1 cells were serum-deprived overnight before treatment with OHT for 6 hours or 2DG (50 mM) and Oligomycin A (1 µM) for 1 hour. UPLC was performed to measure the indicated nucleotides concentrations. Total protein amount was used to normalize the results. B) The ratio AMP/ATP was calculated. Error bars represent S.E.M. for n=3. Statistical significance is represented as: * p < 0.05, ns: p > 0.05. C) U2OS ER-E2F1 cells were serum- deprived overnight before treatment with OHT for the indicated times. Indicated proteins were analyzed by Western Blot.

Glucose increases glycolytic flux and consequently induces other metabolic pathways such as mitochondrial oxidation, an essential route for energy production. To evaluate the contribution of either glycolytic pathway or mitochondrial oxidation onE2F1- induced mTORC1 activation, we measured mTORC1 activity after 2DG, Oligomycin A or their combination in control and E2F1-activated cells and analyzed protein expression by Western Blot.

RESULTS 69 Our results showed that in control cells both inhibitors reduce basal mTORC1 activation to the same extend. However, upon E2F1 induction, mTORC1 activity is strongly inhibited by 2DG and only slightly by Oligomycin A (Figure 18A). The combination of Oligomycin A and 2DG inhibitors completely blocks mTORC1 activation in both, control and E2F1-induced cells. Interestingly, Oligomycin A treatment tends to increase E2F1-induced mTORC1 activationFigure ( 18A and B). These results suggest that E2F1-induced mTORC1 activation rather relies on glycolysis than mitochondrial oxidation. AMPK activity, however, is increased upon 2DG and Oligomycin A treatment reflecting the energetic unbalance under these conditions (Figure 18A). However, E2F1-induced mTORC1 activation, even in cells with defective glycolysis or mitochondrial oxidation, is independent on AMPK, indicating an alternative regulatory mechanism downstream of E2F1 (Figure 18A).

Figure 17. Acute glycolysis blockade inhibits E2F1-driven mTORC1 activation. U2OS ER-E2F1 cells were serum starved overnight, treated with OHT for 5.30 hours prior to 2DG (50 mM) treatment for 30 minutes. Indicated proteins were analyzed by Western Blot.

Previous results from our laboratory demonstrated that E2F1-induced mTORC1 activation occurs through the recruitment of the complex to the lysosome, which is necessary for its activation (Meo-Evoli et al., 2015; Real et al., 2011). To investigate whether glucose induced mTORC1 translocation to the lysosome leading to kinase activation, LAMP2 and mTORC1 intracellular distribution and potential colocalization were analyzed by immunofluorescence in the absence or presence of glucose in cells treated or not with OHT. Our results confirm the effect of E2F1 on mTORC1 translocation in the presence of glucoseFigure ( 19A and B). Furthermore, E2F1- induced mTORC1 lysosomal translocation is abolished upon glucose starvation. Indeed, glucose itself significantly increases mTORC1 recruitment to the lysosomal

70 RESULTS Figure 18. E2F1-induced mTORC1 activation relies on glycolysis rather than mitochondrial oxidation. U2OS ER-E2F1 cells were serum-deprived overnight, treated or not with 2DG (50 mM) or Oligomycin A (1 µM) (OLIGO) for 1hour prior to treatment with OHT for 6 hours. A) Indicated proteins were analyzed by Western Blot. B) Densitometry of protein gels was performed. Error bars represent S.E.M. for n=3 independent experiments. Statistical significance is represented as: * p < 0.05, ns: p > 0.05. a.u.: arbitrary units (Fold change to CTRL).

RESULTS 71 72 RESULTS Figure 19. mTORC1 translocation to the lysosome is sensitive to glucose levels. U2OS ER-E2F1 cells were serum-deprived overnight, glucose-starved for 1 hour prior to treatment with OHT for 6 hours (A and B). A) Immunofluorescence using LAMP2 antibody for lysosomes detection (green), mTOR antibody (red) and DAPI staining for nucleus detection. White squares in the merged image show the image zone used for the magnification. Scale bar, 20 µm. B) Quantification of mTOR/LAMP2 colocalization is shown. Glucose 25 mM (w/ glucose) and Glucose 0 mM (wo glucose). Error bars correspond to S.D. (10 images). Statistical significance is represented as: * p < 0.05, *** p < 0.001, ns: p > 0.05. compartment in both control and E2F1-induced cells (Figure 19A and B). Since mTORC1 translocation to the lysosome is mediated by Rag GTPases, we evaluated the effect of glucose on mTORC1-Rag B binding. To this end, Rag B-FLAG ER-E2F1 stable U2OS cells were treated or not with OHT, in the presence or absence of glucose, and mTORC1 components interaction to Rag B was measured by immunoprecipitation. In line with immunofluorescence analysis, results showed that the absence of glucose reduces mTOR and Raptor binding to Rag B in both E2F1-induced and control cells (Figure 20). Together, these results demonstrate that glucose induces the translocation of mTORC1 to lysosomes through RagB GTPase. Furthermore, our data supports that E2F1 action on mTORC1 requires glucose.

Overall, these results demonstrate that glucose potentiates E2F1-induced mTORC1 activation independently of energetic balance. Moreover, we show that glucose induces the translocation and activation of mTORC1 to lysosomes through Rag B GTPase and suggest that E2F1 uses this mechanism to activate mTORC1.

Figure 20. mTORC1 lysosomal recruitment is affected by glucose levels. U2OS FLAG-Rag B ER-E2F1 stable cells were serum-deprived overnight, glucose-starved for 1 hour prior to treatment with OHT for 6 hours. Immunoprecipitation of FLAG-Rag B was performed, and the indicated proteins were analyzed by Western Blot. Densitometry of TOR signal is shown. A representative immunoblot of n=2 independent experiments is shown.

RESULTS 73 E2F1 induces glycolysis

To understand how glucose availability regulates E2F1-induced mTORC1 activity, we measured the effect of E2F1 on glycolytic flux by SeaHorse GlycoStress analysis. This technique consists on measuring the extracellular acidification rate (ECAR) on living cells. The first measurement is performed in cells incubated without glucose for 1 hour to shut down glycolysis. After this, cells are supplemented with 25 mM of glucose and the ECAR increases due to lactate extrusion (Glycolysis). After this, Oligomycin A is added to the medium to increase glycolysis to the maximum as cells cannot use the mitochondria for energy production (Glycolytic Capacity). Finally, 2DG is added to shut down glycolysis (Figure 21).

Figure 21. Schematic representation of GlycoStress Test. Interpretation of the results is shown. First measurements correspond to non-glycolytic acidification. After glucose addition, glycolysis canbe extrapolated and maximum glycolysis corresponding to glycolytic capacity of cells is obtained after Oligomycin A treatment. The last treatment is 2DG to recover the non-glycolytic acidification which is used as a base line.

Optimal cell number and drug concentrations were established in preliminary experiments. Glycolysis measurements were done in cells treated or not with OHT for 6 hours before the experiment. Results showed that cells with induced E2F1 display an increased basal glycolysis and glycolytic capacity, suggesting that E2F1 regulates glycolysis in our system (Figure 22A, B and C). Given that E2F1 activates mTORC1 and that mTORC1 has been shown to induce glycolysis (Moon et al., 2015; Sun et al., 2011), we proposed to test whether E2F1 effect on glycolysis potentiation was due to mTORC1 activation. To address this question, cells were treated with the mTORC1 allosteric inhibitor RAD001 and the ATP-site competitive inhibitor BEZ235 and E2F1 was induced by OHT treatment. As expected, mTORC1 inhibition instigates a robust decrease on the glycolytic capacity (Figure 23A and B). Strikingly, E2F1 effect on glycolysis is dominant over mTORC1 inhibition as it

74 RESULTS is still able to increase glycolytic flux upon mTORC1 inhibitionFigure ( 23), suggesting that E2F1 induction of glycolysis does not rely on mTORC1 activation.

In all, our results demonstrate that E2F1 induces glycolysis independently on mTORC1 activity.

Figure 22. E2F1 induces glycolysis. U2OS ER-E2F1 cells were serum-deprived overnight and treated with OHT for 6 hours prior to experiment. A) SeaHorse GlycoStress Test was performed. Representative graph of n=3 independent experiments is shown. ECAR rate was normalized by total protein amount in the different experimental conditions. B) Area under the curve (AUC) corresponding to glycolysis (measurements 5 – 9) is shown. C) Area under the curve (AUC) corresponding to glycolytic capacity is shown (measurements 5 – 9). Error bars correspond to S.E.M. for 5 replicates. Statistical significance is represented as: ** p < 0.005, *** p < 0.001.

Figure 23. E2F1-induced glycolysis potentiation is independent of mTORC1 activation. U2OS ER-E2F1 cells were serum starved overnight and treated with OHT, RAD001 and BEZ235 (RB) for 6 hours prior to experiment. A) SeaHorse GlycoStress Test was performed. Representative graph of n=3 independent experiments is shown. B) Area under the curve (AUC) corresponding to glycolysis (measurements 5 – 9) is shown. Error bars correspond to S.E.M. for 4 replicates. Statistical significance is represented as: * p < 0.05.

RESULTS 75 E2F1 modulates cytosolic pH

As protons are one of the final products of glycolysis and SeaHorse experiments showed an increased extracellular acidification upon E2F1 inductionFigure ( 22), we investigated the change on cytosolic pH in this condition, which has been correlated by other groups with mTORC1 activity (Balgi et al., 2011). To tackle this point, cells were treated or not with OHT overnight and stained with the cell-permeant pH indicator SNARF-AM, which exhibits a pH-dependent emission shift from yellow- orange to deep red fluorescence under acidic and basic conditions respectively (Figure 24A). Our results showed that E2F1 activation induces the alkalinization of the intracellular space ranging from 6.9 to 7.1 (Figure 24B).

Figure 24. E2F1 activation increases cytosolic pH. A) Standard curve to extrapolate cytosolic pH was performed. Graph shows the fluorescence of SNARF-AM at different cytosolic pH. B) U2OS ER-E2F1 cells were serum starved overnight and treated with OHT for 16 hours. Cytosolic pH was determined. Statistical significance is represented as: ***p < 0.001.

An increase in glycolytic flux and cytosolic pH suggest an increased extrusion of protons from the cytosol under E2F1 control. Analysis of microarray data openly available (GSE39136 - Shats et al., 2013) allowed us to study the expression of relevant pH regulatory families of proton transporters or carbonic anhydrases: Na+/ H+ exchangers (NHEs), monocarboxylate transporters (MCTs) and the carbonic anhydrases (CAs) upon E2F1 induction. From this study, we selected the pH regulator candidates MCT1, MCT7 and CA2 as E2F1 target candidates for the cytosolic pH regulation Figure( 25). In order to investigate whether cytosolic pH was implicated on E2F1-induced mTORC1 activation, we planned to decrease cytosolic pH by genetic depletion of MCT1, MCT7 and CA2 and mTORC1 activity was measured. CA2 and MCT7 depletion do not affect mTORC1 activity. Oppositely, MCT1 depletion reduces mTORC1 activity

76 RESULTS Figure 25. E2F1 modulates the expression of pH regulators. The heat map was generated from GSE39136 data and represents NHEs, MCTs and CAs expression. Red (low expression), green (high expression), black (intermediate expression). Blue squares indicate selected candidates: CA2, MCT1 and MCT7. both in control and in OHT treated cells (Figure 26). We used ACC phosphorylation as a reflection of energetic stress and showed that energetic balance is mainained upon these treatments (Figure 26). MCT1 mRNA levels upon E2F1 gain- and loss-of- function were determined by qPCR showing that effectively MCT1 is a target of E2F1 (Figure 27A). In parallel, intracellular pH was measured to evaluate the impact of MCT1 depletion on cytosolic protons. Unexpectedly, cytosolic pH is not affected by MCT1 depletion and the E2F1-induced intracellular alkalinization occurs even under these conditions suggesting that compensatory mechanisms could counteract MCT1 genetic deletion at the level of proton gradientFigure ( 27B). In fact, further analysis on MCT4 and CA2 mRNA levels confirmed this hypothesis, as both proteins were up-regulated after MCT1 knock-down indicating a complex regulatory network of cytosolic pH (Figure 27C and D).

RESULTS 77 Figure 26. E2F1 up-regulates pH regulators role on mTORC1 activation.U2OS ER-E2F1 were transfected with siRNA against CA2, MCT7 or MCT1 for 48 hours, serum-starved over-night and treated or not with OHT for 6 hours. Indicated proteins were analyzed by Western Blot. Representative immunoblots of n=3 experiments are shown.

Figure 27. MCT1 depletion does not affect cytosolic pH. U2OS ER-E2F1 were transfected with siRNA against MCT1 for 48 hours, serum-starved over-night and treated or not with OHT for 6 hours. A) Quantitative PCR was performed and the mRNA levels of MCT1 were determined. B) Cytosolic pH measurement was performed using SNARF-AM fluorescent probe. C-D) Quantitative PCR was performed, and the mRNA level of the indicated genes was determined. Actin was used as housekeeping gene. Statistical significance is represented as: ***p < 0.001. a.u.: arbitrary units (Fold change to CTRL).

78 RESULTS Overall, these results demonstrate that E2F1 induces cytosolic alkalinization, a common feature of cancer cells likely by the up-regulation of several proton transporters, at least MCT1, MCT4 and CA2.

While implication of pH on E2F1-driven mTORC1 activation was tested by genetic depletion of some of these proteins, none of them was able to reduce the cytosolic pH. Therefore a direct implication of protons on mTORC1 activation cannot be excluded from these studies. Finally, MCT1 depletion hampers mTORC1 activity suggesting a connection beyond pH regulation. As MCT1 transports lactate and protons through the plasma membrane, we hypothesize that the decreased mTORC1 activity found after MCT1 deletion could potentially be a consequence of glycolic flux reduction due to lactate accumulation.

PFKFB3 regulates E2F1-driven mTORC1 activity

The molecular mechanism by which E2F1 drives glycolysis might be broad, as this transcription factor regulates more than 30% genes of human genome (Bieda et al., 2006) and some E2F1 targets have been linked to its ability to modulate glycolysis in other cellular models. To understand which glycolytic enzymes are up-regulated by E2F1 in U2OS cells, we analyzed a set of microarray data openly available (GSE39136 - Shats et al., 2013).

Glycolysis-focused analysis was performed to elucidate the variation of expression of each family of glycolytic enzymes in U2OS cells Figure( 28A). This analysis showed that no major alterations at mRNA expression level are induced by E2F1 but some isoenzymes are specifically up-regulated. Fold increase upon OHT treatment was calculated and showed that PFKFB3 is the most up-regulated glycolytic enzyme in E2F1-activated cellsFigure ( 28B). PFKFB3 is an isoenzyme of PFK-2/FBPase-2 family, a bifunctional enzyme comprising both kinase and bisphosphatase activities. PFKFB3 is able to reversibly convert fructose 6-phosphate (Fru 6-P) to fructose 2,6-bisphosphate (Fru 2,6-P2), an activator of PFK-1. We validated PFKFB3 microarray data by quantitative PCR and measured its protein expression by Western Blot. Our results confirmed that E2F1 is able to induce PFKFB3 expression at mRNA and protein level (Figure 29A and B).

Overall, these results suggest that E2F1-induced glycolysis might be mediated by increased expression of key enzymes of glycolytic flux such as PFKFB3.

RESULTS 79 Figure 28. E2F1 modifies the expression of glycolytic enzymes. A) The heat map was generated from GSE39136 data and represents glycolytic enzymes expression. Red (low expression), green (high expression), black (intermediate expression). Gene names are grouped by families. Blue square indicates PFKFB3. B) Representation of fold increase of the highly expressed glycolytic isoenzymes. Red/green line indicate 5% of decrease/increase threshold.

80 RESULTS Figure 29. E2F1 regulates the expression of PFKFB3. A) U2OS ER-E2F1 cells were serum-deprived overnight before treatment with OHT for 6 hours. Quantitative PCR was performed, and the mRNA level of the indicated gene was determined. Actin was used as housekeeping gene. Statistical significance is represented as: ** p < 0.005. B) Indicated proteins were analyzed by Western Blot. Densitometry of bands over time is represented in the graphic.

The fact that E2F1 induces PFKFB3 expression raises the possibility that this isoenzyme mediates the E2F1-driven mTORC1 activity. To tackle this hypothesis, we tested the effect 3-(3-pyridinyl)-1-(4-pyridinyl)-2-propen-1-one (3PO), a PFKFB3 specific inhibitor, on E2F1-induced mTORC1 activity. To this end, cells were treated with 3PO, with or without OHT and mTORC1 activity was analyzed by Western Blot. Results showed that mTORC1 activation is hampered by PFKFB3 inhibition, suggesting that PFKFB3 activity might be essential for E2F1-driven mTORC1 activationFigure ( 30A and B). To complement this result, genetic depletion of PFKFB3 was performed by silencing RNA. Accordingly, PFKFB3 depletion decreases the mTORC1 activityFigure ( 30C). Similar results were obtained in A549 cells (Figure 30D). AMPK phosphorylation is not affected by either genetic or chemical PFKFB3 inhibition indicating that mTORC1 regulation is independent on the energetic balanceFigure ( 30A and C).

PFKFB3 chemical or genetic inhibition should promote a decrease on Fru 2,6-P2 levels. To assess the impact of the reduction of this metabolite on mTORC1 signaling pathway we overexpressed TP53-induced glycolysis and apoptosis regulator (TIGAR), whose bisphosphatase activity hydrolyzes Fru 2,6-P2 into Fru 6-P. Our results showed that TIGAR overexpression reduces mTORC1 activity both in control and E2F1-activated cells (Figure 31). Furthermore, we asked whether PFKFB3 expression was sufficient to induce mTORC1 activity. To this end, U2OS cells were transfected with PFKFB3 expression vector and mTORC1 activity was analyzed by measuring S6K phosphorylation. Our results demonstrate that short PFKFB3 overexpression produces a striking increase of mTORC1 activity at 16 hours after transfectionFigure ( 32A). Of note, E2F1 over- activation effect on mTORC1 activation is milder (Figure 32A). In contrast, longer PFKFB3 overexpression shows a different behavior. It increases mTORC1 activity but the effect of E2F1 is dominant Figure( 32B). These results suggest that PFKFB3 could mediate E2F1-driven mTORC1 activation. Noteworthy, PFKFB3 over-expression does not largely modify AMPK phosphorylation, indicating that mTORC1 regulation by PFKFB3 activity is independent of the energetic balance.

Overall, these results suggest that PFKFB3 activity triggers mTORC1 activity and Fru 2,6-P2 levels could be sufficient to stimulate mTORC1 activity. Furthermore, these results suggest that E2F1, by increasing PFKFB3 expression, could favor mTORC1 activation by glucose.

RESULTS 81 Figure 30. PFKFB3 inhibition reduces mTORC1 activity. A) U2OS ER-E2F1 were serum-deprived overnight and treated or not with OHT or 3PO 25 µM for 6 hours. Indicated proteins were analyzed by Western Blot. B) Densitometry of protein gels was performed for n=3 independent experiments. Error bars correspond to S.E.M. for the 3 experiments. Statistical significance is represented as: * p < 0.05, ** p < 0.005, *** p < 0.001, ns: p > 0.05. a.u.: arbitrary units (Fold change to CTRL). C-D) U2OS ER-E2F1 (C) and A549 cells (D) were transfected with siRNA against PFKFB3 for 48 hours, serum-deprived overnight and treated or not with OHT for 6 hours (C). Indicated proteins were analyzed by Western Blot.

82 RESULTS Figure 31. TIGAR overexpression reduces mTORC1 activity. A) U2OS ER-E2F1 were transfected with TIGAR expression plasmid for 48 hours, serum-starved over-night and treated or not with OHT for 6 hours. Indicated proteins were analyzed by Western Blot. B) Densitometry of protein gels was performed for n=3 independent experiments. Error bars correspond to S.E.M. for the 3 experiments. Statistical significance is represented as: * p < 0.05, ** p < 0.005. a.u.: arbitrary units (Fold change to CTRL).

Figure 32. PFKFB3 overexpression induces mTORC1 activity. A) U2OS ER-E2F1 were transfected with PFKFB3 expression plasmid for 16 hours, serum-starved over-night and treated or not with OHT for 6 hours. Indicated proteins were analyzed by Western Blot. Densitometry of S6K phosphorylation is shown. B) U2OS ER-E2F1 were transfected with PFKFB3 expression plasmid for 48 hours, serum-starved over-night and treated or not with OHT for 6 hours. Indicated proteins were analyzed by Western Blot. Densitometry of S6K phosphorylation is shown.

RESULTS 83 PFK-1, but not aldolase or GAPDH, affects mTORC1 activity

PFKFB3 regulates glycolysis by modulating Fru 2,6-P2 levels, which allosterically activates the rate-limiting enzyme PFK-1. The fact that PFKFB3 expression modulates mTORC1 activity, suggests that PFK-1 activity could play a role in this process. To challenge this hypothesis, triple genetic depletion of the three PFK-1 enzymes, PFKM, PFKP and PFKL, was performed by silencing RNA. Results showed that PFK-1 depletion induces a slight decrease in S6K and subsequently S6 phosphorylation, suggesting that PFK-1 might participate on E2F1-driven mTORC1 activation. In contrast, no changes on AMPK activity are detected upon PFK-1 genetic depletionFigure ( 33A). Efficiency of the knock-downs was analyzed by quantitative PCR showing 80% reduction in all the genes mRNA (Figure 33B). Curiously, while PFK-1 isoenzymes mRNA decreased upon OHT treatment (Figure 33B), protein expression was not affected after E2F1 induction Figure( 33A). Knowing that PFK-1 and PFKFB3 are key regulators of glycolytic flux, we asked whether deletion of other glycolytic enzymes had a similar effect on mTORC1 activity. To address this question, we investigated the effect of GAPDH and aldolase deletion on mTORC1 activity; both enzymes acting downstream of PFK-1 on the glycolytic pathway. As glycolytic enzymes present redundancy in the human genome, we selected the most expressed forms of these enzymes for the following experiments based on the analyzed microarray data (Figure 28A). Based on this, GAPDH and aldolase A were studied. Results showed that, in contrast to PFK-1 and PFKFB3, GAPDH and aldolase A deletion do not prevent the activation of mTORC1 by E2F1, on the contrary, an increased phosphorylation of mTORC1 target S6K is detectable (Figure 34A and B). AMPK phosphorylation results increased under these conditions, which could indicate glycolysis slow down due to GAPDH and aldolase A depletion (Figure 34A and B).

Altogether, these results suggest that PFK-1 activity could participate on the regulation of E2F1-driven mTORC1 activation by PFKFB3. Moreover, the opposite effects of aldolase A / GAPDH and PFK-1 / PFKFB3 on E2F1-induced mTORC1 activity suggest that glycolytic regulation of mTORC1 occurs upstream of aldolase activity, likely at PFK-1 step.

84 RESULTS Figure 33. PFK-1 plays a role on E2F1-driven mTORC1 activation.U2OS ER-E2F1 were transfected with siRNA against PFK1 (isoenzymes PFKM, PFKL, PFKP) for 48 hours, serum-starved over-night and treated or not with OHT for 6 hours. A) Indicated proteins were analyzed by Western Blot. Densitometry of S6K phosphorylation is shown. B) Quantitative PCR was performed, and the mRNA levels of the indicated genes were determined. Actin was used as housekeeping gene. Statistical significance is represented as: * p < 0.05, ** p < 0.005, ***p < 0.001, ns: p > 0.05. a.u.: arbitrary units (Fold change to CTRL).

Figure 34. GAPDH or aldolase A deletion increase mTORC1 activity. A and B) U2OS ER-E2F1 were transfected with siRNA against GAPDH for 72 hours (A) or aldolase A (ALDO A) for 48 hours (B), serum- starved over-night and treated or not with OHT for 6 hours. Indicated proteins were analyzed by Western Blot. Representative blots of n=3 independent experiments are shown.

RESULTS 85 Glucose availability determines E2F1-dependent lysosome trafficking regulation

Lysosome positioning has been shown to determine mTORC1 activity (Korolchuk et al., 2011) and previous results from our laboratory showed that E2F1 activation induced the localization of lysosomes towards the cell periphery. As E2F1 drives both mTORC1 activity and lysosome peripheral distribution, and glucose affects mTORC1 activation, we tested whether glucose also affects lysosome positioning. To this end, cells were glucose starved for 1 hour prior to E2F1 activation by OHT for 6 hours and LAMP2-positive lysosomes were detected by immunofluorescence.

Our results showed that E2F1 induces lysosome peripheral localization Figure( 35A and B). Glucose starvation did not impact on lysosome-nucleus distance in control conditions, oppositely, lack of glucose inhibits E2F1-induced lysosomes distribution towards cell periphery, confining them to the perinuclear area Figure( 35A and B). These results suggest that E2F1-driven lysosome peripheral distribution might be dependent on glucose availability.

E2F1 and v-ATPase drive lysosome-dependent exocytosis and cell migration

Lysosomes are calcium-enriched organelles and their peripheral distribution is needed for processes such as lysosome-dependent exocytosis, which is associated with increased migration capabilities. As E2F1 has been shown to modulate cell migration (Wang et al., 2017), we studied whether E2F1 increased lysosome- dependent exocytosis. To tackle this question, cells were subjected to E2F1 genetic depletion and extracellular β-hexosaminidase (HEX) activity was measured. Results showed that E2F1 depletion reduces β-hexosaminidase release, indicating a role of E2F1 on lysosome-dependent exocytosis (Figure 36A). As lysosome acidification and positioning are linked events, we studied the role of v-ATPase on E2F1-induced lysosome-dependent exocytosis. To this end, V0c v-ATPase subunit was genetically depleted and E2F1 activity was induced by OHT Figure( 36B). Our results point out that E2F1 over-activation increases lysosome-dependent exocytosis and suggest that v-ATPase activity is essential for this process.

86 RESULTS Figure 35. Glucose availability affects lysosomes peripheral positioning induced by E2F1. ER-E2F1 stable U2OS cells were serum-deprived overnight and glucose-starved one hour prior to OHT treatment for 6 hours. A) Immunofluorescence was performed using LAMP2 antibody for lysosome detection and DAPI for nuclear staining. Scale bar, 10 µm. B) Quantification of the distance between lysosomes and the nucleus was performed. A total of 15 cells per condition was quantified. Statistical significance is represented as: ***p < 0.001.

RESULTS 87 To verify the activity of lysosomes in the different conditions we stained cells using Magic Red dye that emits fluorescence when lysosomal cathepsin B is active. Our results show that, although peripheral lysosomes are smaller in size, they are acidic and catalytically activeFigure ( 37). As expected, v-ATPase inhibition completely blocks cathepsin B activity, as shown by Magic Red signal shut-down both in control and OHT-treated cells (Figure 37).

Figure 36. E2F1 induces lysosome-dependent exocytosis in a v-ATPase-dependent manner. A) U2OS cells were transfected with silencing RNA against E2F1 (A) or V0c and treated or not with OHT for 6 hours (B) and extracellular β-hexosaminidase (HEX) activity was measured. Results were normalized by total cellular β-hexosaminidase activity as detailed in materials and methods. Error bars represent Mean +/- SEM for n>3 independent experiments. Statistical significance is represented as: * p < 0.05, ** p < 0.005, ns: p > 0.05.

Figure 37. Lysosome activity is preserved from perinuclear to peripheral lysosomes. U2OS ER-E2F1 stable cells were treated or not with Concanamycin A (10 nM) or OHT (400 nM) and living cells were immediately stained with Magic Red (red) for active cathepsin B and Hoechst dye (blue) for DNA staining. Representative epifluorescence images of cells is shown.

88 RESULTS Since E2F1 regulates lysosomal trafficking and lysosome-dependent exocytosis and both processes are related to malignancy and migration abilities, we tested whether E2F1 was implicated in cell mobility. To understand this, cells were depleted for E2F1 and cell migration was measured by live imaging. Results showed that E2F1 depletion reduces cell migration of U2OS cells as we previously proposed (Meo-Evoli et al., 2015). Furthermore, to assess the implication of v-ATPase, we depleted V0c subunit. In accordance with a reduced lysosome-dependent exocytosis, v-ATPase depletion reduces cell migration after 24 hoursFigure ( 38).

Figure 38. E2F1 and v-ATPase affect cell motility. U2OS cells were transfected with silencing RNA against E2F1 or V0c subunit of the v-ATPase for 24 hours and seeded into PDMS device to generate the cellular front. Confocal microscopy was used to perform live imaging for 24 hours every 20 minutes. A) Representative images of the indicated time points are shown. B) Quantification of migrating cells normalized by proliferating and death cells is shown. Error bars represent S.E.M. for n=3 experiments. Statistical significance is represented as: ** p < 0.005.

We confirmed these results using xCELLigence system (Roche), an impedance-based method and measured cell migration through collagen type I matrix. As expected, E2F1 depletion and v-ATPase chemical inhibition by Bafilomycin A shows a significant reduction in cell migration after 10 Figure hours( 39A). As lysosome-dependent exocytosis releases protons and hydrolytic enzymes to the extracellular medium that might be able to degrade extracellular matrix, we assessed the ability of cells to invade through matrigel matrix. To this end, the classical Boyden chamber was used and the number of invading cells was quantified. Results showed that cell ability to invade is reduced at the same extend than migration when either E2F1 or V0c subunit of v-ATPase were depleted (Figure 39B). These results suggest that E2F1 and v-ATPase increase both cell migration and invasion. From these studies, the implication of glucose on cell migration and invasion remains unresolved and will be addressed in near future.

RESULTS 89 Altogether these results indicate that E2F1 and v-ATPase are implicated on cell migration in U2OS cells and suggest a putative implication of lysosome-dependent exocytosis on this process.

Figure 39. E2F1 and v-ATPase promote cell migration and invasion. A) U2OS cells were seeded on xCELLigence devices previously coated with collagen type I. Live cell migration was measured for 24 hours. Cell migration after 10 hours is represented. B) U2OS cells were seeded on Boyden chambers previously coated with Matrigel. Number of invading cells was quantified after 24 hours. Error bars represent S.E.M. for n=3 experiments. Statistical significance is represented as: * p < 0.05, ** p < 0.005, ***p < 0.001.

90 RESULTS 2. LYSOSOMES PREVENT DEFECTIVE MITOSIS AND MEDIATE MITOTIC PROTEINS DEGRADATION

Lysosomal intracellular positioning is an important parameter for cellular metabolism as we characterized in the previous chapter of this thesis. As cytosolic vesicles with high capacity of degradation, the role of lysosomes has been extensively studied in interphase cells. Here, we aim to test whether lysosomes have a function specifically during the timely and highly orchestrated cell division process. Lysosomes are acidic vesicles active during cell division

Lysosomes implication during mitosis has been controversial for the last years of research (Liu et al., 2009; Mathiassen et al., 2017). Here, we propose to pinpoint LAMP2- positive lysosomes exclusively during the mitotic subphases by immunofluorescence in U2OS cells. Interestingly, lysosomes are present and dynamic in all mitotic subphases (Figure 40). Moreover, lysosome morphology and distribution change during mitosis as cells undergo extensive structural rearrangements. Indeed, after prophase lysosomes increase in size and decrease in number. Lysosomes maintain an

Figure 40. Lysosomes shape and number are modified in dividing cells. A) U2OS cells were fixed and immunofluorescence was performed using LAMP2 antibody (green) for lysosomes detection and DAPI (blue) for nuclear staining. Representative confocal images show different mitotic subphases and interphasic cell. Scale bar, 10 µm.

RESULTS 91 increased size from metaphase to late telophase when their shape and distribution is recovered at the interphase status, i.e. a cytosolic dispersed location Figure( 40). Mitotic lysosomes surround the mitotic spindle in metaphase and are moved to the edge of chromosomes in anaphase and telophase until chromosomes decondense.

To analyze whether mitotic lysosomes maintain their degradation capacity during cell division, we stained live Histone 2B-GFP (H2B-GFP) stable U2OS cells with both Lysotracker and Magic Red dyes. Lysotracker is a deep red-fluorescent dye for labeling and tracking acidic organelles such as lysosomes, whereas Magic Red probe contains a cathepsin B target sequence peptide RR2 linked to a red (Cresyl Violet) fluorescent probe that emits fluorescence when lysosomal cathepsin B is active. Thus, high colocalization between lysotracker and Magic Red implies the presence of an active lysosome with activated enzymes. Our results show thatmitotic cells, from prophase to telophase, contain catalytically active lysosomes (Figure 41).

Figure 41. Lysosomes are active vesicles in mitotic cells. U2OS H2B-GFP stable cells were stained with Lysotracker (green – arbitrary color) for lysosomes, Magic Red (red) for active cathepsin B, H2B-GFP (blue – arbitrary color) for DNA staining. Representative confocal images of cells in interphase cell or in mitosis are shown. Scale bar, 10 µm.

92 RESULTS Lysosomes are the end-point of recycling pathways such as macroautophagy, the process by which cells engulf cytoplasmic content to degrade and recycle biomass. Macroautophagy is characterized by the formation of double-membrane autophagosomes containing LC3 protein (Figure 5 - Introduction). To assess whether U2OS mitotic cells have an active autophagic flux, we synchronized cells at late G2 cell cycle phase using RO3306 (CDK1 inhibitor) and released them for 7 hours into normal medium with or without a v-ATPase inhibitor, Concanamycin A. Our results showed that both autophagic proteins, LC3 and p62 accumulate specifically during mitosis when the autophagic flux is inhibited by Concanamycin A treatment Figure( 42A).

Figure 42. Autophagy persists during mitosis. A) U2OS cells were synchronized at late G2 and released for 7 hours with or without Concanamycin A 10 nM treatment to allow the transition to G1 phase. The indicated proteins were analyzed by Western Blot. B) Representative confocal image of Concanamycin A-treated U2OS cell undergoing metaphase. The accumulation of LC3 (red) colocalizing with LAMP2 (green) is shown in yellow. Actin fibers were stained using phalloidin (gray) and DAPI for the nucleus (blue). Scale bar, 10 µm.

To corroborate that autophagic vesicles are present in dividing cells we directly assessed their presence by immunofluorescence detecting the colocalization of LC3 with LAMP2 lysosome marker after autophagic flux blockade. As expected, lysosomal compartment expands under Concanamycin A treatment and LC3-positive autophagosomes are detected next and within those lysosomes (Figure 42B). Thus, the accumulation of autophagic cargos clearly indicates that autophagic flux is functional during mitosis. In addition, Transmission Electron Microscopy (TEM) revealed the presence of double-membrane vesicles (autophagosomes) as well as dense single membrane vesicles (autolysosomes / lysosomes) in anaphase/telophase (Figure 43).

Collectively, these results show that both autophagic vesicles and functional lysosomes are present in mitotic cells.

RESULTS 93 Figure 43. Autophagic vesicles and lysosomes are present in dividing cells. TEM images show two mitotic cells undergoing early telophase. White squares indicate amplificated regions, black arrowheads indicate autophagosomes, blue arrowheads lysosomes and yellow asterisks point out mitochondria. Scale bars, as indicated.

Lysosome function and trafficking impairment delays mitotic progression

During mitosis, several morphological changes need to sequentially occur. Ubiquitin- dependent proteasome degradation is the major player involved in the orchestration of these events. However, lysosomes might be another important degradation route during mitosis thanks to their capacity to degrade all kind of biological material. Nonetheless, the implication of lysosomes in cell division has not been extensively documented. Here we proposed to test whether lysosomes are directly implicated in mitotic progression. To this end, we analyzed and quantified the total time of mitosis, from prophase to telophase, in cells with impaired lysosome function compared to control cells.

Mitotic timing is strictly regulated and remarkably constant among different cell types and species (Araujo et al., 2016), taking approximately one hour in normal mitosis. Live-imaging studies of H2B-GFP stable U2OS cells, treated or not with Concanamycin A, revealed that mitosis takes 22% more time to progress from prophase (chromosome condensation) to telophase (chromosome decondensation) when lysosomes are inactive Figure( 44A and B). As lysosome activity relies on both acidification capacity and intracellular positioning, measurement of cell division timing was also determined upon KIF5B depletion, an essential motor protein for lysosome anterograde movement. Interestingly, impaired lysosomal trafficking

94 RESULTS induced by KIF5B depletion also delays mitotic progression to the same extent than Concanamycin A treatment (Figure 44A and B). Moreover, the combination of lysosomotropic drug Concanamycin A and KIF5B silencing RNA has an additive impact on mitosis progression timing that can then last up to 3 hours instead of1hour (Figure 44B). These results indicate that integrity of both lysosome transport and functionality is a key factor for preserving mitotic timing.

Figure 44. Lysosome impairment delays mitotic progression. A) U2OS H2B-GFP stable cells were subjected to time-lapse imaging for 24 hours every 6 minutes 33 seconds. Representative images of control cells, cells treated with Concanamycin A 10 nM (Conc A) or depleted for KIF5B (siKIF5B) for 48 hours are shown. Scale bars, 5 µm. B) Quantification of mitotic timing of the indicated experimental conditions was performed from prophase to telophase. Error bars represent 5-95 percentiles of 90 ≤ mitosis. Statistical significance is represented as: ** p < 0.005, *** p < 0.001, ns: p > 0.05.

To better characterize in which mitotic phase lysosomes are involved, we separately analyzed the time from prophase to metaphase (mitotic entry) and from metaphase to telophase (mitotic exit). Our results indicate that both phases are delayed, although

RESULTS 95 mitotic exit seems to be the more altered Figure( 45A and B). Indeed, in our system, mitotic entry lasts for about 20 minutes in control condition (20.23 ± 5.8), whereas Concanamycin A treatment or KIF5B depletion leads to a 6 minutes delay to reach metaphase (25.52 ± 7.45 for Conc A; 27.31 ± 10.32 for siKIF5B). Similarly, mitotic exit last about 35 minutes in control condition (35.43 ± 8.3), while inhibition of either lysosomal acidification or trafficking generates about 8 minutes delay at the mitotic exit, lasting approximatively 40-45 minutes (42.5 ± 13.4 for Conc A; 44.001 ± 10.9 for siKIF5B).

Figure 45. Lysosome impairment delays mitotic entry and exit. Quantification of mitotic timing of the indicated experimental conditions was performed from prophase to metaphase (A) or metaphase to cytokinesis (B). Error bars represent 5-95 percentiles of 55 ≤ mitosis. Statistical significance is represented as: * p < 0.05, ** p < 0.005, *** p < 0.001, ns: p > 0.05.

As mitotic progression per se was delayed upon lysosome inhibition, we analyzed whether lysosome impairment also impacts on mitotic rate. To tackle this point, live-imaging of H2B-GFP stable U2OS cells was performed and total number of mitosis was quantified for 24 hours. Concanamycin A treatment impairs cell cycle progression, as fewer mitotic cells were detected compared to control. Interestingly, this phenomenon was exclusive of impaired lysosomal acidification, as KIF5B- depleted cells face further replication cycles like control cells Figure( 46A). To better understand how impaired lysosomal acidification impact on cell cycle progression, we evaluated whether cells treated with Concanamycin A underwent DNA damage response and consequently G1 cell cycle arrest. To this end, cells were synchronized at late G2 and released for 7 hours with or without the lysosomotropic drug. Histone 2AX phosphorylation (gH2AX) and p21 expression were evaluated by immunofluorescence (Figure 46D). Concanamycin A-induced lysosomal inhibition during mitosis induces a significant increase of cells with positive gH2AX and p21 stainingFigure ( 46B and C). Indeed, about 10% of Concanamycin A-treated cells elicit DNA damage response

96 RESULTS and consequent p21 increase indicating a G1 arrest Figure( 46B and C). Our results point out that Concanamycin A affects both cell cycle reentry and mitotic timing, whereas lysosome trafficking impairment specifically delays mitotic progression.

Figure 46. Number of mitosis is reduced upon v-ATPase inhibition. A) Number of mitotic cells was quantified from live imaging of H2B-GFP stable U2OS cells performed for 24 hours scanned every 9 minutes. Bar graph shows the total number of mitotic cells compared to total cell number in the indicated conditions. B) Representative confocal images of U2OS cells illustrating a “negative” vsa “positive” cell for γH2AX (in green) and p21 (in red) staining. Both proteins were detected with specific antibody by immunofluorescence. DAPI staining for nucleus detection in blue. Scale bar, 5 µm. C - D) U2OS cells were synchronized at late G2 phase and released with or without Concanamycin A for 7 hours. Immunofluorescence was performed using gH2AX antibody for DNA damage detection or p21 antibody and DAPI for nuclear staining. γH2AX (C) and p21-positive (D) cells were quantified compared to total number of cells. Error bars represent S.E.M. of 50 images (1416 cells for CTRL and 1317 for Conc A). Statistical significance is represented as: * p < 0.05, ** p < 0.005, *** p < 0.001.

One of the main causes of mitotic delay is the accumulation of mitotic errors such as lagging chromosomes in metaphase plate or chromosomal bridges in anaphase. These events affect the metaphase-anaphase checkpoint and increase the duration of cell division (Rieder and Maiato, 2004). Therefore, we detected and quantified the

RESULTS 97 number of mitotic errors upon lysosome inhibition. As expected, Concanamycin A treatment leads to the accumulation of several mitotic errors related to chromosomes segregation defect, such as misaligned metaphase platesFigure ( 47A and E) and chromosome bridges in anaphase (Figure 47B and E). Normal mitotic cells generate two symmetrical mitotic spindles poles for correct separation of sister chromatids. We tested whether lysosome impairment also affects mitotic spindle organization and number by detecting by immunofluorescence pericentrin, a centrosome marker.

Figure 47. Lysosome impairment increases mitotic errors.A-B) U2OS cells were synchronized at late G2 phase and released with or without Concanamycin A treatment for 1 hour (A) or 3 hours (B). DAPI staining was used for DNA detection. Frequency of misaligned metaphase plates and chromosomal bridges over total number of metaphases and anaphases, respectively were quantified. Error bars represent S.E.M. of n=3 independent experiments (>100 cells). C) U2OS cells were synchronized at late G2 phase and released with or without Concanamycin A treatment for 3 hours. Immunofluorescence was performed using pericentrin antibody to stain spindle poles and DAPI for DNA staining. Frequency of multipolar spindles was quantified. Error bars represent S.E.M. of n=3 independent experiments (>100 cells). D) H2B-GFP stable U2OS cells were subjected to live imaging for 24 hours every 9 minutes. Frequency of anaphase bridges was quantified. Error bars represent S.E.M. of n > 4 fields. Statistical significance is represented as: * p < 0.05. E) Representative images of mitotic errors are shown. Scale bar, 5 µm.

98 RESULTS Indeed, dividing cells present an increase of multipolar spindles under lysosomal impairment condition (Figure 47C and E). Thus, our results indicate that lysosome acidification blockade induces mitotic errors. To test whether the protective role of lysosomes against mitotic errors is specific of their acidification capacity or if their trafficking is also essential for correct mitosis, cells were depleted for KIF5B motor protein. Consistent with the observed mitotic delay, KIF5B-depleted cells show a higher frequency of chromosome bridges (Figure 47D).

In all, our results suggest that active lysosomes are key organelles to ensure correct mitosis progression.

Mitotic errors correlate with a novel nuclear phenotype: the toroidal nucleus

Defective mitotic progression generates a robust and striking nuclear phenotype in interphasic cells. Precisely, it corresponds to a DAPI-stained nucleus with a hole devoid of chromatin (Figure 48A). To identify the origin of this structure, we performed live- imaging studies using H2B-GFP stable U2OS cells and analyzed cells undergoing cell cycle. Our results showed that this nuclear phenotype forms upon mitosis, being detectable once chromosomes decondense (Figure 48B).

Figure 48. Nucleus with chromatin-free compartment is formed upon mitosis.A) U2OS cells were fixed and stained with DAPI for DNA detection. Scale bar, 5 µm. B) Live imaging of U2OS H2B-GFP stable cells for 16 hours every 93 seconds. Mosaic images indicate aberrant nucleus formation upon cell division. Yellow arrowhead indicates a lagging chromosome. Scale bar, 10 µm.

As this structure could eventually correspond to an enlarged nucleolus, we tested three different nucleolar markers: fibrillarin (dense fibrillar component), nucleolin (intranucleolar chromatin and pre-ribosomal particles) and UBF (Nucleolar transcription factor 1). Immunofluorescence analysis clearly discard that this

RESULTS 99 compartment is the nucleolus, as none of the three nucleolar markers stains the DAPI- free area (Figure 49). We further proposed to confirm a normal nuclear envelope architecture by immunofluorescence using Lamin B1 antibodyFigure ( 50A) and NPC (Figure 50B). Interestingly, nuclear envelope is correctly disposed around DNA and is excluded from this “hole”. As this area is distinct from the nucleus, we tested whether it engulfs cytosolic material. Indeed, LAMP2-positive lysosomes as well as Phalloidin-stained actin fibers are detected by confocal microscopy within this unconventional nucleus (Figure 50C). To confirm that this structure was not simply a nuclear invagination and/ or an artefact due to microscopy scanning limitations, immunofluorescence assay was

Figure 49. Chromatin-free structure does not correspond to enlarged nucleolus.A-C) U2OS cells were fixed and immunofluorescence was performed using fibrillarin nucleolar marker (A), nucleolin nucleolar marker (B) or UBF nucleolar marker (C) and DAPI for DNA detection. Scale bars, 5 µm.

100 RESULTS performed to stain both nucleus and lysosomes. Images were acquired with optimal settings to generate a 3D reconstruction using IMARIS softwareFigure ( 50D). High- quality images resulted in the characterization of a novel nuclear phenotype. Due to its donut-like shape, we named this nuclear structure “toroidal nucleus”. Ultrastructure analysis of the toroidal nucleus by TEM confirmed that cytosolic material, including autophagic vesicles and lysosomes are present within (Figure 50E). Therefore our results show that toroidal nucleus is an aberrant nuclear phenotype formed upon mitosis.

RESULTS 101 Figure 50. Toroidal nucleus characterization. A-B) U2OS cells were fixed and immunofluorescence was performed using lamin B1 antibody (A) or NPC antibody (B) for nuclear envelope staining and DAPI for DNA detection. Scale bar, 5 µm. C) U2OS cells were fixed and immunofluorescence was performed using LAMP2 antibody (green) for lysosomes, phalloidin (red) for actin filaments and DAPI forDNA detection. Scale bar, 5 µm. D) 3D reconstruction using IMARIS software was performed for images stack from toroidal nucleus (z every 0.1 µm for 11 µm). DAPI stained nucleus in blue and LAMP2-positive lysosomes in yellow. E) U2OS cells were treated with Concanamycin A for 24 hours, fixed and prepared for TEM. Images show toroidal nucleus surrounded by the nuclear envelope, arrowheads indicate autophagosomes. Scale bars, as indicated.

Interestingly, cells with toroidal nucleus can go through cell cycle, condensate their DNA and divide again (Figure 51). We observed, however, that toroidal nucleus division tends to generate more mitotic errors as well as a mitotic delay when compared to cells without this phenotype. As toroidal nucleus is formed after chromosome decondensation, we investigated whether this phenotype was a consequence of impaired nuclear envelope reformation. To tackle this question, we performed live imaging on cells stably expressing mCherry-Lamin A/C, a protein of the nuclear envelope. We could detect that recruitment of nuclear envelope proteins occurs prior to toroidal nucleus detection, indicating that nuclear envelope reformation can occur independently of toroidal nucleus (Figure 52).

Figure 51. Toroidal nucleus containing cells can undergo DNA replication and cell division. Live imaging of U2OS H2B-GFP cells for 16 hours every 93 seconds. Mosaic images indicate toroidal nucleus undergoing cell division. Arrowheads indicate mitotic errors leading to chromosomal instability (yellow for lagging chromosome, orange for chromosomal bridge and green for micronuclei). Scale bar, 10 µm.

Given that toroidal nucleus was formed at late mitosis, we asked whether it could be directly related to mitotic defects. To tackle this question, we analyzed cells undergoing mitosis and quantified the correlation between mitotic error occurrence and toroidal nucleus formation. This analysis showed a significant correlation between dividing cells with detectable mitotic errors and the presence of toroidal nucleus in at least

102 RESULTS Figure 52. Nuclear envelope formation precedes toroidal nucleus appearance. Live imaging of mCherry-Lamin A/C stable U2OS cell undergoing mitosis for 16 hours every 5 minutes. Nuclear envelope reformation can be observed. Scale bar, 5 µm. one of the daughter cells (Figure 53A). Indeed, more than 80% of toroidal nuclei are the result of cells undergoing detectable defective mitosis (Figure 53B). As toroidal nucleus is mostly maintained in interphase, this nuclear phenotype can be scored as a novel read-out for mitotic errors. This observation gave rise to the question whether specific mitotic stressors also induce toroidal nucleus formation. To address this point, we analyzed cells depleted for two kinesins implicated in chromosome segregation: KIF11, required for establishing a bipolar spindle during mitosis and mitotic centromere-associated kinesin (MCAK/ KIF2C), which regulates the turnover of microtubules at the kinetochore. Results showed that either KIF11 or MCAK depletion induce a significant increase in toroidal nucleus formation (2.2 and 3.3-fold increase, respectively)Figure ( 54A). Consistently, depletion of the Centromere-associated protein E (CENPE), which drives chromosome congression (alignment of chromosomes at the spindle equator for the metaphase plate formation) produces a 3.8-fold increase on toroidal nucleus formation. Furthermore, we studied the effect of chemical-induced mitotic impairment on

RESULTS 103 Figure 53. Toroidal nucleus formation correlates with mitotic errors occurrence. A) Live imaging of U2OS H2B-GFP stable cells for 24 hours every 9 minutes. Mitotic cells were followed, and the presence of mitotic errors and toroidal nucleus was analyzed. Lineal regression was represented with the corresponding Pearson coefficient. B) Analysis of the percentage of toroidal nucleus formed upon normal mitosis or as a consequence of mitotic errors. Error bars represent S.E.M. (295 mitosis). Statistical significance is represented as: ***p < 0.001.

Figure 54. Mitosis impairment generates toroidal nucleus. U2OS cells were depleted for KIF11 or MCAK for 48 hours using silencing RNA or treated with Monastrol 100 µM or Nocodazole 1 µM for 24 hours. Immunofluorescence was performed and DAPI staining was used to detect the nucleus. Representative images are shown in panel A. Quantification of toroidal nuclei formed under those conditions is shown in panel B. Error bars represent S.E.M. of n > 3 experiments (10 fields / experiment). Statistical significance is represented as: ***p < 0.001.

104 RESULTS toroidal nucleus formation. Thus, cells were treated with mitotic poisons suchas Monastrol, a KIF11-specific inhibitor, or Nocodazole, a microtubule de-polymerizing agent. Results showed an increase in the prevalence of toroidal nucleus by 4.5-fold in both cases (Figure 54A). All these insults had none or minor effects on lysosomes shape or position (Figure 54B) confirming that mitotic impairment drives toroidal nucleus formation.

All together these results suggest a new way to detect mitotic progression defects in interphase cells by using toroidal nucleus as readout.

Lysosome disruption induces toroidal nucleus

The advantage that toroidal nucleus is present in interphase cells as a consequence of defective mitosis allowed us to examine whether lysosome impairment stimuli affect mitotic progression by analyzing the whole population instead of the reduced percentage of mitotic cells.

To characterize the implication of lysosome acidification on mitotic progression we treated cells with the v-ATPase specific inhibitor Concanamycin A, the lysosomotropic drug Chloroquine, or the genetic depletion of the V0c subunit of the v-ATPase (ATP6V0c) induced by silencing RNA. Effectively, lysosome acidification inhibition increased the formation of toroidal nucleus by approximately 2.5-3-foldFigure ( 55A). To control treatment efficiency, lysosome positioning and size were assessed by immunofluorescence and their inhibitory effect was confirmed as lysosomal compartment size increases in volume and is restricted to the perinuclear area (Figure 55B). Results showed that both lysosomotropic drugs increase toroidal nucleus formation but interestingly, their kinetics differFigure ( 55C). Indeed, Concanamycin A treatment acts faster and produces a pick effect at 24 hours, while Chloroquine produces a similar effect when incubated for 48 hours and induces the maximum of percentage of cells with toroidal nucleus at 72 hours.

Next, we proposed to test whether lysosomal inhibition directly applied during mitosis would affect toroidal nucleus formation rate. To this end, cells were synchronized at late G2 and released with or without Concanamycin A for only 7 hours. A two-fold increase in toroidal nucleus formation was detected confirming that the inhibition of lysosomal acidification capacity specifically in mitotic cells induces toroidal nucleus formation (Figure 55D). Of note, synchronization protocol was optimized to generate the minimum stress level possible, however basal level of toroidal nucleus population is slightly increased (from 2.5-3% to 4%).

RESULTS 105 Figure 55. Lysosome acidification inhibition increases toroidal nucleus formation. A) U2OS cells were treated with Concanamycin A (10 nM) for 24 hours, Chloroquine (10 µM) for 48 hours or depleted for the V0c subunit of the v-ATPase (ATP6V0c) for 48 hours. Immunofluorescence was performed, and nuclei were stained with DAPI for toroidal nucleus detection. Quantification of toroidal nucleus compared to total cell number is shown. B) Immunofluorescence under the indicated conditions was performed using LAMP2 antibody for lysosome detection and DAPI for nuclear staining. Scale bar, 10 µm. C) Toroidal nucleus quantification of U2OS cells were treated with Chloroquine (10 µM) for the indicated time points and DAPI staining was performed for nucleus detection. D) U2OS cells were synchronized at late G2 and released with or without Concanamycin A (10 nM) for 7 hours. Toroidal nucleus quantification was performed by DAPI staining. Error bars represent S.E.M. of n > 3 experiments (10 fields/experiment). Statistical significance is represented as: * p < 0.05, ***p < 0.001, ns: p > 0.05.

As lysosomotropic drugs are currently being used in the clinical setting, we aimed to decipher the potential correlation between drug sensitivity, delayed proliferation and toroidal nucleus formation in other models. To this end, we investigated whether toroidal nucleus was a common feature of cancer cell lines and not only specific of

106 RESULTS U2OS cells. We screened a panel of cell lines by comparing control versus 24 hours Concanamycin A treatment and evaluated the impact on toroidal nucleus formation (Table 2). Among the cell lines tested, A549 and HeLa exhibited toroidal nucleus and responded to Concanamycin A treatment like in U2OS cells. In contrast, we were not able to detect toroidal nucleus in colon carcinoma cell lines (RKO, LoVo or HCT116), breast adenocarcinoma MCT7 and normal kidney HEK293T cells, either in basal conditions or after 24 hours Concanamycin A treatmentTable ( 2). Interestingly, the two tested non-transformed cell lines (human skin fibroblasts HFF-1 and mouse embryonic fibroblasts MEF) display toroidal nucleus and elicit the response to Concanamycin A treatment (Table 2).

Name Treatment Toroidal nucleus (%) Fold increase Karyotype Human Control 2,32 U2OS 3,16 Highly altered Conc A 7,34 Control 0 HeLa 3,57 Aneuploidy Conc A 3,57 Control 0,08 A549 8,38 > 60 Chromosomes Conc A 0,67 Control ND HCT116 Mostly diploid Conc A ND Control ND MCF7 hypertriploidy to hypotetraploidy Conc A ND Control ND RKO Diploid Conc A ND Control ND LOVO Hyperdiploid Conc A ND Control ND HEK293T Hypotriploid Conc A ND Control 1,46 HFF-1 2,19 Diploid Conc A 3,21 Mouse Control 0,21 MEF 5,11 Aneuploidy Conc A 1,06

Table 2. Toroidal nucleus can be detected in a variety of cell lines. The indicated cell lines were treated or not with Concanamycin A (10 nM) for 24 hours, fixed and stained with DAPI for nuclear morphology detection. Quantification of toroidal nucleus was performed by microscopy in n=3 independent experiments (n > 200 cells/experiment). ND: not detected.

RESULTS 107 We previously observed that cells with toroidal nucleus are still able to progress into cell cycle as we detected toroidal nucleus-containing cells undergoing chromosome condensation and mitotic divisionFigure ( 50). On the other hand, Concanamycin A treatment significantly induces the formation of toroidal nucleusFigure ( 55) but also generates DNA response and G1 arrest (Figure 45). Therefore, we hypothesized that toroidal nucleus-enriched cell population might become more sensitive to cell death program activation. Thus, we designed a wash-out experiment in which G2- synchronized U2OS cells were released with or without Concanamycin A for 7 hours. As previously shown, Concanamycin A induces a significant increase in cell population with toroidal nucleus when applied specifically during mitosis. After the initial treatment, cells were either continuously treated with the lysosomotropic drug or released in normal medium for an additional 24 hours. Toroidal nucleus quantification uncovered thattoroidal nucleus percentage is restored to basal level after 24 hours without the treatment( Figure 56). Moreover, we discarded that cells with toroidal nucleus are more susceptible to elicit G1 arrest. Our results indicate that Concanamycin A effect on toroidal nucleus formation is reversibleFigure ( 56).

12 s u

e *

c l 10 u n

l ** a 8 d i o r

o 6 t

ns h t i

w 4 s l l

c e 2

f o 0 % Ctrl ConcA Ctrl ConcA Ctrl ConcA Syncro T24h T24h Washout

Figure 56. Toroidal nucleus population does not persist after a replication cycle. U2OS cells were synchronized at G2 phase and then treated or not with Concanamycin A (10 nM) for 7 hours (Syncro). Control and Concanamycin A treatment was maintained for 24 hours (T24h) or wash-out was performed in the Concanamycin A treated fraction for 24 hours (T24h Washout). DAPI staining was performed in fixed cells to quantify toroidal nucleus. Error bars represent S.E.M. of n=3 experiments. Statistical significance is represented as: * p < 0.05, ** p < 0.005, ns: p > 0.05.

Attempting to analyze the predisposition of Concanamycin A-treated cells to apoptotic cell death, we stained cells with Cell Event fluorescent probe containinga peptide of four amino acids (DEVD) conjugated to nucleic acids. DEVD peptide is cleaved by

108 RESULTS caspase-3/7 upon apoptotic cascade triggering. Once cleaved, the fluorescent probe is dissociated and interacts with DNA emitting fluorescence. Our live-imaging results show that in control cells there is no significant correlation between proliferating cells and apoptotic cells. However, in cells treated with the lysosomotropic drug, caspase- dependent cell death correlates with mitosis suggesting that impaired lysosome acidification promotes apoptotic cell death during cell divisionFigure ( 57A and B).

Figure 57. Cell death and mitosis correlate in cells treated with Concanamycin A. U2OS H2B-GFP stable cells were incubated with the cell-permeant Cell Event fluorescent dye and live-imaging every 9 minutes 29 seconds for 28 hours was performed in cells treated or not with Concanamycin A (10 nM) A) Representative images corresponding to initial and final time point are shown. Scale bar, 50 µm. B) Quantification of linear correlation between cell death and mitotic cells normalized by the total number of cells per field. Spearman coefficient was calculated for control and Concanamycin A-treated cells (p-value < 0.05 for Concanamycin A).

Finally, as we previously characterized the siKIF5B-induced mitotic errors occurrence and consequent mitotic delay, we asked whether defect in lysosome trafficking by other insults also induces toroidal nucleus formation Figure ( 58). To this end, we scored for toroidal nucleus formation in cells genetically depleted of the components of the lysosome motor core machinery Arl8b, SKIP and KIF5B and showed a significant increase of cells harboring toroidal nucleus (2.3-, 4.1- and 3.0- fold increase respectively) Figure( 58A). To determine the efficacy of the depletion, lysosome positioning was assessed by immunofluorescence and as expected cells depleted for BORC-associated proteins showed a perinuclear clustering of LAMP-2 positive lysosomes Figure( 58B). In addition, we examined the specific role of KIF5B in toroidal nucleus formation by overexpressing the neuronal form KIF5A in control cells or cells depleted for KIF5B. Effectively, KIF5A overexpression rescues the formation of

RESULTS 109 toroidal nucleus in both control and KIF5B-depleted cells (Figure 58C) suggesting that KIF5B-dependent lysosomal transport is necessary for correct mitotic progression and prevent toroidal nucleus formation. Interestingly, the depletion of other motor proteins involved in lysosomal trafficking such as KIF3A, KIF2A or KIF1A induced a significant increase in toroidal nucleus formationFigure ( 58D). In all, these results show that lysosomal trafficking is important for precise mitosis.

Figure 58. Lysosome trafficking impairment increases toroidal nucleus formation. A – B) U2OS cells were depleted for KIF5B, SKIP, or ARL8B using silencing RNA for 48 hours. Immunofluorescence was performed, and nuclei were stained with DAPI for toroidal nucleus detection. Quantification of toroidal nucleus compared to total cell number is shown (panel A). Representative images of intracellular distribution of LAMP2-positive lysosomes under the indicated conditions (panel B). Scale bar, 10 µm. C) U2OS cells were transfected with KIF5B silencing RNA and/or with KIF5A-GFP overexpression plasmid for rescue experiment. Quantification of toroidal nucleus compared to total cell number is shown. Error bars represent SEM of n > 3 independent experiments. Statistical significance is represented as: * p < 0.05, ***p < 0.001, ns: p > 0.05.

110 RESULTS As both lysosome acidification and trafficking are necessary for correct mitosis progression and the combined inhibition of both further delays mitotic timingFigure ( 44), we analyzed whether impairment of both acidification and trafficking also have an additive effect on toroidal nucleus formation. As expected, KIF5B depletion combined with Concanamycin A treatment induces an increased percentage of cells exhibiting toroidal nucleus when compared with single treatments (Figure 59). In all, our results suggest that both, lysosome acidification and trafficking are required to preserve mitotic integrity.

20 **

s ( % ) 18 u

e 16 c l

u 14 * n

l

a 12 d i

o 10 r * o t

8 h t i 6 w s

l 4 l

C e 2 0

Figure 59. Combination of lysosome trafficking impairment and lysosome acidification blockade are additive on toroidal nucleus formation. U2OS cells were depleted for KIF5B and the next day treated or not with Concanamycin A or Chloroquine for 24 hours. Cells were then fixed and stained with DAPI for toroidal nucleus detection. Error bars represent S.E.M of n > 3 independent experiments. Statistical significance is represented as: * p < 0.05, ** p < 0.005, ***p < 0.001.

Macroautophagy prevents defective mitosis

As lysosome impairment by acidification blockade or trafficking impairment provokes defective mitosis and knowing that macroautophagy is also activated during cell division (Figure 42 and 44), we asked whether autophagy inhibition induces toroidal nucleus formation. To this end, U2OS cells were genetically depleted for Atg5,a protein involved in LC3 lipidation during autophagosome elongation, and p62, an autophagy-specific adaptor protein involved in targeting ubiquitinated proteins for lysosome degradation. Interestingly, genetic depletion of both Atg5 and p62 in U2OS cells induces a significant increase in toroidal nucleus formation (1.6 and 2.8- fold increase, respectively) Figure ( 60A). The efficiency of protein depletion was

RESULTS 111 controlled by Western Blot (Figure 60A and B). Results show that effective inhibition of autophagosome formation and p62-dependent lysosomal degradation induce toroidal nucleus formation Figure( 60).

Figure 60. Autophagic proteins depletion causes toroidal nucleus formation. U2OS cells were transfected with siRNA against Atg5 and p62 proteins for 48 hours. A) Immunofluorescence was performed and DAPI staining was used for toroidal nucleus detection. Error bars represent S.E.M. of n > 3 independent experiments. Statistical significance is represented as: ** p < 0.005. B-C) Indicated protein expression was analyzed by Western Blot.

As U2OS is a cancer cell line that is already susceptible to genomic instability and chromosome alterations, we took advantage of mouse embryonic fibroblasts (MEFs) to analyze the effect of defective autophagy on mitotic progression. Of note, we chose MEF cells as we previously tested them for Concanamycin A-induced toroidal nucleus formation Figure( 54). To this end, we used Atg5 wt and Atg5 KO mouse embryonic fibroblasts (MEF) and analyzed the percentage of toroidal nucleus in the whole population. In agreement with Atg5 depletion in U2OS cells, Atg5 KO MEFs present a significant increased percentage of toroidal nucleus in normal growing conditions, supporting that basal autophagy prevents cells from mitotic errorsFigure ( 61A). Moreover, Concanamycin A-induced toroidal nucleus formation was abolished in Atg5 KO MEFs (Figure 61A). To confirm that the system was working, we analyzed the expression of Atg5 levels as well as the autophagic response to Concanamycin A treatment. As expected, Atg5 KO MEFs harbor a defective autophagic fluxFigure ( 61B).

Altogether, our results point out the involvement of macroautophagic machinery in mitotic progression.

112 RESULTS Figure 61. Atg5 KO MEFs exhibit a higher percentage of toroidal nucleus. MEF wt and Atg5 KO were subjected or not to Concanamycin A treatment for 24 hours. A) After treatment cells were fixed and stained with DAPI for nuclear detection. Toroidal nucleus was quantified. Error bars represent S.E.M. of n=3 independent experiments. Statistical significance is represented as: * p < 0.05, ns: p > 0.05. B) Indicated proteins were analyzed by Western Blot.

Novel lysosome protein substrates are implicated in cell division

Based on the discovery of a novel role of lysosomes and macroautophagy in mitotic progression, we next asked whether specific proteins implicated in mitosis were in fact degraded by the autophagic-lysosomal pathway. To address this question, we proposed to identify lysosomal substrates by mass spectrometry analysis. G2- synchronized U2OS cells were released with or without Concanamycin A for 8 hours and G1-enriched cell fractions were processed for mass spectrometry analysis. As we characterized that inhibition of lysosomal acidification capacity induces a mitotic delay (Figure 44B), we confirmed that the majority of the cells are in G1 phase by FACS cell cycle analysis (Figure 62A). Of note, the mitotic delay induced by the lysosomotropic drug is detectable by FACS after 8 hours (increased G2/M fraction compared to control). However more than 85% of the cells are in G1 phase in both fractions Figure( 62A). Of note, we could detect the accummulation of the autophagic proteins LC3 and p62 in this fraction confirming that autophagic flux is inhibited by Concanamycin A (Figure 62B). We hypothesized that Concanamycin A fraction will be enriched in putative lysosomal protein substrates compared to control fraction, in which lysosomal-dependent degradation would have normally occurred during mitosis. Proteomic data analysis

RESULTS 113 Figure 62. Validation of cellular fractions for Mass Spectrometry analysis.U2OS cells were synchronized at G2 as described in materials and methods. A) Propidium iodide staining was used to follow the DNA amount. The percentage of cells in each cell cycle phase (G1, S, G2/M) is represented. B) Indicated proteins were analyzed by Western Blot. identified more than 250 proteins that are either accumulated or exclusively present in Concanamycin A-treated fraction.

The obtained list of candidates was subjected to functional characterization using Reactome, a free database of biological pathways (Fabregat et al., 2018; Stein, 2004) that allows protein clustering into cellular and organism functions. Reactome analysis of our candidates highlights that proteins accumulated after lysosome inhibition are mainly involved in cell cycle, vesicle-mediated transport, DNA repair or cell death (Figure 63). Later, we performed a gene ontology (GO) analysis using David GO web tool (Huang et al., 2009a, 2009b) describing more than 100 GO annotations significantly increased in our data set. Of the enriched GO terms with a false discovery rate bigger than 10-5, four functional annotations related to cell cycle or mitotic progression were enlighted (Figure 64). Subsequently, we selected the 56 protein candidates that clustered in functional annotations related to mitosis and cell cycle regulation (Figure 65). Among them, we found members of the cohesin complex (PDS5 and WAPL) responsible of chromatids congression during mitosis, regulatory kinases such as PLK1, which is involved in centrosome maturation and spindle assembly, removal of cohesins from chromosome arms and regulation of mitotic exit and components of the ESCRT-III complex related to membrane fusion events (CHMP5, CHMP2B and CHMP1A). To validate mass spectrometry data, cells were synchronized at G2 and released with or without Concanamycin A. After 3 hours, mitotic cells were isolated by shake-off and protein expression was determined by Western Blot. Results show that PDS5, WAPL and PLK1 proteins accumulate upon lysosome inhibition Figure( 66). Interestingly, none of these proteins was already described as a lysosomal target.

114 RESULTS 0 p-value 0.05

Figure 63. Representation of pathways affected by lysosome impairment during cell division.Results from mass spectrometry were subjected to functional annotation analysis using online Reactome tool. (https://reactome.org/). Yellow lines represent significantly affected pathways (p-value < 0.05).

Figure 64. Gene ontology analysis of all proteins up-regulated or only present in Concanamycin A treated cells. Mass spectrometry results were filtered and protein hits that presented a fold change of more than 1.2 (Conc A vs ctrl) and those only present in Conc A cells were analyzed using David GO gene ontology web tool (https://david.ncifcrf.gov/). False discovery rate threshold was settled at 10-5. Functional annotations related to cell cycle and mitosis are represented in red.

RESULTS 115 Figure 65. Functional annotation clustering of proteins related to cell cycle regulation and mitotic progression. Mass spectrometry results were filtered and protein hits that presented a fold change of more than 1.2 (Conc A vs ctrl) and those only present in Conc A cells were analyzed using David GO gene ontology web tool. Functional annotation clustering was performed. The heat map represents all the candidate proteins involved in cell cycle regulation and mitotic progression. Yellow squares indicate matching with the GO term described; Black squares for missing relation. Blue squares indicate candidates indicated in the text.

116 RESULTS Figure 66. Pds5, WAPL and PLK1 are accummulated in mitotic cells upon lysosome inhibition. U2OS cells were synchronized at late G2 and treated or not with Concanamycin A 10 nM for 3 hours. Mitotic cells were isolated by shake off and indicated proteins were analyzed by Western Blot.

In all, these results firmly suggest that lysosome-dependent degradation occurs during cell division and suggest that Pds5, WAPL and PLK1 are novel putative lysosomal targets.

RESULTS 117 Dust storm. Immunostaining of U2OS cells actin cytoskelleton DISCUSSION

Tumorigenesis is a multistep process driven by genetic alterations that induce the progressive transformation of a normal cell into a malignant tumor cell. During this process, cells undergo profound adaptations to grow into a tumor mass which are known as the hallmarks of cancer and account for mechanisms of cell death resistance, self-sufficiency in growth signals, anti-growth signals evasion, angiogenesis induction, replicative immortality and invasion/metastasis activation (Hanahan and Weinberg, 2000, 2011).

Lysosomes biology has been studied in the context of cancer alterations as they play essential functions in the maintenance of cellular homeostasis, especially based on their classical role as degradative organelles at the last step of autophagy. By promoting cell survival, autophagy plays a controversial role in cancer as this process promotes or suppresses tumorigenesis in a cell-, tissue- and stage-dependent manner (Levy et al., 2017). Lysosomes present a similar controversial role as they are also related to cell death through their role in lysosome-dependent cell death which triggers the apoptotic cascade. Thanks to their capacity to degrade all kind of biological material as well as voluminous organelles, lysosomes are considered the degradative organelles of mammalian cells, sharing this function with the UPS used to get rid of unwanted proteins. Nowadays, however, lysosomes are studied beyond their degradative role, as they are now also considered major cytosolic signaling hubs that gather information on nutrients status and forward it to signaling networks to dictate cellular metabolism.

In this thesis, we focus on the study of lysosomes in cancer cells considering two different contexts. On one hand, the regulation of mTORC1 and lysosome positioning under the regulation of E2F1 transcription factor, involved in the G1/S transition of the cell cycle. In this part, we develop the idea of E2F1 as a coordinator of glycolytic metabolism and cell growth at the lysosomal surface. On the other hand, the study of the status, function and importance of lysosome degradation in mitosis, the shortest, but nonetheless crucial, phase of cell cycle where cells finally divide into the two daughter cells. In this section, we work on the hypothesis that lysosome-dependent degradation could potentially support mitotic progression and complement proteasome-dependent degradation pathway, which has been largely studied in the context of cell division.

DISCUSSION 121 E2F1 regulates mTORC1 and lysosomal trafficking through glycolysis induction

Many cancer cells display a markedly altered energy metabolism which relies on aerobic glycolysis rather than mitochondrial oxidation for energy production. This metabolic adaptation is commonly known as the Warburg effect and facilitates the generation of metabolic intermediates used for biosynthetic processes. In this context, glucose emerges as a key nutrient, not only essential for cell energy production, but also to promote cell growth and proliferation. Therefore, organisms must sense the abundance of glucose to adapt its energetic and biosynthetic status to the environmental changes. Glucose sensing has been broadly studied in yeast, where many interconnections exist between sophisticated glucose sensing systems that enable yeast cells to fine tune cell growth, cell cycle and their coordination in response to nutritional changes. The aim of the present study is to uncover a potential participation of glucose on E2F1-driven mTORC1 activation and lysosomal trafficking regulation.

Here we demonstrate that glucose potentiates mTORC1 activation induced by E2F1 and show for the first time that glucose increases mTORC1 recruitment to the lysosome in osteosarcoma cells. In accordance, Zhang et al. showed the direct relation between glucose and mTORC1 lysosomal localization under Axin and v-ATPase control (Zhang et al., 2014). We uncover that glycolysis, and not glucose itself, activates mTORC1 upon E2F1 induction suggesting the presence of a sensing machinery that transduces glucose scarcity to the v-ATPase/Ragulator/Rags thus inducting mTORC1 translocation and consequent activation.

Our data also shows that glucose is essential for lysosomal trafficking towards cell periphery induced by E2F1. In this regard, mTORC1 activity has been correlated to lysosome localization and some reports point out that lysosome peripheral localization promotes mTORC1 activation (Korolchuk et al., 2011). This correlation suggest that lysosome perinuclear localization induced by glucose scarcity could be responsible of the reduced mTORC1 activity. However, other stimuli that induce lysosome perinuclear clustering such as KIF5B knock-down have no effect on mTORC1 activation (data not shown) suggesting that the relation between lysosome positioning and mTORC1 activity is not direct. Interestingly, amino acid starvation has been shown to affect lysosomal trafficking inducing perinuclear clustering. Recently, the molecular mechanism underlying amino acid-dependent lysosomal trafficking modulation has been uncovered. Indeed, it has been proposed that Ragulator interacts with BORC,

122 DISCUSSION directly transducing amino acid scarcity to KIF5 motor protein and inducing lysosome perinuclear localization (Pu et al., 2017). We show for the first time that not only amino acids but also glucose starvation promotes perinuclear clustering of lysosomes, suggesting a novel interconnection between the nutritional status of the cell and lysosomal trafficking. Thus, E2F1, by increasing glycolytic flux, is able to increase both mTORC1 activity and lysosome trafficking towards cell periphery.

As glycolysis blockade generates energetic stress and consequent AMPK activation, glycolysis induction could potentially reduce basal AMPK and consequently increase mTORC1 activity. However, we show here that E2F1 induction do not alter the adenylate energy charge nor AMPK phosphorylation excluding the implication of this kinase on E2F1-induced mTORC1 activation. Moreover, these results suggest that the ATP produced by glycolysis induction might be used by the cell for other processes thus maintaining the global energy balance. Interestingly, E2F1 induction tends to increase the global amount of all adenyl nucleotides possibly due to the up-regulation of nucleotide synthesis required for S phase.

Attempting to uncover the mechanism by which glycolysis potentiates E2F1-induced mTORC1 activation and regulates lysosomal trafficking, we have investigated the alteration of cytosolic pH induced by E2F1. Aerobic glycolysis ends up with the generation of lactate and protons that could lead to cytosolic acidification, thus inducing the peripheral localization of lysosomes as suggested before (Korolchuk et al., 2011). Oppositely, here we describe the alkalinization of the intracellular pH in E2F1-activated cells. An increased pH supports cell proliferation and evasion of apoptosis, facilitates metabolic adaptation and helps efficient directed cell migration (Lagadic-Gossmann et al., 2004; Matsuyama et al., 2000; Webb et al., 2011). Indeed, it has been described that intracellular acidification inhibits mTORC1 activity (Balgi et al., 2011), suggesting that the E2F1-induced mTORC1 activation could rely on cytosolic pH modulation. The increase in the cytosolic pH together with the decreased extracellular pH showed by SeaHorse experiments, suggests that E2F1 increases the extrusion of protons from the cell likely by the up-regulation of proton transporters. In this regard, we identified MCT1, MCT7 and CA2 as potential E2F1-target genes implicated in cytosolic pH regulation. However, none of these transporters’ depletion affects cytosolic pH. Interestingly, we describe here that MCT1 depletion induces the expression of MCT4 and CA2 suggesting compensatory pathways on the regulation of pH equilibrium. Despite MCT1 deletion does not acidify the cytosol, its genetic depletion reduces mTORC1 activity. Therefore, we hypothesize that MCT1 deletion could increase intracellular lactate thus reducing the glycolytic flux and inhibiting mTORC1.

DISCUSSION 123 From this study, the PFK-2/PFK-1 glycolytic branch emerges as the putative regulatory machinery controlling glucose-induced E2F1-driven mTORC1 activation. We demonstrate that E2F1 increases the glycolytic flux and identify PFKFB3 as the most up-regulated glycolytic enzyme upon E2F1 induction. PFKFB3 has been widely studied in the context of cancer as this isoenzyme has the highest kinase:phosphatase activity towards the PFK-2/FBPase-2 family (700:1) which makes it extremely effective as glycolysis inducer (Shi et al., 2017). Accordingly, PFKFB3 is overexpressed in several malignancies such as ovarian and thyroid carcinomas, colon adenocarcinoma or breast cancer (Bartrons et al., 2018). Interestingly, other PFK-2 isoenzymes are regulated upon proliferative stimulation in other systems (Dupriez et al., 1993; Joaquin et al., 1997a, 1997b) and E2F1 has been shown to up-regulate the PFKFB1 gene product fetal PFK-2 (fPFK-2) (Fernández de Mattos et al., 2002). Moreover, PFKFB3 expression is modulated in a cell cycle-dependent manner and specifically up-regulated during G1 phase of the cell cycle corresponding to high E2F1 activity (Duran et al., 2009).

Noteworthy, we found that PFKFB3 loss- and gain-of function causes a concomitant change on E2F1-induced mTORC1 activity. Moreover, TIGAR overexpression, which decreases the Fru-2,6-P2 levels inhibiting glycolysis (Bensaad et al., 2006), reduces mTORC1 activation. Our results suggest a potential regulation of glycolysis-dependent mTORC1 activation by the PFK-2 product Fru-2,6-P2 or by the activity of the glycolytic enzyme PFK-1, which is allosterically activated by this metabolite. In this regard, PFK- 1 deletion slightly reduces mTORC1 activity. Interestingly, depletion of other down- stream glycolytic enzymes has opposite effects. Indeed, aldolase A depletion has no effect on mTORC1 activity and GAPDH deletion increases it. GAPDH interaction with Rheb GTPase upon glucose starvation has been already described supporting our results (Lee et al., 2009). Thus, E2F1 glucose-driven mTORC1 regulation likely occurs at the PFK-2/Fru-2,6P2/PFK-1 glycolytic node.

Despite the direct implication of PFKFB3 and PFK-1 on glycolysis progression, their depletion do not activate AMPK, excluding any role of this kinase in the effect of glycolysis on E2F1-induced mTORC1 regulation. We suggest that cells can overcome PFKFB3 and PFK-1 depletion at the level of energy status, but inhibit cell growth upon glucose scarcity. Oppositely, deletion of both aldolase A and GAPDH, down-stream enzymes of PFK-1, highly increases AMPK activity. This activation can be either a consequence of energetic imbalance due to glycolysis inhibition (Gwinn et al., 2008; Lin and Hardie, 2018) or to the direct regulation of AMPK at the lysosomal surface by aldolase (Zhang et al., 2017).

Surprisingly, aldolase A- and GAPDH-dependent AMPK activation does not inhibit mTORC1, suggesting that under these conditions mTORC1 and AMPK are

124 DISCUSSION independently regulated. One explanation of this controversy would be that different pools of lysosomes act to activate either AMPK or mTORC1. It has been shown that Ragulator, which constitutively binds lysosomal membrane, is a common anchor for mTORC1 and AMPK at the lysosome (Zhang et al., 2014). Therefore, it is possible that two different machineries under glycolysis control are present in two different lysosome pools. Complementary experiments to asses this differential regulation should be done in the future.

Results from this work support a novel model of glycolysis-induced mTORC1 activation by the PFK-2/Fru-2,6P2/PFK-1 node. Upon glucose uptake, glycolysis gets activated and converted into Fru-6P. We propose that PFK-1 might be located at the lysosomal surface and act as sensor for the glycolytic flux (i. e. glucose levels) (Figure 67), thus increasing v-ATPase activity and mTORC1. In this context, E2F1, by increasing the expression of PFKFB3, would increase Fru-2,6-P2 metabolite, thus increasing PFK-1 activity. Interestingly, the interaction between PFK-1 and v-ATPase has already been described to induce V1/V0 assembly in yeast and renal cells (Chan and Parra, 2014; Su et al., 2003). Moreover, the alkalinization of the cytosolic pH induced by E2F1 could also potentiate the PFK-1 activity, thus activating mTORC1. This model is supported by other publications describing a modulation of PFK-1 activity by small changes in pH, so intracellular alkalinization would promote higher affinity of this enzyme for its substrate (Andrés et al., 1990; Trivedi and Danforth, 1966). Importantly, mTORC1 activation at the lysosomal surface requires v-ATPase activity, as its genetic and chemical inhibition hampers mTORC1 recruitment to the lysosome upon amino acid or glucose stimulation (Bar-Peled et al., 2012; Zhang et al., 2014). However, the implication of v-ATPase in this regulation is still controversial, as its modulation by glucose availability or amino acids is poorly understood, hindering the link with mTORC1 activation. In this sense, there are supportive evidences for mTORC1 activation by increased v-ATPase assembly (Bar-Peled et al., 2012; Zoncu et al., 2011) but also some reports point out that v-ATPase disassembly is triggered by nutrients such as glucose or amino acids, common activators of mTORC1 (McGuire and Forgac, 2018; Parra and Hayek, 2018; Stransky and Forgac, 2015).

Finally, our studies point out a relevant role of E2F1 on regulating lysosome-dependent exocytosis, a calcium-regulated process that allows the extrusion of lysosomal content, the acidification of the extracellular space and the enrichment of v-ATPase in the cell surface which facilitates cytosolic protons extrusion. Our data suggest that v-ATPase mediates E2F1-induced lysosome-dependent exocytosis. However, an important question about the effect of glucose starvation on lysosome-dependent exocytosis remains to be resolved. E2F1 does not only increases lysosome-dependent exocytosis, but also the expression of proton transporters and aerobic glycolysis

DISCUSSION 125 inducing the acidification of the extracellular medium. Extracellular acidification promotes extracellular matrix remodeling and stimulates the activity of acid-activated proteases to facilitate cell dissemination (Kato et al., 2013; Webb et al., 2011). In this direction, E2F1 and v-ATPase depletion reduce cell migration and invasionFigure ( 67). Lysosomes are emerging as key organelles for cellular function regulating essential processes for cell survival, proliferation and dissemination. In this project, we uncovered a novel oncogenic regulatory branch of regulation of E2F1 that links lysosomal function with glycolysis, mTORC1 activation and lysosome-dependent exocytosis. Cells harboring these adaptations are potentially progressing better in the tumor context. By modulating v-ATPase activity, lysosome motility and mTORC1 translocation and being able to use glucose catabolism in favor of these events, E2F1 appear to exert a broad regulation of cancer cell functions promoting their malignant transformation. Although this study uncovers a new scenario of growth and metastatic abilities coordination by lysosomes, it also let several open questions that would be worth to address in the near future.

Figure 67. Model of E2F1-induced cellular processes. E2F1 induces the expression of the PFKFB3 isoenzyme, which in turn activates PFK-1 and increases glycolytic flux. This activation positively regulates mTORC1 likely through its recruitment to the lysosomal surface. In parallel, E2F1 induces the peripheral localization of lysosomes and their fusion with the plasma membrane. Together with a higher expression of proton transporters, this process induces the acidification of the extracellular space facilitating cell motility and invasion.

126 DISCUSSION Lysosomes prevent defective mitosis and mediate mitotic proteins degradation

Intracellular signaling, metabolic adaptations or cell death evasion are common alterations found in cancer cells. While implication of lysosome biology in cancer is still controversial, lysosomotropic drug’s efficacy has been proved in several malignancies either alone or in combination and indeed some current clinical trials are based on lysosome acidification blockers (Chude and Amaravadi, 2017). Chloroquine and derivatives, for instance, induce tumor shrinkage in a broad spectrum of cancers (Jiang and Mizushima, 2014). Cancer is a complex disease, accounting for a variety of tumor types, heterogeneity in aggressiveness and multiple mechanisms of resistance to therapy. Nonetheless, all cancer cells rely on uncontrolled cell division to sustain tumor growth. In this context, several cancer therapies intend to hamper cell division thus aiming to slow-down tumor cell proliferation (Diogo et al., 2017; Mross et al., 2016). The second part of this PhD project aims to elucidate the implication of lysosomes in mitotic progression, based on the idea that lysosome inhibition could potentially affect cell division and induce genomic instability. Better understanding of lysosomes implication during cell division could be of importance to clarify their role during tumorigenesis and to find novel targets for cancer therapy.

Our results demonstrate that lysosomal activity is required for correct mitotic progression as both lysosome acidification blockade and anterograde transport impairment slow down cell division, which is associated to malfunctioning of SAC, the checkpoint ensuring correct chromosome segregation. In agreement, our live imaging studies suggest that metaphase-to-anaphase timing is determinant for the mitotic delay promoted by lysosome impairment. Indeed, we show that lysosomotropic drugs as well as trafficking impairment induce mitotic errors such as chromosomal bridges and lagging chromosomes, common mechanisms of CIN (Bakhoum and Cantley, 2018). CIN is considered a hallmark of cancer but its specific role in tumor progression is still under debate. CIN can be considered a tumor strength, due to the induction of mutations thus facilitating the acquisition of novel functions andthe tumor diversification, and oppositely it can be considered a tumor vulnerability, as cells with defective chromosome segregation can eventually trigger apoptotic cell death (Bakhoum and Cantley, 2018; Simonetti et al., 2018). This double-edge sword role in cancer progression is also attributed to the autophagic pathway, which ends- up with the autophagosome-lysosome fusion and consequent lysosome-dependent degradation (Kimura et al., 2013; Levy et al., 2017; Singh et al., 2018). As a tumor suppressor, autophagy can get rid of oncogenic protein substrates, toxic unfolded

DISCUSSION 127 proteins and damaged organelles. Alternatively, as a tumor promoter in established cancers, autophagy is used as a source of energy under unfavorable environmental conditions and to generate and recycle metabolites used for biomass production. These evidences support the idea that autophagy role in cancer is context-dependent (White, 2012). Here we show that autophagic flux is not only active, but also necessary for correct mitotic progression. In this regard, recent research reported a link between defective autophagy and increased CIN (Vessoni et al., 2013) and in yeast is has been shown that autophagy is necessary for mitotic progression under starvation conditions (Matsui et al., 2013).

Our data support that lysosomal acidification and trafficking impairment affect lysosomal function. Indeed, anterograde trafficking impairment induces a perinuclear clustering of lysosomes and increase in size indicating that fusion events can still be active. Inhibition of lysosome acidification also affects lysosomal trafficking as vesicle volume increase, while fusion is preserved (Mauvezin et al., 2015). In this regard, physical forces needed to sustain lysosomal trafficking might not be sufficient to track them. In the context of cell division, we found only one feature attributable to the lysosomotropic drug and not to the genetic depletion of KIF5B and is the ability of Concanamycin A to elicit DNA damage response (DDR), which likely affects consecutive replication cycles. Further assays, such as genetic depletion of v-ATPase subunits, are needed to characterize whether this is due to the acidification defect or whether this effect is attributable to side-effects of the drug itself. Anyway, this dual effect of Concanamycin A on both mitotic delay and DNA damage could potentially be therapeutically exploited as, dependent on the source and extent of the DNA damage, it can result in the execution of cell death pathways.

Study of mitosis is challenging, due to the velocity and architectural changes characteristics of this cell cycle phase. Aiming to escape these complications, most studies of mitosis in human cells have taken advantage of synchronization protocols to increase the proportion of dividing cells. However, drugs used for cell synchronization, such as Nocodazole, do not specifically target mitosis but essential cellular components such as the microtubular network, thus potentially altering other basic cellular functions. In this study we propose the use of toroidal nucleus as a novel read-out to score for mitotic impairment in cells in interphase. Currently, the only marker of defective mitosis detectable in non-mitotic cells is micronuclei frequency. Micronuclei form when mitotic errors produce lagging chromosomes as a consequence of impaired chromosome segregation (Crasta et al., 2012). It has been described, however, that micronuclei can be reabsorbed during the next cell division

128 DISCUSSION or destroyed by autophagy (Rello-Varona et al., 2012), therefore complementary tools are needed to study mitosis impairment in interphase cells. Here we show that toroidal nucleus forms upon mitosis and that it linearly correlates with mitotic errors occurrence. Noteworthy, toroidal nucleus detection is a new convenient tool to detect mitotic errors as it is easily detectable in DAPI stained cells using epifluorescence microscopy. We give some insights into the fate of cells harboring toroidal nucleus and describe that these cells can either reenter cell cycle or undergo cell death and that they are more susceptible to successive mitotic errors and chromosomal instability.

However, as we showed, this phenotype cannot be detected in all the cell lines tested. Thus the use of toroidal nucleus as a screening tool might not be universal and need previous confirmation in the cell line of study. The reason of this variability is still an unresolved question, as it might be due to cell type specificities and/or their own genetic background (i.e. susceptibility to genomic instability) or maybe due to the insensibility to the stimulus. To finally verify the applicability of toroidal nucleus, some complementary experiments are needed directly impairing chromosome segregation in the different cell lines and analyzing their nuclear morphology. We propose that both read-outs (micronuclei and toroidal nucleus) can be complementary scored for initial testing of drug or siRNA library screenings related to cell division defects.

During mitosis, dramatic cellular rearrangement occurs to achieve proper chromosome segregation and productive sharing of intracellular organelles to future generations. Many membranous compartments undergo a massive spatiotemporal rearrangement throughout mitosis. For instance, ER and Golgi apparatus are fragmented during cell division and taken apart from the mitotic spindle a feature that is shared by cytoplasmic vesicles, as the spindle poles require an organelle- free path for chromosomal separation by microtubule-based forces (Jongsma et al., 2015). However, both the functionality of endosomes, autophagosomes and lysosomes during mitosis is still under debate. For mitosis coordination, scheduled degradation of proteins is crucial to link chromosome alignment and segregation. While the role of proteasome degradation has been extensively studied through the regulation of the APC/C complex, the implication of lysosomes in cell division is still a matter of debate. Indeed, some studies claim that autophagy is inhibited during cell division (Eskelinen et al., 2002; Furuya et al., 2010). Of note, as cells undergo profound structural rearrangements, the inhibition of autophagy has been speculated to act as a protective mechanism against unintended loss of organelles and chromosomes when those are vulnerable due to the nuclear envelope break down. However, lysosomes have been involved in cyclin A2 degradation during mitosis

DISCUSSION 129 (Loukil et al., 2014) as well as midbody ring degradation through autophagy during cytokinesis (Pohl and Jentsch, 2009). In addition, two recent publications declare that autophagy and mitophagy are active during cell division (Li et al., 2016; Liu et al., 2009). Here, we describe that robust autophagic flux as well as lysosomal-dependent degradation persist during mitotic progression. Moreover, we observe an increase in size of LAMP2-positive vesicles indicating that fusion events occur in early mitosis and normal size is not recovered until late telophase. These results suggest that, in contrast to Golgi or ER, lysosomes play an active role during cell division and that mitosis does not only rely on proteasome-dependent degradation. The finding that Atg5 KO MEFs present higher percentage of cells with toroidal nucleus compared to their counterparts also corroborates that autophagy is active and necessary in natural-occurring mitosis even in non-transformed cells. We identified more than 300 proteins being up-regulated or only present upon Concanamycin A treatment from G2 phase to G1 phase by mass spectrometry. Interestingly, one of the major accumulated protein was p62, the autophagic adaptor for ubiquitinated cargos. p62 accumulation confirms that autophagic flux is active during cell division and points out a potential alternative route for mitotic factors degradation essential for correct mitotic progression.

In the present work, we focus on the study of proteins involved in mitotic progression either as structural components of the mitotic machinery or key regulatory factors that control signaling networks. Indeed, cohesin complex emerges from this analysis as a candidate for the effects of lysosome inhibition in mitosis. Two negative regulators of the cohesin complex: Cohesin associated factor B (PDS5) and Wing-apart like protein (WAPL) are accumulated upon lysosome acidification blockade. Interestingly, both proteins are implicated in sister chromatids cohesion in mitotic entry. Furthermore, we found a LIR motif in the sequence of WAPL suggesting the ability of this protein to directly interact with LC3 (data not shown). This finding suggests that WAPL could serve as a bridge between cohesin complex and the autophagic machinery during cohesin degradation at the onset of anaphase. Studies performed in PDS5 and WAPL-deficient cells showed a direct link between cohesin defects and chromosomal segregation defects, aneuploidies and even cancer progression (Haarhuis et al., 2013b; Misulovin et al., 2018; Oikawa et al., 2004). The master regulator of mitotic progression PLK1 was also found up-regulated in Concanamycin A-treated cells. PLK1 is involved in several steps of cell division from mitotic entry by phosphorylating Cdc25c and activating the Cdk1/cyclin B complex to mitotic exit by the direct phosphorylation of APC/C subunits. Furthermore, PLK1 has been

130 DISCUSSION also described to control centrosome duplication and maturation from S phase to prophase. Interestingly, PLK1 modulation directly impacts on chromosomal instability and in accordance its role in cancer progression is controversial (Cárcer et al., 2018; Smith et al., 1997). Misregulation of PLK1 could explain, at least in part, some of the mitotic errors occurring during lysosome inhibition such as multipolar spindles or the mitotic delay. Another up-regulated protein families found correspond to charged multivesicular body proteins (CHMPs): CHMP5, CHMP2B and CHMP1A, which are known to be part of the ESCRT-III complex. Interestingly, modifications in ESCRT- III levels have been related to multipolar spindles, NE reformation after telophase and abscission mediation during cytokinesis (Hurley and Hanson, 2010; Vietri et al., 2015). Thus, our study proposes that lysosomal function during cell division might be not restricted to a single regulatory road but instead lysosomes might be implicated in multiple mitotic phases for the faithful course of mitosis.

All our candidates are potential lysosomal targeted proteins that should have been degraded during the mitotic process or might be the consequence of inappropriate degradation of their up-stream regulators. Fractionation experiments will be needed to discriminate proteins that are located either in the lysosome or in the cytosol upon Concanamycin A treatment. While the molecular mechanisms controlled by lysosomes during mitosis are still obscure, our data clearly reveals that lysosomes are active during cell division and that some mitotic proteins are degraded by this organelle, which impacts on mitosis progression (Figure 68). Notably, defects in centrosome replication, sister chromatids cohesion or chromosome attachment to microtubules have been shown to induce chromosome instability (Bakhoum and Cantley, 2018). Given promising results of lysosomotropic drugs for cancer treatment, our data about their effects in cell division offers a comprehensive regulatory mechanism for effective antitumoral therapy.

DISCUSSION 131 Figure 68. Lysosomes as gatekeepers of cellular homeostasis and cell cycle progression in the context of cancer. Five cellular functions can be attributed to lysosomes in tumorigenesis. By their fusion with the plasma membrane, lysosomes undergoing exocytosis, enable the extrusion of protons and acid proteases facilitating cell motility. Lysosomes are also signaling hubs used by cells to sense nutritional status and regulate mTORC1 and AMPK, thus dictating the switch between anabolism and catabolism. As the end-point of autophagy, lysosomes are essential for cellular material recycling. Lysosomes can also eventually trigger cell death pathways upon lysosomal membrane permeabilization. Notably, during mitosis, lysosomes also exert specifically their degradative function ensuring chromosome segregation fidelity, thus contributing to genomic integrity.

132 DISCUSSION Integrative map of lysosomal function in cancer

The implication of lysosomes in tumorigenesis is still unresolved, but there are emerging evidences linking this organelle to malignancy. In this PhD thesis, the lysosomal function has been evaluated in two different contexts related to cancer progression. We have uncovered a novel link between glycolytic metabolism and mTORC1 regulation at the lysosomal surface under oncogenic stimulation by E2F1 transcription factor. Here, lysosome serves as a scaffold platform for metabolic signaling pathways dictating the cell response to glucose scarcity. Thus, lysosome intracellular positioning becomes a crucial factor for the nutrient sensing and response to extracellular signaling. The ability of the lysosome to integrate the nutritional status of the cell, turns it into a perfect candidate for the modulation of master regulators of cell anabolism and catabolism such as mTORC1 or AMPK. While we focus here on the specific role of E2F1 oncogenic signaling, we propose that the molecular mechanisms involving PFK-1/PFKFB3 regulation of mTORC1 can also take part in other oncogenic contexts as a way to sense glucose levels and regulate mTOR kinase activity Figure( 69). Moreover, we briefly focus on the ability of lysosomes to promote cell motility through lysosome-dependent exocytosis. We demonstrate that lysosome acidification is essential for this process to facilitate cell dissemination, thus potentially promoting metastasisFigure ( 69). In nutrient-rich conditions, mTORC1 inhibits autophagy thus reducing lysosomal degradation. Autophagy is essential to maintain cellular homeostasis and therefore survival, but its rolein oncogenesis is context-dependent and still controversial. A similar argumentation can be developed for the role of chromosome instability during cell division upon lysosome inhibition. Here, we demonstrate the activity and function of lysosomes during cell division and identify several novel lysosome protein substrates that could be crucial to maintain genomic integrity (Figure 69). Cellular metabolism regulation and cell cycle progression are in several ways interdependent. Cell cycle critically depends on nutrients availability which directly impacts on the G1/S transition and their restriction can stop cell proliferation. Conversely, cell cycle regulators govern cellular metabolic networks to ensure the production of all the components needed to support cell division. In this work we present an overview of lysosome functions across cell cycle by (1) defining the role of E2F1 on mTORC1 regulation and lysosomal intracellular trafficking and (2) describing a robust lysosome degradation during cell division. In this regard, lysosome functions would act as a gatekeeper of cellular homeostasis and cell cycle progression.

DISCUSSION 133 Figure 69. Lysosomes as gatekeepers of cellular homeostasis and cell cycle progression in the context of cancer. Five cellular functions can be attributed to lysosomes in tumorigenesis. By their fusion with the plasma membrane, lysosomes undergoing exocytosis, enable the extrusion of protons and acid proteases facilitating cell motility. Lysosomes are also signaling hubs used by cells to sense nutritional status and regulate mTORC1 and AMPK, thus dictating the switch between anabolism and catabolism. As the end-point of autophagy, lysosomes are essential for cellular material recycling. Lysosomes can also eventually trigger cell death pathways upon lysosomal membrane permeabilization. Notably, during mitosis, lysosomes also exert specifically their degradative function ensuring chromosome segregation fidelity, thus contributing to genomic integrity.

134 DISCUSSION

Die like a star. Immunostaining of U2OS cells actin cytoskelleton CONCLUSIONS

E2F1 regulated mTORC1 and lysosomal trafficking through glycolysis induction

1. Glucose metabolization potentiates E2F1-induced mTORC1 activation independently of energetic imbalance.

2. Glucose is needed for E2F1-induced mTORC1 translocation to the lysosome.

3. E2F1 induces glycolysis independently of mTORC1 activity.

4. E2F1 promotes the alkalinization of the cytosol and regulates the expression of several proton transporters, importantly MCT1.

5. E2F1 induces the expression of PFKFB3 enzyme to mediate mTORC1 activation.

6. Glucose scarcity compromises E2F1-induced lysosome peripheral localization.

7. E2F1 and v-ATPase increase lysosome-dependent exocytosis, which correlates with migration and invasion capacity.

Lysosomes prevent defective mitosis and mediate mitotic proteins degradation

8. Autophagosomes and lysosomes are active in mitotic cells.

9. Lysosome acidification and trafficking impairment delays mitotic progression and promotes mitotic defects.

10. Toroidal nucleus is novel read-out for defective mitosis in interphase cells.

11. Nuclear envelope reformation occurs normally in cells presenting toroidal nucleus.

12. Toroidal nucleus can be detected in multiple cell lines.

13. Impairment of lysosome acidification and trafficking contributes to toroidal nucleus formation in an additive manner.

14. Toroidal nucleus can undergo cell division but correlates with cell death

15. Macroautophagy pathway is required for normal mitotic progression.

16. Lysosome acidification blockade promotes the accumulation of mitotic proteins mostly involved in chromosomal segregation.

CONCLUSIONS 139 Climbing dams. Immunostaining of U2OS cells actin cytoskelleton MATERIALS AND METHODS

Cell culture

U2OS, A549, HeLa, MCF7, HCT116, LoVo, RKO and HEK293T cell lines were obtained from American Type Culture Collection (ATCC). MEFs Atg5 Wt and Atg5 -/- were a gift from Dr. Zorzano’s lab (IRB) and U2OS H2B-GFP cell line was a gift from Dr. Neus Agell (UB). HFF human fibroblasts were obtained from the IDIBELL Biobank.

U2OS ER-E2F1 and U2OS FLAG-RagB ER-E2F1 stable cell lines were previously generated and selected using Geneticin Selective Antibiotic (G418) (Sigma-Aldrich) at 500 μg/mL (U2OS). ER-E2F1 system allows the translocation of E2F1 protein from the cytosol to the nucleus when cells are treated with 400nM OHT (Calbiochem) inducing its transcriptional activity. U2OS Lamin A/C-mRFP and U2OS H2B-GFP LAMP1-RFP were generated in our laboratory by transient transfection and selected by Fluorescence-activated cell sorter (MoFlo Atrios SORTER).

Cells were grown in DMEM high glucose (Gibco) [4mM L-Glutamine, 4.5 g/L glucose, 1mM Pyruvate] supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Sigma Aldrich). Cells were incubated at 37ºC, 5% CO2 and 90-95% of relative humidity. All the chemicals and drugs used in these studies are listed below (Table 3).

Drugs Concentration Use Reference / Brand OHT 400 nM All Calbiochem Concanamycin A 10 nM All Santa Cruz SC-20211A Chloroquine 10 μM >48 hours SIGMA C6628 Monastrol 100 μM All SIGMA M8515 Nocodazole 1 μM All SIGMA M1404 RO3306 9 μM All SIGMA SML0569 NOVARTIS (under MTA BEZ235 50 nM All with S. Kozma) RAD001 5 nM All DELTACLON S1120 2DG 50 mM/ 100 mM 1-6 hours / SeaHorse SIGMA D8375 Oligomycin A 1 μM 6 hours SIGMA 75351 Ionomycin 10 μM All SIGMA I0634 3PO 25 μM All MERCK 525330

Table 3. Chemicals and drugs used.

MATERIALS AND METHODS 143 Protein extraction and Western Blot analysis

Cells were washed twice with ice-cold PBS, scraped and lysed on Lysis Buffer [20 mM Tris-HCl pH 8, 10 mM EDTA, 150 mM NaCl, 1% Triton-X100] supplemented with protease inhibitors cocktail (1/100 Sigma-Aldrich) and phosphatase inhibitors cocktail 2 and 3 (1/100 Sigma-Aldrich). Cell lysates were incubated for 5 minutes in ice and centrifuged at maximum speed for 10 minutes at 4ºC to get rid of membranous components.

Protein concentration was analyzed using Pierce BCA Protein Assay kit (ThermoFisher Scientific) following manufacturer’s instructions.

Equal amounts of protein lysates were resuspended in Laemnli SDS-sample buffer and incubated at 95ºC for 5 minutes. Proteins were separated on SDS-PAGE and transferred to PVDF membranes (Millipore) previously activated with methanol (PanReac AppliChem) with a semi-dry transfer apparatus for 70 minutes at 110 mA (small gels) or 220 mA (precast gels or two small gels). Membranes were blocked with 5% non-fat dry milk (BioRad) in Tris-buffered saline containing 0.1% Tween-20 (TBS-T) for 1 hour at room temperature. Incubation of primary antibodiesTable ( 4) was performed overnight at 4ºC in 5% non-fat dry milk or 3.5% BSA (Sigma-Aldrich) in TBS-T solution. After three washes in TBS-T, membranes were incubated for1 hour at room temperature with secondary horseradish-coupled antibodies Table ( 4) diluted in 5% non-fat milk in TBS-T. Upon incubation, membranes were washed three times with TBS-T and protein detection was performed by using enhanced chemiluminiscence kit (GE Healthcare). Blots were scanned with iBright detection system (Thermo Fisher).

Quantification of band intensities by densitometry was carried out using FIJI Image J software.

Primary antibodies used for WB Antigen Dilution Source Catalog Number A, B, C Aldolase A 1/1000 Abcam ab169544 ATG5 1/1000 Abgent AP1812b ß-Actin 1/10000 Sigma A2228 ß-Actin 1/2000 Cell Signaling 4967 D, E, F FLAG 1/1000 Sigma F7425

144 MATERIALS AND METHODS G, H, I GAPDH (14C10) mAb 1/1000 Cell signaling 2118 J, K, L LC3 (pAb) 1/1000 MBL pm036 M, N, O mTOR 1/1000 Cell Signaling 2983 P, Q, R p62 / SQSTM1 1/1000 MBL M162-3 PFKFB3 1/1000 Proteintech 13763-1-AP Acetyl-CoA Carboxylase 1/1000 Cell Signaling 3661S Phospho S79 Phospho AMPK alpha 1/1000 Cell Signaling 2535S t172 Phospho p70 S6K 1/1000 Cell signaling 9206 (Thr389) Phospho Raptor (S792) 1/1000 Cell Signaling 2083 Phospho S6 1/1000 Cell Signaling 2211 Phospho ULK1 (Ser555) 1/1000 Cell Signaling 5869 Phospho 4EBP (Ser65) 1/1000 Cell Signaling 9451 Raptor 1/1000 Millipore 09-217 Secondary antibodies used for WB Antigen Dilution Source Reference Anti-Mouse Ig HRP 1/5000 Dako P0260 Anti-Rabbit Ig HRP 1/5000 Dako P0399

Table 4. List of primary and secondary antibodies used for Western Blotting. Immunoprecipitation

After treatment, RagB-FLAG ER-E2F1 stable U2OS cells were washed twice with ice-cold PBS. Cross-linking was performed by incubating cells with 1 mM dithiobis(succinimidyl propionate) (DSP) cross-linker reagent (Thermo Scientific) in PBS supplemented with protease and phosphatase inhibitors for 7 minutes at room temperature. 1 M Tris-HCl (pH 7.5) was added 1:10 to quench DSP activity for 5 minutes at room temperature. After crosslinking, cells were washed twice with ice-cold PBS and lysed in ice-cold IP- RIPA buffer [150 mM NaCl, 50 mM HEPES (pH 7.4), 1 mM EDTA, 1% NP-40, 1% sodium deoxycholate, 0.1 % SDS] supplemented with 2x protease and phosphatase inhibitors cocktails. Collected samples were centrifuged at 13000 rpm for 5 minutes at 4 ºC and soluble fraction was subjected to BCA Pierce protein quantification. 1000 µg of

MATERIALS AND METHODS 145 each lysate were separated for immunoprecipitation in a final volume of 900 µl. For the immunoprecipitation, 100 µl of a 20% slurry of anti-FLAG M2 beads was added to each lysate and incubated by rotation overnight at 4 ºC. Immunoprecipitates were washed three times with high salt RIPA [500 mM NaCl, 50 mM HEPES (pH 7.4), 1 mM EDTA, 1% NP-40, 1% sodium deoxycholate, 0.1 % SDS] and once with IP-RIPA. Immunoprecipitated proteins were denatured by the addition of 2X sample buffer followed by boiling for 5 minutes, resolved by 4% - 20% Criterion TGX Gel (BIO-RAD) electrophoresis and analyzed by immunoblotting. 20 µg of the total lysate were loaded as input to control.

Transfection procedures

Plasmid transfections

DNA transfection was performed following manufacturer’s instructions using Lipofectamine 2000 (Life Technologies) in 1:5 opti-MEM: DMEM medium. All the plasmids used in these studies are listed below (Table 5).

Plasmid Reference / Gift from LAMP1-GFP Addgene #34831 LAMP1-mRFP-FLAG Addgene #34611 KIF5A-GFP Dr. Bonifacino (NIH, USA) PFKFB3 (uPFK-2) Dr. Bartrons (UB) FLAG-RagB Dr. Thomas (IDIBELL) TIGAR Dr. Bartrons (UB) Lamin A/C-mRFP Addgene #55068

Table 5. List of plasmids used.

Silencing RNA (siRNA) transfections siRNA transfections were performed following manufacturer’s instructions in Opti- MEM medium (Life Technologies) using Lipofectamine RNA-iMAX (Life Technologies). Unless otherwise indicated, transfections were performed for 48 hours. siRNA sequences and concentrations used in these studies are listed below Table( 6).

146 MATERIALS AND METHODS Gene Sequence [dT][dT] (5' - 3') [siRNA] Brand / Gift from ALDO A CCAACAGCCUUGCCUGUCAAGGAAA 20 nM SIGMA ALDO B CCAAAGCCCUGGAAACUAA 20 nM SIGMA Atg5 CATCTGAGCTACCCGGATA 50 nM gift from Dr. Muñoz (IDIBELL) CENPE GGAAUUAAAGGCUAAAAGA 20 nM SIGMA GAPDH CGGGAAGCUCACUGGCAUG 50 nM SIGMA KIF1A GGAAACAGAGAAGAUCAUA 20 nM SIGMA KIF2C GCAAGCAACAGGUGCAAGU 20 nM SIGMA KIF3A GGUGUUCGAGCUAUUCCUG 20 nM SIGMA KIF5B CGGCGACAAGUACAUCGCCAAGUUU 20 nM SIGMA KIF11 GCUACUCUGAUGAAUGCAU 20 nM SIGMA p62/SQSTM1 GCAUUGAAGUUGAUAUCGAU 50 nM gift from Dr. Muñoz (IDIBELL) PFKFB3 GCUGUGAAGCAGUACAGCUCCUAC 20 nM SIGMA PFKL CGAGAACAACUGGAACAUUUA 20 nM SIGMA PFKM CCUCCAGAAAGCAGGUAAGAUC 20 nM SIGMA PFKP AGGAACGGCCAGAUCGAUA 20 nM SIGMA SKIP CUUCUGAACUGGACCGAUU 20 nM SIGMA SLC16A1/MCT1 GCUCCGUAUUGUUUGAAACAU 50 nM SIGMA Gene Target Sequences Source Catalog Number ARL8B GAUAGAAGCUUCCCGAAAU Dharmacon J-020294-09-0005 ATP6V0C GGCACAGCCAAGAGCGGUA Dharmacon L-017620-01-0005 GCUCUGUGUAUGCGGAUGA CCCGACUAUUCGUGGGCAU CCAGCUAUCUAUAACCUUA E2F1 * Not available Santa Cruz sc-29297

Table 6. List of siRNA used.

Immunofluorescence (IF)

Cells were grown as monolayers on coverslips and subjected to the indicated conditions. Cells were fixed with 4% paraformaldehyde for 10 minutes atroom temperature. After 5 minutes washing with PBS 1x, cells were permeabilized using 0.1% Triton X-100 in PBS 1X for 10 minutes on a shaker incubator at room temperature. Next, cells were blocked with 1% BSA/0.01% Triton X-100 in PBS 10 mM Glycine for 30 minutes on a shaker at room temperature. Cells were incubated with primary antibodies for one hour at room temperature or overnight at 4ºC. Following a series of PBS 1x washes, cells were incubated with the secondary antibody for 45 minutes at room temperature. After two washing steps of 5 minutes with PBS 1x, coverslips were

MATERIALS AND METHODS 147 mounted using Vectashield Mounting Solution containing DAPI.

For detailed analysis and colocalization studies, images were taken with the Leica spectral confocal microscope TCS SP5 using a 63x N.A 1.4 objective and LAS AF software. Fluorophores were excited with Argon 488 for 488 nm, DPSS 561 for 555 nm, Diode 640 for 647 nm and Diode 405 nm laser for 405 nm. For toroidal nucleus and mitotic errors quantification, images were acquired using Nikon Epifluorescence microscope using a 40x dry objective. Image analysis was performed using FIJI Image J software (NIH USA). All the antibodies used in these studies are reported below (Table 7).

Primary antibodies used for IF Antigen Dilution Source Catalog Number A - F Fibrillarin 1/75 Santa Cruz sc-25397 G - L LAMP-2 (CD107b) 1/300 BD Biosciences 555803 LAMIN B1 1/100 ON Abcam ab16048 LC3 (pAb) 1/100 MBL International pm036 M - O mTOR 1/100 Cell Signaling 2983 NPC 1/100 ON Abcam Ab-24609 P - U p21 1/100 Abcam ab109520 p62/SQSTM1 (mAb) 1/200 MBL International M162-3 Phalloidin - Rhodamine 1/500 Cytoskeleton Phdh1 UBF 1/100 ON Santa Cruz sc-13125 Secondary antibodies used for WB Antigen Dilution Source Reference Anti-Mouse (goat), IgG (H+L), Alexa Fluor 488 1/400 Invitrogen A11001 Anti-Mouse (goat), IgG (H+L), Alexa Fluor 647 1/400 Invitrogen A21235 Anti-Rabbit (goat), IgG, (H+L), Rhodamine 1/400 Invitrogen R6394 Red-X Anti-Rabbit (goat), IgG (H+L), Alexa Fluor 546 1/400 Thermo A11010 Anti-Rabbit (goat), IgG (H+L), Alexa Fluor 647 1/400 Thermo A21245 Anti-Rabbit (donkey), IgG, (H+L), Alexa Fluor 1/400 Thermo A21206 488 Anti-Rabbit (donkey), IgG, (H+L), Alexa Fluor 1/400 Thermo A31572 555

Table 7. List of primary and secondary antibodies used for immunofluorescence.

148 MATERIALS AND METHODS Image processing and analysis

3D reconstitution

Stacks of images acquired with confocal microscope were processed with IMARIS Software (Bitplane) for 3D image reconstruction and to generate in silico animations. Lysosome localization

Briefly, individual cells were selected using actin cytoskeleton as a reference and the distance between lysosome center and nucleus center was calculated using the pixels coordinates. Data from each cell was introduced to Prism4 software to process and obtain the corresponding boxplot. Colocalization coefficient

Images were acquired using fixed settings and analyzed using ZEN software. Thresholds for the red and green channels were adjusted and maintained during all the analysis. Percentage of red-green overlapping pixels over total red pixels was calculated. Vesicle acidification assay: Magic Red

U2OS cells were seeded onto glass cover-slips and subjected to the indicated conditions. Live cells were washed once with PBS and 100 µL of Magic Red was added to cover the cell layer. Cells were incubated for 10 minutes at room temperature. Two PBS washes were performed and one last wash in H2O. Coverslips were mounted using Vectashield Mounting Solution containing DAPI. Images were acquired at the moment. Cytosolic pH measurement

Pre-treated U2OS cells were resuspended in PBS and incubated with 5 μM SNARF- AM for 30 minutes at 37 ºC. Samples were centrifuged, and the stained pellet was resuspended in PBS for flow cytometric analysis. pH extrapolation was performed using a standard curve from cells incubated at different pH, the protocol used was as follows. After incubation with SNARF-AM and centrifugation, cells were resuspended in high potassium buffer [135 mM KH2PO4, 29 mM NaCl] supplemented with 10 μM Nigericin. Cell suspension was incubated for 20 minutes to equilibrate the intracellular pH and analyzed by flow cytometry. Bandpass filters were centered at 580 nm and 640 nm for ratio calculation.

MATERIALS AND METHODS 149 Live-cell time-lapse videos

Migration assay

Cells were grown onto poly(dimethylsiloxane) (PDMS) devices specifically designed for this purpose. After indicated treatments, PDMS device was extracted and time- lapse imaging was performed using Zeiss Apotome microscope. Images were taken every 20 minutes during a total time of 24 hours using the 10x objective.

Mitotic delay analysis

U2OS H2B-GFP cells were grown onto glass bottom 8-well slides (IBIDI). After indicated treatments, medium was replaced by FluoroBright medium (Gibco) and cells were placed on the Leica SP5 microscope. Images were taken every 6 minutes and 33 seconds for a total time of 24 hours using the 63x glycerol objective.

Cell death analysis

U2OS H2B-GFP cells were grown onto glass bottom 24-well slides (MatTek). Before the live-imaging, medium was replaced by FluoroBright medium (Gibco) supplemented with Cell Event (Thermo Fisher) (5 μL of Cell Event solution per ml) and treated as indicated. Cells were placed on the Leica SP5 microscope. Images were taken every 9 minutes and 29 seconds for a total time of 24 hours using a 40x glycerol objective. Apoptotic cells were detected using Diode 640 nm laser for 650 nm signal detection (Figure 70).

Figure 70. Validation of Cell Event dye for cell death detection. U2OS H2B-GFP cells were incubated with Cell Event and Staurosporine. At Time 0 (T0), green nuclei correspond to living cells and at Final Time (Tf) all death cells are stained in red (Cell Event).

150 MATERIALS AND METHODS Toroidal nucleus and nuclear lamina reformation

U2OS cells stably expressing H2B-GFP and Lamin A/C-mCherry were grown onto bottom-glass 8 chambers slides. After addition of FluoroBrite with or without 10 nM Concanamycin A, mitotic cell division was analyzed in each condition with Carl Zeiss LSM880 confocal microscope. Images were taken every 5 minutes for 16 hours using the 63x glycerol objective. Image analysis and processing was performed with FIJI ImageJ software (NIH).

Real time quantitative PCR (qPCR)

Total RNA was extracted using TRIzol (Invitrogen) following manufacturer’s instructions. RNA was resuspended in DEPC-treated water and quantified using NanoDrop 1000 Spectrophotometer (Thermo Scientific). 5 μg of total RNA were subjected to DNAase I treatment (Roche) using the following protocol [37 ºC for 30 minutes – 65ºC for 10 minutes – 4ºC ∞]. 1μg of total DNAase I-treated RNA was reverse-transcribed using the M-MLV Reverse Transcriptase (Invitrogen) following the indicated protocol [65 ºC for 5 minutes – 25 ºC for 10 minutes – 37ºC for 50 minutes – 85ºC for 5 minutes – 4ºC ∞]. The resulting cDNA samples were amplified using LightCycler 96 SYBR Green I Master Mix (Roche Molecular Systems). The PCR protocol used was: 95ºC for 5 minutes followed by 45 cycles [95 ºC for 15 seconds; 58 ºC for 15 seconds; 72 ºC for 20 seconds]. Data analysis was performed using LightCycler 96 Software (Roche Molecular Systems). Gene expression was normalized to the endogenous control β-actin. The sequences of the PCR primers used are reported in the table below (Table 8).

Gene Forward Reverse Efficiency ACTIN AATGTGGCCGAGGACTTTGATTGC AGGATGGCAAGGGACTTCCTGTAA 1,88 ALDO A TCACCGCATCGTGGCACCTG GAAGCGCCGGTTCTCCTCGG 2,16 CA2 TGTGCAGCAACCTGATGGACTG ATCCAAGGATTCAGGAAGGAGG 1,62 MCT1/SLC6A1 CATGTATGGTGGAGGTCCTATC CAGAAAGAAGCTGCAATCAAGCC 2,4 MCT4 TTTTGCTGCTGGGCAACTTCTTCTG TCACGTTGTCTCGAAGCATGGGTTT 2,57 PFKFB3 TGTTCAACGTGGGGAGTAT GCAGCTAAGGCACATTGCTT 2,1 PFKL GGCATTTATGTGGGTGCCAAAGTC CAGTTGGCCTGCTTGATGTTCTCA 2,03 PFKM GCCATCAGCCTTTGACAGA CTCCAAAAGTGCCATCACTG 2,03 PFKP GCATGGGTATCTACGTGGGG CTCTGCGATGTTTGAGCCTC 2,13

Table 8. List of primers used in real time quantitative PCR.

MATERIALS AND METHODS 151 Transmission electron microscopy (TEM)

U2OS cells with or without Concanamycin A (10 nM) treatment for 24 hours were fixed with Glutaraldehyde 2.5% in Sodium Cacodylate Trihydrate 0.1 M pH 7.2 for 2 hours at 4 ºC. After three washes with Sodium Cacodylate Trihydrate 0.1 M of 15 minutes each, samples were incubated for 2 hours with Osmium tetroxide 1% in Sodium Cacodylate Trihydrate 0.1 M at room temperature. Samples were washed three times for 15 minutes with Sodium Cacodylate Trihydrate 0.1 M. Then, samples were processed through dehydration with Ethanol 30% to 100% gradually. Samples were then embedded into Resin EPOXY. After sample orientation, polymerization occurred at 60 ºC for 48 hours. Samples were cut using ultramicrotome EM UC6 Leica, first in semithin of 250 nm and then ultrathin of 70 nm. Image acquisition was performed with TEM microscope JEOL JEM-1011. Images were analyzed and processed with FIJI ImageJ software (NIH). Cell cycle analysis

Cells were trypsinized, counted and placed on ice. 5•105 cells were transferred into a Falcon tube and centrifuged at 1200 rpm at 4ºC for 5 minutes. Cell pellet was washed with ice-cold PBS, fixed with 70% ethanol and placed at -20ºC for 24 hours. Cells were then washed with ice-cold FACS buffer (BSA 0.1%, EDTA 5 mM in PBS] and centrifuged for 5 minutes at 1000 rpm at 4ºC. Supernatant was discarded and the cellular pellet was resuspended in propidium iodide (PI) staining solution [PBS, 0.1% NP40, 20 µg/ mL RNAse A (Invitrogen), 40 µg/mL PI (Invitrogen)]. The cellular suspension was transferred to a 5 mL tube with cell strainer cap (Corning) and maintained at room temperature protected from light for at least 15 minutes. Samples were analyzed using FACS Canto System.

Cell synchronization at G2

Cells were seeded the day before starting the synchronization in DMEM complete medium supplemented with FBS. Thymidine was added at a final concentration of 2 μM for 24 hours to synchronize cells at late G1. After PBS washing, cells were released to S phase by adding normal DMEM supplemented with 10% FBS for 2 hours. Finally, RO3306 (Cdk1 inhibitor) was added to cell culture medium at a final concentration of 9 μM for 12 hours to arrest cells at late G2 (Figure 71). Experiments with synchronized cells were performed by releasing cells in normal DMEM supplemented with 10% FBS for the indicated times.

152 MATERIALS AND METHODS Figure 71. Schematic representation of cell synchronization at G2. Lysosome-dependent exocytosis measurement β-hexosaminidase release assay

Cells were grown onto 6-well plates for this assay. Pre-treated cells were incubated with 10 μM Ionomycin/4 mM CaCl2 in 500 μL of HBSS for 10 minutes at 37ºC. Reaction was stopped by placing cells on ice. Incubation media was collected in Eppendorf tubes stored in ice. 500 μL of 1% IGEPAL in dH2O was added to cell monolayer and cells were collected by scrapping into Eppendorf tubes stored in ice. Samples were centrifuged for 5 minutes at 11000 g before performing the enzymatic assay. Total extract lysate was diluted 1:5 in dH2O for protein quantification. Enzymatic assay was performed in a 96-well plate by adding 100 μL of supernatant and 100 μL of the diluted cell lysate of each sample in each well in triplicates. 100 μL of 6 mM 4-methyl-umbelliferyl-N-acetyl-β-d-glucosaminide (Glycosynth) freshly prepared in substrate buffer [40 mM sodium citrate, 88 mM Na2HPO4 pH 4.5] was added. Enzymatic reaction was incubated for 15 minutes at 37ºC and stop reaction solution was added. Fluorescence was analyzed in a plate-spectrofluorimeter at excitation 365 nm/emission 450 nm. Protein concentration was determined using micro BCA protein reagent kit (Pierce laboratories) following manufacturer’s instructions. The following calculations were performed to determine β-hexosaminidase activity:

MATERIALS AND METHODS 153 Ultra-Performance Liquid Chromatography (UPLC) for adenine nucleotides measurements

U2OS cells were treated and harvested in triplicate. Cells were rinsed twice in cold PBS and collected in 400 µL of perchloric acid 2 M followed by incubation for 15 minutes at 4ºC. Supernatants were used for nucleotide measurement and pellets for protein determination. 100 µL of Bicine 1 M and 100 µL K2CO3 4 M were added to the supernatants. Nucleotide solution was subjected to vortex and centrifuged. The supernatant was frozen until UPLC analysis.

For UPLC determination, samples were thawed and filtered using Nylon membrane 4 mm 0.45 µm Syring Filter (National Scientific). Samples were analyzed by UPLC on a Acquity UPLC system with a Kinetex 2.6 µm C18000 Column (Phenomenex). Analyzes were performed using 15 µL sample injection volume at 35 ºC at a flow rate of 0.5 mL/min. Buffer A: MeOH 30%; Buffer B KH2PO4 0,05 M, 4 mM tetrabutyl ammonium hydrogen (TBA) pH=6 with 50% KOH. The UPLC program was: [0-5 min, 0% A; 5 min, 30% A; 10 min, 40% A; 19 min, 100% A; 22 min, 100% A; 23 min 0% A; 30 min, stop]. UV detection was set at 260 nM wavelength. Peaks were identified by retention times and compared to the peak spectrum of ATP/ADP/AMP standards. Area under the curve was analyzed using Empower Software (Waters) and values were normalized by protein amount.

Glycolytic rate measurement

U2OS cells were seeded at optimal confluence of 700000 cells per well. Measurements from pre-treated U2OS cells were performed using a XF24-Extracellular Flux Analyzer (Seahorse Bioscience), following manufacturer standard Glycolysis Stress Test. Reagent’s concentrations were optimized for U2OS cells.

XF-Extracellular Flux Analyzer injection ports were used to inject assay reagents. Extracellular acidification rate (ECAR) was measured in unbuffered DMEM supplemented with 2 mM Glutamine. U2OS cells were starved of glucose for 1 hour prior to experiment followed by the addition of 25 mM D-glucose (glycolysis induction), Oligomycin A (mitochondrial inhibition) and 2-DG (glycolysis inhibition). The protein amount was determined by BCA Pierce and used to normalize the results.

154 MATERIALS AND METHODS Mass Spectrometry

Sample preparation and digestion

Three control and three Concanamycin A treated samples were processed for protein extraction in RIPA (2% SDS) buffer. Then, samples were quantified using the BCA Pierce method and 50 μg of every sample condition was digested using a FASP (Filter-Aided Sample Preparation) approach. Briefly, proteins were reduced with dithiothreitol 10 mM (60 minutes, 32 ºC) and alkylated with iodoacetamide 20 mM (30 minutes at 25 ºC in the dark). Then, the samples were loaded onto an Amicon Ultra (filter 10 KDa, 0.5 mL, Millipore, Billerica, MA, USA) device to remove interfering agents with 2 rounds of centrifugations/washes with 100 mM ammonium bicarbonate buffer (13600 g; 25 minutes at room temperature). Digestion was carried out in two steps: first, samples were digested (1:50 w sample/w enzyme) with Lys-C (Wako, Richmond, VA, USA) in 6 M urea buffer for 3 hours at 35 ºC, second, the samples were diluted 10-fold with 100 mM ammonium bicarbonate buffer and digested with modified porcine trypsin (Promega-Gold, Madison, WI, USA) (1/25w sample/w enzyme) for 16 hours at 37ºC. The resulting peptide mixture was recovered by centrifuging the filter. Then, the filter was washed twice with 300 μL of 50 mM ammonium bicarbonate and once with 200 μL of 20% acetonitrile/50 mM ammonium bicarbonate (13600g for 25 min at room temperature). All the fractions were pooled, and the final peptide mixture was acidified with formic acid. Finally, the final volume of the acidified peptide solution was reduced on a SpeedVac vacuum system (Thermo Fisher Scientific, Barcelona, Spain), and the peptide solution was desalinated with a C18 spin column (Thermo Fisher Scientific) following the indications of the supplier.

Sample acquisition

Samples were analyzed in a Proxeon 1000 liquid chromatographer coupled to an Orbitrap Fusion Lumos (Thermo Fisher Scientific) mass spectrometer. Samples were re-suspended in 0.5% formic acid in water, and 2 μL (1 μg/ μL) were injected for LC-MSMS analysis. Peptides were trapped on an NTCC-360/75-3-123 LC column and separated using a C18 reverse phase LC column-Easy Spray (Thermo Fisher Scientific). The gradient used for the elution of the peptides was 1% to 35% in 90 minutes followed by a gradient from 35% to 85% in 10 minutes with 250 nL/min flow rate. Eluted peptides were subjected to electrospray ionization in an emitter needle

MATERIALS AND METHODS 155 (PicoTipTM, New Objective, Scientific Instrument Services, Ringoes, NJ, USA) with an applied voltage of 2000 V. Peptide masses (m/z 300-1700) were analyzed in data- dependent mode where a full scan MS was acquired on the Orbitrap with a resolution of 60000 FWHM at 400 m/z. Up to the 10 most abundant peptides (minimum intensity of 500 counts) were selected from each MS scan and then fragmented using CID (collision-induced dissociation) in the linear ion trap using helium as collision gas with 38% normalized collision energy. The scan time settings were: full MS at 250 ms and MSn at 120 ms. Generated .raw data files were collected with Thermo Xcalibur (v.2.2) (Termo Fisher Scientific).

Data Analysis

MaxQuant 1.6.1.0 Software (Department for Proteomics and Signal Transduction, Max-Planck Institute for Biochemistry) was used to search the .raw data obtained in the MS analyses against a SwissProt/Uniprot human database with Andromeda Search engine (1.5.6.0). A target and decoy database was used to assess the false discovery rate (FDR). Trypsin was chosen as enzyme and a maximum of two misscleavages were allowed. Carbamidomethylation (C) was set as a fixed modification, whereas oxidation (M) and acetylation (N-terminal) were used as variable modifications. Searches were performed using a peptide tolerance of 7 ppm and a product ion tolerance of 0.5 Da. Resulting data files were filtered for FDR <1%. Statistical analysis was performed in Perseus 1.6.2.1 (Department for Proteomics and Signal Transduction, Max-Planck Institute for Biochemistry). Statistical analysis

Data was analyzed by Excel or GraphPad Prism4 software. Results are presented as Mean ± S.E.M., for the indicated n independent experiments. Experimental data-sets were compared by different statistic tests:

Two-sampled, two-tailed Student’s t-test: two experimental conditions sharing normal distribution and variance.

One-way ANOVA test: more than 2 conditions, data with normal distribution and equal variance. Multiple comparisons corrected using Bonferroni test.

Kruskal-Wallis test: more than 2 conditions and data without assuming normal distribution. Multiple comparisons corrected using Dunn’s test.

156 MATERIALS AND METHODS MATERIALS AND METHODS 157 Ghost. Immunostaining of U2OS cells actin cytoskelleton REFERENCES

Alber, F., Dokudovskaya, S., Veenhoff, L.M., Zhang, W., Kipper, J., Devos, D., Suprapto, A., Karni- Schmidt, O., Williams, R., Chait, B.T., et al. (2007). The molecular architecture of the nuclear pore complex. Nature 450, 695–701.

Alzamora, R., Thali, R.F., Gong, F., Smolak, C., Li, H., Baty, C.J., Bertrand, C.A., Auchli, Y., Brunisholz, R.A., Neumann, D., et al. (2010). PKA Regulates Vacuolar H+-ATPase Localization and Activity via Direct Phosphorylation of the A Subunit in Kidney Cells. J. Biol. Chem. 285, 24676–24685.

Andrejewski, N., Punnonen, E.L., Guhde, G., Tanaka, Y., Lüllmann-Rauch, R., Hartmann, D., von Figura, K., and Saftig, P. (1999). Normal lysosomal morphology and function in LAMP-1- deficient mice. J. Biol. Chem. 274, 12692–12701.

Andrés, V., Carreras, J., and Cussó, R. (1990). Regulation of muscle phosphofructokinase by physiological concentrations of bisphosphorylated hexoses: Effect of alkalinization. Biochem. Biophys. Res. Commun. 172, 328–334.

Araujo, A.R., Gelens, L., Sheriff, R.S.M., and Santos, S.D.M. (2016). Positive Feedback Keeps Duration of Mitosis Temporally Insulated from Upstream Cell-Cycle Events. Mol. Cell 64, 362– 375.

Axe, E.L., Walker, S.A., Manifava, M., Chandra, P., Roderick, H.L., Habermann, A., Griffiths, G., and Ktistakis, N.T. (2008). Autophagosome formation from membrane compartments enriched in phosphatidylinositol 3-phosphate and dynamically connected to the endoplasmic reticulum. J. Cell Biol. 182, 685–701.

Bakhoum, S.F., and Cantley, L.C. (2018). The Multifaceted Role of Chromosomal Instability in Cancer and Its Microenvironment. Cell 174, 1347–1360.

Ballabio A. and Gieselmann V. (2009). Lysosomal disorders: from storage to cellular damage. Biochem Biophys Acta 1793(4), 684-696.

Balachandran, R.S., Heighington, C.S., Starostina, N.G., Anderson, J.W., Owen, D.L., Vasudevan, S., and Kipreos, E.T. (2016). The ubiquitin ligase CRL2 ZYG11 targets cyclin B1 for degradation in a conserved pathway that facilitates mitotic slippage. J. Cell Biol. 215, 151–166.

Balgi, A.D., Diering, G.H., Donohue, E., Lam, K.K.Y., Fonseca, B.D., Zimmerman, C., Numata, M., and Roberge, M. (2011). Regulation of mTORC1 Signaling by pH. PLoS ONE 6.

Bar-Peled, L., Schweitzer, L.D., Zoncu, R., and Sabatini, D.M. (2012). Ragulator Is a GEF for the Rag GTPases that Signal Amino Acid Levels to mTORC1. Cell 150, 1196–1208.

Bartrons, R., Simon-Molas, H., Rodríguez-García, A., Castaño, E., Navarro-Sabaté, À., Manzano, A., and Martinez-Outschoorn, U.E. (2018). Fructose 2,6-Bisphosphate in Cancer Cell Metabolism. Front. Oncol. 8.

REFERENCES 161 Bensaad, K., Tsuruta, A., Selak, M.A., Vidal, M.N.C., Nakano, K., Bartrons, R., Gottlieb, E., and Vousden, K.H. (2006). TIGAR, a p53-inducible regulator of glycolysis and apoptosis. Cell 126, 107–120.

Ben-Sahra, I., Howell, J.J., Asara, J.M., and Manning, B.D. (2013). Stimulation of de Novo Pyrimidine Synthesis by Growth Signaling Through mTOR and S6K1. Science 339, 1323–1328.

Bertoli, C., Skotheim, J.M., and de Bruin, R.A.M. (2013). Control of cell cycle transcription during G1 and S phases. Nat. Rev. Mol. Cell Biol. 14, 518–528.

Bieda, M., Xu, X., Singer, M.A., Green, R., and Farnham, P.J. (2006). Unbiased location analysis of E2F1-binding sites suggests a widespread role for E2F1 in the human genome. Genome Res. 16, 595–605.

Blanchet, E., Annicotte, J.-S., Lagarrigue, S., Aguilar, V., Clapé, C., Chavey, C., Fritz, V., Casas, F., Apparailly, F., Auwerx, J., et al. (2011). E2F transcription factor-1 regulates oxidative metabolism. Nat. Cell Biol. 13, 1146–1152.

Bloom, J., and Cross, F.R. (2007). Multiple levels of cyclin specificity in cell-cycle control. Nat. Rev. Mol. Cell Biol. 8, 149–160.

Bowman, E.J., Graham, L.A., Stevens, T.H., and Bowman, B.J. (2004). The bafilomycin/ concanamycin binding site in subunit c of the V-ATPases from Neurospora crassa and Saccharomyces cerevisiae. J. Biol. Chem. 279, 33131–33138.

Boya, P., and Kroemer, G. (2008). Lysosomal membrane permeabilization in cell death. Oncogene 27, 6434–6451.

Brozzi, A., Urbanelli, L., Luc Germain, P., Magini, A., and Emiliani, C. (2013). hLGDB: a database of human lysosomal genes and their regulation. Database 2013.

Burkhardt, J.K., Echeverri, C.J., Nilsson, T., and Vallee, R.B. (1997). Overexpression of the dynamitin (p50) subunit of the dynactin complex disrupts dynein-dependent maintenance of membrane organelle distribution. J. Cell Biol. 139, 469–484.

Cantalupo, G., Alifano, P., Roberti, V., Bruni, C.B., and Bucci, C. (2001). Rab-interacting lysosomal protein (RILP): the Rab7 effector required for transport to lysosomes. EMBO J. 20, 683–693.

Capecci, J., and Forgac, M. (2013). The Function of Vacuolar ATPase (V-ATPase) a Subunit Isoforms in Invasiveness of MCF10a and MCF10CA1a Human Breast Cancer Cells.

Cárcer, G. de, Venkateswaran, S.V., Salgueiro, L., Bakkali, A.E., Somogyi, K., Rowald, K., Montañés, P., Sanclemente, M., Escobar, B., Martino, A. de, et al. (2018). Plk1 overexpression induces chromosomal instability and suppresses tumor development. Nat. Commun. 9, 3012.

162 REFERENCES Carroll, B., and Dunlop, E.A. (2017). The lysosome: a crucial hub for AMPK and mTORC1 signalling. Biochem. J. 474, 1453–1466.

Caviston, J.P., and Holzbaur, E.L.F. (2006). Microtubule motors at the intersection of trafficking and transport. Trends Cell Biol. 16, 530–537.

Cerami, E., Gao, J., Dogrusoz, U., Gross, B.E., Sumer, S.O., Aksoy, B.A., Jacobsen, A., Byrne, C.J., Heuer, M.L., Larsson, E., et al. (2012). The cBio Cancer Genomics Portal: An Open Platform for Exploring Multidimensional Cancer Genomics Data. Cancer Discov. 2, 401–404.

Chan, C.-Y., and Parra, K.J. (2014). Yeast Phosphofructokinase-1 Subunit Pfk2p Is Necessary for pH Homeostasis and Glucose-dependent Vacuolar ATPase Reassembly. J. Biol. Chem. 289, 19448–19457.

Chen, H.-Z., Tsai, S.-Y., and Leone, G. (2009). Emerging roles of E2Fs in cancer: an exit from cell cycle control. Nat. Rev. Cancer 9, 785–797.

Cheung, P.C., Salt, I.P., Davies, S.P., Hardie, D.G., and Carling, D. (2000). Characterization of AMP-activated protein kinase gamma-subunit isoforms and their role in AMP binding. Biochem. J. 346 Pt 3, 659–669.

Chude, C.I., and Amaravadi, R.K. (2017). Targeting Autophagy in Cancer: Update on Clinical Trials and Novel Inhibitors. Int. J. Mol. Sci. 18.

Conner, E.A., Lemmer, E.R., Omori, M., Wirth, P.J., Factor, V.M., and Thorgeirsson, S.S. (2000). Dual functions of E2F-1 in a transgenic mouse model of liver carcinogenesis. Oncogene 19, 5054–5062.

Corthésy-Theulaz, I., Pauloin, A., and Pfeffer, S.R. (1992). Cytoplasmic dynein participates in the centrosomal localization of the Golgi complex. J. Cell Biol. 118, 1333–1345.

Cotter, K., Stransky, L., McGuire, C., and Forgac, M. (2015a). Recent Insights into the Structure, Regulation and Function of the V-ATPases. Trends Biochem. Sci. 40, 611–622.

Cotter, K., Capecci, J., Sennoune, S., Huss, M., Maier, M., Martinez-Zaguilan, R., and Forgac, M. (2015b). Activity of Plasma Membrane V-ATPases Is Critical for the Invasion of MDA-MB231 Breast Cancer Cells. J. Biol. Chem. 290, 3680–3692.

Couto, N.F., Pedersane, D., Rezende, L., Dias, P.P., Corbani, T.L., Bentini, L.C., Oliveira, A.C.S., Kelles, L.F., Castro-Gomes, T., and Andrade, L.O. (2017). LAMP-2 absence interferes with plasma membrane repair and decreases T. cruzi host cell invasion. PLoS Negl. Trop. Dis. 11, e0005657.

REFERENCES 163 Crasta, K., Ganem, N.J., Dagher, R., Lantermann, A.B., Ivanova, E.V., Pan, Y., Nezi, L., Protopopov, A., Chowdhury, D., and Pellman, D. (2012). DNA breaks and chromosome pulverization from errors in mitosis. Nature 482, 53–58.

Deacon, S.W., Serpinskaya, A.S., Vaughan, P.S., Fanarraga, M.L., Vernos, I., Vaughan, K.T., and Gelfand, V.I. (2003). Dynactin is required for bidirectional organelle transport. J Cell Biol 160, 297–301.

Dechant, R., Binda, M., Lee, S.S., Pelet, S., Winderickx, J., and Peter, M. (2010). Cytosolic pH is a second messenger for glucose and regulates the PKA pathway through V-ATPase. EMBO J. 29, 2515–2526.

Denechaud, P.-D., Fajas, L., and Giralt, A. (2017). E2F1, a Novel Regulator of Metabolism. Front. Endocrinol. 8.

Di Fiore, B., Wurzenberger, C., Davey, N.E., and Pines, J. (2016). The Mitotic Checkpoint Complex Requires an Evolutionary Conserved Cassette to Bind and Inhibit Active APC/C. Mol. Cell 64, 1144–1153.

Dikic, I., and Elazar, Z. (2018). Mechanism and medical implications of mammalian autophagy. Nat. Rev. Mol. Cell Biol. 19, 349–364.

Diogo, V., Teixeira, J., Silva, P.M.A., and Bousbaa, H. (2017). Spindle Assembly Checkpoint as a Potential Target in Colorectal Cancer: Current Status and Future Perspectives. Clin. Colorectal Cancer 16, 1–8.

Dou, Y., Wu, H., Li, H., Qin, S., Wang, Y., Li, J., Lou, H., Chen, Z., Li, X., Luo, Q., et al. (2012). Microglial migration mediated by ATP-induced ATP release from lysosomes. Cell Res. 22, 1022–1033.

Dupriez, V.J., Darville, M.I., Antoine, I.V., Gegonne, A., Ghysdael, J., and Rousseau, G.G. (1993). Characterization of a hepatoma mRNA transcribed from a third promoter of a 6-phosphofructo- 2-kinase/fructose-2,6-bisphosphatase-encoding gene and controlled by ets oncogene-related products. Proc. Natl. Acad. Sci. 90, 8224–8228.

Duran, J., Obach, M., Navarro‐Sabate, A., Manzano, A., Gómez, M., Rosa, J.L., Ventura, F., Perales, J.C., and Bartrons, R. (2009). Pfkfb3 is transcriptionally upregulated in diabetic mouse liver through proliferative signals. FEBS J. 276, 4555–4568. de Duve, C. (2005). The lysosome turns fifty. Nat. Cell Biol. 7, 847–849.

Egan, D.F., Shackelford, D.B., Mihaylova, M.M., Gelino, S.R., Kohnz, R.A., Mair, W., Vasquez, D.S., Joshi, A., Gwinn, D.M., Taylor, R., et al. (2011). Phosphorylation of ULK1 (hATG1) by AMP- activated protein kinase connects energy sensing to mitophagy. Science 331, 456–461.

164 REFERENCES Eskelinen, E.-L. (2006). Roles of LAMP-1 and LAMP-2 in lysosome biogenesis and autophagy. Mol. Aspects Med. 27, 495–502.

Eskelinen, E.-L., Prescott, A.R., Cooper, J., Brachmann, S.M., Wang, L., Tang, X., Backer, J.M., and Lucocq, J.M. (2002). Inhibition of autophagy in mitotic animal cells. Traffic Cph. Den. 3, 878–893.

Eskelinen, E.-L., Schmidt, C.K., Neu, S., Willenborg, M., Fuertes, G., Salvador, N., Tanaka, Y., Lüllmann-Rauch, R., Hartmann, D., Heeren, J., et al. (2004). Disturbed cholesterol traffic but normal proteolytic function in LAMP-1/LAMP-2 double-deficient fibroblasts. Mol. Biol. Cell 15, 3132–3145.

Eymin, B., Gazzeri, S., Brambilla, C., and Brambilla, E. (2001). Distinct pattern of E2F1 expression in human lung tumours: E2F1 is upregulated in small cell lung carcinoma. Oncogene 20, 1678– 1687.

Fabregat, A., Jupe, S., Matthews, L., Sidiropoulos, K., Gillespie, M., Garapati, P., Haw, R., Jassal, B., Korninger, F., May, B., et al. (2018). The Reactome Pathway Knowledgebase. Nucleic Acids Res. 46, D649–D655.

Fennelly, C., and Amaravadi, R.K. (2017). Lysosomal Biology in Cancer. Methods Mol. Biol. Clifton NJ 1594, 293–308.

Ferenz, N.P., Gable, A., and Wadsworth, P. (2010). Mitotic Functions of Kinesin-5. Semin. Cell Dev. Biol. 21, 255–259.

Fernández de Mattos, S., Lam, E.W.-F., and Tauler, A. (2002). An E2F-binding site mediates the activation of the proliferative isoform of 6-phosphofructo-2-kinase/fructose-2,6- bisphosphatase by phosphatidylinositol 3-kinase. Biochem. J. 368, 283–291.

Forgac, M. (2007). Vacuolar ATPases: rotary proton pumps in physiology and pathophysiology. Nat. Rev. Mol. Cell Biol. 8, 917–929.

Fred Dice, J. (1990). Peptide sequences that target cytosolic proteins for lysosomal proteolysis. Trends Biochem. Sci. 15, 305–309.

Furuya, T., Kim, M., Lipinski, M., Li, J., Kim, D., Lu, T., Shen, Y., Rameh, L., Yankner, B., Tsai, L.-H., et al. (2010). Negative regulation of Vps34 by Cdk mediated phosphorylation. Mol. Cell 38, 500–511.

Gadde, S., and Heald, R. (2004). Mechanisms and molecules of the mitotic spindle. Curr. Biol. CB 14, R797-805.

REFERENCES 165 Gao, J., Aksoy, B.A., Dogrusoz, U., Dresdner, G., Gross, B., Sumer, S.O., Sun, Y., Jacobsen, A., Sinha, R., Larsson, E., et al. (2013). Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci. Signal. 6, pl1.

Gatenby, R.A., and Gillies, R.J. (2004). Why do cancers have high aerobic glycolysis? Nat. Rev. Cancer 4, 891–899.

Gillies, R.J., and Gatenby, R.A. (2007). Hypoxia and adaptive landscapes in the evolution of carcinogenesis. Cancer Metastasis Rev. 26, 311–317.

Goncalves, M.D., and Cantley, L.C. (2018). A Glycolysis Outsider Steps into the Cancer Spotlight. Cell Metab. 28, 3–4.

Gonzalez, A., Valeiras, M., Sidransky, E., and Tayebi, N. (2014). Lysosomal integral membrane protein-2: a new player in lysosome-related pathology. Mol. Genet. Metab. 111, 84–91.

Gorgoulis, V.G., Zacharatos, P., Mariatos, G., Kotsinas, A., Bouda, M., Kletsas, D., Asimacopoulos, P.J., Agnantis, N., Kittas, C., and Papavassiliou, A.G. (2002). Transcription factor E2F-1 acts as a growth-promoting factor and is associated with adverse prognosis in non-small cell lung carcinomas. J. Pathol. 198, 142–156.

Goshima, G., and Vale, R.D. (2003). The roles of microtubule-based motor proteins in mitosis. J. Cell Biol. 162, 1003–1016.

Guardia, C.M., Farías, G.G., Jia, R., Pu, J., and Bonifacino, J.S. (2016). BORC Functions Upstream of Kinesins 1 and 3 to Coordinate Regional Movement of Lysosomes along Different Microtubule Tracks. Cell Rep. 17, 1950–1961.

Gwinn, D.M., Shackelford, D.B., Egan, D.F., Mihaylova, M.M., Mery, A., Vasquez, D.S., Turk, B.E., and Shaw, R.J. (2008). AMPK Phosphorylation of Raptor Mediates a Metabolic Checkpoint. Mol. Cell 30, 214–226.

Haarhuis, J.H.I., Elbatsh, A.M.O., van den Broek, B., Camps, D., Erkan, H., Jalink, K., Medema, R.H., and Rowland, B.D. (2013a). WAPL-mediated removal of cohesin protects against segregation errors and aneuploidy. Curr. Biol. CB 23, 2071–2077.

Haarhuis, J.H.I., Elbatsh, A.M.O., van den Broek, B., Camps, D., Erkan, H., Jalink, K., Medema, R.H., and Rowland, B.D. (2013b). WAPL-mediated removal of cohesin protects against segregation errors and aneuploidy. Curr. Biol. CB 23, 2071–2077.

Hamasaki, M., Furuta, N., Matsuda, A., Nezu, A., Yamamoto, A., Fujita, N., Oomori, H., Noda, T., Haraguchi, T., Hiraoka, Y., et al. (2013). Autophagosomes form at ER-mitochondria contact sites. Nature 495, 389–393.

166 REFERENCES Hanahan, D., and Weinberg, R.A. (2000). The Hallmarks of Cancer. Cell 100, 57–70.

Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of Cancer: The Next Generation. Cell 144, 646–674.

Hastie, E.L., and Sherwood, D.R. (2016). A new front in cell invasion: The invadopodial membrane. Eur. J. Cell Biol. 95, 441–448.

Hayden, J.H., Bowser, S.S., and Rieder, C.L. (1990). Kinetochores capture astral microtubules during chromosome attachment to the mitotic spindle: direct visualization in live newt lung cells. J. Cell Biol. 111, 1039–1045.

He, J., Chao, W.C.H., Zhang, Z., Yang, J., Cronin, N., and Barford, D. (2013). Insights into Degron Recognition by APC/C Coactivators from the Structure of an Acm1-Cdh1 Complex. Mol. Cell 50, 649–660.

Heald, R., and McKeon, F. (1990). Mutations of phosphorylation sites in lamin A that prevent nuclear lamina disassembly in mitosis. Cell 61, 579–589.

Hinton, A., Sennoune, S.R., Bond, S., Fang, M., Reuveni, M., Sahagian, G.G., Jay, D., Martinez- Zaguilan, R., and Forgac, M. (2009). Function of a Subunit Isoforms of the V-ATPase in pH Homeostasis and in Vitro Invasion of MDA-MB231 Human Breast Cancer Cells. J. Biol. Chem. 284, 16400–16408.

Hirokawa, N., and Noda, Y. (2008). Intracellular transport and kinesin superfamily proteins, KIFs: structure, function, and dynamics. Physiol. Rev. 88, 1089–1118.

Hirokawa, N., Noda, Y., Tanaka, Y., and Niwa, S. (2009). Kinesin superfamily motor proteins and intracellular transport. Nat. Rev. Mol. Cell Biol. 10, 682–696.

Höök, P., and Vallee, R.B. (2006). The dynein family at a glance. J. Cell Sci. 119, 4369–4371.

Huang, D.W., Sherman, B.T., and Lempicki, R.A. (2009a). Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 37, 1–13.

Huang, D.W., Sherman, B.T., and Lempicki, R.A. (2009b). Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 4, 44–57.

Hurley, J.H., and Hanson, P.I. (2010). Membrane budding and scission by the ESCRT machinery: it’s all in the neck. Nat. Rev. Mol. Cell Biol. 11, 556–566.

Huss, M., Ingenhorst, G., König, S., Gaßel, M., Dröse, S., Zeeck, A., Altendorf, K., and Wieczorek, H. (2002). Concanamycin A, the Specific Inhibitor of V-ATPases, Binds to the Vo Subunit c. J. Biol. Chem. 277, 40544–40548.

REFERENCES 167 Ichimura, Y., Kirisako, T., Takao, T., Satomi, Y., Shimonishi, Y., Ishihara, N., Mizushima, N., Tanida, I., Kominami, E., Ohsumi, M., et al. (2000). A ubiquitin-like system mediates protein lipidation. Nature 408, 488–492.

Inoki, K., Li, Y., Xu, T., and Guan, K.-L. (2003). Rheb GTPase is a direct target of TSC2 GAP activity and regulates mTOR signaling. Genes Dev. 17, 1829–1834.

Ishida, Y., Nayak, S., Mindell, J.A., and Grabe, M. (2013). A model of lysosomal pH regulation. J. Gen. Physiol. 141, 705–720.

Ishikawa, T. (2012). Structural biology of cytoplasmic and axonemal dyneins. J. Struct. Biol. 179, 229–234.

Jaiswal, J.K., Andrews, N.W., and Simon, S.M. (2002). Membrane proximal lysosomes are the major vesicles responsible for calcium-dependent exocytosis in nonsecretory cells. J Cell Biol 159, 625–635.

Jiang, P., and Mizushima, N. (2014). Autophagy and human diseases. Cell Res. 24, 69–79.

Joaquin, M., Salvado, C., Bellosillo, B., Lange, A.J., Gil, J., and Tauler, A. (1997a). Effect of Growth Factors on the Expression of 6-Phosphofructo2-kinase/Fructose-2,6-bisphosphatase in Rat-1 Fibroblasts. J. Biol. Chem. 272, 2846–2851.

Joaquin, M., Rosa, J.L., Bartrons, R., and Tauler, A. (1997b). Expression of the F-type 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase mRNA during liver regeneration. Biochim. Biophys. Acta BBA - Gen. Subj. 1334, 256–260.

Johansson, M., Rocha, N., Zwart, W., Jordens, I., Janssen, L., Kuijl, C., Olkkonen, V.M., and Neefjes, J. (2007). Activation of endosomal dynein motors by stepwise assembly of Rab7-RILP- p150Glued, ORP1L, and the receptor betalll spectrin. J. Cell Biol. 176, 459–471.

Johnson, D.G., Cress, W.D., Jakoi, L., and Nevins, J.R. (1994). Oncogenic capacity of the E2F1 gene. Proc. Natl. Acad. Sci. U. S. A. 91, 12823–12827.

Jongsma, M.L.M., Berlin, I., and Neefjes, J. (2015). On the move: organelle dynamics during mitosis. Trends Cell Biol. 25, 112–124.

Jordens, I., Fernandez-Borja, M., Marsman, M., Dusseljee, S., Janssen, L., Calafat, J., Janssen, H., Wubbolts, R., and Neefjes, J. (2001). The Rab7 effector protein RILP controls lysosomal transport by inducing the recruitment of dynein-dynactin motors. Curr. Biol. CB 11, 1680– 1685.

Jovic, M., Sharma, M., Rahajeng, J., and Caplan, S. (2010). The early endosome: a busy sorting station for proteins at the crossroads. Histol. Histopathol. 25, 99–112.

168 REFERENCES Kalender, A., Selvaraj, A., Kim, S.Y., Gulati, P., Brûlé, S., Viollet, B., Kemp, B.E., Bardeesy, N., Dennis, P., Schlager, J.J., et al. (2010). Metformin, Independent of AMPK, Inhibits mTORC1 in a Rag GTPase-Dependent Manner. Cell Metab. 11, 390–401.

Kallunki, T., Olsen, O.D., and Jäättelä, M. (2013). Cancer-associated lysosomal changes: friends or foes? Oncogene 32, 1995–2004.

Karakaidos, P., Taraviras, S., Vassiliou, L.V., Zacharatos, P., Kastrinakis, N.G., Kougiou, D., Kouloukoussa, M., Nishitani, H., Papavassiliou, A.G., Lygerou, Z., et al. (2004). Overexpression of the replication licensing regulators hCdt1 and hCdc6 characterizes a subset of non-small- cell lung carcinomas: synergistic effect with mutant p53 on tumor growth and chromosomal instability--evidence of E2F-1 transcriptional control over hCdt1. Am. J. Pathol. 165, 1351– 1365.

Karanasios, E., Stapleton, E., Manifava, M., Kaizuka, T., Mizushima, N., Walker, S.A., and Ktistakis, N.T. (2013). Dynamic association of the ULK1 complex with omegasomes during autophagy induction. J. Cell Sci. 126, 5224–5238.

Kato, Y., Ozawa, S., Miyamoto, C., Maehata, Y., Suzuki, A., Maeda, T., and Baba, Y. (2013). Acidic extracellular microenvironment and cancer. Cancer Cell Int. 13, 89.

Kaushik, S., and Cuervo, A.M. (2012). Chaperone-mediated autophagy: a unique way to enter the lysosome world. Trends Cell Biol. 22, 407–417.

Khodjakov, A., Copenagle, L., Gordon, M.B., Compton, D.A., and Kapoor, T.M. (2003). Minus- end capture of preformed kinetochore fibers contributes to spindle morphogenesis. J. Cell Biol. 160, 671–683.

Kim, J., Kundu, M., Viollet, B., and Guan, K.-L. (2011). AMPK and mTOR regulate autophagy through direct phosphorylation of Ulk1. Nat. Cell Biol. 13, 132–141.

Kimura, T., Takabatake, Y., Takahashi, A., and Isaka, Y. (2013). Chloroquine in Cancer Therapy: A Double-Edged Sword of Autophagy. Cancer Res. 73, 3–7.

Korolchuk, V.I., Saiki, S., Lichtenberg, M., Siddiqi, F.H., Roberts, E.A., Imarisio, S., Jahreiss, L., Sarkar, S., Futter, M., Menzies, F.M., et al. (2011). Lysosomal positioning coordinates cellular nutrient responses. Nat. Cell Biol. 13, 453–460.

Koyama-Honda, I., Itakura, E., Fujiwara, T.K., and Mizushima, N. (2013). Temporal analysis of recruitment of mammalian ATG proteins to the autophagosome formation site. Autophagy 9, 1491–1499.

Kraya, A.A., Piao, S., Xu, X., Zhang, G., Herlyn, M., Gimotty, P., Levine, B., Amaravadi, R.K., and Speicher, D.W. (2015). Identification of secreted proteins that reflect autophagy dynamics within tumor cells. Autophagy 11, 60–74.

REFERENCES 169 Kroemer, G., and Jäättelä, M. (2005). Lysosomes and autophagy in cell death control. Nat. Rev. Cancer 5, 886–897.

Kroemer, G., Mariño, G., and Levine, B. (2010). Autophagy and the integrated stress response. Mol. Cell 40, 280–293.

Lagadic-Gossmann, D., Huc, L., and Lecureur, V. (2004). Alterations of intracellular pH homeostasis in apoptosis: origins and roles. Cell Death Differ. 11, 953–961.

Lee, B.S. (2012). Regulation of V-ATPase expression in mammalian cells. Curr. Protein Pept. Sci. 13, 107–116.

Lee, M.N., Ha, S.H., Kim, J., Koh, A., Lee, C.S., Kim, J.H., Jeon, H., Kim, D.-H., Suh, P.-G., and Ryu, S.H. (2009). Glycolytic Flux Signals to mTOR through Glyceraldehyde-3-Phosphate Dehydrogenase-Mediated Regulation of Rheb. Mol. Cell. Biol. 29, 3991–4001. de Leeuw, R., Gruenbaum, Y., and Medalia, O. (2018). Nuclear Lamins: Thin Filaments with Major Functions. Trends Cell Biol. 28, 34–45.

Levy, J.M.M., Towers, C.G., and Thorburn, A. (2017). Targeting autophagy in cancer. Nat. Rev. Cancer 17, 528–542.

Li, Z., and Zhang, X. (2016). Autophagy in mitotic animal cells. Sci. Bull. 61, 105–107.

Li, W., Li, J., and Bao, J. (2012). Microautophagy: lesser-known self-eating. Cell. Mol. Life Sci. CMLS 69, 1125–1136.

Li, Z., Ji, X., Wang, D., Liu, J., and Zhang, X. (2016). Autophagic flux is highly active in early mitosis and differentially regulated throughout the cell cycle. Oncotarget 7, 39705–39718.

Lin, S.-C., and Hardie, D.G. (2018). AMPK: Sensing Glucose as well as Cellular Energy Status. Cell Metab. 27, 299–313.

Linder, M.I., Köhler, M., Boersema, P., Weberruss, M., Wandke, C., Marino, J., Ashiono, C., Picotti, P., Antonin, W., and Kutay, U. (2017). Mitotic Disassembly of Nuclear Pore Complexes Involves CDK1- and PLK1-Mediated Phosphorylation of Key Interconnecting Nucleoporins. Dev. Cell 43, 141-156.e7.

Liu, L., Xie, R., Nguyen, S., Ye, M., and McKeehan, W.L. (2009). Robust autophagy/mitophagy persists during mitosis. Cell Cycle Georget. Tex 8, 1616–1620.

Loukil, A., Zonca, M., Rebouissou, C., Baldin, V., Coux, O., Biard-Piechaczyk, M., Blanchard, J.- M., and Peter, M. (2014). High-resolution live-cell imaging reveals novel cyclin A2 degradation foci involving autophagy. J Cell Sci 127, 2145–2150.

170 REFERENCES Lu, L., Ladinsky, M.S., and Kirchhausen, T. (2011). Formation of the postmitotic nuclear envelope from extended ER cisternae precedes nuclear pore assembly. J. Cell Biol. 194, 425– 440.

Lu, M., Holliday, L.S., Zhang, L., Dunn, W.A., and Gluck, S.L. (2001). Interaction between Aldolase and Vacuolar H+-ATPase. Evidence for direct coupling of glycolysis to the ATP- hydrolizing proton pump. J. Biol. Chem. 276, 30407–30413.

Lu, M., Ammar, D., Ives, H., Albrecht, F., and Gluck, S.L. (2007). Physical Interaction between Aldolase and Vacuolar H + -ATPase Is Essential for the Assembly and Activity of the Proton Pump. J. Biol. Chem. 282, 24495–24503.

Lübke, T., Lobel, P., and Sleat, D. (2009). Proteomics of the Lysosome. Biochim. Biophys. Acta 1793, 625–635.

Luzio, J.P., Pryor, P.R., and Bright, N.A. (2007). Lysosomes: fusion and function. Nat. Rev. Mol. Cell Biol. 8, 622–632.

Magistris, P.D., and Antonin, W. (2018). The Dynamic Nature of the Nuclear Envelope. Curr. Biol. 28, R487–R497.

Maiato, H., Rieder, C.L., and Khodjakov, A. (2004). Kinetochore-driven formation of kinetochore fibers contributes to spindle assembly during animal mitosis. J Cell Biol 167, 831–840.

Marshansky, V., Rubinstein, J.L., and Grüber, G. (2014). Eukaryotic V-ATPase: Novel structural findings and functional insights. Biochim. Biophys. Acta BBA - Bioenerg. 1837, 857–879.

Martinez-Zaguilan, R., Lynch, R.M., Martinez, G.M., and Gillies, R.J. (1993). Vacuolar-type H(+)-ATPases are functionally expressed in plasma membranes of human tumor cells. Am. J. Physiol.-Cell Physiol. 265, C1015–C1029.

Mathiassen, S.G., De Zio, D., and Cecconi, F. (2017). Autophagy and the Cell Cycle: A Complex Landscape. Front. Oncol. 7.

Matsui, A., Kamada, Y., and Matsuura, A. (2013). The Role of Autophagy in Genome Stability through Suppression of Abnormal Mitosis under Starvation. PLOS Genet. 9, e1003245.

Matsuyama, S., Llopis, J., Deveraux, Q.L., Tsien, R.Y., and Reed, J.C. (2000). Changes in intramitochondrial and cytosolic pH: early events that modulate caspase activation during apoptosis. Nat. Cell Biol. 2, 318–325.

Mauvezin, C., Nagy, P., Juhász, G., and Neufeld, T.P. (2015). Autophagosome–lysosome fusion is independent of V-ATPase-mediated acidification. Nat. Commun. 6, 7007.

REFERENCES 171 McClelland, S.E. (2017). Role of chromosomal instability in cancer progression. Endocr. Relat. Cancer 24, T23–T31.

McEwan, D.G., Popovic, D., Gubas, A., Terawaki, S., Suzuki, H., Stadel, D., Coxon, F.P., Miranda de Stegmann, D., Bhogaraju, S., Maddi, K., et al. (2015). PLEKHM1 regulates autophagosome- lysosome fusion through HOPS complex and LC3/GABARAP proteins. Mol. Cell 57, 39–54.

McGuire, C.M., and Forgac, M. (2018). Glucose Starvation Increases V-ATPase Assembly and Activity in Mammalian Cells through AMP Kinase and Phosphatidylinositide 3-Kinase/Akt Signaling. J. Biol. Chem. jbc.RA117.001327.

McGuire, C., Cotter, K., Stransky, L., and Forgac, M. (2016). Regulation of V-ATPase assembly and function of V-ATPases in tumor cell invasiveness. Biochim. Biophys. Acta BBA - Bioenerg. 1857, 1213–1218.

Mellman, I. (1996). Endocytosis and molecular sorting. Annu. Rev. Cell Dev. Biol. 12, 575–625.

Menon, S., Dibble, C.C., Talbott, G., Hoxhaj, G., Valvezan, A.J., Takahashi, H., Cantley, L.C., and Manning, B.D. (2014). Spatial Control of the TSC Complex Integrates Insulin and Nutrient Regulation of mTORC1 at the Lysosome. Cell 156, 771–785.

Meo-Evoli, N., Almacellas, E., Massucci, F.A., Gentilella, A., Ambrosio, S., Kozma, S.C., Thomas, G., and Tauler, A. (2015). V-ATPase: a master effector of E2F1-mediated lysosomal trafficking, mTORC1 activation and autophagy. Oncotarget 6, 28057–28070.

Meyer, H.-J., and Rape, M. (2011). Processive Ubiquitin Chain Formation by the Anaphase- Promoting Complex. Semin. Cell Dev. Biol. 22, 544–550.

Mijaljica, D., Prescott, M., and Devenish, R.J. (2011). Microautophagy in mammalian cells: Revisiting a 40-year-old conundrum. Autophagy 7, 673–682.

Miki, H., Setou, M., Kaneshiro, K., and Hirokawa, N. (2001). All kinesin superfamily protein, KIF, genes in mouse and human. Proc. Natl. Acad. Sci. U. S. A. 98, 7004–7011.

Mindell, J.A. (2012). Lysosomal acidification mechanisms. Annu. Rev. Physiol. 74, 69–86.

Misulovin, Z., Pherson, M., Gause, M., and Dorsett, D. (2018). Brca2, Pds5 and Wapl differentially control cohesin chromosome association and function. PLoS Genet. 14, e1007225.

Mizushima, N. (2007). Autophagy: process and function. Genes Dev. 21, 2861–2873.

Moon, J.-S., Hisata, S., Park, M.-A., DeNicola, G.M., Ryter, S.W., Nakahira, K., and Choi, A.M.K. (2015). mTORC1-Induced HK1-Dependent Glycolysis Regulates NLRP3 Inflammasome Activation. Cell Rep. 12, 102–115.

172 REFERENCES Morgan, D.O. (2013). The D Box Meets Its Match. Mol. Cell 50, 609–610.

Mross, K., Richly, H., Frost, A., Scharr, D., Nokay, B., Graeser, R., Lee, C., Hilbert, J., Goeldner, R.- G., Fietz, O., et al. (2016). A phase I study of BI 811283, an Aurora B kinase inhibitor, in patients with advanced solid tumors. Cancer Chemother. Pharmacol. 78, 405–417.

Naegeli, K.M., Hastie, E., Garde, A., Wang, Z., Keeley, D.P., Gordon, K.L., Pani, A.M., Kelley, L.C., Morrissey, M.A., Chi, Q., et al. (2017). Cell Invasion In Vivo via Rapid Exocytosis of a Transient Lysosome-Derived Membrane Domain. Dev. Cell 43, 403-417.e10.

Nascimbeni, A.C., Giordano, F., Dupont, N., Grasso, D., Vaccaro, M.I., Codogno, P., and Morel, E. (2017). ER-plasma membrane contact sites contribute to autophagosome biogenesis by regulation of local PI3P synthesis. EMBO J. 36, 2018–2033.

Nasmyth, K. (2002). Segregating Sister Genomes: The Molecular Biology of Chromosome Separation. Science 297, 559–565.

Nickel, W., and Rabouille, C. (2009). Mechanisms of regulated unconventional protein secretion. Nat. Rev. Mol. Cell Biol. 10, 148–155.

Nishimura, T., Tamura, N., Kono, N., Shimanaka, Y., Arai, H., Yamamoto, H., and Mizushima, N. (2017). Autophagosome formation is initiated at phosphatidylinositol synthase-enriched ER subdomains. EMBO J. 36, 1719–1735.

Nordmann, M., Cabrera, M., Perz, A., Bröcker, C., Ostrowicz, C., Engelbrecht-Vandré, S., and Ungermann, C. (2010). The Mon1-Ccz1 complex is the GEF of the late endosomal Rab7 homolog Ypt7. Curr. Biol. CB 20, 1654–1659.

Novikoff, A.B., Beaufay, H., and de Duve, C. (1956). Electron microscopy of lysosome-rich fractions from rat liver. J. Biophys. Biochem. Cytol. 2, 179–184.

Oikawa, K., Ohbayashi, T., Kiyono, T., Nishi, H., Isaka, K., Umezawa, A., Kuroda, M., and Mukai, K. (2004). Expression of a novel human gene, human wings apart-like (hWAPL), is associated with cervical carcinogenesis and tumor progression. Cancer Res. 64, 3545–3549.

Olmos, Y., Hodgson, L., Mantell, J., Verkade, P., and Carlton, J.G. (2015). ESCRT-III controls nuclear envelope reformation. Nature 522, 236–239.

Olsvik, H.L., Lamark, T., Takagi, K., Larsen, K.B., Evjen, G., Øvervatn, A., Mizushima, T., and Johansen, T. (2015). FYCO1 Contains a C-terminally Extended, LC3A/B-preferring LC3- interacting Region (LIR) Motif Required for Efficient Maturation of Autophagosomes during Basal Autophagy. J. Biol. Chem. 290, 29361–29374.

REFERENCES 173 Onorati, A.V., Dyczynski, M., Ojha, R., and Amaravadi, R.K. (2018). Targeting autophagy in cancer. Cancer 124, 3307–3318.

Pankiv, S., Clausen, T.H., Lamark, T., Brech, A., Bruun, J.-A., Outzen, H., Øvervatn, A., Bjørkøy, G., and Johansen, T. (2007). p62/SQSTM1 binds directly to Atg8/LC3 to facilitate degradation of ubiquitinated protein aggregates by autophagy. J. Biol. Chem. 282, 24131–24145.

Parra, K.J., and Hayek, S.R. (2018). A lysosomal proton pump turns on when glucose runs out. J. Biol. Chem. 293, 9124–9125.

Parra, K.J., and Kane, P.M. (1998). Reversible Association between the V1and V0 Domains of Yeast Vacuolar H+-ATPase Is an Unconventional Glucose-Induced Effect. Mol. Cell. Biol. 18, 7064–7074.

Parry, D.H., and O’Farrell, P.H. (2001). The schedule of destruction of three mitotic cyclins can dictate the timing of events during exit from mitosis. Curr. Biol. CB 11, 671–683.

Persi, E., Duran-Frigola, M., Damaghi, M., Roush, W.R., Aloy, P., Cleveland, J.L., Gillies, R.J., and Ruppin, E. (2018). Systems analysis of intracellular pH vulnerabilities for cancer therapy. Nat. Commun. 9, 2997.

Peters, J.-M. (2006). The anaphase promoting complex/cyclosome: a machine designed to destroy. Nat. Rev. Mol. Cell Biol. 7, 644–656.

Pfarr, C.M., Coue, M., Grissom, P.M., Hays, T.S., Porter, M.E., and McIntosh, J.R. (1990). Cytoplasmic dynein is localized to kinetochores during mitosis. Nature 345, 263–265.

Pfleger, C.M., and Kirschner, M.W. (2000). The KEN box: an APC recognition signal distinct from the D box targeted by Cdh1. Genes Dev. 14, 655–665.

Pierce, A.M., Fisher, S.M., Conti, C.J., and Johnson, D.G. (1998). Deregulated expression of E2F1 induces hyperplasia and cooperates with ras in skin tumor development. Oncogene 16, 1267–1276.

Pierce, A.M., Schneider-Broussard, R., Gimenez-Conti, I.B., Russell, J.L., Conti, C.J., and Johnson, D.G. (1999). E2F1 has both oncogenic and tumor-suppressive properties in a transgenic model. Mol. Cell. Biol. 19, 6408–6414.

Pohl, C., and Jentsch, S. (2009). Midbody ring disposal by autophagy is a post-abscission event of cytokinesis. Nat. Cell Biol. 11, 65–70.

Pu, J., Guardia, C.M., Keren-Kaplan, T., and Bonifacino, J.S. (2016). Mechanisms and functions of lysosome positioning. J. Cell Sci. 129, 4329–4339.

Pu, J., Keren-Kaplan, T., and Bonifacino, J.S. (2017). A Ragulator-BORC interaction controls lysosome positioning in response to amino acid availability. J. Cell Biol. 216, 4183–4197.

174 REFERENCES Rabanal-Ruiz, Y., and Korolchuk, V.I. (2018). mTORC1 and Nutrient Homeostasis: The Central Role of the Lysosome. Int. J. Mol. Sci. 19.

Real, S., Meo-Evoli, N., Espada, L., and Tauler, A. (2011). E2F1 regulates cellular growth by mTORC1 signaling. PloS One 6, e16163.

Reddy, A., Caler, E.V., and Andrews, N.W. (2001). Plasma Membrane Repair Is Mediated by Ca2+-Regulated Exocytosis of Lysosomes. Cell 106, 157–169.

Rello-Varona, S., Lissa, D., Shen, S., Niso-Santano, M., Senovilla, L., Mariño, G., Vitale, I., Jemaá, M., Harper, F., Pierron, G., et al. (2012). Autophagic removal of micronuclei. Cell Cycle Georget. Tex 11, 170–176.

Rieder, C.L., and Alexander, S.P. (1990). Kinetochores are transported poleward along a single astral microtubule during chromosome attachment to the spindle in newt lung cells. J. Cell Biol. 110, 81–95.

Roczniak-Ferguson, A., Petit, C.S., Froehlich, F., Qian, S., Ky, J., Angarola, B., Walther, T.C., and Ferguson, S.M. (2012). The Transcription Factor TFEB Links mTORC1 Signaling to Transcriptional Control of Lysosome Homeostasis. Sci. Signal. 5, ra42.

Rodríguez, A., Webster, P., Ortego, J., and Andrews, N.W. (1997). Lysosomes Behave as Ca2+- regulated Exocytic Vesicles in Fibroblasts and Epithelial Cells. J. Cell Biol. 137, 93–104.

Rosa-Ferreira, C., and Munro, S. (2011). Arl8 and SKIP act together to link lysosomes to kinesin-1. Dev. Cell 21, 1171–1178.

Rosenbloom, B.E., and Weinreb, N.J. (2013). Gaucher disease: a comprehensive review. Crit. Rev. Oncog. 18, 163–175.

Rowland, T.J., Sweet, M.E., Mestroni, L., and Taylor, M.R.G. (2016). Danon disease- dysregulation of autophagy in a multisystem disorder with cardiomyopathy. J. Cell Sci. 129, 2135–2143.

Saftig, P., and Klumperman, J. (2009). Lysosome biogenesis and lysosomal membrane proteins: trafficking meets function. Nat. Rev. Mol. Cell Biol. 10, 623–635.

Sancak, Y., Peterson, T.R., Shaul, Y.D., Lindquist, R.A., Thoreen, C.C., Bar-Peled, L., and Sabatini, D.M. (2008a). The Rag GTPases bind raptor and mediate amino acid signaling to mTORC1. Science 320, 1496–1501.

Sancak, Y., Peterson, T.R., Shaul, Y.D., Lindquist, R.A., Thoreen, C.C., Bar-Peled, L., and Sabatini, D.M. (2008b). The Rag GTPases Bind Raptor and Mediate Amino Acid Signaling to mTORC1. Science 320, 1496–1501.

REFERENCES 175 Schiefermeier, N., Scheffler, J.M., de Araujo, M.E.G., Stasyk, T., Yordanov, T., Ebner, H.L., Offterdinger, M., Munck, S., Hess, M.W., Wickström, S.A., et al. (2014). The late endosomal p14-MP1 (LAMTOR2/3) complex regulates focal adhesion dynamics during cell migration. J. Cell Biol. 205, 525–540.

Schröder, B.A., Wrocklage, C., Hasilik, A., and Saftig, P. (2010). The proteome of lysosomes. Proteomics 10, 4053–4076.

Schwake, M., Schröder, B., and Saftig, P. (2013). Lysosomal Membrane Proteins and Their Central Role in Physiology. Traffic 14, 739–748.

Seibenhener, M.L., Babu, J.R., Geetha, T., Wong, H.C., Krishna, N.R., and Wooten, M.W. (2004). Sequestosome 1/p62 is a polyubiquitin chain binding protein involved in ubiquitin proteasome degradation. Mol. Cell. Biol. 24, 8055–8068.

Settembre, C., Zoncu, R., Medina, D.L., Vetrini, F., Erdin, S., Erdin, S., Huynh, T., Ferron, M., Karsenty, G., Vellard, M.C., et al. (2012). A lysosome-to-nucleus signalling mechanism senses and regulates the lysosome via mTOR and TFEB. EMBO J. 31, 1095–1108.

Settembre, C., Fraldi, A., Medina, D.L., and Ballabio, A. (2013). Signals for the lysosome: a control center for cellular clearance and energy metabolism. Nat. Rev. Mol. Cell Biol. 14, 283– 296.

Sevenich, L., and Joyce, J.A. (2014). Pericellular proteolysis in cancer. Genes Dev. 28, 2331– 2347.

Shen, D., Wang, X., Li, X., Zhang, X., Yao, Z., Dibble, S., Dong, X., Yu, T., Lieberman, A.P., Showalter, H.D., et al. (2012). Lipid storage disorders block lysosomal trafficking by inhibiting a TRP channel and lysosomal calcium release. Nat. Commun. 3, 731.

Shi, L., Pan, H., Liu, Z., Xie, J., and Han, W. (2017). Roles of PFKFB3 in cancer. Signal Transduct. Target. Ther. 2, 17044.

Simonetti, G., Bruno, S., Padella, A., Tenti, E., and Martinelli, G. (2018). Aneuploidy: Cancer strength or vulnerability? Int. J. Cancer.

Singh, S.S., Vats, S., Chia, A.Y.-Q., Tan, T.Z., Deng, S., Ong, M.S., Arfuso, F., Yap, C.T., Goh, B.C., Sethi, G., et al. (2018). Dual role of autophagy in hallmarks of cancer. Oncogene 37, 1142– 1158.

Slobodkin, M.R., and Elazar, Z. (2013). The Atg8 family: multifunctional ubiquitin-like key regulators of autophagy. Essays Biochem. 55, 51–64.

Smith, M.R., Wilson, M.L., Hamanaka, R., Chase, D., Kung, H., Longo, D.L., and Ferris, D.K. (1997). Malignant Transformation of Mammalian Cells Initiated by Constitutive Expression of thePolo-like Kinase1. Biochem. Biophys. Res. Commun. 234, 397–405.

176 REFERENCES Son, S.W., Kim, S.-H., Moon, E.-Y., Kim, D.-H., Pyo, S., and Um, S.H. (2016). Prognostic significance and function of the vacuolar H+-ATPase subunit V1E1 in esophageal squamous cell carcinoma. Oncotarget 7, 49334–49348.

Stein, L.D. (2004). Using the Reactome database. Curr. Protoc. Bioinforma. Chapter 8, Unit 8.7.

Stolz, A., Ernst, A., and Dikic, I. (2014). Cargo recognition and trafficking in selective autophagy. Nat. Cell Biol. 16, 495–501.

Stransky, L.A., and Forgac, M. (2015). Amino Acid Availability Modulates Vacuolar H+-ATPase Assembly. J. Biol. Chem. 290, 27360–27369.

Stuurman, N., Heins, S., and Aebi, U. (1998). Nuclear lamins: their structure, assembly, and interactions. J. Struct. Biol. 122, 42–66.

Stypmann, J., Janssen, P.M.L., Prestle, J., Engelen, M.A., Kögler, H., Lüllmann-Rauch, R., Eckardt, L., von Figura, K., Landgrebe, J., Mleczko, A., et al. (2006). LAMP-2 deficient mice show depressed cardiac contractile function without significant changes in calcium handling. Basic Res. Cardiol. 101, 281–291.

Su, Y., Zhou, A., Al-Lamki, R.S., and Karet, F.E. (2003). The a-subunit of the V-type H+-ATPase interacts with phosphofructokinase-1 in humans. J. Biol. Chem. 278, 20013–20018.

Sullivan, M., and Morgan, D.O. (2007). Finishing mitosis, one step at a time. Nat. Rev. Mol. Cell Biol. 8, 894–903.

Sun, Q., Chen, X., Ma, J., Peng, H., Wang, F., Zha, X., Wang, Y., Jing, Y., Yang, H., Chen, R., et al. (2011). Mammalian target of rapamycin up-regulation of pyruvate kinase isoenzyme type M2 is critical for aerobic glycolysis and tumor growth. Proc. Natl. Acad. Sci. 108, 4129–4134.

Tan, J.M.M., Wong, E.S.P., Kirkpatrick, D.S., Pletnikova, O., Ko, H.S., Tay, S.-P., Ho, M.W.L., Troncoso, J., Gygi, S.P., Lee, M.K., et al. (2008). Lysine 63-linked ubiquitination promotes the formation and autophagic clearance of protein inclusions associated with neurodegenerative diseases. Hum. Mol. Genet. 17, 431–439.

Tanaka, Y., Guhde, G., Suter, A., Eskelinen, E.-L., Hartmann, D., Lüllmann-Rauch, R., Janssen, P.M.L., Blanz, J., Figura, K. von, and Saftig, P. (2000). Accumulation of autophagic vacuoles and cardiomyopathy in LAMP-2-deficient mice. Nature 406, 902–906.

Tang, Y., Wang, X.-W., Liu, Z.-H., Sun, Y.-M., Tang, Y.-X., and Zhou, D.-H. (2017). Chaperone- mediated autophagy substrate proteins in cancer. Oncotarget 8, 51970–51985.

Tarangelo, A., Lo, N., Teng, R., Kim, E., Le, L., Watson, D., Furth, E.E., Raman, P., Ehmer, U., and Viatour, P. (2015). Recruitment of Pontin/Reptin by E2f1 amplifies E2f transcriptional response during cancer progression. Nat. Commun. 6, 10028.

REFERENCES 177 Thompson, S.L., Bakhoum, S.F., and Compton, D.A. (2010). Mechanisms of chromosomal instability. Curr. Biol. CB 20, R285-295.

Toyomura, T., Murata, Y., Yamamoto, A., Oka, T., Sun-Wada, G.-H., Wada, Y., and Futai, M. (2003). From Lysosomes to the Plasma Membrane. Localization of vacuolar type H+-ATPase with the a3 isoform during osteoclast differentiation. J. Biol. Chem. 278, 22023–22030.

Trivedi, B., and Danforth, W.H. (1966). Effect of pH on the Kinetics of Frog Muscle Phosphofructokinase. J. Biol. Chem. 241, 4110–4114.

Tsantoulis, P.K., and Gorgoulis, V.G. (2005). Involvement of E2F transcription factor family in cancer. Eur. J. Cancer 41, 2403–2414.

Vessoni, A.T., Filippi-Chiela, E.C., Menck, C.F., and Lenz, G. (2013). Autophagy and genomic integrity. Cell Death Differ. 20, 1444–1454.

Vietri, M., Schink, K.O., Campsteijn, C., Wegner, C.S., Schultz, S.W., Christ, L., Thoresen, S.B., Brech, A., Raiborg, C., and Stenmark, H. (2015). Spastin and ESCRT-III coordinate mitotic spindle disassembly and nuclear envelope sealing. Nature 522, 231–235.

Villa-Hernández, S., and Bermejo, R. (2018). Replisome-Cohesin Interfacing: A Molecular Perspective. BioEssays News Rev. Mol. Cell. Dev. Biol.

Wang, L.-Y., Hung, C.-L., Chen, Y.-R., Yang, J.C., Wang, J., Campbell, M., Izumiya, Y., Chen, H.-W., Wang, W.-C., Ann, D.K., et al. (2016). KDM4A Coactivates E2F1 to Regulate the PDK-Dependent Metabolic Switch between Mitochondrial Oxidation and Glycolysis. Cell Rep. 16, 3016–3027.

Wang, Z., Sun, X., Bao, Y., Mo, J., Du, H., Hu, J., and Zhang, X. (2017). E2F1 silencing inhibits migration and invasion of osteosarcoma cells via regulating DDR1 expression. Int. J. Oncol. 51, 1639–1650.

Webb, B.A., Chimenti, M., Jacobson, M.P., and Barber, D.L. (2011). Dysregulated pH: a perfect storm for cancer progression. Nat. Rev. Cancer 11, 671–677.

White, E. (2012). Deconvoluting the context-dependent role for autophagy in cancer. Nat. Rev. Cancer 12, 401–410.

Wordeman, L. (2010). How Kinesin Motor Proteins Drive Mitotic Spindle Function: Lessons from Molecular Assays. Semin. Cell Dev. Biol. 21, 260–268.

Wu, M., Seto, E., and Zhang, J. (2015). E2F1 enhances glycolysis through suppressing Sirt6 transcription in cancer cells. Oncotarget 6, 11252–11263.

Wu, Y.E., Huo, L., Maeder, C.I., Feng, W., and Shen, K. (2013). The Balance Between Capture and Dissociation of Presynaptic Proteins Controls the Spatial Distribution of Synapses. Neuron 78, 994–1011.

178 REFERENCES Xiao, B., Sanders, M.J., Underwood, E., Heath, R., Mayer, F.V., Carmena, D., Jing, C., Walker, P.A., Eccleston, J.F., Haire, L.F., et al. (2011). Structure of mammalian AMPK and its regulation by ADP. Nature 472, 230–233.

Xu, H., and Ren, D. (2015). Lysosomal Physiology. Annu. Rev. Physiol. 77, 57–80.

Yamasaki, L., Jacks, T., Bronson, R., Goillot, E., Harlow, E., and Dyson, N.J. (1996). Tumor induction and tissue atrophy in mice lacking E2F-1. Cell 85, 537–548.

Yamazaki, K., Yajima, T., Nagao, T., Shinkawa, H., Kondo, F., Hanami, K., Asoh, A., Sugano, I., and Ishida, Y. (2003). Expression of transcription factor E2F-1 in pancreatic ductal carcinoma: an immunohistochemical study. Pathol. Res. Pract. 199, 23–28.

Yecies, J.L., and Manning, B.D. (2011). Transcriptional Control of Cellular Metabolism by mTOR Signaling. Cancer Res. 71, 2815–2820.

Yonehara, R., Nada, S., Nakai, T., Nakai, M., Kitamura, A., Ogawa, A., Nakatsumi, H., Nakayama, K.I., Li, S., Standley, D.M., et al. (2017). Structural basis for the assembly of the Ragulator-Rag GTPase complex. Nat. Commun. 8, 1625.

Zaffagnini, G., and Martens, S. (2016). Mechanisms of Selective Autophagy. J. Mol. Biol. 428, 1714–1724.

Zhang, C.-S., Jiang, B., Li, M., Zhu, M., Peng, Y., Zhang, Y.-L., Wu, Y.-Q., Li, T.Y., Liang, Y., Lu, Z., et al. (2014). The lysosomal v-ATPase-Ragulator complex is a common activator for AMPK and mTORC1, acting as a switch between catabolism and anabolism. Cell Metab. 20, 526–540.

Zhang, C.-S., Hawley, S.A., Zong, Y., Li, M., Wang, Z., Gray, A., Ma, T., Cui, J., Feng, J.-W., Zhu, M., et al. (2017). Fructose-1,6-bisphosphate and aldolase mediate glucose sensing by AMPK. Nature 548, 112–116.

Zhang, S.Y., Liu, S.C., Al-Saleem, L.F., Holloran, D., Babb, J., Guo, X., and Klein-Szanto, A.J. (2000). E2F-1: a proliferative marker of breast neoplasia. Cancer Epidemiol. Biomark. Prev. Publ. Am. Assoc. Cancer Res. Cosponsored Am. Soc. Prev. Oncol. 9, 395–401.

Zhu, C., Zhao, J., Bibikova, M., Leverson, J.D., Bossy-Wetzel, E., Fan, J.-B., Abraham, R.T., and Jiang, W. (2005). Functional Analysis of Human Microtubule-based Motor Proteins, the Kinesins and Dyneins, in Mitosis/Cytokinesis Using RNA Interference. Mol. Biol. Cell 16, 3187–3199.

Zoncu, R., Bar-Peled, L., Efeyan, A., Wang, S., Sancak, Y., and Sabatini, D.M. (2011). mTORC1 senses lysosomal amino acids through an inside-out mechanism that requires the vacuolar H(+)-ATPase. Science 334, 678–683.

REFERENCES 179 A picture is worth a thousand words Immunostaining of U2OS cells actin cytoskelleton ACKNOWLEDGEMENTS AGRAÏMENTS

Tot i ser aparentment una col·lecció de dades, aquesta tesi explica la història d’algú que va arribar a un territori desconegut amb un sac buit, hi va viure, créixer i gaudir i ara s’endú un sac ben ple.

Com tota bona història, comença amb el recolzament incondicional d’un pare i una mare que a base d’estima van regant la seva llavor i d’un germà que sempre em dóna la mà. Gràcies família per ser la espurna que encén la flama, pel recolzament diari. A tu, Eduard, per ajudar- me amb tot problema informàtic, per qüestionar-me amb altres punts de vista i per fer-me sentir intel·ligent.

El context d’aquesta tesi és un ecosistema molt divers, el laboratori de metabolisme del càncer.

Gràcies Albert, per haver-me obert les portes i per ensenyar-me des del primer dia a gaudir amb el que faig. Ets un gran exemple d’estima a la ciència. I a tu, Caroline per haver aparegut al bell mig del meu doctorat amb una panxa i un somriure que no cabien per la porta, per totes les discussions científiques i les hores compartides, ha sigut genial aprendre amb tu i tinc la certesa que ho seguirà sent per molt de temps. Gràcies a tu Santi pel teu temps i per les aportacions científiques que han enriquit aquests projectes.

Gràcies també a tots els membres de l’LMC (els actuals i els que han anat marxant), heu fet de la meva estada aquí unes convivències contínues. Al Ferran, per les seves imitacions i les llargues converses a la plaça de Sants, a Sónia por tu toque ácido que tanto se parece al mío, to Joffrey for all the stupid jokes late in the afternoon, a Antonio por todas las técnicas extrañas que has puesto a punto y que han cautivado mi curiosidad, a la Cristina, les Martes i la Mariona per organitzar la nostra entropia expansiva, a la Sandra per ser el viu exemple d’escolta activa i d’acompanyament, al Pedro que tot i que va marxar sempre hi va ser,a

Carmele por sus pocas (pero relevantes) palabras, i a la resta Ana, Fran, Gui, Giulio... por haber sido parte de esta creación.

ACKNOWLEDGEMENTS / AGRAÏMENTS 183 Al grupo Tauler, Const y Nathalie que me enseñasteis como funcionaba todo cuando entré aquí. Gracias por todas la cenas y vermuts compartidos cuando estábais en el lab y todos los que han venido después. Nat, gracias por enseñarme todo lo que pudiste siempre con predisposición y ganas.

A les meves estudiants, Alícia, Àlex i Belén, m’ha encantat ensenyar-vos i aprendre al vostre costat. He descobert amb vosaltres que la ciència és un conte ple d’intrigues per resoldre, i m’ha encantat ser-ne la narradora.

Thank you Sara and George for starting this lab in Barcelona and for giving me the opportunity to be part of the LMC; I learned everything I could during this years (even English). Thank you

George for all the ‘‘It is what Eugènia said’’ in the lab meetings, you made me feel confident to share my opinion. I will always remember you looking at me the first time I asked a question in a journal club, it made me feel that my scientific comments were worth hearing.

Gràcies a tot el personal dels serveis tècnics, sense vosaltres part d’aquests projectes mai hagués tirat endevant. A l’Esther ‘del sorter’ per estar predisposada a tot disseny experimental boig des del primer dia que vaig aparèixer amb dues hores de retard. Al Benja, la Carme, la

Maria, l’Anna i l’Elisenda, per fer que la cova del confocal estigui sempre plena de llum, pel bon humor i per tota l’ajuda.

Agraïda estic també a Nescafè per totes les càpsules amb orígens sorprenents que m’han mantingut desperta durant aquests quatre anys. I al personal del bar de l’IDIBELL que amb un somriure sempre m’ha preparat els dos entrepans de rigor per mantenir la meva dieta saludable.

Sembla que se m’acaben les paraules però no les persones, gràcies a les meves amistats que m’heu acompanyat aquests anys, les birres després del lab han sigut una gran part del meu doctorat. Gràcies Júlia, tu en sí ja ets la meva història, t’agraeixo especialment la teva

184 ACKNOWLEDGEMENTS / AGRAÏMENTS creativitat i el disseny d’aquesta tesi. Gràcies Olga i Heura com a principals responsables de que el meu sou de predoc pràcticament no arribés a final de mes. A tu Carlos, per ser part a nivell científic i personal, per donar-me esperances de que algún dia publicaré i per ser un mirall on admirar el que significa estimar el que fas.

I també els que vau començar amb mi aquest camí, a la Tati, per decidir venir a fer les pràctiques a l’LMC quan no sabíem res, brindo pel primer Western que ens va sortir! A l’Amanda, la meva família i amiga que va decidir un bon dia estudiar la mateixa carrera que jo. Gràcies per ser tant a prop tot i que sou lluny. I al Ferran, la Marta Alay, els Andreus que veu veure els inicis i ara veieu el final. I a tu, Martí per descobrir-me durant tot aquest temps la meva capacitat d’entusiasmar-me per aprendre coses noves.

Gràcies als meus múltiples companys de pis, una casa la fa la gent que hi viu. A tu Marc per sorgir del no res i proposar-me buscar un piset, i a tu Núria per decidir venir a fer una família, amb vosaltres, les vostres grips, les tardes amb la guitarra, la terrasseta i les experiències quotidianes he gaudit del millor entorn per seguir creixent com a persona. Gràcies per aguantar- me quan no em sortia res. Gràcies al Martí pels tres mesos compartits, hem descobert que som l’un per l’altre. I actualment, gràcies a la Fiona, la Cèlia, la Clàudia i l’Andreu, per convertir

Maridiana Palace en casa meva i pels descansos de l’escriptura a la terrassa durant l’estiu.

Penso que ha sigut el millor lloc on podia haver escrit aquest llibre.

Gràcies a tu Andreu, per formar part de l’estructura i el contingut en tot moment i per haver fet esforços sobrenaturals per entendre cada petit detall d’aquesta tesi (i també de mi mateixa).

GRÀCIES GRACIAS GRAZIE THANK YOU MERCI OBRIGADA

ACKNOWLEDGEMENTS / AGRAÏMENTS 185 Y a ti, yaya, a tu, àvia i a tu, iaio que tot i que no l’arribareu a llegir mai formeu part de mi en

cada gest, cada paraula i cada mirada.

186 ACKNOWLEDGEMENTS / AGRAÏMENTS