Novel Roles of the Ack1 Kinase in Epithelial Biology

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1469

GIULIA PILIA

ACTA UNIVERSITATIS UPSALIENSIS ISSN 1651-6206 ISBN 978-91-513-0355-0 UPPSALA urn:nbn:se:uu:diva-349384 2018 Dissertation presented at Uppsala University to be publicly examined in B42, BMC, Husargatan 3, Uppsala, Tuesday, 21 August 2018 at 13:00 for the degree of Doctor of Philosophy (Faculty of Medicine). The examination will be conducted in English. Faculty examiner: Professor Bengt Hallberg (Department of Medical Biochemistry and Cell biology at Institute of Biomedicine, University of Gothenburg).

Abstract Pilia, G. 2018. Novel Roles of the Ack1 Kinase in Epithelial Biology. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1469. 85 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0355-0.

Epithelial homeostasis is maintained through integration of diverse signals that regulate cell fate. A strict control of such signals is required to prevent overproliferation and, ultimately, oncogenesis. In this thesis we identify novel roles of Activated Cdc42-associated kinase 1 (Ack1) in maintenance of epithelial homeostasis. Ack1 has been previously linked to cytoskeletal remodeling, signal transduction and gene expression regulation. Interestingly, our work reveals that Ack1 is also important for I) promoting extrinsic apoptosis, II) mediating mechanically-induced inhibition of proliferation and III) attenuating mitogenic signals, fundamental functions to prevent aberrant tissue growth. Apoptosis is a program of regulated cell death that can be triggered by several pathways. Among them, the TNF-related apoptosis-inducing ligand (TRAIL)-induced apoptosis cascade has raised interest for cancer treatment, as many cancer cell lines are susceptible to it. We found that Ack1 increases sensitivity to TRAIL by promoting translocation of ligand-bound TRAIL- Receptor to lipid rafts. Localisation at the lipid rafts, in turn, favors recruitment of downstream signalling effectors, enhancing the apoptotic response. Yap and Taz are transcriptional co-factors that integrate mechanical and soluble cues to regulate cell proliferation and differentiation. Yap/Taz regulation is mediated by cytoplasmic sequestration and, particularly for Taz, proteasomal degradation via ubiquitination by the E3 ligase β-TrCP. We discovered that Ack1 is activated by mechanical signals and promotes nuclear exclusion of Yap/Taz. Ack1 promotes Yap/Taz interaction with β-TRCP and it is required for efficient degradation of Taz. Consequently, Ack1 limits Yap/Taz-dependent gene expression and cell proliferation. The ErbB family of receptor tyrosine kinases mediates pro-survival and proliferative signals of crucial importance in development and cancer. Among the ErbB family members, ErbB3 has significant oncogenic properties as it potently activates the PI3K/Akt signaling pathway. We observe that Ack1 depletion increases ErbB3 total levels, but not EGFR and ErbB2, and is required for both basal turnover of ErbB3 and its ligand-induced degradation. Consequently, Ack1 attenuates ErbB3-dependent signalling upon Neuregulin-1β treatment. Additionally, Ack1 reduces ErbB3 gene expression both at steady state and upon stimulation, revealing its importance as multi-level regulator of ErbB3. Taken together, our data depict new roles for Ack1 in epithelial cells, highlighting its multifaceted role in maintenance of epithelial homeostasis.

Keywords: Ack1, Epithelial homeostasis, Cell Signalling, Cancer, Apoptosis, TRAIL-R, Yap/ Taz, Mechanotransduction, ErbB3

Giulia Pilia, Department of Medical Biochemistry and Microbiology, Box 582, Uppsala University, SE-75123 Uppsala, Sweden.

ISSN 1651-6206 ISBN 978-91-513-0355-0 urn:nbn:se:uu:diva-349384 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-349384)

To my father, from whom I inherited the curiosity and the sense of humour needed to complete this thesis

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Linderoth E.*, Pilia G.*, Mahajan N.P., Ferby I. (2013) Acti- vated Cdc42-associated kinase 1 (Ack1) is required for tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) receptor recruitment to lipid rafts and induction of cell death. Journal of Biological Chemistry, 288(46):32922-31.

II Pilia G., Darai E., Ferby I. (2018) ACK1 is a mechanoresponsive kinase required for SCF (β-TrCP)-mediated degradation of YAP/TAZ. Manuscript.

III Pilia G.*, Sáez-Ibáñez A.R.*, Linderoth E., Ferby I. (2018) Ack1 is a negative regulator of ErbB3 that acts both by promoting its degradation and suppressing its expression. Manuscript.

*Asterisks indicate equal contribution

Reprints were made with permission from the respective publishers.

The following paper, by the same author, was not included in this thesis:

Hopkins S., Linderoth E., Hantschel O., Suarez-Henriques P., Pilia G., Kendrick H., Smalley M.J., Superti-Furga G., Ferby I. (2012) Mig6 is a sensor of EGF receptor inactivation that directly activates c- Abl to induce apoptosis during epithelial homeostasis. Dev Cell. 2012 Sep 11; 23(3):547-59.

Contents

1. Introduction ...... 13 2. Ack1 is a versatile non-receptor kinase of critical importance for epithelial homeostasis ...... 15 2.1 The unique domain structure of Ack1 supports binding with several interaction partners ...... 15 2.2 The activity of Ack1 is regulated by auto-inhibition ...... 17 2.3 Ack1 is a multifunctional kinase and a signalling hub ...... 19 2.3.1. Roles in cytoskeletal remodelling and trafficking ...... 19 2.3.2. Roles in promoting pro-survival signals and regulation of gene transcription ...... 20 2.3.3. Specialised functions of Ack1 in the nervous and immune systems ...... 21 2.4 Ack1 is involved in oncogenesis and cancer progression ...... 22 3. ErbB receptors-mediated signalling is central for cell survival, proliferation and migration ...... 24 3.1 The ErbB family of receptors is activated by asymmetric dimerization ...... 24 3.2 The ErbB receptors activate signalling cascades implicated in cell survival, differentiation and migration ...... 26 3.3 ErbB-dependent signalling drives development and neoplastic transformation ...... 28 3.4 A wide range of mechanisms modulate ErbB receptors-dependent signalling ...... 29 3.4.1 ErbB receptors signalling is spatiotemporally controlled via regulated trafficking ...... 30 3.4.2 ErbB signalling can be attenuated by competitive binding, phosphatases and adaptor proteins ...... 31 3.5 The catalytically-impaired ErbB3 holds signalling capacity and dedicated mechanisms of regulation ...... 32 4. Mechanical regulation of Yap and Taz controls cell fate ...... 35 4.1 Mechanical cues are crucial messengers in the regulation of cellular functions ...... 35 4.1.1 Cells sense mechanical signals ...... 36 4.2 Yap and Taz are mechano-responsive transcriptional co-factors ...... 38

4.2.1 Yap and Taz’s gene transcription signature directs cell fate ...... 38 4.2.2 Yap and Taz are structurally similar, yet not identical ...... 39 4.2.3 The regulatory network of Yap and Taz is dominated by mechanotransduction ...... 40 4.2.4 Mechanical regulation of Yap and Taz governs cell fate ...... 43 5. Apoptosis: the regulated cell death program that controls tissue stability 45 5.1 Apoptosis can be triggered by both internal and external cellular signals ...... 45 5.1.1 The intrinsic pathway of apoptosis is executed via permeabilization of the mitochondria ...... 46 5.1.2 The extrinsic pathway of apoptosis is triggered by receptors at the plasma membrane ...... 47 5.2 The regulation of TRAIL-Receptors signalling is of special interest in cancer ...... 50 6. Present investigations ...... 52 Paper I: Activated Cdc42-associated kinase 1 (Ack1) is required for tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) receptor recruitment to lipid rafts and induction of cell death ...... 52 Paper II: Ack1 is a mechanoresponsive kinase required for SCF (β- TrCP)-mediated degradation of Yap/Taz ...... 53 Paper III: Ack1 is a negative regulator of ErbB3 that acts both by promoting its degradation and suppressing its expression ...... 54 7. Future perspectives ...... 55 Is Ack1-mediated TRAIL-R recruitment to lipid rafts an aspect of a more general mechanism? ...... 55 What are the molecular details of Ack1 regulation of Yap/Taz downstream of mechanical stimulation? ...... 56 Which mechanisms underlie Ack1-dependent regulation of ErbB3? .. 56 How are Ack1 functions integrated? ...... 57 Acknowledgements ...... 58 References ...... 63

Abbreviations

Ack1 Activated Cdc42-associated Kinase 1 AJ Adherens junction Akt RAC-alpha serine/threonine-protein kinase Alk Anaplastic lymphoma kinase AMOT Angiomotin aP Adipocyte fatty acid-binding protein AP-2 Adaptor protein 2 APAF Apoptotic peptidase activating factor APC Adenomatosis polyposis coli APO Apoptosis antigen AR Androgen receptor ARE Androgen responsive enhancer BAD Bcl-2-associated death promoter BAK BCL-2 homologous antagonist killer BAX BCL-2-associated X protein BCL B-cell limphoma BH BCL-2 homology BID BH3 interacting-domain death agonist BIM BCL-2 interacting mediator Cas Crk-associated substrate Casp- Caspase Cbl Casitas B-linage lymphoma proto-oncogene CD95 Cluster of differentiation 95 Cdc42 Cell division control protein 42 homolog c-Fos cellular-FBJ murine osteosarcoma viral oncogene homolog c-IAP Inhibitor of apoptosis protein CIP Contact inhibition of proliferation CK1 Casein kinase 1 CRIB Cdc42/Rack interactive binding Crk CT10 regulator of kinase DAG Diacylglycerol dATP Deoxyadenosine triphosphate Dbl Diffuse B-cell lymphoma DcR Decoy receptor DD Death domain

DISC Death inducing signalling complex DNA Deoxyribonucleic acid DR Death receptor EBD EGFR binding domain ECM Extracellular matrix EEA Early endosome antigen 1 EGF Epithelial growth factor EGFR EGF Receptor EMT Epithelial to mesenchymal transition ER Estrogen receptor ErbB Erythroblastic leukemia viral oncogene homolog ERRP EGFR Related Peptide FA Focal adhesion FADD Fas-associated protein with death domain FAK Focal adhesion kinase Fas First apoptosis signal receptor FLICE FADD-like Interleukin-1β-converting enzyme FLIP FLICE-like inhibitory protein FOXO Forkhead box O FRS2β Fibroblast growth factor receptor substrate 2β GPCR G-protein coupled receptor Grb2 Growth factor receptor-bound protein 2 GSK3 Glycogen synthase kinase 3 GTP Guanosine tri-phosphate H3K9 Histone 3 lysine 9 hnRNPU Heterogeneous nuclear ribonuclear protein U HOXA Homeobox A IP3 Inositol 1,4,5-triphosphate JAK Janus kinase KDM3A Lysine demethylase 3A Lats Large tumour suppressor kinase Lrig1 Leucine-rich repeats and immunoglobulin-like domains 1 LINC Linker of nucleoskeleton and cytoskeleton Ltk Leucocyte receptor tyrosine kinase LUBAC Linear ubiquitin chain assembly complex MAPK Mitogen-activated protein kinase ERK Extracellular signal–regulated kinase Mdm2 Mouse double minute 2 homolog MHR Mig6 homology region Mig6 Mitogen-inducible gene 6 protein MLL2 Myeloid lymphoid leukemia 2 MOB Mps one binder protein MOMP Mitochondrial outer membrane permeabilization MRTF Myocardin-related transcription factor

MSC Mesenchymal stem cells Mst Mammalian Ste20-like serine/threonine kinase mTOR Mammalian target of rapamycin Nedd Neural precursor cell expressed developmentally down-regulated protein Neuregulin Nrg NFAT5 Nuclear factor of activated T-cells 5 NF-κB Nuclear factor κ B NGF Nerve growth factor Nrdp1 Neuregulin receptor degradation protein 1 NSCLC Non-small-cell lung cancers OPG Osteoprotegerin PDGFR Platelet-derived growth factor receptor PI3K Phosphatidylinositol-4,5-bisphosphate 3’-kinase PIP3 Phosphatidylinositol 3,4,5 triphosphate PKC Protein kinase C PKD Protein kinase D PLC-γ Phospholipase C-γ PPARγ Peroxisome proliferator-activated receptor γ PTB Phosphotyrosine binding PUMA p53 upregulated modulator of apoptosis Rac Ras-related C3 botulinum toxin substrate Raf Rapidly accelerated fibroma RALT Receptor associated late transducer Ras Rat sarcoma protein RCD Regulated cell death Rho Ras homolog RING Really interesting new gene RIPK Receptor interacting serine/threonine protein kinase ROCK Rho-associated coiled-coil containing kinase RUNX Runt-related transcription factor SAM Sterile alpha motif SAV1 Salvador 1 SH2 Src homology domain 2 SH3 Src homology domain 3 Shc SH2-containing collagen-related protein SIAH Seven in absentia homolog SLP-76 SH2-domain containing leukocyte protein of 76 kDa SMAC Second mitochondria derived activator of caspases SMAD Sma mother against decapentaplegic SNX9 Sortin nexin 9 SOCS Suppressors of the cytokine signalling Sos Son of sevenless SQSTM1 Sequestosome 1

Src Sarcoma kinase STAT Signal transducer and activator of transcription TAD Trans-activation domain Taz Transcriptional co-activator with PDZ-binding motif TEAD TEA domain protein TGFα Transforming growth factor α TNF Tumour necrosis factor TNF-R TNF receptor TRADD TNF-R associated death domain protein TRAF TNFR associated factor TRAIL TNF-related apoptosis-inducing ligand TRAIL-R TRAIL receptor Trk Tropomyosin receptor kinase UBA Ubiquitin associated WASP Wiskott-Aldrich syndrome protein WDR5 WD repeat domain 5 WWox WW domain-containing oxidoreductase XIAP X-linked inhibitor of apoptosis Yap Yes-associated kinase ZO-2 Zonula occludens-2 β-TrCP β-Transducin repeat-containing protein

1. Introduction

The healthy functioning of a living organism requires, beside the integrity of its parts, their ability to work together as a whole. The maintenance of tissue homeostasis depends on the coordination of the behaviour of all the cells that compose it. This is accomplished through a complex system of communication between cells and their environment or from cell to cell, that makes use of signals of diverse nature. The integration of such signals can direct cell fate towards survival, proliferation, or regulated death. How cells sense these signals and translate them into biochemical changes, a process collectively called signal transduction, is a fascinating field of study in cell biology, with crucial implications in both normal physiology and pathological conditions. Persistent perturbations of the system of cell communication are indeed involved in the etiology of various diseases. Cancer is a classic paradigm of the catastrophic outcome deriving from an autonomous behaviour of cells that evade the tissue-level control of their fate1,2. Given the fact that a large major- ity of human malignancies originate from epithelia, the importance of approaching the study of signal transduction in epithelial cells appears compelling. A large number of molecular events determine and sustain cancer transformation at the single-cell level. Additionally, cancer is an intrinsically hetero- genous disease; therefore, these events vary not only in between different cancers, but also among the same type of cancer arising in different individuals and in different cells in the same tumour. However, this complex view can be simplified by conceptually framing these molecular events into functional cat- egories, as exemplified by the now classic review of Hanahan and Weinberg that defined them: The Hallmarks of Cancer1,2. By this classification the complex series of events at the root of carcinogenesis and cancer progression are categorised into a few groups, based on the selective advantage they confer to the cell. The cancer properties defined in the original review comprised: evasion of apoptosis, self-sufficiency in (or hypersensitivity to) growth signals, resistance to anti-growth signals, sustained angiogenesis, limitless replicative potential, and the ability to invade tissue and metastasize1. Later additions to this list included deregulation of cellular energetics and the ability to avoid immune destruction, highlighting at the same time the importance of genome instability and tumour-promoting inflammation in enabling the occurrence of all the aforementioned hallmarks2. It is however important to note that despite the usefulness of this simplified view, all these cancer determinants are in

13 strict relationship with one-another. Thus, deregulation of a single cellular process often affects more than one of them. Three of the originally presented hallmarks of cancer are of particular interest for this thesis: evasion of apoptosis, increased sensitivity to growth signals and resistance to anti-growth signals. In the work presented in this thesis we will, in fact, describe three novel and independent mechanisms by which a single protein, called Activated Cdc42-associated Kinase 1 (Ack1), prevents the emergence of these cancer traits.

14 2. Ack1 is a versatile non-receptor kinase of critical importance for epithelial homeostasis

Ack1 is a tyrosine kinase primarily studied for its roles in remodelling of the cytoskeleton, trafficking of receptors and regulation of gene transcription3. The regulation of Ack1 activity downstream many different signals, and its multifaceted roles in epithelial cells (further discussed in section 2.3), make it a very interesting signalling hub, whose complete network of interaction is yet to be uncovered. In the work presented in this thesis we explored novel functions of Ack1, discovering an ambivalent side of this versatile kinase in the context of epithelial biology and oncogenesis. To set the ground for presenting our novel findings, in this chapter we will present an overview of the current understanding of Ack1 biology.

2.1 The unique domain structure of Ack1 supports binding with several interaction partners Ack1 (also known as Tnk2, for tyrosine non-receptor kinase 2) belongs to the Ack family of non-receptor tyrosine kinases. In humans, the family is formed by two members, Ack1 and Tnk1, and it is characterized by the distinctive presence of a SH3 (Src homology 3) domain located C-terminal to the catalytic site and of a neighbouring CRIB (Cdc42/Rack interactive binding) domain. Additionally, Ack1 and Tnk1 both contain a SAM (Sterile Alpha Motif) domain at the N-terminus and, at the C-terminus, a proline-rich sequence including a clathrin-binding (CB) motif and a Ubiquitin-Associated (UBA) Do- main. However, Ack1 is larger than Tnk1 and its 1038 amino acids-long sequence features a more extended proline-rich region, which includes a WW- binding motif and a Mig6-Homology Region (MHR) (Fig. 1). The presence of several different functional domains reflects the multiple interaction partners and, consequently, physiological roles of Ack1.

Figure 1. Ack1 domain architecture. Representation of the main functional domains of Ack1. The yellow vertical lines indicate the position of the tyrosine auto-phosphorylation sites. WWb stands for WW domain binding region, see text above for other abbreviations.

Walking down the Ack1 sequence from the N-terminus, the first characterised functional motif is the SAM domain, which has been proposed to have an important role in promoting Ack1 activation. The SAM domain directs Ack1 localisation to the plasma membrane upon cell adhesion, causing an increase of its local concentration, and mediates dimerization and activation of the pro- tein4,5. In particular, a short sequence ranging from residue 10 to 14 has been found to be critical for adhesion-mediated activation6. However, it has not been clarified if its importance is dependent on the stabilisation of the alpha- 1 helix of the SAM domain or if it constitutes on its own an interacting surface crucial for homodimerization. More recently, the SAM domain of Ack1 has been proposed to mediate association with the SAM domain of another protein, the SH2-domain-containing leukocyte protein of 76 kDa (SLP-76)7. The SAM domain is followed by the kinase domain (discussed in detail in section 2.2), and by a SH3 domain. The SH3 domain interacts with binding sequences in the proline-rich region, stabilising the auto-inhibited conformation of Ack1 (further discussed in section 2.2). Moreover, the SH3 domain mediates interaction with p130Cas to promote invasion and motility8. The CRIB domain was the earliest characterised region of Ack1. As its extended name suggests, Ack1 was first isolated as an interactor of GTP- bound Cdc429. This interaction, which engages the CRIB domain, is important for Ack1 activation and its cytoskeletal functions and it has been reported to be specific for activated Cdc42 over other GTPases such as Rac and Rho9. The second half of Ack1 encloses a large proline-rich domain that serves as a binding hub for different SH3 domain-containing binding partners, including Grb26,10, Sortin nexin 9 (SNX9)11, and cortactin12. Moreover, the proline-rich region comprises the MHR, two CB motifs and a WW-binding domain. The presence of two CB domains highlights the importance of Ack1 in regulating vesicular trafficking. The two motifs are different in sequence, and mutation of both of them is required to completely abolish clathrin binding to Ack113,14. In close proximity to the CB boxes, the classical PPxY consensus sequence mediates interaction with type I WW domain-containing proteins, such as the E3 ligases Nedd 4-1 and 215,16 and the WW domain-containing oxidoreductase (WWox)17.

16 The Mig6-homology region (also known as EBD for EGFR-binding domain) mediates direct binding to Grb2 and EGFR. Ack1 binding to EGFR is dependent on the kinase activity of the latter, while activation of Ack1 is dispensable for the interaction18. Ack1 binding to EGFR regulates intracellular trafficking of the receptor, the most explored and yet debated function of Ack118–21 (discussed in section 2.4). Last, the UBA domain at the C-terminus confers affinity for mono- and poly-ubiquitin chains and it is required for interaction with the receptor for selective autophagy sequestosome 1 (p62/SQSTM1)18,21. The interaction between Ack1 and p62/SQSTM1 is suppressed by EGF stimulation and regulates the sorting of EGFR-containing vesicles.

2.2 The activity of Ack1 is regulated by auto-inhibition The crystal structure of full length Ack1 is still missing, but partial structures have been resolved for the kinase domain, for a fragment spanning through the kinase and SH3 domains, and for the CRIB domain in complex with Cdc4222–24. The kinase domain of Ack1 displays outstanding similarities to the one of EGFR. Not only do they share 39% sequence identity, but they also are both stabilised in an active conformation in the absence of phosphorylation22,23. Many kinases are relieved from auto-inhibition by means of phosphorylation of the so-called activation loop. The activation loop of Ack1 contains one single tyrosine at residue 284 that has been characterised as a site of auto-phosphorylation and as a Src substrate25,26. Although one study has reported the importance of phosphorylation by Src for Ack1 activation26, the available structural information and most of the experimental data suggest that the kinase domain is constitutively stabilised in an active conformation. Due to the presence of critical hydrophobic residues, the unphosphorylated activation loop is well ordered and does not block the ATP binding pocket nor the substrate recognition site22. Analogously, phosphorylation of the tyrosine in the activation loop does not induce substantial re-organisation of the catalytic sites, suggesting that allosteric interactions might instead be responsible for auto-inhibition. The similarity of the kinase domains of EGFR and Ack1 first suggested the possibility that an interaction in cis between the kinase domain and the MHR, which binds the kinase domain of EGFR, could mask the active site. Experi- mental evidence has corroborated this hypothesis by showing that the isolated MHR interacts with the kinase domain and the association is prevented in presence of the somatic cancer mutation E346K, which constitutively activates Ack127.

17 A large amount of data suggests that the SH3 domain could stabilise the closed (auto-inhibited) conformation of Ack1. The MHR of Ack1 has a proline-rich 10 amino acids-long insertion with respect to the homologous region in Mig6 that constitutes a consensus motif for SH3 binding, and mutations in the SH3 domain enhance Ack1 activity4,6,23. The mechanism of release of the auto-inhibitory loop that leads to Ack1 activation is still debated. Self-association could be displaced by competitive binding of Grb2 to the MHR or by the conformational changes induced by binding of Cdc42 to the CRIB domain6,12,28 (Fig. 2a and b). Interestingly, three tyrosine phosphorylation sites of Ack1 recurrently detected in phosphoprote- omic studies are located in the MHR at residues Y827, Y859, Y860, suggesting a role of phosphorylation in maintaining the open conformation (Phos- phoSitePlus®29). Additionally, homodimerization via SAM-SAM domain interaction is reportedly the main mechanism of adhesion-dependent activation of Ack1 (Figure 2c)5. How exactly dimerization leads to release of the self- association loop has not been resolved in detail. However, the possibility that the formation of symmetric dimers stabilises the active conformation is sup- ported by crystallographic data23. Overall, it is possible to hypothesize that all the proposed modes of activation of Ack1 coexist in the cell and are alternatively mediated by different activation stimuli6.

Figure 2. Ack1 activation model, according to Lin et al., 20126. Abbreviations as in Figure 1.

No specific phosphatase has been described to inactivate Ack1 but both proteasomal and lysosomal degradation of active Ack1 have been re- ported15,16. Interestingly, lysosomal degradation of activated Ack1 is dependent on ubiquitination by Nedd4-1, which acts as a sorting signal. The main site of Ack1 ubiquitination has not been described, but deletion studies and the high concentration of lysine in the SAM domain indicate it could be the main ubiquitin acceptor. Additionally, SIAH ubiquitin ligases control the levels of

18 Ack1, promoting its proteasomal degradation independently on its activation status30.

2.3 Ack1 is a multifunctional kinase and a signalling hub Ack1 is activated downstream of many different membrane receptors, such as integrins, M3 muscarinic receptor, and a wide group of receptor tyrosine kinases, namely PDGFR, EGFR, Axl and Trk4,31–33. In addition, both Ltk and Alk have been reported to interact with Ack1 in an activation-dependent manner, suggesting an additional involvement in signalling from these receptors32. However, sensitivity to stimuli and activation kinetics have been reported to be cell type-specific4. The vast variety of activator pathways that converge on Ack1 is mirrored by the diversity of cellular processes in which it is involved.

2.3.1. Roles in cytoskeletal remodelling and trafficking Ack1 has a well-established role in cytoskeletal remodelling, acting both downstream and upstream of Cdc4228. As mentioned above, activated Cdc42 can interact with and activate Ack1. Conversely, activated Ack1 promotes activation of Rho-GTPase family members via phosphorylation of the guanine exchange factor Dbl28,34. Furthermore, Cdc42-dependent activation of Ack1 upon stimulation by laminin or chondroitin-sulphate proteoglycan results in phosphorylation of p130Cas at three tyrosine residues. These phosphorylation events promote the association of p130Cas and Crk, in turn causing the remodelling of the actin cytoskeleton in the context of cell migration8,35,36. Moreover, Ack1 can promote branching of the cytoskeleton via direct phosphorylation of two additional substrates: WASP and cortactin. Surprisingly, Ack1 can phosphorylate WASP at both tyrosine and serine residues, promoting actin polymerisation in vitro17. This uncommon double-specificity has been explained as the result of the high flexibility of the Ack1 kinase domain. However, it is not clear if Ack1-dependent phosphorylation of serine residues could happen in other contexts, as no other serine substrate has been described to date. Upon EGF stimulation, Ack1 phosphorylates cortactin at three tyrosine residues (Y421, Y466, Y482) that are also substrates of many other non- receptor tyrosine kinases12. In this context, phosphorylated cortactin triggers cytoskeletal remodelling linked to clathrin-mediated endocytosis (CME) of EGFR. The involvement of Ack1 in the CME of EGFR has been confirmed by a significant amount of data. First, inhibition of endocytosis by dynamin depletion induces accumulation of activated Ack1 in clathrin-coated pits14. Addi- tionally, Ack1 overexpression generates clathrin aggregates, compromising

19 clathrin-dependent trafficking13. Moreover, Ack1 has been found to co-localise with AP-2, the early endosome marker EEA-1 and SNX9, potentially broadening its connection to the trafficking machinery11,13,18,20. More specifically, Ack1 can directly bind EGFR via its MHR, and the interaction requires the kinase activity of EGFR, but not that of Ack118,19. Conflicting conclusions have been drawn on the effect of Ack1 on EGFR trafficking, as both roles in preserving localisation of the receptor at the membrane19, and in promoting internalisation and degradation18,20 have been proposed. More recently, it has been suggested that Ack1 limits lysosomal targeting of activated EGFR by re- routing it to a non-canonical autophagosome-associated degradation path- way21.

2.3.2. Roles in promoting pro-survival signals and regulation of gene transcription Besides roles related to cytoskeleton and endosomal trafficking, Ack1 has been found to promote pro-survival signals by direct phosphorylation of different substrates. Notably, Ack1 phosphorylates the key signalling component Akt at tyrosine 176, thereby mediating an activation mode that is alternative to the classical PI3K-dependent mechanism. The traditional Akt mode of activation includes translocation to the plasma membrane by binding to PI3K- generated phosphatidylinositol 3,4,5-triphosphate (PIP3). Once bound to the membrane, Akt can be activated by phosphorylation at serine and threonine residues by PKD1/2. In contrast, tyrosine-phosphorylated Akt can be recruited to the membrane by binding phosphatidic acid instead of PIP3, leading to its activation by phosphorylation at the same sites37. Another important substrate of Ack1 phosphorylation is the androgen receptor (AR). Activation of AR by ligand binding promotes its translocation to the nucleus, activating androgen-dependent transcription. However, in the context of prostate cancer, aberrant ligand-independent activation occurs38. It was found that, in absence of androgen stimulation, Ack1 phosphorylates AR at residues Y267 and Y363 in the transactivation domain39. The phosphorylated receptor was consequently able to localise at androgen-responsive enhancer (ARE) regions of the DNA and promote androgen-dependent transcription. Interestingly, it was observed that activated Ack1 localised at AREs together with phosphorylated AR. Furthermore, it was recently found that Ack1 sustains AR expression, acting as an epigenetic writer40. Ack1 can directly phosphorylate histone 4 at Y88, promoting recruitment of epigenetic modulators such as WDR5 and MLL2, which, in turn, methylate histone 3 at K4, inducing expression of the gene encoding AR. The effect of Ack1 on AR expression is reminiscent of another identified epigenetic function, played in the context of estrogen receptor (ER)-dependent transcription. In this case, however, Ack1 does not act as a direct writer of the

20 epigenetic code, but instead activates the eraser KDM3A, a histone 3 lysine 9 (H3K9) demethylase. Upon ER association, Ack1 phosphorylates KDM3A at Y1114, activating its catalytic function to reduce H3K9 methylation, and in turn inducing HOXA1 expression41. Moreover, Ack1 can directly activate gene transcription by yet another mechanism. STAT (Signal Transducer and Activator of Transcription) 1 and 3 have been identified as Ack1 substrates. Ack1-dependent phosphorylation, analogously to the better described JAK-dependent one, induces STATs dimerization and translocation to the nucleus for the transcriptional activation of their target genes42,43. On a different note, Ack1 has been identified as a pro-apoptotic molecule. The most characterised mechanism in this respect is the phosphorylation of WWox, a proapoptotic factor involved in stress-induced cell death. Ack1-dependent phosphorylation, predominantly at Y287, leads to WWox ubiquitination and degradation, resulting in decreased apoptosis44,45. More recently, a role of Ack1 in inhibiting TNF-α (Tumour Necrosis Factor α)-dependent apoptosis has also been proposed46.

2.3.3. Specialised functions of Ack1 in the nervous and immune systems Ack1 is ubiquitously expressed in human tissues, with higher expression levels in thymus, spleen and brain4,47. The expression of Ack1 in the nervous system has been reported both in adult and developing tissues and in neural cell cultures; Ack1 shows a distinct localisation in presynaptic terminals and developing dendrites and axon47,48. Moreover, Ack1 regulates neural arborisa- tion both by perturbing neurotrophin-dependent Trk signalling, and by neurotrophin-independent processes33. Additionally, Ack1 has an important role at pre-synapses, by regulating endocytosis of the dopamine transporter49. Inter- estingly, a missense mutation inhibiting Nedd 4-1 and -2 association and therefore Ack1 degradation was found in a family displaying severe infantile epilepsy50, indicating that Ack1 could be important for a correct functioning of the nervous system. In the immune system, Ack1 supports the function of T-cells by phosphor- ylating SLP-76 and promoting early signalling events downstream T-cells receptor activation7. This suggests that Ack1 may play an important role in the appropriate unfolding of immune responses. As many of the known Ack1 functions have a pro-proliferative and anti- apoptotic outcome, Ack1 has been classically considered a pro-oncogenic protein. A substantial amount of studies strengthens this view, as summarized in the following section (2.4).

21 2.4 Ack1 is involved in oncogenesis and cancer progression An amplification of the gene encoding Ack1 has been found in a variety of cancers, including breast, ovarian, endometrial, cervical, prostate, lung, gastric, head and neck tumours3,35,51–54. Additionally, over 70 somatic mutations affecting its coding sequence were detected in exomes in patient samples (COSMIC database55). Six missense mutations discovered in a big kinase-ori- ented study of leukaemia samples were characterised in functional studies. Despite their spread distribution in the sequence (R34L, R99Q, D163E, E346K, M409I and R806Q), they all resulted in enhanced activation of Ack1 by relieving its auto-inhibited conformation27,56. The importance of Ack1 in neoplastic development and tumour progression has been described for several cancer subtypes. In pancreatic cancer, Ack1 contributes to resistance to therapy by promoting AR expression and ligand-independent activation39,40. Accordingly, Ack1 overexpression and Y284 phosphorylation were found to correlate with tumour progression and poor prognosis35,57,58. Additionally, the Ack1/Akt activation module could be responsible for early stages of tumour formation, as suggested by a study in which mice conditionally expressing Ack1 in the prostate showed overactivated Akt and developed prostatic intraepithelial neo- plasia37. Phospho-Y284 Ack1 and phospho-Y176 Akt levels correlated with tumour progression and lower survival rate in a tissue array of breast cancers37. Fur- thermore, Ack1 overexpression in breast cancer cells enhances tumour formation, metastasis at the lung and mortality in mice35. Immunohistochemistry data also revealed Ack1 overexpression in lung adenocarcinoma. Interestingly, Ack1 overexpression in tumour-surrounding tissue independently correlated with poor prognosis59. Ack1 overexpression in hepatocellular carcinoma in comparison to adjacent non-tumour tissue has also been detected, and correlated with invasiveness, relapse and, once more, lower survival rates45,60,61. Moreover, Ack1 overexpression increased tumour size and metastatic potential in a mouse model61. Given the array of evidence shown above, a clear role of Ack1 has been established in oncogenesis and tumour progression. A therapeutic approach targeting Ack1, in particular in prostate cancer, has been proposed, and there is currently an effort to develop specific, water-soluble, Ack1 kinase inhibitor suitable for use in the clinic40,62. In the work presented in this thesis, however, an ambivalent role of Ack1 function in cancer emerges, which underlines the importance of this protein for central physiological processes. Particularly, our research has uncovered novel independent roles of Ack1 in promoting extrinsic apoptosis (Paper I) and inhibiting mitogenic signals by transducing mechanical cues (Paper II) or controlling growth factor signalling (Paper III).

22 To highlight the significance of these findings, the next chapters will describe the cellular processes of apoptosis and mechanotransduction (especially the Yap/Taz pathway), as well as the role of ErbB receptors in various signalling pathways linked to oncogenesis and cancer progression.

23 3. ErbB receptors-mediated signalling is central for cell survival, proliferation and migration

The ability of cells to sense signals from the environment is crucial in regulating their function. The nature of these signals is extremely broad, ranging from soluble molecules to changes of the composition of the extracellular matrix or mechanical stimuli. Growth factors are one important type of soluble messengers that determine cell fate and behaviour by binding to specialised receptors at the plasma membrane. The first discovery of a growth factor dates back to the 1950s, with the identification of a protein able to promote growth of nerve fibers: the Nerve Growth Factor or NGF63,64. The epithelial growth factor (EGF) was isolated shortly after from mouse submaxillary gland, as a 53-amino acid polypeptide able to promote eyelid opening and teeth eruption in the new-born pups65. Ever since, EGF and its receptor have been extensively studied and proved to be crucial for promoting cell growth and survival, differentiation and migration. They cover, therefore, a central role in the regulation of development and in cancer biology.

3.1 The ErbB family of receptors is activated by asymmetric dimerization The EGF receptor (EGFR, also known as ErbB-1 for Erythroblastic leukemia viral oncogene homolog-1or HER-1 for Human EGF Receptor Kinase) belongs to a family of receptor tyrosine kinases that includes 3 additional members: ErbB2, ErbB3 and ErbB4 (known, as well, as HER-2, -3 and -4, respectively). The ErbBs share a common structural arrangement that includes an extracellular N-terminus, a single transmembrane segment and a C-terminal intracellular region consisting of the juxtamembrane part, the kinase domain and a terminal tail. The extracellular domain is composed of four domains, named I, II, III and IV. I and III are homologous β-helical domains that comprise the ligand binding site, while domain II and IV are – also homologous – cysteine-

24 rich regions that interact with each other in the closed, non ligand-bound, con- formation66,67. Binding of a ligand molecule to the receptor induces a rearrangement of the extracellular domains that leads to the exposure of a hairpin loop of domain II66,68. This region constitutes the dimerization interface, which binds the same region of a second ErbB receptor. In ErbB2, however, the interaction between domain II and IV is impaired due to substitutions of critical residues in the latter, and this receptor is therefore stabilized in an open – dimerization prone – conformation69,70. ErbB2 is indeed the only orphan receptor of the family, with no ligand known, or needed, to cause its activation.

Figure 3. A model of EGFR ligand-induced homo-dimerization as an archetype of ErbB receptors activation.

Homo- or hetero-dimerization of the ErbB receptors is necessary for their activation via a characteristic asymmetric interaction of their kinase domains71 (Fig. 3). In the dimeric conformation, the C-lobe of the kinase domain of one of the receptors involved (the activator) binds a specific region in the N-lobe of the other (the receiver) via a conserved hydrophobic patch. This interaction induces an active conformation of the receiver, culminating in trans-phosphorylation of various tyrosine residues in the C-terminal tail of the activator. Binding of the juxtamembrane segment of the receiver receptor to the homologous segment and the kinase C-lobe of the activator further stabilizes the dimer72,73.

25 Notably, ErbB3, which has a reduced kinase activity due to substitutions of key residues in the catalytic pocket, can still act as an activator in the asymmetric dimer71. However, due to the inability of ErbB3 to act as a fully active kinase, the N-lobe sequence that stabilizes the interaction with the activator has diverged compared to the other members of the family. Thus, an arrangement in which ErbB3 would cover the receiver position is prevented. As a consequence of the activation by dimerization of the ErbB receptor, the multiple phosphorylations of the C-terminal tails generate docking sites for adaptor proteins containing SH2 or Phosphotyrosine Binding (PTB) domains, whose binding initiates downstream signalling cascades.

3.2 The ErbB receptors activate signalling cascades implicated in cell survival, differentiation and migration Eleven known ligands bind, with different selectivity, the ErbB receptors74. EGF, transforming growth factor α (TGFα), amphiregulin and epigen bind exclusively to EGFR; while heparin-binding EGF, betacellulin and epiregulin can bind both EGFR and ErbB4. ErbB4 is activated, as well, by neuregulin (Nrg) -1 to -4, while ErbB3 binds Nrg-1 and -2. The ErbB ligands can be synthetized either as transmembrane proteins (prone, however, to proteolytic cleavage into soluble forms), or as shorter secreted isoforms75–77, both of which can bind to the receptors. It has been reported that, despite binding to the same receptor, different ligands have the potential to induce different activation kinetics and divergent endocytic sorting of the bound receptor78. Ad- ditionally, the signalling outcome can also be affected by cleavage of the lig- and79. To further complicate the picture, although the pathways activated by the four ErbBs largely overlap, the particular identity of the dimerization partners can also modulate the triggered signalling80. As previously mentioned, the phosphorylation of the C-terminal tail of activated ErbB receptors promotes binding of adaptor proteins. Among them, Shc (SH2-containing Collagen-related protein) and Grb2 are of critical importance in recruiting the GEF Sos (Son Of Sevenless), which in turn activates Ras81,82. Activated Ras binds to Raf and relieves its autoinhibition, initiating the MAPK/ERK (Mitogen-Activated Protein Kinase/Extracellular signal– Regulated Kinase) pathway74. The MAPK/ERK pathway culminates in phosphorylation of transcription factors such as c-Fos (cellular-FBJ murine Oste- osarcoma viral oncogene homolog), which promotes entry into the cell cycle and induction of mitogenic genes83,84. Additionally, MAPK/ERK signalling stimulates cell migration-inducing cytoskeletal rearrangements85. Ras activation can also orchestrate cytoskeletal dynamics directly, by supporting the GEF-dependent activation of Cdc4286.

26 Another important branch of the ErbB-activated pathways is initiated by the binding and direct phosphorylation of Phospholipase C-γ (PLCγ)87. Phos- phorylated PLCγ then catalyses the hydrolyzation of PIP2 (phosphatidylinositol 4,5-bisphosphate) into two secondary messengers: Diacylglycerol (DAG) and Inositol 1,4,5-triphosphate (IP3)88. DAG subsequently stimulates PKC (Protein Kinase C) activity at the membrane. The activation of PKC is further potentiated by the release of intracellular calcium stores promoted by the binding of IP3 to dedicated receptors at the endoplasmic reticulum89. PKC, in turn, phosphorylates a number of substrates in the cell, and the best studied functional consequences of its activation are cytoskeletal remodelling and mi- gration90. Another branch of ErbB signalling that promotes cell migration is initiated by the direct binding of JAK to the receptors, and the consequent activation of the STAT-dependent transcriptional program91. ErbB receptors, and in particular heterodimers containing ErbB3, also promote the activation of the PI3K/Akt pathway. PI3K is recruited at the cellular membrane by direct binding of its regulatory subunit p85 to specific phosphorylated residues on the C-terminal tail of ErbBs92. The enzymatically active subunit of PI3K, p110, can then phosphorylate PIP2 to PIP3, generating a docking site for pleckstrin-homology domain-containing proteins93,94. Among them, Akt is the most important downstream effector of PI3K. The activation of Akt (whose mechanism has been described in section 2.3.2) results in the phosphorylation of a series of proteins involved in inhibiting apoptosis and activating proliferation. Akt inhibits apoptosis by phosphorylation of the pro-apoptotic protein BAD (Bcl-2-associated death promoter) and by promoting the nuclear translocation of Mdm2, which in turns ubiquitinates p53 and primes it for degradation 95,96. Additionally, Akt reprograms gene expression by inhibiting the Forkhead-box transcription factor FOXO and promoting degradation of the NF-κB inhibitor IκB, thus enhancing transcription of anti-apoptotic genes97,98. Moreover, Akt induces proliferation by mediating exclusion of the cell cycle inhibitor p27 from the nucleus, and by inhibiting Glycogen Synthase Kinase 3 (GSK3), which controls the cell cycle by down- regulating Cyclin D and E99,100. Altogether, the PI3K/Akt signalling pathway promotes survival and proliferation and suppresses apoptosis, accounting for its strong implications in cancer101. Collectively, the numerous signalling cascades triggered by ErbBs activation result in increased motility, decreased apoptosis and cell cycle progression, determining the pivotal roles of this family of receptors in development, physiology and disease.

27 3.3 ErbB-dependent signalling drives development and neoplastic transformation As previously mentioned, EGF – the founder ErbB ligand – was first discovered for its role in promoting eyelid opening and teeth eruption in mice65. Ever since, a large number of studies, fuelled also by the advent of genetic engi- neering, highlighted the importance of ErbB receptors in development. The deletion of either of the genes encoding ErbB2, ErbB3 and ErbB4 in mice is lethal in utero102–104. In particular, lack of ErbB2 causes abnormalities in cardiac development and defects in formation of sensory nerves derived from the neural crest102. ErbB2-/- embryos also suffer impairment in the development of motor nerves, and lack Schwann cells105. Analogously, the deletion of the ErbB4 gene results in aborted ventricular formation and generates al- tered nervous system phenotypes, in particular deficient innervation of the hindbrain104. Mice embryos lacking ErbB3 also display degeneration of sensory and motor neurons and lack of Schwann cells103. The outcome of EGFR knockout phenotypes, on the other hand, is dependent on the genetic background and results in death between mid-gestation and post-natal day 20106–109. When born alive, mice display severe phenotypes including defects in brain, skin, lungs, kidneys, liver and gastrointestinal tract. Moreover, EGFR knock-out mice also display placental defects, which cause death in utero in some strains107,108. The expression of dominant negative EGFR has additionally unveiled its role in post-natal development of the mammary gland110. In adult life, ErbB receptors have a critical pathophysiological role in a wide range of human cancers. In particular, EGFR has been found to be overexpressed or overactivated in a high percentage of lung cancers111. In non- small-cell lung cancers (NSCLC), EGFR overactivation is associated with active STAT3, ERK and PI3K/Akt pathways112,113. The most common mutation of EGFR spontaneously occurring in tumours is a deletion of exons 2 to 7, which generates a constitutively active truncated form named EGFRvIII114. Expression of EGFRvIII has been found to correlate with poor prognosis in glioblastoma, breast and lung cancers114–116. Additionally, activating mutations in the kinase domain have frequently been reported in NSCLCs and are important in the clinic, as they identify a subset of tumours responsive to therapy with the EGFR inhibitors gefitinib and erlotinib117–119. The constitutive activation of EGFR often occurs in tumours as a result of autocrine loops induced by overexpression of its ligands, which has been reported to occur in breast, lung, prostate and gastrointestinal tumors120–123. As the activity of ErbB2 does not depend on direct ligand binding, its overexpression alone is potentially oncogenic124,125. The tumorigenic role of the overexpression of ErbB2 has been extensively characterized in breast cancer, with an incidence of over 20% in malignant tumours, and it constitutes a widely recognized marker of poor prognosis and resistance to ER-directed

28 therapies126–128. However, its overexpression has also been detected in bladder, NSCLC and stomach cancers129–131. Interestingly, ErbB2 and ErbB3 are often co-overexpressed in cancer, and form a very potent mitogenic com- plex124,132,133. It has been reported that the overexpression of ErbB3 alone cannot drive tumorigenesis134. However, it becomes of critical importance when accompanied with dysregulation of other members of the family or transactivation by the receptor Met135. In particular, ErbB3 overexpression has been found in melanoma and colorectal cancers136,137. Increased ErbB3 expression is also a common feature in breast cancers, where it correlates with tumour size, metastasis and relapse138–140. Additionally enhanced activity of ErbB3 generates therapeutic resistance to ErbBs kinase inhibitors, chemotherapy and castration in prostate cancers141, suggesting the importance of targeting ErbB3 to increase the efficacy of anti-cancer therapy. The role of ErbB4 in human cancers has been less studied in comparison to the other members of the family. However, ErbB4 is overexpressed in colon cancer and its expression correlates with poor prognosis of breast cancer patients, regardless of ErbB2 expression status142,143. Interestingly, ErbB4-activating mutations have also been registered in NSCLC144. The overactivation of EGFR and other family members in various malignancies has prompted the development of drugs directed to block ErbB signalling, and a number of small kinase inhibitors and monoclonal antibodies have successfully been introduced in the clinic in the past years114,145,146. The effectiveness of these therapeutic approaches, however somewhat hindered by the emergence of resistance, highlights the importance of the study of ErbB receptors and their regulation in the fight against cancer.

3.4 A wide range of mechanisms modulate ErbB receptors-dependent signalling Given the physiological importance of ErbB receptors, tight control of the strength, duration and localisation of their activities is required to prevent high-risk aberrant behaviours of this system. Due to its primary role, most of what we know about the regulation of ErbB receptors comes from studies on EGFR, which will therefore be the focus of this section. The specific features of ErbB3 regulation will be instead addressed in section 3.5.

29 3.4.1 ErbB receptors signalling is spatiotemporally controlled via regulated trafficking An important mechanism that prevents uncontrolled activation of ErbBs signalling is the spatial segregation of receptors and ligands at the cell mem- brane147. In polarized epithelial cells, ErbBs are prevalently trafficked from the trans-Golgi network or recycling endosomes to the basolateral membrane, due to the presence of a sorting signal in the juxtamembrane domain148–150. Their ligands, however, are primarily exposed on the apical surface147,151. These physical constraints prevent autocrine stimulation under physiological settings. ErbBs activation to promote restoration of integrity is however al- lowed when the tight junctions (which preserve polarisation) are disrupted by an injury147. Notably, perturbations of the trafficking route of newly synthetized receptors to the appropriate membrane compartment causes transformation in epithelial cells151. Many factors contribute to limiting the signalling kinetics of activated ErbB receptors. The spatiotemporal control of ErbBs signalling is particularly tight at the level of endocytosis and intracellular trafficking of the receptors152. In absence of ligand stimulation, ErbBs are subjected to slow constitutive internalisation followed by rapid recycling, resulting in a prevalent membrane localisation of the receptors153,154. The rate of ErbB receptors internalisation is increased upon ligand binding and dimerization, which prevents high concentrations of activated receptors to stay in the plasma membrane and cause sustained signalling. Certain clathrin-independent mechanisms of receptor endocytosis have been proposed, and they seem to be particularly important in presence of high concentrations of ligand155–157. However, the regulatory process of ErbB receptors internalisation is thought to be predominantly mediated by CME152,158,159. Phosphorylation of EGFR has been considered for a long time to be required for localisation of EGFR in clathrin-coated pits, possibly due to the reported dependency on the interaction with Grb2159–164. However, it has been more recently suggested that it is the ligand-induced dimerization, rather than kinase activity, which is required for CME165,166 Additionally, the identity of the partner of dimerization influences the per- manence at the membrane. Overexpression of ErbB2, for example, favours the formation of ErbB2-EGFR over homo-dimers, thereby promoting sustained signalling167–169. This has been ascribed to both resistance of ErbB2 to endocytosis and its tendency to undergo rapid recycling170. As previously mentioned, Ack1 has been extensively implicated in the regulation of CME of EGFR. However, no clear conclusion regarding its mechanism of action in this context has been drawn yet18,19,171. Upon CME internalization, EGFR-containing clathrin-coated pits fuse with early endosomes. Interestingly, EGFR and its ligand do not dissociate at the mildly acidic pH of endosomes, preserving dimerization, association with

30 adaptor proteins and, consequently, signalling capacity of the complex172–174. The subcellular localisation of endosomal vesicles seems to be important to determine the specificity of signalling, since artificially perturbing their distribution in the cell modulates the activation of ERK 175. The EGFR-containing endosomes are either recycled to the plasma membrane or trafficked to the multivesicular bodies, from which they are directed to lysosomes for degradation or, once more, recycled at the membrane176–178. The ubiquitination of EGFR by Cbl (Casitas B-linage Lymphoma proto-oncogene), although not essential for endocytosis itself, seems to be crucial for the subsequent sorting steps and promotes receptor degradation over recycling at the membrane179,180. Interestingly, it has been demonstrated that the choice of whether EGFR traf- fics towards the lysosomal route rather than being recycled can depend on the identity of the activating ligand78. However, the details of this mechanism have not been elucidated yet. Finally, it has more recently been reported that Ack1 can direct EGFR to yet another pathway, by promoting its degradation at the autophagosome21.

3.4.2 ErbB signalling can be attenuated by competitive binding, phosphatases and adaptor proteins Other mechanisms exist that attenuate ErbB receptors signalling without in- terfering with their trafficking. Among those, a series of secreted splice vari- ants (the EGFR Related Peptide ERRP, herstatin and P85s, transcribed, respectively from the EGFR, ErbB2 and ErbB3 genes) retain the ability to di- merize with the ErbBs, sequestering them at the membrane, inhibiting their activation and, consequently, hampering cell proliferation181–184. Upon activation of the ErbB receptors, the propagation of the signalling cascade can be blocked by dephosphorylation of the C-terminal tails of the receptors. A number of phosphatases have been identified that are involved in regulating ErbB signalling. Moreover, these phosphatases can act both at the cell membrane and at different steps of the internalisation process, suggesting that their differential action could modulate receptor-mediated signalling in different ways185. ErbB-dependent signalling can also be downregulated by ubiquitination and degradation the receptors. This mechanism can be mediated both in Cbl- dependent (as explained in the previous subsection), and -independent man- ners. Ubiquitination by Cbl is enhanced at the membrane by the binding of the receptor to Lrig1 (Leucine-Rich repeats and Immunoglobulin-like domains 1), whose expression is induced by EGFR stimulation186,187. Lrig1 acts therefore in a negative feedback loop, to switch off the pathways activated by EGFR. Ubiquitin-dependent degradation of EGFR can also be stimulated by SOCS (Suppressors Of the Cytokine Signalling), which binds the phosphorylated receptor and promotes its association with E3-ligases other than Cbl188–190. It has

31 been shown that SOCS and STAT3 compete for binding to the same site on the activated receptor, suggesting that SOCS might additionally dampen EGFR signalling by displacing STAT3 association189. Moreover, EGFR can be inhibited by scaffolding proteins such as FRS2β and Mig6. FRS2β (Fibroblast growth factor Receptor Substrate 2β) binds EGFR independently of the presence of ligand, and recruits phosphorylated ERK to form a ternary complex that is signalling-inactive191. This system has been proposed to act both as a negative feedback mechanism and to prevent EGFR signalling when ERK is phosphorylated by other pathways. Addition- ally, FRS2β has been reported to regulate ErbB2, via a different mechanism that induces degradation of the receptor192. Mig6 (also known as RALT for Receptor Associated Late Transducer) can also associate with EGFR, ErbB2 and ErbB4, and, by doing so, suppresses their signalling193,194. Binding of Mig6 to the kinase domain of the ErbBs inhibits dimerization and, therefore, catalytic activation of the receptors195. Ad- ditionally, Mig6 can induce endocytosis and degradation of the receptor by scaffolding the interaction with AP-2196. The importance of Mig6-dependent regulation of ErbBs-signalling in development and tumour formation has been highlighted by the severe phenotype caused by its deletion in mice197. Most of the studies on the regulation of ErbB signalling have been conducted on EGFR and, less frequently, ErbB2. However, due to the strong contribution of all the members of the family to development and cancer, mechanisms of ErbB3- and ErbB4-specific regulation should not be overlooked. In the work presented in this thesis (Paper III), we report a novel mechanism of ErbB3 regulation involving the Ack1 kinase, that is not shared with the other members of the family. Therefore, as a frame of reference, the distinctive traits of ErbB3 function and regulation will be presented in the following section.

3.5 The catalytically-impaired ErbB3 holds signalling capacity and dedicated mechanisms of regulation Due to its impaired kinase activity, ErbB3 signalling is mediated by association with other members of the family. This can lead to the wrong assumption that ErbB3 regulation is always dependent on the mechanisms involving the kinase-competent members of the dimer. However, this view is challenged by the existence of many ErbB3-specific regulatory systems that are starting to be unveiled. ErbB3 is a heavily glycosylated 180 kDa receptor protein that binds the different isoforms of Nrg-1 and -276. The most striking characteristic of ErbB3 compared to the other ErbBs is, as already mentioned, its impaired catalytic activity198. This difference has been attributed to the substitution of important

32 residues in the catalytic pocket that are conserved in tyrosine kinases199. How- ever, more recent studies reported the possibility of residual enzymatic activity of ErbB3200,201. By virtue of the dimeric mode of activation of ErbBs, however, the lack of a functional kinase in ErbB3 would not hinder its signalling potential. Instead, phosphorylated ErbB3 is actually the most potent of the ErbBs at activating the PI3K/Akt pathway. This is due to the presence of six docking sites for the p85 regulatory subunit of PI3K202,203. Additionally, its important roles in development and tumour formation (mentioned in section 3.3) demonstrate the importance of ErbB3-dependent signalling. As a result of substitutions of critical amino acids at the N-terminal dimerization surface, ErbB3 was thought to be the only ErbB unable of homodimer- ization71,204. However, homodimerization and formation of homo-oligomers of ErbB3 has been reported195,201,205. In particular, the formation of high-order molecular association of ErbB3 alone or together with other members of the family has been recently proposed to contribute to a non-canonical non-dimerization-dependent mode of activation of the ErbB receptors, which will need further studies to be fully understood206. ErbB3 is the member of the ErbB family that displays the shortest half- life152. Its fast turnover depends both on constitutive endocytosis leading to lysosomal degradation, and on a proteasome-dependent system of quantity control of newly synthetized ErbB3207–209. The latter mechanism, in particular, is unique in the ErbB family and involves ubiquitination of newly produced and correctly folded ErbB3 by the RING finger E3-ligase Nrdp1 (Neuregulin Receptor Degradation Protein-1)209. The purpose of this level of control of ErbB3 and the molecular players involved in its regulation are, however, still unclear. An early study reported ErbB3 to be endocytosis-impaired upon ligand binding, due to the lack of interaction with AP-2 of a EGFR/ErbB3 chimeric protein167. More recent reports have, however, tracked ErbB3 endocytic pathways under more physiological settings, revealing a different scenario. ErbB3 can be internalised in a clathrin-dependent way and degraded mainly via the lysosomal pathway, both in ligand-dependent and -independent man- ners207,208,210. Despite its inability to bind AP-2, ErbB3 can undergo CME via interaction with the adaptor epsin 1, which in turn mediates recruitment of the receptor to clathrin-coated pits210. Recently, the importance of clathrin-independent internalisation of ErbB3 has emerged, and has interestingly been linked with trafficking to the nucleus211. Localisation in the nucleus of full-length EGFR and ErbB2 and both full- length and cleaved forms of ErbB4 has been reported212. Nuclear ErbB receptors have been shown to activate transcription and contribute to oncogenic potential and resistance to therapy of cancer cells212. Full-length ErbB3 has also been found in the nucleus, and it was first proposed that ligand stimulation would decrease nuclear localization and target the receptor at the mem- brane213. Later reports, however, describe neuregulin-induced translocation of

33 phosphorylated ErbB3 in the nucleus, where it associates with the transcription factors STAT3, STAT5 and E2F1 and the transcription machinery, suggesting an important role in the regulation of gene transcription211. Despite the existence of studies linking the regulator Ack1 to EGFR, the former has not been associated with ErbB3 before. In this thesis (Paper III), we present a role for Ack1 in regulating ErbB3 at multiple levels, independent on EGFR or ErbB2, unveiling its new dedicated role in modulating of ErbB3- dependent signalling.

34 4. Mechanical regulation of Yap and Taz controls cell fate

4.1 Mechanical cues are crucial messengers in the regulation of cellular functions

Beside receptor recognition of soluble signals, cells can exploit other senses to probe their surrounding and adapt to change. In the last years the importance of mechanical cues in directing cell behaviour has raised increasing interest within the field of molecular signalling. The concept of an intersection between mechanical forces and biology is however not new. A little more than a century ago, D’Arcy Thompson’s seminal book On Growth and Form proposed that a “diagram of forces” underlies biological shapes and directs development and cell patterns214. Thompson’s theories attracted new attention a few decades later, with the demonstration that cancer transformation modifies cell adaptability for growth on soft substrates215–217. The subsequent development of devices to measure and modify the mechanical microenvironment at the subcellular level lead to the discovery that cells actively test their sur- roundings for mechanical resistance218, highlighting its importance for regulation of cell function. Cells in the body are indeed exposed to a variety of mechanical cues, that originate from the stiffness of the extracellular matrix (ECM), the contraction of neighbouring muscle cells, shear stress caused by blood flow on endothelial cells, or the strain experienced by infiltrating immune cells when they intrav- asate into tissue. A vast amount of studies has now established that cells sense these mechanical stimuli, transduce it into biochemical messengers, and respond by modifying their behaviour, in a process that goes under the name of mechanotransduction219–221. Far from being a negligible aspect, the ability of cells to respond to these forces has decisive consequences in controlling cell fate. In particular, mechanical stimuli have been proven to be important in determining the rates of cell proliferation, in directing mesenchymal and hem- atopoietic cell differentiation, and even in defining proper organ architecture at the developmental level220–227. As a result of technological advances that allow the studies of mechanical changes at the molecular level, the study of mechanotransduction is experi- encing a rapid expansion in the last years228. Despite the exponential growth

35 of studies on the topic, several critical questions remain to be answered to untangle the molecular details of mechanical-driven signalling and their physiological and pathological implications. Two proteins in particular have gained the spotlight for their ability to integrate many diverse mechanical stimuli into gene expression regulation: Yap and Taz. Several studies have corroborated the importance of Yap and Taz in responding to mechanical cues for lineage determination of stem cells and their dysregulation in many pathological conditions, particularly in cancer221. However, the mechanisms underlying the transduction of physical forces into control of the function of Yap and Taz are still largely unknown. In the work presented in this thesis (Paper II), we identify Ack1 as a novel actor in mechanotransduction, which regulates Yap and Taz subcellular localisation and stability downstream mechanical signals. To put our findings in context, the following section (4.1.1) will review the most understood mechanisms of mechanosensing. Thereafter, the focus will be shifted to the regulation of Yap and Taz and its importance in mechanobiology (4.2).

4.1.1 Cells sense mechanical signals

Cells detect mechanical cues via primary mechanosensors: structures whose change of conformation in response to physical forces determines a functional activation. Such structures must therefore be reactive to the extracellular mechanical microenvironment (such as stiffness of the extracellular matrix and pulling or pushing forces exerted by neighbouring cells or spatial constraints) or to intracellular shape changes (such as cellular swelling or cytoskeletal rearrangements). For these reasons, membrane proteins, junctional complexes and determinants of cell architecture make good candidates as sensors of physical stimuli. The fastest response to mechanical stimulation of the cell comes from the gating of mechanosensitive ion channels229. This diverse group of channels is stabilised in the open conformation by a local increase of membrane tension, generating a flux of ions across the membrane. As a result, mechanical perturbations are quickly translated into changes in the electrical potential or the vol- ume of the cell. Changes in stiffness of the extracellular microenvironment are sensed by integrins, transmembrane proteins which act as receptors for components of the ECM230,231. The binding of integrins to their ligands nucleates the formation, on the cytoplasmic side, of multimolecular structures named focal adhesions (FAs)232. FAs act both as bridges between the ECM and the cytoskeleton and as signalling platforms, via the recruitment of tyrosine kinases. It has been demonstrated that force loading reinforces the association of integrins to their substrate, by inducing a conformational change that increases the affinity of binding230,231. In parallel, mechanical stimulation enhances the association

36 of FA components, such as focal adhesion kinase (FAK), talin, vinculin, pax- illin and p130Cas, resulting in activation of signalling233–236. This increased protein recruitment is mediated by mechanical unfolding of some of the FA components, which promotes the exposure of phosphorylation sites (as in the case of p130Cas, whose stretching unmasks tyrosine residues for Src phosphorylation) or motifs for binding to other proteins (such is the case of talin, whose extension promotes the binding to vinculin)237,238. Talin and vinculin, at the FA, are actin-binding proteins that connect the ECM to the cytoskeleton, allowing force transmission along the cell body. Interestingly, it has been recently shown that integrins can respond to mechanical cues independently of their binding to ECM, via tension-induced association with other activated integrins or vitronectins239,240. Cell-to-cell junctions constitute one more hub for mechanosensing and transduction241. Adherens junctions (AJs), in particular, form by assembly of homophilic interaction of cadherin at the plasma membrane of two neighbouring cells. Each cadherin then recruits, at the cytoplasmic side, β- and α-caten- ins, which in turn couple the AJ to the actin cytoskeleton. Force application modulates AJs strength, indicating a role in mechanically coupling adjacent cells241–243. Similarly to the mechanism of FA reinforcement, mechanical tension unfolds α-catenin, exposing vinculin-binding sites and promoting association to the cytoskeleton. The cytoskeleton is itself a crucial bidirectional actor in mechanotransduction, implicated in both the generation and the sensing of forces. Actin polymerization and myosin II-mediated contractility dynamically modify the mechanical properties of the cell244. Moreover, the actin cytoskeleton not only integrates mechanical signals coming from cell-to-cell and cell-to-ECM adhesion complexes, but it is also a bona fide mechanosensor that rearranges its architecture by recruiting modifying proteins in response to force load- ing245,246. Interestingly, in addition, the ratio of G- (monomeric) to F- (pol- ymerized) actin determines the translocation of G-actin-binding transcription factors, such as MRTF, to the nucleus247. The mechanisms by which force-dependent activation of the primary sensors is transduced into transcriptional responses is not always clear, and varies with the nature of the stimulus and the responding factors215,219. However, general mechanisms that exploit the mechanical properties of the nucleus are also emerging219. For example, mechanical cues can be directly transmitted to the nucleus via the LINC (Linker of Nucleoskeleton and Cytoskeleton) complexes that physically connect the cytoskeleton with the nucleoskeleton. Indeed, cell growth on substrates of different mechanical properties has been shown to influence both the nuclear shape and its stiffness, via regulating nucleoskeleton and LINC-interacting proteins219. These changes have surprisingly been linked to spatial rearrangement of chromosomes, which allows the concentration of gene clusters resulting in modification of transcriptional patterns219,248.

37 Additionally, more specific mechanoresponses can be triggered by controlling the nuclear concentration of transcriptional regulators. Not unlike cascades initiated by receptor-activation, mechanical signals can result in post- translational modification of selected proteins, determining their retention in the cytoplasm, nuclear transport, or modifying their stability. All these mechanisms have been found to be important in the regulation of Yap and Taz, two key players in mechanotransduction, as described in the following section (4.2).

4.2 Yap and Taz are mechano-responsive transcriptional co-factors 4.2.1 Yap and Taz’s gene transcription signature directs cell fate Yap (Yes-Associated Protein) and Taz (Transcriptional co-Activator with PDZ-binding motif) are homologous transcriptional regulators that have been extensively connected to promotion of cell survival, proliferation and differ- entiation221,249–251. The two proteins are at least partially redundant in function and are often studied together, therefore they will be often referred in this thesis as Yap/Taz. It is however worth mentioning that they display structural differences and different regulatory mechanisms (see sections 4.2.2 and 4.2.3), suggesting that their roles might not be completely overlapping252–254. Knock- outs of these two proteins in mouse have indeed very different outcomes. Yap deletion is lethal in utero at embryonic day 8.5, due to defective vasculogenesis of the yolk sac255. Taz-knockout mice, even if born at less than the expected mendelian ratio, are instead viable but develop minor skeletal abnormalities, a more severe kidney phenotype reminiscent of glomerulocystic disease, and lung defects similar to human emphysema256–258. The main cellular function of Yap and Taz resides in their ability to promote gene transcription by association with transcription factors of the TEAD (TEA Domain protein) family259,260. The importance of this interaction is confirmed by the observation that Yap/Taz mutants unable to bind to TEAD fail to induce proliferation and tumour formation260–262. Yap/Taz can, however, partner with many other transcription factors such as RUNX (Runt-related transcription factor), p73, the cleaved intracellular domain of ErbB4 and SMADs (Sma Mother Against Decapentaplegic)263–266. Interestingly, Yap/Taz can also inhibit the transcriptional activity of certain binding partners. Yap/Taz can in fact sequester SMAD2/3 at the Crumbs polarity complex in a cell-density-dependent manner. Moreover, Taz inhibits the transcriptional activity of PPARγ (Peroxisome Proliferator-Activated Receptor γ), despite not diminishing its nuclear localisation or binding to the aP2 promoter267,268.

38 Altogether, Yap/Taz binding to different transcription factors regulates the expression of over 400 genes, collectively promoting a pro-growth and anti- apoptotic signature221,269.

4.2.2 Yap and Taz are structurally similar, yet not identical As briefly mentioned, despite high sequence and structural similarity, Yap and Taz display distinctive features that should not be overlooked (Fig. 4)249. The affinity to TEAD, for example, is mediated by a conserved N-terminal domain of interaction. Structural modelling, however, underlined that the critical TEAD-interacting residues are not the same in Yap and Taz, accounting for a putative difference in regulation of their association254. Partially overlapping with the TEAD domain, Yap/Taz display a consensus sequence for phosphorylation-dependent binding of members of the 14-3-3 family, which have been shown to be involved in their cytoplasmic sequestration270–272. The WW domains (two in the most common isoform of Yap, one in Taz) are also of fundamental importance for the function and regulation of Yap/Taz. The WW domains mediate association with most of the known upstream regulators (Lats, AMOTs, β-catenin) and transcription factors (RUNX, the cleaved intracellular domain of ErbB4, p73, SMAD1) that interact with Yap/Taz. An ex- ception to this second group is SMAD2/3, which binds instead to a different coiled-coil structure266. Additionally, Yap/Taz both contain a poorly struc- tured transactivation domain (TAD) and a PDZ-binding domain at the C-terminus. The PDZ-binding domain mediates interaction with ZO-2, which has surprisingly been shown to regulate nuclear localisation of Yap and Taz in opposite directions252,273,274. A striking structural difference between the two proteins resides in the proline-rich sequence at the N-terminus of Yap, which is missing in Taz. This region has been reported to be a binding surface for the heterogeneous nuclear ribonuclear protein U (hnRNPU), which inhibits Yap- dependent transcription275.

Figure 4. Structural similarities and differences between Yap and Taz, based on Varelas, 2014249 and Piccolo et al., 2014250. The main functional domains of Yap and Taz are represented. The yellow vertical lines indicate the position of the serine (pS) and tyrosine (pY) sites of regulatory phosphorylation. Pro rich stands for proline- rich; BD for binding domain; pDeg for phosphodegron. For other abbreviations see text.

Yap/Taz function is governed by a number of different upstream pathways, which act via two main mechanisms: regulation of nuclear localisation and modulation of protein stability221,250,251, as explained in the following section.

4.2.3 The regulatory network of Yap and Taz is dominated by mechanotransduction Yap/Taz were first studied for their homology with Yorkie, effector of the Hippo pathway, which controls tissue overgrowth and organ size in Drosoph- ila. The core Hippo pathway is a widely conserved kinase cascade consisting, in human, of Mst1/2 (Mammalian Ste20-like serine/threonine kinase, Hippo in Drosophila), Lats1/2 (Large Tumour Suppressor kinase) and the two adaptor proteins SAV1 and MOB (Salvador 1 and Mps One Binder protein, respectively), which promote interaction of the kinases272,276,277. The activation mechanism of the mammalian Hippo pathway is still poorly understood278. However, it has been shown that its activation is enhanced by the formation of polarity complexes that recruit the upstream kinase Mst1/2 to the plasma membrane272,279,280. Once activated, Mst1/2 phosphorylates Lats1/2 at key serine residues, triggering its activation. Active Lats1/2, in turn, phosphorylates Yap and Taz at serine residues (five and four, respectively) enclosed in HXRXXS motifs272,277. Phosphorylation of one of these residues in particular (S127 in Yap and S89 in Taz) generates a recognition site for 14-

40 3-3s, whose binding will anchor Yap/Taz in the cytoplasm271,272,277,281. Inter- estingly, the S127 site in Yap can also be phosphorylated by Akt and has been linked to an anti-apoptotic outcome via the inhibition of the interaction of Yap with p73271. Lats1/2-dependent phosphorylation also targets a conserved serine enclosed in the TAD (S311 in Taz and S407 in Yap), which is part of a phosphodegron motif. This modification primes for further phosphorylation by CK1ε/δ (Caseine Kinase 1 ε/δ) a neighbouring serine, fostering recruitment of β-TrCP (β-Transducin repeat-Containing Protein), a E3 ubiquitin ligase that will in turn ubiquitinate Yap/Taz and promote their proteasome dependent degradation. Interestingly, Taz contains at the N-terminal a second phosphodegron, which promotes β-TrCP-dependent ubiquitination downstream GSK3 phos- phorylation253. This additional degradation pathway seems, however, to be Hippo-independent, and might account for the decreased stability of Taz over Yap. The recruitment of β-TrCP downstream of GSK3-dependent phosphorylation is best known for its importance in triggering β-catenin degradation, a process in which Yap/Taz have been recently implicated. In unstimulated cells, β-catenin is sequestered in the cytoplasm by the formation of the so- called destruction complex, which includes also axin, APC (Adenomatosis Polyposis Coli), CK1 and GSK3. The destruction complex phosphorylates β- catenin and promotes its proteasomal degradation via β-TrCP-mediated ubiq- uitination282. Upon Wnt stimulation the complex dissociates, allowing translocation of β-catenin in the nucleus, where it promotes gene expression. Yap/Taz have been found to take part in the destruction complex and favour β-TrCP recruitment, which in turn promotes both β-catenin and Taz degrada- tion283,284. Epithelial polarity components can regulate Yap/Taz both by activating the Hippo pathway and in Hippo-independent ways. The apical polarity complex Crumbs, for example, sequesters Yap/Taz at the plasma membrane of polar- ised cells267. Additionally, the Crumbs-associated proteins angiomotins (AMOTs) have also been found to direct Yap/Taz localisation. However conflicting data have been published on this matter. While all AMOTs were found to promote Yap/Taz cytoplasmic and cytoskeleton-associated localisation, p130AMOT has also been identified to co-localise with Yap in the nucleus285– 289. At the same time, AMOTs have been found to associate with Yap and Hippo kinase proteins and promote their phosphorylation, suggesting a complex role for AMOTs in Yap/Taz regulation290. The basolateral polarity determinant Scribble can also promote association and activation of Hippo pathway components and its displacement from the membrane was reported to promote Yap/Taz contribution to epithelial-to-mesenchymal transition (EMT) of cancer cells291.

41 Yap/Taz can also be sequestered at the membrane by binding components of the AJs and the importance of these interactions has been confirmed in vivo, as the presence of α-catenin is crucial to prevent aberrant Yap activation and tumour formation in the epidermis292,293. In addition, Yap/Taz localisation can be also controlled by activation of ligand-activated receptors. The family of G-protein coupled receptors (GPCRs), for example, encodes for both positive and negative regulators of Yap/Taz activity. Binding of lysophosphatidic acid or sphingosine 1-phosphate to their respective GPCR, for example, promotes nuclear localisation of Yap/Taz by inhibiting Lats1/2 activity294,295. In contrast, hormone activated- GPCRs, such as the receptors for glucagon and epinephrine which are coupled 294 to GS proteins, increase Lats1/2-dependent phosphorylation of Yap/Taz . Moreover, EGFR signalling inhibits Lats1/2-dependent phosphorylation of Yap and promotes its expression via the PI3K/PDK1 pathway296,297. Additionally, the metabolic activity of the cell can impact Yap/Taz activation. Increase in glycolytic activity commonly observed in many cancer cells has indeed been found to promote, via phosphofructokinase 1, the interaction between TEAD and Yap/Taz, enhancing their nuclear translocation and transcriptional activity298. Additionally, Yap levels are increased by the blockage of the mTOR-induced autophagic program, revealing a mechanism by which nutrient scarcity opposes Yap activation299. Moreover, Yap/Taz can be phosphorylated at a conserved tyrosine residue in the TAD (Y321 in Taz and Y397 in Yap) by Abl, Src and Yes kinases300– 302. Tyrosine phosphorylation by this set of kinases can be promoted by different activating signals such as DNA damage and osmotic shock. Conversely to what was observed for serine phosphorylation, tyrosine phosphorylation does not prevent nuclear localisation nor impairs protein stability, but seems instead to have a role in directing transcription factor selectivity. In particular, tyrosine-phosphorylated Yap/Taz have increased affinity for RUNX and NFAT5 binding, with negative effects on their transcriptional activity, and for association with p73, and subsequent expression of pro-apoptotic genes300–302. Intriguingly, in the last years the importance of mechanical cues in regulating Yap/Taz has gained increasing recognition221. Conditions that induce ten- sional stress on the cell, such as growth on stiff substrates, spreading on large surfaces or stretching, promote Yap/Taz translocation to the nucleus303. Con- versely, soft growth matrices, a confined growth area or cell crowding all con- fine Yap/Taz to the cytoplasm. The integrity of the actin cytoskeleton is determinant for transduction of the signal. In fact, Yap/Taz localise in the cytoplasm regardless of the application of mechanical forces when actin is force- fully depolymerised or when cytoskeletal integrity is compromised such as by depletion of actin-capping proteins or by inhibition of ROCK (Rho-associated Coiled-coil containing Kinase)303–305. Cytoskeleton-transduced signals seem indeed to dominate over other regulatory pathways, constituting a central checkpoint directing Yap/Taz localisation. Consistently, modulators of Rho

42 activity, such as the geranyl-geranylation step, necessary for its targeting to the membrane, influence Yap/Taz activity as well306,307. How the mechanical message is transduced from the actin cytoskeleton to Yap/Taz is still an open question. In some studies, cytoskeletal rearrangements were found to regulate Yap/Taz in a phosphorylation-dependent manner, via the activation of the Hippo pathway or direct downstream engagement of Lats1/2308–312. However, in other settings, Lats1/2-dependent Yap/Taz phosphorylation was not sufficient to override the cytoskeletal-mediated effect, suggesting the existence of two branches (a Lats1/2-dependent and a Hippo- independent one) of Yap/Taz regulation downstream of mechanical sig- nals303,305,313. Importantly, the mechanical responsiveness of Yap/Taz has been connected to a number of biological processes221,250,278.

4.2.4 Mechanical regulation of Yap and Taz governs cell fate The mechanical control of Yap/Taz activity has important biological implica- tions221,250,278. Yap/Taz were early on linked to the establishment of contact inhibition of proliferation (CIP), the mechanism that arrests proliferation upon cell crowding272. CIP was first observed in cell cultures as early as 50 years ago, but it is nowadays considered of outstanding importance in vivo, where it prevents pathological overproliferation and is modulated to restore tissue integrity in wound healing1,314. Yap/Taz sequestration in the cytoplasmic compartment is required for CIP, and this process was initially thought to be fully dependent on the Hippo pathway272. However, despite the fact that CIP is accompanied by phosphorylation of Yap/Taz, it is reinforced by mechanical Hippo-independent signalling and can be overridden by increasing mechanical tension of the cells305,315,316. In the body, the activity of Yap/Taz is generally restricted to progenitor cells and healing tissues, suggesting a role in promoting renewal capacity of adult stem cells221,250. In particular, nuclear localization of Yap/Taz is typical of progenitor cells of the intestine crypts and is reportedly important for intestine regeneration after injury317,318. Similarly, Yap levels were crucial for tissue repair upon myocardial infarction319,320. Their importance in this last system has been ascribed to Yap’s ability to direct the response of the cardiac progenitor to changes in the stiffness of the ECM, as occurring in result of an injury321. Yap/Taz function has been linked to the dictation of lineage commitment of mesenchymal stem cells (MSC). MSC differentiation is a classic example of the importance of the mechanical microenvironment in guiding cell fate. In fact, it has been observed that the stiffness of the growth matrix, the shape of the adhesion surface, and manipulation of cell contractility, all can direct differentiation into specific cell types226,322. Interestingly, Taz is fundamental for osteogenic commitment of MSC, and Yap/Taz localization responds to mechanical stimulation of these cells268,303,311.

43 Additionally, mechanical regulation of Yap/Taz is observed already during embryonic development. Differential localisation of Yap/Taz occurs as early as at the blastocyst stage, where the external cells, which will form the trophoblast, have nuclear Yap/Taz, as opposed to cells in the inner cell mass that will generate the embryo tissues323. Importantly, it has been shown that the contractility of the blastomers determines their specification as inner cell mass rather than trophoblast cells, by controlling the localization of Yap independently of their position in the blastocyst324,325. Beside their physiological roles, Yap/Taz has been found to be overactivated in many cancers, such as lung, breast, liver, colorectal tumours and gliomas250,326. Notably, the activation of Yap/Taz and the transcription of their target genes correlate in many instances with tumour progression and poor prognosis. The importance of Yap/Taz in tumour formation is explained by their ability to promote transcription of pro-proliferative and anti-apoptotic genes326. The activation of Yap/Taz can additionally foster self-renewal capa- bility of tumour cells, enriching the pool of cancer stem cells, a subpopulation characterized by chemoresistance and ability to form highly aggressive undif- ferentiated tumors291,327. Interestingly, there is no report of Yap/Taz activating mutations in cancers and their overactivation in tumours is rarely accompanied by mutations in the Hippo-pathway221. On the other hand, cancer is accompanied by mechanical anomalies caused by tissue architecture perturbation, inflammation and ECM deposition by the cancer associate fibroblasts328. The mechanical tumour microenvironment is therefore subject to compression forces, elevated interstitial pressure and stiffening of the ECM. This has led to the suggestion that the switch of Yap/Taz reactivation in tumour could reside in these mechanical changes250. As the molecular details of Yap/Taz regulation downstream mechanical forces are still unclear, the quest for mechanotransduction players in this context is still open. In the study hereby presented in Paper II, we identified for the first time a role for Ack1 as an actor in this process. We discovered that Ack1 is activated by changes in mechanical tension. Moreover, we found that this activation promotes the confinement of Yap/Taz in the cytoplasm and Taz degradation in a Hippo-independent manner. This discovery sheds light on the currently poorly understood mechanism of control of Yap/Taz function downstream mechanical stimuli. Taking into consideration the presented data, this could have important implications in our understanding of the way by which Yap/Taz dysregulation plays a role in cancer.

44 5. Apoptosis: the regulated cell death program that controls tissue stability

5.1 Apoptosis can be triggered by both internal and external cellular signals Appropriate development and maintenance of homeostasis rely on the possibility to discard damaged or unnecessary cells by regulated cell death (RCD). This process can take place in different ways and a classification based on morphology and effector killer mechanisms recognises three groups of RCD: apoptosis, autophagy and necrosis329. The distinctive traits of apoptosis include the involvement of a distinct group of proteases (caspases), chromatin condensation (pyknosis), nuclear fragmentation (karyorrhexis) and membrane blebbing329,330. The apoptotic process culminates in the dismemberment of the cell into apoptotic bodies, small vesicles with intact membrane. Exposure on the external side of the cell membrane of the phospho-lipid phosphatidylserine, which normally faces the cytoplasm, constitute an attraction signal for macrophages, which clear the apoptotic bodies in a process named efferocytosis331. In vitro, in absence of phagocytic activity, apoptosis culminates instead in secondary necrosis332. Apoptosis plays a fundamental role in defining tissue architecture during development and it keeps a pivotal homeostatic role in the adult organism. Over- active apoptosis is an etiological agent in infertility, immunodeficiency and degenerative diseases, while insufficient apoptotic response can promote autoimmune response and cancer development329,333. Apoptosis can be triggered by two major activation pathways: intrinsic, or mitochondrial, and extrinsic, or receptor-mediated. Despite the distinct activation mechanisms, both pathways converge at the level of the effector caspases. Caspases are cysteine proteases with specificity for aspartate-containing target sites (the name “caspase” derives from “cysteine aspartases”) that are expressed as inactive zymogens334. Initiator caspases are differentially activated by the intrinsic or extrinsic pathways by their recruitment at specific protein complexes where they get trans-activated by oligomerisation. Initiator caspases can thus proteolytically activate effector caspases (3 and 7) that, in turn, process critical proteins leading to the aforementioned apoptosis hallmarks and, eventually, cell death335,336. Crosstalk between the mitochondrial

45 and receptor-mediated apoptosis pathways exist and, in some cell types, it is necessary for efficient cell death induction337,338.

5.1.1 The intrinsic pathway of apoptosis is executed via permeabilization of the mitochondria

The intrinsic apoptosis pathway (Fig. 5) is triggered by a variety of stimuli including growth factor deprivation, endoplasmic reticulum stress, oxidative stress and mitotic defects329,333,339–342. Regulation of the intrinsic apoptosis pathway is orchestrated by proteins of the BCL-2 (B-cell lymphoma 2) superfamily, a group of more than 20 pro- and anti-apoptotic proteins containing BH (BCL-2 homology) domains342,343. The initiation of the pathway is mediated by pro-apoptotic BH3-only proteins such as PUMA and BIM, which get transcriptionally upregulated in response to cellular stress. Increased levels of initiator BH3-only proteins promote their localisation at the outer mitochondrial membrane, where they interact with the effector BCL-2 proteins BAX and BAK. This transient interaction blocks BAX retro-translocation to the cytoplasm and induces structural rearrangements in both BAX and BAK, thereby promoting their association in homo- and hetero-dimers and subsequent higher order oligomerisation344– 348.

Figure 5. Schematic view of the activation of the intrinsic pathway of apoptosis. Casp- stands for caspase, see the text for other abbreviations.

46 BAX/BAK oligomers form pores that lead to mitochondrial outer membrane permeabilization (MOMP) and release in the cytoplasm of cytochrome c and other pro-apoptotic proteins such as SMAC (second mitochondria-derived activator of caspases). Once in the cytoplasm, cytochrome c associates with the apoptotic peptidase activating factor-1 (APAF-1) in a dATP- dependent manner and subsequently to Pro-caspase 9349. The resulting complex, called apoptosome, promotes dimerization and self-activation of the initiator caspase-9 and the subsequent fatal proteolytic cascade. The process is facilitated by cytoplasmic SMAC, which binds to XIAP (X-linked inhibitor of apoptosis) pre- venting it from inhibiting the effector caspases350,351. Inhibitor members of the BCL-2 family can antagonise apoptosis activation by binding to BH3-only members, sequestering BAK and BAX at the mitochondrial membrane or promoting BAX retro-translocation to the cyto- plasm342,343,352.

5.1.2 The extrinsic pathway of apoptosis is triggered by receptors at the plasma membrane Apoptosis can also be triggered by membrane receptor-mediated signalling. There are two groups of apoptosis-inducing receptors that act in opposite fash- ions: the dependence receptors, which trigger apoptosis when deprived of lig- and, and the death receptors, which are activated by ligand binding to induce apoptosis329. Dependence receptors are a diverse group of approximately 20 trophic factor receptors including netrin-1 receptors, TrkA and C, Alk, the insulin receptor and some integrins329,353. The activation mode of dependence receptors usually includes cleavage of their intracellular domains and involvement of caspases. However, the molecular details on dependence receptor-activated apoptosis are still unclear and might vary with the receptor type. For the purpose of this thesis, we will focus on death receptor-induced apoptosis and refer exclusively to this subtype when discussing extrinsic apoptosis353. Death receptors (DRs) are the apoptosis-inducing members of the tumour necrosis factor receptor (TNFR) superfamily. The most studied – and potent – members of the family are its founder TNFR1, the first apoptosis signal receptor (Fas-R, also known as APO-1 for apoptosis Antigen-1 or CD95 for Cluster of Differentiation 95) and the TNF-related apoptosis-inducing ligand receptor (TRAIL-R) 1 and 2 (also, respectively, DR4 and 5). DRs are transmembrane proteins displaying several cysteine-rich domains on the extracellular region and a characteristic 80 residues-long adaptor domain on the intracellular side that is critical for recruitment of downstream proteins, known as the death domain (DD). The TNRFR family also encloses non-apoptosis-inducing members, which are membrane-bound or soluble receptors that lack the DD and thus act as inhibitors of the extrinsic pathway by competitive binding to the ligand329,333,354.

47 The DRs bind death-inducing ligands that are usually expressed as transmembrane proteins, but can be proteolytically cleaved and released in a soluble form355–357. However, differences in the potency of apoptosis activation might exist between the membrane-bound and soluble forms of DR ligands. In particular for Fas, this difference might switch the signalling from pro- apoptotic to growth-promoting358–360. DR ligands form homo-trimeric complexes that, upon binding with their cognate receptors, stabilize their trimeri- zation and consequent high-order multimerization361,362. Ligand-dependent activation induces conformational reorganisation of the receptor, leading to binding of the adaptor protein FADD (Fas-associated protein with death domain) at the DD. DR-bound FADD exposes the death-effector domain which recruits pro-caspase 8 to the complex, completing the death-inducing signalling complex (DISC) (Fig. 6). This activation mechanism, common to Fas-R and TRAIL-Rs, differs in the TNFR1-induced cascade. In particular, DISC formation upon TNF-R activation requires additional steps mediated by binding of TRADD (TNF-R associated death domain protein) and by the formation of a multimolecular signalling complex (further discussed below)363. Once the pro-caspase 8 is recruited to the DISC, proximity-induced multimerization induces its activation and cleavage364–367, in a manner that is analogous to the processing of caspase 9 at the apoptosome. Active caspase 8 can then process the effector caspases and other critical proteins, causing the pre- cipitation of the apoptosis cascade. Various proteins regulate the action of caspase 8. In particular, the cellular FLICE-like inhibitory protein (c-FLIP) competes with caspase 8 for binding 368,369 at the DISC . The binding of the short or long isoform of c-FLIP (c-FLIPS and c-FLIPL) has opposite outcomes on caspase 8 activation. C-FLIPS occu- pies the pro-caspase binding sites, dampening the apoptotic response. C- 370 FLIPL, on the other hand, can promote caspase 8 activation . When its expression is increased, however, c-FLIPL can also function as a caspase 8 in- hibitor371. Caspase 10 was also originally described as an extrinsic apoptosis initiator caspase366; however, its importance in this role has recently been challenged, and an anti-apoptotic function has instead been proposed372. Only one subgroup of cells (including thymocytes and lymphocytes), clas- sified as type I, can execute the apoptotic program downstream DRs activation without the involvement of the mitochondrial pathway. The type II subgroup (comprising most of the cancer cells) requires instead transactivation of the intrinsic apoptosis cascade for efficient death response373,374. This crosstalk is mediated by proteolytic cleavage of the BCL-2 pro-apoptotic protein BID (BH3-interacting domain death agonist) by caspase 8. The resulting truncated BID can in turn promote BAX and BAX activation and MOMP337,375. An elevated expression ratio of XIAP versus the effector caspase 3, renders cells dependent on this second mechanism of extrinsic apoptosis376.

Figure 6. Death receptor-activated apoptotic signalling at a glance. Casp- stands for caspase, see the text above for other abbreviations.

Interestingly, DRs can also mediate survival signals. This mechanism has been extensively studied for TNRF-R1, where ligand-induced TRADD association to the DD mediates first the formation of the so-called complex I. Complex I consists of the E3 ligases TRAF (TNFR associated factor) 2 and 5, c-IAP (cellular Inhibitor of Apoptosis Protein) 1 and 2, the multiprotein linear ubiquitin chain assembly complex (LUBAC)377 and the receptor interacting serine/threonine protein kinase 1 (RIPK1). The formation of complex I leads to ubiquitination of RIPK1, a fundamental step in promoting the downstream activation of NF-κB and consequent pro-survival signalling. When RIPK1 is in a de- ubiquitinated state, it instead dissociates from complex I and together with TRADD and FADD constitute the complex II, that recruits c-FLIP and pro- caspase 8 leading to activation of the latter and apoptosis329,333,377. Analo- gously, TRAIL-R and Fas-R can trigger a non-canonical pro-survival/pro-in- flammatory pathway by interaction with de-ubiquitinated RIPK1, mediating the activation of the pro-survival NF-κB pathway378–383. Interestingly, the preferential activation of the apoptotic over the pro-survival pathway might depend, among other factors, on the degree of multimerization of the receptors, where higher-order multimerization enhances apoptosis, while impaired oligomerisation results in pro-survival signals.360.

49 5.2 The regulation of TRAIL-Receptors signalling is of special interest in cancer The discovery that cancer cells are sensitive to extrinsic apoptosis initially suggested DRs as potential targets in tumour therapy354,384,385. However, early studies discouraged this idea, as agonistic treatment directed at both TNFR and Fas-R resulted in lethal toxicity386–388. New interest in extrinsic apoptosis activation as a therapeutic approach rose with the discovery of TRAIL-R, as TRAIL treatment induces cancer cells apoptosis in absence of severe side ef- fects389–391. Remarkably, TRAIL seems to be physiologically linked to anti-cancer im- munosurveillance, as TRAIL-knock-out mice show increased susceptibility to tumour initiation and metastasis392. Additionally, TRAIL deletion induces im- munoregulation defects and it is linked to development of autoimmune diseases, suggesting a central role in clearance of autoreactive immune cells393. TRAIL ligand binds to five different receptors in human cells. However, only TRAIL-R1 and 2 contain a cytoplasmic DD and are therefore able to induce apoptosis via the previously described DRs general mechanism of activation. Receptor trimers are stabilised by the binding of trimeric ligand to a region comprised between cysteine-rich domains 2 and 3 362,394. Subsequent multimerization is critical for the formation of the DISC and apoptotic signalling. The functions of the two TRAIL-Rs are highly redundant, but some cell lines show preferential activation of one of the receptors over the other395–397. Interestingly, the presence of two TRAIL-binding DR proteins is exclusive to primates398. Decoy receptor (DcR) 1 and 2 are membrane associated proteins that bind TRAIL and have structural similarity to TRAIL-Rs. However, they lack a complete DD and, consequently, cannot activate death-inducing pathways399– 403. DcRs instead counteract the response to TRAIL by either sequestering the ligand or by directly associating with the activated TRAIL-Rs and thus im- pairing effective DISC formation404,405. In addition to antagonising apoptosis signalling, DcRs have also been found to promote pro-survival signalling via NF-κB and Akt379,406,407. Lastly, Osteoprotegerin (OPG) is a low affinity soluble TRAIL receptor that is able to sequester the ligand extracellularly408,409. Reduced expression of DcR and OPG could account for the increased sensitivity to TRAIL of many cancer cells410, however a role for Ras activation in this context has also been described411. To date, despite the testing of many promising compounds in clinical trials, the quest for a suitable drug that targets cancer via TRAIL-R-mediated apoptosis is still ongoing (reviewed in Naoum et al., 2017410). One important aspect hindering the development of a successful therapeutic strategy is the fact that many cancer cells exhibit –or acquire after treatment– resistance to TRAIL. Common mechanisms of TRAIL resistance include constitutive endocytosis

50 of the receptor, downregulation of caspase 8, c-FLIP upregulation, mutation of components of the intrinsic pathway in type II cells and upregulation of NF- κB410,412. Additionally, failure of recruitment of the receptor to lipid rafts, has been associated with resistance to apoptosis and enhanced pro-survival signalling in different experimental settings413–419. Lipid rafts are specialised regions of the membrane that hold a key role as signalling hubs. They are membrane subdomains enriched in low-density components such as sphingolipids, cholesterol and highly saturated fatty acids 420–422. The structure of lipid rafts is stabilised by association with the cytoskeleton via different transmembrane proteins and, in the case of invaginated rafts (or caveolae), by high-order oligomerisation of caveolin-1 and -2. Cave- olin multimers are also responsible for the characteristic curvature of the membrane at these sites. The discovery of these specialised subdomains was initially met with scep- ticism, since it challenged the liquid mosaic model of the plasma membrane by implying that lateral diffusion of transmembrane proteins can be re- stricted422. However, it is now becoming clear that the existence of such compartments is important for local concentration of specific proteins and regulation of their function. The significance of lipid rafts in signalling has been demonstrated, in particular, for PI3K/Akt activation by the insulin-like growth factor receptor and for DRs-mediated apoptosis418,423–425. In particular, the multimerization step of activated TRAIL-R, critical in triggering the apoptotic response, is thought to be promoted by receptor localisation at the lipid rafts413,414,426. Interestingly, it has also been reported that localisation of TRAIL-Rs at lipid rafts could have a role in prioritizing the pro-apoptotic pathway over the pro-survival one414. Although what exactly determines the localisation of TRAIL-Rs at the lipid rafts is still unclear, it has been proposed that post-translational modification, such as O- or N- glycosylation (for DR5 and 4 respectively) and palmitoylation, may play a role157–159. Moreover, the work presented in this thesis suggests that Ack1 represents an overlooked player in this process. Particularly, Paper I describes a novel role for Ack1 in increasing the localisation of TRAIL-Rs to lipid rafts and the potency of apoptotic response upon TRAIL treatment.

51 6. Present investigations

As addressed in the previous chapters, the integrity and functionality of tissues require harmonization of the behaviour of the individual cells that compose them. Deregulation of the system of communication in charge of this coordination has catastrophic consequences on homeostasis and is at the root of cancer. It becomes therefore evident that increasing our knowledge on how this complex system works is of the highest importance in the fight against cancer. In order to untangle the intricate processes of signal sensing and signal transduction in the cell, it can be useful to break the problem in smaller pieces and focus on understanding the specific role of the single molecular players. This was the approach adopted in this thesis, where we focused on deepening the knowledge on the biological role of a single protein, Ack1, with the aim of understanding its implications in epithelial homeostasis. As extensively described in Chapter 2, Ack1 is most prominently known for its roles in remodelling the cytoskeleton, trafficking of EGFR and activating transcription of mitogenic genes. This knowledge, together with occurrence of Ack1 overexpression or overactivation in a wide range of cancers, portrayed Ack1 as an oncogene. Our findings, however, endorse a bivalent function of Ack1 in epithelial biology, since we discovered three independent roles of this kinase in limiting cell survival and proliferation. This thesis provides for the first time evidence of Ack1’s function in promoting extrinsic apoptosis (Paper I), mediating the exclusion of Yap/Taz from the nucleus in the context of mechanotransduction (Paper II), and attenuating ErbB3-mediated pro-survival and pro-proliferative signalling (Paper III).

Paper I: Activated Cdc42-associated kinase 1 (Ack1) is required for tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) receptor recruitment to lipid rafts and induction of cell death The family of death receptors, including TNFR, Fas-R and TRAIL-Rs, mediates apoptotic response upon ligand binding, which is important for immuno- regulation and cancer immuno-surveillance. TRAIL-Rs are of particular interest in the cancer field since agonistic ligands induce death of cancer cells in

52 the absence of severe side effects. The success of therapeutic approaches targeting TRAIL-Rs is, however, hindered by the emergence of resistance mechanisms whose molecular players are still poorly understood. In this study, we demonstrate that Ack1 is required, in different epithelial cell lines, for efficient apoptosis upon TRAIL stimulation. Depletion of Ack1 dampened the apoptotic response without diminishing the levels of TRAIL-Rs exposed at the membrane. However, we found Ack1 to be essential for the translocation of the activated receptor to the lipid rafts, specialised membrane subdomains that favour the multimerization of TRAIL-Rs and the consequent formation of an effective death-inducing signalling complex. Interestingly, inhibition of the kinase activity of Ack1 did not recapitulate the effects of its depletion, suggesting that the regulation of the membrane trafficking of the TRAIL-Rs is independent of Ack1-mediated phosphorylation. Our findings reveal a previously unknown function of Ack1 in mediating the targeting of TRAIL-Rs to the lipid rafts, which results in an enhanced sensitivity to TRAIL, thus facili- tating extrinsic apoptosis.

Paper II: Ack1 is a mechanoresponsive kinase required for SCF (β-TrCP)-mediated degradation of Yap/Taz Yap/Taz-regulated gene expression promotes cell growth, proliferation and differentiation. The control of Yap/Taz’s function was initially ascribed to Hippo pathway-dependent phosphorylation, which resulted in anchorage in the cytoplasm via interaction with 14-3-3s. Recent research, however, demonstrated that the localisation of Yap/Taz also strongly depend on alternative pathways and, especially, on the dominant role of mechanical signals. In this paper we discovered a novel function of Ack1 in controlling the response of Yap/Taz to mechanical cues. We observed that Ack1 is required for the exclusion of Yap/Taz from the nucleus at low mechanical stress, as observed in crowded cell cultures, in cells grown on soft ECM substrates or upon depolymerisation of the actin cytoskeleton. We found that Ack1 was activated by relieving mechanical tension, and regulated the function of Yap/Taz independently of the Hippo pathway. Finally, we demonstrated that Ack1 interacts with Yap/Taz via their WW domains, and that it is required, in particular, for proteasomal degradation of Taz. Specifically, we found that Ack1 fostered the interaction of Yap/Taz with the E3 ubiquitin ligase βTrCP, resulting in reduced Taz stability upon depolymerization of the actin cytoskeleton. Our study sheds new light on the still unknown mechanisms that modulates the activation of Yap/Taz downstream mechanical stimulation.

53 Paper III: Ack1 is a negative regulator of ErbB3 that acts both by promoting its degradation and suppressing its expression Activation of the ErbB family of tyrosine kinase receptors, which includes EGFR, ErbB2, ErbB3 and ErbB4, initiates molecular signalling pathways of elevated mitogenic potential and is frequently deregulated in cancer. Despite its impaired kinase activity, ErbB3 retains signalling capacity, particularly via the PI3K/Akt pathway. Ack1 has been previously linked to the regulation of EGFR trafficking upon ligand binding, but its putative effect on the other members of the family has not been explored. We found that depletion of Ack1 increases ErbB3 protein levels, but not those of other ErbB receptors, in different epithelial cell lines. Upon Neuregulin treatment, Ack1 attenuated ErbB3 phosphorylation and, accordingly, the activation of downstream signalling pathways. The depletion of Ack1 impaired both constitutive and lig- and-induced ErbB3 degradation, suggesting a general role in modulating its stability. Additionally, the depletion of Ack1 resulted in increased expression of the gene encoding ErbB3; an effect that was further enhanced by Neuregu- lin treatment. Taken together, our data demonstrate, for the first time, that Ack1 controls ErbB3 at multiple levels, and uncouples its regulation from the other members of the ErbB family.

54 7. Future perspectives

The work presented in this thesis unveils the importance of Ack1 in modulating the transduction of different kinds of signals for the maintenance of epithelial homeostasis. At the same time, the reported observations raise new compelling questions regarding the precise mechanisms by which Ack1 co- vers these functions, and how they could be integrated in a coherent view of its role in cell signalling.

Is Ack1-mediated TRAIL-R recruitment to lipid rafts an aspect of a more general mechanism? In Paper I, we discovered the importance of Ack1 in enhancing the apoptotic response to TRAIL in different epithelial cell lines by directing TRAIL-R membrane trafficking towards the lipid rafts. However, it is not clear by which mechanisms Ack1 mediates this translocation. Despite the fact that we were not able to detect direct interaction between Ack1 and TRAIL-Rs, we observed their co-localisation in the lipid rafts upon TRAIL treatment. What directs both Ack1 and TRAIL-Rs translocation to this membrane compartments has not yet been clarified. The recruitment of Ack1 to the lipid rafts is unlikely to depend on its activation, since the kinase activity is dispensable for TRAIL-R-dependent apoptosis. Therefore, the involvement of unknown interaction partners is expected and would be worth further investigation. Notably, lipid rafts constitute a signalling hub for activation of different pathways422. Hence, it would be intriguing to explore the possibility of a more general role of Ack1 in enhancing activation of such pathways by promoting the clustering of lipid rafts or the association of various signalling proteins to these membrane regions. From a more general perspective, the study of TRAIL-Rs dynamics is rel- evant in cancer therapy, as there is an ongoing effort in developing effective treatments that trigger TRAIL-R-mediated apoptotic response410. Our results suggest that the presence of Ack1 is critical for sensitivity to TRAIL. It would be of interest to explore the possibility that, accordingly, the overexpression of Ack1, which is common in tumours, increases the responsiveness to treatment in vivo. If this hypothesis would be validated, Ack1 overexpression

55 could constitute a prognostic marker of effective response to TRAIL-Rs ago- nists.

What are the molecular details of Ack1 regulation of Yap/Taz downstream of mechanical stimulation? The study presented in Paper II describes for the first time Ack1 as an important molecular player in mechanotransduction. As addressed in Chapter 2, Ack1 has a documented role in promoting actin remodelling. Surprisingly, our findings revealed mechanosensitive properties of Ack1, indicating that its activity is engaged both upstream and downstream cytoskeleton dynamics. How relieving mechanical tension activates Ack1 is, however, still an open question that requires further investigation. Our results underline the importance of the kinase activity of Ack1 in promoting Yap/Taz’s exclusion from the nucleus. Nevertheless, the target of Ack1 phosphorylation that mediates this mechanism is yet to be identified. It has been reported that different kinases can phosphorylate Yap/Taz at a conserved tyrosine residue300–302. This modification has not, however, been found to correlate with cytoplasmic sequestration. Additionally, in our system, we did not detect tyrosine phosphorylation of Yap/Taz, suggesting that Ack1-dependent phosphorylation of an alternative substrate is required for the regulation of Yap/Taz. Interestingly, we found that Ack1 promotes the interaction of Yap/Taz with β-TrCP upon depolymerisation of the cytoskeleton, resulting, in particular, in the proteasomal degradation of Taz. Taz is notably a much less stable protein than Yap and it has be found to contain one more phosphodegron motif, which is a target of GSK3-dependent serine phosphorylation284,430. It is therefore tempting to speculate that Ack1 could promote Taz degradation via the GSK3- dependent branch. Intriguingly, GSK3 phosphorylation at a tyrosine residue is required for full activation, although the connection of this modification to regulation of the kinase is still unclear431. It would be interesting to test if Ack1 can modulate GSK3 regulation of Taz by phosphorylation at this residue, promoting, in turn, serine phosphorylation of Taz and its affinity for binding of β-TrCP.

Which mechanisms underlie Ack1-dependent regulation of ErbB3? In Paper III we approached the role of Ack1 in the regulation of the spatiotemporal dynamics of the ErbB family of receptor tyrosine kinases. Notably, we discovered that Ack1 regulates ErbB3 at multiple levels and independently

56 of the control of EGFR. The mechanistic details of this regulation are however still unclear. The function of Ack1 has been linked with many proteins involved in receptor trafficking (as described in Chapter 2). It is therefore invit- ing to hypothesise that it attenuates ErbB3 signalling by directing the sorting of receptor-containing vesicles upon activation. Such a mechanism could have consequences for both signalling activation and the dynamics of ErbB3 degradation, offering a unifying explanation to the phenotype observed when Ack1 is depleted. Additionally, it is not clear if the Ack1-dependent regulation of ErbB3 requires their physical interaction. Ack1 binds the catalytic domain of EGFR and this association requires kinase activity of the receptor. Due to the kinase impairment of ErbB3, an analogous interaction is unlikely, and it would therefore be important to test for alternative direct or indirect association modes. Interestingly, Ack1 inhibits the expression of the gene encoding ErbB3, in particular upon Neuregulin treatment. As reported in Chapter 2 Ack1 has been found to regulate gene transcription by different means. It could be worth exploring whether any of the known functions of Ack1 in transcriptional modulation could have an impact on the expression of ErbB3. Alternatively, it is possible to speculate that Ack1-mediated regulation of ErbB3 gene expression depends on the attenuation of the ErbB3-dependent signalling cascade, by in- terfering with an unknown integral feedforward mechanism. To verify this hypothesis, it could be possible to assess the effect of Ack1 depletion on ErbB3 gene expression in presence of inhibitors for downstream components of the ErbB3-mediated signalling cascade.

How are Ack1 functions integrated? In a broader perspective, our studies bring to light a putative tumour-suppres- sive function of Ack1 in epithelial tissues. This new scenario is seemingly at odds with the previously documented roles of Ack1 in cancer biology. Thus, it is necessary to understand what, during tumorigenesis, switches Ack1 from its pro-homeostatic function to promote malignancy instead. To shed light on this central issue, future research needs to focus on reconstructing a complete picture of Ack1 signalling network. A deeper knowledge of Ack1 interactome is at this moment essential to unveil the critical checkpoints of regulation of this still mysterious, yet clearly crucial, multifunctional kinase.

57 Acknowledgements

I have no words to express the incredible gratitude I feel when I look back at these years in Uppsala, for the opportunity I got and the amazing people I met on the way. Although I am aware this section cannot be enough to convey half of what I would like to, I feel the need to say a few words to the people that helped me along the way. First of all, I would like to express my gratitude to my main supervisor, Ingvar. The time spent with you during my Master’s project made me really understand how fun it is to look into the tangle of signal regulation in the cell, and really set my mind into doing a PhD. Thank you for believing in me back then and giving me the opportunity of embarking on this journey, and for everything I learned from you in these years. Thanks, as well, to my co-supervisor Calle; you are the scientist every stu- dent should aspire to become. I will always be impressed by your knowledge and dedication to science and by the fact that you always find time on your busy schedule for discussions and advice. Your kind attitude is something that sets the pace in the corridor and makes everyone work with a smile! I would also like to thank here Aive, which was the first person to take care of me in the lab. Your help in navigating the first months in Sweden and in Ludwig was incredible, and you always had a “secret trick” that made everything run smoothly. Thank you for your words of wisdom and the many laughs. You have been missed since the very moment you left us! During my thesis time I had the pleasure to work shoulder to shoulder with really nice colleagues, very valuable scientist who also made every day special. I would like to start by expressing my gratitude to the past and present members of the Signalling Dynamics group. First and foremost, Ana Rosa. You were a wonderful blessing. You have been a great colleague, collabora- tor, flatmate and, best of all, an amazing friend. I do not know how I could have made it without you. Thanks for the daily words of encouragement, for our therapeutic coffee breaks and for all the stupid jokes (and weird songs) we shared in these years. You are a great scientist and you will always be, whatever you do in the future! Also, Mari, it was great to work with you! You are such a hardworking and positive person! Plus, you know so much that it makes me a bit sad we did not collaborate more. I really wish you the best for your future! Many people have been with us in the lab during my stay. Julia, despite causing so much name confusion, your period with us was great! Thank you for your always sharp and constructive criticism at the meetings, and for

58 the cakes that were always absolutely fantastic. And, of course, thank you very much for introducing me to bouldering, we had such a good time together! Noura, you have been very helpful and were, hands down, the best at spreading the good mood (and lots of chocolate!) in the lab. Your current lab mates are so lucky to have you. Also, thanks to Takeshi. Your kindness and wise scientific advices are unforgettable, I hope we cross paths again. And thanks to Jasminka, who has been an endless source of knowledge and encouragement. And, of course, I will also never forget the students that spent some time which us: Iryna, with whom I had so much fun, and from whom I learnt a lot; Maria J. and her contagious laugh; the always resourceful Alex; kind Anton; supernice Marie; and my dear Eva D., who helped me a lot with the project presented in Paper II. I have been very lucky to be part not only of a really nice group, but also of the Ludwig/B11:3 corridor community. The great atmosphere I experienced every day has been a fuel to keep going even when things got rough and experiments crashed. I am really obliged to the people that compose the back- bone of the corridor: Lasse, who can fix everything and has solved so many emergencies in these years; Eva H. and her patience in explaining me the pa- perwork world; and Uffe for all his help. Also, the technician ladies: Anita, Aino, Lotti, Maryia and Ulla: you have always been incredibly nice with me, helped me solve any problem, advised me when I needed it and, even better, cracked a joke with me every day. I am sure no one has ever left this place without realising how important your contribution is, not only for your dedication to your work, but also for the nice environment you take care of creating (and, those of you who have left in the past years: you are missed!). Also, I had the pleasure of sharing my time in B11:3 with a group of impressively talented PhD students with whom I could share struggles and good memories. A special mention should go to Mahsa, I am so glad to have you in my life. I always have fun with you and even when we disagree we can spend hours talking our little and big problems and never get tired of it. Plus, I know you always have my back, and it feels great. Thanks also for all the help during the application time, I do not know how many times you had to answer the same questions or forward me the same emails, but hey, I made it! Merima, you never got me with your gym obsession and I hope you forgive me for that, but I will always admire your determination and the energy you put in everything you do. Thank you for all the crazy fun and the helpful discussions in this years. Maria T., you are so sweet and caring and we spent an amazing time together, especially our non-stop chats at pre-Cat’s dinners. I wish we will have more time to spend together! You are a bright and dedicated scientist that I truly admire! Claudia, I am glad to have shared so many laughs and good moments with you; I know you will reach whatever goal you set for yourself. Costas, you are so smart and hardworking, you always made sure the cell lab does not feel lonely! But you are also so kind, helpful and fun, thank you for all of that! Panos, we had lots of fun together, you really made

59 my teaching time be a pleasure, thanks for approaching everything with a smile, it is contagious! Kallia, thank you for your sweet attitude and your amazing sense of humour, it is always great to be around you. Varun, thanks for all the chats about science and life, good luck with your new path! Chun, you are always incredibly nice, and I always enjoy your science and your humour. As our “generation” is approaching the end of their PhDs, I know we are leaving the corridor in the hands of a very good young wave. Deepesh, you have been very patient and helpful with me, and no one else can turn a microscope introduction into such an entertaining time, I will never forget that. Niki, I hope my end-of-PhD craziness did not manage to scare you yet! Keep the smiles and the hardwork up! Melanie, Kehuan and all the newcom- ers, I am sure you have an amazing path in front of you: enjoy every minute. I also had the help and the support of many more experienced scientists around me. First of all, my office mate Laia, who has always been supportive, has been there in tears and laughter, and is a fantastic scientist – and person– by whom to get inspired. Thanks also to Oleks, for all the scientific and non- scientific chats, and for making the electrophoresis room an enjoyable place after all! Also, I am grateful to dear Anahita, who has always made me smile, to Natalia for the nice chats and advice, to Ria and her irresistible energy, to Anders for always being a source of good advice, to Ihor for being always helpful and fun, and to Per for being always extremely kind with me. Also thanks to Siamak, Sebastian, Yutaro, Jessica and Yanyu, with whom I did not have much time to share, but who contribute to a great work environment. Also thanks to Inna, Masato, Helena, Berit, Sinnisky, Haisha and all the others who were an example for me in the beginning of my PhD and left the corridor already: you are always remembered fondly for your help and nice- ness. In particular, my deepest gratitude goes to Glenda, you are always so good at everything you do and do not even know it. You and Andrea first, and with Agata and Nino now, have been a family to me, should I need a place to sleep or some help building furniture, you were always ready, and thank you for making ABSOLUTELY sure I would not lose weight in this years – that was a risk that should not be taken! I am extremely glad to have had the examples of many bright senior scientists around me. Aris, I truly admire you and it has been a pleasure to learn from you in many occasions; thanks for being always helpful whenever I asked. Evi, thank you for all the comments during my presentations, it has been great to get inspired from the passion you put in your science. Johan, I would like to thank you for all the feedback you gave me in our shared meetings and during my halftime, and for the help whenever I needed any, on top of all the fun you spread around. Staffan, you are incredibly knowledgeable and never hold back your time; thank you for your precious advice and for the kindness that always accompanies it. Anna-Karin, I really like your science and your positive attitude; you and your group have been a great addition to the corridor. Additionally, I am very grateful to Taija and her group, that have

60 been always friendly and extremely helpful for the development of my pro- jects. I would also like to express all my gratitude to Alexis, Eva G. and all the IMBIM administrative staff, who have been incredibly helpful in smoothing many problems in this last year, and are really the nicest people too! One of the best gifts that Ludwig/B11:3 gave to me was you, Carmen. You have been my first friend in Uppsala, and ever since the moment we met you have not stopped being helpful. I used your arms to move me (a zillion times…), your shoulder to cry on and your incredible energy to fuel me every time I needed it and you never refused a minute of your time, no matter how busy your day was. Thanks for introducing me to the Uppsaliensis crew, for your comments on my thesis and for the incredible friend you have been, even when you “looked at me and you felt pity”. My time in Uppsala would not have been so amazing without the friends I met here. I am so grateful to have been having you guys, I could surely write pages and pages about it and never getting to express fully what you have meant to me. My beloved Ana, we have been sharing a lot during this time together. You are a loyal friend that I truly value, and the most reliable person. Plus, you make me laugh a lot! What would I do without you? Laura and Fernando: you guys are great, you really know how to fire a party, there has not been a dull time with you, but I also really admire your values. That little one really is lucky! The same goes for Maria C. and Alberto: you keep going through the most amazing things in life with an effortless smile on your face. I am sure Axel will pretty soon be having his baby food from a melon! Guillermo, you are as great as your beard, and I cannot think of a better com- pliment in life! Thanks for helping me so many times in these years and sharing so many good moments together. Jose, mi favorito, you are the most ador- ably inappropriate person I know, and that says a lot. Pieter, you are a very good friend and I miss not being able to hang out with you as often anymore, but I am sure that with Elin and Hanna you really are in good hands in the city! Davide, thanks for the parties, the fun, the dances and the shared hotel rooms. I just wished you would answer your phone a bit more often! Fred, I admire that you are always up to something interesting and since “our house/in the middle of our street” it has been a blast having you around. Also, thanks to Huan and Emilien – you guys rock – and all the other Uppsaliensis that I cannot fit here but know how much they made my day by day! Additionally, I cannot forget all the people to whom I got closer in these very last years. Joran, we had so much fun together either climbing, dancing or being eaten alive by mosquitos in Herräng, thanks for each moment of it! And you were right: I survived the PhD! Gianni and Sophia, you are two awesome people singularly, but together you are incredible! Thanks for all the encouragement inside and outside the bouldering hall! Emilio, I always get in a good mood with you! Alejo, even when life feels “meh” you always remind me that I should just “go up!”, thanks for that. Andrea you are so caring and fun, Ale,

61 I love your humour and energy, I really have amazing neighbours! Also, thanks to Guilherme for all the laughs and all the rest of the MolEvo crew for letting me crush your after-works and invade your lunch table, and for always being both funny and inspiring! Also, thanks to my swing dancing buddies: Sol, you are always so sweet and energetic that one cannot help it but smile while being with you; Ilektra, I really appreciate your enthusiasm and positive attitude; Anna and Victoria, I love your humour and I always enjoyed being with you; and to all the rest with whom I shared triple steps and smiles with, it has been a blast! Finally, Thank you Dani. Beside covering a big role in making me get alive to this day – and making me happy along the way – you were of invaluable help in the writing of this thesis, by giving me tireless feedback despite all those awful acronyms! And last but not least, I am truly thankful to my whole family and all my friends back home and around the globe (especially to my cufetti) who sup- ported me and where there despite the distance. In particolare, grazie a Franca, perché mi hai sempre spronato ad esigere il massimo da me stessa, e sei stata un esempio di come si fa. E perché, anche quando ti manco, mi supporti e sei orgogliosa di me. Ad Ilaria, perché mi hai sempre ispirato ad essere coraggiosa e a pormi grandi obiettivi, ma a farlo con leggerezza. E ad Ornello, perché hai creduto in me più di chiunque altro, abbastanza da essermi di sostegno per sempre.

62 References

1. Hanahan, D. & Weinberg, R. A. The hallmarks of cancer. Cell 100, 57–70 (2000). 2. Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144, 646–674 (2011). 3. Mahajan, K. & Mahajan, N. P. ACK1/TNK2 tyrosine kinase: molecular signaling and evolving role in cancers. Oncogene 34, 4162–4167 (2015). 4. Galisteo, M. L., Yang, Y., Ureña, J. & Schlessinger, J. Activation of the non- receptor protein tyrosine kinase Ack by multiple extracellular stimuli. Proc. Natl. Acad. Sci. U. S. A. 103, 9796–9801 (2006). 5. Prieto-Echagüe, V., Gucwa, A., Brown, D. A. & Miller, W. T. Regulation of Ack1 localization and activity by the amino-terminal SAM domain. BMC Bi- ochem. 11, 42 (2010). 6. Lin, Q., Wang, J., Childress, C. & Yang, W. The activation mechanism of ACK1 (activated Cdc42-associated tyrosine kinase 1). Biochem. J. 445, 255– 264 (2012). 7. Thaker, Y. R. et al. Activated Cdc42-associated kinase 1 (ACK1) binds the sterile α motif (SAM) domain of the adaptor SLP-76 and phosphorylates prox- imal tyrosines. J. Biol. Chem. 292, 6281–6290 (2017). 8. Modzelewska, K., Newman, L. P., Desai, R. & Keely, P. J. Ack1 mediates Cdc42-dependent cell migration and signaling to p130Cas. J. Biol. Chem. 281, 37527–37535 (2006). 9. Manser, E., Leung, T., Salihuddin, H., Tan, L. & Lim, L. A non-receptor tyrosine kinase that inhibits the GTPase activity of p21cdc42. Nature 363, 364– 367 (1993). 10. Satoh, T., Kato, J., Nishida, K. & Kaziro, Y. Tyrosine phosphorylation of ACK in response to temperature shift-down, hyperosmotic shock, and epidermal growth factor stimulation. FEBS Lett. 386, 230–234 (1996). 11. Yeow-Fong, L., Lim, L. & Manser, E. SNX9 as an adaptor for linking synap- tojanin-1 to the Cdc42 effector ACK1. FEBS Lett. 579, 5040–5048 (2005). 12. Kelley, L. C. & Weed, S. A. Cortactin is a substrate of activated Cdc42-associated kinase 1 (ACK1) during ligand-induced epidermal growth factor receptor downregulation. PloS One 7, e44363 (2012). 13. Teo, M., Tan, L., Lim, L. & Manser, E. The tyrosine kinase ACK1 associates with clathrin-coated vesicles through a binding motif shared by arrestin and other adaptors. J. Biol. Chem. 276, 18392–18398 (2001). 14. Shen, H. et al. Constitutive activated Cdc42-associated kinase (Ack) phosphorylation at arrested endocytic clathrin-coated pits of cells that lack dynamin. Mol. Biol. Cell 22, 493–502 (2011). 15. Chan, W. et al. Down-regulation of active ACK1 is mediated by association with the E3 ubiquitin ligase Nedd4-2. J. Biol. Chem. 284, 8185–8194 (2009).

63 16. Lin, Q. et al. HECT E3 ubiquitin ligase Nedd4-1 ubiquitinates ACK and regulates epidermal growth factor (EGF)-induced degradation of EGF receptor and ACK. Mol. Cell. Biol. 30, 1541–1554 (2010). 17. Yokoyama, N., Lougheed, J. & Miller, W. T. Phosphorylation of WASP by the Cdc42-associated kinase ACK1: dual hydroxyamino acid specificity in a tyrosine kinase. J. Biol. Chem. 280, 42219–42226 (2005). 18. Shen, F., Lin, Q., Gu, Y., Childress, C. & Yang, W. Activated Cdc42-associated kinase 1 is a component of EGF receptor signaling complex and regulates EGF receptor degradation. Mol. Biol. Cell 18, 732–742 (2007). 19. Howlin, J., Rosenkvist, J. & Andersson, T. TNK2 preserves epidermal growth factor receptor expression on the cell surface and enhances migration and invasion of human breast cancer cells. Breast Cancer Res. BCR 10, R36 (2008). 20. Grøvdal, L. M., Johannessen, L. E., Rødland, M. S., Madshus, I. H. & Stang, E. Dysregulation of Ack1 inhibits down-regulation of the EGF receptor. Exp. Cell Res. 314, 1292–1300 (2008). 21. Jones, S., Cunningham, D. L., Rappoport, J. Z. & Heath, J. K. The non-receptor tyrosine kinase Ack1 regulates the fate of activated EGFR by inducing trafficking to the p62/NBR1 pre-autophagosome. J Cell Sci 127, 994–1006 (2014). 22. Lougheed, J. C., Chen, R.-H., Mak, P. & Stout, T. J. Crystal structures of the phosphorylated and unphosphorylated kinase domains of the Cdc42-associated tyrosine kinase ACK1. J. Biol. Chem. 279, 44039–44045 (2004). 23. Gajiwala, K. S., Maegley, K., Ferre, R., He, Y.-A. & Yu, X. Ack1: activation and regulation by allostery. PloS One 8, e53994 (2013). 24. Mott, H. R. et al. Structure of the small G protein Cdc42 bound to the GTPase- binding domain of ACK. Nature 399, 384–388 (1999). 25. Yokoyama, N. & Miller, W. T. Biochemical properties of the Cdc42-associated tyrosine kinase ACK1. Substrate specificity, authphosphorylation, and interaction with Hck. J. Biol. Chem. 278, 47713–47723 (2003). 26. Chan, W., Sit, S.-T. & Manser, E. The Cdc42-associated kinase ACK1 is not autoinhibited but requires Src for activation. Biochem. J. 435, 355–364 (2011). 27. Prieto-Echagüe, V., Gucwa, A., Craddock, B. P., Brown, D. A. & Miller, W. T. Cancer-associated mutations activate the nonreceptor tyrosine kinase Ack1. J. Biol. Chem. 285, 10605–10615 (2010). 28. Kato-Stankiewicz, J., Ueda, S., Kataoka, T., Kaziro, Y. & Satoh, T. Epidermal growth factor stimulation of the ACK1/Dbl pathway in a Cdc42 and Grb2- dependent manner. Biochem. Biophys. Res. Commun. 284, 470–477 (2001). 29. Phosphosite Ack (human). Available at: https://www.phosphosite.org/protein- Action.action?id=1883&showAllSites=true. (Accessed: 8th April 2018) 30. Buchwald, M. et al. SIAH ubiquitin ligases target the nonreceptor tyrosine kinase ACK1 for ubiquitinylation and proteasomal degradation. Oncogene 32, 4913–4920 (2013). 31. Linseman, D. A., Heidenreich, K. A. & Fisher, S. K. Stimulation of M3 muscarinic receptors induces phosphorylation of the Cdc42 effector activated Cdc42Hs-associated kinase-1 via a Fyn tyrosine kinase signaling pathway. J. Biol. Chem. 276, 5622–5628 (2001). 32. Pao-Chun, L., Chan, P. M., Chan, W. & Manser, E. Cytoplasmic ACK1 Inter- action with Multiple Receptor Tyrosine Kinases Is Mediated by Grb2 An Anal- ysis of ACK1 Effects on Axl Signaling. J. Biol. Chem. 284, 34954–34963 (2009). 33. La Torre, A. et al. A role for the tyrosine kinase ACK1 in neurotrophin signaling and neuronal extension and branching. Cell Death Dis. 4, e602 (2013).

64 34. Kato, J., Kaziro, Y. & Satoh, T. Activation of the guanine nucleotide exchange factor Dbl following ACK1-dependent tyrosine phosphorylation. Biochem. Bi- ophys. Res. Commun. 268, 141–147 (2000). 35. van der Horst, E. H. et al. Metastatic properties and genomic amplification of the tyrosine kinase gene ACK1. Proc. Natl. Acad. Sci. U. S. A. 102, 15901– 15906 (2005). 36. Eisenmann, K. M. et al. Melanoma chondroitin sulphate proteoglycan regulates cell spreading through Cdc42, Ack-1 and p130cas. Nat. Cell Biol. 1, 507– 513 (1999). 37. Mahajan, K. et al. Ack1 Mediated AKT/PKB Tyrosine 176 Phosphorylation Regulates Its Activation. PLoS ONE 5, (2010). 38. Grossmann, M. E., Huang, H. & Tindall, D. J. Androgen receptor signaling in androgen-refractory prostate cancer. J. Natl. Cancer Inst. 93, 1687–1697 (2001). 39. Mahajan, N. P. et al. Activated Cdc42-associated kinase Ack1 promotes prostate cancer progression via androgen receptor tyrosine phosphorylation. Proc. Natl. Acad. Sci. U. S. A. 104, 8438–8443 (2007). 40. Mahajan, K. et al. ACK1/TNK2 Regulates Histone H4 Tyr88-phosphorylation and AR Gene Expression in Castration-Resistant Prostate Cancer. Cancer Cell 31, 790–803.e8 (2017). 41. Mahajan, K., Lawrence, H. R., Lawrence, N. J. & Mahajan, N. P. ACK1 tyrosine kinase interacts with histone demethylase KDM3A to regulate the mammary tumor oncogene HOXA1. J. Biol. Chem. 289, 28179–28191 (2014). 42. Mahendrarajah, N. et al. HSP90 is necessary for the ACK1-dependent phosphorylation of STAT1 and STAT3. Cell. Signal. 39, 9–17 (2017). 43. Fujimoto, Y. et al. A single nucleotide polymorphism in activated Cdc42 associated tyrosine kinase 1 influences the interferon therapy in hepatitis C patients. J. Hepatol. 54, 629–639 (2011). 44. Mahajan, N. P., Whang, Y. E., Mohler, J. L. & Earp, H. S. Activated tyrosine kinase Ack1 promotes prostate tumorigenesis: role of Ack1 in polyubiquitina- tion of tumor suppressor Wwox. Cancer Res. 65, 10514–10523 (2005). 45. Xie, B. et al. ACK1 promotes hepatocellular carcinoma progression via down- regulating WWOX and activating AKT signaling. Int. J. Oncol. 46, 2057–2066 (2015). 46. Zhao, X., Lv, C., Chen, S. & Zhi, F. A role for the non-receptor tyrosine kinase ACK1 in TNF-alpha-mediated apoptosis and proliferation in human intestinal epithelial caco-2 cells. Cell Biol. Int. (2017). doi:10.1002/cbin.10875 47. Ureña, J. M. et al. Expression, synaptic localization, and developmental regulation of Ack1/Pyk1, a cytoplasmic tyrosine kinase highly expressed in the developing and adult brain. J. Comp. Neurol. 490, 119–132 (2005). 48. La Torre, A., del Rio, J. A., Soriano, E. & Ureña, J. M. Expression pattern of ACK1 tyrosine kinase during brain development in the mouse. Gene Expr. Patterns GEP 6, 886–892 (2006). 49. Wu, S., Bellve, K. D., Fogarty, K. E. & Melikian, H. E. Ack1 is a dopamine transporter endocytic brake that rescues a trafficking-dysregulated ADHD coding variant. Proc. Natl. Acad. Sci. U. S. A. 112, 15480–15485 (2015). 50. Hitomi, Y. et al. Mutations in TNK2 in severe autosomal recessive infantile onset epilepsy. Ann. Neurol. 74, 496–501 (2013). 51. Kiyose, S. et al. Detection of kinase amplifications in gastric cancer archives using fluorescence in situ hybridization. Pathol. Int. 62, 477–484 (2012).

65 52. Rudd, M. L. et al. Mutational analysis of the tyrosine kinome in serous and clear cell endometrial cancer uncovers rare somatic mutations in TNK2 and DDR1. BMC Cancer 14, (2014). 53. Shinmura, K. et al. TNK2 gene amplification is a novel predictor of a poor prognosis in patients with gastric cancer. J. Surg. Oncol. 109, 189–197 (2014). 54. Xu, S.-H. et al. Amplification of ACK1 promotes gastric tumorigenesis via ECD-dependent p53 ubiquitination degradation. Oncotarget 8, 12705–12716 (2017). 55. Cancer Sanger ACK1. Available at: https://cancer.sanger.ac.uk/cosmic/search?q=ack1. (Accessed: 8th April 2018) 56. Maxson, J. E. et al. Identification and Characterization of Tyrosine Kinase Nonreceptor 2 Mutations in Leukemia through Integration of Kinase Inhibitor Screening and Genomic Analysis. Cancer Res. 76, 127–138 (2016). 57. Mahajan, K. et al. Effect of Ack1 tyrosine kinase inhibitor on ligand-independent androgen receptor activity. The Prostate 70, 1274–1285 (2010). 58. Mahajan, K. et al. Ack1 tyrosine kinase activation correlates with pancreatic cancer progression. Am. J. Pathol. 180, 1386–1393 (2012). 59. Tan, D. S. W. et al. Bosutinib inhibits migration and invasion via ACK1 in KRAS mutant non-small cell lung cancer. Mol. Cancer 13, 13 (2014). 60. Wang, B. et al. Overexpression of activated Cdc42-associated kinase1 (Ack1) predicts tumor recurrence and poor survival in human hepatocellular carcinoma. Pathol. Res. Pract. 210, 787–792 (2014). 61. Lei, X. et al. Ack1 overexpression promotes metastasis and indicates poor prognosis of hepatocellular carcinoma. Oncotarget 6, 40622–40641 (2015). 62. Lawrence, H. R. et al. Development of novel ACK1/TNK2 inhibitors using a fragment-based approach. J. Med. Chem. 58, 2746–2763 (2015). 63. Cohen, S., Levi-Montalcini, R. & Hamburger, V. A Nerve Growth-Stimulating Factor Isolated from Sarcomas 37 and 180. Proc. Natl. Acad. Sci. U. S. A. 40, 1014–1018 (1954). 64. Cohen, S. & Levi-Montalcini, R. Purification and properties of a nerve growth- promoting factor isolated from mouse sarcoma 180. Cancer Res. 17, 15–20 (1957). 65. Cohen, S. Isolation of a Mouse Submaxillary Gland Protein Accelerating Inci- sor Eruption and Eyelid Opening in the New-born Animal. J. Biol. Chem. 237, 1555–1562 (1962). 66. Ogiso, H. et al. Crystal Structure of the Complex of Human Epidermal Growth Factor and Receptor Extracellular Domains. Cell 110, 775–787 (2002). 67. Cho, H.-S. & Leahy, D. J. Structure of the extracellular region of HER3 reveals an interdomain tether. Science 297, 1330–1333 (2002). 68. Ferguson, K. M. et al. EGF activates its receptor by removing interactions that autoinhibit ectodomain dimerization. Mol. Cell 11, 507–517 (2003). 69. Cho, H.-S. et al. Structure of the extracellular region of HER2 alone and in complex with the Herceptin Fab. Nature 421, 756–760 (2003). 70. Burgess, A. W. et al. An Open-and-Shut Case? Recent Insights into the Acti- vation of EGF/ErbB Receptors. Mol. Cell 12, 541–552 (2003). 71. Zhang, X., Gureasko, J., Shen, K., Cole, P. A. & Kuriyan, J. An Allosteric Mechanism for Activation of the Kinase Domain of Epidermal Growth Factor Receptor. Cell 125, 1137–1149 (2006). 72. Jura, N., Shan, Y., Cao, X., Shaw, D. E. & Kuriyan, J. Structural analysis of the catalytically inactive kinase domain of the human EGF receptor 3. Proc. Natl. Acad. Sci. 106, 21608–21613 (2009).

66 73. Brewer, M. R. et al. The Juxtamembrane Region of the EGF Receptor Func- tions as an Activation Domain. Mol. Cell 34, 641–651 (2009). 74. Wee, P. & Wang, Z. Epidermal Growth Factor Receptor Cell Proliferation Sig- naling Pathways. Cancers 9, (2017). 75. Lee, D. C. et al. TACE/ADAM17 processing of EGFR ligands indicates a role as a physiological convertase. Ann. N. Y. Acad. Sci. 995, 22–38 (2003). 76. Falls, D. L. Neuregulins: functions, forms, and signaling strategies. Exp. Cell Res. 284, 14–30 (2003). 77. Willem, M. Proteolytic processing of Neuregulin-1. Brain Res. Bull. 126, 178– 182 (2016). 78. Roepstorff, K. et al. Differential effects of EGFR ligands on endocytic sorting of the receptor. Traffic Cph. Den. 10, 1115–1127 (2009). 79. Warren, C. M., Kani, K. & Landgraf, R. The N-terminal Domains of Neureg- ulin 1 Confer Signal Attenuation. J. Biol. Chem. 281, 27306–27316 (2006). 80. Olayioye, M. A. et al. ErbB-1 and ErbB-2 acquire distinct signaling properties dependent upon their dimerization partner. Mol. Cell. Biol. 18, 5042–5051 (1998). 81. Lowenstein, E. J. et al. The SH2 and SH3 domain-containing protein GRB2 links receptor tyrosine kinases to ras signaling. Cell 70, 431–442 (1992). 82. Pelicci, G. et al. A novel transforming protein (SHC) with an SH2 domain is implicated in mitogenic signal transduction. Cell 70, 93–104 (1992). 83. Okazaki, K. & Sagata, N. The Mos/MAP kinase pathway stabilizes c-Fos by phosphorylation and augments its transforming activity in NIH 3T3 cells. EMBO J. 14, 5048–5059 (1995). 84. Chen, R. H., Juo, P. C., Curran, T. & Blenis, J. Phosphorylation of c-Fos at the C-terminus enhances its transforming activity. Oncogene 12, 1493–1502 (1996). 85. Cargnello, M. & Roux, P. P. Activation and Function of the MAPKs and Their Substrates, the MAPK-Activated Protein Kinases. Microbiol. Mol. Biol. Rev. MMBR 75, 50–83 (2011). 86. Jullien-Flores, V. et al. Bridging Ral GTPase to Rho pathways. RLIP76, a Ral effector with CDC42/Rac GTPase-activating protein activity. J. Biol. Chem. 270, 22473–22477 (1995). 87. Rotin, D. et al. SH2 domains prevent tyrosine dephosphorylation of the EGF receptor: identification of Tyr992 as the high-affinity binding site for SH2 domains of phospholipase C gamma. EMBO J. 11, 559–567 (1992). 88. Nakamura, Y. & Fukami, K. Regulation and physiological functions of mammalian phospholipase C. J. Biochem. (Tokyo) 161, 315–321 (2017). 89. Streb, H., Irvine, R. F., Berridge, M. J. & Schulz, I. Release of Ca2+ from a nonmitochondrial intracellular store in pancreatic acinar cells by inositol- 1,4,5-trisphosphate. Nature 306, 67–69 (1983). 90. Larsson, C. Protein kinase C and the regulation of the actin cytoskeleton. Cell. Signal. 18, 276–284 (2006). 91. Andl, C. D. et al. EGFR-induced cell migration is mediated predominantly by the JAK-STAT pathway in primary esophageal keratinocytes. Am. J. Physiol. Gastrointest. Liver Physiol. 287, G1227–1237 (2004). 92. Soltoff, S. P., Carraway, K. L., Prigent, S. A., Gullick, W. G. & Cantley, L. C. ErbB3 is involved in activation of phosphatidylinositol 3-kinase by epidermal growth factor. Mol. Cell. Biol. 14, 3550–3558 (1994). 93. Whitman, M., Downes, C. P., Keeler, M., Keller, T. & Cantley, L. Type I phosphatidylinositol kinase makes a novel inositol phospholipid, phosphatidylinositol-3-phosphate. Nature 332, 644–646 (1988).

67 94. Franke, T. F., Kaplan, D. R., Cantley, L. C. & Toker, A. Direct regulation of the Akt proto-oncogene product by phosphatidylinositol-3,4-bisphosphate. Science 275, 665–668 (1997). 95. Datta, S. R. et al. Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91, 231–241 (1997). 96. Ogawara, Y. et al. Akt enhances Mdm2-mediated ubiquitination and degradation of p53. J. Biol. Chem. 277, 21843–21850 (2002). 97. Tang, E. D., Nuñez, G., Barr, F. G. & Guan, K. L. Negative regulation of the forkhead transcription factor FKHR by Akt. J. Biol. Chem. 274, 16741–16746 (1999). 98. Bai, D., Ueno, L. & Vogt, P. K. Akt-mediated regulation of NFκB and the essentialness of NFκB for the oncogenicity of PI3K and Akt. Int. J. Cancer J. Int. Cancer 125, 2863–2870 (2009). 99. Fujita, N., Sato, S., Katayama, K. & Tsuruo, T. Akt-dependent Phosphoryla- tion of p27Kip1Promotes Binding to 14-3-3 and Cytoplasmic Localization. J. Biol. Chem. 277, 28706–28713 (2002). 100. Stambolic, V. & Woodgett, J. R. Mitogen inactivation of glycogen synthase kinase-3 beta in intact cells via serine 9 phosphorylation. Biochem. J. 303 ( Pt 3), 701–704 (1994). 101. Thorpe, L. M., Yuzugullu, H. & Zhao, J. J. PI3K in cancer: divergent roles of isoforms, modes of activation and therapeutic targeting. Nat. Rev. Cancer 15, 7–24 (2015). 102. Lee, K.-F. et al. Requirement for neuregulin receptor erbB2 in neural and cardiac development. Nature 378, 394–398 (1995). 103. Riethmacher, D. et al. Severe neuropathies in mice with targeted mutations in the ErbB3 receptor. Nature 389, 725–730 (1997). 104. Gassmann, M. et al. Aberrant neural and cardiac development in mice lacking the ErbB4 neuregulin receptor. Nature 378, 390–394 (1995). 105. Lin, W. et al. Aberrant development of motor axons and neuromuscular synapses in erbB2-deficient mice. Proc. Natl. Acad. Sci. U. S. A. 97, 1299–1304 (2000). 106. Miettinen, P. J. et al. Epithelial immaturity and multiorgan failure in mice lacking epidermal growth factor receptor. Nature 376, 337–341 (1995). 107. Sibilia, M. & Wagner, E. F. Strain-dependent epithelial defects in mice lacking the EGF receptor. Science 269, 234–238 (1995). 108. Threadgill, D. W. et al. Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype. Science 269, 230–234 (1995). 109. Sibilia, M., Steinbach, J. P., Stingl, L., Aguzzi, A. & Wagner, E. F. A strain- independent postnatal neurodegeneration in mice lacking the EGF receptor. EMBO J. 17, 719–731 (1998). 110. Xie, W., Paterson, A. J., Chin, E., Nabell, L. M. & Kudlow, J. E. Targeted Expression of a Dominant Negative Epidermal Growth Factor Receptor in the Mammary Gland of Transgenic Mice Inhibits Pubertal Mammary Duct Devel- opment. Mol. Endocrinol. 11, 1766–1781 (1997). 111. Hirsch, F. R., Scagliotti, G. V., Langer, C. J., Varella-Garcia, M. & Franklin, W. A. Epidermal growth factor family of receptors in preneoplasia and lung cancer: perspectives for targeted therapies. Lung Cancer Amst. Neth. 41 Suppl 1, S29–42 (2003). 112. Mukohara, T. et al. Differential effects of gefitinib and cetuximab on non- small-cell lung cancers bearing epidermal growth factor receptor mutations. J. Natl. Cancer Inst. 97, 1185–1194 (2005).

68 113. Haura, E. B., Zheng, Z., Song, L., Cantor, A. & Bepler, G. Activated epidermal growth factor receptor-Stat-3 signaling promotes tumor survival in vivo in non-small cell lung cancer. Clin. Cancer Res. Off. J. Am. Assoc. Cancer Res. 11, 8288–8294 (2005). 114. Seshacharyulu, P. et al. Targeting the EGFR signaling pathway in cancer therapy. Expert Opin. Ther. Targets 16, 15–31 (2012). 115. Batra, S. K. et al. Epidermal growth factor ligand-independent, unregulated, cell-transforming potential of a naturally occurring human mutant EGFRvIII gene. Cell Growth Differ. Mol. Biol. J. Am. Assoc. Cancer Res. 6, 1251–1259 (1995). 116. Lee, J. C. et al. Epidermal growth factor receptor activation in glioblastoma through novel missense mutations in the extracellular domain. PLoS Med. 3, e485 (2006). 117. Lynch, T. J. et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N. Engl. J. Med. 350, 2129–2139 (2004). 118. Pao, W. et al. EGF receptor gene mutations are common in lung cancers from ‘never smokers’ and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc. Natl. Acad. Sci. U. S. A. 101, 13306–13311 (2004). 119. Paez, J. G. et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 304, 1497–1500 (2004). 120. Salomon, D. S. et al. The role of amphiregulin in breast cancer. Breast Cancer Res. Treat. 33, 103–114 (1995). 121. Cai, Y.-C. et al. Expression of transforming growth factor-α and epidermal growth factor receptor in gastrointestinal stromal tumours. Virchows Arch. 435, 112–115 (1999). 122. Seth, D., Shaw, K., Jazayeri, J. & Leedman, P. J. Complex post-transcriptional regulation of EGF-receptor expression by EGF and TGF-alpha in human prostate cancer cells. Br. J. Cancer 80, 657–669 (1999). 123. Hsieh, E. T., Shepherd, F. A. & Tsao, M. S. Co-expression of epidermal growth factor receptor and transforming growth factor-alpha is independent of ras mutations in lung adenocarcinoma. Lung Cancer Amst. Neth. 29, 151–157 (2000). 124. Holbro, T., Civenni, G. & Hynes, N. E. The ErbB receptors and their role in cancer progression. Exp. Cell Res. 284, 99–110 (2003). 125. Slamon, D. J. et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244, 707–712 (1989). 126. Hayes, D. F. & Thor, A. D. c-erbB-2 in breast cancer: development of a clini- cally useful marker. Semin. Oncol. 29, 231–245 (2002). 127. Masood, S. & Bui, M. M. Prognostic and predictive value of HER2/neu oncogene in breast cancer. Microsc. Res. Tech. 59, 102–108 (2002). 128. Shou, J. et al. Mechanisms of tamoxifen resistance: increased estrogen receptor-HER2/neu cross-talk in ER/HER2-positive breast cancer. J. Natl. Cancer Inst. 96, 926–935 (2004). 129. Lemoine, N. R. et al. Amplification and overexpression of the EGF receptor and c-erbB-2 proto-oncogenes in human stomach cancer. Br. J. Cancer 64, 79– 83 (1991). 130. Sauter, G. et al. Heterogeneity of erbB-2 gene amplification in bladder cancer. Cancer Res. 53, 2199–2203 (1993). 131. Ricciardi, G. R. R. et al. NSCLC and HER2: Between Lights and Shadows. J. Thorac. Oncol. 9, 1750–1762 (2014).

69 132. Yakes, F. M. et al. Herceptin-induced inhibition of phosphatidylinositol-3 kinase and Akt Is required for antibody-mediated effects on p27, cyclin D1, and antitumor action. Cancer Res. 62, 4132–4141 (2002). 133. Lee-Hoeflich, S. T. et al. A central role for HER3 in HER2-amplified breast cancer: implications for targeted therapy. Cancer Res. 68, 5878–5887 (2008). 134. Amin, D. N., Campbell, M. R. & Moasser, M. M. The role of HER3, the un- pretentious member of the HER family, in cancer biology and cancer therapeutics. Semin. Cell Dev. Biol. 21, 944–950 (2010). 135. Engelman, J. A. et al. MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science 316, 1039–1043 (2007). 136. Reschke, M. et al. HER3 is a determinant for poor prognosis in melanoma. Clin. Cancer Res. Off. J. Am. Assoc. Cancer Res. 14, 5188–5197 (2008). 137. Ciardiello, F. et al. Differential expression of epidermal growth factor-related proteins in human colorectal tumors. Proc. Natl. Acad. Sci. U. S. A. 88, 7792– 7796 (1991). 138. Quinn, C. M. et al. c-erbB-3 protein expression in human breast cancer: comparison with other tumour variables and survival. Histopathology 25, 247–252 (1994). 139. Naidu, R., Yadav, M., Nair, S. & Kutty, M. K. Expression of c-erbB3 protein in primary breast carcinomas. Br. J. Cancer 78, 1385–1390 (1998). 140. Holbro, T. et al. The ErbB2/ErbB3 heterodimer functions as an oncogenic unit: ErbB2 requires ErbB3 to drive breast tumor cell proliferation. Proc. Natl. Acad. Sci. U. S. A. 100, 8933–8938 (2003). 141. Ma, J., Lyu, H., Huang, J. & Liu, B. Targeting of erbB3 receptor to overcome resistance in cancer treatment. Mol. Cancer 13, 105 (2014). 142. Williams, C. S. et al. ERBB4 is over-expressed in human colon cancer and enhances cellular transformation. Carcinogenesis 36, 710–718 (2015). 143. Kim, J.-Y. et al. Prognostic value of ERBB4 expression in patients with triple negative breast cancer. BMC Cancer 16, 138 (2016). 144. Kurppa, K. J., Denessiouk, K., Johnson, M. S. & Elenius, K. Activating ERBB4 mutations in non-small cell lung cancer. Oncogene 35, 1283–1291 (2016). 145. Salomon, D. S., Brandt, R., Ciardiello, F. & Normanno, N. Epidermal growth factor-related peptides and their receptors in human malignancies. Crit. Rev. Oncol. Hematol. 19, 183–232 (1995). 146. Montor, W. R., Salas, A. R. O. S. E. & Melo, F. H. M. de. Receptor tyrosine kinases and downstream pathways as druggable targets for cancer treatment: the current arsenal of inhibitors. Mol. Cancer 17, 55 (2018). 147. Vermeer, P. D. et al. Segregation of receptor and ligand regulates activation of epithelial growth factor receptor. Nature 422, 322–326 (2003). 148. Hobert, M. & Carlin, C. Cytoplasmic juxtamembrane domain of the human EGF receptor is required for basolateral localization in MDCK cells. J. Cell. Physiol. 162, 434–446 (1995). 149. Hobert, M. E., Kil, S. J., Medof, M. E. & Carlin, C. R. The cytoplasmic juxtamembrane domain of the epidermal growth factor receptor contains a novel autonomous basolateral sorting determinant. J. Biol. Chem. 272, 32901–32909 (1997). 150. Shelly, M. et al. Polar expression of ErbB-2/HER2 in epithelia. Bimodal regulation by Lin-7. Dev. Cell 5, 475–486 (2003). 151. Singh, B. & Coffey, R. J. Trafficking of epidermal growth factor receptor ligands in polarized epithelial cells. Annu. Rev. Physiol. 76, 275–300 (2014).

70 152. Sorkin, A. & Goh, L. K. Endocytosis and intracellular trafficking of ErbBs. Exp. Cell Res. 314, 3093–3106 (2008). 153. Resat, H., Ewald, J. A., Dixon, D. A. & Wiley, H. S. An integrated model of epidermal growth factor receptor trafficking and signal transduction. Biophys. J. 85, 730–743 (2003). 154. Austin, C. D. et al. Endocytosis and sorting of ErbB2 and the site of action of cancer therapeutics trastuzumab and geldanamycin. Mol. Biol. Cell 15, 5268– 5282 (2004). 155. Wiley, H. S. Anomalous binding of epidermal growth factor to A431 cells is due to the effect of high receptor densities and a saturable endocytic system. J. Cell Biol. 107, 801–810 (1988). 156. Yamazaki, T. et al. Role of Grb2 in EGF-stimulated EGFR internalization. J. Cell Sci. 115, 1791–1802 (2002). 157. Sigismund, S. et al. Clathrin-mediated internalization is essential for sustained EGFR signaling but dispensable for degradation. Dev. Cell 15, 209–219 (2008). 158. Hanover, J. A., Beguinot, L., Willingham, M. C. & Pastan, I. H. Transit of receptors for epidermal growth factor and transferrin through clathrin-coated pits. Analysis of the kinetics of receptor entry. J. Biol. Chem. 260, 15938– 15945 (1985). 159. Sorkina, T., Huang, F., Beguinot, L. & Sorkin, A. Effect of tyrosine kinase inhibitors on clathrin-coated pit recruitment and internalization of epidermal growth factor receptor. J. Biol. Chem. 277, 27433–27441 (2002). 160. Wiley, H. S. et al. The role of tyrosine kinase activity in endocytosis, compart- mentation, and down-regulation of the epidermal growth factor receptor. J. Biol. Chem. 266, 11083–11094 (1991). 161. Sorkin, A., Waters, C., Overholser, K. A. & Carpenter, G. Multiple autophos- phorylation site mutations of the epidermal growth factor receptor. Analysis of kinase activity and endocytosis. J. Biol. Chem. 266, 8355–8362 (1991). 162. Lamaze, C. & Schmid, S. L. Recruitment of epidermal growth factor receptors into coated pits requires their activated tyrosine kinase. J. Cell Biol. 129, 47– 54 (1995). 163. Jiang, X., Huang, F., Marusyk, A. & Sorkin, A. Grb2 regulates internalization of EGF receptors through clathrin-coated pits. Mol. Biol. Cell 14, 858–870 (2003). 164. Johannessen, L. E., Pedersen, N. M., Pedersen, K. W., Madshus, I. H. & Stang, E. Activation of the epidermal growth factor (EGF) receptor induces formation of EGF receptor- and Grb2-containing clathrin-coated pits. Mol. Cell. Biol. 26, 389–401 (2006). 165. Wang, Q., Villeneuve, G. & Wang, Z. Control of epidermal growth factor receptor endocytosis by receptor dimerization, rather than receptor kinase activation. EMBO Rep. 6, 942–948 (2005). 166. Jaramillo, M. L. et al. Effect of the anti-receptor ligand-blocking 225 monoclonal antibody on EGF receptor endocytosis and sorting. Exp. Cell Res. 312, 2778–2790 (2006). 167. Baulida, J., Kraus, M. H., Alimandi, M., Fiore, P. P. D. & Carpenter, G. All ErbB Receptors Other Than the Epidermal Growth Factor Receptor Are Endo- cytosis Impaired. J. Biol. Chem. 271, 5251–5257 (1996). 168. Hommelgaard, A. M., Lerdrup, M. & van Deurs, B. Association with Mem- brane Protrusions Makes ErbB2 an Internalization-resistant Receptor. Mol. Biol. Cell 15, 1557–1567 (2004).

71 169. Citri, A. & Yarden, Y. EGF-ERBB signalling: towards the systems level. Nat. Rev. Mol. Cell Biol. 7, 505–516 (2006). 170. Hendriks, B. S., Opresko, L. K., Wiley, H. S. & Lauffenburger, D. Coregula- tion of epidermal growth factor receptor/human epidermal growth factor receptor 2 (HER2) levels and locations: quantitative analysis of HER2 overexpression effects. Cancer Res. 63, 1130–1137 (2003). 171. Grøvdal, L. M., Johannessen, L. E., Rødland, M. S., Madshus, I. H. & Stang, E. Dysregulation of Ack1 inhibits down-regulation of the EGF receptor. Exp. Cell Res. 314, 1292–1300 (2008). 172. Sorkin, A. D., Teslenko, L. V. & Nikolsky, N. N. The endocytosis of epidermal growth factor in A431 cells: A pH of microenvironment and the dynamics of receptor complex dissociation. Exp. Cell Res. 175, 192–205 (1988). 173. Di Guglielmo, G. M., Baass, P. C., Ou, W. J., Posner, B. I. & Bergeron, J. J. Compartmentalization of SHC, GRB2 and mSOS, and hyperphosphorylation of Raf-1 by EGF but not insulin in liver parenchyma. EMBO J. 13, 4269–4277 (1994). 174. Fukazawa, T., Miyake, S., Band, V. & Band, H. Tyrosine phosphorylation of Cbl upon epidermal growth factor (EGF) stimulation and its association with EGF receptor and downstream signaling proteins. J. Biol. Chem. 271, 14554– 14559 (1996). 175. Taub, N., Teis, D., Ebner, H. L., Hess, M. W. & Huber, L. A. Late Endosomal Traffic of the Epidermal Growth Factor Receptor Ensures Spatial and Tem- poral Fidelity of Mitogen-activated Protein Kinase Signaling. Mol. Biol. Cell 18, 4698–4710 (2007). 176. Sorkin, A. et al. Recycling of epidermal growth factor-receptor complexes in A431 cells: identification of dual pathways. J. Cell Biol. 112, 55–63 (1991). 177. Carpenter, G. & Cohen, S. 125I-labeled human epidermal growth factor. Bind- ing, internalization, and degradation in human fibroblasts. J. Cell Biol. 71, 159–171 (1976). 178. Stoscheck, C. M. & Carpenter, G. Down regulation of epidermal growth factor receptors: direct demonstration of receptor degradation in human fibroblasts. J. Cell Biol. 98, 1048–1053 (1984). 179. Haglund, K. et al. Multiple monoubiquitination of RTKs is sufficient for their endocytosis and degradation. Nat. Cell Biol. 5, 461–466 (2003). 180. Huang, F., Kirkpatrick, D., Jiang, X., Gygi, S. & Sorkin, A. Differential regulation of EGF receptor internalization and degradation by multiubiquitination within the kinase domain. Mol. Cell 21, 737–748 (2006). 181. Doherty, J. K., Bond, C., Jardim, A., Adelman, J. P. & Clinton, G. M. The HER-2/neu receptor tyrosine kinase gene encodes a secreted autoinhibitor. Proc. Natl. Acad. Sci. U. S. A. 96, 10869–10874 (1999). 182. Yu, Y. et al. Cloning of a novel EGFR-related peptide: a putative negative regulator of EGFR. Am. J. Physiol. Cell Physiol. 280, C1083–1089 (2001). 183. Lee, H., Akita, R. W., Sliwkowski, M. X. & Maihle, N. J. A naturally occurring secreted human ErbB3 receptor isoform inhibits heregulin-stimulated activation of ErbB2, ErbB3, and ErbB4. Cancer Res. 61, 4467–4473 (2001). 184. Fry, W. H. D., Kotelawala, L., Sweeney, C. & Carraway, K. L. Mechanisms of ErbB receptor negative regulation and relevance in cancer. Exp. Cell Res. 315, 697–706 (2009). 185. Monast, C. S., Furcht, C. M. & Lazzara, M. J. Computational analysis of the regulation of EGFR by protein tyrosine phosphatases. Biophys. J. 102, 2012– 2021 (2012).

72 186. Gur, G. et al. LRIG1 restricts growth factor signaling by enhancing receptor ubiquitylation and degradation. EMBO J. 23, 3270–3281 (2004). 187. Segatto, O., Anastasi, S. & Alemà, S. Regulation of epidermal growth factor receptor signalling by inducible feedback inhibitors. J Cell Sci 124, 1785–1793 (2011). 188. Kario, E. et al. Suppressors of cytokine signaling 4 and 5 regulate epidermal growth factor receptor signaling. J. Biol. Chem. 280, 7038–7048 (2005). 189. Bullock, A. N., Rodriguez, M. C., Debreczeni, J. É., Songyang, Z. & Knapp, S. Structure of the SOCS4-ElonginB/C Complex Reveals a Distinct SOCS Box Interface and the Molecular Basis for SOCS-Dependent EGFR Degradation. Struct. England1993 15, 1493–1504 (2007). 190. Nicholson, S. E. et al. Suppressor of cytokine signaling (SOCS)-5 is a potential negative regulator of epidermal growth factor signaling. Proc. Natl. Acad. Sci. U. S. A. 102, 2328–2333 (2005). 191. Huang, L. et al. Unique role of SNT-2/FRS2beta/FRS3 docking/adaptor protein for negative regulation in EGF receptor tyrosine kinase signaling pathways. Oncogene 25, 6457–6466 (2006). 192. Minegishi, Y. et al. Adaptor protein complex of FRS2β and CIN85/CD2AP provides a novel mechanism for ErbB2/HER2 protein downregulation. Cancer Sci. 104, 345–352 (2013). 193. Fiorentino, L. et al. Inhibition of ErbB-2 mitogenic and transforming activity by RALT, a mitogen-induced signal transducer which binds to the ErbB-2 kinase domain. Mol. Cell. Biol. 20, 7735–7750 (2000). 194. Anastasi, S. et al. Feedback inhibition by RALT controls signal output by the ErbB network. Oncogene 22, 4221–4234 (2003). 195. Zhang, X. et al. Inhibition of the EGF receptor by binding of MIG6 to an activating kinase domain interface. Nature 450, 741–744 (2007). 196. Frosi, Y. et al. A two-tiered mechanism of EGFR inhibition by RALT/MIG6 via kinase suppression and receptor degradation. J. Cell Biol. 189, 557–571 (2010). 197. Ferby, I. et al. Mig6 is a negative regulator of EGF receptor-mediated skin morphogenesis and tumor formation. Nat. Med. 12, 568–573 (2006). 198. Guy, P. M., Platko, J. V., Cantley, L. C., Cerione, R. A. & Carraway, K. L. Insect cell-expressed p180erbB3 possesses an impaired tyrosine kinase activity. Proc. Natl. Acad. Sci. U. S. A. 91, 8132–8136 (1994). 199. Citri, A., Skaria, K. B. & Yarden, Y. The deaf and the dumb: the biology of ErbB-2 and ErbB-3. Exp. Cell Res. 284, 54–65 (2003). 200. Shi, F., Telesco, S. E., Liu, Y., Radhakrishnan, R. & Lemmon, M. A. ErbB3/HER3 intracellular domain is competent to bind ATP and catalyze au- tophosphorylation. Proc. Natl. Acad. Sci. U. S. A. 107, 7692–7697 (2010). 201. Steinkamp, M. P. et al. erbB3 Is an Active Tyrosine Kinase Capable of Homo- and Heterointeractions. Mol. Cell. Biol. 34, 965–977 (2014). 202. Prigent, S. A. & Gullick, W. J. Identification of c-erbB-3 binding sites for phosphatidylinositol 3’-kinase and SHC using an EGF receptor/c-erbB-3 chi- mera. EMBO J. 13, 2831 (1994). 203. Olayioye, M. A., Neve, R. M., Lane, H. A. & Hynes, N. E. New EMBO Mem- bers’ Review: The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J. 19, 3159 (2000). 204. Berger, M. B., Mendrola, J. M. & Lemmon, M. A. ErbB3/HER3 does not ho- modimerize upon neuregulin binding at the cell surface. FEBS Lett. 569, 332– 336 (2004).

73 205. Kani, K., Park, E. & Landgraf, R. The extracellular domains of ErbB3 retain high ligand binding affinity at endosome pH and in the locked conformation. Biochemistry (Mosc.) 44, 15842–15857 (2005). 206. Lengerich, B. van, Agnew, C., Puchner, E. M., Huang, B. & Jura, N. EGF and NRG induce phosphorylation of HER3/ERBB3 by EGFR using distinct oligomeric mechanisms. Proc. Natl. Acad. Sci. 114, E2836–E2845 (2017). 207. Sak, M. M. et al. The oncoprotein ErbB3 is endocytosed in the absence of added ligand in a clathrin-dependent manner. Carcinogenesis 33, 1031–1039 (2012). 208. Fosdahl, A. M. et al. ErbB3 interacts with Hrs and is sorted to lysosomes for degradation. Biochim. Biophys. Acta BBA - Mol. Cell Res. 1864, 2241–2252 (2017). 209. Fry, W. H. D., Simion, C., Sweeney, C. & Carraway, K. L. Quantity Control of the ErbB3 Receptor Tyrosine Kinase at the Endoplasmic Reticulum. Mol. Cell. Biol. 31, 3009–3018 (2011). 210. Szymanska, M. et al. Interaction with epsin 1 regulates the constitutive clathrin-dependent internalization of ErbB3. Biochim. Biophys. Acta 1863, 1179– 1188 (2016). 211. Reif, R. et al. Activated ErbB3 Translocates to the Nucleus via Clathrin-independent Endocytosis, Which Is Associated with Proliferating Cells. J. Biol. Chem. 291, 3837–3847 (2016). 212. Wang, Y.-N. & Hung, M.-C. Nuclear functions and subcellular trafficking mechanisms of the epidermal growth factor receptor family. Cell Biosci. 2, 13 (2012). 213. Offterdinger, M., Schöfer, C., Weipoltshammer, K. & Grunt, T. W. c-erbB-3: a nuclear protein in mammary epithelial cells. J. Cell Biol. 157, 929–939 (2002). 214. Thompson, D. W. On growth and form. (Cambridge : University Press ; New York : Macmillan, 1917). 215. Iskratsch, T., Wolfenson, H. & Sheetz, M. P. Appreciating force and shape — the rise of mechanotransduction in cell biology. Nat. Rev. Mol. Cell Biol. 15, 825–833 (2014). 216. Temin, H. M. & Rubin, H. Characteristics of an assay for Rous sarcoma virus and Rous sarcoma cells in tissue culture. Virology 6, 669–688 (1958). 217. Sanford, K. K., Likely, G. D. & Earle, W. R. The Development of Variations in Transplantability and Morphology Within a Clone of Mouse Fibroblasts Transformed to Sarcoma-Producing Cells In Vitro. JNCI J. Natl. Cancer Inst. 15, 215–237 (1954). 218. Harris, A. K., Wild, P. & Stopak, D. Silicone rubber substrata: a new wrinkle in the study of cell locomotion. Science 208, 177–179 (1980). 219. Uhler, C. & Shivashankar, G. V. Regulation of genome organization and gene expression by nuclear mechanotransduction. Nat. Rev. Mol. Cell Biol. 18, 717– 727 (2017). 220. Petridou, N. I., Spiró, Z. & Heisenberg, C.-P. Multiscale force sensing in development. Nat. Cell Biol. 19, 581–588 (2017). 221. Panciera, T., Azzolin, L., Cordenonsi, M. & Piccolo, S. Mechanobiology of YAP and TAZ in physiology and disease. Nat. Rev. Mol. Cell Biol. 18, 758– 770 (2017). 222. Folkman, J. & Moscona, A. Role of cell shape in growth control. Nature 273, 345–349 (1978). 223. Chen, C. S., Mrksich, M., Huang, S., Whitesides, G. M. & Ingber, D. E. Geo- metric control of cell life and death. Science 276, 1425–1428 (1997).

74 224. Emerman, J. T. & Pitelka, D. R. Maintenance and induction of morphological differentiation in dissociated mammary epithelium on floating collagen membranes. In Vitro 13, 316–328 (1977). 225. Lee, E. Y., Parry, G. & Bissell, M. J. Modulation of secreted proteins of mouse mammary epithelial cells by the collagenous substrata. J. Cell Biol. 98, 146– 155 (1984). 226. McBeath, R., Pirone, D. M., Nelson, C. M., Bhadriraju, K. & Chen, C. S. Cell shape, cytoskeletal tension, and RhoA regulate stem cell lineage commitment. Dev. Cell 6, 483–495 (2004). 227. Wozniak, M. A. & Chen, C. S. Mechanotransduction in development: a grow- ing role for contractility. Nat. Rev. Mol. Cell Biol. 10, 34–43 (2009). 228. Roca-Cusachs, P., Conte, V. & Trepat, X. Quantifying forces in cell biology. Nat. Cell Biol. 19, 742–751 (2017). 229. Sukharev, S. & Sachs, F. Molecular force transduction by ion channels: diversity and unifying principles. J. Cell Sci. 125, 3075–3083 (2012). 230. Choquet, D., Felsenfeld, D. P. & Sheetz, M. P. Extracellular matrix rigidity causes strengthening of integrin-cytoskeleton linkages. Cell 88, 39–48 (1997). 231. Xiang, X. et al. Structural basis and kinetics of force-induced conformational changes of an αA domain-containing integrin. PloS One 6, e27946 (2011). 232. Campbell, I. D. & Humphries, M. J. Integrin structure, activation, and interactions. Cold Spring Harb. Perspect. Biol. 3, (2011). 233. Galbraith, C. G., Yamada, K. M. & Sheetz, M. P. The relationship between force and focal complex development. J. Cell Biol. 159, 695–705 (2002). 234. Sawada, Y. & Sheetz, M. P. Force transduction by Triton cytoskeletons. J. Cell Biol. 156, 609–615 (2002). 235. Giannone, G., Jiang, G., Sutton, D. H., Critchley, D. R. & Sheetz, M. P. Talin1 is critical for force-dependent reinforcement of initial integrin-cytoskeleton bonds but not tyrosine kinase activation. J. Cell Biol. 163, 409–419 (2003). 236. Tamada, M., Sheetz, M. P. & Sawada, Y. Activation of a signaling cascade by cytoskeleton stretch. Dev. Cell 7, 709–718 (2004). 237. Rio, A. del et al. Stretching Single Talin Rod Molecules Activates Vinculin Binding. Science 323, 638–641 (2009). 238. Sawada, Y. et al. Force sensing by mechanical extension of the Src family kinase substrate p130Cas. Cell 127, 1015–1026 (2006). 239. Petridou, N. I. & Skourides, P. A. A ligand-independent integrin β1 mecha- nosensory complex guides spindle orientation. Nat. Commun. 7, 10899 (2016). 240. Ferraris, G. M. S. et al. The interaction between uPAR and vitronectin triggers ligand-independent adhesion signalling by integrins. EMBO J. 33, 2458–2472 (2014). 241. Bosch-Fortea, M. & Martín-Belmonte, F. Mechanosensitive adhesion complexes in epithelial architecture and cancer onset. Curr. Opin. Cell Biol. 50, 42–49 (2018). 242. Liu, Z. et al. Mechanical tugging force regulates the size of cell–cell junctions. Proc. Natl. Acad. Sci. 107, 9944–9949 (2010). 243. Ladoux, B. et al. Strength Dependence of Cadherin-Mediated Adhesions. Bi- ophys. J. 98, 534–542 (2010). 244. Fletcher, D. A. & Mullins, R. D. Cell mechanics and the cytoskeleton. Nature 463, 485–492 (2010). 245. Higashida, C. et al. F- and G-actin homeostasis regulates mechanosensitive actin nucleation by formins. Nat. Cell Biol. 15, 395–405 (2013). 246. Luo, T., Mohan, K., Iglesias, P. A. & Robinson, D. N. Molecular mechanisms of cellular mechanosensing. Nat. Mater. 12, 1064–1071 (2013).

75 247. Vartiainen, M. K., Guettler, S., Larijani, B. & Treisman, R. Nuclear actin regulates dynamic subcellular localization and activity of the SRF cofactor MAL. Science 316, 1749–1752 (2007). 248. Wang, Y., Nagarajan, M., Uhler, C. & Shivashankar, G. V. Orientation and repositioning of chromosomes correlate with cell geometry-dependent gene expression. Mol. Biol. Cell 28, 1997–2009 (2017). 249. Varelas, X. The Hippo pathway effectors TAZ and YAP in development, homeostasis and disease. Dev. Camb. Engl. 141, 1614–1626 (2014). 250. Piccolo, S., Dupont, S. & Cordenonsi, M. The biology of YAP/TAZ: hippo signaling and beyond. Physiol. Rev. 94, 1287–1312 (2014). 251. Dupont, S. Role of YAP/TAZ in cell-matrix adhesion-mediated signalling and mechanotransduction. Exp. Cell Res. 343, 42–53 (2016). 252. Oka, T. & Sudol, M. Nuclear localization and pro-apoptotic signaling of YAP2 require intact PDZ-binding motif. Genes Cells Devoted Mol. Cell. Mech. 14, 607–615 (2009). 253. Huang, W. et al. The N-terminal Phosphodegron Targets TAZ/WWTR1 Pro- tein for SCFβ-TrCP-dependent Degradation in Response to Phosphatidylino- sitol 3-Kinase Inhibition. J. Biol. Chem. 287, 26245–26253 (2012). 254. Hau, J. C. et al. The TEAD4-YAP/TAZ protein-protein interaction: expected similarities and unexpected differences. Chembiochem Eur. J. Chem. Biol. 14, 1218–1225 (2013). 255. Morin-Kensicki, E. M. et al. Defects in yolk sac vasculogenesis, chorioallan- toic fusion, and embryonic axis elongation in mice with targeted disruption of Yap65. Mol. Cell. Biol. 26, 77–87 (2006). 256. Hossain, Z. et al. Glomerulocystic kidney disease in mice with a targeted inactivation of Wwtr1. Proc. Natl. Acad. Sci. U. S. A. 104, 1631–1636 (2007). 257. Tian, Y. et al. TAZ promotes PC2 degradation through a SCFbeta-Trcp E3 ligase complex. Mol. Cell. Biol. 27, 6383–6395 (2007). 258. Makita, R. et al. Multiple renal cysts, urinary concentration defects, and pul- monary emphysematous changes in mice lacking TAZ. Am. J. Physiol. Renal Physiol. 294, F542–553 (2008). 259. Vassilev, A., Kaneko, K. J., Shu, H., Zhao, Y. & DePamphilis, M. L. TEAD/TEF transcription factors utilize the activation domain of YAP65, a Src/Yes-associated protein localized in the cytoplasm. Genes Dev. 15, 1229– 1241 (2001). 260. Zhang, H. et al. TEAD Transcription Factors Mediate the Function of TAZ in Cell Growth and Epithelial-Mesenchymal Transition. J. Biol. Chem. 284, 13355–13362 (2009). 261. Zhao, B. et al. TEAD mediates YAP-dependent gene induction and growth control. Genes Dev. 22, 1962–1971 (2008). 262. Lamar, J. M. et al. The Hippo pathway target, YAP, promotes metastasis through its TEAD-interaction domain. Proc. Natl. Acad. Sci. U. S. A. 109, E2441–2450 (2012). 263. Strano, S. et al. Physical interaction with Yes-associated protein enhances p73 transcriptional activity. J. Biol. Chem. 276, 15164–15173 (2001). 264. Cui, C. B., Cooper, L. F., Yang, X., Karsenty, G. & Aukhil, I. Transcriptional coactivation of bone-specific transcription factor Cbfa1 by TAZ. Mol. Cell. Biol. 23, 1004–1013 (2003). 265. Omerovic, J. et al. Ligand-regulated association of ErbB-4 to the transcriptional co-activator YAP65 controls transcription at the nuclear level. Exp. Cell Res. 294, 469–479 (2004).

76 266. Varelas, X. et al. TAZ controls Smad nucleocytoplasmic shuttling and regulates human embryonic stem-cell self-renewal. Nat. Cell Biol. 10, 837–848 (2008). 267. Varelas, X. et al. The Crumbs complex couples cell density sensing to Hippo- dependent control of the TGF-β-SMAD pathway. Dev. Cell 19, 831–844 (2010). 268. Hong, J.-H. et al. TAZ, a Transcriptional Modulator of Mesenchymal Stem Cell Differentiation. Science 309, 1074–1078 (2005). 269. Mohri, Z., Del Rio Hernandez, A. & Krams, R. The emerging role of YAP/TAZ in mechanotransduction. J. Thorac. Dis. 9, E507–E509 (2017). 270. Kanai, F. et al. TAZ: a novel transcriptional co-activator regulated by interactions with 14-3-3 and PDZ domain proteins. EMBO J. 19, 6778–6791 (2000). 271. Basu, S., Totty, N. F., Irwin, M. S., Sudol, M. & Downward, J. Akt phosphorylates the Yes-associated protein, YAP, to induce interaction with 14-3-3 and attenuation of p73-mediated apoptosis. Mol. Cell 11, 11–23 (2003). 272. Zhao, B. et al. Inactivation of YAP oncoprotein by the Hippo pathway is involved in cell contact inhibition and tissue growth control. Genes Dev. 21, 2747–2761 (2007). 273. Oka, T. et al. Functional complexes between YAP2 and ZO-2 are PDZ domain-dependent, and regulate YAP2 nuclear localization and signalling. Bio- chem. J. 432, 461–472 (2010). 274. Remue, E. et al. TAZ interacts with zonula occludens-1 and -2 proteins in a PDZ-1 dependent manner. FEBS Lett. 584, 4175–4180 (2010). 275. Howell, M., Borchers, C. & Milgram, S. L. Heterogeneous nuclear ribonuclear protein U associates with YAP and regulates its co-activation of Bax transcription. J. Biol. Chem. 279, 26300–26306 (2004). 276. Callus, B. A., Verhagen, A. M. & Vaux, D. L. Association of mammalian sterile twenty kinases, Mst1 and Mst2, with hSalvador via C-terminal coiled-coil domains, leads to its stabilization and phosphorylation. FEBS J. 273, 4264– 4276 (2006). 277. Lei, Q.-Y. et al. TAZ promotes cell proliferation and epithelial-mesenchymal transition and is inhibited by the hippo pathway. Mol. Cell. Biol. 28, 2426– 2436 (2008). 278. Yu, F.-X. & Guan, K.-L. The Hippo pathway: regulators and regulations. Genes Dev. 27, 355–371 (2013). 279. Zhang, N. et al. The Merlin/NF2 tumor suppressor functions through the YAP oncoprotein to regulate tissue homeostasis in mammals. Dev. Cell 19, 27–38 (2010). 280. Yu, J. et al. Kibra functions as a tumor suppressor protein that regulates Hippo signaling in conjunction with Merlin and Expanded. Dev. Cell 18, 288–299 (2010). 281. Dong, J. et al. Elucidation of a universal size-control mechanism in Drosophila and mammals. Cell 130, 1120–1133 (2007). 282. MacDonald, B. T., Tamai, K. & He, X. Wnt/β-Catenin Signaling: Components, Mechanisms, and Diseases. Dev. Cell 17, 9–26 (2009). 283. Azzolin, L. et al. Role of TAZ as Mediator of Wnt Signaling. Cell 151, 1443– 1456 (2012). 284. Azzolin, L. et al. YAP/TAZ Incorporation in the β-Catenin Destruction Com- plex Orchestrates the Wnt Response. Cell 158, 157–170 (2014). 285. Chan, S. W. et al. Hippo pathway-independent restriction of TAZ and YAP by angiomotin. J. Biol. Chem. 286, 7018–7026 (2011).

77 286. Zhao, B. et al. Angiomotin is a novel Hippo pathway component that inhibits YAP oncoprotein. Genes Dev. 25, 51–63 (2011). 287. Mana-Capelli, S., Paramasivam, M., Dutta, S. & McCollum, D. Angiomotins link F-actin architecture to Hippo pathway signaling. Mol. Biol. Cell 25, 1676– 1685 (2014). 288. Adler, J. J. et al. Amot130 Adapts Atrophin-1 Interacting Protein 4 to Inhibit Yes-associated Protein Signaling and Cell Growth. J. Biol. Chem. 288, 15181– 15193 (2013). 289. Yi, C. et al. The p130 Isoform of Angiomotin Is Required for Yap-Mediated Hepatic Epithelial Cell Proliferation and Tumorigenesis. Sci Signal 6, ra77– ra77 (2013). 290. Paramasivam, M. et al. Angiomotin family proteins are novel activators of the LATS2 kinase tumor suppressor. Mol. Biol. Cell 22, 3725–3733 (2011). 291. Cordenonsi, M. et al. The Hippo transducer TAZ confers cancer stem cell- related traits on breast cancer cells. Cell 147, 759–772 (2011). 292. Silvis, M. R. et al. α-Catenin Is a Tumor Suppressor That Controls Cell Accu- mulation by Regulating the Localization and Activity of the Transcriptional Coactivator Yap1. Sci Signal 4, ra33–ra33 (2011). 293. Schlegelmilch, K. et al. Yap1 acts downstream of α-catenin to control epidermal proliferation. Cell 144, 782–795 (2011). 294. Yu, F.-X. et al. Regulation of the Hippo-YAP pathway by G-protein-coupled receptor signaling. Cell 150, 780–791 (2012). 295. Miller, E. et al. Identification of serum-derived sphingosine-1-phosphate as a small molecule regulator of YAP. Chem. Biol. 19, 955–962 (2012). 296. Fan, R., Kim, N.-G. & Gumbiner, B. M. Regulation of Hippo pathway by mitogenic growth factors via phosphoinositide 3-kinase and phosphoinositide- dependent kinase-1. Proc. Natl. Acad. Sci. U. S. A. 110, 2569–2574 (2013). 297. Xia, H. et al. EGFR-PI3K-PDK1 pathway regulates YAP signaling in hepatocellular carcinoma: the mechanism and its implications in targeted therapy. Cell Death Dis. 9, 269 (2018). 298. Enzo, E. et al. Aerobic glycolysis tunes YAP/TAZ transcriptional activity. EMBO J. 34, 1349–1370 (2015). 299. Liang, N. et al. Regulation of YAP by mTOR and autophagy reveals a therapeutic target of tuberous sclerosis complex. J. Exp. Med. 211, 2249–2263 (2014). 300. Zaidi, S. K. et al. Tyrosine phosphorylation controls Runx2-mediated subnu- clear targeting of YAP to repress transcription. EMBO J. 23, 790–799 (2004). 301. Levy, D., Adamovich, Y., Reuven, N. & Shaul, Y. Yap1 phosphorylation by c-Abl is a critical step in selective activation of proapoptotic genes in response to DNA damage. Mol. Cell 29, 350–361 (2008). 302. Jang, E. J. et al. TAZ suppresses NFAT5 activity through tyrosine phosphorylation. Mol. Cell. Biol. 32, 4925–4932 (2012). 303. Dupont, S. et al. Role of YAP/TAZ in mechanotransduction. Nature 474, 179– 183 (2011). 304. Sansores-Garcia, L. et al. Modulating F-actin organization induces organ growth by affecting the Hippo pathway. EMBO J. 30, 2325–2335 (2011). 305. Aragona, M. et al. A mechanical checkpoint controls multicellular growth through YAP/TAZ regulation by actin-processing factors. Cell 154, 1047– 1059 (2013). 306. Wang, Z. et al. Interplay of mevalonate and Hippo pathways regulates RHAMM transcription via YAP to modulate breast cancer cell motility. Proc. Natl. Acad. Sci. U. S. A. 111, E89–98 (2014).

78 307. Sorrentino, G. et al. Metabolic control of YAP and TAZ by the mevalonate pathway. Nat. Cell Biol. 16, 357–366 (2014). 308. Wada, K.-I., Itoga, K., Okano, T., Yonemura, S. & Sasaki, H. Hippo pathway regulation by cell morphology and stress fibers. Dev. Camb. Engl. 138, 3907– 3914 (2011). 309. Zhao, B. et al. Cell detachment activates the Hippo pathway via cytoskeleton reorganization to induce anoikis. Genes Dev. 26, 54–68 (2012). 310. Mo, J.-S., Yu, F.-X., Gong, R., Brown, J. H. & Guan, K.-L. Regulation of the Hippo–YAP pathway by protease-activated receptors (PARs). Genes Dev. 26, 2138–2143 (2012). 311. Tang, Y. et al. MT1-MMP-Dependent Control of Skeletal Stem Cell Commit- ment via a β1-Integrin/YAP/TAZ Signaling Axis. Dev. Cell 25, 402–416 (2013). 312. Feng, X. et al. Hippo-Independent Activation of YAP by the GNAQ Uveal Melanoma Oncogene through a Trio-Regulated Rho GTPase Signaling Cir- cuitry. Cancer Cell 25, 831–845 (2014). 313. Das, A., Fischer, R. S., Pan, D. & Waterman, C. M. YAP Nuclear Localization in the Absence of Cell-Cell Contact Is Mediated by a Filamentous Actin-dependent, Myosin II- and Phospho-YAP-independent Pathway during Extracel- lular Matrix Mechanosensing. J. Biol. Chem. 291, 6096–6110 (2016). 314. Eagle, H. & Levine, E. M. Growth regulatory effects of cellular interaction. Nature 213, 1102–1106 (1967). 315. Kim, J.-H. & Asthagiri, A. R. Matrix stiffening sensitizes epithelial cells to EGF and enables the loss of contact inhibition of proliferation. J Cell Sci 124, 1280–1287 (2011). 316. Benham-Pyle, B. W., Pruitt, B. L. & Nelson, W. J. Mechanical strain induces E-cadherin–dependent Yap1 and β-catenin activation to drive cell cycle entry. Science 348, 1024–1027 (2015). 317. Camargo, F. D. et al. YAP1 Increases Organ Size and Expands Undifferenti- ated Progenitor Cells. Curr. Biol. 17, 2054–2060 (2007). 318. Cai, J. et al. The Hippo signaling pathway restricts the oncogenic potential of an intestinal regeneration program. Genes Dev. 24, 2383–2388 (2010). 319. Re, D. P. D. et al. Yes-associated Protein Isoform 1 (Yap1) Promotes Cardio- myocyte Survival and Growth to Protect against Myocardial Ischemic Injury. J. Biol. Chem. 288, 3977–3988 (2013). 320. Xin, M. et al. Hippo pathway effector Yap promotes cardiac regeneration. Proc. Natl. Acad. Sci. 110, 13839–13844 (2013). 321. Mosqueira, D. et al. Hippo Pathway Effectors Control Cardiac Progenitor Cell Fate by Acting as Dynamic Sensors of Substrate Mechanics and Nanostructure. ACS Nano 8, 2033–2047 (2014). 322. Engler, A. J., Sen, S., Sweeney, H. L. & Discher, D. E. Matrix Elasticity Di- rects Stem Cell Lineage Specification. Cell 126, 677–689 (2006). 323. Nishioka, N. et al. The Hippo signaling pathway components Lats and Yap pattern Tead4 activity to distinguish mouse trophectoderm from inner cell mass. Dev. Cell 16, 398–410 (2009). 324. Kono, K., Tamashiro, D. A. A. & Alarcon, V. B. Inhibition of RHO-ROCK signaling enhances ICM and suppresses TE characteristics through activation of Hippo signaling in the mouse blastocyst. Dev. Biol. 394, 142–155 (2014). 325. Maître, J.-L. et al. Asymmetric division of contractile domains couples cell positioning and fate specification. Nature 536, 344–348 (2016). 326. Zanconato, F., Cordenonsi, M. & Piccolo, S. YAP/TAZ at the Roots of Cancer. Cancer Cell 29, 783–803 (2016).

79 327. Bartucci, M. et al. TAZ is required for metastatic activity and chemoresistance of breast cancer stem cells. Oncogene 34, 681–690 (2015). 328. Broders-Bondon, F., Ho-Bouldoires, T. H. N., Fernandez-Sanchez, M.-E. & Farge, E. Mechanotransduction in tumor progression: The dark side of the force. J Cell Biol jcb.201701039 (2018). doi:10.1083/jcb.201701039 329. Galluzzi, L. et al. Molecular mechanisms of cell death: recommendations of the Nomenclature Committee on Cell Death 2018. Cell Death Differ. 25, 486– 541 (2018). 330. Kerr, J. F. R., Wyllie, A. H. & Currie, A. R. Apoptosis: A Basic Biological Phenomenon with Wide-ranging Implications in Tissue Kinetics. Br. J. Cancer 26, 239–257 (1972). 331. Fadok, V. A. et al. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J. Im- munol. Baltim. Md 1950 148, 2207–2216 (1992). 332. Vanden Berghe, T. et al. Necroptosis, necrosis and secondary necrosis converge on similar cellular disintegration features. Cell Death Differ. 17, 922– 930 (2010). 333. Green, D. R. & Llambi, F. Cell Death Signaling. Cold Spring Harb. Perspect. Biol. 7, (2015). 334. Miura, M., Zhu, H., Rotello, R., Hartwieg, E. A. & Yuan, J. Induction of apoptosis in fibroblasts by IL-1 beta-converting enzyme, a mammalian homolog of the C. elegans cell death gene ced-3. Cell 75, 653–660 (1993). 335. Enari, M. et al. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 391, 43–50 (1998). 336. Sakahira, H., Enari, M. & Nagata, S. Cleavage of CAD inhibitor in CAD activation and DNA degradation during apoptosis. Nature 391, 96–99 (1998). 337. Luo, X., Budihardjo, I., Zou, H., Slaughter, C. & Wang, X. Bid, a Bcl2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cell 94, 481–490 (1998). 338. Yin, X. M. et al. Bid-deficient mice are resistant to Fas-induced hepatocellular apoptosis. Nature 400, 886–891 (1999). 339. Nuñez, G. et al. Deregulated Bcl-2 gene expression selectively prolongs survival of growth factor-deprived hemopoietic cell lines. J. Immunol. Baltim. Md 1950 144, 3602–3610 (1990). 340. Nakagawa, T. et al. Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-beta. Nature 403, 98–103 (2000). 341. Buttke, T. M. & Sandstrom, P. A. Oxidative stress as a mediator of apoptosis. Immunol. Today 15, 7–10 (1994). 342. Shamas-Din, A., Brahmbhatt, H., Leber, B. & Andrews, D. W. BH3-only proteins: Orchestrators of apoptosis. Biochim. Biophys. Acta 1813, 508–520 (2011). 343. Letai, A. et al. Distinct BH3 domains either sensitize or activate mitochondrial apoptosis, serving as prototype cancer therapeutics. Cancer Cell 2, 183–192 (2002). 344. Dai, H. et al. Transient binding of an activator BH3 domain to the Bak BH3- binding groove initiates Bak oligomerization. J. Cell Biol. 194, 39–48 (2011). 345. Gavathiotis, E., Reyna, D. E., Davis, M. L., Bird, G. H. & Walensky, L. D. BH3-triggered structural reorganization drives the activation of proapoptotic BAX. Mol. Cell 40, 481–492 (2010). 346. Oh, K. J. et al. Conformational changes in BAK, a pore-forming proapoptotic Bcl-2 family member, upon membrane insertion and direct evidence for the

80 existence of BH3-BH3 contact interface in BAK homo-oligomers. J. Biol. Chem. 285, 28924–28937 (2010). 347. Dewson, G. et al. Bax dimerizes via a symmetric BH3:groove interface during apoptosis. Cell Death Differ. 19, 661–670 (2012). 348. Subburaj, Y. et al. Bax monomers form dimer units in the membrane that further self-assemble into multiple oligomeric species. Nat. Commun. 6, 8042 (2015). 349. Li, P. et al. Cytochrome c and dATP-dependent formation of Apaf-1/caspase- 9 complex initiates an apoptotic protease cascade. Cell 91, 479–489 (1997). 350. Verhagen, A. M. et al. Identification of DIABLO, a mammalian protein that promotes apoptosis by binding to and antagonizing IAP proteins. Cell 102, 43– 53 (2000). 351. Du, C., Fang, M., Li, Y., Li, L. & Wang, X. Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102, 33–42 (2000). 352. Edlich, F. et al. Bcl-x(L) retrotranslocates Bax from the mitochondria into the cytosol. Cell 145, 104–116 (2011). 353. Gibert, B. & Mehlen, P. Dependence Receptors and Cancer: Addiction to Trophic Ligands. Cancer Res. 75, 5171–5175 (2015). 354. von Karstedt, S., Montinaro, A. & Walczak, H. Exploring the TRAILs less travelled: TRAIL in cancer biology and therapy. Nat. Rev. Cancer 17, 352– 366 (2017). 355. Schulte, M. et al. ADAM10 regulates FasL cell surface expression and modulates FasL-induced cytotoxicity and activation-induced cell death. Cell Death Differ. 14, 1040–1049 (2007). 356. Black, R. A. et al. A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature 385, 729–733 (1997). 357. Mariani, S. M. & Krammer, P. H. Differential regulation of TRAIL and CD95 ligand in transformed cells of the T and B lymphocyte lineage. Eur. J. Immu- nol. 28, 973–982 (1998). 358. De Miguel, D. et al. Liposomes decorated with Apo2L/TRAIL overcome chemoresistance of human hematologic tumor cells. Mol. Pharm. 10, 893–904 (2013). 359. Nair, P. M. et al. Enhancing the antitumor efficacy of a cell-surface death lig- and by covalent membrane display. Proc. Natl. Acad. Sci. U. S. A. 112, 5679– 5684 (2015). 360. O’ Reilly, L. A. et al. Membrane-bound Fas ligand only is essential for Fas- induced apoptosis. Nature 461, 659–663 (2009). 361. Eck, M. J., Beutler, B., Kuo, G., Merryweather, J. P. & Sprang, S. R. Crystal- lization of trimeric recombinant human tumor necrosis factor (cachectin). J. Biol. Chem. 263, 12816–12819 (1988). 362. Hymowitz, S. G. et al. Triggering cell death: the crystal structure of Apo2L/TRAIL in a complex with death receptor 5. Mol. Cell 4, 563–571 (1999). 363. Hsu, H., Xiong, J. & Goeddel, D. V. The TNF receptor 1-associated protein TRADD signals cell death and NF-kappa B activation. Cell 81, 495–504 (1995). 364. Chinnaiyan, A. M., O’Rourke, K., Tewari, M. & Dixit, V. M. FADD, a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 81, 505–512 (1995).

81 365. Boldin, M. P. et al. A novel protein that interacts with the death domain of Fas/APO1 contains a sequence motif related to the death domain. J. Biol. Chem. 270, 7795–7798 (1995). 366. Kischkel, F. C. et al. Apo2L/TRAIL-dependent recruitment of endogenous FADD and caspase-8 to death receptors 4 and 5. Immunity 12, 611–620 (2000). 367. Scott, F. L. et al. The Fas-FADD death domain complex structure unravels signalling by receptor clustering. Nature 457, 1019–1022 (2009). 368. Irmler, M. et al. Inhibition of death receptor signals by cellular FLIP. Nature 388, 190–195 (1997). 369. Scaffidi, C., Schmitz, I., Krammer, P. H. & Peter, M. E. The role of c-FLIP in modulation of CD95-induced apoptosis. J. Biol. Chem. 274, 1541–1548 (1999). 370. Sharp, D. A., Lawrence, D. A. & Ashkenazi, A. Selective knockdown of the long variant of cellular FLICE inhibitory protein augments death receptor-mediated caspase-8 activation and apoptosis. J. Biol. Chem. 280, 19401–19409 (2005). 371. Hughes, M. A. et al. Co-operative and Hierarchical Binding of c-FLIP and Caspase-8: A Unified Model Defines How c-FLIP Isoforms Differentially Control Cell Fate. Mol. Cell 61, 834–849 (2016). 372. Horn, S. et al. Caspase-10 Negatively Regulates Caspase-8-Mediated Cell Death, Switching the Response to CD95L in Favor of NF-κB Activation and Cell Survival. Cell Rep. 19, 785–797 (2017). 373. Scaffidi, C. et al. Two CD95 (APO-1/Fas) signaling pathways. EMBO J. 17, 1675–1687 (1998). 374. Barnhart, B. C., Alappat, E. C. & Peter, M. E. The CD95 type I/type II model. Semin. Immunol. 15, 185–193 (2003). 375. Li, H., Zhu, H., Xu, C. J. & Yuan, J. Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 94, 491–501 (1998). 376. Jost, P. J. et al. XIAP discriminates between type I and type II FAS-induced apoptosis. Nature 460, 1035–1039 (2009). 377. Micheau, O. & Tschopp, J. Induction of TNF Receptor I-Mediated Apoptosis via Two Sequential Signaling Complexes. Cell 114, 181–190 (2003). 378. Schneider, P. et al. TRAIL receptors 1 (DR4) and 2 (DR5) signal FADD-dependent apoptosis and activate NF-kappaB. Immunity 7, 831–836 (1997). 379. Degli-Esposti, M. A. et al. The novel receptor TRAIL-R4 induces NF-kappaB and protects against TRAIL-mediated apoptosis, yet retains an incomplete death domain. Immunity 7, 813–820 (1997). 380. Ehrhardt, H. et al. TRAIL induced survival and proliferation in cancer cells resistant towards TRAIL-induced apoptosis mediated by NF-kappaB. Onco- gene 22, 3842–3852 (2003). 381. Faouzi, S. et al. Anti-Fas induces hepatic chemokines and promotes inflammation by an NF-kappa B-independent, caspase-3-dependent pathway. J. Biol. Chem. 276, 49077–49082 (2001). 382. Choi, C. et al. Fas-induced expression of chemokines in human glioma cells: involvement of extracellular signal-regulated kinase 1/2 and p38 mitogen-activated protein kinase. Cancer Res. 61, 3084–3091 (2001). 383. Barnhart, B. C. et al. CD95 ligand induces motility and invasiveness of apoptosis-resistant tumor cells. EMBO J. 23, 3175–3185 (2004). 384. Carswell, E. A. et al. An endotoxin-induced serum factor that causes necrosis of tumors. Proc. Natl. Acad. Sci. U. S. A. 72, 3666–3670 (1975).

82 385. Itoh, N. et al. The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 66, 233–243 (1991). 386. Ogasawara, J. et al. Lethal effect of the anti-Fas antibody in mice. Nature 364, 806–809 (1993). 387. Leist, M. et al. The 55-kD tumor necrosis factor receptor and CD95 independently signal murine hepatocyte apoptosis and subsequent liver failure. Mol. Med. Camb. Mass 2, 109–124 (1996). 388. Kondo, T., Suda, T., Fukuyama, H., Adachi, M. & Nagata, S. Essential roles of the Fas ligand in the development of hepatitis. Nat. Med. 3, 409–413 (1997). 389. Wiley, S. R. et al. Identification and characterization of a new member of the TNF family that induces apoptosis. Immunity 3, 673–682 (1995). 390. Walczak, H. et al. Tumoricidal activity of tumor necrosis factor-related apoptosis-inducing ligand in vivo. Nat. Med. 5, 157–163 (1999). 391. Ashkenazi, A. et al. Safety and antitumor activity of recombinant soluble Apo2 ligand. J. Clin. Invest. 104, 155–162 (1999). 392. Cretney, E. et al. Increased susceptibility to tumor initiation and metastasis in TNF-related apoptosis-inducing ligand-deficient mice. J. Immunol. Baltim. Md 1950 168, 1356–1361 (2002). 393. Lamhamedi-Cherradi, S.-E., Zheng, S.-J., Maguschak, K. A., Peschon, J. & Chen, Y. H. Defective thymocyte apoptosis and accelerated autoimmune diseases in TRAIL-/- mice. Nat. Immunol. 4, 255–260 (2003). 394. Mongkolsapaya, J. et al. Structure of the TRAIL-DR5 complex reveals mechanisms conferring specificity in apoptotic initiation. Nat. Struct. Biol. 6, 1048– 1053 (1999). 395. MacFarlane, M., Kohlhaas, S. L., Sutcliffe, M. J., Dyer, M. J. S. & Cohen, G. M. TRAIL receptor-selective mutants signal to apoptosis via TRAIL-R1 in primary lymphoid malignancies. Cancer Res. 65, 11265–11270 (2005). 396. Kelley, R. F. et al. Receptor-selective mutants of apoptosis-inducing ligand 2/tumor necrosis factor-related apoptosis-inducing ligand reveal a greater contribution of death receptor (DR) 5 than DR4 to apoptosis signaling. J. Biol. Chem. 280, 2205–2212 (2005). 397. Lemke, J. et al. TRAIL signaling is mediated by DR4 in pancreatic tumor cells despite the expression of functional DR5. J. Mol. Med. Berl. Ger. 88, 729–740 (2010). 398. Micheau, O. Regulation of TNF-Related Apoptosis-Inducing Ligand Signaling by Glycosylation. Int. J. Mol. Sci. 19, (2018). 399. Pan, G. et al. The receptor for the cytotoxic ligand TRAIL. Science 276, 111– 113 (1997). 400. MacFarlane, M. et al. Identification and molecular cloning of two novel receptors for the cytotoxic ligand TRAIL. J. Biol. Chem. 272, 25417–25420 (1997). 401. Marsters, S. A. et al. A novel receptor for Apo2L/TRAIL contains a truncated death domain. Curr. Biol. 7, 1003–1006 (1997). 402. Sheridan, J. P. et al. Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors. Science 277, 818–821 (1997). 403. Degli-Esposti, M. A. et al. Cloning and characterization of TRAIL-R3, a novel member of the emerging TRAIL receptor family. J. Exp. Med. 186, 1165–1170 (1997). 404. Mérino, D. et al. Differential inhibition of TRAIL-mediated DR5-DISC formation by decoy receptors 1 and 2. Mol. Cell. Biol. 26, 7046–7055 (2006). 405. Neumann, S., Hasenauer, J., Pollak, N. & Scheurich, P. Dominant negative effects of tumor necrosis factor (TNF)-related apoptosis-inducing ligand

83 (TRAIL) receptor 4 on TRAIL receptor 1 signaling by formation of heter- omeric complexes. J. Biol. Chem. 289, 16576–16587 (2014). 406. Lalaoui, N. et al. TRAIL-R4 promotes tumor growth and resistance to apoptosis in cervical carcinoma HeLa cells through AKT. PloS One 6, e19679 (2011). 407. Anees, M. et al. The potential evasion of immune surveillance in mucosa associated lymphoid tissue lymphoma by DcR2-mediated up-regulation of nuclear factor-κB. Leuk. Lymphoma 56, 1440–1449 (2015). 408. Emery, J. G. et al. Osteoprotegerin is a receptor for the cytotoxic ligand TRAIL. J. Biol. Chem. 273, 14363–14367 (1998). 409. Neville-Webbe, H. L. et al. Osteoprotegerin (OPG) produced by bone marrow stromal cells protects breast cancer cells from TRAIL-induced apoptosis. Breast Cancer Res. Treat. 86, 269–279 (2004). 410. Naoum, G. E., Buchsbaum, D. J., Tawadros, F., Farooqi, A. & Arafat, W. O. Journey of TRAIL from Bench to Bedside and its Potential Role in Immuno- Oncology. Oncol. Rev. 11, 332 (2017). 411. Nesterov, A., Nikrad, M., Johnson, T. & Kraft, A. S. Oncogenic Ras sensitizes normal human cells to tumor necrosis factor-alpha-related apoptosis-inducing ligand-induced apoptosis. Cancer Res. 64, 3922–3927 (2004). 412. Jin, Z., McDonald, E. R., Dicker, D. T. & El-Deiry, W. S. Deficient tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) death receptor transport to the cell surface in human colon cancer cells selected for resistance to TRAIL-induced apoptosis. J. Biol. Chem. 279, 35829–35839 (2004). 413. Delmas, D. et al. Redistribution of CD95, DR4 and DR5 in rafts accounts for the synergistic toxicity of resveratrol and death receptor ligands in colon carcinoma cells. Oncogene 23, 8979–8986 (2004). 414. Song, J. H. et al. Lipid rafts and nonrafts mediate tumor necrosis factor related apoptosis-inducing ligand induced apoptotic and nonapoptotic signals in non small cell lung carcinoma cells. Cancer Res. 67, 6946–6955 (2007). 415. Dumitru, C. A., Carpinteiro, A., Trarbach, T., Hengge, U. R. & Gulbins, E. Doxorubicin enhances TRAIL-induced cell death via ceramide-enriched membrane platforms. Apoptosis Int. J. Program. Cell Death 12, 1533–1541 (2007). 416. Xu, L. et al. Oxaliplatin enhances TRAIL-induced apoptosis in gastric cancer cells by CBL-regulated death receptor redistribution in lipid rafts. FEBS Lett. 583, 943–948 (2009). 417. Xu, L. et al. Epirubicin enhances TRAIL-induced apoptosis in gastric cancer cells by promoting death receptor clustering in lipid rafts. Mol. Med. Rep. 4, 407–411 (2011). 418. Ouyang, W. et al. Redistribution of DR4 and DR5 in lipid rafts accounts for the sensitivity to TRAIL in NSCLC cells. Int. J. Oncol. 39, 1577–1586 (2011). 419. Ouyang, W. et al. Absence of death receptor translocation into lipid rafts in acquired TRAIL-resistant NSCLC cells. Int. J. Oncol. 42, 699–711 (2013). 420. Simons, K. & Ikonen, E. Functional rafts in cell membranes. Nature 387, 569– 572 (1997). 421. Simons, K. & Toomre, D. Lipid rafts and signal transduction. Nat. Rev. Mol. Cell Biol. 1, 31–39 (2000). 422. Pike, L. J. Rafts defined: a report on the Keystone Symposium on Lipid Rafts and Cell Function. J. Lipid Res. 47, 1597–1598 (2006). 423. Gao, X. & Zhang, J. Spatiotemporal analysis of differential Akt regulation in plasma membrane microdomains. Mol. Biol. Cell 19, 4366–4373 (2008). 424. Gao, X. et al. PI3K/Akt signaling requires spatial compartmentalization in plasma membrane microdomains. Proc. Natl. Acad. Sci. U. S. A. 108, 14509– 14514 (2011).

84 425. Scheel-Toellner, D. et al. The death-inducing signalling complex is recruited to lipid rafts in Fas-induced apoptosis. Biochem. Biophys. Res. Commun. 297, 876–879 (2002). 426. Vanoosten, R. L., Moore, J. M., Ludwig, A. T. & Griffith, T. S. Depsipeptide (FR901228) enhances the cytotoxic activity of TRAIL by redistributing TRAIL receptor to membrane lipid rafts. Mol. Ther. J. Am. Soc. Gene Ther. 11, 542–552 (2005). 427. Wagner, K. W. et al. Death-receptor O-glycosylation controls tumor-cell sensitivity to the proapoptotic ligand Apo2L/TRAIL. Nat. Med. 13, 1070–1077 (2007). 428. Yoshida, T., Shiraishi, T., Horinaka, M., Wakada, M. & Sakai, T. Glycosyla- tion modulates TRAIL-R1/death receptor 4 protein: different regulations of two pro-apoptotic receptors for TRAIL by tunicamycin. Oncol. Rep. 18, 1239– 1242 (2007). 429. Rossin, A., Derouet, M., Abdel-Sater, F. & Hueber, A.-O. Palmitoylation of the TRAIL receptor DR4 confers an efficient TRAIL-induced cell death signalling. Biochem. J. 419, 185–192, 2 p following 192 (2009). 430. Huang, W. et al. The N-terminal phosphodegron targets TAZ/WWTR1 protein for SCFβ-TrCP-dependent degradation in response to phosphatidylinositol 3- kinase inhibition. J. Biol. Chem. 287, 26245–26253 (2012). 431. Beurel, E., Grieco, S. F. & Jope, R. S. Glycogen synthase kinase-3 (GSK3): regulation, actions, and diseases. Pharmacol. Ther. 148, 114–131 (2015).

85 Acta Universitatis Upsaliensis Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1469 Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine. (Prior to January, 2005, the series was published under the title “Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine”.)

ACTA UNIVERSITATIS UPSALIENSIS Distribution: publications.uu.se UPPSALA urn:nbn:se:uu:diva-349384 2018