Functional Analysis of the Caspase-Mediated Cleavage of the Pro-Apoptotic Tumor Suppressor Protein PAR-4”

“Functional analysis of the Caspase-mediated cleavage of the pro-apoptotic tumor suppressor protein PAR-4”

Von der Fakultät für Mathematik, Informatik und Naturwissenschaften der RWTH Aachen University genehmigte Dissertation zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften

vorgelegt von

Master of Science Fabian Treude aus Herten

Berichter: Univ.-Prof. Dr. rer. nat. Bernhard Lüscher Prof. Dr. rer. nat. Christian Liedtke Univ.-Prof. Dr. rer. nat. Michael Huber

Tag der mündlichen Prüfung: 22.07.2016

Diese Dissertation ist auf den Internetseiten der Universitätsbibliothek online verfügbar

Meinen Eltern, Manuela und Frank Abstract

Abstract

The pro-apoptotic tumor suppressor protein prostate apoptosis response-4 (PAR-4) is silenced in a well-defined subset of human cancers including lung, prostate and breast cancer and its down-regulation serves as a mechanism for cancer cell survival following chemotherapy. PAR-4 re-expression selectively causes apoptosis in cancer cells but how its pro-apoptotic functions are controlled and executed precisely is poorly understood. In the present study it is demonstrated that UV- or TNFα-induced apoptosis results in a rapid Caspase-8-specific PAR-4 cleavage at EEPD131↓G, which separates the unstructured N-terminal part from the C-terminal region that contains the NLS, SAC and LZ domains. I further demonstrate that TNFα-mediated hydrolysis of PAR-4 requires Caspase-8 and leads to nuclear accumulation of the C-terminal cleavage fragment of the tumor suppressor and thereby induces apoptosis. Taken together, these results indicate that Caspase-8-mediated cleavage induced nuclear translocation of the C- terminal part of PAR-4 is critical in regulating cell death triggered by TNFα. Recent evidence implicates down-regulation of PAR-4 as a critical step for breast cancer recurrence. Down-regulation of PAR-4 allows tumor cells to survive tumor regression following targeted therapy and chemotherapy and is both a necessary and sufficient step for tumor recurrence. Hence, PAR-4 expression was analyzed in a panel of breast cancer cell lines and low PAR-4 expression was observed in the majority of triple negative breast cancer (TNBC) cells. Interestingly, Caspase-8-mediated PAR-4 cleavage was also observed in TNBC cells following DNA-damage-induced apoptosis using genotoxic drugs. I further demonstrate that loss of PAR-4 in TNBC cells mediates resistance to DNA-damage induced apoptosis and surprisingly also interferes with Caspase-8 activation, indicating that PAR-4 is capable to amplify Caspase-8 activation via an unknown feedback mechanism. Low PAR-4 expression promotes tumor cell survival following therapy but the pathways controlling PAR-4 expression are not identified. Using an unbiased mass spectrometry approach I was able to identify nine physical PAR-4 interaction partners out of which two belong to the family of E3 ubiquitin-protein ligases (MYCBP2 and UBR5) and one to the family of deubiquitylases (USP7). These novel PAR-4 binding proteins suggest that down-regulation of PAR-4 might be specifically linked to disordered ubiquitin signaling in cancer.

I Abstract

In summary, the results provide evidence that the mechanism by which PAR-4 orchestrates the apoptotic process requires cleavage by Caspase-8 following TNFα- or genotoxic drug-induced apoptosis and functions by cleavage-induced nuclear translocation of the C-terminal part.

II Zusammenfassung

Zusammenfassung

Prostate apoptosis response 4 (PAR-4) ist ein pro-apoptotischer Tumorsuppressor, dessen Proteinexpression in einer Vielzahl von Krebsgeweben, darunter Lungen-, Prostata- und Brustkrebs herunter reguliert ist. Die meisten durch Brustkrebs verursachten Todesfälle entstehen durch Rückfälle nach primärer Behandlung des primären Tumors. Neue Daten zeigen, dass das Ausschalten der Expression des Tumorsuppressorproteins PAR-4 notwendig und ausreichend ist, um einen Tumorrezidiv bei Brustkrebspatienten zu fördern. Die Expression von PAR-4 hingegen initiiert selektiv Apoptose in Tumorzellen, wobei die molekularen Mechanismen, die den pro-apoptotischen Eigenschaften des Tumorsuppressors zu Grunde liegen, weitestgehend unbekannt sind. In der vorliegenden Arbeit konnte gezeigt werden, dass PAR-4 nach UV-Behandlung oder TNFα-Stimulation durch Caspase-8 an Position EEPD131↓G gespalten wird und sich das C-terminale Spaltprodukt, welches die funktionell wichtigen NLS, SAC und LZ aufweist, im Zellkern anreichert. Weiterhin konnte gezeigt werden, dass TNFα- induzierte PAR-4-Spaltung in der Brustkrebs-Zellline MCF-7 Caspase-8-abhängig ist und die Expression von PAR-4 für TNFα-induzierte Apoptose notwendig ist. Diese Daten suggerieren, dass die Caspase-8-vermittelte Prozessierung und nukleäre Translokation von PAR-4 eine wichtige Voraussetzung für die pro-apoptotische Funktion von PAR-4 ist. Das Ausschalten des Tumorsuppressors PAR-4 ist ein wichtiger und ausreichender molekularer Prozess, der zu Generierung von Sekundärtumoren in der Brust nach gezielter Krebstherapie oder chemotherapeutischer Behandlung maßgeblich beiträgt. Daher wurde die Expression von PAR-4 in einer Anzahl von Brusttumor-Zelllinien getestet und niedrige Expressionsraten des Tumorsuppressors in 75% der triple negativen Brustkrebs (TNBC)-Zellen festgestellt. Interessanterweise konnte die Caspase-8-abhängige PAR-4-Spaltung in diesen TNBC-Zellen nach DNA-Schädigung durch Chemotherapeutika beobachtet werden. Das Ausschalten der PAR-4-Expression in den TNBC-Zellen hatte Resistenzbildung gegenüber den verwendeten genotoxischen Agentien zu Folge, die sich durch den Verlust der Caspase-8-Aktivität auszeichnete. Dies lässt wiederum Rückschlüsse auf eine potentielle biologische Regulation von Caspase-8 durch das pro-apoptotische Protein PAR-4 zu.

III Zusammenfassung

Niedrige Expressionsraten von PAR-4 gehen mit schlechter Prognose für Patienten mit Brustkrebs einher, jedoch sind die dafür verantwortlichen molekularen Mechanismen unzureichend erforscht. Mit Hilfe von massenspektrometrischen Analysen konnte ich neun potentielle Interaktionspartner identifizieren, worunter sich zwei E3-Ligasen (MYCBP2 und UBR5) und ein Deubiquitinierungsenzym (USP7) befanden. Diese Interaktionspartner schlagen vor, dass das Ausschalten von PAR-4 auf Deregulationen innerhalb von Ubiquitin-Signalwegen zurückzuführen sein könnte. Zusammengefasst zeigen die Daten, dass ein grundlegender Mechanismus der TNFα- vermittelten oder der durch DNA-Schädigung-induzierten Apoptose durch die Caspase- 8-abhängige Spaltung des Tumorsuppressors PAR-4 und der nukleären Translokation des C-terminalen Spaltprodukts charakterisiert ist.

IV Table of Contents

Abstract ...... I Zusammenfassung...... III 1 Introduction...... 1 1.1 Hallmarks of cancer...... 1 1.2 Apoptosis...... 2 1.3 PAR-4...... 20 1.4 Aims of this study...... 28 2 Experimental procedures ...... 29 2.1 Materials...... 29 2.2 Conventional Cloning...... 32 2.3 Work with prokaryotic cells...... 35 2.4 Work with eukaryotic cells...... 36 2.5 Methods in biochemistry...... 40 2.5.1 Cell lysis...... 40 2.6 Confocal laser scanning microscopy...... 45 2.7 Flow cytometry...... 46 2.8 Mass spectrometry approaches...... 47 3 Results...... 48 3.1 UV-induced apoptosis results in Caspase-dependent PAR-4 cleavage at ...... 48 EEPD131↓G...... 48 3.2 Inducible expression of PAR-4 but not PAR-4 D131G interferes with cell ...... 50 proliferation...... 50 3.3 PAR-4 is a substrate of Caspase-8 in vitro and in vivo...... 52 3.4 TNFα-induced apoptosis requires Caspase-8-mediated processing of PAR-4....56 3.5 PAR-4 is down-regulated in a subset of breast cancer cell lines...... 58 3.6 PAR-4 is cleaved by Caspase-8 following DNA-damage...... 60 3.7 Caspase-8-mediated PAR-4 cleavage is required for DNA-damage-induced cell ...... 63 death...... 63 3.8 The C-terminal PAR-4 fragment translocates to the nucleus upon DNA-...... 65 damage...... 65 3.9 C-terminal PAR-4 fragment is sufficient to induce apoptosis...... 67 3.10 PAR-4 binds to the deubiquitylating enzyme USP7...... 68 4 Discussion...... 71 4.1 Caspase-dependent PAR-4 cleavage interferes with its pro-apoptotic activities..72 4.2 PAR-4 is a bona fide Caspase-8 substrate...... 73 4.3 Caspase-8 is activated in breast cancer cells upon genotoxic drug treatment ...75 and is required for apoptosis induction...... 75 4.4 PAR-4 cleavage is required for genotoxic drug-induced apoptosis and ...... 77 interferes with Caspase-8 activation...... 77 4.5 Nuclear translocation of the C-terminal PAR-4 fragment is a key regulatory ...... 78 event in DNA-damage-induced apoptosis...... 78 4.6 USP7 – a novel interaction partner of PAR-4...... 80 5 Summary and Outlook...... 83 6 References...... 85 7 Appendix...... 103 7.1 Abbreviations...... 103 7.2 Curriculum Vitae...... 108 7.3 Scientific contributions...... 109 7.4 Eidesstattliche Erklärung...... 111 7.5 Danksagung...... 112 Introduction

1 Introduction

1.1 Hallmarks of cancer

The development of cancer is characterized by the malignant transformation of healthy cells, which is inseparably connected with a multi-step acquisition of genetic alterations. Although tumorigenesis is a complex process initiated by various events, some mutation patterns share redundancy in many tumor entities, resulting in similar pathobiochemical and pathophysiological phenotypes. For this reason Hanahan and Weinberg proposed that the hallmarks of cancer comprise of six distinct biological capabilities gained by malignant cells during the multistep formation of tumors, which were expanded by two new emerging hallmarks in 2011 [Hanahan and Weinberg 2000; Hanahan and Weinberg 2011]. These eight hallmarks provide a logical framework for understanding the complexities of neoplastic disease. They include firstly sustaining proliferative signaling, secondly evading growth suppressors, thirdly resisting cell death, fourthly enabling replicative immortality, fifthly inducing angiogenesis, sixthly activating invasion and metastasis, seventhly reprogramming of energy metabolism and lastly evading immune destruction [Hanahan and Weinberg 2011]. Therefore, the multistep development of human tumor pathogenesis could be explained by the requirement of incipient cancer cells to gain a succession of these biological hallmark capabilities that enable them to become tumorigenic and in the end malignant. Evading programmed cell death is one of the crucial hallmarks of cancer development. The tumor suppressor protein PAR-4 is frequently down-regulated in several cancer entities as it antagonizes tumorigenesis through the induction of apoptosis in cancer cells [Hebbar, Wang, and Rangnekar 2012]. Apoptosis is triggered by a variety of stress signals and can be divided into two major signaling pathways that respond to extracellular or intracellular death signals [Adams and Cory 2010]. Resistance to apoptosis by cancer cells remains an important clinical problem, because it facilitates tumorigenesis and renders tumor cells resistant to therapy, as many therapeutic strategies act primarily by inducing apoptosis [Igney and Krammer 2002; Adams and Cory 2010]. The following chapters describe the molecular mechanisms of apoptosis and list strategies to induce apoptotic cell death in therapeutic approaches. Finally, a brief overview of the pro-apoptotic activities of PAR-4, a tumor suppressor protein that is capable to drive cancer cells into apoptosis is outlined.

1 Introduction

1.2 Apoptosis

1.2.1 Introduction to apoptosis

Apoptosis is defined as a highly regulated form of controlled cell death and represents an important cellular process in the regulation of mammalian cell populations. With the observation and documentation of dying cells, Kerr and colleagues introduced the term “apoptosis”, which is derived from a Greek word describing the dropping off of petals from flowers or leaves from trees [Kerr, Wyllie, and Currie 1972]. Apoptosis represents the final stage in the development of a cell and serves for cell homeostasis in healthy tissues. Besides elimination of damaged, infected or aged cells, many biological processes like cell lineage differentiation or complete organogenesis are controlled by apoptotic cell death [Elmore 2007]. Apoptotic signaling is a cellular response to various stress signals and is based on a tightly regulated genetic program. Hence, mutational alterations in the genetic program provide mechanisms for the deregulation of biochemical processes coupled to apoptosis and defects in apoptosis are associated with human diseases ranging from neurodegenerative disorders to cancer [Lowe and Lin 2000]. Apoptotic cells can be identified by typical morphological changes characterized by cytoplasmic condensation, nuclear fragmentation, cell rounding and blebbing of the plasma membrane. Another hallmark of an apoptotic phenotype is described by the formation of membrane-enclosed vesicles containing diverse cell contents called apoptotic bodies. Since cell death has been progressively subdivided in several intermediate forms with overlapping characteristics, discrimination of apoptosis from other forms of cell death by classic morphological criteria turned out to be challenging. Therefore, definition of cell death by morphological changes was more and more replaced by specific biochemical markers occurring in dying cells [Kroemer et al. 2009].

1.2.2 Caspases

Cysteine-dependent aspartate-directed proteases (Caspases) are extensively characterized biochemical marker proteins activated within programmed cell death. Caspases belong to a family of endopeptidases comprising of 12 family members in humans, which serve as key players in apoptosis, but are also involved in several other

2 Introduction cellular processes. Interleukin-1β-converting enzyme (ICE) was the first cysteine protease found in human monocytes and cleaves interleukin-1β (IL-1β) precursor to mature IL-1β [Thornberry et al. 1992]. ICE represents the mammalian homolog of the Caenorhabditis elegans death protein 3 (CED-3), a central regulator of cell death during the development of the nematode C. elegans [Yuan et al. 1993]. Since more and more ICE/CED-3 related proteins were discovered and individual protein names would have been confusing, a unified nomenclature was introduced in 1996 by a committee of scientists, who were all involved in the identification of these homologs, which use a cysteine (“C”) for hydrolyzing a peptide bond exclusively after aspartic acid residues (“aspase”). In contrast to the human Caspase family consisting of Caspase-1 to -10, Caspase-14 and -16, the mouse genome encodes for 11 murine Caspases, Caspase-1, -2, -3, -6, -7, -8, -9, -11, -12, -14 and -16. Human Caspase-10 is absent in mice and human Caspase-4 and -5 are functional orthologs of murine Caspase-11 and -12 as well as the remaining Caspases, which share the same numbers. Caspase-13 is the bovine homolog of human Caspase-4 [Li and Yuan 2008]. Human Caspase-16 was found in a comprehensive evolutionary analysis in mammalian Caspases and its sequence is similar to Caspase-14. The biochemical function of human Caspase-16 remains unclear [Eckhart et al. 2008]. Human Caspases can be subdivided into three distinct groups, regarding their structure and biochemical function (Table 1). In general, Caspases are expressed as inactive zymogens and these pro-Caspases consist of an N-terminal located pro-domain, a large subunit and a small subunit at the C-terminus. Group 1, referred to as apical or initiator Caspases comprises Caspase-2, -8, -9 and -10. Initiator Caspases feature either a death effector domain (DED), or a Caspase recruitment domain (CARD) at the N- terminus, which allow protein-protein interaction following the induction of apoptosis. Once the initiator Caspases are activated by autoproteolysis, they activate downstream effector Caspases, Caspase-3, -6 and -7. This process is termed Caspase cascade. Effector Caspases, also called executioner Caspases, cleave several cellular proteins, thereby triggering distinct forms of cell death. They are characterized by the lack of the DED or CARD domain in the N-terminal region. Caspases-1, -4, and -5 are inflammatory Caspases and represent the third subgroup of Caspases. They all possess an N-terminal CARD domain and are critical mediators during immune responses, but are not involved in apoptotic processes [Li and Yuan 2008]. Human Caspase-14 is associated with the differentiation of keratinocytes and therefore is not

3 Introduction listed in the previous described groups as well as Caspase-16, whose biochemical function remains elusive [Denecker et al. 2008; Eckhart et al. 2008].

Table 1: The human Caspase family.

Group Function Member Domains Recognition motif I Initiator Caspases Caspase-2 CARD DEHD Caspase-8 DED LETD Caspase-9 CARD LEHD Caspase-10 DED VEHD II Executioner Caspases Caspase-3 - DEVD Caspase-6 - VEHD Caspase-7 - DEVD III Inflammatory Caspases Caspase-1 CARD IEPD Caspase-4 CARD WEHD Caspase-5 CARD WEHD extra Keratinocyte differentiation Caspase-14 - WEHD Unknown Caspase-16 Unknown Unknown DED = Death effector domain; CARD = Caspase recruitment domain [Modified from Thornberry et al. 1997; Garcia-Calvo et al. 1999; Lamkanfi, Kalai, and Vandenabeele 2004; Mikolajczyk et al. 2004; Eckhart et al. 2008]

Inactive pro-Caspases possess limited enzymatic activity and are in general activated through oligomerization followed by autoproteolysis or proteolytic cleavage by another Caspase family member. Close proximity is sufficient for initiator Caspases to activate themselves upon dimerization of the precursor proteins [Salvesen and Dixit 1999]. Within all Caspases, cleavage of the pro-domain and internal processing of the zymogens separates the large subunit from the small subunit, referred to as p20 and p10. Processing and quaternary rearrangement of the fragments is essential for activation. Activated Caspase fragments p20 and p10 both form a heterodimer and together with another dimer they build an activated Caspase heterotetramer [Fuentes- Prior and Salvesen 2004]. The large subunit p20 encompasses a highly conserved peptide motif in the catalytic center, QACXG (where X can be R, Q or D), characterized by a central cysteine, which is required for the enzymatic activity [Fan et al. 2005] (Figure 1).

4 Introduction

D D

Pro-domain p20 p10

QACXG Catalytic motif

Activation

p20

p10

p20

Figure 1: Proteolytic activation of Caspases. Simplified illustration of an inactive and active Caspase. Upon proteolytic activation Caspase precursors are cleaved after D, leading to the removal of the pro-domain and the formation of a large (p20) and a small (p10) subunit. Subsequent rearrangement of two p20 subunits comprising the catalytically essential QACXG motif (X being R, Q or D) and two p10 subunits result in the formation of an active hetero- tetrameric Caspase [Modified from Clarke and Tyler 2009].

Although the substrate specificity is different for every Caspase (Table 1), processing of many substrates is executed by more than one Caspase. In general, no distinct definition exists for a protein to be a Caspase substrate or not, but there are some rules to declare a protein as a rather bad, intermediate or good Caspase substrate. Regarding to the standard nomenclature of Schechter and Berger Caspase substrates expose a peptide of at least four amino acids P4-P3-P2-P1, which is cleaved after the C-terminal residue P1 [Schechter and Berger 1967]. The adjacent amino acids of the C- terminus are marked with an apostrophe (Figure 2). All Caspases prefer D at position P1, with some exceptions reported, in which Caspase substrates are processed after E at position P1 [Stennicke et al. 2000; Krippner-Heidenreich et al. 2001; Snipas et al. 2008]. Independent from the subclass the individual Caspase belongs to, all Caspases prefer E at position P3 [Timmer and Salvesen 2007]. In addition, Caspases require a small and uncharged amino acid at position P1’, with G, S or A being the common residues observed [Stennicke et al. 2000]. P4-P3-P2 residues are complementary for the interaction with the binding pocket of Caspases and the complete cleavage motif P4-P1’ is exposed to the intracellular environment [Pop and Salvesen 2009]. More recent studies even use the surrounding amino acids P10-P10’ of the cleavage site to define Caspase substrate specificity [Shen et al. 2010]. Caspase recognition of cellular

5 Introduction target proteins is further based on secondary, tertiary and quaternary structures, which display important substrate characteristics to predict Caspase recognition [Crawford and Wells 2011].

Caspase cleavage site

N-Terminal flanking region C-Terminal flanking region

P10……………………………….P4 P3 P2 P1 P1 P2 P3 P4…………………………...P10

Figure 2: Nomenclature of the Caspase recognition motif. Overview of the Caspase cleavage site nomenclature of a Caspase substrate. Amino acids N-terminally located from the cleavage site are named and counted P1, P2, P3 etc., whereas the C-terminal amino acids are depicted with an apostrophe, to communicate the amino acid preference of different Caspases [Modified from Shen et al. 2010].

As previously mentioned apoptotic cell death is executed by the Caspase-dependent proteolytic processing of a broad spectrum of target proteins, including building blocks of the cytoskeleton, transcription factors, mediators of translation, cytokines and components of several cellular organelles, resulting in an apoptotic phenotype [Martin and Green 1995; Fischer, Jänicke, and Schulze-Osthoff 2003; Taylor, Cullen, and Martin 2008]. To date, several hundred proteins in humans exhibit a recognition sequences that match the consensus Caspase substrate specificity [Crawford and Wells 2011]. Indeed, the list of in vitro Caspase substrates is increasing rapidly, but in most cases the biochemical or pathological relevance of Caspase-mediated processing of specific target proteins remain unsolved. Beside Caspase substrates that show either a gain- or a loss-of-function phenotype, many processed proteins probably represent innocent bystanders during apoptosis [Lüthi and Martin 2007; Timmer and Salvesen 2007; Taylor, Cullen, and Martin 2008]. Other studies have demonstrated that domain- containing parts of Caspase targets were separated upon Caspase-mediated cleavage, leading to the generation of stable polypeptides fragments with potentially functional activities [Johnson and Kornbluth 2008]. The best known Caspase substrates that become activated and represent a clear gain- of-function mode are Caspases themselves as they are expressed as inactive zymogens, which get activated through proteolytic processing [Fuentes-Prior and Salvesen 2004]. Effector Caspase-3 for example is a substrate of Caspase-8 and the

6 Introduction cleavage site fits the consensus of the apical Caspase [Stennicke et al. 1998]. Once activated, effector Caspase-3 is capable of executing the processing of downstream targets [Walsh et al. 2008]. Another example for a protein that is activated following Caspase-8-mediated cleavage is the BH3-interacting-domain death agonist (BID). Cleavage of BID by Caspase-8 leads to the translocation of the truncated form of BID (tBID) to the mitochondria, whereupon BCL-2-associated X protein (BAX) and BCL-2 antagonist or killer (BAK) oligomerize. Penetration of the mitochondrial outer membrane by the integration of oligomerized BAX and BAK molecules results in the collapse of mitochondrial structure and the release of Cytochrome c, additionally accelerating the apoptotic process [Tait and Green 2010]. Several reports have demonstrated a loss-of-function role for the proteolysis of Caspase substrates, including many constituents of the cytoskeleton. Caspase-dependent cleavage of components of the microfilament such as actin and myosin, as well as intermediate filament proteins, like vimentin and nuclear lamins contributes to the retraction and shrinkage of the cell [Mashima, Naito, and Tsuruo 1999; Communal et al. 2002; Byun et al. 2001; Cowling and Downward 2002]. Two more Caspase substrates, which belong to the loss-of-function category are the poly(ADP-ribose) polymerase 1 (PARP-1) and the Inhibitor of Caspase-activated DNase (ICAD). Both proteins are processed by Caspase-3 and PARP-1 loses its protective role in DNA-repair and other important cellular processes such as proliferation and differentiation [Lazebnik et al. 1994]. ICAD is known to be the antagonist of the endonuclease Caspase-activated DNase (CAD), hence the name. The complex of CAD and ICAD is disrupted upon Caspase-3 activation, allowing CAD to form catalytically competent homodimers and translocation to the nucleus, where it cleaves the DNA into oligonucleosomal fragments [Liu et al. 1997; Sakahira, Enari, and Nagata 1998]. Active Caspases are capable to cleave various proteins and Caspase-dependent hydrolysis of specific target proteins is an important biochemical mechanism. Caspase activation occurs in distinct cellular processes, including inflammation and apoptosis. The following chapter describes two major signaling pathways of apoptosis based on the activation of different Caspases.

7 Introduction

1.2.3 Intrinsic and extrinsic pathway

Caspase-mediated apoptosis can be subdivided into two major signaling pathways and several players enable crosstalk between both biochemical processes (Figure 3). The intrinsic pathway initiates apoptosis upon intracellular signals and involves mitochondria as central cellular compartments [Long and Ryan 2012]. A great variety of intracellular stress signals ranging from DNA-damage, endoplasmic reticulum stress, to growth factor deprivation or hypoxia can trigger the induction of the intrinsic pathway [Pereira and Amarante-Mendes 2011]. Independent from the initiating stimulus, all intracellular stress signals cause mitochondrial outer membrane permeabilization (MOMP), which is executed by mitochondrial lipids, regulators of the bioenergetic metabolic flux and two pro-apoptotic family members of the B-cell lymphoma 2 (BCL-2) family, BAX and BAK [Green and Kroemer 2004]. Mitochondrial pore formation by multimerization of BAX and BAK finally leads to the release of pro-apoptotic factors such as Cytochrome c, second mitochondrial activator of Caspases (Smac/DIABLO), serine protease high-temperature requirement protein A2 (HtrA2/Omi), apoptosis inducing factor (AIF) and endonuclease G from the mitochondrial inter-membrane space [Saelens et al. 2004; Vaux 2011]. Once Cytochrome c is released from mitochondria it binds deoxyadenosine triphosphate (dATP) and subsequently induces Caspase-3 activation through the formation of a multi-protein complex, called the apoptosome, comprising of Cytochrome c, apoptotic protease activating factor 1 (Apaf-1) and Caspase-9 [Cain et al. 2000]. Binding of Cytochrome c to Apaf-1 leads to heptameric oligomerization and the conformational change of each Apaf-1 protein results in the exposure of its CARD domain, which allows interaction with the CARD domain of pro- Caspase-9 [Zhou et al. 1999; Acehan et al. 2002; Yuan et al. 2010]. Within the apoptosome, apical pro-Caspase-9 molecules are brought into close proximity activating themselves by autoproteolysis, whereupon effector Caspases are cleaved and trigger apoptosis through processing of several cellular target proteins [Riedl and Salvesen 2007] (Figure 3). Activation of the Caspase cascade is accompanied with MOMP and Cytochrome c release. In addition, cells have developed other regulatory mechanisms to provoke apoptosis through the release of pro-apoptotic proteins from the mitochondrial inter- membrane space. Efflux of Smac induces apoptosis by antagonizing members of the inhibitors of apoptosis (IAP) family [Verhagen et al. 2000]. Endonuclease G and AIF

8 Introduction cause apoptosis in a Caspase-independent manner, as both proteins enter the nucleus upon mitochondrial release, where they are involved in the fragmentation of genomic DNA [L. Y. Li, Luo, and Wang 2001; Candé et al. 2002; Guido Kroemer and Martin 2005]. The extrinsic pathway is stimulated by the engagement of death receptors at the plasma membrane, upon binding of their cognate ligands. To date, the best characterized death receptors, which belong to the tumor necrosis factor receptor (TNFR) superfamily, include Fas (DR2/CD95/APO-1), TNFR1 (DR1/CD120a), TNF-related apoptosis- inducing ligand receptor 1 (TRAILR1/DR4/APO-2) and TNF-related apoptosis-inducing ligand receptor 2 (TRAILR2/DR5/KILLER) [Walczak and Krammer 2000; Lavrik, Golks, and Krammer 2005; Fulda and Debatin 2006]. These death receptors share a significant homology in their C-terminal extracellular located part and feature cysteine- rich domains. The N-terminal intracellular protein structure is defined by a so called death domain, comprising a conserved sequence of ~80 amino acids, which plays a crucial role in the transduction of extracellular signals to intracellular signaling pathways [Ashkenazi and Dixit 1998; Fulda and Debatin 2006]. As the name indicates activation of death receptors is coupled to cell death, but stimulation of TNFR1 is known to initiate cytokine production and pro-inflammatory signaling delivered by nuclear factor kappa- light-chain-enhancer of activated B-cells (NF-κB), unless the inhibition of these survival pathways uncovers the cytotoxic potential of TNFR1 [Wajant, Pfizenmaier, and Scheurich 2003; Brenner, Blaser, and Mak 2015]. Binding of their respective ligands promotes trimerization of death receptors and clustering of intracellular adaptor proteins. Mediated by the receptors’ death domain, TNFR-associated death domain (TRADD) binds to trimeric TNFR1, while Fas-associated death domain (FADD) is associated with activated Fas receptors, TRAILR1 or TRAILR2 complexes. The adapter proteins in turn, lead to the recruitment of pro-Caspase-8 or its closest homolog Caspase-10 through their DED to form the death-inducing signaling complex (DISC) [Chinnaiyan et al. 1995; Hsu, Xiong, and Goeddel 1995]. Assembly of the DISC complex leads to subsequent autoproteolysis of the apical pro-Caspases and the initiation of the Caspase cascade. The intrinsic and the extrinsic pathways represent two independent signaling cascades within the cell for the regulation of apoptosis. However, there is an inseparable connection between both death-inducing signaling pathways as several players of each signaling cascade allow crosstalk. For instance, the precise mechanism of how Fas-

9 Introduction mediated apoptosis is executed depends on the cell type. In type I cells such as lymphocytes, Fas-mediated signaling is delivered through the DISC complex followed by the Caspase cascade, similar to TNFR1 or TRAIL1. In comparison, type II cells, like hepatocytes and pancreatic ß-cells, also implicate the intrinsic pathway, instead of strictly activating the Caspase cascade following Fas-induced apoptosis [Scaffidi et al. 1998; Fulda et al. 2002]. In this scenario the formation of DISC and Caspase-8 activation is insufficient to activate executioner Caspase-3 and in addition activated Caspase-8 has to cleave the pro-apoptotic factor BID to tBID, whereupon the latter translocates to the mitochondrial membrane (Chapter 1.2.2) (Figure 3). The integration of tBID is accompanied with the oligomerization of BAX or BAK, which initiates the intrinsic pathway of apoptosis via MOMP followed by the activation of Caspase-9 and executioner Caspase-3 and -7 [Igney and Krammer 2002]. In addition, implication of the intrinsic pathway has also been demonstrated for TRAIL signaling [Fulda et al. 2002]. Other studies have demonstrated that intrinsic stress signals, like DNA-damage can directly activate Caspase-8, which is typically implicated in death receptor-mediated cell death [Wesselborg et al. 1999; Chandra et al. 2004; Sohn, Schulze-Osthoff, and Jänicke 2005]. Recently, it was reported that low levels of DNA-damage drive NF-κB activation and expression of anti-apoptotic factors in order to promote cell survival in HeLa cells. If the DNA-damage is extensive, a feedforward signaling loop is activated with the expression of TNFα as an initial step. Autocrine TNFα production supports prolonged NF-κB activity, allowing the feedforward signaling loop to continue. This promotes a second TNFR-dependent signaling phase, including the receptor-interacting serine/threonine kinase 1 (RIPK1) as a key regulatory factor. TNFR-mediated autophosphorylation of RIPK1 leads to the recruitment of the adapter protein FADD, resulting in the initiation of the Caspase cascade upon Caspase-8 activation. The study demonstrates that DNA-damage can convert pro-inflammatory cytokines into pro- apoptotic signals in cells expressing death receptors [Biton and Ashkenazi 2011]. Observations from cells that are insensitive to autocrine death receptor ligands demonstrate that DNA-damage leads to the formation of a death-inducing signaling platform, referred to as the Ripoptosome. It consists of activated RIPK1, FADD and Caspase-8 and their assembly is facilitated upon genotoxic stress-mediated depletion of several negative regulators of the core components of the Ripoptosome. They further demonstrate that the formation of the Ripoptosome is independent from death receptor signaling and mitochondrial pathways. Once RIPK1 is recruited to the complex it is

10 Introduction capable to stimulate Caspase-8-mediated apoptosis [Tenev et al. 2011; Feoktistova et al. 2011]. The intrinsic and extrinsic pathways are two major signaling processes involved in the induction of apoptosis. Both signaling cascades are activated by specific cell death- inducing stimuli and possess several individual regulatory proteins besides Caspases. The intrinsic pathway is triggered through intracellular stress signals, including DNA- damage, which leads to MOMP, activation of Caspase-9 within the apoptosome and Caspase-3 activation. The extrinsic pathway is characterized by death receptor- mediated activation of Caspase-8 followed by Caspase-3 activation. Recent studies observed an important role for DNA-damage-induced Caspase-8 activation, which can be death receptor-dependent or -independent, enabling novel opportunities to selectively kill apoptosis-resistant cancer cells.

11 Introduction

Extracellular death signal TNFα

TNFR1

Plasma membrane TRADD

Intracellular death signal

Caspase-8

BAX BAK (inactive)

Apaf-1 Cyt. c Caspase-8 Mitochondrium tBID (active) BID Caspase-9 (inactive)

Caspase-3 (inactive)

Caspase-9 (active)

Caspase-3 Apoptosis (active)

Figure 3: Intrinsic and extrinsic pathway. Schematic depiction of the intrinsic and extrinsic pathway of apoptosis. Intracellular death signals (e.g. DNA-damage) result in the oligomerization of BAX and BAK in the mitochondrial outer membrane and MOMP. Cytochrome c (Cyt. c) is released from the mitochondria into the cytosol and forms together with Apaf-1 and Caspase-9 a heptameric complex called the apoptosome (only two Caspase-9 molecules are shown). Activated Caspase-9 itself activates effector Caspases (e.g. Caspase-3), whereupon apoptosis is executed. The extrinsic pathway is initiated by binding of extracellular death ligands (e.g. TNFα) to their cognate death receptors (e.g. TNFR1), leading to their trimerization. Adapter proteins (e.g. TRADD) are recruited to the receptors and Caspase-8 binds to the adapter proteins, thereby forming the DISC platform. Formation of DISC brings Caspase-8 molecules into close proximity, which leads to their autoactivation. Active Caspase-8 cleaves Caspase-3, resulting in apoptotic cell death. Caspase-8- mediated cleavage of BID results in the integration of truncated BID into the mitochondrial membrane and allows crosstalk between the extrinsic and intrinsic pathway as truncated BID leads to the oligomerization of BAX and BAK, thereby initiating the intrinsic pathway [Modified from Tait and Green 2010]. 12 Introduction

1.2.4 Regulation of apoptosis

Apoptosis is a powerful cellular process and activation of the Caspase cascade marks the so called “point of no return” for the cell. Therefore, programmed cell death is tightly controlled within the cell by various proteins that either inhibit or promote the apoptotic machinery. The BCL-2 family consists of critical regulators of the intrinsic pathway and can be separated in a pro-apoptotic and an anti-apoptotic group, each of which comprising several protein members. Anti-apoptotic BCL-2 family members are B-cell lymphoma 2 (BCL-2), BCL extra-large (BCL-XL), myeloid leukemia cell differentiation protein 1 (MCL-1), BCL-W and BCL-2A1. Pro-apoptotic BCL-2 family members can be further subdivided into direct effectors comprising of BAX, BAK and BCL-2 ovarian killer protein (BOK) and sensitizing BCL-2 homology domain 3 (BH3)-only proteins with BID, BCL-2- interacting mediator of cell death (BIM), BCL-2 antagonist of cell death (BAD), NOXA and p53 up-regulated modulator of apoptosis (PUMA) being the best characterized family members among others [Tait and Green 2010]. A central biochemical process during the activation of the intrinsic pathway is MOMP, initiated by oligomerization of the pro-apoptotic BCL-2 family members BAX and BAK, whereupon the death-inducing factors Cytochrome c and Smac are released (Chapter 1.2.3). On the one hand, oligomerization of BAX and BAK is facilitated by BH3-only family members such as BID and BIM [Eskes et al. 2000; Kuwana et al. 2005; Strasser 2005]. On the other hand, mitochondrial pore formation mediated by BAX and BAK is inhibited by binding of anti-apoptotic BCL-2 family members BCL-2, BCL-XL and MCL-1, thereby preserving the mitochondrial integrity [Green and Kroemer 2004; Youle and Strasser 2008]. Pro-survival BCL-2, BCL-XL and MCL-1 are neutralized by BID, BIM, BAD, which are activated upon initiation of the intrinsic pathway via irradiation, growth factor depletion, or chemotherapeutic agents [Kaufmann and Vaux 2003]. NOXA and PUMA are transcriptional targets of p53, which is a critical mediator in DNA- damage signaling. Transcriptional induction of both pro-apoptotic proteins interfere with the anti-apoptotic BCL-2 family members, thereby reverting their pro-survival potential [Oda et al. 2000; Yu and Zhang 2008]. A negative regulator of the extrinsic pathway is the cellular FLICE-like inhibitory protein (cFLIP), which is also capable to bind to the adapter proteins, thereby preventing the recruitment of initiator Caspases to the DISC complex [Krueger et al. 2001; Benedict,

13 Introduction

Norris, and Ware 2002; Safa 2012]. This mechanism is exploited in different tumor types expressing increased cFLIP protein levels to overcome induction of cell death and explains the promising role of cFLIP as a target candidate for cancer therapy [Safa and Pollok 2011]. cFLIP is a negative regulator during genotoxic drug-induced apoptosis, as it prevents Caspase-8 activation and different isoforms of cFLIP appear to be critical regulators of the Ripoptosome, a death-inducing signaling platform [Tenev et al. 2011; Feoktistova et al. 2011]. Another regulatory mechanism occurring in apoptosis is delivered by the IAP family. IAPs are E3 ubiquitin ligases, which block both the extrinsic and intrinsic pathway at converging points, as they antagonize Caspase activation and interfere with several regulatory factors associated with cell death [Kocab and Duckett 2016]. The IAP family consists of eight proteins, with X-linked inhibitor of apoptosis (XIAP), cellular inhibitor of apoptosis 1 (cIAP-1) and cellular inhibitor of apoptosis 2 (cIAP-2) being the best characterized with overlapping cellular activities, although functionally non-equivalent [Gyrd-Hansen and Meier 2010]. XIAP is capable to ubiquitinate Caspase-3 through its E3 ligase activity, which results in proteasomal degradation of the executioner Caspase [Suzuki, Nakabayashi, and Takahashi 2001]. XIAP can directly bind initiator Caspase-9 and prevent its activity within the apoptosome complex, and can also inhibit effector Caspase-3 and -7 by preventing substrate accessibility [Srinivasula et al. 2001; Bratton et al. 2001; Eckelman, Salvesen, and Scott 2006]. In contrast, cIAP-1 and -2 are able to bind Caspase-3 and -7, but fail to inhibit enzymatic activity [Eckelman and Salvesen 2006]. cIAP-1 induces proteasomal degradation through ubiquitination of both Caspases [Choi et al. 2009]. Furthermore, Caspase-8 activation is negatively regulated by cIAP-1 and -2 within death receptor signaling mediated by TNFR1 [Salvesen and Duckett 2002]. IAPs decrease Caspase stability through their E3 ligase activity and can further interfere with distinct molecular factors during programmed cell death. For example, expression of cIAP-1 and -2 promotes ubiquitination and proteasomal degradation of RIPK1 [Bertrand et al. 2008]. RIPK1 is known to be implicated in death receptor-mediated activation of Caspase-8 upon autocrine TNFα signaling mediated by extensive DNA- damage [Biton and Ashkenazi 2011]. In cells that are insensitive to death receptor signaling, DNA-damage-induced apoptosis is executed by the Ripoptosome, a death- inducing signaling platform that includes RIPK1. IAP-mediated proteasomal degradation

14 Introduction of RIPK1 and other components of the Ripoptosome negatively regulate the death inducing signaling complex [Tenev et al. 2011, Feoktistova et al. 2011]. Anti-apoptotic signaling via IAP proteins is eliminated by the interaction with Smac following its release from the mitochondrial inner membrane space into the cytosol [Du et al. 2000; Verhagen et al. 2000; Huang et al. 2003]. In contrast, turnover of Smac is implemented by cIAP-1 and -2 through their ubiquitin-protein ligase activity [Hu and Yang 2003]. Another regulatory protein of XIAP and cIAP-1 and -2 is HtrA2, which is also released during MOMP and inactivates these proteins through irreversible proteolytic processing [Verhagen et al. 2002]. Apoptosis is a tightly regulated biochemical process and the interplay of pro-apoptotic or anti-apoptotic factors determine cell fate. Deregulation of these factors can alter the apoptotic process and is associated with human diseases. The role of apoptosis in human diseases is described in the following chapter.

1.2.5 Role of apoptosis in human diseases

Apoptosis has a critical role in various developmental processes, including the immune system and the nervous system, in which initial overproduction of cells is followed by death of those cells that are not functional [Opferman and Korsmeyer 2003; Nijhawan, Honarpour, and Wang 2000]. It is necessary to eliminate pathogen-invaded or auto- aggressive immune cells and apoptosis represents a central aspect in wound healing through the removal of inflammatory cells [Lamkanfi and Dixit 2010; Elmore 2007; Greenhalgh 1998]. Apoptosis illustrates an important cellular process that requires tight regulatory mechanisms to enable development and homeostasis in adult tissues. Therefore, deregulation of apoptotic processes characterized by insufficient or excessive apoptotic activities within the cell is often coupled to distinct human diseases such as autoimmune diseases, neurodegenerative diseases and cancer [Fadeel and Orrenius 2005]. Acquired immunodeficiency syndrome (AIDS) is an example, where infection of CD4 + T- cells with the human immunodeficiency virus (HIV) results in excessive apoptosis rates. Penetration of the target cells and subsequent internalization of the virus causes mitochondrial catastrophe and elevated expression of Fas and its cognate ligand [Selliah and Finkel 2001].

15 Introduction

Deregulation of apoptosis is frequently linked to neurodegenerative diseases, like amyotrophic lateral sclerosis, Huntington´s disease and Alzheimer’s disease. Alzheimer’s disease as an example is characterized by the deposition of amyloid plaques formed by accumulated Aβ peptides in the brain. Aβ peptides are generated by Caspase-3-mediated processing of the amyloid precursor protein (APP) and exposure of Aβ peptides to cultured neurons can directly initiate apoptosis by causing oxidative stress or by triggering accelerated expression of Fas ligand [Gervais et al. 1999; Weidemann et al. 1999; Loo et al. 1993; Ethell and Buhler 2003]. Furthermore, APP serves as a substrate of Caspase-6, -8 and -9, whereupon the C-terminal peptide C31 is formed and induces neuronal cell death independently from Aβ-induced neuronal apoptosis [Lu et al. 2000; Bredesen 2009]. Deregulation of apoptosis is also associated with tumorigenesis. Human tumors feature genetic alterations and evading programmed cell death is a significant component in the development and progression of cancer [Hanahan and Weinberg 2011]. Programmed cell death can be prevented by the suppression of pro-apoptotic factors or by overexpression of anti-apoptotic proteins. More than 50% of human cancers harbor mutations in the p53 gene, encoding for a pro-apoptotic protein, which acts as a tumorsuppressor and is referred to as “the guardian of the genome” [Lane 1992; Soussi et al. 2006]. As a transcription factor, p53 initiates the expression of the pro-apoptotic protein PUMA, as well as NOXA, BAX and BID upon genotoxic stress [Nakano and Vousden 2001; Oda et al. 2000; Toshiyuki and Reed 1995; Sax et al. 2002]. Fas and its respective ligand, two key components of the extrinsic pathway are additional target genes of p53 [Owen-Schaub et al. 1995; Maecker, Koumenis, and Giaccia 2000]. Furthermore, p53 is involved in cell cycle arrest during the DNA-damage response and activates DNA repair proteins, thereby maintaining the genomic integrity or initiating apoptotic cell death if DNA-damage is too far advanced [Reinhardt and Schumacher 2012]. Another mechanism that is used by cancer cells to overcome the apoptotic program is an increased expression of pro-survival factors. Anti-apoptotic proteins such as BCL-2 or survivin have been shown to be highly expressed during carcinogenesis in distinct tumor tissues [Kitada et al. 2002; Adams and Cory 2010; Fukuda and Pelus 2006]. Expression of oncogenic Ras is known to stimulate anti-apoptotic phosphatidylinositol- 4,5-bisphosphate 3-kinase/AKT (PI3K/AKT) signaling [Marte and Downward 1997]. PI3K/AKT signaling is also activated by the loss of the tumor suppressor phosphatase

16 Introduction and tensin homolog (PTEN), that normally attenuates the PI3K/AKT survival signal [Cantley and Neel 1999].

1.2.6 Apoptosis in cancer treatment

Treatment of human malignancies can be subdivided into local and systemic approaches. Combinational treatment regimes are often applied to achieve a greater likelihood of cure. Surgery, the oldest method for cancer treatment, belongs to the local cancer treatments as it specifically targets the tumor tissue, while the rest of the body remains unaffected. Another local approach is radiation therapy, which is characterized by the usage of high-energy X-rays to treat tumor tissues in patients at local sites. Besides, radiation therapy is used to treat pain from cancerous origin when cures fail [Chabner and Roberts 2005]. In contrast, systemic approaches are defined by the administration of specific drugs, enzymes, antibodies and hormones, because they are delivered via the blood stream throughout the whole organism. Targeted therapy is a systemic approach and is characterized by the capability of administered compounds to interfere with specific target molecules necessary for tumor growth, progression and spread of cancer [Hanahan and Weinberg 2011] (Chapter 1.1). Resistance to cell death is considered to be one hallmark of cancer development and overcoming this barrier using targeted therapy approaches is currently the focus of most anticancer drug development. For instance, inhibition of poly(ADP-ribose) polymerases (PARPs), enzymes involved in the repair of damaged DNA, is carried out in breast and ovarian cancers containing mutated breast cancer-1 and -2 (BRCA1 and BRCA2) genes [Weil and Chen 2011]. PARP inhibition is accompanied with collapse of the replication fork during DNA replication causing double strand breaks. In BRCA-deficient cells homologous recombination is not restored, double strand breaks remain and accumulate, ultimately inducing cell death [Livraghi and Garber 2015]. Another option to reactivate the apoptotic machinery is enabled by the inhibition of anti- apoptotic proteins of the BCL-2 family using BH3 analogues, which mimic the binding of BH3-only proteins to the hydrophobic pocket of anti-apoptotic BCL-2 family members [Delbridge and Strasser 2015]. ABT-737 for example inhibits BCL-2, BCL-XL and BCL- W, thereby exhibiting synergistic cytotoxicity with radiation in solid tumors and primary patient-derived cells [Oltersdorf et al. 2005]. ABT-737 has been validated to act alone or 17 Introduction in combination with various cytotoxic agents against many malignancies such as small lung cancers, multiple myeloma and distinct forms of leukemia [Shoemaker et al. 2006; Tahir et al. 2007; Chauhan et al. 2007; Kojima et al. 2006; Konopleva et al. 2006; Kang et al. 2007; Del Gaizo Moore et al. 2008]. Targeted therapy methods can also interfere with sustained proliferative signaling, another biological capability of cancers acquired during tumor formation. The epidermal growth factor receptor (EGFR) plays a vital role in the regulation of cell proliferation and is aberrantly activated in cancer cells [Normanno et al. 2006]. Among several EGFR targeting strategies, monoclonal antibodies and tyrosine kinase inhibitors have been developed and demonstrated clinical efficacy in patients with lung cancer, pancreatic cancer or metastatic breast cancer [Yewale et al. 2013]. Hormonal therapy also interferes with proliferative signaling in cancers. It can specifically target tumors derived from certain tissues, which require hormones to develop such as breast and prostate. Hormone therapy can be subdivided into two groups: those that block the production of hormones and those that interfere with their signaling to alter tumor growth that respond to hormones [Prat and Baselga 2008; Chang et al. 2014]. When no targeted approach is available, chemotherapy remains the standard systemic treatment for human malignancies. Chemotherapy comprises of various cytotoxic drugs, including anthracyclines and Topoisomerase (TOP) inhibitors among others and targets rapid dividing normal and malignant cells. Anthracyclines such as Doxorubicin are known to intercalate with DNA base pairs, thereby affecting several DNA processes, ultimately resulting in extensive DNA-damage and cell death [Tacar, Sriamornsak, and Dass 2013]. Although Doxorubicin has become one of the most effective cytotoxic drugs to treat cancer, it also possesses a lot of side effects, including cardiotoxicity [Volkova and Russell 2011]. Inhibition of TOPs, enzymes that rewinds supercoiled DNA, is used to drive cancer cells into apoptosis and these drugs either target TOP I or II. The chemotherapeutic agent Etoposide for example targets TOP II, thereby inhibiting DNA replication and transcription and inducing DNA double strand breaks [Nitiss 2009]. Although it is accompanied with various side effects, chemotherapy is an efficient approach to treat breast cancer patients. Breast cancer is a heterogeneous disease and the degree of heterogeneity correlates with disease progression and therapeutic resistance [Brooks, Burness, and Wicha 2015]. Breast cancer can be categorized by histological aspects, tumor grade, lymph node status or predictive molecular markers

18 Introduction such as expression of the estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2/neu) [Holliday and Speirs 2011]. Expression profiles comprising these receptors allows further classification of breast cancers in normal, luminal A, luminal B, HER2/neu positive and basal-like, also referred to as triple negative, with each subtype exhibiting different prognosis and treatment response [Perou et al. 1999; Perou et al. 2000; Sorlie et al. 2001]. Triple negative breast cancer cells (TNBCs) lack the expression of ER, PR and HER2/neu and are characterized by an aggressive phenotype and poor prognosis [Dietze et al. 2015]. TNBCs are insensitive to hormonal therapy targeting ER and PR, limiting this cancer subtype to cytotoxic chemotherapy and classical approaches [Isakoff 2010]. Hanahan and Weinberg proposed that the inhibition of biological capabilities important for tumorigenesis should impair tumor growth and progression. Strategies targeting apoptosis in the treatment of cancer have been demonstrated to be beneficial in multiple studies for several tumor entities. The development of chemical agents, which are capable to induce cell death, offers a broad spectrum of specific therapeutic approaches to treat cancer patients, besides surgery and radiation therapy. Cancer is a heterogeneous disease and individual subtypes among cancer entities exhibit limited access for targeted therapy approaches. Hence, different treatment strategies are combined to overcome the heterogeneity of cancer and to avoid drug resistance as demonstrated for neoadjuvant and adjuvant chemotherapy.

19 Introduction

1.3 PAR-4

1.3.1 Identification and structure of PAR-4

In 1994 prostate apoptosis response 4 (Par-4) was initially discovered in a differential hybridization screen as an early up-regulated gene in the androgen-independent rat prostate cancer cell line AT-3 undergoing apoptosis forced by intracellular elevation of calcium ions due to ionomycin exposure [Sells et al. 1994]. In addition, increased Par-4 mRNA levels were also observed in apoptotic androgen-dependent cells upon castration resulting in androgen ablation, while inhibition of calcium channels with nifedepine prior to castration prevented Par-4 gene expression [Sells et al. 1994]. Par-4 has been shown to be coupled specifically to calcium-dependent apoptotic signaling as serum stimulation, growth arrest, oxidative stress and necrosis failed to induce Par-4 gene expression in prostate cells [Sells et al. 1994]. PAR-4, also known as PRKC, Apoptosis, WT1, Regulator (PAWR), is evolutionary conserved in vertebrates and ubiquitously expressed in mammalian tissues to varying degrees [El-Guendy and Rangnekar 2003]. Even though PAR-4 gene expression is not restricted to any specific organ, some maturated cell types such as terminally differentiated ductal cells from the epithelia of the mammary and the prostate glands in rats show low PAR-4 protein levels, indicating that the expression of PAR-4 is tightly regulated during differentiation [Boghaert et al. 1997]. The gene encoding for human PAR-4 is localized to the minus strand of chromosome 12q21 and comprises seven exons and six introns [Johnstone et al. 1998]. PAR-4 consists of 340 amino acids with a predicted molecular weight of approximately 42 kDa, while rat and mouse Par-4 are built up of 332 or 333 amino acids respectively. Rat and mouse Par-4 display amino acid similarity of 93%, whereas human PAR-4 exhibits 75% sequence homology with its rat counterpart and shares 84% functional identical amino acids [Shrestha-Bhattarai and Rangnekar 2010]. PAR-4 possesses a unique and central “selective for apoptosis in cancer cells” (SAC) domain (amino acids 146 – 203), encompassing a nuclear localization sequence (NLS) (amino acids 147 - 163) and experiments with several deletion mutants demonstrated that nuclear translocation of the SAC domain is sufficient to induce apoptosis in cancer cells but not in normal cells [El-Guendy et al. 2003] (Figure 4). At the C-terminus PAR-4 features a leucine zipper (LZ) domain (amino acids 300 – 340), which is important for

20 Introduction the pro-apoptotic activities of PAR-4 as it allows PAR-4 to interact with other proteins, including Wilms tumor protein 1 (WT1), atypical isoforms of protein kinase C (aPKC), DAP-like kinase (DLK), Protein kinase B (PKB/AKT), Androgen receptor (AR), TOP I or F-box protein 45 (FBXO45) [Johnstone et al. 1996; Diaz-Meco et al. 1996; Page et al. 1999; Goswami et al. 2005; Gao et al. 2006; Goswami et al. 2008; Chen et al. 2014] (Figure 4). Both key domains are conserved to 100% across human, rat and mouse homologs, indicating their importance for the pro-apoptotic activities of PAR-4 [Hebbar, Wang, and Rangnekar 2012]. The protein sequence of PAR-4 offers several consensus sites for post-translational modifications such as phosphorylation. aPKC was one of the first identified PAR-4 interacting proteins and PAR-4 exhibits several potential aPKC phosphorylation sites [Diaz-Meco et al. 1996]. Rat Par-4 is phosphorylated by Protein kinase A (PKA) at T155 in vitro and in vivo. The PKA phosphorylation site is conserved in humans (T163) and it was shown that this posttranslational modification within the SAC domain is required for the pro-apoptotic properties of the protein [Gurumurthy et al. 2005]. In contrast, phosphorylation of S249 of rat Par-4 by AKT has been described to inhibit the apoptotic activities of Par-4 through the recruitment of 14-3-3 proteins, which sequesters Par-4 from its nuclear translocation [Goswami et al. 2005]. However, this AKT phosphorylation site of Par-4 is not conserved in mouse and human PAR-4 and other AKT phosphorylation sites have not been reported. Furthermore, recent studies have demonstrated that PAR-4 from human and rat origins both are new substrates of Casein kinase 2 (CK2). CK2-mediated phosphorylation of S223 in rat Par-4 and S231 in the human counterpart both strongly impair the apoptotic functions [de Thonel et al. 2014].

21 Introduction

WT1 1 146 163 203 231 300 340 aPKC N DLK L SAC LZ AKT S AR TOP I Human PAR-4 FBXO45

PKA CK2

Figure 4: Structure of human PAR-4. Simplified illustration of human PAR-4 protein. PAR-4 consists of 340 amino acids and is characterized by two important domains, which play a central role for its pro-apoptotic activities. In the C-terminal part PAR-4 features a leucine zipper (LZ, amino acids 300 - 340) domain, which allows potential interaction with Wilms tumor protein 1 (WT1), atypical isoforms of protein kinase C (aPKC), DAP-like kinase (DLK), Protein kinase B (PKB/AKT), Androgen receptor (AR), Topoisomerase I (TOP I) or F-box protein 45 (FBXO45). The middle part of the protein harbors a unique domain, called selective for apoptosis in cancer cells (SAC, amino acids 146 - 203) domain. The SAC domain is of great importance to induce apoptosis, as the name indicates. The SAC domain encompasses a nuclear localization sequence (NLS, amino acids 147 - 163), which enables PAR-4 to shuttle into the nucleus. Phosphorylation of PAR-4 by PKA at T163 promotes the pro-apoptotic functions, whereas phosphorylation at S231 by CK2 impairs the apoptotic activity [Modified from Johnstone et al. 1996; Diaz-Meco et al. 1996; Page et al. 1999; Gurumurthy et al. 2005; Goswami et al. 2005; Gao et al. 2006; Goswami et al. 2008; de Thonel et al. 2014; Chen et al. 2014].

1.3.2 Pro-apoptotic signaling of PAR-4

PAR-4 is a ubiquitously expressed protein that was initially discovered in androgen- independent prostate cancer cells undergoing apoptotic cell death in response to accelerated levels of intracellular calcium ions forced by ionomycin treatment. Moreover, this study monitored an increase in Par-4 mRNA levels in rat ventral prostates upon androgen ablation caused by castration, while uncastrated rats lack expression of the protein in this tissue [Sells et al. 1994]. Consistent with its pro-apoptotic activities increased PAR-4 gene expression was also observed in apoptotic tissues, including granulose cells of atretic ovarian follicles, the web area between the digits in the development of extremities in mice and in the tadpole tail during involution [El-Guendy and Rangnekar 2003]. PAR-4 mRNA and protein levels are increased in vulnerable neurons of patients with Alzheimer’s disease and cultured primary hippocampal cells from rats also exhibit elevated Par-4 expression during neuronal cell death. In contrast, knockdown of Par-4 in these primary cells

22 Introduction results in decreased apoptosis rates following treatment, whereas overexpression of Par-4 in neuronal PC12 cells sensitizes the cells to different apoptotic stimuli [Mattson et al. 1999]. Studies from Bieberich and colleagues reported asymmetric distribution of PAR-4 during the mitosis of neuronal progenitor cells. Daughter cells that exhibit high levels of PAR-4 undergo apoptosis, while those with low levels of PAR-4 survive and differentiate into neurons [Bieberich et al. 2003]. In general, ectopic expression of PAR- 4 has been shown to sensitize a variety of cells to apoptosis-inducing stimuli or chemotherapeutic agents, again underscoring its pro-apoptotic functions [Hebbar, Wang, and Rangnekar 2012]. The apoptotic program can be manipulated either by the activation of pro-apoptotic factors, or by the inhibition of pro-survival signaling molecules. Since PAR-4 has been linked to apoptotic cell death, the precise mechanisms by which PAR-4 orchestrates its pro-apoptotic activities were investigated. PAR-4 interferes with the pro-survival signaling of NF-κB, thereby preventing its anti-apoptotic activities. NF-κB is a transcription factor that regulates gene expression of a number of pro-survival genes, including the BCL-2 family members BCL-XL and BCL-2, cFLIP and IAP proteins [Chen, Edelstein, and Gélinas 2000; Catz and Johnson 2001; Micheau et al. 2001; You et al. 1997; Stehlik, de Martin, Kumabashiri, et al. 1998; Stehlik, de Martin, Binder, et al. 1998]. PAR-4 can translocate to the nucleus, where it abrogates transcriptional activity of NF-κB [Nalca et al. 1999; Gurumurthy et al. 2005]. These findings were confirmed in overexpression studies, in which the RelA/p65 subunit of NF-κB was capable to revert the inhibitory effect of PAR-4 on the NF-κB activity [Chakraborty et al. 2001]. Overexpression of PAR-4 activates the extrinsic pathway in prostate cancer cells by enabling the translocation of Fas and its cognate ligand to the plasma membrane, whereupon the DISC platform is generated and the Caspase cascade initiated [Chakraborty et al. 2001]. It was further demonstrated that PAR-4 expression prevents nuclear entry of RelA/p65. Cytoplasmic retention of NF-κB is mediated by the interaction of PAR-4 with aPKCs, whereupon the activity of the inhibitor of NF-κB kinase (IKK) accompanied with the degradation of the inhibitor of NF-κB (IκB) is abolished [Diaz- Meco et al. 1999]. Furthermore, it was reported that Nucleoporin/p62 protects cells from PAR-4-mediated inhibition of NF-κB and apoptosis, in an aPKC-dependent manner [Chang, Kim, and Shin 2002]. Another mechanism how PAR-4 can overcome pro-survival signaling within the cell is mediated by the interaction with the transcription factor WT-1, one of the first binding

23 Introduction partners of PAR-4 [Johnstone et al. 1996]. Like WT-1, PAR-4 is able to bind to a WT-1 binding site on the BCL-2 promoter in vitro and in vivo upon nuclear translocation, resulting in decreased expression of the anti-apoptotic protein BCL-2 [Cheema et al. 2003]. Down-regulation of BCL-2 protein levels were observed upon PAR-4-mediated apoptosis before [Qiu et al. 1999]. PAR-4 can also bind to the pro-apoptotic kinase DLK, also known as zipper interacting protein kinase (ZIPK). DLK has been shown to interact and recruit death-associated protein 6 (Daxx) to promyelocytic leukemia (PML) nuclear bodies following induction of apoptosis with arsenic trioxide or interferon gamma. Interestingly, association of DLK and Daxx was accelerated by PAR-4, resulting in enhanced Caspase activity and apoptosis and suppression of each of these proteins reveals a decrease in apoptosis mediated by arsenic trioxide or interferon gamma [Kawai, Akira, and Reed 2003]. In parallel, PAR-4 was found in PML nuclear bodies, where it co-localizes with the pro- apoptotic protein THAP domain-containing protein 1 (THAP1) in primary endothelial cells, fibroblasts and in blood vessels in vivo [Roussigne et al. 2003]. Although the precise mechanism is unknown, forced PAR-4 gene expression also leads to the relocation of DLK from the nucleus to the cytoplasm and association to actin filaments, provoking reorganization of the cytoskeleton and an apoptotic phenotype [Page et al. 1999]. Moreover, binding of PAR-4 to actin filaments results in the generation of actin bundles in vitro and in vivo and is proposed to serve as a scaffold to recruit DLK, which then phosphorylates its substrate myosin light chain (MLC), upon induction of apoptosis [Vetterkind et al. 2005; Boosen et al. 2009; Vetterkind and Morgan 2009]. In a more recent study Alvarez and co-workers investigated the role of PAR-4 during apoptosis in diverse breast cancer models in mice and its relevance in tumor formation. PAR-4 was observed to be up-regulated upon oncogenic HER2/neu inhibition and treatment with chemotherapeutic drugs induced DLK-mediated phosphorylation of MLC in a PAR-4- dependent manner. Finally, they demonstrated that deregulated MLC phosphorylation is linked to multinucleation and ultimately apoptotic cell death due to cytokinesis failure [Alvarez et al. 2013].

1.3.3 Tumor suppressor functions of PAR-4 and deregulation in human diseases

Because of its pro-apoptotic activities and tumor suppressor functions, PAR-4 is often deregulated in human diseases, especially during cancer formation and progression. 24 Introduction

PAR-4 knockout mice show reduced lifespan, are prone to increased tumor formation and develop spontaneous carcinomas in various tissues, including lungs, liver, urinary bladder and endometrium. In addition, prostate intraepithelial neoplasia was observed in this study. These PAR-4 knockout mice showed also increased susceptibility to chemically induced urinary bladder carcinomas as well as hormone driven endometrial and prostatic lesions [García-Cao et al. 2005]. PAR-4 is known to possess tumor suppressor functions, which are closely connected with its pro-apoptotic activities. Experiments using xenograft mice models demonstrated that a single injection of an adenoviral construct of PAR-4 leads to regression of subcutaneous solid tumors within three weeks. These tumors were generated by implanting PC-3 cells into the flanks of the animals and PAR-4-mediated inhibition of the tumor development correlated with induction of apoptosis [Chakraborty et al. 2001]. In a parallel study, overexpression of PAR-4 in xenotransplanted human melanoma cells in mice resulted in decreased tumor volume due to apoptosis, indicating its tumor suppressor functions [Lucas et al. 2001]. Delivery of PAR-4 via nanoliposome application intravenously in mice harboring transplanted colon cancer cells sensitized the tumors to therapeutic treatment with fluorouracil [Kline et al. 2009]. In hematopoietic cells PAR-4 showed anti-transforming capacity by antagonizing oncogenic signaling mediated by breakpoint cluster region Abelson kinase (BCR-ABL) [Kukoc-Zivojnov et al. 2004]. Furthermore, endogenous PAR-4 has been implicated to prevent Ras-induced cellular transformation, validating PAR-4 as a tumor suppressor protein [Goswami et al. 2008]. It has been reported that the central SAC domain of PAR-4 facilitates its pro-apoptotic activities and overexpression of this unique domain, which is 100% conserved among humans and many other species, including rodents, specifically induces apoptosis in cancer cells [El-Guendy et al. 2003]. Consistent with the pro-apoptotic activities, ubiquitous expression of a SAC transgene in mice leads to an increased life span and resistance to spontaneous and oncogene-induced adenocarcinoma of the prostates. However, mice expressing the SAC domain show normal development, designated by fertility, viability and aging [Zhao et al. 2007]. Recent studies reported that lentiviral- mediated expression of the SAC domain significantly extended the survival of patient- derived glioblastoma stem cell mice xenografts [Liu et al. 2014]. PAR-4 was found to localize to the minus strand of chromosome 12q21, a region, which is often deleted or shows increased instability in pancreatic and gastric cancers [Kimura

25 Introduction et al. 1998; Schneider et al. 2003]. Down-regulation of PAR-4 was observed in a variety of cancer tissues such as renal cell carcinoma, neuroblastoma, acute lymphoblastic and chronic lymphocytic leukemia, endometrial cancer, pancreatic cancer, as well as lung and prostate cancers [Cook et al. 1999; Kögel et al. 2001; Boehrer et al. 2001; Moreno- Bueno et al. 2007; Ahmed et al. 2008; Joshi et al. 2008; Fernandez-Marcos et al. 2009]. PAR-4 expression levels were markedly decreased in proliferating tumorigenic cholangiocytes and in patient-derived high-grade gliomas [Franchitto et al. 2010; Y. Liu et al. 2014]. Mutational analysis in endometrial cancer cell lines and carcinomas of the gene sequence encoding for PAR-4 by Moreno-Bueno and colleagues revealed a point mutation in exon 3, resulting in a stop codon and a truncated protein without any functions. They also demonstrated that down-regulation of PAR-4 in endometrial cancer cell lines is due to promoter hyper-methylation. In addition, hyper-methylation of the PAR-4 promoter was found in 19 of 59 endometrial tumors [Moreno-Bueno et al. 2007]. Similar results were obtained in Ras-transformed epithelial cells, which exhibit decreased PAR-4 protein levels achieved by hyper-methylation of the PAR-4 promoter in a mitogen-activated protein kinase kinase (MEK)-dependent manner [Pruitt et al. 2005]. Down-regulation of PAR-4 has been shown to be critical for Ras-induced survival and tumor progression before [Barradas et al. 1999]. However, down-regulation of PAR- 4 was also observed in glioblastomas with mutant isocitrate dehydrogenase 1 (IDH1) and was associated with PAR-4 mRNA degradation and decreased promoter activity. Conversely, PAR-4 promoter hyper-methylation was not increased in these cells, although mutant IDH1 is known to promote global gene methylation [Liu et al. 2014; Horbinski 2013]. Taken together, there is evidence for a wide range of molecular mechanisms that regulate the expression of the pro-apoptotic tumor suppressor protein PAR-4 in different tissues. Since PAR-4 is known to exhibit pro-apoptotic functions in cancer cells and serves as a tumor suppressor, mechanisms that regulate its protein expression have been investigated in a variety of tumor entities as described previously. Over the last years, the role of PAR-4 in breast cancer was investigated as breast cancer is the most common malignancy and the second leading cause of cancer deaths among women, besides lung cancer [Theriault et al. 2013]. First reports analyzing the expression of PAR-4 demonstrated that the tumor suppressor is down-regulated in patient-derived breast cancer tissues [Zapata-Benavides et al. 2009; Méndez-López et al. 2010]. Functional inactivation of PAR-4 correlates with an aggressive tumor phenotype and

26 Introduction serves as a prognostic and predictive marker for breast cancer [Méndez-López et al. 2010; Nagai et al. 2010]. In a study from 2013 Alvarez and co-workers focused on PAR- 4 during tumor relapse as most deaths from breast cancer result from tumor recurrence following primary therapy. Using three different mice models for breast cancer, they observed an almost complete loss of PAR-4 mRNA as well as protein levels in residual tumors, which have been derived from oncogene-induced primary tumors. Furthermore, they compared primary tumor biopsies and recurrent tumors from patients and found that PAR-4 expression decreased following neoadjuvant chemotherapy. Therefore, PAR-4 down-regulation serves as molecular mechanism that is necessary and sufficient to promote recurrence. As a negative regulator of breast cancer recurrence low amounts of PAR-4 are associated with poor response to neoadjuvant chemotherapy accompanied with an increased risk of relapse in patients suffering from breast cancer [Alvarez et al. 2013]. These observations provide a crucial inside into the molecular heterogeneity of breast cancer, underscoring the potential role of PAR-4 as a prognostic marker of breast cancer recurrence [Hebbar, Wang, and Rangnekar 2012].

27 Introduction

1.4 Aims of this study

Breast cancer has become the leading cause of cancer mortality among women worldwide. Most deaths from breast cancer result from disease relapse upon treatment of the primary tumor. The pro-apoptotic tumor suppressor protein PAR-4 is silenced in several tumor entities and recent studies have demonstrated that down-regulation of PAR-4 serves as an important mechanism for breast tumor recurrence. Furthermore, loss of PAR-4 expression allows tumor cells to survive tumor regression following targeted therapy and chemotherapy. Re-expression of PAR-4 can directly drive cancer cells into apoptosis or renders them sensitive to a broad spectrum of death-inducing stimuli such as genotoxic stress. PAR-4 exhibits a unique SAC domain required for its pro-apoptotic and tumor-suppressor activities. An important mechanism by which PAR- 4 orchestrates its molecular functions relies on the ability of its SAC domain to enter the nucleus during apoptosis, as it encompasses a NLS sequence. The C-terminal part of the PAR-4 protein comprises a LZ domain, allowing PAR-4 to interact with several other proteins and interfere with their function. PAR-4 is known to be modified at the posttranslational level and reveals various mechanisms to induce apoptosis in different cancer cells, but the precise mechanism how PAR-4 enables its pro-apoptotic functions in breast cancer cells is poorly understood. Therefore, a key objective of this work was to analyze how PAR-4’s pro-apoptotic functions are controlled and executed. In this study PAR-4 was identified as a Caspase-8 substrate, which separates the unstructured N-terminal part from the C-terminal part including its SAC and LZ domain. The consequences on cell growth of PAR-4 hydrolysis are investigated by the creation of stable cell lines that express wild-type PAR-4 or the Caspase cleavage resistant mutant PAR-4 D131G under the control of a Doxycycline-inducible promoter. To study the effects of Caspase-8-dependent PAR-4 cleavage on subcellular localization confocal microscopy is employed. In addition, to address the question whether the Caspase-8/PAR-4 axis is required for TNFα-induced or DNA-damage-mediated induction of apoptosis in breast cancer cells knockdown experiments using lentiviral- mediated shRNAs or transient transfected siRNAs targeting both proteins are analyzed by immunoblotting and validated via flow cytometry. Finally, identification of novel interaction partners are investigated using an unbiased mass spectrometry approach and potential interaction partners are validated by reciprocal co-immunoprecipitation experiments. 28 Experimental procedures

2 Experimental procedures

Materials and methods are described according to standard protocols used in the Institute of Biochemistry and Molecular Biology, RWTH Aachen University and modified regarding individual differences in experimental procedures.

2.1 Materials

2.1.1 Antibodies

Table 2: Antibodies

Antigen Species Company/ Cat no. Information Primary antibodies α-Caspase-3 rabbit Cell Signaling polyclonal #9662 IB α-Caspase-8 mouse Cell Signaling monoclonal #9746 IB α-GAPDH mouse Santa Cruz monoclonal sc-32233 IB α-GFP/CFP/YFP goat Rockland polyclonal 600-101-215 IB α-HA rabbit Santa Cruz polyclonal sc-805 IB α-PAR-4 rabbit Santa Cruz polyclonal sc-1807 IB α-PAR-4 rabbit Cell Signaling polyclonal #2328 IB, IF α-PAR-4 rabbit Abcam polyclonal ab5787 IB, IF α-PARP-1 rabbit Cell Signaling polyclonal #9542 IB α-T7 Novagen monoclonal #69522 IB Secondary antibodies α-rabbit-IgG-HRP goat DAKO polyclonal P0448 IB 29 Experimental procedures

α-goat-IgG-HRP rabbit DAKO polyclonal P0449 IB α-mouse-IgG- goat DAKO polyclonal HRP P0447 IB α-rabbit-IgG- donkey Thermo Fisher Scientific polyclonal Alexa-Fluor-488 A-21206 IF α-rabbit-IgG- donkey Thermo Fisher Scientific polyclonal Alexa-Fluor-555 A-31572 IF

2.1.2 Plasmids

Plasmid encoding for wild-type Caspase-8 was purchased from addgene and provided by Prof. Dr. Guy Salvesen’s laboratory. Empty pcDNA5/FRT/TO(+) special vector construct was a kind gift from Prof. Dr. Gerhard Müller-Newen. Empty pcDNA3, wild- type Caspase-3, rat Par-4 WT, pcDNA3 PAR-4 WT and D131G were provided by Dr. Jörg Hartkamp.

Table 3: Plasmids

cDNA Vector Epitope tag Source empty pcDNA3 - J. Hartkamp Aachen, Germany Caspase-3 WT pcDNA3 - J. Hartkamp Aachen, Germany Caspase-8 WT pcDNA3 HA, G. Salvesen C-terminal San Diego, USA rat Par-4 WT pcDNA3 - J. Hartkamp Aachen, Germany PAR-4 WT pcDNA3 T7, J. Hartkamp N-terminal Aachen, Germany PAR-4 D131G pcDNA3 T7, J. Hartkamp, N-terminal Aachen, Germany empty pcDNA5/FRT/TO(+) special - G. Müller-Newen, Aachen, Germany PAR-4 WT pcDNA5/FRT/TO(+) special - F. Treude PAR-4 WT pcDNA5/FRT/TO(+) special eCFP, F. Treude C-terminal

30 Experimental procedures

PAR-4 WT pcDNA5/FRT/TO(+) special FLAG, F. Treude C-terminal PAR-4 D131G pcDNA5/FRT/TO(+) special - F. Treude PAR-4 D131G pcDNA5/FRT/TO(+) special eCFP, F. Treude C-terminal PAR-4 132-340 pcDNA5/FRT/TO(+) special - F. Treude PAR-4 132-340 pcDNA5/FRT/TO(+) special eCFP, F. Treude C-terminal PAR-4 1-131 pcDNA5/FRT/TO(+) special eYFP, F. Treude N-terminal

2.1.3 Bacterial strains

In the following experimental procedures, E. coli DH5α bacterial strain was used for plasmid preparation and conventional cloning.

E. coli DH5α

F- endA1 glnV44 thi-1 recA1 relA1 gyrA96 deoR nupG Ф80dlacZΔM15 Δ(lacZYA- argF)U169, hsdR17(rΚ- mΚ+), λ-

2.1.4 Eukaryotic cell lines

In the following experimental procedures several human, epithelial and adherent cell lines were used.

Table 4: Cell lines

Cell line Tissue Disease BT-20 Mammary gland/breast carcinoma HeLa cervix adenocarcinaoma HeLa Flp-In™ T-REx cervix adenocarcinaoma HeLa S3 cervix adenocarcinaoma HEK293 embryonic kidney - HEK293-T embryonic kidney - 31 Experimental procedures

Hs 578T Mammary gland/breast carcinoma MCF-7 Mammary gland/breast adenocarcinoma MCF-10A Mammary gland/breast fibrocystic disease MDA-MB-231 Mammary gland/breast adenocarcinoma MDA-MB-468 Mammary gland/breast adenocarcinoma SK-BR-3 Mammary gland/breast adenocarcinoma T-47D Mammary gland/breast ductal carcinoma ZR-75-1 Mammary gland/breast ductal carcinoma

2.2 Conventional Cloning

Cloning procedures were planned and performed in close cooperation with Hildegard Schmitz-Van de Leur.

2.2.1 PCR

Polymerase chain reaction (PCR) was carried out for mutagenesis, or to insert restriction sites to the gene of interest (GOI) to be able to subclone it in the desired expression vector. Sense and anti-sense primers were designed and purchased from MWG-Biotech. PCR was performed in an automated thermal cycler using Phusion® High-Fidelity DNA Polymerase (NEB) according to the manufacturer’s instructions. Oligo Calculator version 3.26 [Kibbe 2007] was used to determine primer annealing temperatures by using the nearest-neighbor method [Breslauer et al. 1986]. The general pipetting scheme and PCR program is depicted below (Table 5 and 6). Generated DNA fragments were analyzed via agarose gel electrophoresis.

32 Experimental procedures

Table 5: PCR pipetting scheme

PCR pipetting scheme Component Volume Phusion® GC buffer (5x, NEB) 10 µl Phusion® High-Fidelity DNA Polymerase (2 U/µl, NEB) 0.5 µl Sense/Antisense primer (10 pmol/µl, MWG-Biotech) 1.5 µl dNTP’s (2.5 mM, Qiagen) 4 µl DNA template (1 ng/µl cDNA) 20 µl

ddH2O ad 50 µl

Table 6: PCR program

PCR program Step Temperature Time Cycles

Denaturation (initial) 98°C 30 sec 1 Denaturation 98°C 10 sec Annealing 50°C*1 40 sec 30 Extension 72°C 30 sec*2 Extension (final) 72°C 10 min 1 *1 Depending on the GC content of the primer *2 Depending on amplicon length

2.2.2 Agarose gel electrophoresis

Depending on their specific electrophoretic mobility amplified DNA fragments were separated in a 1% (w/v) agarose gel dissolved in TAE buffer supplemented with ethidium bromide (Roth) to a final concentration of 0.1 µg/ml. PCR probes were mixed with 5 µl of 10x DNA loading buffer and agarose gel electrophoresis was performed at 110 V for 45 min at RT. A 1 kB DNA ladder (Life Technologies) was applied as a marker for size. DNA fragments were visualized using a UV-Transilluminator (Intas).

33 Experimental procedures

TAE buffer 40 mM Tris (AppliChem) 20 mM Acetic acid (Merck) 1 mM EDTA (Sigma) pH 8.0

DNA loading buffer 25% (w/v) Ficoll 400 (Sigma) 0.4% (w/v) Bromphenol blue (Sigma) 0.4% (w/v) Xylencyanol (Serva)

2.2.3 Extraction/Purification of DNA fragments from agarose gel

Separated DNA amplicons were extracted and purified from agarose gels using the QIAquick® Gel Extraction Kit (Qiagen) according to the supplied manual. Instead of using elution buffer supplied with the kit DNA fragments were eluted with 50 µl ddH2O.

2.2.4 Restriction digest

Restriction enzymes from NEB were used for restriction digests to generate compatible ends for DNA ligation. Therefore, either 2 µg of plasmid DNA or 300 ng of purified PCR products were incubated with the individual restriction enzymes in a range of 2-10 units (U) and incubated at 37°C for 90 min. For efficient restriction digests the amount of enzymes did not exceed 10% of the total volume and supplements were added as described by the manufacturer.

2.2.5 DNA ligation

T4 DNA ligase (Fermentas) was used for the ligation of purified, compatible DNA fragments as described in the instructions attached. Therefore, typically 30 ng of the digested destination vector was incubated with the purified DNA fragments in a ratio of 3:1. The ligation reaction was carried out in 20 µl of 1x ligation buffer, incubated for 90 min at RT and finally was transformed into competent E.coli DH5α.

34 Experimental procedures

2.3 Work with prokaryotic cells

2.3.1 Bacterial cell cultivation

Microbial cell cultivation of E.coli DH5α was carried out in LB medium (Roth) or on LB agar plates supplemented with 100 µg/ml Ampicillin (Roth), respectively. Upon bacterial transformation, individual colonies were picked from LB agar plates and grown in 3 ml starter cultures of LB medium. Cells were incubated for 6 h at 37°C under permanent agitation at 175 rpm. Large scale purification of plasmid DNA was carried out from 250 ml LB medium that were inoculated with 2 ml of the starter culture and incubated o/n at 37°C under permanent agitation at 175 rpm.

LB medium 0.5% (w/v) LB medium (Roth) 100 µg/ml Ampicillin (Roth)

LB agar plates LB medium (Roth) 1.5% Bacto-Agar (Difco) 100 µg/ml Ampicillin (Roth)

2.3.2 Bacterial transformation and plasmid preparation

Transformation of individual plasmids into chemically competent E.coli DH5α was executed by heat shock. Therefore, microbial cells were thawed on ice and 50 µl of bacterial cell suspension was gently mixed in a microcentrifuge tube with either 1 ng of plasmid DNA or with 10 µl of ligation products on ice. After 30 min of incubation heat shock was performed for 90 s at 42°C and cells were immediately placed back on ice for 1 min. Finally, the bacterial cell suspension was transferred to LB agar plates and incubated o/n at 37°C. Small-scale plasmid DNA preparations from single picked colonies grown in a starter culture were generated with the QIAprep ® Spin Miniprep Kit

(Qiagen), according to the supplier’s manual, but eluted in 30 µl ddH 2O instead of elution buffer. Large-scale preparations of plasmid DNA were generated from 250 ml

35 Experimental procedures cultures using the HiSpeed® Plasmid Maxi Kit (Qiagen), according to the manufacturer’s instructions. Plasmid DNA was eluted in 500 µl ddH2O.

2.3.3 Determination of DNA concentrations

The yield of purified nucleic acid was estimated by measuring their absorbance at 260 nm using a Nanodrop™ ND-1000 (Peqlab). DNA purity was evaluated by the ratio of absorbance at 260 nm and 280 nm. A ratio of about 1.8 is generally accepted as “pure” for DNA.

2.4 Work with eukaryotic cells

2.4.1 Cell culture

All cell lines were cultured in humidified atmosphere at 37°C with 5% CO 2. HeLa, HeLa S3, HeLa Flp-In™ T-REx, HEK293, HEK293-T, MDA-MB-468, MCF-10A and Hs 578T cells were kept in DMEM-GlutaMAX I (Gibco). BT-20, MCF-7, MDA-MB-231, T-47D and ZR-75-1 cells were kept in RPMI 1640 medium (Gibco). McCoy´s 5A-GlutaMAX I (Gibco) medium was used for SK-BR-3 cells. All media were supplemented with 10% (v/v) heat-inactivated fetal calf serum (FCS). For the selection of stable HeLa Flp- In™ T-REx cells 15 µg/ml Blasticidin S (InvivoGen) and 100 µg/ml Hygromycin B (PAA) were added to the medium. Caspase-3 reconstituted MCF-7 cells (kind gift from Prof. Dr. Reiner Udo Jänicke, Düsseldorf, Germany) and lentiviral transduced HeLa S3, MCF-7, BT-20 and MDA-MB-468 cells were supplemented with 2 µg/ml Puromycin (Sigma). 100 ng/ml Doxycycline (Sigma) was used to trigger protein expression of stably introduced GOIs in HeLa Flp-In™ T-REx cells. Prior to UV treatment (20 mJ x cm-2 at 254 nm) cells were washed once with phosphate buffered saline (PBS). Cells were treated with 10 ng/ml human TNFα (Peprotech) in combination with 0.5 μg/ml Cycloheximide (Sigma). Pretreatment with 20 μM pan Caspase inhibitor Z- VAD-FMK (Calbiochem) or 50 μM Caspase-8 inhibitor Z-IETD-FMK (Santa Cruz) was performed 30 min before further stimulation. Cells were incubated with 10 µM Doxorubicin (Sigma) or 100 µM Etoposide (Sigma). Pre-treatment with 10 µM Erlotinib (Santa Cruz) was conducted 16 h prior to further stimulation. At a confluency of about

36 Experimental procedures

90-100%, adherent cells were detached from 10 cm dishes (Sarstedt). Cells were once washed with PBS and incubated with 2 ml Trypsin/EDTA (PAA) in order to passage cells. Cells were then diluted in fresh medium and transferred to a new dish.

PBS 140 mM NaCl (VWR) 2.7 mM KCl (AppliChem)

10 mM Na2HPO4 (Merck)

1.8 mM KH2PO4 (Merck)

2.4.2 Cryoconservation

At a confluency of around 90-100%, adherent cells were trypsinized from a 10 cm plate and pelleted by centrifugation (200 g, 4 min, RT). Cells were subsequently resuspended in 1ml fresh medium containing 10% heat-inactivated FCS and 10% DMSO (Merck) and transferred to a cryotube. A Styrofoam box was used to cool down the tube slowly to -80°C for 24 h. For long term storage cells were transferred to the -150°C freezer. For thawing, cells were warmed in a waterbath at 37°C and plated onto a 10 cm dish containing 9 ml of preheated medium.

2.4.3 Calcium phosphate transfection

For transient transfections, HEK293 cells were seeded in 6-well plates one day prior to transfections. 1 µg of plasmid DNA was diluted in 160 µl HBS and afterwards 8.5 µl of

2.5 µM CaCl2 (Merck) were added. The transfection mixture was vortexed and incubated for 25 min at RT. Finally, the transfection mixture was added drop-wise to the cells at a confluency of about 50% and the plate was rocked gently. Cells were transfected for 16-24 h.

37 Experimental procedures

HBS 137 mM NaCl (VWR) 5 mM KCl (AppliChem) 21 mM Hepes (Roth)

0.89 mM Na2HPO4 (Merck)

2.4.4 Lipid based transfection of DNA

According to the instructions supplied, Lipofectamine® 2000 (Invitrogen) was used for lipid based transient transfections in 6-well plates. Per well 1 µg of plasmid DNA and 4 µl of Lipofectamine® 2000 transfection reagent were mixed with 250 µl Opti-MEM (Gibco), respectively. After 5 min of incubation transfection reagent was added drop- wise to DNA solution and incubated for 20 min at RT. Finally, transfection mixture was added drop-wise to the cells cultured in 2 ml medium at a density of about 90% and mixed by gentle rocking. Medium was exchanged after 6 h post transfection and cells were analyzed after 24 h.

2.4.5 Lipid based transfection of siRNA

Transient PAR-4 knockdown was achieved using two pre-designed siRNAs constructs (Qiagen) directed against human PAR-4 (SI02628997 and SI03071782, referred as siPAR-4 #5 and siPAR-4 #8) at a final concentration of 20 nM and compared to non coding siRNA (SI03650325, referred to as siControl). Cells were seeded in a 6-well plate and transfected at a density of 30% using Oligofectamine™ transfection reagent according to the supplier’s manual. Before transfection, culture medium was exchanged with 800 ml Opti-MEM. For transfection, 183 µl of Opti-MEM was mixed with 2 µl of 10 µM siRNA oligos and 11 µl Opti-MEM was supplemented with 4 µl of Oligofectamine™ transfection reagent at RT. After 10 min of incubation both mixtures were mixed together, incubated for 20 min at RT and added drop-wise to the cells. After 4 h of incubation 500 µl Opti-MEM containing 30% heat-inactivated FCS was added to the cells to get a final concentration of 10% FCS. After 24 h transfection medium was exchanged and cells were incubated for another 24 h.

38 Experimental procedures

2.4.6 Generation of stable cell lines with the Flp-In™ T-REx system

The Flp-In™ T-REx system (Life Technologies) is a tool for generating stable, inducible cells expressing the GOI under a tetracycline-inducible promoter from a defined genomic locus. Utilizing Flp-recombinase-mediated integration of pcDNA5/FRT/TO vector, including the GOI, into the FRT site containing host cell line HeLa Flp-In™ T- REx, allows rapid and efficient generation of stable cells by homologous recombination. Therefore, 0.5 µg pcDNA5/FRT/TO harboring the GOI was mixed with 4.5 µg of pOG44 Flp-recombinase expression plasmid and transfected with Lipofectamine® 2000 using supplier’s manual. 24 h after transfection, cells were exposed to selection medium containing 100 µg/ml Hygromycin B (PAA) and 15 µg/ml Blasticidin S (InvivoGen). Single colonies were picked and cultured in 96-well plates to establish monoclonal cell lines. At a confluency of about 80%, cells were transferred to larger culture dishes and prepared for long-term storage by cryoconservation.

2.4.7 Lentivirus-mediated transduction of cell lines

For the generation of stable HeLa S3, MCF-7, BT-20 and MDA-MB-468 cell lines constitutively expressing short hairpin RNA (shRNA), self-inactivating GIPZ™ lentiviral constructs (Thermo Fisher Scientific) were used according to the guidelines of the RNAi consortium. In brief, HEK293-T packaging cells were seeded in 6 cm dishes and transfected the next day at a confluency of about 70%. Transient co-transfection of 1 µg of the lentiviral pGIPZ vector containing shRNA targeting Caspase-8 (V2LHS_112731 – 5´-TTCCTTCTCCCAGGATGAC-3´, referred to as shCaspase-8 #1 and V2LHS_112733 – 5´-TTCTTAGTGTGAAAGTAGG-3´, referred to as shCaspase-8 #3) or PAR-4 (V2LHS_152662 – 5´-TGTGTAATTGCATCTTCTC-3´, referred to as shPAR-4 #2) or non-silencing control (RHS4346, referred as shControl), 100 ng packaging plasmid psPAX2 and 1 µg envelope plasmid pCMV-VSV-G was carried out using Lipofectamine® 2000 as described previously. For viral harvests, culture medium is replaced by high serum growth medium containing 30% FCS on day three and viruses were incubated for another 24 h. Two days post-transfection lentiviral particles were harvested. To avoid contamination packaging cells were pelleted by centrifugation (200 g, 4 min, RT) and virus-containing medium was supplemented with 8 µg/ml

39 Experimental procedures

Ploybrene (Sigma). Viral transduced target cells were selected with 2 µg/ml Puromycin (Sigma).

2.4.8 Colony formation assay

For the colony formation assay 2x102 HeLa Flp-In™ T-REx cells were seeded in 6 cm dishes in duplicates. Addition of 100 ng/ml Doxycycline was used to induce individual protein expression. Culture medium was changed consecutively every three days. On day twelve cells were washed once with PBS and subsequently stained with 2 ml methanol (VWR) supplemented with 0.2% (w/v) methylene blue (Sigma-Aldrich) for

30 min. Afterwards cells were washed three times in ddH2O and dishes were dried. Finally, dishes were pictured for documentation and colonies were counted.

2.5 Methods in biochemistry

2.5.1 Cell lysis

For the preparation of cell lysates, medium from adherent cultured cells was aspirated. Before adding an appropriate volume of triton lysis buffer (TLB), freshly supplemented with protease and phosphatase inhibitors, cells were washed once with ice-cold PBS to remove remaining culture medium. Depending on cell density in each experiment the volume of TLB lysis buffer was adjusted individually. In general, 100 µl TLB lysis buffer was added to a well of a 6-well plate, while 1 ml TLB lysis buffer was added to a 10 cm dish during immunoprecipitation experiments. Cells were harvested by scraping and cells were transferred to a 1.5 ml microcentrifuge tube pre-cooled on ice. Cell lysates were incubated for 30 min on ice and afterwards cleared by centrifugation at 15000 x g for 15 min at 4°C. Supernatants were then transferred to a new microcentrifuge tube and analyzed by SDS-PAGE or stored at -20°C.

40 Experimental procedures

TLB lysis buffer 50 mM Tris-HCl pH 7.4 (AppliChem) 150 mM NaCl (VWR) 1 mM EDTA (Sigma) 1% (v/v) Triton X-100 (Sigma-Aldrich)

Protease Inhibitors 0.25 mM PMSF (Sigma) 2 µg/ml Aprotinin (Sigma) 1 µg/ml Leupeptin (Sigma)

Phosphatase Inhibitors 0.5 mM NaF (Merck)

1 mM Na3VO4 (Merck)

2.5.2 Measurement of protein concentrations

Concentration of solubilized proteins from cell lysates was determined by using the BIO- RAD Protein Assay (BioRad Laboratories), a method based on Bradford. Dye reagent was mixed with ddH2O in a ratio of 1:4 by volume. 1 ml of the dilution was mixed with 3 µl of cell lysate and incubated for 5 min at RT. Absorbance of protein samples was measured in a spectrophotometer at a wavelength of 595 nm. Protein concentration was evaluated by comparing the absorbance of each sample with standards of known BSA concentrations.

2.5.3 Immunoprecipitation

For immunoprecipitation of FLAG-tagged proteins cells were lysed in 1 ml TLB lysis buffer as described above. After clearing of the lysates an aliquot of 100 µl was set aside to analyze all proteins from whole cell lysate (WCL). The rest of the lysates was supplemented with anti-FLAG® M2 affinity gel (Sigma), a purified murine IgG1 monoclonal antibody covalently attached to agarose. After rotating the probes for 2.5 h at 4°C, FLAG-tagged proteins coupled to FLAG® M2 beads were pelleted by centrifugation at 6000 x g for 5 min at 4°C. Probes were washed three times in TLB 41 Experimental procedures lysis buffer to remove unbound proteins. Finally, TLB lysis buffer was removed and dry beads were analyzed by mass spectrometry approaches, prepared for Caspase assays or mixed with 20 µl of 2 x SDS loading buffer, boiled at 95°C for 5 min for SDS-PAGE analysis.

2 x SDS loading buffer 20 mM Tris-HCl pH 6.8 (AppliChem) 4% (w/v) SDS (Serva) 20% (v/v) Glycerol (Merck) 0.2% (w/v) Bromphenol blue (Sigma) 200 mM DTT (Merck)

2.5.4 Caspase assay

Protein lysates for Caspase assay were generated in TLB lysis buffer and immunoprecipitated with anti-FLAG® M2 affinity gel as described above. After FLAG® M2 beads were dried, 20 µl of Caspase reaction buffer was added and probes were incubated with 0.1 U/enzyme recombinant human Caspases 1-10 (PromoKine) at 37°C. One unit of the recombinant human Caspases is the enzyme activity that cleaves 1 nmol of the individual Caspase substrate at 37°C in 1 h. Reaction was stopped after 30 min by adding equal amounts of 2 x SDS loading buffer. Probes were boiled for 5 min at 95°C and subsequently analyzed by SDS-PAGE and immunoblotting.

Caspase reaction buffer 50 mM Hepes pH 7.2 (Roth) 50 mM NaCl (VWR) 0.1% Chaps (Sigma-Aldrich) 10 mM EDTA (Sigma) 5% (v/v) Glycerol (Merck) 10 mM DTT (Merck)

42 Experimental procedures

2.5.5 SDS-PAGE

Proteins were separated by sodium dodecyl sulphate polyacrylamide gel (SDS-PAGE) in BIO-RAD electrophoresis chambers, according to their electrophoretic mobility. Depending on the molecular size of the protein of interest, separating gels from 10% to 12.5% acrylamide were prepared and 5% stacking gel was used. Cell lysates were mixed with 2 x SDS loading buffer and boiled for 5 min at 95°C before loading onto the gel. Proteins were collected in the stacking gel applying 20 mA per gel for 20 min and afterwards separation was performed at 120 V for 1.5 h in Laemmli running buffer. To identify the approximate size of the protein of interest, a molecular weight marker (Thermo Fisher Scientific) was applied on each gel. In the case of immunoprecipitation experiments gels were incubated with 0.006% (w/v) Coomassie Brilliant Blue G-250 diluted in 10% (v/v) acetic acid (Merck) for visualization of purified proteins.

Laemmli running buffer 25 mM Tris (AppliChem) 192 mM Glycine (AppliChem) 0.1% (w/v) SDS (Serva)

2.5.6 Immunoblotting and immunodetection

Following separation by SDS-PAGE proteins were transferred from the acrylamide gel to a PVDF membrane (GE Healthcare) by semi-dry blotting. At first, the PVDF membrane was activated in methanol for about 1 min, afterwards equilibrated in transfer buffer for 5 min and put on a blotting paper (A. Hartenstein) soaked with transfer buffer. The acrylamide gel was positioned on the PVDF membrane and another wet transfer paper was placed on top. Transfer of proteins was achieved within 1.5 h by using 0.5 mA/cm2. After blotting, the PVDF membrane was washed once in PBST for 5 min and then incubated with blocking buffer for 1 h to avoid unspecific binding of antibodies to the membrane. Depending on the quality of each antibody, dilutions ranged from 1:1000 to 1:5000. Membranes were incubated with primary antibodies o/n at 4°C on a rocking platform. Membranes were then washed three times in PBST for 5 min, followed by 60 min incubation time with HRP coupled secondary antibodies, diluted 1:8000 in

43 Experimental procedures blocking buffer at RT. Membranes were washed again three times in PBST for 5 min to remove unbound secondary antibodies. Visualization was carried out by adding enhanced chemiluminescence (ECL) substrate solution supplemented freshly with 0,03% (v/v) hydrogen peroxide for 2 min. Signal detection was determined using the LAS-4000 reader (Fujifilm Global) and image editing was performed with Multi Gauge version 2 (Fujifilm) analysis software.

Transfer buffer 25 mM Tris (AppliChem) 192 mM Glycine (AppliChem) 20% (v/v) Methanol (VWR)

PBST 140 mM NaCl (VWR) 2.7 mM KCl (AppliChem)

10 mM Na2HPO4 (Merck)

1.8 mM KH2PO4 (Merck) 0.1% (v/v) Tween® 20 (Sigma-Aldrich)

Blocking buffer PBST 5% (w/v) non fat-dried milk powder (AppliChem)

ECL substrate solution 100 mM Tris-HCl pH 8.8 (AppliChem) 2.5 mM Luminol (Sigma-Aldrich) 0.2 mM poly-Coumaric acid (Sigma-Aldrich)

0.03% (v/v) H2O2 (Roth)

44 Experimental procedures

2.6 Confocal laser scanning microscopy

Microscopy analysis was planned and performed in close cooperation with Dr. Dirk Fahrenkamp from the Confocal Microscopy Facility of the RWTH Aachen University.

2.6.1 Fixation of cells for confocal microscopy

One day before transient transfection with expression vectors, cells were seeded on coverslips (Ø 18 mm) in 12-well plates. Culture medium was aspirated and cells were washed three times with PBS+/+. Fixation was carried out by incubating the cells with 500 µl 3.7% (w/v) paraformaldehyde (PFA) diluted in PBS for 20 min at RT. PFA was aspirated and cells were washed once with PBS+/+ to remove residual PFA. Aldehyde group mediated auto-fluorescence was quenched by treating the cells with 1 ml quenching buffer for 5 min at RT, where upon the cells were washed once with PBS +/+ and twice with ddH2O. Cells were then mounted on microscope slides using Immu- Mount (Thermo Fisher Scientific) and stored at 4°C protected from light. Image visualization was performed with the Zeiss LSM 710 confocal microscope using a LDC- apochromat 40 x /1.1 water objective. ZEN 2009 (Zeiss) software was used for image editing.

PBS+/+

PBS, 1mM MgCl2 (Sigma)

0.1 mM CaCl2 (Sigma)

Quenching buffer PBS+/+

50 mM NH4Cl (Merck)

2.6.2 Immunofluorescent staining of cells

Cells were seeded and fixed as described above, where upon the cells were permeabilized with 500 µl PBS supplemented with 0.2% (v/v) Triton X-100 for 30 min at RT. After washing the cells with PBST, cells were incubated with 5% (w/v) BSA (Serva)

45 Experimental procedures diluted in PBST for 1 h at RT to avoid unspecific binding of antibodies. For incubation of primary antibodies parafilm was placed on water-soaked blotting paper and 50 µl of the individual antibody dilution was applied on top. Coverslips were transferred upside- down onto the antibodies and incubated o/n at 4°C. Afterwards coverslips were transferred back to the 12-well plate and washed three times with PBST to remove unbound antibodies. Cells were incubated with secondary Alexa-Fluor-coupled antibodies diluted 1:1000 in 5% (w/v) BSA diluted in PBST for 1 h at RT in a light protected chamber to avoid bleaching of the fluorescent dye. Cells were washed three times with PBST and nuclear staining was carried out using 2 µg/ml Hoechst 33258

(Sigma) diluted in ddH2O for 5 min at RT. Finally, cells were washed once in PBST and twice in ddH2O and coverslips were fixed on microscope slides using Immu-Mount (Thermo Fisher Scientific). Probes were stored at 4°C protected from light and images were examined with the Zeiss LSM 710 confocal microscope with a LDC-apochromat 40 x /1.1 water objective. ZEN 2009 (Zeiss) software was used for image editing.

2.7 Flow cytometry

2.7.1 Fixation of cells for cell cycle analysis

Cells were seeded in 6-well plates and transfected as described previously. Cells were washed once with PBS and harvested by adding 500 µl Trypsin. By adding 1 ml preheated culture medium the reaction was stopped and the cells were transferred to a 15 ml reaction tube. Cells were washed once in PBS after centrifugation at 1200 x g for 5 min at 4°C. Cells were resuspended in 500 µl ice-cold PBS and 5 ml 80% (v/v) Ethanol (Merck) stored at -20°C was added drop-wise to the cells during gentle vortexing for fixation. After 30 min of incubation cells were pelleted at 4°C for 5 min at 1200 x g and resuspended in 500 µl PBS. To avoid unspecific staining of RNA DNAse free RNAse (Roche) was added to a final concentration of 20 µg/ml and incubated for 5 min at RT. Finally, propidium iodide (Sigma-Aldrich) was applied to a final concentration of 50 µg/ml and cells were incubated protected from light for 20 min at RT. Cell cycle analysis was performed using the BD FACSCanto II flow cytometer. Flowing Software version 2.4.1 was used for data editing.

46 Experimental procedures

2.8 Mass spectrometry approaches

Mass spectrometry analysis and probe preparation were planned and performed in close cooperation with Dr. Christian Preisinger from the Proteomics Facility of the RWTH Aachen University.

2.8.1 Digestion and analysis of precipitated protein complexes

On-beads digestion was carried out as recently described and partially modified [Turriziani et al. 2014]. Briefly, immunoprecipitated anti-FLAG® M2 beads were digested for 1 h at RT using 60 µl digestion buffer 1, followed by two washing steps with 25 µl digestion buffer 2. Supernatants were pooled and left to digest o/n at RT. Samples were subsequently modified using 20 µl of 5 mg/ml iodoacetamide (Sigma-Aldrich) and then acidified with formic acid (Sigma-Aldrich) with a final concentration of 1% (v/v) in order to stop the digestion. Afterwards, peptides were desalted using homemade C18-tips and lyophilized peptides were resuspended in 20 µl 10% (v/v) formic acid (2 runs with 10 µl each). Tryptic peptides were then subjected to analysis by reversed phase nano LC-MS/MS using a nano Ultimate 3000 liquid chromatography system (Thermo Scientific) and an Orbitrap Elite mass spectrometer (Thermo Scientific). The raw data was analyzed using MaxQuant version 1.5.2.8 [Cox and Mann 2008]. The spectra were searched against the human SwissProt database version 06/2015 using the built-in Andromeda search engine [Cox et al. 2011].

Digestion buffer 1 2 M Urea (Serva) 50 mM Tris-HCl pH 7.5 (AppliChem) 5 µg/ml Trypsin

Digestion buffer 2 2 M Urea (Serva) 50 mM Tris-HCl pH 7.5 (AppliChem) 1 mM DTT (Merck)

47 Results

3 Results

3.1 UV-induced apoptosis results in Caspase-dependent PAR-4 cleavage at

EEPD131↓G

The tumor suppressor PAR-4 is a pro-apoptotic protein, which is required for the induction of apoptosis in multiple contexts [Hebbar, Wang, and Rangnekar 2012] (Chapter 1.3.2). PAR-4 expression is silenced in a subset of human cancers and its re- expression induces apoptosis in cancer cells [Shrestha-Bhattarai and Rangnekar 2010] (Chapter 1.3.3). Although PAR-4 functions as a critical regulator of tumor cell survival the mechanisms by which PAR-4 regulates the induction of apoptosis is poorly understood. Previous findings indicated that PAR-4 selectively induces apoptosis in cancer cell lines including HeLa cells [Hebbar, Wang, and Rangnekar 2012]. To further evaluate these findings, HeLa cells were treated with UV light and cell lysates were analyzed after the indicated time points using PARP-1 cleavage as a marker for Caspase activity (Figure 5A). Efficient PARP-1 cleavage was detectable after 3 hours of UV treatment and at the same time a PAR-4 fragment of ~17 kDa became visible using a PAR-4 amino-terminal antibody, suggesting that this protein may be cleaved during apoptosis (Figure 5A). To analyze whether PAR-4 is hydrolyzed by Caspases during UV-induced cell death, HeLa cells were pre-incubated with the potent and pan-specific Caspase inhibitor Z-VAD-FMK. Pre-treatment with Z-VAD-FMK prevented UV-mediated PAR-4 and PARP-1 cleavage in HeLa cells, indicating that this event is Caspase-dependent (Figure 5B). To investigate if UV-induced PAR-4 processing was species specific human (h) and rat (r) PAR-4 constructs were expressed in HeLa cells followed by UV exposure. By using two different antibodies against the amino-terminal (N) and the carboxy-terminal (C) part of PAR-4, generation of a ~17 kDa and a ~28 kDa fragment for human PAR-4 and a ~15 kDa and a ~30 kDa fragment for rat Par-4 was observed upon UV-treatment, indicating the existence of a single cleavage site in both species (Figure 5C). To identify the cleavage site, the protein sequence of PAR-4 was scanned for potential Caspase cleavage sites on the CASVM server, a web server for predicting Caspase cleavage sites on protein sequences based on support vector machine (SVM) algorithms [Wee, Tan, and Ranganathan 2007]. The CASVM server revealed a potential cleavage site at EEPD131↓G in the human PAR-4 protein and a potential cleavage site

48 Results at EEPD122↓S for rat Par-4. To validate this finding D at position 131 was mutated to G in the human PAR-4 sequence, to generate a mutant that cannot be cleaved by Caspases [Stennicke et al. 2000; Chaudhry et al. 2012]. Compared to the wild-type (WT) protein expression of PAR-4 D131G in HeLa cells was resistant to UV-induced processing as no cleavage fragments were generated, confirming the existence of a single Caspase cleavage site at position EEPD131↓G in human PAR-4 (Figure 5D). The cleavage site separates the N-terminal region from the SAC and leucine zipper domains, which are both conserved in rat and murine Par-4, albeit slightly shifted towards the N-terminus, explaining at least in part the altered mobility of rat Par-4 cleavage fragments (Figure 5C and 5E).

A) HeLa B) HeLa UV (h) UV (6 h) UV (10 h) kDa 1 3 6 10 kDa + + + z-VAD-FMK

PAR-4 PAR-4 40- 40-

35- 35-

25- 25-

 cl. PAR-4 (N)  cl. PAR-4 (N) 15- 15- 130- PARP-1 130- PARP-1 100-  cl. PARP-1 100-  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

C) HeLa D) HeLa PAR-4 (h) Par-4 (r) T7-PAR-4 WT T7-PAR-4 D131G + + UV (3 h) + + UV (3 h) kDa + + UV (6 h) kDa + + UV (6 h) PAR-4 T7-PAR-4 40- 40- 35- 35- 25- 25-

 cl. PAR-4 (N; h)  cl. PAR-4 (N) 15-  cl. Par-4 (N; r) 15-

PAR-4 PAR-4 40- 40- 35- 35-  cl. Par-4 (C; r)  cl. PAR-4 (C) 25-  cl. PAR-4 (C; h) 25-

40- 40- GAPDH GAPDH 35- 35-

E) Caspase cleavage site

1 131 146 163 203 300 340 N L SAC LZ 49 S

P4 P3 P2 P1 P1´ P2´ P3´

human PAR-4 D E E E P D131 G132V P rat Par-4 D E E E P D122 S123 A P mouse Par-4 D E E E P D123 S124 A R A) HeLa B) HeLa UV (h) UV (6 h) UV (10 h) kDa 1 3 6 10 kDa + + + z-VAD-FMK

PAR-4 PAR-4 40- 40-

35- 35-

25- 25-

 cl. PAR-4 (N)  cl. PAR-4 (N) 15- 15- 130- PARP-1 130- PARP-1 100-  cl. PARP-1 100-  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

C) HeLa D) HeLa PAR-4 (h) Par-4 (r) T7-PAR-4 WT T7-PAR-4 D131G + + UV (3 h) + + UV (3 h) kDa + + UV (6 h) kDa + + UV (6 h) PAR-4 T7-PAR-4 40- 40- 35- 35- 25- 25-

 cl. PAR-4 (N; h)  cl. PAR-4 (N) 15-  cl. PAR-4 (N; r) 15-

PAR-4 PAR-4 40- 40- 35- 35-  cl. PAR-4 (C; r)  cl. PAR-4 (C) 25-  cl. PAR-4 (C; h) 25-

40- 40- GAPDH GAPDH 35- 35- Results

E) Caspase cleavage site

1 131 146 163 203 300 340 N L SAC LZ S

P4 P3 P2 P1 P1´ P2´ P3´

human PAR-4 D E E E P D131 G132V P rat Par-4 D E E E P D122 S123 A P mouse Par-4 D E E E P D123 S124 A R

Figure 5: PAR-4 is cleaved during UV-induced apoptosis at position 131 by Caspases. (A) HeLa cells were treated with UV (20 mJ x cm-2 at 254 nm) and incubated for the indicated time points. Cell lysates were analyzed by immunoblotting using antibodies against PAR-4, PARP-1 and GAPDH. A processed form recognized by an antibody that preferentially recognizes the N-terminal part of PAR-4 is indicated (cl. PAR-4). (B) HeLa cells were pre-treated with the pan-Caspase inhibitor Z-VAD-FMK (20 µM) for 30 min, then treated as in (A) and incubated for the indicated time points. Lysates were analyzed by immunoblotting as above (A). (C) HeLa cells were transiently transfected with plasmids expressing human (h) PAR-4 or rat (r) Par-4 (1 µg each). After incubation for 24 h cells were treated with UV light as in (A). After further incubation for the indicated time points, cell lysates were analyzed by immunoblotting using two PAR-4 specific antibodies that preferentially recognize the N-terminal (N) part of the protein (upper panel) or the C-terminal (C) part of PAR-4 (middle panel). (D) HeLa cells were transiently transfected with plasmids encoding T7-tagged PAR-4 or T7-tagged PAR-4 D131G (1 µg each). 24 h after transfection and treatment with UV for the indicated times, whole cell lysates were prepared and analyzed by immunoblotting with T7-specific antibodies recognizing the N-terminus of PAR-4 (upper panel) or PAR-4 specific antibodies that recognize the C-terminal part of the protein (middle panel). (E) Schematic overview of human PAR-4 protein: the Caspase cleavage site at position 131 adjacent to the “selective for apoptosis in cancer cells” domain (SAC, amino acids 146 - 203); nuclear localization sequence (NLS, amino acids 147 - 163); leucine zipper (LZ, amino acids 300 - 340). The Caspase cleavage site of PAR-4 is conserved between human and rodent orthologs with the exception of S instead of G at position P1’ in mouse and rat Par-4.

3.2 Inducible expression of PAR-4 but not PAR-4 D131G interferes with cell

proliferation

The processing of PAR-4 upon UV-mediated cell death by Caspases at position EEPD131↓G (Figure 5E) raised the question if this posttranslational cleavage event exhibits any biological effects. Therefore, multiple HeLa Flp-In™ T-REx cell clones were established, which either express PAR-4 WT or PAR-4 D131G from the identical locus under the control of a Doxycycline-inducible promoter (Figure 6A) (Chapter 2.4.6).

50 Results

Subsequent analysis of the growth characteristics of two stable cell clones in colony formation assays revealed a significant reduction of colony number and also colony size upon induced expression of wild-type PAR-4 but not PAR-4 D131G (Figure 6B). Although the inducible expression of PAR-4 WT and PAR-4 D131G was comparable, expression of PAR-4 wild-type led to the generation of the N-terminal PAR-4 cleavage fragment (Figure 6A). This suggested that moderate overexpression of PAR-4 was sufficient to induce Caspase activation as observed previously [El-Guendy et al. 2003; Zhao et al. 2007]. Therefore, the capacity of wild-type PAR-4 and PAR-4 D131G to induce apoptosis was analyzed. Doxycycline-inducible protein expression of PAR-4 WT but not the Caspase cleavage resistant mutant in HeLa Flp-In™ T-REx cells led to increased PARP-1 cleavage, indicating that Caspase processing of PAR-4 is necessary to activate its pro-apoptotic properties (Figure 6C).

A) HeLa T-REx B) HeLa T-REx Control PAR-4 WT PAR-4 D131G Control PAR-4 WT PAR-4 D131G #2 #4 #4 #6 #1 #2 #2 #4 #4 #6 #1 #2 kDa + + + + + + Doxycycline PAR-4 40- 35-

25-

 cl. PAR-4 (N) 15- Doxycycline 40- p < 0.005 GAPDH 35- 150 p < 0.005

C) HeLa T-REx 120 WT #6 D131G #2

kDa + + Doxycycline PAR-4 90 40-

35- number Colony 60 25-

 cl. PAR-4 (N) 30 15- 130- PARP-1 Doxycycline 100-  cl. PARP-1 40- GAPDH Control Control WT WT D131G D131G 35- #2 #4 #4 #6 #1 #2

FigureFig. 6: PAR 6:- 4PAR-4 cleavage cleavageinterferes with interferes cell proliferation. with cell proliferation. (A) Individual established clones of HeLa Flp-In PAR-4 wild-type (WT), PAR-4 D131G and control (empty vector) T-REx cells were treated with (A)Doxycycline Individual (100 ng/ml) generated for 72 h. Protein clones expression of HeLa was Flp-In™analyzed by PAR-4immunoblotting WT, using PAR-4 the indicated D131G antibodies. and control N-terminal (empty PAR-4 fragments vector) are T- indicated. REx(B) For cells the colony were formation treated assay, with200 cells Doxycycline were seeded and (100 grown ng/ml)for 12 days for ± Doxycycline72 h. Protein (100 ng/ml) expression prior to staining waswith methylene analyzed blue. by Medium was changed everey three days. Two different clones each for PAR-4 WT, PAR-4 D131G and control cells were analyzed (upper panel). immunoblottingAverage values from threeusing independent the indicated experiments antibodies. are displayed. (B) Doxycycline For the treated colony cells areformation shown in light assay, grey. Error 200 bar cellss indicate were ± SD. seeded P-value was obtained by two-tailed Student’s t-Test, comparing Doxycycline treated with non-treated cells (lower panel). and(C) Hela grown Flp-In PARfor- 412 wild days-type and ± PAR Doxycycline-4 D131G cells (100were grown ng/ml) for 2 daysprior ± Doxycycline.to staining Protein with lysates methylene were analyzed blue. with theMedium indicated was antibodies. changed every three days. Two different clones each for PAR-4 WT, PAR-4 D131G and control cells

51 Results were analyzed (upper panel). Average values from three independent experiments are displayed. Doxycycline treated cells are shown in light grey. Error bars indicate ± SD. P-value was obtained by two- tailed Student’s t-Test, comparing Doxycycline treated with non-treated cells (lower panel). (C) Hela Flp- In™ PAR-4 WT and PAR-4 D131G cells were grown for 2 days ± Doxycycline. Protein lysates were analyzed with the indicated antibodies.

3.3 PAR-4 is a substrate of Caspase-8 in vitro and in vivo

In order to identify Caspases that are capable to process PAR-4, immunoprecipitated Flag-tagged PAR-4 was incubated with human recombinant Caspases 1 to 10. Caspase-1, -7 and -8 were able to cleave PAR-4 to various degrees in vitro, with Caspase-8 being the most efficient Caspase to hydrolyze full length PAR-4 (Figure 7A). The tumor necrosis factor (TNFα) receptor family is an established mediator of the extrinsic pathway and stimulates apoptosis through DISC formation, which includes recruitment and activation of Caspase-8 [Weinlich, Dillon, and Green 2011] (Chapter 1.2.3). To study the role of Caspase-8 in PAR-4 processing HeLa S3 cells were stimulated with TNFα for various time points. In addition, the protein synthesis inhibitor Cycloheximide (CHX) was added to the cells to switch the inflammatory response of mammalian cells to TNFα to apoptotic signaling by inhibiting NF-κB- mediated gene expression of cFLIP, a negative regulator of Caspase-8 [Wang, Du, and Wang 2008]. TNFα/CHX-induced signaling led to simultaneous PAR-4 and PARP-1 cleavage (Figure 7B). To investigate if Caspase-8 is required for TNFα/CHX induced PAR-4 cleavage, HeLa S3 cell lines were generated using lentiviral delivery of shRNA constructs either expressing two Caspase-8-specific shRNAs (shCaspase-8 #1 and #3) or a non-silencing shRNA, which serves as a control (shControl) (Chapter 2.4.5). The expression of apical Caspase-8 was reduced in HeLa S3 cells transduced with shCaspase-8 #1 and #3, with shCaspase-8 #3 being the most efficient shRNA to silence Caspase-8 expression (Figure 7C). Stimulation with TNFα/CHX induced PAR-4 and PARP-1 cleavage in the presence of Caspase-8, while cleavage in cells transduced with shCaspase-8 #1 was decreased. PAR-4 cleavage was totally prevented in cells expressing shCaspase-8 #3, indicating that PAR-4 is downstream of Caspase-8 (Figure 7D). Together these findings suggest that PAR-4 is a direct target of Caspase-8. In contrast to these findings it was shown recently that PAR-4 is a substrate of Caspase-3 and PAR-4 cleavage does not occur after Cisplatin treatment of MCF-7

52 Results breast cancer cells, which lack the expression of Caspase-3 [Chaudhry et al. 2012; Jänicke 2009]. As the in vitro experiment showed only very weak activity of Caspase-3 towards PAR-4 (Figure 7A), the role of Caspase-3 was further addressed in MCF-7 cells. Therefore, TNFα-induced PAR-4 cleavage in Caspase-3-deficient MCF-7 cells and in Caspase-3 reconstituted cells was measured. Stimulation of MCF-7 cells with TNFα led to PAR-4 cleavage regardless whether Caspase-3 was absent or present, indicating that TNFα-induced PAR-4 processing is Caspase-3 independent. The disappearance of full length Caspase-3 after TNFα stimulation serves as a control for its activation (Figure 7E). Pre-treatment of Caspase-3-deficient MCF-7 cells with the Caspase-8-specific inhibitor Z-IETD-FMK demonstrated that TNFα-induced PAR-4 cleavage was Caspase-8 dependent (Fig 3F). To confirm the data in another cell line HeLa S3 cells were pre-incubated either with Z-IETD-FMK or Z-DEVD-FMK, a specific inhibitor for Caspase-3 and stimulated with TNFα/CHX treatment. Upon Caspase-3 inhibition TNFα-mediated PAR-4 and PARP-1 processing was only slightly decreased, whereas Caspase-8 inhibition completely prevented PAR-4 and PARP-1 processing, indicating that cleavage of PAR-4 is upstream of Caspase-3 (Figure 7G). In addition, co- expression experiments demonstrated that expression of Caspase-8 in combination with PAR-4 induced PAR-4 and PARP-1 cleavage, indicating induction of apoptotic cell death. In contrast, co-expression of Caspase-3 and PAR-4 did neither result in PAR-4 cleavage nor in the induction of cell death. Expression of Caspase-8 and Caspase-3 alone has been shown to initiate apoptosis [Miura et al. 1993] and therefore amounts of both Caspases were carefully titrated to generate conditions under which overexpression of each did not result in the induction of apoptosis (Figure 7H). Taken together, these results strongly indicate that PAR-4 is a substrate of Caspase-8 in vitro and gets cleaved following initiation of the extrinsic pathway within cells during apoptosis in vivo.

53 Results

A) Caspase C) HeLa S3 kDa 1 2 3 4 5 6 7 8 9 10 + shControl + shCaspase-8 #1 PAR-4-FLAG kDa + s h Caspase -8 #3 40- Caspase-8 35- 55-  cl. PAR-4 (C) 40- 25- GAPDH 35-

B) HeLa S3 D) HeLa S3 TNFα/CHX (h) CHX (h) TNFα/CHX + + shControl kDa 3 6 9 9 + + shCaspase-8 #1 kDa + + shCaspase-8 #3 PAR-4 40- PAR-4 40- 35- 35- 25- 25-

15-  cl. PAR-4 (N) 130-  cl. PAR-4 (N) PARP-1 15- 110-  cl. PARP-1 130- PARP-1 40- 100- GAPDH  cl. PARP-1 35- 40- GAPDH 35-

E) MCF-7 F) MCF-7 Caspase-3 TNFα/CHX

TNFα/CHX (h) TNFα/CHX (h) kDa + z-IETD-FMK

kDa 2 4 6 8 2 4 6 8 PAR-4 40- PAR-4 40- 35-

35- 25-

25-  cl. PAR-4 (N) 15-  cl. PAR-4 (N) 130- 15 - PARP-1 100-  cl. PARP-1 Caspase-3 35- 40- GAPDH 40- GAPDH 35- 35-

54 E) MCF-7 F) MCF-7 Caspase-3 TNFα/CHX

TNFα/CHX (h) TNFα/CHX (h) kDa + z-IETD-FMK

kDa 2 4 6 8 2 4 6 8 PAR-4 40- PAR-4 40- 35-

35- 25-

25-  cl. PAR-4 (N) 15-  cl. PAR-4 (N) 130- 15 - PARP-1 100-  cl. PARP-1 Caspase-3 35- 40- GAPDH 40- GAPDH 35- 35- Results

G) HeLa S3 H) HEK293 TNFα/CHX (4 h) PAR-4 WT + + Caspase-8-HA + + Z-IETD-FMK kDa + + Caspase - 3 kDa + + Z-DEVD-FMK PAR-4 PAR-4 40- 40- 35- 35-  cl. PAR-4 (C) 25- 25- 70- Caspase-8-HA 55- 15-  cl. PAR-4 (N) 35- Caspase-3 130- PARP-1 130- 100- PARP-1  cl. PARP-1 110- 40-  cl. PARP-1 GAPDH 40- 35- GAPDH 35-

Figure 7: PAR-4 is a substrate of Caspase-8. (A) In vitro Caspase assay using purified FLAG-tagged PAR-4 wild-type and recombinant human Caspases 1-10. The assay was stopped after 30 min by adding sample buffer, and the proteins were subjected to immunoblotting using a PAR-4 antibody recognizing the C-terminal part of the protein. (B) HeLa S3 cells were stimulated with TNFα (10 ng/ml) in combination with CHX (0.5 µg/ml) for the indicated time points and with CHX alone for 9 h. Cell lysates were analyzed by immunoblotting with the indicated antibodies and PAR-4 cleavage is indicated. (C) Cell lysates from HeLa S3 control (shControl) or Caspase-8-deficient (shCaspase-8 #1 and #3) were analyzed by immunoblotting with Caspase-8 and GAPDH antibodies. (D) HeLa S3 shControl or shCaspase-8 #1 and #3 cells were incubated with TNFα (10 ng/ml) in combination with CHX (0.5 µg/ml) for 3 h. Protein lysates were analyzed by protein immunoblotting with antibodies against PAR-4, PARP-1 and GAPDH. (E) Caspase-3-deficient MCF-7 cells and MCF-7 cells stably expressing Caspase-3 were stimulated with TNFα (10 ng/ml) and CHX (0.5 µg/ml) for the indicated time points and protein lysates were analyzed by immunoblotting with antibodies against PAR-4, Caspase-3 and GAPDH. PAR-4 cleavage products are indicated. (F) MCF-7 cells were pre-incubated with 50 µM Caspase-8 inhibitor (Z-IETD-FMK) for 30 min and subsequently stimulated with TNFα (10 ng/ml) and CHX (0.5 µg/ml) for 16 h. Cell lysates were analyzed with the indicated antibodies via immunoblotting and cleavage of PAR-4 is indicated. (G) HeLa S3 cells were pre- incubated with 50 µM Caspase-8 inhibitor (Z-IETD-FMK) or 20 µM Caspase-3 inhibitor (Z-DEVD-FMK) for 30 min and subsequently treated with TNFα (10 ng/ml) and CHX (0.5 µg/ml) for 4 h. Cell lysates were analyzed with antibodies against PAR-4, PARP-1 and GAPDH. (H) HEK293 cells were transiently transfected with plasmids driving the expression of PAR-4 wild-type (1 µg), Caspase-8 or Caspase-3 (50 ng each) for 16 h. Protein lysates were analyzed by protein immunoblotting with antibodies recognizing HA, Caspase-3, PARR-1 and GAPDH. PAR-4 cleavage was analyzed with antibodies generated against the C-terminal part of the protein.

55 Results

3.4 TNFα-induced apoptosis requires Caspase-8-mediated processing of PAR-4

The Caspase-8 cleavage site of PAR-4 at position 131 is adjacent to the central SAC domain encompassing a NLS sequence (Figure 5E) and induction of apoptosis in cancer cell lines by expression of the SAC domain of PAR-4 has been shown to require nuclear localization [El-Guendy et al. 2003; Zhao et al. 2007]. To investigate the cellular distribution of the C-terminal PAR-4 fragment comprising the functional important SAC and leucine zipper domains, PAR-4 mutants with a C-terminal eCFP-tag were generated (Figure 8A, left panel). The constructs were transiently expressed in HEK293 cells and analyzed by confocal microscopy (Figure 4A right panel). Wild-type PAR-4 (PAR-4 WT-eCFP) and the non cleavable point mutant PAR-4 D131G (PAR-4 D131G- eCFP) localized to the cytosol, while the PAR-4 mutant lacking amino acids 1-131 (PAR-4 132-340-eCFP) localized to the nuclear compartment (Fig 8A, right panel). Stimulation with TNFα in combination with CHX for 8 h or exposure to UV light resulted in the nuclear accumulation of wild-type PAR-4, but was prevented in cells expressing PAR-4 D131G (Figure 8B). To expand on these findings, Caspase-8-deficient MCF-7 cells were generated using lentiviral-mediated shRNA constructs expressing two Caspase-8-specific shRNAs (shCaspase-8 #1 and #3). Knock down efficiency was tested by immunoblotting and compared to cells expressing a non-silencing shRNA (shControl), which serves as a control (Figure 8C). Localization of endogenous PAR-4 was then analyzed in the presence or absence of TNFα/CHX using a PAR-4 antibody recognizing the C-terminal part of the protein. Results from confocal microscopy demonstrated an accumulation of PAR-4 in the nucleus upon TNFα/CHX treatment, while nuclear translocation was not monitored in Caspase-8-deficient MCF-7 cells (Figure 8D). These observations suggest that Caspase-8-mediated PAR-4 cleavage result in the nuclear translocation of the C-terminal PAR-4 fragment, thereby inducing apoptosis. To address the question whether PAR-4 cleavage mediated by Caspase-8 is required to trigger TNFα-induced apoptosis, Caspase-8-deficient MCF-7 cells were incubated with TNFα in combination with CHX or CHX alone for 10 h and induction of apoptosis was measured by PARP-1 cleavage. While MCF-7 shControl cells underwent apoptosis and showed PAR-4 processing following TNFα/CHX stimulation, Caspase-8-deficient cells failed to do so (Figure 8E). CHX treatment alone was not sufficient to induce apoptosis in Caspase-8-deficient MCF-7 cells or control cells. To investigate if PAR-4 expression

56 Results was required for the induction of apoptosis in response to TNFα/CHX, PAR-4-deficient MCF-7 cells (shPAR-4 #2) were created via a lentiviral approach (Chapter 2.4.5) and incubated with TNFα combined with CHX for 10 h. In comparison to MCF-7 control cells (shControl), in which PAR-4 processing was obtained upon TNFα/CHX-induced apoptosis, PARP-1 cleavage was prevented in PAR-4-depleted MCF-7 cells (Figure 8F). Taken together, these data demonstrate that PAR-4 cleavage is a direct consequence of Caspase-8 activation and is required for nuclear translocation and induction of apoptosis mediated by the C-terminal PAR-4 fragment.

A) HEK293 B) HEK293 PAR-4 WT-eCFP WT-eCFP D131G-eCFP N S L A LZ eCFP S C

PAR-4 D131G-eCFP N S L A LZ eCFP S C TNFα/CHX

PAR-4 132-340-eCFP N S L A LZ eCFP S C UV

C) MCF-7 D) MCF-7 + shControl shControl shCaspase-8 #3 + shCaspase-8 #1 kDa + shCaspase-8 #3 + + TNFα/CHX

55- Caspase-8 40- GAPDH PAR-4 35- Hoechst

E) MCF-7 F) MCF-7

CHX TNFα/CHX CHX TNFα/CHX

+ + + shControl + + + shControl kDa + + + shCaspase-8 #3 kDa + + + shPAR-4 #2

PAR-4 PAR-4 40- 40-

35- 35-

25- 25-

15-  cl. PAR-4 (N) 15-  cl. PAR-4 (N) 130- 130- PARP-1 100- PARP-1 100-  cl. PARP-1  cl. PARP-1 57 40- 40- GAPDH GAPDH 35- 35- A) HEK293 B) HEK293 PAR-4 WT-eCFP WT-eCFP D131G-eCFP N S L A LZ eCFP S C

PAR-4 D131G-eCFP N S L A LZ eCFP S C TNFα/CHX

PAR-4 132-340-eCFP N S L A LZ eCFP S C UV

C) MCF-7 D) MCF-7 + shControl shControl shCaspase-8 #3 + shCaspase-8 #1 kDa + shCaspase-8 #3 + + TNFα/CHX

55- Caspase-8 40- GAPDH PAR-4 35- HoechstResults

E) MCF-7 F) MCF-7

CHX TNFα/CHX CHX TNFα/CHX

+ + + shControl + + + shControl kDa + + + shCaspase-8 #3 kDa + + + shPAR-4 #2

PAR-4 PAR-4 40- 40-

35- 35-

25- 25-

15-  cl. PAR-4 (N) 15-  cl. PAR-4 (N) 130- 130- PARP-1 100- PARP-1 100-  cl. PARP-1  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

Figure 8: PAR-4 is required for TNFα-induced apoptosis. (A) Schematic representation of wild-type PAR-4 (PAR-4 WT-eCFP), a non cleavable PAR-4 point mutant (PAR-4 D131G-eCFP, mutation marked in red) and a PAR-4 deletion mutant comprising amino acids 132-340 (PAR-4 132-340-eCFP) C-terminal fused to CFP (left panel). HEK293 cells were transiently transfected with the plasmids (1 µg) and analyzed by immunofluorescence via confocal microscopy (right panel). (B) HEK293 cells expressing PAR-4 WT-eCFP or PAR-4 D131G-eCFP (1 µg) were treated with TNFα (10 ng/ml) combined with CHX (0.5 µg/ml) for 8 h or exposed to UV light (20 mJ x cm -2 at 254 nm) and incubated for 8 h. Cells were analyzed by immunofluorescence using confocal microscopy. (C) Caspase-8-deficient MCF-7 cells (shCaspase-8 #1 and #3) were generated using lentivirus-mediated shRNA constructs. Cell lysates were analyzed by immunoblotting using antibodies against Caspase-8 and GAPDH and compared to cell lysates from cells expressing a non silencing shRNA construct (shControl). (D) Caspase-8-deficient MCF-7 (shCaspase-8 #3) and control cells (shControl) were treated with TNFα (10 ng/ml) combined with CHX (0.5 µg/ml) for 10 h and immunostained with a C-terminal PAR- 4 antibody (red). Hoechst was used for nuclear staining (blue) and probes were analyzed by confocal microscopy. (E) MCF-7 shCaspase-8 #3 cells and shControl cells were incubated with CHX (0.5 µg/ml) or in combination with TNFα (10 ng/ml) for 10 h. Lysates were analyzed by immunoblotting using the indicated antibodies. (F) MCF-7 cells that express a lentiviral-mediated shRNA targeting PAR-4 (shPAR-4 #2) were treated and analyzed as in (E) and compared with shControl cells.

3.5 PAR-4 is down-regulated in a subset of breast cancer cell lines

PAR-4 exhibits tumor suppressor activities and is silenced in a well-defined subset of human cancers, including, neuroblastoma, endometrial carcinoma, prostate and lung carcinoma [Kögel et al. 2001; Moreno-Bueno et al. 2007; Fernandez-Marcos et al. 2009; Joshi et al. 2008] (Chapter 1.3.3). Several studies also demonstrated that PAR-4

58 Results is down-regulated in breast cancer [Zapata-Benavides et al. 2009; Méndez-López et al. 2010; Nagai et al. 2010] (Chapter 1.3.3). Recent studies from Alvarez and colleagues further demonstrated that oncogene-induced primary tumors in mice exhibit PAR-4 expression, while secondary recurrent tumors significantly show reduced PAR-4 mRNA and protein levels. Compared to patients that express higher levels of PAR-4, low PAR- 4 levels in women with breast cancers show an increased risk of tumor relapse. Furthermore, they demonstrated that PAR-4 down-regulation serves as a mechanism for tumor recurrence and is associated with poor response to targeted therapy and chemotherapy [Alvarez et al. 2013] (Figure 9A). As down-regulation of PAR-4 serves as a mechanism for tumor cell survival, PAR-4 and Caspase-8 expression was analyzed in a panel of breast cancer cell lines by immunoblotting. Compared to the immortalized, non-transformed mammary epithelial cell line MCF-10A, PAR-4 expression was found to be low in ZR-75-1 cells and in three out of four TNBC cell lines, MDA-MB-468, Hs 578T and BT-20. However, in none of the used cell lines PAR-4 expression was fully abolished, as shown for recurrent tumors. Low protein levels of Caspase-8 were detected in MCF-7, T-47D, ZR-75-1, MDA-MB-231 and Hs 578T cells (Figure 9B). 231

A) B) 468 - - 3 - 1 - MB 10A 7 MB - - - - BR 75 20 - - Chemotherapy - 47D

Primary Recurrent - SK MCF MCF ZR 578T Hs BT T MDA

kDa MDA tumor tumor PAR-4 40-

55- Caspase-8

40- Intensity GAPDH 35- Time - + + + - - - - - ER PAR-4 level - + + + - - - - - PR Chemoresistance - - - - + - - - - HER2/neu + - + - + + - - + EGFR

Figure 9: PAR-4 is down-regulated in TNBC cells. (A) Simplified graph of PAR-4 mRNA and protein levels (blue) during chemoresistance (red) associated with tumor recurrence following therapy of the primary tumor. (B) Lysates from a panel of breast cancer cell lines including MCF-7, T-47D, ZR-75-1, SK-BR-3 and triple negative breast cancer (TNBC) cell lines MDA-MB-468, MDA-MB-231, Hs 578T and BT-20 were analyzed by immunoblotting using antibodies against PAR-4, Caspase-8 and GAPDH and compared with the non-tumorigenic, immortalized breast epithelial cell line MCF-10A. Expression status of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor (HER2/neu) and epidermal growth factor receptor (EGFR) is depicted for each cell line.

59 Results

3.6 PAR-4 is cleaved by Caspase-8 following DNA-damage

PAR-4 down-regulation marks a central role in the development of recurrent mammary tumors and is associated with chemoresistance following therapy [Alvarez et al. 2013]. The previous results demonstrate that PAR-4 is a substrate of Caspase-8 and its processing is required for TNFα/CHX-induced apoptosis in breast cancer cells (Figure 8E and 8F). Genotoxic drug treatment of breast cancer cells in turn has been demonstrated to activate Caspase-8 [Wesselborg et al. 1999; Chandra et al. 2004; Sohn, Schulze-Osthoff, and Jänicke 2005]. To analyze whether DNA-damage-induced apoptosis causes Caspase-8-dependent PAR-4 cleavage, HeLa and MCF-7 cells were incubated with TNFα/CHX and the genotoxic drugs Doxorubicin and Etoposide. The anthracycline Doxorubicin and the antineoplastic agent Etoposide both prevent DNA religation and induce DNA-damage. Doxorubicin causes DNA-damage by intercalation with the DNA’s double helix, thereby affecting a broad range of DNA processes [Pommier et al. 2010]. Although Doxorubicin is highly toxic and exhibits diverse side effects, treatment of many types of cancer, including metastatic breast cancer, remains highly effective [Moreno-Aspitia and Perez 2009]. Etoposide is known to inhibit TOP II, leading to single and double strand breaks, which ultimately causes apoptosis [Pommier et al. 2010]. Regimens including Etoposide are typically used for the treatment of small- cell lung carcinomas and metastatic breast cancer [Vansteenkiste et al. 2013; P. Yuan et al. 2012]. Figure 10A demonstrates that genotoxic drug treatment and TNFα-induced apoptosis all resulted in the generation of an N-terminal PAR-4 fragment of ~17 kDa and coincided with Caspase-8 and PARP-1 cleavage, suggesting that DNA-damage may also result in Caspase-8-mediated PAR-4 processing (Figure 10A). To analyze if DNA-damage-induced PAR-4 cleavage is Caspase-8-dependent, MCF-7 cells were pre- incubated with Z-IETD-FMK to inhibit Caspase-8 activity followed by Doxorubicin or Etoposide stimulation. Pre-incubation with Z-IETD-FMK prevented PAR-4, Caspase-8 and PARP-1 cleavage in MCF-7 cells, indicating that DNA-damage-induced PAR-4 processing is Caspase-8-dependent (Figure 10B). PAR-4 is down-regulated in recurrent breast cancers [Alvarez et al. 2013]. TNBCs lack ER, PR and HER2/neu expression (Figure 9B) and patients with TNBCs have an increased likelihood of tumor recurrence and death within five years of diagnosis [Dent et al. 2007; Liedtke et al. 2008] (Chapter 1.2.6). To investigate if Caspase-8 is required to induce PAR-4-mediated apoptosis upon exposure to genotoxic drugs Caspase-8 was

60 Results silenced in the TNBC cell lines BT-20 and MDA-MB-468 using lentiviral transduced shRNA targeting Caspase-8 (shCaspase-8 #3) (Chapter 2.4.5). Compared to control cells, PAR-4 and PARP-1 cleavage was strongly reduced in both Caspase-8-deficient cell lines upon Doxorubicin or Etoposide treatment (Figure 10C and 10D), indicating that Caspase-8-mediated PAR-4 cleavage occurs during DNA-damage-induced apoptosis, despite low PAR-4 protein levels in both TNBCs (Figure 9B). Combined drug application is often used to treat cancer, including aggressive metastatic mammary tumors and TNBCs [Lee and Nan 2012; Clark et al. 2014]. Patients suffering from TNBCs initially tend to respond to genotoxic chemotherapy [Dent et al. 2007]. A system-based approach from Lee and colleagues investigated targeted inhibition of a variety of oncogenic signaling pathways in combination with chemotherapeutic drugs. In this study it was demonstrated that time-staggered inhibition of the EGFR using Erlotinib, a selective EGFR tyrosine kinase inhibitor, sensitizes triple negative BT-20 cells to the genotoxic agent Doxorubicin through the extrinsic pathway of apoptosis, including Caspase-8 activation [Lee et al. 2012]. To confirm these findings, BT-20 and MDA-MB-468 cells, both of which overexpress EGFR (Figure 9B), were pre-incubated with Erlotinib and were further stimulated with either Doxorubicin or Etoposide. Indeed, sequential application of Erlotinib followed by genotoxic drug treatment using Doxorubicin or Etoposide led to re-activation of Caspase-8, coincided with accelerated PAR-4 processing and enhanced induction of apoptosis as measured by PARP-1 cleavage (Figure 10E and 10F). The observations suggest that Caspase-8 is required for PAR-4 cleavage within DNA- damage-induced apoptosis (Figure 10B, 10C and 10D) and re-activation of Caspase-8 is accompanied with an increased PAR-4 cleavage (Figure 10E and 10F), raising the question whether Caspase-8-mediated PAR-4 cleavage is required for genotoxic drug- induced apoptosis.

A) HeLa MCF-7 B) MCF-7 + + TNFα/CHX + + + Z-IETD-FMK + + Doxorubicin + + Doxorubicin kDa + + Etoposide kDa + + Etoposide

 cl. PAR-4 (N) 15-  cl. PAR-4 (N) 15-

25-  cl. Caspase-8 25-  cl. Caspase-8

130- 130- PARP-1 PARP-1 100- 100-  cl. PARP-1  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

C) BT-20 MDA-MB-468 D) BT-20 MDA-MB-468 61 + + + + shControl + + + + shControl + + + + shCaspase-8 #3 + + + + shCaspase 8 #3 kDa + + + + Doxorubicin kDa + + + + Etoposide

Caspase-8 55- Caspase-8 55-

40-  cl. Caspase-8 40-  cl. Caspase-8

 cl. PAR-4 (N)  cl. PAR-4 (N) 15- 15- 130- PARP-1 130- PARP-1 100-  cl. PARP-1 100-  cl. PARP-1

40- 40- GAPDH GAPDH 35- 35- E) BT-20 MDA-MB-468 F) BT-20 MDA-MB-468 + + + + Erlotinib + + + + Erlotinib kDa + + + + Doxorubicin kDa + + + + Etoposide

55- Caspase-8 55- Caspase-8

  40- cl. Caspase-8 40- cl. Caspase-8  cl. PAR-4 (N) 15- 15-  cl. PAR-4 (N) 130- 130- PARP-1 PARP-1 100-  cl. PARP-1 100-  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35- A) HeLa MCF-7 B) MCF-7 + + TNFα/CHX + + + Z-IETD-FMK + + Doxorubicin + + Doxorubicin kDa + + Etoposide kDa + + Etoposide

 cl. PAR-4 (N) 15-  cl. PAR-4 (N) 15-

25-  cl. Caspase-8 25-  cl. Caspase-8

130- 130- PARP-1 PARP-1 100- 100-  cl. PARP-1  cl. PARP-1 40- 40- GAPDH GAPDH Results 35- 35-

C) BT-20 MDA-MB-468 D) BT-20 MDA-MB-468 + + + + shControl + + + + shControl + + + + shCaspase-8 #3 + + + + shCaspase 8 #3 kDa + + + + Doxorubicin kDa + + + + Etoposide

Caspase-8 55- Caspase-8 55-

40-  cl. Caspase-8 40-  cl. Caspase-8

 cl. PAR-4 (N)  cl. PAR-4 (N) 15- 15- 130- PARP-1 130- PARP-1 100-  cl. PARP-1 100-  cl. PARP-1

40- 40- GAPDH GAPDH 35- 35- E) BT-20 MDA-MB-468 F) BT-20 MDA-MB-468 + + + + Erlotinib + + + + Erlotinib kDa + + + + Doxorubicin kDa + + + + Etoposide

55- Caspase-8 55- Caspase-8

 cl. Caspase-8  cl. Caspase-8 40- 40-  cl. PAR-4 (N) 15- 15-  cl. PAR-4 (N) 130- 130- PARP-1 PARP-1 100-  cl. PARP-1 100-  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

Figure 10: PAR-4 is cleaved by Caspase-8 following DNA-damage. (A) HeLa and MCF-7 cells were treated with TNFα (10 ng/ml) in combination with CHX (0.5 µg/ml) for 8 h or Doxorubicin (10 µM) for 8 h (HeLa) or 16 h (MCF-7) or Etoposide (100 µM) for 24 h. Cell lysates were analyzed by immunoblotting using antibodies against cleaved Caspase-8, PARP-1 and GAPDH. PAR-4 antibodies were used to detect the N-terminal cleavage fragment. (B) MCF-7 cells were pre-incubated with the Caspase-8 inhibitor Z-IETD-FMK (20 µM) for 30 min followed by treatment with Doxorubicin (10 µM) for 16 h or Etoposide (100 µM) for 24 h. Lysates were analyzed via immunoblotting with the indicated antibodies. (C) BT-20 and MDA-MB-468 control (shControl) or Caspase-8-deficient (shCaspase-8 #3) cells were treated with Doxorubicin (10 µM) for 8 h and lysates were analyzed with the indicated antibodies. (D) Caspase-8-deficient cells from (C) were treated with Etoposide (100 µM) for 24 h and compared to control cells. Protein lysates were analyzed as in (C). (E) BT-20 and MDA-MB-468 cells were pre-treated with Erlotinib (10 µM) for 16 h and then incubated with Doxorubicin (10 µM) for 6 h. Cell lysates were analyzed by immunoblotting using antibodies against Caspase-8, PAR-4, PARP-1 and GAPDH. (F) Cells were pre-treated as in (E) followed by Etoposide (100 µM) treatment for 6 h and analyzed via immunoblotting as described in (E).

62 Results

3.7 Caspase-8-mediated PAR-4 cleavage is required for DNA-damage-induced cell

death

Consistent with its role in the regulation of apoptosis, exogenous expression of PAR-4 has been demonstrated to sensitize a large subset of cancer cells to apoptotic stimuli and low PAR-4 expression is associated with poor response to neoadjuvant chemotherapy [Hebbar, Wang, and Rangnekar 2012; Alvarez et al. 2013]. To investigate whether PAR-4 can sensitize breast cancer cells to DNA damage, wild-type PAR-4 was overexpressed in the TNBC cell lines BT-20 and MDA-MB-468 and the expression efficiency was analyzed (Figure 11A). Forced expression of PAR-4 WT alone did result in moderate PAR-4 and PARP-1 cleavage in these TNBC cells. Additional treatment of cells expressing exogenous PAR-4 with Doxorubicin or Etoposide resulted in elevation of PAR-4, Caspase-8 and PARP-1 processing, which indicates that overexpression of PAR-4 renders these cells sensitive towards DNA- damage (Figure 11B and 11C). To verify if PAR-4 is required for DNA damage induced apoptosis in breast cancer cells, PAR-4 expression was silenced in BT-20 and MDA- MB-468 cells using two siRNAs (siPAR-4 #5 and #8) specifically targeting PAR-4. Knock down efficiency in both cell lines was tested and compared with control cells containing a non targeting siRNA (Figure 11D). PAR-4-deficient cells were then treated either with Doxorubicin or Etoposide and induction of apoptosis was measured by Caspase-8 activation and PARP-1 processing. PAR-4 cleavage was monitored simultaneously to Caspase-8 and PARP-1 cleavage in PAR-4-expressing cells upon genotoxic drug treatment, while Caspase-8 and PARP-1 cleavage was prevented in PAR-4-deficient cells (Figure 11E and 11F). To validate these findings, apoptotic cells were quantified by flow cytometric detection of the sub-G1 fraction during cell cycle analysis. BT-20 and MDA-MB-468 control cells showed enhanced apoptosis rates following Etoposide treatment, while PAR-4 deficiency rendered cells significantly insensitive to DNA-damage-induced apoptosis (Figure 11G and 11H). The combined data indicate that Caspase-8-mediated PAR-4 cleavage is required to induce DNA- damage-induced apoptosis and that PAR-4 processing further promotes Caspase-8 activation.

63 Results

A) BT-20 MDA-MB-468 + + Control kDa + + PAR-4 WT

PAR-4 40- 40- GAPDH 35-

B) BT-20 MDA-MB-468 C) BT-20 MDA-MB-468 + + + + Control + + + + Control + + + + PAR-4 WT + + + + PAR-4 WT kDa + + + + Doxorubicin kDa + + + + Etoposide

15-  cl. PAR-4 (N) 15-  cl. PAR-4 (N)

25-  cl. Caspase-8 25-  cl. Caspase-8 130- 130- PARP-1 PARP-1 100- 100-  cl. PARP-1  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

D) BT-20 MDA-MB-468 + + siControl + + siPAR - 4#5 kDa + + siPAR-4#8

PAR-4 40-

40- GAPDH 35-

E) BT-20 MDA-MB-468 F) BT-20 MDA-MB-468 + + + + siControl + + + + siControl + + + + siPAR-4 #5 + + + + siPAR-4 #5 + + + + siPAR-4 #8 + + + + siPAR-4 #8 kDa + + + + + + Doxorubicin kDa + + + + + + Etoposide

15-  cl. PAR-4 (N) 15-  cl. PAR-4 (N)

25-  cl. Caspase-8 25-  cl. Caspase-8 130- 130- PARP-1 PARP-1 100- 100-  cl. PARP-1  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

G) BT-20 H) MDA-MB-468 p < 0.005 p < 0.005 35 p < 0.005 p < 0.005 30 40 35 25 30 20 25 G1 [%] G1 [%] G1 - 15 - 20 15

Sub 10 Sub 10 5 5 + + siControl + + siControl + + siPAR-4 #5 + + siPAR-4 #5 + + siPAR-4 #8 + + siPAR-4 #8 + + + Etoposide + + + Etoposide Figure 11: PAR-4 is required for DNA-damage-induced apoptosis. (A) Overexpression of PAR-4 wild-type (WT) in BT-20 and MDA-MB-468 cells for 24h. Cell lysates were analyzed by using antibodies against PAR-4 and GAPDH. (B) Cells were transfected as in (A), incubated with Doxorubicin (10 µM) for 8 h and lysates were analyzed with the indicated antibodies. (C) BT-20 and MDA-MB-468 cells were transfected as in (A), stimulated with Etoposide (100 µM) for 24 h and analyzed 64 Results as described above. (D) siRNA-mediated PAR-4 knockdown (siPAR-4 #5 and #8) cells were compared to control cells (siControl) and analyzed by immunoblotting using antibodies against PAR-4 and GAPDH. (E) BT-20 and MDA-MB-468 cells were treated as in (D) and incubated with Doxorubicin (10 µM) for 7 h. Cell lysates were analyzed by immunoblotting using the indicated antibodies. (F) Cells were treated as in (D) and incubated with Etoposide (100 µM) for 16 h. Cell lysates were analyzed by immunoblotting using antibodies against PAR-4, PARP-1 and GAPDH. (G) BT-20 cells from (F) were prepared for cell cycle analysis and sub-G1population was measured via flow cytometry. Average values from three independent experiments are shown. Error bars indicate ± SD. P-value was obtained by two-tailed Student’s t-Test. (H) MDA-MB-468 cells from (F) were treated and analyzed as in (G).

3.8 The C-terminal PAR-4 fragment translocates to the nucleus upon DNA-

damage

TNFα-induced apoptosis requires Caspase-8-mediated PAR-4 processing and resulted in the nuclear translocation of the C-terminal PAR-4 fragment containing the functional important SAC and LZ domains (Figure 8B and 8D). To study the localization of the C- terminal PAR-4 cleavage product upon DNA-damage, PAR-4 mutants comprising either a C-terminal eCFP tag or an N-terminal located eYFP tag were generated and expressed in BT-20 and MDA-MB-468 cells. Both TNBC cell lines showed cytosolic localization of PAR-4 WT-CFP and the Caspase-8 cleavage resistant mutant PAR-4 D131G-CFP, whereas the PAR-4 mutant lacking the amino-terminal part (PAR-4 132- 340-CFP) localized to the nuclear compartment. The PAR-4 mutant lacking the C- terminal part (YFP-PAR-4 1-131) was equally distributed throughout the cell (Figure 12A and 12B). Moreover, stimulation with either Doxorubicin or Etoposide resulted in nuclear accumulation of wild-type PAR-4, but was prevented in cells expressing PAR-4 D131G- CFP (Figure 12A and 12B), as it was shown for this mutant in HEK293 cells following TNFα/CHX treatment or UV irradiation (Figure 8B). In contrast, DNA-damage had no influence on the localization of both deletion mutants as PAR-4 132-340-CFP remained nuclear and cellular distribution of YFP-PAR-4 1-131 was also unchanged (Figure 12A and 12B). To confirm these observations, endogenous PAR-4 was stained additionally using antibodies recognizing the C-terminal part of the protein. In the absence of Etoposide, PAR-4 mainly localized to the cytosolic compartment, whereas an accumulation in the nucleus was monitored after Etoposide treatment (Figure 12C). This indicates that DNA-damage-induced cleavage of PAR-4 by Caspase-8 results in nuclear accumulation of the C-terminal PAR-4 fragment and subsequent induction of cell death.

65 Results

A) PAR-4 WT-eCFP PAR-4 D131G-eCFP PAR-4 131-240-eCFP eYFP-PAR-4 1-131

eCFP eCFP eCFP eYFP

PAR-4 WT-eCFP PAR-4 D131G-eCFP PAR-4 132-340-eCFP eYFP-PAR-4 1-131

+ + + + Doxorubicin 20 - BT 468 - MB - MDA

B) PAR-4 WT-eCFP PAR-4 D131G-eCFP PAR-4 132-340-eCFP eYFP-PAR-4 1-131 + + + + Etoposide 20 - BT 468 - MB - MDA

C) BT-20 MDA-MB-468 + + Etoposide

PAR-4 Hoechst

Figure 12: The C-terminal PAR-4 fragment translocates to the nucleus upon DNA-damage. (A) Schematic depiction of C-terminal CFP-tagged wild-type PAR-4 (PAR-4 WT-eCFP), a non cleavable PAR-4 point mutant (PAR-4 D131G-eCFP, mutation marked in red), a deletion mutant encompassing amino acids 132-340 (PAR-4 132-340-eCFP) and an N-terminal YFP-tagged PAR-4 deletion mutant containing amino acids 1-131 (eYFP-PAR-4 1-131) (upper panel). BT-20 and MDA-MB-468 cells were transiently transfected with the indicated plasmids (1 µg), treated with Doxorubicin (10 µM) for 8 h and analyzed by immunofluorescence using confocal microscopy (lower panel). (B) BT-20 and MDA-MB-468 cells were transfected as in (A), incubated with Etoposide (100 µM) for 24 h and analyzed as described above. (C) BT-20 and MDA-MB-468 cells were treated with Etoposide (100 µM) for 24 h followed by immunostaining with PAR-4 antibodies recognizing the C-terminal part of the protein (green). Cells were additionally treated with Hoechst for nuclear staining (blue) and analyzed by confocal microscopy.

66 Results

3.9 C-terminal PAR-4 fragment is sufficient to induce apoptosis

In structure-function analysis El-Guendy and colleagues reported that the SAC domain of PAR-4 is sufficient to induce apoptosis in cancer cells following nuclear translocation [El-Guendy et al. 2003; Zhao et al. 2007]. To analyze whether the C-terminal cleavage product of PAR-4 containing the SAC and the LZ domains is sufficient to induce apoptotic cell death, both cleavage fragments were expressed in BT-20 and MDA-MB- 468 cells and additionally stimulated with Doxorubicin or Etoposide. Caspase-8 and PARP-1 cleavage, implying apoptosis, was not detectable following expression of the N- terminal cleavage fragment of PAR-4 (YFP-PAR-4 1-131) in both cell lines and DNA- damage-induced cell death by Doxorubicin or Etoposide was unchanged (Figure 13A and 13B), suggesting that the N-terminal PAR-4 cleavage fragment is not involved in apoptotic processes. In contrast, expression of the C-terminal PAR-4 fragment (PAR-4 132-340-CFP) alone resulted in enhanced Caspase-8 activation and PARP-1 cleavage and DNA-damage-induced apoptosis was further enhanced by forced expression of PAR-4 132-340-CFP (Figure 13C and 13D). To validate these findings, apoptotic cells were quantified by flow cytometric measurement of the sub-G1 population in cell cycle analysis. Apoptosis was induced by the expression of the C-terminal PAR-4 fragment, whereas Etoposide-induced cell death was increased in BT-20 and MDA-MB-468 cells, indicating that the C-terminal PAR-4 cleavage fragment is sufficient to induce apoptosis by its own and promotes DNA-damage-mediated cell death (Figure 13E and 13F).

A) BT-20 MDA-MB-468 B) BT-20 MDA-MB-468 + + + + Control + + + + Control + + + + eYFP-PAR-4 1-131 + + + + eYFP-PAR-4 1-131 kDa + + + + Doxorubicin kDa + + + + Etoposide

40- eYFP-PAR-4 1-131 40- YFP-PAR-4 1-131

25-  cl. Caspase-8 25-  cl. Caspase-8 130- 130- PARP-1 PARP-1 100-  cl. PARP-1 100-  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

C) BT-20 MDA-MB-468 D) BT-20 MDA-MB-468 + + + + Control + + + + Control + + + + PAR-4 132-340-eCFP + + + + PAR-4 132-340-eCFP kDa + + + + Doxorubicin kDa + + + + Etoposide

PAR-4 132-340-eCFP 55- PAR-4 132-340-eCFP 55-

25-  cl. Caspase-8 25-  cl. Caspase-8 67 130- PARP-1 130- PARP-1 100-  cl. PARP-1 100-  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

E) BT-20 F) MDA-MB-468

p < 0.005 p < 0.005 p < 0.005 50 50 40 40 p < 0.005 30 G1 [%] G1 30 [%] G1 - - 20 20 Sub Sub 10 10

+ + Control + + Control + + PAR-4 132-340-eCFP + + PAR-4 132-340-eCFP + + Etoposide + + Etoposide A) BT-20 MDA-MB-468 B) BT-20 MDA-MB-468 + + + + pcDNA5 + + + + pcDNA5 + + + + eYFP-PAR-4 1-131 + + + + eYFP-PAR-4 1-131 kDa + + + + Doxorubicin kDa + + + + Etoposide

40- eYFP-PAR-4 1-131 40- YFP-PAR-4 1-131

25-  cl. Caspase-8 25-  cl. Caspase-8 130- 130- PARP-1 PARP-1 100-  cl. PARP-1 100-  cl. PARP-1 40- 40- GAPDH GAPDH Results 35- 35-

C) BT-20 MDA-MB-468 D) BT-20 MDA-MB-468 + + + + Control + + + + Control + + + + PAR-4 132-340-eCFP + + + + PAR-4 132-340-eCFP kDa + + + + Doxorubicin kDa + + + + Etoposide

PAR-4 132-340-eCFP 55- PAR-4 132-340-eCFP 55-

25-  cl. Caspase-8 25-  cl. Caspase-8

130- PARP-1 130- PARP-1 100-  cl. PARP-1 100-  cl. PARP-1 40- 40- GAPDH GAPDH 35- 35-

E) BT-20 F) MDA-MB-468

p < 0.005 p < 0.005 p < 0.005 50 50 40 40 p < 0.005 30 G1 [%] G1 30 [%] G1 - - 20 20 Sub Sub 10 10

+ + Control + + Control + + PAR-4 132-340-eCFP + + PAR-4 132-340-eCFP + + Etoposide + + Etoposide

Figure 13: C-terminal PAR-4 fragment is sufficient to induce apoptosis. (A) BT-20 and MBA-MB-468 cells were transiently transfected with 1 µg of the N-terminal YFP-tagged PAR-4 1-131 deletion mutant (YFP-PAR-4 1-131) and treated with Doxorubicin (10 µM) for 8 h. Cell lysates were analyzed by immunoblotting using antibodies recognizing GFP, cleaved Caspase-8, PARP-1 and GAPDH. (B) Cells were transfected as in (A), incubated with Etoposide (100 µM) for 16 h and analyzed as described before. (C) BT-20 and MBA-MB-468 cells were transiently transfected with 1 µg of the C-terminal CFP-tagged PAR-4 132-340 deletion mutant (PAR-4 132-340-CFP) and treated with Doxorubicin (10 µM) for 8 h. Cell lysates were analyzed by immunoblotting using antibodies recognizing GFP, cleaved Caspase-8, PARP-1 and GAPDH. (D) Cells were transfected as in (C), incubated with Etoposide (100 µM) for 16 h and analyzed as described above. (E) BT-20 cells were transfected and treated as in (D) and sub-G1 population was quantified using flow cytometry analysis. Average values from three independent experiments are displayed. Error bars indicate ± SD. P-value was calculated by two-tailed Student’s t-Test. (F) MDA-MB-468 cells were transfected and analyzed as in (E).

3.10 PAR-4 binds to the deubiquitylating enzyme USP7

Caspase-8-mediated PAR-4 processing and nuclear translocation of the C-terminal cleavage fragment is important to activate its pro-apoptotic function, but how PAR-4 orchestrates the apoptotic process remains elusive. Several studies report that PAR-4 can interact with diverse proteins. Because PAR-4 lacks enzymatic activity interaction

68 Results with other proteins is required to attenuate pro-survival signaling or augment molecular pathways coupled to pro-apoptotic signals [Hebbar, Wang, and Rangnekar 2012]. To gain further insights into the molecular functions of PAR-4, mass spectrometry analysis was conducted to search for potential PAR-4 interaction partners. For this approach expression of FLAG-labeled wild-type PAR-4 (PAR-4 WT-FLAG) was induced in HeLa Flp-In™ T-REx cells with Doxycycline and PAR-4-incorporated protein complexes were immunoprecipitated via FLAG® M2 beads. Protein expression was analyzed by SDS-PAGE and Coomassie Brilliant Blue staining (Figure 14A). In parallel, tryptic digestion of precipitated PAR-4 protein complexes was carried out and analyzed by an unbiased mass spectrometry approach. Proteins with an intensity >5 fold over background (empty vector) were considered significant. The analysis revealed nine potential PAR-4 interaction partners and identified Caldesmon (CALD1), mRNA export factor (RAE1) and the heat shock protein 105 kDa (HSPH1). I further identified protein phosphatase 1 (PPP1) regulatory subunits 12A and 12B (PPP1R12A/B) and the PP1- beta catalytic subunit (PPP1CB). Among these nine potential PAR-4 interaction partners the analysis also revealed three ubiquitin-related factors, including the E3 ubiquitin ligases Myc binding protein 2 (MYCBP2), Ubiquitin protein ligase E3 component N- recognin 5 (UBR5) and the deubiquitinating enzyme Ubiquitin specific peptidase 7 (USP7) (Figure 14B). Due to the fact that USP7 has critical functions in the regulation of proteins involved in tumor suppression and DNA repair [Pfoh, Lacdao, and Saridakis 2015], reciprocal co-immunoprecipitation experiments were carried out to validate USP7 as a PAR-4 interacting protein. Indeed, endogenous USP7 was co-immunoprecipitated in HeLa Flp-In™ T-REx cells expressing PAR-4 WT-FLAG, whereas expression of FLAG-tagged USP7 interacted with endogenous PAR-4 (Figure 14C and 14D). Taken together, the experiments indicate that PAR-4 binds to the deubiquitinating enzyme USP7 in vivo.

69 Results

A) HeLa T-REx B) PAR-4 WT-FLAG

kDa + Doxycycline 170- 130- 100- 70- 55- heavy chain 40- PAR-4 WT-FLAG 35- 25- light chain

C) HeLa T-REx D) HeLa T-REx + Control + Control kDa + PAR-4-FLAG kDa + FLAG-USP7

PAR-4-FLAG FLAG-USP7 40- 130- IP: FLAG IP: FLAG 130- USP7 40- PAR-4

PAR-4-FLAG 130- FLAG-USP7 40- USP7 WCL WCL 130- 40- PAR-4 40- 40- GAPDH GAPDH 35- 35-

Figure 14: PAR-4 interacts with USP7. (A) FLAG-labeled wild-type PAR-4 (PAR-4 WT-FLAG) was immunoprecipitated from HeLa Flp-In™ T- REx cell lysates using FLAG® M2 beads after 24 h incubation with Doxycycline (100 ng/ml) to force gene expression. Precipitates were separated via SDS-PAGE followed by Coomassie Brilliant Blue staining. (B) HeLa Flp-In™ T-REx cells were treated as in (A) and proteins captured from immunoprecipitated PAR-4 WT-FLAG expressing cells were identified by mass spectrometry analysis after tryptic digestion. Proteins identified in two independent experiments with an intensity >5 fold over background (empty vector) were considered significant and are listed. (C) HeLa Flp-In™ T-REx PAR-4 WT-FLAG cells and control cells were treated with Doxycycline (100 ng/ml) for 24 h. PAR-4 was immunoprecipitated from cell lysates using FLAG® M2 beads. Whole cell lysates (WCL) and purified proteins were analyzed by immunoblotting using antibodies against PAR-4, USP7 and GAPDH. (D) HeLa Flp-In™ T-REx cells were transiently transfected with FLAG-tagged USP7 or empty vector (1 µg each) for 24 h. USP7 was immunoprecipitated from cell lysates using FLAG® M2 beads, whereupon whole cell lysates and precipitates were analyzed as in (C).

70 Discussion

4 Discussion

The pro-apoptotic tumor suppressor protein PAR-4 is silenced in a specific subset of human cancers, including lung, prostate and breast cancer. Within the last few years PAR-4 gained attention from scientists working on breast cancers, the most common cancer in women worldwide. Among other cancer entities, breast cancer is a heterogeneous disease, thereby limiting therapeutic approaches and making its clinical progression difficult to predict. Most patients with breast cancer die from disease relapse following treatment of the primary tumor tissue. Because breast cancer recurrence is typically an incurable disease, understanding the mechanisms by which residual cancer cells survive treatment is required to improve therapies for breast cancer patients following disease relapse. Interestingly, down-regulation of PAR-4 was demonstrated to be a key regulatory event that is necessary and sufficient to promote breast cancer recurrence following chemotherapy. Re-expression of PAR-4 renders tumor cells sensitive to a broad spectrum of cell death-inducing stimuli, but how the pro- apoptotic activities of the tumor suppressor protein are controlled and executed precisely is poorly understood. In this study I demonstrate that PAR-4 is cleaved upon UV-, TNFα- and DNA-damage- mediated induction of cell death at EEPD131↓G and this cleavage site was preferentially recognized by Caspase-8. Caspase-8-dependent processing of PAR-4 separates the unstructured N-terminal part from the C-terminal region containing the NLS, SAC and leucine zipper domains. I further observe nuclear accumulation of the C- terminal PAR-4 cleavage fragment as a critical step for the induction of apoptosis mediated by TNFα or genotoxic drug treatment. In addition, I show that Caspase-8- mediated PAR-4 cleavage is required for TNFα- or DNA-damage-induced apoptosis in breast cancer cells, and loss of PAR-4 renders breast cancer cells insensitive to TNFα- or DNA-damage-induced apoptosis. Interestingly, I demonstrate that PAR-4 is not only a bona fide Caspase-8 substrate it also interferes with Caspase-8 activation following genotoxic drug treatment. Low protein levels of PAR-4 promote tumor cell survival following therapy but the pathways controlling PAR-4 expression remain elusive. I was able to identify MYCBP2, UBR5 and USP7 as physical interaction partners of PAR-4 using an unbiased mass spectrometry approach. MYCBP2 and UBR5 belong to the family of E3 ubiquitin-protein

71 Discussion ligases, while USP7 is a deubiquitinating enzyme, suggesting that down-regulation of PAR-4 might be associated with disordered ubiquitin signaling.

4.1 Caspase-dependent PAR-4 cleavage interferes with its pro-apoptotic activities

Expression of PAR-4 sensitizes cancer cells towards diverse cell death-inducing factors and the initial experiments demonstrate that PAR-4 is cleaved upon UV exposure (20 mJ x cm-2 at 254 nm) in a Caspase-dependent manner (Figure 5B). The Caspase cleavage site of PAR-4 at position EEPD131↓G is conserved in human and rodent orthologs and separates the unstructured N-terminal region from the C-terminal part, which comprises the NLS-containing SAC domain and the LZ domain (Figure 5E). Under these conditions UV exposure results in DNA-damage by formation of cyclobutane pyrimidine dimers, as the maximum absorption spectrum of DNA (260 nm) lies in the UVC spectrum (100-280 nm) [Kulms and Schwarz 2000]. In response to DNA-damage, the intrinsic pathway of apoptosis is activated via MOMP (Chapter 1.2.3). Moreover, UV light is capable to activate the extrinsic pathway of apoptosis via upregulation and clustering of death receptors. In addition, UV exposure results in upregulation of death receptor ligands, which initiates apoptosis following their autocrine release and activation of the Caspase cascade [Kulms and Schwarz 2000; Lee et al. 2013]. To investigate the effect on cell growth for this cleavage event stable cell lines were established that express wild-type PAR-4 or the Caspase cleavage resistant mutant PAR-4 D131G under the control of a Doxycycline-inducible promoter (Figure 6A). Moderate overexpression of the wild-type protein but not the mutant interfered with cell proliferation, predominantly through the induction of apoptosis (Figure 6B and 6C). Interestingly, PAR-4 cleavage was observed upon overexpression without any additional stress signal, indicating that PAR-4 somehow leads to the activation Caspases (Figure 6A and 6C). The combined results demonstrate that Caspase- dependent PAR-4 cleavage is an important mechanism to initiate its pro-apoptotic activities and that PAR-4 is not a bystander substrate of Caspases during the execution of apoptosis, as observed for many proteins [Crawford and Wells 2011].

72 Discussion

4.2 PAR-4 is a bona fide Caspase-8 substrate

Caspase-dependent PAR-4 cleavage is a key regulatory event to activate the pro- apoptotic functions of the tumor suppressor protein, although PAR-4 cleavage was observed to occur moderately over time upon UV-induced apoptosis or overexpression studies (Figure 5A and 6A). This is supported by the idea of a gain of function event, in which a small portion of the nascent product is competent to signal, compared to a loss of function event, which is characterized by a substantial loss of the protein [Timmer and Salvesen 2007]. Indeed, my observations revealed that overexpression of wild-type PAR-4, but not the non cleavable PAR-4 mutant, resulted in a minimal PAR-4 cleavage, which was sufficient to induce apoptosis, suggesting a gain of function event (Figure 6A and 6C). In order to identify Caspases that are capable of processing PAR-4, immunoprecipitated PAR-4 was subjected to an in vitro Caspase cleavage assay using human recombinant Caspases 1 to 10. PAR-4 was observed to be processed in a Caspase assay by Caspase-1, -7 and -8 to various degrees with Caspase-8 being the most efficient to cleave PAR-4 (Figure 7A). All Caspases display substrate specificity with a strict requirement for D at position P1 and a recognition sequence of at least four amino acids N-terminal located to the site to be cleaved, mainly determining substrate specificity of Caspases (P4-P3-P2-P1-↓-P1’) [Pop and Salvesen 2009]. Besides D at position P1, the optimal Caspase-8 recognition site was defined to contain L, V, D, or E at P4, E at P3 and P2 being T, V, or I [Thornberry et al. 1997]. Additionally, Caspase-8 prefers G or S at position P1’ [Stennicke et al. 2000]. The PAR-4 cleavage site EEPD↓G fulfills these requirements except for residue P2. Instead of T, V, or I PAR-4 possesses P at position P2, a hydrophobic amino acid as V and I. The P1’ residue was determined to require small and uncharged amino acids (G, S, A) and instead of G at P1 in human PAR-4, mouse and rat PAR-4 display S at this position, which fits with the amino acid required for a bona fide Caspase-8 substrate. The primary structure of potential Caspase substrates displays a critical feature to predict Caspase recognition, but substrate recognition occurs spatially within the cell based on secondary, tertiary and quaternary structures [Crawford and Wells 2011]. Classical Caspase-8 activation is facilitated by death receptor trimerization upon binding of their cognate ligands such as TNFα, at the cell surface, whereupon Caspase-8 is recruited to the DISC complex, thereby activating itself via autoproteolytic cleavage.

73 Discussion

Once Caspase-8 gets activated it initiates the processing of Caspase-3 and -7 accompanied with the execution of apoptosis [Oberst and Green 2011] (Chapter 1.2.3). Activation of the extrinsic pathway via TNFα/CHX resulted in PAR-4 cleavage during apoptosis (Figure 7B), which was prevented by siRNA-mediated depletion or enzymatic inhibition of Caspase-8 (Figure 7D and 7F), indicating that PAR-4 cleavage is Caspase- 8-dependent in vivo. Importantly, PAR-4 was reported to comprise a potential death domain within its C-terminal region similar to the death domains of Fas, FADD, TNFR1 and TRADD [Diaz-Meco et al. 1996]. The potential death domain of PAR-4 might be responsible for the recruitment of PAR-4 to Fas or TNFR1 and could explain the physiological interaction of PAR-4 and Caspase-8. Chemical inhibitors targeting Caspase-8 might also interfere with the catalytic activity of Caspase-10, as it is the closest homolog of Caspase-8 in humans with overlapping cleavage preferences for several substrates [Fischer, Stroh, and Schulze-Osthoff 2006]. But, since human recombinant Caspase-10 was not able to cleave immunoprecipitated PAR-4 in vitro (Figure 7A), potential processing of PAR-4 by Caspase-10 was not further investigated. In 2012 Chaudry and colleagues identified PAR-4 as a substrate of Caspase-3 in apoptotic cells and demonstrated that Cisplatin-induced PAR-4 cleavage is prevented in Caspase-3-deficient MCF-7 breast cancer cells [Chaudhry et al. 2012]. Surprisingly, Caspase-3 was not able to hydrolyze PAR-4 in a Caspase assay in vitro (Figure 7A) and re-expression of Caspase-3 in MCF-7 cells demonstrated that PAR-4 cleavage is independent of Caspase-3 (Figure 7E). Moreover, inhibition of Caspase-3 was observed to only partially abrogate TNFα/CHX-induced PAR-4 cleavage, whereas inhibition of Caspase-8 totally blocked PAR-4 processing following treatment (Figure 7G). Co- expression of PAR-4 with Caspase-8, but not with Caspase-3 resulted in PAR-4 cleavage and induction of apoptosis (Figure 7H). Thus, my results strongly suggest that PAR-4 requires Caspase-8 to be processed upon TNFα/CHX-induced apoptosis, but is independent of Caspase-3 activity. The effector Caspases Caspase-7 and -3 share close homology and display an overall sequence identity of 56%. Both Caspases share similar enzymatic activity towards certain synthetic peptides, but it is also reported that Caspase-3 and -7 exhibit differential activities regarding multiple cellular substrates [Walsh et al. 2008]. Besides Caspase-7 and -8, Caspase-1 was capable to hydrolyze PAR-4 in vitro (Figure 7A). Caspase-1 belongs to the group of inflammatory Caspases with key regulatory

74 Discussion functions in innate immunity. It is known that Caspase-1 is involved in the response to inflammation by activating pro-inflammatory cytokines [Sollberger et al. 2014]. Interestingly, Caspase-1-dependent activation of Caspase-7 was demonstrated to regulate pro-inflammatory gene expression via PARP-1 cleavage independent from apoptotic signaling [Erener et al. 2012]. A regulatory role of PAR-4 processing mediated by Caspase-1 or -7 cannot be excluded and future experiments will analyze a potential function of PAR-4 downstream of Caspase-1 and -7.

4.3 Caspase-8 is activated in breast cancer cells upon genotoxic drug treatment

and is required for apoptosis induction

The data so far demonstrates that PAR-4 is a substrate of Caspase-8 in vitro and in vivo and that Caspase-8-mediated PAR-4 cleavage is required to induce TNFα-induced apoptosis in MCF-7 breast cancer cells (Figure 8E and 8F). Surprisingly, recent advances in analyzing apoptotic signaling networks have uncovered an important role for the activation of Caspase-8 following DNA damage caused by genotoxic drugs [Wesselborg et al. 1999; Chandra et al. 2004; Sohn, Schulze-Osthoff, and Jänicke 2005]. Consistent with this, Caspase-8-dependent PAR-4 cleavage was observed in MCF-7 breast cancer cells upon genotoxic drug treatment with Doxorubicin or Etoposide (Figure 10B), two chemotherapeutic agents used to treat cancer patients (Chapter 1.2.6). However, the mechanisms how DNA-damage results in Caspase-8 activation are not fully understood and need to be investigated in more detail. Biton and Ashkenazi demonstrated that extensive DNA-damage in HeLa cells drives prolonged NF-κB-mediated gene expression of TNFα, whereupon autocrine signaling of TNFα induces RIPK1-dependent recruitment of FADD and activation of Caspase-8 [Biton and Ashkenazi 2011]. Cancer cells that are independent of death receptor signaling deliver DNA-damage-induced Caspase-8 activation via a signaling platform, referred to as the Ripoptosome. The Ripoptosome comprises RIPK1, FADD and Caspase-8 and is formed upon genotoxic drug-induced depletion of IAP proteins or cFLIP, antagonists of this death-inducing signaling complex [Tenev et al. 2011; Feoktistova et al. 2011]. These observations demonstrate an unexpected but important role for the activation of Caspase-8 for DNA damage-induced apoptosis in tumor cells. My own data so far suggests an important function for PAR-4 as a Caspase-8 substrate after TNFα-induced apoptosis in various cancer cells. Interestingly, PAR-4 expression was demonstrated to 75 Discussion be strongly down-regulated on mRNA and protein level in recurrent breast cancer tumors (Figure 9A) and down-regulation of PAR-4 is necessary and sufficient to promote tumor recurrence following chemotherapy [Alvarez et al. 2013]. To answer the question whether the Caspase-8/PAR-4 axis is important to drive breast cancer cells into genotoxic drug-induced apoptosis, the TNBC cell lines BT-20 and MDA-MB-468 were used, as TNBCs lack expression of hormone receptors and are therefore limited to chemotherapeutic approaches. Both TNBC cell lines display Caspase-8 and PAR-4 expression (Figure 9B) and silencing of Caspase-8 in these cells prevents induction of apoptosis and PAR-4 cleavage upon DNA-damage mediated by Doxorubicin or Etoposide (Figure 10C and 10D), which confirms the requirement for Caspase-8 activation in cytotoxic drug-induced apoptosis in TNBC cells. MDA-MB-468 and BT-20 cells have genetic amplification of the EGFR gene and exhibit high levels of the EGFR protein [Filmus et al. 1985; Lebeau and Goubin 1987]. My observations demonstrate that time-shifted incubation of both TNBC cell lines with the EGFR inhibitor Erlotinib and Doxorubicin or Etoposide promotes accelerated PAR-4 cleavage accompanied with an increase of apoptosis (Figure 10E and 10F). This is supported by the idea that targeted inhibition of oncogenic signaling pathways can be rewired by drugs, whereupon malignant tissues can be treated with high specificity and efficacy, as it was demonstrated for aberrant EGFR signaling [Kang et al. 2010; Morgillo et al. 2011]. Intriguingly, Lee and colleagues reported that only sequential application of Erlotinib followed by Doxorubicin treatment in the TNBC cell line BT-20 leads to the activation of the extrinsic pathway of apoptosis, including Caspase-8 activation. Moreover, they also observed Caspase-8-mediated apoptosis upon time-staggered treatment with Erlotinib and Doxorubicin in different lung cancer cell lines, which express high levels of EGFR [Lee et al. 2012]. Therefore, it can be speculated that EGFR-mediated suppression of Caspase-8 activation prevents PAR-4 cleavage by Caspase-8 and induction of genotoxic drug-induced apoptosis in other malignant tissues that express high levels of EGFR, including TNBCs and lung cancers. Taken together, the observations indicate that Caspase-8 is activated in TNBC cells upon DNA-damage and its activation is required for cytotoxic drug-induced apoptosis mediated by Doxorubicin and Etoposide. Caspase-8 in turn cleaves PAR-4, raising the question, whether Caspase-8-dependent PAR-4 cleavage is required to deliver genotoxic drug-induced apoptosis in TNBC cells.

76 Discussion

4.4 PAR-4 cleavage is required for genotoxic drug-induced apoptosis and

interferes with Caspase-8 activation

Overexpression of PAR-4 has been demonstrated to sensitize various cancer cells to a broad spectrum of cell death-inducing stimuli and the pro-apoptotic activities of PAR-4 are closely connected with its tumor suppressor functions [Hebbar, Wang, and Rangnekar 2012]. Down-regulation of PAR-4 was observed in several cancer entities, explaining a molecular mechanism provided by cancer cells to avoid induction of an apoptotic program, which serves as a hallmark of cancer during tumorigenesis. PAR-4 was observed to be silenced in breast cancer tissues and is associated with an aggressive tumor phenotype [Zapata-Benavides et al. 2009; Méndez-López et al. 2010]. Moreover, PAR-4 serves as a prognostic marker during breast cancer development, as down-regulation of PAR-4 promotes tumor recurrence [Nagai et al. 2010; Alvarez et al. 2013]. To investigate if Caspase-8-dependent PAR-4 cleavage is required for genotoxic drug- induced apoptosis, overexpression of PAR-4 should sensitize TNBC cells following treatment. Indeed, PAR-4 overexpression in BT-20 and MDA-MB-468 cells promotes apoptosis upon genotoxic drug application (Figure 11B and 11C). Consistent with these findings, knockdown of PAR-4 renders cells insensitive to DNA-damage-induced cell death mediated by Doxorubicin or Etoposide (Figure 11E, 11F, 11G and 11H), indicating that Caspase-8-dependent PAR-4 cleavage is required to induce genotoxic drug-induced apoptosis in TNBC cells. Regulation of Caspase-8 activity following cytotoxic drug treatment was expected to be independent of PAR-4 expression, as Caspase-8 regulation is upstream of PAR-4. Surprisingly, PAR-4-depleted cells show no Caspase-8 activity following DNA-damage (Figure 11E and 11F), indicating that PAR-4 interferes with the activation of the apical Caspase-8 via an unknown feedback mechanism. This feedback loop might also explain why overexpression of PAR-4 in both TNBC cell lines display increased levels of activated Caspase-8 upon DNA- damage (Figure 11B and 11C). As previously discussed DNA-damage-induced apoptosis can be enabled through the formation of the Ripoptosome, which is activated by loss of IAP proteins or cFLIP [Tenev et al. 2011; Feoktistova et al. 2011]. Intriguingly, enhanced protein levels of XIAP were observed in primary embryonic fibroblasts derived from PAR-4 knockout mice and in the uteri of PAR-4 female knockout mice, which are prone to develop spontaneous endometrial carcinomas [Garcia-Cao et al.

77 Discussion

2003; García-Cao et al. 2005]. Notably, overexpression of PAR-4 has been demonstrated to be associated with XIAP down-regulation in human apoptotic renal cells [Lee et al. 2008]. These findings indicate that PAR-4 might interfere with the regulation of IAP proteins during genotoxic drug-induced apoptosis in TNBC cells. To investigate this hypothesis I started to analyze the influence of PAR-4 on cIAP-1, cIAP-2 and XIAP. Observations from preliminary experiments demonstrate that these IAP proteins were downregulated upon genotoxic drug treatment in TNBC cells. I demonstrate that PAR-4 down-regulation prevents cIAP-1 depletion and renders TNBC cells insensitive to DNA-damage-induced apoptosis. Interestingly, treatment of TNBC cells with Smac mimetics sensitizes PAR-4-depleted cells to genotoxic drugs, indicating that PAR-4 has to deplete IAPs for apoptosis induction. I further observed a strong decrease of cIAP-1 on the protein level following overexpression of the C-terminal PAR- 4 cleavage fragment, which is located in the nucleus [Treude et al., manuscript in preparation]. Therefore, it is tempting to speculate that nuclear entry of the C-terminal PAR-4 cleavage fragment negatively regulates transcription of cIAP-1. Taken together, my observations indicate that PAR-4 is required for genotoxic drug- induced apoptosis in TNBC cells and interferes with Caspase-8 activation. The data suggest that loss of PAR-4 prevents cIAP-1 depletion and Caspase-8 activation following genotoxic drug treatment, indicating a molecular mechanism in TNBC cells to overcome DNA-damage-induced apoptosis. Loss of PAR-4 is associated with chemoresistance but in combination with genotoxic drug treatment Smac mimetics can overcome this chemoresistance mediated by PAR-4 depletion in TNBC cells. Therefore, low PAR-4 expression might be used as a biomarker for breast cancer patients as low PAR-4 expressing tumors should benefit from combinatorial treatment of Smac mimetics and chemotherapy.

4.5 Nuclear translocation of the C-terminal PAR-4 fragment is a key regulatory

event in DNA-damage-induced apoptosis

The cleavage of Caspase substrates can cause a loss, gain or even a change of function of the protein with very different effects on the cell. Caspase-mediated cleavage can further result in a change of the subcellular localization, as shown for many proteins involved in signaling, including kinases [Kurokawa and Kornbluth 2009]. Nuclear translocation of the SAC domain encompassing a NLS sequence has been 78 Discussion demonstrated to be an essential regulatory mechanism to deliver the pro-apoptotic and tumor suppressor activities of PAR-4 [El-Guendy et al. 2003, Zhao et al. 2007]. Caspase-8 cleaves PAR-4 at position D131, thereby separating the unstructured N- terminal region from the C-terminal part, which includes the NLS-containing SAC domain and the LZ. Therefore, the C-terminal fragment exhibits all the domains required for nuclear translocation and induction of apoptosis (Figure 5E). Indeed, PAR-4 cleavage markedly enhances nuclear targeting of the C-terminal fragment following UV- mediated or TNFα/CHX-induced apoptosis in HEK293 and in MCF-7 breast cancer cells (Figure 8B and 8D). Nuclear accumulation of the C-terminal cleavage fragment was also monitored in TNBC cells treated with Doxorubicin or Etoposide (Figure 12), but how this shuttling process is regulated is still unknown. One possible mechanism is provided by 14-3-3 proteins, which are capable to bind to phosphorylated S and T residues [Reinhardt and Yaffe 2013]. PAR-4 has been shown to associate with the mainly cytosolic 14-3-3σ isoform, which serves as a scaffolding protein and sequesters PAR-4 in the cytoplasm [Goswami et al. 2005; Kline and Irby 2011]. 14-3-3σ was demonstrated to interact with the transcriptional co-activator Yes-associated protein (YAP), thereby preventing it from nuclear translocation, which would result in p73- mediated BAX expression and apoptosis [Basu et al. 2003]. It can therefore be speculated that Caspase-8-activated cleavage of PAR-4 interferes with 14-3-3-mediated cytosolic retention of PAR-4, thereby inducing nuclear accumulation of the C-terminal cleavage product. Nuclear entry of the Caspase-8-mediated C-terminal PAR-4 cleavage fragment, comprising the SAC domain and the LZ marks an important regulatory mechanism to induce apoptosis. This is confirmed by the fact that overexpression of the C-terminal cleavage product of PAR-4, which is exclusively located to the nucleus, but not the N- terminal fragment is sufficient to induce apoptosis and renders BT-20 and MDA-MB-468 cells sensitive to genotoxic drug-induced apoptosis (Figure 13). PAR-4 is not documented to exhibit any enzymatic activities. It has been demonstrated that PAR-4 delivers its pro-apoptotic activities through binding of several regulatory proteins within the nuclear compartment, including the transcription factor WT1 and TOP I, thereby interfering with their cellular functions [Johnstone et al. 1996; Goswami et al. 2008]. Therefore, Caspase-8-mediated PAR-4 cleavage and nuclear translocation of the C- terminal fragment might provide a regulatory mechanism to allow PAR-4 to interact with nuclear proteins to deliver its pro-apoptotic functions. This idea is supported by my own

79 Discussion preliminary data, demonstrating that overexpression of the nuclear located C-terminal PAR-4 fragment leads to Caspase-8 activation accompanied with cIAP-1 depletion, suggesting a regulatory mechanism at transcriptional levels provided by PAR-4 [Treude et al., manuscript in preparation].

4.6 USP7 – a novel interaction partner of PAR-4

PAR-4 facilitates its pro-apoptotic properties through physical interaction with other proteins located in the cytoplasm or the nuclear compartment. In order to discover critical binding partners of the tumor suppressor, PAR-4 was immunoprecipitated from HeLa cells and subjected to unbiased mass spectrometry approaches. Interestingly, the analysis identified three out of nine proteins being ubiquitin-related factors, including the E3 ubiquitin ligases MYCBP2, UBR5 and the deubiquitinating enzyme USP7 (Figure 14A and 14B). USP7, which primarily localizes to the nucleus, is a critical regulator during the DNA damage response and mediates deubiquitination of many proteins, including the tumor suppressor proteins p53 and PTEN [Pfoh, Lacdao, and Saridakis 2015]. USP7 might target PAR-4 in vivo, thereby stabilizing the protein. Although the cellular binding of PAR-4 and USP7 was confirmed via reciprocal co-immunoprecipitation experiments (Figure 14C and 14D), the domains that are responsible for the interaction between PAR-4 and USP7 need further analysis and would give detailed insights into the molecular regulation of PAR-4 by USP7. Alternatively, PAR-4 might regulate USP7- mediated substrate recognition and modification. The idea is supported by the findings that PAR-4 cooperates with the tumor suppressor activities of PTEN in prostate carcinomas, as PTEN haploinsuffiency accompanied with PAR-4 loss leads to fully invasive prostate carcinomas in mice [Fernandez-Marcos et al. 2009; Diaz-Meco and Abu-Baker 2009]. Despite its commonly accepted role in the cytoplasm, cytosolic retention of PTEN is associated with tumor progression, which is regulated via ubiquitination and tightly controlled by a signaling network, including USP7 [Baker 2007; Song et al. 2008]. Based on these results, it is tempting to speculate that PAR-4 might be involved in USP7-mediated PTEN deubiquitination and trafficking to regulate PTEN during tumor development. Moreover, upon TNFα-induced apoptosis USP7 has been demonstrated to bind to TRIM27, a tripartite motif (TRIM) protein containing RING finger, B-box and coiled-coil domains. RIPK1 is then deubiquitinated by USP7 in a 80 Discussion

TRIM27-dependent manner, whereupon RIPK1 is stabilized and integrated in a complex, comprising FADD and pro-Caspase-8, resulting in Caspase-8 activation and induction of apoptosis [Zaman et al. 2013]. These results indicate that PAR-4 might interfere with the USP7/TRIM27 complex, explaining its capability to activate Caspase- 8, as discussed above. PAR-4 is down-regulated in various cancers, including primary and recurrent breast cancers [Zapata-Benavides et al. 2009; Méndez-López et al. 2010; Nagai et al. 2010; Alvarez et al. 2013]. Depending on the cell type PAR-4 depletion is provided by aberrant gene expression, including complete gene depletion, gene mutations, or promoter hyper-methylation [Kimura et al. 1998; Schneider et al. 2003; Pruitt et al. 2005; Moreno- Bueno et al. 2007]. The E3 ubiquitin ligase Tripartite motif containing 37 (TRIM37) is amplified in breast cancers and was shown to mono-ubiquitinate histone H2A, a chromatin modification linked to transcriptional repression. Interestingly, TRIM37 was found to be enriched at promoters of several putative tumor suppressors, including PAR-4, which was accompanied with increased levels of ubiquitinated H2A. As the authors expected, knockdown of TRIM37 in these breast cancer cells resulted in increased expression of PAR-4 and the other target genes resulting in decreased tumor growth in mouse xenografts [Bhatnagar et al. 2014]. Besides transcriptional regulation of the tumor suppressor, PAR-4 was demonstrated to interact with the E3 ubiquitin ligase FBXO45, which facilitates PAR-4 ubiquitination and proteasomal degradation in prostate cancer cells [Chen et al. 2014]. Interestingly, I identified MYCBP2, an E3 protein ligase, as a novel PAR-4 interaction partner (Figure 14B) and MYCBP2 has been demonstrated to specifically associate with FBXO45 [Saiga et al. 2009]. UBR5 belongs to the E3 ligases of the homologous to E6-AP carboxyl terminus (HECT) domain family. The UBR5 gene is located to chromosome 8q22, a region that is frequently amplified in breast cancers and linked with increased tumor recurrence [Hu et al. 2009; Y. Li et al. 2010]. Moreover, a whole-genome RNA interference screening approach identified that elevated expression of UBR5 in breast, pancreas and lung cancer cells mediates resistance to death receptor-induced apoptosis [Dompe et al. 2011]. Interestingly, a proteome wide, quantitative survey of in vivo ubiquitination sites revealed PAR-4 ubiquitination on K333 [Wagner et al. 2011]. Hence, the interaction of PAR-4 with the E3 ubiquitin ligases MYCBP2 and UBR5 might explain a regulatory mechanism to control PAR-4 protein levels in order to prevent its tumor suppressor

81 Discussion activities. In addition, PAR-4 might interfere with the molecular signaling of MYCBP2 or UBR5. Intriguingly, for UBR5 it was demonstrated by Goncharov and colleagues that the E3 ligase binds to cIAP-1, which is down-regulated upon overexpression of the C- terminal PAR-4 cleavage fragment and stabilized following PAR-4 depletion, as discussed previously [Goncharov et al. 2013]. Although the authors did not investigate the biological relevance of the interaction between UBR5 and cIAP-1, it is tempting to speculate that UBR5 might ubiquitinate cIAP-1 in a PAR-4-dependent manner, thereby destabilizing cIAP-1, which results in the induction of apoptosis due to Caspase-8 activation. The pro-apoptotic tumor suppressor protein PAR-4 is down-regulated in a specific subset of cancers, including lung cancer, prostate cancer and breast cancer. PAR-4 expression is low in residual breast cancers and PAR-4 down-regulation serves as a mechanism for cancer cell survival following chemotherapy. Strategies that increase PAR-4 expression in tumors could be beneficial for cancer therapy. Because PAR-4 is neither an enzyme nor a signaling receptor, it is not a classical drug target. Hence, identification of molecular pathways that control or modulate PAR-4 stability is an important step to treat primary and recurrent tumors with higher efficacy.

82 Summary and Outlook

5 Summary and Outlook

Breast cancer is the most common cause of cancer death for women worldwide and most deaths from breast cancer result from disease relapse upon treatment of the primary tumor. Because breast cancer recurrence is typically an incurable disease, understanding the mechanisms by which residual cancer cells survive treatment is necessary to improve therapies for breast cancer patients following disease relapse. Down-regulation of the tumor suppressor protein PAR-4 was shown to be a critical step for breast cancer recurrence and allows tumor cells to survive tumor regression following cytotoxic drug treatment The results of this study demonstrate that processing of PAR-4 as a bona fide Caspase- 8 substrate is required to induce apoptosis in breast cancer cells following stimulation with TNFα or DNA-damage. Furthermore, cleavage-induced nuclear translocation of the C-terminal part of PAR-4 by Caspase-8 is necessary to promote apoptosis upon chemotherapeutic treatment in breast cancer cells (Figure 15). Moreover, loss of PAR-4 mediates resistance to DNA-damage-induced apoptosis and also interferes with Caspase-8 activation in breast cancer cells. Several negative regulators of Caspase-8, including cIAP-1, are currently investigated as putative downstream targets of PAR-4 and might explain the capability of PAR-4 to stimulate Caspase-8 activation through a novel feedback mechanism. In this regard, I aim to identify nuclear targets of the C- terminal cleavage product, which mediate PAR-4’s pro-apoptotic function in breast cancer using mass spectrometry. Resistance to cell death has been proposed to be an essential hallmark of cancer cells for the development of malignant tissues and low PAR-4 expression promotes survival of breast cancer cells following therapy. However, the pathways controlling PAR-4 expression are not identified. Unbiased mass spectrometry analysis discovered three ubiquitin-related factors, including the E3 ligases MYCBP2 and UBR5 and the deubiquitinating enzyme USP7, suggesting that down-regulation of PAR-4 might be directly linked to aberrant ubiquitin signaling. Restoration of PAR-4 expression as a potential therapeutic intervention may hold significant clinical promise for relapsing breast cancer patients. To confirm this hypothesis, biochemical mechanisms that control PAR-4 are currently investigated and could be exploited for drug targeting approaches.

83 Summary and Outlook

TNFα

TNFR1

Plasma membrane TRADD

cFLIP

CHX Caspase-8 Doxorubicin/Etoposide (inactive)

* * Caspase-8 DNA* * (active) PAR-4

Nucleus PAR-4 Caspase-3 (inactive)

Caspase-3 Apoptosis (active)

Figure 15: PAR-4 signaling in breast cancer cells. Schematic illustration of the Caspase-8-dependent PAR-4 activation. Combinational treatment of TNFα and Cycloheximide (CHX) switches pro-inflammatory TNFα signaling into a death signal. CHX prevents the expression of cFLIP, an antagonist of the DISC complex, resulting in TNFα-induced Caspase-8 activation. Active Caspase-8 cleaves PAR-4 at position 131, whereupon the full length protein is divided into an N-terminal region and a C-terminal (C) part, comprising the functional important LZ domain and the SAC domain, encompassing the NLS sequence. Following processing, the latter translocates to the nucleus and induces apoptosis. Genotoxic drug treatment with Doxorubicin or Etoposide causes DNA- damage (*), resulting in Caspase-8 activation by an unknown mechanism (dotted arrow). Caspase-8 activation in turn leads to PAR-4 processing and nuclear entry of the C-terminal PAR-4 fragment. Translocation of the C-terminal PAR-4 fragment to the nuclear compartment promotes Caspase-8 activity and apoptotic cell death, indicating a positive feedback mechanism orchestrated by PAR-4 to sensitize cells to genotoxic stress stimuli.

84 References

6 References

Acehan, D. et al. 2002. “Three-Dimensional Structure of the Apoptosome: Implications for Assembly, Procaspase-9 Binding, and Activation.” Molecular Cell 9(2): 423–32.

Adams, J.M., and S. Cory 2010. “The Bcl-2 Apoptotic Switch in Cancer Development and Therapy.” Oncogene 26(9): 1324–37.

Ahmed, M. M. et al. 2008. “Downregulation of PAR-4, a pro-Apoptotic Gene, in Pancreatic Tumors Harboring K-Ras Mutation.” International Journal of Cancer 122(1): 63–70.

Alvarez, J. V. et al. 2013. “Par-4 Downregulation Promotes Breast Cancer Recurrence by Preventing Multinucleation Following Targeted Therapy.” Cancer Cell 24(1): 30– 44.

Ashkenazi, A., and V. M. Dixit 1998. “Death Receptors: Signaling and Modulation.” Science 281(5381): 1305–8.

Baker, S. J. 2007. “PTEN Enters the Nuclear Age.” Cell 128(1): 25–28.

Barradas, M. et al. 1999. “The Downregulation of the pro-Apoptotic Protein Par-4 Is Critical for Ras-Induced Survival and Tumor Progression.” The EMBO Journal 18(22): 6362–69.

Basu, S. et al. 2003. “Akt Phosphorylates the Yes-Associated Protein, YAP, to Induce Interaction with 14-3-3 and Attenuation of p73-Mediated Apoptosis.” Molecular Cell 11(1): 11–23.

Benedict, C. A., P. S. Norris, and C. F. Ware 2002. “To Kill or Be Killed: Viral Evasion of Apoptosis.” Nature Imunology 3(11): 1013–18.

Bertrand, M. J. M. et al. 2008. “cIAP1 and cIAP2 Facilitate Cancer Cell Survival by Functioning as E3 Ligases That Promote RIP1 Ubiquitination.” Molecular Cell 30(6): 689–700.

Bhatnagar, S. et al. 2014. “TRIM37 Is a New Histone H2A Ubiquitin Ligase and Breast Cancer Oncoprotein.” Nature 516(7529): 116–20.

Bieberich, E. et al. 2003. “Regulation of Cell Death in Mitotic Neural Progenitor Cells by Asymmetric Distribution of Prostate Apoptosis Response 4 (PAR-4) and Simultaneous Elevation of Endogenous Ceramide.” Journal of Cell Biology 162(3): 469–79.

Biton, S., and A. Ashkenazi 2011. “NEMO and RIP1 Control Cell Fate in Response to Extensive DNA Damage via TNF-α Feedforward Signaling.” Cell 145(1): 92–103.

85 References

Boehrer, S. et al. 2001. “Deregulated Expression of Prostate Apoptosis Response Gene-4 in Less Differentiated Lymphocytes and Inverse Expressional Patterns of Par-4 and Bcl-2 in Acute Lymphocytic Leukemia.” The Hematology Journal 2(2): 103–7.

Boghaert, R. et al. 1997. “Immunohistochemical Analysis of the Proapoptotic Protein Par-4 in Normal Rat Tissues.” Cell Growth and Differentiation 8(8): 881–90.

Boosen, M. et al. 2009. “Par-4 Is an Essential Downstream Target of DAP-like Kinase (Dlk) in Dlk/Par-4-Mediated Apoptosis.” Molecular Biology of the Cell 20(18): 4010– 20.

Bratton, S. B. et al. 2001. “Recruitment, Activation and Retention of Caspases-9 and-3 by Apaf-1 Apoptosome and Associated XIAP Complexes.” The EMBO Journal 20(5): 998–1009.

Bredesen, D. E. 2009. “Neurodegeneration in Alzheimer’s Disease: Caspases and Synaptic Element Interdependence.” Molecular Neurodegeneration 4: 27.

Brenner, D., H. Blaser, and T. W. Mak 2015. “Regulation of Tumour Necrosis Factor Signalling: Live or Let Die.” Nature Reviews - Immunology 15(6): 362–74.

Breslauer, K. J., R. Frank, H. Blöcker, and L. A. Marky 1986. “Predicting DNA Duplex Stability from the Base Sequence.” Proceedings of the National Academy of Sciences USA 83(11): 3746–50.

Brooks, M. D., M. L. Burness, and M. S. Wicha 2015. “Therapeutic Implications of Cellular Heterogeneity and Plasticity in Breast Cancer.” Cell - Stem Cell 17(3): 260–71.

Byun, Y. et al. 2001. “Caspase Cleavage of Vimentin Disrupts Intermediate Filaments and Promotes Apoptosis.” Cell Death and Differentiation 8(5): 443–50.

Cain, K. et al. 2000. “Apaf-1 Oligomerizes into Biologically Active Approximately 700- kDa and Inactive Approximately 1.4-MDa Apoptosome Complexes.” Journal of Biological Chemistry 275(20): 6067–70.

Candé, C. et al. 2002. “Apoptosis-Inducing Factor (AIF): A Novel Caspase-Independent Death Effector Released from Mitochondria.” Biochimie 84(2-3): 215–22.

Cantley, L. C., and B. G. Neel 1999. “New Insights into Tumor Suppression: PTEN Suppresses Tumor Formation by Restraining the Phosphoinositide 3-kinase/AKT Pathway.” Proceedings of the National Academy of Sciences USA 96(8): 4240–45.

Catz, S. D., and J. L. Johnson 2001. “Transcriptional Regulation of Bcl-2 by Nuclear Factor Kappa B and Its Significance in Prostate Cancer.” Oncogene 20(50): 7342– 51.

Chabner, B. A., and T. G. Roberts 2005. “Timeline: Chemotherapy and the War on Cancer.” Nature Reviews - Cancer 5(1): 65–72.

86 References

Chakraborty, M., S. G. Qiu, K. M. Vasudevan, and V. M. Rangnekar 2001. “Par-4 Drives Trafficking and Activation of Fas and FasL to Induce Prostate Cancer.” Cancer Research 61(19): 7255–63.

Chandra, D. et al. 2004. “Association of Active Caspase 8 with the Mitochondrial Membrane during Apoptosis: Potential Roles in Cleaving BAP31 and Caspase 3 and Mediating Mitochondrion-Endoplasmic Reticulum Cross Talk in Etoposide- Induced Cell Death.” Molecular and Cellular Biology 24(15): 6592–6607.

Chang, A. J., K. A. Autio, M. Roach, and H. I. Scher 2014. “High-Risk Prostate Cancer- Classification and Therapy.” Nature Reviews - Clinical Oncology 11(6): 308–23.

Chang, S., J. H. Kim, and J. Shin 2002. “p62 Forms a Ternary Complex with PKCζ and PAR-4 and Antagonizes PAR-4-Induced PKCζ Inhibition.” FEBS Letters 510(1-2): 57–61.

Chaudhry, P., M. Singh, S. Parent, and E. Asselin 2012. “Prostate Apoptosis Response 4 (Par-4), a Novel Substrate of Caspase-3 during Apoptosis Activation.” Molecular and Cellular Biology 32(4): 826–39.

Chauhan, D. et al. 2007. “A Novel Bcl-2/Bcl-X(L)/Bcl-W Inhibitor ABT-737 as Therapy in Multiple Myeloma.” Oncogene 26(16): 2374–80.

Cheema, S. K. et al. 2003. “Par-4 Transcriptionally Regulates Bcl-2 through a WT1- Binding Site on the Bcl-2 Promoter.” Journal of Biological Chemistry 278(22): 19995–5.

Chen, C., L. C. Edelstein, and C. Gélinas 2000. “The Rel/NF-kappaB Family Directly Activates Expression of the Apoptosis Inhibitor Bcl-XL.” Molecular and Cellular Biology 20(8): 2687–95.

Chen, X. et al. 2014. “Fbxo45-Mediated Degradation of the Tumor-Suppressor Par-4 Regulates Cancer Cell Survival.” Cell Death and Differentiation 21(10): 1535–45.

Chinnaiyan, A. M., K. O’Rourke, M. Tewari, and V. M. Dixit 1995. “FADD, a Novel Death Domain-Containing Protein, Interacts with the Death Domain of Fas and Initiates Apoptosis.” Cell 81(4): 505–12.

Choi, Y. E. et al. 2009. “The E3 Ubiquitin Ligase cIAP1 Binds and Ubiquitinates Caspase-3 and -7 via Unique Mechanisms at Distinct Steps in Their Processing.” Journal of Biological Chemistry 284(19): 12772–82.

Clark, O., T. E. A. Botrel, L. Paladini, and M. B. A. Ferreira 2014. “Targeted Therapy in Triple-Negative Metastatic Breast Cancer: A Systematic Review and Meta- Analysis.” Core Evidence 9: 1–11.

Clarke, P., and K. L. Tyler 2009. “Apoptosis in Animal Models of Virus-Induced Disease.” Nature Reviews - Microbiology 7(2): 144–55.

Communal, C. et al. 2002. “Functional Consequences of Caspase Activation in Cardiac Myocytes.” Proceedings of the National Academy of Sciences USA 99(9): 6252–56.

87 References

Cook, J et al. 1999. “Decreased Expression of the pro-Apoptotic Protein Par-4 in Renal Cell Carcinoma.” Oncogene 18(5): 1205–8.

Cowling, V., and J. Downward 2002. “Caspase-6 Is the Direct Activator of Caspase-8 in the Cytochrome c-Induced Apoptosis Pathway: Absolute Requirement for Removal of Caspase-6 Prodomain.” Cell Death and Differentiation 9(10): 1046–56.

Cox, J. et al. 2011. “Andromeda: A Peptide Search Engine Integrated into the MaxQuant Environment.” Journal of Proteome Research 10(4): 1794–1805.

Cox, J., and M. Mann 2008. “MaxQuant Enables High Peptide Identification Rates, Individualized P.p.b.-Range Mass Accuracies and Proteome-Wide Protein Quantification.” Nature Biotechnology 26(12): 1367–72.

Crawford, E. D., and J. A. Wells 2011. “Caspase Substrates and Cellular Remodeling.” Annual Review of Biochemistry 80: 1055–87.

Delbridge, A. R. D., and A. Strasser 2015. “The BCL-2 Protein Family, BH3-Mimetics and Cancer Therapy.” Cell Death and Differentiation 22(7): 1071–80.

Denecker, G., P. Ovaere, P. Vandenabeele, and W. Declercq 2008. “Caspase-14 Reveals Its Secrets.” Journal of Cell Biology 180(3): 451–58.

Dent, R. et al. 2007. “Triple-Negative Breast Cancer: Clinical Features and Patterns of Recurrence.” Clinical Cancer Research 13(15): 4429–34.

Diaz-Meco, M .T., and S. Abu-Baker 2009. “The Par-4/PTEN Connection in Tumor Suppression.” Cell Cycle 8(16): 2518–22.

Diaz-Meco, M. T. et al. 1996. “The Product of Par-4, a Gene Induced during Apoptosis, Interacts Selectively with the Atypical Isoforms of Protein Kinase C.” Cell 86(5): 777–86.

Diaz-Meco, M. T. et al. 1999. “Inactivation of the Inhibitory κB Protein kinase/Nuclear Factor κB Pathway by Par-4 Expression Potentiates Tumor Necrosis Factor α- Induced Apoptosis.” Journal of Biological Chemistry 274(28): 19606–12.

Dietze, E. C. et al. 2015. “Triple-Negative Breast Cancer in African-American Women: Disparities versus Biology.” Nature Reviews - Cancer 15(4): 248–54.

Dompe, N. et al. 2011. “A Whole-Genome RNAi Screen Identifies an 8q22 Gene Cluster That Inhibits Death Receptor-Mediated Apoptosis.” Proceedings of the National Academy of Sciences USA 108(43): 943–51.

Du, C. et al. 2000. “Smac, a Mitochondrial Protein That Promotes Cytochrome c- Dependent Caspase Activation by Eliminating IAP Inhibition.” Cell 102(1): 33–42.

Eckelman, B. P., and G. S. Salvesen 2006. “The Human Anti-Apoptotic Proteins cIAP1 and cIAP2 Bind but Do Not Inhibit Caspases.” Journal of Biological Chemistry 281(6): 3254–60.

88 References

Eckelman, B. P., G. S. Salvesen, and F. L. Scott 2006. “Human Inhibitor of Apoptosis Proteins: Why XIAP Is the Black Sheep of the Family.” EMBO Reports 7(10): 988– 94.

Eckhart, L. et al. 2008. “Identification of Novel Mammalian Caspases Reveals an Important Role of Gene Loss in Shaping the Human Caspase Repertoire.” Molecular Biology and Evolution 25(5): 831–41.

El-Guendy, N., and V. M. Rangnekar 2003. “Apoptosis by Par-4 in Cancer and Neurodegenerative Diseases.” Experimental Cell Research 283(1): 51–66.

El-Guendy, N., Y. Zhao, S. Gurumurthy, and R. Burikhanov 2003. “Identification of a Unique Core Domain of Par-4 Sufficient for Selective Apoptosis Induction in Cancer Cells.” Molecular and Cellular Biology 23(16): 5516–25.

Elmore, S. 2007. “Apoptosis: A Review of Programmed Cell Death.” Toxicologic Pathology 35(4): 495–516.

Erener, S. et al. 2012. “Inflammasome-Activated Caspase-7 Cleaves Cleaves PARP1 to Enhance the Expression of a Subset of NF-κB Target Genes.” Molecular Cell 46(2): 200–211.

Eskes, R., S. Desagher, B. Antonsson, and J. C. Martinou 2000. “Bid Induces the Oligomerization and Insertion of Bax into the Outer Mitochondrial Membrane.” Molecular and Cellular Biology 20(3): 929–35.

Ethell, D. W., and L. A. Buhler 2003. “Fas Ligand-Mediated Apoptosis in Degenerative Disorders of the Brain.” Journal of Clinical Immunology 23(6): 439–46.

Fadeel, B., and S. Orrenius 2005. “Apoptosis: A Basic Biological Phenomenon with Wide-Ranging Implications in Human Disease.” Journal of Internal Medicine 258(6): 479–517.

Fan, T. J., L. H. Han, R. S. Cong, and J. Liang 2005. “Caspase Family Proteases and Apoptosis.” Acta Biochimica et Biophysica Sinica 37(11): 719–27.

Feoktistova, M. et al. 2011. “CIAPs Block Ripoptosome Formation, a RIP1/Caspase-8 Containing Intracellular Cell Death Complex Differentially Regulated by cFLIP Isoforms.” Molecular Cell 43(3): 449–63.

Fernandez-Marcos, P. J. et al. 2009. “Simultaneous Inactivation of Par-4 and PTEN in Vivo Leads to Synergistic NF-κB Activation and Invasive Prostate Carcinoma.” Proceedings of the National Academy of Sciences USA 106(31): 12962–67.

Filmus, J., M. N. Pollak, R. Cailleau, and R. N. Buick 1985. “MDA-468, a Human Breast Cancer Cell Line with a High Number of Epidermal Growth Factor (EGF) Receptors, Has an Amplified EGF Receptor Gene and Is Growth Inhibited by EGF.” Biochemical and Biophysical Research Communications 128(2): 898–905.

89 References

Fischer, U., R. U. Jänicke, and K. Schulze-Osthoff 2003. “Many Cuts to Ruin: A Comprehensive Update of Caspase Substrates.” Cell Death and Differentiation 10(1): 76–100.

Fischer, U., C. Stroh, and K. Schulze-Osthoff 2006. “Unique and Overlapping Substrate Specificities of Caspase-8 and Caspase-10.” Oncogene 25(1): 152–59.

Franchitto, A. et al. 2010. “Prostate Apoptosis Response-4 Is Expressed in Normal Cholangiocytes, Is down-Regulated in Human Cholangiocarcinoma, and Promotes Apoptosis of Neoplastic Cholangiocytes When Induced Pharmacologically.” The American Journal of Pathology 177(4): 1779–90.

Fuentes-Prior, P., and G. S. Salvesen 2004. “The Protein Structures That Shape Caspase Activity, Specificity, Activation and Inhibition.” The Biochemical Journal 384(Pt 2): 201–32.

Fukuda, S., and L. M. Pelus 2006. “Survivin, a Cancer Target with an Emerging Role in Normal Adult Tissues.” Molecular Cancer Therapeutics 5(5): 1087–98.

Fulda, S., and K. M. Debatin 2006. “Extrinsic versus Intrinsic Apoptosis Pathways in Anticancer Chemotherapy.” Oncogene 25(34): 4798–4811.

Fulda, S., W. Wick, M. Weller, and K. M. Debatin 2002. “Smac Agonists Sensitize for Apo2L/TRAIL- or Anticancer Drug-Induced Apoptosis and Induce Regression of Malignant Glioma in Vivo.” Nature Medicine 8(8): 808–15.

Del Gaizo Moore, V. et al. 2008. “BCL-2 Dependence and ABT-737 Sensitivity in Acute Lymphoblastic Leukemia BCL-2 Dependence and ABT-737 Sensitivity in Acute Lymphoblastic Leukemia.” Leukemia 111(4): 2300–2309.

Gao, S. et al. 2006. “Androgen Receptor and Prostate Apoptosis Response Factor-4 Target the c-FLIP Gene to Determine Survival and Apoptosis in the Prostate Gland.” Journal of Molecular Endocrinology 36(3): 463–83.

Garcia-Calvo, M. et al. 1999. “Purification and Catalytic Properties of Human Caspase Family Members.” Cell Death and Differentiation 6(4): 362–69.

Garcia-Cao, I. et al. 2003. “Genetic Inactivation of Par4 Results in Hyperactivation of NF-KB and Impairment of JNK and p38.” EMBO Reports 4(3): 307–12.

García-Cao, I. et al. 2005. “Tumour-Suppression Activity of the Proapoptotic Regulator Par4.” EMBO Reports 6(6): 577–83.

Gervais, F. G. et al. 1999. “Involvement of Caspases in Proteolytic Cleavage of Alzheimer’s Amyloid-ß Precursor Protein and Amyloidogenic Aß Peptide Formation.” Cell 97(3): 395–406.

Goncharov, T. et al. 2013. “OTUB1 Modulates c-IAP1 Stability to Regulate Signalling Pathways.” The EMBO Journal 32(8): 1103–14.

90 References

Goswami, A. et al. 2005. “Binding and Phosphorylation of Par-4 by Akt Is Essential for Cancer Cell Survival.” Molecular Cell 20(1): 33–44.

Goswami, A. et al. 2008. “Par-4 Binds to Topoisomerase 1 and Attenuates Its DNA Relaxation Activity.” Cancer Research 68(15): 6190–99.

Green, D. R., and G. Kroemer 2004. “The Pathophysiology of Mitochondrial Cell Death.” Science 305(5684): 626–29.

Greenhalgh, D. G. 1998. “The Role of Apoptosis in Wound Healing.” The International Journal of Biochemistry & Cell Biology 30(9): 1019–30.

Gurumurthy, S., A. Goswami, K. M. Vasudevan, and V. M. Rangnekar 2005. “Phosphorylation of Par-4 by Protein Kinase A Is Critical for Apoptosis.” Molecular and Cellular Biology 25(3): 1146–61.

Gyrd-Hansen, M., and P. Meier 2010. “IAPs: From Caspase Inhibitors to Modulators of NF-κB, Inflammation and Cancer.” Nature Reviews - Cancer 10(8): 561–74.

Hanahan, D., and R. A. Weinberg 2000. “The Hallmarks of Cancer.” Cell 100(7): 57–70.

Hanahan, D., and R. A. Weinberg 2011. “Hallmarks of Cancer: The next Generation.” Cell 144(5): 646–74.

Hebbar, N., C. Wang, and V. M. Rangnekar 2012. “Mechanisms of Apoptosis by the Tumor Suppressor Par-4.” Journal of Cellular Physiology 227(12): 3715–21.

Holliday, D. L., and V. Speirs 2011. “Choosing the Right Cell Line for Breast Cancer Research.” Breast Cancer Research 13(4): 215.

Horbinski, C. 2013. “What Do We Know about IDH1/2 Mutations so Far, and How Do We Use It?” Acta Neuropathologica 125(5): 621–36.

Hsu, H., J. Xiong, and D. V. Goeddel 1995. “The TNF Receptor 1-Associated Protein TRADD Signals Cell Death and NF-κB Activation.” Cell 81(4): 495–504.

Hu, G. et al. 2009. “MTDH Activation by 8q22 Genomic Gain Promotes Chemoresistance and Metastasis of Poor-Prognosis Breast Cancer.” Cancer Cell 15(1): 9–20.

Hu, S., and X. Yang 2003. “Cellular Inhibitor of Apoptosis 1 and 2 Are Ubiquitin Ligases for the Apoptosis Inducer Smac/DIABLO.” Journal of Biological Chemistry 278(12): 10055–60.

Huang, Y., R. L. Rich, D. G. Myszka, and H. Wu 2003. “Requirement of Both the Second and Third BIR Domains for the Relief of X-Linked Inhibitor of Apoptosis Protein (XIAP)-Mediated Caspase Inhibition by Smac.” The Journal of Biological Chemistry 278(49): 49517–22.

Igney, F. H., and P. H. Krammer 2002. “Death and Anti-Death: Tumour Resistance to Apoptosis.” Nature Reviews - Cancer 2(4): 277–88.

91 References

Isakoff, S. J. 2010. “Triple Negative Breast Cancer: Role of Specific Chemotherapy Agents.” Cancer Journal 16(1): 53–61.

Jänicke, R. U. 2009. “MCF-7 Breast Carcinoma Cells Do Not Express Caspase-3.” Breast Cancer Research and Treatment 117(1): 219–21.

Johnson, C. E., and S. Kornbluth 2008. “Caspase Cleavage Is Not for Everyone.” Cell 134(5): 720–21.

Johnstone, R. W. et al. 1996. “A Novel Repressor, Par-4, Modulates Transcription and Growth Suppression Functions of the Wilms’ Tumor Suppressor WT1.” Molecular and Cellular Biology 16(12): 6945–56.

Johnstone, R. W. et al. 1998. “Mapping of the Human PAWR (par-4) Gene to Chromosome 12q21.” Genomics 53(2): 241–43.

Joshi, J. et al. 2008. “Par-4 Inhibits Akt and Suppresses Ras-Induced Lung Tumorigenesis.” The EMBO Journal 27(16): 2181–93.

Kang, M. H. et al. 2007. “Activity of Vincristine, L-ASP, and Dexamethasone against Acute Lymphoblastic Leukemia Is Enhanced by the BH3-Mimetic ABT-737 in Vitro and in Vivo.” Blood 110(6): 2057–66.

Kang, N. et al. 2010. “Inhibition of EGFR Signaling Augments Oridonin-Induced Apoptosis in Human Laryngeal Cancer Cells via Enhancing Oxidative Stress Coincident with Activation of Both the Intrinsic and Extrinsic Apoptotic Pathways.” Cancer Letters 294(2): 147–58.

Kaufmann, S. H., and D. L. Vaux 2003. “Alterations in the Apoptotic Machinery and Their Potential Role in Anticancer Drug Resistance.” Oncogene 22(47): 7414–30.

Kawai, T., S. Akira, and J. C. Reed 2003. “ZIP Kinase Triggers Apoptosis from Nuclear PML Oncogenic Domains ZIP Kinase Triggers Apoptosis from Nuclear PML Oncogenic Domains.” Molecular and Cellular Biology 23(17): 6174–86.

Kerr, J. F. R., A. H. Wyllie, and A. R. Currie 1972. “Apoptosis: A Basic Biological Phenomenon with Wide-Ranging Implications in Human Disease.” British Journal of Cancer 26(4): 239–57.

Kibbe, W. A. 2007. “OligoCalc: An Online Oligonucleotide Properties Calculator.” Nucleic Acids Research 35(Web Server issue): 43–46.

Kimura, M. et al. 1998. “Identification of Two Common Regions of Allelic Loss in Chromosome Arm 12q in Human Pancreatic Cancer.” Cancer Research 58(11): 2456–60.

Kitada, S., I. M. Pedersen, A. D. Schimmer, and J. C. Reed 2002. “Dysregulation of Apoptosis Genes in Hematopoietic Malignancies.” Oncogene 21(21): 3459–74.

92 References

Kline, C. L. B., and R. B. Irby 2011. “The pro-Apoptotic Protein Prostate Apoptosis Response Protein-4 (Par-4) Can Be Activated in Colon Cancer Cells by Treatment with Src Inhibitor and 5-FU.” Apoptosis 16(12): 1285–94.

Kline, C. L., S. S. Shanmugavelandy, M. Kester, and R. B. Irby 2009. “Delivery of PAR- 4 Plasmid in Vivo via Nanoliposomes Sensitizes Colon Tumor Cells Subcutaneously Implanted into Nude Mice to 5-FU.” Cancer Biology and Therapy 8(19): 1831–37.

Kocab, A. J., and C. S. Duckett 2016. “Inhibitor of Apoptosis Proteins as Intracellular Signaling Intermediates.” FEBS Journal 283(2): 221–31.

Kögel, D. et al. 2001. “Dlk/ZIP Kinase-Induced Apoptosis in Human Medulloblastoma Cells: Requirement of the Mitochondrial Apoptosis Pathway.” British Journal of Cancer 85(11): 1801–8.

Kojima, K. et al. 2006. “Concomitant Inhibition of MDM2 and Bcl-2 Protein Function Synergistically Induce Mitochondrial Apoptosis in AML.” Cell Cycle 5(23): 2778–86.

Konopleva, M. et al. 2006. “Mechanisms of Apoptosis Sensitivity and Resistance to the BH3 Mimetic ABT-737 in Acute Myeloid Leukemia.” Cancer Cell 10(5): 375–88.

Krippner-Heidenreich, A. et al. 2001. “Targeting of the Transcription Factor Max during Apoptosis: Phosphorylation-Regulated Cleavage by Caspase-5 at an Unusual Glutamic Acid Residue in Position P1.” The Biochemical Journal 358(Pt 3): 705–15.

Kroemer, G. et al. 2009. “Classification of Cell Death: Recommidations of the Nomenclature Committee on Cell Death 2009.” Cell Death and Differentiation 16(1): 3–11.

Kroemer, G., and S. J. Martin 2005. “Caspase-Independent Cell Death.” Nature Medicine 11(7): 725–30.

Krueger, A. et al. 2001. “Cellular FLICE-Inhibitory Protein Splice Variants Inhibit Different Steps of Caspase-8 Activation at the CD95 Death-Inducing Signaling Complex.” Journal of Biological Chemistry 276(23): 20633–40.

Kukoc-Zivojnov, N. et al. 2004. “Prostate Apoptosis Response Gene-4 (par-4) Abrogates the Survival Function of p185BCR-ABL in Hematopoietic Cells.” Experimental Hematology 32(7): 649–56.

Kulms, D., and T. Schwarz 2000. “Molecular Mechanisms of UV-Induced Apoptosis.” Photodermatology, Photoimmunology & Photomedicine 16(5): 195–201.

Kurokawa, M., and S. Kornbluth 2009. “Caspases and Kinases in a Death Grip.” Cell 138(5): 838–54.

Kuwana, T. et al. 2005. “BH3 Domains of BH3-Only Proteins Differentially Regulate Bax-Mediated Mitochondrial Membrane Permeabilization Both Directly and Indirectly.” Molecular Cell 17(4): 525–35.

93 References

Lamkanfi, M., and V. M. Dixit 2010. “Manipulation of Host Cell Death Pathways during Microbial Infections.” Cell Host and Microbe 8(1): 44–54.

Lamkanfi, M., M. Kalai, and P. Vandenabeele 2003. “Caspase-12: An Overview.” Cell Death and Differentiation 11(4): 365–68.

Lane, D. P. 1992. “p53, Guardian of the Genome.” Nature 358(6381): 15–16.

Lavrik, I. N., A. Golks, and P. H. Krammer 2005. “Review Series Caspases: Pharmacological Manipulation of Cell Death.” The Journal of Clinical Investigation 115(10): 2665–72.

Lazebnik, Y. A. et al. 1994. “Cleavage of poly(ADP-Ribose) Polymerase by a Proteinase with Properties like ICE.” Nature 371(6495): 346–47.

Lebeau, J., and G. Goubin 1987. “Amplification of the Epidermal Growth Factor Receptor Gene in the BT20 Breast Carcinoma Cell Line.” International Journal of Cancer 40(2): 189–91.

Lee, C. et al. 2013. “Molecular Mechanisms of UV-Induced Apoptosis and Its Effects on Skin Residential Cells : The Implication in UV-Based Phototherapy.” International Journal of Molecular Sciences 14(3): 6414–35.

Lee, J. H., and A. Nan 2012. “Combination Drug Delivery Approaches in Metastatic Breast Cancer.” Journal of Drug Delivery 2012: 915375.

Lee, M. J. et al. 2012. “Sequential Application of Anticancer Drugs Enhances Cell Death by Rewiring Apoptotic Signaling Networks.” Cell 149(4): 780–94.

Lee, T. et al. 2008. “Overexpression of Par-4 Enhances Thapsigargin-Induced Apoptosis via down-Regulation of XIAP and Inactivation of Akt in Human Renal Cancer Cells.” Journal of Cellular Biochemistry 103(2): 358–68.

Li, J., and J. Yuan 2008. “Caspases in Apoptosis and Beyond.” Oncogene 27(48): 6194–6206.

Li, L. Y., X. Luo, and X. Wang 2001. “Endonuclease G Is an Apoptotic DNase When Released from Mitochondria.” Nature 412(6842): 95–99.

Li, Y. et al. 2010. “Amplification of LAPTM4B and YWHAZ Contributes to Chemotherapy Resistance and Recurrence of Breast Cancer.” Nature Medicine 16(2): 214–18.

Liedtke, C. et al. 2008. “Response to Neoadjuvant Therapy and Long-Term Survival in Patients with Triple-Negative Breast Cancer.” Journal of Clinical Oncology 26(8): 1275–81.

Liu, X., H. Zou, C. Slaughter, and X. Wang 1997. “DFF, a Heterodimeric Protein That Functions Downstream of Caspase-3 to Trigger DNA Fragmentation during Apoptosis.” Cell 89(2): 175–84.

94 References

Liu, Y. et al. 2014. “The Tumor Suppressor Prostate Apoptosis Response-4 (Par-4) Is Regulated by Mutant IDH1 and Kills Glioma Stem Cells.” Acta Neuropathologica 128(5): 723–32.

Livraghi, L., and J. E. Garber 2015. “PARP Inhibitors in the Management of Breast Cancer: Current Data and Future Prospects.” BMC Medicine 13(1): 188.

Long, J. S., and K. M. Ryan 2012. “New Frontiers in Promoting Tumour Cell Death: Targeting Apoptosis, Necroptosis and Autophagy.” Oncogene 31(49): 5045–60.

Loo, D. T. et al. 1993. “Apoptosis Is Induced by ß-Amyloid in Cultured Central Nervous System Neurons.” Proceedings of the National Academy of Sciences USA 90(17): 7951–55.

Lowe, S. W., and A. W. Lin 2000. “Apoptosis in Cancer.” Carcinogenesis 21(3): 485–95.

Lu, D. C. et al. 2000. “A Second Cytotoxic Proteolytic Peptide Derived from Amyloid ß- Protein Precursor.” Nature Medicine 6(4): 397–404.

Lucas, T. et al. 2001. “Differential Expression Levels of Par-4 in Melanoma.” Melanoma Research 11(4): 379–83.

Lüthi, A. U., and S. J. Martin 2007. “The CASBAH: A Searchable Database of Caspase Substrates.” Cell Death and Differentiation 14(4): 641–50.

Maecker, H. L., C. Koumenis, and A. J. Giaccia 2000. “p53 Promotes Selection for Fas- Mediated Apoptotic Resistance.” Cancer Research 60(16): 4638–44.

Marte, B. M., and J. Downward 1997. “PKB/Akt: Connecting Phosphoinositide 3-Kinase to Cell Survival and beyond.” Trends in Biochemical Sciences 22(9): 355–58.

Martin, S. J., and D. R. Green 1995. “Protease Activation during Apoptosis: Death by a Thousand Cuts?” Cell 82(3): 349–52.

Mashima, T., M. Naito, and T. Tsuruo 1999. “Caspase-Mediated Cleavage of Cytoskeletal Actin Plays a Positive Role in the Process of Morphological Apoptosis.” Oncogene 18(15): 2423–30.

Mattson, M. P., W. Duan, S. L. Chan, and S. Camandola 1999. “Par-4 - An Emerging Pivotal Player in Neuronal Apoptosis and Neurodegenerative Disorders.” Journal of Molecular Neuroscience 13(1-2): 17–30.

Méndez-López, L. F. et al. 2010. “Immunohistochemical Analysis of Prostate Apoptosis Response-4 (Par-4) in Mexican Women with Breast Cancer: A Preliminary Study.” Archives of Medical Research 41(4): 261–68.

Micheau, O. et al. 2001. “NF-κB Signals Induce the Expression of c-FLIP.” Molecular and Cellular Biology 21(16): 5299–5305.

Mikolajczyk, J. et al. 2004. “Activation and Substrate Specificity of Caspase-14.” Biochemistry 43(32): 10560–69.

95 References

Miura, M. et al. 1993. “Induction of Apoptosis in Fibroblasts by IL-1ß-Converting Enzyme, a Mammalian Homolog of the C. Elegans Cell Death Gene Ced-3.” Cell 75(4): 653–60.

Moreno-Aspitia, A., and E. A. Perez 2009. “Treatment Options for Breast Cancer Resistant to Anthracycline and Taxane.” Mayo Clinic Proceedings 84(6): 533–45.

Moreno-Bueno, G. et al. 2007. “Inactivation of the Candidate Tumor Suppressor Par-4 in Endometrial Cancer.” Cancer Research 67(5): 1927–34.

Morgillo, F. et al. 2011. “Antitumor Activity of Bortezomib in Human Cancer Cells with Acquired Resistance to Anti-Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitors.” Lung Cancer 71(3): 283–90.

Nagai, M. A. et al. 2010. “Down-Regulation of the Candidate Tumor Suppressor Gene PAR-4 Is Associated with Poor Prognosis in Breast Cancer.” International Journal of Oncology 37(1): 41–49.

Nakano, K., and K. H. Vousden 2001. “PUMA, a Novel Proapoptotic Gene, Is Induced by p53.” Molecular Cell 7(3): 683–94.

Nalca, A. et al. 1999. “Oncogenic Ras Sensitizes Cells to Apoptosis by Par-4.” Journal of Biological Chemistry 274(42): 29976–83.

Nijhawan, D., N. Honarpour, and X. Wang 2000. “Aoptosis in Neural Development and Disease.” Proteins 23: 73–87.

Nitiss, J. L. 2009. “Targeting DNA Topoisomerase II in Cancer Chemotherapy.” Nature Reviews - Cancer 9(5): 338–50.

Normanno, N. et al. 2006. “Epidermal Growth Factor Receptor (EGFR) Signaling in Cancer.” Gene 366(1): 2–16.

Oberst, A., and D. R. Green 2011. “It Cuts Both Ways: Reconciling the Dual Roles of Caspase 8 in Cell Death and Survival.” Nature Reviews - Molecular Cell Biology 12(11): 757–63.

Oda, E. et al. 2000. “Noxa, a BH3-Only Member of the Bcl-2 Family and Candidate Mediator of p53-Induced Apoptosis.” Science 288(5468): 1053–58.

Oltersdorf, T. et al. 2005. “An Inhibitor of Bcl-2 Family Proteins Induces Regression of Solid Tumours.” Nature 435(7042): 677–81.

Opferman, J. T., and S. J. Korsmeyer 2003. “Apoptosis in the Development and Maintenance of the Immune System.” Nature Immunology 4(5): 410–15.

Owen-Schaub, L. B. et al. 1995. “Wild-Type Human p53 and a Temperature-Sensitive Mutant Induce Fas/APO-1 Expression.” Molecular and Cellular Biology 15(6): 3032–40.

96 References

Page, G., D. Ko, V. M. Rangnekar, and K. H. Scheidtmann 1999. “Interaction Partners of Dlk/ZIP Kinase : Co-Expression of Dlk/ZIP Kinase and Par-4 Results in Cytoplasmic Retention and Apoptosis.” Oncogene 18(51): 7265–73.

Pereira, W. O., and G. P. Amarante-Mendes 2011. “Apoptosis: A Programme of Cell Death or Cell Disposal?” Scandinavian Journal of Immunology 73(5): 401–7.

Perou, C. M. et al. 1999. “Distinctive Gene Expression Patterns in Human Mammary Epithelial Cells and Breast Cancers.” Proceedings of the National Academy of Sciences USA 96(16): 9212–17.

Perou, C. M. et al. 2000. “Molecular Portraits of Human Breast Tumours.” Nature 406(6797): 747–52.

Pfoh, R., I. K. Lacdao, and Vi. Saridakis 2015. “Deubiquitinases and the New Therapeutic Opportunities Offered to Cancer.” Endocrine-Related Cancer 22(1): T35–54.

Pommier, Y., E. Leo, H. Zhang, and C. Marchand 2010. “DNA Topoisomerases and Their Poisoning by Anticancer and Antibacterial Drugs.” Chemistry and Biology 17(5): 421–33.

Pop, C., and G. S. Salvesen 2009. “Human Caspases: Activation, Specificity, and Regulation.” Journal of Biological Chemistry 284(33): 21777–81.

Prat, A., and J. Baselga 2008. “The Role of Hormonal Therapy in the Management of Hormonal-Receptor-Positive Breast Cancer with Co-Expression of HER2.” Nature Clinical Practice - Oncology 5(9): 531–42.

Pruitt, K. et al. 2005. “Ras-Mediated Loss of the pro-Apoptotic Response Protein Par-4 Is Mediated by DNA Hypermethylation through Raf-Independent and Raf- Dependent Signaling Cascades in Epithelial Cells.” Journal of Biological Chemistry 280(24): 23363–70.

Qiu, G. et al. 1999. “Mutually Exclusive Expression Patterns of Bcl-2 and Par-4 in Human Prostate Tumors Consistent with down-Regulation of Bcl-2 by Par-4.” Oncogene 18(3): 623–31.

Reinhardt, H. C., and B. Schumacher 2012. “The p53 Network: Cellular and Systemic DNA Damage Responses in Aging and Cancer.” Trends in Genetics 28(3): 128–36.

Reinhardt, H. C., and M. B. Yaffe 2013. “Phospho-Ser/Thr-Binding Domains: Navigating the Cell Cycle and DNA Damage Response.” Nature Reviews - Molecular Cell Biology 14(9): 563–80.

Riedl, S. J., and G. S. Salvesen 2007. “The Apoptosome: Signalling Platform of Cell Death.” Nature Reviews - Molecular Cell Biology 8(5): 405–13.

Roussigne, M. et al. 2003. “THAP1 Is a Nuclear Proapoptotic Factor That Links Prostate-Apoptosis-Response-4 (Par-4) to PML Nuclear Bodies.” Oncogene 22(16): 2432–42.

97 References

Saelens, X. et al. 2004. “Toxic Proteins Released from Mitochondria in Cell Death.” Oncogene 23(16): 2861–74.

Safa, A. R. 2012. “C-FLIP, a Master Anti-Apoptotic Regulator.” Experimental Oncology 34(3): 176–84.

Safa, A. R., and K. E. Pollok 2011. “Targeting the Anti-Apoptotic Protein c-FLIP for Cancer Therapy.” Cancers 3(2): 1639–71.

Saiga, T. et al. 2009. “Fbxo45 Forms a Novel Ubiquitin Ligase Complex and Is Required for Neuronal Development.” Molecular and Cellular Biology 29(13): 3529–43.

Sakahira, H., M. Enari, and S. Nagata 1998. “Cleavage of CAD Inhibitor in CAD Activation and DNA Degradation during Apoptosis.” Nature 391(6662): 96–99.

Salvesen, G. S., and V. M. Dixit 1999. “Caspase Activation: The Induced-Proximity Model.” Proceedings of the National Academy of Sciences USA 96(20): 10964–67.

Salvesen, G. S., and C. S. Duckett 2002. “Iap Proteins : Blocking the Road to Death ’ S Door.” Nature Reviews - Molecular Cell Biology 3(6): 1–10.

Sax, J. K. et al. 2002. “BID Regulation by p53 Contributes to Chemosensitivity.” Nature Cell Biology 4(11): 842–49.

Scaffidi, C. et al. 1998. “Two CD95 (APO-1/Fas) Signaling Pathways.” The EMBO Journal 17(6): 1675–87.

Schechter, I., and A. Berger 1967. “On the Size of the Active Site in Proteases. I., Papain.” Biochemical and Biophysical Research Communications 27(2): 157–62.

Schneider, B. G. et al. 2003. “Regions of Allelic Imbalance in the Distal Portion of Chromosome 12q in Gastric Cancer.” Molecular Pathology 56(3): 141–49.

Selliah, N., and T. H. Finkel 2001. “Biochemical Mechanisms of HIV Induced T Cell Apoptosis.” Cell Death and Differentiation 8(2): 127–36.

Sells, S. F. et al. 1994. “Commonality of the Gene Programs Induced and by Effectors of Apoptosis in Androgen-Dependent Prostate Cells.” Cell Growth and Differentiation 5(4): 457–66.

Shen, J. et al. 2010. “Caspase-1 Recognizes Extended Cleavage Sites in Its Natural Substrates.” Atherosclerosis 210(2): 422–29.

Shoemaker, A. R. et al. 2006. “A Small-Molecule Inhibitor of Bcl-XL Potentiates the Activity of Cytotoxic Drugs in Vitro and in Vivo.” Cancer Research 66(17): 8731–39.

Shrestha-Bhattarai, T., and V. M. Rangnekar 2010. “Cancer-Selective Apoptotic Effects of Extracellular and Intracellular Par-4.” Oncogene 29(27): 3873–80.

98 References

Snipas, S. J., M. Drag, H. R. Stennicke, and G. S. Salvesen 2008. “Activation Mechanism and Substrate Specificity of the Drosophila Initiator Caspase DRONC.” Cell Death and Differentiation 15(5): 938–45.

Sohn, D., K. Schulze-Osthoff, and R. U. Jänicke 2005. “Caspase-8 Can Be Activated by Interchain Proteolysis without Receptor-Triggered Dimerization during Drug- Induced Apoptosis.” Journal of Biological Chemistry 280(7): 5267–73.

Sollberger, G. et al. 2014. “Caspase-1: The Inflammasome and Beyond.” Innate Immunity 20(2): 115–25.

Song, M. S et al. 2008. “The Deubiquitinylation and Localization of PTEN Are Regulated by a HAUSP-PML Network.” Nature 455(7214): 813–17.

Sorlie, T. et al. 2001. “Gene Expression Patterns of Breast Carcinomas Distinguish Tumor Subclasses with Clinical Implications.” Proceedings of the National Academy of Sciences USA 98(19): 10869–74.

Soussi, T. et al. 2006. “Meta-Analysis of the p53 Mutation Database for Mutant p53 Biological Activity Reveals a Methodologic Bias in Mutation Detection.” Clinical Cancer Research 12(1): 62–69.

Srinivasula, S. M. et al. 2001. “A Conserved XIAP-Interaction Motif in Caspase-9 and Smac/DIABLO Regulates Caspase Activity and Apoptosis.” Nature 410(6824): 112–16.

Stehlik, C., R. de Martin, I. Kumabashiri, et al. 1998. “Nuclear Factor (NF)-κB-Regulated X-Chromosome-Linked Iap Gene Expression Protects Endothelial Cells from Tumor Necrosis Factor α-Induced Apoptosis.” Journal of Experimental Medicine 188(1): 211–16.

Stehlik, C., R. de Martin, B. R. Binder, and J. Lipp 1998. “Cytokine Induced Expression of Porcine Inhibitor of Apoptosis Protein (iap) Family Member Is Regulated by NF- κB.” Biochemical and Biophysical Research Communications 243(8): 827–32.

Stennicke, H. R. et al. 1998. “Pro-Caspase-3 Is a Major Physiologic Target of Caspase- 8.” Journal of Biological Chemistry 273(42): 27084–90.

Stennicke, H. R., M. Renatus, M. Meldal, and G. S. Salvesen 2000. “Internally Quenched Fluorescent Peptide Substrates Disclose the Subsite Preferences of Human Caspases 1, 3, 6, 7 and 8.” The Biochemical Journal 350(Pt 2): 563–68.

Strasser, A. 2005. “The Role of BH3-Only Proteins in the Immune System.” Nature Reviews - Immunology 5(3): 189–200.

Suzuki, Y., Y. Nakabayashi, and R. Takahashi 2001. “Ubiquitin-Protein Ligase Activity of X-Linked Inhibitor of Apoptosis Protein Promotes Proteasomal Degradation of Caspase-3 and Enhances Its Anti-Apoptotic Effect in Fas-Induced Cell Death.” Proceedings of the National Academy of Sciences USA 98(15): 8662–67.

99 References

Tacar, O., P. Sriamornsak, and C. R. Dass 2013. “Doxorubicin: An Update on Anticancer Molecular Action, Toxicity and Novel Drug Delivery Systems.” Journal of Pharmacy and Pharmacology 65(2): 157–70.

Tahir, S. K. et al. 2007. “Influence of Bcl-2 Family Members on the Cellular Response of Small-Cell Lung Cancer Cell Lines to ABT-737.” Cancer Research 67(3): 1176–83.

Tait, S. W., and D. R. Green 2010. “Mitochondria and Cell Death: Outer Membrane Permeabilization and beyond.” Nature Reviews - Molecular Cell Biology 11(9): 621–32.

Taylor, R. C., S. P. Cullen, and S. J. Martin 2008. “Apoptosis: Controlled Demolition at the Cellular Level.” Nature Reviews - Molecular Cell Biology 9(3): 231–41.

Tenev, T. et al. 2011. “The Ripoptosome, a Signaling Platform That Assembles in Response to Genotoxic Stress and Loss of IAPs.” Molecular Cell 43(3): 432–48.

Theriault, R. L. et al. 2013. “Breast Cancer, Version 3.2013: Featured Updates to the NCCN Guidelines.” Journal of the National Comprehensive Cancer Network 11(7): 753–60. de Thonel, A. et al. 2014. “Regulation of the Proapoptotic Functions of Prostate Apoptosis Response-4 (Par-4) by Casein Kinase 2 in Prostate Cancer Cells.” Cell Death and Diseases 5: e1016.

Thornberry, N. A. et al. 1992. “A Novel Heterodimeric Cysteine Protease Is Required for Interleukin-1ß Processing in Monocytes.” Nature 355(6372): 242–44.

Thornberry, N. A. et al. 1997. “A Combinatorial Approach Defines Specificities of Members of the Caspase Family and Granzyme B.” Journal of Biological Chemistry 272(29): 17907–11.

Timmer, J. C., and G. S. Salvesen 2007. “Caspase Substrates.” Cell Death and Differentiation 14(1): 66–72.

Toshiyuki, M., and J. C. Reed 1995. “Tumor Suppressor p53 Is a Direct Transcriptional Activator of the Human Bax Gene.” Cell 80(2): 293–99.

Turriziani, B. et al. 2014. “On-Beads Digestion in Conjunction with Data-Dependent Mass Spectrometry: A Shortcut to Quantitative and Dynamic Interaction Proteomics.” Biology 3(2): 320–32.

Vansteenkiste, J. et al. 2013. “Early and Locally Advanced Non-Small-Cell Lung Cancer (NSCLC): ESMO Clinical Practice Guidelines for Diagnosis, Treatment and Follow- Up.” Annals of Oncology 24(Suppl.6): vi89–98.

Vaux, D. L. 2011. “Apoptogenic Factors Released from Mitochondria.” Biochimica et Biophysica Acta - Molecular Cell Research 1813(4): 546–50.

100 References

Verhagen, A. M. et al. 2000. “Identification of DIABLO, a Mammalian Protein That Promotes Apoptosis by Binding to and Antagonizing IAP Proteins.” Cell 102(1): 43– 53.

Verhagen, A. M. et al. 2002. “HtrA2 Promotes Cell Death through Its Serine Protease Activity and Its Ability to Antagonize Inhibitor of Apoptosis Proteins.” Journal of Biological Chemistry 277(1): 445–54.

Vetterkind, S. et al. 2005. “Binding of Par-4 to the Actin Cytoskeleton Is Essential for Par-4/Dlk-Mediated Apoptosis.” Experimental Cell Research 305(2): 392–408.

Vetterkind, S., and K. G. Morgan 2009. “The pro-Apoptotic Protein Par-4 Facilitates Vascular Contractility by Cytoskeletal Targeting of ZIPK.” Journal of Cellular and Molecular Medicine 13(5): 887–95.

Volkova, M., and R. Russell 2011. “Anthracycline Cardiotoxicity: Prevalence, Pathogenesis and Treatment.” Current Cardiology Reviews 7(4): 214–20.

Wagner, S. A. et al. 2011. “A Proteome-Wide, Quantitative Survey of in Vivo Ubiquitylation Sites Reveals Widespread Regulatory Roles.” Molecular & Cellular Proteomics 10(10): M111.013284.

Wajant, H., K. Pfizenmaier, and P. Scheurich 2003. “Non-Apoptotic Fas Signaling.” Cytokine and Growth Factor Reviews 14(1): 53–66.

Walczak, H., and P. H. Krammer 2000. “The CD95 (APO-1/Fas) and the TRAIL (APO- 2L) Apoptosis Systems.” Experimental Cell Research 256(1): 58–66.

Walsh, J. G. et al. 2008. “Executioner Caspase-3 and Caspase-7 Are Functionally Distinct Proteases.” Proceedings of the National Academy of Sciences USA 105(35): 12815–19.

Wang, L., F. Du, and X. Wang 2008. “TNF-α Induces Two Distinct Caspase-8 Activation Pathways.” Cell 133(4): 693–703.

Wee, L. J. K., T. W. Tan, and S. Ranganathan 2007. “CASVM: Web Server for SVM- Based Prediction of Caspase Substrates Cleavage Sites.” Bioinformatics 23(23): 3241–43.

Weidemann, A. et al. 1999. “Proteolytic Processing of the Alzheimer’s Disease Amyloid Precursor Protein within Its Cytoplasmic Domain by Caspase-like Proteases.” Journal of Biological Chemistry 274(9): 5823–29.

Weil, M. K., and A. P. Chen 2011. “PARP Inhibitor Treatment in Ovarian and Breast Cancer.” Current Problems in Cancer 35(1): 7–50.

Weinlich, R., C. P. Dillon, and D. R. Green 2011. “Ripped to Death.” Trends in Cell Biology 21(11): 630–37.

101 References

Wesselborg, S. et al. 1999. “Anticancer Drugs Induce Caspase-8/FLICE Activation and Apoptosis in the Absence of CD95 Receptor/ligand Interaction.” Blood 93(9): 3053– 63.

Yewale, C. et al. 2013. “Epidermal Growth Factor Receptor Targeting in Cancer: A Review of Trends and Strategies.” Biomaterials 34(34): 8690–8707.

You, M., P. T. Ku, R. Hrdlicková, and H. R. Bose 1997. “Ch-IAP1, a Member of the Inhibitor-of-Apoptosis Protein Family, Is a Mediator of the Antiapoptotic Activity of the v-Rel Oncoprotein.” Molecular and Cellular Biology 17(12): 7328–41.

Youle, R. J., and A. Strasser 2008. “The BCL-2 Protein Family: Opposing Activities That Mediate Cell Death.” Nature Reviews - Molecular Cell Biology 9(1): 47–59.

Yu, J., and L. Zhang 2008. “PUMA, a Potent Killer with or without p53.” Oncogene 27(Suppl. 1): S71–83.

Yuan, J. et al. 1993. “The C. Elegans Cell Death Gene Ced-3 Encodes a Protein Similar to Mammalian Interleukin-1ß-Converting Enzyme.” Cell 75(4): 641–52.

Yuan, P. et al. 2012. “Oral Etoposide Monotherapy Is Effective for Metastatic Breast Cancer with Heavy Prior Therapy.” Chinese Medical Journal 125(5): 775–79.

Yuan, S. et al. 2010. “Structure of an Apoptosome-Procaspase-9 CARD Complex.” Structure 18(5): 571–83.

Zaman, M. M. et al. 2013. “Ubiquitination-Deubiquitination by the TRIM27-USP7 Complex Regulates Tumor Necrosis Factor Alpha-Induced Apoptosis.” Molecular and Cellular Biology 33(24): 4971–84.

Zapata-Benavides, P. et al. 2009. “Expression of Prostate Apoptosis Response (Par-4) Is Associated with Progesterone Receptor in Breast Cancer.” Archives of Medical Research 40(7): 595–99.

Zhao, Y. et al. 2007. “Cancer Resistance in Transgenic Mice Expressing the SAC Module of Par-4.” Cancer Research 67(19): 9276–85.

Zhou, P. et al. 1999. “Solution Structure of Apaf-1 CARD and Its Interaction with Caspase-9 CARD: A Structural Basis for Specific Adaptor/caspase Interaction.” Proceedings of the National Academy of Sciences USA 96(20): 11265–70.

102 Appendix

7 Appendix

7.1 Abbreviations

A AIDS Acquired immunodeficiency syndrome AIF Apoptosis inducing factor Apaf-1 Apoptotic protease activating factor 1 aPKC Atypical Protein Kinase C APP Amyloid precursor protein AR Androgen receptor B BAD BCL-2 antagonist of cell death BAK BCL-2 antagonist or killer BAX BCL-2-associated X protein BCL-2 B-cell lymphoma 2 BCL-XL BCL extra-large BCR-ABL Breakpoint cluster region Abelson kinase BH3 BCL-2 homology domain BID BH3-interacting-domain death agonist BIM BCL-2-interacting mediator of cell death BOK BCL-2 ovarian killer protein BRCA1 Breast cancer 1 BRCA2 Breast cancer 2 BSA Bovine serum albumin C C Carboxy-terminal CAD Caspase-activated DNase CALD1 Caldesmon CARD Caspase recruitment domain Caspase Cysteine-dependent aspartate-directed protease CED-3 Caenorhabditis elegans death protein 3 cFLIP Cellular FLICE-like inhibitory protein CHX Cycloheximide

103 Appendix cIAP-1 Cellular inhibitor of apoptosis 1 cIAP-2 Cellular inhibitor of apoptosis 2 CK2 Casein kinase 2 D dATP Deoxyadenosine triphosphate Daxx Death-associated protein 6 ddH2O Double distilled water DED Death efeector domain DISC Death-inducing signaling complex DLK DAP-like kinase E eCFP Enhanced cyan fluorescent protein ECL Enhanced chemiluminescence EDTA Ethylenediamine tetra-acetic acid ELISA Enzyme-linked immunosorbent assay EGFR Epidermal growth factor receptor ER Estrogen receptor eYFP Enhanced yellow fluorescent protein F FADD Fas-associated death domain FBXO45 F-box protein 45 FCS Fetal calf serum G GOI Gene of interest H h Human HA Hemaglutinin HBS Hepes buffered saline HECT Homologous to E6-AP carboxyl terminus HER2 Human epidermal growth factor receptor 2 HIV Human immundeficiancy virus HRP Horseradish peroxidas HSPH1 Heat shock protein 105 kDa HtrA2 High-temperature requirement protein A2

104 Appendix

I IAP Inhibitors of apoptosis IB Immunoblotting IBM IAP binding motif ICAD Inhibitor of Caspase-activated DNase ICE Interleukin-1β-converting enzyme IDH1 Isocitrate dehydrogenase 1 IF Immunofluorescence IHC Immunohistochemistry IκB Inhibitor of NF-κB IKK IκB kinase IL Interleukin IP Immunoprecipitation Y YAP Yes-associated protein

K kDa Kilodalton L LB Lysogeny broth LZ Leucine zipper M MCL-1 Induced myeloid leukemia cell differentiation protein 1 MEK Mitogen-activated protein kinase kinase MLC Myosin light chain MOMP Mitochondrial outer membrane permeabilization MYCBP2 Myc binding protein 2 N N Amino-terminal NF-κB Nuclear factor kappa-light-chain-enhancer of activated B-cells NLS Nuclear localisation sequence O o/n Over night

105 Appendix

P PAGE Polyacrylamid gel electrophoresis PAR-4 Prostate apoptosis response protein 4 PARP-1 Poly(ADP-ribose) polymerase 1 PAWR PRKC, Apoptosis, WT1, Regulator PBS Phosphate buffered saline PCR Polymerase chain reaction PFA Paraformaldehyde PI3K Phosphatidylinositol-4,5-bisphosphate 3-kinase PKA Protein kinase A PKB Protein kinase B PML Promyelocytic leukemia PMSF Phenylmethylsulfonyl fluoride PPP1CB Protein phosphatase 1 catalytic subunit beta PPP1R12A Protein phosphatase 1 regulatory subunits 12A PPP1R12B Protein phosphatase 1 regulatory subunits 12B PR Progesterone receptor PTEN Phosphatase and tensin homolog PUMA p53 up-regulated modulator of apoptosis PVDF Polyvenylidene fluoride R r Rat RING Really interesting new gene RIPK1 Receptor-interacting serine/threonine kinase 1 RNA Ribonucleic acid rpm Rounds per minute RT Room temperature S SAC Selective for apoptosis in cancer cells SDS Sodium dodecyl sulphate siRNA Small interfering RNA Smac Second mitochondrial activator of Caspases SVM Support vector machine

106 Appendix

T TAE Tris; Acetic acid; EDTA tBID Truncated BID THAP1 THAP domain containing protein 1 TLB Triton lysis buffer TNBC Triple negative breast cancer TNFR Tumor necrosis factor receptor TOP Topoisomerase TRADD TNFR-associated death domain TRAILR TNF-related apoptosis-inducing ligand receptor TRIM Tripartite motif U U Units UBR5 Ubiquitin protein ligase E3 component N-recognin 5 USP7 Ubiquitin specific peptidase 7 UV Ultra violet V v Volume W w Weight WCL Whole cell lysate WT Wild-type WT1 Wilms tumor protein 1 X XIAP X-linked inhibitor of apoptosis Y YAP Yes-associated protein Z ZIPK Zipper interacting protein kinase

107 Appendix

7.2 Curriculum Vitae

Personal information Name Treude, Fabian Date of birth June 10, 1988 Place of birth Herten, Germany Citizenship German

Education Since 2012 Doctoral studies (Dr. rer. nat.), Biology Institute of Biochemistry and Molecular Biology, RWTH Aachen University, Germany

2010-2012 Master of Science Molecular Biology, Westfälische Hochschule, Germany Thesis at the Institute of Biochemistry and Molecular Biology, RWTH Aachen University, Germany: Funktionelle Analyse der Caspase-abhängigen Spaltung des Tumorsuppressors PAR-4.

2007-2010 Bachelor of Science Molecular Biology, Westfälische Hochschule Germany Thesis at the Institute for Pharmaceutical Science, Albert-Ludwigs University of Freiburg, Germany: Einfluss partikulärer Emissionen auf den NF-κB-Signalweg in der Lungenzelllinie A549.

1998-2007 High school Leibniz-Gymnasium, Gelsenkirchen, Germany

1994-1998 Elementary school KGS Liebfrauenschule, Gelsenkirchen, Germany

Language skills German native language English fluently

108 Appendix

7.3 Scientific contributions

Essential parts of this thesis are published or will be submitted for publication.

Treude F., Plaster J., Krämer O., Lüscher B., Hartkamp J. PAR-4 sensitizes breast cancer cells to DNA damage by antagonizing cIAP-1. Manuscript in preparation.

Treude F., Kappes F., Fahrenkamp D., Müller-Newen G., Dajas-Bailador F., Krämer O.H., Lüscher B., Hartkamp J. (2014). Caspase-8-dependent PAR-4 cleavage is required for TNFα-induced apoptosis. Oncotarget, 5(10):2988-98.

Further publications in scientific journals:

Mancarella D.‡, Plaster J.‡, Vaes R.‡, Treude F., Floss D., Romanowski A., Bug G., Scheller J., Lüscher B., Krämer O.H., Hartkamp J. DNA damage induces caspase-8-mediated cleavage of WT1 in acute myeloid leukemia. Manuscript in preparation. ‡ equal contributing first authors

Treude F., Gladbach T., Plaster J., Hartkamp J. Assessment of HDACi-induced Protein Cleavage by Caspases. Methods in Molecular Biology, manuscript submitted.

Herzog N., Hartkamp J.D., Verheugd P., Treude F., Forst A.H., Feijs K.L., Lippok B.E., Kremmer E., Kleine H., Lüscher B. (2013). Caspase-dependent cleavage of the mono-ADP-ribosyltransferase ARTD10 interferes with its pro-apoptotic function.

109 Appendix

FEBS J., 280(5):1330-43.

Presentations at scientific meetings:

Treude F., Gladbach T., Lüscher B., Hartkamp J. (2015). Caspase-8-mediated PAR-4 cleavage is required for DNA-damage-induced apoptosis in triple negative breast cancer cells. Poster presentation and talk given at the 19th Joint Meeting: Signal Transduction – Receptors, Mediators and Genes, 2-4 November 2015, Weimar, Germany.

Treude F., Kappes F., Fahrenkamp D., Müller-Newen G., Dajas-Bailador F., Krämer O., Lüscher B., Hartkamp J. (2014). Caspase-8-dependent PAR-4 cleavage is required for TNFα-induced apoptosis. Poster presentation and talk given at the 18th Joint Meeting: Signal Transduction – Receptors, Mediators and Genes, 5-7 November 2014, Weimar, Germany.

110 Appendix

7.4 Eidesstattliche Erklärung

Ich erkläre eidesstattlich, dass ich die vorliegende Dissertation selbständig verfasst und alle in Anspruch genommenen Hilfen in der Dissertation kenntlich gemacht habe.

Aachen, 22.08.2016 Fabian Treude

111 Appendix

7.5 Danksagung

Uns Menschen aus dem Ruhrgebiet bzw. Ruhrpott - so nennen wir unsere Heimat - wird oft nachgesagt, dass wir eine eigene Lebensart und Kultur aufweisen, die stark mit unserer Geschichte verwurzelt ist. Tatsächlich hat uns unsere Industriekultur drastisch geprägt. Wir wachsen multikulturell auf, sind weltoffen und direkt zueinander. Wir geben uns offenherzig und ziehen in jedem Fall das Du dem Sie vor. Wir lieben unsere Buden um die Ecke, essen gerne Pommes Schranke und pflegen unsere kleinen Schrebergärten. Wir besitzen ein ausgeprägtes Maß an Lokalpatriotismus, stehen jedoch zu jeder Zeit hinter der gesamten Region. Bei uns gibt es lediglich zwei Religionen, nämlich die Fußballvereine FC Schalke 04 und Lüdenscheid-Nord, wobei der Anhang an letztere Institution dann doch eher an Ketzerei grenzt. Vor allem aber zeichnen wir Menschen aus dem Ruhrpott uns durch unsere Dankbarkeit aus. Wir bedanken uns nicht zwingend, um der Höflichkeit gerecht zu werden. Wir sagen Danke, weil es von Herzen kommt!

Als erstes möchte ich mich bei Jörg bedanken, der mir die Möglichkeit gegeben hat zu promovieren und an einem äußerst spannenden Projekt forschen zu dürfen. Danke Jörg für die vielen Dinge, die ich von dir lernen konnte und vor allem Danke für die zwischenmenschliche Harmonie. Es war eine richtig geile Zeit!

Bernhard danke ich nicht nur für die Funktion des Doktorvaters meiner Promotion und für die wissenschaftlichen Anregungen. Vielmehr möchte ich mich für die offene und herzliche Art bedanken, die zu jeder Zeit zu einem tollen Arbeitsklima beigetragen hat.

Desweiteren möchte ich Christian für die freundliche Übernahme des Koreferats und den regelmäßigen wissenschaftlichen Austausch danken.

Meinem Dank gilt ebenso Michael, der sich für das Ausstellen eines 3. Gutachtens spontan bereit erklärt hat und mir damit viel Zeit erspart hat.

Ich bedanke mich auch bei Herrn Prof. Dr. Lothar Elling für die Funktion des mündlichen Prüfers und Herrn Prof. Dr. Hermann Wagner für den Vorsitz meiner Verteidigung.

112 Appendix

Als nächstes möchte ich mich bei allen Personen aus Labor 15 bedanken, die mich während meiner gesamten Promotion begleitet haben. Danke an die „jungen Wilden“ Haihong, Malte, Ferdi und Chris für die unglaublich tolle Atmosphäre während der Arbeit und darüber hinaus. Ich danke natürlich auch Jacky, Franziska, Manfred, Tobi, Elke, Irina, Rianne, Daniela, Mathias, Pia, Nahleen und Sandra, die über die Jahre zu diesem angenehmen Klima beigetragen haben. Ich wünsche euch weiterhin alles Gute!

Mein Dank gilt ebenfalls Gerhard und seiner Arbeitsgruppe von nebenan. Ein großer Dank geht an Hildegard, die mich in allen Belangen rund ums Klonieren unterstützt hat. Ich danke auch Andrea für ihr offenes Ohr bzgl. all meiner Fragen. Danke Dirk für die Unterstützung am konfokalen Mikroskop und die unzähligen amüsanten Gespräche. Weiterhin möchte ich mich bei Dieter, Sabrina, Anton, Natalie, Tamas, Nico, allen Studenten aus eurem Labor und den zuvor genannten für die vielen witzigen Momente bedanken.

Ich möchte mich auch bei der aktuellen Arbeitsgruppe und den ehemaligen Mitarbeitern von Michael bedanken, als da wären: Marlies, Tanja, Karin, Willi, Carolin, Vrinda, Magdalena, Mathias, Fabian, Ronja, Jörn, Laura, Thomas, Marcel, Katrin, Susanna, Oundrilla, Allison und unzählige Studenten. Danke für die geselligen und entspannten Mittagspausen und die freundliche Art von jedem Einzelnen von euch.

Natürlich bedanke ich mich auch bei der Arbeitsgruppe von Bernhard. Ich danke Barbara, Patrice, Mareike, Laura, Caro, Sarah, Marc, Jorgo, Agnieszka, Juliane, Jürgen, Jörg, Elena, Christian, den Ehemaligen Andrea, Kai, Alex, Karla, Nico, Weili, Franziska und auch allen Studenten für unzählige Tipps und die vielen lustigen Momente. Danke für die schöne Zeit!

Ein großes Dankeschön geht auch an Monika, Patricia, Angelina und Marcel, die uns allen im Institut die tägliche Arbeit ungemein erleichtern und immer hilfsbereit sind. Danke für euren Einsatz!

Ich hoffe, dass ich niemanden vergessen habe und wünsche euch allen weiterhin viel Erfolg und persönlich alles Gute für eure Zukunft! Oder wie wir im Ruhrpott sagen: Glück auf!

113 Appendix

Desweiteren möchte ich an dieser Stelle auch meinen Freunden und meiner Fußballmannschaft danken, die mich trotz der Entfernung immer unterstützt und respektiert haben. Danke für die neugierigen Fragen und das Interesse an meiner Arbeit, auch wenn manche von euch immer noch denken, dass ich an Amöben geforscht und das Protein an dem ich gearbeitet habe „voll das Endgegner-Protein ist“. Die Wahrheit liegt vermutlich irgendwo dazwischen.

Mein besonderer Dank gilt Birgit und Uli für die jahrelange Unterstützung.

Ich möchte mich auch bei meiner Schwester Sarah und André bedanken. In vielen Dingen seid ihr große Vorbilder für mich und ich hoffe, dass ich das auch für euch und später einmal für Jonas sein kann. Danke!

Ich möchte natürlich die Chance ergreifen mich meinen Eltern, Manuela und Frank, zu widmen und euch meinen größten Dank auszusprechen. Ihr habt mir mein Studium ermöglicht und mir jederzeit alle Freiheiten gelassen, um mich frei entwickeln zu können und diesen Weg einzuschlagen. Ihr habt mir immer wieder geholfen und mich motiviert. Danke für alles!

Zuletzt möchte ich mich bei meiner Freundin bedanken. Liebe Julia, ich bin froh, dass du all die Jahre an meiner Seite stehst und mich in jeglicher Hinsicht unermüdlich unterstützt. Ich hoffe, dass wir beide die nächsten Hürden unseres Lebens zusammen genauso meistern, wie bisher. Aber da mach ich mir keine Sorgen. Im Gegenteil, ich freue mich schon auf das nächste Abenteuer mit dir. Vom ganzen Herzen: Danke!

114