Master Thesis

The role of EAPP in cell cycle regulation and DNA damage control

Master Thesis

For the attainment of the academic degree

Master of Science

From the University of Applied Sciences FH Campus Wien

Submitted by: Sarah Gobetzky

Personal identity code 1410544011

Supervisor: Ao. Prof. Dr. Johann Rotheneder Max F. Perutz Laboratories; Medical University of Vienna; Dept. of medical Biochemistry Vienna Austria

Submitted on: 10. 04. 2017 1.1 Abstract in English

Cell cycle checkpoints are an important tool of eukaryotic cells to ensure the correct and error-free replication of the DNA and division of the cells. A cell undergoes apoptosis or remains in cell cycle arrest until any damage is corrected before re-entering the cell-cycle. Mutations bypassing this system or other interferences often lead to the generation of carcinogenic structures.

The E2F associated phosphoprotein (EAPP), discovered by our group, interacts with E2F 1-3, alters in an E2F-dependent manner the activity of cell cycle regulated promotors and is a co-factor in E2F dependent transcription. The levels of this protein are elevated in various human cancers. It could be observed that overexpression but also a knockout of EAPP seems to lead to a slowdown of the cell cycle and great G1 arrest. An interaction of EAPP with p53, MAO B and Chk2 could already be uncovered. Furthermore it could be shown that EAPP levels are elevated after double strand breaks and stimulates p21 expression causing cell cycle arrest in G1.

In this Master´s thesis the impact of EAPP was further investigated. Total knockouts of EAPP in NIH 3T3 cells revealed that EAPP is not necessary for survival for those cells like it is for example for U2OS cells. Cells expressing various variants or truncations of EAPP were generated and used for examination on their impact on stability or protein-protein interactions or to determine their functional domains.

It could be shown that the loss of EAPP in HAP1 cells leads to a drastic increase and reduced amounts of a truncated version of EAPP lead to a moderate increase in their rate of expansion compared to the wt. In addition it could be observed that cells with higher levels of EAPP show a higher fraction of cells in G1 phase after DNA damage introduced by Etoposide. It could also be observed that truncated versions of EAPP seem to show increased stability in comparison to the wt. Analysis revealed that the loss of EAPP or reduced amounts of a truncated version lead to a dramatic increase of P-Akt levels. This effect is even enhanced when treated

2 with Etoposide. Also U2OS cells show elevated P-Akt levels in cells with reduced amounts of EAPP and rather low levels in cells with high EAPP levels indicating a correlation between low P-Akt levels and high EAPP and vice versa. The level of EAPP seems further having an impact on the recovery speed as NIH 3T3 cells lacking EAPP seem to re-enter the cell cycle much faster after serum deprivation than NIH 3T3 cells with normal levels of EAPP. In addition to the obtained results we could identify Helic2, PRP8, E6AP, p53, Sm B/B´/N, GSK 3β, YB1 and AAR2 as putative interaction partners of EAPP.

Our results show clearly an increase of cell fractions in G1 arrest after DNA damage with elevated EAPP levels, alterations of expression levels of different proteins not to mention interaction of EAPP with various splice factors underlining the important role EAPP seems to play in the cell cycle and its regulation and control.

1.2 Abstract in German

Zellzykluskontrollpunkte sind ein wichtiges Mittel eukaryotischer Zellen um sicherzustellen dass sich Zellen in einem einwandfreien Zustand replizieren. Die Zelle unterliegt der Apoptose oder verharrt im Zellzyklusarrest bis die aufgetretenen Schäden behoben sind ehe der Zellzyklus weiter voranschreitet. Mutationen die dieses System umgehen oder anderweitige Störungen führen oft zu der Ausbildung kanzerogener Strukturen.

Das in unserer Forschungsgruppe entdeckte E2F associated phosphoprotein (EAPP) interagiert mit E2F 1-3, hat mit deren Hilfe Einfluss auf zellzyklusabhängige Promotoren und ist ein Co-faktor in der E2F abhängigen Transkription. Es ist ein Protein dessen Levels in verschiedensten humanen Krebsarten deutlich erhöht sind. Es konnte sowohl in Zellen mit Überexpression als auch mit knockout eine Verlangsamung des Zellzyklus sowie eine erhöhte Zellzahl in G1 beobachtet werden. Eine Interaktion von EAPP mit p53, MAO B oder auch Chk2 wurde bereits nachgewiesen. Zudem wurde gezeigt dass EAPP Levels nach Doppelstrangbrüchen erhöht sind und durch eine Stimulation der p21 Expression zu Zellzyklusarrest in G1 führen.

In dieser Masterarbeit wurde die Bedeutung von EAPP weiter untersucht. Totale Knockouts von EAPP in NIH 3T3 Zellen haben bewiesen dass in diesen Zellen EAPP nicht für ein Überleben notwendig ist, wie es beispielsweise in U2OS Zellen der Fall ist. Zellen welche verschiedenste Varianten oder Verkürzungen von EAPP exprimieren wurden hergestellt und dazu verwendet deren Einfluss auf Stabilität, Protein-Protein Interaktionen festzustellen oder mögliche funktionelle Domänen auszumachen.

Es konnte gezeigt werden dass der Verlust von EAPP in HAP1 Zellen zu einer deutlich erhöhten Proliferationsrate und verringerte EAPP Mengen einer verkürzten Form zu einer leicht erhöhten Proliferationsrate führen. Zudem konnte beobachtet werden dass sich nach DNS Schädigung durch Etoposid deutlich mehr Zellen mit erhöhten Mengen an EAPP in G1 befinden. Eine weitere Beobachtung war die Tatsache dass verkürzte EAPP Versionen im Vergleich zum endogenen EAPP stabiler zu sein scheinen. Analysen haben

4 gezeigt dass der Verlust von EAPP oder verringerte Mengen einer verkürzten Form in HAP1 Zellen zu einer drastischen Erhöhung der P-Akt Level führen. Dieser Effekt ist durch Etoposid noch einmal verstärkt. Auch U2OS Zellen zeigen bei verringerten Mengen an EAPP erhöhte Mengen an P-Akt und bei erhöhten EAPP Mengen verringerte P-Akt Mengen, was eine Korrelation zwischen geringem P-Akt und hohem EAPP und umgekehrt vermuten lässt. Das Level an EAPP scheint zudem Auswirkungen auf die Erholungsfähigkeit der Zellen zu haben, da NIH 3T3 Zellen ohne EAPP nach Serumentzug deutlich beschleunigten Wiedereintritt in den Zellzyklus gezeigt haben. Zudem konnten Helic2, PRP8, E6AP, p53, Sm B/B´/N, GSK 3β, YB1 und AAR2 als mögliche Interaktionspartner identifiziert werden.

Unsere Ergebnisse zeigen deutlich eine erhöhte Anzahl an Zellen in G1 nach DNS Schädigung durch Etoposid in Zellen mit erhöhten EAPP Levels, Veränderungen in den Expressionslevels verschiedener Proteine sowie Interaktionen von EAPP mit verschiedenen Splicefaktoren die unterstreichen dass EAPP eine wichtige Rolle im Zellzyklus und seiner Kontrolle zu spielen scheint.

2 Table of Content

1.1 Abstract in English ...... 2 1.2 Abstract in German ...... 4 2 Table of Content ...... 6 3 List of Abbreviations ...... 8 4 List of Figures ...... 11 5 Introduction ...... 13 5.1 Cell cycle...... 13 5.2 p53 and its role in cell cycle ...... 17 5.3 E2F and its role in cell cycle progression ...... 18 5.4 EAPP ...... 20 5.5 CRISPR/Cas9 ...... 24 5.6 Aims ...... 27 6 Materials and Methods ...... 28 6. 1 Materials ...... 28 6.2 Methods ...... 36 7 Results ...... 43 7.1 Generation of cell lines with different version of EAPP ...... 43 7.1.1 Generation of NIH 3T3 cells with knockouts of EAPP or introduction of truncated EAPP versions...... 43 7.1.2 Generation of U2OS cells expressing HA-EAPP 10-240 or 20-240 in addition to endogenous EAPP ...... 48 7.1.3 Generation of U2OS cells with p53 knockout and re-transfection of p53 ...... 49 7.2 Cell proliferation assays ...... 50 7.2.1 Cell proliferation of HAP1 cells with different EAPP knockouts .... 50 7.2.2 Cell proliferation of NIH 3T3 cells with different EAPP versions ... 52 7.3 FACS analysis ...... 53 7.3.1 FACS analysis of U2OS cells with different EAPP versions after treatment with different concentrations of Etoposide ...... 53 7.3.2 FACS analysis of NIH 3T3 versions after serum starvation and reactivation ...... 56

7.4 Analysis of protein stability of EAPP versions in U2OS cells via treatment with Cycloheximide ...... 58 7.5 Alteration of protein expression of various proteins due to different EAPP expression ...... 60 7.5.1 General alterations of protein expression influenced by Etoposide and ATMi in HAP1 cells with EAPP knockouts...... 60 7.5.2 Influence of EAPP on the expression of ATMi-sensitive proteins in HAP1 and NIH 3T3 cells ...... 63 7.5.3. Influence of EAPP on P-Akt levels ...... 65 7.6 Analysis of EAPP protein interaction via immunoprecipitation ...... 70 7.6.1 Direct immunoprecipitation with antibody coupled Protein A beads in U2OS mEAPP (J)...... 70 7.6.2 Direct immunoprecipitation with HA coupled Protein A beads in U2OS cells with different versions of EAPP...... 73 8 Discussion / Outlook ...... 76 9 References ...... 82 10 Statutory declaration ...... 86

3 List of Abbreviations

Abbreviation Full word A Ampere °C Degree Celsius µg Microgram µl Microliter aa Amino Acid AAR2 AAR2 splicing factor Homolog / Protein AAR2 homolog Ab Antibody Akt Protein kinase B Amp Ampicilin ATM Ataxia Telangiectasia Mutated ATMi ATM inhibitor = KU 55933 bp Base pair Cas CRISPR associated nucleases Cdk Cyclin dependent kinase CDK1 Cdc2 / Cyclin dependent kinase 1 CHX Cycloheximide CK2 Casein kinase IIα CKI Cdk inhibitor CRISPR Clustered interspaced palindromic repeats CRISPRi CRISPR interference crRNA Short CRISPR RNA DMSO Dimethyl sulfoxide DSB Double strand break DNA Deoxyribonucleic acid DNS Desoxyribonukleinsäure DP Transcription factor DP - transcription factor dimerization partner E6AP E6AP ubiquitin-protein ligase / UBE3A / Ubiquitin-protein ligase E3A EAPP E2F associated Protein EtOH Ethanol FACS Fluorescence-activated cell sorting FCS Fetal calf serum g Gram Go Gap 0 phase G1 Gap 1 phase G2 Gap 2 phase G418 Geneticin GSK3 Glycogen synthase kinase 3

8 gRNA Guide RNA h Hours HA Hemagglutinin HBS HEPES buffered saline HDAC Histone deacetlyase HDR Homology directed repair HELIC2 SNRNP200 / small nuclear ribonucleoprotein 200kDa HR Homologous recombination HUR ELAVL1 / ELAV like RNA binding protein 1 IP Immunoprecipitation JAB1 c-Jun activation domain binding protein -1 KO Knockout Da Dalton kDa Kilo Dalton LB Lysogeny broth M Molar/ mol per liter mA Miliampere MAO Monoamine oxidase min Minute ml Milliliter mg Milligram mm Milimeter M phase Mitosis phase NHEJ Non-homologous end joining NLS Nuclear localization signal nm Nanometer nM Nanomolar / nanomol per liter ON Over night p16 Cyclin-dependent kinase inhibitor 2A p53 Cellular tumor antigen p53 PAA Poly acrylamide PAM Protospacer adjacent motive PARP Poly [ADP-ribose] polymerase 1 PBS Phosphate buffered saline PCR Polymerase chain reaction PEI Polyethlyenimine PI Propidium iodide Pin1 Peptidylpropyl cis/trans isomerase NIMA-interacting 1 P-Pras Phospo PRAS / phospo proline-rich Akt substrate pRB Retinoblastoma protein PTEN Phospatase and tensin homolog rcf Relative centrifugal force rH Relative humidity

RNA Ribonucleic acid RNP Ribonucleoprotein rpm Rotations per minute RT Room temperature siRNA Small interference RNA Sm B/B´/N Small nuclear ribonucleoprotein polypeptides B, B1, N S phase Synthesis phase Sp1 Simian virus 40 promotor factor TBH Tetrahydrobiopterin / sapropterin TBS Tris buffered saline TBS-T TBS containing Tween 20 TSA Trichostatin A V Volt wt Wildtype YB1 YBX1 / Y-box binding protein 1

4 List of Figures

Figure 1: Stages of the cell cycle…………………………………………………………. p.13 Figure 2: Cdk–cyclin complex……………………………………………………………… p.14 Figure 3: Cdk-cyclin complex distribution throughout the cell cycle… p.15 Figure 4: Scheme of Cdk activation………………………………………………….… p.15 Figure 5: Inactivation of Cdk-cyclin complex by CKI….……………………… p.16 Figure 6: Cell cycle control through Cdk-cyclin complexes. ……………… p.16 Figure 7: G1 arrest after DNA damage through Cdk inactivation……… p.17 Figure 8: Structure of E2F…………………………………………………………………… p.18 Figure 9: Activation of E2F by Cyclin D-Cdk4 complex. ……….…………… p.19 Figure 10: Subcellular locations of EAPP…………………………………..………… p.20 Figure 11 Scheme of EAPP dependent p21 regulation …………….………… p.21 Figure 12: Interaction of EAPP with P-Chk2. ……………………………………… p.22 Figure 13: Structure of CRISPR loci……………………………………………………… p.24 Figure 14: Structure of p53 CRISPR/Cas9 knockout plasmid. …………… p.40 Figure 15: Structure of p53 HDR plasmid. …………………………………………. p.40 Figure 16: Structure of CRISPR-Cas9 PX459 vector…………………………... p.41 Figure 17: Cartoon of EAPP versions used for transfection……………….. p.43 Figure 18: Overview of single cell clones after transfection and knockout. ……………………………………………………………………………………………. p.44 Figure19: Sequence alignment of EAPP with EAPP clone B……………… p.45 Figure 20: Sequence alignment of EAPP with EAPP clone E……………… p.45 Figure 21: Sequence alignment of EAPP with EAPP clone G…………… p.46 Figure 22: Sequence alignment of EAPP with EAPP clone Y….………… p.46 Figure 23: Sequence alignment of EAPP with EAPP clone BC…………… p.47 Figure 24: Sequence alignment of EAPP with EAPP clone BR……………… p.47 Figure 25: Western Blot Analysis of U2OS cells with different transfections of EAPP…………………………………………………………………………… p.48 Figure 26: U2OS cells with p53 knockout and insertion of p53………… p.49 Figure 27: Western Blot Analysis of HAP1 cells with different Knockouts of EAPP.…………………………………………………………………………….… p.50

Figure 28: Cell Proliferation Assay of HAP1 cells with different Knockouts of EAPP………………………………………………………………………………. p.50 Figure 29: Cell Proliferation Assay of HAP1 cells with different Knockouts of EAPP………………………………………………………………………………… p.51 Figure 30: Cell Proliferation Assay of NIH cells with different versions of EAPP……………………………………………………………………………………………….… p.52 Figure 31: FACS analysis of U2OS cells with different EAPP versions after treatment with different concentrations of Etoposide………………. p.53 Figure 32: FACS analysis of NIH versions after serum starvation and reactivation…………………………………………………………………………………………… p.56 Figure 33: Western Blot analysis of levels of EAPP versions after CHX treatment…………………………………………………………………………………………… p.58 Figure 34: Western Blot analysis of EAPP levels after CHX treatment: p.59 Figure 35: Western Blot analysis of EAPP levels after treatment with triple amount of CHX…………………………………………………………………………… p.59 Figure 36: Protein expression influenced by Etoposide and ATMi in HAP1 cells with EAPP knockouts……………………………………………………….… p.61 Figure 37: Protein expression influenced by ATMi in HAP1 and NIH 3T3 cells with EAPP knockout and knock-in………………………………………… p.63 Figure 38: Influence of EAPP Levels on pAkt…………………………………..… p.65 Figure 39: Influence of EAPP Levels on P-Akt in stable cell lines carrying EAPP alterations ………………………………………………………………….… p.67 Figure 40: Influence of EAPP Levels on P-Akt in stable cell lines carrying EAPP alterations ……………………………………………………………….…… p.69 Figure 41: Protein interaction partners of EAPP……………………………….…. p.70 Figure 42: Protein interaction partners of EAPP……………………………..…… p.72 Figure 43: Protein interaction partners of different EAPP truncations. p.73 Figure 44: Protein interaction partners of different EAPP versions or truncations…………………………………………………………….……………………………… p.75

5 Introduction

5.1 Cell cycle

The keystone for analysis of cell cycle control was led by Howard and Pelc in 1953 when they first published their work about cell proliferation. They were the first ones describing a distinct timeframe and proposed the existence of four parts composing the cell cycle. A cytokinesis part, a synthesis stage (S phase), a period prior to S phase called Gap1 phase (G1) and a period prior to the mitotic phase (M phase) called Gap2 (G2). With that first proposal the concept of cell cycle was created (Dubrovsky, Ivanov 2003).

These four stages of the cell cycle, seen in Figure 1, are continually traversed during growth for the generation of new daughter cells. During the S phase DNA is replicated condensed and the chromosomes are then separated during M phase and the cell divides in two daughter cells (Lodish 2004).

Figure 1: Stages of the cell cycle. Scheme showing the 4 stages of the cell cycle. During S phase replication takes place. After that phase every cell contains 4 chromosome copies. Separation of the cells takes place in M phase. During the G phases RNA and protein synthesis takes place and the cells are checked for correctness. It no mitogenic stimulant occurs the cell rests in G0. Molecular Cell Biology. p.17 Lodish, H. et. al. Freeman publisher´s. 13

The cell cycle can be sub divided into mitosis and interphase. Interphase consists of G1, S and G2 phase. The initiation, the process and the end of each phase are tightly regulated. During G1 phase the cells are preparing for S phase. Every chromosome builds a single non-replicated structure (Markl et al. 2011). During that phase many proteins necessary for the following DNA synthesis get expressed. At the restriction point the cells either progress into S phase or remain quiescent.

The cells remain resting in G0 in cases of low cyclin dependent kinase (Cdk) activity (Matson, Cook 2017) or the absence of a mitogenic stimulant (Davies et al. 2015). When the conditions change the cells can re-enter the cell cycle and progress into S-phase (Lodish 2004). During S phase every chromosome gets duplicated and later consists of two sister-chromatids linked by proteins. During G2 phase the cell prepares for mitosis for example by synthesising microtubules essential for the division of the chromatids (Markl et al. 2011).

Different checkpoints control the correct progression of the cell cycle. The restriction point is found at the G1/S transition and passing it commits the cell to DNA replication, mitosis and cytokinesis. Other checkpoints in S phase ensure the activation of DNA damage response if necessary. A checkpoint at the G2/M transition ensures that the DNA is fully replicated and not damaged before cytokinesis (Humphrey 2005).

Cyclin dependent kinases (Cdk) play an important role in the control of the cell cycle. Cdks depend on binding of cyclin subunits for their activity (Figure 2). Whereas the levels of different Cdks remain at the same level, their activities fluctuate during cell cycle and by phosphorylation of critical proteins initiate or Figure 2: Cdk–cyclin complex. Cyclin binds to regulate cell cycle progression. The progression Cdk and activates the from G1 to S phase for example is caused by the kinase. Molecular Biology of THE CELL. p.1062 activation of Cdks leading for example to Alberts, B. et. al. Garland unwinding of the DNA or activation of the helicase science Taylor & Francis group.

14 for initiation of the S-phase (Lewis, Alberts 2007).

Whereas the levels of cyclins massively fluctuate during the different phases, seen in Figure 3, the amounts of Cdks remain constant during the different stages of the cell cycle and only their activation status changes (Lewis, Alberts 2007).

Figure 3: Cdk-cyclin complex distribution throughout the cell cycle. Scheme of the concentrations of the three main cyclin types. Levels fluctuate during the different stages of the cell cycle. Molecular Biology of THE CELL. p.1062 Alberts, B. et. al. Garland science Taylor & Francis group.

Cdk activating kinases (CAK) phosphorylate an amino acid next to the active site of the Cdk leading to a conformational change allowing more effective phosphorylation of target proteins and in that way enhance the activity of the Cdk (Figure 4) (Lewis, Alberts 2007).

Figure 4: Scheme of Cdk activation. Binding of cyclin causes T-loop conformation alteration and liberation of the active site causing partial activation of Cdk. Phosphorylation by CAK in the T-loop alters the shape of it further resulting in better protein substrate binding capability. Molecular Biology of THE CELL. p.1063 Alberts, B. et. al. Garland science Taylor & Francis group. 15

Binding of a Cdk inhibitor protein (CKI) leads to a rearrangement resulting in an inactive Cdk-cyclin complex, as seen in Figure 5 (Lewis, Alberts 2007). Figure 5: Inactivation of Cdk-cyclin complex by CKI. Binding of a CKI, in this case p27, leads to rearrangement causing distortion of the active site and insertion into the ATP- binding site leading to inhibition of the enzyme activity.

Molecular Biology of THE CELL. p.1064 Alberts, B. et. al. Garland science Taylor & Francis group.

Whenever something disturbs the correct progression of the cell cycle, observable in Figure 6, the Cdk-cyclin complexes get blocked and therefore the cell cycle progression is stopped until the problem is solved. After completion of mitosis the Cdk activity is blocked leading to a stable G1 phase (Lewis, Alberts 2007).

Figure 6: Cell cycle control through Cdk-cyclin complexes. The activity of Cdk-cyclin complexes is influenced by various factors like the environment or condition of DNA damage causing block of cell cycle progression or DNA replication. Molecular Biology of THE CELL. p.1066 Alberts, B. et. al. Garland science Taylor & Francis group.

5.2 p53 and its role in cell cycle

One of the most important functions among the numerous functions of p53 is the role of a checkpoint sensor leading to G1 or G2 arrest in damaged cells via transcription of essential genes (Lodish 2004). Mutations hindering its ability of DNA binding and activation of gene expression is mutated in more than half of human tumors (Purves 2004).

The levels of p53 in unstressed cells are rather low due to its rapid degradation by the proteasome due to ubiquitination catalysed by the ubiquitin ligase mdm2 bound to p53 (Lewis, Alberts 2007). Its levels are rapidly elevated after DNA damage or other kind of stress due to stabilisation of p53 via phosphorylation on Ser15 by ATM (Humphrey 2005). It can activate various genes. The CKI p21 is one of them and is able to inhibit the G1 cyclin-Cdk complexes via binding, seen in Figure 7, and in that way leads to G1 arrest Figure 7: G1 arrest after DNA damage through Cdk inactivation. DNA damage leads to phosphorylation and until successful DNA therefore activation of p53 by Chk1 and Chk2 recruited by ATM or ATR. p53 activates transcription of the CKI p21. p21 protein repair (Lodish 2004). then binds and inactivates G1/S-Cdk and S-Cdk complexes leading to cell cycle stop in G1. Molecular Biology of THE CELL. p.1106 Alberts, B. et. al. Garland science Taylor & Francis group.

5.3 E2F and its role in cell cycle progression

The E2F family of transcription factors has an essential part in cell cycle control, progression and proliferation by interaction with important regulators like cyclins, Cdks or the pocket protein Retinoblastoma protein (pRb) (Tsai et al. 1998) (Andorfer, Rotheneder 2013).

Until now 8 members of the E2F family are known. E2F proteins form heterodimers with the DP branch of the family (Rotheneder et al. 2007). The activating members E2Fs 1, 2 and 3a play an important role in S phase entry and carry a pRb binding site in their transactivation domain (Humphrey 2005). E2F3b, E2F4 and E2F5 form a second subgroup of E2Fs and are repressors of pocket proteins (Rotheneder et al. 2007). Whereas E2F1-3 carry a nuclear localization signal (NLS) E2F 4 and E2F5 carry a nuclear export signal (NES) (Gaubatz et al. 2001). E2F6, as can be seen in Figure 8, contains a repression domain but no transactivation domain. E2F7a, E2F7b and E2F8 Figure 8: Structure of E2F. Scheme of do not a carry pocket protein different classes of E2F with functional domains like cyclin A/ cdk2 binding, DNA binding, binding site or a dimerization domain, pocket potein, and transativaiton domain. Rotheneder, H. et. al. heterodimerization domain but a Dynamic Cell Biology p.2, 2007. second DNA binding domain enabling DNA binding without dimerization with DP. The levels of the activating members of E2F fluctuate during the cell cycle with their greatest levels in G1 phase. These members only interact with the pocket protein pRb. In comparison to that E2F4 interacts with all three members of pocket proteins (pRb, p107, p130) whereas E2F5 shows only interaction with p130 (Rotheneder et al. 2007).

E2F transcription factors are mandatory for the transcription of genes needed for S phase entry like G1/S-cyclins, S-cyclins as well as proteins required for DNA synthesis. In the absence of a mitotic stimulus this transcription is inhibited by the binding of pRb to an un-phosphorylated E2F dimer (Lewis, Alberts 2007). In that way it sterically hinders its activity and converts E2F transcription factors into repressors of transcription (Humphrey 2005).

During G1 the cyclin D- Cdk4/6 complex gets free from Cdk-inhibitor 2A (p16) and therefore can phosphorylate pRb liberating E2F, seen in Figure 9. E2F can then activate gene expression of its own genes generating a positive feedback loop, as well as Figure 9: Activation of E2F by cyclin D-Cdk4 complex. Removal of the CKI p16 leads to activation of genes encoding for cyclin the cyclin D-Cdk4 complex. This complex is now able to E, Cdk2 and several phosphorylate pRb and in that way liberates and activates E2F. Molecular Cell Biology. p.958 Lodish, H. et. al. other proteins required Freeman publisher´s. for S phase. As cyclin E–Cdk2 complexes further phosphorylate pRb it comes to a positive feedback loop causing a massive increase of E2F in a short period of time committing the cells to DNA synthesis. An oncogenic mutation in this pathway leading to cyclin D overproduction, loss of p16 or pRb causing E2F overproduction is relatively often found in several tumors (Lodish 2004).

5.4 EAPP

The E2F binding protein EAPP (E2F- associated phosphoprotein) showing interaction with the activating E2F members is a highly phosphorylated protein. Its levels are often elevated in human transformed cells (Novy et al. 2005) indicating a potential role of EAPP in oncogenesis (Schwarzmayr et al. 2008).

The gene located on human chromosome 14 more precisely at 14q13.1 (Ensembl 2017) has a TATA-less promotor getting stimulated by Sp1 and Egr-1 and strongly repressed by Sp3 (Schwarzmayr et al. 2008).

EAPP is ubiquitously expressed with very high concentrations in heart, placenta or pancreas. (Weizmann Institute of Science. 2017) Within the cell it is mostly located in the nucleus as EAPP contains a NLS (Novy et al. 2005). In addition it is also strongly accumulated in the Golgi Figure 10: Subcellular locations of EAPP. EAPP shows high concentrations in nucleolus and Golgi apparatus, but also detectable but also present in the cytoplasm. The endoplasmic in the cytoplasm, as reticulum or mitochondria instead show no EAPP accumulation. [online] available at: observable in Figure 10 http://www.genecards.org/cgi- bin/carddisp.pl?gene=EAPP [last accessed (Weizmann Institute of 11.01.2017] Science. 2017).

EAPP is present in every stage of the cell cycle but mitosis (Novy et al. 2005).

EAPP interacts with E2F 1, 2 and 3a and alters E2F dependently the activity of cell cycle regulated promotors. It could be shown that EAPP acts as a co- factor in E2F dependent transcription. Whereas EAPP expression in most

20 cases leads to activation of E2F dependent promotors the E1β-promotor of the p14/ARF gene gets repressed (Schwarzmayr et al. 2008).

EAPP has been shown to interact with p53 (Rotheneder unpublished) as well as many other proteins like Chk2 (Andorfer et al. 2011).

EAPP gets upregulated following DNA damage like double strand breaks and stimulates p21 expression leading to an increased resistance against apoptosis. Elevated EAPP levels also lead to p21 dependent slowdown of the cell cycle or arrest in G1 phase in the mode of action already described and observable in Figure 7. EAPP modulates the p21 promotor activity by direct binding to its DNA and in that way competes with Sp1 for the same binding site, seen in Figure 11 (Andorfer, Rotheneder 2011). Studies revealed a binding of EAPP to two of the Sp1 binding sites in proximity to the TATA box implying a possible role of EAPP in the formation of the transcription pre- initiation complex as EAPP enhances the promotor activity of p21 but an EAPP knockdown leads to a decrease of several promotor bound elements (Andorfer et al. 2011). Furthermore, it could be shown that EAPP transcriptionally regulates p21 independently of p53. For that the binding sites 3 and 4 of Sp1 are crucial (Andorfer 2011).

Figure 11: Scheme of EAPP dependent p21 regulation. Proposed model of regulation of p21 by EAPP showing Sp binding sites and their interaction partners as well as assembly of pre-initiation factors like TAFs or PolII. EAPP competes with Sp1 for the Sp binding sites and binds to Sp binding sites number 3 and 4. Cartoon generated by Peter Andorfer.

Increased EAPP levels protected cells from apoptosis. This effect could be hindered by RNAi mediated knockdown of p21. RNAi induced knockdown of EAPP on the other hand led to apoptosis (Andorfer, Rotheneder 2011).

In addition to p21 regulation also the monoamine oxidase (MAO) B gene is regulated by EAPP. EAPP down-regulates MAO B by repression of the MAO B core promotor by competition of the Sp1 binding site (Chen et al. 2011).

EAPP also plays a role in multidrug resistance as is can increase the expression of MDR1 via increased activation of the MDR1 promotor independently from E2F (Andorfer, Rotheneder 2013).

In 2011 the Rotheneder group proposed a model explaining the role of the observed interaction of Chk2 with EAPP, seen in Figure 12. After DNA damage ATM phosphorylates Chk2 at Thr68 leading to its full activation. P- Chk2 can now phosphorylate various substrates like p53 or E2F1 leading to apoptosis or cell cycle arrest. It could be shown that Chk2 forms a complex with EAPP which further recruits phosphatases like Wip1, PP2A or PP1 facilitating the dephosphorylation of P-Chk2 and disassembly of the complex (Andorfer et al. 2011).

Figure 12: Interaction of EAPP with P-Chk2. EAPP binds to phosphorylated Chk2 generated by ATM. This complex further recruits Wip1, PP2a and PP1, phosphatases which constantly remove the phosphate moiety. After dephosphorylation EAPP frees Chk2. Andorfer, P. et. al. Cell cycle p.2081, 2011.

It has also been shown that cells with lower amounts of EAPP show an impaired dephosphorylation of Chk2 compared to wt cells. EAPP over- expressing cells on the other hand exhibit a faster dephosphorylation

22 resulting in lower levels of active Chk2. This indicates a requirement of EAPP for Chk2 inactivation during check point recovery. Since the levels of EAPP also rise after DNA damage but delayed compared to P-Chk2 EAPP seems to prevent premature inactivation of P-Chk2 (Andorfer et al. 2011).

5.5 CRISPR/Cas9

For protection against bacteriophages bacteria and also archaea have evolved a RNA mediated adaptive defence system called clustered interspaced palindromic repeats (CRISPR) / CRISPR associated nucleases (Cas) protecting against viruses and conjugative plasmids silencing their nucleic acid (Marraffini, Sontheimer 2010).

After Jinek et al. firstly described the CRISPR/Cas system as a method for genome engineering in 2012 (Jinek et al. 2012) this system has developed into a widely used method allowing easy, affordable and precise genome editing with a relatively wide application range. With the help of the RNA guided endonuclease Cas9 any location of the genome can be targeted through the selection of a short RNA guide (Hsu et al. 2014).

CRISPR loci always show a certain structure, observable in Figure 13. After a leader sequence a hundreds of base pairs long AT-rich repeat cluster with intraspecies conservation can be found. The numbers of repeats differ from at least two to a few hundred increasing with every infection. Each of them has a length of 23 to 50 nucleotides and is separated by a non-repetitive spacer sharing sequence identity with plasmid or phage fragments. Directly followed by the repeats are Cas genes encoding the proteins responsible for the activity of CRISPR (Marraffini, Sontheimer 2010).

Figure 13: Structure of CRISPR loci. CRISPRs are carrying an AT rich leader followed by different numbers of repeats (min. of two up to hundreds) each between 23 and 50 nucleotides long. They are separated by spacers carrying sequences of bacteriophage‟s or plasmids specifying the targets of CRISPR interference. In the end conserved Cas genes encode for proteins necessary for CRISPR activity. Marraffini, L., Sontheimer, E. Nat Rev Genet. p18, 2010.

Cas genes get directly translated into proteins, whereas CRISPR repetitive arrays including the viral DNA fragments in the spacers are transcribed into long RNAs which are then further processed into short CRISPR RNAs (crRNAs) directing the Cas enzyme activity for nucleic acid degradation (Hsu et al. 2014).

There are three types of CRISPR systems (I-III) known. Type I and type III CRSIPR loci encode for numerous Cas genes whose proteins are known for complexation with crRNA to enable the targeting and cleavage of foreign nucleic acids (Hsu et al. 2014).

Type II CRISPR Cas system being the one of choice for genome editing consists of much less Cas proteins and differs completely (Hsu et al. 2014). The RNP complex consists of the endonuclease Cas9, the crRNAs encoding for the guide RNAs (gRNA) and a trans-activating CRISPR RNA (tracrRNA) mediating interference. After loading of the gRNA it comes to a conformational change forming a channel binding the target DNA. The complex scans the DNA for the correct protospacer adjacent motif (PAM). The interaction with PAM is crucial for binding to the DNA and initiation of R- loop formation (van der Oost et al. 2014). Cas9 binds to any DNA sequences whose DNA is complementary to the guide RNA and the PAM in case of the Cas9 of S. pyogenes the 5`-NGG 3´of the target sequence and cleaves around 3bp upstream of it (Anders et al. 2014).

After the DSB either non-homologous end joining (NHEJ) or homologous recombination (HR) is performed. Both outcomes are used depending on the wanted outcome. NHEJ is performed to generate knock outs as NHEJ leads to small indel (insertions or deletions) mutations which can cause frameshifts and therefore a premature termination. HR occurs much less frequent and therefore the cell has to be driven into HR in the case of a donor DNA is to be inserted. The repair template with homology arms is then easily inserted (Ran et al. 2013).

Another approach on genome editing with CRISPR/Cas9 is CRISPR interference (CRISPRi). With a catalytically dead Cas9 the expression of

25 target genes can be repressed via interference with RNA polymerase or transcription factor binding. With CRISPRi it is possible to target various genes at the same time in a reversible manner (Qi et al. 2013).

5.6 Aims

The aims of this Master´s thesis project were the evaluation whether EAPP is necessary in NIH 3T3 cells. As it could be shown that a knockout of EAPP is lethal in U2OS cells it was to be examined if it is also essential in NIH 3T3 cells and if so, which parts of it are necessary.

For evaluation if one can rescue the knockout or which parts of it are obligatory cells with a total knockout of EAPP as well as cells obtaining truncations of EAPP were to be generated and investigated. These various truncations could also be used for determination of the functional domains of EAPP, their impact on stability or interaction with other proteins.

6 Materials and Methods

6. 1 Materials

Cell lines

 HAP1  NIH 3T3  U2OS  HAP1 EAPP KO Exon 3 (CRISPR/Cas9 introduced frameshift in Exon 3 of the EAPP gene resulting in loss of EAPP protein)  HAP1 EAPP KO Exon 6 (CRISPR/Cas9 introduced frameshift in Exon 3 of the EAPP gene resulting in expression of truncated EAPP protein in a reduced amount)  U2OS HA EAPP wt

Sequence:MNRLPDDYDPYAVEEPSDEEPALSSSEDEVDVLLHGTPDQKRKLIRECLTGESESSSEDEFE

KEMEAELNSTMKTMEDKLSSLGTGSSSGNGKVATAPTRYYDDIYFDSDSEDEDRAVQVTKKKKKKQHKIPT NDELLYDPEKDNRDQAWVDAQRRGYHGLGPQRSREQQPVPNSDAVLNCPACMTTLCLDCQRHESYKTQYR AMFVMNCSINKEEVLRYKASENRKKRRVHKKMRSNQEDAAEKAETDVEEIYHPVMCTECSTEVAVYDKDEV FHFFNVLASHS.  U2OS HA EAPP pm -some phosphorylatable Serines and Threonines replaced by Alanines.

Sequence:MNRLPDDYDPYAVEEPADEEPALAAAEDEVDVLLHGAPDQKRKLIRECLAGEAEAAAEDEF

EKEMEAELNATMKTMEDKLSSLGTGSSSGNGKVATAPTRYYDDIYFDADAEDEDRAVQVTKKKKKKQHKIP TNDELLYDPEKDNRDQAWVDAQRRGYHGLGPQRSREQQPVPNSDAVLNCPACMTTLCLDCQRHESYKTQY RAMFVMNCSINKEEVLRYKASENRKKRRVHKKMRSNQEDAAEKAETDVEEIYHPVMCTECSTEVAVYDKDE VFHFFNVLASHS  U2OS ps EAPP (constitutively expressing an siRNA resulting in decreased EAPP protein amount)  U2OS HA mEAPP(J) (knockout of endogenous EAPP and introduction of murine cDNA)

Established cell lines (production further explained in methods section)

 U2OS with EAPP 10-240  U2OS with EAPP 20-240  U2OS p53 KO clone C  U2OS p53 KO clone C + HA p53

 NIH 3T3 Clone B, E, G, P, Y, BA, BC, BF, BN, BQ, BR, BS, BT, BX, CI, IIB

Explanation of further used clones:

o NIH 3T3 Clone B: Western blot analysis shows no endogenous EAPP a deletion of 23 nucleotides resulting in the loss of its start codon. Instead of endogenous EAPP a truncation version (either 95-c or 135-c) is observable. o NIH 3T3 Clone E: The endogenous EAPP seems rather normal but sequencing revealed D7G missense mutation in one allele. In addition a 55-c truncated EAPP version is observable in clone E. o NIH 3T3 Clone Y: Clone Y shows no EAPP protein caused by a frameshift after Q5 due to a deletion of 2 nucleotides o NIH 3T3 Clone 4J: 4J shows as clone Y no EAPP protein.

Media

 DMEM high glucose with 10% FCS, 1% Penicillin/Streptomycin and 1% GlutaMAX Gibco  DMEM high glucose with 0,2% FCS, 1% Penicillin/Streptomycin and 1% GlutaMAX Gibco  DMEM high glucose with 1% FCS, 1% Penicillin/Streptomycin and 1% GlutaMAX Gibco  IMDM with 10% FCS, 1% Penicillin/Streptomycin and 1% GlutaMAX Gibco

Antibodies

 AAR2 C20 ORF (Y18) sc-85419 rabbit polyclonal IgG Antibody 100µg/µl - Santa Cruz Biotechnologies

 Akt 1 (B-1) sc-5298 mouse monoclonal IgG1 Antibody 200µg/ml – Santa Cruz Biotechnologies  Phospho-Akt (T308) 244F9 rabbit monoclonal Antibody #4056 Cell Signalling

 Phospho-Akt (S473) D9E XPR rabbit monoclonal Antibody #4060 Cell Signalling

 CDK1 = Cdc2 p34 (17) sc-54 mouse monoclonal IgG2a 200µg/ml– Santa Cruz Biotechnologies

 CK2 casein Kinase IIα (D-10) sc-365762 mouse monoclonal IgG2b 200µg/ml–Santa Cruz Biotechnologies

 E6AP (E-4) sc-166689 mouse monoclonal IgG2b 200µg/ml–Santa Cruz Biotechnologies  EAPP Antibody generated by 1E4 cells  GAPDH mouse monoclonal Antibody – Gift from Egon Orgis

 GSK 3α/β (0011-A) sc-7291 mouse monoclonal IgG2a 200µg/ml– Santa Cruz Biotechnologies  HA 16B12 Antibody for immunostaining  HA 12CR5 Antibody for immunoprecipitation

 Helic2 (G-9) sc-393170 mouse monoclonal IgG1 200µg/ml–Santa Cruz Biotechnologies

 HUR (3A2) sc-5261 mouse monoclonal IgG1 200µg/ml - Santa Cruz Biotechnologies

 JAB1 (B7) sc-13157 mouse monoclonal IgG2b 200µg/ml - Santa Cruz Biotechnologies

 P53 (DO1) sc-126 monoclonal IgG2a 200µg/ml - Santa Cruz Biotechnologies  PTEN (A2B1) sc-7974 mouse monoclonal IgG 200µg/ml - Santa Cruz Biotechnologies

 PARP1 (F2) sc-8007 mouse monoclonal IgG2a 200µg/ml–Santa Cruz Biotechnologies

 Pin1 (G-8) sc-46660 mouse monoclonal IgG2a 200µg/ml - Santa Cruz Biotechnologies  Phospho-Pras (T246) c77D7 #2997P Cell Signalling  pRb (IF8) sc-102 mouse antibody 200µg/µl –Santa Cruz Biotechnologies

 PRP8 (E5) sc-55533 mouse monoclonal IgG1 200µg/ml - Santa Cruz Biotechnologies

 Sm B/B´/N (A10) sc-271094 mouse monoclonal IgG2a 200µg/ml– Santa Cruz Biotechnologies  YB1 rabbit polyclonal Ab produced and gifted by Franz Wohlrab Lab

Used expression vectors: pCI-neo expression vector from Clontech

 pCI-neo-HA-EAPP wt  pCI-neo-HA-EAPP wt maxi  pCI-neo-HA-EAPP 1-120  pCI-neo-HA-EAPP 1-140  pCI-neo-HA-EAPP 1-180  pCI-neo-HA-EAPP 1-240  pCI-neo-HA-EAPP 10-240  pCI-neo-HA-EAPP 20-240  pCI-neo-HA-EAPP 55-c  pCI-neo-HA-EAPP 95-c  pCI-neo-HA-EAPP 135-c  pCI-neo-HA-EAPP 175-c  pCI-neo-HA-EAPP pm  pCI-neo-HA-EAPP mEAPP(J)  pCI-neo-myc-EAPP  pCl-neo-HA-p53

Oligonucleotides – all primer bought at Microsynth the Swiss DNA company AG

 HA EAPP 10-285 5´GATGAATTCCCCTACGCGGTTGA  HA EAPP 20-285 5´GATGAATTCGAGCCGGCTTTGAG  mEAPP seq1 sense 5´GGAGTCTCCGGGAAGCTTTG  mEAPP seq1 as 5ÁTCTAAGGCTTCACGCCTGG  mEAPP seq8 sense 5ĆGGAGTACACAGACGCCG  mEAPP seq8 as 5ĆGTAATTCCGGAACATCCCTCT  rev. primer nr. 5 5ÁTGAGGTCTAACCAGGAATAGCGGCCGCTAAG 3´  mEAPP Ex1 sense(4) 5´-CAC CGG TGG AAG AGC CGA GCG ACG-3´  mEAPP Ex1 as(4) 5´-AAA CCG TCG CTC GGC TCT TCC ACC-3´

 mEAPP Ex1 sense(5) 5´- CAC CCG TAC GGG TCA TAG TCA TCC -3´  mEAPP Ex1 as(5) 5´- AAA CGG ATG ACT ATG ACC CGT ACG -3´

Treatments

 Cycloheximide 10-30µg/ml final concentration  Etoposide 0.1-10µM final concentration  ATM inhibitor Ku 55933 10µM final concentration  Nocodazole 1µM final concentration  Hydroxyurea 2mM final concentration  TSA 100ng/ml final concentration  TBH 4µM final concentration

Used assays or kits

 CloneJET PCR Cloning” kit from ThermoFisher scientific  EXTRACTme DNA tissue kit from Blirt S.A.  Perfectprep ® Gel Cleanup kit from Eppendorf  Monarch® PCR & DNA Cleanup Kit from New England BioLabs  Monarch® Plasmid Miniprep Kit from New England BioLabs  Perfect prep plasmid mini kit from Eppendorf  Perfect prep plasmid midi kit from Eppendorf

®  Cell Titer 96 AQueous One Solution Cell Proliferation Assay – Promega  Trans-Blot® Turbo™ Mini Nitrocellulose Transfer Packs  Trans-Blot® Turbo™ RTA Midi Nitrocellulose Transfer Kit

Additional materials except standard consumables

 E. coli DH5α competent made after Nishimura et. al protocol  CRISPR/Cas9 PX459 vector: pSpCas9(BB)-2A-Puro (PX459) V2.0  Single guide RNA 4: GGTGGAAGAGCCAGCACG AGG - nr. of target sites: 5 (4 in genes)  Single guide RNA 5: CGTACGGGTCATAGTCATCC TGG - nr. of target sites: 1 (1 in genes)  PI-RNase solution – 50µg/ml PI +100µg/ml RNase Type I-A in PBS  X Ray films – FUJI medical X-RAY Film 100 NIF 18X24

 Protein marker Precision Plus ProteinTM Standards all blue Catalog nr#161-0373 from BIO-RAD  DNA marker Lambda HindIII cn. SM0102 from Thermofischer scientific

Concentrations of buffers and other used consumables

 Huntbuffer (pH 8.0) o 20mM TRIS o 100mM NaCl o 1mM EDTA o 0.5% NP-40 o 1mM DTT o 1mM PMSF o 50mM NaF

 Huntbuffer mix o 1/100 of 2mM Sodium Vanadate o 1/100 cOmplete ( 1 cOmplete™ Mini Protease Inhibitor Cocktail ROCHE Tablet diluted in 500µl huntbuffer) o 1/100 1M Butyrate o 1/500 500nM PMSF o With Huntbuffer to 100%

 TBS (10X) o 121.1g TRIS o 175.2g NaCl

o With H2O to 2 liter (with HCL to pH 7.5)

 TBS-T (1X) o 200ml 10X TBS

o 1800ml H2O o 4ml Tween 20

 5% blocking solution o 300ml TBS-T o 15g milk powder o 200µl sodium azide

 PAA Stacking gel o 600µl Acrylamide (19/1)

o 2.86ml H2O o 500µl TRIS pH 6.8 o 20µl SDS (10%) o 4µl TEMED o 20µl APS

 PAA Running gels

10% Gels 12% Gels 15% Gels

Acrylamide (19/1) 2.4ml 2.88ml 3.6ml (Gerbu biotech)

H2O 2.5ml 2.02ml 1.3ml

TRIS pH 8.7 3ml 3ml 3ml

SDS (10%) 40µl 40µl 40µl

TEMED 8µl 8µl 8µl

APS (20%) 40µl 40µl 40µl

 Electrophoresis running buffer o 60.5g TRIS o 288g Glycine o 20g SDS

o With H2O to 2 liter

 Transfer buffer for Tank Blot (buffer for semi dry from Trans-Blot® Turbo™ RTA Midi Nitrocellulose Transfer Kit) o 1.68l EtOH o 115.62g Glycin o 24.2g TRIS

o With H2O to 8 liter

Equipment

In addition to standard equipment the following instruments were used

 Nanodrop  Victor3V 1420 multilabel counter  Curix 60 AGFA X-ray Film developer  Trans-Blot® Turbo™ Transfer System- Biorad

Programs

 1420 software v 3.4 of Victor3V 1420 multilabel counter  Adobe Photoshop  Adobe Illustrator

6.2 Methods

Cell culture

All experiments shown in this thesis were performed with U2OS, NIH 3T3 and HAP1 cells with different knockouts and stable as well as transient transfections of EAPP as well as p53. U2OS and NIH cells were cultured in DMEM with 10% FSC, 1% Penicillin/Streptomycin and 1% GlutaMAX Gibco. HAP1 were cultured in IMDM with 10% FCS, 1% Penicillin/Streptomycin and

Gibco 1% GlutaMAX all at 37°C and 7.5% CO2 as well as 90.0% rH.

To freeze cells the culture medium was aspirated, the cells were washed with PBS and trypsinized. The cells were resuspended in 1ml medium and transferred to a cryotube containing 1/10 DMSO. After cooling on ice the cells were frozen at -80°C. After complete freezing the cells were transferred into liquid nitrogen for long term storage. To thaw cells the cryotube was held into a water bath until only an ice core was observable. The cells were transferred into a cell culture dish with preheated medium and further medium was used to melt the resisting ice.

Protein extraction

All steps were performed on ice. The medium of the cells in the cell culture dish was aspirated and 1ml PBS was added. Afterwards the cells were scraped of the dish and transferred to a 1.5ml Eppendorf tube. The cells were spinned down for 5 seconds in a microcentrifuge. Afterwards the supernatant was aspirated and the pellet was re-suspended in 10 times the amount of Huntbuffer-mix in comparison to the pellet size and frozen at - 80°C. The lysate was re-thawed and centrifuged auf 25.000 rcf for 10min at 2°C. The supernatant was transferred to a fresh Eppendorf tube and was used for concentration determination or analysis via Bradford assay or Nanodrop. The protein extracts were then further used for Western blot analysis and the rest stored at -80°C.

Western blotting

Gel electrophoresis

The amount of protein – most of the time 20µg, was mixed with 10µl of sample buffer and shacked at 95°C for 4min on an Eppendorf Thermomix. After cooling on ice the extracts were spinned down for 5 seconds in a microcentrifuge and then frozen at -20°C until further use or instantly applied on a PAA gel and run on 110V until the loading dye run out.

Western blot

Tank blot

The gel was transferred into a tank blot gel electrophoresis apparatus and transferred at 250mA for 2h at 4°C

Semi dry

The gel was transferred into the semi dry gel electrophoresis apparatus and transferred at 1.3A 25V for 10min with the Biorad 1.5mm program.

After transfer check with Ponceau staining the blots were ready for immunodetection.

Immunodetection

After blocking with 5% blocking solution for at least 30 minutes blots for phospho-antibodies were washed for 5 minutes with TBS-T and then treated with primary antibody overnight. Regular antibodies could be applied directly after blocking without further washing. The next day the blots were washed 3 times with TBS-T for 10 minutes and treated with secondary antibodies for at least 1 hour. After 3 washing steps with TBS-T and treatment with ECL detection solution the blots could be analysed with X- ray films and the use of Curix 60 AGFA developing machine.

Generation of NIH 3T3 cells with knockouts of EAPP or introduction of truncated EAPP versions

For transfection a total of 40µg DNA was used consisting of 4µg of each of the following versions of EAPP: 1-120, 1-140, 1-180, 1-240, 55-c, 95-c,

135-c, 175-c, pm and myc EAPP. (Cartoon in results part) The DNA was mixed with HBS and this solution was mixed with Polyethlyenimine (PEI) mix (PEI (3 times the amount of DNA) + HBS) vortexing between every step, and after 10 minutes of incubation time applied on the cell culture medium of NIH 3T3 cells. After selection for stable integration with G418 the cells were transfected with 20µg single guide RNA 4 and 5 alone or in combination also via PEI. Afterwards the cells were selected with puromycin and single cell clones were generated. The protein status of the clones was examined via Western blot analysis and the DNA was examined via sequencing. Therefore the DNA was amplified via PCR using either seq1 or seq8 primer. The PCR products were applied on an agarose gel, the bands were cut out and the DNA eluted with the “perfectprep Gel Cleanup” kit from Eppendorf, or the “PCR and DNA Cleanup Kit” from NEB was applied. The elutes were sent in for sequencing. Additionally the PCR product was ligated into a pCl-neo vector and transferred into competent bacteria (E.coli DH5α). The DNA of single clones was obtained by Miniprep Kit from Eppendorf or NEB and the generated DNA examined via sequencing by mycrosynth.

Generation of U2OS cells with EAPP 10-240 or 20-240 in addition to endogenous EAPP

For the generation of 10-240 or 20-240 truncations a PCR with either a forward primer starting at the nucleotides corresponding to amino acid 10 or 20 and the reverse primer nr. 5 was performed with pCl-neo-HA-EAPP wt DNA. The generated PCR product was cleaned up with the “PCR and DNA Cleanup” Kit from NEB and sequenced for first impression. Afterwards 10- 240, 20-240 as well as a pCl neo vector were cut with EcoRI and NotI HF and the use of buffer nr. 2 and applied to an agarose gel. After cut out the DNA was eluted from the gel with “perfectprep Gel Cleanup” from Eppendorf and the truncations were ligated into the vector with the “CloneJET PCR Cloning” kit from ThermoFisher scientific. The transfer into and growth of single colony bacteria was performed as described in Vector ligation and amplification in bacteria. After check of the DNA the bacteria were

38 produced in large scale and a midiprep with an Eppendorf kit was performed. 11µg DNA of each of the generated 10-240 of 20-240 truncations was transfected into U2OS wt cells via PEI transfection after the same scheme as already explained. After selection with G418 single cell colonies were established and the protein expression examined via Western blot analysis.

DNA extraction from mammalian cells

Performed either with “EXTRACTme DNA tissue kit” from Blirt S.A. after its protocol

300µl of water was added to the cell pellet, and heated for 15min to 95°C. After chilling on ice 4µl of RNase A was added and after 5min at 37°C 25µl of proteinase K was added. This solution was heated up again for 15min at 95°C and chilled on ice.

Vector ligation and amplification in bacteria

Performed either with Ligation via “CloneJET PCR Cloning Kit” from ThermoFisher Scientific after its protocol

 10µl of Insert DNA  7µl pCl-neo vector  2µl ligation buffer (#B0202S new England biolabs)  1µl T4 ligase

Stand on RT for 2 hours before application. 5-10µl of the ligation mix was pipetted into 100µl of competent bacteria. After 30 minutes on ice a heat shock of 42°C for 2min was performed. After cooling on ice, 900µl of LB medium was added to the bacteria. After 2h at 37°C the solution was applied to ampicilin-agar plates and stored at 37°C. For the analysis of single clones a bit of a colony was transferred into 5ml of LB-Amp medium

39 and shacked overnight at 37°C. The generated solutions were ready for DNA preparation.

Generation of U2OS cells with p53 knockout and re-transfection of p53

The knock out was performed with 1µg of a p53 CRISPR/Cas9 KO plasmid from Santa Cruz Biotechnology (Figure 14) via PEI transfection in U2OS cells in a 6cm cell culture dish. In addition a co- transfection with a 1µg of a p53 homology directed DNA repair (HDR) plasmid from Santa Cruz Biotechnology (Figure 15) was Figure 14: Structure of p53 CRISPR/Cas9 performed for repair of the site knockout plasmid. Scheme of used specific cleavage of the DNA induced knockout plasmid used for knockout of p53 in U2OS wt cells. A pool of 3 plasmids encoding by Cas9 via homologous for a Cas9 nuclease as well as different specific 20 nt long gRNAs are used causing a recombination and the introduction DSB in a 5´exon within the human TP53 of a puromycin resistance. After gene. Incorporation of GFP for visual transfection verification. p53 CRISPR/Cas9 KO puromycin selection a single cell Plasmid (h): sc-416469 Santa Cruz selection was performed and the cells Biotechnology. were examined via Western blot analysis. 10µg of HA p53 was then transfected into the U2OS p53 KO clone C with a stable knockout of p53 via PEI transfection. After selection with G418 the protein status was examined via Western blot analysis. Figure 15: Structure of p53 HDR plasmid. Scheme of used HDR plasmid used for site specific cleavage in p53 knockout. A pool of 2- 3 plasmids containing HDR templates each consisting of two 800 bp long homology arms binding to the TP53 gene in the surrounding of the DSB caused by the p53 CRISPR/Cas9 knockout plasmid. Incorporation of RFP for visual transfection verification.p35 HDR Plasmid (h): sc-416469-HDR Santa Cruz Biotechnology. 40

Generation of CRISPR/Cas9 sg4 and sg5 EAPP KO vectors

For the generation of sg4 and 5 vectors the 30µl of antisense and sense oligos bought from Mycrosynth were hybridized resulting in a concentration of 50pmol/µl. After spinning down the samples were brought to 96°C and slowly cooled down. In the meantime the pX-459 vector, seen in Figure 16, was cut with BbsI and the DNA was eluted with

“perfect prep gel cleanup” from Figure 16: Structure of CRISPR/Cas9 PX459 Eppendorf after separation on an vector. pSpCas9(BB)-2A-Puro (PX459) V2.0 vector used for EAPP knockout containing a Cas9 agarose gel. 5µl of the vector nuclease as well as a puromycin and ampicillin resistance. It contains a BbsI cutting site. were ligated with 1µl of either sg4 [online] available at: of 5 as described in Vector https://www.addgene.org/62988/ last accessed 11.01.2017 ligation and amplification in bacteria and the DNA of single colony bacteria was ready to use after preparation with the “Perfect prep plasmid midi kit” from Eppendorf and positive sequencing results.

Cell proliferation assay

® Cell Proliferation assays were performed with “Cell Titer 96 AQueous One Solution Cell Proliferation Assay” from Promega. For each cell line of interest 16 wells of a 96 well Plate for each day of observation was seeded with 3,000 or 5,000 cells in 100µl cell culture medium. The day after seeding

® 20µl of “Cell Titer 96 AQueous One solution Reagent” was pipetted into each well. After 2 or 3 hours at 37°C in the dark the absorbance at 490nm was measured by the 96- well Plate reader VIKTOR3V 1420 multilabel reader.

Serum starvation – cell cycle arrest

For serum starvation the media of the cells was aspirated and replaced by DMEM high glucose with 0,2% FSC, 1% Penicillin/Streptomycin and 1% GlutaMAX Gibco instead of 10% FCS. After 48h of serum deprivation the cells were re-stimulated for either 16h or 24h. After every indicated time point cells were harvested for FACS analysis ad described in FACS.

FACS

For FACS analysis the cells of interest were washed, trypsinized, resuspended in 5ml medium and transferred to a 15ml falcon tube. After centrifugation for 5min at 1200rpm the supernatant was aspirated, the cells were resuspended in 5ml PBS and re-centrifuged. The generated pellet was resuspended in 1ml PBS. 0.5ml were used for general protein extraction and treated as already explained. The other 0.5ml used for the FACS were mixed with 0.5ml of cold 96% EtOH while vortexing. These fractions were stored at -20°C until the cells from each time point were collected. After the samples were centrifuged for 7min at 100rpm the EtOH was decanted and the pellet resuspended in 500µl PI mix containing 50µg/ml PI and 10µg/ml RNase in PBS. After 20min storage in the dark the samples got analysed.

7 Results

7.1 Generation of cell lines with different version of EAPP

As EAPP is differently expressed in tissues and tumours its role was examined in different cell lines and not only one. As the total knockout of EAPP was proven to be lethal in human U2OS cells the knockout was also performed in murine NIH 3T3 cells. The fact that the knockout in this cell line results in viable cells again underlines the differences in the dependence or role of EAPP in these cells.

7.1.1 Generation of NIH 3T3 cells with knockouts of EAPP or introduction of truncated EAPP versions.

Truncated versions of EAPP are a useful tool to identify the functional domains of EAPP. As a knockout of EAPP has shown to be lethal for U2OS cells the NIH cells used were transfected with expression vectors coding for truncations of human EAPP followed by a knockout of the endogenous EAPP. In that way the newly introduced EAPP has the chance to rescue the cell.

The first step was the transfection via PEI of 10 different truncations of human EAPP into NIH 3T3 cells, observable in Figure 17.

Figure 17: Cartoon of EAPP versions used for transfection. Scheme showing the length of EAPP protein including PEST sequence containing either HA tag or myc tag. HA EAPP pm showing mutations altering phosphorylatable Serines and Threonines into Alanines.

After selection for stable integration a knockout of the endogenous EAPP was performed using CRISPR/Cas9 system either with sgRNA4 and 5 combined or alone. After selection with puromycin colonies resulting from single cells were isolated and analysed via Western blotting. As seen in Figure 18, Western blot analysis revealed many different variants including complete loss of endogenous EAPP without additional EAPP indicating that EAPP is not essential in NIH cells as it is in U2OS cells. Clones with a truncated EAPP version in addition to the endogenous EAPP were also observable.

Figure 18: Overview of single cell clones after transfection and knockout. Western blot analysis of some of the generated NIH single cell cones after transfection and knockout of EAPP. 4 or 5 in the clone name indicate clones generated with sg4 or sg5 respectively. Clones with no number were generated by knockout with both sg4 and sg5 showing expression patterns of EAPP and HA. [EAPP(1E4), HA (16B12)]

Clone B for example shows no protein expression at the height of endogenous EAPP but a protein band much lower than the endogenous one. Further analysis revealed an insertion of either 95-c or 135-c. 4J or Y for example show in contrast protein expression of neither endogenous EAPP nor a truncated ectopic version.

The interesting clones were sequenced afterwards to further investigate their alterations in the EAPP gene. The DNA was extracted and a PCR was performed with seq8 or seq1 primer pair. The amplified DNA was loaded on an agarose gel and the band excised and purified. The eluted DNA was sent to an external company (Mycrosynth) for sequencing. For allele specific results the PCR products were cloned into a suitable vector. The DNA of single clone bacteria was then sequenced. The detailed protocol is shown in the methods section.

This list of some of the sequencing results shows clearly how differing the results of a knockout with CRISPR/Cas9 can be.

In Figure 19, Clone B for example shows a deletion of 23 nucleotides resulting in the loss of its start codon. Another deletion of 2 nucleotides further downstream could also be examined. The deletion of the translational start codon results in the complete loss of EAPP expression from this allele. This fits very well to the result of the Western blot analysis.

Figure 19: Sequence alignment of EAPP with EAPP clone B. Analysis of alterations of the EAPP gene of clone B (lower line) compared to the wt (upper line) showing a deletion of 23 nucleotides and a second deletion of two nucleotides.

Clone E shows a D7G missense mutation in one allele, observable in Figure 20, but as it seemed rather normal in Western blot analysis, seen in Figure 18 this seems to have only little impact.

Figure 20: Sequence alignment of EAPP with EAPP clone E. Analysis of alterations of the EAPP gene of clone E (upper line) compared to the wt (lower line) as well as its effect on the amino acid level (cone E lower line) revealing a D7G missense mutation.

Clone G depicted in Figure 21 shows an insertion of one Alanine leading to a frameshift in one allele. The second allele shows a D6 frameshift due to deletion of 4 nucleotides but a second deletion of 2 nucleotides leading to replacement of amino acids 7-18 by 10 missense amino acids restores the original reading frame.

Clone G allele 1

Clone G allele 2

Figure 21: Sequence alignment of EAPP with EAPP clone G. Analysis of alterations of the EAPP gene of clone G. Insertion of one Alanine in allele 1 leading to a frameshift. Comparison of wt (lower line) with allele 2 of clone G (upper line) as well as its effect on the amino acid level (clone G lower line) revealing a D6 frameshift restored by a second deletion of 2 nucleotides in allele 2.

As shown in Figure 22, one allele of clone Y shows a frameshift after Q5 due to a deletion of 2 nucleotides. In addition a deletion of 2 other nucleotides further downstream is also observable.

Figure 22: Sequence alignment of EAPP with EAPP clone Y. Analysis of alterations of the EAPP gene of clone Y (upper line) compared to the wt (lower line) revealing a frameshift after Q5. 46

The sequence of BC EAPP shown in Figure 23 shows a Q5 frameshift generated by deletion of one nucleotide. This frameshift is compensated by an insertion of one nucleotide restoring the original reading frame after amino acid 17.

Figure 23: Sequence alignment of EAPP with EAPP clone BC. Analysis of alterations of the EAPP gene of clone BC (upper line) compared to the wt (lower line) as well as its effect on the amino acid level (clone BC lower line) revealing a Q5 frameshift later re-corrected.

Clone BR depicted in Figure 24 shows in one allele a deletion of 24 nucleotides after P10 resulting in a loss of 8 amino acids but no frameshift.

Figure 24: Sequence alignment of EAPP with EAPP clone BR. Analysis of alterations of the EAPP gene of clone BR (upper line) compared to the wt (lower line) as well as its effect on the amino acid level (clone BR lower line) showing a del of 24 nucleotides after P10.

7.1.2 Generation of U2OS cells expressing HA-EAPP 10-240 or 20- 240 in addition to endogenous EAPP

As EAPP contains a PEST region in its N-terminus leading to increased degradation (Rechsteiner, Rogers 1996) we wanted to establish cells with EAPP versions with increased stability due to a lack of the first 10 or 20 amino acids (aa). For the generation of U2OS cells expressing truncated HA-tagged EAPP (either aa 10-240 or 20-240) in addition to the endogenous version the respective truncated cDNAs had to be generated in the first line.

A PCR with a forward primer starting at aa 10 or 20 and a reverse primer from aa 240 was performed with a plasmid comprising the full length EAPP cDNA. The generated PCR products were purified and sequenced. Afterwards 10-240, 20-240 PCR products as well as a pCl-neo vector were cut with EcoRI and NotI and separated by gel electrophoresis. After cut out the DNA fragments were eluted from the gel and ligated into the vector. After growth in bacteria the DNA of single clone bacteria were sequenced. For a stable integration in U2OS wt cells the cells were

Figure 25: Western Blot transfected with 11µg of vector DNA according to analysis of U2OS cells the PEI method. After selection with G418 with different transfections of EAPP. colonies resulting from single cells were isolated Western blot analysis and immunostaining of 20µg and analysed via Western blotting for protein of extracts after expression seen in Figure 25. The detailed transfection with 10-/20- 240 EAPP showing protocol is shown in the methods section. expression patterns of EAPP and HA. [EAPP(1E4), HA (16B12)]

7.1.3 Generation of U2OS cells with p53 knockout and re- transfection of p53

EAPP has been shown to interact with p53 (Rotheneder unpublished) and might therefore influence the activity of EAPP. The three cell lines examined in this project differ in their levels of p53. With NIH showing normal levels of p53, U2OS showing reduced levels and Hap1 cells expressing large amounts of a mutated and inactive form of p53 the influence of p53 on the possible results was to be examined and therefore a knockout cell line was generated.

The knockout was performed with a p53 CRISPR/Cas9 KO plasmid from Santa Cruz via PEI transfection. In addition a co-transfection with a p53 HDR plasmid was performed to allow homologous recombination for the repair of the site specific cleavage of the DNA induced by Cas9 as well as the introduction of a puromycin resistance. After puromycin selection a single cell selection was performed and the cells were examined via Western blot analysis, observable in Figure 26, revealing a total knockout of endogenous p53.

To ensure that the observed alterations in various experiments are caused by p53 knockout, p53 was re-transfected into the cells as verification control. After transfection of the plasmid via PEI and integration selection the expression was again examined via Western blot analysis, as seen in Figure 26: U2OS cells with p53 knockout Figure 26, showing high levles of and insertion of p53. Western blot analysis and immunodetection of U2OS wt, U2OS wt with p53 protein. stable p53 KO and U2OS wt with stable p53 KO but re-transfected with HA-p53 showing loss and re-insertion of p53. [p53 Ab (DO1)]

7.2 Cell proliferation assays

7.2.1 Cell proliferation of HAP1 cells with different EAPP knockouts

With the help of the “Solution Cell Proliferation Assay” of Promega as explained in the methods part, the growth pattern of HAP1 cells with CRISPR/Cas9 introduced frameshifts either in Exon 3 or Exon 6 of the EAPP gene compared to the wild type was examined. As observable in Figure 27 HAP1 cells with a frameshift in the EAPP Figure 28: Western Blot analysis of gene resulting in loss of EAPP show a HAP1 cells with different Knockouts of EAPP. Western blot analysis and massive increase in their rate of immunostaining of 20µg of EAPP of expansion with almost 3 times the HAP1 wt, HAP1 EAPP KO Exon 3 and HAP1 EAPP KO Exon 6 extracts showing absorbance in comparison to the wild EAPP expression patterns [EAPP(1E4)] type. Cells with a mutation in Exon 6 of the EAPP gene, which still produce low amounts of a short version of EAPP as seen in Figure 28, showed a weaker proliferation rate than the knockout cells of EAPP Exon 3 but still a much greater rate of expansion than the wild type.

Figure 27: Cell proliferation assay of HAP1 cells with different knockouts of EAPP. Analysis of cell proliferation with Promega Cell Titer 96 AQueous One Solution Proliferation Assay. For every time point and clone 18 wells of cells were cultured and examined 24h, 48h and 72h after seeding. HAP1 Exon 6 cells with a truncation of EAPP and HAP1 Exon 3 cells with no EAPP protein. 50

A relativization of the absorbance of day one, seen in Figure 29, to reduce the risk of the impact of different numbers of seeded cells on the result does not show any difference indicating that the growth pattern observed is generated by the different amounts of EAPP protein and not to different amounts of seeded cells.

Figure 29: Cell proliferation assay of HAP1 cells with different knockouts of EAPP. Analysis of cell proliferation with Promega Cell Titer 96 AQueous One Solution Proliferation Assay. For every time point and clone 18 wells of cells were cultured and examined 24h, 48h and 72h after seeding. Relativization of absorbance for better analysis of relatively growth.

7.2.2 Cell proliferation of NIH 3T3 cells with different EAPP versions

In addition the proliferation patterns of NIH cells with different versions of EAPP were examined.

In contrast to the HAP1 cells the NIH 3T3 cells do not show an EAPP dependent growth pattern as one can see in Figure 30. Clone B carrying a truncated EAPP protein instead of the endogenous EAPP shows greater relative growth compared to the wild type. The NIH clones Y and 4J both with a knockout of endogenous EAPP show differing patterns as clone Y shows a greater growth rate than the wt and 4J in contrast a slightly slower growth pattern than the wt indicating a clonal effect rather than an effect caused by EAPP.

Figure 30: Cell proliferation assay of NIH cells with different versions of EAPP. Analysis of cell proliferation with Promega Cell Titer 96 AQueous One Solution Proliferation Assay. For every time point and clone 18 wells of cells were cultured and examined 24h, 48h and 72h after seeding. Relativization of absorbance for better analysis of relatively growth

7.3 FACS analysis

7.3.1 FACS analysis of U2OS cells with different EAPP versions after treatment with different concentrations of Etoposide

For further analysis of the impact of different EAPP versions and expression levels on the DNA damage response U2OS cells with different EAPP versions were treated with increasing amounts of Etoposide for 24 hours and the impact on the cell cycle was examined via FACS analysis seen in Figure 31.

Figure 31: FACS analysis of U2OS cells with different EAPP versions after treatment with different concentrations of Etoposide. Treatment of U2OS wt, U2OS HA EAPP, mEAPP (J), ps EAPP and HA EAPP pm with 0.1µM to 3µM Etoposide for 24h followed by FACS analysis after Propidium Iodide DNA labelling and ethanol fixation.

Etoposide acts on cells in late S and G2 phase and is able to inhibit DNA synthesis (LIPMAN 1993). Etoposide induces DNA DSBs (Florey 1989) as it complexes with the DNA and topoisomerase II making re-ligation impossible (Pommier et al. 2010).

It could be observed that U2OS HA EAPP cells, containing more EAPP protein than the wt U2OS cells, showed with 61.66% of the cells a greater amount of cells in G1 phase compared to the wt with only 37.3% of the cells in G1, and much less cells in S phase after treatment with only 0.1µM Etoposide. With the same amount of Etoposide U2OS mEAPP (J) containing the murine instead of the human version showed also an increased fraction of the cells in G1 but no decrease in S phase as HA EAPP but in G2 phase compared to the wt. after treatment with 0.1µM Etoposide.

After treatment with 0.3µM Etoposide mEAPP (J) showed a prominent G1 peak with 58.97% of the cells compared to the wt with only 23.19% of the cells in G1. In addition the amount of cells in G2 phase is very much decreased from 64.41% in wt to 16.44%. U2OS ps EAPP containing less EAPP protein than the wt showed slightly decreased fractions in both G1 and S phase but more cells in G2 phase compared to the wt.

1µM Etoposide led with 93% of the cells to an almost complete stop of cell cycle progression in G2 phase and 6% in G1 and less that 1% in S phase in U2OS wt. This pattern is also observable in U2OS HA EAPP as well as HA EAPP pm with the exception that these cell lines do not show any S phase cells anymore. U2OS mEAPP (J) in comparison showed prominent peaks in both G1 and G2 with about 41% each and with over 18% a significant fraction of cells in S phase. Also U2OS ps EAPP cells showed a strongly different pattern than the wt but also than mEAPP (J). The majority of the cells are in G2 phase as the wt, but with almost 15% the amount of cells in G1 phase is more than doubled. In addition also the amount of cells in S phase is with 12.33% much greater than 0.98% in the wt.

After treatment of the cells with 3µM Etoposide the wt cells showed with over 25% of the cells a much greater fraction of cells in S phase compared

54 to 1µM of Etoposide. HA EAPP cells show almost the same percentage of cells in S phase but with almost 16% much more cells in G1 than the wt with less than 3%. mEAPP (J) cells show with over 22% even more cells in G1 but decreased fractions in both G2 and S phase. U2OS HA EAPP pm cells show with 83.36% over 10% more cells in G2 compared to the wt but at the same time almost 10% less in S phase and no cells in G1 phase. U2OS ps EAPP cells instead showed with 99.36% of the cells the most prominent G2 peak.

7.3.2 FACS analysis of NIH 3T3 versions after serum starvation and reactivation

For further analysis of cell cycle progression and reaction on serum deprivation NIH 3T3 wt as well as clones B and 4J were cultured in medium containing only 0.2% FCS instead of 10% for 48h. After that time the cells were re-induced for 16 or 24h with 10% FCS medium. After each time point cells were harvested for FACS analysis, seen in Figure 32. Analysis of NIH clone Y was not possible as cells were not viable after treatment with 0.2% or even 1% FCS medium and in that way showed the strongest alteration compared to the wt.

Figure 32: FACS analysis of NIH versions after serum starvation and reactivation. Serum starvation of NIH wt, clone B and 4J for 48 hours with 0.2% FCS medium followed by serum activation with 10% FCS medium for 16h or 24h followed by FACS analysis after Propidium Iodide DNA labelling and ethanol fixation.

In logarithmic and serum starved state there could not be significant alterations observed. After 16h of re-stimulation the fraction of 4J cells in S phase was with almost 55% much greater than in NIH wt cells with only 22.52% of the cells in S phase. The FACS pattern of NIH 4J after 16h looks much more like the NIH wt 24h than the 16h indicating a much faster response to serum induction. Also NIH clone B seem to contain much more S phase cells compared to the wt after 16 hours of re activation.

After 24h of serum stimulation clone B shows almost normal logarithmic pattern again, whereas the wt shows with over a third of the cells a great fraction in S phase.

7.4 Analysis of protein stability of EAPP versions in U2OS cells via treatment with Cycloheximide

For analysis of the stability of the different EAPP versions U2OS cells with different truncations of EAPP were treated with 10µg/ml Cycloheximide (CHX) for different durations. As CHX inhibits the translation (Obrig et al. 1971) the different protein levels after each time point, and therefore indirectly their stability, were examined.

It could be observed that the levels of endogenous EAPP in wt NIH as well as wt U2OS and all U2OS versions with an EAPP truncation showed a significant decrease over a timeframe of 32h, seen in Figure 33. This effect is also observable in HA EAPP wt levels.

In comparison the levels of murine mEAPP (J) in human U2OS cells seem to increase relatively to the total protein concentration over time. In addition also the HA EAPP pm and 55-c levels show an increase Figure 33: Western blot analysis of levels of EAPP versions after CHX treatment: EAPP levels of over the time period of different EAPP versions after treatment with 10µg/ml 32h. CHX for 8-32h analysed by Western blot analysis and immunostaining with EAPP Ab (1E4). Panels 1 and 2 show the endogenous EAPP the others show truncated or mutated versions of EAPP better described in the materials section. [EAPP(1E4), GAPDH]

A different pattern is observable in HA EAPP 1-180, 10-240 as well as 20- 240 where the proteins levels at first drop down but then relatively increase again. To further investigate this effect the protein levels were examined in a tighter timeframe during Cycloheximide treatment.

Figure 34 shows that EAPP 10-240 seems to decrease up to 8h of treatment but shows a significant relative increase again after 16h of treatment. This effect is not observable in HA EAPP wt at all where a consistent decrease of protein is observable.

Figure 34: Western blot analysis

of EAPP levels after CHX treatment: EAPP levels after treatment with 10µg/ml CHX for 8- 16h analysed by Western blotting analysis and immunostaining. [EAPP(1E4), GAPDH]

To exclude the risk of non-total protein synthesis block due to too little amount of CHX the experiment was re-performed with three times the amount of CHX. As seen in Figure 35 this does not change the expression pattern of neither HA EAPP wt nor HA EAPP 10-240 indicating that this is a specific result. Figure 35: Western blot analysis of EAPP levels after treatment with triple amount of CHX: EAPP levels after treatment with 30µg/ml CHX for 8-20h analysed by Western blotting analysis and immunostaining. [EAPP(1E4), GAPDH]

7.5 Alteration of protein expression of various proteins due to different EAPP expression

7.5.1 General alterations of protein expression influenced by Etoposide and ATMi in HAP1 cells with EAPP knockouts.

EAPP has been shown to be anti-apoptotic. This effect is to a certain extent caused by p21 but the total mode of action is still relatively unclear.

In this experiment the effects of Etoposide, causing DNA DSBs (Florey 1989) were examined. In addition to Etoposide the cells were also treated with or without ATMi for evaluation whether the effects caused by Etoposide are ATM independent or not. The experiments were carried out in HAP1 cells with EAPP frameshifts in Exon 3 or 6 to further investigate the way EAPP influences apoptosis.

Akt, a serine/threonine kinase plays an important role in the regulation of cell survival, cell cycle progression, apoptosis or cellular metabolism (Song et al. 2005) after its activation through phosphorylation. Components of its signalling pathway are frequently mutated (Fresno Vara et al. 2004).

Experiments examining the impact of EAPP on Akt were performed and Figure 36 shows very clear that P-Akt S473 is not visible under any tested conditions in HAP1 wt cells but clearly observable in both HAP1 EAPP KO Exon 3 and Exon 6. 10µM ATMi leads in Exon 3 KO as well as Exon 6 KO to a decrease of P-Akt no matter if treated alone or in combination with 10 µM Etoposide. Etoposide alone leads to a significant increase of P-Akt levels. This effect is abrogated if the cells are co treated with ATMi.

Figure 36: Protein expression influenced by Etoposide and ATMi in HAP1 cells with EAPP knockouts. Western blot analysis of HAP1 wt, EAPP KO Exon 3 and EAPP KO Exon 6 treated with 10µM ATMi and 10µM Etoposide for 24h for analysis of expression levels of following proteins. [Phospho-Akt S473 (D9E), PTEN, pRb, p53, PARP1, JAB1] 61

PTEN is a tumor suppressor which negatively regulates the Akt signalling pathway (Song et al. 2012) by dephosphorylation of the second messenger phosphatidylinositol 3,4,5-trisphosphate (Maehama, Dixon 1998). Its levels are drastically reduced after treatment with Etoposide or Etoposide in combination with ATMi in HAP1 wt cells. HAP1 cells with a knockout in Exon 3 or 6 instead show only a reduction in amount but still significant amount or protein remaining. In addition these cells show under normal conditions or after treatment with ATMi a second smaller band indicating cleavage. This band is not observable in HAP1 wt cells under the same conditions.

Also pRb shows no protein bands in HAP1 wt cells after Etoposide treatment with or without combination of ATMi. In comparison to that HAP1 EAPP KO Exon 3 and 6 show a decrease but no drastic reduction of pRb.

Western blot analysis of Parp1 revealed the presence of various versions or truncations of Parp1 as bands with different heights are observable. All three versions of HAP1 cells showed expression of the full length protein only under normal conditions or treatment with ATMi alone. After treatment with Etoposide or Etoposide and ATMi a smaller band at around 100 kDa is observable. These bands are much stronger in HAP1 wt extracts than the EAPP knockouts. The same treatment leads to disappearance of the full length protein. In addition to these signals bands at around 50 kDa and 20 kDa were observable in every extract. Whereas the HAP1 wt cells did not show any differences in the signal strength at 50kDa due to any treatment HAP1 Exon 3 show enhanced signals after Etoposide treatment either alone or in combination with ATMi. EAPP KO Exon 6 cells in comparison showed a decrease of intensity at that height after treatment with ATMi both alone and combined with Etoposide. At around 20 kDa strong bands are only visible in untreated HAP1 EAPP wt as well as KO Exon 6 but not in KO Exon 3. HAP1 wt cells are also the only ones showing a strong band at that height after treatment with ATMi.

Neither the different EAPP levels nor the different drugs had any impact on the Jab1 levels of the cells.

7.5.2 Influence of EAPP on the expression of ATMi-sensitive proteins in HAP1 and NIH 3T3 cells

As it was shown that EAPP seems to have impact on the protein levels of different proteins further analysis with additional antibodies and the NIH 3T3 cell line in addition was performed. This time the cells were treated only with ATMi (10µM).

As can be seen in Figure 37, the Hap1 cells show the same P-Akt pattern as already examined in the last experiment seen in Figure 36. NIH 3T3 cells instead show a different pattern as the wt cells also show strong expression. The only NIH clone showing an alteration to the wt is NIH clone 4J showing a reduction of Akt phosphorylation. But both cell lines show a reduction of P-Akt after treatment with ATMi.

The levels of Akt1 instead did not show any altered expression due to the amount of EAPP but NIH clone B shows a reduction of Akt1 after treatment with ATMi.

Figure 37: Protein expression influenced by ATMi in HAP1 and NIH 3T3 cells with EAPP knockout and knock-in. Western blot analysis of NIH 3T3 cells with EAPP knockout and knock-in treated with 10µM ATMi for 24h for analysis of expression levels of following proteins [Phospho-Akt S473 (D9E), Akt1, P-Pras, GSK 3α/β, GAPDH] 63

In HAP1 wt cells the levels of P-Pras are rather low. As P-Pras gets phosphorylated by Akt (Kovacina et al. 2003) this result is not surprising as the P-Akt levels in those cells are very low. The P-Pras amount is even further decreased by ATMi. HAP1 EAPP KO Exon 3 and 6 levels of P-Pras are multiple times higher than the wt. A slight decrease of P-Pras is observable only in Exon 6 but not in Exon 3, with no EAPP left, anymore. NIH cells also show a strong expression of P-Pras in the wt but also the other versions. The reduction of P-Pras by ATMi is observable in every clone but the strongest in clone Y and 4J expressing no EAPP and therefore the opposite result to the HAP1 cells in which a reduction is visible everywhere but the clone with no EAPP expression.

GSK 3α a target of Akt that gets inactivated by it via phosphorylation (Pap, Cooper 1998) seems to be slightly reduced in HAP1 Exon 3 and 6 as well as in NIH B, Y and 4J compared to their wild types. An altered expression of GSK 3β is not observable.

7.5.3. Influence of EAPP on P-Akt levels

The last experiments revealed a possible impact of EAPP on the phosphorylation of Akt P-Akt in Hap1 and NIH 3T3 cells. For further examination research was performed on the possible influence on U2OS cells.

To evaluate the impact of different levels of EAPP on the protein concentrations of Akt, transfections with different amount of HA EAPP DNA were performed in U2OS cells.

P-Akt seems to be influenced by EAPP levels, as can be seen in Figure 38. Extracts of cells transfected with 10µg of DNA show a much lighter band compared to the wt. The effects of 3µg of DNA are hard to examine as probably more protein extract was used indicated by GAPDH. Etoposide

Figure 38: Influence of EAPP Levels on P-Akt. Western blot analysis and immunostaining of U2OS wt cells transfected with different amounts of HA EAPP. Analysis of expression levels of following proteins [Phospho-Akt S473 (D9E), Akt1(B-1), EAPP (1E4), GAPDH]

65 alone does not seem to have influence in P-Akt levels but in combination with 10µM ATMi it comes to a strong reduction of the phosphorylation. The levels of Akt protein are not significantly altered due to the different HA EAPP amounts.

For further analysis the experiment was repeated but this time with different cell lines with stable mutations or transfections of EAPP.

It could be observed, as seen in Figure 39, that the strongest P-Akt T308 bands are observable in extracts obtained from U2OS ps EAPP cells containing the least amount of EAPP of those cells. The lowest signals instead could be detected in U2OS HA EAPP wt cells containing the highest amount of EAPP protein indicating a correlation between low EAPP levels and high P-Akt levels and vice versa.

Figure 39 Influence of EAPP Levels on P-Akt in stable cell lines. Western blot analysis and immunostaining of U2OS cells with different stable EAPP versions treated with 10µM ATMi and 10µM Etoposide. Analysis of their impact on P-Akt and Akt1. [Phospho-Akt S473 (D9E), Phospho-Akt T308 (244F9), Akt1(B-1), EAPP (1E4)]

The observation that P-Akt S473 of the first three cell types seems much stronger than the last three of Figure 39 is misleading, as the antibody was applied at two different time points indicating that the relative pattern is the same as the P-Akt T308 pattern.

In addition to P-Akt also Akt1 seems to be reduced in HA EAPP cells.

The levels of P-Akt in U2OS wt cells seem only reduced after treatment with both Etoposide and ATMi. Treatment with either one of these substances alone does not show any effect. Cells with a knockout in p53 also show a reduction of P-Akt after a combined treatment but also with Etoposide alone. Every other tested cell linage showed clear reduction after treatment with ATMi either alone or in combination with Etoposide.

Akt1 levels of U2OS wt seem to be increased after treatment with ATMi with or without Etoposide. This pattern was also observable in HA EAPP wt and pm cells. The murine mEAPP (J) as well as p53 KO C showed a strong decrease after treatment with Etoposide with or without combination with ATMi.

For further evaluation of the differences in P-Akt levels the experiment was re-performed, but this time only with ATMi.

It seems that the amount of P-Akt slightly increases in U2OS wt cells shows no alteration in strength whereas it seems to decrease in HA EAPP wt and mEAPP (J) cells after treatment with ATMi. The Akt1 levels in comparison show an increase in HA EAPP and mEAPP (J) extracts after treatment and therefore showing the opposite pattern compared to P-Akt.

When one compared the EAPP levels with the P-Akt levels in general, observable in Figure 40, it again comes clear that extracts with higher EAPP levels show lower P-Akt signals and the other way round.

Figure 40: Influence of EAPP Levels on P-Akt in stable cell lines carrying EAPP alterations. Western blot analysis and immunostaining of U2OS cells with different stable EAPP versions treated with 10µM ATMi. Analysis of their impact on P-Akt and Akt1. [Phospho-Akt T308 (244F9), Akt1(B-1), EAPP (1E4)]

7.6 Analysis of EAPP protein interaction via immunoprecipitation

7.6.1 Direct immunoprecipitation with antibody coupled Protein A beads in U2OS mEAPP (J)

To examine the possible interaction partners of EAPP immunoprecipitation (IP) with Protein A beads coupled with the 12CA5 anti-HA-antibody was performed with U2OS mEAPP(J) extracts both untreated and treated with 10µM Etoposide.

Helic2 showed strong bands in IP extracts, seen in Figure 41, indicating an interaction of Helic2 with EAPP. It could also be observed that Helic2 expression increased after Etoposide treatment in both normal extracts as wells as in the IPs.

PRP8 could not be observed in the total protein extracts probably due to its low amount but was clearly observable in the IPs. A difference between untreated and Etoposide treated cells was not observable.

A probable interaction was also in the Figure 41: Protein interaction partners of EAPP. case of E6AP observable. Whereas in Immunoprecipitation with HA (12CR5) the total extracts many different coupled beads with 500µg mEAPP (J) protein extract +/- 10µM Etoposide, versions with different heights of the 20µg total protein extract as control; Western blot analysis for determination protein could be observed, the IPs only of possible interaction partners. [Helic2, PRP8, E6AP, p53, Sm B/B´/N, GSK 3α/β, YB1, HA (16B12)]

70 showed positive signals for the largest version of the protein.

Sm B/B´/N, p53 as well as YB1 showed weak but still significant bands after immunoprecipitation indicating a binding of EAPP also with these proteins. Analysis of GSK 3α/β revealed a possible binding of EAPP only with GSK 3β.

To evaluate the possible influence of not only Etoposide but also Nocodazole which results in a G2 arrest (Ng et al. 1998) and Hydroxyurea which arrests cells in G1 (Cress, Gerner 1977) an additional immunoprecipitation with protein extracts of U2OS mEAPP (J) cells treated with either 2mM Hydroxyurea or 1µM Nocodazole was performed.

As can be seen in Figure 42, a possible interaction between EAPP and PRP8 could be observed. Treatment with Nocodazole seems to even enhance this effect. Also YB1 showed an interaction with EAPP which seems not to be enhanced by Nocodazole as in the case of PRP8 and slightly reduced by Hydroxyurea.

Figure 42: Protein interaction partners of EAPP. Immunoprecipitation with HA (12CR5) coupled beads with 500µg mEAPP (J) protein extract +/- 2mM

Hydroxyurea and +/- 1µM Nocodazole, 20µg total protein extract as control; Western blot analysis for determination of possible interaction partners. [PRP8, YB1, HA (16B12)]

7.6.2 Direct immunoprecipitation with HA coupled Protein A beads in U2OS cells with different versions of EAPP.

For the determination of the binding sites within EAPP immunoprecipitation assays with HA coupled Protein A beads in U2OS cells with different versions of EAPP were performed, seen in Figure 43.

PRP8 shows only significant bands in the IPs with extracts from U2OS HA EAPP as well as mEAPP (J) cells both containing full length EAPP proteins, either human or murine. The different clones with different truncations of EAPP show slight PRP8 bands but as these are also observable in the negative control this seems to be insignificant.

Figure 43: Protein interaction partners of different EAPP truncations. Immunoprecipitation with HA (12CR5) coupled beads with 500µg protein extract of U2OS cells with different truncations or alterations of EAPP, 20µg total protein extract as control; Western blot analysis for determination of possible interaction partners. [PRP8, AAR2]

Also AAR2 is strongly observable in the IPs of U2OS HA EAPP as well as U2OS mEAPP (J) cells but this time the truncations show results too. There is a rather strong band in EAPP 55-c and light bands in 1-120, 1-140, 1-180 as well as 95-c observable. Δ 170-180 and 135-c do not show any or show very light bands.

For further analysis immunoprecipitation of different U2OS cell lines with different versions or truncations of EAPP treated with not only Etoposide but also with TSA a deacetylase inhibitor (Yoshida et al. 1990) or TBH an important co-factor in the synthesis of nitric oxide (Scott-Burden 1995) as seen in Figure 44 war performed.

PRP8 could be observed in every cell line under every condition indicating a strong interaction with EAPP.

Under untreated conditions p53 is only observable in U2OS mEAPP (J) extracts. After Etoposide treatment slight p53 interaction is observable in all cell lines. In addition a light p53 signal is observable after treatment with 4µM TBH in all cell lines with the weakest in U2OS HA EAPP cells.

Figure 44: Protein interaction partners of different EAPP versions or truncations. Immunoprecipitation with HA (12CR5) coupled beads with 500µg protein extract of U2OS cells with different truncations or alterations of EAPP treated with 10µM Etoposide, 100ng/ml TSA, 4µM TBH; Western blot analysis for determination of possible interaction partners. [PRP8(E5), p53(FL 393), HA (16B12)]

8 Discussion / Outlook

EAPP a highly phosphorylated protein which is known to interact with E2F 1, 2 and 3a is known to often be elevated in human cancer cells (Novy et al. 2005). As EAPP levels are elevated after DNA double strand breaks and stimulate p21 expression leading to G1 arrest, EAPP seems to play an important role in cell cycle regulation and DNA damage response (Andorfer, Rotheneder 2011).

Analysis of DNA damage induced by Etoposide revealed an interaction of EAPP with Chk2 which resulted in the dephosphorylation and inactivation of Chk2. This suggests a requirement for EAPP during checkpoint recovery and correct cell cycle progression (Andorfer et al. 2011). In addition an interaction with p53 and anti-apoptotic effects, which can partly be explained by the EAPP-induced stimulation of p21, could also be observed.

The focus of this master thesis was led on further investigation of the role of EAPP in the cell cycle and its influence after DNA damage. To this end we generated cell lines with stable knockouts of EAPP or introduced EAPP truncations. These cell lines were then further used for analysis of the alterations in cell proliferation, their response to DNA damage or serum deprivation via FACS, protein stability or protein expression via Western blot analysis or protein interaction via direct immunoprecipitation caused by different levels or length of EAPP.

Our results show a certain influence of EAPP but not always consistently over all tested cell lines indicating a complex mode of action which is until now not fully understood.

As the knockout of EAPP in murine NIH 3T3 cells was generated without any major issues and was also possible in human HAP1 cells but the complete loss of the EAPP protein was shown to be lethal in U2OS cells it became clear that the dependence on EAPP seems to vary between not only these cell lines but probably more cell lines and/or tissues.

Analysis of the proliferation pattern of HAP1 cells with different EAPP status showed the highest proliferation rate in cells expressing no EAPP protein. A weaker increase than in the cells with a complete knockout could be observed in cells expressing a truncated version of EAPP protein in lower levels. This data implies a correlation between great cell cycle progression and low EAPP levels and vice versa. One probable explanation for these results could be that the alteration is caused by reduced checkpoint control as higher EAPP levels lead to greater p21 activation and in that way to G1 arrest (Andorfer et al. 2011). Unfortunately this correlation could not be observed in NIH 3T3 cells.

FACS analysis for investigation of the cell cycle after DNA damage with increasing amounts of Etoposide revealed that U2OS HA EAPP containing greater amounts of EAPP compared to the wt showed with 61.66% a larger fraction of cells in G1 compared to the wt with only 37.3% after 0.1µM Etoposide treatment. These results support the thought of p21 being the cause of a higher proliferation rate in the previous experiment. HA EAPP cells were in addition the only ones having lower fractions of cells in S phase compared to the wt. The greatest alterations from the wt concerning G1 fractions are observable in U2OS mEAPP (J) containing the murine EAPP protein instead of the human. These cells show after any treatment concentrations elevated amounts of cells in G1 phase and reduced fractions of G2 cells.

Furthermore, it could be observed that the reaction after serum deprivation and re-induction seems to be altered depending on the EAPP status. Our data shows almost no alteration in logarithmic or serum starved state but after 16h of re-stimulation clone 4J expressing no EAPP shows with about 55% a much larger fraction of cells in S phase compared to the wt with less than 23%. As the 4J pattern after 16h looks very similar to the wt pattern after 24h a much faster serum induction response might be performed in clone 4J.

Our results show that the stability of the different EAPP versions seems to vary. The endogenous EAPP in both U2OS and NIH 3T3 cells shows rather

77 low stability. Moreover mEAPP (J), HA EAPP pm as well as HA EAPP 55-c show even increasing levels compared to the total protein amount indicating a great stability probably caused by conformation change or loss of a protease binding site. In the case of U2OS HA EAPP 1-180, 10-240 and 20- 240 the patterns show a decrease in their protein levels after 8h of CHX treatment but increasing levels after 16-24h. One possible explanation of that picture is that a protease attacking these EAPP versions or an ubiquitin ligase mediating their degradation is active in the beginning of the treatment explaining the decreasing amounts in the beginning. After continued treatment this factor may get destroyed itself giving the remaining EAPP the chance to survive longer times and therefore resulting in that pattern. As until now no specific interaction of a repressor or inhibitor of EAPP is known to cause such a pattern this effect has to be further investigated to come to clear prove.

Analysis of protein expression revealed that the amounts of P-Akt which levels are not detectable or P-Pras with very low levels in wt HAP1 cells is drastically increased in HAP1 EAPP Exon 3 and 6 indicating that reduced or absent levels of EAPP lead to increased expression or decreased dismantling of these proteins. Furthermore it could also be observed that Akt phosphorylation is even enhanced after treatment with Etoposide. This effect is repealed with co-treatment with ATMi. ATMi alone leads to a decrease in Akt phosphorylation. To be noted also the levels of other proteins seem to be influenced by EAPP. Under normal conditions all HAP1 versions express PTEN, p53 and pRb. After treatment with Etoposide whether alone or in combination with ATMi the levels of PTEN and pRb become non-detectable and the levels of p53 become reduced in HAP1 wt cells but only reduced or slightly reduced in EAPP Exon 3 and 6 cells. As PTEN is regulated by p53 (Stambolic et al. 2001) the down regulation of PTEN may be caused indirectly through the down regulation of p53. Up until now no interactions or pathways are known to explain these observations.

Comparison of the P-Akt amount with the EAPP levels of different U2OS versions revealed a possible interconnection. It could be shown that U2OS

78 ps EAPP containing reduced amounts of EAPP showed the highest P-Akt levels whereas lowest P-Akt levels could be observed in U2OS HA EAPP cells containing the highest amounts of EAPP indicating a correlation between high EAPP levels and low P-Akt and vice versa. One possible explanation for that could be that kinases of phosphatases regulating Akt phosphorylation may be influenced by EAPP. The interconnection of EAPP and P-Akt could also be caused indirectly via the cell cycle. As the levels of pAkt decrease during cell cycle stop (Liu et al. 2014) and high levels of EAPP cause that stop of cell cycle progression (Andorfer, Rotheneder 2011) EAPP levels could indirectly affect P-Akt levels.

Via immunoprecipitation putative interaction partners of EAPP could be identified revealing interactions of EAPP in U2OS cells with Helic2, PRP8, E6AP, p53, Sm B/B´/N, GSK 3β, YB1 and AAR2 protein. As many of those proteins are necessary for splicing like AAR2 which is involved in mRNA splicing (Nakazawa et al. 1991), or PRP8 which is located in the catalytic core of the spliceosome and is mandatory for rearrangement (Grainger, Beggs 2005), Helic2 an RNA helicase important for RNA-RNA conformation alteration necessary for second step of splicing (Laggerbauer et al. 1998), or GSK3β which has a strong influence on Tau splicing (Mukai et al. 2002), EAPP seems to play a critical role in the splicing machinery.

An alteration of the protein-protein interaction caused by Etoposide treatment could be observed in interactions of Helic2 with EAPP. The interaction seems to be enhanced after treatment with Etoposide. This increased interaction could be caused through a splice factor as splicing is reduced in cell cycle stop (Blencowe 2003), (Suvorova et al. 2013) as induced by Etoposide. Another explanation could be a modification e.g. a phosphorylation leading to enhanced interaction.

Treatment with Etoposide in addition revealed an interaction of EAPP with p53 which is not observable under unstressed conditions. As p53 gets acetylated when treated with Etoposide (Solomon et al. 2006) and it could be observed that EAPP stronger binds to the acetylated p53 this observation is not surprising. An interaction is observable with human as well as murine

EAPP. To be noted probably not the whole protein is required for interaction as interaction is observable with EAPP 1-180 as well as 55-c.

An altered interplay could also be observed after treatment with Hydroxyurea leading to a decreased interaction of EAPP with YB1. As Hydroxyurea leads to an arrest of cells in G1 (Cress, Gerner 1977) the simplest explanation is that EAPP and YB1 interact only at distinct phases of the cell cycle showing reduced interactions in G1.

An increase of PRP8 binding to EAPP could be observed after Etoposide as well as Nocodazole treatment. Analysis of the EAPP truncations revealed that PRP8 seems to require full length EAPP no matter whether it is murine or human for successful interaction as no interaction was observable in any truncated version. This indicates that probably one binding site is missing in the truncated versions. An interaction was neither observable in truncations missing parts of the C-terminus nor the N-terminus. Therefore it may be the case that one binding site is missing in every version one being before nucleotide 55 and the other after nucleotide 180. Another possible explanation could be an altered conformation making an interaction impossible. The fact that clone Δ170-180 also shows no interaction anymore might speak for that theory but until now no information about the specific binding sites, the allostery or the 3D structure of EAPP are known which could further resolve this cause.

Experiments revealed strong interactions of AAR2 with the full length EAPP, again murine and human, but this time also weak interactions with truncations. Whereas the loss of the first 55 nucleotides seems to have rather low impact on the ability of protein-protein interaction formation a greater loss like in 135-c or the loss of the C-terminus even from nucleotide 180 or a deletion in clone Δ170-180 seems to have rather high impact on the ability of EAPP to bind to AAR2. Reasons for that altered interactions could be the same as in PRP8 interaction namely the loss of a binding site or problems in the formation of the correct and needed conformation.

With the results obtained in this thesis we could further investigate the role of EAPP. Experiments revealed an increase in G1 arrest with elevated EAPP levels suggesting a crucial role of EAPP in cell cycle control. We could identify possible interaction partners of EAPP and proteins influenced by EAPP expression levels.

Further experiments will be performed to clarify the differences and common ground of EAPP between different cell lines to further investigate its role. In the following analyses of possible interaction partners via protein pulldown assays will be performed to verify the obtained data gathered via immunoprecipitation. Although we already know certain things about EAPP there are still many things about its mode of action or way of interaction unclear which have to be investigated in the future.

9 References Anders, C.; Niewoehner, O.; Duerst, A.; Jinek, M. (2014): Structural basis of PAM-dependent target DNA recognition by the Cas9 endonuclease. In Nature 513 (7519), pp. 569–573. Andorfer, P. (2011): EAPP: At the crossroad of apoptosis and p21-mediated cell cycle arrest. Dissertation. Universität Wien, Wien. Naturwissenschaften Genetik-Mikrobiologie. Andorfer, P.; Rotheneder, H. (2011): EAPP: gatekeeper at the crossroad of apoptosis and p21-mediated cell-cycle arrest. In Oncogene 30 (23), pp. 2679–2690. Andorfer, P.; Rotheneder, H. (2013): Regulation of the MDR1 promoter by E2F1 and EAPP. In FEBS letters 587 (10), pp. 1504–1509. Andorfer, P.; Schwarzmayr, L.; Rotheneder, H. (2011): EAPP modulates the activity of p21 and Chk2. In Cell cycle (Georgetown, Tex.) 10 (13), pp. 2077–2082. Blencowe, B. (2003): Splicing regulation: the cell cycle connection. In Current biology : CB 13 (4), pp. R149-51. Chen, K.; Ou, X.; Wu, J.; Shih, J. (2011): Transcription factor E2F- associated phosphoprotein (EAPP), RAM2/CDCA7L/JPO2 (R1), and simian virus 40 promoter factor 1 (Sp1) cooperatively regulate glucocorticoid activation of monoamine oxidase B. In Molecular pharmacology 79 (2), pp. 308–317. Cress, A.; Gerner, E. (1977): Hydroxyurea treatment affects the G1 phase in next generation CHO cells. In Experimental Cell Research 110 (2), pp. 347–353. Davies, C.; Brown, C.; Westphal, D.; Ward, J.; Ward, V. (2015): Murine norovirus replication induces G0/G1 cell cycle arrest in asynchronously growing cells. In Journal of virology 89 (11), pp. 6057–6066. Dubrovsky, J.; Ivanov, V. (2003): Celebrating 50 years of the cell cycle. In Nature 426 (759). Ensembl (2017): Gene: EAPP. Available online at http://www.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00000129 518;r=14:34515929-34539711. Florey, K. (1989): Analytical profiles of drug substances. San Diego [Calif.]: Academic Press (Analytical profiles of drug substances, 18). Fresno Vara, J.; Casado, E.; Castro, J.; Cejas, P.; Belda-Iniesta, C.; Gonzalez-Baron, M. (2004): PI3K/Akt signalling pathway and cancer. In Cancer treatment reviews 30 (2), pp. 193–204.

Gaubatz, S.; Lees, J.; Lindeman, G.; Livingston, D. (2001): E2F4 is exported from the nucleus in a CRM1-dependent manner. In Molecular and cellular biology 21 (4), pp. 1384–1392. Grainger, R.; Beggs, J. (2005): Prp8 protein: at the heart of the spliceosome. In RNA (New York, N.Y.) 11 (5), pp. 533–557. Hsu, P.; Lander, E.; Zhang, F. (2014): Development and applications of CRISPR-Cas9 for genome engineering. In Cell 157 (6), pp. 1262–1278. Humphrey, T. (Ed.) (2005): Cell cycle control. Mechanisms and protocols. Totowa NJ: Humana Press (Methods in molecular biology, 296). Jinek, M.; Chylinski, K.; Fonfara, I.; Hauer, M.; Doudna, J.; Charpentier, E. (2012): A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. In Science (New York, N.Y.) 337 (6096), pp. 816–821. Kovacina, K.; Park, G.; Bae, S.; Guzzetta, A.; Schaefer, E.; Birnbaum, M.; Roth, R. (2003): Identification of a proline-rich Akt substrate as a 14-3-3 binding partner. In The Journal of biological chemistry 278 (12), pp. 10189– 10194. Laggerbauer, B.; Achsel, T.; Luhrmann, R. (1998): The human U5-200kD DEXH-box protein unwinds U4/U6 RNA duplices in vitro. In Proceedings of the National Academy of Sciences of the United States of America 95 (8), pp. 4188–4192. Lewis, J.; Alberts, B. (2007): Molecular biology of the cell. 5. ed. New York: Garland Science. LIPMAN, A. (1993): MARTINDALE. „Martindale - the Extra Pharmacopoeia‟ (30th ed), edited by J. E. F. Reynolds. In International Journal of Pharmacy Practice 2 (2), p. 124. Liu, P.; Begley, M.; Michowski, W.; Inuzuka, H.; Ginzberg, M.; Gao, D. et al. (2014): Cell-cycle-regulated activation of Akt kinase by phosphorylation at its carboxyl terminus. In Nature 508 (7497), pp. 541–545. Lodish, H. (2004): Molecular cell biology. 5. ed. New York: Freeman. Maehama, T.; Dixon, J. E. (1998): The tumor suppressor, PTEN/MMAC1, dephosphorylates the lipid second messenger, phosphatidylinositol 3,4,5- trisphosphate. In The Journal of biological chemistry 273 (22), pp. 13375– 13378. Markl, J.; Purves, W.; Sadava, D. (Eds.) (2011): Biologie. With assistance of Jürgen Markl. 9. Aufl. Heidelberg: Spektrum Akad. Verl. Marraffini, L.; Sontheimer, E. (2010): CRISPR interference: RNA-directed adaptive immunity in bacteria and archaea. In Nature reviews. Genetics 11 (3), pp. 181–190.

Matson, J.; Cook, J. (2017): Cell cycle proliferation decisions: the impact of single cell analyses. In The FEBS journal 284 (3), pp. 362–375. Mukai, F.; Ishiguro, K.; Sano, Y.; Fujita, S. (2002): Alternative splicing isoform of tau protein kinase I/glycogen synthase kinase 3beta. In Journal of neurochemistry 81 (5), pp. 1073–1083. Nakazawa, N.; Harashima, S.; Oshima, Y. (1991): AAR2, a gene for splicing pre-mRNA of the MATa1 cistron in cell type control of Saccharomyces cerevisiae. In Molecular and cellular biology 11 (11), pp. 5693–5700. Ng, C.; Lam, C.; Yeung, P.; Wong, J. (1998): Flow cytometric analysis of nocodazole-induced cell-cycle arrest in the pennate diatom Phaeodactylum tricornutum Bohlin. In Journal of Applied Phycology 10 (6), pp. 569–572. Novy, M.; Pohn, R.; Andorfer, P.; Novy-Weiland, T.; Galos, B.; Schwarzmayr, L.; Rotheneder, H. (2005): EAPP, a novel E2F binding protein that modulates E2F-dependent transcription. In Molecular biology of the cell 16 (5), pp. 2181–2190. Obrig, T.; Culp, W.; McKeehan, W.; Hardesty, B. (1971): The Mechanism by which Cycloheximide and Related Glutarimide Antibiotics Inhibit Peptide Synthesis on Reticulocyte Ribosomes. In J. Biol. Chem. 246 (1), pp. 174– 181. Available online at http://www.jbc.org/content/246/1/174.full.pdf. Pap, M.; Cooper, G. (1998): Role of glycogen synthase kinase-3 in the phosphatidylinositol 3-Kinase/Akt cell survival pathway. In The Journal of biological chemistry 273 (32), pp. 19929–19932. Pommier, Y.; Leo, E.; Zhang, H.; Marchand, C. (2010): DNA topoisomerases and their poisoning by anticancer and antibacterial drugs. In Chemistry & biology 17 (5), pp. 421–433. Purves, W. (Ed.) (2004): Life. The science of biology. 7. ed., 1. print. Sunderland Mass.: Sinauer [u.a.]. Qi, L.; Larson, M.; Gilbert, L.; Doudna, J.; Weissman, J.; Arkin, A.; Lim, W. (2013): Repurposing CRISPR as an RNA-guided platform for sequence- specific control of gene expression. In Cell 152 (5), pp. 1173–1183. Ran, A.; Hsu, P.; Wright, J.; Agarwala, V.; Scott, D.; Zhang, F. (2013): Genome engineering using the CRISPR-Cas9 system. In Nature protocols 8 (11), pp. 2281–2308. Rechsteiner, M.; Rogers, S. W. (1996): PEST sequences and regulation by proteolysis. In Trends in biochemical sciences 21 (7), pp. 267–271. Rotheneder, H.; Schwarzmayr, L.; Andorfer, P.; Sarr, A. (2007): Transcription Factors of the E2F Family and DNA Damage Control. In Dynamic Cell Biology.

Schwarzmayr, L.; Andorfer, P.; Novy, M.; Rotheneder, H. (2008): Regulation of the E2F-associated phosphoprotein promoter by GC-box binding proteins. In The international journal of biochemistry & cell biology 40 (12), pp. 2845–2853. Scott-Burden, T. (1995): Regulation of Nitric Oxide Production by Tetrahydrobiopterin. In Circulation 91 (1), pp. 248–250. Solomon, J.; Pasupuleti, R.; Xu, L.; McDonagh, T.; Curtis, R.; DiStefano, P.; Huber, L. (2006): Inhibition of SIRT1 catalytic activity increases p53 acetylation but does not alter cell survival following DNA damage. In Molecular and cellular biology 26 (1), pp. 28–38. Song, G.; Ouyang, G.; Bao, S. (2005): The activation of Akt/PKB signaling pathway and cell survival. In Journal of cellular and molecular medicine 9 (1), pp. 59–71. Song, M.; Salmena, L.; Pandolfi, P. (2012): The functions and regulation of the PTEN tumour suppressor. In Nature reviews. Molecular cell biology 13 (5), pp. 283–296. Stambolic, V.; MacPherson, D.; Sas, D.; Lin, Y.; Snow, B.; Jang, Y. et al. (2001): Regulation of PTEN transcription by p53. In Molecular cell 8 (2), pp. 317–325. Suvorova, E.; Croken, M.; Kratzer, S.; Ting, L.; Conde de Felipe, M.; Balu, B. et al. (2013): Discovery of a splicing regulator required for cell cycle progression. In PLoS genetics 9 (2), pp. e1003305. Tsai, K.; Hu, Y.; Macleod, K.; Crowley, D.; Yamasaki, L.; Jacks, T. (1998): Mutation of E2f-1 suppresses apoptosis and inappropriate S phase entry and extends survival of Rb-deficient mouse embryos. In Molecular cell 2 (3), pp. 293–304. van der Oost, J.; Westra, E.; Jackson, R.; Wiedenheft, B. (2014): Unravelling the structural and mechanistic basis of CRISPR-Cas systems. In Nature reviews. Microbiology 12 (7), pp. 479–492. Weizmann Institute of Science. (2017): EAPP gene. Available online at http://www.genecards.org/cgi-bin/carddisp.pl?gene=EAPP. Yoshida, M.; Kijima, M.; Akita, M.; Beppu, T. (1990): Potent and specific inhibition of mammalian histone deacetylase both in vivo and in vitro by trichostatin A. In The Journal of biological chemistry 265 (28), pp. 17174– 17179.

10 Statutory declaration

Erklärung:

Ich erkläre, dass die vorliegende Diplomarbeit/Masterarbeit von mir selbst verfasst wurde und ich keine anderen als die angeführten Behelfe verwendet bzw. mich auch sonst keiner unerlaubter Hilfe bedient habe. Ich versichere, dass ich diese Diplomarbeit/Masterarbeit bisher weder im In- noch im Ausland (einer Beurteilerin/einem Beurteiler zur Begutachtung) in irgendeiner Form als Prüfungsarbeit vorgelegt habe. Weiters versichere ich, dass die von mir eingereichten Exemplare (ausgedruckt und elektronisch) identisch sind.

Declaration:

I hereby declare that the submitted Master thesis was written by myself and that I did not use any aids other than those indicated, none of which are unauthorised. I assure that I have not previously submitted this Master thesis or its contents in any form for assessment as part of an examination either in Austria or abroad. Furthermore, I assure that all copies submitted by myself (electronic and printed) are identical. date: ...... signature: ......