The Fate of Endoplasmic Reticulum-Targeted Proteins in the Face of Proteasome Failure

Home , ERLIN1, Protein topology, SEC63, SYVN1

The fate of endoplasmic reticulum-targeted proteins in the face of proteasome failure

Dissertation zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.)

vorgelegt von: Franziska Höfer

an der

Mathematisch-Naturwissenschaftliche Sektion Fachbereich Biologie

Tag der mündlichen Prüfung: 18. Mai 2015 1. Referent: Prof. Dr. Martin Scheffner 2. Referent: Prof. Dr. Marcus Groettrup 3. Referent: Prof. Dr. Jörg Tatzelt

Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-0-291257

“Imagination is more important than knowledge. For knowledge is limited, whereas imagination embraces the entire world, and all there ever will be to know and understand, stimulating progress, giving birth to evolution. It is, strictly speaking, a real factor in scientific research.”

Albert Einstein, 1931

Table of Content

1. SUMMARY ...... 1

2. ZUSAMMENFASSUNG ...... 2

3. INTRODUCTION ...... 3

3.1 The endoplasmic reticulum (ER) ...... 3 3.1.1 Composition and function of the endoplasmic reticulum ...... 3 3.1.2 Protein transport across the mammalian ER membrane ...... 3 3.1.3 ER signal peptides and the ER signal peptidase complex ...... 9 3.1.4 ER-associated degradation (ERAD) ...... 11 3.1.5 ER under stress: the unfolded protein response, preemptive quality control in the ER and mislocalized proteins...... 16 3.1.6 Antigen processing and the endoplasmic reticulum ...... 21

3.2 The (immuno-) proteasome ...... 23 3.2.1 Composition and assembly of the mammalian (immuno-) proteasome ...... 23 3.2.2 Ubiquitin and the proteasomal protein degradation process ...... 26 3.2.3 Modification and inhibition of the proteasome ...... 28 3.2.4 Oxidative stress and proteasomes ...... 31

3.3 The brain ...... 33 3.3.1 Anatomy of the brain ...... 33 3.3.2 The (immuno-) biology of astrocytes ...... 34 3.3.3 The role of the (immuno-) proteasome in the central nervous system (CNS) ...... 34

4. OBJECTIVES ...... 36

5. MATERIAL & METHODS ...... 38

5.1 Prokaryotic cells ...... 38 5.1.1 Bacterial strain and culture conditions ...... 38 5.1.2 Preparation and transformation of competent bacteria ...... 38

5.2 Eukaryotic Cells ...... 38 5.2.1 Cell lines ...... 38 5.2.2 Heat shock of eukaryotic cells ...... 39 5.2.3 Cytokine stimulation of cell lines ...... 39 5.2.4 Lymphocytic choriomeningitis virus (LCMV) infection of cells ...... 39 5.2.5 Chemical inhibitors and inducers ...... 39 5.2.6 RNA silencing ...... 40 5.2.7 Transient transfection of eukaryotic cells with DNA plasmids ...... 41 5.2.8 Generation of stable transfected cell lines ...... 44 5.2.9 Lysis of cells ...... 44

5.2.10 Fluorescence-activated cell sorting ...... 44 5.2.11 Confocal fluorescence microscopy of cells ...... 44 5.2.12 Magnetic cell separation ...... 45 5.2.13. Measurement of cell death ...... 45

5.3 Proteins ...... 46 5.3.1 Immunoprecipitation (IP) ...... 46 5.3.2 Deglycosylation of proteins ...... 46 5.3.3 Radioactive labeling of proteins with [35S]-methionine/cysteine (pulse-chase) ...... 47 5.3.4 Separation of proteins with SDS-PAGE ...... 47 5.3.6 Analysis of radioactive proteins on SDS-gels...... 49 5.3.7 Non-equilibrium pH gel electrophoresis (NEPHGE) ...... 50 5.3.8 Fractionating cellular proteins using osmotic pressure ...... 50 5.3.9 Proteasome activity assay ...... 51 5.3.10 Sucrose gradient density centrifugation ...... 51 5.3.11 Purification of antibodies from rabbit sera ...... 51

5.4. DNA ...... 51 5.4.1 Preparation of DNA ...... 51 5.4.2 Agarose gel electrophoresis ...... 52 5.4.4 Polymerase Chain Reaction – PCR ...... 52 5.4.5 Restriction enzyme digestion ...... 53 5.4.6 Ligation of DNA fragments ...... 53 5.4.7 Site directed mutagenesis of plasmids ...... 53 5.4.8 Quantitative real-time polymerase chain reaction (qRT-PCR) ...... 54

5.5 RNA ...... 54 5.5.1 RNA extraction from eukaryotic cells ...... 54 5.5.2 Synthesis of cDNA from RNA samples – Reverse Transcriptase Reaction ...... 54

5.6 Animals ...... 54 5.6.1 Mice ...... 54 5.6.2 Liver sections for immunohistochemical analysis ...... 55 5.6.3 Immunization of rabbits for antibody generation ...... 55

5.7 Virus ...... 55 5.7.1 Lymphocytic choriomeningitis virus - LCMV ...... 55

6. RESULTS ...... 56

6.1 The fate of ER-targeted proteins in the face of proteasome impairment ...... 56 6.1.1 PSCA precursor protein is stabilized during proteasome inhibition ...... 56 6.1.1.1 Treatment with chemical proteasome inhibitors stabilizes FLAG-PSCA-HA precursor protein ...... 56 6.1.1.2 Genetic proteasome silencing stabilizes FLAG-PSCA-HA precursor protein ...... 61 6.1.1.3 Overexpression of aggregation-prone proteins stabilizes FLAG-PSCA-HA precursor protein 64 6.1.1.4 Oxidative stress stabilizes FLAG-PSCA-HA precursor protein ...... 67 6.1.1.5 Glycosylation of FLAG-PSCA-HA protein during proteasome inhibition ...... 68 6.1.1.6 Cellular localization of FLAG-PSCA-HA protein during proteasome inhibition ...... 69 6.1.1.7 Stabilization of PSCA-HA precursor protein during proteasome inhibition ...... 70

6.1.2 Prolactin precursor stability during proteasome inhibition ...... 71 6.1.3 Proteasome inhibition stabilizes Leptin precursor protein ...... 72 6.1.3.1. FLAG-Leptin-HA precursor protein is stabilized during proteasome inhibition and oxidative stress ...... 72 6.1.3.2 Stabilization of Leptin-HA protein during proteasome inhibition ...... 75 6.1.4 Investigation of the mechanisms of cytosolic ER precursor protein accumulation during proteasome inhibition or oxidative stress ...... 75 6.1.4.1. The role of ubiquitinated FLAG-PSCA-HA ER signal peptide in precursor protein stabilization during proteasome inhibition ...... 75 6.1.4.2 Stabilization of FLAG-PSCA-HA precursor protein during different stress conditions ...... 76 6.1.4.3 Ubiquitin and stabilization of FLAG-PSCA-HA precursor protein ...... 79 6.1.4.4 Stability of FLAG-PSCA-HA precursor protein during VCP/p97 inhibition, heat shock and Nrf1 or Nrf2 overexpression ...... 81 6.1.4.5 Identification of ER signal peptidase-associated regulator proteins ...... 82 6.1.4.6 Phosphorylation of ER signal peptidase subunits during proteasome inhibition...... 84 6.1.4.7 Co-localization of ER signal peptidase subunits and Derlin-1 or Synoviolin1 ...... 85 6.1.4.8 Bag6 siRNA knock down and Flag-PSCA-HA precursor processing ...... 86 6.1.5 Stabilization of endogenously expressed ER-targeted precursor proteins during proteasome inhibition...... 87

6.1.5.1 α1-Antitrypsin (AAT) ...... 87 6.1.5.2 β2-microglobulin (B2M) ...... 88 6.1.5.3 Envelope glycoprotein of Lymphocytic Choriomeningitis Virus (LCMV GP) ...... 89 6.1.5.4 Carbonic anhydrase 4 (CA4) ...... 90 6.1.5.5 Human cytomegalovirus gene product US11 ...... 90 6.1.5.6 C-C chemokine receptor type 7 (CCR7) ...... 91 6.1.5.7 Murine MHC class I molecules H-2Dd, H-2Db, H-2Kd and H-2Ld ...... 92

6.2. Immunoproteasome precursor organization in murine astrocytes ...... 94 6.2.1 Establishing polyclonal rabbit antibodies against LMP2, LMP7 and MECL-1 ...... 94 6.2.2 The immunoproteasome assembly in IMA2.1 murine astrocyte cell line...... 95 6.2.3 PI31 expression in murine astrocytes and immunoproteasome assembly ...... 98 6.2.4 Establishing stable FLAG-tagged Mecl-1 expressing cell lines ...... 98

7. DISCUSSION ...... 101

7.1 The fate of ER-targeted proteins in the face of proteasome inhibition ...... 101 7.1.1 Mislocalized proteins and protein aggregates in the cytoplasm of mammalian cells ...... 101 7.1.2 Mechanisms of ER-guided precursor protein accumulation during proteasome inhibition ...... 104 7.1.3 Final remarks and outlook ...... 109

7.2. Immunoproteasome precursor organization in murine astrocytes ...... 111

8. APPENDIX ...... 112

8.1 References ...... 112

8.2 Abbreviations ...... 135

8.3 Amino acid abbreviations ...... 138

1. Summary Secreted and membrane-bound proteins are generated in the cytosol and transported into the endoplasmic reticulum (ER) lumen of cells for further processing. The information that guides ER-targeted proteins to their destination is encoded in the first 5-30 amino acids of their sequence. The ER signal peptidase is a complex of five subunits essential for the conversion of secretory and some membrane-bound proteins to their mature form. The enzyme removes the hydrophobic, N-terminal signal sequences of ER-targeted proteins, while they are translocated into the ER-lumen. The proteasome is a barrel-shaped protein complex and its main function is the degradation of unneeded or damaged proteins that are labeled with ubiquitin chains. So far, it has been shown for a few overexpressed proteins that inhibition of the proteasome causes stabilization of ER-targeted precursor proteins, but the underlying molecular mechanisms have remained elusive. In order to further investigate this mechanism, we screened for ER-guided, transiently- and endogenously expressed proteins that show detectable precursor stabilization during proteasome impairment. In addition, we established an important tool for our further work: an N-terminally FLAG-tagged and C-terminally HA-tagged prostate stem cell antigen (FLAG- PSCA-HA) overexpression construct. Thereby, we were able to analyze cellular localization and glycosylation patterns of accumulated precursor proteins during proteasome inhibition. We also verified that genetical-, as well as protein aggregate-induced proteasome inhibition is able to stabilize the precursor proteins, which excludes side effects of chemical proteasome inhibitors. In another approach, we tested different intracellular stress conditions, e.g. nitrogen stress, ER stress, oxidative stress and heat shock, and their effect on the stability of ER-targeted precursor proteins. We were able to show that only oxidative stress, which eventually leads to proteasome impairment, induces accumulation of ER-guided precursor proteins. Finally, we examined the regulation of the ER-signal peptidase during proteasome disturbance in detail, to discover the role of this enzyme for the stabilization of ER-targeted precursor proteins during defective proteasomal protein degradation.

Exposure of cells to inflammatory signals, like the cytokines IFNγ or TNFα, leads to incorporation of β1i (LMP2), β2i (MECL-1) and β5i (LMP7) subunits into the proteasomes of hematopoietic cells. These so-called immunoproteasomes generate more peptides that exhibit hydrophobic or basic C-terminal residues, which are better suited for MHC class I binding and that expand the pool of antigens. In a second, independent project, we studied the processing of immunoproteasome subunits in an astrocyte cell line and generated antibodies against the three enzymatically-active immunoproteasome subunits.

2. Zusammenfassung Sekretierte und membrangebundene Proteine werden im Zytoplasma eukaryotischer Zellen gebildet und danach zum endoplasmatischen Retikulum (ER) transportiert. Die ersten 5-30 Aminosäuren in der Proteinsequenz kodieren deren endgültiges Ziel. Die ER Signalpeptidase ist ein Proteinkomplex der aus fünf Untereinheiten besteht. Sie ist essentiell für die Prozessierung von Proteinen die für das ER bestimmt sind. Die ER-Signalpeptidase entfernt die hydrophobe, N-terminale Signalsequenz der Proteine während deren Translokation in das ER Lumen. Das Proteasom ist ein Proteinkomplex, der im Zytoplasma und Zellkern fast aller Zellen zu finden ist. Die Hauptaufgabe des Proteasoms ist der Abbau von nicht benötigten oder nicht funktionellen Proteinen. Bisher konnte für einige überexprimierte, sekretorische Proteine gezeigt werden, dass sich ihr unprozessiertes Vorläuferprotein während der Inhibition des Proteasoms anreichert. Die Mechanismen die dem zugrunde liegen sind bisher noch ungeklärt. Um diese Fragestellung weiter zu erforschen haben wir zuerst untersucht, welche weiteren Proteine mit ER Signalsequenz eine Anreicherung des Vorläuferproteins bei Proteasominhibition aufweisen. Dadurch haben wir sowohl über- als auch endogen exprimierte Proteine gefunden, auf die dies zutrifft. Eine N-terminal FLAG- und C-terminal HA-markierte Form des Prostate Stem Cell Antigen Proteins (FLAG-PSCA-HA) erwies sich als hervorragendes Werkzeug, um sowohl die zelluläre Lokalisation als auch die Glykosylierungsmuster der Vorläuferproteine bei fehlender Proteasomaktivität zu verfolgen. Neben der chemischen Proteasominhibition interessierte uns auch das Verhalten der Proteine bei genetischer- oder proteinbasierter Inaktivierung des Proteasoms, sowie bei unterschiedlichen intrazellulären Stressoren, wie Hitzeschock, ER-Stress und oxidativem Stress. Nur oxidativer Stress, welcher ebenfalls das Proteasom inhibiert, führte zu einer Anreicherung der Vorläuferproteine. Wir untersuchten auch den Einfluss der Proteasominhibition auf die Regulation der ER-Signalpeptidase.

Bei der Behandlung von Zellen mit Cytokinen, wie IFNγ oder TNFα, kommt es zu einer Expression von alternativen Untereinheiten des Proteasoms, β1i (LMP2), β2i (MECL-1) und β5i (LMP7), welche in hämatopoetischen Zellen konstitutiv exprimiert werden. Sie sorgen für die Generierung von Peptiden mit exponierten, hydrophoben oder basischen C-terminalen Aminosäuren, welche sehr gut zur Beladung von MHC Klasse I Molekülen geeignet sind, und den Antigenpool der Zelle erweitern. In einem zweiten, unabhängigen Projekt haben wir Antikörper gegen die drei enzymatisch aktiven Immunoproteasomuntereinheiten generiert und analysiert. Ebenfalls wurde in dieser Arbeit die Prozessierung dieser Untereinheiten bei einer Astrozyten Zellinie (IMA 2.1) eruiert.

3. Introduction 3.1 The endoplasmic reticulum (ER) 3.1.1 Composition and function of the endoplasmic reticulum The endoplasmic reticulum is a compartment that exists in almost every type of eukaryotic cell, except for red blood cells and spermatozoa. This organelle is a network of flattened, membrane- enclosed tubes inside the cell. These so-called cisternae are held together by the cytoskeleton. There is a smooth transition from the outer membrane of the nucleus to the ER membrane (Soltys et al. 1996). The endoplasmic reticulum is divided into two types, the smooth endoplasmic reticulum (SER) and the rough endoplasmic reticulum (RER). The smooth endoplasmic reticulum is an important place for lipid, steroid and carbohydrate metabolism, as well as detoxification of the cell (Berg et al. 2010). The outer membrane of the rough endoplasmic reticulum is covered with ribosomes, which are the site of protein synthesis. The ribosomes are constantly bound and released from the RER membrane and their binding point is the translocon (Gorlich et al. 1992). The more active a specific cell type is in protein synthesis the more prominent is the rough endoplasmic reticulum in these cells. It is a specialized compartment for protein folding, quality control and maturation. Most of the proteins targeted into the ER are glycosylated in the ER lumen or in the lumina of the cis-, medial- or trans-Golgi. It is estimated that approximately 20% of all proteins produced by eukaryotic cells are transported into the ER to be part of the secretory pathway, secreted or bound on the cell surface (Lander et al. 2001; Redman et al. 1966).

3.1.2 Protein transport across the mammalian ER membrane SRP-dependent protein transport The targeting information that guides membrane-bound or secreted proteins to the ER is encoded in the first 5-30 amino acids of the protein sequence. These hydrophobic signal sequences are located at the N-terminus of the growing polypeptide chain (Milstein C, Brownlee GG, Harrison TM 1972; Blobel & Sabatini 1971). Signal sequences, originated from the ribosome exit tunnel, are co-translationally recognized by the signal recognition particle (SRP), a complex of a 7S RNA and six proteins (Walter et al. 1981). The SRP does not specifically bind to ribosomes loaded with ER-guided proteins. It scans accessible ribosomes, even unloaded ones, and, if it recognizes an emerging signal peptide, a stable complex is formed (Holtkamp et al. 2012). The translocation across the ER membrane is best understood for secretory proteins. The transient binding of the SRP to the newly synthesized signal peptide of the protein arrests the translation of the protein and guides the ribosome with the nascent polypeptide chain

(ribosome-nascent chain complex, RNC) to the ER membrane-bound SRP-receptor (SR). The receptor is located in close proximity to the so-called translocon (Gilmore et al. 1982) built by the heterotrimeric Sec61αβγ complex. The pore of this aqueous, protein-conducting channel is formed by the Sec61α subunit (Fig 3.1; Oliver et al. 1995; Simon & Blobel 1991).

Figure 3.1│ Recognition of a transmembrane Helix by the Sec61 translocon. (A) A non-polar amino-acid segment (purple) of a co-translationally inserted nascent polypeptide chain (dashed orange line) moving through the 60S ribosomal tunnel during chain elongation. Depending on the segments hydrophobicity, Sec61 initiates either lateral insertion into the ER membrane by transmembrane helix formation or membrane passing of the segment and transport into the ER lumen. (B) Structure of the Sec61 translocon of Methanococcus jannaschii viewed from above (left) or in the plane of the membrane (right). The purple cylinder depicts the hypothetical position of the nascent chain transmembrane helix moving through the lateral gate between helices TM2b-3 (light blue) and TM7-8 (yellow and orange). The green plug helix closes the channel from the periplasmic site and is thought to move out of the way by ribosome binding to the translocon (adapted from von Heijne 2006).

Binding of the ribosome to the cytosolic loops of Sec61α and the release of the signal sequence from the SRP is promoted by GTP hydrolysis of the SRP and its receptor. With this step the pore 4

of the protein conduction channel (PCC) is aligned with the ribosome exit tunnel. Conformational changes in Sec61α open its axial pore, thereby establishing a stable passageway for the nascent polypeptide chain from the ribosome peptidyltransferase center to the inside of the ER lumen (Raden et al. 2000; Holtkamp et al. 2012). It has been shown that, even after binding of the ribosome to the Sec61 complex, a large gap between both is existent. In some cases emerging proteins can form a loop on the cytoplasmic side of the ER membrane (Raden et al. 2000; Pool et al. 2002; Park et al. 2014). The diameter of the translocon pore varies dynamically between 15Å and 50Å, making it impossible to transport already folded proteins. However, the pore size seems to be sufficient for the transport of alpha helical structures (Johnson & van Waes 1999; Meyer et al. 1982). Sec61 has an hourglass-like shape and, without ongoing translation, the constriction between the two funnels of the hourglass is plugged with a short helix (Van den Berg et al. 2004; Fig 3.1). There are several additional proteins known that are connected with the mammalian ER translocase complex, like translocon-associated protein (TRAP) complex and translocating chain-associated membrane protein (TRAM; Meacock et al. 2002; Yamaguchi et al. 1999; Görlich et al. 1992; Voigt et al. 1996; Hartmann et al. 1993). The signal peptide remains stationary in its position during the movement of the growing polypeptide chain through the ER membrane (Gouridis et al. 2009). It is cleaved co-translationally by the ER signal peptidase (Kutay et al. 1995; Fig 3.2). Afterwards, the cleaved signal peptide is further degraded by the ER signal peptidase peptidase (Weihofen et al. 2002). Proteins imported into the ER are N-terminally glycosylated from ER luminal resident glycosidases, if required. Furthermore, folding and refolding of the translocated proteins is mediated by luminal chaperones. Subsequently, proteins are transported to the Golgi apparatus via vesicles that arise from the ER and can finally be secreted (Johnson & van Waes 1999).

Figure 3.2│ Co-translational targeting by the SRP system. (A) Targeting of secreted or membrane bound proteins. Step I: SRP interaction with a signal sequence emerging from the ribosomal exit tunnel. Elongation is retarded until SRP-RNC (RNC, ribosome nascent chain complex) formation. Step II: The SRP-RNC complex is targeted to the ER membrane via the SR. A prerequisite for function of SRP-SR complex formation is GTP, binding to both of them. Step III: The RNC is transferred to the protein-conducting channel (PCC, translocon) in the ER membrane. Step IV: GTP hydrolysis in SRP and SR triggers SRP-SR complex dissociation. For color definition see B. (B) Scheme of the mammalian SRP bound to signal sequence carrying 80S ribosome. SRP54 and SRP helix 8 build the SRP core as part of the S domain. It is positioned near the ribosomal exit tunnel. The 40S ribosomal subunit (yellow) and the 60S ribosomal subunit (grey) are depicted. The SRP RNA is colored in red and the SRP proteins as follows: SRP54NG – light blue, SRP54M – dark blue, signal sequence – green, SRP19 and SRP68/72 – pink. SRP9 – light blue and SRP14 – dark blue. The peptidyl tRNA is labeled and is located between the ribosomal subunits and the nascent chain in the exit tunnel (adapted from Wild et al. 2004).

SRP-dependently transported, membrane-bound proteins with one or more transmembrane (TM) segments are co-translationally integrated into the lipid bilayer. The hydrophilic parts of the protein move either through the Sec61 channel to the ER luminal site or remain in the cytoplasm, in a gap between ribosome and ER membrane, as shown in electron microscopy studies (Frauenfeld et al. 2011; Ménétret et al. 2008). The transmembrane protein segments leave the Sec61 complex via its lateral channel one by one, or in pair, into the lipid phase (Egea & Stroud 2010). If Sec61-integrated polypeptide segments are hydrophilic enough, they will exit through the lateral gate into the lipid phase (Hessa et al. 2005; Heinrich et al. 2000). The hydrophobic signal sequences of membrane-inserted proteins can feature as cleavable signal peptides or stay uncleaved and function as transmembrane domains (signal anchor). The Sec61 complex binds cleavable signal peptides and the nascent polypeptide chain is inserted in a “loop-like” fashion. This leads to a cytosolic orientation of the N-terminus of the protein, creating so-called type II membrane proteins (Shaw et al. 1988). Signal anchor sequences are more hydrophobic than signal sequences and the orientation of the N-terminus (cytosolic or luminal) depends on the protein sequence. The “positive inside rule” predicts that proteins are inserted with their N-terminus towards the cytosol (type II membrane protein) if the segment in front of the hydrophobic region is long and positively charged (Hartmann et al. 1989; Heijne 1986). Integration with the N-terminus towards the ER lumen (type I membrane protein) occurs if the N- 6

terminus of the protein lacks positive charges, the hydrophobic sequence is long and the foregoing protein segment is loosely folded (Rapoport et al. 2004).

SRP-independent protein transport Beside the SRP-dependent pathway mentioned above, there are evidences for several other post-translational pathways that transport proteins into the ER lumen or integrate them into the ER membrane (Johnson et al. 2012; Kutay et al. 1995; Zimmermann & Mollay 1986; Schlenstedt et al. 1990; Shao & Hegde 2011; Yabal et al. 2003). ER targeting sequences that are not available for the SRP or fail binding because of physical restriction require an alternative transport mechanism. The SRP is only engaged in ER targeting if the corresponding signal sequences exhibit a basic interaction threshold with the SRP cleft. The extent of hydrophobicity and the length of the hydrophobic part is important for the interaction (Hikita & Mizushima 1992; Yamamoto et al. 1987). Correspondingly, it has been shown in yeast that proteins with mildly hydrophobic signal sequences are transported in an SRP-independent manner (Ng et al. 1996). Mutations of signal sequences that make them inaccessible for the SRP, for example by inducing α-helical structures, can also attenuate SRP-dependent ER import (Rothe & Lehle 1998). The SRP is unable to detect signal sequences that are not exposed to the cytosol during translocation. Corroboratively, it has been shown that SRP-dependent transport requires proteins with a minimum size of 78-80 amino acids in order to emerge from the ribosome exit tunnel and expose their signal sequence to the cytosol (Schlenstedt & Zimmermann 1987; Zimmermann et al. 1990; Schlenstedt et al. 1992). Tail anchored (TA) proteins possess a single TM domain at their very C-terminus. According to this, TA proteins are C-terminally anchored in the membrane, while their N-terminus functions in the cytosol. Protein folding and degradation, as well as targeting of vesicles in the cell, represent the main functions of TA proteins (Wattenberg & Lithgow 2001). The ER targeting sequences of tail-anchored proteins are still hidden in the ribosome after translation and therefore not accessible to the SRP. Interestingly, the integration of TA proteins in the ER does not involve the Sec61 complex (Yabal et al. 2003).

Figure 3.3│ SRP independent pathways employ a number of cytosolic proteins and chaperones to target and insert their substrates into the ER. Proteins that carry ER targeting sequences that are, due to hydrophobicity or structure, not engaged by the SRP and proteins that cannot associate with the SRP co-translationally (due to length or location of the signal sequence) utilize SRP-independent pathways. Directly after their translation, chaperoning of these proteins is needed to avoid aggregation and misfolding. An ER-targeting pathway, rapidly transporting the substrates to the ER, has been described for some of the chaperones. Eventually, the controlled and unidirectional insertion of the proteins into or through the lipid bilayer is mediated by a membrane bound machinery. Notably, SRP-independent substrates could use several pathways to mediate their journey from the cytosol to the ER (adapted from Ast & Schuldiner 2013).

Different ways of SRP-independent transport routes exist (Fig 3.3). In order to keep the translated proteins in a translocation-competent state, it is important that they are protected through the binding of specialized factors directly after their appearance in the cytosol. This protection prevents premature folding or damage of the protein domains and keeps them in a loosely folded state (Plath & Rapoport 2000). Heat shock protein 70 (Hsp70) and its co- chaperone heat shock protein 40 (Hsp40) function as a “holdase” until the bound proteins have reached the translocon (Ngosuwan et al. 2003). Another preinsertional-binding factor is calmodulin, which can expedite the translocation of short secretory proteins (Shao & Hegde 2011). Calmodulin has been shown to engage amphipathic α-helices because of a binding pocket with large hydrophobic patches that is flanked by regions with highly negative electrostatic potential (Babu et al. 1988; Kretsinger et al. 1986). By binding to the hydrophobic signal sequence, calmodulin shields it from the cytosolic environment, thereby preventing it from degradation or aggregation. After binding to Hsp70/Hsp40 or calmodulin, proteins are guided to

Sec61. Sec61 translocates the proteins across the ER membrane by forming an auxiliary complex with Sec62, Sec63, Sec66 and Sec72 (Deshaies et al. 1991; Panzner et al. 1995; Feldheim et al. 1993; Green et al. 1992). Sec63 consists of a J-domain at its luminal loop, which is able to recruit BiP (Binding immunoglobulin protein, Brodsky et al. 1995; Sadler et al. 1989), an Hsp70-like molecular chaperone. It functions as a “molecular ratchet” providing the force for the insertion of proteins into the ER lumen with the help of Brownian movement (Matlack et al. 1999). A third known SRP-independent transport mechanism in mammalian cells is the ER import of proteins with Bag6 (encoded by BCL-2 associated athanogene 6) involvement, named GET pathway (F. Wang et al. 2010). Bag6 captures newly synthesized TA proteins and prevents protein aggregation or misleading to the mitochondria (Mariappan et al. 2010; Leznicki et al. 2010). It is part of a stable mediator complex formed by TRC35 and Ubl4A. This complex bridges the transfer of the TA proteins from Bag6 to TRC40, a 40 kDa cytosolic ATPase (Wang et al. 2011). WRB, a tryptophan-rich basic protein, is located at the ER membrane and functions as a receptor for TRC40, binding it with a coiled coil motif (Vilardi et al. 2011). The link between TRC40 and WRB is the protein CAML (calcium-signal modulating cyclophilin ligand), which binds WRB and mediates the insertion of mammalian proteins from TRC40 (Yamamoto & Sakisaka 2012). However, there is evidence that TRC-dependent transport is also utilized for ER import of short secretory proteins (Haßdenteufel et al. 2011) and GPI-anchored proteins (Ast et al. 2013).

3.1.3 ER signal peptides and the ER signal peptidase complex Signal sequences play a major role in targeting of newly synthesized proteins to their cellular destination. Eukaryotic signal sequences are, in general, located at the N-termini of proteins and essential for their transport from the cytosol to translocation sites in the membrane of the ER or the mitochondria (Horst & Kronidou 1995). Signal sequences are cleaved from the precursor proteins after or during the transport. There also exist signal sequences that are located in the middle of the protein or at its C-terminus (Kutay et al. 1995). It has been shown that the efficiency of signal sequence mediated protein segregation into the secretory pathway varies between more than 95% and less than 60% (Levine et al. 2005). Proteins that reside in the ER own a typical KDEL amino acid sequence, which guides them to their destination and is cleaved off before the protein leaves the endoplasmic reticulum (Munro & Pelham 1987). Typical ER signal sequences are characterized by a hydrophobic core (h-) region, which is composed of six to fifteen amino acids. This region has been shown to be the most essential part of the signal sequence for targeting and membrane insertion of the transported proteins (von Heijne 1985). A 9

polar (c-) region, often containing helix-breaking proline and glycine residues, flanks the h-region C-terminally. This region contains small, uncharged residues at position -3 and -1, defining the cleavage site of the precursor protein (von Heijne 1990; Hegner et al. 1992). Usually, a polar n- region with a net positive charge, contributing the most to the ranging size from signal sequences between 15 and more than 50 amino acids, is N-terminally connected to the h-region (von Heijne 1986). Signal sequences can target proteins to more than one location, as shown for plasminogen activator inhibitor (PAI; Belin et al. 1996). They also regulate how close the ribosome and translocon complex are associated during translocation and thereby the engagement of other factors into this process, like TRAM protein and TRAP complex (Fons et al. 2003; Voigt et al. 1996). Beside the targeting of proteins and the binding affinity to translocation complexes, signal sequences can also influence the final glycosylation state (Anjos et al. 2002; Rutkowski et al. 2003) and the sensitivity to translocon inhibitors of their corresponding proteins (Besemer et al. 2005; Garrison et al. 2005). The mammalian ER signal peptidase is way more complex than the bacterial enzyme. It is built of five subunits, termed SPC12, SPC18, SPC 21, SPC22/23 and SPC25, according to their molecular weight (Fig 3.4; Evans et al. 1986). The ER signal peptidase belongs to the group of type I signal peptidases, which comprises several other peptidases found in the mitochondrial inner membrane and the cytoplasmic membrane of bacterial cells (Dalbey et al. 1997). It has been shown that the catalytic mechanism of the ER signal peptidase is distinct from the mechanism of most other type I signal peptidases, lacking a common active lysine and using a serine, a histidine and two aspartic acids for the catalytic mechanism (VanValkenburgh et al. 1999). The catalytic centers of SPC18 and SPC21 possess the active serine residues and both proteins are homologous to each other. Human, dog, rat and mouse SPC18 and SPC21 are highly homologous to the bacterial Sec11 polypeptide. SPC12, SPC 22/23 and SPC25 are not essential, but it is known that SPC25 interacts with the translocon complex (Kalies et al. 1998). The C-terminal domains of SPC18, SPC21 and SPC22/23 are located at the luminal side of the ER and span the membrane once, whereas SPC 25 and SPC12 span the membrane twice with their N- and C-terminus facing the cytoplasm (Kalies & Hartmann 1996).

Figure 3.4│ Membrane topologies of the ER signal peptidase subunits. Five subunits build the ER signal peptidase complex. Two of the subunits (SPC18 and SPC21) are weakly homologous to the bacterial, chloroplast and mitochondrial type I SPases. The active serine is depicted. It is located near the membrane surface on the luminal side of the ER membrane (adapted from Dalbey et al. 1997).

3.1.4 ER-associated degradation (ERAD) The process of protein synthesis is not free of errors. In fact, a majority of newly synthesized proteins is misfolded or misassembled (Hoseki et al. 2010). To make sure that none of these defective proteins is secreted or transported to the cell membrane, the ER possesses a quality control system that senses and disposes terminally misfolded proteins. This process, called ER- associated degradation (ERAD), is conserved in eukaryotes (Merulla et al. 2013; Vembar & Brodsky 2008). ERAD is defined by four steps: (I) recognition of misfolded proteins, (II) retrotranslocation across the ER membrane, (III) ubiquitination of the substrates in the cytoplasm and (IV) the proteasomal degradation by the 26S proteasome (Fig 3.5).

Figure 3.5│The four steps of endoplasmic reticulum-associated degradation (ERAD). Recognition: A misfolded region may reside in the cytoplasmic, ER luminal or transmembrane domain of a protein (red star) during synthesis and translation. ER luminal and cytoplasmic chaperones recognize the misfolded region as depicted, depending on its location. Lectins (pink) interact with N-glycans of glycoproteins and monitor their folding state in some cases. Ubiquitination: Upon recognition of misfolded protein regions, ER luminal or cytoplasmic chaperones recruit the ubiquitination machinery to the substrates. A ubiquitin-activating E1 enzyme transfers ubiquitin (gray circle) to the active site cysteine of a ubiquitin-conjugating E2 enzyme in an ATP-dependent manner. Then, most commonly, ubiquitin is transferred to a lysine residue on the substrate protein via an ubiquitin-ligase (E3). These E3 ligases can be cytoplasmic or ER resident. Retrotranslocation: Polytopic membrane proteins are presumably removed from the membrane through retrotranslocation via a channel (retrotranslocon) or via removal of the protein and the surrounding membrane (not depicted). The retrotranslocation depends on the p97/Cdc48 complex. It includes Ufd1 and Npl4 and can interact with ubiquitin and the misfolded regions of the substrate. The mechanical force for substrate removal is provided by p97/Cdc48 via ATP hydrolysis. Degradation: Following retrotranslocation, the misfolded proteins are delivered to the 26S proteasome and must be kept soluble to prevent aggregation during the transport. N-glycans are removed via N-glycanases (not shown) while deubiquitinating enzymes remove ubiquitin moieties. They are located either in the cytosol or the proteasome cap. The three enzymatically-active subunits of the proteasome (with trypsin- like, chymotrypsin-like or caspase-like activities) degrade the protein into short fragments (adapted from Guerriero & Brodsky 2012).

Step I: Recognition The majority of ER-guided proteins receives multiple, N-linked glycans during their maturation pathway (Apweiler et al. 1999). The composition of these glycans functions as a quality control feature that provides information about the folding state of the modified protein (Hessa et al.

2005; Hebert & Molinari 2007). Nascent, N-glycosylated proteins consist of a glucose3- mannose9-N-acetylglucosamine2-asparagine three branched structure (Aebi et al. 2010). Calnexin (CNX) and Calreticulin (CRT), two lectin-type chaperones, are able to interact with nascent proteins after trimming of the first two glucose residues by glucosidases I and II. CNX and CRT recruit protein disulfide isomerase cofactor ERp57 and facilitate oxidative folding of the proteins (Brodsky 2012). After the removal of the third glucose residue by further deglucosylation, the binding of CRT and CNX is abrogated and the glycosylated proteins could exit the ER (Olzmann et al. 2013; Smith et al. 2011). UDP-glucose-glycoprotein- glycosyltransferase is a key quality control sensor during this step. It modifies the structure and composition of glycosylated proteins according to their structural integrity, which means that proteins that are not able to acquire native structures are reglucosylated (Caramelo et al. 2004). Thereby, proteins reassemble with CNX and CRT and undergo new rounds of oxidative folding (Hebert et al. 2010). Simultaneously with this re- and deglucosylation steps, mannose residues of these proteins are progressively removed by ER luminal mannosidases, e.g. the ER mannosidase I (ERmanI) or ER-degradation-enhancing α-mannosidase-like proteins (EDEMs). The removal of three to four mannose residues leads to the escape from the CNX/CRT cycle. These, “terminally misfolded” proteins are recognized by the proteins OS-9 and XTP3-B/Erlectin and committed to the next ERAD steps (Aebi et al. 2010; Lederkremer 2009; Hosokawa et al. 2008). The detection of misfolded, non-glycosylated proteins is independent of CNX and CRT (Brodsky 2012). Non-glycosylated, as well as glycosylated ERAD-prone proteins, are sent to degradation with BiP involvement (Ushioda et al. 2013). EDEM1 is also involved in targeting misfolded, non-glycosylated, luminal proteins to the proteasome. There is evidence that targeting for both protein types, glycosylated and non-glycosylated ones, is similar (Shenkman et al. 2013). The protein disulfide isomerase (PDI) family, whose members can form, break or rearrange disulfide bonds as well as perform chaperone functions, is another important protein factor family for the recognition of ERAD substrates (Benham 2012). Peptidyl-prolyl cis/trans isomerases (PPIs) can detect and eliminate secondary protein structures that result from peptidyl-prolyl bonds before the retrotranslocation of misfolded proteins is initiated (Määttänen et al. 2010; Byun et al. 2014).

Step II: Retrotranslocation Upon recognition of misfolded proteins, the retrotranslocation step is initiated. It is proposed that Derlins (Derlin1-3), a group of proteins with multiple membrane spanning domains, are part of the retrotranslocation channel and/or represent regulatory factors of this process (Ye et al. 2004). They are related to rhomboid proteases, but lack their proteolytic activity. They are suggested to bind ERAD substrates and guide them to their corresponding E3 ligases and VCP/p97 protein for ubiquitination and membrane extraction. (Brodsky 2012). Some ERAD substrates seem to be cleaved prior to retrotranslocation (Tsai & Weissman 2012). Rhomboid- related protein 4 (RHBDL4; Fleig et al. 2012), signal peptide peptidase (SPP, Loureiro et al. 2006) and proteases that are associated with OS-9 and XTP3-B (Olzmann et al. 2013) have been shown to be involved in the proteolysis of ERAD-targeted proteins. Derlin-1 and Derlin-2 are bound to the VCP/p97 AAA-ATPase via their SHP domains (Greenblatt et al. 2011), and provide access of ERAD substrates to the ubiquitination process. It is assumed that the suppressor/enhancer of Lin12-like (SEL1L) protein is an adaptor between proteins that detect misfolding or inappropriate glycosylation (like ERdj5, PDI protein ERp90 and EDEMs) and the retrotranslocon (Williams et al. 2013). SEL1 participates in the EDEMosome synthesis (Bernasconi et al. 2012) and is important for the stability of hydroxymethylglutaryl reductase degradation protein1 (Hrd1) E3 ligase, which ubiquitinates soluble luminal and membrane proteins with misfolded luminal domains or ER-membrane proteins with misfolded transmembrane domains (Plemper et al. 1999). Erlin1, Erlin2 and TMUB are other adaptor proteins that have been identified, acting between polytopic membrane substrates and E3 ligases (Olzmann et al. 2013).

Step III: Ubiquitination Once proteins are recruited to the ER membrane and retrotranslocated into the cytoplasm, ubiquitination is needed to guide them to the proteasomal degradation machinery. Various E3 ligases are proposed to be part of the translocation channel. They span the ER membrane multiple times, and possess cytosolic RING domains (Smith et al. 2011; Ruggiano et al. 2014). It has been shown that Hrd1 E3 ligase-mediated ERAD requires Hrd1 transmembrane domains, its oligomerization and its E3 ligase activity (Carvalho et al. 2010). Hrd1-dependence is determined by the composition of target proteins. For example, only the luminal splice variant, but not the transmembrane domain-containing variant, of the human beta-site amyloid precursor cleaving enzyme (BACE) is degraded in a Hrd1-dependent manner (Bernasconi et al. 2010). There are numerous membrane-spanning E3 ligases, including gp78/AMFR (Fairbank et al. 2009), TRC8 (Stagg et al. 2009), RMA1/RNF5 (El Khouri et al. 2013), MARCH6/TEB4 (Kreft & 14

Hochstrasser 2011) and CHIP (Matsumura et al. 2013), that participate in ERAD. Additionally, there exist cytosolic ERAD-associated E3 ligases that recognize already retrotranslocated, misfolded glycoproteins (Yoshida et al. 2005). They belong to the cytosolic SFC (S-phase kinase-associated protein 1 (Skp1) – Cullin 1 (Cul1)-F-box) family. These proteins, e.g. Fbs1 and Fbs2, recognize N-glycans of retrotranslocated substrates with their F-box components (Yoshida 2007) and probably work together to direct them to degradation (Olzmann et al. 2013). Intriguing results from Ploegh and colleagues imply that there are two ubiquitination steps required for complete processing of ERAD substrates. The first one seems to be necessary for dislocation and, after subsequent deubiquitination, the second ubiquitin attachment leads to substrate direction to the 26S proteasome (Ernst et al. 2009; Sanyal et al. 2012).

Step IV: Delivery to the 26S proteasome and proteasomal degradation VCP/p97 is a member of the AAA-ATPase protein family (Erzberger & Berger 2006) and a key factor in ERAD pathway. It consists of two C-terminal ATPase domains (D1 and D2, Meyer 2012) which are each arranged in stacked hexameric rings. Its ATPase activity is necessary to provide the energy for protein remodeling and substrate extraction from the ER membrane or the retrotranslocon (Hampton & Sommer 2012). Although VCP/p97 itself has an ubiquitin-binding activity (Ye et al. 2003; Meyer et al. 2012), it functions in complexes with ubiquitin-X (UBX) or UBX-like binding domain-containing cofactors (Schuberth & Buchberger 2008), similarly to the nuclear protein localization homolog 4 (Npl4) and ubiquitin fusion degradation protein 1 (Ufd1) heterodimer (Meyer et al. 2012; Wolf & Stolz 2012; Nowis et al. 2006), p47,UBXD1, UBXD7, Ufd3/PLAA, VSIP135 and Ataxin 3 (Meyer et al. 2012). These cofactors are able to bind an amino-terminal, regulatory domain of VCP/p97. The Ufd1-Npl4 heterodimer works as a substrate adapter between VCP/p97 and the retrotranslocon (Bays & Hampton 2002). E3 ligases add K48- linked or K63-linked ubiquitin chains to the ERAD substrates which can be recognized and bound by Npl4 or Ufd1 (Ye et al. 2003; Komander et al. 2009). The combination of VCP/p97 with different adaptor proteins enables it to recognize different substrates and to function in a cofactor-dependent manner, e.g. for membrane protein segregation and trafficking or direction of substrates to the proteasome (Ritz et al. 2011). In other models, Derlins are proposed to unfold substrates and establish contact between VCP/p97 and its associated factors (Greenblatt et al. 2011). Ubiquitin chain editors, elongating short ubiquitin chains, as well as deubiquitinating enzymes (Dubs) are bound by VCP/p97 (Jentsch & Rumpf 2007; Sowa et al. 2009). Peptide-N- glycanase (PNGase) is recruited to ERAD substrates to cleave N-linked glycans after their extraction from the ER membrane (Hirsch et al. 2003; Li et al. 2006). VCP/p97 binds ubiquitin thioesterase 1 (Yod1), a hydrolase, which prevents polyubiquitin chains of ERAD substrates 15

from interfering with proteasomal integration by removal of the ubiquitin moieties (Ernst et al. 2009). The 26S proteasome is a multimeric, catalytic protein complex (Chapter 3.2) build of a 20S barrel-shaped, core complex and a 19S regulatory particle with an ATPase function, which is similar to the one of VCP/p97 (Lipson et al. 2008; Matouschek & Finley 2012). Either several enzymes work synergistically together (Hampton & Sommer 2012) or VCP/p97 delivers the substrates directly to the 20S proteasome core (Matouschek & Finley 2012). Proteins must enter the center of the barrel shaped proteasome to become degraded. It is required that the proteins are unfolded to fit into the 20S core complex (Groll et al. 2000). Finally, the proteins are degraded into peptides by the 26S proteasome.

3.1.5 ER under stress: the unfolded protein response, preemptive quality control in the ER and mislocalized proteins. ERAD Chaperones function in a calcium- and ATP-dependent manner and the oxidizing environment in the ER lumen is important for establishing disulfide bonds in maturing proteins. This leads to sensitivity of the ER to redox stress, changes in calcium homeostasis, nutrient deprivation, high loads of secreted proteins and impairment of glycosylation or vesicular trafficking (Ron 2002; Kaufman et al. 2002). These ER stressors can overwhelm the ER capacity for protein folding, which provokes accumulation of unfolded proteins in the ER lumen. Thereby, the unfolded protein response (UPR) is activated to restore the ER luminal homeostasis. Prolonged unfolded protein stress could induce apoptosis, mainly via the mitochondrial pathway, which is controlled by the Bcl-2 gene family (Szegezdi et al. 2006; Samali et al. 2010). The UPR is characterized by the expansion of the secretory apparatus, an increase of the ER volume and a decrease in the amount of newly synthesized luminal proteins. Additionally, the UPR leads to enhanced removal of unfolded proteins via ERAD (Friedlander et al. 2000) and autophagy induction (Deegan et al. 2013). Three ER stress sensor proteins with distinct functions: inositol- requiring enzyme 1 (IRE1), PKR-like ER kinase (PERK) and activating transcription factor 6 (ATF6), are known in mammalian cells. The luminal domains of these three proteins interact with BiP, which means that they compete with all unfolded proteins in the ER for BiP binding. Accumulation of unfolded proteins leads to dissociation of BiP from IRE1, PERK and ATF6, activation of the three sensor proteins and induction of the UPR (Lisa Vincenz et al. 2013).

Figure 3.6│Overview of signaling pathways of the unfolded protein response (UPR). Grp78 binds unfolded proteins accumulating in the ER lumen. Grp78 binding leads to activation of the ER stress sensors ATF6, PERK and IRE1. This induces a signal cascade termed UPR. It involves downregulation of mRNA translation and the activation of transcription factors regulating genes important for ER homeostasis and cell survival. During prolonged or severe ER stress, apoptosis inducing genes are upregulated (adapted from L Vincenz et al. 2013).

PERK: PERK is a serine-threonine kinase and phosphorylates the eukaryotic initiation factor 2α (eIF2α) in its activated state. The phosphorylation inhibits the ability of eIF2α to initiate mRNA biosynthesis and the overall protein load of the ER is lowered. At the same time, the translation of activation transcription factor 4 (ATF4) is induced and leads to the transcription of a gene set connected to apoptotic and adaptive responses during ER stress (Harding et al. 2000; Bi et al. 2005). Additionally, nuclear factor erythroid 2-related factor 2 (Nrf2) is phosphorylated by ATF4. Nrf2 phosphorylation induces its release from kelch-like ECH-associated protein 1 (Keap1),

which normally inhibits the Nrf2-induced expression of genes in oxidative stress and redox signaling that are now transcribed (Cullinan et al. 2003).

ATF6: ATF6 is tethered to the ER membrane and synthesized as a membrane protein to exclude it from the nucleus under healthy cell conditions. The dissociation from BiP because of ER stress induces the transport of ATF6 to the Golgi where the transmembrane domain is cleaved off and the protein is able to translocate to the nucleus (Shen et al. 2002). ATF6 regulates genes involved in protein quality control and ERAD (Adachi et al. 2008) and it stimulates the expression of X-box binding protein 1 (XBP1) that is targeted by IRE1α (Yoshida et al. 2001).

IRE1α: IRE1 is a type I ER transmembrane protein with a kinase- and an endoribonuclease activity, and it exists in two isoforms. IRE1α is ubiquitously expressed, whereas IRE1β expression is restricted to gastrointestinal epithelial cells (Tirasophon et al. 1998). ER stress-induced dissociation of BiP from the ER membrane induces IRE1α autophosphorylation. By following oligomerization, IRE1α’s RNAse activity is activated (Korennykh et al. 2009). Simultaneously, TNF receptor-associated factor 2 (TRAF2) E3 ligase and NFκB are recruited via the kinase domain of IRE1α. Additionally, C-jun-NH2-kinase (JNK) signaling pathways are mediated, affecting cell death induction or expression of prosurvival genes and/or cytokines (Urano et al. 2000; Kaneko et al. 2003). The unspliced XBP1 mRNA encodes an unstable, mainly cytoplasmic protein (XBP1u) with a DNA binding domain. The mRNA is spliced by the activated IRE1α form, which removes an intron of the XBP1 mRNA, generating an open reading frame shift. The shift of the mRNA open reading frame causes the translation of an alternative XBP1 C- terminus. This activated XBP1 form is a transcription factor with high potency for the control of genes that are related to ER membrane biosynthesis, protein import, ERAD, chaperoning functions and cell type-specific genetic programs (Acosta-Alvear et al. 2007). ER stress, induced by a mass of unfolded or misfolded proteins, is supposed to directly affect the protein transport through ER membranes. It is beneficial for cells to stop the insertion of newly synthesized proteins immediately with emerging ER stress, to deal with the misfolded proteins already inserted. It has been shown for proteins with “weak” signal sequences, especially for major prion protein (PrP), that they are rerouted to the cytosol immediately after the induction of misfolded protein stress (Kang et al. 2006). Proteins with “weak” signal sequences are predominantly dependent on SRP-independent, chaperone-ratcheting mechanisms to reach the ER. During acute ER stress, the pre-emptive quality control pathway 18

(pQC) is induced and stops the translocation of several membrane-bound and secreted proteins, before they fully enter the ER. Finally, the proteins are guided to proteasomal degradation during acute ER stress (Orsi et al. 2006; Kang et al. 2006). The limitation of the pool of luminal chaperones, like BiP, is one potential explanation for that mechanism. If the load of misfolded proteins arises in the ER lumen, the chaperones are excluded from their function in translocation, thereby providing a fast and effective delay in protein insertion. The processing capacity of the ER seems to be tightly intertwined with its translocation competence for SRP- independently transported proteins. Additionally, the SRP-dependent and the SRP-independent transport mechanism are both sensitive for energy depletion stress, although using different energy expenditures. The SRP and SRP-receptor are GTPases (Rapiejko & Gilmore 1997). The SRP-independent pathway, in contrast, is dependent on ATP as energy source. ATP is required for the chaperone function of Hsp70, the Sec62-Sec63/Kar2 ratcheting mechanism and TRC40/Get3 regulation. This is the reason why SRP-independent transport is more adjusted to the energy reserves of the cell. According to this, it was recently shown that upon glucose starvation, Get3 is capable to work as a holdase that could move with tail-anchored proteins to cytosolic aggregation sites (Powis et al. 2013). It is known since several years that ER-guided proteins appear and enrich unexpectedly in the cytosol. This phenomenon was extensively investigated with mammalian PrP, as 10-20% of the protein can be found in the cytoplasm of cells, in vivo (Drisaldi et al. 2003; Rane et al. 2004). PrP was shown to bear a signal sequence with average efficiency (Kim & Hegde 2002). The protein is translocated into the ER, glycosylated and GPI-anchored under healthy cell conditions. If the translocation initiation of the signal sequence fails There are two potential outcomes, which depend on downstream sequence elements (Kim et al. 2001; Kim & Hegde 2002; Stewart & Harris 2003). The first possibility is the afore mentioned release from the ribosome into the cytosol, where the protein is rapidly degraded by the proteasome to avoid aggregation (Ma & Lindquist 2001; Yedidia et al. 2001; Drisaldi et al. 2003). The arising mislocalized proteins (MLPs) are protected from aggregation by the heterotrimeric Bag6 complex, which binds specifically to their unprocessed, hydrophobic domains (Hessa et al. 2011). The Bag6/TRC35/Ubl4A complex seems to play a role in multiple other quality control pathways (Kawahara et al. 2013), e.g. it seems to be associated with newly synthesized proteins that are required for MHC class I loading (Minami et al. 2010). The pathway that leads the captured mislocalized proteins from Bag6 complex to the proteasome is poorly understood. The second possible outcome for PrP, if the sequence features a highly conserved downstream hydrophobic domain (HD), is its engagement in the translocon. This leads to the production of a 19

transmembrane isoform PrPctm (Kim & Hegde 2002). The emergence of both PrP isoforms seems to be due to signal sequence inefficiencies and can be prevented, if the signal sequence is substituted with a more efficient one (Rane et al. 2004; Rane et al. 2010). In 2014, Hegde and colleagues were able to investigate a mechanism, called Rapid ER stress-induced export (RESET), which leads to exclusion of GPI-anchored PrP isoforms under ER stress conditions. During RESET the proteins access the cell surface transiently and are then degraded in lysosomes (Satpute-Krishnan et al. 2014). There is evidence for the existence of a crosstalk between degradation pathways in the ER lumen and degradation pathways in the lysosomal compartment, which protects the cell from toxic effects of protein accumulation (Bustamante et al. 2013).

3.1.6 Antigen processing and the endoplasmic reticulum Specialized immune cells, called T-lymphocytes or T-cells, are the key players in cell-mediated immunity. Two distinct ways of antigen processing, with the ER as a central point, exist: an endogenous and an exogenous one. Both ways enable cells to present peptides on their surface for T-cell recognition (Fig 3.7.).

Figure 3.7│ The MHC class I and MHC class II antigen-presenting pathways. (A) Intracellular antigens (e.g. viral or tumor antigens) are processed by the immunoproteasome into peptides. The small peptides are transported into the ER and loaded on MHC class I molecules with the help of transporter associated with antigen processing (TAP). MHC class I complex is composed of a heavy chain and a β2-microglobulin (β2-m) molecule. MHC class I molecules on the cell surface are recognized by CD8+ T-cells. (B) Antigens from extracellular sources (e.g. bacterial antigens) are transferred into endosomes and degraded by endolysosomal enzymes into peptides. These peptides are able to bind the MHC-class-II groove by displacing the MHC class II-associated invariant chain peptides (CLIP) derived from MHC class II-associated invariant chain (Ii). CD4+ T-cells recognize peptides on MHC class II molecules located on the cell surface of antigen presenting cells. TCR, T-cell receptor; MIIC, MHC class II compartment (adapted from Kobayashi & van den Elsen 2012).

MHC class I The classical antigen processing pathway to present intracellular peptide fragments depends on MHC class I molecules. CD8+ T-cells are able to recognize aberrant cellular protein fragments presented on MHC class I molecules. Deviant intracellular peptides can originate from virus infected- or mutated cells. If identified as abnormal, these cells are killed by the T-cells. Nearly all nucleated cells carry MHC class I molecules on their surface. Misfolded, damaged or 21

unneeded proteins, ubiquitinated for proteasomal degradation, are cleaved into peptides, some of around nine amino acids length, which are able to fit in the binding cleft of MHC class I molecules. Another source for antigenic peptides are defective ribosomal products (DRIPs). It is assumed, that the translation of many proteins is prematurely terminated and therefore they are destined for proteasomal degradation (Yewdell et al. 1996). The ER import of peptides designed for MHC class I loading is dependent on transporter-associated with antigen processing (TAP), an ER transmembrane protein (Skipper et al. 1996; Hammond et al. 1995). MHC class I molecules are folded in the ER lumen with the help of luminal chaperones, interacting with TAP (through a complex also including tapasin, calreticulin, ERp57 and calnexin) during this process and, finally, they are loaded with peptides (reviewed in Raghavan et al. 2008). Peptide binding stabilizes MHC class I molecules and the complex is transported to the cell surface by the Golgi apparatus. There are multiple processes described that ensure the availability of glycosylated, ER-guided proteins (which are normally not present in the cytosol) for MHC class I processing. It has been shown for different glycoproteins that they are translocated into the ER, glycosylated, retrotranslocated into the cytoplasm via ERAD pathway (Chapter 3.1.4.) and loaded via TAP on MHC class I molecules (Mosse et al. 1998; Selby et al. 1999; Ferris et al. 1999; Bacik et al. 1997). Seven years ago, we discovered a so far unknown pathway, demonstrating that an MHC class I-targeted PSCA epitope was not guided into the ER at all. It seems that a proportion of ER-targeted proteins is degraded by the proteasome and loaded on MHC class I molecules before they reach the lumen of the ER (Schlosser et al. 2007).

MHC class II Specialized antigen presenting cells, like macrophages and dendritic cells, utilize MHC class II presentation, or the exogenous pathway. Antigens are taken up by endocytosis, loaded on MHC class II molecules and recognized by CD4+ T-lymphocytes. Their degradation is executed by acid-dependent proteases in endosomal compartments. The binding cleft of MHC class II molecules is blocked with a small protein, called invariant chain, to prevent binding of endogenous proteins during maturation. This protein also promotes the subsequent transport of MHC class II molecules in vesicles, to fuse them with late endosomes that contain exogenous peptides. Then, the invariant chain is broken down to a smaller fragment called CLIP and finally removed and replaced by an exogenous peptide stored in the endosome. MHC class II molecules are stabilized by peptide binding and the exogenous protein fragments are presented on the cell surface for T-cell recognition (Murphy et al. 2009). If endocytosed proteins are transported into the cytoplasm and presented on MHC class I molecules, the process is called

cross presentation, but the molecular details of this pathway are barely understood (Bevan 2006).

3.2 The (immuno-) proteasome 3.2.1 Composition and assembly of the mammalian (immuno-) proteasome Barrel-shaped, multimeric proteasomes can be found in all eukaryotes and archaea, and in some bacteria. The protein complexes are located in the nucleus and the cytoplasm of eukaryotic cells (Peters et al. 1994). The main function of this multi-catalytic, ATP-dependent, cylindrical protease complex is to degrade unneeded or damaged proteins in collaboration with the ubiquitin system (Chapter 3.2.2). It also modulates transcription factors and signal transducers, is responsible for the generation of a large portion of peptides presented on MHC class I molecules (Chapter 3.1.6), involved in regulation of cell cycle progression (Monaco & Nandi 1995) and important for adaption of cells to oxidative stress (Pickering et al. 2012). Four stacked, hetero heptameric rings form the core of the eukaryotic 20S proteasome. Each ring is composed of seven non-identical subunits. The two inner rings consist of seven β-subunits, whereas the two outer rings are composed of seven α-subunits. The three active proteases are arranged in the two β-subunit rings and named β1, β2 and β5. These subunits incorporate threonine residues and act as N-terminal nucleophilic proteases (Orlowski & Wilk 2000; Seemüller et al. 1995). The α-subunit rings preserve the “gate” structure of the proteasome through which the proteins have to enter the barrel. The 19S regulatory particle binds to the α- subunit ring of the 20S proteasome and this “cap” structure recognizes poly-ubiquitinated protein substrates and manages the initiation of degradation. Two bound 19S regulatory particles and the 20S proteasome form the 26S proteasome (Fig 3.8., Lodish et al. 2008). An 11S alternative regulatory particle (11S activator, PA28) can associate with the 20S proteasome as well. This particle is important for the degradation of antigenic, MHC class I presented peptides (Wang & Maldonado 2006). A third proteasome regulator involved in DNA repair, PA200, was discovered a decade ago (Ortega et al. 2005; Ustrell et al. 2002).

Figure 3.8│Scheme of the 26S proteasome. The 20S core complex is built by two inner α-rings (light green) and two outer β-rings (blue-green). The 19S regulatory particle is composed of a nine subunits base structure (dark blue and yellow) and a 10 subunits lid structure (red). The enzymatically active subunits β1, β2 and β3 are depicted in purple (adapted from Gomes 2013).

The 20S core particle The 20S core particle is equipped with three enzymatically active subunits β1, β2 and β5. All of them are arranged in the inner 20S core and operate with the same mechanism, but slightly differing substrate specificities. They are classified as chymotrypsin-like (β5, cleavage after hydrophobic, aromatic amino acids), trypsin-like (β2, cleavage after basic amino acids, lysine & arginine) and caspase-like (β1, peptidyl-glutamyl peptide-hydrolyzing, PHGH, cleavage after acidic and branched-chain amino acids) activities (Heinemeyer et al. 1997). The three subunits are constitutively expressed, but alternative β-forms have been identified. The surrogate β1i (LMP2), β2i (MECL-1) and β5i (LMP7) subunits are expressed in hematopoietic cells in response to stimulation with cytokines, like interferon γ (IFNγ) or tumor necrosis factor α (TNFα), thus exposure to inflammatory signals. Proteasomes with these alternatively incorporated subunits are called immunoproteasomes (Groettrup et al. 1996; Nandi et al. 1996; Aki et al. 1994). Immunoproteasomes generate more peptides that exhibit hydrophobic or basic C- terminal residues, thus are better suited for MHC class I binding, and with that the antigenic pool is expanded (Cascio et al. 2001). Interestingly, another β-subunit, β5t, was identified in 2007 and 24

is solely expressed in the thymic cortex. Incorporation of β5t reduces the chymotrypsin-like protease activity and seems to play a role in the positive selection of CD8+ T-cells. β5t-holding proteasomes are known as thymo-proteasomes (Takahama et al. 2012). All active β-subunits are expressed as precursors with N-terminal propeptides, which participate in the assembly mechanism of the 20S core, prevent premature activation and are autoproteolytically cleaved whereby the catalytic protein parts are exposed and active (Chen & Hochstrasser 1996; Schmidt et al. 1999). The proteasome assembly begins with the generation of an α-ring, which forms a template to associate with a corresponding heptameric β-ring. PAC1/PAC2 and PAC3/PAC4 (proteasome-associated chaperones) are two dimeric complexes associated with α-ring assembly. PAC1/PAC2 complex is known to bind the heptameric α-ring and is associated until the whole proteasome is assembled (Hirano et al. 2005). The proteasome maturation factor UMP1 is an assembly initiation factor for the β-rings in yeast cells (Ramos et al. 1998). The mammalian homologue is called proteasome maturation protein (POMP) and recruits precursor complexes to the endoplasmic reticulum, the main site for proteasome formation (Fricke et al. 2007). In the next step, two “half-proteasomes” associate their β-rings, forming the 20S proteasome. This assembly triggers the threonine-dependent auto-proteolysis of the β-subunits and they switch into their active forms. Salt bridges and hydrophobic interactions between conserved alpha-helices mediate the β-interactions (Witt et al. 2006).

19S regulatory particle The association of the bottom and top of the barrel-shaped 20S core particle with one or two respective 19S regulatory particles leads to the 26S proteasome formation. The 19S regulatory particle is made up of 19 subunits – nine subunits build a base that is bound to the α-subunits of the core particle and a lid-like structure is composed of the other ten subunits (Glickman et al. 1998; Glickman & Ciechanover 2002). Six of the nine base subunits hold ATPase-associated activity (AAA-family), are called proteasome-associated nucleotidases (PAN) and have archaebacterial homologues (Zwickl et al. 1999). ATP hydrolysis is needed to unfold substrates before they enter the 20S core particle, whereas the binding of ATP catalyzes all other steps required for degradation of polyubiquitinated proteins, for instance complex assembly, opening of the α-ring gate, translocation of the proteins and proteolysis (Smith et al. 2005; Liu et al. 2006). The molecular architecture of the 26S proteasome was elucidated in yeast in 2012 (Lasker et al. 2012; Lander et al. 2012). AAA-ATPases assemble to a heterohexameric ring, which lies adjacent to the 20S subunit. Rpt1/Rpt2, Rpt6/Rpt3 and Rpt4/Rpt5 dimers form a trimeric structure and dimerize via coiled coil N-terminal domains. The non-ATPase subunits Rpn1, Rpn2 and the ubiquitin receptor Rpn13 complete the base structure of the 19S regulator. 25

Via Rpn6 and Rpn5, the lid structure directly contacts the 20S core particle. The Rpn8/Rpn11 heterodimer is enclosed by a u-shaped structure, consistent of the Rpn9, Rpn5, Rpn6, Rpn7, Rpn3 and Rpn12 subunits. The Rpn11 deubiquitinating subunit is ideally located on the top of the lid to remove ubiquitin residues from the substrates right in front of translocation into the 20S core particle (Verma et al. 2002). The regulation of 20S gate opening by the 19S regulatory particle has recently been enlightened in archaeal cells (Rabl et al. 2008). It was described as a “key-in-a-lock” mechanism, where the C-termini of the 19S ATPases bind in pockets of the 20S core particle and induce gate opening.

11S activator The heptameric 11S activator, also known as PA28α/β, is able to bind the two ends of the 20S proteasome, as an alternative to 19S regulatory particle binding. It has no ATPase activity, but can strongly increase the peptidase activity of all three catalytically active β-subunits, promoting the degradation of short peptides, but not of larger proteins or ubiquitin-conjugated proteins (Ma Chu-Ping et al. 1992). PA28α/β expression is inducible with IFNγ, and it plays a role in MHC class I peptide generation together with the cytokine-induced assembly of the immunoproteasome (Ahn et al. 1995; Rechsteiner et al. 2000).

3.2.2 Ubiquitin and the proteasomal protein degradation process Ubiquitin (Ub) is a highly conserved 76 amino acid protein and involved in regulatory processes in the cytoplasm and nucleus of all eukaryotic cells (Schlesinger et al. 1975). Via a C-terminal critical glycine, it can be conjugated to other ubiquitin molecules or substrate proteins. Internal lysine residues are needed to create polyubiquitin chains. Substrate degradation, cell cycle progression, regulation of DNA repair and induction of immune response are the main tasks conducted by ubiquitin conjugation (Glickman & Ciechanover 2002). Tagging of proteins with Ub leads to their recognition by the 26S proteasome, and with that to their degradation into small peptides (Baumeister et al. 1998). There are three major groups of proteins needed to link polyubiquitin chains to proteins that are destined for degradation (Fig. 3.9).

Figure 3.9│Diversity and process of ubiquitination. (A) The process of substrate protein ubiquitination by E1, E2 and E3 enzymes. The E1 enzyme activates a ubiquitin molecule using ATP hydrolysis. The activated ubiquitin is conjugated to an E2 enzyme and finally ligated to the substrate protein with the help of a specific E3 enzyme containing either a HECT or a RING domain. (B) Different possibilities of ubiquitination. Letters refer to the amino acid residues used for ubiquitin chain elongation (adapted from Randles 2012).

Ubiquitin is prepared for conjugation in an ATP-dependent manner by ubiquitin activating enzymes, called E1s. ATP is utilized to generate a thioester at the C-terminus of ubiquitin with the help of an E1 enzyme, and with that the protein is activated. First, the active-site cysteine of the E1 enzyme receives the ATP-activated ubiquitin molecule, a thioester linkage is formed and AMP is released (Haas et al. 1982). So far, there is only a single mammalian E1 enzyme known (Handley et al. 1991). One of numerous E2 enzymes receives the ubiquitin at a conserved cysteine via a transesterification reaction from the E1 enzyme (Jentsch 1992). In the last step of this reaction, the ubiquitin forms an isopeptide bond composed of its C-terminal glycine residue (Gly76) and the epsilon amino group of a lysine side chain of the target protein. Additionally, an isopeptide bond can be formed between two ubiquitin molecules. The activated Ub (again via the Gly76) is attached to the Lys48 residue of a previously bound ubiquitin molecule. Ubiquitin- protein ligases (E3 ligases) are the key enzymes in this process. There are hundreds of E3 enzymes, containing either a HECT domain or a RING finger motif (Jackson et al. 2000). Ub conjugation can occur, dependent on the substrate, even without the involvement of the E1 enzyme. Typically, four ubiquitin molecules are attached via Lys48, to mark a target protein for degradation. Specific 19S subunits possess ubiquitin interacting motifs (UIM), e.g. Rpn10 has been demonstrated to bind multi-ubiquitin chains with the length of at least four ubiquitin moieties (Deveraux et al. 1994; Young et al. 1998). Other subunits, like Rpn13, are also known 27

to associate with ubiquitin (Husnjak et al. 2008). Instead of Lys48 polyubiquitination, leading to the above-mentioned proteasomal degradation, it has been shown that Lys11 can also be used for polyubiquitination. This kind of modification targets proteins to the lysosomal degradation pathway (Barriere et al. 2007) and influences the cellular tolerance to DNA damage, inflammatory immune responses, endocytotic processes and defective ribosomal protein synthesis (Pickart & Fushman 2004). Mono-ubiquitination is used to regulate DNA-repair, transcriptional regulation and protein localization (Hershko & Ciechanover 1998; Sloper-Mould et al. 2001; Weissman 2001).

3.2.3 Modification and inhibition of the proteasome As the main protease in the cytoplasm and nucleus of eukaryotic cells, the proteasome is the most important enzymatic complex for the degradation of a large number of proteins. The involvement of the proteasome in depletion of abnormal and misfolded proteins, as well as its participation in the regulation of cell cycle proteins, oncogenic factors and transcription factors, predestines it to be a sensitive part of cellular homeostasis (Glickman & Ciechanover 2002). It is no surprise that defects in the ubiquitin proteasome system are connected with a number of human diseases, for example neurodegenerative disorders (such as Alzheimer’s and Parkinson’s disease) and several types of cancer (Alves-Rodrigues et al. 1998; Mani & Gelmann 2005).

Co- and post-translational modifications of the proteasome Structural- and physiochemical properties of proteins, as well as their cellular localization, biochemical activity, turnover rate and possible protein-protein interactions, can be regulated by post-translational modifications (Mann & Jensen 2003; Witze et al. 2007). The proper assembly, function and activity of the 26S proteasome are likewise dependent on several co- and post- translational modifications of its subunits. Numerous post-translational modifications of the proteasome subunits have been reported in mammalian proteasomes. Phosphorylation, acetylation, ubiquitination, oxidative modifications and myristoylation were identified by large scale analyses, mostly on proteome level or in purified proteasomes (Wang et al. 2007; Ballif et al. 2008; Mayya et al. 2009; Rush et al. 2005; Beausoleil et al. 2004; Besche et al. 2014; Farout et al. 2006).

Proteasome inhibitors In 2003, the US Food and Drug Administration (FDA) approved the proteasome inhibitor bortezomib (Velcade, PS-341) for treatment of multiple myeloma, and since then five other 28

proteasome inhibitors have entered clinical trials (Molineaux 2012). Multiple myeloma cells are the cell type with the highest amount of protein secretion, because they synthesize and secret high levels of IgG and IgA (Bianchi et al. 2009; Cenci et al. 2006). This leads to permanently high intracellular protein load and secretion, accompanied by high proteasomal burden due to misfolded immunoglobulin chains, resulting in constantly active ERAD because of permanent ER stress. Hence, the unfolded protein response can be easily induced via proteasome inhibition in multiple myeloma cells (Obeng et al. 2006). As a consequence, amounts of bortezomib that kill multiple myeloma cells are not toxic to other cells. In general, proteasome inhibitors arise from different chemical structure classes and inhibit either subunits of the 20S core or the 19S regulatory particle. 20S core inhibitors are divided into two major classes, whether or not they form a covalent bond with the active site threonine of one of the three enzymatically active subunits. Inhibitor classes that form covalent bonds are peptide aldehydes, peptide boronates, peptide α’,β’-epoxyketones, peptide ketoaldehydes, β-lactones and oxatiazol-2-ones (Fig 3.10). Non-covalent inhibitor classes are cyclic peptides, noncyclic peptides and peptide isosteres and non-peptide inhibitors (Kisselev et al. 2012).

Figure 3.10│Structural classes of proteasome inhibitors. Chemical structure with inhibitory active groups depicted in red. PR39 and PR11: amino acid structure shown, binding sites so far unknown (adapted from Kreidenweiss et al. 2008).

MG132 belongs to the proteasome inhibitor class of peptide aldehydes and blocks the proteasome by forming a hemiacetal with the hydroxyl group of active site threonines. It is cell permeable and blocks the proteasome activity very efficiently and reversibly. Unfortunately, aldehydes are known to inhibit serine and cysteine proteases, too (Tsubuki et al. 1993) and are therefore rather unspecific. In contrast, peptide boronates, like bortezomib, are much more potent than aldehydes and more specific (Adams et al. 1998). Boronates form tetrahedral adducts with the active site threonines, stabilized by an hydrogen bond between an N-terminal 30

amino group of the threonine and the hydroxyl groups of the boronic acid (Groll et al. 2006). Boronate inhibition is likewise reversible, but the dissociation rate of bortezomib-proteasome complexes is much slower than the one of aldehyde-proteasome complexes. To date, the peptide α’,β’-epoxyketones are the most specific and potent proteasome inhibitors known. Epoxomicin was discovered in 1999 and holds no off-target effects. The N-terminal threonines of the proteasome active sites and the epoxyketon moiety of the inhibitor form a six-membered morpholino ring (Groll et al. 2000). The inhibition is highly threonine-protease-specific, because of the fact that serine and cysteine proteases do not have an α-amino group that is needed to form the morpholino adduct. Lactacystin is a lactam, a cyclic amide. It was the first non-peptidic proteasome inhibitor discovered. Lactacystin inhibits all three peptidase activities of the proteasome, the chymotrypsin- and trypsin-like activities irreversibly and the caspase-like activity reversibly (Fenteany et al. 1995).

3.2.4 Oxidative stress and proteasomes All aerobic organism produce reactive oxygen species (ROS) as a byproduct of their aerobic metabolism and oxidative phosphorylation (Chance et al. 1979; Hansford et al. 1997; Turrens & Boveris 1980). Certain chemical agents, as well as other exogenous stressors (e.g. ionizing- or non-ionizing radiation), induce ROS production (Vile et al. 1995; Leach et al. 2001; Chou et al. 2010; Watanabe et al. 2003). Moreover, ROS are generated during the pathogenesis of some diseases, e.g. Alzheimer’s disease (Hureau & Faller 2009) and the natural aging process (Leutner et al. 2001; Finkel & Holbrook 2000). If cells fail to neutralize intracellular ROS in time, the oxidative damage of lipids, proteins and DNA results in their functional abnormalities (Sedelnikova et al. 2010; Fruhwirth et al. 2007; Adibhatla & Hatcher 2010). Cells have evolved several defense mechanisms for oxidative stress reduction, due to the omnipresent risk of oxidative damage effects. Mechanisms for oxidative stress reduction are the production of antioxidants (like glutathione, vitamin A, C and E and flavonoids) and the expression of ROS neutralizing enzymes (like superoxide dismutase (SOD), catalase and glutathione peroxide). The appearance of oxidatively damaged proteins in the cytosol triggers the induction of protein repair pathways, trying to restore their function and the cellular viability (Finkel 2000; Barford 2004; Martindale & Holbrook 2002; Chen et al. 2008). If this fails, the proteins are designated for degradation or controlled accumulation to prevent a disruption of cellular processes (Davies 2001; Davies et al. 1987). It has been shown that unrestricted accumulation of oxidatively damaged proteins is associated with an impairment of the ubiquitin proteasome system (Bence et al. 2001). There are strong hints that neurological diseases are highly linked to oxidation- triggered protein-aggregation, partially because of raised ROS level in the brain (Butterfield & 31

Kanski 2001; Keller & Mattson 1998; Sayre et al. 2001). Keap1/Nrf2 pathway is known as the major regulator of an cytoprotective response against ROS (Kansanen et al. 2012). Nrf2 (nuclear factor erythroid 2-related factor 2) is a transcription factor that binds in antioxidant response elements (ARE) of its target genes. Under physiological conditions, Nrf2 is bound by kelch-like ECH-associated protein 1 (Keap1). This association leads to cytoplasmic Nrf2 localization and promotes its degradation by the ubiquitin proteasome system. Keap1 is a cysteine rich protein, and senses oxidative stress. The cysteine residues can be easily modified by oxidants and electrophiles, which leads to structural changes in Keap1 and release of Nrf2 (Taguchi et al. 2011). Thereby, Nrf2 is no longer degraded, can translocate into the nucleus and activate its antioxidant target genes (e.g. NAD(P)H quinone oxidoreductase 1 (NQO1), heme oxygenase 1 (HMOX1), glutamate-cysteine ligase (GCL) and glutathione S transferases (GSTs)).

Under moderate oxidative stress conditions, the 26S proteasome activity is stimulated by an unknown mechanism to clear the cytosol from mildly oxidized proteins (Grune et al. 2004; Ding et al. 2003). Upon persistent or acute oxidative stress, the 26S proteasome is inhibited maybe by accumulating aggregates of oxidized proteins, but definitely due to the fact that, in yeast and mammals, the 26S proteasome disassembles under oxidative stress conditions. 90% of all intracellular oxidatively damaged proteins are degraded by the 20S proteasome ATP- and ubiquitin-independent (Jung & Grune 2008). Disassembly of the 26S proteasome into 19S regulatory particle and 20S core leads to a more efficient clearance of irreparably damaged proteins (Shringarpure et al. 2003; Reinheckel et al. 2000). At this stage, the disassembly is reversible and the 26S proteasome reassembles if the oxidative stress ceases, leading to cellular recovery (X. Wang et al. 2010). 19S and 20S proteasome subunits are susceptible to oxidative modifications triggered by increased ROS in the cytosol, including 4-hydroy-2-nonenal (HNE) modification, carbonylation and S-glutathionylation. In general, all of these modifications result in a loss of proteolytic function (Bulteau et al. 2001; Ishii et al. 2005; Demasi et al. 2001; Farout et al. 2006). Therefore, throughout prolonged oxidative stress, proteasomal activities are inhibited and de novo proteasome synthesis starts, resulting in the formation of more functional 20S proteasome- and immunoproteasome complexes. They associate with 19S or PA28 regulatory particles, leading to a heterogeneous population of proteasomes, degrading oxidized proteins more efficiently and save the cell from oxidative damage. Activated 20S proteasomes, 20S- and 26S immunoproteasomes degrade oxidatively damaged proteins much more efficient than the standard 26S proteasome (Seifert et al. 2010; Pickering et al. 2010; Ding et al. 2003; Hussong et al. 2010).

3.3 The brain 3.3.1 Anatomy of the brain In all vertebrates, and most invertebrates, the brain represents the center of the nervous system. A few invertebrates, like sponges, jellyfishes and starfishes, do not have a brain, but diffuse neural tissue is present. The brain in vertebrates is protected by the skull and in close proximity to the primary sensory organs that encounter for the senses: hearing, tasting, viewing, smelling and balance. Two types of cells compose the brain of all species: neurons and glia cells. Neurons have the unique ability to send signals to specific target cells over a long distance via thin protoplasmic fibers, called axons. Signals travel as electrochemical pulses, called action potentials, along the axons. There exist several types of glia cells and they operate in structural support, metabolic support and insulation and guidance of development. Human brain consists of six main regions, namely telencephalon (cerebral hemisphere), diencephalon (thalamus and hypothalamus), mesencephalon (midbrain), cerebellum, pons and medulla. The medulla is involved in a wide variety of sensory and involuntary functions, like controlling the heart rate and contains many small nuclei (Kandel et al. 2000). The pons lies above the medulla, controlling sleep, respiration, swallowing, equilibrium, eye movement, facial expression and bladder function (Flemming 2006). The hypothalamus is a small, but highly complex and very important region, containing numerous small nuclei, with distinct neurochemistry and connection. It is involved in eating and drinking, regulation of sleep-wake cycles, temperature and the release of several hormones (Hofman & Swaab 1992). The thalamus is a collection of nuclei with various functions, too. It relays informations from and to the cerebral hemispheres and plays a role for the individual’s motivation. An action generating system, responsible for behavior like eating, drinking, excretion and copulation exists in the subthalamic area (Jones 2007). The cerebellum modulates the outcome of all other brain systems and precises them (Kandel et al. 2000). The cerebral cortex is involved in smelling and spatial memory and adopts the function of many different brain areas in humans (Puelles 2001). The hippocampus is only found in mammals and needed for complex events such as spatial memory and orientation (Salas et al. 2003, Fig 3.11).

Figure 3.11│Anatomy of the human brain. Main regions are depicted in different colors (adapted from The Anatomy of the Nervous System: From the Standpoint of Development and Function", by SW Ranson, WB Saunders, 1920).

3.3.2 The (immuno-) biology of astrocytes Astrocytes are highly differentiated cells of the nervous system, serving numerous functions, like nutrient supply of neurons, regulation of cerebral blood flow, orchestration of neuronal growth and differentiation, maintenance of extracellular glutamate levels and ion- and liquid balance (Freeman 2010; Benveniste 1998; Bush et al. 1999; Haydon 2000). Moreover, their involvement in inflammatory processes in the CNS, following pathogen invasion or chronic neurodegenerative diseases (Alzheimer’s disease, Parkinson’s disease), has been shown (Falsig et al. 2006). Primary astrocyte cell cultures are difficult to handle and go along with very high numbers of animals. The IMA 2.1 cell line was generated from astrocytes of wild type BALB/c mice and immortalized with the SV40 large T antigen. Those cells are able to respond to inflammatory cytokines and express functional monoamine oxidase-B (MAO-B; Schildknecht et al. 2011).

3.3.3 The role of the (immuno-) proteasome in the central nervous system (CNS) A series of different cells in the mammalian brain express proteasomes in their cytoplasm, nuclei, axons, dendrites and synaptic buttons, among them glial cells, pyramidal cells, Purkinje cells and hippocampal granular cells (Ding & Keller 2001; Mishto, Bellavista, et al. 2006; Mishto, Santoro, et al. 2006; Keller et al. 2002). Proteasome activity is elevated during neuronal differentiation in the brain. Decreased proteasome activity goes along with neuronal disorders 34

and aging (Keller et al. 2002; Keller et al. 2000). The induction of immunoproteasome subunits in the human brain is very low and the immunoproteasomes show an increased caspase-like and a lowered chymotrypsin-like activity (Piccinini et al. 2003). Immunoproteasomes are localized in hippocampal and cerebellar brain areas, more specifically in neurons, astrocytes and endothelial cells (Mishto, Bellavista, et al. 2006; Mishto, Santoro, et al. 2006). This indicates that the brain is, at least with the appropriate stimuli, able to trigger an immune response and to generate antigenic peptides. Microglial cells are known to become major antigen presenting cells and can respond to various pathogenic events (Xiao & Link 1998). In 2010, reduced immunoproteasome formation and accumulation of immunoproteasome precursors in the brain of LCVM-infected mice was shown (Kremer et al. 2010).

3.3.4 Lymphocytic choriomeningitis virus (LCMV) The LCMV is a negative single stranded RNA virus, which belongs to the Arenaviridae family. This family also comprises human-pathogenic viruses, such as Lassa-, Machupo- or Junin-virus (Emonet et al. 2006; Charrel & De Lamballerie 2003). Lymphocytic choriomeningitis virus is a natural occurring human and murine pathogen and it is transmitted, for instance, via human organ transplantation (Fischer et al. 2006) or contact with infected laboratory animals (Chadwick 2005; Rousseau et al. 1997). Infection of adults could lead to fever, malaises, headache, seizures and seldom to fatal meningitis (Barton & Hyndman 2000; Roebroek et al. 1994). LCMV infection of mice induces an efficient immune response in murine brains (Kang & McGavern 2008; McGavern et al. 2002).

4. Objectives Previous work by different researchers has shown that the inhibition of the proteasomal degradation machinery stabilizes ER-targeted precursor proteins (Rebello et al. 2004; Mitchell et al. 2006; Rane et al. 2004; Drisaldi et al. 2003). In 2003, Drisaldi et al. revealed that a small fraction (10-20%) of overexpressed cytosolic prion protein (PrP) is fully translated in the cytoplasm, degraded by the proteasome and has never been translocated into the ER lumen. This effect was not seen if cells endogenously express PrP. So far, there is no evidence that endogenously expressed, secretory proteins are stabilized by proteasome inhibition. Finally, they assumed that the precursor stabilization during proteasome inhibition is due to overexpression effects. For some proteins, it is known that the cellular mislocalization is enhanced because of ER stress (Orsi et al. 2006; Kang et al. 2006), rare mutations in signal peptides (Arnold et al. 1990; Hussain et al. 2013) or can be an effect of a mutant translocation machinery (Zimmermann et al. 2006). Interestingly, our group identified a strikingly new mechanism for MHC class I peptide generation, which utilizes ER-targeted proteins that have never been in the ER and are rapidly degraded by the 26S proteasome after their translation (Schlosser et al. 2007).

To study the fate of ER-targeted proteins in the face of proteasome failure, we generated an excellent model protein construct. N-terminally FLAG-tagged and C-terminally HA-tagged Prostate Stem Cell Antigen (FLAG-PSCA-HA) was overexpressed and used in [35S]-pulse-chase label experiments. With that construct, we could investigate PSCA processing from either the N- terminal (including its ER signal sequence) or the C-terminal part of the protein. With the help of this overexpression construct and different other approaches, we wanted to address the following questions:

 Are there other ER-targeted proteins whose precursors are detectably stabilized under proteasome inhibition? Can we find this effect in endogenously expressed protein?

 What happens with the enriched precursor proteins and where are they located in the cell?

 Does proteasome malfunction, induced by other methods than chemical inhibition, result in precursor stabilization, or is it a side effect?

 Can we find other cellular stress conditions that lead to the precursor stabilization?

 Which cellular mechanism leads to the precursor stabilization? Is the ER signal peptidase involved in this regulation?

In a second, independent project, we wanted to shed light on the phenomenon of enriched immunoproteasome subunit precursor proteins and less immunoproteasome formation in murine, LCMV-infected brains (Kremer et al. 2010).

For this purpose, we generated and analyzed antibodies against the immunoproteasome subunits β1i (LMP2), β2i (MECL-1) and β5i (LMP7). We used IMA 2.1, an immortalized murine astrocyte cell line, which is suitable as an in vitro model for the murine brain (Schildknecht et al. 2011). Our goal was to further identify and investigate the mechanisms that delay immunoproteasome subunit processing in cytokine-induced astrocyte cells. Additionally, we started to establish stably transfected astrocyte cell lines that overexpress MECL-1, LMP2 and LMP7 to have a permanent, cytokine-independent immunoproteasome subunit expression in those cells.

5. Material & Methods 5.1 Prokaryotic cells 5.1.1 Bacterial strain and culture conditions E. coli DH5α strain was used for all experiments. Cells were cultured in LB Medium (5 g/l yeast extract, 10 g/l peptone (both BD biosciences, Franklin Lakes, USA), 5 g/l NaCl (Roth, Karlsruhe, Germany; pH 7) or on LB agar plates (1.5% agar (BD biosciences, Franklin Lakes, USA). Ampicillin (100 µg/ml final) and kanamycin (50µg/ml final, both Sigma-Aldrich, St. Louis, USA) were added as antibiotics, if necessary.

5.1.2 Preparation and transformation of competent bacteria

E. Coli DH5α cells were grown until an OD600 of 0.8. Then, cells were centrifuged and the pellet was resuspended in cold 0.5 M CaCl2 (Roth, Karlsruhe, Germany) and placed on ice for 40 minutes. The competent cells were aliquoted and frozen at -80°C after pelleting and resuspending in CaCl2 and 30% glycerol (Roth, Karlsruhe, Germany). Competent bacteria were transformed by adding appropriate DNA amounts after an incubation time of 30 minutes on ice. Next, cells were heat shocked for one minute at 42°C. Afterwards, the cells were incubated for 2 minutes on ice and 1 ml of LB medium was added. Now, the cells were incubated under shaking conditions for one hour at 37°C and finally plated on LB agar plates with the respective antibiotic drug. After 12 hours of incubation at 37°C, colonies could be detected.

5.2 Eukaryotic Cells 5.2.1 Cell lines Eukaryotic cell lines used in this study are given in table 5.1. Cell line Description Medium ACHN Renal cell adenocarcinoma cell line EMEM, 10% FCS, 1% FCS B8-wt Mouse fibroblast cell line derived from BALB/c DMEM, 10% FCS, 1% P/S, mice B8-Db Mouse fibroblast cell line derived from BALB/c DMEM, 10% FCS, 1% P/S, mice, stably overexpressing H-2Db COS-7 Monkey kidney tissue cells (fibroblast-like) DMEM, 10% FCS, 1% P/S HEK293 Human embryonic kidney cell line DMEM, 10% FCS, 1% P/S HEK293T HEK293 stably expressing large T antigen of DMEM, 10% FCS, 1% P/S SV40

HEK293T-GFPu HEK293T cell line stable expressing GFP fused DMEM, 10% FCS, 1% P/S, c-terminal to an unstable CL-1 degron 0,5mg/ml G418 HeLa Human cervical cancer cell line DMEM, 10% FCS, 1% P/S HepG2 Human hepatocyte cell line DMEM, 10% FCS, 1% P/S HUVEC Umbilical vein/ vascular epithelium cell line F-12K, 10% FCS, 1% P/S IMA 2.1 Mouse astrocyte cell line derived from BALB/c DMEM, 10% FCS, 1% P/S mice LCL721 Human lymphoblastic cell line DMEM, 10% FCS, 1% P/S LCL721.174 Human lymphoblastic cell line without LMP2, DMEM, 10% FCS, 1% P/S LMP7 and less MECL-1 expression MC57 Mouse fibro sarcoma cell line derived from MEM, 5% FCS, 1% P/S C57BL/6 mice SW620 Colorectal adenocarcinoma cell line Leibovitz’s L15 medium, 10% FCS, 1% SDS Table 5.1│ List of eukaryotic cell lines. All cell culture media and additions were obtained from

Gibco Life Technologies (Carlsbad, USA). Cells were incubated at 37°C and 5% CO2.

5.2.2 Heat shock of eukaryotic cells

For heat shock procedure, cells were incubated at 42°C with 5% CO2 for one hour.

5.2.3 Cytokine stimulation of cell lines Cells were stimulated with either 200 U/ml interferon γ (IFNγ, Preprotech, Hamburg, Germany), 400 U/ml tumor necrosis factor alpha (TNFα, Preprotech, Hamburg, Germany), or both for 72 hours.

5.2.4 Lymphocytic choriomeningitis virus (LCMV) infection of cells To infect cells with LCMV, they were treated with a virus MOI (multiplicity of infection) of 1 in appropriate amounts of cell culture media for 18 hours.

5.2.5 Chemical inhibitors and inducers Name Function Concentration (Incubation time) 3-Methyladenine Autophagy inhibition 10 mM (2 hours) Brefeldin A Inhibits protein transport from the ER to the 20 µM (6 hours) Golgi Canavanine Arginine analogue, leads to misfolding of 5 µg/µl (4 hours)

proteins Dithiothreitol (DTT) Reduction of protein disulfide bonds, induction 3 mM (12 hours) of unfolded protein response Epoxomycin Proteasome inhibition 10 µM (30 min) Eeyarestatin-1 (Eey-1) Inhibition of VCP/p97 10 µM (2 hours) Epigallocatechin gallate (EGCG) Inhibition of BiP/GRP78 ATPase domain 10 µM (24 hours)

Hydrogen peroxide (H2O2) Induction of oxidative stress 100 µM - 1 mM (30 min) Lactacystin Proteasome inhibition 20 µM (30 min) MG132 Proteasome inhibition 10 µM (30 min) N-(Methoxysuccinyl)-Ala-Ala-Pro- ER signal peptidase inhibitor 250 µM (1 hour) Val-chloromethyl ketone Puromycin Inhibition of ribosomal translation 1 µg/µl (4 hours) Rapamycin Induction of autophagy 100 µM (2 hours) SIN-1 chloride Induction of peroxynitrite stress 1 mM (30 min) Spermine NONO-ate Induction of nitric oxide stress 1 mM (30 min) D,L-Sulforaphane Induction of Nrf2 10 µM (18 hours) 150 µM (30 min) Thapsigargin Inhibition of autophagosomal processes and 5 µM (6 hours) induction of ER stress Tunicamycin Inhibition of N-glycan synthesis and induction 10 µg/ml (6 hours) of unfolded protein response Table 5.2│ Inhibitors and inducers

5.2.6 RNA silencing For RNA knockdown in cells, SMARTpool ON-TARGETplus siRNA was used (Thermo Scientific, Waltham, USA). Every SMARTpool siRNA targets four different regions in the respective mRNA. The sequences are given in table 5.3. Cells were transfected with DharmaFECT transfection reagent 1 (Thermo Scientific, Waltham, USA) according to the manufacturer’s protocol. 24 to 48 hours after transfection, silencing was proven with real-time PCR analysis (Chapter 5.4.9). As a control, non-targeting siRNA against noncoding mRNAs (Thermo Scientific, Waltham, USA) was used.

Target gene RNA Sequences hPAC1 AGAAUGAUGAAAACGGAAUA CGGCCACGUUCUUCGGAGA UCGACAUGUUACCGAUUAU UGCUAGAACAACCGAAUAU hPSMB6 GGAGAUGUGUUACCGAUAC UAUCAUGGCCGUGCAGUUU GUAGAGCGGCAAGUACUUU CGAGAAGUUUCCACUGGGA hSPC18 CUAUUAGGAAUGGCAUAUA GUGAAAGUCCGAUUGUAGU UGACUAUCCUAAAAUUUAAG GCACUGGUUUUGCAUUAGUU hSPC21 GACAUCAAAUUUCUGACUA UGUAAUGGGUGCAUAUGUG UGAUAGAGGCUUGUACAAA UCCAAUAGUUCACAGAGUA mBAG6 GAUAUCAUCCGGACAAAUU CUGAAUGGGUCCCUAUUAU CCACGGGUCAUCCGGAUUU GAUCUGCGCUGCAAUCUAG Table 5.3│ siRNA used in this study for gene silencing, four RNAs mixed per gene

5.2.7 Transient transfection of eukaryotic cells with DNA plasmids All cells were grown until 90% confluence before transfection. HEK293T, HeLa and CoS-7 cells were transfected with Trans-IT®-LT1 transfection reagent (Mirus Bio, Madison, USA) according to the manufacturer’s protocol. In contrast, B8- and IMA2.1 cells were transfected with X-fect transfection reagent (Clontech, Mountainview, USA) according to the manufacturer’s protocol. After 24 to 48 hours, cells were harvested and used for subsequent experiments. Transfection efficiency was confirmed by transfection of a GFP-coding plasmid (pmax-GFP, Lonza, Basel, Switzerland) and FACS-analysis (Chapter 5.2.10). Plasmids used in this study are listed in table 5.4.

Plasmid Description Reference pmaxGFP Expression plasmid for GFP Lonza, Basel, Switzerland pcDNA3.1-PSCA-HA Expression plasmid for human PSCA, C-terminally Schlosser et al. HA-tagged pcDNA3.1-FLAG- Expression plasmid for human PSCA, C-terminally Schlosser et al. PSCA-HA HA-tagged and N-terminally FLAG-tagged pcDNA3-.1FLAG- Expression plasmid for human PSCA, C-terminally This study PSCA-HA-K13R HA-tagged and N-terminally FLAG-tagged, lysine in the ER signal peptide was mutated to arginine pcDNA3.1-Leptin-HA Expression plasmid for human Leptin, C-terminally This study HA-tagged pcDNA3.1-FLAG- Expression plasmid for human Leptin, C-terminally This study Leptin-HA HA-tagged and N-terminally FLAG-tagged pcDNA3.1-Prolactin- Expression plasmid for human Prolactin, C-terminally This study HA HA-tagged pcDNA3.1-FLAG- Expression plasmid for human Prolactin, C-terminally This study Prolactin-HA HA-tagged and N-terminally FLAG-tagged pCMV6-XL5-Leptin Expression plasmid for human Leptin Origene, Amsbio, Abingdon Oxon, UK pCMV6-XL5- Expression plasmid for human Prolactin Origene, Amsbio, Prolactin Abingdon Oxon, UK pCMV6-hVCP/p97 Expression plasmid for human VCP/p97 Ricarda Schwab, PhD pCMV6-hVCP/p97- Expression plasmid for dominant negative human Ricarda Schwab, PhD EQ VCP/p97 pcDNA3.1-hSPC18- Expression plasmid for human SPC18, N-terminally Kathrin Rothfelder, MA HA HA-tagged pcDNA3.1-hSPC18- Expression plasmid for human SPC18, N-terminally This study HA-S65A HA-tagged, dominant negative pcDNA3.1-hSPC21- Expression plasmid for human SPC21, N-terminally Kathrin Rothfelder, MA HA HA-tagged pcDNA3.1-hSPC21- Expression plasmid for human SPC21, N-terminally This study HA-S68A HA-tagged, dominant negative pcDNA3.1- Expression plasmid for human SPC22/23, N- Kathrin Rothfelder, MA hSPC22/23-HA terminally HA-tagged pcDNA3.1-His- Expression plasmid for 3 times FLAG-tagged FAT10 Chiu et al., 2007 3xFLAGFAT10

pcDNA3-Htt25- Expression plasmid for human Huntington, containing Mark Steffen Hipp, Cherry 25 CAG repeats and Cherry-tag Martinsried, Germany pcDNA3-Htt97- Expression plasmid for human Huntington, containing Mark Steffen Hipp, Cherry 97 CAG repeats and Cherry-tag Martinsried, Germany pcDNA3-hPA28α- Expression plasmid for human PA28α, N-terminally Andrea Kniepert, PhD myc myc-tagged pcDNA6-hPA28α- Expression plasmid for human PA28α, C-terminally Andrea Kniepert, PhD myc-his myc- and his-tagged pMACS-CD4 Expression plasmid for CD4 Miltenyi, Bergisch Glattbach, Germany +1 +1 pcDNA3-Ub2-Ubb Expression plasmid for UB2-UBB Nico Dantuma, Stockholm, Sweden pDest-myc-hNrf2 Expression plasmid for human NRF2, N-terminally Terje Johansen, myc-tagged Tromsø, Norway pcDNA3-hNrf1- Expression plasmid for human NRF1, 3 times N- Kobayashi et al., 2011 FLAG3 terminally FLAG-tagged pCMV6-Der1-myc Expression plasmid for human Derlin1, C-terminally Addgene, Cambridge, myc-tagged USA pCMV6-Syvn1-myc Expression plasmid for human Synoviolin1, C- Addgene, Cambridge, terminally myc-tagged USA pcDNA3-Mecl1- Expression plasmid for murine Mecl-1, 3 times This study FLAG3 FLAG-tagged pcDNA3-hCA4- Expression plasmid for human carbonic anhydrase 4, Rebello et al, 2001 secreted no GPI anchor pLMP2 Expression plasmid for murine LMP2 Steffen Frentzel pPuro Expression plasmid for puromycin resistance K. Strebhardt et al., 2004 pMECL-1 Expression plasmid for murine MECL-1 Marcus Groettrup, Konstanz, Germany pLMP7 Expression plasmid for murine LMP7 Marcus Groettrup, Konstanz, Germany pGPC Expression plasmid for LCMV glycoprotein Marcus Groettrup, Konstanz, Germany pcDNA3-HA- Expression plasmid for Ubiquitin, N-terminally HA- Kamitani et al., 1997 Ubiquitin tagged Table 5.4│ Plasmids used in this study.

5.2.8 Generation of stable transfected cell lines Cells were transiently transfected with the respective plasmids (Chapter 5.2.7). After 24 hours, puromycin (5 µg/µl, Sigma-Aldrich, St Louis, USA) was added to the cell culture media and exchanged every day. Five days later, cells were pooled and seeded in a density of 3 cells per well in 96-well plates containing growth medium supplemented with puromycin. After two weeks, the plates were screened for monoclonal cell colonies and labeled. The colonies were grown until density needed for the experiments.

5.2.9 Lysis of cells Cells were harvested and pelleted by centrifugation. After washing with PBS, an appropriate amount of lysis buffer (20 mM Tris (Sigma-Aldrich St. Louis, USA), 50 mM NaCl, 0,1% Triton X- 100 (Sigma-Aldrich, St. Louis, USA, pH 7.7) was added and the lysates were incubated on ice for 30 minutes. After 10 minutes of centrifugation (14000 rpm), the lysates were used for further experiments. Cells were lysed in RIPA buffer (50 mM Tris, 150 mM NaCl, 1% NP-40 (Sigma- Aldrich, St. Louis, USA), 0.5% sodium desoxycholate (Affymetrix, Santa Clara, USA), 0.1% SDS, PH 8), digitonin lysis buffer (1% digitonin (Calbiochem, Darmstadt, Germany), 25 mM Tris, 150 mM NaCl, 5 mM MgCl2) or high-salt lysis buffer (150 mM Tris, 50 mM NaCl, 0.1% Triton X-100, pH 7.7). For lysis of cells expressing CCR7, a special buffer was used (10 mM Tris-HCl, pH 7.5,

150 mM NaCl, 2 mM MgCl2, 10% Glycerol, 0.4% N-dodecyl maltoside (Sigma-Aldrich, St Louis, USA)). Protease inhibitor cocktail (Roche, Basel, Switzerland) and iodoacetamid (10 mM final, Sigma-Aldrich, St Louis, USA) were added to all lysis buffers directly before use. For phosphorylation monitoring, we used an additional phosphatase inhibitor (Phosphatase Inhibitor Cocktail 2, Sigma-Aldrich, St Louis, USA).

5.2.10 Fluorescence-activated cell sorting The fluorescence of GFP-expressing cells was measured with the Accuri C6 Flow cytometer. Cherry-expressing cells were measured and sorted for their red fluorescence with the FACSAria IIIu Flow Cytometer (both instruments: BD biosciences, Franklin Lakes, USA). Before the measurement, cells were harvested and washed three times with PBS. Cells were measured in appropriate amounts of FACS buffer (PBS, 5% FCS, 0.1% sodium azide (Sigma-Aldrich, St. Louis, USA).

5.2.11 Confocal fluorescence microscopy of cells Cells were grown on glass coverslips and stained with the desired antibodies (Table 5.5 and 5.6). In detail, cells were fixed with 4% paraformaldehyde solution for 15 minutes. After two 44

times of washing with PBS supplemented with 20 mM glycine, cells were permeabilized with PBS/ 20 mM glycine (Sigma-Aldrich, St. Louis, USA) / 3% BSA (Applichem, Chicago, USA) / 0.2% Triton X-100 for four minutes. Then, cells were washed two times with PBS/ 20 mM glycine/ 3% BSA and incubated with the first antibody in a 1:1000 dilution in PBS/ 20 mM glycine/ 3% BSA at room temperature for 30 minutes. After three times of washing with PBS/ 3% BSA, the cells were incubated with the secondary antibody in PBS/ 3% BSA in a 1:400 dilution at room temperature for another 30 minutes. Cells were washed three times with PBS and mounted for analysis with Fluoromount G (Southern Biotech, Alabama, USA). Afterwards, the fluorescent signals were measured in a confocal fluorescence microscope (LSM 510 META, Zeiss, Oberkochen, Germany). epitope type origin clone supplier Hemagglutinin monoclonal mouse 12CA5 Roche, Basel, (HA)-Tag Switzerland Climp-63 monoclonal rabbit G1/296 Enzo Lifescience, Lörrach, Germany Table 5.5│ Primary confocal microscopy antibodies used in this study. antibody type conjugate supplier Mouse anti-rabbit IgG polyclonal Alexa Fluor 488 Invitrogen, Oregon, USA Rabbit anti-mouse IgG polyclonal Alexa Fluor 546 Invitrogen, Oregon, USA

Table 5.6│ Secondary confocal microscopy antibodies used in this study.

5.2.12 Magnetic cell separation CD4+ cells were magnetically sorted by using anti-CD4 micro beads and magnetic LS columns (both Miltenyi, Bergisch-Glattbach, Germany) according to the manufacturer’s protocol.

5.2.13. Measurement of cell death Propidium iodide (PI) stainings, with a final concentration of 1 µg/ml PI in PBS, were performed to quantify cell death (Sigma-Aldrich, St. Louis, USA). Cells were harvested and washed three times in PBS and then incubated in the staining solution for five minutes. After this period, cells were measured by the Accuri C6 flow cytometer (BD biosciences, Franklin Lakes, USA). Secondary, cell death was quantified with the Apo Alert™ DNA Fragmentation Assay Kit (Clontech, Mountainview, USA) according to the manufacturer’s protocol and the fluorescence was measured by the Accuri C6 flow cytometer.

5.3 Proteins 5.3.1 Immunoprecipitation (IP) After cell lysis and lysate centrifugation, the supernatants were pre-cleared by adding 20 µl Protein A/G Beads (Santa Cruz, Dallas, USA) and incubated under agitation at 4°C for one hour. Next, the beads were removed via centrifugation. Following, either directly HA- or FLAG antibody coupled beads (Sigma-Aldrich, St. Louis, USA) were added for 18 hours or the respective unbound antibody (Table 5.7) was added for 1 hour. After 1 hour of antibody incubation, 40 µl of protein A/G beads were added and incubated rotating at 4°C over night. Afterwards, the beads were washed three times with RIPA buffer (50 mM Tris, 150 mM NaCl, 1% NP40, 0.5% sodium desoxycholate, 0.1% SDS) and then heated with Laemmli buffer (50 mM Tris-HCl, pH 6.8, 2% SDS, 0.0025% Bromphenol Blue (Applichem, Chicago, USA), 10% glycerol and 5% beta-mercaptoethanol (Merck, Darmstadt, Germany) at 95°C for five minutes. epitope type origin clone supplier/source Anti-alphatrypsin-1 polyclonal rabbit H-203 Santa Cruz, Dallas, USA Β2-microglobulin monoclonal mouse B2M-01 Santa Cruz, Dallas, USA Carbonic anhydrase IV polyclonal rabbit W. S. Sly, Missouri, USA FLAG-tag monoclonal mouse M2 Sigma-Aldrich, St. Louis, USA H-2Db,H-2Dd,H-2Kd,H-2Ld polyclonal rabbit G. Schmidtke, Konstanz, Germany HA-tag monoclonal mouse HA-7 Sigma-Aldrich, St. Louis, USA LCMV glycoprotein monoclonal mouse KL-25 F. Lehmann-Grube, Hamburg, Germany Leptin polyclonal rabbit A-20 Santa Cruz, Dallas, USA LMP2 polyclonal rabbit this study LMP7 polyclonal rabbit this study MECL-1 polyclonal rabbit this study myc-tag monoclonal mouse 9E10 Santa Cruz, Dallas, USA US11 polyclonal rabbit E. Wiertz, Utrecht, Netherlands Table 5.7│ Antibodies used for immunoprecipitation.

5.3.2 Deglycosylation of proteins For deglycosylation, either recombinant N-Glycosidase-F (Roche, Basel, Switzerland) or Endoglycosidase H (New England Biolabs, Danvers, USA) were used according to the manufacturer’s instructions. Then, protease inhibitor was added and the samples were

incubated at 37°C overnight. After deglycosylation, samples were boiled in Laemmli Buffer and 10 mM iodoacetamid for five minutes.

5.3.3 Radioactive labeling of proteins with [35S]-methionine/cysteine (pulse-chase) Radioactive sulfur ([35S])-labeled methionine/cysteine (Hartmann Analytical, Braunschweig, Germany) was used for metabolic labeling. Cells were starved in RPMI-1640 medium lacking methionine and glutamine (Sigma-Aldrich, St. Louis, USA) supplemented with 4 mM L-glutamine (Promega, Madison, USA) for one hour. Radio labeled [35S]-methionine/cysteine (0.25 mCi/ ml) was added for five minutes to the cells (pulse). After labeling, [35S]-methionine/cysteine was removed and the cells were washed twice with pre-warmed growth medium and further incubated in pre-warmed growth medium. After incubation times (chase), cells were washed with ice-cold PBS and lysed on ice for 30 minutes. The cell debris was removed by centrifugation (10min, 13000 rpm) and 5 µl of the radioactive supernatant were quantified with the Top Count NXTTM scintillation counter (Packard eBioscience, Waltham, USA) and the protein amount of the samples was equalized for the following experiments.

5.3.4 Separation of proteins with SDS-PAGE Proteins were separated by their size with sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). According to the size range of the proteins, 10-15% SDS gels were used (Table 5.8). “Spectra Multicolor Low Range” (Pierce Biotechnology, Rockford, USA) and “Pageruler Prestained” (Fermentas, Waltham, USA) protein ladders were used to estimate the molecular weight of the separated proteins. Gels were run with 80V until the samples arrive the border between the stacking- and the resolving gel. Then the run was completed with 120V.

Resolving Gels (15 ml) 10% 12% 15% 1.5 M Tris, pH 8.8, 0.4% SDS 5 ml 5 ml 5 ml 30% acrylamide (Applichem, Chicago, USA) 6.7 ml 8 ml 10 ml Water 8.3 ml 7 ml 5 ml 10% APS (Sigma-Aldrich, St. Louis, USA) 120 µl 120 µl 120 µl TEMED (Roth, Karlsruhe, Germany) 25 µl 25 µl 25 µl

Stacking Gels (10 ml) 0.5M Tris, pH 6.8, 0,4% SDS 2.5 ml 30% acrylamide 1 ml Water 6.5 ml 10% APS 80 µl TEMED 25 µl Table 5.8│ SDS gel recipe

After gel electrophoretic separation, SDS gels were either coomassie stained (coomassie stain solution: 0.1% Coomassie R250 (Sigma-Aldrich, St. Louis, USA), 10% acetic acid (VWR, Darmstadt, Germany), 40% methanol (Roth, Karlsruhe, Germany)) for 1 hour, destained with coomassie destainer (20% methanol, 10% acetic acid) overnight and dried (Model 583, Bio-Rad, Hercules, USA) or Western blotted (0.45µm, 120mA, 12Vh) on nitrocellulose membranes (Schleicher & Schuell BioSciences, Dassel, Germany). To evaluate the transfer efficiency and visualize the proteins, membranes were stained with Ponceau S (0.1% Ponceau S (Applichem, Chicago, USA) in 1% Acetic Acid) and afterwards destained with water. Then, membranes were blocked with blocking solution (5% milk powder (Roth, Darmstadt, Germany) in TBS with 0.5% Tween-20 (Sigma-Aldrich, St. Louis, USA; TBS-T) at room temperature for 30 minutes. After blocking, membranes were incubated with the according primary antibodies (Table 5.10) diluted in blocking solution at 4°C overnight. Subsequently, the membranes were washed three times for five minutes with TBS-Tween and the appropriate HRP-conjugated second antibody (Table 5.11) was incubated at room temperature for 2 hours. The membranes were washed three times with TBS-Tween and developed by using chemiluminescent substrate (SuperSignal West Pico or Femto, Thermo Scientific, Waltham, USA) and Chem Doc chemiluminescent imaging system (Bio-Rad, Hercules, USA). epitope type origin clone supplier/source dilution Derlin1 polyclonal rabbit Abcam, Cambridge, UK 1:500 His-tag monoclonal mouse 4A12E4 Invitrogen, Oregon, USA 1:1000 Leptin polyclonal rabbit Santa Cruz, Dallas, USA 1:1000 LMP2 polyclonal rabbit this study 1:1000 LMP7 polyclonal rabbit this study 1:1000 MECL-1 polyclonal rabbit this study 1:1000 myc-tag monoclonal mouse 9E10 Santa Cruz, Dallas, USA 1:1000 PAC1 polyclonal goat Abcam, Cambridge, UK 1:1000

Phospho- monoclonal mouse 4G10 Merck Millipore, Darmstadt, 1:1000 tyrosine Germany PI31 polyclonal goat Thermo Scientific, Waltham, 1:1000 USA poly monoclonal mouse 3B5H10 Sigma-Aldrich, St. Louis, USA 1:1000 glutamines PSMA6 (iota) monoclonal mouse K. Scherrer, Paris, France 1:2000 PSCA polyclonal rabbit Santa Cruz, Dallas, USA 1:1000 SPC18 polyclonal rabbit Proteintech, Chicago, USA 1:1000 Synoviolin-1 polyclonal rabbit Abcam, Cambridge, UK 1:500 α-tubulin monoclonal mouse AA13 Sigma-Aldrich, St. Louis, USA 1:2000 Ubiquitin monoclonal mouse FK2 Enzo Lifescience, Lörrach, 1:1000 Germany Table 5.9│ Primary antibodies used for Western blotting. epitope type origin clone supplier dilution mouse Ig polyclonal goat DakoCytomation, Baar, 1:2000 Switzerland rabbit Ig polyclonal swine DakoCytomation, Baar, 1:3000 Switzerland goat Ig polyclonal rabbit DakoCytomation, Baar, 1:2000 Switzerland HA-tag monoclonal mouse H6533 Sigma-Aldrich, St. Louis, USA 1:2000 FLAG-tag monoclonal mouse M2 Sigma-Aldrich, St. Louis, USA 1:1000 His-tag monoclonal mouse HIS-1 Sigma-Aldrich, St. Louis, USA 1:1000 myc-tag polyclonal rabbit Sigma-Aldrich, St. Louis, USA 1:1000 Table 5.10│ HRP-coupled antibodies used for Western blotting.

16,5% Tricine-SDS-polyacrylamide gels with 6 M urea (Roth, Darmstadt, Germany) were used to separate small sized proteins between 1-50 kDa (Schägger & von Jagow 1987). Before drying, gels were incubated in 5% glutaraldehyde for 30 minutes and then washed three times in 40% EtOH/ 10% acetic acid for five minutes each.

5.3.6 Analysis of radioactive proteins on SDS-gels After gel electrophoresis, gels were dried and exposed to a radiosensitive photo plate. After 24 hours, the radioactive bands were visualized using a phosphoimager (Molecular Imager® FX, Bio-Rad, Hercules, USA). 49

5.3.7 Non-equilibrium pH gel electrophoresis (NEPHGE) The respective samples were dissolved in 80µl NEPHGE sample buffer (9.5M Urea, 2% NP-40, 5% Servalyt® ampholines, pH 3-10 (Serva, Fitchburg, USA), 0.3% SDS and 5% beta- mercaptoethanol) at room temperature overnight. For separation of proteins by their isoelectric point, gel rods were prepared in 2 mm x 12,5 cm glass tubes (Bio-Rad, Hercules, USA). To a filtered (0,22 µm) solution of 5.5 g urea in 1.32 ml acrylamide stock solution (28.38% acrylamide; 1.62% bisacrylamide (Applichem, Chicago, USA), 2 ml 10% NP40, 0.5 ml Servalyt® pH 3-10, 2 ml H20, 28 µl 10% APS and 19 µl TEMED were added. Gel overlay solution (8 M urea, 2.5% Servalyt®, pH 3-10) was used to top the gel and it was left for polymerization for one hour. The samples were applied at the top of the gel and overlaid with 20 µl sample overlay solution (5 M urea, 5% glycerol, 2% ampholytes pH 2-4). The electrophoretic separation was performed in a 2D gel apparatus (Bio-Rad, Hercules, USA) at 400V for four hours. The anode tank was filled with 0.01 M H3PO4 (Riedel deHäen, Seelze, Germany) buffer and the cathode tank with 0.02 M NaOH (Roth, Darmstadt, Germany). After the run, the gel cylinder was loosened from the walls of the glass bar and equilibrated in 25 ml equilibration buffer (10% glycerol, 10% beta- mercaptoethanol, 2.3% SDS, 90 mM Tris-HCl, pH 6.8) for 30 minutes. The gel rod was fixed on the top of a 15% SDS gel with 1% agarose in Laemmli sample buffer (10% glycerol, 2.3% SDS, 0.0625 M Tris, pH 6.8, 5% beta-mercaptoethanol, 0.05 % Bromephenol blue) and the gel run was performed at 1100Vh overnight. For visualization of the proteins, the gel was stained with coomassie staining solution and dried (as described in 5.3.4).

5.3.8 Fractionating cellular proteins using osmotic pressure Cells were grown in 6cm plates and transfected with the respective plasmids. On the next day, they were starved in RPMI-1640 medium lacking methionine, cysteine and glutamine supplemented with 4mM L-glutamine, with or without MG132, for one hour. Radio labeled [35S]- methionine/cysteine (0.25 mCi/ml) was added to the cells (pulse) for 30 minutes. After the labeling, [35S]-methionine/cysteine was removed and the cells were washed once with 5 ml ice cold PBS. PBS was removed and hypotonic buffer 1 (20 mM HEPES, 15 mM KCl, 250 mM Sucrose, pH 7.4) was added and immediately removed. This step was repeated with hypertonic buffer 2 (20 mM HEPES, 300mM KCl, 250 mM Sucrose, pH 7.4). After removal of the hypertonic buffer, 500 µl of buffer 1 (plus protease inhibitors) were added and the cells were carefully scraped off the plate. The lysate was centrifuged at 800g at 4°C for 8 minutes. The supernatant was transferred in a new tube and the pellet washed with 150 µl buffer 1 and both tubes were centrifuged at 2000g for five minutes. The nuclei were removed in this step. Both supernatant 50

fractions were pooled and centrifuged at 21000g for 15 minutes. This step separates membrane and cytosolic fractions. The supernatant (cytosolic fraction) was used for immunoprecipitation. The pellet (membrane fraction) was lysed on ice in 500 µl lysis buffer (20 mM Tris, 50 mM NaCl, 0,1% Triton X-100, pH 7.7) with protease inhibitors for 30 minutes. The radioactive lysates were further used for immunoprecipitation.

5.3.9 Proteasome activity assay To test the activity of immunoprecipitated or purified proteasomes, the fluorogenic peptide substrate Suc-LLVY-AMC (Bachem, Bubendorf, Switzerland) diluted 1:150 in buffer S (50 mM

Tris, pH 7.5, 25 mM KCl, 10 mM NaCl, 1 mM MgCl2, 1 mM dithiothreitol, 0.1 mM EDTA) was added to the sample. Eventually, 100 μl of the substrate solution were added to 20 μl of each probe, in flat bottomed, black 96-well, plates and incubated at 37°C for 60 minutes. The fluorescence, emitted when the substrate was cleaved by active proteasomes, was measured with SpectraFluor Plus® plate reader (TECAN, Grödig, Austria; excitation wavelength: 360nm; emission wavelength: 465nm).

5.3.10 Sucrose gradient density centrifugation Lysates were loaded on a gradient of 15% to 40% sucrose in 0.1 M KCl buffer. The gradient was centrifuged at 40000 rpm at 4°C in a Beckman SW40Ti rotor for 16 hours. Then, the gradient was separated in 20 fractions with 600 μl each. The fractions were further methanol-chloroform precipitated according to the protocol of Wessel and Flügge (Wessel and Fluegge, 1984). The protein pellets were dried, dissolved in Laemmli buffer and heated to 95°C for 5 minutes.

5.3.11 Purification of antibodies from rabbit sera KLH-conjugated LMP2, LMP7 and MECL-1 peptides were immobilized with Sulfolink® Immobilization Kit (Thermo Scientific, Waltham, USA) and the newly generated antibodies were purified from sera of immunized rabbits as described in the manufacturer’s protocol.

5.4. DNA 5.4.1 Preparation of DNA Preparation of plasmid DNA from E. coli bacteria cultures was performed with the QIAprep™ Plasmid Mini Kit and the QIAprep™ Plasmid Maxi Kit (Qiagen, Venlo, Netherlands).

5.4.2 Agarose gel electrophoresis DNA and RNA were separated according to their size with 1% agarose gels (in TAE buffer, 40 mM Tris-acetate and 1 mM EDTA). The samples were mixed with 6x loading dye (0.1% Bromephenol blue, 30% glycerol in 10 mM Tris, pH 7) and loaded on the gel. Ethidium bromide was used to visualize the DNA or RNA after electrophoretic separation with 90V for one hour.

5.4.3 Extraction of DNA from agarose gels DNA was extracted from agarose gels by using the NucleoSpin® Gel and PCR clean-up kit from Macherey-Nagel (Düren, Germany).

5.4.4 Polymerase Chain Reaction – PCR Polymerase chain reaction was used to amplify DNA fragments with different sizes. Phusion high fidelity DNA polymerase (New England Biolabs, Danvers, USA) was chosen for the experiments, because of its efficiency and proof reading quality. Buffer conditions, cycling parameters, and primer temperature were chosen according to the manufacturer’s protocol. Afterwards, DNA fragments were purified with a PCR Cleanup Kit (Macherey and Nagel, Düren, Germany) as described in the protocol. A list of primers used in this study is given in table 5.11.

Primer Sequence FLAG-fwd GATCGAGAATTCGACTACAAAGACCATGACGGTGATTATAAAGATCATGACATCGAT TA

FLAG-rev CATGCGTCTAGATCATTCCTTGTCATCGTCATCCTTGTAATCGATGTCATGATCTTTA T

Mecl-1-fwd GATCGAAAGCTTATGCTGAAGCAGGCAGTGGAACC

Mecl-1-rev CATGCGGAATTCTTCCACCTCCATGGCCTGCACAG Prolactin fwd GGATCCATGAACATCAAAGG Prolactin rev GAATTCGCAGTTGTTGTTGTG Leptin fwd GCGGATCCATGCATTGGGGAACCCTG Leptin rev GCGAATTCGCACCCAGGGCTGAGGTC

Table 5.11│ Primers used for cloning in this study. All primers were ordered from Microsynth (Balgach, Switzerland). Newly introduced restriction enzymes cutting sites were shown in bold writing. 52

5.4.5 Restriction enzyme digestion Plasmids or DNA fragments were digested with digestion enzymes from Fermentas (Waltham, USA) according to the incubation conditions given in the manufacturer’s protocol.

5.4.6 Ligation of DNA fragments T4 DNA Ligase (Fermentas, Waltham, USA) was used to ligate DNA according to the manufacturer’s protocol.

5.4.7 Site directed mutagenesis of plasmids Single nucleotide mutations were inserted into DNA plasmids with the Quick Change® site directed mutagenesis Kit (Agilent, Santa Clara, California, USA). Mutagenesis primers are given in table 5.12.

Primer Sequence SPC18 S56A fwd TAGGGTGCTCAGTGGCGCCATGGAACCTGCATTTC SPC18 S56A rev GAAATGCAGGTTCCATGGCGCCACTGAGCACCACTA SPC21 S68A fwd GGTGCTGAGTGGCGCTATGGAGCCGGCC SPC21 S68A rev GGCCGGCTCCATAGCGCCACTCAGCACC PSCA K2A fwd AGGGATCCATGCGGGCTGTGCTGCTT PSCA K2A rev AAGCAGCACAGCCCGCATGGATCCCT Prolactin A28T fwd AACTTAAGCTTGGTACCTTGGACTACAAGGACGAC Prolactin A28T rev GTCGTCCTTGTAGTCCAAGGTACCAAGCTTAAGTT Prolactin A47T fwd GACTACAAGGACGACGTTGACAAGGGATCCATG Prolactin A47T rev CATGGATCCCTTGTCAACGTCGTCCTTGTAGTC Prolactin mut fwd CCTCAGAAATGTTCAGCGAGTTCGATAAACGGTATACCC Prolactin mut rev GGGTATACCGTTTATCGAACTCGCTGAACATTTCTGAGG Table 5.12│ Primers designed for Quick Change mutagenesis used in this study.

5.4.8 Quantitative real-time polymerase chain reaction (qRT-PCR) For real-time PCR the LightCycler® FastStart DNA Master SYBR Green I kit from Roche (Basel, Switzerland) was used according to the user manual. Primers are given in table 5.13.

Primer Sequence PAC1 fwd AGTGCAGTTGCTATGTTGCAG PAC1 rev GTAGCTTAGAACTGCTGCAGG Leptin fwd TGACTGACAAGCTGACACCT Leptin rev GACTGCCTTACCATCATACCC SPC21 fwd CAACGCCCAGCTCTATTACC SPC21 rev CTCTGTGAACTATTGGAATGTCTC SPC18 fwd ACTAGCAAGGTCGTGAGTCTC SPC18rev GACTTGATAATAGAGCTGCCGC PSMB6 fwd TGACTGACAAGCTGACACCT PSMB6 rev GACTGCCTTACCATCATACCC Table 5.13│ Primers used in real-time PCR experiments used in this study.

5.5 RNA 5.5.1 RNA extraction from eukaryotic cells All RNA samples were extracted with the NucleoSpin® RNA II Kit (Macherey-Nagel, Düren, Germany) according to the manufacturer’s protocol.

5.5.2 Synthesis of cDNA from RNA samples – Reverse Transcriptase Reaction The cDNA was synthesized from RNA by using the Promega Reverse Transcription System (Promega, Madison, USA) according to the manufacturer’s manual. 1 µg of RNA per reaction was used and the reaction mix was incubated at 42°C for one hour. Then, samples were heated to 95°C for five minutes. Finally, the cDNA was cooled down to 4°C.

5.6 Animals 5.6.1 Mice C57BL/6 (H-2b) mice were originally obtained from Charles River Laboratories (Kisslegg, Germany) and further bred in the animal facilities of the University of Konstanz. MECL-1 (Basler et al. 2006), LMP2 (Van Kaer et al. 1994) and LMP7 (Fehling et al. 1994) gene-targeted mice were kindly provided by Dr. John J. Monaco (Department of Molecular Genetics, Cincinnati Medical Center, Cincinnati, USA). For the experiments in this study, 8- to 12-week-old and sex-

matched mice were used. The review board of Regierungspräsidium Freiburg approved all animal experiments. For the analysis of immuno proteasome assembly, spleens of mice were removed and smashed.

Red blood cell lysis was performed by addition of red blood cell lysis buffer (155 mM NH4Cl, 10 mM KHCO3, 0.1 mM EDTA, pH 7.3) at room temperature for ten minutes.

5.6.2 Liver sections for immunohistochemical analysis Livers were removed from mice and immediately embedded in Tissue-Tec OCT (Sakura). Then, the livers were snap-cap frozen in liquid nitrogen and stored at –80°C until use. Cryostat sections (10-15 μm) were mounted on Superfrost plus® slides (Thermo Scientific, Waltham, USA), air dried and circled with a liquid blocker. Samples were immunologically stained as described in 5.2.11.

5.6.3 Immunization of rabbits for antibody generation Immunization of rabbits was performed in the animal facility of the University of Konstanz and sera were retrieved after initial priming and three times antigen boost every two weeks. Antigens are listed in table 5.14 and were purified as described in 5.3.10.

Name Peptide LMP2 CLPKFYDE LMP7 CKYGEAAL MECL-1 CAMEVE Table 5.14│ Antigenic peptides used for rabbit immunization.

5.7 Virus 5.7.1 Lymphocytic choriomeningitis virus - LCMV LCMV-WE was originally obtained from F. Lehmann-Grube (Hamburg, Germany) and propagated on the fibroblast line L929. Mice were infected with 200 pfu of LCMV-WE intravenously.

6. Results 6.1 The fate of ER-targeted proteins in the face of proteasome impairment 6.1.1 PSCA precursor protein is stabilized during proteasome inhibition 6.1.1.1 Treatment with chemical proteasome inhibitors stabilizes FLAG-PSCA-HA precursor protein Previously, it has been shown that prostate stem cell antigen (PSCA), residing in the ER, is cleaved rapidly after co-translational insertion by the ER signal peptidase (Schlosser et al., 2007). There was evidence that proteasome inhibition leads to a stabilization of the PSCA precursor protein. In this study, we used an N-terminally FLAG-tagged human PSCA construct. The C-terminal GPI anchor was replaced with an HA-tag (Fig 6.1A). This double-tagging strategy was necessary to follow the cleavage of the premature protein unambiguously, as there are no antibodies against the hydrophobic ER leader peptide for immunoprecipitation available. The CMV promoter-containing plasmid pcDNA3.1 was used as a vector for overexpression. We evaluated the translocation into the ER and the cellular localization of the overexpressed protein with confocal fluorescence microscopy. FLAG-PSCA-HA and N-terminally untagged PSCA-HA co-localize in the same way with the rough ER marker Climp-63 in transiently transfected cells. This indicates that FLAG-PSCA-HA is delivered physiologically into the ER and the N-terminal FLAG-tag does not interfere with the cellular localization of the protein (Fig 6.1B).

Fig 6.1│ N-terminal FLAG-tagging does not alter the localization of PSCA-HA protein. (A) Scheme of N-terminally FLAG-tagged and C-terminally HA-tagged PSCA precursor protein. (B) Comparison of PSCA localization pattern in HEK293T cells expressing PSCA-HA or FLAG-PSCA-HA. Cells were stained with polyclonal antibodies against the HA-tag (red) and Climp63 (green) and visualized by confocal fluorescence microscopy; bar: 10 μm. The experiments were repeated twice with similar outcome.

Next, we wanted to investigate the processing of FLAG-PSCA-HA in the ER and Golgi network by detecting its glycosylation. For this reason, we examined the glycosylation pattern of FLAG- PSCA-HA protein in comparison to PSCA-HA protein with short-term pulse-chase label experiments for up to 60 minutes. HEK293T cells transiently expressing FLAG-PSCA-HA or PSCA-HA were starved in cell culture medium lacking methionine and cysteine for one hour. After 5 minutes of labeling with [35S]-methionine/cysteine, we followed the processing of the ER leader for up to one hour of chase time. After anti-HA immunoprecipitation, the glycosylated PSCA proteins were separated by size with the help of 16.5% Tricine-SDS-polyacrylamide gels (Schägger & von Jagow 1987) and visualized by autoradiography. Unglycosylated FLAG-PSCA- HA and its three glycosylated isoforms were detectable and there was no difference in the glycosylation pattern of FLAG-PSCA-HA as compared to FLAG-HA (Fig 6.2A). Endoglycosidase H (EndoH) is a deglycosylase that cleaves very specifically asparagine-linked and mannose-rich, but not highly processed, complex oligosaccharides from proteins. This means that proteins are resistant to Endoglycosidase H cleavage after their final processing in the Golgi (where high complex glycosylation steps took place), but not as long as they reside in the ER lumen. Endoglycosidase H was used to clarify if FLAG-PSCA-HA protein is transported into the ER

lumen and glycosylated in an appropriate way in untreated and MG132 treated cells. In untreated cells, we detected three PSCA protein bands (13 kDa, 16 kDa and 19 kDa) after immunoprecipitation against the HA-tag. EndoH treatment caused deglycosylation of the protein species and an 11 kDa band was detectable. Immediately after the labeling and after 5 minutes of chase, a 13 kDa band was detectable in EndoH treated cells, too. This band seems to represent the precursor protein, which is not glycosylated and located in the cytoplasm. MG132 treated cells displayed a slightly reduced amount of glycosylated PSCA species as compared to untreated cells after 30, 60 and 120 minutes of chase time. Additionally, the 13 kDa PSCA precursor protein band is stabilized in the cytosol for up to 120 minutes during proteasome inhibition (Fig 6.2B). It seems that FLAG-PSCA-HA protein is sufficiently glycosylated, indicating that the N-terminal FLAG-tag does not interfere with the transport of the protein into the ER lumen.

Fig 6.2│ N-terminal FLAG-tagging does not alter the glycosylation of PSCA-HA protein. (A) HEK293T cells transfected with FLAG-PSCA-HA or PSCA-HA were pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Proteins were then immunoprecipitated with mAb against the HA-tag and separated on 16.5% Tricine-SDS-polyacrylamide gels. The proteins were visualized by autoradiography. (B) HEK293T cells transfected with FLAG-PSCA-HA or PSCA-HA were pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. During starvation, labeling and chase period, cells were treated with the proteasome inhibitor MG132 or left untreated. Then, proteins were immunoprecipitated with mAb against the HA-tag, deglycosylated with Endoglycosidase H or left untreated and separated on 16.5% Tricine-SDS- polyacrylamide gels. The proteins were visualized by autoradiography. The experiments were repeated twice with similar outcome.

To further identify the effect of proteasome inhibition on FLAG-PSCA-HA precursor, we pulse- chase labeled HEK293T cells transiently overexpressing FLAG-PSCA-HA with [35S]-methionine/ cysteine (5 min label time) and chased the proteins for up to one hour. We monitored the stability of FLAG-PSCA-HA with anti-FLAG immunoprecipitation. After precipitation, samples were deglycosylated with PNGase F (to completely deglycosylate the proteins), separated by 16.5% Tricine-SDS-polyacrylamide gels and visualized with autoradiography. A 13 kDa band represents full length FLAG-PSCA-HA precursor protein. It disappears during the chase because of the removal of the ER signal peptide by the ER signal peptidase. The FLAG-tag is located N-terminally of the ER signal peptide. Anti-FLAG immunoprecipitation precipitates the uncleaved precursor protein but not the mature protein. In untreated cells, 92% of FLAG-PSCA- HA protein, produced in the 5 minutes labeling time, is cleaved after 10 minutes of chase. The preprotein was stabilized for up to 60 minutes in cells treated with MG132, epoxomycin or lactacystin (Fig 6.3A). We detected the same result by using HA-specific immunoprecipitation. The upper band represents the 13 kDa precursor protein of FLAG-PSCA-HA and the lower 11 kDa band the mature protein without ER leader and FLAG-tag. Proteasome inhibition leads to precursor stabilization for up to 60 minutes (Fig 6.3B). To exclude a cell line-specific effect, the experiment was repeated in HeLa cells and the same effect was visible (Fig 6.3C,D). Notably, some of the pulse-chase labeling experiments showed a diffuse protein band with the size of 2-3 kDa. Because of the inconsistence of this effect it was impossible to determine if the band represents the FLAG-tagged ER leader. Mass spectrometry analysis of the band revealed no result.

Fig 6.3│ Inhibition of the proteasome leads to stabilization of prostate stem cell antigen (FLAG-PSCA-HA) precursor protein. (A) HEK293T cells transfected with FLAG-PSCA-HA were starved for one hour, pulse-labeled with [35S]- methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were additionally treated with the indicated proteasome inhibitors for the last 30 minutes of starvation, during labeling and chase. Lysates were immunoprecipitated with mAb against the FLAG-tag and deglycosylated with PNGase F. Then, proteins were separated on 16.5% Tricine-SDS-polyacrylamide gels and the visualized by autoradiography. (B) The same experimental setup as in B was performed, but with using mAb against the HA-tag for immunoprecipitation. (C) Experiments were performed as in A and B but with HeLa cells. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

Brefeldin A inhibits the formation of transport vesicles from the ER to the Golgi, thereby interfering with secretion of ER resident proteins. 6 hours prior to the 35[S] pulse-chase labeling experiments described above, cells were pretreated with brefeldin A. The FLAG-PSCA-HA precursor protein is stabilized only by addition of MG132 (Fig 6.4A) indicating that enrichment of premature FLAG-PSCA-HA protein caused by proteasome inhibition is not an effect of aberrant

secretion. Pulse-chase experiments require one hour of methionine/cysteine starvation of cells that could induce autophagy pathways. To eliminate potential side effects of this pathway, we skipped the starvation and labeled the cells directly. The labeling efficiency was very low, but FLAG-PSCA-HA precursor protein was clearly stabilized during proteasome inhibition (Fig 6.4B).

Fig 6.4│FLAG-PSCA-HA precursor protein stabilization upon proteasome inhibition is not caused by changes in protein secretion or starvation-induced. (A) HEK293T cells expressing FLAG-PSCA-HA were treated with brefeldin A for 6 hours. Cells were starved for one hour, pulse-labeled for 5 minutes with [35S]-methionine/cysteine and chased for the indicated time periods in the presence of brefeldin A. For the last 30 minutes of starvation and during the labeling and chase period, cells were treated with the proteasome inhibitor MG132 as indicated. Lysates were immunoprecipitated with a mAb against the FLAG-tag. Proteins were separated on 16.5% Tricine-SDS- polyacrylamide gels and visualized by autoradiography. (B) Experiments were performed as in A, but without brefeldin A treatment and 1h of starvation before [35S]-labeling. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.1.2 Genetic proteasome silencing stabilizes FLAG-PSCA-HA precursor protein We used siRNA mediated proteasome silencing to exclude the unlikely possibility that stabilization of the FLAG-PSCA-HA precursor is caused by side effects of chemical proteasome inhibition, e.g. inhibition of the ER signal peptidase catalytic activity. In a first approach, we used an siRNA pool against proteasome assembling chaperone 1 (PAC1), a mammalian chaperone that plays a role in the initial assembly of the 20S proteasome. PAC1 siRNA knockdown has been shown to inhibit the 26S proteasome (Hirano et al. 2005). We treated HEK293T cells with PAC1 siRNA and analyzed intracellular PAC1 mRNA levels with qRT-PCR analysis. PAC1 mRNA amount was reduced to less than 5% in comparison to control siRNA treated cells, 48 hours after siRNA silencing (data not shown). The reduction of PAC1 protein after siRNA knockdown was confirmed in anti-PAC1 Western blots. After PAC1 siRNA knockdown, the 29 kDa PAC1 band disappeared (Fig 6.5A). 48 hours of PAC1 silencing in HEK293T cells had no effect on FLAG-PSCA-HA precursor protein stability (Fig 6.5B). We identified the amount of high molecular weight polyubiquitin conjugates in HEK293T cell lysates in anti-ubiquitin Western Bots after 24, 48 and 72 hours of PAC1 siRNA knockdown. The process of siRNA transfection itself 61

caused a high amount of polyubiquitin conjugates in the cells, but the amount of polyubiquitinated conjugates was not enhanced after 24, 48 or 72 hours of PAC1 after siRNA knockdown in comparison to control siRNA knockdown (Fig 6.5C). According to this, cells treated three times with PAC1 siRNA in 48 hours intervals showed low levels of cell death (20% dead cells measured by PI staining), implying that the 26S proteasome is not properly inhibited. By comparison, after 8 hours of MG132 treatment 95% of HEK293T cells are dead. The results indicate that, with our experimental setup, PAC1 siRNA knockdown did not impair the 26S proteasome activity.

Fig 6.5│Genetic silencing of proteasome assembly chaperone 1 (PAC1) did not stabilize FLAG-PSCA-HA precursor protein. (A) HEK293T cells were transfected with PAC-1-specific- or control siRNA pools. After 48h, the PAC1 expression was visualized with Western blot analysis using a polyclonal antibody against PAC1. Tubulin was used as a loading control. (B) HEK293T cells were treated two times with PAC1-specific- or control siRNA pools (48 hours interval) and transfected with FLAG-PSCA-HA. Then, cells were pulse-chase labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Lysates were immunoprecipitated with a mAb against the FLAG-tag and deglycosylated with PNGase F. Afterwards, the proteins were separated on 16.5% Tricine-SDS- polyacrylamide gels and the proteins were visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. (C) HEK293T cells were transfected with PAC-1-specific- or control siRNA pools. After 24 h, 48 h and 96 h, cells were lysed. Proteins were separated on 10% SDS-polyacrylamide gels and analyzed with Western blots using a mAb against ubiquitin. Control cells were incubated with the proteasome inhibitor MG132 for 6 hours. Tubulin was used as a loading control. The experiments were repeated twice with similar outcome. 62

It has been shown that proteasome subunit beta type-6 (PSMB6) siRNA knockdown inhibits proteasomal degradation significantly (oral communication with Mark Steffen Hipp, Munich). An siRNA pool directed against PSMB6 was used to silence the proteasome activity. The intracellular amount of PSMB6 mRNA was reduced to 9% in cells treated with PSMB6 siRNA as compared to cells treated with control siRNA (data not shown). 24 hours after PSMB6 siRNA knockdown, an accumulation of polyubiquitin conjugates was detectable in HEK293T cells as compared to cells treated with control siRNA in anti-ubiquitin Western blots (Fig 6.6A). A HEK293-GFPu cell line, which stably expresses a GFP variant containing an CL-1 degron (Bence et al. 2001), was used to further monitor the proteasome inhibition efficiency of PSMB6 siRNA knock down via flow cytometric analysis of GFP fluorescence. At physiological levels of proteasome activity, overexpressed GFP is degraded directly after its expression. During proteasome impairment, the GFP load increases in cells. HEK293T-GFPu cells treated with PSMB6 siRNA showed an enhanced GFP fluorescence, in contrast to control siRNA treated cells, after 48 hours of transfection (Fig 6.6B). The knockdown was performed three times (with 48 hours intervals) before we monitored the stability of FLAG-PSCA-HA precursor protein in pulse-chase experiments with anti-FLAG immunoprecipitation. The precursor was stabilized for up to 60 minutes in cells after PSBM6 knockdown. Next, we silenced the two catalytically active ER signal peptidase subunits (SPC21 and SPC18) by using siRNA pools against them. The mRNA levels were reduced to 10% as compared to control siRNA treated cells (data not shown). We silenced the subunits three times every 48 hours prior to the pulse-chase labeling experiments. Inhibition of the ER signal peptidase activity stabilized FLAG-PSCA-HA precursor protein was stabilized for up to 60 (Fig 6.6C).

Fig 6.6│ Genetic depletion of 20S proteasome subunit beta type-6 (PSMB6) stabilizes FLAG-PSCA-HA precursor protein. (A) HEK293T cells were treated withPSMB6-specific- or control siRNA pools. 24 hours after transfection, cells were lysed and ubiquitin expression was visualized in Western Blots probed with a mAb against ubiquitin. Tubulin was used as a loading control. (B) Stably GFPu transfected HEK293 cells were treated two times (48 hours interval) with the indicated siRNA pools and analyzed for GFP fluorescence by flow cytometry. (C) HEK293T cells were treated with SPC18- and SPC21-, PSMB6-specific- or unspecific siRNA pools two times (48h interval) and transfected with FLAG- PSCA-HA. Cells were pulse-labeled with [35S]-Methionine/Cysteine for 5 minutes and chased for the indicated time periods. Lysates were immunoprecipitated with a mAb against the FLAG-tag and deglycosylated with PNGase F. Then, proteins were separated on 16.5% Tricine-SDS-polyacrylamide gels and visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.1.3 Overexpression of aggregation-prone proteins stabilizes FLAG-PSCA-HA precursor protein We used a third, distinct approach to inhibit the function of the proteasome: protein aggregates. Aggregates of Huntingtin (Htt) proteins are known to impair the proteasome (Bence et al. 2001). We overexpressed a C-terminally Cherry-tagged Huntingtin exon 1, containing either a short, nonpathogenic glutamine repeat part (HttQ25) or a highly aggregation-prone, pathogenic repeat

region (HttQ96). Overexpression of ChFP-HttQ96 in HEK293T induced aggregate formation of Huntington proteins in the cytoplasm of cells, which were detectable with confocal microscopy analysis. In contrast, there were no protein aggregates detectable in ChFP-HttQ25 overexpressing cells (Fig 6.7A/B).

Fig 6.7│Overexpression of highly aggregation-prone, CherryFP-tagged HuntingtinQ96 protein provokes protein aggregates. (A, B) Visualization of HttQ96-ChFP or HttQ25ChFP (control) in transiently transfected HEK293T cells by confocal fluorescence microscopy; bar: 10 μm. The experiments were repeated twice with similar outcome.

HEK293T cells were transiently transfected with HttQ97 and sorted for ChFP expression by fluorescence-associated cell sorting. Anti-polyQ Western blot analysis revealed that ChFP-Htt97 protein is expressed with the calculated size of 50 kDa and we detected aggregates with the molecular size of 80 kDa (Fig 6.8A). To visualize the impairment of proteasome activity after ChFP-HttQ97 aggregate formation, we transiently transfected cells with ChFP-HttQ97 (or as a control with ChFP-HttQ25) and HA-tagged ubiquitin for 48 hours. We visualized the ubiquitination state of cellular proteins with anti-ubiquitin immuno blots. A slight increase in the amount of ubiquitinated proteins in ChFP-HttQ97 transfected cells in contrast to ChFP-HttQ25 transfected cells was detectable (Fig 6.8B). Unfortunately, this result was not reproducible and the transfection efficiency of ChFP-HttQ97 and ChFP-HttQ25 fluctuated between 5% and 50% in different experiments. To enrich cells positive for ChFP-HttQ97 (and ChFP-HttQ25) expression, we tried to use fluorescence-associated cell sorting (FACS) and coexpression of CD4 with

subsequent magnetic-associated cell sorting (MACS) against CD4. Both enrichment methods impaired the cell viability and prevented subsequent [35S]-pulse-chase labeling (data not shown).

Fig 6.8│ Overexpression of highly aggregation-prone, CherryFP-tagged HuntingtinQ96 protein does not induce an enrichment of poly-ubiquitinated protein conjugates. (A) HEK293T cells transfected with HttQ96-ChFP or HttQ25- ChFP (control) were analyzed on Western blots probed with a mAb against poly glutamines. The experiments were repeated twice with similar outcome. (B) HEK293T cells were transfected with Ubiquitin-HA or Ubiquitin-HA and HttQ96-ChFP or HttQ25ChFP. Lysates were immunoprecipitated with a mAb against the HA-tag followed by Western blot analysis with a mAb against mono- and polyubiquitinated proteins. Tubulin was used as a loading control.

+1 C-terminally extended mutant ubiquitin (Ub2-UBB ) is known to form aggregates in Alzheimer’s disease, which inhibit the proteasome activity (Lindsten et al. 2002). An enrichment of ubiquitin- conjugated proteins was detectable in anti-ubiquitin Western blot analysis of cells +1 overexpressing Ub2-UBB for 36 hours (Fig 6.9A). Beside the polyubiquitin conjugates, mono-, +1 di- and tri-ubiquitinated forms of Ub2-UBB were detectable. 36 hours of Ub2-

UBB+1overexpression lead to FLAG-PSCA-HA precursor protein stabilization for up to 60 minutes in contrast to untreated cells (10 min; Fig 6.9B).

+1 Fig 6.9│ Overexpression of highly aggregation-prone Ub2-UBB protein stabilizes FLAG-PSCA-HA precursor protein. +1 (A) HEK293T cells were transfected with FLAG-PSCA-HA and/or Ub2-UBB . Lysates were analyzed on Western blots using a mAb against mono- and polyubiquitinated proteins. Tubulin was used as a loading control. (B) HEK293T +1 35 cells were transfected with FLAG-PSCA-HA and/or Ub2-UBB 36h prior to pulse-labeling with [ S]- methionine/cysteine (5 min labeling time) and chased for the indicated time periods. Lysates were immunoprecipitated using a mAb against the FLAG-tag followed by deglycosylation with PNGase F. Then, the proteins were separated on 16.5% Tricine-SDS-polyacrylamide gels and visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.1.4 Oxidative stress stabilizes FLAG-PSCA-HA precursor protein In the next set of experiments, we investigated the effect of oxidative stress on FLAG-PSCA-HA precursor protein stability in HEK293T cells. During oxidative stress, an overload of misfolded or ROS-damaged proteins in the cytosol inhibits the degradation capacity of the 26S proteasome (reviewed in Aiken et al. 2011). We treated HEK293T (Fig 6.10A) or HeLa (Fig 6.10B) cells, transiently expressing FLAG-PSCA-HA, with 750 µM H2O2 and performed pulse-chase labeling experiments as described before for MG132. FLAG-PSCA-HA precursor protein was stabilized for up to 60 minutes in cells treated with 750 µM H2O2. The stabilization was dose-dependent and started at a concentration 200 µM H2O2 (Fig 6.10C).

Fig 6.10│Oxidative stress leads to stabilization of FLAG-PSCA-HA precursor protein. (A) HEK293T cells transfected with FLAG-PSCA-HA were pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. If indicated, cells were treated with H2O2 for the last 30 minutes of starvation, during labeling and chase. Lysates were immunoprecipitated using a mAb against the FLAG-tag, deglycosylated with PNGase F and separated on 16.5% Tricine-SDS-polyacrylamide gels. Proteins were visualized by autoradiography. (B) The experiment was performed as described in A but with HeLa cells. (C) The experiment was performed as described in A but with different H2O2 concentrations. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.1.5 Glycosylation of FLAG-PSCA-HA protein during proteasome inhibition Now, we investigated the glycosylation pattern of FLAG-PSCA-HA precursor protein during proteasome inhibition. We visualized the glycosylation of FLAG-PSCA-HA over time by using a short-term pulse-chase labeling setup (described above) and anti-FLAG immunoprecipitation and skipped the PnGase F deglycosylation step (Fig 6.11). An enrichment of a fully glycosylated FLAG-PSCA-HA form (18 kDa, upper band) was observable in MG132 treated and untreated cells over time. This demonstrates that at least a part of synthesized FLAG-PSCA-HA protein is transported into the ER and glycosylated there during proteasome inhibition. In contrast, two bands (13 kDa and 16 kDa) are stabilized during MG132 treatment, presenting the precursor protein and/or not fully glycosylated FLAG-PSCA-HA forms.

Fig 6.11│Glycosylation pattern of FLAG-PSCA-HA protein during proteasome inhibition. HEK293T cells transfected with FLAG-PSCA-HA were pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Lysates were immunoprecipitated with a mAb against the HA-tag and separated on 16.5% Tricine-SDS- polyacrylamide gels. The proteins were visualized by autoradiography. The experiment was repeated twice with similar outcome.

6.1.1.6 Cellular localization of FLAG-PSCA-HA protein during proteasome inhibition HEK293T cells transiently expressing FLAG-PSCA-HA protein were starved for one hour and radioactively labeled for 30 minutes. Simultaneously, cells were incubated with MG132 for 90 minutes or left untreated. Cellular compartments were fractionated using osmotic pressure, immediately after labeling (Chapter 5.3.8). Lysates of endoplasmic reticulum- and cytoplasmic fractions were immunoprecipitated with anti-FLAG beads and deglycosylated. After separation on SDS-gels and autoradiographical analysis, a 13 kDa protein band was detectable in the cytoplasmic fraction of HEK293T cells treated with MG132. A faint band was observable in the endoplasmic reticulum fraction of MG132 treated cells (Fig 6.12).

Fig 6.12│Cellular localization of FLAG-PSCA-HA protein during proteasome inhibition. HEK293T cells were transfected with FLAG-PSCA-HA and pulse-labeled with [35S]-methionine/cysteine for 30 minutes. Where indicated, cells were treated with the proteasome inhibitor MG132 during starvation and labeling. Afterwards, cells were fractioned into cellular and ER-resident proteins. Then, samples were immunoprecipitated with a mAb against the FLAG-tag, deglycosylated with PNGase F and separated on 16.5% Tricine-SDS-polyacrylamide gels. The proteins were visualized by autoradiography. ctr = untransfected cells. The experiment was repeated twice with similar outcome. 69

6.1.1.7 Stabilization of PSCA-HA precursor protein during proteasome inhibition We repeated the short-term pulse-chase labeling experiments in cells transiently overexpressing PSCA-HA protein (no FLAG-tag in front of the ER leader sequence). The separation of PSCA into premature and mature protein was not detectable after anti-HA immunoprecipitation of the radioactively labeled proteins. A single band band with the size of 11 kDa (correlating with the size of the matured protein) was detected (Fig 6.13A). It is possible that the N-terminal FLAG-tag stabilizes PSCA precursor protein, e.g. because of interactions with the SRP that slow down the processing. We repeated PSCA-HA pulse-chase labeling experiments at room temperature, instead of incubating cells at 37°C, but this did not change the experimental result of the experiment (Fig 6.13B).

Fig 6.13│Effect of proteasome inhibition on PSCA-HA precursor protein stability. (A) HEK293T cells transfected with PSCA-HA were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes, and chased for the indicated time periods. Cells were treated for 30 minutes of starvation and during the labeling and chase period with the proteasome inhibitor lactacystin or left untreated. Then, lysates were immunoprecipitated with a mAb against the HA-tag and separated on 16.5% Tricine-SDS-polyacrylamide gels. The proteins were visualized by autoradiography. (B) Experiments were performed as described in A but samples were kept at room temperature instead of 37°C during the experiment. The experiments were repeated twice with similar outcome.

6.1.2 Prolactin precursor stability during proteasome inhibition Prolactin is a peptide hormone secreted from the pituitary gland with a large number of functions. It is best known for its function to enable lactation in female mammals (Bole-Feysot et al. 1998). We used the double-tag strategy described before for PSCA with human Prolactin cDNA (Fig 6.14A). Anti-FLAG Western blot analysis of cells overexpressing FLAG-Prolactin-HA showed a band detectable at the expected size of 26 kDa (Fig 6.14B). FLAG-Prolactin-HA overexpression construct was used in short-term pulse-chase labeling experiments with anti- FLAG immunoprecipitation. A slight stabilization of FLAG-Prolactin-HA precursor protein was detected during MG132 treatment of cells (Fig 6.14C). Immunoprecipitation with antibodies against the HA-tag resulted in two bands with the size of 28 kDa (precursor) and 26 kDa (mature protein). There was no detectable difference in the stability of FLAG-Prolactin-HA precursor over time during proteasome inhibition (Fig 6.14D). Subsequently, we removed the FLAG-tag and overexpressed Prolactin-HA in HEK293T cells. After radioactive labeling and anti-HA immunoprecipitation, the upper band disappeared and only the putative processed Prolactin-HA protein was observable, as seen for PSCA-HA (Chapter 6.1.1.7).

Fig 6.14│ Effect of proteasome inhibition on Prolactin precursor protein stability. (A) Scheme of N-terminally FLAG- tagged and C-terminally HA-tagged Prolactin precursor protein. (B) HEK293T cells transfected with FLAG-Prolactin- HA were lysed and used for Western blot analysis with a mAb against the FLAG-tag. Tubulin was used as a loading control. (C) HEK293T cells transfected with FLAG-Prolactin-HA were starved for one hour, pulse-labeled with [35S]- methionine/cysteine for 5 minutes and chased for the indicated time periods. Where indicated, cells were treated with proteasome inhibitor lactacystin for the last 30 minutes of starvation, during labeling and chase. Lysates were used for immunoprecipitation with a mAb against the FLAG-tag and deglycosylated with PNGase F. Afterwards, the proteins were separated on 16.5% Tricine-SDS-polyacrylamide gels and the proteins were visualized by autoradiography. (D) The experiments were performed as described in C but with a mAb against the HA-tag used for immunoprecipitation. (E) The experiments were performed as described in C but with HEK293T cells transfected with Prolactin-HA. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.3 Proteasome inhibition stabilizes Leptin precursor protein

6.1.3.1. FLAG-Leptin-HA precursor protein is stabilized during proteasome inhibition and oxidative stress Leptin is a peptide hormone that regulates the amount of fat in the body (Brennan & Mantzoros 2006). Because of the inexplicit results using FLAG-Prolactin-HA, we decided to N-terminally FLAG- and C-terminally HA-tag human Leptin cDNA (Fig 6.15A). We used the same

experimental conditions as described for FLAG-PSCA-HA before with anti-FLAG immunoprecipitation. FLAG-Leptin-HA precursor protein (21 kDa band) was stabilized for up to 60 minutes when we inhibited the proteasome with MG132 (Fig 6.15B). In control cells, the precursor protein band disappeared after 10 minutes of chase. We immunoprecipitated overexpressed FLAG-Leptin-HA with antibodies against the HA-tag in untreated cells after pulse-chase labeling and detected the mature protein with a band size of 18 kDa at every given time point. The 21 kDa precursor band was detectable for up to 10 minutes after labeling. Treatment of cells with MG132 or lactacystin stabilized the precursor protein for up to 60 minutes

(Fig 6.15C). Next, we treated cells with H2O2 to induce oxidative stress. Leptin precursor protein was stabilized for up to 60 minutes in cells treated with H2O2 (Fig 6.15D). Thereafter, we monitored the cellular localization and translocation of FLAG-Leptin-HA and Leptin-HA into the ER. Co-localization with Climp63 was detected by confocal fluorescence microscopy. There was no difference between the cellular localization of FLAG-Leptin-HA and Leptin-HA detectable (Fig 15E).

Fig 6.15│ Proteasome inhibition stabilizes FLAG-Leptin-HA precursor protein. (A) Scheme of N-terminally FLAG- tagged and C-terminally HA-tagged Leptin precursor protein. (B) HEK293T cells transfected with FLAG-Leptin-HA were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with proteasome inhibitors lactacystin and MG132 for the last 30 minutes of starvation, during labeling and chase, as indicated. Lysates were immunoprecipitated with a mAb against the FLAG- tag followed by deglycosylation with PNGase F. Then, proteins were separated on 16.5% Tricine-SDS-polyacrylamide gels and visualized by autoradiography. (C) The same experimental setup was performed as in B, but with a mAb against the HA-tag used for immunoprecipitation. (D) Experiments were performed as described in B but with H2O2 treatment. Bold numbers show the percentage of lane intensity compared to given 100% band. (E) Comparison of Leptin localization pattern in HEK293T cells expressing Leptin-HA or FLAG-Leptin-HA. Cells were stained with polyclonal antibodies against the HA-tag (red) and Climp63 (green) and visualized by confocal fluorescence microscopy; bar: 10 μm. The experiments were repeated twice with similar outcome. 74

6.1.3.2 Stabilization of Leptin-HA protein during proteasome inhibition We transfected HEK293T cells with N-terminally untagged Leptin-HA and performed short-term pulse-chase label experiments as described before. After immunoprecipitation with antibodies against the HA-tag, deglycosylation and separation of the proteins, there was a single band with the size of 20 kDa. No separated bands, which could be assigned to precursor or mature protein, were detectable (Fig 6.16A). Next, we used an antibody against Leptin for immunoprecipitation in the same experimental setup and failed to detect two distinguishable bands, too (Fig 6.16B).

Fig 6.16│ Effect of proteasome inhibition on Leptin-HA precursor protein. (A) HEK293T cells transfected with Leptin- HA were starved for one hour, pulse-labeled for 5 minutes with [35S]-methionine/cysteine, and chased for the indicated time periods. Cells were treated for 30 minutes of starvation and during the labeling and chase period with the proteasome inhibitor lactacystin or left untreated. Next, lysates were immunoprecipitated with a mAb against the HA- tag and separated on 16.5% Tricine-SDS-polyacrylamide gels. The Proteins were visualized by autoradiography. (B) Experiments were performed as described in A but with a polyclonal antibody against Leptin used for immunoprecipitation. The experiments were repeated twice with similar outcome.

6.1.4 Investigation of the mechanisms of cytosolic ER precursor protein accumulation during proteasome inhibition or oxidative stress

6.1.4.1. The role of ubiquitinated FLAG-PSCA-HA ER signal peptide in precursor protein stabilization during proteasome inhibition The enrichment of ubiquitinated FLAG-PSCA-HA ER leader proteins in the cytoplasm due to the impairment of proteasome activity could be a signal for the induction of its accumulation. We replaced the single lysine residue in the PSCA ER leader sequence with an arginine, which can not be ubiquitinated, and performed short-term pulse-chase label experiments as described in 6.1.1.1. FLAG-PSCA-HA-K2A was stabilized during proteasome inhibition with MG132 even

though the lysine in the ER leader peptide was missing and no accumulation of its ubiquitinated leader peptide was possible (Fig 6.17).

Fig 6.17│Ubiquitination of FLAG-PSCA-HA ER leader peptide and PSCA precursor protein stabilization during proteasome inhibition. HEK293T cells transfected with FLAG-PSCA-K2A-HA were starved for one hour, pulse-labeled for 5 minutes with [35S]-methionine/cysteine, and chased for the indicated time periods. Cells were treated for 30 minutes of starvation and during the labeling and chase period with the proteasome inhibitor MG132 or left untreated. Then, lysates were immunoprecipitated with a mAb against the FLAG-tag and separated on 16.5% Tricine-SDS- polyacrylamide gels. The proteins were visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.4.2 Stabilization of FLAG-PSCA-HA precursor protein during different stress conditions We induced various stress conditions in HEK293T cells and followed the FLAG-PSCA-HA precursor protein stability over time. We used short-term pulse-chase label experiments and immunoprecipitated with antibodies against the FLAG-tag as described in 6.1.1.1. In untreated cells, FLAG-PSCA-HA precursor protein was cleaved after 10 minutes of chase (Fig 6.18A). 3- Methyladenine is an autophagy inhibitor that blocks the autophagosome formation via the inhibition of type III Phosphatidylinositol 3-kinases (Wu et al. 2010). HEK293T cells were treated with 10 mM 3-Methyladenine for 2 hours. FLAG-PSCA-HA precursor protein was not stabilized (Fig 6.18B). In a contrary approach, we used rapamycin (100 µM, for 2 hours) to induce autophagy. Rapamycin-induced inhibition of mTOR mimics cellular starvation by blocking signals required for cell growth and proliferation (Jung et al. 2010). Cells treated with 100 µM rapamycin for two hours could not be labeled with [35S]-methionine/cysteine (data not shown). L-canavanine is an L-amino acid, naturally occurring in in leguminous plants. L-canavanine competes against L-arginine in enzyme binding because of its structural similarity, which leads to a high amount misfolded proteins in the cell. It was almost impossible to treat cells with L-canavanine (5 µg/µl,

for 4 hours) and maintain their ability to produce [35S]-labeled proteins. No stabilization of FLAG- PSCA-HA precursor protein was detectable in L-canavanine treated cells (Fig 6.18C). Dithiothreitol (DTT) is a small, cell permeable molecule, which is used as a reducing agent that reduces disulfide bonds, producing a high amount of misfolded proteins in cells (Klonne & Johnson 1988). No FLAG-PSCA-HA precursor protein stabilization was observable in cells treated with 3 mM DTT for 12 hours (Fig 6.18D). Eeyarestatin I inhibits ERAD, specifically targeting the VCP/p97-associated deubiquitinating process and inhibiting ataxin-3-dependent deubiquitination. Impairment of the ERAD pathway by treatment of cells with 10 µM Eeyarestatin I for 2 hours had no detectable effect on the stability of FLAG-PSCA-HA precursor protein (Fig 6.18E). Puromycin generates chain termination of premature amino acid chains during their translation at the ribosome. The 3' end of an aminoacylated tRNA is resembled by a part of the molecule. It causes the formation of a puromycylated nascent chain and leads to premature chain release (Pestka 1971). FLAG-PSCA-HA precursor protein was not stabilized in cells treated 4 hours with puromycin (1 µg/µl; Fig 6.18F). SIN-1 chloride treatment, leads to production of NO and superoxide in the cytoplasm of cells, therefore generating peroxynitrite (Rosenkranz et al. 1996). We treated HEK293T cells for 30 minutes with SIN-1 chloride (1 mM) and short-term pulse-chase labeled them. FLAG-PSCA-HA precursor protein exhibited the same time-dependent band pattern as visible in control cells (Fig 6.18G). Spermine NONO-ate is a radical generator that is used to generate nitric oxide in aqueous solutions (Maragos et al. 1991). HEK293T cells treated with 1mM of this nitrogen stress inducer for 30 minutes showed no aberration in the stability of FLAG-PSCA-HA precursor protein (Fig 6.18H). D,L-sulforaphane reacts with the cysteine-rich intervening region of Keap1 and disrupts the Nrf2/Keap1 complex, freeing Nrf2. This leads to translocation of Nrf2 into the nucleus and activation of genes of the antioxidant pathway, which is the primary cellular defense against cytotoxic effects triggered by oxidative stress (Gold et al. 2012). Neither HEK293T cells treated with 10 µM D, L-sulforaphane for 18 hours (data not shown) nor cells treated with 150 µM D, L-sulforaphane for 30 minutes (Fig 6.18I) showed stabilization of FLAG-PSCA-HA precursor protein. Thapsigargin is an inhibitor of ubiquitous sarco-endoplasmic reticulum Ca2+-ATPases in mammalian cells. Disruption of cellular Ca2+ homeostasis impairs ER function (Treiman et al. 1998). A second, ER stress-inducing agent is Tunicamycin which inhibits the N-glycosylation of ER-guided proteins (Bull & Thiede 2012). 6 hours treatment of HEK293T cells with either 5 µM thapsigargin or 1 µg/ml tunicamycin revealed no aberration in FLAG-PSCA-HA precursor protein stability (Fig 6.18J,K). N-methoxysuccinyl-Ala-Ala-Pro-Val-chloromethyl ketone is published as an inhibitor the ER signal peptidase, if used on microsomal membrane proteins (Lemberg et al. 2001). We incubated HEK293T with N-methoxysuccinyl-Ala-Ala-Pro-Val-chloromethyl ketone (250 µM) for 77

1 hour, prior to pulse-chase labeling experiments. There was no effect on the FLAG-PSCA-HA precursor stabilization detectable (Fig 6.18L).

Fig 6.18│Stability of FLAG-PSCA-HA precursor protein during different stress conditions (A-L) HEK293T cells transfected with FLAG-PSCA-HA were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with the indicated chemicals for the given periods of time. The lysates were immunoprecipitated with a mAb against the FLAG-tag and deglycosylated with PNGase F and separated on 16.5% Tricine-SDS-polyacrylamide gels. The proteins were visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

(-)-Epigallocatechin gallate (EGCG) is reported to inhibit the molecular chaperone glucose regulated protein 78, known as BiP or GRP78 (Ermakova et al. 2006). Cells were treated with 10 µM EGCG 24 hours before the pulse-chase labeling experiments. During proteasome inhibition, FLAG-PSCA-HA precursor protein was stabilized in EGCG treated cells in the same range as detected in untreated cells (Fig 6.19A-D).

Fig 6.19│Stability of FLAG-PSCA-HA precursor protein during Grp78 inhibition. (A-D) HEK293T cells expressing FLAG-PSCA-HA were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated for the last 30 minutes of starvation, during labeling and chase with the proteasome inhibitor MG132. Cells were treated with epigallocatechin gallate (EGCG) for 24 hours before the pulse- labeling or left untreated. Cell lysates were immunoprecipitated with a mAb against the FLAG-tag and deglycosylated with PNGase F. Then, proteins were separated on 16.5% Tricine-SDS-polyacrylamide gels and visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.4.3 Ubiquitin and stabilization of FLAG-PSCA-HA precursor protein Stabilization of FLAG-PSCA-HA precursor protein was detectable 40 minutes after proteasome inhibition. We wanted to investigate the effect of short-term proteasome inhibition (and oxidative stress induction) on the amount of polyubiquitinated protein conjugates in HEK293T and HeLa cells. Cells were treated with MG132 or lactacystin for 40 minutes. After lysis, the proteins were separated on 10% SDS-polyacrylamide gels, transferred onto nitrocellulose membranes and probed with anti-ubiquitin antibody. An enrichment of high molecular weight, polyubiquitinated conjugates was detectable with both proteasome inhibitors (Fig 6.20A). Treatment of cells with

350 µM or 750 µM H2O2 for 40 minutes resulted in less accumulation of polyubiquitinated proteins (Fig 6.20B). Next, we overexpressed HA-tagged ubiquitin together with FLAG-PSCA- HA construct for 36 hours and performed short-term pulse-chase label experiments. There was

no difference in FLAG-PSCA-precursor protein stability detectable in HEK293T cells overexpressing mono-ubiquitin (Fig 6.20C).

Fig 6.20│Short-term MG132- and H2O2 treatment induces accumulation of intracellular poly-ubiquitin conjugates. Ubiquitin overexpression does not lead to FLAG-PSCA-HA precursor protein stabilization. (A) HEK293T or HeLa cells were treated with proteasome inhibitors lactacystin and MG132 as indicated. Cell lysates were analyzed on Western blots using a mAb against mono- and polyubiquitinated proteins. Tubulin was used as a loading control. (B) Cells were treated as described in A but with H2O2 instead of MG132. (C) HEK293T cells transfected with FLAG-PSCA-HA were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes, and chased for the indicated time periods. 36 hours prior to pulse-labeling, cells were transfected with ubiquitin-HA or left untreated. Next, lysates were immunoprecipitated with a mAb against the FLAG-tag and separated on 16.5% Tricine-SDS-polyacrylamide gels. The proteins were visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.4.4 Stability of FLAG-PSCA-HA precursor protein during VCP/p97 inhibition, heat shock and Nrf1 or Nrf2 overexpression The VCP/p97-associated deubiquinating process is an important step in endoplasmic reticulum- associated protein degradation (ERAD). The dominant negative variant of VCP/p97 is an E578Q mutant. The mutation abolishes ATP hydrolysis and leads to trapping of ubiquitin conjugates along with substrate proteins (Pye et al. 2007). Wild type VCP/p97 or VCP/p97-E578Q were transiently expressed in HEK293T cells 48 hours prior to short-term pulse-chase experiments. Cells overexpressing dominant negative VCP/p97 showed no difference in FLAG-PSCA-HA precursor stability as compared to wild type VCP/p97 expressing cells (Fig 6.21A). Then, cells were heat shocked during starving and labeling, incubating them at 42°C instead of 37°C. We detected no stabilization of FLAG-PSCA-HA precursor protein in heat shocked cells (Fig 6.21B).

Fig 6.21│Overexpression of dominant negative VCP and heat shock do not stabilize FLAG-PSCA-HA precursor protein. (A) HEK293T cells transfected with FLAG-PSCA-HA were starved for one hour, pulse-labeled with [35S]- methionine/cysteine for 5 minutes, and chased for the indicated time periods. Cells were transfected with hVCP or hVCP-E578Q 36 hours prior to pulse-labeling experiments. Lysates were immunoprecipitated with a mAb against the FLAG-tag and separated on 16.5% Tricine-SDS-polyacrylamide gels. The proteins were visualized by autoradiography. (B) HEK293T cells transfected with FLAG-PSCA-HA were pulse-labeled with [35S]- methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were incubated at 42°C (heat shock) or 37°C (control) during starvation, labeling and chase periods. Then, lysates were immunoprecipitated with a mAb against the FLAG-tag and separated on 16.5% Tricine-SDS-polyacrylamide gels. The proteins were visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

We overexpressed the transcription factors Nrf1 and Nrf2 in HEK293T cells to induce expression of antioxidant genes (Biswas & Chan 2010; Gold et al. 2012). Transient transfection of cells and subsequent Western blot analysis revealed appropriate Nrf1 and Nrf2 expression (Fig 6.22A). Nrf1 or Nrf2 were overexpressed in HEK293T cells together with FLAG-PSCA-HA. Pulse-chase label experiments and following immunoprecipitation against the Flag-tag revealed no difference

of FLAG-PSCA-HA precursor protein stability as compared to cells not overexpressing Nrf1 or Nrf2 (Fig 6.22B).

Fig 6.22│Overexpression of Nrf1 or Nrf2 transcription factor does not stabilize FLAG-PSCA-HA precursor protein. (A) HEK293T cells were transfected with Nrf1-FLAG or Nrf2-myc and analyzed on Western blots probed with a mAb against FLAG- or myc-tag. Tubulin was used as a loading control. (B) HEK293T cells transfected with FLAG-PSCA- HA were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes, and chased for the indicated time periods. 48 hours prior to the pulse-labeling, cells were transfected with Nrf1-FLAG or Nrf2-myc or left untreated. Lysates were immunoprecipitated with a mAb against the FLAG-tag and separated on 16.5% Tricine-SDS- polyacrylamide gels. The proteins were visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. The experiments were repeated twice with similar outcome.

6.1.4.5 Identification of ER signal peptidase-associated regulator proteins One possible explanation for FLAG-PSCA-HA precursor protein accumulation during proteasome inhibition would be a negative feedback mechanism, inhibiting the expression or activity of the ER signal peptidase. We determined the amount of SPC18 protein in cells treated with MG132 in Western blots probed with antibodies against SPC18. The expression of SPC18 during proteasome impairment was unchanged (Fig 6.23A). Unfortunately, no anti-SPC21 antibodies are commercially available to measure the SPC21 protein amount during proteasome inhibition. To identify proteins that interact with the ER signal peptidase during proteasome inhibition and not during physiological conditions, we overexpressed N-terminally HA-tagged ER signal peptidase subunits SPC18, SPC21 and SPC22/23 for 48 hours. After 40 minutes of treatment with MG132, cells were radioactively pulse-labeled, lysed and immunoprecipitation with antibodies against the HA-tag was performed in treated and control cells. 2D SDS gel analysis revealed changes in the abundance of seven proteins (Fig 6.23B, marked with arrows). 82

Fig 6.23│ Effect of proteasome inhibition on SPC18 and SPC22/23 ER signal peptidase subunits. (A) HEK293T or HeLa cells were treated with the proteasome inhibitor MG132 as indicated. SPC18 expression was analyzed with Western blots using a polyclonal antibody against SPC18. Tubulin was used as a loading control. (B) HEK293T cells were transfected with SPC22/23-HA and treated with MG132 for 40 minutes. The lysates were separated on 2D SDS- polyacrylamide gel, pH range 3-10 in the first dimension and 15% SDS-PAGE in the second dimension. Gels were silver stained; showing seven protein spots exclusively in MG132 treated samples (labeled with arrows). The experiments were repeated twice with similar outcome.

The proteins we co-immunoprecipitated during MG132 treatment were excised and analyzed with mass spectrometry (Table 6.1).

Score Protein 839,54 Malate dehydrogenase, mitochondrial 823,62 Proliferating cell nuclear antigen 810,68 40S ribosomal protein 657,38 Triosephosphate isomerase 545,99 Nucleophosmin 435,51 3-hydroxyacyl-CoA dehydrogenase type 2 429,02 Elongation factor 1-beta 369,61 Peroxiredoxin-6 358,02 GTP-binding nuclear protein Ran 305,85 14-3-3 protein epsilon 273,78 Heterogeneous nuclear ribonucleoproteins C1/C2 268,02 Hypoxanthine-guanine phosphoribosyltransferase 239,20 Heterogeneous nuclear ribonucleoproteins A2/B1 226,95 Poly(ADP-ribose) glycohydrolase ARH3 225,70 UPF0368 protein Cxorf26 Complement component 1 Q subcomponent-binding 168,65 protein, mitochondrial 159,92 L-lactate dehydrogenase B chain 147,14 Glyceraldehyde-3-phosphate dehydrogenase 126,55 Carbonic anhydrase 2 120,42 Tropomyosin alpha-4 chain 116,32 Annexin A5

Table 6.1│ Analysis of proteins bound to SPC22/23 only during MG132 treatment.

There were no differences in protein abundance of SPC18 and SPC21 detectable during proteasome inhibition. Next, we analyzed the ubiquitination state of ER signal peptidase subunits after MG132 treatment. We overexpressed the HA-tagged signal peptidase complex subunits and treated HEK293T cells for one hour with MG132, lysed them and immunoprecipitated the samples with anti-HA beads. There was no ubiquitination of SPC18, SPC21 or SPC22/23 detectable in anti-ubiquitin Western blots (data not shown).

6.1.4.6 Phosphorylation of ER signal peptidase subunits during proteasome inhibition Phosphorylation of proteins at serine, threonine or tyrosine residues is a fast mechanism to alter their activity (Barford et al. 1998). This modification would be an efficient way to impair the function of the ER signal peptidase in situations of unfolded protein stress, e.g. because of proteasome impairment. We overexpressed HA-tagged SPC18, SPC21 and SPC22/23 in

HEK293T cells for 48 hours. Cells were starved for 2 hours and labeled in the presence or absence of MG132 for 1 hour. After lysis, immunoprecipitation against phosphorylated tyrosines, 15% SDS gel separation and gel drying, the proteins were visualized via autoradiography. There were no tyrosine-phosphorylated SPC18, SPC21 or SPC22/23 proteins precipitated. We immunoprecipitated with antibodies against the HA-tag to control the sufficient expression of SPC18 (18 kDa), SPC21 (21 kDa) and SPC22/23 (23 kDa; Fig 6.24).

Fig 6.24| Proteasome inhibition does not lead to tyrosine phosphorylation of SPC18, SPC21 or SPC22/23 ER signal peptidase subunits. HEK293T cells were transfected with SPC18-HA, SPC21-HA and SPC22/23-HA 48 hours before the pulse-labeling or left untreated. Then, cells were starved for two hours and labeled with [35S]-methionine/cysteine in the presence or absence of the proteasome inhibitor MG132 for one hour. Lysates were immunoprecipitated with a polyclonal antibody against phospho-tyrosines or a monoclonal antibody against the HA-tag (control) and separated on 15% SDS-polyacrylamide gels. Proteins were visualized by autoradiography. The experiment was repeated twice with similar outcome.

6.1.4.7 Co-localization of ER signal peptidase subunits and Derlin-1 or Synoviolin1 The induction of ERAD could be a signal to down regulate the activity of the ER signal peptidase. A co-localization of ER signal peptidase subunits and proteins of the ERAD machinery would point to this hypothesis. Derlin-1 (Der-1) is an ER membrane protein and it functions in ERAD as a detector of misfolded proteins and targets them to degradation (Schaheen et al. 2009). Synoviolin1 (Syvn1) is an E3 ligase, that retrotranslocates accumulated, unfolded proteins from the ER lumen into the cytosol during ER stress (Amano et al. 2003). We used C-terminally myc-tagged Der-1 and Syvn-1 overexpression constructs and monitored their expression in Western blots probed with antibodies against the myc-tag. Derlin-1 and Syvn-1 were expressed with expected molecular mass of 29 kDa and 68 kDa, respectively (Fig 6.25A). We focused on Der-1 for further experiments, because repetitions of Syvn-1 expression analysis revealed an unsteady expression profile of the protein. Either, Der-1 (myc-tagged), SPC18-HA, SPC21-HA or SPC22/23-HA alone or Der-1 with each of the ER signal peptidase subunits together was overexpressed in HEK293T cells for 48 hours. Cells were radioactively labeled for 1 hour, lysed and immunoprecipitated with antibodies against the myc-tag overnight. After 85

washing, the samples were loaded on 15% SDS gels and the gels were dried. Proteins were visualized by autoradiography. Der-1 was not co-immunoprecipitated with any of the ER signal peptidase subunits SPC18, SCP21 or SPC22/23 (Fig 6.25B).

Fig 6.25│ Derlin-1 does not co-immunoprecipitate with SPC18, SPC21 or SPC22/23 ER signal peptidase subunits. (A) HEK293T cells were transfected with Derlin-1-myc (Der-1-myc) or Synoviolin1-myc (Syvn1-myc) and analyzed on western blots with a mAb against the myc-tag. Tubulin was used as a loading control. (B) HEK293T cells were transfected with Der-1-myc, SPC18-HA, SPC21-HA and/or SPC22/23-HA 48 hours before the pulse-labeling. Then, cells were starved for one hour, labeled with [35S]-methionine/cysteine for one hour, lysed and immunoprecipitated with a mAb against the myc-tag or the HA-tag. Proteins were separated on 15% SDS-polyacrylamide gels and visualized by autoradiography. The experiments were repeated twice with similar outcome.

6.1.4.8 Bag6 siRNA knock down and Flag-PSCA-HA precursor processing Bag6 has been shown to be an important factor for the degradation of mislocalized PrP protein (Hessa et al. 2011). Bag6 siRNA knockdown was performed two times every 48 hours, prior to the pulse-labeling experiments. Bag6 mRNA amount was reduced to less than 15% in comparison to control siRNA treated cells. We monitored the stability of FLAG-PSCA-HA

precursor protein in pulse-chase experiments with anti-FLAG immunoprecipitation as described in 6.1.1.1. There was no stabilization of FLAG-PSCA-HA precursor protein observable (Fig 6.26).

Fig 6.26│Bag6 siRNA knock down does not stabilize FLAG-PSCA-HA precursor protein. HEK293T cells were treated two times with Bag6-specific- or control siRNA pools (48 hours interval) and transfected with FLAG-PSCA-HA. Cells were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated periods of time. Lysates were immunoprecipitated with a mAb against the FLAG-tag and deglycosylated with PNGase F. Then, proteins were separated on 16.5% Tricine-SDS-polyacrylamide gels and visualized by autoradiography. Bold numbers show the percentage of lane intensity compared to given 100% band. This experiment was performed once.

6.1.5 Stabilization of endogenously expressed ER-targeted precursor proteins during proteasome inhibition.

6.1.5.1 α1-Antitrypsin (AAT)

α1-Antitrypsin is a secretory protein that functions as a protease inhibitor and belongs to the serpin superfamily (Gettins 2002). AAT is endogenously expressed in the human liver carcinoma cell line Hep G2. We used this cell line in short-term pulse-chase label experiments. Cells were starved in methionine/cysteine free medium for one hour, labeled with [35S]-methionine/cysteine for 5 minutes and chased for up to 60 minutes. After lysis, we performed anti-AAT immunoprecipitation and deglycosylated the samples overnight. Subsequently, proteins were separated on 12% SDS-polyacrylamide gels and visualized by autoradiography. Cells were either treated with MG132 for the last 30 minutes of starvation and during the chase times or left untreated. Treated and control samples showed one single AAT1 band with 41 kDa in size. No precursor band was detectable (Fig 6.27).

Fig 6.27│ Effect of proteasome inhibition on endogenous α1-Antitrypsin (AAT1) precursor protein. HepG2 cells were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with the proteasome inhibitor MG132 for the last 30 minutes of starvation, during labeling and chase or left untreated. Lysates were immunoprecipitated with a polyclonal antibody against AAT and deglycosylated with PNGase F. Then, proteins were separated on 15% SDS-polyacrylamide gels and the proteins were visualized by autoradiography. The experiment was repeated twice with similar outcome.

6.1.5.2 β2-microglobulin (B2M) β2-microglobulin is a component of the class I major histocompatibility complex (MHC), which is involved in the peptide antigen presentation of the immune system (Güssow et al. 1987). We used HEK293T cells and induced B2M expression with 16 hours of IFNγ treatment. The experimental conditions are the same as mentioned in 6.1.5.1 with using Tricine-SDS- polyacrylamide gels instead of standard Tris-SDS-polyacrylamide gels for separation because of the small size of B2M protein (10 kDa). Proteins were immunoprecipitated with an anti-B2M antibody. The autoradiographic visualization showed a single band with the size of 10 kDa. There was no difference between MG132 treated and control cells and no double-band detectable (Fig 6.28).

Fig 6.28│ β2-microglobulin (B2M) precursor protein stabilization during proteasome inhibition. HEK293T cells were stimulated with IFNγ for 16 hours, starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with proteasome inhibitor MG132 for the last 30 minutes of starvation, during labeling and chase or left untreated. Lysates were immunoprecipitated with a mAb against B2M and deglycosylated with PNGase F. Then, proteins were separated on 16.5% Tricine-SDS-polyacrylamide gels and visualized by autoradiography. The experiment was repeated twice with similar outcome.

6.1.5.3 Envelope glycoprotein of Lymphocytic Choriomeningitis Virus (LCMV GP) LCMV GP is initially expressed as a precursor polypeptide, called GP-C. GP-C is post- translationally processed into two proteins, GP-1 and GP-2. GP-1 interacts with an LCMV receptor on the cell surface. The fusion peptide and the transmembrane domain of the virus are enclosed in GP-2 (Kunz et al. 2003). We used MC57 fibrosarcoma cell line in these experiments. Cells were infected with LCMV for 16 hours with an MOI of 1. After pulse-chase labeling experiments (as described in 6.1.5.1.) and immunoprecipitation with antibodies against LCMV GP, samples were deglycosylated and 10% SDS-polyacrylamide gels were used to separate proteins. We detected the 50 kDa precursor protein and the 44 kDa mature protein. There is no stabilization of the precursor protein detectable in MG132 treated cells as compared to untreated cells (Fig 6.29A). In a next approach, we overexpressed LCMV GP in HEK293T cells for 48 hours and performed short-term pulse-chase labeling as described before. We detected two bands at the right size (44 kDa and 50 kDa) and one additional background band (upper band). Unfortunately, we were not able to separate the band of precursor and mature protein in an appropriate way. After 120 minutes, cleavage of GP-C to GP-1 and GP-2 could be monitored and both bands showed appropriate sizes of 20 kDa and 24 kDa (Fig 6.29B).

Fig 6.29│ LCMV GP precursor protein stabilization during proteasome inhibition. (A) MC57 cells were infected with LCMV (MOI = 1) overnight. Next, cells were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with the proteasome inhibitor MG132 for the last 30 minutes of starvation, during labeling and chase or left untreated. Lysates were immunoprecipitated with a mAb against LCMV-GP and deglycosylated with PNGase F. Then, proteins were separated on 15% SDS-polyacrylamide gels visualized by autoradiography. (B) Experiments were performed as described in A, but without LCMV infection. HEK293T cells were transfected with LCMV GP 48 hours before pulse-labeling. The experiments were repeated twice with similar outcome.

6.1.5.4 Carbonic anhydrase 4 (CA4) Carbonic anhydrases catalyze the reversible hydration of carbon dioxide and belong to a large family of zinc-metallo enzymes (Okuyama et al. 1993). We screened HeLa (cervix carcinoma), HUVEC (endothelium of veins from the umbilical cord), ACHN (renal adenocarcinoma) and SW620 (colorectal adenocarcinoma) cells for endogenous CA4 protein expression in anti-CA4 Western blots. None of the cell lines showed endogenous CA4 expression (data not shown). We overexpressed untagged CA4 in COS-7 cells for 24 hours and performed short-term pulse- chase label experiments with adjacent anti-CA4 immunoprecipitation, deglycosylation and 15% SDS-polyacrylamide gel separation. MG132 treatment of cells for 30 minutes of starvation and for the full chase time period leads to a stabilization of the CA4 precursor protein (31 kDa) for up to 60 minutes. The mature protein was visible at every time point with the right band size of 29 kDa (Fig 6.30).

Fig 6.30│Inhibition of the proteasome stabilizes carbonic anhydrase 4 (CA4) precursor protein. HEK293T were transfected with CA4, starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with proteasome inhibitor MG132 for the last 30 minutes of starvation, during labeling and chase or left untreated. Lysates were immunoprecipitated with a polyclonal antibody against CA4 and deglycosylated with PNGase F. Then, proteins were separated on 15% SDS-polyacrylamide gels and visualized by autoradiography. The experiment was repeated twice with similar outcome.

6.1.5.5 Human cytomegalovirus gene product US11 US11 is a glycoprotein of human cytomegalovirus (HCMV) that targets MHC class I molecules for destruction in a proteasome-dependent manner (Rehm et al. 2001). We overexpressed US11 in HEK293T cells for 24 hours and performed short-term pulse-chase labeling experiments. After lysis, immunoprecipitation with antibodies against US11, deglycosylation and 15% SDS- polyacrylamide gel electrophoresis, autoradiographic analysis revealed a single band of about 23 kDa. No precursor protein was detected (Fig 6.31).

Fig 6.31│Human cytomegalovirus gene product US11 precursor protein stabilization during proteasome inhibition. HEK293T cells transfected with US11 were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with the proteasome inhibitor MG132 for the last 30 minutes of starvation, during labeling and chase or left untreated. Lysates were immunoprecipitated with a polyclonal antibody against US11 and deglycosylated with PNGase F. Finally, proteins were separated on 15% SDS- polyacrylamide gels and visualized by autoradiography. The experiment was repeated twice with similar outcome.

6.1.5.6 C-C chemokine receptor type 7 (CCR7) CCR7 has two ligands that are identified: the chemokine (C-C motif) ligand 19 (CCL19/ELC) and (C-C motif) ligand 21 (CCL21; Birkenbach et al. 1993). We overexpressed C-terminally HA- tagged CCR7 in HEK293T cells for 24 hours. Cells were short-term pulse-chase labeled with [35S]-methionine/cysteine and proteins were separated on 15% SDS-polyacrylamide gels after lysis. We detected three bands with the sizes of 39 kDa, 41 kDa and 43 kDa. They represent CCR7 (39 kDa) and its glycosylated isoforms. No difference could be detected between MG132 treated (for 30 minutes starvation time as well as for the following chase time) and control cells. We were not able to analyze deglycosylated CCR7 because there were no bands detectable after the deglycosylation process with PNGase F (Fig 6.32).

Fig 6.32│C-C chemokine receptor type 7 (CCR7) precursor protein stabilization during proteasome inhibition. HEK293T cells transfected with CCR7-HA were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with proteasome inhibitor MG132 for the last 30 minutes of starvation, during labeling and chase or left untreated. Lysates were immunoprecipitated with a mAb against the HA-tag and separated on 15% SDS-polyacrylamide gels. The proteins were visualized by autoradiography. The experiment was repeated twice with similar outcome.

6.1.5.7 Murine MHC class I molecules H-2Dd, H-2Db, H-2Kd and H-2Ld Major histocompatibility complexes (MHC) can be divided into two groups, MHC class I and class II. Mouse MHC antigen is also known as H-2 antigen. H-2Dd, H-2Db, H-2Kd and H-2Ld belong to the subgroup of “classical MHC class I”. We used B8-Db fibroblast cells and induced MHC class I upregulation by addition of IFNγ for 72 hours. After short-term pulse-chase labeling, cells were immunoprecipitated with antibodies against H-2Dd, H-2Db, H-2Kd or H-2Ld. Next, samples were deglycosylated, separated on 15% SDS-polyacrylamide gels and autoradiographically visualized. In H-2Db, H-2Kd and H-2Ld samples a single band with the size of the corresponding H-2Db, H-2Kd or H-2Ld mature protein was detectable (Fig 6.33A,B,D). The precursor protein and the mature protein were detectable in H-2Dd immunoprecipitated samples. The precursor band was stabilized for up to 60 minutes in MG132 treated cells. The precursor protein disappeared after 10 minutes of chase in untreated cells (Fig 6.33C).

Fig 6.33│Stabilization of MHC class I haplotypes H-2Db, H-2Dd, H-2Kd and H-2Ld precursor proteins during proteasome inhibition. (A-D) B8-Db cells were incubated with IFNγ for 72h. Next, cells were starved for one hour, pulse-labeled with [35S]-methionine/cysteine for 5 minutes and chased for the indicated time periods. Cells were treated with proteasome inhibitor MG132 for the last 30 minutes of starvation, during labeling and chase or left untreated. Lysates were immunoprecipitated with polyclonal antibodies against HA-2Dd, -H-2Db, -H-2Kd or -H-2Ld and separated on 15% SDS-polyacrylamide gels. The proteins were visualized by autoradiography. The experiments were repeated twice with similar outcome.

6.2. Immunoproteasome precursor organization in murine astrocytes

6.2.1 Establishing polyclonal rabbit antibodies against LMP2, LMP7 and MECL-1 To analyze the immunoproteasome assembly in murine brains, we had to raise antibodies against murine immunoproteasome subunits LMP2, LMP7 and MECL-1. We used already published LMP2, LMP7 and MECL-1 peptides (Guillaume et al. 2010) to immunize rabbits (Chapter 5.6.3). The sera were purified with the help of KLH-conjugated peptides (Chapter 5.3.10). We analyzed sera and purified antibodies in Western blots of LCMV-infected C57BL/6 mice spleen lysates. We took spleen lysates of LMP2-/- (Van Kaer et al. 1994), LMP7-/- (Fehling et al. 1994) or MECL-1-/- (Khan et al. 2001) mice as a control (Fig 6.34). The generated antibodies were able to detect the murine immunoproteasome subunits (LMP2: 21 KDa, LMP7 23 kDa, MECL-1: 25 kDa). After antibody purification, the background bands disappeared. We tried to use the antibodies for immunoprecipitation and immunohistochemistry but they could not be used for these approaches.

Fig 6.34│Western blot analysis of polyclonal rabbit anti-mouse LMP2, LMP7 and MECL-1 antibodies. C57BL/6 (+), LMP2-/-, LMP7-/- or MECL-1-/- (-) mice were infected with LCMV. After 72 hours of infection, mice were sacrificed and spleen lysates were separated on 15% SDS-polyacrylamide gels. Samples were analyzed on Western blots probed with polyclonal antibodies against LMP2, LMP7 or MECL-1. The sera of immunized rabbits were compared to the respective purified antibodies. The experiments were repeated twice with similar outcome.

Next, we analyzed the ability of the antibodies to detect human immunoproteasome subunits in Western blots. For this application, LCL721 (human lymphoblastic cell line) and LCL721.174 (human lymphoblastic cell line without LMP2, LMP7 and less MECL-1 expression) cell lines were used. Cells were incubated with IFNγ for 72 hours and lysed. We were able to detect human LMP7 and MECL-1 immunoproteasome subunits in Western blots probed with MECL-1 and LMP2-specific sera or purified antibodies (Fig 6.35). LMP7 antibodies were not able to detect human LMP7 (data not shown).

Fig 35│ Western blot analysis of polyclonal rabbit anti-mouse LMP2 and MECL-1 antibodies. LCL721.145 (+) or LCL.721.174 (LMP2-/- , LMP7-/- and less MECL-1 protein; (-)) cells were stimulated with IFNγ for 72 hours, lysed and the proteins were separated on 15% SDS-polyacrylamide gels. Lysates were analyzed on Western blots probed with polyclonal antibodies against LMP2 or MECL-1. The sera of immunized rabbits were compared to the respective purified antibodies. The experiments were repeated twice with similar outcome.

6.2.2 The immunoproteasome assembly in IMA2.1 murine astrocyte cell line. Former studies revealed that immunoproteasome subunit precursor proteins accumulate in murine astrocytes treated with IFNγ for 72 hours (Kremer et al. 2010). Experiments were performed as described in the publication with the murine astrocyte cell line IMA2.1. Cells were either induced with IFNγ for 72 hours or left untreated. Samples were analyzed on Western blots probed with LMP2-, LMP7- or MECL-1-specific antibodies. We used purified immunoproteasomes as positive control. LMP2, LMP7 and MECL-1 precursor protein accumulation was detectable after 72 hours IFNγ treatment. LMP2 precursor protein was detectable with the size of 23 kDa, LMP7 precursor protein with 26 kDa and MECL-1 precursor protein with 29 kDa (Fig 36).

Fig 6.36│Immunoproteasome subunit expression in immortalized astrocyte cells after IFNγ-stimulation. IMA2.1 cells were either left untreated (-) or stimulated with IFNγ for 72 hours (+). Lysates were analyzed on Western blots probed with polyclonal antibodies against MECL-1, LMP2 and LMP7. Tubulin was used as a loading control. Purified immunoproteasome from LCMV-infected mouse liver served as a positive control (C). The experiment was repeated twice with similar outcome.

We performed 10-40% sucrose density gradient centrifugation with lysates of IMA2.1 cells after 72 hours of IFNγ induction to explore the mechanism of immunoproteasome assembly and precursor formation in murine brains. We used untreated IMA2.1 cells and IFNγ stimulated B8 cells as controls. The density sucrose gradient was separated into 20 fractions. Fractions were chloroform-methanol precipitated (Wessel & Flügge 1984) and loaded on 15% SDS- polyacrylamide gels to separate the proteins by size. After Western blotting on nitrocellulose membranes and probing with anti-LMP2, -LMP7 or -MECL-1 antibodies, the protein distribution in the 20 fractions was analyzed. We detected mature LMP2 protein in fractions 13-19 in untreated IMA2.1 and IFNγ treated B8 cells. In IFNγ treated IMA2.1 cells, mature LMP2 protein was detectable in fraction 12-15. Additionally, in cells treated with IFNγ, two bands with the sizes of the precursor and the mature protein were detectable in the lower density fractions 2-4 (Fig 6.37A). Mature LMP7 protein was detected in fractions 14-19 in IFNγ-stimulated B8 and IMA2.1 cells and in unstimulated IMA2.1 cells. Precursor and mature protein were detected in IFNγ- induced IMA2.1 sample fractions 3-9 (Fig 6.37B). MECL-1 Western blots revealed an expression of the protein in high-density fractions 14-19 in IFNγ treated IMA2.1 and B8 cells. There was no MECL-1 expression in high-density fractions of untreated IMA2.1 cells detectable. MECL-1 precursor protein was observable in fractions 3-7 in untreated and IFNγ treated IMA2.1 cells (Fig 6.37C).

Fig 6.37│Density gradient analysis of MECL-1, LMP2 and LMP7 immunoproteasome precursor proteins in immortalized astrocyte cells. (A-C) IMA 2.1 cells were either stimulated with IFNγ for 72 hours or left untreated. Lysates were subjected to ultracentrifuge density gradient centrifugation and fractions were analyzed for immunoproteasome subunit expression on Western blots probed with polyclonal antibodies against MECL-1, LMP2 and LMP7. Purified immunoproteasome from LCMV-infected mouse liver served as a positive control (C). The gradient density increases from 15 to 40% toward fraction 20. Lysates of B8 cells stimulated with IFNγ were used as a negative control. The experiments were repeated twice with similar outcome.

6.2.3 PI31 expression in murine astrocytes and immunoproteasome assembly PI31 protein is known to partly suppress the function of 20S and 26S proteasomes (M Chu-Ping et al. 1992). We thought about a potential regulation of the immunoproteasome assembly in astrocytes by PI31. To test this hypothesis, we analyzed PI31 mRNA levels with quantitative RT- PCR analysis in IMA2.1 and B8 cells (treated with IFNγ for 72 hours or left untreated). There was no difference between IFNγ treated or untreated cells. Then, we used anti-PI31 Western blot analysis with the same experimental setup and there was slightly less PI31 detectable in IMA2.1 cells treated with IFNγ (Fig 6.38).

Fig 6.38│PI31 expression in IFNγ-stimulated, immortalized astrocyte and fibroblast cells. IMA2.1 and B8 cells were either treated with IFNγ for 72 hours or left untreated. After lysis, proteins were separated on 15% SDS- polyacrylamide gels and analyzed on western blots probed with a polyclonal antibody against PI31. Tubulin was used as a loading control. The experiment was repeated twice with similar outcome.

6.2.4 Establishing stable FLAG-tagged Mecl-1 expressing cell lines For further investigation, immunoprecipitation of immunoproteasome subunits was inevitable but no appropriate antibodies for immunoprecipitation of LMP2, LMP7 and MECL-1 were available. We cloned a C-terminally three times FLAG-tagged MECL-1 overexpression construct (Fig 6.39A). We transiently overexpressed MECL-1-FLAG3 construct in IMA2.1 and B8 cells and analyzed the expression with anti-FLAG Western blots. Mature MECL-1-FLAG3 protein was detectable with a size of 31 kDa. MECL-1-FLAG3 precursor protein was detectable with a size of 33 kDa in lower amounts. In untreated IMA2.1 cells there was an additional 36 kDa protein band detectable (Fig 6.39B).

Fig 6.39│Overexpression of FLAG-tagged MECL-1 in immortalized astrocyte- and fibroblast cells. (A) Scheme of C- terminally three-times FLAG-tagged MECL-1 protein. (B) IMA2.1 and B8 cells were transfected with MECL-1-FLAG3 and lysed. Proteins were separated on 15% SDS-polyacrylamide gels and analyzed on Western blots probed with a mAb against the FLAG-tag or PSMA6. The experiments were repeated twice with similar outcome.

Next, we performed 10-40% sucrose density gradient centrifugation to separate the overexpressed MECL-1-FLAG3 protein and its precursor by their density. We used IMA2.1 cells treated for 72 hours with IFNγ in this experiment. Cells were lysed, anti-FLAG immunoprecipitated and separated in glucose gradients, as describe before. The density gradient fractions were analyzed on anti-FLAG Western blots. We detected a 34 kDa MECL-1- FLAG3 precursor protein in the low-density fractions 2-5 and the mature MECL-1-FLAG3 protein in high-density fractions 9-16 (Fig 6.40A). We used B8 cells with the same protocol as a control and detected the mature MECL-1-FLAG3 protein in fractions 1-6 and 11-20 but no precursor protein (Fig 6.40B).

Fig 6.40│Density gradient analysis of transiently overexpressed MECL-1-FLAG3 protein in immortalized astrocyte- and fibroblast cells. (A,B) IMA2.1 and B8 cells were transfected with MECL-1-FLAG3 and stimulated for 72 hours with IFNγ. Lysates were subjected to ultracentrifuge density gradient centrifugation. Fractions were analyzed by western blotting with a mAb against the FLAG-tag or PSMA6. The gradient density increases from 15 to 40% toward fraction 20. The experiments were repeated twice with similar outcome.

For future studies, we started to establish IMA2.1 and B8 cell lines stably expressing MECL-1- FLAG3, LMP2 and LMP7 to avoid cost-intensive IFNγ stimulation of cells (data not shown).

100

7. Discussion

7.1 The fate of ER-targeted proteins in the face of proteasome inhibition

7.1.1 Mislocalized proteins and protein aggregates in the cytoplasm of mammalian cells The most diverse proteome resides in the cytosolic compartment of cells. A tremendous amount of proteins and the myriad ways in which they can be defective request a flexible quality control system, because cellular homeostasis depends highly on protein quality surveillance. Additionally, various secreted or membrane-bound proteins targeted to the secretory pathway have been shown to play functional or pathological roles when they accumulate at alternative cellular compartments because of improper translocation. There is a wide range of diseases known to be associated with the failure of prompt disposal of undesired proteins aggregating in the cytosol, e.g. neurodegenerative diseases and cancer (Selkoe 2003; Rubinsztein 2006; Soto et al. 2006). Mislocalized proteins, that have been inappropriately released to the cytosol, are susceptible to aggregation, misfolding, inappropriate interaction with other proteins and interfere with different cellular pathways (Lansbury & Lashuel 2006; Ross 2002; Chakrabarti et al. 2011). This necessitates several flexible and robust possibilities to degrade them efficiently. Pathways to remove defective cytoplasmic proteins are diverse, including ribosome-associated systems for nascent proteins (Pechmann et al. 2013), chaperone-assisted pathways of degradation (Kettern et al. 2010) and autophagy-based pathways for large multimeric protein aggregates (Jimenez- Sanchez et al. 2012). Defining their full complement is a current goal in the quality control field. There is a need to understand the limits of these pathways and the relevant circumstances to exceed them. Moreover, further investigation of consequences of aberrant proteins and their aggregates for cellular physiology with new model systems and endogenously expressed proteins is required.

Major prion protein (PrP) is one of the best-studied proteins and known to cause neurotoxic aggregates when accumulated at alternative cellular destinations (Drisaldi et al. 2003; Rane et al. 2004; Ma et al. 2002). PrP has been shown to aggregate in the cytoplasm during situations of acute ER stress because of attenuated translocation (Kang et al. 2006). Moreover, unglycosylated PrP precursor protein accumulates during proteasome inhibition in the cytoplasm of cells, indicating the existence of a cytoplasmic proportion of this protein that is degraded in a proteasome dependent manner. Precursor protein stabilization of PrP during proteasome

101

inhibition was not detectable in endogenously expressed PrP (Drisaldi et al. 2003; Rane et al. 2004). Yet, we stand at the beginning of understanding the mechanisms of mislocalized PrP targeting to the proteasome in vitro. The targeting involves, among others, chaperone Bag6 and depends on unprocessed or non-membrane inserted, hydrophobic domains that characterize cytoplasmic PrP forms as mislocalized. Ubiquitination and degradation of mislocalized PrP proteins is only slightly attenuated during Bag6 knockdown in vivo, presumably because of overlapping cytosolic quality control pathways (Rodrigo-Brenni et al. 2014; Hessa et al. 2011).

An epitope of prostate stem cell antigen (PSCA), which is MHC class I presented, is not processed by the ER signal peptidase, demonstrating that a proportion of ER-targeted proteins is degraded by the proteasome and loaded on MHC class I molecules before entering the ER (Schlosser et al. 2007). Prostate stem cell antigen was N-terminally FLAG- and C-terminally HA- tagged, which allows us to completely follow the maturation of the ER-guided precursor protein. The C-terminal HA-tag replaced the GPI anchor of PSCA. Analysis of the cellular localization and glycosylation pattern of FLAG-PSCA-HA protein ensured that the N-terminal FLAG-tag does not interfere with translocation of PSCA into the ER (Fig 6.1 and 6.2). In this study, we found that PSCA precursor protein is stabilized during proteasome inhibition in the cytosol of cells, regardless of type of inhibition and cell line used (Fig 6.3). This correlates with results seen for overexpressed PrP by other research groups. 5-10% of emerging PrP molecules fail to enter the ER and are not originated from the ER lumen (Drisaldi et al. 2003; Rane et al. 2004; Chakrabarti et al. 2011; Rane et al. 2010; Levine et al. 2005). We detected a precursor-product relationship between PSCA precursor and mature protein. The premature, unglycosylated protein is stabilized more than five times during proteasome inhibition as compared to untreated cells. Its localization is mainly cytosolic as shown by Endoglycosidase H sensitivity assay and cellular fractionating experiments (Fig 6.2, 6.11 and 6.12). Furthermore, we detected a stabilization of the PSCA precursor protein in cells challenged with oxidative stress (Fig 6.10), which is known to inhibit the 26S proteasome (reviewed in Aiken et al. 2011).

We applied the FLAG-PSCA-HA precursor protein double-tagging strategy for two ER-guided peptide hormones, human Prolactin and Leptin, to show that the effect is not limited to PSCA protein. FLAG-tagged Prolactin precursor protein is slightly stabilized during proteasome inhibition (Fig 6.14). Double-tagged Leptin precursor protein is detectable up to five times longer in cells treated with proteasome inhibitor or hydrogen peroxide as compared to control cells. N- terminally FLAG-tagging, as previously shown for PSCA, does not alter co-localization of Leptin and endoplasmic reticulum marker Climp63 (Fig 6.15). This qualifies our double-tagging strategy 102

as an excellent tool for further experiments in deciphering the fate of precursor proteins during proteasome inhibition. We demonstrate here that FLAG-PSCA-HA protein is transported into the ER and glycosylated in the same way in untreated and MG132 treated cells, even though an unglycosylated protein species clearly accumulates in cells incubated with the proteasome inhibitor MG132 (Fig 6.2). This finding argues against a general precursor protein ER import deficiency during proteasome inhibition. It would be of interest to compare glycosylation pattern of PSCA in hydrogen peroxide challenged cells as compared to MG132 treated cells, to further identify the mechanism of hydrogen peroxide-induced precursor protein stabilization. Despite an unaltered ER luminal localization of N-terminally FLAG-tagged PSCA, it seems that the FLAG- tag changes the translocation rate of the protein. Pulse-chase studies with N-terminally untagged PSCA revealed no detectable separation of protein species into precursor and mature protein (Fig 6.13). One possible explanation is that FLAG-tagging slows down a physiologically rapid processing of PSCA precursor protein, or enhances a very slow one, making it detectable with the time frame used in our system (60 minutes of chase). It was shown before, that only a minority of secretory proteins, e.g. Prolactin and PSCA, are processed efficiently (Levine et al. 2005; Schlosser et al. 2007). Furthermore, the possibility exists, that the additional one kilo Dalton in size, added by the FLAG-tag to the signal peptide, enables us to separate both bands appropriately. The analysis of the glycosylation pattern of N-terminally untagged PSCA-HA protein during proteasome inhibition, and allows for judging its translation efficiency, could help to answer the question.

In order to circumvent potential side effects of protein tagging, we investigated the process of precursor stabilization during ER stress with untagged and endogenously expressed proteins.

We were not able to separate precursor and mature proteins of α1-Antitrypsin (AAT, Fig 6.27), β2-microglobulin (B2M, Fig 6.28), Human cytomegalovirus gene product US11 (Fig 6.31) and C- C chemokine receptor type 7 (CCR7, Fig 6.32). During proteasome inhibition, a stabilization of LCMV glycoprotein precursor was visible, but there was no precursor-product relationship (Fig 6.29). Our results prove that untagged carbonic-anhydrase 4 (CA4) precursor protein is stabilized in transfected cells during proteasome inhibition (Fig 6.30), as reported before (Rebello et al. 2004). We could not continue the experiments with endogenously expressed CA4 because it is not expressed in all common available cell lines that we tested (data not shown). H- 2Dd, H-2Db, H-2Kd and H-2Ld are murine MHC class I molecules and their expression is enhanced in cytokine stimulated cells. Endogenously expressed H-2Dd precursor protein is stabilized in cells treated with proteasome inhibitors. Interestingly, we were not able to separate precursor and mature protein species of the other three MHC class I molecules mentioned 103

before, although we used the same experimental conditions and samples (Fig 6.33). It becomes clear that susceptibility to proteasome inhibition and precursor processing characteristics are distinct from protein to protein. A possible explanation for this observation is the varying signal sequence composition of the examined proteins. Protein sequence features (signal sequence, length, hydrophobicity, amino acid composition, general domain structure) and functional properties (like the interaction with translocon factors) can influence the behavior of ER-guided proteins. It has already been shown that differences among signal sequences facilitate substrate-specific modulation of protein translocation during acute ER stress (Kang et al. 2006). In that study, there was evidence that ER stress-induced accumulation of cytoplasmic protein species strongly depends on the type of protein that is investigated. Angiotensin, IFNγ and corticotropin-releasing factor receptor (CRFR) accumulate in the cytosol during ER stress, but TRAPα, Frizzled-7, vesicular stomatitis virus glycoprotein (VSVG) or vascular cell adhesion molecule (VCAM) do not accumulate. Replacement of CRFR signal sequence with the signal sequence of Prolactin rescued its ER stress-induced translocation attenuation. Hence, domain swapping experiments with FLAG-PSCA-HA and Prolactin (or other) precursor protein signal sequences, followed by precursor stability analysis during proteasome inhibition, would help to further investigate the effect of signal peptide composition on proteins in quality control pathways and degradation. Pathways of protein quality control, degradation and translocation are interwoven and overlapping, which enhances the difficulties to identify single regulation mechanisms. Accordingly, it was shown by Hegde and colleagues that even different cell line and cell culture conditions affect the translocation efficiency of cytoplasmic PrP during DTT- induced ER stress (Levine et al. 2005).

7.1.2 Mechanisms of ER-guided precursor protein accumulation during proteasome inhibition Translational, translocational and post-translocational mechanisms contribute to cytosolic localization of ER-guided proteins. The Kozak’s consensus sequence initiates the mRNA translation and defines its starting point (Kozak 1992). From time to time, the consensus sequence is skipped and translation initiates from an internal AUG codon due to leaky ribosomal scanning. This results in proteins lacking signal peptides, which are not recognized as targets for translocation, hence are degraded in the cytosol (Kozak 2002). It was shown that 20-50% of the cytosolic enriched fraction of an overexpressed protein with Prolactin signal sequence is produced by this translational effect (Levine et al. 2005). Protein species developed from skipping the ideal Kozak’s sequence are not detected in our FLAG-PSCA-HA pulse-chase label

104

experiments because they miss the FLAG-tag and are not immunoprecipitated. Every step between the transport of the RNC complex to the ER membrane and the complete translocation of the protein is potentially error-prone. Malfunction results in release of physiologically ER- guided proteins into the cytosol. 10-20% of overexpressed mammalian prion protein (PrP) is mislocalized to the cytosol and rapidly degraded by the proteasome (Drisaldi et al. 2003; Rane et al. 2004). It is not trivial to determine the exact translocation step that fails and causes cytosolic enrichment of those secretory proteins. SRP binding to the signal peptide could fail or the following RNC targeting to the translocon complex is insufficient. Sec61 signal recognition could be defective or the insertion into the translocon could be inadequate. The translocation process or signal peptidase processing could be inefficient. Problems in translocon channel opening or restrained interaction with luminal chaperones would also lead to a cytosolic accumulation of untranslocated proteins. ERAD delivery of proteins can be hijacked, especially from pathogens to facilitate their various replication strategies (Byun et al. 2014). Escape of secretory proteins from ERAD could lead to the accumulation of ER-targeted proteins in the cytosol, despite the evidence that cytoplasmic PrP has never been translocated into the ER lumen (Drisaldi et al. 2003; Rane et al. 2004). This mechanism could play a role in PSCA precursor protein accumulation during proteasome inhibition. Our newly established FLAG- PSCA-HA construct emerges as an excellent tool to verify the involvement of the different translocations steps in this process and shed light on this question. During the last years, initial steps to elucidate the phenomenon of PrP mislocalization and degradation were made. Recent studies suggested a role for Bag6 chaperone and ubiquitin ligase RNF126 in the delivery of cytosolic PrP to the 26S proteasome (Rodrigo-Brenni et al. 2014; Hessa et al. 2011). In addition, Bag6 is proposed to generate peptides for MHC class I presentation (Minami et al. 2010), which would fit into the hypothesis of MHC class I peptide generation due to secretory protein mislocalization. We used Bag6 siRNA knockdown to investigate its role as a chaperone in the targeting of FLAG-PSCA-HA precursor protein to the 26S proteasome. FLAG-PSCA-HA precursor protein does not accumulate during Bag6 knockdown (Fig 6.26), indicating that its way to destruction is different than the one seen for PrP. To completely exclude a Bag6 involvement, the result must be reproduced. The identification of chaperones and E3 ligases involved in PSCA preprotein degradation is one of the most important tasks in future work with our FLAG- PSCA-HA overexpression technique. Interestingly, an increase of overexpressed but not endogenously expressed cytosolic PrP mRNA was detected in proteasome inhibitor treated cells (Drisaldi et al. 2003; Rane et al. 2004). Thus, it would be of interest to determine the amount of PSCA, Leptin, CA4 and H-2Dd mRNA during MG132 proteasome inhibitor treatment.

105

Our observations imply the existence of various ER-guided precursor proteins that are efficiently degraded by the 26S proteasome right after their translation, and not translocated into the ER lumen, which leads to the cytosolic abundance of the ER-guided proteins and finally their respective peptides. There are several possible reasons why the partial cytoplasmic localization of secretory proteins would be advantageous for cells. Peptides loaded onto MHC class I molecules facilitate cells to present degraded, cytoplasmic protein fragments to CD8+ T-cells. In fact, it is assumed that 30% of all proteins are degraded immediately after their synthesis where they account for a large part of the antigenic peptide repertoire (Reits et al. 2000; Turner & Varshavsky 2000; Schubert et al. 2000). Moderately inefficient ER targeting could generate a large pool of cytosolic substrates, which are degraded into peptides and can be used as an abundant source for MHC class I presented epitopes. A direct use of these proteins for the purpose of generating self-antigens is more efficient and profitable than a two-step mechanism using ERAD after translocation of proteins into the ER. Some years ago, we found evidence for the existence of MHC class I peptides derived from proteins that have never entered the ER. An MHC class I loaded peptide with an intact ER signal peptidase cleavage site (which means it was not inserted into the ER) originated from PSCA precursor protein was identified (Schlosser et al. 2007). A related hypothesis was postulated from Jonathan Yewdell in 1996, supposing defective ribosomal products as a source for antigenic peptides (mDRiPs; Yewdell et al. 1996). Inefficiently segregated proteins with independent functions in cytosol and ER (or other cellular locations to which they can be transported from there, e.g. nucleus or mitochondria) increase the functional diversity of a single mRNA. If only 1-5% of all translated secretory proteins compartmentalize inefficiently, a cytosolic function is considerable and the major ER-guided population is not biased. There is further evidence for such a mechanism, for example the ER luminal chaperone calreticulin was initially identified as a cytosolic integrin-binding protein (Coppolino et al. 1995; Coppolino et al. 1997) and later found to be a nuclear export factor (Holaska et al. 2001), a regulator of steroid hormone receptor activity (Burns et al. 1994; Dedhar 1994) and an mRNA binding protein regulating p21 translation (Iakova et al. 2004). Other examples for secretory proteins with alternative localization and function include the β subunit of glucosidase II (Trombetta et al. 1996), cytosolic adaptor protein Grb3 (Goh et al. 1996; Kanai et al. 1997), growth factor-β (Heine et al. 1991), mitochondrial Slit3 (Little et al. 2001), mitochondrial cytochrome P450 (Avadhani et al. 2011; Burns et al. 1994), cytosolic IL15 (Tagaya et al. 1997; Kurys et al. 2000), nucleocytoplasmic cathepsin L (Goulet et al. 2004) and nuclear HBV precore protein (Garcia et al. 1988; Ou et al. 1989). There are many pathways of protein translocation into the ER with different components, for instance proteins with weak signal peptides require TRAP and TRAM to be transferred into the ER lumen, while efficiently 106

transported proteins are translocated in a TRAP and TRAM-independent manner (Fons et al. 2003; Voigt et al. 1996). The existence of different translocation pathways gives the cell an opportunity to adjust compartmentalization of distinct secretory proteins. Slight inefficiencies in translocation of specific proteins can be beneficial for the cell, but on the other hand, the constant presence of some ER-targeted proteins in the cytosol is in many cases associated with pathological consequences. This was, for example, shown for PrP (Ma & Lindquist 2002), ApoE (Huang et al. 2001) and APP (Lustbader et al. 2004). Evolutionary conservation of several signal sequences and high variation in translocation requirements for ER-targeted proteins allow cells to handle the delicate balance between pathological- and beneficial consequences of inefficiently segregated, secretory proteins.

The initial results of this study allow two different ways of interpretation. 1) Precursor protein stabilization might be due to a physiological effect in which ER-guided proteins are constantly degraded in the cytosol directly after translation (e.g. for mentioned above MHC class I peptide generation) and 2) there is a feedback mechanism which is required during accumulation of misfolded, defective proteins in the cytosol and/or the ER to prevent their further transport into the ER. However, it is unclear which proteins or regulation pathways are involved and, furthermore, cellular homeostasis pathways are highly overlapping and associated with each other. We already excluded the contribution of autophagy- and aberrant secretion mechanisms to PSCA precursor protein stabilization (Fig 6.4). Induction of heat shock response, unfolded protein response and nitrogen stress as well as inhibition of autophagy, ERAD and ribosomal translation does not stabilize PSCA precursor protein. Activation of oxidative stress response had no effect on PSCA precursor protein stability, implying that hydrogen peroxide-induced stabilization of premature FLAG-PSCA-HA relies rather on its 26S proteasome inhibition effect than on induction of oxidative stress response pathways with Nrf1- or Nrf2 involvement (Fig. 6.18 and 6.22). In contrast to PrP precursor protein (Kang et al. 2006), induction of ER stress did not stabilize cytosolic PSCA precursor protein. Further studies are required to determine the effect of ER stress on CA4 and H-2Dd precursor stabilization. Accumulation of ubiquitinated signal peptides may serve as an alert for impaired proteasomal protein degradation and initiate a delay of protein translocation into the ER, but we could demonstrate that stabilization of PSCA precursor protein is independent of ubiquitination of its signal peptide. The accumulation of polyubiquitinated protein conjugates because of proteasome inhibitor- or hydrogen peroxide treatment may induce an inhibition of protein translocation into the ER. In this study, we could show that overexpression of mono-ubiquitin had no effect on PSCA precursor protein appearance (Fig 6.20). Analysis of the ubiquitination state of FLAG-PSCA-HA precursor protein 107

during proteasome inhibition would provide further information about its cytosolic localization and elucidate its pathway of degradation. Impaired degradation and accumulation of misfolded ER resident proteins due to ERAD inhibition could trigger a translocation stop. However, inhibition of VCP/p97 did not stabilize PSCA precursor protein (Fig 6.18 and 6.21). BiP is a chaperone known for binding hydrophobic patches of nascent polypeptides within UPR signaling. Inhibition of BiP did not interfere with FLAG-PSCA-HA precursor stabilization, arguing against a feedback mechanism that depends on BiP as an ER stress sensor that inhibits translocation during proteasome inhibition to prevent further import of defective proteins (Fig 6.21).

The ER signal peptidase is a key factor for protein import into the ER lumen because it processes precursor proteins to their mature forms before they can be completely released into the ER. Therefore, we analyzed its participation in FLAG-PSCA-HA precursor protein accumulation during proteasome inhibition in detail. Signal peptidase siRNA knock down and proteasome inhibition stabilize FLAG-PSCA-HA precursor protein to the same extent (Fig 6.6). Further studies, comparing glycosylation patterns of precursor proteins in cells with silenced ER signal peptidase activity as compared to MG132 treated ones would help to identify a potential common mechanism. Stabilization of precursor proteins occurs to fast to require translation or splicing of regulatory protein factors (e.g. spliced Xbp1 mRNA is mostly expressed after 6h; Van Schadewijk et al. 2012). Inhibition of the ER signal peptidase activity as a mechanism to pause protein translocation into the ER would assume binding of regulatory proteins or structural modifications, like phosphorylation. However, we could not detect tyrosine-phosphorylation of ER signal peptidase subunits resulting from proteasome inhibition (data not shown). Radioactive experiments using orthophosphate to detect phosphorylation of the subunits would be more sensitive and could additionally detect serine and threonine phosphate modifications. Experiments, detecting FLAG-PSCA-HA precursor protein stabilization during inhibition of cytoplasmic kinases would clarify this aspect, too. By mass spectrometry, we identified 21 proteins that are specifically bound to the ER signal peptidase in the course of proteasome inhibition (Table 6.1). Almost all identified proteins are highly abundant translation- or metabolism-associated proteins. An interesting candidate is UPF0368 (Cxorf26), whose function is not known but which was shown to interact with ubiquitinated proteins (Sowa et al. 2009). Such a protein could be the missing link between proteasome impairment and translocation inhibition and further interaction studies of UPF0368 and the ER signal peptidase ought to be performed. Close proximity of ER signal peptidase subunits and ERAD machinery proteins would support the hypothesis of a feedback regulation mechanism. Nonetheless, by using immunoprecipitation, we could not detect an interaction between of SPC18, SPC21 or SPC22/23 108

subunits and Derlin1 (Fig 6.24). In order to confirm this finding, the results should be validated by crosslinking experiments or proximity ligation assays that are more sensitive. Additionally, other ERAD cofactors (like Hrd1 or VCP/p97) should be tested for their association with the ER signal peptidase. Double-tagged PSCA is a perfect tool for future studies, because it allows us to immunoprecipitate PSCA efficiently and to distinguish between precursor and mature PSCA protein. Moreover, one could use crosslinking experiments with FLAG-PSCA-HA and Sec61/SRP receptor to detect differences of PSCA precursor targeting and translocon binding during proteasome inhibition.

The quantity of glycosylated FLAG-PSCA-HA isoforms in MG132 treated cells is slightly reduced after 30 minutes of chase, which supports the feedback mechanism hypothesis, at least for FLAG-PSCA-HA precursor protein (Fig 6.2). We used Concanavanin A lectines to immunoprecipitate N-glycosylated proteins and to determine the ratio of glycosylated proteins to cytosolic proteins in cells treated with proteasome inhibitors. There were no differences in this ratio between untreated and MG132 treated cells, contradicting the hypothesis of a general secretory protein translocation inhibition due to proteasome impairment. However, some protein bands were reduced during MG132 treatment (data not shown) and we cannot exclude the existence of such a mechanism for individual proteins, defined by properties of their signal sequences.

7.1.3 Final remarks and outlook Taken together, our results point to a mechanism that is rather based on the degradation of secreted proteins right after translation than on a feedback mechanism that inhibits translocation of maturated proteins when misfolded proteins are accumulating in the cytosol. We proved the cytosolic accumulation of FLAG-PSCA-HA to be independent of side effects from chemical proteasome inhibitors and its relevance for endogenously expressed proteins. Now, we begin to understand the mechanism behind this observation, but there are still many open questions. First, it is necessary to reproduce Bag6 knockdown effects on FLAG-PSCA-HA precursor protein degradation and to identify other chaperones and E3 ligases leading to recruitment of PSCA to the 26S proteasome. It was shown for cytoplasmic PrP that N- or C-terminally located hydrophobic sequences are needed for its ubiquitination and proteasomal degradation (Hessa et al. 2011) and we should prove such a mechanism for PSCA. It was shown that proteins with extremely efficient signal sequences (e.g. Osteopontin) did not accumulate during MG132 treatment (Rane et al. 2004; Kang et al. 2006). Domain swapping experiments using

109

Osteopontin signal peptide and for FLAG-PSCA-HA protein during proteasome inhibition would help to understand the role of translocation efficiencies with respect to precursor compartmentalization and resulting MHC class I peptide assembly. Moreover, analyzing the accumulation of FLAG-PSCA-HA precursor protein during kifunensine treatment, which inhibits access of glycoproteins to ERAD, could shed more light on the participation of ERAD to the mislocalization process. Integration of proteins into Sec61 is inhibited by cotransin and investigation of FLAG-PSCA-HA glycosylation pattern during cotransin treatment, as compared to MG132 treatment, could help to identify possible overlapping pathways. Mislocalized PrP aggregates are believed to contribute to translocation inhibition of cytoplasmic PrP (Ma et al. 2002; Chakrabarti et al. 2011). Hence, it would be worthwhile to investigate whether cytosolic FLAG-PSCA-HA proteins aggregate during MG132 treatment, too. Additionally, the reversibility of FLAG-PSCA-HA precursor protein accumulation after removal of MG132 should be analyzed.

Huge proteasomal burden because of misfolded immunoglobulin chains, permanently activates the unfolded protein response and ERAD in multiple myeloma cells (Bianchi et al. 2009; Cenci et al. 2006), resulting in high sensitivity to proteasome inhibitors like bortezomib. An intracellular mechanism that degrades secretory proteins right after translation, to enhance the antigenic repertoire of MHC class I molecules, would further explain the extensive cytotoxic effect of proteasome inhibition. Identifying the mechanism of protein mislocalization would help to better understand the major developing processes of many human diseases. Several neurodegenerative disorders, like Alzheimer’s disease, Parkinson’s disease or Huntington’s disease, and various cancer types are caused by cellular accumulation of misfolded proteins (Selkoe 2003; Rubinsztein 2006; Soto et al. 2006; Dai et al. 2007; Morimoto 2008). Additionally, selective transport inhibition of ER-targeted proteins determines polycystic liver disease, which is characterized by progressive development of biliary epithelial cysts throughout the liver (Davila et al. 2004). Pathologies, like lysosomal storage disease (Sawkar et al. 2006) and cystic fibrosis (Koulov et al. 2010), result from misfolded proteins, too. Elucidating the reason why cells occasionally fail to compartmentalize misfolded cytotoxic species, endorsing them to interfere with normal cellular protein homoeostasis, will assist to shed light on the etiology of amyloid diseases.

110

7.2. Immunoproteasome precursor organization in murine astrocytes

The assembly of immunoproteasomes is an event that includes multiple steps. N-terminal prosequences and C-terminal extensions of the subunits render the proteasomes catalytically inactive until they are cleaved off in the final maturation step. This autocatalytic cleavage generates the functional active conformation of the 20S proteasome complex (Schmidtke et al. 1996). The assembly of immunoproteasome subunits seems to be partially impeded in the brain, which is indicated by the accumulation of immunoproteasome precursors (Kremer et al. 2010). In this study, we generated polyclonal antibodies against the immunoproteasome subunits LMP2, LMP7 and MECL-1 and were able to confirm the findings of Kremer et al. This points to the existence of a post-translational mechanism, which regulates immunoproteasome formation in areas where uncontrolled inflammatory responses could cause considerable harm destroying fundamental, non-renewable cells. An enrichment of unprocessed immunoproteasome precursor proteins in lower fractions of density gradients of cytokine-induced astrocytes supports the hypothesis of incomplete proteasome formation and precursor proteins that accumulate in half- proteasomes (Chen & Hochstrasser 1995), stated by Kremer et al. Next, we used FLAG-tagged MECL-1 and detected the same effect, which proves the construct as an excellent tool for further research and immunoprecipitation experiments. Establishing a cell line that stably expresses the FLAG-tagged MECL-1 construct would be the next step towards the elucidation of this mechanism. Factors that regulate the suppression of immunoproteasome formation in brains are unknown, but such a mechanism may exist in other cells. PI31 is a protein that was identified as a suppressor of 20S and 26S proteasomes function (M Chu-Ping et al. 1992) and overexpression of PI31 in astrocytes could explain the precursor accumulation. Our results are contrary to this hypothesis. The amount of P31 in astrocytes was slightly reduced as compared to control cells. Clarifying the role of immunoproteasome expression and the mechanism of its reduced formation in brains could provide new insights into immune response regulation in immunoprivileged organs and help to understand how active immune tolerance can be achieved.

111

8. Appendix 8.1 References

Acosta-Alvear, D. et al., 2007. XBP1 Controls Diverse Cell Type- and Condition-Specific Transcriptional Regulatory Networks. Molecular Cell, 27, pp.53–66.

Adachi, Y. et al., 2008. ATF6 is a transcription factor specializing in the regulation of quality control proteins in the endoplasmic reticulum. Cell structure and function, 33, pp.75–89.

Adams, J. et al., 1998. Potent and selective inhibitors of the proteasome: Dipeptidyl boronic acids. Bioorganic and Medicinal Chemistry Letters, 8, pp.333–338.

Adibhatla, R.M. & Hatcher, J.F., 2010. Lipid oxidation and peroxidation in CNS health and disease: from molecular mechanisms to therapeutic opportunities. Antioxidants & redox signaling, 12, pp.125–169.

Aebi, M. et al., 2010. N-glycan structures: recognition and processing in the ER. Trends in Biochemical Sciences, 35, pp.74–82.

Ahn, J.Y. et al., 1995. Primary structures of two homologous subunits of PA28, a gamma-interferon-inducible protein activator of the 20S proteasome. FEBS letters, 366, pp.37–42.

Aiken, C.T. et al., 2011. Oxidative stress-mediated regulation of proteasome complexes. Molecular & cellular proteomics : MCP, 10, p.R110.006924.

Aki, M. et al., 1994. Interferon-gamma induces different subunit organizations and functional diversity of proteasomes. Journal of biochemistry, 115, pp.257–269.

Alves-Rodrigues, A., Gregori, L. & Figueiredo-Pereira, M.E., 1998. Ubiquitin, cellular inclusions and their role in neurodegeneration. Trends in Neurosciences, 21, pp.516–520.

Amano, T. et al., 2003. Synoviolin/Hrd1, an E3 ubiquitin ligase, as a novel pathogenic factor for arthropathy. Genes & development, 17, pp.2436–2449.

Anjos, S. et al., 2002. A common autoimmunity predisposing signal peptide variant of the cytotoxic T-lymphocyte antigen 4 results in inefficient glycosylation of the susceptibility allele. Journal of Biological Chemistry, 277, pp.46478–46486.

Apweiler, R., Hermjakob, H. & Sharon, N., 1999. On the frequency of protein glycosylation, as deduced from analysis of the SWISS-PROT database. Biochimica et Biophysica Acta - General Subjects, 1473, pp.4–8.

Arnold, A. et al., 1990. Mutation of the signal peptide-encoding region of the preproparathyroid hormone gene in familial isolated hypoparathyroidism. Journal of Clinical Investigation, 86, pp.1084–1087.

Ast, T., Cohen, G. & Schuldiner, M., 2013. A network of cytosolic factors targets SRP-independent proteins to the endoplasmic reticulum. Cell, 152, pp.1134–1145.

Ast, T. & Schuldiner, M., 2013. All roads lead to Rome (but some may be harder to travel): SRP-independent translocation into the endoplasmic reticulum. Critical reviews in biochemistry and molecular biology, 48, pp.273– 88. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23530742.

Avadhani, N.G. et al., 2011. Bimodal targeting of cytochrome P450s to endoplasmic reticulum and mitochondria: The concept of chimeric signals. FEBS Journal, 278, pp.4218–4229.

Babu, Y.S., Bugg, C.E. & Cook, W.J., 1988. Structure of calmodulin refined at 2.2 A resolution. Journal of molecular biology, 204, pp.191–204. 112

Bacik I, Snyder HL, Antón LC, Russ G, Chen W, Bennink JR, Urge L, Otvos L, Dudkowska B, Eisenlohr L, Y.J., 1997. Introduction of a Glycosylation Site into a Secreted Protein Provides Evidence for an Alternative Antigen Processing Pathway : Transport of Precursors of Major from the Endoplasmic Reticulum to the Cytosol. , 186(4), pp.479–487.

Ballif, B.A. et al., 2008. Large-scale identification and evolution indexing of tyrosine phosphorylation sites from murine brain. Journal of Proteome Research, 7, pp.311–318.

Barford, D., 2004. The role of cysteine residues as redox-sensitive regulatory switches. Current Opinion in Structural Biology, 14, pp.679–686.

Barford, D., Das, A.K. & Egloff, M.P., 1998. The structure and mechanism of protein phosphatases: insights into catalysis and regulation. Annual review of biophysics and biomolecular structure, 27, pp.133–164.

Barriere, H. et al., 2007. Plasticity of polyubiquitin recognition as lysosomal targeting signals by the endosomal sorting machinery. Molecular biology of the cell, 18, pp.3952–3965.

Barton, L.L. & Hyndman, N.J., 2000. Lymphocytic choriomeningitis virus: reemerging central nervous system pathogen.,

Basler, M. et al., 2006. An altered T cell repertoire in MECL-1-deficient mice. Journal of immunology (Baltimore, Md. : 1950), 176, pp.6665–6672.

Baumeister, W. et al., 1998. The proteasome: Paradigm of a self-compartmentalizing protease. Cell, 92, pp.367–380.

Bays, N.W. & Hampton, R.Y., 2002. Cdc48-Ufd1-Npl4: Stuck in the middle with Ub. Current Biology, 12.

Beausoleil, S.A. et al., 2004. Large-scale characterization of HeLa cell nuclear phosphoproteins. Proceedings of the National Academy of Sciences of the United States of America, 101, pp.12130–12135.

Belin, D. et al., 1996. A two-step recognition of signal sequences determines the translocation efficiency of proteins. The EMBO journal, 15, pp.468–478.

Bence, N.F., Sampat, R.M. & Kopito, R.R., 2001. Impairment of the ubiquitin-proteasome system by protein aggregation. Science (New York, N.Y.), 292, pp.1552–1555.

Benham, A.M., 2012. The Protein Disulfide Isomerase Family: Key Players in Health and Disease. Antioxidants & Redox Signaling, 16, pp.781–789.

Benveniste, E.N., 1998. Cytokine actions in the central nervous system. Cytokine and Growth Factor Reviews, 9, pp.259–275.

Van den Berg, B. et al., 2004. X-ray structure of a protein-conducting channel. Nature, 427, pp.36–44.

Berg, J.M., Tymoczko, J.L. & Stryer, L., 2010. Biochemistry, Available at: http://books.google.com/books?id=jQKGAAAACAAJ.

Bernasconi, R. et al., 2012. Role of the SEL1L:LC3-I Complex as an ERAD Tuning Receptor in the Mammalian ER. Molecular Cell, 46, pp.809–819.

Bernasconi, R. et al., 2010. Stringent requirement for HRD1, SEL1L, and OS-9/XTP3-B for disposal of ERAD-LS substrates. Journal of Cell Biology, 188, pp.223–235.

Besche, H.C. et al., 2014. Autoubiquitination of the 26S Proteasome on Rpn13 Regulates Breakdown of Ubiquitin Conjugates. EMBO Journal, 33, pp.1159–1176.

113

Besemer, J. et al., 2005. Selective inhibition of cotranslational translocation of vascular cell adhesion molecule 1. Nature, 436, pp.290–293.

Bevan, M.J., 2006. Cross-priming. Nature immunology, 7, pp.363–365.

Bi, M. et al., 2005. ER stress-regulated translation increases tolerance to extreme hypoxia and promotes tumor growth. The EMBO journal, 24, pp.3470–3481.

Bianchi, G. et al., 2009. The proteasome load versus capacity balance determines apoptotic sensitivity of multiple myeloma cells to proteasome inhibition. Blood, 113, pp.3040–3049.

Birkenbach, M. et al., 1993. Epstein-Barr virus-induced genes: first lymphocyte-specific G protein-coupled peptide receptors. Journal of virology, 67, pp.2209–2220.

Biswas, M. & Chan, J.Y., 2010. Role of Nrf1 in antioxidant response element-mediated gene expression and beyond. Toxicology and Applied Pharmacology, 244, pp.16–20.

Blobel, G. & Sabatini, D.D., 1971. Ribosome-Membrane Interaction in Eukaryotic Cells. Biomembranes, pp.193–195. Available at: http://www.springerlink.com/index/10.1007/978-1-4684-3330-2.

Bole-Feysot, C. et al., 1998. Prolactin (PRL) and its receptor: actions, signal transduction pathways and phenotypes observed in PRL receptor knockout mice. Endocrine reviews, 19, pp.225–268.

Brennan, A.M. & Mantzoros, C.S., 2006. Drug Insight: the role of leptin in human physiology and pathophysiology-- emerging clinical applications. Nature clinical practice. Endocrinology & metabolism, 2, pp.318–327.

Brodsky, J.L., 2012. Cleaning Up: ER-associated degradation to the rescue. Cell, 151, pp.1163–1167.

Brodsky, J.L., Goeckeler, J. & Schekman, R., 1995. BiP and Sec63p are required for both co- and posttranslational protein translocation into the yeast endoplasmic reticulum. Proceedings of the National Academy of Sciences of the United States of America, 92, pp.9643–9646.

Bull, V.H. & Thiede, B., 2012. Proteome analysis of tunicamycin-induced ER stress. Electrophoresis, 33, pp.1814– 1823.

Bulteau, A.L. et al., 2001. Oxidative Modification and Inactivation of the Proteasome during Coronary Occlusion/Reperfusion. Journal of Biological Chemistry, 276, pp.30057–30063.

Burns, K. et al., 1994. Modulation of gene expression by calreticulin binding to the glucocorticoid receptor. Nature, 367, pp.476–480.

Bush, T.G. et al., 1999. Leukocyte infiltration, neuronal degeneration, and neurite outgrowth after ablation of scar- forming, reactive astrocytes in adult transgenic mice. Neuron, 23, pp.297–308.

Bustamante, H.A. et al., 2013. Turnover of C99 is controlled by a crosstalk between ERAD and ubiquitin-independent lysosomal degradation in human neuroglioma cells. PLoS ONE, 8.

Butterfield, D.A. & Kanski, J., 2001. Brain protein oxidation in age-related neurodegenerative disorders that are associated with aggregated proteins. Mechanisms of Ageing and Development, 122, pp.945–962.

Byun, H. et al., 2014. ERAD and how viruses exploit it. Frontiers in Microbiology, 5.

Caramelo, J.J. et al., 2004. The endoplasmic reticulum glucosyltransferase recognizes nearly native glycoprotein folding intermediates. Journal of Biological Chemistry, 279, pp.46280–46285.

Carvalho, P., Stanley, A.M. & Rapoport, T.A., 2010. Retrotranslocation of a misfolded luminal ER protein by the ubiquitin-ligase hrd1p. Cell, 143, pp.579–591. 114

Cascio, P. et al., 2001. 26S proteasomes and immunoproteasomes produce mainly N-extended versions of an antigenic peptide. EMBO Journal, 20, pp.2357–2366.

Cenci, S. et al., 2006. Progressively impaired proteasomal capacity during terminal plasma cell differentiation. The EMBO journal, 25, pp.1104–1113.

Chadwick, D.R., 2005. Viral meningitis. British Medical Bulletin, 75-76, pp.1–14.

Chakrabarti, O., Rane, N.S. & Hegde, R.S., 2011. Cytosolic aggregates perturb the degradation of nontranslocated secretory and membrane proteins. Molecular biology of the cell, 22, pp.1625–1637.

Chance, B., Sies, H. & Boveris, A., 1979. Hydroperoxide metabolism in mammalian organs. Physiological reviews, 59, pp.527–605.

Charrel, R.N. & De Lamballerie, X., 2003. Arenaviruses other than Lassa virus. Antiviral Research, 57, pp.89–100.

Chen, D. et al., 2008. Multiple pathways differentially regulate global oxidative stress responses in fission yeast. Molecular biology of the cell, 19, pp.308–317.

Chen, P. & Hochstrasser, M., 1996. Autocatalytic subunit processing couples active site formation in the 20S proteasome to completion of assembly. Cell, 86, pp.961–972.

Chen, P. & Hochstrasser, M., 1995. Biogenesis, structure and function of the yeast 20S proteasome. The EMBO journal, 14, pp.2620–2630.

Chou, A.P. et al., 2010. Mechanisms of rotenone-induced proteasome inhibition. NeuroToxicology, 31, pp.367–372.

Chu-Ping, M., Slaughter, C.A. & DeMartino, G.N., 1992. Identification, purification, and characterization of a protein activator (PA28) of the 20 S proteasome (macropain). Journal of Biological Chemistry, 267, pp.10515–10523.

Chu-Ping, M., Slaughter, C.A. & DeMartino, G.N., 1992. Purification and characterization of a protein inhibitor of the 20S proteasome (macropain). Biochimica et biophysica acta, 1119, pp.303–311.

Coppolino, M. et al., 1995. Inducible Interaction of Integrin alpha2beta1 with calreticulin: Dependence on the activation state of the integrin. Journal of Biological Chemistry, 270, pp.23132–23138.

Coppolino, M.G. et al., 1997. Calreticulin is essential for integrin-mediated calcium signalling and cell adhesion. Nature, 386, pp.843–847.

Cullinan, S.B. et al., 2003. Nrf2 is a direct PERK substrate and effector of PERK-dependent cell survival. Molecular and cellular biology, 23, pp.7198–7209.

Dai, C. et al., 2007. Heat Shock Factor 1 Is a Powerful Multifaceted Modifier of Carcinogenesis. Cell, 130, pp.1005– 1018.

Dalbey, R.E. et al., 1997. The chemistry and enzymology of the type I signal peptidases. Protein science : a publication of the Protein Society, 6, pp.1129–1138.

Davies, K.J., Lin, S.W. & Pacifici, R.E., 1987. Protein damage and degradation by oxygen radicals. IV. Degradation of denatured protein. Journal of Biological Chemistry, 262, pp.9914–9920.

Davies, K.J.A., 2001. Degradation of oxidized proteins by the 20S proteasome. In Biochimie. pp. 301–310.

Davila, S. et al., 2004. Mutations in SEC63 cause autosomal dominant polycystic liver disease. Nature genetics, 36, pp.575–577.

115

Dedhar, S., 1994. Novel functions for calreticulin: Interaction with integrins and modulation of gene expression? Trends in Biochemical Sciences, 19, pp.269–271.

Deegan, S. et al., 2013. Stress-induced self-cannibalism: On the regulation of autophagy by endoplasmic reticulum stress. Cellular and Molecular Life Sciences, 70, pp.2425–2441.

Demasi, M., Shringarpure, R. & Davies, K.J., 2001. Glutathiolation of the proteasome is enhanced by proteolytic inhibitors. Archives of biochemistry and biophysics, 389, pp.254–263.

Deshaies, R.J. et al., 1991. Assembly of yeast Sec proteins involved in translocation into the endoplasmic reticulum into a membrane-bound multisubunit complex. Nature, 349, pp.806–808.

Deveraux, Q. et al., 1994. A 26 S protease subunit that binds ubiquitin conjugates. Journal of Biological Chemistry, 269, pp.7059–7061.

Ding, Q. et al., 2003. Role of the proteasome in protein oxidation and neural viability following low-level oxidative stress. FEBS Letters, 546, pp.228–232.

Ding, Q. & Keller, J.N., 2001. Proteasomes and proteasome inhibition in the central nervous system. Free Radical Biology and Medicine, 31, pp.574–584.

Drisaldi, B. et al., 2003. Mutant PrP is delayed in its exit from the endoplasmic reticulum, but neither wild-type nor mutant PrP undergoes retrotranslocation prior to proteasomal degradation. Journal of Biological Chemistry, 278, pp.21732–21743.

Egea, P.F. & Stroud, R.M., 2010. Lateral opening of a translocon upon entry of protein suggests the mechanism of insertion into membranes. Proceedings of the National Academy of Sciences of the United States of America, 107, pp.17182–17187.

Emonet, S. et al., 2006. Phylogeny and evolution of old world arenaviruses. Virology, 350, pp.251–257.

Ermakova, S.P. et al., 2006. (-)-Epigallocatechin gallate overcomes resistance to etoposide-induced cell death by targeting the molecular chaperone glucose-regulated protein 78. Cancer Research, 66, pp.9260–9269.

Ernst, R. et al., 2009. The Otubain YOD1 Is a Deubiquitinating Enzyme that Associates with p97 to Facilitate Protein Dislocation from the ER. Molecular Cell, 36, pp.28–38.

Erzberger, J.P. & Berger, J.M., 2006. Evolutionary relationships and structural mechanisms of AAA+ proteins. Annual review of biophysics and biomolecular structure, 35, pp.93–114.

Evans, E.A., Gilmore, R. & Blobel, G., 1986. Purification of microsomal signal peptidase as a complex. Proceedings of the National Academy of Sciences of the United States of America, 83, pp.581–585.

Fairbank, M., St-Pierre, P. & Nabi, I.R., 2009. The complex biology of autocrine motility factor/phosphoglucose isomerase (AMF/PGI) and its receptor, the gp78/AMFR E3 ubiquitin ligase. Molecular bioSystems, 5, pp.793– 801.

Falsig, J. et al., 2006. The inflammatory transcriptome of reactive murine astrocytes and implications for their innate immune function. Journal of Neurochemistry, 96, pp.893–907.

Farout, L. et al., 2006. Inactivation of the proteasome by 4-hydroxy-2-nonenal is site specific and dependant on 20S proteasome subtypes. Archives of biochemistry and biophysics, 453, pp.135–142.

Fehling, H.J. et al., 1994. MHC class I expression in mice lacking the proteasome subunit LMP-7. Science (New York, N.Y.), 265, pp.1234–1237.

116

Feldheim, D. et al., 1993. Structural and functional characterization of Sec66p, a new subunit of the polypeptide translocation apparatus in the yeast endoplasmic reticulum. Molecular biology of the cell, 4, pp.931–939.

Fenteany, G. et al., 1995. Inhibition of proteasome activities and subunit-specific amino-terminal threonine modification by lactacystin. Science (New York, N.Y.), 268, pp.726–731.

Ferris, R.L. et al., 1999. Processing of HIV-1 envelope glycoprotein for class I-restricted recognition: dependence on TAP1/2 and mechanisms for cytosolic localization. Journal of immunology (Baltimore, Md. : 1950), 162, pp.1324–1332.

Finkel, T., 2000. Redox-dependent signal transduction. FEBS Letters, 476, pp.52–54.

Finkel, T. & Holbrook, N.J., 2000. Oxidants, oxidative stress and the biology of ageing. Nature, 408, pp.239–247.

Fischer, S.A. et al., 2006. Transmission of lymphocytic choriomeningitis virus by organ transplantation.,

Fleig, L. et al., 2012. Ubiquitin-Dependent Intramembrane Rhomboid Protease Promotes ERAD of Membrane Proteins. Molecular Cell, 47, pp.558–569.

Flemming, K.D., 2006. Essential Neuroscience. Mayo Clinic Proceedings, 81, p.1409.

Fons, R.D., Bogert, B.A. & Hegde, R.S., 2003. Substrate-specific function of the translocon-associated protein complex during translocation across the ER membrane. Journal of Cell Biology, 160, pp.529–539.

Frauenfeld, J. et al., 2011. Cryo-EM structure of the ribosome-SecYE complex in the membrane environment. Nature structural & molecular biology, 18, pp.614–621.

Freeman, M.R., 2010. Specification and morphogenesis of astrocytes. Science (New York, N.Y.), 330, pp.774–778.

Fricke, B. et al., 2007. The proteasome maturation protein POMP facilitates major steps of 20S proteasome formation at the endoplasmic reticulum. EMBO reports, 8, pp.1170–1175.

Friedlander, R. et al., 2000. A regulatory link between ER-associated protein degradation and the unfolded-protein response. Nature cell biology, 2, pp.379–384.

Fruhwirth, G.O., Loidl, A. & Hermetter, A., 2007. Oxidized phospholipids: From molecular properties to disease. Biochimica et Biophysica Acta - Molecular Basis of Disease, 1772, pp.718–736.

Garcia, P.D. et al., 1988. Targeting of the hepatitis B virus precore protein to the endorplasmic reticulum membrane: After signal peptide cleavage translocation can be aborted and the product released into the cytoplasm. Journal of Cell Biology, 106, pp.1093–1104.

Garrison, J.L. et al., 2005. A substrate-specific inhibitor of protein translocation into the endoplasmic reticulum. Nature, 436, pp.285–289.

Gettins, P.G.W., 2002. Serpin structure, mechanism, and function. Chemical Reviews, 102, pp.4751–4803.

Gilmore, R., Walter, P. & Blobel, G., 1982. Protein translocation across the endoplasmic reticulum. II. Isolation and characterization of the signal recognition particle receptor. The Journal of cell biology, 95, pp.470–477.

Glickman, M.H. et al., 1998. A subcomplex of the proteasome regulatory particle required for ubiquitin-conjugate degradation and related to the COP9-signalosome and elF3. Cell, 94, pp.615–623.

Glickman, M.H. & Ciechanover, A., 2002. The ubiquitin-proteasome proteolytic pathway: destruction for the sake of construction. Physiological reviews, 82, pp.373–428.

117

Goh, K.C. et al., 1996. Identification of p90, a prominent tyrosine-phosphorylated protein in fibroblast growth factor- stimulated cells, as 80K-H. Journal of Biological Chemistry, 271, pp.5832–5838.

Gold, R. et al., 2012. Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis. The New England journal of medicine, 367, pp.1098–107. Available at: http://www.ncbi.nlm.nih.gov/pubmed/22992073.

Gomes, A. V, 2013. Genetics of Proteasome Diseases. Scientifica, 2013, p.637629. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3892944&tool=pmcentrez&rendertype=abstract.

Gorlich, D. et al., 1992. A mammalian homolog of SEC61p and SECYp is associated with ribosomes and nascent polypeptides during translocation. Cell, 71, pp.489–503.

Görlich, D. et al., 1992. A protein of the endoplasmic reticulum involved early in polypeptide translocation. Nature, 357, pp.47–52.

Goulet, B. et al., 2004. A cathepsin L isoform that is devoid of a signal peptide localizes to the nucleus in S phase and processes the CDP/Cux transcription factor. Molecular Cell, 14, pp.207–219.

Gouridis, G. et al., 2009. Signal peptides are allosteric activators of the protein translocase. Nature, 462, pp.363–367.

Green, N., Fang, H. & Walter, P., 1992. Mutants in three novel complementation groups inhibit membrane protein insertion into and soluble protein translocation across the endoplasmic reticulum membrane of Saccharomyces cerevisiae. Journal of Cell Biology, 116, pp.597–604.

Greenblatt, E.J., Olzmann, J.A. & Kopito, R.R., 2011. Derlin-1 is a rhomboid pseudoprotease required for the dislocation of mutant α-1 antitrypsin from the endoplasmic reticulum. Nature Structural & Molecular Biology, 18, pp.1147–1152.

Groettrup, M. et al., 1996. A third interferon-gamma-induced subunit exchange in the 20S proteasome. European Journal of Immunology, 26, pp.863–869.

Groll, M. et al., 2000. A gated channel into the proteasome core particle. Nature structural biology, 7, pp.1062–1067.

Groll, M. et al., 2000. Crystal structure of epoxomicin:20S proteasome reveals a molecular basis for selectivity of alpha’,beta'-epoxyketone proteasome inhibitors. Journal of the American Chemical Society, 122, pp.1237–1238.

Groll, M. et al., 2006. Crystal structure of the boronic acid-based proteasome inhibitor bortezomib in complex with the yeast 20S proteasome. Structure, 14, pp.451–456.

Grune, T. et al., 2004. Decreased proteolysis caused by protein aggregates, inclusion bodies, plaques, lipofuscin, ceroid, and “aggresomes” during oxidative stress, aging, and disease. International Journal of Biochemistry and Cell Biology, 36, pp.2519–2530.

Guerriero, C.J. & Brodsky, J.L., 2012. The delicate balance between secreted protein folding and endoplasmic reticulum-associated degradation in human physiology. Physiological reviews, 92, pp.537–76. Available at: http://www.ncbi.nlm.nih.gov/pubmed/22535891.

Guillaume, B. et al., 2010. Two abundant proteasome subtypes that uniquely process some antigens presented by HLA class I molecules. Proceedings of the National Academy of Sciences of the United States of America, 107, pp.18599–18604.

Güssow, D. et al., 1987. The human beta 2-microglobulin gene. Primary structure and definition of the transcriptional unit. Journal of immunology (Baltimore, Md. : 1950), 139, pp.3132–3138.

Haas, A.L. et al., 1982. Ubiquitin-activating enzyme. Mechanism and role in protein-ubiquitin conjugation. Journal of Biological Chemistry, 257, pp.2543–2548.

118

Hammond, S.A. et al., 1995. An epitope-selective, transporter associated with antigen presentation (TAP)-1/2- independent pathway and a more general TAP-1/2-dependent antigen-processing pathway allow recognition of the HIV-1 envelope glycoprotein by CD8+ CTL. Journal of immunology (Baltimore, Md. : 1950), 154, pp.6140– 6156.

Hampton, R.Y. & Sommer, T., 2012. Finding the will and the way of ERAD substrate retrotranslocation. Current Opinion in Cell Biology, 24, pp.460–466.

Handley, P.M. et al., 1991. Molecular cloning, sequence, and tissue distribution of the human ubiquitin-activating enzyme E1. Proceedings of the National Academy of Sciences of the United States of America, 88, pp.258– 262.

Hansford, R.G., Hogue, B.A. & Mildaziene, V., 1997. Dependence of H2O2 formation by rat heart mitochondria on substrate availability and donor age. Journal of Bioenergetics and Biomembranes, 29, pp.89–95.

Harding, H.P. et al., 2000. Regulated translation initiation controls stress-induced gene expression in mammalian cells. Molecular cell, 6, pp.1099–1108.

Hartmann, E. et al., 1993. A tetrameric complex of membrane proteins in the endoplasmic reticulum. European journal of biochemistry / FEBS, 214, pp.375–381. Available at: http://www.ncbi.nlm.nih.gov/pubmed/7916687.

Hartmann, E., Rapoport, T.A. & Lodish, H.F., 1989. Predicting the orientation of eukaryotic membrane-spanning proteins. Proceedings of the National Academy of Sciences of the United States of America, 86, pp.5786–5790.

Haßdenteufel, S. et al., 2011. Ca2+-calmodulin inhibits tail-anchored protein insertion into the mammalian endoplasmic reticulum membrane. FEBS Letters, 585, pp.3485–3490.

Haydon, P.G., 2000. Neuroglial networks: Neurons and glia talk to each other. Current Biology, 10.

Hebert, D.N., Bernasconi, R. & Molinari, M., 2010. ERAD substrates: Which way out? Seminars in Cell and Developmental Biology, 21, pp.526–532.

Hebert, D.N. & Molinari, M., 2007. In and out of the ER: protein folding, quality control, degradation, and related human diseases. Physiological reviews, 87, pp.1377–1408.

Hegner, M. et al., 1992. Single amino acid substitutions can convert the uncleaved signal-anchor of sucrase- isomaltase to a cleaved signal sequence. Journal of Biological Chemistry, 267, pp.16928–16933.

Heijne, G., 1986. The distribution of positively charged residues in bacterial inner membrane proteins correlates with the trans-membrane topology. The EMBO journal, 5, pp.3021–3027.

Von Heijne, G., 2006. Membrane-protein topology. Nature reviews. Molecular cell biology, 7, pp.909–918.

Von Heijne, G., 1986. Net N-C charge imbalance may be important for signal sequence function in bacteria. Journal of molecular biology, 192, pp.287–290.

Von Heijne, G., 1985. Signal sequences. The limits of variation. Journal of molecular biology, 184, pp.99–105.

Von Heijne, G., 1990. The signal peptide. The Journal of membrane biology, 115, pp.195–201.

Heine, U.I. et al., 1991. Localization of transforming growth factor-beta 1 in mitochondria of murine heart and liver. Cell regulation, 2, pp.467–477.

Heinemeyer, W. et al., 1997. The active sites of the eukaryotic 20 S proteasome and their involvement in subunit precursor processing. Journal of Biological Chemistry, 272, pp.25200–25209.

119

Heinrich, S.U. et al., 2000. The Sec61p complex mediates the integration of a membrane protein by allowing lipid partitioning of the transmembrane domain. Cell, 102, pp.233–244.

Hershko, A. & Ciechanover, A., 1998. The ubiquitin system. Annual review of biochemistry, 67, pp.425–479.

Hessa, T. et al., 2011. Protein targeting and degradation are coupled for elimination of mislocalized proteins. Nature, 475, pp.394–397.

Hessa, T. et al., 2005. Recognition of transmembrane helices by the endoplasmic reticulum translocon. Nature, 433, pp.377–381.

Hikita, C. & Mizushima, S., 1992. Effects of total hydrophobicity and length of the hydrophobic domain of a signal peptide on in vitro translocation efficiency. Journal of Biological Chemistry, 267, pp.4882–4888.

Hirano, Y. et al., 2005. A heterodimeric complex that promotes the assembly of mammalian 20S proteasomes. Nature, 437, pp.1381–1385.

Hirsch, C., Blom, D. & Ploegh, H.L., 2003. A role for N-glycanase in the cytosolic turnover of glycoproteins. EMBO Journal, 22, pp.1036–1046.

Hofman, M.A. & Swaab, D.F., 1992. The human hypothalamus: comparative morphometry and photoperiodic influences. Progress in brain research, 93, pp.133–147; discussion 148–149.

Holaska, J.M. et al., 2001. Calreticulin is a receptor for nuclear export. J Cell Biol, 152, pp.127–140. Available at: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=11149926.

Holtkamp, W. et al., 2012. Dynamic switch of the signal recognition particle from scanning to targeting. Nature structural & molecular biology, 19, pp.1332–7. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23142984.

Horst, M. & Kronidou, N.G., 1995. Protein import into mitochondria. Membrane Protein Transport, 1, pp.109–143.

Hoseki, J., Ushioda, R. & Nagata, K., 2010. Mechanism and components of endoplasmic reticulum-associated degradation. Journal of Biochemistry, 147, pp.19–25.

Hosokawa, N. et al., 2008. Human XTP3-B forms an endoplasmic reticulum quality control scaffold with the HRD1- SEL1L ubiquitin ligase complex and BiP. Journal of Biological Chemistry, 283, pp.20914–20924.

Huang, Y. et al., 2001. Apolipoprotein E fragments present in Alzheimer’s disease brains induce neurofibrillary tangle- like intracellular inclusions in neurons. Proceedings of the National Academy of Sciences of the United States of America, 98, pp.8838–8843.

Hureau, C. & Faller, P., 2009. Aβ-mediated ROS production by Cu ions: Structural insights, mechanisms and relevance to Alzheimer’s disease. Biochimie, 91, pp.1212–1217.

Husnjak, K. et al., 2008. Proteasome subunit Rpn13 is a novel ubiquitin receptor. Nature, 453, pp.481–488.

Hussain, S. et al., 2013. Permanent neonatal diabetes due to a novel insulin signal peptide mutation. Pediatric Diabetes, 14, pp.299–303.

Hussong, S.A. et al., 2010. Immunoproteasome deficiency alters retinal proteasome’s response to stress. Journal of Neurochemistry, 113, pp.1481–1490.

Iakova, P. et al., 2004. Competition of CUGBP1 and calreticulin for the regulation of p21 translation determines cell fate. The EMBO journal, 23, pp.406–417.

Ishii, T. et al., 2005. Oxidative modification of proteasome: Identification of an oxidation-sensitive subunit in 26 S proteasome. Biochemistry, 44, pp.13893–13901. 120

Jackson, P.K. et al., 2000. The lore of the RINGs: Substrate recognition and catalysis by ubiquitin ligases. Trends in Cell Biology, 10, pp.429–439.

Jentsch, S., 1992. The ubiquitin-conjugation system. Annual review of genetics, 26, pp.179–207. Available at: http://www.ncbi.nlm.nih.gov/pubmed/1336336.

Jentsch, S. & Rumpf, S., 2007. Cdc48 (p97): a “molecular gearbox” in the ubiquitin pathway? Trends in Biochemical Sciences, 32, pp.6–11.

Jimenez-Sanchez, M. et al., 2012. Autophagy and polyglutamine diseases. Progress in Neurobiology, 97, pp.67–82.

Johnson, A.E. & van Waes, M.A., 1999. The translocon: a dynamic gateway at the ER membrane. Annual review of cell and developmental biology, 15, pp.799–842.

Johnson, N. et al., 2012. TRC40 can deliver short secretory proteins to the Sec61 translocon. Journal of Cell Science, 125, pp.3612–3620.

Jones, E.G., 2007. The Thalamus,

Jung, C.H. et al., 2010. MTOR regulation of autophagy. FEBS Letters, 584, pp.1287–1295.

Jung, T. & Grune, T., 2008. The proteasome and its role in the degradation of oxidized proteins. IUBMB Life, 60, pp.743–752.

Van Kaer, L. et al., 1994. Altered peptidase and viral-specific T cell response in LMP2 mutant mice. Immunity, 1, pp.533–541.

Kalies, K.U. & Hartmann, E., 1996. Membrane topology of the 12- and the 25-kDa subunits of the mammalian signal peptidase complex. Journal of Biological Chemistry, 271, pp.3925–3929.

Kalies, K.U., Rapoport, T.A. & Hartmann, E., 1998. The beta subunit of the Sec61 complex facilitates cotranslational protein transport and interacts with the signal peptidase during translocation. Journal of Cell Biology, 141, pp.887–894.

Kanai, M. et al., 1997. Signal transduction pathway of human fibroblast growth factor receptor 3. Identification of a novel 66-kDa phosphoprotein. The Journal of biological chemistry, 272, pp.6621–6628.

Kandel, E.R., Schwartz, J.H. & Jessell, T.M., 2000. Principles of Neural Science,

Kaneko, M., Niinuma, Y. & Nomura, Y., 2003. Activation signal of nuclear factor-kappa B in response to endoplasmic reticulum stress is transduced via IRE1 and tumor necrosis factor receptor-associated factor 2. Biological & pharmaceutical bulletin, 26, pp.931–935.

Kang, S.S. & McGavern, D.B., 2008. Lymphocytic choriomeningitis infection of the central nervous system. Front Biosci, 13, pp.4529–4543.

Kang, S.W. et al., 2006. Substrate-Specific Translocational Attenuation during ER Stress Defines a Pre-Emptive Quality Control Pathway. Cell, 127, pp.999–1013.

Kansanen, E., Jyrkkänen, H.K. & Levonen, A.L., 2012. Activation of stress signaling pathways by electrophilic oxidized and nitrated lipids. Free Radical Biology and Medicine, 52, pp.973–982.

Kaufman, R.J. et al., 2002. The unfolded protein response in nutrient sensing and differentiation. Nature reviews. Molecular cell biology, 3, pp.411–421.

Kawahara, H., Minami, R. & Yokota, N., 2013. BAG6/BAT3: Emerging roles in quality control for nascent polypeptides. Journal of Biochemistry, 153, pp.147–160. 121

Keller, J.N., Gee, J. & Ding, Q., 2002. The proteasome in brain aging. Ageing Research Reviews, 1, pp.279–293.

Keller, J.N., Huang, F.F. & Markesbery, W.R., 2000. Decreased levels of proteasome activity and proteasome expression in aging spinal cord. Neuroscience, 98, pp.149–156.

Keller, J.N. & Mattson, M.P., 1998. Roles of lipid peroxidation in modulation of cellular signaling pathways, cell dysfunction, and death in the nervous system. Reviews in the neurosciences, 9, pp.105–116.

Kettern, N. et al., 2010. Chaperone-assisted degradation: Multiple paths to destruction. Biological Chemistry, 391, pp.481–489.

Khan, S. et al., 2001. Immunoproteasomes largely replace constitutive proteasomes during an antiviral and antibacterial immune response in the liver. Journal of immunology (Baltimore, Md. : 1950), 167, pp.6859–6868.

El Khouri, E. et al., 2013. RNF185 is a novel E3 ligase of Endoplasmic Reticulum Associated Degradation (ERAD) that targets Cystic Fibrosis Transmembrane conductance Regulator (CFTR). The Journal of biological chemistry. Available at: http://www.ncbi.nlm.nih.gov/pubmed/24019521.

Kim, S.J. & Hegde, R.S., 2002. Cotranslational partitioning of nascent prion protein into multiple populations at the translocation channel. Molecular biology of the cell, 13, pp.3775–3786.

Kim, S.J., Rahbar, R. & Hegde, R.S., 2001. Combinatorial Control of Prion Protein Biogenesis by the Signal Sequence and Transmembrane Domain. Journal of Biological Chemistry, 276, pp.26132–26140.

Kisselev, A.F., Van Der Linden, W.A. & Overkleeft, H.S., 2012. Proteasome inhibitors: An expanding army attacking a unique target. Chemistry and Biology, 19, pp.99–115.

Klonne, D.R. & Johnson, D.R., 1988. Enzyme activity and sulfhydryl status in rat renal cortex following mercuric chloride and dithiothreitol administration. Toxicology letters, 42, pp.199–205.

Kobayashi, K.S. & van den Elsen, P.J., 2012. NLRC5: a key regulator of MHC class I-dependent immune responses. Nature reviews. Immunology, 12, pp.813–20. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23175229.

Komander, D. et al., 2009. Molecular discrimination of structurally equivalent Lys 63-linked and linear polyubiquitin chains. EMBO reports, 10, pp.466–473.

Korennykh, A. V et al., 2009. The unfolded protein response signals through high-order assembly of Ire1. Nature, 457, pp.687–693.

Koulov, A. V et al., 2010. Biological and structural basis for Aha1 regulation of Hsp90 ATPase activity in maintaining proteostasis in the human disease cystic fibrosis. Molecular biology of the cell, 21, pp.871–884.

Kozak, M., 2002. Pushing the limits of the scanning mechanism for initiation of translation. Gene, 299, pp.1–34.

Kozak, M., 1992. Regulation of translation in eucaryotic systems. Annual Review of Cell Biology, 8, pp.197–225.

Kreft, S.G. & Hochstrasser, M., 2011. An unusual transmembrane helix in the endoplasmic reticulum ubiquitin ligase Doa10 modulates degradation of its cognate E2 enzyme. Journal of Biological Chemistry, 286, pp.20163– 20174.

Kreidenweiss, A., Kremsner, P.G. & Mordmüller, B., 2008. Comprehensive study of proteasome inhibitors against Plasmodium falciparum laboratory strains and field isolates from Gabon. Malaria journal, 7, p.187.

Kremer, M. et al., 2010. Reduced immunoproteasome formation and accumulation of immunoproteasomal precursors in the brains of lymphocytic choriomeningitis virus-infected mice. Journal of immunology (Baltimore, Md. : 1950), 185, pp.5549–5560.

122

Kretsinger, R.H., Rudnick, S.E. & Weissman, L.J., 1986. Crystal structure of calmodulin. Journal of inorganic biochemistry, 28(2-3), pp.289–302. Available at: http://www.ncbi.nlm.nih.gov/pubmed/3806094.

Kunz, S. et al., 2003. Mechanisms for lymphocytic choriomeningitis virus glycoprotein cleavage, transport, and incorporation into virions. Virology, 314, pp.168–178.

Kurys, G. et al., 2000. The long signal peptide isoform and its alternative processing direct the intracellular trafficking of interleukin-15. Journal of Biological Chemistry, 275, pp.30653–30659.

Kutay, U. et al., 1995. Transport route for synaptobrevin via a novel pathway of insertion into the endoplasmic reticulum membrane. The EMBO journal, 14, pp.217–223.

Lander, E. et al., 2001. Initial sequencing and analysis of the human genome. Nature, 409(February). Available at: http://www.nature.com/nature/journal/v409/n6822/abs/409860a0.html [Accessed September 2, 2014].

Lander, G.C. et al., 2012. Complete subunit architecture of the proteasome regulatory particle. Nature.

Lansbury, P.T. & Lashuel, H.A., 2006. A century-old debate on protein aggregation and neurodegeneration enters the clinic. Nature, 443, pp.774–779.

Lasker, K. et al., 2012. Inaugural Article: Molecular architecture of the 26S proteasome holocomplex determined by an integrative approach. Proceedings of the National Academy of Sciences, 109, pp.1380–1387.

Leach, J.K. et al., 2001. Ionizing radiation-induced, mitochondria-dependent generation of reactive oxygen/nitrogen. Cancer Research, 61, pp.3894–3901.

Lederkremer, G.Z., 2009. Glycoprotein folding, quality control and ER-associated degradation. Current Opinion in Structural Biology, 19, pp.515–523.

Lemberg, M.K. et al., 2001. Intramembrane proteolysis of signal peptides: an essential step in the generation of HLA- E epitopes. Journal of immunology (Baltimore, Md. : 1950), 167, pp.6441–6446.

Leutner, S., Eckert, A. & Müller, W.E., 2001. ROS generation, lipid peroxidation and antioxidant enzyme activities in the aging brain. Journal of neural transmission (Vienna, Austria : 1996), 108, pp.955–967.

Levine, C.G. et al., 2005. The efficiency of protein compartmentalization into the secretory pathway. Molecular biology of the cell, 16, pp.279–291.

Leznicki, P. et al., 2010. Bat3 promotes the membrane integration of tail-anchored proteins. Journal of cell science, 123, pp.2170–2178.

Li, G. et al., 2006. The AAA ATPase p97 links peptide N-glycanase to the endoplasmic reticulum-associated E3 ligase autocrine motility factor receptor. Proceedings of the National Academy of Sciences of the United States of America, 103, pp.8348–8353.

Lindsten, K. et al., 2002. Mutant ubiquitin found in neurodegenerative disorders is a ubiquitin fusion degradation substrate that blocks proteasomal degradation. The Journal of cell biology, 157, pp.417–427.

Lipson, C. et al., 2008. A proteasomal ATPase contributes to dislocation of Endoplasmic Reticulum-associated Degradation (ERAD) substrates. Journal of Biological Chemistry, 283, pp.7166–7175.

Little, M.H. et al., 2001. Dual trafficking of Slit3 to mitochondria and cell surface demonstrates novel localization for Slit protein. American journal of physiology. Cell physiology, 281, pp.C486–C495.

Liu, C.W. et al., 2006. ATP Binding and ATP Hydrolysis Play Distinct Roles in the Function of 26S Proteasome. Molecular Cell, 24, pp.39–50.

123

Lodish, H. et al., 2008. Molecular Cell Biology,

Loureiro, J. et al., 2006. Signal peptide peptidase is required for dislocation from the endoplasmic reticulum. Nature, 441, pp.894–897.

Lustbader, J.W. et al., 2004. ABAD directly links Abeta to mitochondrial toxicity in Alzheimer’s disease. Science (New York, N.Y.), 304, pp.448–452.

Ma, J. & Lindquist, S., 2002. Conversion of PrP to a self-perpetuating PrPSc-like conformation in the cytosol. Science (New York, N.Y.), 298, pp.1785–1788.

Ma, J. & Lindquist, S., 2001. Wild-type PrP and a mutant associated with prion disease are subject to retrograde transport and proteasome degradation. Proceedings of the National Academy of Sciences of the United States of America, 98, pp.14955–14960.

Ma, J., Wollmann, R. & Lindquist, S., 2002. Neurotoxicity and neurodegeneration when PrP accumulates in the cytosol. Science (New York, N.Y.), 298, pp.1781–1785.

Määttänen, P. et al., 2010. Protein quality control in the ER: The recognition of misfolded proteins. Seminars in Cell and Developmental Biology, 21, pp.500–511.

Mani, A. & Gelmann, E.P., 2005. The ubiquitin-proteasome pathway and its role in cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology, 23, pp.4776–4789.

Mann, M. & Jensen, O.N., 2003. Proteomic analysis of post-translational modifications. Nature biotechnology, 21, pp.255–261.

Maragos, C.M. et al., 1991. Complexes of .NO with nucleophiles as agents for the controlled biological release of nitric oxide. Vasorelaxant effects. Journal of medicinal chemistry, 34, pp.3242–3247.

Mariappan, M. et al., 2010. A ribosome-associating factor chaperones tail-anchored membrane proteins. Nature, 466, pp.1120–1124.

Martindale, J.L. & Holbrook, N.J., 2002. Cellular response to oxidative stress: Signaling for suicide and survival. Journal of Cellular Physiology, 192, pp.1–15.

Matlack, K.E.S. et al., 1999. BIP acts as a molecular ratchet during posttranslational transport of prepro-α factor across the ER membrane [2]. Cell, 97, pp.553–564.

Matouschek, A. & Finley, D., 2012. An Ancient Portal to Proteolysis. Science, 337, pp.813–814.

Matsumura, Y., Sakai, J. & Skach, W.R., 2013. Endoplasmic reticulum protein quality control is determined by cooperative interactions between Hsp/c70 protein and the CHIP E3 ligase. Journal of Biological Chemistry, 288, pp.31069–31079.

Mayya, V. et al., 2009. Quantitative phosphoproteomic analysis of T cell receptor signaling reveals system-wide modulation of protein-protein interactions. Science signaling, 2, p.ra46.

McGavern, D.B., Homann, D. & Oldstone, M.B.A., 2002. T cells in the central nervous system: the delicate balance between viral clearance and disease. The Journal of infectious diseases, 186 Suppl , pp.S145–S151.

Meacock, S.L. et al., 2002. Different Transmembrane Domains Associate with Distinct Endoplasmic Reticulum Components during Membrane Integration of a Polytopic Protein. , 13(December), pp.4114–4129.

Ménétret, J.F. et al., 2008. Single Copies of Sec61 and TRAP Associate with a Nontranslating Mammalian Ribosome. Structure, 16, pp.1126–1137.

124

Merulla, J. et al., 2013. Specificity and Regulation of the Endoplasmic Reticulum-Associated Degradation Machinery. Traffic, 14, pp.767–777.

Meyer, D.I., Krause, E. & Dobberstein, B., 1982. Secretory protein translocation across membranes - the role of the “docking protein.” Nature, 297, pp.647–650.

Meyer, H., 2012. p97 complexes as signal integration hubs. BMC Biology, 10, p.48.

Meyer, H., Bug, M. & Bremer, S., 2012. Emerging functions of the VCP/p97 AAA-ATPase in the ubiquitin system. Nature Cell Biology, 14, pp.117–123.

Milstein C, Brownlee GG, Harrison TM, M.M., 1972. A possible precurser of immunoglobulin light chains. Nat New Biol., 239, pp.117–20.

Minami, R. et al., 2010. BAG-6 is essential for selective elimination of defective proteasomal substrates. Journal of Cell Biology, 190, pp.637–650.

Mishto, M., Santoro, A., et al., 2006. A structural model of 20S immunoproteasomes: Effect of LMP2 codon 60 polymorphism on expression, activity, intracellular localisation and insight into the regulatory mechanisms. Biological Chemistry, 387, pp.417–429.

Mishto, M., Bellavista, E., et al., 2006. Immunoproteasome and LMP2 polymorphism in aged and Alzheimer’s disease brains. Neurobiology of Aging, 27, pp.54–66.

Mitchell, W.B. et al., 2006. αIIbβ3 biogenesis is controlled by engagement of αIIb in the calnexin cycle via the N15- linked glycan. Blood, 107, pp.2713–2719.

Molineaux, S.M., 2012. Molecular pathways: Targeting proteasomal protein degradation in cancer. Clinical Cancer Research, 18, pp.15–20.

Monaco, J.J. & Nandi, D., 1995. The genetics of proteasomes and antigen processing. Annual review of genetics, 29, pp.729–754.

Morimoto, R.I., 2008. Proteotoxic stress and inducible chaperone networks in neurodegenerative disease and aging. Genes and Development, 22, pp.1427–1438.

Mosse, C.A. et al., 1998. The class I antigen-processing pathway for the membrane protein tyrosinase involves translation in the endoplasmic reticulum and processing in the cytosol. The Journal of experimental medicine, 187, pp.37–48.

Munro, S. & Pelham, H.R., 1987. A C-terminal signal prevents secretion of luminal ER proteins. Cell, 48, pp.899–907.

Murphy, K.M., Travers, P. & Walport, M., 2009. Janeway Immunologie,

Nandi, D., Jiang, H. & Monaco, J.J., 1996. Identification of MECL-1 (LMP-10) as the third IFN-gamma-inducible proteasome subunit. Journal of immunology (Baltimore, Md. : 1950), 156, pp.2361–2364.

Ng, D.T.W., Brown, J.D. & Walter, P., 1996. Signal sequences specify the targeting route to the endoplasmic reticulum membrane. Journal of Cell Biology, 134, pp.269–278.

Ngosuwan, J. et al., 2003. Roles of cytosolic Hsp70 and Hsp40 molecular chaperones in post-translational translocation of presecretory proteins into the endoplasmic reticulum. Journal of Biological Chemistry, 278, pp.7034–7042.

Nowis, D., McConnell, E. & Wójcik, C., 2006. Destabilization of the VCP-Ufd1-Npl4 complex is associated with decreased levels of ERAD substrates. Experimental Cell Research, 312, pp.2921–2932.

125

Obeng, E.A. et al., 2006. Proteasome inhibitors induce a terminal unfolded protein response in multiple myeloma cells. Blood, 107, pp.4907–4916.

Okuyama, T., Batanian, J.R. & Sly, W.S., 1993. Genomic organization and localization of gene for human carbonic anhydrase IV to chromosome 17q. Genomics, 16, pp.678–684.

Oliver, J. et al., 1995. The Sec61 complex is essential for the insertion of proteins into the membrane of the endoplasmic reticulum. FEBS letters, 362, pp.126–130.

Olzmann, J.A., Kopito, R.R. & Christianson, J.C., 2013. The mammalian endoplasmic reticulum-associated degradation system. Cold Spring Harbor Perspectives in Biology, 5.

Orlowski, M. & Wilk, S., 2000. Catalytic activities of the 20 S proteasome, a multicatalytic proteinase complex. Archives of biochemistry and biophysics, 383, pp.1–16.

Orsi, A. et al., 2006. Conditions of endoplasmic reticulum stress favor the accumulation of cytosolic prion protein. Journal of Biological Chemistry, 281, pp.30431–30438.

Ortega, J. et al., 2005. The axial channel of the 20 S proteasome opens upon binding of the PA200 activator. Journal of Molecular Biology, 346, pp.1221–1227.

Ou, J.H., Yeh, C.T. & Yen, T.S., 1989. Transport of hepatitis B virus precore protein into the nucleus after cleavage of its signal peptide. Journal of virology, 63, pp.5238–5243.

Panzner, S. et al., 1995. Posttranslational protein transport in yeast reconstituted with a purified complex of Sec proteins and Kar2p. Cell, 81, pp.561–570.

Park, E. et al., 2014. Structure of the SecY channel during initiation of protein translocation. Nature, 506, pp.102–6. Available at: http://www.ncbi.nlm.nih.gov/pubmed/24153188.

Pechmann, S., Willmund, F. & Frydman, J., 2013. The Ribosome as a Hub for Protein Quality Control. Molecular Cell, 49, pp.411–421.

Pestka, S., 1971. Inhibitors of ribosome functions. Annual review of microbiology, 25, pp.487–562.

Peters, J.M., Franke, W.W. & Kleinschmidt, J.A., 1994. Distinct 19 S and 20 S subcomplexes of the 26 S proteasome and their distribution in the nucleus and the cytoplasm. Journal of Biological Chemistry, 269, pp.7709–7718.

Piccinini, M. et al., 2003. Interferon-gamma-inducible subunits are incorporated in human brain 20S proteasome. Journal of Neuroimmunology, 135, pp.135–140.

Pickart, C.M. & Fushman, D., 2004. Polyubiquitin chains: Polymeric protein signals. Current Opinion in Chemical Biology, 8, pp.610–616.

Pickering, A.M. et al., 2012. Nrf2-dependent induction of proteasome and Pa28αβ regulator are required for adaptation to oxidative stress. Journal of Biological Chemistry, 287, pp.10021–10031.

Pickering, A.M. et al., 2010. The immunoproteasome, the 20S proteasome and the PA28αβ proteasome regulator are oxidative-stress-adaptive proteolytic complexes. The Biochemical journal, 432, pp.585–594.

Plath, K. & Rapoport, T.A., 2000. Spontaneous release of cytosolic proteins from posttranslational substrates before their transport into the endoplasmic reticulum. Journal of Cell Biology, 151, pp.167–178.

Plemper, R.K. et al., 1999. Genetic interactions of Hrd3p and Der3p/Hrd1p with Sec61p suggest a retro-translocation complex mediating protein transport for ER degradation. Journal of cell science, 112 ( Pt 2, pp.4123–4134.

126

Pool, M.R. et al., 2002. Distinct modes of signal recognition particle interaction with the ribosome. Science (New York, N.Y.), 297, pp.1345–1348.

Powis, K. et al., 2013. Get3 is a holdase chaperone and moves to deposition sites for aggregated proteins when membrane targeting is blocked. Journal of cell science, 126, pp.473–83. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3613179&tool=pmcentrez&rendertype=abstract.

Puelles, L., 2001. Thoughts on the development, structure and evolution of the mammalian and avian telencephalic pallium. Philosophical transactions of the Royal Society of London. Series B, Biological sciences, 356, pp.1583– 1598.

Pye, V.E. et al., 2007. Structural insights into the p97-Ufd1-Npl4 complex. Proceedings of the National Academy of Sciences of the United States of America, 104, pp.467–472.

Rabl, J. et al., 2008. Mechanism of Gate Opening in the 20S Proteasome by the Proteasomal ATPases. Molecular Cell, 30, pp.360–368.

Raden, D., Song, W. & Gilmore, R., 2000. Role of the cytoplasmic segments of Sec61?? in the ribosome-binding and translocation-promoting activities of the Sec61 complex. Journal of Cell Biology, 150, pp.53–64.

Raghavan, M. et al., 2008. MHC class I assembly: out and about. Trends in Immunology, 29, pp.436–443.

Ramos, P.C. et al., 1998. Ump1p is required for proper maturation of the 20S proteasome and becomes its substrate upon completion of the assembly. Cell, 92, pp.489–499.

Randles, L., 2012. Ubiquitin and its binding domains. Frontiers in Bioscience, 17, p.2140.

Rane, N.S. et al., 2010. Signal sequence insufficiency contributes to neurodegeneration caused by transmembrane prion protein. Journal of Cell Biology, 188, pp.515–526.

Rane, N.S., Yonkovich, J.L. & Hegde, R.S., 2004. Protection from cytosolic prion protein toxicity by modulation of protein translocation. The EMBO journal, 23, pp.4550–4559.

Rapiejko, P.J. & Gilmore, R., 1997. Empty site forms of the SRP54 and SR alpha GTPases mediate targeting of ribosome-nascent chain complexes to the endoplasmic reticulum. Cell, 89, pp.703–713.

Rapoport, T.A. et al., 2004. Membrane-protein integration and the role of the translocation channel. Trends in Cell Biology, 14, pp.568–575.

Rebello, G. et al., 2004. Apoptosis-inducing signal sequence mutation in carbonic anhydrase IV identified in patients with the RP17 form of retinitis pigmentosa. Proceedings of the National Academy of Sciences of the United States of America, 101, pp.6617–6622.

Rechsteiner, M., Realini, C. & Ustrell, V., 2000. The proteasome activator 11 S REG (PA28) and class I antigen presentation. The Biochemical journal, 345 Pt 1, pp.1–15.

Redman, C.M., Siekevitz, P. & Palade, G.E., 1966. MACROMOLECULES : Synthesis and Transfer of Amylase in Pigeon Pancreatic Microsomes Synthesis and Transfer Pancreatic Microsomes * of Amylase in Pigeon.

Rehm, A. et al., 2001. Signal peptide cleavage of a type I membrane protein, HCMV US11, is dependent on its membrane anchor. EMBO Journal, 20, pp.1573–1582.

Reinheckel, T., Grune, T. & Davies, K., 2000. The Measurement of Protein Degradation in Response to Oxidative Stress. METHODS IN MOLECULAR BIOLOGY-CLIFTON THEN TOTOWA-. Available at: http://www.springerlink.com/index/lx5356163401j331.pdf.

127

Reits, E.A. et al., 2000. The major substrates for TAP in vivo are derived from newly synthesized proteins. Nature, 404, pp.774–778.

Ritz, D. et al., 2011. Endolysosomal sorting of ubiquitylated caveolin-1 is regulated by VCP and UBXD1 and impaired by VCP disease mutations. Nature cell biology, 13, pp.1116–1123.

Rodrigo-Brenni, M.C., Gutierrez, E. & Hegde, R.S., 2014. Cytosolic Quality Control of Mislocalized Proteins Requires RNF126 Recruitment to Bag6. Molecular Cell, 55, pp.227–237.

Roebroek, R.M., Postma, B.H. & Dijkstra, U.J., 1994. Aseptic meningitis caused by the lymphocytic choriomeningitis virus.,

Ron, D., 2002. Translational control in the endoplasmic reticulum stress response. Journal of Clinical Investigation, 110, pp.1383–1388.

Rosenkranz, B., Winkelmann, B.R. & Parnham, M.J., 1996. Clinical pharmacokinetics of molsidomine. Clinical pharmacokinetics, 30, pp.372–384.

Ross, C.A., 2002. Polyglutamine pathogenesis: Emergence of unifying mechanisms for Huntington’s disease and related disorders. Neuron, 35, pp.819–822.

Rothe, C. & Lehle, L., 1998. Sorting of invertase signal peptide mutants in yeast dependent and independent on the signal-recognition particle. European journal of biochemistry / FEBS, 252, pp.16–24.

Rousseau, M. et al., 1997. Lymphocytic choriomeningitis virus in southern France: four case reports and a review of the literature. European journal of …, 13, pp.817–823. Available at: http://link.springer.com/article/10.1023/A:1007434521082.

Rubinsztein, D.C., 2006. The roles of intracellular protein-degradation pathways in neurodegeneration. Nature, 443, pp.780–786.

Ruggiano, A., Foresti, O. & Carvalho, P., 2014. ER-associated degradation: Protein quality control and beyond. Journal of Cell Biology, 204, pp.869–879.

Rush, J. et al., 2005. Immunoaffinity profiling of tyrosine phosphorylation in cancer cells. Nature biotechnology, 23, pp.94–101.

Rutkowski, D.T. et al., 2003. Signal sequences initiate the pathway of maturation in the endoplasmic reticulum lumen. Journal of Biological Chemistry, 278, pp.30365–30372.

Sadler, I. et al., 1989. A yeast gene important for protein assembly into the endoplasmic reticulum and the nucleus has homology to DnaJ, an Escherichia coli heat shock protein. Journal of Cell Biology, 109, pp.2665–2675.

Salas, C., Broglio, C. & Rodríguez, F., 2003. Evolution of forebrain and spatial cognition in vertebrates: Conservation across diversity. In Brain, Behavior and Evolution. pp. 72–82.

Samali, A. et al., 2010. Mechanisms of ER stress-mediated mitochondrial membrane permeabilization. International Journal of Cell Biology.

Sanyal, S., Claessen, J.H.L. & Ploegh, H.L., 2012. A viral deubiquitylating enzyme restores dislocation of substrates from the endoplasmic reticulum (ER) in semi-intact cells. Journal of Biological Chemistry, 287, pp.23594–23603.

Satpute-Krishnan, P. et al., 2014. ER stress-induced clearance of misfolded GPI-anchored proteins via the secretory pathway. Cell, 158, pp.522–533.

Sawkar, A.R., D’Haeze, W. & Kelly, J.W., 2006. Therapeutic strategies to ameliorate lysosomal storage disorders--a focus on Gaucher disease. Cellular and molecular life sciences : CMLS, 63, pp.1179–1192. 128

Sayre, L.M., Smith, M.A. & Perry, G., 2001. Chemistry and biochemistry of oxidative stress in neurodegenerative disease. Current medicinal chemistry, 8, pp.721–738.

Van Schadewijk, A. et al., 2012. A quantitative method for detection of spliced X-box binding protein-1 (XBP1) mRNA as a measure of endoplasmic reticulum (ER) stress. Cell Stress and Chaperones, 17, pp.275–279.

Schägger, H. & von Jagow, G., 1987. Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Analytical biochemistry, 166, pp.368–379.

Schaheen, B., Dang, H. & Fares, H., 2009. Derlin-dependent accumulation of integral membrane proteins at cell surfaces. Journal of cell science, 122, pp.2228–2239.

Schildknecht, S. et al., 2011. Characterization of Mouse Cell Line IMA 2 . 1 as a Potential Model System to Study Astrocyte Functions. , pp.261–274.

Schlenstedt, G. et al., 1990. A large presecretory protein translocates both cotranslationally, using signal recognition particle and ribosome, and post-translationally, without these ribonucleoparticles, when synthesized in the presence of mammalian microsomes. Journal of Biological Chemistry, 265, pp.13960–13968.

Schlenstedt, G. et al., 1992. Structural requirements for transport of preprocecropinA and related presecretory proteins into mammalian microsomes. Journal of Biological Chemistry, 267, pp.24328–24332.

Schlenstedt, G. & Zimmermann, R., 1987. Import of frog prepropeptide GLa into microsomes requires ATP but does not involve docking protein or ribosomes. The EMBO journal, 6, pp.699–703.

Schlesinger, D.H., Goldstein, G. & Niall, H.D., 1975. The complete amino acid sequence of ubiquitin, an adenylate cyclase stimulating polypeptide probably universal in living cells. Biochemistry, 14, pp.2214–2218.

Schlosser, E. et al., 2007. A novel cytosolic class I antigen-processing pathway for endoplasmic-reticulum-targeted proteins. EMBO reports, 8, pp.945–951.

Schmidt, M. et al., 1999. Sequence information within proteasomal prosequences mediates efficient integration of beta-subunits into the 20 S proteasome complex. Journal of molecular biology, 288, pp.117–128.

Schmidtke, G. et al., 1996. Analysis of mammalian 20S proteasome biogenesis: the maturation of beta-subunits is an ordered two-step mechanism involving autocatalysis. The EMBO journal, 15, pp.6887–6898.

Schubert, U. et al., 2000. Rapid degradation of a large fraction of newly synthesized proteins by proteasomes. Nature, 404, pp.770–774.

Schuberth, C. & Buchberger, A., 2008. UBX domain proteins: Major regulators of the AAA ATPase Cdc48/p97. Cellular and Molecular Life Sciences, 65, pp.2360–2371.

Sedelnikova, O.A. et al., 2010. Role of oxidatively induced DNA lesions in human pathogenesis. Mutation Research - Reviews in Mutation Research, 704, pp.152–159.

Seemüller, E. et al., 1995. Proteasome from Thermoplasma acidophilum: a threonine protease. Science (New York, N.Y.), 268, pp.579–582.

Seifert, U. et al., 2010. Immunoproteasomes preserve protein homeostasis upon interferon-induced oxidative stress. Cell, 142, pp.613–624.

Selby, M. et al., 1999. Hepatitis C virus envelope glycoprotein E1 originates in the endoplasmic reticulum and requires cytoplasmic processing for presentation by class I MHC molecules. Journal of immunology (Baltimore, Md. : 1950), 162, pp.669–676.

Selkoe, D.J., 2003. Folding proteins in fatal ways. Nature, 426, pp.900–904. 129

Shao, S. & Hegde, R.S., 2011. A calmodulin-dependent translocation pathway for small secretory proteins. Cell, 147, pp.1576–1588.

Shaw, A.S., Rottier, P.J. & Rose, J.K., 1988. Evidence for the loop model of signal-sequence insertion into the endoplasmic reticulum. Proceedings of the National Academy of Sciences of the United States of America, 85, pp.7592–7596.

Shen, J. et al., 2002. ER stress regulation of ATF6 localization by dissociation of BiP/GRP78 binding and unmasking of golgi localization signals. Developmental Cell, 3, pp.99–111.

Shenkman, M. et al., 2013. A shared endoplasmic reticulum-associated degradation pathway involving the EDEM1 protein for glycosylated and nonglycosylated proteins. Journal of Biological Chemistry, 288, pp.2167–2178.

Shringarpure, R. et al., 2003. Ubiquitin conjugation is not required for the degradation of oxidized proteins by proteasome. Journal of Biological Chemistry, 278, pp.311–318.

Simon, S.M. & Blobel, G., 1991. A protein-conducting channel in the endoplasmic reticulum. Cell, 65, pp.371–380.

Skipper, J.C. et al., 1996. An HLA-A2-restricted tyrosinase antigen on melanoma cells results from posttranslational modification and suggests a novel pathway for processing of membrane proteins. The Journal of experimental medicine, 183, pp.527–534.

Sloper-Mould, K.E. et al., 2001. Distinct Functional Surface Regions on Ubiquitin. Journal of Biological Chemistry, 276, pp.30483–30489.

Smith, D.M. et al., 2005. ATP binding to PAN or the 26S ATPases causes association with the 20S proteasome, gate opening, and translocation of unfolded proteins. Molecular Cell, 20, pp.687–698.

Smith, M.H., Ploegh, H.L. & Weissman, J.S., 2011. Road to Ruin: Targeting Proteins for Degradation in the Endoplasmic Reticulum. Science, 334, pp.1086–1090.

Soltys, B.J., Falah, M. & Gupta, R.S., 1996. Identification of endoplasmic reticulum in the primitive eukaryote Giardia lamblia using cryoelectron microscopy and antibody to Bip. Journal of cell science, 109 ( Pt 7, pp.1909–1917.

Soto, C., Estrada, L. & Castilla, J., 2006. Amyloids, prions and the inherent infectious nature of misfolded protein aggregates. Trends in Biochemical Sciences, 31, pp.150–155.

Sowa, M.E. et al., 2009. Defining the Human Deubiquitinating Enzyme Interaction Landscape. Cell, 138, pp.389–403.

Stagg, H.R. et al., 2009. The TRC8 E3 ligase ubiquitinates MHC class I molecules before dislocation from the ER. Journal of Cell Biology, 186, pp.685–692.

Stewart, R.S. & Harris, D.A., 2003. Mutational Analysis of Topological Determinants in Prion Protein (PrP) and Measurement of Transmembrane and Cytosolic PrP during Prion Infection. Journal of Biological Chemistry, 278, pp.45960–45968.

Szegezdi, E. et al., 2006. Mediators of endoplasmic reticulum stress-induced apoptosis. EMBO reports, 7, pp.880– 885.

Tagaya, Y. et al., 1997. Generation of secretable and nonsecretable interleukin 15 isoforms through alternate usage of signal peptides. Proceedings of the National Academy of Sciences of the United States of America, 94, pp.14444–14449.

Taguchi, K., Motohashi, H. & Yamamoto, M., 2011. Molecular mechanisms of the Keap1-Nrf2 pathway in stress response and cancer evolution. Genes to Cells, 16, pp.123–140.

130

Takahama, Y. et al., 2012. β5t-containing thymoproteasome: Specific expression in thymic cortical epithelial cells and role in positive selection of CD8+ T cells. Current Opinion in Immunology, 24, pp.92–98.

Tirasophon, W., Welihinda, A.A. & Kaufman, R.J., 1998. A stress response pathway from the endoplasmic reticulum to the nucleus requires a novel bifunctional protein kinase/endoribonuclease (Ire1p) in mammalian cells. Genes and Development, 12, pp.1812–1824.

Treiman, M., Caspersen, C. & Christensen, S.B., 1998. A tool coming of age: Thapsigargin as an inhibitor of sarco- endoplasmic reticulum Ca2+-ATPases. Trends in Pharmacological Sciences, 19, pp.131–135.

Trombetta, E.S., Simons, J.F. & Helenius, A., 1996. Endoplasmic reticulum glucosidase II is composed of a catalytic subunit, conserved from yeast to mammals, and a tightly bound noncatalytic HDEL- containing subunit. Journal of Biological Chemistry, 271, pp.27509–27516.

Tsai, Y.C. & Weissman, A.M., 2012. A Ubiquitin-Binding Rhomboid Protease Aimed at ERADication. Developmental Cell, 23, pp.454–456.

Tsubuki, S. et al., 1993. Purification and characterization of a Z-Leu-Leu-Leu-MCA degrading protease expected to regulate neurite formation: a novel catalytic activity in proteasome. Biochemical and biophysical research communications, 196, pp.1195–1201.

Turner, G.C. & Varshavsky, A., 2000. Detecting and measuring cotranslational protein degradation in vivo. Science (New York, N.Y.), 289, pp.2117–2120.

Turrens, J.F. & Boveris, A., 1980. Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. The Biochemical journal, 191, pp.421–427.

Urano, F. et al., 2000. Coupling of stress in the ER to activation of JNK protein kinases by transmembrane protein kinase IRE1. Science (New York, N.Y.), 287, pp.664–666.

Ushioda, R., Hoseki, J. & Nagata, K., 2013. Glycosylation-independent ERAD pathway serves as a backup system under ER stress. Molecular biology of the cell, 24, pp.3155–63. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3806662&tool=pmcentrez&rendertype=abstract.

Ustrell, V. et al., 2002. Pa200, a nuclear proteasome activator involved in DNA repair. EMBO Journal, 21, pp.3516– 3525.

VanValkenburgh, C. et al., 1999. The catalytic mechanism of endoplasmic reticulum signal peptidase appears to be distinct from most eubacterial signal peptidases. Journal of Biological Chemistry, 274, pp.11519–11525.

Vembar, S.S. & Brodsky, J.L., 2008. One step at a time: endoplasmic reticulum-associated degradation. Nature reviews. Molecular cell biology, 9, pp.944–957.

Verma, R. et al., 2002. Role of Rpn11 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science (New York, N.Y.), 298, pp.611–615.

Vilardi, F., Lorenz, H. & Dobberstein, B., 2011. WRB is the receptor for TRC40/Asna1-mediated insertion of tail- anchored proteins into the ER membrane. Journal of cell science, 124, pp.1301–1307.

Vile, G.F., Tanew-Ilitschew, A. & Tyrrell, R.M., 1995. Activation of NF-kappa B in human skin fibroblasts by the oxidative stress generated by UVA radiation. Photochemistry and photobiology, 62, pp.463–468.

Vincenz, L. et al., 2013. Endoplasmic reticulum stress and the unfolded protein response: targeting the Achilles heel of multiple myeloma. Mol Cancer Ther, 12, pp.831–843. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23729400.

131

Vincenz, L. et al., 2013. Endoplasmic reticulum stress and the unfolded protein response: targeting the Achilles heel of multiple myeloma. Molecular cancer therapeutics, 12, pp.831–43. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23729400.

Voigt, S. et al., 1996. Signal sequence-dependent function of the TRAM protein during early phases of protein transport across the endoplasmic reticulum membrane. Journal of Cell Biology, 134, pp.25–35.

Walter, P., Ibrahimi, I. & Blobel, G., 1981. Translocation of proteins across the endoplasmic reticulum. I. Signal recognition protein (SRP) binds to in-vitro-assembled polysomes synthesizing secretory protein. The Journal of cell biology, 91, pp.545–550.

Wang, F. et al., 2010. A chaperone cascade sorts proteins for posttranslational membrane insertion into the endoplasmic reticulum. Molecular Cell, 40, pp.159–171.

Wang, J. & Maldonado, M.A., 2006. The ubiquitin-proteasome system and its role in inflammatory and autoimmune diseases. Cellular & molecular immunology, 3, pp.255–261.

Wang, Q. et al., 2011. A Ubiquitin Ligase-Associated Chaperone Holdase Maintains Polypeptides in Soluble States for Proteasome Degradation. Molecular Cell, 42, pp.758–770.

Wang, X. et al., 2007. Mass spectrometric characterization of the affinity-purified human 26S proteasome complex. Biochemistry, 46, pp.3553–3565.

Wang, X. et al., 2010. Regulation of the 26S proteasome complex during oxidative stress. Science signaling, 3, p.ra88.

Watanabe, Y. et al., 2003. Interferon-gamma induces reactive oxygen species and endoplasmic reticulum stress at the hepatic apoptosis. Journal of cellular biochemistry, 89, pp.244–253.

Wattenberg, B. & Lithgow, T., 2001. Targeting of C-terminal (tail)-anchored proteins: understanding how cytoplasmic activities are anchored to intracellular membranes. Traffic (Copenhagen, Denmark), 2, pp.66–71.

Weihofen, A. et al., 2002. Identification of signal peptide peptidase, a presenilin-type aspartic protease. Science (New York, N.Y.), 296, pp.2215–2218.

Weissman, A.M., 2001. Themes and variations on ubiquitylation. Nature reviews. Molecular cell biology, 2, pp.169– 178.

Wessel, D. & Flügge, U.I., 1984. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Analytical biochemistry, 138, pp.141–143.

Wild, K. et al., 2004. SRP meets the ribosome. Nature structural & molecular biology, 11, pp.1049–1053.

Williams, J.M. et al., 2013. The ERdj5-Sel1L complex facilitates cholera toxin retrotranslocation. Molecular biology of the cell, 24, pp.785–95. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3596249&tool=pmcentrez&rendertype=abstract.

Witt, S. et al., 2006. Proteasome Assembly Triggers a Switch Required for Active-Site Maturation. Structure, 14, pp.1179–1188.

Witze, E.S. et al., 2007. Mapping protein post-translational modifications with mass spectrometry. Nature methods, 4, pp.798–806.

Wolf, D.H. & Stolz, A., 2012. The Cdc48 machine in endoplasmic reticulum associated protein degradation. Biochimica et Biophysica Acta - Molecular Cell Research, 1823, pp.117–124.

132

Wu, Y.-T. et al., 2010. Dual role of 3-methyladenine in modulation of autophagy via different temporal patterns of inhibition on class I and III phosphoinositide 3-kinase. The Journal of biological chemistry, 285, pp.10850– 10861.

Xiao, B.G. & Link, H., 1998. Immune regulation within the central nervous system. Journal of the neurological sciences, 157, pp.1–12.

Yabal, M. et al., 2003. Translocation of the C terminus of a tail-anchored protein across the endoplasmic reticulum membrane in yeast mutants defective in signal peptide-driven translocation. Journal of Biological Chemistry, 278, pp.3489–3496.

Yamaguchi, A. et al., 1999. Stress-associated endoplasmic reticulum protein 1 (SERP1)/ribosome- associated membrane protein 4 (RAMP4) stabilizes membrane proteins during stress and facilitates subsequent glycosylation. Journal of Cell Biology, 147, pp.1195–1204.

Yamamoto, Y. et al., 1987. Engineering of the hydrophobic segment of the signal sequence for efficient secretion of human lysozyme by Saccharomyces cerevisiae. Biochemical and biophysical research communications, 149, pp.431–436.

Yamamoto, Y. & Sakisaka, T., 2012. Molecular Machinery for Insertion of Tail-Anchored Membrane Proteins into the Endoplasmic Reticulum Membrane in Mammalian Cells. Molecular Cell, 48, pp.387–397.

Ye, Y. et al., 2004. A membrane protein complex mediates retro-translocation from the ER lumen into the cytosol. Nature, 429, pp.841–847.

Ye, Y., Meyer, H.H. & Rapoport, T.A., 2003. Function of the p97-Ufd1-Npl4 complex in retrotranslocation from the ER to the cytosol: Dual recognition of nonubiquitinated polypeptide segments and polyubiquitin chains. Journal of Cell Biology, 162, pp.71–84.

Yedidia, Y. et al., 2001. Proteasomes and ubiquitin are involved in the turnover of the wild-type prion protein. EMBO Journal, 20, pp.5383–5391.

Yewdell, J.W., Antón, L.C. & Bennink, J.R., 1996. Defective ribosomal products (DRiPs): a major source of antigenic peptides for MHC class I molecules? Journal of immunology (Baltimore, Md. : 1950), 157, pp.1823–1826.

Yoshida, H. et al., 2001. XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell, 107, pp.881–891.

Yoshida, Y., 2007. F-box proteins that contain sugar-binding domains. Bioscience, biotechnology, and biochemistry, 71, pp.2623–2631.

Yoshida, Y. et al., 2005. Glycoprotein-specific ubiquitin ligases recognize N-glycans in unfolded substrates. EMBO reports, 6, pp.239–244.

Young, P. et al., 1998. Characterization of two polyubiquitin binding sites in the 26 S protease subunit 5a. Journal of Biological Chemistry, 273, pp.5461–5467.

Zimmermann, R. et al., 1990. Ribonucleoparticle-independent transport of proteins into mammalian microsomes. Journal of Bioenergetics and Biomembranes, 22, pp.711–723.

Zimmermann, R. & Mollay, C., 1986. Import of honeybee prepromelittin into the endoplasmic reticulum. Requirements for membrane insertion, processing, and sequestration. Journal of Biological Chemistry, 261, pp.12889–12895.

Zimmermann, R., Müller, L. & Wullich, B., 2006. Protein transport into the endoplasmic reticulum: mechanisms and pathologies. Trends in Molecular Medicine, 12, pp.567–573.

133

Zwickl, P. et al., 1999. An archaebacterial ATPase, homologous to ATPases in the eukaryotic 26 S proteasome, activates protein breakdown by 20 S proteasomes. Journal of Biological Chemistry, 274, pp.26008–26014.

134

8.2 Abbreviations % per cent °C Degrees Celsius [35S] Sulfur-35 aa amino acid

AAT α1-Antitrypsin Ag Antigen Amp Ampicillin AMP Adenosine Monophosphate APS Ammonium Persulfate ATP Adenosine Triphosphate B2M β2-Microglobulin bp base pair BSA Bovine Serum Albumin CA4 Carbonic Anhydrase 4 CaCl2 Calcium Chloride CCR7 C-C Chemokine Receptor –type 7 CD Cluster of Differentiation cDNA complementary Deoxyribonucleic Acid CNS Central Nervous System cpm counts per minute DC Dendritic Cell ddH2O double-distilled Water DMEM Dulbecco’s Modified Eagle Medium DMSO Dimethyl Sulfoxide DNA Deoxyribonucleic Acid dNTPs Deoxyribonucleoside Triphosphates dsDNA double-stranded DNA DTT Dithiothreitol E. coli Escherichia coli EDEM ER-Degradation Enhancing α-Mannosidase-like Protein EDTA Ethylenediaminetetraacetic Acid ER Endoplasmic Reticulum ER SP Endoplasmic Reticulum Signal Peptidase ERAD Endoplasmic Reticulum-Associated Degradation et al. et alii (and others) EtBr Ethidium Bromide EtOH Ethanol FACS Fluorescence-Activated Cell Sorting FCS Fetal Calf Serum FD Fast Digest Fig Figure FSC Forward Scatter

135

Fwd Forward g gravity, gram GAPDH Glyceraldehyde 3-phosphate dehydrogenase GP Glycoprotein GPI Glycosylphosphatidylinositol h hour HA Haemagglutinin HECT Homologous to the E6-AP Carboxyl Terminus HEK Human Embryonic Kidney HF High Fidelity HRP Horseradish Peroxidase Hsp Heat Shock Protein ICAM Intercellular Adhesion Molecule IFN Interferon Ig Immunoglobulin IP Immunoprecipitation KCl Potassium Chloride Kan Kanamycin kb kilo base kbp kilo base pairs kDa kilo Dalton l liter LB lysogen broth LCMV Lymphocytic Choriomeningitis Virus LMP2/7 Low Molecular Mass Protein 2/7 m murine M Molarity Man Mannose MECL Multicatalytic Endopeptidase Complex Subunit MeOH Methanol mg milligram MHC Major Histocompatibility Complex min minute ml milliliter MLP Mislocalized Protein mM milliMolar mRNA messenger RNA NAc N-Acetylglucosamine NaCl Sodium Chloride NET-TO NaCl-EDTA-[-Triton-X100-Ovalbumin ng nanogram nm nanometer OD Optical Density o.n. over night PA Proteasome Activator 136

PAGE Polyacrylamide Gel Electrophoresis PBS Phosphate-Buffered Saline PCR Polymerase Chain Reaction Pen/Strep Penicillin/Streptomycin PFA Paraformaldehyde Pfu Plaque-forming Units pmol picomol PMSF Phenylmethanesulfonyl Fluoride POMP Proteasome Maturation Protein PSCA Prostate Stem Cell Antigen PSMA Proteasome Subunit Alpha PSMB Proteasome Subunit Beta qRT-PCR quantitative Real-Time PCR Rev Reverse RING Really Interesting New Gene RNA Ribonucleic Acid RNC Ribosome Nascent Chain RP Regulatory Particle rpm revolutions per minute RPN Regulatory Particle non-ATPase RPT Regulatory Particle triple-A RT Room Temperature, Reverse Transcription S Svedberg SDS Sodium Dodecyl Sulfate SP Signal Peptide SPase Signal Peptidase SPC Signal Peptidase Complex SPCS Signal Peptidase Complex Subunit SR SRP Receptor SRP Signal Recognition Particle SS Signal Sequence SSC Sideward Scatter SV40 Simian Virus 40 TAE Tris-Acetate-EDTA TBS Tris-Buffered Saline TBS-T Tris-Buffered Saline-Tween-20 TE Tris-EDTA TEMED Tetramethylethylenediamine TM Transmembrane TNF Tumor Necrosis Factor U Unit Ub Ubiquitin UV Ultraviolet V Volt VCAM1 Vascular Cell Adhesion Molecule 1 137

WT Wild Type μg microgram μl microliter

8.3 Amino acid abbreviations

Amino Acid Three-Letter Abbreviation One-Letter Abbreviation Alanine Ala A Arginine Arg R Asparagine Asn N Aspartate Asp D Cysteine Cys C Glutamate Glu E Glutamine Gln Q Glycine Gly G Histidine His H Isoleucine Ile I Leucine Leu L Lysine Lys K Methionine Met M Phenylalanine Phe F Proline Pro P Serine Ser S Threonine Thr T Tryptophan Trp W Tyrosine Tyr Y Valine Val V

138

Acknowledgement

First of all, I would like to thank Prof. Marcus Groettrup for giving me the opportunity to perform this PhD thesis in his group. Thanks for the constant interest in my projects and the fruitful discussions, the permanent motivation and the chance to follow my own theories. Even during difficult phases and when reorientation was needed, your support was always there. Thank you for this.

Next, I like to thank Prof. Martin Scheffner for his kind agreement of reviewing my thesis and, together with Prof. Jörg Tatzelt, for serving as a member of my thesis committee.

I thank the Konstanz Research School Chemical Biology (KoRS-CB) for accepting and supporting me. I thank the Bioimaging Center of Konstanz University for educating me in confocal microscopy.

Thanks to the entire group of Immunology Konstanz for making this chapter of my life so enjoyable. Thanks to Gunter for his biochemical support, great jokes and all the numberless signatures on my orders for radioactive stuff. Thanks to Michi, for his immunological support and some entertaining discussions. Another special thanks goes to the other two “Tussi-Box” members, Sarah & Valerie, you made this time very special! Thanks to you, Christian, Gretel, Konsti, Richi and Valentina for being great co-workers and even better friends on sunny and rainy days. Special thanks to Sarah, Valerie and Gunter for the proofreading of this thesis.

I also thank my former Master student Kathrin for giving me the opportunity to consider my supervising qualities and for her excellent lab work.

Very big thanks goes to a very special person I met during my PhD: Sarah (alias Homie). Well, you know, for everything.

I like to thank my family for their love. I thank Bernd for traveling a long part of this journey with me, supporting me all the time without doubts.

Thanks to Max, for giving me the opportunity to write this thesis under the best possible conditions and for everything else. <3

139