The Chaperones P97 and BAG6 in MHC Class I‐ Restricted Antigen Processing

The chaperones p97 and BAG6 in MHC class I‐ restricted antigen processing

Dissertation

zur Erlangung des akademischen Grades

eines Doktors der Naturwissenschaften (Dr. rer. nat.)

vorgelegt von

Annegret Bitzer

an der

Mathematisch‐Naturwissenschaftliche Sektion

Fachbereich Biologie

Tag der mündlichen Prüfung: 14.4.2016 1. Referent: Prof. Dr. Alexander Bürkle 2. Referent: Prof. Dr. Marcus Gröttrup

Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-0-329445 Table of contents

Zusammenfassung ...... 1

Abstract ...... 3

1. Introduction ...... 5 1.1. The immune system ...... 5 1.2 Cellular protein homeostasis ...... 5 Ubiquitin proteasome system (UPS) ...... 6 Immunoproteasome ...... 9 p97 ...... 10 BAG6 ...... 18 1.4 Antigen processing and presentation on MHC class I ...... 22 The MHC class I‐restricted pathway ...... 22 The source of antigenic peptides ...... 23 1.4 NF‐κB activation ...... 25 1.5 Virus strains ...... 28 Lymphocytic choriomeningitis virus ...... 28 Vaccinia virus ...... 29 Aim of this study ...... 29

2. Material and Methods ...... 31 2.1 Molecular cloning and plasmid preparation ...... 31 2.2 Agarose gel electrophoresis ...... 31 2.3 Mice ...... 32 2.4 Virus amplification and titer determination ...... 32 LCMV strain WE (LCMV‐WE) ...... 32 VV Western Reserve strain (VV‐WR) and rVV strains ...... 32 2.5 Cell culture ...... 33 2.6 Preparation of peritoneal macrophages and MEFs ...... 34 2.7 Stimulation of cells ...... 34 2.8. Treatment of cells with chemical inhibitors ...... 35 2.9 Transfection of cells ...... 35 Transient transfection with expression constructs ...... 35 Stable transfection of HEK293 cells ...... 36 Transfection with siRNA ...... 36 2.10 SDS‐PAGE (sodium dodecylsulfate polyacrylamide gel electrophoresis) and western blot ...... 37 2.11 Immunoprecipitation (IP) ...... 39 2.12 Proteasome IP ...... 39 2.13 2D‐gel electrophoresis ...... 40 2.14 Electrophoretic mobility shift assay (EMSA) ...... 40 2.15 Ribosome isolation ...... 41 2.16 Real‐time RT‐PCR ...... 42 2.17 Acid wash ...... 42 2.18 Flow cytometry ...... 43 Surface staining ...... 43 GFP fluorescence ...... 43 Dead cell staining ...... 43 Intracellular IFN‐g staining ...... 43 2.19 Generation of epitope‐specific CTLs ...... 44 2.20 Antigen presentation assay using CTLs ...... 45 2.21 Antigen presentation assay using T cell hybridoma lines ...... 45 2.22 Magnetic‐activated cell sorting (MACS) ...... 46 2.23 Statistical analysis ...... 46

3. Results ...... 47 3.1 The role of p97 in antigen processing and presentation on MHC class I ...... 47 Chemical inhibition of p97 blocks MHC class I cell surface expression ...... 47 Expression of a dominant negative p97 mutant interferes with MHC class I cell surface expression ...... 48 Presentation of virus‐derived epitopes on MHC class I depends on p97 activity ...... 50 Expression of a DN p97 mutant inhibits presentation of virus‐derived epitopes ...... 52 Ribosome‐associated degradation via the E3 ligase LTN1 is dispensable for MHC class I‐ restricted antigen presentation ...... 53 3.2 The role BAG6 in antigen processing and presentation on MHC class I ...... 55 MHC class I surface expression is independent of BAG6 ...... 55 BAG6 is not transcriptionally regulated after IFN‐γ stimulation...... 56 Presentation of LCMV‐derived epitopes is independent of BAG6 ...... 57 Presentation of retrotranslocated tyrosinase‐derived epitope is not affected by BAG6 knockdown ...... 59 3.3 The influence of immunoproteasome deficiency on NF‐kB activation ...... 60 Expression of immunoproteasome subunits in peritoneal macrophages and MEFs ...... 60 Degradation of IkBα is not affected in immunoproteasome knockout cells ...... 61 Nuclear translocation of free NF‐kB and transactivation of target genes is normal in immunoproteasome knockout cells ...... 63

4. Discussion ...... 65 4.1 Protein homeostasis and immune regulation ...... 65 Antigen processing – which factors are involved? ...... 65 4.2 Major role of p97 in antigen processing and presentation on MHC class I ...... 67 4.3 BAG6 is dispensable for antigen processing and presentation on MHC class I ...... 70 4.3 Immunoproteasome deficiency has no influence on NF‐kB activation ...... 72

References ...... 76

Appendix ...... 96 Abbreviations ...... 96 Abbreviations of amino acids ...... 102 Acknowledgments ...... 103

Zusammenfassung

Antigenpräsentation auf MHC‐Klasse‐I Molekülen ermöglicht immunologische Kontrolle intrazellulär synthetisierter Proteine. Bei den präsentierten Epitopen handelt es sich im Allgemeinen um kurze Peptide, die durch proteasomalen Abbau entstehen. Als Quelle antigener Peptide dient sehr wahrscheinlich ein Teil der schnell abgebauten Proteine, die auch als defekte ribosomale Produkte (DRiPs) bezeichnet werden. Die essentiellen Schritte zu Beginn der Antigenprozessierung, nämlich Proteinsynthese am Ribosom und der Abbau über das Proteasom, sind bereits seit Jahren bekannt. Dennoch konnten bis jetzt nur sehr wenige Faktoren identifiziert werden, die einen Einfluss auf die Entstehung und Prozessierung von DRiPs haben.

Die AAA‐ATPase p97 ist Teil verschiedener Protein‐Kontrollmechanismen und unterstützt den Abbau von Substraten durch das Ubiquitin‐Proteasom‐System. Mithilfe von chemischer Inhibition und der Expression einer dominant negativen p97 Variante konnte gezeigt werden, dass die MHC‐Klasse‐I‐restringierte Präsentation von Virusepitopen und endogenen Epitopen sowie die Gesamtmenge an MHC‐Klasse‐I Molekülen an der Zelloberfläche von der Aktivität von p97 abhängig ist. Die schnelle Akkumulierung von polyubiquitylierten Proteinen in Zellen mit gestörter p97 Aktivität deutet auf eine dem Proteasom vorangestellte Rolle von p97 hin. Zusammengefasst konnte p97 als einen essentiellen Faktor in der Antigenprozessierung für MHC‐Klasse‐I Präsentation identifiziert werden und dadurch das Repertoire an p97‐abhängigen Funktionen in der Zelle erweitert werden.

Das Chaperon BAG6 ist ein weiterer Faktor, dessen Einfluss auf die MHC‐Klasse‐I Präsentation untersucht wurde. Obwohl zuvor eine essentielle Rolle von BAG6 postuliert wurde, zeigen die Experimente, dass BAG6 für die Oberflächenexpression von MHC‐Klasse‐I Molekülen und die MHC‐Klasse‐I‐restringierte Präsentation von Virusepitopen entbehrlich ist. Da die Interaktion von BAG6 und dem Modellantigen Tyrosinase durch Inhibition des Proteasoms verstärkt wird, könnte BAG6 dennoch eine Rolle im Abbau von Antigenen spielen. Redundante Chaperon‐ Netzwerke überdecken möglicherweise den Beitrag von BAG6 an der Antigen Prozessierung und Präsentation.

Weiterhin wurde der Einfluss der Untereinheiten des Immunproteasoms auf die Aktivierung von NF‐κB innerhalb dieser Arbeit untersucht. Das Immunproteasom kann unter inflammatorischen Bedingungen induziert werden und zeichnet sich, im Vergleich zum

Zusammenfassung konstitutiven Proteasom durch eine veränderte Spezifität der Proteinspaltung aus. Die Aktivierung des pro‐inflammatorischen Transkriptionsfaktors NF‐κB erfordert den proteasomalen Abbau des Inhibitors von NF‐κB (IκB), um die Translokation in den Zellkern zu ermöglichen. Um den Effekt des Immunproteasoms auf die Aktivierung von NF‐κB zu untersuchen, wurden Peritonealmakrophagen und embryonische Fibroblasten genutzt. Diese wurden aus Mäusen isoliert, die für die Immunproteasom Untereinheiten LMP2 oder LMP7 und MECL‐1 defizient sind. Im kanonischen Signalweg konnten jedoch keine Unterschiede im Abbau von IκB, der Translokation von NF‐κB in den Kern, oder der Expression von Zielgenen beobachtet werden. Das Immunprotasom hat daher keinen Einfluss auf die Aktivierung von NF‐ κB.

Abstract

Antigen presentation on MHC class I molecules allows immunosurveillance of proteins synthesized within cells. Generally, epitopes presented derive from proteasomal degradation of proteins into small peptides. One source of antigenic peptides is most likely a fraction of rapidly degraded proteins termed defective ribosomal products (DRiPs). The essential steps at the beginning of antigen processing, namely protein synthesis at the ribosome and degradation via the proteasome, have been known for years. Nevertheless, there is a considerable paucity of identified factors, which influence DRiP formation and processing.

The AAA‐ATPase p97 is involved in different cellular protein control pathways and supports substrate degradation via the ubiquitin‐proteasome system. By chemical inhibition of p97 and expression of a dominant negative p97 mutant it is shown here that MHC class I‐restricted presentation of virus‐derived and endogenous peptides as well as bulk MHC class I surface expression is dependent on p97 activity. Rapid accumulation of polyubiquitylated proteins in cells with disrupted p97 activity points towards a role upstream of the proteasome. Taken together, p97 is identified as an essential factor for MHC class I‐restricted antigen processing, which further extends the repertoire of p97‐dependent cellular functions.

Another factor analyzed for its impact on antigen processing and presentation is the chaperone BAG6. Although an essential role of BAG6 in this pathway has been proposed previously, it was found to be dispensable for bulk MHC class I cell surface expression and presentation of virus‐ derived peptides. Still, interaction of BAG6 and the model antigen tyrosinase was enhanced during proteasome inhibition suggesting a role of BAG6 in antigen degradation. Also, redundant chaperone pathways could potentially mask the contribution of BAG6 to antigen processing and presentation.

Further, the influence of immunoproteasome subunits on NF‐κB activation was examined during this thesis. Inducible under inflammatory conditions, the immunoproteasome is mainly characterized through an altered cleavage specificity compared to the constitutive proteasome. Activation of the pro‐inflammatory transcription factor NF‐κB requires the signal induced proteasomal degradation of the inhibitor of NF‐κB (IκB) in order to allow nuclear translocation. To study the effect of immunoproteasomes on NF‐κB activation, peritoneal macrophages and mouse embryonic fibroblasts derived from mice deficient in the immunoproteasome subunits LMP2 or LMP7 and MECL‐1 were used. Along the canonical

Abstract signaling pathway no differences in IκB degradation, nuclear translocation of NF‐κB, or transactivation activity could be observed. Hence, immunoproteasome subunits have no influence on NF‐κB activation.

1. Introduction

1.1. The immune system The immune system protects organisms from infection by pathogens and development of cancer. Displaying several layers of defense, the vertebrate immune system can broadly be divided into physical barriers, innate immunity and adaptive immunity. The innate immune system provides immediate but non‐specific defense mechanisms and is needed to activate adaptive immune responses. A main pillar of the innate immune system is the discrimination between self and non‐self by recognizing pathogen‐associated molecular patterns (PAMPs) shared by many pathogens. Since the receptors involved are produced non‐clonally there is no need for clonal expansion of effector cells (Blach‐Olszewska, 2005). The effector functions of innate immunity comprise phagocytosis, cytokine‐induced resistance, complement and major histocompatibility complex (MHC)‐independent killing. In contrast to innate immunity, adaptive immunity relies on clonally‐distributed antigen‐specific receptors produced through rearrangement of DNA (Tonegawa, 1983). Although highly specific, adaptive responses are time delayed compared to innate responses due to the requirement of clonal expansion.

The adaptive immune response can be further subdivided into humoral and cell‐mediated response. While antibody‐secreting B cells are the mediators of the humoral arm, T cells mediate the effector functions of the cellular response. A central feature of T cells is the recognition of small peptides presented on MHC molecules via their T cell receptor (TCR) (Townsend et al., 1985; Zinkernagel and Doherty, 1974). Cytotoxic T lymphocytes (CTLs) recognize peptides presented on MHC class I molecules with their TCR and the co‐receptor cluster of differentiation 8 (CD8) (Swain, 1983). Upon recognition, infected or malignant cells are killed through the release of cytotoxic effector molecules and/or expression of receptor ligands triggering apoptosis (Peters et al., 1991; Squier and Cohen, 1994). T helper cells recognize peptides presented on MHC class II with the help of the co‐receptor CD4 (Swain, 1983). In contrast to CTLs, the effector function of T helper cells is the secretion of cytokines, which modulate the immune response (Luckheeram et al., 2012).

1.2 Cellular protein homeostasis One of the fundamental characteristics of a living cell is the constant flow of information from DNA to RNA to protein. The proteome is the entire set of proteins expressed in a cell and maintenance of proteome integrity, also termed proteostasis, is directly linked to cellular

1. Introduction health and functionality. Cellular networks regulating proteostasis essentially comprise protein synthesis, maintenance (folding and handling of aggregates) and degradation (Powers et al., 2009; Wickner et al., 1999).

Figure 1: Central role of molecular chaperones in proteostasis. Chaperones connect all cellular networks regulating proteostasis. Pathways of each network are indicated. UPS = Ubiquitin proteasome system. Figure adapted from (Kim et al., 2013).

Molecular chaperones are centrally involved in each regulatory network of proteostasis (Kim et al., 2013; Wickner et al., 1999) (Fig. 1). As soon as a nascent chain emerges from the ribosome, it is subjected to co‐translational interaction with chaperones and other protein biogenesis factors (Kramer et al., 2009). The common feature of chaperones is the ability to recognize unfolded or misfolded proteins (Hartl and Hayer‐Hartl, 2002). Functionally, chaperones can be classified into holding, folding and unfolding chaperones (Brehme et al., 2014; Stirling et al., 2003).

Finally, proteases remove proteins in a highly specific fashion and replenish the cellular amino acid pool (Sauer and Baker, 2011). This process not only removes terminally inactivated proteins (Goldberg, 2003), it also regulates signal induced proteome re‐organization (Conaway et al., 2002; Reed, 2003). The bulk of cellular proteins is degraded by the proteasome (Rock et al., 1994). However, aggregated polypeptides unable to unfold can be removed by lysosomal degradation via autophagy (Mizushima, 2007).

Ubiquitin proteasome system (UPS) The UPS is the main selective degradation pathway in the cytoplasm and nucleus of eukaryotic cells (Fig. 2). Besides degrading proteins of cytoplasmic and nuclear origin (Nielsen et al., 2014),

1. Introduction it also mediates degradation of proteins derived from the endoplasmic reticulum (ER) (Buchberger et al., 2010; Stolz and Wolf, 2010) and mitochondria (Taylor and Rutter, 2011).

In order to be degraded by the proteasome, the majority of proteins have to be post‐ translationally modified with ubiquitin. Ubiquitin is a highly conserved 76 amino acid (aa) protein, which is found in the cytoplasm and nucleus of eukaryotic cells (Finley and Chau, 1991; Schlesinger and Goldstein, 1975). It is expressed as linear ubiquitin fusion protein (Wiborg et al., 1985) or fused to ribosomal subunits (Finley et al., 1989) and is subsequently cleaved by deubiquitylases (DUBs) into single ubiquitin molecules (Grou et al., 2015).

Figure 2: The ubiquitin proteasome system. Degradation of substrate proteins is achieved through a series of enzymatic reactions. Ubiquitin is activated and bound to an E1 enzyme in an ATP‐dependent manner (step 1). Next, the ubiquitin moiety is transferred to an E2 enzyme in a conjugation reaction (step 2). Together with a substrate‐specific E3 enzyme (ligase) the ubiquitin moiety is then covalently linked to a lysine residue of the substrate (step 3). Repeated ubiquitin ligation reactions with a lysine residue from ubiquitin leads to chain formation. Deubiquitylating enzymes can also catalyze the deconjugation of ubiquitin (step 4). Depending on the type of chain, substrates are degraded by the proteasome in an ATP‐dependent manner (Lys48‐ and Lys11‐linked chains) (step 5) or mediate assembly of signaling complexes (Lys11‐, Lys63‐linked or linear chains) (step 6). X, Y, Z = ubiquitin binding proteins, Pi = inorganic phosphate, PPi = inorganic diphosphate, Ub = ubiquitin. Figure from (Vucic et al., 2011).

Covalent attachment of ubiquitin to target substrates is achieved through the consecutive activity of three different classes of enzymes: ubiquitin‐activating enzymes (E1), ubiquitin‐ conjugating enzymes (E2), and ubiquitin ligases (E3) (Hershko and Ciechanover, 1998). At the expense of ATP‐hydrolysis, the C‐terminal glycine motif of ubiquitin is activated and bound to the active site cysteine of the E1 enzyme. Next, the ubiquitin moiety is transferred to the active site cysteine of an E2 enzyme. From there, it is linked to the ε‐amino group of a substrate lysine via an isopeptide bond with the help of E3 ligases. Whether the E3 ligase involved during this last step of conjugation transiently takes over the ubiquitin moiety with its own active site cysteine or if it works as a scaffold for the E2 enzyme and substrate depends on the type of E3

1. Introduction ligase involved (homologous to E6AP carboxyl terminus (HECT)‐type E3 ligase or really interesting new gene (RING)‐type E3 ligase) (Metzger et al., 2012). With over 600 different enzymes recognizing one or several substrates (higher eukaryotes), E3 ligases also determine the specificity of the ubiquitylation machinery (Ardley and Robinson, 2005). While conjugation to a lysine residue is by far the most common form of modification, attachment to the N‐ terminus or to other nucleophilic amino acids (serine, threonine, cysteine) of the substrate has been reported (Kravtsova‐Ivantsiv and Ciechanover, 2012). Besides the three typical classes of enzymes, E4 enzymes (also termed ubiquitin chain assembly factors) were recognized as a separate class of enzymes mediating chain elongation (Hoppe, 2005). Moreover, DUBs can catalyze the deconjugation of ubiquitin making ubiquitylation a highly dynamic process (Eletr and Wilkinson, 2014).

Multiple rounds of ubiquitylation lead to the formation of chains through conjugation to an internal lysine or the N‐terminal methionine of a previously attached ubiquitin moiety. Lys48‐ linked chains are generally considered as proteasomal degradation signal (Chau et al., 1989; Pickart, 2000), whereas some of the other linkages (Lys6, Lys11, Lys29, Lys33, Lys63, Met1 (linear)) may participate in proteasomal targeting but are mainly implicated in other cellular processes like protein‐protein interactions (Komander and Rape, 2012). Apart from ubiquitylation, modification with the ubiquitin‐like modifier FAT10 can deliver substrates to the proteasome (Schmidtke et al., 2014) and even modification‐independent degradation can occur (Erales and Coffino, 2014).

Following their conjugation to ubiquitin, substrates are recognized and degraded by the 26S proteasome (Finley, 2009). Structurally, the 26S proteasome can be divided into the 20S core particle and its associated regulatory particles. The 20S core particle is a cylindrical complex of four stacked rings, each consisting of seven subunits. The outer rings contain seven different catalytically inactive alpha subunits (α1‐α7) mediating the binding to regulatory particles. In contrast, the inner rings contain seven different beta subunits (β1‐β7), three of which display N‐terminal threonine protease activity (β1, β2, β5) (Tanaka, 1998). Together, the beta subunits enclose a central catalytic chamber (Unno et al., 2002). The cleavage specificities of the active subunits can be generally classified as caspase‐like for β1 (C‐terminal of acidic residues), trypsin‐like for β2 (C‐terminal of basic residues) and chymotrypsin‐like for β5 (C‐terminal for hydrophobic residues) (Borissenko and Groll, 2007). The cleavage products of the proteasome are peptides of 2‐25 aa (Kisselev et al., 1999; Toes et al., 2001). These peptides can be further

1. Introduction degraded by cytosolic peptidases and/or enter the MHC class I‐restricted antigen processing pathway (Weimershaus et al., 2013).

Since the active sites of the proteasome are facing towards the inside of the core particle, substrates have to gain access to the central pore in order to be degraded. Therefore, 19S regulatory particles (PA700) are associated with proteasomes mediating substrate recognition, deubiquitylation, unfolding, 20S gate opening and substrate translocation (Stadtmueller and Hill, 2011). The 19 S regulatory particle consists of a lid and base (Glickman et al., 1998). While the lid harbors DUB activity, the base contains ubiquitin receptors and six AAA‐ATPases (ATPase associated with divers cellular activities) mediating substrate unfolding and gate opening (Ehlinger and Walters, 2013). Other regulatory particles, like the interferon (IFN)‐γ inducible 11S regulatory particle (PA28), can replace the 19S regulatory particle and influence the proteasomal activity (Dubiel et al., 1992; Ma et al., 1992).

Figure 3: Catalytically active subunits of constitutive proteasome and immunoproteasome. The constitutive proteasome and immunoproteasome contain a different set of catalytically active subunits. Modified from (Groettrup et al., 2010).

Immunoproteasome Cells of hematopoietic origin and cells stimulated with IFN‐γ or, to a lesser extent, with type I interferons express an additional set of catalytically active proteasome subunits: low molecular mass polypeptide 2 (LMP2 or β1i), multicatalytic endopeptidase complex‐like 1 (MECL‐1, or β2i), and LMP7 (β5i) (Barton et al., 2002; Khan et al., 2001b; Shin et al., 2006). During de novo synthesis of proteasomes, these subunits are incorporated and form the “immunoproteasome”, whereas expression of constitutive subunits is reduced (Griffin et al., 1998) (Fig. 3).

1. Introduction

The immunoproteasome is mainly characterized by an altered cleavage specificity compared to the constitutive proteasome. LMP2 changes the proteolytic activity from caspase‐like to chymotrypsin‐like and thereby produces more cleavage products with hydrophobic C‐termini (Driscoll et al., 1993; Gaczynska et al., 1994; Van Kaer et al., 1994). Likewise, the specificity of the β5 position to cleave after hydrophobic residues increases through incorporation of LMP7 (Driscoll et al., 1993; Gaczynska et al., 1994). Overall, more peptides with a hydrophobic C‐ terminus are produced, which are considered to be suited better for the binding to MHC class I molecules than other peptides (Strehl et al., 2005). It also allows the presentation of a variety of peptides that are distinct from non‐inflamed tissue. By providing a better discrimination between tissues, these differences could potentially protect against autoimmune pathology (Groettrup et al., 2010). The existence of mixed proteasomes (LMP7 only, LMP7 and LPM2) can even further increase peptide variety and influence immunodominance (Guillaume et al., 2010; Zanker et al., 2013).

LMP2‐/‐ mice and MECL‐1‐/‐ mice both show an altered CTL repertoire, however, the MHC class I cell surface expression is not influenced (Basler et al., 2006; Chen et al., 2001; Van Kaer et al., 1994). In contrast, lymphocytes and monocytes in LMP7‐/‐ mice show about a 50% reduction in MHC class I cells surface expression (Fehling et al., 1994). The same phenotype was also observed in mice deficient for all immunoproteasome subunits (Kincaid et al., 2012). The immunological relevance of immunoproteasome subunits was further strengthened by the finding that mice deficient for any of the immunoproteasome subunits are protected from dextran sulphate sodium‐induced colitis (Basler et al., 2010). Indeed, inhibition of LMP7 has been proven effective for the treatment of autoimmune conditions in different mouse models (Basler et al., 2015). However, inhibition of the chymotrypsin‐like activity in hematopoietic cells rather than the catalytic specificity of LMP7 seems to be the underlying mechanism in these mouse models (Basler et al., 2014). p97 The AAA‐ATPase p97 (also called valosin‐containing protein (VCP), Cdc48 in yeast) was originally discovered because of its sheer abundance, indicating its broad cellular role (Peters et al., 1990). Since its discovery 25 years ago, p97 has emerged as a central regulator of the UPS that functionally bridges ubiquitylation and degradation (Meyer et al., 2012; Richly et al., 2005). The principle mechanism of action is the conversion of energy, released through ATP hydrolysis, into a mechanical force that disassembles protein complexes, partially unfolds substrate proteins, or segregates substrates from membranes and other cellular structures 10

1. Introduction

(DeLaBarre and Brunger, 2005; Huang et al., 2012; Kobayashi et al., 2007; Li et al., 2012; Rape et al., 2001; Yeung et al., 2014). This “segregase” function has been linked to many diverse cellular processes including cytosolic protein turnover (Dai and Li, 2001), ER‐associated degradation (ERAD) (Ye et al., 2001), mitochondrial protein degradation (Heo et al., 2010), ubiquitin‐fusion degradation (UFD) (Ghislain et al., 1996), autophagy (Ju et al., 2009), transcription factor regulation (Dai et al., 1998), homotypic membrane fusion (Latterich et al., 1995), cell cycle regulation (Moir and Botstein, 1982), and chromatin‐associated processes (Meerang et al., 2011). In almost all pathways involving p97, the substrates engaged are modified with ubiquitin. Thus, p97 appears to be a ubiquitin‐selective chaperone (Meyer et al., 2002b; Rape et al., 2001; Richly et al., 2005). Although p97 has some affinity to ubiquitin itself, it binds to substrates largely via cofactors containing dedicated ubiquitin‐binding domains (Schuberth and Buchberger, 2008; Ye, 2006). At the same time, p97 serves as a platform for substrate modification by deubiquitylating enzymes and E3/E4 ligases, which “edit” polyubiquitin chains on substrates (Jentsch and Rumpf, 2007). In many cases, the combined segregase and ubiquitin‐editing function of p97 facilitates the degradation of substrates via the proteasome (Buchberger, 2013; Franz et al., 2014) (Fig. 4). The extent of how much a substrate depends on p97 for degradation seems to be determined by substrate localization, structure and solubility (Beskow et al., 2009; Gallagher et al., 2014).

Figure 4: Model for p97 segregase activity and p97‐associated ubiquitin chain editing. A substrate (S) is modified with ubiquitin (orange) by the consecutive action of E1, E2, and E3 enzymes. p97 binds to the modified substrate via cofactors (C) and the energy released through ATP hydrolysis is converted into a mechanical force that remodels substrates and segregates it from binding partners (B) or other cellular structures. Associated substrate‐processing factors, like E4 enzymes or DUBs, may further process the ubiquitin chain and either recycle the substrate or direct it to the proteasome (Pr) for degradation. Modified from (Meyer et al., 2012).

Structurally, p97 can be divided into an N‐terminal domain (N domain), two highly conserved ATPase domains (D1 and D2), flexible linkers connecting all three domains, and an unstructured C‐terminal domain. It builds homohexameric complexes in which the D1 and D2 domains form

1. Introduction two stacked rings with a central pore (Brunger and DeLaBarre, 2003) (Fig. 5). The N domain mediates substrate recognition and binding to almost all cofactors (Dreveny et al., 2004). The D1 ATPase domain mediates oligomerization and confers stability to the complex (Wang et al., 2003). Recent data also suggested that the D1 domain is an active ATPase whose activity is coupled to that of the D2 domain (Chou et al., 2014). In contrast, data obtained by crystallography points towards low ATPase activity at the D1 domain versus strong activity at the D2 domain (Davies et al., 2008). Each of the AAA domains contain conserved nucleotide binding (Walker A) and nucleotide hydrolysis (Walker B) motifs (Ogura and Wilkinson, 2001).

Figure 5: Structure of the hexameric AAA‐ATPase p97. (A) Each subunit consist of a globular N domain, the ATPase domains D1 and D2, and an unstructured C‐terminus. (B and C) Hexameric structure of p97 with N domain (green), D1 ATPase (cyan), and D2 ATPase (blue). The ATPases form two stacked rings with a central pore. The N domain is positioned at the periphery of the D1 domain. (D) Molecular surface of a cross section through the central pore. His317 residues of the D1 domain (green) obstruct the pore. Positively charged Arg586 and Arg599 residues (purple) lining the pore in the D2 domain together with the hydrophobic residues Trp551 and Phe552 (yellow) have strong protein denaturing power. Modified from (Buchberger et al., 2015; Meyer et al., 2012).

ATP hydrolysis drives major conformational changes of the complex creating a mechanical force that is transmitted onto the substrate (Davies et al., 2008; DeLaBarre and Brunger, 2005; Huang et al., 2012; Li et al., 2012; Yeung et al., 2014). Whether a substrate is thereby threaded through the central pore of p97, looping transiently into the pore, or remodeled by other mechanisms is still a matter of debate (Davies et al., 2005; Halawani and Latterich, 2006). Because substrate ubiquitylation generally precedes association with p97, movement through the narrow D1 pore seems to be rather unlikely, unless the substrate would be completely deubiquitylated by DUBs beforehand. For a subset of ERAD substrates it could, nevertheless, be possible (Ernst et al., 2011). The D2 pore is wider and contains twelve Arg residues with a

1. Introduction calculated protein denaturing power of 8 M guanidine. Together with Phe and Trp residues in close proximity, these Arg residues might be responsible for the protein unfolding activity of p97 (DeLaBarre et al., 2006).

Mutations in the gene encoding for p97 are associated with the degenerative disease IBMPFD/ALS (Watts et al., 2004). IBMPFD/ALS refers to the four most common phenotypes that can affect patients solitarily or in any possible combination: inclusion body myopathy (IBM), Paget's disease of the bone (PDB), frontotemporal dementia (FTD), and amyotrophic lateral sclerosis (ALS). Most of the known mutations are located within the N domain or the linker region between N domain and D1 domain (Watts et al., 2004). Several reports could link some of these mutations to a higher basal ATPase activity, while others found cofactor association to be affected (Meyer and Weihl, 2014). A recent study suggests that cofactor binding plays a critical role in controlling ATPase activity and thus would explain both observations (Zhang et al., 2015). On a cellular level, autophagic clearance of aggregated proteins is comprised in affected tissues leading to accumulation of ubiquitin‐positive inclusion bodies (Ju et al., 2009; Ju et al., 2008; Tresse et al., 2010).

Besides the disease‐associated mutations, mutations within the catalytic centers of the D1 and D2 domain are being used to further elucidate p97‐mediated processes. Especially the Walker B mutation E578Q, which disrupts the ATP hydrolysis in the D2 domain, has been studied in various contexts (Song et al., 2003; Ye et al., 2003). p97‐E578Q builds oligomers and is able to interact with cofactors (Dalal et al., 2004; Ye et al., 2003). Nevertheless, cells expressing this mutant accumulate ubiquitylated proteins in the cytoplasm and at the ER membrane indicating a general defect in protein turnover (Dalal et al., 2004; Song et al., 2003; Weihl et al., 2006).

The cofactors associated with p97 can be classified topologically into N domain binding and C‐ terminus binding factors (Buchberger et al., 2015; Yeung et al., 2008). Functionally, cellular localization factors, substrate recruiting factors, and substrate processing factors can be distinguished (Buchberger, 2010; Jentsch and Rumpf, 2007). In mammals, about 40 cofactors have been identified so far. Most of these cofactors bind to the N domain, while only a few interact with the C‐terminal tail. In most cases, the interaction is mediated via a small number of conserved binding motifs (Buchberger et al., 2015) (Fig. 6). Cofactors typically interact with the N domain via ubiquitin‐regulatory X (UBX) or UBX‐L (UBX‐like) domains. Alternatively, the short linear binding motifs VCP‐interacting motif (VIM), VCP‐binding motif (VBM), or SHP box (SH2 containing phosphatase) can mediate binding to p97 (Buchberger et al., 2015; Yeung et al., 2008). In contrast to the N domain, interaction with the C‐terminal tail has only been 13

1. Introduction reported for a few cofactors. Here, the PUB (PNGase/UBA or UBX) domain and the PUL (PLAP, Ufd3p, and Lub1p) domain are implicated to mediate binding to p97 (Doerks et al., 2002; Iyer et al., 2004; Suzuki et al., 2001). Given the high number of cofactors and the participation in many different cellular pathways, it is still not completely understood, how cofactor association is regulated to ensure the formation of productive complexes. The mechanisms proposed are: competition for binding sites, hierarchical binding of major and auxiliary cofactors, and nucleotide dependent binding (Buchberger et al., 2015). Interestingly, protein interaction with the C‐terminus appears to be regulated by phosphorylation at Tyr805 (Li et al., 2008; Zhao et al., 2007).

Figure 6: Domain structure of selected p97 cofactors. UBX or UBX‐like (UBX‐L) domains, or the short motifs VCP‐interaction motif (VIM), VCP‐binding motif (VBM), and SH2 containing phosphatase (SHP) box interact with the p97 N domain while the PUB (PNGase/UBA or UBX) domain binds the C‐terminal tail of p97. Most cofactors also contain ubiquitin‐binding domains like the ubiquitin‐associated (UBA), the NPL4 zinc finger (NZF), the ubiquitin‐interacting motif (UIM) or the coupling of ubiquitin to ER degradation (CUE) domain. The RING domain has E3 ligase activity, whereas OTU (ovarian tumor) and Josephin domains have deubiquitinating activity. The PNGase domain removes carbohydrates. Modified from (Meyer et al., 2012).

The UBX domain closely resembles the structure of ubiquitin, whereas the UBX‐L domain is structurally similar to UBX despite a low sequence similarity (Buchberger et al., 2001; Kim et al., 2014). With 13 mammalian proteins identified, UBX domain proteins represent the largest group of cofactors (Alexandru et al., 2008; Kloppsteck et al., 2012; Schuberth and Buchberger, 2008). A subset of these cofactors also contains a ubiquitin‐associated (UBA) domain which

1. Introduction mediates binding to ubiquitylated substrates. Therefore, these cofactors are classified as substrate recruiting factors (Alexandru et al., 2008; Schuberth et al., 2004). The heterodimeric UFD1 (ubiquitin fusion degradation 1)‐NPL4 (nuclear protein localization homolog 4) complex is one of the best‐characterized substrate recruiting factors and is required for several proteasomal degradation pathways (Meyer et al., 2000; Shcherbik and Haines, 2007). It interacts with p97 via the UBX‐L domain in NPL4 and the SHP box in UFD1 (Yeung et al., 2008). Since only one UFD1‐NPL4 dimer binds per p97 hexamer, further cofactors can bind to this complex (Alexandru et al., 2008; Bruderer et al., 2004). Another important adapter is p47, which is implicated in autophagy and membrane fusion (Kondo et al., 1997; Krick et al., 2010). Similar to the UFD1‐NPL4 complex, p47 uses a bipartite binding mode with a UBX and SHP domain mediating the binding to p97 (Bruderer et al., 2004). Since p47 is a trimeric complex, it occupies all six N domains of p97 (Beuron et al., 2006). Not surprisingly, the two important adapters UFD1‐NPL4 and p47 show mutually exclusive binding to p97 and confine the complex to different cellular pathways (Bruderer et al., 2004; Meyer et al., 2000).

One of the first pathways described explicitly depending on p97 activity is the ERAD pathway. ERAD is a sophisticated quality control mechanism that ensures proper folding of ER‐targeted proteins and removes terminally misfolded or unassembled proteins from the ER for proteasomal degradation. Since the ER is devoid of any component of the UPS, substrates have to be transported back into cytosol in order to be degraded (Hiller et al., 1996). The ERAD machinery can be divided into functionally distinct modules that process substrates sequentially: (1) recognition, (2) retrotranslocation initiation, (3) ubiquitination, (4) extraction, (5) delivery, and (6) degradation (Christianson et al., 2012; Christianson and Ye, 2014). As soon as a protein enters the ER it is subjected to folding and glycosylation at consensus Asn residues. The attached glycans are further processed during maturation and serve as sorting signal or, in case of unsuccessful folding, degradation signal (Helenius and Aebi, 2004). Depending on the location of the misfolded domain, three types of substrates can be distinguished: ERAD‐L (luminal), ERAD‐M (membrane), and ERAD‐C (cytosolic) substrates (Carvalho et al., 2006; Vashist and Ng, 2004). A precise complex moving ERAD substrates across the membrane has not been identified yet. One of the fist candidates proposed was the same channel that is also used for ER entry, namely the Sec61 channel (Pilon et al., 1997). Ever since, contradicting results about the contribution of Sec61 to retrotranslocation have been published (Hampton and Sommer, 2012). Generally, it is believed that large membrane protein complexes assemble around one or several E3 ligases and mediate retrotranslocation (Hampton and Sommer, 2012).

1. Introduction

In yeast, the membrane‐anchored enzymes Hrd1p and Doa10p are the main E3 ligases involved in ERAD (Bays et al., 2001; Swanson et al., 2001). The former ubiquitylates ERAD‐L and ERAD‐ M substrates, whereas the latter targets ERAD‐C substrates (Carvalho et al., 2006). Although Hrd1p and Doa10p are conserved in evolution, mammalian cells use an expanded set of E3 ligases to modify ERAD substrates (Mehnert et al., 2010). Moreover, the substrate‐E3 relationship is not as clear as in yeast with often multiple E3s targeting the same substrate (Christianson et al., 2012; Morito et al., 2008; Younger et al., 2006). Once ERAD substrates are ubiquitylated, the different pathways merge at p97. It is not yet clear, if ubiquitylation precedes p97 recruitment or whether p97 first interacts with the unmodified substrate to prevent backsliding into the membrane (Thoms, 2002; Ye et al., 2003). In any case, p97 is recruited to the ERAD E3 ligases by the membrane anchored adapter proteins UBXD2 and UBXD8 (Liang et al., 2006; Mueller et al., 2008). Moreover, the mammalian E3 ligases HRD1 and gp78 contain p97‐binding motifs themselves and directly recruit p97 (Ballar and Fang, 2008; Morreale et al., 2009). p97 further appears to provide the energy needed for retrotranslocation, especially for the extraction of hydrophobic domains (Garza et al., 2009; Jarosch et al., 2002; Rabinovich et al., 2002; Ye et al., 2001). The substrate recognition of p97 is thereby mediated in most cases by the dimeric UFD1‐NPL4 complex (Flierman et al., 2003; Ye et al., 2003). At the same time, the trimeric p97‐UFD1‐NPL4 complex serves as a platform for DUBs like Ataxin‐3 or YOD1 indicating a dynamic processing of ubiquitin chains (Ernst et al., 2009; Sowa et al., 2009; Wang et al., 2006). Other proposed functions of DUB activity are protection of the retrotranslocation machinery itself (Liu et al., 2014) or even rescue of substrates (Hassink et al., 2009). Besides DUBs, mammalian p97 also associates with peptide:N‐glycanase (PNGase) which removes carbohydrates bound to Asn residues (Allen et al., 2006). Since PNGase in an amidase, Asn residues are also deamidated to Asp residues (Hirsch et al., 2003). Finally, the substrates are handed over to shuttling factors like Ubiquilin‐1, hHR23A or hHR23B (Rad23p in yeast), which deliver substrates to the proteasome for degradation (Lim et al., 2009; Richly et al., 2005).

Recent studies could also link p97 to the extraction of nascent chains from stalled ribosomes (Brandman et al., 2012; Defenouillere et al., 2013; Verma et al., 2013). Stalling of ribosomes can be caused by translation of truncated or non‐stop mRNAs, rare codons, amino acid insufficiency, mRNA secondary structure, or polybasic stretches in the nascent chain (Bengtson and Joazeiro, 2010; Dimitrova et al., 2009; Ito‐Harashima et al., 2007). After stalling, ribosome subunits dissociate and the E3 ligase listerin (LTN1), together with NEMF and RQC1, associates with the 60S ribosomal subunit in order to ubiquitylate the nascent chain (Bengtson and

1. Introduction

Joazeiro, 2010; Shao et al., 2013). Subsequently, the ribosome quality control complex (RQC) consisting of LTN1, NEMF, RQC1, and p97 extracts the nascent chain (Brandman et al., 2012; Defenouillere et al., 2013; Verma et al., 2013) (Fig. 7). The nascent chain is then degraded via the UPS, whereas the 60S ribosomal subunit is recycled.

Figure7: Model of RQC‐mediated degradation of nascent chains. 80S ribosomes can stall due to a polybasic substrate (++++) interacting with the ribosomal exit tunnel. The ribosomal subunits dissociate and the E3 ligase LTN1, NEMF, and RQC1 are recruited to the 60S ribosomal subunit. LTN1 ubiquitylates the nascent chain, which further recruits p97 together with its cofactors UFD1‐NPL4. The nascent chain is extracted from the ribosomal subunit and degraded via the proteasome. Modified from (Brandman et al., 2012).

In 2004, the first small molecule inhibitor of p97, eeyarestatin I (EerI), was described (Fiebiger et al., 2004). EerI was identified in a high throughput screen for molecules inhibiting the ERAD pathway. Further studies confirmed that EerI inhibits dislocation from the ER membrane, turnover of cytosolic proteins, and p97‐associated DUB activity while proteasomal activity is unaffected (Wang et al., 2008; Wang et al., 2010). However, some off‐target effects are likely since EerI also affects Sec61‐mediated protein translocation into the ER, protein synthesis, and vesicular transport (Aletrari et al., 2011; Cross et al., 2009a). Wang et al. described EerI as a bifunctional molecule whose aromatic domain leads to membrane targeting while the nitrofuran moiety binds irreversibly to p97, potentially at the D1 domain (Wang et al., 2010).

Differences in IC50 values obtained with in vitro (IC50 = 70 μM) and cellular experiments (IC50 = 8 μM) further suggest that the molecule has to be metabolized in order to inhibit p97 (Cross et al., 2009a; Wang et al., 2008). Longer incubation times with EerI lead to the activation of the unfolded protein response and finally apoptosis (Wang et al., 2009). This effect is even enhanced when combined with proteasome inhibitors and might be a promising strategy for cancer therapy (Auner et al., 2013; Wang et al., 2009). Overall, EerI is still an inhibitor with a poorly defined mechanism but a powerful tool to study p97‐mediated processes.

1. Introduction

The first reversible and selective small molecule p97 inhibitor N2,N4‐dibenzylquinazoline‐2,4‐ diamine (DBeQ) was identified in a high throughput screen using an ATPase assay (Chou et al., 2011). Chou et al. could show that DBeQ inhibited degradation of UFD and ERAD substrates whereas degradation of a p97‐independent substrate was not affected. Moreover, it inhibited a second major pathway of cellular protein degradation by interfering with the maturation of autophagosomes. The mechanism of action of DBeQ is competing with ATP binding at both

ATPase domains (Chou et al., 2014). Although DBeQ has a low IC50 value in in vitro ATPase activity assays (IC50 = 1.5 μM) and in cellular degradation assays (IC50 = 2.6 μM), binding of the cofactor p47 to p97 significantly reduced the potency (Chou et al., 2011; Fang et al., 2015). These results indicate that the potency of DBeQ might vary between different p97 complexes. Similar to EerI, treatment with DBeQ induces the unfolded protein response and caspase‐9 mediated apoptosis (Chou et al., 2011).

BAG6 The chaperone BCL2‐associated athanogene 6 (BAG6) (also termed HLA‐B‐associated transcript 3 (BAT3) or Scythe) is conserved in higher eukaryotes, ubiquitously expressed, and encoded in the MHC class III locus (Banerji et al., 1990; Ozaki et al., 1999; Wang and Liew, 1994). The first described cellular function of BAG6 was the regulation of cell death (Thress et al., 1999; Thress et al., 1998). BAG6 is able to sequester pro‐apoptotic factors or promote their degradation via the proteasome and thus seems to be an anti‐apoptotic regulator (Colon‐Ramos et al., 2003; Minami et al., 2007; Thress et al., 1999). Indeed, genetic ablation of BAG6 causes embryonic lethality resulting from developmental defects associated with increased apoptosis and aberrant cell proliferation (Desmots et al., 2005). Interestingly, BAG6 itself contains a caspase‐ 3 cleavage site, which might be important for cell death induced by certain pathogenic bacteria (Grover and Izzo, 2012; Wu et al., 2004). Hence, BAG6 can potentially do both, inhibit apoptosis or contribute to it. A recent study could even extend the anti‐apoptotic activity of BAG6 to a T cell‐specific pathway, where it prevented apoptosis due to exhaustion (Rangachari et al., 2012).

The major isoform of the human bag6 gene encodes a protein of 1,132 aa containing an N‐ terminal ubiquitin‐like (UBL) domain, an intrinsically disordered, proline‐rich region, a zinc finger‐like domain, and at the C‐terminus a nuclear localization sequence (NLS) together with a conserved BAG domain (Banerji et al., 1990; Doong et al., 2002; Manchen and Hubberstey, 2001) (Fig. 8). The BAG domain qualifies BAG6 as a member of the BAG family, which contains five more members apart from BAG6 (Doong et al., 2002). The consensus BAG domain consists three anti‐parallel α‐helices (Briknarova et al., 2001; Sondermann et al., 2001). The second and 18

1. Introduction third helix can bind to the ATPase domain of heat shock protein 70 (HSP70)/heat shock cognate 70 (HSC70) and either positively or negatively influence its folding activity (Takayama and Reed, 2001). Therefore, BAG proteins can also be classified as co‐chaperones. It was indeed shown that purified BAG6 inhibits HSP70‐mediated folding in vitro (Thress et al., 2001). However, two recently published studies stirred up some controversy about the structure and function of the BAG domain in BAG6. The combined biochemical and structural data obtained indicated that BAG6 does not contain a canonical BAG domain and that the main function of the C‐terminus is the binding to cofactors (Kuwabara et al., 2015; Mock et al., 2015). Experiments of Leznicki et al. suggested that the central proline‐rich region of BAG6 mediates binding to substrates (Leznicki et al., 2013). The central region is also sufficient to keep denatured luciferase in a soluble state, which further supports a role in substrate binding (Xu et al., 2013). Despite the NLS in BAG6 (Manchen and Hubberstey, 2001), it is primarily found in the cytoplasm of cells. In part, this is due to the cofactor transmembrane domain recognition complex 35 (TRC35) masking the NLS in BAG6 (Wang et al., 2011). Moreover, BAG6 isoforms lacking an NLS can be generated by alternative splicing (Kamper et al., 2012b). Finally, the UBL domain found at the N‐terminus indicates a role of BAG6 in proteasome‐mediated protein degradation (Banerji et al., 1990).

Figure 8: The domain structure of human BAG6. UBL = ubiquitin‐like, Pro‐rich = proline‐rich, NLS = nuclear localization sequence.

Long after BAG6 had been linked to apoptosis, its role in protein synthesis and quality control was recognized (Fig. 9). Minami et al. could show for the first time that BAG6 associates with a model substrate carrying a hydrophobic degradation signal (Minami et al., 2010). Knockdown of BAG6 inhibited the turnover of this substrate by the proteasome. Moreover, ubiquitylated proteins and subunits of the 26S proteasome co‐immunoprecipitated with BAG6 if cells were treated with proteasome inhibitors. Together, these results indicated a functional requirement of BAG6 in proteasomal degradation.

Next, BAG6 could be identified as factor involved in the biogenesis of tail‐anchored (TA) proteins (Leznicki et al., 2010; Mariappan et al., 2010). TA proteins have a single C‐terminal transmembrane domain (TMD) and are post‐translationally inserted into the ER membrane

1. Introduction

(Rabu et al., 2009). In order to prevent aggregation or inappropriate interactions, the hydrophobic TMD requires constant chaperoning until inserted in the membrane. The heterotrimeric complex consisting of BAG6 and its cofactors UBL4A and TRC35 associates with TMDs directly at the ribosome. Subsequently, the nascent chain is handed over to the dimeric chaperone TRC40 (transmembrane domain recognition complex of ~40 kDa) (Mariappan et al., 2010). Upon substrate binding, TRC40 adopts an ATP‐bound “closed” state allowing the transport of substrate to the ER membrane. Interaction with its membrane receptors leads to an ATP‐dependent release of TA proteins and membrane insertion (Favaloro et al., 2010). Besides regulating substrate loading onto TRC40, the BAG6 complex could also regulate the nucleotide cycle. While BAG6 is most likely not essential for TA protein targeting, it potentially increases efficiency and fidelity.

The heterotrimeric BAG6 complex is not only involved in the biogenesis of TA proteins, it also mediates degradation of mislocalized proteins (Hessa et al., 2011). Depending on the type of polypeptide, insertion into the ER is either accomplished in a co‐translational manner via the Sec61 channel or via the post‐translational TRC40 pathway (Cross et al., 2009b). However, protein import into the ER is not always perfectly efficient, resulting in at least some polypeptides being mislocalized in the cytoplasm (Levine et al., 2005). Most importantly, mislocalization can cause disease, which emphasizes the importance of an efficient quality control pathway (Zimmermann et al., 2006). The BAG6 complex recognizes exposed hydrophobic TMDs, prevents aggregation or inappropriate interactions, and further regulates protein triaging (Hessa et al., 2011). BAG6‐associated polypeptides are either targeted to the ER (TA proteins, via TRC40), or channeled into proteasomal degradation. The BAG6 complex is necessary for maximum ubiquitylation of mislocalized proteins, which points towards a BAG6‐ associated E3 ligase. Indeed, a recent study identified RNF126 as a BAG6‐dependent E3 ligase associating with the UBL domain of BAG6 (Rodrigo‐Brenni et al., 2014). Depletion of RNF126 stabilized a BAG6 client protein indicating that RNF126 is the main E3 ligase in this pathway.

Finally, BAG6 could also be linked to the proteasomal degradation of ERAD substrates. BAG6 was first found to co‐purify with deglycosylated ERAD substrates after enzymatic blockage of the proteasome (Ernst et al., 2011). Wang et al. could subsequently confirm that the BAG6‐ TRC35‐UBL4A complex increases ERAD efficiency by keeping aggregation‐prone substrates in a soluble state, which allows proteasomal degradation (Wang et al., 2011). Recruitment of BAG6 to the ER membrane can likely be mediated by several different interaction partners, since BAG6 can associate with the E3 ligases gp78 and HRD1 (Wang et al., 2011), with the putative 20

1. Introduction channel protein Derlin2 (Claessen and Ploegh, 2011), and the adapter protein UBXD8 (Xu et al., 2013). The exact binding partners could potentially vary and depend on the ERAD substrate and the precise retrotranslocation machinery involved.

Figure 9: The role of BAG6 in protein synthesis and quality control. Misfolded proteins are removed from the ER in a process known as ER‐associated degradation (ERAD), which involves E3 ligases and the chaperone p97 (1). BAG6 is recruited to the ERAD machinery, associates with hydrophobic patches (orange) in the ERAD substrate to keep it in a soluble state (2) and potentially shuttle it to the proteasome for degradation (3). Directly at the ribosome, BAG6 can also associate with exposed hydrophobic domains of nascent chains (4). BAG6 can then regulate protein triaging that either results in protein degradation or insertion into the ER membrane (5). Mislocalized polypeptides can be ubiquitylated by the BAG6‐associated E3 ligase RNF126 and subsequently be degraded by the proteasome. In case of transmembrane‐anchored proteins, BAG6 hands over its cargo to TRC40 for post‐translational insertion into the ER membrane (6). Ub(n) = polyubiquitin.

Taken together, BAG6 and its binding partners TRC35 and UBL4 associate with long hydrophobic patches in client proteins and prevent aggregation (Leznicki et al., 2013; Mariappan et al., 2010; Minami et al., 2010; Wang et al., 2011). Even though BAG6 is classified as a chaperone, it has no apparent folding activity (Wang et al., 2011). Therefore, BAG6 has also been described as “holdase” which keeps its substrate in a soluble state (Wang et al., 2011). Being able to furthermore interact with ribosomes and proteasomes (Minami et al., 2010), BAG6 truly is a factor acting at the interface of protein biogenesis and degradation (Kawahara et al., 2013; Lee and Ye, 2013).

1. Introduction

1.4 Antigen processing and presentation on MHC class I The MHC class I‐restricted pathway Generally, antigen processing and presentation is accomplished by a multi‐step process, which results in the generation and presentation of peptide‐MHC‐class‐I complexes. Intracellular viral and bacterial proteins as well as endogenous proteins are degraded to peptides of 8‐10 aa, bound to MHC class I molecules, and presented to CTLs on the cell surface (Rock and Goldberg, 1999; York et al., 1999) (Fig. 10).

Figure 10: Antigen processing and presentation on MHC class I. The main steps of antigen processing are translation (1), polyubiquitylation (2), proteasomal degradation (3), transport into the ER via transporter associated with antigen processing (TAP) (4), loading onto MHC class I with the help of the peptide‐loading complex (5), and transport to the cells surface (6). Modified from (Vyas et al., 2008).

MHC class I molecules are heterodimeric membrane complexes consisting of the MHC class I heavy chain (α‐chain), a β2‐microglobulin (β2m), and a short peptide (Bijlmakers and Ploegh, 1993; Madden, 1995). There are three class I α‐chains encoded in highly polymorphic genes (human leukocyte antigen (HLA)‐A, ‐B, and ‐C in humans; histocompatibility‐2 (H‐2) K, D, and L in mice). The α‐chain consists of the three domains α1, α2, and α3. While α3 mediates interaction with the CD8 co‐receptor of the TCR, α1 and α2 together form the peptide‐binding groove and display their cargo to TCRs. The highest polymorphism is observed among the

1. Introduction residues forming the peptide‐binding groove allowing each molecule to present a different set of peptides.

Proteasomal degradation results in peptides ranging from about 2‐25 aa in size (Kisselev et al., 1999), which can be further trimmed or destroyed by cytosolic peptidases (Lazaro et al., 2015). Peptides can be selectively transported into the ER via the transporter associated with antigen processing (TAP) in an ATP‐dependent manner (Neefjes et al., 1993). TAP shows the highest transport efficiency for peptides of 8‐10 aa with a hydrophobic C‐terminus and thus for peptides ideal for binding to MHC class I molecules (Momburg and Hammerling, 1998). Peptides are loaded onto MHC class I molecules with the help of the MHC class I loading complex consisting of TAP, the chaperones calreticulin and tapasin, the thiol oxidoreductase

ERp57, as well as the α‐chain/β2m dimer (Wearsch and Cresswell, 2008). Tapasin mediates peptide “editing” and ensures binding of a high‐affinity peptide (Praveen et al., 2010). N‐ terminally extended peptides can be further trimmed by ER aminopeptidases (ERAP1 and ERAP2 in humans; ERAAP in mice) until optimal length is reached (Saric et al., 2002; Serwold et al., 2002). Moreover, C‐terminal trimming is possible although potentially less frequent (Shen et al., 2011). Peptide‐loaded MHC class I molecules are subsequently released from the peptide loading complex and transported to the cell surface via the Golgi complex.

The source of antigenic peptides The peptide array displayed by MHC class I molecules (also termed immunopeptidome) reflects the physiological state of the cell, which is monitored by CTLs (Caron et al., 2011). The source of antigenic peptides can be any cellular protein, including proteins of pathogenic origin (Hunt et al., 1992; Yewdell and Bennink, 1992). One of the first factors identified to play a major role in antigen processing was the proteasome. Since it degrades the bulk of cellular proteins, the immune system takes advantage of this system by using the peptides it produces for presentation. By the use of proteasome inhibitors it was confirmed that MHC class I cell surface expression and peptide‐presentation strongly depend on proteasomal activity (Cerundolo et al., 1997; Harding et al., 1995; Rock et al., 1994). The fact that conventional and immunoproteasomes imprint differently on the peptide repertoire further underscores the importance of the proteasome for antigen processing (Vigneron and Van den Eynde, 2014). While the majority of antigenic peptides produced by the proteasome are likely to be derived from ubiquitylated polypeptides, several studies also support an important role for ubiquitin‐ independent antigen processing by the proteasome (Cox et al., 1995; Huang et al., 2011; Qian et al., 2006). Apart from the proteasome, other cellular pathways can contribute to the 23

1. Introduction immunopeptidome (Oliveira and van Hall, 2015). Although most peptidases in the cytosol and secretory pathway cooperate with the proteasome in antigen processing, a few are implicated in proteasome‐independent antigen processing (Lazaro et al., 2015). The peptides presented can also derive from leader sequences cleaved off from ER‐targeted polypeptides (El Hage et al., 2008; Henderson et al., 1992; Wolfel et al., 2000). Moreover, autophagy, induced under various stress conditions, can be a source of antigenic peptides (Demachi‐Okamura et al., 2012; English et al., 2009; Tey and Khanna, 2012).

Initially, antigenic peptides were expected to derive from the turnover of senescent proteins according to their half‐life, until it was discovered that blocking translation affected the presentation on MHC class I instantly and greatly (Jensen, 1988; Reits et al., 2000; Schubert et al., 2000). Essentially the same result could be obtained using tetracyclin‐regulated expression systems (Fiebiger et al., 2012; Khan et al., 2001a). Moreover, it could be observed that, despite the long half‐life of mature viral proteins, virus‐derived peptides are rapidly presented after the onset of expression (Croft et al., 2013; Khan et al., 2001a; Probst et al., 2003). Together, these results indicate a direct connection between protein synthesis and antigen presentation and imply the existence of a fraction of rapidly degraded proteins (RDPs). In order to explain this phenomenon, Yewdell postulated the defective ribosomal product (DRiP) hypothesis (Yewdell et al., 1996). DRiPs are polypeptides that are unable to achieve functionality due to errors occurring during transcription, translation, targeting, folding, or oligomer assembly and as a consequence, are rapidly degraded. Using DRiPs as a source of antigenic peptides would enable cells to reflect protein synthesis rates rather than protein concentration on their surface. This becomes eminently advantageous during infections with viruses, which tend to hijack most of a cell’s translation machinery (Komarova et al., 2009). Although RDPs are an important source of antigen, senescent proteins and proteins that have gained a functional state can still contribute to antigen presentation (Colbert et al., 2013; Dolan et al., 2011; Farfan‐ Arribas et al., 2012).

In an attempt to quantify RDPs, initial studies suggested that up to 30% of all proteins are rapidly degraded (Schubert et al., 2000; Wheatley et al., 1980). Later, this number was revised to about 25% (Princiotta et al., 2003; Qian et al., 2006; Shaffer et al., 2004) or even found to only make up a few percent at all (Bulik et al., 2005; Vabulas and Hartl, 2005). In a more recent report, RDP fractions varied between 1% and 30% depending on the cell line and the activation of cells (Cenci et al., 2012). These studies reflect the difficulty to quantify rapidly degraded substrates since this proportion is likely to vary greatly between different proteins, cell types, 24

1. Introduction and physiological states of cells. An important factor influencing the proportion of RPSs is most likely the overall availability of cellular chaperones (Duttler et al., 2013; Qian et al., 2006). In an approach to further characterize the mechanisms behind RDP degradation, Wang et al. could show that up to 15% of nascent chains were co‐translationally ubiquitylated with Lys48‐linked chains (Wang et al., 2013). Moreover, the proportion of modified nascent chains increased in response to agents inducing misfolding or ribosome stalling.

In order to characterize the peptide ligands presented by MHC class I molecules, an increasing number of studies relies on large‐scale mass spectrometry‐based data sets. The combined results from these analyses point towards an immunopeptidome that is plastic and influenced by cell‐intrinsic and extrinsic factors (Granados et al., 2015). Moreover, these studies show indeed that there is only a limited overlap between the proteome and the immmunopeptidome (Berlin et al., 2015; Croft et al., 2013; Goodenough et al., 2014; Hassan et al., 2013; Milner et al., 2006).

Although the importance of RDPs as a source of antigen is well accepted the cause for rapid degradation are not yet completely understood. Mechanisms proposed to contribute RDPs include premature termination of translation (Cardinaud et al., 2010; Lacsina et al., 2012; Schwab et al., 2003), downstream initiation (Berglund et al., 2007), destabilization of mRNAs by miRNAs (Granados et al., 2012) or shRNA (Gu et al., 2009), surplus oligomer subunits (Bassani‐Sternberg et al., 2015; Bourdetsky et al., 2014), out‐of‐frame translation (Bullock and Eisenlohr, 1996; Malarkannan et al., 1999), and degradation of polypeptides produced during the pioneer round of translation (Apcher et al., 2015). While all of the aforementioned mechanisms are certainly able to produce substrates for antigen processing their relative contribution and significance still remains mostly elusive. Moreover, mere accessibility of nascent chains to proteasomal degradation rather than defectiveness could render newly synthesized polypeptides as the predominant source of antigen presentation (Rock et al., 2014).

1.4 NF‐κB activation Nuclear factor‐kB (NF‐kB) is a central pro‐inflammatory transcription factor, which is ubiquitously expressed and a master switch in the initiation and maintenance of inflammation. In unstimulated cells, NF‐kB is sequestered in the cytoplasm as it is bound to inhibitors of kB (IkBs) (Hinz et al., 2012). Extensive signal transduction cascades integrate extracellular and/or intracellular signals ultimately leading to phosphorylation, polyubiquitylation, and proteasomal

1. Introduction degradation of IkBs (Hayden and Ghosh, 2008). Free NF‐kB then migrates into the nucleus where it binds to consensus sites in promotors of numerous genes (Brasier, 2006). Generally, NF‐kB signaling pathways are subdivided into two distinct pathways, i.e. canonical and noncanoncial/alternative pathway.

Figure 11: Domain structure of the mammalian NF‐kB transcription factor family. All family members share the Rel homology domain (RHD) and contain a nuclear localization sequence (NLS). The N‐terminal part of the RHD mediates DNA binding while the C‐terminal part is responsible for the formation of dimers and binding to IkBs. Only RelA, c‐Rel and RelB contain a C‐terminal transactivation domain (TAD). The p50 precursor p105 and the p52 precursor p100 contain ankyrin repeats, which bind to Rel proteins. Ankyrin repeats mask the NLS and thereby retain complexes in the cytosol. GRR (glycine‐rich region), LZ (leucine zipper). Modified from (Chen and Greene, 2004).

The mammalian NF‐kB transcription factor family consists of the five members p65 (RelA), RelB, c‐Rel, p105/p50, and p100/p52 that all share the N‐terminal Rel homology domain (RHD) (Baldwin, 1996; Ghosh et al., 1998) (Fig. 11). Distinct segments of the RHD mediate homo‐ and heterodimerization of the NF‐kB family members, binding to IkBs, nuclear translocation, and DNA binding. Among the different dimeric complexes formed between the Rel proteins, p65/p50 is by far the most abundant complex found in almost all types of cells. Only three of the five NF‐kB family members, namely p65, RelB and c‐Rel, contain a C‐terminal transactivation domain (TAD). The other two members, p50 and p52, are produced via limited proteolysis from the precursor proteins p105 and p100, respectively. The two precursor proteins contain the RHD at the C‐terminus, while the N‐terminal half harbors multiple copies of ankyrin repeats, which are characteristic for the IkB family.

In the cytoplasm of resting cells, NF‐kB dimers are associated with one of the prototypical IkB proteins IkBα, IkBβ, or IkBε, or the precursor proteins p100 and p105 (Hinz et al., 2012). IkB proteins not only prevent nuclear translocation of bound NF‐kB dimers, they also determine the localization within the cytosol. The interaction between IkB proteins and the RHD of NF‐kB

1. Introduction is mediated by six to seven ankyrin repeats. As individual IkB proteins seem to preferentially associate with particular dimers, e.g. p65/p50 is often associated with IkBα. IkBα is the best‐ studied member of the IkB family and its degradation can be triggered by many typical pro‐ inflammatory stimuli like Toll‐like receptor (TLR) ligands or cytokines (Hayden and Ghosh, 2008; Vallabhapurapu and Karin, 2009). Due to strong NF‐kB‐mediated transcriptional induction of IkBα, it forms part of a negative feedback loop (Brown et al., 1993; Sun et al., 1993).

Figure 12: The canonical NF‐kB signaling pathway. Under resting conditions, NF‐kB dimers are sequestered in the cytoplasm through association with IB proteins. Pro‐inflammatory signals stimulate receptors like the tumor necrosis factor receptor (TNFR) or Toll‐like receptors (TLR). Signaling cascades originating from engaged receptors activate the IkB kinase (IKK) complex consisting of the subunits IKKα and IKKβ and the NF‐ kB essential modulator (NEMO). The IKK complex phosphorylates IB proteins on specific serine residues, thereby triggering their polyubiquitylation (Ub) and proteasome‐dependent degradation. Free NF‐kB dimers then migrate into the nucleus, bind to promotors of NF‐kB‐responsive genes, and initiate gene transcription. Modified from (Pasparakis, 2009).

A key step in the canonical NF‐kB signaling pathway is activation of the inhibitor of kB kinase (IKK) complex (Fig. 12). The IKK complex is a serine‐specific IkB kinase consisting of the catalytic subunits IKKα and IKKβ together with the regulatory subunit NF‐kB essential modulator (NEMO, or IKKγ) (Hinz and Scheidereit, 2014). Activation of the kinase complex depends on the phosphorylation of key serine residues in the kinase subunits through autophosphorylation or

1. Introduction by an upstream kinase. Once activated, the IKK complex phosphorylates IkBα at Ser32 and Ser36. Phosphorylated IkBα is then recognized and ubiquitylated by the SCF‐β‐TrCp complex and subsequently degraded by the proteasome (Kanarek and Ben‐Neriah, 2012). The free NF‐ kB dimer translocates into the nucleus and activates gene transcription.

The canonical pathway can be activated by a variety of signals including cytokines, pathogens, stress signals, and radiation (Gilmore, 2008). Generally, receptor proximal signaling adapter molecules initiate a signaling cascade ultimately leading to IKK activation (Hayden and Ghosh, 2008). Non‐degradable ubiquitin chains are involved as scaffold molecules at several signaling steps and are critical for IKK complex activation (Iwai, 2012). Therefore, DUBs play an important role in the termination of NF‐kB signaling (Harhaj and Dixit, 2012).

NF‐kB is a central mediator of immune responses and coordinates both, the innate and adaptive arm. Consistent with the variety of stimuli triggering NF‐kB activation, NF‐kB controls expression of a large number of diverse genes (Pahl, 1999). Among the target genes are immunregulatory proteins like cytokines, chemokines and their receptors, regulators of apoptosis and proliferation, as well as negative regulators of the NF‐kB response.

1.5 Virus strains Lymphocytic choriomeningitis virus Lymphocytic choriomeningitis virus (LCMV) belongs to the family of Arenaviridae and is a well‐ characterized model for studying various aspects of the antiviral immune response as well as antigen presentation on MHC class I. The natural host of LCMV is the mouse, however, infection of other mammals, including humans, is possible. LCMV is an enveloped negative‐strand RNA virus, which noncytopathically buds from infected cells. Hence, pathogenic effects are a sole consequence of the antiviral immune response.

The viral genome encodes four proteins on two ambisense RNA sequences, a 3.4 kb short (S) RNA and a 7.2 kb long (L) RNA (Meyer et al., 2002a; Salvato and Shimomaye, 1989). The S RNA encodes the 75 kDa glycoprotein (GP) precursor and the 63 kDa nucleoprotein (NP) (Riviere et al., 1985). The L RNA encodes a 200 kDa virus‐specific RNA polymerase and the 11 kDa RING finger protein (Salvato and Shimomaye, 1989). Processing of the GP precursor results in an ER signal peptide as well as the glycoproteins GP‐1 (44 kDa) and the GP‐2 (35 kDa), which cover the virion envelope as spikes (Borrow, 1997).

1. Introduction

Intravenous (i.v.) infection of C57BL/6 mice with a low dose of LCMV (200 PFU) induces a strong CTL response mainly dominated by four epitopes: GP‐derived GP33‐41 (H‐2Db), GP34‐41 (H‐ 2Kb), GP276‐286 (H‐2Db) and NP‐derived NP396‐404 (H‐2Db). CTL responses in BALB/c mice are strongly dominated by NP118‐126 (H‐2Ld) (Gallimore et al., 1998; Sourdive et al., 1998; van der Most et al., 1998; van der Most et al., 1996). Despite the metabolic stability of NP, presentation of NP‐derived epitopes depends on neosynthesis and thus, NP can be classified as DRiP substrate (Khan et al., 2001a). Moreover, faster presentation of NP‐derived epitopes compared to GP‐derived epitopes was revealed in kinetic studies in vivo (Probst et al., 2003).

Vaccinia virus Vaccinia virus (VV) belongs to the Poxviridaea, a family of large DNA viruses replicating in the cytoplasm of susceptible host cells (Moss, 2006). After replicating, the majority of infectious virus particles remains in the host cell until released through cell lysis. VV was formerly used in vaccinations against variola virus, the causative agent of smallpox. Today, VV is used as a heterologous expression vector in immunological animal studies and tissue cultures (Moss, 1996).

The doublestranded DNA genome of VV is approximately 190 kb in length and encodes for about 200 genes (Goebel et al., 1990). The large genome of VV allows the insertion of up to 25 kb of heterologous DNA via homologous recombination in vivo (recombinant VV, rVV) (Mackett and Smith, 1986). Since rVV‐based expression allows de novo synthesis of antigen in many different cell types, it became a useful tool in antigen processing and presentation research. Moreover, infection of mice with rVV can induce CTL responses directed against peptides derived from heterologous antigens (Moss, 2011). While this made rVV a useful vector to study vaccination strategies on the one hand, it also allows the generation of CTL lines used in antigen presentation assays.

Aim of this study The aim of the present study was to elucidate the influence of the chaperones p97 and BAG6 on antigen processing and presentation on MHC class I molecules. Both factors build higher order complexes with cofactors and are implicated in the turnover of a broad spectrum of defective nascent polypeptides. Hence, an involvement in antigen degradation seems likely.

Previous studies concerned with the role of p97 in MHC class I‐restricted antigen processing produced controversial results (Ackerman et al., 2006; Dolan et al., 2011; Palmer and Dolan, 2013). However, these studies are either based on very limited data sets or on unconvincing 29

1. Introduction experimental setups. Here, a combined pharmacological and genetic approach was used to test the influence of p97 activity on MHC class I surface expression and presentation of virus‐ derived and endogenous peptides. To identify the potential mechanism behind p97 activity in antigen processing, the influence of the RQC complex, whose substrates depend on p97 for degradation, was tested.

Minami et al. not only found BAG6 associated with chemically induced DRiPs, they also recorded reduced MHC class I surface expression after BAG6 knockdown in HeLa cells (Minami et al., 2010). In order to test the hypothesis of BAG6 being an essential factor in DRiP degradation, bulk MHC class I surface expression and presentation of virus‐derived peptides was analyzed in cells transfected with BAG6 siRNA. Furthermore, the influence of IFN‐γ stimulation on BAG6 expression was tested. The cytokine IFN‐γ is critical for the anti‐viral immune response and known to induce many genes involved in antigen processing and presentation (Le Page et al., 2000).

This study is further concerned with the influence of immunoproteasome subunits on the activation of the inflammatory transcription factor NF‐kB. Proteasomal activity is crucial for NF‐ kB precursor processing and signal induced degradation of IkBs (Palombella et al., 1994). However, the influence of immunoproteasome expression on these processes is controversial (McCarthy and Weinberg, 2015). Here, activation of NF‐kB in cells derived from immunoproteasome‐deficient mice was analyzed. The signal transduction along the canonical pathway was monitored at three different stages, namely degradation of IkBα, nuclear translocation of p65/p50 dimers, and transactivation of target genes.

2. Material and Methods

2.1 Molecular cloning and plasmid preparation The DNA fragment encoding human Tyrosinase (Tyr) was amplified from a pre‐existing vector using oligonucleotides including AsiSI or Mlu‐I restriction sites (5’‐ GCGGGGCGATCGCCATGCTCCTGGCTG‐3’, 5’CGCCCACGCGTTATAAATGGCTCTGATAC‐3’) and Phusion High‐Fidelity DNA Polymerase (NEB) according to the manufacturer’s instruction. The amplified fragment was purified via agarose gel electrophoresis and the NucleoSpin Extract II kit (Macherey‐Nagel). The purified DNA fragment and the target vector pCMV6‐Entry (OriGene Technologies) were both digested with AsiSI and Mlu‐I (NEB) according to the manufacturer’s instruction and again purified via agarose gel electrophoresis. Ligation was performed in a molar ratio of 1:3 of vector to insert using T4 DNA ligase (NEB) according to the manufacturer’s instruction. The ligation reaction was carefully stirred into 50 µl of chemically competent E.coli TOP10F’ (originally obtained from Invitrogen) and incubated on ice for 30 min. Bacteria were then heat shocked for 30 sec at 42°C, cooled on ice for several minutes, and mixed with 300 µl LB medium (1% (w/w) tryptone, 1% (w/w) NaCl, 0.5% yeast extract). Antibiotic resistance of transformed bacteria was allowed to develop for 1 h at 37°C while shaking. Bacteria were plated on LB agar plates (2% (w/w) agar in LB medium) containing 25 µg/ml kanamycin and incubated overnight at 37°C. For sequence analysis, 5 ml LB medium containing kanamycin were inoculated with a single bacterial colony picked from LB‐agar plates. Cultures were grown overnight at 37°C and 150 rpm and plasmid DNA was extracted using the NucleoSpin Extract II kit (Macherey‐Nagel). Plasmids were analyzed by restriction digest and agarose gel electrophoresis. DNA sequences were further analyzed by sequencing at GATC Biotech AG (Köln, Germany).

Plasmid DNA used for transfections was prepared from overnight cultures in 150 ml LB medium containing appropriate antibiotics. Cultures were incubated at 37°C and 150 rpm. Plasmid DNA was extracted using the NucleoBond PC 500 kit (Macherey‐Nagel).

2.2 Agarose gel electrophoresis DNA or RNA was analyzed or purified via agarose gel electrophoresis. Depending on the size of the construct to be analyzed, gels with 1‐2% (w/v) agarose in TAE buffer (40 mM Tris‐acetate, 40 mM acetic acid, 2 mM EDTA, pH8.3) containing 0.5 µg/ml ethidium bromide were used. Samples were mixed with 6x DNA loading dye (30% (v/v) glycerol, 0.25% (w/v) bromphenol

2. Material and Methods blue, 0.25% (w/v) xylene cyanol) and loaded onto the gel together with an appropriate DNA ladder (Eurogenetic). Electrophoretic separation was performed at 90 V for approximately 45 min. DNA was visualized by UV light.

2.3 Mice C57BL/6 mice (H‐2b), BALB/c mice (H‐2d), and AAD mice (H‐2b, HLA‐A2.1/H2‐Dd) were originally purchased from Charles River. LMP2 (Van Kaer et al., 1994), LMP7 (Fehling et al., 1994) and MECL‐1 (Basler et al., 2006) gene‐targeted mice were provided by J. Monaco (Cincinnati, OH, USA). LMP7‐/‐/MECL‐1‐/‐ double knockout mice (L7M‐/‐) were generated by crossing the F1 generation of LMP7‐/‐ with MECL‐1‐/‐ mice. Mice were kept in a specific pathogen‐free facility and used at 8–12 weeks of age. Animal experiments were approved by the review board of Regierungspräsidium Freiburg.

2.4 Virus amplification and titer determination LCMV strain WE (LCMV‐WE) LCMV‐WE was originally obtained from F. Lehmann‐Grube (Hamburg, Germany) and propagated on the fibroblast line L929. At 80% confluence, L929 cells were infected with LCMV with a multiplicity of infection (MOI) of 0.01. Supernatant containing virus particles was harvested on day two post infection and kept at ‐80°C in aliquots. To determine virus titers, a tenfold serial dilution of each batch was prepared and transferred onto monolayers of MC57 cells. Infected cell were detected after 48 h of incubation at 37°C and 5% CO2 by immunofocus assay using the LCMV‐NP‐specific monoclonal antibody VL‐4 as previously described (Battegay et al., 1991).

VV Western Reserve strain (VV‐WR) and rVV strains VV‐WR was originally obtained from H. Hengartner (University Hospital Zürich, Switzerland). rVV expressing full‐length LCMV‐NP and LCMV‐GP were a kind gift from D. Bishop (Oxford, UK) while the virus strains rVV‐NP118, rVV‐NP396, and rVV‐GP276 expressing LCMV‐derived epitopes as minigenes as well as rVV‐eGFP expressing enhanced green fluorescent protein (eGFP) were originally constructed in the laboratory of M. van den Broek (Zürich, Switzerland). Virus strains expressing a mouse male minor antigen, rVV‐UTY or a UTY‐derived epitope as minigene, rVV‐UTY246 were a kind gift from V. Cerundolo (Oxford, UK). Virus strains rVV‐OVA and rVV‐SIINFEKL encoding the model antigen OVA and its corresponding epitope were kind

2. Material and Methods gifts form J. W. Yewdell (Bethesda, MD, USA). The virus strain expressing tyrosinase (rVV‐Tyr) was a kind gift from V. H. Engelhard (University of Virginia, VA, USA).

All vaccinia virus strains were propagated on BSC‐40 cells. At 80% confluence, BSC‐40 cells were infected with virus with an MOI of 0.01. After 40 h of incubation at 37°C, supernatant was almost completely removed and cells were lysed by repeated freeze‐thaw cycles. Aliquots were stored at ‐80°C. A tenfold serial dilution of each batch was prepared and transferred onto BSC‐

40 monolayers in order to determine virus titers. Cells were incubated at 37°C and 5% CO2 for two days to allow plaque formation. Plaques were visualized by staining with crystal violet solution (0.5% (w/v) crystal violet, 3.7% (v/v) formaldehyde, 0.8% (w/v) NaCl, 50% (v/v) ethanol) for 45 min at room temperature. Excess staining solution was washed off with Milli‐Q water and the number of plaques was counted.

2.5 Cell culture B8‐Db (H‐2d) is a BALB/c‐derived fibroblast cell line stably transfected with H‐2Db and was cultured in IMDM (Basler et al., 2004). MC57 (H‐2b) is a C57BL/6‐derived methylcholantrene‐ induced fibrosarcoma cell line that was maintained in MEM. The dendritic cell line DC2.4 (H‐2b) was maintained in RPMI and was a kind gift from K. Rock (Worsecter, MA, USA). HeLa cells, HEK293 wild type cells and HEK293 cells stably transfected with expression constructs for the murine MHC molecules H‐2Db (kindly provided by M. Basler, Konstanz, Germany) or H‐2Ld were all cultured in DMEM. The T cell hybridoma cell line B3Z specifically recognizing H‐2Kb‐SIINFEKL complexes was grown in IMDM while the UTY hybridoma recognizing UTY246‐254 presented on H‐2Db was maintained in RPMI supplemented with 100 µM 2‐mercaptoethanol. BSC‐40 cells used for rVV amplification is an African green monkey‐derived cell line that was maintained in MEM. The mouse fibroblast cell line L929 was used to propagate LCMV‐WE and was maintained in DMEM. CTLs were cultured in IMDM supplemented with 40 U/ml interleukin (IL)‐2 and 100 µM 2‐mercaptoethanol. Peritoneal macrophages and mouse embryonic fibroblasts (MEFs) from AAD mice (H‐2b, HLA‐A2.1/H‐2Dd), C57BL/6 mice, L7M‐/‐ mice, and LMP2‐/‐ mice were also cultured in DMEM. All media were purchased from Invitrogen‐Life Technologies and contained GlutaMAX, 10% FCS and 100 U/ml penicillin/streptomycin.

All cell lines were cultured at 37°C and 5% CO2 until confluence was reached. T cell hybridoma cell lines growing in suspension were resuspended and further cultured in desired densities. Adherently growing cells were washed with PBS and detached using trypsin/EDTA (Invitrogen‐

2. Material and Methods

Life Technologies). Trypsin was inactivated with cell culture medium and cells were split in appropriate densities.

2.6 Preparation of peritoneal macrophages and MEFs Peritoneal macrophages (pMΦs) were generated by intraperitoneal (i.p.) injection of 1 ml 3% (w/v) thioglycolate broth. After 4 days, mice were sacrificed and cells were washed out of the abdominal cavity by peritoneal lavage using 10 ml cold PBS. PBS was removed by centrifugation at 400 x g for 5 min, cells were resuspended in medium and plated overnight. Adherent cells were further cultured in the presence of 200 U/ml IFN‐γ (Peprotech) for 2 days or left untreated.

MEFs were prepared from embryos on day 14 of gestation. Embryos were removed from the uterus and carefully rinsed with PBS. Next, head and liver was removed from each embryo. Carcasses were again rinsed with PBS, finely minced, and resuspended in 2 ml trypsin/EDTA solution (Invitrogen‐Life Technologies) per embryo. Tissues were digested for 15 min at 37°C before trypsin was inactivated with two volumes of culture medium. Cell suspensions were vigorously pipetted up and down to separate cells and then centrifuged at 400 x g for 5 min. Supernatant was removed and cells resuspended in fresh medium. The cell suspension was passed through a 100 µm filter and plated on one 15 cm tissue culture dish per embryo for two days at 37°C and 5% CO2. Aliquots of cells were stored at ‐150 °C and a different batch of cells was used for each replication of an experiment. MEFs derived from crossed hemizygous AAD mice were prepared from single embryos. Subsequent to preparation, a small fraction of cells was stimulated with 200 U/ml IFN‐γ (Peprotech) to induce MHC class I cell surface expression. Expression of HLA‐A*0201 was analyzed by flow cytometry and positive cells were further used for antigen presentation assays.

2.7 Stimulation of cells In order to induce upregulation of MHC class I surface expression, expression of immunoproteasome, and other immune regulatory genes cells were stimulated with 200 U/ml IFN‐γ (Peprotech) for two days. Cells were further subjected to real‐time RT‐PCR, ribosome isolation or used in antigen presentation assays.

For analysis of NF‐κB activation, pMΦs or MEFs were seeded into 12‐ or 6‐well plates and stimulated with 200 U/ml IFN‐γ (Peprotech) for two days to induce immunoproteasome expression. Cells were then stimulated with 100 U/ml TNF‐α (Peprotech) or 200 ng/ml 34

2. Material and Methods lipopolysaccharide (LPS) (Sigma‐Aldrich) and harvested after different time points. Cells were washed with cold PBS and either used for SDS‐PAGE and western blot or EMSA. Cells used for expression analysis of immunoproteasome subunits were directly harvested after IFN‐γ treatment.

To analyze cytokine secretion, pMΦs stimulated with 200 U/ml IFN‐γ (Peprotech) for 2 days were seeded into 96‐well plates. Two hours after seeding, cells were stimulated with 200 ng/ml LPS (Sigma‐Aldrich) or left untreated. Supernatants of triplicates were collected 24 h later and analyzed using mouse IL‐6, tumor necrosis factor (TNF)‐α, or IL‐10 ELISA kits (Ready‐Set‐Go!, eBiosciences).

2.8. Treatment of cells with chemical inhibitors In order to inhibit p97 activity, cells were treated with the covalent inhibitor EerI (Tocris Bioscience) or the non‐covalent p97‐specific ATP‐competitor DBeQ (Sigma‐Aldrich). Both inhibitors were used at a concentration of 10 μM unless indicated differently. The proteasome inhibitor MG132 (Sigma‐Aldrich) was used at a concentration of 10 μM. All inhibitors were dissolved in DMSO at a concentration of 10 mM. Control samples were treated with solvent DMSO only.

2.9 Transfection of cells Transient transfection with expression constructs HEK293 cells were transfected at 80% confluence using Trans‐IT‐LT1 transfection reagent (Mirus Bio LLC.) according to the manufacturer’s instruction. pCMV expression plasmids encoding p97‐Myc or p97‐E578Q‐Myc were transfected together with a CD4 expression plasmid (pMACS 4.1, Miltenyi Biotec) at a molar ratio of 3:1. After 20 h of incubation, cells were subjected to acid wash or infected with rVV‐eGFP. Cells used in antigen presentation experiments were further purified with MACS using anti‐CD4 antibody‐conjugated magnetic microbeads. Transient overexpression of tyrosinase (Tyr) in HEK293 cells was achieved likewise using a pCMV6‐Entry expression construct encoding Tyr‐Myc‐Flag. Here, cells were incubated for 24 h before 10 μM MG132 or solvent DMSO was applied for 5 h. Cell lysates prepared from these cells were further subjected to immunoprecipitation and SDS‐PAGE.

For transient overexpression of BAG6, HeLa cells were transfected with a pCMV6‐Entry expression plasmid encoding BAG6‐Myc‐FLAG (OriGene Technologies Inc.). Control cells were

2. Material and Methods transfected with empty vector. Transfection was performed at 80% confluence using FuGENE HD transfection reagent (Promega) according to the manufacturer’s instruction. Cells were further analyzed 24 h post transfection.

Stable transfection of HEK293 cells HEK293 cells grown in a 6‐well plate were transfected at 80% confluence with the bicistronic expression vector pIREShyg2 encoding for the mouse MHC molecule H‐2Ld (kind gift from T. Hansen, St. Louis, MO, USA) using Trans‐IT‐LT1 transfection reagent (Mirus Bio LLC.) according to the manufacturer’s instruction. After two days of incubation at 37°C, cells were harvested and three tenfold serial dilutions were prepared. Cell suspensions were plated onto 96‐well flat bottom plates and incubated in the presence of 100 µg/ml hygromycin at 37°C and 5% CO2 until confluence was reached. Clones were further cultured without hygromycin and tested for transgene expression by flow cytometry using an anti‐H‐2Ld antibody. A clone expressing high levels of H‐2Ld was selected and used for antigen presentation assays.

Transfection with siRNA Knockdown of BAG6 in HeLa and B8‐Db cells and knockdown of LTN1 in B8‐Db cells was performed through simultaneous transfection of four different siRNAs targeting the respective mRNA. At 80% confluence, cells were transfected with siRNA (ON‐TARGET plus SMART pool siRNA, Thermo Scientific) using Dharmafect 1 transfection reagent (Thermo Scientific) according to the manufacturer’s instruction. Target sequences of siRNA oligonucleotides are listed below. After 48 h, cells were transfected a second time to achieve maximum efficiency. Cells were analyzed or treated further after another 48 h of incubation. Control cells were transfected twice with a mix of four non‐targeting siRNAs (ON‐TARGET plus Non‐Targeting pool siRNA, Thermo Scientific). Knockdown efficiency was confirmed by SDS‐PAGE and western blot. Cells were further subjected to analysis of MHC class I cell surface expression by flow cytometry or used in antigen presentation assays.

Table 1: Target sequences of siRNA

Oligonucleotide Target Sequence Supplier

Human BAG6 siRNA ACAUUCAGAGCCAGCGGAA Thermo Scientific

(ON‐TARGETplus SMART pool) UCUCUAUGGUGGACGUAGU

UGUUAUCAAUGGCCGAAUU

GAGGAGGAUCAGCGGUUGA

2. Material and Methods

Mouse BAG6 siRNA GAUCUGCGCUGCAAUCUAG Thermo Scientific

(ON‐TARGETplus SMART pool) CCACGGGUCAUCCGGAUUU

CUGAAUGGGUCCCUAUUAU

GAUAUCAUCCGGACAAAUU

Mouse LTN1 siRNA ACAUAGAGAUCAUGCGGUU Thermo Scientific

(ON‐TARGETplus SMART pool) UUUGAAUGCUUGCGGUUUA

CCACAAGCUCCCUGCGAGA

CCGUUAAGCCACUGAGCGA

Non‐targeting siRNA Not available Thermo Scientific

(ON‐TARGETplus non‐targeting pool)

2.10 SDS‐PAGE (sodium dodecylsulfate polyacrylamide gel electrophoresis) and western blot Cells were lysed in RIPA buffer (150 mM NaCl, 50mM tris pH 8, 1% (v/v) Triton X‐100, 0.5% (v/v) sodiumdesoxycholate, 0.1% (w/v) SDS) including protease inhibitors (cOmplete EDTA‐free, Roche) for 30 min on ice. Lysates were centrifuged at 20,000 x g and 4°C for 15 min and supernatants were transferred into fresh Eppendorf tubes. Protein concentration was analyzed using DC protein assay (Bio‐Rad) according to the manufacture’s instruction before samples were boiled with SDS sample buffer (8 ml glycerol, 6 ml 20% (w/v) SDS, 5.6 ml Milli‐Q, 0.4 ml 1 M Tris pH 6.8, 0.05% (w/v) bromphenol blue, 10% (v/v) 2‐mercaptoethanol) for 5 min at 95°C.

Proteins were separated by SDS‐PAGE according to standard procedures and blotted onto nitrocellulose membranes (Whatman) by wet blot (Laemmli, 1970). After blocking in Roti‐Block solution (Roth) for 1 h at room temperature, membranes were incubated with primary antibodies diluted in Roti‐Block solution at 4°C overnight. Membranes were washed three times for 5 min with TBST buffer (50 mM Tris, 150 mM NaCl, 0.05% (v/v) Tween 20, pH 7.6) and incubated with appropriate peroxidase‐conjugated secondary antibodies diluted in Roti‐Block solution for 2 h at room temperature. Membranes were washed again and proteins were visualized with enhanced chemiluminescence using SuperSignal West Pico or Femto Chemiluminescent Substrate (Thermo Scientific) together with the ChemiDocTM XRS system 37

2. Material and Methods

(Bio‐Rad) and the Quantity One software (Bio‐Rad). All primary and secondary antibodies used are listed below.

Table 2: Primary antibodies

Antibody Clone Specification Dilution Supplier

Anti‐α6 IB5 Mouse K. Scherrer

Paris, France

Anti‐BAG6 Polyclonal Rabbit 1:10000 R. S. Hegde,

Cambridge, UK

Anti‐β1 E1K9O Rabbit 1:1000 Cell Signaling Technology

Anti‐β5 D1H6B Rabbit 1:1000 Cell Signaling Technology

Anti‐Flag M2 Mouse Sigma‐Aldrich

Anti‐IκBα L35A5 Mouse 1:5000 Cell Signaling Technology

Anti‐Lamin A/C 4C11 Mouse 1:2000 Cell Signaling Technology

Anti‐LMP2 Polyclonal Rabbit 1:2000 Produced in house

(Kremer et al., 2010)

Anti‐LPM7 Polyclonal Rabbit 1:5000 Produced in house

(Kremer et al., 2010)

Anti‐LTN1 Polyclonal Rabbit 1:1000 C. Joazeiro,

San Diego, USA

Anti‐Myc 9E10 Mouse 1:1000 Sigma‐Aldrich

Anti‐p97 5 Mouse 1:1000 Santa Cruz Biotechnology

Anti‐ubiquitin FK2 Mouse 1:1000 Enzo Life Sciences

Anti‐RPL‐7 Polyclonal Rabbit 1:2000 Abcam

Anti‐tubulin AA13 Mouse 1:5000 Sigma‐Aldrich

2. Material and Methods

Table 3: Secondary antibodies

Antibody Clone Specification Dilution Supplier

Anti‐Mouse HRP Polyclonal Goat 1:1000 Dako

Anti‐Rabbit HRP Polyclonal Swine 1:3000 Dako

2.11 Immunoprecipitation (IP) Cells transiently transfected with an expression construct for Tyr‐Myc‐Flag or with empty vector were incubated with 10 µM MG132 (Sigma‐Aldrich) or solvent DMSO for 5 h at 37°C and

5% CO2. Cells were then harvested and lysed in IP lysis buffer (20 mM Tris pH7.8, 50 mM NaCl,

10 mM MgCl2, 1% (v/v) NP‐40) including protease inhibitors (cOmplete EDTA‐free, Roche) for 30 min on ice. Lysates were centrifuged at 20,000 x g and 4°C for 15 min. Supernatants were loaded onto protein G affinity gel (EZview, Sigma‐Aldrich) together with anti‐Myc antibody and incubated overnight at 4°C on a rotator. The affinity gel was separated from unbound protein by centrifugation at 800 x g and 4°C for 1 min. The supernatant was removed and the affinity gel was washed three times in high salt IP buffer and three times in low salt IP buffer buffer (650 mM or 150 mM NaCl, 50 mM Tris pH 8, 5 mM EDTA, 0.5% (v/v) Triton X‐100). Then, precipitated proteins were boiled in SDS sample buffer for 5 min at 95°C and subsequently analyzed by SDS‐PAGE and western blot.

2.12 Proteasome IP Prior to the IP, anti‐proteasome immune serum was cross‐linked to protein A affinity gel (EZview, Sigma‐Aldrich) in order to elute proteasomes with little antibody contamination. About 300 μl protein A affinity gel was resuspended in 1 ml PBS and centrifuged at 800 x g for 1 min. The supernatant was discarded and the beads were washed again with PBS. After incubation in 1 ml PBS overnight at 4°C, beads were blocked in 300 μl dilution buffer (1 mg/ml BSA in PBS) for 10 min followed by incubation with 150 μl polyclonal rabbit anti‐20S proteasome immune serum (Schwarz et al., 2000) diluted 1:1 with dilution buffer for 1 h at 4°C on a rotator. Beads were washed with blocking buffer and PBS and then cross‐linked for with 300 μl of 13 ml/ml DMP dissolved in wash buffer (0.2 M triethanolamine in PBS) for 30 min at room temperature while rotating. Beads were washed with wash buffer and treated with DMP two more times. Finally, beads were treated twice with quenching buffer (50 mM ethanolamine in PBS), washed with PBS, and cleared off unlinked antibody by incubation in 1 M glycin pH 3

2. Material and Methods for 10 min at room temperature while rotating. After two more washing steps with PBS, beads were stored at 4°C.

About 5*107 pMΦs were lysed in 1 ml proteasome lysis buffer for 30 min on ice. Cell debris was removed by centrifugation at 20,000 x g and 4°C for 15 min. The supernatant was precleared with 30 µl protein A affinity gel for 30 min at 4°C, then loaded onto cross‐linked beads and incubated overnight while rotating. Unbound protein was removed by centrifugation at 800 x g and 4°C for 1 min followed three washing steps in high salt and low salt buffer (650 mM or 150 mM NaCl, 50 mM Tris pH 8, 5 mM EDTA, 0.5% (v/v) Triton X‐100), respectively. Bound proteasome was eluted by overnight incubation in 80 µl 2D sample buffer without 2‐mercaptoethanol (9.5 M urea, 2% NP‐40, 2% (w/v) SERVALYT carrier ampholines pH 3‐10 (Serva), 0.3% (w/v) SDS). After centrifugation, the supernatant was mixed with 4 µl 2‐ mercaptoethanol and subjected to 2D‐gelelectrophoresis.

2.13 2D‐gel electrophoresis Nonequilibrium pH gel electrophoresis (NEPHGE) was used to separate proteins in the first dimension. Gel rods (9.5 M urea, 3.75% (w/v) acrylamide, 0.21% (w/v) bisacrylamide, 2% (v/v) NP‐40, 2% (w/v) SERVALYT carrier ampholines pH 3‐10 (Serva), APS, TEMED) were poured into glass tubes, topped with overlay solution (9 M urea, 2% (w/v) SERVALYT carrier ampholines pH 3‐10 (Serva)) and left to polymerize for 1 h. Samples were loaded onto gel rods and separated for 4 h at 400 V with 0.01 M H3PO4 used as anode buffer at the top and 0.02 M NaOH used as cathode buffer at the bottom of the rod. Gels were washed out, equilibrated twice for 20 min in equilibration buffer (10% /v/v) glycerol, 10% (v/v) 2‐mercaptoethanol, 2.3% (w/v) SDS, 90 mM Tris pH 8.8), and fixed to 15% SDS‐PAGE gel with agarose buffer. The second dimension was run at 1100 Vh and stained using the Pierce Silver Stain Kit (Thermo Scientific).

2.14 Electrophoretic mobility shift assay (EMSA) Nuclear extracts prepared from MEFs were used for EMSA. MEFs were lysed in hypotonic lysis buffer (10 mM HEPES pH 7.9, 10 mM KCl, 1 mM DTT, 0.1 mM EDTA) including protease inhibitors (cOmplete EDTA‐free, Roche) and phosphatase inhibitors (PhosSTOP, Roche) for 10 min on ice. NP‐40 was added to reach a final concentration of 0.2%, lysates were vortexed and centrifuged at 12,000 x g and 4°C for 20 sec. Pellets containing the nuclear faction were lysed in nuclear extraction buffer (20% (v/v) glycerol, 20 mM HEPES pH 7.9, 0.4 M NaCl, 1 mM DTT, 1 mM EDTA, 1 mM EGTA) including phosphatase and protease inhibitors for 30 min on ice. Nuclear lysates were centrifuged at 20,000 x g at 4°C for 14 min. Protein concentration of 40

2. Material and Methods supernatants was analyzed using DC protein assay (Bio‐Rad) according to the manufacture’s instruction. To ensure equal loading of EMSA reactions, aliquots of the nuclear lysates were boiled with SDS sample buffer at 95°C for 5 min and subjected to SDS‐PAGE and western blot using anti‐Lamin A/C antibody.

EMSA was performed using a 32P‐labeled double‐stranded DNA a probe containing the NF‐κB binding site from the mouse H‐2K promoter (5’‐CAGGGCTGGGGATTCCCCATCTCCACAGG‐3’). Complementary oligonucleotides containing 5’‐GATC overhangs were mixed in annealing buffer (50 mM Tris pH 8, 70 mM NaCl) at a concentration of 200 ng/ml, incubated at 90°C for 10 min and slowly cooled down to allow oligonucleotide annealing. Oligonucleotides were then radioactively labeled with dATP[α‐32P] using DNA polymerase I Klenow fragment (NEB) according to the manufactures instruction. Labeled probes were purified using QIAquick nucleotide removal kit (Qiagen). Shift reactions (20 µl) contained 5 µg nuclear extracts, 0.5 µg BSA, 5 mM DTT, 0.1 µg/µl poly(dI‐dC) (AffymetriX) and about 0.5 ng probe in shift buffer (20 mM HEPES pH 7.9, 60 mM KCL, 5% (v/v) Ficoll) and were incubated for 30 min at room temperature. The complexes were separated on a 5% native polyacrylamide gel in TBE buffer (100 mM Tris, 90 mM boric acid, 1 mM EDTA, pH 8.2). Gels were vacuum‐dried onto whatman paper at 80°C for 1.5 h, exposed to phosphor screens, and the bands visualized using a phosphorimager.

2.15 Ribosome isolation About 1*107 HEK293 cells either stimulated with IFN‐γ of left untreated were lysed in 300 µl ribosome lysis buffer (50 mM Tris pH 7.4, 250 mM sucrose, 25 mM KCl, 5 mM MgCl2, 0.7 % (v/v) NP‐40, 2 mM DTE) including protease inhibitors (cOmplete EDTA‐free, Roche) for 10 min on ice. Lysates were centrifuged at 800 x g and 4°C for 10 min. Supernatants containing the post‐ nuclear fraction were centrifuged again at 12,500 x g and 4°C for 10 min. The post‐ mitochondrial supernatant was layered on top of 1 ml sucrose cushion (50 mM Tris pH 7.4, 1

M sucrose, 25 mM KCl, 5 mM MgCl2) and centrifuged at 305,000 x g and 4°C for 2 h. A fraction of the post‐mitochondrial fraction was boiled with SDS sample buffer for 5 min at 95°C and used as input control. Ribosomal pellets were carefully washed with water, resuspended in ribosome buffer (50 mM Tris pH 7.4, 25 mM KCl, 5 mM MgCl2, 2 mM DTE) including protease inhibitors, and then boiled with SDS sample buffer for 5 min at 95°C. Post‐mitochondrial and ribosomal fractions were analyzed by SDS‐PAGE and western blot.

2. Material and Methods

2.16 Real‐time RT‐PCR Total RNA was extracted using NucleoSpin RNA II kit (Macherey‐Nagel) according to the manufacturer’s instruction. Extracted RNA was reverse‐transcribed using the Reverse Transcription System (Promega) according to the manufacturer’s instruction. Quantitative real‐ time PCR was performed with the LightCycler Instrument II (Roche) and LightCycler FastStart DNA Master SYBR Green I kit (Roche). Gene expression was calculated relative to control cells and normalized to HPRT (hypoxanthine guanine phosphoribosyl transferase) expression using REST‐384O software version 2 (Gene Quantification). RNA quality and amplicon size of PCR reactions was verified by agarose gel electrophoresis. Sequences of oligonucleotides used for Real‐time RT‐PCR are listed in below.

Table 4: Oligonucleotides used in real‐time RT‐PCR

Oligonucleotide Sequence

BAG6 (human) forward TACAATAACAATCACGAGGGCC

BAG6 (human) reverse GGTGGTGTAGTGAGACATAGG

HPRT (human) forward TGGACAGGACTGAACGTCTTG

HPRT (human) reverse CCAGCAGGTCAGCAAAGAATTTA

LMP7 (human) forward AATGCAGGCTGTACTATCTGCG

LMP7 (human) reverse TGCAGCAGGTCACTGACATCTG

2.17 Acid wash Harvested MC57 cells were centrifuged at 400 x g for 5 min and supernatant was discarded. The cell pellet was chilled on ice and then gently resuspended in 1.5 ml ice‐cold citric acid buffer for 1.5 min. Two volumes of ice‐cold PBS were added followed by ten volumes of cell culture medium. Cells were centrifuged at 400 x g and 4°C for 5 min. Supernatant was aspirated and cells were washed two more times with cold cell culture medium. Cell suspensions were further incubated at 37°C in the presence of chemical inhibitors or solvent DMSO and analyzed for MHC class I surface expression by flow cytometry after different time points. Samples taken directly after acid wash served as time point zero for all groups. HEK293 cells were treated essentially like MC57 cells except that cells were treated while still being adherently attached to tissue culture plates. After treatment, cells were harvested after different time points and analyzed. Whenever transiently transfected HEK293 cells were used, cells were left untreated at 37°C and 5% CO2 for the indicated time periods. In some approaches cells were additionally 42

2. Material and Methods supplemented with peptides at a concentration of 10‐6 M during the entire recovery phase. HEK293 cells were supplemented with the HLA‐A02‐restricted peptide Tyr369‐377 and MC57 cells were supplemented with the H‐2Db‐restricted NP396‐404. Sequences of peptides used are listed in 3.19.

2.18 Flow cytometry All antibodies used in flow cytometric analysis are listed below.

Surface staining MHC class I surface expression after acid wash, after transfection with siRNA or transient overexpression was analyzed by flow cytometry using fluorescent‐labeled antibodies. About 1‐ 5*105 cells were incubated together with antibodies diluted 1:150 in flow cytometry buffer (PBS, 2% (v/v) FCS, 2 mM EDTA) for 20 min at 4°C. Cells were centrifuged at 620 x g for 90 sec and supernatant containing excess antibody was discarded. After washing with flow cytometry buffer twice cells were acquired with the use of the Accuri C6 flow cytometer (BD Biosciences). Data were analyzed with FlowJo software (Tree Star).

GFP fluorescence Cells transfected with p97‐Myc or p97‐E578Q‐Myc together with a CD4 expression constructs were infected with rVV‐eGFP with an MOI of 10 for 3 h. Cells were then first stained with anti‐ CD4 APC (BD Biosciences) as described above. Before acquiring, cells were fixed for 5 min in cold 4% paraformaldehyde (PFA) in PBS and washed twice in flow cytometry buffer.

Dead cell staining HEK293 cells were treated with 1, 5, or 10 µM DBeQ or solvent DMSO for 3 or 6 h and then stained with anti‐annexin V APC and PI using the Annexin V Apotosis Detection Kit APC (eBioscience) according to the manufacturer’s instruction.

Intracellular IFN‐g staining CTLs were stained with anti‐CD8a APC (eBioscience) antibody (1:150) in flow cytometry buffer for 20 min at 4°C. Cells were centrifuged at 620 x g for 90 sec and supernatant containing excess antibody was discarded. After washing with flow cytometry buffer twice, cells were fixed for 5 min with cold 4% PFA in PBS. Cells were then washed with permeabilization buffer (2% (w/v) saponin in flow cytometry buffer) twice and incubated with anti‐IFN‐γ FITC (1:1000) in permeabilization buffer overnight at 4°C. Cells were again washed with flow cytometry buffer and acquired.

2. Material and Methods

Table 5: Flow cytometry antibodies

Antibody Clone Specification Dilution Supplier

Anti‐Annexin V APC VAA‐27 Mouse 1:20 eBioscience

Anti‐CD4 APC (human) RPA‐T4 Mouse 1:150 eBioscience

Anti‐CD8a APC (mouse) 53‐6.7 Rat 1:150 eBioscience

Anti‐HLA‐ABC APC G46‐2.6 Mouse 1:150 BD Biosciences

Anti‐HLA‐A02 APC BB7.2 Mouse 1:150 eBioscience

Anti‐H‐2Db FITC KH95 Mouse 1:150 BD Biosciences

Anti‐H‐2Dd PE 34‐2‐12 Mouse 1:150 BD Biosciences

Anti‐H‐2Kb PE AF6‐88.5 Mouse 1:150 BD Biosciences

Anti‐IFN‐γ FITC (mouse) AN‐18 Rat 1:1000 M. Basler,

Konstanz, Germany

2.19 Generation of epitope‐specific CTLs LCMV‐specific CTL lines were generated from female C57BL/6 or BALB/c mice infected i.v. with 200 PFU LCMV‐WE. Four weeks post infection mice were sacrificed and the spleens were homogenized in medium. Released splenocytes were cultured in IMDM 10% FCS, penicillin/streptomycin, supplemented with 40 U/ml IL‐2, 100 µM 2‐mercaptoethanol and 10‐6 M peptide at about 1*107 cells per 6‐well. Sequences of peptides used for restimulation are listed below. Cytokine‐supplemented medium was added every other day for 8‐14 days. Before CTLs were used in presentation assays, dead cells were removed by Ficoll density centrifugation. CTLs were collected, layered on top of 3 ml Ficoll‐Paque PLUS (GE Healthcare) per well and centrifuged for 15 min at 960 x g (accelerate: 66% of max, decelerate 33% of max). CTLs, forming a discrete fraction between medium and Ficoll, were washed with medium and directly used in presentation assays.

Tyr‐specific CTL‐lines were generated from male AAD mice infected i.p. with 2*10^6 PFU rVV‐ Tyr. Four weeks post transfection, splenocytes were cultured for three to four weeks as described for LCMV‐specific CTLs. To increase percentage of specific T cells, CTLs were in addition restimulated twice with peptide‐loaded, irradiated (20 Gy) AAD splenocytes on day 8 and 16 at a 1:10 ratio of loaded splenocytes to CTLs.

2. Material and Methods

Table 6: Synthetic peptides

Oligonucleotide Target Sequence Supplier

GP276‐286 SGVENPGGYCL P. Henklein (Charité, Berlin, Germany)

NP118‐126 RPQASGYVM P. Henklein (Charité, Berlin, Germany)

NP396‐404 FQPQNGQF P. Henklein (Charité, Berlin, Germany)

Tyr369‐377 YMDGTMSQV Sigma‐Aldrich

2.20 Antigen presentation assay using CTLs B8‐Db cells were pulsed with 1, 5, or 10 μM EerI or DMSO for 30 min, washed with PBS, and infected with indicated rVVs or LCMV‐WE with an MOI of 10 for 3 h. Transfected HEK293 cells, B8‐Db cells, and AAD MEFs were infected likewise. Serial dilution of infected cells was performed to achieve different effector to stimulator ratios (E/S). Epitope‐specific CTL lines were added to infected cells and incubated in presence of 10 µg/ml brefeldin A (Sigma‐Aldrich) for 3 h at 37°C. Activation of CTLs was determined by intracellular IFN‐γ staining and flow cytometry. All samples were prepared as duplicates.

2.21 Antigen presentation assay using T cell hybridoma lines DC2.4 cells were harvested and the cell suspension adjusted to a volume of 1.5 ml medium per 1*106 cells. Cells were then pulsed with 1, 5, or 10 μM EerI or solvent DMSO for 30 min at 37°C. Cells were centrifuged at 400 x g for 5 min and supernatants were removed. After washing with PBS, cells were infected with indicated rVVs with an MOI of 10 for 3 h at 37°C and 450 rpm in about 150‐200 μl medium per 1*106 cells. Cells were washed with 1 ml PBS and fixed for 5 min in 200 μl cold 4% PFA in PBS. After extensive washing with PBS, cell suspensions were adjusted to 2*106 cells/ml medium. A threefold serial dilution of infected cells was performed to achieve different effector to stimulator ratios (E/S). Epitope‐specific T cell hybridoma lines were also adjusted 2*106 cells/ml medium and 100 μl cell suspension was added to 100 μl of diluted infected cells in 96‐well round bottom plates. All samples were prepared as triplicates. Cells were incubated for 20 h at 37°C and 5% CO2 and supernatants were analyzed for IL‐2 concentrations using a mouse IL‐2 ELISA kit (Ready‐Set‐Go!, eBioscience) according to the manufacturer’s instruction. In UTY246‐254 presentation assays, DC2.4 cells were stimulated with 200 U/ml IFN‐γ for two days before the start of the experiment.

2. Material and Methods

2.22 Magnetic‐activated cell sorting (MACS) After 20 h of transfection with expression plasmids encoding p97‐Myc or p97‐E578Q‐Myc together with a CD4 expression plasmid, CD4 expressing HEK293 cells were further purified by positive sorting using human CD4 MicroBeads (Miltenyi) according to the manufacture’s instruction. Purified cells were further used in antigen presentation assays. Transgene expression and ubiquitin accumulation of positively sorted cells was analyzed by SDS‐PAGE and western blot.

2.23 Statistical analysis The unpaired two‐tailed Student’s t test and two‐way ANOVA was applied for statistical analysis using GraphPad Prism software.

3. Results

3.1 The role of p97 in antigen processing and presentation on MHC class I Chemical inhibition of p97 blocks MHC class I cell surface expression MHC class I complexes at the cell surface are relatively stable with a half‐life in the range of several hours (Zehn et al., 2004). To study the impact of p97 on antigen presentation, preexisting complexes on the human embryonic kidney cell line HEK293 were therefore removed using citric acid (acid wash) and their reappearance on the cell surface was monitored over time. During recovery phase cells were incubated in the presence of the reversible p97 inhibitor DBeQ, proteasome inhibitor MG132, or solvent only. Acid treatment caused a strong down‐shift in MHC class I cell surface expression as determined by flow cytometry using a pan‐ MHC class I antibody against HLA‐A,B,C (Fig. 13A, left panel). While control cells regained substantial class I surface expression during the 4.5 h monitored, cells treated with DBeQ did not (Fig. 13A, middle panel). A similar pattern was obtained when expression of the MHC class I molecule HLA‐A02 was analyzed (Fig. 13A, right panel). Strikingly, HLA‐A02 expression of DBeQ treated cells could be restored when cells were supplemented with an HLA‐A02‐restricted peptide covering residues 369‐377 of human tyrosinase. These results indicate that inhibition of p97 limits the peptide supply for MHC class I‐restricted presentation. Since DBeQ and MG132 had almost the same impact on pan MHC class I as well as HLA‐A02 surface expression, p97 and the proteasome seem to be equally important for antigen processing and presentation.

To not only rely on one human cell line the previous finding was verified with the mouse fibroblast cell line MC57. Here, surface expression of the MHC molecules H‐2Kb and H‐2Db after acid wash and inhibitor treatment was analyzed (Fig. 13B). Again, treatment with DBeQ or MG132 blocked re‐expression of H‐2Kb and H‐2Db on the cell surface while supplementation with the H‐2Db‐restricted peptide NP396‐404 derived from LCMV‐NP restored surface expression of H‐2Db of DBeQ treated cells.

The central role of p97 and the proteasome in protein turnover is also reflected in the accumulation of polyubiquitylated proteins within only three hours of inhibiting these enzymes (Fig. 13C). It has been shown previously that interference with p97 activity leads to accumulation of Lys11‐ and Lys48‐linked ubiquitin chains (Locke et al., 2014). Both types of chains are implicated in proteasomal targeting of substrates (Grice et al., 2015). Thus, the effect of p97 inhibition on antigen presentation is likely caused through inhibition of antigen

3. Results degradation via the UPS. Since treatment with 10 µM DBeQ, the concentration used throughout experiments, caused only a mild reduction in the number of viable cells (Fig. 13D), cytotoxicity can be excluded as the main mechanism of action.

Figure 13: Chemical inhibition of p97 blocks MHC class I cell surface expression. (A) HEK293 cells and (B) MC57 cells were acid washed to remove MHC complexes or left untreated. Cells were further cultured in the presence of 10 µM DBeQ, 10 µM DBeQ + peptide, 10 µM MG132, or DMSO and MHC class I surface expression was monitored over time. Unstained cells were included as negative control. Values represent mean ± SD of duplicate determinations. (A) Peptide supplemented: Tyr369‐377 (HLA‐A02), MHC molecules: HLA‐A,B,C and HLA‐A02. (B) Peptide supplemented: NP396‐404 (H‐2Db), MHC molecules: H‐2Kb and H‐2Db. (C) Western blot analysis of whole cell lysates prepared from HEK293 cells treated for 3 h with 10 µM DBeQ, 10 µM MG132, or DMSO. Tubulin was used as loading control. (D) Percentage of viable cells (annexin V‐ and PI‐) of cells treated for 3 or 6 h with 10 µM, 5 µM, 1 µM DBeQ, or DMSO. Values represent mean ± SD of triplicate determinations. *, P < 0.05 (two‐way ANOVA). For all experiments one out of three independent experiments is shown. MFI = median fluorescence intensity.

Expression of a dominant negative p97 mutant interferes with MHC class I cell surface expression To genetically confirm the results obtained with chemical inhibition, expression of the dominant negative (DN) mutant p97‐E578Q containing a catalytically dead D2 domain was used (Dalal et al., 2004). The effect of this mutant on class I surface expression was tested in HEK293 cells transiently transfected for 20 h followed by acid wash. Since p97 is one of the most abundant proteins in the cell p97 expression constructs were co‐transfected in threefold excess

3. Results compared to a CD4 expression construct. Gating on the CD4 positive population in flow cytometric analysis then allowed to further enrich transfected cells.

Figure 14: DN p97 mutant blocks MHC class I surface expression. (A) HEK293 cells transfected for 20 h with p97 or p97‐E578Q in threefold molecular excess compared to CD4 were acid washed to remove MHC complexes or left untreated. HLA‐A,B,C and HLA‐A02 surface expression of CD4+ cells was monitored over time. One part of the p97‐E578Q‐transfected cells was cultured in the presence of peptide (Tyr369‐377, HLA‐ A02). Unstained cells were included as negative control. Values represent mean ± SD of duplicate determinations. (B) Western blot analysis of whole cell lysates prepared from HEK293 cells transfected as described in (A) and further separated through magnetic sorting for CD4. Tubulin was used as loading control. For both experiments one out of three independent experiments is shown. FSC = forward scatter, MFI = median fluorescence intensity.

Before and directly after acid wash p97‐ and p97‐E578Q‐transfected cells did not differ in their pan‐MHC class I cell surface expression (Fig. 14A, upper panels). However, similar to chemical inhibition of p97, expression of p97‐E578Q reduced the recovery of HLA‐A,B,C complexes on the cell surface after acid wash (Fig. 14A, lower panels). This effect was even more prominent for HLA‐A02 complexes. Moreover, surface expression of HLA‐A02 could be restored by supplementing cells with an HLA‐A02 peptide ligand. To further verify the DN effect of p97‐ E578Q on protein turnover, HEK293 cells were co‐transfected with constructs for p97‐E578Q and CD4 or wild type p97 and CD4, respectively, and magnetically enriched CD4 positive cells were subjected to SDS‐PAGE western blot analysis. Indeed, accumulation of polyubiquitylated proteins was only visible in cells expressing DN but not wild type p97 (Fig. 14B). Hence, expression of a DN p97 mutant fully recapitulates the results obtained with chemical p97 inhibition.

3. Results

Presentation of virus‐derived epitopes on MHC class I depends on p97 activity Next, the influence of p97 inhibition on the presentation of single MHC class I epitopes after virus infection was investigated. The murine dendritic cell line DC2.4 (H‐2b) was pulsed for 30 min with various concentrations of the covalent p97 inhibitor EerI and then infected for 3 h with rVV expressing either the model antigen ovalbumin (OVA) or its H‐2Kb‐restricted epitope SIINFEKL as minigene. To avoid cytotoxic effects of extended EerI treatment, cells were fixed before overnight incubation with the SIINFEKL‐specific T cell hybridoma cell line B3Z. Secreted interleukin (IL)‐2, produced by B3Z cells upon encountering their cognate antigen, was used as readout. EerI had a strong dose‐dependent effect on SIINFEKL presentation in rVV‐OVA infected cells, which was clearly less pronounced in cells infected with rVV‐SIINFEKL (Fig. 15A). Another epitope analyzed with this approach was the male minor antigen UTY‐derived peptide UTY246‐254. Since the production of this epitope depends on the immunoproteasome (Basler et al., 2012), DC2.4 cells were first stimulated with IFN‐γ for two days to induce immunoproteasome formation. Presentation of UTY246‐254 produced from full length UTY protein was entirely blocked in cells treated with 10 µM EerI while presentation could be restored substantially if cells were infected with a virus encoding the UTY246‐254 epitope only (Fig. 15B). Similar to DBeQ, EerI caused accumulation of polyubiquitylated proteins in DC2.4 cells implicating its inhibiting effect on UPS dependent protein degradation (Fig. 15C).

The impact of EerI on antigen processing was further tested with LCMV‐NP, which was previously identified as a DRiP substrate depending on neosynthesis for presentation on MHC class I (Khan et al., 2001a). The mouse fibroblast cell line B8‐Db expressing endogenous H‐2d and stably transfected H‐2Db was pulsed for 30 min with various concentrations of EerI before being infected for 3 h with rVVs expressing full‐length NP or the NP‐derived epitopes NP118‐ 126/Ld and NP396‐404/Db as minigenes. Peptide presentation was monitored with NP118‐126‐ or NP396‐404‐specific CTLs lines by CD8 and intracellular IFN‐γ staining after 3 h of co‐culture with infected cells. CTLs produce IFN‐γ upon stimulation of their receptor with the appropriate peptide‐MHC complex. Thus, the proportion of IFN‐γ producing CTLs of all CD8‐bearing CTLs reflects the amount of peptides presented by infected cells. Inhibition of p97 had a strong dose dependent effect on the presentation of both NP‐derived epitopes (Fig. 15D). Similar to DC2.4 cells (Fig. 15C), inhibition of p97 with EerI increased the amount of polyubiquitylated substrates in B8‐Db cells (Fig. 15E).

3. Results

Figure 15: Inhibition of p97 blocks presentation of virus‐derived epitopes. (A and B) DC2.4 cells were pulsed for 30 min with 1 µM, 5 µM, 10 µM EerI, or DMSO and then infected with rVV or VV‐WR for 3 h (MOI = 10). Cells were fixed and epitope‐specific T cell hybridoma cells were added to infected cells in indicated effector to stimulator ratios (E/S). IL‐2 concentration was determined as a measure for epitope presentation. Values represent mean ± SD of triplicate determinations. (A) Presentation of SIINFEKL on cells infected with rVV‐OVA (left) or rVV‐SIINFEKL (right). (B) Presentation of UTY246‐254 on cells infected with rVV‐UTY (left) or rVV‐ UTY246‐254 (right). DC2.4 cells were pre‐stimulated with IFN‐γ for 2 d. (C and E) Western blot analysis of whole cell lysates prepared from DC2.4 (C) or B8‐Db (E) cells treated with 10 µM DBeQ, 10 µM MG132, or DMSO for 3 h. Tubulin was used as loading control. (D) Presentation of virus‐derived epitopes on B8‐Db cells pulsed for 30 min with 1 µM, 5 µM, 10 µM EerI, or DMSO and then infected with rVV or VV‐WR for 3 h. Peptide‐specific CTLs were added to infected cells in indicated effector to stimulator ratios (E/S) and percentage of IFN‐γ+ of CD8+ cells was determined as a measure for presentation. Upper graphs: Infection with rVV‐NP (left) or rVV NP396‐404 (right), detection with NP396‐404‐specific CTLs. Lower graphs: Infection with rVV‐NP (left) or rVV‐NP118‐126 (right), detection with NP118‐126‐specific CTLs. Values represent mean ± SD of duplicate determinations. For all experiments one out of three independent experiments is shown.

Taken together, EerI influences antigen processing and to a lesser extent antigen presentation of peptide epitopes, which do not need further processing. However, the possibility that the effects seen with EerI are in part due to inhibition of other cellular pathways cannot be entirely excluded. EerI is still an inhibitor with poorly defined mechanism and apart from inhibiting p97

3. Results it also affects protein synthesis and Sec61‐mediated protein translocation into the endoplasmic reticulum (Aletrari et al., 2011; Cross et al., 2009a).

Expression of a DN p97 mutant inhibits presentation of virus‐derived epitopes Presentation of epitopes derived from LCMV‐NP and ‐GP was also tested in HEK293 cells transfected with DN p97 to further strengthen the results obtained with chemical inhibition. HEK293‐Ld cells stably expressing murine H‐2Ld were transiently transfected with p97 or p97‐ E578Q together with CD4 for 20 h. Magnetically enriched CD4‐expressing HEK293‐Ld cells were infected with rVV‐NP or rVV‐NP118‐126 for 3 h and then analyzed for NP118‐126 presentation using NP118‐126‐specific CTLs and intracellular IFN‐γ staining.

Figure 16: DN p97 inhibits presentation of virus‐derived epitopes. (A and B) Presentation of virus‐derived epitopes on HEK293 cells transfected for 20 h with p97 or p97‐E578Q in threefold molecular excess compared to CD4. CD4+ cells were magnetically enriched and infected with rVV or VV‐WR (MOI = 10). Epitope‐specific CTLs were added to infected cells in indicated effector to stimulator ratios (E/S) and percentage of IFN‐γ+ of CD8+ cells was determined as a measure for epitope presentation. Values represent mean ± SD of duplicate determinations. One out of three independent experiments is shown (A) Presentation of NP118‐126 on HEK293‐Ld cells infected with rVV‐NP (left) or rVV‐NP118‐126 (right). (B) Presentation of GP276‐286 on HEK293‐Db cells infected with rVV‐GP (left) or rVV‐GP276‐286 (right). (C) HEK293 cells were transfected for 20 h with p97 or p97‐E578Q in threefold molecular excess compared to CD4 and infected for 3 h with rVV‐eGFP. CD4+ cells were analyzed for median GFP fluorescence and percentage of GFP+ cells. Bar graphs show means and ± SD of three independent experiments. ns, P > 0.05 (unpaired two‐tailed Student’s t test).

3. Results

Strong reduction of NP118‐126 antigen presentation was detected in cells transfected with DN p97 when infected with rVV‐NP while the reduction was less pronounced in cells producing the NP118‐126 epitope from rVV‐NP118‐126 (Fig. 16A). Likewise, transiently transfected HEK293‐ Db cells were infected with rVV‐GP or rVV‐GP276‐286 and analyzed for their potential to present GP276‐286/Db. Antigen processing from GP was reduced in p97‐E578Q‐transfected cells, however, GP276‐286 presentation of cells infected with rVV‐GP276‐286 was comparable between cells transfected with DN and wild type p97 (Fig. 16B). To exclude an effect of p97‐ E578Q on protein synthesis and rVV infection, transfected cells were infected with rVV expressing enhanced GFP (eGFP) and analyzed by flow cytometry. The median GFP fluorescence and number of GFP positive cells within the CD4 expressing population was not significantly altered in p97‐E578Q expressing cells (Fig. 16C). Compared to the previous results, the effect of DN p97 construct on the presentation of virus‐derived epitopes fully recapitulated chemical inhibition with EerI.

Ribosome‐associated degradation via the E3 ligase LTN1 is dispensable for MHC class I‐ restricted antigen presentation p97 is recruited to stalled ribosomes as part of the RQC and assists in ribosome‐associated degradation of nascent chains (Brandman et al., 2012; Verma et al., 2013). A specialized functional connection between ribosomes and proteasomes as part of the MHC class I‐ restricted antigen processing pathway has been discussed for years (Yewdell and Nicchitta, 2006). In order to test if the RQC is an important part of the antigen processing machinery siRNA‐mediated knockdown of one of its central components was performed. B8‐Db cells were transfected with siRNA targeting the E3 ligase LTN1 and then tested for their potential to present NP118‐128 after infection with rVV‐NP (Fig. 17A). No difference could be observed between control cells and cells devoid of LTN1. Furthermore, surface expression of the MHC class I molecules H‐2Db and H‐2Dd was unaffected in LTN1 knockdown cells (Fig. 17B). The full depletion of LTN1 by knockdown was confirmed by western blot (Fig. 17C) supporting the conclusion that LTN1 is not essential for class I peptide generation. Finally, it was addressed whether ribosome‐association of p97 is affected by IFN‐γ stimulation. Ribosomes isolated from HEK293 cells stimulated with IFN‐γ for two days or left untreated did, however, not differ in their amount of associated p97 (Fig. 17D). In western blot analysis of cell lysates and isolated ribosomes the immunoproteasome subunit LMP7 was used as positive control for IFN‐γ stimulation while the ribosome subunit RPL‐7 was used as a ribosome marker. Taken together,

3. Results these data do not support a central role of the RQC in antigen processing. Most likely, bulk antigen derives from other sources than from stalled ribosomes.

Figure 17: E3 ligase LTN1 is dispensable for antigen processing. (A) Presentation of NP118‐126 on B8‐Db cells transfected with LTN1 siRNA or control siRNA infected with rVV‐NP (left), rVV‐NP118‐126 (right), or VV‐WR (MOI = 10). NP118‐126‐specific CTLs were added to infected cells in indicated effector to stimulator ratios (E/S) and percentage of IFN‐γ+ of CD8+ cells was determined as a measure for epitope presentation. Values represent mean ± SD of duplicate determinations (B) MHC class I surface expression of B8‐Db cells transfected with LTN1 siRNA or control siRNA. Control cells were left unstained. (C) Western blot analysis of whole cell lysates prepared from cells transfected with LTN1 or control siRNA. Tubulin was used as loading control. (D) Western blot analysis of whole cell lysates and ribosomes prepared from HEK293 cells stimulated with IFN‐γ for 2 d or left untreated. RPL‐7 was used as loading control. For all experiments one out of three independent experiments is shown.

3. Results

3.2 The role of BAG6 in antigen processing and presentation on MHC class I MHC class I surface expression is independent of BAG6 If DRiPs result from incomplete capture of nascent polypeptides by chaperones, deletion of chaperones should increase DRiP rates. A higher proportion of DRiPs should in turn increase MHC class I surface expression due to increased peptide supply. On the other hand, deletion of chaperones could lead to inefficient shuttling and degradation of antigen.

Figure 18: Knockdown or overexpression of BAG6 has no influence on MHC class I surface expression. (A and C) B8‐Db cells (A) and HeLa cells (C) were transfected with BAG6 siRNA or control siRNA and surface expression of MHC class I molecules H‐2Db and H‐2Dd (A) or HLA‐A,B,C (C) was analyzed by flow cytometry. Control cells were left unstained. (B and D) Western blot analysis of whole cell lysates prepared from B8‐Db cells (B) or HeLa cells (D) transfected with BAG6 or control siRNA. (E) HeLa cells were transfected with a BAG6 expression construct or an empty plasmid (control) and surface expression of MHC class I molecules HLA‐ A,B,C was analyzed by flow cytometry. Control cells were left unstained. (F) Western blot analysis of whole cell lysates prepared from HeLa cells transfected with a BAG6 expression construct or an empty control plasmid. Tubulin was used as loading control. For all experiments one out of two independent experiments is shown.

To analyze the influence of BAG6 on MHC class I surface expression the murine fibroblast cell line B8‐Db (H‐2d + H‐2Db) and HeLa cells were subjected to BAG6 knockdown. Subsequently, surface expression of MHC class I molecules was determined by flow cytometry. No difference in surface expression of the MHC class I molecules H‐2Db and H‐2Dd could be observed between

3. Results

B8‐Db cells transfected with BAG6 siRNA and cells transfected with control siRNA (Fig. 18A). Successful knockdown leading to ablation of the BAG6 protein below the detection limit was confirmed by SDS‐PAGE and immunoblotting (Fig. 18B). The same result was obtained with HeLa cells in which total MHC class I surface expression (HLA‐A,B,C) did not differ between BAG6 knockdown and control cells (Fig. 18C). Very successful knockdown in HeLa cells was again confirmed by SDS‐PAGE and immunoblotting (Fig. 18D).

To rule out limiting effects of BAG6 on MHC class I surface expression, overexpression was performed in HeLa cells. Cells were transiently transfected with an expression construct encoding for BAG6 and subjected to flow cytometric analysis 24 h post transfection (Fig. 18E). No influence on HLA‐A,B,C surface expression could be detected although lysates of transfected cells subjected to SDS‐PAGE and immunoblotting revealed very high expression levels of BAG6 compared to control cells (Fig. 18F).

Taken together, no influence of BAG6 on MHC class I surface expression could be observed. Loss of BAG6 might be effectively compensated by other cytosolic chaperones in the setting applied here. On the other hand, BAG6 is not a limiting factor in antigen processing since its overexpression had no effect on MHC class I surface expression.

BAG6 is not transcriptionally regulated after IFN‐γ stimulation Stimulation of cells with the cytokine IFN‐γ induces or enhances transcription of several genes whose products delay viral replication and enhance peptide presentation on MHC class I (Platanias, 2005; Sen and Lengyel, 1992). Like many of these genes, BAG6 is encoded in the MHC locus and cytokine‐regulated expression control could be possible (Banerji et al., 1990). Yet, stimulation of HeLa cells with IFN‐γ for 24 h or 48 h did not induce upregulation of BAG6 mRNA (Fig. 19, upper left panel). In contrast, expression of the immunoproteasome subunit LMP7, a well‐known IFN‐regulated gene, was strongly upregulated after stimulation (Fig. 19A, upper right panel). Although Kamper et al. found expression of the bag6 gene to be upregulated after stimulation of the melanoma cell line MelJuSo and the lung fibroblast cell line IMRS with IFN‐γ, this finding could neither be confirmed for HeLa cells, nor for the other two cell lines (Fig. 19B and C) (Kamper et al., 2012a). Overall, these results do not point towards a central role of BAG6 in an antiviral immune response.

3. Results

Figure 19: BAG6 is not transcriptionally regulated after IFN‐γ stimulation. HeLa cells (A), IMRS cells (B) or MelJuSo cells (C) were stimulated with IFN‐γ for 24 h or 48 h or were left unstimulated and mRNA levels of BAG6 and LMP7 were determined by real‐time RT‐PCR. Expression levels were calculated relative to unstimulated cells and normalized to HPRT expression. Bar graphs show means ±SD from three independent experiments. ns (not significant) P > 0.05, **P < 0.01, ***P < 0.001 (unpaired two‐tailed Student’s t test).

Presentation of LCMV‐derived epitopes is independent of BAG6 To address whether generation of virus‐derived peptide epitopes is dependent on BAG6, antigen presentation in LCMV‐infected cells after BAG6 knockdown was investigated. B8‐Db cells were transfected with BAG6 siRNA and subsequently infected with LCMV‐WE for 2, 4, or 5 h. Peptides presented on MHC class I were detected with peptide‐specific CTL lines against the epitopes NP118‐126 (H‐2Dd), NP396‐404 (H‐2Db) and GP276‐286 (H‐2Db) by staining for the CTL‐surface marker CD8 and intracellular IFN‐γ staining. GP276‐286 represents an epitope derived from the secretory pathway, whereas NP118‐126 and NP396‐404 are derived from a cytosolic protein. No difference in presentation of any of these epitopes between BAG6 knockdown cells and control cells could be observed (Fig. 20A).

3. Results

Figure 20: BAG6 is dispensable for presentation of virus‐derived epitopes. (A and B) Cells were transfected with BAG6 siRNA or control siRNA and infected with virus (MOI = 10). Epitope‐specific CTLs were added to infected cells in different effector to stimulator ratios (E/S) for 3 h and percentage of IFN‐γ+ of CD8+ cells was determined by flow cytometry. Values represent mean ± SD of duplicate determinations. (A) B8‐Db cells infected with LCMV‐WE for 2, 4, or 5 h were analyzed for presentation of NP396‐404, GP276‐286, or NP118‐ 126. Control cells were left uninfected. (B) Transfected AAD MEFs stimulated with IFN‐γ for 2 d and infected with rVV‐Tyr or VV‐WR for 5 h were analyzed for the presentation of Tyr‐369‐377(D). Control cells were infected with VV‐WR. (C) Western blot analysis of whole cell lysates prepared from AAD MEFs transfected with BAG6 or control siRNA and stimulated with IFN‐γ for 2 d. Tubulin was used as loading control. (D) HEK293 cells were transiently transfected with an expression construct for tyrosinase (Tyr)‐Myc‐Flag or empty plasmid (control) and treated with 10 μM MG132 or solvent DMSO for 5 h. Cell lysates were subjected to immunoprecipitation (IP) with anti‐Myc antibody followed by western blot (WB) analysis. Load indicates protein expression levels in total protein lysates. For all experiments one out of two independent experiments is shown.

3. Results

Presentation of retrotranslocated tyrosinase‐derived epitope is not affected by BAG6 knockdown BAG6 maintains retrotranslocated ER‐associated protein degradation (ERAD) substrates in a soluble state and thereby allows proteasomal degradation of these substrates (Claessen and Ploegh, 2011; Wang et al., 2011). To study the presentation of an epitope derived from a retrotranslocated protein we chose the tyrosinase epitope Tyr369‐377(D). Tyrosinase is a membrane‐associated glycoprotein produced in melanocytes and melanoma cells that is targeted to the ER (Jimbow et al., 2000; Petrescu et al., 2000). Misfolded tyrosinase is exported back to the cytosol where it is deglycosylated leading to the conversion of N371 to D371 through deamination (Skipper et al., 1996). The HLA‐A*0201‐restricted epitope Tyr369‐377(D) spans across this modified amino acid and allows analysis of an epitope derived exclusively from a retrotranslocated antigen pool. To study presentation of Tyr369‐377(D), the transgenic mouse strain AAD (H‐2b, HLA‐A2.1/H2‐Dd) expressing a hybrid MHC class I molecule consisting of the α‐1 and α‐2 domain of human HLA‐A*0201 and the α‐3 transmembrane and cytoplasmic domain of mouse H‐2Dd was used (Newberg et al., 1996). AAD mouse embryonic fibroblasts (MEFs) were transfected with BAG6 siRNA and stimulated with IFN‐γ to induce MHC class I expression. Cells were then infected with recombinant vaccinia virus expressing tyrosinase (rVV‐Tyr) for 5 h. Tyr369‐377(D)‐MHC complexes were detected with specific CTL lines generated from AAD mice and analyzed by staining for CD8 and intracellular IFN‐γ. BAG6 knockdown did not alter Tyr369‐377(D) presentation (Fig. 20B) while no detectable BAG6 protein remained after knockdown (Fig. 20C). Although BAG6 seemed to be dispensable for Tyr369‐377(D) presentation, endogenous BAG6 could be pulled down with transiently expressed tyrosinase in HEK293 cells (Fig. 20D). This interaction was even enhanced when proteasomal degradation was blocked by MG132 treatment indicating a role of BAG6 in tyrosinase degradation.

Overall, no evidence for an essential role of BAG6 in generating LCMV‐ or tyrosinase‐derived ligands for the presentation on MHC class I could be found. However, the possibility that other epitopes might depend on BAG6 for effective processing and presentation cannot be formerly excluded.

3. Results

3.3 The influence of immunoproteasome deficiency on NF‐kB activation Expression of immunoproteasome subunits in peritoneal macrophages and MEFs As an approach for assessing the influence of immunoproteasome subunits on NF‐kB activation two different types of primary cells were isolated from wild type, LMP7‐/‐/MECL‐1‐/‐ double knockout mice (L7M‐/‐), and LMP2‐/‐ mice. Thioglycollate‐elicited peritoneal macrophages (pMΦs) and MEFs were chosen in order to include cell types of the hematopoietic and non‐ hematopoietic lineage, respectively. Isolated pMΦs were cultured in the presence of IFN‐γ for two days to further upregulate immunoproteasome expression. To determine proteasome subunit composition, a new combination of immunoprecipitation and 2D‐gel electrophoresis was tested for its applicability. Immunoprecipitation using anti‐proteasome antibodies cross‐ linked to protein A gel was used to rapidly enrich proteasomes and circumvent the previously reported purification procedure requiring a large number of cells (Schmidtke et al., 2000). The subunits of core particles were separated by 2D‐gel electrophoresis and visualized by silver staining (Fig. 21A). From the catalytically active subunits only β1 and LMP7 could be detected reliably. Therefore, the subunit composition was analyzed with SDS‐PAGE and western blot instead (Fig. 21B). Unstimulated wild type pMΦs strongly express LMP7, which seems not to be further upregulated after stimulation with IFN‐γ. In contrast, expression of LMP2 is rather low in unstimulated cells and is strongly induced after stimulation. These results indicate that a proportion of proteasomes in unstimulated pMΦs has a mixed subunit composition consisting of LMP7 together with the constitutive subunits β1 and β2. This fits well with data published by Guillaume et al., who found about 40% of proteasomes present in dendritic cells to have incorporated only LMP7 but not LMP2 or MECL‐1 (Guillaume et al., 2010). After stimulation, the subunit composition shifts further towards immunoproteasomes. However, expression of MECL‐1 and β2 was not tested and it is therefore not possible to discriminate the proportion of full immunoproteasomes and complexes containing LMP7 and LMP2 together with β2. In agreement with previous reports, L7M‐/‐ pMΦs show a strongly reduced incorporation of LMP2 and accumulation of the unprocessed LMP2 precursor (Griffin et al., 1998). Hence, L7M‐/‐ pMΦ almost exclusively express constitutive proteasomes. In contrast, LMP2‐/‐ cells show LMP7 expression levels similar to wild type cells.

3. Results

Figure 21: Proteasome subunit composition in pMΦs and MEFs. (A) pMΦs isolated from C57BL/6, L7M‐/‐, and LMP2‐/‐ mice were stimulated with 200 U/ml IFN‐γ for 2 days. 20S proteasome core particles were immunoprecipitated from whole cell lysates using anti‐proteasome antibodies and resolved by 2D nonequilibrium pH gel electrophoresis (NEPHGE)/SDS‐PAGE. Proteins were visualized by silver staining. Arrows indicate position of the subunits β1 and LMP7. Position and identity of other spots have been reported previously (Groettrup et al., 1996). (B and C) pMΦs (B) or MEFs (C) isolated from C57BL/6, L7M‐/‐, and LMP2‐ /‐ mice were cultured in the presence of 200 U/ml IFN‐γ for 2 d or left untreated. Whole cell lysates were prepared and analyzed by western blot. The catalytically inactive subunit α6 was used as loading control. For both experiments one out of three independent experiments is shown.

Unlike pMΦs, MEFs do not express immunoproteasome subunits in the absence of IFN‐γ stimulation (Fig. C). After stimulation, expression of LMP7 and LMP2 is induced, while the respective constitutive subunits β5 and β1 are downregulated. Similar to pMΦs, L7M‐/‐ MEFs display strongly reduced incorporation of LMP2, whereas LMP2‐/‐ MEFs show reduced expression of LMP7.

Degradation of IkBα is not affected in immunoproteasome knockout cells After having analyzed the proteasome composition in pMΦs and MEFs, stimulus‐induced degradation of IkBα was investigated next. At this point, the proteasome is directly involved in the signaling cascade. Incomplete or delayed degradation of IkBα leads to retention of NF‐κB in the cytoplasm. Thus, any alteration affecting the extent or kinetic of IkBα degradation would imminently affect all downstream signaling events.

3. Results

Figure 22: Stimulus‐induced degradation of IkBα in pMΦs and MEFs. pMΦs (A and B) or MEFs (C) isolated from C57BL/6, L7M‐/‐, and LMP2‐/‐ mice were incubated in the presence of 200 U/ml IFN‐γ for 2 days. Cells were then stimulated with 200 ng/ml LPS (A) or 100 U/ml TNF‐α (B and C). Whole cell lysates were prepared at the indicated time points and subjected to western blot analysis. Tubulin was used as loading control. For all experiments one out of three independent experiments is shown. pMΦs from C57BL/6, L7M‐/‐, and LMP2‐/‐ mice were incubated in the presence of IFN‐γ for two days to further upregulate immunoproteasome expression. Degradation of IkBα was induced by stimulating the cells with the TLR4 ligand LPS and monitored by SDS‐PAGE and western blot. No difference in IkBα degradation could be detected between cells derived from wild type C57BL/6 mice and either L7M‐/‐ or LMP2‐/‐ mice (Fig. 22A). To not only rely on a single stimulus, pMΦs were also stimulated with tumor necrosis factor (TNF)‐α. Due to a different upstream signaling cascade, receptor binding of TNF‐α triggers a faster response compared to LPS. Here, maximum IkBα degradation is already reached after 15 min (Fig. 22B). Nevertheless, immunoproteasome deficiency had no apparent influence on IkBα degradation.

3. Results

Similar to pMΦs, MEFs stimulated with TNF‐α displayed a fast response, which is evidenced by the rapid degradation of IkBα (Fig. 22 C). Although MEFs were pre‐stimulated with IFN‐γ to upregulate immunoproteasome expression, IkBα degradation was not influenced by either LMP7 and MECL‐1 or LMP2 deficiency.

Even though incorporation of immunoproteasome subunits alters the cleavage specificity and substrate binding pockets of the 20S core particle (Huber et al., 2012; Toes et al., 2001) there seems to be no influence of these subunits on the degradation of IkBα.

Nuclear translocation of free NF‐kB and transactivation of target genes is normal in immunoproteasome knockout cells The final steps in NF‐κB signaling are nuclear translocation of free NF‐κB dimers and transactivation of target genes. To test whether the immunoproteasome has any influence on NF‐κB signaling downstream of IkBα, active nuclear NF‐κB was quantified using an electrophoretic mobility shift assay (EMSA). MEFs derived from C57BL/6, L7M‐/‐, and LMP2‐/‐ mice were incubated in the presence of IFN‐γ for two days followed by stimulation with TNF‐ α. Nuclear extracts were prepared after different time points and subjected to EMSA using an NF‐κB binding site probe (Fig. 23A). As expected from the results obtained for IkBα degradation, no difference in the amount of active NF‐κB could be detected between wild type and immunoproteasome knockout cell lines at any time point.

Cytokines mediate important effector functions of the immune system. Hence, many genes encoding cytokines contain NF‐κB responsive elements in their promotor region. As a measure for transactivation, secretion of the pro‐inflammatory cytokines TNF‐α and IL‐6 as well as the anti‐inflammatory cytokine IL‐10 by IFN‐γ pre‐conditioned and LPS‐stimulated pMΦs was quantified using ELISA (Fig. 23B). Concentration of none of the cytokines differed in a statistically significant way between supernatants of C57BL/6‐, L7M‐/‐‐, and LMP2‐/‐‐derived cells. Thus, transactivation is not influenced by immunoproteasome deficiency.

Overall, IkB degradation and signaling events further downstream of this step are not influenced by the immunoproteasome subunits LMP7, MECL‐1, and LMP2. Collectively, these data strongly support a model of NF‐κB activation being independent of the 20S proteasome subunit composition.

3. Results

Figure 23: Nuclear translocation of free NF‐kB and transactivation of target genes. (A) MEFs isolated from C57BL/6, L7M‐/‐, and LMP2‐/‐ mice were incubated in the presence of 200 U/ml IFN‐γ for 2 days. Cells were then stimulated with 100 U/ml TNF‐α for the indicated time points. Nuclear extracts were prepared and analyzed by EMSA using an NF‐κB binding site probe. As a loading control, nuclear extracts were also subjected to western blot (WB) analysis using a Lamin A/C antibody. One out of three independent experiments is shown. (B) pMΦs isolated from C57BL/6, L7M‐/‐, and LMP2‐/‐, mice were incubated in the presence of 200 U/ml IFN‐γ for 2 days. Cells were then stimulated with 200 ng/ml LPS for 20 h or left untreated (control). Supernatants were collected and concentration of the cytokines TNF‐α, IL‐6, and IL‐10 was analyzed by ELISA in triplicates. Graphs show pooled data from at least five independent experiments including mean. ns (not significant) P > 0.05 (unpaired two‐tailed Student’s t test). EMSA = electrophoretic mobility shift assay.

4. Discussion

4.1 Protein homeostasis and immune regulation The proteome of a cell is in constant flux and its composition is defined by the coordinated action of the cellular proteostasis networks. Human diseases associated with imbalances or failure of these networks further underscore the importance of proteostasis for cellular and organismal health. As a consequence, modulation of proteostasis has emerged as a promising strategy for the treatment of various diseases such as cancer, autoimmune conditions, or neurodegeneration (Anderson et al., 2015; Balch et al., 2008; Basler et al., 2015; Lindquist and Kelly, 2011; Powers et al., 2009).

As proteins are continually turned over the bulk of cellular proteins is degraded via the UPS (Rock et al., 1994). This route of degradation is especially employed for the removal of short‐ lived regulatory proteins and defective polypeptides. Besides amino acid recycling, peptides produced through proteasomal degradation can be used for antigen presentation on MHC class I molecules. However, antigen presentation is a nonessential proteasome function, which was added during vertebrate evolution (Plytycz and Seljelid, 1998). Nevertheless, existence of the immunoproteasome, which produces more peptides suited for the binding to MHC class I molecules, indicates a certain adaptation of the UPS machinery for the purpose of antigen presentation (Sijts and Kloetzel, 2011). This study was focused on elucidating the influence of factors upstream of the proteasome on antigen processing so as to further broaden the knowledge about the antigen processing machinery.

Another immunity‐relevant function of the proteasome is the degradation of signaling molecules. One of the best‐studied signal transduction pathways ultimately depending on the proteasome is the NF‐κB pathway. Here, proteasomes are required for the processing of NF‐κB precursors as well as the signal‐induced degradation of IkBα. This study was further concerned with the influence of immunoproteasome deficiency on canonical NF‐κB activation.

Antigen processing – which factors are involved? Antigen processing and presentation have crucial impacts on the efficiency and outcome of an immune response. Especially during infection with intracellular pathogens, rapid CTL‐mediated killing of infected cells can be critical for the containment of an infection. The consequences of failure in recognition of infected cells can be persistent or even fatal infection. A possibility to rapidly detect translation of foreign proteins within a cell is the use of DRiPs as substrates for

4. Discussion antigen processing. As a consequence, the “immunopeptidome” surveyed by CTLs mirrors translation rates rather than protein concentrations (Caron et al., 2011; Fortier et al., 2008; Khan et al., 2001a). Bourdetsky and colleagues could indeed show that peptides presented on MHC class I shifted faster from light to heavy isotope forms than their source proteins when cells were switched to media containing amino acids with heavy isotopes (Bourdetsky et al., 2014). This raises the question whether DRiPs and DRiP‐derived peptides are simple by‐ products of protein synthesis or if cells have evolved mechanisms regulating DRiP formation and/or processing. Yewdell and Niccitta suggested the existence of “immunoribsomes” specialized on the production of substrates for the MHC class I‐restricted antigen processing pathway (Yewdell and Nicchitta, 2006). Such immunoribosomes may have a different subunit composition or localize in a different intracellular compartment than conventional ribosomes. Although intellectually attractive, this hypothesis has not yet been confirmed experimentally. Instead, ribosomes are more likely to function as a platform for associated factors acting upon the nascent chain and influencing its fate (Kramer et al., 2009). While more and more factors are being identified to mediate co‐translational and post‐translational protein quality control little is known about the contribution of these factors to antigen processing.

Specifically designed to recognize non‐native polypeptides chaperones are amongst the candidates ideally suited for DRiP processing. Chaperones mediate an important layer of quality control and can facilitate degradation of non‐native polypeptides through interaction with factors of the ubiquitin‐proteasome system (Hohfeld et al., 2001; Rodrigo‐Brenni and Hegde, 2012). For example, the well‐studied chaperones HSP70/HSP40 and HSP90 can capture unfolded polypeptides and mediate substrate ubiquitylation through interaction with the E3 ligase carboxyl terminus of HSC70 interacting protein (CHIP) (Murata et al., 2003; Murata et al., 2001). Rodrigo‐Brenni and Hegde suggested that chaperones associated with a ubiquitylated substrate become a stabilized targeting complex (Rodrigo‐Brenni et al., 2014). With potential help of other cofactors like BAG1, these complexes might then shuttle substrates directly to the proteasome while safely sequestering the substrate from the cytosol at the same time (Arndt et al., 2007). The triaging decision between folding and degradation not only affects cellular proteostasis, it most likely also influences the potential pool of substrates for antigen presentation. The mechanisms governing this decision are, however, still a matter of speculation. Parameters discussed include the dwell time of binding to chaperones (possibly influenced by the extent of hydrophobicity exposed), stochastic association of factors favoring either folding or degradation, and a special conformation that is adopted by the chaperone if

4. Discussion bound to a difficult to fold substrate (Buchberger et al., 2010). In favor of the stochastic model are results obtained through modifying the availability of folding or degradation factors. Knockdown or inhibition of CHIP increased the proportion of a difficult to fold substrate to mature (Grove et al., 2009; Younger et al., 2004). Along the same line, overexpression of HSC70 decreased the level of RDP degradation, whereas expression of a DN mutant increased it (Qian et al., 2006). While chaperones are certainly important for recognition and degradation of DRiPs, it is important to mention that some E3 ligases are able to directly recognize nonnative features of substrates (Fredrickson et al., 2011; Rosenbaum et al., 2011; Sato et al., 1998).

4.2 Major role of p97 in antigen processing and presentation on MHC class I The ATP‐driven chaperone p97 is a critical factor for the handling of ubiquitylated proteins and teams up with the proteasome for substrate degradation. The rapid accumulation of polyubiquitylated proteins upon disruption of p97 activity seen in this and many other studies implicates a significant contribution of p97 to efficient proteasomal degradation. The fact that expression of p97 and the 26S proteasome can be induced coordinately by the transcription factor NRF1 further suggests that both enzymes build a functional entity (Sha and Goldberg, 2014).

While the contribution of proteasomes to antigen processing is well established, the role of p97, in particular for direct MHC class I presentation, is discussed quite controversially. Ackermann et al. showed that overexpression of DN p97 inhibited cross‐presentation of ovalbumin‐derived SIINFEKL while direct presentation on MHC class I was not affected (Ackerman et al., 2006). However, expression of DN p97 may only partially inhibit cellular p97 activity below a threshold sufficient to suppress direct presentation (Beskow et al., 2009). Using a stabilized precursor protein containing the SIINFEKL epitope, Dolan et al. found that presentation on H‐2Kb, classified as DRiP‐presentation in this study, was suppressed by EerI. Still, only a marginal effect of EerI on H‐2Kb and H‐2Db cell surface expression was apparent (Dolan et al., 2011). A follow‐up study by Palmer and Dolan extended investigations in the same experimental system by using the p97 inhibitor DBeQ and overexpression of DN p97. In contrast to the first study, no effect of p97 inhibition or DN expression on SIINFEKL/Kb presentation or class I surface expression was recorded (Palmer and Dolan, 2013). However, DBeQ was used at a concentration of 1 µM or 5 µM, which only partially inhibits p97 (Chou et al., 2011). Moreover, cells expressing transfected DN p97 were not enriched for subsequent experiments. In the present study, it was found that antigen processing depends on active p97

4. Discussion in order to give rise to peptides presented on MHC class I molecules. In contrast to previous studies, bulk MHC class I surface expression as well as the presentation of several different peptide epitopes were analyzed. In both setups, contribution of peptide availability was controlled by external peptide supply or minigene expressing rVVs, respectively. To not only rely on one approach to interfere with p97 activity, a combined strategy using two chemical p97 inhibitors and expression of transfected DN p97 followed by magnetic isolation or gating on transfected cells was applied.

Inhibition of p97 is currently one of the avenues followed in search of new chemotherapeutic reagents. As such, it can rapidly disturb proteostasis and induce cell death (Chou et al., 2011; Chou et al., 2013; Magnaghi et al., 2013). In order to avoid cytotoxic effects, the incubation times with inhibitors and expression of the p97 mutant were kept as short as possible. Indeed, within incubation times applied for the inhibitor DBeQ no enhanced cell death could be observed. Moreover, MHC class I surface expression in acid‐washed and DBeQ‐treated cells could be restored through external peptide application indicating undisturbed trafficking of MHC class I molecules throughout the cellular secretory pathway. The same result was observed for cells transfected with DN p97. Unfortunately, cell viability after inhibitor exposure could not be assessed with EerI due to high emission of fluorescence. In antigen presentation assays, EerI suppressed generation of peptide‐MHC complexes indicating a strong relation between p97 activity and antigen processing. Still, presentation could not be entirely restored when only the corresponding peptide epitope was expressed. Hence, the possibility that the effects seen with EerI are in part due to inhibition of other cellular pathways cannot be entirely excluded. EerI is still an inhibitor with poorly defined mechanism and apart from inhibiting p97 it also affects protein synthesis and Sec61‐mediated protein translocation into the endoplasmic reticulum (Aletrari et al., 2011; Cross et al., 2009a). Similar to EerI treatment, expression of DN p97 in presentation assays did not always allow full restoration of presentation when only the peptide epitope was expressed. Since expression of GFP was not reduced in these cells a negative effect of DN p97 on protein synthesis could be ruled out. Interestingly, restoration of NP118‐126 was least in EerI treated cells as well as in cells transfected with the mutant compared to other peptides. This could be a hint for trafficking or stability of the corresponding MHC class I molecule H‐2Ld being affected. Although some side effects of disrupted p97 activity on other cellular processes than antigen processing are not unlikely, the robust effects seen throughout the experiments clearly indicate an important role of p97 in MHC class I‐restricted antigen processing.

4. Discussion

Interestingly, the antigens tested were affected by p97 inhibition or ablation to a different degree. Indeed, the extent of how much a UPS substrate depends on p97 for its degradation seems to be determined by substrate localization, structure and solubility (Beskow et al., 2009; Gallagher et al., 2014). Hence, dependence on p97 most likely also varies among DRiPs and is dictated by the nature of these defective and/or newly synthesized chains.

While the majority of antigenic peptides produced by the proteasome are likely to be derived from ubiquitylated polypeptides, several studies also support an important role for ubiquitin‐ independent antigen processing by the proteasome (Cox et al., 1995; Huang et al., 2011; Qian et al., 2006). Since p97 seems to be part of the antigen processing machinery and its inhibition causes accumulation of polyubiquitylated proteins, our results argue in favor of a ubiquitin‐ dependent model. Interestingly, introduction of a C‐terminal unstructured peptide renders the degradation of a ubiquitin‐fusion degradation protein independent of p97 and reduces the amount of further ubiquitin moieties needed (Beskow et al., 2009; Godderz et al., 2015). Gödderz and colleagues concluded from their data, that the degree of ubiquitylation required for degradation could depend on factors upstream of the proteasome rather than on the proteasome itself (Godderz et al., 2015). Indeed, p97 conferred a higher unfolding activity onto a penta‐ubiquitylated substrate compared to a mono‐ubiquitylated substrate (Song et al., 2015). Thus, a certain degree of ubiquitylation seems to qualify substrates for engagement of the p97 unfolding activity, which prepares substrates for proteasomal degradation.

Even though p97 has some affinity to ubiquitin itself, it binds to substrates largely via cofactors containing dedicated ubiquitin‐binding domains (Schuberth and Buchberger, 2008; Ye, 2006). Depending on the cofactors bound, the activity of p97 can be directed towards different cellular activities (Bruderer et al., 2004; Meyer et al., 2000). Similar to the exchange of proteasome subunits, expression of p97 cofactors could potentially be adapted under inflammatory conditions. Thereby, p97 activity could be concentrated to overall enhance substrate degradation or to specifically enhance degradation of a certain set of substrates like DRiPs. However, this hypothesis remains to be tested.

Ever since the DRiP hypothesis was postulated, origin and processing of DRiPs are being debated (Anton and Yewdell, 2014; Rock et al., 2014). Coupling of protein synthesis and protein degradation followed by presentation on MHC class I would be the fastest way to ensure detection of pathogenic translation products. In fact, about 12‐15% of nascent chains in HEK293 cells are co‐translationally modified with ubiquitin (Wang et al., 2013). The best‐ characterized pathway mediating co‐translational ubiquitylation and substrate extraction is the 69

4. Discussion

RQC pathway acting upon stalled ribosomes (Bengtson and Joazeiro, 2010; Brandman et al., 2012; Verma et al., 2013). Since p97 is an essential component of the RQC needed for the extraction of nascent chains from stalled ribosomes, the impact of this pathway on antigen presentation was analyzed. Knockdown of the RQC‐specific E3 ligase LTN1 affected neither bulk MHC class I surface expression nor presentation of a virus derived epitope. Moreover, p97 association with ribosomes was not enhanced after stimulation of cells with IFN‐γ. Together, these results fail to support a major contribution of nascent chains derived from stalled ribosomes to antigen presentation. Ribosomes stall due to translation of non‐stop or damaged mRNA and initiate mRNA quality control pathways referred to as nonstop‐decay (NSD) and no‐ go decay (NGD), respectively (Shoemaker and Green, 2012). NSD, NGD, and other RNA surveillance mechanisms were suggested to serve as an alternative source of antigen for presentation on MHC class I (Anton and Yewdell, 2014; Apcher et al., 2015). After all, the contribution of these sources might only be of minor importance.

Based on the combined pharmacological and genetic data a major and essential role for p97 in class I antigen processing can be inferred. Defective polypeptides (including DRiPs) are detected by the various cellular quality control pathways and are subjected to ubiquitylation (Amm et al., 2014). Then, the different pathways likely merge at p97, which extracts and unfolds substrates if necessary and allows ubiquitin chain editing before substrates are degraded by the proteasome. Without p97 activity, substrates depending on p97 to gain access to the cytosol or to become separated from binding partner would be entirely missing in the antigen pool. On the other hand, substrates unable to be processed by the proteasome due to tight folding would potentially clog the degradation machinery. Thus, the role of p97 in antigen processing could be explained as a combination of making substrates available, preparing the substrates of degradation, and keeping the proteasome accessible.

4.3 BAG6 is dispensable for antigen processing and presentation on MHC class I The chaperone BAG6 is involved in post‐translational insertion of tail‐anchored proteins into the ER membrane (Leznicki et al., 2010; Mariappan et al., 2010). While assisting in protein biosynthesis on the one hand, BAG6 also mediates degradation of mislocalized nascent chains (Hessa et al., 2011) and ERAD substrates (Claessen and Ploegh, 2011; Claessen et al., 2014; Payapilly and High, 2014; Wang et al., 2011). Moreover, Rodrigo‐Brenni et al. recently described RNF126 as an E3 ligase interacting with BAG6 and specifically ubiquitylating

4. Discussion chaperone bound client proteins (Rodrigo‐Brenni et al., 2014). The fact that BAG6 has been described as a “holdase” conferring no evident folding activity onto its substrates makes this chaperone even more interesting with regard to protein quality control and degradation (Wang et al., 2011). The absence of a yeast homolog could furthermore point towards a coevolution with the antigen processing machinery.

Here, it was addressed whether BAG6 can influence antigen processing and presentation in the MHC class I‐restricted pathway. Knockdown of BAG6 had no influence on MHC class I cell surface expression indicating that loss of BAG6 can be effectively counterbalanced by other cytosolic chaperones under physiological conditions. This result is in contrast to a previous report by Minami et al. who found reduced MHC class I cell surface expression in BAG6 knockdown cells (Minami et al., 2010). There is currently no explanation for this discrepancy, especially, since no residual BAG6 expression could be detected after transfection with BAG6 siRNA. Presentation of three LCMV‐derived epitopes was entirely unaffected although one of the epitopes analyzed (GP276‐286) derives from an ER‐targeted glycoprotein. Hence, degradation of the GP276‐286 precursor could be handled by the ERAD pathway or the mislocalized protein disposal pathway where BAG6 is involved. Strikingly, the tyrosinase‐ derived epitope Tyr369‐377(D), which strictly requires retrotranslocation from the ER in order to be presented, was also not affected by BAG6 knockdown. Together, these results indicate that BAG6 is dispensable for antigen processing and presentation on MHC class I. Enhanced interaction of BAG6 and tyrosinase under proteasome inhibition on the other hand still indicates that BAG6 could be a potential player in antigen processing. The redundancy in cellular chaperone networks could, however, mask the contribution of BAG6 to this pathway. BAG6 together with its E3 ligase RNF126 indeed resembles the HSP70/HSP90 chaperone system, which cooperates with the E3 ligase CHIP to promote substrate degradation (Connell et al., 2001; Rodrigo‐Brenni et al., 2014). The finding that BAG6 substrates could still be degraded in the absence of BAG6 and/or RNF126, albeit more slowly and less efficiently, further indicates the existence of one or several redundant pathways (Rodrigo‐Brenni et al., 2014). Moreover, BAG6 and HSP70 seem to compete for substrate binding as BAG6 could prevent HSP70‐mediated refolding of denatured luciferase in an in vitro assay (Wang et al., 2011). Although BAG6 preferably binds to longer hydrophobic patches (Mariappan et al., 2010) and HSP70/HSP90 to shorter hydrophobic patches (Jackson, 2013; Rudiger et al., 1997), considerable overlap in both systems is likely. Definition of the potential proportion of antigen that is handled by BAG6 would therefore be possible if BAG6‐mediated degradation could be

4. Discussion inhibited after substrate binding to the chaperone. Thereby, antigen could be trapped on BAG6 and degradation via a different route would be blocked. Possible strategies to interrupt BAG‐ mediated degradation were recently described by Rodrigo‐Brenni and colleagues and included knockdown of the E3 ligase RNF126 and expression of an N‐terminal fragment of RNF126 lacking ligase activity. Moreover, a BAG6 mutant lacking the UBL domain was still able to bind substrates while lacking the interaction with RNF126 (Rodrigo‐Brenni et al., 2014). Testing the effect of these approaches on antigen processing and presentation would be an interesting avenue to follow up on this project.

Compartmentalized antigen processing was first suggested by Lev et al. (Lev et al., 2010) and could be another explanation for our findings. In this case, BAG6 might be involved in degradation of bulk proteins, while being dispensable for epitope generation. Since no work to date has identified such a compartment, this possibility can be regarded as rather unlikely.

BAG6 is encoded in a cluster of immune‐relevant genes in the MHC class III locus (Banerji et al., 1990). Yet, stimulation with the anti‐viral cytokine IFN‐γ had no impact on mRNA levels of BAG6. In contrast, the immunoproteasome subunit LMP7 was highly upregulated after stimulation. This finding further supports the hypothesis of a mainly non‐immune cellular function of BAG6. Besides transcriptional regulation, chaperone activity could also be regulated through altered subcellular distribution. BAG6 is, to a large extent, retained in the cytoplasm though association with its cofactor TCR35 that covers up the nuclear localization sequence in BAG6 (Wang et al., 2011). Transcriptional regulation of TRC35 after stimulation with IFN‐γ was tested once (data not shown) and also appeared unaltered in comparison to unstimulated cells. A second possibility to regulate cellular localization of BAG6 can be alternative splicing of BAG6 mRNA. Several alternative splice variants with different cellular localization pattern have been reported previously (Kamper et al., 2012b). However, if alternative splicing of BAG6 can be induced through IFN‐γ stimulation remains to be tested.

Taken together, the results obtained during this study indicate that BAG6 is dispensable for antigen processing and presentation on MHC class I. Nevertheless, it might become essential under conditions of limited chaperone availability.

4.3 Immunoproteasome deficiency has no influence on NF‐kB activation Two decades of immunoproteasome research solidified the hypothesis of immunoproteasome particles shaping the immunopeptideome presented on MHC class I molecules (Basler et al., 2011; Basler et al., 2006; Basler et al., 2004; Kincaid et al., 2012; Osterloh et al., 2006). 72

4. Discussion

Moreover, it became evident that the immunoproteasome has, apart from antigen processing, additional immunological functions. Immunoproteasome deficiency or inhibition affects T cells survival, expansion, and differentiation (Basler et al., 2004; Chen et al., 2001; Kalim et al., 2012; Moebius et al., 2010; Muchamuel et al., 2009; Zaiss et al., 2008), cytokine production (Basler et al., 2011; Basler et al., 2010; Basler et al., 2014; Muchamuel et al., 2009), and progression of autoimmune conditions (Basler et al., 2015). The impact of the immunoproteasome on various immunological aspects made this complex to an emerging pharmacolocigal target for various diseases and raised an even higher interest in understanding its exact cellular role.

The inherent difficulty of studying the function of immunoproteasome subunits is the fact that deficiency in one or several subunits can affect assembly of core particles (Bai et al., 2014; De et al., 2003; Joeris et al., 2012). In order to effectively mature into precursor particles, the chaperone‐like function of the propeptides of β‐subunits is required (Ramos et al., 1998; Schmidt et al., 1999). During full maturation of particles the propeptides are then autocatalytically cleaved, which finally activates the catalytic sites of these subunits (Groettrup et al., 2010). Griffin et al. could show that preparticles containing MECL‐1 and LMP2 need LMP7 for efficient maturation (Griffin et al., 1998). On the other hand, MECL‐1 depends on LMP2 for incorporation (De et al., 2003). Moreover, LMP7 exerts higher chaperone activity compared to β5 and hence, has a higher capacity to promote particle maturation (Heink et al., 2005; Joeris et al., 2012). The combination of subunits present in the particles of knockout cells could affect various aspects of proteasomal activity. Discrimination between artifacts potentially caused by “mixed” proteasomes and effects of immunoproteasome subunit deficiency is crucial yet difficult to put into practice. Under inflammatory conditions, cells devoid of LMP7 could furthermore have an overall decreased proteasomal capacity due to less efficient particle maturation. Decreased capacity and/or compensatory mechanisms to balance decreased proteasomal capacity could both influence the results obtained with knockout cells.

The last part of this study was focused on the activation of the transcription factor NF‐κB in MEFs and pMΦs derived from L7M‐/‐ and LMP2‐/‐. In order to define the proteasomal composition of these cells, a combined approach of proteasome immunoprecipitation and 2D‐ gel electrophoresis was tested for its applicability. The possibility to better assess the proportion of immunoproteasome expression relative to total proteasome content constitutes the clear advantage of this technique compared to western blot analysis. Immunorecipitation of proteasomes proved to be an easy way to rapidly enrich particles, especially if cells to be analyzed are limited in number. However, the visualization of proteins on 2D‐gels still has to 73

4. Discussion be optimized in order to detect all relevant subunits. A fluorescent dye applied to the enriched proteasomes before the electrophoretic separation could be an approach to be tested in the future. It would then be interesting to apply this technique for the analysis of proteasome composition of different mouse and human tissues.

Expression of different proteasome subunits was also determined by SDS‐PAGE and western blot. After stimulation with IFN‐γ, both pMΦs and MEFs expressed immunoproteasomes. Still, expression of constitutive subunits was not completely downregulated indicating a mix of constitutive and immunoproteasomes present in the cells. Due to the assembly defect in L7M‐ /‐ cells the particles present are almost exclusively composed of constitutive subunits. In contrast, LMP2‐/‐ cells seem to express LMP7 but due to the defect in MECL‐1 incorporation most likely contain constitutive proteasomes as well as mixed particles containing β1, β2, and LMP7. It would be interesting to further determine the proteasomal activity of the cells to potentially rule out differences in the overall degradative capacity between the knockout cells.

The literature about the role of the immunoproteasome in NF‐κB activation is controversial. An early report by Hayashi and Faustman suggested that LMP2 is required for the generation of p50 from the p105 precursor (Hayashi and Faustman, 1999). This finding could, however, not be reproduced by two other groups (Kessler et al., 2000; Runnels et al., 2000). Along the same line, Visekruna et al. proposed enhanced processing of p105 and degradation of IkBα in cytosolic extracts or purified proteasomes derived from mucosa of patients with Crohn’s disease or ulcerative colitis, which both differ in the proportion of immunoproteasome expressed (Visekruna et al., 2006). Unfortunatley, poor experimental setup and missing controls leave these data impossible to interpret. A follow‐up study by Hayashi and Faustman found reduced IkBα degradation in LMP2‐/‐ lymphocytes upon stimulation with TNF‐α (Hayashi and Faustman, 2000). This finding could be confirmed with LPS‐stimulated B cells derived from LMP2‐/‐ mice, although the effect observed was rather minor (Hensley et al., 2010). In contrast, a recent report by Maldonado and colleagues claimed that activation of the canonical pathway was not affected in LMP2‐/‐ and L7M‐/‐ cells while the alternative pathway seemed to be “aberrant” in LMP2‐/‐ cells (Maldonado et al., 2013). However, data presented in this study is inconsistent and rather an overinterpretation of minor effects.

The results presented here clearly argue against an influence of immunoproteasomes on the canonical pathway of NF‐κB activation. Two different types of primary cells prepared from LMP2‐/‐, L7M‐/‐, and wild type mice displayed no differences in the extent and kinetic of IκBα degradation when stimulated with LPS or TNF‐α. Consistent with this finding, the amount of 74

4. Discussion active NF‐κB dimers in the nucleus of knockout cells as well as the transactivation activity was normal. Although generation of p50 from the p105 precursor was not analyzed here, these results do clearly not indicate a deficit in NF‐κB subunits in the knockout cells. A study conducted with small molecule inhibitors targeting LMP2 or LMP7 further supports the model of NF‐κB activation being independent of proteasome composition. Inhibition of LMP2, LMP7, or even both had no influence on IκBα degradation in cells stimulated with TNF‐α (Jang et al., 2012).

Until now, the immunological impact of immunoproteasomes is still not fully understood. It has been suggested that the immunoproteasome has a higher capacity to clear ubiquitylated proteins accumulating after stimulation with IFN‐γ (Seifert et al., 2010). However, this finding could not be reproduced by others (Nathan et al., 2013). Taken together, more research is needed to improve the mechanistic understanding of immunoproteasomes in immune regulation.

References

Ackerman, A.L., A. Giodini, and P. Cresswell. 2006. A role for the endoplasmic reticulum protein retrotranslocation machinery during crosspresentation by dendritic cells. Immunity 25:607‐ 617. Aletrari, M.O., C. McKibbin, H. Williams, V. Pawar, P. Pietroni, J.M. Lord, S.L. Flitsch, R. Whitehead, E. Swanton, S. High, and R.A. Spooner. 2011. Eeyarestatin 1 interferes with both retrograde and anterograde intracellular trafficking pathways. PLoS One 6:e22713. Alexandru, G., J. Graumann, G.T. Smith, N.J. Kolawa, R. Fang, and R.J. Deshaies. 2008. UBXD7 binds multiple ubiquitin ligases and implicates p97 in HIF1alpha turnover. Cell 134:804‐816. Allen, M.D., A. Buchberger, and M. Bycroft. 2006. The PUB domain functions as a p97 binding module in human peptide N‐glycanase. J Biol Chem 281:25502‐25508. Amm, I., T. Sommer, and D.H. Wolf. 2014. Protein quality control and elimination of protein waste: the role of the ubiquitin‐proteasome system. Biochimica et biophysica acta 1843:182‐196. Anderson, D.J., R. Le Moigne, S. Djakovic, B. Kumar, J. Rice, S. Wong, J. Wang, B. Yao, E. Valle, S. Kiss von Soly, A. Madriaga, F. Soriano, M.K. Menon, Z.Y. Wu, M. Kampmann, Y. Chen, J.S. Weissman, B.T. Aftab, F.M. Yakes, L. Shawver, H.J. Zhou, D. Wustrow, and M. Rolfe. 2015. Targeting the AAA ATPase p97 as an Approach to Treat Cancer through Disruption of Protein Homeostasis. Cancer Cell 28:653‐665. Anton, L.C., and J.W. Yewdell. 2014. Translating DRiPs: MHC class I immunosurveillance of pathogens and tumors. Journal of leukocyte biology 95:551‐562. Apcher, S., C. Daskalogianni, and R. Fahraeus. 2015. Pioneer translation products as an alternative source for MHC‐I antigenic peptides. Molecular immunology Ardley, H.C., and P.A. Robinson. 2005. E3 ubiquitin ligases. Essays Biochem 41:15‐30. Arndt, V., C. Rogon, and J. Hohfeld. 2007. To be, or not to be‐‐molecular chaperones in protein degradation. Cell Mol Life Sci 64:2525‐2541. Auner, H.W., A.M. Moody, T.H. Ward, M. Kraus, E. Milan, P. May, A. Chaidos, C. Driessen, S. Cenci, F. Dazzi, A. Rahemtulla, J.F. Apperley, A. Karadimitris, and N. Dillon. 2013. Combined inhibition of p97 and the proteasome causes lethal disruption of the secretory apparatus in multiple myeloma cells. PLoS One 8:e74415. Bai, M., X. Zhao, K. Sahara, Y. Ohte, Y. Hirano, T. Kaneko, H. Yashiroda, and S. Murata. 2014. Assembly mechanisms of specialized core particles of the proteasome. Biomolecules 4:662‐ 677. Balch, W.E., R.I. Morimoto, A. Dillin, and J.W. Kelly. 2008. Adapting proteostasis for disease intervention. Science 319:916‐919. Baldwin, A.S., Jr. 1996. The NF‐kappa B and I kappa B proteins: new discoveries and insights. Annu Rev Immunol 14:649‐683. Ballar, P., and S. Fang. 2008. Regulation of ER‐associated degradation via p97/VCP‐interacting motif. Biochemical Society transactions 36:818‐822. Banerji, J., J. Sands, J.L. Strominger, and T. Spies. 1990. A gene pair from the human major histocompatibility complex encodes large proline‐rich proteins with multiple repeated motifs and a single ubiquitin‐like domain. Proc Natl Acad Sci U S A 87:2374‐2378. Barton, L.F., M. Cruz, R. Rangwala, G.S. Deepe, Jr., and J.J. Monaco. 2002. Regulation of immunoproteasome subunit expression in vivo following pathogenic fungal infection. J Immunol 169:3046‐3052. Basler, M., U. Beck, C.J. Kirk, and M. Groettrup. 2011. The antiviral immune response in mice devoid of immunoproteasome activity. J Immunol 187:5548‐5557. Basler, M., M. Dajee, C. Moll, M. Groettrup, and C.J. Kirk. 2010. Prevention of experimental colitis by a selective inhibitor of the immunoproteasome. J Immunol 185:634‐641.

References

Basler, M., C. Lauer, J. Moebius, R. Weber, M. Przybylski, A.F. Kisselev, C. Tsu, and M. Groettrup. 2012. Why the structure but not the activity of the immunoproteasome subunit low molecular mass polypeptide 2 rescues antigen presentation. J Immunol 189:1868‐1877. Basler, M., J. Moebius, L. Elenich, M. Groettrup, and J.J. Monaco. 2006. An altered T cell repertoire in MECL‐1‐deficient mice. J Immunol 176:6665‐6672. Basler, M., S. Mundt, A. Bitzer, C. Schmidt, and M. Groettrup. 2015. The immunoproteasome: a novel drug target for autoimmune diseases. Clin Exp Rheumatol Basler, M., S. Mundt, T. Muchamuel, C. Moll, J. Jiang, M. Groettrup, and C.J. Kirk. 2014. Inhibition of the immunoproteasome ameliorates experimental autoimmune encephalomyelitis. EMBO Mol Med 6:226‐238. Basler, M., N. Youhnovski, M. Van Den Broek, M. Przybylski, and M. Groettrup. 2004. Immunoproteasomes down‐regulate presentation of a subdominant T cell epitope from lymphocytic choriomeningitis virus. J Immunol 173:3925‐3934. Bassani‐Sternberg, M., S. Pletscher‐Frankild, L.J. Jensen, and M. Mann. 2015. Mass spectrometry of human leukocyte antigen class I peptidomes reveals strong effects of protein abundance and turnover on antigen presentation. Molecular & cellular proteomics : MCP 14:658‐673. Battegay, M., S. Cooper, A. Althage, J. Banziger, H. Hengartner, and R.M. Zinkernagel. 1991. Quantification of lymphocytic choriomeningitis virus with an immunological focus assay in 24‐ or 96‐well plates. Journal of virological methods 33:191‐198. Bays, N.W., R.G. Gardner, L.P. Seelig, C.A. Joazeiro, and R.Y. Hampton. 2001. Hrd1p/Der3p is a membrane‐anchored ubiquitin ligase required for ER‐associated degradation. Nat Cell Biol 3:24‐29. Bengtson, M.H., and C.A. Joazeiro. 2010. Role of a ribosome‐associated E3 ubiquitin ligase in protein quality control. Nature 467:470‐473. Berglund, P., D. Finzi, J.R. Bennink, and J.W. Yewdell. 2007. Viral alteration of cellular translational machinery increases defective ribosomal products. J Virol 81:7220‐7229. Berlin, C., D.J. Kowalewski, H. Schuster, N. Mirza, S. Walz, M. Handel, B. Schmid‐Horch, H.R. Salih, L. Kanz, H.G. Rammensee, S. Stevanovic, and J.S. Stickel. 2015. Mapping the HLA ligandome landscape of acute myeloid leukemia: a targeted approach toward peptide‐based immunotherapy. Leukemia 29:647‐659. Beskow, A., K.B. Grimberg, L.C. Bott, F.A. Salomons, N.P. Dantuma, and P. Young. 2009. A conserved unfoldase activity for the p97 AAA‐ATPase in proteasomal degradation. Journal of molecular biology 394:732‐746. Beuron, F., I. Dreveny, X. Yuan, V.E. Pye, C. McKeown, L.C. Briggs, M.J. Cliff, Y. Kaneko, R. Wallis, R.L. Isaacson, J.E. Ladbury, S.J. Matthews, H. Kondo, X. Zhang, and P.S. Freemont. 2006. Conformational changes in the AAA ATPase p97‐p47 adaptor complex. EMBO J 25:1967‐ 1976. Bijlmakers, M.J., and H.L. Ploegh. 1993. Putting together an MHC class I molecule. Current opinion in immunology 5:21‐26. Blach‐Olszewska, Z. 2005. Innate immunity: cells, receptors, and signaling pathways. Arch Immunol Ther Exp (Warsz) 53:245‐253. Borissenko, L., and M. Groll. 2007. 20S proteasome and its inhibitors: crystallographic knowledge for drug development. Chem Rev 107:687‐717. Borrow, P. 1997. Mechanisms of viral clearance and persistence. J Viral Hepat 4 Suppl 2:16‐24. Bourdetsky, D., C.E. Schmelzer, and A. Admon. 2014. The nature and extent of contributions by defective ribosome products to the HLA peptidome. Proc Natl Acad Sci U S A 111:E1591‐ 1599. Brandman, O., J. Stewart‐Ornstein, D. Wong, A. Larson, C.C. Williams, G.W. Li, S. Zhou, D. King, P.S. Shen, J. Weibezahn, J.G. Dunn, S. Rouskin, T. Inada, A. Frost, and J.S. Weissman. 2012. A ribosome‐bound quality control complex triggers degradation of nascent peptides and signals translation stress. Cell 151:1042‐1054. Brasier, A.R. 2006. The NF‐kappaB regulatory network. Cardiovasc Toxicol 6:111‐130.

References

Brehme, M., C. Voisine, T. Rolland, S. Wachi, J.H. Soper, Y. Zhu, K. Orton, A. Villella, D. Garza, M. Vidal, H. Ge, and R.I. Morimoto. 2014. A chaperome subnetwork safeguards proteostasis in aging and neurodegenerative disease. Cell Rep 9:1135‐1150. Briknarova, K., S. Takayama, L. Brive, M.L. Havert, D.A. Knee, J. Velasco, S. Homma, E. Cabezas, J. Stuart, D.W. Hoyt, A.C. Satterthwait, M. Llinas, J.C. Reed, and K.R. Ely. 2001. Structural analysis of BAG1 cochaperone and its interactions with Hsc70 heat shock protein. Nat Struct Biol 8:349‐352. Brown, K., S. Park, T. Kanno, G. Franzoso, and U. Siebenlist. 1993. Mutual regulation of the transcriptional activator NF‐kappa B and its inhibitor, I kappa B‐alpha. Proc Natl Acad Sci U S A 90:2532‐2536. Bruderer, R.M., C. Brasseur, and H.H. Meyer. 2004. The AAA ATPase p97/VCP interacts with its alternative co‐factors, Ufd1‐Npl4 and p47, through a common bipartite binding mechanism. J Biol Chem 279:49609‐49616. Brunger, A.T., and B. DeLaBarre. 2003. NSF and p97/VCP: similar at first, different at last. FEBS letters 555:126‐133. Buchberger, A. 2010. Control of ubiquitin conjugation by cdc48 and its cofactors. Subcell Biochem 54:17‐30. Buchberger, A. 2013. Roles of Cdc48 in regulated protein degradation in yeast. Subcell Biochem 66:195‐222. Buchberger, A., B. Bukau, and T. Sommer. 2010. Protein quality control in the cytosol and the endoplasmic reticulum: brothers in arms. Mol Cell 40:238‐252. Buchberger, A., M.J. Howard, M. Proctor, and M. Bycroft. 2001. The UBX domain: a widespread ubiquitin‐like module. Journal of molecular biology 307:17‐24. Buchberger, A., H. Schindelin, and P. Hanzelmann. 2015. Control of p97 function by cofactor binding. FEBS letters Bulik, S., B. Peters, and H.G. Holzhutter. 2005. Quantifying the contribution of defective ribosomal products to antigen production: a model‐based computational analysis. J Immunol 175:7957‐7964. Bullock, T.N., and L.C. Eisenlohr. 1996. Ribosomal scanning past the primary initiation codon as a mechanism for expression of CTL epitopes encoded in alternative reading frames. J Exp Med 184:1319‐1329. Cardinaud, S., S.R. Starck, P. Chandra, and N. Shastri. 2010. The synthesis of truncated polypeptides for immune surveillance and viral evasion. PLoS One 5:e8692. Caron, E., K. Vincent, M.H. Fortier, J.P. Laverdure, A. Bramoulle, M.P. Hardy, G. Voisin, P.P. Roux, S. Lemieux, P. Thibault, and C. Perreault. 2011. The MHC I immunopeptidome conveys to the cell surface an integrative view of cellular regulation. Molecular systems biology 7:533. Carvalho, P., V. Goder, and T.A. Rapoport. 2006. Distinct ubiquitin‐ligase complexes define convergent pathways for the degradation of ER proteins. Cell 126:361‐373. Cenci, S., L. Oliva, F. Cerruti, E. Milan, G. Bianchi, M. Raule, A. Mezghrani, E. Pasqualetto, R. Sitia, and P. Cascio. 2012. Pivotal Advance: Protein synthesis modulates responsiveness of differentiating and malignant plasma cells to proteasome inhibitors. Journal of leukocyte biology 92:921‐931. Cerundolo, V., A. Benham, V. Braud, S. Mukherjee, K. Gould, B. Macino, J. Neefjes, and A. Townsend. 1997. The proteasome‐specific inhibitor lactacystin blocks presentation of cytotoxic T lymphocyte epitopes in human and murine cells. European journal of immunology 27:336‐ 341. Chau, V., J.W. Tobias, A. Bachmair, D. Marriott, D.J. Ecker, D.K. Gonda, and A. Varshavsky. 1989. A multiubiquitin chain is confined to specific lysine in a targeted short‐lived protein. Science 243:1576‐1583. Chen, L.F., and W.C. Greene. 2004. Shaping the nuclear action of NF‐kappaB. Nat Rev Mol Cell Biol 5:392‐401.

References

Chen, W., C.C. Norbury, Y. Cho, J.W. Yewdell, and J.R. Bennink. 2001. Immunoproteasomes shape immunodominance hierarchies of antiviral CD8(+) T cells at the levels of T cell repertoire and presentation of viral antigens. J Exp Med 193:1319‐1326. Chou, T.F., S.J. Brown, D. Minond, B.E. Nordin, K. Li, A.C. Jones, P. Chase, P.R. Porubsky, B.M. Stoltz, F.J. Schoenen, M.P. Patricelli, P. Hodder, H. Rosen, and R.J. Deshaies. 2011. Reversible inhibitor of p97, DBeQ, impairs both ubiquitin‐dependent and autophagic protein clearance pathways. Proc Natl Acad Sci U S A 108:4834‐4839. Chou, T.F., S.L. Bulfer, C.C. Weihl, K. Li, L.G. Lis, M.A. Walters, F.J. Schoenen, H.J. Lin, R.J. Deshaies, and M.R. Arkin. 2014. Specific inhibition of p97/VCP ATPase and kinetic analysis demonstrate interaction between D1 and D2 ATPase domains. Journal of molecular biology 426:2886‐2899. Chou, T.F., K. Li, K.J. Frankowski, F.J. Schoenen, and R.J. Deshaies. 2013. Structure‐Activity Relationship Study Reveals ML240 and ML241 as Potent and Selective Inhibitors of p97 ATPase. ChemMedChem Christianson, J.C., J.A. Olzmann, T.A. Shaler, M.E. Sowa, E.J. Bennett, C.M. Richter, R.E. Tyler, E.J. Greenblatt, J.W. Harper, and R.R. Kopito. 2012. Defining human ERAD networks through an integrative mapping strategy. Nat Cell Biol 14:93‐105. Christianson, J.C., and Y. Ye. 2014. Cleaning up in the endoplasmic reticulum: ubiquitin in charge. Nature structural & molecular biology 21:325‐335. Claessen, J.H., and H.L. Ploegh. 2011. BAT3 guides misfolded glycoproteins out of the endoplasmic reticulum. PLoS One 6:e28542. Claessen, J.H., S. Sanyal, and H.L. Ploegh. 2014. The chaperone BAG6 captures dislocated glycoproteins in the cytosol. PLoS One 9:e90204. Colbert, J.D., D.J. Farfan‐Arribas, and K.L. Rock. 2013. Substrate‐induced protein stabilization reveals a predominant contribution from mature proteins to peptides presented on MHC class I. J Immunol 191:5410‐5419. Colon‐Ramos, D.A., P.M. Irusta, E.C. Gan, M.R. Olson, J. Song, R.I. Morimoto, R.M. Elliott, M. Lombard, R. Hollingsworth, J.M. Hardwick, G.K. Smith, and S. Kornbluth. 2003. Inhibition of translation and induction of apoptosis by Bunyaviral nonstructural proteins bearing sequence similarity to reaper. Mol Biol Cell 14:4162‐4172. Conaway, R.C., C.S. Brower, and J.W. Conaway. 2002. Emerging roles of ubiquitin in transcription regulation. Science 296:1254‐1258. Connell, P., C.A. Ballinger, J. Jiang, Y. Wu, L.J. Thompson, J. Hohfeld, and C. Patterson. 2001. The co‐ chaperone CHIP regulates protein triage decisions mediated by heat‐shock proteins. Nat Cell Biol 3:93‐96. Cox, J.H., P. Galardy, J.R. Bennink, and J.W. Yewdell. 1995. Presentation of endogenous and exogenous antigens is not affected by inactivation of E1 ubiquitin‐activating enzyme in temperature‐sensitive cell lines. J Immunol 154:511‐519. Croft, N.P., S.A. Smith, Y.C. Wong, C.T. Tan, N.L. Dudek, I.E. Flesch, L.C. Lin, D.C. Tscharke, and A.W. Purcell. 2013. Kinetics of antigen expression and epitope presentation during virus infection. Plos Pathog 9:e1003129. Cross, B.C., C. McKibbin, A.C. Callan, P. Roboti, M. Piacenti, C. Rabu, C.M. Wilson, R. Whitehead, S.L. Flitsch, M.R. Pool, S. High, and E. Swanton. 2009a. Eeyarestatin I inhibits Sec61‐mediated protein translocation at the endoplasmic reticulum. J Cell Sci 122:4393‐4400. Cross, B.C., I. Sinning, J. Luirink, and S. High. 2009b. Delivering proteins for export from the cytosol. Nat Rev Mol Cell Biol 10:255‐264. Dai, R.M., E. Chen, D.L. Longo, C.M. Gorbea, and C.C. Li. 1998. Involvement of valosin‐containing protein, an ATPase Co‐purified with IkappaBalpha and 26 S proteasome, in ubiquitin‐ proteasome‐mediated degradation of IkappaBalpha. J Biol Chem 273:3562‐3573. Dai, R.M., and C.C. Li. 2001. Valosin‐containing protein is a multi‐ubiquitin chain‐targeting factor required in ubiquitin‐proteasome degradation. Nat Cell Biol 3:740‐744.

References

Dalal, S., M.F. Rosser, D.M. Cyr, and P.I. Hanson. 2004. Distinct roles for the AAA ATPases NSF and p97 in the secretory pathway. Mol Biol Cell 15:637‐648. Davies, J.M., A.T. Brunger, and W.I. Weis. 2008. Improved structures of full‐length p97, an AAA ATPase: implications for mechanisms of nucleotide‐dependent conformational change. Structure 16:715‐726. Davies, J.M., H. Tsuruta, A.P. May, and W.I. Weis. 2005. Conformational changes of p97 during nucleotide hydrolysis determined by small‐angle X‐Ray scattering. Structure 13:183‐195. De, M., K. Jayarapu, L. Elenich, J.J. Monaco, R.A. Colbert, and T.A. Griffin. 2003. Beta 2 subunit propeptides influence cooperative proteasome assembly. J Biol Chem 278:6153‐6159. Defenouillere, Q., Y. Yao, J. Mouaikel, A. Namane, A. Galopier, L. Decourty, A. Doyen, C. Malabat, C. Saveanu, A. Jacquier, and M. Fromont‐Racine. 2013. Cdc48‐associated complex bound to 60S particles is required for the clearance of aberrant translation products. Proc Natl Acad Sci U S A 110:5046‐5051. DeLaBarre, B., and A.T. Brunger. 2005. Nucleotide dependent motion and mechanism of action of p97/VCP. Journal of molecular biology 347:437‐452. DeLaBarre, B., J.C. Christianson, R.R. Kopito, and A.T. Brunger. 2006. Central pore residues mediate the p97/VCP activity required for ERAD. Mol Cell 22:451‐462. Demachi‐Okamura, A., H. Torikai, Y. Akatsuka, H. Miyoshi, T. Yoshimori, and K. Kuzushima. 2012. Autophagy creates a CTL epitope that mimics tumor‐associated antigens. PLoS One 7:e47126. Desmots, F., H.R. Russell, Y. Lee, K. Boyd, and P.J. McKinnon. 2005. The reaper‐binding protein scythe modulates apoptosis and proliferation during mammalian development. Mol Cell Biol 25:10329‐10337. Dimitrova, L.N., K. Kuroha, T. Tatematsu, and T. Inada. 2009. Nascent peptide‐dependent translation arrest leads to Not4p‐mediated protein degradation by the proteasome. Journal of Biological Chemistry 284:10343‐10352. Doerks, T., R.R. Copley, J. Schultz, C.P. Ponting, and P. Bork. 2002. Systematic identification of novel protein domain families associated with nuclear functions. Genome Res 12:47‐56. Dolan, B.P., L. Li, C.A. Veltri, C.M. Ireland, J.R. Bennink, and J.W. Yewdell. 2011. Distinct pathways generate peptides from defective ribosomal products for CD8+ T cell immunosurveillance. J Immunol 186:2065‐2072. Doong, H., A. Vrailas, and E.C. Kohn. 2002. What's in the 'BAG'?‐‐A functional domain analysis of the BAG‐family proteins. Cancer Lett 188:25‐32. Dreveny, I., V.E. Pye, F. Beuron, L.C. Briggs, R.L. Isaacson, S.J. Matthews, C. McKeown, X. Yuan, X. Zhang, and P.S. Freemont. 2004. p97 and close encounters of every kind: a brief review. Biochemical Society transactions 32:715‐720. Driscoll, J., M.G. Brown, D. Finley, and J.J. Monaco. 1993. MHC‐linked LMP gene products specifically alter peptidase activities of the proteasome. Nature 365:262‐264. Dubiel, W., G. Pratt, K. Ferrell, and M. Rechsteiner. 1992. Purification of an 11 S regulator of the multicatalytic protease. J Biol Chem 267:22369‐22377. Duttler, S., S. Pechmann, and J. Frydman. 2013. Principles of cotranslational ubiquitination and quality control at the ribosome. Mol Cell 50:379‐393. Ehlinger, A., and K.J. Walters. 2013. Structural insights into proteasome activation by the 19S regulatory particle. Biochemistry 52:3618‐3628. El Hage, F., V. Stroobant, I. Vergnon, J.F. Baurain, H. Echchakir, V. Lazar, S. Chouaib, P.G. Coulie, and F. Mami‐Chouaib. 2008. Preprocalcitonin signal peptide generates a cytotoxic T lymphocyte‐defined tumor epitope processed by a proteasome‐independent pathway. Proc Natl Acad Sci U S A 105:10119‐10124. Eletr, Z.M., and K.D. Wilkinson. 2014. Regulation of proteolysis by human deubiquitinating enzymes. Biochimica et biophysica acta 1843:114‐128. English, L., M. Chemali, J. Duron, C. Rondeau, A. Laplante, D. Gingras, D. Alexander, D. Leib, C. Norbury, R. Lippe, and M. Desjardins. 2009. Autophagy enhances the presentation of

References

endogenous viral antigens on MHC class I molecules during HSV‐1 infection. Nat Immunol 10:480‐487. Erales, J., and P. Coffino. 2014. Ubiquitin‐independent proteasomal degradation. Biochimica et biophysica acta 1843:216‐221. Ernst, R., J.H. Claessen, B. Mueller, S. Sanyal, E. Spooner, A.G. van der Veen, O. Kirak, C.D. Schlieker, W.A. Weihofen, and H.L. Ploegh. 2011. Enzymatic blockade of the ubiquitin‐proteasome pathway. PLoS biology 8:e1000605. Ernst, R., B. Mueller, H.L. Ploegh, and C. Schlieker. 2009. The otubain YOD1 is a deubiquitinating enzyme that associates with p97 to facilitate protein dislocation from the ER. Mol Cell 36:28‐38. Fang, C.J., L. Gui, X. Zhang, D.R. Moen, K. Li, K.J. Frankowski, H.J. Lin, F.J. Schoenen, and T.F. Chou. 2015. Evaluating p97 inhibitor analogues for their domain selectivity and potency against the p97‐p47 complex. ChemMedChem 10:52‐56. Farfan‐Arribas, D.J., L.J. Stern, and K.L. Rock. 2012. Using intein catalysis to probe the origin of major histocompatibility complex class I‐presented peptides. Proc Natl Acad Sci U S A 109:16998‐ 17003. Favaloro, V., F. Vilardi, R. Schlecht, M.P. Mayer, and B. Dobberstein. 2010. Asna1/TRC40‐mediated membrane insertion of tail‐anchored proteins. J Cell Sci 123:1522‐1530. Fehling, H.J., W. Swat, C. Laplace, R. Kuhn, K. Rajewsky, U. Muller, and H. von Boehmer. 1994. MHC class I expression in mice lacking the proteasome subunit LMP‐7. Science 265:1234‐1237. Fiebiger, B.M., A. Moosmann, U. Behrends, and J. Mautner. 2012. Mature proteins derived from Epstein‐Barr virus fail to feed into the MHC class I antigenic pool. European journal of immunology 42:3167‐3173. Fiebiger, E., C. Hirsch, J.M. Vyas, E. Gordon, H.L. Ploegh, and D. Tortorella. 2004. Dissection of the dislocation pathway for type I membrane proteins with a new small molecule inhibitor, eeyarestatin. Mol Biol Cell 15:1635‐1646. Finley, D. 2009. Recognition and processing of ubiquitin‐protein conjugates by the proteasome. Annu Rev Biochem 78:477‐513. Finley, D., B. Bartel, and A. Varshavsky. 1989. The tails of ubiquitin precursors are ribosomal proteins whose fusion to ubiquitin facilitates ribosome biogenesis. Nature 338:394‐401. Finley, D., and V. Chau. 1991. Ubiquitination. Annu Rev Cell Biol 7:25‐69. Flierman, D., Y. Ye, M. Dai, V. Chau, and T.A. Rapoport. 2003. Polyubiquitin serves as a recognition signal, rather than a ratcheting molecule, during retrotranslocation of proteins across the endoplasmic reticulum membrane. J Biol Chem 278:34774‐34782. Fortier, M.H., E. Caron, M.P. Hardy, G. Voisin, S. Lemieux, C. Perreault, and P. Thibault. 2008. The MHC class I peptide repertoire is molded by the transcriptome. J Exp Med 205:595‐610. Franz, A., L. Ackermann, and T. Hoppe. 2014. Create and preserve: proteostasis in development and aging is governed by Cdc48/p97/VCP. Biochimica et biophysica acta 1843:205‐215. Fredrickson, E.K., J.C. Rosenbaum, M.N. Locke, T.I. Milac, and R.G. Gardner. 2011. Exposed hydrophobicity is a key determinant of nuclear quality control degradation. Mol Biol Cell 22:2384‐2395. Gaczynska, M., K.L. Rock, T. Spies, and A.L. Goldberg. 1994. Peptidase activities of proteasomes are differentially regulated by the major histocompatibility complex‐encoded genes for LMP2 and LMP7. Proc Natl Acad Sci U S A 91:9213‐9217. Gallagher, P.S., S.V. Clowes Candadai, and R.G. Gardner. 2014. The requirement for Cdc48/p97 in nuclear protein quality control degradation depends on the substrate and correlates with substrate insolubility. J Cell Sci 127:1980‐1991. Gallimore, A., H. Hengartner, and R. Zinkernagel. 1998. Hierarchies of antigen‐specific cytotoxic T‐ cell responses. Immunol Rev 164:29‐36. Garza, R.M., B.K. Sato, and R.Y. Hampton. 2009. In vitro analysis of Hrd1p‐mediated retrotranslocation of its multispanning membrane substrate 3‐hydroxy‐3‐methylglutaryl (HMG)‐CoA reductase. J Biol Chem 284:14710‐14722.

References

Ghislain, M., R.J. Dohmen, F. Levy, and A. Varshavsky. 1996. Cdc48p interacts with Ufd3p, a WD repeat protein required for ubiquitin‐mediated proteolysis in Saccharomyces cerevisiae. EMBO J 15:4884‐4899. Ghosh, S., M.J. May, and E.B. Kopp. 1998. NF‐kappa B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu Rev Immunol 16:225‐260. Gilmore, T.D. 2008. NF‐kB Transcription Factors. In. Glickman, M.H., D.M. Rubin, O. Coux, I. Wefes, G. Pfeifer, Z. Cjeka, W. Baumeister, V.A. Fried, and D. Finley. 1998. A subcomplex of the proteasome regulatory particle required for ubiquitin‐ conjugate degradation and related to the COP9‐signalosome and eIF3. Cell 94:615‐623. Godderz, D., C. Heinen, F.P. Marchese, T. Kurz, K. Acs, and N.P. Dantuma. 2015. Cdc48‐independent proteasomal degradation coincides with a reduced need for ubiquitylation. Scientific reports 5:7615. Goebel, S.J., G.P. Johnson, M.E. Perkus, S.W. Davis, J.P. Winslow, and E. Paoletti. 1990. The complete DNA sequence of vaccinia virus. Virology 179:247‐266, 517‐263. Goldberg, A.L. 2003. Protein degradation and protection against misfolded or damaged proteins. Nature 426:895‐899. Goodenough, E., T.M. Robinson, M.B. Zook, K.M. Flanigan, J.F. Atkins, M.T. Howard, and L.C. Eisenlohr. 2014. Cryptic MHC class I‐binding peptides are revealed by aminoglycoside‐ induced stop codon read‐through into the 3' UTR. Proc Natl Acad Sci U S A 111:5670‐5675. Granados, D.P., C.M. Laumont, P. Thibault, and C. Perreault. 2015. The nature of self for T cells‐a systems‐level perspective. Current opinion in immunology 34:1‐8. Granados, D.P., W. Yahyaoui, C.M. Laumont, T. Daouda, T.L. Muratore‐Schroeder, C. Cote, J.P. Laverdure, S. Lemieux, P. Thibault, and C. Perreault. 2012. MHC I‐associated peptides preferentially derive from transcripts bearing miRNA response elements. Blood Grice, G.L., I.T. Lobb, M.P. Weekes, S.P. Gygi, R. Antrobus, and J.A. Nathan. 2015. The Proteasome Distinguishes between Heterotypic and Homotypic Lysine‐11‐Linked Polyubiquitin Chains. Cell Rep Griffin, T.A., D. Nandi, M. Cruz, H.J. Fehling, L.V. Kaer, J.J. Monaco, and R.A. Colbert. 1998. Immunoproteasome assembly: cooperative incorporation of interferon gamma (IFN‐ gamma)‐inducible subunits. J Exp Med 187:97‐104. Groettrup, M., C.J. Kirk, and M. Basler. 2010. Proteasomes in immune cells: more than peptide producers? Nat Rev Immunol 10:73‐78. Groettrup, M., R. Kraft, S. Kostka, S. Standera, R. Stohwasser, and P.M. Kloetzel. 1996. A third interferon‐gamma‐induced subunit exchange in the 20S proteasome. European journal of immunology 26:863‐869. Grou, C.P., M.P. Pinto, A.V. Mendes, P. Domingues, and J.E. Azevedo. 2015. The de novo synthesis of ubiquitin: identification of deubiquitinases acting on ubiquitin precursors. Scientific reports 5:12836. Grove, D.E., M.F. Rosser, H.Y. Ren, A.P. Naren, and D.M. Cyr. 2009. Mechanisms for rescue of correctable folding defects in CFTRDelta F508. Mol Biol Cell 20:4059‐4069. Grover, A., and A.A. Izzo. 2012. BAT3 regulates Mycobacterium tuberculosis protein ESAT‐6‐ mediated apoptosis of macrophages. PLoS One 7:e40836. Gu, W., M. Cochrane, G.R. Leggatt, E. Payne, A. Choyce, F. Zhou, R. Tindle, and N.A. McMillan. 2009. Both treated and untreated tumors are eliminated by short hairpin RNA‐based induction of target‐specific immune responses. Proc Natl Acad Sci U S A 106:8314‐8319. Guillaume, B., J. Chapiro, V. Stroobant, D. Colau, B. Van Holle, G. Parvizi, M.P. Bousquet‐Dubouch, I. Theate, N. Parmentier, and B.J. Van den Eynde. 2010. Two abundant proteasome subtypes that uniquely process some antigens presented by HLA class I molecules. Proc Natl Acad Sci U S A 107:18599‐18604. Halawani, D., and M. Latterich. 2006. p97: The cell's molecular purgatory? Mol Cell 22:713‐717. Hampton, R.Y., and T. Sommer. 2012. Finding the will and the way of ERAD substrate retrotranslocation. Curr Opin Cell Biol 24:460‐466.

References

Harding, C.V., J. France, R. Song, J.M. Farah, S. Chatterjee, M. Iqbal, and R. Siman. 1995. Novel dipeptide aldehydes are proteasome inhibitors and block the MHC‐I antigen‐processing pathway. J Immunol 155:1767‐1775. Harhaj, E.W., and V.M. Dixit. 2012. Regulation of NF‐kappaB by deubiquitinases. Immunol Rev 246:107‐124. Hartl, F.U., and M. Hayer‐Hartl. 2002. Molecular chaperones in the cytosol: from nascent chain to folded protein. Science 295:1852‐1858. Hassan, C., M.G. Kester, A.H. de Ru, P. Hombrink, J.W. Drijfhout, H. Nijveen, J.A. Leunissen, M.H. Heemskerk, J.H. Falkenburg, and P.A. van Veelen. 2013. The human leukocyte antigen‐ presented ligandome of B lymphocytes. Molecular & cellular proteomics : MCP 12:1829‐ 1843. Hassink, G.C., B. Zhao, R. Sompallae, M. Altun, S. Gastaldello, N.V. Zinin, M.G. Masucci, and K. Lindsten. 2009. The ER‐resident ubiquitin‐specific protease 19 participates in the UPR and rescues ERAD substrates. EMBO Rep 10:755‐761. Hayashi, T., and D. Faustman. 1999. NOD mice are defective in proteasome production and activation of NF‐kappaB. Mol Cell Biol 19:8646‐8659. Hayashi, T., and D. Faustman. 2000. Essential role of human leukocyte antigen‐encoded proteasome subunits in NF‐kappaB activation and prevention of tumor necrosis factor‐ alpha‐induced apoptosis. J Biol Chem 275:5238‐5247. Hayden, M.S., and S. Ghosh. 2008. Shared principles in NF‐kappaB signaling. Cell 132:344‐362. Heink, S., D. Ludwig, P.M. Kloetzel, and E. Kruger. 2005. IFN‐gamma‐induced immune adaptation of the proteasome system is an accelerated and transient response. Proc Natl Acad Sci U S A 102:9241‐9246. Helenius, A., and M. Aebi. 2004. Roles of N‐linked glycans in the endoplasmic reticulum. Annu Rev Biochem 73:1019‐1049. Henderson, R.A., H. Michel, K. Sakaguchi, J. Shabanowitz, E. Appella, D.F. Hunt, and V.H. Engelhard. 1992. HLA‐A2.1‐associated peptides from a mutant cell line: a second pathway of antigen presentation. Science 255:1264‐1266. Hensley, S.E., D. Zanker, B.P. Dolan, A. David, H.D. Hickman, A.C. Embry, C.N. Skon, K.M. Grebe, T.A. Griffin, W. Chen, J.R. Bennink, and J.W. Yewdell. 2010. Unexpected role for the immunoproteasome subunit LMP2 in antiviral humoral and innate immune responses. J Immunol 184:4115‐4122. Heo, J.M., N. Livnat‐Levanon, E.B. Taylor, K.T. Jones, N. Dephoure, J. Ring, J. Xie, J.L. Brodsky, F. Madeo, S.P. Gygi, K. Ashrafi, M.H. Glickman, and J. Rutter. 2010. A stress‐responsive system for mitochondrial protein degradation. Mol Cell 40:465‐480. Hershko, A., and A. Ciechanover. 1998. The ubiquitin system. Annu Rev Biochem 67:425‐479. Hessa, T., A. Sharma, M. Mariappan, H.D. Eshleman, E. Gutierrez, and R.S. Hegde. 2011. Protein targeting and degradation are coupled for elimination of mislocalized proteins. Nature 475:394‐397. Hiller, M.M., A. Finger, M. Schweiger, and D.H. Wolf. 1996. ER degradation of a misfolded luminal protein by the cytosolic ubiquitin‐proteasome pathway. Science 273:1725‐1728. Hinz, M., S.C. Arslan, and C. Scheidereit. 2012. It takes two to tango: IkappaBs, the multifunctional partners of NF‐kappaB. Immunol Rev 246:59‐76. Hinz, M., and C. Scheidereit. 2014. The IkappaB kinase complex in NF‐kappaB regulation and beyond. EMBO Rep 15:46‐61. Hirsch, C., D. Blom, and H.L. Ploegh. 2003. A role for N‐glycanase in the cytosolic turnover of glycoproteins. EMBO J 22:1036‐1046. Hohfeld, J., D.M. Cyr, and C. Patterson. 2001. From the cradle to the grave: molecular chaperones that may choose between folding and degradation. EMBO Rep 2:885‐890. Hoppe, T. 2005. Multiubiquitylation by E4 enzymes: 'one size' doesn't fit all. Trends in biochemical sciences 30:183‐187.

References

Huang, C., G. Li, and W.J. Lennarz. 2012. Dynamic flexibility of the ATPase p97 is important for its interprotomer motion transmission. Proc Natl Acad Sci U S A 109:9792‐9797. Huang, L., J.M. Marvin, N. Tatsis, and L.C. Eisenlohr. 2011. Cutting Edge: Selective role of ubiquitin in MHC class I antigen presentation. J Immunol 186:1904‐1908. Huber, E.M., M. Basler, R. Schwab, W. Heinemeyer, C.J. Kirk, M. Groettrup, and M. Groll. 2012. Immuno‐ and constitutive proteasome crystal structures reveal differences in substrate and inhibitor specificity. Cell 148:727‐738. Hunt, D.F., R.A. Henderson, J. Shabanowitz, K. Sakaguchi, H. Michel, N. Sevilir, A.L. Cox, E. Appella, and V.H. Engelhard. 1992. Characterization of peptides bound to the class I MHC molecule HLA‐A2.1 by mass spectrometry. Science 255:1261‐1263. Ito‐Harashima, S., K. Kuroha, T. Tatematsu, and T. Inada. 2007. Translation of the poly(A) tail plays crucial roles in nonstop mRNA surveillance via translation repression and protein destabilization by proteasome in yeast. Genes & development 21:519‐524. Iwai, K. 2012. Diverse ubiquitin signaling in NF‐kappaB activation. Trends Cell Biol 22:355‐364. Iyer, L.M., E.V. Koonin, and L. Aravind. 2004. Novel predicted peptidases with a potential role in the ubiquitin signaling pathway. Cell Cycle 3:1440‐1450. Jackson, S.E. 2013. Hsp90: structure and function. Top Curr Chem 328:155‐240. Jang, E.R., N.R. Lee, S. Han, Y. Wu, L.K. Sharma, K.C. Carmony, J. Marks, D.M. Lee, J.O. Ban, M. Wehenkel, J.T. Hong, K.B. Kim, and W. Lee. 2012. Revisiting the role of the immunoproteasome in the activation of the canonical NF‐kappaB pathway. Mol Biosyst 8:2295‐2302. Jarosch, E., C. Taxis, C. Volkwein, J. Bordallo, D. Finley, D.H. Wolf, and T. Sommer. 2002. Protein dislocation from the ER requires polyubiquitination and the AAA‐ATPase Cdc48. Nat Cell Biol 4:134‐139. Jensen, P.E. 1988. Protein synthesis in antigen processing. J Immunol 141:2545‐2550. Jentsch, S., and S. Rumpf. 2007. Cdc48 (p97): a "molecular gearbox" in the ubiquitin pathway? Trends in biochemical sciences 32:6‐11. Jimbow, K., J.S. Park, F. Kato, K. Hirosaki, K. Toyofuku, C. Hua, and T. Yamashita. 2000. Assembly, target‐signaling and intracellular transport of tyrosinase gene family proteins in the initial stage of melanosome biogenesis. Pigment cell research / sponsored by the European Society for Pigment Cell Research and the International Pigment Cell Society 13:222‐229. Joeris, T., N. Schmidt, D. Ermert, P. Krienke, A. Visekruna, U. Kuckelkorn, S.H. Kaufmann, and U. Steinhoff. 2012. The proteasome system in infection: impact of beta5 and LMP7 on composition, maturation and quantity of active proteasome complexes. PLoS One 7:e39827. Ju, J.S., R.A. Fuentealba, S.E. Miller, E. Jackson, D. Piwnica‐Worms, R.H. Baloh, and C.C. Weihl. 2009. Valosin‐containing protein (VCP) is required for autophagy and is disrupted in VCP disease. J Cell Biol 187:875‐888. Ju, J.S., S.E. Miller, P.I. Hanson, and C.C. Weihl. 2008. Impaired protein aggregate handling and clearance underlie the pathogenesis of p97/VCP‐associated disease. J Biol Chem 283:30289‐30299. Kalim, K.W., M. Basler, C.J. Kirk, and M. Groettrup. 2012. Immunoproteasome subunit LMP7 deficiency and inhibition suppresses Th1 and Th17 but enhances regulatory T cell differentiation. J Immunol 189:4182‐4193. Kamper, N., S. Franken, S. Temme, S. Koch, T. Bieber, and N. Koch. 2012a. gamma‐Interferon‐ regulated chaperone governs human lymphocyte antigen class II expression. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 26:104‐116. Kamper, N., J. Kessler, S. Temme, C. Wegscheid, J. Winkler, and N. Koch. 2012b. A Novel BAT3 Sequence Generated by Alternative RNA Splicing of Exon 11B Displays Cell Type‐Specific Expression and Impacts on Subcellular Localization. PLoS One 7:e35972.

References

Kanarek, N., and Y. Ben‐Neriah. 2012. Regulation of NF‐kappaB by ubiquitination and degradation of the IkappaBs. Immunol Rev 246:77‐94. Kawahara, H., R. Minami, and N. Yokota. 2013. BAG6/BAT3: emerging roles in quality control for nascent polypeptides. Journal of biochemistry 153:147‐160. Kessler, B.M., A.M. Lennon‐Dumenil, M.L. Shinohara, M.A. Lipes, and H.L. Ploegh. 2000. LMP2 expression and proteasome activity in NOD mice. Nat Med 6:1064; author reply 1065‐1066. Khan, S., R. de Giuli, G. Schmidtke, M. Bruns, M. Buchmeier, M. van den Broek, and M. Groettrup. 2001a. Cutting edge: neosynthesis is required for the presentation of a T cell epitope from a long‐lived viral protein. J Immunol 167:4801‐4804. Khan, S., M. van den Broek, K. Schwarz, R. de Giuli, P.A. Diener, and M. Groettrup. 2001b. Immunoproteasomes largely replace constitutive proteasomes during an antiviral and antibacterial immune response in the liver. J Immunol 167:6859‐6868. Kim, S.J., J. Cho, E.J. Song, S.J. Kim, H.M. Kim, K.E. Lee, S.W. Suh, and E.E. Kim. 2014. Structural basis for ovarian tumor domain‐containing protein 1 (OTU1) binding to p97/valosin‐containing protein (VCP). J Biol Chem 289:12264‐12274. Kim, Y.E., M.S. Hipp, A. Bracher, M. Hayer‐Hartl, and F.U. Hartl. 2013. Molecular chaperone functions in protein folding and proteostasis. Annu Rev Biochem 82:323‐355. Kincaid, E.Z., J.W. Che, I. York, H. Escobar, E. Reyes‐Vargas, J.C. Delgado, R.M. Welsh, M.L. Karow, A.J. Murphy, D.M. Valenzuela, G.D. Yancopoulos, and K.L. Rock. 2012. Mice completely lacking immunoproteasomes show major changes in antigen presentation. Nat Immunol 13:129‐135. Kisselev, A.F., T.N. Akopian, K.M. Woo, and A.L. Goldberg. 1999. The sizes of peptides generated from protein by mammalian 26 and 20 S proteasomes. Implications for understanding the degradative mechanism and antigen presentation. J Biol Chem 274:3363‐3371. Kloppsteck, P., C.A. Ewens, A. Forster, X. Zhang, and P.S. Freemont. 2012. Regulation of p97 in the ubiquitin‐proteasome system by the UBX protein‐family. Biochimica et biophysica acta 1823:125‐129. Kobayashi, T., A. Manno, and A. Kakizuka. 2007. Involvement of valosin‐containing protein (VCP)/p97 in the formation and clearance of abnormal protein aggregates. Genes Cells 12:889‐901. Komander, D., and M. Rape. 2012. The ubiquitin code. Annu Rev Biochem 81:203‐229. Komarova, A.V., A.L. Haenni, and B.C. Ramirez. 2009. Virus versus host cell translation love and hate stories. Adv Virus Res 73:99‐170. Kondo, H., C. Rabouille, R. Newman, T.P. Levine, D. Pappin, P. Freemont, and G. Warren. 1997. p47 is a cofactor for p97‐mediated membrane fusion. Nature 388:75‐78. Kramer, G., D. Boehringer, N. Ban, and B. Bukau. 2009. The ribosome as a platform for co‐ translational processing, folding and targeting of newly synthesized proteins. Nature structural & molecular biology 16:589‐597. Kravtsova‐Ivantsiv, Y., and A. Ciechanover. 2012. Non‐canonical ubiquitin‐based signals for proteasomal degradation. J Cell Sci 125:539‐548. Kremer, M., A. Henn, C. Kolb, M. Basler, J. Moebius, B. Guillaume, M. Leist, B.J. Van den Eynde, and M. Groettrup. 2010. Reduced immunoproteasome formation and accumulation of immunoproteasomal precursors in the brains of lymphocytic choriomeningitis virus‐ infected mice. J Immunol 185:5549‐5560. Krick, R., S. Bremer, E. Welter, P. Schlotterhose, Y. Muehe, E.L. Eskelinen, and M. Thumm. 2010. Cdc48/p97 and Shp1/p47 regulate autophagosome biogenesis in concert with ubiquitin‐ like Atg8. J Cell Biol 190:965‐973. Kuwabara, N., R. Minami, N. Yokota, H. Matsumoto, T. Senda, H. Kawahara, and R. Kato. 2015. Structure of a BAG6 (Bcl‐2‐associated athanogene 6)‐Ubl4a (Ubiquitin‐like protein 4a) Complex Reveals a Novel Binding Interface that Functions in Tail‐Anchored Protein Biogenesis. J Biol Chem

References

Lacsina, J.R., O.A. Marks, X. Liu, D.W. Reid, S. Jagannathan, and C.V. Nicchitta. 2012. Premature translational termination products are rapidly degraded substrates for MHC class I presentation. PLoS One 7:e51968. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680‐685. Latterich, M., K.U. Frohlich, and R. Schekman. 1995. Membrane fusion and the cell cycle: Cdc48p participates in the fusion of ER membranes. Cell 82:885‐893. Lazaro, S., D. Gamarra, and M. Del Val. 2015. Proteolytic enzymes involved in MHC class I antigen processing: A guerrilla army that partners with the proteasome. Molecular immunology Le Page, C., P. Genin, M.G. Baines, and J. Hiscott. 2000. Interferon activation and innate immunity. Rev Immunogenet 2:374‐386. Lee, J.G., and Y. Ye. 2013. Bag6/Bat3/Scythe: a novel chaperone activity with diverse regulatory functions in protein biogenesis and degradation. BioEssays : news and reviews in molecular, cellular and developmental biology 35:377‐385. Lev, A., M.F. Princiotta, D. Zanker, K. Takeda, J.S. Gibbs, C. Kumagai, E. Waffarn, B.P. Dolan, A. Burgevin, P. Van Endert, W. Chen, J.R. Bennink, and J.W. Yewdell. 2010. Compartmentalized MHC class I antigen processing enhances immunosurveillance by circumventing the law of mass action. Proc Natl Acad Sci U S A 107:6964‐6969. Levine, C.G., D. Mitra, A. Sharma, C.L. Smith, and R.S. Hegde. 2005. The efficiency of protein compartmentalization into the secretory pathway. Mol Biol Cell 16:279‐291. Leznicki, P., A. Clancy, B. Schwappach, and S. High. 2010. Bat3 promotes the membrane integration of tail‐anchored proteins. J Cell Sci 123:2170‐2178. Leznicki, P., Q.P. Roebuck, L. Wunderley, A. Clancy, E.M. Krysztofinska, R.L. Isaacson, J. Warwicker, B. Schwappach, and S. High. 2013. The association of BAG6 with SGTA and tail‐anchored proteins. PLoS One 8:e59590. Li, G., C. Huang, G. Zhao, and W.J. Lennarz. 2012. Interprotomer motion‐transmission mechanism for the hexameric AAA ATPase p97. Proc Natl Acad Sci U S A 109:3737‐3741. Li, G., G. Zhao, H. Schindelin, and W.J. Lennarz. 2008. Tyrosine phosphorylation of ATPase p97 regulates its activity during ERAD. Biochem Biophys Res Commun 375:247‐251. Liang, J., C. Yin, H. Doong, S. Fang, C. Peterhoff, R.A. Nixon, and M.J. Monteiro. 2006. Characterization of erasin (UBXD2): a new ER protein that promotes ER‐associated protein degradation. J Cell Sci 119:4011‐4024. Lim, P.J., R. Danner, J. Liang, H. Doong, C. Harman, D. Srinivasan, C. Rothenberg, H. Wang, Y. Ye, S. Fang, and M.J. Monteiro. 2009. Ubiquilin and p97/VCP bind erasin, forming a complex involved in ERAD. J Cell Biol 187:201‐217. Lindquist, S.L., and J.W. Kelly. 2011. Chemical and biological approaches for adapting proteostasis to ameliorate protein misfolding and aggregation diseases: progress and prognosis. Cold Spring Harb Perspect Biol 3: Liu, Y., N. Soetandyo, J.G. Lee, L. Liu, Y. Xu, W.M. Clemons, Jr., and Y. Ye. 2014. USP13 antagonizes gp78 to maintain functionality of a chaperone in ER‐associated degradation. Elife 3:e01369. Locke, M., J.I. Toth, and M.D. Petroski. 2014. Lys11‐ and Lys48‐linked ubiquitin chains interact with p97 during endoplasmic‐reticulum‐associated degradation. The Biochemical journal 459:205‐216. Luckheeram, R.V., R. Zhou, A.D. Verma, and B. Xia. 2012. CD4(+)T cells: differentiation and functions. Clin Dev Immunol 2012:925135. Ma, C.P., C.A. Slaughter, and G.N. DeMartino. 1992. Identification, purification, and characterization of a protein activator (PA28) of the 20 S proteasome (macropain). J Biol Chem 267:10515‐10523. Mackett, M., and G.L. Smith. 1986. Vaccinia virus expression vectors. The Journal of general virology 67 ( Pt 10):2067‐2082. Madden, D.R. 1995. The three‐dimensional structure of peptide‐MHC complexes. Annu Rev Immunol 13:587‐622.

References

Magnaghi, P., R. D'Alessio, B. Valsasina, N. Avanzi, S. Rizzi, D. Asa, F. Gasparri, L. Cozzi, U. Cucchi, C. Orrenius, P. Polucci, D. Ballinari, C. Perrera, A. Leone, G. Cervi, E. Casale, Y. Xiao, C. Wong, D.J. Anderson, A. Galvani, D. Donati, T. O'Brien, P.K. Jackson, and A. Isacchi. 2013. Covalent and allosteric inhibitors of the ATPase VCP/p97 induce cancer cell death. Nat Chem Biol Malarkannan, S., T. Horng, P.P. Shih, S. Schwab, and N. Shastri. 1999. Presentation of out‐of‐frame peptide/MHC class I complexes by a novel translation initiation mechanism. Immunity 10:681‐690. Maldonado, M., R.J. Kapphahn, M.R. Terluk, N.D. Heuss, C. Yuan, D.S. Gregerson, and D.A. Ferrington. 2013. Immunoproteasome deficiency modifies the alternative pathway of NFkappaB signaling. PLoS One 8:e56187. Manchen, S.T., and A.V. Hubberstey. 2001. Human Scythe contains a functional nuclear localization sequence and remains in the nucleus during staurosporine‐induced apoptosis. Biochem Biophys Res Commun 287:1075‐1082. Mariappan, M., X. Li, S. Stefanovic, A. Sharma, A. Mateja, R.J. Keenan, and R.S. Hegde. 2010. A ribosome‐associating factor chaperones tail‐anchored membrane proteins. Nature 466:1120‐1124. McCarthy, M.K., and J.B. Weinberg. 2015. The immunoproteasome and viral infection: a complex regulator of inflammation. Front Microbiol 6:21. Meerang, M., D. Ritz, S. Paliwal, Z. Garajova, M. Bosshard, N. Mailand, P. Janscak, U. Hubscher, H. Meyer, and K. Ramadan. 2011. The ubiquitin‐selective segregase VCP/p97 orchestrates the response to DNA double‐strand breaks. Nat Cell Biol 13:1376‐1382. Mehnert, M., T. Sommer, and E. Jarosch. 2010. ERAD ubiquitin ligases: multifunctional tools for protein quality control and waste disposal in the endoplasmic reticulum. BioEssays : news and reviews in molecular, cellular and developmental biology 32:905‐913. Metzger, M.B., V.A. Hristova, and A.M. Weissman. 2012. HECT and RING finger families of E3 ubiquitin ligases at a glance. J Cell Sci 125:531‐537. Meyer, B.J., J.C. de la Torre, and P.J. Southern. 2002a. Arenaviruses: genomic RNAs, transcription, and replication. Curr Top Microbiol Immunol 262:139‐157. Meyer, H., M. Bug, and S. Bremer. 2012. Emerging functions of the VCP/p97 AAA‐ATPase in the ubiquitin system. Nat Cell Biol 14:117‐123. Meyer, H., and C.C. Weihl. 2014. The VCP/p97 system at a glance: connecting cellular function to disease pathogenesis. J Cell Sci 127:3877‐3883. Meyer, H.H., J.G. Shorter, J. Seemann, D. Pappin, and G. Warren. 2000. A complex of mammalian ufd1 and npl4 links the AAA‐ATPase, p97, to ubiquitin and nuclear transport pathways. EMBO J 19:2181‐2192. Meyer, H.H., Y. Wang, and G. Warren. 2002b. Direct binding of ubiquitin conjugates by the mammalian p97 adaptor complexes, p47 and Ufd1‐Npl4. EMBO J 21:5645‐5652. Milner, E., E. Barnea, I. Beer, and A. Admon. 2006. The turnover kinetics of major histocompatibility complex peptides of human cancer cells. Molecular & cellular proteomics : MCP 5:357‐365. Minami, R., A. Hayakawa, H. Kagawa, Y. Yanagi, H. Yokosawa, and H. Kawahara. 2010. BAG‐6 is essential for selective elimination of defective proteasomal substrates. J Cell Biol 190:637‐ 650. Minami, R., M. Shimada, H. Yokosawa, and H. Kawahara. 2007. Scythe regulates apoptosis through modulating ubiquitin‐mediated proteolysis of the Xenopus elongation factor XEF1AO. The Biochemical journal 405:495‐501. Mizushima, N. 2007. Autophagy: process and function. Genes & development 21:2861‐2873. Mock, J.Y., J.W. Chartron, M. Zaslaver, Y. Xu, Y. Ye, and W.M. Clemons, Jr. 2015. Bag6 complex contains a minimal tail‐anchor‐targeting module and a mock BAG domain. Proc Natl Acad Sci U S A 112:106‐111. Moebius, J., M. van den Broek, M. Groettrup, and M. Basler. 2010. Immunoproteasomes are essential for survival and expansion of T cells in virus‐infected mice. European journal of immunology 40:3439‐3449.

References

Moir, D., and D. Botstein. 1982. Determination of the order of gene function in the yeast nuclear division pathway using cs and ts mutants. Genetics 100:565‐577. Momburg, F., and G.J. Hammerling. 1998. Generation and TAP‐mediated transport of peptides for major histocompatibility complex class I molecules. Advances in immunology 68:191‐256. Morito, D., K. Hirao, Y. Oda, N. Hosokawa, F. Tokunaga, D.M. Cyr, K. Tanaka, K. Iwai, and K. Nagata. 2008. Gp78 cooperates with RMA1 in endoplasmic reticulum‐associated degradation of CFTRDeltaF508. Mol Biol Cell 19:1328‐1336. Morreale, G., L. Conforti, J. Coadwell, A.L. Wilbrey, and M.P. Coleman. 2009. Evolutionary divergence of valosin‐containing protein/cell division cycle protein 48 binding interactions among endoplasmic reticulum‐associated degradation proteins. The FEBS journal 276:1208‐1220. Moss, B. 1996. Genetically engineered poxviruses for recombinant gene expression, vaccination, and safety. Proc Natl Acad Sci U S A 93:11341‐11348. Moss, B. 2006. Poxviridae: the viruses and their replication. In Fields Virology. D.M. Kinpe, editor Lippicott Williams & Wilkins, Philadelphia. 2905‐2946. Moss, B. 2011. Smallpox vaccines: targets of protective immunity. Immunol Rev 239:8‐26. Muchamuel, T., M. Basler, M.A. Aujay, E. Suzuki, K.W. Kalim, C. Lauer, C. Sylvain, E.R. Ring, J. Shields, J. Jiang, P. Shwonek, F. Parlati, S.D. Demo, M.K. Bennett, C.J. Kirk, and M. Groettrup. 2009. A selective inhibitor of the immunoproteasome subunit LMP7 blocks cytokine production and attenuates progression of experimental arthritis. Nat Med 15:781‐787. Mueller, B., E.J. Klemm, E. Spooner, J.H. Claessen, and H.L. Ploegh. 2008. SEL1L nucleates a protein complex required for dislocation of misfolded glycoproteins. Proc Natl Acad Sci U S A 105:12325‐12330. Murata, S., T. Chiba, and K. Tanaka. 2003. CHIP: a quality‐control E3 ligase collaborating with molecular chaperones. Int J Biochem Cell Biol 35:572‐578. Murata, S., Y. Minami, M. Minami, T. Chiba, and K. Tanaka. 2001. CHIP is a chaperone‐dependent E3 ligase that ubiquitylates unfolded protein. EMBO Rep 2:1133‐1138. Nathan, J.A., V. Spinnenhirn, G. Schmidtke, M. Basler, M. Groettrup, and A.L. Goldberg. 2013. Immuno‐ and constitutive proteasomes do not differ in their abilities to degrade ubiquitinated proteins. Cell 152:1184‐1194. Neefjes, J.J., F. Momburg, and G.J. Hammerling. 1993. Selective and ATP‐dependent translocation of peptides by the MHC‐encoded transporter. Science 261:769‐771. Newberg, M.H., D.H. Smith, S.B. Haertel, D.R. Vining, E. Lacy, and V.H. Engelhard. 1996. Importance of MHC class 1 alpha2 and alpha3 domains in the recognition of self and non‐self MHC molecules. J Immunol 156:2473‐2480. Nielsen, S.V., E.G. Poulsen, C.A. Rebula, and R. Hartmann‐Petersen. 2014. Protein quality control in the nucleus. Biomolecules 4:646‐661. Ogura, T., and A.J. Wilkinson. 2001. AAA+ superfamily ATPases: common structure‐‐diverse function. Genes Cells 6:575‐597. Oliveira, C.C., and T. van Hall. 2015. Alternative Antigen Processing for MHC Class I: Multiple Roads Lead to Rome. Front Immunol 6:298. Osterloh, P., K. Linkemann, S. Tenzer, H.G. Rammensee, M.P. Radsak, D.H. Busch, and H. Schild. 2006. Proteasomes shape the repertoire of T cells participating in antigen‐specific immune responses. Proc Natl Acad Sci U S A 103:5042‐5047. Ozaki, T., E. Hanaoka, M. Naka, A. Nakagawara, and S. Sakiyama. 1999. Cloning and characterization of rat BAT3 cDNA. DNA Cell Biol 18:503‐512. Pahl, H.L. 1999. Activators and target genes of Rel/NF‐kappaB transcription factors. Oncogene 18:6853‐6866. Palmer, A.L., and B.P. Dolan. 2013. MHC class I antigen presentation of DRiP‐derived peptides from a model antigen is not dependent on the AAA ATPase p97. PLoS One 8:e67796.

References

Palombella, V.J., O.J. Rando, A.L. Goldberg, and T. Maniatis. 1994. The ubiquitin‐proteasome pathway is required for processing the NF‐kappa B1 precursor protein and the activation of NF‐kappa B. Cell 78:773‐785. Pasparakis, M. 2009. Regulation of tissue homeostasis by NF‐kappaB signalling: implications for inflammatory diseases. Nat Rev Immunol 9:778‐788. Payapilly, A., and S. High. 2014. BAG6 regulates the quality control of a polytopic ERAD substrate. J Cell Sci Peters, J.M., M.J. Walsh, and W.W. Franke. 1990. An abundant and ubiquitous homo‐oligomeric ring‐shaped ATPase particle related to the putative vesicle fusion proteins Sec18p and NSF. EMBO J 9:1757‐1767. Peters, P.J., J. Borst, V. Oorschot, M. Fukuda, O. Krahenbuhl, J. Tschopp, J.W. Slot, and H.J. Geuze. 1991. Cytotoxic T lymphocyte granules are secretory lysosomes, containing both perforin and granzymes. J Exp Med 173:1099‐1109. Petrescu, S.M., N. Branza‐Nichita, G. Negroiu, A.J. Petrescu, and R.A. Dwek. 2000. Tyrosinase and glycoprotein folding: roles of chaperones that recognize glycans. Biochemistry 39:5229‐ 5237. Pickart, C.M. 2000. Ubiquitin in chains. Trends in biochemical sciences 25:544‐548. Pilon, M., R. Schekman, and K. Romisch. 1997. Sec61p mediates export of a misfolded secretory protein from the endoplasmic reticulum to the cytosol for degradation. EMBO J 16:4540‐ 4548. Platanias, L.C. 2005. Mechanisms of type‐I‐ and type‐II‐interferon‐mediated signalling. Nat Rev Immunol 5:375‐386. Plytycz, B., and R. Seljelid. 1998. MHC molecules and lymphocytes: evolutionary perspective. Arch Immunol Ther Exp (Warsz) 46:137‐142. Powers, E.T., R.I. Morimoto, A. Dillin, J.W. Kelly, and W.E. Balch. 2009. Biological and chemical approaches to diseases of proteostasis deficiency. Annu Rev Biochem 78:959‐991. Praveen, P.V., R. Yaneva, H. Kalbacher, and S. Springer. 2010. Tapasin edits peptides on MHC class I molecules by accelerating peptide exchange. European journal of immunology 40:214‐ 224. Princiotta, M.F., D. Finzi, S.B. Qian, J. Gibbs, S. Schuchmann, F. Buttgereit, J.R. Bennink, and J.W. Yewdell. 2003. Quantitating protein synthesis, degradation, and endogenous antigen processing. Immunity 18:343‐354. Probst, H.C., K. Tschannen, A. Gallimore, M. Martinic, M. Basler, T. Dumrese, E. Jones, and M.F. van den Broek. 2003. Immunodominance of an antiviral cytotoxic T cell response is shaped by the kinetics of viral protein expression. Journal of Immunology 171:5415‐5422. Qian, S.B., M.F. Princiotta, J.R. Bennink, and J.W. Yewdell. 2006. Characterization of rapidly degraded polypeptides in mammalian cells reveals a novel layer of nascent protein quality control. J Biol Chem 281:392‐400. Rabinovich, E., A. Kerem, K.U. Frohlich, N. Diamant, and S. Bar‐Nun. 2002. AAA‐ATPase p97/Cdc48p, a cytosolic chaperone required for endoplasmic reticulum‐associated protein degradation. Mol Cell Biol 22:626‐634. Rabu, C., V. Schmid, B. Schwappach, and S. High. 2009. Biogenesis of tail‐anchored proteins: the beginning for the end? J Cell Sci 122:3605‐3612. Ramos, P.C., J. Hockendorff, E.S. Johnson, A. Varshavsky, and R.J. Dohmen. 1998. Ump1p is required for proper maturation of the 20S proteasome and becomes its substrate upon completion of the assembly. Cell 92:489‐499. Rangachari, M., C. Zhu, K. Sakuishi, S. Xiao, J. Karman, A. Chen, M. Angin, A. Wakeham, E.A. Greenfield, R.A. Sobel, H. Okada, P.J. McKinnon, T.W. Mak, M.M. Addo, A.C. Anderson, and V.K. Kuchroo. 2012. Bat3 promotes T cell responses and autoimmunity by repressing Tim‐ 3‐mediated cell death and exhaustion. Nat Med

References

Rape, M., T. Hoppe, I. Gorr, M. Kalocay, H. Richly, and S. Jentsch. 2001. Mobilization of processed, membrane‐tethered SPT23 transcription factor by CDC48(UFD1/NPL4), a ubiquitin‐ selective chaperone. Cell 107:667‐677. Reed, S.I. 2003. Ratchets and clocks: the cell cycle, ubiquitylation and protein turnover. Nat Rev Mol Cell Biol 4:855‐864. Reits, E.A., J.C. Vos, M. Gromme, and J. Neefjes. 2000. The major substrates for TAP in vivo are derived from newly synthesized proteins. Nature 404:774‐778. Richly, H., M. Rape, S. Braun, S. Rumpf, C. Hoege, and S. Jentsch. 2005. A series of ubiquitin binding factors connects CDC48/p97 to substrate multiubiquitylation and proteasomal targeting. Cell 120:73‐84. Riviere, Y., R. Ahmed, P.J. Southern, M.J. Buchmeier, F.J. Dutko, and M.B. Oldstone. 1985. The S RNA segment of lymphocytic choriomeningitis virus codes for the nucleoprotein and glycoproteins 1 and 2. J Virol 53:966‐968. Rock, K.L., D.J. Farfan‐Arribas, J.D. Colbert, and A.L. Goldberg. 2014. Re‐examining class‐I presentation and the DRiP hypothesis. Trends in immunology 35:144‐152. Rock, K.L., and A.L. Goldberg. 1999. Degradation of cell proteins and the generation of MHC class I‐ presented peptides. Annu Rev Immunol 17:739‐779. Rock, K.L., C. Gramm, L. Rothstein, K. Clark, R. Stein, L. Dick, D. Hwang, and A.L. Goldberg. 1994. Inhibitors of the proteasome block the degradation of most cell proteins and the generation of peptides presented on MHC class I molecules. Cell 78:761‐771. Rodrigo‐Brenni, M.C., E. Gutierrez, and R.S. Hegde. 2014. Cytosolic Quality Control of Mislocalized Proteins Requires RNF126 Recruitment to Bag6. Mol Cell Rodrigo‐Brenni, M.C., and R.S. Hegde. 2012. Design principles of protein biosynthesis‐coupled quality control. Developmental cell 23:896‐907. Rosenbaum, J.C., E.K. Fredrickson, M.L. Oeser, C.M. Garrett‐Engele, M.N. Locke, L.A. Richardson, Z.W. Nelson, E.D. Hetrick, T.I. Milac, D.E. Gottschling, and R.G. Gardner. 2011. Disorder targets misorder in nuclear quality control degradation: a disordered ubiquitin ligase directly recognizes its misfolded substrates. Mol Cell 41:93‐106. Rudiger, S., A. Buchberger, and B. Bukau. 1997. Interaction of Hsp70 chaperones with substrates. Nat Struct Biol 4:342‐349. Runnels, H.A., W.A. Watkins, and J.J. Monaco. 2000. LMP2 expression and proteasome activity in NOD mice. Nat Med 6:1064‐1065; author reply 1065‐1066. Salvato, M.S., and E.M. Shimomaye. 1989. The completed sequence of lymphocytic choriomeningitis virus reveals a unique RNA structure and a gene for a zinc finger protein. Virology 173:1‐10. Saric, T., S.C. Chang, A. Hattori, I.A. York, S. Markant, K.L. Rock, M. Tsujimoto, and A.L. Goldberg. 2002. An IFN‐gamma‐induced aminopeptidase in the ER, ERAP1, trims precursors to MHC class I‐presented peptides. Nat Immunol 3:1169‐1176. Sato, S., C.L. Ward, and R.R. Kopito. 1998. Cotranslational ubiquitination of cystic fibrosis transmembrane conductance regulator in vitro. J Biol Chem 273:7189‐7192. Sauer, R.T., and T.A. Baker. 2011. AAA+ proteases: ATP‐fueled machines of protein destruction. Annu Rev Biochem 80:587‐612. Schlesinger, D.H., and G. Goldstein. 1975. Molecular conservation of 74 amino acid sequence of ubiquitin between cattle and man. Nature 255:423‐424. Schmidt, M., D. Zantopf, R. Kraft, S. Kostka, R. Preissner, and P.M. Kloetzel. 1999. Sequence information within proteasomal prosequences mediates efficient integration of beta‐ subunits into the 20 S proteasome complex. Journal of molecular biology 288:117‐128. Schmidtke, G., A. Aichem, and M. Groettrup. 2014. FAT10ylation as a signal for proteasomal degradation. Biochimica et biophysica acta 1843:97‐102. Schmidtke, G., S. Emch, M. Groettrup, and H.G. Holzhutter. 2000. Evidence for the existence of a non‐catalytic modifier site of peptide hydrolysis by the 20 S proteasome. J Biol Chem 275:22056‐22063.

References

Schubert, U., L.C. Anton, J. Gibbs, C.C. Norbury, J.W. Yewdell, and J.R. Bennink. 2000. Rapid degradation of a large fraction of newly synthesized proteins by proteasomes. Nature 404:770‐774. Schuberth, C., and A. Buchberger. 2008. UBX domain proteins: major regulators of the AAA ATPase Cdc48/p97. Cell Mol Life Sci 65:2360‐2371. Schuberth, C., H. Richly, S. Rumpf, and A. Buchberger. 2004. Shp1 and Ubx2 are adaptors of Cdc48 involved in ubiquitin‐dependent protein degradation. EMBO Rep 5:818‐824. Schwab, S.R., K.C. Li, C. Kang, and N. Shastri. 2003. Constitutive display of cryptic translation products by MHC class I molecules. Science 301:1367‐1371. Schwarz, K., M. van Den Broek, S. Kostka, R. Kraft, A. Soza, G. Schmidtke, P.M. Kloetzel, and M. Groettrup. 2000. Overexpression of the proteasome subunits LMP2, LMP7, and MECL‐1, but not PA28 alpha/beta, enhances the presentation of an immunodominant lymphocytic choriomeningitis virus T cell epitope. J Immunol 165:768‐778. Seifert, U., L.P. Bialy, F. Ebstein, D. Bech‐Otschir, A. Voigt, F. Schroter, T. Prozorovski, N. Lange, J. Steffen, M. Rieger, U. Kuckelkorn, O. Aktas, P.M. Kloetzel, and E. Kruger. 2010. Immunoproteasomes preserve protein homeostasis upon interferon‐induced oxidative stress. Cell 142:613‐624. Sen, G.C., and P. Lengyel. 1992. The interferon system. A bird's eye view of its biochemistry. J Biol Chem 267:5017‐5020. Serwold, T., F. Gonzalez, J. Kim, R. Jacob, and N. Shastri. 2002. ERAAP customizes peptides for MHC class I molecules in the endoplasmic reticulum. Nature 419:480‐483. Sha, Z., and A.L. Goldberg. 2014. Proteasome‐Mediated Processing of Nrf1 Is Essential for Coordinate Induction of All Proteasome Subunits and p97. Current biology : CB Shaffer, A.L., M. Shapiro‐Shelef, N.N. Iwakoshi, A.H. Lee, S.B. Qian, H. Zhao, X. Yu, L. Yang, B.K. Tan, A. Rosenwald, E.M. Hurt, E. Petroulakis, N. Sonenberg, J.W. Yewdell, K. Calame, L.H. Glimcher, and L.M. Staudt. 2004. XBP1, downstream of Blimp‐1, expands the secretory apparatus and other organelles, and increases protein synthesis in plasma cell differentiation. Immunity 21:81‐93. Shao, S., K. von der Malsburg, and R.S. Hegde. 2013. Listerin‐dependent nascent protein ubiquitination relies on ribosome subunit dissociation. Mol Cell 50:637‐648. Shcherbik, N., and D.S. Haines. 2007. Cdc48p(Npl4p/Ufd1p) binds and segregates membrane‐ anchored/tethered complexes via a polyubiquitin signal present on the anchors. Mol Cell 25:385‐397. Shen, X.Z., S. Billet, C. Lin, D. Okwan‐Duodu, X. Chen, A.E. Lukacher, and K.E. Bernstein. 2011. The carboxypeptidase ACE shapes the MHC class I peptide repertoire. Nat Immunol 12:1078‐ 1085. Shin, E.C., U. Seifert, T. Kato, C.M. Rice, S.M. Feinstone, P.M. Kloetzel, and B. Rehermann. 2006. Virus‐induced type I IFN stimulates generation of immunoproteasomes at the site of infection. J Clin Invest 116:3006‐3014. Shoemaker, C.J., and R. Green. 2012. Translation drives mRNA quality control. Nature structural & molecular biology 19:594‐601. Sijts, E.J., and P.M. Kloetzel. 2011. The role of the proteasome in the generation of MHC class I ligands and immune responses. Cell Mol Life Sci 68:1491‐1502. Skipper, J.C., R.C. Hendrickson, P.H. Gulden, V. Brichard, A. Van Pel, Y. Chen, J. Shabanowitz, T. Wolfel, C.L. Slingluff, Jr., T. Boon, D.F. Hunt, and V.H. Engelhard. 1996. An HLA‐A2‐restricted tyrosinase antigen on melanoma cells results from posttranslational modification and suggests a novel pathway for processing of membrane proteins. J Exp Med 183:527‐534. Sondermann, H., C. Scheufler, C. Schneider, J. Hohfeld, F.U. Hartl, and I. Moarefi. 2001. Structure of a Bag/Hsc70 complex: convergent functional evolution of Hsp70 nucleotide exchange factors. Science 291:1553‐1557.

References

Song, C., Q. Wang, and C.C. Li. 2003. ATPase activity of p97‐valosin‐containing protein (VCP). D2 mediates the major enzyme activity, and D1 contributes to the heat‐induced activity. J Biol Chem 278:3648‐3655. Song, C., Q. Wang, C. Song, and T.J. Rogers. 2015. Valosin‐containing protein (VCP/p97) is capable of unfolding polyubiquitinated proteins through its ATPase domains. Biochem Biophys Res Commun 463:453‐457. Sourdive, D.J., K. Murali‐Krishna, J.D. Altman, A.J. Zajac, J.K. Whitmire, C. Pannetier, P. Kourilsky, B. Evavold, A. Sette, and R. Ahmed. 1998. Conserved T cell receptor repertoire in primary and memory CD8 T cell responses to an acute viral infection. J Exp Med 188:71‐82. Sowa, M.E., E.J. Bennett, S.P. Gygi, and J.W. Harper. 2009. Defining the human deubiquitinating enzyme interaction landscape. Cell 138:389‐403. Squier, M.K., and J.J. Cohen. 1994. Cell‐mediated cytotoxic mechanisms. Current opinion in immunology 6:447‐452. Stadtmueller, B.M., and C.P. Hill. 2011. Proteasome activators. Mol Cell 41:8‐19. Stirling, P.C., V.F. Lundin, and M.R. Leroux. 2003. Getting a grip on non‐native proteins. EMBO Rep 4:565‐570. Stolz, A., and D.H. Wolf. 2010. Endoplasmic reticulum associated protein degradation: a chaperone assisted journey to hell. Biochimica et biophysica acta 1803:694‐705. Strehl, B., U. Seifert, E. Kruger, S. Heink, U. Kuckelkorn, and P.M. Kloetzel. 2005. Interferon‐gamma, the functional plasticity of the ubiquitin‐proteasome system, and MHC class I antigen processing. Immunol Rev 207:19‐30. Sun, S.C., P.A. Ganchi, D.W. Ballard, and W.C. Greene. 1993. NF‐kappa B controls expression of inhibitor I kappa B alpha: evidence for an inducible autoregulatory pathway. Science 259:1912‐1915. Suzuki, T., H. Park, E.A. Till, and W.J. Lennarz. 2001. The PUB domain: a putative protein‐protein interaction domain implicated in the ubiquitin‐proteasome pathway. Biochem Biophys Res Commun 287:1083‐1087. Swain, S.L. 1983. T cell subsets and the recognition of MHC class. Immunol Rev 74:129‐142. Swanson, R., M. Locher, and M. Hochstrasser. 2001. A conserved ubiquitin ligase of the nuclear envelope/endoplasmic reticulum that functions in both ER‐associated and Matalpha2 repressor degradation. Genes & development 15:2660‐2674. Takayama, S., and J.C. Reed. 2001. Molecular chaperone targeting and regulation by BAG family proteins. Nat Cell Biol 3:E237‐241. Tanaka, K. 1998. Molecular biology of the proteasome. Biochem Biophys Res Commun 247:537‐541. Taylor, E.B., and J. Rutter. 2011. Mitochondrial quality control by the ubiquitin‐proteasome system. Biochemical Society transactions 39:1509‐1513. Tey, S.K., and R. Khanna. 2012. Autophagy mediates transporter associated with antigen processing‐independent presentation of viral epitopes through MHC class I pathway. Blood 120:994‐1004. Thoms, S. 2002. Cdc48 can distinguish between native and non‐native proteins in the absence of cofactors. FEBS letters 520:107‐110. Thress, K., E.K. Evans, and S. Kornbluth. 1999. Reaper‐induced dissociation of a Scythe‐sequestered cytochrome c‐releasing activity. EMBO J 18:5486‐5493. Thress, K., W. Henzel, W. Shillinglaw, and S. Kornbluth. 1998. Scythe: a novel reaper‐binding apoptotic regulator. EMBO J 17:6135‐6143. Thress, K., J. Song, R.I. Morimoto, and S. Kornbluth. 2001. Reversible inhibition of Hsp70 chaperone function by Scythe and Reaper. EMBO J 20:1033‐1041. Toes, R.E., A.K. Nussbaum, S. Degermann, M. Schirle, N.P. Emmerich, M. Kraft, C. Laplace, A. Zwinderman, T.P. Dick, J. Muller, B. Schonfisch, C. Schmid, H.J. Fehling, S. Stevanovic, H.G. Rammensee, and H. Schild. 2001. Discrete cleavage motifs of constitutive and immunoproteasomes revealed by quantitative analysis of cleavage products. J Exp Med 194:1‐12.

References

Tonegawa, S. 1983. Somatic generation of antibody diversity. Nature 302:575‐581. Townsend, A.R., F.M. Gotch, and J. Davey. 1985. Cytotoxic T cells recognize fragments of the influenza nucleoprotein. Cell 42:457‐467. Tresse, E., F.A. Salomons, J. Vesa, L.C. Bott, V. Kimonis, T.P. Yao, N.P. Dantuma, and J.P. Taylor. 2010. VCP/p97 is essential for maturation of ubiquitin‐containing autophagosomes and this function is impaired by mutations that cause IBMPFD. Autophagy 6:217‐227. Unno, M., T. Mizushima, Y. Morimoto, Y. Tomisugi, K. Tanaka, N. Yasuoka, and T. Tsukihara. 2002. The structure of the mammalian 20S proteasome at 2.75 A resolution. Structure 10:609‐ 618. Vabulas, R.M., and F.U. Hartl. 2005. Protein synthesis upon acute nutrient restriction relies on proteasome function. Science 310:1960‐1963. Vallabhapurapu, S., and M. Karin. 2009. Regulation and function of NF‐kappaB transcription factors in the immune system. Annu Rev Immunol 27:693‐733. van der Most, R.G., K. Murali‐Krishna, J.L. Whitton, C. Oseroff, J. Alexander, S. Southwood, J. Sidney, R.W. Chesnut, A. Sette, and R. Ahmed. 1998. Identification of Db‐ and Kb‐restricted subdominant cytotoxic T‐cell responses in lymphocytic choriomeningitis virus‐infected mice. Virology 240:158‐167. van der Most, R.G., A. Sette, C. Oseroff, J. Alexander, K. Murali‐Krishna, L.L. Lau, S. Southwood, J. Sidney, R.W. Chesnut, M. Matloubian, and R. Ahmed. 1996. Analysis of cytotoxic T cell responses to dominant and subdominant epitopes during acute and chronic lymphocytic choriomeningitis virus infection. J Immunol 157:5543‐5554. Van Kaer, L., P.G. Ashton‐Rickardt, M. Eichelberger, M. Gaczynska, K. Nagashima, K.L. Rock, A.L. Goldberg, P.C. Doherty, and S. Tonegawa. 1994. Altered peptidase and viral‐specific T cell response in LMP2 mutant mice. Immunity 1:533‐541. Vashist, S., and D.T. Ng. 2004. Misfolded proteins are sorted by a sequential checkpoint mechanism of ER quality control. J Cell Biol 165:41‐52. Verma, R., R.S. Oania, N.J. Kolawa, and R.J. Deshaies. 2013. Cdc48/p97 promotes degradation of aberrant nascent polypeptides bound to the ribosome. elife 2:e00308. Vigneron, N., and B.J. Van den Eynde. 2014. Proteasome subtypes and regulators in the processing of antigenic peptides presented by class I molecules of the major histocompatibility complex. Biomolecules 4:994‐1025. Visekruna, A., T. Joeris, D. Seidel, A. Kroesen, C. Loddenkemper, M. Zeitz, S.H. Kaufmann, R. Schmidt‐Ullrich, and U. Steinhoff. 2006. Proteasome‐mediated degradation of IkappaBalpha and processing of p105 in Crohn disease and ulcerative colitis. J Clin Invest 116:3195‐3203. Vucic, D., V.M. Dixit, and I.E. Wertz. 2011. Ubiquitylation in apoptosis: a post‐translational modification at the edge of life and death. Nat Rev Mol Cell Biol 12:439‐452. Vyas, J.M., A.G. Van der Veen, and H.L. Ploegh. 2008. The known unknowns of antigen processing and presentation. Nat Rev Immunol 8:607‐618. Wang, F., L.A. Durfee, and J.M. Huibregtse. 2013. A cotranslational ubiquitination pathway for quality control of misfolded proteins. Mol Cell 50:368‐378. Wang, Q., L. Li, and Y. Ye. 2006. Regulation of retrotranslocation by p97‐associated deubiquitinating enzyme ataxin‐3. J Cell Biol 174:963‐971. Wang, Q., L. Li, and Y. Ye. 2008. Inhibition of p97‐dependent protein degradation by Eeyarestatin I. J Biol Chem 283:7445‐7454. Wang, Q., Y. Liu, N. Soetandyo, K. Baek, R. Hegde, and Y. Ye. 2011. A ubiquitin ligase‐associated chaperone holdase maintains polypeptides in soluble states for proteasome degradation. Mol Cell 42:758‐770. Wang, Q., H. Mora‐Jensen, M.A. Weniger, P. Perez‐Galan, C. Wolford, T. Hai, D. Ron, W. Chen, W. Trenkle, A. Wiestner, and Y. Ye. 2009. ERAD inhibitors integrate ER stress with an epigenetic mechanism to activate BH3‐only protein NOXA in cancer cells. Proc Natl Acad Sci U S A 106:2200‐2205.

References

Wang, Q., B.A. Shinkre, J.G. Lee, M.A. Weniger, Y. Liu, W. Chen, A. Wiestner, W.C. Trenkle, and Y. Ye. 2010. The ERAD inhibitor Eeyarestatin I is a bifunctional compound with a membrane‐ binding domain and a p97/VCP inhibitory group. PLoS One 5:e15479. Wang, Q., C. Song, and C.C. Li. 2003. Hexamerization of p97‐VCP is promoted by ATP binding to the D1 domain and required for ATPase and biological activities. Biochem Biophys Res Commun 300:253‐260. Wang, R., and C.C. Liew. 1994. The human BAT3 ortholog in rodents is predominantly and developmentally expressed in testis. Mol Cell Biochem 136:49‐57. Watts, G.D., J. Wymer, M.J. Kovach, S.G. Mehta, S. Mumm, D. Darvish, A. Pestronk, M.P. Whyte, and V.E. Kimonis. 2004. Inclusion body myopathy associated with Paget disease of bone and frontotemporal dementia is caused by mutant valosin‐containing protein. Nature genetics 36:377‐381. Wearsch, P.A., and P. Cresswell. 2008. The quality control of MHC class I peptide loading. Curr Opin Cell Biol 20:624‐631. Weihl, C.C., S. Dalal, A. Pestronk, and P.I. Hanson. 2006. Inclusion body myopathy‐associated mutations in p97/VCP impair endoplasmic reticulum‐associated degradation. Hum Mol Genet 15:189‐199. Weimershaus, M., I. Evnouchidou, L. Saveanu, and P. van Endert. 2013. Peptidases trimming MHC class I ligands. Current opinion in immunology 25:90‐96. Wheatley, D.N., M.R. Giddings, and M.S. Inglis. 1980. Kinetics of degradation of "short‐" and "long‐ lived" proteins in cultured mammalian cells. Cell Biol Int Rep 4:1081‐1090. Wiborg, O., M.S. Pedersen, A. Wind, L.E. Berglund, K.A. Marcker, and J. Vuust. 1985. The human ubiquitin multigene family: some genes contain multiple directly repeated ubiquitin coding sequences. EMBO J 4:755‐759. Wickner, S., M.R. Maurizi, and S. Gottesman. 1999. Posttranslational quality control: folding, refolding, and degrading proteins. Science 286:1888‐1893. Wolfel, C., I. Drexler, A. Van Pel, T. Thres, N. Leister, W. Herr, G. Sutter, C. Huber, and T. Wolfel. 2000. Transporter (TAP)‐ and proteasome‐independent presentation of a melanoma‐ associated tyrosinase epitope. Int J Cancer 88:432‐438. Wu, Y.H., S.F. Shih, and J.Y. Lin. 2004. Ricin triggers apoptotic morphological changes through caspase‐3 cleavage of BAT3. J Biol Chem 279:19264‐19275. Xu, Y., Y. Liu, J.G. Lee, and Y. Ye. 2013. A ubiquitin‐like domain recruits an oligomeric chaperone to a retrotranslocation complex in endoplasmic reticulum‐associated degradation. J Biol Chem 288:18068‐18076. Ye, Y. 2006. Diverse functions with a common regulator: ubiquitin takes command of an AAA ATPase. Journal of structural biology 156:29‐40. Ye, Y., H.H. Meyer, and T.A. Rapoport. 2001. The AAA ATPase Cdc48/p97 and its partners transport proteins from the ER into the cytosol. Nature 414:652‐656. Ye, Y., H.H. Meyer, and T.A. Rapoport. 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. J Cell Biol 162:71‐84. Yeung, H.O., A. Forster, C. Bebeacua, H. Niwa, C. Ewens, C. McKeown, X. Zhang, and P.S. Freemont. 2014. Inter‐ring rotations of AAA ATPase p97 revealed by electron cryomicroscopy. Open Biol 4:130142. Yeung, H.O., P. Kloppsteck, H. Niwa, R.L. Isaacson, S. Matthews, X. Zhang, and P.S. Freemont. 2008. Insights into adaptor binding to the AAA protein p97. Biochemical Society transactions 36:62‐67. Yewdell, J.W., L.C. Anton, and J.R. Bennink. 1996. Defective ribosomal products (DRiPs): a major source of antigenic peptides for MHC class I molecules? J Immunol 157:1823‐1826. Yewdell, J.W., and J.R. Bennink. 1992. Cell biology of antigen processing and presentation to major histocompatibility complex class I molecule‐restricted T lymphocytes. Advances in immunology 52:1‐123.

References

Yewdell, J.W., and C.V. Nicchitta. 2006. The DRiP hypothesis decennial: support, controversy, refinement and extension. Trends in immunology 27:368‐373. York, I.A., A.L. Goldberg, X.Y. Mo, and K.L. Rock. 1999. Proteolysis and class I major histocompatibility complex antigen presentation. Immunol Rev 172:49‐66. Younger, J.M., L. Chen, H.Y. Ren, M.F. Rosser, E.L. Turnbull, C.Y. Fan, C. Patterson, and D.M. Cyr. 2006. Sequential quality‐control checkpoints triage misfolded cystic fibrosis transmembrane conductance regulator. Cell 126:571‐582. Younger, J.M., H.Y. Ren, L. Chen, C.Y. Fan, A. Fields, C. Patterson, and D.M. Cyr. 2004. A foldable CFTR{Delta}F508 biogenic intermediate accumulates upon inhibition of the Hsc70‐CHIP E3 ubiquitin ligase. J Cell Biol 167:1075‐1085. Zaiss, D.M., N. de Graaf, and A.J. Sijts. 2008. The proteasome immunosubunit multicatalytic endopeptidase complex‐like 1 is a T‐cell‐intrinsic factor influencing homeostatic expansion. Infect Immun 76:1207‐1213. Zanker, D., J. Waithman, J.W. Yewdell, and W. Chen. 2013. Mixed proteasomes function to increase viral peptide diversity and broaden antiviral CD8+ T cell responses. J Immunol 191:52‐59. Zehn, D., C.J. Cohen, Y. Reiter, and P. Walden. 2004. Extended presentation of specific MHC‐peptide complexes by mature dendritic cells compared to other types of antigen‐presenting cells. European journal of immunology 34:1551‐1560. Zhang, X., L. Gui, X. Zhang, S.L. Bulfer, V. Sanghez, D.E. Wong, Y. Lee, L. Lehmann, J.S. Lee, P.Y. Shih, H.J. Lin, M. Iacovino, C.C. Weihl, M.R. Arkin, Y. Wang, and T.F. Chou. 2015. Altered cofactor regulation with disease‐associated p97/VCP mutations. Proc Natl Acad Sci U S A 112:E1705‐ 1714. Zhao, G., X. Zhou, L. Wang, G. Li, H. Schindelin, and W.J. Lennarz. 2007. Studies on peptide:N‐ glycanase‐p97 interaction suggest that p97 phosphorylation modulates endoplasmic reticulum‐associated degradation. Proc Natl Acad Sci U S A 104:8785‐8790. Zimmermann, R., L. Muller, and B. Wullich. 2006. Protein transport into the endoplasmic reticulum: mechanisms and pathologies. Trends Mol Med 12:567‐573. Zinkernagel, R.M., and P.C. Doherty. 1974. Restriction of in vitro T cell‐mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature 248:701‐702.

Appendix

Abbreviations AAA ATPases associated with divers cellular activities aa Amino acid ALS Amyotrophic lateral sclerosis ANOVA Analysis of variance APC Allophycocyanin APS Ammonium persulfate ATP Adenosine triphosphate

β2m β2‐microglobulin BAG BCL‐2 associated athanogene BAT HLA‐B‐associated transcript bp Base pair BSA Bovine serum albumin °C Degree Celsius CD Cluster of differentiation Cdc Cell division cycle CHIP Carboxyl terminus of HSC70 interacting protein

CO2 Carbon dioxide CTL Cytotoxic T lymphocyte C‐terminus Carboxy‐terminus CUE Coupling of ubiquitin to ER degradation d Day dATP Deoxyadenosine triphosphate DBeQ N2,N4‐debenzylquinazoline‐2,‐4‐diamine DMEM Dulbecco’s Modified Eagle Medium DMP Dimethyl pimelimidate DMSO Dimethyl sulfoxide DNA Deoxyribonucleic acid Doa Degradation of alpha 2 DRiP Defective ribosomal product 96

Appendix

DTE Dithioerythritol DTT Dithiotreitol DUP Deubiquitylase E1 Ubiquitin‐activating enzyme E2 Ubiquitin‐conjugating enzyme E3 Ubiquitin ligase E4 Ubiquitin chain assembly factor E.coli Escherichia coli EDTA Ethylenediaminetetraacetic acid EerI Eeyarestatin I eGFP Enhanced GFP EGTA Ethylene glycol‐bis(2‐aminoethylether)‐N,N,N′,N′‐tetraacetic acid ELISA Enzyme‐linked immunosorbent assay EMSA Electrophoretic mobility shift assay ER Endoplasmic reticulum ERAAP ER aminopeptidase ERAD (‐C, ‐L, ‐M) ER‐associated degradation (cytosolic, luminal, membrane) ERAP ER aminopeptidase ERp ER protein E/S Effector to stimulator FAT HLA‐F adjacent protein FCS Fetal calf serum FITC Fluorescein isothiocyanate g Gravity GFP Green fluorescent protein gp Glycoprotein GP Glycoprotein GRR Glycine‐rich region Gy Gray h Hour H‐2 Histocompatibility‐2 97

Appendix

HECT Homologous to E6AP carboxyl terminuS HEPES 4‐(2‐hydroxyethyl)‐1‐piperazineethanesulfonic acid hHR23 Human homolog of RAD23 HLA Human leukocyte antigen

H3PO4 Phosphoric acid HPRT Hypoxanthin guanine phosphoribosyl transferase Hrd, HRD HMG‐CoA reductase degradation HRP Horseradish peroxidase HSC Heat‐shock cognate HSP Heat shock protein IBMPFD Inclusion body myopathy, Paget's disease of the bone, frontotemporal dementia and amyotrophic lateral sclerosis

IC50 Half maximal inhibitory concentration IFN Interferon IκB Inhibitor of kB IKK Inhibitor of kB kinase IL Interleukin IMDM Iscove’s Modified Dulbecco’s Medium i.p. Intraperitoneal IP Immunoprecipitation i.v. Intravenous kb Kilobase KCl Potassium chloride kDa Kilo Dalton L Long LB Luria broth LCMV Lymphocytic choriomeningitis virus LMP Low molecular mass polypeptide LPS Lipopolysaccharid LTN1 Listerin LZ Leucine zipper M Molar 98

Appendix

MACS Magnetic‐activated cell sorting MECL Multicatalytic endopeptidase complex‐like

MgCl2 Magnesium chloride ml Milliliter mM Millimolar MOI Multiplicity of infection mRNA Messenger RNA NaCl Sodium chloride NEPHGE Nonequilibrium pH gel electrophoresis NEMF Nuclear export mediator factor NEMO Nuclear factor‐kB essential modulator NF‐κB Nuclear factor‐kB MEF Mouse embryonic fibroblast MEM Minimum Essential Medium MHC Major histocompatibility min Minute NLS Nuclear localization sequence NP Nucleoprotein NPL Nuclear protein localization homolog NRF Nuclear respiratory factor N‐terminus Amino‐termiuns NZF NPL4 zinc finger OTU Ovarian tumor OVA Ovalbumin p Protein PA Proteasome activator PAGE Polyacrylamide gel electrophoresis PAMP pathogen‐associated molecular pattern PBS Phosphate buffered saline PCR Polymerase chain reaction PFA Paraformaldehyde PFU Plaque forming unit 99

Appendix

Pi Inorganic phosphate PI Propidium Iodide pMΦs Peritoneal macrophages poly(dI‐dC) Poly(deoxyinosinic‐deoxycytidylic) acid PPi Inorganic diphosphate PSMB Proteasome subunit beta PUB PNGase/UBA or UBX PUL PLAP, Ufd3p, and Lub1p Rad Radiation absorbed dose RDP Rapidly degraded protein RHD Rel homology domain RING Really interesting new gene RIPA Radioimmunoprecipitation assay buffer RNA Ribonucleic acid RNF Ring finger protein RPL Ribosomal protein large subunit rpm Revolutions per minute RPMI Rosewell Park Memorial Institute RT‐PCR Reverse transcription polymerase chain reaction RQC Ribosome quality control complex S Short SCF SKP1‐Cullin1‐F‐box SDS Sodium dodecylsulfate Sec Second SHP SH2 containing phosphatase siRNA Small interfering RNA TA Tail‐anchored TAD Transactivation domain TAE Tris‐acetate‐EDTA TAP Transporter‐associated with antigen processing TBST Tris buffered saline with Tween 20 TCR T cell receptor 100

Appendix

TEMED N,N,N’,N’‐Tetramethylethylenediamine TLR Toll‐like receptor TMD Transmembrane domain TNF Tumor necrosis factor TNFR Tumor necrosis factor receptor TRC Transmembrane domain recognition complex TRCP Transducin repeat containing protein Tyr Tyrosinase U Unit Ub Ubiquitin UBA Ubiquitin‐associated UBL Ubiquitin‐like UBX Ubiquitin‐regulatory X UBXD UBX domain protein UBX‐L UBX‐like UDF Ubiquitin‐fusion degradation UFD ubiquitin fusion degradation UIM Ubiquitin‐interacting motif μg Milligram μm Micrometer μM Micromolar μl Microliter UPS Ubiquitin proteasome system UTY Ubiquitously transcribed tetratricopeptide repeat containing, Y‐ linked UV Ultraviolet VBM VCP‐binding motif VCP Valsoin‐containing protein VIM VCP‐interacting motif V Volt Vh Volt hour VV Vaccinia virus 101

Appendix v/v Volume per volume rVV recombinant vaccinia virus WR Western Reserve w/v Weight per volume

Abbreviations of amino acids Alanine Ala A Arginine Arg R Asparagine Asn N Aspartic acid Asp D Cysteine Cys C Glutamine Gln Q Glutamic acid Glu E 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

102

Appendix

Acknowledgments I would like to express my special appreciation and thanks to my advisor Professor Dr. Marcus Gröttrup for giving me the opportunity to pursue my PhD thesis in his group. Thank you for your endless enthusiasm, motivation, and support and for the freedom you gave me to grow as a scientist.

A very special thank you goes to my supervisor Dr. Michael Basler who was a tremendous support and inspiration to me. It was always great to share ideas, jokes, and chocolate.

I would like to thank Professor Dr. Alexander Bürkle and Professor Dr. Maries van den Broek for taking the time to read my thesis and for my defence, the research training group 1331 for funding of travel expenses, courses, and retreats, and Dr. Daniel Krappmann and Dr. Michelle Vincendau for teaching me the secrets of EMSAs.

I would also like to thank the whole group of immunology for the great time I had on P11, for all the help, occasional roof top coffee brakes, funny parties, trips to New York, and great conversations.

Most of all, I would like to express my deepest gratitude to my family for the wonderful friendship we have, for their love, support, and interest.

Finally, I want to thank Chrischti for his support and love!

103