Over a decade, proteasome inhibitors (PIs), bortezomib, carfilzomib (Cfz) and ixazomib, have contributed to a significant improvement in the overall survival for multiple myeloma (MM) patients. However, the response rate of PI was fairly low, leaving a huge gap in MM patient care. Given this, mechanistic understanding of PI resistance is crucial towards developing new therapeutic strategies for refractory/relapsed MM patients. In this dissertation work, we found H727 human bronchial carcinoid cells are inherently resistant to Cfz, yet susceptible to other PIs and inhibitors targeting upstream components of the ubiquitin-proteasome system (UPS). It indicated H727 cells may serve as a cell line model for de novo Cfz resistance and remains UPS dependent for survival. To examine the potential link between proteasome catalytic subunit composition and cellular response to Cfz, we altered the composition of proteasome catalytic subunits via interferon-γ treatment or siRNA knockdown in H727 cells. Our results showed alteration in composition of proteasome catalytic subunits results in sensitization of H727 cells to Cfz. It supported that proteasome inhibition by alternative PIs may still be a valid therapeutic strategy for patients with relapsed MM after having received treatment with Cfz. With this in mind, we designed and synthesized a small library of epoxyketone-based PIs by structural modifications at the P1′ site. We observed that a Cfz analog, harboring a hydroxyl substituent at its P1′ position was cytotoxic against cancer cell lines with de novo or acquired resistance to Cfz. These results suggested that peptide epoxyketones incorporating P1′-targeting moieties may have the potential to overcome Cfz resistance mechanisms in cells. The immunoproteasome (IP), an inducible proteasome variant which is harboring distinct catalytic subunits, LMP2, MECL1 and LMP7 of the proteasome typically expressed in cells of hematopoietic origin, plays a role in immune response and is closely linked to inflammatory diseases. It has been reported that the IP is upregulated in reactive glial cells surrounding amyloid β (Aβ) deposits in brains of Alzheimer’s disease (AD) patients and AD animal models. To investigate whether the IP is involved in the pathogenesis of AD, we examined the impact of IP inhibition on cognitive function in AD mouse models. We observed that YU102, an epoxyketone peptide targeting the IP catalytic subunit LMP2, improved cognitive dysfunction in AD mice without clearance of Aβ deposition or tau aggregation. Our cell line model study also showed a potential mode of action of YU102 which is suppressing pro-inflammatory cytokine production in microglial cells. It suggested that LMP2 contributes to microglia-mediated inflammatory response. These findings supported that LMP2 may offers a valuable therapeutic target for treatment of Alzheimer’s disease, expanding the therapeutic potential of the LMP2-targeting strategy.

β1i is a major catalytic subunit of the immunoproteasome, an alternative form of the constitutive proteasome, and its upregulation has been demonstrated in a variety of disease states including cancer. Our lab has developed a small molecule inhibitor of β1i, dubbed UK-101. While UK-101 causes apoptosis in cancer cell lines, it was not clear whether this apoptotic effect was directly mediated by its irreversible inhibition of β1i. Since off-target effects are major roadblocks for the development of new and effective pharmaceuticals, target validation studies in this system would assist in the further progression of β1i inhibitors towards preclinical trials. Our hypothesis was that the expression and catalytic activity of β1i is important for the growth and proliferation of the PC-3 prostate cancer cell line, therefore the apoptotic effect seen upon treatment of PC-3 cells with UK-101 was due solely to its covalent inhibition of β1i. To test this hypothesis, a number of complementary approaches were used. The expression of β1i in PC-3 cells was increased by the treatment of these cells with interferon-gamma or tumor necrosis factor-alpha, natural inducers of the immunoproteasome. The expression of β1i in PC-3 cells was decreased using small interfering RNA or short hairpin RNA, in a transient or stable manner, respectively. All of these cells were then treated with UK-101. The efficacy of UK-101 decreased in the interferon-gamma treated cells but did not change in any other the other cell lines, suggesting that UK-101 was not specific for β1i. This was confirmed using a molecular probe of the proteasome and demonstrated that UK-101 bound to other proteasome catalytic subunits. Additional experiments were performed to determine the effect of β1i on the proliferation of PC-3 cells. Simply removing the β1i using small interfering RNA reduces the viability of these cells. Other studies demonstrated that a mutation of β1i which inhibited its catalytic activity reduced the viability of cells when compared to those containing the wild type protein. Overall, our data indicate that β1i is a potential therapeutic target in prostate cancer. Further medicinal chemistry efforts will be required develop UK-101 into a truly selective proteasome inhibitor.

The immunoproteasome, known to play an important role in MHC class I antigen processing and presentation, have been linked to neurodegenerative diseases and hematological cancers. However, the pathophysiological functions of the immunoproteasome in these diseases are still not very well established. This can be attributed mainly to the lack of appropriate molecular probes that selectively target the immunoproteasome catalytic subunits. Herein, we report the development of a small molecular inhibitor (AM) that selectively targets the major catalytic subunit, LMP2, of the immunoproteasome. We show that the compound covalently modifies the LMP2 subunit with high specificity in human prostate cancer cell. AM was also shown to selectively inhibit the chymotrypsin-like activity of LMP2 subunit. More importantly, the anti-proliferative activity of AM is more pronounced in prostate cancer cells that highly express LMP2 without inducing toxicity in normal cells. These results implicate an important role of LMP2 in regulating cell growth of malignant tumors that highly express LMP2. Subsequently, the modes of action of AM were investigated. Prostate cancer cells that highly express LMP were shown to induce G2/M cell cycle arrest and apoptosis via PARP cleavage when treated with the compound. Similar to epoxomicin, the treatment of AM induced the accumulation of poly-ubiquitination in prostate cancer cells, which indicates the inhibition of proteolysis. However, unlike epoxomicin, the treatment of AM did not appear to inhibit the activation of inflammation. In conclusion, these results suggest that the LMP2 inhibitor, AM, may induce cytotoxicity prostate cancer cells that highly express LMP2 catalytic subunit in similar modes of action as epoxomicin but it does not involve the inflammatory pathway.

Le protéasome et l’immunoprotéasome possèdent trois activités catalytiques différentes et dégradent une grande variété de protéines. Leur rôle central dans de nombreux processus tels que le contrôle du cycle cellulaire et circadien, la régulation des voies de défense contre le stress font d’eux des cibles majeures dans le processus de vieillissement et dans le traitement des cancers. Aussi, le lien établi entre l'horloge circadienne et les défenses anti-oxydantes suggère que l'homéostasie redox des protéines et leur élimination par le protéasome soient modulées par l'horloge circadienne. L’étude d’inhibiteurs bivalents et non covalents du protéasome a montré que de cibler deux sites actifs à la fois permet d’augmenter l’efficacité et la sélectivité. Nous avons également effectué l'évaluation de la pénétration cellulaire de ces molécules à l’aide de sondes fluorescentes non covalentes conçues par le laboratoire. Nous avons montré que les activités peptidases du protéasome présentent une rythmicité circadienne en antiphase avec le taux de protéines carbonylées. L'expression circadienne de NRF2 et de l'activateur PA28 expliquerait cette stimulation de l'activité du protéasome. De plus, contrairement aux fibroblastes jeunes, aucune modulation circadienne du protéasome et du niveau de ROS n’a été mise en évidence dans les fibroblastes sénescents. Des résultats préliminaires de nos inhibiteurs sur le modèle cellulaire HEK293 synchronisé montrent que leur efficacité varie en fonction du rythme circadien et qu’elle est meilleure lorsque les activités du protéasome sont minimales. Ces résultats suggèrent que le protéasome puisse être une nouvelle cible pour la chronothérapie. / Proteasome and immunoproteasome have three different catalytic activities, and degrade a wide variety of proteins. Their central role in many pathways such as control of cellular and circadian cycles, regulation of the defense pathways against stress make them major targets in the aging process and in the treatment of cancer. Also, the link between the circadian clock and antioxidant defenses suggests that the redox homeostasis of proteins and their removal by the proteasome could be modulated by the circadian clock.Studying non-covalent and bivalent proteasome inhibitors showed that targeting two active sites at the same time results in increasing their efficiency and selectivity. We also carried out the evaluation of the cell penetration of these molecules using non-covalent fluorescent probes designed in the laboratory.We have shown that the peptidase activities of proteasome exhibit circadian rhythmicity in antiphase with protein carbonyl intracellular content. The circadian expression of NRF2 and PA28 activator can explain this stimulation of proteasome activity. Moreover, in contrast to young fibroblasts, no circadian modulation of the proteasome and the level of ROS have been observed in senescent fibroblasts.Preliminary results obtained with our inhibitors on the synchronized HEK293 cell model suggest that their effectiveness varies depending on the circadian rhythm and they are more efficient when proteasome activities are minimal. These results suggest that the proteasome could be a new target for chronotherapy.

Die Studie beinhaltet eine biochemisch-molekularbiologische Analyse des 20S-Proteasoms und seiner Aktivierung durch Proteine der PA28-Familie. Das 20S-Proteasom ist die zentrale Epitop-prozessierende cytosolisch-nukleäre Protease des MHC-Klasse-I-Antigenpräsentationsweges. In Mikroglia, wie auch in anderen Zellen, unterliegt das Proteasom einer Interferon-g-(IFN-g)-vermittelten strukturellen Plastizität, d.h. einer Substitution der Untereinheiten der Aktiven Zentren. Durch diesen Austauschmechanismus werden proteolytische Schnittpräferenzen modifiziert, was für die Hierarchie von cytotoxischen T-Zellantworten von Bedeutung ist. Lipopolysaccharide (LPS) bewirken in Mikroglia ebenfalls Veränderungen der proteasomalen Zusammensetzung des 20S-Komplexes und seines PA700-Aktivators. Dies ist ein Hinweis auf die Rolle des Proteasoms auch bei der Prozessierung von Antigenen des endolysosomalen Antigenpräsentationsweges. Die Modulation der Zusammensetzung des 20S-Proteasoms in Mikroglia durch IFN-g und LPS ist ein weiterer Beleg für die Rolle der Mikroglia bei der zellulären Immunantwort im Zentralnervensystem. Funktionen des Proteasoms in der MHC-Klasse-I-Antigenpräsentation werden durch den Proteasomaktivator PA28 optimiert. Die PA28-Proteinfamilie besteht aus den Proteinen PA28a, PA28b und PA28g. Diese Studie trägt - basierend auf kinetischen Modellierungen, Mutagenese- und Protein-Protein-Interaktionsstudien und Untersuchungen zur MHC-Klasse-I-Antigen-präsentation in PA28-Transfektanten - zur funktionellen Neubewertung dieser drei Proteine bei. Die drei PA28-Proteine sind autonome Aktivatoren des Proteasoms. Heteromere PA28ab-Komplexe verursachen eine stärkere Aktivierung des Proteasoms als die homomeren PA28a-oder PA28b-Komplexe. Das PA28g-Protein ist in vitro ein schwacher Aktivator verschiedener proteasomaler Peptidaseaktivitäten, der dennoch in unserem in vivo-Modell eine verbesserte MHC-Klasse-I-Antigenpräsentation bewirkt. Kinetische Argumente sprechen für Funktionen der PA28a und PA28b-Proteine als Translokasen für Peptidsubstate und -produkte des Proteasoms. Anhand des HBx-Proteins des Hepatitis B-Virus wird die Möglichkeit illustriert, das vorgestellte Modell zur Analyse viraler Faktoren anzuwenden, die mit der Aktivierung des Proteasoms durch PA28 interferieren. / The 20S proteasome is the proteolytic core complex of the main cytoplasmic and nuclear protein degradation system. One of the numerous functions assigned to the 20S proteasome is the generation of antigenic peptides from intracellular proteins. MHC class I surface presentation of antigenic peptides is one key event of the cellular immune response. The presented study was aimed on the biochemical and molecular-biological analysis of the 20S proteasome and its activation by regulatory proteins of the PA28 family. In the brain, microglial cells are the major antigen presenting cells and they respond sensitive to pathologic events. Cultured mouse microglia was used as a model to study the correlation between microglial activation parameters and structural plasticity of the 20S/26S proteasome. In response to interferon-g , constitutive active site subunits were replaced by inducible subunits as described for other cellular systems. These replacements result in altered proteolytic cleavage preferences, indicating that activated microglia adapts its proteasomal subunit composition to the requirements of an optimized MHC class I epitope processing. Since lipopolysaccharide (LPS), a glycolipid of bacterial pathogens, also alters proteasome subunit composition of the 20S proteasome and its activator PA700, a function of microglial proteasome in processing of antigens of the endolysosomal pathway has been postulated. The modulation of the 20S proteasome subunit composition indicates that microglia is part of the cellular immune response in the central nervous system. The proteasome activator PA28 has been reported to optimize MHC class I antigen presentation. The PA28 protein family is composed of three proteins: PA28a, PA28b and PA28b. Based on kinetic modelling approaches, mutagenesis of activator proteins, protein-protein interaction studies and investigation of MHC class I antigen presentation in PA28-transfected cell lines a evaluation of functions of these three proteins has been performed. In vitro investigations revealed that the recombinant PA28 proteins are activators of peptidase activities of the proteasome. Heteromeric PA28ab complexes are stronger activators than homomeric PA28a o PA28b complexes. Recombinant PA28g is a weak activator of several peptidase activities. Nevertheless, in PA28g transfected B8-fibroblasts preliminary results indicate an improved MHC class I presentation. Based on kinetic evidence, a model has been presented indicating that PA28a and PA28b proteins act as peptide translocases, performing the import of substrates or the export of products into the catalytic chamber of the 20S proteasome. Finally, as examplified with the HBx protein of Hepatitis B virus, the opportunity is presented to use an in vitro reconstitution system of proteasomal activation to examine viral proteins interfering with proteasomal activation.

Le protéasome (CP) contrôle la majorité des fonctions cellulaires par la dégradation des protéines intracellulaires. En plus d’exprimer le CP, les vertébrés expriment également l’immunoprotéasome (IP), caractérisé par des préférences de dégradation distinctes. Le rôle le mieux caractérisé pour l’IP est la génération d’antigènes adaptés pour la liaison au complexe majeur d’histocomptabilité de classe I (CMH-I). Cependant, les nombreux phénotypes observés au niveau de cellules déficientes en IP ou avec une mutation révèlent que l’IP influence des fonctions immunitaires indépendamment de la génération d’antigènes et peut atténuer le stress présent au niveau de cellules non-immunitaires. L’objectif de cette thèse était de caractériser les rôles de l’IP qui ne sont pas reliés à la génération d’antigènes associés au CMH-I. L’analyse du transcriptome de cellules dendritiques IP-déficientes en cours de maturation révèle que l’IP affecte l’expression de plus de 8 000 transcrits. L’IP affecte l’expression génique principalement au niveau transcriptionnel en contrôlant l’abondance de régulateurs de transcriptions tels que NF-κB et les membres des familles IRF et STAT. Les cellules dendritiques IP-déficientes sont également moins efficaces pour activer des lymphocytes T CD8+, même chargées artificiellement avec des quantités optimales d’antigènes associés au CMH-I. En outre, nos études montrent que l’IP est fortement exprimé au niveau de cellules de patients atteints de leucémie myéloïde aigue. L’expression de l’IP est intrinsèque aux leucémies, puisque qu’elle n’est pas corrélée à la présence de lymphocytes sécréteurs d’IFN-γ. De plus, l’expression d’IP est particulièrement élevée au niveau de leucémies monocytaires et/ou possédant un réarrangement MLL. Notamment, des analyses de corrélation montrent que l’IP est connecté à des gènes impliqués dans le métabolisme, l’activité mitochondriale et la réponse au stress. En effet l’inhibition de la sous-unité PSMB8 de l’IP mène à l’accumulation de protéines ubiquitinées et la mort de cellules leucémiques monocytaires. Globalement, nos travaux montrent que le rôle de l’IP n’est pas limité à la génération d’antigènes, mais qu’il peut contrôler l’expression génique et la survie des leucémies. / By regulating protein degradation, constitutive proteasomes (CP) control practically all cellular functions. In addition to CP, vertebrates express immunoproteasomes (IP), which display distinct substrate preferences. The first non-redundant role ascribed to IP is its enhanced ability to generate MHC I-associated antigens. However, deletion or inhibition of IP subunits can affect several immune cell functions independently of MHC-I antigen generation. Moreover, recent work has shown that IP can be expressed in non-immune cells to deal with cell stress. Thus, we wished to investigate the roles of IP that are not related to antigen generation and that are not redundant with the CP. Based on profiling of WT and IP-deficient maturing mouse dendritic cells (DCs), we report that IP regulate the expression of more than 8,000 transcripts. The broad impact of IP on gene expression is cell-autonomous, mediated mainly at the transcriptional level, and involves major signaling pathways including IRFs, NF-kB and STATs. Moreover, even when engineered to present optimal amounts of antigenic peptides, IP-deficient DCs are inefficient for in vivo T-cell priming. In addition, consistent with the fact that cancer cells endure proteotoxic stress, we report that acute myeloid leukemia (AML) cells from patients express high levels of IP genes. Expression of IP genes in AML is a cell-autonomous and IFN-independent feature that correlates with the methylation status of IP genes, and is particularly high in AML with a monocytic phenotype and/or MLL rearrangement. Notably, IP inhibition leads to accumulation of polyubiquitinated proteins and cell death in IPhigh but not IPlow AML cells. Co-clustering analysis reveals that genes correlated with IP subunits in monocytic AMLs are primarily implicated in cell metabolism and proliferation, mitochondrial activity and stress responses. Overall, our studies show that the role of IP is not limited to antigen processing and reveals major non-redundant roles for IP in transcription regulation and resistance to cell stress in AML.

Le système ubiquitine-protéasome est le principal mécanisme par lequel les protéines intracellulaires sont dégradées. Le protéasome dit constitutif (PC) est donc essentiel à l’homéostasie mais aussi à la régulation de la majorité des processus cellulaires importants. La découverte d’un deuxième type de protéasome, appelé immunoprotéasome (IP), soulève toutefois de nouvelles questions. Pourquoi existe-t-il plus d’un type de protéasome ? L’IP a-t-il des rôles redondants ou complémentaires avec le PC ? L’IP étant présent principalement dans les cellules immunitaires ou stimulées par des cytokines, plusieurs groupes ont tenté de définir son rôle dans la réponse immunitaire. Or, l’implication de son homologue constitutif dans un éventail de processus non spécifiquement immunitaires nous laisse croire que l’IP pourrait lui aussi avoir un impact beaucoup plus large. L’objectif de cette thèse était donc de caractériser certains rôles cellulaires de l’IP dans les cellules dendritiques. Nous avons d’abord étudié l’impact global de l’IP sur la présentation antigénique de classe I. Ce faisant, nous avons pu déterminer ses deux contributions principales, soit l’augmentation drastique du nombre et de la diversité des peptides présentés sur les complexes majeurs d’histocompatibilité de classe I. Les différences de clivage entre le PC et l’IP pourraient expliquer en partie cette diversité du répertoire peptidique, notamment par l’affinité apparente de l’IP pour les régions protéiques non structurées. Dans un deuxième temps, nous avons dévoilé un nouveau rôle de l’IP sur un processus dépassant le cadre immunitaire : la transcription. Nous avons découvert que l’IP modifie l’abondance des ARNm en agissant principalement au niveau de leur synthèse. L’impact de l’IP sur le transcriptome est majeur et serait dû en partie à une dégradation différente de facteurs de transcription des familles IRF, STAT et NF-kB. Les cellules dendritiques IP-déficientes activent moins efficacement les lymphocytes T CD8+ et nous croyons que cette défaillance est causée (du moins en partie) par la perturbation transcriptomique provoquée par l’absence d’IP. Il importe donc de comprendre les différents rôles moléculaires de l’IP afin de mieux définir sa contribution globale au fonctionnement de la cellule et comprendre l’avantage évolutif, au niveau de l’organisme, procuré par une telle plasticité du système ubiquitine-protéasome. / The ubiquitin-proteasome system is the major mechanism by which intracellular proteins get degraded. Constitutive proteasomes (CPs) are thus essential for cellular homeostasis but also to regulate the majority of important cellular processes. However, the discovery of a second type of proteasome, named immunoproteasome (IP), raises new questions. Why are there more than one type of proteasome? Does the IP perform redundant or complementary roles with the CP? The IP is predominantly expressed in immune or cytokine-stimulated cells and several groups worked at defining its role during the immune response. Yet, the implication of its constitutive homolog in a variety of processes suggests that the IP may also have a much broader impact. The objective was to characterize cellular roles of the IP in dendritic cells. We first studied the global impact of the IP on class I antigen presentation. We discovered that the IP drastically increases the number and the diversity of peptide presented by class I major histocompatibility complexes. Cleavage differences between the CP and the IP are likely part of the explanation for this peptide repertoire diversity, notably due to IP’s apparent affinity for unstructured protein regions. Second, we discovered a new role for the IP in a process unrestricted to the immune system: transcription. We found that the IP affects transcript abundance mostly at the level of mRNA synthesis. The impact of IPs on the transcriptome is major and would be partly based on a different degradation of IRF, STAT and NF-kB transcription factor family members by the two types of proteasomes. IP-deficient dendritic cells are less potent activators of CD8+ T cells and we believe that this defect is at least partly caused by the transcriptome alterations induced by the absence of IPs. It is therefore important to understand the different molecular roles of the IP in order to better define its global contribution to cellular functions and to understand the evolutionary advantage, at the level of the organism, brought by such plasticity of the ubiquitin- proteasome system.

No description available.

Le système ubiquitine-protéasome est le principal mécanisme par lequel les protéines intracellulaires sont dégradées. Le protéasome dit constitutif (PC) est donc essentiel à l’homéostasie mais aussi à la régulation de la majorité des processus cellulaires importants. La découverte d’un deuxième type de protéasome, appelé immunoprotéasome (IP), soulève toutefois de nouvelles questions. Pourquoi existe-t-il plus d’un type de protéasome ? L’IP a-t-il des rôles redondants ou complémentaires avec le PC ? L’IP étant présent principalement dans les cellules immunitaires ou stimulées par des cytokines, plusieurs groupes ont tenté de définir son rôle dans la réponse immunitaire. Or, l’implication de son homologue constitutif dans un éventail de processus non spécifiquement immunitaires nous laisse croire que l’IP pourrait lui aussi avoir un impact beaucoup plus large. L’objectif de cette thèse était donc de caractériser certains rôles cellulaires de l’IP dans les cellules dendritiques. Nous avons d’abord étudié l’impact global de l’IP sur la présentation antigénique de classe I. Ce faisant, nous avons pu déterminer ses deux contributions principales, soit l’augmentation drastique du nombre et de la diversité des peptides présentés sur les complexes majeurs d’histocompatibilité de classe I. Les différences de clivage entre le PC et l’IP pourraient expliquer en partie cette diversité du répertoire peptidique, notamment par l’affinité apparente de l’IP pour les régions protéiques non structurées. Dans un deuxième temps, nous avons dévoilé un nouveau rôle de l’IP sur un processus dépassant le cadre immunitaire : la transcription. Nous avons découvert que l’IP modifie l’abondance des ARNm en agissant principalement au niveau de leur synthèse. L’impact de l’IP sur le transcriptome est majeur et serait dû en partie à une dégradation différente de facteurs de transcription des familles IRF, STAT et NF-kB. Les cellules dendritiques IP-déficientes activent moins efficacement les lymphocytes T CD8+ et nous croyons que cette défaillance est causée (du moins en partie) par la perturbation transcriptomique provoquée par l’absence d’IP. Il importe donc de comprendre les différents rôles moléculaires de l’IP afin de mieux définir sa contribution globale au fonctionnement de la cellule et comprendre l’avantage évolutif, au niveau de l’organisme, procuré par une telle plasticité du système ubiquitine-protéasome. / The ubiquitin-proteasome system is the major mechanism by which intracellular proteins get degraded. Constitutive proteasomes (CPs) are thus essential for cellular homeostasis but also to regulate the majority of important cellular processes. However, the discovery of a second type of proteasome, named immunoproteasome (IP), raises new questions. Why are there more than one type of proteasome? Does the IP perform redundant or complementary roles with the CP? The IP is predominantly expressed in immune or cytokine-stimulated cells and several groups worked at defining its role during the immune response. Yet, the implication of its constitutive homolog in a variety of processes suggests that the IP may also have a much broader impact. The objective was to characterize cellular roles of the IP in dendritic cells. We first studied the global impact of the IP on class I antigen presentation. We discovered that the IP drastically increases the number and the diversity of peptide presented by class I major histocompatibility complexes. Cleavage differences between the CP and the IP are likely part of the explanation for this peptide repertoire diversity, notably due to IP’s apparent affinity for unstructured protein regions. Second, we discovered a new role for the IP in a process unrestricted to the immune system: transcription. We found that the IP affects transcript abundance mostly at the level of mRNA synthesis. The impact of IPs on the transcriptome is major and would be partly based on a different degradation of IRF, STAT and NF-kB transcription factor family members by the two types of proteasomes. IP-deficient dendritic cells are less potent activators of CD8+ T cells and we believe that this defect is at least partly caused by the transcriptome alterations induced by the absence of IPs. It is therefore important to understand the different molecular roles of the IP in order to better define its global contribution to cellular functions and to understand the evolutionary advantage, at the level of the organism, brought by such plasticity of the ubiquitin- proteasome system.

Das Proteasom spielt eine zentrale Rolle beim Proteinabbau und der Antigen-Generierung für die adaptive Immunantwort. Vertebraten exprimieren zwei Typen des proteolytischen 20S-Kernkomplexes: das konstitutive c20S (mit den aktiven Untereinheiten beta 1, 2, 5) und das Immunoproteasom i20S (mit den Immunountereinheiten LMP2, MECL-1, LMP7). Die i20S-Expression wird durch Interferon_gamma (IFNg) induziert, was die Antigen-Präsentation auf MHC Klasse I und die Immunantwort gegen infizierte bzw. maligne entartete Zellen durch cytotoxische T-Zellen steigert. Proteasomen werden über komplexe, bisher unvollständig verstandene Biogenese-Prozesse formiert. Die initialen Schritte der humanen 20S-Formation wurden in dieser Arbeit untersucht und eine Methode zur Isolation früher Assemblierungsintermediate (EPIs) etabliert. Die 20S-Biogenese bedarf der Assistenz von Hilfsfaktoren wie dem Proteasom-Maturierungsprotein POMP. Diese Komponente von Precursorkomplexen stellt das erste Substrat gereifter c20S dar. In dieser Arbeit konnte erstmalig gezeigt werden, dass POMP ebenfalls die i20S-Formation vermittelt und sich die Biogenese von c20S und i20S hinsichtlich der Maturierungskinetik unterscheidet. POMP wird durch IFNg induziert und interagiert mit der Immunountereinheit LMP7. Dieses molekulare Zusammenspiel bewirkt eine schnellere Maturierung von i20S- im Vergleich zu c20S- Precursorkomplexen, wodurch POMP einem schnelleren Abbau unterliegt. Die forcierte i20S-Biogenese ist eine intrinsische Eigenschaft und unabhängig von weiteren, IFNg-induzierten Faktoren. Nur die LMP7_E2-Variante vermittelt die schnelle Degradation von POMP, während das nicht funktionelle LMP7_E1 mit einer anderen Prosequenz nicht in i20S-Vorläufer inkorporiert wird. Somit führt die alleinige LMP7_E1-Expression in IFNg-stimulierten Carcinom-Zellen zu einer i20S-Defizienz, was eine mögliche Immunevasions-Strategie darstellt. Weiterhin besitzen beide 20S-Typen unterschiedliche Halbwertszeiten: i20S sind, unabhängig von weiteren IFNg-induzierten Proteinen, signifikant instabiler als c20S. Somit werden i20S sowohl schneller formiert als auch zügiger wieder abgebaut als c20S, womit sie typische Eigenschaften cytokin-regulierter Proteine aufweisen. Die i20S-Formation ist also eine transiente Antwort und stellt ein effizientes Instrument zur schnellen Reaktion auf immunologische Herausforderungen wie z.B. eine Infektion dar. Nach einer wirksamen Immunantwort erlaubt die geringere i20S-Stabilität eine schnelle Rückkehr zur standardmäßigen c20S-Expression. / The proteasome plays a crucial role in protein degradation and antigen generation for the adaptive immune response. Vertebrates express two types of the proteolytic 20S core complex: the constitutive proteasome c20S (with the active subunits beta 1, 2 and 5) and the immunoproteasome i20S (with the immunosubunits LMP2, MECL-1 and LMP7). Interferon_gamma (IFNg) induces the i20S expression, that supports a more efficient MHC class I antigen presentation and an effective immune response against infected or malignant cells by cytotoxic T-cells. Proteasomes are formed by a complex and not well understood biogenesis program. The initial steps in the human 20S formation have been analyzed in this thesis and a method for the isolation of ´early proteasome assembly intermediates´ (EPIs) has been established. The 20S biogenesis requires the assistance of accessory factors like the proteasome maturation protein POMP. This component of precursor complexes becomes the first substrate of the matured c20S. The described experiments demonstrate for the first time that POMP mediates the i20S formation and that biogenesis of c20S and i20S differ in their maturation kinetics. POMP is induced by IFNg and interacts with the immunosubunit LMP7. This molecular interplay provokes a faster maturation of i20S compared to c20S precursor complexes, whereby POMP becomes subject to a faster degradation. The accelerated i20S biogenesis is an intrinsic characteristic and independent of additional IFNg-induced factors. Exclusively the LMP7_E2 variant causes the rapid degradation of POMP, whereas the non-functional LMP7_E1 bearing another prosequence is not incorporated into i20S precursor complexes. Thus, LMP7_E1 expression in IFNg-stimulated carcinoma cells leads to a i20S deficiency pointing out a possible immune evasion strategy. In addition, both 20S types display different half-life values: i20S are, independent of other IFNg-induced proteins, significantly less stable than c20S. Thus, i20S are not only faster assembled, but also more quickly decomposed compared to c20S, showing typical attributes of proteins regulated by cytokines. Consequently, i20S formation is a transient response and represents an efficient instrument for a rapid adjustment to varying immunological challenges like an infection. Once the immune response has been effective, the lower stability of i20S permits an expeditious return to the standard c20S expression.