Doctoral Thesis
Refine
Document Type
- Doctoral Thesis (4) (remove)
Has Fulltext
- yes (4)
Is part of the Bibliography
- no (4)
Keywords
- Proteasom (4) (remove)
Despite the extensive ongoing research, there still exist plenty of diseases whose mechanisms have not yet been fully understood, one such example being proteasome-related disorders. Over the last few years, an increasing number of studies have been initiated
to elucidate their driving pathophysiological mechanisms. Determining the systematic effects of genomic alterations occurring in genes encoding 19S proteasome subunits is a key to comprehend the molecular basis of syndromic intellectual disability (ID) pathogenesis and
the subsequent design of new targeted therapies. Therefore, the main objective of my research was to contribute to the identification of potential drivers of syndromic ID, and thereby pave the way for the development of new targeted therapy approaches. In this regard, my aim was to characterize tissue, proteomic and metabolomic changes in cells from patients with PSMC5 mutations and uncover a potential dysregulation of various biochemical and/or inflammatory pathways.
To this end, I undertook a comparative examination of control and patient T cells expanded from peripheral blood mononuclear cells (PBMCs). First, I assessed the proteasome composition in these samples (both in its denaturized and native form), by means of
SDS-PAGE, native PAGE and western-blotting. Moreover, I determined proteasome chymotrypsin-like activity by measure of Suc-LLVY-AMC peptidase activity assay. In addition, I analysed the activation status of the ER stress and mTOR pathway by RT-PCR and SDS-PAGE /western-blotting prior to a subsequent analysis of T-cell markers.
The data show that the investigated p.(Pro320Arg) and p.(Arg201Trp) de novo heterozygous missense mutations in the PSMC5 gene do not cause haploinsufficiency as the steady-state expression level of the PSMC5/Rpt6 full-length protein does not vary between control and patient cells. Further analysis of control and patient T cells under non-reducing conditions revealed that PSMC5/Rpt6 mutants were less efficiently incorporated into 26S proteasome complexes than their wild-type counterparts. The failure to assemble PSMC5/Rpt6 into fully mature proteasomes was associated with a reduced proteasome chymotrypsin-like activity in patient T cells, as determined by in-plate assays. These data unambiguously demonstrate that both of the p.(Pro320Arg) and p.(Arg201Trp) PSMC5 mutations identified in patients suffering from syndromic ID are loss-of-function mutations. Interestingly, my data further show that proteasome dysfunction in these patients was accompanied by abnormalities in mTOR signalling and T-cell differentiation, as determined by western-blotting and flow cytometry, respectively.
Altogether, our data identified for the first time PSMC5 as a disease-causing gene for
a syndromic form of ID. How proteasome dysfunction caused by PSMC5 variants contributes to disease pathogenesis, remains to be fully determined.
Zusammenfassung Das Ubiquitin-Proteasom-System (UPS) ist das wichtigste nicht-lysosomale proteolytische System für den Abbau intrazellulärer Proteine. Die Inhibition des UPS kann dosisabhängig den apoptotischen Zelltod oder die Induktion einer protektiven Stressantwort auslösen. In der therapeutischen Anwendung der Proteasominhibition sind neben der Wirksamkeit auch exakte Kenntnisse der Wirkungsweise essentiell, um die primären Effekte durch die Proteasominhibition besser zu erfassen und die initialen Effekte der Vermittlung dieser zellulären Reaktion besser zu verstehen. In dieser Arbeit wurden Methoden der Proteom- und Transkriptomebene kombiniert, um komplexe zelluläre Veränderungen von Endothelzellen nach Proteasomhemmung zu charakterisieren. Weiterhin wurde mittels Immunpräzipitation Ubiquitin-bindender Proteine untersucht, welche Substrate des Proteasoms nach Proteasomhemmung in Endothelzellen stabilisiert werden. Primäre humane Endothelzellen (Huvec) wurden als Modell gewählt und mit niedrigen bzw. hohen Dosierungen des Proteasominhibitors MG132 behandelt. Das Expressionsprofil wurde in einer Zeitkinetik innerhalb der ersten 6h mittels Affymetrix Chipanalysen bestimmt. Die differentielle Proteinexpression nach zwei Stunden Proteasomhemmung wurde im Gesamtzelllysat durch 2D Gelelektrophorese und anschließende Silberfärbung visualisiert. Dabei konnten mehr als 20 regulierte Proteine identifiziert werden, welche zuvor nicht direkt im Zusammenhang mit der Vermittlung einer protektiven Stressantwort nach niedrig dosierter Proteasomhemmung bekannt waren. Durch Korrelation mit den parallel durchgeführten Expressionsarrays konnte die Regulation dieser Proteine als unabhängig von der Transkription erfasst werden. Die funktionelle Annotation der Daten zeigte dabei eine Anreicherung von Proteinen der zellulären Stressantwort, der intrazellulären Signaltransduktion und des oxidativen Stresses. Diese waren differentiell reguliert nach niedrig bzw. hoch dosierter Proteasominhibition. Während die niedrig dosierte Proteasominhibition ein protektives Genmuster zeigte, induzierten hohe Dosen den apoptotischen Zelltod. Auch konnte, in Abhängigkeit vom Grad der Proteasominhibition, ein deutlicher Anstieg der freien Radikale innerhalb der Zellen nachgewiesen werden, was auf die Vermittlung der Apoptose durch freie Radikale hinweist. Mit DJ-1, Peroxiredoxin-1 und -6 wurden mehrere Sensorproteine für oxidativen Stress identifiziert. Um, neben der Proteom- und Transkriptomanalyse, auch gezielt Substrate der Proteasominhibition zu identifizieren und als mögliche Mediatoren der protektiven Stressantwort zu identifizieren, wurden Gesamtzelllysate mittels Ubiquitin-Immunpräzipitation getrennt und Ubiquitin-bindende Proteine per Massenspektrometrie identifiziert. Dabei konnte ein Set von 22 Proteinen identifiziert werden, welche spezifisch nach 2h an der Ubiquitin-spezifischen Matrix gebunden wurden. In diesem Set finden sich vor allem RNA bindende Proteine, Bestandteile des UPS und ribosomale Proteine. Um gezielt transkriptionelle Mediatoren der protektiven Stressantwort nach partieller Proteasominhibition zu identifizieren, wurde eine Subproteom-Analyse mit nukleären Extrakten von Huvec durchgeführt. Nach Visualisierung mittels DIGE-Labeling und quantitativer Auswertung konnten 361 regulierte Spots nach 2 stündiger Proteasominhibition erfasst werden. Von diesen Spots konnten 319 per MALDI-MS identifiziert werden und 152 verschiedenen Proteinen zugeordnet werden. Die funktionelle Auswertung ergab eine deutliche Überrepräsentation von RNA-bindenden und Splicing-relevanten Proteinen. Auch konnte für HnRNP A1, einem RNA-bindenden Protein, eine funktionelle Methylierung sowie eine Veränderung der subzellulären Lokalisierung nachgewiesen werden. Diese Ergebnisse deuten auf die Regulation des zellulären Splicings hin. In parallel dazu angefertigten Exon-sensitiven Affymetrix Arrays wurden über 600 Gene mit differentiell regulierten Exons identifiziert, welche vor allem Gene mit Rezeptor- und Transportproteinen kodieren. Auch eine erhöhte Anzahl von Genen mit Methyltransferase-/Kinaseaktivität wurde identifiziert. Insbesondere die Methyltransferasen, welche auch bei der posttranslationalen Modifikation von HnRNP A1 eine Rolle spielen, zeigten dabei eine signifikante Regulation unter niedrig dosierter Proteasominhibition. Zusammengefasst tragen die in dieser Arbeit vorgestellten Ergebnisse dazu bei, einen detaillierten Überblick über die initialen Prozesse niedrig dosierter Proteasominhibition in Endothelzellen zu geben. Die Daten deuten dabei auf eine schnelle Adaptation der Zellen nach partieller Proteasomhemmung mittels Aktivierung einer anti-oxidativen Stressantwort sowie durch Regulation von Splicingvorgängen hin, die letztendlich in einem veränderten Expressionsprofil der Endothelzellen münden. Mit diesem systematischen Ansatz und der Kombination von Proteom- und Transkriptomanalyse konnten dabei einzelne Targets für die Mediation einer protektiven zellulären Stressantwort nach partieller Proteasomhemmung identifiziert werden.
The maintenance of protein homeostasis in muscle by degradation systems, e.g. the autophagy lysosomal pathway (ALP) and the ubiquitin-proteasome system (UPS), is of great importance. It prevents the accumulation of nonfunctioning and not properly folded proteins, which can lead to protein aggregate myopathies (PAMs) and several other protein storage diseases. Degradation by the UPS depends on the transfer of ubiquitin to a target protein. This happens in a cascade of E1-E2-E3 proteins. This process is also involved in protein location and regulation of protein activity. E3 ligases are often tissue specific. Muscle RING-finger proteins (MuRFs) are a family of really interesting new gene (RING)-Finger E3 ubiquitin ligases, that are almost exclusively expressed in the striated muscle. They play a role in muscle wasting, but are also important for the maintenance of the structure of striated muscle. MuRF proteins are also involved in the regulation of the striated muscle energy metabolism. Previous work has demonstrated that MuRF1/MuRF3 DKO mice show a protein surplus myopathy characterized by an accumulation of myosin heavy chain proteins in striated muscles and a reduction in function of both heart and skeletal muscle. The aim of this study was to test the hypothesis that the myopathic phenotype of MuRF1/MuRF3 DKO mice is mediated by a disturbed energy homeostasis in the heart and skeletal muscle, with focus on mitochondrial function. Because sex-specific differences have not been investigated in these mice so far, a further aim was to investigate any differences between male and female mice.
To test these hypotheses, we measured the weight of the heart and the hindlimb muscles tibialis anterior and soleus to detect a possible hypertrophy in the DKO mice. Hematoxylin and eosin staining of histological cross sections of the tibialis anterior were performed to investigate protein accumulations. Muscle function was quantitated via grip strength and specific force measurements. Possible changes in protein amounts were detected via mass spectrometry analyses and western blot analyses. Changes in gene expression were investigated by qRT-PCR. Coimmunoprecipitation was used to determine direct interactions between proteins. Protein stability and ubiquitination were investigated by cycloheximide (CHX) and ubiquitination assays, respectively.
DKO mice showed an increase in heart and skeletal muscle weights. Grip strength assays revealed limb weakness of DKO mice. H&E staining of histological cross sections of the tibialis anterior muscle (TA) showed protein aggregates within myofibers. Mass spectrometry analyses of proteins isolated from TA and heart muscle revealed an increase of muscle stress markers and structural proteins in DKO mice, while proteins involved in the energy metabolism were reduced. Especially interesting here were the proteins of the mitochondrial electron transport chain (ETC), which play a major role in the energy production of the mitochondria by catalyzing the phosphorylation of ADP to ATP, the universal energy carrier in all living organisms. These changes were more pronounced in TA compared to heart. Western blot and qRT-PCR results of ETC subunits supported our proteome data. They also revealed a sex-specific difference, in which the reduction ETC subunits was more pronounced in females than males. In female
TA NDUFB8, SDHB, UQCRC2, MTCO1 and ATP5 were significantly reduced compared to controls, while only UQCRC2 and ATP5 were decreased in male TA compared to controls. A significant reduction in gene expression of Ndufb8, Sdhb, Mtco1 and Atp5 was detected in TA of female mice compared to controls, while only Ndufb8, Sdhb and Atp5 were decreased in male TA compared to controls. We observed the same pattern in Heart of male (protein: NDUFB8; mRNA: Mtco1) and female (protein: UQCRC2, MTCO1, ATP5; mRNA: Sdhb, Mtco1) DKO mice compared to their controls. The reduction in ETC subunits was paralleled by a reduction in complex I and complex III activity in the TA of DKO mice, but not in heart. However, this was only significant in the TA of female but not male mice. Mechanistical analyses using coimmunoprecipitation, cycloheximide chase and ubiquitination assays showed that MuRF1 physically interacted with the transcriptional repressor histone deacetylase 5 (HDAC5), mediated its ubiquitination as well as its UPS-dependent degradation. The absence of MuRF1 and MuRF3 in DKO mice let to an increase in the amounts of HDAC5 in TA. Because HDAC5 binds to PGC-1α, the master regulator of mitochondrial biogenesis (encoded by Ppargc1a), we investigated its gene expression in DKO muscle and found it to be reduced.
These data connect MuRF1 and MuRF3 directly to the striated muscle energy metabolism, by regulating mitochondrial function. The results provide insights into the development of PAMs and possibly other protein storage diseases, where a decrease of mitochondrial function has already been described.
The proteasome is a major part of the ubiquitin-proteasome-system playing an important role in cell homeostasis due to its protein quality control function. Moreover, the proteasome is involved in cell cycle regulation and in the regulation of transcription factors. Upon induction of interferons, or treatment with lipopolysaccharides, an isoform of the standard-proteasome is composed, named immunoproteasome (i-proteasome). The i-proteasome is constitutively expressed in immune cells and deficiency of proteolytic subunits of this multiprotein complex has been associated with a poor outcome during infectious diseases. I-proteasome-deficiency has been shown to result in reduced MHC class I presentation. Using mice which are deficient for all three proteolytic active subunits LMP2, MECL-1 and LMP7, we could demonstrate that i-proteasome-deficiency lead to an altered recruitment of immune cells to the CNS when challenged with the intracellular parasite Toxoplasma gondii, resulting in increased frequencies of neutrophils and other cells of myeloid origin. The shift to reduced frequencies of CD45highCD11blow lymphocytes can be further explained by a decreased migratory capacity of i-proteasome-deficient CD8+ T cells. In contrast to previous studies using other pathogens, effector function of CD8+ as well as CD4+ T cells, measured by frequencies of IFNγ, TNF, IL-2 and granzyme B producing cells, were not impaired in these mice, whereas induction of CD4+ Tregs was strongly reduced. In addition, we found that parasite control was comparable to control mice and that i-proteasome deletion caused an overall pro-inflammatory cytokine milieu within the brain. Our results indicate that i-proteasome-deficiency lead to prolonged tissue inflammation during T. gondii infection which could be an explanation for the more severe course of disease observed in these mice.
The Src homology domain containing phosphatase 2 (SHP2) is a tyrosine phosphatase modulating several signaling pathways and therefore has an influence in cell cycle, differentiation, proliferation and cell activation. However, SHP2 is assumed to play a negative role during T-cell activation as the phosphatase has been shown to inhibit T-cell receptor-induced signaling cascades. Although, various gain-of-function mutations in the SH2 or PTP domain of this phosphatase, such as D61Y, have been associated with myeloproliferative diseases such as juvenile myelomonocytic leukemia (JMML), effects of such mutations on T cells have not been addressed in scientific literature so far. Therefore, in the second part of this thesis we could demonstrate that D61Y mutation in the SH2 domain of SHP2 did not cause JMML pathology when only introduced into T cells. Especially in aged mice, T cells of SHP2 mutant mice showed an increased expression of cell adhesion molecule CD44. In accordance with these findings, we observed increased influenza A virus-specific T cells in the bone marrow of SHP2 D61Y mutant mice, indicating a role of the phosphatase in memory formation or maintenance of CD8+ Tem. Although SHP2D61Y mice revealed a comparable viral clearance, IFNγ production of virus experienced CD4+ and CD8+ T cells was diminished compared to control mice, underlining a negative involvement of the phosphatase in the JAK/STAT1 signaling axis as suggested before by studies using mice with SHP2-/- T cells.