Refine
Year of publication
Document Type
- Doctoral Thesis (95)
Has Fulltext
- yes (95)
Is part of the Bibliography
- no (95)
Keywords
- Heterologe Genexpression (3)
- Endothelin (2)
- G-Protein gekoppelte Rezeptoren (2)
- Gentherapie (2)
- HIV (2)
- Membrane Proteins (2)
- NMR-Spektroskopie (2)
- gene therapy (2)
- 5-Lipoxygenase (1)
- APOBEC3G (1)
Institute
- Biochemie und Chemie (63)
- Biochemie, Chemie und Pharmazie (28)
- Pharmazie (5)
- Biowissenschaften (1)
- Georg-Speyer-Haus (1)
To overcome poor treatment response of pediatric high-risk acute lymphoblastic leukemia (ALL), novel treatment strategies are required to reactivate programmed cell death in this malignancy. Therefore, we take advantage of using small-molecule antagonists of Inhibitor of apoptosis (IAP) proteins, so called Smac mimetics such as BV6, which are described to overcome apoptosis resistance and thereby sensitize tumor cells for several apoptotic stimuli. To address the question whether redox alterations can sensitize leukemic cells for Smac mimetic-mediated cell death, we interfered with the cellular redox status in different ALL cell lines. Here, we show for the first time that redox alterations, mediated by the glutathione depleting agent Buthioninesulfoximine (BSO), prime ALL cells for BV6-induced apoptosis. Besides ALL cell lines, BV6/BSO cotreatment similarly synergizes in cell death induction in patient-derived primary leukemic samples. In contrast, the combination treatment does not exert any cytotoxicity against peripheral blood lymphocytes (PBLs) or mesenchymal stroma cells (MSCs) from healthy donors, suggesting some tumor selectivity of this treatment. We also identify the underlying molecular mechanism of the novel synergistic drug interaction of BSO and BV6. We demonstrate that both agents act in concert to increase reactive oxygen species (ROS) production, lipid peroxidation and finally apoptotic cell death. Enhanced ROS levels in the combination treatment account for cell death induction, since several ROS scavengers, like NAC, MnTBAP and Trolox attenuate BSO/BV6-induced apoptosis. BSO/BV6-induced ROS can be mainly classified as lipid peroxides, since the vitamin E derivate α-Tocopherol as well as Glutathione peroxidase 4 (GPX4), which both specifically reduce lipid-membrane peroxides, prevent lipid peroxidation, caspase activation and cell death induction. Vice versa, GPX4 knockdown and pharmacological inhibition of GPX4 by RSL3 or Erastin enhance BV6-induced cell death. Importantly, cell death induction critically depends on the formation of a complex consisting of RIP1/FADD/Caspase-8, since all complex components are required for ROS production, lipid peroxidation and cell death induction. Taken together, we demonstrate that BSO and BV6 cooperate to induce ROS production and lipid peroxidation which are eventually required for caspase activation and cell death execution. Collectively, findings of this study indicate that BV6-induced apoptosis is mediated via redox alterations offering promising new treatment strategy to overcome apoptosis resistance in ALL.
Recently, two of the most common types of bone cancers in children and young adults have been proven to exhibit vulnerability to poly(ADP)-ribose polymerase, (PARP) inhibitors (e.g. olaparib, talazoparib). Ewing’s sarcoma (ES) are reported to harbor a fusion gene EWS-FLI1 (85%), inducing tumorigenesis. Additional, as the fusion gene acts as aberrant transcription factor, it similarly induces elevated PARP expression levels sensitizing ES to PARP inhibition. Second, by an exome sequencing approach in a set of primary osteosarcomas (OS) we identified mutation signatures being reminiscent of BRCA deficiency. Therefore, the sensitivity of a panel of OS cell lines to either talazoparib single treatment or in combination with several chemotherapeutic drugs was investigated.
To screen ES tumor cell lines against PARP inhibitors we applied four different PARP inhibitors (talazoparib, olaparib, niraparib and veliparib) that are frequently being used for clinical studies. We combined those PARP inhibitors with a set of chemotherapeutics (temozolomide (TMZ), SN-38, etoposide, ifosfamide, doxorubicin, vincristine and actinomycin D) that are part of the first-line therapy of ES patients. Here, we demonstrate how PARP inhibitors synergize with TMZ or SN-38 to induce apoptosis, whereas the combination of PARP inhibitors with the other drugs are not favorable. By investigation of key checkpoints in the molecular mechanisms of cell death, the pivotal role of the mitochondrial pathway of apoptosis mediating the synergy between olaparib and TMZ was revealed.
Employing talazoparib monotherapy in combination with or without several chemotherapeutic drugs (TMZ, SN-38, cisplatin, doxorubicin, methotrexate and etoposide/carboplatin), the correlation between homologous recombination (HR) repair deficiency (BRCAness) and the response to talazoparib as prototypical PARP inhibitor was validated in different OS cell lines. By calculation of combination indices (CI) and fraction affected (Fa) values, we identified TMZ as the most potent chemotherapeutic drug in combination with talazoparib inducing the mitochondrial apoptotic pathway in OS.
In our studies of two independent tumor entities with contrary genetic background we identified the combination of PARP inhibitor and TMZ as being most effective. Our studies point out that after TMZ induced DNA methylation and concomitant PARP trapping, DNA damage-imposed checkpoint kinase activation consequently induces G2-cell cycle arrest. Subsequent, PARP inhibitor/TMZ causes MCL-1 degradation, followed by activation of BAK and BAX, succeeding in loss of mitochondrial outer membrane potential (LMMP) and activation of downstream effector-caspases in mitochondrial apoptosis. Our findings emphasize the importance of PARP inhibition in order to chemosensitize ES, which express high PARP levels, or OS that bear features of BRCAness.
Three types of post-translation modifications (PTMs) containing N-glycosylation, phosphorylation, ubiquitylation were characterized in diffuse large B-cell lymphoma (DLBCL) on a global scale using quantitative mass spectrometry based proteomics technology in this study.
DLBCL is the most common type of malignant lymphomas and has a heterogeneous gene expression profiling, phenotype and clinical response to chemotherapy. DLBCL is a good model for the correct classification of cancers into molecularly different subtypes, which benefits for the selection of rational therapeutic strategies. It resulted in two histologically indistinguishable subtypes-activated B-cell-like (ABC) subgroup and germinal center B-cell-like (GCB) subgroup according to gene expression profiling. Signals originating from the B-cell receptor (BCR), the key protein on the surface of B cells, promote growth and survival of DLBCL. Antigen-dependent/independent BCR signaling is found in DLBCL subtypes.
Recent researches reveal that glycosylation plays role in human cells via site-specific regulation. Aberrant N-glycosylation in BCR-related effectors, such as, CD79a, immunoglobulin M or G (IgM or IgG), has been found to be associated with lymphoid malignancies. However, accurate quantification of intact glycopeptides and their individual glycan moieties in a cell-wide manner is still challenging. Here we established a site-specific quantitative N-glycoproteomics platform termed SugarQuant. It included a fast sample preparation workflow using Protein Aggregation Capture (PAC), an optimized multi-notch MS3 acquisition workflow (Glyco-SPS-MS3), a self-developed R-based tool (GlycoBinder). The robustness and accuracy of quantitation in SugarQuant were proved in a study using the different amounts of TMT-labelled IgM N-glycopeptides spiked into a background of TMT-labelled yeast peptides. Next, we used SugarQuant to identify and quantify more than 5000 unique glycoforms in Burkitt’s lymphoma cells treated with a series of doses of 2-deoxy-2-fluoro-L-fucose (2FF) and determine the more accurate site-specific glycosylation changes that occurred upon inhibition of fucosylation compared to using MS2 analysis. It revealed that 2FF-sensitive N-glycosylation on key players in BCR-mediated signaling in DG75. Furthermore, 2FF treatment also affects phosphorylation of the key players involving in B cell receptor signaling.
Then we investigated the site-specific quantitative N-glycoproteome in the cell lines of DLBCL subtypes using SugarQuant. More than 7000 unique intact glycopeptides (glycoforms) were quantified in five ABC DLBCL and four GCB DLBCL cell lines. The glycoproteome mapping (intact glycopeptide expressions) in each cell line allows to segregate DLBCL subtypes. The majority of these glycoforms were from the key cell-surface BCR effectors, such as IgM, CD79 and PTPRC. Lastly, we investigated the change of fucosylated glycopeptides in TMD8 cell line upon knockout of the fucosyltransferase FUT8, which is responsible for core-fucose synthesis, and by the treatment with 2FF. The results revealed that FUT8 might also regulate the synthesis of sub/terminal fucose on glycan chain and the inhibition of fucosylation increased the sialyated glycopeptide expression.
Phosphorylation is involved in regulating multiple processes as an important mediator in BCR signaling. Likewise, ubiquitylation plays vital roles in the activation of the nuclear factor-kappaB (NF-κB) pathway in BCR signaling. There are two vital upstream BCR-proximal tyrosine kinases, Bruton’s tyrosine kinase (BTK) and spleen tyrosine kinase (SYK), which regulate the auto-phosphorylation and phosphorylation of other proteins in BCR signaling pathway. Here we investigated the dynamics of downstream phosphorylation and ubiquitylation signaling in ABC DLBCL and GCB DLBCL cell lines upon the inhibitions of BTK and SYK using quantitative proteomics strategy. In the phosphoproteome analysis, a large dataset of quantified phosphorylation sites was obtained in the three ABC and four GCB DLBCL cell lines. BCR signaling in the subtypes of DLBCL cell lines was found to be highly individual in distinct cell lines. These significantly regulated phosphorylation events in each cell line with individual treatment were involved in multiple Reactome pathways, such as, M phase, signaling by Rho GTPases and diseases of signal transduction. Moreover, the gene regulation-related biological processes including chromosome organization and medication, DNA metabolic process, nuclear export, were involved in the DLBCL cell lines. In the ubiquitinome analysis, we identified more than 15,000 ubiquitylation sites in two ABC and one GCB cell lines upon the inhibition of BTK and SYK. The different ubiquitylation events observed in ABC and GCB subtypes revealed distinct BCR signaling pathways in two subtypes. The similar signaling perturbations across each cell line upon BTK and SYK inhibition, which were obtained from the significantly regulated ubiquitylated peptides expression, revealed the cell-type-specific concordance in ubiquitylation regulation upon BTK and SYK inhibition. These ubiquitylation modified proteins who bore the significantly regulated ubi-peptides in the samples were also found to be highly involved in gene regulatory processes.
...
Protein quality control (PQC) machinery is in charge of ensuring protein homeostasis in the cell, i.e. proteostasis. Chaperones assist polypeptides throughout their maturation until functionality is achieved. This process might be disrupted in the presence of mutations or external damaging agents that affect the folding and stability of proteins. In this case, proteins can be efficiently recognized and targeted for degradation in a controlled manner. Ubiquitylation refers to the covalent attachment of one or more ubiquitin moieties to faulty proteins, thus triggering their degradation by the 26S proteasome.
More than 30% of proteins need cofactor molecules. Lack of cofactors renders proteins non-functional. We wanted to understand how the PQC deals with wild-type proteins in the absence of their cofactors. Several studies have indicated the importance of the riboflavin-derived cofactor FAD in the stability of individual flavoproteins, and hence we assumed that loss of flavin should mediate a targeted degradation of this group of proteins. Indeed, our mass spectrometry experiments showed that flavoproteome levels decreased under riboflavin starvation. The oxidoreductase NQO1 was used as a model enzyme to further investigate the mechanism of flavoproteome targeting by the PQC. We showed that cofactor loading determines ubiquitylation of NQO1 by the co-chaperone CHIP, both in vivo and in vitro. Furthermore, subtle changes in the C-terminus of NQO1 in the absence of FAD seemed to be crucial for this recognition event. ApoNQO1 interactome differed from holoNQO1. Chaperones and degradation factors were enriched on NQO1 upon cofactor withdrawal, probably to support maturation and prevent aggregation of the enzyme.
Loss of protein folding and stability, even to a small extent, can enhance the aggregating behavior of proteins. Proper loading with FAD reduced the co-aggregation of NQO1 with Aβ1-42 peptide. We assumed that the flavoproteome might represent aggregating-prone species under riboflavin deprivation. Supportingly, reversible apoNQO1 aggregates were observed in vivo in the absence of cofactor. General amyloidogenesis in vivo also increased under these conditions, apparently as a result of flavoproteome destabilization. In this context, we think that our data might have important implications considering the onset and development of conformational diseases.
This work has shed some light on the therapeutic implications of riboflavin deficiency as well. The sensitivity of melanoma cells towards the alkylating agent methyl methanesulfonate (MMS) increased under riboflavin starvation. Subsequent analyses indicated that a complex metabolic reorganization, mostly affecting proliferation and energy metabolism, occurs in response to starvation. What we suggest to call “flavoaddiction” can be understood as the dependence of melanoma cells on the flavoproteome structural and functional intactness to survive chemotherapy. Understanding this cellular reprogramming in detail might reveal new possibilities for future therapies.
Acute lymphoblastic leukemia (ALL), a neoplastic disorder of blood cells of the lymphoid lineage, is the most frequent childhood cancer. In spite of increasing survival rates, the outcome for adults, infants or relapsed patients is still less favorable, highlighting the need for novel treatment options. Reactive oxygen species (ROS) are important signaling molecules that are involved in a variety of cellular pathways. As high ROS levels lead to oxidative stress and irreversible oxidation of cellular macromolecules, the production and elimination of ROS is tightly controlled. Therefore, cells express several antioxidant molecules and enzymes, including glutathione, catalase and the thioredoxin (Trx) system, to balance ROS levels. As cancer cells were found to have increased ROS levels that could contribute to tumor progression and metastasis, they rely strongly on these antioxidant systems to prevent oxidative damage, making cancer cells especially vulnerable to ROS-inducing treatments. ROS and oxidative stress have been shown to induce programmed cell death via different pathways, however the exact mechanisms that couples oxidative signaling and cell death is not completely understood.
As a disturbance of the cellular redox homeostasis was reported during leukemia development and progression, we wanted to determine the potential of Trx inhibitors for ALL therapy. Additionally, we aimed to further understand the role of ROS and subsequent protein oxidation in the induction and execution of programmed cell death.
First, we demonstrated that the Trx1 inhibitor PX-12 induced cell death in three ALL cell lines. Further analysis of the events leading to PX-12-induced cell death in FADD-deficient (FD) Jurkat cells revealed an increase in ROS levels and oxidation-mediated dimer formation of peroxiredoxin 3 (PRDX3). Interestingly cell death was inhibited by the thiol-containing antioxidant N-acetylcysteine (NAC), but not by non-thiol-containing ROS scavengers. PX-12 treatment further induced cleavage of caspase-9 and -3 and activation of the pro-apoptotic BCL-2 protein BAK, leading us to the conclusion that mitochondria-dependent apoptosis was induced. Interestingly, we could demonstrate an important role for the BH3-only protein NOXA in the mediation of PX-12-induced apoptosis as knock-down of NOXA prevented cell death induction and BAK activation. Our findings give novel insights into the mechanism of PX-12-induced cell death in ALL cell lines and underscores the potential of PX-12 for the treatment of ALL.
To further understand the processes leading to cell death upon inhibition of the Trx system, we analyzed global protein oxidation in Jurkat FD cells upon treatment with the Trx reductase inhibitor Auranofin. In line with previous results, Auranofin induced intrinsic apoptosis that was dependent on BAK and accompanied by increased ROS levels. Using a BIAM Switch Assay followed by mass spectrometry, we demonstrated that Auranofin treatment induced oxidation of over 200 proteins. We identified several proteins whose oxidation upon Auranofin treatment was expected, like Trx1, Trx2 and several peroxiredoxins. Additionally, we verified oxidation of APAF1-interacting protein (APIP) and protein arginine N-methyltransferase (PRMT1) that are both implicated in the regulation of apoptosis. With this analysis we were able to demonstrate that Auranofin treatment leads to changes in global protein oxidation. Whether oxidation of the determined proteins changes their functionality and contributes to apoptosis induction remains to be elucidated.
As we identified BAK as an important player in PX-12- and Auranofin-induced cell death in the previous parts of this study, we wanted to further understand its involvement in ROS-mediated cell death. First analyses in wild-type (WT) and BAK-/- murine embryonic fibroblasts (MEFs) revealed that BAK was essential for Auranofin-induced cell death and that this cell death was caspase-independent in MEFs. Interestingly, BAK oxidation was induced upon treatment with Auranofin, but not upon stimulation with the apoptosis-inducing compound Etoposide. Expression of mutated BAK, with either one or both oxidation-sensitive cysteines mutated to oxidation-insensitive serines, revealed that mutating already one cysteine protected cells from Auranofin , but not Etoposide-induced cell death. Of note, mutation of the BAK BH3 domain rescued MEFs from both, Auranofin- and Etoposide-mediated cell death. The presence of cysteine residues also altered BAK interactions as observed by a mass spectrometric analysis of Auranofin-treated MEFs expressing either WT or cysteine-less BAK. We identified interactions of WT BAK with proteins involved in mitochondrial fission and vesicle transport upon Auranofin treatment. Of note, interaction with proteins involved in apoptosis, like BAX or BCL-XL, was not changed between WT and cysteine-less BAK. Our results demonstrate a critical role for BAK oxidation in Auranofin-induced cell death. Furthermore, we identified novel oxidation-dependent BAK interaction partners.
To conclude, this study highlights the potential of ROS-inducing treatments for ALL therapy and provides novel insights into the redox regulation of programmed cell death.
Protein ubiquitination is a post-translational modification that typically involves the conjugation of ubiquitin to substrate proteins via a three-enzyme cascade and regulates a wide variety of cellular processes. Recent studies have revealed that SidE family of Legionella effectors such as SdeA catalyzes novel phosphoribosyl-linked ubiquitination (PR-ubiquitination) of serines in host substrate proteins utilizing NAD+, without the need of E2, E3. The catalytic core of SdeA comprises a mono-ADP-ribosyltransferase (mART) domain that functions to ADP-ribosylate ubiquitin, and a phosphodiesterase (PDE) domain that processes ADP-ribosylated ubiquitin and transfers the resulting phosphoribosylated ubiquitin to serines of substrates.
To date, extensive efforts have been made to study the function of SdeA and mechanism of SdeA mediated PR-ubiquitination, however, the cellular effects of this novel ubiquitination and phosphoribosylation of ubiquitin remained poorly understood. In our study, using biochemical and cell biological approaches, we explored the biological effect of phosphoribosylation of ubiquitin caused by SdeA in cells. We found that phosphoribosylated ubiquitin is not available for conventional ubiquitination, thereby phosphoribosylation of ubiquitin impairs numerous classical ubiquitination related cellular processes including mitophagy, TNF-α signaling and proteasomal degradation.
The precise temporal regulation of the functions of bacterial effectors during Legionella infection by other effectors with antagonizing activities has been well studied so far. Not surprisingly, PR-ubiquitination catalyzed by SidE family effecters is tightly controlled as well, it has been long known that effector SidJ counteracts the toxicity of SdeA to yeast cells. Interestingly, in an experiment for verifying the activity of SidJ, we found that Legionella lysate lacking SidJ was still able to remove ubiquitin from PR-ubiquitinated substrates. Using biochemical approach we identified DupA and DupB, two Legionella bacterial effectors that specifically reverse the novel serine PR-ubiquitination catalyzed by SdeA. We found that DupA and DupB possess a highly homologous PDE domain that removes ubiquitin from PR-ubiquitinated substrates by cleaving the phosphodiester bond between the phosphoribosylated-ubiquitin and serines of substrates. Catalytically deficient mutant DupA H67A strongly binds to PR-ubiquitinated proteins but not capable of cleaving PR-ubiquitin, using it as a trapping bait we identified over 180 substrates of PR-ubiquitination, including a number of ER and Golgi proteins.
In particular, we found that exogenously expressed SdeA localizes to the Golgi apparatus via its C-terminal region and disrupts the Golgi. We validated the identified potential substrates of SidE effectors and found that SdeA modifies Golgi tethering proteins GRASP55 and GRASP65. Using mass spectrometry analyses we identified four serine targets (S3, S408, S409, S449) of GRASP55 PR-ubiquitinated by SdeA in vitro. Ubiquitination of GRASP55 serine mutant in cells co-expressing SdeA or infected with Legionella was markedly decreased, compared with that of the wild-type GRASP55. In addition, with co-immunoprecipitation analyses we found that SdeA-catalyzed ubiquitination regulates the function of GRASP55. PR-ubiquitinated GRASP55 exhibited reduced self-interaction compared to unmodified GRASP55, expression of GRASP55 serine mutant in cells in part rescued Golgi damage caused by SdeA. Furthermore, our study reveals that Golgi structure disruption caused by SdeA does not result in the recruitment of Golgi membranes to the Legionella-containing vacuoles. Instead, it affects cellular secretory pathway including cytokine secretion in cells.
Taken all together, this work expands the understanding of this unconventional PR-ubiquitination catalyzed by Legionella effectors and sheds light on the functions of PR-ubiquitination by which Legionella regulates the Golgi function and secretion pathway during bacterial infection.
Inhibitoren der Apoptose (IAP, inhibitor of apoptosis) Proteine spielen eine wichtige Rolle in Bezug auf Zelltodregulation und es ist anzunehmen, dass eine Dysregulation dieser Proteine zu einer Tumorentwicklung und Tumorprogression beiträgt. Erhöhte Expressionslevel von IAP Proteinen verhindern die Aktivierbarkeit des Zelltodprogrammes von Tumorzellen und eine Reihe von Studien konnte bereits erhöhte IAP Level in Tumorzelllinien sowie in primären Tumorproben nachweisen. Des Weiteren korrelieren erhöhte Expressionslevel von IAPs in Tumoren mit Behandlungsresistenzen und schlechten Prognosen für die Patienten.
Das diffuse großzellige B-Zell Lymphom (DLBCL, diffuse large B-cell lymphoma) zählt zu den häufigsten Subtypen der Non-Hodgkin Lymphome (NHL) mit 40 % aller neu diagnostizierten NHL Fälle. DLBCL ist eine sehr heterogene Erkrankung die in drei verschiedene Gruppen klassifiziert wurde: aktivierter B-Zell Typ (ABC, activated B-cell), Keimzentrum B-Zell Typ (GCB, germinal center B-Cell) und Mediastinaler großzelliger B-Zell Typ (PMBL, primary mediastinal B-cell lymphoma). Erhöhte Expressionslevel von zellulärem IAP1 (cIAP, cellular IAP) und cIAP2 wurden ebenfalls in primären Tumorproben von DLBCL Patienten nachgewiesen. Smac mimetics wurden entwickelt, um IAPs zu antagonisieren und stellen damit eine Behandlungsstrategie für DLBCL Patienten dar, denn ca. 40 % aller DLBCL Patienten entwickeln ein Rezidiv oder erreichen gar keine Remission unter Standardtherapie. Jedoch ist der Effekt von Smac mimetics in einer Einzelbehandlung limitiert, weswegen Kombinationstherapien mit Smac mimetics eine vielversprechende Strategie für ihren klinischen Einsatz darstellen. Aus diesem Grund haben wir in dieser Arbeit den Effekt von Smac mimetic in Kombination mit Proteasom-Inhibitoren analysiert und einen speziellen Fokus auf den molekularen Mechanismus des ausgelösten Zelltodsignalweges gelegt.
Die Kombination verschiedener Konzentrationen des Smac mimetics BV6 mit dem Proteasom-Inhibitor carfilzomib (CFZ) löst in allen drei getesteten DLBCL Subtypen (ABC, GCB und PMBL) Zelltod aus. Die Kalkulation des Kombinationsindexes (CI, combination index) sowie des Bliss Scores, zwei quantitative Parameter zur Bestimmung eines Synergismus, zeigen, dass fast alle getesteten Kombinationen einen Synergismus aufweisen. Dies verdeutlicht, dass eine Co-Behandlung von BV6 und CFZ eine wirksame Kombination ist um Zelltod in DLBCL Zelllinien auszulösen. Außerdem zeigt eine Kombination von BV6 mit anderen Proteasom-Inhibitoren wie ixazomib (IXA) oder oprozomib (OPR), ebenfalls eine synergistische Reduktion der Zellviabilität. Diese Ergebnisse deuten darauf hin, dass der detektierte Effekt nicht auf eine Substanz limitiert ist, sondern, dass ein genereller Effekt von Smac mimetic und Proteasom-Inhibitoren vorliegt, um Zellviabilität in DLBCL zu reduzieren. BV6 und CFZ induzieren einen apoptotischen Zelltod, da sie die Spaltung und Aktivierung von Initiator- und Effektorcaspasen (Caspasen-3, -7, -8 und -9) initiieren und sich der induzierte Zelltod mit Hilfe des Caspasen-Inhibitors zVAD.fmk verhindern lässt. Die Behandlung mit BV6 und CFZ führt zu einer Akkumulation von NIK, ein Protein welches zur Aktivierung des non-kanonischen NF-kB Signalweges benötigt wird. Weitere Untersuchungen zeigen jedoch, dass NIK nicht an der Zelltodinduktion beteiligt ist, da eine siRNA-basierte Herunterregulierung des NIK Proteins keinen Einfluss auf die Zelltodinduktion nimmt. Ebenfalls ist der Zelltod unabhängig von dem TNFa Signalweg, da weder eine Behandlung mit dem TNFa Inhibitor Enbrel den Zelltod verringern kann noch eine zusätzliche Gabe von TNFa den Zelltod erhöht. Weitere mechanistische Studien zeigen eine kritische Rolle der mitochondrialen Apoptose für den BV6/CFZ-vermittelten Zelltod. Unter Behandlung mit BV6/CFZ wurde eine Aktivierung von BAX und BAK nachgewiesen, welche beide mit verantwortlich für die Porenbildung in der mitochondrialen Membran sind. Eine Herunterregulation dieser beiden Proteine mittels siRNA reduziert signifikant den durch BV6/CFZ-induzierten Zelltod auf ein Minimum. Gleichzeitig löst eine Co-Behandlung mit BV6/CFZ einen Verlust des mitochondrialen Membranpotentials (LOMMP, loss of mitochondrial membrane potential) aus. In Übereinstimmung mit den vorherigen Experimenten, zeigen wir eine Akkumulation von mitochondrialen reaktiven Sauerstoffspezies (ROS; reactive oxygen species), sowie einen generellen Anstieg des allgemeinen ROS Levels. Eine Behandlung mit BV6/CFZ zeigt eine deutliche Akkumulation des pro-apoptotischen Proteins NOXA. Um dessen funktionelle Relevanz zu überprüfen, wurde die Proteinmenge von NOXA mittels siRNA stark reduziert. Eine Behandlung mit der Kombination aus BV6 und CFZ zeigt daraufhin eine signifikant reduzierte Zelltodinduktion, was die funktionelle Relevanz von NOXA für den BV6/CFZ-vermittelten Zelltod unterstreicht. Immunopräzipitationsstudien zeigen, dass in RIVA und U2932 Zellen NOXA konstitutiv an seinen anti-apoptotischen Bindungspartner MCL-1 gebunden ist, was die Zellen bereits darauf vorbereitet Apoptose zu durchlaufen. Dieses sogenannte „primen“ für Apoptose wird durch die Behandlung mit BV6 und CFZ weiter verstärkt, da es die Bindung zwischen NOXA und MCL-1 weiter erhöht. Dadurch wird die Balance zwischen pro- und anti-apoptotischen Proteinen zu Gunsten der pro-apoptotischen Proteine verschoben und die Induktion von Apoptose begünstigt.
Insgesamt zeigen die Ergebnisse, dass DLBCL Zelllinien sensitiv auf eine Behandlung mit Smac mimetic und Proteasom-Inhibitor reagieren und damit eine mögliche neue Behandlungsstrategie für diese heterogene Tumorerkrankung darstellt.
Inducing cell death in tumor cells is a major goal of anti-cancer therapy. However, the preferable mode of cell death to induce is under debate. Apoptosis is known to be an anti-inflammatory and pro-resolving type of programmed cell death, whereas necroptosis results in the release of danger-associated molecular patterns (DAMPs) and is pro-inflammatory. Efferocytosis of apoptotic cells by macrophages results in a pro-resolving switch of macrophages polarization and is required to induce resolution of inflammation. This impact of apoptotic cells on macrophages is a non-desired consequence of cell death in tumors, which are often characterized by an overshooting wound healing response. Moreover, apoptosis resistance is frequently observed in cancer cells. To overcome apoptosis resistance in cancer cells, necroptosis can be induced as an alternative mechanism for cancer treatment. Interferons (IFNs) play an important role in tumor immune responses and act by inducing the expression of IFN-stiumlated genes (ISGs). Furthermore, IFNs were shown to be able to induce necroptosis together with Smac-mimetics when caspases are inhibited in different cancer cell lines. Necroptosis is induced by phosphorylation and activation of receptor-interacting serine/threonine-protein kinase 1 (RIPK1), RIPK3 and pseudokinase mixed lineage kinase domain-like (MLKL).
In my thesis, we first identified MLKL as an ISG in various cancer cell lines. MLKL upregulation was found to be a general feature of IFN signaling since both type I and type II IFNs increase the expression of MLKL. IFNy was able to upregulate MLKL at messenger ribonucleic acid (mRNA) and protein level indicating that MLKL is elevated transcriptionally. Indeed, Actinomycin D chase experiments showed that inhibition of transcription abolished MLKL upregulation upon IFN treatment. Both, knockdown of the IFNy-activated transcription factors interferon regulatory factor 1 (IRF1) and signal transducer and activator of transcription 1 (STAT1) as well as knockout of IRF1 significantly dampened MLKL mRNA upregulation, demonstrating that STAT1 and especially IRF1 are necessary to induce MLKL expression. This first part of the study highlights the upregulation of MLKL by IFNy as valuable tool to sensitize cells towards necroptosis and by that overcome apoptosis resistance in cancers.
When compared to apoptosis, the immune response to necroptotic cells and the polarization of macrophages phagocytosing necroptotic cells is not well studied. In most studies, cell death was induced by biological or chemical compounds, which may lead to artifacts by affecting the macrophages and triggering of unrelated signaling pathways. Therefore, in the second part of my thesis we used a pure cell death system of NIH 3T3 cells expressing either dimerizable caspase 8 or oligomerizable RIPK3 to induce cell death. Addition of B/B-Homodimerizer (dimerizer) to the cells resulted in apoptosis or necroptosis, which was confirmed by caspase 3/7 activation, phosphorylation of MLKL and inhibitor experiments, respectively. We analyzed the effect of dying cells on peritoneal macrophages by establishing a co-culture in a transwell system. The genetic profile of macrophages co-cultured with dying cells was evaluated by whole transcriptome RNA sequencing. In macrophages co-cultured with necroptotic cells genes corresponding to chemotaxis and hypoxia pathways were upregulated. A significant proportion of hypoxia-related pathways are mediated by hypoxia-inducible factor 1-alpha (HIF-1α), which also induces metabolic changes in polarized macrophages. We could show that macrophages co-cultured with necroptotic cells showed a decreased mitochondrial respiration, indicating an inflammatory (M1) polarization. Protein levels of chemokine C-X-C motif ligand 1 (CXCL1), which was increased in the RNA sequencing data, were also upregulated in supernatant of co-cultured macrophages and of necroptotic cells, demonstrating that necroptotic cells both secrete CXCL1 and induce gene expression of CXCL1 in peritoneal macrophages. This may influence the recruitment of neutrophils as inhibition of necroptosis during Zymosan-A-induced peritonits in mice decreased the levels of neutrophils at day 1 of this model of self-resolving inflammation.
Furthermore, RNA sequencing revealed an unexpected impact of apoptotic cells on macrophage biology as cell cycle and cell division pathways were increased. Enhanced proliferation of macrophages was confirmed by two functional assay with peritoneal macrophages isolated from mice and IC-21 macrophages. Inhibition of apoptosis during Zymosan-A-induced peritonits in mice demonstrated decreased mRNA levels of cell cycle mediators in peritoneal macrophages. Simultaneously with cell cycle activation, gene sets of prostaglandin E2 (PGE2) signaling were upregulated during RNA sequencing. In the second part of my thesis we could demonstrate, that apoptotic cells induce transcription of cell cycle genes and proliferation of macrophages and necroptotic cells are able to influence the chemokine profile of macrophages and thereby the recruitment of neutrophils.
Cell-free-synthesized voltage-gated proton channels: Approaches to the study of protein dynamics
(2018)
We often only realize how important health is when diseases manifest themselves through their symptoms and, ultimately, in a diagnosis. Over time, we suffer from many diseases starting with the first childhood disease to colds to gastrointestinal infections. Most diseases pass harmlessly and symptoms fade away. However, not all diseases are so harmless. Alzheimer’s disease, breast cancer, Parkinson’s disease, and colorectal cancer usually cause severe illness with high mortality rates. In pharmaceutical research, efforts are therefore being made to determine the molecular basis of them in order to provide patients with potential relief and, at best, healing. A special group of regulators, involved in the previously mentioned diseases, are voltage-gated proton channels. Thus, the understanding of their structure, function, and potential drug interaction is of great importance for humanity.
Voltage-gated proton channels are localized in the cell membrane. As their name indicates, they are controlled by voltage changes. Depolarization of the cell membrane induces conformational changes that open these channels allowing protons to pass through. Here, the transfer is based on a passive process driven by a concentration gradient between two individual compartments separated by the cell membrane. Voltage-gated proton channels are highly selective for protons and show a temperature- and pH-dependent gating behavior. However, little is known about their channeling mechanism. Previous experimental results are insufficient for understanding the key features of proton channeling.
In this thesis, for the first time, the cell-free production of voltage-sensing domains (VSD) of human voltage-gated proton channels (hHV1) and zebrafish voltage-sensing phosphatases (DrVSP) is described. Utilizing the cell free approach, parameters concerning protein stability, folding and labeling can be easily addressed. Furthermore, the provision of a membrane mimetic in form of detergent micelles, nanodiscs, or liposomes for co-translational incorporations of these membrane proteins is simple and efficient. Both VSDs were successfully produced up to 3 mg/ml. Furthermore, the cell-free synthesis enabled for the first time studies of lipid-dependent co-translational VSD insertions into nanodiscs and liposomes. Cell-free produced VSDs were shown to be active, and to exist mainly as dimers. In addition, also their activation was stated to be lipid-dependent, which has not been described so far. Solution-state NMR experiments were performed with fully and selectively labeled cell-free produced VSDs. With respect to the development of potential drug candidates, I could demonstrate the inhibition of the VSDs by 2-guanidinobenzimidazole (2GBI). Determined KD values were comparable to literature data for the human construct. For the first time, a low affinity for 2GBI of the zebrafish VSD could be described.
In future, the combination of a fast, easy and cheap cell-free production of fully or selectively labeled VSDs and their analysis by solution state NMR will enable structure determinations as well as inhibitor binding studies and protein dynamic investigations of those proteins. The results of these investigations will serve as a basis for example for the development of new drugs. In addition, a detailed description of the lipid-dependent activity might be helpful in controlling the function of voltage-gated proton channels in cancer cells and thereby reducing their growth or disturbing their cell homeostasis in general.
Membrane proteins (MPs) constitute about 30% of the genome and are essential in many cellular processes. In particular structural characterisation of MPs is challenged by their hydrophobic nature resulting in expression difficulties and structural instability upon extraction from the membrane. Despite these challenges, progress in sample preparation and the techniques to solve MP structures has led to 281 unique MP structures as of January 2011. Through the combination of a cell-free expression system and selective labelling strategies, this thesis aimed to advance the structure determination of α-helical MPs by NMR spectroscopy and resulted in the structure determination of a seven-ransmembrane-helix protein. Results were obtained for the 5-lipoxygenase-activating protein (FLAP) and proteorhodopsin (PR). The detergent-based cell-free expression mode proved most efficient for production of both targets, but optimisation of FLAP and PR followed different routes. The presence of a retinal cofactor in PR greatly facilitated the search for an appropriate hydrophobic environment. For structural studies, NMR spectra of FLAP indicated favourable properties of the lysolipid LPPG. In contrast, PR was stable and homogenous in the short-chain lipid diC7PC. As NMR spectra of α-helical MPs are generally characterised by broad lines and signal overlap, selective labelling strategies were essential in the assignment process of both targets. For the backbone assignment of FLAP the transmembrane segment-enhanced (TMS) labelling was developed, employing the six amino acids AFGILV. These residues cluster predominantly in transmembrane helices and form long stretches allowing a large extent of backbone assignment. Besides that, the combinatorial labelling enables identification of unique pairs in the sequence based on a mixture of 15N and 1-13C-labelled amino acids. To find the optimal labelling pattern for a given primary structure, the UPLABEL algorithm has been made available and successfully applied in the backbone assignment of PR. Both selective labelling approaches greatly benefitted from the use of a cell-free expression system to reduce isotope scrambling. Additionally, the de novo structure of PR was determined with an average backbone rmsd of 1.2 Å based on TALOS-derived backbone torsion angles, intrahelical hydrogen bond restraints and distance restraints from the NOE and paramagnetic relaxation enhancement (PRE). A major bottleneck in the NMR structure determination of MPs concerns the number of long-range distances which are often limited. In PR, side chain assignment was enabled by stereo-array isotope labelling as well as selective labelling which provided 33 long-range NOEs. These NOEs stabilised the symmetry of the seven helix bundle. With a total number of 1031, the majority of long-range distances were derived from PREs. The structure of PR reveals differences to its homologues such as the absence of an anti-parallel β-sheet between helices B and C and allows conclusions towards the mechanism of colour tuning.
Proteostasis stressors that destabilize the cellular proteome, like heat shock, trigger transcription and translational reactions leading to the accumulation of heat shock proteins, also called molecular chaperones. During stress, induction of stress response genes is prioritized so that molecular chaperones and other stress response proteins are synthesized to cope with proteome misfolding and aggregation. In order to promote the selective translation of stress-specific genes, translation of others genes that are nonessential for cell survival has to stop. Nonessential protein-coding mRNAs accumulate in the cytosol with the associated proteins to form granular structures called stress granules (SG). These membrane-less organelles are thought to be involved in cell survival, mRNA stabilization and mRNA triage. They were proposed to form via the liquid-liquid phase separation which can be triggered by the high local concentration of RNA-binding proteins. mRNAs were long thought to simply play a scaffolding role by bringing RNA-binding proteins together and allowing their concentration and local aggregation. Recently, the active role of mRNAs in the SG assembly became apparent, too. For example, the spontaneous assembly of total yeast RNA into granules was observed, and these RNA granules showed a large overlap with SG transcriptome. Furthermore, cytosolic mRNAs can be released from polyribosomes under stress and be exposed to the cytosolic contents as free mRNAs. It has been suggested that this massive increase of free mRNA in the cytosol might overload the capacities of RNA-stabilizing proteins. The remaining free mRNA molecules would then become exposed to misfolded and aggregation-prone proteins and trigger granulation.
We investigated the role of free mRNAs in different stress conditions during the early and chronic phases of stress response and explored their involvement in SGs assembly and amlyoidogenesis. We identified and studied the interactome of a free mRNA probe incubated with heat shocked cell lysate by means of quantitative mass spectrometry. Proteomics analysis allowed us to identify 79 interactors of free mRNA. Among these interactors, we focused on the translation initiation factor eIF2α and on the RNA methyltransferase TRMT6/61A. Both interactions were verified biochemically, which confirmed that the association is enhanced in heat shocked lysate. In vitro reconstitution showed that free mRNA and TRMT6 interact directly. Ex vivo pulldowns revealed that eIF2α and TRMT6/61A interact under stress conditions and that this interaction is RNA-dependent.
TRMT6/61A is a tRNA methytransferase responsible for the methylation of the adenosine 58 at the position 1 producing m1A. However, also mRNAs have been recently found to be methylated by TRMT6/61A. Our bioinformatics analyses revealed that significantly more mRNAs enriched in SG contain the motif for methylation than SG-depleted mRNAs. We hypothesized that m1A methylation of mRNAs could constitute a tag for the mRNAs targeting to SGs. TRMT61A knock-down (KD) cell lines were generated using the CRISPR-Cas9 technique. In TRMT61A KD cells, m1A was significantly reduced on mRNAs, which correlated with an increased sensitivity of the cells to proteostasis stress. KD cells also showed defects in SG assembly. In heat shocked cells, an m1A motif-containing mRNA recovered better after returning to normal temperature than a control mRNA with mutated motif. In addition, we could isolate SGs and analyze their m1A and m6A content by mass spectrometry. While m6A content in SG mRNAs was very similar to cytosolic mRNAs, m1A was almost 8 times enriched in SGs. Thus, we could confirm experimentally the results of the bioinformatics analysis and directly support the hypothesis that m1A is a tag to direct mRNAs for sequestration. Finally, we compared amyloidogenesis in wild-type and TRMT61A KD cell lines. Cells with reduced levels of TRMT61A demonstrated an increased accumulation of transfected Aβ and an impaired aggregate clearance. Various assays led us to conclude that the lack of m1A deposition on mRNAs enhanced RNA co-aggregation with amyloids.
Based on our results, we propose a model explaining the fate of free mRNA during proteostasis stress. Upon polysome disassembly, free mRNA is released and becomes free to interact with other proteins, including the methyltransferase TRMT6/61A. TRMT6/61A methylates the freed mRNAs containing the cognate motif. The m1A tag then targets mRNAs to SGs promoting sequestration. Upon stress release, SGs disassemble, thus releasing rescued mRNAs which could now reenter translation and support cell recovery. On the other hand, non-sequestered mRNAs increasingly co-aggregate with aggregating proteins. Thus, deficiency of the N1-adenine methylation of mRNAs due to the lack of TRMT6/61A increases the amount of unpacked mRNAs. The deposition of m1A on mRNAs could then be a way to protect them during exposure to stress, to limit their co-aggregation with misfolded proteins and to allow a faster recovery upon stress release.
All lifeforms have to sense changes in their environment and adapt to possibly detrimental conditions. On a cellular level, the highly elaborate proteostasis network (PN) consisting of housekeeping and stress-induced proteins, confers this tolerance against stress and maintains cellular protein homoestasis. This is essential for survival, as an accumulation of stress-induced protein aggregation will eventually affect the functionality of crucial cellular components and ultimately lead to cell death. The guardians of this balance are the molecular chaperones and their activity-regulating co-haperones. They are engaged in all aspects of protein biogenesis, maintenance and degradation, especially during stress.
The heat shock proteins (HSPs) are the major chaperones in mammals and encompass constitutive and stress-induced isoforms. Among them, the HSP70 and the HSP90 family are the most abundant HSPs and their activity is involved in a great variety of homoestasis and stress-induced tasks.
As part of the protein triage the E3 ligase CHIP (C-terminal HSC70-interacting protein) is an essential activity regulating co-chaperone of HSP70 and HSP90 which provides a link between chaperone mediated protein-folding and various degradation pathways. Due to its decisive function, CHIP is involved in a wide array of cellular processes, especially in clearing misfolded HSP70 client proteins that are prone to aggregate. As a consequence, CHIP was reported to confer protection against many aggregation-induced pathologies of the neuronal system. Additionally, CHIP has been identified as a critical factor in various types of cancer and is implied to affect the development and the longevity of mammals.
Despite the significant progress in the understanding of CHIP’s structure and function, many aspects surrounding its chaperone dependency and its substrate recognition remain unclear. Moreover, due to the variety of substrates in diverse cellular pathways, there are yet many connections to elucidate between CHIP and components of the cellular proteostasis network.
The work of this thesis was focused on the role of CHIP in acute stress response and the corresponding status of chaperone association. Moreover, it was investigated if CHIP, as the connecting ligase of folding and degradation systems, might also provide a link between the PN and the reorganisation of the cellular architecture upon stress exposure.
This has become of increasing interest as recent reports highlight the importance of spatial sequestration in protein quality control.
To this end, subcellular distribution of CHIP was analysed by live-cell microscopy during heat stress. It became obvious that during the heat-induced challenge of the chaperone system, CHIP migrated to new cellular sites. Further experiments suggested that the observed migration to the plasma membrane is a chaperone-independent process and in vitro reconstitution of membrane association confirmed the competitive nature of membranes and chaperones for CHIP binding. A detailed in vivo and in vitro analysis of the newly observed membrane association of CHIP revealed a distinct lipid specificity and a novel direct association with lipids. Binding experiments with recombinantly purified deletion mutants of CHIP identified the TPR domain and a positive patch in the coiled-coil domain as main determinants for the lipid association. Through biochemical and biophysical approaches, the structural integrity and functionality of CHIP upon membrane binding was confirmed and further characterised.
Moreover, mass spectrometry analysis provided a high confidence identification of chaperone-free interactors of CHIP at the plasma membrane and other membranous compartments.
In accordance with the lipid specificity, the Golgi apparatus was one of these sites. Only chaperone-free CHIP had a significant effect on the morphology of the organelle, again confirming the competitive role of chaperones and lipids. With respect to the physiological consequences of the changed localisation of CHIP, preliminary results indicated increased cell death when the ligase localises to cellular membranes. The results lead to the conclusion that CHIP acts as an initiator of early stress adaptation and as a sensor for the severity and strength of the stress reaction.
Reactive oxygen species (ROS) are involved in various signalling mechanisms. Redox homeostasis is important in cancer cells, since they are dependent on upregulated antioxidant defence pathways to cope with elevated ROS levels. Therefore, targeting the antioxidant defence system and/ or increasing ROS to a lethal level may be a feasible strategy to counteract cancer cell progression.
Acute lymphoblastic leukaemia (ALL) is the most frequent malignant childhood cancer, displaying on one side resistance to cell death induction and on the other side elevated ROS levels. Therefore, inducing ferroptosis, a ROS- and iron-dependent cell death pathway might be useful to trigger cell death in ALL as a novel treatment strategy. In the first study of this thesis we observed that RSL3, a glutathione (GSH) peroxidase 4 (GPX4) inhibitor, triggered ROS accumulation and lipid peroxidation which contributed to ferroptotic cell death. These observations were based on suppression of RSL3 stimulated cell death using different ferroptosis inhibitors like Ferrostatin-1 (Fer-1), Liproxstatin-1 (Lip-1), as well as iron chelator Deferoxamine (DFO) and the vitamin E derivate α-Tocopherol (α-Toc). RSL3-triggered ROS and lipid peroxide production were also inhibited through Fer-1 and α-Toc. Furthermore, lipoxygenases (LOX) were activated upon RSL3 stimulation and contributed to ferroptotic cell death in ALL as well. Selective inhibition of LOX with the 12/15-LOX inhibitor Baicalein and the pan-LOX inhibitor nordihydroguaiaretic acid (NDGA) abolished RSL3-induced ROS production, lipid peroxidation and cell death. In addition, RSL3 induced lipid peroxide-dependent ferroptotic cell death in FAS-associated Death Domain (FADD)-deficient, death receptor-induced apoptosis resistant cells, demonstrating that ferroptosis might circumvent apoptosis resistance.
The second part of the study revealed that RSL3 and Erastin (Era), a GSH-depleting agent, inhibiting the cystine/glutamate antiporter system xc- and ferroptosis inducer, cooperated with the Smac mimetic BV6 to trigger cell death in ALL cells. RSL3/BV6 and Era/BV6 combination-induced cell death was dependent on ROS accumulation, but independent of caspases and key modulators of necroptosis. RSL3/BV6-treated ALL cells exhibited classical features of ferroptotic cell death with iron-dependency, ROS accumulation and lipid peroxidation which was diminished through either pharmacological inhibition (Fer-1, DFO, α-Toc) or genetic inhibition by overexpressing GPX4. Interestingly, Era/BV6-induced cell death in ALL cells was independent of iron but dependent on ROS accumulation, since α-Toc rescued from Era/BV6-triggered ROS production, lipid peroxidation and cell death. Moreover, inhibition of lipid peroxide formation through the addition of Fer-1 or by overexpressing GPX4 failed to rescue from Era/BV6-triggered cell death, even if Era/BV6-stimulated lipid peroxidation was diminished. Likewise, Fer-1 protected from RSL3/BV6-, but not from Era/BV6-generated ROS production, leading to the assumption that other ROS besides lipid-based ROS contributed to cell death in Era/BV6-treated cells. In summary, while RSL3/BV6 induced ferroptosis in ALL, Era/BV6 stimulated a ROS dependent cell death, which was neither dependent on iron nor caspases or receptor-interacting protein (RIP) kinase 1 nor 3. Additionally, using Erastin alone did not trigger ferroptotic cell death in ALL. Finally, with these two studies we tried to unravel the molecular pathway of ferroptosis by using RSL3 and Erastin as well described ferroptosis stimulators. Here, we demonstrate the possibility of a novel treatment strategy to reactivate programmed cell death by impeding redox homeostasis in ALL.
Since ALL failed to induce ferroptosis upon Erastin treatment, we investigated in the third part of this thesis a new model system to induce ferroptosis upon Erastin and RSL3 exposure. Previous studies revealed that rhabdomyosarcoma (RMS) cells might be susceptible to oxidative stress-induced compounds. To this end, we used Erastin as a prototypic ferroptosis stimulus and GSH-depleting agent and demonstrated that GSH depletion, ROS and lipid ROS accumulation contributed to cell death. Additionally, Fer-1, Lip-1, DFO, lipophilic vitamin E derivate α-Toc and GSH, a cofactor of GPX4, protected from Erastin stimulated ROS accumulation, lipid peroxidation and cell death. Also, the use of a broad spectrum protein kinase C (PKC) inhibitor Bisindolylmaleimide I (Bim1), a PKCα and ß selective inhibitor Gö6976 and siRNA-mediated knockdown of PKCα suppressed Erastin-mediated cell death in RMS. Moreover broad spectrum nicotinamide-adenine dinucleotide phosphate (NADPH) oxidase (NOX) inhibitor Diphenyleneiodonium (DPI) and a more selective NOX1/4 isoform inhibitor GKT137831 abrogated Erastin-generated ROS formation, lipid peroxidation and cell death. With this, we demonstrate that RMS are vulnerable to ferroptotic cell death and investigated the molecular mechanism of ferroptosis by unravelling that PKC and NOX could have a pivotal role in ROS-mediated ferroptosis signalling in RMS. In this regard, ferroptosis inducers may act as a possible novel treatment strategy for RMS, especially those with poor clinical outcome.
Der 2‘-Desoxyguanosin-Riboschalter gehört zur unter Bakterien weit verbreiteten Klasse der Purin-Riboschalter. Allerdings wurden 2‘-Desoxyguanosin-bindende Riboschalter bisher ausschließlich in M. florum gefunden, damit stellt diese RNA eine Ausnahme unter den ansonsten verbreiteten Purin-Riboschaltern dar. In der vorliegenden Arbeit wurde ein NMR-Strukturmodell des IA-Aptamer-2‘-Desoxyguanosinkomplexes erstellt und anhand der mittels NMRSpektroskopie zugänglichen strukturellen Informationen sowohl Struktur und Dynamik des freien RNA-Aptamers als auch des 2‘-Desoxyguanosinkomplexes charakterisiert. Dabei wurde insbesondere der Einfluss von Mg2+ auf Struktur und Dynamik der jeweiligen Zustände sowie auf den durch 2‘-Desoxyguanosin induzierten Faltungsprozess untersucht.
Mg2+-Ionen modulieren die Faltungstrajektorien von sensorischen RNA-Domänen. Die Übertragbarkeit von Mg2+-abhängigen Charakteristika der RNA-Faltung innerhalb verschiedener Messmethoden ist durch die schlechte Vergleichbarkeit der relativen Konzentrationsverhältnisse eingeschränkt. Die NMR-spektroskopisch beobachtbaren Mg2+-Einflüsse sollten also unter besonderer Berücksichtigung der für NMR benötigten vergleichsweise sehr hohen RNAKonzentrationen mit Ergebnissen aus kalorimetrischen oder fluoreszenzspektroskopischen Messungen interpretiert werden. Die in der NMR-Spektroskopie üblichen hohen Probenkonzentrationen befinden sich in dem Regime, in dem auch der physikalische Effekt des verdrängten Volumens eine Rolle zu spielen beginnt. Demnach ist es für die RNA-Moleküle im NMR-Probenröhrchen bei Konzentrationen von 5-10 mg/ml auch ohne Zugabe von Mg2+ entropisch günstiger, kompakte Konformationen einzunehmen. Die Relevanz des Effekts des verdrängten Volumens für die RNA-Faltung unter NMR-Bedingungen und unter zellulären Bedingungen ist Gegenstand der aktuellen Forschung und wird in dieser Arbeit am Beispiel des IA-Aptamers diskutiert.
Der oft einzigartige Bindungsmodus ubiquitärer Metaboliten durch bakterielle Riboschalter (Montange and Batey, 2006) ermöglicht prinzipiell den Einsatz von RNA-Aptameren in vivo, ohne mit zellulären Proteinsystemen zu interferieren (Mulhbacher et al., 2010). Therapeutische Ziele sind beispielsweise die Anwendung von Riboschaltern gegen bakterielle Pathogene beziehungsweise gegen pathogene Bakterien selbst. Eine weitere Rolle wird RiboschalterElementen zukünftig als Bausteine in der synthetischen Biologie zukommen (Dixon et al., 2010; Knight, 2003; Topp and Gallivan, 2008). Hierfür ist es von grundlegender Bedeutung, Charakterisierung von Struktur als Basis für das Verständnis von Funktion unter zellulären Bedingungen zu etablieren. Im Rahmen einer Zusammenarbeit mit Robert Hänsel aus dem Arbeitskreis von Prof. Dr. Volker Doetsch wurde am Beispiel des IA-Aptamers und einer nichtnatürlichen Sequenzvariante gezeigt, dass eine strukturelle Charakterisierung von Riboschaltern mittels in cell NMR-Spektroskopie möglich ist. In Zusammenarbeit mit Karl von Laer aus der Arbeitsgruppe von Prof. Dr. Beatrix Suess wurden beide RNA-Aptamer hinsichtlich ihrer Funktion in einem biologischen Assay getestet. Die Ergebnisse dieser Experimente zeigten eine deutliche Korrelation von Struktur und Funktion in vivo, während Diskrepanzen zwischen Struktur in vitro und Funktion in vivo demonstriert werden.
Weiterhin wurde im Rahmen dieser Arbeit gezeigt, dass eine gewisse strukturelle Flexibilität der Bindungstaschen regulatorischer RNA-Motive für Selektion und Adaption während Evolution nötig ist. Beispielsweise wurde für den Guanin-Riboschalter gezeigt, dass der nicht-native Ligand 2‘-Desoxyguanosin zur Komplexbildung des Aptamers führt. Demnach könnte die Bindung von 2‘-Desoxyguanosin im Guanin-Riboschalter bereits evolutionär angelegt sein und die Entstehung des IA-Aptamers nach Genomreduktion der Mesoplasmen begünstigt haben. Das IA-Aptamer dagegen bindet Guanin nicht, stattdessen besitzt M. florum auf Guanin spezialisierte Sequenzvarianten dieses Riboschalters (Kim et al., 2007). Strukturell hochauflösende Einblicke in unterschiedliche Zustände der Bindungstasche im G-Aptamer-Thioguaninkomplex, die durch die Lösung der Kristallstruktur des GLoop-Aptamers ermöglicht wurden, unterstützen die Hypothese einer anpassungsfähigen Bindungstasche im G-Aptamer. Für B. subtilis wäre es interessant, die physiologische Bedeutung der Komplexbildung des G-Aptamers mit 2‘-Desoxyguanosin zu untersuchen.
G-protein coupled receptors (GPCRs) are a predominant class of cell-surface receptors in eukaryotic life. They are responsible for the perception of a broad range of ligands and involved in a multitude of physiological functions. GPCRs are therefore of crucial interest for biological and pharmaceutical research. Molecular analysis and functional characterisation of GPCRs is frequently hampered by challenges in efficient large-scale production, non-destructive purification and long-term stability. Cell-free protein synthesis (CFPS) provides new production platforms for GPCRs by extracting the protein synthesis machinery of the cell in an open system that allows target-oriented modulations of the synthesis process and direct access to the nascent polypeptide chain. CFPS is fast, reliable and highly adaptable. Unfortunately, highly productive cell-free synthesis of GPCRs is often opposed by low product quality. This thesis was aimed to adapt and improve some of the new possibilities for the cell-free production of GPCRs in high yield and quality for structural and pharmaceutical analysis. An E. coli based CFPS system was applied to synthesise various turkey and human Beta-adrenergic receptor (Beta1AR) derivatives as well as human Endothelin receptors type A and B (ETA and ETB) constructs. Both receptor families are important drug targets and pharmacologically addressed in the treatment of several cardiovascular diseases. CF-synthesis was mainly performed in presence of nanodiscs (ND), which are reconstituted high density lipoprotein particles forming discoidal bilayer patches with a diameter varyring from 6 to approx. 15 nm. The supplementation of ND in the CF-synthesis reaction caused the co-translational solubilisation of the freshly synthesised GPCRs. The fraction of the solubilised GPCR that was correctly folded was analysed by the competence to bind its ligand alprenolol or Endothelin-1, respectively. Both the solubilisation efficiency and the ability to fold in a ligand binding competent state was strongly affected by the lipid composition of the supplied ND. Best results were generally achieved with lipids having phosphoglycerol headgroups and unsaturated fatty acid chains with 18 carbon atoms. Furthermore, thermostabilisation by introduction of point mutations had a large positive impact on the folding efficiency of both Beta1AR and ETB receptor. Formation of a conserved disulphide bridge in the extracellular region was additionally found to be crucial for the function of the ETB receptor. Disulphide bridge formation could be enhanced by applying a glutathione-based redox system in the CFPS. Further improvements in the quality of ETB receptor could be made by the enrichment of heat-shock chaperones in the CF-reaction. Depending on the receptor type and DNA-template, roughly 10 – 30 nmol (350 – 1500 µg) of protein could be synthesised in 1 ml of CF-reaction mixture. After the applied optimisation steps, the fractions of correctly folded receptor could be improved by several orders of magnitude and were finally in between 35% for the thermostabilised turkey Beta1AR, 9% for the thermostabilised ETB receptor, 6.5% for the non-stabilised ETB receptor, 1 - 5% for non-stabilised turkey Beta1AR and for human Beta1AR isoforms and 0.1% for ETA receptor. Therefore, between 2 and 120 µg of GPCR could be synthesised in a ligand binding competent form, depending on the receptor and its modifications. Correctly folded turkey Beta1AR and ETB receptors were thermostable at 30°C and could be stored at 4°C for several weeks after purification. Yields of the thermostabilised turkey Beta1AR were sufficient to purify the receptor in a two-step process by ligand-binding chromatography to obtain pure and correctly folded receptor in the lipid bilayer of a ND. Furthermore, a lipid dependent ligand screen could be demonstrated with the turkey Beta1AR and significant alterations in binding affinities to currently in-use pharmaceuticals were found. The established protocols are therefore suitable and highly competetive for a variety of applications such as screening of GPCR ligands, analysis of lipid effects on GPCR function or for the systematical biochemical characterisation of GPCRs. Most promising for future approaches appears to address the suspected bottlenecks of intial insertion of the GPCR-polypeptide chain in the ND bilayer and the thermal stability of the receptors. Nevertheless, the estabilised protocols for the analysed targets in this thesis are already highly competitive to previously published production protocols either in cell-based or cell-free systems with regard to yield of functional protein, speediness and costs. Moreover, the direct accessibility and other general characteristics of cell-free synthesis open a large variety of possible applications and this work can therefore contribute to the molecular characterisation of this important receptor type and to the development of new pharmaceuticals.
Die Biosynthese der Fettsäuren (FS) ist in Eukaryoten und Bakterien ein hochkonserviert zentraler Stoffwechselweg, der in zwei strukturell verschiedenen Systemen ausgeführt wird. Die meisten Bakterien, Parasiten, Pflanzen und Mitochondrien nutzen ein Fettsäuresesynthase Typ-II (FAS-II) System. Bei FAS II Systemen sind alle katalytischen Domänen separate lösliche Proteine. In Eukaryoten wie auch den Bakterien Corynebakteria, Mycobakteria, Nocardia (Klasse der CMN Bakterien) liegen die katalytischen Domänen fusioniert auf einer Polypeptidkette vor, die zu einem Multienzymkomplex der Fettsäuresynthase Typ I (FAS-I) assemblieren. Die Architektur der FAS-I zeigt große Unterschiede; die X förmige Säuger-FAS-I (Maier et al., 2006), sowie die fassartigen Enzyme der Pilz FAS-I (Jenni et al., 2007; Leibundgut et al., 2007; Lomakin et al., 2007; Johansson et al., 2008) und der bakteriellen FAS-I (Boehringer et al., 2013; Ciccarelli et al., 2013). Zwischen Pilz- und bakterieller FAS-I gibt es trotz des ähnlichen Aufbaus bedeutende Unterschiede. Mycobakterium tuberculosis, der Auslöser von Tuberkulose (TB), an der jährlich über eine Million Menschen weltweit sterben (WHO, 2014), synthetisiert durch eine Symbiose von FAS-I, FAS-II und der Polyketidsynthase-13 Mykolsäuren. Durch die Mykolsäuren ist M. tuberculosis resistent gegen äußere Einflüsse. FAS-I ist in die Synthese der Vorstufen der Mykolsäuren involviert. Sie stellt im Kampf gegen TB ein potentielles Inhibierungstarget dar.
Strukturell war die bakterielle FAS-I beim Beginn der vorliegenden Arbeit, nur durch negative-stain-Elektronenmikroskopie (EM) Aufnahmen aus dem Jahr 1982 charakterisiert (Morishima et al., 1982). In dieser Arbeit konnte die bakteriellen FAS I aus M. tuberculosis (MtFAS), sowie Corynebacterium ammoniagenes (CaFAS) und Corynebacterium efficiens (CeFAS) strukturell untersucht werden. Dies geschah mit den Methoden negative-stain-EM, Einzelmolekül-Cryo-EM (Cryo-EM), Cryo EM Tomographie (CET) und Röntgenkristallographie.
Anhand von CeFAS-Kristallen konnte erstmals durch Röntgenkristallographie die Struktur einer bakteriellen FAS-I bestimmt werden. Zudem wurde die hohe konformationelle Flexibilität der bakteriellen FAS-I mit mehreren Methoden gezeigt. Für die CaFAS konnte mit Cryo-EM initiale Prozesse der Proteinkristallbildung abgebildet werden.
Die 5-Lipoxygenase (5-LOX) stellt den Startpunkt des Leukotrienstoffwechsels dar, da sie Arachidonsäure (AA) über die 5(S)-Hydroperoxy-6-trans-8,11,14-cis-eicosatetraensäure (5-HpETE) in Leukotrien A4 (LTA4) umwandelt. 5-HpETE kann zum korrespondierenden Alkohol 5(S)-Hydroxy-6-trans-8,11,14-cis-eicosatetraensäure (5-HETE) reduziert werden. LTA4 dient als Zwischenprodukt für die Synthese von LTB4 und den Cysteinyl-gebundenden LTs LTC4, LTD4 und LTE4. LTs nehmen eine wichtige Funktion in der Immunabwehr ein, sind jedoch auch an einer Vielzahl von Krankheitsgeschehen wie z. B. Asthma bronchiale, Atherosklerose und einiger Tumorarten beteiligt. Die 5-LOX teilt sich in zwei Domänen auf: der reglatorischen, N-terminalen Domäne und der katalytischen, C-terminalen Domäne. Ihre Aktivität unterliegt einer komplexen allosterischen Regulation und kinetischen Besonderheiten wie einer Substratinhibition. In vielen Fällen ist die regulatorische PLAT-(Polycystin-1, Lipoxygenase, alpha-Toxin)-Domäne involviert. Sie ist essentiell an der Bindung von Calcium, Membranen und weiterer Faktoren wie dem Coactosin-like protein (CLP) und Dicer beteiligt. Auch eine zweite Bindungsstelle für das Substrat oder einen seiner Metaboliten wird dort vermutet. Letztlich bleibt jedoch die Regulation der 5-LOX-Aktivität durch die PLAT-Domäne unzureichend geklärt. Diese Tatsache und die fortwährende Suche nach neuen Ansatzpunkten für die 5-LOX-Inhibition bilden den Hintergrund, vor dem diese Arbeit angefertigt wurde.
Das Ziel lag in der Entwicklung einer stabilen, isolierten PLAT-Domäne und deren Charakterisierung. Es stellte sich jedoch heraus, dass sich die isolierte Domäne durch eine hohe thermische Instabilität und starke Aggregationsneigung auszeichnet. Mittels Mutationsstudien auf Basis der 5-LOX AS 1-115, verbunden mit Gelfiltrationsläufen zur Analyse der Proteinaggregation, wurde schließlich ein Konstrukt entwickelt, das in Konzentrationen < 0,5 mg/ml als Monomer vorlag: die sogenannte PLAT1-115 W75G. Ein Austausch des W75 in Glycin erhöhte ebenfalls die thermische Stabilität, so dass Versuche bei 20°C durchgeführt werden konnten. Zunächst wurden jedoch die grundlegenden Eigenschaften der Mutante untersucht. Dies umfasste die Beantwortung der Frage, ob auch die PLAT1-115 W75G Calcium bindet, sowie die Aufnahme eines Circulardichroismus-(CD)-Spektrums. Der erste Aspekt konnte mit mehreren Methoden bestätigt werden. Eine Calciumzugabe zum Laufpuffer 20 mM MOPS, 50 mM KCl pH 7,4 erhöhte konzentrationsabhängig das Elutionsvolumen der PLAT1-115 W75G auf der analytischen Gelfiltrationssäule – vermutlich durch den bekannten Einfluss von Calcium auf die Hydrophobizität der PLAT-Domäne. Zusätzlich wurde die Interaktion durch differential scanning fluorimetry (DSF) und Oberflächen-Plasmonen-Resonanz-Spektroskopie (SPR) nachgewiesen. Allerdings gelang aus verschiedenen Gründen keine Quantifizierung der Bindungsaffinität. Das CD-Spektrum bestätigte die Struktur der PLAT-Domäne als sogenanntes all-beta_protein und ermöglichte die Einordnung der PLAT1-115 W75G in die Gruppe der betaII-Proteine.
Ein weiterer Fokus dieser Arbeit lag auf der vermuteten allosterischen Fettsäurebindungsstelle in der PLAT-Domäne. Es wurde versucht, die Interaktion mittels SPR nachzuweisen. Zur Vorbereitung wurde im 5-LOX-Aktivitätstest und im DSF an der isolierten Domäne ein Detergens bestimmt, das einen möglichst geringen Einfluss auf das Protein ausübt. Dabei zeigte Octyl-beta-D-glucopyranosid (beta-OG) das vorteilhafteste Profil. Auf dieser Basis wurde die kritische Mizellbildungs-Konzentration (CMC) der AA und einiger HETEs in beta-OG-haltigen Puffern bestimmt. Die SPR-Studien ergaben jedoch keine reproduzierbaren Ergebnisse. In einem weiteren Schritt wurden die Substrathemmung des Gesamtproteins 5-LOX und der Einfluss von Calcium charakterisiert. Sowohl in Gegenwart von ~ 1 mM freiem Calcium als auch von 1 mM EDTA lag mit 20 µM AA die höchste Produktbildung nach 10-minütiger Reaktion vor. Das Detergens Tween20 (T20) hob in einer Konzentration unter seiner CMC (0,001 % m/V) in Anwesenheit von Calcium die Inhibition auf. Ohne Calcium zeigte sich auch in Gegenwart von T20 die bekannte Substratinhibition der 5-LOX einschließlich ihrer Maximalaktivität bei 20 µM AA. Diese Ergebnisse deuten darauf hin, dass Calcium eine Bindung der 5-LOX an eventuell vorhandene, negativ geladene Vesikel aus AA und Detergens vermitteln und dadurch die Substratinhibition aufheben kann. In Fällen, in denen die Substratinhibition vor dem Erreichen der AA-CMC auftritt, hat Calcium folglich keinen Einfluss.
Zuletzt wurde die Interaktion der PLAT1–115 W75G mit CLP und einem C-terminalen Fragment von Dicer untersucht. Im Crosslinking ließ sich nicht auf eine Interaktion der isolierten PLAT-Domäne mit CLP schließen. Dagegen ergaben Diamid-Crosslinking-Studien, dass die isolierte PLAT-Domäne in der Lage ist, das Dicer-Fragment zu binden. Dieses Ergebnis wurde im SPR bestätigt.
Identifizierung des vertebraten-spezifischen Proteins C7orf43 als neue TRAPPII Komplexuntereinheit
(2016)
Bei den transport protein particle (TRAPP) Komplexen handelt es sich um eine Familie von Protein Komplexen, die jeweils aus mehreren Untereinheiten bestehen. In der vorliegenden Arbeit konnte das Protein C7orf43 als neue potenzielle TRAPPII Untereinheit identifiziert werden, die - wie auch die beiden anderen TRAPPII-spezifischen Komponenten TRAPPC9 und TRAPPC10 - sowohl für die Erhaltung von ERGIC, Golgi-Apparat und COPI Vesikel als für den ER zu Golgi Transportweg benötigt wird.
Nichtribosomale Peptid Synthetasen sind Quelle für eine Vielzahl an Sekundärmetaboliten mit antibiotischer Wirkung. Jede Synthetase besteht aus einer Abfolge von Modulen, wobei jedes Modul die nötigen Domänen für den Einbau eines Bausteins in das gebildeten Peptids enthält. Ein Ansatz zur Gewinnung neuer Peptidantibiotika, die angesichts der steigenden Zahl multiresistenter Keime dringend benötigt werden, ist der Austausch von Domänen oder Modulen. Aufgrund bisher noch nicht verstandener Selektivitäten, entweder zwischen den Domänen oder zwischen einzelnen Domänen und Zwischenstufen des gebildeten Peptids, führt dieser Ansatz jedoch in der Praxis oft zu keiner oder nur geringer Ausbeute.
Ziel der vorgelegten Arbeit war es, einige dieser Selektivitäten zu untersuchen, wobei der Fokus auf Peptidyl Carrier Proteinen Domänen (PCPs) lag. An diese Domänen sind alle Intermediate während der Reifung des Peptids kovalent über einen Phosphopantethein-Kofaktor (Ppan-Arm) gebunden.
Im ersten Teil der Arbeit sollte die Struktur einer mit einem Heptapeptid beladenen PCP mittels Lösungs-Kernspinresonanzspektroskopie (NMR) bestimmt werden. Hierbei konnte die natürliche Verknüpfung zwischen Ppan-Arm und Peptid über einen Thioester nicht verwendet werden, da diese Bindung zu Hydrolyse-anfällig war. Es konnte jedoch gezeigt werden, dass die Substitution des Thioesters durch eine nicht hydrolysierbare Amidbindung keinen Einfluss auf die Struktur hat, wodurch die Strukturbestimmung möglich war. Hierbei zeigte sich, dass die Peptid-beladene PCP in der sogenannten A/H state Konformation vorliegt, wobei das an sie gebundene Peptid frei beweglich ist. Somit scheint es wahrscheinlich, dass die PCP keine Selektivität für das an sie gebundene Peptid aufweist. Dies ist ein Unterschied zu den strukturell ähnlichen Acyl Carrier Proteinen (ACPs) aus der bakteriellen Fettsäurebiosynthese, da diese eine Bindungstasche für die an sie gebundenen Fettsäuren ausbilden.
Untersuchungen der Selektivität der Kondensationsdomäne (C Domäne) für das PCP gebundene Peptid mittels NMR-Titrationen und biochemischer Analysen konnten nicht durchgeführt werden, da sich im Laufe des Projekts zeigte, dass die aus der Synthetase herausgetrennte C Domäne katalytisch nicht aktiv war. Stattdessen sollte die Kristallstruktur einer Peptid-beladenen PCP-C Bidomäne, für welche eine katalytische Aktivität bereits gezeigt worden war, gelöst werden. Da aber bereits ein signifikanter Anteil der Bidomäne während der Expression mit dem Ppan-Arm beladen wurde, war die nötige quantitative Beladung mit dem Peptid gekoppelten Ppan-Arm in vitro nicht möglich. Eine quantitative Modifizierung mit dem Ppan-Arm in vitro war hingegen erfolgreich, und die Struktur der Ppan-beladenen Bidomäne konnte gelöst werden. Aufgrund des großen Abstands zwischen den aktiven Zentren der beiden Domänen kann es sich bei der beobachteten Orientierung nicht um jene handeln, die die beiden Domänen zueinander annehmen, wenn die C Domäne das PCP-gebundene Peptid bindet.
Im zweiten Teil der Arbeit wurde die Modifizierung einer PCP durch eine Gruppe II Phosphopantetheintransferase (PPT) untersucht. PPTs katalysieren die Übertragung des Ppan Arms auf die Seitenkette eines in PCPs konservierten Serins. In dieser Magnesium-abhängigen Reaktion dient Coenzym A (CoA) als Quelle für den Ppan-Arm. Durch Mutation des konservierten Serins in der PCP zu Alanin konnte ein stabiler Komplex aus PCP und PPT in Anwesenheit von CoA und Magnesium kristallisiert und seine Struktur bestimmt werden.
In einem Strukturmodell für den PCP/PPT Komplex war eine andere Konformation für die PCP postuliert worden, als sie in der Kristallstruktur des Komplexes zu beobachten ist. Durch Strukturbestimmung der PCP mittels Lösungs-NMR und anschließender Titrationsexperimente konnte jedoch gezeigt werden, dass sowohl die freie als auch die komplexierte PCP in Lösung ebenfalls die in der Kristallstruktur beobachtete Konformation einnehmen.
Aufgrund der gelösten Kristallstruktur konnten zwei Bereiche identifiziert werden, in denen die beiden Proteine im Komplex in direktem Kontakt zueinander stehen. Der eine Bereich ist durch eine intermolekulare Wasserstoffbrücke, der andere durch hydrophobe Wechselwirkungen zwischen den Proteinen gekennzeichnet. Durch ortsspezifische Mutagenese konnten beide Wechselwirkungen gestört werden, was sich in einer Abnahme der Komplexstabilität und einer veränderten Geschwindigkeit der Übertragung des Ppan-Arms äußerte.
Die große strukturelle Ähnlichkeit zwischen dem in dieser Arbeit untersuchten Komplex aus zwei in Bacillus vorkommenden Proteinen und einem humanen ACP/PPT Komplex legt die Vermutung nahe, dass die beobachteten Wechselwirkungen in vielen Organismen konserviert sind.
Ribosomes are the central cellular assembly lines for protein synthesis. To cope with the translational needs, a proliferating mammalian cell can produce up to 7500-ribosomes per minute. However, under growth limiting conditions, such as nutrient depletion, ribosome synthesis is rapidly shut down exemplifying the importance of a tight coordination between ribosome supply and cellular energy status. In addition to the quantitative regulation, a strict quality control of ribosome synthesis is equally important, because alterations in the composition or function of ribosomes can lead to a variety of pathologies. To cope with these challenges a highly regulated, multi-step pathway of ribosome biogenesis has evolved. In mammals this pathway generates the mature 80S ribosomes that comprise the large 60S and the small 40S subunits. Together they contain around 80 ribosomal proteins and the 28S, 18S, 5.8S and 5S rRNAs. The 28S, 5.8S and 5S rRNAs are assembled into the large subunit, while the 18S rRNA is part of the small subunit. The pathway of ribosome biogenesis is a multi-step cellular process, where specific stages occur in distinct subcellular compartments. Transcription of the 47S rRNA, which is the precursor for the 28S, 18S and 5.8S species, occurs in the nucleolus. Modification of distinct bases and early processing of this precursor also take place in the nucleolus. Subsequently, the 40S and 60S pre-ribosomes take separate maturation routes through the nucleoplasm before their export and final assembly in the cytoplasm. The various stages of preribosomal maturation require the constant and sequential action of a large number of non-ribosomal proteins, known as trans-acting factors. These factors coordinate the delicate remodeling of the pre-ribosomal intermediates and thereby ensure proper progression of the maturation process. The remodeling events largely depend on the dynamics of post-translational modifications, such as phosphorylation or SUMOylation. This requires that the enzymes controlling these modifications are properly targeted to their sites of activity as they fulfill their functions within specific compartments. Here we studied the regulatory principles that govern the subcellular partitioning of the SUMO-specific isopeptidase SENP3 and its associated factor PELP1. Previous work from our laboratory has delineated the importance of the SUMO system for proper ribosome biogenesis in mammalian cells. In particular, we have shown that SENP3 is critically involved in 28S rRNA formation, which is a key step for pre-60S subunit maturation. A critical involvement of SENP3 at this stage of the maturation process is in agreement with the observed enrichment of SENP3 in the nucleolus, since 28S rRNA processing is considered to occur in the nucleolus. Our subsequent work identified the nucleolar scaffold protein NPM1 and the ribosomal trans-acting factor PELP1 as bona fide substrates of SENP3. For both proteins we could demonstrate modification by SUMO2/3 and define SENP3 as the demodifying enzyme. Depletion of SENP3 enhanced the conjugation of SUMO to both proteins and concomitantly reduced conversion of the 32S pre-rRNA to the mature 28S rRNA. PELP1 is part of a larger protein complex consisting of the core components PELP1, TEX10 and WDR18. We could show that the balanced SUMOylation/deSUMOylation of PELP1 controls the nucleolar/nucleoplasmic distribution of this complex. Enhanced SUMOylation, which is observed in the absence of SENP3, triggers the nucleolar release of the complex suggesting that SENP3-mediated deSUMOylation controls the dynamics of nucleolar trans-acting factors. Based on these findings we first wanted to understand, in which cellular compartment(s) SENP3 exerts its function on 28S maturation. Next, we wanted to tackle the question how the subcellular distribution of SENP3 is controlled. Finally
we addressed the question how the SUMOylation of PELP1 determines the subnuclear distribution of the PELP1 complex. This work initially revealed that the nucleolar localization of SENP3 is crucial for proper 28S rRNA formation and 60S ribosome maturation. Importantly, we could demonstrate that the nucleolar compartmentalization of SENP3 depends on its direct physical interaction with NPM1. Further, we could show that the amino-terminal region of SENP3 is necessary for its binding to NPM1 and nucleolar recruitment. Strikingly, this interaction requires the phosphorylation of SENP3, which is brought about by the mTOR kinase. By in-vitro kinase assays and mass-spectrometric approaches we identified five serine/threonine residues within the amino-terminal region of SENP3 that are targeted by mTOR (S/T 25, 26, 141, 142, 143). We could further demonstrate by mutagenesis that these sites in SENP3 are in fact critical for the phospho-dependent binding of SENP3 to NPM1 and its nucleolar recruitment.
Consistent with these data, we found that chemical inhibitors of the mTOR kinase trigger the nucleolar release of SENP3 and impair its interaction with NPM1. Strikingly, this goes along with severe 28S rRNA maturation defects demonstrating the physiological importance of mTOR signaling in the regulation SENP3 function and rRNA processing. By specifically depleting components of the either mTORC1 or mTORC2, we could attribute the observed effects to signaling by mTORC1 rather than mTORC2. In an attempt to find the negative regulators of SENP3 phosphorylation, we identified PP1-γ as the candidate phosphatase in this pathway. We found a strong physical interaction of SENP3 with PP1-γ and observed a loss of SENP3 nucleolar localization upon ectopic expression of PP1-γ. Thus we could define mTOR/PP1-γ mediated phosphorylation/dephosphorylation of SENP3 as an important
mechanism in the control of ribosome maturation. Given that mTOR activity is controlled by nutrient availability, SENP3 functions as a sensor that couples ribosome synthesis with nutrient availability. The second part of this work delineated the role of SUMOylated PELP1 in nucleoplasmic partitioning of the SENP3-PELP1 complex. It was revealed that the AAA-ATPase MDN1 binds preferentially to SUMO modified PELP1 and likely segregates SUMOylated PELP1 from nucleolar pre-60S particles. We initially found that the PELP1 complex associates with MDN1, a factor known to be involved in the 28S rRNA maturation. Notably, depletion of MDN1 led to an enhanced accumulation of the PELP1 complex in the nucleolus and a strong association of PELP1 with pre-60S particles, suggesting that MDN1 is required for the release of this complex from the pre-ribosomes. Intriguingly, the interaction of PELP1 with MDN1 requires SUMO2/3 and SUMOylated PELP1 shows enhanced binding to MDN1 when compared to unmodified PELP1. Taken together this work provides new insights in the control of the SENP3-PELP1 complex dynamics. We could define several layers for the coordinated spatial regulation of SENP3 and the PELP1 complex. This work therefore underscores the crucial importance of dynamic post-translational modifications for the control of ribosome maturation.
Ubiquitin is a highly conserved protein involved in several cellular processes like protein degradation, endocytosis, signal transduction and DNA repair. The discovery of ubiquitin-like proteins (UBL) and ubiquitin-like domains (ULD) increases the number of regulation pathways where the property of the ubiquitin-fold is profitable.
Autophagy is the catabolic pathway used in cells to deliver cytosolic components and dysfunctional organelles to the lysosome for degradation. MAP1LC3 proteins are ubiquitin-like proteins involved in one hand for the expansion of the autophagosome, which sequesters cytosolic substrates. In the other hand, these proteins (LC3- and GABARAP- subfamilies) bind to autophagic receptors linked to polyubiquitinated proteins aggregates. For this project, the 3D structure of the GABARAPL-1/NBR1-LIR complex was determined and confirmed that GABARAPL-1 belongs to the MAP1LC3 proteins family, structurally characterized by an ubiquitin-fold, consisting of a central beta-sheet formed by four beta-strands and two alpha-helices on one side of the beta-sheet, preceded N terminally by two alpha-helices, resulting in the formation of two hydrophobic pockets, hp1 and hp2. The autophagic receptor NBR1 interacts with GABARAPL-1 through the hp1 and hp2 with its LIR motif taking an extended beta conformation upon binding, forming an intermolecular beta-sheet with the second beta-strand of GABARAPL 1. This LC3- interacting region (LIR) consists of an Theta XX Gamma sequence preceded by acidic amino acids, with Theta and Gamma represented by any aromatic and hydrophobic residues, respectively. Interaction studies of the LIR domains of p62, Nix and NBR1 with different members of the MAP1LC3 proteins family indicate that the presence of a tryptophan in the LIR motif increases the binding affinity. Substitution to other aromatic amino acids or increasing the number of negatively charged residues at the N-terminus of the LIR motif, however, has little effect on the binding affinity due to enthalpy-entropy compensation, suggesting that effector proteins can interact with a wide variety of different sequences with similar and moderate binding affinities.
Additionally to be present in proteins dealing with protein folding and degradation, ubiquitin-like domain were found protein involved in the regulation of signal transduction like TBK1, a serine/threonine kinase responsible for induction of immune response. In this second project, based on the NMR chemical shifts of the TBK1 domain contained between amino acids 302 and 383, secondary structure prediction programs (TALOS and CSI) confirmed the presence of an Ubiquitin-like domain in TBK1 by identifying one alpha-helix and four beta-strands sequentially aligned like following beta-beta-alpha-beta-beta. This alignment corresponds perfectly with the secondary structure elements of Ubiquitin and proved that TBK1_ULD belongs to the UBL protein superfamily. The similarity to ubiquitin was even bigger by the presence in addition of a small beta-strand and a short helix, which are observed as the beta 5-strand and a 310-helix in Ubiquitin, respectively. The first attempts on the 3D structure determination confirmed the Ub-fold but due to the lack of assignment in TBK1_ULD, only a structure based on ubiquitin as a model was determined. Interaction studies of TBK1_ULD with the IAD-SRR domain of IRF3 showed that both side of the molecule seems involved and that the TBK1/IRF3 interaction is more complex than a one to one binding process. Unfortunately, the instability of TBK1_ULD associated to the difficulty in the purification of IAD-SRR did not allow to further study this interaction more precisely.
Finally, to overcome the difficulty encountered in NMR experiments because of low expression and/or poor solubility, an expression vector using the intrinsic property of ubiquitin was designed. Fused to proteins or peptides targets, this construct produced proteins and peptides in a larger amount than with traditional expression vectors and also with a less cost than chemical synthesis for pure labeled peptides for NMR structural studies. The presence of a hexa histidine tag was useful for the isolation and the purification of the constructs. The existence of a TEV cleavage site was created to keep the possibility of releasing the ubiquitin moiety from the expressed protein or peptide. Moreover, the ubiquitin-tag could also still be attached to the protein/peptide of interest when biophysical methods like NMR, ITC or CD spectroscopy are applied, providing the same results than for the protein/peptide moiety alone.
Since Inhibitor of Apoptosis (IAP) proteins are frequently dysregulated in different cancer entities and contribute to apoptosis resistance, pharmacological IAP antagonists are considered to be promising agents for the future development of cancer treatment strategies. IAP antagonists are small-molecule drugs that have been designed to mimic the interaction site of IAP proteins with their endogenous inhibitor Second mitochondrial activator of caspases (SMAC). Thus, they are frequently referred to as SMAC mimetics. Treatment with SMAC mimetics engages an apoptotic program in cancers by affecting different components of the apoptotic machinery. Besides disinhibition of caspases, SMAC mimetics trigger non-canonical nuclear factor-κB (NF-κB) signaling, which induces upregulation of tumor necrosis factor (TNF) α and other NF-κB target genes. In particular, TNFα production has been closely linked to the induction of SMAC mimetic-mediated cell death. The TNFα-dependent para/autocrine loop facilitates the formation of a cytosolic complex consisting of caspase-8, Fas-associated death domain (FADD) and Receptor-interacting protein (RIP) 1, which serves as caspase-8 activation platform and ultimately triggers induction of apoptosis. In the present study, we use the small-molecule bivalent SMAC mimetic BV6 to analyze SMAC-stimulated NF-κB signaling in cancer cell lines of different entities. Interestingly, we identify two novel NF-κB-regulated factors that are both required for SMAC mimetic-induced apoptosis in a context-dependent manner. First, we show that NF-κB-dependent upregulation of death receptor 5 (DR5) can serve as an alternative mechanism of BV6-mediated cell death. We demonstrate that BV6 treatment induces NF-κB-dependent but largely TNFα -independent apoptosis in A172 glioblastoma cells. By using an unbiased whole genome expression analysis approach, we identify DR5 as a critical NF-κB target gene, which substitutes TNFα and is indispensable for BV6-initated cell death in A172 cells. Second, we demonstrate that Interferon regulatory factor (IRF) 1 is required for BV6-induced TNFα production and apoptosis. Our study provides evidence that IRF1 closely cooperates with the NF-κB network in BV6-mediated cell death and additionally alters expression of selective SMAC mimetic-induced target genes. Furthermore, we show that BV6 treatment triggers secretion of a set of proinflammatory cytokines and increases attraction of monocytes to BV6-treated tumor cells in an IRF1-dependent manner. In summary, our work supports the notion that NF-κB-regulated factors are critically required for SMAC mimetic-initiated apoptosis. We show that IRF1 is indispensable for TNFα production and cell death in BV6-sensitive cell lines and that also DR5 can serve as a proapoptotic NF-κB-controlled factor in BV6-induced apoptosis besides TNFα. Furthermore, this study contributes to an improved understanding on non-apoptotic functions of SMAC mimetics, as IRF1 additionally influences expression levels of proinflammatory cytokines and attraction of immune cells. Thus, our work provides novel insights into the regulation of SMAC mimetic-induced signaling events, which is crucial for the translation of SMAC mimetics for use in clinical application.
Structural biology often employs a combination of experimental and computational approaches to unravel the structure-function paradigm of biological macromolecules. This thesis aims to approach this combination by the application of Pulsed Electron-Electron Double Resonance (PELDOR/DEER) spectroscopy and structural modelling. In this respect, PELDOR spectroscopy in combination with site-directed spin labelling (SDSL) of proteins is frequently used to gain distance restraints in the range from 1.8 to 8 nm. The inter-spin distance and the flexibility of the spin labelled protein domains are encoded in the oscillation and the dampening of the PELDOR signal. The intrinsic flexibility of the commonly used MTSSL (1-Oxyl-2,2,5,5-tetramethylpyrroline-3-methyl) spin label itself can be an obstacle for structural modelling if the flexibility of the label is large compared to the flexibility of the protein domains. In this thesis the investigation of two multi-domain proteins by the 4-pulse PELDOR sequence is presented. At first, the N-terminal polypeptide transport-associated (POTRA) domains of anaOmp85, a rigid three domain protein, giving well-defined PELDOR distance restraints, is investigated. The experimental restraints are used for structure refinement of the X-ray structure and reveal a strong impact of the intrinsic flexibility of MTSSL on the accuracy of structural refinement. The second example, K48-linked diubiquitin, is a highly flexible multi-domain protein on which the flexibility of MTSSL is of minor impact on structural modelling. In this case, the distance restraints are utilized to determine conformational ensembles. Due to the high intrinsic flexibility already characterizing diubiquitin the recently developed 7-pulse Carr-Purcell (CP) PELDOR sequence was applied to investigate longer ubiquitin chains. This sequence enables to measure dipolar oscillations with an extended time window, allowing a good separation between inter- and intramolecular contributions even for long distance and broad conformational distributions, thereby providing an increased accuracy of the obtained distance distributions.
The human endothelin receptors, ETA and ETB, are two members of the G-protein coupled receptors family (GPCRs) and they are key players in cardiovascular regulation. The characterization of their functionality in vitro has been limited by the possibility to obtain high quality samples using conventional expression systems. The Cell-Free expression system is an alternative technique for the production of membrane protein as well as GPCRs and can overcome some of the limitations that are commonly encountered using an in vivo approach. Cell-Free expression protocols for the two receptors ETA and ETB have been optimized by implementing post- and co-translational association to lipid bilayers. The efficiency of the reconstitution or association to liposomes and nanodiscs has systematically been studied and the ligand binding properties of the two receptors have been analyzed using a set of different complementary techniques. In several different conditions a high affinity binding of the peptide ligand ET-1 to both endothelin receptors could be obtained and the highest activity values were detected in sample prepared using a co-translational approach in presence of nanodiscs. Furthermore, the characteristic differential binding pattern of selected agonists and antagonists to the two receptors was confirmed. In samples obtained from several Cell-Free expression conditions, two intrinsic properties of the functionally folded ETB receptor, such as the proteolytic processing based on conformational recognition as well as the formation of SDS-resistant complexes with the peptide ligand ET-1, were detected. ETA and ETB are able to induce in vivo the activation of hetrotrimeric G proteins upon stimulation with an agonist, leading to the dissociation of the heterotrimeric complex and the exchange of GDP to GTP in the Galpha subunit. The Cell-Free expression system was chosen for the production of two G alpha subunit, Galpha s and Galpha q. Soluble expression of the two proteins was achieved and the production of active Galpha s was confirmed using fluorescent as well as radioactive assays. In conclusion, the obtained results document a new process for the production of ligand binding competent endothelin receptors, as well as Galpha proteins, using a Cell-Free expression system. The combination of this expression system and the nanodiscs technology appears to be a promising tool for the further characterization of membrane proteins as well as GPCRs.
Cancer cells, in general and especially Rhabdomyosarcoma (RMS) cells have been reported to be highly susceptible to oxidative stress. Based on this knowledge we examined whether the inhibition of the two main antioxidant defense pathways, i.e. the thioredoxin (TRX) and the glutathione (GSH) system, represents a possible new strategy to induce cell death in RMS. To do so, we combined the -glutamylcysteine synthetase (γGCL) inhibitor buthionine sulfoximine (BSO) or the cystine/glutamate antiporter (xc-) inhibitor erastin (ERA), both GSH depleting enzymes, with the thioredoxinreductase (TrxR) inhibitor auranofin (AUR) to evaluate synergistic cell death in the alveolar RMS (ARMS) cell line RH30 and the embryonal RMS (ERMS) cells RD.
Furthermore, we tried to unravel the underlying molecular mechanisms of AUR/BSO or AUR/ERA treatment in RMS cells. Thereby we showed that AUR/BSO as well as AUR/ERA treatment leads to proteasome inhibition characterized by the accumulation of ubiquitinated proteins, which is in agreement with the already published ability of AUR to inhibit proteasomeassociated deubiquitinases (DUBs) aside from TrxR. As a consequence, the protein levels of ubiquitinated short-lived proteins, like NOXA and MCL-1, increase upon treatment with AUR/BSO or AUR/ERA. Consistently, we could detect an increased binding of NOXA to MCL-1. Interestingly, not only NOXA protein levels but also mRNA levels rise upon treatment, pointing to a transcriptional regulation of pro-apoptotic NOXA through AUR/BSO or AUR/ERA combination treatment. The fact that siRNA mediated knockdown of NOXA rescues cells from combination treatment-induced cell death strengthens the role of NOXA as an important regulator of cell death induction. Apart from proteasome inhibition and subsequent NOXA accumulation, AUR cooperates with BSO or ERA to trigger BAX/BAK activation, which is needed for cell death induction, too. Additionally, loss of mitochondrial membrane potential (MMP) as well as caspase activation and PARP cleavage is detected after treatment of RMS cells with AUR/BSO or AUR/ERA.
Except of apoptotic cell death we also detected features of iron-dependent ferroptosis after treatment with AUR/BSO or AUR/ERA. This is not surprising, since BSO and ERA already have been described to induce ferroptotic cell death. Although lipid peroxidation takes place in both cell lines, only in RH30 cells, cell death seems to be partially ferroptosis-dependent, since especially in this cell line AUR/BSO- or AUR/ERA-induced cell death can be rescued with different ferroptosis inhibitors.
Although both combination treatments, AUR/BSO as well as AUR/ERA, induce production of reactive oxygen species (ROS), only the thiol-containing ROS scavengers GSH and its precursor N-acetylcysteine (NAC), but not the non-thiolcontaining antioxidant α-Tocopherol (α-Toc), consistently prevent proteasome inhibition, NOXA accumulation and cell death.
Additionally, we demonstrated that BSO and ERA abolish AUR-mediated upregulation of GSH thereby releasing the AUR cytotoxic effect on RMS cells, in line with the described ability of cysteines to inhibit the function of AUR. Together, this points to the conclusion that GSH depletion, rather than an increase in ROS levels, is important for AUR/BSO- or AUR/ERA-induced cell death.
In conclusion, through revealing that the antitumor activity of AUR is enhanced in combination with GSH depleting agents, we identified redox homeostasis as a new and promising target for the treatment of RMS cells.
In this thesis, molecular dynamics (MD) simulations are used to study the interaction of different proteins with lipid bilayers. MD simulations can be used as a “computational microscope” to gain atomistic insights into the interactions between proteins and lipids that can barely be accessed in such detail by experimental methods. The different chapters of this thesis address the lipid sensing functionality of amphipathic helices (AHs) when bound to membranes, the folding of AHs at lipid-water interfaces as well as the conformational dynamics of the HIV-1 Env glycoproteins in viral-like and experimental bilayers. In the last chapter the possibilities to enhance the performance of MD simulations are explored, leading to a more efficient usage of computational resources.
In Reaktion auf zellulären Stress wie etwa Schädigungen der DNA oder die vermehrte Aktivität von Onkogenen aktivieren vorgeschaltete Signalkaskaden den Transkriptionsfaktor (TF) p53. Dieser kann über die Aktivierung der Expression von Zielgenen wiederum die Zellteilung stoppen, die Reparatur von DNA Schäden initiieren oder in schweren Fällen die Eliminierung der Zelle durch Apoptose einleiten. Ist p53 durch Mutationen deaktiviert, können sich entartete somatische Zellen vermehren und in der Folge Krebs entstehen.
In Wirbeltieren finden sich neben p53 mit p63 und p73 zwei weitere TFs, welche während der Evolution aus dem gleichen gemeinsamen Vorläufer durch Genduplikationen hervorgegangen sind. Die drei TFs sind modular aufgebaut und alle Isoformen verfügen jeweils minimal über eine DNA Bindungsdomäne (DBD) und eine Tetramerisierungsdomäne (TD). Werden die p53 ähnlichen TFs aktiviert, lagern sie sich über die TD vermittelt zu Tetrameren zusammen, wodurch ihre DBDs kooperativ an DNA Sequenzmotive binden können. Die DBD ist auch über große phylogenetische Abstände hinweg hoch konserviert, wodurch bereits gezeigt werden konnte, dass auch primitive vielzellige Tiere bereits Homologe dieser TF Familie besitzen. Im Vergleich zur DBD variiert die Proteinsequenz der TD deutlich stärker, was andeutet, dass deren Struktur im Laufe der Evolution erhebliche Veränderungen durchlaufen hat. Diese Veränderungen aufzuklären ist das übergeordnete Forschungsvorhaben zu dem diese Dissertationsschrift beiträgt.
Ciona intestinalis (C.int.) ist eine Spezies aus dem Unterstamm der Manteltiere. Diese sind die engsten lebenden Verwandten der Wirbeltiere und C.int. ist ein populärer Modelorganismus für die Erforschung der Embryonalentwicklung. Sein Genom kodiert für zwei p53 ähnliche TFs, welche mit p53/p73-a und p53/p73-b bezeichnet werden. Die Struktur ihrer TDs wurde im Rahmen der vorliegenden Arbeit mittels Kernspinresonanz (NMR) Spektroskopie untersucht.
Die TD von menschlichem p53 (hp53) ist ein Dimer aus Dimeren. Jedes Monomer formt einen beta-Strang und eine alpha-Helix. Im primären Dimer lagern diese sich so zusammen, dass ein beta-Faltblatt entsteht und die alpha-Helices mit entgegen gesetzter Orientierung der Länge nach aneinander packen. Zwei dieser Dimer lagern sich dann so zum Tetramer zusammen, dass zwischen pol-ständigen beta-Faltblättern ein Bündel aus vier Helices entsteht. Dieses Motiv ist auch in den TDs der Ciona Proteine hochkonserviert und wird im Folgenden als Kern?TD bezeichnet. In den TDs von menschlichem p63 und p73 (hp63 und hp73) verfügt jedes Monomer an seinem C-terminus noch über eine zweite Helix. Die zweiten Helices eines jeden Dimers greifen wie Klammern um das jeweils andere primäre Dimer und stabilisieren so das Tetramer. Entscheidend für die stabile Anbindung an die Kern?TD ist dabei ein charakteristisches Tyrosin-Arginin (YR) Motiv in der zweiten Helix, welches sich auch in der Sequenz der TD von C.int. p53/p73-a wiederfindet. Analysen der Sekundärstruktur auf Basis von NMR Experimenten ergaben jedoch, dass die TD von C.int. p53/p73-a bei 25°C keine zweite Helix ausbildet. Mit Hilfe von chimären TD Peptiden, in denen Teile der Ciona Sequenz gegen die entsprechenden Abschnitte von hp73 ausgetauscht wurden, konnte gezeigt werden, dass die Kern TD von C.int. p53/p73-a fähig ist eine zweite Helix zu stabilisieren und hierfür neben dem YR Motiv auch der Sequenzabschnitt zwischen erster und zweiter Helix entscheidend ist. Stabilisierende Substitutionen in diesem Bereich bewirkten ebenso wie ein Absenken der Temperatur die Ausbildung einer zweiten Helix, welche jedoch im Gegensatz zu jener in hp73 nur transient faltet und auch nicht essentiell für die Bildung des Tetramers ist, wohl aber dessen Stabilität erhöht.
Spezifisch in der Entwicklungslinie von Ciona kam es dazu, dass eine, für eine entsprechende Vorläuferversion von C.int. p53/p73-a kodierende, mRNA spontan zurück in DNA übersetzt und ins Genom eingefügt wurde. Die durch diese Retrotransposition erzeugte neue Genkopie C.int. p53/p73-b muss demnach ursprünglich einmal für die gleiche Proteinsequenz kodiert haben, innerhalb der TD finden sich konservierte Reste jedoch nur im Bereich der Kern TD.
Von der TD von C.int. p53/p73-b wurde die molekulare Struktur in freier Lösung mittels NMR ermittelt. Diese zeigte, dass interessanterweise in der TD von C.int. p53/p73-b jedes Monomer am C-terminus eine stabil gefaltete, zweite Helix besitzt. Obwohl diese zweite Helix sich aus einer Sequenz faltet, die keinerlei Sequenzhomologie zu homologen Proteinen aus Wirbeltieren aufweist, lagert sie sich in einer Position auf die Kern TD, welche der in hp73 sehr nahe kommt. Da die primären Dimere der Kern TD aber anders als in hp63 und hp73 durch Salzbrücken miteinander verbunden sind, ist die zweite Helix jedoch nicht essentiell, um das Tetramer zu stabilisieren. Vermutlich kommt der zweiten Helix von C.int. p53/p73-b vielmehr u.a. die Aufgabe zu die Bildung von Heterotetrameren aus C.int. p53/p73-a und –b zu unterbinden.
Zusammengenommen zeigen die Ergebnisse, dass die Architektur der TD mit zweiter Helix bereits der Prototyp für die TDs aller p53 ähnlichen Proteine der Wirbel- und Manteltiere war und die als eine Art Klammer das Tetramer stabilisierende zweite Helix sich nicht erst während der Evolution der Wirbeltiere entwickelt hat.
Caspase-2 is the evolutionary most conserved member of the caspase family and was shown to be involved in genotoxic stress induced apoptosis, control of aneuploidy, and ageing related metabolic changes. However, its role in apoptosis seems redundant due to the observation, that knockout does not inhibit apoptotic signalling exclusively. Instead, knockout of caspase-2 leads to tumor susceptibility in vivo, which led to the assumption, that caspase-2 has non-apoptotic functions and can act as a tumor suppressor. The underlying mechanism of the tumor suppressor activity of caspase-2 has not been clarified so far. Furthermore, caspase-2, has a prominent, and as pro-enzyme exclusive localisation in the nucleus and other subcellular compartments, implicating a distinct and location specific role.
In this study, a novel caspase-2 specific substrate, termed p54nrb, was identified. P54nrb is harbouring a caspase-2 specific cleavage site at the aspartate residue D422, and cleavage of p54nrb leads apparently to disruption of its putative DNA binding domain at the C-terminus.
P54nrb is a nuclear multifunctional RNA and DNA binding protein, known for roles in transcriptional regulation, DNA unwinding and repair, RNA splicing, and retention of defective RNA. Overexpression of p54nrb has been observed in several human cancers, such as cervix carcinoma, melanoma, and colon carcinoma.
Data from this study revealed, that depletion of p54nrb in tumor cell lines results in a loss of resistance to drug induced cell death and to reduced capability of anchorage independent growth, which is functionally equivalent to a reduced tumorigenic potential. Meanwhile, p54nrb depletion alone is not cytotoxic.
The investigation of p54nrb dependent gene regulations by high resolution quantitative proteomics uncovered an altering expression of multiple tumorigenic genes. For two of these candidates, the tumorigenic protease cathepsin-Z and the anti-apoptotic gelsolin, p54nrb dependent expression was detected universally in all three investigated tumor cell lines, cervix carcinoma, melanoma, and colon carcinoma. Additionally, a direct interaction of p54nrb with the cathepsin Z and gelsolin encoding DNA, but not with their corresponding mRNA, could be demonstrated.
Conjointly, this study unveils a novel mechanistic feature of caspase-2 as a tumor suppressor. The caspase-2—p54nrb axis can orchestrate the levels of several tumorigenic proteins and thereby determine the cell death susceptibility and long-term tumor survival. These findings might be of great value for future therapeutic interventions and for overcoming drug resistance of tumors.
The transcription factor p63 is part of the p53 protein family, which consists of three members, p53, p63 and p73. P63 shares structural similarity with all family members, but is associated to different biological functions than p53 or p73. While p53 is mainly linked to tumor suppression and p73 is connected with neuronal development, p63 has been connected to critical biological roles within ectodermal development and skin stem cell biology as well as supervision of the genetic stability of oocytes. Due to its gene structure p63 is expressed as at least six different isoforms, three of them containing a N-terminal transactivation domain. The isoforms that are of biological relevance both have a C-terminal inhibitory domain that negatively regulates the transcriptional activity. This inhibitory domain is supposed to contain two individual components of which one is internally binding and masking the transactivation domain while the other one can be sumoylated. To further investigate this domain a mutational analysis with the help of transactivation assays in SAOS2 cells was carried out to identify the critical amino acids within the inhibitory domain and the impact on transcriptional activity of TAp63alpha, the p63-isoform which is essential for the integrity of the female germline. The results of these experiments show that a stretch of approximately 13 amino acids seems to be important for the regulation of transcriptional activity in TAp63alpha, due to the increased transcriptional activity occurring in this region after mutation. Additional experiments showed that this mechanism is distinct from sumoylation, which seems to have only implications for the intracellular level of TAp63alpha. As a conclusion, the C-terminus of the Tap63alpha is essential for two different mechanisms, which control the transcriptional activity of the protein. Both regulatory elements are independent from each other and can now be restricted to certain amino acids. Activation of the wild type protein might take place in the identified region via post-translational modification. Furthermore an inhibition assay was carried out to test if the same region might have implications on the second biological relevant isoform deltaNp63alpha. The results show that the same amino acids which show an impact on transcriptional activity in Tap63alpha lead to a significant change in functional behaviour of deltaNp63alpha. There is a possibility that both proteins are regulated with opposite effects via the same mechanisms, based at the C-terminus of the p63alpha-isoforms. In both cases a modification of these residues could lead to a more opened conformation of the protein with consequences on promoter binding, which can be even important for deltaNp63alpha with respect to promoter squelching. Both alpha-isoforms seem to be regulated via the C-terminus and to elucidate if that is also the case for TAp63gamma a deletion analysis was carried out. The results show that there are also amino acids within the C-terminus of TAp63gamma, which have implications on the transcriptional activity of the protein. Therefore the C-terminus seems to play a major role for regulation of diverse p63 isoforms.
The multistep-processes leading to the formation of tumors have been extensively studied in the past decades, leading to the identification of “hallmarks of cancer”. They are characteristic changes in biological processes that discriminate tumor cells from healthy cells. Increasing knowledge on the molecular structures associated with tumorigenesis allowed their specific inhibition in targeted anti-cancer therapy. However, successful targeted anti-cancer therapy is only available for a limited subset of diseases, so the continuous investigation of tumorigenic mechanisms is required to tackle the immense diversity of neoplastic entities.
AVEN and FUSE binding protein 1 (FUBP1) display the ability to regulate apoptosis and cell cycle progression. Thus, the proteins are associated with hallmarks of cancer (resisting cell death and uncontrolled proliferation). Indeed, aberrant expression of AVEN and FUBP1 could be demonstrated in multiple cancers. In contrast, there is only little knowledge on the physiological function of AVEN and FUBP1. The lack of knowledge results in part from the embryonic lethality of the homozygous knockout of Aven and Fubp1 in mouse models, limiting the gain of information by analyzing these animals.
In this study, I generated conditional Aven and Fubp1 knockout mice to investigate their physiological function.
By analyzing reporter mice expressing β-galactosidase under the control of the endogenous Aven promoter, I identified Aven promoter activity to be both tissue- and cell type-specific and dependent on the developmental stage. Detecting apoptotic cell death by immunohistochemistry did not reveal increased apoptosis in Aven knockout mice, suggesting a functional role of AVEN besides apoptosis inhibition during embryogenesis.
Basing on the significant Aven promoter activity detected in the adult brain and in the mammary gland, I generated and characterized conditional Aven knockout mice with Aven deletion restricted to cells within the brain or the mammary gland. AVEN depletion in these tissues was not embryonic lethal and the affected tissues displayed a normal histology.
Since aberrant Aven expression had been associated with hematologic malignancies, I also analyzed mice with an Aven knockout in the hematopoietic system. Depletion of AVEN in the blood cells had no effect on hematopoietic stem and progenitor cell frequencies. Consequently, AVEN seems to be dispensable for the maintenance and differentiation of stem, progenitor and mature blood cells, at least as far as the expression of particular differentiation markers was concerned.
As loss of AVEN in the analyzed tissues did not affect the viability of mice and did not produce any other obvious phenotype, the exact role of AVEN that is essential for embryo survival remains to be identified.
To study the oncogenic potential of AVEN, I investigated the role of AVEN in a mouse model for breast carcinogenesis. While AVEN expression seemed to be increased in breast tumors, tumor onset and progression were not altered in mice with depleted AVEN expression in the mammary gland. Consistently, Aven knockout tumor cells were neither less proliferative nor more prone to undergo apoptosis than Aven wildtype tumor cells. Cell culture experiments demonstrated that AVEN expression is upregulated by estrogen. Knockdown of AVEN in the breast cancer cell line MCF-7 slightly increased UV irradiation-induced apoptosis and accelerated metabolism. So while AVEN does not promote development or progression of breast tumors, enhanced AVEN expression in ER+ breast cancers might contribute to chemotherapy resistance.
To study the physiological role of FUBP1, I generated a conditional Fubp1 knockout mouse model. While the insertion of loxP sites into the Fubp1 locus was occasionally embryonic lethal, some mice with a cell type-specific deletion of Fubp1 in hematopoietic cells or EPO receptor expressing cells were born alive. In these mice, frequencies of hematopoietic stem and progenitor cells as well as erythrocytes were unaltered. These results conflict with previous publications. However, compensating mechanisms might be responsible for the discrepancies between the observed phenotypes and reported FUBP1 function.
In cell culture studies, I could demonstrate that the previously reported upstream regulation of FUBP1 by TAL1 depended on an intact GATA motif in the FUBP1 promoter and that binding of GATA1 to the FUBP1 promoter increased during erythropoiesis.
To identify new FUBP1 target genes with relevance for erythropoiesis, I performed differential gene expression analysis in cells with wildtype and depleted FUBP1 expression. RNA-sequencing and PCR-arrays revealed only moderate differences in the expression of genes that are components of the EPO receptor signaling pathway as well as genes associated with apoptosis and proliferation of hematopoietic cells. By regulating the transcription of these genes, FUBP1 could contribute to efficient erythropoiesis.
The detailed mechanism of the 20 S proteasome from Thermoplasma acidophilum is unknown. Substrates are degraded processively to small fragments without the release of intermediates, but the basis for this unique degradation mode remains obscure. The proteasome is a molecular machine, but how the different nanocompartments interplay and whether more than one substrate can be treated simultaneously has not been elucidated yet. To address these questions we had to disable the functionality of one aperture in order to dissect whether the other pore can compensate for the loss. As it is challenging to introduce mutations solely around one pore aperture of the highly symmetrical construct, we chose a novel approach by unique orientation of the proteasome at interfaces. For this purpose we purified recombinant 20 S proteasomes, where hexahistidine tags were fused either around the entrances or at the sides. According to electron microscopic studies we immobilized these constructs uniformly either end-on or side-on at metal-chelating interfaces (lipid vesicles, lipid monolayers and self-assembled thiol monolayers). Degradation of small fluorogenic peptides and large proteins like casein was analyzed. Small substrates were degraded with comparable activity by free and immobilized proteasomes, irrespective of their orientation. Thus it can be assumed that peptides can pass the sealed entrance of the 'dead-end' proteasome. However, larger substrates like fluorescently labeled casein were processed near the temperature optimum by side-on immobilized and soluble proteasomes with threefold activity compared to end-on immobilized proteasomes. Hence it can be concluded that one pore is sufficient for substrate entry and product release. In other words, the pore and antechamber can fulfil a triple function in the import and unwinding of substrates and the egress of products. With means of surface plasmon resonance the exact substrate/proteasome stoichiometry could be determined to ~1 for 'dead-end' proteasomes and ~2 for side-on immobilized (active and inactive) proteasomes. Most importantly, a fit with the Hill equation revealed positive cooperativity for side-on immobilized (Hill coefficient ~2) in contrast to end-on immobilized proteasomes (Hill coefficient ~1). Thus in case of soluble proteasomes two substrates bind presumably in opposite antechambers with positive cooperativity. The off-rate of casein as substrate is twofold for the active side-on immobilized proteasome in comparison to the end-on immobilized proteasome. The exact 2:1 stoichiometry of the off-rates equals the ratio of exit pathways amenable in case of side-on orientated versus 'dead-end' immobilized proteasomes. Thus crevices along the cylindrical body of the 20 S proteasome seem not to participate in the egress of small products. An inactive proteasome mutant displays a concentration-dependent off-kinetic against casein. Accordingly, the off-rate of the bisubstrate:proteasome complex can be attributed around half the value of the monosubstrate:proteasome complex. Consequently, substrates exit the inactive proteasome via the route of access due to obstruction of the trans side with an entering substrate. Hence the active proteasomes have to chop substrates down to small fragments prior to release through both pores. Thus the processive degradation mode might result from positive binding cooperativity. The on-rate constants for casein suggested that substrate association represents a two-step process comprising a rate-limiting translocation step and a fast binding step. As fluorescence cross-correlation revealed that two substrates can be co-localized in the proteasome and bind successively with increasing affinity (KD,1 = 8 µM versus KD,2 = 700 nM), an allosteric transition in the proteasome can be assumed. Combining our results with the data from other research groups led to a mechanistic model for the 20 S proteasome. Accordingly, the first substrate undergoes a slow translocation step, binds in the antechamber and diffuses subsequently to the catalytic centers, where it is degraded. By switching on the catalytic activity, the pores at both termini are dilated via conformational changes. Hence entry of the second substrate into the proteasome is facilitated due to omission of the rate-determining translocation step. The second substrate is either accommodated in the antechamber before it is processed (alternating degradation) or, most probably, is directly threaded into the central cavity (simultaneous degradation). As effusing peptides compete with entering proteins for binding in the antechamber, the pores are kept in an open state. After finishing digestion the pores are closed and a new degradation cycle can be reinitiated. In summary, substrate association with the proteasome underlies an ordered alternating binding mechanism in contrast to the random mode of degradation. Thus the two-stroke engine offers the advantage of speeding up degradation without enhancing complexity.
Evaluierung der zellfreien Produktion sekundär aktiver Transporter für die Proteinkristallisation
(2013)
Resistance in glucocorticoid-induced apoptosis is associated with poor prognosis for long term survival in childhood acute lymphoblastic leukemia (ALL). As Smac mimetics have been shown to reactivate apoptosis by antagonizing Inhibitor of Apoptosis (IAP) proteins, we investigate the potential of the Smac mimetic BV6 to overcome glucocorticoid-resistance in ALL. This study shows that BV6 synergistically cooperates with glucocorticoids to trigger apoptosis and to suppress clonogenic growth of pediatric ALL cells. Of note, the BV6/glucocorticoid combination treatment also induces cell death in cells having defects in the apoptotic signaling cascade by inducing a switch from apoptotic to necroptotic cell death. The clinical relevance of our novel combination treatment is underscored by parallel experiments in primary pediatric ALL samples, in which glucocorticoids and BV6 act together to induce cell death in a synergistic manner. Importantly, the addition of BV6 enhances the anti-leukemic effects of glucocorticoids in an in vivo mouse model of pediatric ALL without causing substantial side effects, highlighting the potency of a BV6/glucocorticoid combination treatment. In contrast, BV6 does not increase cytotoxicity of glucocorticoids against several non-malignant cell types of the lympho-hematopoietic system. Furthermore, we have identified the novel underlying mechanism of BV6/glucocorticoid-induced apoptosis by showing that BV6 and glucocorticoids synergistically act together to promote assembly of the ripoptosome, a RIP1/FADD/caspase-8-containing cell death complex. Ripoptosome assembly is critically required for BV6/Dexamethasone-induced cell death, since genetic silencing of its members, i.e. RIP1, reduces ROS production, caspase activation and most importantly cell death induction. BV6/glucocorticoid combination treatment promotes ripoptosome assembly by inhibition of both of its negative regulators, IAP proteins and cFLIP. Thus, we identify that BV6 and glucocorticoids cooperate together to reduce cIAP1, cIAP2 and XIAP protein levels and cFLIP expression. Ripoptosome formation occurs independently of autocrine/paracrine loops of death receptor ligands, since blocking antibodies for TNFα, TRAIL or CD95L or genetic silencing of their corresponding receptors fail to rescue BV6/glucocorticoid-induced cell death. In summary, this study shows that the Smac mimetic BV6 sensitizes for glucocorticoid-induced apoptosis by promoting ripoptosome assembly with important implications for the treatment of childhood ALL.
Small molecule inhibitors sensitize neuroblastoma cells for chemotherapeutic drug-induced apoptosis
(2015)
Neuroblastoma (NB) is one of the most common solid extracranial pediatric tumors, deriving from undifferentiated cells of the peripheral nervous system. It accounts for approximately 10% of all childhood cancers. High stage tumors usually show poor prognosis despite aggressive treatment such as radiotherapy or chemotherapy. Therefore, it is of utmost importance to find novel treatment strategies in order to improve existing chemotherapy protocols. Combination treatment offers advantages, as chemotherapeutic drugs can be applied in low and subtoxic doses, reducing possible side-effects. Here, we report in a two-part study that small molecule inhibitors (SMI), namely BI 2536, a PLK1 inhibitor and BV6, a SMAC mimetic (SM), sensitize neuroblastoma cells for chemotherapeutic drug-induced cell death. By using i) BI 2536 in combination with vinca alkaloids and ii) BV6 in combination with either doxorubicin or vinca alkaloids, we show that cell death is synergistically enhanced compared to monotherapy. Furthermore, combination treatment significantly reduces survival of NB cells in long-term assays, compared to single treatment. We identify that vinca alkaloid/SMI combinations induce mitotic arrest, as shown by phosphorylation of histone H3, which results in the induction of intrinsic apoptosis and inhibition of CDK1 by RO-3306 could abolish these findings. Mechanistically, upon vinca alkaloid/SMI-induced mitotic arrest, anti-apoptotic BCL-2 proteins such as MCL-1, BCL-2 or BCL-XL are degraded or inactivated by phosphorylation, which induces the activation of the proapoptotic BCL-2 family proteins BAX and BAK. The importance of the mitochondrial apoptosis pathway in vinca alkaloid/SMI-induced cell death was further highlighted by the fact that ectopic expression of BCL-2 inhibits vinca alkaloid/SMI-induced DNA fragmentation and BAK- and caspase-activation. In contrast to the vinca alkaloid/SMI cotreatment, DOX/SMI (DOX/BV6)-induced apoptosis only partially involves the mitochondrial pathway. Instead, we clarify that RIP1 is required for DOX/BV6-induced apoptosis, as pharmacological and genetic inhibition of RIP1 rescues from apoptosis induction. Although it has been shown in previous studies that SM-treatment (e.g. BV6) can induce the NF-κB pathway and auto-/paracrine TNFα production through cIAP1/2 depletion, DOX/BV6-induced apoptosis is completely independent of NF-κB activation in our setting, despite fast cIAP1 depletion. This conclusion is based on the fact that inhibition of the NF-κB pathway by exogenously expressed dominant-negative IκBα as well as application of a TNFα blocking antibody does not reduce DOX/BV6-induced cell death. In summary, we unravel two new promising treatment strategies for neuroblastoma patients by using a combination treatment of two different small molecule inhibitors, combined with well-characterized chemotherapeutic agents. Furthermore we give detailed insights into cell death pathways induced by these combination treatments, in which mitochondria and RIP1 have a differential role in chemotherapeutic drug-induced apoptosis.
Structural characterization of stressosome complexes by single-particle cryo-electron microscopy
(2015)
The stressosome is a Mega Dalton macromolecular complex involved in stress adaptation in bacteria. Stressosomes are considered as stress signaling hubs. They are able to perceive a variety of different stress stimuli and transduce them into one single cellular answer, which is the initialization of a transcriptional up-regulation of hundreds of different genes encoding for universal but also very specific stress response proteins.
The stressosome of Bacillus subtilis became a prime example for this intriguing stress-triggered transcriptional regulation when its architecture was determined by Single-particle cryo-electron microscopy (cryo-EM) in 2008. In Gram-positive Bacillus species, the stressosome complex senses changes in salt concentration, ethanol content, blue-light, heat or acid stress contributing to the general stress response by activation of the alternative σB factor. σB is a transcriptional promoter that initiates the transcription of over 150 general stress genes, e.g., genes that encode osmolyte transporters to counteract osmotic and chill stress. The B. subtilis stressosome (stressosome_Bc) is composed of multiple copies of the 3 proteins: RsbR, RsbS and RsbT. These three Rsb proteins (Regulator of Sigma B) are found clustered in one operon forming the conserved RST module. RsbS and RsbR are scaffold proteins comprising a STAS domain, respectively. Because these domains are dominantly associated to sulfate transporters and anti-sigma antagonist they were named STAS domains, however, they were also identified in other sensor proteins. In the stressosome they form the internal ball-shaped core, while the N-terminal globin-fold sensor domain of RsbR, protruding to the outside, facilitates stress sensing. It is assumed that the stress signal is transduced to the stressosome core via the STAS domain resulting in conformational changes of the core. These changes affect the binding of the third protein, RsbT, a serin-threonine kinase. As a direct consequence of stress sensing the RsbT kinase is released from the complex to start an activation cascade involving the stepwise activation of RsbU, V, W, and X, which are all part of the same operon, and finally of σB. In Bacillus species, several RsbR orthologs were identified varying mainly in the sequence of the N-terminal sensor domains. It is assumed that the stressosome_Bc assembles with a still unknown combination of RsbR orthologs allowing for the broad spectrum of stress stimuli that can be processed in vivo. The pathogenic bacteria Listeria monocytogenes is a close relative of Bacillus. Its potent stress response allows Listeria to survive the harsh environmental conditions during host infection and therefore the stress regulation machinery is contributing heavily to the virulence of this pathogen. In Listeria the Rsb operon is conserved and highly homologous to the Bacillus one. In the frame of this thesis, the in vitro assembly of Listeria innocua stressosomes was shown for the first time by Single-particle (SP) negative stain EM. Moreover, binding of Listeria RsbT to the assembled RsbR-RsbS complex was demonstrated biochemically.
Despite the conservation of the RST-module the entire Rsb operon is not conserved in the bacterial kingdom suggesting that signal transduction and regulation of gene expression might occur by very different mechanisms in stressosomes of different species. We have focused here on a stressosome type from the Gram-negative pathogen Vibrio vulnificus that is quite distinct from the Bacillus ones with respect to (1) the missing conservation of the Rsb operon, (2) the role of RsbT, (3) the activation of a different transcriptional promoter, and (4) the absence of additional RsbR orthologs. Interestingly, there is only one RsbR protein encoded in the genome. This one contains a Haem-group in its N-terminal domain being oxygen sensitive. It is assumed that the Vibrio stressosome perceive only oxidative stress and that regulation occurs via a diguanylate cyclase with a GAF domain that synthesizes the second messenger c-di-GMP from GTP.
We have started a structure determination of the Vibrio vulnificus stressosome by SP cryo-EM to elucidate the differences in the molecular mechanism of stress sensing in divers stressosome types. A 3D map of the oxidized (activated) Vibrio vulnificus stressosome was determined to 7.6 Å resolution revealing an increased flexibility of both the core and the N-terminal sensor domains in comparison to the Bacillus stressosome suggesting that our structure has trapped for the first time an active state of a stressosome complex. A 3D map of the stressosome core to 7 Å resolution allowed fitting of a homology model of the Vibrio stressosome based on the Bacillus stressosome as template. The conformational changes could be attributed to the entire core, which was confirmed by MD simulations.
Membrane proteins are biological macromolecules that are located in a cell’s membrane and are responsible for essential functions within an organism, which makes them to prominent drug targets. The extraction of membrane proteins from the hydrophobic membrane bilayer to determine high-resolution crystal structures is a difficult task and only 2% of all solved proteins structures are membrane proteins. Computational methods may help to gain deeper insights into membrane protein structures and their functions. This study will give an overview of such computational methods on a representative set of membrane proteins and will provide ideas for future computational and experimental research on membrane proteins.
In a first step (chapter 2), I updated an earlier, manually-curated data set of homologous membrane proteins (HOMEP) to more recent versions in 2010 (HOMEP2) and 2013 (HOMEP3) using an automated clustering approach. High-resolution structures of membrane proteins listed in the PDB_TM database were structurally aligned and subsequently clustered using structural similarity scores. Both data sets were used as a standard gold reference set for subsequent work.
Subsequently, I have updated and applied the sequence alignment program AlignMe to determine protein descriptors that are suitable for detecting evolutionary relationship between homologous a-helical membrane proteins. Single input descriptors were tested alone and in combination with each other in different modes of AlignMe by optimizing gap penalties on the HOMEP2 data set. Most accurate alignments and homology models on the HOMEP2 data set were observed when using position-specific substitution information (P), secondary structure propensities (S) and transmembrane propensities (T) in the AlignMe PST mode. An evaluation on an independent reference set of membrane protein sequence alignments from the BAliBASE collection showed that different modes of AlignMe are suitable for different sequence similarity levels. The AlignMe PST mode improved the alignment accuracy significantly for distantly related proteins, whereas for closely-related proteins from the BAliBASE set the AlignMe PS mode was more suitable. This work was published in March 2013 in PLOS ONE. In order to allow also an easier usage of the AlignMe program, I have implemented a web server of AlignMe (chapter 4) that provides the optimized settings and gap penalties for the AlignMe P, PS and PST modes. A comparison to other recent alignment web server shows that the alignments of AlignMe are similar or even more accurate than those of other methods, especially for very distantly related proteins for which the inclusion of membrane protein information has been shown to be suitable. This work was published in the NAR web server issue in July 2014.
Although membrane-specific information has been shown to be suitable for aligning distantly related membrane proteins on a sequence level, such information was not incorporated into structural alignment programs making it unclear which method is the most suitable for aligning membrane proteins. Thus, I compared 13 widely-used pairwise structural alignment methods on an updated reference set of homologous membrane protein structures (HOMEP3) and evaluated their accuracy by building models based on the underlying sequence alignments and used scoring functions (e.g., AL4 or CAD-score) to rate the model accuracy (chapter 5). The analysis showed that fragment-based approaches such as FR-TM-align are the most useful for aligning structures of membrane proteins that have undergone large conformational changes whereas rigid approaches were more suitable for proteins that were solved in the same or a similar state. However, no method showed a significant higher accuracy than any other. Additionally, all methods lack a measure to rate the reliability of the accuracy for a specific position within a structure alignment. In order to solve these problems, I propose a consensus-type approach that combines alignments from four different methods, namely FR-TM-align, DaliLite, MATT and FATCAT and assigns a confidence value to each position of the alignment that describes the agreement between the methods. This work has been published 2015 in the journal “PROTEINS: structure, function and bioinformatics”.
Consensus alignments were then generated for each pair of proteins of the HOMEP3 data set and subsequently analyzed for single evolutionary events within membrane spanning segments and for irregular structures (e.g., 310- and p-helices) (chapter 6). Interestingly, single insertions and deletions could be observed with the help of consensus alignments in the conserved membrane-spanning segments of membrane proteins in four protein families. The detection of such single InDels might help to identify crucial residues for a proteins function.
Throughout their life cells of eukaryotic organisms can be confronted with a variety of proteotoxic stresses and in order to survive, corresponding resistance mechanisms had to evolve. Proteotoxic stresses can cause misfolding of proteins and accumulation of toxic protein aggregates. Failure to remove aggregates of misfolded proteins compromises cellular function and can ultimately cause cell death and disease. To deal with this challenge, cells utilize a complex network of protein quality control pathways, including chaperones, the ubiquitin-proteasome system and the autophagy system.
Another mechanism to cope with proteotoxic stresses is the stalling of translation initiation in order to save valuable resources and prevent faulty translation. Upon stress, intrinsically disordered RNA-binding proteins such as TIA-1 or G3BP1/2 are recruited to stalled preinitiation complexes and a network of multivalent interactions between RNAs and proteins is formed. These mRNP networks can merge with each other and phase separate into membraneless liquid-like structures called stress granules (SGs). Once stress is released, SGs are quickly resolved and translation continues. Yet, chronic stress or mutations of SG-associated proteins can cause persistent SGs, which can sequester misfolded proteins and have been linked to neurodegenerative diseases such as amyotrophic lateral sclerosis or frontotemporal dementia.
In mammalian cells, three isoforms of the small ubiquitin-related modifier (SUMO), SUMO1, SUMO2 and SUMO3 are covalently attached to lysine residues of target proteins. SUMO conjugation is catalyzed via an enzymatic cascade of an heteromeric E1 activating enzyme, the E2 conjugating enzyme Ubc9 and in some cases one of a limited number of E3 SUMO ligases. SUMOylation is a dynamic modification and can be reversed by SUMO isopeptidases, the best characterized of which belong to the SENP family. Cellular stresses such as heat or oxidative stress strongly induce SUMOylation resulting in increased numbers of poly-SUMOylation (formation of SUMO2/3 chains) on nuclear proteins.
The SUMO-targeted ubiquitin ligase (STUbL) RNF4 harbors four SUMO interaction motifs in its N-terminal domain. This feature allows RNF4 to specifically bind poly-SUMOylated proteins and catalyze their proteolytic or non-proteolytic ubiquitylation.
A variety of substrate proteins have been shown to undergo SUMO-primed ubiquitylation by RNF4 in response to stress or DNA damage. RNF4-mediated ubiquitylation is often a signal for proteolytic degradation of these substrates.
In this work we aimed by identify novel RNF4 targets, in heat-stressed cells in order to gain a wider understanding of the nuclear proteotoxic stress response. Analysis by mass spectrometry revealed that a large fraction of RNF4-interacting proteins in heatstressed cells are nuclear RNA-binding proteins, many of which shuttle outside the nucleus and associate with SGs upon stress. We validated, that nuclear RNA-binding proteins, such as TDP-43 and hnRNP M are indeed heat-induced targets of SUMOprimed ubiquitylation by RNF4.
These initial results led us to further investigate the links between the SUMO/RNF4-mediated, nuclear protein quality control and the dynamics of cytosolic heat- or arsenite-induced SGs. SUMO2/3 and RNF4 are mainly nuclear proteins and we confirmed that they do not associate with SGs. Yet, we could demonstrate that depletion of SUMO2/3, the E3 SUMO ligase PML or RNF4 as well as chemical inhibition of SUMOylation strongly delayed SG clearance upon stress release, indicating that a functional STUbL pathway is essential for the timely clearance of SGs.
Next, we investigated how stress-induced poly-SUMOylation is regulated. Our data shows that SENP levels and activities are reduced in response to heat and arsenite stress, which allows the buildup of poly-SUMO chains on nuclear proteins. Limitation of poly-SUMOylation by overexpression of the SUMO chain-specific isopeptidases SENP6 and SENP7 induced SG formation. In contrast, poly-SUMO-priming by chemical depletion of SENP6 with the drug hinokiflavone drastically limited SG formation upon stress treatment. These results indicate a clear role of chain-specific SENPs in the regulation of stress-induced poly-SUMOylation and SG dynamics.
Last, we investigated whether the STUbL pathway could affect the phase separation of FUSP525 (an ALS-linked mutant of the SG-associated protein FUS) and observed that perturbations of the STUbL pathway lead to an increased phase separation of FUSP525L.
Thus, our work connects the SUMO/RNF4 protein quality control mechanism to the dynamics of SGs supporting the hypothesis that release of proteotoxic stress in the nucleus facilitates the clearance of cytosolic SGs. Thereby, we discovered a previously unknown link between the nuclear and cytosolic axis of proteotoxic stress response.
Necroptosis is a programmed cell death pathway that is implicated in a variety of human diseases. In recent years, increasing knowledge has been gained on the necroptotic signaling cascade. Nevertheless, the role of reactive oxygen species (ROS) in necroptosis is still ambiguous. In this study, we reveal that ROS critically regulate BV6/TNFα-induced necroptotic signaling in FADD-deficient Jurkat cells and in zVAD-treated MV4-11 cells. We show that several ROS scavengers such as butylated hydroxyanisole (BHA), N-acetylcysteine (NAC), α-tocopherol (αToc) and ethyl pyruvate (EP) significantly reduce ROS production and BV6/TNFα–induced cell death. Importantly, ROS are produced prior to cell death induction and promote the assembly of the Receptor-interacting protein kinase (RIP)1/RIP3 necrosome complex via a potential positive feedback loop since on the one hand radical scavengers diminish RIP1/RIP3 necrosome formation and since on the other hand RIP1 or RIP3 silencing attenuates ROS production. Furthermore, the deubiquitinase CYLD contributes to BV6/TNFα-induced ROS generation, necrosome assembly and cell death since CYLD knockdown attenuates all these events. Of note, knockdown of the downstream effector protein mixed lineage kinase domain like (MLKL) only partly reduces BV6/TNFα-triggered ROS production and cell death and does not affect necrosome formation. Contrary to expectations, the MLKL inhibitor Necrosulfonamide (NSA) not only decreases BV6/TNFα-stimulated ROS production and cell death but also attenuates RIP1/RIP3 necrosome assembly pointing to additional and MLKL-independent anti-necroptotic effects of NSA. Interestingly, silencing of the potential necroptotic excecutors mitochondrial proteins phosphoglycerate mutase family member 5 (PGAM5) or Dynamin-related protein 1 (Drp1) does not affect BV6/TNFα-induced cell death. Consistently, mitochondrial perturbations are not implicated in BV6/TNFα-induced cell death since mitochondrial membrane potential and respiration remain stable along with to BV6/TNFα-triggered necroptosis induction. Interference with the mitochondrial potential by depolarizing agents such as FCCP reduces BV6/TNFα-induced necroptosis indicating that proper mitochondrial function or a well-defined redox status is required for necroptotic cell death execution. This study demonstrates that ROS are critically involved in BV6/TNFα-induced necroptosis and thus provides novel insights into the redox regulation of necroptotic signaling.
Electron microscopy (EM) demarcates itself from other structural biology techniques by its applicability to a large range of biological objects that spans from whole cells to individual macromolecules. In single-particle cryo-EM, frozen-hydrated samples, prepared by vitrification with liquid ethane, retain macromolecules in a medium that approximates their natural aqueous environment and that, in this way, preserves high-resolution structural information. Nonetheless, the sensitivity of biological specimens to the high-energy electron beam introduces restrictions on the total dose that can be used during imaging while avoiding significant radiation damage. Consequently, the signal-to-noise ratio attained in each individual image is very low, and structures with high-resolution detail must be recovered by averaging thousands of projections in random orientations. This is achieved through the use of image processing algorithms capable of aligning and classifying particle images through the evaluation of cross-correlation functions between each particle and a reference.
In recent years, several innovations took place in the field of single-particle cryo-EM, among which the development of direct electron detectors must be highlighted. Direct electron detectors have a better detective quantum efficiency (DQE) than both photographic film and CCD cameras, and offer a fast readout, compatible with the acquisition of movie stacks. Additionally, new image processing software has become available, with more sophisticated algorithms and designed to take advantage of the specific characteristics of the movies produced with direct electron detectors. These technological advances in both hardware and software catalyzed a revolution in single-particle cryo-EM, which is now routinely used for the determination of near-atomic structures. As a result, the range of macromolecules accessible to cryo-EM has increased drastically, as targets that were unsuitable before for imaging due to their small dimensions can now be adequately visualized and refined to high-resolution.
During my doctoral work, I have used single-particle cryo-EM to structurally characterize challenging membrane proteins, with a strong emphasis on protein complexes from aerobic respiratory chains. In chapter I of this thesis, I present my results on the bovine respirasome, a mitochondrial supercomplex composed of complexes I, III and IV. Chapter II is dedicated to the analysis of the structure of alternative complex III (ACIII) from Rhodothermus marinus, a bacterial quinol:cytochrome c/HiPIP oxidoreductase unrelated to the canonical cytochrome bc1 complex (complex III). In addition, in chapter III I describe the structure of KimA, a high-affinity potassium transporter that drives the transport of its substrate by using the energy stored in the form of a proton gradient. These three membrane proteins, with molecular weights ranging from 140 kDa to 1.7 MDa, illustrate the possibilities and limitations faced in single-particle cryo-EM.
The aerobic respiratory chain is responsible for the generation of a transmembrane difference of electrochemical potential that is then used by ATP synthase for the production of ATP or for driving solute transport over the membrane. They catalyze the transfer of electrons from a substrate, such as NADH or succinate, to molecular oxygen and use the chemical energy released in these redox reactions to drive the translocation of protons, or in some cases sodium ions, to the intermembrane space in mitochondria or the periplasm in bacteria.
In mitochondria, the respiratory chain is composed of four complexes: complex I (NADH:ubiquinone oxidoreductase), complex II (succinate dehydrogenase), complex III (cytochrome bc1 complex) and complex IV (cytochrome c oxidase). While it was for a long time believed that these complexes existed as single entities in the membrane, the use of milder procedures for protein purification and analysis revealed that respiratory complexes associate into well-ordered structures, known as supercomplexes. These have been proposed to offer different structural and functional advantages that are still controversial, including substrate channeling, stabilization of individual complexes and reduction of reactive oxygen species (ROS) production. The most thoroughly studied respiratory supercomplex has been the respirasome, conserved in higher eukaryotes and composed of one copy of complex I, a complex III dimer and one complex IV. By single-particle cryo-EM analysis, I retrieved a 9 Å map of the respirasome from Bos taurus, which allowed the accurate docking of atomic models of the three component complexes. The structure shows that complex III associates to the concave side of the membrane arm of complex I, while complex IV is located between the end of the complex I hydrophobic arm and complex III. Several defined protein-protein contacts are observed between the component complexes, which are mediated predominantly by supernumerary subunits and close to the membrane surfaces. The interactions established between complex I and complex III are extensive and may support the argument that the association of complex I into supercomplexes is required for the stabilization or even the biogenesis of this complex.
...
The universal biological energy currency adenosine triphosphate (ATP) is synthesized by the F1Fo-ATP synthase in most living organisms. The overall structure and function of F-type ATPases is conserved in the different organisms. The F1Fo-ATP synthase consist of two domains; the soluble F1 complex has the subunit stoichiometry α3β3γδε and the membrane embedded Fo complex consists of subunits ab2c10-15 in its simplest form found in bacteria. F1 and Fo both function as reversible rotary motors that are connected by a central stalk (γε) and a peripheral stalk (b2δ).
For ATP synthesis, the electrochemical energy formed by a proton or sodium ion gradient is required. The ion translocation across the Fo subcomplex induces torque in the motor part of the enzyme (cnγε), which causes conformational changes in the α3β3 domain leading to ATP synthesis from ADP and inorganic phosphate (Pi) catalyzed in the β-subunits. ATP hydrolysis causes a reverse torque in the Fo subcomplex triggering uphill ion translocation from cytoplasm to periplasm, and the enzyme functions as an ion pump.
The ATP synthesis mechanism is well understood, since several high-resolution structures of F1 are available. In contrast, the ion translocation mechanism across the membrane, mediated by the Fo subcomplex, is not understood in its structural detail.
Subunit a and the c-ring form an ion pathway, but subunit b is needed to form an active ion translocation pathway in both H+- and Na+-dependent systems. Several high-resolution structures of c-rings have provided insights in the ion translocation mechanism. The different ion translocation models based on biochemical, biophysical and structural analysis are in agreement in the fact that ions are translocated through a periplasmic ion access pathway in subunit a to the middle of the membrane and there to the binding site of a c-subunit. After almost a whole rotation of the c-ring the ion returns into the a-c interface, where it can be released to the cytoplasm. In the different models the cytoplasmic access pathway has been proposed to be located in subunit a, at the a-c interface or within the c-ring. The driving force of torque generation has been proposed to be the pH gradient or membrane potential. Several biochemical studies show that a conserved arginine in helix four of subunit a (R226 in Ilyobacter tartaricus or R210 in Escherichia coli)plays a critical role in the ion translocation. The arginine has been proposed to function as an electrostatic separator between the cytoplasmic and periplasmic pathways and as a mediator of the ion exchange into the c-ring ion-binding site.
Structural data of a related enzyme (V1Vo-ATPase from Thermus thermophilus) has provided insight into the helical arrangement of the ion translocating subunits I and Lring (related to subunit a and the c-ring). These structures indicated a small interface between subunit I and the L-ring, and two four-helix bundles in the N-terminal domain of subunit I were proposed to build the periplasmic and cytoplasmic ion pathways. To comprehend the ion-translocation and torque generation mechanism in F1Fo-ATP synthase, structural data of an intact a-c complex is needed.
The goal of this work was to obtain structural data of subunit a, most preferably in a complex with the c-ring or additionally with subunit b. Therefore, a new purification procedure for the I. tartaricus Fo-subcomplex, heterologously expressed in E. coli cells, was established. The purified Fo was characterized biochemically and by Laserinduced liquid bead ion desorption mass spectrometry (LILBID-MS). These analyses showed that pure and completely assembled Fo containing all its subunits in the correct stoichiometry (ab2c11) was obtained. The purified Fo complex was stable at 4°C for several months and at room temperature in the presence of lipids for several weeks. A lipid analysis was performed by thin-layer chromatography (TLC) to investigate the qualitative lipid composition of I. tartaricus whole lipid extract and various I. tartaricus F1Fo isolates. The whole lipid extract contained PC, PG and PE lipids and probably cardiolipin. PC, PG and PE lipids were bound to wild type I. tartaricus F1Fo, whereas recombinant I. tartaricus F1Fo did not have any bound lipids, but was able to bind the synthetic lipids POPC and POPG if they were provided during the purification.
For subsequent structural studies the purified Fo was subjected to two-dimensional (2D) crystallization trials. Vesicles and sheets tightly packed with protein and crystals with a rare plane group for I. tartaricus c11 (p121) were obtained. The c-ring was visible in the CCD images, and immunogold-labeling revealed the presence of the His-tagged a-subunit in the reconstituted vesicles. Furthermore, atomic force microscopy (AFM) imaging showed protein densities next to the c-rings, which protruded less from the membrane (0.4±0.1 nm) than the c-ring (0.7±0.1 nm). These protein densities presumably belonged to subunit a.
Cryo-electronmicroscopy (cryo-EM) was used to collect data of the p121 crystals and a merged projection density map was calculated to 7.0 Å resolution. The unit cell of the crystals (81 × 252 Å) contained two asymmetric units with three c-rings in each and next to the c11-rings new prominent densities were visible. In each extra density up to 7 transmembrane helices were visible, belonging to the stator subunit a and/or subunit b. To elucidate whether there are conserved elements in the three extra densities non-crystallographic averaging was applied using a single-particle approach.
Six possible arrangements for the c-rings and the extra densities were identified and used for the averaging. The extra densities were enhanced only in one of the possible arrangements. The average showed a four-helix bundle and a fifth helix in close proximity to the c-ring. Two more helices were present in each position but their position was ambivalent. The data obtained in this work provides the first insight in the helical arrangement in the a-c interface of F1Fo-ATP synthase.
During my thesis, I worked on two different membrane proteins. One is a bacterial secondary transporter and the second is a human mitochondrial calcium channel.
The first part of my thesis involves the structural and biochemical characterization of an L-carnitine/ γ-butyrobetaine antiporter from bacteria called CaiT. The aim of the project was to understand the Na+ independence of CaiT and to determine the crystal structures of CaiT in different conformations to expand the mechanistic understanding of substrate/ product antiport in CaiT.
The study revealed how a positively charged amino acid side chain (arginine 262) in CaiT could structurally and functionally mimic a sodium ion. Additionally, various crystal structures of CaiT obtained in this study demonstrate that the central substrate-binding site is highly dynamic and can accommodate the substrate in various orientations.
In the second part of my thesis, I was able to optimize the expression and purification conditions for the human mitochondrial calcium uniporter or the MCU. Understanding how this channel functions can help us unravel the mechanism of calcium uptake by mitochondria. Secondary structure prediction analysis in combination with mass spectrometry of degraded MCU products obtained during the purification of the full-length protein led to the identification of a stable MCU construct. This study resulted in the successful purification of milligram quantities of stable MCU protein for the first time. Further optimization may be required to obtain more homogenous protein that is amenable for crystallization.
In der vorliegenden Dissertation stand die Aufklärung der Funktion und Regulation von p21 in der Mitose im Mittelpunkt. p21 ist als Cdk-Inhibitor und Schlüsselregulator bekannt, der in viele fundamentale zelluläre Prozesse involviert ist: Zellzyklusregulation, Apoptose, Seneszenz, Zellmigration und Dynamik des Zytoskeletts, Transkription, Differenzierung sowie DNA-Reparatur, aber auch in die Umprogrammierung induzierter pluripotenter Stammzellen (Besson et al. 2008; Abbas und Dutta 2009; Jung et al. 2010).
Die unkontrollierte Proliferation von Zellen ist mit der Tumorgenese assoziiert und wird unter anderem durch die Fehlregulation von p21, aber auch durch die wichtigen mitotischen Kinasen Cdk1, im Komplex mit ihrer regulatorischen Untereinheit Cyclin B1, sowie Plk1 bedingt. Zudem ist das Fehlen von p21 oder die Fehllokalisation in das Zytoplasma mit einer schlechteren Prognose für den Patienten und Chemotherapie-Resistenz von Tumoren verbunden (Abukhdeir und Park 2008). Aufgrund der zunehmenden Inzidenz und Mortalität von Krebserkrankungen ist es daher von besonderem klinischem Interesse, die molekularen Ursachen für die Entstehung maligner Tumorerkrankungen aufzuklären. Bislang existieren kaum Studien über welche molekularen Mechanismen die Funktionen von p21, dem wichtigsten Cdk-Inhibitor, der zum Beispiel durch die Anwendung niedermolekularer Inhibitoren wie BI 2536, das sich bereits in klinischen Phase II Studien befindet (Strebhardt 2010), beeinflusst wird, während der Mitose reguliert werden.
In der vorliegenden Dissertation wurde daher die physiologische Rolle des Cdk-Inhibitors bzw. Regulators p21 während der Mitose untersucht und mit der Kinaseaktivität von Cdk1/Cyclin B1, wie auch Plk1 korreliert. Es konnte gezeigt werden, dass p21 während der Mitose stark exprimiert wird und dass mitotisches p21 in verschiedenen Krebszelllinien unabhängig von dem p53-Status in einer phosphorylierten Form vorkommt, welche mit der Aktivität von Cdk1 und weniger mit der von Cdk2 assoziiert ist. Durch Untersuchungen der isogenen HCT116-Zelllinien mit und ohne p21 wurde aufgezeigt, wie wichtig p21 für den ordnungsgemäßen Ablauf der Mitose ist. Ohne p21 sind sowohl die Anaphase wie auch die Zytokinese verlängert, die Zellen ordnen die Chromosomen fehlerhaft in der Metaphaseplatte an (congression Fehler), besitzen weitaus mehr lagging Chromosomen und fast 20 % der Zellen weisen im Versuchsverlauf Polyploidie auf. Durch den Verlust des Cdk-Regulators p21 kommt es zur Fehlregulation von Cdk1 und seiner Substrate (wie MCAK) und es treten die oben beschriebenen Probleme auf.
Weiterhin phosphoryliert Cdk1/Cyclin B1 p21 an Ser-130 in vitro und ex vivo in der frühen Phase der Mitose, der Prophase bzw. Prometaphase. Die nicht phosphorylierbare p21 Form S130A befindet sich hauptsächlich im Zellkern und führt zu vermehrtem Auftreten von congression Fehlern, während die S130D-Mutante, die die Phosphorylierung durch Cdk1 vortäuscht, schneller degradiert wird und zudem den Phänotyp der HCT116 p21-/- Zellen verstärkt. Zellen, die S130D exprimieren, benötigen mehr Zeit für das Durchlaufen der Mitose. Hier ist vor allem die Metaphase stark verlängert, aber auch Anaphase und Zytokinese. Dies führt zu congression Fehlern und zu Polyploidie. Diese Ergebnisse bestätigen, wie wichtig die zeitlich korrekte Phosphorylierung von p21 und die dadurch vermittelte Aktivierung von Cdk1/Cyclin B1 ist.
Darüber hinaus stabilisiert die Suppression von Plk1 das p21 Protein, was darauf hinweist, dass die Degradation von p21 während der Prometaphase von Plk1 kontrolliert wird. Dies wird von der Tatsache unterstützt, dass Ser-114, wie auch Ser116 von Plk1 in vitro phosphoryliert wird. Die Deregulation von p21 durch Plk1, SS114/116AA bzw. SS114/116DD induziert Chromosomenfehler, wodurch die molekularen Mechanismen, warum fehlreguliertes Plk1 die Tumorgenese fördert, hervorgehoben werden.
Nach Abschluss der bisherigen Untersuchungen steht fest, dass man sich von der starren Rolle von p21 als Tumorsuppressor und Akteur während der G1/S-Phase lösen muss. Der Cdk-Inhibitor p21 trägt entscheidend zur mitotischen Progression bei, vor allem bedingt durch die zeitlich ordnungsgemäße Inaktivierung bzw. Aktivierung von Cdk1/Cyclin B1, der Kinase, die wiederum zahlreiche für die Mitose essentielle Proteine reguliert. In Zukunft muss zum besseren Verständnis der Rolle von p21 in der Mitose die genaue Abfolge der Ereignisse unter Einbeziehung der Degradationsmechanismen eingehender untersucht werden.
Rhabdomyosarcoma is the most common paediatric soft-tissue sarcoma, and for tumour recurrence, the prognosis is still unfavourable. The current standard therapy consisting of surgery, radiation and combined chemotherapy does not consider the specific biology of this tumour.
Histone deacetylases (HDACs) and the Lysine-specific demethylase-1 (LSD1) are two epigenetic modifiers which are both part of repressor complexes leading to transcriptional silencing of target genes. Whereas HDACs lead to deacetylation of several lysine-residues within the histone tail, LSD1 is specific for demethylation of H3K4me2 and H3K4me1, as well as in a different context for H3K9me2. Rhabdomyosarcoma is reported to harbour high levels of LSD1, but the functional relevance is yet unclear. HDAC inhibition proved to be effective as single agent treatment, however, the proximity of HDAC1/2 and LSD1 in repressor complexes at the DNA implies a suitable rationale for a combination therapy potentially leading to cooperative effects on target gene transcription. In this study, we aimed to evaluate the potential of a combined LSD1 and HDAC inhibition for cell death induction in rhabdomyosarcoma cell lines. Whereas LSD1 inhibitors failed to induce cell death on their own, the combined inhibition of HDACs and LSD1 resulted in highly synergistic cell death induction. This effect extended to several combinations of LSD1 and HDAC inhibitors as well as to four different rhabdomyosarcoma cell lines, two of embryonal and two of alveolar histology.
With the use of the HDAC inhibitor JNJ-26481585 and the reversible LSD1 inhibitor GSK690, we demonstrated that the cell death induced by the combination matches with the details of intrinsic mitochondrial apoptosis. JNJ-26481585/GSK690-induced cell death is partially caspase-dependent and leads to caspase cleavage, followed by substrate cleavage as shown for PARP, as well as loss of the mitochondrial membrane potential.
Furthermore, JNJ-26481585 and GSK690 acted together to transcriptionally upregulate the proapoptotic proteins NOXA, BIM and BMF, which resulted in respective changes on protein level for both cell lines. However, the antiapoptotic BCL-2 family proteins BCL-2, MCL-1 and BCL-xL displayed only minor changes in protein levels upon treatment with GSK690 and JNJ-26481585, which did not rely on transcriptional activity. Therefore, the increase in proapoptotic proteins induces a shift towards proapoptotic signalling at the mitochondrial membrane. This shift is functionally relevant since knockdown of a proapoptotic protein or overexpression of one of the antiapoptotic proteins BCL-2 and MCL-1, as well as a stabilized mutant MCL-1, can significantly protect from GSK690/JNJ-26481585-induced cell death.
Knockdown of the mitochondrial membrane protein BAK, which is directly guarding the mitochondrial membrane integrity, potently protected from GSK690/JNJ-26481585- induced cell death, directly linking the shift in the BCL-2 family proteins to the observed loss of mitochondrial membrane potential and the further downstream activation of caspases. Furthermore, treatment with JNJ-26481585 and GSK690 resulted in a cell cycle arrest in G2/M phase, indicating additional effects on the tumour cells beside apoptosis induction. Taken together, the combined inhibition of LSD1 and HDACs is a promising strategy for rhabdomyosarcoma treatment.
Im Rahmen dieser Arbeit sollte untersucht werden, ob eine Zellzyklusabhängigkeit der CD95- vermittelten Apoptose besteht. Dazu wurde ein ecdysoninduzierbares Genexpressionsystem für die induzierte Überexpression der CDK-Inhibitoren p21 und p27 in RKO-Zellen (Kolonkarzinomzellen) zur Herbeiführung eines Zellzyklusarrests in der G1-Phase benutzt. Nach Induktion mit dem Ecdysonhomolog Muristeron wurde durch Zugabe von rekombinanten hCD95-Liganden Apoptose ausgelöst und anschließend untersucht. Die erzielten Ergebnisse zeigen, dass der Induktor Muristeron an sich und nicht die p21- bzw. p27-Überexpression die anti-apoptotische Akt-Kinase aktiviert, die Expression des anti-apoptotischen Bcl-xL erhöht, die Caspase-8-Aktivierung (entweder am CD95-DISC oder durch "Feedback"-Aktivierung durch Caspase-3) und die darauf folgenden Ereignisse verhindert und somit die hCD95L-induzierte Apoptose blockiert. Zusätzlich beeinflusst der Induktor auch das Genexpressionsmuster der behandelten Zellen, was ebenfalls für die Hemmung der Apoptose mit verantwortlich sein könnte. Somit ist das ecdysoninduzierbare Genexpressionsystem zur Apoptoseuntersuchung in RKO-Zellen nicht verwendbar. Mit der Untersuchung des Apoptoseverhaltens proliferierender RKO-Zellen konnte gezeigt werden, dass überlebende Zellen nach hCD95L-Behandlung vermehrt in der G0/G1-Zellzyklusphase nachweisbar sind, während apoptotische (Caspase-3-positive) Zellen aus der G2/M-Phase heraus sterben. Allerdings weisen die apoptotischen Zellen kaum Cyclin B1 auf, ein für die G2-Phase wichtiges und typisches Cyclin. Somit bleibt die genaue Verknüpfung von Zellzyklusregulation und Apoptose auch nach diesen Analysen ungeklärt. In einem dritten Ansatz - Zellzyklusarrest durch Dichtearretierung - konnte eine Hemmung der CD95- vermittelten Apoptose in der arretierten Zellpopulation nachgewiesen werden. Allerdings sekretieren RKO-Zellen einen anti-apoptotischen Faktor in ihr Medium, dessen Konzentration und Wirkung mit größerer Zelldichte zunimmt und somit für die Protektion, unabhängig von Zellzyklusarrest oder Proliferation, verantwortlich ist. Konfluente und auch mit konditioniertem Medium behandelte RKO-Zellen zeigen im Vergleich zu dünn ausgesäten RKO-Zellen Veränderungen, die denen sehr ähnlich sind die beim Übergang einer epithelverankerten Zelle zu einer migrierenden Einzelzelle (EMT) auftreten. Beispielsweise verändert sich die Zusammensetzung des Zytoskeletts, die Zellen verlieren den Zell-Zell-Kontakt und lösen sich ab, bleiben aber am Leben. Zusätzlich steigt die Sekretion von Zytokinen an, die Angiogenese, Migration und Invasion positiv beeinflussen. Sowohl konfluente als auch mit konditioniertem Medium behandelte sub-konfluente Zellen sind apoptoseresistent (hCD95L, TRAIL, UV, Staurosporin), woran u.a. die Kinasen PKC und PI3K, aber auch das anti-apoptotische Bcl-xL beteiligt sind. Die Zellen sterben interessanterweise, wenn ein agonistischer anti-CD95-Antikörper statt des rekombinanten CD95-Liganden verwendet wird, was vermuten lässt, dass eine mangelhafte Vernetzung der einzelnen DISC-Komplexe zur Apoptosehemmung führt, welche durch den Antikörper dann aber erzwungen wird. Zwar handelt es sich hierbei um ein reines Zellkulturmodell, dennoch könnte es bedeuten, dass die Umgebung in einer dichten RKO-Zellkultur vergleichbar ist mit der in größeren soliden Tumoren. Die Zellen brauchen Nährstoffe, versuchen über eine Neovaskularisierung Anschluss an ein Blutsystem zu finden und sekretieren Lockstoffe, Wachstumsfaktoren sowie Proteasen, um die Metastasierung zu erleichtern. PI3K, cPKCs und Bcl-xL tragen dabei zu einer Apoptoseresistenz bei, welche die Zellen zum einen resistent gegenüber Anoikis, Nährstoffmangel, aber auch gegen angreifende zytotoxische T-Zellen macht. Eine weitere Aufklärung der hier ablaufenden Prozesse würde es erleichtern, Möglichkeiten zu finden, in diese Signalwege einzugreifen, um die Apoptosesensitivität wieder herzustellen und die Metastasierung zu verhindern. Insbesondere ist die Identifizierung des für die Apoptoseprotektion verantwortlichen Zytokins das nächste wichtige Ziel bei der Fortsetzung dieser Arbeiten.
Biomoleküle, insbesondere Membranproteine (MPs), sind oftmals sehr sensitiv gegenüber ihrer chemischen Umgebung, wie pH-Wert, Puffer, Salzkonzentration und vielen weiteren Faktoren. MPs stabil und funktional in Lösung zu halten ist nicht trivial. Sie stellen deshalb eine besondere Herausforderung bei der Analyse von biologischen Systemen dar. Aus diesem Grund wurden und werden nach wie vor sogenannte membrane mimicking-(MM-) Systeme, wie beispielsweise Nanodiscs (NDs) oder styrene-maleic acid lipid particles (SMALPs), untersucht und entwickelt, um MPs eine naturähnliche Umgebung in Form einer Lipid-Doppelschicht zu bieten und sie so in ihrer natürlichen Konformation und natürlichen Funktionsweise/Aktivität in Lösung zu halten.
Laser induced liquid bead ion desorption (LILBID) Massenspektrometrie (MS) hat sich als hervorragende analytische Methode herausgestellt, um MPs in Kombination mit MM-Systemen zu untersuchen. LILBID-MS bietet nicht nur die Möglichkeit Proteine an sich zu identifizieren, sondern ermöglicht ebenfalls eine zerstörungsfreie Analyse von nicht-kovalent gebundenen Proteinkomplexen, sowie die Detektion einzelner Subkomplexe eines Proteinkomplexes. Auch die Analyse von Protein-Ligand-Wechselwirkungen ist möglich. Bei der LILBID-Ionisationsmethode werden kleine Tröpfchen erzeugt, die einen wässrig gelösten Analyt enthalten. Die Analyt-Tröpfchen werden anschließend mittels IR-Laser bestrahlt, wodurch der Analyt freigesetzt und massenspektrometrisch analysiert werden kann.
Diese Dissertation beschäftigt sich zum einen mit der Analyse des Lyse-Proteins ΦX174-E der Bakteriophage ΦX174, zum anderen mit Untersuchungen zur Histidinkinase SpaK aus B. subtilis in Kombination mit MMs. Weiterhin wird die Frage geklärt, ob und wie gut sich LILBID-MS zur Analyse von Saposin-Nanopartikel-(SapNPs)-solubilisierten MPs eignet. Darüber hinaus wird in dieser Dissertation die Darstellung von SapNP-solubilisierten MPs mittels zellfreier Proteinsynthese näher charakterisiert und untersucht welche Parameter aus präparativer Sicht optimiert werden können.
In vorausgegangenen Analysen von ND-solubilisierten MPs mittels LILBID-MS zeigte sich, dass manche in Verbindung mit NDs genutzten Lipide unerwünschte Signale im Spektrum zur Folge haben, die aus massiven Lipid-Anhaftungen am MSP oder dem Analyten resultieren. Überlappungen der m/z-Signale verschiedener Analyt- und/oder Komplexkomponenten mit diesen Lipid-Cluster-Signalen kann wiederum zum Verlust von Informationen führen. Daher beschäftigt sich ein weiterer Teil dieser Arbeit mit der Frage, ob durch den Einsatz von UV-schaltbaren Lipiden der Anwendungsbereich und/oder die Auflösung von LILBID-MS erweitert und verbessert werden kann.
Um biologische Prozesse zu verstehen ist es ebenfalls wichtig die zeitlichen/kinetischen Aspekte einer Reaktion zu untersuchen/kennen, sowie molekulare Prozesse gezielt zu kontrollieren. Licht hat sich hierbei als ein hervorragendes Werkzeug in der Analytik, sowie in der molekularen Prozesskontrolle etabliert. Licht bietet den Vorteil sehr selektiv eingesetzt werden zu können und sowohl orts- als auch zeitaufgelöst Informationen liefern zu können. Das gezielte Triggern einer Reaktion oder einer Protein-Protein-Interaktion kann beispielsweise durch sog. photo-cleaving von photolabilen Schutzgruppen ermöglicht werden. Bisweilen bietet die native MS nur wenig Möglichkeiten schnelle Reaktionen zu analysieren und kinetische Informationen zu gewinnen. Daher beschäftigt sich ein weiterer Teil dieser Dissertation damit zu untersuchen, ob und wie sich lichtgesteuerte Reaktionen im LILBID-Ionisationsprozess induzieren und gegebenenfalls auch zeitlich analysieren und charakterisieren lassen können.
Mitochondria are important for cellular health and their dysfunction is linked to a variety of diseases, especially neurodegeneration. Thus, the renewal and degradation of dysfunctional mitochondria is crucial for the well-being of organisms. The selective digestion of damaged mitochondria via the lysosome (mitophagy), is the main pathway to do so.
In my dissertational work, I investigated the connection between protein misfolding, protein import into mitochondria and the degradation of mitochondria via mitophagy. Here, I present a new model for the initiation of mitophagy without collapse of the membrane potential. This model provides the link between protein import into mitochondria, stress signal transduction to the cytosol and the mitochondrial stress sensor PINK1. To comprehensively examine how mitophagy can be triggered, I performed a genome-wide CRISPR knockout screen utilizing the mitophagy reporter mitochondrial mKEIMA. Thereby, I observed numerous novel gene deletions that induce mitophagy. Prominently, I identified an accumulation of gene deletions of the protein import and of protein quality control factors. I validated several of those and examined HSPA9 (mitochondrial HSP70) and LONP1 (a mitochondrial matrix AAA protease) in more detail, regarding their effect on mitophagy and protein import. For this, I used an established fluorescence-based, mitochondrial-targeted EGFP, as well as a newly-developed pulsed-SILAC mass spectrometry approach (mePRODmt). Depletions of both genes resulted in reduced protein import and PINK1-dependent mitophagy. Strikingly, I did not observe any loss of mitochondrial membrane potential, which was hitherto believed to be essential for activation of PINK1-mediated mitophagy. Literature shows that certain mitochondrial stressors can also induce mitophagy without mitochondrial membrane depolarization, which I confirmed with my assays. Next, I characterized the impact of LONP1 and HSPA9 depletion, which are involved in proteostasis maintenance, and the mtHSP90 inhibitor GTPP on mitochondrial protein folding in more detail. GTPP treatment and LONP1 depletion both resulted in the accumulation of an insoluble protein fraction, as judged by proteomic analysis. This insoluble protein fraction enriched several components of the presequence translocase-associated motor PAM, including TIMM44. TIMM44 acts as a link between the translocon, the import pore of the inner mitochondrial membrane (TIM) complex and the PAM complex. Thus, I hypothesized that TIMM44 dissociates from the TIM complex upon protein folding stress, when it becomes part of the insoluble protein fraction. To validate this model, I measured the TIMM44 interactome upon proteostasis disturbance using proximity labeling. Indeed, interaction of TIMM44 with the import pore was almost completely abolished, explaining the loss of matrix-targeted import upon protein folding stress. From these findings, I reasoned that an import reduction mediated by the PAM complex would likely also inhibit the degradation of PINK1. Consistent with this hypothesis, I observed that mitophagy induced by HSPA9 or LONP1 deletion was prevented when PINK1 was genetically deleted. In comparison, non-processed PINK1 was stabilized on mitochondria in wild type cells when mitochondrial protein import was impaired. On this basis, I drew the conclusion that the loss of mitochondrial import was the stress signal, which leads to the stabilization of PINK1, as it could not be processed anymore via the inner mitochondrial membrane protease PARL. PINK1 auto-activates itself upon accumulation and signals to the cytosol that this mitochondrion is damaged. Mitophagy is subsequently initiated by the ubiquitin kinase activity of PINK1. As a result, the autophagy apparatus gets activated, damaged mitochondria are engulfed by a double membrane and removed via lysosomal digestion. This proposed model is, to the best of my knowledge, the first to provide an explanation for protein folding stress-induced and protein import inhibition-triggered mitophagy without mitochondrial depolarization. The model thus extends the PINK1/PARKIN-dependent mitophagy pathway to milder stresses and clears some of the open questions in the field. Furthermore, this work is also important, because protein misfolding stress and dysfunctional mitochondria are two hallmarks of neurodegeneration. In particular, mitochondrial protein import inhibition during Parkinson’s and Huntington disease might be driver of mitochondrial dysfunction. Hence, I hope and anticipate that the newly developed protein import method, mePRODmt, and the proposed model will be beneficial to further characterize underlying processes and to establish which factors prevent or drive these disorders on molecular level.
Solute carrier (SLC) are related to various diseases in human and promising pharmaceutical targets but more structural and functional information on SLCs is required to expand their use for drug design and therapy. The 7-transmembrane segment inverted (7-TMIR) fold was identified for the SLC families 4, 23 and 26 in the last decade thus detailed analysis of the structure function relationship of one of these families might also yield insights for the other two. SVCT1 and SVCT2 from the SLC23 family are sodium dependent ascorbic acid transporters in human but structural analysis of the SLC23 family is exclusively based on two homologs – UraA from E. coli and UapA from A. nidulans – yielding two inward-facing and one occluded conformation. In combination with outward-facing conformations from SLC4 transporters, and additional information from the SLC26 family, an elevator transport mechanism for all 7-TMIR proteins was identified but detailed mechanistic features of the transport remain elusive due to the lack of multiple conformations from individual transporters.
To increase the understanding of 7-TMIR protein structure and function in this study, the transport mechanism of SLC23 transporters was analyzed by two strategies including selection of alpaca derived nanobodies and synthetic nanobodies against UraA as prokaryotic model protein of the SLC23 family. The second strategy involved mutagenesis of UraA at functional relevant positions regarding the conformational change during transport. Therefore, available structures of 7-TMIR proteins and less related elevator transporters were analyzed and a common motif identified – the alpha helical inter-domain linkers. The proposed rigid body movement for transport in combination with the characteristic alpha helical secondary structure of the linkers connecting both rigid bodies led to the hypothesis of functional relevance of the linkers and a conformational hinge being located in close proximity to the linkers. These positions were identified and used to modulate the biophysical properties of the transporter. Mutagenesis at three relevant positions led to loss of transport functionality and these UraA variants could be recombinantly produced and purified to further examine the underlying mechanistic effects. The variants UraAG320P and UraAP330G from the periplasmic inter-domain linker showed increased dimerization and thermal stability as well as substrate binding in solution. The substrate affinity of UraAG320P was identified to be 5-fold higher compared to the wildtype. The solvent accessibility of the substrate binding site in UraAG320P and UraAP330G revealed reduced open probability that indicated an altered conformational space compared to UraAWT. This phenomenon was analyzed in more detail by differential hydrogen-deuterium exchange mass spectrometry and the results supported the hypothesis of a reduced open probability and gave further insights into the impact of the two mutations in the periplasmic inter-domain linker in UraA.
This thesis further presents strategies for phage display selection of nanobodies with epitope bias and a post selection analysis pipeline to identify nanobodies with desired binding characteristics. Thereby, whole cell transport inhibition highlighted periplasmic epitope binders and conformational selectivity. A cytoplasmic epitope could be identified by pulldown with inside-out membrane vesicles for one cytoplasmic side binder. Thermal stabilization analysis of the target protein in differential scanning fluorometry was performed in presence of two different nanobodies to identify simultaneous binding by additional thermal stabilization respectively competition by intermediate melting temperatures. Combination of epitope information with simultaneous DSF could be used to identify the stabilization of different UraA conformations by a set of binders and presents a general nanobody selection strategy for other SLCs. Synthetic nanobodies (sybodies) were also included in the analysis pipeline and Sy45 identified as promising candidate for co-crystallization that gave rise to UraAWT crystals in several conditions in presence or absence of uracil. Similar crystals could be obtained in combination with UraAG320P that were further optimized to gain structural information on this mutant. The structure was solved by molecular replacement and the model refined at 3.1 Å resolution confirming the cytoplasmic epitope of Sy45 as predicted by the selection pipeline. The stabilized conformation was inward-facing similar to the reported UapA structure but significantly different to the previously reported inward-facing structure of UraA. The structure further confirmed the structural integrity of the UraA mutant G320P. Despite the monomeric state of UraA in the structure, the gate domain aligned reasonably well with the gate domain of the previously published dimeric UraA structure in the occluded conformation and allowed detailed analysis of the conformational transition in UraA from inward-facing to occluded by a single rigid body movement. Thereby little movement in the gate domain of UraA was observed in contrast to a previously reported transport mechanism. Core domain rotation around a rotation axis parallel to the substrate barrier was found to explain the major part of conformational transition from inward-facing to occluded and experimentally supported the hypothesized mechanism by Chang et al. (2017). Additionally, the conformational hinge around position G320 in UraA could be identified as well as the impact of the backbone rigidity introduced by the highly conserved proline residue at position 330 in UraA on the conformational transition. This position was found to serve as anchoring point the inter-domain linker and determines the coordinated movement of inter-domain linker and core domain. The functional analysis further highlighted the requirement of alpha helical secondary structure within the inter-domain linker that serves as amphipathic structural entity that can adjust to changed core-gate domain distances and angles during transport by extension/compression or bending while preserving the rigid linkage.
The applied strategies to modulate the conformational space of UraA by mutagenesis at the hinge positions in the inter-domain linkers is transferrable to other transporters and might facilitate their structural and functional characterization.
Further, this study discusses the conformational thermostabilization of UraA that is based on increased melting temperatures upon restriction of its conformational freedom. The term ‘conformational thermostabilization’ introduced by Serrano-Vega et al. (2007) could be experimentally supported and the direct correlation between the conformational freedom and thermostabilization was qualitatively analyzed for UraA. The concept of conformational thermostabilization might help in characterization of other dynamic transport systems as well.
Ubiquitination is regarded as one of the key post-translational modifications in nearly all biological processes, endowed with numerous layers of complexity. Deubiquitinating enzymes (DUBs) dynamically counterbalance ubiquitination events by deconjugating ubiquitin signals from substrates. Dysregulation of the ubiquitin code and its negative regulators drive various pathologies, such as neurological disorders and cancer.
The DUB ubiquitin-specific peptidase 22 (USP22) is well-known for its essential role in the human Spt-Ada-Gcn5 acetyltransferase (SAGA) complex, mediating the removal of monoubiquitination events from Histone 2A and 2B (H2A and -B), thereby regulating gene transcription. In cancer, USP22 was initially described as a part of an 11-gene expression signature profile, predicting tumor metastasis, reoccurrence and death after therapy in a wide range of tumor cells. However, novel roles for USP22 have emerged recently, accrediting USP22 essential roles in regulating tumor development as well as apoptotic cell death signaling.
One of the hallmarks of cancer is the evasion of cell death, especially apoptosis, a form of programmed cell death (PCD). Necroptosis, a regulated form of necrosis, is regarded as an attractive therapeutic strategy to overcome apoptosis-resistance in tumor cells, although a profound understanding of the exact signaling cascade still remains elusive. Nevertheless, several ubiquitination and deubiquitination events are described in fine-tuning necroptotic signaling.
In this study, we describe a novel role for USP22 in regulating necroptotic cell death signaling in human tumor cell lines. USP22 depletion significantly delayed TNFa/Smac mimetic/zVAD.fmk (TBZ)-induced necroptosis, without affecting TNFa-induced nuclear factor-kappa B (NF-KB) signaling or TNFa-mediated extrinsic apoptosis. Intriguingly, re-expression of USP22 wildtype in the USP22 knockout background could re-sensitize HT-29 cells to TBZ-induced necroptosis, whereas re-constitution with the catalytic inactive mutant USP22 Cys185Ser did not rescue susceptibility to TBZ-induced necroptosis, confirming the USP22 DUB-function a pivotal role in regulating necroptotic cell death. USP22 depletion facilitated ubiquitination and unexpectedly also phosphorylation of Receptor-interacting protein kinase 3 (RIPK3) during necroptosis induction, as shown by Tandem Ubiquitin Binding Entities (TUBE) pulldowns and in vivo (de)ubiquitination immunoprecipitations. To substantiate our findings, we performed mass-spectrometric ubiquitin remnant profiling and identified the three novel USP22-regulated RIPK3 ubiquitination sites Lysine (K) 42, K351 and K518 upon TBZ-induced necroptosis. Further assessment of these ubiquitination sites unraveled, that mutation of K518 in RIPK3 reduced necroptosis-associated RIPK3 ubiquitination and additionally affected RIPK3 phosphorylation upon necroptosis induction. At the same time, genetic knock-in of RIPK3 K518R sensitizes tumor cells to TNFa-induced necroptotic cell death and amplified necrosome formation.
In summary we identified USP22 as a new regulator of TBZ-induced necroptosis in various human tumor cell lines and further unraveled the distinctive role of DUBs and (de)ubiquitination events in controlling programmed cell death signaling.
Das Hauptziel dieser Dissertation lag in der Verbesserung einzelner Schritte im Prozess der automatischen Proteinstrukturbestimmung mittels Kernmagnetischer Resonanz (NMR). Dieser Prozess besteht aus einer Reihe von sequenziellen Schritten, welche zum Teil bereits erfolgreich automatisiert wurden. CYANA ist ein Programmpaket, welches routinemäßig zur automatischen Zuordnung der chemischen Verschiebungen, der Nuclear Overhauser Enhancement (NOE) Signalen und der Strukturrechnung von Proteinen verwendet wird. Einer der Schritte, der noch nicht erfolgreich automatisiert wurde, stellt die Signalidentifizierung von NMR Spektren dar. Dieser Schritt ist besonders wichtig, da Listen von NMR-Signalen Grundlage aller Folgeschritte sind. Fehler in den Signallisten pflanzen sich in allen Folgeschritten der Datenauswertung fort und können am Ende in falschen Strukturen resultieren. Daher war ein Ziel dieser Arbeit, einen robusten und verlässlichen Algorithmus zur Signalidentifizierung von NMR Spektren in CYANA zu implementieren. Dieser Algorithmus sollte mit dem in FLYA implementierten Ansatz zur automatischen Resonanzzuordnung, der automatischen NOE-Zuordnung und der Strukturrechnung mit CYANA kombiniert werden. Der in CYANA implementierte CYPICK Algorithmus ahmt den von Hand durchgeführten Ansatz nach. Bei der manuellen Methode schaut sich der Wissenschaftler zweidimensionale Konturliniendarstellungen der NMR Spektren an und entscheidet anhand verschiedener Geomtrie- und Ähnlichkeitskriterien, ob es sich um ein Signal des Proteins oder um einen Artefakt handelt. Proteinsignale sind ähnlich zu konzentrischen Ellipsen und erfüllen bestimmte geometrische Kriterien, wie zum Beispiel ungefähr kreisförmiges Aussehen nach entsprechender Skalierung der spektralen Achsen und gänzlich konvexe Formen, die Artefakte nicht aufzeigen. CYPICK bewertet die Konturlinien lokaler Extrema nach diesen Bedingungen und entscheidet anhand dieser, ob es sich um ein echtes Signal handelt oder nicht. Das zweite Ziel dieser Arbeit war es ein Maß zur Quantifizierung der Information von strukturellen NMR Distanzeinschränkungen zu entwickeln. Der sogenannte Informationsgehalt (I) ist vergleichbar mit der Auflösung in der Röntgenkristallographie. Ein weiteres Projekt dieser Dissertation beschäftigte sich mit der strukturbasierten Medikamentenentwicklung (SBDD). SBDD wird meist von der Röntgenkristallographie durchgeführt. NMR hat jedoch einige Vorteile gegenüber der Röntgenkristallographie, welche interessant für SBDD sind. Daher wurden Strategien entwickelt, die NMR für SBDD zugänglicher machen sollen.