Refine
Has Fulltext
- yes (22)
Is part of the Bibliography
- yes (22)
Year of publication
Document Type
- Doctoral Thesis (22) (remove)
Language
- English (22) (remove)
Keywords
- Vaccinia-Virus (5)
- Biogenese (3)
- oncolytic virotherapy (3)
- Onkolyse (2)
- SMN (2)
- Small nuclear RNP (2)
- Tumor (2)
- Vaccinia virus (2)
- electron microscopy (2)
- onkolytische Virotherapie (2)
Institute
- Lehrstuhl für Biochemie (22) (remove)
RNA helicases are key players in the regulation of gene expression. They act by remodeling local RNA secondary structures as well as RNA-protein interactions to enable the dynamic association of RNA binding proteins to their targets. The putative RNA helicase DHX30 is a member of the family of DEAH-box helicases with a putative role in the ATP-dependent unwinding of RNA secondary structures. Mutations in the DHX30 gene causes the autosomal dominant neuronal disease “Neurodevelopmental Disorder with severe Motor Impairment and Absent Language” (NEDMIAL;OMIM#617804). In this thesis, a strategy was established that enabled the large-scale purification of enzymatically active DHX30. Through enzymatic studies performed in vitro, DHX30 was shown to act as an ATP-dependent 3’ → 5’ RNA helicase that catalyzes the unwinding of RNA:RNA and RNA:DNA substrates. Using recombinant DHX30, it could be shown that disease-causing missense mutations in the conserved helicase core caused the disruption of its ATPase and helicase activity. The protein interactome of DHX30 however, was unchanged indicating that the pathogenic missense-mutations do not cause misfolding of DHX30, but rather specifically affect its catalytic activity. DHX30 localizes predominantly in the cytoplasm where it forms a complex with ribosomes and polysomes. Using a cross-linking mass spectrometry approach, a direct interaction of the N-terminal double strand RNA binding domain of DHX30 with sites next to the ribosome’s mRNA entry channel and the subunit interface was uncovered. RNA sequencing of DHX30 knockout cells revealed a strong de-regulation of mRNAs involved in neurogenesis and nervous system development, which is in line with the NEDMIAL disease phenotype. The knockdown of DHX30 results in a decreased 80S peak in polysome gradients, indicating that DHX30 has an effect on the translation machinery. Sequencing of the pool of active translating mRNAs revealed that upon DHX30 knockout mainly 5’TOP mRNAs are downregulated. These mRNAs are coding for proteins of the translational machinery and translation initiation factors. This study identified DHX30 as a factor of the translation machinery that selectively impacts the expression of a subset of proteins and provides insight on the etiology of NEDMIAL.
Cancer is one of the major causes of mortality in developed countries. In 2020, there were more than 19.3 million new cases of tumor malignancies worldwide, with more than 10 million deaths. The high rates of cancer cases and mortality necessitate extensive research and the development of novel cancer treatments and antitumor agents. In most cases, conventional treatment strategies for tumor therapy are based on chemotherapeutic treatment, which is supplemented with radiotherapy and/or surgical resection of solid tumors [1]. The use of chemotherapy for the treatment of cancer has significant side effects, the most dangerous of which is toxicity [2] [3].
Modern methods of treating tumors focus on specific drug delivery to the tumor site, actively targeting the tumor cells, as well as the reduction of side effects. One of the most promising current approaches is based on oncolytic viruses. Antitumor properties of viruses were documented at the beginning of the 20th century when some cancer patients recovered after acute viral infections, particularly influenza [4]. Vaccinia virus (VACV) is a member of the Poxviridae family, has natural antitumor properties, and provides a good basis for generating efficient recombinant oncolytic strains. Furthermore, VACV has never been shown to integrate into the host genome [5]. VACV is likely one of the safest and well-studied viruses due to extensive research being done in molecular biology and pathophysiology to investigate its potential as a vaccine for smallpox eradication programs. It has been administered to over 200 million people worldwide. VACV antitumor therapeutic effectiveness has been established in xenograft models with a variety of tumor types for human and canine cancers. Furthermore, recombinant oncolytic VACVs expressing genes encoding light-emitting proteins are a big improvement in a treatment strategy that combines tumor-specific therapies and diagnostics.
Oncolytic virus treatments are effective in xenograft cancer models in mice, however, the significant improvements found in mice do not always translate to human cancer patients. These therapies should be tested in dogs with spontaneous cancer not only to offer well translatable information regarding the possible efficiency of viral therapy for human cancers but also to improve the health of our household pets as well. Spontaneous canine tumors are starting to be regarded as an essential model of human cancers that can reproduce the tumor microenvironment and immune response of cancer patients [6]. Just as data obtained in dog experiments can improve cancer therapy for human patients, these findings can also be used to improve treatment protocols in canine patients.
Hundreds of studies and dozens of reviews have been published regarding the antitumor effects of various recombinants of VACV, but information on the anticancer features of initial, genetically-unmodified “naïve” VACV is still limited. In the first studies, we compared different wild-type, non-modified strains of VACV and tested their oncolytic properties on a panel of various cancer cells derived from different organs. In addition, we also tested a protection system based on the “Trojan horse” concept - using a combination of human Adipose tissue-derived Stem Cells (hADSC) and three different wild-type single plaque purified Vaccinia virus strains: W1, L1, and T1. We showed that all tested human cell lines (FaDu, MDA MB 231, HNT-13, HNT-35, and PC-3) are permissive to L0, W0, T0, L1, W1, and L1 infection. Furthermore, we tested the cytotoxicity of VACV in different cancer cell lines (A549, PC-3, MDA-MB 231, FaDu, HNT-13, HNT-25, and HNT-35). All strains lysed the cells, which was most visible at 96 hpi. We also showed that all tested strains could efficiently infect and multiply in hADSC at a high level. In our in vivo study, we tested the therapeutic efficacy of the wild-type Vaccinia viruses L1, W1, and T1 alone or in combination with hADSC. Wild-type VACV strains were tested for their oncolytic efficiency in human lung adenocarcinoma (A549) in a xenograft model. Treatment of A549 tumors with different doses of L1 and W1 as well as with a L1/ADSC or W1/ADSC combination led to significant tumor regression compared to the PBS control. Additionally, the treatment with L1 and W1 and the combination of L1/ADSC and W1/ADSC was well tolerated by the animals. In the case of the wild-type Tian Tan strain, results were not obtained due to the high cytotoxicity of this strain. Therefore, it should be attenuated for further studies.
In the second part of the current study, we investigated the oncolytic effect of C1-opt1, W1 opt1, and L3-opt1 strains based on the wild-type Copenhagen, Wyeth, and Lister vaccines with additional expression of turboFP635. Replication and cytotoxicity assays demonstrated that all 3 viruses were able to infect, replicate in and kill canine tumor cell lines STSA-1 and CT1258 in a virus dose- and time- dependent fashion. Cytotoxicity and replication assays were also performed on cultured canine Adipose-derived Mesenchymal Stem Cells (cAdMSC). The results showed that the cells were lysed much slower than the tumor cells. It suggests that these cells can harbour the virus for a long-term period, allowing the virus to spread into the body and there is enough time to reach the primary tumor or metastases before the cell carrier is destroyed. The viral replication in cAdMSC in our study was lower than in canine cancer cells (STSA-1 and CT1258) at the same MOI. After being studied in cell culture, C1 opt1 and their combination with cAdMSC (C1-opt1/cAdMSC) were used in canine STSA 1 tumor bearing nude mice. We tested the oncolytic effect of the C1-opt1 virus alone and in combination with cAdMSC in the canine STSA-1 xenograft mouse model. Altogether, our findings have shown that both C1-opt1 and cAdMSC/C1-opt1 significantly reduced tumor size or eliminated the tumor. There was no significant difference between C1-opt1 alone and cAdMSC/C1-opt1. The virus particles were mostly found within the tumor after 24 dpi, some amount of virus particles were found in the lungs of mice injected with a combination of cAdMSC/C1-opt1 but not in the group injected with virus alone (cAdMSC might get stuck in the lungs and cause virus propagation there).
Taken together, this study provided a proof-of-concept that hADSC/cAdMSC can be used as a carrier system for the “Trojan horse” concept. However, it should be confirmed in another experimental model system, such as canine patients. Moreover, these findings suggest that wild-type, non-modified strains of Vaccinia virus isolates can be considered promising candidates for oncolytic virotherapy, especially in combination with mesenchymal stem cells.
Interleukin 2 (IL-2) was the first cytokine applied for cancer treatment in human history. It has been approved as monotherapy for renal cell carcinoma and melanoma by the FDA and does mediate the regression of the tumors in patients. One of the possible mechanisms is that the administration of IL-2 led to T lymphocytes expansion, including CD4+ and CD8+ T cells. In addition, a recent study demonstrated that antigen-specific T cells could also be expanded through the induction of IL-2, which plays a crucial role in mediating tumor regression. However, despite the long-term and extensive use of IL-2 in the clinic, the ratio of patients who get a complete response was still low, and only about one-fifth of patients showed objective tumor regression. Therefore, the function of IL-2 in cancer treatment should continue to be optimized and investigated. A study by Franz O. Smith et al. has shown that the combination treatment of IL-2 and tumor-associated antigen vaccine has a strong trend to increased objective responses compared to patients with melanoma receiving IL-2 alone. Peptide vaccines are anti-cancer vaccines able to induce a powerful tumor antigenspecific immune response capable of eradicating the tumors. According to the type of antigens, peptide vaccines can be classified into two distinct categories: Tumor-associated antigens (TAA) vaccine and tumor-specific neoantigens (TSA) vaccine. Currently, Peptide vaccines are mainly investigated in phase I and phase II clinical trials of human cancer patients with various advanced cancers such as lung cancer, gastrointestinal tumors, and breast cancers. Vaccinia virus (VACV) is one of the safest viral vectors, which has been wildly used in cancer treatment and pathogen prevention. As an oncolytic vector, VACV can carry multiple large foreign genes, which enable the virus to introduce diagnostic and therapeutic agents without dramatically reducing the viral replication. Meanwhile, the recombinant vaccinia virus (rVACV) can be easily generated by homologous recombination. Here, we used the vaccinia virus as the therapeutic cancer vector, expressing mouse Interleukin 2 (IL-2) and tumor-associated antigens simultaneously to investigate the combined effect of anti-tumor immune response in the 4T1 mouse tumor model. As expected, the VACV driven mIL-2 expression remarkably increased both CD4+ and CD8+ populations in vivo, and the virus-expressed tumor-associated peptides successfully elicited theantigen-specific T cell response to inhibit the growth of tumors. Furthermore, the experiments with tumor-bearing animals showed that the mIL-2 plus tumor antigens expressing VACV vector gave a better anti-cancer response than the mIL-2 alone expressing vector. The combinations did significantly more inhibit tumor growth than mIL-2 treatment alone. Moreover, the results confirmed our previous unpublished data that the mIL-2 expression driven by synthetic early/late promoter in the Lister strain VACV could enhance the tumor regression in the 4T1 mouse model.
Bacteria thrive and survive in many different environments, and as a result, they have developed robust mechanisms to adapt rapidly to alterations in their surroundings. The protection against osmotic forces is provided by mechanosensitive channels: their primary function is to maintain the integrity of the cell upon a hypoosmotic shock. The mechanosensitive channel of small conductance (MscS) is not only the smallest common structural unit of a diverse family that allows for a tailored response in osmoregulation; it is also the most intensively studied homologue. Mechanosensitive channels directly sense elevated membrane tension levels generated by increased pressure within the cell and open transiently. Escherichia coli has six paralogues that differ in their gating properties and the number of additional transmembrane (TM) helices. These TM helices, termed sensor paddles, are essential for sensing, as they directly contact the surrounding membrane; however, the role of the additional TM helices is still unclear. Furthermore, lipids occupy hydrophobic pockets far away from the membrane plane. A recent gating model for MscS states that increased membrane tension triggers the expulsion of lipids out of those pockets, modulating different conformational states of MscS. This model focuses on bound lipids, but it is still unclear to what extent the direct interaction with the membrane influences sensing and how relevant it is for the larger paralogues.
In the herein described work, structural studies on two larger paralogues, the medium-sized channel YnaI and the large channel YbiO were realised using electron cryomicroscopy (cryo-EM). Lipids were identified in YnaI in the pockets in a similar position and orientation as in MscS, suggesting a conserved sensing mechanism. Moreover, the copolymer diisobutylene/maleic acid (DIBMA) allowed the extraction of artificially activated YnaI from plasma membranes, leading to an open-like form of this channel. This novel conformation indicated that the pore helices bend at a GGxGG motif during gating, which is unique among the Escherichia coli paralogues, concomitant with a structural reorganisation of the sensor paddles. Thus, despite a high similarity of their closed states, the gating mechanisms of MscS and YnaI are surprisingly different. Furthermore, the comparison of MscS, YnaI, and YbiO accentuates variations and similarities between the differently sized family members, implying fine-tuning of channel properties in the pore regions and the cytosolic lateral entry sides into the channel. Structural analyses of MscS reconstituted into different systems showed the advantages and disadvantages of certain polymers and detergents. The novel DIBMA copolymer and the more conventional amphiphilic polymers, so-called Amphipols, perturb contacting transmembrane helices or lead to their denaturation. Due to this observation, the obtained structures of YnaI must also be cautiously considered. The structures obtained in detergents resulted in unaffected channels; however, the applicability of detergents for MscS-like channels is limited by the increased required sample concentration.
The role of lipids for gating MscS in the absence of a membrane was examined by deliberately removing coordinated lipid molecules from MscS using different amounts and kinds of detergent. The effects on the channel were inspected by cryo-EM. These experiments showed that closed MscS adopts the open conformation when it is enough delipidated by incubation with the detergent n-dodecyl-β-D-maltoside, and adding lipids to the open channel reverses this process. The results agree with the state-of-the-art model that the amount of lipid molecules in the pockets and grooves is responsible for the conformational state of MscS. Furthermore, incubation with the detergent lauryl maltose neopentyl glycol, which has stabilising and delipidating characteristics, resulted in a high-resolution structure of open MscS exhibiting an intricate network of ligands. Based on this structure, an updated gating model is proposed, which states that upon opening, lipids from the pockets migrate into the cytosolic membrane leaflet, while lipids from the periplasmic leaflet enter the grooves that arise between the sensor paddles.
∆Np63 is a master regulator of squamous cell identity and regulates several signaling pathways that crucially
contribute to the development of squamous cell carcinoma (SCC) tumors. Its contribution to coordinating the
expression of genes involved in oncogenesis, epithelial identity, DNA repair, and genome stability has been
extensively studied and characterized. For SCC, the expression of ∆Np63 is an essential requirement to
maintain the malignant phenotype. Additionally, ∆Np63 functionally contributes to the development of cancer
resistance toward therapies inducing DNA damage.
SCC patients are currently treated with the same conventional Cisplatin therapy as they would have been
treated 30 years ago. In contrast to patients with other tumor entities, the survival of SCC patients is limited,
and the efficacy of the current therapies is rather low. Considering the rising incidences of these tumor entities,
the development of novel SCC therapies is urgently required. Targeting ∆Np63, the transcription factor, is a
potential alternative to improve the therapeutic response and clinical outcomes of SCC patients.
However, ∆Np63 is considered “undruggable.” As is commonly observed in transcription factors, ∆Np63 does
not provide any suitable domains for the binding of small molecule inhibitors. ∆Np63 regulates a plethora of
different pathways and cellular processes, making it difficult to counteract its function by targeting
downstream effectors. As ∆Np63 is strongly regulated by the ubiquitin–proteasome system (UPS), the
development of deubiquitinating enzyme inhibitors has emerged as a promising therapeutic strategy to target
∆Np63 in SCC treatment.
This work involved identifying the first deubiquitinating enzyme that regulates ∆Np63 protein stability. Stateof-the-art SCC models were used to prove that USP28 deubiquitinates ∆Np63, regulates its protein stability,
and affects squamous transcriptional profiles in vivo and ex vivo. Accordingly, SCC depends on USP28 to
maintain essential levels of ∆Np63 protein abundance in tumor formation and maintenance. For the first time,
∆Np63, the transcription factor, was targeted in vivo using a small molecule inhibitor targeting the activity of
USP28. The pharmacological inhibition of USP28 was sufficient to hinder the growth of SCC tumors in
preclinical mouse models.
Finally, this work demonstrated that the combination of Cisplatin with USP28 inhibitors as a novel therapeutic
alternative could expand the limited available portfolio of SCC therapeutics. Collectively, the data presented
within this dissertation demonstrates that the inhibition of USP28 in SCC decreases ∆Np63 protein abundance,
thus downregulating the Fanconi anemia (FA) pathway and recombinational DNA repair. Accordingly, USP28
inhibition reduces the DNA damage response, thereby sensitizing SCC tumors to DNA damage therapies, such
as Cisplatin.
The eukaryotic gene expression requires extensive regulations to enable the homeostasis of the cell and to allow dynamic responses due to external stimuli. Although many regulatory mechanisms involve the transcription as the first step of the gene expression, intensive regulation occurs also in the post-transcriptional mRNA metabolism. Thereby, the particular composition of the mRNPs plays a central role as the components associated with the mRNA form a specific “mRNP code” which determines the fate of the mRNA. Many proteins which are involved in this regulation and the mRNA metabolism are affected in diseases and especially neurological disorders often result from an aberrant mRNP code which leads to changes in the regulation and expression of mRNPs.
The focus of this work was on a trimeric protein complex which is termed TTF complex based on its subunits TDRD3, TOP3β and FMRP. Biochemical investigations revealed that the three components of the TTF complex are nucleo-cytosolic shuttle proteins which localize in the cytoplasm at the steady-state, associate with mRNPs and are presumably connected to the translation. Upon cellular stress conditions, the TTF components concentrate in stress granules. Thus, the TTF complex is part of the mRNP code, however its target RNAs and function are still completely unknown. Since the loss of functional FMRP results in the fragile X syndrome and TOP3β is associated with schizophrenia and intellectual disability, the TTF complex connects these phenotypically related neuro-psychiatric disorders with each other on a molecular level.
Therefore, the aim of this work was to biochemically characterize the TTF complex and to define its function in the mRNA metabolism. In this work, evidence was provided that TDRD3 acts as the central unit of the TTF complex and directly binds to FMRP as well as to TOP3β. Thereby, the interaction of TDRD3 and TOP3β is very stable, whereas FMRP is a dynamic component. Interestingly, the TTF complex is not bound directly to mRNA, but is recruited via the exon junction complex (EJC) to mRNPs. This interaction is mediated by a specific binding motif of TDRD3, the EBM. Upon biochemical and biological investigations, it was possible to identify the interactome of the TTF complex and to define the role in the mRNA metabolism. The data revealed that the TTF complex is mainly associated with “early” mRNPs and is probably involved in the pioneer round of translation. Furthermore, TOP3β was found to bind directly to the ribosome and thus, establishes a connection between the EJC and the translation machinery. A reduction of the TTF components resulted in selective changes in the proteome in cultured cells, whereby individual protein subsets seem to be regulated rather than the global protein expression.
Moreover, the enzymatic analysis of TOP3β indicated that TOP3β is a type IA topoisomerase which can catalytically attack not only DNA but also RNA. This aspect is particularly interesting with regard to the connection between early mRNPs and the translation which has been revealed in this work.
The data obtained in this work suggest that the TTF complex plays a role in regulating the metabolism of an early mRNP subset possibly in the course of the pioneer round of translation. Until now, the link between an RNA topoisomerase and the mRNA metabolism is thereby unique and thus provides a completely new perspective on the steps in the post-transcriptional gene expression and its regulation.
The degradation of poly-adenosine tails of messenger RNAs (mRNAs) in the eukaryotic cells is a determining step in controlling the level of gene expression. The highly conserved Ccr4-Not complex was identified as the major deadenylation complex in all eukaryotic organisms. Plenty of biochemical studies have shown that this complex is also involved in many aspects of the mRNA metabolism, but we are still lacking the detailed structural information about its overall architecture and conformational states that could help to elucidate its multifunction and the way it is coordinated in the cells. Such information can also provide a basis to finding a possible way of intervention since the complex is also involved in some diseases such as cancer and cardiovascular disorders in humans. Meanwhile, the single particle Cryo-EM method has been through a “resolution revolution” recently due to the use of the newly developed direct electron detectors and has since resolved the high-resolution structures of many macromolecular protein complexes in their near-native state. Therefore, it was employed as a suitable method for studying the Ccr4-Not complex here. In this work, the Falcon 3EC direct detector mounted on the 300kV Titan Krios G3i Cryo-EM was evaluated for its practical performance at obtaining high-quality Cryo-EM data from protein samples of different molecular sizes. This served as a proof of principle for this detector’s capabilities and as a data collection guidance for studying the macromolecular complexes, such as the Ccr4-Not, when using an advanced high-performance microscope system. Next, the endogenous yeast Ccr4-Not complex was also purified via the immunoaffinity purification method and evaluated using negative staining EM to assess the conditions of the complex before proceeding to sample preparation for Cryo-EM. This has shown that the complex had an unexpected inherently dynamic property in vitro and extra optimisation procedures were needed to stabilise the complex during the purification and sample preparation. In addition, by using the label-free quantitative Mass spectrometry to examine the coimmunoprecipitated complex via different tagged subunits, it was deduced that two of the subunits (Not3/Not5) that shared some sequence similarity might compete for association with the scaffold subunit of the complex. An uncharacterised protein was also identified coimmunoprecipitating with the Caf130 subunit of the yeast complex. Cryo-EM data from the purified complex provided a low-resolution map that represents a surprisingly smaller partial complex as compared to 3D structures from previous studies, although gel electrophoresis and Mass spectrometry data have identified all of the nine subunits of the Ccr4-Not core complex in the sample. It was concluded that due to the presence of many predicted unstructured regions VI in the subunits and their dynamic composition in solution, the native complex could have been spontaneously denatured at the air/water interface during the sample preparation thus limiting the resolution of the Cryo-EM reconstruction. The purified complex was also examined for its deadenylase and ubiquitin ligase activity by in vitro assays. It was shown that the native complex has a different rate of activity and possibly also a different mode of action compared to the recombinant complexes from other species under similar reaction conditions. The Not4 E3 ligase was also shown to be active in the complex and was likely auto-ubiquitinated in the absence of a substrate. Both types of assays have also shown that the conformational flexibility does not seem to affect the enzymatic reactions when using a chemically crosslinked form of the complex for the assay, which implies that there can be other underlying mechanisms coordinating its structural and functional relationship. The findings from this work have therefore moved our understanding of the Ccr4-Not complex forward by looking at the different structural and functional behaviours of the endogenous complex, especially highlighting the obstacles in sample preparation for the native complex in high-resolution Cryo-EM. This would serve as foundation for future studies on the mechanism of this complex’s catalytic functions and also for optimising the Cryo-EM sample to generate better data that could eventually resolve the structure to a high-resolution.
The biogenesis of spliceosomal UsnRNPs is a highly elaborate cellular process that occurs both in the nucleus and the cytoplasm. A major part of the process is the assembly of the Sm-core particle, which consists of a ring shaped heptameric unit of seven Sm proteins (SmD1•D2•F•E•G•D3•B) wrapped around a single stranded RNA motif (termed Sm-site) of spliceosomal UsnRNAs. This process occurs mainly in the cytoplasm by the sequential action of two biogenesis factors united in PRMT5- and SMN-complexes, respectively. The PRMT5-complex composed of the three proteins PRMT5, WD45 and pICln is responsible for the symmetric dimethylation of designated arginine residues in the C-terminal tails of some Sm proteins. The action of the PRMT5- complex results in the formation of assembly incompetent Sm-protein intermediates sequestered by the assembly chaperone pICln (SmD1•D2•F•E•G•pICln and pICln•D3•B). Due to the action of pICln, the Sm proteins in these complexes fail to interact with UsnRNAs to form the mature Sm-core. This kinetic trap is relieved by the action of the SMN-complex, which removes the pICln subunit and facilitates the binding of the Sm-core intermediates to the UsnRNA, thus forming the mature Sm-core particle. The human SMN complex consists of 9 subunits termed SMN, Gemin2-8 and Unrip. So far, there are no available atomic structures of the whole SMN-complex, but structures of isolated domains and subunits of the complex have been reported by several laboratories in the past years. The lack of structural information about the entire SMN complex most likely lies in the biophysical properties of the SMN complex, which possesses an oligomeric SMN core, and many unstructured and flexible regions. These were the biggest roadblocks for its structural elucidation using traditional methods such as X-ray crystallography, NMR or CryoEM. To circumvent these obstacles and to obtain structural insight into the SMN-complex, the Schizosaccharomyces pombe SMN complex was used as a model system in this work. In a collaboration with the laboratory of Dr. Remy Bordonne (IGMM, CNRS, France), we could show that the SpSMN complex is minimalistic in its composition, consisting only of SpSMN, SpGemin2, SpGemin8, SpGemin7 and SpGemin6. Using biochemical experiments, an interaction map of the SpSMN complex was established which was found to be highly similar to the reported map of the human SMN complex. The results of this study clearly show that SpSMN is the oligomeric core of the complex and provides the binding sites for the rest of the subunits. Through biochemical and X-ray scattering experiments, the properties of the SpSMN subunit such as oligomerization viii and intrinsic disorder, were shown to determine the overall biophysical characteristics of the whole complex. The structural basis of SpSMN oligomerization is presented in atomic detail which establishes a dimeric SpSMN as the fundamental unit of higher order SpSMN oligomers. In addition to oligomerization, the YG-box domain of SpSMN serves as the binding site for SpGemin8. The unstructured region of SpSMN imparts an unusual large hydrodynamic size, intrinsic disorder, and flexibility to the whole complex. Interestingly, these biophysical properties are partially mitigated by the presence of SpGemin8•SpGemin7•SpGemin6 subunits. These results classify the SpSMN complex as a multidomain entity connected with flexible linkers and characterize the SpSMN subunit to be the central oligomeric structural organizer of the whole complex.
G-quadruplex structures are highly stable alternative DNA structures that can, when not properly regulated, impede replication fork progression and cause genome instability (Castillo Bosch et al, 2014; Crabbe et al, 2004; Koole et al, 2014; Kruisselbrink et al, 2008; London et al, 2008; Lopes et al, 2011; Paeschke et al, 2013; Paeschke et al, 2011; Piazza et al, 2015; Piazza et al, 2010; Piazza et al, 2012; Ribeyre et al, 2009; Sabouri et al, 2014; Sarkies et al, 2012; Sarkies et al, 2010; Schiavone et al, 2014; Wu & Spies, 2016; Zimmer et al, 2016). The aim of this thesis was to identify novel G-quadruplex interacting proteins in Saccharomyces cerevisiae and to unravel their regulatory function at these structures to maintain genome integrity. Mms1 and Rtt101 were identified as G-quadruplex binding proteins in vitro via a pull-down experiment with subsequent mass spectrometry analysis. Rtt101, Mms1 and Mms22, which are all components of an ubiquitin ligase (Rtt101Mms1/Mms22), are important for the progression of the replication fork following fork stalling (Luke et al, 2006; Vaisica et al, 2011; Zaidi et al, 2008). The in vivo binding of endogenously tagged Mms1 to its target regions was analyzed genome-wide using chromatin-immunoprecipitation followed by deep-sequencing. Interestingly, Mms1 bound independently of Mms22 and Rtt101 to G-rich regions that have the potential to form G-quadruplex structures. In vitro, formation of G-quadruplex structures could be shown for the G-rich regions Mms1 bound to. This binding was observed throughout the cell cycle. Furthermore, the deletion of MMS1 caused replication fork stalling as evidenced by increased association of DNA Polymerase 2 at Mms1 dependent sites. A gross chromosomal rearrangement assay revealed that deletion of MMS1 results in a significantly increased genome instability at G-quadruplex motifs compared to G-rich or non-G-rich regions. Additionally, binding of the helicase Pif1, which unwinds G4 structures in vitro (Paeschke et al, 2013; Ribeyre et al, 2009; Sanders, 2010; Wallgren et al, 2016), to Mms1 binding sites was reduced in mms1 cells. The data presented in this thesis, together with published data, suggests a novel mechanistic model in which Mms1 binds to G-quadruplex structures and enables Pif1 association. This allows for replication fork progression and genome integrity.
Using viruses to treat cancer is a novel approach to an age-old disease. Oncolytic viruses are native or recombinant viruses that have the innate or enhanced capability to infect tumour cells, replicate within the tumour microenvironment and subsequently lyse those cells. One representative, the vaccinia virus (VACV), belongs to the orthopoxvirus genus of the Poxviridae family. GLV-1h68, a recombinant and attenuated vaccinia virus devel- oped by the Genelux Corporation, is a member of this family currently being tested in various phase I/II clinical trials under the name GL-ONC1. It has been shown to specif- ically replicate in tumour cells while sparing healthy tissue and to metabolise prodrug at or transport immunological payloads to the site of affliction. Since imaging modalities offer little insight into viral replication deep within the body, and because oncolytic virotherapy is dependent on replication within the target tissue, the need for a monitoring system is evident. Pharmacokinetic analysis of this oncolytic agent was to give insight into the dynamics present in tumours during treatment. This, in turn, would give clinicians the opportunity to monitor the efficacy as early as possible after the onset of treatment, to observe treatment progression and possibly to gauge prognosis, without resorting to invasive procedures, e.g. biopsies. A criteria for viable biomarkers was that it had to be directly dependent on viral replica- tion. Ideally, a marker for treatment efficacy would be specific to the treatment modality, not necessarily the treatment type. Such a marker would be highly detectable (high sen- sitivity), specific for the treatment (high specificity), and present in an easily obtained specimen (blood). Taking this into consideration, the biomarkers were chosen for their potential to be indicators of viral replication. Thus, the biomarkers analysed in this thesis are: the native proteins expressed by the viral genes A27L and B5R, the virally encoded recombinant proteins β-galactosidase, β-glucuronidase, green fluorescent protein (GFP), carboxypeptidase G2 (CPG2) and carcinoembryonic antigen (CEA). Each marker is under the control of one of five different promoters present. All recombinant viruses used in this thesis express A27L, B5R, GFP and β-glucuronidase and all are derived from the parental virus GLV-1h68. In addition to these markers, GLV-1h68 expresses β-galactosidase; GLV-1h181 expresses CPG2. [...]
Most protein-encoding genes in Eukaryotes are separated into alternating coding and non-coding sequences (exons and introns). Following the transcription of the DNA into pre-messenger RNA (pre-mRNA) in the nucleus, a macromolecular complex termed spliceosome removes the introns and joins the exons to generate mature mRNA that is exported to the cytoplasm. There, it can be interpreted by ribosomes to generate proteins. The spliceosome consists of five small nuclear ribonucleic acids (snRNAs) and more than 150 proteins. Integral components of this complex are RNA-protein particles (RNPs) composed of one or two snRNAs, seven common (Sm) and a various number of snRNP-specific proteins. The Sm proteins form a ring-structure around a conserved site of the snRNA called Sm site. In vitro, Sm proteins (B/B', D1, D2, D3, E, F, G) and snRNA readily assemble to form snRNPs. In the context of the cell, however, two macromolecular trans-acting factors, the PRMT5 (protein arginine methyltransferases type 5) and the SMN (survival motor neuron) complex, are needed to enable this process. Initially, the Sm proteins in the form of heterooligomers D1/D2, D3/B and F/E/G are sequestered by the type II methyltransferase PRMT5. pICln, a component of the PRMT5 complex, readily interacts with Sm proteins to form two distinct complexes. Whereas the first one comprises pICln and D3/B the second one forms a ring consisting of pICln, D1/D2 and F/E/G (6S). It has been found that pICln prevents the premature interaction of snRNAs with the Sm proteins in these complexes and thus functions as an assembly chaperone imposing a kinetic trap upon the further assembly of snRNPs. PRMT5 catalyzes the symmetrical dimethylation of arginine residues in B/B', D1 and D3 increasing their affinity towards the SMN complex. Finally, the SMN complex interacts with the pICln-Sm protein complexes, expels pICln and mediates snRNP assembly in an ATP-dependent reaction. So far, only little is known about the action of PRMT5 in the early phase of snRNP assembly and especially how the 6S complex is formed. Studies of this have so far been hampered by the unavailability of soluble and biologically active PRMT5 enzyme. The composition of the SMN complex and possible functions of individual subunits have been elucidated or hypothesized in recent years. Still, the exact mechanism of the entire machinery forming snRNPs is poorly understood. In vivo, reduced production of functional SMN protein results in the neurodegenerative disease spinal muscular atrophy (SMA). How specific SMN mutations that have been found in SMA patients cause the disease remains elusive, yet, are likely to interfere with either SMN complex stability or snRNP assembly. The aim of this work was to establish an in vitro system to recapitulate the cytoplasmic assembly of snRNPs. This was enabled by the recombinant production of all PRMT5 and SMN complex components as well as Sm proteins in a combination of bacterial and insect cell expression systems. Co-expression of human PRMT5 and its direct interaction partner WD45 (WD-repeat domain 45) in Sf21 (Spodoptera frugiperda 21) insect cells resulted for the first time in soluble and biologically active enzyme. Recombinant PRMT5/WD45 formed complexes with Sm protein heterooligomers as well as pICln-Sm protein complexes but not with F/E/G alone. Also, the enzyme exhibited a type II methyltransferase activity catalyzing the mono- (MMA) and symmetrical dimethylation (sDMA) of Sm proteins B, D1 and D3. Two experimental setups were devised to quantitatively analyze the overall methylation of substrates as well as to identify the type and relative abundance of specific methylation types. Methylation of Sm proteins followed Michaelis-Menten kinetics. Complex reconstitutions and competition of the methylation reaction indicate that 6S is formed in a step-wise manner on the PRMT5 complex. The analysis of the methylation type could be applied to deduce a model of sequential MMA and sDMA formation. It was found that large Sm protein substrate concentrations favored monomethylation. Following a distributive mechanism this leads to the conclusion that PRMT5 most likely confers partial methylation of several different substrate proteins instead of processing a single substrate iteratively until it is completely dimethylated. Finally, the human SMN complex was reconstituted from recombinant sources and was shown to be active in snRNP formation. The introduction of a modified SMN protein carrying a mutation (E134K) present in spinal muscular atrophy (SMA) proved that mutated complexes can be generated in vitro and that these might be applied to elucidate the molecular etiology of this devastating disease.
Formation oft the central nervous system (CNS) from multipotent neuronal stem cells (NSCs) requires a tightly controlled, step-wise activation of the neuronal gene expression program. Expression of neuronal genes at the transition from neural stem cell to mature neuron (i. e. neuronal cell differentiation) is controlled by the Repressor element 1 (RE1) silencing transcription factor (REST) complex. As a master transcriptional regulator, the REST-complex specifically inhibits expression of neuronal genes in non-neuronal tissues and neuronal progenitor cells. Differentiation of NSCs to mature neurons requires the activation of genes controlled by the REST-complex, but how abrogation of REST-complex mediated repression is achieved during neurogenesis is only poorly understood. MicroRNAs (miRNAs) are a class of small regulatory RNAs that posttranscriptionally control target gene expression. Binding of miRNAs to target sequences in the 3’UTR of mRNAs, leads either to degradation or translational inhibition of the mRNA. Distinct neuronal miRNAs (e.g. miR-124) were shown to modulate REST-complex activity by silencing expression of REST-complex components. Interestingly, these miRNAs are also under transcriptional control of the REST-complex and inactivation of the REST-complex precedes their expression. Hence, additional factors are required for derepression of neuronal genes at the onset of neurogenesis. In this study function of the miR-26 family during neurogenesis of the zebrafish (Danio rerio) was analyzed. Computational target prediction revealed a number of REST-complex components as putative miR-26 targets. One of these predicted target genes, the C-terminal domain small phosphatase 2 (Ctdsp2) was validated as an in vivo target for miR-26b. Ctdsps are important cofactors of REST and suppress neuronal gene expression by dephosphorylating the C-terminal domain (CTD) of RNA polymerase II (Pol II). Interestingly, miR-26b is encoded in an intron of the ctdsp2 primary transcript and is cotranscribed together with its host gene. Hence, miR-26b modulates expression of its host gene ctdsp2 in an intrinsic negative autoregulatory loop. This negative autoregulatory loop is inactive in NSCs because miR-26b biogenesis is inhibited at the precursor level. Generation of mature miR-26b is activated during neurogenesis, where it suppresses Ctdsp2 protein expression and is required for neuronal cell differentiation in vivo. Strikingly, miR-26b is expressed prior to miR-124 during neuronal cell differentiation. Thus, it is reasonable to speculate about a function of miR-26b in early events of neurogenesis. In line with this assumption, knockdown of miR-26b in zebrafish embryos results in downregulation of REST-complex controlled neuronal genes and a block in neuronal cell differentiation, most likely due to aberrant regulation of Ctdsp2 expression. This is evident by reduced numbers of secondary motor neurons compared to control siblings. In contrast, motor neuron progenitor cells and glia cells were not affected by depletion of miR-26b.This study identifies the ctdsp2/miR-26b autoregulatory loop as the first experimentally validated interaction between an intronic miRNA and its host gene transcript. Silencing of ctdsp2 by miR-26b in neurons is possible because biogenesis of the ctdsp2 mRNA and mature mir-26b is uncoupled at the posttranscriptional level. Furthermore the obtained data indicate a cell type specific role for miR-26b in vertebrate neurogenesis and CNS development.
In initial experiments, the well characterized VACV strain GLV-1h68 and three wild-type LIVP isolates were utilized to analyze gene expression in a pair of autologous human melanoma cell lines (888-MEL and 1936 MEL) after infection. Microarray analyses, followed by sequential statistical approaches, characterized human genes whose transcription is affected specifically by VACV infection. In accordance with the literature, those genes were involved in broad cellular functions, such as cell death, protein synthesis and folding, as well as DNA replication, recombination, and repair. In parallel to host gene expression, viral gene expression was evaluated with help of customized VACV array platforms to get better insight over the interplay between VACV and its host. Our main focus was to compare host and viral early events, since virus genome replication occurs early after infection. We observed that viral transcripts segregated in a characteristic time-specific pattern, consistent with the three temporal expression classes of VACV genes, including a group of genes which could be classified as early-stage genes. In this work, comparison of VACV early replication and respective early gene transcription led to the identification of seven viral genes whose expression correlated strictly with replication. We considered the early expression of those seven genes to be representative for VACV replication and we therefore referred to them as viral replication indicators (VRIs). To explore the relationship between host cell transcription and viral replication, we correlated viral (VRI) and human early gene expression. Correlation analysis revealed a subset of 114 human transcripts whose early expression tightly correlated with early VRI expression and thus early viral replication. These 114 human molecules represented an involvement in broad cellular functions. We found at least six out of 114 correlates to be involved in protein ubiquitination or proteasomal function. Another molecule of interest was the serine-threonine protein kinase WNK lysine-deficient protein kinase 1 (WNK1). We discovered that WNK1 features differences on several molecular biological levels associated with permissiveness to VACV infection. In addition to that, a set of human genes was identified with possible predictive value for viral replication in an independent dataset. A further objective of this work was to explore baseline molecular biological variances associated with permissiveness which could help identifying cellular components that contribute to the formation of a permissive phenotype. Therefore, in a subsequent approach, we screened a set of 15 melanoma cell lines (15-MEL) regarding their permissiveness to GLV-1h68, evaluated by GFP expression levels, and classified the top four and lowest four cell lines into high and low permissive group, respectively. Baseline gene transcriptional data, comparing low and highly permissive group, suggest that differences between the two groups are at least in part due to variances in global cellular functions, such as cell cycle, cell growth and proliferation, as well as cell death and survival. We also observed differences in the ubiquitination pathway, which is consistent with our previous results and underlines the importance of this pathway in VACV replication and permissiveness. Moreover, baseline microRNA (miRNA) expression between low and highly permissive group was considered to provide valuable information regarding virus-host co-existence. In our data set, we identified six miRNAs that featured varying baseline expression between low and highly permissive group. Finally, copy number variations (CNVs) between low and highly permissive group were evaluated. In this study, when investigating differences in the chromosomal aberration patterns between low and highly permissive group, we observed frequent segmental amplifications within the low permissive group, whereas the same regions were mostly unchanged in the high group. Taken together, our results highlight a probable correlation between viral replication, early gene expression, and the respective host response and thus a possible involvement of human host factors in viral early replication. Furthermore, we revealed the importance of cellular baseline composition for permissiveness to VACV infection on different molecular biological levels, including mRNA expression, miRNA expression, as well as copy number variations. The characterization of human target genes that influence viral replication could help answering the question of host cell response to oncolytic virotherapy and provide important information for the development of novel recombinant vaccinia viruses with improved features to enhance replication rate and hence trigger therapeutic outcome.
Glioblastoma multiforme (GBM) represents the most aggressive form of malignant brain tumors and remains a therapeutically challenge. Intense research in the field has lead to the testing of oncolytic viruses to improve tumor control. Currently, a variety of different oncolytic viruses are being evaluated for their ability to be used in anti-cancer therapy and a few have entered clinical trials. Vaccinia virus, is one of the viruses being studied. GLV-1h68, an oncolytic vaccinia virus engineered by Genelux Corporation, was constructed by insertion of three gene cassettes, RUC-GFP fusion, β-galactosidase and β- glucuronidase into the genome of the LIVP strain. Since focal tumor radiotherapy is a mainstay for cancer treatment, including glioma therapy, it is of clinical relevance to assess how systemically administered oncolytic vaccinia virus could be combined with targeted ionizing radiation for therapeutic gain. In this work we show how focal ionizing radiation (IR) can be combined with multiple systemically delivered oncolytic vaccinia virus strains in murine models of human U-87 glioma. After initial experiments which confirmed that ionizing radiation does not damage viral DNA or alter viral tropism, animal studies were carried out to analyze the interaction of vaccinia virus and ionizing radiation in the in vivo setting. We found that irradiation of the tumor target, prior to systemic administration of oncolytic vaccinia virus GLV-1h68, increased viral replication within the U-87 xenografts as measured by viral reporter gene expression and viral titers. Importantly, while GLV-1h68 alone had minimal effect on U-87 tumor growth delay, IR enhanced GLV-1h68 replication, which translated to increased tumor growth delay and mouse survival in subcutaneous and orthotopic U-87 glioma murine models compared to monotherapy with IR or GLV-1h68. The ability of IR to enhance vaccinia replication was not restricted to the multi-mutated GLV-1h68, but was also seen with the less attenuated oncolytic vaccinia, LIVP 1.1.1. We have demonstrated that in animals treated with combination of ionizing radiation and LIVP 1.1.1 a strong pro-inflammatory tissue response was induced. When IR was given in a more clinically relevant fractionated scheme, we found oncolytic vaccinia virus replication also increased. This indicates that vaccinia virus could be incorporated into either larger hypo-fraction or more conventionally fractionated radiotherapy schemes. The ability of focal IR to mediate selective replication of systemically injected oncolytic vaccinia was demonstrated in a bilateral glioma model. In mice with bilateral U-87 tumors in both hindlimbs, systemically administered oncolytic vaccinia replicated preferentially in the focally irradiated tumor compared to the shielded non- irradiated tumor in the same mouse We demonstrated that tumor control could be further improved when fractionated focal ionizing radiation was combined with a vaccinia virus caring an anti-angiogenic payload targeting vascular endothelial growth factor (VEGF). Our studies showed that following ionizing radiation expression of VEGF is upregulated in U-87 glioma cells in culture. We further showed a concentration dependent increase in radioresistance of human endothelial cells in presence of VEGF. Interestingly, we found effects of vascular endothelial growth factor on endothelial cells were reversible by adding purified GLAF-1 to the cells. GLAF-1 is a single- chain antibody targeting human and murine VEGF and is expressed by oncolytic vaccinia virus GLV-109. In U-87 glioma xenograft murine models the combination of fractionated ionizing radiation with GLV-1h164, a vaccinia virus also targeting VEGF, resulted in the best volumetric tumor response and a drastic decrease in vascular endothelial growth factor. Histological analysis of embedded tumor sections 14 days after viral administration confirmed that blocking VEGF translated into a decrease in vessel number to 30% of vessel number found in control tumors in animals treated with GLV-164 and fractionated IR which was lower than for all other treatment groups. Our experiments with GLV-1h164 and fractionated radiotherapy have shown that in addition to ionizing radiation and viral induced tumor cell destruction we were able to effectively target the tumor vasculature. This was achieved by enhanced viral replication translating in increased levels of GLAF-2 disrupting tumor vessels as well as the radiosensitization of tumor vasculature to IR by blocking VEGF. Our preclinical results have important clinical implications of how focal radiotherapy can be combined with systemic oncolytic viral administration for highly aggressive, locally advanced tumors with the potential, by using a vaccinia virus targeting human vascular endothelial growth factor, to further increase tumor radiation sensitivity by engaging the vascular component in addition to cancer cells.
Effects of stem cell transcription factor-expressing vaccinia viruses in oncolytic virotherapy
(2012)
Cancer remains the second leading cause of death in the industrialized. The data from many different studies investigating the nature of cancer-initiating cells coined the description ‘cancer stem cells’ and has major implications on conventional cancer therapy. Thus, to improve the outcome of cancer treatment and to lower negative side effects, the development of novel therapeutic regimens is indispensable. It has been demonstrated in many preclinical studies that oncolytic virotherapy using vaccinia virus may provide a powerful and well-tolerable new tool in cancer therapy which is currently investigated in several clinical trials (Phase I & II) as stand-alone treatment or in combination with conventional cancer therapy. Cancer-initiating cells and stem cells share a variety of characteristics like the ability to self-renew, differentiation potential, quiescence, drug and radiation resistance, activation and inhibition of similar signaling pathways as well as expression of cell surface markers and stem cell-related genes. In this work, two new recombinant vaccinia viruses expressing the transcription factors Nanog (GLV-1h205) and Oct4 (GLV-1h208) were engineered to provide deeper insight of these stem cell master regulators in their significance of cancer-initiation and their impact on oncolytic virotherapy. Both viruses were analyzed for their replication potential in A549 and PC-3 human cancer cells. Marker gene expression was assessed by RT-PCR, SDS-PAGE and Western blotting, ELISA or immunocytochemistry.Furthermore, the effect of GLV-1h205 infection on the cell cycle in A549 cells was analyzed. Next, the effects of virus-mediated expression of stem cell transcription factors on therapeutic efficacy and survival rates in A549 xenograft mouse models was analyzed. A non-functional Nanog mutant-expressing virus strain (GLV-1h321) was engineered to analyze whether the observed therapeutic benefits were promoter- or payload-driven. Furthermore, this study analyzed the potential of GLV-1h68 to infect, replicate in, and lyse colorectal cancer cell lines to study whether oncolytic vaccinia viruses can be potential new and less invasive treatment regimens for late stage colorectal cancer. Marker gene expression was assessed by fluorescence microscopy and FACS. The transcription factor Klf4 is highly expressed in quiescent, terminally differentiated cells in the colonic epithelium whereas it is dramatically downregulated in colon cancers. Klf4 expression leads to cell growth arrest and inhibits Wnt signaling by binding to beta-catenin. To further improve the treatment of colorectal cancers, new recombinant vaccinia viruses (GLV-1h290-292) mediating the expression of differing amounts of the tumor suppressor Klf4 by using different promoter strengths were engineered. Initial characterization of recombinant vaccinia viruses expressing Klf4 by replication assay, cell viability assay, SDS-PAGE and Western blotting, immuncytochemistry and analysis of protein functionality by qPCR and ELISA analysis for cellular beta-catenin expression, demonstrated promoter strength-dependent expression of and impact of Klf4. To further boost the effects of tumor suppressor Klf4, a vaccinia virus strain expressing Klf4 with a C-terminal fusion of the TAT transduction domain (GLV-1h391) was engineered. Treatment of HT-29 non-responder tumors in vivo with GLV-1h291 and GLV-1h391 led to significant tumor growth inhibition and improved overall survival compared to GLV-1h68. This makes the Klf4-TAT expressing GLV-1h391 a promising candidate for the treatment of colorectal cancer in man.
Vaccinia virus plays an important role in human medicine and molecular biology ever since the 18th century after E. Jenner discovered its value as a vaccination virus against smallpox. After the successful eradication of smallpox, vaccinia virus, apart from its use as a vaccine carrier, is today mainly used as a viral vector in molecular biology and increasingly in cancer therapy. The capability to specifically target and destroy cancer cells makes it a perfect agent for oncolytic virotherapy. Furthermore, the virus can easily be modified by inserting genes encoding therapeutic or diagnostic proteins to be expressed within the tumor. The emphasis in this study was the diagnosis of tumors using different vaccinia virus strains. Viruses with metal-accumulating capabilities for tumor detection via MRI technology were generated and tested for their usefulness in cell culture and in vivo. The virus strains GLV-1h131, GLV-1h132, and GLV-1h133 carry the gene encoding the two subunits of the iron storage protein ferritin under the control of three different promoters. GLV-1h110, GLV-1h111, and GLV-1h112 encode the bacterial iron storage protein bacterioferritin, whereas GLV-1h113 encodes the codon-optimized version of bacterioferritin for more efficient expression in human cells. GLV-1h22 contains the transferrin receptor gene, which plays an important role in iron uptake, and GLV-1h114 and GLV-1h115 contain the murine transferrin receptor gene. For possibly better iron uptake the virus strains GLV-1h154, GLV-1h155, GLV-1h156, and GLV-1h157 were generated, each with a version of a ferritin gene and a transferrin receptor gene. GLV-1h154 carries the genes that encode bacterioferritin and human transferrin receptor, GLV-1h155 the human ferritin H-chain gene and the human transferrin receptor gene. GLV-1h156 and GLV-1h157 infected cells both express the mouse transferrin receptor and bacterioferritin or human ferritin H-chain, respectively. The virus strains GLV-1h186 and GLV-1h187 were generated to contain a mutated form of the ferritin light chain, which was shown to result in iron overload and the wildtype light chain gene, respectively. The gene encoding the Divalent Metal Transporter 1, which is a major protein in the uptake of iron, was inserted in the virus strain GLV-1h102. The virus strain GLV-1h184 contains the magA gene of the magnetotactic bacterium Magnetospirillum magnetotacticum, which produces magnetic nanoparticles for orientation in the earth’s magnetic field. Initially the infection and replication capability of all the virus strains were analyzed and compared to that of the parental virus strain GLV-1h68, revealing that all the viruses were able to infect cells of the human cancer cell lines A549 and GI-101A. All constructs exhibited a course of infection comparable to that of GLV-1h68. Next, to investigate the expression of the foreign proteins in GI-101A and A549 cells with protein analytical methods, SDS-gelelectrophoresis, Western blots and ELISAs were performed. The proteins, which were expressed under the control of the strong promoters, could be detected using these methods. To be able to successfully detect the protein expression of MagA and DMT1, which were expressed under the control of the weak promoter, the more sensitive method RT-PCR was used to at least confirm the transcription of the inserted genes. The determination of the iron content in infected GI-101A and A549 cells showed that infection with all used virus strains led to iron accumulation in comparison to uninfected cells, even infection with the parental virus strain GLV-1h68. The synthetic phytochelatin EC20 was also shown to enhance the accumulation of different heavy metals in bacterial cultures. In vivo experiments with A549 tumor-bearing athymic nude mice revealed that 24 days post infection virus particles were found mainly in the tumor. The virus-mediated expression of recombinant proteins in the tumors was detected successfully by Western blot. Iron accumulation in tumor lysates was investigated by using the ferrozine assay and led to the result that GLV-1h68-infected tumors had the highest iron content. Histological stainings confirmed the finding that iron accumulation was not a direct result of the insertion of genes encoding iron-accumulating proteins in the virus genome. Furthermore virus-injected tumorous mice were analyzed using MRI technology. Two different measurements were performed, the first scan being done with a seven Tesla small animal scanner seven days post infection whereas the second scan was performed using a three Tesla human scanner 21 days after virus injection. Tumors of mice injected with the virus strains GLV-1h113 and GLV-1h184 were shown to exhibit shortened T2 and T2* relaxation times, which indicates enhanced iron accumulation. In conclusion, the experiments in this study suggest that the bacterioferritin-encoding virus strain GLV-1h113 and the magA-encoding virus strain GLV-1h184 are promising candidates to be used for cancer imaging after further analyzation and optimization.
Macromolecular complexes, also termed molecular machines, facilitate a large spectrum of biological reactions and tasks crucial to the survival of cells. These complexes are composed of either protein only, or proteins bound to nucleic acids (DNA or RNA). Prominent examples for each class are the proteosome, the nucleosome and the ribosome. How such units are assembled within the context of a living cell is a central question in molecular biology. Earlier studies had indicated that even very large complexes such as ribosomes could be reconstituted from purified constituents in vitro. The structural information required for the formation of macromolecular complexes, hence, lies within the subunits itself and, thus, allow for self- assembly. However, increasing evidence suggests that in vivo many macromolecular complexes do not form spontaneously but require assisting factors (“assembly chaperones”) for their maturation. In this thesis the assembly of RNA-protein (RNP) complexes has been studied by a combination of biochemical and structural approaches. A resourceful model system to study this process is the biogenesis pathway of the uridine-rich small nuclear ribonucleoproteins (U snRNPs) of the spliceosome. This molecular machine catalyzes pre-mRNA splicing, i.e. the removal of non-coding introns and the joining of coding exons to functional mRNA. The composition and architecture of U snRNPs is well defined, also, the nucleo-cytoplasmic transport events enabling the formation of these particles in vivo have been analyzed in some detail. Furthermore, recent studies suggest that the formation of U snRNPs in vivo is mediated by an elaborate assembly machinery consisting of protein arginine methyltransferase (PRMT5)- and survival motor neuron (SMN)-complexes. The elucidation of the reaction mechanism of cellular U snRNP assembly would serve as a paradigm for our understanding of how RNA-protein complexes are formed in the cellular environment. The following key findings were obtained as part of this study: 1) Efforts were made to establish a full inventory of the subunits of the SMN-complex. This was achieved by the biochemical definition and characterization of an atypical component of this complex, the unrip protein. This protein is associated with the SMN-complex exclusively in the cytoplasm and influences its subcellular localization. 2) With a full inventory of the components in hand, the architecture of the SMN-complex was defined on the basis of an interaction map of all subunits. This study elucidated that the proteins SMN, Gemin7 and Gemin8 form a backbone, onto which the remaining subunits adhere in a modular manner. 3) The two studies mentioned above formed the basis to elucidate the reaction mechanism of cellular U snRNP assembly. Initially, an early phase in the SMN-assisted formation of U snRNPs was analyzed. Two subunits of the U7 snRNP (LSm10 and 11) were found to interact with the PRMT5-complex, without being methylated. This report suggests that the stimulatory role of the PRMT5-complex is independent of its methylation activity. 4) Key reaction intermediates in U snRNP assembly were found and characterized by a combination of biochemistry and structural studies. Initially, a precursor to U snRNPs with a sedimentation coefficient of 6S is formed by the pICln subunit of the PRMT5-complex and Sm proteins. This intermediate was shown to constitute a kinetic trap in the U snRNP assembly reaction. Progression towards the assembled U snRNP depends on the activity of the SMN-complex, which acts as a catalyst. The formation of U snRNPs is shown to be structurally similar to the way clamps are deposited onto DNA to tether poorly processive polymerases. 5) The human SMN-complex is composed of several subunits. However, it is unknown whether all subunits of this entity are essential for U snRNP assembly. A combination of bioinformatics and biochemistry was applied to tackle this question. By mining databases containing whole-genome assemblies, the SMN-Gemin2 heterodimer is recognized as the most ancestral form of the SMN-complex. Biochemical purification of the Drosophila melanogaster SMN-complex reveals that this complex is composed of the same two subunits. Furthermore, evidence is provided that the SMN-Gemin2 heterodimer is necessary and sufficient to promote faithful U snRNP assembly. Future studies will adress further details in the reaction mechanism of cellular U snRNP assembly. The results obtained in this thesis suggest that the SMN and Gemin2 subunits are sufficient to promote U snRNP formation. What then is the function of the remaining subunits of the SMN-complex? The reconstitution schemes established in this thesis will be instrumental to address this question. Furthermore, additional mechanistic insights into the U snRNP assembly reaction will require the elucidation of structures of the assembly machinery trapped at various states. The prerequisite for these structural studies, the capability to generate homogenous complexes in sufficient amounts, has been accomplished in this thesis.
Spinal muscular atrophy with respiratory distress type 1 (SMARD1) is an autosomal recessive neuronal disorder in infants. The disease is marked by early onset of respiratory distress and predominantly distal muscle weakness, as consequences of diaphragmatic paralysis and progressive degeneration of  motor neurons in the spinal cord, respectively. Genetically, SMARD1 is caused by mutations in the single gene encoding Immunoglobulin µ-Binding Protein 2 (IGHMBP2). Despite the tissue specific degeneration observed in SMARD1 patients, the disease gene product IGHMBP2 is ubiquitously expressed in human and mouse tissues. Therefore, SMARD1 appears to be a motor neuron disease caused by the malfunction of a “housekeeping” protein, rather than a neuron specific factor. IGHMBP2 harbors an N-terminal DEXDc-type helicase/ATPase domain and has been classified as a member of the Superfamily 1 (SF1) of helicases. This protein has been assigned to various cellular activities such as DNA replication, pre-mRNA splicing and transcription. However its precise function in either process has remained elusive. The study presented here aimed at the enzymatic characterization of IGHMBP2, the identification of a specific cellular process to which IGHMBP2 is connected and the role of this factor in the pathophysiology of SMARD1. As a first step toward this end, a two-step purification strategy was established, which enabled the large-scale purification of properly folded and enzymatically active IGHMBP2. In vitro enzymatic studies using this recombinant protein defined IGHMBP2 as an ATP-dependent helicase that catalyzes unwinding of duplices composed of either DNA or RNA in a 5’→3’ direction. In contrast to previous reports, indirect immunofluorescence studies revealed a predominantly cytoplasmic localization of IGHMBP2. Size-fractionation studies and affinity-purification experiments further showed that IGHMBP2 is part of an RNase-sensitive macromolecular complex, which was identified as the ribosome. Interestingly, IGHMBP2 was abundantly detected in both subunits as well as to 80S ribosomes but only in small amounts in actively translating polysomes. These data strongly point to a role of IGHMBP2 in ribosomes-associated gene regulation control, such as in mRNA stabilization or mRNA translation. However, its precise function in those pathways remains to be identified. The biochemical and enzymatic characterization of IGHMBP2 allowed for the first time insights into the pathomechanism of SMARD1. SMARD1-causing pathogenic IGHMBP2 variants were investigated for their enzymatic activities and interaction with ribosomal subunits. Interestingly, among all missense mutations that have been tested thus far, none obstructs association with ribosomal subunits. However, these mutants exhibit specific defects in either the ATPase or RNA helicase activity or both. The data suggest that defects in the enzymatic activity of IGHMBP2 directly correlate with the pathogenesis of SMARD1. Furthermore, these data also raise the possibility that the disease SMARD1 is caused by alterations in the cellular translation machinery.
Directed cortical actin assembly is the driving force for intercellular adhesion. Vasodilator-stimulated phosphoprotein (VASP) participates in actin-fiber formation and VASP activity is regulated by phosphorylations. We screened for endothelial cell proteins, which bind to VASP dependent on its phosphorylation status. Differential proteomics identified αII-spectrin as novel VASP-interacting protein. αII-spectrin binds to the triple GP5-motif in VASP via its SH3 domain. cAMP-dependent protein kinase-mediated VASP phosphorylation at Ser157 inhibits αII-spectrin/VASP complex formation. VASP becomes dephosphorylated upon formation of cell-cell contacts and in confluent but not in sparse endothelial cells αII-spectrin colocalizes with non-phosphorylated VASP at cell-cell junctions. Ectopic expression of the αII-spectrin SH3 domain fused to claudin-5 translocates VASP to cell-cell contacts and is sufficient to initiate the formation of cortical actin cytoskeletons. αII-spectrin SH3 domain overexpression stabilizes cell-cell contacts and decreases endothelial permeability. Conversely, permeability of VASP-deficient endothelial cells is elevated. In a skin edema model, microvascular leakage is increased in VASP-deficient over wild-type mice. We propose that αII-spectrin/VASP complexes regulate cortical actin cytoskeleton assembly with implications for formation of endothelial cell-cell contacts and regulation of vascular permeability.
In physiological conditions platelets have a major role in maintaining haemostasis. Platelets prevent bleeding from wounds by distinguishing normal endothelial cells in vasculature from areas with lesions to which they adhere. Interaction of platelet agonists and their receptors is controlled by intracellular signaling molecules that regulate the activation state of platelets. Very important intracellular signaling molecules are cyclic nucleotides (cGMP and cAMP), both involved in inhibition of platelet activation. Formation of cGMP and cAMP in platelets is stimulated by endothelial-derived NO and prostacyclin (PGI2), which then mediate inhibition of platelets by activating protein kinase G (PKG) and protein kinase A (PKA). Recently, it has been suggested that reactive oxygen species (ROS) represent new modulators of cell signaling within different cell types. The work summarized here describes the involvement of platelet ROS production in platelet activation, the relation of NO/cGMP/PKG I pathway to ROS and to mitogen-activated protein kinases (MAP kinase) signaling, and the involvement of cyclic nucleotides in megakaryocyte and platelet development. Platelets activated with different agonists produce intracellular but not extracellular ROS by activation of NAD(P)H oxidase. In addition, ROS produced in platelets significantly affects αIIbβ3 integrin activation but not alpha/dense granule secretion and platelet shape change. Thrombin induced integrin αIIbβ3 activation is significantly decreased after pretreatment of platelets with NAD(P)H oxidase inhibitors and superoxide scavengers. These inhibitors also reduce platelet aggregation and thrombus formation on collagen under high shear and achieve their effects independently of the NO/cGMP pathway. ADP secreted from platelet dense granules with subsequent activation of P2Y12 receptors as well as thromboxane A2 release are found to be important upstream mediators of p38 MAP kinase activation by thrombin. However, p38 MAP kinase activation does not significantly contribute to calcium mobilization, P-selectin expression, αIIbβ3 integrin activation and aggregation of human platelets in response to thrombin. Finally, PKG activation does not stimulate, but rather inhibit, p38 and ERK MAP kinases in human platelets. Further study revealed that cyclic nucleotides not only inhibit platelet activation, but are also involved, albeit differentially, in megakaryocyte and platelet development. cAMP is engaged in haematopoietic stem cell differentiation to megakaryocytes, and cGMP has no impact on this process. While PKA is already present in stem cells, expression of proteins involved in cGMP signaling (soluble guanylyl cyclase, sGC; PKG) increases with maturation of megakaryocytes. In the final step of megakaryocyte maturation that includes release of platelets, cGMP and cAMP have mild but opposing effects: cGMP increases platelet production while cAMP decreases it indicating a finely regulated process that could depend on stimulus coming from adjacent endothelial cells of sinusoids in bone marrow. The results of this thesis contribute to a better understanding of platelet regulation and of the possible molecular mechanisms involved in megakaryocyte maturation in bone marrow vascular microenvironment.