Universitätspublikationen
Refine
Year of publication
Document Type
- Article (651)
- Doctoral Thesis (154)
- Preprint (32)
- Conference Proceeding (11)
- Part of a Book (2)
- Other (1)
Language
- English (851) (remove)
Has Fulltext
- yes (851) (remove)
Is part of the Bibliography
- no (851)
Keywords
- crystal structure (37)
- hydrogen bonding (11)
- NMR spectroscopy (7)
- RNA (7)
- structural biology (7)
- Optogenetics (6)
- Biochemistry (5)
- Bioenergetics (5)
- Membrane proteins (5)
- NMR (5)
Institute
- Biochemie und Chemie (851) (remove)
Dynamic imaging of landmark organelles, such as nuclei, cell membrane, nuclear envelope, and lipid droplets enables image-based phenotyping of functional states of cells. Multispectral fluorescent imaging of landmark organelles requires labor-intensive labeling, limits throughput, and compromises cell health. Virtual staining of label-free images with deep neural networks is an emerging solution for this problem. Multiplexed imaging of cellular landmarks from scattered light and subsequent demultiplexing with virtual staining saves the light spectrum for imaging additional molecular reporters, photomanipulation, or other tasks. Published approaches for virtual staining of landmark organelles are fragile in the presence of nuisance variations in imaging, culture conditions, and cell types. This paper reports model training protocols for virtual staining of nuclei and membranes robust to cell types, cell states, and imaging parameters. We developed a flexible and scalable convolutional architecture, named UNeXt2, for supervised training and self-supervised pre-training. The strategies we report here enable robust virtual staining of nuclei and cell membranes in multiple cell types, including neuromasts of zebrafish, across a range of imaging conditions. We assess the models by comparing the intensity, segmentations, and application-specific measurements obtained from virtually stained and experimentally stained nuclei and membranes. The models rescue the missing label, non-uniform expression of labels, and photobleaching. We share three pre-trained models, named VSCyto3D, VSCyto2D, and VSNeuromast, as well as VisCy, a PyTorch-based pipeline for training, inference, and deployment that leverages the modern OME-Zarr format.
Dynamic imaging of landmark organelles, such as nuclei, cell membrane, nuclear envelope, and lipid droplets enables image-based phenotyping of functional states of cells. Multispectral fluorescent imaging of landmark organelles requires labor-intensive labeling, limits throughput, and compromises cell health. Virtual staining of label-free images with deep neural networks is an emerging solution for this problem. Multiplexed imaging of cellular landmarks from scattered light and subsequent demultiplexing with virtual staining saves the light spectrum for imaging additional molecular reporters, photomanipulation, or other tasks. Published approaches for virtual staining of landmark organelles are fragile in the presence of nuisance variations in imaging, culture conditions, and cell types. This paper reports model training protocols for virtual staining of nuclei and membranes robust to label-free imaging parameters, cell states, and cell types. We developed a flexible and scalable convolutional architecture, named UNeXt2, for supervised training and self-supervised pre-training. The strategies we report here enable robust virtual staining of nuclei and cell membranes in multiple cell types, including neuromasts of zebrafish, across a range of imaging conditions. We assess the models by comparing the intensity, segmentations, and application-specific measurements obtained from virtually stained and experimentally stained nuclei and membranes. The models rescue the missing label, non-uniform expression of labels, and photobleaching. We share three pre-trained models, named VSCyto3D, VSCyto2D, and VSNeuromast, as well as VisCy, a PyTorch-based pipeline for training, inference, and deployment that leverages the modern OME-Zarr format.
The combination of high-throughput sequencing and in vivo crosslinking approaches leads to the progressive uncovering of the complex interdependence between cellular transcriptome and proteome. Yet, the molecular determinants governing interactions in protein-RNA networks are not well understood. Here we investigated the relationship between the structure of an RNA and its ability to interact with proteins. Analysing in silico, in vitro and in vivo experiments, we find that the amount of double-stranded regions in an RNA correlates with the number of protein contacts. This relationship —which we call structure-driven protein interactivity— allows classification of RNA types, plays a role in gene regulation and could have implications for the formation of phase-separated ribonucleoprotein assemblies. We validate our hypothesis by showing that a highly structured RNA can rearrange the composition of a protein aggregate. We report that the tendency of proteins to phase-separate is reduced by interactions with specific RNAs.
Ribosome recycling orchestrated by ABCE1 is a fundamental process in protein translation and mRNA surveillance, connecting termination with initiation. Beyond the plenitude of well-studied translational GTPases, ABCE1 is the only essential factor energized by ATP, delivering the energy for ribosome splitting via two nucleotide-binding sites by a yet unknown mechanism. Here, we define how allosterically coupled ATP binding and hydrolysis events in ABCE1 empower ribosome recycling. ATP occlusion in the low-turnover control site II promotes formation of the pre-splitting complex and facilitates ATP engagement in the high-turnover site I, which in turn drives the structural reorganization required for ribosome splitting. ATP hydrolysis and ensuing release of ABCE1 from the small subunit terminate the post-splitting complex. Thus, ABCE1 runs through an allosterically coupled cycle of closure and opening at both sites, consistent with a processive clamp model. This study delineates the inner mechanics of ABCE1 and reveals why various ABCE1 mutants lead to defects in cell homeostasis, growth, and differentiation.
Life and biological resilience rely on the execution of precise gene expression profiles. A key mechanism to ensure cellular homeostasis is the regulation of protein synthesis. Recent studies have unveiled an intrinsic regulatory capacity of ribosomes, previously considered mere executors of mRNA translation. Neurons in particular finely regulate protein synthesis, at both global and local levels. This sustains their complex morphology and allows them to rapidly transmit, integrate, and respond to external stimuli. In this thesis, I investigated the neuronal ribosome and how subcellular environments and physiological perturbations shape it, by profiling its molecular composition, functional interconnections, and cellular distribution.
First, I used genetic engineering, biochemical purification, and mass spectrometry, to characterize in an unbiased manner the translation machinery specifically from excitatory and inhibitory neurons of the mouse cortex. I found that neuronal ribosomes commonly interact with RNA-binding proteins, components of the cytoskeleton, and proteins associated with the endoplasmic reticulum and vesicles. In line with the requirement for local protein synthesis in the distal parts of neurons, we observed that neuronal ribosomes preferentially interact with proteins involved in cellular transport. Remarkably, I observed a strong association between ribosomes and pre-synaptic vesicles, which suggests a potential regulatory interaction between local translation and neuronal activity.
Intriguingly, I and others have observed mRNAs encoding for core ribosomal proteins (RPs) among the genes most enriched in neuronal processes. This observation challenges two historical assumptions of ribosome biology: (1) new RPs are incorporated only into newly forming ribosomes, and (2) this incorporation occurs only in the nucleus and perinuclear region. In my PhD, I aimed to directly test these two assumptions and if proven wrong ask whether and why neurons would localize RP mRNAs far from their known assembly site.
Employing a combination of metabolic labeling and highly sensitive mass spectrometry techniques, I discovered that a subset of RPs rapidly and dynamically binds on and off mature ribosomes. Strikingly, this incorporation does not depend on the supply of new ribosomes from the nucleus. Therefore, my data refuted the assumption that ribosomes are built and degraded as a unit and revealed a more dynamic view of these machines, which can actively exchange core components. In particular, I found that the association of certain exchanging RPs is influenced by location (e.g., cell body versus neurites) and cellular state (e.g., post-oxidative stress). Neurons may use this mechanism to repair and/or specialize their protein synthesis machinery in a rapid and context-dependent manner.
Finally, I asked whether some steps of ribosome biogenesis could also take place in distal processes. Although most steps of ribosome assembly occur within the nucleus, the final stages of maturation are known to occur in the cytosol. By combining several imaging and biochemical approaches, I found that cytosolic (but not nuclear) pre-ribosomal particles are present in neuronal processes. Through the incorporation of new RPs into these immature particles, neurons may be able to locally “turn on” previously incompetent ribosomes. This may enable regions near synapses to enhance and customize their translational capacity, independently of the central pool of ribosomes from the cell body. Indeed, I observed that synaptic plasticity induces a maturation of cytosolic pre-ribosomes.
In summary, this thesis shows how neuronal ribosomes can sense cellular states, respond by adjusting their core composition, and in doing so influence the local capacity for protein synthesis. By overturning long-held assumptions in ribosome biology, this work highlights new molecular mechanisms of gene expression and enriches our understanding of the rapid and dynamic strategies cells employ to operate, thrive, and adaptively respond to environmental changes.
The ribonucleic acid of reovirus was extracted with 2 M sodium perchlorate solution and spread by the protein monolayer technique. Areas of the monolayer were transferred to support films, rotary shadowed, and observed in the electron microscope. Filaments of RNA obtained by extraction prior to spreading were similar in appearance and in distribution of contour lengths (0.2 to 1.2 μ) to those obtained by phenol extraction of the virus. Most of the filaments resulting from extraction of the virus suspension during spreading on a sodium perchlorate solution, however, were longer than 1 μ. The lengths of the longest filaments exceeded the 5 μ length predicted from chemical data for one single piece of complementary-stranded RNA in the reovirus particle.
The short filaments, 1.2 μ and less in length, fell into a tri-modal pattern of length distribution with peaks at 0.35 μ, 0.60 μ and 1.10 μ. These shorter lengths probably resulted from breakage of the intact RNA during the extraction procedures. The consistently observed pattern of length distribution suggests that they represent relatively stable subunits of the molecule.
Sodium perchlorate extracted reovirus RNA was thermally denatured in formaldehyde prior to spreading by the protein monolayer technique. Length distributions and relative numbers of filaments in the peaks of the tri-modal distribution pattern were similar to those found for unheated material when extracted prior to spreading. This similarity indicates that heating subsequent to extraction produced no further filament breakage. The thin, kinky appearance of the heated filaments, and the appearance of congruent pairs, indicated that heating had separated the strands of the complementary-stranded RNA subunits.
Rhodopsin-based voltage imaging tools for use in muscles and neurons of Caenorhabditis elegans
(2019)
Genetically encoded voltage indicators (GEVIs) based on microbial rhodopsins utilize the voltage-sensitive fluorescence of all-trans retinal (ATR), while in electrochromic FRET (eFRET) sensors, donor fluorescence drops when the rhodopsin acts as depolarization-sensitive acceptor. In recent years, such tools have become widely used in mammalian cells but are less commonly used in invertebrate systems, mostly due to low fluorescence yields. We systematically assessed Arch(D95N), Archon, QuasAr, and the eFRET sensors MacQ-mCitrine and QuasAr-mOrange, in the nematode Caenorhabditis elegans ATR-bearing rhodopsins reported on voltage changes in body wall muscles (BWMs), in the pharynx, the feeding organ [where Arch(D95N) showed approximately 128% ΔF/F increase per 100 mV], and in neurons, integrating circuit activity. ATR fluorescence is very dim, yet, using the retinal analog dimethylaminoretinal, it was boosted 250-fold. eFRET sensors provided sensitivities of 45 to 78% ΔF/F per 100 mV, induced by BWM action potentials, and in pharyngeal muscle, measured in simultaneous optical and sharp electrode recordings, MacQ-mCitrine showed approximately 20% ΔF/F per 100 mV. All sensors reported differences in muscle depolarization induced by a voltage-gated Ca2+-channel mutant. Optogenetically evoked de- or hyperpolarization of motor neurons increased or eliminated action potential activity and caused a rise or drop in BWM sensor fluorescence. Finally, we analyzed voltage dynamics across the entire pharynx, showing uniform depolarization but compartmentalized repolarization of anterior and posterior parts. Our work establishes all-optical, noninvasive electrophysiology in live, intact C. elegans.
In optogenetics, rhodopsins were established as light-driven tools to manipulate neuronal activity. However, during long-term photostimulation using channelrhodopsin (ChR), desensitization can reduce effects. Furthermore, requirement for continuous presence of the chromophore all-trans retinal (ATR) in model systems lacking sufficient endogenous concentrations limits its applicability. We tested known, and engineered and characterized new variants of de- and hyperpolarizing rhodopsins in Caenorhabditis elegans. ChR2 variants combined previously described point mutations that may synergize to enable prolonged stimulation. Following brief light pulses ChR2(C128S;H134R) induced muscle activation for minutes or even for hours (‘Quint’: ChR2(C128S;L132C;H134R;D156A;T159C)), thus featuring longer open state lifetime than previously described variants. Furthermore, stability after ATR removal was increased compared to the step-function opsin ChR2(C128S). The double mutants C128S;H134R and H134R;D156C enabled increased effects during repetitive stimulation. We also tested new hyperpolarizers (ACR1, ACR2, ACR1(C102A), ZipACR). Particularly ACR1 and ACR2 showed strong effects in behavioral assays and very large currents with fast kinetics. In sum, we introduce highly light-sensitive optogenetic tools, bypassing previous shortcomings, and thus constituting new tools that feature high effectiveness and fast kinetics, allowing better repetitive stimulation or investigating prolonged neuronal activity states in C. elegans and, possibly, other systems.
F1Fo‐ATP synthase is one of the best studied macromolecular machines in nature. It can be inhibited by a range of small molecules, which include the polyphenols, resveratrol and piceatannol. Here, we introduce Photoswitchable Inhibitors of ATP Synthase, termed PIAS, which were synthetically derived from these polyphenols. They can be used to reversibly control the enzymatic activity of purified yeast Yarrowia lipolyticaATP synthase by light. Our experiments indicate that the PIAS bind to the same site in the ATP synthase F1 complex as the polyphenols in their trans form, but they do not bind in their cis form. The PIAS could be useful tools for the optical precision control of ATP synthase in a variety of biochemical and biotechnological applications.
NK cells are part of the innate immune system, and are important players in the body’s first defence line against virus-infected and malignantly transformed cells. While T cells recognize neoplastic cells in an MHC-restricted fashion, NK cells do not require prior sensitization and education about the target. In leukemia and lymphoma patients undergoing allogeneic hematopoietic stem cell transplantation not only T cells but also NK cells have been found to mediate potent graft-versus-tumor effects. Hence, autologous or donor-derived NK cells hold great promise for cancer immunotherapy. Since the generation of highly purified NK cell products for clinical applications is labor-intensive and time consuming, established human NK cell lines such as NK-92 are also being considered for clinical protocols. NK-92 cells display phenotypic and functional characteristics similar to activated primary NK cells. While NK-92 cells are highly cytotoxic towards malignant cells of hematologic origin, they do not affect healthy human tissues. NK-92 cells can be expanded under GMP-compliant conditions, and can therefore be provided in sufficient numbers with defined phenotypic characteristics for clinical applications. Safety of NK-92 cells for adoptive immunotherapy was already shown in two phase I/II clinical trials...