Refine
Has Fulltext
- yes (5)
Is part of the Bibliography
- yes (5)
Document Type
- Doctoral Thesis (5)
Language
- English (5)
Keywords
- 18S rRNA (1)
- Agentenbasierte Modellierung (1)
- Bildverarbeitung (1)
- Caenorhabditis elegans (1)
- Cell cycle (1)
- Compressed Sensing (1)
- Datenanalyse (1)
- Differentialgleichung (1)
- Elektronenmikroskopie (1)
- Embryonalentwicklung (1)
Because of its complexity and intricacy, studying the nervous system is often challenging. Fortunately, the small nematode roundworm Caenorhabditis elegans is well established as a model system for basic neurobiological research. The C. elegans model is also the only organism with a supposedly complete connectome, an organism-wide map of synaptic connectivity resolved by electron microscopy, which provides some understanding of how the nervous system works as a whole. However, the number of available data-sets is small and the connectome contains errors and gaps. One example of this concerns electrical synapses. Electrical synapses are formed by gap junctions and difficult to map due to their often ambiguous morphology in electron micrographs, leading to misclassification or omission. On the other hand, chemical synapses are more easily mapped, but many aspects of their mode of operation remain elusive and their role in the C. elegans connectome is oversimplified. A comprehensive understanding of signal transduction of neurons between each other and other cells will be indispensable for a comprehensive understanding of the nervous system. In this thesis, I approach these challenges with a combination of advanced light and electron microscopy techniques.
First, this thesis describes a strategy to increase synaptic specificity in connectomics. Specifically, I classify gap junctions with a high degree of confidence. To achieve this, I utilized array tomography (AT). In this thesis, AT is adapted for high-pressure freezing to optimize for structure preservation and for super-resolution light microscopy; in this manner, I aim to bridge the gap between light and electron microscopy resolutions. I call this adaptation super-resolution array tomography (srAT). The srAT approach made it possible to clearly identify and map gap junctions with high precision and accuracy. The results from this study showcased the feasibility of incorporating electrical synapses into connectomes in a systematic manner, and subsequent studies have used srAT for other models and questions.
As mentioned above, the C. elegans connectomic model suffers from a shortage of datasets. For most larval stages, including the special dauer larval stage, connectome data is completely missing up to now. To obtain the first partial connectome data-set of the C. elegans dauer larva, we used focused ion-beam scanning electron microscopy (FIB-SEM). This technique offers an excellent axial resolution and is useful for acquiring large volumes for connectomics. Together with our collaborators, I acquired several data-sets which enable the analysis of dauer stage-specific “re-wiring” of the nervous system and thus offer valuable insights into connectome plasticity/variability.
While chemical synapses are easy to map relative to electrical synapses, signal transduction via chemical transmitters requires a large number of different proteins and molecular processes acting in conjunction in a highly constricted space. Because of the small spatial scale of the synapse, investigating protein function requires very high resolution, which electron tomography provides. I analyzed electron tomograms of a worm-line with a mutant synaptic protein, the serine/threonine kinase SAD-1, and found remarkable alterations in several architectural features. My results confirm and re-contextualize previous findings and provide new insight into the functions of this protein at the chemical synapse.
Finally, I investigated the effectiveness of our methods on “malfunctioning,” synapses, using an amyotrophic lateral sclerosis (ALS) model. In the putative synaptopathy ALS, the mechanisms of motor neuron death are mostly unknown. However, mutations in the gene FUS (Fused in Sarcoma) are one known cause of the disease. The expression of the mutated human FUS in C. elegans was recently shown to produce an ALS-like phenotype in the worms, rendering C. elegans an attractive disease model for ALS. Together with our collaboration partners, I applied both srAT and electron tomography methods to “ALS worms” and found effects on vesicle docking. These findings help to explain electrophysiological recordings that revealed a decrease in frequency of mini excitatory synaptic currents, but not amplitudes, in ALS worms compared to controls. In addition, synaptic endosomes appeared larger and contained electron-dense filaments in our tomograms. These results substantiate the idea that mutated FUS impairs vesicle docking and also offer new insights into further molecular mechanisms of disease development in FUS-dependent ALS. Furthermore, we demonstrated the broader applicability of our methods by successfully using them on cultured mouse motor neurons.
Overall, using the C. elegans model and a combination of light and electron microscopy methods, this thesis helps to elucidate the structure and function of neuronal synapses, towards the aim of obtaining a comprehensive model of the nervous system.
Mammalian embryonic development is subject to complex biological relationships that need to be understood. However, before the whole structure of development can be put together, the individual building blocks must first be understood in more detail. One of these building blocks is the second cell fate decision and describes the differentiation of cells of the inner cell mass of the embryo into epiblast and primitive endoderm cells. These cells then spatially segregate and form the subsequent bases for the embryo and yolk sac, respectively. In organoids of the inner cell mass, these two types of progenitor cells are also observed to form, and to some extent to spatially separate. This work has been devoted to these phenomena over the past three years. Plenty of studies already provide some insights into the basic mechanics of this cell differentiation, such that the first signs of epiblast and primitive endoderm differentiation, are the expression levels of transcription factors NANOG and GATA6. Here, cells with low expression of GATA6 and high expression of NANOG adopt the epiblast fate. If the expressions are reversed, a primitive endoderm cell is formed. Regarding the spatial segregation of the two cell types, it is not yet clear what mechanism leads to this. A common hypothesis suggests the differential adhesion of cell as the cause for the spatial rearrangement of cells. In this thesis however, the possibility of a global cell-cell communication is investigated. The approach chosen to study these phenomena follows the motto "mathematics is biology's next microscope". Mathematical modeling is used to transform the central gene regulatory network at the heart of this work into a system of equations that allows us to describe the temporal evolution of NANOG and GATA6 under the influence of an external signal. Special attention is paid to the derivation of new models using methods of statistical mechanics, as well as the comparison with existing models. After a detailed stability analysis the advantages of the derived model become clear by the fact that an exact relationship of the model parameters and the formation of heterogeneous mixtures of two cell types was found. Thus, the model can be easily controlled and the proportions of the resulting cell types can be estimated in advance. This mathematical model is also combined with a mechanism for global cell-cell communication, as well as a model for the growth of an organoid. It is shown that the global cell-cell communication is able to unify the formation of checkerboard patterns as well as engulfing patterns based on differently propagating signals. In addition, the influence of cell division and thus organoid growth on pattern formation is studied in detail. It is shown that this is able to contribute to the formation of clusters and, as a consequence, to breathe some randomness into otherwise perfectly sorted patterns.
The fusion of methods from several disciplines is a crucial component of scientific development. Artificial Neural Networks, based on the principle of biological neuronal networks, demonstrate how nature provides the best templates for technological advancement. These innovations can then be employed to solve the remaining mysteries of biology, including, in particular, processes that take place on microscopic scales and can only be studied with sophisticated techniques. For instance, direct Stochastic Optical Reconstruction Microscopy combines tools from chemistry, physics, and computer science to visualize biological processes at the molecular level. One of the key components is the computer-aided reconstruction of super-resolved images. Improving the corresponding algorithms increases the quality of the generated data, providing further insights into our biology. It is important, however, to ensure that the heavily processed images are still a reflection of reality and do not originate in random artefacts.
Expansion microscopy is expanding the sample by embedding it in a swellable hydrogel. The method can be combined with other super-resolution techniques to gain additional resolution. We tested this approach on microtubules, a well-known filamentous reference structure, to evaluate the performance of different protocols and labelling techniques.
We developed LineProfiler an objective tool for data collection. Instead of collecting perpendicular profiles in small areas, the software gathers line profiles from filamentous structures of the entire image. This improves data quantity, quality and prevents a biased choice of the evaluated regions. On the basis of the collected data, we deployed theoretical models of the expected intensity distribution across the filaments. This led to the conclusion that post-expansion labelling significantly reduces the labelling error and thus, improves the data quality. The software was further used to determine the expansion factor and arrangement of synaptonemal complex data.
Automated Simple Elastix uses state-of-the-art image alignment to compare pre- and post-expansion images. It corrects linear distortions occurring under isotropic expansion, calculates a structural expansion factor and highlights structural mismatches in a distortion map. We used the software to evaluate expanded fungi and NK cells. We found that the expansion factor differs for the two structures and is lower than the overall expansion of the hydrogel.
Assessing the fluorescence lifetime of emitters used for direct Stochastic Optical Reconstruction Microscopy can reveal additional information about the molecular environment or distinguish dyes emitting with a similar wavelength. The corresponding measurements require a confocal scanning of the sample in combination with the fluorescent switching of the underlying emitters. This leads to non-linear, interrupted Point Spread Functions. The software ReCSAI targets this problem by combining the classical algorithm of compressed sensing with modern methods of artificial intelligence. We evaluated several different approaches to combine these components and found, that unrolling compressed sensing into the network architecture yields the best performance in terms of reconstruction speed and accuracy.
In addition to a deep insight into the functioning and learning of artificial intelligence in combination with classical algorithms, we were able to reconstruct the described non-linearities with significantly improved resolution, in comparison to other state-of-the-art architectures.
Most of the studies in cell biology primarily focus on models from the opisthokont group of eukaryotes. However, opisthokonts do not encompass the full diversity of eukaryotes. Thus, it is necessary to broaden the research focus to other organisms to gain a comprehensive understanding of basic cellular processes shared across the tree of life. In this sense, Trypanosoma brucei, a unicellular eukaryote, emerges as a viable alternative. The collaborative efforts in genome sequencing and protein tagging over the past two decades have significantly expanded our knowledge on this organism and have provided valuable tools to facilitate a more detailed analysis of this parasite. Nevertheless, numerous questions still remain.
The survival of T. brucei within the mammalian host is intricately linked to the endo-lysosomal system, which plays a critical role in surface glycoprotein recycling, antibody clearance, and plasma membrane homeostasis. However, the dynamics of the duplication of the endo-lysosomal system during T. brucei proliferation and its potential relationship with plasma membrane growth remain poorly understood. Thus, as the primary objective, this thesis explores the endo-lysosomal system of T. brucei in the context of the cell cycle, providing insights on cell surface growth, endosome duplication, and clathrin recruitment. In addition, the study revisits ferritin endocytosis to provide quantitative data on the involvement of TbRab proteins (TbRab5A, TbRab7, and TbRab11) and the different endosomal subpopulations (early, late, and recycling endosomes, respectively) in the transport of this fluid-phase marker. Notably, while these subpopulations function as distinct compartments, different TbRabs can be found within the same region or structure, suggesting a potential physical connection between the endosomal subpopulations. The potential physical connection of endosomes is further explored within the context of the cell cycle and, finally, the duplication and morphological plasticity of the lysosome are also investigated. Overall, these findings provide insights into the dynamics of plasma membrane growth and the coordinated duplication of the endo-lysosomal system during T. brucei proliferation. The early duplication of endosomes suggests their potential involvement in plasma membrane growth, while the late duplication of the lysosome indicates a reduced role in this process. The recruitment of clathrin and TbRab GTPases to the site of endosome formation supports the assumption that the newly formed endosomal system is active during cell division and, consequently, indicates its potential role in plasma membrane homeostasis.
Furthermore, considering the vast diversity within the Trypanosoma genus, which includes ~500 described species, the macroevolution of the group was investigated using the combined information of the 18S rRNA gene sequence and structure. The sequence-structure analysis of T. brucei and other 42 trypanosome species was conducted in the context of the diversity of Trypanosomatida, the order in which trypanosomes are placed. An additional analysis focused on Trypanosoma highlighted key aspects of the group’s macroevolution. To explore these aspects further, additional trypanosome species were included, and the changes in the Trypanosoma tree topology were analyzed. The sequence-structure phylogeny confirmed the independent evolutionary history of the human pathogens T. brucei and Trypanosoma cruzi, while also providing insights into the evolution of the Aquatic clade, paraphyly of groups, and species classification into subgenera.
Machine-Learning-Based Identification of Tumor Entities, Tumor Subgroups, and Therapy Options
(2023)
Molecular genetic analyses, such as mutation analyses, are becoming increasingly important in the tumor field, especially in the context of therapy stratification. The identification of the underlying tumor entity is crucial, but can sometimes be difficult, for example in the case of metastases or the so-called Cancer of Unknown Primary (CUP) syndrome. In recent years, methylome and transcriptome utilizing machine learning (ML) approaches have been developed to enable fast and reliable tumor and tumor subtype identification. However, so far only methylome analysis have become widely used in routine diagnostics.
The present work addresses the utility of publicly available RNA-sequencing data to determine the underlying tumor entity, possible subgroups, and potential therapy options. Identification of these by ML - in particular random forest (RF) models - was the first task. The results with test accuracies of up to 99% provided new, previously unknown insights into the trained models and the corresponding entity prediction. Reducing the input data to the top 100 mRNA transcripts resulted in a minimal loss of prediction quality and could potentially enable application in clinical or real-world settings.
By introducing the ratios of these top 100 genes to each other as a new database for RF models, a novel method was developed enabling the use of trained RF models on data from other sources.
Further analysis of the transcriptomic differences of metastatic samples by visual clustering showed that there were no differences specific for the site of metastasis. Similarly, no distinct clusters were detectable when investigating primary tumors and metastases of cutaneous skin melanoma (SKCM).
Subsequently, more than half of the validation datasets had a prediction accuracy of at least 80%, with many datasets even achieving a prediction accuracy of – or close to – 100%.
To investigate the applicability of the used methods for subgroup identification, the TCGA-KIPAN dataset, consisting of the three major kidney cancer subgroups, was used. The results revealed a new, previously unknown subgroup consisting of all histopathological groups with clinically relevant characteristics, such as significantly different survival. Based on significant differences in gene expression, potential therapeutic options of the identified subgroup could be proposed.
Concludingly, in exploring the potential applicability of RNA-sequencing data as a basis for therapy prediction, it was shown that this type of data is suitable to predict entities as well as subgroups with high accuracy. Clinical relevance was also demonstrated for a novel subgroup in renal cell carcinoma. The reduction of the number of genes required for entity prediction to 100 genes, enables panel sequencing and thus demonstrates potential applicability in a real-life setting.