Combining High-throughput Data and CRISPR-based Approaches to Investigate E. coli Metabolism

DSpace Repositorium (Manakin basiert)


Dateien:

Zitierfähiger Link (URI): http://hdl.handle.net/10900/181038
http://nbn-resolving.org/urn:nbn:de:bsz:21-dspace-1810388
http://dx.doi.org/10.15496/publikation-122362
Dokumentart: Dissertation
Erscheinungsdatum: 2028-05-15
Sprache: Deutsch
Englisch
Fakultät: 7 Mathematisch-Naturwissenschaftliche Fakultät
Fachbereich: Informatik
Gutachter: Nahnsen, Sven (Prof. Dr.)
Tag der mündl. Prüfung: 2026-05-15
DDC-Klassifikation: 004 - Informatik
500 - Naturwissenschaften
570 - Biowissenschaften, Biologie
Freie Schlagwörter: CRISPR
Bioinformatik
Escherichia Coli
Daten Analyse
High-throughput data
Lizenz: http://tobias-lib.uni-tuebingen.de/doku/lic_ohne_pod.php?la=de http://tobias-lib.uni-tuebingen.de/doku/lic_ohne_pod.php?la=en
Zur Langanzeige

Inhaltszusammenfassung:

High-throughput experimental data acquisition and the generation of large datasets improved our possibilities to explore various fields of microbial research. One of these aspects is the metabolism of bacteria, such as Escherichia coli. CRISPR-based techniques combine high-throughput methods together with data driven analysis and enable the construction of large genetic libraries. These libraries can serve as resources, allowing a detailed look into condition dependent phenotypes. The aim of this thesis is to combine high-throughput datasets together with CRISPR-based methods to investigate condition-based metabolism of Escherichia coli and how it is connected to regulatory mechanisms and antibiotic resistance. First, I provide a broader background of all relevant topics. In Chapter 3 we introduce a workflow to mine and introduce amino acid mutations into Escherichia coli using CRISPR-assisted recombineering. The first step is mining genomes for amino acid mutations. Next, we provide a web application to design sgRNA-insert pairs for the collected mutations. Using CRISPR-assisted recombineering the mutations can be introduced into the wild-type Escherichia coli in a pooled approach. We can distinguish individual strains with their respective mutations by sequencing the plasmids of recombineered clones. We developed a bioinformatic workflow to analyze the sequencing results in an automated way, to make identification of pooled strains faster. As a proof of concept, we mined 9,370 clinical Escherichia coli isolates from the NCBI pathogens database and constructed a CRISPR library with 43,086 strains containing 16,723 metabolic mutations. After treating the library with two antibiotics, namely ciprofloxacin and carbenicillin we found 389 putative resistant mutations for ciprofloxacin and 164 for carbenicillin with clinical relevance. Next, in Chapter 4 we have a closer look into the metabolic mutations of the clinical isolates. We identified 213,450 mutations and investigated their distribution across 38 metabolic pathways. Further, we address the issue of data availability and quality. After normalizing the data, we highlight the pathways and genes with the highest number of mutations. In Chapter 5 we investigate another CRISPR mutant library and how they reduce antibiotic susceptibility. This library contained 15,120 Escherichia coli mutants, each of them have one amino acid mutation in one of 346 proteins. After treating the library with carbenicillin and gentamicin, we observed a twofold to tenfold increase in the minimal inhibitory concentrations. The highest number of mutations that reduced susceptibility against carbenicillin were found in the purine nucleotide biosynthesis, and against gentamicin in the respiratory chain. Moving on from antibiotic resistances, in Chapter 6 we investigate regulatory mechanisms and pathway wise interactions of a CRISPR interference library. To this end, we measured the metabolome and proteome of 281 Escherichia coli strains after 6.5 h of initial induction of the CRISPR interference system. We could observe buffering mechanisms inside target pathways and upregulated branchpoint enzymes, revealing the importance of metabolic flux control upon perturbation. Additionally, we integrated our metabolome and proteome data using a random forest regression model, enabling the prediction of 20% of measured protein concentrations.

Das Dokument erscheint in: