Schneider, Christopher (2017)
Design, implementation and characterization of synthetic riboswitches in Saccharomyces cerevisiae.
Technische Universität Darmstadt
Dissertation, Erstveröffentlichung
Kurzbeschreibung (Abstract)
The driving force of synthetic biology is the desire to build programmable cellular systems de novo modeled on their natural counterparts. To achieve this, environmental signals must be sensed and integrated to generate an appropriate output. Sensors and actuators can be created according to naturally occurring riboswitches. The implementation of systems that can process many inputs initially requires the availability of the corresponding sensor domains, called aptamers. Many of the currently used synthetic riboswitches rely on only a handful of ligands as their external input. It is therefore desirable to expand the toolbox by adding other ligand specificities. Harnessing the potential of in vitro selected aptamers to function as in vivo regulators in yeast, the current literature was screened for suitable synthetic and natural aptamers. Candidates were then assayed for their ability to reduce reporter gene expression at the translational level in presence of their cognate ligands, essentially functioning as a roadblock to the scanning ribosome in the 5’-UTR. The results revealed a wide range of basal expressions, but no switching phenotypes. Since the aptamers had not been adapted to the expression system, the results confirmed either the inability of most highly in vitro evolved aptamers to exhibit the structural changes necessary for riboswitching or the requirement of natural aptamers for a dedicated expression platform. In case of the in vitro generated neomycin aptamer, an in vivo screening yielded aptamers with regulatory activity. Following this example, ciprofloxacin-binding aptamers were subjected to the screening and one aptamer was found that exhibited 2-fold regulation. This aptamer was then subjected to further analysis. In an approach to improve existing riboswitches and derive rules for riboswitch design, dimers of the tetracycline riboswitch were set to control reporter expression by the described regulatory mechanisms. Modifying one riboswitch was sufficient to obtain a range of basal expressions levels and switching factors that could then be used to train a computational model based on data from more than 100 different constructs. After setting desired output parameters with respect to basal expression and switching factors the model computed corresponding sequences that were tested in a second iterative approach. These results partly mirrored the first data recording, but also showed that the model could reproduce the sequence-activity relationships learned from the first recording. Moreover, non-regulating sequences were almost completely erased and the abundance of switches with desired switching factors was increased. Leaving the refinement of riboswitches at the nucleotide level, a third study was conducted to elucidate the performance of a logic NOR gate constructed from the neomycin and tetracycline riboswitches at the device level. A descriptor for the performance of this RNA logic gate was sought to be applied to assess its functionality at the single-cell level, since the correct functionality of (RNA) devices in each single cell is a prerequisite for their adaption to a genetic circuit. To this end the Receiver-Operator-Characteristics (ROC) analysis was identified as a suitable performance measure to account for cell-to-cell variability compromising the full separation of two Boolean states by overlapping populations. Complemented by the transient induction and repression kinetics of the constructed NOR gate recorded by flow cytometry and microfluidics-based fluorescence microscopy, data was derived that allowed mapping the performance of the NOR gate with respect to speed and accuracy of the logic operation. In most studies only steady state data from bulk cultures are collected and these information are lost. A hierarchical stochastic model was built and calibrated with the transient single-cell data that enabled the in silico completion of the device characterization at the steady state level for various induction levels and repressor concentrations. In general, such data can then be used to simulate the experimentally inaccessible gate performance. Additionally, the model facilitated the computational investigation of a possible re-design of the gate for which the model prediction proved to be correct in vivo, highlighting both the appropriateness of the experiment design to obtain meaningful data and the applicability of in silico approaches to predict the correct in vivo phenotype.
Typ des Eintrags: | Dissertation | ||||
---|---|---|---|---|---|
Erschienen: | 2017 | ||||
Autor(en): | Schneider, Christopher | ||||
Art des Eintrags: | Erstveröffentlichung | ||||
Titel: | Design, implementation and characterization of synthetic riboswitches in Saccharomyces cerevisiae | ||||
Sprache: | Englisch | ||||
Referenten: | Süß, Prof. Dr. Beatrix ; Köppl, Prof. Dr. Heinz | ||||
Publikationsjahr: | 2017 | ||||
Ort: | Darmstadt | ||||
Datum der mündlichen Prüfung: | 4 April 2017 | ||||
URL / URN: | http://tuprints.ulb.tu-darmstadt.de/6133 | ||||
Kurzbeschreibung (Abstract): | The driving force of synthetic biology is the desire to build programmable cellular systems de novo modeled on their natural counterparts. To achieve this, environmental signals must be sensed and integrated to generate an appropriate output. Sensors and actuators can be created according to naturally occurring riboswitches. The implementation of systems that can process many inputs initially requires the availability of the corresponding sensor domains, called aptamers. Many of the currently used synthetic riboswitches rely on only a handful of ligands as their external input. It is therefore desirable to expand the toolbox by adding other ligand specificities. Harnessing the potential of in vitro selected aptamers to function as in vivo regulators in yeast, the current literature was screened for suitable synthetic and natural aptamers. Candidates were then assayed for their ability to reduce reporter gene expression at the translational level in presence of their cognate ligands, essentially functioning as a roadblock to the scanning ribosome in the 5’-UTR. The results revealed a wide range of basal expressions, but no switching phenotypes. Since the aptamers had not been adapted to the expression system, the results confirmed either the inability of most highly in vitro evolved aptamers to exhibit the structural changes necessary for riboswitching or the requirement of natural aptamers for a dedicated expression platform. In case of the in vitro generated neomycin aptamer, an in vivo screening yielded aptamers with regulatory activity. Following this example, ciprofloxacin-binding aptamers were subjected to the screening and one aptamer was found that exhibited 2-fold regulation. This aptamer was then subjected to further analysis. In an approach to improve existing riboswitches and derive rules for riboswitch design, dimers of the tetracycline riboswitch were set to control reporter expression by the described regulatory mechanisms. Modifying one riboswitch was sufficient to obtain a range of basal expressions levels and switching factors that could then be used to train a computational model based on data from more than 100 different constructs. After setting desired output parameters with respect to basal expression and switching factors the model computed corresponding sequences that were tested in a second iterative approach. These results partly mirrored the first data recording, but also showed that the model could reproduce the sequence-activity relationships learned from the first recording. Moreover, non-regulating sequences were almost completely erased and the abundance of switches with desired switching factors was increased. Leaving the refinement of riboswitches at the nucleotide level, a third study was conducted to elucidate the performance of a logic NOR gate constructed from the neomycin and tetracycline riboswitches at the device level. A descriptor for the performance of this RNA logic gate was sought to be applied to assess its functionality at the single-cell level, since the correct functionality of (RNA) devices in each single cell is a prerequisite for their adaption to a genetic circuit. To this end the Receiver-Operator-Characteristics (ROC) analysis was identified as a suitable performance measure to account for cell-to-cell variability compromising the full separation of two Boolean states by overlapping populations. Complemented by the transient induction and repression kinetics of the constructed NOR gate recorded by flow cytometry and microfluidics-based fluorescence microscopy, data was derived that allowed mapping the performance of the NOR gate with respect to speed and accuracy of the logic operation. In most studies only steady state data from bulk cultures are collected and these information are lost. A hierarchical stochastic model was built and calibrated with the transient single-cell data that enabled the in silico completion of the device characterization at the steady state level for various induction levels and repressor concentrations. In general, such data can then be used to simulate the experimentally inaccessible gate performance. Additionally, the model facilitated the computational investigation of a possible re-design of the gate for which the model prediction proved to be correct in vivo, highlighting both the appropriateness of the experiment design to obtain meaningful data and the applicability of in silico approaches to predict the correct in vivo phenotype. |
||||
Alternatives oder übersetztes Abstract: |
|
||||
URN: | urn:nbn:de:tuda-tuprints-61339 | ||||
Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 500 Naturwissenschaften und Mathematik > 570 Biowissenschaften, Biologie | ||||
Fachbereich(e)/-gebiet(e): | 10 Fachbereich Biologie 10 Fachbereich Biologie > Synthetic Genetic Circuits (2020 umbenannt in "Synthetic RNA biology") LOEWE > LOEWE-Schwerpunkte > CompuGene – Computer-gestützte Verfahren zur Generierung komplexer genetischer Schaltkreise LOEWE > LOEWE-Schwerpunkte LOEWE |
||||
Hinterlegungsdatum: | 09 Apr 2017 19:55 | ||||
Letzte Änderung: | 09 Apr 2017 19:55 | ||||
PPN: | |||||
Referenten: | Süß, Prof. Dr. Beatrix ; Köppl, Prof. Dr. Heinz | ||||
Datum der mündlichen Prüfung / Verteidigung / mdl. Prüfung: | 4 April 2017 | ||||
Export: | |||||
Suche nach Titel in: | TUfind oder in Google |
Frage zum Eintrag |
Optionen (nur für Redakteure)
Redaktionelle Details anzeigen |