Engineering and testing RNA-circuits in cell-free systems

Abstract

RNA molecules lie at the heart of living organisms where they are associated with most of the cellular processes. They have recently emerged as one of the most promising elements for developing programmable genetic regulatory systems. RNA regulators have been shown to offer great advantages to harness the power of synthetic biology. Versatility of functions, predictability of design, and light metabolic cost have turned RNA-based devices into components of primordial importance for therapeutic, diagnostic and biotechnological applications. However, advanced tasks require the use of sequential logic circuits that embed many constituents in the same system. Combining RNA-parts into more complex circuits remains experimentally challenging and difficult to predict. Contrary to protein-based networks, little work has been performed regarding the integration of RNA components to multi-level regulated circuits. In the first part of this thesis, combinations of variety of small transcriptional activator RNAs (STARs) and toehold switches were built into highly effective AND-gates. To characterise the components and their dynamic range, an Escherichia coli (E. coli) cell-free transcription-translation (TX-TL) system dispensed via nanoliter droplets was used. Cell-free systems, which constitute an open environment, have removed many of the complexities linked to the traditional use of living cells and have led to exciting opportunities for the rational design of genetic circuits. A modelling framework based on ordinary differential equations (ODEs), where parameters were inferred via parallel tempering, was established to analyse the expression construct in a qualitative and quantitative manner. Based on this analysis, nine additional AND-gates were built and tested in vitro. The functionality of the gates was found to be highly dependent on the concentration of the activating RNA for either the STAR or the toehold switch. All gates were successfully implemented in vivo, displaying a dynamic range comparable to the level of protein circuits. Subsequent spacer screening experiment enabled isolation of a gate mutant with dynamic range up to 1087 fold change, paving the way towards multi-layered devices where tight OFF-stages are required for efficient computation. Expanding the repertoire of RNA regulatory parts with efficient inhibitors would complete the set of logic operations necessary for the building of dynamiccircuits, such as memory devices or oscillators. The TX-TL system was functionalized with pre-expressed dSpyCas9, a mutated version of Cas9 without endonuclease activity. Four functional small guide RNAs (sgRNAs) targeting the sfGFP reporter were engineered and characterized, all resulting in high repression efficiency. A three-inputs logic circuit containing toehold, STAR and sgRNA was successfully co-expressed, validating the orthogonality of NOT and AND gates based solely on RNA-based regulation. In order to minimize interactions which could arise from RNA-circuit of increasing complexity, the TX-TL system was functionalized with a second protein, the Csy4 endoribonuclease, which selectively binds and cuts a small RNA hairpin. Normalization of gene expression from various untranslated region contexts and enhanced processing of three-inputs small RNA operon were demonstrated via the use of Csy4. Finally, characterizing complex RNA-based circuits requires techniques that resolves dynamics. To overcome the batch-format limitations inherent to TX-TL systems, a microfluidic nanoliter-scaled reactor was implemented, enabling synthesis rates to stay constant over time. Dynamic control of RNA circuitry was demonstrated by modulating the concentration of ligands, reversing the gene state through the conformational change of riboswitches. This thesis shows the potential of a rapid prototyping approach for RNA circuit design in TX-TL systems combined with a predicting model framework. Taken together, the characterization of a variety of RNA-parts : activators, repressors, or controllers culminating into logic modules; and augmented cell-extracts; form a complete RNA-toolbox for cell-free systems. The leveraging of this unique prototyping platform will ultimately enable the engineering and the study of highly dynamical RNA-circuits in vitro.