Synthetic genetic circuits to monitor nanomaterial triggered toxicity

Abstract

In the past decades, nanomaterial (NM) usage in various fields has been of great interest because of their unique properties that show tuneable optical and physical properties depending on their size. Yet, safety concerns of NMs on human or environment arise with increased NM usage. Thanks to their small size, NMs can easily penetrate through cellular barriers and their high surface-to-volume ratio makes them catalytically active creating stress on cells such as protein unfolding, DNA damage, ROS generation etc. Hence, biocompatibility assessment of NMs has been analyzed before their field application such as drug delivery and imaging which requiring human exposure. Yet, conventional biocompatibility tests fall short of providing a fast toxicity report. One aspect of the present thesis is to develop a living biosensor to report biocompatibility of NMs with the aim of providing fast feedback to engineer them with lower toxicity levels before applying on humans. For this purpose, heat shock response (HSR), which is the general stress indicator, was engineered utilizing synthetic biology approaches. Firstly, four highly expressed heat shock protein (HSP) promoters were selected among HSPs. In each construct, a reporter gene was placed under the control of these HSP promoters to track signal change upon stress (i.e., heat or NMs) exposure. However, initial results indicated that native HSPs are already active in cells to maintain cellular homeostasis. Moreover, they need to be engineered to create a proper stress sensor. Thus, these native HSP promoters were engineered with riboregulators and results indicated that these new designs eliminated unwanted background signals almost entirely. Yet, this approach also led to a decrease in expected sensor signal upon stress treatment. To increase the sensor signal, a positive feedback loop using bacterial communication, quorum sensing, method was constructed. HSR was integrated with QS circuit showed that signal level increased drastically. Yet, background signal also increased. Moreover, instead of using activation based HSR system as in Escherichia coli, repression based system was hypothesized to solve the problem. Thus, a repression based genetic circuit, inspired by the HSR mechanism of Mycobacterium tuberculosis, was constructed. These circuits could report the toxicity of quantum dots (QDs) in 1 hour. As a result, these NM toxicity sensors can provide quick reports, which can lower the demand for additional experiments with more complex organisms. As part of this study, a source detection circuit coupling HSR mechanism with metal induced transcription factors (TFs) has been constructed to report the source of the toxic compound. For this purpose, gold and cadmium were selected as model ions. In the engineered circuits, stress caused by metal ions activates expression of regulatory elements such as TFs of specific ions (GolS for gold and CadR and MerR(mut) for cadmium) and a site-specific recombinase. In the system, the recombinase inverts the promoter induced by TF-metal ion complex, and a reporter has been expressed based on the inducer showing the source of the stress as either gold or cadmium. Finally, a mammalian cellular toxicity sensor has been developed using similar approaches used in bacterial sensors. To begin with, two HSP families have been selected: HSP70 and α-Bcrystallin. Initial circuits were designed using promoter regions of both protein families to control the expression of a reporter, gfp. Both circuits were tested with heat and cadmium ions with varying concentrations and results showed that HSP70-based sensor had high background signal because of its active role in cellular homeostasis and protein folding in cells. Additionally, a slight increase was observed after heat treatment. Similar results were observed for α-Bcrystallin-based sensor; yet, these outcomes were not suitable for a desirable sensor requiring tight control. Thus, we decided to transfer the bacterial repression based toxicity sensor into mammalian cells. At the beginning, expression of the repressor, HspR, from M. tuberculosis was checked in HEK293T cell line and modified with nuclear localization signal (NLS) to localize the repressor in the nucleus. Further, a minimal promoter (SV40) controlling the expression of a reporter was engineered with single and double inverted repeats (IRs) for HspR binding. Then, HspR and engineered reporter circuits were co-trasfected to track signals at normal growth conditions and upon stress. Each circuit was tested with heat and cadmium treatment and results were showed repression of GFP expression by HspR at normal conditions, but no significant signal increase was observed upon stress. Hence, constructed mammalian circuits require more optimization to find optimum working conditions of sensors. To sum up, in this study, a powerful candidate to manufacture ordered gene circuits to detect nanomaterial-triggered toxicity has been demonstrated. Unlike previous studies utilizing HSR mechanism as stress biosensors, we re-purposed the HSR mechanism of both bacteria and mammalian cells with different engineering approaches (i.e., riboregulators, quorum sensing mechanism, promoter engineering). As a result, an easy-to-use, cheap and fast acting nanomaterial-triggered toxicity assessment tool has been developed. Also, initial principles of mammalian whole cell biosensor design for the same purpose have been indicated to expand the limited toxicity detection strategies utilizing mammalian cells. This study contributed for the detection of toxic NMs providing a feedback about the fate of these NMs so that one can engineer them to make biocompatible before field application.