Engineering E. coli for Dual Color Detection of 2,4,6-Trinitrotoluene

Genetically engineered microorganisms have been applied for accurate and sensitive detection of target analytes in a diverse variety of fields such as medicine, environmental monitoring, defense, food processing and safety. Here we combine in vitro selection and computational design to identify a novel RNA regulatory sensing element: a synthetic riboswitch that initiates a recombinase-based memory switch in E. coli cells in response to 2,4,6-trinitrotoluene (TNT).  TNT is one of the most commonly used explosives for military and industrial applications; its detection is important from a security perspective and due to its high toxicity.

Riboswitches are typically composed of two functional domains: an aptamer domain that binds to a specific analyte; and an expression platform that controls the expression of a downstream gene via conformational changes that are induced by analyte binding to the aptamer domain. First, we identified a TNT-binding aptamer using a magnetic bead-based SELEX (systematic evolution of ligands by exponential enrichment) technique specifically developed for selection of structure-switching aptamers for small molecule analytes. This method is advantageous because it does not require modification of the analyte. Modifying a small molecule can block groups that are critical for binding to an aptamer, which can affect the selectivity of identified aptamers. Next, we applied computational RNA design to facilitate and expedite converting the resulting TNT aptamer into a functional riboswitch. This was based on a recently developed statistical thermodynamic model that predicts the sequence-structure-function relationship for translation-regulating riboswitches that activate gene expression inside cells. In contrast to commonly used in vivo selection and screening approaches that require analysis of a large number of riboswitch clones, computational design resulted in two functional riboswitches selected from the top five predicted sequences when tested in E. coli cells.  These two riboswitches, TNTrs14 and TNTrs15, demonstrated activation ratios of 19.2 (± 5.0) and 18.8 (± 0.3), respectively. In the cell-based sensor, TNTrs14 and TNTrs15 were placed upstream of FimE recombinase coding sequence in one plasmid, and a second plasmid carried a fluorescent reporter composed of two fluorescent protein genes (mKate2 and GFPa1) flanking an invertible DNA segment (fimS) containing a constitutively active promoter. Without an analyte, the fluorescent reporter constitutively expresses green fluorescent protein (GFPa1). Addition of TNT initiates translation of FimE causing unidirectional inversion of the fimS segment and constitutive expression of red fluorescent protein (mKate2). This dual color cell-based sensor allows not only the detection of the presence of the target analyte, but also determination of the viability and localization of an uninduced sensor. By leveraging parallel computational modeling with construction and testing we were able to expedite the design, build, test cycle to develop a sensor-circuit device with ‘real-world’ applications.

This research is funded by the Air Force Office of Scientific Research.