Identifying, isolating, and obtaining naturally occurring transcription factors (TFs) is crucial for developing transcription-dependent biosensors. However, identifying and optimizing TFs for given molecules requires extensive time and effort. Accordingly, here, we report a strategy for the de novo design of a nonnatural TF, DLA, on the basis of a subtle conformational change of the ligand-binding domain (LBD) after the binding of a target molecule with its receptor. For the de novo design of DLA, we applied molecular dynamics to simulate different conformational states of DLA in order to understand the complete activity of DLA, which involves shortening of the distance between the DNA-binding domain (DBD) and ... More
Identifying, isolating, and obtaining naturally occurring transcription factors (TFs) is crucial for developing transcription-dependent biosensors. However, identifying and optimizing TFs for given molecules requires extensive time and effort. Accordingly, here, we report a strategy for the de novo design of a nonnatural TF, DLA, on the basis of a subtle conformational change of the ligand-binding domain (LBD) after the binding of a target molecule with its receptor. For the de novo design of DLA, we applied molecular dynamics to simulate different conformational states of DLA in order to understand the complete activity of DLA, which involves shortening of the distance between the DNA-binding domain (DBD) and the activation domain (AD) after progesterone binds to its LBD within DLA. The simulated results suggested that prokaryotic LexA, a truncated LBD from the progesterone receptor, and prokaryotic B42 together constitute DLA with a TF function. As a proof of concept, DLA was used as a transcription activator controlling the transcription of green fluorescent protein to construct an S. cerevisiae biosensor for progesterone detection. The progesterone-specific biosensor was successfully constructed with a sensitivity index EC of 27 μg/L, working range (0.16-60 μg/L), and time-to-detection (2.5 h). Ultimately, a low-cost, user-friendly kit was developed for the rapid detection of progesterone in the clinic. Theoretically, this work can also be used to develop a variety of other biosensors by employing the same strategy.
