Evolving Small-Molecule Biosensors with Improved Performance and Reprogrammed Ligand Preference Using OrthoRep

Abstract

Genetically encoded biosensors are valuable for the optimization of small-molecule biosynthesis pathways, because they transduce the production of small-molecule ligands into a readout compatible with high-throughput screening or selection in vivo. However, engineering biosensors with appropriate response functions and ligand preferences remains challenging. Here, we show that the continuous hypermutation system, OrthoRep, can be effectively applied to evolve biosensors with a high dynamic range, reprogrammed activity toward desired noncognate ligands, and proper operational range for coupling to biosynthetic pathways. In particular, we encoded the allosteric transcriptional factor, BenM, on OrthoRep such that the propagation of host yeast cells resulted in BenM’s rapid and continuous diversification. When these cells were subjected to cycles of culturing and sorting on BenM activity in the presence and absence of its cognate ligand, muconic acid, or the noncognate ligand, adipic acid, we obtained multiple BenM variants that respond to their corresponding ligands. These biosensors outperform previously engineered BenM-based biosensors by achieving a substantially greater dynamic range (up to ∼180-fold induction) and broadened operational range. The expression of select BenM variants in the presence of a muconic acid biosynthetic pathway demonstrated sensitive biosensor activation without saturating response, which should enable pathway and host engineering for higher production of muconic and adipic acids. Given the streamlined manner in which high-performance and versatile biosensors were evolved using OrthoRep, this study provides a template for generating custom biosensors for metabolic pathway engineering and other biotechnology goals.

GroovDB: A Database of Ligand-Inducible Transcription Factors

“…4,7 In general, activator-regulated sensor circuits can be generated by placing GroovDB-annotated binding sequences upstream of reporters (such as an RBS and GFP gene), and this approach has been used to successfully build circuits responsive to erythritol with EryD, 3-hydroxypropionate with HpdR, and cis−cis muconate with BenM, among others. 3,22,23 These general design strategies are described in more detail within the documentation section of GroovDB https://groov. bio/howToUse.…”

“…For example, by placing GroovDB-sourced operator sequences for the CamR and RamR repressors downstream of the −10 site of a strong E. coli promoter driving GFP, we were able to build reporter circuits under control of CamR and RamR biosensors that were responsive to camphor and tetrahydropapaverine, respectively. , In general, activator-regulated sensor circuits can be generated by placing GroovDB-annotated binding sequences upstream of reporters (such as an RBS and GFP gene), and this approach has been used to successfully build circuits responsive to erythritol with EryD, 3-hydroxypropionate with HpdR, and cis–cis muconate with BenM, among others. ,, These general design strategies are described in more detail within the documentation section of GroovDB .…”

Tackling the Catch-22 Situation of Optimizing a Sensor and a Transporter System in a Whole-Cell Microbial Biosensor Design for an Anthropogenic Small Molecule

“…Snoek et al used the ccMA and benzoate-responsive BenM as a scaffold for rapidly altering its response to ADA in Saccharomyces cerevisiae . The same scaffold protein was also used by Javanpour and Liu to select sequences with a very high dynamic range with ADA . The low detection range for ADA was 1.4 and 0.44 mM in these studies and may be attributed to transport limitations of the molecules because, in these studies, the authors depended on the native transporter or diffusion of ADA into the cell.…”

GroovDB: A database of ligand-inducible transcription factors

“…For example, by placing GroovDB-sourced operator sequences for the CamR and RamR repressors downstream of the -10 site of a strong E. coli promoter driving GFP, we were able to build reporter circuits under control of CamR and RamR biosensors that were responsive to camphor and tetrahydropapaverine, respectively 4,7 . In general, activator-regulated sensor circuits can be generated by placing GroovDB-annotated binding sequences upstream of reporters (such as an RBS and GFP gene), and this approach has been used to successfully build circuits responsive to erythritol with EryD, 3-hydroxypropionate with HpdR, and cis-cis muconate with BenM, among others 3,22,23 . These general design strategies are described in more detail within the documentation section of GroovDB https://groov.bio/howToUse.…”