Fig. 4: DRIVER-selected biosensors function directly as in vivo gene-regulatory switches.

a Biosensors evolved from DRIVER are validated to regulate reporter gene expression in yeast cells in response to exogenous ligand addition. Flow cytometry measurement of a fluorescent reporter under the control of b gardiquimod, d theophylline, e trans-zeatin, f aciclovir, and g noscapine biosensors in yeast in the presence and absence of exogenously added ligands. Each biosensor and control was tested in n = 3–8 biologically independent samples, with error bars corresponding to the standard error of mean across the replicates. Mean GFP/mCherry was normalized such that an inactive ribozyme control (sTRSVCtl) was set to 100 RFU (relative fluorescence unit). Biosensors exhibiting significant changes in gene-regulatory activity in response to their cognate ligand are denoted with asterisk(s), where ***p value <1e − 5, ****p value <1e − 6, from two-tailed, unpaired t tests between the with and without ligand conditions. Error bars are the standard error of the mean over n = 3–8 biologically independent samples for b–g, i, k, where individual dots indicate measurements for individual replicates. c. Dose–response curves of activation ratios at varying concentrations of ligand for gardiquimod switches as measured by flow cytometry. h Biosensors evolved from DRIVER are validated to regulate reporter gene expression in response to endogenous production of a metabolite through a heterologous pathway. i Flow cytometry measurement of a fluorescent reporter under the control of (S)-reticuline biosensors in yeast engineered with a heterologous metabolic pathway for synthesizing (S)-reticuline from l-DOPA. l-DOPA is fed to the yeast cells to increase the production of (S)-reticuline. j Density plot showing the distribution of relative fluorescence levels across a population of yeast cells harboring a fluorescent reporter under the control of SRet-499 as in (c) under different concentrations of l-DOPA. k Comparison of in vitro cleavage fraction determined via CleaveSeq and in vivo gene regulatory activity via flow cytometry in yeast in the absence of ligand. Each point represents an individual biosensor sequence. Error bars are the standard error of the mean of measurements over at least n = 3 biologically independent experiments in each dimension. The dotted line shows the expected relationship assuming that GFP expression is proportional to the fraction of the mRNA that does not undergo cleavage. Source data are provided as a Source Data file.