Protein-based sensors and switches provide attractive tools for the real-time monitoring and control of molecular processes in complex biological environments. Fluorescent sensor proteins have been developed for a wide variety of small molecules, but the construction of genetically encoded light-responsive ligand binding proteins remains mostly unexplored. Here we present a generic approach to reengineer a previously developed FRET-based Zn(2+) sensor into a light-activatable Zn(2+) binding protein using a design strategy based on mutually exclusive domain interactions. These so-called VividZn proteins consist of two light-responsive Vivid domains that homodimerize upon illumination with blue light, thus pr... More
Protein-based sensors and switches provide attractive tools for the real-time monitoring and control of molecular processes in complex biological environments. Fluorescent sensor proteins have been developed for a wide variety of small molecules, but the construction of genetically encoded light-responsive ligand binding proteins remains mostly unexplored. Here we present a generic approach to reengineer a previously developed FRET-based Zn(2+) sensor into a light-activatable Zn(2+) binding protein using a design strategy based on mutually exclusive domain interactions. These so-called VividZn proteins consist of two light-responsive Vivid domains that homodimerize upon illumination with blue light, thus preventing the binding of Zn(2+) between two Zn(2+) binding domains, Atox1 and WD4. Following optimization of the linker between WD4 and the N-terminus of one of the Vivid domains, VividZn variants were obtained that show a 9- to 55-fold decrease in Zn(2+) affinity upon illumination, which is fully reversible following dark adaptation. The Zn(2+) affinities of the switch could be rationally tuned between 1 pM and 2 nM by systematic variation of linker length and mutation of one of the Zn(2+) binding residues. Similarly, introduction of mutations in the Vivid domains allowed tuning of the switching kinetics between 10 min and 7 h. Low expression levels in mammalian cells precluded the demonstration of light-induced perturbation of cytosolic Zn(2+) levels. Nonetheless, our results firmly establish the use of intramolecular Vivid dimerization as an attractive light-sensitive input module to rationally engineer light-responsive protein switches based on mutually exclusive domain interactions.