Massively Multiplexed Engineering of Photo-Responsive Protein Actuators

Optogenetic tools have transformed modern biology by massively increasing the level of spatiotemporal control over cellular and protein systems through interfacing them with both programmable light-delivery devices and ambient environmental light stimuli. Their recent application to metabolic engineering, bacterial growth patterning, stem cell differentiation, and even the development of photo-programmable biomaterials, demonstrates their range of applications and operating environments, from microbes, to animals, to in vitro systems. However, despite their immense promise, engineering new optogenetic tools is still low-throughput and lacks predictive design tools.

Here we combine two innovations–one conceptual and one technical–to build hundreds of optogenetic tools simultaneously and, in doing so, create the data necessary to predict functional designs of arbitrary optogenetic proteins. Conceptually we are motivated by the modularity found in naturally occurring, photo-responsive proteins. When evolving new light-responsive behaviors organisms re-use a core set of photo-responsive domains rather than evolving new light-responsive protein domains from scratch. This implies that this core set of domains can endow a wide range of proteins with light-gated functionality. Indeed, we recently demonstrated this to be the case in engineering photo-switchable protein binding using nanobodies from llama and alpaca genomes. To take advantage of this modularity and engineer new classes of photo-switchable proteins we propose to leverage a massively multiplexed CRISPR engineering platform, which we recently adopted from its original intended purpose of creating single base-pair mutations to making large (~500bp) insertions into the yeast genome. This new high-throughput genome engineering technology will enable us to rapidly construct and screen through libraries of tens of thousands of potential photo-switchable proteins in a single experiment. This platform will rely on foundational computational and molecular engineering tools supported by our previous ICB work.