Researchers at EPFL's Laboratory of the Physics of Biological Systems have introduced a new method called "optovolution," which uses light to guide the evolution of proteins with dynamic and multi-state functions. The technique is designed to better mimic how proteins behave in living cells, where they often need to switch between different states over time.
Traditional laboratory methods for directed protein evolution usually apply constant selection pressure, resulting in proteins that are always strongly active. This approach does not capture the dynamic behavior seen in natural biological systems, where proteins may need to turn on and off at specific times or respond to multiple signals. As a result, existing methods can fail to produce proteins capable of complex switching or multi-state behaviors.
The team led by Sahand Jamal Rahi developed their system using the yeast Saccharomyces cerevisiae. They modified the yeast’s cell cycle so that its progression depended on the performance of the protein being evolved, requiring it to switch between off and on states. If the protein remained in one state too long, the yeast cell would either stall or die. Only those cells with proteins able to oscillate correctly continued dividing.
Light was used as an external control through optogenetics—a method that switches genes on and off using light pulses. Each approximately 90-minute cell cycle served as a test for whether the protein switched at the right moment. This allowed researchers to favor variants with improved dynamics without manual screening or repeated intervention.
The researchers applied optovolution to several types of proteins. They enhanced a widely used light-controlled transcription factor, obtaining 19 new variants that showed increased sensitivity to light, reduced activity in darkness, or responsiveness to green rather than just blue light—a property previously thought difficult to achieve due to how these proteins absorb light.
Additionally, they evolved a red-light optogenetic system so that it no longer required supplementation with a chemical cofactor in yeast. A discovered mutation allowed this system to use naturally occurring light-sensitive molecules inside yeast cells, simplifying its use in experiments.
The method also proved effective beyond light-sensing proteins. The team evolved a transcription factor acting as a single-protein computer—activating genes only when both a light signal and a chemical signal were present simultaneously.
"Dynamic protein functions are at the heart of sensing, decision‑making, and control in biology, from how cells respond to stress to how they commit to divide. By making these behaviors continuously evolvable inside living cells, optovolution opens new possibilities for synthetic biology, biotechnology, and basic research," said Sahand Jamal Rahi.
According to the study published in Cell, this approach could help scientists design more advanced cellular circuits and develop optogenetic systems controllable by different colors of light.
The findings aim to advance synthetic biology by enabling researchers to explore how complex protein behaviors emerge through evolution within living cells.
