In the relentless pursuit of mastering biological systems, evolution has long been humanity’s unparalleled engineer, capable of sculpting life through natural selection. From the earliest agricultural practices where farmers selectively bred livestock and crops, to modern laboratories engineering proteins for diverse applications, the evolutionary process remains central to biological innovation. However, traditional laboratory approaches to directed protein evolution have often fallen short in replicating the dynamic, multi-state functions intrinsic to living organisms, constraining the complexity of proteins that can be engineered.
Directed evolution in the lab typically imposes a static selection pressure, favoring protein variants that exhibit constant, robust activity. This one-dimensional approach overlooks a crucial aspect of biological function: proteins often operate as switches or logic gates, continually shifting their states over time in response to cellular cues. Such proteins may need to toggle between on and off states multiple times or integrate multiple signals before triggering a response. Conventional methods, by prioritizing singular active states, inadvertently eliminate variants capable of these nuanced behaviors, compromising the very dynamism fundamental to cellular life.
Addressing this challenge, scientists at the École Polytechnique Fédérale de Lausanne (EPFL), led by Sahand Jamal Rahi, have pioneered a revolutionary technique dubbed “optovolution.” This method harnesses the precision of light to direct the evolution of proteins with dynamic, multi-state, and computational functionalities—allowing proteins to make intricate yes-or-no decisions based on defined inputs. Published in the journal Cell, this breakthrough bridges the divide between artificial protein evolution and the temporal complexities of natural cellular processes.
The research team designed an ingenious system within the budding yeast Saccharomyces cerevisiae, a model organism widely utilized in both industry and molecular biology due to its well-characterized genetics and ease of manipulation. By rewiring the yeast’s cell cycle machinery, they linked cellular progression directly to the functional oscillations of the protein under evolution. This novel setup created a biological “pass-fail” test: if the protein failed to switch states accurately within the cell cycle, the yeast cells would either halt division or die, effectively selecting only for proteins with optimal dynamic switching capabilities.
Critical to this system was coupling the output of the protein of interest to a cell-cycle regulator protein that is beneficial during one phase but toxic during another. This elegant genetic linkage ensured that proteins remaining persistently on or off would be detrimental to the cell, whereas those capable of timely oscillations would support continuous proliferation. Through this mechanism, cellular fitness became a direct proxy for the temporal functionality of the evolving protein, a feat that dramatically enhances the fidelity and relevance of selection in protein engineering.
Light served as the external control lever for protein activity through optogenetics — a cutting-edge technology that provides reversible, non-invasive, and high-resolution regulation of gene expression using tailored wavelengths of light. By delivering precisely timed light pulses synchronized with the yeast cell cycle, researchers could impose rigorous, dynamic selection pressures. Each roughly 90-minute cell cycle functioned as a rapid screening round, accelerating the evolutionary process and favoring variants exhibiting optimal temporal behavior without requiring manual intervention or laborious screening.
Optovolution’s capabilities were demonstrated across several protein classes. Initially, the researchers refined a widely studied light-responsive transcription factor, producing nineteen novel variants with distinctive properties. Some demonstrated heightened sensitivity to light, others showed reduced baseline activity in the absence of light, and remarkably, a subset exhibited responsiveness to green light rather than the commonly utilized blue spectrum. Achieving green-light responsiveness represents a significant breakthrough, as the engineering of optogenetic tools sensitive to warmer wavelengths has historically been constrained by proteins’ intrinsic light-absorption characteristics.
Further innovation came from evolving a red-light optogenetic system functional in yeast without the need for supplemental chemical cofactors. Through evolution, a key mutation disabled a yeast transport protein, enabling the system to utilize endogenously available light-sensitive molecules. This discovery simplifies experimental conditions significantly, reducing reliance on external compounds and broadening the applicability of red-light optogenetics in vivo.
The team’s ambitions extended beyond light-sensing proteins. They successfully evolved a transcription factor capable of acting as a molecular logic gate—a single protein computing element that activates gene expression only when both a light signal and a chemical input coexist. This achievement exemplifies the potential for building intricate, programmable biological circuits at the protein level, mirroring the decision-making processes of living cells with unprecedented precision.
Dynamic protein behaviors orchestrated by such molecular switches underpin critical biological functions—ranging from stress responses to cell cycle control and differentiation. By enabling the continuous evolution of these dynamic properties within living cells, optovolution provides a powerful platform for synthetic biology, allowing scientists to develop smarter, more adaptable cellular systems. These engineered proteins could pave the way toward programmable therapeutics, biosensors, and adaptive biomanufacturing processes driven by multi-layered control mechanisms.
Optovolution not only enriches our toolbox for synthetic biology but also illuminates fundamental evolutionary principles governing protein dynamics. By mimicking the temporal selection pressures that living cells impose, the platform offers a window into how complex protein functions may have emerged through natural evolution, offering clues to the development and optimization of biological computation over millennia.
The ability to evolve proteins that respond to multiple stimuli with temporal precision could transform optogenetic methodologies, enabling simultaneous independent control of various cellular pathways using different light colors. Such multiplexed control mechanisms would dramatically enhance our capacity to dissect cellular networks and engineer complex biological behaviors, inspiring new therapies and biotechnological innovations.
This novel approach promises to accelerate advances in biotechnology, enabling the creation of synthetic systems with unprecedented complexity and functionality. As directed evolution moves closer to recapitulating the nuanced operations of living cells, the boundaries of protein engineering are set to expand, opening a new era of dynamic, versatile bioengineering made possible through the fusion of light and evolution.
Subject of Research: Directed evolution of dynamic and computational protein functionalities using light-controlled selection systems.
Article Title: Light-directed evolution of dynamic, multi-state, and computational protein functionalities.
News Publication Date: March 6, 2026
Web References:
https://doi.org/10.1016/j.cell.2026.02.002
References:
Vojislav Gligorovski, Marco Labagnara, Lorenzo Scutteri, Marius Blackholm, Andreas Möglich, Nahal Mansouri, Sahand Jamal Rahi. Light-directed evolution of dynamic, multi-state, and computational protein functionalities. Cell, March 6, 2026.
Image Credits:
Sahand Rahi (EPFL)
Keywords: Directed evolution, optogenetics, light-controlled proteins, protein dynamics, synthetic biology, molecular computation, Saccharomyces cerevisiae, dynamic switching, protein engineering, multi-state proteins, optovolution, cellular circuits.
