Abstract
We leverage AI-powered de novo protein design to create a new generation of luciferase catalysts, termed the neoLux series, which exhibit superior properties over native luciferases. These features include compact size, robust stability, cofactor independence, efficient cellular expression, higher catalytic efficiency, and unique substrate orthogonality, marking a significant advancement beyond the limitations of native luciferases. Additionally, we computationally designed highly efficient neoLux-fluorescent protein FRET fusions capable of simultaneous multi-parametric imaging in cellulo and in vivo. Our pioneering approach has created a unified luminescent toolkit to allow for multi-colored tracking of cancer heterogeneity in vivo, paving the way for complex biological discovery.
Keywords:
FRET; bioassay; deep learning; enzyme design; luciferase; multiplexed bioluminescence imaging.