The generation and validation of a de novo-designed titin binder for use in fluorescence microscopy

Abstract

Advances in the generation of proteins in silico has enabled the efficient design of such that can bind to a specified target. Here, we demonstrate the use of a fluorescently-labelled de novo-designed protein to bind its target in situ and be imaged using fluorescence microscopy, a widely used experimental technique that typically relies on antibodies or similar evolutionary derived binders to identify the presence and location of targets in their native environment. Our de novo-designed protein binds the C-terminal domain M10 (Ig-169) of the giant muscle protein titin, which spans half a sarcomere, the basic contractile unit of striated muscle. M10 antibodies suitable for fluorescence microscopy are unavailable. Confocal microscopy of muscle sections shows the binder localises to the M-band of the sarcomere - where M10 is found - and fails to label muscle in competition experiments and in mutant muscle where M10 is absent. These results demonstrate the utility of de novo-designed proteins in immunostaining-like experiments and suggest a future where targets can be routinely identified in complex biological samples by in silico-generated binders. Such an approach avoids the need to generate antibodies or similar binders either in vivo or in vitro, which can have technical, financial and ethical challenges.