De novo design of high-affinity protein binders to bioactive helical peptides

Abstract

Many peptide hormones form an alpha-helix upon binding their receptors, and sensitive detection methods for them could contribute to better clinical management. De novo protein design can now generate binders with high affinity and specificity to structured proteins. However, the design of interactions between proteins and short helical peptides is an unmet challenge. Here, we describe parametric generation and deep learning-based methods for designing proteins to address this challenge. We show that with the RFdiffusion generative model, picomolar affinity binders can be generated to helical peptide targets either by noising and then denoising lower affinity designs generated with other methods, or completely de novo starting from random noise distributions; to our knowledge these are the highest affinity designed binding proteins against any protein or small molecule target generated directly by computation without any experimental optimization. The RFdiffusion designs enable the enrichment of parathyroid hormone or other bioactive peptides in human plasma and subsequent detection by mass spectrometry, and bioluminescence-based protein biosensors. Capture reagents for bioactive helical peptides generated using the methods described here could aid in the improved diagnosis and therapeutic management of human diseases.

Competing Interest Statement

The authors have declared no competing interest.

Footnotes

* Fixed misspelled author name

* https://www.bakerlab.org/wp-content/uploads/2022/11/diffusion_animation_PTHbinder.gif