Precise de novo Design Principle of Antifreeze Peptides

Abstract

De novo design of antifreeze peptides (AFPTs) represents a formidable challenge due to the unclarified active structure of AFPTs. Here, we describe a “Site to Distance” principle for de novo design of AFPTs, in terms of understanding their structure–activity relationships. The first step is to point E, identified as the most potent ice-binding site (IBS) possessing at least 4-fold binding energy than natural IBSs, into the candidate backbones. The second step, based on the IBS (E), is to judiciously adjust the distances of sites to match the favorable number of the ice crystal lattice to achieve the strongest ice-binding, relying on a newly established low-temperature AFPT structure prediction platform. The resultant AFPTs show a substantial reduction in single ice crystal growth rates, much superior to >100 natural or designed AFPTs, including all that have been reported. Cryopreservation of therapeutic cells further confirms the accuracy of this design principle.