Computational design of superstable proteins through maximized hydrogen bonding

Hydrogen bonds are fundamental chemical interactions that stabilize protein structures, particularly in β sheets, enabling resistance to mechanical stress and environmental extremes. Here, inspired by natural mechanostable proteins with shearing hydrogen bonds, such as titin and silk fibroin, we de novo designed superstable proteins by maximizing hydrogen-bond networks within force-bearing β strands. Using a computational framework combining artificial intelligence-guided structure and sequence design with all-atom molecular dynamics MD simulations, we systematically expanded protein architecture, increasing the number of backbone hydrogen bonds from 4 to 33. The resulting proteins exhibited unfolding forces exceeding 1,000 pN, about 400% stronger than the natural titin immunoglobulin domain, and retained structural integrity after exposure to 150 °C. This molecular-level stability translated directly to macroscopic properties, as demonstrated by the formation of thermally stable hydrogels. Our work introduces a scalable and efficient computational strategy for engineering robust proteins, offering a generalizable approach for the rational design of resilient protein systems for extreme environments.