Transferrin receptor targeting by de novo sheet extension

The de novo design of polar protein–protein interactions is challenging because of the thermodynamic cost of stripping water away from the polar groups. Here, we describe a general approach for designing proteins which complement exposed polar backbone groups at the edge of beta sheets with geometrically matched beta strands. We used this approach to computationally design small proteins that bind to an exposed beta sheet on the human transferrin receptor (hTfR), which shuttles interacting proteins across the blood–brain barrier (BBB), opening up avenues for drug delivery into the brain. We describe a design which binds hTfR with a 20 nM K_d, is hyperstable, and crosses an in vitro microfluidic organ-on-a-chip model of the human BBB. Our design approach provides a general strategy for creating binders to protein targets with exposed surface beta edge strands.

While most protein–protein interfaces are composed primarily of sidechain–sidechain interactions, backbone hydrogen bonding can also play a role. For example, beta sheet hydrogen bonds across protein–protein interfaces are present in complexes of PDZ domains with their peptide targets, SUMO–SIM interactions, and Serpin-protease complexes among others (1, 2); in each case, the result is an extended beta sheet that spans both partners. Such backbone interactions can contribute to interaction specificity even though backbone hydrogen bonding groups are present on all residues: Formation of hydrogen bonds with the correct geometry requires precise alignment of often twisted or curved beta strands—the structure of the interacting edge strand in a binder must be matched to the structure of the edge strand it interacts with in the target. Pathological processes such as the formation of amyloid fibrils also involve beta sheet extension (3), and inhibitors that hydrogen bond to the beta sheet have been developed to block such extension (4, 5). Design approaches have been used to create homodimeric structures with extended beta sheets (6, 7) that rely on symmetrical/self-docking of scaffolds to form homodimers. But to date, methods for designing heterodimeric complexes through beta sheet extension in which one component (the target protein) is fixed have not been described, even though targets with an exposed edge strand constitute a substantial class of therapeutically interesting molecules (1, 2).A challenge in designing binding proteins is how to form an extensive binding interface while avoiding the energetically unfavorable burial of nonhydrogen-bonded polar groups on the target. Indeed, previous de novo binder design efforts have focused primarily on sidechain–sidechain interactions with hydrophobic patches on target protein surfaces with few polar groups (8, 9). With this approach, polar regions of a target protein surface are difficult to design binders against as it is very challenging to make sidechain-mediated hydrogen bonds to all the exposed polar groups simultaneously; in particular, the many exposed C=O and N–H groups at the edges of beta sheets are difficult to fully engage with sidechain hydrogen bonds. We reasoned that designed binding proteins with edge beta strands complementary in shape to an exposed beta strand in the target protein could overcome this challenge, as the multiple strand–strand hydrogen bonds could compensate for the loss of interactions with water.