Ras is at the hub of signal transduction pathways controlling cell proliferation and survival. Its mutants, present in about 30% of human cancers, are major drivers of oncogenesis and render tumors unresponsive to standard therapies. Here we report the engineering of a protein scaffold for preferential binding to K-Ras G12D. This is the first reported inhibitor to achieve nanomolar affinity while exhibiting specificity for mutant over wild type (WT) K-Ras. Crystal structures of the protein R11.1.6 in complex with K-Ras WT and K-Ras G12D offer insight into the structural basis for specificity, highlighting differences in the switch I conformation as the major defining element in the higher affinity interaction. R11.1.6 directly blocks interaction with Raf and reduces signaling through the Raf/MEK/ERK pathway. Our results support greater consideration of the state of switch I and provide a novel tool to study Ras biology. Most importantly, this work makes an unprecedented contribution to Ras research in inhibitor development strategy by revealing details of a targetable binding surface. Unlike the polar interfaces found for Ras/effector interactions, the K-Ras/R11.1.6 complex reveals an extensive hydrophobic interface that can serve as a template to advance the development of high affinity, non-covalent inhibitors of K-Ras oncogenic mutants.

The Ras proteins are essential GTPases involved in the regulation of cell proliferation and survival. Mutated oncogenic forms of Ras alter effector binding and innate GTPase activity, leading to deregulation of downstream signal transduction. Mutated forms of Ras are involved in approximately 30% of human cancers. Despite decades of effort to develop direct Ras inhibitors, Ras has long been considered ‘undruggable’ due to its high affinity for GTP and its lack of hydrophobic binding pockets. Herein, we report a total chemical synthesis of all L- and all D-amino acid biotinylated variants of oncogenic mutant KRas(G12V). The protein is synthesized using Fmoc-based solid-phase peptide synthesis and assembled using combined native chemical ligation and isonitrile-mediated activation strategies. We demonstrate that both KRas(G12V) enantiomers can successfully fold and bind nucleotide substrates and binding partners with observable enantiodiscrimination. By demonstrating the functional competency of a mirror-image form of KRas bound to its corresponding enantiomeric nucleotide triphosphate, this study sets the stage for further biochemical studies with this material. In particular, this protein will enable mirror-image yeast surface display experiments to identify all-D peptide ligands for oncogenic KRas, providing a useful tool in the search for new therapeutics against this challenging disease target.

The Ras proteins are aberrantly activated in a wide range of human cancers, often endowing tumors with aggressive properties and resistance to therapy. Decades of effort to develop direct Ras inhibitors for clinical use have thus far failed, largely because of a lack of adequate small-molecule–binding pockets on the Ras surface. Here, we report the discovery of Ras-binding miniproteins from a naïve library and their evolution to afford versions with midpicomolar affinity to Ras. A series of biochemical experiments indicated that these miniproteins bind to the Ras effector domain as dimers, and high-resolution crystal structures revealed that these miniprotein dimers bind Ras in an unprecedented mode in which the Ras effector domain is remodeled to expose an extended pocket that connects two isolated pockets previously found to engage small-molecule ligands. We also report a Ras point mutant that stabilizes the protein in the open conformation trapped by these miniproteins. These findings provide new tools for studying Ras structure and function and present opportunities for the development of both miniprotein and small-molecule inhibitors that directly target the Ras proteins.

The α-helix is one of the most common protein surface recognition motifs found in nature, and its unique amide-cloaking properties also enable α-helical polypeptide motifs to exist in membranes. Together, these properties have inspired the development of α-helically constrained (Helicon) therapeutics that can enter cells and bind targets that have been considered “undruggable”, such as protein–protein interactions. To date, no general method for discovering α-helical binders to proteins has been reported, limiting Helicon drug discovery to only those proteins with previously characterized α-helix recognition sites, and restricting the starting chemical matter to those known α-helical binders. Here, we report a general and rapid screening method to empirically map the α-helix binding sites on a broad range of target proteins in parallel using large, unbiased Helicon phage display libraries and next-generation sequencing. We apply this method to screen six structurally diverse protein domains, only one of which had been previously reported to bind isolated α-helical peptides, discovering 20 families that collectively comprise several hundred individual Helicons. Analysis of 14 X-ray cocrystal structures reveals at least nine distinct α-helix recognition sites across these six proteins, and biochemical and biophysical studies show that these Helicons can block protein–protein interactions, inhibit enzymatic activity, induce conformational rearrangements, and cause protein dimerization. We anticipate that this method will prove broadly useful for the study of protein recognition and for the development of both biochemical tools and therapeutics for traditionally challenging protein targets.

Mirror-image biological systems have the potential for broad-reaching impact in health and diagnostics, but their study has been greatly limited by the lack of routine access to synthetic D-proteins. We demonstrate that automated fast flow peptide synthesis (AFPS) can reliably produce novel mirror-image protein targets without prior sequence engineering. We synthesized 12 D-proteins, along with their L-counterparts. All 24 synthetic proteins were folded into active structures in vitro, and characterized using biochemical and biophysical techniques. From these chiral protein pairs, we chose MDM2 and CHIP to carry forward into mirror-image phage display screens, and identified macrocyclic D-peptides that bind the recombinant targets. We report 6 mirror-image peptide ligands with unique binding modes: three to MDM2, and three to CHIP, each confirmed with X-ray co-crystal structures. Reliable production of mirror-image proteins with AFPS stands to enable not only the discovery of D-peptide drug leads, but to the study of mirror-image biological systems more broadly.