Sampling of structure and sequence space of small protein folds.

Nature only samples a small fraction of the sequence space that can fold into stable proteins. Furthermore, small structural variations in a single fold, sometimes only a few amino acids, can define a protein's molecular function. Hence, to design proteins with novel functionalities, such as molecular recognition, methods to control and sample shape diversity are necessary. To explore this space, we developed and experimentally validated a computational platform that can design a wide variety of small protein folds while sampling shape diversity. We designed and evaluated stability of about 30,000 de novo protein designs of eight different folds. Among these designs, about 6,200 stable proteins were identified, including some predicted to have a first-of-its-kind minimalized thioredoxin fold. Obtained data revealed protein folding rules for structural features such as helix-connecting loops. Beyond serving as a resource for protein engineering, this massive and diverse dataset also provides training data for machine learning. We developed an accurate classifier to predict the stability of our designed proteins. The methods and the wide range of protein shapes provide a basis for designing new protein functions without compromising stability.

De novo design of potent and resilient hACE2 decoys to neutralize SARS-CoV-2.

We developed a de novo protein design strategy to swiftly engineer decoys for neutralizing pathogens that exploit extracellular host proteins to infect the cell. Our pipeline allowed the design, validation, and optimization of de novo human angiotensin-converting enzyme 2 (hACE2) decoys to neutralize severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The best monovalent decoy, CTC-445.2, bound with low nanomolar affinity and high specificity to the receptor-binding domain (RBD) of the spike protein. Cryo-electron microscopy (cryo-EM) showed that the design is accurate and can simultaneously bind to all three RBDs of a single spike protein. Because the decoy replicates the spike protein target interface in hACE2, it is intrinsically resilient to viral mutational escape. A bivalent decoy, CTC-445.2d, showed ~10-fold improvement in binding. CTC-445.2d potently neutralized SARS-CoV-2 infection of cells in vitro, and a single intranasal prophylactic dose of decoy protected Syrian hamsters from a subsequent lethal SARS-CoV-2 challenge.

De novo design of ACE2 protein decoys to neutralize SARS-CoV-2.

There is an urgent need for the ability to rapidly develop effective countermeasures for emerging biological threats, such as the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) that causes the ongoing coronavirus disease 2019 (COVID-19) pandemic. We have developed a generalized computational design strategy to rapidly engineer de novo proteins that precisely recapitulate the protein surface targeted by biological agents, like viruses, to gain entry into cells. The designed proteins act as decoys that block cellular entry and aim to be resilient to viral mutational escape. Using our novel platform, in less than ten weeks, we engineered, validated, and optimized de novo protein decoys of human angiotensin-converting enzyme 2 (hACE2), the membrane-associated protein that SARS-CoV-2 exploits to infect cells. Our optimized designs are hyperstable de novo proteins (∼18-37 kDa), have high affinity for the SARS-CoV-2 receptor binding domain (RBD) and can potently inhibit the virus infection and replication in vitro. Future refinements to our strategy can enable the rapid development of other therapeutic de novo protein decoys, not limited to neutralizing viruses, but to combat any agent that explicitly interacts with cell surface proteins to cause disease.

Accurate de novo design of hyperstable constrained peptides.

Naturally occurring, pharmacologically active peptides constrained with covalent crosslinks generally have shapes that have evolved to fit precisely into binding pockets on their targets. Such peptides can have excellent pharmaceutical properties, combining the stability and tissue penetration of small-molecule drugs with the specificity of much larger protein therapeutics. The ability to design constrained peptides with precisely specified tertiary structures would enable the design of shape-complementary inhibitors of arbitrary targets. Here we describe the development of computational methods for accurate de novo design of conformationally restricted peptides, and the use of these methods to design 18-47 residue, disulfide-crosslinked peptides, a subset of which are heterochiral and/or N-C backbone-cyclized. Both genetically encodable and non-canonical peptides are exceptionally stable to thermal and chemical denaturation, and 12 experimentally determined X-ray and NMR structures are nearly identical to the computational design models. The computational design methods and stable scaffolds presented here provide the basis for development of a new generation of peptide-based drugs.

Discovery of structurally-diverse inhibitor scaffolds by high-throughput screening of a fragment library with dimethylarginine dimethylaminohydrolase.

Potent and selective inhibitors of the enzyme dimethylarginine dimethylaminohydrolase (DDAH) are useful as molecular probes to better understand cellular regulation of nitric oxide. Inhibitors are also potential therapeutic agents for treatment of pathological states associated with the inappropriate overproduction of nitric oxide, such as septic shock, selected types of cancer, and other conditions. Inhibitors with structures dissimilar to substrate may overcome limitations inherent to substrate analogs. Therefore, to identify structurally-diverse inhibitor scaffolds, high-throughput screening (HTS) of a 4000-member library of fragment-sized molecules was completed using the Pseudomonas aeruginosa DDAH and human DDAH-1 isoforms. Use of a substrate concentration equal to its K(M) value during the primary screen allowed for the detection of inhibitors with different modes of inhibition. A series of validation tests were designed and implemented in the identification of four inhibitors of human DDAH-1 that were unknown prior to the screen. Two inhibitors share a 4-halopyridine scaffold and act as quiescent affinity labels that selectively and covalently modify the active-site Cys residue. Two inhibitors are benzimidazole-like compounds that reversibly and competitively inhibit human DDAH-1 with Ligand Efficiency values ≥0.3 kcal/mol/heavy (non-hydrogen) atom, indicating their suitability for further development. Both inhibitor scaffolds have available sites to derivatize for further optimization. Therefore, use of this fragment-based HTS approach is demonstrated to successfully identify two novel scaffolds for development of DDAH-1 inhibitors.

On the mechanism of dimethylarginine dimethylaminohydrolase inactivation by 4-halopyridines.

Small molecules capable of selective covalent protein modification are of significant interest for the development of biological probes and therapeutics. We recently reported that 2-methyl-4-bromopyridine is a quiescent affinity label for the nitric oxide controlling enzyme dimethylarginine dimethylaminohydrolase (DDAH) (Johnson, C. M.; Linsky, T. W.; Yoon, D. W.; Person, M. D.; Fast, W. J. Am. Chem. Soc. 2011, 133, 1553-1562). Discovery of this novel protein modifier raised the possibility that the 4-halopyridine motif may be suitable for wider application. Therefore, the inactivation mechanism of the related compound 2-hydroxymethyl-4-chloropyridine is probed here in more detail. Solution studies support an inactivation mechanism in which the active site Asp66 residue stabilizes the pyridinium form of the inactivator, which has enhanced reactivity toward the active site Cys, resulting in covalent bond formation, loss of the halide, and irreversible inactivation. A 2.18 Å resolution X-ray crystal structure of the inactivated complex elucidates the orientation of the inactivator and its covalent attachment to the active site Cys, but the structural model does not show an interaction between the inactivator and Asp66. Molecular modeling is used to investigate inactivator binding, reaction, and also a final pyridinium deprotonation step that accounts for the apparent differences between the solution-based and structural studies with respect to the role of Asp66. This work integrates multiple approaches to elucidate the inactivation mechanism of a novel 4-halopyridine "warhead," emphasizing the strategy of using pyridinium formation as a "switch" to enhance reactivity when bound to the target protein.

Discovery of halopyridines as quiescent affinity labels: inactivation of dimethylarginine dimethylaminohydrolase.

In an effort to develop novel covalent modifiers of dimethylarginine dimethylaminohydrolase (DDAH) that are useful for biological applications, a set of "fragment"-sized inhibitors that were identified using a high-throughput screen are tested for time-dependent inhibition. One structural class of inactivators, 4-halopyridines, show time- and concentration-dependent inactivation of DDAH, and the inactivation mechanism of one example, 4-bromo-2-methylpyridine (1), is characterized in detail. The neutral form of halopyridines is not very reactive with excess glutathione. However, 1 readily reacts, with loss of its halide, in a selective, covalent, and irreversible manner with the active-site Cys249 of DDAH. This active-site Cys is not particularly reactive (pK(a) ca. 8.8), and 1 does not inactivate papain (Cys pK(a) ca. ≤4), suggesting that, unlike many reagents, Cys nucleophilicity is not a predominating factor in selectivity. Rather, binding and stabilization of the more reactive pyridinium form of the inactivator by a second moiety, Asp66, is required for facile reaction. This constraint imparts a unique selectivity profile to these inactivators. To our knowledge, halopyridines have not previously been reported as protein modifiers, and therefore they represent a first-in-class example of a novel type of quiescent affinity label.

Promiscuous partitioning of a covalent intermediate common in the pentein superfamily.

Many enzymes in the pentein superfamily use a transient covalent intermediate in their catalytic mechanisms. Here we trap and determine the structure of a stable covalent adduct that mimics this intermediate using a mutant dimethylarginine dimethylaminohydrolase and an alternative substrate. The interactions observed between the enzyme and trapped adduct suggest an altered angle of attack between the nucleophiles of the first and second half-reactions of normal catalysis. The stable covalent adduct is also capable of further reaction. Addition of imidazole rescues the original hydrolytic activity. Notably, addition of other amines instead yields substituted arginine products, which arise from partitioning of the intermediate into the evolutionarily related amidinotransferase reaction pathway. The enzyme provides both selectivity and catalysis for the amidinotransferase reaction, underscoring commonalities among the reaction pathways in this mechanistically diverse enzyme superfamily. The promiscuous partitioning of this intermediate may also help to illuminate the evolutionary history of these enzymes.