Proteins are molecular machines whose function depends on their ability to achieve complex folds with precisely defined structural and dynamic properties. The rational design of proteins from first-principles, or de novo, was once considered to be impossible, but today proteins with a variety of folds and functions have been realized. We review the evolution of the field from its earliest days, placing particular emphasis on how this endeavor has illuminated our understanding of the principles underlying the folding and function of natural proteins, and is informing the design of macromolecules with unprecedented structures and properties. An initial set of milestones in de novo protein design focused on the construction of sequences that folded in water and membranes to adopt folded conformations. The first proteins were designed from first-principles using very simple physical models. As computers became more powerful, the use of the rotamer approximation allowed one to discover amino acid sequences that stabilize the desired fold. As the crystallographic database of protein structures expanded in subsequent years, it became possible to construct proteins by assembling short backbone fragments that frequently recur in Nature. The second set of milestones in de novo design involves the discovery of complex functions. Proteins have been designed to bind a variety of metals, porphyrins, and other cofactors. The design of proteins that catalyze hydrolysis and oxygen-dependent reactions has progressed significantly. However, de novo design of catalysts for energetically demanding reactions, or even proteins that bind with high affinity and specificity to highly functionalized complex polar molecules remains an importnant challenge that is now being achieved. Finally, the protein design contributed significantly to our understanding of membrane protein folding and transport of ions across membranes. The area of membrane protein design, or more generally of biomimetic polymers that function in mixed or non-aqueous environments, is now becoming increasingly possible.

Cytokine release syndrome (CRS), or ‘cytokine storm’, is the leading side effect during chimeric antigen receptor (CAR)-T therapy that is potentially life-threatening. It also plays a critical role in viral infections such as Coronavirus Disease 2019 (COVID-19). Therefore, efficient removal of excessive cytokines is essential for treatment. We previously reported a novel protein modification tool called the QTY code, through which hydrophobic amino acids Leu, Ile, Val and Phe are replaced by Gln (Q), Thr (T) and Tyr (Y). Thus, the functional detergent-free equivalents of membrane proteins can be designed. Here, we report the application of the QTY code on six variants of cytokine receptors, including interleukin receptors IL4Rα and IL10Rα, chemokine receptors CCR9 and CXCR2, as well as interferon receptors IFNγR1 and IFNλR1. QTY-variant cytokine receptors exhibit physiological properties similar to those of native receptors without the presence of hydrophobic segments. The receptors were fused to the Fc region of immunoglobulin G (IgG) protein to form an antibody-like structure. These QTY code-designed Fc-fusion receptors were expressed in Escherichia coli and purified. The resulting water-soluble fusion receptors bind to their respective ligands with Kd values affinity similar to isolated native receptors. Our cytokine receptor–Fc-fusion proteins potentially serve as an antibody-like decoy to dampen the excessive cytokine levels associated with CRS and COVID-19 infection.

Residue Asn47 at position L1 of a type II′ β-turn of the α-spectrin SH3 domain is located in a disallowed region of the Ramachandran plot (φ = 56 ± 12, ψ = −118 ± 17). Therefore, it is expected that replacement of Asn47 by Gly should result in a considerable stabilization of the protein. Thermodynamic analysis of the N47G and N47A mutants shows that the change in free energy is small (∼0.7 kcal/mol; ∼3 kJ/mol) and comparable to that found when mutating a Gly to Ala in a α-helix or β-sheet. X-ray structural analysis of these mutants shows that the conformation of the β-turn does not change upon mutation and, therefore, that there is no relaxation of the structure, nor is there any gain or loss of interactions that could explain the small energy change. Our results indicate that the energetic definition of II′ region of the Ramachandran plot (φ = 60 ± 30, ψ = −115 ± 15) should be revised for at least Ala and Asn in structure validation and protein design.

An important goal of protein design is to understand the forces that stabilize a particular fold in preference to alternative folds. Here, we describe an extension of earlier studies in which we successfully designed a stable, native-like helical protein that is 50% identical in sequence to a predominantly β-sheet protein, the B1 domain of Streptococcal IgG-binding protein G. We report the characteristics of a series of variants of our original design that have even higher sequence identity to the B1 domain. Their properties illustrate the extent to which protein stability and conformation can be modulated through careful manipulation of key amino acid residues. Our results have implications for understanding conformational change phenomena of central biological importance and in probing the malleability of the sequence/structure relationship.

Six helix surface positions of protein G (Gβ1) were redesigned using a computational protein design algorithm, resulting in the five fold mutant Gβ1m2. Gβ1m2 is well folded with a circular dichroism spectrum nearly identical to that of Gβ1, and a melting temperature of 91 °C, ∼6 °C higher than that of Gβ1. The crystal structure of Gβ1m2 was solved to 2.0 Å resolution by molecular replacement. The absence of hydrogen bond or salt bridge interactions between the designed residues in Gβ1m2 suggests that the increased stability of Gβ1m2 is due to increased helix propensity and more favorable helix dipole interactions.

We have developed a method for the prediction of an amino acid sequence that is compatible with a three-dimensional backbone structure. Using only a backbone structure of a protein as input, the algorithm is capable of designing sequences that closely resemble natural members of the protein family to which the template structure belongs. In general, the predicted sequences are shown to have multiple sequence profile scores that are dramatically higher than those of random sequences, and sometimes better than some of the natural sequences that make up the superfamily. As anticipated, highly conserved but poorly predicted residues are often those that contribute to the functional rather than structural properties of the protein. Overall, our analysis suggests that statistical profile scores of designed sequences are a novel and valuable figure of merit for assessing and improving protein design algorithms.

A new computer program (CORE) is described that predicts core hydrophobic sequences of predetermined target protein structures. A novel scoring function is employed, which for the first time incorporates parameters directly correlated to free energies of unfolding (ΔGu), melting temperatures (Tm), and cooperativity. Metropolis-driven simulated annealing and low-temperature Monte Carlo sampling are used to optimize this score, generating sequences predicted to yield uniquely folded, stable proteins with cooperative unfolding transitions. The hydrophobic core residues of four natural proteins were predicted using CORE with the backbone structure and solvent exposed residues as input. In the two smaller proteins tested (Gβ1, 11 core amino acids; 434 cro, 10 core amino acids), the native sequence was regenerated as well as the sequence of known thermally stable variants that exhibit cooperative denaturation transitions. Previously designed sequences of variants with lower thermal stability and weaker cooperativity were not predicted. In the two larger proteins tested (myoglobin, 32 core amino acids; methionine aminopeptidase, 63 core amino acids), sequences with corresponding side-chain conformations remarkably similar to that of native were predicted.

We present direct evidence for a change in protein structural specificity due to hydrophobic core packing. High resolution structural analysis of a designed core variant of ubiquitin reveals that the protein is in slow exchange between two conformations. Examination of side-chain rotamers indicates that this dynamic response and the lower stability of the protein are coupled to greater strain and mobility in the core. The results suggest that manipulating the level of side-chain strain may be one way of fine tuning the stability and specificity of proteins.

We describe here a systematic investigation into the role of position a in the hydrophobic core of a model coiled-coil protein in determining coiled-coil stability and oligomerization state. We employed a model coiled coil that allowed the formation of an extended three-stranded trimeric oligomerization state for some of the analogs; however, due to the presence of a Cys-Gly-Gly linker, unfolding occurred from the same two-stranded monomeric oligomerization state for all of the analogs. Denaturation from a two-stranded state allowed us to measure the relative contribution of 20 different amino acid side chains to coiled-coil stability from chemical denaturation profiles. In addition, the relative hydrophobicity of the substituted amino acid side chains was assessed by reversed-phase high-performance liquid chromatography and found to correlate very highly (R = 0.95) with coiled-coil stability. We also determined the effect of position a in specifying the oligomerization state using ultracentrifugation as well as high-performance size-exclusion chromatography. We found that nine of the analogs populated one oligomerization state exclusively at peptide concentrations of 50 μM under benign buffer conditions. The Leu-, Tyr-, Gln-, and His-substituted analogs were found to be exclusively three-stranded trimers, while the Asn-, Lys-, Orn-, Arg-, and Trp-substituted analogs formed exclusively two-stranded monomers. Modeling results for the Leu-substituted analog showed that a three-stranded oligomerization state is preferred due to increased side-chain burial, while a two-stranded oligomerization state was observed for the Trp analog due to unfavorable cavity formation in the three-stranded state.

The tumor suppressor function of the wild-type p53 protein is transdominantly inhibited by tumor-derived mutant p53 proteins. Such transdominant inhibition limits the prospects for gene therapy approaches that aim to introduce wild-type p53 into cancer cells. The molecular mechanism for transdominant inhibition involves sequestration of wild-type p53 subunits into inactive wild-type/mutant hetero-tetramers. Thus, p53 proteins, whose oligomerization specificity is altered so they cannot interact with tumor-derived mutant p53, would escape transdominant inhibition. Aided by the known three-dimensional structure of the p53 tetramerization domain and by trial and error we designed a novel domain with seven amino acid substitutions in the hydrophobic core. A full-length p53 protein bearing this novel domain formed homo-tetramers and had tumor suppressor function, but did not hetero-oligomerize with tumor-derived mutant p53 and resisted transdominant inhibition. Thus, hydrophobic core residues influence the oligomerization specificity of the p53 tetramerization domain.

Electrostatic interactions are often critical for determining the specificity of protein-protein complexes. To study the role of electrostatic interactions for assembly of helical bundles, we previously designed a thermostable, heterotrimeric coiled coil, ABC, in which charged residues were employed to drive preferential association of three distinct, 34-residue helices. To investigate the basis for heterotrimer specificity, we have used multiwavelength anomalous diffraction (MAD) analysis to determine the 1.8 Å resolution crystal structure of ABC. The structure shows that ABC forms a heterotrimeric coiled coil with the intended arrangement of parallel chains. Over half of the ion pairs engineered to restrict helix associations were apparent in the experimental electron density map. As seen in other trimeric coiled coils, ABC displays acute knobs-into-holes packing and a buried anion coordinated by core polar amino acids. These interactions validate the design strategy and illustrate how packing and polar contacts determine structural uniqueness.