Design strategies for the construction of independently folded polypeptide motifs

Miniprotein engineering for inhibition of PD-1/PD-L1 interaction

Miniproteins constitute an excellent basis for the development of structurally demanding functional molecules. The engrailed homeodomain, a three-helix-containing miniprotein, was applied as a scaffold for constructing programmed cell death protein 1/programmed death-ligand 1 (PD-1/PD-L1) interaction inhibitors. PD-L1 binders were initially designed using the computer-aided approach and subsequently optimized iteratively. The conformational stability was assessed for each obtained miniprotein using circular dichroism spectroscopy, indicating that numerous mutations could be introduced. The formation of a sizable hydrophobic surface at the inhibitor that fits the molecular target imposed the necessity for the incorporation of additional charged amino acid residues to retain its appropriate solubility. Finally, the miniprotein effectively binding to PD-L1 (KD = 51.4 nM) that inhibits PD-1/PD-L1 interaction in cell-based studies with EC50 = 3.9 μM, was discovered.

Design and Engineering of Miniproteins

The potential of miniproteins in the biological and chemical sciences is constantly increasing. Significant progress in the design methodologies has been achieved over the last 30 years. Early approaches based on propensities of individual amino acid residues to form individual secondary structures were subsequently improved by structural analyses using NMR spectroscopy and crystallography. Consequently, computational algorithms were developed, which are now highly successful in designing structures with accuracy often close to atomic range. Further perspectives include construction of miniproteins incorporating non-native secondary structures derived from sequences with units other than α-amino acids. Noteworthy, miniproteins with extended structures, which are now feasibly accessible, are excellent scaffolds for construction of functional molecules.

Exploring the molecular design of protein interaction sites with molecular dynamics simulations and free energy calculations

The significant work that has been invested toward understanding protein-protein interaction has not translated into significant advances in structure-based predictions. In particular redesigning protein surfaces to bind to unrelated receptors remains a challenge, partly due to receptor flexibility, which is often neglected in these efforts. In this work, we computationally graft the binding epitope of various small proteins obtained from the RCSB database to bind to barnase, lysozyme, and trypsin using a previously derived and validated algorithm. In an effort to probe the protein complexes in a realistic environment, all native and designer complexes were subjected to a total of nearly 400 ns of explicit-solvent molecular dynamics (MD) simulation. The MD data led to an unexpected observation: some of the designer complexes were highly unstable and decomposed during the trajectories. In contrast, the native and a number of designer complexes remained consistently stable. The unstable conformers provided us with a unique opportunity to define the structural and energetic factors that lead to unproductive protein-protein complexes. To that end we used free energy calculations following the MM-PBSA approach to determine the role of nonpolar effects, electrostatics and entropy in binding. Remarkably, we found that a majority of unstable complexes exhibited more favorable electrostatics than native or stable designer complexes, suggesting that favorable electrostatic interactions are not prerequisite for complex formation between proteins. However, nonpolar effects remained consistently more favorable in native and stable designer complexes reinforcing the importance of hydrophobic effects in protein-protein binding. While entropy systematically opposed binding in all cases, there was no observed trend in the entropy difference between native and designer complexes. A series of alanine scanning mutations of hot-spot residues at the interface of native and designer complexes showed less than optimal contacts of hot-spot residues with their surroundings in the unstable conformers, resulting in more favorable entropy for these complexes. Finally, disorder predictions revealed that secondary structures at the interface of unstable complexes exhibited greater disorder than the stable complexes.

Malleability of protein folding pathways: a simple reason for complex behaviour

Although the structures of native proteins are generally unique, the pathways by which they form are often free to vary. Some proteins fold by a multitude of different pathways, whereas others seem restricted to only one choice. An explanation for this variation in folding behaviour has recently emerged from studies of transition state changes: the number of accessible pathways is linked to the number of nucleation motifs contained within the native topology. We refer to these nucleation motifs as 'foldons', as they approach the size of an independent cooperative unit. Thus, with respect to pathway malleability and the composition of the folding funnel, proteins can be seen as modular assemblies of competing foldons. For the split beta-alpha-beta fold, these foldons are two-strand-helix motifs coupled by spatial overlap.

D-amino acid residues in peptides and proteins

We have investigated the D-amino acid residues present in Protein Data Bank (PDB) entries, categorizing them into "real" D-residues and artifacts. In polypeptide chains of more than 20 residues, only a single instance of a "real" D-residue, other than those deliberately designed or engineered, was found. This example was the result of a slow chemical epimerization process. Another 12 designed D-residues were found in these longer polypeptide chains. Smaller peptides of 20 or fewer residues contained 479 "real" D-residues, the majority in various gramicidin, actinomycin, or cyclosporin structures. We found 148 PDB entries with "real" D-residues and a further 186, in which all apparent D-residues are artifacts. Investigating the (phi, psi) preferences of the "real" D-residues, we found that the region around (-60 degrees, -45 degrees ) was almost completely unoccupied, even though it is not formally disallowed. We link the low propensity to occupy this region with the alpha-helix destabilizing properties of D-residues.

Hydrogen exchange, core modules, and new designed proteins

A strategy for design of new proteins that mimic folding properties of native proteins is based on peptides modeled on the slow exchange cores of natural proteins. We have synthesized peptides, called core modules, that correspond to the elements of secondary structure that carry the very slowest exchanging amides in a protein. The expectation is that, if soluble in water, core modules will form conformational ensembles that favor native-like structure. Core modules modeled on natural bovine pancreatic trypsin inhibitor have been shown by NMR studies to meet this expectation. The next step toward production of a native state mimic is to further shift the conformational bias of a core module toward more ordered structure by promoting module-module interactions that are mutually stabilizing. For this, two core modules were incorporated into a single molecule by means of a long cross-link. From a panel of several two-module peptides, one very promising lead emerged; it is called BetaCore. BetaCore is monomeric in water and forms a new fold composed of a four-stranded, antiparallel beta-sheet. The single, dominant conformation of BetaCore is characterized by various NMR experiments. Here we compare the individual core module to the two-module BetaCore and discuss the progressive stabilization of intramodule structure and the formation of new intermodule interactions.

Optimal N-caps for N-terminal helical templates: effects of changes in H-bonding efficiency and charge

A family of efficient helix-initiating N-terminal caps X-Hel is introduced that expand the scope and versatility of the previously reported reporting conformational template Ac-Hel, (Kemp, D. S.; Allen, T. J.; Oslick, S. J. Am. Chem. Soc. 1995, 117, 6641-6657) and a working principle for predicting cap performance is described, based on structurally specific intramolecular hydrogen bond formation. Replacement of the N-acetyl by urethane, urea, or sulfonamide generated less efficient polypeptide helix inducers. The N-formyl cap is found to be equivalent to the N-acetyl and may provide more convenient quantitative helix reporting properties. Anionic N-caps derived from the series X = (-)O(2)C-(CH(2))(n)-CO, 0 < or = n < or = 3, are superior to N-acetyl, as are N-acylglycyl and N-acyl-beta-aspartyl. The latter pair of caps permit introduction of the X-Hel functionality within a polypeptide chain, allowing control of helicity of a peptide sub-sequence. Applications of these capping functions are discussed. This work has been focused primarily on immediate practical goals directed toward enhancing the maximum helicity of isolated short to medium-sized peptides in aqueous solution, but its developing concepts and working hypotheses are likely to significantly enhance our understanding at a chemical level of the protein folding problem.

Effect of surface charges on the rates of intermolecular electron-transfer between de novo designed metalloproteins

A de novo designed coiled-coil metalloprotein was prepared that uses electrostatic interactions to control both its conformational and bimolecular electron-transfer properties. The title protein exists as a coiled-coil heterodimer of the [Ru(trpy)(bpy)-KK(37-mer)] and [Ru(NH(3))(5)-EE(37-mer)] polypeptides which is formed by interhelix electrostatic attractions. Circular dichroism studies show that the electrostatic heterodimer has K(d) = 0.19 +/- 0.03 microM and is 96% helical at high concentrations. Intercomplex electron-transfer reactions were studied that involve the [Ru(NH(3))(5)-H21](2+) electron-donor and the [Ru(trpy)(bpy)-H21](3+) electron-acceptor belonging to different electrostatic dimers. An important feature of the designed metalloprotein is its two cationic redox centers embedded within protein surfaces having opposite charge. Thus, the Ru(II)(NH(3))(5)-H21 site was placed on the surface of one chain of the coiled-coil which was made to be positively charged, and the Ru(III)(trpy)(bpy)-H21 site was placed on the surface of the other chain which was negatively charged. The rates of intermolecular electron-transfer increased from (1.9 +/- 0.4) x 10(7) M(-1) s(-1) to (3.7 +/- 0.5) x 10(7) M(-1) s(-1) as the ionic strength was increased from 0.01 to 0.20 M. This indicates that the electrostatic repulsion between the ruthenium centers dominates the kinetics of these reactions. However, the presence of the oppositely charged protein surfaces in the coiled-coils creates an electrostatic recognition domain that substantially ameliorates the effects of this repulsion.

A novel artificial loop scaffold for the noncovalent constraint of peptides

BACKGROUND

Few examples exist of peptides of < 35 residues that form a stable tertiary structure without disulfide bonds. A method for stabilization and noncovalent constraint of relatively short peptides may allow the construction and use of intracellular peptide libraries containing protein minidomains.

RESULTS

We have examined a novel method for the noncovalent constraint of peptides by attaching the peptide EFLIVKS (single-letter amino acid code), which forms dimers, to the amino and carboxyl termini of different peptide inserts. An 18 residue random coil taken from the inhibitor loop of barley chymotrypsin inhibitor 2 was inserted between the peptides to produce a 32-mer minidomain that is attacked only slowly by elastase, has numerous slowly exchanging protons, contains a high beta-structure content and has a T(m) above 37 degrees C. A point mutation disrupting the hydrophobic interior in both dimerizing peptides causes a loss of all slowly exchanging protons and of secondary structure. Adding specific charged residues to each terminus substantially increased the T(m), as did point mutants designed to add interdimerizer ion pairs. Three flexible epitope tag inserts and a nonamer insert do not appear to be folded in a stable structure by EFLIVKS. The properties of two peptides selected for expression in HeLa cells suggest they do form a stable tertiary structure.

CONCLUSIONS

Attaching short dimerizing peptides to both the amino and carboxyl termini of several 18-mer peptides appears to create stable monomeric tertiary structures. Mutations in the dimerizers can either destabilize or significantly stabilize a standard 18-mer insert. Dimerizing peptides flanking random insert sequences could be used as a strategy to generate heterogeneous peptide libraries with both extended and folded members.

Construction and expression of functional multi-domain polypeptides in Escherichia coli: expression of the Neurospora crassa metallothionein gene

A system for the construction of polymeric peptides in Escherichia coli was utilized to prepare a library of plasmids coding for tandem repeats of the Neurospora crassa metallothionein gene. Selected oligomeric metallothionein clones were expressed and targeted to the periplasm as a fusion with the maltose-binding protein. Bacterial cells harbouring the expressed oligopeptides were characterized for their ability to bind 109Cd2+. The metal-binding ability was enhanced for all the oligomeric constructs tested and, in the best case, a 6.5-fold increased capacity for metal uptake was achieved with cells expressing a tandem 9-mer in comparison with cells expressing a monomer. Plateauing of the metal uptake ability occurred at between six and nine tandem repeats, possibly due to a combination of lowered translation levels, inefficient export and prematurely terminated translation products. The overall enhancement of the heavy metal removal capacity was approximately 65-fold relative to non-recombinant cells. The use of this strategy for the design and expression of de novo polypeptides containing multiple functional domains for use in bioremediation is discussed.