The determination of multiple energy minima on complex potential energy surfaces is challenging. A systematic desymmetrization procedure was employed to find stationary points on the copper(I) + chloride + water potential energy surface using HF, MP2, and B3LYP methods in conjunction with the 6-31G*, 6-31+G*, and 6-311+G* basis sets. Comparison with experimental results demonstrated that the speciation of copper(I) in the presence of chloride and water may be formulated as [CuCl(H2O)]0, [CuCl2]−, and [CuCl3]2−. Our results indicate that the combination of the MP2 method along with basis sets containing diffuse functions gives excellent agreement with experimental Cu-Cl distances and vibrational frequencies. Poorer results were obtained at the HF levels and/or using the 6-31G* basis set.

MOTIVATION:Predicting the structure of antibody-antigen complexes is a challenging task with significant implications for the design of better antibody therapeutics. However, the levels of success have remained dauntingly low, particularly when high standards for model quality are required, a necessity for efficient antibody design. Artificial intelligence (AI) has significantly impacted the landscape of structure prediction for antibodies, both alone and in complex with their antigens. METHODS:We utilized AI-guided antibody modeling tools to generate ensembles displaying diversity in the complementarity-determining region (CDR) and integrated those into our previously published AlphaFold2-rescored docking pipeline, a strategy called AI-augmented physics-based docking. In this study, we also compare docking performance with AlphaFold and Boltz-1, the new state-of-the-art. We distinguish between two types of success tailored to specific downstream applications: (i) criteria sufficient for epitope mapping, where gross quality is adequate and can complement experimental techniques, and (ii) criteria for producing higher-quality models suitable for engineering purposes. RESULTS:We highlight that the quality of the ensemble is crucial for docking performance, that including too many models can be detrimental, and that prioritization of models is essential for achieving good performance. In a scenario analogous to docking using a crystallized antigen, our results robustly demonstrate the advantages of AI-augmented docking over AlphaFold2, further accentuated when higher standards in quality are imposed. Docking also shows improvements over Boltz-1, but those are less pronounced. Docking performance is still noticeably lower than AlphaFold3 in both epitope mapping and antibody design use cases. We observe a strong dependence on CDR-H3 loop length for physics-based tools on their ability to successfully predict. This helps define an applicability range where physics-based docking can be competitive to the newer generation of AI tools. AVAILABILITY AND IMPLEMENTATION:The AF2 rescoring scripts are available at github.com/gaudreaultfnrc/AF2-Rescoring.

Scoring functions are ubiquitous in structure-based drug design as an aid to predicting binding modes and estimating binding affinities. Ideally, a scoring function should be broadly applicable, obviating the need to recalibrate and refit its parameters for every new target and class of ligands. Traditionally, drugs have been small molecules, but in recent years biologics, particularly antibodies, have become an increasingly important if not dominant class of therapeutics. This makes the goal of having a transferable scoring function, i.e., one that spans the range of small-molecule to protein ligands, even more challenging. One such broadly applicable scoring function is the Solvated Interaction Energy (SIE), which has been developed and applied in our lab for the last 15 years, leading to several important applications. This physics-based method arose from efforts to understand the physics governing binding events, with particular care given to the role played by solvation. SIE has been used by us and many independent labs worldwide for virtual screening and discovery of novel small-molecule binders or optimization of known drugs. Moreover, without any retraining, it is found to be transferrable to predictions of antibody-antigen relative binding affinities and as accurate as functions trained on protein-protein binding affinities. SIE has been incorporated in conjunction with other scoring functions into ADAPT (Assisted Design of Antibody and Protein Therapeutics), our platform for affinity modulation of antibodies. Application of ADAPT resulted in the optimization of several antibodies with 10-to-100-fold improvements in binding affinity. Further applications included broadening the specificity of a single-domain antibody to be cross-reactive with virus variants of both SARS-CoV-1 and SARS-CoV-2, and the design of safer antibodies by engineering of a pH switch to make them more selective towards acidic tumors while sparing normal tissues at physiological pH.

The ADAPT (Assisted Design of Antibody and Protein Therapeutics) platform guides the selection of mutants that improve/modulate the affinity of antibodies and other biologics. Predicted affinities are based on a consensus z-score from three scoring functions. Computational predictions are interleaved with experimental validation, significantly enhancing the robustness of the design and selection of mutants. A key step is an initial exhaustive virtual single-mutant scan that identifies hot spots and the mutations predicted to improve affinity. A small number of proposed single mutants are then produced and assayed. Only the validated single mutants (i.e., having improved affinity) are used to design double and higher-order mutants in subsequent rounds of design, avoiding the combinatorial explosion that arises from random mutagenesis. Typically, with a total of about 30-50 designed single, double, and triple mutants, affinity improvements of 10- to 100-fold are obtained.

Accurate protein-protein docking remains challenging, especially for artificial biologics not coevolved naturally against their protein targets, like antibodies and other engineered scaffolds. We previously developed ProPOSE, an exhaustive docker with full atomistic details, which delivers cutting-edge performance by allowing side-chain rearrangements upon docking. However, extensive protein backbone flexibility limits its practical applicability as indicated by unbound docking tests. To explore the usefulness of ProPOSE on systems with limited backbone flexibility, here we tested the engineered scaffold DARPin, which is characterized by its relatively rigid protein backbone. A prospective screening campaign was undertaken, in which sequence-diversified DARPins were docked and ranked against a directed epitope on the target protein BCL-W. In this proof-of-concept study, only a relatively small set of 2,213 diverse DARPin interfaces were selected for docking from the huge theoretical library from mutating 18 amino-acid positions. A computational selection protocol was then applied for enrichment of binders based on normalized computed binding scores and frequency of binding modes against the predefined epitope. The top-ranked 18 designed DARPin interfaces were selected for experimental validation. Three designs exhibited binding affinities to BCL-W in the nanomolar range comparable to control interfaces adopted from known DARPin binders. This result is encouraging for future screening and engineering campaigns of DARPins and possibly other similarly rigid scaffolds against targeted protein epitopes. Method limitations are discussed and directions for future refinements are proposed.

Predicting the structure of antibody-antigen complexes has tremendous value in biomedical research but unfortunately suffers from a poor performance in real-life applications. AlphaFold2 (AF2) has provided renewed hope for improvements in the field of protein–protein docking but has shown limited success against antibody-antigen complexes due to the lack of co-evolutionary constraints. In this study, we used physics-based protein docking methods for building decoy sets consisting of low-energy docking solutions that were either geometrically close to the native structure (positives) or not (negatives). The docking models were then fed into AF2 to assess their confidence with a novel composite score based on normalized pLDDT and pTMscore metrics after AF2 structural refinement. We show benefits of the AF2 composite score for rescoring docking poses both in terms of (1) classification of positives/negatives and of (2) success rates with particular emphasis on early enrichment. Docking models of at least medium quality present in the decoy set, but not necessarily highly ranked by docking methods, benefitted most from AF2 rescoring by experiencing large advances towards the top of the reranked list of models. These improvements, obtained without any calibration or novel methodologies, led to a notable level of performance in antibody-antigen unbound docking that was never achieved previously.

Single-domain antibodies (sdAbs) are a promising class of biotherapeutics with unique structural traits within their paratope region. The distribution of canonical conformations explored by their complementarity determining region (CDR) loops differs to some extent from conventional two-chain Fv fragments of monoclonal antibodies (mAbs). In this study, we explored in detail the canonical structures of sdAb CDR-H1 and CDR-H2 loops and compared those with mAbs from the IGHV3 and IGHV1 gene families. We surveyed the antibody structures catalogued in SAbDab and clustered the CDR canonical loops in Cartesian space. While most of the sdAb clusters were sub-populations of previously defined canonical Fv conformations of CDR-H1 and CDR-H2, our stricter clustering approach defined narrower clusters in sequence-space. Meticulous visual inspection of sub-populations allowed a clearer understanding of sequence-structure relationships. The packing densities within structural pockets contacted by CDR-H1 and CDR-H2 canonical conformations were analyzed on the premise that these pockets cannot be left vacant as they would leave exposed supportive hydrophobic residues. The fine resolution of the canonical clusters defined here revealed unique signatures within these pockets, including distinct structural complementarities between CDR-H1 and CDR-H2 canonical clusters, which could not be perceived with the previous coarser clusters. We highlight examples where a single residue change in CDR-H1 sequence is sufficient to induce a dramatic population shift in CDR-H2 conformation. This suggests that preferences in combining CDR-H1 and CDR-H2 emerged naturally during antibody evolution, leading to preferred sets of conserved amino acids at key positions in the framework as well as within the CDR loops. We outline a game of musical chairs that is necessary to maintain the integrity of the antibody structures that arose during evolution. Our study also provides refined CDR-H1 and CDR-H2 structural templates for sdAb homology modeling that could be leveraged for improved antibody design.

Antibody-based therapeutics for treatment of various tumors have grown rapidly in recent years. Unfortunately, safety issues, attributed to off-tumor effects and cytotoxicity, are still a significant concern with the standard of care. Improvements to ensure targeted delivery of antitumor pharmaceuticals are desperately needed. We previously demonstrated that incorporating histidyl pH-switches in an anti-HER2 antibody induced selective antigen binding under acidic pH conditions (MAbs 2020;12:1682866). This led to an improved safety profile due to preferential targeting of the oncoprotein in the acidic solid tumor microenvironment. Following this success, we expanded this approach to a set of over 400 antibody structures complexed with over 100 different human oncoproteins, associated with solid tumors. Calculations suggested that mutations to His of certain residue types, namely Trp, Arg, and Tyr, could be significantly more successful for inducing pH-dependent binding under acidic conditions. Furthermore, 10 positions within the complementarity-determining region were also predicted to exhibit greater successes. Combined, these two accessible metrics could serve as the basis for a sequence-based engineering of pH-selective binding. This approach could be applied to most anticancer antibodies, which lack detailed structural characterization.

To date, only a limited number of cutinases have been isolated and characterized from nature, the majority being sourced from phytopathogenic fungi and thermophilic bacteria. The significance of our research relates to the identification and characterization of a unique member of microbial cutinases, named KrCUT, that was derived from the genome of the Gram-positiveKineococcus radiotoleransSRS30216, a highly radiation-resistant actinobacterium.

The design of superior biologic therapeutics, including antibodies and engineered proteins, involves optimizing their specific ability to bind to disease-related molecular targets. Previously, we developed and applied the Assisted Design of Antibody and Protein Therapeutics (ADAPT) platform for virtual affinity maturation of antibodies (Vivcharuk et al. in PLoS One 12(7):e0181490, https://doi.org/10.1371/journal.pone.0181490 , 2017). However, ADAPT is limited to point mutations of hot-spot residues in existing CDR loops. In this study, we explore the possibility of wholesale replacement of the entire H3 loop with no restriction to maintain the parental loop length. This complements other currently published studies that sample replacements for the CDR loops L1, L2, L3, H1 and H2. Given the immense sequence space theoretically available to H3, we focused on the virtual grafting of over 5000 human germline-derived H3 sequences from the IGMT/LIGM database increasing the diversity of the sequence space when compared to using crystalized H3 loop sequences. H3 loop conformations are generated and scored to identify optimized H3 sequences. Experimental testing of high-ranking H3 sequences grafted into the framework of the bH1 antibody against human VEGF-A led to the discovery of multiple hits, some of which had similar or better affinities relative to the parental antibody. In over 75% of the tested designs, the re-designed H3 loop contributed favorably to overall binding affinity. The hits also demonstrated good developability attributes such as high thermal stability and no aggregation. Crystal structures of select re-designed H3 variants were solved and indicated that although some deviations from predicted structures were seen in the more solvent accessible regions of the H3 loop, they did not significantly affect predicted affinity scores.

Abstract Group theoretical arguments are used to find the subgroup corresponding to symmetry reduction along a particular irreducible representation of a group. The results are used to guide the search for stationary points on the potential energy surface of hydrated copper(I) ion at the HF/6-31G∗, HF/6-31+G∗, HF/6-311+G∗, MP2/6-31G∗, MP2/6-31+G∗, MP2/6-311+G∗, B3LYP/6-31G∗, B3LYP/6-31+G∗, and B3LYP/6-311+G∗ levels. The better levels give the most stable coordination number of two. The effect of desymmetrization on the Cu-O distances and stretching frequencies has been examined.

Recent development of monoclonal antibodies as mainstream anticancer agents demands further optimization of their safety for use in humans. Potent targeting and/or effector activities on normal tissues is an obvious toxicity concern. Optimization of specific tumor targeting could be achieved by taking advantage of the extracellular acidity of solid tumors relative to normal tissues. Here, we applied a structure-based computational approach to engineer anti-human epidermal growth factor receptor 2 (Her2) antibodies with selective binding in the acidic tumor microenvironment. We used an affinity maturation platform in which dual-pH histidine-scanning mutagenesis was implemented for pH selectivity optimization. Testing of a small set of designs for binding to the recombinant Her2 ectodomain led to the identification of antigen-binding fragment (Fab) variants with the desired pH-dependent binding behavior. Binding selectivity toward acidic pH was improved by as much as 25-fold relative to the parental bH1-Fab. In vitro experiments on cells expressing intact Her2 confirmed that designed variants formatted as IgG1/k full-size antibodies have high affinity and inhibit the growth of tumor spheroids at a level comparable to that of the benchmark anti-Her2 antibody trastuzumab (Herceptin®) at acidic pH, whereas these effects were significantly reduced at physiological pH. In contrast, both Herceptin and the parental bH1 antibody exhibited strong cell binding and growth inhibition irrespective of pH. This work demonstrates the feasibility of computational optimization of antibodies for selective targeting of the acidic environment such as that found in many solid tumors.

Effective biologic therapeutics require binding affinities that are fine-tuned to their disease-related molecular target. The ADAPT (Assisted Design of Antibody and Protein Therapeutics) platform aids in the selection of mutants that improve/modulate the affinity of antibodies and other biologics. It uses a consensus z-score from three scoring functions and interleaves computational predictions with experimental validation, significantly enhancing the robustness of the design and selection of mutants. The platform was tested on three antibody Fab-antigen systems that spanned a wide range of initial binding affinities: bH1-VEGFA- (44 nM), bH1-HER2 (3.6 nM) and Herceptin-HER2 (0.058 nM). Novel triple mutants were obtained that exhibited 104-, 46- and 32-fold improvements in binding affinity for each system, respectively. Moreover, for all three antibody-antigen systems over 90% of all the intermediate single and double mutants that were designed and tested showed higher affinities than the parent sequence. The contributions of the individual mutants to the change in binding affinity appear to be roughly additive when combined to form double and triple mutants. The new interactions introduced by the affinity-enhancing mutants included long-range electrostatics as well as short-range nonpolar interactions. This diversity in the types of new interactions formed by the mutants was reflected in SPR kinetics that showed that the enhancements in affinities arose from increasing on-rates, decreasing off-rates or a combination of the two effects, depending on the mutation. ADAPT is a very focused search of sequence space and required only 20-30 mutants for each system to be made and tested to achieve the affinity enhancements mentioned above.

The Farnesoid X receptor (FXR) exhibits significant backbone movement in response to the binding of various ligands and can be a challenge for pose prediction algorithms. As part of the D3R Grand Challenge 2, we tested Wilma-SIE, a rigid-protein docking method, on a set of 36 FXR ligands for which the crystal structures had originally been blinded. These ligands covered several classes of compounds. To overcome the rigid protein limitations of the method, we used an ensemble of publicly available structures for FXR from the PDB. The use of the ensemble allowed Wilma-SIE to predict poses with average and median RMSDs of 2.3 and 1.4 Å, respectively. It was quite clear, however, that had we used a single structure for the receptor the success rate would have been much lower. The most successful predictions were obtained on chemical classes for which one or more crystal structures of the receptor bound to a molecule of the same class was available. In the absence of a crystal structure for the class, observing a consensus binding mode for the ligands of the class using one or more receptor structures of other classes seemed to be indicative of a reasonable pose prediction. Affinity prediction proved to be more challenging with generally poor correlation with experimental IC50s (Kendall tau ~ 0.3). Even when the 36 crystal structures were used the accuracy of the predicted affinities was not appreciably improved. A possible cause of difficulty is the internal energy strain arising from conformational differences in the receptor across complexes, which may need to be properly estimated and incorporated into the SIE scoring function.

Despite decades of development, protein-protein docking remains a largely unsolved problem. The main difficulties are the immense space spanned by the translational and rotational degrees of freedom and the prediction of the conformational changes of proteins upon binding. FFT is generally the preferred method to exhaustively explore the translation-rotation space at a fine grid resolution, albeit with the trade-off of approximating force fields with correlation functions. This work presents a direct search alternative that samples the states in Cartesian space at the same resolution and computational cost as standard FFT methods. Operating in real space allows the use of standard force field functional forms used in typical non-FFT methods as well as the implementation of strategies for focused exploration of conformational flexibility. Currently, a few misplaced side chains can cause docking programs to fail. This work specifically addresses the problem of side chain rearrangements upon complex formation. Based on the observation that most side chains retain their unbound conformation upon binding, each rigidly docked pose is initially scored ignoring up to a limited number of side chain overlaps which are resolved in subsequent repacking and minimization steps. On test systems where side chains are altered and backbones held in their bound state, this implementation provides significantly better native pose recovery and higher quality (lower RMSD) predictions when compared with five of the most popular docking programs. The method is implemented in the software program ProPOSE (Protein Pose Optimization by Systematic Enumeration).

Conjugation of small molecules to proteins through N-hydroxysuccinimide (NHS) esters results in a random distribution of small molecules on lysine residues and the protein N-terminus. While mass spectrometry methods have improved characterization of these protein conjugates, it remains a challenge to quantify the occupancy at individual sites of conjugation. Here, we present a method using Tandem Mass Tags (TMT) that enabled the accurate and sensitive quantification of occupancy at individual conjugation sites in the NIST monoclonal antibody. At conjugation levels relevant to antibody drug conjugates in the clinic, site occupancy data was obtained for 37 individual sites, with average site occupancy data across 2 adjacent lysines obtained for an additional 12 sites. Thus, altogether, a measure of site occupancy was obtained for 98% of the available primary amines. We further showed that removal of the Fc-glycan on the NIST mAb increased conjugation at two specific sites in the heavy chain, demonstrating the utility of this method to identify changes in the susceptibility of individual sites to conjugation. This improved site occupancy data allowed calibration of a bi-parametric linear model for predicting the susceptibility of individual lysines to conjugation from 3D-structure based on their solvent exposures and ionization constants. Trained against the experimental data for lysines from the Fab fragment, the model provided accurate predictions of occupancies at lysine sites from the Fc region and the protein N-terminus (R-2=0.76). This predictive model will enable improved engineering of antibodies for optimal labeling with fluorophores, toxins, or crosslinkers.

Affinity modulation of antibodies and antibody fragments of therapeutic value is often required in order to improve their clinical efficacies. Virtual affinity maturation has the potential to quickly focus on the critical hotspot residues without the combinatorial explosion problem of conventional display and library approaches. However, this requires a binding affinity scoring function that is capable of ranking single-point mutations of a starting antibody. We focus here on assessing the solvated interaction energy (SIE) function that was originally developed for and is widely applied to scoring of protein-ligand binding affinities. To this end, we assembled a structure-function data set called Single-Point Mutant Antibody Binding (SiPMAB) comprising several antibody-antigen systems suitable for this assessment, i.e., based on high-resolution crystal structures for the parent antibodies and coupled with high-quality binding affinity measurements for sets of single-point antibody mutants in each system. Using this data set, we tested the SIE function with several mutation protocols based on the popular methods SCWRL, Rosetta, and FoldX. We found that the SIE function coupled with a protocol limited to sampling only the mutated side chain can reasonably predict relative binding affinities with a Spearman rank-order correlation coefficient of about 0.6, outperforming more aggressive sampling protocols. Importantly, this performance is maintained for each of the seven system-specific component subsets as well as for other relevant subsets including non-alanine and charge-altering mutations. The transferability and enrichment in affinity-improving mutants can be further enhanced using consensus ranking over multiple methods, including the SIE, Talaris, and FOLDEF energy functions. The knowledge gained from this study can lead to successful prospective applications of virtual affinity maturation.

Computational methods for docking small molecules to proteins are prominent in drug discovery. There are hundreds, if not thousands, of documented examples-and several pertinent cases within our research program. Fifteen years ago, our first docking-guided drug design project yielded nanomolar metalloproteinase inhibitors and illustrated the potential of structure-based drug design. Subsequent applications of docking programs to the design of integrin antagonists, BACE-1 inhibitors, and aminoglycosides binding to bacterial RNA demonstrated that available docking programs needed significant improvement. At that time, docking programs primarily considered flexible ligands and rigid proteins. We demonstrated that accounting for protein flexibility, employing displaceable water molecules, and using ligand-based pharmacophores improved the docking accuracy of existing methods-enabling the design of bioactive molecules. The success prompted the development of our own program, Fitted, implementing all of these aspects. The primary motivation has always been to respond to the needs of drug design studies; the majority of the concepts behind the evolution of Fitted are rooted in medicinal chemistry projects and collaborations. Several examples follow: (1) Searching for HDAC inhibitors led us to develop methods considering drug-zinc coordination and its effect on the pKa of surrounding residues. (2) Targeting covalent prolyl oligopeptidase (POP) inhibitors prompted an update to Fitted to identify reactive groups and form bonds with a given residue (e.g., a catalytic residue) when the geometry allows it. Fitted-the first fully automated covalent docking program-was successfully applied to the discovery of four new classes of covalent POP inhibitors. As a result, efficient stereoselective syntheses of a few screening hits were prioritized rather than synthesizing large chemical libraries-yielding nanomolar inhibitors. (3) In order to study the metabolism of POP inhibitors by cytochrome P450 enzymes (CYPs)-for toxicology studies-the program Impacts was derived from Fitted and helped us to reveal a complex metabolism with unforeseen stereocenter isomerizations. These efforts, combined with those of other docking software developers, have strengthened our understanding of the complex drug-protein binding process while providing the medicinal chemistry community with useful tools that have led to drug discoveries. In this Account, we describe our contributions over the past 15 years-within their historical context-to the design of drug candidates, including BACE-1 inhibitors, POP covalent inhibitors, G-quadruplex binders, and aminoglycosides binding to nucleic acids. We also remark the necessary developments of docking programs, specifically Fitted, that enabled structure-based design to flourish and yielded multiple fruitful, rational medicinal chemistry campaigns.

Temperature-dependent Raman studies of aqueous copper(I) chloride complexes have been carried out up to 80 °C, along with supporting ab initio calculations for the species [CuCl(n)(H2O)m](1-n), n = 0-4 and hydration numbers m = 0-6. Normalized reduced isotropic Raman spectra were obtained from perpendicular and parallel polarization measurements, with perchlorate anion, ClO4(-), as an internal standard. Although the Raman spectra were not intense, spectra could be corrected by solvent baseline subtraction, to yield quantitative reduced molar scattering coefficients for the symmetric vibrational bands at 297 ± 3 and 247 ± 3 cm(-1). The intensity variations of these bands with concentration and temperature provided strong evidence that these arise from the species [CuCl2](-) and [CuCl3](2-), respectively. The results from ab initio calculations using density functional theory predict similar relative peak positions and intensities for the totally symmetric Cu-Cl stretching bands of the species [CuCl2(H2O)6](-) and [CuCl3(H2O)6](2-), in which the water is coordinated to the chloride ions. A less intense Raman band at 350 ± 10 cm(-1) is attributed to the symmetric Cu-Cl stretching mode of hydrated species [CuCl(H2O)](0) with six waters of hydration. Temperature- and concentration-independent quantitative Raman molar scattering coefficients (S) are reported for the [CuCl2](-) and [CuCl3](2-)species.