Comparative Performance of High-Throughput Methods for Protein pKa Predictions.

The medically relevant field of protein-based therapeutics has triggered a demand for protein engineering in different pH environments of biological relevance. In silico engineering workflows typically employ high-throughput screening campaigns that require evaluating large sets of protein residues and point mutations by fast yet accurate computational algorithms. While several high-throughput pKa prediction methods exist, their accuracies are unclear due to the lack of a current comprehensive benchmarking. Here, seven fast, efficient, and accessible approaches including PROPKA3, DeepKa, PKAI, PKAI+, DelPhiPKa, MCCE2, and H++ were systematically tested on a nonredundant subset of 408 measured protein residue pKa shifts from the pKa database (PKAD). While no method outperformed the null hypotheses with confidence, as illustrated by statistical bootstrapping, DeepKa, PKAI+, PROPKA3, and H++ had utility. More specifically, DeepKa consistently performed well in tests across multiple and individual amino acid residue types, as reflected by lower errors, higher correlations, and improved classifications. Arithmetic averaging of the best empirical predictors into simple consensuses improved overall transferability and accuracy up to a root-mean-square error of 0.76 pKa units and a correlation coefficient (R2) of 0.45 to experimental pKa shifts. This analysis should provide a basis for further methodological developments and guide future applications, which require embedding of computationally inexpensive pKa prediction methods, such as the optimization of antibodies for pH-dependent antigen binding.

Antibody mutations favoring pH-dependent binding in solid tumor microenvironments: Insights from large-scale structure-based calculations.

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.

Serum albumin-binding VH Hs with variable pH sensitivities enable tailored half-life extension of biologics.

Prolonged serum half-life is required for the efficacy of most protein therapeutics. One strategy for half-life extension is to exploit the long circulating half-life of serum albumin by incorporating a binding moiety that recognizes albumin. Here, we describe camelid single-domain antibodies (VH Hs) that bind the serum albumins of multiple species with moderate to high affinity at both neutral and endosomal pH and significantly extend the serum half-lives of multiple proteins in rats from minutes to days. We serendipitously identified an additional VH H (M75) that is naturally pH-sensitive: at endosomal pH, binding affinity for human serum albumin (HSA) was dramatically weakened and binding to rat serum albumin (RSA) was undetectable. Domain mapping revealed that M75 bound to HSA domain 1 and 2. Moreover, alanine scanning of HSA His residues suggested a critical role for His247, located in HSA domain 2, in M75 binding and its pH dependence. Isothermal titration calorimetry experiments were suggestive of proton-linked binding of M75 to HSA, with differing binding enthalpies observed for full-length HSA and an HSA domain 1-domain 2 fusion protein in which surface-exposed His residues were substituted with Ala. M75 conferred moderate half-life extension in rats, from minutes to hours, likely due to rapid dissociation from RSA during FcRn-mediated endosomal recycling in tandem with albumin conformational changes induced by M75 binding that prevented interaction with FcRn. Humanized VH Hs maintained in vivo half-life extension capabilities. These VH Hs represent a new set of tools for extending protein therapeutic half-life and one (M75) demonstrates a unique pH-sensitive binding interaction that can be exploited to achieve modest in vivo half-life.

Role of the non-hypervariable FR3 D-E loop in single-domain antibody recognition of haptens and carbohydrates.

Single-domain antibodies (sdAbs), the variable domains of camelid heavy chain-only antibodies, are generally thought to poorly recognize nonproteinaceous small molecules and carbohydrates in comparison with conventional antibodies. However, the structures of anti-methotrexate, anti-triclocarban and anti-cortisol sdAbs revealed unexpected contributions of the non-hypervariable "CDR4" loop, formed between ß-strands D and E of framework region 3, in binding. Here, we investigated the potential role of CDR4 in sdAb binding to a hapten, 15-acetyl-deoxynivalenol (15-AcDON), and to carbohydrates. We constructed and panned a phage-displayed library in which CDR4 of the 15-AcDON-specific sdAb, NAT-267, was extended and randomized. From this library, we identified one sdAb, MA-232, bearing a 14-residue insertion in CDR4 and showing improved binding to 15-AcDON by ELISA and surface plasmon resonance. On the basis of these results, we constructed a second set of phage-displayed libraries in which the CDR4 and other regions of three hapten- or carbohydrate-binding sdAbs were diversified. With the goal of identifying sdAbs with novel glycan-binding specificities, we panned the library against four tumor-associated carbohydrate antigens but were unable to enrich binding phages. Thus, we conclude that while CDR4 may play a role in binding of some rare hapten-specific sdAbs, diversifying this region through molecular engineering is probably not a general solution to sdAb carbohydrate recognition in the absence of a paired VL domain.

An engineered transforming growth factor ß (TGF-ß) monomer that functions as a dominant negative to block TGF-ß signaling.

The transforming growth factor ß isoforms, TGF-ß1, -ß2, and -ß3, are small secreted homodimeric signaling proteins with essential roles in regulating the adaptive immune system and maintaining the extracellular matrix. However, dysregulation of the TGF-ß pathway is responsible for promoting the progression of several human diseases, including cancer and fibrosis. Despite the known importance of TGF-ßs in promoting disease progression, no inhibitors have been approved for use in humans. Herein, we describe an engineered TGF-ß monomer, lacking the heel helix, a structural motif essential for binding the TGF-ß type I receptor (TßRI) but dispensable for binding the other receptor required for TGF-ß signaling, the TGF-ß type II receptor (TßRII), as an alternative therapeutic modality for blocking TGF-ß signaling in humans. As shown through binding studies and crystallography, the engineered monomer retained the same overall structure of native TGF-ß monomers and bound TßRII in an identical manner. Cell-based luciferase assays showed that the engineered monomer functioned as a dominant negative to inhibit TGF-ß signaling with a Ki of 20-70 nm Investigation of the mechanism showed that the high affinity of the engineered monomer for TßRII, coupled with its reduced ability to non-covalently dimerize and its inability to bind and recruit TßRI, enabled it to bind endogenous TßRII but prevented it from binding and recruiting TßRI to form a signaling complex. Such engineered monomers provide a new avenue to probe and manipulate TGF-ß signaling and may inform similar modifications of other TGF-ß family members.

Binding pose and affinity prediction in the 2016 D3R Grand Challenge 2 using the Wilma-SIE method.

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.

Structural Studies of Trypanosoma brucei RNA Editing Ligases and Their Binding Partner Proteins.

To study the mechanism of ligating nicked RNA strands, we conducted molecular dynamics simulations of Trypanosoma brucei RNA editing ligases L1 and L2 complexed with double-stranded RNA (dsRNA) fragments. In each resulting model, a Mg(2+) ion coordinates the 5'-PO4 of the nicked nucleotide and the 3'-OH of the terminal nucleotide for a nucleophilic reaction consistent with the postulated step 3 chemistry of the ligation mechanism. Moreover, coordination of the 3'-OH to the Mg(2+) ion may lower its pKa, thereby rendering it a more effective nucleophile as an oxyanion. Thus, Mg(2+) may play a twofold role: bringing the reactants into the proximity of each other and activating the nucleophile. We also conducted solvated interaction energy calculations to explore whether ligation specificities can be correlated to ligase-dsRNA binding affinity changes. The calculated dsRNA binding affinities are stronger for both L1 and L2 when the terminal nucleotide is changed from cytosine to guanine, in line with their experimentally measured ligation specificities. Because the ligation mechanism is also influenced by interactions of the ligase with partner proteins from the editosome subcomplex, we also modeled the structure of the RNA-bound L2 in complex with the oligonucleotide binding (OB) domain of largest editosome interacting protein A1. The resulting L2-dsRNA-A1 model, which is consistent with mutagenesis and binding data recorded to date, provides the first atomic-level glimpse of plausible interactions around the RNA ligation site in the presence of an OB domain presented in-trans to a nucleic acid ligase.

Evaluation of the Wilma-SIE Virtual Screening Method in Community Structure-Activity Resource 2013 and 2014 Blind Challenges.

Prospective assessments of the Wilma-SIE (solvated interaction energy) platform for ligand docking and ranking were performed during the 2013 and 2014 editions of the Community Structure-Activity Resource (CSAR) blind challenge. Diverse targets like a steroid-binding protein, a serine protease (factor Xa), a tyrosine kinase (Syk), and a nucleotide methyltransferase (TrmD) were included. Pose selection was achieved with high precision; in all 24 tests Wilma-SIE top-ranked the native pose among carefully generated sets of decoy conformations. Good separation for the native pose was also observed indicating robustness in pose scoring. Cross-docking was also accomplished with high accuracy for the various systems, with ligand median-RMSD values around 1 Å from the crystal structures. Larger deviations were occasionally obtained due to the rigid-target approach even if multiple target structures were used. Affinity ranking of congeneric ligands after cross-docking was reasonable for three of the four systems, with Spearman ranking coefficients around 0.6. Poor affinity ranking for FXa is possibly due to missing structural domains, which are present during measurements. Assignment of protonation states is critical for affinity scoring with the SIE function, as shown here for the Syk system. Including the FiSH model improved cross-docking but worsened affinity predictions, pointing to the need for further fine-tuning of this newer solvation model. The consistently strong performance of the Wilma-SIE platform in recent CSAR and SAMPL blind challenges validates its applicability for virtual screening on a broad range of molecular targets.

Assessment of Solvated Interaction Energy Function for Ranking Antibody-Antigen Binding Affinities.

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.

Sulfonation of glycopeptide antibiotics by sulfotransferase StaL depends on conformational flexibility of aglycone scaffold.

Although glycopeptide antibiotics (GPAs), including vancomycin and teicoplanin, represent the most important class of anti-infective agents in the treatment of serious gram-positive bacterial infections, their usefulness is threatened by the emergence of resistant strains. GPAs are complex natural products consisting of a heptapeptide skeleton assembled via nonribosomal peptide synthesis and constrained through multiple crosslinks, with diversity resulting from enzymatic modifications by a variety of tailoring enzymes, which can be used to produce GPA analogues that could overcome antibiotic resistance. GPA-modifying sulfotransferases are promising tools for generating the unique derivatives. Despite significant sequence and structural similarities, these sulfotransferases modify distinct side chains on the GPA scaffold. To provide insight into the spatial diversity of modifications, we have determined the crystal structure of the ternary complex of bacterial sulfotransferase StaL with the cofactor product 3'-phosphoadenosine 5'-phosphate and desulfo-A47934 aglycone substrate. Desulfo-A47934 binds with the hydroxyl group on the 4-hydroxyphenylglycine in residue 1 directed toward the 3'-phosphoadenosine 5'-phosphate and hydrogen-bonded to the catalytic His67. Homodimeric StaL can accommodate GPA substrate in only one of the two active sites because of potential steric clashes. Importantly, the aglycone substrate demonstrates a flattened conformation, in contrast to the cup-shaped structures observed previously. Analysis of the conformations of this scaffold showed that despite the apparent rigidity due to crosslinking between the side chains, the aglycone scaffold displays substantial flexibility, important for enzymatic modifications by the GPA-tailoring enzymes. We also discuss the potential of using the current structural information in generating unique GPA derivatives.

Small-molecule inhibitors of the pseudaminic acid biosynthetic pathway: targeting motility as a key bacterial virulence factor.

Helicobacter pylori is motile by means of polar flagella, and this motility has been shown to play a critical role in pathogenicity. The major structural flagellin proteins have been shown to be glycosylated with the nonulosonate sugar, pseudaminic acid (Pse). This glycan is unique to microorganisms, and the process of flagellin glycosylation is required for H. pylori flagellar assembly and consequent motility. As such, the Pse biosynthetic pathway offers considerable potential as an antivirulence drug target, especially since motility is required for H. pylori colonization and persistence in the host. This report describes screening the five Pse biosynthetic enzymes for small-molecule inhibitors using both high-throughput screening (HTS) and in silico (virtual screening [VS]) approaches. Using a 100,000-compound library, 1,773 hits that exhibited a 40% threshold inhibition at a 10 µM concentration were identified by HTS. In addition, VS efforts using a 1.6-million compound library directed at two pathway enzymes identified 80 hits, 4 of which exhibited reasonable inhibition at a 10 µM concentration in vitro. Further secondary screening which identified 320 unique molecular structures or validated hits was performed. Following kinetic studies and structure-activity relationship (SAR) analysis of selected inhibitors from our refined list of 320 compounds, we demonstrated that three inhibitors with 50% inhibitory concentrations (IC50s) of approximately 14 µM, which belonged to a distinct chemical cluster, were able to penetrate the Gram-negative cell membrane and prevent formation of flagella.

Exhaustive docking and solvated interaction energy scoring: lessons learned from the SAMPL4 challenge.

We continued prospective assessments of the Wilma-solvated interaction energy (SIE) platform for pose prediction, binding affinity prediction, and virtual screening on the challenging SAMPL4 data sets including the HIV-integrase inhibitor and two host-guest systems. New features of the docking algorithm and scoring function are tested here prospectively for the first time. Wilma-SIE provides good correlations with actual binding affinities over a wide range of binding affinities that includes strong binders as in the case of SAMPL4 host-guest systems. Absolute binding affinities are also reproduced with appropriate training of the scoring function on available data sets or from comparative estimation of the change in target's vibrational entropy. Even when binding modes are known, SIE predictions lack correlation with experimental affinities within dynamic ranges below 2 kcal/mol as in the case of HIV-integrase ligands, but they correctly signaled the narrowness of the dynamic range. Using a common protein structure for all ligands can reduce the noise, while incorporating a more sophisticated solvation treatment improves absolute predictions. The HIV-integrase virtual screening data set consists of promiscuous weak binders with relatively high flexibility and thus it falls outside of the applicability domain of the Wilma-SIE docking platform. Despite these difficulties, unbiased docking around three known binding sites of the enzyme resulted in over a third of ligands being docked within 2 Å from their actual poses and over half of the ligands docked in the correct site, leading to better-than-random virtual screening results.

Enhanced antibody-antigen structure prediction from molecular docking using AlphaFold2.

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.

Optimizing Antibody-Antigen Binding Affinities with the ADAPT Platform.

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.

Solvated interaction energy: from small-molecule to antibody drug design.

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.

Effects of Two Soluble ACE2-Fc Variants on Blood Pressure and Albuminuria in Hypertensive Mice: Research Letter.

Background: Angiotensin-converting enzyme 2 (ACE2) hydrolyzes angiotensin (Ang) II to Ang-(1-7), promoting vasodilatation, and inhibiting oxidative stress and inflammation. Plasma membrane ACE2 is the receptor for all known SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) viral variants. In COVID-19 infection, soluble ACE2 variants may act as decoys to bind and neutralize the coronavirus, reducing its tissue infectivity. Furthermore, soluble ACE2 variants have been proposed as potential therapeutics for kidney disease and hypertensive disorders. Objective: Soluble ACE2 variants conjugated to human Fc domains and selected for high-potency viral SARS-CoV-2 neutralization were prepared and evaluated for ACE2 activity in vitro. Lead candidates were then tested for systemic ACE2 activity, stability, and effects on blood pressure and albuminuria in mice with Ang II-induced hypertension. Methods: ACE2 activity of 10 soluble ACE2 variants was first assessed in cell-free conditions using a fluorogenic substrate, or by Ang II hydrolysis to Ang-(1-7). Hypertension was induced in male or female mice by implantation of osmotic minipumps containing Ang II. Two lead ACE2 variants were injected intravenously (i.v.) into hypertensive mice, followed by measurements of blood pressure (tail-cuff plethysmography), albuminuria, and tissue ACE2 activity and protein (immunoblots). Results: Soluble ACE2-Fc variants demonstrated significant ACE2 enzymatic activity, with kinetics comparable with human recombinant ACE2. In hypertensive mice, single dose i.v. injection of ACE2-Fc variant K (10 mg/kg) significantly decreased systolic blood pressure at 24 hours, with partial lowering sustained to 48 hours, and tendency to reduce albuminuria at 72 hours. By contrast, ACE2-Fc variant I had no effect on blood pressure or albuminuria in hypertensive mice; ACE2-Fc variant K was detected by immunoblotting in plasma, kidney, heart, lung, liver, and spleen lysates 72 hours after injection, associated with significantly increased ACE2 activity in all tissues except kidney and spleen. Angiotensin-converting enzyme 2-Fc variant I had no effect on plasma ACE2 activity. Conclusions: Soluble ACE2-Fc variant K reduces blood pressure and tends to lower albuminuria in hypertensive mice. Furthermore, soluble ACE2-Fc variant K has prolonged tissue retention, associated with increased tissue ACE2 activity. The results support further studies directed at the therapeutic potential of soluble ACE2-Fc variant K for cardiovascular and kidney protection.

Contexte: L'enzyme de conversion de l'angiotensine 2 (ACE2) hydrolyse l'angiotensine (Ang) II en angiotensine (Ang)-(1-7), ce qui favorise la vasodilatation et inhibe le stress oxydatif et l'inflammation. L'ACE2 de la membrane plasmique est le récepteur de tous les variants connus du SARS-COV-2. Dans les cas d'infection à la COVID-19, les variants solubles de l'ACE2 peuvent agir comme leurres pour lier et neutraliser le coronavirus, et réduire ainsi son infectiosité dans les tissus. Des variants solubles de l'ACE2 ont également été proposés comme agents thérapeutiques potentiels pour l'insuffisance rénale et les troubles liés à l'hypertension. Objectif: Des variants solubles de l'ACE2 conjugués au domaine Fc humain ont été sélectionnés pour leur fort potentiel neutralisant du virus SARS-COV-2, puis préparés et évalués pour la mesure de l'activité de l'ACE2 in vitro. Les meilleurs candidats ont ensuite été testés chez des souris souffrant d'hypertension induite par l'Ang II afin de mesurer l'activité d'ACE2, ainsi que leur stabilité et leurs effets sur la pression artérielle et l'albuminurie. Méthodologie: L'activité de 10 variants solubles de l'ACE2 a d'abord été évaluée en conditions acellulaires à l'aide d'un substrat fluorogène, ou par hydrolyse de l'Ang II en Ang-(1-7). L'hypertension a été induite chez des souris mâles ou femelles par l'implantation de minipompes osmotiques contenant de l'Ang II. Deux des meilleurs variants de l'ACE2 ont été injectés par voie intraveineuse (i.v.) à des souris hypertendues, puis des mesures de la pression artérielle (pléthysmographie par manchon caudal), de l'albuminurie, de l'activité de l'ACE2 dans les tissus et des protéines (immunobuvardage) ont été effectuées. Résultats: Les variants solubles ACE2-Fc ont montré une activité enzymatique significative, avec une cinétique comparable à celle de l'ACE2 recombinante humaine. Chez les souris hypertendues, l'injection i.v. d'une dose unique (10 mg/kg) du variant K ACE2-Fc a abaissé significativement la pression artérielle systolique après 24 heures­une réduction partielle s'étant prolongée jusqu'à 48 heures­et a montré une tendance à réduire l'albuminurie après 72 heures. En revanche, le variant I ACE2-Fc n'a eu aucun effet sur la pression artérielle ou l'albuminurie des souris hypertendues. Après 72 heures, le variant K ACE2-Fc a été détecté par immunobuvardage dans le plasma, ainsi que dans des lysats de reins, de cœur, de poumon, de foie et de rate, ce qui a été associé à une augmentation significative de l'activité de l'ACE2 dans tous les tissus sauf dans les reins et la rate. Le variant I ACE2-Fc n'a montré aucun effet sur l'activité de l'ACE2 dans le plasma. Conclusion: Le variant soluble K ACE2-Fc abaisse la pression artérielle et tend à diminuer l'albuminurie chez les souris hypertendues. Il présente en outre une rétention tissulaire prolongée, laquelle est associée à une plus grande activité de l'ACE2 dans les tissus. Ces résultats appuient d'autres études portant sur le potentiel thérapeutique du variant soluble K ACE2-Fc dans la protection cardiovasculaire et rénale.

Exploring rigid-backbone protein docking in biologics discovery: a test using the DARPin scaffold.

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.

A glyco-engineering approach for site-specific conjugation to Fab glycans.

Effective processes for synthesizing antibody-drug conjugates (ADCs) require: 1) site-specific incorporation of the payload to avoid interference with binding to the target epitope, 2) optimal drug/antibody ratio to achieve sufficient potency while avoiding aggregation or solubility problems, and 3) a homogeneous product to facilitate approval by regulatory agencies. In conventional ADCs, the drug molecules are chemically attached randomly to antibody surface residues (typically Lys or Cys), which can interfere with epitope binding and targeting, and lead to overall product heterogeneity, long-term colloidal instability and unfavorable pharmacokinetics. Here, we present a more controlled process for generating ADCs where drug is specifically conjugated to only Fab N-linked glycans in a narrow ratio range through functionalized sialic acids. Using a bacterial sialytransferase, we incorporated N-azidoacetylneuraminic acid (Neu5NAz) into the Fab glycan of cetuximab. Since only about 20% of human IgG1 have a Fab glycan, we extended the application of this approach by using molecular modeling to introduce N-glycosylation sites in the Fab constant region of other therapeutic monoclonal antibodies. We used trastuzumab as a model for the incorporation of Neu5NAz in the novel Fab glycans that we designed. ADCs were generated by clicking the incorporated Neu5NAz with monomethyl auristatin E (MMAE) attached to a self-immolative linker terminated with dibenzocyclooctyne (DBCO). Through this process, we obtained cetuximab-MMAE and trastuzumab-MMAE with drug/antibody ratios in the range of 1.3 to 2.5. We confirmed that these ADCs still bind their targets efficiently and are as potent in cytotoxicity assays as control ADCs obtained by standard conjugation protocols. The site-directed conjugation to Fab glycans has the additional benefit of avoiding potential interference with effector functions that depend on Fc glycan structure.

Predicting hydration free energies of polychlorinated aromatic compounds from the SAMPL-3 data set with FiSH and LIE models.

Next-generation solvation models are devised to mimic the accuracy and generality of explicit solvation models at the speed of current popular implicit solvation models. One such method is the first-shell of hydration (FiSH) continuum model that was trained on hydration energetics from LIE calculations and molecular dynamics simulations in explicit solvent. Here we tested prospectively the FiSH model on the SAMPL-3 hydration data set that zooms in the effect of chlorination on solvation. We compare these FiSH predictions with those from retrospective LIE calculations. We find that neither FiSH nor LIE can reproduce well the absolute values and the trend of hydration free energies in the biphenyl and dioxin aromatic chlorination series. Some of the hypotheses behind this performance are discussed and tested. The LIE explicit-solvent model shows some improvement relative to the FiSH continuum model, and we correct a systematic deviation in the continuum van der Waals term of FiSH associated with aromatic Cl atom type.

Exhaustive search and solvated interaction energy (SIE) for virtual screening and affinity prediction.

We carried out a prospective evaluation of the utility of the SIE (solvation interaction energy) scoring function for virtual screening and binding affinity prediction. Since experimental structures of the complexes were not provided, this was an exercise in virtual docking as well. We used our exhaustive docking program, Wilma, to provide high-quality poses that were rescored using SIE to provide binding affinity predictions. We also tested the combination of SIE with our latest solvation model, first shell of hydration (FiSH), which captures some of the discrete properties of water within a continuum model. We achieved good enrichment in virtual screening of fragments against trypsin, with an area under the curve of about 0.7 for the receiver operating characteristic curve. Moreover, the early enrichment performance was quite good with 50% of true actives recovered with a 15% false positive rate in a prospective calculation and with a 3% false positive rate in a retrospective application of SIE with FiSH. Binding affinity predictions for both trypsin and host-guest complexes were generally within 2 kcal/mol of the experimental values. However, the rank ordering of affinities differing by 2 kcal/mol or less was not well predicted. On the other hand, it was encouraging that the incorporation of a more sophisticated solvation model into SIE resulted in better discrimination of true binders from binders. This suggests that the inclusion of proper Physics in our models is a fruitful strategy for improving the reliability of our binding affinity predictions.