Blinded evaluation of cathepsin S inhibitors from the D3RGC3 dataset using molecular docking and free energy calculations

Reliability of AlphaFold2 Models in Virtual Drug Screening: A Focus on Selected Class A GPCRs

Protein three-dimensional (3D) structure prediction is one of the most challenging issues in the field of computational biochemistry, which has overwhelmed scientists for almost half a century. A significant breakthrough in structural biology has been established by developing the artificial intelligence (AI) system AlphaFold2 (AF2). The AF2 system provides a state-of-the-art prediction of protein structures from nearly all known protein sequences with high accuracy. This study examined the reliability of AF2 models compared to the experimental structures in drug discovery, focusing on one of the most common protein drug-targeted classes known as G protein-coupled receptors (GPCRs) class A. A total of 32 representative protein targets were selected, including experimental structures of X-ray crystallographic and Cryo-EM structures and their corresponding AF2 models. The quality of AF2 models was assessed using different structure validation tools, including the pLDDT score, RMSD value, MolProbity score, percentage of Ramachandran favored, QMEAN Z-score, and QMEANDisCo Global. The molecular docking was performed using the Genetic Optimization for Ligand Docking (GOLD) software. The AF2 models' reliability in virtual drug screening was determined by their ability to predict the ligand binding poses closest to the native binding pose by assessing the Root Mean Square Deviation (RMSD) metric and docking scoring function. The quality of the docking and scoring function was evaluated using the enrichment factor (EF). Furthermore, the capability of using AF2 models in molecular docking to identify hits with key protein-ligand interactions was analyzed. The posing power results showed that the AF2 models successfully predicted ligand binding poses (RMSD < 2 Å). However, they exhibited lower screening power, with average EF values of 2.24, 2.42, and 1.82 for X-ray, Cryo-EM, and AF2 structures, respectively. Moreover, our study revealed that molecular docking using AF2 models can identify competitive inhibitors. In conclusion, this study found that AF2 models provided docking results comparable to experimental structures, particularly for certain GPCR targets, and could potentially significantly impact drug discovery.

Azetidinimines as a novel series of non-covalent broad-spectrum inhibitors of β-lactamases with submicromolar activities against carbapenemases KPC-2 (class A), NDM-1 (class B) and OXA-48 (class D)

The occurrence of resistances in Gram negative bacteria is steadily increasing to reach extremely worrying levels and one of the main causes of resistance is the massive spread of very efficient β-lactamases which render most β-lactam antibiotics useless. Herein, we report the development of a series of imino-analogues of β-lactams (namely azetidinimines) as efficient non-covalent inhibitors of β-lactamases. Despite the structural and mechanistic differences between serine-β-lactamases KPC-2 and OXA-48 and metallo-β-lactamase NDM-1, all three enzymes can be inhibited at a submicromolar level by compound 7dfm, which can also repotentiate imipenem against a resistant strain of Escherichia coli expressing NDM-1. We show that 7dfm can efficiently inhibit not only the three main clinically-relevant carbapenemases of Ambler classes A (KPC-2), B (NDM-1) and D (OXA-48) with Ki's below 0.3 μM, but also the cephalosporinase CMY-2 (class C, 86% inhibition at 10 μM). Our results pave the way for the development of a new structurally original family of non-covalent broad-spectrum inhibitors of β-lactamases.

D3R grand challenge 4: blind prediction of protein-ligand poses, affinity rankings, and relative binding free energies

The Drug Design Data Resource (D3R) aims to identify best practice methods for computer aided drug design through blinded ligand pose prediction and affinity challenges. Herein, we report on the results of Grand Challenge 4 (GC4). GC4 focused on proteins beta secretase 1 and Cathepsin S, and was run in an analogous manner to prior challenges. In Stage 1, participant ability to predict the pose and affinity of BACE1 ligands were assessed. Following the completion of Stage 1, all BACE1 co-crystal structures were released, and Stage 2 tested affinity rankings with co-crystal structures. We provide an analysis of the results and discuss insights into determined best practice methods.

Docking rigid macrocycles using Convex-PL, AutoDock Vina, and RDKit in the D3R Grand Challenge 4

The D3R Grand Challenge 4 provided a brilliant opportunity to test macrocyclic docking protocols on a diverse high-quality experimental data. We participated in both pose and affinity prediction exercises. Overall, we aimed to use an automated structure-based docking pipeline built around a set of tools developed in our team. This exercise again demonstrated a crucial importance of the correct local ligand geometry for the overall success of docking. Starting from the second part of the pose prediction stage, we developed a stable pipeline for sampling macrocycle conformers. This resulted in the subangstrom average precision of our pose predictions. In the affinity prediction exercise we obtained average results. However, we could improve these when using docking poses submitted by the best predictors. Our docking tools including the Convex-PL scoring function are available at https://team.inria.fr/nano-d/software/.

Correlation of cathepsin S with coronary stenosis degree, carotid thickness, blood pressure, glucose and lipid metabolism and vascular endothelial function in atherosclerosis

Correlation of cathepsin S with coronary stenosis degree, carotid thickness, blood pressure, glucose and lipid metabolism, and vascular endothelial function in patients with atherosclerosis was investigated. Data from 120 patients with increased cathepsin S levels (increased group) and 120 subjects with normal cathepsin S levels (normal group) were retrospectively analyzed. The serum cathepsin S level and Gensini score were compared between the healthy subjects and patients with coronary atherosclerotic heart disease, and the correlation between serum cathepsin S level and Gensini score was analyzed. The carotid thickness, mean arterial pressure, and indexes related to glucose and lipid metabolism, as well as vascular endothelial function were compared between the two groups. The correlation of the serum cathepsin S level with carotid intima-media thickness (IMT), mean arterial pressure, fasting blood glucose, total cholesterol (TC) and nitric oxide (NO) was also investigated. Patients with multi-vessel coronary artery disease (CAD) had higher serum cathepsin S level than those with double-vessel and single-vessel disease, and higher level than that of healthy subjects. The Gensini score of patients with multi-vessel CAD was higher than that of patients with double-vessel and single-vessel disease, as well as that of healthy subjects. The serum cathepsin S level was positively correlated with the Gensini score. Patients with increased cathepsin S level had greater IMT, higher mean arterial pressure, fasting blood glucose, fasting insulin (FINS), triglyceride (TG), TC, and endothelin-1 (ET-1), however, lower NO level than those of healthy subjects. The serum cathepsin S level was positively correlated with IMT, mean arterial pressure, fasting blood glucose, and TC, however, it was negatively correlated with the NO level. In conclusion, as the serum cathepsin S level is elevated, the coronary stenosis is aggravated, the carotid thickness and blood pressure are increased, and the glucose and lipid metabolism, as well as vascular endothelial function are significantly abnormal.

Performance evaluation of molecular docking and free energy calculations protocols using the D3R Grand Challenge 4 dataset

Using the D3R Grand Challenge 4 dataset containing Beta-secretase 1 (BACE) and Cathepsin S (CatS) inhibitors, we have evaluated the performance of our in-house docking workflow that involves in the first step the selection of the most suitable docking software for the system of interest based on structural and functional information available in public databases, followed by the docking of the dataset to predict the binding modes and ranking of ligands. The macrocyclic nature of the BACE ligands brought additional challenges, which were dealt with by a careful preparation of the three-dimensional input structures for ligands. This provided top-performing predictions for BACE, in contrast with CatS, where the predictions in the absence of guiding constraints provided poor results. These results highlight the importance of previous structural knowledge that is needed for correct predictions on some challenging targets. After the end of the challenge, we also carried out free energy calculations (i.e. in a non-blinded manner) for CatS using the pmx software and several force fields (AMBER, Charmm). Using knowledge-based starting pose construction allowed reaching remarkable accuracy for the CatS free energy estimates. Interestingly, we show that the use of a consensus result, by averaging the results from different force fields, increases the prediction accuracy.

Modelling the binding mode of macrocycles: Docking and conformational sampling

Drug discovery is increasingly tackling challenging protein binding sites regarding molecular recognition and druggability, including shallow and solvent-exposed protein-protein interaction interfaces. Macrocycles are emerging as promising chemotypes to modulate such sites. Despite their chemical complexity, macrocycles comprise important drugs and offer advantages compared to non-cyclic analogs, hence the recent impetus in the medicinal chemistry of macrocycles. Elaboration of macrocycles, or constituent fragments, can strongly benefit from knowledge of their binding mode to a target. When such information from X-ray crystallography is elusive, computational docking can provide working models. However, few studies have explored docking protocols for macrocycles, since conventional docking methods struggle with the conformational complexity of macrocycles, and also potentially with the shallower topology of their binding sites. Indeed, macrocycle binding mode prediction with the mainstream docking software GOLD has hardly been explored. Here, we present an in-depth study of macrocycle docking with GOLD and the ChemPLP scores. First, we summarize the thorough curation of a test set of 41 protein-macrocycle X-ray structures, raising the issue of lattice contacts with such systems. Rigid docking of the known bioactive conformers was successful (three top ranked poses) for 92.7% of the systems, in absence of crystallographic waters. Thus, without conformational search issues, scoring performed well. However, docking success dropped to 29.3% with the GOLD built-in conformational search. Yet, the success rate doubled to 58.5% when GOLD was supplied with extensive conformer ensembles docked rigidly. The reasons for failure, sampling or scoring, were analyzed, exemplified with particular cases. Overall, binding mode prediction of macrocycles remains challenging, but can be much improved with tailored protocols. The analysis of the interplay between conformational sampling and docking will be relevant to the prospective modelling of macrocycles in general.