Discovery and optimization of a new class of potent and non-chiral indole-3-carboxamide-based renin inhibitors

Palladium-Catalyzed C3-Carbaldehyde Directed Regioselective C2-Thioarylation of Indoles

Palladium-catalyzed thioarylation of indoles by diaryl disulfides in the presence of phenyliododiacetate is reported. The directing potential of weakly coordinating aldehyde group present at 3-position of indole was exploited for regioselective C2-H thioarylation over the possible C4-H functionalization. Mechanistic studies reveal that the process involves initial generation of thioaryl radical followed by sequential C-H activation, thiolate transfer, and reductive elimination.

Impact of Applicability Domains to Generative Artificial Intelligence

Molecular generative artificial intelligence is drawing significant attention in the drug design community, with several experimentally validated proof of concepts already published. Nevertheless, generative models are known for sometimes generating unrealistic, unstable, unsynthesizable, or uninteresting structures. This calls for methods to constrain those algorithms to generate structures in drug-like portions of the chemical space. While the concept of applicability domains for predictive models is well studied, its counterpart for generative models is not yet well-defined. In this work, we empirically examine various possibilities and propose applicability domains suited for generative models. Using both public and internal data sets, we use generative methods to generate novel structures that are predicted to be actives by a corresponding quantitative structure-activity relationships model while constraining the generative model to stay within a given applicability domain. Our work looks at several applicability domain definitions, combining various criteria, such as structural similarity to the training set, similarity of physicochemical properties, unwanted substructures, and quantitative estimate of drug-likeness. We assess the structures generated from both qualitative and quantitative points of view and find that the applicability domain definitions have a strong influence on the drug-likeness of generated molecules. An extensive analysis of our results allows us to identify applicability domain definitions that are best suited for generating drug-like molecules with generative models. We anticipate that this work will help foster the adoption of generative models in an industrial context.

Optimizing interactions to protein binding sites by integrating docking-scoring strategies into generative AI methods

The identification and optimization of promising lead molecules is essential for drug discovery. Recently, artificial intelligence (AI) based generative methods provided complementary approaches for generating molecules under specific design constraints of relevance in drug design. The goal of our study is to incorporate protein 3D information directly into generative design by flexible docking plus an adapted protein-ligand scoring function, thereby moving towards automated structure-based design. First, the protein-ligand scoring function RFXscore integrating individual scoring terms, ligand descriptors, and combined terms was derived using the PDBbind database and internal data. Next, design results for different workflows are compared to solely ligand-based reward schemes. Our newly proposed, optimal workflow for structure-based generative design is shown to produce promising results, especially for those exploration scenarios, where diverse structures fitting to a protein binding site are requested. Best results are obtained using docking followed by RFXscore, while, depending on the exact application scenario, it was also found useful to combine this approach with other metrics that bias structure generation into "drug-like" chemical space, such as target-activity machine learning models, respectively.

Interpretation of Structure-Activity Relationships in Real-World Drug Design Data Sets Using Explainable Artificial Intelligence

In silico models based on Deep Neural Networks (DNNs) are promising for predicting activities and properties of new molecules. Unfortunately, their inherent black-box character hinders our understanding, as to which structural features are important for activity. However, this information is crucial for capturing the underlying structure-activity relationships (SARs) to guide further optimization. To address this interpretation gap, "Explainable Artificial Intelligence" (XAI) methods recently became popular. Herein, we apply and compare multiple XAI methods to projects of lead optimization data sets with well-established SARs and available X-ray crystal structures. As we can show, easily understandable and comprehensive interpretations are obtained by combining DNN models with some powerful interpretation methods. In particular, SHAP-based methods are promising for this task. A novel visualization scheme using atom-based heatmaps provides useful insights into the underlying SAR. It is important to note that all interpretations are only meaningful in the context of the underlying models and associated data.

Palladium-catalyzed carbonylative synthesis of indole-3-carboxamides from 2-ethynylanilines and nitroarenes

A facile and straightforward approach for the expedite construction of indole-3-carboxamide skeletons via a palladium-catalyzed carbonylative cyclization of 2-ethynylanilines with nitroarenes has been developed.

Cp*Co(III)-catalyzed C2-thiolation and C2,C3-dithiolation of substituted indoles with N-(arylthio)succinimide

A general and efficient Cp*Co^III-catalyzed C2-thiolation and C2,C3-dithiolation of indole derivatives has been achieved employing N-(aryl/alkylthio)succinimide as a thiolating reagent. This external oxidant-free method utilizes only catalytic amounts of additive and tolerates various functional groups to afford various thiolated products in good yields. Control experiments revealed the importance of the Cp*Co^III-catalyst for both C2- and C3-thiolation.

Application of Deep Neural Network Models in Drug Discovery Programs

In silico driven optimization of compound properties related to pharmacokinetics, pharmacodynamics, and safety is a key requirement in modern drug discovery. Nowadays, large and harmonized datasets allow to implement deep neural networks (DNNs) as a framework for leveraging predictive models. Nevertheless, various available model architectures differ in their global applicability and performance in lead optimization projects, such as stability over time and interpretability of the results. Here, we describe and compare the value of established DNN-based methods for the prediction of key ADME property trends and biological activity in an industrial drug discovery environment, represented by microsomal lability, CYP3A4 inhibition and factor Xa inhibition. Three architectures are exemplified, our earlier described multilayer perceptron approach (MLP), graph convolutional network-based models (GCN) and a vector representation approach, Mol2Vec. From a statistical perspective, MLP and GCN were found to perform superior over Mol2Vec, when applied to external validation sets. Interestingly, GCN-based predictions are most stable over a longer period in a time series validation study. Apart from those statistical observations, DNN prove of value to guide local SAR. To illustrate this important aspect in pharmaceutical research projects, we discuss challenging applications in medicinal chemistry towards a more realistic picture of artificial intelligence in drug discovery.

Artificial intelligence in the early stages of drug discovery

Although the use of computational methods within the pharmaceutical industry is well established, there is an urgent need for new approaches that can improve and optimize the pipeline of drug discovery and development. In spite of the fact that there is no unique solution for this need for innovation, there has recently been a strong interest in the use of Artificial Intelligence for this purpose. As a matter of fact, not only there have been major contributions from the scientific community in this respect, but there has also been a growing partnership between the pharmaceutical industry and Artificial Intelligence companies. Beyond these contributions and efforts there is an underlying question, which we intend to discuss in this review: can the intrinsic difficulties within the drug discovery process be overcome with the implementation of Artificial Intelligence? While this is an open question, in this work we will focus on the advantages that these algorithms provide over the traditional methods in the context of early drug discovery.

Decades-old renin inhibitors are still struggling to find a niche in antihypertensive therapy. A fleeting look at the old and the promising new molecules

Hypertension is a diverse illness interlinked with cerebral, cardiovascular (CVS) and renal abnormalities. Presently, the malady is being treated by focusing on Renin- angiotensin system (RAS), voltage-gated calcium channels, peripheral vasodilators, renal and sympathetic nervous systems. Cardiovascular and renal abnormalities are associated with the overactivation of RAS, which can be constrained by angiotensin- converting enzyme inhibitors (ACEIs), angiotensin II (Ang-II) -AT1 receptor blockers (ARBs) and renin inhibitors. The latter is a new player in the old system. The renin catalyzes the conversion of angiotensinogen to Angiotensin I (Ang-I). This can be overcome by inhibiting renin, a preliminary step, eventually hinders the occurrence of the cascade of events in the RAS. Various peptidomimetics, the first-generation renin inhibitors developed six decades ago have limited drug-like properties as they suffered from poor intestinal absorption, high liver first-pass metabolism and low oral bioavailability. The development of chemically diverse molecules from peptides to nonpeptides expanded the horizon to achieving direct renin inhibition. Aliskiren, a blockbuster drug that emerged as a clinical candidate and got approved by the US FDA in 2007 was developed by molecular modeling studies. Aliskiren indicated superior to average efficacy and with minor adverse effects relative to other RAS inhibitors. However, its therapeutic use is limited by poor oral bioavailability of less than 2% that is similar to first-generation peptidic compounds. In this review, we present the development of direct renin inhibitors (DRIs) from peptidic to nonpeptidics that lead to the birth of aliskiren, its place in the treatment of cardiovascular diseases and its limitations.

Synthetic strategies, SAR studies, and computer modeling of indole 2 and 3-carboxamides as the strong enzyme inhibitors: a review

Abstract

Indole derivatives have been the focus of many researchers in the study of pharmaceutical compounds for many years. Researchers have investigated the effect of carboxamide moiety at positions 2 and 3, giving unique inhibitory properties to these compounds. The presence of carboxamide moiety in indole derivatives causes hydrogen bonds with a variety of enzymes and proteins, which in many cases, inhibits their activity. In this review, synthetic strategies of indole 2 and 3-carboxamide derivatives, the type, and mode of interaction of these derivatives against HLGP, HIV-1, renin enzyme, and structure–activity studies of these compounds were investigated. It is hoped that indole scaffolds will be tested in the future for maximum activity in pharmacological compounds.

Graphic abstract

Sequence, Structural Analysis and Metrics to Define the Unique Dynamic Features of the Flap Regions Among Aspartic Proteases

Aspartic proteases are a class of hydrolytic enzymes that have been implicated in a number of diseases such as HIV, malaria, cancer and Alzheimer's. The flap region of aspartic proteases is a characteristic unique structural feature of these enzymes; and found to have a profound impact on protein overall structure, function and dynamics. Flap dynamics also plays a crucial role in drug binding and drug resistance. Therefore, understanding the structure and dynamic behavior of this flap regions is crucial in the design of potent and selective inhibitors against aspartic proteases. Defining metrics that can describe the flap motion/dynamics has been a challenging topic in literature. This review is the first attempt to compile comprehensive information on sequence, structure, motion and metrics used to assess the dynamics of the flap region of different aspartic proteases in "one pot". We believe that this review would be of critical importance to the researchers from different scientific domains.

Discovery of highly potent renin inhibitors potentially interacting with the S3' subsite of renin

To exploit the S3' subsite of renin active site for renin inhibitor design, 42 aliskiren derivatives with modified P2' portion were designed, synthesized and biologically evaluated. Some highly potent renin inhibitors (IC₅₀ < 3 nM) were identified, among which compounds 38 (IC₅₀ = 0.9 nM) and 39 (IC₅₀ = 0.7 nM) were over 2.5-fold more potent than aliskiren (IC₅₀ = 2.3 nM). SAR analysis indicated that incorporation of polar hydrophilic moieties into the P2' portion of renin inhibitors generally enhanced the potency. Consistently with this, molecular modeling study revealed that the triazole part of 39 could provide additional interactions to the S3' subsite of renin active site. Moreover, in vivo evaluation in the double transgenic mouse hypertension model demonstrated that 39 produced greater reduction of the mean arterial blood pressure than ariskiren at the doses of 17.0 and 34.0 μmol/kg, respectively. Taken together, the S3' subsite of renin active site merits further consideration for renin inhibitor design.

Synthesis of 2-arylindole derivatives and evaluation as nitric oxide synthase and NFκB inhibitors

Development of small molecule drug-like inhibitors blocking both nitric oxide synthase and NFκB could offer a synergistic therapeutic approach in the prevention and treatment of inflammation and cancer. During the course of evaluating the biological potential of a commercial compound library, 2-phenylindole (1) displayed inhibitory activity against nitrite production and NFκB with IC(50) values of 38.1 ± 1.8 and 25.4 ± 2.1 μM, respectively. Based on this lead, synthesis and systematic optimization have been undertaken in an effort to find novel and more potent nitric oxide synthase and NFκB inhibitors with antiinflammatory and cancer preventive potential. First, chemical derivatizations of 1 and 2-phenylindole-3-carboxaldehyde (4) were performed to generate a panel of N-alkylated indoles and 3-oxime derivatives 2–7. Second, a series of diversified 2-arylindole derivatives (10) were synthesized from an array of substituted 2-iodoanilines (8) and terminal alkynes (9) by applying a one-pot palladium catalyzed Sonogashira-type alkynylation and base-assisted cycloaddition. Subsequent biological evaluations revealed 3-carboxaldehyde oxime and cyano substituted 2-phenylindoles 5 and 7 exhibited the strongest nitrite inhibitory activities (IC(50) = 4.4 ± 0.5 and 4.8 ± 0.4 μM, respectively); as well as NFκB inhibition (IC(50) = 6.9 ± 0.8 and 8.5 ± 2.0 μM, respectively). In addition, the 6′-MeO-naphthalen-2′-yl indole derivative 10at displayed excellent inhibitory activity against NFκB with an IC(50) value of 0.6 ± 0.2 μM.

The discovery of novel potent trans-3,4-disubstituted pyrrolidine inhibitors of the human aspartic protease renin from in silico three-dimensional (3D) pharmacophore searches

The small-molecule trans-3,4-disubstituted pyrrolidine 6 was identified from in silico three-dimensional (3D) pharmacophore searches based on known X-ray structures of renin-inhibitor complexes and demonstrated to be a weakly active inhibitor of the human enzyme. The unexpected binding mode of the more potent enantiomer (3S,4S)-6a in an extended conformation spanning the nonprime and S1' pockets of the recombinant human (rh)-renin active site was elucidated by X-ray crystallography. Initial structure-activity relationship work focused on modifications of the hydrophobic diphenylamine portion positioned in S1 and extending toward the S2 pocket. Replacement with an optimized P3-P1 pharmacophore interacting to the nonsubstrate S3(sp) cavity eventually resulted in significantly improved in vitro potency and selectivity. The prototype analogue (3S,4S)-12a of this new class of direct renin inhibitors exerted blood pressure lowering effects in a hypertensive double-transgenic rat model after oral administration.

The catalytic mechanism of mouse renin studied with QM/MM calculations

Hypertension is a chronic condition that affects nearly 25% of adults worldwide. As the Renin-Angiotensin-Aldosterone System is implicated in the control of blood pressure and body fluid homeostasis, its combined blockage is an attractive therapeutic strategy currently in use for the treatment of several cardiovascular conditions. We have performed QM/MM calculations to study the mouse renin catalytic mechanism in atomistic detail, using the N-terminal His6-Asn14 segment of angiotensinogen as substrate. The enzymatic reaction (hydrolysis of the peptidic bond between residues in the 10th and 11th positions) occurs through a general acid/base mechanism and, surprisingly, it is characterized by three mechanistic steps: it begins with the creation of a first very stable tetrahedral gem-diol intermediate, followed by protonation of the peptidic bond nitrogen, giving rise to a second intermediate. In a final step the peptidic bond is completely cleaved and both gem-diol hydroxyl protons are transferred to the catalytic dyad (Asp32 and Asp215). The final reaction products are two separate peptides with carboxylic acid and amine extremities. The activation energy for the formation of the gem-diol intermediate was calculated as 23.68 kcal mol(-1), whereas for the other steps the values were 15.51 kcal mol(-1) and 14.40 kcal mol(-1), respectively. The rate limiting states were the reactants and the first transition state. The associated barrier (23.68 kcal mol(-1)) is close to the experimental values for the angiotensinogen substrate (19.6 kcal mol(-1)). We have also tested the influence of the density functional on the activation and reaction energies. All eight density functionals tested (B3LYP, B3LYP-D3, X3LYP, M06, B1B95, BMK, mPWB1K and B2PLYP) gave very similar results.

Exploring the substructural space of indole-3-carboxamide derivatives binding to renin: a novel active-site spatial partitioning approach

Renin has recently attracted much attention in the antihypertensive community, since this enzyme starts the angiotensin-converting cascade and forms the rate-limiting step in this cascade. In the present study, we describe a new method called active-site spatial partitioning (ASSP) for quantitatively characterizing the nonbonding interaction profile between renin and the substructures of indole-3-carboxamide derivatives-a novel class of achiral renin inhibitors that exhibit both high affinity and strong specificity for renin, thus blocking its active state-on the basis of structural models of protein-ligand complexes. It is shown that the ASSP-derived potential parameters are highly correlated with the experimentally measured activities of indole-3-carboxamides; the statistical models linking the parameters and activities using a sophisticated partial least squares regression technique show much promise as an effective and powerful tool for generalizing and predicting the pharmaceutical potencies and the physicochemical properties of other modified derivatives. Furthermore, by visually examining substructure-color plots generated by the ASSP procedure, it is found that the relative importance of nonbonding contributions to the recognition and binding of a ligand by renin is as follows: steric < hydrophobic < electrostatic. The polar and charged moieties that float on the surface of the ligand molecule play a critical role in conferring electrostatic stability and specificity to renin-ligand complexes, whereas the aromatic rings embedded in the core region of the ligand are the main source of hydrophobic and steric potentials that lead to substantial stabilization of the complex architecture.

Docking alignment-3D-QSAR of a new class of potent and non-chiral indole-3-carboxamide-based renin inhibitors

Using generated conformations from docking analysis by CDOCKER algorithm, some 3D-QSAR models; CoMFA region focusing (CoMFA-RF) and CoMSIA have been created on a series of a new class of potent and non-chiral renin inhibitors. The satisfactory predictions were obtained by CoMFA-RF and CoMSIA based on docking alignment in comparison to CoMFA. Robustness and predictability of the models were further verified by using the test set, cross validation (leave one out and leave ten out), bootstrapping, and progressive scrambling. All-orientation search (AOS) strategy was used to acquire the best orientation and minimize the effect of the initial orientation of aligned compounds. The results of 3D-QSAR models are in agreement with docking results. Moreover, the resulting 3D CoMFA-RF/ CoMSIA contour maps and corresponding models were applied to design new and more active inhibitors.

Discovery of VTP-27999, an Alkyl Amine Renin Inhibitor with Potential for Clinical Utility

Structure guided optimization of a series of nonpeptidic alkyl amine renin inhibitors allowed the rational incorporation of additional polar functionality. Replacement of the cyclohexylmethyl group occupying the S1 pocket with a (R)-(tetrahydropyran-3-yl)methyl group and utilization of a different attachment point led to the identification of clinical candidate 9. This compound demonstrated excellent selectivity over related and unrelated off-targets, >15% oral bioavailability in three species, oral efficacy in a double transgenic rat model of hypertension, and good exposure in humans.