Alchemical Free Energy Calculations on Membrane-Associated Proteins

Membrane proteins have diverse functions within cells and are well-established drug targets. The advances in membrane protein structural biology have revealed drug and lipid binding sites on membrane proteins, while computational methods such as molecular simulations can resolve the thermodynamic basis of these interactions. Particularly, alchemical free energy calculations have shown promise in the calculation of reliable and reproducible binding free energies of protein-ligand and protein-lipid complexes in membrane-associated systems. In this review, we present an overview of representative alchemical free energy studies on G-protein-coupled receptors, ion channels, transporters as well as protein-lipid interactions, with emphasis on best practices and critical aspects of running these simulations. Additionally, we analyze challenges and successes when running alchemical free energy calculations on membrane-associated proteins. Finally, we highlight the value of alchemical free energy calculations calculations in drug discovery and their applicability in the pharmaceutical industry.

Broadening the Scope of Binding Free Energy Calculations Using a Separated Topologies Approach

Binding free energy calculations predict the potency of compounds to protein binding sites in a physically rigorous manner and see broad application in prioritizing the synthesis of novel drug candidates. Relative binding free energy (RBFE) calculations have emerged as an industry-standard approach to achieve highly accurate rank-order predictions of the potency of related compounds; however, this approach requires that the ligands share a common scaffold and a common binding mode, restricting the methods' domain of applicability. This is a critical limitation since complex modifications to the ligands, especially core hopping, are very common in drug design. Absolute binding free energy (ABFE) calculations are an alternate method that can be used for ligands that are not congeneric. However, ABFE suffers from a known problem of long convergence times due to the need to sample additional degrees of freedom within each system, such as sampling rearrangements necessary to open and close the binding site. Here, we report on an alternative method for RBFE, called Separated Topologies (SepTop), which overcomes the issues in both of the aforementioned methods by enabling large scaffold changes between ligands with a convergence time comparable to traditional RBFE. Instead of only mutating atoms that vary between two ligands, this approach performs two absolute free energy calculations at the same time in opposite directions, one for each ligand. Defining the two ligands independently allows the comparison of the binding of diverse ligands without the artificial constraints of identical poses or a suitable atom-atom mapping. This approach also avoids the need to sample the unbound state of the protein, making it more efficient than absolute binding free energy calculations. Here, we introduce an implementation of SepTop. We developed a general and efficient protocol for running SepTop, and we demonstrated the method on four diverse, pharmaceutically relevant systems. We report the performance of the method, as well as our practical insights into the strengths, weaknesses, and challenges of applying this method in an industrial drug design setting. We find that the accuracy of the approach is sufficiently high to rank order ligands with an accuracy comparable to traditional RBFE calculations while maintaining the additional flexibility of SepTop.

DeltaDelta neural networks for lead optimization of small molecule potency

The capability to rank different potential drug molecules against a protein target for potency has always been a fundamental challenge in computational chemistry due to its importance in drug design. While several simulation-based methodologies exist, they are hard to use prospectively and thus predicting potency in lead optimization campaigns remains an open challenge. Here we present the first machine learning approach specifically tailored for ranking congeneric series based on deep 3D-convolutional neural networks. Furthermore we prove its effectiveness by blindly testing it on datasets provided by Janssen, Pfizer and Biogen totalling over 3246 ligands and 13 targets as well as several well-known openly available sets, representing one the largest evaluations ever performed. We also performed online learning simulations of lead optimization using the approach in a predictive manner obtaining significant advantage over experimental choice. We believe that the evaluation performed in this study is strong evidence of the usefulness of a modern deep learning model in lead optimization pipelines against more expensive simulation-based alternatives.

Large scale relative protein ligand binding affinities using non-equilibrium alchemy

Ligand binding affinity calculations based on molecular dynamics (MD) simulations and non-physical (alchemical) thermodynamic cycles have shown great promise for structure-based drug design. However, their broad uptake and impact is held back by the notoriously complex setup of the calculations. Only a few tools other than the free energy perturbation approach by Schrödinger Inc. (referred to as FEP+) currently enable end-to-end application. Here, we present for the first time an approach based on the open-source software pmx that allows to easily set up and run alchemical calculations for diverse sets of small molecules using the GROMACS MD engine. The method relies on theoretically rigorous non-equilibrium thermodynamic integration (TI) foundations, and its flexibility allows calculations with multiple force fields. In this study, results from the Amber and Charmm force fields were combined to yield a consensus outcome performing on par with the commercial FEP+ approach. A large dataset of 482 perturbations from 13 different protein-ligand datasets led to an average unsigned error (AUE) of 3.64 ± 0.14 kJ mol-1, equivalent to Schrödinger's FEP+ AUE of 3.66 ± 0.14 kJ mol-1. For the first time, a setup is presented for overall high precision and high accuracy relative protein-ligand alchemical free energy calculations based on open-source software.

The size matters? A computational tool to design bivalent ligands

Motivation

Bivalent ligands are increasingly important such as for targeting G protein-coupled receptor (GPCR) dimers or proteolysis targeting chimeras (PROTACs). They contain two pharmacophoric units that simultaneously bind in their corresponding binding sites, connected with a spacer chain. Here, we report a molecular modelling tool that links the pharmacophore units via the shortest pathway along the receptors van der Waals surface and then scores the solutions providing prioritization for the design of new bivalent ligands.

Results

Bivalent ligands of known dimers of GPCRs, PROTACs and a model bivalent antibody/antigen system were analysed. The tool could rapidly assess the preferred linker length for the different systems and recapitulated the best reported results. In the case of GPCR dimers the results suggest that in some cases these ligands might bind to a secondary binding site at the extracellular entrance (vestibule or allosteric site) instead of the orthosteric binding site.

Availability and implementation

Freely accessible from the Molecular Operating Environment svl exchange server (https://svl.chemcomp.com/).

Supplementary information

Supplementary data are available at Bioinformatics online.

Computational Drug Design Applied to the Study of Metabotropic Glutamate Receptors

Metabotropic glutamate (mGlu) receptors are a family of eight GPCRs that are attractive drug discovery targets to modulate glutamate action and response. Here we review the application of computational methods to the study of this family of receptors. X-ray structures of the extracellular and 7-transmembrane domains have played an important role to enable structure-based modeling approaches, whilst we also discuss the successful application of ligand-based methods. We summarize the literature and highlight the areas where modeling and experiment have delivered important understanding for mGlu receptor drug discovery. Finally, we offer suggestions of future areas of opportunity for computational work.

Computationally Guided Identification of Allosteric Agonists of the Metabotropic Glutamate 7 Receptor

The metabotropic glutamate 7 (mGlu₇) receptor belongs to the group III of mGlu receptors. Since the mGlu₇ receptor can control excitatory neurotransmission in the hippocampus and cortex, modulation of the receptor may have therapeutic benefit in several CNS diseases. However, mGlu₇ remains relatively unexplored among the eight known mGlu receptors partly because of the limited availability of tool compounds to interrogate its potential therapeutic utility. Here we report the discovery of a new class of mGlu₇ allosteric agonists. Hits originating from virtual screening were followed up with further analogue searching and screening, leading to a novel series of mGlu₇ allosteric agonists. Guided by docking into a structural model of the mGlu₇ receptor the initial hit 5 was successfully optimized to analogues with comparable potencies and more attractive drug-like attributes than AMN082.

Mechanisms Underlying Allosteric Molecular Switches of Metabotropic Glutamate Receptor 5

The metabotropic glutamate 5 (mGlu₅) receptor is a class C G protein-coupled receptor (GPCR) that is implicated in several CNS disorders making it a popular drug discovery target. Years of research have revealed allosteric mGlu₅ ligands showing an unexpected complete switch in functional activity despite only small changes in their chemical structure, resulting in positive allosteric modulators (PAM) or negative allosteric modulators (NAM) for the same scaffold. Up to now, the origins of this effect are not understood, causing difficulties in a drug discovery context. In this work, experimental data was gathered and analyzed alongside docking and Molecular Dynamics (MD) calculations for three sets of PAM and NAM pairs. The results consistently show the role of specific interactions formed between ligand substituents and amino acid side chains that block or promote local movements associated with receptor activation. The work provides an explanation for how such small structural changes lead to remarkable differences in functional activity. While this work can greatly help drug discovery programs avoid these switches, it also provides valuable insight into the mechanisms of class C GPCR allosteric activation. Furthermore, the approach shows the value of applying MD to understand functional activity in drug design programs, even for such close structural analogues.

Design of a True Bivalent Ligand with Picomolar Binding Affinity for a G Protein-Coupled Receptor Homodimer

Bivalent ligands have emerged as chemical tools to study G protein-coupled receptor dimers. Using a combination of computational, chemical, and biochemical tools, here we describe the design of bivalent ligand 13 with high affinity ( K_DB1 = 21 pM) for the dopamine D₂ receptor (D₂R) homodimer. Bivalent ligand 13 enhances the binding affinity relative to monovalent compound 15 by 37-fold, indicating simultaneous binding at both protomers. Using synthetic peptides with amino acid sequences of transmembrane (TM) domains of D₂R, we provide evidence that TM6 forms the interface of the homodimer. Notably, the disturber peptide TAT-TM6 decreased the binding of bivalent ligand 13 by 52-fold and had no effect on monovalent compound 15, confirming the D₂R homodimer through TM6 ex vivo. In conclusion, by using a versatile multivalent chemical platform, we have developed a precise strategy to generate a true bivalent ligand that simultaneously targets both orthosteric sites of the D₂R homodimer.

Covalent Allosteric Probe for the Metabotropic Glutamate Receptor 2: Design, Synthesis, and Pharmacological Characterization

Covalent labeling of G protein-coupled receptors (GPCRs) by small molecules is a powerful approach to understand binding modes, mechanism of action, pharmacology, and even facilitate structure elucidation. We report the first covalent positive allosteric modulator (PAM) for a class C GPCR, the mGlu₂ receptor. Three putatively covalent mGlu₂ PAMs were designed and synthesized. Pharmacological characterization identified 2 to bind the receptor covalently. Computational modeling combined with receptor mutagenesis revealed T791^7.29×30 as the likely position of covalent interaction. We show how this covalent ligand can be used to characterize the PAM binding mode and that it is a valuable tool compound in studying receptor function and binding kinetics. Our findings advance the understanding of the mGlu₂ PAM interaction and suggest that 2 is a valuable probe for further structural and chemical biology approaches.

Cross-communication between Gi and Gs in a G-protein-coupled receptor heterotetramer guided by a receptor C-terminal domain

Background

G-protein-coupled receptor (GPCR) heteromeric complexes have distinct properties from homomeric GPCRs, giving rise to new receptor functionalities. Adenosine receptors (A₁R or A_2AR) can form A₁R-A_2AR heteromers (A₁-A_2AHet), and their activation leads to canonical G-protein-dependent (adenylate cyclase mediated) and -independent (β-arrestin mediated) signaling. Adenosine has different affinities for A₁R and A_2AR, allowing the heteromeric receptor to detect its concentration by integrating the downstream G_i- and G_s-dependent signals. cAMP accumulation and β-arrestin recruitment assays have shown that, within the complex, activation of A_2AR impedes signaling via A₁R.

Results

We examined the mechanism by which A₁-A_2AHet integrates G_i- and G_s-dependent signals. A₁R blockade by A_2AR in the A₁-A_2AHet is not observed in the absence of A_2AR activation by agonists, in the absence of the C-terminal domain of A_2AR, or in the presence of synthetic peptides that disrupt the heteromer interface of A₁-A_2AHet, indicating that signaling mediated by A₁R and A_2AR is controlled by both G_i and G_s proteins.

Conclusions

We identified a new mechanism of signal transduction that implies a cross-communication between G_i and G_s proteins guided by the C-terminal tail of the A_2AR. This mechanism provides the molecular basis for the operation of the A₁-A_2AHet as an adenosine concentration-sensing device that modulates the signals originating at both A₁R and A_2AR.

Electronic supplementary material

The online version of this article (10.1186/s12915-018-0491-x) contains supplementary material, which is available to authorized users.

Identification of Allosteric Modulators of Metabotropic Glutamate 7 Receptor Using Proteochemometric Modeling

Proteochemometric modeling (PCM) is a computational approach that can be considered an extension of quantitative structure-activity relationship (QSAR) modeling, where a single model incorporates information for a family of targets and all the associated ligands instead of modeling activity versus one target. This is especially useful for situations where bioactivity data exists for similar proteins but is scarce for the protein of interest. Here we demonstrate the application of PCM to identify allosteric modulators of metabotropic glutamate (mGlu) receptors. Given our long-running interest in modulating mGlu receptor function we compiled a matrix of compound-target bioactivity data. Some members of the mGlu family are well explored both internally and in the public domain, while there are much fewer examples of ligands for other targets such as the mGlu7 receptor. Using a PCM approach mGlu7 receptor hits were found. In comparison to conventional single target modeling the identified hits were more diverse, had a better confirmation rate, and provide starting points for further exploration. We conclude that the robust structure-activity relationship from well explored target family members translated to better quality hits for PCM compared to virtual screening (VS) based on a single target.

Acylguanidine Beta Secretase 1 Inhibitors: A Combined Experimental and Free Energy Perturbation Study

A series of acylguanidine beta secretase 1 (BACE1) inhibitors with modified scaffold and P3 pocket substituent was synthesized and studied with free energy perturbation (FEP) calculations. The resulting molecules showed potencies in enzymatic BACE1 inhibition assays up to 1 nM. The correlation between the predicted activity from the FEP calculations and the experimental activity was good for the P3 pocket substituents. The average mean unsigned error (MUE) between prediction and experiment was 0.68 ± 0.17 kcal/mol for the default 5 ns lambda window simulation time improving to 0.35 ± 0.13 kcal/mol for 40 ns. FEP calculations for the P2' pocket substituents on the same acylguanidine scaffold also showed good agreement with experiment and the results remained stable with repeated simulations and increased simulation time. It proved more difficult to use FEP calculations to study the scaffold modification from increasing 5 to 6 and 7 membered-rings. Although prediction and experiment were in agreement for short 2 ns simulations, as the simulation time increased the results diverged. This was improved by the use of a newly developed "Core Hopping FEP+" approach, which also showed improved stability in repeat calculations. The origins of these differences along with the value of repeat and longer simulation times are discussed. This work provides a further example of the use of FEP as a computational tool for molecular design.

Discovery of 8-Trifluoromethyl-3-cyclopropylmethyl-7-[(4-(2,4-difluorophenyl)-1-piperazinyl)methyl]-1,2,4-triazolo[4,3-a]pyridine (JNJ-46356479), a Selective and Orally Bioavailable mGlu2 Receptor Positive Allosteric Modulator (PAM)

Positive allosteric modulators of the metabotropic glutamate 2 receptor have generated great interest in the past decade. There is mounting evidence of their potential as therapeutic agents in the treatment of multiple central nervous system disorders. We have previously reported substantial efforts leading to potent and selective mGlu2 PAMs. However, finding compounds with the optimal combination of in vitro potency and good druglike properties has remained elusive, in part because of the hydrophobic nature of the allosteric binding site. Herein, we report on the lead optimization process to overcome the poor solubility inherent to the advanced lead 6. Initial prototypes already showed significant improvements in solubility while retaining good functional activity but displayed new liabilities associated with metabolism and hERG inhibition. Subsequent subtle modifications efficiently addressed those issues leading to the identification of compound 27 (JNJ-46356479). This new lead represents a more balanced profile that offers a significant improvement on the druglike attributes compared to previously reported leads.

Application of Free Energy Perturbation for the Design of BACE1 Inhibitors

Novel spiroaminodihydropyrroles probing for optimized interactions at the P3 pocket of β-secretase 1 (BACE1) were designed with the use of free energy perturbation (FEP) calculations. The resulting molecules showed pIC50 potencies in enzymatic BACE1 inhibition assays ranging from approximately 5 to 7. Good correlation was observed between the predicted activity from the FEP calculations and experimental activity. Simulations run with a default 5 ns approach delivered a mean unsigned error (MUE) between prediction and experiment of 0.58 and 0.91 kcal/mol for retrospective and prospective applications, respectively. With longer simulations of 10 and 20 ns, the MUE was in both cases 0.57 kcal/mol for the retrospective application, and 0.69 and 0.59 kcal/mol for the prospective application. Other considerations that impact the quality of the calculations are discussed. This work provides an example of the value of FEP as a computational tool for drug discovery.

Molecular mechanism of positive allosteric modulation of the metabotropic glutamate receptor 2 by JNJ-46281222

BACKGROUND AND PURPOSE

Allosteric modulation of the mGlu2 receptor is a potential strategy for treatment of various neurological and psychiatric disorders. Here, we describe the in vitro characterization of the mGlu2 positive allosteric modulator (PAM) JNJ-46281222 and its radiolabelled counterpart [(3) H]-JNJ-46281222. Using this novel tool, we also describe the allosteric effect of orthosteric glutamate binding and the presence of a bound G protein on PAM binding and use computational approaches to further investigate the binding mode.

EXPERIMENTAL APPROACH

We have used radioligand binding studies, functional assays, site-directed mutagenesis, homology modelling and molecular dynamics to study the binding of JNJ-46281222.

KEY RESULTS

JNJ-46281222 is an mGlu2 -selective, highly potent PAM with nanomolar affinity (KD = 1.7 nM). Binding of [(3) H]-JNJ-46281222 was increased by the presence of glutamate and greatly reduced by the presence of GTP, indicating the preference for a G protein bound state of the receptor for PAM binding. Its allosteric binding site was visualized and analysed by a computational docking and molecular dynamics study. The simulations revealed amino acid movements in regions expected to be important for activation. The binding mode was supported by [(3) H]-JNJ-46281222 binding experiments on mutant receptors.

CONCLUSION AND IMPLICATIONS

Our results obtained with JNJ-46281222 in unlabelled and tritiated form further contribute to our understanding of mGlu2 allosteric modulation. The computational simulations and mutagenesis provide a plausible binding mode with indications of how the ligand permits allosteric activation. This study is therefore of interest for mGlu2 and class C receptor drug discovery.

Gastrin-stimulated Gα13 Activation of Rgnef Protein (ArhGEF28) in DLD-1 Colon Carcinoma Cells

The guanine nucleotide exchange factor Rgnef (also known as ArhGEF28 or p190RhoGEF) promotes colon carcinoma cell motility and tumor progression via interaction with focal adhesion kinase (FAK). Mechanisms of Rgnef activation downstream of integrin or G protein-coupled receptors remain undefined. In the absence of a recognized G protein signaling homology domain in Rgnef, no proximal linkage to G proteins was known. Utilizing multiple methods, we have identified Rgnef as a new effector for Gα13 downstream of gastrin and the type 2 cholecystokinin receptor. In DLD-1 colon carcinoma cells depleted of Gα13, gastrin-induced FAK Tyr(P)-397 and paxillin Tyr(P)-31 phosphorylation were reduced. RhoA GTP binding and promoter activity were increased by Rgnef in combination with active Gα13. Rgnef co-immunoprecipitated with activated Gα13Q226L but not Gα12Q229L. The Rgnef C-terminal (CT, 1279-1582) region was sufficient for co-immunoprecipitation, and Rgnef-CT exogenous expression prevented Gα13-stimulated SRE activity. A domain at the C terminus of the protein close to the FAK binding domain is necessary to bind to Gα13. Point mutations of Rgnef-CT residues disrupt association with active Gα13 but not Gαq. These results show that Rgnef functions as an effector of Gα13 signaling and that this linkage may mediate FAK activation in DLD-1 colon carcinoma cells.