TYROSINE KINASES: COMPLEX MOLECULAR SYSTEMS CHALLENGING COMPUTATIONAL METHODOLOGIES.

Classical molecular dynamics (MD) simulations based on atomic models play an increasingly important role in a wide range of applications in physics, biology, and chemistry. Nonetheless, generating genuine knowledge about biological systems using MD simulations remains challenging. Protein tyrosine kinases are important cellular signaling enzymes that regulate cell growth, proliferation, metabolism, differentiation, and migration. Due to the large conformational changes and long timescales involved in their function, these kinases present particularly challenging problems to modern computational and theoretical frameworks aimed at elucidating the dynamics of complex biomolecular systems. Markov state models have achieved limited success in tackling the broader conformational ensemble and biased methods are often employed to examine specific long timescale events. Recent advances in machine learning continue to push the limitations of current methodologies and provide notable improvements when integrated with the existing frameworks. A broad perspective is drawn from a critical review of recent studies.

Diversity of Long-Lived Intermediates along the Binding Pathway of Imatinib to Abl Kinase Revealed by MD Simulations.

Imatinib, a drug used for the treatment of chronic myeloid leukemia and other cancers, works by blocking the catalytic site of pathological constitutively active Abl kinase. While the binding pose is known from X-ray crystallography, the different steps leading to the formation of the complex are not well understood. The results from extensive molecular dynamics simulations show that imatinib can primarily exit the known crystallographic binding pose through the cleft of the binding site or by sliding under the αC helix. Once displaced from the crystallographic binding pose, imatinib becomes trapped in intermediate states. These intermediates are characterized by a high diversity of ligand orientations and conformations, and relaxation timescales within this region may exceed 3-4 ms. Analysis indicates that the metastable intermediate states should be spectroscopically indistinguishable from the crystallographic binding pose, in agreement with tryptophan stopped-flow fluorescence experiments.

Conformational Changes in Tyrosine 11 of Neurotensin Are Required to Activate the Neurotensin Receptor 1.

Cell-cell communication via endogenous peptides and their receptors is vital for controlling all aspects of human physiology and most peptides signal through G protein-coupled receptors (GPCRs). Disordered peptides bind GPCRs through complex modes for which there are few representative crystal structures. The disordered peptide neurotensin (NT) is a neuromodulator of classical neurotransmitters such as dopamine and glutamate, through activation of neurotensin receptor 1 (NTS1). While several experimental structures show how NT binds NTS1, details about the structural dynamics of NT during and after binding NTS1, or the role of peptide dynamics on receptor activation, remain obscure. Here saturation transfer difference (STD) NMR revealed that the binding mode of NT fragment NT10-13 is heterogeneous. Epitope maps of NT10-13 at NTS1 suggested that tyrosine 11 (Y11) samples other conformations to those observed in crystal structures of NT-bound NTS1. Molecular dynamics (MD) simulations confirmed that when NT is bound to NTS1, residue Y11 can exist in two χ1 rotameric states, gauche plus (g+) or gauche minus (g-). Since only the g+ Y11 state is observed in all the structures solved to date, we asked if the g- state is important for receptor activation. NT analogues with Y11 replaced with 7-OH-Tic were synthesized to restrain the dynamics of the side chain. P(OH-TIC)IL bound NTS1 with the same affinity as NT10-13 but did not activate NTS1, instead acted as an antagonist. This study highlights that flexibility of Y11 in NT may be required for NT activation of NTS1.

Markov State Model Analysis of Haloperidol Binding to the D3 Dopamine Receptor.

We have developed Markov state models (MSMs) and hidden Markov models (HMMs) that describe the binding of haloperidol to the D3 dopamine receptor. Haloperidol is an antipsychotic drug that binds with nanomolar affinity to the D3 dopamine receptor, where it functions as an inverse agonist. The models were constructed using an adaptive sampling approach from 519 individual molecular dynamics simulations totaling 122 µs of simulated time and encompass the entire drug binding process. They reveal short-lived metastable bound states and two distinct long-lived bound conformations that cannot be separated in affinity using our current methodology. This work extends the use of MSMs and HMMs to study ligand binding, which thus far has been limited to simpler systems.

How kanamycin A interacts with bacterial and mammalian mimetic membranes.

Biological membranes are natural barriers to the transport of molecules and drugs within human bodies. Many antibacterial agents need to cross these membranes to reach their target and elicit specific effects. Kanamycin A belongs to the family of aminoglycoside antibiotics that target cellular RNA to inhibit bacterial and viral replication. Previous studies have shown that aminoglycosides bind to mammalian but disrupt bacterial membranes. In this study, molecular dynamics (MD) simulations and infrared (IR) spectroscopy were applied to investigate the initial, first key interactions of kanamycin A, as a representative aminoglycoside, with both bacterial and mammalian lipid bilayers at the molecular level. Computational studies revealed strong hydrogen bonding interactions between the hydroxyl and amino groups of the aminoglycoside with the ester carbonyl and phosphate groups of the lipids. IR spectroscopy provided experimental verification of the important role of the lipid's ester carbonyl, phosphate and hydroxyl groups for aminoglycoside binding. The bacterial membrane became disordered upon aminoglycoside addition, whereas the mammalian membrane became stiffer and more ordered. This indicates the bacterial membrane disruption observed by previous studies.

Role of the Second Extracellular Loop of the Adenosine A1 Receptor on Allosteric Modulator Binding, Signaling, and Cooperativity.

Allosteric modulation of adenosine A1 receptors (A1ARs) offers a novel therapeutic approach for the treatment of numerous central and peripheral disorders; however, despite decades of research, there is a relative paucity of structural information regarding the A1AR allosteric site and mechanisms governing cooperativity with orthosteric ligands. We combined alanine-scanning mutagenesis of the A1AR second extracellular loop (ECL2) with radioligand binding and functional interaction assays to quantify effects on allosteric ligand affinity, cooperativity, and efficacy. Docking and molecular dynamics (MD) simulations were performed using an A1AR homology model based on an agonist-bound A2AAR structure. Substitution of E172ECL2 for alanine reduced the affinity of the allosteric modulators PD81723 and VCP171 for the unoccupied A1AR. Residues involved in cooperativity with the orthosteric agonist NECA were different in PD81723 and VCP171; positive cooperativity between PD81723 and NECA was reduced on alanine substitution of a number of ECL2 residues, including E170ECL2 and K173ECL2, whereas mutation of W146ECL2 and W156ECL2 decreased VCP171 cooperativity with NECA. Molecular modeling localized a likely allosteric pocket for both modulators to an extracellular vestibule that overlaps with a region used by orthosteric ligands as they transit into the canonical A1AR orthosteric site. MD simulations confirmed a key interaction between E172ECL2 and both modulators. Bound PD81723 is flanked by another residue, E170ECL2, which forms hydrogen bonds with adjacent K168ECL2 and K173ECL2. Collectively, our data suggest E172ECL2 is a key allosteric ligand-binding determinant, whereas hydrogen-bonding networks within the extracellular vestibule may facilitate the transmission of cooperativity between orthosteric and allosteric sites.

Extracellular Loop 2 of the Adenosine A1 Receptor Has a Key Role in Orthosteric Ligand Affinity and Agonist Efficacy.

The adenosine A1 G protein-coupled receptor (A1AR) is an important therapeutic target implicated in a wide range of cardiovascular and neuronal disorders. Although it is well established that the A1AR orthosteric site is located within the receptor's transmembrane (TM) bundle, prior studies have implicated extracellular loop 2 (ECL2) as having a significant role in contributing to orthosteric ligand affinity and signaling for various G protein-coupled receptors (GPCRs). We thus performed extensive alanine scanning mutagenesis of A1AR-ECL2 to explore the role of this domain on A1AR orthosteric ligand pharmacology. Using quantitative analytical approaches and molecular modeling, we identified ECL2 residues that interact either directly or indirectly with orthosteric agonists and antagonists. Discrete mutations proximal to a conserved ECL2-TM3 disulfide bond selectively affected orthosteric ligand affinity, whereas a cluster of five residues near the TM4-ECL2 juncture influenced orthosteric agonist efficacy. A combination of ligand docking, molecular dynamics simulations, and mutagenesis results suggested that the orthosteric agonist 5'-N-ethylcarboxamidoadenosine binds transiently to an extracellular vestibule formed by ECL2 and the top of TM5 and TM7, prior to entry into the canonical TM bundle orthosteric site. Collectively, this study highlights a key role for ECL2 in A1AR orthosteric ligand binding and receptor activation.

Ligand Binding Pathways of Clozapine and Haloperidol in the Dopamine D2 and D3 Receptors.

The binding of a small molecule ligand to its protein target is most often characterized by binding affinity and is typically viewed as an on/off switch. The more complex reality is that binding involves the ligand passing through a series of intermediate states between the solution phase and the fully bound pose. We have performed a set of 29 unbiased molecular dynamics simulations to model the binding pathways of the dopamine receptor antagonists clozapine and haloperidol binding to the D2 and D3 dopamine receptors. Through these simulations we have captured the binding pathways of clozapine and haloperidol from the extracellular vestibule to the orthosteric binding site and thereby, we also predict the bound pose of each ligand. These are the first long time scale simulations of haloperidol or clozapine binding to dopamine receptors. From these simulations, we have identified several important stages in the binding pathway, including the involvement of Tyr7.35 in a "handover" mechanism that transfers the ligand between the extracellular vestibule and Asp3.32. We have also performed interaction and cluster analyses to determine differences in binding pathways between the D2 and D3 receptors and identified metastable states that may be of use in drug design.

Homology modeling of human muscarinic acetylcholine receptors.

We have developed homology models of the acetylcholine muscarinic receptors M1R-M5R, based on the ß2-adrenergic receptor crystal as the template. This is the first report of homology modeling of all five subtypes of acetylcholine muscarinic receptors with binding sites optimized for ligand binding. The models were evaluated for their ability to discriminate between muscarinic antagonists and decoy compounds using virtual screening using enrichment factors, area under the ROC curve (AUC), and an early enrichment measure, LogAUC. The models produce rational binding modes of docked ligands as well as good enrichment capacity when tested against property-matched decoy libraries, which demonstrates their unbiased predictive ability. To test the relative effects of homology model template selection and the binding site optimization procedure, we generated and evaluated a naïve M2R model, using the M3R crystal structure as a template. Our results confirm previous findings that binding site optimization using ligand(s) active at a particular receptor, i.e. including functional knowledge into the model building process, has a more pronounced effect on model quality than target-template sequence similarity. The optimized M1R-M5R homology models are made available as part of the Supporting Information to allow researchers to use these structures, compare them to their own results, and thus advance the development of better modeling approaches.