Biasing AlphaFold2 to predict GPCRs and kinases with user-defined functional or structural properties

Hederagenin is a Highly Selective Antagonist of the Neuropeptide FF Receptor 1 that Reveals Mechanisms for Subtype Selectivity

RF-amide peptide receptors including the neuropeptide FF receptor 1 (NPFFR1) are G protein-coupled receptors (GPCRs) that modulate diverse physiological functions. High conservation of endogenous ligands and receptors makes the identification of selective ligands challenging. Previously identified antagonists mimic the C-terminus of peptide ligands and lack selectivity towards the closely related neuropeptide FF receptor 2 (NPFFR2) or the neuropeptide Y1 receptor (Y1R). In a high-throughput screening, we identified the pentacyclic triterpenoid hederagenin (1) as a novel selective antagonist for the NPFFR1. Hederagenin (1) is a natural product isolated from Hedera helix (ivy). We characterized its mode of activity using in vitro and in silico methods, revealing an overlapping binding site of the small molecule with the orthosteric peptide agonists. Despite the high similarity of the orthosteric binding pockets of NPFFR1 and NPFFR2, hederagenin (1) shows strong subtype selectivity, particularly caused by slight differences in the shape of the binding pockets and the rigidity of the small molecule. Several residues inhibiting the activity of hederagenin (1) at the NPFFR2 were identified. As NPFFR1 antagonists are discussed as potential candidates for the treatment of chronic pain, these insights into the structural determinants governing subtype specificity will facilitate the development of next-generation analgesics with improved safety and efficacy.

An outlook on structural biology after AlphaFold: tools, limits and perspectives

AlphaFold and similar groundbreaking, AI-based tools, have revolutionized the field of structural bioinformatics, with their remarkable accuracy in ab-initio protein structure prediction. This success has catalyzed the development of new software and pipelines aimed at incorporating AlphaFold's predictions, often focusing on addressing the algorithm's remaining challenges. Here, we present the current landscape of structural bioinformatics shaped by AlphaFold, and discuss how the field is dynamically responding to this revolution, with new software, methods, and pipelines. While the excitement around AI-based tools led to their widespread application, it is essential to acknowledge that their practical success hinges on their integration into established protocols within structural bioinformatics, often neglected in the context of AI-driven advancements. Indeed, user-driven intervention is still as pivotal in the structure prediction process as in complementing state-of-the-art algorithms with functional and biological knowledge.

Leveraging Artificial Intelligence in GPCR Activation Studies: Computational Prediction Methods as Key Drivers of Knowledge

G protein-coupled receptors (GPCRs) are key molecules involved in cellular signaling and are attractive targets for pharmacological intervention. This chapter is designed to explore the range of algorithms used to predict GPCRs' activation states, while also examining the pharmaceutical implications of these predictions. Our primary objective is to show how artificial intelligence (AI) is key in GPCR research to reveal the intricate dynamics of activation and inactivation processes, shedding light on the complex regulatory mechanisms of this vital protein family. We describe several computational strategies that leverage diverse structural data from the Protein Data Bank, molecular dynamic simulations, or ligand-based methods to predict the activation states of GPCRs. We demonstrate how the integration of AI into GPCR research not only enhances our understanding of their dynamic properties but also presents immense potential for driving pharmaceutical research and development, offering promising new avenues in the search for newer, better therapeutic agents.

Modeling Alternative Conformational States of Pseudo-Symmetric Solute Carrier Transporters using Methods from Deep Learning

The Solute Carrier (SLC) superfamily of integral membrane proteins function to transport a wide array of small molecules across plasma and organelle membranes. SLC proteins also function as important drug transporters and as viral receptors. Despite being classified as a single superfamily, SLC proteins do not share a single common fold classification; however, most belong to multi-pass transmembrane helical protein fold families. SLC proteins populate different conformational states during the solute transport process, including outward-open, intermediate (occluded), and inward-open conformational states. For some SLC fold families this structural "flipping" corresponds to swapping between conformations of their N-terminal and C-terminal symmetry-related sub-structures. Conventional AlphaFold2, AlphaFold3, or Evolutionary Scale Modeling methods typically generate models for only one of these multiple conformational states of SLC proteins. Several modifications of these AI-based protocols for modeling multiple conformational states of proteins have been described recently. These methods are often impacted by "memorization" of one of the alternative conformational states, and do not always provide both the inward and outward facing conformations of SLC proteins. Here we describe a combined ESM - template-based-modeling process, based on a previously described template-based modeling method that relies on the internal pseudo-symmetry of many SLC proteins, to consistently model alternate conformational states of SLC proteins. We further demonstrate how the resulting multi-state models can be validated experimentally by comparison with sequence-based evolutionary co-variance data (ECs) that encode information about contacts present in the various conformational states adopted by the protein. This simple, rapid, and robust approach for modeling conformational landscapes of pseudo-symmetric SLC proteins is demonstrated for several integral membrane protein transporters, including SLC35F2 the receptor of a feline leukemia virus envelope protein required for viral entry into eukaryotic cells.

Targeting adhesion G protein-coupled receptors. Current status and future perspectives

G protein-coupled receptors (GPCRs) orchestrate many physiological functions and are a crucial target in drug discovery. Adhesion GPCRs (aGPCRs), the second largest family within this superfamily, are promising yet underexplored targets for treating various diseases, including obesity, psychiatric disorders, and cancer. However, the receptors' unique and complex structure and miscellaneous interactions complicate comprehensive pharmacological studies. Despite recent progress in determining structures and elucidation of the activation mechanism, the function of many receptors remains to be determined. This review consolidates current knowledge on aGPCR ligands, focusing on small molecule orthosteric ligands and allosteric modulators identified for the ADGRGs subfamily (subfamily VIII), (GPR56/ADGRG1, GPR64/ADGRG2, GPR97/ADGRG3, GPR114/ADGRG5, GPR126/ADGRG6, and GPR128/ADGRG7). We discuss challenges in hit identification, target validation, and drug discovery, highlighting molecular compositions and recent structural breakthroughs. ADGRG ligands can offer new insights into aGPCR modulation and have significant potential for novel therapeutic interventions targeting various diseases.

AlphaFold2-Based Characterization of Apo and Holo Protein Structures and Conformational Ensembles Using Randomized Alanine Sequence Scanning Adaptation: Capturing Shared Signature Dynamics and Ligand-Induced Conformational Changes

Proteins often exist in multiple conformational states, influenced by the binding of ligands or substrates. The study of these states, particularly the apo (unbound) and holo (ligand-bound) forms, is crucial for understanding protein function, dynamics, and interactions. In the current study, we use AlphaFold2, which combines randomized alanine sequence masking with shallow multiple sequence alignment subsampling to expand the conformational diversity of the predicted structural ensembles and capture conformational changes between apo and holo protein forms. Using several well-established datasets of structurally diverse apo-holo protein pairs, the proposed approach enables robust predictions of apo and holo structures and conformational ensembles, while also displaying notably similar dynamics distributions. These observations are consistent with the view that the intrinsic dynamics of allosteric proteins are defined by the structural topology of the fold and favor conserved conformational motions driven by soft modes. Our findings provide evidence that AlphaFold2 combined with randomized alanine sequence masking can yield accurate and consistent results in predicting moderate conformational adjustments between apo and holo states, especially for proteins with localized changes upon ligand binding. For large hinge-like domain movements, the proposed approach can predict functional conformations characteristic of both apo and ligand-bound holo ensembles in the absence of ligand information. These results are relevant for using this AlphaFold adaptation for probing conformational selection mechanisms according to which proteins can adopt multiple conformations, including those that are competent for ligand binding. The results of this study indicate that robust modeling of functional protein states may require more accurate characterization of flexible regions in functional conformations and the detection of high-energy conformations. By incorporating a wider variety of protein structures in training datasets, including both apo and holo forms, the model can learn to recognize and predict the structural changes that occur upon ligand binding.

ESM-scan-A tool to guide amino acid substitutions

Protein structure prediction and (re)design have gone through a revolution in the last 3 years. The tremendous progress in these fields has been almost exclusively driven by readily available machine learning algorithms applied to protein folding and sequence design problems. Despite these advancements, predicting site-specific mutational effects on protein stability and function remains an unsolved problem. This is a persistent challenge, mainly because the free energy of large systems is very difficult to compute with absolute accuracy and subtle changes to protein structures are hard to capture with computational models. Here, we describe the implementation and use of ESM-Scan, which uses the ESM zero-shot predictor to scan entire protein sequences for preferential amino acid changes, thus enabling in silico deep mutational scanning experiments. We benchmark ESM-Scan on its predictive capabilities for stability and functionality of sequence changes using three publicly available datasets and proceed by experimentally testing the tool's performance on a challenging test case of a blue-light-activated diguanylate cyclase from Methylotenera species (MsLadC), where it accurately predicted the importance of a highly conserved residue in a region involved in allosteric product inhibition. Our experimental results show that the ESM-zero shot model is capable of inferring the effects of a set of amino acid substitutions in their correlation between predicted fitness and experimental results. ESM-Scan is publicly available at https://huggingface.co/spaces/thaidaev/zsp.

Calcineurin-fusion facilitates cryo-EM structure determination of a Family A GPCR

Advances in singe-particle cryo-electron microscopy (cryo-EM) have made it possible to solve the structures of numerous Family A and Family B G protein-coupled receptors (GPCRs) in complex with G proteins and arrestins, as well as several Family C GPCRs. Determination of these structures has been facilitated by the presence of large extramembrane components (such as G protein, arrestin, or Venus flytrap domains) in these complexes that aid in particle alignment during the processing of the cryo-EM data. In contrast, determination of the inactive state structure of Family A GPCRs is more challenging due to the relatively small size of the seven transmembrane domain (7TM) and to the surrounding detergent micelle that, in the absence of other features, make particle alignment impossible. Here, we describe an alternative protein engineering strategy where the heterodimeric protein calcineurin is fused to a GPCR by three points of attachment, the cytoplasmic ends of TM5, TM6, and TM7. This three-point attachment provides a more rigid link with the GPCR transmembrane domain that facilitates particle alignment during data processing, allowing us to determine the structures of the β2 adrenergic receptor (β2AR) in the apo, antagonist-bound, and agonist-bound states. We expect that this fusion strategy may have broad application in cryo-EM structural determination of other Family A GPCRs.

Probing Functional Allosteric States and Conformational Ensembles of the Allosteric Protein Kinase States and Mutants: Atomistic Modeling and Comparative Analysis of AlphaFold2, OmegaFold, and AlphaFlow Approaches and Adaptations

This study reports a comprehensive analysis and comparison of several AlphaFold2 adaptations and OmegaFold and AlphaFlow approaches in predicting distinct allosteric states, conformational ensembles, and mutation-induced structural effects for a panel of state-switching allosteric ABL mutants. The results revealed that the proposed AlphaFold2 adaptation with randomized alanine sequence scanning can generate functionally relevant allosteric states and conformational ensembles of the ABL kinase that qualitatively capture a unique pattern of population shifts between the active and inactive states in the allosteric ABL mutants. Consistent with the NMR experiments, the proposed AlphaFold2 adaptation predicted that G269E/M309L/T408Y mutant could induce population changes and sample a significant fraction of the fully inactive I2 form which is a low-populated, high-energy state for the wild-type ABL protein. We also demonstrated that other ABL mutants G269E/M309L/T334I and M309L/L320I/T334I that introduce a single activating T334I mutation can reverse equilibrium and populate exclusively the active ABL form. While the precise quantitative predictions of the relative populations of the active and various hidden inactive states in the ABL mutants remain challenging, our results provide evidence that AlphaFold2 adaptation with randomized alanine sequence scanning can adequately detect a spectrum of the allosteric ABL states and capture the equilibrium redistributions between structurally distinct functional ABL conformations. We further validated the robustness of the proposed AlphaFold2 adaptation for predicting the unique inactive architecture of the BSK8 kinase and structural differences between ligand-unbound apo and ATP-bound forms of BSK8. The results of this comparative study suggested that AlpahFold2, OmegaFold, and AlphaFlow approaches may be driven by structural memorization of existing protein folds and are strongly biased toward predictions of the thermodynamically stable ground states of the protein kinases, highlighting limitations and challenges of AI-based methodologies in detecting alternative functional conformations, accurate characterization of physically significant conformational ensembles, and prediction of mutation-induced allosteric structural changes.

A Multi-Angle Approach to Predict Peptide-GPCR Complexes: The N/OFQ-NOP System as a Successful AlphaFold Application Case Study

With nearly 700 structures solved and a growing number of customized structure prediction algorithms being developed at a fast pace, G protein-coupled receptors (GPCRs) are an optimal test case for validating new approaches for the prediction of receptor active state and ligand bioactive conformation complexes. In this study, we leveraged the availability of hundreds of peptide GPCRs in the active state and both classical homology and artificial intelligence (AI) based protein modeling combined with docking and AI-based peptide structure prediction approaches to predict the nociceptin/orphanin FQ-NOP receptor active state complex (N/OFQ-NOPa). The In Silico generated hypotheses were validated via the design, synthesis, and pharmacological characterization of novel linear N/OFQ(1-13)-NH2 analogues, leading to the discovery of a novel antagonist (3B; pKB = 6.63) bearing a single ring-constrained residue in place of the Gly2-Gly3 motif of the N/OFQ message sequence (FGGF). While the experimental validation was ongoing, the availability of the Cryo-EM structure of the predicted complex enabled us to unambiguously validate the generated hypotheses. To the best of our knowledge, this is the first example of a peptide-GPCR complex predicted with atomistic accuracy (full complex Cα RMSD < 1.0 Å) and of the N/OFQ message moiety being successfully modified with a rigid scaffold.

Harnessing AlphaFold to reveal hERG channel conformational state secrets

To design safe, selective, and effective new therapies, there must be a deep understanding of the structure and function of the drug target. One of the most difficult problems to solve has been resolution of discrete conformational states of transmembrane ion channel proteins. An example is KV11.1 (hERG), comprising the primary cardiac repolarizing current, I Kr. hERG is a notorious drug anti-target against which all promising drugs are screened to determine potential for arrhythmia. Drug interactions with the hERG inactivated state are linked to elevated arrhythmia risk, and drugs may become trapped during channel closure. However, the structural details of multiple conformational states have remained elusive. Here, we guided AlphaFold2 to predict plausible hERG inactivated and closed conformations, obtaining results consistent with multiple available experimental data. Drug docking simulations demonstrated hERG state-specific drug interactions in good agreement with experimental results, revealing that most drugs bind more effectively in the inactivated state and are trapped in the closed state. Molecular dynamics simulations demonstrated ion conduction for an open but not AlphaFold2 predicted inactivated state that aligned with earlier studies. Finally, we identified key molecular determinants of state transitions by analyzing interaction networks across closed, open, and inactivated states in agreement with earlier mutagenesis studies. Here, we demonstrate a readily generalizable application of AlphaFold2 as an effective and robust method to predict discrete protein conformations, reconcile seemingly disparate data and identify novel linkages from structure to function.

Click, Compute, Create: A Review of Web-based Tools for Enzyme Engineering

Enzyme engineering, though pivotal across various biotechnological domains, is often plagued by its time-consuming and labor-intensive nature. This review aims to offer an overview of supportive in silico methodologies for this demanding endeavor. Starting from methods to predict protein structures, to classification of their activity and even the discovery of new enzymes we continue with describing tools used to increase thermostability and production yields of selected targets. Subsequently, we discuss computational methods to modulate both, the activity as well as selectivity of enzymes. Last, we present recent approaches based on cutting-edge machine learning methods to redesign enzymes. With exception of the last chapter, there is a strong focus on methods easily accessible via web-interfaces or simple Python-scripts, therefore readily useable for a diverse and broad community.

Predicting Mutation-Induced Allosteric Changes in Structures and Conformational Ensembles of the ABL Kinase Using AlphaFold2 Adaptations with Alanine Sequence Scanning

Despite the success of AlphaFold2 approaches in predicting single protein structures, these methods showed intrinsic limitations in predicting multiple functional conformations of allosteric proteins and have been challenged to accurately capture the effects of single point mutations that induced significant structural changes. We examined several implementations of AlphaFold2 methods to predict conformational ensembles for state-switching mutants of the ABL kinase. The results revealed that a combination of randomized alanine sequence masking with shallow multiple sequence alignment subsampling can significantly expand the conformational diversity of the predicted structural ensembles and capture shifts in populations of the active and inactive ABL states. Consistent with the NMR experiments, the predicted conformational ensembles for M309L/L320I and M309L/H415P ABL mutants that perturb the regulatory spine networks featured the increased population of the fully closed inactive state. The proposed adaptation of AlphaFold can reproduce the experimentally observed mutation-induced redistributions in the relative populations of the active and inactive ABL states and capture the effects of regulatory mutations on allosteric structural rearrangements of the kinase domain. The ensemble-based network analysis complemented AlphaFold predictions by revealing allosteric hotspots that correspond to state-switching mutational sites which may explain the global effect of regulatory mutations on structural changes between the ABL states. This study suggested that attention-based learning of long-range dependencies between sequence positions in homologous folds and deciphering patterns of allosteric interactions may further augment the predictive abilities of AlphaFold methods for modeling of alternative protein sates, conformational ensembles and mutation-induced structural transformations.

Empowering AlphaFold2 for protein conformation selective drug discovery with AlphaFold2-RAVE

Small-molecule drug design hinges on obtaining co-crystallized ligand-protein structures. Despite AlphaFold2's strides in protein native structure prediction, its focus on apo structures overlooks ligands and associated holo structures. Moreover, designing selective drugs often benefits from the targeting of diverse metastable conformations. Therefore, direct application of AlphaFold2 models in virtual screening and drug discovery remains tentative. Here, we demonstrate an AlphaFold2-based framework combined with all-atom enhanced sampling molecular dynamics and Induced Fit docking, named AF2RAVE-Glide, to conduct computational model-based small-molecule binding of metastable protein kinase conformations, initiated from protein sequences. We demonstrate the AF2RAVE-Glide workflow on three different mammalian protein kinases and their type I and II inhibitors, with special emphasis on binding of known type II kinase inhibitors which target the metastable classical DFG-out state. These states are not easy to sample from AlphaFold2. Here, we demonstrate how with AF2RAVE these metastable conformations can be sampled for different kinases with high enough accuracy to enable subsequent docking of known type II kinase inhibitors with more than 50% success rates across docking calculations. We believe the protocol should be deployable for other kinases and more proteins generally.

Hit-to-Lead Optimization of Heterocyclic Carbonyloxycarboximidamides as Selective Antagonists at Human Adenosine A3 Receptor

Antagonism of the human adenosine A3 receptor (hA3R) has potential therapeutic application. Alchemical relative binding free energy calculations of K18 and K32 suggested that the combination of a 3-(2,6-dichlorophenyl)-isoxazolyl group with 2-pyridinyl at the ends of a carbonyloxycarboximidamide group should improve hA3R affinity. Of the 25 new analogues synthesized, 37 and 74 showed improved hA3R affinity compared to K18 (and K32). This was further improved through the addition of a bromine group to the 2-pyridinyl at the 5-position, generating compound 39. Alchemical relative binding free energy calculations, mutagenesis studies and MD simulations supported the compounds' binding pattern while suggesting that the bromine of 39 inserts deep into the hA3R orthosteric pocket, so highlighting the importance of rigidification of the carbonyloxycarboximidamide moiety. MD simulations highlighted the importance of rigidification of the carbonyloxycarboximidamide, while suggesting that the bromine of 39 inserts deep into the hA3R orthosteric pocket, which was supported through mutagenesis studies 39 also selectively antagonized endogenously expressed hA3R in nonsmall cell lung carcinoma cells, while pharmacokinetic studies indicated low toxicity enabling in vivo evaluation. We therefore suggest that 39 has potential for further development as a high-affinity hA3R antagonist.

The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins

Artificial intelligence-driven advances in protein structure prediction in recent years have raised the question: has the protein structure-prediction problem been solved? Here, with a focus on nonglobular proteins, we highlight the many strengths and potential weaknesses of DeepMind's AlphaFold2 in the context of its biological and therapeutic applications. We summarize the subtleties associated with evaluation of AlphaFold2 model quality and reliability using the predicted local distance difference test (pLDDT) and predicted aligned error (PAE) values. We highlight various classes of proteins that AlphaFold2 can be applied to and the caveats involved. Concrete examples of how AlphaFold2 models can be integrated with experimental data in the form of small-angle X-ray scattering (SAXS), solution NMR, cryo-electron microscopy (cryo-EM) and X-ray diffraction are discussed. Finally, we highlight the need to move beyond structure prediction of rigid, static structural snapshots toward conformational ensembles and alternate biologically relevant states. The overarching theme is that careful consideration is due when using AlphaFold2-generated models to generate testable hypotheses and structural models, rather than treating predicted models as de facto ground truth structures.

Identification of flavor peptides based on virtual screening and molecular docking from Hypsizygus marmoreuss

Hypsizygus marmoreuss is an under-explored source of flavor peptides that can enhance the flavor of NaCl or MSG, allowing products to be reformulated in line with reduction policies. This study utilized advanced techniques, including UPLC-Q-TOF MS/MS and molecular docking, to identify H. marmoreuss peptides. Sensory evaluations revealed 10 peptides with pronounced umami flavors and seven with dominantly salty tastes. VLPVPQK scored highest for umami intensity (5.2), and EGNPAHQK for salty intensity (6.2). Further investigation influenced by 0.35 % MSG or 0.35 % NaCl exposed peptides with elevated umami and salty thresholds. LDSPATPEK, VVEGEPSLK, and QKLPEKPER had umami-enhancing thresholds of 0.18, 0.18, and 0.35 mM, while LDSPATPEK and VVEGEPSLK had similar thresholds for salt (0.09 mM). Molecular docking revealed that taste receptor proteins interacted with umami peptides through hydrogen, carbon-hydrogen, alkyl, and van der Waals forces. Specific amino acids in the umami receptor T1R1 had roles in bonding with umami peptides through hydrogen and carbon-hydrogen interactions. In conclusion, molecular docking proved to be an effective and efficient method for flavor peptide screening. Further, this study demonstrated that flavor peptides from H. marmoreuss had the capacity to enhance NaCl and MSG flavours and might be useful tools for reformulation, reducing salt and MSG contents.

Integration of AlphaFold with Molecular Dynamics for Efficient Conformational Sampling of Transporter Protein NarK

Transporter proteins carry their substrate across the cell membrane by changing their conformation. Thus, conformational dynamics are crucial for transport function. However, clarifying the complete transport cycle is challenging even with the current structural biology approach. Molecular dynamics (MD) simulation is a computational approach that can provide the time-resolved conformational dynamics of transporter proteins in atomic details but suffers from a high computational cost. Here, we integrate state-of-the-art protein structure prediction AI, AlphaFold2 (AF2), with MD simulation to reduce the computational cost. Focusing on the transporter protein NarK, we first show that AF2 sampled broad conformations of NarK, including the inward-open, occluded, and outward-open states. We also applied the coevolution-informed mutation in AF2, identifying state-shifting mutations. Then, we show that MD simulations from AF2-generated outward-open conformation, which is experimentally unresolved, captured the essence of the conformational state. We also found that MD simulations from AF2-generated intermediates showed transient dynamics like a transition state connecting two conformational states. This study paves the way for efficient conformational sampling of transporter proteins.

Empowering AlphaFold2 for protein conformation selective drug discovery with AlphaFold2-RAVE

Small molecule drug design hinges on obtaining co-crystallized ligand-protein structures. Despite AlphaFold2's strides in protein native structure prediction, its focus on apo structures overlooks ligands and associated holo structures. Moreover, designing selective drugs often benefits from the targeting of diverse metastable conformations. Therefore, direct application of AlphaFold2 models in virtual screening and drug discovery remains tentative. Here, we demonstrate an AlphaFold2 based framework combined with all-atom enhanced sampling molecular dynamics and induced fit docking, named AF2RAVE-Glide, to conduct computational model based small molecule binding of metastable protein kinase conformations, initiated from protein sequences. We demonstrate the AF2RAVE-Glide workflow on three different protein kinases and their type I and II inhibitors, with special emphasis on binding of known type II kinase inhibitors which target the metastable classical DFG-out state. These states are not easy to sample from AlphaFold2. Here we demonstrate how with AF2RAVE these metastable conformations can be sampled for different kinases with high enough accuracy to enable subsequent docking of known type II kinase inhibitors with more than 50% success rates across docking calculations. We believe the protocol should be deployable for other kinases and more proteins generally.

A comprehensive exploration of the druggable conformational space of protein kinases using AI-predicted structures

Protein kinase function and interactions with drugs are controlled in part by the movement of the DFG and ɑC-Helix motifs that are related to the catalytic activity of the kinase. Small molecule ligands elicit therapeutic effects with distinct selectivity profiles and residence times that often depend on the active or inactive kinase conformation(s) they bind. Modern AI-based structural modeling methods have the potential to expand upon the limited availability of experimentally determined kinase structures in inactive states. Here, we first explored the conformational space of kinases in the PDB and models generated by AlphaFold2 (AF2) and ESMFold, two prominent AI-based protein structure prediction methods. Our investigation of AF2's ability to explore the conformational diversity of the kinome at various multiple sequence alignment (MSA) depths showed a bias within the predicted structures of kinases in DFG-in conformations, particularly those controlled by the DFG motif, based on their overabundance in the PDB. We demonstrate that predicting kinase structures using AF2 at lower MSA depths explored these alternative conformations more extensively, including identifying previously unobserved conformations for 398 kinases. Ligand enrichment analyses for 23 kinases showed that, on average, docked models distinguished between active molecules and decoys better than random (average AUC (avgAUC) of 64.58), but select models perform well (e.g., avgAUCs for PTK2 and JAK2 were 79.28 and 80.16, respectively). Further analysis explained the ligand enrichment discrepancy between low- and high-performing kinase models as binding site occlusions that would preclude docking. The overall results of our analyses suggested that, although AF2 explored previously uncharted regions of the kinase conformational space and select models exhibited enrichment scores suitable for rational drug discovery, rigorous refinement of AF2 models is likely still necessary for drug discovery campaigns.

Exploring conformational landscapes and binding mechanisms of convergent evolution for the SARS-CoV-2 spike Omicron variant complexes with the ACE2 receptor using AlphaFold2-based structural ensembles and molecular dynamics simulations

In this study, we combined AlphaFold-based approaches for atomistic modeling of multiple protein states and microsecond molecular simulations to accurately characterize conformational ensembles evolution and binding mechanisms of convergent evolution for the SARS-CoV-2 spike Omicron variants BA.1, BA.2, BA.2.75, BA.3, BA.4/BA.5 and BQ.1.1. We employed and validated several different adaptations of the AlphaFold methodology for modeling of conformational ensembles including the introduced randomized full sequence scanning for manipulation of sequence variations to systematically explore conformational dynamics of Omicron spike protein complexes with the ACE2 receptor. Microsecond atomistic molecular dynamics (MD) simulations provide a detailed characterization of the conformational landscapes and thermodynamic stability of the Omicron variant complexes. By integrating the predictions of conformational ensembles from different AlphaFold adaptations and applying statistical confidence metrics we can expand characterization of the conformational ensembles and identify functional protein conformations that determine the equilibrium dynamics for the Omicron spike complexes with the ACE2. Conformational ensembles of the Omicron RBD-ACE2 complexes obtained using AlphaFold-based approaches for modeling protein states and MD simulations are employed for accurate comparative prediction of the binding energetics revealing an excellent agreement with the experimental data. In particular, the results demonstrated that AlphaFold-generated extended conformational ensembles can produce accurate binding energies for the Omicron RBD-ACE2 complexes. The results of this study suggested complementarities and potential synergies between AlphaFold predictions of protein conformational ensembles and MD simulations showing that integrating information from both methods can potentially yield a more adequate characterization of the conformational landscapes for the Omicron RBD-ACE2 complexes. This study provides insights in the interplay between conformational dynamics and binding, showing that evolution of Omicron variants through acquisition of convergent mutational sites may leverage conformational adaptability and dynamic couplings between key binding energy hotspots to optimize ACE2 binding affinity and enable immune evasion.

Integration of a Randomized Sequence Scanning Approach in AlphaFold2 and Local Frustration Profiling of Conformational States Enable Interpretable Atomistic Characterization of Conformational Ensembles and Detection of Hidden Allosteric States in the ABL1 Protein Kinase

Despite the success of AlphaFold methods in predicting single protein structures, these methods showed intrinsic limitations in the characterization of multiple functional conformations of allosteric proteins. The recent NMR-based structural determination of the unbound ABL kinase in the active state and discovery of the inactive low-populated functional conformations that are unique for ABL kinase present an ideal challenge for the AlphaFold2 approaches. In the current study, we employ several adaptations of the AlphaFold2 methodology to predict protein conformational ensembles and allosteric states of the ABL kinase including randomized alanine sequence scanning combined with the multiple sequence alignment subsampling proposed in this study. We show that the proposed new AlphaFold2 adaptation combined with local frustration profiling of conformational states enables accurate prediction of the protein kinase structures and conformational ensembles, also offering a robust approach for interpretable characterization of the AlphaFold2 predictions and detection of hidden allosteric states. We found that the large high frustration residue clusters are uniquely characteristic of the low-populated, fully inactive ABL form and can define energetically frustrated cracking sites of conformational transitions, presenting difficult targets for AlphaFold2. The results of this study uncovered previously unappreciated fundamental connections between local frustration profiles of the functional allosteric states and the ability of AlphaFold2 methods to predict protein structural ensembles of the active and inactive states. This study showed that integration of the randomized sequence scanning adaptation of AlphaFold2 with a robust landscape-based analysis allows for interpretable atomistic predictions and characterization of protein conformational ensembles, providing a physical basis for the successes and limitations of current AlphaFold2 methods in detecting functional allosteric states that play a significant role in protein kinase regulation.

Benchmarking AlphaFold-Generated Structures of Chemokine-Chemokine Receptor Complexes

AlphaFold and AlphaFold-Multimer have become two essential tools for the modeling of unknown structures of proteins and protein complexes. In this work, we extensively benchmarked the quality of chemokine-chemokine receptor structures generated by AlphaFold-Multimer against experimentally determined structures. Our analysis considered both the global quality of the model, as well as key structural features for chemokine recognition. To study the effects of template and multiple sequence alignment parameters on the results, a new prediction pipeline called LIT-AlphaFold (https://github.com/LIT-CCM-lab/LIT-AlphaFold) was developed, allowing extensive input customization. AlphaFold-Multimer correctly predicted differences in chemokine binding orientation and accurately reproduced the unique binding orientation of the CXCL12-ACKR3 complex. Further, the predictions of the full receptor N-terminus provided insights into a putative chemokine recognition site 0.5. The accuracy of chemokine N-terminus binding mode prediction varied between complexes, but the confidence score permitted the distinguishing of residues that were very likely well positioned. Finally, we generated a high-confidence model of the unsolved CXCL12-CXCR4 complex, which agreed with experimental mutagenesis and cross-linking data.

Prediction of Conformational Ensembles and Structural Effects of State-Switching Allosteric Mutants in the Protein Kinases Using Comparative Analysis of AlphaFold2 Adaptations with Sequence Masking and Shallow Subsampling

Despite the success of AlphaFold2 approaches in predicting single protein structures, these methods showed intrinsic limitations in predicting multiple functional conformations of allosteric proteins and have been challenged to accurately capture of the effects of single point mutations that induced significant structural changes. We systematically examined several implementations of AlphaFold2 methods to predict conformational ensembles for state-switching mutants of the ABL kinase. The results revealed that a combination of randomized alanine sequence masking with shallow multiple sequence alignment subsampling can significantly expand the conformational diversity of the predicted structural ensembles and capture shifts in populations of the active and inactive ABL states. Consistent with the NMR experiments, the predicted conformational ensembles for M309L/L320I and M309L/H415P ABL mutants that perturb the regulatory spine networks featured the increased population of the fully closed inactive state. On the other hand, the predicted conformational ensembles for the G269E/M309L/T334I and M309L/L320I/T334I triple ABL mutants that share activating T334I gate-keeper substitution are dominated by the active ABL form. The proposed adaptation of AlphaFold can reproduce the experimentally observed mutation-induced redistributions in the relative populations of the active and inactive ABL states and capture the effects of regulatory mutations on allosteric structural rearrangements of the kinase domain. The ensemble-based network analysis complemented AlphaFold predictions by revealing allosteric mediating centers that often directly correspond to state-switching mutational sites or reside in their immediate local structural proximity, which may explain the global effect of regulatory mutations on structural changes between the ABL states. This study suggested that attention-based learning of long-range dependencies between sequence positions in homologous folds and deciphering patterns of allosteric interactions may further augment the predictive abilities of AlphaFold methods for modeling of alternative protein sates, conformational ensembles and mutation-induced structural transformations.

Predicting Functional Conformational Ensembles and Binding Mechanisms of Convergent Evolution for SARS-CoV-2 Spike Omicron Variants Using AlphaFold2 Sequence Scanning Adaptations and Molecular Dynamics Simulations

In this study, we combined AlphaFold-based approaches for atomistic modeling of multiple protein states and microsecond molecular simulations to accurately characterize conformational ensembles and binding mechanisms of convergent evolution for the SARS-CoV-2 Spike Omicron variants BA.1, BA.2, BA.2.75, BA.3, BA.4/BA.5 and BQ.1.1. We employed and validated several different adaptations of the AlphaFold methodology for modeling of conformational ensembles including the introduced randomized full sequence scanning for manipulation of sequence variations to systematically explore conformational dynamics of Omicron Spike protein complexes with the ACE2 receptor. Microsecond atomistic molecular dynamic simulations provide a detailed characterization of the conformational landscapes and thermodynamic stability of the Omicron variant complexes. By integrating the predictions of conformational ensembles from different AlphaFold adaptations and applying statistical confidence metrics we can expand characterization of the conformational ensembles and identify functional protein conformations that determine the equilibrium dynamics for the Omicron Spike complexes with the ACE2. Conformational ensembles of the Omicron RBD-ACE2 complexes obtained using AlphaFold-based approaches for modeling protein states and molecular dynamics simulations are employed for accurate comparative prediction of the binding energetics revealing an excellent agreement with the experimental data. In particular, the results demonstrated that AlphaFold-generated extended conformational ensembles can produce accurate binding energies for the Omicron RBD-ACE2 complexes. The results of this study suggested complementarities and potential synergies between AlphaFold predictions of protein conformational ensembles and molecular dynamics simulations showing that integrating information from both methods can potentially yield a more adequate characterization of the conformational landscapes for the Omicron RBD-ACE2 complexes. This study provides insights in the interplay between conformational dynamics and binding, showing that evolution of Omicron variants through acquisition of convergent mutational sites may leverage conformational adaptability and dynamic couplings between key binding energy hotspots to optimize ACE2 binding affinity and enable immune evasion.

AlphaFold2 Predicts Whether Proteins Interact Amidst Confounding Structural Compatibility

Predicting whether two proteins physically interact is one of the holy grails of computational biology, galvanized by rapid advancements in deep learning. AlphaFold2, although not developed with this goal, is promising in this respect. Here, I test the prediction capability of AlphaFold2 on a very challenging data set, where proteins are structurally compatible, even when they do not interact. AlphaFold2 achieves high discrimination between interacting and non-interacting proteins, and the cases of misclassifications can either be rescued by revisiting the input sequences or can suggest false positives and negatives in the data set. AlphaFold2 is thus not impaired by the compatibility between protein structures and has the potential to be applied on a large scale.

Interpretable Atomistic Prediction and Functional Analysis of Conformational Ensembles and Allosteric States in Protein Kinases Using AlphaFold2 Adaptation with Randomized Sequence Scanning and Local Frustration Profiling

The groundbreaking achievements of AlphaFold2 (AF2) approaches in protein structure modeling marked a transformative era in structural biology. Despite the success of AF2 tools in predicting single protein structures, these methods showed intrinsic limitations in predicting multiple functional conformations of allosteric proteins and fold-switching systems. The recent NMR-based structural determination of the unbound ABL kinase in the active state and two inactive low-populated functional conformations that are unique for ABL kinase presents an ideal challenge for AF2 approaches. In the current study we employ several implementations of AF2 methods to predict protein conformational ensembles and allosteric states of the ABL kinase including (a) multiple sequence alignments (MSA) subsampling approach; (b) SPEACH_AF approach in which alanine scanning is performed on generated MSAs; and (c) introduced in this study randomized full sequence mutational scanning for manipulation of sequence variations combined with the MSA subsampling. We show that the proposed AF2 adaptation combined with local frustration mapping of conformational states enable accurate prediction of the ABL active and intermediate structures and conformational ensembles, also offering a robust approach for interpretable characterization of the AF2 predictions and limitations in detecting hidden allosteric states. We found that the large high frustration residue clusters are uniquely characteristic of the low-populated, fully inactive ABL form and can define energetically frustrated cracking sites of conformational transitions, presenting difficult targets for AF2 methods. This study uncovered previously unappreciated, fundamental connections between distinct patterns of local frustration in functional kinase states and AF2 successes/limitations in detecting low-populated frustrated conformations, providing a better understanding of benefits and limitations of current AF2-based adaptations in modeling of conformational ensembles.

Computational Workflow for Refining AlphaFold Models in Drug Design Using Kinetic and Thermodynamic Binding Calculations: A Case Study for the Unresolved Inactive Human Adenosine A3 Receptor

A structure-based drug design pipeline that considers both thermodynamic and kinetic binding data of ligands against a receptor will enable the computational design of improved drug molecules. For unresolved GPCR-ligand complexes, a workflow that can apply both thermodynamic and kinetic binding data in combination with alpha-fold (AF)-derived or other homology models and experimentally resolved binding modes of relevant ligands in GPCR-homologs needs to be tested. Here, as test case, we studied a congeneric set of ligands that bind to a structurally unresolved G protein-coupled receptor (GPCR), the inactive human adenosine A3 receptor (hA3R). We tested three available homology models from which two have been generated from experimental structures of hA1R or hA2AR and one model was a multistate alphafold 2 (AF2)-derived model. We applied alchemical calculations with thermodynamic integration coupled with molecular dynamics (TI/MD) simulations to calculate the experimental relative binding free energies and residence time (τ)-random accelerated MD (τ-RAMD) simulations to calculate the relative residence times (RTs) for antagonists. While the TI/MD calculations produced, for the three homology models, good Pearson correlation coefficients, correspondingly, r = 0.74, 0.62, and 0.67 and mean unsigned error (mue) values of 0.94, 1.31, and 0.81 kcal mol-1, the τ-RAMD method showed r = 0.92 and 0.52 for the first two models but failed to produce accurate results for the multistate AF2-derived model. With subsequent optimization of the AF2-derived model by reorientation of the side chain of R1735.34 located in the extracellular loop 2 (EL2) that blocked ligand's unbinding, the computational model showed r = 0.84 for kinetic data and improved performance for thermodynamic data (r = 0.81, mue = 0.56 kcal mol-1). Overall, after refining the multistate AF2 model with physics-based tools, we were able to show a strong correlation between predicted and experimental ligand relative residence times and affinities, achieving a level of accuracy comparable to an experimental structure. The computational workflow used can be applied to other receptors, helping to rank candidate drugs in a congeneric series and enabling the prioritization of leads with stronger binding affinities and longer residence times.

Computational drug development for membrane protein targets

The application of computational biology in drug development for membrane protein targets has experienced a boost from recent developments in deep learning-driven structure prediction, increased speed and resolution of structure elucidation, machine learning structure-based design and the evaluation of big data. Recent protein structure predictions based on machine learning tools have delivered surprisingly reliable results for water-soluble and membrane proteins but have limitations for development of drugs that target membrane proteins. Structural transitions of membrane proteins have a central role during transmembrane signaling and are often influenced by therapeutic compounds. Resolving the structural and functional basis of dynamic transmembrane signaling networks, especially within the native membrane or cellular environment, remains a central challenge for drug development. Tackling this challenge will require an interplay between experimental and computational tools, such as super-resolution optical microscopy for quantification of the molecular interactions of cellular signaling networks and their modulation by potential drugs, cryo-electron microscopy for determination of the structural transitions of proteins in native cell membranes and entire cells, and computational tools for data analysis and prediction of the structure and function of cellular signaling networks, as well as generation of promising drug candidates.

Rosetta Energy Analysis of AlphaFold2 models: Point Mutations and Conformational Ensembles

There has been an explosive growth in the applications of AlphaFold2, and other structure prediction platforms, to accurately predict protein structures from a multiple sequence alignment (MSA) for downstream structural analysis. However, two outstanding questions persist in the field regarding the robustness of AlphaFold2 predictions of the consequences of point mutations and the completeness of its prediction of protein conformational ensembles. We combined our previously developed method SPEACH_AF with model relaxation and energetic analysis with Rosetta to address these questions. SPEACH_AF introduces residue substitutions across the MSA and not just within the input sequence. With respect to conformational ensembles, we combined SPEACH_AF and a new MSA subsampling method, AF_cluster, and for a benchmarked set of proteins, we found that the energetics of the conformational ensembles generated by AlphaFold2 correspond to those of experimental structures and explored by standard molecular dynamic methods. With respect to point mutations, we compared the structural and energetic consequences of having the mutation(s) in the input sequence versus in the whole MSA (SPEACH_AF). Both methods yielded models different from the wild-type sequence, with more robust changes when the mutation(s) were in the whole MSA. While our findings demonstrate the robustness of AlphaFold2 in analyzing point mutations and exploring conformational ensembles, they highlight the need for multi parameter structural and energetic analyses of these models to generate experimentally testable hypotheses.

Accelerated Molecular Dynamics and AlphaFold Uncover a Missing Conformational State of Transporter Protein OxlT

Transporter proteins change their conformations to carry their substrate across the cell membrane. The conformational dynamics is vital to understanding the transport function. We have studied the oxalate transporter (OxlT), an oxalate:formate antiporter from Oxalobacter formigenes, significant in avoiding kidney stone formation. The atomic structure of OxlT has been recently solved in the outward-open and occluded states. However, the inward-open conformation is still missing, hindering a complete understanding of the transporter. Here, we performed a Gaussian accelerated molecular dynamics simulation to sample the extensive conformational space of OxlT and successfully predicted the inward-open conformation where cytoplasmic substrate formate binding was preferred over oxalate binding. We also identified critical interactions for the inward-open conformation. The results were complemented by an AlphaFold2 structure prediction. Although AlphaFold2 solely predicted OxlT in the outward-open conformation, mutation of the identified critical residues made it partly predict the inward-open conformation, identifying possible state-shifting mutations.

Representing structures of the multiple conformational states of proteins

Biomolecules exhibit dynamic behavior that single-state models of their structures cannot fully capture. We review some recent advances for investigating multiple conformations of biomolecules, including experimental methods, molecular dynamics simulations, and machine learning. We also address the challenges associated with representing single- and multiple-state models in data archives, with a particular focus on NMR structures. Establishing standardized representations and annotations will facilitate effective communication and understanding of these complex models to the broader scientific community.

Breaking the conformational ensemble barrier: Ensemble structure modeling challenges in CASP15

For the first time, the 2022 CASP (Critical Assessment of Structure Prediction) community experiment included a section on computing multiple conformations for protein and RNA structures. There was full or partial success in reproducing the ensembles for four of the nine targets, an encouraging result. For protein structures, enhanced sampling with variations of the AlphaFold2 deep learning method was by far the most effective approach. One substantial conformational change caused by a single mutation across a complex interface was accurately reproduced. In two other assembly modeling cases, methods succeeded in sampling conformations near to the experimental ones even though environmental factors were not included in the calculations. An experimentally derived flexibility ensemble allowed a single accurate RNA structure model to be identified. Difficulties included how to handle sparse or low-resolution experimental data and the current lack of effective methods for modeling RNA/protein complexes. However, these and other obstacles appear addressable.

From Byte to Bench to Bedside: Molecular Dynamics Simulations and Drug Discovery

Molecular dynamics (MD) simulations and computer-aided drug design (CADD) have advanced substantially over the past two decades, thanks to continuous computer hardware and software improvements. Given these advancements, MD simulations are poised to become even more powerful tools for investigating the dynamic interactions between potential small-molecule drugs and their target proteins, with significant implications for pharmacological research.

Is the Functional Response of a Receptor Determined by the Thermodynamics of Ligand Binding?

For an effective drug, strong binding to the target protein is a prerequisite, but it is not enough. To produce a particular functional response, drugs need to either block the proteins' functions or modulate their activities by changing their conformational equilibrium. The binding free energy of a compound to its target is routinely calculated, but the timescales for the protein conformational changes are prohibitively long to be efficiently modeled via physics-based simulations. Thermodynamic principles suggest that the binding free energies of the ligands with different receptor conformations may infer their efficacy. However, this hypothesis has not been thoroughly validated. We present an actionable protocol and a comprehensive study to show that binding thermodynamics provides a strong predictor of the efficacy of a ligand. We apply the absolute binding free energy perturbation method to ligands bound to active and inactive states of eight G protein-coupled receptors and a nuclear receptor and then compare the resulting binding free energies. We find that carefully designed restraints are often necessary to efficiently model the corresponding conformational ensembles for each state. Our method achieves unprecedented performance in classifying ligands as agonists or antagonists across the various investigated receptors, all of which are important drug targets.

Interplay of thermodynamics and evolution within the ternary ligand-GPCR-G protein complex

Recent studies on G protein-coupled receptors (GPCRs) dynamics report that GPCRs adopt a wide range of conformations that coexist in equilibrium, with the apo state of a GPCR having a high entropy. The formation of a ligand-GPCR-transducer complex comes with a reduction of conformational space and therefore with an entropic cost. We hypothesize that the availability of binding partners, their binding affinity and the rigidity of the respective binding sites are reflected in a distinct degree of sequence conservation to balance the energetic cost of intra- and extracellular binding events. Here, we outline the current findings in delineating the conformational space and include sequential conservation of many-to-many ligand-receptor systems to discuss the entropic cost that comes with GPCR signal transduction.

Investigating the conformational landscape of AlphaFold2-predicted protein kinase structures

Summary

Protein kinases are a family of signaling proteins, crucial for maintaining cellular homeostasis. When dysregulated, kinases drive the pathogenesis of several diseases, and are thus one of the largest target categories for drug discovery. Kinase activity is tightly controlled by switching through several active and inactive conformations in their catalytic domain. Kinase inhibitors have been designed to engage kinases in specific conformational states, where each conformation presents a unique physico-chemical environment for therapeutic intervention. Thus, modeling kinases across conformations can enable the design of novel and optimally selective kinase drugs. Due to the recent success of AlphaFold2 in accurately predicting the 3D structure of proteins based on sequence, we investigated the conformational landscape of protein kinases as modeled by AlphaFold2. We observed that AlphaFold2 is able to model several kinase conformations across the kinome, however, certain conformations are only observed in specific kinase families. Furthermore, we show that the per residue predicted local distance difference test can capture information describing structural flexibility of kinases. Finally, we evaluated the docking performance of AlphaFold2 kinase structures for enriching known ligands. Taken together, we see an opportunity to leverage AlphaFold2 models for structure-based drug discovery against kinases across several pharmacologically relevant conformational states.

Availability and implementation

All code used in the analysis is freely available at https://github.com/Harmonic-Discovery/AF2-kinase-conformational-landscape.

Modeling conformational states of proteins with AlphaFold

Many proteins exert their function by switching among different structures. Knowing the conformational ensembles affiliated with these states is critical to elucidate key mechanistic aspects that govern protein function. While experimental determination efforts are still bottlenecked by cost, time, and technical challenges, the machine-learning technology AlphaFold showed near experimental accuracy in predicting the three-dimensional structure of monomeric proteins. However, an AlphaFold ensemble of models usually represents a single conformational state with minimal structural heterogeneity. Consequently, several pipelines have been proposed to either expand the structural breadth of an ensemble or bias the prediction toward a desired conformational state. Here, we analyze how those pipelines work, what they can and cannot predict, and future directions.