Discovery and synthesis of hydroxy-L-proline based blockers of the neutral amino acid transporters SLC1A4 (ASCT1) and SLC1A5 (ASCT2).

The conformationally restricted heterocycle hydroxy-ʟ-proline is a versatile scaffold for the synthesis of diverse multi-functionalized pyrrolidines for probing the ligand binding sites of biological targets. With the goal to develop new inhibitors of the widely expressed amino acid transporters SLC1A4 and SLC1A5 (also known as ASCT1 and ASCT2), we synthesized and functionally screened a series of hydroxy-ʟ-proline derivatives or 'prolinols' using electrophysiological and radio-labeled uptake assays on amino acid transporters from the SLC1, SLC7, and SLC38 solute carrier families. We identified a number of synthetic prolinols that act as selective high-affinity inhibitors of the SLC1 functional subfamily comprising the neutral amino acid transporters SLC1A4 and SLC1A5. The active and inactive prolinols were computationally docked into a threaded homology model and analyzed with respect to predicted molecular orientation and observed pharmacological activity. The series of hydroxy-L-proline derivatives identified here represents a new class of potential agents to pharmacologically modulate SLC1A4 and SLC1A5, amino acid exchangers that play important roles in a wide range of physiological and pathophysiological processes.

Avidity observed between a bivalent inhibitor and an enzyme monomer with a single active site

Although myriad protein–protein interactions in nature use polyvalent binding, in which multiple ligands on one entity bind to multiple receptors on another, to date an affinity advantage of polyvalent binding has been demonstrated experimentally only in cases where the target receptor molecules are clustered prior to complex formation. Here, we demonstrate cooperativity in binding affinity (i.e., avidity) for a protein complex in which an engineered dimer of the amyloid precursor protein inhibitor (APPI), possessing two fully functional inhibitory loops, interacts with mesotrypsin, a soluble monomeric protein that does not self-associate or cluster spontaneously. We found that each inhibitory loop of the purified APPI homodimer was over three-fold more potent than the corresponding loop in the monovalent APPI inhibitor. This observation is consistent with a suggested mechanism whereby the two APPI loops in the homodimer simultaneously and reversibly bind two corresponding mesotrypsin monomers to mediate mesotrypsin dimerization. We propose a simple model for such dimerization that quantitatively explains the observed cooperativity in binding affinity. Binding cooperativity in this system reveals that the valency of ligands may affect avidity in protein–protein interactions including those of targets that are not surface-anchored and do not self-associate spontaneously. In this scenario, avidity may be explained by the enhanced concentration of ligand binding sites in proximity to the monomeric target, which may favor rebinding of the multiple ligand binding sites with the receptor molecules upon dissociation of the protein complex.

De novo protein fold families expand the designable ligand binding site space

A major challenge in designing proteins de novo to bind user-defined ligands with high affinity is finding backbones structures into which a new binding site geometry can be engineered with high precision. Recent advances in methods to generate protein fold families de novo have expanded the space of accessible protein structures, but it is not clear to what extend de novo proteins with diverse geometries also expand the space of designable ligand binding functions. We constructed a library of 25,806 high-quality ligand binding sites and developed a fast protocol to place (“match”) these binding sites into both naturally occurring and de novo protein families with two fold topologies: Rossman and NTF2. Each matching step involves engineering new binding site residues into each protein “scaffold”, which is distinct from the problem of comparing already existing binding pockets. 5,896 and 7,475 binding sites could be matched to the Rossmann and NTF2 fold families, respectively. De novo designed Rossman and NTF2 protein families can support 1,791 and 678 binding sites that cannot be matched to naturally existing structures with the same topologies, respectively. While the number of protein residues in ligand binding sites is the major determinant of matching success, ligand size and primary sequence separation of binding site residues also play important roles. The number of matched binding sites are power law functions of the number of members in a fold family. Our results suggest that de novo sampling of geometric variations on diverse fold topologies can significantly expand the space of designable ligand binding sites for a wealth of possible new protein functions.

Characterizing disease-associated human proteins without available protein structures or homologues

Mutations in human proteins lead to diseases. The structure of these proteins can help understand the mechanism of such diseases and develop therapeutics against them. With improved deep learning techniques such as RoseTTAFold and AlphaFold, we can predict the structure of these proteins even in the absence of structural homologues. We modeled and extracted the domains from 553 disease-associated human proteins. We noticed that the model quality was higher and the RMSD lower between AlphaFold and RoseTTAFold models for domains that could be assigned to CATH families as compared to those which could be assigned to Pfam families of unknown structure or could not be assigned to either. We predicted ligand-binding sites, protein-protein interfaces, conserved residues and destabilising effects caused by residue mutations in these predicted structures. We then explored whether the disease-associated mutations were in the proximity of these predicted functional sites or if they destabilized the protein structure based on ddG calculations. We could explain 80% of these disease-associated mutations based on proximity to functional sites or structural destabilization. Usage of models from the two state-of-the-art techniques provide better confidence in our predictions, and we explain 93 additional mutations based on RoseTTAFold models which could not be explained based solely on AlphaFold models.

Identification of ligand binding sites in intrinsically disordered proteins with a differential binding score

Screening ligands directly binding to an ensemble of intrinsically disordered proteins (IDP) to discover potential hits or leads for new drugs is an emerging but challenging area as IDPs lack well-defined and ordered 3D-protein structures. To explore a new IDP-based rational drug discovery strategy, a differential binding score (DIBS) is defined. The basis of DIBS is to quantitatively determine the binding preference of a ligand to an ensemble of conformations specified by IDP versus such preferences to an ensemble of random coil conformations of the same protein. Ensemble docking procedures performed on repeated sampling of conformations, and the results tested for statistical significance determine the preferential ligand binding sites of the IDP. The results of this approach closely reproduce the experimental data from recent literature on the binding of the ligand epigallocatechin gallate (EGCG) to the intrinsically disordered N-terminal domain of the tumor suppressor p53. Combining established approaches in developing a new method to screen ligands against IDPs could be valuable as a screening tool for IDP-based drug discovery.

Identification of Iron-Sulfur (Fe-S) and Zn-binding Sites Within Proteomes Predicted by DeepMind′s AlphaFold2 Program Dramatically Expands the Metalloproteome

DeepMind′s AlphaFold2 software has ushered in a revolution in high quality, 3D protein structure prediction. In very recent work by the DeepMind team, structure predictions have been made for entire proteomes of twenty-one organisms, with >360,000 structures made available for download. Here we show that thousands of novel binding sites for iron-sulfur (Fe-S) clusters and zinc ions can be identified within these predicted structures by exhaustive enumeration of all potential ligand-binding orientations. We demonstrate that AlphaFold2 routinely makes highly specific predictions of ligand binding sites: for example, binding sites that are comprised exclusively of four cysteine sidechains fall into three clusters, representing binding sites for 4Fe-4S clusters, 2Fe-2S clusters, or individual Zn ions. We show further: (a) that the majority of known Fe-S cluster and Zn-binding sites documented in UniProt are recovered by the AlphaFold2 structures, (b) that there are occasional disputes between AlphaFold2 and UniProt with AlphaFold2 predicting highly plausible alternative binding sites, (c) that the Fe-S cluster binding sites that we identify in E. coli agree well with previous bioinformatics predictions, (d) that cysteines predicted here to be part of Fe-S cluster or Zn-binding sites show little overlap with those shown via chemoproteomics techniques to be highly reactive, and (e) that AlphaFold2 occasionally appears to build erroneous disulfide bonds between cysteines that should instead coordinate a ligand. These results suggest that AlphaFold2 could be an important tool for the functional annotation of proteomes, and the methodology presented here is likely to be useful for predicting other ligand-binding sites.

Locating ligand binding sites in G-protein coupled receptors using combined information from docking and sequence conservation

GPCRs (G-protein coupled receptors) are the largest family of drug targets and share a conserved structure. Binding sites are unknown for many important GPCR ligands due to the difficulties of GPCR recombinant expression, biochemistry, and crystallography. We describe our approach, ConDockSite, for predicting ligand binding sites in class A GPCRs using combined information from surface conservation and docking, starting from crystal structures or homology models. We demonstrate the effectiveness of ConDockSite on crystallized class A GPCRs such as the beta2 adrenergic and A2A adenosine receptors. We also demonstrate that ConDockSite successfully predicts ligand binding sites from high-quality homology models. Finally, we apply ConDockSite to predict the ligand binding sites on a structurally uncharacterized GPCR, GPER, the G-protein coupled estrogen receptor. Most of the sites predicted by ConDockSite match those found in other independent modeling studies. ConDockSite predicts that four ligands bind to a common location on GPER at a site deep in the receptor cleft. Incorporating sequence conservation information in ConDockSite overcomes errors introduced from physics-based scoring functions and homology modeling.

Compatibility of the Ligand Binding Sites in the Spike Glycoprotein of COVID-19 with those in the Aminopeptidase and the Caveolins 1, 2 Proteins

A novel severe viral pneumonia emerged in Wuhan city, China, in December 2019. The spike glycoprotein of the SARS-CoV-2 plays a crucial role in the viral entry to the host cell and eliciting a strong response for antibody-mediated neutralization in mice. Caveolins 1,2 are scaffolding proteins dovetailed as a co-stimulatory signal essential for T-cell receptor and activation. Aminopeptidase is a membrane protein acting as a receptor for human coronavirus within the S1 subunit of the spike glycoprotein. Vaccines for COVID-19 have become a priority for predisposition against the outbreak, so that our study aimed to find interaction sites between SP of SARS-CoV-2 and CAV1, CAV2, and AMPN. Methods: Amino acids motif search was employed to predict the possible CAV1, CAV2, and AMPN related interaction domains in the SARS-CoV-2 SP In silico analysis. Results: Interactions between proteins revealed 5 and16 residues. ZN ligand binding site is matched between AMPN and SARS- CoV-2 SP. HLA-A*74:01 allele is the best CTL epitope for SP. We identified seven B-cell epitopes specifically for SARS-CoV-2 SP. Conclusions: SARS-CoV-2 SP binding sites might be compatible with AMPN ligand binding sites. The limit score was detected for ligand binding sites of CAV1 and CAV2. Our findings might be critical for the further substantial study of vaccine production strategy.

PUResNet: prediction of protein-ligand binding sites using deep residual neural network

Background Predicting protein-ligand binding sites is a fundamental step in understanding the functional characteristics of proteins, which plays a vital role in elucidating different biological functions and is a crucial step in drug discovery. A protein exhibits its true nature after binding to its interacting molecule known as a ligand that binds only in the favorable binding site of the protein structure. Different computational methods exploiting the features of proteins have been developed to identify the binding sites in the protein structure, but none seems to provide promising results, and therefore, further investigation is required. Results In this study, we present a deep learning model PUResNet and a novel data cleaning process based on structural similarity for predicting protein-ligand binding sites. From the whole scPDB (an annotated database of druggable binding sites extracted from the Protein DataBank) database, 5020 protein structures were selected to address this problem, which were used to train PUResNet. With this, we achieved better and justifiable performance than the existing methods while evaluating two independent sets using distance, volume and proportion metrics.