GPCRdb in 2021: integrating GPCR sequence, structure and function.
Kooistra AJ, Mordalski S, Pándy-Szekeres G, Esguerra M, Mamyrbekov A, Munk C, Keserű GM, Gloriam DE. · 2021
Read GPCRdb in 2021: integrating GPCR sequence, structure and function. by Kooistra AJ, Mordalski S, Pándy-Szekeres G, Esguerra M, Mamyrbekov A, Munk C, Keserű GM, Gloriam DE. (2021) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc7778909.txt.
Introduction
Kooistra and co-authors update GPCRdb, a database for a major drug target family. G protein-coupled receptors form the largest family of membrane proteins. They mediate the action of one-third of medicines. The database serves more than 4,000 researchers each month with reference data and design tools.
It holds all human non-olfactory receptors, plus more than 27,000 counterparts in other species. Drug records top 2,000 agents, with nearly 200,000 ligands and their activity data. All published receptor structures are listed and refined. Models cover each human receptor in three states. New tools help spot key residues and design mutations for lab work.
Catalog record
| CREATOR | Kooistra AJ, Mordalski S, Pándy-Szekeres G, Esguerra M, Mamyrbekov A, Munk C, Keserű GM, Gloriam DE. |
|---|---|
| YEAR | 2021 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
GPCRdb in 2021: integrating GPCR sequence, structure and function By Albert J Kooistra, Stefan Mordalski, Gáspár Pándy-Szekeres, Mauricio Esguerra, Alibek Mamyrbekov, Christian Munk, György M Keserű, David E Gloriam Published in: Nucleic Acids Research DOI: 10.1093/nar/gkaa1080 ABSTRACT Abstract G protein-coupled receptors (GPCRs) form both the largest family of membrane proteins and drug targets, mediating the action of one-third of medicines. The GPCR database, GPCRdb serves >4 000 researchers every month and offers reference data, analysis of own or literature data, experiment design and dissemination of published datasets. Here, we describe new and updated GPCRdb resources with a particular focus on integration of sequence, structure and function. GPCRdb contains all human non-olfactory GPCRs (and >27 000 orthologs), G-proteins and arrestins. It includes over 2 000 drug and in-trial agents and nearly 200 000 ligands with activity and availability data. GPCRdb annotates all published GPCR structures (updated monthly), which are also offered in a refined version (with re-modeled missing/distorted regions and reverted mutations) and provides structure models of all human non-olfactory receptors in inactive, intermediate and active states. Mutagenesis data in the GPCRdb spans natural genetic variants, GPCR-G protein interfaces, ligand sites and thermostabilising mutations. A new sequence signature tool for identification of functional residue determinants has been added and two data driven tools to design ligand site mutations and constructs for structure determination have been updated extending their coverage of receptors and modifications. The GPCRdb is available at https://gpcrdb.org. INTRODUCTION G protein-coupled receptors (GPCRs) account for 4% (799 (1) / 20 595 (2)) of human genes and mediate the actions of two-thirds (342/515) of hormones and neurotransmitters (3). Their abundant modulation of human physiology is mirrored in medicine where 34% of marketed drugs act on GPCRs (4). Hence, GPCRs are studied by very large basic receptor research and drug discovery communities. The GPCR database, GPCRdb currently serves >4 000 researchers every month with multi-disciplinary reference data, data visualisation/analysis, tools to design experiments and dissemination of datasets upon publication. GPCRdb is compliant with the FAIR principles (5), provides all its code as open source, gives full free access to all its resources and publishes open access whenever possible. GPCR sequence, structure and function data is increasing rapidly from advances in e.g. cryo-electron microscopy (cryo-EM) (6), deep mutational scanning (7), genome sequencing (8) and signal protein profiling (9,10). The curation and integration of such large datasets require extensive work—typically beyond the capacity in the individual research group or study. In this article, we describe recent updates of GPCRdb infrastructure—section by section—that integrate and infer GPCR sequence, structure and function data at the receptor and residue levels. We hope that this will aid the community to reach a better understanding of receptor mechanisms and how they may be exploited for drug discovery. MATERIALS AND METHODS Structure annotation and sequence alignment New GPCR structures are imported to a development database instance upon release in the Protein Data Bank (PDB) (11) along with draft data on e.g. ligands, auxiliary fusion and signalling proteins, and preferred chain. These data are manually inspected, and further annotations are added before (now monthly) release in GPCRdb to allow researchers to select relevant receptor structures. The first and last residue positions are annotated in each receptor segment—the transmembrane helices 1–7 (TM1–7), helix 8 (H8) and intra-/extra-cellular loops 1–2 (ICL/ECL1–2)—and used to define their length in the sequence alignments and homology models in GPCRdb. For all human GPCRs lacking a structure we infer each segment's start and end from receptors with either similar sequence motives/lengths or overall sequence. When a new structure is published for an already annotated receptor, the structural annotation is updated, if necessary. For example, the first structure with a native ICL3 sequence can correct the cytosolic ends of the flanking TM5 and TM6 previously affected by the use of a protein fusion facilitating crystallography (12). The sequence annotation of species orthologs is inferred from the human ortholog based on an automated sequence alignment. Generic residue numbering The generic residue numbering uses a structural investigation to confirm or correct the receptor residue that corresponds to the reference position (designated the number 50) in each of TM1–7 using the sequence-based numbering scheme for the given GPCR class (A: Ballesteros-Weinstein (13), B1: Wootten (14), C: Pin (15), D1 (16) and F: Wang (17)) and in H8 and ICL/ECL1–2 (first defined by GPCRdb) (18,19). To ensure that the residue numbers are generic also when one of two compared receptor structures contains a helix bulge or constriction that cause an offset in a sequence-based numbering of the following residues, the observed structural single-position gap is represented by a corresponding gap in a structure-based sequence alignment (18). In contrast to sequence-based alignments, which align residues by their position in the gene product and therefore are suited for evolutionary studies, the unique alignment and generic numbering of residues that are equivalent in their structural position are designed for residue-structure-function correlation. Extending amino acid identity with property conservation in sequence alignments and signatures Amino acid properties were defined based on polarity, helical propensity and size. Subsequently, we defined a distance for each property from the receptor backbone by counting the bonds from each residue's Cα atom to the sidechain atom with the given property (for multiple atoms with the same property we used the most terminal). Finally, residue property conservation was measured in percent for each of the in total 56 groups of or single amino acids that share a property and differ by none or one in their backbone distance. Furthermore, we implemented numeric amino acid descriptors—specifically frequently used ‘z-scales’ derived from principle component analysis of 10 experimental and 16 calculated properties (20). For each of the residue positions and five z-scales, the average and standard deviation value is calculated and the separation between the conserved and non-conserved residues is evaluated with a two-tailed t-test. Structure statistics We implemented a ‘Structure model statistics’ page containing root-mean-square deviation (RMSD) values of the latest GPCRdb model (based on another receptor template) to the first experimental structure of the given receptor and state as well as a link to documentation and scripts used for the calculations. For this publication, we also downloaded models from the latest release (2019-12-20) of the RosettaGPCR’s Github page (https://github.com/benderb1/rosettagpcr) and database site (http://www.meilerlab.org/index.php/gpcrmodeldb) which have identical models. Thirty seven receptors had attained their first experimental inactive state structures since the launch of the GPCRdb homology modelling pipeline and could therefore be evaluated. However, 16 models could not be used for validation, as RosettaGPCR uses alignment.fasta files (from https://github.com/benderb1/rosettagpcr) that include the target structure among its templates. For the remaining 21 GPCRdb and corresponding RosettaGPCR models, RMSD values were calculated to the experimental structure for the following receptor segments and atoms: 7TM all atoms, 7TM backbone atoms, ICL1, ICL2, ECL1, ECL2, ECL3 and H8 (Supplementary Spreadsheet 1). To ensure a comparable measure, only atoms present in the experimental target structure and in both models were included in the RMSD calculation. The superposition was based on the 7TM backbone atoms defined by GPCRdb sequence alignments. Ligand database update All GPCR ligands and their biological activities were imported from ChEMBL version 27 (21) using the ChEMBL web services (22) which were queried using ChEMBL identifiers. These