The G protein database, GproteinDb

Abstract

Two-thirds of signaling substances, several sensory stimuli and over one-third of drugs act via receptors coupling to G proteins. Here, we present an online platform for G protein research with reference data and tools for analysis, visualization and design of scientific studies across disciplines and areas. This platform may help translate new pharmacological, structural and genomic data into insights on G protein signaling vital for human physiology and medicine. The G protein database is accessible at https://gproteindb.org.

Figure 1.

G protein couplings and selectivity. (A) G protein couplings encompass Guide to Pharmacology (GtP) primary/secondary transducers (22) and log(E_max/EC₅₀), pEC₅₀ and E_max values from profiling studies (3–5) and a means thereof. The confidence and coverage of quantitative couplings can be adjusted based on counts of supporting datasets (default 2) and standard deviations from basal signal (default 1.4). (B) G protein selectivity profiles can be intersected in a Venn diagram counting the receptors in each class. (C), G protein selectivity trees map couplings to trees classifying receptors by class, ligand type and receptor families sharing endogenous ligand(s). (B, C) Dashed arrows illustrate paths to obtain coupling-based selection of receptor sets for further study.

Figure 2.

Structures and interface interactions. (A) β₂-adrenoceptor-Gs structure (35), its refined structure and β₂-G_i1 and -G_o models. (B) Snakeplot mapping residues topological segments and highlighting GPCR–G protein interface positions (red residues). (C) Interface interactions shown in % and grayscale across class A GPCR–G_i/o family structures (interaction presence/absences in each structure is shown upon mouse hover). GPCR and G protein residues are indexed with generic residue numbers (25,26) to compare of structurally corresponding positions and color-coded to visualize properties or interaction types. A structure-based interface fingerprint can be profiled against GPCR lacking structures (Supplementary Figure S2).

Figure 3.

Sequence data in the online G protein research platform. (A) Snakeplot visualization of amino acid topological positions and properties. All amino acids are colored by property: polarity for most but backbone modifying ability for proline and glycine and disulphide-formation for cysteine. (B) Sequence alignment of the receptor-interacting helix 5 of all human 16 G_α proteins along with conservation measures for amino acids and residue groups with similar property and size, and numeric amino acid descriptors, ‘z-scales’ (29) (from https://gproteindb.org/alignment/gproteinselection). (C) Protein sequence mapping onto segments by secondary structures (helices, β-sheet and loops). (D) Protein-specific and common residue numbers (25) for G_αs, G_αi1, G_αq and G_α12 (from https://gproteindb.org/residue/residuetable_gprot). These number tables can also be downloaded in Excel format or retrieved programmatically via a RESTFUL-API web service to integrate the numbering in any dataset and analysis method. (A, B) Taken from the ‘G protein page’ (https://gproteindb.org/signprot) for G_αs. (A–C), Common residue numbers (25) can be shown by mouse hover.