Extended Data Fig. 6: Receptor isoforms and drug targets.
From: Combinatorial expression of GPCR isoforms affects signalling and drug responses
a, Number of receptors categorized by the number of tissue-expression signatures for all GPCRs (grey) and for those belonging to the 111 targets of 474 FDA-approved drugs (in green). b, The number of approved drugs varies greatly for receptors with the same number of tissue-expression signatures. Count plot of the 111 GPCRs that are targeted by 474 FDA-approved drugs. The radius of each circle indicates the number of GPCR targets for each combination of ‘number of tissue-expression signatures’ (x-axis) versus ‘number of approved drugs’ (y-axis). c, Disease-related phenotypes that can be exclusively linked to non-reference receptor isoforms were extracted from Gene ATLAS (see Methods and Supplementary Table 3). Non-reference isoform sequences for the N-terminal segment of the chemokine receptor CXCR3 and C-terminal segment of the metabotropic glutamate receptor 8 (GRM8) are represented in blue; the reference isoform is in grey. Polymorphisms are indicated in bold, together with their associated phenotypes in light grey boxes. d, Filtering based on structural and expression considerations identifies non-reference isoforms with changes in extracellular structure and tissue distribution with respect to the reference isoform, potentially allowing the development of ligands that specifically target them. e, Alignment of the N-terminal segment of the reference and non-reference GPR35 isoforms. Varying regions in isoform 1 are highlighted in blue. f, pEC50 and Emax values for the coupling of Gα13 and β-arrestin-2 to the reference GPR35 isoform and isoform 1 in response to lodoxamide, pamoic acid and zaprinast. (n = 3 biologically independent samples; error values correspond to s.e.m.) g, BRET signals were monitored after treatment of HEK293T cells with varying concentrations of pamoic acid and zaprinast using a GPR35–Gα13 SPASM sensor (left panel; n = 3 biologically independent samples) or eYFP-tagged GPR35 isoforms and β-arrestin-2 tagged with nanoluciferase (right panel; n = 3 biologically independent samples; Methods). Error values correspond to s.e.m.
