Functional relevance of naturally occurring mutations in adhesion G protein-coupled receptor ADGRD1 (GPR133)

Abstract

Background: A large number of human inherited and acquired diseases and phenotypes are caused by mutations in G protein-coupled receptors (GPCR). Genome-wide association studies (GWAS) have shown that variations in the ADGRD1 (GPR133) locus are linked with differences in metabolism, human height and heart frequency. ADGRD1 is a Gs protein-coupled receptor belonging to the class of adhesion GPCRs.

Results: Analysis of more than 1000 sequenced human genomes revealed approximately 9000 single nucleotide polymorphisms (SNPs) in the human ADGRD1 as listed in public data bases. Approximately 2.4 % of these SNPs are located in exons resulting in 129 non-synonymous SNPs (nsSNPs) at 119 positions of ADGRD1. However, the functional relevance of those variants is unknown. In-depth characterization of these amino acid changes revealed several nsSNPs (A448D, Q600stop, C632fs [frame shift], A761E, N795K) causing full or partial loss of receptor function, while one nsSNP (F383S) significantly increased basal activity of ADGRD1.

Conclusion: Our results show that a broad spectrum of functionally relevant ADGRD1 variants is present in the human population which may cause clinically relevant phenotypes, while being compatible with life when heterozygous.

Keywords: ADGRD1; Adhesion GPCR; Database; GPR133; Mutations; SNP.

Fig. 1

nsSNPs in ADGRD1 influence cell surface expression and cAMP responsive element activity. a Schematic amino acid structure of ADGRD1 which consists of 874 amino acids. The signal peptide (SP) is followed by a pentraxin domain (PTX), the GPCR autoproteolysis inducing (GAIN) domain (including the two subdomains) harboring a GPCR proteolysis site (GPS), the 7TM region (I-VII) with 3 intra and 3 extracellular loops and a C-terminal tail (CT). The receptor is divided into two fragments: an N-terminal fragment (NTF) and a C-terminal fragment (CTF) after autoproteolytic cleavage at the GPS. Each vertical line indicates a position influenced by amino acid changing single nucleotide variants (data collection: 21-04-2015). b Each dot represents the mean of one inspected nsSNPs in ADGRD1 (for details see Additional file 1). The dotted lines indicate the one- and twofold standard deviation (SD) from wt in CRE-SeAP activity. Data are given as % of wt activity and cell surface expression as mean ± SD. Empty vector (EV) served as negative control (basal activity: EV: 568,311 ± 59,100 counts; wt: 786,125 ± 85,787 counts; n = 4; basal expression: EV: 0.01 ± 0.02 OD_{492 nm}; wt: 0.70 ± 0.11 OD_{492 nm}; n = 12)

Fig. 2

nsSNPs in ADGRD1 influence signaling activity in compartment-depending manner. a N-terminal nsSNPs of ADGRD1 have different influences on signaling activity and expression. The positions of nsSNPs within the N-terminal fragment (NTF) are indexed through the pictogram under the graph. b nsSNPs in 7TM core and C terminus influence cAMP response element (CRE)-signaling activity. The schema under the graph indicates the position of signaling activity relevant nsSNPs: S667L in intracellular loop (ICL) 2, V764M in TM6 (VI), N795K in TM7 (VII) and A816T and T827M in C terminus. c Constitutive activity of P2Y₁₂-CTF-mutant is lost after insertion of N795K. (A-C) Graphs show the percentage of wildtype (wt) after normalization to mock control of functional relevant nsSNPs in CRE-SeAP activity (EV: 568,311 ± 59,100 counts; wt: 786,125 ± 85,787 counts; n = 4), cell surface expression (EV: 0.02 ± 0.03 OD_{492 nm}; wt: 0.70 ± 0.11 OD_{492 nm}; n = 9) and whole cell expression (EV: 0.08 ± 0.07 OD_492nm; wt: 0.97 ± 0.16 OD_492nm; n = 7). To compare differences of receptor mutants to wt basal activity or expression levels an unpaired two-tailed t-test was performed with * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001. d Activation analysis of nsSNPs upon stimulation with the tethered agonist-derived peptide p13. After normalization to mock control (EV: 872 ± 200 counts; n = 7) data are shown as percentage of wildtype stimulated with p13 (wt + p13: 5.3 ± 1.3 x-fold over EV; n = 7). To analyze significance of receptor mutant activation after p13 stimulation a two-way ANOVA with Bonferroni as post-test was used: * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001. a-d Data are given as means ± SD

Fig. 3

nsSNPs leading to length variations do not elicit intracellular signal transduction. a Schematic structure of the cloned constructs for the length variation Q600stop and the frameshift C632fs are shown in the middle. The signal peptide (dark grey triangle) is followed by an HA-tag (light grey box) and the functional domains of the NTF: a PTX (grey box labeled ‘PTX’) and GAIN domain (rectangle including GPS as circle). Transmembrane units are indicated as grey boxes downstream the GPS circle. The C terminus features a hexagonal FLAG-tag. The star indicates the newly cloned 3′UTR region. b Two days after transfection CRE-SeAP assay, cell surface and whole cell ELISA were performed. Graphs show the percentage of wildtype (wt) after normalization to mock control of functional relevant nsSNPs in CRE-SeAP activity (EV: 568,311 ± 59,100 counts; wt: 786,125 ± 85,787 counts; n = 4), cell surface expression (EV: 0.02 ± 0.03 OD_{492 nm}; wt: 0.70 ± 0.11 OD_492nm; n = 9) and whole cell expression (EV: 0.08 ± 0.07 OD_{492 nm}; wt: 0.97 ± 0.16 OD_{492 nm}; n = 7). c Stimulation with p13 could not activate Q600stop. Data are shown as x-fold over mock control (EV: CREB-Luciferase: 1320.6 ± 628.1 counts; n = 4). a-c Data are given as means ± SD. To compare differences of receptor mutants to wildtype basal activity or expression levels an unpaired two-tailed t-test was performed, for receptor activation after p13 stimulation two-way ANOVA with Bonferroni as post-test was used, ** p ≤ 0.01 *** p ≤ 0.001; n.d., not detectable