Search

Recently three orphan G-protein coupled receptors, RAIG1, GPRC5B and GPRC5C, with homology to members of family C (metabotropic glutamate receptor-like) have been identified. Using the protein sequences of these receptors as queries we identified overlapping expressed sequence tags which were predicted to encode an additional subtype. The full length coding regions of mouse mGprc5d and human GPRC5D were cloned and shown to contain predicted open reading frames of 300 and 345 amino acids, respectively. GPRC5D has seven putative transmembrane segments and is expressed in the cell membrane. The four human receptor subtypes, which we assign to group 5 of family C GPCRs, show 31-42% amino acid sequence identity to each other and 20-25% sequence identity to the transmembrane domains of metabotropic glutamate receptor subtypes 2 and 3 and other family C members. In contrast to the remaining family C members, the group 5 receptors have short amino terminal domains of some 30-50 amino acids. GPRC5D was shown to be clustered with RAIG1 on chromosome 12p13.3 and like RAIG1 and GPRC5B to consist of three exons, the first exon being the largest containing all seven transmembrane segments. GPRC5D mRNA is widely expressed in the peripheral system but all four receptors show distinct expression patterns. Interestingly, mRNA levels of all four group 5 receptors were found in medium to high levels in the kidney, pancreas and prostate and in low to medium levels in the colon and the small intestine, whereas other organs only express a subset of the genes. In an attempt to delineate the signal transduction pathway(s) of the orphan receptors, a series of chimeric receptors containing the amino terminal domain of the calcium sensing receptor or metabotropic glutamate receptor subtype 1, and the seven transmembrane domain of the orphan receptors were constructed and tested in binding and functional assays.

The metabotropic glutamate receptors (mGluRs) belong to family C of the G-protein-coupled receptor (GPCR) superfamily. The receptors are characterized by having unusually long amino-terminal domains (ATDs), to which agonist binding has been shown to take place. Previously, we have constructed a molecular model of the ATD of mGluR1 based on a weak amino acid sequence similarity with a bacterial periplasmic binding protein. The ATD consists of two globular lobes, which are speculated to contract from an "open" to a "closed" conformation following agonist binding. In the present study, we have created a Zn(2+) binding site in mGluR1b by mutating the residue Lys(260) to a histidine. Zinc acts as a noncompetitive antagonist of agonist-induced IP accumulation on the K260H mutant with an IC(50) value of 2 microm. Alanine mutations of three potential "zinc coligands" in proximity to the introduced histidine in K260H knock out the ability of Zn(2+) to antagonize the agonist-induced response. Zn(2+) binding to K260H does not appear to affect the dimerization of the receptor. Instead, we propose that binding of zinc has introduced a structural constraint in the ATD lobe, preventing the formation of a "closed" conformation, and thus stabilizing a more or less inactive "open" form of the ATD. This study presents the first metal ion site constructed in a family C GPCR. Furthermore, it is the first time a metal ion site has been created in a region outside of the seven transmembrane regions of a GPCR and the loops connecting these. The findings offer valuable insight into the mechanism of ATD closure and family C receptor activation. Furthermore, the findings demonstrate that ATD regions other than those participating in agonist binding could be potential targets for new generations of ligands for this family of receptors.

The metabotropic glutamate receptors belong to family C of the G-protein coupled receptor superfamily. These receptors all possess large extracellular amino terminal domains, where agonist binding takes place. We have previously constructed a molecular model of the amino terminal domain of the mGlu(1) receptor based on a weak amino acid sequence similarity with a family of bacterial periplasmic binding proteins (PBPs). The residues Ser(165) and Thr(188) were demonstrated to be involved in agonist binding to the receptor. Here, we report that mutation of Arg(78) in the mGlu(1b) receptor to leucine or glutamate completely knocks out [3H]quisqualic acid binding to the receptor. The constructed mutants, R78L and R78E, have also been characterized in a inositol phosphate assay. Here, the potency of (S)-glutamic acid and (S)-quisqualic acid was reduced 1000- and 100-fold, respectively, on R78L compared to the wild type (WT) receptor. (S)-Quisqualic acid was as potent on mutant R78E as it was on R78L, whereas (S)-glutamic acid was unable to activate R78E significantly at concentrations up to 10 mM. In conclusion, Arg(78) appears to be essential for agonist binding to the mGlu(1) receptor, most likely, through the formation of an ionic bond between its positively charged side chain and the distal acid group of the agonists. Furthermore, the different impact of the two mutations on (S)-glutamic acid and (S)-quisqualic acid potencies strongly indicates that while Arg(78) appears to be a common site of interaction for the agonists, the Group I subtype selectivity of (S)-quisqualic acid is probably determined by other residues in the amino terminal domain.