Structure And Functions Of Signal-transducing G-proteins

Deafness &Other Communication Disorders

Investigators

Linked publications & trials

Abstract

We have completed establishing the methods for measuring G-protein-coupled receptor (GPCR) interactions with G-proteins by surface plasmon resonance spectroscopy (SPR). For these studies we have utilized the binding of extracellular carbohydrates of bovine rhodopsin by mitogenic lectins (concanavalin A, ConA) to immobilize a prototype GPCR on the metal surface substrate for SPR spectroscopy. Rhodopsin immobilized in this manner retained full biochemical activity to catalyze the activation of retinal transducin (Gt). The binding interactions of retinal Galpha and Gbeta-gamma subunits with rhodopsin were profoundly synergistic. Binding of Gbeta-gamma dimers with distinct gamma subunits to rhodopsin, independent of alpha subunit, was readily observable by SPR. Further, these dimers displayed dramatically different binding affinities and kinetics. The physiologically appropriate retinal dimer displayed rapid association and dissociation kinetics, while the other beta-gamma dimers dissociated at more than 100-fold slower rates. These data suggest that the duration of a G-protein-coupled receptor-signaling event is an intrinsic property of the G-protein coupling partners, in particular, the beta-gamma dimer. These findings are being followed up by analysis of the binding interactions of mutant Gt alpha subunits with alterred guanine nucleotide binding properties and of beta-gamma dimers with chimeric gamma subunit chains, as described below. We have continued to explore the structural determinants of Gbeta-gamma selectivity of receptors by analysis of the independent contributions of isoprenoid modification and primary protein structure of gamma subunits. Dimers of beta1gamma1 proteins have lowered affinity for rhodopsin than that for beta1gamma2. However, gamma1 and gamma2 proteins also differ in isoprenoid modification. To test the relative contributions of protein and isoprenoid structures, we constructed mutants of gamma1 and gamma2 subunits encoding the C-terminal recognition sequences for both farnesyl and geranylgeranyl transferases to introduce C15 (farnesyl) or C20 (geranylgeranyl) isoprenoids into gamma1 and gamma2 proteins. Further, we constructed a set of chimeric protein structures with gamma1 and gamma2 sequences replacing homologous sequence all with the identical C-terminal geranylgeranyl transferase recognition sequence. All proteins were expressed as dimers with the beta1 gene product using baculoviral vectors in Sf9 cells, and the expressed beta-gamma dimmers were purified to near homogeneity for in vitro biochemical and biophysical analyses. The biochemical activity tests reveal that all constructs display equivalent interaction with the retinal G-protein alpha-t, expect a mutant constructed to contain no isoprenoid modification, which is severely reduced in affinity for alpha. However, both isoprenoid and protein modifications result in differing affinities for rhodopsin interaction, tested either for activation of the alpha-t or by surface plasmon resonance measurement of binding to rhodopsin. The isoprenoid modifications of the expressed mutant and chimeric gamma chains were confirmed by mass spectrometry, revealing that the prenyltransferase specificities are not uniquely determined by the carboxyl terminal 3 amino acids (CAAX motif). However, the products were sufficiently correctly modified to allow the assignment of the C-terminal third of the gamma chain as a major determinant of rhodopsin affinity of the beta-gamma dimmer. In addition, they reveal that isoprenoid modification has a more significant impact on gamma1 affinity for rhodopsin than for gamma2. This past year we have extended our characterization of the mechanism of activation of class3 GPCR using the metabotropic glutamate receptor (mGluR1a) and the calcium sensing receptor (CaR). Our initial experiments were designed to address the function(s) of the large amino-terminal domains of these receptors, by expressing chimeric receptors (mGluR1a/CaR and CaR/mGluR1a) bearing the N-terminal domain of the rat mGluR1a replacing homologous sequence of the human CaR and the converse. We also constructed a soluble, secreted N-terminal domain of mGluR1a (mGluR-N-term) and a membrane anchored form of the N-terminal domain consisting of the amino-terminal domain and the first transmembrane helix of the mGluR1a (mGluR1a-tm). The properties of the mGluR1a, CaR, mGluR1a/CaR and mGluR1a-tm were examined by expression in HEK293 cells either transiently or stably transformed. Quantitative examination of the binding and signaling properties of these receptors reveals that the chimeric receptors are considerably less efficient in signaling, with an EC50/Kd ratio some 7-fold higher that the wild-type mGluR1a. Further, maximal PI-hydrolysis response in HEK293 cells expressing equivalent abundances of mGluR1a/CaR or CaR/mGluR1a are some 5-fold lower than for mGluR1a or CaR parent structures. These data indicate that signaling initiated by binding of ligand to a class3 GPCR is transmitted by amino-terminal domain interaction with the transmembrane helix bundle of the receptors. In addition, the contact sites for this interaction are sufficiently, but not completely, conserved between mGluR1a and CaR such that the domain swaps produce a functional receptor with reduced signaling efficiency. To examine the intramolecular mechanism(s) of class3 GPCR, we have expressed the transmembrane helix bundles of the CaR and mGluR1a without their amino-terminal domains. The signaling activities of these structures are readily measured in membranes from HEK293 cells expressing them as measured by catalytic activation of Gq protein. Further, the CaR 7TM structure contains multiple, interactive ligand regulatory sites for divalent metal ion, basic amino acids and the novel allosteric ligand (NPS568). These allosteric sites are functioning within the full-length structure, since the metal and NPS regulation also is exhibited by the mGluR1a/CaR, but not the CaR/mGluR1a chimera. These data indicate that the regulation of class3 GPCR is more complex than that for rhodopsin-homologous receptors, with the structures exhibiting unique allosteric regulatory sites and properties.