The interaction between the GTP-binding protein, its receptor, and its effector at the plasma membrane is well characterized. In contrast, the specific interaction and function of similar systems in the Golgi membranes is still not clear. A G Alpha Interacting Protein (GAIP) was chosen as a model to study this interaction. GAIP interacts specifically with the Gai3 which has been localized to the Golgi membranes. A plasmid construct containing the core domain (150 residues) of GAIP was constructed. The core domain of GAIP contains a homology domain found in a novel family of regulators of G protein signaling (RGS proteins). The three dimensional fold of human GAIP has been determined using NMR spectroscopy. The refinement of the structure of GAIP is in progress. Human GAIP at a concentration higher than 0.1 mM exists as a dimer in solution. This results in an effective MW of roughly 34 kD. The initial fold was determined without further deuteration of the protein which is typically done for structure determination of proteins of this size by solution NMR. A backbone dynamic study of human GAIP has also been carried out using NMR. This confirms our finding that GAIP exists as a dimer in solution, at least at concentrations higher than 0.1mM. The dynamic data also reveals the regions which have flexibility. Initial comparison of GAIP and the X-ray structure of RGS4 complexed to Gai1 reveals some conformational changes upon binding to the G protein. The dynamic data suggests possible flexibility that allows the conformational change in the structure. A parallel project to express the G ai3 subunit has been initiated. The goal is to be able to reconstruct human GAIP and its G protein complement in vitro and observe the biochemical properties. We have completed the solution structure of human GAIP and carried out detailed comparison to the structure of RGS4 complexed to Galpha-i. We concluded that the activation of catalysis by RGS protein is through stabilization of the complex structure, not by direct interaction of RGS to the active site of Galpha. Furthermore, we have shown that the loop between helix Vand VI which contacts the Galpha differs in structure only for the N-terminal portion. The C-terminal portion of this loop does not adopt a different conformation upon binding the Galpha. We are finishing the dynamic study of this protein. We have also initiated a structural study of a calcium binding protein, CALNUC. This protein in the calcium loaded state binds Galpha in the Golgi. It is believed that CALNUC is regulated through its interaction with Galpha to modulate calcium concentration in the Golgi apparatus. CALNUC does not seem to effect the GTP hydrolysis in Galpha. Therefore we hypothesize thatthere are several different modes of binding to the Galpha. These different modes govern a subset of different functions that the Galpha would undertake to respond to a certain stimulus. We have constructed the CALNUC plasmid which encompasses the two EF hands. We have successfully expressed the protein and have carried out experiments on calcium binding as well as peptide binding. The peptide used represents the C-terminal helix of the Gai. Our results so far show that the protein undergoes a certain degree of exchange between two conformations that results in broadening of the resonance signals. However, we have been able to establish the specificity of calcium binding as well as peptide binding. We are currently identifying the different conformations that simultaneously exist. It is interesting that this exchange process does not seem to effect CALNUC's ability to bind calcium or its target peptide.
Showing the most recent 10 out of 15 publications
