Catalytic Core of Adenylyl Cyclase: Structure and Modeling of Its Interaction With Gsa. Vibha D. Rao¹, Gongyi Zhang¹, Yu Liu², Arnold E. Ruoho², Wei-Jen Tang³ and James H. Hurley¹, ¹Laboratory of Molecular Biology, National Institute of Diabetes, Digestive and Kidney Diseases, National Institute of Health, Bethesda, MD 20892-0580, ²Dept. of Pharmacology, University of Wisconsin School of Medicine, Madison, WI 52705, ³Dept. of Pharmacological and Physiological Sciences, University of Chicago, 947 E. 58th St., MC 0926, Chicago, IL 60637

Adenylyl cyclases play a central role in signal transduction by synthesizing 3'5' cyclic AMP when stimulated by G-protein coupled hormone receptors. Mammalian adenylyl cyclases contain two conserved homologous domains, C1 and C2, which together constitute the catalytic machinery of the enzyme. The two domains when expressed separately and mixed are functionally active. The C2 domain alone also exhibits catalytic activity, albeit at a much lower rate. The formation of a single catalytic pocket at the dimer interface of the two domains is clearly revealed by the crystal structure of the C2 homodimer at 2.6 Å. The polypeptide chain folds into a three layer a/ß structure arranged in a wreath like dimer. Two forskolin molecules bridge the hydrophobic ends of the cleft . The central part of the cleft is lined by highly polar and charged residues and is implicated in ATP binding. The active site is formed by the forskolin catalyzed dimerization of the monomers. The carboxy terminus which is also implicated in ATP binding is disordered in the crystal structure. The C1 and C2 are roughly 30% identical, the structure of the C2 domain homodimer presents a plausible model for modeling of the intact catalytic core with a view to understanding the higher activity of the intact enzyme. Differences seen in the residues in the binding cleft along with the asymmetry in the structure in the heterodimer model would possibly affect the mode of binding favoring an optimum accomodation of the substrate and activator, forskolin as compared to the C2 homodimer. Based on the crystal structure of the C2 homodimer, mutagenesis experiments on the adenylyl cyclase and the G protein a subunit, a model of Gs_a was docked onto adenylyl cyclase to study the interaction between the G protein and cyclase. The Gs_a was modeled based on the crystal structure of G transducin a. The model shows the negatively charged groove is complemented by a positively charged region in Gs_a.