Comparison of the mammalian adenylyl cyclase 8 and the Drosophila rutabaga adenylyl cyclase in lipid rafts

Jessica L Sorge1, Brian P Head2, Hemal H Patel2, Dermot MF Cooper1. 1University of Cambridge, Department of Pharmacology, Tennis Court Road, Cambridge, CB2 1PD, UK, 2University of California, San Diego, Department of Anesthesiology, La Jolla, 92093, USA

Ca2+-stimulated adenylyl cyclases (ACs) play key roles in learning and memory in both mammals and flies. Whereas much attention has focussed on the mammalian Ca2+-stimulated enzymes in terms of their dependence on calmodulin (CaM) regulation in vitro and regulation by Ca2+-entry in vivo, little is known of these aspects for Drosophila rutabaga AC, with the exception that Drosophila rutabaga AC is also known to be stimulated by Ca2+/CaM. For instance, it is known that the mammalian Ca2+/CaM stimulated adenylyl cyclase 8 (AC8) is required to reside in lipid rafts to be susceptible to the entry of Ca2+ (Fagan et al., 2000), but nothing is known of the regulation of the Drosophila rutabaga AC by Ca2+ nor indeed whether it resides in lipid rafts. Therefore, we asked whether similar behaviour was shown by the Drosophila enzyme to the mammalian AC8.

To address this issue, the mammalian AC8, the Drosophila rutabaga AC and a truncated form were expressed in HEK293 cells. The Drosophila rutabaga AC has a similar overall amino acid sequence to the mammalian AC8, except at the C-terminus, where an additional 1037 amino acids are included. We deleted these extra 1037 amino acids from the C-terminus to produce a truncated form. Cells were fractionated into raft and non-raft fractions via a sucrose gradient after extraction with cold Triton X-100 (about 70% of the total caveolin remained in light fractions; n = 3). In further experiments, the lipid rafts were disrupted in all cells by cholesterol depletion using methyl-β-cyclodextrin (about 10% of the total caveolin remained in light fractions; n = 3). We find that, like the mammalian enzyme, there is an absolute dependence on Ca2+-entry for the regulating the Drosophila rutabaga AC by Ca2+. The Drosophila rutabaga AC is also found in lipid rafts along with the lipid raft marker, caveolin, like the mammalian AC8 (80% of the ACs remained in light fractions; n = 3). However, the truncated Drosophila rutabaga AC is present in both rafts and non-raft fractions (70% of the AC remained in light fractions and 30% in the heavy fractions; n = 3). Using cholesterol depletion, the mammalian and Drosophila rutabaga is moved out of lipid rafts, whereas the truncated Drosophila rutabaga AC persists in rafts and non-raft fractions.

The results lead us to conclude that the Drosophila enzyme behaves like the mammalian enzyme unlike the truncated Drosophila enzyme. This could mean that the Drosophila enzyme might have the same targeting mechanism as the mammalian AC8 and that the C-terminus in the Drosophila enzyme might play a key role in this mechanism because the truncated Drosophila enzyme does not show the same characteristics in our hands.

To confirm the similar targeting mechanism of the Drosophila rutabaga AC and mammalian AC8, ongoing experiments involve caveolae disruption by caveolin-1 knockdown using caveolin-1 shRNA, examination of the consequences of the knockdown on AC fractionation, and measurements of the effect of Ca2+-entry on the AC activity in these caveolin-1 knockdown cells.

K.A. Fagan, K.E. Smith and D.M.F. Cooper (2000) Regulation of the Ca2+-inhibitable adenylyl cyclase type VI by capacitative Ca2+-entry requires localization in cholesterol-rich domains. J. Biol. Chem. 275, 26530-26537