ABSTRACT
A new strategy has been successfully applied to reconstitute the brain specific serotonin 5-HT1A receptor-G protein-adenylate cyclase complex. A mild method of tissue preparation gave a stable, membrane-bound form of the receptor (SBP) which retained its natural lipid content. Treatment of SBP with serotonin (1 microM) and 3-[(3-cholamidopropyl) dimethyl ammonio]-1-propanesulphonate (CHAPS) (2%) solubilized the ligand-receptor-G protein-ligand complex along with the associated phospholipids and cholesterol. Dialysis of this extract (SBDS) against buffer containing 25% ethylene glycol produced a stable, reconstituted and active preparation (SBDSE) of vesicles which upon centrifugal separation followed by gentle resuspension retained 95-100% [3H] 8-OH-DPAT binding activity as well as 60% [3H] GppNHp binding and adenylate cyclase activities of SBDSE. The reconstituted receptor preparation compared well with the membrane-bound form in displaying a similar value for KD (2.1 nM) and a single affinity state for [3H] 8-OH-DPAT binding (Bmax = 118 fmol/mg). However, in sharp contrast to the membrane-bound receptor which was negatively coupled to adenylate cyclase, agonist treatment of the solubilized and reconstituted receptor resulted in an increase in adenylate cyclase. This change in receptor-adenylate cyclase coupling following reshuffling of membrane lipids during solubilization and reconstitution suggested that membrane lipids could have a profound effect on receptor-effector coupling. To study the effect of membrane lipid composition on receptor-mediated signal transduction in a stabler and more natural system, neural cells derived from different parts of the brain (hippocampus, HN2; CNS, NCB-20; dorsal root ganglion, F-11) and a non-neural cell line (CHO), all with differing membrane lipid compositions, were selected. Since no known cell line contains the serotonin 5-HT1A receptor (5-HT1A-R), stable transfection of the selected cell lines with a DNA construct encoding the human 5-HT1A-R was carried out and this resulted in a late increase of [3H] 8-OH-DPAT binding in the stationary phase only in the cell lines of neural origin. In the non-neural cell line (CHO), which also displayed marked difference in membrane lipids, the receptor was positively coupled to the phospholipase C-IP3-[Ca2+]i cascade. Even though GPLC was present in the NCB-20 and F-11 cells as evidenced by a bradykinin receptor-mediated increase in inositol phosphates in these cells 8-OH-DPAT treatment resulted in no change in phospholipase C in any of the cell lines of neural origin.(ABSTRACT TRUNCATED AT 400 WORDS)