1. Adenylate cyclase (EC 4.6.1.1) activity was characterized in human liver, and its subcellular distribution compared with that of three other potential enzyme markers of the pericellular membrane: leucine aminopeptidase (EC 3.4.11.1), γ-glutamyltransferase (EC 2.3.2.2) and 5′-nucleotidase (EC 3.1.3.5). Although these three enzyme activities were detected in each of the subcellular fractions studied, 85% of the total adenylate cyclase activity was found in the 1000 g pellet (‘nuclear’ fraction) with a threefold increase in specific activity as compared with the homogenate. No adenylate cyclase activity existed in the 150 000 g supernatant fraction.
2. In the ‘nuclear’ fraction, adenylate cyclase activity was increased in a dose-dependent fashion by glucagon with a half-maximal stimulation at 10 nmol/l and a maximal four- to seven-fold increase at 1 μmol/l. Catecholamines activated adenylate cyclase 2·5- to three-fold, with an order of potency (protokylol > isoprenaline > adrenaline > noradrenaline) typical of a β2-adrenoreceptor. Prostaglandin E1 and NaF also stimulated cyclase two- and four-fold respectively. Insulin, serotonin, dopamine, thyroid-stimulating hormone and ACTH had no effect. Adenosine provoked a weak inhibition at 0·1 mmol/l. Finally guanosine triphosphate and 5′-guanylyl imidodiphosphate induced a marked increase in basal activity, four- and eight-fold respectively, but both reduced the relative increase in enzyme activity due to glucagon or adrenaline.
3. Cyclase from foetal liver (12–16 weeks old) and cirrhotic adult liver appeared to behave similarly to that from normal liver; however, foetal cyclase was more active, and cirrhotic enzyme less active than normal adult liver. Both systems responded to catecholamines via a β2-adrenoreceptor.
4. These results validate the use of rat liver adenylate cyclase as a tool for pharmacological and physiological studies.