Cyclic AMP-dependent and -independent protein kinases in HeLa cells

Abstract

The protein kinase activities of HeLa cell cytosol have been characterized. Differentiation between cyclic AMP-dependent and -independent kinase activities was achieved by the use of purified specific cyclic AMP-dependent protein kinase inhibitor. HeLa cells were found to contain high levels of soluble cyclic AMP-dependent protein kinase activity, as well as of cyclic AMP-independent kinases which phosphorylate casein and endogenous substrates. Also present were low levels of cyclic AMP-independent histone kinase activity. These various activities were characterized by differences in their substrate specificities and response to NaCl. The cyclic AMP-dependent protein kinase was characterized further and conditions achieved for the in vitro stabilization of its activity ratio (activity in absence of cAMP/activity in presence of cAMP). Type I and type II cyclic AMP-dependent protein kinases, as well as casein kinase, were separable by DEAE-cellulose column chromatography. Asynchronously growing cells had 60% type I and 40% type II cyclic AMP-dependent kinase. Type I, but not type II, holoenzyme was readily dissociable by NaCl. NaCl prevented reassociation of type II subunits. Type I and type II activity ratios in crude supernatants were stable in 150 m m NaCl. Activity ratios in log phase cells were found to be low (about 0.1), despite relatively high concentrations of cellular cyclic AMP, and a low k a for cyclic AMP in extracts. This may be explained by the relatively high enzyme content, since the k a for cyclic AMP increased with protein kinase concentration. When cells were incubated with increasing concentrations of 1-methyl-3-isobutyl-xanthine, with or without isoproterenol, cyclic AMP levels and activity ratios rose in a parallel fashion, though activity ratio changes were rather insensitive to cyclic AMP changes. As synchronized cells progressed through the cell cycle cyclic AMP-independent histone kinase and casein kinase both underwent a peak at mitosis. Cyclic AMP levels and activity ratios of the cyclic AMP-dependent protein kinase both declined to their lowest level at mitosis, rising again during the G 1-phase. The results are consistent with the hypothesis that cyclic AMP may exert regulatory influence in mitosis via alteration of the activation state of cyclic AMP-dependent protein kinase.