Adenosine 3',5'-monophosphate-dependent protein kinase associated with the cytoskeleton of adrenal tumor cells

Preparations of cytoskeleton from Y-1 cells were found to phosphorylate various cytoskeletal proteins when incubated with [gamma-32P]ATP. When cAMP was added to the cytoskeleton, a rapid increase in phosphorylation of cytoskeletal protein was observed, and changes were seen in the phosphorylation of individual proteins; four additional proteins were phosphorylated (mol wt, 165,000, 92,000, 45,000, and 24,000) and three proteins were more intensely phosphorylated than without cAMP (mol wt, 125,000, 51,000, and 38,000). In addition, one protein (mol wt, 96,000) that was intensely phosphorylated without cAMP was not phosphorylated with the cyclic nucleotide, and a second (mol wt, 48,000) was less phosphorylated. The increased level of total phosphorylation returned to the unstimulated level within 10 min. The increased phosphorylation of proteins produced by cAMP was inhibited by protein kinase inhibitor. cAMP-dependent protein kinase activity was closely associated with the cytoskeleton, since it was not removed by Triton X-100 (1%, wt/vol), although some activity could be extracted with buffer containing high concentrations of salt. When the cytoskeleton of Y-1 cells was subjected to treatments that disrupt the cytoskeleton before the cells were extracted (cytochalasin B, colchicine, and sonication), no change was seen in cAMP-dependent protein kinase activity. However, cytochalasin B increased phosphorylation of two proteins that were not phosphorylated by cAMP-dependent kinase (mol wt, 63,000 and 43,000). Sonication of the cytoskeleton before addition of [gamma-32P]ATP caused a number of changes in cAMP-independent phosphorylation, but did not affect cAMP-dependent phosphorylation. cAMP-dependent phosphorylation required Mg2+ and was inhibited by Ca2+. It is concluded that the cytoskeleton of Y-1 cells contains bound cAMP-dependent protein kinase that phosphorylates certain cytoskeleton proteins. The cytoskeleton also contains one or more cAMP-independent kinase systems. It is suggested that the cAMP-dependent protein kinase described here may be important in the cytoskeletal responses to ACTH.     

Adenosine 3',5'-monophosphate-dependent protein kinase in supernatants from dispersed bovine parathyroid cells

The presence and characteristics of protein kinase(s) were studied in supernatants of sonicates of dispersed bovine parathyroid cells. cAMP caused a 3- to 5-fold stimulation of protein kinase activity in such extracts, with half of the maximal activation at 4--5 x 10(-8) M cAMP. Protein kinase inhibitor nearly totally abolished both basal and cAMP-stimulated activity, suggesting that most of the activity was cAMP dependent. About 90% of the cAMP-stimulated protein kinase activity eluted from a DEAE-cellulose column at 0.15 m NaCl, consistent with a type II enzyme. The presence of a type II enzyme was also supported by the effects of histone and salt concentrations on enzyme activity; both the basal and cAMP-stimulated activity ratios (activity minus cAMP divided by activity plus 10(-6) M cAMP) were stable in 0.4 M NaCl. The basal activity ratio was not increased by concentrations of histone as high as 10 mg/ml in the protein kinase assay. The predominance of the type II enzyme in dispersed bovine parathyroid cells made it possible to develop conditions for extracting the enzyme from intact intact cells (0.4 m NaCl and 5 mg/ml charcoal), whereby the state of activation of the enzyme remained relatively constant. These studies demonstrate the presence of cAMP-dependent protein kinase activity in dispersed bovine parathyroid cells and define conditions which make it possible to assess the effects of various secretagogues on protein kinase activation in intact parathyroid cells.     

Adenosine 3',5'-monophosphate-dependent protein kinase activity in soluble thyrotropin receptor complex.

cAMP-dependent protein kinase activity was present in a soluble TSH receptor fraction. The Km of this enzyme was 2.2 X 10(-6) M for casein substrate in the absence or presence of 10(-5) M cAMP. A [3H]cAMP-binding protein was also found in this fraction. The Ka for [3H]cAMP-binding was 0.11 X 10(6) M-1, with a total binding capacity of 3 nmol/mg protein. After fractionation using a continuous sucrose density gradient, one of the several [125I]iodobovine TSH-binding peaks corresponded to a [3H]cAMP-binding peak. After fractionation on a sucrose density gradient containing 0.4 M NaCl at pH 6.5, a major peak of protein kinase activity was shown. This protein kinase activity was stimulated by adding 10(-5) M cAMP. A peak of [3H]cAMP-binding activity corresponded to the same peak. Protein kinase activity in the receptor fraction was stimulated by adding 6 mg/ml bovine TSH. The soluble TSH receptor fraction also has an adenylate cyclase activity stimulated by TSH. These results suggest that some TSH receptors released from thyroid plasma membranes have associated adenylate cyclase activity and cAMP-dependent protein kinase activity. The receptor, cyclase, and kinase activities may exist in a functional primary receptor unit which is spontaneously released from plasma membranes.     

Adenosine 3',5'-monophosphate-dependent protein kinase activity decreases in human muscle after insulin infusion

We have previously reported that the increase in glycogen synthase activity in human muscle during a euglycemic clamp was not associated with a measured increase in glycogen synthase phosphatase activity after a 200-min insulin administration. To investigate further the mechanism of the regulation of human muscle glycogen synthase by insulin, we measured the activity of cAMP-dependent protein kinase before and after a 200-min hyperinsulinemic euglycemic clamp in Southwest American Indians. Insulin infusion resulted in a decreased cAMP-dependent protein kinase activity assayed at physiological cAMP concentration with increased glycogen synthase activity in all subjects (n = 5; P less than 0.01). No significant change was observed in cAMP-independent protein kinase activity. These results suggest that 200 min of insulin administration during a euglycemic clamp may regulate human muscle glycogen synthase activity by mechanisms other than the stimulation of phosphatase; one probable mechanism is by decreasing the activity of cAMP-dependent protein kinase.     

Adenosine 3',5'-monophosphate-dependent protein kinase and granulosa cell responsiveness to gonadotropins

The mechanisms by which estradiol enhances the actions of FSH (and cAMP), including the induction of LH receptors in rat ovarian granulosa cells, remain unclear. These studies were conducted to determine the extent to which changes in the activity, content, or intracellular distribution of the catalytic subunit of cAMP-dependent protein kinase might be altered in granulosa cells as a consequence of estradiol, FSH, and hCG administration in vivo. Dose-dependent stimulation of protein kinase activity (measured by histone phosphorylation in the presence of [gamma 32P]ATP and cAMP) demonstrated that the EC50 for cAMP was consistently 20 X 10(-8) M in cytosols prepared from granulosa cells of hypophysectomized rats before and after treatment with estradiol alone or estradiol and FSH. However, estradiol alone caused a 1.5 to 2.0-fold increase in the total amount of enzyme activity. When the cytosol content of the catalytic subunit (C) was quantitated directly, using immunoblotting procedures, the amount of C was 40 pmol/mg protein in all tissues, regardless of hormone treatments in vivo. When the content of RII, the regulatory subunit of type II cAMP-dependent protein kinase, was measured by similar immunoblotting procedures, a 10-fold increase was observed in granulosa cells exposed to both estradiol and FSH compared to that in cells exposed to estradiol alone. Greater than 80% of the intracellular content of both C and RII was present in the cytosol fraction (30,000 X g supernatant) rather than in the particulate nuclear fraction (30,000 X g pellet) of granulosa cells. This distribution of subunits was not altered by rapidly elevating intracellular concentrations of cAMP in vivo with 10 IU hCG, iv. We conclude that the catalytic subunit of protein kinase is a constitutive component of granulosa cells and that the sensitivity of the enzyme for cAMP is not affected by hormones or by a 10-fold increase in RII. Thus, the ability of estradiol to enhance FSH and cAMP action in granulosa cells appears to come primarily from the induction of specific substrates for the enzyme and a small increase in the catalytic activity but not from a change in the content of the catalytic subunit.     

Adenosine 3',5'-monophosphate-dependent phosphoproteins in human placenta

cAMP modulates estrogen, hCG, and lactate syntheses by the human placenta. The major effects of cAMP are presumably mediated through the phosphorylation of specific regulatory phosphoproteins after cAMP activation of cAMP-dependent protein kinase. cAMP-dependent phosphoproteins have not been identified in the placenta. Homogenates and cytosol from term human placenta were subjected to an endogenous protein phosphorylation assay using [gamma-32P]ATP in the presence and absence of 1.0 microM cAMP. Protein phosphorylation was assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography. cAMP stimulated 32P incorporation into proteins with mol wt of 25,000, 27,000, 39,000, 45,000, 52,000, 58,000, and 73,000 (P less than 0.02). Half-maximal 32P incorporation was observed with 1.0 X 10(-7) M cAMP, which was similar to the concentration required for half-maximal histone kinase activity (8.5 +/- 2.9 X 10(-8) M). cGMP induced 32P incorporation into the same placental proteins as cAMP, but 2 orders of magnitude greater cGMP concentrations were required to achieve phosphorylation levels similar to those caused by cAMP. cAMP-dependent protein kinase inhibitor completely blocked cGMP-induced phosphorylation, even when histone protein was added. Therefore, no evidence of a cGMP-dependent protein kinase or specific cGMP-dependent phosphoproteins were detected. CaCl2 (10(-8) - 10(-4) M) had no effect on cAMP-induced 32P incorporation into the seven cAMP-dependent phosphoproteins. However calcium induced 32P incorporation into four other proteins (mol wt, 97,000, 90,000, 20,000, and 19,000). Regulation of placental metabolism by catecholamines and other hormones known to mediate intracellular cAMP or calcium levels may be accomplished by phosphorylation of these phosphoproteins.     

Adenosine 3'5'-cyclic monophosphate dependent protein kinase in human adrenocortical tumors.

Protein kinase activity has been studied in four human adrenocortical tumors and compared to the one of the normal human adrenal. In two cases where the lack of action of ACTH was related to an anomaly of ACTH receptor, the protein kinase activity was normal. In the other two cases the ACTH receptor was normal, but the protein kinase activity was different from that of the normal adrenal. In one of these cases where the steroidogenesis response of isolated tumor cells to ACTH and DcAMP was higher than in normal adrenal, basal and cAMP stimulated protein kinase activities were significantly higher than those of the normal adrenal, but the activation constants of both nucleotides were similar to those of the normal gland. In the other case, the basal and the cAMP stimulated protein kinase activities were significantly lower, as well as the activation constant of cAMP. However, the binding affinity of 3H-cAMP was normal. Normal adrenal cytosol contains three protein kinases, as resolved by DEAE-cellulose, two of which designated I and II, are cAMP-dependent. The DEAE-cellulose chromatography of the last tumor showed a loss of isoenzyme II. In addition, the protein kinase eluted at the same molarity as that of isoenzyme I of the normal adrenal was not activated by cAMP. Therefore, the lack of response to ACTH of some adrenocortical human tumors may be attributed either to an anomaly of the ACTH receptor or to some defect of the cAMP-dependent protein kinase.     

Ultrastructural localization of cyclic adenosine 3',5'-monophosphate-dependent protein kinase after adrenocorticotropin stimulation in adrenal cortical tumor cells

To increase our knowledge of the molecular details of peptide hormone action, a specific immunogold staining procedure for the ultrastructural localization of cAMP-dependent protein kinase regulatory (RI) and catalytic (C) subunits was used in Y-1 adrenal cortical tumor cells. The Y-1 adrenal cell responds to ACTH (40 mU/ml) with a decrease in cell division and an increase in steroid production. Corresponding to the decrease in rate of cell division and the increase in steroid production, there was a 2-fold increase in both nuclear and cytoplasmic localization of the C subunit after 60-min ACTH (40 mU/ml) treatment of the culture. The amount of immunogold staining in the adrenal tumor cells after localization with antiserum of the RI subunit decreased after 60-min ACTH (40 mU/ml) stimulation. A 2-fold decrease in labeling in the cytoplasm and nucleus was observed for the RI subunit. The loss of RI subunit from the soluble fraction of the cytoplasm of the cell or an alteration of this RI subunit is suggested by this investigation. Since there was no increase in the RI subunit in the nuclear compartment, a loss of RI subunit from the cytoplasm into the nucleus seems unlikely. The observed immunogold changes in the C subunit after ACTH treatment correspond to the reported changes observed with light microscopic techniques with a fluorescein-coupled inhibitor as a probe for the localization of free C. The immunogold technique allows for the ultra-structural identification and quantification of nuclear as well as the cytoplasmic sites of cAMP-dependent protein kinase after hormonal stimulation. These results support the proposed role of protein kinase as a mediator of the ACTH response.     

Adenosine 3',5'-cyclic monophosphate dependent protein kinase activity in differentiating germ cells of the mouse testis.

Protein kinase activity was studied in middle-late pachytene spermatocytes, round spermatids (steps 1–8 of spermiogenesis) and elongating spermatids (steps 9–13) isolated by velocity sedimentation from the seminiferous tubules of the mouse testis. The 3 cell fractions obtained by this method were 80–90% homogeneous and devoid of somatic cells as judged by light microscopy.cAMP-dependent protein kinase activity was present both in pachytene spermatocytes and in spermatids and was mostly recovered in the soluble fraction of the cell homogenate.The phosphorylative activity expressed per cell was comparable to that described in somatic testicular cells, thus excluding that the activity measured was due to minor somatic contamination undetected with the morphological analysis. Further studies on the phosphokinase activity have shown a stimulation of several fold by cAMP, a preference for protamine as substrate (protamine > total histone > histone F2a) and ATP and Mg²⁺ requirements similar to those of kinase characterized in other compartments of the testis and in other tissues. Fractionation on DEAE-cellulose of the cytosol prepared from pachytene spermatocytes and round spermatids demonstrated two major peaks of activity eluted with 40 and 180 mM KCl; both enzymes were stimulated by cAMP and specifically bound ³H-cAMP. The same type of analysis performed on elongating spermatids demonstrated that most of the kinase activity eluted with 180 mM KCl.     

Hormonal regulation of chick kidney inhibitor of adenosine 3',5'-monophosphate-dependent protein kinase

The endogenous inhibitor of cAMP-dependent protein kinase (PKI) in chick kidney is regulated by the vitamin D status of the animal. To determine the specific factors that are involved in the regulation of chick kidney PKI, chicks were raised on a low (0.05%), normal (1%), or high (3%) calcium diet and given vitamin D3 or vehicle three times a week orally. The results from this experimental protocol show that vitamin D3 or one or more of its metabolites and serum calcium levels are both involved in the regulation of chick kidney PKI in vivo. Measurement of PKI activity in primary cultures of chick kidney cells revealed treatment with 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3) led to a 90-95% decrease in PKI activity. This effect of 1,25-(OH)2D3 was dose dependent, and neither PTH nor insulin was able to reverse it completely. Treatment with PTH caused 30-60% increase in PKI activity, and cell cultures that were grown in medium containing either 0.5 or 2 mM calcium chloride had similar PKI activities. Taken together, these results indicate that 1,25-(OH)2D3, the most physiologically active form of vitamin D3, is the predominant regulator of PKI, but serum calcium, indirectly through the regulation of PTH secretion, is also involved.     

Adenosine 3',5'-monophosphate (cAMP)-binding protein and cAMP-dependent protein kinase in human placenta

cAMP modulates estrogen, hCG, and lactate syntheses by human placenta, cAMP presumably exerts its major intracellular effect by binding to cAMP-dependent protein kinase (cAMP-PK), which, in turn, phosphorylates regulatory proteins within the target cell. cAMP binding and cAMP-PK have not been previously identified in placenta. [3H]cAMP binding to crude cytosol fractions of term placenta was rapid, saturable, and reversible. Scatchard analyses of saturation experiments of [3H]cAMP binding to placental cytosol were linear (Kd = 1.13 +/- 0.11 x 10(-8) M; n = 5). The binding capacity was 1.27 +/- 0.18 pmol/mg protein. Competition for the [3H]cAMP-binding site followed the potency order cAMP much greater than cGMP much greater than (Bu)2cAMP, analogous to cAMP binding to cAMP-PK in other tissues. ADP, ATP, and adenosine did not compete for the [3H]cAMP-binding site. cAMP significantly enhanced phosphorylation of histone protein by placental cytosol (activity ratio, 0.57 +/- 0.04; P less than 0.01). Two peaks of [3H]cAMP binding and coincident cAMP-PK activity were identified by DEAE-cellulose column chromatography of placental cytosol corresponding to classical type I and type II cAMP-PK. While the majority of the cAMP-PK was found in placental cytosol, cAMP-PK was also demonstrated in crude microsomal and microvillous brush border membranes of human placenta after solubilization with Triton X-100 (P less than 0.05). Regulation of placental function by catecholamines and other hormones known to mediate cAMP levels may be accomplished through the phosphorylation of cellular proteins by cAMP-dependent protein kinases.     

Adenosine 3',5'-monophosphate-dependent loss of growth hormone binding in rat adipocytes

GH exerts a number of metabolic effects on adipose tissue. Depending on the circumstances, it may increase or decrease glucose metabolism and lipolysis. These effects appear to be mediated by a single class of receptors, which bind GH with high affinity. Incubation of isolated rat adipocytes with a variety of lipolytic agents, including catecholamines, forskolin, or (Bu)2cAMP, decreased the specific binding of [125I]human (h) GH within 10 min. In the presence of 10 microM forskolin, GH binding declined to less than 20% of the control value within 50 min. Cholera and pertussis toxins, which increase cAMP secondary to ADP ribosylation of guanine nucleotide-binding proteins associated with hormone receptors, also decreased the binding of GH. None of these agents affected the rate of loss of cell-associated 125I when added to cells that had previously equilibrated with [125I]hGH. The inhibitory effects of forskolin and (Bu)2cAMP were at least as great when binding was measured in the presence of the protease inhibitor leupeptin, suggesting that increased rates of internalization and processing of bound hormone could not account for the decline in binding. Scatchard plots of data obtained in the presence of forskolin or (Bu)2cAMP were linear and parallel to control plots, indicating that the decline in binding could be accounted for by a decrease in the number of binding sites, with no change in affinity. To determine whether phosphorylation affected binding to receptors already present in the membrane or modified the turnover of receptors, we studied adipocyte ghosts, whose cellular apparatus for receptor turnover is disrupted. Incubation of adipocyte ghosts with cAMP-dependent protein kinase decreased the binding of [125I]hGH by 25%. The data suggest that cAMP-dependent phosphorylation of the GH receptor or a closely associated membrane protein renders the receptor incapable of binding GH.     

Activity ratio measurements reflect intracellular activation of adenosine 3',5'-monophosphate-dependent protein kinase in osteoblasts

Parathyroid hormone, prostaglandin E2, and prostacyclin activate cAMP-dependent protein kinase in osteoblast-rich normal rat calvarial cells and in clonal rat osteogenic sarcoma cells of osteoblastic phenotype. The present study was undertaken to determine the activation of the enzyme in relation to cellular cAMP concentrations at increasing doses of the three hormones and also to test that the activity ratio measurement of the enzyme (ratio of the activity in the absence of cAMP to the activity in the presence of excess cAMP) was a true reflection of intracellular activation of the enzyme. With each hormone, using either normal or malignant osteoblasts, activation of the enzyme took place at hormone concentrations lower than those required to produce detectable changes in cAMP concentrations in the incubations. Stimulation of activity was abolished by addition of the heat-stable inhibitor of cAMP-dependent protein kinase, indicating that activation was of cAMP-dependent protein kinase alone. To demonstrate that protein kinase activation occurred intracellularly and not during sample preparation, charcoal was added at the time of cell disruption to absorb free cAMP. Under these conditions, no change was observed in the concentration of bovine parathyroid hormone required to cause activation of cAMP-dependent protein kinase. Finally, addition of purified cAMP-dependent protein kinase type I or type II to treated cells at the time of lysis did not result in significant activation of added isoenzyme, except at hormone concentrations sufficient to increase the total cAMP concentration of incubations. It is concluded that activity ratio measurement reflects the intracellular state of activation of cAMP-dependent protein kinase in the osteoblast-like cells treated by hormones and, furthermore, that only a fraction of the maximally generated cAMP is necessary for full enzyme activation.     

Distribution and localization of adenosine 3':5'-monophosphate-dependent protein kinase in mammalian artery

Among the three major vascular layers (the intima-inner, the media-middle smooth muscle, and the adventitia-outer connective tissue) over 90% of the total protein kinase activity was observed in the middle layer. Of various subcellular fractions of the vascular smooth muscle, the 105,000 g supernatant (cytosol fraction) showed the highest specific activity and represented more than two thirds of the total kinase present in this layer. DEAE-cellulose chromatography of the soluble enyzme revealed the existence of two major forms of cyclic AMP-dependent protein kinase, type I and type II, of which 60% of the total enzymatic activity was found in type II. A divalent cation was found to be essential for their phosphotransferase activity. Only Mg2+ and Co2+, but not Zn2+, Mn2+ or Ca2+ could satisfy the cation requirement. The phosphorylated substrate had the characteristics of a protein with a phosphoester bond.