Rat brain guanylate cyclase. Purification, amphiphilic properties and immunological characterization

Nucleotidyl cyclase activity of soluble guanylyl cyclase α1β1

Soluble guanylyl cyclase (sGC) regulates several important physiological processes by converting GTP into the second-messenger cGMP. sGC has several structural and functional properties in common with adenylyl cyclases (ACs). Recently, we reported that membranous ACs and sGC are potently inhibited by 2',3'-O-(2,4,6-trinitrophenyl)-substituted purine and pyrimidine nucleoside 5'-triphosphates. Using a highly sensitive high-performance liquid chromatography-tandem mass spectrometry method, we report that highly purified recombinant sGC of rat possesses nucleotidyl cyclase activity. As opposed to GTP, ITP, XTP and ATP, the pyrimidine nucleotides UTP and CTP were found to be sGC substrates in the presence of Mn(2+). When Mg(2+) is used, sGC generates cGMP, cAMP, cIMP, and cXMP. In conclusion, soluble "guanylyl" cyclase possesses much broader substrate specificity than previously assumed. Our data have important implications for cyclic nucleotide-mediated signal transduction.

Effects of chronic hypoxia on soluble guanylate cyclase activity in fetal and adult ovine cerebral arteries

A broad variety of evidence obtained largely in pulmonary vasculature suggests that chronic hypoxia modulates vasoreactivity to nitric oxide (NO). The present study explores the general hypothesis that chronic hypoxia also modulates cerebrovascular reactivity to NO, and does so by modulating the activity of soluble guanylate cyclase (sGC), the primary target for NO in vascular smooth muscle. Pregnant and nonpregnant ewes were maintained at either sea level or at 3,820 m for the final 110 days of gestation, at which time middle cerebral arteries from term fetal lambs and nonpregnant adults were harvested. In both fetal and adult arteries, NO-induced vasodilatation was attenuated by chronic hypoxia and completely inhibited by 10 microM 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ), a selective inhibitor of sGC. sGC abundance (in ng sGC/mg protein) measured via Western immunoblots was approximately 10-fold greater in fetal (17.6 +/- 1.6) than adult (1.7 +/- 0.3) arteries but was not affected by chronic hypoxia. The specific activity of sGC (in pmol cGMP.microg sGC(-1).min(-1)) was similar in fetal (255 +/- 64) and adult (280 +/- 75) arteries and was inhibited by chronic hypoxia in both fetal (120 +/- 10) and adult (132 +/- 26) arteries. Rates of cGMP degradation (in pmol cGMP.mg protein(-1).min(-1)) were similar in fetal (159 +/- 59) and adult (134 +/- 36) arteries but were not significantly depressed by chronic hypoxia in either fetal (115 +/- 25) or adult (108 +/- 25) arteries. The cGMP analog 8-(p-chlorophenylthio)-cGMP was a more potent vasorelaxant in fetal (pD(2) = 4.7 +/- 0.1) than adult (pD(2) = 4.3 +/- 0.1) arteries, but its ability to promote vasodilatation was not affected by chronic hypoxia in either age group. Together, these results reveal that hypoxic inhibition of NO-induced vasodilatation is attributable largely to attenuation of the specific activity of sGC and does not involve significant changes in sGC abundance, cGMP-phosphodiesterase activity, or the vasorelaxant activity of protein kinase G.

Differential inhibition of adenylyl cyclase isoforms and soluble guanylyl cyclase by purine and pyrimidine nucleotides

Mammals express nine membranous adenylyl cyclase isoforms (ACs 1-9), a structurally related soluble guanylyl cyclase (sGC) and a soluble AC (sAC). Moreover, Bacillus anthracis and Bacillus pertussis produce the AC toxins, edema factor (EF), and adenylyl cyclase toxin (ACT), respectively. 2'(3')-O-(N-methylanthraniloyl)-guanosine 5'-[gamma-thio]triphosphate is a potent competitive inhibitor of AC in S49 lymphoma cell membranes. These data prompted us to study systematically the effects of 24 nucleotides on AC in S49 and Sf9 insect cell membranes, ACs 1, 2, 5, and 6, expressed in Sf9 membranes and purified catalytic subunits of membranous ACs (C1 of AC5 and C2 of AC2), sAC, sGC, EF, and ACT in the presence of MnCl(2). N-Methylanthraniloyl (MANT)-GTP inhibited C1.C2 with a K(i) of 4.2 nm. Phe-889 and Ile-940 of C2 mediate hydrophobic interactions with the MANT group. MANT-inosine 5'-[gamma-thio]triphosphate potently inhibited C1.C2 and ACs 1, 5, and 6 but exhibited only low affinity for sGC, EF, ACT, and G-proteins. Inosine 5'-[gamma-thio]triphosphate and uridine 5'-[gamma-thio]triphosphate were mixed G-protein activators and AC inhibitors. AC5 was up to 15-fold more sensitive to inhibitors than AC2. EF and ACT exhibited unique inhibitor profiles. At sAC, 2',5'-dideoxyadenosine 3'-triphosphate was the most potent compound (IC(50), 690 nm). Several MANT-adenine and MANT-guanine nucleotides inhibited sGC with K(i) values in the 200-400 nm range. UTP and ATP exhibited similar affinities for sGC as GTP and were mixed sGC substrates and inhibitors. The exchange of MnCl(2) against MgCl(2) reduced inhibitor potencies at ACs and sGC 1.5-250-fold, depending on the nucleotide and cyclase studied. The omission of the NTP-regenerating system from cyclase reactions strongly reduced the potencies of MANT-ADP, indicative for phosphorylation to MANT-ATP by pyruvate kinase. Collectively, AC isoforms and sGC are differentially inhibited by purine and pyrimidine nucleotides.

Activation of soluble guanylate cyclase through phosphorylation by protein kinase C in intact PC12 cells

Soluble guanylate cyclase was found to be phosphorylated by protein kinase C in intact PC12 pheochromocytoma cells. The phosphate incorporation into guanylate cyclase upon addition of phorbol 12-O-myristate 13-acetate (PMA) to PC12 cells in culture coincided with an increased intracellular cGMP level. A strong correlation between phosphate incorporation into guanylate cyclase and increased cGMP level was also observed by time-course and dose-response studies of the PMA effect, as well as when cells were treated with various phorbol esters and diacylglycerols or with various protein kinase C inhibitors. The cAMP system and the presence of extracellular Ca2+ were found not to be involved in guanylate cyclase phosphorylation. The phosphorylation and activation of guanylate cyclase by protein kinase C represent a possible mechanism whereby agonist-stimulation of receptors coupled to phosphoinositide hydrolysis induces cGMP synthesis.

Studies on cytosolic guanylate cyclase from human placenta

We have purified the soluble form of guanylate cyclase from human placenta greater than 2400-fold. The enzyme shared several characteristics with the enzyme purified from other sources including molecular mass and subunit composition, activation by divalent cations, inhibition by ATP and Michaelis constants. The enzyme, however, had a lower absorption maximum in the Soret region (417 +/- 1 nm) than the enzyme from other sources and was activated only one-fifth as much by nitric oxide as the bovine lung enzyme. It appears that the heme prosthetic group in the human placental enzyme may be hexa-coordinate and in the bovine lung enzyme the heme group may be penta-coordinate.

GABA release triggered by the activation of neuron-like non-NMDA receptors in cultured type 2 astrocytes is carrier-mediated

Kainate (KA), quisqualate (QA), and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) stimulated gamma-aminobutyric acid [3H]gamma-aminobutyric acid (GABA) release from cultured cerebellar type 2 astrocytes and from their bipotential precursors. The evoked release was prevented by the antagonist 6-cyano-2,3-dihydroxy-7-nitro-quinoxaline (CNQX). AMPA and QA applied together with KA at concentrations around or above their EC50S (20-50 microM) antagonized the stimulatory effect of KA on [3H]GABA release. On the other hand, the releasing action of KA was potentiated by concentrations of QA in the low micromolar range (2-5 microM), particularly when the concentration of KA was at the borderline of effectiveness (10 microM). KA and QA did not elevate intracellular cyclic GMP levels in astrocyte cultures, although guanylate cyclase was present in both type 2 and type 1 astrocytes. The inability of KA to elevate cyclic GMP levels in astrocytes was the only major difference in the behavior of this glutamate agonist between astroglial and neuronal cultures. The GABA transport inhibitor nipecotic acid or replacement of NaCl with LiCl abolished [3H]GABA uptake and also KA- and QA-induced release of preaccumulated [3H]GABA. Therefore, [3H]GABA was released from type 2 astrocytes and their progenitors through its Na(+)-dependent transport system, operating in an outward direction when the cells were depolarized by non-NMDA receptor agonists.

Cyclic GMP alterations in fetal rat cerebrum after global intrauterine ischemia: role of guanylate cyclase phosphorylation

Changes in the levels of cyclic AMP (cAMP) and cyclic GMP (cGMP) have been measured in brains of 20-day-old rat fetuses exposed to global intrauterine ischemia. Ischemia of different duration (0.5-30 minutes) did not alter the level of cAMP. In contrast, cGMP levels increased as a result of ischemia. This increase was seen even after a short period of ischemia (less than 5 minutes) and was maximal after 5 minutes, where a threefold increase could be observed. This stimulation was transient: after 30 min of ischemia, cGMP returned to the control level. Accumulation of cGMP can be related to the activation of guanylate cyclase, the activity of which is doubled after 15 minutes of ischemia. Immunoprecipitation of guanylate cyclase after in vivo labeling of the fetal brain with 32Pi revealed a threefold increase in the phosphorylation of the enzyme after 15 minutes of ischemia. The possible role of these modifications in cGMP metabolism during the course of ischemia is discussed.

Purification and characterization of the 180-kDa membrane guanylate cyclase containing atrial natriuretic factor receptor from rat adrenal gland and its regulation by protein kinase C

The original concept that cyclic GMP is one of the mediators of the hormone-dependent process of steroidogenesis has been strengthened by the characterization of a 180-kDa protein from rat adrenocortical carcinoma and rat and mouse testes. This protein appears to have an unusual characteristic of containing both the atrial natriuretic factor (ANF)-binding and guanylate cyclase activities, and appears to be intimately involved in the ANF-dependent steroidogenic signal transduction. In rat adrenal glands we now demonstrate: 1) the direct presence of a 180-kDa ANF-binding protein in GTP-affinity purified membrane fraction as evidenced by affinity cross-linking technique and by the Western blot analysis of the partially purified enzyme; 2) that the enzyme is biochemically and immunologically different from the soluble guanylate cyclase as there is no antigenic cross-reactivity of 180-kDa guanylate cyclase antibody with soluble guanylate cyclase; 3) in contrast to the soluble guanylate cyclase, the particulate enzyme is not stimulated by nitrite-generating compounds and hemin; and 4) protein kinase C inhibits both the basal and ANF-dependent guanylate cyclase activity and phosphorylates the 180-kDa guanylate cyclase. These results reveal the presence of a 180-kDa protein in rat adrenal glands and support the contention that: (a) this protein contains both the guanylate cyclase and ANF receptor; (b) the 180-kDa enzyme is coupled with the ANF-dependent cyclic GMP production; (c) the 180-kDa enzyme is biochemically distinct from the nonspecific soluble guanylate cyclase; and (d) there is a protein kinase C-dependent negative regulatory loop for the operation of ANF-dependent cyclic GMP signal pathway which acts via the phosphorylation of 180-kDa guanylate cyclase.

Regulation of cyclic AMP and cyclic GMP levels by adrenocorticotropic hormone in cultured neurons

The effect of adrenocorticotropic hormone (ACTH) on the intracellular concentration of cyclic nucleotides was studied in cultures of neurons from embryonic chick cerebral hemispheres. Incubation of neurons with ACTH(1-24) in the presence of phosphodiesterase inhibitor isobutylmethylxanthine resulted in a sustained increase in cyclic AMP while rise in cyclic GMP level was transient. The values obtained for half-maximal stimulation were 0.5 microM and 0.03 nM for cyclic AMP and cyclic GMP respectively. Concomitantly, ACTH(1-24) stimulated guanylate cyclase activity (half-maximal stimulation at 0.02 nM). These results suggest the existence of two distinct populations of ACTH receptors in neurons and provide the first evidence that cyclic GMP does mediate the action of ACTH in neurons.

Neuronal and glial localization of guanylate cyclase. Immunohistochemical evidence in cultured cells

A study has been carried out on the localization of guanylate cyclase employing cultured brain cells. Guanylate cyclase has been found to be located in neurons as well as in glial cells. This has been supported by the immunohistochemical as well as biochemical data.

Regulation of the macromolecular transport in the brain microvessels: the role of cyclic GMP

The possible effects of dibutyryl cyclic GMP (db-cGMP), the lipid-soluble derivative of cyclic GMP, on brain microvessels were studied by light microscopic detection of albumin and quantitative electron microscopy. Different concentrations (25, 50, 100 and 200 micrograms) of db-cGMP were given to adult rats by intracarotid infusion, while in the controls, the animals were infused either with Krebs-Ringer solution or with the same solution containing butyrate or cyclic GMP. In contrast to the controls, db-cGMP was found to be able to increase the permeability of brain microvessels to albumin in a dose-dependent manner. At the same time, the number of transport vesicles being indicative of the pinocytotic activity of the endothelial cells was also increased. These results indicate that, similarly to the cyclic AMP whose effects have been revealed earlier, the cyclic GMP-system may also be involved in the regulation of opening mechanisms of the blood-brain barrier.

Antigenic conservation of brain guanylate cyclase during evolution

A comparative study of brain guanylate cyclase from different animal species (including man, bird, fish and amphibian) has been performed using a specific antibody directed against soluble rat brain guanylate cyclase. Analyses were performed on supernatant fractions by the double-immunodiffusion test, by the protein blotting technique after SDS-polyacrylamide gel electrophoresis and by analytical isoelectric focusing on agarose allowing specific immunodetection of isoelectric patterns. Membrane-bound guanylate cyclase from rat brain and soluble guanylate cyclase from several rat tissues cross-reacted with the antibody. All the brain enzymes tested were found to be identical by double-immunodiffusion. The electrophoretic and isoelectrophoretic profiles of the different brain guanylate cyclases were found to exhibit many common features with some differences between mammalian and non-mammalian enzymes. In human brain, guanylate cyclase has been localized in glial and neuronal cells by immunohistochemistry. The results demonstrate that guanylate cyclase has been well conserved during the course of evolution and are consistent with the involvement of guanylate cyclase and cyclic GMP in basic cellular function.

Activation of purified guanylate cyclase by nitric oxide requires heme. Comparison of heme-deficient, heme-reconstituted and heme-containing forms of soluble enzyme from bovine lung

Bovine lung soluble guanylate cyclase was purified to apparent homogeneity in a form that was deficient in heme. Heme-deficient guanylate cyclase was rapidly and easily reconstituted with heme by reacting enzyme with hematin in the presence of excess dithiothreitol, followed by removal of unbound heme by gel filtration. Bound heme was verified spectrally and NO shifted the absorbance maximum in a manner characteristic of other hemoproteins. Heme-deficient and heme-reconstituted guanylate cyclase were compared with enzyme that had completely retained heme during purification. NO and S-nitroso-N-acetylpenicillamine only marginally activated heme-deficient guanylate cyclase but markedly activated both heme-reconstituted and heme-containing forms of the enzyme. Restoration of marked activation of heme-deficient guanylate cyclase was accomplished by including 1 microM hematin in enzyme reaction mixtures containing dithiothreitol. Preformed NO-heme activated all forms of guanylate cyclase in the absence of additional heme. Guanylate cyclase activation was observed in the presence of either MgGTP or MnGTP, although the magnitude of enzyme activation was consistently greater with MgGTP. The apparent Km for GTP in the presence of excess Mn2+ or Mg2+ was 10 microM and 85-120 microM, respectively, for unactivated guanylate cyclase. The apparent Km for GTP in the presence of Mn2+ was not altered but the Km in the presence of Mg2+ was lowered to 58 microM with activated enzyme. Maximal velocities were increased by enzyme activators in the presence of either Mg2+ or Mn2+. The data reported in this study indicate that purified guanylate cyclase binds heme and the latter is required for enzyme activation by NO and nitroso compounds.

Role of lipoxygenase in the O2-dependent activation of soluble guanylate cyclase from rat lung

Guanylate cyclase activity in rat lung supernatant fractions is stimulated 3-4 fold by aerobic incubation at 30 degrees C for approx. 30 min ('O2-dependent activation'). This stimulation was blocked by 20 microM-eicosa-5,8,11,14-tetraynoic acid (ETYA), an inhibitor of lipoxygenase and cyclo-oxygenase, but not by aspirin or indomethacin, which are cyclo-oxygenase inhibitors. The enzyme activator(s) is presumed to be the fatty acid hydroperoxide(s) formed by lipoxygenase. Removal of lipoxygenase from the supernatant fraction by chromatography on Amberlite XAD-4 also prevented activation, which was restored by the addition of soya-bean lipoxygenase. Bovine serum albumin prevented O2-dependent activation or activation by soya-bean lipoxygenase, through its ability to bind the unsaturated fatty acid substrate of lipoxygenase. The lipoxygenase in the supernatant fraction is inhibited by endogenous glutathione peroxidase plus reduced glutathione (GSH); removal of GSH de-inhibits lipoxygenase and activates guanylate cyclase. This was effected by autoxidation, by cumene hydroperoxide (with GSH peroxidase) and by titration with N-ethylmaleimide (NEM). Activation by NEM was inhibited by serum albumin or ETYA, as was activation by low concentrations (less than 50 microM) of cumene hydroperoxide. Activation by higher concentrations was not so inhibited; therefore, cumene hydroperoxide can also activate by a direct effect on guanylate cyclase. A hypothesis for physiological activation is proposed.

Activation of brain guanylate cyclase by phospholipase A2

Various pure snake venom phospholipases A2 were used for studying their effect on guanylate cyclase activity. All the phospholipases A2 tested were found to activate guanylate cyclase from a rat brain homogenate. It was shown that particulate guanylate cyclase was especially affected. Intact glial cells incubated in presence of phospholipase A2 showed also an increased guanylate cyclase activity, demonstrating that the phospholipase effect, observed in disrupted cells, occurs also at the cellular level. These results suggest that in intact cells membrane-bound phospholipase A2 activity could be involved in the modulation of the cellular cyclic GMP content.