Photoaffinity labelling of central-nervous-system myelin. Evidence for an endogenous type I cyclic AMP-dependent kinase phosphorylating the larger subunit of 2'',3''-cyclic nucleotide 3''-phosphodiesterase.

Endogenous cyclic AMP-stimulated phosphorylation of a 49700-Mr Wolfgram protein component in rabbit central nervous system was investigated by using photoaffinity labelling and 2',3'-cyclic nucleotide 3'-phosphodiesterase activity staining after electroblotting on to nitrocellulose paper. Photoaffinity labelling with 8'-azidoadenosine 3',5'-cyclic monophosphate showed a cyclic AMP-binding protein that appeared to be intrinsic to the myelin membrane and appeared to represent the R-subunit of a type I cyclic AMP-dependent protein kinase. This photoaffinity-labelled protein was of larger apparent Mr than the protein showing cyclic AMP-stimulated phosphorylation. Blotting of one-dimensional sodium dodecyl sulphate/polyacrylamide-gel electrophoretograms followed by staining for 2',3'-cyclic nucleotide 3'-phosphodiesterase activity showed two activity bands corresponding to the two components of the Wolfgram protein doublet. Cyclic AMP-stimulated protein phosphorylation corresponded to the upper component of this doublet. Electroblotting of two-dimensional non-equilibrium pH-gradient electrophoretograms also showed co-migration of cyclic AMP-stimulated protein phosphorylation with enzyme activity. It is proposed that central-nervous-system myelin contains an endogenous type I cyclic-AMP dependent protein kinase that phosphorylates the larger subunit of 2',3'-cyclic nucleotide 3'-phosphodiesterase.     

Cyclic nucleotide content of tobacco BY-2 cells

The cyclic nucleotide content of cultured tobacco bright yellow-2 (BY-2) cells was determined, after freeze-killing, perchlorate extraction and sequential chromatography, by radioimmunoassay. The identities of the putative cyclic nucleotides, adenosine 3',5'-cyclic monophosphate (cyclic AMP), guanosine 3',5'-cyclic monophosphate (cyclic GMP) and cytidine 3',5'-cyclic monophosphate (cyclic CMP) were unambiguously confirmed by tandem mass spectrometry. The potential of BY-2 cell cultures as a model system for future investigations of cyclic nucleotide function in higher plants is discussed.     

Induction of an extracellular cyclic nucleotide phosphodiesterase as an accessory ribonucleolytic activity during phosphate starvation of cultured tomato cells

During growth under conditions of phosphate limitation, suspension-cultured cells of tomato (Lycopersicon esculentum Mill.) secrete phosphodiesterase activity in a similar fashion to phosphate starvation-inducible ribonuclease (RNase LE), a cyclizing endoribonuclease that generates 2':3'-cyclic nucleoside monophosphates (NMP) as its major monomeric products (T. Nürnberger, S. Abel, W. Jost, K. Glund [1990] Plant Physiol 92: 970-976). Tomato extracellular phosphodiesterase was purified to homogeneity from the spent culture medium of phosphate-starved cells and was characterized as a cyclic nucleotide phosphodiesterase. The purified enzyme has a molecular mass of 70 kD, a pH optimum of 6.2, and an isoelectric point of 8.1. The phosphodiesterase preparation is free of any detectable deoxyribonuclease, ribonuclease, and nucleotidase activity. Tomato extracellular phosphodiesterase is insensitive to EDTA and hydrolyzes with no apparent base specificity 2':3'-cyclic NMP to 3'-NMP and the 3':5'-cyclic isomers to a mixture of 3'-NMP and 5'-NMP. Specific activities of the enzyme are 2-fold higher for 2':3'-cyclic NMP than for 3':5'-cyclic isomers. Analysis of monomeric products of sequential RNA hydrolysis with purified RNase LE, purified extracellular phosphodiesterase, and cleared -Pi culture medium as a source of 3'-nucleotidase activity indicates that cyclic nucleotide phosphodiesterase functions as an accessory ribonucleolytic activity that effectively hydrolyzes primary products of RNase LE to substrates for phosphate-starvation-inducible phosphomonoesterases. Biosynthetical labeling of cyclic nucleotide phopshodiesterase upon phosphate starvation suggests de novo synthesis and secretion of a set of nucleolytic enzymes for scavenging phosphate from extracellular RNA substrates.     

Higher-plant cyclic nucleotide phosphodiesterases. Resolution, partial purification and properties of three phosphodiesterases from potato tuber.

1. Three phosphodiesterases that are capable of hydrolysing 3':5'-cyclic nucleotides were purified from potato tubers. 2. The phosphodiesterases were fractionated by (NH4)2SO4 precipitation and CM-cellulose chromatography. The phosphodiesterases were resolved from each other and further purified by gel filtration in high- and low-ionic-strength conditions. 3. All three enzymes lacked significant nucleotidase activity. 4. Enzymes I and II had mol. wts. 240,000 and 80,000 respectively, determined by gel filtration, whereas enzyme III showed anomalous behaviour on gel filtration, behaving as a high- or low-molecular-weight protein in high- or low-ionic-strength buffers respectively. 5. All enzymes hydrolysed 2':3'-cyclic nucleotides as well as 3':5'-cyclic nucleotides. The enzymes also had nucleotide pyrophosphatase activity, hydrolysing NAD+ and UDP-glucose to various extents. Enzymes I and II hydrolyse cyclic nucleotides at a greater rate than NAD+, whereas enzyme III hydrolyses NAD+ at a much greater rate than cyclic nucleotides. All three enzymes hydrolysed the artificial substrate bis-(p-nitro-phenyl) phosphate. 6. The enzymes do not require the addition of bivalent cations for activity. 7. Both enzymes I and II have optimum activity at pH6 with 3':5'-cyclic AMP and bis-(p-nitrophenyl) phosphate as substrates. The products of 3':5'-cyclic AMP hydrolysis were 3'-AMP and 5'-AMP, the ratio of the two products being different for each enzyme and varying with pH. 8. Theophylline inhibits enzymes I and II slightly, but other methyl xanthines have little effect. Enzymes I and II were competitively inhibited by many nucleotides containing phosphomonoester and phosphodiester bonds, as well as by Pi. 9. The possible significance of these phosphodiesterases in cyclic nucleotide metabolism in higher plants is discussed.     

Purification and characterization of wheat germ 2',3'-cyclic nucleotide 3'-phosphodiesterase

The 2',3'-cyclic nucleotide 3'-phosphodiesterase which hydrolyzes nucleoside 2',3'-cyclic phosphates (N greater than p) to nucleoside 2'-phosphates has been purified 16,000-fold to near homogeneity from wheat germ. The purified enzyme is a single polypeptide with a molecular weight of 23,000-24,000. It has a pH optimum of 7.0. The apparent Km values for A greater than p, G greater than p, C greater than p, and U greater than p are 13.1, 9.2, 25.2, and 25.3 mM, respectively. Vmax values for A greater than p, G greater than p, C greater than p, and U greater than p are 2090, 280, 2140, and 600 mumol/min/mg of purified protein, respectively. Wheat germ 2',3'-cyclic nucleotide 3'-phosphodiesterase does not hydrolyze 2',3'-cyclic esters in cyclic phosphate-terminated oligoribonucleotides or in nucleoside 5'-phosphate, 2',3'-cyclic phosphate (pN greater than p). This is in contrast to the 3'-phosphodiesterase activity associated with a wheat germ RNA ligase which hydrolyzes cyclic phosphate-terminated oligonucleotides and pN greater than p substrates much more efficiently than nucleoside 2',3'-cyclic phosphates. The enzyme characterized in this work appears to be the only known 2',3'-cyclic nucleotide 3'-phosphodiesterase specific for 2',3'-cyclic mononucleotides.     

Identification of 2':3'-cyclic nucleotide 3'-phosphodiesterase in the vertebrate retina

The enzyme, 2':3'-cyclic nucleotide 3'-phosphodiesterase (2':3'-cNMP-3'-ase) has been used as a marker in the nervous system for the presence of myelin membrane or myelin-producing glial cells. In this study, goldfish and bovine neural retinas are found to have high levels of such a diesterase activity. Analysis of retinal tissue incubated with 2':3'-cAMP shows only 2'-AMP as the reaction product, indicating the selective hydrolysis of the cyclic nucleotide. Microdissection of the goldfish retina demonstrates the highest 2':3'-cNMP-3'-ase activity in the region of the photoreceptors. A fraction enriched in bovine rod outer segments has about a 5-fold increase in specific enzyme activity when compared to whole retina preparations. These data suggest that 2':3'-cNMP-3'-ase is either closely associated with or is an intrinsic feature of vertebrate photoreceptor elements. The retina, which contains this enzyme, may serve as a model to investigate the influence of 2':3'-cyclic nucleotides on a function of the nervous system.     

cGMP、cAMP及两种cAMP衍生物对大鼠肝细胞核体外转录系统的影响

 本文介绍了以α-鹅膏蕈碱和低浓度KCl为手段建立了RNA聚合酶Ⅰ、RNA聚合酶Ⅱ活性的细胞核转录系统进而研究了cGMP、cAMP、cAMP丁酯及cAMP硫代环磷酰二乙胺对大鼠肝细胞核中RNA聚合酶Ⅰ与Ⅱ活性的影响。结果显示cGMP可以提高RNA聚合酶Ⅰ的活性;cAMP主要提高RNA聚合酶Ⅱ的活极,而cAMP分子结构变化产生的丁酯及硫代环磷酰二乙胺衍生物可增强cAMP的这种作用,为深入研究cAMP的构效关系提供了实验依据。     

A RADIOISOTOPIC METHOD FOR MEASURING THE FORMATION OF ADENOSINE 3'',5''-CYCLIC MONOPHOSPHATE IN INCUBATED SLICES OF BRAIN

Abstract— A simple and sensitive method for measuring the effect of neurohormones and other chemical agents on the formation of adenosine 3',5'-cyclic monophosphate (cyclic 3',5'-AMP) in incubated slices of brain was developed. The principle of the method depends on pulse-labelling of adenosine-5'-triphosphate in slices of brain with [8-¹⁴C]adenine, followed by incubation in a medium containing the test substance, separation by thin-layer chromatography of the cyclic nucleotide formed in the slices, and radioassay. The purity of the cyclic nucleotide was confirmed by chromatography in a variety of systems and by hydrolysis with a specific, bovine-heart phosphodiesterase. The method was used to study the effect of histamine, norepinephrine, and adenosine on the accumulation of adenosine 3',5'-cyclic monophosphate in incubated slices of brain.     

Isolation of a 5.3-S calmodulin-deficient 3'':5''-cyclic nucleotide phosphodiesterase from cardiac muscle.

1. In the presence of Ca2+, a 5.3-S 3':5'-cyclic nucleotide phosphodiesterase (EC 3.1.4.17) from bovine ventricle was isolated and purified by (NH4)2SO4 precipitation and DEAE-cellulose and Affi-Gel Blue chromatography. The enzyme activity was enriched 800-fold by these procedures. 2. Sucrose-density gradient centrifugation, gel filtration and non-denaturing polyacrylamide-gel electrophoresis resolved a single enzyme species with an Mr of 89 000. 3. Sodium dodecyl sulphate/polyacrylamide-gel electrophoresis of the purified enzyme demonstrated a prominent protein band at Mr 59000 and a minor band of Mr 28000. Calmodulin was not detected. 4. The hydrolysis of micromolar concentrations of 3':5'-cyclic guanosine monophosphate (cyclic GMP) but not 3':5'-cyclic adenosine monophosphate (cyclic AMP) was stimulated by calmodulin. 5. Anomalous biphasic kinetics plots were observed for both the catalysis of cyclic AMP and cyclic GMP hydrolysis. Kinetic plots became linear in the presence of calmodulin. 6. After several months of storage at -20 degrees C, the 5.3-S enzyme was transformed into a 6.2-S cyclic GMP-specific enzyme and a 4.4-S non-specific form.     

Neurotrophic control of cyclic nucleotide levels during muscle differentiation in cell culture.

The effects of chick brain-spinal cord extract on morphological development and cyclic nucleotide levels of cultured chick embryo skeletal muscle cells were determined. It had previously been shown that the extract stimulated morphological differentiation, protein synthesis, and choliniesterase activity of muscle cells. Myoblasts fused earlier and an increase in number as well as diameter of myotubes were seen in the extract treated cultures. Cyclic nucleotides levels were higher (almost twice the controls for both adenosine 3',5'-cyclic monophosphate and guanosine 3',5'-cyclic monophosphate) and preceded their occurrence in the control cultures. It was suggested that factor(s) in the extract interact with membrane receptor(s) to alter nucleotide levels which, in turn, allow the effects to be expressed.     

A further study on the regulation of cyclic nucleotide phosphodiesterase activity in neuroblastoma cells: effect of growth.

Adenosine 3',5'-cyclic monophosphate (cyclic AMP) phsophodiesterase activity in mouse neuroblastoma cells in culture markedly increased during exponential growth and reached a maximal level at confluency; whereas guanosine 3'5'-cyclic monophosphate (cyclic GMP) phosphodiesterase activity only slightly but significantly increased under a similar experimental condition. The increase in cyclic AMP phosphodiesterase activity was blocked by both cycloheximide and dactinomycin, whereas the increase in cyclic GMP phosphodiesterase was blocked by only cycloheximide. When the confluent cells were replated at low density, the cyclic nucleotide phosphodiesterase activity decreased; however, when they were plated at high cell density which equaled confluency, the enzyme activity did not decrease. Unlike cyclic AMP phosphodiesterase activity, cyclic GMP phosphodiesterase activity did not change significantly in prostaglandin E1-treated cells, but decreased in cells treated with the inhibitor of phosphodiesterase. Like cyclic AMP phosphodiesterase activity, cyclic GMP phosphodiesterase activity also did not change in cells treated with serum-free medium, X-irradiation, sodium butyrate and 6-thioguanine.     

2',3'-cyclic nucleotide 2'-phosphodiesterase from Fusarium culmorum

We describe the properties of a 2',3'-cyclic nucleotide 2'-phosphodiesterase (EC 3.1.4.16), found in Fusarium culmorum, which hydrolyzes nucleoside 2',3'-cyclic monophosphates to nucleoside 3'-phosphates. In contrast with a similar enzyme found in bacteria, the Fusarium enzyme does not exhibit nucleotidase activity and does not show a requirement for metal ions, but is inhibited by micromolar concentrations of Cu++ and Zn++, and is very stable to heat. This cyclic phosphodiesterase hydrolyzes the four major nucleoside 2',3'-cyclic monophosphates and has greater affinity for purine (Kms for Ado-2',3'-P = 0.3 mM and for Guo-2',3'-P = 0.1 mM) than for pyrimidine nucleotides (Kms for Cyd-2',3'-P = 0.6 mM and for Urd-2',3'-P = 2 mM). The respective Vmax for Urd-2',3'-P; Cyd-2',3'-P; Ado-2',3'-P; and Guo-2',3' are 100:45:16:5. The efficacy of the phosphodiesterase to hydrolyze the four major 2',3' cyclic nucleotides (based on the relative values of Vmax/Km) is not significantly different. The Fusarium enzyme differs from a previously described 2',3' cyclic phosphodiesterase from Neurospora, in that it is inactive on 3',5'-nucleoside monophosphates and nucleoside 2' or 3' phosphates.     

Prebiotic phosphorylation of nucleosides in formamide.

A possible prebiotic phosphorylation method has been investigated in which formamide served as the reaction medium. Nucleotides and nucleotide derivatives were formed when nucleosides were allowed to react with different orthophosphate, hydrogen phosphate or dihydrogen phosphate salts or with different condensed phosphate salts. The reaction products obtained from the phosphorylation of adenosine were 2'3' and 5'-AMPs, 2',5' and 3',5'-ADPs and 2',3'-cyclic AMP. The extent of phosphorylation in formamide exceeded 50% under favorable conditions after 15 days at 70 degrees. The acidic dihydrogen phosphates and condensed hydrogen phosphates proved to be the best phosphorylating agents. The presence of water in the medium decreased the yield of nucleotide derivatives, but some phosphorylation of adenosine was detected using dihydrogen phosphate in formamide containing water. The phosphorylation reactions were also observed for deoxynucleosides. Little decompression of the nucleosides was detected during the reaction time needed to form nucleotide derivatives. The facility with which phosphorylation takes place in formamide under very mild conditions may justify further studies both of prebiotic phosphorylation and synthetic phosphorylation using this solvent.     

A rapid direct assay of 3'-5' cyclic nucleotide phosphodiesterase activity using chromatography on immobilized acriflavin

A chromatographic method using immobilized acriflavin has been developed for the separation of unreacted cyclic nucleotides from their corresponding 5'-nucleotides, in a direct assay of 3'-5' cyclic nucleotide phosphodiesterase using [3H]-cyclic nucleotides as substrate. The method based on the so-called charge-transfer overlap recognition between flavin and indol rings, provides a rapid (15-20 min) and sensitive elution of [3H]-5'nucleotides with high recovery (up to 98%) and low blanks, while [3H]-cyclic nucleotides are retarded on the column. By this method, the formation of some secondary products by purine metabolizing enzymes such as 5'-nucleotidases, nucleosidases and/or deaminases is taken into account, using [14C]-5'-AMP thus allowing an accurate determination of phosphodiesterase activity in any preparations.     

Solution structure of the catalytic domain of RICH protein from goldfish

Regeneration-induced CNPase homolog (RICH) is an axonal growth-associated protein, which is induced in teleost fish upon optical nerve injury. RICH consists of a highly acidic N-terminal domain, a catalytic domain with 2',3'-cyclic nucleotide 3'-phosphodiesterase (CNPase) activity and a C-terminal isoprenylation site. In vitro RICH and mammalian brain CNPase specifically catalyze the hydrolysis of 2',3'-cyclic nucleotides to produce 2'-nucleotides, but the physiologically relevant in vivo substrate remains unknown. Here, we report the NMR structure of the catalytic domain of goldfish RICH and describe its binding to CNPase inhibitors. The structure consists of a twisted nine-stranded antiparallel beta-sheet surrounded by alpha-helices on both sides. Despite significant local differences mostly arising from a seven-residue insert in the RICH sequence, the active site region is highly similar to that of human CNPase. Likewise, refinement of the catalytic domain of rat CNPase using residual dipolar couplings gave improved agreement with the published crystal structure. NMR titrations of RICH with inhibitors point to a similar catalytic mechanism for RICH and CNPase. The results suggest a functional importance for the evolutionarily conserved phosphodiesterase activity and hint of a link with pre-tRNA splicing.     

Repressible extracellular phosphodiesterases showing cyclic 2'',3''- and cyclic 3'',5''-nucleotide phosphodiesterase activities in Neurospora crassa.

Two molecular species of repressible extracellular phosphodiesterases showing cyclic 2',3'- and cyclic 3',5'-nucleotide phosphodiesterase activities were detected in mycelial culture media of wild-type Neurospora crassa and purified. The two molecular species were found to be monomeric and polymeric forms of an enzyme constituted of identical subunits having molecular weights of 50,000. This enzyme had the same electrophoretic mobility as repressible acid phosphatase. The enzyme designated repressible cyclic phosphodiesterase showed pH optima of 3.2 to 4.0 with a cyclic 3',5'-AMP substrate and 5.0 to 5.6 with a cyclic 2',3'-AMP substrate. Repressible cyclic phosphodiesterase was activated by MnCl2 and CoCl2 with cyclic 2',3'-AMP as substrate and was slightly activated by MnCl2 with cyclic 3',5'-AMP. The enzyme hydrolyzed cyclic 3',5'- and cyclic 2',3'-nucleotides, in addition to bis-rho-nitrophenyl phosphate, but not certain 5' -and 3'-nucleotides. 3'-GMP and 3'-CMP were hydrolyzed less efficiently. Mutant strains A1 (nuc-1) and B1 (nuc-2), which cannot utilize RNA or DNA as a sole source of phosphorus, were unable to produce repressible cyclic phosphodiesterase. The wild type (74A) and a heterocaryon between strains A1 and B1 produced the enzyme and showed growth on orthophosphate-free media containing cyclic 2',3'-AMP or cyclic 3',5'-AMP, whereas both mutants showed little or no growth on these media.     

Extraction, purification, identification and metabolism of 3'',5''-cyclic UMP, 3'',5''-cyclic IMP and 3'',5''-cyclic dTMP from rat tissues.

The large-scale extraction and partial purification of endogenous 3',5'-cyclic UMP, 3',5'-cyclic IMP and 3',5'-cyclic dTMP are described. Rat liver, kidney, heart, spleen and lung tissues were subjected to a sequential purification procedure involving freeze-clamping, perchlorate extraction, alumina and Sephadex ion-exchange chromatography and preparative electrophoresis. The samples thus obtained co-chromatographed with authentic cyclic UMP, cyclic IMP and cyclic dTMP on t.l.c. and h.p.l.c. and the u.v. spectra of the extracted samples were identical with those of the standards. Fast atom bombardment of the three cyclic nucleotide standards yielded mass spectra containing a molecular protonated ion in each case; mass-analysed ion kinetic-energy spectrometry ('m.i.k.e.s') of these ions produced a spectrum unique to the parent cyclic nucleotide. The extracted putative cyclic UMP, cyclic IMP and cyclic dTMP each produced a m.i.k.e.s. identical with that obtained with the corresponding cyclic nucleotide standard. Rat liver, heart, kidney, brain, intestine, spleen, testis and lung protein preparations were each found capable of the synthesis of cyclic UMP, cyclic IMP and cyclic dTMP from the corresponding nucleoside triphosphate, of the hydrolysis of these cyclic nucleotides and of their binding, with the exception that cyclic dTMP was not synthesized by the kidney preparation.     

Hormonal Control of Nuclease Level in Excised Avena Leaf Tissues1

In excised Avena leaves, kinetin and benzyladenine decreased,while abscisic acid and benzimidazole increased the over-allnuclease level. Significant effects were observed as early as2 h after treatment. Not all the nucleases of the Avena leafwere affected by the growth regulators. Changes in over-allnuclease activity were accounted for almost entirely by changesin the amount of a relatively purine-specific endo-ribonuclease,which produces 2',3'-cyclic phosphates as breakdown products.Slight changes induced by the growth regulators were also detectedin the amount of a sugar non-specific endo-nuclease which produces5'-nucleotides and has a relative specificity for adenylic acid.The level of an alkaline phosphodiesterase, an exo-nucleasewhich produces 5'-nucleotides, was not affected by any of thegrowth regulators tested. Gibberellic acid and indol-3yl-aceticacid did not influence the level of Avena nucleases.     

2',3'-cyclic nucleotide 3'-phosphohydrolase in rat liver mitochondrial membranes

2',3'-Cyclic nucleotide 3'-phosphohydrolase (nucleoside-2':3'-cyclic-phosphate 2'-nucleotidohydrolase, EC 3.1.4.37) activity has been demonstrated in rat liver mitochondria. The enzyme was localized in both the outer and inner mitochondrial membranes but was absent from the intermembrane space and matrix. The mitochondrial (cyclic nucleotide) phosphohydrolase was activated by freezing and thawing and by treatment with digitonin or detergents. It is suggested that (cyclic nucleotide) phosphohydrolase is an integral membrane protein which is buried to a significant degree within the membrane. Atractyloside was found to be a noncompetitive inhibitor of the enzyme both in intact mitochondria and in preparations of the mitochondrial membranes. The enzyme substrate, 2',3'-cyclic adenosine monophosphate, had no effect on the oxidation of exogenous beta-hydroxybutyrate or succinate by intact mitochondria. These findings suggest that 2',3'-cyclic nucleotide 3'phosphohydrolase is more widely distributed than was previously thought and that the enzyme may play a fundamental role in membranes, independent of their specialized structure or functions.     

Human blood platelet 3': 5'-cyclic nucleotide phosphodiesterase. Isolation of low-Km and high-Km phosphodiesterase.

Human blood platelet contained at least three kinetically distinct forms of 3': 5'-cyclic nucleotide phosphodiesterase (3': 5'-cyclic-AMP 5'-nucleotidohydrolase, EC 3.1.4.17) (F I, F II, and F III) which were clearly separated by DEAE-cellulose column chromatography. Although a few properties of the platelet phosphodiesterases such as their substrate affinities and DEAE-cellulose profile resembled somewhat those of the three 3': 5'-cyclic nucleotide phosphodiesterase in rat liver reported by Russell et al. [10], there were pronounced differences in some properties between the platelet and the liver enzymes: (1) the platelet enzymes hydrolyzed both cyclic nucleotides and lacked a highly specific cyclic guanosine 3': 5'-monophosphate (cyclic GMP) phosphodiesterase and (2) kinetic data of the platelet enzymes indicated that cyclic adenosine 3': 5'-monophosphate (cyclic AMP) and cyclic GMP interact with a single catalytic site on the enzyme. F I was a cyclic nucleotide phosphodiesterase with a high Km for cyclic AMP and a negatively cooperative low Km for cyclic GMP. F II hydrolyzed cyclic AMP and cyclic GMP about equally with a high Km for both substrates. F III was low Km phosphodiesterase which hydrolyzed cyclic AMP faster than cyclic GMP. Each cyclic nucleotide acted as a competitive inhibitor of the hydrolysis of the other nucleotide by these three fractions with Ki values similar to the Km values for each nucleotide suggesting that the hydrolysis of both cyclic AMP and cyclic GMP was catalyzed by a single catalytic site on the enzyme. However, cyclic GMP at low concentration (below 10 muM) was an activator of cyclic AMP hydrolysis by F I. Papaverine and EG 626 acted as competitive inhibitors of each fraction with virtually the same Ki value in both assays using either cyclic AMP or cyclic GMP as the substrate. The ratio of cyclic AMP hydrolysis to cyclic GMP hydrolysis by each fraction did not vary significantly after freezing/thawing or heat treatment. These facts also suggest that both nucleotides were hydrolyzed by the same catalytic site on the enzyme. The differences in apparent Ki values for inhibitors such as cyclic nucleotides, papaverine and EG 626 would indicate that three enzymes were different from each other. Centrifugation in a continuous sucrose gradient revealed sedimentation coefficients F I and II had 8.9 S and F III 4.6 S. The molecular weight of these forms, determined by gel filtration on a Sepharose 6B column, were approx. 240 000 (F I and II) and 180 000 (F III). F III was purified extensively (70-fold) from homogenate, with a recovery of approximately 7%.