Subclasses of cyclic AMP-specific phosphodiesterase in left ventricular muscle and their involvement in regulating myocardial contractility

Ventricular muscle contains a low Km, cyclic AMP-specific form of phosphodiesterase (PDE III), which is believed to represent the site of action for several of new cardiotonic agents including imazodan (CI-914), amrinone, cilostamide, and enoximone. However, species differences in the inotropic response to these agents have raised questions about the relationship between PDE III inhibition and cardiotonic activity. The present study demonstrates that these differences can be accounted for by the presence of two subclasses of PDE III in ventricular muscle and variations in the intracellular localization of these two enzymes. For these experiments, PDE III was initially isolated from canine, guinea pig, and rat left ventricular muscle. The results demonstrate that canine left ventricular muscle contains two functional subclasses of PDE III: an imazodan-sensitive form, which is membrane bound, and an imazodan-insensitive form, which is soluble. Although only weakly inhibited by imazodan, this latter enzyme is potently inhibited by the selective PDE III inhibitors, Ro 20-1724 and rolipram. Guinea pig ventricular muscle also contains the imazodan-sensitive subclass of PDE III. Unlike canine left ventricle, however, thi enzyme is soluble in the guinea pig. No membrane-bound subclass of PDE III was observed in the guinea pig. Rat left ventricle possesses only the soluble form of PDE III, which apparently represents a mixture of the imazodan-sensitive and imazodan-insensitive subclasses of PDE III. Measurement of in vivo contractility in these three species showed that imazodan exerts a potent positive inotropic effect only in the dog, in which the imazodan-sensitive subclass of PDE III is membrane bound.(ABSTRACT TRUNCATED AT 250 WORDS)     

Cellular distribution of phosphodiesterase isoforms in rat cardiac tissue.

We have resolved multiple forms of cyclic nucleotide phosphodiesterase (PDE) in whole rat ventricle and in isolated rat ventricular myocytes by use of anion-exchange high-performance liquid chromatography. One major form, the soluble calmodulin-stimulated PDE, is apparently absent from isolated myocytes. We discern four peaks of PDE activity (designated A-D in the order of their elution) in a soluble fraction obtained from whole rat ventricle. Peak A is stimulated twofold to threefold by the addition of calcium and calmodulin (Ca2+/CalM) and preferentially hydrolyzes cGMP over cAMP (in the presence of Ca2+/CalM, KmcGMP = 1.5 microM, KmcAMP = 17 microM). Peak B has similar affinities for both cAMP and cGMP (half-maximum velocities achieved at 30 microM substrate) and demonstrates positive cooperativity with cAMP but not with cGMP. The hydrolysis of cAMP by peak B is stimulated by cGMP at substrate concentrations up to 20 microM; the maximum effect is seen at 1 microM cAMP (25-fold stimulation by 3 microM cGMP). This pattern of stimulation by cGMP results from two kinetic changes: an increase in the enzyme's apparent affinity for cAMP (apparent KmcAMP decreases from 33 to 11 microM) and the abolition of positive cooperativity. Peaks C and D selectively hydrolyze cAMP, are not stimulated by Ca2+/CalM or cGMP, and differ in their affinities for substrate (peak C, apparent KmcAMP = 7.2 microM; peak D, 0.44 microM). In addition, peak D is much more sensitive than peak C to inhibition by cGMP, cilostamide, rolipram, and milrinone. Ro20-1724 is a slightly more potent inhibitor of peak D than of peak C. Peak D appears to consist of two different enzyme activities, one inhibited by cGMP, cilostamide, and cardiotonic drugs and the other potently inhibited by rolipram. In contrast to whole ventricle, the soluble fraction of isolated rat ventricular myocytes lacks peak A. Three major peaks in myocytes are entirely analogous to peaks B, C, and D of whole ventricle in terms of the NaCl concentration at which they elute, substrate affinities, and stimulation or inhibition by various drugs and effectors. We conclude that the soluble Ca2+/CalM-stimulated PDE in whole rat ventricle is present in nonmyocyte cells.     

Antilipolytic action of insulin: role of cAMP phosphodiesterase activation

Exposure of 3T3-L1 adipocytes to 1 nM insulin for 10 min results in activation of particulate cAMP phosphodiesterase and suppression of lipolysis stimulated by 10 nM isoproterenol. When lipolysis was increased by cilostamide, a selective inhibitor of the particulate phosphodiesterase, the antilipolytic effect of insulin was not observed. Insulin did suppress lipolysis stimulated by Ro 20-1724, an inhibitor of soluble cAMP phosphodiesterase activity. Cilostamide did not interfere with insulin stimulation of glucose uptake, nor did it have any direct effect on cAMP-dependent protein kinase. Thus, inhibition of particulate but not soluble cAMP phosphodiesterase blocked the antilipolytic effect of insulin. Our findings support the idea that insulin inhibits lipolysis, perhaps in large part by activating particulate "low Km" cAMP phosphodiesterase, which seems to be functionally closely coupled with the hormone-sensitive lipase-regulatory system influencing primarily a pool of cAMP utilized by the relevant protein kinase.     

Biochemical aspects of inhibition of cardiovascular low (Km) cyclic adenosine monophosphate phosphodiesterase

Multiple isozymes of cyclic nucleotide phosphodiesterase (PDE) exist in mammalian cells. At least 5 major types of PDE isozymes have been identified; they differ by substrate affinity, maximal activity, intracellular regulation or mechanism of pharmacologic inhibition. A low Michaelis constant (Km) cyclic adenosine monophosphate (cAMP) PDE, whose activity is inhibited by submicromolar concentrations of cyclic guanosine monophosphate and stimulated by cAMP-mediated phosphorylation, is present in both cardiac muscle and vascular smooth muscle. This PDE isozyme (referred to as peak IIIc PDE) is sensitive to selective inhibition by amrinone, milrinone, imazodan, CI-930, piroximone, and numerous other PDE inhibitors. The subcellular distribution of cardiac PDE IIIc varies according to species; it is found in the soluble fraction of guinea pig myocardium, in the particulate fraction of canine myocardium, and in both fractions of primate (simian and human) myocardium. Another PDE isozyme, which is sensitive to inhibition by rolipram and is less sensitive to inhibition by PDE IIIc inhibitors, is found in cardiac muscle of some species (i.e., soluble fractions of rat and canine myocardium) and is apparently not related to direct regulation of positive inotropy. Both positive inotropy and vasorelaxation by milrinone and other PDE IIIc inhibitors can be linked to inhibition of PDE IIIc and activation of the cAMP system. These significant relations are similar to those obtained for other cAMP-related positive inotrope/vasodilators (such as beta-adrenoreceptor agonists). Moreover, an increased rate of ventricular relaxation (lusitropy), which is apparent with PDE IIIc inhibitors, may also be attributable to activation of the cAMP system.(ABSTRACT TRUNCATED AT 250 WORDS)     

Heat-stable low molecular weight form of phosphodiesterases from bovine pineal gland.

The 105,000 X g supernatant fraction of bovine pineal gland contains a phosphodiesterase activity that hydrolyzes both cyclic AMP and cyclic GMP. The rate of hydrolysis is 4-5 times greater with cyclic GMP as substrate than with cyclic AMP. Chromatography of supernatant fraction on Sephadex G-150 resolves phosphodiesterase activity into two fractions designated PDE I and PDE II. These are distinguishable on the basis of their molecular size, substrate specificity, and kinetic parameters. PDE I hydrolyzes cyclic GMP at a faster rate than cyclic AMP and has a molecular weight of 163,000. PDE II appears to be a smaller protein with a molecular weight of 24,400 and is specific for cyclic AMP. PDE I has apparent Km values of 83 and 53 micron for cyclic AMP and cyclic GMP, respectively, whereas PDE II exhibits an apparent Km value of 330 micron for cyclic AMP. With subsaturating concentrations of cyclic AMP as substrate, the phosphodiesterase activity of PDE I is inhibited by the addition of cyclic GMP. However, PDE II activity remains unaffected by cyclic GMP even at concentrations up to 125 micron. PDE II appears to be thermostable, losing only 20% of its activity on heating at 80 degrees for 2 min. Similar treatment completely abolishes the enzyme activity of PDE I.     

Regulation by phosphodiesterase isoforms of protein kinase A-mediated attenuation of myocardial protein kinase D activation

Protein kinase D (PKD) targets several proteins in the heart, including cardiac troponin I (cTnI) and class II histone deacetylases, and regulates cardiac contraction and hypertrophy. In adult rat ventricular myocytes (ARVM), PKD activation by endothelin-1 (ET1) occurs via protein kinase Cε and is attenuated by cAMP-dependent protein kinase (PKA). Intracellular compartmentalisation of cAMP, arising from localised activity of distinct cyclic nucleotide phosphodiesterase (PDE) isoforms, may result in spatially constrained regulation of the PKA activity that inhibits PKD activation. We have investigated the roles of the predominant cardiac PDE isoforms, PDE2, PDE3 and PDE4, in PKA-mediated inhibition of PKD activation. Pretreatment of ARVM with the non-selective PDE inhibitor isobutylmethylxanthine (IBMX) attenuated subsequent PKD activation by ET1. However, selective inhibition of PDE2 [by erythro-9-(2-hydroxy-3-nonyl) adenine, EHNA], PDE3 (by cilostamide) or PDE4 (by rolipram) individually had no effect on ET1-induced PKD activation. Selective inhibition of individual PDE isoforms also had no effect on the phosphorylation status of the established cardiac PKA substrates phospholamban (PLB; at Ser16) and cTnI (at Ser22/23), which increased markedly with IBMX. Combined administration of cilostamide and rolipram, like IBMX alone, attenuated ET1-induced PKD activation and increased PLB and cTnI phosphorylation, while combined administration of EHNA and cilostamide or EHNA and rolipram was ineffective. Thus, cAMP pools controlled by PDE3 and PDE4, but not PDE2, regulate the PKA activity that inhibits ET1-induced PKD activation. Furthermore, PDE3 and PDE4 play redundant roles in this process, such that inhibition of both isoforms is required to achieve PKA-mediated attenuation of PKD activation.     

Pharmacology of a new cyclic nucleotide phosphodiesterase type 4 inhibitor,V11294

V11294 is a new cyclic nucleotide phosphodiesterase type 4 (PDE4) inhibitor of the rolipram class. In this report we present the pharmacological profile of V11294. V11294 inhibited PDE4 isolated from human lung with IC(50) 405 nM, compared to 3700 nM for rolipram. In contrast, V11294 inhibition of human PDE3 and PDE5 occurred only at concentrations greater than 100,000 nM. Like rolipram, V11294 inhibited PDE4D more potently than other PDE4 subtypes. V11294, when incubated with human anticoagulated whole blood in vitro, or administered to mice, caused increased cAMP concentration, consistent with inhibition of PDE4. V11294 inhibited lectin-induced proliferation and lipopolysaccharide-induced TNFalpha synthesis by human adherent monocytes in vitro and inhibited lipopolysaccharide-induced TNFalpha synthesis in mice. V11294 caused relaxation of guinea pig isolated trachea and inhibited allergen-induced bronchoconstriction and eosinophilia in guinea pigs at doses of 1 and 3 mg/kg, p.o. In ferrets, V11294 was not emetogenic at doses up to 30 mg/kg, p.o., despite plasma concentration reaching 10-fold the IC(50) for PDE4. In contrast, rolipram induced severe retching and vomiting at 10 mg/kg, p.o. In conclusion, V11294 is an orally active PDE4 inhibitor that exhibits antiinflammatory activity in vitro, and in vivo at doses that are not emetogenic.     

Characterization of a liver low Michaelis-Menten constant 3',5'-cyclic adenosine monophosphate phosphodiesterase activity sensitive to thyroid status

These studies were undertaken to assess the subcellular distribution and some biochemical properties of the hepatic cAMP phosphodiesterase(s) whose activity is modulated by the thyroid status in the rat. Thyroidectomy led to a 2-fold increase in low Michaelis-Menten constant (Km) cAMP phosphodiesterase activity in Golgi-endosomal fractions, but little affected this activity in crude particulate fractions. On analytical sucrose density gradients, an increase in cAMP phosphodiesterase activity in particulate elements which equilibrated at densities 1.17-1.22 was also observed. Acute insulin treatment did not further increase cAMP phosphodiesterase activity in Golgi-endosomal fractions of thyroidectomized rats. Up to 75% of the cAMP phosphodiesterase activity associated with Golgi-endosomal fraction of euthyroid and hypothyroid rats was inhibited by cGMP (IC50, 10 microM and 1 microM, respectively). Activity was also potently inhibited by griseolic acid, cilostamide, and cilostazole (IC50, less than 1 microM) but was much less sensitive to R0-20-1724 (IC50, 1 mM). Treatment of Golgi-endosomal fractions by a hypotonic extract of rat liver lysosomes led to the solubilization of 50% of low Km cAMP phosphodiesterase activity. On sucrose density gradients, the solubilized activity migrated as a slightly asymmetrical peak with a sedimentation coefficient of 6 S in euthyroid rats and 6.9 S in hypothyroid rats. On nondenaturing polyacrylamide gel electrophoresis, the activity migrated as two majors peaks with Rf values of 0.23 and 0.50; only the activity associated with the fast-moving peak was increased by thyroidectomy. On diethylaminoethyl-Sephacel chromatography, four peaks of cAMP phosphodiesterase activity, two of which were cGMP-inhibitable, were resolved. Thyroidectomy increased the activity associated with one of the cGMP-inhibitable peaks (eluted at 0.7-0.9 M sodium acetate) and led to the appearance of a new peak of activity (eluted at 0.4 M), which was not sensitive to cGMP. These results show that the low Km phosphodiesterase activity associated with liver Golgi-endosomal fractions, previously shown to be increased in hyperinsulinemic rats, is also increased in hypothyroid animals. They also suggest that, based on pharmacological and physical criteria, the enzyme species affected by the thyroid status belongs to the cGMP-and cilostamide-inhibited subclass of low Km cAMP phosphodiesterases.     

Purification and characterization of guanosine 3',5'-monophosphate-inhibited low K(m) adenosine 3',5'-monophosphate phosphodiesterase from human placental cytosolic fractions.

We previously characterized human placental cytosolic cAMP phosphodiesterase (PDE) and found that two low K(m) cAMP PDE isoforms that were very sensitive to inhibition by cGMP and cilostamide were activated by insulin. As a first step toward understanding the mechanisms by which insulin activates this enzyme, we purified the cGMP-inhibited low K(m) cAMP PDE (cGI-PDE) from human placentas. The enzyme was purified 11,700-fold from a pool of 100,000 x g supernatant fractions of 10-15 placentas by ammonium sulfate precipitation, diethylaminoethyl-cellulose chromatography, and affinity chromatography, using an isothiocyanate derivative of cilostamide (CIT-agarose). The specific activity of the affinity-purified enzyme was 432 +/- 17 nmol/min.mg (mean +/- SD; n = 4). Gel permeation chromatography of the CIT-agarose eluates revealed one protein peak that coincided with PDE activity at an elution position of 135,000 daltons. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of this protein peak and CIT-agarose eluates revealed the same patterns, indicating that the purified PDE preparations contained multiple proteins with apparent mol wt of 138K, 83K, 72K, 67K, 63K, and 44K. The 138K form appears to be an intact enzyme; an analogous approximately 135K form has recently been identified in rat adipocyte particulate fractions by specific immunoprecipitation or Western immunoblots. In addition, other smaller forms eluted at 135,000 daltons on gel permeation chromatography, suggesting that, although proteolyzed, they must have been associated by either noncovalent interactions or disulfide bonds. All of the protein bands observed on the sodium dodecyl sulfate-polyacrylamide electrophoresis gel reacted with rabbit antibodies raised against human platelet cGI-PDE. Ten peptides from endoproteinase Lys-C-digests of the affinity-purified placental cGI-PDE were isolated and sequenced; sequences of eight peptides were identical to the deduced amino acid sequences in the C-terminal half of a human heart cGI-PDE cDNA, while those of two peptides were not found in the heart enzyme. The sequences of the eight peptides also matched peptide sequences derived from a purified human platelet cGI-PDE. These results provide evidence that the catalytic C-terminal half domain of the placental insulin-sensitive cGI-PDE shares homology with those of human heart and platelet cGI-PDEs. K(m) and maximum velocity values for cAMP and cGMP were 0.57 microM and 862 nmol/min.mg, and 15 microM and 467 nmol/min.mg, respectively. ED50 values for cGMP, cilostamide, and Ro 20-1724 were 0.12, 0.22, and 120 microM, respectively.(ABSTRACT TRUNCATED AT 400 WORDS)     

Subcellular distribution of high-affinity type IV cyclic AMP phosphodiesterase activity in rabbit ventricular myocardium: relations to the effects of cardiotonic drugs

Rabbit ventricular myocardium contains distinct cytosolic and particulate cyclic AMP (cAMP) phosphodiesterase activities that exhibit characteristics ascribed to a high-affinity type IV cAMP phosphodiesterase activity found in several tissues. The particulate activity associated with sarcoplasmic reticulum vesicles has an apparent Km for cAMP of about 0.3 microM and a Vmax of 2.45 +/- 0.55 nmol/min/mg. Cyclic GMP (cGMP) inhibits hydrolysis measured at 0.25 microM cAMP with an IC50 value of 0.28 microM. In comparison, a ventricular cytosolic high-affinity cAMP phosphodiesterase activity obtained by anion exchange chromatography (Peak III) has an apparent Km of 0.93 microM and a Vmax of 17 +/- 1 nmol/min/mg. Hydrolysis of 0.25 microM cAMP by this cytosolic activity is weakly inhibited by cGMP with an IC50 value of 142 microM. Particulate enzyme activity is 60-fold more sensitive to inhibition by milrinone than is the cytosolic form (Ki = 0.18 versus 11 microM, respectively); the pyridazinone imazodan is a 12-fold more potent inhibitor of the particulate activity than of the cytosolic form (Ki = 1.5 versus 18 microM, respectively). Inhibition of both cytosolic and particulate enzyme activities appears competitive in nature. Solubilization of particulate activity did not significantly alter its affinity for substrate or sensitivity to inhibition by cGMP. In the presence of a submaximally activating concentration of forskolin (0.4 microM), selective phosphodiesterase inhibitors potentiated the activation of protein kinase in isolated ventricular septal slices. Under these conditions, changes in cAMP-dependent protein kinase activity ratios correlated more closely with contractile responses than did changes in intracellular content of cAMP.(ABSTRACT TRUNCATED AT 250 WORDS)     

Subcellular distribution of high-affinity type IV cyclic AMP phosphodiesterase activities in rabbit ventricular myocardium: relations to post-natal maturation

Cytosolic and particulate Type IV (high-affinity) cAMP phosphodiesterase (PDE) activities were isolated from the ventricular myocardium of newborn (NB; 24 to 48 h), immature (IM; 14 to 16 days) and adult (AD; 6 to 8 months) rabbits. Cytosolic activity from each age group was resolved into three distinct peaks of activity by DEAE cellulose anion exchange chromatography. Type IV PDE activity was identified as a predominant activity in the cytosolic peak III activity in all three age groups when measured with 0.25 microM cAMP as substrate. A particulate Type IV PDE activity was associated with the sarcoplasmic reticulum (SR) fractions in each age group. No significant age-related changes in the affinity of the particulate enzyme for cAMP (apparent Km = 0.3 to 0.5 microM) were evident, but the Vmax for this SR-associated activity increased from 553 +/- 7 pmol/min/mg in the NB to 725 +/- 9 pmol/min/mg in the IM and 2450 +/- 33 pmol/min/mg in the AD. In each age group, milrinone, imazodan, piroximone and indolidan were more potent inhibitors of the SR-associated activity as compared with the cytosolic peak III activity. In contrast, RO 20-1724 and rolipram were relatively more selective inhibitors of the cytosolic peak III activity. Age-related differences in the sensitivity of type IV PDE to inhibition was dependent upon the selectivity of the inhibitor and the subcellular enzymic distribution. Cytosolic peak III PDE activity was further resolved by gel filtration chromatography into two peaks. Hydrolysis of cAMP by the higher molecular weight peak was inhibitable by cGMP (IC50 = 0.25 +/- 0.07 microM in NB and 0.07 +/- 0.01 microM in AD) whereas the lower molecular weight peak activity was relatively insensitive to inhibition by cGMP (IC50 greater than 100 microM). The lower molecular weight peak constituted a relatively greater proportion of the total peak III activity in the NB as compared to the AD. Analysis of the kinetics of cGMP inhibition of high-affinity cAMP hydrolysis was consistent with the presence of a greater number of high-affinity (presumably drug-sensitive) binding sites in the SR-associated activity as compared to the cytosolic peak III activity in both NB and AD. These results support the hypothesis that the cGMP-inhibitable Type IV PDE activity may be the primary site of action for certain newer cardiotonic drugs. Differences in drug action in young versus adult myocardium may be related to the selectivity of the cardiotonic drugs for this specific isozyme and its lower specific activity during the early stages of maturation.     

Spatiotemporal dynamics of beta-adrenergic cAMP signals and L-type Ca2+ channel regulation in adult rat ventricular myocytes: role of phosphodiesterases

Steady-state activation of cardiac beta-adrenergic receptors leads to an intracellular compartmentation of cAMP resulting from localized cyclic nucleotide phosphodiesterase (PDE) activity. To evaluate the time course of the cAMP changes in the different compartments, brief (15 seconds) pulses of isoprenaline (100 nmol/L) were applied to adult rat ventricular myocytes (ARVMs) while monitoring cAMP changes beneath the membrane using engineered cyclic nucleotide-gated channels and within the cytosol with the fluorescence resonance energy transfer-based sensor, Epac2-camps. cAMP kinetics in the two compartments were compared to the time course of the L-type Ca(2+) channel current (I(Ca,L)) amplitude. The onset and recovery of cAMP transients were, respectively, 30% and 50% faster at the plasma membrane than in the cytosol, in agreement with a rapid production and degradation of the second messenger at the plasma membrane and a restricted diffusion of cAMP to the cytosol. I(Ca,L) amplitude increased twice slower than cAMP at the membrane, and the current remained elevated for approximately 5 minutes after cAMP had already returned to basal level, indicating that cAMP changes are not rate-limiting in channel phosphorylation/dephosphorylation. Inhibition of PDE4 (with 10 micromol/L Ro 20-1724) increased the amplitude and dramatically slowed down the onset and recovery of cAMP signals, whereas PDE3 blockade (with 1 micromol/L cilostamide) had a minor effect only on subsarcolemmal cAMP. However, when both PDE3 and PDE4 were inhibited, or when all PDEs were blocked using 3-isobutyl-l-methylxanthine (300 micromol/L), cAMP signals and I(Ca,L) declined with a time constant >10 minutes. cAMP-dependent protein kinase inhibition with protein kinase inhibitor produced a similar effect as a partial inhibition of PDE4 on the cytosolic cAMP transient. Consistently, cAMP-PDE assay on ARVMs briefly (15 seconds) exposed to isoprenaline showed a pronounced (up to approximately 50%) dose-dependent increase in total PDE activity, which was mainly attributable to activation of PDE4. These results reveal temporally distinct beta-adrenergic receptor cAMP compartments in ARVMs and shed new light on the intricate roles of PDE3 and PDE4.     

Characterization of human placental cytosolic adenosine 3',5'-monophosphate phosphodiesterase by inhibitors and insulin treatment

To gain insight into the regulation of low Km cAMP phosphodiesterases (PDE) by insulin in human tissues, PDEs in human placenta were studied. Human placenta contained cAMP PDEs in particulate and cytosolic fractions. More than 99% of the total activity was localized in the cytosolic fraction. The cytosolic fraction exhibited at least four cyclic nucleotide PDEs when fractionated by DEAE-cellulose chromatography. The first form was a calmodulin-activated PDE which hydrolyzed both cGMP and cAMP. The second form was a high affinity cAMP PDE with a nonlinear kinetic characteristic, but was not inhibited by either cGMP or cilostamide (either compound is known to specifically inhibit rat insulin-sensitive cAMP PDE). The third form was a low Km cAMP PDE, but was only modestly sensitive to inhibition by cGMP or cilostamide. The fourth form was a cAMP PDE which showed high sensitivity to inhibition by cGMP or cilostamide. The IC50 values of the fourth form were comparable to those of rat adipose insulin-sensitive PDE. However, its Km for cAMP was 2 microM, which is about 10 times higher than that of the rat enzyme. Insulin treatment on placenta tissues stimulated at least two PDEs, the third and fourth forms. To our knowledge, this is the first report to describe insulin-sensitive cAMP PDEs in the cytosolic fraction of human placenta.     

Differential Modulation of Cytokine Production by Drugs: Implications for Therapy in Heart Failure

We studied the effects of various phosphodiesterase (PDE) III inhibitors: amrinone, pimobendan and vesnarinone: a PDE IV inhibitor (Ro 20-1724) and a PDE V inhibitor (E-4021) on the production of cytokines which have been shown to depress myocardial function. Recently developed inotropic agents which inhibit PDE III activity have produced short-term hemodynamic benefits in patients with advanced heart failure, but long-term treatment with these agents has an adverse effect on survival. However, vesnarinone, which has been shown to improve survival dramatically, has an immunomodulating effect and inhibits the production of cytokines. Peripheral blood mononuclear cells obtained from healthy human subjects were stimulated with lipopolysaccharide and each PDE inhibitor was added. After 24 h of incubation, tumor necrosis factor alpha (TNF-α), interleukin 1 beta (IL-1β) and IL-6 in the culture supernatants were measured by an enzyme-linked immunosorbent assay. All three PDE III inhibitors, amrinone, pimobendan and vesnarinone, inhibited TNF-αproduction, but vesnarinone's inhibitory effect was the most prominent. Amrinone and pimobendan enhanced IL-1βproduction, whereas vesnarinone had no effect. Vesnarinone inhibited IL-6 production and pimobendan slightly decreased IL-6 production, whereas amrinone had no significant effect on IL-6 production. The PDE IV inhibitor, Ro 20-1724, decreased the production of IL-1βand TNF-αand also tended to inhibit IL-6 production; its modulation of cytokine production was similar to the effects of vesnarinone. Because 8Br-cAMP or 8Br-cGMP did not suppress cytokine production, the modulating effects were not considered to result from an increase in cAMP or cGMP. Differential modulation of cytokine production may play a role in the therapeutic effect in heart failure patients who are treated with drugs that have PDE-inhibitory actions. It may be important to study whether the use of dual inhibitors of PDE III and PDE IV is therapeutically more useful for the treatment of heart failure due to their immunomodulating properties.     

New phosphodiesterase inhibitors as therapeutics for the treatment of chronic lung disease

Phosphodiesterase 4 (PDE4) is a member of the growing family cyclic AMP and cyclic GMP. Earliest described inhibitors of PDE4, such as rolipram, demonstrate marked anti-inflammatory and bronchodilatory effects in vitro and in vivo. The clinical utility of these earlier compounds was limited by their propensity to elicit gastrointestinal side effects. This has led to an extensive effort to identify novel PDE4 inhibitors that maintain the anti-inflammatory activity and bronchodilatory activity of rolipram but with a reduced potential to produce side effects. This article summarizes the evidence supporting the utility of selective PDE4 inhibitors in the treatment of asthma and chronic obstructive pulmonary disease, discusses the recent results obtained in clinical trials with second-generation inhibitors, and presents two approaches designed to identify additional novel selective PDE4 inhibitors.     

Cyclic AMP phosphodiesterases in human lymphocytes

The function of lymphocytes, like platelets, has been shown to be inhibited by agents which increase intracellular cyclic AMP. Two high-affinity cAMP phosphodiesterases (PDEs), the cyclic GMP-inhibited cAMP phosphodiesterase, PDE3, and the cAMP-specific phosphodiesterase PDE4, are known to regulate cAMP concentration in haemopoietic cells by degrading cAMP to AMP. We characterized the relative contribution of the two PDEs to total lymphocyte PDE activity. We then determined which of the different gene products, PDE3A, typical of myocardium and platelets, or PDE3B, typical of adipocytes, were present in lymphocytes. The PDE3-specific inhibitor, milrinone, and the PDE4 inhibitor, rolipram, suppressed hydrolysis by 70% and 30% respectively, which indicated that both PDE4 and PDE3 were present, and that PDE3 was predominant. RT-PCR yields the expected size fragment for the primer pair PDE3B and not for PDE3A. The DNA sequence obtained had > 95% identity with PDE3B. PDE3B appears to be the major cAMP PDE in lymphocytes. In contrast to human platelets, human lymphocytes appear to contain the PDE3B subtype. Since PDE3B in adipocytes is subject to hormonal regulation, lymphocytes may be similarly modulated. Understanding the role of cAMP regulation and the involvement of cAMP in lymphocyte function may have important implications in drug development.     

A biochemical and functional assessment of monocyte phosphodiesterase activity in healthy and asthmatic subjects

The aim of this study was to investigate whether cyclic adenosine 3'5-monophosphate (cAMP) phosphodiesterase (PDE) activity is altered in monocytes from mild asthmatic subjects. Total cAMP PDE activity (pmol/min per mg protein) was significantly greater in homogenates prepared from monocytes from asthmatic subjects (68.3 +/- 7.0, n=9) compared to healthy individuals (46.3 +/- 3.3, n=14, P<0.05). The PDE inhibitors siguazodan (PDE3-selective), rolipram (PDE4-selective) and theophylline (non-selective) produced a concentration-dependent inhibition of cAMP PDE activity in homogenates from monocytes from normal and asthmatic subjects. However, siguazodan produced significantly greater (P<0.05), and rolipram significantly less (P<0.05), inhibition of total cAMP PDE activity in monocytes from asthmatics (n=4) than from healthy individuals (n=5). cAMP PDE activity was inhibited with equal potency by theophylline in monocytes from healthy and asthmatic subjects. We also investigated the functional consequences of the changes in PDE activity in mononuclear cells obtained from asthmatic subjects. There was no significant difference in the ability of PDE4 inhibitors to attenuate TNF alpha release from monocytes obtained from asthmatic compared with healthy subjects (P>0.05). Despite a significant increase in the biochemical activity of PDE3 in monocytes from asthmatic subjects, the PDE3 inhibitor siguazodan, failed to significantly reduce TNF alpha release from human monocytes. Thus, total cAMP PDE activity is increased in monocytes taken from mild asymptomatic asthmatics compared to healthy subjects and is reflected by an increase in the proportion of PDE3 and a decrease in the proportion of PDE4. This augmented enzyme activity was not associated with an alteration in the ability of PDE4 inhibitors to attenuate mononuclear cell function from asthmatics compared to healthy individuals.     

Rolipram, a Specific Type IV Phosphodiesterase Inhibitor, Ameliorates Aspirin-Induced Gastric Mucosal Injury in Rats

Inhibition of type IV phosphodiesterase (PDE IV) activity reduces the production of various proinflammatory cytokine and suppresses neutrophil activation. Nonsteroidal anti-inflammatory drugs such as aspirin induce gastric mucosal lesions. In the pathogenesis of aspirin-induced gastric mucosal lesion, the contributions, of activated inflammatory cells and proinflammatory cytokine production are critical. The specific PDE IV inhibitor rolipram is known to be a potent inhibitor of inflammation by increasing intracellular cyclic AMP in leukocytes. The aim of the present study was to determine whether rolipram can ameliorate aspirin-induced gastric mucosal lesions in rats and whether the agent can inhibit the inrease in neutrophil accumulation and the production of proinflammatory cytokines. Gastric lesions were produced by administration of aspirin (200 mg/kg) and HCl (0.15 N; 8.0 ml/kg). Rolipram was injected 30 min before aspirin administration. The tissue myeloperoxidase concentration in gastric mucosa was measured as an indicat or of neutrophil infiltration. The gastric mucosal concentrations of tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) were determined by ELISA. The intragastric administration of aspirin induced multiple hemorrhagic erosions in rat gastric mucosa. Gastric mucosal lesions induced by aspirin were significantly inhibited by treatment with rolipram. The mucosal myeloperoxidase concentration was also suppressed by rolipram. Increases in the gastric content of TNF-α and IL-1β after aspirin administration were inhibited by pretreatment with rolipram. We demonstrated that the specific type IV PDE inhibitor, rolipram, could have a potent antiulcer effect, presumably mediated by its anti-inflammatory properties.     

Inhibition of type 4 phosphodiesterase by rolipram and Ginkgo biloba extract (EGb 761) decreases agonist-induced rises in internal calcium in human endothelial cells

The effects of Gingko biloba extract EGb 761 on 5 isolated, vascular, cyclic nucleotide phosphodiesterase (PDE) isoforms were evaluated. EGb 761 preferentially inhibited PDE4 (IC(50)=25.1 mg/L), the isoform that is mainly present in endothelial cells, in a competitive manner (K:(i)=12.5 mg/L). Because changes in cyclic nucleotide levels may affect intracellular calcium ([Ca(2+)](i)) levels in endothelial cells, we examined the effects of EGb 761 on both resting [Ca(2+)](i) levels and agonist-induced rises in [Ca(2+)](i) in single human umbilical vein endothelial cells (HUVECs) in culture. The effects of EGb 761 were compared with those of rolipram, a selective PDE4 inhibitor that increases cellular cAMP levels, and the cAMP analogue dibutyryl cAMP (db-cAMP). EGb 761 (20 and 100 mg/L), rolipram (50 micromol/L), and db-cAMP (100 micromol/L) significantly inhibited histamine-, ATP-, and thrombin-induced [Ca(2+)](i) increases in HUVECs without modifying resting [Ca(2+)](i) levels. Similar results were obtained by using a Ca(2+)-free bath solution. EGb 761 (100 mg/L), but not rolipram (50 micromol/L) or db-cAMP (100 micromol/L), also inhibited Ca(2+) influx into cells having thapsigargin-depleted internal Ca(2+) stores and bathed in a Ca(2+)-free external solution. Our results are consistent with an inhibition of PDE activity that causes a reduction of agonist-induced increases in [Ca(2+)](i) in HUVECs, mainly by inhibition of Ca(2+) mobilization from internal stores. It thus may be that the cardiovascular effects of EGb 761 involve inhibition of PDE4 activity and subsequent modification of Ca(2+) signaling in endothelial cells.     

Myocardial protein kinases: I. Properties of soluble and membrane-bound catalytic subunits.

The particulate fraction of rat left ventricle, under certain specific assay conditions, was found to contain as much protein phosphotransferase activity as the soluble fraction. Both the soluble and the membrane-bound holoenzyme could be activated by cyclic AMP. The properties of the catalytic subunit of the soluble enzyme differed from those of the catalytic subunit of the particulate fraction with respect to substrate specificity and apparent K_M values for ATP. The soluble enzyme was found to have a very low apparent K_M for ATP (5.0 × 10^?6m) and was inhibited by ATP concentrations above 0.2 mm. The particulate enzyme, on the other hand, showed a high apparent K_M for ATP (1.3 × 10^?4m) and the activity increased with increasing ATP concentrations up to 0.4 mm. The activities in the soluble and particulate fractions were observed to vary inversely between ATP concentrations of 0.1 mm and 0.4 mm. It is concluded that ATP concentration may play a role in controlling the activity of the catalytic subunit after its dissociation from the holoenzyme (by cyclic AMP binding to the regulatory subunit).