Primary sequence of cyclic nucleotide phosphodiesterase isozymes and the design of selective inhibitors

Primary sequence information has been reported for more than 15 different mammalian cyclic nucleotide phosphodiesterases. Moreover, recent observations suggest that many of these isozymes are selectively expressed in a limited number of cell types. The fact that nearly all these different phosphodiesterases have unique primary sequences in their catalytic or regulatory domains and that they are often selectively expressed implies that it may be possible to modulate individual isozymes using specific drugs. Joe Beavo and David Reifsnyder summarize much of the evidence that has led to our current understanding of multiple isozymes of phosphodiesterase, with emphasis on aspects that may be relevant to drug design. They also discuss why many previous attempts to isolate isozyme-selective inhibitors may have failed.     

选择性磷酸二酯酶抑制剂研究进展

磷酸二酯酶 (PDEs)迄今已报道有 9个基因家族 ,每个家族又包括多个亚家族。PDEs分布于多个组织中 ,其抑制剂具有广泛的生理作用。其中 ,PDE 4抑制剂被认为是作用于细胞内靶点的新型免疫调节、抗炎药物。新型PDE 5抑制剂sildenafil通过NO/cGMP通路舒张海绵体血管及平滑肌 ,应用于阳痿治疗取得了较好的临床效果。该文简要综述选择性PDEs抑制剂近年研究进展。     

Characterization and selective inhibition of cyclic nucleotide phosphodiesterase isozymes in canine tracheal smooth muscle

Cyclic nucleotide phosphodiesterases (PDEs) from canine trachealis were characterized with respect to their kinetic properties, sensitivity to selective inhibitors, and subcellular distribution. Extracts from whole tissue homogenates were applied to DEAE-Sepharose anion exchange columns and eluted with a linear sodium acetate gradient. Three major peaks of PDE activity were resolved. The first (PDE I), which eluted at 0.2 M sodium acetate, was applied to a calmodulin (CaM)-Sepharose affinity column and resolved into CaM-insensitive and CaM-sensitive PDEs. The CaM-insensitive isozyme (PDE Ia) had apparent Km values of 135 microM (cAMP) and 4 microM (cGMP) and was potently inhibited by zaprinast (Ki = 0.1 microM). The CaM-sensitive isozyme (PDE Ic) had apparent Km values of 1 microM (cAMP) and 2 microM (cGMP) and was inhibited by zaprinast with an apparent Ki of 35 microM. The second peak of activity (PDE II) from the anion exchange column eluted at 0.3 M sodium acetate and had apparent Km values of 93 microM (cAMP) and 60 microM (cGMP). The enzyme displayed positive cooperativity with respect to the hydrolysis of cAMP (nH = 1.7). Low concentrations of cGMP (0.1-1 microM) reduced cooperativity (nH = 1.1) and increased the hydrolysis of 1 microM cAMP. The third peak of activity from the anion exchange column eluted at 0.6 M sodium acetate and displayed anomalous kinetics that suggested the presence of two isozymes. This was supported by the observation that enzyme activity was only partially inhibited by SK&F 94120 or Ro 20-1724 but was abolished by the combination of the two PDE inhibitors. Subsequent studies confirmed the existence of two isozymes. The first, PDE III, had apparent Km values of 0.3 microM (cAMP) and 8 microM (cGMP) and was inhibited by cGMP (IC50 = 0.1 microM), SK&F 94120 (Ki = 7.8 microM), and SK&F 94836 (Ki = 0.4 microM). The second, PDE IV, had apparent Km values of 4 microM (cAMP) and 40 microM (cGMP) and was inhibited by Ro 20-1724 (Ki = 5.2 microM) and rolipram (Ki = 0.5 microM) but not by cGMP. Assessment of the 100,000 x g soluble and particulate PDE activity revealed that all five isozymes were present in the soluble fraction, but only four isozymes (PDEs Ia, Ic, III, and IV) were present in the particulate fraction. These results indicate that five distinct PDE isozyme exist in canine trachealis and that these isozymes differ in their kinetic characteristics, sensitivity to activators and inhibitors, and subcellular distribution.     

Role of cyclic nucleotide phosphodiesterase isozymes in intact canine trachealis.

The role of individual cyclic nucleotide phosphodiesterase (PDE) isozymes in regulating cAMP and cGMP content in intact canine trachealis was examined using isozyme-selective and nonselective PDE inhibitors. The inhibitors used in this study were characterized previously [Mol. Pharmacol. 37:206-214 (1990)] and included: 1) zaprinast, an inhibitor (Ki = 0.1 microM) of the cGMP-specific PDE (cAMP Km = 135 microM; cGMP Km = 4 microM); 2) SK&F 94120, an inhibitor (Ki = 7 microM) of the cGMP-inhibited PDE (cAMP Km = 0.3 microM; cGMP Km = 8 microM); 3) Ro 20-1724, an inhibitor (Ki = 5 microM) of the cAMP-specific PDE (cAMP Km = 4 microM; cGMP Km = 40 microM); and 4) 3-isobutyl-1-methylxanthine (IBMX), a nonselective PDE inhibitor (IC50 = 1-30 microM). In addition to the aforementioned isozymes, canine trachealis contains a Ca2+/calmodulin-stimulated PDE (cAMP Km = 1 microM; cGMP Km = 2 microM) and a GMP-stimulated PDE (cAMP Km = 93 microM; cGMP Km = 60 microM), for which selective inhibitors are not available. Isolated canine trachealis strips were contracted with methacholine and exposed to various concentrations of PDE inhibitors, before being relaxed by the cumulative addition of isoproterenol, an adenylate cyclase activator, or sodium nitroprusside, a guanylate cyclase activator. At the completion of the concentration-response studies, tissues were flash-frozen and assayed for cyclic nucleotide content. Neither isoproterenol-induced relaxation nor cAMP accumulation was altered by zaprinast, but both of these responses were potentiated by pretreatment of tissues with either SK&F 94120 or Ro 20-1724. The effects of SK&F 94120 and Ro 20-1724 were additive, and the combination of SK&F 94120, Ro-1724, and IBMX had no greater effect on the responses to isoproperenol than did either IBMX alone or the combination of SK&F 94120 plus Ro 20-1724. In contrast, zaprinast potentiated sodium nitroprusside-induced relaxation and cGMP accumulation, whereas neither SK&F 94120 nor Ro 20-1724 altered these responses. IBMX produced a greater potentiation than did zaprinast, and the combination of zaprinast and IBMX had a greater effect than either agent alone. The results of this study suggest that the cGMP-inhibited and cAMP-specific PDEs are responsible for cAMP hydrolysis in intact canine trachealis, whereas cGMP hydrolysis is mediated by the cGMP-specific PDE as well as the Ca2+/calmodulin-stimulated PDE and/or the cGMP-stimulated PDE.     

Tissue selective inhibition of cyclic nucleotide phosphodiesterase by denbufylline

The effects of the novel alkylxanthine denbufylline (1,3-d-n-butyl-7-(2'-oxopropyl)-xanthine), theophylline and 3-isobutyl-1-methyl-xanthine (IBMX), on the breakdown of cyclic AMP in homogenates of rat erythrocytes, abdominal aorta, adipocytes and cardiac and skeletal muscle were studied. Theophylline and IBMX inhibited cyclic nucleotide phosphodiesterase in all tissue extracts. In contrast, denbufylline was a tissue selective inhibitor of cyclic nucleotide phosphodiesterase. In skeletal muscle and erythrocytes, denbufylline (10 mumol/l) inhibited cyclic nucleotide phosphodiesterase activity by at least 80%. In these tissues, denbufylline was 10-30 and 100-300fold more potent than IBMX and theophylline, respectively. In adipocytes and cardiac and smooth muscle, denbufylline was not an effective inhibitor of cyclic AMP breakdown. Hofstee analysis of phosphodiesterase activity revealed that denbufylline selectively inhibited high affinity cyclic AMP phosphodiesterase in erythrocytes and skeletal muscle. In adipocytes, cardiac and vascular smooth muscle, denbufylline did not effectively inhibit the composite cyclic nucleotide phosphodiesterase activities either with high or with low affinity for cyclic AMP.     

Human bronchial cyclic nucleotide phosphodiesterase isoenzymes: biochemical and pharmacological analysis using selective inhibitors.

1 The aims of the present study were to characterize the cyclic nucleotide phosphodiesterase (PDE) isoenzyme activities present in human bronchi and to examine the ability of selective isoenzyme inhibitors to relax histamine and methacholine precontracted preparations of human bronchi. 2 Three separations of pooled human bronchial tissue samples were performed. Ion-exchange chromatography showed that the soluble fraction of human bronchial preparations contains PDE I, II, III, IV and V isoenzyme activities. Multiple forms of PDE I and PDE IV were observed and PDE IV was the main cyclic AMP hydrolytic activity. 3 3-Isobutyl-l-methylxanthine (IBMX) non-selectively inhibited all separated isoenzyme activities. Zaprinast selectively inhibited PDE V, but also effectively inhibited one of the two PDE I isoforms identified. The PDE IV selective inhibitors rolipram and RO-201724, inhibited the PDE IV activities as did the dual PDE III/IV inhibitor, Org 30029. Org 9935, a PDE III selective inhibitor, potently attenuated part of the PDE IV activity peak in one of three separations performed, indicating that some PDE III activity may co-elute with PDE IV under the experimental conditions employed. 4 PDE IV-selective (rolipram), PDE III-selective (Org 9935) and dual PDE III/IV (Org 30029) inhibitors were effective relaxants of human bronchial smooth muscle. The PDE V/PDE I inhibitor, zaprinast was relatively ineffective. 5 The present study demonstrates in human bronchi, as in animal airways smooth muscle, that inhibitors of PDE III, PDEIV and dual PDE III/IV have potentially useful bronchodilator activity and are worthy of further consideration as anti-asthma drugs.     

Therapeutic potential of cyclic nucleotide phosphodiesterase inhibitors in heart failure

There are several reasons to believe that agents that augment cAMP-mediated signalling in cardiac myocytes should have beneficial effects in patients with heart failure. However, clinical trials of first-generation cyclic nucleotide phosphodiesterase (PDE3) inhibitors, which raise cAMP content by blocking its hydrolysis, have shown that chronic administration of these drugs affect survival adversely. The problem may be the non-selective activation of a broad spectrum of cAMP-regulated cellular responses these agents elicit. More selective (or alternatively selective) cyclic nucleotide PDE inhibitors might improve results by evoking a more restricted set of cellular responses.     

Therapeutic potential of cyclic nucleotide phosphodiesterase inhibitors in heart failure

There are several reasons to believe that agents that augment cAMP-mediated signalling in cardiac myocytes should have beneficial effects in patients with heart failure. However, clinical trials of first-generation cyclic nucleotide phosphodiesterase (PDE3) inhibitors, which raise cAMP content by blocking its hydrolysis, have shown that chronic administration of these drugs affect survival adversely. The problem may be the non-selective activation of a broad spectrum of cAMP-regulated cellular responses these agents elicit. More selective (or alternatively selective) cyclic nucleotide PDE inhibitors might improve results by evoking a more restricted set of cellular responses.     

Differential modulation of tissue function and therapeutic potential of selective inhibitors of cyclic nucleotide phosphodiesterase isoenzymes.

Since the discovery of cyclic nucleotide phosphodiesterase 30 years ago, there have been major advances in our knowledge of this group of isoenzymes. Five families, each composed of several isoforms and having differing tissue distributions, have been described. David Nicholson and colleagues compare the tissue distribution of phosphodiesterase isoenzymes and discuss the differential effects of inhibition of particular isoenzymes, with differing subcellular localization, on tissue function. They also review the potential use of isoenzyme selective phosphodiesterase inhibitors in a range of clinical disorders such as heart failure, asthma, depression and dementia.     

Comparison of cyclic nucleotide phosphodiesterase isoforms from rat heart and bovine aorta. Separation and inhibition by selective reference phosphodiesterase inhibitors

The resolution as well as the biochemical properties of the multiple molecular forms of cyclic nucleotide phosphodiesterase, in a given tissue, may be strongly dependent upon experimental conditions of preparation (extraction of crude enzyme from tissues and fractionation procedures). In the present study, we compare the different molecular forms of cardiac (rat heart ventricle) and vascular (bovine aorta) phosphodiesterase isolated from crude extracts prepared either in sucrose medium or in hypotonic medium (in the presence of protease inhibitors and ion chelators) using two different fractionation procedures: isoelectric focusing on flat gel bed and DEAE-Trisacryl anion exchange chromatography. Both the calmodulin-dependent and the cAMP-specific forms exhibited close IEF and chromatographic patterns and showed similar sensitivities towards reference inhibitors regardless of the tissue of origin. In marked contrast, the cGMP-specific isoform notably differed from one to another tissue with respect to its biochemical properties (only the cardiac tissue being capable of stimulation by cGMP) and sensitivities to xenobiotics. Thus the possibility exists that pharmacological agents may modulate phosphodiesterase activity differently in cardiac and vascular target tissues.     

Effects of nucleotide cyclase activators on the depressor response to selective low Km cyclic AMP phosphodiesterase inhibitors in rats

The potential interaction between agents which promote vasodilation through cyclic AMP (cAMP)- and cyclic GMP (cGMP)-dependent mechanisms was examined in conscious, spontaneously hypertensive rats (SHR). Selective inhibitors of low Km cAMP phosphodiesterase (PDE), CI-930 (0.1?10 mg/kg) and milrinone (0.03?3 mg/kg), were administered i.v. to SHR during a steady-state i.v. infusion of either the guanylate cyclase activator sodium nitroprusside (SNP, 5 or 15 μg/kg/min) or the adenylate cyclase activator fenoldopam (20 μg/kg/min). The depressor responses to Cl-930 and milrinone were significantly attenuated in the presence of SNP. This attenuation of the depressor response to Cl-930 was apparently not due to a hemodynamic floor effect since an equihypotensive dose of the calcium channel blocker verapamil (20 μg/kg/min infusion preceeded by a loading dose of 1.2 mg/kg i.v.) did not alter the response to Cl-930. The depressor response to Cl-930 was also attenuated during an infusion of fenoldopam, although this interaction was dependent upon the order of administration of these agents since pretreatment with Cl-930 (0.3 mg/kg, i.v.) potentiated the depressor response to infusions of fenoldopam (10 and 20 μg/kg/min). These data suggest that activation of either adenylate or guanylate cyclase may alter the intrinsic activity of low Km cAMP PDE and functionally impair the ability of inhibitors of this PDE isozyme to lower blood pressure. Furthemore, teh interactive effects of fenoldoparm and Cl-930 on blood pressure are dependent upon the order of administration.     

A sensitive assay of human blood platelet cyclic nucleotide phosphodiesterase activity by HPLC using fluorescence derivatization and its application to assessment of cyclic nucleotide phosphodiesterase inhibitors

A selective and sensitive HPLC measurement of 3',5'-cyclic nucleotide phosphodiesterase (PDE) activity in human platelets using (3,4-dimethoxyphenyl)glyoxal (DMPG) as a fluorogenic reagent for guanine and its nucleosides and nucleotides is described. cGMP, a substrate for PDE, and GMP, which was produced by the enzyme reaction, are selectively converted by the reaction with DMPG to the fluorescent derivatives. The derivatives were separated by reversed-phase HPLC. Human platelet PDE activity was measured and the inhibitory effects of several compounds were investigated.     

Effects of selective inhibitors on cyclic nucleotide phosphodiesterases of rabbit aorta

In this study three forms of cyclic nucleotide phosphodiesterase (PDE) isolated from rabbit aorta were pharmacologically characterized, and the consequence of selective inhibition of calmodulin-stimulated PDE (CaM-PDE) and cGMP specific PDE (cG-PDE) was evaluated using PDE inhibitors. The cG-PDE (F1) was selectively inhibited by M&B 22948 (IC50 = 0.5 microM) and dipyridamole (IC50 = 7 microM). The cAMP-PDE (cA-PDE, F3) was inhibited more effectively by the cA-PDE inhibitor milrinone than by other PDE inhibitors. The cA-PDE preparation appeared to contain both cG-inhibited PDE and cG-insensitive PDE based on an additive inhibition of the activity by milrinone and SQ 65442, respective inhibitors of these enzymes. Vinpocetine, 8-methoxymethyl isobutylmethylxanthine (8-MeOMeMIX) and M&B 22948 effectively inhibited CaM-PDE (F2). Vinpocetine was a more selective inhibitor of CaM-PDE than M&B 22948 or 8-MeOMeMIX. CaM-PDEs isolated from rabbit aorta and bovine brain exhibited a similar sensitivity to these inhibitors. Seventy-two percent of the cGMP-hydrolyzing activity of this rabbit aortic CaM-PDE preparation was immunoadsorbed to monoclonal antibody (ACC-1) against CaM bound to brain CaM-PDE. Vinpocetine, 8-MeOMeMIX and M&B 22948 at concentrations (30 and 100 microM) which inhibit CaM-PDE greater than 60% increased cGMP but not cAMP levels in l-norepinephrine (NE) preincubated rabbit aortic slices. At concentrations selectively inhibiting cG-PDE, dipyridamole and M&B 22948 increased cGMP levels in untreated slices but failed to increase cGMP levels significantly in NE-treated slices. By contrast, vinpocetine failed to increase cGMP significantly in untreated slices, although it increased cGMP levels in NE or KCl preincubated slices. These data indicate that, in activated (precontracted) aorta, CaM-PDE is a major enzyme, whereas in untreated aorta cG-PDE is a predominant enzyme for the hydrolysis of cGMP. This study also shows a usefulness of selective inhibitors in identifying different forms of PDE and similar drug sensitivities and immunoadsorption of aortic and brain CaM-PDEs by a monoclonal antibody.     

Inotropic responses to cyclic nucleotide phosphodiesterase inhibitors in immature and adult rabbit myocardium

Contractile dose-response relationships for amrinone, milrinone, piroximone, and sulmazole were compared in right ventricular papillary muscles isolated from adult and 14-16-day-old immature rabbits. These drugs were effective inotropic agents in immature myocardium as evidenced by substantial increases in the maximal rate of tension development. The rank order of maximum inotropic effect in the immature muscles was milrinone = sulmazole greater than piroximone greater than amrinone. Compared with adults, the relative magnitude of the inotropic response for each drug was greater in immature myocardium. Profiles of ventricular cytosolic cyclic nucleotide phosphodiesterase (PDE) activities resolved by anion exchange chromatography were similar for the two age groups. Immature myocardium was found to contain the Type IV (Peak III) high-affinity cAMP PDE that has been implicated in the mechanism of action of these drugs. Partially purified cytosolic Type IV PDE from immature and adult myocardium exhibited similar kinetic characteristics (cAMP Km = 0.9 microM; Vmax = 17 nmol/min/mg) and sensitivity to inhibition by cGMP. Except for piroximone, the inhibitory effect of each drug on cytosolic Type IV PDE activity from immature myocardium did not differ from the adult, as indexed by comparable concentrations required to inhibit activity by 50% (IC50) and Ki values (piroximone IC50 and Ki values were higher in the immature compared with the adult group). Thus, these studies demonstrated significant age-related differences in the contractile responses to amrinone, milrinone, piroximone, and sulmazole. These differences are not attributable to differences in myocardial cytosolic Type IV high-affinity cAMP PDE activity.(ABSTRACT TRUNCATED AT 250 WORDS)     

Endothelium-dependent and independent relaxation of the rat aorta by cyclic nucleotide phosphodiesterase inhibitors.

1. The effects of selective inhibitors of adenosine 3':5'-cyclic monophosphate (cyclic AMP) and guanosine 3':5'-cyclic monophosphate (cyclic GMP) phosphodiesterases (PDEs) were investigated on PDEs isolated from the rat aorta and on relaxation of noradrenaline (1 microM) precontracted rat aortic rings, with and without functional endothelium. 2. Four PDE forms were isolated by DEAE-sephacel chromatography from endothelium-denuded rat aorta: a calmodulin-activated PDE (PDE I) which hydrolyzed preferentially cyclic GMP, two cyclic AMP PDEs (PDE III and PDE IV) and one cyclic GMP-specific PDE (PDE V). The latter was selectively and potently inhibited by zaprinast. The two cyclic AMP PDEs were discriminated by specific inhibitors: one was inhibited by cyclic GMP (PDE III) and by new cardiotonic agents (milrinone, CI 930, LY 195115 and SK&F 94120); the other was inhibited by denbufylline and rolipram (PDE IV). None of these drugs significantly inhibited PDE I. 3. The PDE III inhibitors caused endothelium-independent relaxations of rat aortic rings with the following EC50 values (microM concentration producing 50% relaxation): LY 195115: 3.4, milrinone: 5.7, CI 930; 7.8, SK&F 94120: 14.7. Neither NG-monomethyl-L-arginine (L-NMMA, 300 microM), an inhibitor of the L-arginine-NO pathway, nor L-arginine (1 mM) modified the effect of PDE III inhibitors. However, methylene blue (10 microM) an inhibitor of soluble guanylate cyclase abolished relaxation induced by PDE III inhibitors except in the case of compound CI 930. 4. The specific PDE IV and PDE V inhibitors both produced endothelium-dependent relaxations which were inhibited by L-NMMA and by methylene blue (10 microM).(ABSTRACT TRUNCATED AT 250 WORDS)