ried out the present experiments in endothelium-denuded rat aorta. Materials and methods. Phosphodiesterase inhibition. Cytosolic cyclic nucleotide phos-.
Naunyn-Schmiedeberg’s Arch Pharmacol (2001) 363 : 612–619 DOI 10.1007/s002100100397
O R I G I N A L A RT I C L E
Maria Antonia Noguera · Maria Dolores Ivorra · Claire Lugnier · Pilar D’Ocon
Role of cyclic nucleotide phosphodiesterase isoenzymes in contractile responses of denuded rat aorta related to various Ca2+ sources Received: 16 December 1999 / Accepted: 4 January 2001 / Published online: 20 March 2001 © Springer-Verlag 2001
Abstract We have examined the cyclic nucleotide phosphodiesterase isoforms (PDE) involved in the contractile response of rat aorta to different agonists and different experimental procedures for use in functional studies. The inhibitory effect of AAL 05 on the different PDEs isolated from bovine aortic smooth muscle was examined. Compound AAL 05 appeared to be a selective PDE3 inhibitor. We analyzed the ability of the non-selective inhibitor IBMX (3-isobutyl-1-methylxanthine) and the isoenzyme selective inhibitors nimodipine (type1), AAL 05 (6-(Nmethyl-N-cyclohexyl butyl carboxamide) quinolin-2-one) and SK&F 94120 (5-(4-acetamidophenyl) pyrazin-2(1H)-one; type3), rolipram (type4) and zaprinast (type5) to affect the contractile responses of denuded rat aortic rings to KCl (80 mM) and noradrenaline (NA, 1 µM) in the presence or absence of extracellular Ca2+. Rolipram (10–100 µM) and zaprinast (1–100 µM) failed to relax the aortic strips, but IBMX (0.1–30 µM), nimodipine (1 fM– 10 µM), AAL 05 (0.01–100 µM) and SK&F 94120 (0.1– 100 µM) produced a concentration-dependent relaxation or inhibition of contractile responses to the different agonists, but the pIC50 obtained for each inhibitor was different depending on the experimental procedure. Except for nimodipine (a Ca2+ channel blocker), all the PDE inhibitors showed the following rank of potency: pIC50 on NAinduced contractions in Ca2+-free medium > pIC50 on NAinduced contractions in Ca2+-containing solution > pIC50 on depolarizing solution-induced contraction. This ranking apparently depends on the differences in the Ca2+ sources.
M. A. Noguera · M. D. Ivorra · P. D’Ocon (✉) Departamento de Farmacología, Facultat de Farmacia, Universitat de València, Avda. Vicent Andrés Estelles s/n 46100 Burjassot, Valencia, Spain e-mail: m.pilar.docon@uv.es, Tel.: +34-963864828, Fax: +34-963864943 C. Lugnier Laboratoire de Pharmacologie et Physicochimie des Interactions Cellulaires et Moléculaires, UMR CNRS 7034, Facultè de Pharmacie, Université Louis Pasteur, Strasbourg, France
We obtained a good correlation between the pKi of PDE3 inhibitors in biochemical studies and the pIC50 on NA-induced contraction in Ca2+-free medium. In conclusion, PDE1 and PDE3 isoenzymes play an important role as modulators of rat aortic smooth muscle contractility regardless of the experimental procedure used. Since intracellular mechanisms are more dependent on PDE activity, experimental procedures performed in absence of extracellular calcium are the most suitable for analyzing the modulatory role of PDE inhibitors. Keywords KCl · Noradrenaline · Cyclic nucleotide phosphodiesterases · Biochemical-functional correlation · Ca2+ sources
Introduction In the cardiovascular system, modulation of cyclic AMP and cyclic GMP levels is involved in the regulation of diverse functions (Stoclet et al. 1995), and agents that increase cyclic nucleotides produce vasodilatation by decreasing intracellular calcium and reducing the myofilament Ca2+ sensitivity (Ito et al. 1993; Shiraishi et al. 1998). The cellular levels of these nucleotides reflect a balance between their synthesis and catabolism, the latter of which is regulated by cyclic nucleotide phosphodiesterases (PDE), the family of enzymes that catalyse the hydrolysis of the 3’-5’ phosphodiester bond of cyclic nucleotides and terminate cyclic nucleotide-mediated signalling (Tomkinson and Raeburn 1996). The presence of different PDE isozymes in a given tissue might reflect their differential regulation by a variety of signals, thus permitting a fine tuning of cyclic nucleotides (Beavo 1990). In vascular smooth muscle, PDEs belonging to three families were initially isolated: PDE1 (from the calmodulin-dependent family that hydrolyzes cyclic GMP and cyclic AMP), PDE4 (cyclic AMP-specific family) and PDE5 (cyclic GMP-specific family; Lugnier et al. 1986; Weishaar et al. 1986). Later, the presence of another PDE,
613
PDE3 (from the cGMP-inhibited family that hydrolyzes cyclic AMP preferentially but not exclusively), was reported (Komas et al. 1991a, 1991b; Ivorra et al. 1992; Rascon et al. 1992; Saeki and Saito 1993; Coste and Grondin 1995). However, until now, the biochemical evidence that different families of PDEs are present in vascular systems has not been confirmed by definitive studies that clarify their involvement in underlying mechanisms governing vascular contractility (Stoclet et al. 1995) because the functionality of smooth muscle includes a variety of mechanisms modulated by cyclic nucleotides that make it difficult to correlate biochemical analysis and functional studies. The aims of the present work were to determine the PDE isoforms involved in the contractile response of rat aorta to different agonists and to optimize experimental procedures for use in functional studies in order to define the role of each PDE isozyme in the vascular contractility. The four PDE isoenzymes described above were identified from the rat aortic media-adventitia layer (Komas et al. 1991a). We examined the inhibitory effect of AAL 05 on the different PDEs isolated from bovine aortic smooth muscle and analyzed the ability of the non-selective inhibitor IBMX and the isoenzyme selective inhibitors nimodipine (type1), SK&F 94120 and AAL 05 (type3), rolipram (type4) and zaprinast (type5) to affect the contractile responses of rat aorta to different agents acting in different experimental conditions including the presence and absence of extracellular calcium. In order to avoid the interference of PDE isoforms present in the endothelium or an endothelial regulation of the cyclic nucleotide levels of smooth muscle cells, we carried out the present experiments in endothelium-denuded rat aorta.
Materials and methods Phosphodiesterase inhibition. Cytosolic cyclic nucleotide phosphodiesterase activities (PDE) were isolated from the media layer of bovine aorta by a modification of the method of Lugnier et al. (1986). Briefly, the tissue was homogenized in 10 vol (w/v) isotonic buffer A [250 mM sucrose; 25 mM phosphate buffer, pH 6.6; 2 mM magnesium acetate (MgAc); 1 mM dithiothreitol; 5 mM ethylene glycol bis-(β-aminoethyl ether) N,N,N’,N’-tetraacetic acid (EGTA); 2000 u/ml aprotinin; 10 µg/ml soya bean trypsin inhibitor; 10 µg/ml leupeptin] using an Ultraturrax (six times, 10 s) and a glass pestle homogenizer, and then centrifuged at 105,000 g for 60 min. The resulting supernatant was applied to a diethylaminoethyl (DEAE)-Sephacel ion-exchange column and eluted with buffer B (25 mM phosphate buffer, pH 6.6; MgAc 2 mM; dithiothreitol 1 mM) until no absorbance was detected in the eluate at 280 nm. Elution was then continued with a linear gradient of 0–0.55 M NaCl in buffer B (flow rate 25 ml/h). Each fraction was tested for PDE activity. Since DEAE-Sephacel did not allow the separation of calmodulin-PDE (type I) from guanosine 3’-5’-cyclic monophosphate (cyclic GMP)-PDE (typeV), but DEAE-Trysacryl cromatography did (Lugnier et al. 1986), a further high performance liquid chromatography (HPLC) purification step was carried out. The first peak was injected into an HPLC column (TSK-DEAE-5PW), washed for 20 min with elution buffer (25 mM phosphate buffer, pH 6.6) and eluted (0.8 ml/min) by a linear NaCl gradient (0.05–0.3 M) in elution buffer. Fractions under each PDE activity
peak were pooled, dialysed against buffer (20 mM Tris-HCl; 2 mM MgAc, pH 7.5) and stored in aliquots at –80°C with bovine serum albumin (BSA, 1 mg/ml). PDE activities were measured as previously described by Keravis et al. (1980) at a substrate ([3H]cyclic AMP or [3H]cyclic GMP) concentration of 1 µM in the following buffer: 650 mM Tris-HCl, pH 7.5; 2 mM MgAc; BSA; in the presence of 10 µM CaCl2 and 18 nM calmodulin or in the absence of Ca2+ and calmodulin but with 1 mM EGTA. To prevent the influence of cross-contamination between PDE3 and PDE4, the studies performed with these forms were always carried out in the presence of 50 µM rolipram or 100 µM cyclic GMP, respectively. Analysis of results. The IC50 (concentration which produced 50% inhibition of substrate hydrolysis) for the compounds studied was determined from the concentration-response curves obtained with five concentrations of inhibitor and calculated using a non-linear regression. The apparent Ki values were obtained according to Cheng and Prusoff (1973). In each case inhibition was competitive with respect to substrate. Results are expressed as means ± SEM of three determinations done in duplicate with three enzymatic preparations. Functional studies. Endothelium-denuded aortic rings (3–5 mm in length) of Wistar rats (200–220 g) were prepared by rubbing the entire intimal surface, suspended in a 10-ml organ bath containing physiological solution, maintained at 37°C and gassed with 95% O2 and 5% CO2. An initial load of 1 g was applied to each preparation and maintained throughout a 75–90 min equilibration period. This pre-tension was kept constant, but there was a loss of tension (
