- Email: [email protected]
Synergistic actions of nitrovasodilators and isoprenaline on rat aortic smooth muscle
l-uropean Journal o/Pharmacology. 192 (1991) 235-242 ',: 1991 Elsevier Science Publishers B.V. (Biomedical Division)0014-2999/91/$03.50 A D O N I S 001429999100090A
235
EJP 51632
Synergistic actions of nitrovasodilators and isoprenaline on rat aortic smooth muscle D o n a l d H. M a u r i c e 1, Denis C r a n k s h a w 2 a n d R i c h a r d J. H a s l a m i Department of Pathoh~gv and " Department of Obstetrics atut (~vnecoloKv. McMm'ter U'mcersitv. Ilamilton. Ontario. ('anada LS,V 3Z5
Received 18 June 1990, revised MS received 11 September 1990. accepted 25 September 1990
Previous studies have established that nitrovasodilators potentiate the inhibition of platelet function by activators of adenylyl cyclase, but uncertainty exists as to whether a comparable effect is seen in vascular smooth muscle. We initially studied the effects of the nitrovasodilators, sodium nitroprusside (SNP) and 3-morpholinosydnonimine (SIN-l), on the relaxation by isoprenaline of rat aortic smooth muscle that had been precontracted by phenylephrine. Concentrations of SNP (0.25 nM) and SIN-I (30 nM) that relaxed aortic smooth musclc less than 30% alone, caused significant (3-fold) decreases in the ICso values for isoprenaline. The cAMP phosphodiesterase inhibitors, cilostamide (20 nM) and Ro 20-1724 (10 p,M), caused comparable reductions in the IC5o values for isoprenaline. At these concentrations, each of the four compounds also increased the maximum relaxation achieved with isoprenaline. Even more marked synergistic interactions were observed between isoprenaline and either the nitrovasodilators or the cAMP phosphodiesterase inhibitors when these compounds were added simultaneously before contraction of aortic smooth muscle by phenylephrine. Thus, concentrations of SNP (5 nM), SIN-1 (1 #M), cilostamide (1 p.M) and Ro 20-1724 (100 btM) that inhibited contraction by less than 30% decreased the IC50 values for isoprenalinc by 8- to 10-fold. At the above concentrations, these compounds each caused a supra-additive inhibition of contraction when added with 100 nM isoprenaline. Thus. synergism between nitrovasodilators and isoprenaline, an activator of adenylyl cyclasc, could be detected in vascular smooth muscle and was particularly marked when inhibition of contraction was studied. This action of nitrovasodilators resembled that of inhibitors of cAMP phosphodiesterasc. Smooth muscle (aortic); Isoprenaline: Nitrovasodilator; Synergism; (Rat)
1. Introduction The nitrovasodilators, sodium nitroprusside (SNP) and 3-morpholinosydnonimine (SIN-l), and endothelium-derived relaxing factor ( E D R F ) inhibit platelet aggregation (Glusa et al., 1974; Nishikawa et al., 1982: Azuma et al., 1986) in addition to relaxing vascular smooth muscle (Hashimoto et al., 1971; Furchgott and Zawadzki, 1980). Both of these effects are associated with the activation of soluble guanylyl cyclase and increases in intracellular cyclic G M P (cGMP) (Murad, 1986). In contrast, vasodilators such as isoprenaline and prostacyclin (PGI2) exert their effects on vascular smooth muscle by stimulating adenylyl cyclase and increasing intracellular cyclic AMP (cAMP) (Hardman, 1984). Again, platelets are similar to vascular smooth muscle in that cAMP, as well as cGMP, mediates an inhibition of aggregation (Haslam et al., 1978). The most potent inhibitors of platelet function that activate
Correspondence to: R.J. Haslam, Department of Pathology, McMaster University. 120(I Main Street West. Hamilton. Ontario. Canada LgN 3Z5.
adenylyl cyclase are PGI~ and its analogues, such as iloprost (Tateson et al., 1977; SchrOr et al., 1981). Several reports have documented synergistic inhibitory effects on platelet aggregation of PGI~ or its analogues in combination with nitrovasodilators or E D R F (Levin et al., 1982: Radomski et al.. 1987; Macdonald et al., 1988: De Caterina et al., 1988; Willis et al., 1989). However, comparable synergistic effects were not observed in a study of the effects of SNP and iloprost on rabbit vascular smooth muscle (Lidbury et al., 1989), though evidence for synergism between E D R F and PGI 2 has been obtained in experiments with pig coronary arteries (Shimokawa et al., 1988). Recently, we determined the molecular basis of the synergistic inhibition of platelet aggregation by nitrovasodilators and activators of adenylyl cyclase (Maurice and Haslam, 1990). We showed that it resulted from the ability of c G M P to enhance the accumulation of cAMP in platelets by inhibiting a specific cGMP-inhibited cAMP phosphodiestcrase. This enzyme has been most fully characterized in platelets (Grant and Colman, 1984; MacPhee et al., t986) and cardiac muscle (Harrison et al., 1986), but a cAMP phosphodiesterase very similar in its kinetics and sensitivity to inhibitors is known to be present in vascular smooth muscle (Schoeffter et al.,
236 1987; Silver et al.. 1988; Souness et al., 1989: Ahn et al., 1989). We therefore re-investigated the possibility that synergistic effects similar to those seen in platelets could be demonstrated in vascular smooth muscle. The results showed that nitrovasodilators do potentiate the relaxation of rat aortic smooth muscle by isoprenaline and that these compounds exert even more potent synergistic effects when added prior to the contractile stimulus.
2. Materials and methods
2. I. Materials SNP, ( - )-isoprenaline ( + )-bitartrate. phenylephrine and carbachol were obtained from Sigma (St. Louis, MO, U.S,A.). Male Wistar-Kyoto (WKY) rats were obtained from Harlan Sprague-Dawley (Indianapolis, IN, U.S.A.). SIN-1 and Ro 20-1724 were provided by Casella-Riedel (Frankfurt am Main, F.R.G.) and F. Hoffmann La Roche (Nutley, N J, U.S.A.), respectively. Cilostamide was generously supplied by Professor H. Hidaka of Nagoya University (Nagoya, Japan). 2.2. Preparation of rat aortic rings Male WKY rats (225-350 g) were killed by a blow to the head. The thoracic aorta was rapidly removed and placed in a physiological salt solution (PSS), containing 1.16 mM MgSO 4, 2.5 mM CaCI~, 21.9 mM N a H C O > 1.16 mM N a H 2 P O 4, 4.6 mM KCI, 115.5 mM NaCI and 11.1 mM dextrose, that had been equilibrated with 95% 02-5% CO2. The aorta was cleaned of adherent fat and connective tissue and cut into 4.0 mm transverse rings. The endothelium was removed by gently rubbing the luminal surfaces with a wooden stick (Furchgott and Zawadzki, 1980). Rings were mounted in 10 ml jacketed organ baths using stainless steel wire and bathed at 3 7 ° C in PSS, which was gassed with 95% 02-5% CO,. All experiments were carried out with an initial tension of 2 g. Tension development was measured isometrically using a Grass FTO3 transducer and a Grass Model 7D polygraph. Rings were allowed to equilibrate for 2 h, during which they were repeatedly washed with PSS. During this period, the rings were also contracted 3 to 4 times with 30 mM KCI (until a maximal contraction was achieved). Removal of the endothelium was then verified by the inability of 1 p.M carbachol to induce relaxation of rings submaximally contracted with 100 nM phenylephrine. 2.3. Studies on the relaxation of aortic smooth rnuscle The concentration-response curves for the relaxant effects of isoprenaline were determined by cumulative
additions of the compound to submaximally contracted rat aortic smooth muscle. In all experiments, contraction was induced by phenylephrine: the concentration used (I00 nM) caused on average 80% of the maximal contraction, which was seen with 1 p.M. Each addition of isoprenaline was allowed to have its full effect before further additions were made (2 to 3 min). The relaxation caused by each addition was recorded and calculated as a percentage of the maximum possible relaxation (i.e. of the full increase in tension caused by phenylephrine). After 4 washes with PSS, the aortic rings were allowed to equilibrate for 90 min in fresh PSS before thev were used again. With this time interval, successive concentration-response curves for isoprenaline were identical. The effects of nitrovasodilators (SNP or SIN-I) or c A M P phosphodiesterase inhibitors (cilostamide or Ro 20-1724) on the ability' of isoprenaline to relax precontracted aortic smooth muscle were determined as follows. Each compound studied was added to the contracted rings at a specific concentration that caused less than 30% relaxation on all occasions. When the ring tension reached a new plateau (5-15 min), isoprenaline was added cumulatively and the total relaxation caused by each concentration was determined as a percentage of the initial increase in tension caused by phenylephrine. To reduce errors due to the inherent variability, of responses of aortic smooth muscle to vasodilators, concentration-response curves for isoprenaline in the absence and presence of each compound were in each experiment constructed front paired records obtained from the same aortic rings. 2.4.
Studies
on
inhibition of the contraction of aortic
smooth muscle The effects of nitrovasodilators or of cAMP phosphodiesterase inhibitors on the ability of isoprenaline to inhibit the contractions of rat aortic smooth muscle induced by phenylephrine were studied as follows. Aortic rings were incubated for 30 s with vasodilator compounds, either singly or in combination. Phenylephrine (100 nM) was then added and the tension developed after 2 min was measured. Following each contraction, the PSS was changed 4 times and the rings were allowed to equilibrate in fresh PSS for 30 min. This time interval permitted the generation of identical control contractions and virtually identical responses to isoprenaline throughout the experiment, provided low concentrations of isoprenaline were studied first. Using this protocol, it was possible to test up to 15 different combinations of compounds on each aortic ring and to obtain concentration-response curves for the inhibition of contraction by isoprenaline in the presence and absence of single concentrations of other compounds. The latter were used at selected concentrations that caused less than a 30~ inhibition of contraction. Inhibitions of
237
contraction were calculated as percentages of the control contractions for each aortic ring; 3 or 4 rings received identical treatments in each experiment.
1
loo
-
\~x
I
I
I
A
80
2.5. Analysis of results Concentration-response curves were constructed for the relaxation or inhibition of the contraction of rat aortic smooth muscle by isoprenaline in the absence and presence of nitrovasodilators or inhibitors of c A M P phosphodiesterase. Curves were fitted to the experimental results using a computer program (SigmaPlot 4.02A, Jandel Scientific, Corte Madera, CA, U.S.A.) and the following relationship, E = E .... • xn/(x n + I n) + C, where E is the effect, Em~x the maximum effect, x the concentration of isoprenaline, I the ICs0 of isoprenaline, n the Hill slope and C the effect observed in the absence of isoprenaline. This computer program also gave values for the parameters, Em~x, I, n and C, that minimized the sum of the squares of the differences between the dependent variable (E) in the above relationship and the experimental observations. In addition, a theoretical additive concentration-response curve for isoprenaline in the presence of the other compound tested was constructed as described by P~Sch and Holzmann (1980). A leftward shift of the experimental concentration-response curve relative to the theoretical additive curve implies synergism with isoprenaline. Significant changes in the ICs0 values of isoprenaline caused by the presence of a nitrovasodilator or an inhibitor of cAMP phosphodiesterase were evaluated by two-sided paired t-tests after logarithmic transformation of the results. Supra-additive effects of single concentrations of isoprenaline and other compounds were detected by comparison of the sums of the individual effects with the combined responses, using a repeated measures analysis of the variance of the results from individual aortic rings in at least 3 separate experiments.
3. Results
3.1. Effects on the relaxation of aortic smooth muscle Isoprenaline (30 nM-2 ktM) caused concentration-dependent relaxation of precontracted aortic rings with an IC50 of 211 _+ 21 nM (mean_+ S.E.M., 13 determinations, each based on 3 aortic rings). However, the maximum relaxation varied in different rings, ranging from 30 to 80% of the phenylephrine-induced contraction. The relaxant effect of isoprenaline was freely reversible, as demonstrated by the addition of propranolol and by the reproducibility in the same aortic ring of successive contractions induced by phenylephrine, following re-
T ±
oo 4 ()
~ -~ X
eo o
v
I O0
"g
I3
T
T
80
~-
60 -
20 - ~
0
.~ 0
\
,, l0
I 1000
100
Isoprenalinc
(riM)
Fig. 1. Effects of SNP and of cilostamide on isoprenaline-induced relaxation of rat aortic rings contracted by 100 nM phenylephrine. Cumulative concentration-response curves for isoprenaline were obtained in both the absence (O) and presence (11) of 0.25 nM SNP (A) or 20 nM cilostamide (B), as described in Materials and methods. Values are means:l:S.E.M, from 3 aortic rings. The theoretical additive concentration-response curves are also shown ( . . . . . . ) (see Materials and methcxts).
moval of both phenylephrine and isoprenaline by washing. lsoprenaline caused more marked relaxations in the presence of 0.25 nM SNP or of 30 nM SIN-l, concentrations chosen to have small relaxant effects alone. The potentiation of the action of isoprenaline by the nitrovasodilators was demonstrated by a leftward and upward shift of the isoprenaline concentration-response curves relative to the theoretical additive curves (e.g. fig. 1A); this is indicative of synergism (PtSch and Holzmann, 1980). The corresponding 3-fold reductions in the ICs0 values for isoprenaline, which were observed in the presence of either SNP or SIN-l, were statistically significant (table 1). In the presence of 0.25 nM SNP, which on average caused only 16% relaxation alone, the maximum relaxation observed with isoprenaline increased from 55 to 84% (mean values from 3 experiments). However, both under these conditions and with a lower isoprenaline concentration (96 nM, fig. 2), the effects of isoprenaline and SNP were not significantly supra-additive. Similar results were obtained with SIN-1 (fig. 2).
238 TABLt" 1 Effects of low concentrations of nitrovasodilators and of inhibitors of cAMP phosphodiesterase on the ability of isoprenaline to relax rat aortic rings previously contracted by 100 nM phenylephrine. IC5. values for isoprenaline were determined from cumulative concentration-response curves, as described under Materials and methods. Each experiment was similar to that shown in fig. 1 and yielded paired values for the ICso of isoprcnaline in the absence and presence of the indicated compounds. Mean values _+S.E.M. are gi',en from the numbers of experiments indicated: the significance of changes in I('s. for isoprenaline were evaluated by two-sided paired t-tests after logarithmic transformation of the results: " P < 0.05, ~' P < 0.0l. The ratio of IC50 values for isoprenaline obtained in the absence and presence of each compound studied indicates the extent of synergism. Compound
No. of
I('s of isoprenaline (nM) Ratio of
studied
expts.
Without
With
compound
conlpound
181- 42 232 L3I 251 + 59 193._+45
69_--20~' 74+ 4 ~' 68 ~- 6 " 101 +25 ~'
SNP(0.25 nM) SIN-I (30nM) ('ilostamide(20 nM) Ro 20-1724(10 gM)
4 3 3 3
I(',, values 2.6 3.1 3.7 1.9
Since the effects of the nitrovasodilators on isoprenaline-induced relaxation were relatively modest, we compared them with those obtained using the cAMP phosphodiesterase inhibitors, cilostamide and Ro 201724, which are known to inhibit cAMP breakdown in rat aortic smooth muscle (Schoeffter et al., 1987). Concentrations of these compounds selected to have relaxant effects similar to those obtained with 0.25 nM SNP or 30 nM SIN-1 alone, also shifted the isoprenaline concentration-response curves to the left of the theoretical additive curves and increased the maximum relaxation obtained with isoprenaline (e.g. fig. I B). In
these studies, cilostamide appeared to be slightly more effective than Ro 20-1724. The 2- to 4-fold shifts in the ICs~~ for isoprenalinc caused by these cAMP phosphodiesterasc inhibitors were also statistically significant and similar to those seen with SNP or SIN-1 (table 1). In 3 experiments, similar to that shown in fig. I B, 20 nM cilostamidc increased the maximum relaxation achieved with isoprcnalinc from an average of 61 to 92%, although alone it relaxed aortic smooth muscle by only 19%. With cilostamidc, but not Ro 20-1724, a significant supra-additive effect was observed in the presence of 96 nM isoprenaline (fig. 2). Although the potentiating effect,,, of the nitrovasodilators or c A M P phosphodiestcrasc inhibitors on the action of isoprenaline were always significant with respect to changes in the 1Cs~~values for isoprenaline, and supra-additive effects of low concentrations of the compounds were sometimes seen, the synergism was not as strong as that observed when the effects of nitrovasodilators and prostaglandins on platelet aggregation were studied (Lcvin ct al., 1982: De Caterina et al.. 1988: Willis ct al., 1989: Mauricc and ttaslam. 1990). To address this possible discrepancy, we carried out experiments on the inhibition of aortic smooth muscle contraction, which is probably more closely comparable to inhibition of platclct aggregation than smooth muscle relaxation. 3.2. Effects on the contraction of aortic smooth muscle
Incubation of aortic smooth muscle with isoprcnaline for 30 s caused a concentration-dependent inhibition of contraction induced by phcnylephrine (e.g. fig. 3). This
1 O0
0
,I 0 ¢.)
20
,'; N I '
,'-:,IN - 1
Ciloslarnide
I,~o 2 0 - 1 7 2 , 1
Fig. 2. Effects of nitrovasodilators and of cAMP phosphodiesterase inhibitors on the relaxation by 96 nM isoprenalinc of rat aortic rings precontracted by I(X) nM phenylephrine. "['he compounds studied were SN P (0.25 n M), SIN- 1 (30 nM), cilostamide (20 nM) and Ro 20- ] 724 ( I 0 ~M). Results with 96 nM isoprenaline are from cumulative concentration-response curves similar to those shown in fig. 1. The first bar in each group shows the percent relaxation with the indicated compound alone ([]), the second, the relaxation of the same aortic rings with isoprenaline alone (11~)and the third, the relaxation induced by the combined action of both compounds (I). Results are means + S.E.M. from 3 experiments. each with 3 or 4 aortic rings. The supra-additive effect of cilostamide and isoprenaline is indicated: * P < 0.05.
239
action of isoprenaline required slightly higher concentrations of the compound than relaxation of precontracted aortic smooth muscle. An lC.so value for isoprenaline of 349 + 26 nM was obtained (mean + S.E.M., 15 determinations each based on 3 or 4 aortic rings). Under the conditions of our experiments, 10- to 50-fold higher concentrations of nitrovasodilators (and cAMP phosphodiesterase inhibitors) were required to induce a weak ( < 30%) inhibition of contraction than to induce a similar extent of relaxation. Simultaneous addition of such concentrations of SNP or SIN-1 with isoprenaline caused marked leftward shifts in the isoprenaline concentration-response curves relative to the theoretical additive curves (e.g. fig. 3A). This resulted in 9- to 10-fold reductions in the IC~() values for isoprenaline (table 2), about 3-fold the effect observed on relaxation (table 1). Since the maximum inhibition of contraction obtained with isoprenaline was close to 100c/c, no upward shift in the concentration-response curves was possible. However, clear supra-additive effects of either SNP or SIN-I were observed in the presence of 100 nM isoprenaline (fig. 4). Both of the c A M P phosphodiesterase inhibitors studied (cilostamide and Ro 20-1724) interacted synergistically with isoprenaline to inhibit phenylephrine-induced contractions. A concentration of cilostamide (1 gM), which alone had a similar effect to 5 nM SNP or 1 ~M SIN-l, shifted the isoprenaline concentration-response curve to essentially the same extent as did these nitrovasodilators (table 2). Ro 20-1724 (100 gM) was only slightly less effective than cilostamide (table 2). Supra-additive effects on contraction were also seen with 100 nM isoprenaline and the above concentrations c)f these inhibitors of
I 10 0
-
\\
v
i
i
A
60
~<
40
2O a.,
! 00 -
It
_
"z
-2 •="=
60
'1;,
,
_
4(1
2O
~
\
0 ~====~
0
10
lsoprerlaline
I
I
100
1000
(riM)
F i g 3. F,ffects of SNP and of cilostanlide on the ability of isoprenMine to inhibit phcny[ephrine-induced contractions of rat aortic rings. The indicated concentrations of isoprenaline were added without (O) or ',,,ith (In) 5 nM SNP IA) or 1 p.M cilostamide (B). After 30 s, rings were contracted with I(X) nM phenylephrine. Contractions were measured after a further 2 rain, and were expressed as percentages of controls with phenylephrine alone. Values are means + S.E.M. from 4 rings. The theoretical additive concentration-response curves are also shown f . . . . . . ) (see Materials and methods1.
I O0
~-
80
60
411
SNI'
SIN-I
Cil,)st amide
Ro 2 0 - 1724
Fig. 4. Effects of nitrovas(xtilators and of cAMP phosphodiesterase inhibitors on the abilily of 100 nM isoprenaline to inhibit the contraction of rat aortic rings by 100 nM phenylephrine. The compounds studied were SNP (5 nM), SIN-I (1 #M), cilostamJde (1 p.M) and Ro 20-1724 (100 p.M). Rings were incubated for 30 s with isoprenaline wilhout or with one of the above compounds. Phenylephrme was then added and the resulting contraction measured after a further 2 min. The first bar in each group shows the inhibition of contraction b,, the indicated compound alone ([]), the second, the inhibition of contraction by isoprenaline alone (~) and the third, the combined action of both compounds (11). Values are means 4: S.E.M. from 3 to 6 experiments, each with 3 or 4 aortic rings. Supra-additive effects of the drug combinations studied are indicated: * P < 0.05: * * P
< 0.01.
240 "FABLE 2 it'ffects of low concentrations of nitrovasodilators and of inhibitors of cAMP phosphtxzliesterase on the ability of isoprenaline to inhibit the contraction of rat aortic rings by I00 nM phenylephrine, lsoprenaline was added without or with each of the listed compounds 30 s before phenylephrine and the contraction of the aortic rings was measured after a further 2 min. In each experiment. I('sq, values for isoprenaline in the absence and presence of one of the compounds listed were calculated from concentration-response curves obtained from the same aortic rings, as described under Materials and methods and shown in fig. 2. Results are means _+.SE.M. from the numbers of experiments shown; the significance of changes in [('5~ values for isoprenaline were evaluated by two-sided paired t-tests after logarithmic transformation of the results; " P < 0.02. h p < 0.001. The ratio of I('s~ values for isoprenaline in the absence or presence of other drugs indicates the extent of synergism. ('ompounds
No. of
IC5oof isoprenaline (nM) Ratio of
studied
expts.
Without
With
compound
compound
334__-49 398 k55 318L57 356_+57
37__+ 5 h 4 2 - 1 0 ;' 34± 4 " 44_+ 3 h
SNP(5 nM) SIN-1 (1 p.M) Cilostamide(1 p.M) Ro 20-1724 (100 #M)
5 3 3 4
1('5" values 90 9.5 9.4 g.l
c A M P phosphodiesterase (fig. 4). Our observation that the nitrovasodilators and cAMP phosphodiesterase inhibitors potentiated the inhibition of aortic smooth muscle contraction by isoprenaline to essentially the same extent, and that these compounds also had equal, though weaker, effects on the relaxant action of isoprenaline, suggests that these two different groups of compounds may ultimately act through a common inhibitory mechanism.
4. Discussion
In experiments with rat aortic rings from which the endothelium had been removed, we showed that the nitrovasodilators, SNP and SIN-l, potentiate the ability of isoprenaline to relax precontracted aortic smooth muscle and, when added with isoprenaline before the contractile stimulus, exert potent synergistic inhibitory effects on contraction. In contrast, a study of the interactions between SNP and iloprost in vascular smooth muscle failed to detect synergism (Lidbury et al., 1989). Although use of rabbit mesenteric and coeliac arteries in the latter study, rather than rat aorta, could account for the different results obtained, other considerations may be more important. In our experiments, detection of the synergistic effects of nitrovasodilators and isoprenaline on relaxation depended on comparison of results obtained in the same aortic rings incubated with and without test compounds. Moreover, potentiation of the effects of isoprenaline by either the nitrovasodilators or the cAMP phosphodiesterase inhibitors was always much more marked when inhibition of contraction,
rather than relaxation, was studied. The reasons for the latter difference are not clear, but both methodological and physiological factors could be important. First, prolonged exposure of precontracted smooth muscle to nitrovasodilators or cAMP phosphodiesterase inhibitors could have inhibitory, as well as stimulatory, effects on the ability of the tissue to respond to isoprenaline. Thus, pretreatment of other rat tissues with c A M P analogues has been shown to increase cAMP phosphodiesterase activity and to reduce the effects of agonists that stimulate adenylyl cyclase (Gettys et al., 1987). In addition, the platelet cGMP-inhibited cAMP phosphodiesterase appears to be activated by cAMP-dependent phosphorylation (MacPhee et al., 1988: Grant et al., 1988). The continued presence of phenylephrine could also have adversely affected the actions of the other compounds studied, for example through the activation of protein kinase C. Second, there may be critical differences in the state of the vascular smooth muscle before and after induction of contraction that affect the responses to the vasodilator drugs studied. Since increases in intracellular C a : " concentration and the phosphorylation of myosin light chain are important in the contraction of vascular smooth muscle ( K a m m and Stull, 1985: Murphy', 1989), compounds that block these processes will inhibit contraction. On the other hand, relaxation of smooth muscle, which involves reversal of a latch state, is less dependent on the inhibition of ('a e" mobilization or on dephosphorylation of myosin (l)illon et al.. 1981; Hal and Murphy, 1988). Increases in cGMP, presumably acting through cGMP-dependent protein kinase, reduce the level of myosin phosphorylation in smooth muscle (Rapoport et al.. 1983), activate the plasmalemmal Ca 2 '-ATPase (Rashatwar et al., 1987: Vrolix et al.. 1988, Cornwell and l,incoln, 1989) and inhibit the G-protein-mediated stimulation of phosphoinositide hydrolysis (Hirata et al., 1990). ()n the other hand, cAMP-dependent protein kinase activity' increases the concentration of ('a2'-calmodulin required for activation of myosin light-chain kinase (Adelstein et al., 1978) and the C a 2" sensitivity of smooth muscle C a 2 ' - d e p e n d e n t potassium channels (Sadoshima et al., 1988). Thus, in view of the different biochemical mechanisms that may be involved in the inhibition of contraction and relaxation, it is entirely conceivable that activation of one or the other cyclic nucleotide system, or both together, could have distinct effects. The marked synergism between the actions of the nitrovasodilators and isoprenaline reported here for the inhibition of rat smooth muscle contraction is analogous to that observed in platelets between nitrovasodilatots and activators of platelet adenylyl cyclase (Levin et al., 1982: De Caterina et al., 1988: Willis et al., 1989: Maurice and Haslam. 1990). In platelets, we recently'
241
showed that this synergism resulted from the inhibition by cGMP of cAMP breakdown by a cGMP-inhibited low K mcAMP phosphodiesterase (Maurice and Haslam, 1990). An enzyme with kinetic characteristics and a sensitivity to inhibitors very similar to those of the platelet enzyme has been shown to be present in rat, guinea pig, and rabbit aorta (Schoeffter et al., 1987: Silver et al., 1988; Souness et al., 1989; Ahn et al., 1989). The possibility therefore exists that the synergism between nitrovasodilators and isoprenaline reported here results from inhibition by c G M P of this same enzyme in rat aortic smooth muscle. Much of our data is consistent with this. First, both SNP and SIN-I had potentiating effects on the action of isoprenaline that were virtually indistinguishable from those obtained using cilostamide, whether relaxation or inhibition of contraction was studied. The latter compound has been shown to be a selective inhibitor of the cGMP-inhibited low K m cAMP phosphodiesterase in platelets (Hidaka et al., 1979; MacPhee et al., 1986) and to increase cAMP, but not cGMP levels in rat aortic smooth muscle at concentrations that caused 50% relaxation (Schoeffter et al., 1987). The possibility that the nitrovasodilators do potentiate the effects of isoprenaline by inhibiting the cGMP-inhibited enzyme is also consistent with evidence that cilostamide appears to decrease relaxation by SNP in rat aortic smooth muscle (Schoeffter et al., 1987). Ro 20-1724, a compound known to inhibit a different low K m cAMP phosphodiesterase that is unaffected by c G M P (Beavo, 1988), was slightly less effective in potentiating the actions of isoprenaline. This observation suggests that cAMP breakdown in vascular smooth muscle depends on the activities of both low K m cAMP phosphodiesterases. E D R F has been identified as nitric oxide (Palmer et al., 1987) or more probably an S-nitrosothiol (Myers et al., 1990) and appears to relax vascular smooth muscle by the same cGMP-dependent mechanism as do the nitrovasodilators (Murad, 1986). A synergism similar to that seen with nitrovasodilators should therefore be observed between the effects of E D R F and activators of adenylyl cyclase on vascular smooth muscle. It is therefore significant that physiological stimuli can cause a coupled release of both E D R F and PGI2 from the vascular endothelium (De Nucci et al., 1988) and that an interaction has been observed between E D R F in an endothelial superfusate and PGI 2 in causing the relaxation of pig coronary artery (Shimokawa et al., 1988). These results are consistent with the known synergistic effects of E D R F and PGI 2, as inhibitors of platelet aggregation (Radomski et al., 1987; Macdonald et al., 1988). The present results extend evidence of significant synergistic interactions between nitrovasodilators and activators of adenylyl cyclase in platelets to vascular smooth muscle and provide indirect evidence that the
same mechanism, namely inhibition by cGMP of a low K m cAMP phosphodiesterase, may be responsible. Although no synergism was observed between the effects of nitroglycerin and a PGI 2 analogue on the mean blood pressure and heart rate of spontaneously hypertensive rats (,Willis et al., 1989), the possibility of effects specific to particular vascular beds has not yet been investigated. Since a cGMP-inhibited cAMP phosphodiesterase similar or identical to the platelet enzyme has also been shown to be present in both cardiac muscle (Harrison et al., 1986) and tracheal smooth muscle (Torphy and Cieslinski, 1990), synergistic interactions between the effects of activators of guanylyl and adenylyl cyclases are also likely to be found in these tissues. In general, synergism may be expected whenever the cGMP-inhibited phosphodiesterase accounts for a large fraction of cAMP breakdown. However, the physiological significance and pharmacological value of synergism between activators of guanylyl and adenylyl cyclases in platelets, vascular smooth muscle and other tissues remain to be determined. In principle, selective synergistic effects of therapeutic interest might be obtained through the use of combinations of pharmacological agents, each of limited specificity, that together activate both adenylyl and guanylyl cyclases only in the target tissue.
Acknowledgement This work was supported by a Grant-in-aid (1".1265) from the Heart and Stroke Foundation of Ontario.
References Adelstein, R.S., M.A. Conti, D.R. Hathaway and C.B. Klee, 1978, Phosphorylation of smooth muscle myosin light chain kinase by the catalytic subunit of adenosine 3':5'-monophosphate-dependent protein kinase, J. Biol. Chem. 253, 8347. Ahn, H.S., W. Crim, M. Romano, E. Sybertz and B. Pitts, 1989, Effects of selective inhibitors on cyclic nucleotide phosphtnliesterases of rabbit aorta, Biochem Pharmacol. 38, 3331. Azuma, H., M. Ishikawa and S. Sekizaki, 1986, Endothelium-dependent inhibition of platelet aggregation, Br. J. Pharmacol. 88, 411. Beavo, J.A., 1988, Multiple isozymes of cyclic nucleotide phosphodiesterase, Adv. Sec. Mess. Phosphoprot. Res. 22, 1. Cornwell, T.L. and T.M. Lincoln, 1989, Regulation of intracellular Ca 2+ levels in cultured vascular smooth muscle cells. Reduction of Ca 2~ by atriopeptin and 8-bromo-cyclic GMP is mediated by cyclic GMP-dependent protein kinase. J. Biol. Chem. 264. 1146. De Caterina, R., D. Giannessi, W. Bernini and A. Mazzone, 1988, Organic nitrates: direct antiplatelet effects and synergism with prostacyclin. Antiplatelet effects of organic nitrates, Thromb. Iiaemost. 59, 207. De Nucci, G., R.J. Gryglewski, T.D. Warner and J.R. Vane, 1988, Receptor-mediated release of endothelium-derived relaxing factor and protacyclin from bovine aortic endothelial cells is coupled, Prec. Natl. Acad. Sci. U.S.A. 85, 2334. Dillon, P.F., M.O. Aksoy, S.P. Driska and R.A. Murphy, 1981,
242
Myosin phosphorylation and the cross-hridge cycle in arterial smooth muscle, Science 211, 495. Furchgott. R.F. and J.V. Zav, adTki, 1980, ] h e obligatory role of endothelial cells in the relaxation of arterial smooth muscle h~, acetylcholine, Nature 289, 373. (,}ettys, T.W., P.F. Blackmore, J.B. Redmon, S.J. Beebe and .1.I). Corbin, 1987, Short-term feedback regulation of c A M P bv accelerated degradation in rat tissues, J. Biol. ('hem. 262, 333. (Jlusa. t!.. F. Markwardt and J. Sti]rzehecher, 1974, I'ffects of sodium nitroprusside and other pentacyanonitrosyl complexes on platelet aggregation. Haemostasis 3, 249. Orant, P.G. anti R.W. Colman, 1984, Purification and characteri,'alion of a human platelet cyclic nucleotide phosphc,diesterase. Bi,x:hemistry 23, lg01. Grant, P.G., A.F. Mannarino and R.W. ( o l m a n , 1988, cAMP-mediated phosphorylation of the low K,,, cAMP phosphodiesterase markedly stimulates its catalvtic activity, Proc. Natl. A c a d Sci. U.S.A. 85, 9071. ttai, ('.-M. and R.A. Murphy. 1989, Cross-bridge phosphorylation and regulathm of latch state m sm(x~th muscle, Am..J. Phvsiol 254, 4"99. l|ardman, J.C}., 1994, Cyclic nucleotides and regt, lation of vascular smooth muscle. J. Cardiovasc. Pharmacol. 6. $639. tlarrison, S.A.. D.H. Reifsnyder, B. Gallis. G.(;. Cadd and J.A. Bcavo, 1986, Isolation and characteri:,'ation of bovine cardiac muscle cOMP-inhibited phosphodiesterase: A receptor for new cardiotonic drugs, Mol. Pharmacol. 29. 506. Ilashimoto. K., N. Taira, M. Hirata and M. Kokuhun, 1971, The mode of hypotensive action of newly synthesized svdnonimine derivati',es, Arzneim. Forsch. 21. 1329. tlashnn, R.J., M.M.I,. Davidson, '1. l)avies, J.A. l,ynllam and M.I). McClenaghan, 1979, Regulation of b],,)od platelet function b', cvclic nucleotides. Ad',. ( ' y c l Nucl. Res. 9, 533. tlidaka, I1., H. ttayashi, It. Kohri, Y. Kimura, T. Hosokawa, "r. lgav.a and Y. Saitoh, 1979, Selective inhibitor of platelet cyclic adenosine monnphosphate phosphodiesterase, cilostamide, inhibits platelet aggregation, J. Pharmacol. Exp. "fher. 211.26. ttirata, M., K.P. Kohse, C.-tl. ('hang, T. lkebe and t r. Murad, 1990, Mechanism of cyclic (.;MP inhibition of inositol phosphate formalion in rat aorta segments and cultured bovine aortic smooth muscle cells. J. Biol. Chem. 265, 1268. K a m m , K.f!. and J.'l. Stull, 1985. -['he function of myosin and myosin light chain kinase phosphorylation in smoc, th muscle. Ann. Re,,'. Pharmacol. Toxicol. 25, 593. Levin, R.I., B.B. Weksler and E.A. Jaffe, 1982. The interaction of sodium nitroprusside with h u m a n endothelial cells and platelets: Nitroprusside and prostacyclin synergistically inhibit platelet function, Circulation 66, 1299. Lidhury. P.S., E. Antunes, G. De Nucci and J . R Vane, 1999. Interactions of iloprost and sodium nitroprusside on vascular smooth muscle and platelet aggregation. Br. J. Pharmacol. 99, 1275. Macdonald, P.S., M.A. Read and G.J. Dusting, 1998, Synergistic inhibition of platelet aggregation by endothelium-derived relaxing factor and prostacyclin, Thromb. Res. 49. 437. Mad"hee, C.tt., S.A. llarrison and J.A. Beavo, 1986, Immunological identification of the major platelet low K m cAMP phosphodiesterase: Probable target for anti-thrombotic agents, Proc. Natl. Acad. ,~i. U.S.A. 83, 6660. MacPhee, ('.H.. D.H. Reifsnyder, T.A. Moore, K.M. I,erea and J.A. Beavo, 19g8, Phosphory'lation results in activation of a c A M P phosphodiesterase in h u m a n platelets, J. Biol. Chem. 263, 10353. Maurice, I).1t. and R.J. Haslam, 1990, Molecular basis of the synergistic inhibition of platelet function by nitrovasodilators and activators of adenylate cyclase: Inhibition of cyclic A M P breakdown by cyclic GMP, Mol. Pharmacol. 37, 671. Murad, F., 1986. Cyclic guanosine monophnsphate as a mediator of vas¢~:lilation, J. Clin. Invest. 78, 1.
Murphy. R.A.. 1999, (,'ontraclion hi :,.in,,uath muscle cells. Ann. Re',. Physiol. 51. 275. M'.'er~,, P.R., R.I.. Minor, R. (;uerra, J.N. Bales and 11.(;. Harrison, 1990, Vasorelaxant properties of the endothelium-derived relaxing factor more closely resemble S-nilrosocvsteine than nitric oxide, Nature 345, 161. Ni~,hikawa. M.. M. Kanamon and H. 11idaka. 1982, Inhibition of platelet aggregation and stimulation of guanylate c',clasc b~ an antianginal agent molsid~mline alld its metabolites, J. Pharmacol. Exp. Thor. 220. 183. Palmer, R.M.J., A.(;. I"errige and S. Moncada, 1987. Nitric oxide release accounts for the biological activity of endothelium-dcrived relaxing fi,ctor. Nature 327. 524. P,Sch. G. and S. ttolzmann. 1980, Quantitative estimation of overadditive and underadditive drug effects bv means of theoretical. additi,.'e dose-response curves, J. Pharmacol. Moth 4. 179 Radomski, M.W.. R.M.J. Palmer anti S. Moncada. 1997. "['he anti-aggregating properties of vascular endothelium: interactions bet,*een prostac?,clin anti nitric oxide, Br. J. Pharmacol. 92. 6.t9. Rapoport, R.M., M.B. l)raznin and F. Murad, 1983. I'-indothelium-dependent relaxation in rat ,:lorta may he mediated through cyclic GMP-dependent protein phosphor',lation, Nature 306. 174. Rashat,*,ar, S.S.. "I'.L. ('ornwell and I.M. l,mcoln, 1997, Effects of 8-bromo-cGMP on ('a z" levels in ',ascular smooth muscle cells: Possible regulation of Ca" '-ATPase b', cGMP-dependent protein kinase, Proc. Nat[. Acad. Sci. I, LS.A. 84, 5685. Sadoshinla, J.-I., N. Akaike, I-I. Kana~de and M. Nakamura, 198., ('relic A M P modulates Ca-activated K channel in cultured sm~u~th muscle cells of rat aortas, Am. J. Phvsiol. 255. H754. SchrGr, K.. It. l)arius, R. Matzkv and R. ()hlendorf. 1981, "[he antiplatelet anti cardiovascular actions of a ne',~, carbacvclin tterivative (ZK 363741 equipotent to I)GI, in vitro, Nat.n',nSchmiedeb. Arch. Pharmacol. 316, 252. Schoeffter, P., ('. l,ugnier, F. Demcs~,.-Waeldele and J.('. Stoclet, 1987, Role of c~,clic AMP- and cyclic (;MP-phosphodiesterases in the control of cvclic nucleotide levels and smooth muscle tone in rat isolated aorta. A studv with ~,elective inhibitors. Biochcm Pharmacol. 36. 3965. Shimokawa. H., N.A. Flavahan, R.R. Lorenz and P.M. Vanhoutte. 19gg, Prostac,.clm releases endothelium-derived relaxing factor and potent)ates its action m coronary, arteries of the pig, Br. ,I. Pharmacol. 95, 1197+ Silver, P.J., R.IL kepore. B. ()'Connor, B.M+ l+emp, I..T. Ilamel. R.G. Bentley, and A.L. H a m s , 1989, Inhibition of the Io'a, Kn, cyclic A M P phosph~uJiesterase and activation of the c','clic A M P system in vascular smooth muscle by milrinone, J. Pharmacol. I'×p. Thor. 247, 34. Souness, J.l'.. R. Brazdil, B.K. l)ic,cee and R. Jordan. 1989, Role of selecti~,e cyclic ( i M P phosphodiesterase inhibition in the mvorelaxant actions of M&B 22,948, MY-5445, vinpc~etine and 1-methyl-3-isobut~,l-S-(methylamim~)xanthme, Br J. Pharmacol. 98, 725. Tateson, J.I-., S. Moncada and J.R. Vane, 1977. Effects of prostac~.clin (P(;X) on cyclic A M P concentrations in human platelets. Prostaglandin.s 13, 389. Torphy, T.J. and L.B. ('ieslinski. 1990, ('haracterl;'ation and selective inhibition of cyclic nudeotide phosphodiesterase isozvmes m canine tracheal snlootll muscle, Mol. Pharmacol. 37, 2(16. Vrolix, M., I,. Rae,,maekers, F. Wuvtack, F. t t o f m a n n and R. ('asteel,. 1988. Cyclic (}MP-dependent protein kinase stimulates the plasmalemmal Ca 2 " pump of sm(x',th muscle via phosphorylatzon of phosphatidylinositol, t~.it~.'hem. J. 255, 855. Willis, A t . , I).L. Smith, M. I,oveday, J. Fulks, C.lt. Lee, L. ttedlev and I). VanAntwerp, 1999. Selecti',e anti-platelet aggregation synergism bet'~,een a prostacyclin-mimetic, RS93427 and the nitrodilators s
235
EJP 51632
Synergistic actions of nitrovasodilators and isoprenaline on rat aortic smooth muscle D o n a l d H. M a u r i c e 1, Denis C r a n k s h a w 2 a n d R i c h a r d J. H a s l a m i Department of Pathoh~gv and " Department of Obstetrics atut (~vnecoloKv. McMm'ter U'mcersitv. Ilamilton. Ontario. ('anada LS,V 3Z5
Received 18 June 1990, revised MS received 11 September 1990. accepted 25 September 1990
Previous studies have established that nitrovasodilators potentiate the inhibition of platelet function by activators of adenylyl cyclase, but uncertainty exists as to whether a comparable effect is seen in vascular smooth muscle. We initially studied the effects of the nitrovasodilators, sodium nitroprusside (SNP) and 3-morpholinosydnonimine (SIN-l), on the relaxation by isoprenaline of rat aortic smooth muscle that had been precontracted by phenylephrine. Concentrations of SNP (0.25 nM) and SIN-I (30 nM) that relaxed aortic smooth musclc less than 30% alone, caused significant (3-fold) decreases in the ICso values for isoprenaline. The cAMP phosphodiesterase inhibitors, cilostamide (20 nM) and Ro 20-1724 (10 p,M), caused comparable reductions in the IC5o values for isoprenaline. At these concentrations, each of the four compounds also increased the maximum relaxation achieved with isoprenaline. Even more marked synergistic interactions were observed between isoprenaline and either the nitrovasodilators or the cAMP phosphodiesterase inhibitors when these compounds were added simultaneously before contraction of aortic smooth muscle by phenylephrine. Thus, concentrations of SNP (5 nM), SIN-1 (1 #M), cilostamide (1 p.M) and Ro 20-1724 (100 btM) that inhibited contraction by less than 30% decreased the IC50 values for isoprenalinc by 8- to 10-fold. At the above concentrations, these compounds each caused a supra-additive inhibition of contraction when added with 100 nM isoprenaline. Thus. synergism between nitrovasodilators and isoprenaline, an activator of adenylyl cyclasc, could be detected in vascular smooth muscle and was particularly marked when inhibition of contraction was studied. This action of nitrovasodilators resembled that of inhibitors of cAMP phosphodiesterasc. Smooth muscle (aortic); Isoprenaline: Nitrovasodilator; Synergism; (Rat)
1. Introduction The nitrovasodilators, sodium nitroprusside (SNP) and 3-morpholinosydnonimine (SIN-l), and endothelium-derived relaxing factor ( E D R F ) inhibit platelet aggregation (Glusa et al., 1974; Nishikawa et al., 1982: Azuma et al., 1986) in addition to relaxing vascular smooth muscle (Hashimoto et al., 1971; Furchgott and Zawadzki, 1980). Both of these effects are associated with the activation of soluble guanylyl cyclase and increases in intracellular cyclic G M P (cGMP) (Murad, 1986). In contrast, vasodilators such as isoprenaline and prostacyclin (PGI2) exert their effects on vascular smooth muscle by stimulating adenylyl cyclase and increasing intracellular cyclic AMP (cAMP) (Hardman, 1984). Again, platelets are similar to vascular smooth muscle in that cAMP, as well as cGMP, mediates an inhibition of aggregation (Haslam et al., 1978). The most potent inhibitors of platelet function that activate
Correspondence to: R.J. Haslam, Department of Pathology, McMaster University. 120(I Main Street West. Hamilton. Ontario. Canada LgN 3Z5.
adenylyl cyclase are PGI~ and its analogues, such as iloprost (Tateson et al., 1977; SchrOr et al., 1981). Several reports have documented synergistic inhibitory effects on platelet aggregation of PGI~ or its analogues in combination with nitrovasodilators or E D R F (Levin et al., 1982: Radomski et al.. 1987; Macdonald et al., 1988: De Caterina et al., 1988; Willis et al., 1989). However, comparable synergistic effects were not observed in a study of the effects of SNP and iloprost on rabbit vascular smooth muscle (Lidbury et al., 1989), though evidence for synergism between E D R F and PGI 2 has been obtained in experiments with pig coronary arteries (Shimokawa et al., 1988). Recently, we determined the molecular basis of the synergistic inhibition of platelet aggregation by nitrovasodilators and activators of adenylyl cyclase (Maurice and Haslam, 1990). We showed that it resulted from the ability of c G M P to enhance the accumulation of cAMP in platelets by inhibiting a specific cGMP-inhibited cAMP phosphodiestcrase. This enzyme has been most fully characterized in platelets (Grant and Colman, 1984; MacPhee et al., t986) and cardiac muscle (Harrison et al., 1986), but a cAMP phosphodiesterase very similar in its kinetics and sensitivity to inhibitors is known to be present in vascular smooth muscle (Schoeffter et al.,
236 1987; Silver et al.. 1988; Souness et al., 1989: Ahn et al., 1989). We therefore re-investigated the possibility that synergistic effects similar to those seen in platelets could be demonstrated in vascular smooth muscle. The results showed that nitrovasodilators do potentiate the relaxation of rat aortic smooth muscle by isoprenaline and that these compounds exert even more potent synergistic effects when added prior to the contractile stimulus.
2. Materials and methods
2. I. Materials SNP, ( - )-isoprenaline ( + )-bitartrate. phenylephrine and carbachol were obtained from Sigma (St. Louis, MO, U.S,A.). Male Wistar-Kyoto (WKY) rats were obtained from Harlan Sprague-Dawley (Indianapolis, IN, U.S.A.). SIN-1 and Ro 20-1724 were provided by Casella-Riedel (Frankfurt am Main, F.R.G.) and F. Hoffmann La Roche (Nutley, N J, U.S.A.), respectively. Cilostamide was generously supplied by Professor H. Hidaka of Nagoya University (Nagoya, Japan). 2.2. Preparation of rat aortic rings Male WKY rats (225-350 g) were killed by a blow to the head. The thoracic aorta was rapidly removed and placed in a physiological salt solution (PSS), containing 1.16 mM MgSO 4, 2.5 mM CaCI~, 21.9 mM N a H C O > 1.16 mM N a H 2 P O 4, 4.6 mM KCI, 115.5 mM NaCI and 11.1 mM dextrose, that had been equilibrated with 95% 02-5% CO2. The aorta was cleaned of adherent fat and connective tissue and cut into 4.0 mm transverse rings. The endothelium was removed by gently rubbing the luminal surfaces with a wooden stick (Furchgott and Zawadzki, 1980). Rings were mounted in 10 ml jacketed organ baths using stainless steel wire and bathed at 3 7 ° C in PSS, which was gassed with 95% 02-5% CO,. All experiments were carried out with an initial tension of 2 g. Tension development was measured isometrically using a Grass FTO3 transducer and a Grass Model 7D polygraph. Rings were allowed to equilibrate for 2 h, during which they were repeatedly washed with PSS. During this period, the rings were also contracted 3 to 4 times with 30 mM KCI (until a maximal contraction was achieved). Removal of the endothelium was then verified by the inability of 1 p.M carbachol to induce relaxation of rings submaximally contracted with 100 nM phenylephrine. 2.3. Studies on the relaxation of aortic smooth rnuscle The concentration-response curves for the relaxant effects of isoprenaline were determined by cumulative
additions of the compound to submaximally contracted rat aortic smooth muscle. In all experiments, contraction was induced by phenylephrine: the concentration used (I00 nM) caused on average 80% of the maximal contraction, which was seen with 1 p.M. Each addition of isoprenaline was allowed to have its full effect before further additions were made (2 to 3 min). The relaxation caused by each addition was recorded and calculated as a percentage of the maximum possible relaxation (i.e. of the full increase in tension caused by phenylephrine). After 4 washes with PSS, the aortic rings were allowed to equilibrate for 90 min in fresh PSS before thev were used again. With this time interval, successive concentration-response curves for isoprenaline were identical. The effects of nitrovasodilators (SNP or SIN-I) or c A M P phosphodiesterase inhibitors (cilostamide or Ro 20-1724) on the ability' of isoprenaline to relax precontracted aortic smooth muscle were determined as follows. Each compound studied was added to the contracted rings at a specific concentration that caused less than 30% relaxation on all occasions. When the ring tension reached a new plateau (5-15 min), isoprenaline was added cumulatively and the total relaxation caused by each concentration was determined as a percentage of the initial increase in tension caused by phenylephrine. To reduce errors due to the inherent variability, of responses of aortic smooth muscle to vasodilators, concentration-response curves for isoprenaline in the absence and presence of each compound were in each experiment constructed front paired records obtained from the same aortic rings. 2.4.
Studies
on
inhibition of the contraction of aortic
smooth muscle The effects of nitrovasodilators or of cAMP phosphodiesterase inhibitors on the ability of isoprenaline to inhibit the contractions of rat aortic smooth muscle induced by phenylephrine were studied as follows. Aortic rings were incubated for 30 s with vasodilator compounds, either singly or in combination. Phenylephrine (100 nM) was then added and the tension developed after 2 min was measured. Following each contraction, the PSS was changed 4 times and the rings were allowed to equilibrate in fresh PSS for 30 min. This time interval permitted the generation of identical control contractions and virtually identical responses to isoprenaline throughout the experiment, provided low concentrations of isoprenaline were studied first. Using this protocol, it was possible to test up to 15 different combinations of compounds on each aortic ring and to obtain concentration-response curves for the inhibition of contraction by isoprenaline in the presence and absence of single concentrations of other compounds. The latter were used at selected concentrations that caused less than a 30~ inhibition of contraction. Inhibitions of
237
contraction were calculated as percentages of the control contractions for each aortic ring; 3 or 4 rings received identical treatments in each experiment.
1
loo
-
\~x
I
I
I
A
80
2.5. Analysis of results Concentration-response curves were constructed for the relaxation or inhibition of the contraction of rat aortic smooth muscle by isoprenaline in the absence and presence of nitrovasodilators or inhibitors of c A M P phosphodiesterase. Curves were fitted to the experimental results using a computer program (SigmaPlot 4.02A, Jandel Scientific, Corte Madera, CA, U.S.A.) and the following relationship, E = E .... • xn/(x n + I n) + C, where E is the effect, Em~x the maximum effect, x the concentration of isoprenaline, I the ICs0 of isoprenaline, n the Hill slope and C the effect observed in the absence of isoprenaline. This computer program also gave values for the parameters, Em~x, I, n and C, that minimized the sum of the squares of the differences between the dependent variable (E) in the above relationship and the experimental observations. In addition, a theoretical additive concentration-response curve for isoprenaline in the presence of the other compound tested was constructed as described by P~Sch and Holzmann (1980). A leftward shift of the experimental concentration-response curve relative to the theoretical additive curve implies synergism with isoprenaline. Significant changes in the ICs0 values of isoprenaline caused by the presence of a nitrovasodilator or an inhibitor of cAMP phosphodiesterase were evaluated by two-sided paired t-tests after logarithmic transformation of the results. Supra-additive effects of single concentrations of isoprenaline and other compounds were detected by comparison of the sums of the individual effects with the combined responses, using a repeated measures analysis of the variance of the results from individual aortic rings in at least 3 separate experiments.
3. Results
3.1. Effects on the relaxation of aortic smooth muscle Isoprenaline (30 nM-2 ktM) caused concentration-dependent relaxation of precontracted aortic rings with an IC50 of 211 _+ 21 nM (mean_+ S.E.M., 13 determinations, each based on 3 aortic rings). However, the maximum relaxation varied in different rings, ranging from 30 to 80% of the phenylephrine-induced contraction. The relaxant effect of isoprenaline was freely reversible, as demonstrated by the addition of propranolol and by the reproducibility in the same aortic ring of successive contractions induced by phenylephrine, following re-
T ±
oo 4 ()
~ -~ X
eo o
v
I O0
"g
I3
T
T
80
~-
60 -
20 - ~
0
.~ 0
\
,, l0
I 1000
100
Isoprenalinc
(riM)
Fig. 1. Effects of SNP and of cilostamide on isoprenaline-induced relaxation of rat aortic rings contracted by 100 nM phenylephrine. Cumulative concentration-response curves for isoprenaline were obtained in both the absence (O) and presence (11) of 0.25 nM SNP (A) or 20 nM cilostamide (B), as described in Materials and methods. Values are means:l:S.E.M, from 3 aortic rings. The theoretical additive concentration-response curves are also shown ( . . . . . . ) (see Materials and methcxts).
moval of both phenylephrine and isoprenaline by washing. lsoprenaline caused more marked relaxations in the presence of 0.25 nM SNP or of 30 nM SIN-l, concentrations chosen to have small relaxant effects alone. The potentiation of the action of isoprenaline by the nitrovasodilators was demonstrated by a leftward and upward shift of the isoprenaline concentration-response curves relative to the theoretical additive curves (e.g. fig. 1A); this is indicative of synergism (PtSch and Holzmann, 1980). The corresponding 3-fold reductions in the ICs0 values for isoprenaline, which were observed in the presence of either SNP or SIN-l, were statistically significant (table 1). In the presence of 0.25 nM SNP, which on average caused only 16% relaxation alone, the maximum relaxation observed with isoprenaline increased from 55 to 84% (mean values from 3 experiments). However, both under these conditions and with a lower isoprenaline concentration (96 nM, fig. 2), the effects of isoprenaline and SNP were not significantly supra-additive. Similar results were obtained with SIN-1 (fig. 2).
238 TABLt" 1 Effects of low concentrations of nitrovasodilators and of inhibitors of cAMP phosphodiesterase on the ability of isoprenaline to relax rat aortic rings previously contracted by 100 nM phenylephrine. IC5. values for isoprenaline were determined from cumulative concentration-response curves, as described under Materials and methods. Each experiment was similar to that shown in fig. 1 and yielded paired values for the ICso of isoprcnaline in the absence and presence of the indicated compounds. Mean values _+S.E.M. are gi',en from the numbers of experiments indicated: the significance of changes in I('s. for isoprenaline were evaluated by two-sided paired t-tests after logarithmic transformation of the results: " P < 0.05, ~' P < 0.0l. The ratio of IC50 values for isoprenaline obtained in the absence and presence of each compound studied indicates the extent of synergism. Compound
No. of
I('s of isoprenaline (nM) Ratio of
studied
expts.
Without
With
compound
conlpound
181- 42 232 L3I 251 + 59 193._+45
69_--20~' 74+ 4 ~' 68 ~- 6 " 101 +25 ~'
SNP(0.25 nM) SIN-I (30nM) ('ilostamide(20 nM) Ro 20-1724(10 gM)
4 3 3 3
I(',, values 2.6 3.1 3.7 1.9
Since the effects of the nitrovasodilators on isoprenaline-induced relaxation were relatively modest, we compared them with those obtained using the cAMP phosphodiesterase inhibitors, cilostamide and Ro 201724, which are known to inhibit cAMP breakdown in rat aortic smooth muscle (Schoeffter et al., 1987). Concentrations of these compounds selected to have relaxant effects similar to those obtained with 0.25 nM SNP or 30 nM SIN-1 alone, also shifted the isoprenaline concentration-response curves to the left of the theoretical additive curves and increased the maximum relaxation obtained with isoprenaline (e.g. fig. I B). In
these studies, cilostamide appeared to be slightly more effective than Ro 20-1724. The 2- to 4-fold shifts in the ICs~~ for isoprenalinc caused by these cAMP phosphodiesterasc inhibitors were also statistically significant and similar to those seen with SNP or SIN-1 (table 1). In 3 experiments, similar to that shown in fig. I B, 20 nM cilostamidc increased the maximum relaxation achieved with isoprcnalinc from an average of 61 to 92%, although alone it relaxed aortic smooth muscle by only 19%. With cilostamidc, but not Ro 20-1724, a significant supra-additive effect was observed in the presence of 96 nM isoprenaline (fig. 2). Although the potentiating effect,,, of the nitrovasodilators or c A M P phosphodiestcrasc inhibitors on the action of isoprenaline were always significant with respect to changes in the 1Cs~~values for isoprenaline, and supra-additive effects of low concentrations of the compounds were sometimes seen, the synergism was not as strong as that observed when the effects of nitrovasodilators and prostaglandins on platelet aggregation were studied (Lcvin ct al., 1982: De Caterina et al.. 1988: Willis ct al., 1989: Mauricc and ttaslam. 1990). To address this possible discrepancy, we carried out experiments on the inhibition of aortic smooth muscle contraction, which is probably more closely comparable to inhibition of platclct aggregation than smooth muscle relaxation. 3.2. Effects on the contraction of aortic smooth muscle
Incubation of aortic smooth muscle with isoprcnaline for 30 s caused a concentration-dependent inhibition of contraction induced by phcnylephrine (e.g. fig. 3). This
1 O0
0
,I 0 ¢.)
20
,'; N I '
,'-:,IN - 1
Ciloslarnide
I,~o 2 0 - 1 7 2 , 1
Fig. 2. Effects of nitrovasodilators and of cAMP phosphodiesterase inhibitors on the relaxation by 96 nM isoprenalinc of rat aortic rings precontracted by I(X) nM phenylephrine. "['he compounds studied were SN P (0.25 n M), SIN- 1 (30 nM), cilostamide (20 nM) and Ro 20- ] 724 ( I 0 ~M). Results with 96 nM isoprenaline are from cumulative concentration-response curves similar to those shown in fig. 1. The first bar in each group shows the percent relaxation with the indicated compound alone ([]), the second, the relaxation of the same aortic rings with isoprenaline alone (11~)and the third, the relaxation induced by the combined action of both compounds (I). Results are means + S.E.M. from 3 experiments. each with 3 or 4 aortic rings. The supra-additive effect of cilostamide and isoprenaline is indicated: * P < 0.05.
239
action of isoprenaline required slightly higher concentrations of the compound than relaxation of precontracted aortic smooth muscle. An lC.so value for isoprenaline of 349 + 26 nM was obtained (mean + S.E.M., 15 determinations each based on 3 or 4 aortic rings). Under the conditions of our experiments, 10- to 50-fold higher concentrations of nitrovasodilators (and cAMP phosphodiesterase inhibitors) were required to induce a weak ( < 30%) inhibition of contraction than to induce a similar extent of relaxation. Simultaneous addition of such concentrations of SNP or SIN-1 with isoprenaline caused marked leftward shifts in the isoprenaline concentration-response curves relative to the theoretical additive curves (e.g. fig. 3A). This resulted in 9- to 10-fold reductions in the IC~() values for isoprenaline (table 2), about 3-fold the effect observed on relaxation (table 1). Since the maximum inhibition of contraction obtained with isoprenaline was close to 100c/c, no upward shift in the concentration-response curves was possible. However, clear supra-additive effects of either SNP or SIN-I were observed in the presence of 100 nM isoprenaline (fig. 4). Both of the c A M P phosphodiesterase inhibitors studied (cilostamide and Ro 20-1724) interacted synergistically with isoprenaline to inhibit phenylephrine-induced contractions. A concentration of cilostamide (1 gM), which alone had a similar effect to 5 nM SNP or 1 ~M SIN-l, shifted the isoprenaline concentration-response curve to essentially the same extent as did these nitrovasodilators (table 2). Ro 20-1724 (100 gM) was only slightly less effective than cilostamide (table 2). Supra-additive effects on contraction were also seen with 100 nM isoprenaline and the above concentrations c)f these inhibitors of
I 10 0
-
\\
v
i
i
A
60
~<
40
2O a.,
! 00 -
It
_
"z
-2 •="=
60
'1;,
,
_
4(1
2O
~
\
0 ~====~
0
10
lsoprerlaline
I
I
100
1000
(riM)
F i g 3. F,ffects of SNP and of cilostanlide on the ability of isoprenMine to inhibit phcny[ephrine-induced contractions of rat aortic rings. The indicated concentrations of isoprenaline were added without (O) or ',,,ith (In) 5 nM SNP IA) or 1 p.M cilostamide (B). After 30 s, rings were contracted with I(X) nM phenylephrine. Contractions were measured after a further 2 rain, and were expressed as percentages of controls with phenylephrine alone. Values are means + S.E.M. from 4 rings. The theoretical additive concentration-response curves are also shown f . . . . . . ) (see Materials and methods1.
I O0
~-
80
60
411
SNI'
SIN-I
Cil,)st amide
Ro 2 0 - 1724
Fig. 4. Effects of nitrovas(xtilators and of cAMP phosphodiesterase inhibitors on the abilily of 100 nM isoprenaline to inhibit the contraction of rat aortic rings by 100 nM phenylephrine. The compounds studied were SNP (5 nM), SIN-I (1 #M), cilostamJde (1 p.M) and Ro 20-1724 (100 p.M). Rings were incubated for 30 s with isoprenaline wilhout or with one of the above compounds. Phenylephrme was then added and the resulting contraction measured after a further 2 min. The first bar in each group shows the inhibition of contraction b,, the indicated compound alone ([]), the second, the inhibition of contraction by isoprenaline alone (~) and the third, the combined action of both compounds (11). Values are means 4: S.E.M. from 3 to 6 experiments, each with 3 or 4 aortic rings. Supra-additive effects of the drug combinations studied are indicated: * P < 0.05: * * P
< 0.01.
240 "FABLE 2 it'ffects of low concentrations of nitrovasodilators and of inhibitors of cAMP phosphtxzliesterase on the ability of isoprenaline to inhibit the contraction of rat aortic rings by I00 nM phenylephrine, lsoprenaline was added without or with each of the listed compounds 30 s before phenylephrine and the contraction of the aortic rings was measured after a further 2 min. In each experiment. I('sq, values for isoprenaline in the absence and presence of one of the compounds listed were calculated from concentration-response curves obtained from the same aortic rings, as described under Materials and methods and shown in fig. 2. Results are means _+.SE.M. from the numbers of experiments shown; the significance of changes in [('5~ values for isoprenaline were evaluated by two-sided paired t-tests after logarithmic transformation of the results; " P < 0.02. h p < 0.001. The ratio of I('s~ values for isoprenaline in the absence or presence of other drugs indicates the extent of synergism. ('ompounds
No. of
IC5oof isoprenaline (nM) Ratio of
studied
expts.
Without
With
compound
compound
334__-49 398 k55 318L57 356_+57
37__+ 5 h 4 2 - 1 0 ;' 34± 4 " 44_+ 3 h
SNP(5 nM) SIN-1 (1 p.M) Cilostamide(1 p.M) Ro 20-1724 (100 #M)
5 3 3 4
1('5" values 90 9.5 9.4 g.l
c A M P phosphodiesterase (fig. 4). Our observation that the nitrovasodilators and cAMP phosphodiesterase inhibitors potentiated the inhibition of aortic smooth muscle contraction by isoprenaline to essentially the same extent, and that these compounds also had equal, though weaker, effects on the relaxant action of isoprenaline, suggests that these two different groups of compounds may ultimately act through a common inhibitory mechanism.
4. Discussion
In experiments with rat aortic rings from which the endothelium had been removed, we showed that the nitrovasodilators, SNP and SIN-l, potentiate the ability of isoprenaline to relax precontracted aortic smooth muscle and, when added with isoprenaline before the contractile stimulus, exert potent synergistic inhibitory effects on contraction. In contrast, a study of the interactions between SNP and iloprost in vascular smooth muscle failed to detect synergism (Lidbury et al., 1989). Although use of rabbit mesenteric and coeliac arteries in the latter study, rather than rat aorta, could account for the different results obtained, other considerations may be more important. In our experiments, detection of the synergistic effects of nitrovasodilators and isoprenaline on relaxation depended on comparison of results obtained in the same aortic rings incubated with and without test compounds. Moreover, potentiation of the effects of isoprenaline by either the nitrovasodilators or the cAMP phosphodiesterase inhibitors was always much more marked when inhibition of contraction,
rather than relaxation, was studied. The reasons for the latter difference are not clear, but both methodological and physiological factors could be important. First, prolonged exposure of precontracted smooth muscle to nitrovasodilators or cAMP phosphodiesterase inhibitors could have inhibitory, as well as stimulatory, effects on the ability of the tissue to respond to isoprenaline. Thus, pretreatment of other rat tissues with c A M P analogues has been shown to increase cAMP phosphodiesterase activity and to reduce the effects of agonists that stimulate adenylyl cyclase (Gettys et al., 1987). In addition, the platelet cGMP-inhibited cAMP phosphodiesterase appears to be activated by cAMP-dependent phosphorylation (MacPhee et al., 1988: Grant et al., 1988). The continued presence of phenylephrine could also have adversely affected the actions of the other compounds studied, for example through the activation of protein kinase C. Second, there may be critical differences in the state of the vascular smooth muscle before and after induction of contraction that affect the responses to the vasodilator drugs studied. Since increases in intracellular C a : " concentration and the phosphorylation of myosin light chain are important in the contraction of vascular smooth muscle ( K a m m and Stull, 1985: Murphy', 1989), compounds that block these processes will inhibit contraction. On the other hand, relaxation of smooth muscle, which involves reversal of a latch state, is less dependent on the inhibition of ('a e" mobilization or on dephosphorylation of myosin (l)illon et al.. 1981; Hal and Murphy, 1988). Increases in cGMP, presumably acting through cGMP-dependent protein kinase, reduce the level of myosin phosphorylation in smooth muscle (Rapoport et al.. 1983), activate the plasmalemmal Ca 2 '-ATPase (Rashatwar et al., 1987: Vrolix et al.. 1988, Cornwell and l,incoln, 1989) and inhibit the G-protein-mediated stimulation of phosphoinositide hydrolysis (Hirata et al., 1990). ()n the other hand, cAMP-dependent protein kinase activity' increases the concentration of ('a2'-calmodulin required for activation of myosin light-chain kinase (Adelstein et al., 1978) and the C a 2" sensitivity of smooth muscle C a 2 ' - d e p e n d e n t potassium channels (Sadoshima et al., 1988). Thus, in view of the different biochemical mechanisms that may be involved in the inhibition of contraction and relaxation, it is entirely conceivable that activation of one or the other cyclic nucleotide system, or both together, could have distinct effects. The marked synergism between the actions of the nitrovasodilators and isoprenaline reported here for the inhibition of rat smooth muscle contraction is analogous to that observed in platelets between nitrovasodilatots and activators of platelet adenylyl cyclase (Levin et al., 1982: De Caterina et al., 1988: Willis et al., 1989: Maurice and Haslam. 1990). In platelets, we recently'
241
showed that this synergism resulted from the inhibition by cGMP of cAMP breakdown by a cGMP-inhibited low K mcAMP phosphodiesterase (Maurice and Haslam, 1990). An enzyme with kinetic characteristics and a sensitivity to inhibitors very similar to those of the platelet enzyme has been shown to be present in rat, guinea pig, and rabbit aorta (Schoeffter et al., 1987: Silver et al., 1988; Souness et al., 1989; Ahn et al., 1989). The possibility therefore exists that the synergism between nitrovasodilators and isoprenaline reported here results from inhibition by c G M P of this same enzyme in rat aortic smooth muscle. Much of our data is consistent with this. First, both SNP and SIN-I had potentiating effects on the action of isoprenaline that were virtually indistinguishable from those obtained using cilostamide, whether relaxation or inhibition of contraction was studied. The latter compound has been shown to be a selective inhibitor of the cGMP-inhibited low K m cAMP phosphodiesterase in platelets (Hidaka et al., 1979; MacPhee et al., 1986) and to increase cAMP, but not cGMP levels in rat aortic smooth muscle at concentrations that caused 50% relaxation (Schoeffter et al., 1987). The possibility that the nitrovasodilators do potentiate the effects of isoprenaline by inhibiting the cGMP-inhibited enzyme is also consistent with evidence that cilostamide appears to decrease relaxation by SNP in rat aortic smooth muscle (Schoeffter et al., 1987). Ro 20-1724, a compound known to inhibit a different low K m cAMP phosphodiesterase that is unaffected by c G M P (Beavo, 1988), was slightly less effective in potentiating the actions of isoprenaline. This observation suggests that cAMP breakdown in vascular smooth muscle depends on the activities of both low K m cAMP phosphodiesterases. E D R F has been identified as nitric oxide (Palmer et al., 1987) or more probably an S-nitrosothiol (Myers et al., 1990) and appears to relax vascular smooth muscle by the same cGMP-dependent mechanism as do the nitrovasodilators (Murad, 1986). A synergism similar to that seen with nitrovasodilators should therefore be observed between the effects of E D R F and activators of adenylyl cyclase on vascular smooth muscle. It is therefore significant that physiological stimuli can cause a coupled release of both E D R F and PGI2 from the vascular endothelium (De Nucci et al., 1988) and that an interaction has been observed between E D R F in an endothelial superfusate and PGI 2 in causing the relaxation of pig coronary artery (Shimokawa et al., 1988). These results are consistent with the known synergistic effects of E D R F and PGI 2, as inhibitors of platelet aggregation (Radomski et al., 1987; Macdonald et al., 1988). The present results extend evidence of significant synergistic interactions between nitrovasodilators and activators of adenylyl cyclase in platelets to vascular smooth muscle and provide indirect evidence that the
same mechanism, namely inhibition by cGMP of a low K m cAMP phosphodiesterase, may be responsible. Although no synergism was observed between the effects of nitroglycerin and a PGI 2 analogue on the mean blood pressure and heart rate of spontaneously hypertensive rats (,Willis et al., 1989), the possibility of effects specific to particular vascular beds has not yet been investigated. Since a cGMP-inhibited cAMP phosphodiesterase similar or identical to the platelet enzyme has also been shown to be present in both cardiac muscle (Harrison et al., 1986) and tracheal smooth muscle (Torphy and Cieslinski, 1990), synergistic interactions between the effects of activators of guanylyl and adenylyl cyclases are also likely to be found in these tissues. In general, synergism may be expected whenever the cGMP-inhibited phosphodiesterase accounts for a large fraction of cAMP breakdown. However, the physiological significance and pharmacological value of synergism between activators of guanylyl and adenylyl cyclases in platelets, vascular smooth muscle and other tissues remain to be determined. In principle, selective synergistic effects of therapeutic interest might be obtained through the use of combinations of pharmacological agents, each of limited specificity, that together activate both adenylyl and guanylyl cyclases only in the target tissue.
Acknowledgement This work was supported by a Grant-in-aid (1".1265) from the Heart and Stroke Foundation of Ontario.
References Adelstein, R.S., M.A. Conti, D.R. Hathaway and C.B. Klee, 1978, Phosphorylation of smooth muscle myosin light chain kinase by the catalytic subunit of adenosine 3':5'-monophosphate-dependent protein kinase, J. Biol. Chem. 253, 8347. Ahn, H.S., W. Crim, M. Romano, E. Sybertz and B. Pitts, 1989, Effects of selective inhibitors on cyclic nucleotide phosphtnliesterases of rabbit aorta, Biochem Pharmacol. 38, 3331. Azuma, H., M. Ishikawa and S. Sekizaki, 1986, Endothelium-dependent inhibition of platelet aggregation, Br. J. Pharmacol. 88, 411. Beavo, J.A., 1988, Multiple isozymes of cyclic nucleotide phosphodiesterase, Adv. Sec. Mess. Phosphoprot. Res. 22, 1. Cornwell, T.L. and T.M. Lincoln, 1989, Regulation of intracellular Ca 2+ levels in cultured vascular smooth muscle cells. Reduction of Ca 2~ by atriopeptin and 8-bromo-cyclic GMP is mediated by cyclic GMP-dependent protein kinase. J. Biol. Chem. 264. 1146. De Caterina, R., D. Giannessi, W. Bernini and A. Mazzone, 1988, Organic nitrates: direct antiplatelet effects and synergism with prostacyclin. Antiplatelet effects of organic nitrates, Thromb. Iiaemost. 59, 207. De Nucci, G., R.J. Gryglewski, T.D. Warner and J.R. Vane, 1988, Receptor-mediated release of endothelium-derived relaxing factor and protacyclin from bovine aortic endothelial cells is coupled, Prec. Natl. Acad. Sci. U.S.A. 85, 2334. Dillon, P.F., M.O. Aksoy, S.P. Driska and R.A. Murphy, 1981,
242
Myosin phosphorylation and the cross-hridge cycle in arterial smooth muscle, Science 211, 495. Furchgott. R.F. and J.V. Zav, adTki, 1980, ] h e obligatory role of endothelial cells in the relaxation of arterial smooth muscle h~, acetylcholine, Nature 289, 373. (,}ettys, T.W., P.F. Blackmore, J.B. Redmon, S.J. Beebe and .1.I). Corbin, 1987, Short-term feedback regulation of c A M P bv accelerated degradation in rat tissues, J. Biol. ('hem. 262, 333. (Jlusa. t!.. F. Markwardt and J. Sti]rzehecher, 1974, I'ffects of sodium nitroprusside and other pentacyanonitrosyl complexes on platelet aggregation. Haemostasis 3, 249. Orant, P.G. anti R.W. Colman, 1984, Purification and characteri,'alion of a human platelet cyclic nucleotide phosphc,diesterase. Bi,x:hemistry 23, lg01. Grant, P.G., A.F. Mannarino and R.W. ( o l m a n , 1988, cAMP-mediated phosphorylation of the low K,,, cAMP phosphodiesterase markedly stimulates its catalvtic activity, Proc. Natl. A c a d Sci. U.S.A. 85, 9071. ttai, ('.-M. and R.A. Murphy. 1989, Cross-bridge phosphorylation and regulathm of latch state m sm(x~th muscle, Am..J. Phvsiol 254, 4"99. l|ardman, J.C}., 1994, Cyclic nucleotides and regt, lation of vascular smooth muscle. J. Cardiovasc. Pharmacol. 6. $639. tlarrison, S.A.. D.H. Reifsnyder, B. Gallis. G.(;. Cadd and J.A. Bcavo, 1986, Isolation and characteri:,'ation of bovine cardiac muscle cOMP-inhibited phosphodiesterase: A receptor for new cardiotonic drugs, Mol. Pharmacol. 29. 506. Ilashimoto. K., N. Taira, M. Hirata and M. Kokuhun, 1971, The mode of hypotensive action of newly synthesized svdnonimine derivati',es, Arzneim. Forsch. 21. 1329. tlashnn, R.J., M.M.I,. Davidson, '1. l)avies, J.A. l,ynllam and M.I). McClenaghan, 1979, Regulation of b],,)od platelet function b', cvclic nucleotides. Ad',. ( ' y c l Nucl. Res. 9, 533. tlidaka, I1., H. ttayashi, It. Kohri, Y. Kimura, T. Hosokawa, "r. lgav.a and Y. Saitoh, 1979, Selective inhibitor of platelet cyclic adenosine monnphosphate phosphodiesterase, cilostamide, inhibits platelet aggregation, J. Pharmacol. Exp. "fher. 211.26. ttirata, M., K.P. Kohse, C.-tl. ('hang, T. lkebe and t r. Murad, 1990, Mechanism of cyclic (.;MP inhibition of inositol phosphate formalion in rat aorta segments and cultured bovine aortic smooth muscle cells. J. Biol. Chem. 265, 1268. K a m m , K.f!. and J.'l. Stull, 1985. -['he function of myosin and myosin light chain kinase phosphorylation in smoc, th muscle. Ann. Re,,'. Pharmacol. Toxicol. 25, 593. Levin, R.I., B.B. Weksler and E.A. Jaffe, 1982. The interaction of sodium nitroprusside with h u m a n endothelial cells and platelets: Nitroprusside and prostacyclin synergistically inhibit platelet function, Circulation 66, 1299. Lidhury. P.S., E. Antunes, G. De Nucci and J . R Vane, 1999. Interactions of iloprost and sodium nitroprusside on vascular smooth muscle and platelet aggregation. Br. J. Pharmacol. 99, 1275. Macdonald, P.S., M.A. Read and G.J. Dusting, 1998, Synergistic inhibition of platelet aggregation by endothelium-derived relaxing factor and prostacyclin, Thromb. Res. 49. 437. Mad"hee, C.tt., S.A. llarrison and J.A. Beavo, 1986, Immunological identification of the major platelet low K m cAMP phosphodiesterase: Probable target for anti-thrombotic agents, Proc. Natl. Acad. ,~i. U.S.A. 83, 6660. MacPhee, ('.H.. D.H. Reifsnyder, T.A. Moore, K.M. I,erea and J.A. Beavo, 19g8, Phosphory'lation results in activation of a c A M P phosphodiesterase in h u m a n platelets, J. Biol. Chem. 263, 10353. Maurice, I).1t. and R.J. Haslam, 1990, Molecular basis of the synergistic inhibition of platelet function by nitrovasodilators and activators of adenylate cyclase: Inhibition of cyclic A M P breakdown by cyclic GMP, Mol. Pharmacol. 37, 671. Murad, F., 1986. Cyclic guanosine monophnsphate as a mediator of vas¢~:lilation, J. Clin. Invest. 78, 1.
Murphy. R.A.. 1999, (,'ontraclion hi :,.in,,uath muscle cells. Ann. Re',. Physiol. 51. 275. M'.'er~,, P.R., R.I.. Minor, R. (;uerra, J.N. Bales and 11.(;. Harrison, 1990, Vasorelaxant properties of the endothelium-derived relaxing factor more closely resemble S-nilrosocvsteine than nitric oxide, Nature 345, 161. Ni~,hikawa. M.. M. Kanamon and H. 11idaka. 1982, Inhibition of platelet aggregation and stimulation of guanylate c',clasc b~ an antianginal agent molsid~mline alld its metabolites, J. Pharmacol. Exp. Thor. 220. 183. Palmer, R.M.J., A.(;. I"errige and S. Moncada, 1987. Nitric oxide release accounts for the biological activity of endothelium-dcrived relaxing fi,ctor. Nature 327. 524. P,Sch. G. and S. ttolzmann. 1980, Quantitative estimation of overadditive and underadditive drug effects bv means of theoretical. additi,.'e dose-response curves, J. Pharmacol. Moth 4. 179 Radomski, M.W.. R.M.J. Palmer anti S. Moncada. 1997. "['he anti-aggregating properties of vascular endothelium: interactions bet,*een prostac?,clin anti nitric oxide, Br. J. Pharmacol. 92. 6.t9. Rapoport, R.M., M.B. l)raznin and F. Murad, 1983. I'-indothelium-dependent relaxation in rat ,:lorta may he mediated through cyclic GMP-dependent protein phosphor',lation, Nature 306. 174. Rashat,*,ar, S.S.. "I'.L. ('ornwell and I.M. l,mcoln, 1997, Effects of 8-bromo-cGMP on ('a z" levels in ',ascular smooth muscle cells: Possible regulation of Ca" '-ATPase b', cGMP-dependent protein kinase, Proc. Nat[. Acad. Sci. I, LS.A. 84, 5685. Sadoshinla, J.-I., N. Akaike, I-I. Kana~de and M. Nakamura, 198., ('relic A M P modulates Ca-activated K channel in cultured sm~u~th muscle cells of rat aortas, Am. J. Phvsiol. 255. H754. SchrGr, K.. It. l)arius, R. Matzkv and R. ()hlendorf. 1981, "[he antiplatelet anti cardiovascular actions of a ne',~, carbacvclin tterivative (ZK 363741 equipotent to I)GI, in vitro, Nat.n',nSchmiedeb. Arch. Pharmacol. 316, 252. Schoeffter, P., ('. l,ugnier, F. Demcs~,.-Waeldele and J.('. Stoclet, 1987, Role of c~,clic AMP- and cyclic (;MP-phosphodiesterases in the control of cvclic nucleotide levels and smooth muscle tone in rat isolated aorta. A studv with ~,elective inhibitors. Biochcm Pharmacol. 36. 3965. Shimokawa. H., N.A. Flavahan, R.R. Lorenz and P.M. Vanhoutte. 19gg, Prostac,.clm releases endothelium-derived relaxing factor and potent)ates its action m coronary, arteries of the pig, Br. ,I. Pharmacol. 95, 1197+ Silver, P.J., R.IL kepore. B. ()'Connor, B.M+ l+emp, I..T. Ilamel. R.G. Bentley, and A.L. H a m s , 1989, Inhibition of the Io'a, Kn, cyclic A M P phosph~uJiesterase and activation of the c','clic A M P system in vascular smooth muscle by milrinone, J. Pharmacol. I'×p. Thor. 247, 34. Souness, J.l'.. R. Brazdil, B.K. l)ic,cee and R. Jordan. 1989, Role of selecti~,e cyclic ( i M P phosphodiesterase inhibition in the mvorelaxant actions of M&B 22,948, MY-5445, vinpc~etine and 1-methyl-3-isobut~,l-S-(methylamim~)xanthme, Br J. Pharmacol. 98, 725. Tateson, J.I-., S. Moncada and J.R. Vane, 1977. Effects of prostac~.clin (P(;X) on cyclic A M P concentrations in human platelets. Prostaglandin.s 13, 389. Torphy, T.J. and L.B. ('ieslinski. 1990, ('haracterl;'ation and selective inhibition of cyclic nudeotide phosphodiesterase isozvmes m canine tracheal snlootll muscle, Mol. Pharmacol. 37, 2(16. Vrolix, M., I,. Rae,,maekers, F. Wuvtack, F. t t o f m a n n and R. ('asteel,. 1988. Cyclic (}MP-dependent protein kinase stimulates the plasmalemmal Ca 2 " pump of sm(x',th muscle via phosphorylatzon of phosphatidylinositol, t~.it~.'hem. J. 255, 855. Willis, A t . , I).L. Smith, M. I,oveday, J. Fulks, C.lt. Lee, L. ttedlev and I). VanAntwerp, 1999. Selecti',e anti-platelet aggregation synergism bet'~,een a prostacyclin-mimetic, RS93427 and the nitrodilators s