Acta Physiol Scand 1992, 146, 21-30

Effects of Amrinone on shortening velocity, force development and ATPase activity of demembranated preparations of rat ventricular myocardium S. E. J. N. M O R N E R " , M. CANEPARI, R. B O T T I N E L L I , V. CAPPELLI and C. R E G G I A N I Institute of H u m a n Physiology, University of Pavia, Pavia, Italy and "Department of Pharmacology, University of Lund, Lund, Sweden M., BOTTINELLI, R., CAPPELLI,V. & REGGIANI, C. MORNER,S.E. J. N., CANEPARI, 1992. Effects of Amrinone on shortening velocity, force development and ATPase activity of demembranated preparations of rat ventricular myocardium. Acta Physiol Scand 146, 21-30. Received 20 December 1991, accepted 19 March 1992. ISSN 0001-6772. Institute of Human Physiology, University of Pavia, Italy, and Department of Pharmacology, University of Lund, Sweden. This study analyses the effects of Amrinone (bipyridine derivative with phosphodiesterase inhibitor properties) on the myofibrillar apparatus of rat myocardium. Thin trabeculae were isolated from the right ventricle and chemically demembranated. Force development and shortening velocity were measured during maximal calcium activations (pCa = 4.45) in control conditions and in the presence of 1-3 mM Amrinone. Maximum shortening velocity was obtained both from extrapolation of the force-velocity curve and with the slack test method. Amrinone was found to significantly reduce maximum shortening velocity and force development. Myofibrils and myosin were prepared from rat ventricular myocardium and their ATPase activity was assessed in control conditions and in the presence of Amrinone (0.3-6 mM). Ca-Mg dependent myofibrillar ATPase activity which was determined at low ionic strength was depressed by Amrinone in a dose-dependent way. Ca-stimulated ATPase activity determined at high ionic strength in myofibril or myosin preparations was not affected. Furthermore, Amrinone did not influence the pCa-ATPase activity curve of the myofibrillar preparations. A comparison between the inhibitory effects of Amrinone on myofibrils prepared from euthyroid rats and myofibrils prepared from hypothyroid rats was carried out. The ATPase activity was significantly less depressed in myofibrils prepared from hypothyroid rats than in those prepared from euthyroid rats. These results provide the first evidence of an effect of Amrinone on ATP splitting and force generation in the myofilament system of cardiac muscle. Key words : Amrinone, cardiac muscle, force-velocity relation, myofibrillar ATPase, myosin, phosphodiesterase inhibitors.

Amrinone is a bipyridine derivative with phosphodiesterase inhibitor properties (Honerjager et al. 1981, Endoh et al. 1982). In isolated cardiac preparations Amrinone has been shown to increase myocardial contractility (Onuaguluchi Correspondence : C. Reggiani, Institute of Human Physiology, Via Forlanini 6, 1-27100 Pavia, Italy.

& T a n z 1981, Gaide et al. 1981, Kondo et al.

1983) and to reduce the duration of the action potential (Malecot et al. 1985, Morner & Wohlfart 1990). Furthermore, it increases the aequorin signal (Morgan et al. 1980), thus indicating an increased concentration of intracellular free calcium. Convincing evidence that the inotropic effect depends on phospho-

21

22

S . E . J . N . Morner et al.

diesterase inhibition and CAMP increase has velocity was associated with an increased tetanic been presented (Honerjager et a/. 1981, Endoh tension (Minsson & Edman 1985, Minsson et al. et al. 1982, Weishaar et al. 1987). I n vascular 1989). smooth muscle -4mrinone has been found to reduce calcium available for contraction and to determine vasodilation (Morgan et al. 1986). In M E T H 0 D S frog and mammalian skeletal muscle Amrinone Animals. .All experiments were carried out on male exhibits a caffeine-like potentiating effect on Wistar rats 2-3 months old. The animals were born twitch response (Minsson et al. 1989, Morner & and maintained at the facilities of the Institute of Mansson 1990). Moreover, in tetanic contrac- Human Physiology, University of Pavia. Hypothytions Amrinone decreases maximum velocity of roidism was induced by an iodine free diet and by shortening and increases isometric tension adding KClO, to the drinking water for 8 weeks (Poggesi et a/. 1987). The animals were anesthetized (MPnsson & Edman 1985, Minsson et al. 1989). with ethylic ether, the chest was opened and the heart T h e latter effects cannot be attributed to the quickly removed. action on excitation-contraction coupling. -4s discussed by Minsson & Edman (1985), there are reasons to believe that Amrinone interferes .Lfechanical studies with the mechanism of force generation at Preparation and mounting. The heart was excised myofibrillar level. from the chest and quickly perfused with ice cold T h e aim of the present study was to investigate skinning solution via an aortic cannula and then whether a specific action of Amrinone at immersed in a dissection chamber containing the same myofibrillar level also exists in cardiac muscle. solution at 12-15 "C. The free wall of the right I n order to avoid any interference with well ventricle was opened and the trabeculae placed along known effects on excitation-contraction coup- the insertion of the free wall on the septum and close ling, skinned or demembranated preparations to the pulmonary cone were carefully dissected. After 2-3 h the solution of the dissection chamber was were employed. O n these preparations the effects replaced by a second skinning solution containing lo/& of Amrinone on the force-velocitv curve u-ere Triton X-100. The preparations w-ere kept in this determined during maximal calcium activations. solution for 1 h. Aluminium clips were placed at both Furthermore, variations of ATPase activity of ends of the trabeculae which were then transferred to purified myofibrils and isolated myosin in the the experimental setup. presence of Amrinone were studied. ConcentraSolutions. Skinning, relaxing, activating and pretions of the compound between 0.3 and 6 m ~activating solutions were similar to those used for were employed. I n this range of high con- skeletal muscle fibres (Bottinelli et al. 1991) with some centration, Amrinone has been found to display modifications (see Table 1). Two skinning solutions its inotropic effect on isolated cardiac prepar- were utilized. Their composition was identical except ations. Although threshold effects have been that the second contained Triton X-100 (10 pl ml-') to ensure complete skinning. Both pre-activating and found at 3 @ 5 0 p ~ ,the 50°0 effective conactivating solutions were prepared without amrinone centration (EC 5000) has been found between or with appropriate concentrations (1-3 mM) of the 0.5 and 1.5 mM depending on the animal species drug. The pCa of the activating solution was estimated (Siegl et al. 1984, Gibbs et al. 1985, Morner &- to be 4.45 by means of a computer program designed Wohlfart 1990). D u e to the lower potency of by Fabiato (1988). A rephosphorylating system based hmrinone, these concentrations are one or two on creatine phosphokinase (Boehringer, Mannheim, orders of magnitude higher than those at which Germany) and creatine phosphate was present in both other phosphodiesterase inhibitors are eff'ective. pre-activating and activating solutions. Amrinone was T h e results obtained support the conclusion dissolved in 0.5 M HCI to a concentration of 100 mM that at the above mentioned concentrations (18.8 mg ml-I). The stock solution was then diluted in Amrinone had a direct action on the force relaxing solution to the desired concentrations and the pH was adjusted by buffering with KOH. generating mechanism also in rat cardiac muscle. Experimental setup. The experimental setup was In demembranated cardiac preparations, how- composed by three chambers milled in an aluminium ever, not only maximum shortening velocity but plate which was placed on the stage of an inverted also developed tension and myofibrillar ATPase microscope (Axiovert 10, Zeiss, Oberkochcn, Geractivity were depressed, whereas in skeletal man>-). The three chambers could be lowered and muscle the reduction of maximum shortening translated and raised to immerse the specimen quickly

Amrinone efects on cardiac muscle

23

Table 1. Composition of the solutions (mM unless otherwise indicated)

Skinning Relaxing Preactivating Activating

K propionate

KH,PO,

Mg acetate

Na,ATP

EGTA

CaCl,

150 KCI 100 100 100

5 imidazole 20

5 MgCI, 5 5

3 Na,ATP

5 EGTA

-

5 5

5

-

5

-

5

5

5

5

20 20

CaCl,

~~~

~

DTT (dithiothreitol, 1 mM) was present in all the solutions. Two skinning solutions were used: their composition was identical, except for the presence of Triton X-100 (1 yo V/V) in the second one. CP (creatine phosphate, 30 mM) and CPK (creatine phosphokinase, 300 U ml-l) were added to the preactivating and activating solutions. The pH of all the solutions was set to 7. Activating solution exhibited a pCa of 4.45, a pMg of 3.14 and a pMgATP of 2.41. Table 2. Cross-sectional area (c.s.a.), length and sarcomere length of the demembranated ventricular preparations employed to test the effects of Amrinone. Data is expressed as means and SE Mean c.s.a. (pm2) length (mm) sarcomere length (pm)

32590 1.37 2.11

Standard error

7046 0.09 0.03

n

14 14 14

in any of them. The first chamber (volume, 3 ml) contained relaxing solution, the second and the third chamber (volume, 1 ml) pre-activating and activating solution respectively with or without Amrinone. Temperature was maintained constant at 12 "C throughout each experiment by circulating a water glycol solution cooled by means of a thermostat (Endocal, Neslab Newington, NH, USA) through channels engraved into the aluminium plate. The trabeculate were mounted horizontally between an electromagnetic puller (Ling 101, Ling Dynamic Systems, Royston, UK) and a force transducer (AE 801 Aksjeselskapet, Mikroelektronikk, Horten, Norway). The puller, which was provided with an inductance position transducer, was driven by a feed-back circuit whose input was either the output of the position transducer (length control mode) or the output of the force transducer (load control mode). In the former, the length of the trabecula was kept constant in order to produce isometric contraction. In the latter, the load was controlled so that isotonic shortening against prefixed load could be obtained. During the experiment the trabeculae could be inspected through a stereomicroscope (Wild M3B, Wild, Heenbrugg, Switzerland) mounted above the setup and through the inverted microscope. The latter was utilized to

measure the thickness and the width of the trabeculae and to determine sarcomere length by counting the number of striations over a 30 p m interval. Experimental procedure. The trabecula was mounted in the chamber containing relaxing solution and stretched to a sarcomere length of about 2.1 pm. It was then transferred to the chamber containing preactivating solution and finally activated by rapid translation to the chamber containing activating solution. The tension rose to a steady level which was generally well maintained during the activation which lasted 2-3 min. During this time, 15-20 load clamps were performed to obtain a forcevelocity curve. In each load clamp manoeuvre, the trabecula was released against a prefixed load. After a phase of isotonic shortening, the trabecula was restretched to its initial length. The trabecula was then relaxed by returning it to the chamber containing relaxing solution. Once relaxed, the preparation was examined through the inverted microscope and the preservation of the striation pattern was taken as an index of its acceptable quality. The complete activation procedure was repeated after 10-15 min. Amrinone was present in the pre-activating and in the activating solutions either during the first or the second activation. A group of six trabeculae was employed to determine the unloaded shortening velocity according to a method first designed by Edman (1979) for skeletal muscle fibres and indicated as slack test procedure. These trabeculae were activated first under control conditions, then in the presence of AMR, and then again under control conditions. During each activation, 3-5 releases of amplitude ranging from 8 to 15% of initial length were applied. All these releases were sufficient to introduce slack in the trabecula. Each release was followed by a slow elongation to the initial length. Average values of length, cross sectional area and sarcomere length of the trabeculae employed in this study are reported in Table 2.

24

S. E.3. hi. Morner et a1

preparations was determined from the rate of release of inorganic phosphate in an assay medium at low ionic strength which had the following composition (mbi): KCI 50, MgCI, 2, NaATP 2, imidazole 20, CaCI, 0-1.5, EGTA (L-1.6, p H 7, temperature 27 "C. T h e details of the assay procedure have been previous1)- described (Cappelli et a / . 1988, 1989). T h e concentrations of CaCI, and EGTA were varied in order to obtain different values of pCa between 8 and 4, T h e pCa value was calculated with a computer program designed by Fabiato (1988). Amrinone was added to the assay medium in concentrations between 0.3 and 6 mi4 in order to test its effect on enzymatic activity. In some cases the Ca-activated ATPase activity of the myofibrils was also determined. In these experiments the assay medium was the high ionic strength solution used to determine purified myosin ATPase (P/P,,+u/P,,) ( V + b ) = (P,/P,+a/P,)b = acti1-ity and described below. a/f'n (V,,,+b) (1) ,II.yosin preparation. Purified myosin was prepared ..icomputer program derived from that described from ventricular myocardium with a method derived b! Edman et a/. (1976) was used to determine the from that proposed by Barany (1967) and described numerical values of the parameters Ynlar, Pi/P,, by Cappelli et al. (1988). Determination of myosin ATPase activity. Ca actia/P,. C;,,,, was the intercept of the curve on the ordinate and represented maximum shortening vel- vated myosin ATPase activity was determined at high ocity. P i / P o was the intercept on the force axis. a / P , ionic strength in a medium of the following composition (mhl): KCI 300, Tris-C1 50, CaCI, 10, and b represented the asymptotes of the hyperbola. Calculation of unloaded shortening velocity ( V,,) NaATP 2, p H 7, temperature 27 "C. Appropriate and extension of series elasticit!. was performed on concentrations of Amrinone were added to the test data collected with the slack test procedure, according tubes in order to determine the effect of the compound to Edman (1979). A linear regression analysis of the on myosin ATPase activity. T h e assay procedure has release amplitude versus the time necessary to take up been described in a previous paper (Cappelli et ill. the slack provided a value of unloaded shortening 1989). Statistical ana!ysis. Data was expressed as means velocit! corresponding to the slope and a value of extension of series elasticity corresponding to the and standard errors. Student's t-test was employed to intercept on the ordinate. Since the amplitude of the assess the statistical significance of the variations of mot-ement &-as kept below l j o 0 initial length, no mechanical and biochemical parameters caused by variation of the slope with the amplitude could be Amrinone. A4value of P < 0.05 was considered to be statistically significant. observed (Moss 1986, Morano et ai. 1988). r. I he values of isometric tension measured before load clamp manoeuvres (Po), the values of the R E S U L T S parameters P;,,,, and a / P , calculated bl- the computer program and the value of V,, obtained with the slack Effect of Amrrnone on the contractile properties test method in each trabecula in a given condition of skinned ventricular trabeculae were treated as single observations for statistical purposes. T h e effects of Amrinone on the f o r c e v e l o c i t y curve were examined in eight chemically skinned ventricular trabeculae. In each trabecula, the Biozheniical studies forcevelocity curve was determined during .,tf.yojibril preparation. Freshll- excised hearts were maximal calcium activation u n d e r control conquickly immersed in ice cold saline (0.900) solution ditions a n d i n the presence of 1 mM Amrinone. and the atria and blood vessels were dissected out. The entricular myocardium was homogenized and In four trabeculae, t h e f o r c e v e l o c i t y curve was washed and purified ventricular myofibrils were determined first u n d e r control conditions a n d prepared according to the method previously described then in the presence of the compound. In the other four trabeculae, the opposite sequence was (Cappelli et a f . 1988). Determination of m.yofibrillar '4 TPase activity. T h e followed. As can be seen in F i g u r e 1 shortening Ca-hlg dependent ATPase activity of the m!-ofibrillar velocity was reduced in the presence of Amri-

llatu rrrordrng and analysrs. Signals from tension and position transducers were displayed on a storage oscilloscope (Tektronix 5. 113, Tektronis, Beaverton, OR, USi\) and fed to a chart recorder (Graphtek WR3701, Graphtek, Japan) and to a digital oscilloscope (Nicolet 1094, Nicolet, Madison, M'I, USA). Xleasurenients were performed on chart records or on the Nicolet oscilloscope. In each load clamp manoeuvre, load ( P ) ivas measured and expressed as a fraction of the isometric tension (Po)just prior the manoeuvre. T h e shortening velocity ( V ] rvas measured by linear interpolation in the interval between 30 and 40 ms after the beginning of load clamping and expressed in trabecula length per second ( I s-'). .ill force-velocity pairs obtained in each trabecula in a given condition (control or AMR) were fitted to the Hill's hl-perbolic equation:

Amrinone efects on cardiac muscle

'I

25

I

Relative

load

P/P,

Fig. 1. Force-velocity curve of a skinned ventricular trabecula in control conditions (B) and in Each point represents a pair of force-velocity data. The lines the presence of 1 mM Amrinone (A). represent the hyperbolas obtained by fitting the Hill's equation to the force-velocity data. The fitting yielded the following parameters. Control, V,,, = 0.92 1 s-'; P,/P, = 0.86; a/P, = 0.23. Amrinone, V,,, = 0.69 1 s-'; Po/Po= 0.83; a/P, = 0.33.

Table 3. Effects of AMR on the contractile properties of demembranated rat ventricular trabeculae. Data is expressed as means and SE

Force-velocity relation Maximum isometric tension (Po) (mN mm-') Maximum speed of shortening (VrntX) (1 s-7 Curvature index (alp,) Intercept on the force axis (PO"/PO) Slack test manoeuvre Unloaded shortening velocity

Values in control conditions

% variation due to AMR (1 mM)

58.67k 13.01

-7.48

1.115 k0.061

-8.93f3.51""

8

0.21 1 f 0.022 0.922 k 0.035

+ 14.2f

8 8

(% 0

Tension redevelopment rate? (mN mm-' s-')

* P < 0.05; *"

t

13.1

8

yo variation 1.838 k0.159

due to AMR (3 mM) - 17.6f 5.8"

6

6.809 .t 0.879

+ 5.32k8.61

6

7.040 k0.460

- 26.5 4.3""

6

(1 s-1)

Series elasticity extension

+ 2.67""

n

P < 0.01. Determined after a shortening of 10% 1.

26

5'. E . 3 . N . Morner et al.

1 "0

25

50

75

time ms

Fig. 2 . Determination of r;, with the slack test method. I R ~ S determined in control conditions ( 0 ) in - the presence of .-\mrinone (+) and again in The linear regression analgis control conditions ( 0). provided the folloxing d u e s of slope (mm ms- ', corresponding to 1,) and intercept (mm, corresponding to extension of series elasticity). 1st control, if= 0.1062+0.0041.~; .\mrinone, ,y = 0.1011 + 0.0035 Y ; 2nd control. ,I' = O.O'H3 + 0.0041.~; the length of the trabccula n-as 1.91 mm.

:,

none. T h e results are summarized in Table 3. .4mrinone caused a significant reduction of maximum shortening velocit!. as well as of isometric tension. There was also a small, albeit non-significant increase of the parameter alp,. T h e effkct of -4mrinone on unloaded shortening velocity, determined according to the slack test method. w s studied in six trabeculae. Each trabecula was activated first under control conditions, then in the presence of .\mrinone and finall!. a p i n under control conditions. :lmrinone u a s used at a concentration of 3 mxt to maximize its effects on contractile properties. .4 representative example is shown in Figure 2 . I n the presence of Amrinone the slope of the regression line, which corresponds to unloaded shortening L-elocitJ-, was significantly reduced (see Table 3). 'The intercept, which g i x s a measure of series elasticit!., was not modified. The depressant effect was completely reversible as can be seen in Figure 2 : the slope obtained in the third series of tests was 1012 5 2 4.16 ( n = 6, P < 0.05) O,] of the slope obtained in the first series. In the same six trabeculae, the rate of tension redevelopment after a release of approximately loo,, initial length was measured under control conditions and in the presence of r\mrinone. -4s shown in 'Table 3, the rate of tension re-

0

1

2

3

4

5

6

Arnrinone rnM 3. Effects of Amrinone on ATPase activity of m!'ofibrillar and myosin preparations. ATPase activity in the presence of Amrinone is expressed in relation to the .\TPase actiritk- in control conditions (no '4mrinone in the incubation medium). (:a-Mg dependent A'l'Pase activity is indicated by 0; its control value is 0.294i0.007 pmol m g ' protein min '; each point and bar is the mean and the SE of seven determinations. Ca-stimulated ATPase activity of myosin preparations is indicated by 4 ; its control value is 1.037+0.149 pmol mg-' protein min-'; each point and bar represent the mean and the SE of four determinations. Ca-stimulated ATPase activity of myofibrils is indicated by [?; its control value is 0.163 k0.017 p n o l n i g ' protein min-'; the effect of 3 mlt rimrinone was assessed in four determinations. C:a-\Ig dependent ,.\TPase activity is significantly inhibited at 1,3, i and 6 mr\.f Amrinone. The inhibition of Ca stimulated ATPase activity does not reach statistical significance at any concentration of Arnrinone.

development was significantly reduced by Amrinone.

T h e action of .lmrinone was tested on purified isolated myofibrils and myosin from rat ventricular mi-ocardium. T h e results are shown in Figure 3. Ca-hlg dependent ATPase activity determined at low ionic strength in myofibrillar preparations was inhibited in a dose dependent way by Amrinone. O n the contrary little or n o inhibitory effect was observed on the Ca stimulated ATPase activity determined at high ionic strength in myosin or myofibrillar preparations. T h i s discrepancy might be attributed to an effect of Amrinone on the calcium affinity of the regulatory protein system. I n the myofibrillar preparations the latter is preserved and is

Amrinone efecfs on cardiac muscle

1

6

5

4

0

1

2

3

4

5

27

6

Arnrinone rnM

pCa

Fig. 4. Effect of Amrinone on calcium sensitivity of the myofibrillar preparations. Ca-Mg dependent ATPase activity was determined at different pCa levels in control conditions (no Amrinone in the medium) and in the presence of 3 mM Amrinone. Relative ATPase activity was obtained by subtracting the value of ATPase activity at pCa = 7.8 and normalizing to the maximal ATPase activity (at pCa = 4.4). Each point indicates a determination carried out in duplicate: 0 , control; 0, Amrinone. The curves were obtained by fitting a sigmoid equation y = loo/( 1 ( lOPc")"/( lOpCaSoCqn) by means of the software package GRAPHPAD (IS1 1988). No attempt at statistical comparison between the two curves was carried out.

Fig. 5 . Different inhibitory effect of Amrinone on Ca-Mg-dependent ATPase activity of ventricular myofibrils prepared from euthyroid and hypothyroid rats. ATPase activity in the presence of Amrinone is expressed as a percentage of the ATPase activity in control conditions. ATPase activity of myofibrils from euthyroid rats is indicated by 0; its control value was 0.294 0.007 pmol mg-' protein min-'; each point and bar represents mean and SE of seven determinations. ATPase activity of myofibrils from hypothyroid rats is indicated by 0 ; its control value was 0.181 kO.101 pmol mg-' protein min-l; each point and bar represents mean and standard errors of four determinations. The differences between euthyroid and hypothyroid myofibrils are significant at both Amrinone concentrations (3 and 5 mM).

essential for ATPase activity at low ionic strength. T h e regulatory protein system is not present in myosin preparations and not required for Ca stimulated ATPase activity of myofibrils at high ionic strength. For this reason the calcium sensitivity of ATPase activity at low ionic strength was examined under control conditions and in the presence of 3 mM Amrinone. As can be seen in Figure 4, no difference was present between the curve under control conditions and the curve in the presence of Amrinone, although the absolute ATPase activity was inhibited by 11.3 1.47% ( n = 24) of control values. Since the results of previous studies (Minsson & Edman 1985) on Amrinone action in skeletal muscle suggest that the compound might act directly on the myosin molecule, possible differences between myosin isoforms were taken into consideration. For this reason, the inhibitory action of Amrinone was also tested on myofibrils prepared from ventricular myocardium of hypothyroid rats. As shown in our previous studies (Cappelli et al'. 1988, 1989) ventricular myocardium of young adult (2-3 months old)

euthyroid rats of our strain contains about 80% of the CI isoform of myosin heavy chain (V1 isomyosin). In hypothyroid rats ventricular myocardium has a homogeneous beta isoform (V3 isomyosin) composition. T h e determination of Ca-Mg dependent myofibrillar ATPase activity showed that the inhibitory effect of Amrinone was significantly less in myofibrils prepared from hypothyroid rats than in myofibrils from euthyroid rats at any of the concentrations examined (see Fig. 5).

+

DISCUSSION T h e results obtained in this study provide evidence of a direct action of Amrinone on the myofibrillar system of rat ventricular myocardium. Amrinone was found to reduce maximum shortening velocity, isometric tension and maxim u m rate of tension development in skinned cardiac trabeculae. Furthermore, the Ca-Mgdependent ATPase activity (acto-myosin ATPase) of purified myofibrils was depressed, whereas the Ca stimulated ATPase activity of purified myosin or myofibrils was not affected. These

28

S. E.3. N . Morner et al.

effects were observed at concentrations ranging from 0.3 to 6 mxi. i2t the same concentrations -4mrinone shows a marked inotropic effect in isolated cardiac preparations (Onuaguluchi & Tanz 1981, Gaide et al. 1981, Kondo et al. 1983, Siegl et al. 1984). Values of E C 50°, from 0.5 to 1.5 mM have been found in papillary muscles of different animal species (Siegl et al. 1984). Due to its lower potency, EC 50°., of Amrinone are a t least ten times higher than those of other phosphodiesterase (PDE) inhibitors (Weishaar et al. 198i). Amrinone is generally believed to influence cardiac contractility by modifying intracellular calcium kinetics (,Morgan et al. 1980, Rendig & Amsterdam 1984, IMorner & Wohlfart 1990). The inhibitory effect on PDE and the consequent increase of intracellular cAMP has been shown to correlate well with the positive inotropic action of the drug (Endoh et al. 1982, Honerjager et al. 1981, Weishar et al. 1987). In this study both mechanical and biochemical determinations were carried out on preparations where sarcoplasmic reticulum and sarcolemma were disrupted and removed by the combined action of EGTA and Triton X (Kentish 1988). Activation was directly controlled by setting the pCa of the activating solution and of the ATPase assay medium by means of a CaC1,-EGTA buffer system. Any action via altered Ca kinetics can therefore be excluded. The determination of the pCa-ATPase activity curve for the myofibrillar preparations also demonstrated that calcium sensitivity of the regulatory protein system was not affected by Amrinone even at high concentrations of the drug. The effects of Amrinone described in this study would cause a reduction in the ability of the cardiac muscle to develop force. Actually, Amrinone has been found to have a negative inotropic effect on rat heart both in ziz*o and in ritro (Azari et al. 1980, Weishar et al. 1987) in contrast to the positive inotropic effect observed in other species (rabbit, guinea pig, cat and dog). This difference can be accounted for either by differences in excitation-contraction coupling betn-een rat and other species, by the different intracellular distribution of phosphodiesterase I11 (Weishar et al. 1987) or, finally, in the light of this study, by differences in ventricular isomyosin distribution (V1 is prevailing in the rat, whereas V3 is prevailing in the other above listed species).

Two hypotheses can be forwarded to explain the effects of Amrinone described in this study: (1) due to its high concentration (0.3-6 mM) AMR might interfere with one of the components of the experimental medium, thus altering its ionic composition or ATP availability; and ( 2 ) Amrinone might act directly on the contractile proteins affecting the ATP splitting and the mechanism of shortening and force generation. The first alternative can be ruled out on the basis of the following considerations : Amrinone ionizes both with basic ( p K 1.5 and 3) and acidic ( p K 12) functions. At p H 7, i.e. the p H of all experimental solutions, it behaves as a weak base. Its possible binding to anions which are present at very high concentration, such as C1and creatinephosphate would not bear any significant effect. A possible binding to ATP would reduce MgATP availability. It seems, however, very unlikely that such binding can explain the observed effects. For example, to depress myofibrillar ATPase activity significantly, 1 mM Amrinone would need to reduce the concentration of the available MgATP from about 1.9 mM to less than 0.2 mM (see Glynn & Sleep 1984). The second hypothesis seems more acceptable. Amrinone, which behaves as a weak base, might interact directly with the contractile proteins. A likely target for Amrinone action is myosin which is responsible for ATP splitting and for shortening and force generation. Different myosin heavy chain isoforms exhibit different sensitivity to Amrinone: its depressant effect was found to be lower for the slow V3 myosin than for the fast V1 myosin. An action on the contractile proteins mediated by CAMP-dependent phosphorylation might also be considered. Effects of cAMP on ATPase activity and on force development have been shown by Winegrad et al. (1983) and Winegrad & Weisberg (1987) in rat ventricular myocardium. These effects were, however, demonstrated in cryosections where it is very likely that all enzymatic systems were preserved (Winegrad & Weisberg 1987) or in gently skinned (hyperpermeable) preparations in the presence of PDE inhibitors (Winegrad et al. 1983). T h e preparations used in this study were deeply demembranated by the combined treatment with EGTA and Triton X and were therefore very likely devoid of soluble cytoplasmic enzymes and membrane bound regulatory systems such as kinases and PDEs. As

Amrinone efects on cardiac muscle mentioned above, in rat myocardium only soluble forms of PDE 111 are present (Weishaar et al. 1987). In conclusion, any action mediated by CAMPseems unlikely. Previous studies (Minsson & Edman 1985) carried out on skeletal muscles fibres have analysed Amrinone effects on tetanic isometric tension and maximum shortening velocity in the frame of Huxley’s (1957) cross-bridge model. It was assumed that the force generated by a single cross bridge and the total number of available cross bridges were unaltered by Amrinone. This analysis led to the conclusion that Amrinone reduced both cross-bridge attachment and crossbridge detachment rates, the latter being more affected than the former. This caused a reduction of maximum shortening velocity which is critically dependent on cross-bridge detachment rate. It also caused an increase of the fraction of the attached cross-bridge, resulting in a higher isometric tension. This situation closely resembles the effects of a decrease of MgATP concentration (Cooke & Bialek 1979, Ferenczi et al. 1984, Stienen et al. 1988) and suggests a hypothetical interference of Amrinone with ATP binding on the myosin head. An attempt to analyse Amrinone effects on rat ventricular myocardium along the same lines must take into account that there are some differences between Amrinone effects in cardiac muscle and in skeletal muscle. I n both muscles Amrinone depresses maximum shortening velocity, maximum rate of tension rise (Minsson & Edman 1985) and Ca-Mg dependent myofibrillar ATPase activity (Bottinelli et al. 1989). Unlike skeletal muscle no potentiation of isometric tension was observed in cardiac muscle. The difference can be explained with a quantitative modification of the above reported model analysis. If the attachment rate constant is reduced more than the detachment rate constant, the number of attached cross-bridge will decrease. This might explain the depression of isometric tension observed in cardiac preparations. This study supports the conclusion that the action of Amrinone at myofilament level, previously observed in skeletal muscle (Minsson & Edman 1985), exists also in cardiac muscle. T o our knowledge, only two substances, BDM (Mulieri & Alpert 1984) and ARL 115 BS (Cecchi et al. 1984), have demonstrated the ability to exert a direct effect on the contractile mechanism in the myofilament system. Very

29

recently there have been indications (Beier et al. 1991) that other phosphodiesterase inhibitors, besides Amrinone, are active not only as calcium sensitizers but also as modulators of maximally activated myofibrillar ATPase activity. I n our opinion, the development of molecules capable of modifying the force generation mechanism is a worthy target for future cardiovascular pharmacology. REFERENCES AZARI,J. & HUXTABLE, R.J. 1980. Differential effects of amrinone on contractility and taurine influx in rat and guinea pig hearts. Europ 3 Pharmacol 67, 347-353. BARANY, M. 1967. ATPase activity of myosin correlated with speed of muscle shortening. 3 Gen Physiol 50, 197-216. BEIER,N. & WOLF,H.P. 1991. Comparison of the effects ofEMD 53998 on bovine cardiacmyofibrillar ATPase with those of other calcium sensitisers. 3 Mol Cell Cardiol 23, Suppl V, P39. R., CAPPELLI, V., MORNER, J. & REGGIBOTTINELLI, ANI, C. 1989. Amrinone decreases maximum velocity of shortening and myofibrillar ATPase activity in fast mammalian skeletal muscle. Proc Internat Union Physiol Sci 17, P2445. BOTTINELLI, R., SCHIAFFINO, S. & REGGIANI, C. 1991. Force-velocity relations and myosin heavy chain isoform compositions of skinned fibres from rat skeletal muscle. 3 Physiol437, 655-672. CAPPELLI,V., MOGGIO, R., POLLA,B., BOTTINELLI, R., POGGESI, C. & REGGIANI, C. 1988. The dual effect of thyroid hormones on contractile properties of rat myocardium. Pfiigers Arch 41 1, 62C627. V., BOTTINELLI, R., POGGESI, C., MOGGIO, CAPPELLI, R. & REGGIANI, C. 1989. Shortening velocity and myosin and myofibrillar ATPase activity related to myosin isoenzyme composition during postnatal development in rat myocardium. Circ Res 65, 44-57, G., LOMBARDI, V. & MENCHETTI, G. 1984. CECCHI, Development of force-velocity relation and rise of isometric tetanic tension measure the time course of different processes. Pjiigers Arch 401, 39-01, COOKE,R. & BIALEK,W. 1979. Contraction of glycerinated muscle fibres as a function of the MgATP concentration. Biophys 3 28, 241-258. L. & SCUBON MULIERI, B. EDMAN, K.A.P., MULIERI, 1976. Non-hyperbolic forcevelocity relationship in single muscle fibres. Acta Physiol Scand 98, 143-156. EDMAN,K.A.P. 1979. The velocity of unloaded shortening and its relation to sarcomere length and isometric force in vertebrate muscle fibres. JPhysiol 291, 143-159.

30

S . E . 3 . N . Mfirner et al.

E\DOH, 41., ~ - . A X ~ % S H I T ,s. A , & T.AIR.4, 1. 1982. Positive inotropic effect of amrinone in relation to cyclic nucleotide metabolism in the canine ventricular muscle. j' Phurttiarul 221, 7 - 7 8 3 . F.As1.2~0,-1.1088. (hmputer programs for calculating total from specific free or free from specified totam ionic concentrations in aqucous solutions containing multiple metals and ligands. ,ZIPtli Etiqnrol 157, 387-417. I . II..l.. \l.AV, I . E . & SI\1\iONS, R.XI. 19x4. The dependence of force and shortening velocit! on substrate concentration in skinned rnuscle fibres from Ratru trmporariiz. J Ph.ysrul 350, .ilO 543. GIIDE,hI.s.,BAKER, S.P., EZRIN, .1.11., GELBAUD, H. ? H.~SSE.IT, i . A L . 1981. .Amrinone, isoproterenol and ouahain modification of cardiac K contractures. Eirr J' Phurniac-01 73, 253-260. Gtsns, C L . & KENDI,I.R. 198.i. .1mrinone and energ!- output of rahhit papillary muscles, 3 Curi1ioru.w Phurmaiol 7 , 267-272. C ; I . ~ ZtI. , & SLEEP, J. 1984. Dependence of adenosin triphosphatase acti\ it! of rabbit psoas muscle fibres and m!ofihrils on substrate concentration. .7 P/i,lssiol 36.5 1.59---176. HOSERJ.AG~R, P., SH.AFF,R-KORTIUG, hl. & REITER, 11. 10x1. Iniolvement of cyclic AlIP in the direct inotropic action of amrinone. Biochemical and functional evidence. ~~[z1~ti,),ti-Sihttiienrhrr~'s -4rL.h Phtrritruroi 318, 112-120. HLLF.I. -1.F. 19.57. 1luscle structure and theories of contraction. P r q r Biop/?j,s Btophj's Chetti 7 , 2.5.5 318. I(ESTISH, J.C. 1984. The inhibitor! effects of monovalent ions on force development in detergent skinned ventricular muscle from guinea pig. Pi~,ysiol352, 35.3-374. KONDO, x., SHIBATA, s., k;ODA\1.4, I. & YAXLADA, K. 108.3. Electrical and mechanical effects of amrinone o n isolated guinea pig 1-entricular muscle. .7 Cmfio;,asc.Phartmiol j3903-9 I?. ~ I A I . E C OC.A., ~ I . ARLOCK, P. & K.ATZUNG, B.G. 1985. Amrinonc effects on electromechanical coupling and depolarization-induced automaticitl; in ventricular muscle of guinea pigs and ferrets. J Phurrtiircol rind E v p Ther 232, 10-19. ~IAsSSON, -1.& EDCl.AN, K.A.P. 1985. Effects of amrinone on the contractile behaviour of frog striated muscle fibres. .4cra Ph~ysrol Scntid 125, 48 1---493. llb~ssou,.I.,MORSER,J. & EDLIAN,K..A.P. 1989. Effects of amrinone on twitch, tetanus and shortening kinetics in mammalian skeletal muscle. Acta Phyiol Scand 136, 37-4-i. !V~OR.ANO. I., ARSDT,H., GARTSER, C. & RVEGG, J .C:. ~

1988. Skinned fibres of human atrium and ventricle : ml-soin isoenzyme and contractility. Circ Res 62, 632-6.5 9. MORGAN, J.P., LEE, N.K.M. & BLINKS,J.R. 1980. Alechanism of inotropic action of amrinone: unusual pattern of Ca transients as detected with aequorin. Fed Pro(- 39, 854. MORGAN, J.P., GWATHMEY, J.K., DEFEO, T.T. & ~ I O R G AK.G. N , 1986. T h e effects of amrinone and related drugs on intracellular calcium in isolated mammalian cardiac and vascular smooth muscle. Circ.ulririorr 7 3 (Suppl III), 65-77. ~ I O R N E RS.E.J.N. , & M~NSSON, A. 1990. Effects of amrinone on the electromechanical coupling in frog skeletal muscle fibres. Acta Ph.ysiol Scand 139, 289-295. ~ I O R N E R S.E.J.N. , & WOHLFART, B. 1990. Inotropic mechanisms of amrinone in papillary muscles from guinea pig hearts. -4ctu Ph,ysiol Scand 139, 575-581. Moss, R.L. 1986. Effects on shortening velocity of rabbit skeletal muscle due to variations in the level of thin filament activation. 3Physiul377, 187-505. ~ I C L I E RL.&4. I , & h P E R T , N.R. 1984. Differential effects of ZJ-butanedione monoxime (BDM) on activation and contraction. Biuplqts 3 45, 47a. ONL-AGL-LUCHI, G. & TANZ, R.D. 1981. Cardiac effects of amrinone on rabbit papillary muscle and guinea pig Langendorff heart preparations. 3 Curdiooasc Pharniazol 3, 1342-1 355. POGGESI,C., EVERTS, M., POLLA,B., TANZI, F. & REGGIANI, C. 1987. Influence of thyroid state on mechanical restitution of rat myocardium. Circ Res 60, 142-1.51. RESDIG,S.V. & AMSTERDAM, E.A. 1984. Positive inotropic action of amrinone: effect of elevated external calcium. J Cardini>ascPharmacol 6, 293299. SIEGL,P.K.S., MORGAN,G. & SEET, C.S. 1984. Responses to amrinone in isolated cardiac muscles from cat, rabbit and guinea pig. 3 Cardiozlasc Pharniacol6, 281-287. STIENES, G .J.M., VAN DER LAARSE, W. & ELZINGA, G. 1988. Dependency of the force-velocity relationships on MgATP in different types of muscle fibres from .Yetinpus laeris. Biuph.ys -7 53, 849-8.55. LVEISHAAR, R.E., KORYLARZ-SINGER, D.C., STEFFEN, R.P. & KAPLAN,H.R. 1987. Subclasses of cyclic specific phosphodiesterase in left ventricular muscle and their involvement in regulating myocardial contractility. Circ Res 61, 539-547. WINEGRAD, S., MCCLELLAN, G., TUCKER, M . & LIN, L.E. 1983. Cyclic AMP regulation of myosin isoenzymes in mammalian cardiac muscle. J Gen PIgtsiol 81, 749-765. \t71NEGR.4D, S . & WEISBERG, A. 1987. Isozyme specific modification of myosin ATPase by CAMP in rat heart. C i r r Res 60, 3 8 5 3 9 2 .