- Email: [email protected]
Anthracyclines Enhance Tension Development in Cardiac Muscle by Direct Interaction with the Contractile System
J Mol Cell Cardiol 29, 1001–1008 (1997)
Anthracyclines Enhance Tension Development in Cardiac Muscle by Direct Interaction with the Contractile System∗ Antonio E. Bottone1‡, Evert L. de Beer1 and Emile E. Voest2 1
Department of Medical Physiology and Sports Medicine, Utrecht University, P.O. Box 80043, 3508 TA Utrecht, The Netherlands, 2Department of Internal Medicine, University Hospital Utrecht, P.O. Box 85500, 3508 GA Utrecht, The Netherlands (Received 6 August 1996, accepted in revised form 25 November 1996) A. E. B, E. L. D B E. E. V. Anthracyclines Enhance Tension Development in Cardiac Muscle by Direct Interaction with the Contractile System. Journal of Molecular and Cellular Cardiology (1997) 29, 1001–1008. Anthracyclines are highly effective anticancer agents which induce a well described but incompletely understood cardiac toxicity. In this study, a direct action of several anthracyclines on the force generating mechanism of heart muscle preparations is described. To allow discrimination between membrane related effects and a direct action of anthracyclines on the actin–myosin contractile system, both inner and outer membranes of cardiac fibres were permeabilized. All anthracyclines tested in this study [doxorubicin (Dox), epirubicin, daunorubicin and idarubicin] showed positive inotropic actions. Dox and epirubicin, which are considered the most cardiotoxic drugs of the anthracycline family, significantly increased the maximal calcium activated tension by 33% (n=8, P<0.01) and by 26% (n=8, P<0.01) respectively. Daunorubicin and idarubicin increased the maximal tension by 12% and 9% respectively (P=n.s.). Other chemotherapeutic drugs (Taxol and 5-FU) had no effect on maximal tension. To elucidate the mechanism behind this Dox-induced increase in maximal tension, calcium sensitivity curves were measured and rigor experiments were performed. A small but significant increase in pCa50 value (+0.14±0.03, P<0.05) was observed only after incubation with 20 l Dox. Dox acted during the transition to force generating cross-bridges as reflected by the significant increase in rigor tension (12%, P<0.05) after preincubation of cardiac fibres with Dox. Cycling of cross-bridges is a prerequisite for Dox to increase tension because no effect on tension was seen after Dox was added to fibres in an established rigor. In summary, anthracyclines increased the maximal tension in cardiac muscle fibres by direct interaction with the actin–myosin cross-bridges. Changes in calcium sensitivity are unlikely to contribute to the observed increase in maximal tension. The rise in tension as is seen in this experimental set-up may contribute to destruction of the contractile machinery of cardiac muscle. In agreement with this hypothesis is the observation that the more cardiotoxic anthracyclines induced the largest increase in maximal tension of the cardiac fibres. 1997 Academic Press Limited
K W: Doxorubicin; Anthracycline; Cardiomyopathy; Isometric tension; Muscle.
Introduction Anthracyclines are highly effective anticancer drugs which are used in the treatment of haematological malignancies as well as a variety of solid tumours (Young et al., 1981). Their clinical use, however, is limited by the development of a dosedependent chronic cardiomyopathy (Doroshow,
1991). In spite of extensive research efforts over the past decade, the precise mechanism of anthracycline induced cardiotoxicity remains unclear (Singal et al., 1987). The cardiotoxicity of doxorubicin (Dox), the most frequently used anthracycline drug, is morphologically characterized by myofibrillar loss, vacuolization of the SR, and swelling of the mitochondria (Doroshow et al., 1985). Several
‡ Author for correspondence. ∗ This research was supported by grant 93.074 from the Netherlands Heart Foundation
0022–2828/97/031001+08 $25.00/0
mc960345
1997 Academic Press Limited
1002
A. E. Bottone et al.
hypotheses to explain the cardiomyopathy have been proposed. The most popular ones include free radical formation by anthracyclines (Doroshow, 1983), an impaired myocardial calcium homeostasis through alteration of the function of cardiac sarcoplasmic reticulum (SR) (Holmberg and Williams, 1990; Pessah et al., 1990), and the formation of C-13 hydroxy anthracycline metabolites (Olson et al., 1988). Although treatment with Dox eventually leads to chronic heart failure, Dox increases left ventricular systolic and diastolic function 4 and 24 h after treatment of patients with normal ventricular function (Unverferth et al., 1981; Brown et al., 1989). To obtain insight into the effects of Dox on myocardial contraction, a variety of isolated muscle preparations have been studied over the past years. Exposure of intact heart muscle preparations to Dox has been reported to produce positive inotropic effects in relatively low concentrations and negative inotropic effects in higher concentrations (Azuma et al., 1981; Singal et al., 1987). Others as well have reported an increase of the contractile force at various concentrations of Dox (Van Boxtel et al., 1978; Kim et al., 1980; Temma et al., 1992). In contrast, Hofling and Bolte (1981), Politi et al. (1985), Singal and Pierce (1986) and Voest et al. (1994) reported that doxorubicin produces only negative inotropic effects, while others found no inotropic effect at all (for review see Singal et al., 1987). These conflicting data with regard to the acute contractile effects of Dox may result, at least partially, from differences in drug dosage, the use of different myocardial preparations, the use of various animal species, and differences in the experimental protocol. Most research on the inotropic effect of anthracyclines is done in electrically stimulated intact isolated muscle preparations, providing information on the final effect of anthracyclines on the generation of force. In these models the sarcolemma and the membrane of the SR are intact. This makes it difficult to separate membrane-related effects from direct effects on the actin-myosin interaction. In order to eliminate membrane-related events, we used muscle fibre preparations in which both outer and inner membranes were permeabilized. This allows free entry of small molecules into the muscle fibre and direct access of added drugs to the contractile system. This experimental procedure has provided evidence that in addition to the effect on calcium release, anthracyclines have a direct effect on the actin-myosin contractile system of skinned skeletal muscle preparations (De Beer et al., 1992). The present study provides evidence that a strong
positive inotropic effect of members of the anthracycline family occurs in skinned cardiac preparations. This effect is time-, dose- and anthracycline structure dependent and is specific for the anthracycline family.
Materials and Methods Animals and preparations All experimental animals were given water and standard chow ad libitum, and were kept on a 12-h light–dark cycle. The experiments were approved by the University Experimental Animal Committee. In order to obtain heart preparations, male Wistar rats (250–350 g) were anaesthetized with Nembutal (60 mg/kg bodyweight, i.p.). After tracheotomy, the thorax was opened and the aorta was cannulated. The heart was rapidly removed and connected to a Langendorff perfusion system. The tyrode solution had the following composition (in m): NaCl 130, KCl 4.7, Na2HPO4 0.42, NaHCO3 20.2, glucose 10.1, MgCl2 1.0, CaCl2 2.0, and was continuously gassed with a mixture of 95% O2 and 5% CO2 to maintain pH at 7.4 at 30°C. 2,3 Butanedione monoxime (BDM, 10 m) was added to the tyrode solution to prevent contraction and muscle damage during dissection (Mulieri et al., 1989). Free running trabeculae ranging from 50 to 150 lm in dia. and 1 to 2 mm in length were dissected carefully from the right ventricle wall and skinned by exposure to Triton X-100 (1% v/v) for 30 min. We used only one trabecula out of each experimental animal. Triton X-100 rendered both the sarcolemma and inner membrane structures permeable for small ions and molecules. The skinning solution was removed by washing with relaxation solution. Table 1 summarizes the various solutions used in the experimental protocol.
Apparatus The muscle preparation was mounted with a fast setting glue between a force transducer (Scientific Instruments, Heidelberg, Germany) and a fixed support for isometric force measurement. Rigor experiments were performed using a Sensonor AE801 (Horten, Norway) force transducer. Sarcomere length was continuously measured by means of laser diffraction in which the first order diffraction pattern was monitored on a screen placed behind the preparation. The output of the force transducer
1003
Anthracyclines Enhance Tension Development Table 1 Composition of solutions (in m)
MOPS EGTA HDTA EDTA CP MgATP pCa Mg2+
Relaxation
Jump
Activation
Rigor
10.0 7.0 – – 10.0 5.0 10 1.0
10.0 0.2 6.8 – 10.0 5.0 10 1.0
10.0 7.0 – – 10.0 5.0 4 1.0
10.0 7.0 – 1.0 – – 10 –
All concentrations are expressed in m. Ionic strength is 160 m at pH 7.0 and at 22°C. All solutions, except rigor, contained 50 U/ml creatine kinase (where one unit transfers 1.0 lmol of phosphate from creatine phosphate to ADP per min at pH 7.4 and at 30°C).
was recorded on a flat bed recorder and a digital voltmeter. The signals were also digitized by a computer with a AD-card (Keithley DAS 1602). The width and depth of the muscles were measured immediately after the experiments using an eyepiece graticule fitted in the dissecting microscope. The cross-sectional area was calculated on the assumption that the cross-sections were perfectly circular or ellipsoid. The experimental solutions were carried in a series of wells (1.25 ml each) in a temperature controlled stainless steel block. The sides of the wells were made of glass allowing laser diffraction measurement at any moment during the experiment. Solutions were changed by lifting the muscle out of the solution and sliding the block horizontally to bring the next solution under the muscle. The emergence of the muscle through the solution meniscus, as the preparation was transferred to a next solution, removes any significant adhering droplets. These movements took about 1 s each for all present experimental protocols. All experiments were performed at 22°C and at pH 7.0.
of the preparation. The pCa values of these activation solutions ranged from 6.17 to 4.00, which was obtained by mixing relaxation and activation stock solutions. The effect of incubation time (range: 3 min–120 min) with anthracyclines on the calcium sensitivity response was tested.
Evaluation of rigor tension in trabeculae Rigor was forced by immersing the fibre in rigor solution without ATP and calcium (pCa=10). In order to study the effect of Dox on the level of stable rigor tension, fibres were incubated with Dox (20 l) in relaxation solution for 30 min and subsequently transferred to rigor solution containing the same Dox concentration. To study the effect of Dox on the plateau of stable rigor tension, Dox was added to control rigor solution as soon as a stable rigor tension was reached. All tension responses were digitized by a computer for further analysis.
Data analysis Experimental protocols
Measurement of maximal tension and calcium sensitivity of trabeculae The muscles were placed in a high-buffering relaxation fluid with a pCa>9 for stabilization. In this initial period the sarcomere length was set at 2.15 lm and controlled throughout the experiment. Before activation, the fibre was placed in a lowbuffering pre-activating “jump” solution. During every activation cycle the fibre was immersed in solutions with increasing free Ca2+ concentrations resulting in a stepwise increase in the force response
Results are expressed as mean±... where appropriate. The normalized Ca2+ sensitivity curves, which were obtained by stepwise increasing the calcium concentration surrounding the preparation, were fitted to the Hill equation: T([Ca2+])/Tmax=[Ca2+]n/(Kn + [Ca2+]n) In this Hill equation, Tmax is the maximal isometric tension at saturated Ca2+ binding sites, [Ca2+] the Ca2+ concentration, T([Ca2+]) is the isometric tension at a certain Ca2+ concentration and K and n are constants to be fitted. K is the Ca2+ concentration necessary to develop half maximal tension (T(K)= 0.5.×Tmax), while n represents the total number of
1004
A. E. Bottone et al.
150
Tmax (% of control)
140 130
† †
120
‡ 110 100 90
0
30 60 90 Incubation time (min)
120
Figure 1 Shows a time- and dose-dependent effect of incubation with Dox on the calcium activated tension at pCa 4. Each tension value is taken relative to the control tension before incubation. Values are means of eight contractions±standard errors of means. Legend: (Β) control, (Χ) 5 l Dox, (Ε) 10 l Dox, (Ο) 20 l Dox. At all concentrations of Dox the tension curves were significantly higher than the control tension curve; †, P<0.01; ‡, P<0.05.
binding sites involved. Data values are given as mean±... A Student’s t-test was used to test significance of differences at a 0.05 level of significance (P<0.05). The tension curves measured during incubation with anthracyclines were compared with the control tension curve using an analysis of variance (ANOVA) to test for an effect at the 0.05 level of significance.
Results The effect of anthracyclines on maximal tension and Ca2+ sensitivity of trabeculae The trabeculae had a mean diameter (mean of the width and the depth measurements) of 79±4 lm and generated a peak tension of 65±4 kN/m2 (n= 58) under control conditions. Incubation with anthracyclines resulted in a time- and dose-dependent increase of the maximal calcium activated tension of skinned cardiac muscles, as is shown for Dox in Figure 1. This increase in tension was significant for Dox (5 l: P<0.05; 10 lm and 20 l: P<0.01). The resting tension of the trabeculae was not increased upon incubation with anthracyclines. To determine whether the inotropic effect was specific for the anthracycline family, two commonly used
non-anthracycline anti-cancer agents, Taxol and 5-fluoro-uracil (5-FU), were included in the experimental protocol. At a concentration of 20 l no effect on the contractile force was observed. The next step involved comparison of a variety of clinically available anthracyclines for their effect on contractile force. Dox and its stereoisomere epirubicin significantly increased the maximal tension, whereas daunorubicin and idarubicin showed a lesser positive inotropic action (Table 2). In order to investigate whether this positive inotropic effect was the result of an increased calcium sensitivity, partial activation curves were determined before and during anthracyline incubation. Figure 2 shows a typical registration of calcium sensitivity curves before and during incubation with 20 l Dox. The calcium sensitivity of heart preparations was significantly (P<0.05) increased upon 3, 30 and 60 min of incubation with 20 l Dox. At 60 min this effect was maximal and at that point calcium sensitivity was increased by 0.14±0.03 (n=8, P<0.05). Incubation with the other anthracyclines had no effect on calcium sensitivity (Table 2). The Hill coefficient of the calcium sensitivity curves was not significantly altered following anthracycline incubation, indicating that the cooperativity between the calcium binding sites of troponin-C was not changed by treatment with anthracyclines.
Effect of doxorubicin on rigor tension of trabeculae To determine whether the increase in active tension upon anthracycline incubation was the result of a change in the conformation of the strongly bound cross-bridges, Dox was applied to trabeculae which were already in a stable rigor contraction. We found that Dox (20 l) had no effect on the level of maintained tension when applied to the preparation during an established rigor contraction (data not shown). Next, the effect of Dox on the stable rigor tension was investigated after incubation with 20 l Dox for 30 min and subsequently transferring the preparation to rigor solution containing the same Dox concentration. After 30 min incubation in control relaxing solution, the level of maintained rigor tension was diminished as compared to the rigor tension that was measured before the start of the incubation [Fig. 3(a)]. After incubation with 20 l Dox, however, the level of maintained rigor tension was increased as compared to the rigor tension that was measured before the start of the incubation [Fig. 3(b)]. The difference between the decrease in rigor tension in the control
1005
Anthracyclines Enhance Tension Development Table 2 Effect of 60 min incubation with several drugs on maximal calcium activated tension, calcium sensitivity and Hill coefficient of rat trabeculae
Control Doxorubicin Epirubicin Daunorubicin Idarubicin Taxol 5-FU
(n=8) (n=8) (n=5) (n=5) (n=5) (n=6) (n=5)
Tmax
D pCa50
D nHill
102.5±2.8 133.3±7.6† 126.5±7.8† 111.6±1.4 108.9±6.3 99.4±1.0 105.5±4.4
– +0.14±0.03‡ +0.04±0.02 +0.01±0.04 −0.11±0.07 −0.07±0.04 −0.07±0.08
– +0.11±0.17 +0.04±0.11 −0.04±0.03 −0.14±0.14 +0.24±0.08 +0.10±0.11
Shows the effect of incubation with several anthracyclines and other chemotherapeutic agents (concentration: 20 l) on the calcium activated tension at pCa 4 (Tmax), on the calcium sensitivity (D pCa50) and on the Hill coefficient (D nHill). Tmax is expressed as the percentage of the control value before the start of incubation with the drugs. The effect on pCa50 and nHill is expressed as the change in value as a result of incubation with drugs. The average pCa50 value was 6.02±0.03 (n=42), and the average nHill coefficient was 1.73±0.09 (n=42) at the start of the experiments. †, P<0.01; ‡, P<0.05.
Tension (kN/m2)
experiment and the increase in rigor tension in the fibres exposed to Dox was 12% (n=7, P<0.05).
Discussion
0
3
30 60 Time (min)
90
120
Figure 2 Shows calcium sensitivity curves measured before and after several incubation times with 20 l Dox, shown by a registration of a typical experiment. 0: control calcium sensitivity curve. The subsequent curves were measured after 3, 30, 60, 90 and 120 min incubation with 20 l Dox. Partial activations were performed at pCa 9 (relaxation), 6.17, 5.96, 5.69, 5.22 and 4.00. pCa50 value was 5.87 and nHill coefficient was 2.20 for the control curve. Calibration: horizontal 2 min, vertical 10 kN/m2.
(a)
(b)
Figure 3 Shows examples of tracings obtained in two typical rigor experiments. (a) Control experiment; (b) effect of incubation with Dox. Solid line shows the rigor contraction after 30 min of incubation with either control relaxation solution (a) or 20 l Dox (b); the dotted line shows the rigor contraction within the same fibre before the incubation. Calibration: horizontal 30 s, vertical 10 kN/m2.
Anthracycline cardiomyopathy still presents an enormous clinical problem. Patients may be deprived of a very potent anti-cancer drug if they have reached the cumulative dose after which there is a high risk for developing congestive heart failure. The complexity of anthracycline-induced cardiomyopathy is reflected by the many different actions of anthracyclines. The targets of these drugs include the sarcolemma, intracellular membranes such as the SR, and the contractile proteins, the subject of the present study. In this study, a novel mechanism is described. Anthracyclines caused a strong inotropic effect by direct interaction with the actinmyosin system. We used rather thin trabeculae (diam: 79±4 lm, n=58) in order to avoid diffusion limited processes within the preparations. Even in thin muscle preparations, however, diffusion limitations might interfere with processes involving muscle contraction, and they might therefore be of influence for the magnitude of the tension response (Kentish and Jewell, 1984; Stienen et al., 1990). In the control measurements, we found a small increase in tension between the contraction before the solution change and contraction after immersion in fresh superfusion fluid (at t=3 min). We hypothesize that this slight increase in maximal calcium activated tension may be the result of organic phosphate accumulation within the fibre and/or inside the well during the first test contraction (Stienen et al., 1990). Therefore, any drug-related effect on calcium activated tension has to be well beyond this limit.
1006
A. E. Bottone et al.
The observed inotropic effect seems specific for anthracyclines, because the other chemotherapeutic agents tested in this study had no effect on tension development. Among the anthracyclines that were studied, Dox and epirubicin were the most potent in enhancing the maximal tension of heart preparations. We used concentrations which closely match plasma concentrations found in patients during anthracycline chemotherapy. Epirubicin is a stereoisomer of Dox, with an opposite orientation of the hydroxyl group at carbon atom 4′. The other anthracyclines tested in this study, idarubicin and daunorubicin, also had a positive inotropic effect, but they did not significantly increase the maximal tension. Anthracyclines vary in their chemical structure in distinct places inside the molecule (Arcamone, 1985), which might reflect their different ability in augmenting the maximal tension of skinned heart preparations in this experimental set-up. Modifications in the chemical structure of anthracyclines have been used to reduce their cardiotoxic side effects. Epirubicin, daunorubicin, and idarubicin are examples of structural analogues which are now used in the treatment of patients with cancer. These analogues have been compared with Dox for their effects on the heart. Randomized clinical trials have shown that, of the clinically used anthracyclines, Dox is the most cardiotoxic anthracycline and that epirubicin and idarubicin are less cardiotoxic (Jain et al., 1985; Cersosimo, 1992). The position of daunorubicin as a cardiotoxic anthracycline is less clear but may be considered slightly less or equal cardiotoxic as Dox (Gilladoga et al., 1976; Jaenke et al., 1980). Our results are in agreement with the observed rise in systolic and diastolic bloodpressure during the first 24 h in patients treated with Dox (Unverferth et al., 1981; Brown et al., 1989). Chronic anthracycline cardiomyopathy, however, is characterized by a gradual loss in contractile force. One hypothesis for this biphasic action of Dox might be that the positive inotropic effect of Dox eventually results in a disruption of the contractile machinery, thereby leading to an attenuation of tension. Calcium plays, among other pathways, an important role in anthracycline cardiotoxicity (Singal et al., 1987) although application of the calcium entry-blocker verapamil did not prevent the cardiotoxic effect of Dox (Rabkin, 1983; Rabkin et al., 1983; Rabkin and Godin, 1985). We investigated whether the observed increase in contractile force was the result of an increased calcium sensitivity of the trabeculae. Determination of the calcium sensitivity of the preparations at the start of each experiment resulted in an average control pCa50
value of 6.02±0.03 (n=42) and an average Hill coefficient of 1.73±0.09 (n=42). These values are in close range with values found in the literature (Harrison et al., 1988). However, it is important to note that differences in solution parameters like pH, ionic strength, temperature and several concentrations of important ions all influence the shape and the position of the pCa-tension curve (Fabiato and Fabiato, 1978; Kentish, 1984; Harrison and Bers, 1990; Kentish, 1986). The average maximal tension developed in the control experiments was 65±4 kN/m2 (n=58). This value is in close agreement with values of EGTA-treated trabeculae reported by Kentish and Jewell (1984). Although incubation with 20 l Dox gave rise to a small, but significant leftward shift of the calcium sensitivity curve after 3, 30 and 60 min of incubation, it seems unlikely that this limited increase of about 0.10 of calcium sensitivity of the contractile system can explain for the large inotropic effect observed after incubation with Dox. Since we used preparations in which all membrane structures were eliminated, it is more likely that the inotropic effect of Dox is mediated by direct interaction with the force generating filaments, i.e. actin and/or myosin, rather than being mediated by alteration of the calcium sensitivity of the troponin/tropomyosin complex on the thin filament. The inotropic effect of Dox may be explained by several mechanisms: an increase of the tension per cross-bridge, an altered distribution of the crossbridges, or an altered distribution of the length of the cross-bridges in the force generating state. The direct effect of Dox on the quality of the crossbridges was investigated in muscles that were forced into the rigor state. Because even a very small length change (> 0.1%) of a trabecula during an isometric rigor contraction will affect the force development during a rigor contraction (De Winkel et al., 1995), we focused on the stable tension level of an established rigor contraction. The results of the rigor experiments showed that Dox had no effect on the level of tension of an established rigor contraction. During normal activation, the crossbridges cycle through bound and unbound states leading to an equilibrium giving a stable tension. In the absence of ATP, however, myosin binds actin with high affinity forming rigor cross-bridges. Because of the absence of ATP this leads to a situation in which cross-bridges are arrested in a strongly bound stage and in which cycling of crossbridges no longer occurs. The inability of Dox to increase stable rigor force implies that the stiffness of the already equilibrated strongly bound rigor bridges was not affected. If the cardiac muscles
Anthracyclines Enhance Tension Development
were incubated with 20 l Dox in relaxation solution, and subsequently transferred to rigor solution to which the same Dox concentration was added, the level of maintained rigor tension was significantly increased by 12% as compared with the effect of control incubation on the level of rigor force. The results of the rigor experiments indicate that Dox acts during the transition to force-generating cross-bridges rather than on the end product, i.e. that the cross-bridges need to be actively cycling for Dox to potentiate force. One explanation might be that Dox alters the coupling and/or the decoupling of cross-bridges and thus the distribution of cross-bridges during the activation cycle. During a rigor contraction, the level of tension is determined by the ratio of “strained” and “unstrained” crossbridges (Kawai and Brandt, 1976). If the coupling and/or decoupling of cross-bridges is affected by Dox, the ratio between “strained” and “unstrained” cross-bridges will be altered. In our experiments this may have lead to a higher mean tension development by the cross-bridges in rigor and might explain for the increase in tension upon incubation with Dox. However, more study is necessary to further elucidate the mechanism of this positive inotropic action. In summary, insight into the mechanisms underlying anthracycline-induced cardiotoxicity is essential in developing strategies to prevent the development of a disabling cardiomyopathy. We showed that Dox and other anthracyclines exert a positive inotropic action in skinned heart muscles by direct interaction with the contractile system. This effect may contribute to the observed cardiotoxic effect after Dox treatment and warrants further investigation.
References A F, 1985. Properties of antitumour anthracyclines and new developments in their application: Cain memorial award lecture. Cancer Res 45: 5995–5999. A J, S N, H H, T T, V S, O K, A N, S A, H H, I T, M Y, Y Y, 1981. Adriamycin cardiotoxicity: possible pathogenic mechanisms. J Mol Cell Cardiol 13: 381–397. B KA, B AJ, W RM, S JA, 1989. Acute effects of doxorubicin on human left ventricular systolic and diastolic function. Am Heart J 118: 979– 982. C RJ, 1992. Idarubicin: an anthracycline antineoplastic agent. Clin Pharm 11: 152–167. D B EL, F H, V EE, V H BGV, S P, 1992. Doxorubicin interacts directly with
1007
skinned single skeletal muscle fibers. Eur J Pharmacol 214: 97–100. D W ME, B T, T BW, 1995. Viscoelastic properties of cross bridges in cardiac muscle. Am J Physiol 268: H987–H998. D JH, 1983. Effect of anthracycline antibiotics on oxygen radical formation in rat heart. Cancer Res 43: 460–472. D JH, 1991. Doxorubicin-induced cardiac toxicity. N Engl J Med 324: 843–845. D JH, T C, S JE, 1985. Ultrastructural features of Adriamycin-induced skeletal and cardiac muscle toxicity. Am J Pathol 118: 288–297. F A, F F, 1978. Effects of pH on myofibrils and sarcoplasmic reticulum of skinned cells from cardiac and skeletal muscles. J Physiol (Lond) 276: 233– 256. G AC, M C, T CT, W N, S SS, M ML, 1976. The cardiotoxicity of adriamycin and daunomycin in children. Cancer 37: 1070–1078. H SM, L C, M DJ, 1988. Hysteresis and the length dependence of calcium sensitivity in chemically skinned rat cardiac muscle. J Physiol (Lond) 401: 115–143. H SM, B DM, 1990. Temperature dependence of myofibrillar Ca2+-sensitivity of rat, guinea pig and frog ventricular muscle. Am J Physiol 258: C241–C281. H B, B HD, 1981. Acute negative inotropic effect of adriamycin (doxorubicin). Naunyn-Schmied Arch Pharmacol 317: 252–256. H SRM, W AJ, 1990. Patterns of interaction between anthraquinone drugs and the adriamycin-release channel from cardiac sarcoplasmic reticulum. Circ Res 67: 272–283. J RS, D-DC D, T A, 1980. Cardiotoxicity and comparative pharmacokinetics of six anthracyclines in the rabbit. Cancer Res 40: 3530– 3536. J KK, C ES, G NL, H TB, K RJ, C V, S W, C C, P GR, Y CW, W RE, 1985. A prospective randomized comparison of epirubicin and doxorubicin in patients with advanced breast cancer. J Clin Oncol 3: 818–826. K M, B PW, 1976. Two rigor states in skinned crayfish single muscle fibres. J Gen Physiol 68: 267– 280. K JC, 1984. The inhibitory effects of monovalent ions on force development in detergent-skinned ventricular muscle from guinea pig. J Physiol (Lond) 352: 353–374. K JC, 1986. The effects of inorganic phosphate and creatine phosphate on force production in skinned muscles from rat ventricle. J Physiol (Lond) 370: 585– 604. K JC, J BR, 1984. Some characteristics of Ca2+-regulated force production in EGTA-treated muscle from rat heart. J Gen Physiol 84: 83–99. K DH, A T, B TM, 1980. Inotropic actions of doxorubicin in isolated guinea pig atria. Evidence for lack of involvement of Na+, K+-adenosine triphosphatase. J Pharmacol Exp Ther 214: 368–374. M LA, H G, I F, B EM, A NR, 1989. Protection of human left ventricular myocardium from cutting injury with 2,3 butanedione monoxime. Circ Res 65: 1441–1444.
1008
A. E. Bottone et al.
O RD, M PS, B DE, F S, C BJ, C BK, B RJ J, 1988. Doxorubicin cardiotoxicity may be caused by its metabolite, doxorubicinol. Proc Natl Acad Sci USA 85: 3585–3589. P IN, D EL, S MJ, Z I, 1990. Anthraquinone-sensitized Ca2+ release channel from rat cardiac sarcoplasmic reticulum: possible receptormediated mechanism of doxorubicin cardiomyopathy. Mol Pharmacol 37: 503–514. P PM, R RP, P JEB, 1985. Acute effects of doxorubicin on chronotropic and inotropic mechanisms in guinea pig atria. Cancer Treat Rep 69: 859– 865. R SW, 1983. Interaction of external calcium concentration and verapamil on the effects of doxorubicin (adriamycin) in the isolated heart preparation. J Cardiovasc Pharmacol 5: 848–855. R SW, O M, P PI, 1983. Increased mortality with cardiotoxic doses of Adriamycin after verapamil pretreatment despite prevention of myocardial calcium accumulation. Can J Physiol Pharmacol 61: 1050–1056. R SW, G DV, 1985. Adriamycin cardiotoxicity and calcium entry blockers: the need for caution in the combination. Can J Cardiol 1: 4–7. S PK, P GN, 1986. Adriamycin stimulates low-affinity Ca2+ binding and lipid peroxidation but
depresses myocardial function. Am J Physiol 250: H419–H425. S PK, D CMR, W LE, 1987. Subcellular effects of adriamycin in the heart: a concise review. J Mol Cell Cardiol 19: 817–828. S GJM, R MCM, W MGA, E G, 1990. Depression of force by phosphate in skinned skeletal muscle fibres of the frog. Am J Physiol 259: C349–C357. T K, A T, C A, O M, Y M, K H, 1992. Doxorubicin: an antagonist of muscarinic receptors in guinea pig heart. Eur J Pharmacol 220: 63–69. U DV, M RD, U BP, T RL, B SP, B N, 1981. Human myocardial morphological and functional changes in the first 24 h after doxorubicin administration. Canc Treat Rep 65: 1093–1097. V B CJ, O RD, B RC, O JA, 1978. Doxorubicin: inotropic effects and inhibitory action of ouabain. J Pharmacol Exp Ther 207: 277–283. V EE, A S, V WJF, A BS, B A, 1994. Comparison of different iron chelators as protective agents against doxorubicin-induced cardiotoxicity. J Mol Cell Cardiol 26: 1179–1185. Y RC, O RF, M CE, 1981. The anthracycline antineoplastic drugs. N Engl J Med 305: 139–153.
Anthracyclines Enhance Tension Development in Cardiac Muscle by Direct Interaction with the Contractile System∗ Antonio E. Bottone1‡, Evert L. de Beer1 and Emile E. Voest2 1
Department of Medical Physiology and Sports Medicine, Utrecht University, P.O. Box 80043, 3508 TA Utrecht, The Netherlands, 2Department of Internal Medicine, University Hospital Utrecht, P.O. Box 85500, 3508 GA Utrecht, The Netherlands (Received 6 August 1996, accepted in revised form 25 November 1996) A. E. B, E. L. D B E. E. V. Anthracyclines Enhance Tension Development in Cardiac Muscle by Direct Interaction with the Contractile System. Journal of Molecular and Cellular Cardiology (1997) 29, 1001–1008. Anthracyclines are highly effective anticancer agents which induce a well described but incompletely understood cardiac toxicity. In this study, a direct action of several anthracyclines on the force generating mechanism of heart muscle preparations is described. To allow discrimination between membrane related effects and a direct action of anthracyclines on the actin–myosin contractile system, both inner and outer membranes of cardiac fibres were permeabilized. All anthracyclines tested in this study [doxorubicin (Dox), epirubicin, daunorubicin and idarubicin] showed positive inotropic actions. Dox and epirubicin, which are considered the most cardiotoxic drugs of the anthracycline family, significantly increased the maximal calcium activated tension by 33% (n=8, P<0.01) and by 26% (n=8, P<0.01) respectively. Daunorubicin and idarubicin increased the maximal tension by 12% and 9% respectively (P=n.s.). Other chemotherapeutic drugs (Taxol and 5-FU) had no effect on maximal tension. To elucidate the mechanism behind this Dox-induced increase in maximal tension, calcium sensitivity curves were measured and rigor experiments were performed. A small but significant increase in pCa50 value (+0.14±0.03, P<0.05) was observed only after incubation with 20 l Dox. Dox acted during the transition to force generating cross-bridges as reflected by the significant increase in rigor tension (12%, P<0.05) after preincubation of cardiac fibres with Dox. Cycling of cross-bridges is a prerequisite for Dox to increase tension because no effect on tension was seen after Dox was added to fibres in an established rigor. In summary, anthracyclines increased the maximal tension in cardiac muscle fibres by direct interaction with the actin–myosin cross-bridges. Changes in calcium sensitivity are unlikely to contribute to the observed increase in maximal tension. The rise in tension as is seen in this experimental set-up may contribute to destruction of the contractile machinery of cardiac muscle. In agreement with this hypothesis is the observation that the more cardiotoxic anthracyclines induced the largest increase in maximal tension of the cardiac fibres. 1997 Academic Press Limited
K W: Doxorubicin; Anthracycline; Cardiomyopathy; Isometric tension; Muscle.
Introduction Anthracyclines are highly effective anticancer drugs which are used in the treatment of haematological malignancies as well as a variety of solid tumours (Young et al., 1981). Their clinical use, however, is limited by the development of a dosedependent chronic cardiomyopathy (Doroshow,
1991). In spite of extensive research efforts over the past decade, the precise mechanism of anthracycline induced cardiotoxicity remains unclear (Singal et al., 1987). The cardiotoxicity of doxorubicin (Dox), the most frequently used anthracycline drug, is morphologically characterized by myofibrillar loss, vacuolization of the SR, and swelling of the mitochondria (Doroshow et al., 1985). Several
‡ Author for correspondence. ∗ This research was supported by grant 93.074 from the Netherlands Heart Foundation
0022–2828/97/031001+08 $25.00/0
mc960345
1997 Academic Press Limited
1002
A. E. Bottone et al.
hypotheses to explain the cardiomyopathy have been proposed. The most popular ones include free radical formation by anthracyclines (Doroshow, 1983), an impaired myocardial calcium homeostasis through alteration of the function of cardiac sarcoplasmic reticulum (SR) (Holmberg and Williams, 1990; Pessah et al., 1990), and the formation of C-13 hydroxy anthracycline metabolites (Olson et al., 1988). Although treatment with Dox eventually leads to chronic heart failure, Dox increases left ventricular systolic and diastolic function 4 and 24 h after treatment of patients with normal ventricular function (Unverferth et al., 1981; Brown et al., 1989). To obtain insight into the effects of Dox on myocardial contraction, a variety of isolated muscle preparations have been studied over the past years. Exposure of intact heart muscle preparations to Dox has been reported to produce positive inotropic effects in relatively low concentrations and negative inotropic effects in higher concentrations (Azuma et al., 1981; Singal et al., 1987). Others as well have reported an increase of the contractile force at various concentrations of Dox (Van Boxtel et al., 1978; Kim et al., 1980; Temma et al., 1992). In contrast, Hofling and Bolte (1981), Politi et al. (1985), Singal and Pierce (1986) and Voest et al. (1994) reported that doxorubicin produces only negative inotropic effects, while others found no inotropic effect at all (for review see Singal et al., 1987). These conflicting data with regard to the acute contractile effects of Dox may result, at least partially, from differences in drug dosage, the use of different myocardial preparations, the use of various animal species, and differences in the experimental protocol. Most research on the inotropic effect of anthracyclines is done in electrically stimulated intact isolated muscle preparations, providing information on the final effect of anthracyclines on the generation of force. In these models the sarcolemma and the membrane of the SR are intact. This makes it difficult to separate membrane-related effects from direct effects on the actin-myosin interaction. In order to eliminate membrane-related events, we used muscle fibre preparations in which both outer and inner membranes were permeabilized. This allows free entry of small molecules into the muscle fibre and direct access of added drugs to the contractile system. This experimental procedure has provided evidence that in addition to the effect on calcium release, anthracyclines have a direct effect on the actin-myosin contractile system of skinned skeletal muscle preparations (De Beer et al., 1992). The present study provides evidence that a strong
positive inotropic effect of members of the anthracycline family occurs in skinned cardiac preparations. This effect is time-, dose- and anthracycline structure dependent and is specific for the anthracycline family.
Materials and Methods Animals and preparations All experimental animals were given water and standard chow ad libitum, and were kept on a 12-h light–dark cycle. The experiments were approved by the University Experimental Animal Committee. In order to obtain heart preparations, male Wistar rats (250–350 g) were anaesthetized with Nembutal (60 mg/kg bodyweight, i.p.). After tracheotomy, the thorax was opened and the aorta was cannulated. The heart was rapidly removed and connected to a Langendorff perfusion system. The tyrode solution had the following composition (in m): NaCl 130, KCl 4.7, Na2HPO4 0.42, NaHCO3 20.2, glucose 10.1, MgCl2 1.0, CaCl2 2.0, and was continuously gassed with a mixture of 95% O2 and 5% CO2 to maintain pH at 7.4 at 30°C. 2,3 Butanedione monoxime (BDM, 10 m) was added to the tyrode solution to prevent contraction and muscle damage during dissection (Mulieri et al., 1989). Free running trabeculae ranging from 50 to 150 lm in dia. and 1 to 2 mm in length were dissected carefully from the right ventricle wall and skinned by exposure to Triton X-100 (1% v/v) for 30 min. We used only one trabecula out of each experimental animal. Triton X-100 rendered both the sarcolemma and inner membrane structures permeable for small ions and molecules. The skinning solution was removed by washing with relaxation solution. Table 1 summarizes the various solutions used in the experimental protocol.
Apparatus The muscle preparation was mounted with a fast setting glue between a force transducer (Scientific Instruments, Heidelberg, Germany) and a fixed support for isometric force measurement. Rigor experiments were performed using a Sensonor AE801 (Horten, Norway) force transducer. Sarcomere length was continuously measured by means of laser diffraction in which the first order diffraction pattern was monitored on a screen placed behind the preparation. The output of the force transducer
1003
Anthracyclines Enhance Tension Development Table 1 Composition of solutions (in m)
MOPS EGTA HDTA EDTA CP MgATP pCa Mg2+
Relaxation
Jump
Activation
Rigor
10.0 7.0 – – 10.0 5.0 10 1.0
10.0 0.2 6.8 – 10.0 5.0 10 1.0
10.0 7.0 – – 10.0 5.0 4 1.0
10.0 7.0 – 1.0 – – 10 –
All concentrations are expressed in m. Ionic strength is 160 m at pH 7.0 and at 22°C. All solutions, except rigor, contained 50 U/ml creatine kinase (where one unit transfers 1.0 lmol of phosphate from creatine phosphate to ADP per min at pH 7.4 and at 30°C).
was recorded on a flat bed recorder and a digital voltmeter. The signals were also digitized by a computer with a AD-card (Keithley DAS 1602). The width and depth of the muscles were measured immediately after the experiments using an eyepiece graticule fitted in the dissecting microscope. The cross-sectional area was calculated on the assumption that the cross-sections were perfectly circular or ellipsoid. The experimental solutions were carried in a series of wells (1.25 ml each) in a temperature controlled stainless steel block. The sides of the wells were made of glass allowing laser diffraction measurement at any moment during the experiment. Solutions were changed by lifting the muscle out of the solution and sliding the block horizontally to bring the next solution under the muscle. The emergence of the muscle through the solution meniscus, as the preparation was transferred to a next solution, removes any significant adhering droplets. These movements took about 1 s each for all present experimental protocols. All experiments were performed at 22°C and at pH 7.0.
of the preparation. The pCa values of these activation solutions ranged from 6.17 to 4.00, which was obtained by mixing relaxation and activation stock solutions. The effect of incubation time (range: 3 min–120 min) with anthracyclines on the calcium sensitivity response was tested.
Evaluation of rigor tension in trabeculae Rigor was forced by immersing the fibre in rigor solution without ATP and calcium (pCa=10). In order to study the effect of Dox on the level of stable rigor tension, fibres were incubated with Dox (20 l) in relaxation solution for 30 min and subsequently transferred to rigor solution containing the same Dox concentration. To study the effect of Dox on the plateau of stable rigor tension, Dox was added to control rigor solution as soon as a stable rigor tension was reached. All tension responses were digitized by a computer for further analysis.
Data analysis Experimental protocols
Measurement of maximal tension and calcium sensitivity of trabeculae The muscles were placed in a high-buffering relaxation fluid with a pCa>9 for stabilization. In this initial period the sarcomere length was set at 2.15 lm and controlled throughout the experiment. Before activation, the fibre was placed in a lowbuffering pre-activating “jump” solution. During every activation cycle the fibre was immersed in solutions with increasing free Ca2+ concentrations resulting in a stepwise increase in the force response
Results are expressed as mean±... where appropriate. The normalized Ca2+ sensitivity curves, which were obtained by stepwise increasing the calcium concentration surrounding the preparation, were fitted to the Hill equation: T([Ca2+])/Tmax=[Ca2+]n/(Kn + [Ca2+]n) In this Hill equation, Tmax is the maximal isometric tension at saturated Ca2+ binding sites, [Ca2+] the Ca2+ concentration, T([Ca2+]) is the isometric tension at a certain Ca2+ concentration and K and n are constants to be fitted. K is the Ca2+ concentration necessary to develop half maximal tension (T(K)= 0.5.×Tmax), while n represents the total number of
1004
A. E. Bottone et al.
150
Tmax (% of control)
140 130
† †
120
‡ 110 100 90
0
30 60 90 Incubation time (min)
120
Figure 1 Shows a time- and dose-dependent effect of incubation with Dox on the calcium activated tension at pCa 4. Each tension value is taken relative to the control tension before incubation. Values are means of eight contractions±standard errors of means. Legend: (Β) control, (Χ) 5 l Dox, (Ε) 10 l Dox, (Ο) 20 l Dox. At all concentrations of Dox the tension curves were significantly higher than the control tension curve; †, P<0.01; ‡, P<0.05.
binding sites involved. Data values are given as mean±... A Student’s t-test was used to test significance of differences at a 0.05 level of significance (P<0.05). The tension curves measured during incubation with anthracyclines were compared with the control tension curve using an analysis of variance (ANOVA) to test for an effect at the 0.05 level of significance.
Results The effect of anthracyclines on maximal tension and Ca2+ sensitivity of trabeculae The trabeculae had a mean diameter (mean of the width and the depth measurements) of 79±4 lm and generated a peak tension of 65±4 kN/m2 (n= 58) under control conditions. Incubation with anthracyclines resulted in a time- and dose-dependent increase of the maximal calcium activated tension of skinned cardiac muscles, as is shown for Dox in Figure 1. This increase in tension was significant for Dox (5 l: P<0.05; 10 lm and 20 l: P<0.01). The resting tension of the trabeculae was not increased upon incubation with anthracyclines. To determine whether the inotropic effect was specific for the anthracycline family, two commonly used
non-anthracycline anti-cancer agents, Taxol and 5-fluoro-uracil (5-FU), were included in the experimental protocol. At a concentration of 20 l no effect on the contractile force was observed. The next step involved comparison of a variety of clinically available anthracyclines for their effect on contractile force. Dox and its stereoisomere epirubicin significantly increased the maximal tension, whereas daunorubicin and idarubicin showed a lesser positive inotropic action (Table 2). In order to investigate whether this positive inotropic effect was the result of an increased calcium sensitivity, partial activation curves were determined before and during anthracyline incubation. Figure 2 shows a typical registration of calcium sensitivity curves before and during incubation with 20 l Dox. The calcium sensitivity of heart preparations was significantly (P<0.05) increased upon 3, 30 and 60 min of incubation with 20 l Dox. At 60 min this effect was maximal and at that point calcium sensitivity was increased by 0.14±0.03 (n=8, P<0.05). Incubation with the other anthracyclines had no effect on calcium sensitivity (Table 2). The Hill coefficient of the calcium sensitivity curves was not significantly altered following anthracycline incubation, indicating that the cooperativity between the calcium binding sites of troponin-C was not changed by treatment with anthracyclines.
Effect of doxorubicin on rigor tension of trabeculae To determine whether the increase in active tension upon anthracycline incubation was the result of a change in the conformation of the strongly bound cross-bridges, Dox was applied to trabeculae which were already in a stable rigor contraction. We found that Dox (20 l) had no effect on the level of maintained tension when applied to the preparation during an established rigor contraction (data not shown). Next, the effect of Dox on the stable rigor tension was investigated after incubation with 20 l Dox for 30 min and subsequently transferring the preparation to rigor solution containing the same Dox concentration. After 30 min incubation in control relaxing solution, the level of maintained rigor tension was diminished as compared to the rigor tension that was measured before the start of the incubation [Fig. 3(a)]. After incubation with 20 l Dox, however, the level of maintained rigor tension was increased as compared to the rigor tension that was measured before the start of the incubation [Fig. 3(b)]. The difference between the decrease in rigor tension in the control
1005
Anthracyclines Enhance Tension Development Table 2 Effect of 60 min incubation with several drugs on maximal calcium activated tension, calcium sensitivity and Hill coefficient of rat trabeculae
Control Doxorubicin Epirubicin Daunorubicin Idarubicin Taxol 5-FU
(n=8) (n=8) (n=5) (n=5) (n=5) (n=6) (n=5)
Tmax
D pCa50
D nHill
102.5±2.8 133.3±7.6† 126.5±7.8† 111.6±1.4 108.9±6.3 99.4±1.0 105.5±4.4
– +0.14±0.03‡ +0.04±0.02 +0.01±0.04 −0.11±0.07 −0.07±0.04 −0.07±0.08
– +0.11±0.17 +0.04±0.11 −0.04±0.03 −0.14±0.14 +0.24±0.08 +0.10±0.11
Shows the effect of incubation with several anthracyclines and other chemotherapeutic agents (concentration: 20 l) on the calcium activated tension at pCa 4 (Tmax), on the calcium sensitivity (D pCa50) and on the Hill coefficient (D nHill). Tmax is expressed as the percentage of the control value before the start of incubation with the drugs. The effect on pCa50 and nHill is expressed as the change in value as a result of incubation with drugs. The average pCa50 value was 6.02±0.03 (n=42), and the average nHill coefficient was 1.73±0.09 (n=42) at the start of the experiments. †, P<0.01; ‡, P<0.05.
Tension (kN/m2)
experiment and the increase in rigor tension in the fibres exposed to Dox was 12% (n=7, P<0.05).
Discussion
0
3
30 60 Time (min)
90
120
Figure 2 Shows calcium sensitivity curves measured before and after several incubation times with 20 l Dox, shown by a registration of a typical experiment. 0: control calcium sensitivity curve. The subsequent curves were measured after 3, 30, 60, 90 and 120 min incubation with 20 l Dox. Partial activations were performed at pCa 9 (relaxation), 6.17, 5.96, 5.69, 5.22 and 4.00. pCa50 value was 5.87 and nHill coefficient was 2.20 for the control curve. Calibration: horizontal 2 min, vertical 10 kN/m2.
(a)
(b)
Figure 3 Shows examples of tracings obtained in two typical rigor experiments. (a) Control experiment; (b) effect of incubation with Dox. Solid line shows the rigor contraction after 30 min of incubation with either control relaxation solution (a) or 20 l Dox (b); the dotted line shows the rigor contraction within the same fibre before the incubation. Calibration: horizontal 30 s, vertical 10 kN/m2.
Anthracycline cardiomyopathy still presents an enormous clinical problem. Patients may be deprived of a very potent anti-cancer drug if they have reached the cumulative dose after which there is a high risk for developing congestive heart failure. The complexity of anthracycline-induced cardiomyopathy is reflected by the many different actions of anthracyclines. The targets of these drugs include the sarcolemma, intracellular membranes such as the SR, and the contractile proteins, the subject of the present study. In this study, a novel mechanism is described. Anthracyclines caused a strong inotropic effect by direct interaction with the actinmyosin system. We used rather thin trabeculae (diam: 79±4 lm, n=58) in order to avoid diffusion limited processes within the preparations. Even in thin muscle preparations, however, diffusion limitations might interfere with processes involving muscle contraction, and they might therefore be of influence for the magnitude of the tension response (Kentish and Jewell, 1984; Stienen et al., 1990). In the control measurements, we found a small increase in tension between the contraction before the solution change and contraction after immersion in fresh superfusion fluid (at t=3 min). We hypothesize that this slight increase in maximal calcium activated tension may be the result of organic phosphate accumulation within the fibre and/or inside the well during the first test contraction (Stienen et al., 1990). Therefore, any drug-related effect on calcium activated tension has to be well beyond this limit.
1006
A. E. Bottone et al.
The observed inotropic effect seems specific for anthracyclines, because the other chemotherapeutic agents tested in this study had no effect on tension development. Among the anthracyclines that were studied, Dox and epirubicin were the most potent in enhancing the maximal tension of heart preparations. We used concentrations which closely match plasma concentrations found in patients during anthracycline chemotherapy. Epirubicin is a stereoisomer of Dox, with an opposite orientation of the hydroxyl group at carbon atom 4′. The other anthracyclines tested in this study, idarubicin and daunorubicin, also had a positive inotropic effect, but they did not significantly increase the maximal tension. Anthracyclines vary in their chemical structure in distinct places inside the molecule (Arcamone, 1985), which might reflect their different ability in augmenting the maximal tension of skinned heart preparations in this experimental set-up. Modifications in the chemical structure of anthracyclines have been used to reduce their cardiotoxic side effects. Epirubicin, daunorubicin, and idarubicin are examples of structural analogues which are now used in the treatment of patients with cancer. These analogues have been compared with Dox for their effects on the heart. Randomized clinical trials have shown that, of the clinically used anthracyclines, Dox is the most cardiotoxic anthracycline and that epirubicin and idarubicin are less cardiotoxic (Jain et al., 1985; Cersosimo, 1992). The position of daunorubicin as a cardiotoxic anthracycline is less clear but may be considered slightly less or equal cardiotoxic as Dox (Gilladoga et al., 1976; Jaenke et al., 1980). Our results are in agreement with the observed rise in systolic and diastolic bloodpressure during the first 24 h in patients treated with Dox (Unverferth et al., 1981; Brown et al., 1989). Chronic anthracycline cardiomyopathy, however, is characterized by a gradual loss in contractile force. One hypothesis for this biphasic action of Dox might be that the positive inotropic effect of Dox eventually results in a disruption of the contractile machinery, thereby leading to an attenuation of tension. Calcium plays, among other pathways, an important role in anthracycline cardiotoxicity (Singal et al., 1987) although application of the calcium entry-blocker verapamil did not prevent the cardiotoxic effect of Dox (Rabkin, 1983; Rabkin et al., 1983; Rabkin and Godin, 1985). We investigated whether the observed increase in contractile force was the result of an increased calcium sensitivity of the trabeculae. Determination of the calcium sensitivity of the preparations at the start of each experiment resulted in an average control pCa50
value of 6.02±0.03 (n=42) and an average Hill coefficient of 1.73±0.09 (n=42). These values are in close range with values found in the literature (Harrison et al., 1988). However, it is important to note that differences in solution parameters like pH, ionic strength, temperature and several concentrations of important ions all influence the shape and the position of the pCa-tension curve (Fabiato and Fabiato, 1978; Kentish, 1984; Harrison and Bers, 1990; Kentish, 1986). The average maximal tension developed in the control experiments was 65±4 kN/m2 (n=58). This value is in close agreement with values of EGTA-treated trabeculae reported by Kentish and Jewell (1984). Although incubation with 20 l Dox gave rise to a small, but significant leftward shift of the calcium sensitivity curve after 3, 30 and 60 min of incubation, it seems unlikely that this limited increase of about 0.10 of calcium sensitivity of the contractile system can explain for the large inotropic effect observed after incubation with Dox. Since we used preparations in which all membrane structures were eliminated, it is more likely that the inotropic effect of Dox is mediated by direct interaction with the force generating filaments, i.e. actin and/or myosin, rather than being mediated by alteration of the calcium sensitivity of the troponin/tropomyosin complex on the thin filament. The inotropic effect of Dox may be explained by several mechanisms: an increase of the tension per cross-bridge, an altered distribution of the crossbridges, or an altered distribution of the length of the cross-bridges in the force generating state. The direct effect of Dox on the quality of the crossbridges was investigated in muscles that were forced into the rigor state. Because even a very small length change (> 0.1%) of a trabecula during an isometric rigor contraction will affect the force development during a rigor contraction (De Winkel et al., 1995), we focused on the stable tension level of an established rigor contraction. The results of the rigor experiments showed that Dox had no effect on the level of tension of an established rigor contraction. During normal activation, the crossbridges cycle through bound and unbound states leading to an equilibrium giving a stable tension. In the absence of ATP, however, myosin binds actin with high affinity forming rigor cross-bridges. Because of the absence of ATP this leads to a situation in which cross-bridges are arrested in a strongly bound stage and in which cycling of crossbridges no longer occurs. The inability of Dox to increase stable rigor force implies that the stiffness of the already equilibrated strongly bound rigor bridges was not affected. If the cardiac muscles
Anthracyclines Enhance Tension Development
were incubated with 20 l Dox in relaxation solution, and subsequently transferred to rigor solution to which the same Dox concentration was added, the level of maintained rigor tension was significantly increased by 12% as compared with the effect of control incubation on the level of rigor force. The results of the rigor experiments indicate that Dox acts during the transition to force-generating cross-bridges rather than on the end product, i.e. that the cross-bridges need to be actively cycling for Dox to potentiate force. One explanation might be that Dox alters the coupling and/or the decoupling of cross-bridges and thus the distribution of cross-bridges during the activation cycle. During a rigor contraction, the level of tension is determined by the ratio of “strained” and “unstrained” crossbridges (Kawai and Brandt, 1976). If the coupling and/or decoupling of cross-bridges is affected by Dox, the ratio between “strained” and “unstrained” cross-bridges will be altered. In our experiments this may have lead to a higher mean tension development by the cross-bridges in rigor and might explain for the increase in tension upon incubation with Dox. However, more study is necessary to further elucidate the mechanism of this positive inotropic action. In summary, insight into the mechanisms underlying anthracycline-induced cardiotoxicity is essential in developing strategies to prevent the development of a disabling cardiomyopathy. We showed that Dox and other anthracyclines exert a positive inotropic action in skinned heart muscles by direct interaction with the contractile system. This effect may contribute to the observed cardiotoxic effect after Dox treatment and warrants further investigation.
References A F, 1985. Properties of antitumour anthracyclines and new developments in their application: Cain memorial award lecture. Cancer Res 45: 5995–5999. A J, S N, H H, T T, V S, O K, A N, S A, H H, I T, M Y, Y Y, 1981. Adriamycin cardiotoxicity: possible pathogenic mechanisms. J Mol Cell Cardiol 13: 381–397. B KA, B AJ, W RM, S JA, 1989. Acute effects of doxorubicin on human left ventricular systolic and diastolic function. Am Heart J 118: 979– 982. C RJ, 1992. Idarubicin: an anthracycline antineoplastic agent. Clin Pharm 11: 152–167. D B EL, F H, V EE, V H BGV, S P, 1992. Doxorubicin interacts directly with
1007
skinned single skeletal muscle fibers. Eur J Pharmacol 214: 97–100. D W ME, B T, T BW, 1995. Viscoelastic properties of cross bridges in cardiac muscle. Am J Physiol 268: H987–H998. D JH, 1983. Effect of anthracycline antibiotics on oxygen radical formation in rat heart. Cancer Res 43: 460–472. D JH, 1991. Doxorubicin-induced cardiac toxicity. N Engl J Med 324: 843–845. D JH, T C, S JE, 1985. Ultrastructural features of Adriamycin-induced skeletal and cardiac muscle toxicity. Am J Pathol 118: 288–297. F A, F F, 1978. Effects of pH on myofibrils and sarcoplasmic reticulum of skinned cells from cardiac and skeletal muscles. J Physiol (Lond) 276: 233– 256. G AC, M C, T CT, W N, S SS, M ML, 1976. The cardiotoxicity of adriamycin and daunomycin in children. Cancer 37: 1070–1078. H SM, L C, M DJ, 1988. Hysteresis and the length dependence of calcium sensitivity in chemically skinned rat cardiac muscle. J Physiol (Lond) 401: 115–143. H SM, B DM, 1990. Temperature dependence of myofibrillar Ca2+-sensitivity of rat, guinea pig and frog ventricular muscle. Am J Physiol 258: C241–C281. H B, B HD, 1981. Acute negative inotropic effect of adriamycin (doxorubicin). Naunyn-Schmied Arch Pharmacol 317: 252–256. H SRM, W AJ, 1990. Patterns of interaction between anthraquinone drugs and the adriamycin-release channel from cardiac sarcoplasmic reticulum. Circ Res 67: 272–283. J RS, D-DC D, T A, 1980. Cardiotoxicity and comparative pharmacokinetics of six anthracyclines in the rabbit. Cancer Res 40: 3530– 3536. J KK, C ES, G NL, H TB, K RJ, C V, S W, C C, P GR, Y CW, W RE, 1985. A prospective randomized comparison of epirubicin and doxorubicin in patients with advanced breast cancer. J Clin Oncol 3: 818–826. K M, B PW, 1976. Two rigor states in skinned crayfish single muscle fibres. J Gen Physiol 68: 267– 280. K JC, 1984. The inhibitory effects of monovalent ions on force development in detergent-skinned ventricular muscle from guinea pig. J Physiol (Lond) 352: 353–374. K JC, 1986. The effects of inorganic phosphate and creatine phosphate on force production in skinned muscles from rat ventricle. J Physiol (Lond) 370: 585– 604. K JC, J BR, 1984. Some characteristics of Ca2+-regulated force production in EGTA-treated muscle from rat heart. J Gen Physiol 84: 83–99. K DH, A T, B TM, 1980. Inotropic actions of doxorubicin in isolated guinea pig atria. Evidence for lack of involvement of Na+, K+-adenosine triphosphatase. J Pharmacol Exp Ther 214: 368–374. M LA, H G, I F, B EM, A NR, 1989. Protection of human left ventricular myocardium from cutting injury with 2,3 butanedione monoxime. Circ Res 65: 1441–1444.
1008
A. E. Bottone et al.
O RD, M PS, B DE, F S, C BJ, C BK, B RJ J, 1988. Doxorubicin cardiotoxicity may be caused by its metabolite, doxorubicinol. Proc Natl Acad Sci USA 85: 3585–3589. P IN, D EL, S MJ, Z I, 1990. Anthraquinone-sensitized Ca2+ release channel from rat cardiac sarcoplasmic reticulum: possible receptormediated mechanism of doxorubicin cardiomyopathy. Mol Pharmacol 37: 503–514. P PM, R RP, P JEB, 1985. Acute effects of doxorubicin on chronotropic and inotropic mechanisms in guinea pig atria. Cancer Treat Rep 69: 859– 865. R SW, 1983. Interaction of external calcium concentration and verapamil on the effects of doxorubicin (adriamycin) in the isolated heart preparation. J Cardiovasc Pharmacol 5: 848–855. R SW, O M, P PI, 1983. Increased mortality with cardiotoxic doses of Adriamycin after verapamil pretreatment despite prevention of myocardial calcium accumulation. Can J Physiol Pharmacol 61: 1050–1056. R SW, G DV, 1985. Adriamycin cardiotoxicity and calcium entry blockers: the need for caution in the combination. Can J Cardiol 1: 4–7. S PK, P GN, 1986. Adriamycin stimulates low-affinity Ca2+ binding and lipid peroxidation but
depresses myocardial function. Am J Physiol 250: H419–H425. S PK, D CMR, W LE, 1987. Subcellular effects of adriamycin in the heart: a concise review. J Mol Cell Cardiol 19: 817–828. S GJM, R MCM, W MGA, E G, 1990. Depression of force by phosphate in skinned skeletal muscle fibres of the frog. Am J Physiol 259: C349–C357. T K, A T, C A, O M, Y M, K H, 1992. Doxorubicin: an antagonist of muscarinic receptors in guinea pig heart. Eur J Pharmacol 220: 63–69. U DV, M RD, U BP, T RL, B SP, B N, 1981. Human myocardial morphological and functional changes in the first 24 h after doxorubicin administration. Canc Treat Rep 65: 1093–1097. V B CJ, O RD, B RC, O JA, 1978. Doxorubicin: inotropic effects and inhibitory action of ouabain. J Pharmacol Exp Ther 207: 277–283. V EE, A S, V WJF, A BS, B A, 1994. Comparison of different iron chelators as protective agents against doxorubicin-induced cardiotoxicity. J Mol Cell Cardiol 26: 1179–1185. Y RC, O RF, M CE, 1981. The anthracycline antineoplastic drugs. N Engl J Med 305: 139–153.