OUP user menu

Cardiotonic drugs differentially alter cytosolic [Ca2+] to left ventricular relationships before and after ischemia in isolated guinea pig hearts

Qun Chen, Amadou K.S Camara, Samhita S Rhodes, Matthias L Riess, Enis Novalija, David F Stowe
DOI: http://dx.doi.org/10.1016/S0008-6363(03)00524-8 912-925 First published online: 1 October 2003

Abstract

Objective: Cardiotonic agents may differentially alter indices of the cytosolic [Ca2+]/left ventricular pressure (LVP) relationship when given before and after ischemia. We measured and calculated systolic–diastolic [Ca2+], systolic–diastolic LVP, velocity ratios (VRs) d[Ca2+]/dtmax to dLVP/dtmax (VRmax), d[Ca2+]/dtmin to dLVP/dtmin (VRmin), and area ratio (AR, area Ca2+]/area LVP per beat) before and after 30 min global ischemia in guinea pig hearts. Methods: Hearts were perfused with levosimendan, dobutamine, dopamine, or digoxin. Ca2+ transients were recorded by indo-1 fluorescence via a fiber optic probe placed on the LV free wall. [Ca2+]/LVP loops were acquired by plotting LVP time as a function of [Ca2+] at multiple time points during the cardiac cycle. Results: Ischemia reperfusion increased [Ca2+] and decreased contractility and relaxation and produced a flatter and broader [Ca2+]/LVP loop. All drugs shifted the [Ca2+]/LVP loop rightward and upward when given before and after ischemia. Dobutamine increased [Ca2+] and contractility more than other drugs. Digoxin increased [Ca2+] the least but increased contractility similar to dopamine and levosimendan. Before ischemia dopamine and digoxin both decreased VRmax and VRmin, whereas dobutamine increased VRmin, but not VRmax, and levosimendan had no effect on VR. VRmax and VRmin were markedly elevated after ischemia, but again decreased with dopamine and digoxin; dobutamine again increased VRmin, but not VRmax, and levosimendan decreased both VRmax and VRmin. Before ischemia dopamine and digoxin both decreased AR, dobutamine increased AR, and levosimendan had no effect; after ischemia AR was markedly elevated but dopamine and digoxin decreased AR, dobutamine increased AR, and levosimendan decreased AR. Conclusion: Although each drug enhanced contractility and relaxation both before and after ischemia by increasing cytosolic [Ca2+] and Ca2+ flux, dopamine and digoxin improved, and dobutamine worsened responsiveness to Ca2+, i.e., velocity ratio and area ratio, whereas levosimendan had no net effect before ischemia but improved responsiveness after ischemia.

Keywords
  • Calcium-pressure loops
  • Cardiotonic drugs
  • Digoxin
  • Dopamine
  • Dobutamine
  • Guinea pig

Time for primary review 21 days.

☆ Portions of this work have appeared in abstract form Biophys J 2002;82:66A, 653A and 654A; FASEB J 2002;16:A857; Anesthesiology 2001;95:A622.

1 Introduction

Positive inotropic agents such as catecholamines are frequently used to treat acute heart failure or to facilitate contractility and relaxation in stunned hearts after cardiac surgery. Most cardiotonic agents in current use, including cardiac glycosides, catecholamines and phosphodiesterase (PDE) inhibitors, ultimately act to enhance contractility and relaxation by increasing the cyclic change in cytosolic [Ca2+] during the cardiac cycle [1–3]. However if cytosolic [Ca2+] exceeds an optimum there is no further increase in contractility and diastolic tonus and dysrhythmias can occur [4]. In contrast to these commonly used cardiotonic agents, so-called calcium sensitizers, such as levosimendan [1,4,5], are believed to increase contractility in part by improving responsiveness of the contractile apparatus to Ca2+. Specifically, levosimendan is thought to bind to troponin C in a Ca2+-dependent manner to stabilize the Ca–troponin C bond and enhance contractility at a given [Ca2+] [6,7]. Levosimendan also opens myocyte KATP channels [8], which may lead to a cardioprotective effect, and may increase cAMP and myocyte [Ca2+] as an inotropic mechanism by selectively inhibiting PDE III [9].

Contractile responsiveness to Ca2+ can be altered by pharmacologic and pathologic conditions. Hypothermia [10], β-adrenergic agonists [11], PDE inhibitors [3], and ischemia reperfusion injury [12–15] each decrease contractile responsiveness to Ca2+. Digoxin, a sarcolemmal Na+, K+-ATPase inhibitor, dobutamine, a β1-receptor activator [16,17], dopamine, a β1 and dopamine receptor activator [18], and levosimendan [19,20] are variably effective for treating acute and chronic heart failure and stunned myocardium [3,18,21]. However, if one of these drugs increases [Ca2+] above a threshold, and especially if [Ca2+] is already elevated due to myocyte injury, cardiac function will most likely be impaired. Because the selected drugs are known to enhance contractility by different mechanisms, we postulated that they differentially alter the cyclic [Ca2+]/left ventricular pressure (LVP) relationship and have different effects on this relationship after ischemia reperfusion injury. To test this we simultaneously measured cytosolic [Ca2+] and LVP in guinea pig isolated hearts and constructed several indices of this relationship to explore and compare differences.

2 Methods

2.1 Isolated heart preparation and measurements

The investigation conformed to the Guide for the Care and Use of Laboratory Animals (National Institutes of Health No. 85-23, revised 1996). Prior approval was obtained from the Medical College of Wisconsin Animal Studies Committee. Our preparation and measurements have been described in detail [10,14,15,22–27]. In brief, hearts were isolated from anesthetized guinea pigs (300–350 g) and perfused with crystalloid solution by the Langendorff method. Hearts were perfused at 55 mmHg and 37°C with a modified Krebs–Ringer (KR) solution equilibrated with 97% O2, 3% CO2 and containing, in mmol/l, Na+ 137, K+ 4.5, Mg2+ 2.4, Ca2+ 1.25, Cl 134, HCO3 15.5, H2PO4 1.2, glucose 11.5, pyruvate 2, mannitol 16, EDTA (ethylenediaminetetraacetic acid) 0.05, probenecid 0.1, and insulin 5 units/l. Extracellular [Ca2+] was half-normal so that positive inotropic drugs had a better range of contractile responses starting at a lower LVP.

Phasic LVP was measured with a transducer connected to a thin, saline-filled latex balloon inserted into the left ventricle through the mitral valve from a cut in the left atrium. Developed LVP was defined as systolic LVP minus diastolic LVP. The first derivative of LVP, dLVP/dt, was derived on line and maximum and minimum values determined. Balloon volume was adjusted to maintain a diastolic LVP of zero mmHg during the initial control period so that any increase in diastolic LVP indicated an increase in left ventricle (LV) wall stiffness or diastolic contracture. Two pairs of bipolar electrodes were placed in each heart to monitor intracardiac electrograms from which spontaneous atrial heart rate (HR) was determined from the right atrial beat-to-beat interval.

Coronary inflow was measured at constant temperature (37.1 °C) with a self-calibrating in-line, ultrasonic flow meter. CF and coronary effluent Na+, K+, Ca2+, PO2, PCO2, and pH were measured off-line with an intermittently self-calibrating analyzer system (Radiometer Copenhagen ABL 505, Copenhagen, Denmark). Coronary sinus effluent was collected through a cannula inserted into the right ventricle through the pulmonary artery after ligating the venae cavae. Coronary outflow (coronary sinus) O2 tension was also measured continuously on-line with an O2 Clark-type electrode placed in the pulmonary artery. Because myocardial metabolism is altered by cardiotonic drugs and ischemia reperfusion [18,21,23], we also measured myocardial O2 consumption (MVO2) and cardiac efficiency. MVO2 was calculated as (coronary flow/g)×(arterial PO2−venous PO2)×24 μl O2/ml at 760 mmHg; cardiac efficiency was calculated as developed LVP×HR/MVO2.

2.2 Measurement of cytosolic and non-cytosolic free Ca2+ in isolated hearts

We have described details of our method to monitor and calibrate indo-1 fluorescence signals as a measure of cytosolic [Ca2+] in the left ventricle of isolated hearts [10,14,15,22,23,25–27]. Experiments were carried out in a light-blocking Faraday cage. Briefly, the heart was partially immobilized by hanging it from the aortic cannula, the pulmonary artery catheter, and the left ventricular balloon catheter. The heart was immersed continuously in the bath at 37°C. The distal end of a trifurcated silica fiberoptic cable was placed gently against the LV epicardial surface through a hole in the bath to excite the tissue with light filtered at 350 nm and to record emitted light filtered at 385 and 456 nm. A rubber O ring was placed over the fiberoptic tip to seal the hole and netting was applied around the heart for optimal contact with the fiber optic tip. Background auto fluorescence was determined for each heart after initial perfusion and equilibration at 37°C.

Each heart was then loaded with indo-1 AM for 20–30 min with the re-circulated KR solution at a final indo-1 AM concentration of 6 μM. Residual interstitial indo-1 AM was washed out by perfusing the heart with standard perfusate for at least another 20 min. Additional experiments (three hearts for each of the five groups) were undertaken to assess changes in tissue autofluorescence due to changes in the redox state (primarily a measure of NADH) and drug autofluorescence. None of the drugs exhibited a significant change in autofluorescence; ischemia and reperfusion as noted previously, caused an increase and decrease in NADH as we have reported previously [28], these values were subtracted from the basal fluorescent signal obtained with indo-1.

The fluorescence emissions at 385 and 456 nm (F385 and F456) were recorded using a modified luminescence spectrophotometer (SLM Aminco-Bowman II, Spectronic Instruments, Urbana, IL, USA). The LV region of the heart was excited with light filtered at 350 nm through the in-going fibers of the optic bundle. The arc lamp shutter was opened only for 2.5 s recording intervals to prevent photobleaching. Emission fluorescence was collected by fibers of the remaining two limbs of the cable and filtered by square interference filters at 385 nm and 456 nm. The F385/F456 ratio remains stable during the 3-h course of these studies indicating no change in effective measured [Ca2+]. After indo-1 loading developed LVP was not significantly altered in non-ischemic hearts over this time period.

2.3 Protocol

Forty hearts were divided randomly and equally among five groups: non-treated ischemia controls, dobutamine, dopamine, digoxin and levosimendan groups. Each experiment lasted 210 min (Fig. 1). Initial background (before indo-1 loading) measurements were obtained after 30 min of stabilization. After loading and residual washout of indo-1, on-line recordings were sampled and stored every 1 to 5 min during drug perfusion, normothermic ischemia and reperfusion. The 30-min period of ischemia was effected by clamping the coronary inflow cannula. Drugs were perfused at the approximate ED50 concentrations for LVP established from pilot studies. Dobutamine (4 μM), dopamine (8 μM), digoxin (1 μM) or levosimendan (1 μM) was perfused for 2 min 30 min before ischemia (at 80 min) in each of the drug groups and again for 2 min at the same concentration beginning at 30 min reperfusion (at 170 min). Each drug produced a steady-state and sub-maximal increases in LVP beginning by at least the first minute of perfusion and lasting over the second minute. All measured variables returned statistically to control values between drug infusions. LVP and dLVP/dt, coronary flow, and coronary sinus oxygen tension (PO2) were recorded continuously. All analog signals were digitized (PowerLab®/8 SP; ADInstruments, Castle, Hills, Australia) and recorded at 125 Hz (Chart & Scope v3.63, ADInstruments, Castle Hills, Australia) on Power Macintosh® G4 computers (Apple, Cupertino, CA, USA) for later analysis using MATLAB® (Mathworks, Natick, MA, USA) and Microsoft Excel® (Microsoft, Redmond, WA, USA) software.

Fig. 1

Experimental protocol. Hearts were perfused and stabilized with Krebs–Ringer’s (KR) solution for 30 min, loaded with indo-1 AM for 30 min, and residual indo-1 AM washed out for 20 min. In treated groups, 8 μM dopamine, 4 μM dobutamine, 1 μM levosimendan, or 1 μM digoxin was perfused for 2 min before ischemia (time 80 min) and at 30 min reperfusion (time 170 min). CON hearts were perfused only with KR at the same time points. Hearts were subjected to 30 min global ischemia at 37°C (time 110 to 140 min) and reperfused for 60 min (time 140 to 200 min). MnCl2 (50 mmol/l) was perfused at the end (time 200–210 min) to quench non-cytosolic Ca2+.

We obtained simultaneous [Ca2+] and LVP recordings at designated time points. Exposure at the 350 nm excitation wavelength light was 55 s. Customized software was developed in MATLAB® for off-line signal processing of recorded data. LVP and fluorescence data were digitally low-pass filtered using a fourth-order bi-directional Butterworth filter at 25 Hz. Data were analyzed for peak systolic, peak diastolic, and systolic–diastolic LVP (mmHg) and [Ca2+] (nM). First derivatives of [Ca2+] (d[Ca2+]/dt) and LVP (dLVP/dt) were derived on-line and values for d[Ca2+]/dtmax and dLVP/dtmax (contractility and peak rate of Ca2+ influx), as well as d[Ca2+]/dtmin and dLVP/dtmin (relaxation and peak rate of Ca2+ efflux) were determined. Area [Ca2+] and area LVP (systolic–diastolic time integral), i.e., total LV pressure (potential work) and total cytosolic [Ca2+] averaged during one beat were computed.

The index of d[Ca2+]/dtmax to dLVP/dtmax (velocity ratio, VRmax), was used to assess the fastest cytosolic Ca2+ influx to generate maximal contractility, and the index of d[Ca2+]/dtmin to dLVP/dtmin (velocity ratio, VRmin), was used to assess the fastest cytosolic Ca2+ efflux that allowed maximal relaxation. The index of area [Ca2+]/area LVP (area ratio, AR) was used to assess the net amount of cytosolic free Ca2+ moved in and out to generate the cardiac (potential) work over one beat. Concentration–response curves were not obtained in this study, so direct comparisons among drugs for a given variable were not considered valid. However, the velocity and area ratios were utilized to compare responses to drugs because the ratios normalized the individual values for Ca2+ and LVP for each drug. Ratios were calculated from raw rather than from grouped data.

2.4 Statistical analysis

All data were expressed as mean±standard error of mean (S.E.M.). One-way analysis of variance for repeated measures (Super Anova 1.11® software for Macintosh® from Abacus Concepts, Berkeley, CA, USA) was used to assess within group differences over time at selected time points: 80 min (baseline) versus peak response of drugs given before ischemia, reperfusion 2 min (142 min), 30 min (170 min), and peak response of drugs given after 30 min reperfusion (Fig. 1). Two-way analysis of variance was used to assess among group differences at baseline, 30 min reperfusion and the peak response of drug given before and after ischemia. If F-values for the analysis of variance were significant, Tukey’s multiple-comparison post-hoc tests was used to differentiate within or among group differences. Differences among means were considered significant when P<0.05.

3 Results

Table 1 summarizes drug-induced changes in heart rate, cytosolic [Ca2+], and mechanical function before and after ischemia. Table 2 summarizes drug-induced changes in coronary flow, O2 consumption and cardiac efficiency before and after ischemia. Each drug increased heart rate, systolic [Ca2+], diastolic [Ca2+] and systolic–diastolic [Ca2+] compared to baseline and 30 min reperfusion except for digoxin, which did not increase systolic or systolic–diastolic [Ca2+] after ischemia. There were few significant differences in the relative changes from controls (30 min reperfusion and baseline) induced by drugs after ischemia than before ischemia.

View this table:
Table 2

Cardiac effects of different positive inotropic agents, given before ischemia (BI) and after ischemia (AI), on metabolic indices in isolated guinea pig hearts

GroupBaselineDrug BIRP 2 minRP 30 minDrug AI
(units)(% increase)(units)(units)(% increase)
Coronary flow (ml/min/g)
CON9.3±0.3−3±27.6±0.7*5.7±0.4*−6±2
DA10.5±0.617±7†7.5±0.7*6.9±0.5*13±4‡
DB10.4±0.918±6†8.9±0.6*5.8±0.4*49±5‡§
LV10.9±0.611±6†7.9±0.6*5.6±0.4*14±3‡
DG10.5±0.6−4±47.7±0.5*5.8±0.4*1±5
MVO2 (0.1 μl min−1 g−1)
CON116±75±594±1470±9*−5±5
DA132±1015±8†112±1299±9*19±6‡
DB110±1049±12†109±1676±6*46±4‡
LV114±641±10†93±861±6*36±10‡
DG120±1413±11102±772±7*13±11
Cardiac efficiency (mmHg beat 0.1 μl−1 g−1)
CON51±413±613±8*46±619±11
DA47±620±6†23±7*40±444±13‡
DB59±9147±26†30±1*47±199±16‡
LV47±246±18†18±7*48±726±7‡
DG54±048±10†17±7*51±839±14‡
  • MVO2, Myocardial O2 consumption; cardiac efficiency=LVDP×HR/MVO2; temperature was 37°C. Values are means±SE.

    * P<0.05 vs. baseline within each group; † P<0.05, drug BI vs. baseline; ‡ P<0.05, drug AI vs. RP 30 min; § P<0.05, drug AI vs. drug BI. See Table 1 for other details.

View this table:
Table 1

Cardiac effects of different positive inotropic agents, given before ischemia (BI) and after ischemia (AI), on cytosolic [Ca2+] and mechanical indices in isolated guinea pig hearts

GroupBaselineDrug BIRP 2 minRP 30 minDrug AI
(units)(% increase)(units)(units)(% increase)
Heart rate (beats min−1)
CON220±52±1148±24*223±71±1
DA211±623±6†155±17*216±617±3‡
DB213±648±2†152±14*206±650±5‡
LV209±613±2†157±13*219±89±5‡
DG210±511±2†142±18*217±311±7‡
Systolic [Ca2+] (nM)
CON242±207±4646±35*260±29−5±5
DA226±2425±5†526±46*242±249±3‡
DB216±6134±13†562±26*240±20128±22‡
LV238±2248±12†576±28*256±2239±16‡
DG232±1418±6†610±35*242±186±9
Systolic LVP (mmHg)
CON45±25±231±5*29±2*0±3
DA49±330±6†35±4*35±2*11±2‡§
DB47±395±16†35±3*29±2*44±7‡§
LV48±241±7†31±4*28±1*32±5‡§
DG51±628±2†30±2*32±3*19±2‡§
Diastolic [Ca2+] (nM)
CON124±44±5208±14*100±12−5±4
DA120±88±2†156±13*94±122±3
DB112±234±6†164±16*90±1042±9*
LV112±67±2†172±22*100±49±12
DG122±83±3182±20*92±123±10
Diastolic LVP (mmHg)
CON9±1−9±716±2*15±2*−11±8‡
DA8±1−20±7†21±3*17±2*−14±3‡
DB9±1−25±6†22±4*18±2*−26±4‡
LV9±2−23±7†21±2*18±2*−13±2‡
DG8±1−23±7†19±1*17±2*−13±5‡§
Systolic–diastolic [Ca2+] (nM)
CON130±189±4378±29*262±28*2±7
DA118±1244±9†306±28*154±12*14±4‡§
DB104±4245±35†356±28*152±16*179±30‡
LV126±1867±15†314±27*170±12*52±19‡
DG110±834±10†378±29*150±10*9±13
Systolic–diastolic LVP (mmHg)
CON36±1−4±215±5*16±1*3±4
DA39±361±12†14±3*18±2*35±9‡
DB37±3157±27†13±2*17±2*143±8‡
LV36±284±19†10±3*11±3*71±13‡
DG40±656±6†14±2*15±3*70±10‡
  • Values are means±S.E. for baseline, RP 2 min and RP 30 min; Values (in bold) are percent change (means±S.E.) for drug BI vs. baseline and drug AI vs. RP 30 min; n=8 for each group; baseline, perfusion 80 min; RP, reperfusion; drug BI is drug before ischemia; drug AI is drug given drug after ischemia; CON, control; DA, dopamine; DB, dobutamine; LV, levosimendan; DG, digoxin.

    * P<0.05 vs. baseline within each group; † P<0.05, drug BI vs. baseline; ‡ P<0.05, drug AI vs. RP 30 min; § P<0.05, drug AI vs. drug BI.

Each drug increased systolic LVP and systolic–diastolic LVP, and decreased diastolic LVP compared to baseline and 30 min reperfusion, but the relative changes from their 30 min reperfusion or baseline values were comparable or smaller for systolic LVP when drugs were given after ischemia than before ischemia. Each drug increased coronary flow before and after ischemia expect for digoxin which did not alter flow; dobutamine increased coronary flow more after ischemia than before ischemia. Each drug, except for digoxin, increased MVO2. All drugs increased cardiac efficiency compared to baseline and 30 min reperfusion, respectively; this increase was greater by dobutamine largely because of the higher heart rate. Averaged inflow (arterial) pH was 7.45±0.01, pO2 was 695±12 mmHg, and pCO2 was 20±2 mmHg. Control venous (v) pHv, pO2v, and pCO2v values were 7.36±0.01, 250±16, and 30±2, respectively. All drugs decreased pHv and pO2v, and increased pO2v, 30 min before and after ischemia but the effect was greatest (P<0.5) for dobutamine before (7.22±0.02, 149±12, 34±2) and after (7.25±0.07, 112±18, 36±4) ischemia. There were no appreciable differences among the drugs for venous [Na+], [K+], or [Ca2+].

Fig. 2 displays average [Ca2+]/LVP loops before 30 min ischemia (80 min) and at 30 min reperfusion in absence and presence of the four positive inotropic drugs. At 30 min reperfusion, the cardiac loop was lower, smaller and shifted slightly rightward than before ischemia. Before ischemia each drug increased loop size and shifted the loop increasingly rightward in the order digoxin, dopamine, levosimendan, dobutamine; dobutamine increased loop height (systolic LVP) more than other drugs. After ischemia loop height was increased by each drug to approximately the level observed before ischemia without drugs. Fig. 2A shows that after ischemia dopamine continued to shift the cardiac loop rightward but it was smaller than before ischemia. Fig. 2B shows that dobutamine (DB) after ischemia, as before, shifted the cardiac loop far rightward compared to drug-free controls but that the loop was much smaller (flatter) after ischemia. Fig. 2C shows that levosimendan (LV) after ischemia, as before, shifted the cardiac loop rightward but that it was smaller after ischemia. Fig. 2D shows that digoxin (DG) after ischemia shifted the cardiac loop only slightly rightward compared to drug-free baseline and reperfusion controls.

Fig. 2

Average cytosolic [Ca2+]/LVP loops (6–10 cardiac cycles per heart; eight hearts per group). Loop shape and position acquired by instantaneous plot of LVP and [Ca2+] coordinates over one averaged beat. Loop area acquired by integration of [Ca2+]/LVP over one beat. Reperfusion after ischemia shifted the cardiac loop slightly rightward and downward. Dopamine (panel A) and especially dobutamine (panel B) given before and after ischemia shifted the cardiac loop rightward and upward. Levosimendan (panel C) and digoxin (panel D) given before and after ischemia shifted the cardiac loop rightward and upward, the rightward more so by levosimendan.

Fig. 3 displays maximal time derivatives of Ca2+ and LVP, d[Ca2+]/dtmax (A) and dLVP/dtmax (B), at baseline before ischemia and during peak responses to drugs before ischemia, at 2 and 30 min reperfusion, and during peak responses of drugs after 30 min reperfusion. Before ischemia dLVP/dtmax increased with each drug while d[Ca2+]/dtmax increased with each drug except digoxin. Part of this effect was due to increased heart rate by these drugs, especially dobutamine. After 2 min reperfusion dLVP/dtmax decreased markedly in each group compared to baseline values, while d[Ca2+]/dtmax was higher than baseline values; by 30 min reperfusion dLVP/dtmax was equivalent to baseline values, while d[Ca2+]/dtmax remained higher than baseline values. Each drug given after ischemia increased d[Ca2+]/dtmax and dLVP/dtmax compared to 30 min reperfusion, but increases in dLVP/dtmax were much smaller than before ischemia. Fig. 4 displays d[Ca2+]/dtmin (A) and dLVP/dtmin (B) at the same time points as Fig. 3. These changes were inverse but qualitatively similar to findings of d[Ca2+]/dtmax and dLVP/dtmax, except that at 2 min reperfusion d[Ca2+]/dtmin was about four times greater than d[Ca2+]/dtmax, i.e., velocity of Ca2+ efflux was slower than velocity of Ca2+ influx.

Fig. 4

d[Ca2+]/dtmin (A) and dLVP/dtmin (B) at baseline, during peak response of drugs before ischemia, at 2 and 30 min reperfusion, and during peak response of drugs after ischemia. Reperfusion increased d[Ca2+]/dtmin before and after ischemia, but increased dLVP/dtmin less after ischemia than before.

Fig. 3

d[Ca2+]/dtmax (A) and dLVP/dtmax (B) at baseline, during peak response of drugs before ischemia, at 2 and 30 min reperfusion, and during peak response of drugs after ischemia. Reperfusion increased d[Ca2+]/dtmax but decreased dLVP/dtmax in each group. Each drug significantly increased d[Ca2+]/dtmax before and after ischemia, but increased dLVP/dtmax less after ischemia than before.

Figs. 5–7 display the ratios of d[Ca2+]/dtmax to dLVP/dtmax (VRmax), d[Ca2+]/dtmin to dLVP/dtmin (VRmin), and area [Ca2+]/area LVP (AR) at baseline, during peak response of drugs given before ischemia, at 30 min reperfusion, and during peak response of drugs given after 30 min reperfusion. These ratios allow a qualitative comparison of the relative effects of each drug before and after ischemia. Fig. 5 shows that dopamine and digoxin decreased VRmax, whereas dobutamine and levosimendan did not affect the VRmax before ischemia compared to baseline values; this indicates improved contractile responsiveness to Ca2+ influx to dopamine and digoxin. VRmax at 30 min reperfusion was much higher than before ischemia in each group, indicating a worsened contractile responsiveness to Ca2+ influx. However, prior treatment with each of these drugs resulted in a better return in contractile responsiveness compared to the control group. As before ischemia, dopamine and digoxin, and in addition levosimendan, each decreased VRmax after 30 min reperfusion but these values were higher than before ischemia. This indicates improved but substandard contractile responsiveness by these drugs after ischemia. Dobutamine did not change VRmax before and after ischemia compared to the baseline and 30 reperfusion controls; this indicates no change in contractile responsiveness.

Fig. 7

Ratio of area Ca2+ to area LVP (AR) at baseline, during peak response of drugs before ischemia, at 30 min reperfusion, and during peak response of drugs after ischemia. Reperfusion alone increased AR in each group. Dopamine and digoxin decreased AR when given before and after ischemia, whereas dobutamine increased AR and levosimendan did not alter AR before ischemia; levosimendan decreased AR only when given after ischemia. The relative change in AR was greater when drugs were given after ischemia than before. Dobutamine induced the highest increase in AR; this increase was similar to that of the control group.

Fig. 6

Ratio of d[Ca2+]/dtmin to dLVP/dtmin (VRmin) at baseline, during peak response of drugs before ischemia, at 30 min reperfusion, and during peak response of drugs after ischemia. Reperfusion alone increased VRmin in each group. Dopamine and digoxin decreased VRmin when given before and after ischemia, whereas dobutamine and levosimendan did not alter VRmin before ischemia; levosimendan decreased VRmin only when given after ischemia. The relative change in VRmin was greater when drugs were given after ischemia than before. Dobutamine induced the highest increase in VRmin.

Fig. 5

Ratio of d[Ca2+]/dtmax to dLVP/dtmax (VRmax) at baseline, during peak response of drugs before ischemia, at 30 min reperfusion, and during peak response of drugs after ischemia. Reperfusion alone increased VRmax in each group. Dopamine and digoxin decreased VRmax when given before and after ischemia, whereas dobutamine and levosimendan did not alter VRmax before ischemia; levosimendan decreased VRmax only when given after ischemia. The relative change in VRmax was greater when drugs were given after ischemia than before. Dobutamine induced the highest increase in VRmax.

Changes in VRmin (Fig. 6) were qualitatively similar to findings of Fig. 5 except that VRmin was relatively greater than VRmax after ischemia and that dobutamine increased VRmin before and after ischemia. In general, this indicates worsened relaxation responsiveness to Ca2+ efflux by dobutamine, improved relaxation responsiveness to dopamine and digoxin before ischemia, and to dopamine, digoxin and levosimendan after ischemia. Fig. 7 shows decreased AR by dopamine and digoxin before ischemia and by dopamine, digoxin and levosimendan after ischemia. Dobutamine increased this ratio before and after ischemia compared to its baseline and 30 min reperfusion controls. This indicates the net amount of Ca2+ moved in and out of the myoplasm to produce (potential) contractile work during a cardiac cycle is improved by dopamine and digoxin before ischemia, and by these drugs and levosimendan after ischemia, and worsened by dobutamine before and after ischemia.

4 Discussion

This study shows that the cardiotonic drugs dopamine, digoxin, levosimendan and dobutamine shift the [Ca2+]/LVP loop rightward and upward and increase loop area; differences among loops were greatest with dobutamine at the single concentrations selected. After ischemia and 30 min reperfusion the [Ca2+]/LVP loop returned nearly to its pre-ischemia baseline location but remained flatter. Drugs had similar effects on loop location and shape as before ischemia although loops were flatter. Computation of velocity and area ratios allowed comparison of drug contractility and relaxation responses to Ca2+ before and after ischemia. Ischemia reperfusion injury caused a twofold decrease in the contractile response (increased control VRmax) and a fourfold decrease in the relaxation response (increased control VRmin) but prior treatment with any drug reduced the decreases in these responses. Dopamine and digoxin enhanced these responses before ischemia and these drugs and levosimendan did so after ischemia. Total LVP generated per total change of [Ca2+] during one cycle was enhanced by dopamine and digoxin before ischemia and by all but dobutamine after ischemia (decreased AR). This index was increased threefold by ischemia reperfusion injury (increased control AR) but prior drug treatment with dopamine, digoxin and levosimendan attenuated these increases indicating improved LVP per change in [Ca2+]. Dobutamine, dopamine and levosimendan, but not digoxin, increased coronary flow before and after ischemia. Each drug, especially dobutamine and levosimendan, increased O2 consumption and cardiac efficiency, but there were no significant differences in the relative increases in MVO2 and cardiac efficiency between drugs given before and after ischemia. This indicated that ischemia reperfusion does not blunt metabolic responses to drugs after ischemia reperfusion injury.

Overall, this study suggests (a) dopamine and digoxin similarly improve ratios of contractility, relaxation, and total LVP responsiveness to cytosolic Ca2+, (b) ischemia reperfusion injury markedly worsens responsiveness, (c) prior treatment with any of these positive inotropic drugs attenuates depression of these responses after ischemia, (d) levosimendan more effectively improves responsiveness after than before ischemia, and (e) dobutamine does not alter responsiveness before or after ischemia. Therefore, cardiotonic drugs that have different mechanisms of action appear to have different effects on the Ca2+/LVP relationship.

We have examined the cyclic Ca2+/LVP relationship in detail in two recent studies. In one, the phasic change in Ca2+ (system input) was used to model actinomyosin cross-bridge kinetics by best fitting the corresponding best-fit phasic change in isovolumetric LVP (system output) at different temperatures [26]. In another, we constructed several dynamic and static indices of the Ca2+/LVP relationship and identified those indices that were most useful to describe drug-induced alterations in the Ca2+/LVP relationship [27].

4.1 Selection and comparison of cardiotonic drugs

Criteria for selecting and comparing these four agents were that (a) they stimulated different sarcolemmal receptors (β, dopamine) or altered enzyme function (Na/KATPase, troponin C/PDE III) to increase myoplasmic Ca2+, (b) they produced a sub-maximal contractile response, (c) they could be easily washed out, and (d) they did not interfere with the fluorescent probe used to measure Ca2+. We chose the isolated heart model so that the influence of cardiac preload and afterload, blood-borne factors, and autonomic nervous function would be minimized. We did not pace hearts because it is difficult to destroy the SA and AV nodes and a faster rate for the control hearts would obscure the influence of heart rate on the measured variables in the treated groups. Digoxin and levosimendan each had a direct effect to increase heart rate by approximately 12%. This suggests that they have direct effects on SA node automaticity or that they modulate endogenous catecholamines or acetylcholine release or effects.

We selected one concentration near the EC50 for each drug. We did not conduct concentration response curves to compare efficacy of these drugs, so we could not directly compare effects of different drugs on altering cytosolic [Ca2+] and contraction and relaxation independently. Thus for much of the results (Table 2, Figs. 3,4) we compared changes in drug responses within a drug group only. The three ratios (VRmax, VRmin, and AR) furnish normalized values for indices of LVP and Ca2+ and so allow comparison among groups. We selected these ratios for comparison because they each yield information on the cost of performing work, i.e., velocity of Ca2+ influx and efflux required to produce the velocity of contraction and relaxation, as well as the net amount of Ca2+ utilized to produce a net contractile effect over the cardiac cycle. Thus an increase in any ratio meant that contractile responsiveness was decreased, and vice versa. We found that most drugs enhanced responsiveness and that ischemia reperfusion injury attenuated responsiveness. Moreover, these drugs remained capable of enhancing responsiveness, albeit at a lesser response level, after ischemia. Dobutamine alone did not alter these indices of responsiveness before or after ischemia.

It should be noted that the Ca2+/LVP loops and the above analyses of various indices of [Ca2+] plotted against LVP do not furnish any temporal information. The variable reductions in cycle length induced by the different drugs likely alter the rates of entry and exit or Ca2+ and so the rates of LVP development and relaxation. The heart rate effect is normalized to some degree by analysis of the ratios of an index of Ca2+ to the same index of LVP. We have recently addressed in detail the effect of altering heart rate on several static and dynamic indices of the Ca2+/LVP relationship [26] and on modeling the actin myosin cross bridge interaction in the isolated beating heart [27].

4.2 Ischemia reperfusion and cytosolic [Ca2+]/LVP relationship

Ischemia reperfusion injury is known to alter the myoplasmic [Ca2+]/LVP relationship but this has not been examined in detail. Global ischemia for 20 min followed by 20 min reperfusion in isolated rat hearts at 37°C led to myocardium stunning and decreased maximal extracellular Ca2+-activated force and myocardial Ca2+ responsiveness [12]. Stunned myocytes isolated from pig hearts subjected to 90 min regional ischemia and 60 min reperfusion displayed impaired contractile function and decreased Ca2+ transients [29]. We reported that 30 min global ischemia at 37°C and 60 min reperfusion increased cytosolic [Ca2+] and decreased contractility in isolated beating guinea pig hearts [14,23]. Moreover, the [Ca2+]/LVP response curves generated by changing extracellular [Ca2+] before and after ischemia exhibited a reduced maximal contractile response and a rightward shift of the normalized [Ca2+]/LVP relationship indicating reduced maximal activated contractile force and reduced sensitivity to Ca2+ [14,23]. In the present study we found that ischemia reperfusion significantly increased VRmax, VRmin and AR. This demonstrated that ischemia reperfusion increased Ca2+ influx, efflux and total cytosolic [Ca2+] but decreased contractility, relaxation and total contractile force over a cardiac cycle.

Ischemia reperfusion injury causes not only cardiac dysfunction in this model but also infarction. In other studies under similar conditions we showed approximately 50–60% left ventricular infraction and 40% recovery of systolic–diastolic LVP after 1–2 h reperfusion in control hearts [14,15,23,24]. For this study positive inotropic agents were given before ischemia. This could precondition hearts. Our results do not indicate any difference in contractile (dLVP/dtmax) or relaxant (dLVP/dtmin) effects between the control group and prior drug treated groups at 30 min reperfusion. However, responsiveness to Ca2+ was improved similarly in each treated group suggesting some preservation of the Ca2+/LVP relationship after ischemia reperfusion injury. These results confirm and extend previous results that ischemia reperfusion decreases myocardial Ca2+ sensitivity and increases Ca2+ influx and efflux associated with impaired contractility and relaxation.

4.3 Effects of cardiotonic drugs on cytosolic [Ca2+] to LVP relationships

Cardiotonic drugs have been shown to improve contractility and relaxation of stunned myocardium [18,30,31]. These drugs work principally by increasing phasic cytosolic [Ca2+]; some may also enhance myocardial responsiveness to Ca2+. Dobutamine was shown to enhance contractility in in vivo infarcted mice hearts [17] and during cardiac hypothermia [32] and to improve contractile and metabolic function in regionally ischemic hearts [21]. Our study furnishes more information on differences between dobutamine and other inotropic drugs. Dobutamine did not increase VRmax and VRmin when given before or after ischemia compared to baseline and 30 min reperfusion. This demonstrates that dobutamine increases Ca2+ influx and efflux in proportion to increases in myocardium contractility and relaxation both before and after ischemia. Dobutamine also caused the highest AR among drugs given either before or after ischemia; this suggests that dobutamine increases cytosolic [Ca2+] much higher than LVP compared to other drugs so that dobutamine had the lowest [Ca2+] efficiency. Our results suggest that although dobutamine is quite effective at augmenting contractility, the costs in terms of Ca2+ loading, particularly during diastole after ischemia are high, and it does not enhance responsiveness like the other drugs.

Levosimendan is reported to be a sensitizer at lower doses but not at higher doses [9,31]. Sato et al. [9] showed that 0.1 μM levosimendan increased contractility of right-ventricular papillary muscles and single ventricular rabbit cardiomyocytes and that these effects were not associated with increased cytosolic [Ca2+] as assessed in aequorin and indo-1 loaded cells, respectively. Levosimendan is thought to enhance binding of Ca2+ to troponin C without increasing diastolic Ca2+ unlike other positive inotropes [33]. However, 3 μM levosimendan similarly increased both contractility and cytosolic [Ca2+] as well as cAMP concentration [9]. Other reports also show that levosimendan increased cAMP concentration and contractility [5,31]. A higher cAMP level may blunt myofilament sensitivity to [Ca2+]. A milrinone-induced increase in LV contractility was associated with higher [Ca2+] in guinea pig hearts [3]. We selected a midrange levosimendan concentration (1 μM) that increased systolic–diastolic LVP by approximately 80% and dLVP/dt by threefold; because it also significantly increased systolic Ca2+, it is probable that levosimendan’s PDE III inhibitory effect overrides the Ca2+ troponin C interaction [9]. The increase in [Ca2+] was much less than for dobutamine, but was greater than for dopamine and digoxin. It was interesting that levosimendan was more effective in utilizing Ca2+ for contractility only after ischemia reperfusion injury. This suggests that levosimendan causes a lesser increase in Ca2+ influx and efflux relative to increases in contractility and relaxation after ischemia than before ischemia. Levosimendan also increased heart rate, as did each of the drugs. Thus it is clear that levosimendan can have other effects in isolated hearts.

Dopamine was shown to improve mechanical and metabolic function in post-ischemic isolated rat hearts [18]. Digoxin was reported to increase [Ca2+] less when compared to milrinone, also a PDE inhibitor, in isolated guinea pig hearts [3]. In our model digoxin was as effective as dopamine in enhancing myocardial responsiveness to Ca2+. Both dopamine and digoxin decreased VRmax and VRmin before ischemia and after ischemia. This indicates these drugs induce relative increases in Ca2+ influx and efflux that were less than the increases in dLVP/dtmax and dLVP/dtmin. Thus dopamine and digoxin had comparable effects to enhance utilization of Ca2+ to generate contractile force both before and after ischemia. Both drugs had a moderate effect to increase [Ca2+], particularly after ischemia and reperfusion, so that myocardial responsiveness to Ca2+ was markedly better than for levosimendan.

In conclusion, brief cardiac perfusion by each cardiotonic agent increased mechanical and metabolic function by increasing cytosolic [Ca2+] when given either before ischemia or after ischemia. Ca2+ responsiveness was not altered by dobutamine but was enhanced by dopamine and digoxin before ischemia and afterward also by levosimendan. It was interesting that drug treatment before ischemia helped to preserve the association of contractility and Ca2+ after ischemia, i.e., that drug effects were relatively well preserved after ischemia. Taken together, our results suggest that digoxin and dopamine increase myocardial contractility with a smaller increase in [Ca2+] than other drugs, and significantly increase myofilament Ca2+ sensitivity. This study in isolated hearts allowed us to better understand the mechanisms and differences among inotropic drugs on the Ca2+/LVP relationship during the cardiac cycle. In in vivo blood perfused and innervated hearts the effects of these drugs given over a longer time period is likely modified from our observations, but the overall effect of phasic cytosolic Ca2+ on beat-to-beat contractility and relaxation is probably quite similar. Thus, if these isolated heart results can be carried over to the clinical arena, they would suggest that digoxin and dopamine are relatively more efficacious than levosimendan, and especially dobutamine, in enhancing contractility without inducing a large increase in [Ca2+], particularly just after ischemia when [Ca2+] is quite elevated.

Acknowledgements

The research was supported in part by National Heart, Lung, and Blood Institute grants R01-HL-58691 and R01-5T32 GM-08377. The authors thank Dr. Gopu Varadarajan, Dr. Ming Tao Jiang, Dr. Leo Kevin, Jim Heisner, and Anita Tredeau for help in this study.

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
View Abstract